The Activation of Small Molecules Employing Main Group and Transition Metal Frustrated Lewis Pairs

Rebecca Claire Neu

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department of Chemistry University of Toronto

Rebecca Claire Neu

Doctor of Philosophy

Department of Chemistry University of Toronto

2012

Abstract

Combinations of sterically encumbered Lewis acids and Lewis bases are precluded from dative

bond formation, failing to yield classical Lewis acid-base adducts. These unique systems are

referred to as frustrated Lewis pairs (FLPs) and demonstrate a plethora of unique small molecule

activations.1 Herein, the syntheses of phosphonium alkoxy- and aryloxyborate salts in addition

to phosphonium thioborate salts are detailed. The scope of Lewis acid and base, alcohol and

thiol, that are effective in this chemistry, is also detailed.

The syntheses of novel borate and boronate esters of the form B(OR)3 and (C6R4O2)B(C6F5) are detailed and applied in metal-free heterolytic dihydrogen activation with phosphines. The further study of a variety of perfluoroboranes in the activation of H2 with tertiary phosphines is also

detailed. Derivatization of triarylboranes, boronate esters, borinic esters and chloroboranes by

reaction with one or two equivalents of alky- or aryldiazomethane is achieved yielding the

corresponding RR’C insertion products. The solid-state structures of the free boranes, in

addition to Lewis base adducts of a sampling of these species, are detailed. Reactivity of the

resulting sterically encumbered boranes in imine hydrogenations, H2 and XeF2 activation are also

detailed.

Combinations of intermolecular frustrated Lewis pairs are found to react collaboratively to

activate greenhouse gases such as carbon dioxide (CO2) and nitrous oxide (N2O), yielding the corresponding zwitterionic compounds R3P(CO2)BR2R’ and R3P(N2O)BR’3. Atom connectivity

has been established via X-ray crystallographic studies and molecular structures and parameters

are discussed. Subsequent exchange chemistry of both CO2 and N2O species with transition

metal and other d-block Lewis acids are investigated and yield zwitterionic compounds and ion

pairs which are otherwise unobtainable employing more conventional methods. Transition metal

containing Lewis acids are employed in conjunction with tri(tert-butyl)phosphine for the FLP-

mediated direct activation of N2O.

iii

Dedication

I dedicate this thesis in memory of my grandfathers “Grampy” and “Opa” Jack White and Hans Neu Whose passion for science and life have inspired me since the beginning.

Acknowledgments

Over the past four and a half years, I have learned that a graduate degree is not a path travelled alone but one travelled in the company of many.

I am indebted to my research supervisor Prof. Douglas Stephan for fruitful discussions, moral support, suggestions and the constant encouragement to believe in myself. I am forever grateful for the knowledge I have acquired and the opportunities that were presented to me over the years. The ability to travel abroad and to attend numerous conferences, ranging from Edmonton to Hawaii, has enriched my graduate experience and has allowed me to establish innumerous connections and relationships. For all these reasons, I will forever be grateful.

To my lab colleagues, both past and present, I thank you for your constant support, inappropriate jokes, lunchtime and coffee dates and friendship. I would especially like to thank Michael Sgro, Dr. Sharonna Greenberg, Dr. Lindsay Hounjet, Fatme Dahcheh and Dr. Jillian Hatnean for dedicating their time to the review of this thesis. Without their assistance this work would never have been finished or rid of its many typographical errors. In particular I would like to acknowledge Dr. Edwin Otten who was an inspiring mentor, who always encouraged me to believe in my abilities and to whom I owe a considerable degree of my success.

Furthermore, I would like to recognize the many support staff (NMR facility, ANALEST, mass spectrometry facility, chemistry stores, machine shop, glass blowing and administration) who have contributed to the material included in this thesis. Particularly, I would like to acknowledge our administrative assistant, Shanna Pritchard, for her continued dedication to the organization and smooth running of our lab. Lastly, thanks to our X-ray crystallographers: Dr. Alan Lough, Michael Sgro, Shokei Zhao, Christopher Brown, Adam McKinty, Christopher Caputo, Stephanie Granville, Dr. Stephen Geier and Dr. Meghan Dureen for their many hours spent mounting crystals and collecting data, which was of essence to my research.

Finally to Mom, Vati and Mike: my three greatest supporters. You have always encouraged me to reach for excellence and to pursue my dreams. When I was doubtful, you offered words of support and encouragement and celebrated with me in my successes. Most of all you have always listened, offered meaningful advice and loved me truly unconditionally. I would have never come this far in your absence. I love and thank you all.

Table of Contents

Acknowledgments ...... v

Table of Contents ...... vi

List of Tables ...... xi

List of Figures ...... xii

List of Schemes ...... xvv

List of Abbreviations ...... xx

Chapter: 1 Introduction ...... 1

1.1 Frustrated Lewis Pairs...... 1

1.1.1 The Beginning ...... 1

1.1.2 Nucleophilic Aromatic Substitution ...... 3

1.1.3 Hydrogen Activation ...... 4

1.1.4 Olefins and Dienes ...... 10

1.1.5 Tetrahydrofuran, Dioxane, Thioxane and Lactone Ring-Opening ...... 12

1.1.6 Activation of Terminal Alkynes ...... 14

1.1.7 C−F Bond Activation ...... 16

1.1.8 Activation of Disulfides ...... 17

1.2 Scope of Thesis ...... 19

Chapter 2: The Activation of O−H and S−H Bonds By Frustrated Lewis Pairs ...... 21

2.1 Introduction ...... 21

2.1.1 Methods for the Conversion of Secondary Alcohols to Ketones ...... 21

2.2 Results and Discussion ...... 26

2.2.1 O-H Bond Activation Employing Intermolecular FLPs ...... 26

2.2.4 S-H Bond Activation Employing Intermolecular FLPs ...... 36 vi

2.2.3 Mechanism of O−H and S−H Bond Cleavage ...... 39

2.2.4 Reactivity of Phosphonium Alkoxyborate Ion Pairs ...... 40

2.3 Conclusions ...... 41

2.4 Experimental Section ...... 41

2.4.1 General Considerations ...... 41

2.4.2 Syntheses...... 42

2.4.3 X-ray Crystallography ...... 50

Chapter 3: Synthesis of Borate Esters, Boronate Esters and Triarylboranes for Application in the Activation of H2 ...... 55

3.1 Introduction ...... 55

3.1.1 H2 Activation by Main Group Species...... 55

3.2 Results and Discussion ...... 58

3.2.1 Synthesis of Borate Esters ...... 58

3.2.2 Synthesis of Boronate Esters ...... 62

3.2.3 Lewis Acidity Determinations ...... 65

3.2.4 Dihydrogen Activation Employing Borate and Boronate Esters ...... 70

3.2.5 Dihydrogen Activation Employing Perfluoroarylboranes ...... 71

3.3 Conclusions ...... 74

3.4 Experimental Section ...... 74

3.4.1 General Considerations ...... 74

3.4.2 Syntheses...... 75

3.4.3 X-ray Crystallography ...... 80

Chapter 4: Borane Derivatization Employing Diazomethanes ...... 83

4.1 Introduction ...... 83

4.1.1 Tris(pentafluorophenyl)borane ...... 83

4.1.2 Current Methods for the Synthesis of Perfluoroarylboranes ...... 84

vii

4.1.3 Reaction of Boranes and Diazomethanes ...... 86

4.2 Results and Discussion ...... 87

4.2.1 Reactions of Diazomethanes with B(C6F5)3 ...... 87

4.2.2 Reactions of Diazomethanes with RB(C6F5)2 and BAr3 ...... 93

4.2.3 Reactions of Diazomethanes with Boronic Acids and Boronate Esters ...... 102

4.2.4 Reactions of Catecholborane and B-Chlorocatecholborane with Diazomethanes ...... 108

4.2.5 Reactivity of HB(C6F5)2 with Diazomethanes and Azides ...... 111

4.2.6 Reaction of Zn(C6F5)2 with Diazomethanes ...... 116

4.2.7 Applications of Derivatized Boranes in FLP Chemistry and Beyond ...... 121

4.3 Conclusions ...... 130

4.4 Experimental Section ...... 130

4.4.1 General considerations ...... 130

4.4.2 Syntheses...... 131

4.4.3 X-ray Crystallography ...... 148

Chapter 5: Activation of Carbon Dioxide by Main Group Frustrated Lewis Pairs and Subsequent Lewis Acid Exchange Chemistry ...... 154

5.1 Introduction ...... 154

5.1.1 Carbon Dioxide and the Environment ...... 154

5.1.2 Sequestration of CO2 by Main Group Frustrated Lewis Pairs ...... 155

5.1.3 Reduction of Activated CO2 ...... 157

5.2 Results and Discussion ...... 158

5.2.1 Reaction of PtBu3 and RB(C6F5)2 with CO2 ...... 158

5.2.2 Reactivity of tBu3P(CO2)B(C6F5)2Cl ...... 162

5.3 Conclusions ...... 170

5.4 Experimental Section ...... 170

5.4.1 General Considerations ...... 170 viii

5.4.2 Syntheses...... 171

5.4.3 X-ray Crystallography ...... 174

Chapter 6: Sequestration of Nitrous Oxide by Main Group Frustrated Lewis Pairs ...... 177

6.1 Introduction ...... 177

6.1.1 The Environmental Implications of Nitrous Oxide ...... 177

6.1.2 Applications of Nitrous Oxide in Synthesis ...... 177

6.1.3 Nature: The Best Model ...... 178

6.1.4 Sequestration of Nitrous Oxide by Main Group Systems ...... 178

6.2 Results and Discussion ...... 179

6.2.1 Reactions of RB(C6F5)2 and PtBu3 with N2O ...... 179

6.2.2 Reactions of BAr3 and PtBu3 with N2O ...... 188

6.2.3 Probing the Range of Lewis Basicities Tolerated in N2O Activation ...... 191

6.2.4 Mechanism of N2O Activation by FLPs ...... 194

6.2.5 Reactivity of tBu3P(N2O)BR3 ...... 196

6.2.6 Mechanism of Lewis Acid Exchange ...... 201

6.3 Conclusions ...... 203

6.4 Experimental Section ...... 203

6.4.1 General Considerations ...... 203

6.4.2 Syntheses...... 204

6.4.3 19F EXSY NMR Experimental Details ...... 208

6.4.4 X-ray Crystallography ...... 209

Chapter 7: The Exchange Chemistry of Main Group N2O Adducts with d-Block Lewis Acids ...... 213

7.1 Introduction ...... 213

2+ 7.1.1 Discovery of [Ru(NH3)5(N2O)] ...... 213

7.1.2 Reaction of N2O with Transition Metal Complexes ...... 214

7.2 Results and Discussion ...... 217

7.2.1 Reaction of Early Metal Metallocenium Cations with tBu3P(N2O)B(p-F-C6H4)3 ...... 217

7.2.2 Zinc and Phosphine Stabilized Complexes of Nitrous Oxide ...... 224

7.3 Conclusion ...... 235

7.4 Experimental Section ...... 236

7.4.1 General Considerations ...... 236

7.4.2 Syntheses...... 236

7.4.3 X-ray Crystallography ...... 241

Chapter 8: Future Work and Summary ...... 245

8.1 Future Work ...... 245

8.2 Summary ...... 246

References ...... 249

List of Tables

Table 2.1 – 11B{1H}, 19F and 31P NMR data for compounds 2-1 through 2-11...... 36

Table 2.2 – 11B{1H}, 19F and 31P NMR data for compounds 2-12 through 2-14...... 39

Table 2.3 – Selected crystallographic data for 2-1, 2-2 and 2-4...... 52

Table 2.4 – Selected crystallographic data for 2-6, 2-8 and 2-12...... 53

Table 2.5 – Selected crystallographic data for 2-13 and 2-14...... 54

11 1 19 Table 3.1 – B{ H} and F NMR data (C6F5 groups only) for 3-4, 3-5 and 3-6...... 64

Table 3.2 – Gutmann-Beckett and Childs Lewis Acidity Tests for compounds 3-1 through 3-6 and other related boranes...... 68

Table 3.3 – Selected crystallographic data for 3-2, 3-3, 3-4 and 3-10...... 82

Table 4.1 – Conversion of N-benzylidene-tert-butylamine to N-benzyl-tert-butylamine over a 30 hour time period, employing 4-1 and H2...... 126

Table 4.2 – Selected crystallographic data for 4-1, 4-2 and 4-4...... 150

Table 4.3 – Selected crystallographic data for compounds 4-5, 4-8 and 4-12 ...... 151

Table 4.4 – Selected crystallographic data for compounds 4-14, 4-17 and 4-18 ...... 152

Table 4.5 – Selected crystallographic data for 4-22, 4-25, 4-26 and 4-30...... 153

Table 5.1 – Selected 13C NMR data, coupling constants and IR stretching frequencies ...... 169

Table 5.2 – Selected crystallographic data for 5-2, 5-3 and 5-6...... 176

Table 6.1 – Comparison of relevant 15N chemical shifts and coupling constants...... 188

Table 6.2 – Lewis acidity test data determined employing the Gutmann-Beckett method ...... 189

Table 6.3 – Selected crystallographic data for 6-1, 6-4 and 6-5...... 211

Table 6.4 – Selected crystallographic data for compounds 6-7 and 6-8...... 212

Table 7.1 – Selected crystallographic data for 7-1, 7-3 and 7-4...... 243

Table 7.2 – Selected crystallographic data for 7-5 and 7-6...... 244

List of Figures

Figure 2.1 – POV-Ray depictions of the molecular structures of 2-1 and 2-2 ...... 29

Figure 2.2 – NMR spectra of B(C6F5)3 and 2-6 ...... 31

Figure 2.3 - POV-Ray depictions of the molecular structures of 2-4 and 2-6 ...... 32

Figure 2.4 - Resonance structures of 2-6 ...... 33

Figure 2.5 – POV-Ray depiction of the molecular structure of 2-8...... 35

Figure 2.6 – POV-Ray depictions of the molecular structures of 2-12 and 2-13 ...... 37

Figure 2.7 – POV-Ray depiction of the molecular structure of 2-14 ...... 39

Figure 3.1 – 19F, 1H and 11B{1H} NMR spectra of 3-1 ...... 60

Figure 3.2 – Aryl group inclination angle relative to the BO3 plane ...... 61

Figure 3.3 – POV-Ray depictions of the molecular structures of 3-2 and 3-3 ...... 62

Figure 3.4 – The effects of electron-rich and poor catechol fragments on the Lewis acidity of boronate esters ...... 63

Figure 3.5 – Correlation between the B centre and the para-F of a C6F5 group on boron ...... 64

Figure 3.6 – POV-Ray depiction of the molecular structure of 3-4 ...... 65

Figure 3.7 – (a) Gutmann-Beckett and (b) Childs Lewis acidity tests ...... 66

1 Figure 3.8 – H NMR spectra of the crotonaldehyde adducts of B(C6F5)3, 3-3 and 3-4 and the 1H NMR spectra of free crotonaldehyde ...... 67

Figure 3.9 – Lewis acidities of 3-1 through 3-6 relative to B(OC6F5)3 ...... 69

Figure 3.10 – p-orbital obstruction by a mesityl substituent...... 73

Figure 3.11 – POV-Ray depiction of the molecular structure of 3-10 ...... 73

Figure 4.1 – Examples of perfluoroarylboranes developed by (a) Marks and (b) Piers. Bisboranes developed by (c) Marks and (d) Piers...... 84

Figure 4.2 – POV-Ray depictions of the molecular structures of 4-1 and 4-2 ...... 90

xii

Figure 4.3 – POV-Ray depiction of the molecular structure of 4-4 ...... 92

Figure 4.4 – POV-Ray depiction of the molecular structure of 4-5 ...... 94

Figure 4.5 – (a) Resonance structures of a perfluorophenyl-containing borane showing deshielding at both ortho- and para- positions. (b) Full 19F NMR spectrum of 4-5. (c) Magnification of the o-F region of the 19F NMR spectrum of 4-5. (d) Magnification of the m-F region of the 19F NMR spectrum of 4-5 ...... 95

Figure 4.6 – POV-Ray depiction of the molecular structure of 4-8 ...... 99

Figure 4.7 – Resonance stabilized carbocation en route to 4-10 ...... 101

Figure 4.8 – (a) B2pπ-O2pπ orbital interaction. (b) Resonance structures of 4-11 ...... 102

Figure 4.9 – POV-Ray depictions of the molecular structures of 4-12 and 4-14 ...... 105

Figure 4.10 – Boroxine resonance contributors (a sampling) ...... 105

Figure 4.11 – POV-Ray depictions of the molecular structures of 4-17 and 4-18 ...... 108

Figure 4.12 – POV-Ray depiction of the molecular structure of 4-22 ...... 113

Figure 4.13 – Resonance structures of 4-22 ...... 113

Figure 4.14 – POV-Ray depictions of the molecular structure of 4-25 (left) and the azine-Zn2 core of 4-25 (right) ...... 117

Figure 4.15 – Electronic structures of 4-25 ...... 119

Figure 4.16 – POV-Ray depiction of the molecular structure of 4-26 ...... 120

Figure 4.17 – POV-Ray depiction of the molecular structure of 4-30 ...... 128

Figure 4.18 – (a) Full 19F NMR spectrum of 4-30. Expansion of the (b) o-F, (c) P-F and B-F and (d) p-F and m-F regions of the 19F NMR spectrum of 4-30 ...... 129

Figure 5.1 – Products of main group-mediated activation of CO2 ...... 157

Figure 5.2 – Excerpts from the 19F, 13C{1H} and 31P{1H} NMR spectra of 5-2 ...... 160

Figure 5.3 – POV-Ray depiction of the molecular structure of 5-2 ...... 161

Figure 5.4 – POV-Ray depiction of the molecular structure of 5-3 ...... 163

Figure 5.5 – Monitoring of the conversion of 5-2 to 5-4 by 31P{1H}NMR spectroscopy...... 164

xiii

Figure 5.6 – POV-Ray depiction of the molecular structure of 5-6 ...... 168

Figure 6.1 – POV-Ray depiction of the molecular structure of 6-1 ...... 183

Figure 6.2 – Structures of (a) a stabilized phosphazide and (b) a B/P adduct of N2O ...... 183

Figure 6.3 – 31P{1H} and 15N NMR spectra for 6-3 ...... 185

Figure 6.4 - POV-Ray depiction of the molecular structure of 6-4 ...... 187

Figure 6.5 – POV-Ray depiction of the molecular structure of 6-5 ...... 191

Figure 6.6 – POV-Ray depictions of the molecular structures of Cy3P(N2O)B(p-H-C6F4)3 and 6-7 ...... 195

19 Figure 6.7 – F NMR spectrum of the reaction of 6-6 with tritylborate in CD2Cl2 ...... 199

Figure 6.8 – POV-Ray depiction of the molecular structure of 6-8 ...... 200

1 19 31 1 15 Figure 6.9 – H, F, P{ H} and N NMR spectra in CD2Cl2 for 6-8 ...... 201

Figure 6.10 – 2D 19F-19F EXSY NMR spectrum ...... 203

2+ Figure 7.1 – (a) [Ru(NH3)5(N2O)] and (b) N2O resonance contributors ...... 214

Figure 7.2 – POV-Ray depiction of the molecular structure of 7-1 ...... 220

Figure 7.3 – POV-Ray depiction of the molecular structure of 7-3 ...... 224

Figure 7.4 – (a) 1:1 and (b) 2:1 adducts of Zn(C6F5)2 with triphenylphosphine and bis(diphenylphosphino)ethane ...... 225

Figure 7.5 – POV-Ray depictions of the molecular structure of 7-5 ...... 228

Figure 7.6 – POV-Ray depictions of molecular structure and of 7-6 ...... 230

Figure 7.7 – Low temperature 19F NMR spectroscopy of 7-6 ...... 231

Figure 7.8 – POV-Ray depictions of the molecular structure of 7-7...... 234

Figure 7.9 – 19F and 31P{1H} spectra for compounds 7-5 through 7-7 ...... 236

xiv

List of Schemes

Scheme 1.1 – (a) Lewis acid-base adduct formation and (b) a Frustrated Lewis pair ...... 1

Scheme 1.2 – Previously observed frustrated Lewis pair reactivity ...... 3

Scheme 1.3 – (a) Lewis acid-based adduct formation and (b) nucleophilic substitution at the

para-carbon of the trityl cation and (c) B(C6F5)3 ...... 4

Scheme 1.4 – Synthesis of the first metal-free catalyst capable of reversible H2 activation ...... 5

Scheme 1.5 – Synthesis of intramolecular FLPs and subsequent H2 activation: (a) B/P and (b) B/N ...... 6

Scheme 1.6 – Heterolytic H2 activation employing intermolecular FLPs ...... 6

Scheme 1.7 – Imine reduction by an intramolecular phosphonium-borate catalyst ...... 8

Scheme 1.8 – Reduction of aldimines and ketimines employing B(C6F5)3 and H2 ...... 9

Scheme 1.9 – Reduction of N-bound arenes employing B(C6F5)3 and H2 ...... 9

Scheme 1.10 – Reaction of intermolecular (a) and (b) and intramolecular (c) and (d) FLPs with olefinic and double bond-containing substrates ...... 10

Scheme 1.11 – Reactions of 1,3-dienes with intermolecular FLPs ...... 11

Scheme 1.12 – THF ring-opening reactions: (a) mechanism (b) trityl anion and borane (c) zirconium Lewis acid and phosphine and (d) borane and phosphine ...... 12

Scheme 1.13 – (a) Reactions of dioxane and thioxane with intermolecular FLPs and (b) the

three thioxane−(B(C6F5)3)n (n = 1 or 2) binding modes observed by nuclear magnetic resonance (NMR) spectroscopy ...... 13

Scheme 1.14 – Reaction of intermolecular FLPs with (a) δ-valerolactone and (b) lactide ...... 14

Scheme 1.15 – Reactions of frustrated Lewis pairs with terminal alkynes: (a) deprotonation and (b) 1,2-addition ...... 14

Scheme 1.16 – Application of (a) Al/P and (b) B/N FLP systems for the activation of terminal alkynes to yield cyclized products ...... 15

Scheme 1.17 – Reaction of intermolecular FLPs with bis-acetylenic substrates: (a)

B(C6F5)3/PPh3 and B(C6F5)3/PtBu3 and (b) B(C6F5)3/P(C6F5)3 ...... 16

Scheme 1.18 – FLP-mediated C−F bond activation ...... 16

Scheme 1.19 – (a) Activation of fluoroalkanes employing P/B FLPs yielding phosphonium

fluoroborate salts and (b) reaction of fluoroalkane, B(C6F5)3 and [tBu3PH][HB(C6F5)3] or Et3SiH to yield free alkanes ...... 17

Scheme 1.20 – Activation of disulfides by: (a) intramolecular and (b) intermolecular FLPs ...... 18

Scheme 1.21 – Disulfide activation employing a strictly carbon-based FLP system ...... 19

Scheme 2.1 – Synthesis of a chlorosulfonium ion ...... 22

Scheme 2.2 – Secondary alcohol oxidation by a chlorosulfonium ion ...... 22

Scheme 2.3 – Dess-Martin oxidation...... 23

Scheme 2.4 – Oppenauer oxidation ...... 24

Scheme 2.5 – Corey-Kim oxidation ...... 25

Scheme 2.6 – Pfitzner-Moffatt oxidation ...... 25

Scheme 2.7 – Proposed cycle for the conversion of alcohols to ketones ...... 27

Scheme 2.8 – Synthesis of 2-1 and 2-2...... 27

Scheme 2.9 –Synthesis of 2-4, 2-5 and 2-6...... 30

Scheme 2.10 – Synthesis of 2-7 ...... 33

Scheme 2.11 – Synthesis of compounds 2-8, 2-9, 2-10 and 2-11...... 34

Scheme 2.12 – Synthesis of 2-12, 2-13 and 2-14...... 38

Scheme 2.13 – Activation of alcohols and thiols by intermolecular FLPs ...... 40

Scheme 2.14 – Reaction of [R3PH][R’OB(C6F5)2R”] with N-benzylidene-tert-butylamine, yielding the corresponding imminium alkoxyborates...... 41

Scheme 3.1 – Main group species capable of H2 activation. (a) benzophenone/KOtBu, (b) difluorovinylidene and (c) acyclic alkyl amino carbenes...... 56

xvi

Scheme 3.2 – Intramolecular frustrated Lewis pairs capable of reversible and irreversible H2 activation. (a) Stephan, (b) Erker and Rieger and Repo ...... 57

Scheme 3.3 – Reversible and irreversible H2 activation (a) Stephan, (b) Erker and (c) Stephan ...... 58

Scheme 3.4 – Synthesis of compounds 3-1, 3-2 and 3-3...... 59

Scheme 3.5 – The synthesis of 3-7...... 70

Scheme 3.6 – Synthesis of (a) 3-8 and 3-9 and (b) 3-10 and 3-11...... 72

Scheme 4.1 – Activation of dimethylzirconocene by B(C6F5)3 ...... 83

Scheme 4.2 – 1,1-Carboboration reactions employing tris(pentafluorophenyl)borane and (a) terminal alkynes yielding a mixture of E/Z isomers and (b) internal alkynes...... 85

Scheme 4.3 – Decomposition of a B(C6F5)3-silylene adduct...... 85

Scheme 4.4 – Reaction of diazoacetate with dialkylchloroborane via the migration of (a) chloride or (b) an alkyl group. (c) Derivatization of B-R-9 borabicyclononane via insertion of trimethylsilyldiazomethane into a B-C bond...... 87

Scheme 4.5 – The synthesis of 4-2 and 4-1...... 88

Scheme 4.6 – Synthesis of 4-4 ...... 91

Scheme 4.7 – Mechanism for diazomethane insertion into B(C6F5)3...... 93

Scheme 4.8 – Synthesis of 4-6, 4-7 and 4-8 ...... 98

Scheme 4.9 – Proposed mechanism for the synthesis of 4-10 ...... 100

Scheme 4.10 – Synthesis of 4-12, 4-13 and 4-14...... 102

Scheme 4.11 – The reversible dehydration/rehydration reaction of a boronic acid to the corresponding boroxine...... 103

Scheme 4.12 – Synthesis of compound 4-15, 4-16, 4-17 and 4-18 ...... 106

Scheme 4.13 – Synthesis of 4-19, 4-20 and 4-21...... 110

Scheme 4.14 – Synthesis of of 4-22 ...... 112

Scheme 4.15 – Proposed mechanism for the synthesis of 4-22 ...... 114

xvii

Scheme 4.16 – Synthesis of compounds 4-23 and 4-24 ...... 115

Scheme 4.17 – Mechanism for the reaction of organic azides N3R (R = non-bulky substituent) with HB(C6F5)2...... 116

Scheme 4.18 – (a) Mechanism for the synthesis of 4-25. (b) Ketazine formation employing a amido-bis(phenolate) Ta(V) complex...... 118

Scheme 4.19 – Synthesis of 4-26 from 4-25 and from tol⋅Zn(C6F5)2 ...... 120

Scheme 4.20 – Proposed catalytic cycle for the reduction of ketones...... 123

Scheme 4.21 – Mechanism for the synthesis of 4-28 ...... 124

Scheme 4.22 – Reduction of N-benzylidene-tert-butylamine employing H2 and 4-1 ...... 125

Scheme 4.23 – Synthesis of compounds 4-29 and 4-30...... 127

Scheme 5.1 – CO2 sequestration by (a) intermolecular FLP and (b) intramolecular FLP ...... 155

Scheme 5.2 – Full hydrogenation of CO2 to methane ...... 158

Scheme 5.3 – Synthesis of (a) 5-1 and (b) 5-2...... 160

Scheme 5.4 – Chloride abstraction from 5-2 by TMSOTf yielding 5-3 and the reaction of 5-2

with Et3SiH to yield the classic Lewis acid-base adduct...... 162

Scheme 5.5 – Synthesis of 5-5 and 5-6...... 166

Scheme 5.6 – Intramolecular activation of CO2 by (a) a metallocenium cation and phosphine and (b) a neutral zirconocene and phosphine (Cp* = C5Me5)...... 168

Scheme 5.7 – Mechanism of Lewis acid exchange in 5-2...... 169

Scheme 6.1 – Examples of the oxidation of (a) alkenes to the related carbonyl compounds

and (b) phosphines to the related phosphine oxides employing N2O ...... 178

Scheme 6.2 – The sequestration of N2O by a basic and bulky N-heterocyclic carbene ...... 179

Scheme 6.3 – Synthesis of 6-1...... 181

Scheme 6.4 – Synthesis of 6-2 and 6-3...... 183

Scheme 6.5 – The synthesis of the bis-zwitterionic compound 6-4...... 185

Scheme 6.6 – The two step activation of N2O by 1,4-(C6F4)[B(C6F5)2]2 and 2PtBu3...... 187

xviii

Scheme 6.7 – The synthesis of 6-5 and 6-6...... 189

Scheme 6.8 – Nucleophilic attack by PCy3 at the para-C of a C6F5 ring of B(C6F5)3...... 192

Scheme 6.9 – Synthesis of Cy3P(N2O)B(p-H-C6F4)3 and 6-7 ...... 192

Scheme 6.10 – Computationally determined mechanism for the Staudinger synthesis of aza-ylides ...... 195

Scheme 6.11 – Proposed mechanism for the activation of N2O by B/P FLPs...... 196

Scheme 6.12 – Mechanism for decomposition of tBu3P(N2O)BR’3 to (tBu3PO)BR’3 ...... 196

Scheme 6.13 – Reaction of tritylborate with PCy3 and PtBu3...... 197

Scheme 6.14 – Exchange reaction of 6-6 with [Ph3C][B(C6F5)4] to yield 6-8...... 199

Scheme 6.15 – Proposed mechanism of Lewis acid (LA) exchange employing 6-6...... 202

Scheme 7.1 – Reaction of N2O with: (a) Cp*2Hf(H)Ph and (b) Cp*2M(PhCCPh) (M = Ti and Zr) ...... 215

Scheme 7.2 – Reaction of (dtbpe)Ni(CPh2) with stoichiometric and excess N2O ...... 216

Scheme 7.3 – N–O and N–N bond scission employing an NHC and a low valent V complex ..217

Scheme 7.4 – Synthesis of 7-1 and 7-2 via Lewis acid exchange...... 218

Scheme 7.5 – Synthesis of Cp*2ZrMe(OMe)...... 221

Scheme 7.6 – Synthesis of 7-3 and 7-4...... 222

Scheme 7.7 – Synthesis of 7-5...... 225

Scheme 7.8 – Synthesis of 7-6 ...... 228

Scheme 7.9 – Proposed mechanism of Zn(C6F5)2 site exchange in 7-6...... 231

Scheme 7.10 – Synthesis of 7-7...... 232

Scheme 7.11 – Process of Zn(C6F5)2 exchange in 7-7...... 233

Scheme 7.12 – Synthesis of 7-5, 7-6 and 7-7 by direct and indirect methods...... 234

xix

List of Abbreviations

Å angstrom, 10-10 m ° Degrees δ chemical shift

Δδp-m para-meta separation ∆ heat ΔG‡ Gibbs free energy of activation ΔH‡ enthalpy of activation ΔS‡ entropy of activation π pi σ sigma λ lambda, wavelength ν wavenumber, cm-1 μL microliters, 10-6 L 6-31G(d) type of basis set Anal analytical atm atmospheres B3LYP a type of DFT exchange-correlational functional br broad C Celsius C Childs

C60 Buckminsterfullerene

C6D5Br deuterated bromobenzene Calcd. calculated

CD2Cl2 deuterated dichloromethane

C4D8O deuterated tetrahydrofuran 5 C5Me5 pentamethylcyclopentadienyl, η -C5(CH3)5 cent centroid CHN carbon, hydrogen, nitrogen CO carbonyl

COD 1,5-cyclooctadiene 5 Cp cyclopentadienyl anion, η -C5H5 5 Cp* pentamethylcyclopentadienyl, η -C5(CH3)5

Cy cyclohexyl, C6H10 d doublet

dcalc calculated density dd doublet of doublets ddd doublet of doublet of doublets DFT density functional theory DMP Dess-Martin periodinane DMSO dimethylsulfoxide dtbpe bis(di-tert-butylphosphino)ethane) equiv. equivalent et al. et alia (and others) etc. et cetera

Et ethyl, C2H5

Et2O diethyl ether EXSY exchange spectroscopy

Fc calculated structure factor

Fo observed structure factor FLP frustrated Lewis pair g grams G-B Gutmann-Beckett GOF goodness of fit h hour

H3 third proton of crotonaldehyde Hz Hertz, s-1 HOMO highest occupied molecular orbital HRMS high resolution mass spectrometry in vacuo under vacuum

iPr isopropyl, (CH3)2CH IR infrared

xxi

J scalar coupling constant K Kelvin kB Boltzmann constant kcal kilocalories, 103 cal LA Lewis acid L.A. Lewis acidity LB Lewis base LUMO lowest unoccupied molecular orbital m meta m multiplet MAO methylalumoxane Me methyl

MeCN acetonitrile, CH3CN

Mes mesityl, 2,3,6-(CH3)3C6H2 MHz megahertz, 106 Hz mg milligram, 10-3 g mL millilitre, 10-3 L mmol millimol, 10-3 mol mol mol m/z mass-to-charge ratio

NEt3 triethylamine, N(C2H5)3 NHC N-heterocyclic carbene NMR Nuclear Magnetic Resonance o ortho o-tol ortho-tolyl, 2-(CH3)C6H4

OTf trifluoromethanesulfonate, OSO2CF3 p para

Ph phenyl, C6H5 pH measure of acidity and basicity (potential hydrogen) POV-Ray Persistence of Vision Raytracer ppb parts per billion, 10-9 ppm parts per million, 10-6 xxii

PR3 tertiary phosphine R chirality configuration R gas constant Rw weighted residual r.t. room temperature, 25 °C s singlet S chirality configuration sim simulated SIMes 1,3-dimesitylimidazolidin-2-ylidene

SiMe4 tetramethylsilane, Si(CH3)4

SMe2 dimethylsulfide, S(CH3)2 T temperature t triplet td triplet of doublets

THF tetrahydrofuran, C4H8O tm triplet of multiplets TMP 2,2,6,6-tetramethylpiperidine TMPH 2,2,6,6-tetramethylpiperidinium ion

TMS trimethylsilyl, Me3Si tol toluene tBu tert-butyl q quartet V volume

xxiii

Chapter 1 Introduction

1.1 Frustrated Lewis Pairs

1.1.1 The Beginning

The foundation for acid-base chemistry was laid by G. N. Lewis in 1923, when he described Lewis acids and bases as species that have the ability to function as either electron pair acceptors or electron pair donors.2 These molecules are characterized by their ability to interact yielding a formal dative bond thereby quenching the respective Lewis acidity and basicity of the two components.1 The lowest unoccupied molecular orbital (LUMO) of the acidic species readily interacts with the highest occupied molecular orbital (HOMO) of the base, yielding the expected acid-base adduct (Scheme 1.1). A classic model of this reactivity is the combination of ammonia

and borane to yield ammonia borane, H3NBH3. Astonishingly, this simplistic model derived by Lewis nearly 90 years ago has since been fundamental to our understanding of chemical reactivity in the fields of transition metal and main group chemistry.1,3,4

Scheme 1.1 – (a) Lewis acid-base adduct formation and (b) a Frustrated Lewis pair.

Contrary to reactivity noted by Lewis, Stephan et al. recently noted that when the acidic and basic moieties are sterically encumbered, the route to classic adduct formation is precluded, rendering the acidic and basic components available for alternative reactivity (Scheme 1.1).1,3,5,6

This concept of sterically induced chemical frustration, preventing Lewis acid-base adduct formation, was established in the late 2000s and coined as frustrated Lewis pairs (FLPs).1,3,4,6 This model, although only recently popularized, finds its roots in the early portion of the last century.

Unique FLP behavior was initially observed in 1942 by H. C. Brown, upon the reaction of 2,6- dimethypyridine (lutidine) with boron-containing Lewis acids. In a much anticipated fashion, stoichiometric reactions of lutidine with the small, hard Lewis acid borontrifluoride (BF3), 7 resulted in the formation of the corresponding adduct, (C5H3(CH3)2N)BF3 (Scheme 1.2, a).

Surprisingly, an analogous reaction with the larger Lewis acid, trimethylborane, (BMe3), resulted in no observable adduct formation, due to the steric protection of the nitrogen atom provided by the ortho-methyl substituents, preventing approach of the borane.

Similarly, in 1959 Wittig and Benz described the reaction of in situ generated benzyne, which in the presence of triphenylphosphine (PPh3) and triphenylborane (BPh3) did not yield the expected acid-base adduct but rather yielded the product of phosphine and borane 1,2-addition across the benzyne triple bond (Scheme 1.2, b).8

Shortly following the discovery by Wittig and Benz, Tochtermann reported similar addition

chemistry upon the reaction of the trityl anion, 1,3-butadiene and BPh3, yielding the product of diene 1,2-addition as opposed to traditionally observed anionic polymerization (Scheme 1.2, c).9

Scheme 1.2 – Previously observed frustrated Lewis pair reactivity.7,8,9

1.1.2 Nucleophilic Aromatic Substitution

Over the years, there have been numerous reports of adduct formation between the strongly 10,11,12 electrophilic borane, tris(pentafluorophenyl)borane B(C6F5)3, and a range of Lewis bases.11,13-20 Investigations conducted by Stephan et al. revealed that the highly electrophilic

carbon-based B(C6F5)3 analogue, triphenylmethylcarbenium ion, ([CPh3] or trityl), reacts in an anticipated donor-acceptor fashion when in the presence of strongly basic donors. Simple

reaction of the salt trityl borate ([Ph3C][B(C6F5)4] with a stoichiometric amount of trimethylphosphine yields the expected donor-acceptor compound (Scheme 1.3, a).21

Reaction of the more basic triisopropylphosphine (PiPr3), (electronic parameters - PiPr3: 2059.2 -1 -1 22 cm ; PMe3: 2064.1 cm ) with increased steric demands (Tolman cone angles - PiPr3: 160°; 22 PMe3: 118°) resulted in unexpected reactivity. Nucleophilic attack of the phosphine is observed at the para-carbon of the trityl cation yielding a cyclohexadienyl-containing species, followed by rearrangement to yield a [p-benzhydryl-phenyl]-phosphonium salt (Scheme 1.3, b).21 Such rearrangements have also previously been reported by Bidan and Genie in 1978.23

Scheme 1.3 – (a) Lewis acid-based adduct formation and (b) nucleophilic substitution at the 21 para-carbon of the trityl cation and (c) B(C6F5)3.

In a related fashion, Stephan et al. reported stoichiometric reactions of bulky secondary and

tertiary phosphines with B(C6F5)3 (Scheme 1.3, c). In these instances, nucleophilic aromatic

substitution occurs at the para-carbon of a C6F5 ring, followed by fluoride migration to the neighbouring boron centre yielding a phosphonium fluoroborate zwitterion.24

1.1.3 Hydrogen Activation

1.1.3.1 H2 Activation by Intramolecular and Intermolecular FLPs

A significant breakthrough in the field of hydrogenation chemistry was contributed by Stephan in 2006, upon the realization that zwitterionic products of nucleophilic aromatic substitution could easily be converted to phosphonium hydridoborate perfluorophenyl-linked zwitterions 24,25 upon reaction with chlorodimethylsilane. The linked system, derived from B(C6F5)3 and

HPMes2, was found to readily undergo H2 liberation at 150 °C yielding a neutral phosphino-

borane species. Most surprisingly was the discovery that upon exposure to a stream of H2 at room temperature, a single molecule of H2 is readily taken up regenerating the phosphonium

hydridoborate zwitterion. This novel advancement marked the first demonstrated instance of

metal-free reversible H2 activation (Scheme 1.4).

Scheme 1.4 – Synthesis of the first metal-free catalyst capable of reversible H2 activation.

Subsequent investigations by Erker in 2007 led to the development of an intramolecular

ethylene-linked frustrated Lewis pair as a product of the anti-Markovnikov addition of 26 bis(pentafluorophenyl)borane (HB(C6F5)2), to a vinyl phosphine. Similarly, Rieger and Repo reported the synthesis of an intramolecular boron-nitrogen frustrated Lewis pair derived from the reaction of 2-bromobenzylbromide, 2,2,6,6-tetramethylpiperidine and 27,28 bis(pentafluorophenyl)chloroborane (ClB(C6F5)2). The systems developed by Erker, Rieger and Repo were found to initiate the heterolytic cleavage of dihydrogen; however, notably only the ammonium hydridoborate zwitterion developed by Repo and Rieger was capable of reversible H2 activation (Scheme 1.5).

26 Scheme 1.5 – Synthesis of intramolecular FLPs and subsequent H2 activation: (a) B/P and (b) B/N.27,28

An evident limitation to these catalysts, and related linked systems,29,30 is the demanding nature of the synthetic preparations. Simple catalyst systems were sought to circumvent the tedious syntheses of these intramolecular precursors. The Stephan group first examined the interaction of intermolecular sterically encumbered frustrated Lewis pairs with molecular hydrogen.

Stoichiometric reactions of B(C6F5)3 and tertiary phosphines (PR3: R = tBu, Mes) resulted in no

interaction as evidenced spectroscopically down to -50 °C. Exposure to a stream of H2 resulted

in the isolation of phosphonium hydridoborate ion pairs which were the product of heterolytic H2 31 cleavage. Despite close P−H⋅⋅⋅H−B contacts, these species are immune to H2 liberation as was previously observed for some related zwitterionic species (Scheme 1.6). These initial observations were further elaborated upon in the years that followed, resulting in a string of related intermolecular FLPs which employ a plethora of Lewis acids and bases.1

Scheme 1.6 – Heterolytic H2 activation employing intermolecular FLPs.

1.1.3.2 Mechanism of H2 Activation

The mechanism of H2 activation employing frustrated Lewis pair systems remains the centre of

discussion, controversy and study. Little reliable kinetic data describing the process of H2 activation employing the phosphino-borane system developed by Stephan has been gathered to date, attributed to the difficulty associated with the accurate control of H2 concentrations in solution in addition to the inability to study the reverse reaction due to the extremely fast nature 1 of H2 liberation.

Despite this, small molecule activation employing intermolecular frustrated Lewis pairs has been the centre of numerous computational studies. Investigations in the mid to late 1990s show 32-35 evidence for the interaction of H2 and BH3 which would indicate that in the presence of

B(C6F5)3, H2 is possibly first activated by the borane, and then subsequently attacked by the phosphine moiety. Similarly, one cannot discount initial activation by the phosphine followed by interaction with the borane, as low-temperature matrix-isolation studies have demonstrated 36,37 the possibility for end-on attack by the nucleophilic phosphine on the H2 moiety.

More recent computational studies by Pápai suggest the existence of an “encounter complex” which loosely holds the Lewis acidic and basic components together via substituent H···F bonding, without the quenching of the respective acidic or basic sites. An electric field is created in the crevice between the boron and phosphorous centres and allows for the entrance of a 38-41 molecule of H2, which is subsequently polarized and heterolytically cleaved. Successive investigations by Grimme et al. cast doubt on the observation of a linear P···H−H···B transition state as postulated by Pápai et al. and suggest that the distance between the P and B centres is considerably contracted as compared to that previously determined via quantum mechanical methods. This new finding would imply that due to steric crowding, a linear orientation of the

H2 molecule within the encounter complex is sterically prohibited; therefore, the limiting step to

hydrogen activation is the entrance of a molecule of H2 into the P···B pocket. Once contained 42 within the phosphine and borane cavity, heterolytic H2 cleavage should be barrierless.

1.1.3.3 Reduction of Unsaturated Substrates

Due to the ease of H2 activation and liberation employing the phosphino-borane system

(Mes2PC6F4B(C6F5)2) developed by Stephan et al. this catalyst was employed for the catalytic

reduction of imines. Reactions employing 5 mol % of the phosphonium hydridoborate zwitterion in conjunction with a variety of imines resulted in the catalytic reduction to the corresponding amines.6,43,44 Sterically encumbered and basic imines were found to undergo complete reduction much more readily as compared to bulky and electron-deficient or small

electron-rich imines, which generally required longer reaction times and increased H2 pressures and temperatures.43 Computational studies revealed that proton transfer from the phosphorous centre to the imine N-atom is the first step in the reaction sequence followed by hydride attack by the borohydride moiety,39 implying that the rate of reduction is in fact a function of the basicity of the imine and not a question of the steric bulk of the substrate (Scheme 1.7).

Scheme 1.7 – Imine reduction by an intramolecular phosphonium-borate catalyst.43

Subsequent investigations revealed that the substrate can serve a dual role in the reduction 45 process, acting as both the H2 receptor and Lewis base in the heterolytic activation of H2. Hydrogenations of aldimines and ketimines were achieved by heating imines in the presence of catalytic amounts of B(C6F5)3 at 120 °C under 4 atmospheres of H2, in as little as one hour or as 45 long as 48 hours. The mechanism of reduction is believed to occur via initial heterolytic H2 activation by the imine and B(C6F5)3 FLP yielding an iminium hydridoborate salt. Subsequent proton and hydride transfer occurs collapsing the ion pair, yielding an amine-borane adduct (Scheme 1.8). Liberation of the free amine is achieved under thermal duress.45

45 Scheme 1.8 – Reduction of aldimines and ketimines employing B(C6F5)3 and H2.

Main group catalysts in the aforementioned paragraphs (Section 1.1.3.1) have since been applied in the reduction of organic substrates such as imines (aldimines and ketimines), diimines, protected nitriles, silyl enol ethers, enamines, dieneamines, aziridines, enones, N-heterocycles, indoles, quinolines and the partial reduction of carbonyl-containing molecules.1,3-6,44

In a very recent communication, Stephan et al. reported the unprecedented stoichiometric hydrogenation of N-bound arenes yielding cyclohexyl derivatives.46 Phenyl-bound secondary amines display resonance structures where delocalization of the nitrogen lone pair into the phenyl aromatic system is possible, formally placing an increased degree of electron density at the para-carbon atom of the phenyl group.46 This basic carbon atom subsequently partakes in heterolytic H2 cleavage in conjunction with the bulky Lewis acid, B(C6F5)3 at 110 °C. Repeated activation and reduction steps provide the fully reduced cyclohexylamine derivative which is isolated as the ammonium hydridoborate salt (Scheme 1.9).46

46 Scheme 1.9 – Reduction of N-bound arenes employing B(C6F5)3 and H2.

1.1.4 Olefins and Dienes

Following the discovery of unique FLP-mediated H2 activations, Stephan et al. examined the interaction of FLPs in the presence of simple olefins. Initial studies revealed no interaction, as detected by spectroscopic methods, between the individual components of the FLP in the presence of olefin. Exposure of two-component FLP systems to olefin (ethylene, propylene and 1-hexene) yielded ethyl-linked phosphonium-borate zwitterionic compounds (Scheme 1.10, b).47 It is of note that phosphine attack routinely occurs at the most hindered carbon atom with addition of the borane to the adjacent unhindered carbon of the double bond. Similarly, reaction of phosphines, bearing tethered olefinic functionalities, with B(C6F5)3, results in addition to the olefinic fragment yielding cyclic phosphonium-borate compounds (Scheme 1.10, a).47 Subsequent transformations have been reported in a collaborative effort by the Erker and Stephan groups and demonstrate how N,N-dimethylanilines, with tethered olefinic fragments, offer a prearranged orientation of the olefin for facile attack by the amine and borane moieties yielding cyclized phosphonium-borate zwitterions.48

Scheme 1.10 – Reaction of intermolecular (a)47 and (b)47 and intramolecular (c)49 and (d)50 FLPs with olefinic and double bond-containing substrates.

Further examples of intramolecular FLP-mediated olefin activations soon followed from the Erker group highlighting the ability of intramolecular ethylene-linked frustrated Lewis pairs to add to the double bond of norbornene, ethyl vinyl ether and trans-cinnamic aldehyde

(Scheme 1.10, c).49 Interestingly, reaction of the analogous ethylene-linked FLP with pentafulvene resulted in the generation of a B/P-containing product of pentafulvene dimerization, which can be liberated from the phosphorous and boron moieties upon heating, yielding a rare [6+4] cycloaddition product (Scheme 1.10, d).50 Recent reports demonstrate the application of an

FLP bearing geminal electron-deficient phosphine (P(C6F5)2) and borane (B(C6F5)2) fragments, which act to intramolecularly activate ethylene yielding a B/P-containing five membered ring.51

Intermolecular FLPs have also found application in the activation of 1,3-dienes, such as 1,3- butadiene, 2,3-dimethyl-1,3-butadiene, 2,3-diphenyl-1,3-butadiene and 1,3-cyclohexadiene, producing butene linked phosphonium-borate zwitterions, the products of 1,4-addition. These products of addition proved to be less high-yielding when compared to other intermolecular FLP-mediated olefin activations, and often yielded mixtures of isomers which were not readily separable (Scheme 1.11).52

Scheme 1.11 – Reactions of 1,3-dienes with intermolecular FLPs.

Since the first report of FLP-mediated olefin activation, numerous questions have arisen regarding the mechanism for the observed termolecular reactions. These systems have been studied computationally by both Guo and Pápai; yet both provide differing rationale for the

observed reactivity. Guo suggests a weakly bound complex of B(C6F5)3 and ethylene, thereby activating the olefin toward attack by the phosphine.53 According to these studies, the olefin is only slightly activated as evidenced by the elongation of the C−C double bond by a mere 0.002 Å.53 Comparatively, computations by Pápai suggest the generation of a pre-organized "encounter complex" held together by secondary interactions between the Lewis acid and base substituents (H···F bonding) which offers a cavity bearing a P···B distance of 4.2 Å, capable of olefin acceptance and activation.38,41 In recent investigations, Stephan et al. provide both spectroscopic and theoretical evidence supporting the existence of a borane-olefin van der Waals complex, thereby further substantiating the conclusions made by Guo.54

1.1.5 Tetrahydrofuran, Dioxane, Thioxane and Lactone Ring-Opening

In 1950, Wittig and Rückert reported the reaction of the tetrahydrofuran (THF) adduct of triphenylborane with the trityl anion. Chemical intuition would point toward the substitution of 55 THF for the stronger donor, yielding the corresponding salt, [Na][Ph3C−BPh3]. To their surprise, addition of the donor resulted in the nucleophilic ring-opening of THF yielding the corresponding B/P alkoxy-linked zwitterion (Scheme 1.12, a). In the years that followed,

Stephan would report similar reactivity of [Cl4Zr(THF)2] and B(C6F5)3 in the presence of basic and bulky phosphines yielding dimeric and monomeric THF ring-opening products (Scheme 1.12, c and d).56,57 In fact, numerous examples of transition metal and main group-based Lewis acids have been implicated in mediating THF ring-openings and include aluminum,58 uranium,59,60 samarium,61 titanium,62,63 zirconium,64 manganese65 and tellurium66,67 Lewis acidic species.

Scheme 1.12 – THF ring-opening reactions: (a) mechanism (b) trityl anion and borane (c) zirconium Lewis acid and phosphine and (d) borane and phosphine.

Subsequent THF ring-opening investigations have been conducted employing a series of Lewis bases such as amines, pyridines and diimines, and have been found to yield phosphonium-borate

zwitterions analogous to those initially reported.68 It is of note that dinuclear bases, such as N,N,N',N'-tetramethyl-p-phenylenediamine, are found to activate a single equivalent of THF in 68 the presence of B(C6F5)3, leaving one basic site unreacted.

Reactions of intermolecular FLPs with dioxane and thioxane operate analogously resulting in the formation of B−O and P−C bonds upon ring-opening, (Scheme 1.13, a). Although thioxane contains a competing nucleophilic sulfur atom, solid-state molecular structures reveal that the

ring-opening products are B−O bound, despite the observance of S−B(C6F5)3 and O−B(C6F5)3 adducts in addition to a bis-Lewis acid adduct of thioxane, in solution prior to Lewis base addition (Scheme 1.13, b).68

Scheme 1.13 – (a) Reactions of dioxane and thioxane with intermolecular FLPs and (b) three

thioxane−(B(C6F5)3)n (n = 1 or 2) binding modes observed by nuclear magnetic resonance (NMR) spectroscopy.68

In a manner analogous to THF, dioxane and thioxane ring-opening reactions, intermolecular frustrated Lewis pairs can be utilized for the facile activation and ring-opening of lactones, 69 generating products of the form LB(CH2)4CO2B(C6F5)3 (LB: Lewis base). Reaction of lactide, however, results in divergent reactivity where ring contraction is observed yielding diastereomers of the ring-contracted anion (Scheme 1.14).

Scheme 1.14 – Reaction of intermolecular FLPs with (a) δ-valerolactone and (b) lactide. LB = Lewis base.69

1.1.6 Activation of Terminal Alkynes

Investigations by Stephan et al. demonstrated that terminal alkynes react with combinations of frustrated Lewis pairs to result in deprotonation reactions yielding phosphonium-borate salts or the products of 1,2-addition.70-72 Preference for deprotonation versus addition is generally

mediated by the nature of the Lewis base. Basic and bulky phosphines such as PtBu3, typically result in deprotonation products yielding phosphonium alkynylborate ion pairs while smaller bases, such as tri-ortho-tolylphosphine (P(o-tol)3) undergo 1,2-addition reactions in the presence

of B(C6F5)3 (Scheme 1.15, a and b).

Scheme 1.15 – Reactions of frustrated Lewis pairs with terminal alkynes: (a) deprotonation and (b) 1,2-addition.

The reactivity of terminal alkynes with FLPs has since been widely elaborated upon by the Berke, Erker and Uhl groups, demonstrating reactions with a variety of Lewis acids such as alanes70,71,73,65,66,68 non-perfluoro-70,71 and vinyl boranes74,75 and Lewis bases such as phosphines70,71,76,65,66,71 imines,71 amines,71 thioethers,71 carbenes,71 enamines72 and pyrroles72 in addition to the application of intramolecular FLPs, yielding cyclized, 1,2-addition products (Scheme 1.16, a and b).48,73

Scheme 1.16 – Application of (a) Al/P and (b) B/N FLP systems for the activation of terminal alkynes to yield cyclized products.

Interestingly, the bis-acetylenic substrate, 1,4-diethynylbenzene, has shown to be active in both deprotonation and 1,2-addition chemistry, where one acetylenic fragment undergoes 1,2-addition while the remaining fragment is subject to deprotonation (Scheme 1.17, a).71 The related bis- acetylenic species 1,2-diethynylbenzene first undergoes a 1,1-carboboration reaction at a single acetylenic group in the presence of B(C6F5)3 yielding a vinylborane followed by 1,2-addition by another molecule of borane and by the electron deficient tris(pentafluorophenyl)phosphine 75 (P(C6F5)3) providing a cyclized product of addition (Scheme 1.17, b).

Scheme 1.17 – Reaction of intermolecular FLPs with bis-acetylenic substrates: (a) 71 75 B(C6F5)3/PPh3 and B(C6F5)3/PtBu3 and (b) B(C6F5)3/P(C6F5)3.

1.1.7 C−F Bond Activation

C−F bonds are among the most inert chemical bonds known and are typically only activated in the presence of transition metal species.77,78 A recent report by Alcarazo demonstrated the first instance of FLP-mediated C−F bond activation of 1-fluoropentane upon reaction of a carbon-

based nucleophile, hexaphenylcarbodiphosphorane, with B(C6F5)3, quantitatively yielding the corresponding alkylphosphonium fluoroborate salt (Scheme 1.18).79

Scheme 1.18 – FLP-mediated C−F bond activation.79

Subsequent investigations by Stephan et al., employing phosphorous and borane-based frustrated Lewis pairs, demonstrated analogous reactivity with alkyl fluorides yielding phosphonium fluoroborate salts (Scheme 1.19).77 Further studies illustrated the ability to activate

fluoroalkanes with B(C6F5)3 alone, with the production of alkane upon reaction with phosphonium hydridoborate salts or silane (Scheme 1.19).77

Scheme 1.19 – (a) Activation of fluoroalkanes employing P/B FLPs yielding phosphonium 77 fluoroborate salts and (b) reaction of fluoroalkane, B(C6F5)3 and [tBu3PH][HB(C6F5)3] or 77 Et3SiH to yield free alkanes.

1.1.8 Activation of Disulfides

In addition to the activation of H−H, C−O, C−H, B−H and C−F bonds,1 frustrated Lewis pairs have shown utility in the heterolytic cleavage of the S−S bonds of disulfides, a task which was previously reserved for transition metal catalysts.80

Stoichiometric reactions of neutral phosphino-boranes with disulfides result in the heterolytic cleavage of the S−S bond yielding a phosphonium thioborate zwitterion.80 Exposure of the

zwitterion to a donor molecule, such as PMe3 or THF, results in the facile regeneration of the disulfide with concomitant generation of a phosphino-borate species. This finding implies that

the activation of disulfides, much like the heterolytic cleavage of H2 employing phosphino- boranes, is reversible in nature (Scheme 1.20, a).

Scheme 1.20 – Activation of disulfides by: (a) intramolecular and (b) intermolecular FLPs.80

Similarly, intermolecular frustrated Lewis pairs composed of combinations of B(C6F5)3 and

PtBu3 readily activate alkyl- and aryl-disulfides yielding the anticipated phosphonium thioborate salts (Scheme 1.20, b). It is of note that reaction of the phosphine with disulfide fails to result in

desulfurization, despite prolonged reaction times. Similarly, B(C6F5)3 is found to react yielding an extremely weak donor-acceptor adduct which is only spectroscopically detectable below -30 °C. Based on these findings, activation of the S−S bond requires cooperative activation, employing both acidic and basic FLP components.80

In investigations carried out by the Alcarazo group, purely carbogenic frustrated Lewis pairs were developed for application in the field of frustrated Lewis pair chemistry. These systems employ an electron deficient allene whose Lewis acidity is extremely limited and experimentally 81 found to be in the range of B(OPh)3. Combinations of the allene with the bulky and basic N- heterocyclic carbene, 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene, resulted in no reaction; yet upon exposure to disulfide, yielded the product of heterolytic S−S bond cleavage (Scheme 1.21).81

Scheme 1.21 – Disulfide activation employing a strictly carbon-based FLP system.81

1.2 Scope of Thesis

The first section of this thesis addresses O−H and S−H activation by intermolecular frustrated Lewis pairs to yield phosphonium alkoxy- and thioborate salts. The syntheses of novel borate and boronate esters are described in addition to the application of these species and other perfluoroarylboranes in the heterolytic cleavage of dihydrogen in the presence of tertiary phosphines. Furthermore, a detailed study of the reaction of triarylboranes, chlorodiarylboranes, borinic esters, boronate esters and boronic acids with alkyl- and aryldiazomethanes is communicated in addition to reactivity studies involving these derivatized Lewis acids. Finally, the interaction of frustrated Lewis pairs with the greenhouse gases, carbon dioxide and nitrous oxide, is detailed and includes a comprehensive assessment of the Lewis acid exchange chemistry of both carbon dioxide and nitrous oxide FLP adducts.

All synthetic work and characterizations cited in this thesis were performed by the author with the exception of elemental analysis and X-ray diffraction experiments and some solutions. Additionally, small portions of sections 6.2.3, 6.2.6 and 7.2.2 were produced as a result of collaborative research with Dr. Edwin Otten.

Portions of the work presented herein have been discussed in the following publications:

Chapter 3: Neu, R. C.; Ouyang, E. Y.; Geier, S. J.; Zhao, X.; Ramos, A.; Stephan, D. W. Dalton Trans. 2010, 39, 4285.

Chapter 4: (a) Neu, R. C.; Stephan, D. W. Organometallics. 2012, 31, 46. (b) Neu, R. C.; Jiang, C.; Stephan, D. W. Chem. Sci. 2012, in preparation.

Chapter 5: (a) Peuser, I.; Neu, R. C.; Zhao, X.; Ullrich, M.; Schirmer, B.; Tannert, J. A.; Kehr, G.; Frölich, R.; Grimme. S.; Erker, G.; Stephan, D. W. Chem. –Eur. J. 2011, 17, 9640. (b) Neu, R. C.; Ménard, G.; Stephan, D. W. Dalton Trans. 2012, DOI: 10.1039/C2DT30206C.

Chapter 6: (a) Otten, E.; Neu, R. C.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 9918. (b) Neu, R. C.; Otten, E.; Stephan, D. W. Chem. Sci. 2011, 2, 170.

Chapter 7: (a) Neu, R. C.; Otten, E.; Stephan, D. W. Chem. Sci. 2011, 2, 170. (b) Neu, R. C. Otten, E.; Stephan, D. W. Angew. Chem. Int. Ed. 2009, 48, 9709.

Chapter 2 The Activation of O−H and S−H Bonds By Frustrated Lewis Pairs

2.1 Introduction

2.1.1 Methods for the Conversion of Secondary Alcohols to Ketones

The transformation of primary and secondary alcohols to aldehydes and ketones is arguably one of the most fundamental and important transformations in the field of organic synthesis.82,83 The resulting carbonyl-containing species are readily over-oxidized to yield a plethora of valuable organic compounds such as esters, amides, nitriles and carboxylic acids in the presence of nucleophiles such as alcohols, 1° amines, ammonia and water.84

Oxidation of primary alcohols can yield the associated aldehydes but also offer the possibility to over oxidize the starting alcohols to yield carboxylic acids.85-87 The mechanism of over oxidation is precluded when oxidizing secondary alcohols, halting the oxidative process at the ketone stage. Some of the most widely applied methodologies for the oxidation of secondary alcohols to the related ketones are the: Swern, Corey-Kim, Dess-Martin, Oppenauer and the Pfitzner-Moffatt oxidations.

One of the most common methodologies for the oxidation of alcohols to ketones is the Swern oxidation. This process involves the activation of dimethylsulfoxide (DMSO) employing oxalyl

chloride yielding an electrophilic chlorosulfonium species, carbon dioxide (CO2) and carbon monoxide (CO) (Scheme 2.1).88,89

Scheme 2.1 – Synthesis of a chlorosulfonium ion.89

The chlorosulfonium species undergoes nucleophilic attack by the secondary alcohol yielding an

alkoxysulfonium chloride intermediate. In the presence of NEt3, this species is decomposed

yielding the corresponding ketone with dimethyl sulfide (SMe2) and two equivalents of triethylammonium chloride as byproducts (Scheme 2.2).90 Although, the chlorosulfonium reagent is useful for the oxidation of a number of organic substrates, it is not without limitations. Notably, dimethylsulfoxide and oxalyl chloride react explosively at room temperature, therefore the reaction must be maintained below -60 °C. Decomposition of the sulfonium species is otherwise observed at approximately -20 °C.91 Additionally, the byproducts of reaction, namely

SMe2 and CO, are toxic in nature and require careful handling.

Scheme 2.2 – Secondary alcohol oxidation by a chlorosulfonium ion.

Numerous variations for the generation of “activated DMSO” have been developed such as the Pfitzner-Moffat92-94 (DMSO/dicyclohexylcarbodiimide), Albright-Goldman95,96 (DMSO/acetic anhydride), Onodera97 (DMSO/phosphorous pentoxide), Parikh-Doering98 (DSMO/pyridine-

99,100 101 sulfur trioxide), Corey-Kim (SMe2/N-chlorosuccinimide) and Liu (DMSO/phenyl dichlorophosphate) procedures which offer routes to ketones from secondary alcohols employing varying reaction conditions.

Oxidation can be achieved employing the Dess-Martin periodinane (DMP) which is derived from the well-known and insoluble hypervalent iodine reagent, iodoxybenzoic acid (IBX).102 DMP offers many added advantages such as the ability to work at room temperature and neutral pH, simple work ups, good functional group tolerance, high yields and good shelf life.103 Additionally, DMP is environmentally benign and is a mild oxidant as compared to other reagents.104 As IBX is insoluble, the improved solubility of the DMP is achieved by the incorporation of acetate groups which results in improved reactivity as compared to IBX. Oxidation of the alcohol readily occurs at the iodine centre yielding the desired ketone and an iodinane, as the final products of reaction (Scheme 2.3).

Scheme 2.3 –Dess-Martin oxidation.105

A simple and selective method for the oxidation of primary and secondary alcohols is readily

accomplished employing an aluminum-based Lewis acid, Al(OR)3 (R = tBu, iPr, Ph), in what is commonly referred to as the Oppenauer oxidation.106 This method involves arrangement of the components to yield a six membered transition state107 encompassing the Lewis acid, reductant and the oxidant. Hydride transfer from the sacrificial reductant to the oxidant yields the desired ketone and the alcohol byproduct (Scheme 2.4).108 This is of particular interest due to the inexpensive and non-toxic nature of the starting materials in addition to the functional group tolerance of the process (alkenes, alkynes, esters, amides, etc.).105 Additionally, the oxidation of

secondary alcohols is much faster as compared to primary alcohols, resulting in complete chemoselectivity.

Scheme 2.4 – Oppenauer oxidation.105

In a reaction related to the Swern oxidation, the chlorosulfonium cation can be generated in the presence of dimethylsulfide and N-chlorosuccinimide, subsequently yielding the active species, S,S-dimethylsuccinimidosulfonium chloride, also referred to as the Corey-Kim reagent.109-111 Subsequent reaction with an alcohol yields an alkoxysulfonium intermediate reminiscent of the Swern oxidation (Scheme 2.5). This method is mild and tolerant of a vast array of functionalities; however, SMe2 remains a product of the reaction.

The Pfitzner and Moffatt method also employs activated DMSO as a reagent for the oxidation of aldehydes and specifically ketones. This method employs readily available and inexpensive carbodiimides for the direct activation of DMSO prior to reaction with alcohol (Scheme 2.6).92-94 This method is tolerant toward a range of functional groups and is performed on both small and large scales, without the use of specialized equipment. Notable drawbacks are the production of dialkylurea as a byproduct in addition to excess carbodiimide, which are both challenging compounds to separate from the reaction medium.105

Scheme 2.5 – Corey-Kim oxidation.105

Scheme 2.6 – Pfitzner-Moffatt oxidation.105

Methods developed by Jones in the late 1940s, gave rise to the Jones oxidation which involves in

situ generation of chromic acid, from chromium trioxide or sodium dichromate in dilute H2SO4, which reacts with alcohol to yield a chromate ester. Subsequent reaction with base yields the desired carbonyl compound.112 This method, however, is limited to species which are tolerant to acidic media, otherwise the alternative pyridinium chlorochromate reagent can be employed.113 In recent years, this oxidative method has fallen out of use due to the difficulty associated with ketone separation from the resulting chromium salts, in addition to the dangers associated with the handling of these carcinogenic materials.103

The aforementioned examples illustrate the principle oxidative methodologies employed in current day organic synthesis. Improvements and substitutions to these procedures have been made to offer a range of organic media, modified catalysts and reaction conditions suitable for the oxidation of unique and specialized substrates.

2.2 Results and Discussion

2.2.1 O-H Bond Activation Employing Intermolecular FLPs

The previous section describes methods employed in the field of organic chemistry to transform simple secondary alcohols into ketones. These methodologies often involve complications such as the production of toxic byproducts, difficulty of product from byproduct separation, in addition to reaction temperature restrictions. Our efforts focused on the application of sterically bulky frustrated Lewis pairs for the activation of secondary alcohols and subsequent attempted thermal elimination of molecular hydrogen, to yield the desired ketones (Scheme 2.7).

Computational studies by Privalov et al. reveal that the hydrogenation of basic ketones is feasible in the presence of tris(pentafluorophenyl)borane and molecular hydrogen.114 Small

energy barriers for H2 activation by B and O were noted with larger ketones and was found to have larger hydride delivery energies due to steric demands of both the substrate and borohydride moieties. Recent experimental investigations conducted by Repo et al. demonstrate

the over-reduction of benzaldehyde to yield diphenylmethane with B(C6F5)3 and H2, while attempted dehydrogenation of diphenylmethanol with B(C6F5)3 and H2 yields product mixtures containing benzophenone and diphenylmethane.115

Scheme 2.7 – Proposed cycle for the conversion of alcohols to ketones.

The prototypical sterically demanding intermolecular FLPs composed of

tris(pentafluorophenyl)borane (B(C6F5)3) and tri-tert-butylphosphine (PtBu3) in addition to

B(C6F5)3 and tris(2,4,6-trimethylphenyl)phosphine (PMes3) were examined for their interaction

with alcohols. Reaction of B(C6F5)3 with PtBu3 and PMes3 yielded brightly coloured yellow and violet solutions which are the commonly observed colours for these species in solution.31,116 Addition of a toluene solution of the bulky alcohol, diphenylmethanol, to each of the above solutions resulted in nearly immediate solution discolouration, indicative of consumption of the FLPs (Scheme 2.8). Precipitation of a white solid was noted from the reaction mixture containing PtBu3 while a clear and colourless oil separated from the solution containing PMes3. Work-up of both solutions followed by isolation yielded white solids in 79 (2-1) and 94 % (2-2) yield.

Scheme 2.8 – Synthesis of 2-1 and 2-2.

Examination of the 1H NMR spectra of 2-1 and 2-2 revealed broad doublets at 4.96 and 8.25 ppm respectively, corresponding to a one bond P−H coupling with coupling constants of 427 and 478 Hz. The PH moiety was further supported by the presence of a doublet in the 31P NMR spectra for 2-1 and 2-2 at 61.70 and -26.29 ppm. Signals assignable to the tBu and Mes moieties were also located within the 1H NMR spectra. Additionally, resonances pertaining to the phenyl

and methine protons of the (C6H5)2CH fragment were observed spectroscopically. Examination of the 11B{1H} NMR spectra revealed a sharp singlet for both 2-1 and 2-2 at an average chemical shift of -2.98 ppm, which is consistent with the quaternization of the boron centre and is 117 19 comparable to the related salt [C10H6(NMe2)2H][(n-C18H37O)B(C6F5)3] (-2.4 ppm). The F

NMR spectra revealed chemically equivalent C6F5 groups at boron, as evidenced by three signals at -132.94, -164.70 and -167.95 ppm, integrating in a 2:1:2 fashion, consistent with the o-, p- and m-fluorine atoms. The sum of these data pointed toward the cleavage of the H−O bond of diphenylmethanol by the FLPs, yielding the corresponding salts [R3PH][(C6H5)2HCOB(C6F5)3] (R = tBu, 2-1; Mes, 2-2). The solid-state structures of both 2-1 and 2-2 were determined via X- ray crystallographic methods (Figure 2.1).

The solid-state data confirmed the pseudo-tetrahedral nature of both the boron and the phosphorous centres in 2-1 and 2-2. The anion in 2-1 was noted to bear one phenyl ring of the

(C6H5)2CH moiety at an angle 19.2° relative to the plane as defined by a neighbouring C6F5 ring. This interaction was indicative of a π-stacking interaction between an electron-rich and an electron-poor arene.118 Two longer C−F···H−P approaches of 2.56 Å were observed in 2-1

between two o-F atoms of two distinct C6F5 rings with the phosphonium P−H. This interaction was absent in the solid-state structure of 2-2.

A stoichiometric mixture of B(C6F5)3 and PtBu3 was prepared in as analogous fashion to the aforementioned reactions. A single equivalent of neat 2-propanol was added to the FLP solution and instantaneous discolouration of the solution was observed (Scheme 2.8). Manipulation with pentane afforded 2-3 as a white solid in 92 % yield. Examination of the 1H NMR spectrum 1 revealed a broad doublet with corresponding JH-P of 427 Hz, consistent with the presence of the tri-tert-butylphosphonium cation. The observance of a septet and doublet at 3.59 and 0.48 ppm, corresponding to the methine and methyl signals, confirmed the presence of the isopropyl 19 functionality. A narrowing of the p-m gap in the F NMR spectrum (∆δp-m = 3.20 ppm) confirmed the four-coordinate nature of the borate centre12, which was further supported by a

sharp signal in the 11B{1H} NMR spectrum, at -3.41 ppm. In combination with elemental

analysis, these data confirmed the nature of 2-3 as [tBu3PH][(CH3)2HCOB(C6F5)3].

Unfortunately, attempts to generate the analogous ion pair employing PMes3 yielded a waxy solid which was only ever isolable in 90 % purity.

Figure 2.1 – POV-Ray depictions of the molecular structures of 2-1 and 2-2. B: yellow-green, C: black, O: red, F: pink, P: orange. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). 2-1: B(1)-O(1), 1.458(6); O(1)-C(19), 1.238(7); B(1)-O(1)-C(19), 131.1(5). 2-2: B(1)-O(1), 1.451(7); O(1)-C(19), 1.417(6); B(1)-O(1)-C(19), 120.5(4).

Similar reaction of B(C6F5)3 with PR3 (R = tBu, Mes) with a stoichiometric amount of phenol

yielded the expected ion pairs [R3PH][(C6H5)OB(C6F5)3] (R = tBu, 2-4; R= Mes, 2-5) in 95 and 73 % yield (Scheme 2.9). The 31P NMR spectra revealed a doublet in each instance, with one

bond P−H couplings of 430 and 479 Hz, indicative of the HPtBu3 and HPMes3 moieties. 1 Aromatic protons were detected by H NMR spectroscopy confirming the O−C6H5 moiety in both 2-4 and 2-5. The anion was confirmed employing both 11B{1H} and 19 F NMR spectroscopy. A narrow and sharp singlet at an average chemical shift of -3.71 ppm was observed for 2-4 and 2-5 by 11B{1H} NMR in addition to an averaged p-m gap of 4.16 ppm, supporting the four-coordinate nature of the borate ion. An X-ray crystallographic study confirmed the formulation of 2-4 (Figure 2.2).

Scheme 2.9 – Synthesis of 2-4, 2-5 and 2-6.

In a related manner, a solution of B(C6F5)3 and PtBu3 was prepared to which a solution of pentafluorophenol was added yielding a white solid, 2-6, in 88 % yield (Scheme 2.9). The

spectroscopic data was found to be related to that of 2-4 and 2-5 (Figure 2.2). Notably, two C6F5 19 environments were observable by F NMR pertaining to the freely rotating C6F5 rings on B and

the O−C6F5 moiety. Upon quaternization of the boron centre, a marked shift of the para-F signal

of the B-bound C6F5 group was noted with an associated p-m gap narrowing of 4.85 ppm, as

compared to that of the free borane (Δδp-m = 18 ppm). This chemical shift difference is indicative of how changes in electron density at the boron centre affect the electronics of the associated perfluoroaryl rings.12 This distinctive narrowing of the p-m gap is illustrated in Figure 2.2. The

formulation of 2-6 was determined to be [tBu3PH][(C6F5)OB(C6F5)3] and was confirmed crystallographically (Figure 2.3).

Figure 2.2 – NMR spectra of B(C6F5)3 and 2-6.

Figure 2.3 - POV-Ray depictions of the molecular structures of 2-4 and 2-6. B: yellow-green, C: black, O: red, F: pink, P: orange. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). 2-4: B(1)-O(1), 1.511(5); C(19)-O(1), 1.343(4); B(1)-O(1)-C(19), 123.4(3). 2-6: B(1)-O(1), 1.511(2); C(19)-O(1), 1.325(2); B(1)-O(1)-C(19), 126.10(5).

It should be noted that the B−O bond lengths in 2-4 and 2-6 were found to be statistically identical, despite the difference in electronics between the two OR moieties. The B−O−C bond angle in 2-6 was noted to be larger as compared to that in 2-4, consistent with ability of the

O−C6F5 group to inductively withdraw electron density from the O-atom (Figure 2.4). The unit cell contents of 2-6 demonstrated a long P−H···F−C contact of 2.601 Å, indicative of H···F

bonding between the phosphonium PH and an ortho-F atom of a B-bound C6F5 group.

Figure 2.4 - Resonance structures of 2-6.

119 In a related fashion, the bisborane 1,4-(C6F4)[B(C6F5)]2 was reacted with two equivalents of

PtBu3 and phenol yielding a white solid, 2-7, which was isolated in quantitatively yield. Signals 1 31 pertaining to the tBu3PH species were observable by both H and P NMR spectroscopy, confirming the phosphonium moiety. Examination of the 19F NMR spectrum revealed four

signals, assignable to the C6F4 linker and the o-, p- and m-F signals of the C6F5 rings, signifying two-fold symmetry. These data supported the cleavage of the O−H bonds of two molecules of phenol employing both Lewis acidic sites and two equivalents of base, yielding

[tBu3PH]2[(C6F4)[(C6H5O)B(C6F5)2]2] (Scheme 2.10). Double activation, employing both Lewis acidic sites, has been previously documented for this species, having been used for the dual activation of two metallocene species yielding two metallocenium cations and a dianion.119

Scheme 2.10 – Synthesis of 2-7.

As has been documented in earlier FLP-based chemistry, the strength of a B−O bond is often a 1 limiting factor in further reactivity of the O-containing moiety. The Lewis acidity of B(C6F5)3 is such that release of an O-containing molecule following transformation will be challenging especially in the event that the substrate is both small and electron rich. In attempts to reduce the

propensity of the borane to hold onto the target ketones, the less Lewis acidic, yet bulky species

PhB(C6F5)2 was investigated for the ability to activate O−H bonds in the presence of phosphine.

Stoichiometric reactions of PhB(C6F5)2 and PtBu3 were prepared in pentane and neat 2-propanol or diphenylmethanol was added (Scheme 2.11). Immediate precipitation of a white solid, 2-8, was noted upon addition of diphenylmethanol whereas addition of 2-propanol yielded an oil which separated from solution and was triturated to a white solid, 2-9. Characteristic doublets 1 31 1 were observed in both the H and P NMR spectra for 2-8 and 2-9 with JP-H coupling constants of 428 Hz. Resonances in the aryl region of the 1H NMR spectrum were observed for 2-8, while methine and methyl signals integrating in a 1:6 fashion were observed for 2-9 confirming the

presence of the (C6H5)2CH and (CH3)2CH moieties. Quaternization of the boron centres was confirmed by sharp resonances in the 11B{1H} NMR spectra in addition to narrowing of the p-m gaps for both compounds. These data supported the formulations of 2-8 and 2-9 as

[tBu3PH][ROB(C6F5)2Ph] (R = (C6H5)2CH, 2-8; (CH3)2CH, 2-9). X-ray crystallography further confirmed the connectivity of 2-8 (Figure 2.5).

Scheme 2.11 – Synthesis of compounds 2-8, 2-9, 2-10 and 2-11.

Figure 2.5 – POV-Ray depiction of the molecular structure of 2-8. B: yellow-green, C: black, O: red, F: pink, P: orange. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). B(2)-O(2), 1.465(6); O(2)-C(50), 1.401(5); B(2)-O(2)-C(50), 122.5(3).

The crystallographic data revealed that two ion pairs were contained within the asymmetric unit. 120 Even with reduction in Lewis acidity upon substitution of B(C6F5)3 for PhB(C6F5)2, the B−O bond lengths remained unchanged and were determined to be statistically equivalent (2-8: 1.465(6) Å; 2-1: 1.458(6) Å). An electron poor-electron rich π-stacking interaction was noted between a phenyl ring of the [(C6H5)2O] fragment and a C6F5 group bound to boron. The planes

defined by the C6H5 and C6F5 rings were determined to be disposed at 26.08° relative to each other.

In an analogous fashion 1:1 reactions of PhB(C6F5)2 and PR3 (R = tBu and Mes) were prepared in pentane to which one equivalent of phenol was added to each combination, with nearly instantaneous precipitation of oils, 2-10 (triturated to a solid) and 2-11, in 99 and 83 % yield. 1H 31 1 and P NMR spectroscopy confirmed the phosphonium moieties with JP-H coupling constants of 429 and 478 Hz, respectively. 19F NMR p-m gap narrowing was observed upon quaternization of the boron centre as was previously observed for compounds 2-1 to 2-9 (Table 2.1). Together, these data supported the assignment of 2-10 and 2-11 as the ion pairs

[R3PH][(C6H5)OB(C6F5)Ph] (R = tBu, 2-10; Mes, 2-11).

Table 2.1 – 11B{1H}, 19F and 31P NMR data for compounds 2-1 through 2-11.

11 1 31 B{ H} (ppm) o-F (ppm) p-F (ppm) m-F (ppm) Δδp-m (ppm) P (ppm) 2-1 -3.01 -132.95 -164.70 -167.95 3.25 61.70 2-2 -2.95 -132.93 -164.70 -167.93 3.23 -26.29 2-3 -3.41 -133.50 -165.23 -168.43 3.20 60.63 2-4 -3.68 -134.80 -163.73 -167.90 4.17 60.07 2-5 -3.73 -134.92 -163.94 -168.09 4.15 -27.34 2-6 -1.87 -134.69 -161.94 -166.76 4.82 61.27 2-7 -3.53 -134.07 -164.39 -168.17 3.78 57.08 2-8 -1.14 -131.29 -165.93 -167.97 2.04 61.61 2-9 -1.56 -131.45 -166.12 -168.19 2.07 60.30 2-10 -1.62 -132.60 -164.62 -167.75 3.13 60.12 2-11 -1.67 -132.80 -164.84 -167.94 3.10 -27.35

2.2.4 S-H Bond Activation Employing Intermolecular FLPs

The activation of S−H bonds is a well-documented process employing transition metal and main group complexes.121-123 Subsequent transfer of SR fragments to unsaturated carbon-based species is the basis of much of the drug development industry, in addition to biologically active organosulfur molecules.124 In recent years, disulfides (RSSR) have been activated employing frustrated Lewis pairs yielding phosphonium thioborate salts.80 Analogously, the reactivity of

thiols with B(C6F5)3 and phosphines warranted investigation.

Stoichiometric reactions of B(C6F5)3 and PR3 (R = tBu and Mes) were prepared in toluene. Addition of a stoichiometric amount of 2-propanethiol resulted in instantaneous discolouration of the solutions, indicative of quenching of the Lewis acidity and basicity of the FLP components (Scheme 2.12). Addition of pentane yielded white solids, 2-12 and 2-13, in 95 and 90 % yield, respectively. Much like compounds 2-1 through 2-11, signals in the 1H and 31P NMR spectrum 1 pertaining to the phosphonium moieties were visible and bore JP-H coupling constants of 433 and 478 Hz, respectively. The borate anion was identical in both instances and bore an average 11B{1H} NMR chemical shift of -10.93 which is in keeping with the 11B{1H} chemical shift for 80 the product of disulfide activation, [tBu3PSiPr][iPrSB(C6F5)3] and the related ammonium

117 thioborate salt [C10H6(NMe2)2H][(n-C18H37S)B(C6F5)3]. Similarly, narrowing of the para- meta gaps were observed upon quaternization of the boron centres and resulted in an average 117 Δδp-m of 3.68 ppm (Δδp-m [C10H6(NMe2)2H][(n-C18H37S)B(C6F5)3]: 3.8 ppm ). The solid-state structures of both ion pairs were elucidated employing X-ray diffraction techniques and

supported the formulations as, [R3PH][(CH3)2CHSB(C6F5)3] (R = tBu, 2-12; Mes, 2-13) (Figure 2.6).

Figure 2.6 – POV-Ray depictions of the molecular structures of 2-12 and 2-13. B: yellow-green, C: black, F: pink, P: orange, S: yellow. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°).2-12: B(1)-S(1), 1.948(4); S(1)-C(19), 1.835(4); B(1)-S(1)-C(19), 105.32(19). 2-13: B(1)-S(1), 1.9798(19); S(1)-C(47), 1.8391(17); B(1)-S(1)-C(47), 109.97(8).

Scheme 2.12 – Synthesis of 2-12, 2-13 and 2-14.

The solid-state structure of 2-12 revealed that the phosphonium PH was orientated toward the thioborate anion, with a long P−H···S contact of 2.645 Å. Additionally, the methine proton of the thiolate fragment in 2-12, was orientated inward toward a B-bound C6F5 ring with a long C−H···F−C contact of 2.671 Å.

Stoichiometric reaction of B(C6F5)3, PMes3 and benzenethiol resulted in the isolation of an oil, 2-14, which was subsequently triturated into a solid and collected in 78 % yield. The NMR data was consistent with the related phosphonium alkylthiolate species 2-12 and 2-13 (Table 2.2). A broad doublet was observable by 1H NMR spectroscopy, indicative of deprotonation of the thiol fragment by the phosphine. A narrow para-meta gap in the 19F NMR spectrum, characteristic of a four-coordinate boron centre in addition to a sharp singlet in the 11B{1H} NMR spectrum 11 1 further supported the formation of the [(C6H5)SB(C6F5)3] ion. The B{ H} NMR chemical shift of the borate ion was noted to be downfield shifted by 1 ppm relative to 2-12 and 2-13, in keeping with the reduced electron rich nature of the [(C6H5)S] moiety as compared to the

[(CH3)2CHS] species. X-ray crystallography revealed the expected ion pair,

[Mes3PH][(C6H5)SB(C6F5)3] (Figure 2.7).

The solid-state structure of 2-14 revealed a stacking of the phenyl group of the thiolate fragment with an adjacent perfluorophenyl ring. These two aromatic fragments were found to be positioned at an angle of 29.7° relative to each other which is consistent with an electron poor- electron rich π-stacking interaction.118 Additionally, the P−H proton is orientated toward the sulfur atom of the anion, with a P−H···S−B contact of 2.888 Å. This contact distance is quite long, however, still falls within the sum of the Van der Waals radii (H: 1.20 Å; S: 1.80 Å).125

Table 2.2 - 11B{1H}, 19F and 31P NMR data for compounds 2-12 through 2-14.

11 1 31 B{ H} (ppm) o-F (ppm) p-F (ppm) m-F (ppm) Δδp-m (ppm) P (ppm) 2-12 -10.91 -131.47 -164.02 -167.68 3.66 59.19 2-13 -10.95 -131.50 -164.12 -167.82 3.70 -26.25 2-14 -9.90 -131.89 -163.64 -167.83 3.99 -26.27

Figure 2.7 – POV-Ray depictions of the molecular structure of 2-14. B: yellow-green, C: black, F: pink, P: orange, S: yellow. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). B(1)-S(1), 1.992(4); S(1)-C(46), 1.771(4); S(1)-B(1)-C(46). 106.08(18).

2.2.3 Mechanism of O−H and S−H Bond Cleavage

Reaction of B(C6F5)3, 1,4-(C6F4)[B(C6F5)2]2 or PhB(C6F5)2 with alcohol or thiol demonstrates some degree of weak adduct formation, resulting in the upfield shift of the 11B{1H} NMR signals. The activation by the borane drains electron density from the alcohol or thiol species rendering the X−H (X = O or S) H-atom more acidic. In the presence of base, such as phosphine, the alcohol or thiol is readily deprotonated yielding the associated phosphonium

alkoxy-, aryloxy- or thioborate ion pairs (Scheme 2.13). This is consistent with the mechanism

for the formation of alkoxy- and thioborate salts upon reaction of B(C6F5)3 with thiol or alcohol followed by deprotonation by 1,8-bis(dimethylamino)naphthalene (proton sponge).117

Scheme 2.13 – Activation of alcohols and thiols by intermolecular FLPs.

2.2.4 Reactivity of Phosphonium Alkoxyborate Ion Pairs

In attempts to induce the liberation of molecular hydrogen from the aforementioned phosphonium alkoxyborate salts, compounds 2-1 and 2-2 were heated between the temperatures of 100 and 150 °C in hopes of generating free benzophenone. Unfortunately, the phosphonium and borate moieties remained present as evidenced by 1H, 31P and 11B{1H} NMR spectroscopy.

It was thought that the hydrogen atom of the (C6H5)2CH moiety was too acidic in nature, due to the electron draining effects of the neighbouring phenyl groups, and unlikely to be driven off to + yield H2 in conjunction with the H from the phosphonium ion. Due to the donating nature of the isopropyl moiety in 2-3, this species was more likely to have a hydridic H atom resulting in

more facile H2 liberation. However, the decreased bulk of the iPr moiety as compared to

(C6H5)2CH would result in a stronger ketone-borane bond which could be difficult to cleave, thereby poisoning the borane. Subsequent attempts to thermally induce H2 liberation employing

2-3 also proved to be unsuccessful. Difficulty associated with H2 liberation is also a product of the highly electrophilic nature of the borane species which contributes to the reduced hydricity of the methine protons.

Alkoxide and proton transfer to an unsaturated acceptor molecule was attempted. Combinations of the phosphonium alkoxyborate salts were heated with N-benzylidene-tert-butylamine between

120 and 150 °C. In all cases, the phosphonium cation was found to protonate the imine indicative of the superior basicity of the imine relative to the phosphine. Unfortunately, the borate functionalities remained untouched, as evidenced by 11B{1H} NMR spectroscopy, indicative of the inability to induce OR transfer.

Scheme 2.14 – Reaction of [R3PH][R’OB(C6F5)2R”] with N-benzylidene-tert-butylamine, yielding the corresponding iminium alkoxyborates.

2.3 Conclusions

Combinations of sterically encumbered Lewis acids and bases can be employed to activate both O−H and S−H bonds of alcohols and thiols yielding phosphonium alkoxy- or aryloxyborate and

phosphonium thioborate salts. Subsequent attempts to thermolytically induce H2 release from these salts proved to be challenging and is likely due to the extremely electrophilic nature of the borane moiety. Further S−H and O−H bond activations employing bulky boranes with reduced electrophilicities are warranted. A balance of appropriate Lewis acidity and basicity should yield systems where alcohol conversion to ketone is achievable.

2.4 Experimental Section

2.4.1 General Considerations

All manipulations were carried out under an atmosphere of dry, O2-free N2 employing standard

Schlenk-line and glovebox techniques. Solvents (CH2Cl2, pentane and toluene) were dried by employing a Grubbs-type column system (Innovative Technology), degassed and stored under

N2. Bromobenzene was dried over CaH2 and distilled under N2 while cyclohexane was dried

over sodium and benzophenone and distilled under N2. CD2Cl2 and C6D5Br were dried over

CaH2 then vacuum transferred or distilled and subsequently stored under N2. B(C6F5)3 (Boulder

Scientific), PtBu3 (Strem Chemicals), PMes3 (Alfa Aesar) and phenol (Sigma-Aldrich) were used as received. Diphenylmethanol (Sigma-Aldrich) was purified by recrystallization from pentane at -35 °C, 2-propanol (Sigma-Aldrich) was dried over Na metal and distilled under N2, thiophenol (Sigma-Aldrich) and 2-propanethiol (Sigma-Aldrich) were dried over CaH2 and distilled under N2 while pentafluorophenol (Apollo Scientific) was vacuum transferred to yield 119 126 pure material. 1,4-(C6F4)[B(C6F5)2]2 and PhB(C6F5)2 were prepared according to literature procedures. 1H, 11B, 13C, 19F and 31P NMR spectra were recorded at 25 °C, unless otherwise stated, on a Varian NMR System 400 MHz or Bruker Avance III 400 MHz spectrometer and 1 13 were referenced using (residual) solvent resonances relative to SiMe4 ( H and C) or relative to 11 19 31 an external standard ( B: (Et2O)BF3, F: CFCl3, P: 85% H3PO4). Chemical shifts are reported in ppm and coupling constants as scalar values in Hz. Combustion analyses were performed in house employing a Perkin-Elmer CHN Analyzer.

2.4.2 Syntheses

Synthesis of 2-1. Compounds 2-1 and 2-2 were prepared employing the same synthetic methodology therefore only one preparation is detailed. A 20 mL scintillation vial was charged

with B(C6F5)3 (0.303 g, 0.59 mmol) and PtBu3 (0.120 g, 0.59 mmol) in toluene (5 mL). A solution of diphenylmethanol (0.110 g, 0.59 mmol) in toluene (2 mL) was added to the solution

of PtBu3 and B(C6F5)3. The reaction was stirred at room temperature for 12 hours. At this time, a solid was noted to have separated from the solution. Pentane (10 mL) was added to ensure complete product precipitation and a white solid was isolated by filtration, washed with pentane (3 x 5mL) and dried in vacuo. Crystals suitable for X-ray diffraction were grown from a layered 1 solution of CH2Cl2 and pentane at 25 °C. Yield: 0.421 g (79 %). H NMR (CD2Cl2): δ 7.23 (d, 1 2H, C6H5); 7.13 (t, 4H, C6H5); 7.01 (t, 4H, C6H5); 5.47 (s, 1H, CH{C6H5}2); 4.96 (d, 1H, JP-H = 3 11 1 13 1 427 Hz, PH); 1.60 (d, 27H, JH-P = 16 Hz, PtBu). B{ H} NMR (CD2Cl2): δ -3.01 (s). C{ H} 1 NMR (CD2Cl2): δ 149.11 (s, ipso-C6H5); 148.24 (br d, JC-F = 227 Hz, o-C6F5); 138.29 (br d, 1 1 JC-F = 235 Hz, p-C6F5); 136.33 (br d, JC-F = 245 Hz, m-C6F5); 127.32 (s, m-C6H5); 126.26 (s, p- 1 C6H5); 125.25 (s, o-C6H5); 79.05 (s, CH{C6H5}2); 37.81 (d, JP-C = 26 Hz, PC{CH3}3); 30.70 (s, 19 3 3 PC{CH3}3). F NMR (CD2Cl2): δ -132.95 (d, 6F, JF-F = 22 Hz, o-F); -164.70 (t, 3F, JF-F = 21

3 31 1 Hz, p-F); -167.95 (t, 6F, JF-F = 20 Hz, m-F). P NMR (CD2Cl2): δ 61.70 ppm (d, JP-H = 427 Hz,

PH). Anal. Calcd. for C44H41Cl2BF15OP: C, 53.74; H, 4.20. Found: C, 53.68; H, 4.26 %.

Synthesis of 2-2. B(C6F5)3 (0.200 g, 0.39 mmol), PMes3 (0.152 g, 0.39 mmol) and diphenylmethanol (0.072 g, 0.39 mmol). An oil precipitated from toluene and was triturated into a white solid with cold pentane. Crystals suitable for X-ray diffraction were grown from a layered solution of 1 1 CH2Cl2 and pentane at 25 °C. Yield: 0.400 g (94 %). H NMR (CD2Cl2): δ 8.25 (d, 1H, JP-H = 3 478 Hz, PH); 7.48 (d, 4H, JH-H = 8.4 Hz, C6H5); 7.17 (s, 3H, C6H2{CH3}3); 7.08 (s, 3H, 3 4 3 4 C6H2{CH3}3); 7.07 (td, 4H, JH-H = 7.6 Hz, JH-H = 2.0 Hz); 6.97 (tt, 3H, JH-H=7.2 Hz, JH-H = 1.6 11 1 Hz); 5.51 (s, 1H), 2.42 (s, 9H, p-CH3); 2.33 (s, 9H, o-CH3); 2.06 (s, 9H, o-CH3). B{ H} NMR 13 1 1 (CD2Cl2): δ -2.95 (s). C{ H} NMR (CD2Cl2): δ 149.13 (s, ipso-C6H5); 148.22 (br d, JC-F = 232 4 3 Hz, C6F5); 147.42 (d, JC-P = 2.5 Hz, p-C6H2{CH3}3); 144.42 (d, JC-P = 10 Hz, m-C6H2{CH3}3); 3 1 1 142.91 (d, JC-P = 9.5 Hz, m-C6H2{CH3}3); 138.31 (br d, JC-F = 241 Hz, C6F5); 136.53 (br d, JC-F 2 2 = 257 Hz, C6F5); 133.39 (d, JC-P = 11 Hz, o-C6H2{CH3}3); 131.96 (d, JC-P = 11 Hz, o- 1 C6H2{CH3}3); 127.29 (s, m-C6H5); 126.29 (s, p-C6H5); 125.21 (s, o-C6H5); 111.65 (d, JC-P = 83

Hz, ipso-C6H2{CH3}3); 79.05 (s, CH(C6H5)2); 22.06 (br s, o-CH3); 21.30 (s, p-CH3); 21.15 (s, o- 19 3 3 CH3). F NMR (CD2Cl2): δ -132.93 (d, 6F, JF-F = 22 Hz, o-F); -164.70 (t, 3F, JF-F = 20 Hz, p- 3 31 1 F); -167.93 (t, 6F, JF-F = 19 Hz, m-F). P NMR (CD2Cl2): δ -26.29 (d, JP-H = 478 Hz, PH).

Anal. Calcd. for C58H45BF15OP: C, 64.22; H, 4.18. Found: C, 64.09; H, 4.35 %.

Synthesis of 2-3. A 20 mL scintillation vial was charged with

B(C6F5)3 (0.200 g, 0.391 mmol), PtBu3 (0.079 g, 0.390 mmol) and toluene (5 mL). 2-propanol (30 μL, 0.390 mmol) was added dropwise to the yellow solution of the borane and phosphine. Upon addition, the yellow colour faded until the solution became clear and colourless. The reaction was allowed to stir for an additional hour. At this time, an oil was observed to have separated from solution and was further precipitated with the addition of pentane (10 mL). The supernatant was decanted and further cold pentane (10 mL) was added. The oil was triturated into a white solid with vigorous stirring. The supernatant was decanted once again and the solid was washed with pentane (2 x 5 1 1 mL) before being dried in vacuo. Yield 0.279 g (92 %). H NMR (CD2Cl2): δ 5.03 (br d, 1H, JH- 3 3 P = 427 Hz, HP); 3.59 (septet, 1H, JH-H = 5.8 Hz, CHMe2); 1.65 (d, 27H, JH-P = 16 Hz,

3 11 1 PC{CH3}3); 0.48 (d, 6 H, JH-H = 5.8 Hz, CH{CH3}). B{ H} NMR (CD2Cl2): δ -3.41 13 1 1 1 (s). C{ H} NMR (CD2Cl2): δ 148.61 (br d, JC-F = 238 Hz, o-C6F5); 138.58 (br d, JC-F = 242 1 Hz, p-C6F5); 136.91 (br d, JC-F = 243 Hz, m-C6F5); 126.59 (br s, ipso-C6F5); 65.61 (s, CH); 1 19 38.26 (d, JC-P = 27 Hz, PC{CH3}3); 30.51 (s, PC{CH3}3); 25.80 (s, CH{CH3}). F NMR 3 3 (CD2Cl2): δ -133.50 (d, 6F, JF-F = 22 Hz, o-C6F5); -165.23 (t, 3F, JF-F = 20 Hz, p-C6F5); -168.43 31 1 (m, 6F, m-C6F5). P NMR (CD2Cl2): δ 60.63 (d, JP-H = 427 Hz, PH). Anal. Calcd. for

C33H35BF15OP: C, 51.15; H, 4.56. Found: C, 50.82; H, 4.83 %.

Synthesis of 2-4. Compounds 2-4, 2-5 and 2-6 were prepared employing the same synthetic methodology therefore only one preparation is detailed. A 20 mL scintillation vial was charged

with B(C6F5)3 (0.200 g, 0.391 mmol), PtBu3 (0.079 g, 0.390 mmol) and toluene (7 mL). A second vial was charged with phenol (0.037 g, 0.393 mmol) and toluene (2 mL). The solution of the phenol was added to the yellow solution of the borane and phosphine in a dropwise fashion. Upon addition, the yellow colour was consumed and a clear oil separated from solution. With stirring the oil became a white solid. The reaction was stirred for a period of 12 hours. At this time the solid was allowed to settle and the supernatant was decanted. The solid was then 1 washed with pentane (2 x 5 mL) and dried in vacuo. Yield: 0.299 g (95 %). H NMR (CD2Cl2): 3 3 3 δ 6.97 (t, 2H, JH-H = 7.7 Hz, m-C6H5); 6.63 (t, 2H, JH-H = 7.9 Hz, o-C6H5); 6.57 (tt, 1H, JH-H = 4 1 3 7.5 Hz, JH-H = 1.1 Hz, p-C6H5); 5.00 (br d, 1H, JH-P = 430 Hz, HP); 1.60 (d, 27H, JH-P = 16 Hz, 11 1 13 1 PtBu3). B{ H} NMR (CD2Cl2): δ -3.68 (s). C{ H} NMR (CD2Cl2): δ 161.17 (s, ipso-C6H5); 1 1 1 148.53 (br d, JC-F = 240 Hz, o-C6F5); 129.12 (br d, JC-F = 244 Hz, p-C6F5); 137.02 (br d, JC-F =

244 Hz, m-C6F5); 128.81 (s, C6H5); 124.41 (br s, ipso-C6F5); 119.10 (s, C6H5); 117.51 (s, C6H5); 1 19 38.15 (d, JC-P = 27 Hz, PC{CH3}3); 30.46 (s, PC{CH3}3). F NMR (CD2Cl2): δ -134.80 (d, 6F, 3 3 31 JF-F = 24 Hz, o-C6F5); -163.73 (d, 3F, JF-F = 24 Hz, p-C6F5); -167.90 (m, 6F, m-C6F5). P 1 NMR (CD2Cl2): δ 60.07 (d, JP-H = 430 Hz, PH). Anal. Calcd. for C36H33BF15OP: C, 53.45; H, 4.11. Found: C, 53.57; H, 4.22 %.

Synthesis of 2-5. B(C6F5)3 (0.200 g, 0.391 mmol), PMes3 (0.152 g, 0.391 mmol), phenol (0.037 g, 0.393 mmol) and toluene (2 mL). A clear oil separated from solution and was 1 triturated into a white solid with pentane. Yield: 0.284 g (73 %). H NMR (CD2Cl2): δ 8.24 (br

1 3 d, 1H, JH-P = 479 Hz, HP); 7.17 (br s, 3H, C6H2Me3); 7.08 (br s, 3H, C6H2Me3); 6.95 (t, 2H, JH- 3 3 H = 7.3 Hz, m-C6H5); 6.62 (d, 2H, JH-H = 7.1 Hz, o-C6H5); 6.54 (t, 1H, JH-H = 7.1 Hz, p-C6H5); 11 1 2.38 (s, 9H, p-CH3); 2.29 (br s, 9H, o-CH3); 2.01 (br s, 9H, o-CH3). B{ H} NMR (CD2Cl2): δ - 13 1 1 3.73 (s). C{ H} NMR (CD2Cl2): δ 161.15 (s, ipso-C6H5); 148.55 (br d, JC-F = 243 Hz, o-C6F5); 4 3 147.76 (d, JC-F = 2.8 Hz, p-C6H2Me3); 144.74 (br d, JC-F = 10 Hz, m-C6H2Me3); 143.23 (br d, 3 1 1 JC-F = 10 Hz, p-C6H2Me3); 130.05 (br d, JC-F = 244 Hz, p-C6F5); 137.02 (br d, JC-F = 246 Hz, 2 2 p-C6F5); 133.72 (br d, JC-F = 12 Hz, o-C6H2Me3); 132.30 (br d, JC-F = 12 Hz, o-C6H2Me3); 1 128.89 (s, C6H5); 124.25 (s, ipso-C6F5); 118.89 (s, C6H5); 117.31 (s, C6H5); 111.77 (d, JC-P = 83 5 19 Hz, ipso-C6H2Me3); 21.71 (d, JC-P = 1.3 Hz, p-CH3); 21.63 (s, o-CH3); 21. 52 (s, o-CH3). F 3 3 NMR (CD2Cl2): δ -134.92 (d, 6F, JF-F = 23 Hz, o-C6F5); -163.94 (t, 3F, JF-F = 21 Hz, p-C6F5); - 31 1 168.09 (m, 6F, m-C6F5). P NMR (CD2Cl2): δ -27.34 (d, JP-H = 479 Hz, PH). Anal. Calcd. for

C51H39BF15OP: C, 61.55; H, 3.95. Found: C, 60.99; H, 4.09 %.

Synthesis of 2-6. B(C6F5)3 (0.253 g, 0.494 mmol), PtBu3 (0.100 g, 0.494 mmol) and phenol (0.091 g, 0.494 mmol). White solid. 1 1 Yield: 0.392 g (88 %). H NMR (CD2Cl2): δ 5.05 (br d, 1H, JH- 3 11 1 P = 428 Hz, HP); 1.66 (d, 27H, JH-H = 16 Hz, PtBu3). B{ H} NMR (CD2Cl2): δ -1.87 (s). 13 1 1 1 C{ H} NMR (CD2Cl2) partial: δ 148.73 (br d, JC-F = 239 Hz, C6F5); 141.95 (br d, JC-F = 247 1 1 Hz, C6F5); 139.49 (br d, JC-F = 243 Hz, C6F5); 138.34 (br d, JC-F = 243 Hz, C6F5); 137.09 (br d, 1 1 JC-F = 238 Hz, C6F5); 123.55 (br s, ipso-C6F5); 38.28 (d, JC-P = 27 Hz, PC{CH3}3); 30.49 (s, 19 3 3 PC{CH3}3). F NMR (CD2Cl2): δ -134.69 (d, 6F, JF-F = 23 Hz, o-BC6F5); -159.39 (dm, 2F, JF- 3 F = 23 Hz, o-OC6F5); -161.94 (t, 3F, JF-F = 20 Hz, p-BC6F5); -166.76 (m, 6F, m-BC6F5); -168.12 3 4 31 (m, 1F, p-OC6F5); -174.23 (tt, 2F, JF-F = 22 Hz, JF-F = 6.7 Hz, m-OC6F5). P NMR (CD2Cl2): δ 1 61.27 (d, JP-H = 428 Hz, PH). Anal. Calcd. for C36H28BF15OP: C, 48.13; H, 3.14. Found: C, 48.22; H, 3.34 %.

Synthesis of 2-7. A 20 mL scintillation vial was

charged with 1,4-(C6F4)[B(C6F5)2]2 (0.0.70 g, 0.084

mmol), PtBu3 (0.034 g, 0.168 mmol) and CH2Cl2 (10 mL). A second vial was charged with phenol (0.016

g, 0.170 mmol) and CH2Cl2 (3 mL) and was added to the suspension of the borane and phosphine. Upon addition of the phenol solution, the

suspension cleared to a clear and colourless solution. The resulting reaction was left stirring overnight. At this time, the solution was reduced until ~2 mL remained. Pentane (10 mL) was then added precipitating a white solid. The supernatant was decanted and the white solid was then washed with pentane (2 x 5 mL) and was dried in vacuo. Yield: 0.113 g (95 %). 1H NMR 3 3 (CD2Cl2): δ 6.93 (t, 2H, JH-H = 6.9 Hz, m-C6H5); 6.61 (d, 2H, JH-H = 6.9 Hz, o-C6H5); 6.51 (t, 3 1 3 1H, JH-H = 6.9 Hz, p-C6H5); 5.24 (br d, 1H, JH-P = 442 Hz, HP); 1.53 (d, 27H, JH-P = 16 Hz, 11 1 13 1 PtBu). B{ H} NMR (CD2Cl2): δ -3.53 (s). C{ H} NMR (CD2Cl2) partial: δ 161.54 (s, ipso- 1 1 C6H5); 148.65 (br d, JC-F = 237 Hz, o-C6F5); 148.04 (br d, JC-F = 241 Hz, o-C6F4); 138.85 (br d, 1 1 JC-F = 245 Hz, p-C6F5); 136.89 (br d, JC-F = 247 Hz, o-C6F5); 128.67 (s, m-C6H5); 125.41 (br s, 1 ipso-C6F5); 119.12 (s, o-C6F5); 117.05 (s, p-C6F5); 37.80 (d, JC-P = 27 Hz, PC{CH3}3); 30.27 (s, 19 3 PC{CH3}3). F NMR (CD2Cl2): δ -134.07 (d, 8F, JF-F = 24 Hz, o-C6F5); -137.46 (s, 4F, C6F4); - 3 31 164.39 (t, 4F, JF-F = 21 Hz, p-C6F5); -168.17 (m, 8F, m-C6F5). P NMR (CD2Cl2): δ 57.08 (d, 1 JP-H = 442 Hz, PH). Anal. Calcd. for C66H66B2F24O2P2: C, 55.37; H, 4.65. Found: C, 54.42; H, 4.82 %. Consistently low on carbon despite repeated combustion attempts, attributable to the formation of boron carbides.

Synthesis of 2-8. A scintillation vial was charged with

PhB(C6F5)2 (0.103 g, 0.20 mmol) and PtBu3 (0.041 g, 0.20 mmol) in toluene (5 mL). A solution of diphenylmethanol (0.037 g, 0.20 mmol) in toluene (2 mL) was added to the solution of borane and phosphine. The mixture was stirred at room temperature for 12 hours. Pentane was added, precipitating a white solid. The solid was isolated by filtration, washed with pentane (3 x 5mL) and dried in vacuo. 1 3 Yield: 0.134 g (74 %). H NMR (CD2Cl2): δ 7.66 (d, 2H, JH-H = 7.2 Hz, C6H5B(C6F5)2); 7.41 (d, 3 1 4H, JH-H = 7.5 Hz, (C6H5)2CHO); 7.06 (m, 6H, C6H5); 6.97 (m, 3H, C6H5); 4.93 (d, 1H, JP-H = 3 11 1 428 Hz, PH); 1.59 (d, 27H, JH-P = 16 Hz, PC{CH3}3). B{ H} NMR (CD2Cl2): δ -1.14 (s). 13 1 1 C{ H} NMR (CD2Cl2) partial: δ 149.98 (s, ipso-C6H5); 148.42 (br d, JC-F = 236 Hz, o-C6F5); 1 1 138.01 (br d, JC-F = 242 Hz, p-C6F5); 136.72 (br d, JC-F = 239 Hz, m-C6F5); 131.87 (s, C6H5);

127.72 (s, m-C6H5); 126.82 (s, C6H5); 126.57 (s, C6H5); 126.57 (s, p-C6H5); 125.58 (s, o-C6H5); 1 124.06 (s, C6H5); 78.62 (s, CH{C6H5}2); 38.18 (d, JP-C = 26 Hz, PC{CH3}3); 30.51 (s, 19 3 3 PC{CH3}3). F NMR (CD2Cl2): δ -131.29 (dd, 4F, JF-F = 25 Hz, JF-F = 7.0 Hz, o-C6F5); -165.93 3 3 31 (t, 2F, JF-F = 20 Hz, p-C6F5); -167.97 (t, 4F, JF-F = 20 Hz, m-C6F5). P NMR (CD2Cl2): δ 61.61

1 (d, JP-H = 427 Hz, PH). Anal. Calcd. for C43H44BF10OP: C, 63.87; H 5.48. Found: C, 63.68; H, 5.52 %.

Synthesis of 2-9. A 20 mL scintillation vial was charged with

PhB(C6F5)2 (0.150 g, 0.355 mmol), PtBu3 (0.072 g, 0.356 mmol) and toluene (5 mL). Neat 2-propanol (27 μL, 0.353 mmol) was added to the solution of borane and phosphine and the resulting reaction mixture became instantaneously cloudy. The solution was allowed to stir for an additional hour over which time a clear and colourless oil separated from solution. Pentane (10 mL) was added and with vigorous stirring the oil was triturated into a solid. The supernatant was then decanted and the product was washed with pentane (2 x 5 mL) then dried in vacuo. Yield: 0.210 g (87 %). 1H 3 3 NMR (CD2Cl2): δ 7.57 (d, 2H, JH-H = 7.4 Hz, o-C6H5); 7.05 (t, 2H, JH-H = 7.3 Hz, m-C6H5); 3 4 1 6.91 (d, 1H, JH-H = 7.4 Hz, JH-H = 1.4 Hz, p-C6H5); 4.98 (br d, JH-P = 428, PH); 3.67 (sept, 1H, 3 3 3 JH-H = 5.9 Hz, CHMe2); 1.61 (d, 27H, JH-P = 16 Hz, PC{CH3}3); 0.92 (d, 6H, JH-H = 5.9 Hz, 11 1 13 1 1 CH{CH3}2). B{ H} NMR (CD2Cl2): δ -1.56 (s). C{ H} NMR (CD2Cl2): δ 148.50 (br d, JC-F 1 1 = 235 Hz, o-C6F5); 138.10 (br d, JC-F = 242 Hz, p-C6F5); 136.92 (br d, JC-F = 246 Hz, o-C6F5);

135.58 (s, C6H5); 131.78 (s, C6H5); 126.64 (s, C6H5); 123.84 (s, C6H5); 64.97 (s, CHMe2); 38.17 1 19 (d, JC-P = 27 Hz, PC{CH3}3); 30.50 (s, PC{CH3}3); 26.22 (s, CH{CH3}2). F NMR (CD2Cl2): δ 3 4 3 -131.45 (d, 4F, JF-F = 26 Hz, JF-F = 8.4 Hz, o-C6F5); -166.12 (t, 2F, JF-F = 21 Hz, p-C6F5); - 31 1 168.19 (m, 4F, m-C6F5). P NMR (CD2Cl2): δ60.30 (d, JP-H = 428 Hz, PH). Anal. Calcd. for

C33H40BF10OP: C, 57.87; H, 5.89. Found: C, 57.84; H, 5.60 %.

Synthesis of 2-10. Compounds 2-10 and 2-11 were prepared in an analogous fashion, therefore only one preparation is detailed. A 20 mL scintillation vial was

charged with PhB(C6F5)2 (0.150 g, 0.355 mmol), PtBu3 (0.072 g, 0.356 mmol) and toluene ( 5 mL). A solution of phenol (0.033 g, 0.351 mmol) in toluene (3 mL) was added to the colourless solution of borane and phosphine and the reaction mixture instantly became cloudy. The reaction was allowed to stir for an hour during which time a clear and colourless oil separated from solution. Addition of pentane (10 mL) and vigorous stirring produced a white solid. The solid was allowed to settle and the supernatant was decanted. The product was washed with 1 pentane (2 x 5 mL) and dried in vacuo. Yield: 0.249 g (99 %). H NMR (CD2Cl2): δ 7.54 (d,

3 3 3 2H, JH-H = 7.4 Hz, C6H5); 7.10 (t, 2H, JH-H = 7.3 Hz, C6H5); 6.99 (t, 3H, JH-H = 7.4 Hz, C6H5); 3 3 4 6.75 (d, 2H, JH-H = 7.7 Hz, C6H5); 6.55 (tt, 1H, JH-H = 7.2 Hz, JH-H = 1.2 Hz, C6H5); 4.94 (br d, 1 3 11 1 1H, JH-P = 429 Hz, PH); 1.59 (d, 27H, JH-P = 16 Hz, PC{CH3}3). B{ H} NMR (CD2Cl2): δ - 13 1 1 1.62 (s). C{ H} NMR (CD2Cl2): δ 162.40 (s, C6H5); 148.39 (br d, JC-F = 235 Hz, o-C6F5); 1 1 138.60 (br d, JC-F = 244 Hz, p-C6F5); 137.05 (br d, JC-F = 244 Hz, m-C6F5); 132.04 (s, C6H5);

128.89 (s, C6H5); 126.84 (s, C6H5); 124.40 (s, C6H5); 118.94 (s, C6H5); 116.82 (s, C6H5); 38.13 1 19 3 (d, JC-P = 27 Hz, PC{CH3}3); 30.48 (s, PC{CH3}3). F NMR (CD2Cl2): δ -132.60 (dd, 4F, JF-F 4 3 = 25 Hz, JF-F = 9.2 Hz, o-C6F5); -164.62 (t, 2F, JF-F = 20 Hz, p-C6F5); -167.75 (m, 4F, m-C6F5). 31 1 P NMR (CD2Cl2): δ 60.12 (d, JP-H = 429 Hz, PH). Anal. Calcd. for C36H38BF10OP: C, 60.15; H, 5.33. Found: C, 59.72; H, 5.19 %.

Synthesis of 2-11. PhB(C6F5)2 (0.150 g, 0.355 mmol), PMes3 (0.138 g, 0.355 mmol) and phenol (0.033 g, 0.351 mmol). 1 Isolated as an oil. Yield: 0.264 g (83 %). H NMR (CD2Cl2): 1 3 δ 8.83 (br d, 1H, JH-P = 478 Hz, PH); 7.53 (d, 2H, JH-H = 7.6 Hz, C6H5); 7.16 (br s, 3H, m- 3 3 C6H2Me3); 7.07 (br s, 3H, m-C6H2Me3); 7.07 (t, 2H, JH-H = 7.4 Hz, C6H5); 6.96 (t, 3H, JH-H = 3 3 4 7.4 Hz, C6H5); 6.73 (d, 2H, JH-H = 7.9 Hz, C6H5); 6.52 (tt, 1H, JH-H = 7.2 Hz, JH-H = 1.1 Hz, 11 1 C6H5); 2.39 (s, 9H, p-CH3); 2.28 (br s, 9H, o-CH3); 2.01 (br s, 9H, o-CH3). B{ H} NMR 13 1 1 (CD2Cl2): δ -1.67 (s). C{ H} NMR (CD2Cl2): δ 162.34 (s, ipso-C6H5); 148.40 (br d, JC-F = 237 4 3 Hz, o-C6F5); 147.73 (d, JC-F = 3.0 Hz, p-C6H2Me3); 144.71 (br d, JC-F = 10 Hz, m-C6H2Me3); 3 1 1 143.22 (br d, JC-F = 10 Hz, p-C6H2Me3); 138.57 (br d, JC-F = 244 Hz, p-C6F5); 137.04 (br d, JC- 2 2 F = 244 Hz, m-C6F5); 133.71 (br d, JC-F = 12 Hz, o-C6H2Me3); 132.31 (br d, JC-F = 12 Hz, o-

C6H2Me3); 132.00 (s, C6H5); 128.81 (s, C6H5); 126.77 (s, C6H5); 124.32 (s, C6H5); 118.81 (s, 1 C6H5); 116.69 (s, C6H5); 111.95 (d, JC-P = 84 Hz, ipso-C6H2Me3); 22.49 (br s, o-CH3); 21.74 (d, 5 19 3 JC-P = 1.4 Hz, p-CH3); 21.59 (br s, o-CH3). F NMR (CD2Cl2): δ -132.80 (dd, 4F, JF-F = 25 Hz, 4 3 31 JF-F = 8.9 Hz, o-C6F5); -164.84 (t, 2F, JF-F = 21 Hz, p-C6F5); -167.94 (m, 4F, m-C6F5). P 1 NMR (CD2Cl2): δ -27.35 (d, JP-H = 478 Hz, PH). Anal. Calcd. for C51H44BF10OP: C, 67.68; H, 4.90. Found: C, 67.38; H, 5.14 %.

Synthesis of 2-12. Compounds 2-12 and 2-13 were prepared in an analogous fashion and therefore only one preparation is detailed. A

solution of B(C6F5)3 (0.154g, 0.30 mmol) and PtBu3 (0.061 g, 0.30

mmol) in toluene (5 mL) was prepared. The solution was stirred at room temperature for 1h. 2- propanethiol (28 μL, 0.30 mmol) was added via syringe. The solution was stirred at room temperature for 12 hours. Pentane was added, precipitating a white solid. The solid was isolated by filtration, washed with pentane (3 x 5mL) and dried in vacuo. Yield: 0.226 g (95 %). Crystals 1 suitable for X-ray diffraction were grown from a layered CH2Cl2/pentane solution at 25°C. H 1 NMR (CD2Cl2): δ 5.29 (d, 1H, JH-P = 433 Hz, PH); 2.12 (br m, 1H, Me2(CH)S); 1.63 (d, 27H, 3 3 11 1 JH-P = 16 Hz, PtBu3); 0.98 (d, 6H, JH-H = 7 Hz, CH3). B{ H} NMR (CD2Cl2): δ -10.91 (s). 13 1 1 1 C{ H} NMR (CD2Cl2) partial: 148.78 (br d, JC-F = 239 Hz, o-C6F5); 148.26 (br d, JC-F = 237 1 1 Hz, p-C6F5); 137.32 (br d, JC-F = 246 Hz, m-C6F5); 38.21 (d, JP-C = 27 Hz, PC{CH3}3); 33.72 (s, 19 Me2(CH)S); 30.53 (s, C(CH3)3); 27.32 (s, (CH3)2(CH)S) . F NMR (CD2Cl2): δ -131.47 (d, 6F, 3 3 3 JF-F = 22 Hz, o-C6F5), -164.02 (t, 3F, JF-F = 20 Hz, p-C6F5), -167.68 (t, 6F, JF-F = 20 Hz, m- 31 1 C6F5). P NMR (CD2Cl2): δ 59.19 (d, JP-H = 435 Hz, PH). Anal. Calcd. for C33H35BF15PS: C, 50.14; H, 4.46. Found: C, 50.10; H, 4.53 %.

Synthesis of 2-13. B(C6F5)3 (0.116, 0.23 mmol), PMes3 (0.088 g, 0.23 mmol) and 2-propanethiol (21 μL, 0.23 mmol). Yield: 0.200 g (90 %). Crystals suitable for X-ray diffraction were grown from 1 1 a layered CH2Cl2/pentane solution at -35°C. H NMR (CD2Cl2): δ 8.25 (d, 1H, JH-P = 479 Hz,

PH); 7.16 (s, 3H, C6H2(CH3)3); 7.08 (s, 3H, C6H2(CH3)3); 2.38 (s, 9H, p-CH3); 2.29 (s, 9H, o- 3 11 1 CH3); 2.13 (br m, 1H, Me2(CH)S); 2.01 (s, 9H, o-CH3); 0.98 (d, 6H, JH-H=7 Hz, CH3). B{ H} 13 1 1 NMR (CD2Cl2): δ -10.95 (s). C{ H} NMR (CD2Cl2): δ 148.21 (br d, JC-F = 237 Hz, o-C6F5) - 4 3 147.78 ( d, JC-P = 2.7 Hz, p-C6H2(CH3)3); 144.83 (d, JC-P = 11 Hz, m-C6H2(CH3)3); 143.33 (d, 3 1 1 JC-P = 10 Hz, m-C6H2(CH3)3); 140.00 (br d, JC-F = 260 Hz, p-C6F5); 137.26 (br d, JC-F = 258 Hz, 2 2 m-C6F5);133.84 (d, JC-P = 12 Hz, o-C6H2(CH3)3); 132.40 (d, JC-P = 11 Hz, o-C6H2(CH3)3); 1 111.97 ( d, JC-P = 86 Hz, ipso-C6H2(CH3)3); 33.33 (s, Me2(CH)S); 27.27 (s, (CH3)2(CH)S); 22.73 19 3 (br s, o-CH3); 21.73 (s, p-CH3); 21.71 (br s, o-CH3). F NMR (CD2Cl2): δ -131.50 (d, 6F, JF-F = 3 3 31 21 Hz, o-F); -164.12 (t, 3F, JF-F = 20 Hz, p-F); -167.82 (t, 6F, JF-F = 19 Hz, m-F); P NMR 1 (CD2Cl2): δ -26.25 (d, JP-H = 478 Hz, PH). Anal. Calcd. for C48H41BF15PS: C, 59.02; H, 4.23. Found: C, 58.50; H, 4.27 %.

Synthesis of 2-14. A 50 mL Schlenk flask was charged with a

solution of B(C6F5)3 (0.200 g, 0.39 mmol) and PMes3 (0.152 g,

0.39 mmol) in toluene (5 mL). Thiophenol (40 μL, 0.39 mmol) was then added via syringe. The solution was stirred at room temperature for 12 hours. The solvent was removed in vacuo and pentane was added, precipitating an oil. The oil was taken up in CH2Cl2 (2 mL) and filtered through a plug of Celite. Trituration with pentane (10 mL) resulted in a white solid. The solid was isolated by filtration, washed with pentane (3 x 5 mL) and dried in vacuo. Yield: 0.310 g (78

%). Crystals suitable for X-ray diffraction were grown from a layered solution of CH2Cl2 and 1 1 pentane at 25°C. H NMR (CD2Cl2): δ 8.25 (d, 1H, JP-H = 479 Hz, PH); 7.16 (s, 3H,

C6H2(CH3)3); 7.06 (s, 3H, C6H2(CH3)3); 7.03 (m, 2H, C6H5); 6.84 (m, 3H, C6H5); 2.38 (s, 9H, p- 11 1 CH3); 2.29 (s, 9H, o-CH3); 2.01 (s, 9H, o-CH3). B{ H} NMR (CD2Cl2): δ -9.90 ppm (s). 13 1 1 4 C{ H} NMR (CD2Cl2): δ 148.22 (br d, JC-F = 235 Hz, o-C6F5); 147.25 (d, JC-P = 3.1 Hz, p- 3 3 C6H2(CH3)3); 144.36 (d, JC-P = 9.4 Hz, m-C6H2(CH3)3); 142.93 (d, JC-P = 10 Hz, m- 1 1 C6H2(CH3)3); 138.48 (br d, JC-F = 237 Hz, p-C6F5); 136.70 (br d, JC-F = 240 Hz, m-C6F5); 2 2 133.35 (d, JC-P = 11 Hz, o-C6H2(CH3)3); 132.91 (s, ipso-C6H5); 131.94 (d, JC-P = 11 Hz, o- 1 C6H2(CH3)3); 127.24 (s, o/m-C6H5); 123.45 (s, p- C6H5); 111.62 (d, JC-P = 83 Hz; ipso- 19 C6H2(CH3)3); 22.07 (br s, o-CH3); 21.32 (s, p-CH3); 21.14 (s, o-CH3). F NMR (CD2Cl2): δ - 3 3 3 131.89 (d, 6F, JF-F = 21 Hz, o-C6F5); -163.84 (t, 3F, JF-F = 19 Hz, p-C6F5); -167.83 (t, 6F, JF-F = 31 1 20 Hz, m-C6F5). P NMR (CD2Cl2): δ -26.27 (d, JP-H = 479 Hz, PH). Anal. Calcd. for

C51H39BF15PS: C, 60.57; H, 3.89. Found: C, 60.38; H, 4.04 %.

2.4.3 X-ray Crystallography

2.4.3.1 X-ray Data Collection and Reduction

In preparation for analysis, crystals were first coated with Paratone-N oil in a glovebox and were

subsequently mounted on a MiTegen Micromount and placed in a N2 stream in order to maintain a dry and oxygen-free sample environment. Data collection was performed on a Bruker Apex II diffractometer and data collection strategies were determined employing Bruker provided Apex software. Optimization was performed in order to yield >99.5 % complete data to a minimum 2θ value of 55 °. All data sets were collected at 150(±2) K unless otherwise noted. The acquired frames were integrated employing the Bruker SAINT software package employing a narrow frame algorithm. Absorption corrections were conducted employing the empirical multi-scan method (SADABS).

2.4.3.2 Structure and Refinement

Non-hydrogen atomic scattering factors were taken from literature tabulations.127 Direct methods were employed to elucidate the positions of the heavy metals using the SHELXTL direct methods procedure. Remaining non-hydrogen atoms were subsequently found from successive difference Fourier map calculations. All cycles of refinement were carried out 2 employing full-matrix least squares techniques on F, minimizing the function ω (Fo-Fc) where 2 2 weight (ω) equates to 4Fo /2σ (Fo ) and Fo and Fc are equal to the observed and the calculated structure factor amplitudes. In the final cycles of each refinement, all non-hydrogen atoms were assigned anisotropic temperature factors in the absence of disorder or insufficient data. In the instance of the latter case the atoms were then treated isotropically. Unless noted, the C-H atom positions were calculated and allowed to ride on the carbon to which they are bonded with an assumed C-H bond length of 0.95 Å. Temperature factors pertaining to the H-atoms were fixed at 1.20 times the isotropic temperature factor of the C-atom to which they are bound. The H- atom contributions were calculated however never refined. The locations of the largest peaks in the final difference Fourier map calculations as well as the magnitude of the residual electron densities in each case were of no chemical significance.

2.4.3.3 Tables of Crystallographic Data

Table 2.3 – Selected crystallographic data for 2-1, 2-2 and 2-4.

2-1 2-2 2-4

Formula C44H41BCl2F15OP C58H45BF15OP C36H33BF15OP Formula wt 983.45 1084.72 808.40 Crystal system Triclinic Monoclinic Monoclinic

Space group P-1 Cc P21/n a (Å) 11.472(2) 23.385(5) 12.0085(15) b (Å) 13.125(3) 12.356(3) 14.8236(15) c (Å) 14.937(3) 20.293(4) 20.397(2) α (deg) 102.93(3) 90 90 β (deg) 95.93(3) 119.70(3) 99.966(4) γ (deg) 90.19(3) 90 90 V (Å3) 2179.6(8) 5093.3(18) 3576.0(7) Z 2 4 4 T (K) 293(2) 293(2) 150(2) d (calc) gcm-3 1.499 1.415 1.502 Abs coeff, μ, mm-1 0.285 0.151 0.185 Data collected 18251 25358 24664 R int 0.0505 0.0732 0.0480 # of indpndt reflns 7638 7909 6297

Reflns Fo≥2.0σ(Fo) 4854 6309 5140 Variables 590 698 500 R (>2σ) 0.0647 0.0700 0.0660

wR2 0.1674 0.1951 0.1937 Goodness of fit 1.017 1.032 1.020

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

Table 2.4 – Selected crystallographic data for 2-6, 2-8 and 2-12.

2-6 2-8 2-12

Formula C36H43BF20OP C86H88B2F20O2P2 C33H35BF15PS Formula wt 913.48 1617.12 790.45 Crystal system Monoclinic Monoclinic Orthorhombic

Space group P21/n P21/c Pbca a (Å) 12.5204(5) 18.5451(10) 18.814(4) b (Å) 15.1275(6) 23.8710(12) 17.177(3) c (Å) 19.9402(8) 17.9898(8) 21.481(4) α (deg) 90 90 90 β (deg) 104.064(2) 96.947(2) 90 γ (deg) 90 90 90 V (Å3) 3663.5(3) 7905.4(7) 6942(2) Z 4 4 8 T (K) 150(2) 296(2) 293(2) d (calc) gcm-3 1.656 1.359 1.513 Abs coeff, μ, mm-1 0.210 0.150 0.245 Data collected 39535 43347 49197 R int 0.0391 0.0413 0.0723 # of indpndt reflns 10013 11310 6110

Reflns Fo≥2.0σ(Fo) 6905 8337 4101 Variables 545 1035 475 R (>2σ) 0.0494 0.0755 0.0578

wR2 0.1399 0.2364 0.1740 Goodness of fit 1.031 1.023 1.026

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

Table 2.5 – Selected crystallographic data for 2-13 and 2-14.

2-13 2-14

Formula C48H41BF15PS C51H39BF15PS Formula wt 976.65 1010.66 Crystal system Triclinic Triclinic Space group P-1 P-1 a (Å) 11.3586(8) 13.582(3) b (Å) 11.3584(7) 13.798(3) c (Å) 17.5301(12) 13.802(3) α (deg) 86.154(3) 80.63(3) β (deg) 80.089(3) 70.07(3) γ (deg) 87.529(3) 73.51(3) V (Å3) 2221.7(3) 2325.4(8) Z 2 2 T (K) 296(2) 293(2) d (calc) gcm-3 1.460 1.443 Abs coeff, μ, mm-1 0.207 0.201 Data collected 27745 24544 R int 0.0234 0.0775 # of indpndt reflns 7759 10460

Reflns Fo≥2.0σ(Fo) 6709 5519 Variables 610 635 R (>2σ) 0.0329 0.0753

wR2 0.0898 0.2365 Goodness of fit 1.018 1.078

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

Chapter 3 Synthesis of Borate Esters, Boronate Esters and Triarylboranes for Application in the Activation of H2

3.1 Introduction

3.1.1 H2 Activation by Main Group Species

The hydrogenation of unsaturated species is arguably one of the most important chemical transformations in organic chemistry. However, typically the catalytic activation of molecular dihydrogen has been the domain of transition metal-based catalysts128 where examples of metal- free systems capable of H2 activation are rare. Organocatalysis employing main group species is well-documented in the literature but depends on hydrogen sources such as Hantzsch’s ester and its related derivatives.129 Additionally, hydrogenations are readily achieved employing the main 130 group reducing agents NaBH4 and LiAlH4, but are unfortunately only stoichiometric in nature. In recent years, research efforts have addressed low molecular weight main group compounds, such as ammonia borane, that have the ability to store H2 for application in the field of energy storage.131-134 Notably, one of the major limitations to these main group hydride sources is that they cannot be readily generated from molecular hydrogen and typically involve syntheses that 135 are energetically intensive. Few main group species to date have the ability to activate H2 directly and are limited to systems of: benzophenone and potassium tert-butoxide136 (Scheme 3.1, a), difluorovinylidene generated in an argon matrix137,138 (Scheme 3.1, b) and acyclic alkyl amino carbenes (Scheme 3.1, c).139

136 Scheme 3.1 – Main group species capable of H2 activation. (a) benzophenone/KOtBu , (b) difluorovinylidene137,138 and (c) acyclic alkyl amino carbenes.139

Recent advances have demonstrated the application of frustrated Lewis pairs for the facile and high yielding synthesis of main group hydrides. Initial studies involved the reaction of the large

perfluoroarylborane, tris(pentafluorophenyl)borane in conjunction with HPMes2 yielding a 24,25 phosphonium hydridoborate zwitterion capable of reversible H2 activation. This initial report was subsequently followed by similar examples of intramolecular FLPs by the Erker26 and Rieger and Repo groups (Scheme 3.2).27

Shortly after, intermolecular FLPs composed of bulky perfluoroarylboranes and tertiary phosphines, were elucidated. Subsequent exposure of these frustrated Lewis pairs to molecular hydrogen resulted in the heterolytic activation of the substrate yielding thermally robust phosphonium hydridoborate salts (Scheme 3.3, a).31

Scheme 3.2 – Intramolecular frustrated Lewis pairs capable of reversible and irreversible H2 activation. (a) Stephan24 (b) Erker26 (c) Rieger and Repo.27

Investigations by Erker et al. resulted in the discovery of an FLP resulting from the interaction of

1,8-bis(diphenylphosphino)naphthalene and B(C6F5)3 which heterolytically activates H2 yielding

the corresponding salt. This ion pair liberates H2 at 60 °C constituting the first example of 140 reversible H2 activation employing an intermolecular main group system (Scheme 3.3, b).

A further example was reported by Stephan et al. where tris(2,3,5,6-tetrafluorophenyl)borane, in

conjunction with tri(ortho-tolyl)phosphine, yielded the expected product of heterolytic H2

cleavage. Subsequent H2 liberation was achieved under vacuum over 9 days (85 % completion) at room temperature.141,142 These examples demonstrate the ability to finely tune both the steric and electronic properties of the Lewis acidic and basic moieties in order to achieve the desired reactivity (Scheme 3.3, c).

Numerous examples of the application of intermolecular frustrated Lewis pairs for the activation of H2 have been published since these seminal discoveries in the late 2000’s. Such examples 143,144 45 involve H2 activation employing a variety of bases such as carbenes, imines, pyridines,145-147 amines28,45,46,148 in addition to a variety of simple and specialized

phosphines.1,149,150 Lesser attention, however, has focused on determining the limits of Lewis acidity of the acidic components in this chemistry.1,3,141,142,149,151-153

31 140 Scheme 3.3 – Reversible and irreversible H2 activations by (a) Stephan, (b) Erker and (c) Stephan.141,142

3.2 Results and Discussion

3.2.1 Synthesis of Borate Esters

As previously described, the salts [R3PH][HB(C6F5)3], derived from the heterolytic activation of

H2, were found to be thermally stable and resistant to H2 liberation. This is likely in part due to

the highly electrophilic nature of the B(C6F5)3 moiety and its propensity to hold onto the hydride. Derivatization of the borane to maintain steric bulk while gaining control of the Lewis acidity of the boron centre has been the focus of investigations by Britovsek.154 These studies examined boranes of the form (C6F5)3-nB(OR)n and showed that they are highly electrophilic boranes which allow for control of the pπ-pπ interaction between the B and O atoms by variation of the R substituents. Our efforts focused on the development of borate esters of the form B(OR)3 where

R are electron-withdrawing arenes, with the intended application in H2 activation chemistry.

A solution containing 3,4,5-trifluorophenol in CH2Cl2 was prepared and cooled to -60 °C to which a solution of a third of an equivalent of BCl3 was added (Scheme 3.4). After warming to room temperature and four hours stirring, a beige solid, 3-1, was recovered in 75 % yield. Examination of the 1H NMR spectrum revealed a multiplet, assignable to six ortho protons, which were found to couple to the neighbouring F-atoms. Two resonances were observable by 19F NMR spectroscopy at -134.34 and -167.33 ppm, pertaining to the meta and para fluorine atoms and integrated in a 2:1 fashion. Notably, the para-F signal was found to be considerably further upfield as compared to the neighbouring meta-F signal. This is related to the electron- withdrawing ability of the borane substituents which delocalize the oxygen lone pair into the ring system. This relocates electron density to the para-carbon, shielding both the carbon and bound fluorine atoms. A broad signal was observed by 11B{1H} NMR spectroscopy at 15.64 ppm which 155 is located in the spectral region for boranes of the form B(OR)3 and is only slightly downfield 156 relative to the related species B(OC6F5)3 (15 ppm). These data and supporting elemental analysis are in keeping with the assignment of the formulation of 3-1 as the borate ester

B(O(3,4,5-F3C6H2))3.

Scheme 3.4 – Synthesis of compounds 3-1, 3-2 and 3-3.

Figure 3.1 – 19F, 1H and 11B{1H} NMR spectra of 3-1.

In a similar fashion, solutions of 4-(trifluoromethyl)phenol or 3,5-bis(trifluoromethyl)phenol

were reacted with one third of an equivalent of BCl3 at -60 °C, yielding off-white solids in 92 (3- 2) and 58 % (3-3) yield (Scheme 3.4). 1H NMR spectroscopy for 3-2 revealed the presence of 3 two doublets with JH-H coupling constants of 9 Hz, pertaining to the ortho- and meta-protons: similarly for 3-3 two singlets, integrating in a 2:1 fashion were observed assignable to the ortho- and para-protons. Nearly identical 19F chemical shifts were observed for both species with singlets at -62.74 and -63.77 ppm for 3-2 and 3-3, respectively. Sharp resonances for 3-2 and 3- 3 were observed by 11B{1H} NMR spectroscopy at 15.98 and 15.94 ppm. These data supported

the formulations of 3-2 and 3-3 as the borate esters B(O(4-(CF3)C6H4))3 and B(O(3,5-

(CF3)2C6H3))3. The solid-state structures of 3-2 and 3-3 were established via X-ray crystallographic techniques and confirmed the assigned formulations (Figure 3.3).

The OR substituents in 3-2 and 3-3 were found to be disposed about the B centre in a propeller- like fashion while the B centers were trigonal planar, with O-B-O angles summing to 360°. The aryl rings in 3-2, as defined by the ipso-carbon atoms C(1), C(8) and C(15) were inclined at

angles of 18.45, 50.02 and 25.67° relative to the BO3 plane while the aryl rings in 3-3 as defined by the ipso-carbons C(1), C(9) and C(17) were found at angles of 47.06, 46.49 and 45.18° relative to the BO3 plane. These are notably different than the related borate ester B(OC6F5)3 which bears aryl ring inclination angles of 52, 77 and 104° relative to the BO3 plane (Figure 3.2).154

Figure 3.2 –Aryl group inclination angle relative to the BO3 plane.

The average B-O bond lengths in 3-2 and 3-3 were found to be statistically identical at 1.360(3) 154 and 1.362(2) Å and indistinguishable from B(OC6F5)3 (1.360(9) Å). Relative bond angles and lengths are listed in Table 3.1.

Figure 3.3 – POV-Ray depictions of the molecular structures of 3-2 and 3-3. B: yellow-green, C: black, O: red, F: pink. SH atoms removed for clarity. Selected bond distances (Å) and angles (°). 3-2: B(1)-O(1), 1.356(3); B(1)-O(2), 1.358(3); B(1)-O(3), 1.367(3); O(3)-C(15), 1.389(3); O(2)- C(8), 1.376(3); O(1)-C(1), 1.385(3); O(1)-B(1)-O(2), 121.0(2); O(2)-B(1)-O(3), 118.0(2); O(3)- B(1)-O(1), 121.0(2); B(1)-O(3)-C(1), 123.99(18); B(1)-O(1)-C(8), 132.06(18); B(1)-O(1)-C(1), 129.69(18). 3-3: B(1)-O(1), 1.364(3); B(1)-O(2), 1.360(3); B(1)-O(3), 1.363(3); O(1)-C(1), 1.384(3); O(2)-C(9), 1.380(3); O(3)-C(17), 1.376(3); O(1)-B(1)-O(2), 119.8(2); O(2)-B(1)-O(3), 119.6(2); O(3)-B(1)-O(1), 120.5(2); B(1)-O(3)-C(17), 126.50(19); B(1)-O(1)-C(1), 123.9(2); B(1)-O(2)-C(9), 125.87(19).

3.2.2 Synthesis of Boronate Esters

Boronate esters find applications in varying fields such as C-C bond formation157-159 and as additives in lithium battery electrolytes.160 The respective Lewis acidity of catechol-containing Lewis acids can easily be tuned via substituent placement on the catechol frame. Addition of electron-donating or withdrawing substituents helps to control the Lewis acidity of the boron centre, by enhancing or dampening π-π interactions between the oxygen donors and the Lewis acidic centre (Figure 3.4). A series of catechol-based boronate esters were subsequently synthesized for application in FLP-type reactivity.

Figure 3.4 – The effects of electron-rich and electron-poor catechol fragments on the Lewis acidity of boronate esters.

Typical boronate ester synthesis involves the stoichiometric reaction of a diol and an alkyl- or arylboronic acid (or the associated trimetric anhydride) in toluene.161 Azeotropic removal of water, employing a Dean-Stark apparatus, drives the reaction forward yielding the desired boronate ester. Addition of molecular sieves or other inorganic drying agents can assist in the removal of the water produced throughout the course of the reaction.161

Synthesis of compounds 3-4, 3-5 and 3-6 involved the stoichiometric reaction of pentafluorophenylboronic acid with 1,2-benzenediol, 3-fluoro-1,2-benzenediol or 3,4,5,6- tetrafluoro-1,2-benzenediol to yield the corresponding boronate esters, (C6H4O2)B(C6F5) 3-4, 11 1 (C6H3FO2)B(C6F5) 3-5 and (C6F4O2)B(C6F5) 3-6 in 78, 75 and 42 % yield. B{ H} NMR spectroscopy gave rise to broad singlets at 29.78, 29.92 and 30.29 ppm pertaining to 3-4, 3-5 and 3-6, respectively. The catechol fragments were confirmed by 1H NMR in 3-4 and 3-5, as evidenced by proton signals in the aromatic region of the spectrum and by 19F NMR spectroscopy for 3-6. Examination of the 19F NMR spectra for compounds 3-4 through 3-6

revealed that despite significant electronic changes at the catecholate fragments, the C6F5 signals are relatively unperturbed (Table 3.1). The para-F signal for 3-6 is however, noted to be deshielded and slightly downfield relative to that of 3-4 and 3-5. This finding is consistent with reduced electron density in the aromatic ring due to the electron-withdrawing effects of the neighbouring catechol ligand. As electronic changes at the boron centre are most reflected at the 12 para position of the bound C6F5 ring, a progressive downfield shift and associated increase in the para-meta gap is anticipated upon an increase in Lewis acidity at the boron centre (Figure 3.5). Based on this notion, compounds 3-4 through 3-6 can be tentatively ordered according to increasing Lewis acidity where 3-4 < 3-5 < 3-6 (Table 3.1). The Lewis acidity of these boronate esters will be discussed further in Chapter 3.2.3.

Figure 3.5 – Correlation between the B centre and the para-F of a C6F5 group on boron.

11 1 19 Table 3.1 – B{ H} and F NMR data (C6F5 groups only) for 3-4, 3-5 and 3-6.

11 1 B{ H} (ppm) ortho-F (ppm) para-F (ppm) meta-F (ppm) Δδp-m (ppm) 3-4 29.78 -129.30 -147.80 -162.13 14.33 3-5 29.92 -128.89 -146.96 -161.83 14.87 3-6 30.29 -127.76 -144.79 -160.81 16.02

Crystals of 3-4 were readily obtained from cold pentane and the solid-state structure was confirmed by X-ray crystallography (Figure 3.6). The B−O bond lengths were found to be indistinguishable at 1.3802(15) and 1.3805(14) Å and are comparable to those observed for the 162 related compound 2-phenyl-1,3,2-benzodioxaborole ((C6H4O2)BPh: B-O, 1.394 Å). The B−O bond lengths fall between that of a B−O single (1.43 Å) and double bond (1.36 Å)163 and are consistent with other three-coordinate boronate esters that bear a considerable degree of double

bond character between the O and B atoms. The boron centre was found to be pseudo-trigonal planar with an O-B-O angle of 112.10(10)° and O-B-C angles of 124.36(10) and 123.36(10)°, while the molecule itself was completely planar. The internal O-B-O angle of 112.10(10)° was determined to be considerably smaller as compared to the related O-B-C angles and approached that expected for tetrahedral geometry (109.5°). The enhanced Lewis acidity of these boronate esters results from the pre-organized tetrahedral-like geometry, which eases the process of boron quaternization upon reaction with a donor molecule.

Figure 3.6 – POV-Ray depiction of the molecular structure of 3-4. B: yellow-green, C: black, O: red, F: pink. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). O(1)-B(1), 1.3820(15); O(2)-B(1), 1.3805(14); O(2)-B(1), 1.3838(13); O(1)-C(7), 1.3917(13); B(1)-C(1), 1.5509(16); O(1)-B(1)-O(2), 112.10(10); O(1)-B(1)-C(1), 124.36(10); C(1)-B(1)- O(2), 123.53(10); C(7)-O(1)-B(1), 104.67(8); C(8)-O(2)-B(1), 104.99(8).

3.2.3 Lewis Acidity Determinations

The quantification of the Lewis acidity of an electrophilic borane is a challenging task and unfortunately, a comprehensive technique for the classification of Lewis acidity which addresses factors such as solvent effects, substrate class and sterics has yet to be elucidated.12 Numerous methodologies exist for the quantification of Lewis acidity such as calorimetry164-167, reactivity168,169 and spectroscopic methods.170 Two of the most broadly applied methodologies for acidity classification are the Gutmann-Beckett and the Childs methods which involve the generation of Lewis-acid base adducts and analysis via either 31P{1H} or 1H NMR spectroscopy (Figure 3.7).

Figure 3.7 – (a) Gutmann-Beckett and (b) Childs Lewis acidity tests.

The Gutmann method was first developed for the classification of solvents according to acceptor number (AN)171,172 and was later modified by Beckett et al. to reflect the Lewis acidity of boranes.173,174 This methodology examines the difference in 31P NMR chemical shift between 175 triphenylphosphine oxide (Et3PO) and the Et3PO−adduct of the borane of study. Similarly, the Childs method involves the generation of the borane adduct of crotonaldehyde.165 The H3 proton of the crotonaldehyde fragment most accurately reflects the Lewis acidity of the borane fragment as it is removed from the locus of coordination (reduced steric interference) but is electronically connected to the site of coordination. Figure 3.8 illustrates the changes in 1H NMR chemical shift of the crotonaldehyde H3 proton upon coordination to a series of boranes.

The Gutmann-Beckett Lewis acidity tests were performed employing a 3:1 Lewis acid:Et3PO 31 1 ratio (excess acid to ensure full complexation of the Et3PO) and were mixed in CD2Cl2. P{ H} NMR spectra of the resulting solutions were collected at room temperature. Similarly, a 1:1 1 Lewis acid:crotonaldehyde ratio in CD2Cl2 was prepared and all H NMR spectra were recorded at -20 °C to prevent polymerization of the aldehyde. It should be noted that fixed acid and base concentrations were employed in all cases.

1 1 Figure 3.8 – H NMR spectra of the crotonaldehyde adducts of B(C6F5)3, 3-3 and 3-4 and the H NMR spectra of free crotonaldehyde.

Table 3.2 – Gutmann-Beckett and Childs Lewis Acidity Tests for compounds 3-1 through 3-6 and other related boranes.

Lewis Acid 31P Δδ L.A. relative H3 Δδ L.A. relative a b (ppm) (ppm) to B(C6F5)3 (ppm) (ppm) to B(C6F5)3

B(C6F5)3 78.06 26.59 1 7.90 1.07 1

BPh3 73.70 22.23 0.83 6.93 0.10 0.09

B(OC6F5)3 82.63 31.16 1.17 7.30 0.47 0.44

B(OPh)3 66.60 15.13 0.57 6.92 0.09 0.08 3-1 81.10 29.63 1.11 6.92 0.09 0.08 3-2 80.21 28.74 1.08 6.90 0.07 0.07 3-3 82.41 30.94 1.16 6.93 0.10 0.09 3-4 78.94 27.47 1.03 6.90 0.07 0.07 3-5 79.81 28.34 1.06 6.89 0.06 0.06 3-6 80.27 28.80 1.08 7.29 0.46 0.43

(C6F5O)B(C6F5)2 81.41 29.94 1.13 7.83 1.00 0.94

MesB(C6F5)2 74.09 22.62 0.85 - - -

PhB(C6F5)2 76.64 25.17 0.95 7.56 0.73 0.68

1,4-C6F4[B(C6F5)2]2 78.58 27.11 1.02 7.86 1.03 0.96

a 31 b 1 P NMR for Et3PO in CD2Cl2: δ 51.47 ppm; H NMR for H3 of crotonaldehyde in CD2Cl2 (-20 °C): δ 6.83 ppm. L.A. = Lewis acidity.

Table 3.2 depicts the spectroscopic results from both the Gutmann-Beckett and Childs Lewis acidity tests. For our purposes, all Lewis acidity measurements have been referenced relative to

B(C6F5)3 as this Lewis acid finds application in many fields of frustrated Lewis pair reactivity. The Lewis acidity of a number of the compounds listed in Table 3.2 have previously been 120 154 154 12,154,176-178 reported in the literature (PhB(C6F5)2, BPh3, (C6F5O)B(C6F5)2, B(C6F5)3 and 154 B(OC6F5)3 ). To ensure identical reaction conditions and for ease of comparison, the Lewis acidities of these species were remeasured as part of the experimental set. The experimental data revealed poor correlation between the Gutmann-Beckett and Childs methods for compounds 3-1 through 3-6. The Gutmann-Beckett method classifies these Lewis acids as more electrophilic as

compared to B(C6F5)3. This counterintuitive result was also noted by Britovsek in the study of

154 the Lewis acidities of compounds of the form (C6F5O)nB(C6F5)3-n (n = 1, 2 or 3). This overestimated Lewis acidity is attributed to the superior binding ability of the hard Lewis acids 154 (3-1 through 3-6) to the hard Lewis base, Et3PO. A weaker interaction is noted between the same hard base and the comparatively soft perfluoroaryl Lewis acids, such as B(C6F5)3. In comparison, the Childs tests indicated that compounds 3-1 through 3-6 are extremely weak

Lewis acids relative to B(C6F5)3 (Table 3.2). This underestimated Lewis acidity is attributable to a weak hard acid-soft base interaction as opposed to the soft acid-soft base interaction noted with perfluoroarylboranes.154 It is evident that the trend in Lewis acidity is strongly dependent on the nature of the donor, and therefore reversals in the Lewis acidity rankings are to be expected upon substitution of the base. Consideration of Pearson’s hard-soft acid-base (HSAB) principle,179 which ranks the hardness or softness of a boron centre as a function of the atoms attached and of the bonding, will assist in choosing an appropriate base for the purposes of Lewis acidity determination. According to this principle, the B-O bonds in 3-1 to 3-6 are more ionic in nature resulting in “hard” Lewis acids. Therefore an appropriate base would be an equally “hard” species such as Et3PO. Comparison of 3-1 to 3-6 is best made employing the Gutmann-Beckett method and is most suitably compared relative to the hard perfluoroaryloxyborane, B(OC6F5)3 (Figure 3.9).

Figure 3.9 – Lewis acidities of 3-1 through 3-6 relative to B(OC6F5)3.

3.2.4 Dihydrogen Activation Employing Borate and Boronate Esters

Borate esters, B(OC6F5)3, 3-1, 3-2 and 3-3 were tested for their ability to function as suitable

Lewis acids in FLP-mediated heterolytic H2 activation. Stoichiometric reactions of B(OC6F5)3,

3-1, 3-2 or 3-3 with PtBu3 were found to yield FLPs in solution as evidenced by no detectable interaction by 11B{1H}, 19F or 31P{1H} NMR spectroscopy. These solutions were subsequently degassed and exposed to 4 atmospheres of H2 in a sealed bomb. The reaction of B(OC6F5)3, 1 PtBu3 and H2 gave a white solid, 3-7, in a mere 25 % yield following manipulation. H NMR 1 spectroscopy revealed a broad doublet at 5.08 with a coupling constant of ( JH-P) 428 Hz, characteristic of a phosphonium ion. The 19F NMR spectrum gave three signals in a 2:1:2 ratio with a p-m gap of 10.67 ppm. A sharp singlet was observable by 11B{1H} NMR spectroscopy at -1.21 ppm and was indicative of a four coordinate boron centre. An X-ray crystallographic study

confirmed the formulation as [tBu3PH][B(OC6F5)4] however, poor data did not allow for the discussion of the metrical parameters.

Scheme 3.5 – The synthesis of 3-7.

The production 3-7 would indicate that heterolytic H2 activation is achieved by the FLP but is followed by ligand redistribution (Scheme 3.5). Analogous reactivity was observed upon

reaction of 3-1, 3-2 and 3-3 with PtBu3 and H2. One equivalent of a hydrogen acceptor molecule

(N-benzylidene-tert-butylamine) was added to the reaction mixture of B(OC6F5)3, PtBu3 and H2 in attempts to provide an H+ and H- sink, which would prohibit ligand redistribution. However, NMR spectroscopy demonstrated that the previously observed reaction pathway is followed in spite of the presence of the unsaturated acceptor molecule.

Due to the chelating catechol ligand in compounds 3-4 through 3-6, ligand scrambling as was

observed in 3-1 to 3-3 with PtBu3 and H2, would likely be precluded. Therefore, stoichiometric 11 1 19 reactions of 3-4, 3-5 and 3-6 with PtBu3 were prepared and followed by B{ H}, F and 31P{1H} NMR spectroscopy. Following six hours reaction, no change was noted and the reagents were deemed to be frustrated Lewis pairs. The solutions were then exposed to a stream

of H2 and stirred for 5 days. No reaction was noted in all three instances and was likely attributable to the insufficient Lewis acidity of the boronate esters.

3.2.5 Dihydrogen Activation Employing Perfluoroarylboranes

A series of perfluoroaryl-containing boranes were tested for their ability to collaboratively activate dihydrogen in the presence of a bulky and basic phosphine.

The Gutmann-Beckett and Childs Lewis acidity tests for 1,4-C6F4[B(C6F5)2]2 show good correlation between the two methods (G-B: 1.02; C: 0.96, relative to B(C6F5)3) which indicates that the dinuclear species possesses nearly identical Lewis acidity compared to the mononuclear

species B(C6F5)3. Therefore 1:2 mixtures of 1,4-C6F4[B(C6F5)2]2 with PR3 (R = tBu and Mes) were prepared in bromobenzene yielding yellow and purple solutions of the respective FLPs

(Scheme 3.6, a). The solutions were exposed to a stream of H2 and stirred for 24 hours. 3-8 and 3-9 were precipitation from solution yielding white solids. The 1H NMR spectra were in keeping 1 with phosphonium ions with broad doublets at 5.25 and 8.26 with JH-P coupling constants of 437 and 480 for 3-8 and 3-9, respectively. Additionally, broad 1:1:1:1 quartets were noted at 3.69 and 3.58 ppm for 3-8 and 3-9, with a coupling constant of 86 Hz, in keeping with the dianion 11 1,4-(C6F4)[HB(C6F5)2]2. This was further supported by a sharp doublet in the B NMR spectra at -25.18 and -24.90 ppm (3-8 and 3-9), which were found to be very similar to the chemical 31 shift for the related species [tBu3PH][HB(C6F5)3] (-25.5 ppm). The presence of four signals in 19 the F NMR spectra confirmed the two fold symmetric nature of the dianion and supported H2 activation at both electrophilic centres. The double activation of H2 employing both boron centres of the bisborane is not entirely surprising due to the electronic similarity to B(C6F5)3. In consideration of the above data 3-8 and 3-9 were determined to be

[R3PH]2[1,4-(C6F4)[HB(C6F5)2]2] (R = tBu, 3-8; Mes, 3-9). Prolonged heating of C6D5Br 11 solutions of both 3-8 and 3-9 for five days at 120 °C showed no evidence of H2 loss by either B 31 or P NMR spectroscopy, analogous to salts of the form [R3PH][HB(C6F5)3] (R = tBu and Mes).31

Scheme 3.6 – Synthesis of (a) 3-8 and 3-9 and (b) 3-10 and 3-11.

Similarly, both PhB(C6F5)2 and MesB(C6F5)2 whose Lewis acidities are reduced as compared to

B(C6F5)3, were reacted with stoichiometric quantities of PtBu3. Following exposure to H2 and 24 hours reaction time, white solids were recovered in 74 (3-10) and 31 % (3-11) yield and assigned as [tBu3PH][HB(C6F5)2R] (R = Ph, 3-10; Mes, 3-11) (Scheme 3.6, b). Anticipated broad 1:1:1:1 quartets in the 1H NMR spectra were observable at 3.66 and 3.70 ppm with one bond H-B couplings of 89 and 84 Hz for 3-10 and 3-11, respectively. Three signals in the 19F NMR spectra confirmed the four-coordinate nature of the boron centre (Δδp-m = 2.32 ppm, 3-10; 1.68 ppm, 4-

11) and suggested free rotation about the B-C6F5 bonds.

The yield for 3-11 was noted to be reduced relative to other known salts of this form. It is likely

that the Lewis acidity of MesB(C6F5)2 is not sufficient to retain the hydride making the back

reaction (loss of H2) facile. However, the low yield of 3-11 may also be attributable to the steric effect imposed by the mesityl group, which obstructs the vacant p orbital of the free borane180 preventing the Lewis acid from partaking in H2 activation with phosphine (Figure 3.10). Studies

by Tilset et al. have shown that the mesityl group in MesB(C6F5)2 is disposed at an angle of 72°

relative to the BC3 plane, effectively obstructing the p-orbital to the extent that only a weak 180 interaction is noted upon reaction with THF or Et2O.

Systems that readily lose H2 often find application in catalytic reductions, in fact recent reports by the Soós group demonstrated the application of ammonium hydridoborate salts, generated

from the heterolytic cleavage of H2 by MesB(C6F5)2 and 1,4-diazabicyclo[2.2.2.]octane, in the catalytic reduction of imines.148

Figure 3.10 – Boron p-orbital obstruction by a mesityl substituent.

A crystallographic study was conducted, confirming the bonding in 3-10 (Figure 3.11). The phosphonium and hydridoborate fragments were noted to be oriented toward each other within the unit cell with a long contact distance of 3.665 Å. The metrical parameters in 3-10 were noted to be unexceptional as compared to other salts of the form [tBu3PH][HBAr3].

Figure 3.11 – POV-Ray depiction of the molecular structure of 3-10. B: yellow-green, C: black, F: pink, P: orange. Select H atoms removed for clarity. Selected bond distances (Å). B(1)-C(13), 1.602(5); B(1)-C(19), 1.636(5); B(1)-C(25), 1.645(5).

3.3 Conclusions

Borate esters of the form B(OR)3 are readily and cleanly synthesized from the reaction of an

appropriate alcohol with BCl3. Similarly, catechol-containing boronate esters are accessible via stoichiometric reactions of boronic acids with appropriate diols. Azeotropic distillation of the resulting water drives the reaction forward to yield the desired products. Borate esters, in the

presence of phosphine are active in the heterolytic activation of H2 and subsequently undergo

ligand redistribution yielding phosphonium-borate salts of the form, [R3PH][B(OR)4]. Attempts to employ chelating ligands to circumvent the process of ligand redistribution yielded boronate esters which were likely of insufficient Lewis acidity to operate in the process of B/P-mediated

H2 activation. Ultimately, perfluoroaryl boranes of the form RB(C6F5)2 maintain appropriate

Lewis acidity to function in the process of H2 activation with bulky and basic phosphines. Ion pairs encompassing weakly bound hydride and proton moieties result in systems which have potential application in the reduction of unsaturated organic compounds (Chapter 4).

3.4 Experimental Section

3.4.1 General Considerations

All manipulations were carried out under an atmosphere of dry, O2-free N2 employing standard

Schlenk-line and glovebox techniques. Solvents (CH2Cl2, pentane and toluene) were dried by employing a Grubbs-type column system (Innovative Technology), degassed and stored under

N2. Bromobenzene was dried over CaH2 and distilled under N2 while CD2Cl2 was vacuum

transferred from CaH2, degassed and stored under N2. 3,4,5-trifluorophenol (Aldrich), 4- (trifluoromethyl)phenol (Aldrich), 3,5-bis(trifluoromethyl)phenol (Aldrich), pentafluorophenylboronic acid (Aldrich), 1,2-benzenediol (Aldrich), 3-fluoro-1,2-benzenediol

(Apollo Scientific), 3,4,5,6-tetrafluoro-1,2-benzendiol (Apollo Scientific), 1.0 M BCl3 solution in hexanes (Aldrich), triethylphosphine oxide (Aldrich), BPh3 (Aldrich) and B(C6F5)3 (Boulder

Scientific) were used as received. Crotonaldehyde (Aldrich) was dried over CaH2, distilled and 180 154 154 181 stored under N2. MesB(C6F5)2, (C6F5O)B(C6F5)2, B(OC6F5)3, B(OPh)3 and 1,4- 119 1 11 13 19 31 C6F4[B(C6F5)2]2 were prepared according to literature methods. H, B, C, F and P NMR spectra were recorded at 25 °C, unless otherwise stated, on a Varian NMR System 400 MHz or Bruker Avance III 400 MHz spectrometer and were referenced using (residual) solvent 1 13 11 19 resonances relative to SiMe4 ( H and C) or to an external standard ( B: (Et2O)BF3, F: CFCl3,

31 P: 85% H3PO4). Chemical shifts are reported in ppm and coupling constants as scalar values in Hz. Combustion analyses were performed in house employing a Perkin-Elmer CHN Analyzer.

3.4.2 Syntheses

Synthesis of 3-1. A 250 mL Schlenk was charged with 3,4,5-

trifluorophenol (1.35 g, 9.12 mmol) in CH2Cl2 (15 mL) while a second

250 mL Schlenk was charged with a 1.0 M solution on BCl3 in hexanes

(3.10 mL, 3.10 mmol) in CH2Cl2 (20 mL). The BCl3 solution was cooled to -60 °C and the solution of 3,4,5-trifluorophenol was added dropwise via cannula. The reaction was maintained at -60 °C, with stirring, for four hours and was then allowed to warm to room temperature. The solvent was removed in vacuo and the residue was left under vacuum for an additional 30 minutes to remove the excess BCl3. The resulting beige solid was purified by recrystallization from pentane at -35 1 11 1 °C. Yield: 1.03 g (75 %). H NMR (CD2Cl2): δ 6.84 (m, 6H, B(OC6H2F3)3). B{ H} NMR 13 1 1 (CD2Cl2): δ 15.64 (s). C{ H} NMR (CD2Cl2): δ 151.62 (dm, JC-F = 249 Hz, p-C6H2F3); 147.62 1 1 (m, ipso-C6H2F3); 137.68 (dm, JC-F = 247 Hz, m-C6H2F3); 105.89 (m, JC-F = 249 Hz, o-C6H2F3). 19 3 3 F NMR (CD2Cl2): δ -134.34 (dd, 6F, JF-F = 21 Hz, JF-H = 8.6 Hz, m-C6H2F3); -167.33 (tt, 3F, 3 4 JF-F = 21 Hz, JF-H = 5.6 Hz, p-C6H2F3). Anal. Calcd. for C18H6BF9O3: C, 47.83; H, 1.34. Found: C, 47.49; H, 1.41 %.

Synthesis of 3-2. A 250 mL Schlenk was charged with 4-

(trifluoromethyl)phenol (1.07 g, 6.62 mmol) in CH2Cl2 (30 mL) while

a second 250 mL Schlenk was charged with a 1.0 M solution of BCl3

in hexanes (2.40 mL, 2.40 mmol) in CH2Cl2 (20 mL). The BCl3 solution was cooled to -60 °C and the solution of 4- (trifluoromethyl)phenol was added dropwise, via cannula. The reaction was maintained at -60 °C with stirring for four hours then allowed to warm to room temperature overnight. The solvent was removed in vacuo and left under active vacuum for an additional 30 minutes to remove the excess BCl3. The resulting product was a yellow oil. An off-white solid was produced upon cooling the oil to -35 °C followed by gentle trituration with cold pentane (10 mL). Purification was achieved by recrystallization of the crude product from pentane at -35 °C. Crystals suitable for X-ray diffraction were grown from a pentane solution at -35 °C. Yield 1.00 g (92 %). 1H

3 3 NMR (CD2Cl2): δ 7.63 (d, 6H, JH-H = 9 Hz, m-C6H4CF3); 7.27 (d, 6H, JH-H = 9 Hz, o- 11 1 13 1 C6H4CF3). B{ H} NMR (CD2Cl2 ): δ 15.98 (s). C{ H} NMR (CD2Cl2): δ155.29 (s, ipso- 3 2 C(CF3)); 127.10 (q, JC-F = 4 Hz, m-C6H4CF3); 126.38 (q, JC-F = 33 Hz, p-C6H4CF3); 124.41 (br 1 19 d, JC-F = 124 Hz, CF3); 120.82 (s, o-C6H4CF3). F NMR (CD2Cl2): δ -62.74 (s). Anal. Calcd.

for C21H12BF9O3: C, 51.05; H, 2.45. Found: C, 50.69; H, 2.56 %.

Synthesis of 3-3. A 250 mL Schlenk was charged with 3,5-

(trifluoromethyl)phenol (2 mL, 13.1 mmol) in CH2Cl2 (15 mL) while a

second 250 mL Schlenk was charged with 1.0 M BCl3 in hexanes (4.5

mL, 4.50 mmol) in CH2Cl2 (20 mL). The BCl3 solution was cooled to -60 °C and the solution of 3,5-bis(trifluoromethyl)phenol was added dropwise via cannula. The reaction was maintained at -60 °C, with stirring, for four hours and was then allowed to warm to room temperature. The solvent was removed in vacuo and left under active vacuum for an additional 30 minutes to remove the excess BCl3. The resulting off-white solid was purified by recrystallization from pentane at -35 °C. Crystals suitable for X-ray diffraction were grown from a pentane solution at -35 °C. Yield: 1 1.80 (58 %). H NMR (CD2Cl2): δ 7.75 (s, 3H, p-C6H3(CF3)2); 7.64 (s, 6H, o-C6H3(CF3)2). 11 1 13 1 B{ H} NMR (CD2Cl2): δ 15.94 (s). C{ H} NMR (CD2Cl2): δ 152.73 (s, ipso-C6H3(CF3)2); 2 1 133.20 (q, JC-F = 34 Hz, m-C6H3(CF3)2); 123.07 (br d, JC-F = 273 Hz, CF3); 121.14 (m, o- 19 C6H3(CF3)2); 137.29 (m, p-C6H3(CF3)2). F NMR (CD2Cl2): δ -63.77 (s). Anal. Calcd. for

C24H9BF18O3: C, 41.29; H, 1.30. Found: C, 41.07; H, 1.24 %.

Synthesis of 3-4. A 250 mL Schlenk flask was charged with 1,2-benzenediol (0.500 g, 4.54 mmol) and pentafluoroboronic acid (0.968 g, 4.54 mmol) in toluene (60 mL) and equipped with a Dean-Stark apparatus. The reaction mixture was heated to reflux for a period of 12 hours. All volatiles were removed under reduced pressure resulting in a maroon solid. The product was taken up in CH2Cl2 (3 mL) and filtered through a Celite plug. Pentane (15 mL) was added to the solution and cooled to -35 °C, resulting in a colourless crystalline product. Crystals suitable for X-ray diffraction were grown from a 1 mixed 1:5 CH2Cl2/pentane solution at -35°C. Yield: 1.01 g (78 %). H NMR (CD2Cl2): δ 7.40 3 4 3 4 (dd, 2H, JH-H = 6 Hz, JH-H = 3 Hz, C6H4); 7.21 (dd, 2H, JH-H = 6 Hz, JH-H = 3 Hz, C6H4). 11 1 13 1 1 B{ H} NMR (CD2Cl2 ): δ 29.78 (s). C{ H} NMR (CD2Cl2): δ 150.45 (br d, JC-F = 254 Hz, o-

1 1 C6F5); 147.78 (s, ipso-C6F5); 144.36 (br d, JC-F = 258 Hz, p-C6F5); 137.86 (br d, JC-F = 253 Hz, 19 m-C6F5); 124.20 (s, m-O2C6H4); 113.52 (s, o-O2C6H4). F NMR (CD2Cl2): δ -129.30 (dt, 2F, 3 4 3 4 JF-F = 19 Hz, JF-F = 8 Hz, o-C6F5); -147.80 (tt, 1F, JF-F = 20 Hz, JF-F = 6 Hz, p-C6F5); -162.13 3 4 (tt, 2F, JF-F = 20 Hz, JF-F = 6 Hz, m-C6F5). Anal. Calcd. for C12H4BF5O2: C, 50.40; H, 1.41. Found: C, 50.20; H, 1.73 %.

Synthesis of 3-5. A 250 mL Schlenk was charged with 3-fluoro-1,2- benzenediol (0.215 g, 1.68 mmol) and pentafluoroboronic acid (0.356 g, 1.68 mmol) in toluene (60 mL) and equipped with a Dean-Stark apparatus. The reaction mixture was heated to reflux for a period of 12 hours. All volatiles were removed under reduced pressure resulting in a white solid. The product was taken up in pentane (5 mL) and filtered through a Celite plug. The crude material was purified by recrystallization from pentane 1 at -35 °C. Yield: 0.382 g (75 %). H NMR (CD2Cl2): δ 7.24 (m, 1H, o-C6H3F); 7.19 (m, 1H, m- 11 1 13 1 C6H3F); 7.04 (m, 1H, m-C6H3F). B{ H} NMR (CD2Cl2 ): δ 29.92 (s). C{ H} NMR ( CD2Cl2) 1 3 partial: δ 150.95 (br d, JC-F = 244 Hz, o-C6F5); 150.60 (d, JC-F = 4.9 Hz, ipso-C6H3F); 149.73 1 1 1 (d, JC-F = 250 Hz, o-C6H3F); 145.05 (br d, JC-F = 242 Hz, p-C6F5); 138.30 (br d, JC-F = 253 Hz, 2 3 m-C6F5); 135.36 (d, JC-F = 14 Hz, ipso-C6H3F); 124.49 (d, JC-F = 7.0 Hz, m-C6H3F); 111.95 (d, 2 4 19 JC-F = 17 Hz, m-C6H3F); 109.60 (d, JC-F = 3.7 Hz, o-C6H3F). F NMR (CD2Cl2): δ -128.89 3 4 (m, o-C6F5); -135.94 (m, 2F, C6H3F); -146.96 (tt, 1F, JF-F = 20 Hz, JF-F = 6 Hz, p-C6F5); -

161.83 (m, 2F, m-C6F5). Anal. Calcd. for C12H3BF6O2: C, 47.37 H, 0.99 %. Found: C, 47.28; H, 1.14 %.

Synthesis of 3-6. A 250 mL Schlenk was charged with 3,4,5,6-tetrafluoro- 1,2-benzenediol (0.410 g, 2.25 mmol) and pentafluoroboronic acid (0.477 g, 2.25 mmol) in toluene (50 mL) and equipped with a Dean-Stark apparatus. The suspension was heated to reflux for a period of 12 hours. All volatiles were removed under reduced pressure resulting in a pink residue. The crude product was sublimed under vacuum at 80 °C, yielding a tan product. The solid was then taken up in pentane, filtered through a bed of Celite. The volatiles were removed yielding a white solid which was 11 subsequently dried for 2 hours in vacuo. Yield: 0.340 g (42 %). B NMR (CD2Cl2): δ 30.29 (s). 13 1 1 1 C{ H} NMR ( CD2Cl2) partial: δ151.06 (br d, JC-F = 256 Hz, o-C6F5); 145.66 (br d, JC-F = 1 1 264 Hz, p-C6F5); 139.11 (br d, JC-F = 250 Hz, C6F4); 138.35 (br d, JC-F = 250 Hz, m-C6F5);

1 19 135.29 (br d, JC-F = 255 Hz, C6F4). F NMR (CD2Cl2): -127.76 (m, 2F, o-C6F5); -144.79 (tt, 3 4 1F, JF-F = 20 Hz, JF-F = 7 Hz, p-C6F5); -160.12 (m, 2F, o-C6F4); -160.81 (m, 2F, m-C6F5); -

163.41 (m, 2F, m-C6F4). Anal. Calcd. for C12BF9O2: C, 40.23; Found: C, 40.20 %.

Synthesis of 3-7. A 100 mL tube bomb was charged B(OC6F5)3 (0.116 g,

0.207 mmol) and PtBu3 (0.042 g, 0.208 mmol) in toluene (10 mL). The reaction was degassed using three freeze-pump-thaw cycles. The reaction bomb was exposed to a continuous flow of H2 for a period of one minute and then sealed and allowed to stir for a period of 24 hours. At this time, pentane (20 mL) was added, precipitating a colourless oil. The resulting oil was triturated with pentane (40 mL) affording a white solid. Yield: 40 mg (25 %). 1 1 3 H NMR (CD2Cl2): δ 5.08 (br d, 1H, JH-P = 428 Hz, HP); 1.64 (d, 27H, JH-P = 16 Hz, PtBu). 11 1 13 1 1 B{ H} NMR (CD2Cl2 ): δ -1.21 (s). C{ H} NMR (CD2Cl2) partial: δ 142.78 (br d, JC-F = 1 1 247 Hz, o-C6F5); 138.35 (br d, JC-F = 246 Hz, p-C6F5); 135.11 (br d, JC-F = 242 Hz, m-C6F5); 1 19 38.25 (d, JC-P = 27 Hz, P{C(CH3)3}); 30.55 (s, P{C(CH3)3}). F NMR (CD2Cl2): δ -158.28 (d, 3 3 3 8F, JF-F = 19 Hz, o-OC6F5); -168.95 (t, 8F, JF-F = 21 Hz, m-OC6F5); (t, 4F, JF-F = 22 Hz, p- 31 1 OC6F5). P NMR (CD2Cl2): δ 61.34 (d, JP-H = 428 Hz, PH). Anal. Calcd. for C36H28BF20O4P: C, 45.69; H, 2.98. Found: C, 45.30; H, 2.83 %.

Synthesis of 3-8. Compounds 3-8 and 3-9 were prepared in a similar fashion and thus only one preparation is detailed. A 100 mL thick walled tube bomb was charged with 1,4-

(C6F4)[B(C6F5)2]2 (0.040 g, 0.048 mmol) and PtBu3 (0.019 g, 0.094 mmol) in bromobenzene (5 mL). The yellow solution was degassed using three freeze-

pump-thaw cycles. The reaction vessel was exposed to a continuous flow of H2 for a period of one minute and then sealed and allowed to stir for a period of 24 h. At this time, pentane (20 mL) was added to the opaque solution precipitating a microcrystalline solid. The solvent was decanted and the product washed with pentane (3 x 5 mL) and dried in vacuo for 2 h. Yield: 1 1 1 0.051 g (93 %). H NMR (CD2Cl2): δ 5.25 (br d, 2H, JH-P = 437 Hz, PH); 3.69 (br q, 2H, JH-B 3 11 1 = 86 Hz, BH); 1.56 (d, 54H, JH-P = 16 Hz, PtBu). B NMR (CD2Cl2): δ -24.9 (d, JB-H = 86 Hz). 13 1 1 1 C{ H} NMR (CD2Cl2): δ149.38 (br dm, JC-F = 234 Hz, o-C6F5); 149.03 (br d, JC-F = 239 Hz, 1 1 C6F4); 138.32 ( br d, JC-F = 243 Hz, p-C6F5); 137.18 (br dm, JC-F = 248 Hz, m-C6F5); 128.74 (br 1 s, ipso-C6F5); 125.16 (br s, ipso-C6F4); 37.94 (d, JC-P = 28 Hz, P{C(CH3)3}); 30.10 (s,

19 3 P{C(CH3)3}). F NMR (CD2Cl2): δ -132.95 (d, 8F, JF-F = 22 Hz, o-C6F5); -137.30 (s, 4F, C6F4); 3 31 -165.24 (t, 4F, JF-F = 22 Hz, p-C6F5); -167.74 (m, 8F, m-C6F5). P NMR (CD2Cl2): δ 59.37 (d, 1 JP-H = 437 Hz). Anal. Calcd. for C54H56B2F24P2: C, 52.04; H, 4.69. Found: C, 51.90; H, 4.87 %.

Synthesis of 3-9. 1,4-(C6F4)[B(C6F5)2]2 (0.060 g, 0.072

mmol) and PMes3 (0.056 g, 0.144 mmol). The pale purple 1 solution before addition of H2. Yield: 0.103 g (88 %). H 1 NMR (CD2Cl2): δ 8.26 (br d, 2H, JH-P = 480 Hz, PH); 7.12 1 (s, 6H, C6H2(CH3)3); 7. 03 (s, 6H, C6H2(CH3)3); 3.58 (br q, 2H, JH-B = 86 Hz, BH); 2.34 (s, 18H, 11 1 p-CH3); 2.24 (s, 18H, o-CH3); 1.98 (s, 18H, o-CH3). B NMR (CD2Cl2): δ -25.18 (d, JB-H = 86 13 1 1 1 Hz). C{ H} NMR (CD2Cl2) partial: δ 148.76 (br d, JC-F = 239 Hz, o-C6F5); 147.83 (br d, JC-F 4 3 = 245 Hz, C6F4); 147.59 (d, JC-F = 2.8 Hz, p-C6H2Me3); 144.64 (d, JC-P = 11 Hz, m-C6H2Me3); 3 1 1 143.45 (d, JC-P = 11 Hz, m-C6H2Me3); 137.30 (br d, JC-F = 249 Hz, p-C6F5); 136.15 (br d, JC-F 2 2 = 245 Hz, m-C6F5); 133.58 (d, JC-P = 12 Hz, o-C6H2Me3); 132.24 (d, JC-F = 11 Hz, o-C6H2Me3); 1 112.02 (d, JC-P = 83 Hz, ipso-C6H2Me3); 22.41 (br s, o-CH3); 21.45 ( br s, p-CH3); 21.34 (br s, 19 3 o-CH3). F NMR (CD2Cl2): δ -133.14 (d, 8F, JF-F = 19 Hz, o-C6F5); -138.28 (s, 4F, C6F4); - 3 31 166.02 (t, JF-F = 20 Hz, 4F, p-C6F5); -168.07 (m, 8F, m-C6F5). P NMR (CD2Cl2): δ -26.38 (d, 1 JP-H = 482 Hz). Anal. Calcd. for C84H70B2F24P2·CH2Cl2: C, 59.88; H, 4.26. Found: C, 59.83; H, 4.67 %.

Synthesis of 3-10. A 100 mL thick walled tube bomb was charged

with PhB(C6F5)2 (0.183 g, 0.434 mmol) and PtBu3 (0.088 g, 0.435 mmol) in toluene (10 mL). The clear and colourless solution was degassed using three freeze-

pump-thaw cycles. The reaction bomb was exposed to a continuous flow of H2 for a period of one minute and then sealed and allowed to stir for a period of 24 h. Pentane (20 mL) was added precipitating a clear and colourless oil. The solvent was decanted, the residue redissolved in

CH2Cl2 (3 mL) and triturated with pentane (40 mL), affording a white powder. The solvent was again decanted, the product washed with pentane (3 x 5 mL) and dried in vacuo for 2 h. Crystals

suitable for X-ray diffraction were grown from a layered solution of CH2Cl2 and pentane at 25 1 3 ºC. Yield: 0.202 g (74 %). H NMR (CD2Cl2): δ 7.16 (d, 2H, JH-H = 6.8 Hz, o-C6H5); 7.04 (t, 3 3 1 2H, JH-H = 6.8 Hz, m-C6H5); 6.94 (t, 1H, JH-H = 7.2 Hz, p-C6H5); 5.12 (br d, 1H, JH-P = 427 Hz, 1 3 11 PH); 3.66 (br q, JH-B = 89 Hz, BH); 1.58 (d, 27H, JH-P = 16 Hz, PtBu). B NMR (CD2Cl2): δ -

1 13 1 1 19.71 (d, JB-H = 89 Hz). C{ H} NMR (CD2Cl2) partial: δ 148.70 (br d, JC-F = 249 Hz, o- 1 1 C6F5); 138.18 (br d, JC-F = 253 Hz, p-C6F5); 137.11 (br d, JC-F = 242 Hz, m-C6F5); 134.00 (s, p- 1 C6H5); 126.91 (s, o-C6H5); 123.44 (s, m-C6H5); 38.06 (d, JC-P = 27 Hz, P{C(CH3)3}); 30.47 (s, 19 3 3 P{C(CH3)3}. F NMR (CD2Cl2): δ -131.68 (d, 4F, JF-F = 23 Hz, o-C6F5); -165.48 (t, 2F, JF-F = 3 31 1 20 Hz, p-C6F5); -167.80 (t, 4F, JF-F = 24 Hz, m-C6F5). P NMR (CD2Cl2): 59.55 (d, JP-H = 427

Hz). Anal. Calcd. for C30H34BF10P: C, 57.53; H, 5.47. Found: C, 57.25; H, 5.19 %.

Synthesis of 3-11. MesB(C6F5)2 (0.133 g, 0.287 mmol) and PtBu3 (0.058 g, 0.287 mmol) in pentane (10 mL). The yellow solution was degassed using three freeze-pump-thaw cycles. The reaction bomb was exposed to a continuous flow of H2 for a period of one minute and then sealed and allowed to stir for a period of 24 h. At this time the solvent was removed yielding a yellow residue. The crude product was washed with 1 pentane (3 x 5mL) resulting in a white powder. Yield: 0.059 g (31 %). H NMR (CD2Cl2): δ 1 1 6.60 (s, 2H, meta-CH2(CH3)3); 5.07 (br d, 1H, JH-P = 430 Hz, PH); 3.70 (br q, JH-B = 84 Hz, 3 BH); 2.17 (s, 3H, para-CH2(CH3)3); 2.07 (s, 6H, ortho-CH2(CH3)3); 1.60 (d, JH-P = 16 Hz, 11 1 13 1 PtBu). B NMR (CD2Cl2): δ -22.08 (d, JB-H = 84 Hz). C{ H} NMR (CD2Cl2) partial: δ 1 1 148.78 (br d, JC-F = 226 Hz, o-C6F5); 142.65 (s, o-C6H2(CH3)3); 137.70 (br d, JC-F = 244 Hz, p- 1 C6F5); 136.87 (br d, JC-F = 254 Hz, m-C6F5); 132.42 (s, m-C6H2(CH3)3); 127.90 (s, p- 1 C6H2(CH3)3); 38.15 (d, JC-P = 27 Hz, P{C(CH3)3}); 30.49 (s, P{C(CH3)3}); 24.25 (s, o- 19 3 C6H2(CH3)3); 21.20 (s, p-C6H2(CH3)3). F NMR (CD2Cl2): δ -133.18 (d, 4F, JF-F = 23 Hz, o- 3 31 C6F5), -166.22 (t, 2F, JF-F = 21 Hz, p-C6F5), -167.90 (m, 4F, m-C6F5). P NMR (CD2Cl2): 60.68 1 (d, JP-H = 430 Hz). Anal. Calcd. for C48H46BF10P: C, 67.42; H, 5.43. Found: C, 67.23; H, 5.33 %.

3.4.3 X-ray Crystallography

3.4.3.1 X-ray Data Collection and Reduction

In preparation for analysis, crystals were first coated with Paratone-N oil in a glovebox and were

frames were integrated employing the Bruker SAINT software package employing a narrow frame algorithm. Absorption corrections were conducted employing the empirical multi-scan method (SADABS).

3.4.3.2 Structure and Refinement

Non-hydrogen atomic scattering factors were taken from literature tabulations.127 Direct methods were employed to elucidate the positions of the heavy metals using the SHELXTL direct methods procedure. Remaining non-hydrogen atoms were subsequently found from successive difference Fourier map calculations. All cycles of refinement were carried out employing full- 2 matrix least squares techniques on F, minimizing the function ω (Fo-Fc) where weight (ω) 2 2 equates to 4Fo /2σ (Fo ) and Fo and Fc are equal to the observed and the calculated structure factor amplitudes. In the final cycles of each refinement, all non-hydrogen atoms were assigned anisotropic temperature factors in the absence of disorder or insufficient data. In the instance of the latter case the atoms were then treated isotropically. Unless noted, the C-H atom positions were calculated and allowed to ride on the carbon to which they are bonded with an assumed C- H bond length of 0.95 Å. Temperature factors pertaining to the H-atoms were fixed at 1.20 times the isotropic temperature factor of the C-atom to which they are bound. The H-atom contributions were calculated however never refined. The locations of the largest peaks in the final difference Fourier map calculations as well as the magnitude of the residual electron densities in each case were of no chemical significance.

3.4.3.3 Tables of Crystallographic Data

Table 3.3 – Selected crystallographic data for 3-2, 3-3, 3-4 and 3-10.

3-2 3-3 3-4 3-10

Formula C21H12BF9O3 C24H9BF18O3 C12H4BF5O2 C30H34BF10P Formula wt 494.12 698.12 285.96 626.35 Crystal system Monoclinic Triclinic Monoclinic Monoclinic

Space group C2/c P-1 P21/n P21/c a (Å) 13.8692(7) 8.9246(4) 7.4706(8) 9.508(2) b (Å) 13.4263(6) 8.9972(4) 6.0946(6) 16.170(4) c (Å) 22.0941(11) 16.6085(7) 23.464(3) 19.191(4) α (deg) 90 86.400(2) 90 90 β (deg) 94.773(3) 77.376(2) 92.603(6) 91.518(10) γ (deg) 90 88.971(2) 90 90 V (Å3) 4099.9(3) 1298.78(10) 1067.24(19) 2949.4(12) Z 8 2 4 4 T (K) 150(2) 150(2) 150(2) 150(2) d (calc) gcm-3 1.601 1.785 1.780 1.411 Abs coeff, μ, mm-1 0.159 0.202 0.176 0.176 Data collected 28124 21720 20781 25914 R int 0.0373 0.0196 0.0507 0.0361 # of indpndt reflns 3789 5928 4203 6830

Reflns Fo≥2.0σ(Fo) 2551 4645 2634 4590 Variables 334 469 181 385 R (>2σ) 0.0442 0.0608 0.0402 0.0674

wR2 0.1271 0.1735 0.1041 0.2054 Goodness of fit 0.998 1.060 0.953 1.048

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

Chapter 4 Borane Derivatization Employing Diazomethanes

4.1 Introduction

4.1.1 Tris(pentafluorophenyl)borane

It was in the early 1960s when pivotal work in the field of borane synthesis was accomplished by

the groups of Massey and Chambers with the discovery of X2B(C6F5) (X = Cl and F) and 10,182 ClB(C6F5)2. Most notably however, was the discovery made by Stone, Massey and Park of 10,11 the sterically encumbered electrophilic Lewis acid, tris(pentafluorophenyl)borane, B(C6F5)3. Studies of this new species revealed that it possessed Lewis acidity comparable to that of the smaller Lewis acid, BF3, and encompassed the added feature of being less susceptible to 183 hydrolysis, contrary to other boron trihalides of the form BX3 (X = F, Cl, Br, I). B(C6F5)3 was found to be a much more versatile Lewis acid mainly due to its increased resistance to B-C bond cleavage in the presence of protic species.12 Although the characteristics of this

perfluoroarylborane are well-known today, B(C6F5)3 remained essentially unutilized from the time of its discovery in 1964 until the early portion of the 1990s.183-185 At this juncture, tris(pentafluorophenyl)borane received a revival due to the discovery by the groups of Marks and Ewen that the bulky and electrophilic borane could serve as a highly effective alkyl abstracting agent in the activation of early metal metallocene derivatives, thereby generating exceptionally active polymerization catalysts (Scheme 4.1).186,187

Scheme 4.1 – Activation of dimethylzirconocene by B(C6F5)3

Unearthing the utility of electrophilic boranes in the activation of metallocene cations prompted a plethora of research in the field of borane synthesis toward the development of novel triarylboranes for the use as olefin polymerization co-catalysts.

4.1.2 Current Methods for the Synthesis of Perfluoroarylboranes

Recent efforts have demonstrated applications of triarylboranes in the fields of organometallic184,188-191 and main group192-194 chemistry, optoelectronic materials,195-206 frustrated Lewis pair chemistry1,3,4,44,207 in addition to the development of sensors used for the detection of anions and toxic species. B(C6F5)3 has also found application in Lewis acid- mediated catalysis183,208, novel metal-free hydrogenations43,45 and in the catalytic synthesis of silicones.209 Due to the vast applicability of these compounds in varying disciplines, it is evident that straightforward and high-yielding synthetic routes must be accessible. This task however, is challenging as the syntheses and subsequent purifications of such triarylboranes are by no means trivial, and are frequently technically challenging. Over the decades, numerous synthetic methods have been developed for the syntheses of triarylboranes, such as the simple 154,185,210-212 derivatization of halide-containing precursors like BCl3, BF3 or ClB(C6F5)2. Reaction

of these synthons with lithium (RLi) or Grignard reagents (RMgX) allows for the installation of

new R functionalities on boron, yielding BR3 or RB(C6F5)2, respectively. The Piers group has

demonstrated the elegant use of ClB(C6F5)2 as a precursor to the hydroborating agent HB(C6F5)2 via H/Cl exchange with triethylsilane.210,213 This species offers a route to the generation of a series of alkyldiarylboranes via the hydroboration of unsaturated substrates. Other elegant methods have been developed in parallel by the groups of Piers176,214-223 and Marks119,164,224-229 which allow access to a catalogue of bulky perfluoroarylboranes and bisboranes (Figure 4.1). A notable drawback to these methodologies is their dependence on the use of mercury or tin reagents, both of which are known to be highly toxic and carcinogenic.12,230

Figure 4.1 – Examples of perfluoroarylboranes developed by (a) Marks231 and (b) Piers.176 Bisboranes developed by (c) Marks227 and (d) Piers.223

Most methodologies focus on the derivatization of B−X (X = F, Cl, Br, I) or B−H bonds; however in recent years progress has been made toward the development of strategies for the facile derivatization of B−C bonds. Simplistic yet elegant strategies have recently been elucidated by the groups of Erker74,75,232-237and Berke238 which demonstrate the facile

derivatization of B(C6F5)3 via reaction with alkynes. B(C6F5)3, in the presence of terminal or internal alkynes, undergoes 1,1−carboboration reactions at room or elevated temperatures yielding the associated vinylboranes (Scheme 4.2). These species, although reduced in Lewis acidity relative to B(C6F5)3, have been found to a possess the degree of Lewis acidity required to reduce unsaturated substrates in frustrated Lewis pair chemistry.239 This is a unique method of derivatization as it offers a simple one-pot synthesis to otherwise synthetically challenging species.

Scheme 4.2 – 1,1−Carboboration reactions employing tris(pentafluorophenyl)borane and (a) terminal alkynes yielding a mixture of E/Z isomers and (b) internal alkynes.

A further example demonstrating B−C bond derivatization is a product of the slow decomposition of a mixture of tris(pentafluorophenyl)borane and bis(amino)silylene over a three

month period of time (Scheme 4.3). During this time, migration of a C6F5 group from boron to 240 silicon is observed generating the new borane (HCNtBu)2Si(C6F5)(B(C6F5)2).

Scheme 4.3 – Decomposition of a B(C6F5)3-silylene adduct.

Although a plethora of synthetic methods is available for B−X and B−H bond derivatization, a comprehensive description of simple and straightforward methods of B−C bond derivatization requires an elaborate chapter.

4.1.3 Reaction of Boranes and Diazomethanes

Diazomethanes tend to be highly reactive species, frequently unstable and often prove challenging to manipulate in a safe fashion. Reactions between diazo-compounds and boranes differ depending on the nature of the reagents.

The straightforward reaction of diazomethane (H2C(N2)) with boron trifluoride (BF3) results in

the generation of a polymer terminated by the BF3 moiety - the result of initial borane and

diazomethane adduct formation followed by decomposition yielding the species [F3B− (CH2)n]. Chain propagation occurs via attack of a second molecule of diazomethane at the carbon centre

with concomitant N2 liberation, and so forth. Chain termination is thought to occur via B-C bond cleavage to yield the free olefin.241

Reactions of trialkylboranes with diazomethanes have been found to be sluggish and poor yielding or completely unfeasible depending on substituent size. Reaction of chlorodialkylboranes with diazoacetates were demonstrated by Brown to have considerably different reactivity as compared to reactions of diazomethanes with boron trihalides.242 In this instance, reactions of the diazo-species with dialkylchloroboranes generate B−C adducts,

yielding quaternary boron species followed by decomposition of the diazo moiety via loss of N2.

The liberation of N2 is either followed by, or occurs concomitantly, with halide or alkyl migration (Scheme 4.4). The speculated products of these reactions were not isolable as they were found to be thermally unstable requiring reduced temperatures during their preparation.242

Scheme 4.4 – Reaction of diazoacetate with dialkylchloroborane via migration of a (a) chloride or (b) alkyl group.242 (c) Derivatization of B-R-9-borabicyclononane via insertion of trimethylsilyldiazomethane into a B−C bond.243

In more recent studies, Soderquist et al. demonstrated the use of trimethylsilyldiazomethane toward the derivatization of B-R-9-borabicyclononanes (B−R−9−BBN). In these reactions, the

Me3SiCH fragment is found to insert into the highly strained B−C bond of the B−R−9−BBN precursor thereby generating the expanded B−R−10−TMS−9−borabicyclodecane (B−R−10−TMS−9−BBD) (Scheme 4.4).243 It is likely that the insertion is assisted by the BBN ring strain, which is alleviated upon insertion of the RCH fragment. In related work conducted

by Shea and Bai, derivatization of the dimethylsulfide-borane adduct, (Me2S·BH3), is achieved by reaction with trimethylsilyldiazomethane yielding tris(trimethylsilylmethyl)borane.244

Despite the examples described above, reactions of diazomethanes with bulky electrophilic triarylboranes remains vastly uncharted territory.

4.2 Results and Discussion

4.2.1 Reactions of Diazomethanes with B(C6F5)3

The simple stoichiometric reaction of tris(pentafluorophenyl)borane and trimethylsilyldiazomethane (Me3SiCH(N2)), at room temperature, was found to yield a complex

mixture of unidentifiable products. Repetition of the reaction where a CH2Cl2 solution of the borane was pre-cooled to -78 °C prior to the addition of a solution of the diazomethane, yielded a pale beige solution. It is of note that even at this temperature the diazomethane is consumed

nearly immediately upon addition, as evidenced by rapid evolution of N2 and the disappearance of the characteristic yellow colour of the diazomethane. This is in keeping with the extremely unstable nature of aliphatic diazo compounds as compared to their aromatic counterparts.245 Following removal of the volatiles, a beige oil was isolated in 65 % yield. Subsequent manipulation of the product allowed for the isolation of a new compound, 4-1, as a white solid. It was suspected that the Me3SiCH fragment had formally been inserted into a B−C bond of the borane with subsequent migration of a perfluorophenyl group to the adjacent carbon atom

yielding (C6F5)2B(Me3SiCH(C6F5)) (Scheme 4.5).

Scheme 4.5 – The synthesis of 4-2 and 4-1.

The 19F NMR spectrum was found to be most diagnostic with respect to the identification of the product, displaying six resonances pertaining to the ortho-, para- and meta-fluorine atoms of the

two equivalent C6F5 rings bound to boron (-130.9, -149.8 and -161.3 ppm, respectively) and the

ortho-, para- and meta-fluorine signals pertaining to the C6F5 ring bound to carbon (-140.8, - 158.9 and -162.9 ppm, respectively). Recrystallization of the crude product allowed for the isolation of single crystals that were analyzed by X-ray diffraction, confirming the formulation of 1 4-1 as a racemic mixture of (C6F5)2B(Me3SiCH(C6F5)). The H NMR spectrum revealed two distinct resonances assignable to the methine and trimethylsilyl fragments at 4.38 and 0.16 ppm, respectively. An examination of the 11B{1H} NMR spectrum at room temperature revealed no discernible signal, between -60 and 85 °C.

An examination of the metrical parameters revealed that the Me3Si group of the tertiary carbon is orientated in such a way as to obstruct one face of the B plane (Figure 4.2). The closest B···C−H

(of the Me3Si group) contact is 3.10 Å from the electron-poor boron centre. This interaction might be expected to limit the reactivity at boron by effectively blocking the face from an incoming substrate. To probe this hypothesis, a low temperature 1H NMR study was conducted between the temperatures of 25 and -85 °C. At no time was there any discernible interaction

observed between the Me3Si fragment and the boron centre.

Reactions of B(C6F5)3 with two equivalents of Me3SiCH(N2) under conditions similar to those

described above, resulted in the straightforward synthesis of 4-2, where Me3SiCH fragments were inserted into two B−C bonds of B(C6F5)3 yielding (C6F5)B(Me3SiCH(C6F5))2 in 67 % yield (Scheme. 4.5). A close examination of the 19F NMR spectrum revealed that two products were present in a 72:28 ratio. The major product was identifiable by eight distinct signals (9 anticipated signals however two signals were found to overlap) pertaining to three chemically

inequivalent C6F5 rings. The minor isomer, comparatively, gave rise to 15 distinct fluorine signals that were resolved at -85 °C. Signals assignable to the methine and trimethylsilyl fragments were observed in the 1H NMR spectrum at 3.79 and 0.22 ppm for the major isomer and 3.58 and -0.17 ppm for the minor isomer. An examination of the 11B{1H} NMR spectrum indicated the presence of a broad peak at 75.4 ppm, contrary to that of 4-1 where the 11B{1H}

NMR spectrum was silent. Large single crystals were obtained from a layered solution of CH2Cl2 and pentane at -35 °C and were analyzed by X-ray crystallography to establish the solid-state structure of 4-2.

Although insertions of Me3SiCH into one or two B−C bonds of B(C6F5)3 are possible, attempts

to insert a third equivalent to generate B(Me3SiCH(C6F5))3 by adding five or ten equivalents of

Me3SiCH(N2) to a pre-cooled solution of the borane gave no reaction. It is likely that the

incorporation of the third Me3SiCH fragment is not feasible for steric reasons. The steric congestion about the boron centre of 4-2 was supported by the inability to generate an acetonitrile adduct of the compound when employing the Lewis base as the solvent.

Figure 4.2 - POV-Ray depictions of the molecular structures of 4-1 and 4-2. B: yellow-green, C: black, F: pink, Si: blue. Select H atoms removed for clarity. Selected bond distances (Å). 4-1: B(1)-C(1), 1.577(2); B(1)-C(7), 1.593(2); B(1)-C(13), 1.545(2); C(13)-B(1)-C(1), 122.28(14); C(1)-B(1)-C(7), 117.83(15); C(7)-B(1)-C(13), 119.64(14). 4-2: B(1)-C(1), 1.556(3); B(1)-C(11), 1.600(3); B(1)-C(23), 1.566(3); C(1)-B(1)-C(23), 120.76(18); C(23)-B(1)-C(11), 118.48(17); C(11)-B(1)-C(1), 120.74(18).

An examination of the solid-state structure of 4-2 revealed that the B-C bond to the arene was 1.600(3) Å, which was lengthened relative to the analogous bonds in 4-1. Similarly, the B−C bond lengths to the tertiary carbons were found to be 1.556(3) and 1.566(3) Å, respectively, both lengthened relative to the analogous B−C bond in 4-1. The lengthening of these bonds is consistent with the reduced Lewis acidity of the B centre upon incorporation of two electron- donating Me3SiCH groups.

The generality of diazomethane reactivity with tris(pentafluorophenyl)borane was further probed by a stoichiometric reaction of pentafluorophenyldiazomethane (C6F5)CH(N2), 4-3, with

B(C6F5)3. Due to the well-known difficulties associated with the handling of diazomethanes and their inherent thermal instability,246,247 extreme precautions were taken in order to avoid injury.

(C6F5)CH(N2) was prepared from the corresponding hydrazone with conversion to the

diazomethane by Swern oxidation in a controlled fashion at -78 °C.248 Clean insertion, as in 4-1 and 4-2, was achieved by addition of a solution of diazomethane to a pre-cooled dichloromethane solution of the borane at -78 °C. Following manipulation with cold pentane, a beige solid, 4-4, was isolated from the reaction medium in 60 % yield and was purified by recrystallization. 4-4 was suspected to be analogous in structure to that of 4-1 (Scheme 4.6).

Scheme 4.6 – Synthesis of 4-4.

A single crystal X-ray diffraction study confirmed the formulation of 4-4 as

(C6F5)2B((CH(C6F5)2) (Figure 4.3). This assignment was supported spectroscopically as well. 19 Six resonances were observed in the F NMR spectrum assignable to two equivalent C6F5 rings

bound to boron and two C6F5 rings bound to a tertiary carbon. As anticipated, the signals for the

two distinct C6F5 groups presented in a 1:1 ratio. A single resonance was observed in the 11B{1H} spectrum at 72.34 ppm indicating a three-coordinate boron centre, and the methine proton presented at 5.28 ppm in the 1H NMR spectrum. The Lewis acidity of 4-4 was suspected to be greater than that of 4-1 due to the electron-withdrawing nature of the (C6F5)CH moiety. The increased Lewis acidity of 4-4 was further substantiated by its solid-state structure, which had shorter B−C6F5 bonds (1.553(4) and 1.568(4) Å) compared to 4-1 (1.577(2) and 1.593(2) Å).

Figure 4.3 – POV-Ray depiction of the molecular structure of 4-4. B: yellow-green, C: black, F: pink. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). C(1)-B(1), 1.553(4); C(7)-B(1), 1.568(4); C(1)-B(1)-C(13), 120.8(2); C(1)-B(1)-C(7), 119.3(2); C(7)-B(1)- C(13), 119.9(2).

Compounds 4-1 and 4-2 result in products bearing one or two stereogenic centres, respectively. Product separation to yield stereochemically pure boranes is desirable for their application to asymmetric catalysis. Conventional methods of separation, including column chromatography, are not feasible due to the air-sensitivity of the boranes. Efforts to generate enantiomerically pure boranes upon diazomethane insertion remain a focus of ongoing work.

In order to eliminate the inherent complication caused by the generation of a stereogenic centre, a symmetrically disubstituted diazomethane was sought for investigating reactions with

B(C6F5)3. Diphenyldiazomethane Ph2C(N2), was a suitable reagent due to the fact that it is isolable, facile to manipulate and stable when stored in the dark at -35 °C.248 Stoichiometric,

room temperature reactions with B(C6F5)3 revealed no interaction with the borane apart from decomposition of the diazomethane moiety to the free carbene and eventual coupling of two carbene moieties to yield tetraphenylethylene. Heating of the reaction mixture to 80 °C with 10

equivalents of Ph2C(N2) appeared to result in a small amount of Ph2C insertion into a B−C bond

of B(C6F5)3, albeit not cleanly. It is likely that the large, electrophilic perfluoroarylborane is

incapable of accommodating the sterically demanding Ph2C fragment, preventing facile insertion

of the Ph2C moiety.

The mechanism by which the R’RC fragment is inserted into the B−C bonds of B(C6F5)3 is of interest. Based on spectroscopic and X-ray crystallographic studies, the diazomethane species is best described with a triple bond between the two N atoms, and an adjacent nucleophilic carbon centre.245 This species initially interacts with a borane via the nucleophilic carbon atom, quaternizing the boron centre. Subsequent decomposition of the “activated” diazomethane, via

cleavage of the relatively weak C−N bond to yield gaseous N2, followed by aryl group migration, provides the observed boranes. Such 1,2-migrations are quite common among organoboranes 249 when the nucleophile bears an appropriate leaving group or electron sink, N2 in this instance. An analogous mechanism was previously proposed by Brown in 1972, for the reaction of chlorodialkylboranes with ethyldiazoacetate.242

Scheme 4.7 – Mechanism for diazomethane insertion into B(C6F5)3.

4.2.2 Reactions of Diazomethanes with RB(C6F5)2 and BAr3

In order to probe the generality of diazomethane insertion, a series of aryl-containing boranes 126 were investigated. PhB(C6F5)2 was selected as a suitable Lewis acid due to the presence of both phenyl and perfluorophenyl substituents. This Lewis acid allowed for the determination of

whether RR’C insertion would be selective for a B−C6H5 bond as opposed to a B−C6F5 bond upon reaction with RR’C(N2).

A stoichiometric reaction of PhB(C6F5)2 and Me3SiCH(N2) was initially conducted under conditions similar to those used to prepare 4-1, 4-2, and 4-4. The 19F NMR spectrum of the reaction mixture appeared to contain at least two products, which were not easily identifiable or separable, however these species were suspected to be the products of both single and double insertion into the B−C bonds of the borane. Repetition of the reaction employing two

equivalents of Me3SiCH(N2) and one equivalent of PhB(C6F5)2 gave a clear and colourless solution, which upon removal of the solvent and subsequent workup yielded a colourless oil in 64 % yield. Cooling a solution of the product in pentane for a week at -35 °C resulted in large rectangular crystals, which were crystallographically analyzed confirming the identity of 4-5 as

(C6F5)B(Me3SiCH(C6F5))(Me3SiCH(C6H5)) (Figure 4.5). The structure of 4-5 was found to be unexceptional, resembling that of 4-2. The solid-state structure of 4-5 bore the same cog-wheel- type orientation of the substituents about the boron centre. The geometry at boron was found to be pseudo-trigonal planar with the C−B−C angles summing to nearly 360 °.

Figure 4.4 – POV-Ray depiction of the molecular structure of 4-5. B: yellow-green, C: black, F: pink, Si: blue. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). C(4)-B(1), 1.5942(18); C(5)-B(1), 1.5743(18); C(15)-B(1), 1.5554(18); C(4)-B(1)-C(5), 117.36(10); C(4)-B(1)-C(15), 121.66(11); C(5)-B(1)-C(15), 120.96(10).

Examination of the 19F NMR spectrum of the crude product indicated that the reaction was analytically pure, requiring no further purification. Ten distinct resonances were observed in the 19 F NMR spectrum pertaining to a C6F5 ring bound to boron and that bound to an adjacent tertiary carbon atom. In this instance, all fluorine atoms were found to be chemically distinct. It

is of note that in compounds bearing a Me3SiCH moiety, the ortho-F signals of a C6F5 ring

bound to boron are quite readily distinguishable from a C6F5 ring bound to a tertiary-carbon as they appear separated by a minimum chemical shift difference of at least 6 ppm in the 19F NMR spectrum. This difference in chemical shift can be attributed to the proximity of the fluorine atom relative to the electron-poor boron centre. Resonance structures demonstrate how electron density can be formally withdrawn from the ortho position of a perfluorophenyl group bound to boron, effectively deshielding these nuclei while similar resonance structures are not feasible for a perfluorophenyl group adjacent to carbon (Figure 4.5).

Compound 4-5 bore a broad signal in its 11B{1H} NMR spectrum at 76.31 ppm. Interestingly, only two methine signals (3.70 and 3.53 ppm) and two SiMe3 signals (0.17 and 0.16 ppm) were observable by 1H NMR spectroscopy. Collectively, the 1H, 19F NMR and crystallographic data indicated the selective formation of the RR/SS diastereomers since there was no evidence for the formation of the RS/SR diastereomers as was observed in the double-insertion product, 4-2. The

chirality induced by the initial Me3SiCH insertion makes the second insertion stereoselective. It

is likely that following the first Me3SiCH insertion, a π-stacking interaction is established between two adjacent aryl groups, which in the transition state assists in orientating the second

insertion of the Me3SiCH moiety in a fashion selective for the RR/SS diastereomers. Similar selectivity was observed in the formation of 4-2, although this resulted in a 72:28 distribution of racemic and mesomeric products.

As previously indicated, the products of diazomethane insertion were derived from highly electrophilic boranes bearing at least two electron-withdrawing perfluoroaryl rings. It was of interest to determine whether similar reactivity would be observed with a less Lewis acidic

borane. Stoichiometric reactions of BPh3 and MeSi3CH(N2) at -78 °C resulted exclusively in the racemic mono-insertion product, Ph2B(Me3SiCH(C6H5)), in 75 % yield. This compound was isolated as a clear and colourless oil, and was converted to the pyridine adduct,

(C5H5N)B(Me3SiCH(C6H5))Ph2, 4-6, (Scheme 4.8). Resonances attributable to aromatic protons 1 were observable in the H NMR spectrum corresponding to the pyridine donor and two BC6H5 groups. These findings were consistent with the 1,2-migration of a C6H5 group to the carbon of

the adjacent Me3SiCH fragment. Similarly, signals pertaining to the methine and trimethylsilyl groups were observed at 2.69 and -0.27 ppm, respectively. A sharp signal was seen by 11B{1H} NMR spectroscopy at 11.98 ppm, consistent with a four-coordinate boron centre.

Although only the stoichiometric reaction of BPh3 with Me3SiCH(N2) was investigated, it is

conceivable that at least one further Me3SiCH fragment could be inserted into a second B−C6H5 bond of the borane. It should be noted that with each insertion of Me3SiCH, there is a marked reduction in the Lewis acidity of the boron centre. This reduction can be attributed to the

incorporation of an electron-donating Me3Si group into the borane framework, in addition to the migration of a moderate electron-withdrawing arene away from B, where its effects are diminished.

Investigations by Brown et al. in the early 1970s proposed that the reaction of ethyldiazoacetate, with chlorodialkylboranes, could proceed by two possible insertion pathways (Scheme 4.4). In the first case, an alkyl group migrates from the boron centre to the adjacent carbon atom, and in the second case the chloride ion migrates instead.242 It was of interest to investigate the reactivity

of ClB(C6F5)2 species to establish whether the migration of aryl or chloro groups would occur preferentially to yield isolable materials.

210,250 A stoichiometric reaction of Me3SiCH(N2) with ClB(C6F5)2 in dichloromethane at -78 °C yielded a series of products by 19F NMR, which were not discernible from each other.

Me3SiCH(N2) was too reactive when subjected to the chloroborane and resulted in uncontrolled

reactivity. This reactivity is not surprising considering the highly reactive nature of aliphatic diazo compounds. Conversely, aromatic diazo compounds are inherently more stable due to 245 strong π-π interactions between the Cα−Nβ atoms of the diazomethane R2CαNβNγ. Due to its

increased stability, the reaction of Ph2C(N2) and ClB(C6F5)2 was investigated at room

temperature. The diazomethane was dissolved in a small portion of CH2Cl2 and was then added to a solution of the borane in a controlled fashion. The resulting reaction was vigorous in nature and foamed considerably as N2 was evolved. The progress of the reaction was easily monitored

as the initially fuchsia-coloured solution turned a bright-yellow colour as N2 was evolved. This reaction was found to be extremely fast and nearly quantitative in nature, and a yellow oil, 4-7, was recovered in 97 % yield.

The 19F NMR spectrum revealed the presence of five signals pertaining to two chemically 11 1 distinct C6F5 groups. A broad signal was observed in the B{ H} NMR spectrum at 60.31 ppm,

which was found to be essentially unchanged from that of the parent borane ClB(C6F5)2 210 (59.1ppm). It had previously been noted that C6F5 migration to an adjacent Me3SiCH fragment produced dramatic chemical shift differences of the o-F, B−C6F5 and C−C6F5 groups.

However, upon reaction with Ph2C(N2) the chemical shift difference between these o-F signals is narrowed considerably, making signal assignment more challenging. Considering this data the

formulation of the unknown compound was proposed as (C6F5)B(Ph2C(C6F5))Cl, 4-7 (Scheme 4.9). This compound was then reacted with 10 equivalents of pyridine isolated as a white solid, the Lewis acid-base adduct, 4-8.

Scheme 4.8 - Synthesis of 4-6, 4-7 and 4-8.

The 19F NMR spectrum was consistent with the generation of a four-coordinate boron centre with characteristic narrowing of the m-p gap.210 The signals were, however, broad at room temperature, indicating a fluxional process in solution, but were successfully resolved at -75 °C. The broad 11B{1H} signal of 4-7 at 6.30 ppm upon coordination of pyridine was characteristic of a B−N adduct (Scheme 4.8). An X-ray crystallographic analysis of 4-8 confirmed its structure as the pyridine-adduct (Figure 4.6). The B−N bond length was found to be 1.6438(15) Å, longer 189 relative to the pyridine adduct of B(C6F5)3, C5H5NB(C6F5)3 (B−N, 1.628(2) Å) in keeping with the reduced Lewis acidity of 4-7. The B−Cl bond was found to be 1.8911(12) Å, which is 251 comparatively longer than the B−Cl bond in the parent borane ClB(C6F5)2 (1.746(5) Å) in accordance with the presence of the donor at B. It should be noted that both 4-7 and 4-8 could not be successfully characterized by elemental analysis, therefore the triethylphosphine adduct of 4-9 was generated, but unfortunately, this compound also proved to be unsatisfactory for analysis.

Figure 4.6 – POV-Ray depiction of the molecular structure of 4-8. B: yellow-green, C: black, N: aquamarine, F: pink, Cl: green. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). B(1)-N(1), 1.6438(15); B(1)-C(1), 1.6484(16); Cl(1)-B(1), 1.8911(12); B(1)- C(16), 1.7229(15); N(1)-B(1)-C(1), 105.14(8); N(1)-B(1)-C(16), 109.52(8); Cl(1)-B(1)-C(16), 109.23(7); Cl(1)-B(1)-C(1), 105.93(7).

Compound 4-7 is the sole product of the reaction of ClB(C6F5)2 with Ph2C(N2), where a C6F5 group has migrated to the neighbouring carbon atom of the Ph2C fragment showing that the first case of Brown’s proposed pathways was operable in this system. No evidence was found to corroborate the migration of the chloride atom to the adjacent carbon atom. Compound 4-7 was

reacted with 2, 3 and 5 equivalents of Ph2C(N2) in order to promote further reaction of the remaining B-C bond, however, this was not observed.

Retention of the B-Cl bond in 4-7 yields the possibility for subsequent derivatization. This species has potential as a unique synthon for a number of prochiral compounds of the form

RB(C6F5)(Ph2C(C6F5)). Reaction of the B−Cl bond to yield a new B−C unit could be achieved by employing nucleophilic aryl- or alkylating agents including lithium, magnesium, zinc, copper or tin reagents. The generation of any number of other B−R bonds (R = N, O or P) would also be easily accessible. As an example, in a synthesis published by Piers et al., the effective hydroborating agent, HB(C6F5)2, commonly referred to as “Piers’ borane”, was generated from 210,213 the reaction of ClB(C6F5)2 with Me2SiClH via chloride/hydride exchange. In attempts to

100

generate a derivative of Piers’ borane, 4-7 was reacted in neat Me2SiClH for a period of four

hours at room temperature. Removal of the excess silane and volatile byproduct, Me2SiCl2, in vacuo, resulted in the quantitative isolation of a microcrystalline, white product, 4-10. An examination of the 19F NMR spectrum revealed three distinct resonances in a 2:1:2 ratio 19 pertaining to a C6F5 group bound to boron. Assuming Cl/H exchange with 4-7, six distinct F 19 resonances should be expected for the two C6F5 groups. The F NMR data indicated migration 1 of a C6F5 group from the quaternary carbon to the boron centre. The H NMR spectrum gave no evidence of a B-H bond; however, a broad singlet was observable at 4.95 ppm, indicative of a CH bond, in addition to the expected aromatic proton resonances. In consideration of the above

data, the identity of 4-10 was proposed to be (C6F5)2B(Ph2CH).

Mechanistically, Cl/H exchange is suspected to be the first step in the reaction. Subsequently a

C6F5 group migrates back to the boron centre, indicating that diazomethane insertion into

HB(C6F5)2 is reversible. The resulting carbocation is resonance-stabilized by its two phenyl substituents and is presumably quite robust (Figure 4.7). The reaction terminates by hydride transfer to the carbocation, yielding 4-10 (Scheme 4.9).

Scheme 4.9 – Proposed mechanism for the synthesis of 4-10.

101

Figure 4.7 – Resonance stabilized carbocation en route to 4-10.

The aforementioned examples illustrate that aryl group migration occurs preferentially over halide migration, in chlorodiarylboranes. It was of interest to probe whether a perfluorophenoxy borane would undergo RR’C insertion, and if so, into which of the B-bound substituents this would occur. Based on previous results, it was suspected that C6F5 migration from B to C would

occur preferentially to OC6F5 migration.

This hypothesis was tested by reacting a pre-cooled solution of the borinic ester 154 (C6F5)2B(OC6F5) with a solution containing one equivalent of Me3SiCH(N2). Upon addition of 19 the diazomethane at -78 °C, immediate evolution of N2 gas was noted. The F NMR spectrum

revealed nine distinct signals assignable to three chemically unique C6F5 environments. This

indicated that insertion of the Me3SiCH fragment had occurred into one B−C bond as opposed to the B−O bond. The product was recovered as a racemic mixture with CH and Me3Si signals observed in the 1H NMR spectrum in a 1:9 ratio at 3.24 and 0.16 ppm, respectively. A broad resonance was observed in the 11B{1H} NMR spectrum at 51.00 ppm which was found to be 154 11 downfield relative to the parent borane (C6F5)2B(OC6F5)2 (41.2 ppm). The B NMR signal for 4-11 is noted to be considerably upfield relative to those of 4-1, 4-2, 4-4, 4-5, 4-7 and 4-10 due to π-interactions between the O and B atoms (Figure 4.8, a and b).252

102

252 Figure 4.8 – (a) B2pπ-O2pπ orbital interaction. (b) Resonance structures of 4-11.

4.2.3 Reactions of Diazomethanes with Boronic Acids and Boronate Esters

Sections 4.2.1 and 4.2.2 described the reactivity of bulky and electrophilic boranes with diazomethanes resulting in simple B−C derivatizations, however, it remains unknown whether similar reactivity would be observed with significantly less Lewis acidic boranes. Curiosity prompted the study of the interaction of diazomethanes with boronic acids and boronate esters.

Scheme 4.10 – Synthesis of 4-12, 4-13 and 4-14.

The stoichiometric reaction of pentafluorophenylboronic acid, (C6F5)B(OH)2, and Ph2C(N2) was

carried out in CH2Cl2 at room temperature. Dropwise addition of a Ph2C(N2) solution to the

boronic acid resulted in vigorous evolution of N2 and complete consumption of the fuchsia- coloured diazomethane over a fifteen minute time period. The resulting crude product was contaminated with residual diphenyldiazomethane and required a series of recrystallizations from

CH2Cl2 and pentane to afford the clean product, 4-12, in 80 % yield. Three signals were observed in by 19F NMR spectroscopy at -133.27, -157.44 and -162.25 ppm, consistent with a

103

single C6F5 ring and a narrow p-m gap, indicative of a C-C6F5 fragment. An examination of the 1H NMR spectrum revealed signals in the aromatic region, consistent with the presence of two phenyl groups; however, the absence of a broad OH signal was noted. This led to the formulation of 4-12 as the trimer [(Ph2C(C6F5))BO]3, the cyclic anhydride resulting from self- condensation of the boronic acid (Scheme 4.10).163

Single crystals of 4-12 were grown from a layered solution of CH2Cl2 and pentane at -35 °C and were analyzed by X-ray crystallography. The molecular structure confirmed the proposed B3O3 ring structure of the boroxine where Ph2C insertion into the B−C6F5 bond of the former boronic acid had occurred. An examination of the X-ray data revealed that only one third of the cyclic

species was contained within the asymmetric unit and that the central B3O3 core was found to be slightly distorted from planarity, which is consistent with the structure of the related species, triphenylboroxine.253 The aromatic rings are disposed about each quaternary carbon in a propeller-like fashion in order to reduce negative steric interactions.

The self-promoted dehydration of the corresponding boronic acid is known to occur under non- aqueous conditions.163,254,255 The conversion between the boronic acid and boroxine species is reversible, which allows the derivatized boronic acid to be accessible from the boroxine in the presence of water (Scheme 4.11).163

Scheme 4.11 – The reversible dehydration/rehydration reaction of a boronic acid to the corresponding boroxine.

Analogous reactions employing 4-fluorophenylboronic acid, (C6H4F)B(OH)2, and phenylboronic acid, PhB(OH)2, in conjunction with Ph2C(N2) were carried out and found to yield products

analogous to 4-12, namely [(Ph2C(C6H4F))BO]3, 4-13, and [(Ph3C)BO]3, 4-14. Reactions of

104

(C6H4F)B(OH)2 and PhB(OH)2 with Ph2C(N2) were not found to be as fast or vigorous as that observed for 4-12. In fact the consumption of the diazomethane was slow in these cases, and the solution colours progressed from fuchsia to colourless only after several hours. As the number of electron-withdrawing aromatic rings is decreased, reaction times with the diazomethane are

increased. An order of reactivity with Ph2C(N2) can be written as follows:

C6F5B(OH)2 > p-C6H4FB(OH)2 > C6H5B(OH)2. This order is presumably related to the difficulty

associated with the initial coordination of the less basic diazomethane Ph2C(N2) to the progressively less Lewis acidic borane. The most rapid reaction is therefore always anticipated to occur with the most Lewis acidic boronic acid. A single resonance was observed in the 11B{1H} NMR spectrum for both 4-13 and 4-14 at 33.20 and 32.50 ppm, respectively. Both of these signals are typical of 11B{1H} NMR chemical shifts for substituted boroxines 256 ([(o-tol)BO]3: 30.0 ppm).

The 19F NMR spectrum of 4-13 revealed a signal at -118.27 ppm, which appeared as a complex 3 4 multiplet due to JF-H and JF-H coupling to the neighbouring aromatic protons. Crystals of 4-13

and 4-14 were grown from solutions of CH2Cl2 layered with pentane at -35 °C. An X-ray

analysis of 4-14 confirmed the proposed formulation as [(CPh3)BO]3, however, the data for 4-13 were of insufficient quality to discuss metrical parameters. Nevertheless, the data still allowed

for determination of atom connectivity, thereby confirming 4-13 as [(Ph2C(p-F-C6H4))BO]3 (Figure 4.9).

The B3O3 core of 4-14 was found to be slightly distorted from planarity and possessed a

staggered orientation of the substituents about the B3O3 plane. Slight variation of the B−O lengths was noted in both 4-12 and 4-14, with three of the B−O bonds slightly elongated relative to each of their two adjacent neighbours. This difference in bond length is in keeping with the notion that boroxines are isoelectronic with benzene and have partial aromatic character due to the vacant p-orbital on B, which can interact with the neighbouring O atoms (Figure 4.10).163

105

Figure 4.9 – POV-Ray depictions of the molecular structures of 4-12 and 4-14. B: yellow-green, C: black, F: pink, O: red. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). 4-12: B(1)-O(1), 1.3697(17); B-C(1), 1.596(2); O(1)-B(1)-O(1), 119.05(12); O(1)- B(1)-C(1), 121.08(11); O(1)-B(1)-C(1), 119.53(11). 4-14: B(1)-O(1), 1.3810(18); B(1)-O(2), 1.3736(18); B(2)-O(2), 1.3830(18); B(2)-O(3), 1.3722(18); B(3)-O(1), 1.3768(17); B(3)-O(3), 1.3689(18); B(1)-C(1), 1.604(2); B(2)-C(20), 1.602(2); B(3)-C(39), 1.5968(19); C(1)-B(1)-O(1), 118.33(12); C(1)-B(1)-O(2), 123.35(12); O(2)-B(1)-O(1), 118.26(12); C(20)-B(2)-O(3), 119.89(12); C(20)-B(2)-O(2), 122.23(12); O(2)-B(2)-O(3), 117.73(13); O(3)-B(3)-C(39), 119.39(12); C(39)-B(3)-O(1), 122.41(12); O(3)-B(3)-O(1), 117.95(12).

Figure 4.10 – Boroxine resonance contributors (a sampling).

106

Boronic acids are mildly Lewis acidic, inherently stable and easy to handle. These properties tend to make boronic acids an attractive class of precursors in synthetic chemistry.163 They have found vast application in the field of organic chemistry as synthons in palladium-catalyzed cross coupling reactions (i.e. Suzuki).157-159 The above-developed methodology provides an easily accessible route to a number of starting materials for subsequent coupling reactions. In order to further extend the library of boron-based synthons that could be applied in C-C cross-coupling reactions, the reactivity of diazomethanes with boronate esters was investigated.

149 A stoichiometric reaction of 2-pentafluorophenyl-1,3,2-benzodioxaborole, (C6H4O2)B(C6F5) and Me3SiCH(N2) was performed in CH2Cl2. After stirring at room temperature overnight, an aliquot of the reaction medium revealed that only partial conversion to a new product, 4-15, had occurred. Two and three subsequent equivalents of the diazomethane were reacted with the boronic ester with 3 equivalents yielding full conversion to 4-15, which was recovered in 90 % 19 yield. The F NMR spectrum revealed that insertion of the Me3SiCH moiety into a B-C6F5 bond of (C6H4O2)B(C6F5) had occurred as evidenced by three distinct resonances at -141.02, -161.29 and -163.96 ppm, respectively, pertaining to the ortho-, para- and meta-F atoms, respectively of 11 1 the C−C6F5 fragment. A narrow resonance was observed in the corresponding B{ H} NMR spectrum at 34.38 ppm. These data collectively supported the formulation of 4-15 as

(C6H4O2)B(Me3SiCH(C6F5)) (Scheme 4.12).

Scheme 4.12 – Synthesis of compound 4-15, 4-16, 4-17 and 4-18.

To further probe substitution of the highly electron-withdrawing perfluorophenyl substituent, the 149 stoichiometric reaction of 2-(4-fluorophenyl)-1,3,2-benzodioxaborole with Me3SiCH(N2) was investigated. Aliquots from the reaction medium revealed slow and incomplete (roughly 20 %) conversion to a new product, 4-16. To encourage full conversion, a solution containing a 5:1

107

ratio of diazomethane to boronate ester was prepared and allowed to stir at room temperature overnight. Subsequent removal of the volatiles revealed an analytically pure, colourless oil,

which upon standing solidified, yielding solid 4-16, in 83 % yield. Insertion of the Me3SiCH moiety was confirmed by two resonances in the 1H NMR at 2.32 and -0.03 ppm pertaining to the methine and trimethylsilyl fragments, respectively. A broad singlet was observed in the 11B{1H} NMR spectrum at 35.20 ppm, while a complex multiplet was observed by19F NMR spectroscopy at -119.97 ppm. These resonances were found to be shifted downfield and upfield, respectively, relative to those of the parent borane (11B{1H}: 31.6 ppm; 19F -106.51 ppm).257 The above data collectively pointed toward the identity of 4-16 as (C6H4O2)B(Me3SiCH(p-F-C6H4)) (Scheme 4.12). Reactions of 4-15 and 4-16 with excess pyridine resulted in adducts, yielding 4-17 and 4-18, respectively, which were isolable as white solids in 78 and 96 % yield, respectively. A crystallographic study confirmed the identities of 4-17 and 4-18 as pyridine adducts,

[(C5H4N)B(C6H4O2)(Me3SiCH(Ar))] (Ar = C6F5, 4-17; Ar = p-F-C6H4, 4-18), also confirming the previously proposed formulations of 4-15 and 4-16 (Figure 4.11).

An examination of the metrical parameters of 4-17 and 4-18 revealed the four-coordinate, pseudo-tetrahedral nature of the B centres. The B−C, B−O and B−N bond lengths of 4-17 and 4- 18 were found to be statistically indistinguishable from each other. Compound 4-17 bore notably longer B-O bond lengths, 1.474(4) and 1.472(4) Å, relative to the parent boronic ester, 149 (C6H4O2)B(C6F5), with analogous bond lengths of 1.3820(15) and 1.3805(14) Å, which is in keeping with the reduced Lewis acidity of 4-17. This finding is consistent with a typical four- coordinate boronic ester-Lewis base adduct where the B−O bond lengths are commonly found in the range of 1.43-1.47 Å.163 The B−O bond lengths were similar to those observed for the 4- picoline adduct of 2-phenyl-1,3,2-benzoxadiborole (B−O, 1.481(3) and 1.464(3) Å).258 The considerably longer B−O bonds reflect the quenched acidity of the borane, which would otherwise possess π-orbital overlap between the B and O, resulting in partial double bond character between these atoms.163 Additionally, the O−B−O bond angles (4-17: 106.5(2)°; 4-18: 106.25(15)°) were compressed relative to those expected for a perfect tetrahedral geometry

(109.5°), as is common for adducts of this class of borane (MeC5H4N(C6H4O2)BPh: 106.0(2)°).258

108

Figure 4.11 – POV-Ray depictions of the molecular structures of 4-17 and 4-18. B: yellow- green, C: black, N: aquamarine, F: pink, O: red, Si: blue. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). 4-17: C(1)-C(7), 1.520(4); C(7)-B(1), 1.623(4); B(1)- N(1), 1.660(4); B(1)-O(1), 1.474(4); B(1)-O(2), 1.472(4); C(7)-B(1)-O(1), 120.3(3); C(7)-B(1)- N(1), 107.5(2); N(1)-B(1)-O(2), 105.6(2); O(2)-B(1)-O(1), 106.5(2). 4-18: C(33)-C(34), 1.509(3); C(33)-B(2), 1.608(3); B(2)-N(2), 1.664(3); B(2)-O(4), 1.474(2); B(2)-O(3), 1.477(2); C(33)-B(2)-O(4), 117.80(16); C(33)-B(2)-N(2), 107.61(15); N(2)-B(2)-O(3), 105.33(15); O(3)- B(2)-O(4), 106.25(15).

It should be noted that combinations of the boronate esters (C6H4O2)B(C6F5) and

(C6H4O2)B(p-F-C6H4) with the bulkier diazomethane Ph2C(N2) led to no reactions under ambient conditions. Upon heating the reaction mixtures to 80 °C for 16 hours, slow conversion to new

maroon-coloured products was observed where Ph2C had been inserted into the B-C bonds, although these reactions were not clean. Subsequent manipulation of the crude materials failed to yield analytically pure products.

4.2.4 Reactions of Catecholborane and B-Chlorocatecholborane with Diazomethanes

Catecholborane commonly finds application in the preparation of alkyl- or vinylboranes via hydroboration of the corresponding alkene or alkyne for subsequent use in Suzuki cross-coupling

109 reactions.157-159 It has also has been used as an effective agent in the selective reduction of β- hydroxy ketones to the corresponding syn 1,3-diols.259 In fact, catecholborane has been found to be one of the most versatile boron-containing species in common-day synthesis.260

Considering the borane/diazomethane reactivity discussed in the previous sections, it was hypothesized that reaction of a diazomethane with either catecholborane (HBCat) or B- chlorocatecholborane (ClBCat) might result in selective RR’C insertion into either the B−H or the B−Cl bond, yielding the corresponding boronate esters. This would provide a fast, room temperature and high-yielding synthetic route to a library of branched boronic esters from commercially available materials.

Low temperature stoichiometric reactions of HBCat and Me3SiCH(N2) resulted in a series of unidentifiable products by 11B{1H} and 1H NMR spectroscopy. The less reactive diazomethane,

Ph2C(N2) was employed to promote controlled reaction with HBCat. A stoichiometric quantity of Ph2C(N2) was added, in a dropwise fashion, to a room temperature solution of HBCat in

CH2Cl2. Immediate effervescence was noted upon addition of the diazomethane. Complete consumption of the reagent occurred over a 30 minute time period from which a yellow oil, 4-19, was isolated (Scheme 4.13). Resonances pertaining to the catechol fragment and two phenyl groups, integrating to a total of 14 H, were observed in the 1H NMR spectrum of the analytically pure sample. Retention of the catechol fragment was not surprising as the chelate effect of this bidentate ligand makes dissociation of an arm unfavorable. A sharp singlet was observed in the aliphatic region at 4.30 ppm and was thought to correlate to a methine proton. A broad signal was observed in the 11B{1H} NMR spectrum at 34.63 ppm, downfield from that of the parent borane HBCat (28.3 ppm).260 Considering these data, the formulation of 4-19 was assigned as

(C6H4O2)B(CPh2H). Subsequent attempts to obtain a molecular structure of the insertion product were unsuccessful. However, high resolution mass spectrometry supported the proposed formulation.

110

Scheme 4.13 – Synthesis of 4-19, 4-20 and 4-21.

In an analogous fashion, a solution of ClBCat was reacted with a stoichiometric amount of

Ph2C(N2) resulting in immediate reaction of the diazomethane as evidenced by the instantaneous disappearance of the fuchsia-coloured reagent. The reaction was found to be vigorous and precautions were taken during diazomethane addition to ensure that the vessel did not overflow. A yellow oil, 4-20, was recovered from the reaction medium, which solidified upon standing. 1H NMR spectroscopy confirmed that the catechol fragment had remained intact during manipulation and that Ph2C had been incorporated into the structure of 4-20. Similar to the 11B{1H} NMR spectrum of 4-19 and the boronate esters mentioned in Section 4.2.3, a new signal was observed at 32.79 ppm by 11B{1H} NMR spectroscopy. The spectroscopic data, in

conjunction with elemental analysis, led to the assignment of this species as (C6H4O2)B(CPh2Cl), 4-20. The new boronate ester demonstrated typical Lewis acid reactivity toward the hard Lewis base triethylphosphine oxide179 generating the anticipated Lewis acid-base adduct, 4-21. This assignment was supported by a sharp 11B{1H} NMR signal at 11.09 ppm in keeping with a four- coordinate boron centre in addition to a sharp and narrow signal in the 31P NMR spectrum at 77.63. The 31P resonance was found to be more than 20 ppm downfield from free triethylphosphine oxide which is in keeping with the reduced electron density at the P centre

upon coordination of Et3PO to the boronate ester.

It should be noted that previous attempts to react boronate esters of the form (C6H4O2)BR (R =

C6F5, p-F-C6H4) with Ph2C(N2) resulted in the generation of derivatized boronate esters of the

form (C6H4O2)B(CPh2R), however, synthesis involved the use of excess reagent and forceful conditions. Discouragingly, the isolated products were impure and subsequent purification proved unsuccessful. The above methodology, involving the direct and fast reaction of HBCat and ClBCat with bulky diazomethanes, provides a synthetic route to the related bulky boronate esters from commercial synthons.

111

4.2.5 Reactivity of HB(C6F5)2 with Diazomethanes and Azides

Bis(pentafluorophenyl)borane, HB(C6F5)2, a powerful hydroborating reagent, was prepared by Piers et al. in the mid-1990s.213 This highly electrophilic borane undergoes facile hydroboration of both simple alkenes and alkynes yielding the corresponding alkyl- or akenylboranes. A notable shortcoming of this reagent is the propensity of the products of hydroboration to undergo retrohydroboration and subsequent rehydroboration to yield isomerized products.210,213 Due to

the previously elucidated reactivity of Ph2C(N2) with the B−H bond of catecholborane, it was

postulated that HB(C6F5)2 and Ph2C(N2) might interact in a fashion analogous to 4-19, yielding a slightly less electrophilic variant of Piers’ borane, which may prove to be less prone to retrohydroboration

A stoichiometric reaction of HB(C6F5)2 with Me3SiCH(N2) at -78 °C yielded a series of unidentifiable products that were not readily isolated from the reaction mixture. Subsequently, a

1:1 reaction of HB(C6F5)2 and Ph2C(N2) was conducted at room temperature in CH2Cl2. The ensuing reaction was found to be slow and took place over four hours. Removal of the volatiles yielded a yellow solid, which was subsequently purified via recrystallization from a solution of

CH2Cl2 layered with pentane at -35 °C. The resulting pure compound, 4-22, was isolated as colourless crystals in 83 % yield. The 1H NMR spectrum revealed resonances in the aromatic

region assignable to the phenyl groups of the Ph2C moiety. A diagnostic resonance at 8.75 ppm, integrating to 1H, drew considerable attention and was consistent with that of an NH fragment. A single, broad resonance was observed at 34.84 ppm in the 11B{1H} NMR spectrum which curiously lacked any B−H coupling, indicating that the H atom was no longer bound to B.

Typically, HB(C6F5)2 is found to exist along with an equilibrium concentration of its dimer in 11 1 solution. The B{ H} NMR spectrum of the HB(C6F5)2 monomer displays a resonance at 60.1 ppm and the dimer manifests itself at 18.0 ppm.213 The observation of a 11B{1H} NMR chemical shift at 34.84 ppm did not represent either of these types of species. Six signals were observed in 19 the F NMR spectrum, assignable to two inequivalent C6F5 rings indicative of some degree of hindered rotation about the B-N bond. These data, in addition to the observation that no effervescence was noted throughout the course of the reaction, led to the formulation of the product as (C6F5)2BNHNCPh2, 4-22 (Scheme 4.14). An X-ray crystallographic study of 4-22 was performed, confirming the proposed formulation (Figure. 4.12).

112

Scheme 4.14 – Synthesis of 4-22.

The molecular structure of 4-22 revealed capture of the intact diazomethane, in contrast to aforementioned compounds wherein decomposition of the diazomethane occurred with carbene insertion into the B−C bond. The B−N bond distance of 1.385(2) Å was found to be a short bond, in comparison to the related amido-borane species (TMS)2NB(C6F5)2 with a B−N bond length of 1.400(3) Å.211 This shortened bond is indicative of considerable π-character in the B-N bond, a result of N lone pair delocalization of the nitrogen lone pair.261 The restricted rotation and subsequent inequivalence of the C6F5 groups in 4-22 can therefore be attributed to a resonance contributor of 4-22 (Figure 4.13).

113

Figure 4.12 – POV-Ray depiction of the molecular structure of 4-22. B: yellow-green, C: black, N: aquamarine, F: pink. Select H atoms removed for clarity. Selected bond distances (Å) and angles (°). B(1)-N(1), 1.385(2); B(1)-C(13), 1.581(2); B(1)-C(19), 1.577(2); N(1)-N(2), 1.3940(18); N(2)-C(25), 1.287(2); C(1)-C(25), 1.484(2); C(7)-C(25), 1.499(2); N(1)-B(1)-C(19), 118.20(15); N(1)-B(1)-C(13), 121.14(4); C(13)-B(1)-C(19), 120.64(14); B(1)-N(1)-N(2), 122.80(14).

Figure 4.13 – Resonance structures of 4-22.

The related species, (Me3Si)HNB(C6F5)2, was found to exist in equilibrium with the dimeric form in solution, displaying two resonances by 11B{1H} NMR spectroscopy at 38.2 and -2.8 ppm, respectively.211 Compound 4-22 however, was found to exist solely as the monomeric species in solution, likely attributable to steric limitations.

The proposed mechanism for the synthesis of the aforementioned boranes involved coordination of the diazomethane fragment via the nucleophilic carbon centre, followed by N2 liberation and

C6F5 migration. However, the mechanism for the synthesis 4-22 was thought to be quite

114

different. The first step is presumed to involve the hydroboration of the Ph2C(N2) moiety by

HB(C6F5)2 followed by a 1,2-proton shift from the C-bound N-atom to the N-atom adjacent to boron, yielding 4-22 (Scheme 4.15).

Scheme 4.15 – Proposed mechanism for the synthesis of 4-22.

Additional studies were conducted to elucidate further reactivity of 4-22. The compound was found to be very thermally robust and showed no signs of decomposition despite heating in bromobenzene at 150 °C for four days. Attempts were made to utilize 4-22 in reactions with disubstituted acetylenes in order to promote Diels-Alder cyclizations. Reactions employing both stoichiometric and large excesses of both diphenylacetylene and dimethyl acetylenedicarboxylate with 4-22 demonstrated no interaction, even under thermal duress.

Furthermore, a sample of 4-22 was taken up in CH2Cl2 and photolyzed employing a water- cooled quartz mercury arc lamp for a period of four hours. During this time a new decomposition product was observed in small quantities but was unfortunately unidentifiable.

Considering of the reactivity of Piers’ borane with Ph2C(N2) it was of interest to investigate whether similar reactivity might be observed upon the reaction of the hydroborating agent with

organic azides. A stoichiometric reaction of HB(C6F5)2 with trimethylsilylazide, Me3SiN3, in

CH2Cl2 was stirred at room temperature for a period of one hour. Following removal of the volatiles an analytically pure white solid, 4-23, was isolated in 93 % yield. The 1H NMR spectrum revealed retention of the B−H bond with a characteristically broad quartet at 3.81 ppm 1 with JH-B = 85 Hz. A sharp signal at 0.49 ppm was also observed and confirmed the presence of the trimethylsilyl group in the product. A narrowing of the m-p gap to 6.66 ppm in the 19F NMR spectrum was characteristic of a four-coordinate anionic boron centre and was further supported by a sharp doublet in the 11B NMR spectrum at -9.64 ppm indicative of a hydridoborate species.

115

These data supported the formulation of 4-23 as the organic azide adduct, [(Me3SiN3)BH(C6F5)2] (Scheme 4.16).

Scheme 4.16 – Synthesis of compounds 4-23 and 4-24.

Similarly, a 1:1 reaction of HB(C6F5)2 and benzylazide, (C6H5)CH2N3, was conducted at room

temperature. Contrary to the former reaction with Me3SiN3, upon addition of (C6H5)CH2N3, vigorous evolution of N2 gas was observed, which ceased after two minutes of stirring. Removal of the volatiles and subsequent manipulation of the crude material with cold pentane yielded a white product, 4-24. The 11B{1H} NMR data was in keeping with a three coordinate B centre, with a broad resonance at 34.36 ppm. Similarly, the 1H NMR spectrum displayed no evidence of a B−H bond; in fact a broad signal integrating to a single proton was visible at 5.92 ppm and was 19 assigned to an N-H fragment. Examination of the F NMR spectrum revealed that the two C6F5 groups were inequivalent as evidenced by six resonances. This data bore a great deal of similarity to the spectroscopic data of 4-22, and accordingly, 4-24 was assigned as the alkylaminodiarylborane (C6F5)2BNH(CH2Ph). The chemical inequivalence of the perfluoroaryl rings could be justified as a result of restricted rotation about the B−N bond, due to pπ-pπ interactions, common among aminoboranes.261,262 Numerous attempts to grow crystals suitable for X-ray crystallography and molecular structure determination proved unsuccessful. The formulation of 4-24 was however, further supported by elemental analysis.

116

Scheme 4.17 – Mechanism for the reaction of organic azides N3R (R = non-bulky substituent)

with HB(C6F5)2.

The differing reactivities of organic azides versus diazomethanes with bis(pentafluorophenyl)borane can be justified by employing a steric argument. As depicted in Scheme 4.17, the organic azide can be described by two resonance structures. When the R group is sterically demanding, as is the case with trimethylsilylazide, the azide may only coordinate end-on via the terminal N-atom. When R is less sterically cumbersome, as with benzylazide, an alternative mode of bonding is accessible. This mode involves coordination of the azide via the

internal N-atom, adjacent to the R group. N2 evolution and subsequent migration of a substituent

to the boron centre generates the new aminoborane (Scheme 4.17). The inability of the Me3SiN3 species to bind to boron via the internal N-atom is not only a result of the reagent steric demands but is also brought about by the sterically encumbered nature of the borane.263 The reaction of trialkylboranes, chlorodialkylboranes and dichloroalkylboranes with organic azides follow the same mechanism yielding intermediate aminoboranes that, following hydrolysis, yield secondary amines.263-267

4.2.6 Reaction of Zn(C6F5)2 with Diazomethanes

The decomposition of diphenyldiazomethane has been studied extensively, dating as far back as 1891, by Curtius and Rauterberg.268 A range of both transition metal and main group Lewis acids have been successfully employed for the decomposition of Ph2C(N2) to products such as tetraphenylazine and tetraphenylethylene.269-275 Of particular note, zinc halides have also been 276 investigated for their reactivity toward the decomposition of Ph2C(N2). The lack of literature

precedence for the interaction between electrophilic diarylzinc species and Ph2C(N2) warranted further investigation.

A stoichiometric reaction of tol·Zn(C6F5)2 and Ph2C(N2) in CH2Cl2 was performed at room temperature. The diazomethane was consumed upon addition as evidenced by the disappearance

117

of the fuchsia colour. The resulting solution was pale yellow in colour, and upon removal of the volatiles yielded a solid, 4-25, in 94 % yield. Examination of the 19F NMR spectrum revealed three distinct resonances at -118.60, -157.68 and -163.61 ppm pertaining to two freely rotating 13 C6F5 rings bound to Zn. A singlet was observed in the C NMR spectrum at 178.28 ppm, indicating that the C=N functionality had remained intact. NMR spectroscopy alone was insufficient for the elucidation of the structure of 4-25, and an X-ray crystallographic study of 4-25 allowed for its assignment as the bis-Lewis acid tetraphenylazine adduct,

[(C6F5)2Zn(Ph2CNNCPh2)Zn(C6F5)2] (Figure 4.14).

Figure 4.14 – POV-Ray depictions of the molecular structure of 4-25 (left) and the azine-Zn2 core of 4-25 (right). C: black, N: aquamarine, F: pink, Zn: red-brown. H atoms removed for clarity in both structures. C6F5 rings removed for clarity (right). Selected bond distances (Å) and angles (°). Zn(1)-N(1), 2.1317(16); N(1)-N(1), 1.416(3); N(1)-C(13), 1.307(2); N(1)-N(1)- C(13), 117.62(19); C(13)-N(1)-Zn(1), 126.20(13); N(1)-N(1)-Zn(1), 115.19(13).

Decomposition of Ph2C(N2) in the presence of zinc halides has been reported in the literature as commonly producing both tetraphenylketazine and tetraphenylethylene in varying proportions during the Lewis-acid catalyzed reaction. Experimental evidence reveals that decomposition of

Ph2C(N2) at Zn followed by coupling with a second equivalent of the diazomethane yields the tetraphenylketazine (Scheme 4.18, a). Similar coupling reactions have been reported employing

118

group 5 and 10 metal complexes. Heyduk demonstrated the application of an amido-

bis(phenolate) tantalum (V) catalyst for the rapid coupling of two Ph2C(N2) moieties yielding a Ta-azine adduct (Scheme 4.18, b).277 Similarly, the group of Zanotta described the slow

conversion of two molecules of Ph2C(N2) to the corresponding azine following 3 days’ reaction with ethylenebis(triphenylphosphine)platinum(0).278

Scheme 4.18 – (a) Mechanism for the synthesis of 4-25. (b) Ketazine formation employing a amido-bis(phenolate) Ta(V) complex.

Interestingly, in the synthesis of 4-25, the ketazine is trapped by two equivalents of the Lewis acid with no evidence for the formation of tetraphenylethylene, the product of diphenylcarbene coupling, indicating that the former reaction is much faster as compared to the latter. An examination of the metrical parameters of 4-25 revealed an N−N bond length of 1.416(3) Å, in keeping with an N−N single bond. Similarly, the C−N bond length was found to be 1.307(2) Å, indicating retention of the C−N double bond. The N centres are slightly distorted from planarity

(angles summing to 359 °) with the two coordinated Zn(C6F5)2 fragments disposed in a mutually trans orientation about to the N−N single bond, in order to best minimize their steric interaction.

119

The Zn centres are best described as pseudo-planar with bond angles summing to 358°. The Zn centre is slightly removed from the plane defined by the ipso-carbons of the Zn-C6F5 rings and the N atom of the ketazine. It is of note that both Zn centres are found to be three-coordinate, which is a rare coordination geometry for the d-block metal, as it prefers to adopt either a linear two-coordinate, or a four- to six-coordinate geometry.279,280

Compound 4-25 is a Lewis diacid stabilized ketazine, but might also be referred to as a Lewis

diacid adduct and biscarbene stabilized N2 fragment (Figure 4.15). It was of interest to investigate whether the presence of Zn Lewis acid could initiate formal exchange of the

diphenylcarbene for a bulkier more basic carbene, trapping N2. Application of bulky N- heterocyclic carbenes for the isolation of small main group species has been thoroughly investigated in the past years.139,281-285 A recent example from the Bertrand group demonstrates

isolation of a P2 fragment from white phosphorous upon employing cyclic alkylamino carbenes as the trapping agents.286-288

Figure 4.15 – Electronic structures of 4-25.

A 1:2 reaction of 4-25 with 1,3-dimesitylimidazolidin-2-ylidene (SIMes) in toluene was undertaken at room temperature. Spectroscopic evidence indicated reaction of the SIMes fragment; however, this occurred independently from the azine fragment. It was suspected that upon the addition of the carbene to 4-25 that the Lewis acid was trapped by the carbene as the adduct, liberating the free azine. Independent stoichiometric reaction of Zn(C6F5)2 and SIMes yielded an adduct, 4-26, whose spectroscopic data was in keeping with the reaction of 4-25 with SIMes (Scheme 4.19). The 19F NMR spectrum displayed three resonances at -116.59, -158.49 and -161.92 ppm, only slightly shifted from those of 4-25, and pertained to two chemically 13 equivalent C6F5 rings. A downfield resonance was observed in the C NMR spectrum for the carbene C bound to zinc at 200.28 ppm.

120

Scheme 4.19 – Synthesis of 4-26 from 4-25 and from tol⋅Zn(C6F5)2.

It was unclear whether the solid-state structure of 4-26 would be monomeric or dimeric in nature, however a crystallographic study of 4-26 confirmed the formulation to be that of the monomer,

(C6F5)2Zn(SIMes) (Figure 4.16).

Figure 4.16 – POV-Ray depiction of the molecular structure of 4-26. C: black, N: aquamarine, F: pink, Zn: light steel blue. H atoms removed for clarity. Selected bond distances (Å) and angles (°). Zn(1)-C(7), 2.048(2); Zn(1)-C(1), 2.0045(15); Zn(1)-C(1), 2.0044(15); C(7)-N(1), 1.3281(16); C(7)-N(1), 1.3282(16); C(1)-Zn(1)-C(1), 131.46(9); C(1)-Zn(1)-C(7), 114.27(4)7; C(1)-Zn(1)-C(7), 114.27(4)7; N(1)-C(7)-N(1), 109.30(17); N(1)-C(7)-Zn(1), 125.35(8); N(1)- C(7)-Zn(1), 125.35(8).

The Zn centre was confirmed to be three-coordinate and planar in nature. The Zn atom lays flush within the plane defined by N(1), N(1) and C(7) with the angles at Zn summing to precisely 360°. Similarly, the carbene-carbon, bound to Zn is found to be planar, in keeping with the three-coordinate nature and sp2 hybridization.

121

4.2.7 Applications of Derivatized Boranes in FLP Chemistry and Beyond

Although the term “frustrated Lewis pairs” was only coined in 2006, researchers such as Brown, Wittig, Benz and Tochtermann have been reporting instances of “strange,” non-classical reactivity since as early as 1942. Since the formal introduction of the concept in 2006, the field of frustrated Lewis pair chemistry has garnered considerable attention and has grown in size. Numerous main group and transition metal systems have been developed and employed in a wide range of small molecule activations.1 A remaining challenge is the derivatization of such activated and sequestered molecules and their subsequent release from catalyst systems. As indicated in the preceding chapters, the ability to cleave strong B−X (X = O, S and N) bonds remains a challenge for synthetic chemists in this field. The approach adopted in this chapter was for the derivatization of boranes in such a way as to reduce the electrophilicity of the B centre and in some cases increase the steric demands about the Lewis acid. In fact, these boranes might find many applications in FLP chemistry as a result of this reduced Lewis acidity and increased steric demand.

As previously indicated, 4-1 is the product of Me3SiCH insertion into a B−C bond of B(C6F5)3.

The inclusion of the electron-donating Me3SiCH fragment rendered the B centre less Lewis

acidic relative to B(C6F5)3, the parent borane, but was suspected to be Lewis acidic enough to

enable heterolytic H2 activation in the presence of a phosphine. A stoichiometric reaction of

PtBu3 and 4-1 gave no spectroscopic evidence for adduct formation and was therefore stirred, in

toluene, under 4 atmospheres of H2 for two days at room temperature. Subsequent manipulation resulted in the recovery of a white solid, 4-27, in 83 % yield. The 1H NMR spectrum revealed a broad doublet at 5.17 ppm and a broad 1:1:1:1 quartet at 3.38 ppm in keeping with the 1 1 phosphonium PH and hydridoborate BH fragments, with JH-P and JH-B coupling constants of

431 and 94 Hz, respectively. Similarly, methine and Me3Si signals were observable at 2.82 and - 11 19 0.20 ppm confirming retention of the Me3SiCH fragment. The B and F NMR spectra were consistent with the generation of a hydridoborate anion in addition to the 31P NMR spectrum, which supported the presence of a phosphonium cation. Compound 4-27 was assigned as the

B/P H2 activation product [tBu3PH][HB(C6F5)2(Me3SiCH(C6F5))].

Reaction of 4-1 with other basic and bulky phosphines such as PMes3, P(o-tol)3 and 1,8- bis(diphenylphosphino)naphthalene also resulted in no adduct formation in solution but were

122

either extremely sluggish or inactive toward H2 activation. Due to the increased steric demands

of 4-1, H2 activation with bulky phosphines is likely to be challenging, as close approach of the Lewis acid and base may be sterically precluded. Recent computational evidence has indicated that the Lewis acid and base form an encounter complex in solution, thereby generating a pocket or cavity that can accommodate a small molecule. The substrate is then polarized within this pocket and activated by the frustrated Lewis pair.39,40,42,144 In the event that the Lewis acid and base are too large, insertion of the substrate may be completely obstructed, resulting in no 22 activation. PMes3 and P(o-tol)3 are found have Tolman cone angles of 212 and 194°, respectively, and are inactive toward H2 activation. When combined with 4-1, however, the basic

and less bulky PtBu3 with a cone angle of 182 ° readily generates the [PH][BH] salt. Unlike

[tBu3PH][HB(C6F5)3] and [Mes3PH][HB(C6F5)3], which fail to liberate H2 even when heated to 31 150 °C, 4-27 was found to release H2 upon heating to 100 °C as evidenced by the disappearance of the PH and BH signals by 1H NMR spectroscopy.

Attempts to activate H2 in the presence of PR3 (R = tBu, Mes, o-tol) with 4-2 and 4-5 proved to be unsuccessful, despite the availability of an FLP in solution. This lack of reactivity can be rationalized by employing a combination of both electronic and steric arguments. The incorporation of two Me3SiCH fragments drastically decreases the Lewis acidity at B while providing increased steric bulk, which may not allow adequate space for substrate approach.

As indicated in Chapter 2, frustrated Lewis pairs have the potential to facilitate the dehydrogenation of alcohols to ketones, releasing hydrogen or the hydrogenation of ketones to alcohols employing molecular hydrogen. In the reduction of a ketone to an alcohol, the B/P pair

must first heterolytically activate H2, generating the phosphonium hydridoborate salt. Insertion of the ketone into the B-H bond of the borohydride follows. The alcohol is finally liberated via protonation of the alkoxyborate by the phosphonium cation (Scheme 4.20). Partial reductions of carbonyl-containing organic compounds have already been achieved, terminating with the formation of phosphonium alkoxyborates.1

123

Scheme 4.20 – Proposed catalytic cycle for the reduction of ketones.

The limiting step in the cycle involves the cleavage of the alcohol fragment from the borane. It was hypothesized that this cleavage might be facilitated employing 4-1, due to the added steric bulk incorporated about the boron centre which would lengthen the resulting B−O bond to the alkoxide fragment. The ability of 4-27 to release both hydride and proton further supported the potential to successfully reduce ketones.

Initial studies revealed that ketones bearing small substituents were not suitable for reduction due to the propensity of these species to quaternize the boron centre, thereby poisoning the frustrated

Lewis pair toward further H2 activation. Benzophenone and its derivatives were found to be adequate candidates in this chemistry as they demonstrated only slight or no interaction with the B centre.

A bomb was charged with stoichiometric quantities of 4-1, PtBu3 and benzophenone and charged with 4 atmospheres of hydrogen. The mixture was left stirring in an oil bath for 24 hours at 80 °C after which time an aliquot was examined by 1H NMR spectroscopy. The 1H NMR spectrum confirmed the presence of a new CH signal at 5.70 ppm characteristic of a methine proton resulting from attack of the hydridoborate species at the carbonyl carbon. A broad signal, observed at 2.90 ppm was thought to correspond to the OH of the liberated alcohol, however, this signal was found to integrate to 2H. A resonance was observed in the 31P NMR spectrum pertaining to a HP fragment, and a sharp peak in the borohydride region of the 11B NMR spectrum was observed, pertaining to a BH fragment. These data were in accordance with capture of a second molecule of H2 by the catalyst following ketone reduction. Subsequent

124

manipulation of the reaction mixture resulted in the isolation of a white solid, 4-28, in 77 % yield. Analysis of the solid material by 1H NMR spectroscopy revealed the characteristic broad doublet and a 1:1:1:1 quartet assignable to the PH and BH ion pair at 5.17 and 2.77 ppm, respectively. The tBu groups were represented by a sharp doublet at 1.64 ppm. There was a distinct absence of signals for the methine and the Me3Si fragments, indicating that these moieties had been cleaved during reaction. Additionally, there were no signals pertaining to the alkoxide fragment. Together, these data, in addition to elemental analysis, confirmed the identity

of 4-28 as the ion pair [tBu3PH][HB(C6F5)2(CH2(C6F5))].

Scheme 4.21 – Mechanism for the synthesis of 4-28.

Mechanistically, activation of molecular H2 by the B/P FLP is the initial step followed by hydride transfer from the hydridoborate ion to the carbonyl species. As opposed to the desired protonation of the alkoxide fragment by phosphonium, the borate undergoes elimination of silane. Deprotonation of the phosphonium ion by the carbon adjacent to boron yields the new borane, (C6F5)2B(CH2(C6F5)), which initiates the heterolytic cleavage of a second equivalent of

H2 with PtBu3 to give 4-28. Unfortunately, this same deactivation pathway was observed when other basic and bulky benzophenone derivatives were employed (Scheme 4.21).

125

Tris(pentafluorophenyl)borane has previously been utilized for the reduction of aldimines employing molecular hydrogen as the source of proton and hydride.45 This reduction is known to be catalytic but has some inherent limitations, namely the reduction of sterically unhindered or electron-poor imines is sluggish or does not proceed at all. It was thought that the reduced Lewis acidity and steric bulk of 4-1 might assist in the cleavage of the product amine from the B centre following hydrogenation, and was therefore investigated for use in reduction chemistry (Scheme 4.22).

Scheme 4.22 – Reduction of N-benzylidene-tert-butylamine employing H2 and 4-1.

A solution of N-benzylidene-tert-butylamine in toluene was prepared and charged with 10 mol %

4-1. The vessel was pressurized with 4 atmospheres of H2 and heated in an oil bath at 80 °C for a period of 30 hours. 80 °C was determined to be the optimal reaction temperature, as higher temperatures resulted in the thermally-induced decomposition of the borane via pentafluorobenzene liberation. The process of reduction to N-benzyl-tert-butylamine was slow, and conversions determined by 1H NMR spectroscopy are listed in Table 4.1. The 1H NMR spectrum showed the gradual appearance of two signals at 3.53 and 1.78 ppm pertaining to the 45 CH and NH signals of the amine. Compared to B(C6F5)3, conversion to the fully reduced amine by 4-1 was slow, and further hydrogenations were not studied for this reason.

126

Table 4.1 – Conversion of N-benzylidene-tert-butylamine to N-benzyl-tert-butylamine over a 30

hour time period, employing 4-1 and H2 in toluene (4 atmospheres, 80 °C).

Time (hours) % Conversion 00 1 2.85 3 9.09 5 12.7 7 16.0 9 20.1 24 69.9 30 100

It has previously been elucidated, that varying combinations of B/P frustrated Lewis pairs can 289 undergo fluorination with xenon difluoride, XeF2. The initial step is oxidative fluorination of

the phosphine generating a phosphorous (V) species PR3F2, with concomitant reduction of Xe(II)

to xenon gas. The borane, B(C6F5)3, serves to abstract a fluoride ion from the PR3F2 intermediate, yielding the fluorophosphonium fluoroborate salt [R3PF][FB(C6F5)3]. It was of interest to elucidate whether analogous reactivity might be accessible employing the more sterically hindered boranes, 4-1 and 4-2.

An equivalent of XeF2 was added to a stoichiometric combination of 4-1 and PtBu3 at room temperature in CH2Cl2 and was stirred for an hour. Following removal of the volatiles and product manipulation with cold pentane, a white solid, 4-29, was recovered in 96 % yield (Scheme 4.23). Examination of the 31P{1H} NMR spectrum supported the presence of the 1 fluorophosphonium cation by a broad doublet at 147.56 ppm with a JP-F coupling constant of 1019 Hz. Similarly, a doublet was observed in the 11B{1H} NMR spectrum at -3.23 ppm, 1 19 diagnostic of a fluoroborate anion with JB-F coupling of 75 Hz. The F NMR spectrum of 4-29

proved complicated. The C6F5 rings at B were chemically inequivalent and presented as six 19 distinct signals in this spectrum while the CC6F5 group gave rise to two o-F signals in the F NMR spectrum, indicating restricted rotation about the B-C bond. One signal was shifted

127

considerably downfield (-136.49 ppm) relative to the second signal (-144.94 ppm) and displayed

a complex multiplet. It is likely that the Me3SiCH(C6F5) fragment is restricted in such a fashion

as to orient an o-F of the C6F5 ring in close proximity to the fluoride atom on boron. Notable features of the 19F NMR spectrum are the P−F and B−F signals at -172.52 and -182.26 ppm, which present as a broad doublet and singlet, respectively. Together with elemental analysis the

formulation of 4-29 was proposed as [tBu3PF][FB(C6F5)2(Me3SiCH(C6F5))].

Scheme 4.23 – Synthesis of compounds 4-29 and 4-30.

An analogous reaction was carried out employing meso 4-2. Following manipulation of the reaction mixture, an analytically pure, white solid, 4-30, was isolated in 88 % yield. 31 1 Examination of the P{ H} NMR spectrum confirmed the presence of tBu3PF as a characteristic 1 doublet at 147.59 ppm, with JP-F of 1018 Hz (Figure 4.18). Two doublets, at 6.59 and 3.56 ppm, were observed in the 11B{1H} NMR spectrum consistent with a set of diastereomers. Consequently, the 19F NMR spectrum was complicated, however, the characteristic F-P and F-B signals were easily assignable supporting the formulation of 4-30 as

[tBu3PF][FB(C6F5)(Me3SiCH(C6F5))2]. The ability to accept a fluoride ion is quite remarkable considering the steric bulk about the B centre, as 4-2 is incapable of forming a stable, isolable adduct with acetonitrile. Crystals suitable for X-ray crystallography confirmed the formulation of 4-30 as the [FP][BF] salt (Figure 4.17).

128

Figure 4.17 – POV-Ray depiction of the molecular structure of 4-30. C: black, B: yellow-green, F: pink, Si: blue, P: orange. Selected H atoms removed for clarity. Selected bond distances (Å). B(1)-C(17), 1.676(4); B(1)-C(1), 1.660(4); B(1)-F(16), 1.434(3); B(1)-C(1), 1.693(4); C(17)- C(21), 1.505(3); C(1)-C(5), 1.509(3); P(1)-F(17), 1.5712(15).

129

Figure 4.18 – (a) Full 19F NMR spectrum of 4-30. Expansion of the (b) o-F, (c) P-F and B-F and (d) p-F and m-F regions of the 19F NMR spectrum of 4-30.

130

4.3 Conclusions

A facile and high-yielding methodology for the derivatization of B−C, B−H and B−Cl bonds using diazomethanes has been developed and employed in the derivatization of triaryl- and diarylboranes, borinic esters and boronic acids. The newly synthesized boranes are generated in moderate to high yields and offer clean products upon solvent removal requiring little or no purification. This methodology offers access to synthetically complex boranes in mere hours and avoids the application of toxic tin and mercury intermediates commonly used in literature preparations. This strategy offers the ability to expand the library of boron-containing synthons that commonly find application in the field of organic chemistry such as in C−C bond forming

reactions like Suzuki cross-couplings. The sterically demanding B(C6F5)3 derivatives have further potential to act as co-catalysts for the activation of early metal polymerization catalysts generating large, non-coordinating anions.

4.4 Experimental Section

4.4.1 General considerations

All manipulations were carried out under an atmosphere of dry, O2-free N2 employing standard

Schlenk-line and glovebox techniques. Solvents (pentane, CH2Cl2, THF, toluene) were dried by employing a Grubbs-type column system (Innovative Technology), degassed and stored under

N2. CD2Cl2 was vacuum transferred from CaH2, degassed and stored under N2 while C6D5Br was

dried over CaH2 and distilled under N2. Pyridine was dried over CaH2 and distilled under N2.

B(C6F5)3 (Boulder Scientific), BPh3 (Aldrich), phenylboronic acid (Aldrich), 4- fluorophenylboronic acid (Aldrich), pentafluorophenylboronic acid (Aldrich), catecholborane (Aldrich), trimethylsilylazide (Aldrich), benzylazide (Alfa Aesar), xenon difluoride (Alfa Aesar),

chlorodimethylsilane (Aldrich), triethylsilane (Aldrich), PtBu3 (Strem Chemicals), and trimethylsilyldiazomethane (2.0 M in hexanes) (Aldrich) were used as received. Commercially

available, yet impure Zn(C6F5)2 (Aldrich) was taken up in toluene, filtered through a plug of Celite and evacuated yielding the toluene adduct of the Lewis acid as a solid. Commercially available, yet impure B-chlorocatecholborane (Aldrich), was taken up in pentane, filtered through a plug of Celite to remove an insoluble black contaminant and evacuated yielding the

131

126 210 154 pure Lewis acid as a white solid. Ph(C6F5)2, ClB(C6F5)2, (C6F5)2B(OC6F5), 149 257 210 248 290 (C6H4O2)BC6F5, (C6H4O2)BC6H4F, HB(C6F5)2, Ph2C(N2) and SIMes were prepared according to literature procedures. It should be noted that conventional methods for the preparation of pentafluorophenyldiazomethane require pyrolysis of the related tosylhydrazone generating the associated diazomethane. A safer and more gentle synthetic route, shown to be

successful in the synthesis of Ph2C(N2) was however, employed in the synthesis of 291 (C6F5)CH(N2). CAUTION. It should be noted that diazo compounds can be quite unstable. Proper eye and face protection is required when handling these species. 1H, 11B, 13C, 19F, 29Si and 31P NMR spectra were recorded at 25 °C, unless otherwise stated, on a Varian NMR Mercury System 300 MHz, Varian NMR System 400 MHz or Bruker Avance III 400 MHz 1 13 spectrometer and were referenced using (residual) solvent resonances relative to SiMe4 ( H, C, 29 11 19 31 Si), or relative to an external standard ( B: (Et2O)BF3, F: CFCl3, P: 85% H3PO4). Chemical shifts are reported in ppm and coupling constants as scalar values in Hz. Combustion analyses were performed in house employing a Perkin-Elmer CHN Analyzer. High-resolution mass spectrometry was performed in house employing electrospray ionization.

4.4.2 Syntheses

Synthesis of 4-1. B(C6F5)3 (0.200 g, 0.391 mmol) in CH2Cl2 (10 mL) was

cooled to -78 °C where a solution of Me3SiCH(N2) (2.0 M solution in

hexanes) (235 μL, 0.470 mmol), diluted in further CH2Cl2 (3 mL) was added, in a dropwise fashion. The resulting colourless solution was stirred at -78 °C for a period of 30 minutes and was warmed to r.t. where the reaction was stirred for a further hour. The volatiles were then removed in vacuo, yielding a sticky tan solid. The residue was taken up in pentane (5 mL) and was placed in the freezer at -35 °C. Colourless crystals were recovered after decanting the cold pentane. The product was washed with cold pentane (2 x 2 mL) and dried in vacuo. Crystals suitable for X-ray diffraction were grown from a pentane solution at -35 °C. Yield: 1 11 1 0.152 g (65 %). H NMR (CD2Cl2): δ 4.38 (s, 1H, CH); 0.16 (s, 9H, Si(CH3)3). B{ H} NMR 13 1 (CD2Cl2/C6D5Br, -60 to 85 °C): no observable signal. C{ H} NMR (CD2Cl2) partial: δ 145.89 1 1 1 (br d, JC-F = 240 Hz, C6F5); 144.73 (br d, JC-F = 240 Hz, C6F5); 143.26 (br d, JC-F = 260 Hz, 1 1 C6F5); 139.27 (br d, JC-F = 251 Hz, C6F5); 138.05 (br d, JC-F = 253 Hz, C6F5); 115.62 (br s, 19 C6F5); 114.09 (br s, C6F5); 41.12 (s, CH); 0.91 (s, Si(CH3)3). F NMR (CD2Cl2): δ -130.87 (s, 3 3 4F, o-BC6F5); -140.75 (d, 2F, JF-F = 22 Hz, o-CC6F5); -149.83 (t, 2F, JF-F = 20 Hz, p-BC6F5); -

132

3 158.91 (t, 1F, JF-F = 21 Hz, p-CC6F5); -161.27 (m, 4F, m-BC6F5); -162.88 (m, 2F, m-CC6F5). 29 Si NMR (CD2Cl2): δ -110.97 (br s). Anal. Calcd. for C22H10BF15Si: C, 44.14; H, 1.69. Found: C, 44.17; H, 1.84 %.

Synthesis of 4-2. B(C6F5)3 (0.200 g, 0.391 mmol) in CH2Cl2 (10 mL) was

cooled to -78 °C where a solution of Me3SiCH(N2) (2 M solution in hexanes)

(586 μL, 1.17 mmol) diluted in further CH2Cl2 (3 mL) was added, in a dropwise fashion. The resulting pale yellow solution was stirred at -78 °C for 30 minutes and warmed to room temperature where the reaction was stirred for a further hour. The volatiles were then removed yielding a sticky pale yellow solid. The residue was taken up in pentane (5 mL) and was placed in the freezer at -35 °C. Colourless crystals were recovered after decanting the cold pentane. The product was then washed with cold pentane (2 x 2 mL) and dried in vacuo. Crystals suitable for X-ray diffraction were obtained from a pentane solution at -35 °C. Yield: 0.189 g (71 %). Product distribution: (S,S/R,R) 72 %, (S,R) 28 %. (S,S/R,R) 1H NMR 1 (CD2Cl2): δ 3.79 (s, 1H, CH); 0.22 (s, 9H, Si(CH3)3). (S,R) H NMR NMR (CD2Cl2): δ 3.58 (s, 11 1 1H, CH); -0.17 (s, 9H, Si(CH3)3). B{ H} NMR (CD2Cl2): δ 75.44 (br s). (S,S/R,R and S,R) 13 1 1 1 C{ H} NMR (CD2Cl2) partial: δ 144.82 (d, JC-F = 239 Hz, C6F5); 144.25 (d, JC-F = 244 Hz, 1 1 C6F5); 138.91 (d, JC-F = 253 Hz, C6F5); 138.12 (d, JC-F = 244 Hz, C6F5); 116.36 (br s, ipso-

C6F5); 115.14 (br s, ipso-C6F5); 35.88 (s, CHR,R/S,S); 35.00 (s, CHRS), 0.95 (s, Si(CH3)3 R,R/S,S); 19 3 0.48 (s, Si(CH3)3 S,R). (S,S/R,R) F NMR (CD2Cl2, -85 °C): δ -131.71 (d, 2F, JF-F = 20 Hz, o- 3 3 BC6F5); -140.53 (d, 2F, JF-F = 23 Hz, o-CC6F5); -143.06 (d, 2F, JF-F = 22 Hz, o-CC6F5); -151.27 3 3 3 (t, 1F, JF-F = 21 Hz, p-BC6F5); -159.35 (t, 2F, JF-F = 22 Hz, 2 x p-CC6F5); -160.47 (t, 2F, JF-F = 3 3 21 Hz, m-C6F5); -161.91 (t, 2F, JF-F = 22 Hz, m-C6F5); -162.99 (t, 2F, JF-F = 22 Hz, m-C6F5). 19 3 3 (S,R) F NMR (CD2Cl2, -85 °C): δ -130.27 (d, 1F, JF-F = 21 Hz, o-BC6F5); -131.91 (d, 1F, JF-F 3 3 = 26 Hz, o-BC6F5); -137.53 (d, 1F, JF-F = 18 Hz, o-CC6F5); -138.59 (d, 1F, JF-F = 21 Hz, o- 3 3 CC6F5); -141.74 (d, 1F, JF-F = 22 Hz, o-CC6F5);-143.35 (d, 1F, JF-F = 24 Hz, o-CC6F5); -150.78 3 3 (d, 1F, JF-F = 21 Hz, p-BC6F5); -158.10 (d, 2F, JF-F = 22 Hz, p-CC6F5); -159.61 (m, 2F, m- 3 3 C6F5); -160.29 (m, 2F, m-C6F5); -160.74 (t, 2F, JF-F = 20 Hz, m-C6F5); -161.59 (t, 2F, JF-F = 20 3 3 29 Hz, m-C6F5); -162.12 (2F, JF-F = 22 Hz, m-C6F5); -163.49 (t, 2F, JF-F = 23 Hz, m-C6F5). Si

NMR (CD2Cl2): δ -109.34 (br s). Anal. Calcd. for C26H20BF15Si2: C, 45.61; H, 2.95. Found: C, 45.67; H, 3.14 %.

133

Synthesis of 4-3. A 500 mL Schlenk was charged with dry DMSO (0.74 mL, 10.4

mmol) in CH2Cl2 (160 mL). The solution was cooled to -60 °C at which time oxalyl chloride (0.85 mL, 10.0 mmol) was added dropwise. The reaction mixture was maintained at -60 °C and stirred until no further gas evolution was observed. A 100 mL

Schlenk was then charged with pentafluorophenylhydrazone (2.00 g, 9.52 mmol), Et3N (2.79 mL, 20.00 mmol) and THF (40 mL). This solution was added in a dropwise fashion to the solution of the preformed sulfonium that had been further cooled to -78 °C. The reaction became deep red-orange upon addition of the hydrazone solution and was stirred at -78 °C for an additional hour then warmed to room temperature. The volatiles were then carefully transferred to a second flask via a flask-to-flask distillation. Careful removal of the solvent is required to prevent transfer of the product. Note that a small quantity of the yellow diazomethane is always transferred over with the solvent. The resulting yellow oil was taken up in pentane, passed through a bed of basic alumina and eluted with excess pentane. Relatively rapid elution is required to prevent product decomposition on the column. The solvents were then removed in a manner parallel to the initial steps. The pure diazomethane was then recovered as a vibrant yellow oil. The product was found to be stable for a prolonged period of time when stored at -35

°C under an atmosphere of N2. In the above synthesis, CH2Cl2 can easily be substituted for THF, similarly trap-to-trap removal of the solvent can be substituted for the direct removal of the solvent using convention distillation techniques. Yield 0.600 g (30 %). 1H and 19F NMR spectroscopic data was in keeping with previously reported data.

Synthesis of 4-4. B(C6F5)3 (0.267 g, 0.912 mmol) in CH2Cl2 (20 mL) was

cooled to -78 °C where (C6F5)CH(N2) (0.190 g, 0.913 mmol) in CH2Cl2 (10 mL) was added in a dropwise fashion. The yellow colour of the diazomethane was consumed immediately upon addition. The solution was then warmed to r.t. and allowed to stir for an additional 30 minutes at which point the volatiles were removed revealing a beige oil. The residue was washed with cold pentane (3 x 5 mL) yielding a beige coloured solid. The product was subsequently dried in vacuo. Crystals suitable for X-ray crystallography were grown 1 from a layered solution of CH2Cl2 and pentane at -35 °C. Yield: 0.379 g (60 %). H NMR 11 1 13 1 (CD2Cl2): δ 5.28. B{ H} NMR (CD2Cl2): δ 72.34. C{ H} NMR (CD2Cl2) partial: δ 146.42 1 1 1 (br d, JC-F = 243 Hz, C6F5); 143.43 (br d, JC-F = 243 Hz, C6F5); 142.58 (br d, JC-F = 249 Hz, 1 1 C6F5); 140.27 (br d, JC-F = 246 Hz, C6F5); 137.098 (br d, JC-F = 253 Hz, C6F5); 114.09 (br s,

134

19 C6F5); 113.88 (br s, C6F5). F NMR (CD2Cl2): δ -129.68 (s, 4F, o-B(C6F5)); -140.04 (s, 4F, o-

C(C6F5)); -146.22 (s, 2F, p-B(C6F5)); -155.24 (s, 2F, p-C(C6F5)); -161.10 (s, 4F, m-C6F5);

-162.41 (s, 4F, m-C6F5). Due to compound instability, attempts to obtain elemental analysis and high-resolution mass spectrometry were unsuccessful.

Synthesis of 4-5. PhB(C6F5)2 (0.300 g, 0.788 mmol) in CH2Cl2 (10 mL) was

cooled to -78 °C and a solution of Me3SiCH(N2) (2 M in hexanes) (789 μL,

1.58 mmol), diluted in further CH2Cl2 (3 mL) was added to the borane solution in a dropwise fashion. The solution was left stirring at -78 °C for an additional twenty minutes and was then warmed to r.t. where it was stirred for a further 2 hours.

During this time the yellow colour of the Me3SiCH(N2) disappeared, indicative of consumption of the reagent. After two hours of stirring, the volatiles were removed from the clear and colourless solution yielding a colourless oil. The oil was taken up in a small quantity of pentane (2 mL) and filtered through a plug of Celite. The filtrate was collected in a vial where the bottom had been thoroughly scratched. Over a number of days large, clear and colourless crystals separated from the solution, which was stored at -35 °C. This method was employed to grow crystals suitable for X-ray diffraction. Yield 0.302 g (64 %). Stereoselective for (R,R/S,S) 1 only. H NMR (CD2Cl2): δ 7.19-7.04 (m, 3H, C6H5); 6.77 (br s, 2H, C6H5); 3.70 (s, 1H, CH); 11 1 3.53 (s, 1H, CH); 0.17 (s, 9H, Si(CH3)3); 0.16 (s, Si(CH3)3). B{ H} NMR (CD2Cl2): δ 76.31 13 1 1 1 (br s). C{ H} NMR (CD2Cl2) partial: δ 143.92 (br d, JC-F = 244 Hz, C6F5); 142.80 (br d, JC-F 1 1 = 240 Hz, C6F5); 140.01 (s, ipso-C6H5); 127.91 (br d, JC-F = 238 Hz, C6F5); 137.28 (br d, JC-F = 1 254 Hz, C6F5); 136.58 (br d, JC-F = 251 Hz, C6F5); 128.11 (s, o-C6H5); 127.71 (s, m-C6H5);

124.49 (s, p-C6H5); 117.32 (m, ipso-C6F5); 51.25 (s, CH(C6H5)); 34.75 (s, CH(C6F5)); 0.83 (s, 19 3 4 Si(CH3)3); -0.34 (s, Si(CH3)3). F NMR (CD2Cl2): δ -128.69 (dd, 1F, JF-F = 25 Hz, JF-F = 8.9 3 4 3 Hz, o-BC6F5); -134.36 (dd, 1F, JF-F = 25 Hz, JF-F = 8.9 Hz, o-BC6F5); -139.49 (d, 1F, JF-F = 23 3 4 3 Hz, o-CC6F5); -143.21 (dd, 1F, JF-F = 24 Hz, JF-F = 7.6 Hz, o-CC6F5); -153.72 (t, 1F, JF-F = 21 3 3 3 Hz, p-BC6F5); -160.41 (t, 1F, JF-F = 22 Hz, p-CC6F5); -161.95 (ddd, 1F, JF-F = 23 Hz, JF-F = 21 4 3 3 4 Hz, JF-F = 9.1 Hz, m-BC6F5); -162.26 (ddd, 1F, JF-F = 24 Hz, JF-F = 21 Hz, JF-F = 9.6 Hz, m- 3 4 3 BC6F5); -163.07 (td, 1F, JF-F = 23 Hz, JF-F = 7.6 Hz, m-CC6F5); -164.02 (td, 1F, JF-F = 23 Hz, 4 29 JF-F = 7.5 Hz, m-CC6F5). Si NMR (CD2Cl2): δ -111.50 (br s). Anal. Calcd. for C26H25BF10Si2: C, 52.51; H, 4.24. Found: C, 52.31; H, 4.42 %.

135

Synthesis of 4-6. BPh3 (0.120 g, 0.496 mmol) in CH2Cl2 (10 mL) was

cooled to -78 °C where Me3SiCH(N2) (2.0 M in hexanes) (248 μL, 0.496

mmol) in additional CH2Cl2 (3 mL) was added in a dropwise fashion. The solution was then warmed to r.t. and was stirred for an additional hour. At this time, the volatiles were removed in vacuo yielding a clear and colourless oil. The oil was taken up in a small portion of pentane (5 mL) and pyridine (40 μL, 0.497 mmol) was added

precipitating a white solid. Crystals were obtained from a layered solution of CH2Cl2 and pentane at -35 °C over a number of weeks, although they were not of adequate quality for X-ray 1 3 4 crystallography. Yield: 0.152 g, (75 %). H NMR (CD2Cl2): δ 8.45 (d, 2H, JH-H = 6.2 Hz, JH-H 3 4 = 1.8 Hz, o-C5H5N); 7.78 (tt, 1H, JH-H = 7.6 Hz, JH-H = 1.5 Hz, p-C5H5N); 7.27 (m, 4H, m- 3 3 C5H5N/C6H5); 7.28-7.19 (m, 8H, C6H5); 6.98 (t, 2H, JH-H = 7.2 Hz, o-CC6H5); 6.89 (tt, 1H, JH-H 4 3 = 7.2 Hz, JH-H = 1.5 Hz, p-CC6H5); 6.82 (d, 2H, JH-H = 7.7 Hz, m-CC6H5); 2.69 (s, 1H, CH); - 11 1 13 1 0.27 (s, 9H, SiMe3). B{ H} NMR (CD2Cl2): δ 11.98 (s). C{ H} NMR (CD2Cl2) partial: δ

147.99 (s, ipso-BC6H5); 147.51 (s, C5H5N); 139.80 (s, C5H5N); 135.10 (s, BC6H5); 131.26 (s,

CC6H5); 127.90 (s, BC6H5); 127.55 (s, BC6H5); 126.30 (s, C5H5N); 124.83 (s, CC6H5); 123.29 (s,

CC6H5); 1.21 (s, Si(CH3)3). Anal. Calcd. for C27H30BNSi: C, 79.57; H, 7.42; N, 3.44. Found: C, 79.05; H, 7.24; N, 3.51 %.

Synthesis of 4-7. Ph2C(N2) (0.051 g, 0.263 mmol) in CH2Cl2 (5 mL) was

added in a dropwise fashion to ClB(C6F5)2 (0.100 g, 0.263 mmol) in CH2Cl2 (5 mL). Slow addition was necessary to prevent overflow of the reaction vessel

as the evolution of N2 was noted to be extremely vigorous. Directly following

addition, the fuchsia colour of the Ph2C(N2) solution was consumed yielding a pale yellow reaction mixture. The resulting solution was allowed to stir at r.t. for a period of 30 minutes. At this time, the volatiles were removed yielding a yellow oil. Crystals were grown in a small portion of pentane (2 mL) over a number of days at -35 °C but were not of adequate quality for 1 X-ray diffraction. Yield: 0.138 g (97 %). H NMR (CD2Cl2): δ 7.48-7.26 (m, 10H, C6H5). 11 1 13 1 1 B{ H} NMR (CD2Cl2): δ 60.31 (s). C{ H} NMR (CD2Cl2): δ 146.06 (br d, JC-F = 247 Hz, 1 1 1 C6F5); 145.59 (br d, JC-F = 247 Hz, C6F5); 143.06 (br d, JC-F = 247 Hz, C6F5); 141.04 (br d, JC-F 1 1 = 254 Hz, C6F5); 138.86 (br d, JC-F = 245 Hz, C6F5); 138.11 (br d, JC-F = 251 Hz, C6F5); 136.69 2 (s, C6H5); 131.64 (s, C6H5); 129.20 (s, C6H5); 128.76 (s, C6H5); 128.85 (t, JC-F = 15 Hz, CC6F5); 19 113.22 (br s, C6F5); 57.06 (s, C(Ph2(C6F5))). F NMR (CD2Cl2, -85 °C): δ -126.73 (m, 2F, o-

136

3 4 BC6F5); -130.26 (m, 2F, o-CC6F5); -151.51 (tt, 1F, JF-F = 20Hz, JF-F = 4.5 Hz, p-BC6F5); - 3 157.22 (t, 1F, JF-F = 21 Hz, p-CC6F5); -162.81 (m, 4F, m-BC6F5/m-CC6F5). Due to compound instability, attempts to obtain elemental analysis and high resolution mass spectrometry were unsuccessful.

Synthesis of 4-8. 4-7 was first prepared employing Ph2C(N2) (0.051 g, 0.263

mmol) and ClB(C6F5)2 (0.100 g, 0.263 mmol). Neat 4-7 was taken up in pentane (10 mL) and filtered through a plug of Celite. At this time, pyridine (100 μL, 1.24 mmol) was added generating an immediately opaque solution. The reaction was allowed to stir for an hour at r.t. and was then filtered through a sintered glass frit to collect the solid. The product was washed with pentane (2 x 5 mL) and dried in vacuo.

Crystals suitable for X-ray diffraction were grown at -35 °C from a layered solution of CH2Cl2 1 3 and pentane. Yield: 0.135 (82 %). H NMR (CD2Cl2): δ. 8.09 (t, 1H, JH-H = 7.1 Hz, p-C5H5N); 11 1 7.62 (br s, 2H, C5H5N); 7.34-7.07 (m, 10H, C6H5); 6.74 (br s, 2H, C5H5N). B{ H} NMR 13 1 1 (CD2Cl2): δ 6.30 (s). C{ H} NMR (CD2Cl2) partial: δ 148.92 (br d, JC-F = 249 Hz, C6F5); 1 148.75 (br s, C5H5N); 147.44 (br d, JC-F = 248 Hz, C6F5); 143.38 (br s, C5H5N); 141.05 (br d, 1 1 1 JC-F = 239 Hz, C6F5); 138.63 (br d, JC-F = 248 Hz, C6F5); 137.91 (br d, JC-F = 249 Hz, C6F5); 19 130.35 (br s, C6H5); 130.03 (br s, C6H5); 127.57 (br s, C6H5); 124.54 (br s, C5H5N). F NMR 3 (CD2Cl2, -75 °C): δ -122.49 (d, 1F, JF-F = 25 Hz, o-C6F5); -122.66 (s 1F, o-C6F5); -123.56 (s, 1F, 3 3 o-C6F5); -125.56 (d, 1F, JF-F = 21 Hz, o-C6F5); -156.96 (t, 2F, JF-F = 22 Hz, p-C6F5); -158.05 (t, 3 2F, JF-F = 22 Hz, p-C6F5); -163.84 (m, 4F, m-C6F5); -164.40 (m, 2F, m-C6F5); -164.70 (m, 2F, m-

C6F5). Due to compound instability, attempts to obtain elemental analysis and high resolution mass spectrometry were unsuccessful.

Synthesis of 4-9. 4-7 was first prepared employing Ph2C(N2) (0.092 g, 0.474

mmol) and ClB(C6F5)2 (0.120 g, 0.315 mmol). Neat 4-7 was taken up in

CH2Cl2 (5 mL) where triethylphosphine oxide (0.042 g, 0.313 mmol) in

CH2Cl2 (2 mL) was added generating a light yellow solution. The volatiles were removed yielding a yellow oil. Trituration with pentane yielded an off-white solid which 1 3 was subsequently dried in vacuo. Yield: 0.162 g (76 %). H NMR (CD2Cl2): δ 7.88 (d, 4H, JH-H

= 8.0 Hz, o-C6H5); 7.11 (m, 4H, m-C6H5); 7.01 (m, 2H, p-C6H5); 2.18 (m, 6H, PCH2CH3); 1.06 3 3 11 1 13 1 (dt, 9H, JH-P = 19 Hz, JH-H = 7.8 Hz, PCH2CH3). B{ H} NMR (CD2Cl2): δ 6.79 (s). C{ H}

137

1 NMR (CD2Cl2) partial: δ 149.21 (s, ipso-C6H5); 148.00 (br s, JC-F = 243 Hz, C6F5); 147.33 (s, 1 1 ipso-C6H5); 146.87 (br s, JC-F = 248 Hz, C6F5); 139.87 (br s, JC-F = 246 Hz, C6F5); 138.24 (br s, 1 1 JC-F = 246 Hz, C6F5); 136.90 (br s, JC-F = 239 Hz, C6F5); 129.07 (s, C6H5); 127.84 (s, C6H5); 1 127.49 (s, C6H5); 125.23 (s, C6H5); 125.16 (s, C6H5); 123.09 (br s, ipso-C6F5); 17.81 (d, JC-P = 2 19 3 65 Hz, PCH2); 5.63 (d, JC-P = 4.3 Hz, PCH2CH3). F NMR (CD2Cl2): δ -122.65 (d, 2F, JF-F = 3 21 Hz, o-BC6F5); -130.79 (br s, 2F, o-CC6F5); -158.76 (t, 1F, JF-F = 21 Hz, p-C6F5); -160.91 (t, 3 31 1 1F, JF-F = 22 Hz, p-C6F5); -166.12 (m, 4F, m-C6F5). P{ H} NMR (CD2Cl2): δ 83.75 (s). Due to compound instability, attempts to obtain elemental analysis and high resolution mass spectrometry were unsuccessful.

Synthesis of 4-10. 4-7 was first prepared employing ClB(C6F5)2 (0.120 g,

0.315 mmol) and Ph2C(N2) (0.061 g, 0.314 mmol). Neat 4-7 was taken up in chlorodimethylsilane (2 mL, 18.0 mmol) and stirred for four hours. At the reaction completion the reaction was clear and colourless. The volatiles were then removed yielding a white microcrystalline material. The product was subsequently dried in vacuo for an hour. Crystals were grown from a solution of pentane at -35 °C but were not however, adequate 1 for X-ray diffraction. Yield: 0.160 g (99 %). H NMR (C6D6): δ 7.04-6.97 (m, 8H, C6H5); 6.93- 11 1 13 1 6.89 (m, 2H, C6H5); 4.95 (s, 1H, CH). B{ H} NMR (C6D6): δ 75.94 (s). C{ H} (C6D6): δ 1 1 146.40 (br d, JC-F = 252 Hz, o-C6F5); 143.58 (br d, JC-F = 242 Hz, p-C6F5); 141.91 (s, ipso- 1 C6H5); 137.91 (br d, JC-F = 254 Hz, m-C6F5); 130.10 (s, C6H5); 129.34 (s, C6H5); 127.33 (s, 19 3 C6H5); 115.02 (br s, ipso-C6F5); 59.08 (s, CH). F NMR (C6D6): δ -130.58 (dd, 4F, JF-F = 19 4 3 Hz, JF-F = 6.5 Hz, o-C6F5); -147.95 (t, 2F, JF-F = 21 Hz, p-C6F5); -161.79 (m, 4F, m-C6F5). Anal.

Calcd. for C25H11BF10: C, 58.63; H, 2.16. Found: C, 58.46; H, 2.48 %.

Synthesis of 4-11. (C6F5)2B(OC6F5) (0.120 g, 0.227 mmol) in CH2Cl2 (10

mL) was cooled to -78 °C where Me3SiCH(N2) (125 μL of a 2 M solution, 0.250 mmol) was added dropwise resulting in a pale yellow solution. The reaction was stirred at -78 °C for twenty minutes, then warmed to r.t. where it was stirred for 2 hours. At this time, the reaction was clear and colourless. Following removal of the volatiles a clear and colourless oil remained. The oil was taken up in a small portion of pentane (1 mL) and placed in the freezer overnight. Small crystals of unreacted (C6F5)2B(OC6F5) was found to crystallize overnight. The supernatant was removed and pumped down yielding the

138

1 pure product as a clear and colourless oil. Yield: 0.126 g (90 %). H NMR (CD2Cl2): δ 3.24 (s, 11 1 13 1 1H, CH); 0.16 (s, 9H, TMS). B{ H} NMR (CD2Cl2): δ 51.00 (s). C{ H} NMR (CD2Cl2): δ 1 1 1 146.13 (br d, JC-F = 248 Hz, C6F5); 144.22 (br d, JC-F = 241 Hz, C6F5); 142.95 (br d, JC-F = 248 1 1 Hz, C6F5); 139.66 (br d, JC-F = 244 Hz, C6F5); 138.26 (br d, JC-F = 248 Hz, C6F5); 139.18- 1 135.93 (overlapping br doublets, 3 x ArC-C6F5); 137.37 (br d, JC-F = 244 Hz, C6F5); 129.65 (t, 2 2 JC-F = 14.58 Hz, ipso-C6F5); (t, JC-F = 18.84 Hz, ipso-C6F5); 107.36 (br s, ipso-C6F5); 29.71 (s, 19 3 CH); -0.18 (s, TMS). F NMR (CD2Cl2): δ -132.16 (d, 2F, JF-F = 21 Hz, o-BC6F5); -141.64 (dd, 3 4 3 2F, JF-F = 22 Hz, JF-F = 5.9 Hz, o-CC6F5); -148.51 (t, 1F, JF-F = 20 Hz, p-BC6F5); -157.15 (d, 3 3 2F, JF-F = 22 Hz, o-OC6F5); -159.81 (m, 2F, m-BC6F5); -160.19 (t, 1F, JF-F = 21 Hz, p-CC6F5); 3 -161.23 (t, 1F, JF-F = 22Hz, p-OC6F5); -162.56 (m, 2F, m-OC6F5); -163.47 (m, 2F, m-CC6F5). 29 Si NMR (CD2Cl2): δ -112.06 (br s).

Synthesis of 4-12. Ph2C(N2) (0.050 g, 0.257 mmol) in CH2Cl2 (4 mL)

was added very slowly to a suspension of C6F5B(OH)2 (0.054 g, 0.255

mmol) in CH2Cl2 (2 mL). The reaction was vigorous, foaming

considerably as the Ph2C(N2) was consumed and N2 was liberated. Addition must be slow to prevent overflowing the reaction vessel. The

fuchsia colour of the Ph2C(N2) faded to a deep orange and finally to a bright yellow as the reaction progressed over a 15 minute period of time. Once gas evolution ceased the reaction was allowed to stir at r.t. for an hour. At this time the volatiles were removed yielding a yellow sticky solid. The solid was washed with pentane (2 x 5 mL) yielding an off-white solid. Pure product was recovered following recrystallization from

a layered solution of CH2Cl2 and pentane. Following recrystallization, the supernatant was decanted and the volatiles removed. A second recrystallization of this crude material was conducted. This procedure was repeated three times. Crystals suitable for X-ray diffraction 1 were grown from a layered solution of CH2Cl2 and pentane at -35 °C. Yield: 0.094 (94 %). H 3 NMR (CD2Cl2): δ 7.24-7.16 (m, 18H, m/p-C6H5); 7.07 (d, 12H, JH-H = 7.4 Hz, o-C6H5). 11 1 13 1 1 B{ H} NMR (CD2Cl2): δ 29.71 (s). C{ H} NMR (CD2Cl2): δ 145.77 (br d, JC-F = 245 Hz, o- 1 1 C6F5); 140.51 (br d, JC-F = 250 Hz, p-C6F5); 139.83 (s, ipso-C6H5); 138.47 (br d, JC-F = 250 Hz, 2 m-C6F5); 129.76 (s, m-C6H5); 128.85 (s, o-C6H5); 127.53 (p-C6H5); 120.11 (t, JC-F = 17 Hz, ipso- 19 3 C6F5); 47.40 (s, CPh2C6F5). F NMR (CD2Cl2): δ -133.27 (d, 6F, JF-F = 18 Hz, o-C6F5); -157.44

139

3 (t, 3F, JF-F = 21 Hz, p-C6F5); -162.25 (m, 6F, m-C6F5). Anal. Calcd. for C57H30B3F15O3: C, 63.32; H, 2.80. Found: C, 63.25; H, 2.85 %.

Synthesis of 4-13. Ph2C(N2) (0.069 g, 0.355 mmol) in CH2Cl2 (4 mL)

was added to (4-FC6H4)B(OH)2 (0.050 g, 0.357 mmol) in CH2Cl2 (2

mL). Reaction of the boronic acid with the Ph2C(N2) was slow and the fuchsia colour of the diphenyldiazomethane was found to fade to a vibrant yellow and then decolourize as the reagent was consumed finally yielding a clear and colourless solution upon reaction completion (5 hours). At this time the volatiles were removed yielding an off-white sticky solid. The product was washed with pentane (2 x 5 mL) and dried in vacuo for two hours. The end product was isolated as an off-white solid. Crystals were grown from a layered solution of

CH2Cl2 and pentane at -35 °C but were not suitable for X-ray diffraction. Yield: 0.083 (81 %). 1 H NMR (CD2Cl2): δ 7.21-7.12 (m, 26H, ArH); 7.91-7.90 (m, 24H, ArH); 6.82-6.77 (m, 7H, 11 1 13 1 1 ArH). B{ H} NMR (CD2Cl2): δ 33.20 (br s). C{ H} NMR (CD2Cl2) partial: δ 161.61 (d, JC- 4 F = 244 Hz, p-C6H4F); 144.76 (s, ipso-C6H5); 140.70 (d, JC-F = 3.3 Hz, ipso-C6H4F); 132.62 (d, 3 2 JC-F = 7.9 Hz, o-C6H4F); 130.83 (s, m-C6H5); 128.73 (s, C6H5); 126.64 (s, C6H5); 115.18 (d, JC-F 19 = 21 Hz, m-C6H4F). F NMR (CD2Cl2): δ -118.87 (m, 3F, p-C6H4F. Anal. Calcd. for

C57H42B3F3O3: C, 79.20; H, 4.90. Found: C, 77.50; H, 5.24. Elemental analysis consistently found to be low on carbon, attributable to the formation of boron-carbides.

Synthesis of 4-14. Prepared in an analogous fashion to 4-13 employing

Ph2C(N2) (0.050 g, 0.257 mmol) and C6H5B(OH)2 (0.031 g, 0.254 mmol).

in CH2Cl2 (5 mL). The end product was isolated as an off-white solid. Crystals suitable for X-ray diffraction were grown from a layered solution 1 of CH2Cl2 and pentane at -35 °C. Yield: 0.052 (75 %). H NMR 3 11 1 (CD2Cl2): δ 7.12-7.09 (m, 27H, m/p-C6H5); 6.92 (d, 18H, JH-H = 7.4 Hz, o-C6H5). B{ H} 13 1 NMR (CD2Cl2): δ 32.50 (s). C{ H} NMR (CD2Cl2): δ 145.04 (s, ipso-C6H5); 131.03 (s, m-

C6H5); 128.56 (s, o-C6H5); 126.38 (s, p-C6H5); 54.72 (s, CPh3). Anal. Calcd. for C57H45B3O3: C, 84.48; H, 5.60. Found: C, 82.43; H, 5.68. Elemental analysis consistently found to be low on carbon, attributable to the formation of boron-carbides.

140

Synthesis of 4-15. Me3SiCH(N2) (629 μL of a 2.0 M solution, 1.26 mmol)

was added to C6H4O2B(C6F5) (0.120 g, 0.420 mmol) in CH2Cl2 (5 mL) resulting in a vibrant yellow mixture. The reaction was allowed to stir at r.t. for 4 hours. At this time, the volatiles were removed and the product

was left under vacuum for two hours to remove any residual Me3SiCH(N2). A pale-yellow oil 1 3 4 was recovered. Yield: 0.140 g (90 %). H NMR (CD2Cl2): δ 7.22 (dd, 2H, JH-H = 5.8 Hz, JH-H = 3 4 3.4 Hz, C6H4); 7.11 (dd, 2H, JH-H = 5.8 Hz, JH-H = 3.4 Hz, C6H4); 2.87 (s, 1H, CH); 0.21 (s, 9H, 11 1 13 1 Si(CH3)3). B{ H} NMR (CD2Cl2): δ 34.38 (br s). C{ H} NMR (CD2Cl2): δ 148.87 (s, ipso- 1 1 C6H4); 145.20 (br d, JC-F = 242 Hz, o-C6F5); 138.95 (br d, JC-F = 248 Hz, p-C6F5); 138.38 (br d, 1 2 JC-F = 248 Hz, m-C6F5); 123.39 (s, C6H4); 115.51 (t, JC-F = 20 Hz, ipso-C6F5); 112.93 (s, C6H4); 19 3 4 12.53 (br s, CH); -0.73 (s. Si(CH3)3). F NMR (CD2Cl2): δ -141.02 (dd, 2F, JF-F = 22 Hz, JF-F 3 29 = 7.3 Hz, o-C6F5); -161.29 (t, 1F, JF-F = 21 Hz, p-C6F5); -163.96 (m, 2F, m-C6F5). Si NMR

(CD2Cl2): δ -114.92 (br s, Si(CH3)3).

Synthesis of 4-16. Prepared in an analogous fashion to 4-15 employing

Me3SiCH(N2) (1.40 mL of a 2.0 M solution, 2.80 mmol) and

C6H4O2B(C6H4F) (0.120 g, 0.561 mmol). The reaction was allowed to stir at r.t. overnight. Analytically pure borane was recovered as a colourless oil. Placement of the oil in a -35 °C freezer resulted in the solidification of the product yielding an 1 3 4 off-white solid. Yield: 0.139 g (83 %). H NMR (CD2Cl2): δ 7.08 (dd, 2H, JH-H = 8.9 Hz, JH-F 3 4 3 = 5.3 Hz, C6H4F); 7.03 (dd, 2H, JH-H = 5.9 Hz, JH-F = 3.3 Hz, C6H4O2); 6.86 (t, 2H, JH-H = 8.9 3 4 Hz, C6H4F); 6.78 (dd, 2H, JH-H = 5.9 Hz, JH-F = 3.3 Hz, C6H4O2); 2.32 (s, 1H, CH); -0.03 (s, 11 1 13 1 9H, Si(CH3)3). B{ H} NMR (CD2Cl2): δ 35.20 (br s). C{ H} NMR (CD2Cl2): δ 160.99 (d, 1 3 JC-F = 241 Hz, p-C6H4F); 148.91 (s, ipso-C6H4O2); 135.80 (s, ipso-C6H4F); 130.32 (d, JC-F = 7.2 2 Hz, o-C6H4F);123.05 (s, C6H4O2); 115.41 (d, JC-F = 21 Hz, m-C6H4F); 112.72 (s, C6H4O2); 19 1 29 26.06 (br s, CH); -1.30 (s, Si(CH3)3). F{ H} NMR (CD2Cl2): δ -119.97 (m, 1F, p-C6H4F). Si

NMR (CD2Cl2): δ -111.11 (br s, Si(CH3)3).

Synthesis of 4-17. Pyridine (247 μL, 3.05 mmol) was added to 4-15 (0.114 g, 0.306 mmol) in pentane (5 mL). Immediate precipitation of an off-white solid was observed. The product was allowed to settle and the supernatant was decanted. The solid was washed with pentane (2 x 5 mL) and dried in vacuo

141

for one hour. Yield: 0. 107 g (78 %). Crystals suitable for X-ray crystallography were obtained 1 3 from a layered solution of CH2Cl2 and pentane at -35 °C. H NMR (CD2Cl2): δ 8.60 (d, 2H, JH- 3 3 H = 7.6 Hz, ortho-C5H5N); 7.82 (t, 1H, JH-H = 7.6 Hz, para-C5H5N); 7.40 (t, 2H, JH-H = 6.7 Hz, meta-C5H5N); 7.00 (m, 2H, C6H4O2); 6.89 (m , 2H, C6H4O2); 2.50 (s, 1H, CH); 0.08 (s, 9H, 11 1 13 1 Si(CH3)3). B{ H} NMR (CD2Cl2): δ 18.53 (s). C{ H} NMR (CD2Cl2): δ 150.89 (s, ipso- 1 C6H4); 145.57 (s, C5H5N); 145.08 (br d, JC-F = 243 Hz, o-C6F5); 140.57 (s, C5H5N); 137.96 (br 2 dm, p/m-C6F5); 125.60 (s, C5H5N); 120.86 (s, C6H4); 118.55 (t, JC-F = 20 Hz, ipso-C6F5); 111.20 19 3 (s, C6H4); 18.21 (br s, CH); -0.46 (s, Si(CH3)3). F NMR (CD2Cl2): δ -141.08 (dd, 2F, JF-F = 23 4 3 29 Hz, JF-F = 6.3 Hz, o-C6F5); -162.82 (1F, JF-F = 21 Hz, p-C6F5); -164.68 (m, 2F, m-C6F5). Si

NMR (CD2Cl2): δ -113.49 (br s, Si(CH3)3). Anal. Calcd. for C21H19BF5NO2Si: C, 55.89; H, 4.24; N, 3.10. Found: C, 55.91; H, 4.45; N, 3.21 %.

Synthesis of 4-18. Pyridine (243 μL, 3.00 mmol) was added to 4-16 (0.114 g, 0.300 mmol) in pentane (5 mL). The reaction mixture became a very pale yellow upon the pyridine addition. The volatiles were removed yielding an off-white solid. The product was dried in vacuo for one hour. Yield: 0. 132 g (96 %). Crystals suitable for X-ray crystallography were obtained from a 1 layered solution of CH2Cl2 and pentane at -35 °C. H NMR (CD2Cl2): δ 8.63 (br s, 2H, o- 3 3 C5H5N); 7.76 (t, 1H, JH-H = 7.7 Hz, p-C5H5N); 7.36 (t, 1H, JH-H = 7.3 Hz, m-C5H5N); 7.14 (m,

2H, ArH); 7.03 (m, 2H, ArH); 6.89 (m, 4H, ArH); 2.21 (s, 1H, CH); 0.17 (s, 9H, Si(CH3)3). 11 1 13 1 1 B{ H} NMR (CD2Cl2): δ 26.41 (s). C{ H} NMR (CD2Cl2): δ 161.65 (br d, JC-F = 240 Hz, p-

C6H4F); 151.05 (s, ipso-C6H4O2); 148.71 (s, C5H5N); 139.37 (C5H5N); 138.94 (s, ipso-C6H4F); 3 2 131.27 (d, JC-F = 7.3 Hz, o-C6H4F); 126.04 (s, C5H5N); 122.70 (s, C6H4O2); 16.11 (d, JC-F = 22 19 Hz, m-C6H4F); 112.79 (s, C6H4O2); 29.69 (br s, CH); 0.00 (s, Si(CH3)3). F NMR (CD2Cl2): δ - 29 121.57 (m, 1F, p-C6H4F). Si NMR (CD2Cl2): δ -113.02 (br s, Si(CH3)3). Anal. Calcd. for

C21H23BFNO2Si: C, 66.50; H, 6.11; N, 3.69. Found: C, 66.77; H, 6.20; N, 3.93 %.

Synthesis of 4-19. Ph2C(N2) (0.199 g, 1.02 mmol) in CH2Cl2 (2 mL) was

added to catecholborane (0.123 g, 1.03 mmol) in CH2Cl2 (5 mL). Immediate

effervescence was observed as N2 was liberated. The solution changed from a pink to a bright yellow solution over a period of 30 minutes at r.t. The reaction was allowed to stir for a total of 2 hours to ensure reaction completion. At this time, the volatiles were removed

142

1 3 yielding a yellow-oil. Yield: 0.285 g (97 %). H NMR (C6D6): δ 7.31 (d, 4H, JH-H = 7.4 Hz, o- 3 3 C6H5); 7.12 (t, 4H, JH-H = 7.4 Hz, m-C6H5); 7.02 (t, 2H, JH-H = 7.4 Hz, p-C6H5); 6.95 (dd, 2H, 3 4 3 4 JH-H = 5.7 Hz, JH-H = 3.2 Hz, p-C6H5); 6.73 (dd, 2H, JH-H = 5.7 Hz, JH-H = 3.2 Hz, p-C6H5); 11 1 13 1 4.30 (s, CH). B{ H} NMR (C6D6): δ br s (34.63). C{ H} NMR (C6D6): δ 148.64 (s, ipso-

C6H2O2); 141.35 (s, ipso-C6H5); 129.64 (s, o-C6H5); 128.95 (s, m-C6H5); 126.39 (s, p-C6H5);

122.97 (s, m-C6H4O2); 112.79 (s, o-C6H4O2); 39.09 (br s, CPh2H). HRMS. Calcd. 12 1 11 14 16 C19 H19 B1 N1 O2: 304.15088 mu. Found: 304.15048 mu.

Preparation of 4-20: Ph2C(N2) (0.050 g, 0.257 mmol) in CH2Cl2 (2 mL)

was added to B-chlorocatecholborane (0.040 g, 0.259 mmol) in CH2Cl2 (5

ml). The Ph2C(N2) was consumed instantaneously upon addition, evidenced by the disappearance of the fuchsia colour. The resulting clear and colourless solution was stirred at r.t. for an hour. At this time, the volatiles were removed and the resulting oil was taken up in pentane and filtered through a plug of Celite. The volatiles were then removed yielding a yellow-coloured oil, which solidified to an off-white solid upon standing. Yield: 0.072 (87 %). 1 3 H NMR (CD2Cl2): δ 7.50-7.48 (m, 4H, C6H5); 7.41-7.34 (m, 6H, C6H5); 7.31 (dd, 2H, JH-H = 4 3 4 11 1 5.9 Hz, JH-H = 3.4 Hz, C6H4O2); 7.17 (dd, 2H, JH-H = 5.9 Hz, JH-H = 3.4 Hz, C6H4O2). B{ H} 13 1 NMR (CD2Cl2): δ 32.79 (s). C{ H} NMR (CD2Cl2): δ 148.60 (s, ipso-C6H4O2); 142.68 (s,

ipso-C6H5); 129.10 (s, m-C6H5); 128.96 (s, o-C6H5); 128.51 (s, p-C6H5); 123.97 (s, m-C6H4O2);

113.55 (s, o-C6H4O2). Anal. Calcd. for C19H14BClO2: C, 71.13; H, 4.40. Found: C, 70.78; H, 4.38 %.

Synthesis of 4-21. 4-20 was prepared employing Ph2C(N2) (0.201 g, 1.03

mmol) and B-chlorocatecholborane (0.160 g, 1.04 mmol) in CH2Cl2 (8 mL).

Triethylphosphine oxide (0.139 g, 1.04 mmol) in CH2Cl2 (2 mL) was added to the solution of the borane. The reaction was stirred for an hour before the volatiles were removed, revealing a viscous yellow oil. Pentane (10 mL) was added and the oil was triturated to a solid. The pentane was decanted and the white solid was washed with pentane (2 x 5 mL) 1 again. The solid was later dried in vacuo. Yield: 0.449 g (90 %). H NMR (CD2Cl2): δ 7.51 (m, 2 3 4H, ArH); 7.25-7.12 (m, 6H, ArH); 6.72-6.63 (m, 4H, ArH); 1.68 (dq, 6H, JH-P = 12 Hz, JH-H = 3 3 11 1 7.8 Hz, PEt); 0.97 (dt, 9H, JH-H = 7.7 Hz, JP-H = 18 Hz, PCH2CH3). B{ H} NMR (CD2Cl2): δ 13 1 11.09 (s). C{ H} NMR (CD2Cl2): δ 152.22 (s, ipso-C6H4O2); 147.79 (s, ipso-C6H5); 130.04 (s,

143

m-C6H5); 127.60 (s, o-C6H5); 126.36 (s, p-C6H5); 119.43 (s, m-C6H4O2); 109.85 (s, o-C6H4O2); 1 2 31 1 117.49 (d, JC-P = 65 Hz, PCH2); 5.45 (d, JC-P = 5.0 Hz, PCH2CH3). P{ H} NMR (CD2Cl2): δ

77.63 (s). Anal. Calcd. for C25H29BClO3P: C, 66.03; H, 6.43. Found: C, 64.84; H, 6.45. Elemental analysis consistently found to be low on carbon, attributable to the formation of boron-carbides.

Preparation of 4-22. Ph2C(N2) (0.112 g, 0.577 mmol) in CH2Cl2 (5 mL)

was added to HB(C6F5)2 (0.100 g, 0.289 mmol) in CH2Cl2 (5 mL). The resulting fuchsia-coloured solution was allowed to stir at r.t. for a period of four hours during which the colour of the reaction faded from fuchsia, to pale pink and finally to bright yellow. The volatiles were removed in vacuo and the recovered yellow solid was washed with hexanes (2 x 5 ml) yielding an off-white solid. The product was dried in vacuo for 1 hour. Crystals suitable for X-ray diffraction were grown from a layered 1 solution of CH2Cl2 and pentane at -35 °C. Yield: 0.129 g (83 %). H NMR (CD2Cl2): δ 8.75 (s, 11 1 1H, NH); 7.64-7.60 (m, 3H, C6H5); 7.45-7.29 (m, 7H, C6H5). B{ H} NMR (CD2Cl2): δ 34.84 13 1 1 (br s). C{ H} NMR (CD2Cl2) partial: δ 158.53 (s, N=C); 148.80 (br d, JC-F = 248 Hz, C6F5); 1 1 1 146.72 (br d, JC-F = 245 Hz, C6F5); 142.85 (br d, JC-F = 249 Hz, C6F5); 141.83 (br d, JC-F = 248 1 Hz, C6F5); 137.35 (br d, JC-F = 245 Hz, C6F5); 136.92 (s, ipso-C6H5); 131.52 (s, ipso-C6H5); 19 130.23 (s, C6H5); 129.74 (s, C6H5); 128.76 (s, C6H5); 128.39 (s, C6H5); 127.90 (s, C6H5). F 3 3 NMR (CD2Cl2): δ 131.52 (d, 2F, JF-F = 18 Hz, o-C6F5); -131.81 (d, 2F, JF-F = 18 Hz, o-C6F5); 3 4 3 (dd, 2F, JF-F = 23 Hz, JF-F = 8 Hz o-C6F5); -149.84 (t, 1F, JF-F = 23 Hz, p-C6F5); -153.93 (t, 1F, 3 JF-F = 20 Hz, p-C6F5); -162.01 (m, 2F, m-C6F5); -163.40 (m, 2F, m-C6F5). HRMS. Calcd. 12 1 11 19 14 C25 H12 B1 F10 N2: 541.09339 mu. Found: 541.09223 mu.

Synthesis of 4-23. Me3SiN3 (30 μL, 0.0228 mmol) was added to HB(C6F5)2

(0.080 g, 0.231 mmol) in CH2Cl2 (5 mL) and stirred for one hour at r.t. The volatiles were removed from the clear and colourless solution yielding a white microcrystalline solid which was analytically pure requiring no further purification. Yield: 1 1 0.103 g, (93 %). H NMR (CD2Cl2): δ 3.81 (br q, 1H, JH-B = 85 Hz, BH); 0.49 (s, 9H, 11 1 13 1 Si(CH3)3). B NMR (CD2Cl2): δ -9.64 (d, JB-H = 85 Hz, BH). C{ H} NMR (CD2Cl2): δ 1 1 1 148.78 (br d, JC-F = 239 Hz, o-C6F5); 140.90 (br d, JC-F = 254 Hz, p-C6F5); 137.82 (br d, JC-F = 19 3 4 247 Hz, m-C6F5); -1.45 (s, Si(CH3)3). F NMR (CD2Cl2): δ -134.70 (dd, 4F, JF-F = 24 Hz, JF-F

144

3 29 1 = 8.0 Hz, o-C6F5); -157.58 (t, 2F, JF-F = 20 Hz, p-C6F5); -164.24 (m, 4F, m-C6F5). Si{ H}

NMR (CD2Cl2): δ -110.99 (s, Si(CH3)3). Anal. Calcd. for C15H10BF10N3Si: C, 39.04; H, 2.19; N, 9.11. Found: C, 39.02; H, 2.21 ; N, 8.72.

Synthesis of 4-24. (C6H5)CH2N3 (35.4 μL, 0.283 mmol) was added to

HB(C6F5)2 (0.098 g, 0.283 mmol) in pentane (5 mL). Liberation of N2 resulted in vigorous bubbling of the solution and was found to cease within two minutes. The reaction mixture was stirred for an hour to ensure reaction completion prior to removal of the volatiles. A white and sticky solid was recovered and was subsequently washed with cold pentane (2 x 2mL). The pentane was decanted and the product was dried in vacuo 1 yielding a white solid. Yield: 0.1221 g (96 %). H NMR (CD2Cl2): δ 7.37 (m, 2H, C6H5); 7.30 3 11 1 (m, 3H, C6H5); 5.92 (br s, 1H, NH); 4.36 (d, 2H, JH-H = 7.4 Hz, CH2). B{ H} NMR (CD2Cl2): 13 1 1 δ 34.36 (s). C{ H} NMR (CD2Cl2) partial: δ 148.52 (br d, JC-F = 247 Hz, o-C6F5); 146.69 (br 1 1 1 d, JC-F = 241 Hz, o-C6F5); 142.71 (br d, JC-F = 251 Hz, p-C6F5); 142.19 (br d, JC-F = 249 Hz, p- 1 C6F5); 139.08 (s, ipso-C6H5); 137.72 (br d, JC-F = 249 Hz, 2 x m-C6F5); 129.14 (s, m-C6H5); 19 128.06 (s, p-C6H5); 127.39 (s, o-C6H5); 49.64 (s, CH2). F NMR (CD2Cl2): δ -133. 40 (dm, 2F, 3 3 3 4 JF-F = 23 Hz, o-C6F5); -134.01 (m, 2F, JF-F = 24 Hz, o-C6F5); -152.69 (tt, 1F, JF-F = 20 Hz, JF-F 3 = 3.4 Hz, p-C6F5); -154.39 (t, 1F, JF-F = 20 Hz, p-C6F5); -163.19 (m, 2F, m-C6F5); -163.45 (m,

2F, m-C6F5). Anal. Calcd. for C19H8BF10N: C, 50.55; H, 1.79; N, 3.10. Found: C, 50.14; H, 2.35; N, 2.87 5.

Synthesis of 4-25. Ph2C(N2) (0.040 g, 0.206 mmol) in CH2Cl2 (3 mL)

was added to tol•Zn(C6F5)2 (0.100 g, 0.203 mmol) in CH2Cl2 (5 mL).

The Ph2C(N2) was instantaneously consumed yielding a pale yellow solution. The reaction was stirred for an hour to ensure reaction completion. The volatiles were removed yielding a yellow solid. The product was washed with pentane (2 x 5 mL) and dried in vacuo. The product was recovered as a white solid. Crystals

suitable for X-ray crystallography were grown from a layered solution of CH2Cl2 and pentane at 1 3 -35 °C. Yield: 0.115 g (94 %). H NMR (CD2Cl2): δ 7.63 (d, 4H, JH-H = 8 Hz, C6H5); 7.51-7.42 13 1 (m, 8H, C6H5); 7.39-7.31 (m, 8H, C6H5). C{ H} NMR (CD2Cl2): δ 178.28 (s, CPh2); 148.78 (br 1 2 1 ddm, JC-F = 229 Hz, JC-F = 25.6 Hz, o-C6F5); 140.43 (br dm, JC-F = 229 Hz, p-C6F5); 138.11 (s, 1 2 C6H5); 136.71 (br ddm, JC-F = 258 Hz, JC-F = 30.6 Hz, m-C6F5); 134.82 (s, C6H5); 134.20 (s,

145

C6H5); 132.58 (s, C6H5); 132.07 (s, C6H5); 130.98 (s, C6H5); 130.63 (s, C6H5); 129.71 (s, C6H5); 2 19 3 118.60 (t, JC-F = 60.86 Hz, ipso-C6F5). F NMR (CD2Cl2): δ -118.60 (d, 8F, JF-F = 18 Hz, o- 3 C6F5); -157.68 (t, 4F, JF-F = 20 Hz, p-C6F5); -163.61 (m, 8F, m-C6F5). Anal. Calcd. for

C52H20F20Zn2: C, 51.77; H, 1.74; N, 2.42. Found: C, 51.21; H, 1.97; N, 2.48 %.

Synthesis of 4-26. SIMes (0.094 g, 0.307 mmol) in toluene (3 mL) was added to

tol•Zn(C6F5)2 (0.150 g, 0.305 mmol) in toluene (5 mL). The resulting solution became cloudy nearly immediately upon addition of the carbene. The reaction was allowed to stir at r.t. for 1 hour. At this time the solution was reduced to one half its volume and pentane (15 mL) was added precipitating a fine white solid. The precipitate was allowed to settle and the beige supernatant was decanted. The solid was then washed with two portions of pentane (2 x 5 mL). The solid was dried in vacuo for 2 hours. The isolated product was analytically pure and required no further purification. Crystals suitable for X-ray diffraction 1 were grown from a layered solution of CH2Cl2 and pentane at -35 °C. Yield: 0.188 g (87 %). H

NMR (CD2Cl2): δ 6.50 (s, 4H, C6H2Me3); 3.55 (s, 4H, C2H4); 2.08 (s, 12H, o-CH3); 2.06 (s, 6H, 13 1 1 p-CH3). C{ H} NMR (CD2Cl2): δ 200.28 (s, Zn=C); 147.74 (br d, JC-F = 230 Hz, o-C6F5); 1 1 139.83 (s, ipso-C6H2Me3); 139.53 (br d, JC-F = 247 Hz, p-C6F5); 136.35 (br d, JC-F = 251 Hz, m-

C6F5); 136.16 (s, p-C6H2Me3); 134.34 (s, o-C6H2Me3); 129.56 (s, m-C6H2Me3); 52.00 (s, C2H4); 19 20.89 (s, p-CH3); 17.74 (s, o-CH3). F NMR (C6D5Br): δ -116.59 (m, 4F, o-C6F5); -158.49 (t, 3 2F, JF-F = 20 Hz, p-C6F5); -161.92 (m, 4F, m-C6F5). Anal. Calcd. for C33H26F10N2Zn: C, 55.77; H, 3.67; N, 3.81. Found: C, 56.12; H, 3.71; N, 3.97 %.

Synthesis of 4-27. A 100 mL round-bottom bomb was charged with

4-1 (0.040 g, 0.0.67 mmol) and tBu3P (0.014 g, 0.069 mmol) in toluene (10 mL). The clear and colourless solution was degassed

employing three freeze-pump-thaw cycles. The bomb was frozen and filled with H2 resulting in an internal pressure of roughly 4 atmospheres of H2. The solution was allowed to stir at r.t. for two days. During this time the reaction mixture became cloudy. Pentane (15 mL) was added precipitating a fine white solid. The solid was allowed to settle and the solvent was decanted. The precipitate was further washed with pentane (2 x 5 mL) and then dried in vacuo for 2 hours. The product was analytically pure and required no further purification. Yield: 0.045 g (83 %) 1H 1 1 NMR (CD2Cl2): δ 5.17 (br d, 1H, JH-P = 431 Hz, PH); 3.38 (br q, 1H, JH-B = 94 Hz, BH); 2.82 3 3 11 (d, 1H, JH-H = 9.2 Hz, CH); 1.69 (d, 27H, JH-P = 16 Hz, tBu); -0.20 (s, 9H, Si(CH3)3). B NMR

146

1 13 1 1 (CD2Cl2): δ -21.65 (d, JB-H = 94 Hz, BH). C{ H} NMR (CD2Cl2) partial: δ 149.19 (d, JC-F = 1 1 1 236 Hz, ArC); 148.63 (d, JC-F = 232 Hz, ArC); 146.00 (d, JC-F = 251 Hz, ArC); 145.28 (d, JC-F 1 = 232 Hz, ArC); 139.52-134.72 (br m, multiple ArC), 128.38 (br s, ArC); 38.27 (d, JC-P = 27 Hz, 19 PC{CH3}3); 30.56 (s, PC{CH3}3); 16.57 (br s, CH); 0.74 (s, Si(CH3)3). F NMR (CD2Cl2): δ -

130.79 (s, 2F, o-B(C6F5)); -132.30 (s, 2F, o-B(C6F5)); -136.96 (s, 1F, o-C(C6F5)); -143.29 (s, 1F, 3 3 3 JF-F = 22 Hz, o-C(C6F5)); -164.73 (t, 1F, JF-F = 20 Hz, p-B(C6F5)); -164.97 (t, 1F, JF-F = 20 Hz,

p-B(C6F5)); -167.24 (m, 4F, m-B(C6F5)); -167.48 (m, 1F, p-C(C6F5)); -167.97 (m, 2F, m- 29 31 1 C(C6F5)). Si NMR (CD2Cl2): δ -115.65 (br s, Si(CH3)3). P NMR (CD2Cl2): δ -59.94 (d, JP-H

= 431 Hz, PH). Calcd. for C34H39BF15PSi: C, 50.86; H, 4.90; N, 2.42. Found: C, 50.69; H, 5.01 %.

Synthesis of 4-28. A 50 mL tube bomb was charged with 4-1 (0.080 g,

0.134 mmol), tBu3P (0.027 g, 0.133 mmol), benzophenone (0.024 g, 0.132 mmol) and toluene (10 mL). The resulting solution was degassed employing three consecutive freeze-pump-thaw cycles. The vessel was then submerged in liquid

N2, freezing the contents and backfilled with H2, generating an internal H2 pressure of ~4 atm upon thawing. The bomb was placed in an oil bath thermostated for 80 °C and was left stirring for a period of 48 hours. At this time, the volatiles were removed. The resulting residue was washed repeatedly with pentane (3 x 5 mL) yielding a white solid. The product was found to be analytically pure and required no further purification. Yield: 0.075 g (77 %). 1H NMR 1 1 (CD2Cl2): δ 5.17 (d, 1H, JH-P = 431 Hz, PH); 2.77 (br q, 1H, JH-B = 90 Hz, BH); 2.43 (s, 2H, 3 11 1 1 CH2); 1.64 (d, 27H, JH-P = 16 Hz, PtBu). B{ H} NMR (CD2Cl2): δ -20.21 (d, JB-H = 90 Hz, 13 1 1 1 BH). C{ H} NMR (CD2Cl2): δ148.75 (br s, JC-F = 238 Hz, C6F5); 144.80 (br s, JC-F = 238 Hz, 1 1 1 C6F5); 138.14 (br s, JC-F = 244 Hz, C6F5); 137.39 (br s, JC-F = 240 Hz, C6F5); 136.95 (br s, JC-F 1 2 = 246 Hz, C6F5); 136.72 (br s, JC-F = 246 Hz, C6F5); 127.97 (ipso-BC6F5); 125.54 (t, JC-F = 19 1 Hz, ipso-CC6F5); 38.20 (d, JC-P = 27 Hz, PC{CH3}3); 30.49 (s, PC{CH3}3); 17.16 (br s, CC6F5). 19 3 3 F NMR (CD2Cl2): δ -134.46 (d, 4F, JF-F = 24 Hz, o-BC6F5); -147.50 (d, 2F , JF-F = 21 Hz, o- 3 3 CC6F5); -165.77 (t, 2F , JF-F = 21 Hz, p-BC6F5); -168.39 (t, 1F , JF-F = 21 Hz, p-CC6F5); -168.46 31 1 (m, 6F, m-BC6F5/m-CC6F5). P NMR (CD2Cl2): δ 59.65 (d, JP-H = 431 Hz, PH). Anal. Calcd.

for C31H31BF15P: C, 50.95; H, 5.28. Found: C, 51.25; H, 5.35 %.

147

Synthesis of 4-29. A 100 mL round-bottom bomb was charged

with 4-1 (0.075 g, 0.125 mmol) and tBu3P (0.025 g, 0.124 mmol)

in CH2Cl2 (10 mL). A solution of XeF2 (0.021 g, 0.124 mmol) in

CH2Cl2 (2 mL) was added to the solution of the borane and the evolution of Xe gas was immediately noted. The solution was allowed to stir at r.t. for two hours. The volatiles were then removed revealing a white solid. The product was washed with pentane 1 (2 x 5 mL) and then dried in vacuo for 2 hours. Yield: 0.100 g (96 %). H NMR (CD2Cl2): δ 3 3 3.16 (d, 1H, JH-F = 28 Hz, C); 1.69 (d, 27H, JH-P = 16 Hz, tBu); -0.08 (s, 9H, Si(CH3)3). 11 1 1 13 1 B{ H} NMR (CD2Cl2): δ -3.23 (d, JB-F = 75 Hz, BF). C{ H} NMR (CD2Cl2): δ 148.59 (br d, 1 1 1 JC-F = 240 Hz, C6F5); 147.38 (br d, JC-F = 233 Hz, C6F5); 145.81 (br d, JC-F = 245 Hz, C6F5); 1 145.02 (br d, JC-F = 229 Hz, C6F5); 140.04-135.15 (br m, multiple C6F5); 126.97 (s, ipso-C6F5); 2 1 2 125.06 (t, JC-F = 22 Hz, ipso-C6F5); 41.83 (dd, JC-P = 26 Hz, JC-F =7.7 Hz, PC{CH3}3); 28.34 3 19 (d, JC-F = 2.2 Hz, PC{CH3}3); 0.14 (s, Si(CH3)3). F NMR (CD2Cl2): δ -134.01 (m, 2F, o- 3 C6F5); -136.49 (m, 1F, o-C6F5); -137.31 (s, 2F, o-C6F5); -144.94 (d, 1F, JF-F = 23 Hz, o-C6F5); - 3 3 164.49 (t, 1F, JF-F = 20 Hz, p-C6F5); -165.05 (t, 1F, JF-F = 20 Hz, p-C6F5); -167.97 (m, 2F, m- 3 C6F5); -168.08 (t, 1F, JF-F = 20 Hz, p-C6F5); -168.26 (m, 1F, m-C6F5); -168.44 (m, 2F, m-C6F5); - 1 29 168.79 (m, 1F, m-C6F5); -172.52 (d, 1F, JF-P = 1019 Hz, PF); -182.26 (br s, BF). Si NMR 31 1 1 (CD2Cl2): δ -112.24 (br s, Si(CH3)3). P{ H} NMR (CD2Cl2): δ 147.56 (d, JP-F = 1019 Hz, PF).

Anal. Calcd. for C34H37BF17PSi: C, 48.68; H, 4.45. Found: C, 48.33; H, 4.33 %.

Synthesis of 4-30. A 20 mL scintillation vial was charged with

(S,R) 4-2 (0.153 g, 0.224 mmol), tBu3P (0.045 g, 0.222 mmol)

and CH2Cl2 (10 mL). XeF2 (0.038 g, 0.224 mmol) in CH2Cl2 (3 mL) was then added to the solution of phosphine and borane. The resultant mixture was stirred for an hour at r.t. At this time the volatiles were removed revealing a white microcrystalline product. The solid was washed with pentane (1 x 5 mL) and dried in vacuo. It is suggested that the isolated be stored away from light and in a cold environment in order to prevent decomposition. Crystals suitable for X-ray diffraction were grown from a layered solution of

CH2Cl2 and pentane. Yield: 0.180 (88 %). Species observed (present in equal proportions and 1 experimentally indistinguishable from one another). H NMR (CD2Cl2): δ 2.23 (s, 1H, CH); 2.21 3 3 4 (s, 1H, CH); 2.05 (d, 2H, JH-F = 26 Hz, CH); 1.71 (dd, 54H, JH-P = 16 Hz, JH-F = 1.7 Hz, tBu); 11 1 1 0.02 (s, 18H, SiMe3); -0.38 (s, 18H, Si(CH3)3). B{ H} NMR (CD2Cl2): δ 6.59 (d, JB-F = 70 Hz,

148

1 13 1 1 BF); 3.56 (d, JB-F = 73 Hz, BF). C{ H} NMR (CD2Cl2) partial: δ 148.10 (br d, JC-F = 241 Hz, 1 1 1 C6F5); 147.46 (br d, JC-F = 227 Hz, C6F5); 145.13 (br d, JC-F = 242 Hz, C6F5); 144.64 (br d, JC-F 1 1 = 233 Hz, C6F5); 144.16 (br d, JC-F = 230 Hz, C6F5); 136.39 (br d, JC-F = 240 Hz, C6F5); 136.41 1 1 (br d, JC-F = 242 Hz, C6F5); 135.19 (br d, JC-F = 236 Hz, C6F5); 125.17 (br s, ipso- 1 2 BC6F5/CC6F5); 41.23 (dd, JC-P = 27 Hz, JC-F = 7.8 Hz, PC{CH3}3); 27.77 (s, PC{CH3}3); 24.98 19 (br s, CC6F5); 23.57 (br s, CC6F5); 0.44 (s, Me3Si); -0.01 (s, Me3Si). F NMR (CD2Cl2): δ- 3 4 3 130.34 (dd, 2F, JF-F = 25 Hz, JF-F = 9.6 Hz, o-BC6F5); -131.81 (d, 2F, JF-F = 25 Hz, o-BC6F5); - 3 4 3 4 135.55 (dd, 2F, JF-F = 25 Hz, JF-F = 9.6 Hz, o-BC6F5); -136.24 (dd, JF-F = 25 Hz, JF-F = 10 Hz, 3 3 4 o-BC6F5); -136.35 (d, 4F, JF-F = 21 Hz, o-CC6F5); -136.39 (dd, 2F, JF-F = 25 Hz, JF-F = 9.6 Hz, 3 3 o-BC6F5); -136.50 (d, 4F, JF-F = 23 Hz, o-CC6F5); -143.60 (d, 4F, JF-F = 25 Hz, o-CC6F5); - 3 3 144.83 (d, 4F, JF-F = 25 Hz, o-CC6F5); -164.46 (t, 2F, JF-F = 21 Hz, p-BC6F5); -165.18 (t, 2F, 3 3 3 JF-F = 21 Hz, p-BC6F5); -167.09 (tm, 2F, JF-F = 23 Hz, m-BC6F5); -167.99 (tm, 2F, JF-F = 22 3 3 Hz, m-BC6F5); -168.15 (tm, 2F, JF-F = 23 Hz, m-BC6F5); -168.54 (tm, 4F, JF-F = 23 Hz, m- 3 3 CC6F5); -168.85 (t, 4F, JF-F = 21 Hz, p-CC6F5); -169.10 (t, 4F, JF-F = 24 Hz, m-CC6F5); -169.43 3 3 3 (t, 4F, JF-F = 21 Hz, p-CC6F5); -169.64 (t, 4F, JF-F = 21 Hz, m-CC6F5); -169.90 (t, 4F, JF-F = 23 3 1 Hz, m-CC6F5); -170.24 (tm, 2F, JF-F = 21 Hz, m-BC6F5); -172.50 (d, 2F, JF-P = 1019 Hz, PF); - 29 175.04 (br s, 2F, BF); -182.04 (br s, 2F, BF). Si NMR (CD2Cl2): δ -111.76 (br s, Si(CH3)3). 31 1 1 P{ H} NMR (CD2Cl2): δ 147.59 (d, JP-F = 1018 Hz, PF). Anal. Calcd. for C38H47BF17PSi2: C, 49.34; H, 5.12. Found: C, 49.30; H, 5.14 %.

4.4.3 X-ray Crystallography

4.4.3.1 X-ray Data Collection and Reduction

In preparation for analysis, crystals were first coated with Paratone-N oil in a glovebox and were

149

4.4.3.2 Structure and Refinement

150

4.4.3.3 Tables of Crystallographic Data Table 4.2 – Selected crystallographic data for 4-1, 4-2 and 4-4.

4-1 4-2 4-4

Formula C22H10BF15Si C26H20BF15Si2 C25HBF20 Formula wt 598.20 684.41 692.04 Crystal system Orthorhombic Monoclinic Monoclinic

Space group Pbca P21/c P21 a (Å) 11.0719(6) 9.7676(15) 10.7172(6) b (Å) 15.3825(8) 15.440(2) 10.7172(6) c (Å) 27.1192(12) 19.306(3) 11.1932(6) α (deg) 90 90 90 β (deg) 90 98.467(6) 107.629(2) γ (deg) 90 90 90 V (Å3) 4618.8(4) 2879.9(7) 1204.24(12) Z 8 4 2 T (K) 150(2) 150(2) 150(2) d (calc) gcm-3 1.721 1.579 1.909 Abs coeff, μ, mm-1 0.234 0.238 0.220 Data collected 14925 24520 10440 R int 0.0185 0.0507 0.0224 # of indpndt reflns 5303 6597 4838

Reflns Fo≥2.0σ(Fo) 4234 4263 4000 Variables 392 403 415 R (>2σ) 0.0364 0.0420 0.0380

wR2 0.0947 0.0934 0.1250 Goodness of fit 1.017 1.010 0.904

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

151

Table 4.3 – Selected crystallographic data for compounds 4-5, 4-8 and 4-12.

4-5 4-8 4-12

Formula C26H25BF10Si2 C30H15BClF10N C57H30B3F15O3 Formula wt 689.45 625.69 1080.24 Crystal system Monoclinic Monoclinic Rhombohedral

Space group P121/n1 P21/c R3 a (Å) 10.1221(7) 8.2719(3) 20.7933(6) b (Å) 25.2357(16) 17.8870(7) 20.7933(6) c (Å) 11.1170(8) 17.4067(7) 9.7004(5) α (deg) 90 90 90 β (deg) 97.856(3) 92.931(2) 90 γ (deg) 90 90 120 V (Å3) 2813.1(3) 2572.12(17) 3632.2(2) Z 4 4 3 T (K) 150(2) 150(2) 150(2) d (calc) gcm-3 1.628 1.616 1.482 Abs coeff, μ, mm-1 0.244 0.245 0.129 Data collected 32811 41779 20296 R int 0.0364 0.0311 0.0286 # of indpndt reflns 8546 10460 4878

Reflns Fo≥2.0σ(Fo) 6698 7587 4108 Variables 358 448 235 R (>2σ) 0.0393 0.0410 0.0368

wR2 0.1393 0.1090 0.0777 Goodness of fit 1.000 1.026 1.026

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

152

Table 4.4 – Selected crystallographic data for compounds 4-14, 4-17 and 4-18.

4-14 4-17 4-18

Formula C57H45B3O3 C21H19BF5NO2Si C21H23BFNO2Si Formula wt 810.36 451.27 379.30 Crystal system Triclinic Triclinic Triclinic Space group P-1 P-1 P-1 a (Å) 9.5553(5) 8.3043(8) 8.3424(5) b (Å) 11.0517(6) 11.1169(10) 11.1791(8) c (Å) 21.8569(13) 12.0808(11) 22.4978(14) α (deg) 92.454(4) 94.495(4) 76.520(4) β (deg) 97.710(4) 107.005(4) 82.063(8) γ (deg) 109.598(2) 98.763(4) 82.054(3) V (Å3) 2145.2(2) 1045.11(17) 2008.5(2) Z 2 2 4 T (K) 150(2) 150(2) 150(2) d (calc) gcm-3 1.255 1.434 1.254 Abs coeff, μ, mm-1 0.075 0.175 0.141 Data collected 48640 18802 32643 R int 0.0623 0.0321 0.0726 # of indpndt reflns 13141 4795 9048

Reflns Fo≥2.0σ(Fo) 8190 3443 5663 Variables 748 319 577 R (>2σ) 0.0540 0.0691 0.0464

wR2 0.1375 0.1974 0.0989 Goodness of fit 1.021 1.036 0.990

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

153

Table 4.5 – Selected crystallographic data for 4-22, 4-25, 4-26 and 4-30.

4-22 4-25 4-26 4-30

Formula C25H11BF10N2 C25H10F10NZn C33H26F10N2Zn C19H23.5B0.5F8.5P0.5Si Formula wt 540.17 579.71 705.93 462.36 Crystal system Triclinic Monoclinic Monoclinic Monoclinic

Space group P1 C2/c C2/c P21/n a (Å) 6.4079(5) 24.8282(16) 10.259(2) 11.2820(7) b (Å) 7.7817(7) 12.1264(8) 12.178(2) 13.4099(11) c (Å) 11.0637(9) 17.1668(12) 24.677(5) 28.099(2) α (deg) 86.754(5) 90 90 90 β (deg) 85.190(4) 124.207(2) 96.00(3) 92.882(3) γ (deg) 83.495(4) 90 90 90 V (Å3) 545.58(8) 4274.4(5) 3066.1(11) 4245.7(5) Z 1 8 4 8 T (K) 150(2) 150(2) 150(2) 150(2) d (calc) gcm-3 1.644 1.802 1.529 1.447 Abs coeff, μ, 0.157 1.252 0.888 0.224 mm-1 Data collected 12070 18197 10061 31820 R int 0.0278 0.0386 0.0324 0.0649 # of indpndt 5544 4910 3518 7473 reflns Reflns 4996 3917 2945 4871 Fo≥2.0σ(Fo) Variables 387 334 212 555 R (>2σ) 0.0343 0.0324 0.0291 0.0416

wR2 0.0829 0.0758 0.0773 0.0909 Goodness of fit 1.048 1.038 1.017 1.011

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

154

Chapter 5 Activation of Carbon Dioxide by Main Group Frustrated Lewis Pairs and Subsequent Lewis Acid Exchange Chemistry

5.1 Introduction

5.1.1 Carbon Dioxide and the Environment

Over the past century, the global surface temperature has risen by 0.8 °C with particular contributions being made during the past three decades.292,293 This increase in temperature (global warming) is above and beyond what can be attributed to natural causes and is directly

related to anthropogenic sources. Evidence has shown that the levels of atmospheric CO2 are higher than have ever been observed in the past 2.1 million years294 and are 35 % higher than pre-industrialization levels (280 ppm),292 directly correlating with a stark increase in human 295 activity. As a result, the serious effects of greenhouse gases such as CO2, on the phenomenon of global warming are now being acknowledged as an issue of concern by both the scientific 296 community and governmental agencies alike. Atmospheric levels of CO2 have rigorously been recorded since the late 1950s onward showing an increase of CO2 concentrations from 315 ppm 292 to today’s current levels of 390 ppm. In consideration of an annual CO2 concentration increase of 2 ppm and continued economic and population growth, it is predicted that by the end st of the 21 century CO2 levels will lie somewhere between a staggering 900 and 1100 ppm, 3.2 to 3.9 times greater than pre-industrial levels.292

Present day efforts focus on reducing the consumption of fossil fuels (coal, oil and natural gas) and limiting the degree of deforestation occurring globally. In attempts to reduce our carbon footprint, fossil fuel-based energy sources are looking to be replaced by renewable sources such as biomass, solar and wind power.297 In addition, research is focused on a three-step

methodology involving CO2 capture, transportation and storage, where storage involves such strategies as mineralization and ocean storage.297

Ultimately, methods to capture CO2 and transform the activated CO2 molecule into value-added

commodities, are desirable. In fact, CO2 can be viewed as a potentially viable C1

155

feedstock,298-302 which upon derivatization could yield hydrocarbons or methanol, a potential fuel source.295,298,303

5.1.2 Sequestration of CO2 by Main Group Frustrated Lewis Pairs

In consideration of the effects of the greenhouse gas CO2 on the phenomenon of global warming,

it is of significant importance to develop technologies able to readily bind and store CO2.

Numerous technologies currently exist for the capture and subsequent storage of CO2 such as zeolites, activated carbons, silica gels, aluminas and metal organic frameworks.304-308

In recent years, the concept of frustrated Lewis pairs has been applied for the simple activation 5 of CO2. In these instances, combinations of sterically encumbered phosphines and boranes or

phosphines and alanes provide catalyst systems capable of CO2 capture and storage, some of which generate extremely stable and robust adducts where subsequent chemistry can be

conducted at the C-atom of the CO2 moiety. An early collective report by Erker and Stephan described the ability of both intermolecular PtBu3/B(C6F5)3 and intramolecular

Mes2PCH2CH2B(C6F5)2 FLPs to successfully activate CO2 in analogous PC(O)OB binding modes (Scheme 5.1).309

Scheme 5.1 – CO2 sequestration by (a) intermolecular FLP and (b) intramolecular FLP.

Many examples have since been published demonstrating successful activation of CO2 by

intermolecular FLPs, employing a series of boranes of the form RB(C6F5)2 (R = Cy, hexyl, norbornyl) in addition to substitution of the Lewis base.310 A significant limitation to the

activation of CO2 by FLPs is the frequent thermal instability of the CO2 adducts toward CO2 loss. Stephan, however, reported the reaction of PMes3 with two equivalents of the simple alane

AlX3 (X = Br, Cl, I), under an atmosphere of CO2 affords products where a single CO2 unit

bridges two AlX3 units and encompasses a central P-C bond to a sterically encumbered phosphine (Figure 5.1, compound a).311 These adducts are stable at room temperature, thermally robust and are subsequently employed in further chemistry (Section 5.1.3)

156

In attempts to generate species capable of more stable adduct formation with CO2, numerous groups have reported the synthesis of novel intramolecular FLPs that have been shown to be

involved in CO2 activation. Recently, the group of Lammertsma has shown that the geminal

intramolecular FLP tBu2PCH2BPh2 is capable of CO2 activation yielding a five-membered ring (Figure 5.1, compound b), which had previously only been documented in the presence of highly electrophilic perfluorinated boranes.312 A similar report from Lammertsma and Uhl documented

the use of a geminal intramolecular phosphine-alane FLPs for analogous activation of CO2 (Figure 5.1, compound c).73

Stephan et al. have recently investigated the use of boron amidinates in prototypical FLP-type

chemistry and have shown that in the presence of CO2, B−N bond scission occurs with formal 313 insertion of the CO2 in a NC(O)OB binding mode (Figure 5.1, compound d). Recently, bisboranes have been employed as the Lewis acidic component in the FLP-mediated activation of CO2. This was achieved employing both sterically encumbered (B(C6F5)2) and simple (BCl2)

Lewis acidic sites yielding either activation at a single boron centre in conjunction with PtBu3, or double activation where the CO2 moiety is bridging the boron centres with the PtBu3 moiety 314 bound to the CO2 carbon (Figure 5.1, compounds e, f and g).

In addition to recent advances and interest in main group frustrated Lewis pairs, transition metal- based frustrated Lewis pair systems have begun to make an appearance in the field. Recently, Wass et al. reported the exploitation of zirconocene-phosphinoaryloxide complexes in the

activation of CO2. These metal complexes were found to contain either weak Zr−P interactions or possess bulky substituents where steric congestion precludes the generation of formal Zr−P

bonds, resulting in facile CO2 take up yielding CO2 adducts bearing PC(O)OZr binding modes.315,316

157

311 Figure 5.1 – Products of main group-mediated activation of CO2 by: (a) Al/P FLP (Stephan), (b) B/P intramolecular FLP (Lammertsma),73(c) Al/P intramolecular FLP (Lammertsma and Uhl),73,312 (d) a boron amidinate (Stephan), (e), (f) and (g) dinuclear bisboranes/P FLPs (Stephan).313

5.1.3 Reduction of Activated CO2

The activation and reduction of CO2 has previously been the domain of transition metal catalyst

systems. Recently, reports have highlighted the use of CO2 as a potential C1 feedstock for the generation of cleaner and more efficient fuels. This in principle is the basis for the “methanol economy” as proposed by George Olah which suggests the application of methanol as a renewable and environmentally friendly carbon neutral fuel in place of conventional fossil fuels and as building blocks for other synthetic hydrocarbons and associated products.295,303

In the past few years, a handful of examples have surfaced demonstrating the application of

frustrated Lewis pairs in the successful sequestration and reduction of CO2 to methanol and methane employing ammonia borane and silane as reductants. In 2009, O’Hare published the hydrogenation of CO2 by heating a stoichiometric reaction of CO2, 2,2,6,6-tetramethylpiperidine

and B(C6F5)3 under an atmosphere of H2 at 160 °C for 6 days. This reaction, employing low gas pressures, resulted in the thermally induced generation of methanol but in a meager yield of 24 %.317 Shortly following, the Stephan et al. similarly published the ability of Al/P FLPs to

mediate the activation of CO2 and its subsequent reduction to methanol in 37-51 % yield upon room temperature reaction with ammonia borane.311 A more recent report from the Piers group

158

documented the reduction of CO2 to methane, by means of reaction of the 2,2,6,6- tetramethylpiperidinium formatoborate salt, as proposed by O’Hare, with the triethylsilylium 318 cation as generated by the B(C6F5)3 activation of triethylsilane (Scheme 5.2).

318 Scheme 5.2 – Full hydrogenation of CO2 to methane.

5.2 Results and Discussion

5.2.1 Reaction of PtBu3 and RB(C6F5)2 with CO2

The previously reported adduct tBu3P(CO2)B(C6F5)3, resulting from the synergistic activation of gaseous CO2 by a B/P FLP, is a thermally robust zwitterion which only liberates CO2 upon heating at 80 °C.309 This adduct encompasses the sterically hindered and extremely electrophilic borane, tris(pentafluorophenyl)borane and the question of whether analogous complexation chemistry could be achieved in the presence of less Lewis acidic centres warranted investigation.

A stoichiometric mixture of PtBu3 and PhB(C6F5)2, in a J. Young NMR tube, was exposed to an 13 1 31 1 atmosphere of CO2 at room temperature. C{ H} and P{ H} NMR spectroscopy supported the

formation of the B/P CO2 adduct in solution, which bore similar spectroscopic data to the related 309 compound tBu3P(CO2)B(C6F5)3. Repetition of the experiment on a larger scale in a sealed bomb, resulted in the isolation of merely the phosphine and borane starting materials. This

159

would imply that the CO2 adduct, tBu3P(CO2)B(C6F5)2Ph, 5-1, was not stable with respect to

CO2 loss at room temperature, unless maintained under a blanket of CO2. The reaction was repeated in pentane in attempts to promote precipitation of the zwitterionic species, 5-1 (Scheme 5.3, a). Solid 5-1 was found to precipitate from solution yielding a white solid in 55 % yield. The 13 1 31 1 presence of the P(CO2) moiety was supported by a doublet in both the C{ H} and P{ H} 1 NMR spectra at 160.79 and 43.29 ppm, respectively, with a corresponding JC-P coupling constant of 92 Hz. The four-coordinate anionic boron centre was supported by a sharp singlet in the 11B{1H} NMR spectrum at 0.83 ppm in addition to the 19F NMR spectrum bearing three signals, with a para-meta gap of 4.68 ppm. The carbonyl functionality was further confirmed by IR spectroscopy as a sharp band, representative of a C−O stretching frequency, at 1695 cm-1. These data in conjunction with elemental analysis confirmed the previously proposed formulation of 5-1. It should be noted that solid 5-1, when stored in the freezer at -35 ° was stable for a number of weeks without undergoing degradation; however, upon dissolution in

CD2Cl2, approximately 20 % conversion to the free FLP was observed in 30 minutes.

Characterization was best achieved in a sealed J. Young NMR tube under an atmosphere of CO2

in order to ensure complete conversion of the FLP to the CO2 adduct.

The decreased stability of 5-1 as compared to tBu3P(CO2)B(C6F5)3 is not surprising due to the

inherently weaker Lewis acidity of PhB(C6F5)2 relative to B(C6F5)3. Employing the Childs

Lewis acidity test, PhB(C6F5)2 was found to be considerably less Lewis acidic (0.54) as 120 compared to its B(C6F5)3 counterpart (0.68). Similarly, competition reactions of

stoichiometric quantities of PhB(C6F5)2 and B(C6F5)3 with 0.90 equivalents of benzaldehyde

resulted in sole observation of the B(C6F5)3−benzaldehyde adduct thereby confirming the 120 reduced Lewis acidity of PhB(C6F5)2 relative to B(C6F5)3. It should be noted that a longer and

therefore weaker B−O bond would contribute to the ease of CO2 liberation from 5-1.

160

Scheme 5.3 – Synthesis of (a) 5-1 and (b) 5-2.

210,250 Analogous reaction of the stronger Lewis acid ClB(C6F5)2 with PtBu3 and CO2 afforded a white solid, 5-2, in 79 % yield (Scheme 5.3, b). Doublets were observed by both 31P{1H} and 13 1 1 C{ H} NMR spectroscopy at 46.09 and 161.18 ppm, respectively, with a JC-P coupling constant of 95 Hz, supporting the presence of the P(CO2) fragment. Similarly, the anionic nature of the boron centre was confirmed as evidenced by a sharp singlet in the 11B{1H} NMR spectrum in addition to a narrow para-meta gap of 6.11 ppm, observed by 19F NMR spectroscopy (Figure 5.2).

Figure 5.2 – Excerpts from the 19F, 13C{1H} and 31P{1H} NMR spectra of 5-2.

161

IR analysis further supported the CO2 moiety with a sharp C−O stretching band visible at 1702 cm-1. This increase in stretching frequency, as compared to that of 5-1, is consistent with the shortening of the C−O double bond upon coordination to the more Lewis acidic species,

ClB(C6F5)2. An X-ray crystallographic study of single crystals of 5-2 was performed confirming the formulation as tBu3P(CO2)B(C6F5)2Cl (Figure 5.3).

Figure 5.3 – POV-Ray depiction of the molecular structure of 5-2. B: yellow-green, C: black, O: red, F: pink, P: orange, Cl: green. H atoms removed for clarity. Selected bond distances (Å) and angles (°). B(1)-Cl(1), 1.8860(17); B(1)-O(1), 1.5301(18); P(1)-C(13), 1.8884(14); O(1)-C(13), 1.2989(18); O(2)-C(13), 1.2032(17), C(18)-P(1)-C(13), 109.87(7); O(1)-C(13)-O(2), 128.02(13); O(2)-C(13)-P(1), 119.07(11); O(1)-C(13)-P(1), 112.76(10).

The B−O bond in 5-2 is noticeably shorter than in the related compound tBu3P(CO2)B(C6F5)3 (B-

O: 1.5474(15) Å), consistent with the increased Lewis acidity of ClB(C6F5)2 relative to B(C6F5)3. Tetrahedral geometries were noted for both the B and P centres (sum of bond angles - B: 656° and P: 657°) while the PCO2 moiety was found to have a distorted trigonal planar geometry.

Comparison of the C−O, C=O and P−C bond lengths of 5-2 with tBu3P(CO2)B(C6F5)3 revealed statistically indistinguishable metrical parameters, despite the substitution of the Lewis acid.

162

This species is also comparable to the inorganic salt [Me4N][Me(CO2)B(C6F5)3] which bears B−O, C−O and C=O bond lengths of 1.514(2), 1.324(2) and 1.217(2) Å319, respectively.

5-2 was found to be stable at room temperature, undergoing slow degradation via liberation of the bound CO2 moiety over a number of weeks. The compound is however, stable for a number of months when stored at -35 °C. This is vastly improved as compared to other

bis(pentafluorophenyl)alkylboranes where loss of CO2 is observed at ~ -15 °C (RB(C6F5)2: R = hexyl, -14 °C; R = Cy, -16 °C).310

5.2.2 Reactivity of tBu3P(CO2)B(C6F5)2Cl

In light of the vastly improved stability of 5-2 (due to the tightly bound CO2 fragment) over 5-1

and other boranes of the form RB(C6F5)2 (hexyl, Cy, norbornyl), 5-2 was investigated for further reactivity.

Initial investigations probed the possibility of halide abstraction to yield a salt of the form

[tBu3P(CO2)B(C6F5)2][X] which would provide a stabilized CO2 adduct by strengthening the

B−O bond to the CO2 moiety. Stoichiometric reactions of 5-2 with Ag(OSO2CF3),

[Ph3C][B(C6F5)4], K[B(C6F5)4] and NaBPh4 yielded no reaction as evidenced by NMR spectroscopy, even following reaction over 5 days. However, reaction of 5-2 with a single - equivalent of Me3Si(OSO2CF3) resulted in the facile and instantaneous extraction of Cl yielding

5-3 and Me3SiCl as a byproduct of reaction (Scheme 5.4).

Scheme 5.4 – Chloride abstraction from 5-2 by Me3SiOSO2CF3 yielding 5-3 and the reaction of

5-2 with Et3SiH to yield the classic Lewis acid-base adduct, 5-4.

Examination of the 11B{1H} NMR spectrum revealed a sharp singlet at 3.76 pm which was indicative of a quaternized boron centre (5-2: 2.13 ppm). This was further supported by a narrow

163

para-meta gap of 7.6 ppm. An additional singlet was observed in the 19F NMR spectrum at - 31 1 13 1 78.59 ppm representative of the OSO2CF3 fragment. P{ H} and C{ H} NMR spectroscopy, in 1 addition to IR analysis, confirmed the continued presence of the PCO2 moiety with a JC-P coupling constant of 93 Hz and a sharp IR band at 1706 cm-1. A crystallographic study confirmed the formulation as the product of Cl-triflate metathesis,

tBu3P(CO2)B(C6F5)2(OSO2CF3), 5-3 (Figure 5.4).

Figure 5.4 – POV-Ray depiction of the molecular structure of 5-3. B: yellow-green, C: black, O: red, F: pink, P: orange. H atoms removed for clarity. Selected bond distances (Å) and angles (°). P(1)-C(13), 1.8916(13); C(13)-O(2), 1.1993(16); C(13)-O(1), 1.3180(15); O(1)-B(1), 1.4978(16); B(1)-O(3), 1.5453(17); P(1)-C(13)-O(2), 116.84(10); P(1)-C(13)-O(1), 115.84(9); O(2)-C(13)-O(1), 127.10(12).

The metrical parameters of 5-3 revealed a considerable shortening of the B−O bond to 1.4978(16) Å, as compared to 5-2, which is consistent with the increased Lewis acidity of the borane B(C6F5)2(OSO2CF3) relative to the parent, 5-2 (B−O: 1.5301(18) Å).

Further attempts were made to derivatize 5-2 by reaction with Et3SiH in hopes of effecting Cl/H

exchange. A stoichiometric mixture of 5-2 and Et3SiH in CH2Cl2 was allowed to stir overnight.

164

Removal of the solvents resulted in the isolation of 5-4 as a white solid (Scheme 5.3). Room temperature NMR analysis revealed extremely broadened signals; therefore, NMR analysis was repeated at -80 °C. Singlets were observable in the 11B{1H} and 31P{1H} NMR spectra at -26.85 and 42.36 ppm, respectively. The resonance in the 11B{1H} NMR spectrum is distinctive of a 19 hydridoborate fragment. Inequivalent C6F5 rings were observed by F NMR spectroscopy which is representative of restricted rotation about the aryl C−B bonds. An absence of P−C coupling

suggested that the CO2 moiety had been lost and that 5-4 was the Lewis acid-base adduct,

(tBu3P)B(C6F5)2H. Restricted rotation about the P-C bond was additionally supported by the appearance of two methyl signals, in the 1H NMR spectrum, presenting in a 1:2 ratio and assignable to the tBu methyl groups.

Figure 5.5 – Monitoring of the conversion of 5-2 to 5-4 by 31P{1H} NMR spectroscopy. (*) 5-2,

(*) tBu3P(CO2)B(C6F5)2H and (*) 5-4.

The reaction was repeated at room temperature and monitored by 31P{1H} NMR spectroscopy for a period of 90 minutes. During this time 5-2 was suspected to undergo Cl/H exchange in the presence of Et3SiH, generating the intermediate species tBu3P(CO2)B(C6F5)2H. This species,

165

which is expected to be unstable with respect to CO2 loss, is short-lived and releases CO2 into

solution. The respective Lewis acidity and basicity of the free HB(C6F5)2 and PtBu3 moieties were quenched yielding the classic Lewis acid-base adduct, 5-4. The 31P{1H} NMR spectra, documenting this conversion, are shown in Figure 5.5. The instability and subsequent liberation of CO2 from the zwitterion tBu3P(CO2)B(C6F5)2H is not surprising as direct exposure of CO2 to the adduct 5-4 results in no reaction.

Due to the observed stability of 5-2, the ability to promote Lewis acid exchange at the oxygen 320 atom of the tBu3P(CO2) fragment was investigated. A stoichiometric reaction of Al(C6F5)3

(complexed with half an equivalent of toluene) and 5-2 was undertaken in CH2Cl2. Examination 19 210 of the crude F NMR spectrum revealed liberation of ClB(C6F5)2. Repetition of the 13 experiment employing the CO2 isotopologue of 5-2 resulted in the isolation of a white solid, 5-5, following manipulation with pentane (Scheme 5.5). A sharp doublet was observable in the 31P{1H} NMR spectrum at 46.65 ppm, which is only slightly downfield relative to the analogous resonance for 5-2 (46.09 ppm). The 27Al{1H} spectrum of 5-5 was found to be completely silent. The 19F NMR spectrum indicated that 5-2 had been completely consumed throughout the course of the reaction but had been replaced by the alane, as evidenced by three new resonances at - 123.12, -155.83 and 163.39 ppm, respectively. The identity of 5-5 was suspected to be the

product of direct Lewis acid exchange of ClB(C6F5)2 to give tBu3P(CO2)Al(C6F5)3. Previous

reports by the Stephan group have demonstrated that AlX3 (X = Cl, Br, I) reacts with PMes3 and 311 CO2 in a 2:1:1 fashion yielding a stable CO2 adduct. As the ratio of Lewis acid to base in 5-5

could not be ambiguously confirmed, a reaction employing two equivalents of Al(C6F5)3 to one 19 equivalent of PtBu3 was conducted. The F NMR spectrum supported the formulation of 5-5 as

the 1:1:1 acid:base:CO2 adduct as evidenced by the generation of the adduct in addition to the

observance of one equivalent of unreacted Al(C6F5)3. It should be noted that 5-5 can also be

synthesized via direct methods by exposing a mixture of the adduct (tBu3P)Al(C6F5)3 to an atmosphere of CO2, affording comparable yields to synthesis via exchange (exchange: 65 %, direct: 62 %).

166

Scheme 5.5 – Synthesis of 5-5 and 5-6.

To further probe the exchange capabilities of 5-2, a stoichiometric reaction was conducted

employing a transition metal-based Lewis acid. The cationic Ti salt [Cp2TiMe][B(C6F5)4] was 321 pre-formed in C6H5Br via the reaction of Cp2TiMe2 with [Ph3C][B(C6F5)4] prior to addition of dry 5-2 (Scheme 5.4). The solution was found to undergo a colour change from a deep red/brown to a vibrant red. Manipulation of the resulting solution yielded an orange solid, 5-6, which was isolated in 96 % yield. An examination of the 1H NMR spectrum confirmed the

presence of the Cp2Ti moiety as evidenced by a singlet at 6.65 ppm, assignable to the cyclopentadienyl rings in addition to a doublet at 1.66 ppm assignable to the tert-butyl groups of the phosphine. Similarly, examination of the 11B{1H} NMR spectrum supported the presence of the tetrakis(pentafluorophenyl)borate ion. Additionally, there was no evidence for unreacted 5-2 13 or residual ClB(C6F5)2. The analogous reaction employing labeled CO2 5-2 was conducted and 31 1 1 revealed a doublet in the P{ H} NMR spectrum at 49.09 ppm with a JP-C coupling constant of 13 1 86 Hz. Retention of the tBu3P(CO2) moiety was further supported by a doublet by C{ H} NMR spectroscopy, at 163.59 ppm and with an absorption band in the IR spectrum at 1670 cm-1,

indicative of the C=O stretching mode. It was noted however that the Me group of the Cp2TiMe was unobservable by 1H NMR which led to the question of whether ligand scrambling had occurred. A crystallographic study of single crystals of 5-6 unambiguously confirmed the formulation as the salt of Lewis acid exchange, [tBu3P(CO2)TiCp2Cl][B(C6F5)4] (Figure 5.6).

Presumably, following Lewis acid substitution and liberation of free ClB(C6F5)2 into solution, a Cl/Me exchange between the B and Ti centres occurred resulting in the generation of the

Cp2TiCl core in addition to MeB(C6F5)2. This exchange is not surprising as the transmetallation

of Cp2ZrR2 and ClB(C6F5)2 is an established route to generating boranes of the form RB(C6F6)2 where R does not bear any β-hydrogens.322 It should be noted that 5-6 is stable at room temperature for approximately seven days in solution without undergoing appreciable decomposition.

167

Figure 5.6 – POV-Ray depiction of the molecular structure of 5-6. C: black, O: red, F: pink, P:

orange, Cl: green, Ti: grey. H atoms and B(C6F5)4 ion removed for clarity. Selected bond distances (Å) and angles (°). Ti(1)-O(1), 2.0156(18); Ti(1)-Cl(1), 2.3171(9); O(1)-C(11), 1.270(3); O(2)-C(11), 1.220(3); P(1)-C(11), 1.896(3); O(1)-Ti(1)-Cl(1), 93.09(6); Ti(1)-O(1)- C(11), 135.77(18); O(1)-C(11)-O(2), 128.2(3); O(2)-C(11)-P(1), 116.3(2); O(1)-C(11)-P(1), 115.5(2).

The coordination sphere of the cation was found to involve the coordination of the tBu3P(CO2) fragment, formerly from 5-2. The Ti−Cl bond length was found to be 2.3171(9) Å while the C−O and C=O bonds were found to be 1.270(3) and 1.220(3) Å, where the former is both considerably shortened as compared to the parent zwitterion 5-2 (C-O: 1.2989(18) Å; C=O: 1.2032(17) Å)

consistent with the increased Lewis acidity of the titanocene fragment relative to ClB(C6F5)2. The P−C bond length was found to be 1.896(3) Å while the Ti−O bond length was found to be 2.0156(18) Å. The Ti−O−C and O−C−O angles were determined to be 135.8(2)° and 128.2(3)° while the O−C−P angles were found to 116.3(2)° and 115.5(2)°, respectively. The Ti−O−C angle of 135.77(18)° is notably larger than the comparable B−O−C angle in either 5-2 or 309 tBu3P(CO2)B(C6F5)3 (5-2: 120.69(11)°; tBu3P(CO2)B(C6F5)3: 119.44(9)°) which would indicate some degree of interaction between the oxygen lone pair and a vacant metal d-orbital, which is absent in the main group species. An examination of the metrical parameters shows

little distortion of the [Cp2TiCl] moiety as evidenced by the Cp(cent)−Ti−Cp(cent) angle of 131.88°,

which is comparable to the known species Cp2Ti(OEt)Cl bearing an analogous 323 Cp(cent) −Ti−Cp(cent) angle of 130.5°.

168

This reactivity is reminiscent of the activation of CO2 employing a cationic zirconocene- phosphinoaryloxide catalyst, in an intramolecular fashion, as described by Wass et al. (Scheme 315 324 5.6, a). Earlier reports by Majoral illustrated similar insertion of CO2 and its heavier

congener CS2 between a Zr Lewis acidic centre and a P Lewis base where no formal Zr−P interaction was previously observable, yielding five-coordinate anionic zirconium complexes (Scheme 5.6, b). Both complexes published by Wass and Majoral display 13C NMR and IR data which are closely related to those experimentally determined for 5-6 (Table 5.1).

Scheme 5.6 – Intramolecular activation of CO2 by (a) a metallocenium cation and phosphine and

(b) a neutral zirconocene and phosphine (Cp* = C5Me5).

169

Table 5.1 – Selected 13C NMR data, coupling constants and IR stretching frequencies.

13 1 -1 Compound δ C (ppm) JC-P (Hz) IR band (cm )

161.9a 85a -

153.9b 88b 1694b

5-6 163.6 86 1670

a 315 b 324 see Wass et al. ; see Majoral et al. ; Cp* = C5Me5

Mechanistically, it is suspected that species 5-5 and 5-6 are generated via the transient + interaction of the incoming stronger Lewis acid, either Al(C6F5)3 or [Cp2TiMe ], with the oxygen

atom of the carbonyl group, generating an intermediate resembling that of Mes3P(CO2)(AlX3)2

(X = Cl, Br, I), prior to departure of the weaker Lewis acid, ClB(C6F5)2 (Scheme 5.7).

Scheme 5.7 – Mechanism of Lewis acid exchange in 5-2.

170

5.3 Conclusions

Weaker boron-containing Lewis acids, as compared to tris(pentafluorophenyl)borane, are shown

to be active in the complexation of CO2 in combination with PtBu3 to yield zwitterionic species

of the general form tBu3P(CO2)BR3. Although these complexes are readily formed in solution

nearly instantaneously upon exposure to an atmosphere of CO2, they have been found to be

generally unstable with respect to CO2 liberation, resulting in unsuitable candidates for further

reactivity. The tBu3P(CO2) moiety has been demonstrated to be a stable and transferable

fragment that can readily undergo transfer from tBu3P(CO2)B(C6F5)2Cl to acids of greater Lewis acidity, yielding highly stable compounds with potential for further functionalization.

5.4 Experimental Section

5.4.1 General Considerations

All manipulations were carried out under an atmosphere of dry, O2-free N2 employing standard

Schlenk-line and glovebox techniques. Solvents (CH2Cl2, pentane and hexanes) were dried by employing a Grubbs-type column system (Innovative Technology), degassed and stored under

N2. Bromobenzene was dried over CaH2 and distilled under N2 while cyclohexane was dried

over sodium and benzophenone and similarly distilled under N2. CD2Cl2 was vacuum transferred

from CaH2, degassed and stored under N2. Me3SiOSO2CF3 (Aldrich), Et3SiH (Aldrich), and 13 [Ph3C][B(C6F5)4] (Aldrich), PtBu3 (Strem Chemicals), CO2 (Aldrich, 99.8 %) and CO2 13 18 126 210,250 (Aldrich, 99 atom % C, <3 atom % O) were used as received. PhB(C6F5), ClB(C6F5)2, 66 320 321 HB(C6F5)2, 0.5tol·Al(C6F5)3 and Cp2TiMe2 were prepared according to literature procedures. 1H, 11B, 13C, 19F, 27Al and 31P NMR spectra were recorded at 25 °C, unless otherwise stated, on a Varian NMR System 400 MHz or Bruker Avance III 400 MHz spectrometer and 1 13 were referenced using (residual) solvent resonances relative to SiMe4 ( H and C) or to an 11 19 27 31 external standard ( B: (Et2O)BF3, F: CFCl3, Al: Al(NO3)3, P: 85% H3PO4). Chemical shifts are reported in ppm and coupling constants as scalar values in Hz. Combustion analyses were performed in house employing a Perkin-Elmer CHN Analyzer.

171

5.4.2 Syntheses

Synthesis of 5-1. A 100 mL Schlenk was charged with PhB(C6F5)2

(0.100 g, 0.237 mmol) and PtBu3 (0.048 g, 0.237 mmol) in pentane (10 mL) and bromobenzene (1 mL). The reaction mixture was degassed

and backfilled with CO2. The reaction instantly became cloudy and was stirred for a further 12 hours. The white precipitate was allowed to settle and the solvent was decanted. The product was washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Yield: 0.087 g (55 %). 1H 3 3 3 NMR (CD2Cl2): δ 7.68 (d, JH-H = 7.4 Hz, o-C6H5); 7.30 (t, JH-H = 7.6 Hz, m-C6H5); 7.15 (t, JH- 3 11 13 1 H = 7.3 Hz, p-C6H5); 1.16 (d, JH-P = 14 Hz, PtBu). B NMR (CD2Cl2): δ 0.83 (s). C{ H} 1 1 NMR (CD2Cl2) partial: δ 160.79 (d, JC-P = 92 Hz, PCO2); 147.35 (br d, JC-F = 240 Hz, o-C6F5); 1 1 138.42 (br d, JC-F = 253 Hz, p-C6F5); 136.42 (br d, JC-F = 243 Hz, m-C6F5); 131.35 (s, p-C6H5); 1 126.49 (s, o-C6H5); 124.99 (s, m-C6H5); 39.44 (d, JC-P = 20 Hz, PC{CH3}3); 29.25 (s, 19 3 3 PC{CH3}3). F NMR (CD2Cl2): δ -130.54 (d, JF-F = 23 Hz, o-C6F5), -160.02 (t, JF-F = 21 Hz, 31 1 1 p-C6F5), -164.70 (m, m-C6F5). P{ H} NMR (CD2Cl2): δ 43.29 (d, JP-C = 92 Hz, PCO2). -1 IR(KBr): ν 1695 cm (C=O). Anal. Calcd. for C31H32BF10O2P: C, 55.67; H, 4.83. Found: C, 55.60; H, 5.09 %.

Synthesis of 5-2. A 50 mL Schlenk was charged with ClB(C6F5)2 (0.150

g, 0.394 mmol) and PtBu3 (0.080 g, 0.395 mmol) in bromobenzene (10

mL). The bright yellow solution was degassed and backfilled with CO2 (1 bar). The reaction was then stirred for two hours at room temperature. At this time, pentane (20 mL) was added precipitating an off-white solid. The solvent was decanted and the crude product was washed with pentane (3 x 5 mL). The product was then dried in vacuo for two hours. Crystals suitable for X-ray diffraction were grown from a layered CH2Cl2/cyclohexane 1 3 solution at 25 °C. Yield: 0.193 g (79 %). H NMR (CD2Cl2): δ 1.68 (d, 27H, JH-P = 14 Hz, 11 1 13 1 1 PtBu). B{ H} NMR (CD2Cl2): δ 2.13 (s). C{ H} NMR (CD2Cl2): δ 161.18 (d, JC-P = 95 Hz, 1 1 PCO2); 148.08 (br d, JC-F = 242 Hz, o-C6F5); 140.31 (br d, JC-F = 250 Hz, p-C6F5); 137.71 (br d, 1 1 JC-F = 242 Hz, m-C6F5); 119.43 (br s ipso-C6F5); 41.55 (d, JC-P = 20 Hz, PC{CH3}3); 30.67 ( s, 19 3 4 PC{CH3}3). F NMR (CD2Cl2): δ -133.78 (dd, JF-F = 24 Hz, JF-F = 8.6 Hz, o-C6F5); -159.48 (t, 3 31 1 1 JF-F = 20 Hz, p-C6F5); 165.59 (m, m-C6F5). P{ H} NMR (CD2Cl2): δ 46.09 (d, JP-C = 95 Hz).

172

-1 IR(KBr): ν 1702 cm (C=O). Anal. Calcd. for C25H27BClF10O2P: C, 47.88; H, 4.34. Found: C, 47.89; H, 4.46 %.

Synthesis of 5-3. A 20 mL scintillation vial was charged with 5-2 (0.080

g, 0.128 mmol) and CH2Cl2 (5 mL). Me3SiOSO2CF3 (23.1 μL, 0.128 mmol) was added to the clear and colourless solution in a dropwise fashion. The resulting reaction was left stirring for a period of one hour. At this time, the reaction was faintly yellow. All volatiles were then removed in vacuo yielding a white solid. The product was washed with hexanes (10 mL) and dried in vacuo for one hour. Yield: 0.094 g

(99 %). Crystals suitable for X-ray diffraction were grown from a layered solution of CH2Cl2 1 3 11 1 and hexanes at -35 °C. H NMR (CD2Cl2): δ 1.68 (d, 27H, JH-P = 15 Hz, PtBu). B{ H} NMR 13 1 1 (CD2Cl2): δ 3.76 (s). C{ H} NMR (CD2Cl2): δ 162.08 (d, JC-P = 93 Hz, PCO2); 148.48 (br d, 1 1 1 JC-F = 245 Hz, o-C6F5); 141.08 (br d, JC-F = 249 Hz, p-C6F5); 137.62 (br d, JC-F = 249 Hz, m- 1 C6F5); 120.62 (s, OSO2CF3); 115.37 (br s, ipso-C6F5); 41.68 (d, JC-P = 19 Hz, PC{CH3}3); 30.64 19 3 ( s, PC{CH3}3). F NMR (CD2Cl2): δ -78.59 (s, 3F, OSO2CF3); -134.40 (dd, 6F, JF-F = 23 Hz, 4 3 31 1 JF-F = 6.8 Hz, o-C6F5); -157.34 (t, 3F, JF-F = 20 Hz, p-C6F5); -164.94 (m, 6F, m-C6F5). P{ H} 1 -1 NMR (CD2Cl2): δ 48.88 (d, JP-C = 93 Hz). IR(KBr): ν 1706 cm (C=O). Anal. Calcd. for

C26H27BF13O5PS: C, 42.16; H, 3.68. Found: C, 41.67; H, 4.04 %.

Synthesis of 5-4: A 20 mL scintillation vial was charged with HB(C6F5)2

(0.050 g, 0.145 mmol) in CH2Cl2 (3 mL). A second vial was charged with

PtBu3 (0.029 g, 0.143 mmol) in CH2Cl2 (2 mL). The phosphine was added to the solution of the borane yielding a clear and colourless solution. The reaction was allowed to stir at room temperature for 10 minutes before the removal of the volatiles. The resulting white solid was 1 dried in vacuo for an hour. Yield: 0.077 g (98 %). H NMR (CD2Cl2, -80 °C): δ 3.66 (br s, 1H, 3 3 11 1 BH); 1.51 (d, 9H, JH-P = 9.5 Hz, PtBu); 1.32 (d, 18H, JH-P = 12.7 Hz, PtBu2). B{ H} NMR 13 1 1 (CD2Cl2, -80 °C): δ -26.85 (br s). C{ H} NMR (CD2Cl2, -80 °C) partial: δ 147.65 (br s, JC-F = 1 1 245 Hz, C6F5); 146.88 (br s, JC-F = 236 Hz, C6F5); 138.56 (br s, JC-F = 246 Hz, C6F5); 136.69 1 1 (br s, JC-F = 245 Hz, C6F5); 119.00 (s, ipso-C6F5); 39.04 (d, JC-P = 20 Hz, PC{CH3}3); 30.96 (s, 19 PC{CH3}3); 29.73 (s, PC{CH3}3). F NMR (CD2Cl2, -80 °C): δ -123.14 (s, 2F, o-C6F5); -126.24 3 (s, 2F, o-C6F5); -159.16 (t, 2F, JF-F = 21 Hz, p-C6F5); -163.06 (m, 2F, m-C6F5); -164.28 (m, 2F,

173

31 1 m-C6F5). P{ H} NMR (CD2Cl2, -80 °C): δ 42.36 (s). Anal. Calcd. for C25H28BF10O2P: C, 52.54; H, 5.15. Found: C, 52.43; H, 5.28 %.

Synthesis of 4-5 Method I. A 20 mL scintillation vial was charged with 5-2 (0.080 g, 0.128 mmol) in bromobenzene (5 mL) while a second 20

mL vial was charged with 0.5tol·Al(C6F5)3 (0.075 g, 0.129 mmol) in bromobenzene (5 mL). The second solution was added to the first in a dropwise fashion at room temperature. The reaction became faintly yellow upon addition. The resulting solution was stirred for 30 minutes. Following, the solution was divided amongst two 20 mL vials and pentane (15 mL) was added slowly precipitating a white solid. The product was allowed to settle and the supernatant was decanted. The material was washed with pentane (2 x 5 mL) and dried in vacuo for 30 minutes. Yield: 0.641 g (65 %). Method II: A 50 mL tube bomb was charged with 0.5tol•Al(C6F5)3 (0.080 g, 0137 g) and PtBu3 (0.028 g, 0.138 mmol) in bromobenzene (7

mL). The solution was degassed and then backfilled with CO2. The reaction was allowed to stir

under an atmosphere of CO2 for 16 hours at room temperature. At this time, the solution was transferred to a vial and pentane (10 mL) was added slowly precipitating a white solid. The precipitate was allowed to settle and then the supernatant was decanted. The product was washed with pentane ( 2 x 5 mL) and dried for 30 minutes in vacuo. Yield: 0.066 g (62 %). 1H 3 13 1 1 NMR (CD2Cl2): δ 1.65 (d, 27H, JH-P = 14 Hz, PtBu3). C{ H} NMR (CD2Cl2): δ 162.61 (d, JP- 1 1 C = 89 Hz, PCO2); 149.98 (br d, JC-F = 239 Hz, o-C6F5); 143.54 (br d, JC-F = 250 Hz, p-C6F5); 1 1 136.80 (br d, JC-F = 253 Hz, m-C6F5); 125.20 (br s, ipso-C6F5); 41.16 (d, JC-P = 20 Hz, 19 PC{CH3}3}); 30.74 (s, PC{CH3}3). F NMR (CD2Cl2): δ -123.12 (m, 6F, o-C6F5); -155.83 (t, 3 27 1 31 1 3F, JF-F = 19 Hz, p-C6F5); -163.39 (m, 6F, m-C6F5). Al{ H} NMR (CD2Cl2): silent. P{ H} 1 -1 NMR (CD2Cl2): δ 47.72 (d, JP-C = 89 Hz). IR (KBr pellet): ν 1686 cm (C=O). Anal. Calcd. for

C31H27AlF15O2P: C, 48.08; H, 3.51. Found: C, 47.69; H, 3.60 %.

Synthesis of 5-6. A 20 mL scintillation vial was charged with Cp2TiMe2 (0.030 g, 0.144 mmol) in bromobenzene (2 mL). A second 10 mL vial

was charged with [Ph3C][B(C6F5)4] (0.133 g, 0.144 mmol) in bromobenzene (5 mL). The solution of trityl borate was added to the titanocene solution yielding a deep red/brown reaction mixture, which was stirred for a further 5 minutes. At this time, dry 5-2 (0.090 g, 0.144 mmol) was added and the resulting solution progressively lightened in colour to a vibrant red over a 30 second period of time. The reaction

174

mixture was allowed to stir at room temperature for an hour after which the reaction was divided amongst two vials and hexanes (15 mL) were added to both fractions precipitating an orange solid. The supernatant was decanted and the resulting solid was washed with further hexanes (2 x 10 mL). The solvents were once again decanted and the solid was dried in vacuo for a period

of 1 hour. Crystals suitable for X-ray diffraction were grown from a solution of CH2Cl2 layered 1 3 with hexanes. Yield: 0.158 g (96 %). H NMR (CD2Cl2): δ 6.65 (s, 10H, Cp); 1.66 (d, 27H, JH- 11 1 13 1 P = 14 Hz, PtBu3). B{ H} NMR (CD2Cl2): δ -16.65 (s). C{ H} NMR (CD2Cl2): δ 163.59 (d, 1 1 1 JC-P = 86 Hz, PCO2); 148.07 (br d, JC-F = 244 Hz, o-C6F5); 138.16 (br d, JC-F = 244 Hz, p- 1 C6F5); 136.23 (br d, JC-F = 244 Hz, m-C6F5); 124.65 (br s, ipso-C6F5); 120.53 (s, Cp); 41.56 (d, 1 19 JC-P = 19 Hz, PC{CH3}3); 30.97 (s, PC{CH3}3). F NMR (CD2Cl2): δ -133.02 (s, 8F, o-C6F5); - 3 31 1 163.03 (t, 4F, JF-F = 20 Hz, p-C6F5); 167.45 (m, 8F, m-C6F5). P{ H} NMR (CD2Cl2): δ 49.09 1 -1 (d, JP-C = 86 Hz ). IR(KBr): ν 1670 cm (C=O). Anal. Calcd. for C47H37BF20O2PTi: C, 49.54; H, 3.28. Found: C, 49.10; H, 3.20 %.

5.4.3 X-ray Crystallography

5.4.3.1 X-ray Data Collection and Reduction

In preparation for analysis, crystals were first coated with Paratone-N oil in a glovebox and were

subsequently mounted on a MiTegen Micromount and placed in a N2 stream in order to maintain a dry and oxygen-free sample environment. Data collection was performed on a Bruker Apex II diffractometer and data collection strategies were determined employing Bruker provided Apex software. Optimization was performed in order to yield >99.5 % complete data to a minimum 2θ value of 55 °. All data sets were collected at 150(±2) K unless otherwise stated. The acquired frames were integrated employing the Bruker SAINT software package employing a narrow frame algorithm. Absorption corrections were conducted employing the empirical multi-scan method (SADABS).

5.4.3.2 Structure and Refinement

175

2 employing full-matrix least squares techniques on F, minimizing the function ω (Fo-Fc) where 2 2 weight (ω) equates to 4Fo /2σ (Fo ) and Fo and Fc are equal to the observed and the calculated structure factor amplitudes. In the final cycles of each refinement, all non-hydrogen atoms were assigned anisotropic temperature factors in the absence of disorder or insufficient data. In the instance of the latter case the atoms were then treated isotropically. Unless noted, the C-H atom positions were calculated and allowed to ride on the carbon to which they are bonded with an assumed C-H bond length of 0.95 Å. Temperature factors pertaining to the H-atoms were fixed at 1.20 times the isotropic temperature factor of the C-atom to which they are bound. The H- atom contributions were calculated however never refined. The locations of the largest peaks in the final difference Fourier map calculations as well as the magnitude of the residual electron densities in each case were of no chemical significance.

176

5.4.3.3 Tables of Crystallographic Data

Table 5.2 – Selected crystallographic data for 5-2, 5-3 and 5-6.

5-2 5-3 5-6

Formula C25H27BClF10O2P C26H27BF13O5PS C47H37BClF20O2Ti Formula wt 626.70 740.32 1138.90 Crystal system Monoclinic Triclinic Triclinic Space group C2/c P-1 P-1 a (Å) 32.2497(14) 9.7343(4) 10.7349(9) b (Å) 8.6303(4) 10.7665(4) 14.4145(12) c (Å) 22.2895(10) 14.9487(6) 15.0880(11) α (deg) 90 98.1230(10) 85.942(4) β (deg) 119.491(2) 97.127(2) 88.205(4) γ (deg) 90 106.6580(10) 81.850(5) V (Å3) 5399.9(4) 1463.34(10) 2304.8(3) Z 8 2 2 T (K) 150(2) 150(2) 150(2) d (calc) gcm-3 1.542 1.680 1.641 Abs coeff, μ, mm-1 0.293 0.287 0.400 Data collected 23861 24536 35285 R int 0.0293 0.0226 0.0578 # of indpndt reflns 6231 6690 10393

Reflns Fo≥2.0σ(Fo) 5111 5865 6346 Variables 370 444 707 R (>2σ) 0.0324 0.0303 0.0483

wR2 0.0861 0.0835 0.1080 Goodness of fit 1.027 1.043 1.011

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

177

Chapter 6 Sequestration of Nitrous Oxide by Main Group Frustrated Lewis Pairs

6.1 Introduction

6.1.1 The Environmental Implications of Nitrous Oxide

Over the past decades, global warming has garnered a considerable degree of international concern due to the inherent environmental implications.325,326 Much focus has been placed on the negative environmental impact of rising carbon dioxide (CO2) and methane (CH4) levels.

However, nitrous oxide (N2O) has been identified as a potent ozone-depleting substance whose presence has been directly linked to the progressive destruction of the stratospheric ozone 325,327 layer. In fact, recent results have demonstrated that N2O is 300 times more potent a 327,328 greenhouse gas than CO2 and can persist for upwards of 150 years in the stratosphere. The levels of N2O have risen steadily from the time of industrialization from 270 ppb to 319 ppb, in 2005.329 This increase is vastly attributed to anthropogenic sources that include, but are not limited to, the agricultural sector, fossil fuel consumption and manufacturing.330 Recently, nitrous oxide has been acknowledged as the utmost destructive ozone depleting substance. As a 325,330 result, the yearly increase in the concentration of N2O by 0.5-0.9 ppb per volume is a cause for both concern and action.

6.1.2 Applications of Nitrous Oxide in Synthesis

Due to the inherent kinetic stability of N2O, very few applications have been elucidated. While

not thoroughly investigated, the high oxidation potential renders N2O an excellent candidate for the oxidation of organic substrates.331 Oxidation can be achieved employing either gaseous or

liquid N2O, resulting in an arguably environmentally green methodology as N2 is the only byproduct of reaction. Indeed, nitrous oxide has been employed for the oxidation of olefins to the corresponding carbonyl compounds under relatively forceful conditions (10 bar N2O and 150- 250 °C) (Scheme 6.1, a).332 This route had a broad substrate scope allowing for the synthesis of carbonyl compounds from aliphatic, cyclic and heterocyclic alkene precursors.333 It has also found application in reaction with reactive species such as triethylborane,334 silaethenes and

178

disilenes.335-340 From an inorganic perspective, the group of Jessop has demonstrated the use of

supercritical N2O (pressures in excess of 100 bar) at more reasonable temperature (below 100 341 °C) for the effective oxidation of PPh3 to the corresponding phosphine oxide (Scheme 6.1, b).

Scheme 6.1 – Examples of the oxidation of (a) alkenes to the related carbonyl compounds333 and 341 (b) phosphines to the related phosphine oxides employing N2O.

6.1.3 Nature: The Best Model

Methods to both sequester and decompose nitrous oxide to inert molecular nitrogen and water are desirable and the focus of ongoing intellectual pursuits. In nature, the metalloenzyme nitrous oxide reductase serves as a crucial component of the microbial denitrification process and is an essential element of the nitrogen cycle.328,342-344 Copper-based metalloenzymes employ

μ-sulfido-tetracopper clusters (Cu4S), for the trapping of, and subsequent breakdown of N2O in the terminal step of the denitrification process.344,345 These bacteria are fundamental to the

agricultural sector for the decomposition of N2O to water, according to the following equation: + - 346 (N2O + 2H + 2e → H2O + N2).

Recently, the Tolman group has successfully demonstrated the reduction of N2O employing a synthetic analogue of the naturally-found nitrous oxide reductase. This species presents as a

tricopper-disulfido cluster capable of the reduction of N2O reduction to N2 at low temperatures.347

6.1.4 Sequestration of Nitrous Oxide by Main Group Systems

To date, few reports (either synthetic or computational) detailing the sequestration of intact nitrous oxide have been documented in the literature. Of those instances, observed complexation has generally been described in the presence of a transition metal centre (Chapter 7).348 Currently, only a handful of examples demonstrating the activation of nitrous oxide, by main group systems, have been documented.349-351

179

Recently, the ability to utilize decorated fullerenes for the purposes of both gas capture and storage has garnered a considerable degree of interest.352-358 Wang et al. employed DFT

computational methods to study the application of C60 fullerenes in the sequestration of both CO2 and N2O. Both gases were found to only weakly adsorb to the surface of a bare C60 molecule 359 whereas in the presence of calcium-doped C60, CO2 and N2O adsorption was greatly enhanced.

Superior N2O adsorption was observed as compared to CO2 and was thought to result from the

more polar nature of the N2O molecule. Despite these findings, no synthetic attempts to employ

C60 as a N2O trapping agent have been documented to date.

In 2011, Severin et al. demonstrated the elegant use of commercially available bulky N- heterocyclic carbenes for the unassisted capture of intact N2O. In this instance, the basic and

sterically demanding carbene acts to bind the N2O fragment via C-N bond formation (scheme 6.2).351 These adducts demonstrated unique reactivity and were easily alkylated resulting in the

N−N bond scission of the activated N2O moiety.

351 Scheme 6.2 – The sequestration of N2O by a basic and bulky N-heterocyclic carbene.

6.2 Results and Discussion

6.2.1 Reactions of RB(C6F5)2 and PtBu3 with N2O

As has been previously documented, the combination of sterically encumbered perfluoroarylboranes and bulky phosphines preclude the formation of typical Lewis acid-base adducts rendering the unquenched acidic and basic sites available for further reactivity.1 As these

180

systems have demonstrated the ability to effect unique activations, the sequestration of the

otherwise unreactive N2O moiety with FLPs was attempted.

Initial investigations of PhB(C6F5)2 under an atmosphere of N2O revealed that there is no 19 11 1 interaction detectable by F and B{ H} NMR spectroscopy. Similarly, studies of PtBu3 and

N2O revealed no immediate interaction, although 50 % conversion to tri-tert-butylphosphine oxide was observed over 16 hours.

Subsequently, a mixture of PhB(C6F5)2 and PtBu3 was exposed to one atmosphere of N2O for a period of 12 hours and resulted in the recovery of a white solid, 6-1, in 76 % (Scheme 6.3). The 19F NMR spectrum revealed the presence of three distinct resonances pertaining to two

equivalent C6F5 rings. The para-meta gap was noted to be quite narrow (∆δm-p = 4.74 ppm), characteristic of a four-coordinate boron centre.12 Additionally, a sharp signal at 3.27 ppm was observed in the 11B{1H} further corroborating the four-coordinate nature of the boron centre. The 31P{1H} NMR spectrum revealed a signal at 67.26 ppm, downfield from the free phosphine at 63.3 ppm.360

In order to further probe the relationship between the N2O moiety and the B/P pair, the above reaction was repeated employing isotopically enriched 15N15NO, allowing for the reaction to be monitored by 15N NMR spectroscopy. Examination of a NMR scale solution of the B/P pair after 15 having been exposed to N2O for a period of 12 hours revealed the presence of a doublet of doublets in the 31P{1H} NMR spectrum. This was in stark contrast to the singlet previously 14 1 observed employing N2O. Coupling constants of 59 and 19 Hz were assigned to both JP-N and 2 JP-N couplings thereby confirming the presence of the P(N2O) moiety. Similarly, two doublets of doublets were observable in the 15N NMR spectrum at 575.35 and 375.57 ppm, respectively, 1 15 with a JN-N coupling constant of 16 Hz. Free N2O gives rise to two signals in the N NMR 1 361 spectrum at 218 and 135 ppm with a JN-N coupling constant of 8.1 Hz. Comparison of the

chemical shifts and coupling constants of free N2O to those of the isotopologue of 6-1 indicates

that substantial perturbation of the N2O moiety has occurred in the presence of the frustrated Lewis pair. Together, the above data were consistent with two chemically inequivalent nitrogen

atoms and signaled toward the formulation of 6-1, tBu3P(N2O)B(C6F5)2Ph.

181

Scheme 6.3 – Synthesis of 6-1.

Recrystallization of the crude product from CH2Cl2 and cyclohexane yielded crystals suitable for

X-ray diffraction, confirming 6-1 as tBu3P(N2O)B(C6F5)2Ph (Figure 6.1). The solid-state

structure of 6-1 revealed that the N2O moiety was trapped as the intact species between the

Lewis acidic and basic centres in a 1,3−PNNOB binding mode where the PtBu3 and the

OB(C6F5)2Ph fragments are orientated in a trans disposition relative to the N−N double bond. The N−N and N−O bond lengths of 6-1 were found to be 1.2602(8) and 1.3270(8) Å, respectively, consistent with both an N−N double bond and N−O single bond. These bond

lengths were lengthened as compared to those of a free molecule of N2O (N-N: 1.127 Å; N-O: 1.186 Å)362, indicating a reduction in bond order upon complexation.

182

Figure 6.1 – POV-Ray depiction of the molecular structure of 6-1. B: yellow-green, C: black, N: aquamarine, F: pink, O: red, P: orange. H atoms removed for clarity. Selected bond distances (Å) and angles (°). P(1)-N(1), 1.707(6); N(1)-N(2), 1.2602(8); N(2)-O(1), 1.3270(8); O(1)-B(1), 1.5475(9); P(1)-N(1)-N(2), 112.85(5); N(1)-N(2)-O(1), 111.68(6); N(1)-O(1)-B(1), 111.61(5).

The B/P−N2O adduct is reminiscent of structurally related phosphazides, of the general form

R3P(N1N2N3)R’. Phosphazides, which were previously thought to be intermediates to Staudinger aza-ylides, are isolable when R’ is an aryl group bearing ortho electron-withdrawing 363 substituents. Among phosphazides of this type, the N1−N2 bond is commonly found to be approximately 1.34 Å,363 consistent with a N−N single bond, which is considerably lengthened as compared to the N−N bond in 6-1 (1.2602(8)) (Figure. 6.2).

Figure 6.2 – Structures of (a) a stabilized phosphazide and (b) a B/P adduct of N2O.

183

As has been documented in numerous examples of small molecule activation by frustrated Lewis pairs, the ability to sequester small molecules lies with the ability to strike a fine balance between the electronic and steric factors of both the Lewis acidic and basic fragments.1 It was therefore of interest to investigate an array of boranes for activation of N2O with PtBu3 in attempts to define the limits of Lewis acidity necessary to effect reactivity.

180 In a parallel fashion, the less Lewis acidic boranes, MesB(C6F5)2 154 (Mes = C6H2(CH3)3) and (C6F5O)B(C6F5)2 were combined with stoichiometric amounts of

PtBu3 in bromobenzene under an atmosphere of N2O, and stirred for 12 hours at room

temperature. Following reaction termination, the zwitterionic products of N2O activation, 6-2 and 6-3, were isolated as white solids in 84 and 85 % yield, respectively. Assuming analogous

reactivity to 6-1; 6-2 and 6-3 were assigned the formulations of tBu3P(N2O)B(C6F5)2Mes and tBu3P(N2O)B(C6F5)2(OC6F5) (Scheme 6.4).

Scheme 6.4 – Synthesis of 6-2 and 6-3.

Examination of the 11B{1H} NMR spectra indicated the expected four-coordinate nature of the

boron centres upon generation of the B−O bonds to the N2O moiety, with sharp singlets observable at 3.56 and 6.43 ppm for both 6-2 and 6-3, respectively. Quaternization of the boron centre was further confirmed by NMR spectroscopy as evidenced by the narrowing of the 12 respective para-meta gaps for the C6F5 rings in both 6-2 and 6-3. Compound 6-3 is unique in that it bears two distinct C6F5 environments (BC6F5 and OC6F5), both of which undergo

meta-para gap narrowing upon quaternization at the boron centre. The respective ∆δm-p of the

boron−bound C6F5 groups, in 6-2 and 6-3, were found to be 4.61 and 6.44 ppm. Study of the 31 1 15 1 P{ H} NMR spectra of the N-isotopologues of both 6-2 and 6-3 revealed similar JP-N and 2 JP-N couplings to 6-1 thereby confirming the same PNNOB formulation. The binding mode was further supported by the presence of two doublets of doublets in the 15N NMR spectra of both

184

15 6-2 and 6-3, confirming the P(N2O) fragment (Figure 6.3). The relevant N NMR chemical shifts and associated coupling constants are listed in Table 6.1. It should be noted that the nitrogen atom bound to phosphorous is considerably further upfield relative to the nitrogen atom two bonds separated from the phosphorous centre.

Figure 6.3 – 31P{1H} and 15N NMR spectra for 6-3.

Crystals of both 6-2 and 6-3 were grown from either layered solutions of bromobenzene and

pentane or CH2Cl2 and pentane and were analyzed by X-ray crystallography. Severe solvent disorder prohibited the discussion of the metrical parameters for 6-2 and 6-3, however, the data was valuable in confirming atom connectivity. Collectively, these data confirmed the proposed formulations of 6-2 and 6-3 and demonstrated that in spite of markedly different Lewis acidities,

N2O activation in the presence of PtBu3 remains feasible.

185

The Lewis acidic bisborane, 1,4-(C6F5)2C6F4(BC6F5)2 was developed in the groups of Marks and Rheingold to investigate the effects of employing a dinuclear Lewis acid as a co-catalyst in the activation of bimetallic constrained geometry catalysts.119 This Lewis acid warranted further investigation as it was of interest to establish whether activation at a single B site would hamper or perhaps enhance the reactivity at the adjacent boron centre.

A 1:2 mixture of 1,4-(C6F5)2C6F4(BC6F5)2 with PtBu3 in bromobenzene, was exposed to an

atmosphere of N2O and stirred overnight. Following reaction completion, a white solid, 6-4 was isolated in 91 % yield (Scheme 6.5).

Scheme 6.5 – The synthesis of the bis-zwitterionic compound 6-4.

The 1H and 31P{1H} NMR spectra confirmed the incorporation of the phosphine moiety in the product with a doublet pertaining to the tBu groups observable by 1H NMR, in addition to a doublet of doublets by 31P{1H} NMR spectroscopy. The presence of a single sharp resonance in 11 1 the B{ H} NMR could not confirm double activation of N2O, as the neutral species is typically silent by 11B{1H} NMR spectroscopy.119 Examination of the 19F NMR spectrum revealed that the

C6F4 linkage presented as a singlet at -137.48 ppm, integrating to 4F, indicative of a two-fold

rotation axis. The signal for the C6F4 fluorine atoms was found to undergo a large upfield shift from -125.7 ppm (free bisborane) to -137.48 ppm (6-4). This notable chemical shift difference

stands in stark contrast to the ortho-fluorine of the B(C6F5)2 fragments, which are observed to undergo upfield shifts of roughly 5 ppm upon quaternization of the boron centres.119 This deviation in chemical shift is consistent with the large quantity of electron density in the C6F4 ring. Free rotation about the B-C bonds of the perfluoroaryl rings was indicated by chemical 19 equivalence of the C6F5 rings as demonstrated by three resonances in the F NMR spectrum at - 133.57, -161.980 and -166.82 ppm pertaining to the ortho-, meta- and para-F atoms,

respectively. Examination of the ∆δm-p gap supported the double quaternization of the B centres

186

with a chemical shift difference of 5.02 ppm.12 Two resonances in the 15N NMR spectrum confirmed two chemically inequivalent N atoms, both of which were found to couple to each other and the neighbouring P atom (Table 6.1). In consideration of the above-noted data, the formulation of 6-4 was proposed as the bis-zwitterionic compound tBu3P(N2O)B(C6F5)2C6F4(C6F5)2B(N2O)PtBu3.

Crystals of 6-4 were grown from a layered solution of bromobenzene and cyclohexane. A crystal structure determination was performed confirming the proposed formulation of 6-4 (Figure 6.4).

Figure 6.4 – POV-Ray depiction of the molecular structure of 6-4. B: yellow-green, C: black, N: aquamarine, F: pink, O: red, P: orange. H atoms removed for clarity. Selected bond distances (Å) and angles (°). P(1)-N(1), 1.6905(16); N(1)-N(2), 1.253(2); N(2)-O(1), 1.3198(19); O(1)- B(1), 1.536(2); P(1)-N(1)-N(2), 116.48(14); N(1)-N(2)-O(1), 109.81(15); N(2)-O(1)-B(1), 111.26(13).

The two PN2O moieties were found to be centrosymmetrically disposed on opposite sides of the bisborate species in order to minimize both steric and electrostatic interactions. The boron centre was found to be pseudo-tetrahedral as was previously determined employing spectroscopic

187 techniques (sum of angles at boron - 656°). The metrical parameters of 6-4 can readily be 349 compared to the benchmark compound tBu3P(N2O)B(C6F5)3 , whose borane bears electronic properties closely related to that of the bisborane. Examination of the pertinent metrical parameters reveals that regardless of the nature of the borane, the B−O bond lengths remain essential unchanged and are statistically indistinguishable from each other (6-4: 1.536(2) Å; 349 tBu3P(N2O)B(C6F5)3: 1.5428(18) Å ). The N−N bond lengths are statistically identical (6-4: 349 1.253(2) Å, tBu3P(N2O)B(C6F5)3: 1.2573(17) Å ) yet show considerable perturbation of the

N2O moiety as compared to free N2O. Notably however, the P−N bond length in 6-4 (1.6905(16) Å) was found to be slightly shorter than in the related zwitterion 349 tBu3P(N2O)B(C6F5)3, which bears a P−N bond length of 1.7087(12) Å .

Mechanistically, the boron centers of the bisborane likely undergo sequential activation of two molecules of N2O with PtBu3. Upon coordination of the first N2O moiety, electron density is centered on the B centre and transferred to the linking C6F4 ring. The increased electron density in the C6F4 ring can be donated to the remaining three coordinate B centre, as demonstrated via resonance. The delocalization of electron density to the second boron atom likely reduces the Lewis acidity of this site. Despite decreased Lewis acidity, activation of a second molecule of

N2O is achieved employing a subsequent equivalent of PtBu3 (Scheme 6.6).

Scheme 6.6 – The two step activation of N2O by 1,4-(C6F4)[B(C6F5)2]2 and 2PtBu3.

188

Table 6.1 – Comparison of relevant 15N chemical shifts and coupling constants.

1 1 2 δN(1) (ppm) δN(2) (ppm) JN-N (Hz) JN-P (Hz) JN-P (Hz) a a a a a tBu3P(N2O)B(C6F5)3 381.7 566.6 16 59 20 6-1 377.03 577.72 16 60 19 6-2 375.33 574.19 15 59 20 6-3 389.37 572.49 16 59 20 6-4 382.51 572.24 16 59 20

a See Otten et al.349

Comparison of the spectroscopic data for the bound N2O moiety as compared to the free molecule indicated a drastic downfield shift for both the N(1) and N(2) nitrogen atoms which were formerly observed at 135 and 218 ppm in the free species. Examination of the N−N and N−P coupling constants for these complexes revealed that the values were essentially indistinguishable irrespective of the Lewis acidity of the coordinated borane.

6.2.2 Reactions of BAr3 and PtBu3 with N2O

The results in the aforementioned section demonstrated the successful application of pentafluorophenyl-containing boranes in the activation of nitrous oxide in cooperation with

PtBu3. As these former boranes contained inductively withdrawing perfluorophenyl groups, it was of interest to probe boranes of lesser Lewis acidity to establish the exact limits of Lewis acidity required for the acidic component of the FLP pair. Due to the overwhelmingly steep cost of perfluorinated boranes and potentially hazardous synthetic methodologies, it was especially desirable to elucidate systems which are either readily synthesized from inexpensive starting materials or are commercially available.

141 156 Parallel stoichiometric mixtures of both B(p-H-C6F4)3 and B(p-F-C6H4)3 with PtBu3 were

prepared and subjected to an atmosphere of N2O in bromobenzene. These boranes were found to be good points for comparison as their respective Lewis acidities vary significantly. The use of the Gutmann-Beckett Lewis acidity experiment172,174,175,364 led to the determination of the Lewis 141 acidity of B(p-H-C6F4)3 and B(p-F-C6H4)3 as 97 and 86 % that of B(C6F5)3 (Table 6.2).

189

Table 6.2 – Lewis acidity test data determined employing the Gutmann-Beckett method.

B(C6F5)3 B(p-H-C6F4)3 B(p-F-C6H4)3 31 b b P NMR (Et3PO)BR3 (ppm) 78.1 77.3 74.3 a b b ∆δ (P(Et3PO)BR3 – PEt3PO) (ppm) 26.6 25.8 22.8 b Lewis acidity relative to B(C6F5)3 (%) - 97.0 85.7

a 31 b 141 P chemical shift of Et3PO in CD2Cl2 determined to be 51.5 ppm; See Ullrich et al.

Following reaction of the respective solutions for 12 hours, the yellow colour of the solution

containing B(p-H-C6F4)3 had faded to colourless, indicative of consumption of the FLP species,

while the yellow solution containing B(p-H-C6F4)3 remained visibly unchanged from that of the starting reaction. Precipitation was induced in both cases affording 6-5 and 6-6 in 91 and 80 % yield, respectively (Scheme 6.7).

Scheme 6.7 – The synthesis of 6-5 and 6-6.

Examination of the spectroscopic data for 6-5 and 6-6 supported the generation of an anionic boron centre with 11B{1H} NMR resonances at 0.69 and 6.69 ppm. The19F NMR resonances were located upfield for 6-5 (-134.27/-143.10 ppm) and 6-6 (-120.87 ppm) relative to those of 141,142 156 the free boranes B(p-H-C6F4)3 (-129.60/-138.74 ppm) and B(p-H-C6F4)3 (-106.00 ppm). Singlets were observable in the 31P{1H} NMR spectrum at 68.33 and 64.27 ppm for 6-5 and 6-6, respectively. The collection of these data could not unambiguously confirm the formulations of 6-5 and 6-6. Generation of the related 15N isotopologues revealed the expected doublet of

190

doublets in both the 31P{1H} and 15N NMR spectra for 6-5 (588.75 and 367.61 ppm) and 6-6 (571.01 and 381.33 ppm), respectively. 6-5 and 6-6 were therefore assigned the chemical formulas tBu3P(N2O)B(p-H-C6F4)3 and tBu3P(N2O)B(p-F-C6H4)3. The bonding in 6-5 was confirmed crystallographically (Figure 6.5).

Figure 6.5 – POV-Ray depiction of the molecular structure of 6-5. B: yellow-green, C: black, N: aquamarine, F: pink, O: red, P: orange. H atoms removed for clarity. Selected bond distances (Å) and angles (°). B(1)-O(1), 1.533(2); P(1)-N(1), 1.7092(16); O(1)-N(2), 1.327(2); N(1)-N(2), 1.255(2); N(2)-O(1)-B(1), 114.86(14); N(2)-N(1)-P(1), 115.32(13); N(1)-N(2)-O(1), 109.49(14).

The weakly Lewis acidic and commercially available borane, triphenylborane (BPh3) was combined with PtBu3 and N2O in attempts to isolate tBu3P(N2O)BPh3. BPh3 has previously been 31 documented to promote the heterolytic cleavage of H2 employing PtBu3 and it was hypothesized that this Lewis acid would be suitable in the present chemistry. The reaction

afforded tBu3P(N2O)BPh3 in poor yields, with a maximal purity of 90 %. Subsequent attempts at purification resulted in the decomposition of the product. This finding suggested that the

threshold of Lewis acidity required for N2O activation lies in the vicinity of BPh3.

191

6.2.3 Probing the Range of Lewis Basicities Tolerated in N2O Activation

Based on the data presented in the aforementioned sections, it is evident that complexes of N2O

can be generated from combinations of PtBu3 with a series of boranes of varying Lewis acidities. Having determined the range of Lewis acidities tolerated, it was of interest to probe the range of

Lewis basicities capable of generating compounds analogous to the aforementioned P/B-N2O complexes.

A series of phosphorous bases, di-tert-butylphosphine (HPtBu2), dimesitylphosphine (HPMes2)

and P(o-tol)3) were probed for their effectiveness in N2O activation based on their documented 24,25,141 application in the heterolytic cleavage of H2. Stoichiometric mixtures of these bases and

B(C6F5)3 or B(p-H-C6F4)3, under N2O, failed to generate the analogous zwitterionic species as those previously observed in sections 6.2.1 and 6.2.2. Similarly, the nitrogen bases, 2,2,6,6- tetramethylpiperidine, lutidine and N-benzylidene-tert-butylamine were tested for their effectiveness in N2O activation, due to previous application in H2 activation and reduction 45,145,365 chemistry. The nitrogen bases were combined with a stoichiometric amount of B(C6F5)3 and exposed to an atmosphere of N2O, however, did not yield N2O adducts. These findings

suggest that despite the flexibility available with respect to the nature of the Lewis acid, N2O

activation is dependent on the nature of the base. This implies that the PN2O fragment is stabilized due to σ-donation provided by the phosphine. The ability of PtBu3 to interact in N2O activation chemistry illustrates how the basic and bulky phosphine strikes the required balance of sterics and electronics.

In 1977, Tolman et al., reported the determination of the electronic parameters for a variety of ligands (L), by the generation of the corresponding Ni(0) complexes, Ni(CO)3L, from the 22 reaction of Ni(CO)4 and L. The carbonyl CO bands, in the IR spectra of the resulting

Ni(CO)3L complexes, were shifted relative to the basicity of the coordinated ligand, L. An

examination of the Tolman parameters revealed that tricyclohexylphosphine (PCy3) and PtBu3 22 are sterically quite different (Tolman cone angles: PCy3, 170° and PtBu3, 182°) but are -1 -1 22 electronically equivalent (PCy3: ν, 2056.4 cm ; PtBu3: ν, 2056.1 cm ). In light of this, PCy3 was hypothesized to be a suitable base for use in the activation of N2O with boranes. However,

previous reports have shown that less sterically demanding phosphines, such as Mes2PH,

tBu2PH, PiPr3 and PCy3, tend to undergo nucleophilic attack at the para-carbon of a C6F5 ring of

192

25 B(C6F5)3, yielding the C6F4-linked phosphonium-fluoroborate zwitterion (Scheme 6.8). Due to this complication, the reaction of PCy3, B(p-H-C6F4)3 and N2O was conducted where substitution

of the para-F of B(C6F5)3 for H, precludes para-attack by the PCy3 moiety.

Scheme 6.8 – Nucleophilic attack by PCy3 at the para-C of a C6F5 ring of B(C6F5)3.

The stoichiometric mixture of PCy3 and B(p-H-C6F4)3 in bromobenzene was subjected to an 350 atmosphere of N2O, yielding the zwitterionic species, Cy3P(N2O)B(p-H-C6F4)3 , in 59 % yield (Scheme 6.9). 1H and 11B{1H} spectroscopic data was consistent with that of 6-5. The 31P{1H} NMR chemical shift was observed at 56.4 ppm, considerably downfield relative to the free phosphine. One bond and two bond N−N and N−P couplings were consistent with the capture of

the intact N2O fragment. It should be noted that the zwitterionic product was isolated in

relatively low yield as compared to the other N2O activation products, with PtBu3 as the base.

Reaction of PCy3 in bromobenzene, under an atmosphere of N2O, resulted in excess of 90 %

conversion to the phosphine oxide (Cy3PO); whereas under parallel conditions employing PtBu3, conversion was only 50 % over the same time period. Given these data, it can be concluded that the oxidation of PCy3 is in competition with the capture of N2O by the frustrated Lewis pair.

350 Scheme 6.9 – Synthesis of Cy3P(N2O)B(p-H-C6F4)3 and 6-7.

193

In consideration of this result, it was of interest to establish whether N2O could similarly be

captured in the presence of the FLP resulting from the reaction of Cy3P and B(p-F-C6H4)3. Despite bearing a para-F atom, previous reactivity studies revealed that no interaction existed between the phosphine and borane, even upon heating at 150 °C for five days. A stoichiometric mixture of PCy3 and B(p-F-C6H4)3 was exposed to an atmosphere of N2O and left stirring for 12 hours. A white solid, 6-7, was isolated from the yellow solution in 78 % yield. Examination of the 11B{1H} NMR spectrum revealed the presence of a single peak at 26.39 ppm, considerably downfield as compared to 6-6. Similarly, a singlet was observed in both the 19F NMR and 31P{1H} NMR spectra at -116.52 ppm and 62.58 ppm, respectively. Repetition of the experiment 15 employing enriched N2O revealed no evidence of P−N and N−N coupling as was previously observed for the aforementioned complexes. In conjunction with elemental analysis, the nature

of 6-7 was formulated to be the phosphine oxide adduct of the borane, (Cy3PO)B(p-F-C6H4)3) (Scheme 6.9). A crystal structure analysis was performed unambiguously confirming this formulation (Figure. 6.6).

194

Figure 6.6 – POV-Ray depictions of the molecular structures of Cy3P(N2O)B(p-H-C6F4)3 and 6-7. B: yellow-green, C: black, N: aquamarine, F: pink, O: red, P: orange. H atoms removed for 350 clarity. Selected bond distances (Å) and angles (°). Cy3P(N2O)B(p-H-C6F4)3 : P(1)-N(1), 1.7146(15); N(1)-N(2), 1.260(2); N(2)-O(1), 1.3351(18); B(1)-O(2), 1.551(1); P(1)-N(1)-N(1), 111.10(12); N(1)-N(1)-O(1), 111.13(14); N(2)-O(1)-B(1), 110.32(13). 6-7: P(1)-O(1), 1.5251(8); O(1)-B(1), 1.5854(15); P(1)-O(1)-B(1), 147.26(8).

It was hypothesized that Cy3P(N2O)(p-F-C6H4)3 might be generated as an intermediate en route to the formation of 6-7. Therefore, the mechanism for the generation of 6-7 was investigated 15 31 1 15 employing N2O and followed spectroscopically by both P{ H} and N NMR. The reaction

was followed for 12 hours and at no time was the species Cy3P(N2O)B(p-F-C6H4)3 detected in solution. 6-7 was therefore determined to be the product of the slow oxidation of PCy3 by N2O, yielding the phosphine oxide, followed by coordination of the Lewis base to the Lewis acid,

B(p-F-C6H4)3.

6.2.4 Mechanism of N2O Activation by FLPs

The reaction between organic azides and phosphines, yielding synthetically valuable aza-ylides, was first elucidated by Staudinger in 1919.366-368 At this time, Staudinger proposed that the phosphazides were transient species generated in situ en route to aza-ylides. These phosphazides

were thought to be highly reactive and unstable toward N2 loss. It was only in the latter two

195

decades of the 20th century that strategies were developed for the isolation of phosphazide adducts, by means of Lewis acid stabilization.363

The Staudinger reaction was simultaneously computational investigated by Rzepa369 and Grutzmacher370 in 1999, 80 years after the primary discovery. These studies proposed initial nucleophilic attack by the phosphine at the terminal nitrogen atom of the azide. The cis phosphazide was determined to be the sole product of reaction and was found to be 6-9 kcal/mol more stable than the corresponding trans isomer. This counterintuitive geometry was rationalized by the electrostatic interaction between the positively charged phosphorous centre 363 and the opposing nitrogen atom. Cyclization, followed by formal elimination of N2 was found to yield the stable aza-ylide (Scheme 6.10).

Scheme 6.10 – Computationally determined mechanism for the Staudinger synthesis of aza- ylides.363

A similar mechanism is suspected to be at play with respect to the activation of N2O by phosphorous and boron-based frustrated Lewis pairs. The zwitterionic compounds 6-1 to 6-7 could be classified as intermediates in Staudinger-type oxidations where they are kinetically stabilized and trapped by a Lewis acid. In fact, a report in 1921 by Staudinger proposes the interaction of triethylphosphine with N2O to yield the species Et3P=N−N=O prior to 371 decomposition to the triethylphosphine oxide and N2. This would imply that in the presence

of an FLP and N2O, The P=N−N=O species is captured in an irreversible fashion by the Lewis acid (Scheme 6.11).

196

Scheme 6.11 – Proposed mechanism for the activation of N2O by B/P FLPs.

6.2.5 Reactivity of tBu3P(N2O)BR3

333 As oxidation of organic substrates with N2O is exceedingly difficult, it was of interest to

establish the general reactivity of the species tBu3P(N2O)BR3, which contains an “activated” form of N2O, and to determine whether the species might be successful in enabling O-atom

transfer to organic species while liberating N2.

Initial studies focused on the thermal stability of the tBu3P(N2O)BR3 zwitterions. NMR samples

of the species 6-1 to 6-6 were prepared in C6D5Br. The respective samples were heated to 125 °C for a period of 60 hours. 15N NMR spectroscopy indicated loss of the P−N and N−N

couplings, indicative of decomposition of the tBu3P(N2O)BR3 moieties. Thermally induced

oxidation of the phosphine oxide was suspected, yielding the corresponding (tBu3PO)BR3 adducts. Independent synthesis of the (tBu3PO)BR3 adducts confirmed this assignment. DFT calculations at the B3LYP/6-31G(d) level of theory demonstrated that the species 349 (tBu3PO)B(C6F5)3 was in fact 60.4 kcal/mol more favorable than the tBu3P(N2O)B(C6F5)3. This is consistent with the experimental results observed upon thermolysis. It is proposed that the trans zwitterion undergoes isomerization to the cis orientation placing the O-atom in close proximity to the P-atom allowing for facile transfer of the OBR3 fragment to the phosphine,

resulting in the generation of the Lewis acid-base adducts, (tBu3PO)BR3 (Scheme 6.12). Notably, the analogous products are accessible via photolysis in five minutes; however, prolonged exposure ultimately results in decomposition to unidentifiable products.

Scheme 6.12 – Mechanism for decomposition of tBu3P(N2O)BR’3 to (tBu3PO)BR’3.

197

Reactions of tBu3P(N2O)BR3 with diphenylacetylene and dimethylacetylene were attempted in attempts to promote O-atom transfer to yield the cyclic compounds, 2,3-diphenyloxirene and 2,3- dimethyloxirene. However, even upon heating for two days, no reaction was observed.

6-6 was investigated for its ability to act as a ligand (chelating species via the N(1) and O atoms) in the generation of novel transition metal-containing complexes. 6-6 was reacted with stoichiometric amounts of Ti(OiPr)4, TiCl4 or Ni(COD)2 but was found to result in series of 31 1 products by P{ H} NMR which were not separable from each other. Reaction with Me3SiBr,

Me3SiI, BF3·OEt2 and trifluoromethanesulfonic acid resulted in either no reaction with the

zwitterionic species or in complete decomposition of the phosphine (HOSO2CF3).

Although never successfully crystallographically characterized, it was suspected that due to the

weak Lewis acidity of B(p-F-C6H4)3, relative to B(C6F5)3, that the zwitterion 6-6 possessed the weakest B-O of the species 6-1 to 6-6 and would perhaps be susceptible to Lewis acid exchange reactions. The trityl cation, a strong carbon-based Lewis acid was employed for exchange as it is incapable of generating an FLP with PtBu3 or PCy3 due to its propensity to undergo nucleophilic aromatic substitution in the presence of either phosphine (Scheme 6.13).21 This results in the

inability of the trityl cation to be used in the direct activation of N2O.

Scheme 6.13 – Reaction of tritylborate with PCy3 and PtBu3.

A stoichiometric reaction of 6-6 and [Ph3C][B(C6F5)4] was prepared in CD2Cl2. The colour of the reaction was noted to transition from deep orange to bright-yellow over 30 seconds. Examination of the 19F NMR spectrum revealed four resonances, assignable to free

B(p-F-C6H4)3 and the o-, p- and m-F signals of the B(C6F5)4 ion (Figure 6.7).

198

19 Figure 6.7 – F NMR spectrum of the reaction of 6-6 with tritylborate in CD2Cl2.

Typically, the 19F NMR resonance for the p-F in 6-6 is found at -120.87 ppm, however, the signal observed upon reaction with tritylborate was at -106 ppm, indicative of release of 156 B(p-F-C6H4)3 (literature value (C6D6): -109 ppm ), from 6-6. The reaction was repeated on a larger scale and allowed to stir at room temperature overnight from which a yellow solid, 6-8, was recovered in 96 % yield. 1H NMR analysis revealed signals in the aromatic region of the

spectrum assignable to the CPh3 fragment. Similarly, a sharp singlet was observable in the 11 1 B{ H} NMR spectrum at -16.68 ppm confirming the presence of the B(C6F5)4 anion. A sharp singlet was also noted in the 31P{1H} NMR spectrum at 77.31 ppm, considerably downfield from the parent species 6-6 (64.27 ppm). NMR analysis of the 15N isotopologue of 6-6 revealed that

the tBu3PN2O fragment was retained throughout the course of the reaction as evidenced by the 31 1 15 1 1 continued presence of a doublet of doublets in the P{ H} and N NMR spectra with JN-N, JN-P 2 and JN-P coupling constants of 15, 58 and 20 Hz, respectively. These coupling constants were consistent with those observed for 6-1 to 6-6. It is worth noting that the chemical shift for the N atom bound to P is considerably downfield (Figure 6.9) at 412.30 ppm as compared to the average chemical shift value for the analogous N atom in compounds 6-1 to 6-6 (average chemical shift - 378.86 ppm). It is important to note that the above reaction was not performed in a sealed J. Young bomb but was conducted in a simple scintillation vial where liberation of

199

gaseous products could readily be achieved. Collectively these data suggest the formulation of

6-8 to be of the salt, [tBu3P(N2O)CPh3][B(C6F5)4] (Scheme 6.14).

Scheme 6.14 – Exchange reaction of 6-6 with [Ph3C][B(C6F5)4] to yield 6-8.

Crystals of 6-8 were grown from a layered solution of CH2Cl2 and cyclohexane at room temperature and were analyzed via X-ray crystallography, thereby unambiguously confirming the formulation of 6-8 (Figure 6.8).

Figure 6-8. POV-Ray depiction of the molecular structure of 6-8. B: yellow-green, C: black, N: aquamarine, F: pink, O: red, P: orange. H atoms removed for clarity. Selected bond distances (Å) and angles (°). P(1)-N(2), 1.7217(13); N(1)-O(1), 1.3752(15); O(1)-C(1), 1.4955; N(1)-N(2), 1.2418(17); P(1)-N(2)-N(1), 115.98(10); N(2)-N(1)-O(1), 108.52(12); N(1)-O(1)-C(1), 110.68(10).

200

Examination of the metrical parameters revealed a lengthening of the P−N and N−O bonds in and the shortening of the N−N bond relative to the related zwitterions, tBu3(N2O)BR3’. The O−C bond was noted to be 1.4955(17) Å, shorter than the corresponding B−O bonds of the aforementioned compounds in keeping with the increased Lewis acidity of the trityl cation.

1 19 31 1 15 Figure 6.9 – H, F, P{ H} and N NMR spectra in CD2Cl2 for 6-8.

It is noted that the tBu3P(N2O) fragment is inherently stable and remains intact throughout the exchange process. Notably, this method of exchange chemistry can be employed for the

synthesis of compounds 6-1 to 6-5 via exchange of 6-6 with a single equivalent of: PhB(C6F5)2,

MesB(C6F5)2, (C6F5)2BOC6F5, B(p-H-C6F4) or half an equivalent of 1,4-(C6F4)[B(C6F5)2]2.

201

Exchange with other main group Lewis acids such as Me3Sn(OSO2CF3), Ph3Si(OSO2CF3) and

Me3Si(OSO2CF3) either resulted in no reaction (Me3Sn(OSO2CF3)), incomplete conversion to

the exchange product [tBu3P(N2O)SiPh3][ (OSO2CF3)] (Ph3Si(OSO2CF3)) or complete

decomposition of 6-6 (Me3Si(OSO2CF3)). As conversion to [tBu3P(N2O)SiPh3][(OSO2CF3)] could never fully be schieved, the product remained unisolable due to parallel solubilities with the starting materials.

6.2.6 Mechanism of Lewis Acid exchange

The mechanism of exchange between 6-6 and Lewis acids of greater acidity warranted further investigation. Both dissociative and associative mechanisms were potentially at play during the process of exchange. According to a dissociative pathway, the weaker Lewis acid B(p-F-C6H4)3 is ejected, releasing the tBu3P(N2O) ion into solution where it is scavenged by a stronger Lewis acid. An associative mechanism involves an intermediate where the O−B bond to the weak

borane B(p-F-C6H4)3 is weakened and lengthened as the stronger Lewis acid approaches the O centre. The transition state would involve the coordination of both boranes to the oxygen of the

tBu3P(N2O) fragment, prior to the departure of B(p-F-C6H4)3 and formal coordination of the new Lewis acid. These mechanisms were further probed via 2D 19F EXSY NMR.

In order to gain kinetic insight into the mechanism of exchange of 6-6 with a stronger Lewis acid, stoichiometric reactions of 6-6 and B(C6F5)3, [Ph3C][B(C6F5)4] and PhB(C6F5)2 were followed by 19F exchange spectroscopy (EXSY). In each instance, the solid starting materials were combined in a J. Young NMR tube and the solvent was vacuum transferred into the vessel. The samples were thawed and immediately inserted into a pre-locked, tuned and shimmed NMR instrument where 19F EXSY spectra were recorded at room temperature. Unfortunately, the first scans indicated that full conversion to the products of exchange had occurred. In attempts to slow and observe the exchange reactions, the analogous reactions were repeated at -80 °C and in all instances complete conversion to the products of exchange had occurred prior to data collection. This indicates that the O−B(p-F-C6H4)3 bond of 6-6 is extremely weak in nature and that the subsequent exchange reactions are too fast to be observed.

However, the stoichiometric reaction of 6-6 with B(p-F-C6H4)3, in a self-exchange reaction,

resulted in the observation of site exchange between the tBu3P(N2O)B(p-F-C6H4)3 moiety with 19 the free borane B(p-F-C6H4)3 in CD2Cl2 as evidenced by 2D F EXSY NMR (Figure 6.10).

202

Figure 6.10 – 2D 19F-19F EXSY NMR spectrum.

Spectroscopic data was acquired for the process of exchange between -25 and +25 °C, from which the enthalpy (ΔH‡ = 71.2(9) kJ·mol-1) and entropy (ΔS‡ = +32(3) J·mol-1·K-1) of activation were determined. The small entropic value indicates that there is some degree of bond weakening of the O−B(p-F-C6H4)3 bond N2O prior to the binding of the incoming Lewis acid (Scheme 6.15). This data is more consistent with a dissociative interchange mechanism.

Scheme 6.15 – Proposed mechanism of Lewis acid (LA) exchange employing 6-6.

203

6.3 Conclusions

Combinations of sterically encumbered electrophilic boranes and tertiary phosphines (PR3: R = tBu and Cy) collaboratively activate nitrous oxide yielding zwitterionic compounds that are 349 stable toward N2O liberation. These species can be thermally or photolytically decomposed to the phosphine oxide adducts of the associated borane. Similarly, the exchange chemistry of

tBu3P(N2O)B(p-F-C6H4)3 can be exploited for the generation of otherwise inaccessible main group compounds via exchange of the weakly bound acidic fragment for a Lewis acid of superior strength.350

6.4 Experimental Section

6.4.1 General Considerations

All manipulations were carried out under an atmosphere of dry, O2-free N2 employing standard

Schlenk-line and glovebox techniques. Solvents (pentane, hexanes, CH2Cl2) were dried by employing a Grubbs-type column system (Innovative Technology), degassed and stored under

N2. Cyclohexane was distilled under N2 from Na/benzophenone and bromobenzene was distilled

under N2 from CaH2. CD2Cl2 was vacuum transferred from CaH2 and d8-THF was vacuum transferred from Na/benzophenone. Both solvents were subsequently degassed and stored under

N2. PtBu3 (Strem Chemicals), PCy3 (Strem Chemicals), [Ph3C][B(C6F5)4] (Strem Chemicals), 15 BPh3 (Aldrich), N2O (Sigma-Aldrich; 99%) and N2O (Cambridge Isotope Laboratories; 99.9%, 15 126 180 98.8% N enriched) were used as received. The reagents PhB(C6F5)2, MesB(C6F5)2, 154 141 156 119 (C6F5)2BOC6F5, B(C6F4-p-H)3, B(C6H4-p-F)3 and 1,4-(C6F5)2BC6F4B(C6F5)2 were prepared according to literature procedures. 1H, 11B, 13C, 15N, 19F and 31P NMR spectra were recorded at 25 °C on a Varian NMR System 400 MHz or Bruker Avance III 400 MHz 1 spectrometer, and were referenced using (residual) solvent resonances relative to SiMe4 ( H, 13 11 19 31 15 C), or to an external standard ( B: (Et2O)BF3; F: CFCl3; P: 85% H3PO4; N: NH3(l) via the 15 372 N resonance of 90% formamide in DMSO-d6 at 112 ppm). Chemical shifts are reported in ppm and coupling constants as scalar values in Hz. Combustion analyses were performed in house employing a Perkin-Elmer CHN Analyzer.

204

6.4.2 Syntheses

Synthesis of 6-1. A 50 mL Schlenk tube was charged with PhB(C6F5)2

(0.119 g, 0.282 mmol) and PtBu3 (0.057 g, 0.282 mmol) in bromobenzene (5 mL). The colourless solution was degassed and

exposed toN2O. The solution was stirred at room temperature for 12 hours. At this time, the solution was clear and colorless. Pentane (10 mL) was added precipitating a white solid. The solid was isolated by filtration, washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours.

Crystals suitable for X-ray diffraction were grown from a layered solution of CH2Cl2 and 1 3 cyclohexane at 25 °C. Yield: 0.144 g (76%). H NMR (CD2Cl2): δ 7.40 (d, 2H, JH-H = 7 Hz, o- 3 3 3 C6H5); 7.17 (t, 2H, JH-H = 8 Hz, m-C6H5); 7.09 (t, 1H, JH-H = 8 Hz, p-C6H5); 1.44 (d, 27H, JH-P 11 1 13 1 1 = 14Hz, PtBu). B{ H} NMR (CD2Cl2): δ 3.27 (s). C{ H} NMR (CD2Cl2): δ 148.25 (br d, JC- 1 1 F = 240 Hz, o-C6F5); 139.68 (br d, JC-F = 218 Hz, p-C6F5); 137.23 (br d, JC-F = 226 Hz, m-C6F5); 1 132.31 (s, p-C6H5); 127.35 (s, o-C6H5); 125.77 (s, m-C6H5); 41.64 (d, JC-P = 30 Hz, PC{CH3}3); 15 2 1 29.71 (s, PC{CH3}3). N NMR (CD2Cl2): δ 577.72 (dd, JN-P = 19.1 Hz, JN-N = 15.9 Hz, 1 1 19 PNNO); 377.03 (dd, JN-P = 59.5 Hz, JN-N = 16.0 Hz, PNNO). F NMR (CD2Cl2): δ -131.87 3 4 3 (dd, 4F, JF-F = 25 Hz, JFF = 9 Hz, o-C6F5); -161.71 (t, 2F, JF-F = 20 Hz, m-C6F5); -166.45 (td, 3 4 31 1 1 2 4F, JF-F = 23 Hz, JFF = 8 Hz m-C6F5). P{ H} NMR (CD2Cl2): δ 67.20 (dd, JP-N = 59.0 Hz, JP-

N = 19.4 Hz). Anal. Calcd for C30H32BF10N2OP: C, 53.83; H, 4.97; N, 4.18. Found: C, 54.06; H, 4.94; N, 4.27 %.

Synthesis of 6-2. Compounds 6-2 and 6-3 were prepared in a similar fashion and thus only one preparation is detailed. A 50 mL

Schlenk tube was charged with MesB(C6F5)2 (0.103 g, 0.222 mmol)

and PtBu3 (0.045 g, 0.222 mmol) in bromobenzene (5 mL). The reaction was degassed and

backfilled with N2O. The solution was stirred for 12 hours at r.t. At this time, the solution was clear and colorless. Pentane (10 mL) was added precipitating a white solid. The solid was isolated by filtration, washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Crystals, although not suitable for X-ray diffraction, were grown from a layered solution of CH2Cl2 and 1 cyclohexane at 25 °C. Yield: 0.133 g (84 %). H NMR (CD2Cl2): δ 6.56 (s, 2H, (CH3)3C6H2); 3 11 1 2.18 (s, 3H, p-CH3); (s, 6H, o-CH3); 1.43 (d, 27H, JH-P = 14 Hz, PtBu). B{ H} NMR 13 1 1 (CD2Cl2): δ 3.56 (s). C{ H} NMR (CD2Cl2) partial: δ 148.42 (br d, JC-F = 235 Hz, o-C6F5); 1 1 141.69 (s, o-(CH3)3C6H2); 139.59 (br d, JC-F = 227 Hz, p-C6F5); 137.19 (br d, JC-F = 227 Hz, m-

205

1 C6F5); 134.85 (s, p-(CH3)3C6H2); 129.78 (s, m-(CH3)3C6H2); 41.59 (d, JC-P = 30 Hz, PC{CH3}3); 15 29.72 (s, PC{CH3}3); 27.49 (s, o-(CH3)3C6H2); 24.85 (s, p-(CH3)3C6H2). N NMR (CD2Cl2): δ 2 1 1 1 574.19 (dd, JN-P = 20 Hz, JN-N=15 Hz, PNNO); 375.33 (dd, JN-P = 59 Hz, JN-N = 15, PNNO). 19 3 4 3 F NMR (CD2Cl2): δ -132.70 (dd, 4F, JF-F = 23 Hz, JF-F = 7 Hz, o-C6F5); -162.10 (t, 2F, JF-F = 3 4 31 1 20 Hz, m-C6F5); -166.71 (td, 4F, JF-F = 24 Hz, JF-F = 8 Hz, p-C6F5). P{ H} NMR (CD2Cl2): δ 1 2 67.07 (dd, JP-N = 59 Hz, JP-N = 20 Hz). Anal. Calcd. for C33H38BF10N2OP: C, 55.79; H, 5.39; N, 3.94. Found: C, 55.88; H, 5.75; N, 3.65 %.

Synthesis of 6-3. MesB(C6F5)2 (0.103 g, 0.222 mmol) and PtBu3 (0.045 g, 0.222 mmol). Yield: 0.121 g (85 %). 1H NMR 3 11 1 (CD2Cl2): δ 1.50 (d, JH-P = 14 Hz, PtBu). B{ H} NMR 13 1 1 (CD2Cl2): δ 6.43 (s). C{ H} NMR (CD2Cl2): δ 148.77 (br d, JC-F = 255 Hz, o-C6F5); 142.25 1 1 1 (br d, JC-F = 244 Hz, o-OC6F5); 140.40 (br d, JC-F = 250 Hz, p-C6F5); 138.23 (br d, JC-F = 239 1 1 Hz, p-OC6F5); 137.50 (br d, JC-F = 255 Hz, m-C6F5); 135.62 (br d, JC-F = 244 Hz, m-OC6F5); 1 119.38 (br s, ipso-C6F5); 115.23 (br s, ipso-OC6F5); 41.93 (d, JC-P = 29 Hz, PC{CH3}3); 29.75 15 2 1 (s, PC{CH3}3). N NMR (CD2Cl2): δ 572.49 (dd, JN-P = 20 Hz, JN-N = 16 Hz, PNNO); 389.37 1 1 19 3 (dd, JN-P = 59 Hz, JN-N = 16 Hz, PNNO). F NMR (CD2Cl2): δ -134.05 (d, 4F, JF-F = 16 Hz, o- 3 3 C6F5); -157.72 (d, 2F, JF-F = 19 Hz, o-OC6F5); -159.03 (t, 2F, JF-F = 20 Hz, p-C6F5); -165.47 (td, 3 4 3 4F, JF-F = 23 Hz, JF-F = 9 Hz, m-C6F5); -166.96 (t, 1F, JF-F = 20 Hz, p-OC6F5); -170.29 (tt, 2F, 3 4 31 1 1 2 JF-F = 22 Hz, JF-F = 6 Hz, m-OC6F5). P{ H} NMR (CD2Cl2): δ 68.98 (dd, JP-N = 59 Hz, JP-N =

20 Hz). Anal. Calcd. for C30H27BF15N2O2P: C, 46.54; H, 3.51; N, 3.62. Found: C, 46.63; H, 3.47; N, 3.47 %.

Synthesis of 6-4. A 50 mL Schlenk tube was charged

with 1,4-(C6F4)[B(C6F5)2]2 (0.100 g, 0.119 mmol) and

PtBu3 (0.048 g, 0.237 mmol) in bromobenzene (5 mL). The pale yellow slurry was degassed and backfilled

with N2O. The solution was stirred under an

atmosphere of N2O for 12 hours at r.t. At this time, the solution was opaque. Pentane (10 mL) was added precipitating a white solid. The solid was isolated by filtration, washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Crystals suitable for X-ray diffraction were grown 1 from a layered solution of C6H5Br and cyclohexane at 25 °C. Yield: 0.144 g (91 %). H NMR

206

3 11 1 13 1 (CD2Cl2): δ 1.46 (d, 27H, JH-P = 14 Hz, PtBu). B{ H} NMR (CD2Cl2): δ 0.67 (s). C{ H} 1 1 NMR (C4D8O): δ 149.22 (br d, JC-F = 238 Hz, C6F4); 148.94 (br d, JC-F = 240 Hz, o-C6F5); 1 1 140.03 (br d, JC-F = 233 Hz, p-C6F5); 137.58 (br d, JC-F = 229 Hz, m-C6F5); 128.01 (br s, ipso- 1 15 C6F4); 123.76 (br s, ipso-C6F5); 41.87 (d, JC-P = 29 Hz, PC{CH3}3); 29.51 (s, PC{CH3}3). N 2 1 1 NMR (CD2Cl2): δ 571.04 (dd, JN-P = 20 Hz, JN-N = 15 Hz, PNNO); 381.31 (dd, JN-P = 59 Hz, 1 19 3 JN-N = 15 Hz, PNNO). F NMR (CD2Cl2): δ -133.57 (d, 8F, JF-F = 17 Hz, o-C6F5); -137.48 (s, 3 3 31 1 4F, C6F4); -161.80 (t, 4F, JF-F = 21 Hz, p-C6F5); -166.82 (t, 8F, JF-F = 18 Hz, m-C6F5). P{ H} 1 2 NMR (CD2Cl2): δ 68.37 (dd, JP-N = 59 Hz, JP-N = 20 Hz). Anal. Calcd. for C54H54B2F24N4O2P2: C, 48.71; H, 4.09; N, 4.21. Found: C, 48.50; H, 4.20; N, 3.80 %.

Synthesis of 6-5. A 50 mL Schlenk tube was charged with

B(p-H-C6F4)3 (0.130 g, 0.284 mmol) and PtBu3 in C6H5Br (5

mL). The yellow solution was degassed and backfilled with N2O.

The reaction was stirred under an atmosphere of N2O for 12 hours at r.t. At this time, the solution was clear and colourless. Pentane (10 mL) was added precipitating a white solid. The product was collected by filtration, washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Crystals suitable for X-ray diffraction were grown from a 1 layered solution of C6H5Br and cyclohexane at 25 °C. Yield: 0.181 g (91 %). H NMR 3 11 1 (CD2Cl2): δ 6.86 (m, 3H, C6F4H), 1.44 (d, 27H, JH-P = 14 Hz, PtBu). B{ H} NMR (CD2Cl2): δ 13 1 1 0.69 (s). C{ H} NMR (CD2Cl2) partial: δ 149.58 (br d, JC-F = 238 Hz, m-C6F4H); 144.91 (br 1 2 3 d, JC-F = 240 Hz, o-C6F4H); 103.61 (t, JC-F = 23 Hz, p-C6F4H); 41.74 (d, JH-P = 29, 15 2 1 PC{CH3}3}); 29.69 (s, PC{CH3}3}). N NMR (CD2Cl2): δ 588.75 (dd, JN-P = 20 Hz, JN-N = 16 1 1 19 Hz, PNNO); 367.61 (dd, JN-P = 59 Hz, JN-N = 16 Hz, PNNO). F NMR (CD2Cl2): δ -134.27 31 1 1 (m, 6F, o-C6F4H); -143.10 (m, 6F, m-C6F4H). P{ H} NMR (CD2Cl2): δ 68.33 (dd, JP-N = 59 2 Hz, JP-N = 20 Hz). Anal. Calcd. for C30H30BF12N2OP: C, 51.16; H, 4.29; N, 3.98. Found: C, 51.24; H, 4.59; N, 4.02 %.

Synthesis of 6-6. A 50 mL Schlenk tube was charged with

B(p-F-C6H4)3 (0.212 g, 0.716 mmol) and PtBu3 (0.145 g, 0.717 mmol) in bromobenzene (10 mL). The pale yellow solution was

degassed and stirred under an atmosphere of N2O for 12 hours at r.t. At this time, the solution was cloudy and pale yellow. Pentane (10 mL) was added precipitating a white solid. The solid

207

was isolated by filtration, washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Yield: 1 0.312 g (80 %). H NMR (CD2Cl2): δ 7.33 (m, 6H, o-C6H4F); 6.87 (m, 6H, m-C6H4F); 1.38 (d, 3 11 1 13 1 27H, JH-P = 14 Hz, PtBu). B{ H} NMR (CD2Cl2): δ 6.69 (s). C{ H} NMR (CD2Cl2) 1 partial: δ 161.25 (br d, JC-F = 247 Hz, p-C6H4F); 135.54 (br s, m-C6H4F); 113.50 (br s, o- 1 15 C6H4F); 41.31 (d, JC-P = 31 Hz, PC{CH3}3); 29.85 (s, PC{CH3}3}. N NMR (CD2Cl2): δ 2 1 1 1 588.75 (dd, JN-P = 19 Hz, JN-N = 18 Hz, PNNO); 367.61 (dd, JN-P = 61 Hz, JN-N = 18 Hz, 19 31 1 1 PNNO). F NMR (CD2Cl2): δ -120.87 (s). P{ H} NMR (CD2Cl2): δ 64.27 (dd, JP-N = 61 Hz, 2 JP-N = 19 Hz). Anal. Calcd. for C30H39BF3N2OP: C, 66.43; H, 7.25; N, 5.16. Found: C, 66.25; H, 7.27; N, 5.17 %.

Synthesis of 6-7. A 50 mL Schlenk tube was charged with

B(p-F-C6H4)3 (0.205 g, 0.692 mmol) and PCy3 (0.194 g, 0.692 mmol)

in C6H5Br (5 mL). The pale yellow solution was degassed and stirred

under an atmosphere of N2O for 12 hours at r.t. At this time, the solution was faintly yellow in color. Pentane (10 mL) was added precipitating a microcrystalline solid. The solid was isolated by filtration, washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Crystals suitable

for X-ray diffraction were grown from a layered solution of CH2Cl2 and pentane at -35 °C. 1 Yield: 0.320 g (78 %). H NMR (CD2Cl2): δ 7.38 (m, 6H, o-C6H4F), 6.97 (m, 6H, m-C6H4F), 11 1 13 1 1.90-1.16 (m, 30H, Cy). B{ H} NMR (CD2Cl2): δ 26.39 (s). C{ H} NMR (CD2Cl2) partial: 1 3 δ 163.64 (d, JC-F = 246 Hz, p-C6H4F); 146.06 (br s, ipso-C6F4H); 138.16 (d, JC-F = 7 Hz, o- 2 1 2 C6H4F); 114.34 (d, JC-F = 20 Hz, m-C6H4F); 35.91 (d, JC-P = 59.77 Hz, PCy3); 27.31 (d, JC-P = 3 4 19 12 Hz, o-C6H10); 26.78 (d, JC-P = 3 Hz, m-C6H10); 26.46 (d, JC-P = 1 Hz, p-C6H10). F NMR 31 1 (CD2Cl2): δ -116.52 (s). P{ H} NMR (CD2Cl2): δ 62.58 (s). Anal. Calcd. for

C36H44BF3PCl2O: C, 65.27; H, 6.69. Found: C, 65.25; H, 7.12 %.

Synthesis of 6-8. A 20 mL scintillation vial was charged with

6-6 (0.081 g, 0.149 mmol) in C6H5Br (5 mL). A separate vial

was charged with [Ph3C][B(C6F5)4] (0.136 g, 0.149 mmol) in

C6H5Br (5 mL). The latter solution was slowly added to the first. The reaction was noted to change from a deep orange to a bright yellow within the first 30 seconds. The reaction was allowed to stir for a period of 12 hours at room temperature. Pentane was added (15 mL) precipitating a yellow solid. The product was isolated by filtration and was

208

subsequently washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Crystals suitable

for X-ray diffraction were grown from a layered solution of CH2Cl2 and cyclohexane at 25 °C. 1 3 Yield: 0.165 g (96 %). H NMR (CD2Cl2): δ 7.37 (m, 15H, C6H5), 1.41 (d, JH-P = 15 Hz, PtBu). 11 13 1 1 B NMR (CD2Cl2): δ -16.68 (s). C{ H} NMR (CD2Cl2): δ 148.74 (br d, JC-F = 235 Hz, o- 1 1 C6F5); 142.73 (s, ipso-C6H5); 138.81 (br d, JC-F = 242 Hz, p-C6F5); 136.90 (br d, JC-F = 241 Hz,

m-C6F5); 128.45 (s, ipso-C6F5); 129.18 (s, p-C6H5); 129.09 (s, m-C6H5); 129.04 (s, o-C6H5); 3 15 98.30 (s, C(C6H5)3); 42.69 (d, JH-P = 24 Hz, PC{CH3}3), 29.63 (s, PC{CH3}3). N NMR 2 1 1 1 (CD2Cl2): δ 548.42 (dd, JN-P = 20 Hz, JN-N = 15 Hz, PNNO); 412.30 (dd, JN-P = 58 Hz, JN-N = 19 3 15 Hz, PNNO). F NMR (CD2Cl2): δ -133.04 (m, 8F, o-C6F5); -163.71 (t, 4F, JF-F = 21, p- 3 31 1 1 C6F5); -167.52 (m, 8F, JF-F = 18, m-C6F5). P{ H} NMR (CD2Cl2): δ 77.31 (dd, JP-N = 58 Hz, 2 JP-N = 20 Hz). Anal. Calcd. for C55H42BF20N2OP: C, 56.52; H, 3.62; N, 2.40. Found: C, 56.40; H, 4.02; N, 2.29 %.

6.4.3 2D 19F-19F EXSY NMR Experimental Details

EXSY spectra were acquired on a Bruker AVANCE III spectrometer operating at 376.7 MHz (19F) in phase-sensitive mode, using the standard Bruker pulse sequence (noesyph). In the indirectly detected dimension 64 complex points were collected with 8 scans and 1024 points per increment. EXSY spectra were recorded with appropriate mixing times at each temperature.

Sample temperatures were calibrated with a 4% CH3OH in CD3OD sample using the standard method implemented in Bruker Topspin 2.1. Integration of diagonal- and cross-peak volumes of the 19F resonances was performed using the Gaussian fit integration method implemented in 373 374 Sparky. Using Mathematica 5.0 , cross-peak volumes of all spectra were normalized (Ix/Id) and plotted against mixing time. The data points were all at once fitted against equation (1)375 by non-linear regression.

‡ ‡ −2kτ ΔH −TΔS I 1− e mix k T − x = k = B e RT −2kτ mix I d 1+ e h

The error in determining the activation parameters from the EXSY data is related to the error in the normalized peak volumes (0 ≤ Ix/Id ≤ 1 ) introduced in the integration routine. An estimate of

the errors in the activation parameters was obtained by generating peak volumes (Ix/Id + R), in which R was randomly chosen from a normal distribution with mean μ = 0 and standard

209

deviation σ = 0.025 (corresponding to ca. 5% of the average Ix/Id). Non-linear regression

(equation (1)) was performed on the simulated peak data (Ix/Id + R), and the procedure independently repeated 1000 times. This gave 1000 simulated values for the activation parameters, for which the mean and standard deviation are reported in the text as ‡ ‡ ‡ ‡ ‡ ‡ ΔH = μ(ΔH sim) ± σ(ΔH sim) and ΔS = μ(ΔS sim) ± σ(ΔS sim).

6.4.4 X-ray Crystallography

6.4.4.1 X-ray Data Collection and Reduction

In preparation for analysis, crystals were first coated with Paratone-N oil in a glovebox and were

subsequently mounted on a MiTegen Micromount and placed in a N2 stream in order to maintain a dry and oxygen-free sample environment. Data collection was performed on a Bruker Apex II diffractometer and data collection strategies were determined employing Bruker provided Apex software. Optimization was performed in order to yield >99.5 % complete data to a minimum 2θ value of 55 °. All data sets were collected at 150(±2) K unless otherwise stated. The acquired frames were integrated employing the Bruker SAINT software package employing a narrow frame algorithm. Absorption corrections were conducted employing the empirical multi-scan method (SADABS).

6.4.4.2 Structure and Refinement

210 atom contributions were calculated however never refined. The locations of the largest peaks in the final difference Fourier map calculations as well as the magnitude of the residual electron densities in each case were of no chemical significance.

211

6.4.4.3 Tables of Crystallographic Data

Table 6.3 – Selected crystallographic data for 6-1, 6-4 and 6-5.

6-1 6-4 6-5

Formula C30H32BF10N2OP C60H66B2F24N4O2P2 C30H30BF12N2OP Formula wt 668.36 1414.73 704.34 Crystal system Triclinic Triclinic Orthorhombic

Space group P-1 P-1 P212121 a (Å) 10.3832(8) 11.6910(5) 11.6507(6) b (Å) 11.9066(9) 12.8326(5) 13.7877(6) c (Å) 14.5601(12) 13.5442(6) 19.5448(8) α (deg) 70.621(4) 63.637(2) 90 β (deg) 76.818(4) 88.082(2) 90 γ (deg) 65.912(4) 72.341(2) 90 V (Å3) 1541.2(2) 1722.62(13) 3139.6(2) Z 2 1 4 T (K) 150(2) 150(2) 150(2) d (calc) gcm-3 1.440 1.364 1.490 Abs coeff, μ, mm-1 0.178 0.171 0.187 Data collected 60932 27807 28943 R int 0.0309 0.0279 0.0426 # of indpndt reflns 16452 7846 7187

Reflns Fo≥2.0 σ (Fo) 12630 5627 6151 Variables 415 433 445 R (>2σ) 0.0377 0.0477 0.0367

wR2 0.1157 0.1337 0.0856 Goodness of fit 1.021 1.061 1.029

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

212

Table 6.4 – Selected crystallographic data for compounds 6-7 and 6-8.

6-7 6-8

Formula C37H47BClF3OP C55H42BF20N2OP Formula wt 672.95 1168.69 Crystal system Monoclinic Triclinic

Space group P21/c P-1 a (Å) 9.5346(5) 12.3838(3) b (Å) 21.1476(9) 12.5415(3) c (Å) 17.6877(8) 18.4914(5) α (deg) 90 94.7030(10) β (deg) 102.6040(10) 103.5460(10) γ (deg) 90 101.7340(10) V (Å3) 3480.5(3) 2708.21(12) Z 4 2 T (K) 150(2) 150(2) d (calc) gcm-3 1.284 1.433 Abs coeff, μ, mm-1 0.247 0.161 Data collected 44879 51443 R int 0.0366 0.0435 # of indpndt reflns 11037 13665

Reflns Fo≥2.0σ(Fo) 8728 9659 Variables 432 730 R (>2σ) 0.0437 0.0424 wR2 0.1165 0.1077 Goodness of fit 1.035 1.047

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

213

Chapter 7 The Exchange Chemistry of Main Group N2O Adducts with d-Block Lewis Acids

7.1 Introduction

2+ 7.1.1 Discovery of [Ru(NH3)5(N2O)]

In 1969, Armor and Taube described the first transition metal nitrous oxide (N2O) complex, 2+ 376 [Ru(NH3)5(N2O)] . Although this complex was not structurally confirmed at the time; mechanistic, theoretical and spectroscopic data have since been reported and support the original 346,377-379 2+ formulation. The coordination of the N2O molecule in [Ru(NH3)5(N2O)] has been the 1 center of much debate as N2O is thought interact with the Ru centre in either a η -N linear coordination mode or a η1-O bent binding mode. Unfortunately the unambiguous confirmation

of the coordination mode of N2O remains uncertain due to the inability to structurally 346 1 characterize the Ru(II) complex (Figure 7.1). Generally, however, the N2O η -N linear coordination mode is most widely supported and accepted.

Initial reports described the pentaamineruthenium(II) dinitrogen oxide complex as being unstable in solution,380 however, subsequent efforts have afforded stable and isolable materials with a 377,378,381-386 range of counter anions, [Ru(NH3)5(N2O)]X2 (X = Cl, Br, BF4) , although none of these complexes have been spectroscopically characterized. Other efforts to isolate and structurally characterize related transition metal complexes have only resulted in the isolation of metal oxide, nitride and nitrosyl species.346

2+ Figure 7.1 – (a) [Ru(NH3)5(N2O)] and (b) N2O resonance contributors.

214

7.1.2 Reaction of N2O with Transition Metal Complexes

Nature is known to efficiently process nitrous oxide by reaction with metal-containing enzymatic

sites. As a result, considerable research has focused on the study of the interaction of N2O with transition metal centres to better understand how N−N and N−O bond scission occurs naturally.

However, since the discovery by Armor and Taube, few transition metal complexes of N2O have been reported in the literature. This lack of stable and isolable complexes is likely due to the inability of the nitrous oxide moiety to function as a good ligand. The kinetically stable N2O molecule is neither a good σ-donor nor a good π-acceptor348 (resonance contributors listed in Figure 7.1). As a result, it was not until 2009 that a transition metal complex containing an intact 387,388 O-bound N2O fragment was crystallographically characterized.

Reaction of N2O with transition metal complexes typically involve oxidation of low-valent metals, as has been demonstrated upon the reaction of various divalent Ti, V, Cr and Ru

complexes with N2O, resulting in the liberation of N2 and the formation of the respective metal 388,389 390-398 oxides. Most commonly, the O atom of N2O undergoes insertion into M−carbon or M−hydride bonds399,400 while N−N and N−O bond scission also typically occurs at transition metal centers.348

In many of these reactions, insertion of the intact N2O moiety is speculated to occur first followed by bond scission; however, there is little experimental or theoretical evidence to support the formation of such intermediates.348 Early investigations by Hillhouse et al. focused

on the reaction of Hafnium metallocene complexes with N2O and provided examples for O-atom insertion into both M−C and M−H bonds. These results stemmed from the study of a solution of

Cp2*HfH(Ph) which was heated at 80 °C in the presence of N2O, yielding both Cp*2Hf(OH)Ph

and Cp2*HfH(OPh) in a 3:2 molar ratio. The N2O moiety is speculated to bind to the Hf centre 1 394 in a η -O fashion followed by N2 liberation. Subsequent migration of an H or Ph group yields oxidation of the coordinated ligands and hence the observed complexes (Scheme 7.1, a).

Subsequent reports by Hillhouse et al. documented the reaction of Ti and Zr diphenylacetylene

bis(pentamethylcyclopentadienyl) complexes with N2O where insertion of N2O into a M-C bond 1 was found to occur, effectively capturing the N2O moiety in a η −N MN(O)NC bonding mode 395 (Scheme 7.1, b). The planar metallocycle, encompassing the intact N2O fragment was

215

crystallographically characterized confirming the bonding motif. This η1-N binding mode has been theoretically shown to be the preferred form of interaction with transition metal species.346 This finding is somewhat counterintuitive considering the oxophilic nature of the early transition metals. This anomaly has been rationalized by the notion that electron donation from the metal to 1 the π* lowest unoccupied molecular orbital of the coordinated η -N N2O molecule stabilizes the ligand, leading to preference over the competing η1-O binding mode. The metallocycle subsequently undergoes decomposition at ambient temperatures, even in the solid-state, to yield oxametallocyclobutenes (Scheme 7.1, b).

394 Scheme 7.1 – Reaction of N2O with: (a) Cp*2Hf(H)Ph and (b) Cp*2M(PhCCPh) (M = Ti and Zr).395

Group 10 metal complexes such as L2NiR2 (L = PMes3; L2 = bipyridine, phenanthroline; R =

alkyl, aryl) have been shown to undergo reaction with nitrous oxide where the O atom of N2O has inserted into a Ni−R bond yielding a new aryloxide or alkoxide (OR) fragment. Cundari and

Hillhouse have also demonstrated that the reaction of (dtbpe)Ni(CPh2)

(dtbpe: bis(di-tert-butylphosphino)ethane) with N2O was found to yield a metallooxirane, the result of 1,3-dipolar cycloaddition. The insertion reaction was achieved employing exactly two equivalents of N2O, however, in the presence of excess N2O, the phosphine ligand was released as the oxide, in addition to the liberation of benzophenone and Ni metal. Theoretical calculations show that the enthalpic barrier to cycloaddition yielding the 5-membered transition state is small at 8.3 kcal/mol (Scheme 7.2).390

216

390 Scheme 7.2 – Reaction of (dtbpe)Ni(CPh2) with stoichiometric and excess N2O.

Although less documented, the direct hydrogenation of transition metal-bound N2O has been

reported by Caulton et al., and details the reduction of N2O to H2O and N2 employing a polyhydride unsaturated osmium complex.401

Many transformations of N2O at transition metal centres involve the scission of the N−O 325,402 bond with O-atom incorporation and N2 liberation. It is much more rare to observe the cleavage of the N−N bond. In a very recent report, Severin described the activation of N2O employing the basic and bulky N-heterocyclic carbene IMes, to yield the stable and isolable 351 zwitterion IMes(N2O). Coordination to the vanadium complex [V(Mes)3(THF)] results in the facile cleavage of the N−O bond, yielding a vanadium oxo-complex.403 The low valent vanadium species is a suitable trapping reagent to its highly oxophilic nature and propensity to react with

N2O. Subsequent heating of the oxo product resulted in N-N bond cleavage and migration to yield a new mesitylamide vanadium species (Scheme 7.3).404-406

217

Scheme 7.3 – N-O and N-N bond scission employing an NHC and a low valent V complex.403

7.2 Results and Discussion

7.2.1 Reaction of Early Metal Metallocenium Cations with tBu3P(N2O)B(p-F-C6H4)3

As previously described, the instances of main group and transition metal stabilized N2O adducts are limited and have only gained momentum in recent years. As precedence for the interaction

of early metal metallocene complexes with N2O was previously documented by Hillhouse et al.,390-395 our efforts focused on related chemistry, employing main group nitrous oxide adducts (chapter 6) as ligands for early transition metal centres. Research as described in chapter 6 revealed the ability to exchange the mild Lewis acid B(p-F-C6H4)3 acid from the

species tBu3P(N2O)B(p-F-C6H4)3 (6-6) for a Lewis acid of superior strength. This chemistry was further extended for the exchange with d-block metal complexes.

Group four metallocene complexes have been shown to react with electrophilic reagents such as arylboranes,187,407,408 the trityl cation409,410 and methylalumoxane (MAO)411 in order to yield effective Ziegler-Natta-type homogeneous single-site polymerization catalysts. An appropriate pairing of the metallocene and the co-catalyst has led to polymerization catalysts that have demonstrated extremely high activities in addition to high selectivities. Studies by Ewan and Marks in the early 1990s demonstrated the application of the highly electrophilic 187,407,408 perfluoroarylborane B(C6F5)3 in the activation of group four metallocenes.

218

The electrophilic nature of these metallocenium cations made excellent candidates for exchange. The first complex was prepared by conventional methods via methide abstraction from dimethyl

zirconocene, by a stoichiometric amount of the co-catalyst B(C6F5)3, yielding the reactive ion

pair [Cp2ZrMe][MeB(C6F5)3]. 6-6 was added to the vibrantly coloured yellow solution of the metallocenium cation and 7-1, was subsequently recovered as an oil in 83 % yield (Scheme 7.4).

Scheme 7.4 – Synthesis of 7-1 and 7-2 via Lewis acid exchange.

1H NMR analysis of the crude material revealed the presence of a singlet at 6.30 ppm, assignable to the cyclopentadienyl protons in addition to a doublet at 1.60 ppm pertaining to the tBu groups of the phosphine. Both a broad and a sharp singlet were similarly observable at 0.48 and 0.46

ppm and confirmed the presence of the MeB(C6F5)3 and ZrMe fragments, respectively. A narrow, sharp signal was observed by 11B{1H} NMR spectroscopy at -14.96, characteristic of the

MeB(C6F5)3 anion. It is noteworthy, that free B(p-F-C6F5)3 and 6-6 were both completely absent following work-up. Repetition of the reaction employing the 15N isotopologue of 6-6 supported

the retention of the P(N2O) fragment as evidenced by the preservation of the N−N and P−N couplings, in both the 15N and 31P{1H} NMR spectra. A doublet of doublets was observed by 31 1 1 2 P{ H } NMR spectroscopy at 67.06 ppm with JP-N and JP-N coupling constants of 61 and 16 Hz, respectively. Two doublets of doublets were also observed in the 15N NMR spectrum at 1 587.60 and 398.46 ppm with an associated JN-N coupling constant of 17 Hz. These data were

consistent with the formal exchange of B(p-F-C6H4)3 for [Cp2ZrMe], yielding

[tBu3P(N2O)ZrCp2Me][MeB(C6F5)3]. The solid-state formulation was confirmed by single crystal X-ray diffraction (Figure 7.2).

219

Figure 7.2 – POV-Ray depiction of the molecular structure of 7-1. C: black, N: aquamarine, F:

pink, O: red, P: orange, Zr: light steel blue. H atoms and the B(C6F5)4 ion were removed for clarity. Selected bond distances (Å) and angles (°). Zr(1)-O(1), 2.105(3); Zr(1)-C(65), 2.261(4); N(1)-N(2), 1.233(5); N(1)-O(1), 1.372(5); N(2)-P(1), 1.719(4); N(2)-O(1)-Zr(1), 126.2(2).

The molecular structure of 7-1 constitutes the first example of the structural characterization of a

group four metal in which the N2O fragment is bound to the zirconium centre in a linear η1-O fashion. The observed bonding mode is notably different from the related intermediate titanium and zirconium metallocyclo complexes as described by Hillhouse,395 which bear a M−N bond and an exocyclic O atom (Scheme 7.1, b). This example is not, however, directly

comparable due to the pre-activated N2O fragment in tBu3P(N2O)B(p-F-C6H4)3 and as a result of the stabilization provided by the phosphine.

The phosphine and the OZrCp2Me units were disposed in a trans orientation relative to the N−N 349,350 double bond, as was observed in the analogous main group compounds, tBu3P(N2O)BR3. The Zr(1)−O and Zr(1)−C(65) bond lengths were found to be 2.105(3) and 2.261(4) Å while the N(1)−N(2), N(1)−O(1) and N(2)−P(1) bond lengths were found to be 1.233(5), 1.372(5) and 1.719(4) Å, respectively. These metrical parameters bear distinct similarities to, and are statistically indistinguishable from, the aforementioned B/P zwitterionic N2O complexes. These

data indicate that regardless of the nature of the Lewis acid, the tBu3P(N2O) moiety is essentially unperturbed. The Zr(1)−O(1)−N(1) bond angle in 7-1 was noted to be 126.2(2)° while the

220

average X−O−N (X = B, C) angle of the related main group species was found to have an average angle of 111.8°. The considerably larger bond angle in 7-1 is indicative of a secondary

interaction between the O lone pair of the N2O fragment and a vacant metal d orbital on zirconium, which is absent in the main group analogues. An examination of the metallocene

fragment revealed a Cp(cent)−Zr−Cp(cent) angle of 130.5°, which is only slightly larger than the 412 413 related zirconocene dichloride Cp2ZrCl2 or the ion pair [Cp2ZrMe(THF)][BPh4] with

Cp(cent)−Zr−Cp(cent) angles of 129.2 and 129.6°, respectively. Comparison of the Zr−Me and Zr−O

bond angles of 7-1 (2.261(4) and 2.105(3) Å) with those of [Cp2ZrMe(THF)][BPh4] (2.256(10) and 2.122(14) Å)413 reveals that the metrical parameters are statistically indistinguishable.

In an analogous fashion, the stoichiometric reaction of Cp2TiMe2 and B(C6F5)3 was performed generating the metallocenium cation in situ. Solid 6-6 was added to the red/brown solution resulting in an intensely red coloured solution. Manipulation afforded 7-2, as an orange solid in 81 % yield. Spectroscopic data revealed features resembling that of the related compound, 7-1. Signals pertaining to the metallocene fragment and the counter anion were visible in the 1H

NMR spectrum at 6.30, 1.06 and 0.47 ppm confirming the presence of the Cp, Ti-CH3 and the 1 methylborate moieties. The Ti-CH3 H NMR resonance was noted to be downfield relative to

the analogous Zr-CH3 resonance (0.46 ppm). This is consistent with the observation made by Rausch, where upon substitution of a 1st row for a 2nd row metal within the same group, a slight 414 upfield shift is noted for the methyl signals of the species Cp2M(CH3)2. A doublet was also 1 3 visible by H spectroscopy at 1.58 ppm with a JH-P coupling of 16 Hz, confirming the presence

of the PtBu3 fragment. The sum of the NMR data and elemental analysis confirmed the

formulation of 7-2 as, [tBu3P(N2O)TiCp2Me][MeB(C6F5)3]. It is of note that much like the B/P

N2O adducts, the group four metal and phosphine stabilized N2O adducts are robust and stable for seven days in solution without appreciable decomposition.

The ability to generate stable early metal adducts of N2O without decomposition to yield O-atom

insertion and N2 liberation, led to the question of whether a transition metal FLP could be applied

in the direct activation of N2O. This would avoid the indirect and uneconomical method of Lewis acid exchange.

Stoichiometric mixtures of the preformed metallocenium cations [Cp2TiMe] and [Cp2ZrMe] with

PtBu3 were subjected to an atmosphere of N2O. Following overnight reaction at ambient

221

temperature, a number of inseparable and unidentifiable products were observed by 31P{1H}

NMR spectroscopy. This finding was not surprising as N2O is known to react with early metal complexes resulting in oxygen insertion into M-C bonds yielding alkoxy- or aryloxide ligands.

Cp*2ZrMe(OMe) was targeted as a suitable Lewis acid as the metallocenium cation would lack any M-alkyl bonds and would therefore circumvent undesirable side reactions in the presence of

N2O. The steric bulk provided by the incorporation of the pentamethylcyclopentadienyl ligands ensures that bridging bimetallic complexes are prohibited. The Cp* ligands, however, are more electron donating than the Cp ligands and therefore contribute to the reduction of the Lewis acidity of the Zr centre.

Cp*2ZrMe(OMe) was synthesized from the reaction of Cp*2ZrMe2 and 1.1 equivalents of

methanol, yielding the desired bent metallocene and an equivalent of CH4 (Scheme 7.5). The

metallocenium species [Cp*2Zr(OMe)][(B(C6F5)4] was generated in situ via reaction of the pre-

catalyst with [Ph3C][B(C6F5)4]. Previous investigations demonstrated that the scope of Lewis

bases that are active in the complexation of N2O is extremely limited. PtBu3 and PCy3 were determined to be the only phosphines capable of N2O activation in the presence of

perfluoroarylboranes. These bases were therefore tested with [Cp*2Zr(OMe)][(B(C6F5)4] in order to establish whether they form frustrated Lewis pairs in solution. The stoichiometric reaction of

PCy3 with [Cp*2Zr(OMe)][(B(C6F5)4] yielded an adduct whereas the analogous reaction with 31 1 PtBu3 demonstrated no interaction by P{ H} NMR spectroscopy (Scheme 7.6).

Scheme 7.5 – Synthesis of Cp*2ZrMe(OMe).

A stoichiometric reaction of in situ generated [Cp*2Zr(OMe)][(B(C6F5)4] and PtBu3 was prepared and subjected to an atmosphere of N2O. Sixteen hours reaction time and manipulation afforded a yellow solid 7-3, in 74 % yield. Signals attributable to the pentamethylcyclopentadienyl rings and Zr-OMe fragment were observed at 1.91 and 3.98 integrating in a 30:3 ratio. 31P{1H} NMR spectroscopy of the 15N isotopologue of 7-3 revealed a

222

doublet of doublets at 64.33 ppm, indicative of one and two bond P-N coupling. Similarly, two resonances were observed in the 15N NMR spectrum at 591.01 and 386.28 ppm, supporting the 1 2 1 presence of the N2O fragment. JP-N, JP-N and JN-N coupling constants of 62, 16 and 17 Hz were 11 1 determined experimentally. The B(C6F5)4 anion was confirmed by a sharp signal in the B{ H} NMR spectrum in addition to three signals in the 19F NMR spectrum with a narrow p-m gap of

3.81 ppm. These data supported the direct activation of N2O by the transition metal and

phosphine-based frustrated Lewis pair, yielding [tBu3P(N2O)ZrCp*2(OMe)[B(C6F5)4]. Similarly,

[tBu3P(N2O)ZrCp*2(OMe)[MeB(C6F5)3] 7-4, was accessible by methide abstraction from

Cp*2ZrMe(OMe) by B(C6F5)3 followed by Lewis acid exchange with 6-6 (Scheme 7.6)

Scheme 7.6 – Synthesis of 7-3 and 7-4.

Crystals of [tBu3P(N2O)ZrCp*2(OMe)[B(C6F5)4] were found to be extremely sensitive upon exposure to air and readily underwent decomposition prior to mounting and data collection.

However, diffraction quality single crystals of 7-3 were grown from a solution of CH2Cl2 and pentane at -35 °C, and an X-ray crystallographic study unambiguously confirmed the molecular structure of 7-3 (Figure 7.3).

223

Figure 7.3 – POV-Ray depiction of the molecular structure of 7-3. C: black, N: aquamarine, F:

pink, O: red, P: orange. Zr: light steel blue. H atoms and B(C6F5)4 ion removed for clarity. Selected bond distances (Å) and angles (°). Zr(1)-O(2), 1.916(2); Zr(1)-O(1), 2.151(3); P(1)- N(2), 1.683(3); O(1)-N(1), 1.300(4); O(2)-C(21), 1.397(4); N(1)-N(2), 1.266(4); O(1)-Zr(2)- O(2), 96.27(11); N(1)-O(1)-Zr(1), 123.4(2); C(21)-O(2)-Zr(1), 169.9(2); N(2)-N(1)-O(1), 111.9(3); N(1)-N(2)-P(1), 116.9(3).

The molecular structure of 7-3 confirmed that N2O was O-bound to the Zr centre and N-bound to the phosphine, while the orientation of the OZr and phosphine moieties remained disposed in a

trans orientation about the N-N double bond. A comparison of the Zr−O bond lengths of 7-3

revealed a considerable degree of variation which indicated that the tBu3P(N2O) and the OMe moieties were interacting with the Zr(IV) centre in differing degrees. The Zr−OMe bond length was found to be 1.916(2) Å, which is consistent with other zirconium alkoxides. The Zr-OMe

bond length in 7-3 is drastically shortened as compared to the Zr−O bond of the Zr−(ON2)PtBu3 fragment, indicating a greater degree of interaction between the O lone pairs of the OMe moiety with a vacant Zr d-orbital. The larger Zr−O−C bond angle of 169.9(2)° (slightly distorted from linearity) further supported the increased donation of the methoxide ligand to the Zr centre as

compared to the Zr−O−N bond angle of 123.4(2)° for the Zr−(ON2)PtBu3 fragment.

Comparison of the molecular structure of 7-3, to that of the related species 7-1, indicated that the Zr−O bond in 7-1 is shorter compared to the analogous bond in 7-3 which is consistent with the

224

increased Lewis acidity of the Cp2ZrMe fragment. B/P-based N2O adducts, as described in the aforementioned chapters, typically do not see a considerable perturbation of the N2O fragment upon substitution of the Lewis acidic fragment. Comparatively, substitution with a transition

metal centre imparts considerable change on the N2O moiety as evidenced by the shortening of the P−N and N−O bonds in 7-3 (1.683(3) and 1.300(4) Å), compared to the analogous bonds in 7-1 (1.719(4) and 1.372(5) Å). In an opposing fashion, the N−N bond in 7-3 (1.266(4) Å) was found to be elongated as compared to the analogous bond in 7-1 (1.233(5) Å).

7.2.2 Zinc and Phosphine Stabilized Complexes of Nitrous Oxide

The aforementioned chemistry illustrated the ability of group 4 metallocenium cations to accommodate the tBu3P(N2O) species into the metal coordination sphere via Lewis acid

exchange with tBu3P(N2O)B(p-F-C6H4)3. In a unique fashion, the [Cp*2Zr(OMe)] cation and

PtBu3 function as a frustrated Lewis pair collaboratively mediating the activation of N2O yielding an ion pair analogous to those generated via exchange (7-1 and 7-2). Due to the reactivity employing early d-block elements, the investigation of late d-block based Lewis acids was warranted.

The species bis(pentafluorophenyl)zinc Zn(C6F5)2, is a highly electrophilic Lewis acid and is 415 commonly employed as a C6F5 transfer reagent in synthesis. Zn(C6F5)2 can be isolated in its base-free and monomeric form where its C−Zn−C bond angle approaches linearity. Like many

perfluoroaryl Lewis acids, Zn(C6F5)2 readily generates adducts with common N, O and P donor 415 molecules. Early studies demonstrated the synthesis of a 2:1 triphenylphosphine:Zn(C6F5)2 416 adduct and a 1:1 diphosphine bis(diphenylphosphino)ethane:Zn(C6F5)2 adduct (Figure 7.4).

Figure 7.4 – (a) 1:2 and (b) 1:1 adducts of Zn(C6F5)2 with triphenylphosphine and bis(diphenylphosphino)ethane.

225

Previous reports demonstrate that N2O activation is achievable employing a plethora of Lewis 350 acids but is limited by the nature of the Lewis base. Our efforts have demonstrated that PtBu3 and PCy3 typically function as suitable Lewis bases in this chemistry. Therefore stoichiometric

mixtures of Zn(C6F5)2 and PR3 (R = tBu, Cy) were prepared. NMR spectroscopy revealed the generation of 1:1 Lewis acid-base adducts in both instances. Stephan et al. recently reported the

application of the strongly bound Lewis acid-base adduct (Ph3P)B(C6F5)3 in reaction with alkynes, yielding 1,2-addition products in analogous methods to “typical” sterically encumbered 71 FLPs. The adducts, (tBu3P)Zn(C6F5)2 and (Cy3P)Zn(C6F5)2, assuming some degree of

equilibrium with the free components, might serve to collaboratively activate N2O. Exposure of

solutions of the 1:1 adducts of Zn(C6F5)2 and PR3 (R = tBu, Cy) yielded no reaction even following prolonged reaction (3 days).

Alternatively, the toluene adduct of Zn(C6F5)2 was reacted with a pre-activated source of N2O,

tBu3P(N2O)B(p-F-C6H4)3 (6-6). The stoichiometric reaction, initially found to be opaque, cleared within a few seconds of stirring. 7-5 was recovered quantitatively as a white solid (Scheme 7.7).

Scheme 7.7 – Synthesis of 7-5.

1 The H NMR spectrum revealed that the product contained the PtBu3 moiety as evidenced by a 3 doublet observable at 1.51 ppm with a JH-P coupling constant of 14 Hz. In a similar fashion, a resonance was observable in the 31P{1H} NMR spectrum at 65.83 ppm, a mere 1.56 ppm downfield compared to the parent zwitterion, 6-6 (64.27 ppm). The resonance of the 15N

isotopologue of 7-5 bore the expected coupling pattern for the R3P(N2O) moiety, with two 1 2 doublet of doublets with a N−N one bond coupling constant of 18 Hz. The JN-P and JN-P coupling constants were found to be 54 and 9.3 Hz, drastically reduced as compared to the B/P derived adducts. An examination of the 19F NMR spectrum revealed no traces of 6-6 or the free

borane B(p-F-C6H4)3 following work-up, which is indicative of Lewis acid exchange. Three new

226

sharp signals were observed at -117.44, -157.71 and -162.64 ppm and were assignable to the two

freely rotating C6F5 rings bound to Zn. Collectively, these data suggested the empirical

formulation of 7-5 to be tBu3P(N2O)Zn(C6F5)2. A crystal structure determination established the

centrosymmetric and dimeric nature of the product in which two tBu3P(N2O) fragments were

bridging two Zn centres, yielding a central Zn2O2 core (Figure 7.5).

Metrical parameters confirmed the pseudo-tetrahedral geometry of the P centre and the distorted tetrahedral geometry at the Zn centre. (sum of the angles: Zn - 643°; P - 656°). The Zn−O bond

distances of the Zn2O2 core were found to be 2.088(2) and 2.144(2) Å, which are comparable to 417 the four-coordinate Zn species [(C6F5)2Zn(THF)2] with Zn−O bond distances of 2.113(3) and 2.093(2) Å. The Zn−C bonds in 7-5 (2.000(3) and 2.011(3) Å) were found to be statistically indistinguishable from those of the bis-THF adduct (1.9999(4) and 2.012(3) Å)417. The Zn−O−Zn’ and O−Zn−O angles were determined to be 107.15(10) and 72.85(8)° while the N−N and N−O bond distances were found to be 1.267(4) and 1.307(3) Å, which is considerably 361 elongated as compared to the analogous bonds in free N2O (N−N: 1.127 Å; N−O: 1.186 Å). Comparison of the metrical parameters of 7-5 to those of the related main group zwitterion tBu3P(N2O)B(C6F5)3, showed a slight shortening of the N−N double and elongation of the N−O bond. In addition, the N−O−Zn angles were found to be 125.46(18) and 125.40(18)° (as

compared to 114.43(1)° in tBu3(N2O)B(C6F5)3) indicative of significant interaction with each of the two Zn centres. The dimeric nature of 7-5 was found to position Zn(1) proximal to N(1) at a

non-bonded distance of 3.035(2) Å and it was noted that the ZnO and PtBu3 moieties were disposed in a trans orientation relative to the N−N double bond in an analogous manner as

observed for main group species of the form R’3P(N2O)BR3.

227

Figure 7.5 – POV-Ray depictions of (a) the full molecular structure and (b) the core structure of 7-5. C: black, N: aquamarine, F: pink, O: red, P: orange. Zn: red-brown. All H atoms removed

for clarity. Additionally, the C6F5 rings on Zn and the tert-butyl methyl groups have been removed in (b) for clarity. Selected bond distances (Å) and angles (°). P(1)-N(1), 1.702(3); N(1)-N(2), 1.267(4); N(2)-O(1), 1.307(3); O(1)-Zn(1), 2.088(2); O(1)-Zn’(1), 2.144(2); Zn(1)- C(1), 2.000(3); Zn(1)-C(7), 2.011(3); P(1)-N(1)-N(2), 114.32(2); N(1)-N(2)-O(1), 11.7(2); N(2)- O(1)-Zn(1), 125.46(18); N(2)-O(1)-Zn’(1), 125.40(18); Zn(1)-O(1)-Zn’(1), 107.15(9); O(1)- Zn(1)-O(1), 72.85(9).

228

Reaction of 6-6 with 1.5 equivalents of Zn(C6F5)2 resulted in the clean formation of 7-6, which was isolated was a white solid in 81 % yield (Scheme 7.8). 31P{1H} NMR spectroscopy of the 15 N isotopologue of 7-6 confirmed the retention of the tBu3P(N2O) linkage as evidenced by a 1 2 doublet of doublets at 68.54 ppm with JP-N and JP-N coupling constants of 54 and 9.3 Hz, respectively. Similarly, two resonances were located in the 15N NMR spectrum at 595.17 and 1 19 323.78 pm with a JN-N coupling constant of 18 Hz. The signals in the F NMR spectrum appeared at nearly identical chemical shifts to 7-5. Based on the similarity of the 31P{1H}, 15N 19 and F NMR data, exchange of the B(p-F-C6H4)3 fragment in 6-6 for Zn(C6F5)2 was suspected although the formulation could not be unambiguously confirmed via spectroscopic techniques alone.

Scheme 7.8 – Synthesis of 7-6.

A crystallographic study of 7-6 revealed a C2 symmetric molecule whose formulation involved

the coordination of two tBu3P(N2O) fragments to a central Zn(C6F5)2 molecule with two

additional Zn(C6F5)2 units bridging the O and N-atoms of the N2O ligands (Figure 7.6). The

geometry about the central Zn(C6F5)2 unit was found to be pseudo-tetrahedral and was linked to

two independent N2OPtBu3Zn(C6F5)2 units by Zn−O linkages with an average length of

2.1176(19) Å. The Zn(2) units which bridge the tBu3P(N2O) fragments were found to be

coordinated to both an O and N atom of the N2O unit, yielding a chelating four-membered

ZnN2O ring. The Zn−O and Zn−N bond distances were found to be 2.1841(18) and 2.242(2) Å while the N−Zn−O bond angle was determined to be 56.91(9)o. Despite the two chelating four- membered rings in 7-6, the metrical parameters for the N−N, N−O and N−P bonds were found to be statistically indistinguishable from those of the dimeric species, 7-5. The greater steric

congestion around the central Zn(1) in 7-6 forces the Zn(C6F5)2 fragment to be almost perpendicular to the PN2O plane (C-Zn(1)-C / N-N-O interplanar angle 7-6: 67.1(3)°. This

results in a close approach of two C6F5 rings in 7-6, with concomitant displacement of Zn(2) away from O(1) towards N(1).

229

Figure 7.6 – POV-Ray depictions of (a) the full molecular structure and (b) the core structure of 7-6. C: black, N: aquamarine, F: pink, O: red, P: orange. Zr: red-brown. All H atoms removed for clarity. Additionally, the C6F5 rings on Zn and the tert-butyl methyl groups have been removed from (b) for clarity. Selected bond distances (Å) and angles (°). Zn(1)-O(1), 2.1176(19); Zn(2)- N(1), 2.242(2); Zn(2)-O(1), 2.1842(18); O(1)-N(2), 1.301(3); N(1)-N(2), 1.287(3); N(1)-P(1), 1.702(2); O(1)-Zn(1)-O(1), 135.57(9); Zn(1)-O(1)-N(2), 119.66(15); O(1)-N(2)-N(1), 109.2(2); N(2)-N(1)-P(1), 114.55(17).

230

Examination of the 19F NMR spectrum of 7-6 revealed three sharp signals assignable to the o-, p-

and m-F atoms of the respective C6F5 rings. This was surprising due to the two Zn(C6F5)2 chemical environments present in the molecule. This pointed toward facile exchange between the two Zn(C6F5)2 environments at room temperature. An NMR sample of

7-6 in CD2Cl2 was cooled to -75 °C and revealed two Zn(C6F5)2 environments, present in a 2:1 ratio, consistent with the solid-state structure of 7-6. A variable temperature NMR study revealed the decoalescence of the ortho-F signals at -19.6 °C (following temperature correction) ‡ corresponding to a ΔG = 11.7 kcal/mol, for the process of Zn(C6F5)2 exchange (Figure 7.7).

Figure 7.7 – Low temperature 19F NMR spectroscopy of 7-6.

231

The exchange is presumed to occur via dissociation of a weak Zn-N bond with rotation about the

N−O bond, followed by coordination to a second Zn(C6F5)2 unit. Employing this process, all three Zn(C6F5)2 units can rotate through the two Zn environments (Scheme 7.9).

Scheme 7.9 – Proposed mechanism of Zn(C6F5)2 site exchange in 7-6.

In a similar fashion, two equivalents of Zn(C6F5)2 were reacted with a single equivalent of 6-6 (Scheme 7.10). Immediate reaction was noted and 7-7, was isolated as a white solid in 80 % yield. Examination of the 1H and 19F NMR spectra revealed resonances that were nearly identical to those observed for 7-5 and 7-6. Repetition of the experiment employing the 15N 1 2 1 isotopologue of 6-6, confirmed the retention of the P(N2O) fragment with JP-N, JP-N and JN-N coupling constants of 54, 11 and 17 Hz, as observed by 31P{1H} and 15N NMR spectroscopy. An X-ray crystallographic study unambiguously confirmed the formulation of 7-7 as

[tBu3P(N2O)Zn(C6F5)2]Zn(C6F5)2 (Figure 7.8).

232

Scheme 7.10 – Synthesis of 7-7.

The solid-state structure of 7-7 indicated that the molecule contained a single tBu3P(N2O)Zn(C6F5)2 fragment and an additional Zn(C6F5)2 unit, which was found to bridge the

N-atom, bound to P, and the O-atom of the N2O moiety. Typically, Zn species adopt stable two- coordinate linear or four-coordinate tetrahedral geometries, making the three-coordinate Zn, bound by two perfluoroaryl rings and an O-atom, quite unusual.279,280 The Z(1)−O bond distance was determined to be 2.0912(9) Å while the C−Zn(1)−C angle was found to be 153.23(6)o. The Zn(2) atom, bound by two perfluoroaryl rings, a N and O atom, was found to

adopt a pseudo-tetrahedral coordination sphere yielding a ZnN2O four-membered chelate, reminiscent of that observed for 7-6. The Zn(2)−O and Zn(2)−N distances were found to be 2.1435(10) and 2.3086(12) Å, respectively, while the O−Zn(2)−N chelate bite-angle at Zn(2) was 56.38(4)o.

It is noted that the C6F5 rings of 7-7 were represented by only three distinct resonances at room

temperature, indicative of rapid exchange between the two Zn(C6F5)2 environments. Low temperature 19F NMR studies of 7-7 revealed the decoalescence of the ortho-F signals at -34.6 °C which allowed for the determination of the ΔG‡ for the process of exchange (10.9 kcal/mol).

Six resonances, pertaining to two separate C6F5 environments, cleanly resolved at -75 °C, integrating in 1:1 fashion. The mechanism for exchange is thought to be similar to that observed for 7-6 where the process is initiated by cleavage of the weak Zn-N bond and is followed by

rotation about the N-O bond prior to coordination of the second Zn(C6F5)2 unit regenerating the

ZnN2O core (Scheme 7.11).

233

Figure 7.8 –POV-Ray depictions of (a) the full molecular structure and (b) the core structure of 7-7. C: black, N: aquamarine, F: pink, O: red, P: orange. Zr: red-brown. All H atoms removed for clarity. Additionally, the C6F5 rings on Zn and the tert-butyl methyl groups have been removed from (b) for clarity. Selected bond distances (Å) and angles (°). P(1)-N(1), 1.7103(11); N(1)- N(2), 1.2793(15); N(2)-O(1), 1.3057(15); Zn(1)-O(1), 2.0912(9); Zn(2)-N(1), 2.3086(12); Zn(2)- O(1), 2.1435(10); Zn(1)-O(1)-Zn(2), 139.95(5); N(1)-Zn(1)-O(1), 56.38(4).

Scheme 7.11 – Process of Zn(C6F5)2 exchange in 7-7.

Compounds 7-5 through 7-7 can be synthesized by direct methods from one equivalent of 6-6

employing either 1, 1.5 or 2 equivalents of Zn(C6F5)2. Compounds 7-6 and 7-7 can alternatively

be made from 7-5 employing 1 and 2 equivalents of Zn(C6F5)2. Finally, 7-7 can be generated

234

from 7-6 employing a single equivalent of Zn(C6F5)2. Due to the facile conversion between compounds 7-5, 7-6 and 7-7, extreme care must be taken when handling the starting materials, as stoichiometry plays a crucial role in product formation (Scheme 7.12).

Scheme 7.12 – Synthesis of 7-5, 7-6 and 7-7 by direct and indirect methods.

The characterization of the solid-state structure of compounds 7-5 through 7-7 was essential in the confirmation of the formulations of these species. Comparison of the 19F and 31P{1H} NMR data for compounds 7-5, 7-6 and 7-7 revealed little perturbation between the resonances for each compound (Figure 7.9). Comparison of the metrical parameters revealed a similar trend, as there is little variation noted in the bond lengths of the PN2O fragment. A marginal elongation of the

N−N double bond is observed upon coordination of a Zn(C6F5)2 group across the N2O moiety to

yield the ZnN2O four-membered chelate. In a related fashion, the N-N-O bond angle is found to

be more acute in compounds bearing the ZnN2O chelate. Based on these observations, it is clear

that coordination of the second Zn(C6F5)2 unit has very little effect on the bound N2O molecule.

235

It was additionally noted that the terminal Zn−O bond in 7-7 is shorter as compared to the bridging Zn in the ZnN2O chelate, in accordance with the coordinatively unsaturated nature of the three-coordinate Zn species.

Figure 7.9 –19F and 31P{1H} spectra for compounds 7-5 through 7-7.

7.3 Conclusions

Combination of a sterically encumbered metallocenium cation, which is precluded from reaction with either phosphine or N2O independently, has the ability to collaboratively activate nitrous oxide in the presence of PtBu3, in an fashion analogous to main group species. This resulted in the elucidation of the first crystallographically characterized group four Lewis acid and 1 phosphine stabilized adduct of η -O bound N2O. In strong Lewis acid-base adducts where FLP-

236

type reactivity is precluded, phosphine and borane stabilized adducts of N2O can be employed in Lewis acid exchange chemistry to yield novel Zr, Ti and novel Zn complexes which are otherwise inaccessible through tradition routes.

7.4 Experimental Section

7.4.1 General Considerations

All manipulations were carried out under an atmosphere of dry, O2-free N2 employing standard

Schlenk-line and glovebox techniques. Solvents (pentane, hexanes, toluene, CH2Cl2) were dried by employing a Grubbs-type column system (Innovative Technology), degassed and stored under

N2. Cyclohexane was distilled under N2 from Na/benzophenone while bromobenzene was

distilled under N2 from CaH2. CD2Cl2 was vacuum transferred from CaH2, degassed and stored 15 under N2. PtBu3 (Strem Chemicals), N2O (Sigma-Aldrich; 99%), N2O (Cambridge Isotope 15 Laboratories; 99.9%, 98.8% N enriched), [Ph3C][B(C6F5)4] (Aldrich) and B(C6F5)3 (Boulder

Scientific) were used as received. Zn(C6F5)2 (Sigma-Aldrich; 97%) was recrystallized from

toluene at -35 °C, yielding the toluene adduct, tol·Zn(C6F5)2. The reagent

tBu3P(N2O)B(p-F-C6H4)3 was prepared according to procedures outlined in chapter 6 (6-6) while 321 414 418 Cp2TiMe2, Cp2ZrMe2 and Cp*2ZrMe(OMe) were prepared according to literature procedures. 1H, 11B, 13C, 15N, 19F and 31P NMR spectra were recorded at 25 °C, unless otherwise stated, on a Varian NMR System 400 MHz or Bruker Avance III 400 MHz spectrometer and 1 13 were referenced using (residual) solvent resonances relative to SiMe4 ( H, C), or to an external 11 19 31 15 15 standard ( B: (Et2O)BF3, F: CFCl3, P: 85% H3PO4, N: NH3(l) via the N resonance of 90% 372 formamide in DMSO-d6 at 112 ppm). Chemical shifts are reported in ppm and coupling constants as scalar values in Hz. Combustion analyses were performed in house employing a Perkin-Elmer CHN Analyzer.

7.4.2 Syntheses

Synthesis of 7-1. A 20 mL scintillation vial was charged with

Cp2ZrMe2 (0.076 g, 0.260 mmol) and toluene (2 mL). A second 10

mL vial was charged with B(C6F5)3 (0.133 g, 0.260 mmol) and

toluene (5 mL). The borane solution was slowly added to the solution of Cp2ZrMe2 at room temperature. The resulting solution was bright yellow and was allowed to stir for an additional 5 minutes. At this time, 6-6 (0.141 g, 0.260 mmol) was added resulting in an opaque pale yellow

237

solution. The reaction was allowed to stir for a period of one hour. Pentane was added (10 mL) precipitating a beige oil. The oil was taken up in CH2Cl2 (2 mL) and triturated with pentane (15 mL). The solvents were then decanted from the solid. The product was washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours. Yield: 0.226 g (83 %). Crystals suitable for X-ray 1 crystallography were grown from a layered solution of CH2Cl2 and pentane at -35 °C. H NMR 3 (CD2Cl2): δ 6.30 (s, 10H, C5H5); 1.60 (d, 27H, JH-P = 14, PtBu); 0.48 (br s, 3H, H3CB(C6F5)3); 11 1 13 1 0.46 (s, 3H, ZrCH3). B{ H} NMR (CD2Cl2): δ -14.96 (s). C{ H} NMR (CD2Cl2): δ 148.88 1 1 1 (br d, JC-F = 238 Hz, o-C6F5); 138.07 (br d, JC-F = 241 Hz, p-C6F5); 136.94 (br d, JC-F = 244 Hz, 1 m-C6F5); 129.10 (s, ipso-C6F5); 113.39 (s, C5H5); 42.07 (d, JC-P = 29 Hz, PC{CH3}3), 29.94 (s, 15 2 1 PC{CH3}3); 10.59 (br s, H3CB(C6F5)3). N NMR (CD2Cl2): δ 587.60 (dd, JN-P = 16 Hz, JN-N = 1 1 19 17 Hz, PNNO); 398.46 (dd, JN-P = 61 Hz, JN-N = 17 Hz, PNNO). F NMR (CD2Cl2): δ -133.11 3 3 (d, 6F, JF-F = 20 Hz, o-C6F5); -165.25 (t, 3F, JF-F = 21 Hz, p-C6F5); -167.81 (m, 6F, m-C6F5). 31 1 1 1 P{ H} NMR (CD2Cl2): δ 67.06 (dd, JP-N = 61 Hz, JP-N = 16 Hz). Anal. Calcd. for

C42H43BF15N2OPZr: C, 49.96; H, 4.29; N, 2.77. Found: C, 49.83; H, 4.50; N, 2.54 %.

Synthesis of 7-2. A 20 mL scintillation vial was charged with

Cp2TiMe2 (0.050 g, 0.240 mmol) and toluene (5 mL). A second 10

mL vial was charged with B(C6F5)3 (0.123 g, 0.240 mmol) and

toluene (5 mL). The borane solution was slowly added to the solution of Cp2TiMe2 at room temperature. The resulting solution dark red and was allowed to stir for an additional 5 minutes. At this time, 6-6 (0.130 g, 0.240 mmol) was added and the reaction was allowed to stir for a period of one hour. Pentane was added (10 mL) precipitating a dark red oil. The oil was taken up in CH2Cl2 (2 mL) and triturated with pentane (15 mL). The solvents were then decanted from the solid. The orange product was washed with pentane (3 x 5 mL) and dried in vacuo for 2 1 3 hours. Yield: 0.232 g (81 %). H NMR (CD2Cl2): δ 6.30 (s, 10H, C5H5); 1.58 (d, 27 H, JH-P = 14 11 1 Hz, PtBu); 1.06 (s, 3H, TiCH3); 0.47 (br s, 3H, H3CB(C6F5)3). B{ H} NMR (CD2Cl2): δ 14.93 13 1 1 1 (s). C{ H} NMR (CD2Cl2): δ 148.88 (br d, JC-F = 240 Hz, o-C6F5); 138.11 (br d, JC-F = 244 1 Hz, p-C6F5); 136.99 (br d, JC-F = 244 Hz, m-C6F5); 129.62 (br s, ipso-C6F5); 116.43 (s, C5H5); 1 19 42.07 (d, JC-P = 29 Hz, PC{CH3}3); 29.93 (s, P{C(CH3)3}); 10.54 (br s, H3CB(C6F5)3). F 3 3 NMR (CD2Cl2): δ 132.21 (d, 6F, JF-F = 20 Hz, o-C6F5); 164.42 (t, 3F, JF-F = 20 Hz, p-C6F5); 31 1 166.97 (m, 6F, m-C6F5). P{ H} NMR (CD2Cl2): δ 64.36 (s). Anal. Calcd. for

C52H43BF15N2OPTi: C, 52.17; H, 4.49; N, 2.90. Found: C, 52.02; H, 4.25; N, 2.80 %.

238

Synthesis of 7-3. A 20 mL scintillation vial was charged with

Cp*2Zr(OMe)Me (0.060 g, 0. 147 mmol) and C6H5Br (5 mL). A

second 10 mL vial was charged with [Ph3C][B(C6F5)4] (0.136 g,

0.147 mmol) and C6H5Br (2 mL). The trityl borate solution was added in a dropwise fashion to

the solution of Cp*2Zr(OMe)Me and was stirred at room temperature for 5 minutes. At this time,

PtBu3 (0.030 g, 0.148 mmol) in bromobenzene (1 mL) was added resulting in a deep yellow solution. The solution was then transferred to a 100 mL bomb, was degassed and backfilled with

N2O (1 atmosphere). The reaction was stirred at r.t. for 12 hours. Pentane (10 mL) was added precipitating a yellow oil which was subsequently taken up in CH2Cl2 (2 mL), filtered through a plug of Celite and triturated with pentane (15 mL). The solvents were decanted from the pale yellow solid. The product was washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours.

Crystals suitable for X-ray diffraction were grown from a layered solution of CH2Cl2 and 1 pentane at -35 °C. Yield: 0.142 g (74 %). H NMR (CD2Cl2): δ 3.98 (s, 3H, OCH3); 1.91 (s, 3 11 1 30H, C5(CH3)5); 1.63 (d, 27H, JH-P = 14 Hz, PtBu). B{ H} NMR (CD2Cl2): δ -16.66 (s). 13 1 1 1 C{ H} NMR (CD2Cl2): δ 148.77 (br d, JC-F = 242 Hz, o-C6F5); 138.85 (br d, JC-F = 244 Hz, p- 1 C6F5); 136.90 (br d, JC-F = 244 Hz, m-C6F5); 124.36 (br s, ipso-C6F5); 122.70 (s, C5(CH3)5); 1 58.92 (s, OCH3); 41.95 (d, JC-P = 31 Hz, PC{CH3}3); 30.04 (s, P{C{CH3}3); 11.27 (s, 15 2 1 C5(CH3)5). N NMR (CD2Cl2): δ 591.01 (dd, JN-P = 16 Hz, JN-N = 17 Hz, PNNO), 386.28 (dd, 1 1 19 3 JN-P = 62 Hz, JN-N = 17 Hz, PNNO). F NMR (CD2Cl2): δ -133.07 (d, 8F, JF-F = 20 Hz, o- 3 31 1 C6F5); -163.73 (t, 4F, JF-F = 20 Hz, p-C6F5); -167.54 (m, 8F, m-C6F5). P{ H} NMR (CD2Cl2): 1 2 δ 64.33 (dd, JP-N = 62 Hz, JP-N = 16 Hz). Anal. Calcd. for C57H60BF20N2O2PZr: C, 51.91; H, 4.59; N, 2.13. Found: C, 51.24; H, 4.75; N, 2.28 %.

Synthesis of 7-4. A 20 mL scintillation vial was charged with

Cp*2Zr(OMe)Me (0.041 g, 0.101 mmol) and toluene (2 mL). A

second 10 mL vial was charged with B(C6F5)3 (0.051 g, 0.100 mmol) and toluene (5 mL). The borane solution was slowly added to the solution of

Cp*2Zr(OMe)Me at r.t. The resulting solution was bright yellow and was allowed to stir for an additional 5 minutes. At this time, 6-6 (0.055 g, 0.101 mmol) was added resulting in an opaque yellow solution. The reaction was stirred for 1 hour at r.t. Pentane was added (10 mL)

precipitating a yellow oil. The oil was taken up in CH2Cl2 (2 mL), filtered through a plug of Celite and triturated with pentane (15 mL). The solvents were then decanted from the bright

239

yellow solid. The product was washed with pentane (3 x 5 mL) and dried in vacuo for 2 hours.

Crystals suitable for X-ray crystallography were grown from a layered solution of CH2Cl2 and 1 pentane at -35 °C. Yield: 0.98 g (86 %). H NMR (CD2Cl2): δ 3.99 (s, 3H, OCH3); 1.91 (s, 3 11 1 30H, C5(CH3)5); 1.62 (d, 27H, JH-P = 14 Hz, PtBu); 0.48 (br s, 3H, H3CB(C6F5)3). B{ H} 13 1 1 NMR (CD2Cl2): δ -14.93 (s). C{ H} NMR (CD2Cl2): δ 148.99 (br d, JC-F = 237 Hz, ortho- 1 1 C6F5); 138.09 (br d, JC-F = 242 Hz, para-C6F5); 136.47 (br d, JC-F = 257 Hz, meta-C6F5); 129.59 1 (br s, ipso-C6F5); 122.74 (s, C5(CH3)5); 58.94 (s, OCH3); 41.99 (d, JC-P = 31 Hz, PC{CH3}3); 15 30.07 (s, PC{CH3}3); 11.29 (s, C5(CH3)5); 10.23 (br s, H3CB(C6F5)3). N NMR (CD2Cl2): δ 2 1 1 1 591.01 (dd, JN-P = 16 Hz, JN-N = 17 Hz, PNNO); 386.28 (dd, JN-P = 62 Hz, JN-N = 17 Hz, 19 3 3 PNNO). F NMR (CD2Cl2): δ -133.10 (d, JF-F = 20 Hz, ortho-C6F5); -165.36 (t, JF-F = 20 Hz, 31 1 1 2 para-C6F5); -167.89 (m, meta-C6F5). P{ H} NMR (CD2Cl2,): δ 64.33 (dd, JP-N = 62 Hz, JP-N =

16 Hz). Anal. Calcd. for C52H63BF15N2O2PZr: C, 53.56; H, 5.45; N, 2.40. Found: C, 53.59; H, 5.60; N, 2.64 %.

Synthesis of 7-5. A 20 mL scintillation vial was charged with 6-

6 (0.100 g, 0.184 mmol) and tol·Zn(C6F5)2 (0.091 g, 0.185

mmol) in CH2Cl2 (5 mL). The solution was initially opaque but cleared after a few seconds of stirring. The reaction was left stirring for 1 hour at room temperature. At this time, the solution was cloudy. Hexanes (10 mL) was added precipitating a fine white solid. The solid was isolated by filtration, washed with hexanes (3 x 5 mL) and dried in vacuo for 2 hours. Crystals suitable for X-ray diffraction were 1 grown from a layered CH2Cl2/pentane solution at -35 °C. Yield: 0.118 g (99 %). H NMR 3 13 1 1 (CD2Cl2): δ 1.51 (d, 27H, JH-P = 14 Hz, PtBu). C{ H} NMR (CD2Cl2): δ 149.30 (br d, JC-F = 1 1 228 Hz, o-C6F5); 140.46 (br d, JC-F = 244 Hz, p-C6F5); 136.90 (br d, JC-F = 259 Hz, m-C6F5); 1 15 120.71 (s, ipso-C6F5); 41.65 (d, JC-P = 31.65, PC{CH3}3); 29.70 (s, PC{CH3}3). N NMR 2 1 1 1 (CD2Cl2): δ 599.07 (dd, JN-P = 9.3 Hz, JN-N = 18 Hz, PNNO); 317.97 (dd, JN-P = 54 Hz, JN-N= 19 3 18 Hz, PNNO). F NMR (CD2Cl2): δ -117.44 (m, o-C6F5); -157.71 (t, JF-F = 19 Hz, p-C6F5); - 31 1 1 2 162.64 (m, m-C6F5). P{ H} NMR (CD2Cl2): 65.83 (dd, JP-N = 54 Hz, JP-N = 9.3 Hz). Anal.

Calcd. for C48H54F20N4O2P2Zn2: C, 44.63; H, 4.21; N, 4.34. Found: C, 44.32; H, 4.14; N, 4.66 %.

240

Synthesis of 7-6. Method I. A 20 mL scintillation vial was

charged with 6-6 (0.060 g, 0.111 mmol) and tol·Zn(C6F5)2

(0.082 g, 0.167 mmol) in CH2Cl2 (10 mL). The clear solution was left stirring for 1 hour at room temperature. At this time, pentane (10 mL) was added precipitating a fine white solid. The product was allowed to settle and the solvent was decanted followed by washing of the solid with pentane (3 x 5 mL). The product was dried in vacuo for 1 hour. Yield: 0.076 g (81 %). Method II. A 20 mL scintillation

vial was charged with 7-5 (0.037 g, 0.029 mmol) and tol·Zn(C6F5)2 (0.014 g, 0.028 mmol) in

CH2Cl2 (10 mL). The clear solution was left stirring for 1 hour at room temperature. At this time, pentane (10 mL) was added precipitating a fine white solid. The product was allowed to settle and the solvent was decanted followed by washing of the solid with pentane (3 x 5 mL). The product was dried in vacuo for 1 hour. Yield: 0.037 g (77 %). Crystals suitable for X-ray 1 diffraction were grown from a layered CH2Cl2/pentane solution at -35 °C. H NMR (CD2Cl2): 3 13 1 1 δ 1.46 (d, 27H, JH-P = 14 Hz, PtBu). C{ H} NMR (CD2Cl2): δ 149.26 (br d, JC-F = 245 Hz, o- 1 1 C6F5); 140.17 (br d, JC-F = 224 Hz, p-C6F5); 136.41 (br d, JC-F = 241 Hz, m-C6F5); 118.95 (s, 1 15 ipso-C6F5); 41.86 (d, JC-P = 30 Hz, PC{CH3}3); 29.59 ( s, PC{CH3}3). N NMR (CD2Cl2): δ 2 1 1 1 595.17 (dd, JN-P = 9.4 Hz, JN-N = 18 Hz, PNNO); 323.78(dd, JN-P = 54 Hz, JN-N = 18 Hz, 19 3 PNNO). F NMR (CD2Cl2, 25 °): δ -117.56 (m, o-C6F5); -156.73 (t, JF-F = 19 Hz, p-C6F5); - 19 162.42 (m, m-C6F5). F NMR (CD2Cl2, -75 °): δ -117.43 (br s, o-C6F5); -117.93 (br s, o-C6F5); - 3 3 156.26 (t, JF-F = 19 Hz, p-C6F5); -157.65 (t, JF-F = 18 Hz, p-C6F5); -161.81 (br s, m-C6F5); - 31 1 1 2 162.30 (br s, m-C6F5). P{ H} NMR (CD2Cl2): δ 68.22 (dd, JP-N = 54 Hz, JP-N = 9.4 Hz). Anal.

Calcd. for C60H54F30N4O2P2Zn3·CH2Cl2: C, 41.25; H, 3.18; N, 3.15. Found: C, 41.73; H, 3.39; N, 3.49 %.

Synthesis of 7-7. Method I. A 20 mL scintillation vial was charged

with 6-6 (0.064 g, 0.118 mmol) and tol·Zn(C6F5)2 (0.116 g, 0.236

mmol) in CH2Cl2 (10 mL). The clear solution was left stirring for 1 hour at room temperature. At this time, pentane (10 mL) was added precipitating a fine white solid. The product was allowed to settle and the solvent was decanted followed by washing with pentane (3 x 5 mL). The solid was dried in vacuo for 1 hour. Yield: 0.099 g, (80 %). Method

II. A 20 mL scintillation vial was charged with 7-5 (0.060 g, 0.046 mmol) and tol·Zn(C6F5)2

(0.046 g, 0.094 mmol) in CH2Cl2 (8 mL). The cloudy, faintly yellow solution was left stirring for

241

1 hour at room temperature. At this time, hexanes (10 mL) was added precipitating a fine white powder. The solid was allowed to settle and the solvent was decanted followed by washing with hexanes (3 x 5 mL). The solid was dried in vacuo for 1 hour. Yield: 0.068 g (70 %). Crystals

suitable for X-ray diffraction were grown from a layered CH2Cl2/cyclohexane solution at 25 °C. 1 3 13 1 H NMR (CD2Cl2): δ 1.43 (d, 27H, JH-P = 14 Hz, PtBu). C{ H} NMR (CD2Cl2): δ 149.41 (br 1 1 d, JC-F = 224 Hz, o-C6F5); 140.97 (br d, JC-F = 248 Hz, p-C6F5); 137.00 (br d, 262 Hz, m-C6F5); 1 15 118.29 (s, ipso-C6F5); 42.04 (d, JC-P = 30 Hz, PC{CH3}3); 29.53 ( s, PC{CH3}3). N NMR 2 1 1 1 (CD2Cl2): δ 582.52 (dd, JN-P = 11 Hz, JN-N = 17 Hz, PNNO); 349.33 (dd, JN-P = 54 Hz, JN-N = 19 3 17 Hz, PNNO). F NMR (CD2Cl2, 25 °C): δ -117.62 (m, o-C6F5); -156.26 (t, JF-F = 20 Hz, p- 19 3 C6F5); -162.18 (m, m-C6F5). F NMR (CD2Cl2, -75 °C): δ -117.61 (d, JF-F = 20 Hz, o-C6F5); - 3 3 3 118.15 (d, JF-F = 20 Hz, o-C6F5); -154.85 (t, JF-F = 20 Hz, p-C6F5); -155.36 (t, JF-F = 21 Hz, p- 3 31 1 1 C6F5); -161.10 (t, JF-F = 19 Hz, 2 x m-C6F5). P{ H} NMR (CD2Cl2): δ 71.99 (dd, JP-N = 54 2 Hz, JP-N = 11 Hz). Anal. Calcd. for C36H27F20N2OPZn2·0.5CH2Cl2: C, 40.30; H, 2.59; N, 2.57. Found: C, 40.05; H, 2.80; N, 2.78 %.

7.4.3 X-ray Crystallography

7.4.3.1 X-ray Data Collection and Reduction

In preparation for analysis, crystals were first coated with Paratone-N oil in a glovebox and were

subsequently mounted on a MiTegen Micromount and placed in a N2 stream in order to maintain a dry and oxygen-free sample environment. Data collection was performed on a Bruker Apex II diffractometer and data collection strategies were determined employing Bruker provided Apex software. Optimization was performed in order to yield >99.5 % complete data to a minimum 2θ value of 55 °. All data sets were collected at 150(±2) K unless otherwise stated. The acquired frames were integrated employing the Bruker SAINT software package employing a narrow frame algorithm. Absorption corrections were conducted employing the empirical multi-scan method (SADABS).

7.4.3.2 Structure and Refinement

242

successive difference Fourier map calculations. All cycles of refinement were carried out 2 employing full-matrix least squares techniques on F, minimizing the function ω (Fo-Fc) where 2 2 weight (ω) equates to 4Fo /2σ (Fo ) and Fo and Fc are equal to the observed and the calculated structure factor amplitudes. In the final cycles of each refinement, all non-hydrogen atoms were assigned anisotropic temperature factors in the absence of disorder or insufficient data. In the instance of the latter case the atoms were then treated isotropically. Unless noted, the C-H atom positions were calculated and allowed to ride on the carbon to which they are bonded with an assumed C-H bond length of 0.95 Å. Temperature factors pertaining to the H-atoms were fixed at 1.20 times the isotropic temperature factor of the C-atom to which they are bound. The H- atom contributions were calculated however never refined. The locations of the largest peaks in the final difference Fourier map calculations as well as the magnitude of the residual electron densities in each case were of no chemical significance.

243

7.4.3.3 Tables of Crystallographic Data

Table 7.1 – Selected crystallographic data for 7-1, 7-3 and 7-4.

7-1 7-3 7-4

Formula C42H42BF15N2OPZr C57H60BF20N2O2PZr C24H27F10N2OPZn Formula wt 1009.78 1318.07 645.82 Crystal system Monoclinic Monoclinic Triclinic

Space group P21/n C2/c P-1 a (Å) 15.6655(7) 36.218(7) 10.6427(5) b (Å) 12.5681(6) 12.304(3) 11.8120(5) c (Å) 21.5094(10) 31.224(10) 11.9009(5) α (deg) 90 90 94.636(2) β (deg) 93.812(2) 122.397(5) 105.821(2) γ (deg) 90 90 108.515(2) V (Å3) 4225.5(3) 11749(5) 1341.73(10) Z 8 8 2 T (K) 159(2) 159(2) 150(2) d (calc) gcm-3 1.587 1.490 1.599 Abs coeff, μ, mm-1 0.401 0.321 1.066 Data collected 35224 13566 20009 R int 0.0551 0.0000 0.0335 # of indpndt reflns 9538 13566 6118

Reflns Fo≥2.0 σ (Fo) 6004 7588 5269 Variables 568 760 361 R (>2σ) 0.0576 0.0587 0.0454

wR2 0.1490 0.1425 0.1150 Goodness of fit 1.037 0.997 1.113

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

244

Table 7.2 – Selected crystallographic data for 7-5 and 7-6.

7-5 7-6

Formula C60H54F30N4O2P2Zn3 C36.5H28ClF20N2OPZn2 Formula wt 1691.12 1087.77 Crystal system Monoclinic Triclinic Space group C2/c P-1 a (Å) 24.1607(10) 11.7815(6) b (Å) 15.3120(6) 12.0554(6) c (Å) 22.3763(13) 15.7080(8) α (deg) 90 95.803(2) β (deg) 117.136(2) 111.223(2) γ (deg) 90 101.752(2) V (Å3) 7366.9(6) 1998.40(17) Z 4 2 T (K) 150(2) 150(2) d (calc) gcm-3 1.525 1.808 Abs coeff, μ, mm-1 1.130 1.436 Data collected 31355 71980 R int 0.0521 0.0335 # of indpndt reflns 8478 19154

Reflns Fo≥2.0 σ (Fo) 4988 13414 Variables 456 601 R (>2σ) 0.0448 0.0350 wR2 0.0986 0.0865 Goodness of fit 0.962 1.044

Data acquired employing Mo Kα radiation (λ = 0.71069 Å).

245

Chapter 8 Future Work and Summary

8.1 Future Work

The work presented in the aforementioned chapters has the potential to be further elaborated upon and applied in fields other than those presented within the confines of this thesis.

The protocol for the derivatization of sterically encumbered electrophilic boranes can be extended to create a range of new Lewis acids from commercially available starting materials which have the potential to act as excellent co-catalysts for the activation of metallocene pre- catalysts for application in the field of olefin polymerization. Methide extraction from a metallocene complex by a derivatized bulky borane could result in large non-coordinating anions, likely demonstrating minimal ion-pairing, resulting in highly effective polymerization catalysts.

Furthermore, the Lewis acids, as derived from the reaction of diazomethanes with electrophilic boranes, have the potential to be applied in the field of asymmetric hydrogenation. Currently, little progress has been made toward the catalytic asymmetric hydrogenation of unsaturated substrates employing frustrated Lewis pair systems. Some efforts have focused on the application of frustrated Lewis pair systems composed of commercially available chiral phosphines ((R)-BINAP, (S,S)-CHIRAPHOS and (S,S)-DIOP) in attempts to induce chirality in the product molecules.44 Such investigations have proven to be vastly unsuccessful yielding racemic product mixtures or only moderate enantiomeric excesses (~25 %, employing (S,S)- DIOP).44 Chen and Klankermayer studied the application of chiral Lewis acids, employing

boranes such as (α-pinenyl)B(C6F5)2 and two camphor derivatives toward the asymmetric hydrogenation of imines.30 These catalysts saw various degrees of success with respect to the hydrogenation of prochiral ketimines yielding amines with enantiomeric excesses ranging from 29 13 ((α-pinenyl)B(C6F5)2) to 85 % (camphor derivatives). The method of borane derivatization employing diazomethanes allows for the facile synthesis of chiral boranes by reaction of dissymmetric diazomethanes with symmetric triarylboranes or by reaction of dissymmetric triarylboranes with symmetric diazomethanes. Enantiomerically pure catalysts could potentially

246

be isolated by diastereomer separation following complexation with a suitable chiral base. These catalysts not only offer the possibility for application in the catalytic asymmetric hydrogenation of prochiral substrates but also offer the possibility for application in the selective carbon-carbon cross-coupling reactions.157-159 Additionally, the sterically encumbered nature of the Lewis acids protects the boron center making the borane potentially more tolerant toward donor molecules such as ketones and aldehydes which are commonly found to be catalyst poisons.1

The derivatized boranes also have the potential to be applied in the synthesis of boron-containing polymers which find applications in: luminescent materials, lithium ion batteries, chemical sensors, medical imaging, optical and electronic devices and so forth.419,420 The pioneering concept of boron integration into the principal chain of non-conjugated polymers was first accomplished by Chujo in the early 1990s.421,422 Since that time, the catalogue of boron containing polymers has grown considerably and these polymers are now found to incorporate a range of boron moieties such as boronic acids, carboranes, ionic borates and triarylboranes.417,418 These polymers have recently drawn considerable attention due to their ease of preparation, wide tunability, plethora of intriguing chemical properties and good stability.417,418 The ease of borane derivatization employing diazomethane reagents makes these Lewis acid potentially attractive synthons for application in the synthesis of boron-containing polymers.

Investigations of the stable CO2 adduct, tBu3P(CO2)B(C6F5)2Cl has recently shown application in

Lewis acid exchange chemistry. This chemistry implies that the tBu3P(CO2) moiety is stable toward the loss of CO2 and is likely a suitable candidate for the subsequent functionalization of

the complexed CO2 moiety. This synthon should further be studied for the reduction of the

complexed CO2 molecule to formate containing species or methanol.

8.2 Summary

Data presented in this thesis demonstrate that small molecule activation is no longer the domain of purely transition metal-based complexes but can be effected equally well employing main group compounds. The unquenched Lewis acidity and basicity in a sterically encumbered frustrated Lewis pair can be harnessed for the unique activation of a series of small molecules. These activations can be achieved by a plethora of Lewis acid and base combinations and are not

limited to the use of the sterically demanding perfluoroborane B(C6F5)3.

247

The activation of O−H and S−H bonds was investigated employing a variety of frustrated Lewis pair systems. Such combinations resulted in the formation of readily isolable phosphonium alkoxy- and thioborate salts. These salts were shown to be thermally robust and to resist the

liberation of H2 and group transfer to unsaturated substrates. The inherent stability of these compounds and reluctance to undergo further reactivity was attributed to the strength of the respective S−B and O−B bonds of the anion.

In FLP-mediated catalytic hydrogenations, the strength of the product-boron bond is typically the limiting factor in product liberation from the catalyst, resulting in sluggish reaction rates or catalyst poisoning. Therefore, boranes with reduced Lewis acidities, yet maintained steric bulk,

warranted investigation. Borate and boronate esters were readily synthesized by reaction of BCl3 or C6F5B(OH)2 with an appropriate alcohol or diol. Borate esters of the form B(OR)3 were found to initiate the heterolytic cleavage of H2 in the presence of a tertiary phosphine and readily

underwent redistribution reactions yielding phosphonium borate salts, [R3PH][B(OR)4]. Boronate esters were, however, found to be inactive under analogous reaction conditions. The Lewis acidities of the newly synthesized boranes were quantified employing both The Gutmann- Beckett and the Childs Lewis acidity determination protocols.

As methods for borane synthesis are frequently complex, requiring numerous synthetic steps, or the use of toxic intermediates, new methods for simple and straightforward derivatization of boranes were sought. Reactions of perfluoroarylboranes such as B(C6F5)3, PhB(C6F5)3 or BPh3 with an alkyldiazomethane resulted in clean insertion of the Me3SiCH moiety into one or two B−C bonds of the respective borane. Similarly, reaction of chlorodiarylboranes with

diphenyldiazomethane resulted in the insertion of the Ph2C fragment, selectively into a B-C bond of the borane. Reaction of boronic acids with diphenyldiazomethane yielded products of B−C bond insertion followed by dehydration to give cyclic boroxines. Reaction of HB(C6F5)2 with diphenyldiazomethane resulted in divergent reactivity where the intact diazomethane molecule was trapped by the borane yielding an amino-borane as the product. Derivatization of a number of classes of boranes employing alkyl- and aryldiazomethanes was readily achieved yielding novel boranes in high yield. These boranes have been subsequently applied in the reversible activation of H2 and imine hydrogenation.

248

In recent years, frustrated Lewis pairs have found application in the activation of the greenhouse gas carbon dioxide. Data presented in this thesis demonstrate the ability to employ boranes of

varying Lewis acidities for the sequestration of CO2, generating stable adducts. The zwitterion

tBu3P(CO2)B(C6F5)2Cl was found to be stable in nature and to readily undergo exchange of the tBu3P(CO2) fragment upon reaction with Lewis acids of superior strength. Exchange employing

the metallocenium cation [Cp2TiMe][B(C6F5)4] with tBu3P(CO2)B(C6F5)2Cl resulted in the facile

and high-yielding synthesis of a transition metal and phosphine stabilized CO2 adduct.

Similarly, reaction of the perfluoroarylalane Al(C6F5)3 with tBu3P(CO2)B(C6F5)2Cl resulted in

the isolation of a 1:1:1 phosphine:CO2:alane adduct, as the product of Lewis acid exchange. This method of Lewis acid exchange offered access to compounds which were otherwise inaccessible by more traditional synthetic methods.

Lastly, the reaction of main group frustrated Lewis pairs with nitrous oxide resulted in the collaborative activation of the greenhouse gas to yield zwitterionic compounds of the form

R3P(N2O)BAr3. These adducts were found to be robust and stable and resisted decomposition for prolonged periods of time. Exchange of the R3P(N2O) fragment was readily achieved upon reaction with a Lewis acid of superior strength such as the trityl or a metallocenium cation.

Reaction with Zn(C6F5)2 with R3P(N2O)BAr3 resulted in the synthesis of novel phosphine and

zinc stabilized N2O complexes, demonstrating a variety of unique bridging binding modes.

In summary, the aforementioned results demonstrate the ability of frustrated Lewis pairs to collectively activate a series of small molecules yielding novel main group compounds. The scope of Lewis acids and bases which are active in small molecule complexation is already quite broad and continues to expand at an astonishing rate. The results presented herein validate the applicability of main group systems to the field of small molecule activation, which prior to very recently, was deemed to be typically achievable in the presence transition metal complexes.

249

References

(1) Stephan, D. W.; Erker, G. Angew. Chem. Int. Ed. 2010, 49, 46.

(2) Lewis, G. N. Valence and the Structure of Atoms and Molecules; Chemical Catalogue Company: New York, 1923.

(3) Stephan, D. W. Org. Biomol. Chem. 2008, 6, 1535.

(4) Stephan, D. W. Dalton Trans. 2009, 3129.

(5) Erker, G. C. R. Chimie 2011, 14, 831.

(6) Stephan, D. W. Org. Biomol. Chem. 2012.

(7) Brown, H. C.; Schlesinger, H. I.; Cardon, S. Z. J. Am. Chem. Soc. 1942, 64, 325.

(8) Wittig, G.; Benz, E. Chem. Ber. Recl. 1959, 92, 1999.

(9) Tochtermann, W. Angew. Chem. Int. Ed. 1966, 5, 351.

(10) Massey, A. G.; Stone, F. G. A.; Park, A. J. Proc. Chem. Soc. London 1963, 212.

(11) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1964, 2, 245.

(12) Piers, W. E. Adv. Organomet. Chem. 2005, 52, 1.

(13) Bradley, D. C.; Harding, I. S.; Keefe, A. D.; Motevalli, M.; Zheng, D. H. J. Chem. Soc., Dalton Trans. 1996, 3931.

(14) Denis, J. M.; Forintos, H.; Szelke, H.; Toupet, L.; Pham, T. N.; Madec, P. J.; Gaumont, A. C. Chem. Commun. 2003, 54.

(15) Lucarini, M.; Pedulli, G. F. J. Organomet. Chem. 1995, 494, 123.

(16) Su, Z. M.; Wang, X. J.; Huang, Z. H.; Wang, R. S.; Feng, J. K.; Sun, J. Z. Synth. Met. 2001, 119, 583.

(17) Yamaguchi, S.; Akiyama, S.; Tamao, K. J. Organomet. Chem. 2002, 652, 3.

(18) Blackwell, J. M.; Piers, W. E.; Parvez, M.; McDonald, R. Organometallics 2002, 21, 1400.

(19) Guidotti, S.; Camurati, I.; Focante, F.; Angellini, L.; Moscardi, G.; Resconi, L.; Leardini, R.; Nanni, D.; Mercandelli, P.; Sironi, A.; Beringhelli, T.; Maggioni, D. J. Org. Chem. 2003, 68, 5445.

(20) Lesley, M. J. G.; Woodward, A.; Taylor, N. J.; Marder, T. B.; Thornton, A.; Bruce, D. W.; Kakkar, A. K. Chem. Mater. 1998, 10, 1355.

250

(21) Cabrera, L.; Welch, G. C.; Masuda, J. D.; Wei, P. R.; Stephan, D. W. Inorg. Chim. Acta. 2006, 359, 3066.

(22) Tolman, C. A. Chem. Rev. 1977, 77, 313.

(23) Bidan, G.; Genies, M. Tetrahedron Lett. 1978, 2499.

(24) Welch, G. C.; Juan, R. R. S.; Masuda, J. D.; Stephan, D. W. Science 2006, 314, 1124.

(25) Welch, G. C.; Cabrera, L.; Chase, P. A.; Hollink, E.; Masuda, J. D.; Wei, P. R.; Stephan, D. W. Dalton Trans. 2007, 3407.

(26) Spies, P.; Erker, G.; Kehr, G.; Bergander, K.; Fröhlich, R.; Grimme, S.; Stephan, D. W. Chem. Commun. 2007, 5072.

(27) Sumerin, V.; Schulz, F.; Atsumi, M.; Wang, C.; Nieger, M.; Leskela, M.; Repo, T.; Pyykko, P.; Rieger, B. J. Am. Chem. Soc. 2008, 130, 14117.

(28) Sumerin, V.; Schulz, F.; Nieger, M.; Atsumi, M.; Wang, C.; Leskela, M.; Pyykko, P.; Repo, T.; Rieger, B. J. Organomet. Chem. 2009, 694, 2654.

(29) Chen, D. J.; Wang, Y. T.; Klankermayer, J. Angew. Chem. Int. Ed. 2010, 49, 9475.

(30) Chen, D. J.; Klankermayer, J. Chem. Commun. 2008, 2130.

(31) Welch, G. C.; Stephan, D. W. J. Am. Chem. Soc. 2007, 129, 1880.

(32) Jursic, B. S. J. Mol. Struc. -Theochem 1999, 492, 97.

(33) Watts, J. D.; Bartlett, R. J. J. Am. Chem. Soc. 1995, 117, 825.

(34) Schreiner, P. R.; Schaefer, H. F.; Schleyer, P. V. J. Chem. Phys. 1994, 101, 7625.

(35) Tague, T. J.; Andrews, L. J. Am. Chem. Soc. 1994, 116, 4970.

(36) Moroz, A.; Sweany, R. L.; Whittenburg, S. L. J. Phys. Chem. 1990, 94, 1352.

(37) Moroz, A.; Sweany, R. L. Inorg. Chem. 1992, 31, 5236.

(38) Rokob, T. A.; Hamza, A.; Stirling, A.; Soós, T.; Papái, I. Angew. Chem. Int. Ed. 2008, 47, 2435.

(39) Rokob, T. A.; Hamza, A.; Stirling, A.; Pápai, I. J. Am. Chem. Soc. 2009, 131, 2029.

(40) Rokob, T. A.; Hamza, A.; Stirling, A.; Soós, T.; Pápai, I. Angew. Chem. Int. Ed. 2008, 47, 2435.

(41) Stirling, A.; Hamza, A.; Rokob, T. A.; Pápai, I. Chem. Commun. 2008, 3148.

(42) Grimme, S.; Kruse, H.; Goerigk, L.; Erker, G. Angew. Chem. Int. Ed. 2010, 49, 1402.

251

(43) Chase, P. A.; Welch, G. C.; Jurca, T.; Stephan, D. W. Angew. Chem. Int. Ed. 2007, 46, 8050.

(44) Stephan, D. W.; Greenberg, S.; Graham, T. W.; Chase, P.; Hastie, J. J.; Geier, S. J.; Farrell, J. M.; Brown, C. C.; Heiden, Z. M.; Welch, G. C.; Ullrich, M. Inorg. Chem. 2011, 50, 12338.

(45) Chase, P. A.; Jurca, T.; Stephan, D. W. Chem. Commun. 2008, 1701.

(46) Mahdi, T.; Heiden, Z. M.; Grimme, S.; Stephan, D. W. J. Am. Chem. Soc. 2012, 134, 4088.

(47) McCahill, J. S. J.; Welch, G. C.; Stephan, D. W. Angew. Chem. Int. Ed. 2007, 46, 4968.

(48) Voss, T.; Chen, C.; Kehr, G.; Nauha, E.; Erker, G.; Stephan, D. W. Chem. -Eur. J. 2010, 16, 3005.

(49) Mömming, C. M.; Fromel, S.; Kehr, G.; Fröhlich, R.; Grimme, S.; Erker, G. J. Am. Chem. Soc. 2009, 131, 12280.

(50) Mömming, C. M.; Kehr, G.; Fröhlich, R.; Erker, G. Chem. Commun. 2011, 47, 2006.

(51) Rosorius, C.; Kehr, G.; Fröhlich, R.; Grimme, S.; Erker, G. Organometallics 2011, 30, 4211.

(52) Ullrich, M.; Seto, K. S. H.; Lough, A. J.; Stephan, D. W. Chem. Commun. 2009, 2335.

(53) Guo, Y.; Li, S. H. Eur. J. Inorg. Chem. 2008, 2501.

(54) Zhao, X.; Stephan, D. W. J. Am. Chem. Soc. 2011, 133, 12448.

(55) Wittig, G.; Ruckert, A. Liebigs Ann. Chem. 1950, 566, 101.

(56) Breen, T. L.; Stephan, D. W. Inorg. Chem. 1992, 31, 4019.

(57) Welch, G. C.; Masuda, J. D.; Stephan, D. W. Inorg. Chem. 2006, 45, 478.

(58) Campbell, J. P.; Gladfelter, W. L. Inorg. Chem. 1997, 36, 4094.

(59) Avens, L. R.; Barnhart, D. M.; Burns, C. J.; McKee, S. D. Inorg. Chem. 1996, 35, 537.

(60) Campello, M. P. C.; Domingos, A.; Santos, I. J. Organomet. Chem. 1994, 484, 37.

(61) Evans, W. J.; Leman, J. T.; Ziller, J. W.; Khan, S. I. Inorg. Chem. 1996, 35, 4283.

(62) Mommertz, A.; Leo, R.; Massa, W.; Harms, K.; Dehnicke, K. Z. Anorg. Allg. Chem. 1998, 624, 1647.

(63) Polamo, M.; Mutikainen, I.; Leskela, M. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1997, 53, 1036.

252

(64) Guo, Z. Y.; Bradley, P. K.; Jordan, R. F. Organometallics 1992, 11, 2690.

(65) GomezSaso, M.; Mullica, D. F.; Sappenfield, E.; Stone, F. G. A. Polyhedron 1996, 15, 793.

(66) Chivers, T.; Schatte, G. Eur. J. Inorg. Chem. 2003, 3314.

(67) Kunnari, S. M.; Oilunkaniemi, R.; Laitinen, R. S.; Ahlgren, M. J. Chem. Soc., Dalton Trans. 2001, 3417.

(68) Birkmann, B.; Voss, T.; Geier, S. J.; Ullrich, M.; Kehr, G.; Erker, G.; Stephan, D. W. Organometallics 2010, 29, 5310.

(69) Kreitner, C.; Geier, S. J.; Stanlake, L. J. E.; Caputo, C. B.; Stephan, D. W. Dalton Trans. 2011, 40, 6771.

(70) Dureen, M. A.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 8396.

(71) Dureen, M. A.; Brown, C. C.; Stephan, D. W. Organometallics 2010, 29, 6594.

(72) Dureen, M. A.; Brown, C. C.; Stephan, D. W. Organometallics 2010, 29, 6422.

(73) Appelt, C.; Westenberg, H.; Bertini, F.; Ehlers, A. W.; Slootweg, J. C.; Lammertsma, K.; Uhl, W. Angew. Chem. Int. Ed. 2011, 50, 3925.

(74) Chen, C.; Eweiner, F.; Wibbeling, B.; Fröhlich, R.; Senda, S.; Ohki, Y.; Tatsumi, K.; Grimme, S.; Kehr, G.; Erker, G. Chem. -Asian J. 2010, 5, 2199.

(75) Liedtke, R.; Fröhlich, R.; Kehr, G.; Erker, G. Organometallics 2011, 30, 5222.

(76) Caputo, C. B.; Geier, S. J.; Ouyang, E. Y.; Kreitner, C.; Stephan, D. W. Dalton Trans. 2012, 41, 237.

(77) Caputo, C. B.; Stephan, D. W. Organometallics 2012, 31, 27.

(78) Johnson, S. A.; Hatnean, J. A.; Doster, M. E. Progress in Inorganic Chemistry; John Wiley & Sons: Hoboken, New Jersey, 2012.

(79) Alcarazo, M.; Gomez, C.; Holle, S.; Goddard, R. Angew. Chem. Int. Ed. 2010, 49, 5788.

(80) Dureen, M. A.; Welch, G. C.; Gilbert, T. M.; Stephan, D. W. Inorg. Chem. 2009, 48, 9910.

(81) Ines, B.; Holle, S.; Goddard, R.; Alcarazo, M. Angew. Chem. Int. Ed. 2010, 49, 8389.

(82) Larock, R. C. Comprehensive Organic Transformations; 2nd ed.; Wiley: New York, 1999.

(83) Tojo, G.; Fernandez, M. Oxidation of Alcohols to Aldehydes and Ketones; A Guide to Current Common Practice; Springer: New York, 2006.

253

(84) Li, X. Q.; Wang, W. K.; Zhang, C. Adv. Synth. Catal. 2009, 351, 2342.

(85) Zhao, M. Z.; Li, J.; Mano, E.; Song, Z. G.; Tschaen, D. M.; Grabowski, E. J. J.; Reider, P. J. J. Org. Chem. 1999, 64, 2564.

(86) Tohma, H.; Takizawa, S.; Maegawa, T.; Kita, Y. Angew. Chem. Int. Ed. 2000, 39, 1306.

(87) Tohma, H.; Maegawa, T.; Takizawa, S.; Kita, Y. Adv. Synth. Catal. 2002, 344, 328.

(88) Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651.

(89) Clayden, J.; Greeves, N.; Warren, S.; Wothers, P. Organic Chemistry; 1st ed.; Oxford University Press: New York, 2001.

(90) Tidwell, T. T. Synthesis 1990, 857.

(91) Mancuso, A. J.; Brownfain, D. S.; Swern, D. J. Org. Chem. 1979, 44, 4148.

(92) Pfitzner, K. E.; Moffatt, J. G. J. Am. Chem. Soc. 1963, 85, 3027.

(93) Pfitzner, K. E.; Moffatt, J. G. J. Am. Chem. Soc. 1965, 87, 5661.

(94) Pfitzner, K. E.; Moffatt, J. G. J. Am. Chem. Soc. 1965, 87, 5670.

(95) Albright, J. D.; Goldman, L. J. Am. Chem. Soc. 1965, 87, 4214.

(96) Albright, J. D.; Goldman, L. J. Am. Chem. Soc. 1967, 89, 2416.

(97) Onodera, K.; Hirano, S.; Kashimur, N. J. Am. Chem. Soc. 1965, 87, 4651.

(98) Parikh, J. R.; Doering, W. V. E. J. Am. Chem. Soc. 1967, 89, 5505.

(99) Corey, E. J.; Kim, C. U. J. Org. Chem. 1973, 38, 1233.

(100) Corey, E. J.; Kim, C. U. Tetrahedron Lett. 1973, 919.

(101) Liu, H. J.; Nyangulu, J. M. Tetrahedron Lett. 1988, 29, 3167.

(102) Wirth, T. Angew. Chem. Int. Ed. 2001, 40, 2812.

(103) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.

(104) Zhdankin, V. V.; Stang, P. J. Chem. Rev. 2002, 102, 2523.

(105) Kürti, L.; Czakó, B.; Corey, E. J.; Nicolaou, K. C. Strategic Applications of Named Reactions in Organic Synthesis; 1st ed.; Elsevier Publishing Press: New York, 2005.

(106) Degraauw, C. F.; Peters, J. A.; Vanbekkum, H.; Huskens, J. Synthesis 1994, 1007.

(107) Ashby, E. C. Acc. Chem. Res. 1988, 21, 414.

254

(108) Oppenauer, R. V. Recl. Trav. Chim. Pays-Bas 1937, 56, 137.

(109) Vilsmaie.E; Sprugel, W. Liebigs Ann. Chem. 1971, 747, 151.

(110) Corey, E. J.; Kim, C. U. J. Am. Chem. Soc. 1972, 94, 7586.

(111) Johnson, C. R.; Bacon, C. C.; Kingsbury, W. D. Tetrahedron Lett. 1972, 501.

(112) Bowden, K.; Heilbron, I. M.; Jones, E. R. H.; Weedon, B. C. L. J. Chem. Soc. 1946, 39.

(113) Corey, E. J.; Suggs, J. W. Tetrahedron Lett. 1975, 2647.

(114) Nyhlen, J.; Privalov, T. Dalton Trans. 2009, 5780.

(115) Lindqvist, M.; Sarnela, N.; Sumerin, V.; Chernichenko, K.; Leskela, M.; Repo, T. Dalton Trans. 2012, 41, 4310.

(116) Marwitz, A. J. V.; Dutton, J. L.; Mercier, L. G.; Piers, W. E. J. Am. Chem. Soc. 2011, 133, 10026.

(117) Drewitt, M. J.; Niedermann, M.; Kumar, R.; Baird, M. C. Inorg. Chim. Acta. 2002, 335, 43.

(118) Williams, J. H. Acc. Chem. Res. 1993, 26, 593.

(119) Li, H. B.; Li, L. T.; Schwartz, D. J.; Metz, M. V.; Marks, T. J.; Liable-Sands, L.; Rheingold, A. L. J. Am. Chem. Soc. 2005, 127, 14756.

(120) Morrison, D. J.; Piers, W. E. Org. Lett. 2003, 5, 2857.

(121) Kondo, T.; Mitsudo, T. Chem. Rev. 2000, 100, 3205.

(122) Meltzer, A.; Inoue, S.; Prasang, C.; Driess, M. J. Am. Chem. Soc. 2010, 132, 3038.

(123) Schaub, T.; Backes, M.; Plietzsch, O.; Radius, U. Dalton Trans. 2009, 7071.

(124) Cao, Y. J.; Lai, Y. Y.; Cao, H.; Xing, X. N.; Wang, X.; Xiao, W. J. Can. J. Chem. 2006, 84, 1529.

(125) Bondi, A. J. Phys. Chem. 1964, 68, 441.

(126) Deck, P. A.; Beswick, C. L.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 1772.

(127) Cromer, D. T.; Waber, J. T. Int. Tables X-Ray Crystallogr. 1974, 4, 71.

(128) De Vries, J. G.; Elsevier, C. J. The Handbook of Homogeneous Hydrogenation; Wiley VCH: Weinheim, 2007.

(129) Hantzsch, A. Liebigts Ann. Chem. 1882, 215, 1.

255

(130) Smith, M. B.; March, J. Advanced Organic Chemistry; 5th ed.; Wiley-Interscience: New York, 2001.

(131) Stephens, F. H.; Pons, V.; Baker, R. T. Dalton Trans. 2007, 2613.

(132) Marder, T. B. Angew. Chem. Int. Ed. 2007, 46, 8116.

(133) Gutowski, M.; Autrey, T. Chem. World 2006, 3, 44.

(134) von Helmolt, R.; Eberle, U. J. Power Sources 2007, 165, 833.

(135) Webb, J. D.; Laberge, V. S.; Geier, S. J.; Stephan, D. W.; Crudden, C. M. Chem. -Eur. J. 2010, 16, 4895.

(136) Berkessel, A.; Schubert, T. J. S.; Muller, T. N. J. Am. Chem. Soc. 2002, 124, 8693.

(137) Sander, W.; Hubert, R.; Kraka, E.; Grafenstein, J.; Cremer, D. Chem. -Eur. J. 2000, 6, 4567.

(138) Sander, W.; Kotting, C. Chem. -Eur. J. 1999, 5, 24.

(139) Frey, G. D.; Lavallo, V.; Donnadieu, B.; Schoeller, W. W.; Bertrand, G. Science 2007, 316, 439.

(140) Wang, H. D.; Fröhlich, R.; Kehr, G.; Erker, G. Chem. Commun. 2008, 5966.

(141) Ullrich, M.; Lough, A. J.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 52.

(142) Ullrich, M.; Lough, A. J.; Stephan, D. W. Organometallics 2010, 29, 3647.

(143) Chase, P. A.; Stephan, D. W. Angew. Chem. Int. Ed. 2008, 47, 7433.

(144) Holschumacher, D.; Bannenberg, T.; Hrib, C. G.; Jones, P. G.; Tamm, M. Angew. Chem. Int. Ed. 2008, 47, 7428.

(145) Geier, S. J.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 3476.

(146) Geier, S. J.; Gille, A. L.; Gilbert, T. M.; Stephan, D. W. Inorg. Chem. 2009, 48, 10466.

(147) Caputo, C. B.; Geier, S. J.; Winkelhaus, D.; Mitzel, N. W.; Vukotic, V. N.; Loeb, S. J.; Stephan, D. W. Dalton Trans. 2012, 41, 2131.

(148) Eros, G.; Mehdi, H.; Pápai, I.; Rokob, T. A.; Kiraly, P.; Tarkanyi, G.; Soós, T. Angew. Chem. Int. Ed. 2010, 49, 6559.

(149) Neu, R. C.; Ouyang, E. Y.; Geier, S. J.; Zhao, X.; Ramos, A.; Stephan, D. W. Dalton Trans. 2010, 39, 4285.

(150) Harhausen, M.; Fröhlich, R.; Kehr, G.; Erker, G. Organometallics 2012, 10.1021/om201076f.

256

(151) Jiang, C. F.; Blacque, O.; Berke, H. Chem. Commun. 2009, 5518.

(152) Jiang, C. F.; Blacque, O.; Berke, H. Organometallics 2009, 28, 5233.

(153) Jiang, C. F.; Blacque, O.; Fox, T.; Berke, H. Dalton Trans. 2011, 40, 1091.

(154) Britovsek, G. J. P.; Ugolotti, J.; White, A. J. P. Organometallics 2005, 24, 1685.

(155) Wrackmeyer, B. Annual Reports on NMR Spectroscopy; Academic Press: London, UK, 1988; Vol. 20.

(156) Naumann, D.; Butler, H.; Gnann, R. Z. Anorg. Allg. Chem. 1992, 618, 74.

(157) Suzuki, A. J. Organomet. Chem. 1999, 576, 147.

(158) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.

(159) Suzuki, A. Pure Appl. Chem. 1991, 63, 419.

(160) Lee, H. S.; Sun, X.; Yang, X. Q.; McBreen, J. J. Electrochem. Soc. 2002, 149, A1460.

(161) Duggan, P. J.; Tyndall, E. M. J. Chem. Soc., Perkin Trans. 1 2002, 1325.

(162) Zettler, F.; Hausen, H. D.; Hess, H. Acta Crystallogr., Sect. B : Struc. Sci. 1974, B 30, 1876.

(163) Hall, D. G.; Murahashi, S.-I.; de Meijere, A.; Diederich, F.; Evans, P. A.; Beller, M.; Bolm, C.; Berkessel, A.; Groeger, H. Boronic Acids: Preparation, Applications in Organic Synthesis and Medicine; Wiley: Weinheim, 2005.

(164) Metz, M. V.; Schwartz, D. J.; Stern, C. L.; Marks, T. J.; Nickias, P. N. Organometallics 2002, 21, 4159.

(165) Childs, R. F.; Mulholland, D. L.; Nixon, A. Can. J. Chem. 1982, 60, 809.

(166) Fenwick, J. T. F.; Wilson, J. W. Inorg. Chem. 1975, 14, 1602.

(167) Luo, L.; Marks, T. J. Top. Catal. 1999, 7, 97.

(168) Kobayashi, S.; Busujima, T.; Nagayama, S. Chem. -Eur. J. 2000, 6, 3491.

(169) Olah, G. A.; Kobayash.S; Tashiro, M. J. Am. Chem. Soc. 1972, 94, 7448.

(170) Branch, C. S.; Bott, S. G.; Barron, A. R. J. Organomet. Chem. 2003, 666, 23.

(171) Mayer, U.; Gutmann, V.; Gerger, W. Monatsh. Chem. 1975, 106, 1235.

(172) Gutmann, V. Coord. Chem. Rev. 1976, 18, 225.

(173) Beckett, M. A.; Strickland, G. C.; Holland, J. R.; Varma, K. S. Polymer 1996, 37, 4629.

257

(174) Beckett, M. A.; Brassington, D. S.; Coles, S. J.; Hursthouse, M. B. Inorg. Chem. Commun. 2000, 3, 530.

(175) Beckett, M. A.; Brassington, D. S.; Light, M. E.; Hursthouse, M. B. J. Chem. Soc., Dalton Trans. 2001, 1768.

(176) Chase, P. A.; Piers, W. E.; Patrick, B. O. J. Am. Chem. Soc. 2000, 122, 12911.

(177) Doring, S.; Erker, G.; Fröhlich, R.; Meyer, O.; Bergander, K. Organometallics 1998, 17, 2183.

(178) Jacobsen, H.; Berke, H.; Doring, S.; Kehr, G.; Erker, G.; Fröhlich, R.; Meyer, O. Organometallics 1999, 18, 1724.

(179) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533.

(180) Cummings, S. A.; Iimura, M.; Harlan, C. J.; Kwaan, R. J.; Trieu, I. V.; Norton, J. R.; Bridgewater, B. M.; Jäkle, F.; Sundararaman, A.; Tilset, M. Organometallics 2006, 25, 1565.

(181) Malkowsky, I. M.; Fröhlich, R.; Griesbach, U.; Putter, H.; Waldvogel, S. R. Eur. J. Inorg. Chem. 2006, 1690.

(182) Chambers, R. D.; Chivers, T. Proc. Chem. Soc. London 1963, 208.

(183) Piers, W. E.; Chivers, T. Chem. Soc. Rev. 1997, 26, 345.

(184) Erker, G. Dalton Trans. 2005, 1883.

(185) Chivers, T. J. Fluorine Chem. 2002, 115, 1.

(186) Ewen, J. A. E., M. J. Chem. Abstr.,1991, 115, 136998g; 256895t.

(187) Yang, X. M.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1991, 113, 3623.

(188) Erker, G.; Kehr, G.; Fröhlich, R. J. Organomet. Chem. 2005, 690, 6254.

(189) Focante, F.; Mercandelli, P.; Sironi, A.; Resconi, L. Coord. Chem. Rev. 2006, 250, 170.

(190) Piers, W. E.; Sun, Y. M.; Lee, L. W. M. Top. Catal. 1999, 7, 133.

(191) Negishi, E. I.; Takahashi, T. Sci. Synth. 2003, 2, 681.

(192) Takeda, N. Organomet. News 2007, 30.

(193) Matteson, D. S. Sci. Synth. 2004.

(194) Miyaura, N. Sci. Synth. 2004, 7, 677.

(195) Entwistle, C. D.; Marder, T. B. Chem. Mater. 2004, 16, 4574.

258

(196) Yamaguchi, S.; Wakamiya, A. Pure Appl. Chem. 2006, 78, 1413.

(197) Kobayashi, H.; Sato, N.; Ichikawa, Y.; Miyata, M.; Chujo, Y.; Matsuyama, T. Synth. Met. 2003, 135, 393.

(198) Hudson, Z. M.; Wang, S. N. Acc. Chem. Res. 2009, 42, 1584.

(199) Mercier, L. G.; Piers, W. E.; Parvez, M. Angew. Chem. Int. Ed. 2009, 48, 6108.

(200) Chai, J. F.; Wang, C.; Jia, L.; Pang, Y.; Graham, M.; Cheng, S. Z. D. Synth. Met. 2009, 159, 1443.

(201) Reitzenstein, D.; Lambert, C. Macromolecules 2009, 42, 773.

(202) Weber, L.; Werner, V.; Fox, M. A.; Marder, T. B.; Schwedler, S.; Brockhinke, A.; Stammler, H. G.; Neumann, B. Dalton Trans. 2009, 1339.

(203) Hoven, C. V.; Wang, H. P.; Elbing, M.; Garner, L.; Winkelhaus, D.; Bazan, G. C. Nat. Mater. 2010, 9, 249.

(204) Cataldo, S.; Fabiano, S.; Ferrante, F.; Previti, F.; Patane, S.; Pignataro, B. Macromol. Rapid Commun. 2010, 31, 1281.

(205) Pron, A.; Baumgarten, M.; Mullen, K. Org. Lett. 2010, 12, 4236.

(206) Caruso, A.; Siegler, M. A.; Tovar, J. D. Angew. Chem. Int. Ed. 2010, 49, 4213.

(207) Stephan, D. W. Chem. Commun. 2010, 46, 8526.

(208) Ishihara, K.; Yamamoto, H. Eur. J. Org. Chem. 1999, 527.

(209) Brook, M. A.; Grande, J. B.; Ganachaus, F. Adv. Polym. Sci. 2010, 235, 161.

(210) Parks, D. J.; Piers, W. E.; Yap, G. P. A. Organometallics 1998, 17, 5492.

(211) Galsworthy, J. R.; Green, M. L. H.; Williams, V. C.; Chernega, A. N. Polyhedron 1998, 17, 119.

(212) Kehr, G.; Fröhlich, R.; Wibbeling, B.; Erker, G. Chem. -Eur. J. 2000, 6, 258.

(213) Parks, D. J.; Spence, R. E. V. H.; Piers, W. E. Angew. Chem. Int. Ed. 1995, 34, 809.

(214) Ghesner, I.; Piers, W. E.; Parvez, M.; McDonald, R. Organometallics 2004, 23, 3085.

(215) Romero, P. E.; Piers, W. E.; Decker, S. A.; Chau, D.; Woo, T. K.; Parvez, M. Organometallics 2003, 22, 1266.

(216) Emslie, D. J. H.; Piers, W. E.; Parvez, M. Angew. Chem. Int. Ed. 2003, 42, 1252.

(217) Roesler, R.; Piers, W. E.; Parvez, M. J. Organomet. Chem. 2003, 680, 218.

259

(218) Chase, P. A.; Henderson, L. D.; Piers, W. E.; Parvez, M.; Clegg, W.; Elsegood, M. R. J. Organometallics 2006, 25, 349.

(219) Kohler, K.; Piers, W. E. Can. J. Chem. 1998, 76, 1249.

(220) Morrison, D. J.; Piers, W. E.; Parvez, M. Synlett 2004, 2429.

(221) Roesler, R.; Har, B. J. N.; Piers, W. E. Organometallics 2002, 21, 4300.

(222) Williams, V. C.; Dai, C. Y.; Li, Z. M.; Collins, S.; Piers, W. E.; Clegg, W.; Elsegood, M. R. J.; Marder, T. B. Angew. Chem. Int. Ed. 1999, 38, 3695.

(223) Williams, V. C.; Piers, W. E.; Clegg, W.; Elsegood, M. R. J.; Collins, S.; Marder, T. B. J. Am. Chem. Soc. 1999, 121, 3244.

(224) Chen, Y. X.; Metz, M. V.; Li, L. T.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 6287.

(225) Chen, Y. X.; Stern, C. L.; Yang, S. T.; Marks, T. J. J. Am. Chem. Soc. 1996, 118, 12451.

(226) Li, L. T.; Stern, C. L.; Marks, T. J. Organometallics 2000, 19, 3332.

(227) Li, H. B.; Li, L. T.; Marks, T. J.; Liable-Sands, L.; Rheingold, A. L. J. Am. Chem. Soc. 2003, 125, 10788.

(228) Li, L. T.; Marks, T. J. Organometallics 1998, 17, 3996.

(229) Metz, M. V.; Schwartz, D. J.; Stern, C. L.; Nickias, P. N.; Marks, T. J. Angew. Chem. Int. Ed. 2000, 39, 1312.

(230) Piers, W. E.; Irvine, G. J.; Williams, V. C. Eur. J. Inorg. Chem. 2000, 2131.

(231) Chen, E. Y. X.; Marks, T. J. Chem. Rev. 2000, 100, 1391.

(232) Dierker, G.; Ugolotti, J.; Kehr, G.; Fröhlich, R.; Erker, G. Adv. Synth. Catal. 2009, 351, 1080.

(233) Chen, C.; Kehr, G.; Fröhlich, R.; Erker, G. J. Am. Chem. Soc. 2010, 132, 13594.

(234) Chen, C.; Fröhlich, R.; Kehr, G.; Erker, G. Chem. Commun. 2010, 46, 3580.

(235) Chen, C.; Voss, T.; Fröhlich, R.; Kehr, G.; Erker, G. Org. Lett. 2011, 13, 62.

(236) Ekkert, O.; Kehr, G.; Fröhlich, R.; Erker, G. J. Am. Chem. Soc. 2011, 133, 4610.

(237) Kehr, G.; Erker, G. Chem. Commun. 2012, 48, 1839.

(238) Jiang, C. F.; Blacque, O.; Berke, H. Organometallics 2010, 29, 125.

(239) Stephan, D. W.; Erker, G. unpublished results 2012.

260

(240) Metzler, N.; Denk, M. Chem. Commun. 1996, 2657.

(241) Bawn, C. E. H.; Ledwith, A.; Matthies, P. J. Polym. Sci. 1959, 34, 93.

(242) Brown, H. C.; Midland, M. M.; Levy, A. B. J. Am. Chem. Soc. 1972, 94, 3663.

(243) Burgos, C. H.; Canales, E.; Matos, K.; Soderquist, J. A. J. Am. Chem. Soc. 2005, 127, 8044.

(244) Bai, J.; Shea, K. J. Macromol. Rapid Commun. 2006, 27, 1223.

(245) Zollinger, H. Diazo Chemistry II: Aliphatic, Inorganic and Organometallic Compounds; 1st ed.; Wiley-VCH: Weinheim, 1995.

(246) Aggarwal, V. K.; Alonso, E.; Hynd, G.; Lydon, K. M.; Palmer, M. J.; Porcelloni, M.; Studley, J. R. Angew. Chem. Int. Ed. 2001, 40, 1430.

(247) Admasu, A.; Gudmundsdottir, A. D.; Platz, M. S. J. Phys. Chem. A 1997, 101, 3832.

(248) Javed, M. I.; Brewer, M. Org. Lett. 2007, 9, 1789.

(249) Thomas, S. E. Organic Synthesis: The Roles of Boron and Silicon; Oxford University Press: Oxford, England, 1991.

(250) Chambers, R. D.; Chivers, T. J. Chem. Soc. 1965, 3933.

(251) Piers, W. E.; Spence, R. E. V.; Macgillivray, L. R.; Zaworotko, M. J. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1995, 51, 1688.

(252) Ashby, M. T.; Sheshtawy, N. A. Organometallics 1994, 13, 236.

(253) Martichonok, V.; Jones, J. B. J. Am. Chem. Soc. 1996, 118, 950.

(254) Lappert, M. F. Chem. Rev. 1956, 56, 959.

(255) Wolfrom, M. L.; Solms, J. J. Org. Chem. 1956, 21, 815.

(256) Islam, T. M. B.; Yoshino, K.; Nomura, H.; Mizuno, T.; Sasane, A. Anal. Sci. 2002, 18, 363.

(257) Carrow, B. P.; Hartwig, J. F. J. Am. Chem. Soc. 2011, 133, 2116.

(258) Clegg, W.; Scott, A. J.; Souza, F. E. S.; Marder, T. B. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1999, 55, 1885.

(259) Evans, D. A.; Hoveyda, A. H. J. Org. Chem. 1990, 55, 5190.

(260) Kanth, J. V. B.; Periasamy, M.; Brown, H. C. Org. Process Res. Dev. 2000, 4, 550.

(261) Brown, C.; Cragg, R. H.; Miller, T. J.; Smith, D. O. J. Organomet. Chem. 1981, 217, 139.

261

(262) Brown, C.; Cragg, R. H.; Miller, T. J.; Smith, D. O. J. Organomet. Chem. 1981, 220, C25.

(263) Brown, H. C.; Midland, M. M.; Levy, A. B. J. Am. Chem. Soc. 1972, 94, 2114.

(264) Brown, H. C.; Midland, M. M.; Levy, A. B. J. Am. Chem. Soc. 1973, 95, 2394.

(265) Brown, H. C.; Singaram, B. Pure Appl. Chem. 1987, 59, 879.

(266) Brown, H. C.; Salunkhe, A. M.; Singaram, B. J. Org. Chem. 1991, 56, 1170.

(267) Suzuki, A.; Sono, S.; Itoh, M.; Brown, H. C.; Midland, M. M. J. Am. Chem. Soc. 1971, 93, 4329.

(268) Curtius, T.; Rauterberg, F. J. Prakt. Chem. 1891, 44, 192.

(269) Bethell, D.; Eeles, M. F. J. Chem. Soc., Perkin Trans. 2 1974, 704.

(270) Hensel, H. R. Chem. Ber. Recl. 1955, 88, 527.

(271) Hellerman, L.; Newman, M. D. J. Am. Chem. Soc. 1932, 54, 2859.

(272) Staudinger, H.; Anthes, E.; Pfenninger, F. Ber. Dtsch. Chem. Ges. 1916, 49, 1928.

(273) Crumrine, D. S.; Haberkamp, T. J.; Suther, D. J. J. Org. Chem. 1975, 40, 2274.

(274) Stojiljk.A; Orbovic, N.; Sredojev, S. Tetrahedron 1970, 26, 1101.

(275) Whitlock, H. W. J. Am. Chem. Soc. 1962, 84, 2807.

(276) Bethell, D.; Brown, K. C. J. Chem. Soc., Perkin Trans. 2 1972, 895.

(277) Zarkesh, R. A.; Heyduk, A. F. Organometallics 2009, 28, 6629.

(278) Benetollo, F.; Bertani, R.; Ganis, P.; Pandolfo, L.; Zanotto, L. J. Organomet. Chem. 2001, 629, 201.

(279) Martin, E.; Hughes, D. L.; Lancaster, S. J. Eur. J. Inorg. Chem. 2006, 4037.

(280) Gais, H. J.; Bulow, G.; Raabe, G. J. Am. Chem. Soc. 1993, 115, 7215.

(281) Wang, Y.; Quillian, B.; Wei, P.; Wannere, C. S.; Xie, Y.; King, R. B.; Schaefer, H. F.; Schleyer, P. V.; Robinson, G. H. J. Am. Chem. Soc. 2007, 129, 12412.

(282) Wang, Y. Z.; Xie, Y. M.; Wei, P. R.; King, R. B.; Schaefer, H. F.; Schleyer, P. V.; Robinson, G. H. J. Am. Chem. Soc. 2008, 130, 14970.

(283) Wang, Y. Z.; Xie, Y. M.; Wei, P. R.; King, R. B.; Schaefer, H. F.; Schleyer, P. V.; Robinson, G. H. Science 2008, 321, 1069.

(284) Dyker, C. A.; Bertrand, G. Science 2008, 321, 1050.

262

(285) Lavallo, V.; Canac, Y.; Donnadieu, B.; Schoeller, W. W.; Bertrand, G. Angew. Chem. Int. Ed. 2006, 45, 3488.

(286) Masuda, J. D.; Schoeller, W. W.; Donnadieu, B.; Bertrand, G. J. Am. Chem. Soc. 2007, 129, 14180.

(287) Masuda, J. D.; Schoeller, W. W.; Donnadieu, B.; Bertrand, G. Angew. Chem. Int. Ed. 2007, 46, 7052.

(288) Back, O.; Kuchenbeiser, G.; Donnadieu, B.; Bertrand, G. Angew. Chem. Int. Ed. 2009, 48, 5530.

(289) Caputo, C. B.; Stephan, D. W. Unpublished Results 2012.

(290) Arduengo, A. J.; Krafczyk, R.; Schmutzler, R.; Craig, H. A.; Goerlich, J. R.; Marshall, W. J.; Unverzagt, M. Tetrahedron 1999, 55, 14523.

(291) Meese, C. O. Liebigs Ann. Chem. 1985, 1711.

(292) Seinfeld, J. H. AIChE J. 2011, 57, 3259.

(293) Intergovernmental Panel on Climate Change,Climate Change 2007: The Physical Science Basis; Cambridge University Press: Cambridge, UK, 2007.

(294) Honisch, B.; Hemming, N. G.; Archer, D.; Siddall, M.; McManus, J. F. Science 2009, 324, 1551.

(295) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. J. Org. Chem. 2009, 74, 487.

(296) Hansen, J.; Nazarenko, L.; Ruedy, R.; Sato, M.; Willis, J.; Del Genio, A.; Koch, D.; Lacis, A.; Lo, K.; Menon, S.; Novakov, T.; Perlwitz, J.; Russell, G.; Schmidt, G. A.; Tausnev, N. Science 2005, 308, 1431.

(297) Pires, J. C. M.; Martins, F. G.; Alvim-Ferraz, M. C. M.; Simoes, M. Chem. Eng. Res. Des. 2011, 89, 1446.

(298) Sakakura, T.; Choi, J. C.; Yasuda, H. Chem. Rev. 2007, 107, 2365.

(299) Aresta, M.; Dibenedetto, A. Dalton Trans. 2007, 2975.

(300) Steeneveldt, R.; Berger, B.; Torp, T. A. Chem. Eng. Res. Des. 2006, 84, 739.

(301) Leitner, W. Coord. Chem. Rev. 1996, 153, 257.

(302) Jessop, P. G.; Ikariya, T.; Noyori, R. Chem. Rev. 1995, 95, 259.

(303) Olah, G. A. Angew. Chem. Int. Ed. 2005, 44, 2636.

(304) Vaidhyanathan, R.; Iremonger, S. S.; Dawson, K. W.; Shimizu, G. K. H. Chem. Commun. 2009, 5230.

263

(305) Lee, K. B.; Beaver, M. G.; Caram, H. S.; Sircar, S. Ind. Eng. Chem. Res. 2008, 47, 8048.

(306) Demessence, A.; D'Alessandro, D. M.; Foo, M. L.; Long, J. R. J. Am. Chem. Soc. 2008, 131.

(307) Fletcher, A. J.; Cussen, E. J.; Prior, T. J.; Rosseinsky, M. J.; Kepert, C. J.; Thomas, K. M. J. Am. Chem. Soc. 2001, 123, 10001.

(308) Surble, S.; Millange, F.; Serre, C.; Duren, T.; Latroche, M.; Bourrelly, S.; Llewellyn, P. L.; Ferey, G. J. Am. Chem. Soc. 2006, 128, 14889.

(309) Mömming, C. M.; Otten, E.; Kehr, G.; Fröhlich, R.; Grimme, S.; Stephan, D. W.; Erker, G. Angew. Chem. Int. Ed. 2009, 48, 6643.

(310) Peuser, I.; Neu, R. C.; Zhao, X. X.; Ulrich, M.; Schirmer, B.; Tannert, J. A.; Kehr, G.; Fröhlich, R.; Grimme, S.; Erker, G.; Stephan, D. W. Chem. -Eur. J. 2011, 17, 9640.

(311) Ménard, G.; Stephan, D. W. J. Am. Chem. Soc. 2010, 132, 1796.

(312) Bertini, F.; Lyaskovskyy, V.; Timmer, B. J.; de Kanter, F. J.; Lutz, M.; Ehlers, A. W.; Slootweg, J. C.; Lammertsma, K. J. Am. Chem. Soc. 2012, 134, 201.

(313) Dureen, M. A.; Stephan, D. W. J. Am. Chem. Soc. 2010, 132, 13559.

(314) Zhao, X.; Stephan, D. W. Chem. Commun. 2011, 47, 1833.

(315) Chapman, A. M.; Haddow, M. F.; Wass, D. F. J. Am. Chem. Soc. 2011, 133, 18463.

(316) Chapman, A. M.; Haddow, M. F.; Wass, D. F. Eur. J. Inorg. Chem. 2012.

(317) Ashley, A. E.; Thompson, A. L.; O'Hare, D. Angew. Chem. Int. Ed. 2009, 48, 9839.

(318) Berkefeld, A.; Piers, W. E.; Parvez, M. J. Am. Chem. Soc. 2010, 132, 10660.

(319) Mitu, S.; Baird, M. C. Can. J. Chem. 2006, 84, 225.

(320) Hair, G. S.; Cowley, A. H.; Jones, R. A.; McBurnett, B. G.; Voigt, A. J. Am. Chem. Soc. 1999, 121, 4922.

(321) Erskine, G. J.; Wilson, D. A.; Mccowan, J. D. J. Organomet. Chem. 1976, 114, 119.

(322) Spence, R. E. V.; Piers, W. E.; Sun, Y. M.; Parvez, M.; MacGillivray, L. R.; Zaworotko, M. J. Organometallics 1998, 17, 2459.

(323) Huffman, J. C.; Moloy, K. G.; Marsella, J. A.; Caulton, K. G. J. Am. Chem. Soc. 1980, 102, 3009.

(324) Cadierno, V.; Zablocka, M.; Donnadieu, B.; Igau, A.; Majoral, J. P. Organometallics 1999, 18, 1882.

264

(325) Ravishankara, A. R.; Daniel, J. S.; Portmann, R. W. Science 2009, 326, 123.

(326) Trogler, W. C. Coord. Chem. Rev. 1999, 187, 303.

(327) Hansen, J.; Sato, M. P. Natl. Acad. Sci. USA 2004, 101, 16109.

(328) Richardson, D.; Felgate, H.; Watmough, N.; Thomson, A.; Baggs, E. Trends Biotechnol. 2009, 27, 388.

(329) Snyder, C. S.; Bruulsema, T. W.; Jensen, T. L.; Fixen, P. E. Agric. Ecosyst. Environ. 2009, 133, 247.

(330) 2009 U.S. Greenhouse Gas Inventory Report, E. P. A., (http://tinyurl.com/emissionsreport).

(331) Leont'ev, A. V.; Fomicheva, O. A.; Proskurnina, M. V.; Zefirov, N. S. Russ. Chem. Rev. 2001, 70.

(332) Panov, G. I.; Dubkov, K. A.; Starokon, E. V.; Parmon, V. N. React. Kinet. Catal. Lett. 2002, 76, 401.

(333) Parmon, V. N.; Panov, G. I.; Uriarte, A.; Noskov, A. S. Catal. Today 2005, 100, 115.

(334) Paetzold, P.; Schimmel, G. Z. Naturforsch, B: Chem. Sci. 1980, 35, 568.

(335) Yao, S. L.; Xiong, Y.; Driess, M. Chem. -Eur. J. 2010, 16, 1281.

(336) Sen, S. S.; Tavcar, G.; Roesky, H. W.; Kratzert, D.; Hey, J.; Stalke, D. Organometallics 2010, 29, 2343.

(337) Xiong, Y.; Yao, S. L.; Driess, M. J. Am. Chem. Soc. 2009, 131, 7562.

(338) Yao, S. L.; Xiong, Y.; Brym, M.; Driess, M. J. Am. Chem. Soc. 2007, 129, 7268.

(339) Wiberg, N.; Preiner, G.; Schurz, K. Chem. Ber. Recl. 1988, 121, 1407.

(340) Yokelson, H. B.; Millevolte, A. J.; Gillette, G. R.; West, R. J. Am. Chem. Soc. 1987, 109, 6865.

(341) Poh, S.; Hernandez, R.; Inagaki, M.; Jessop, P. G. Org. Lett. 1999, 1, 583.

(342) Chen, P.; Gorelsky, S. I.; Ghosh, S.; Solomon, E. I. Angew. Chem. Int. Ed. 2004, 43, 4132.

(343) Trogler, W. C. J. Chem. Educ. 1995, 72, 973.

(344) Zumft, W. G.; Kroneck, P. M. H. Advances in Microbial Physiology; Academic Press: San Diego, California, 2006.

265

(345) Solomon, E. I.; Sarangi, R.; Woertink, J. S.; Augustine, A. J.; Yoon, J.; Ghosh, S. Acc. Chem. Res. 2007, 40, 581.

(346) Paulat, F.; Kuschel, T.; Nather, C.; Praneeth, V. K. K.; Sander, O.; Lehnert, N. Inorg. Chem. 2004, 43, 6979.

(347) Bar-Nahum, I.; Gupta, A. K.; Huber, S. M.; Ertem, M. Z.; Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 2009, 131, 2812.

(348) Tolman, W. B. Angew. Chem. Int. Ed. 2010, 49, 1018.

(349) Otten, E.; Neu, R. C.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 9918.

(350) Neu, R. C.; Otten, E.; Lough, A.; Stephan, D. W. Chem. Sci. 2011, 2, 170.

(351) Tskhovrebov, A. G.; Solari, E.; Wodrich, M. D.; Scopelliti, R.; Severin, K. Angew. Chem. Int. Ed. 2012, 51, 232.

(352) Yildirim, T.; Ciraci, S. Phys. Rev. Lett. 2005, 94.

(353) Sun, Q.; Wang, Q.; Jena, P.; Kawazoe, Y. J. Am. Chem. Soc. 2005, 127, 14582.

(354) Sun, Q.; Jena, P.; Wang, Q.; Marquez, M. J. Am. Chem. Soc. 2006, 128, 9741.

(355) Sun, Q.; Wang, Q.; Jena, P. Appl. Phys. Lett. 2009, 94.

(356) Chandrakumar, K. R. S.; Ghosh, S. K. Nano Lett. 2008, 8, 13.

(357) Yoon, M.; Yang, S. Y.; Hicke, C.; Wang, E.; Geohegan, D.; Zhang, Z. Y. Phys. Rev. Lett. 2008, 100.

(358) Strobel, R.; Garche, J.; Moseley, P. T.; Jorissen, L.; Wolf, G. J. Power Sources 2006, 159, 781.

(359) Gao, B.; Zhao, J. X.; Cai, Q. H.; Wang, X. G.; Wang, X. Z. J. Phys. Chem. A 2011, 115, 9969.

(360) Mann, B. E. J. Chem. Soc., Perkin Trans. 2 1972, 30.

(361) Pileio, G.; Carravetta, M.; Hughes, E.; Levitt, M. H. J. Am. Chem. Soc. 2008, 130, 12582.

(362) Griggs, J. L.; Rao, K. N.; Jones, L. H.; Potter, R. M. J. Mol. Spectrosc. 1968, 25, 34.

(363) Bebbington, M. W. P.; Bourissou, D. Coord. Chem. Rev. 2009, 253, 1248.

(364) Mayer, U.; Gutmann, V.; Gerger, W. Monatsh. Chem. 1975, 106.

(365) Sumerin, V.; Schulz, F.; Nieger, M.; Leskela, M.; Repo, T.; Rieger, B. Angew. Chem. Int. Ed. 2008, 47, 6001.

(366) Staudinger, H.; Meyer, J. Helv. Chim. Acta. 1919, 2, 635.

266

(367) Gololobov, Y. G.; Zhmurova, I. N.; Kasukhin, L. F. Tetrahedron 1981, 37, 437.

(368) Gololobov, Y. G.; Kasukhin, L. F. Tetrahedron 1992, 48, 1353.

(369) Alajarin, M.; Conesa, C.; Rzepa, H. S. J. Chem. Soc., Perkin Trans. 2 1999, 1811.

(370) Widauer, C.; Grutzmacher, H.; Shevchenko, I.; Gramlich, V. Eur. J. Inorg. Chem. 1999, 1659.

(371) Staudinger, H.; Hauser, E. Helv. Chim. Acta. 1921, 4, 861.

(372) Martin, G. E.; Hadden, C. E. Journal of Natural Products 2000, 63, 543.

(373) Goddard, T. D.; Kneller, D. G.; SPARKY3; University of California: San Francisco.

(374) Wolfram-Research; Mathematica-5.2.

(375) Ernst, R. R.; Bodenhausen, G.; Wokaun, A. Principles of Nuclear Magnetic Resonance in One and Two Dimensions; Clarendon Press: Oxford, 1990.

(376) Armor, J. N.; Taube, H. J. Am. Chem. Soc. 1969, 91, 6874.

(377) Bottomley, F.; Crawford, J. R. J. Am. Chem. Soc. 1972, 94, 9092.

(378) Bottomley, F.; Brooks, W. V. F. Inorg. Chem. 1977, 16, 501.

(379) Tuan, D. F.; Hoffmann, R. Inorg. Chem. 1985, 24, 871.

(380) Armor, J. N.; Taube, H. J. Am. Chem. Soc. 1970, 92, 2560.

(381) Diamantis, A. A.; Sparrow, G. J. J. Chem. Soc. D., Chem. Commun. 1970, 819.

(382) Diamantis, A. A.; Sparrow, G. J. J. Colloid Interface Sci. 1974, 47, 455.

(383) Diamantis, A. A.; Sparrow, G. J.; Snow, M. R.; Norman, T. R. Aust. J. Chem. 1975, 28, 1231.

(384) Bottomley, F.; Brooks, W. V. F.; Clarkson, S. G.; Tong, S. B. J. Chem. Soc., Chem. Commun. 1973, 919.

(385) Bottomley, F.; Clarkson, S. G.; Tong, S. B. J. Chem. Soc., Dalton Trans. 1974, 2344.

(386) Bottomley, F. Acc. Chem. Res. 1978, 11, 158.

(387) Neu, R. C.; Otten, E.; Stephan, D. W. Angew. Chem. Int. Ed. 2009, 48, 9709.

(388) Pamplin, C. B.; Ma, E. S. F.; Safari, N.; Rettig, S. J.; James, B. R. J. Am. Chem. Soc. 2001, 123, 8596.

(389) Bottomley, F.; Darkwa, J. J. Chem. Soc., Dalton Trans. 1983, 399.

267

(390) Harrold, N. D.; Waterman, R.; Hillhouse, G. L.; Cundari, T. R. J. Am. Chem. Soc. 2009, 131, 12872.

(391) Koo, K.; Hillhouse, G. L. Organometallics 1995, 14, 4421.

(392) Matsunaga, P. T.; Hillhouse, G. L.; Rheingold, A. L. J. Am. Chem. Soc. 1993, 115, 2075.

(393) Matsunaga, P. T.; Mavropoulos, J. C.; Hillhouse, G. L. Polyhedron 1995, 14, 175.

(394) Vaughan, G. A.; Rupert, P. B.; Hillhouse, G. L. J. Am. Chem. Soc. 1987, 109, 5538.

(395) Vaughan, G. A.; Sofield, C. D.; Hillhouse, G. L.; Rheingold, A. L. J. Am. Chem. Soc. 1989, 111, 5491.

(396) Koo, K.; Hillhouse, G. L. Organometallics 1996, 15, 2669.

(397) Koo, K.; Hillhouse, G. L. Organometallics 1998, 17, 2924.

(398) Bleeke, J. R.; Behm, R. J. Am. Chem. Soc. 1997, 119, 8503.

(399) Kaplan, A. W.; Bergman, R. G. Organometallics 1997, 16, 1106.

(400) Kaplan, A. W.; Bergman, R. G. Organometallics 1998, 17, 5072.

(401) Lee, J. H.; Pink, M.; Tomaszewski, J.; Fan, H. J.; Caulton, K. G. J. Am. Chem. Soc. 2007, 129, 8706.

(402) Dameris, M. Angew. Chem. Int. Ed. 2010, 49, 489.

(403) Tskhovrebov, A. G.; Solari, E.; Wodrich, M. D.; Scopelliti, R.; Severin, K. J. Am. Chem. Soc. 2012, 134, 1471.

(404) Reeds, J. P.; Yonke, B. L.; Zavalij, P. Y.; Sita, L. R. J. Am. Chem. Soc. 2011, 133, 18602.

(405) Cherry, J. P.; Johnson, A. R.; Baraldo, L. M.; Tsai, Y. C.; Cummins, C. C.; Kryatov, S. V.; Rybak-Akimova, E. V.; Capps, K. B.; Hoff, C. D.; Haar, C. M.; Nolan, S. P. J. Am. Chem. Soc. 2001, 123, 7271.

(406) Johnson, A. R.; Davis, W. M.; Cummins, C. C.; Serron, S.; Nolan, S. P.; Musaev, D. G.; Morokuma, K. J. Am. Chem. Soc. 1998, 120, 2071.

(407) Ewen, J. A.; Elder, M. J. Eur. Patent Appl. 0,427,697, 1991; U.S. Pat. 5,561,092, 1996.

(408) Yang, X. M.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10015.

(409) Chien, J. C. W.; Tsai, W. M.; Rausch, M. D. J. Am. Chem. Soc. 1991, 113, 8570.

(410) Ewen, J. A.; Elder, M. J. Eur. Pat. Appl. 0,426, 637 1991.

(411) Sinn, H.; Kaminsky, W.; Vollmer, H. J.; Woldt, R. Angew. Chem. Int. Ed. 1980, 19, 390.

268

(412) Corey, J. Y.; Zhu, X. H.; Brammer, L.; Rath, N. P. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1995, 51, 565.

(413) Jordan, R. F.; Bajgur, C. S.; Willett, R.; Scott, B. J. Am. Chem. Soc. 1986, 108, 7410.

(414) Samuel, E.; Rausch, M. D. J. Am. Chem. Soc. 1973, 95, 6263.

(415) Sun, Y. M.; Piers, W. E.; Parvez, M. Can. J. Chem. 1998, 76, 513.

(416) Noltes, J. G.; Vandenhurk, J. W. G. J. Organomet. Chem. 1963, 1, 377.

(417) Weidenbruch, M.; Herrndorf, M.; Schafer, A.; Pohl, S.; Saak, W. J. Organomet. Chem. 1989, 361, 139.

(418) Devore, D. D.; Timmers, F. J.; Neithamer, D. R. 1997; Vol. WO: 9736937 A1.

(419) Jakle, F. Chem. Rev. 2010, 110, 3985.

(420) Cheng, F.; Jakle, F. Polym. Chem. 2011, 2, 2122.

(421) Chujo, Y. Macromol. Symp. 1997, 118, 111.

(422) Matsumi, N.; Chujo, Y. Polym. J. 2008, 40, 77.