Ligand Modification and F-Block Complexes by Andrea Marie Corrente

UNIVERSITY OF CALGARY

Expanding the Scope of Boraamidinates: Ligand Modification and f-Block Complexes

Andrea Marie Corrente

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF CHEMISTRY

CALGARY, ALBERTA

AUGUST, 2010

Library and Archives Bibliothèque et Canada Archives Canada

Published Heritage Direction du Branch Patrimoine de l’édition

395 Wellington Street 395, rue Wellington Ottawa ON K1A 0N4 Ottawa ON K1A 0N4 Canada Canada

Your file Votre référence ISBN: 978-0-494-69539-5 Our file Notre référence ISBN: 978-0-494-69539-5

NOTICE: AVIS:

The author has granted a non- L’auteur a accordé une licence non exclusive exclusive license allowing Library and permettant à la Bibliothèque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par télécommunication ou par l’Internet, prêter, telecommunication or on the Internet, distribuer et vendre des thèses partout dans le loan, distribute and sell theses monde, à des fins commerciales ou autres, sur worldwide, for commercial or non- support microforme, papier, électronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats. . The author retains copyright L’auteur conserve la propriété du droit d’auteur ownership and moral rights in this et des droits moraux qui protège cette thèse. Ni thesis. Neither the thesis nor la thèse ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent être imprimés ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author’s permission.

In compliance with the Canadian Conformément à la loi canadienne sur la Privacy Act some supporting forms protection de la vie privée, quelques may have been removed from this formulaires secondaires ont été enlevés de thesis. cette thèse.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n’y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.

ii Abstract

A critical aspect of coordination chemistry is ligand design. Boraamidinates

2- - ([RB(NRʹ)2] , bams) are isoelectronic to amidinates ([RC(NRʹ)2] , ams) and guanidinates

- ([R2NC(NRʹ)2] , guans), however, the boron-containing ligands are comparatively less well studied. For instance, all known bams are dianionic with either alkyl or aryl substituents on boron and only complexes of the main group and d-block elements are found in the literature. This thesis focuses on further developing the chemistry of bams through ligand modification and preparation of f-block complexes.

2- The first example of a stable dilithio boraguanidinate ([R2NB(NRʹ)2] , bog) is presented; metathesis reactions reveal that this electron-rich ligand is a strong reducing agent. A novel seven-membered C4OBN heterocycle, isolated while exploring alternative synthetic routes to the bog ligand, represents the first example of a structurally characterized 1,3,2-oxazaborepane.

Altering bams by either incorporating a spacer group on boron or by mono- metallation is achieved. New ferrocenyl (diamino)boranes are prepared and deprotonation is attempted in the pursuit of ferrocenyl bis-boraamidinates. The first example of a

t t tetralithio bis-bam ligand, [Li4][( BuN)2B−C6H4−(N Bu)2], is found to be extremely air- sensitive. Single deprotonation of PhB[N(H)Dipp]2 with either butyllithium or benzyl potassium gives the monoanionic bams [M][PhB{N(H)Dipp}(NDipp)] (M = Li, K). The new dianionic bam reagent [K2][PhB(NDipp)2], which is obtained from double deprotonation of PhB[N(H)Dipp]2 with benzyl potassium, is useful for preventing the inclusion of a salt by-product in metathetical reactions.

iii The coordination chemistry of the bam ligand is extended to the f-block elements in the preparation of “ate” complexes of the form {[Li(THF)4][{PhB(NDipp)2}

LnCl2(THF)]}2 (Ln = Y, Pr, Nd, Sm, Ho, Er, Yb) and 2:1 spirocyclic species

[Li2(THF)8][{PhB(NDipp)2}2LnI] (Ln = La, Sm). The reaction of the yttrium and samarium “ate” complexes with an excess of trimethylsilyl trifluoromethanesulfonate generates {[PhB(NDipp)2]LnCl(THF)2}2. The analogous iodide species are prepared from the reaction of [K2][PhB(NDipp)2] and LnI3(THF)3.5 and metathesis reactions of both neutral complexes are examined. Extensive attempts to form a samarium(II) bam complex reveal that protonolysis is the preferred route over metathesis or reduction.

iv Acknowledgements

I would first like to thank Dr. Tristram Chivers for being an incredible supervisor.

His enthusiasm and involvement in all of my projects is greatly appreciated and the wealth of knowledge I gained from our numerous discussions is immeasurable. I would also like to acknowledge the members of the Chivers group with whom I had the privilege to work over the past four years: Drs. Jari Konu, Stuart Robertson, Jamie Ritch,

Dana Eisler, Maarit Risto and Tracey Roemmele. It was a pleasure being a part of this group of excellent chemists, colleagues and friends.

I would like to extend my appreciation to Dr. Warren Piers and Dr. Thomas

Baumgartner for being members of my supervisory committee, to Dr. George Shimizu who participated in my qualifying exam, and to Dr. Nasser Moazzen-Ahmadi who acted as my internal-external for both examinations. Many thanks to Dr. Alan Cowley for agreeing to make the trip up north to Calgary to serve as my external examiner.

Dr. Sjoerd Harder provided very helpful correspondence when I ventured into the world of lanthanide boraamidinates and I want to express my gratitude to him, especially for his inspirational words that got me through some very frustrating times in the lab:

“This bam chemistry is not easy!”. Thanks as well to Dr. Chuck Macdonald for giving me the push I needed to leave Windsor for grad school - it really was the right decision!

As all chemistry graduate students at UofC know, we would not get through our degrees without the extraordinary efforts of Bonnie King. From lending a sympathetic ear to making sure our i's are dotted and t’s are crossed before a deadline hits, Bonnie keeps everyone on track and her genuine care for the students is something you don’t find just anywhere.

v I’d also like to say thank you to the members of the instrumentation team, including Dr. Masood Parvez, Dr. Yamdagni, Dorothy Fox, Qiao Wu and, in particular,

Johnson Li and Wade White who always gave it their best shot with my less-than- cooperative samples. I truly appreciate your time and efforts! Mike Siewert, Ed Cairns and Keith Collins were essential in keeping our glove box and ovens running, which warrants a huge thank you!

Additionally, I would like to acknowledge NSERC, Alberta Ingenuity and the

University of Calgary for providing ample funding throughout my PhD studies.

Although I came to Calgary knowing not a single person, I am fortunate to now have so many friends here that acknowledging each of them would require me to add

Appendix C to this document; I appreciate everyone who has been a part of my Calgary life and would like to take a moment to mention a few specific people. Thanks to Josh

Smith for wearing a Toronto Maple Leafs shirt during orientation making for easy conversation, for putting up with my homesick *moments* during our first year here and for inviting Korwin to be his Calgary roommate when he came to town. Thank you to

Matt Hobbs for opening my eyes to the greatness of the CFL, pushing me well beyond my limits in the mountains and for our numerous chemistry “discussions”. A very special thanks to Dr. Sean Dalrymple and Alana Keen for being great friends and for all the fun times we’ve shared in Calgary and The ‘Stoon. It has also been nice having fellow

Macdonald group alumnus Paolo Bomben in the department to keep up with what’s going on at home - I wish him luck with the rest of his degree.

I am fortunate to have wonderful family and friends from Windsor who have supported me throughout my University career. I am forever indebted to my incredible

vi parents, whose love and support has gotten me through the roughest of times; even though they probably don’t realize it, I really couldn’t have made it this far without them.

Thank you to my sister and brother, Kimberly and Christopher, for the text messages and iChats that on some days made me homesick and on others made me happy to be 3000 km away! I would also like to thank my Nonna for all of her prayers and for getting excited about every little success I had in grad school.

To my great friends Lindsay, Dana, Melissa and Kendall – thanks for all of the random chats and messages and your visits to CowTown. I would also like to express thanks to Dr. Sharonna Greenberg not only for regular phone dates but also for taking the time to read this thesis and provide me with awesome feedback.

And finally, to my best friend, life co-pilot and soon-to-be-husband Korwin: your daily support and encouragement and your endless tolerance of chemistry talk is appreciated more than I can ever explain. If we made it through grad school together, we can make it through anything!!

vii Dedication

For their endless love and support and for always believing in me, I dedicate this thesis

to my parents, Mary Jean and Silvio.

viii Table of Contents

Approval Page...... ii Abstract...... iii Acknowledgements ...... v Dedication...... viii Table of Contents ...... ix List of Tables ...... xii List of Figures...... xiv List of Schemes...... xvii List of Symbols and Abbreviations...... xviii List of Compounds...... xxii Proem...... xxvi

CHAPTER ONE: INTRODUCTION...... 1 1.1 Preface ...... 1 1.2 Inorganic Chemistry ...... 2 1.2.1 Main Group Chemistry...... 2 1.2.2 Lanthanide Chemistry...... 5 1.3 Coordination Chemistry...... 6 1.3.1 General Concepts...... 6 1.3.2 Ligand Classifications ...... 7 1.4 Amidinates and Guanidinates...... 9 1.4.1 General Considerations...... 9 1.4.2 Highlights of Main Group am and guan Complexes...... 10 1.4.3 Highlights of Transition Metal am and guan Complexes ...... 13 1.4.4 Applications of am and guan Complexes ...... 15 1.5 Boraamidinates...... 19 1.5.1 General Considerations...... 19 1.5.2 Highlights of Main Group bam Complexes...... 21 1.5.3 Novel Redox Behaviour...... 24 1.6 Objectives and Outline of Dissertation...... 27

CHAPTER TWO: INVESTIGATIONS OF BORAGUANIDINATE LIGANDS ...... 29 2.1 Introduction...... 29 2.2 Synthesis, Spectroscopic Characterization and X-ray Structure of [PhB(NDipp)2]PCl, (2.2)...... 31 2.3 Synthesis, Spectroscopic Characterization and X-ray Structures of i i ( Pr2N)B[N(H)Dipp]2 (2.3) and {[Li2][( Pr2N)B(NDipp)2]}2 (2.4)...... 35 2.4 Alternative Syntheses of Li2bog: Synthesis and Characterization of ClB[N(H)Dipp]2 (2.5), BrB[N(H)Dipp]2 (2.6) and B[N(H)Dipp]3 (2.7)...... 41 2.5 Reactions of Li2bog with p-Block Halides ...... 46 2.5.1 Metathesis ...... 46 2.5.2 Reduction ...... 51 2.5.3 Attempted Alkane Elimination...... 53 2.6 Preparation of New Boraguanidinate Ligands ...... 53 2.7 Conclusions...... 55 ix 2.8 Experimental Section...... 57

CHAPTER THREE: FORMATION OF A SEVEN-MEMBERED C4OBN HETEROCYCLE BY A THF RING-EXPANSION PROCESS...... 68 3.1 Introduction...... 68 3.2 Synthesis, Spectroscopic Characterization and X-ray Structure of DippN(H)BO(CH2)4NDipp (3.2a)...... 70 3.3 Synthesis and Spectroscopic Characterization of ClBO(CH2)4NDipp (3.2b)...... 75 3.4 Conclusions...... 77 3.5 Experimental Section...... 78

CHAPTER FOUR: NEW BORAAMIDINATE REAGENTS...... 80 4.1 Introduction...... 80 4.2 Synthesis, Spectroscopic Characterization and X-ray Structures of t t FcB[N(H) Bu]2 (4.1), FcB[N(H)Dipp]2 (4.2) and 1,1ʹ-Fc{B[N(H) Bu]2}2 (4.3)...82 4.2.1 Attempted Lithiation of (Diaminoboryl)ferrocenes ...... 87 4.3 Synthesis, Spectroscopic Characterization, X-ray Structure and Reactions of t t [Li4][(N Bu)2−C6H4−B(N Bu)2]⋅6THF...... 88 4.4 Synthesis and Spectroscopic Characterization [Li][PhB{N(H)Dipp}(NDipp)] (4.6), and [K][PhB{N(H)Dipp}(NDipp)] (4.7), and X-ray Structure of 4.6...... 94 4.5 Synthesis, Spectroscopic Characterization and X-ray Structure of [K2(THF)3][PhB(NDipp)2], (4.8)...... 99 4.6 Conclusions...... 103 4.7 Experimental Section...... 104

CHAPTER FIVE: LANTHANIDE COMPLEXES OF A BORAAMIDINATE LIGAND ...... 111 5.1 Introduction...... 111 5.2 Synthesis, Spectroscopic Characterization and X-ray Structures of 2- {[Li(THF)4][bamLnCl2(THF)]}2, (bam = [PhB(NDipp)2] , Ln = Y, Pr, Nd, Sm, Ho, Er, Yb ...... 114 5.3 Synthesis, Spectroscopic Characterization and X-ray Structures of {bamLnCl(THF)2}2 (Ln = Y, Sm)...... 123 5.4 Synthesis and Spectroscopic Characterization of {bamLnI(THF)n}m...... 128 5.5 Synthesis and Spectroscopic Characterization of [Li2][bam2LnX] (Ln = La, X = I; Ln = Y, Sm, X = Cl)...... 130 5.6 Attempted Synthesis of a bamSmII Complex...... 134 5.6.1 Metathesis Reactions ...... 134 5.6.2 Reduction Reactions ...... 138 5.6.3 Protonolysis (Amine or Alkane Elimination) ...... 140 5.7 Conclusions...... 146 5.8 Experimental Section...... 148

CHAPTER SIX: CONCLUSIONS AND FUTURE WORK ...... 159 6.1 Concluding Remarks ...... 159 6.2 Suggested Future Work ...... 165

x REFERENCES...... 170

APPENDIX A: REAGENTS, PROCEDURES AND INSTRUMENTATION...... 189 A.1. Reagents and General Procedures ...... 189 A.2. Instrumentation...... 190 A.3. X-Ray Structure Determinations ...... 190

APPENDIX B: CRYSTALLOGRAPHIC DATA...... 193

xi List of Tables

Table 2.1. Selected bond lengths (Å) and bond angles (o) for 2.2...... 33

Table 2.2. Selected bond lengths (Å) and bond angles (°) for 2.4 and selected dilithio bams...... 39

Table 2.3. 1H NMR chemical shifts (in ppm) for the iPr groups in 2.3, 2.4, and 2.8- 2.10 ...... 47

(in THF−d8)...... 47

11 Table 2.4 B NMR chemical shifts (in ppm) for 2.3, 2.4, and 2.8-2.10 (in THF-d8)...... 48

1 Table 2.5. Selected H NMR chemical shifts (in THF-d8) for the reactions of 2.4 with LnX3...... 50

Table 3.1. Selected bond lengths (Å) and bond angles (°) for 3.2a...... 71

Table 4.1. Selected bond lengths (Å) and bond angles (o) for 4.1, 4.2 and 4.3...... 85

1 Table 4.2. H NMR chemical shifts (in ppm) for 4.1, 4.2 and 4.3 (in C6D6)...... 87

Table 4.3. Selected bond lengths (Å) and bond angles (o) for 4.5 and t {Li2[PhB(N Bu)2]}2...... 90

1 Table 4.4. H NMR chemical shifts (in THF-d8) for 4.4, 4.5 and 4.5 + 2 MgCl2 ...... 94

o Table 4.5. Selected bond lengths (Å) and bond angles ( ) for 4.6, bamH2 and Li2bam(THF)3...... 95

1 Table 4.6. H NMR chemical shifts (in THF-d8) for monoanionic bams 4.6 and 4.7...... 98

Table 4.7. Selected bond lengths (Å) and bond angles (o) for 4.8...... 101

Table 5.1. Selected bond lengths (Å) for 5.7-5.13...... 116

Table 5.2. Selected bond angles (o) for 5.7-5.13...... 117

1 Table 5.3. The H NMR chemical shifts (in ppm) for complexes 5.7-5.10 (in THF-d8)121

o Table 5.4. Selected bond lengths (Å) and bond angles ( ) for {bamLnCl(THF)2}2...... 126

Table 5.5. The 1H NMR chemical shifts (in ppm) for the isopropyl groups in the neutral bamLnX(THF)2 complexes 5.15-5.17, and “ate” complexes 5.7 and 5.10 (in THF-d8)...... 129

Table 5.6. Selected bond lengths (Å) and bond angles (o) for 5.18...... 132

xii Table 5.7. The 1H NMR chemical shifts (in ppm) for the isopropyl groups 5.18-5.20..133

(in THF-d8)...... 133

Table 5.8. Selected bond lengths (Å) and bond angles (o) for 5.21...... 137

Table 5.9 Formal reduction potentials (V vs. FcH)232 of the reducing agents used in the attempted one electron reduction of 5.17...... 138

Table B.1 Crystallographic data for 2.2-2.4...... 193

Table B.2 Crystallographic data for 2.5, 2.7 and 3.1...... 194

Table B.3 Crystallographic data for 4.1-4.3...... 195

Table B.4 Crystallographic data for 4.5, 4.6 and 4.8...... 196

Table B.5 Crystallographic data for 5.7-5.9...... 197

Table B.6 Crystallographic data for 5.10-5.12...... 198

Table B.7 Crystallographic data for 5.13-5.15...... 199

Table B.8 Crystallographic data for 5.18 and 5.21...... 200

xiii List of Figures

Figure 2.1. Thermal ellipsoid plot (30% probability) of [PhB(NDipp)2]PCl, 2.2. For clarity, hydrogen atoms have been omitted and only α-carbon atoms of Dipp groups are shown...... 32

Figure 2.2. A comparison of electron donation in phosphenium cations (left), carbene 2.1 and bamPCl 2.2...... 35

i Figure 2.3. Thermal ellipsoid plot (30% probability) of ( Pr2N)B[N(H)Dipp]2, 2.3. For clarity, hydrogen atoms other than those on nitrogen have been omitted and only α-carbon atoms of Dipp groups are shown. Selected bond lengths (Å) and angles (°): B1−N1 1.438(2), B1−N2 1.437(2), B1−N3 1.429(2); N1−B1−N2 116.0(1), N1−B1−N3 120.7(1), N2−B1−N3 123.3(1)...... 37

i Figure 2.4. Thermal ellipsoid plot (30% probability) of {[Li2][( Pr2N)B(NDipp)2]}2, 2.4. For clarity, hydrogen atoms and the co-crystallized solvent molecule (hexane) have been omitted and only α-carbon atoms of Dipp groups are shown. Symmetry elements used to generate equivalent atoms: #1, −x + 1, y, −z + ½; #2, −x, −y, −z + 1...... 38

Figure 2.5. Thermal ellipsoid plot (30% probability) of BrB[N(H)Dipp]2, 2.6. For clarity, hydrogen atoms other than those on nitrogen have been omitted and only α-carbon atoms of Dipp groups are shown. Selected bond lengths (Å) and angles (°): B1−N1 1.383(7), B1−N2 1.402(7), B1−Br1 1.951(6); N1−B1−N2 123.5(5), N1−B1−Br1 119.2(4), N2−B1−Br1 117.2(4)...... 43

Figure 2.6. Thermal ellipsoid plot (30% probability) of B[N(H)Dipp]3, 2.7. For clarity, hydrogen atoms other than those on nitrogen have been omitted and only α-carbon atoms of Dipp groups are shown. Selected bond lengths (Å) and angles (°): B1−N1 1.423(3), B1−N2 1.427(3), B1−N3 1.426(3); N1−B1−N2 120.4(2), N1−B1−N3 120.6(2), N2−B1−N3 119.0(2)...... 45

Figure 2.7. Proposed structures of bog complexes 2.8-2.10...... 49

Figure 2.8. Targeted dianionic bog ligands...... 53

Figure 3.1. Thermal ellipsoid plot (30% probability) of one conformation of DippN(H)BO(CH2)4NDipp (3.2a). For clarity, all hydrogen atoms except that on nitrogen have been omitted and only α−carbon atoms of Dipp groups are shown...72

Figure 3.2. Chair (top) and boat (bottom) conformations of the C4NBO ring in 3.2a with hydrogen atoms and substituents on boron and nitrogen omitted for clarity....73

t Figure 4.1. Thermal ellipsoid plots (30% probability) of FcB[N(H) Bu]2 (4.1, left) and FcB[N(H)Dipp]2 (4.2, right). For clarity, most H atoms omitted and only α- carbons of Dipp groups are shown...... 84

xiv t Figure 4.2. Thermal ellipsoid plot (30% probability) of 1,1ʹ-Fc[B(N Bu)2]2 (4.3) with hydrogen atoms omitted for clarity, other than those located on the nitrogen atoms...... 86

Figure 4.3. Thermal ellipsoid plot (30% probability) of t t [Li4][(N Bu)2B−C6H4−B(N Bu)2]⋅6THF, 4.5. Hydrogen atoms omitted for clarity. Symmetry transformations used to generate equivalent atoms: −x+1, −y+1, −z...... 90

Figure 4.4. Solutions of 4.5 under argon (left) and after introduction of air (right)...... 91

Figure 4.5. EPR spectra of the oxidation of 4.5 with half an equivalent of iodine, collected at various temperatures...... 92

Figure 4.6. Thermal ellipsoid plot (30% probability) of [Li(THF)2][PhB(NHDipp)(NDipp)], 4.6. For clarity, hydrogen atoms on carbon atoms have been omitted and only α-carbons of Dipp groups shown...... 95

Figure 4.7. A comparison of 4.6 with other structurally characterized lithium ams.168,169...... 97

Figure 4.8. Thermal ellipsoid plot (30% probability) of a discrete [K2(THF)3][PhB(NDipp)2] unit, K2bam(THF)3 (4.8). For clarity, all hydrogen atoms and carbon atoms of THF solvent molecules have been omitted and only α-carbon atoms on Dipp groups are shown. Symmetry elements used to generate equivalent atoms: −x, y+½, −z+½; −x, y−½, −z+½...... 100

Figure 4.9. Representation of the extended solid state structure of 4.8. For clarity all hydrogen atoms and selected carbon atoms have been omitted...... 101

Figure 5.1. Thermal ellipsoid plot (30% probability) of the anion of 5.11, representative of complexes 5.7-5.13. For clarity, hydrogen atoms, lithium counter ions, and lattice solvent (THF) have been omitted and only the α-carbon atoms of Dipp groups are shown. Symmetry elements used to generate equivalent atoms: #1 −x+2,−y,−z+2...... 116

1 Figure 5.2. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamYCl2(THF)]}2 (5.7) ...121

1 Figure 5.3. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamPrCl2(THF)]}2 (5.8)..122

1 Figure 5.4. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamNdCl2(THF)]}2 (5.9) 122

1 Figure 5.5. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamSmCl2(THF)]}2 (5.10)...... 123

Figure 5.6. Thermal ellipsoid plot (30% probability) of 5.15 (representative of 5.14). For clarity, hydrogen atoms and selected THF carbon atoms have been omitted

xv and only the α-carbon atoms of Dipp groups are shown. Symmetry elements used to generate equivalent atoms: #1 −x,−y+3,−z+1...... 125

Figure 5.7. Thermal ellipsoid plot (30% probability) of the dianion of 5.18. For clarity, hydrogen atoms, lithium counter ions, and solvent (THF) have been omitted and only the α-carbon atoms of Dipp groups are shown. Symmetry elements used to generate equivalent atoms: #1 −x+1,−y+1, z...... 131

Figure 5.8. Proposed structures of the products obtained from the reactions of SmI2 or SmI3 and Li2bam (left) or K2bam (right)...... 135

Figure 5.9. Thermal ellipsoid plot (30% probability) of the dianion of 5.21. For clarity, hydrogen atoms, lithium counter ions, and solvent (THF) have been omitted and only the α-carbon atoms of Dipp groups are shown...... 136

1 Figure 5.10. H NMR spectrum (in THF-d8) of the dark product obtained from the amine elimination reaction bamH2 + Sm[N(SiMe3)2]2(THF)2...... 142

xvi List of Schemes

Scheme 2.1...... 30

Scheme 2.2...... 34

Scheme 2.3...... 36

Scheme 2.4...... 41

Scheme 2.5...... 44

Scheme 2.6...... 52

Scheme 3.1...... 71

Scheme 3.2...... 75

Scheme 4.1...... 83

Scheme 4.2...... 85

Scheme 4.3...... 89

Scheme 4.4...... 99

Scheme 5.1...... 130

Scheme 5.2...... 141

Scheme 5.3...... 143

Scheme 5.4...... 144

Scheme 5.5...... 145

Scheme 5.6...... 146

Scheme 6.1...... 166

Scheme 6.2...... 168

Scheme 6.3...... 169

xvii List of Symbols and Abbreviations

° degrees % percent . unpaired electron α alpha, unit cell angle β beta, unit cell angle γ gamma, unit cell angle π pi Θ theta µ bridging group or absorption coefficient ∑ sum ∠ angle > greater than < less than < less than or equal to λ wavelength δ chemical shift (in parts per million) ρ density a, b, c unit cell edge length (X-ray crystallography) Å angstrom ALD atomic layer deposition am amidinate bam boraamidinate bamam amidinate/boraamidinate bog boraguanidinate br broad nBu n-butyl tBu tert-butyl Bz benzyl

xviii C Celsius ca. about calcd. calculated cf. compare to Cp cyclopentadienyl Cp* pentamethyl cyclopentadienyl CVD chemical vapour deposition Cy cyclohexyl d day(s) or doublet or deuterium DFT density functional theory Dipp 2,6-diisopropylphenyl DMA 2-(dimethylamino)benzyl

DMAT 2-(Me2N-α-Me3Si)benzyl DME dimethoxyethane E element e.g. for example EPR electron paramagnetic resonance eq equivalents Et ethyl Fc ferrocenyl fw formula weight g gram or g value G Gauss guan guanidinate h hours HMDS hexamethyldisilylazide hpp 1,3,4,6,7,8-hexahydro-2H-pyrimido-[1,2- a]pyrimidine Hz hertz I nuclear spin quantum number i.e. that is

xix in situ in the original place in vacuo under vacuum J coupling constant K Kelvin LDA lithium diisopropylamide

Li2bam [Li][PhB(NDipp)2], used in synthesis M metal or moles per litre m multiplet Me methyl Mes 2,4,6-trimethylphenyl mg milligram mL millilitre min minute(s) mmol millimole mol mole MOCVD metal-organic chemical vapour deposition mp melting point m/z mass to charge ratio NMR nuclear magnetic resonance OTf trifluoromethanesulfonate PDI polydispersity index Ph phenyl PMDTA N,N,Nʹ,Nʹʹ,Nʹʹ-pentamethyldiethylenetriamine ppm parts per million p-tolyl para-tolyl [4-(Me)C6H4 group] q quartet iPr iso-propyl R agreement factor (X-ray crystallography) R alkyl or aryl substituent rac racemic ref. references

xx ROP ring-opening polymerization s singlet sept. septet SOMO singly-occupied molecular orbital T temperature t triplet THF tetrahydrofuran TMEDA N,N,Nʹ,Nʹ-tetramethylethylenediamine TMS trimethylsilyl Trip 2,4,6-triisopropylphenyl V volume V volt via by way of wR weighted agreement factor (X-ray crystallography) X halide or other monoanionic ligand Z formula units per unit cell (X-ray crystallography)

xxi List of Compounds

Ph NiPr B 2 Dipp N N Dipp Dipp B Dipp N N P H H Cl 2.2 2.3

i N Pr2 Cl

Dipp B Dipp Dipp B Dipp N N N N H H Li Li 2.4 2.5

Br Dipp H N Dipp B Dipp N N Dipp B Dipp N N H H H H 2.6 2.7

Dipp Dipp

N N i i Pr2N B Ge O O Ge B N Pr2 N N

Dipp 2.8 Dipp

xxii i i Pr2N Pr2N

B B Dipp N N Dipp Dipp N N Dipp Mg Zn

O O O O 2.9 2.10

Dipp Dipp N N i i Pr2N B La(I)(THF)n Pr2N B Y(Cl)(THF)n N N Dipp Dipp m m 2.11 2.12

Ph Me Ph Me N N

Dipp B Dipp B N N Dipp N N Dipp

H H Li Li 2.14H2 Li22.14

Ph Ph Ph Ph N N

Dipp B Dipp Dipp B Dipp N N N N

H H H Na 2.15H2 2.16

xxiii H2 H2 C C H2C CH2 H2C CH2

O CH2 O CH2 B N B N Dipp Dipp(H)N Cl Dipp 3.2a 3.2b

t B[N(H) Bu]2 B[N(H)Dipp]2

Fe Fe

4.1 4.2

N(H)tBu B tBu(H)N t N(H)tBu N(H) Bu tBu(H)N Fe B B B tBu(H)N N(H)tBu tBu(H)N 4.3 4.4

THF - Ph t 4 BuN NtBu Li B Dipp B 4 [Li]+ N N B THF H tBuN NtBu Dipp 4.5 4.6

xxiv Ph Ph

Dipp B Dipp Dipp B Dipp N N N N H K K K 4.7 4.8

+ Li (THF)4 Dipp Dipp Dipp Dipp Cl THF THF N N THF Cl N Cl N Ph B Ln Ln B Ph Ph B Ln Ln B Ph Cl Cl N N N N THF Cl THF THF Dipp Dipp Dipp Dipp + Li (THF)4 Ln = Y (5.7), Pr (5.8), Nd (5.9), Sm (5.10), Ho (5.11), Er (5.12), Yb (5.13) Ln = Y (5.14), Sm (5.15)

2 Dipp Dipp N X N Ph B Ln B Ph 2 [Li] Dipp N N N Dipp Dipp Ph B Ln(I)(THF)n N 5.18, Ln = La, X = I Dipp m 5.19, Ln = Y, X = Cl 5.20, Ln = Sm, X = Cl Ln = Y (5.16), Sm (5.17) 5.21, Ln = Sm, X = I

xxv Proem

The following is a list of publications arising from the work described in this thesis. The candidate was involved in the writing of the first drafts of all manuscripts as well as figure preparation. Additionally, the candidate performed all single crystal X-ray structural determinations.

Chapter Two

Corrente, A.M. and Chivers, T. “Boraguanidinates: Synthesis, X-ray Structures and

i Reactions of {Li2[ Pr2NB(NDipp)2]}2 with p-Block and Group 12 Element Halides”;

Inorg. Chem. 2008, 47, 10073.

The candidate prepared and characterized all compounds reported in this full paper.

Konu, J., Tuononen, H.M., Chivers, T., Corrente, A.M., Boeré, R.T., Roemmele, T.L.

Inorg. Chem. 2008, 47, 3823.

The candidate prepared and characterized [PhB(NDipp)2]PCl (compound 2.2 in this thesis).

Chapter Three

Corrente, A.M. and Chivers, T. “Formation and X-ray Structure of a Seven-Membered

C4OBN Heterocycle by a THF Ring-Expansion Process”; Dalton Trans. 2008, 4840.

xxvi The candidate prepared and characterized all compounds reported in this communication.

Chapter Four

Corrente, A.M. and Chivers, T. “Syntheses and Structures of New Alkali-Metal

Boraamidinates and Ferrocenyl Aminoboranes”; New J. Chem. 2010, 34, 1751.

The candidate prepared and characterized all compounds reported in this full paper.

Chapter Five

Corrente, A.M. and Chivers, T. “Lanthanide Complexes of Boraamidinate Ligands:

Synthesis and X-Ray Structures of {[Li(THF)4][bamLnCl2(THF)]}2, (bam =

2- [PhB(NDipp)2] ; Ln = Y, Pr, Nd, Sm, Ho, Er, Yb), and {bamLnCl(THF)2}2 (Ln = Y,

Sm)”; Inorg. Chem. 2010, 49, 2457.

The candidate prepared and characterized all compounds reported in this full paper.

xxvii 1

Chapter One: Introduction

1.1 Preface

The scope of this thesis encompasses the areas of main group (s- and p-block) and lanthanide (f-block) chemistry. Ligand design and coordination chemistry are examined in the following ways: (i) the development of new nitrogen- and boron-containing ligands and assessment of their metathetical reactions with main group halides; (ii) modification of the known boraamidinate ligand by either changing the substituent on boron to a bridging group or by mono-metallation of the backbone; (iii) expansion of the coordination chemistry of the boraamidinate ligand to the lanthanide elements. In this introductory chapter, an overview of main group and lanthanide chemistry as well as a discussion of relevant concepts in coordination chemistry is provided, along with a more detailed discussion of three ligands that are most significant to this work: amidinates, guanidinates and boraamidinates. To conclude, the overall objectives of this thesis are described.

1.2 Inorganic Chemistry

Inorganic chemistry has evolved dramatically over the years driven in part by its large scope of practical applications, which range from Portland cement and phosphate fertilizers1 to compounds that exhibit anti-tumour properties, such as cis- diamminedichloroplatinum(II), (cisplatin).2 Today, research is dedicated not only to examining the fundamental structure and bonding in inorganic compounds, but also to applying this knowledge to areas such as the development of improved catalysts and medicinal compounds. In order to enhance the efficiency of these inorganic systems, much research is aimed at developing new ligands and studying their coordination chemistry.

The discipline of inorganic chemistry can be divided into the following major constituents: main group (comprised of the s- and p-blocks), transition metal (d-block), lanthanide and actinide (f-block), organometallic (joining the disciplines of organic and inorganic chemistry) and bioinorganic. The following sections provide a brief introduction to the two branches of inorganic chemistry relevant to this dissertation: main group and lanthanide chemistry. Subsequently, aspects of coordination chemistry will be presented, as this facet of inorganic chemistry has evolved over the past century to be a crucial component of industrial and materials science applications.

1.2.1 Main Group Chemistry

Main group chemistry3,4 encompasses the elements of the s- and p-blocks, without consideration of d-electrons. The group 12 elements zinc, cadmium and mercury have an

3 electron configuration of [core]nd10s2 and a 2+ oxidation state generated by removal of the two s electrons; as such they are often regarded as part of the main group. The s- block is comprised of metallic elements (the alkali and alkaline earth metals) while the p- block contains not only metals, but also non-metallic elements and metalloids. Metals are characterized by having low electronegativities and are good conductors of electricity and heat, whereas non-metals (for example, the halogens) have higher electronegativities and are insulators. Metalloids are then classified as those elements with intermediate electronegativities and include the elements boron, silicon, germanium, arsenic, antimony and tellurium, in addition to the radioactive polonium.

The metals of group 1 easily achieve a noble gas configuration by loss of the single s electron and, therefore, their compounds form predominately by ionic bonding in the 1+ oxidation state. Another shared feature between the alkali metals is their high reactivity towards water, forming the metal hydroxide with evolution of hydrogen gas.

Among the group 2 metals, the chemistry of beryllium differs the most owing to its high charge:radius ratio of its 2+ cation, precluding it from forming free Be2+ cations.

The rest of the elements in this group, however, exhibit a strong parallel in their properties. In general, the high charge density of the group two ions favours the formation of coordination complexes and, with the exception of beryllium, it is common to observe six donors coordinated to cations of this group.

With an electron configuration of [core]ns2p1, the group 13 elements can only form three covalent bonds, leaving an empty p-orbital and rendering them electron- deficient. This results in compounds of these elements acting as Lewis acids (electron acceptors), which can either coordinate a Lewis base (electron donor) or participate in π

4 back-donation from an adjoining element with a free lone pair. The chemistry of the lighter elements of group 13 is dominated by the 3+ oxidation state. However, moving down the group there is an increase in stabilization of the 1+ oxidation state, owing to the inert pair effect and relativistic effects. The inert pair effect arises from the fact that low energy s-electrons require greater energy for promotion to form spn hybrid orbitals, and since this energy is not compensated for by the strength of the bond formed, a lower oxidation state is favoured. Relativistic effects occur in heavy elements, where electrons close to the nucleus are traveling at speeds close to the speed of light causing an increase in mass and a decrease in orbital radius. This results in the electrons being held closer to the nucleus and a greater amount of energy being required for electron promotion. Both the inert pair effect and relativistic effects play a role in the oxidation states commonly found for elements of groups 14 and 15 as well.

Group 14 contains the element carbon, upon which the field of organic chemistry is based; however, some molecules entirely based on carbon, (fullerenes, for example), are sometimes viewed as inorganic species. In their native state, group 14 elements are relatively unreactive, with the best example being that of the carbon allotrope diamond.

Elements of group 15, often referred to at the pnictogens, display a wide range of chemistry owing to the diversity in electronic character going down the group: nitrogen is a non-metal whereas bismuth is metallic in nature. This trend of going from a non-metal to a metal also holds true for the elements of group 16, named the chalcogens.

A highlight of the chalcogen elements is the ability to participate in catenation, or the formation of chains and rings of a single element. This is particularly important in

5 the chemistry of sulfur, which exists as cyclo-S8 in the elemental state, although it also plays a role in selenium and tellurium chemistry.

The halogens are the elements of group 17 and contain the most electronegative element, fluorine. Although commonly found in the 1- oxidation state, a wide range of positive oxidation states are available for chlorine, bromine and iodine.

Possessing a completely filled valence shell, the group 18 noble gases are viewed as being inert; however, the two heaviest, non-radioactive elements in this group, krypton and xenon, have been shown to form bonds with other elements such as fluorine and oxygen. Argon is notable in this group, as it is a commonly used atmosphere under which air-sensitive chemistry is conducted.

1.2.2 Lanthanide Chemistry

The lanthanide elements are found in the sixth period of the periodic table, for which the 4f subshell is filled across the series. However, the f-subshell does not begin filling until the second lanthanide element, as the 5d subshell in lanthanum is lower in energy than the 4f, resulting in the electron configuration for La being [Xe]6s25d1.

Additionally, the group 3 elements scandium and yttrium are often classified with the f- elements and referred to as pseudo-lanthanides, as there is a strong parallel in their chemistry.

Unlike transition metals of the d-block, which usually demonstrate various oxidation states, the lanthanide elements predominantly exist in the 3+ oxidation state, although the 2+ and 4+ states are known for some elements. Europium, samarium and ytterbium are the three most common elements in the series existing in the 2+ state,

(although the diiodides of neodymium, dysprosium and thullium exist), whereas chemistry in the 4+ oxidation state is only found frequently for cerium.5

One unique aspect of the lanthanide series is that ionic radii decrease going across the period, referred to as the lanthanide contraction. The 4f orbitals are adequately shielded by the 5s and 5p subshells and can essentially be viewed as core electrons.

Consequently, from left to right across the lanthanides, the increased nuclear charge is not effectively shielded by the 4f subshell, resulting in a higher effective nuclear charge and a decreased radius. Relativistic effects also contribute to this decrease, although to a much smaller extent (ca. 10%).6

Shielding of the 4f-orbitals also means that they do not participate in bonding; consequently, lanthanide complexes are ionic and multiple bonding is not observed, which is in contrast to main group and transition metal chemistry. A wide range of coordination geometries, primarily influenced by the steric effects, is observed for the lanthanides. Typically, donors such as nitrogen or oxygen are preferred for lanthanide metals and, although coordination numbers as low as 2 exist, it is more common for the lanthanide centre to coordinate 6 to 12 donors.

1.3 Coordination Chemistry

1.3.1 General Concepts

A coordination complex contains a central atom or ion surrounded by molecules

- - or atoms, called ligands, which can be neutral (e.g. :NH3, :PR3) or anionic (e.g. Cl , OR ).

If all of the ligands in a complex are the same, the term homoleptic is used; conversely,

7 heteroleptic complexes contain more than one type of ligand. Coordination compounds were first proposed by Alfred Werner in the early 1890s when he examined homoleptic

3+ cobalt ammines of the form [Co(NH3)6] . He suggested that for such compounds both a primary and secondary valence existed, which today correspond to oxidation state and coordination number, respectively.1,7 Werner was later awarded a Nobel Prize for his substantial contributions to this field. A short time after, Gilbert N. Lewis’ establishment of the octet rule and diagrams that used dots to illustrate free electron pairs and bonds led

Nevil Sidgwick to apply these concepts to Werner’s compounds.8

The central atom in a coordination complex is a Lewis acid, as it accepts electrons from the ligands, whereas the electron-donating ligands are Lewis bases. The interaction between a Lewis base and Lewis acid is known as a coordinate covalent or dative bond, differing from a regular covalent bond only in where the electrons originate (i.e. both electrons come from the same atom in a dative bond, whereas in covalent bonding each atom donates one electron). The coordination number of a compound is defined as the number of atoms bonded to the central atom.

1.3.2 Ligand Classifications

The term ligand originates from the Latin word “ligare”, which means “to bind”;9 in chemistry, ligands are defined as atoms or molecules that can donate a pair of electrons. Ligands can be described as being either hard or soft.10 Hard ligands are defined as being small, negatively charged, not easily polarizable and highly electronegative, therefore they are predominantly involved in ionic bonding. Soft ligands possess the opposite characteristics: large, easily polarizable and typically form covalent

8 bonds. The fluoride anion, for example, is classified as a hard ligand, whereas iodide is

- soft; an amide ( NR2) is another example of a hard donor while a neutral amine (HNR2) is considered soft. Similar to the solubility concept of “like dissolves like”, hard ligands are more appropriate for hard cations (for example, group 1) and soft ligands for soft cations

(for instance, group 11).10

The number of ligand atoms bonded to the central Lewis acid defines the denticity of the ligand; a monodentate ligand has one donor atom, two donors correspond to a bidentate ligand and so forth. Common monodentate ligands in inorganic chemistry include water, ammonia, and tetrahydrofuran (THF), while examples of bidentate ligands include the neutral molecules ethylenediamine and 2,2'-bipyridine and the anion acetylacetonate. Macrocycles such as crown ethers, cryptands and porphyrins are examples of polydentate ligands, as they can have 4 or more coordination sites.

When two or more donor atoms in a ligand interact with the same metal centre, a chelate ring is formed; the most stable chelates contain 5 or 6 atoms. Tetramethylethylene diamine (TMEDA) is a common example of a chelating ligand. A ligand may also bridge metal centres, which is denoted by the symbol µn, where n indicates how many centres are being bridged.

In the following two sections, three ligands that are anionic, hard donors that can be chelating or bridging, namely amidinates, guanidinates and boraamidinates, will be explored in detail, as they are the ligands most relevant to this thesis.

1.4 Amidinates and Guanidinates

1.4.1 General Considerations

Amidinates, ams (1.1), and guanidinates, guans (1.2), are monoanionic ligands that have been thoroughly investigated as ligands for elements throughout the periodic table.11-14 Both the steric and electronic properties of ams and guans can be modified by varying the R and R' groups, an important facet of ligand design.

R NR2

R'N NR' R'N NR'

1.1 1.2 R = H, alkyl, aryl R = alkyl, trimethylsilyl R' = H, alkyl, aryl, trimethylsilyl R' = H, alkyl, aryl, trimethylsilyl

Although the first benzamidines were prepared in 1973,15 it was only in the late

1980s that Oakley and co-workers expanded on the versatility of this ligand by reporting an improved synthetic protocol that allowed for steric and electronic tunability.16

Guanidinates were first introduced by the Lappert group in 1970 as ligands for titanium and zirconium.17 Since these initial reports, metal complexes of both ligands have been prepared using similar synthetic approaches, which include the following:11 (i) carbodiimide insertion into a metal-carbon (ams) or metal-nitrogen (guans) bond; (ii) alkane elimination between an amidine and metal alkyl; (iii) metallation of an amidine or guanidine by an alkaline earth element, followed by a metathetical reaction with a metal halide.

The coordination chemistry of ams and guans is diverse: both chelating (i) and bridging structures (ii) are known, with the former being the more common mode of bonding and the latter being known for some transition metal complexes. It is least common for amidinates or guanidinates to bond in a monodentate fashion (iii) with a free, pendant –NR unit.

R R R

R'N NR' R'N NR' R'N NR' M M M M

i ii iii

Since the rich coordination chemistry of amidinates and guanidinates has been detailed in several reviews,11-14 only selected highlights are discussed here.

1.4.2 Highlights of Main Group am and guan Complexes

Group 1 complexes of ams, particularly those of lithium, are commonly used as starting materials for metathetical reactions with metal halides. The diversity of the bonding in alkali metal formamidinates has been reviewed18and thorough examination of the bonding in lithium fluoroarylamidinates has been carried out.19 In the solid state, lithium ams are commonly dimeric, although other structural forms have been observed by single crystal X-ray diffraction, including monomeric20-23 and even monodentate24-26 complexes. Dimerization of the lithium amidinates is prevented by employing chelating, nitrogen-based ligands such as N,N,N',N'',N''-pentamethyldiethylenetriamine

(PMDTA),20 and TMEDA,21-23 while the monodentate complexes are isolated by using amidinates with bulky terphenyl24,25 (1.3) and triptycenyl26 (1.4) substituents on carbon.

Extended solid state structures can also be observed for lithium ams if the carbon substituent is 3- or 4-pyridyl, as the coordination of the nitrogen atom to a lithium centre displaces the TMEDA ligand.27

i iPr Pr

i iPr Pr iPr iPr iPrN NiPr iPrN NiPr Li

Me2N NMe2 Li(THF)2

1.3 1.4

Although amidinate and guanidinate complexes of magnesium and the heavier alkaline earth metals are readily found in the literature, to date there is only one report of beryllium am complexes.28 Both a 2:1 am:Be spirocyclic product (1.5) as well as a six- membered ring containing a bridging am ligand (1.6) were isolated from a single reaction, in which the am ligand was prepared in situ followed by addition of BeCl2. The stabilizing ability of the guan ligand has recently been exploited by Jones and co-workers in the synthesis and crystallographic characterization of a Mg(I) complex with a unique magnesium-magnesium bond (1.7).29

Cl SiMe3 SiMe3 SiMe3 Dipp Dipp

N N Me3Si Be N N N i i Ph Be Ph N Ph Pr2N Mg Mg N Pr2 N N Be N N N Me3Si

SiMe SiMe3 Dipp Dipp 3 Cl SiMe3 1.5 1.6 1.7 Dipp = 2,6-diisopropylphenyl

The Jones group has also reported novel group 13 complexes of ams and guans, including the following: (i) four-membered metal(I) guans, which are analogues of N- heterocyclic carbenes (1.8),30 and their complexes with group 10 metals;31 (ii) isomers of these carbene analogues using the am ligand;32 and (iii) group 13 metal(III) ams and their

33 complexes with CpFe(CO)2. Additionally, the first example of an amido indium-hydride

(1.9) was prepared using a bulky am ligand, which provided the necessary steric protection required for thermal stability of the In−H bond.34 Finally, a rigid guan ligand has recently been used by the group of Himmel to form a complex of the lightest group

13 element, through isolation of the first dinuclear B(II) cation (1.10).35

2+ N

NCy2 Dipp H Dipp N N 2 Cl- N N In DippN NDipp Me2HN B B NHMe2 H H E N N N N

E = Ga, In Dipp Dipp N 1.8 1.9 1.10

Both ams and guans ligands have been used to isolate thermally stable Ge(I) dimers36 and novel diarsenes (compounds involving As=As double bonds)37 from the reduction of the corresponding Ge(II) and As(III) chloride species, respectively. In group

i i i i 15, the antimony guan complex [Sb{( PrN)2CNH Pr}{( PrN)2CN Pr}] was prepared and found to pack in a helical fashion in the solid state owing to hydrogen bonding of the

−NH group on the backbone carbon.38 Additionally, a series of homo- and heteroleptic formamidinate complexes of bismuth39 and heteroleptic am complexes of antimony and

13 bismuth40 exhibiting a variety of bonding modes that varies with the ligand substituents have recently been reported.

1.4.3 Highlights of Transition Metal am and guan Complexes

Both amidinate and guanidinate ligands can stabilize transition metal complexes where the metal centre is either electronically unsaturated or in an unusual oxidation state. For example, a 14 electron tantalum methylidene bis-amidinate complex (1.11) was prepared by Arnold and co-workers41 and found to be reactive towards pyridine N- oxides; this reactivity was not observed for the tantalocene analogue,

42 Cp2Ta(CH2)(CH3). A vanadium(II) dimeric species with a short V−V multiple bond has been prepared from the reaction of VCl2(TMEDA)2 with the bulky lithium formamidinate, [Li][CyNC(H)NCy].43 Crystallographic analysis provided insight into the strength of the multiple bonds and it was concluded that the nature and steric bulk of the ligand was important in their formation. The reaction of an excess of bis(trimethylsilyl)benzamidinate with dimolybdenum tetraacetate produced a low yield two isomers of a dimolybdenum(II) species, one of which contained a bridging am ligand, while in the other the am was chelating.44 This was particularly notable because complexes of molybdenum bearing bidentate ligands typically exhibit bridging in coordination complexes; few examples of chelation were known at the time. Another chelating am complex of quadruply bonded molybdenum (1.12) was later reported and the product obtained from exposure to dry oxygen was structurally characterized.45 More recently, an iron(I) am complex (1.13) was prepared by the reduction of the

14 corresponding iron(II) bromide,46 and gold(II)47 complexes have also been isolated using a rigid am ligand.

Three notable examples of transition metal centres stabilized by guanidinate ligands include molybdenum, cobalt and iridium. In 2002, Cotton and co-workers reported a series of high oxidation state molybdenum paddlewheel complexes containing

m+ a rigid guan ligand, Mo2(hpp)4 , (hpp = 1,3,4,6,7,8-hexahydro-2H-pyrimido-[1,2- a]pyrimidine, m = 0-2), with Mo−Mo bond orders of 4, 3.5 and 3 (for m = 2, 1 and 0, respectively).48 The guan ligand altered the electrochemical potential of the dimolybdenum unit, which resulted in the stabilization of the higher oxidation state molybdenum atoms. The Jones group has prepared a series of monomeric and dimeric cobalt(I) complexes of guan and am ligands49 with the dimeric complexes (1.14) possessing the shortest known Co−Co interaction known to date. Most recently, guans have been used to stabilize complexes of iridium(IV), formed by the oxidation of the

50 corresponding Ir(III) species by [FeCp2][PF6], further demonstrating the utility of this ligand.

R R SiMe t 3 iPr iPr Bu Dipp Dipp MeH C N N 4 6 N Dipp N N CH2 O Dipp N N Ta O Me3SiN CH3 Mo Mo Co Co O Fe NSiMe3 Me3SiN O N N N N Dipp Dipp iPr iPr C6H4Me R R 1.11 1.12 1.13 1.14

1.4.4 Applications of am and guan Complexes

Main group, transition metal and lanthanide am and guan complexes have demonstrated potential utility in catalysis and materials science applications. The catalytic chemistry of group 451 and ruthenium52 am complexes, as well as applications of lanthanide ams and guans53 has been reviewed. This section presents selected examples of am and guan complexes as catalysts for polymerization of polar monomers and olefins and will also highlight the use of complexes in the materials science field of thin films.

Interest in the ring-opening polymerization (ROP) of cyclic esters catalyzed by lanthanide am and guan complexes has increased over the past decade. Both neutral and anionic yttrium(III) ams have been shown to polymerize rac-lactide, with the former being more effective than the latter.54 Homoleptic amidinate complexes of the form

[CyNC(Me)NCy]3Ln, (Ln = Nd, Gd, Yb), were first prepared and reported as initiators for the ROP of ε-caprolactone in 2002 by Shen and co-workers,55 who have also investigated the catalytic potential of bridged bis-am complexes of yttrium and ytterbium

(1.15) 56 as well as the ytterbium complex 1.16.57

Ph Ph

N N i Me3SiN N Pr N NSiMe3 N N N N O Ln Ln Yb Yb O N N N N Me3SiN N iPr N NSiMe3 N N

Ph Ph 1.15 1.16 Ln = Y, Yb

N N N N = Ph Ph N N N N Me3Si SiMe3

Additionally, the Shen group has pursued yttrium, neodymium and ytterbium

i - complexes containing two of the guan ligands [(SiMe3)2NC(N Pr2)2] , and either an amido58 or alkyl group59 for ROP of ε-caprolactone. Although high initiator activity was demonstrated, the resulting polymers had high polydispersity indices (PDIs). Recently, bis-guan metal alkoxide complexes of this ligand have shown good activity towards ROP of rac-lactide and r-β-butyrolactone, with the latter having PDIs in the range of 1.09-

1.69.60 Monoamidinate complexes of lutetium, yttrium and neodymium can also

61 effectively catalyze the ROP of L-lactide, yielding polymers with narrow PDIs.

Aluminum ams have been shown to act as catalysts in olefin polymerization. In

1996, the group of Teuben reported the synthesis and characterization of the aluminum amidinate complex, [PhC(NSiMe3)2]2AlH, which was reactive towards unsaturated olefins.62 Shortly after, Jordan and co-workers investigated cationic aluminum am

+ complexes of the form [RC(NR')2]AlR and their activity in the polymerization of ethylene,63,64 although the precise active species was not conclusively determined.

Highlights of group 5 amidinate complexes as catalysts in olefin oligomerization and polymerization include reports of a mixed Cp/am tantalum(V) complex,65 a vanadium(III) perfluorobenzamidinate66 and a vanadium(III) species with a pendant amine functionality (1.17).67 More recently, when activated with methylaluminoxane, the late metal am [PhC(NSiMe3)2]Ni(acac) was shown to have promise for catalyzing the addition polymerization of norbornene, as well as the oligomerization of ethylene.68

SiMe3 THF Cl N V Ph Cl N N

1.17

Amidinate and guanidinate complexes of the lanthanides have also been pursued as catalysts for olefin polymerization. In 2004, Hessen and co-workers prepared a series of neutral (1.18) and cationic (1.19) am complexes with alkyl substituents on the metal centre, which exhibited a varying range of activity towards ethylene polymerization, depending on the size of the metal.69 The Trifonov group has shown that dimeric lutetium hydrido bis-guan complexes (1.20) can polymerize ethylene, propylene and styrene. This catalyst was novel not only because it was the first report of a dimeric hydrido lanthanide complex containing two guan ligands, but also because it showed activity towards both ethylene and propylene polymerization, an uncommon occurrence with lanthanide catalysts.70 Analogous complexes of yttrium, neodymium, samarium, gadolinium and ytterbium were later prepared and studies revealed that the yttrium and samarium derivatives had the highest activity in ethylene polymerization.71

Ph Ph iPr iPr (Me3Si)2N N(SiMe3)2 Dipp Dipp Dipp Dipp N N N N N N i iPr N H N Pr M M Lu Lu i i (THF)n (THF)n Pr N H N Pr N SiMe3 SiMe3 N iPr iPr N(SiMe ) Me3Si (Me3Si)2N 3 2

1.18 1.19 1.20 M = Sc, Y, Gd, Lu, La, Nd

In the past five years, significant progress has been made towards the use of metal am and guan complexes for the preparation of thin films. The Barry group has investigated aluminum ams72,73 and copper(I) ams74 and guans75 as potential precursors for atomic layer deposition (ALD). Carmalt and co-workers have studied titanium76 and zirconium77 guans for use in low pressure chemical vapour deposition (CVD). The volatility of amidinate complexes of numerous first row transition elements has been examined.78,79 Additionally, tantalum am complexes have recently shown promise for both CVD and ALD applications.80

Lanthanide complexes of am and guan ligands have also generated interest in this area. For example, highly pure Er2O3 films can be grown using atomic layer deposition

t t of the volatile homoleptic am complex, Er[ BuNC(CH3)N Bu]3 with ozone as the oxygen source,81 and homoleptic guan complexes of gadolinium and dysprosium have been used to prepare films of Gd2O3 and Dy2O3 by metal-organic chemical vapour deposition

(MOCVD), using molecular oxygen as the reactive gas.82,83 Additionally, GdN thin films have been prepared from homoleptic guan complexes, signifying the first time a single source precursor has been used for the deposition of a rare-earth nitride by MOCVD.84

Although still in its early stages, initial studies indicate that with the synthesis of new, highly volatile complexes, this field will continue to evolve and become another important area of application of these types of complexes.

1.5 Boraamidinates

1.5.1 General Considerations

Boraamidinates85 (bams) (1.21) are dianionic, boron- and nitrogen-containing ligands that are isoelectronic to ams and guans. Replacing the –CR fragment in the ligand backbone with a –BR unit imposes two unique consequences: fewer anionic ligands are necessary for stabilizing higher oxidation states and novel redox behaviour is observed

(see Section 1.5.4). The bam ligand demonstrates a wide range of bonding modes, with the most common being a single bam chelated to one metal centre (iv); a monomeric bam complex can dimerize via a nitrogen contact with a second metal centre (v). The second most common type of bonding in bam complexes arises when two bam ligands chelate the metal in a spirocyclic fashion, forming a bis-bam complex (vi). When tris-bam complexes are formed (vii), the three bam ligands coordinate to the metal centre in a pseudo-octahedral geometry. Finally, three types of bridging can occur in bam complexes: one bam can coordinate two metal centres (viii), two bams can bridge two metal centres (ix) and three bams have been shown to bridge across a triply-bonded metal core86 (x).

2- R' R' R R N N B RB N M NR' RB M BR B R'N NR' R'N M N BR N N R'N NR' M R' R' R' 1.21 iv v vi R, R' = alkyl, aryl R B BR R'N R' R' N M R'N NR' R' N M N NR' RB NR' M RB M M RB R'N N NR' N M R'N BR NR' R' R' RB NR' R'N M NR' R'N BR B vii viii ix x R

As with ams, guans and many other ligand systems, both the electronic and steric properties of bams can be tuned by varying the substituents on either nitrogen or boron.

However, compared to the carbon analogues, the chemistry of the bam ligand has been less thoroughly investigated, despite bams being known for over thirty years. In 1979,

Fuβstetter and Nöth reported the first metal complex of a boraamidinate ligand,

{[MeB(NSiMe3)2]Sn}2, which was dimeric in the solid state as determined by single crystal X-ray analysis.87 Almost a decade later, Niecke and co-workers isolated and structurally characterized an alkylimino phosphorus(V) bam, t t t t 88 ( BuB(N Bu)2)P(=N Bu)( Bu). Throughout the 1990s, several boraamidinate complexes of groups 489 and 14-1690-100 were reported. The following sections detail more recent research into main group (Section 1.5.2) and transition metal bams (Section 1.5.3), headed by the groups of Chivers and Nocera, respectively.

1.5.2 Highlights of Main Group bam Complexes

The first structurally characterized examples of the bam ligand in s-block

t chemistry were prepared from the reaction of the tris-aminoborane B[N(H)( Bu)]3 and three equivalents of MeLi or nBuLi.101 The alkyl lithium acted not only as a double deprotonating reagent but also nucleophilically displaced the third amino group on boron, confirmed by the isolation of [Li][N(H)(tBu)] from the reaction. This represented a new synthetic route to bam ligands that was also extendable to other lithium reagents, namely tBuLi and PhLi.102

t The dilithio bam [Li2][PhB(N Bu)2] was shown to undergo metathetical reactions with phosphorus and tellurium halides, generating the first examples of bam complexes with an element-halide bond, which could be further functionalized.102 This bam was also used, in varying stoichiometry, to form complexes of the group 13 elements gallium and

103 indium. The equimolar reaction of the dilithio ligand and InCl3 proceeded to give the

t mono-bam dimeric species, [µ-LiCl(THF)2][PhB(µ-N Bu)2InCl]2 (1.22), with incorporation of one equivalent of the LiCl by-product. However, in the case of gallium,

t an inseparable mixture of [PhB(µ-N Bu)2GaCl]2 (1.23) and the spirocyclic complex

t [Li][{PhB(µ-N Bu)2}2Ga] (1.24) was obtained. The 1:1 product was formed using a step-

t wise synthetic approach: the reaction of GaCl3 and [Li2][PhB(N Bu)2] was first performed in 2:1 stoichiometry, producing the ion-separated complex [Li(THF)4][PhB(µ-

t t N Bu)2GaCl2•GaCl3] (1.25) to which was added a second equivalent of [Li][PhB(N Bu)2].

t In both the indium and gallium cases, the 2:1 [Li2][PhB(N Bu)2]:MCl3 reaction gave the expected spirocyclic complexes.

t Bu t N Bu Cl Cl N In BPh t Ga BPh Bu t N N Bu t N N Bu t PhB In Bu Cl PhB Ga N Cl N Cl tBu Li(THF)2 tBu 1.22 1.23

(OEt2) tBu Li N Cl tBuN NtBu Ga BPh [Li(THF)4] PhB M BPh Cl N N N t t t Bu Bu Bu GaCl3 1.24 1.25 M = Ga, In

In 2004, the Chivers group reported the synthesis and structure of the first monomeric dilithio boraamidinate, with bulky 2,6-diisopropylphenyl (Dipp) substituents on both nitrogen centres, as well as the first bam with different substituents on the

104 nitrogen atoms. The dilithio bam [Li2][PhB(NDipp)2] was used in a metathetical reaction with InCl3 in 2:1 stoichiometry to produce a spirocyclic complex, which differed in the solid state from the N-tert-butyl analogue previously reported102 in that the complex crystallized as an ion-separated pair, [Li(Et2O)4][{PhB(NDipp)2}2In], as opposed to having a bridging Li+ cation. This report also included the synthesis of a mono-lithiated hybrid bam/am ligand with a five-membered backbone; later, bamam complexes (for example, 1.26) of magnesium were structurally characterized and their ability to polymerize rac-lactide was investigated.105

nBu tBu

tBu N

B N

Ph Dipp

1.26

t 2- For heavier group 15 elements, complexes of the bam ligand [PhB(N Bu)2] with novel bonding modes have been observed.106 For example, 2:3 complexes of the form bam3M2 with arsenic and antimony have solid-state structures consisting of one bam bridging two [bamM]+ units (1.27).

Ph B tBuN NtBu tBuN t M M N Bu PhB BPh N tBuN tBu 1.27 M = Sb, Bi

Thallium-thallium interactions lead to a unique solid state structure of

i 107 [Tl2][PhB(N Pr2)2], reported by Nocera and co-workers. Whereas the tert-butyl

t derivative [PhB(N Bu)2][Tl2] crystallizes in the same manner as the dimeric dilithio starting material with weak Tl−Tl contacts between monomeric units, the less-bulky isopropyl substituents allow for an increased number of Tl−Tl contacts and Tl−arene interactions, resulting in the aggregation of four dithallium entities.

1.5.3 Transition Metal bam Complexes

Unlike the elements of the s- and p-blocks, recent examples of transition metal bam complexes are limited. Nocera and co-workers have published the synthesis and X-

t 2- 108,109 ray structures of group 4 and 5 complexes of the bam ligand [PhB(N Bu)2] . The

t titanium complex [PhB(N Bu)2]TiBz2 (Bz = CH2C6H5) was investigated as a precursor for the polymerization of 1-hexene but no activity was observed;109 the authors rationalized this as being a result of either olefin insertion into the Ti−Bz bond or the increased electrophilic nature of the Ti centre, a consequence of the boron-containing ligand.

Boraamidinate complexes of the group 6 metals molybdenum and tungsten have also been prepared and structurally characterized.86 The 2:3 complexes of the form

i M2[PhB(NR)2]3 (R = Et, Pr), which feature a metal-metal triple bond, were generated from the reaction of the dilithiated bam with MoCl3(dme) (dme = dimethoxyethane) or

NaW2Cl7. In the solid state, the bam ligands are in an eclipsed conformation about the metal core (cf. x, Section 1.5.1) and the shortest metal-metal bond lengths for a neutral

M2X6 species, to date, were observed.

1.5.3 Novel Redox Behaviour

Novel redox behaviour can be observed in complexes of the boraamidinate ligand.

The isolation of stable spirocyclic group 13 radicals was first communicated by the

110 t Chivers group in 2005. One electron oxidation of [Li][{PhB(N Bu)2}2M], (1.24, M =

Al, Ga), with iodine generated paramagnetic, spirocyclic complexes 1.28 isolated as dark-coloured X-ray quality crystals that were stable under anaerobic conditions (Eq.

1.1). Electron paramagnetic resonance (EPR) spectroscopy and computational investigations determined that these neutral complexes contained a bam2- ligand as well as the radical anion [bam]•¯, coordinated to the group 13 centre.

(OEt ) 2 t t Li Bu Bu N N t t BuN N Bu 0.5 I2, Et2O PhB M BPh (1.1) PhB M BPh - LiI N N N N tBu tBu tBu tBu

1.24, M = Al, Ga 1.28

These results were followed by a report that detailed the synthesis, structural

t 2- characterization, and oxidation chemistry of a variety of complexes of the [PhB(N Bu)2] ligand, including those of magnesium and zinc.111 Intensely coloured pink and purple solutions were observed for magnesium and zinc, respectively, upon the addition of an

t appropriate amount of iodine to {[(Et2O)µ-Li][PhB(N Bu)2]}2M, but no crystalline product could be isolated. These colours were attributed to the radical complexes

t • 111 [Li{PhB(N Bu)2}2M] . The boron and indium analogues were also investigated; the former radical species, although unable to be isolated pure, was stable for weeks under an inert atmosphere, while the latter was thermally unstable and required the reaction to be maintained at −80°C in order to collect EPR data. Overall, the boron, aluminum and gallium radicals demonstrated greater stability than the magnesium and zinc species, a result of delocalization of the unpaired electron over both bam ligands as opposed to being localized on a single ligand, as shown by DFT calculations. Additionally, the one

t 111 electron oxidation of [Li2][PhB(N Bu)2] with ½ equivalent of I2 was carried out. The

EPR spectrum of the oxidized species was consistent with coupling of the unpaired electron to one lithium centre, two nitrogen atoms and one boron, supporting the formation of a cyclic radical (Eq.1.2).

Ph t Li BuN B 0.5 I2 1/2 Ph B tBuN NtBu (1.2) Et2O N Li Li t - LiI Bu 2 (Et O) 2 x

The redox chemistry of group 14 complexes has also recently been examined.112

t Spirocyclic germanium and tin complexes of the form M[PhB(N Bu)2]2 were treated with various oxidizing agents in attempts to produce radical cations to compare with their neutral, isoelectronic group 13 derivatives. Neither complex reacted with molecular

t •+ iodine, but the tin cation radical {Sn[PhB(N Bu)2]2} was generated upon the addition of

SO2Cl2; however, the analogous germanium radical was not obtained. Although the tin radical complex exhibited a thermal stability similar to that of the previously characterized neutral indium radical, the EPR spectral simulation revealed that the nature of the radical was more similar to that of the magnesium and zinc systems, i.e. localization of the unpaired electron on one bam ligand.112

In contrast to the group 13 bam radicals 1.28, DFT calculations predicted that the

SOMO of the analogous paramagnetic species for the group 15 elements P, As, Sb would be localized on the central pnictogen atom.113 To examine this experimentally, the 2:1

t complexes [Li][{PhB(N Bu)2}2E] (E = P, As) were prepared and treated with different

113 oxidizing agents. The reaction of SO2Cl2 with the phosphorus complex resulted in

27 multiple products (by 11B and 31P NMR), although none were EPR-active species.

t However, the addition of sulfuryl chloride to [Li][{PhB(N Bu)2}2As] produced

t t zwitterionic [PhB(N Bu)2]As(N Bu)2B(Cl)(Ph) (1.29), which was the first example of reactivity at the boron centre in a bam ligand.

tBu tBu N N Cl PhB As B N N Ph tBu tBu

1.29

Attempts to generate the target arsenic radical by reduction of 1.29 with various reagents were unsuccessful; additionally, no EPR-active species were detected in situ during electrochemical oxidation and reduction attempts of the anion and 1.29, respectively.

1.6 Objectives and Outline of Dissertation

The primary objective of this work is to expand the chemistry of boraamidinate ligands in two ways: ligand modification and extension to f-block complexes.

First, the bam ligand will be altered by replacing the alkyl or aryl substituent on boron with an amino group to produce a boraguanidinate, bog. Chapter 2 explores this new ligand system, including its synthesis and structural characterization, along with a thorough investigation of its metathetical reactions with main group halides.

In Chapter 3 the serendipitous discovery of a 1,3,2-oxazaborepane is presented.

This seven-membered C4OBN heterocycle is generated from the ring-opening of THF by

28 aminoboron halides. The oxazaborepane contains the NBN framework of a bam ligand with one of the nitrogen atoms involved in the heterocyclic ring.

Chapter 4 highlights further ways in which the bam ligand can be modified, including the addition of a bridging group to form bis-bam ligands and mono-metallation to produce monoanionic bams. The preparation of bis-bam ligands is of interest because of the potential of forming novel biradical species, while the reactivity of mono- metallated bams allows for direct comparison to the monoanionic am and guan ligands.

Additionally, the synthesis and structural characterization of a new bam reagent, K2bam, is discussed. Employing a dipotassiated ligand in metathesis with metal iodides can be advantageous since the salt by-product, KI, is highly insoluble in solvents such as THF.

Finally, this work extends the chemistry of the bam ligand by exploring its coordination chemistry towards the lanthanide metals. Chapter 5 begins by using a dilithio bam to form “ate” complexes of numerous lanthanide(III) metals, through incorporation of the salt by-product, and halide replacement reactions are examined.

Subsequently, the newly prepared K2bam ligand is used to produce neutral lanthanide bam complexes and the metathetical chemistry of these species is explored. A brief investigation of the preparation of 2:1 (ligand:metal) complexes is then described.

Finally, numerous synthetic attempts at the preparation of a samarium(II) boraamidinate complex are detailed.

Chapter Two: Investigations of Boraguanidinate Ligands

2.1 Introduction

Since amidinates and guanidinates are able to stabilize reactive metal centres, (see

Sections 1.4.2 and 1.4.3), careful design of a boraamidinate may afford similar capabilities to this ligand system. In 2006, Bertrand and co-workers reported the synthesis and structural characterization of a stable four-membered carbene (2.1), which contained an NBN fragment with Dipp groups on the nitrogen atoms and a diisopropylamino substituent at the electron-deficient boron centre.114 Electron donation

i from the −N Pr2 group to the empty p-orbital on boron resulted in localization of the π- system on the NCN fragment, preventing the four-membered ring from being antiaromatic. When the substituent on boron was changed to phenyl or an alkyl group, in attempts to promote delocalization of the π-electrons over the entire ring, even the cationic four-π electron carbene precursors were substantially less stable: decomposition occurred in donor solvents and even in weakly coordinating ones such as CH2Cl2 and

CHCl3. These results support the notion that the amino substituent on boron was crucial in stabilizing the system by hindering electron delocalization and, therefore, precluding

30 anti-aromaticity.114 Notably, carbene 2.1 was prepared by reacting dibromo(diisopropylamino)borane with the appropriate formamidine, followed by halide abstraction to close the ring; i.e. the bam-type NBN fragment was completed upon ring closure and not prepared itself (Scheme 2.1).

i SiMe3 Br N Pr2 B i Dipp N N Pr2NBBr2 Dipp Dipp N N Dipp - Me3SiBr H H

- Et SiBr 3 [Et3Si(toluene)][B(C6F5)4] - toluene

i i N Pr2 N Pr2 [K][HMDS] B B TMEDA Dipp N N Dipp Dipp N + N Dipp

- [K(TMEDA)][B(C6F5)4] - H [B(C6F5)4] 2.1

Scheme 2.1 Prior to the work presented here, only a single example of a boraguanidinate, (i.e. a boraamidinate with an amino group as the substituent on boron), [Li2][Ph2NB(NPh)2],

115 was found in the literature. However, this complex readily eliminated [Li][NPh2] in the presence of dimethoxyethane or pyridine and was only characterized on the basis of 11B

115 NMR data. The preparation of a stable dilithio boraguanidinate, [Li2][R2NB(NRʹ2)2]

Li2bog, would provide an opportunity to extend this chemistry to main group elements to form complexes analogous to 2.1, such as cationic group 15 derivatives, for instance.

This chapter begins with the preparation of [PhB(NDipp)2]PCl, bamPCl, accompanied by a discussion of attempted chloride abstraction to form a cationic complex. These efforts were undertaken in order to have a benchmark with which an analogous bog complex could be compared. Subsequently, the synthesis, spectroscopic

i and structural characterization of a boraguanidinate, [Li2][ Pr2NB(NDipp)2] Li2bog, are presented along with alternative synthetic routes to this ligand. A thorough investigation of reactions of Li2bog with main group halides is then detailed, along with reactions with selected lanthanide trihalides. Finally, attempts at complex formation by alkane elimination are outlined, followed by a brief exploration into modifying the amino substituent on boron in the bog ligand.

2.2 Synthesis, Spectroscopic Characterization and X-ray Structure of

[PhB(NDipp)2]PCl, (2.2)

In order to observe the changes imposed by installing an amino group on boron, the phosphorus bam complex [PhB(NDipp)2]PCl (2.2) was first targeted to serve as a comparison point for when the analogous phosphorus bog complex was prepared. Group

t 2- 15 complexes of the bam ligand [PhB(N Bu)2] have been reported and structurally

102,106 t characterized, including the phosphorus complexes [PhB(N Bu)2]PCl and

t [PhB(N Bu)2]PBr, although the molecular structure of the former was not determined by

X-ray analysis;102 however, the analogous complexes of the bulky ligand

2- i [PhB(NDipp)2] are unknown. The equimolar reaction of [Li2][PhB(NDipp)2], Li2bam, and PCl3 in Et2O followed by extraction in hexane to remove LiCl, affords

[PhB(NDipp)2]PCl (2.2) as a yellow oil that solidifies overnight to a pale yellow solid.

Both the 11B and 31P NMR spectra display a single resonance at 32.8 ppm and 189 ppm, respectively, which is consistent with the values observed for the tert-butyl analogue. X- ray quality crystals were grown from a concentrated hexanes solution, cooled to ca.

−18°C; the molecular structure is illustrated in Figure 2.1 and selected bond lengths and angles are found in Table 2.1.

Figure 2.1. Thermal ellipsoid plot (30% probability) of [PhB(NDipp)2]PCl, 2.2. For clarity, hydrogen atoms have been omitted and only α-carbon atoms of Dipp groups are shown.

i The only bam reagent used for syntheses in this thesis is [Li2][PhB(NDipp)2], abbreviated Li2bam.

Table 2.1. Selected bond lengths (Å) and bond angles (o) for 2.2.

B1−N1 1.447(3) B1−N1−P1 91.0(1)

B1−N2 1.454(3) C7−N1−B1 134.8(2)

P1−N1 1.712(2) C7−N1−P1 134.2(1)

P1−N2 1.716(2) B1−N2−P1 90.6(1)

P1−Cl1 2.099(1) C19−N2−B1 132.5(2)

N1−B1−N2 98.0(2) C19−N2−P1 129.4(1)

N1−B1−C1 131.7(2) N1−P1−N2 79.4(1)

N2−B1−C1 130.1(2) N1−P1−Cl1 103.4(1)

N2−P1−Cl1 104.2(1)

The B−N bond lengths in 2.2 are intermediate between a single and double bond

t 102 and, within experimental error, are identical to those found for [PhB(N Bu)2]PBr. The

P−N bond distances, however, are slightly elongated (by ca. 0.03 Å) in 2.2 when compared to the tert-butyl analogue. A search of the Cambridge Structural Database reveals that the P−Cl length falls at the lower end of the range expected for this type of bond. The geometry about boron and N1 is planar (∑ angles = 360°), however, the angles about N2 sum to approximately 353°. Despite the Dipp substituents being more sterically demanding than a tert-butyl group, the angles within the PNBN ring in 2.2 are

t 102 consistent with those in [PhB(N Bu)2]PBr. Additionally, this ring is slightly puckered, as indicated by B1−N1−P1−N2 torsion angle of ca. 8.3°, which was also found in the tert-butyl complex.

The approaches used in attempts to remove the chloride ligand to form a phosphenium cation are summarized in Scheme 2.2.

GaCl3 Ph Multiple 31P NMR resonances

B [Li][B(C6F5)4] Dipp N N Dipp No reaction P [Et Si][B(C F ) ] 3 6 5 4 !31P not in range of Cl phosphenium cations

Scheme 2.2

The use of gallium trichloride as the halide-abstracting reagent in CD2Cl2 resulted in numerous species, as indicated by multiple resonances in the 31P NMR spectrum.

When salt metathesis with [Li][B(C6F5)4] was performed, the predominant species in the

31 P NMR spectrum was starting material, whereas if the silylenium salt [Et3Si][B(C6F5)4] was employed, a single new phosphorus-containing product was generated. However, the

31P NMR chemical shift of this product at 79 ppm was not indicative of a phosphenium cation116 and the identity of this species was unable to be confirmed.

The difficulties encountered in forming a phosphenium cation containing a bam ligand can be rationalized by taking into account the electronic effects imposed by the boron atom in the ligand backbone. As evidenced by the B−N bond lengths in 2.2, the nitrogen atoms donate electron density to the boron centre through N(2p)→B(2p) π- bonding, therefore decreasing the availability of the lone pairs to stabilize the cationic phosphorus centre. This explanation is illustrated conceptually in Figure 2.2, using a general representation of one type of phosphenium cation (left) as well as Bertrand’s

35 carbene, 2.1 in comparison to the structure of 2.2. These results provide further motivation to explore the synthesis of a boraguanidinate since the exocyclic amino group should provide similar stabilizing effects as observed in 2.1; the synthesis of such a ligand is detailed in the following two sections.

Ph R' R' NR2 B B N N Dipp N N Dipp VS Dipp N N Dipp R R P P

X Cl 2.1 2.2

Figure 2.2. A comparison of electron donation in phosphenium cations (left), carbene 2.1 and bamPCl 2.2.

2.3 Synthesis, Spectroscopic Characterization and X-ray Structures of i i ( Pr2N)B[N(H)Dipp]2 (2.3) and {[Li2][( Pr2N)B(NDipp)2]}2 (2.4)

Since the dilithio boraamidinate reagent [Li2][PhB(NDipp)2] has been prepared by reaction of PhBCl2 with an excess of [Li][N(H)Dipp], followed by double deprotonation with nBuLi,104 a similar synthetic approach to dilithio boraguanidinate

i [Li2][ Pr2NB(NDipp)2] was proposed. The reaction of dichloro(diisopropylamino)borane and two equivalents of [Li][N(H)Dipp] in hexanes proceeded cleanly to give i ( Pr2N)B[N(H)Dipp]2, bogH2 (2.3), as a colourless solid, in high yields, after filtration and removal of volatiles (Scheme 2.3). The boraguanidinate precursor 2.3 was

36 characterized by multinuclear NMR spectroscopy (1H, 11B, 13C), CHN analysis and single crystal X-ray diffraction.

i i N Pr2 N Pr2 2 nBuLi Dipp Dipp hexanes B hexanes i B + 2 [Li][N(H)Dipp] N N 0.5 {[Li2][ Pr2NB(NDipp)2]}2 Cl Cl - 2 LiCl - 2 H H 2.3 2.4

Scheme 2.3

The air-stability of 2.3 is exemplified by the fact that X-ray quality crystals can be grown from slow evaporation of a concentrated hexane solution either in the presence or absence of air; the molecular structure of bogH2 is shown in Figure 2.3. There is no

i significant difference in the B−N Pr2 and B−N(H)Dipp bond lengths, which are intermediate between single and double bond values, as expected for N(2p)→B(2p) π- bonding. Similar to PhB[N(H)Dipp]2, the boron atom in 2.3 is planar (∑ angles is 360°) but distorted from a trigonal geometry, with angles ranging from approximately 116-

123°. The nitrogen atoms also display distorted trigonal geometries, having bond angles ranging from ca. 114-131°, though the sum of the angles about each nitrogen is 360°.

i Figure 2.3. Thermal ellipsoid plot (30% probability) of ( Pr2N)B[N(H)Dipp]2, 2.3. For clarity, hydrogen atoms other than those on nitrogen have been omitted and only α- carbon atoms of Dipp groups are shown. Selected bond lengths (Å) and angles (°): B1−N1 1.438(2), B1−N2 1.437(2), B1−N3 1.429(2); N1−B1−N2 116.0(1), N1−B1−N3 120.7(1), N2−B1−N3 123.3(1).

Deprotonation of 2.3 with two equivalents of nBuLi in hexanes produces

i [Li2][( Pr2N)B(NDipp)2], Li2bog (2.4), the first example of a stable dilithio boraguanidinate. During the synthesis of 2.4, the effect of the diisopropylamino group is immediately observed as 2.4 does not precipitate from the hexane solution. In contrast, during the preparation of Li2bam hexanes are used to wash the crude material as the dilithio bam has very limited solubility in this hydrocarbon solvent. The novel dilithio boraguanidinate was characterized by multinuclear NMR spectroscopy (1H, 7Li, 11B, 13C),

CHN analysis and X-ray diffraction. X-ray quality crystals of Li2bog can be grown from an air-protected concentrated toluene/hexanes solution, cooled to ca. −18°C. The molecular structure of 2.4 is shown in Figure 2.4 and selected bond lengths and angles are presented in Table 2.2, along with the corresponding parameters for selected dilithio bams. Additionally, pertinent 1H NMR and 11B NMR data are found in Tables 2.3 and

2.4, respectively.

Table 2.2. Selected bond lengths (Å) and bond angles (°) for 2.4 and selected dilithio bams.

n t 2.4 {[Li2][ BuB {[Li2][ BuB {[Li2][PhB [Li(THF)3]

t 101 t 101 t 102 104 (N Bu)2]}2 (N Bu)2]}2 (N Bu)2]}2 [PhB(NDipp)2] B1−N1 1.471(3) 1.455(4) 1.476(4) 1.448(3) 1.415(5)

B1−N2 1.470(3) 1.464(4) 1.479(4) 1.449(3) 1.446(4)

B1−N3 1.463(3) N/A N/A N/A N/A

Li1−N1 2.055(5) 2.032(5) 1.980(4) 2.077(4) 2.022(6)

Li1−N2 2.121(5) 2.039(6) 2.154(5) 2.052(4) 2.010(6)

Li1−N2A 2.144(5) 2.027(5) 2.125(4) 2.017(4) 2.094(7)

Li2−N1 2.063(5) 2.032(5) 2.154(5) 2.027(4) 2.115(6)

Li2−N2A 2.063(5) 2.022(6) 1.976(4) 2.022(4) N/A

Li2−N1A 2.145(5) 2.091(5) 1.976(4) 2.074(4) N/A

N1−B1−N2 109.1(2) 108.2(2) 105.2(2) 109.5(2) 111.4(3)

N1−B1−N3 124.8(2) N/A N/A N/A N/A

N2−B1−N3 126.1(2) N/A N/A N/A N/A

Boraguanidinate 2.4 forms a dimer in the solid state, which has previously been observed for bam ligands that could be crystallized in the absence of a coordinating solvent.101,102 The B−N bond distances are slightly elongated (by ca. 0.03 Å) with respect to those in diprotio bog 2.3 (see caption to Fig. 2.3), but still fall in the expected range. Although the sum of the bond angles about boron in 2.4 is 360°, as it is in 2.3, the geometry is further distorted from trigonal, with the N1−B1−N2 angle being

40 approximately 7° smaller than in the neutral precursor, presumably a result of the constraints imposed by the dimeric nature of the structure.

Although the ideal structural comparison to establish the effect of the amino group on boron would be between 2.4 and [Li2][PhB(NDipp)2], the dilithio bam is a monomer in the solid state owing to solvation of the lithium cations by THF, a consequence of crystals being grown from this coordinating solvent. As such, differences in the structural parameters between Li2bog and Li2bam are evident: the B−N bond lengths are elongated and the N1−B1−N2 angle is slightly narrower in 2.4 when compared to the dilithio bam (109.1° versus 111.4°). The former observation is a consequence of the third amino substituent on boron, while the latter stems from the limitations imposed by dimer formation. Additionally, the N−Li distances are, on average, slightly longer in the boraguanidinate ligand.

t When compared to other unsolvated, dimeric dilithio bams, {[Li2][RB(N Bu)2]}2

(R = nBu, tBu, Ph), 2.4 displays many structural similarities, despite the more bulky Dipp substituent on the nitrogen atoms. The B−N bond lengths in 2.4 are not significantly

t n t different than those found in dimeric B−alkyl bams {Li2[RB(N Bu)2]}2 (R = Bu, Bu), however, they are slightly elongated when compared to those in the B-phenyl derivative

t {Li2[PhB(N Bu)2]}2. This difference is attributed to the fact that the phenyl substituent is electron-withdrawing, in comparison to the electron-donating alkyl or diisopropylamino groups. A similar trend is found for the N−Li bond lengths, although all fall into the range expected for three-coordinate lithium centres. Finally, the N1−B1−N2 bond angle

n t t in 2.4 is the same as that in both {[Li2][ BuB(N Bu)2]}2 and {[Li2][PhB(N Bu)2]}2, within

41 experimental error. Reactions of Li2bog with a variety of main group halides, as well as selected lanthanide trihalides, will be discussed in Section 2.5.

2.4 Alternative Syntheses of Li2bog: Synthesis and Characterization of

ClB[N(H)Dipp]2 (2.5), BrB[N(H)Dipp]2 (2.6) and B[N(H)Dipp]3 (2.7)

Alternative synthetic approaches are possible for the preparation of Li2bog. In the first approach, bis[(Dipp)amino]chloroborane was synthesized and subsequently treated

i with three equivalents of [Li][N Pr2] (LDA) in order to install the third amino group (with loss of LiCl) and deprotonate both −N(H)Dipp groups in a one-pot process (Scheme 2.4).

i 3 [Li][N Pr2] hexanes hexanes i BX3 + 4 DippNH2 XB[N(H)Dipp]2 0.5 {[Li2][ Pr2NB(NDipp)2]}2 - 2 [DippNH3][X] - LiX i 2.4 2.5: X = Cl - 2 HN Pr2 X = Cl, Br 2.6: X = Br

Scheme 2.4

The reaction of BCl3 and four equivalents of DippNH2 in hexanes results in the precipitation of [DippNH3][Cl], producing ClB[N(H)Dipp]2 (2.5) as a colourless solid in good yields (72%) after filtration and removal of solvent. It is essential to use the excess amine approach rather than performing the reaction of BCl3 with two equivalents of

[Li][N(H)Dipp)] in THF, as the latter reaction results in ring-opening of the solvent, which is the topic of Chapter 3.

Although attempts to grow X-ray quality crystals of 2.5 gave only fibrous needles that were not suitable for diffraction, multinuclear NMR spectroscopy, elemental analysis and high resolution mass spectrometry confirm its identity. In addition to the aryl

42 resonances for the Dipp groups, the 1H NMR spectrum demonstrates two broad singlets integrating to 1 proton each (−NH), as well as two 2 H broad singlets (−CH(CH3)2) and overlapping doublets integrating to 24 H total, indicative of diastereotopic isopropyl groups. Additionally, the 11B NMR spectrum exhibits a single resonance at 26.4 ppm,

i which is downfield-shifted when compared to ( Pr2N)B[N(H)Dipp]2 (δ 23.4).

Since the molecular structure of 2.5 was unable to be determined by single crystal

X-ray analysis, the bromo derivative, BrB[N(H)Dipp]2 (2.6), was prepared by the

1 analogous reaction of BBr3 and four equivalents of DippNH2 (Scheme 2.4). The H NMR spectrum of 2.6 was comparable to that of 2.5 and the 11B NMR spectrum revealed the expected small upfield shift of the 11B resonance (25.7 vs. 26.4 ppm). X-ray quality crystals were grown from a concentrated solution of 2.6 in hexanes and the molecular structure is shown in Figure 2.5.

When comparing the structures of 2.6 and bogH2 (2.3), a slight shortening of the boron-nitrogen distances is observed, which is attributed to the decrease in π-donation from the bromine atom in comparison to the diisopropylamino group. Within experimental error, the B−Br distance of 1.951(6) Å is the same as is found in a bromo- diazaborole,117 which provides a comparable environment about boron (i.e. three- coordinate boron containing a bromine and two nitrogen substituents). The boron centre in 2.6 is planar, with angles ranging from ca. 117-124°, which is comparable to what is found in 2.3.

The addition of a solution of either chloroborane 2.5 or bromoborane 2.6 in toluene to three equivalents of LDA at 0°C yields a pale yellow solid, which was shown by the 1H NMR spectrum to primarily be comprised of the dilithio bog 2.4, together with a minor unidentified impurity. The haloboranes 2.5 and 2.6 are potentially useful reagents for the preparation of other bam or bog ligands via reactions with alkyl-, aryl- or amido- lithium reagents.

The second alternative synthetic route to Li2bog 2.4 that was pursued was to prepare B[N(H)Dipp]3 (2.7) and use three equivalents of LDA to both deprotonate two of the −N(H)Dipp groups and nucleophilically displace the remaining −N(H)Dipp group

t (Scheme 2.5). This method proved successful in the preparation of {[Li2][RB(N Bu)2]}x

n t (R = Me, x = 2,3; R = Bu, x = 2), where B[N(H) Bu]3 was treated with three equivalents of the appropriate alkyllithium reagent.101

hexanes i B[N(H)Dipp]3 + 3 LDA 0.5 {[Li2][ Pr2NB(NDipp)2]}2 2.7 - [Li][N(H)Dipp] 2.4 i - 2 Pr2NH

Scheme 2.5

The tris(amino)borane 2.7 is easily prepared from the reaction of BCl3 and six equivalents of DippNH2 in hexanes. After filtration and removal of volatiles, 2.7 is isolated as a sticky, colourless solid in excellent yields (>90%). Multinuclear NMR spectroscopy supports the replacement of all three chloride ligands: in the 1H NMR spectrum the septet resonances are clearly resolved (in comparison to the broadened signals observed for disubstituted 2.5 and 2.6) and an upfield shift of ca. 3-4 ppm is observed in the 11B NMR spectrum, with respect to the corresponding haloboranes 2.5 and 2.6 (22.6 ppm in 2.7). The molecular structure was established by single crystal X- ray analysis of crystals grown from a concentrated solution of 2.8 in hexanes and is illustrated in Figure 2.6.

The B−N distances in 2.7 are similar to those found for both diprotio bam

104 PhB[N(H)Dipp]2 and bogH2. The boron centre is almost perfectly trigonal planar with

N−B−N bond angles deviating from 120° by one degree or less. To minimize steric interactions, two of the −N(H)Dipp groups are oriented essentially perpendicular to the

N−B−N plane, with torsion angles of 93.0° for B1−N1−C1−C2 and 88.6° for

B1−N2−C13−C14; the third −N(H)Dipp substituent is bent even further with a

B1−N3−C25−C30 dihedral angle of 102.1°.

The addition of three equivalents of LDA to 2.7 in hexanes produces 2.4 by 1H

NMR as the major product, in addition to minor impurities. Consequently, although both of the above-mentioned alternative synthetic approaches can be used to prepare Li2bog, it is preferable to use the initial synthesis from dichloro(diisopropylamino)borane depicted

46 in Scheme 2.3, as that method gives a clean product without the need for purification by recrystallization.

2.5 Reactions of Li2bog with p-Block Halides

The dilithio boraguanidinate 2.4 was reacted with numerous p-block and group 12 halides, as well as selected lanthanide halides, to compare its behaviour with guanidinate ligands. While simple metathesis was observed in certain cases, Li2bog was found to be an extremely strong reducing agent in most instances. Both of these facets will be detailed in the following sub-sections. Attempts to form bog complexes by deprotonation of bogH2 with metal alkyls are also described briefly.

2.5.1 Metathesis

Although Li2bog predominantly displays reductive behaviour towards main group halides, as will be discussed in Section 2.5.2, there were a few cases where simple metathesis occurred in the reactions of 2.4 with p-block, group 12 or lanthanide halides.

The first example was the reaction of Li2bog and PCl3, which was chosen as the starting point in order to observe the effect of the amino group by comparison to the analogous phosphorus bam complex, 2.2.

The equimolar reaction of Li2bog and PCl3 in Et2O gives a yellow oil that solidifies overnight at room temperature. The 31P NMR spectrum of the resulting solid contains two major resonances of approximately equal intensities, at 189 ppm and 164 ppm. The former chemical shift is assigned to the desired phosphorus bog complex,

47 i [ Pr2NB(NDipp)2]PCl, by analogy to bam complex 2.2, while the identity of the latter was unable to be conclusively determined. The production of this second species was reproducible and the two compounds could not be separated by extraction, crystallization or sublimation. Given the fact that the products were present in equal amounts in the crude solid, halide abstraction to produce a cationic group 15 analogue of 2.1 was not attempted and, instead, other main group complexes were pursued.

The reactions of 2.4 with certain main group dihalides, namely GeCl2·dioxane,

MgCl2, and ZnCl2, in THF, proceeded cleanly to give products that are tentatively

i assigned, on the basis of their NMR data, as {Ge[ Pr2NB(NDipp)2]}2·dioxane (2.8) and

M(THF)x[PhB(NDipp)2] (2.9, M = Mg; 2.10, M = Zn). Metathesis is supported by the absence of any metallic precipitate, which would be anticipated from reduction (see

Section 2.5.2). The 1H NMR data for the isopropyl groups of 2.8-2.10 are summarized in

Table 2.3 and compared to those of 2.3 and 2.4; 11B NMR data are presented in Table 2.4.

Table 2.3. 1H NMR chemical shifts (in ppm) for the iPr groups in 2.3, 2.4, and 2.8-2.10

(in THF−d8).

2.3 2.4 2.8 2.9 2.10

−CH(CH3)2 of 3.61 3.84 3.80 4.15 4.15 Dipp groups

−CH(CH3)2 of 3.25 3.65 3.26 3.43 3.30 i −N Pr2 groups

−CH(CH3)2 of 1.22, 1.16 1.20, 1.07 1.30, 1.16 1.18, 1.09 1.21, 1.13 Dipp groups

−CH(CH3)2 of 1.12 0.92 0.86 0.76 0.76 i N Pr2 groups

The methyl and methyne resonances of both the Dipp and amino substituents in

2.8-2.10 are significantly shifted with respect to those in bogH2 2.3; consequently, the presence of even small amounts of the diprotio ligand in the reaction products is readily detected. Attempts to separate LiCl from the reaction products using non-coordinating solvents, e.g. toluene or hexane, consistently resulted in the formation of the diprotonated derivative 2.3. This observation is surprising in view of the lack of decomposition of the dilithio derivative 2.4 in toluene or hexane.

11 Table 2.4 B NMR chemical shifts (in ppm) for 2.3, 2.4, and 2.8-2.10 (in THF-d8).

2.3 2.4 2.8 2.9 2.10

23.4 22.8 28.0 23.2 25.2

The complexes 2.8, 2.9 and 2.10 all show broad singlets in the 11B NMR spectra at 28.0, 23.2 and 25.2 ppm, respectively, cf. 22.8 ppm for the dilithio derivative 2.4. A distinct downfield shift in these resonances is evident upon the replacement of an electropositive metal (Li or Mg) by zinc or germanium.

Figure 2.7 illustrates the proposed structures of complexes 2.8-2.10. The integration of the 1H NMR resonances was used to estimate the solvation in

i {Ge[ Pr2NB(NDipp)2]}2·dioxane (2.8); on this basis, a structure in which the dioxane molecule bridges two germanium centres is suggested. Resonances for THF were also

49 evident in the 1H NMR spectra of the magnesium and zinc complexes and it is most likely that two solvent molecules are coordinated, which would complete the tetrahedral coordination sphere of the metal centre. Attempted crystallizations of 2.9 and 2.10 in the presence of coordinating ligands such as TMEDA or PPh3 gave either glassy non- crystalline materials or decomposition to 2.3.

i Dipp Dipp Pr2N

N N B i i Pr2N B Ge O O Ge B N Pr2 Dipp N N Dipp N N M

Dipp Dipp O O

2.8 2.9, M = Mg 2.10, M = Zn

Figure 2.7. Proposed structures of bog complexes 2.8-2.10.

The reactions of Li2bog with group 13 trihalides were also undertaken, with the most promising results obtained from using AlCl3. In this instance, the 1:1 reaction gave a single new product by 1H NMR; however, as in the cases of main group dihalides, attempts to remove the LiCl by-product were unsuccessful and only bogH2 (2.3) was recovered. When BBr3 or GaCl3 were employed there was typically one major product with numerous impurities present.

Given the propensity for lanthanide complexes to exist in the 3+ oxidation state, it was proposed that these elements may be well-suited for bogLi2 as they should also undergo metathesis rather than reduction. Consequently, reactions with 2.4 and selected f- block halides were performed in a 1:1 stoichiometry on both an NMR and preparative

50 scale in deutero- and protio-THF, respectively. As anticipated, reduction did not occur and in the cases of yttrium and lanthanum, metathesis products

i i {[ Pr2NB(NDipp)2]YCl(THF)n}m (2.11) and {[ Pr2NB(NDipp)2]LaI(THF)n}m (2.12) were able to be assigned on the basis of 1H NMR spectra. Both yttrium(III) and lanthanum(III) are diamagnetic, therefore the proton spectra are well-resolved; selected data are summarized in Table 2.5 and compared with 2.4. Small amounts of bogH2 were also present in the 1H NMR spectrum of the product obtained from the larger scale reactions

(in protio THF), however, washing with hexanes removed most of this impurity. The ionic nature of complexes formed by the lanthanides was expected to promote crystallization; however, no crystals were obtained from either of these reactions.

1 Table 2.5. Selected H NMR chemical shifts (in THF-d8) for the reactions of 2.4 with

LnX3

2.4 2.11 (Ln = Y, 2.12 (Ln = La, X = Cl) X = I) −CH(CH3)2 of 3.84 4.09 4.01 Dipp groups

−CH(CH3)2 of 3.65 3.53 3.45 i −N Pr2 groups

−CH(CH3)2 of 1.20, 1.07 1.23, 1.18 1.29, 1.19 Dipp groups

−CH(CH3)2 of 0.92 0.78 0.75 i N Pr2 groups

The reaction of 2.4 and an equimolar amount of either NdCl3 or PrCl3 requires heating to ca. 60°C in order to proceed. In both cases, resonances in the 1H NMR

51 spectrum were observed at extremely shielded values (i.e. below -10 ppm), which is expected because of the paramagnetic nature of the metal centres; however, more than one species existed in solution and conclusive assignment of a bogLnCl complex was not possible. However, successful structural characterization of lanthanide(III) complexes of

2- the boraamidinate ligand [PhB(NDipp)2] will be the topic of Chapter 5.

2.5.2 Reduction

The strong reducing nature of 2.4 was observed in the reactions of the dilithio bog with numerous p-block and group 12 metal halides, e.g., SnCl2, PbCl2, TeBr4, TeI4,

CdCl2, and TlCl. In these instances, the respective metal was instantly deposited upon mixing of the reagents either at room temperature or at low temperatures (−80°C). In most cases, after filtering off the metal and removing the solvent, the isolated solid contained two or more major products by 1H NMR spectroscopy.

The behaviour of 2.4 in the reactions with tin(II) and lead(II) dichlorides is in contrast to that of dilithio bams, which produce stable dimeric complexes {Mbam}2 (M =

Sn, Pb) via metathesis.87,100 It is important to note that although the lead dimer

t {Pb[PhB(N Bu)2]}2 is conveniently prepared by the metathetical reaction of PbCl2 and

t [Li2][PhB(N Bu)2, it was also obtained from the reduction of PbCl4 with the dilithio bam

100 reagent. The immediate precipitation of lead metal in the reaction of 2.4 with PbCl2 demonstrates the strong reducing power of the bog ligand, resulting from the influence of the diisopropylamino group, in comparison to the dilithio bams.

The reduction of Tl(I) to thallium metal in the reaction of 2.4 and two equivalents of TlCl is not unprecedented, as this observation was also made by Manke and Nocera in

i t 107 the reaction of [Li2][PhB(NR)2] (R = Pr, Bu) with TlCl. However, these researchers were able to isolate the polymeric thallium(I) bam complexes {Tl2[PhB(NR)2]}∞ in moderate yields from these reactions. By contrast, the major product of the reaction of

1 2.4 and TlCl was identified as bogH2 (2.3) by H NMR spectroscopy.

When 2.4 is reacted with an equimolar amount of indium(I) chloride deposition of indium metal occurs immediately upon mixing of the reagents, even at −80°C. Since the analogous reaction with Li2bam was not previously reported, it too was performed and similar observations were made. After removing the indium metal by filtration, the solid isolated from the bam reaction was determined by 1H NMR spectroscopy to be the previously reported spirocyclic indium(III) complex 2.13,104 implying that disproportionation of the initially formed bamIn(I) complex had occurred (Scheme 2.5).

With this finding, it seems reasonable to suggest that a similar disproportionation takes place in the reaction of Li2bog with InCl. However, attempts to grow X-ray crystals of the product of this reaction were unsuccessful.

2- Ph Ph B + Et2O 2 B 4 [Li] 2 + 2 InCl [Li]+ DippN NDipp - 2 LiCl Dipp N N Dipp In

- Li0 - In0

Dipp Dipp N N Ph B In B Ph [Li]+ N N Dipp Dipp

2.13

Scheme 2.6

2.5.3 Attempted Alkane Elimination

Given the strong reducing nature of the dianionic bog ligand, alkane elimination was chosen as an alternative route to metal complexes. This direct metallation approach has been used for the preparation of the magnesium bam complex

n/s 111 Mg(OEt2)2[PhB(NDipp)2] from PhB[N(H)Dipp]2 and an excess of Bu2Mg.

n However, when the analogous reaction was performed with 2.3 and Bu2Mg, only the neutral bog ligand was recovered, even after heating to 60°C for 24 hours. Subsequently, alkane elimination reactions using dimethyl zinc or trimethyl aluminum were carried out in numerous solvents (hexane, toluene, THF) and at various temperatures, but in all cases the bogH2 remained unreacted.

2.6 Preparation of New Boraguanidinate Ligands

In order to reduce the electron-richness of 2.4, new boraguanidinates with electron-withdrawing groups on the secondary amino substituent were proposed. The two target ligands [Li2][RRʹNB(NDipp)2] (2.14, R = Ph, Rʹ = Me; 2.15, R,Rʹ = Ph) were judiciously chosen to systematically reduce the electron-donating ability of the amino substituent and are illustrated in Figure 2.8.

2- 2- Ph Me Ph Ph N N

B B Dipp N N Dipp Dipp N N Dipp

2.14 2.15

Figure 2.8. Targeted dianionic bog ligands.

The reaction of chloro(diamino)borane 2.6 with [Li][N(Ph)(Me)] in hexanes produces Ph(Me)NB[N(H)Dipp]2, Ph(Me)bogH2 (2.14H2) the diprotio derivative of 2.14.

The reaction can also be performed using bromo(diamino)borane 2.7 to yield the same product. Although X-ray quality crystals were unable to be obtained, Ph(Me)bogH2 was characterized by multinuclear NMR spectroscopy. The proton NMR spectrum has the expected aryl, aliphatic and −NH resonances, integrating to the appropriate intensities, and only one signal (δ 24.7) is found in the 11B NMR spectrum. Additionally, elemental analysis supported the formation of Ph(Me)bogH2.

Dilithiation with two equivalents of nBuLi in hexanes at ca. −30°C proceeds to

1 give a major new species by H NMR, presumed to be the dilithio bog ligand Li22.14, along with a minor impurity. Although Li22.14 was unable to be characterized by X-ray analysis, the reactions of this dilithio bog with selected main group halides give similar results to those involving 2.4. For instance the addition of Li22.14 to PbCl2 in THF at either room temperature or −80°C results in immediate precipitation of lead metal; similarly, indium metal is isolated from the analogous reaction with InCl3. As with bog

2.4, the reaction of Li22.14 with ZnCl2 or GeCl2 gives one product (by proton NMR); however, attempted extraction from the LiCl by-product resulted in decomposition to several species, including the diprotio ligand. In contrast to the clean reaction observed between dilithio bog 2.4 and MgCl2, numerous products are obtained from the analogous reaction using Li22.14, as indicated by multiple methyl resonances in the proton NMR

n spectrum. Additionally, the reaction of diprotio Ph(Me)bogH2 with Bu2Mg under the same conditions used for alkane elimination attempts with 2.4 also resulted in numerous products.

Since the introduction of only one electron-withdrawing Ph group on the amino substituent did not result in a significant change in the electronic properties of the bog ligand, boraguanidinate Li22.15 with two Ph substituents in the secondary amine was pursued. The preferable route to the diprotio derivative (2.15H2) is the reaction of bromo(diamino)borane 2.7 with [Li][NPh2] in hexanes. Unlike the previously mentioned bog ligands, lithiation attempts in both coordinating and non-coordinating solvents did not proceed cleanly and Li22.15 was unable to be prepared. Subsequently, the stronger metallating agent n-butylsodium was employed with the intention of preparing the disodiated ligand. However, despite using two equivalents of the metal alkyl, the proton

NMR spectrum of the resulting product suggested that only mono-metallation to give

[Na][Ph2NB{(NDipp)N(H)Dipp}] (2.16) occurred as indicated by the presence of a broad singlet (−NH) integrating to 1 H and the inequivalence of the isopropyl groups (two septet resonances integrating to 2 H and three doublets with relative intensities of 6:6:12).

2.7 Conclusions

In this chapter, the first example of a stable dilithio boraguanidinate,

i {[Li2][ Pr2NB(NDipp)2]}2, Li2bog, was presented. This boraguanidinate can be prepared from three different synthetic approaches: (i) reaction of the commercially available i Pr2NBCl2 with two equivalents of [Li][N(H)Dipp] followed by double deprotonation

n with BuLi; (ii) preparation of XB[N(H)Dipp]2 (X = Cl, Br), and subsequent reaction with three equivalents of LDA; or (iii) reaction of the tris(amino)borane B[N(H)Dipp]3 with three equivalents of LDA. The first route is preferred as the dilithiation step

56 proceeds cleanly and the Li2bog can be used without further purification. The dilithio boraguanidinate is dimeric in the solid state, similar to dilithio bams that are crystallized from non-coordinating solvents.

A survey of reactions of Li2bog with numerous p-block and group 12 metal halides reveals that the bog ligand is a much stronger reducing agent than the analogous dilithio bam, presumably a consequence of the electron donating influence of the diisopropylamino substituent on boron. Metathetical behaviour was found to occur in only certain cases, e.g. with MgCl2, ZnCl2 and GeCl2, but the resulting bog complexes were unstable with respect to bogH2 formation and the diprotio ligand was the major product recovered from attempted extraction or crystallization in non-coordinating solvents. Boraguanidinate complexes of yttrium and lanthanum were able to be prepared

(on the basis of 1H NMR spectroscopy) and could be washed with hexanes to remove small amounts of bogH2 with retention of product integrity. However, no crystals were obtained to unambiguously confirm the molecular structure of these complexes.

n Protonolysis reactions between bogH2 and ZnMe2, Bu2Mg or AlMe3 would avoid

2 using the strongly reducing bog − ligand in metathesis; however, no reaction occurred between any of these reagents and the diprotio ligand, regardless of reaction solvent or temperature.

In response to the highly reducing nature of Li2bog, replacing the diisopropylamino substituent with a less electron-rich amino group was targeted. A new bog ligand, [Li2][Ph(Me)NB(NDipp)2] was prepared using a halo(diamino)borane,

XB[N(H)Dipp]2 (X = Cl, Br), and [Li][NPh(Me)], followed by deprotonation with two equivalents of nBuLi. This new bog displayed metathetical behaviour similar to that of

Li2bog. Further modification of the electronic properties of the amino group was deemed necessary but the lithiation of Ph2NB[N(H)Dipp]2 produced multiple products and this dilithio bog derivative could not be obtained cleanly.

The boraguanidinate backbone played a crucial role in the preparation and characterization of novel carbene 2.1 by the group of Bertrand. However, the electron-

i 2- rich nature of [ Pr2NB(NDipp)2] , which was not isolated in the synthesis of 2.1, made extending the chemistry of this bog ligand to other main group elements challenging. The strongly reducing behaviour of the bog ligands described in this chapter is in contrast to that observed for guanidinates, which readily form complexes with elements throughout the periodic table.11,118

2.8 Experimental Section

Reagents, general procedures and instrumentation are described in Appendix 1.

Preparation of [PhB(NDipp)2]PCl (2.2). A solution of [Li2][PhB(NDipp)2] (0.726 g,

Ph 1.60 mmol) in Et2O (30 mL) was added to a solution of PCl3 B (0.14 mL, 1.60 mmol) in Et2O (10 mL) at −80°C. The reaction Dipp N N Dipp P mixture was stirred for ½ h at −80 oC and 18 h at room Cl temperature. The volatiles were removed in vacuo and the product was then extracted with n-hexane. The precipitate of LiCl was removed by filtration and the solvent was evaporated under vacuum yielding an oily product that solidified to give 2.2 as a pale yellow powder after 24 h at room temperature (0.648 g,

1.30 mmol, 80%). Anal. Calcd. for C30H39BN2PCl: C, 71.37; H, 7.78; N, 5.55. Found: C,

1 71.35; H, 8.24; N, 5.45. H NMR (toluene-d8, 25°C): δ 7.26-637 (m, 11 H, aromatic),

3 3 4.18 (sept., 2 H, −CH(CH3)2, JH-H = 6.8 Hz), 3.64 (2 H, −CH(CH3)2, JH-H = 6.8 Hz),

3 3 1.37 (d, 6 H, −CH(CH3)2, JH-H = 6.8 Hz), 1.27 (d, 6 H, −CH(CH3)2, JH-H = 6.8 Hz), 1.16

3 3 13 (d, 6 H, −CH(CH3)2, JH-H = 6.8 Hz), 1.06 (d, 6 H, −CH(CH3)2, JH-H = 6.8 Hz). C NMR

2 2 (toluene-d8, 25°C): δ 147.9 (d, aryl, J13C-31P = 7.6 Hz), 145.8 (d, aryl J13C-31P = 6.0 Hz),

3 134.7 (aryl), 134.5 (d, aryl J13C-31P = 14.8 Hz), 132.2 (aryl), 128 (aryl), 127.7 (aryl),

4 124.4 (aryl), 29.72 (d, −CH(CH3)2, J13C-31P = 26.4 Hz), 28.76 (−CH(CH3)2), 25.32 (d,

4 −CH(CH3)2, J13C-31P = 16.8 Hz), 24.78 (−CH(CH3)2), 24.5 (−CH(CH3)2), 23.98

11 31 (−CH(CH3)2). B NMR (toluene-d8, 25°C): δ 32.8. P NMR (toluene-d8, 25°C): δ 189.8.

Colourless crystals of 2.2 were grown from n-hexane after 5 d at 5 oC.

i i Preparation of Pr2NB[N(H)Dipp]2 (2.3). A solution of Pr2NBCl2 (1.50 g, 8.20 mmol)

i N Pr2 in hexanes (10 mL) was added to a stirred slurry of

Dipp B Dipp [Li][N(H)Dipp] (3.05 g, 16.6 mmol) in hexanes (25 mL) at N N H H 0°C. The ice bath was removed after addition and the reaction was allowed to warm to room temperature and stirred for 18 h. The reaction mixture was filtered and volatiles were removed in vacuo yielding 2.3 as a pale yellow solid (3.16 g,

6.80 mmol, 82%). Colourless, X-ray quality crystals were grown from two different n- hexane solutions, one exposed to air and the other air-protected. Mp: 123-126°C. Anal.

Calcd. for C30H50BN3: C, 77.73; H, 10.87; N, 9.06. Found: C, 77.35; H, 10.22; N, 8.96.

1 H NMR (C6D6, 25°C): δ 7.13 (m, 6 H, Dipp), 3.68 (sept, 4 H, −CH(CH3)2 of Dipp

3 i groups, JH-H = 6.8 Hz), 3.30 (2 H, br s, −NH), 3.24 (sept, 2 H, −CH(CH3)2 of −N Pr2

3 3 groups, JH-H = 6.8 Hz), 1.28 (d, 12 H, −CH(CH3)2 of Dipp groups, JH-H = 6.8 Hz), 1.18

3 (d, 12 H, −CH(CH3)2 of Dipp groups, JH-H = 6.8 Hz), 1.11 (d, 12 H, −CH(CH3)2 of

i 3 11 13 −N Pr2 groups, JH-H = 6.8 Hz). B NMR (C6D6, 25°C): δ 23.4. C NMR (C6D6, 25°C):

i δ 146.5 (Dipp), 139.8 (Dipp), 126.0 (Dipp), 123.6 (Dipp), 45.3 (−CH(CH3)2 of −N Pr2 groups), 29.1 (−CH(CH3)2 of Dipp groups), 24.5 (−CH(CH3)2 of Dipp groups), 24.3

i (−CH(CH3)2 of -N Pr2 groups), 24.0 (−CH(CH3)2 of Dipp groups).

i n Preparation of [Li2][ Pr2NB(NDipp)2] (2.4). A 2.5 M solution of BuLi in hexanes (2.0

i i N Pr2 mL, 5.0 mmol) was added to a solution of Pr2NB[N(H)Dipp]2

Dipp B Dipp N N (1.16 g, 2.49 mmol) in n-hexane (20 mL) at −30°C. The

Li Li reaction mixture was allowed to warm to room temperature and stirred for 18 h. Volatiles were removed in vacuo to give 2.4 as a pale yellow solid

(1.004 g, 2.10 mmol, 85%). Colourless, X-ray quality crystals were grown from an n- hexane/toluene mixture. Anal. Calcd. for C66H110N6B2Li4 (dimer + co-crystallized hexane molecule): C, 76.44; H, 10.69; N, 8.10. Found: C, 76.12; H, 10.93; N, 7.92. 1H NMR

(THF-d8, 25°C): δ 6.77 (m, 4 H, Dipp), 6.23 (m, 2 H, Dipp), 3.84 (br sept, 4 H,

i −CH(CH3)2 of Dipp groups), 3.65 (br sept, 2 H, −CH(CH3)2 of −N Pr2 groups), 1.20 (d,

3 12 H, −CH(CH3)2 of Dipp groups, JH-H = 6.8 Hz), 1.07 (d, 12 H, −CH(CH3)2 of Dipp

3 i 3 11 groups, JH-H = 6.8 Hz), 0.92 (d, 12 H, −CH(CH3)2 of −N Pr2 groups, JH-H = 6.9 Hz). B

7 13 NMR (THF-d8, 25°C): δ 22.8. Li NMR (THF-d8): δ 0.87. C NMR (THF-d8, 25°C): δ

i 157.9 (Dipp), 139.2 (Dipp), 122.7 (Dipp), 46.9 (−CH(CH3)2 of −N Pr2 groups), 27.9

(−CH(CH3)2 of Dipp groups), 26.5 (−CH(CH3)2 of Dipp groups), 25.8 (−CH(CH3)2 of

i −N Pr2 groups), 24.9 (−CH(CH3)2 of Dipp groups).

Preparation of ClB[N(H)Dipp]2 (2.5). A solution of DippNH2 (4.26 g, 24.0 mmol) in n-

Cl hexane (40 mL) was added to a solution of BCl3 in hexanes (6 Dipp B Dipp N N mL, 6 mmol) at −80°C. A white precipitate formed shortly

H H after the addition. The reaction mixture was warmed to room temperature and stirred for 18 h. It was then filtered and volatiles were removed in vacuo to give 2.5 as a colourless solid (1.71 g, 72%). MP: 67-69°C. Anal. Calcd. for

1 C24H36BN2Cl: C, 72.28; H, 9.10; N, 7.02. Found: C, 73.08; H, 9.45; N, 7.05. H NMR

(C6D6, 25°C): δ 7.17-7.06 (m, 6 H, Dipp), 4.14 (br, 1 H), 3.72 (br, 1 H), 3.57 (br s, 2 H),

11 3.38 (br, 2 H), 1.20 (overlapping doublets, 24 H, −CH(CH3)2). B NMR (C6D6, 25°C): δ

13 26.4. C NMR (C6D6, 25°C): δ 147 (Dipp), 137 (Dipp), 124 (Dipp), 28.9 (−CH(CH3)2),

24.3 (−CH(CH3)2). HRMS: Calculated m/z for [C24H36N2BCl]: 398.2660. Found:

398.2680.

Preparation of BrB[N(H)Dipp]2 (2.6). The bromo derivative 2.6 was obtained as a

Br colourless solid (0.87 g, 4.2 mmol, 47%) from DippNH2 (2.85 Dipp B Dipp N N g, 16.1 mmol) in n-hexane (15 mL) and BBr3 (1.04 g, 4.15

H H mmol) in hexanes (15 mL) at −80°C by using a procedure similar to that described for 2.5. Mp: 109-112°. Anal. Calcd. for C24H36BN2Br: C, 65.03;

1 H, 8.19; N, 6.32. Found: C, 64.64; H, 8.34; N, 6.34. H NMR (C6D6, 25°C): δ 7.19-7.10

(m, 6 H, Dipp), 4.37 (br, 1 H), 3.94 (br, 1 H), 3.60 (br s, 2 H), 3.41 (br, 2 H), 1.19

11 13 (overlapping doublets, 24 H, −CH(CH3)2). B NMR (C6D6, 25°C): δ 25.7. C

NMR(C6D6, 25°C): δ 147 (Dipp), 136.8 (Dipp), 135.9 (Dipp), 123.9 (Dipp), 29.0

(−CH(CH3)2), 24.3 (−CH(CH3)2). LRMS: Calculated m/z for [C24H36N2BBr]: 442.2.

Found: 442.2.

Preparation of B[N(H)Dipp]3 (2.7). A solution of DippNH2 (6.38 g, 36 mmol) in n-

H Dipp hexane (40 mL) was added to a stirred solution of BCl3 in n- N

Dipp B Dipp hexane (6.0 mL, 6.0 mmol) at −80°C. A white precipitate N N began to form upon addition. The reaction mixture was allowed H H to warm to room temperature and stirred for 18 h. After filtration, volatiles were removed in vacuo yielding 2.7 as a white sticky solid (3.0 g, 5.6 mmol, 93%).

Colourless, X-ray quality crystals were grown from an air-protected n-hexane solution.

Mp: 167-170°C. Anal. Calcd. for C36H54BN3: C, 80.12; H, 10.09; N, 7.79. Found: C,

1 79.62; H, 10.12; N, 7.72. H NMR (C6D6, 25°C): δ 7.13 (m, Dipp), 7.02 (m, Dipp), 3.73

3 (sept, 6 H, −CH(CH3)2, JH-H = 6.9 Hz), 3.26 (s, 3 H, −NH), 1.24, 1.19 (overlapping d, 36

11 13 H, −CH(CH3)2). B NMR (C6D6, 25°C): δ 22.6. C NMR (C6D6, 25°C): δ 147.1 (Dipp),

137.4 (Dipp), 126.6 (Dipp), 123.9 (Dipp), 28.9 (−CH(CH3)2), 24.1 (−CH(CH3)2). HRMS:

Calculated m/z for [C36H54BN3]: 539.4411. Found: 539.4416.

i Alternative Synthesis of [Li2][ Pr2NB(NDipp)2] (2.4). A solution of B[N(H)Dipp]3

(0.334 g, 0.62 mmol) in n-hexane (12 mL) was added to a slurry of LDA (0.20 g, 1.87

62 mmol) in n-hexane (12 mL) at −40°C. The reaction was maintained at this temperature for 45 min then allowed to warm to room temperature and stirred for 18 h. The mixture was filtered and volatiles were removed in vacuo to give a yellow solid (0.285 g). The

1 i H NMR spectrum showed the presence of [Li2][ Pr2NB(NDipp)2] (2.4) as the major product, together with a minor unidentified impurity.

i Synthesis of {Ge[ Pr2NB(NDipp)2]}2·dioxane (2.8). A solution of 2.4 (0.012 g, 0.025

Dipp Dipp mmol) in THF-d8 was added to solid N N i i Pr2N B Ge O O Ge B N Pr2 GeCl2·(C4H8O2) (0.006g, 0.026 mmol) in N N

Dipp Dipp an NMR tube. As the GeCl2·(C4H8O2) dissolved and reacted, the colour of the solution changed from pale yellow to golden

1 orange. NMR data were collected on this solution. H NMR (THF-d8, 25°C): δ 7.04 (d,

3 4 H, Dipp), 6.87 (t, 2 H, Dipp), 3.80 (sept, 4 H, −CH(CH3)2 of Dipp groups, JH-H = 6.9

i 3 Hz), 3.57 (s, 4H, dioxane), 3.26 (sept, 2 H, −CH(CH3)2 of −N Pr2 groups, JH-H = 6.9 Hz),

3 1.30 (d, 12 H, −CH(CH3)2 of Dipp groups, JH-H = 6.9 Hz), 1.16, (d, 12 H, −CH(CH3)2 of

3 i 3 Dipp groups, JH-H = 6.9 Hz), 0.86 (d, 12 H, −CH(CH3)2 of −N Pr2 groups, JH-H = 6.9

11 13 Hz). B NMR (THF-d8, 25°C): δ 28.04. C NMR (THF-d8, 25°C): δ 145.2, 123.4,

123.2, 68.0, 45.7, 28.8, 28.0, 25.0, 23.4.

i Synthesis of [Mg][ Pr2NB(NDipp)2] (2.9). A mixture of 2.4 (0.189 g, 0.40 mmol) and

i Pr2N MgCl2 (0.038 g, 0.40 mmol) was dissolved in THF (15 mL) B and heated at 60°C for 4 h. Volatiles were removed in vacuo to Dipp N N Dipp

Mg 1 give 2.9 as an orange solid (0.085 g, 44%). H NMR (THF-d8 O O 25°C): δ 6.67 (d, 4 H, Dipp), 6.25 (t, 2 H, Dipp), 4.15 (sept, 4

3 H, −CH(CH3)2 of Dipp groups, JH-H = 6.7 Hz), 3.43 (sept, 2 H, −CH(CH3)2 of –NiPr2

3 3 3 groups JH-H = 6.7 Hz), 1.18 (d, 12 H, JH-H = 6.7 Hz), 1.09 (d, 12 H, JH-H = 6.7 Hz), 0.76

3 11 13 (d, 12 H, JH-H = 6.9 Hz). B NMR (THF-d8, 25°C): δ 23.2. C NMR (THF-d8, 25°C): δ

157.5, 141.7, 122.5, 115.0, 47.1, 27.9, 26.5, 24.8.

i Synthesis of [Zn][ Pr2NB(NDipp)2] (2.10). A solution of 2.4 (0.372 g, 0.78 mmol) in

i Pr2N THF (10 mL) was added to solid ZnCl2 (0.107 g, 0.79 mmol) B and stirred for 30 min. Volatiles were removed in vacuo and Dipp N N Dipp Zn the product was extracted with Et2O (20 mL) and filtered. O O Removal of volatiles in vacuo gave 2.10 as pale yellow solid

1 (0.335 g, 0.50 mmol, 63%). H NMR (THF-d8, 25°C): δ 6.74 (d, 4 H, Dipp), 6.42 (t, 2

3 H, Dipp), 4.15 (sept, 4 H, CH(CH3)2 of Dipp groups, JH-H = 6.8 Hz), 3.30 (sept, 2 H,

i 3 −CH(CH3)2 of −N Pr2 groups, JH-H = 6.9 Hz), 1.21 (d, 12 H, −CH(CH3)2 of Dipp groups,

3 3 JH-H = 6.8 Hz), 1.13, (d, 12 H, −CH(CH3)2 of Dipp groups, JH-H = 6.8 Hz), 0.76 (d, 12

i 3 11 13 H, −CH(CH3)2 of −N Pr2 groups, JH-H = 6.8 Hz). B NMR (THF-d8, 25°C): δ 25.2. C

NMR (THF-d8, 25°C): δ 154.2, 143.1, 121.9, 117.0, 46.7, 28.2, 25.0, 23.3.

i Synthesis of {[ Pr2NB(NDipp)2]YCl(THF)n}m (2.11). A solution of 2.4 (0.170 g, 0.21

Dipp mmol) in THF (8 mL) was added to a slurry of YCl3 (0.100 N i g, 0.21 mmol) in THF and stirred for 10 min. Volatiles were Pr2N B Y(Cl)(THF)n N removed in vacuo and the product washed with hexanes (7 Dipp m mL). After removal of volatiles in vacuo, a colourless solid

1 was obtained, which still contained the LiCl by-product. H NMR (THF-d8, 25°C): δ

3 6.77 (d, 4 H, Dipp), 6.44 (t, 2 H, Dipp), 4.08 (sept, 4 H, −CH(CH3)2 of Dipp groups, JH-H

i 3 = 6.7 Hz), 3.60 (−OCH2CH2, THF), 3.53 (sept, 2 H, −CH(CH3)2 of −N Pr2 groups, JH-H =

3 7.0 Hz), 1.78 (−OCH2CH2, THF), 1.23 (d, 12 H, −CH(CH3)2 of Dipp groups, JH-H = 6.7

3 Hz), 1.17, (d, 12 H, −CH(CH3)2 of Dipp groups, JH-H = 6.7 Hz), 0.78 (d, 12 H,

i 3 −CH(CH3)2 of −N Pr2 groups, JH-H = 7.0 Hz).

i Synthesis of [ Pr2NB(NDipp)2]LaI(THF)x (2.12). A solution of 2.4 (0.170 g, 0.21

Dipp mmol) in THF (8 mL) was added to a slurry of LaI3(THF)4 N i (0.100 g, 0.21 mmol) in THF and stirred for 10 min. Pr2N B La(I)(THF)n N Volatiles were removed in vacuo and the product was Dipp m washed with hexanes (7 mL). After removal of volatiles in vacuo, a colourless solid was obtained, which still contained the LiI by-product. 1H

NMR (THF-d8, 25°C): δ 6.83 (d, 4 H, Dipp), 6.49 (t, 2 H, Dipp), 4.01 (sept, 4 H,

3 −CH(CH3)2 of Dipp groups, JH-H = 6.8 Hz), 3.60 (−OCH2CH2, THF), 3.45 (sept, 2 H,

i 3 −CH(CH3)2 of −N Pr2 groups, JH-H = 7.0 Hz), 1.78 (−OCH2CH2, THF), 1.29 (d, 12 H,

3 −CH(CH3)2 of Dipp groups, JH-H = 6.8 Hz), 1.19, (d, 12 H, −CH(CH3)2 of Dipp groups,

3 i 3 JH-H = 6.8 Hz), 0.75 (d, 12 H, −CH(CH3)2 of −N Pr2 groups, JH-H = 7.0 Hz).

Reaction of [Li2][PhB(NDipp)2] with InCl. A solution of Li2[PhB(NDipp)2] (0.20 g,

0.45 mmol) in Et2O was added to a slurry of InCl (0.069 g, 0.46 mmol) in Et2O at −70°C.

The reaction mixture immediately turned dark brown with formation of a precipitate.

The mixture was allowed to warm to room temperature and stirred for 4 h. After filtration to remove indium metal and LiCl and evaporation of volatiles in vacuo, an orange-yellow solid was isolated and the major product was determined to be the spirocyclic indium(III)

104 1 complex LiIn[PhB(NDipp)2]2 2.13. H NMR (C6D6, 25°C): δ 7.29-6.73 (m, 22 H, aryl

3 of Dipp and Ph), 4.47 (sept, 4 H, −CH(CH3)2, JH-H = 7 Hz), 3.81 (sept, 4 H, −CH(CH3)2,

3 3 JH-H = 7 Hz), 2.93 (q, Et2O), 1.73 (d, 12 H, −CH(CH3)2, JH-H = 7 Hz), 1.64 (d, 12 H,

3 3 −CH(CH3)2, JH-H = 7 Hz), 0.82 (d, 12 H, −CH(CH3)2, JH-H = 7 Hz), 0.77 (t, Et2O), 0.65

3 11 (d, 12 H, −CH(CH3)2, JH-H = 7 Hz). B NMR (C6D6, 25°C): δ 30 ppm (br, s).

Preparation of Ph(Me)NB[N(H)Dipp]2 (2.14H2). A solution of 2.5 (1.110 g, 2.78

Ph Me mmol) in hexanes (15 mL) was added to a stirred slurry of N

Dipp B Dipp [Li][NPh(Me)] (0.316 g, 2.79 mmol) in hexanes (10 mL) at ca. N N 60 C. The reaction was kept cold for 30 min then warmed to H H − ° room temperature and stirred for an additional 4.5 h. After filtration and removal of volatiles in vacuo, 2.14H2 was obtained as a sticky colourless solid that completely solidifies after 48 h (1.180 g, 2.51 mmol, 90%). Anal. Calcd. for C31H44BN3: C, 79.30;

1 H, 9.45; N, 8.95. Found: C, 79.26; H, 9.52; N, 8.52. H NMR (C6D6, 25°C): δ 7.33-7.12

(aryl resonances of Dipp and Ph groups), 6.92 (t, Dipp group) 3.71 (sept, 2 H,

3 3 −CH(CH3)2, JH-H = 6.9 Hz), 3.66 (sept, 2 H, −CH(CH3)2, JH-H = 6.9 Hz), 3.62 (2 H, br s,

3 11 −NH), 2.61 (3 H, −CH3) 1.17 (d, 24 H, −CH(CH3)2, JH-H = 6.9 Hz). B NMR (C6D6,

13 25°C): δ 24.7. C NMR (C6D6, 25°C): δ 12 resonances in the range of 152 to 120 ppm

(aryl, Dipp and Phenyl) 36.4 (−CH(CH3)2), 29.2 (−CH(CH3)2), 28.9 (−CH(CH3)2), 24.3

(−CH(CH3)2).

n Preparation of [Li2][Ph(Me)NB(NDipp)2] (Li22.14). A 2.5 M solution of BuLi in

Ph Me hexanes (1.96 mL, 4.9 mmol) was added to a solution of N

B Ph(Me)NB[N(H)Dipp]2 (1.153 g, 2.45 mmol) in hexanes (15 Dipp N N Dipp mL) at ca. 0°C. The reaction mixture was allowed to warm to Li Li room temperature and stirred for 1 h. The resulting colourless precipitate was allowed to settle and the solvent was removed via a cannula. Volatiles were removed in vacuo to

1 give Li22.14 as a colourless solid (0.805 g, 1.67 mmol, 68 %). H NMR (THF-d8, 25°C):

δ 7.26-6.69 (aryl, Dipp and Ph), 3.84 (br sept, 4 H, −CH(CH3)2), 2.39 (s, 3 H, −CH3).

11 1.14-1.02 (overlapping doublets, 24 H, −CH(CH3)2). B NMR (THF-d8, 25°C): δ 23.8.

7 Li NMR (THF-d8): δ 0.03.

Preparation of Ph2NB[N(H)Dipp] (2.15H2). A solution of 2.6 (1.259 g, 2.8 mmol) in

Ph Ph hexanes (15 mL) was added to a stirred slurry of [Li][NPh2] N

Dipp B Dipp (0.501 g, 2.85 mmol) in hexanes (15 mL) at ca. −60°C. The N N

H H

67 reaction was kept cold for 30 min then warmed to room temperature and stirred for an additional 3.5 h. After filtration and removal of volatiles in vacuo, 2.15H2 was obtained as a sticky colourless solid that completely solidifies after 3 d at room temperature (1.138 g, 1.67 mmol, 75 %). Anal. Calcd. for C36H46BN3: C, 81.34; H, 8.72; N, 7.90. Found: C,

1 80.27; H, 9.02; N, 7.65 H NMR (C6D6, 25°C): δ 7.02-6.79 (16 H, aryl resonances of

3 Dipp and Ph groups), 3.99 (2 H, br s, −NH), 3.62 (sept, 4 H, −CH(CH3)2, JH-H = 6.8 Hz),

3 11 1.14 (d, 24 H, −CH(CH3)2, JH-H = 6.8 Hz). B NMR (C6D6, 25°C): δ 25.3.

Synthesis of [Na][Ph2NB{(NDipp)N(H)Dipp}] (2.16). A solution of Ph2NB[N(H)Dipp]

Ph Ph (0.344 g, 0.65 mmol) in THF (8 mL) was added to a solution of N n Dipp B Dipp Na Bu (0.104 g, 1.29 mmol) in a mixture of THF:hexanes (10 N N mL: 3 mL) at ca. 78 C. The reaction was kept cold for 1.5 h H − ° Na 1 and volatiles were removed in vacuo to yield a pale yellow solid. H NMR (THF-d8,

25°C): δ 7.05-6.9 (10 H, aryl resonances of Dipp and Ph groups), 6.91-6.50 (6 H, aryl

3 resonances of Dipp and Ph groups), 4.09 (sept., 2 H, −CH(CH3)2, JH-H = 6.9 Hz), 3.41

3 3 (sept., 2 H, −CH(CH3)2, JH-H = 6.8 Hz), 3.18 (s, 1H, −NH), 1.37 (d, 6 H, −CH(CH3)2, JH-

3 3 H = 6.9 Hz), 1.20 (d, 6 H, −CH(CH3)2, JH-H = 6.91 Hz), 0.83 (d, 12 H, −CH(CH3)2, JH-H

= 6.8 Hz).

Chapter Three: Formation of a Seven-Membered C4OBN Heterocycle by a THF Ring-Expansion Process

3.1 Introduction

The ring-opening of THF promoted by Lewis acidic transition metal,119-124 lanthanide125 or actinide complexes126-128 is well known in the literature. Examples of initiation by electrophilic main group element centres are increasing in number. In this category, ring-opening of THF happens when dimethylethylamine-alane is reacted with

129 three equivalents of 2,6-diisopropylphenol in THF. The combination of TeBr4 with

130 triphenylphosphine in THF produces the zwitterionic product Ph3PO(CH2)4TeBr4.

Cooperative ring-opening occurs when the tellurium diimide dimer tBuNTe(µ-

t t N Bu)2TeN Bu is added to a solution of B(C6F5)3 in THF; the 1:1 adduct of these two reagents also initiates this process.131 It has also been shown that sterically bulky phosphines react with the adduct (THF)·B(C6F5)3 in a ring-opening fashion, producing phosphonium-borate zwitterions.132,133 Similarly, the borane complex

134 135 (lutidine)·B(C6F5)3 and a carbene/B(C6F5)3 Lewis pair both promote ring-opening of

THF to give zwitterionic products. Finally, it has recently been demonstrated that

136 penta(organo)[60]fullerenes can open THF in the presence of Me3SiCl.

In the preceding examples, the THF molecule is integrated into the final product as an acyclic fragment; incorporation of THF as part of a ring is much less common, with only a single example previously reported. The group of Sekiguchi found that seven- membered rings (1,3,2-oxasila- and germaborepanes, 3.1a and 3.1b, respectively) are formed by the insertion of in situ >E=B–doubly bonded species (E = Si, Ge) into a C−O bond of THF.137

H2 C H2C CH2

O CH2

B E t SiMe Bu2 Mes SiMetBu2

3.1a: E = Si 3.1b: E = Ge Mes = 2,4,6-trimethylphenyl

It is well known that Lewis acidic boron trihalides BX3 (X = Cl, Br) readily cleave ethers; organoboron halides, for instance Me2BBr, are known to cleave cyclic

138,139 ethers such as 2-methyltetrahydrofuran. By contrast, the reaction of Ph2BCl with a

THF solution of tetralithiated tetra-tert-butyl calix(4)arene does not proceed to give a calixarene-containing product but instead yields the simple adduct Ph2BCl·THF, which was characterized by X-ray analysis.140 This was the first example of THF remaining unchanged in the presence of a boron-halogen bond and suggested that the replacement of two chlorine atoms in BCl3 by phenyl groups substantially reduces the reactivity of the remaining B−Cl bond towards C−O bond cleavage, as boron trichloride readily ring- opens THF.140

In this chapter, the first structurally characterized example of a 1,3,2- oxazaborepane is presented. This novel product, containing a seven-membered C4OBN ring, was inadvertently prepared from the reaction of BCl3 with two equivalents of

[Li][N(H)Dipp] in THF during investigations of alternative syntheses of boraguanidinate

2.4, described in Chapter 2. Evidence supporting a reaction pathway where the ring-

opening is initiated by the (amino)dichloroborane Cl2B[N(H)Dipp] will be discussed.

3.2 Synthesis, Spectroscopic Characterization and X-ray Structure of

DippN(H)BO(CH2)4NDipp (3.2a)

In Section 2.4 it was mentioned that when preparing the (diamino)chloroborane

ClB[N(H)Dipp]2 (2.5) it is imperative to use the reaction of BCl3 and four equivalents of

DippNH2 in hexane as opposed to the reaction of BCl3 with two equivalents of

[Li][N(H)Dipp] in THF (Scheme 2.3). This was emphasized because the latter reaction produces the colourless solid DippN(H)BO(CH2)4NDipp (3.2a) (Scheme 3.1) and a minor oily by-product, instead of the (diamino)chloroborane 2.5. Compound 3.2a was isolated as a pure crystalline solid from a concentrated solution of the crude reaction product in a mixture of hexanes and toluene at ca. −30°C and was characterized by multinuclear NMR spectroscopy (1H, 11B, 13C), CHN analysis and single crystal X-ray diffraction.

H2 C H2C CH2 THF/hexanes BCl + 2 [Li][N(H)Dipp] 3 O CH - 2 LiCl 2 - HCl B N

Dipp(H)N Dipp 3.2a

Scheme 3.1

In addition to the characteristic aryl, methyl and methyne resonances for the Dipp substituents, the 1H NMR spectrum of 3.2a displays four sets of equally intense multiplets (at 3.82, 3.14, 1.69 and 1.52 ppm) with relative intensities that correspond to two protons each, signifying the presence of four inequivalent methylene groups. The relatively deshielded chemical shifts of two of these resonances correlate to the −CH2− units that are linked to the electronegative nitrogen and oxygen atoms. Resonances corresponding to the former-THF carbon atoms are observed in the 13C NMR spectrum at

64.9, 53.5 28.9 and 28.8 ppm and a single broad resonance for the three-coordinate boron centre is found in the 11B NMR spectrum at 23.8 ppm.

Single crystal X-ray diffraction confirmed the molecular structure of 3.2a, which is illustrated in Figure 3.1; selected structural parameters are presented in Table 3.1.

Table 3.1. Selected bond lengths (Å) and bond angles (°) for 3.2a.

B1−O1 1.381(3) O1−B1−N2 116.5(2) B1−N2 1.421(3) O1−B1−N1 122.9(2) B1−N1 1.428(3) N2−B1−N1 120.6(2) N1−C28 1.478(2) B1−N1−C28 123.9(2)

N1−C1 1.440(2) C1−N1−C28 113.9(2) O1−C25 1.431(3) B1−N1 −C1 120.8(2) C28−N1− B1−O1 22.4(3) C28−C27− C26−C25 57.1(4) C28A−C27A− C26A−C25A 39.7(8)

Figure 3.1. Thermal ellipsoid plot (30% probability) of one conformation of

DippN(H)BO(CH2)4NDipp (3.2a). For clarity, all hydrogen atoms except that on nitrogen have been omitted and only α−carbon atoms of Dipp groups are shown.

Carbon atoms C26 and C27 are disordered over two atomic sites resulting in two superimposed seven-membered rings with chair- and twist boat-like conformations in a

60:40 ratio, respectively (Figure 3.2); the latter conformation is observed for the seven-

73 membered rings in 3.1a and 3.1b. The dihedral angles of the two rings in 3.2a are 57.1° for C25−C26−C27−C28 and 39.7o for C25A−C26A−C27A−C28A.

Figure 3.2. Chair (top) and boat (bottom) conformations of the C4NBO ring in 3.2a with hydrogen atoms and substituents on boron and nitrogen omitted for clarity.

The C28−N1−B1−O1 torsion angle of 22.4° is comparable to that of a related seven-membered heterocycle containing nitrogen, boron and oxygen where the boron atom is four-coordinate.141 The boron-oxygen bond length of 1.38 Å is in agreement with the corresponding distance in structures containing a nitrogen-boron-oxygen fragment, which typically range from 1.35 Å 142 to 1.43 Å.143 Specifically, the distance is indistinguishable from that observed for the borazine 3.3,144 which also contains an

N−B−O fragment in a cyclic environment.

Ph O

B N N Ph B B O O N

Ph 3.3

The endocyclic and exocyclic B−N bond lengths in 3.2a are equal and intermediate between single and double bond values. The two N−aryl bonds for the Dipp groups in 3.2a are identical within experimental error, but slightly shorter than the

N−alkyl bond within the ring. The geometry about the three-coordinate boron and nitrogen atoms in the ring is planar (Σ ∠ B = 360o and Σ ∠ N1 = 358.6°).

Heterocycle 3.2a is the first structurally characterized example of a seven- membered ring containing nitrogen, boron and oxygen, in which three-coordinate nitrogen and boron atoms are covalently bonded. A single example of a 1,3,2- oxazaborepane with an alkyl substituent on boron has been reported, obtained from the reaction of 1-hydroxymethyl-2-aminomethylbenzene with dibutyl isopropylboronate.

However, this product was isolated as a red oil and characterized only by 1H NMR and

IR spectroscopy and by mass spectrometry.145 Other ring systems containing these elements within smaller rings are known, including the well-studied 1,3,2- oxazaborolidines, which are used as catalysts for numerous organic transformations,146,147 and the tricyclic borazine 3.3.144

The formation of the C4OBN heterocycle 3.2a can be envisaged to involve the initial coordination of THF to a Lewis acidic chloroborane Cl3-nB[N(H)Dipp]n (n = 1 or

2; shown in detail for n = 1 in Scheme 3.2), followed by concomitant N−C bond formation and ring-opening by C−O bond cleavage, with the elimination of HCl to give the observed product. Investigations to gain evidence in support of this pathway are the topic of the following section.

+ BCl3 + x [Li][N(H)Dipp] O x = 2 x = 1 - 2 LiCl - LiCl - HCl

H2 C H2 H2 C C H2C CH2 H2C CH2 H2C CH2 [Li][N(H)Dipp] O CH2 O CH2 O CH2 - HCl - LiCl Dipp B N B N B N Cl Cl Dipp Dipp(H)N Dipp Cl H 3.2b 3.2a

Scheme 3.2

3.3 Synthesis and Spectroscopic Characterization of ClBO(CH2)4NDipp (3.2b)

To examine the above reaction pathways, disubstituted (diamino)chloroborane

ClB[N(H)Dipp]2 (2.5) was first stirred in THF for 18 h; however, heterocycle 3.2a was not formed, even when an HCl scavenger was added. Given this observation, the monosubstituted dichloroborane Cl2B[N(H)Dipp] (3.4), which is a stronger Lewis acid when compared to 2.5, was proposed to be responsible for initiating the ring-opening

76 process. Consequently, the reaction depicted in Scheme 3.1 was carried out using BCl3 and [Li][N(H)Dipp] in an equimolar ratio in THF in an attempt to generate 3.3 in situ and observe its subsequent reaction in this solvent.

The 1H NMR spectrum of the product of this reaction revealed a major species containing equally intense multiplets at 3.69, 3.13, 1.56 and 1.48 ppm; a singlet was observed at 26.8 ppm in the 11B NMR spectrum. The similarity of the chemical shifts for methylene groups in the 1H NMR spectrum to those observed for 3.2a (at 3.82, 3.14, 1.69 and 1.52 ppm) is consistent with the formation of the ring-opened derivative 3.2b with a

Cl substituent on boron (Scheme 3.2). Additionally, the small downfield shift of 3.0 ppm in the 11B NMR resonance with respect to 3.2a also supports the presence of an exocyclic

Cl substituent attached to boron instead of an −N(H)Dipp group. The reaction mixture

1 contains [DippNH3][Cl] ( H NMR spectrum), which presumably forms by the reaction of the eliminated HCl with [Li][N(H)Dipp] in a 2:1 molar ratio. Further evidence that the oily product obtained from this reaction is 3.2b was drawn from the observation that the subsequent addition of excess [Li][N(H)Dipp] produced 3.2a (1H NMR and 11B NMR spectra).

The isolation of [DippNH3][Cl] in the synthesis of both 3.2a and 3.2b suggests that adding excess amide to scavenge the HCl by-product could increase the yield.

However, carrying out this reaction of BCl3:[Li][N(H)Dipp] in a 2:5 molar ratio gave the ring-opened product in lower yields.

Efforts were also made to isolate the (amino)dichloroborane 3.4 from the reaction of DippNH2 with a solution of BCl3 in a 2:1 molar ratio in hexanes. Although a new boron-containing product was formed, with an 11B NMR chemical shift appropriate for

Cl2B[N(H)Dipp] (32.3 versus 26.4 ppm for ClB[N(H)Dipp]2), it was contaminated by significant amounts of ClB[N(H)Dipp]2; because of the similar compositions, and therefore similar solublities, of the two chloroboranes, these products could not be separated by extraction. However, stirring the mixture containing both 2.5 and 3.4 in

THF for 18 h also produced some 3.2b, as identified by 1H and 11B NMR spectra.

Although the formation of this seven-membered ring formally involves the insertion of the iminoborane ClB=NDipp into a C−O bond of the THF molecule, the generation of an iminoborane under the reactions conditions employed here is unlikely as these compounds are typically obtained from gas-phase thermolysis reactions.148 It has been shown that iminoboranes can also be formed by inducing HX elimination from

149 sterically bulky aminoboranes using Na[N(SiMe3)2] in THF, demonstrating a contrast in reactivity to the (amino)chloroboranes presented here.

3.4 Conclusions

This chapter described the unexpected formation of a novel seven-membered heterocycle from the reaction of BCl3 with two equivalents of [Li][N(H)Dipp] in THF.

Compound 3.2a is the first structurally characterized example of a 1,3,2-oxazaborepane.

Variation of the stoichiometry of this reaction indicates that the THF ring-opening process is likely initiated by the dichloroborane Cl2B[N(H)Dipp] and NMR evidence supports the formation of a second ring-opened product, ClBO(CH2)4NDipp.

3.5 Experimental Section

Reagents, general procedures and instrumentation are described in Appendix 1.

Preparation of DippN(H)BO(CH2)4NDipp (3.2a). A solution of [Li][N(H)Dipp] (1.10

H2 g, 6.0 mmol) in THF (15 mL) was added to a solution of BCl3 (3 C H C 2 CH2 mL, 1 M, 3.0 mmol) in hexanes at ca. −80°C. The reaction

O CH2 mixture was allowed to warm to room temperature and stirred for B N

Dipp(H)N Dipp 18 h. Volatiles were removed in vacuo and the oily product was

extracted with hexanes and filtered. After removal of solvent, a sticky solid was isolated

and colourless, X-ray quality crystals of 3.2a were grown from an n-hexane/toluene

solution (0.488 g, 1.12 mmol, 37 %). Anal. Calcd. for C28H43N2BO: C, 77.41; H, 9.98; N,

1 6.45. Found: C, 76.68; H, 10.02; N, 6.28. H NMR (C6D6; 25°C): δ 7.18−7.09 (m, 6 H,

3 Dipp groups), 3.82 (m, 2 H, ring −OCH2), 3.66 (sept, 2 H, −CH(CH3)2, JH-H = 6.9 Hz),

3 3.40 (sept, 2 H, −CH(CH3)2, JH-H = 6.9 Hz), 3.14 (m, 2 H, −NCH2), 2.92 (1 H, br s,

−NH), 1.69 m, 2 H, −OCH2CH2−), 1.52 (m, 2 H, −NCH2CH2), 1.35 (d, 6 H, −CH(CH3)2,

3 11 JH-H = 6.8 Hz), 1.23 and 1.22 (overlapping d, 18 H, −CH(CH3)2). B NMR (C6D6,

13 25°C): δ 23.8 (br, s). C NMR (C6D6, 25°C): δ 147.7 (Dipp), 145.8 (Dipp), 144.6

(Dipp), 138.1 (Dipp), 127.5 (Dipp), 125.6 (Dipp), 124.9 (Dipp), 123.3 (Dipp), 64.9 (-

CH2-O), 53.5 (-CH2-N), 30.8 (−CH(CH3)2), 28.9 (ring −CH2−), 28.8 (ring −CH2−), 28.5

(−CH(CH3)2), 26.5 (−CH(CH3)2), 24.5 (−CH(CH3)2), 24.4 (−CH(CH3)2).

Preparation of ClBO(CH2)4NDipp (3.2b). A solution of [Li][N(H)Dipp] (0.917 g, 5.0

H2 mmol) in THF (15 mL) was added to a solution of BCl3 (5 mL, 1 M, C H2C CH2 5.0 mmol) in hexanes at ca. −80°C. The reaction mixture was allowed O CH2 B N to warm to room temperature and stirred for 18 h. Volatiles were Dipp Cl removed in vacuo and the product was extracted with hexanes and filtered to remove LiCl. After removal of solvent a peach-coloured oil was isolated. 1H

NMR ( C6D6; 25°C): δ 7.15−6.97 (m, 3H, Dipp), 3.69 (m, 2 H, −OCH2−), 3.36 (sept, 2 H,

−CH(CH3)2), 3.13 (m, 2 H, −NCH2), 1.56 (m, 2 H, −OCH2CH2−), 1.48 (m, 2 H,

11 −NCH2CH2), 1.13 (overlapping d, −CH(CH3)2). B NMR (C6D6, 25°C): δ 25.1 (br, s).

1 The resonances of [DippNH3]Cl were also observed in the H NMR spectrum. The product 3.2b could not be separated from the by-product [DippNH3]Cl. Addition of

[Li][N(H)Dipp] (0.288 g, 1.57 mmol) to this oily product (0.258 g) in either hexane or

THF gave an oil, which exhibited multiplet resonances at δ 3.82, , 3.14, 1.69, 1.52 in the

1H NMR spectrum attributable to 3.2a.

Chapter Four: New Boraamidinate Reagents

4.1 Introduction

Although numerous bams have been prepared (see Section 1.5), the ways in which they have been modified are limited: all are dianionic and although varying the substituents on nitrogen and boron has been explored, the extent of these alterations is narrow.85 The design of new types of bams is of interest for expanding the chemistry of this ligand. For example, the substituents on boron for known bams are always non- bridging; typically these groups are alkyl or aryl, with the first example of an amino group in this position presented in Chapter 2. However, the addition of a linker group would connect two dianionic bams and form a tetraanionic, bis-bam ligand. With the

t 2- 110,111 interesting redox chemistry demonstrated by the dianionic bam [PhB(N Bu)2] , it can be envisioned that isolating a bis-bam could extend this chemistry to novel biradical systems. The isolation of stable radical species is intriguing from both a fundamental and practical perspective150 and inorganic radicals containing main group elements are receiving increased attention.151,152 In particular, biradical systems incorporating p-block elements are significant for their unique electronic structures, chemical reactivity, and

81 physical properties such as magnetic behaviour.153,154 As such, the preparation of bis- bams and the exploration of their oxidation chemistry is appealing.

A second modification of the bam ligand that warrants investigation is single deprotonation of a bamH2 to form a monoanionic ligand that still contains an −NH functionality. New coordination modes would be expected for this type of bam ligand, which could be compared with that of monoanionic ams and guans through preparation of metal(I) or metal(II) complexes, for instance.

Finally, when metathetical reactions with dilithio bams are carried out, the lithium salt by-product that is produced is typically soluble to some extent in the reaction solvent

(i.e. THF, Et2O), necessitating separation of the desired product by extraction or crystallization. Having access to a dipotassiated starting material would be beneficial in cases where the products and by-products have similar solubilities as the resultant potassium salts would be much less soluble. In particular, using iodide starting materials would generate KI as the by-product, which is highly insoluble even in THF. Despite these advantages, a dipotassiated bam reagent has yet to be prepared.

This chapter explores the synthesis, characterization and reactivity of several new bam precursors and reagents, prepared with the above facets in mind. First, the synthesis and structural characterization of two mono-boryl ferrocenes and a 1,1ʹ-bis(boryl) ferrocene with amino groups on boron are presented and lithiation attempts to form ferrocenyl boraamidinates are discussed. Subsequently, the synthesis, spectroscopic

t t characterization and tetra-lithiation of [N(H) Bu]2B−C6H4−B[N(H) Bu]2 to give the first example of a bis-bam ligand is described and the molecular structure of the novel tetraanionic bis-bam is presented. Monoanionic bams are successfully prepared from the

n reaction of PhB[N(H)Dipp]2 and one equivalent of either BuLi or [K][CH2Ph]. The molecular structure of the former monoanionic product is compared to its neutral precursor and its dianionic analogue, which have previously been reported.104 Finally, a new dipotassiated bam reagent will be introduced with a discussion of its synthesis and structural parameters; the utility of this reagent will be demonstrated in Section 5.4.

4.2 Synthesis, Spectroscopic Characterization and X-ray Structures of

t t FcB[N(H) Bu]2 (4.1), FcB[N(H)Dipp]2 (4.2) and 1,1ʹ-Fc{B[N(H) Bu]2}2 (4.3)

155 Ferrocene, (Cp2Fe, Cp = cyclopentadienyl), was the first linker group examined for the preparation of bis-bams. The mono-borylation of ferrocene using BX3 (X = Br, I)

156-158 to give FcBX2 (Fc = ferrocenyl) has been known for over three decades. The chloro

159 derivative is preferably synthesized by reacting FcHgCl and BCl3, although it can also

159,160 be isolated from the reaction of B2Cl4 with ferrocene, while the bromo species is

157,161 easily prepared from heating equimolar amounts of ferrocene and BBr3 in hexanes.

Additionally, 1,1'-bis(bromoboryl) ferrocene, Fc(BBr2)2, can be isolated from either the

2:1 reaction of BBr3 and ferrocene or, in a stepwise manner, by treating FcBBr2 with one

161 equivalent of BBr3. Substitution of the halogen atoms in haloboryl ferrocenes for dialkylamino groups has been accomplished by the reaction 1,1′-Fc(BBr2)2 with different

158 amounts of secondary amine. The dimethyl derivative, 1,1′-Fc(BMe2)2, is obtained

158 using 1,1ʹ-Fc(BBr2)2 and an excess of tetramethyltin. More recently, it was demonstrated that FcBBr2 reacts in a metathetical fashion with two equivalents of lithium

162 anilide to afford FcB[N(H)Ph]2; the mono-substituted tert-butyl derivative,

FcBBr[N(H)tBu], was also prepared. However, no structural characterization for either compound was reported.

In view of these precedents, the new (diaminoboryl)ferrocenes FcB[N(H)R]2,

(4.1, R = tBu; 4.2, R = Dipp) were synthesized as precursors to ferrocenyl boraamidinate

(Fcbam) ligands. By preparing the mono-Fcbam ligands, a direct comparison of their

t 2- redox chemistry could be made with that of [PhB(N Bu)2] to observe the effects of the ferrocene substituent before extending this chemistry to bis-Fcbams.

156 t The reaction of FcBBr2 with two equivalents of [Li][N(H)R], (R = Bu, Dipp) in toluene produces the bis-amino ferrocenyl boranes 4.1 (R = tBu) and 4.2 (R = Dipp), which are isolated in moderate yields after filtration to separate the LiBr by-product and removal of volatiles (Scheme 4.1). Orange crystals of both 4.1 and 4.2, suitable for X-ray diffraction, were grown from a toluene or hexane/toluene solution; the molecular structures are shown in Figure 4.1 with selected bond lengths and angles presented in

Table 4.1.

BBr2 B[N(H)R]2 Toluene Fe + 2[Li][N(H)R] Fe -2 LiBr

4.1 R = tBu 4.2 R = Dipp

Scheme 4.1

t Figure 4.1. Thermal ellipsoid plots (30% probability) of FcB[N(H) Bu]2 (4.1, left) and

FcB[N(H)Dipp]2 (4.2, right). For clarity, most H atoms omitted and only α-carbons of Dipp groups are shown.

The unit cell of 4.1 contains two independent molecules, which is also found for

163 the FcBBr2 precursor. For both 4.1 and 4.2, planar geometry is observed at boron. In

4.2, the narrow range of bond angles at boron (ca. 119° to 122°) indicates an environment that is very close to trigonal, while 4.1 displays a distortion from trigonal geometry, with bond angles in the range of approximately 114° to 123°. The B−N bond lengths in both

4.1 and 4.2 are intermediate between a single and double bond as expected for

N(2p)→B(2p) π-bonding. Crystallographic analysis of 1,1′-bis(dibromoboryl)ferrocene has revealed that Lewis acidic ferrocenyl boranes have weak iron-boron interactions resulting in the tilting of the B−C bond towards the metal centre, as indicated by a

B−C1−C2−C3 dihedral angle of ca. 10°.164 The amino substituents in 4.1 and 4.2 significantly decrease the Lewis acidity of the boron centre, reducing this angle to ca. 4° in 4.1 and even further in 4.2 where the boron moiety lies in the plane of the Cp ring.

Table 4.1. Selected bond lengths (Å) and bond angles (o) for 4.1, 4.2 and 4.3.

4.1 4.2 4.3

B1−N1 1.421(6) 1.418(4) 1.415(3)

B1−N2 1.414(7) 1.419(3) 1.413(3)

* B2−N3 1.437(6) N/A 1.414(3)

* B2−N4 1.394(7) N/A 1.414(3)

N1−B1−N2 122.5(4) 118.8(2) 122.0(2)

N1−B1−C1 114.8(4) 119.1(2) 116.8(2)

N2−B1−C1 122.6(4) 122.1(2) 121.3(2)

* N3−B2−N4 123.2(4) N/A 122.2(2)

* N3−B2−C19 114.0(4) N/A 116.5(2)

* N4−B2−C19 122.7(4) N/A 121.3(2)

* These values correspond to the second molecule in the asymmetric unit of 4.1 but to the

1ʹ-boryl substituent in 4.3.

The disubstituted ferrocene, 1,1'-bis(tert-butylaminoboryl)ferrocene, 4.3, was

161 prepared from the analogous reaction using 1,1'-Fc(BBr2)2 and four equivalents of

[Li][N(H)tBu] in toluene (Scheme 4.2).

t BBr2 B[N(H) Bu]2 Toluene Fe + 4[Li][N(H)tBu] Fe - 4 LiBr t Br2B [ Bu(H)N]2B

4.3

Scheme 4.2

Orange, X-ray quality crystals were obtained by cooling a concentrated toluene/hexanes solution. The molecular structure of 4.3 is shown in Figure 4.2, selected bond lengths and angles are found in Table 4.1. Additionally, 1H NMR data for 4.1, 4.2 and 4.3 are summarized in Table 4.2. In the cases of 4.2 and 4.3, where suitable analyses were not obtained, the proton NMR spectra were used as evidence of both the identity and purity of the compounds; in the former instance, high resolution mass spectrometry was also used to support the assignment. The 1H NMR spectra for 4.1-4.3 contain the expected number of signals, integrating to the appropriate relative intensities, while the

11B chemical shifts are essentially identical. Structurally, 4.3 is very similar to mono- ferrocenyl borane 4.1, with no notable differences in B−N bond lengths or angles.

t Figure 4.2. Thermal ellipsoid plot (30% probability) of 1,1ʹ-Fc[B(N Bu)2]2 (4.3) with hydrogen atoms omitted for clarity, other than those located on the nitrogen atoms.

1 Table 4.2. H NMR chemical shifts (in ppm) for 4.1, 4.2 and 4.3 (in C6D6).

4.1 4.2 4.3

C6H4 N/A 7.21-7.03 N/A

C5H4 4.28, 4.12 4.01, 3.87 4.28, 4.21

C5H5 4.06 4.07 N/A

−NH 3.35 3.41 3.47

−CH(CH3)2 N/A 4.28 N/A

−CH(CH3)2 N/A 1.37-1.04 N/A

−C(CH3)3 1.21 N/A 1.25

4.2.1 Attempted Lithiation of (Diaminoboryl)ferrocenes

The dilithiation of (diaminoboryl)ferrocene 4.1 was first attempted using 2

n t 102 equivalents of BuLi, as this reagent readily deprotonates PhB[N(H) Bu]2. However,

t when carried out under the conditions used to prepare [Li2][PhB(N Bu)2], the lithiation of 4.1 yielded an oily substance comprised of mostly unreacted starting materials along with a mixture of minor products by 1H NMR. Changing the reaction solvent from hexane to either toluene or THF, increasing the temperature or allowing the reaction to proceed for an extended period of time (i.e. > 24 h) produced similar results. These observations suggest that replacing the Ph substituent by the electron-rich metallocene decreases the acidity of the −NH protons to such an extent that nBuLi is insufficient as a metallating reagent.

Subsequently, 4.1 was treated with the stronger base tBuLi. However, the proton

NMR spectrum of the resulting product was very broad and indicated unreacted starting

88 material as well as numerous other species in solution, which likely resulted from competition between metallation of the −NHtBu groups and the Cp ring.165

Since the Dipp substituent on the nitrogen atoms in 4.2 should render the −NH proton more acidic (compared to 4.1), the reaction of 4.2 and two equivalents of nBuLi was carried out in hexane. Although some of the starting material was consumed, the resulting product contained multiple species by 1H NMR. To remove the possibility of the cyclopentadienyl rings undergoing metallation, lithium hexamethyldisilylazide,

166 [Li][N(SiMe3)2], was employed as it does not react with ferrocene; however, this reagent proved to be unreactive towards the –NH protons of 4.2. Given the difficulties encountered in the attempted deprotonation of ferrocenyl boranes 4.1 and 4.2, lithiation of 4.3 was not carried out and, instead, a new linker group was selected in the pursuit of bis-bam ligands.

4.3 Synthesis, Spectroscopic Characterization, X-ray Structure and Reactions of

t t [Li4][(N Bu)2−C6H4−B(N Bu)2]⋅6THF

In light of the observations made when the substituent on boron was a metallocene, phenylene (–C6H4–), was explored as the group connecting the two boron atoms. This spacer was judiciously chosen as the resulting bis-bam is a direct extension

t of [Li2][PhB(N Bu)2] and therefore would be a comparable system with respect to radical generation.

167 The reaction of 1,4-bis(dibromoboryl)benzene, Br2B−C6H4−BBr2, and five equivalents of [Li][N(H)tBu] in toluene, followed by filtration and removal of volatiles in

89 vacuo, produces an oil that solidifies overnight at room temperature to give 4.4 in a good yield (Scheme 4.3). If only four equivalents of amide are used 4.4 is still obtained, however, minor impurities are also present in the reaction mixture. The requirement of excess amide has been found in the synthesis of dilithio bams, where four equivalents are necessary for the reaction to proceed to completion, even though only two amine groups are installed on the boron centre.102,104 In these cases, as well as in that of 4.4, the excess amide is removed by filtering the reaction mixture. Multinuclear NMR and high- resolution mass spectrometry data were consistent with complete replacement of Br substituents by −N(H)tBu groups. The 1H NMR spectrum of 4.4 exhibits singlets at 7.66,

2.88 and 1.16 ppm with relative intensities of 4:4:36, respectively, and a single resonance is observed in the 11B NMR spectrum at 30 ppm.

Tetra-lithiation of 4.4 using 4 equivalents of nBuLi in hexanes results in a colourless precipitate of 4.5, the first example of a tetra-lithio bis-boraamidinate (Scheme

4.3). The molecular structure of 4.5 is illustrated in Figure 4.3 and pertinent bond lengths

t and angles are found in Table 4.3, along with those of {Li2[PhB(N Bu)2]}2 for comparison.

Br Br tBuHN NHtBu t B B + 5 [Li][N(H) Bu] B B + [Li][N(H)tBu] - 4 LiBr Br Br tBuHN NHtBu

4.4

4 nBuLi - 4

4- tBuN NtBu

+ B B 4 [Li]

tBuN NtBu 4.5

Scheme 4.3

Table 4.3. Selected bond lengths (Å) and bond angles (o) for 4.5 and t {Li2[PhB(N Bu)2]}2.

t 102 4.5 {Li2[PhB(N Bu)2]}2

B1−N1 1.437(3) 1.448(3)

B1−N2 1.442(3) 1.449(3)

Li1−N1 1.970(4) 2.077(4)

Li1−N2 1.992(4) 2.052(4)

Li2−N1 2.013(4) 2.027(4)

Li2−N2 1.989(4) 2.022(4)

N1−B1−N2 109.8(2) 109.5(2)

N1−B1−C1 125.5(2) 124.9(2)

N2−B1−C1 124.8(2) 125.5(1)

As depicted in Figure 4.3, the –C6H4− unit of 4.5·6THF lies on an inversion centre. The structural parameters involving the boron centres for 4.5 are nearly identical

t 102 to those of the unsolvated complex {Li2[PhB(N Bu)2]}2. The geometry about boron is planar, but greatly distorted from trigonal, with bond angles in the range 109.8-125.4°.

The B−N bond lengths are intermediate between a single and double bond, as expected.

The arrangement of solvent THF molecules in 4.5 (i.e. one terminal and one bridging

+ THF per Li ion) has been observed previously in the monomeric complex [Li(THF)2(µ-

104 THF)][PhB(NDipp)2]. Although the average N−Li distance in both the unsolvated

t {Li2[PhB(N Bu)2]}2 and solvated [Li(THF)2(µ-THF)][PhB(NDipp)2] are nearly identical, the average length of this bond in bis-bam 4.5 is approximately 0.05 Å shorter than in both of these bam ligands.

Bis-bam 4.5 is extremely air-sensitive and rapidly turns pink upon exposure to air, indicative of oxidation to a radical species (Figure 4.4).111 Even in the presence of only trace amounts of oxygen, (i.e. ~ 8 ppm), a cooled THF solution containing the tetra-lithio reagent turns this pink colour.

Figure 4.4. Solutions of 4.5 under argon (left) and after introduction of air (right).

The controlled oxidation of 4.5 with one-half equivalent of I2 at −80 °C in THF generates a pink solution and the very complex EPR spectrum obtained is indicative of not only what is presumed to be the one-electron oxidation product, but also the presence of a second radical species, even at this low temperature. As the reaction mixture is warmed to room temperature, this latter species is the only paramagnetic entity in solution, which suggests that it is formed by decomposition of the one-electron oxidation product (see Figure 4.5). The 1:1:1 triplet (a = 14.9 G) indicates a radical in which the unpaired electron is coupled to a single 14N centre. Repeated oxidation attempts using iodine consistently resulted in more than one paramagnetic species.

Figure 4.5. EPR spectra of the oxidation of 4.5 with half an equivalent of iodine, collected at various temperatures.

An intricate EPR spectrum has previously been observed for the one electron

t 111 oxidation of [Li2][PhB(N Bu)2]. In that case, an EPR simulation that included interactions with one lithium, two nitrogens and one boron atom best fit the experimentally observed spectrum. On this basis, a structure in which the monoanionic

t - bam radical [PhB(N Bu)2] • coordinates to a solvated lithium cation was assigned (see Eq

1.2); the proposed structure was also supported by DFT calculations.111 Attempts were made to simulate the EPR spectrum of the oxidation of 4.5 collected at 193 K, which was presumed to contain the targeted radical species. The simulation involving coupling of the unpaired electron to two nitrogens, one lithium and one boron (similar to mono-bam radical mentioned above) did not provide a good match for the outer lying peaks in the

EPR spectrum. These were better represented in a simulation that accounted for coupling to two additional nitrogens and one extra boron centre, however, conclusive identity of the radical species was unable to be established.

Metathetical reactions of 4.5 and various main group dihalides were also carried out. Using two equivalents of ZnCl2 and performing the reaction in THF gives 4.4 as the

1 major species while employing CdI2 in THF-d8 results in a H NMR spectrum with multiple broad resonances in the aryl region. The most promising reaction was that involving two equivalents of MgCl2. When carried out on an NMR scale in THF-d8, a new species is formed by 1H NMR spectrum and the chemical shifts of this new product are compared to those of 4.4 and 4.5 in Table 4.4. However, when executed on a preparative scale only bis-bamH4 was isolated, further illustrating the extremely sensitive nature of the tetra-anionic ligand.

1 Table 4.4. H NMR chemical shifts (in THF-d8) for 4.4, 4.5 and 4.5 + 2 MgCl2

4.4 4.5 4.5 + 2 MgCl2

−C6H4− 7.22 6.98 7.48

−C(CH3)3 1.13 0.93 1.00

4.4 Synthesis and Spectroscopic Characterization [Li][PhB{N(H)Dipp}(NDipp)]

(4.6), and [K][PhB{N(H)Dipp}(NDipp)] (4.7), and X-ray Structure of 4.6

The second way in which the bam ligand was chosen to be modified was by single deprotonation to produce a monoanionic bam of the form [M][bam(H)] (M = Li,

n K). The equimolar reaction of PhB[N(H)Dipp]2 with BuLi in hexanes proceeds cleanly to give the mono-lithiated bam, [Li][PhB{N(H)Dipp}(NDipp)], [Li][bam(H)], in high yields. The 1H NMR spectrum is consistent with a mono-lithiated product, i.e. in addition to aryl resonances that integrate to the expected values, there are two septet resonances integrating to 2 H each, a broad singlet assigned to the remaining −NH group and three methyl resonances with relative intensities of 6:6:12. Additionally, both the 11B and 7Li

NMR spectra demonstrate a single resonance at 26.4 and 0.23 ppm, respectively. X-ray quality crystals were obtained from a concentrated THF/hexane solution cooled to ca.

−18°C and the molecular structure is illustrated in Figure 4.6. Selected bond lengths and angles presented in Table 4.5 and compared with PhB[N(H)Dipp]2 (bamH2), and

[Li2(THF)3][PhB(NDipp)2] (Li2bam(THF)3).

Figure 4.6. Thermal ellipsoid plot (30% probability) of

[Li(THF)2][PhB(NHDipp)(NDipp)], 4.6. For clarity, hydrogen atoms on carbon atoms have been omitted and only α-carbons of Dipp groups shown.

o Table 4.5. Selected bond lengths (Å) and bond angles ( ) for 4.6, bamH2 and

Li2bam(THF)3.

104 104 4.6 bamH2 Li2bam(THF)3

B1−N1 1.396(4) 1.420(3) 1.415(5)

B1−N2 1.448(4) 1.414(3) 1.446(4)

N1−Li1 1.908(5) N/A 2.022(6)

N2−Li2 N/A N/A 2.049(7)

N1−B1−N2 121.1(2) 118.5(2) 111.4(3)

N1−B1−C1 119.1(2) 118.7(2) 125.2(3)

N2−B1−C1 119.7(2) 122.8(2) 122.9(3)

B1−N1−C7 120.0(2) 124.6(2) 130.4(3)

B1−N1−Li1 120.0(2) N/A 83.7(3)

C7−N1−Li1 119.1(2) N/A 135.5(3)

C19−N2−B1 130.2(2) 128.0(2) 126.4(3)

C19−N2−Li1 N/A N/A 124.4(3)

B1−N2−Li2 N/A N/A 79.2(2)

N1−Li1−O1 121.1(3) N/A 163.0(4)

N1−Li1−O2 131.3(3) N/A 87.6(2)

O1−Li1−O2 104.3(3) N/A 98.6(3)

In the solid state, the lithium cation in 4.6 coordinates to a single nitrogen atom in a monodentate fashion and is solvated by two THF molecules. As mentioned in Section

1.4.2, monodentate structures of lithium ams are known so this coordination mode is not entirely unexpected; however, it is interesting to note that the lithium ams that crystallize in this manner contain even bulkier groups on nitrogen, such as substituted terphenyls.

Two notable examples for direct comparison with 4.6 are lithium formamidinate

168 169 [Li(THF)2][HC(NDipp)2] and lithium amidinate [Li(THF)2][p-TolC(NDipp)2] (p-

Tol = −C6H5Me), in which the lithium ion interacts with both nitrogen atoms (Figure

4.7). In the formamidinate case, however, monodentate structures were obtained when the THF molecules coordinated to the lithium cation were replaced with TMEDA.170

THF R Ph Li Dipp Dipp B Dipp N N N N THF H Li

Dipp THF THF 4.6 R = H, p-tolyl

Figure 4.7. A comparison of 4.6 with other structurally characterized lithium ams.168,169

The geometry about B1 and N1 in 4.6 is trigonal planar and the distance between these two atoms is shorter than that of B1−N2 by ca. 0.05 Å. There is a small elongation of the B−N bond distance of the protonated nitrogen atom in 4.6 when compared to bamH2, whereas the distance between the anionic nitrogen and boron is slightly shortened compared to the diprotio ligand. As in 4.6, the boron centre in the diprotio starting material is also planar, although there is deviation from a trigonal geometry in the neutral molecule (angles ranging from ca. 118.5°-123°).104

104 When 4.6 is compared to Li2bam(THF)3, the most notable structural difference is the coordination number of the anionic nitrogen and the lithium centre: in the monoanionic bam, both are three-coordinate whereas four-coordinate centres are found in the dianionic ligand. The nitrogen atoms interact with both lithium centres in

Li2bam(THF)3, while the coordination sphere of the lithium cations is completed by two

THF molecules (one terminal and one bridging THF).

There is no notable difference between the B−N bond lengths in 4.6 and the corresponding distances in Li2bam(THF)3, however, the N1−Li length is approximately

0.13 Å shorter in the monoanionic ligand than the average value in the dilithiated bam.

The boron centre in Li2bam(THF)3 is also planar, however, unlike in 4.6 where a trigonal

98 geometry is present, the angles about boron range from approximately 111° to 125° in the dilithio ligand.104

The mono-potassiated derivative, [K][bam(H)] (4.7), was prepared from the 1:1

171 reaction of bamH2 and [K][CH2Ph] in THF. The deep red colour of the benzylpotassium is immediately consumed upon its addition to bamH2, yielding a golden solution. Numerous attempts to crystallize the resulting pale yellow solid failed to produce X-ray quality crystals and, instead, only an amorphous powder precipitated from solution. However, 1H NMR data are consistent with the formation of the mono- potassiated product: aryl resonances integrating to 11 H, overlapping septets integrating to 4 H, a 1 H singlet for the remaining −NH proton and two doublets (6:18 relative intensities) for the methyl protons of the isopropyl groups. Additionally, the 11B NMR spectrum displays a single resonance at 24.9 ppm, which is shifted upfield from the diprotio starting material (29 ppm).104 The 1H NMR data for 4.6 and 4.7 are summarized in Table 4.6.

1 Table 4.6. H NMR chemical shifts (in THF-d8) for monoanionic bams 4.6 and 4.7.

4.6 4.7

Aryl (Ph,Dipp) 7.30, 7.00-6.96, 7.34, 6.95-6.78, 6.80-6.74, 6.58 6.32

−CH(CH3)2 3.91, 3.41 3.79

−NH 3.58 3.62

−CH(CH3)2 1.29, 1.13, 0.81 1.10, 1.03

The mono-potassiated bam can also be prepared using potassium hexamethyldisilylazide (KHMDS) as the deprotonating reagent (see Section 4.5). The utility of both of these new monoanionic bams will be exemplified in Section 5.6.3.

4.5 Synthesis, Spectroscopic Characterization and X-ray Structure of

[K2(THF)3][PhB(NDipp)2], (4.8)

The metathetical chemistry of the boraamidinate ligand has primarily been based on dilithiated species. Because of the less desirable properties of lithium-based ligands outlined in Section 4.1, the dipotassiated bam, [K2][PhB(NDipp)2] (K2bam) was targeted as a new bam reagent; attempted synthetic routes are summarized in Scheme 4.4.

2 KH No Reaction Ph 2 [K][HMDS] Dipp B Dipp [K][bam(H)] + [K][HMDS] N N -HMDS(H) H H

2 [K][CH2Ph] K2bam - 2 PhMe

Scheme 4.4

Potassium hydride was first selected as the deprotonating reagent, but no reaction took place, even upon heating in THF. The reaction of two equivalents of KMHDS and bamH2 gave only the mono-potassiated product, 4.7, with one equivalent of amide remaining. The addition of a deep red solution of benzylpotassium in THF to a solution of bamH2 in the same solvent immediately results in loss of the intense colour, producing

100 a golden coloured solution. The yellow solid that is isolated after removal of volatiles is very soluble in THF and to a lesser extent in toluene, but is insoluble in hexanes and

1 Et2O. The H NMR spectrum of this solid is consistent with double deprotonation and only one resonance is observed in the 11B NMR spectrum at 24.2 ppm. X-ray quality crystals were obtained by cooling a concentrated solution of 4.8 in a mixture of THF and either hexanes or toluene. The structure of a discrete K2bam(THF)3 unit is shown in

Figure 4.8, the extended structure is illustrated in Figure 4.9 and selected structural parameters are presented in Table 4.7.

Figure 4.8. Thermal ellipsoid plot (30% probability) of a discrete

[K2(THF)3][PhB(NDipp)2] unit, K2bam(THF)3 (4.8). For clarity, all hydrogen atoms and carbon atoms of THF solvent molecules have been omitted and only α-carbon atoms on Dipp groups are shown. Symmetry elements used to generate equivalent atoms: −x, y+½, −z+½; −x, y−½, −z+½.

101

Figure 4.9. Representation of the extended solid state structure of 4.8. For clarity all hydrogen atoms and selected carbon atoms have been omitted.

Table 4.7. Selected bond lengths (Å) and bond angles (o) for 4.8

B1−N1 1.407(5) N1−B1−N2 124.7(4)

B1−N2 1.422(5) N1−B1−C1 116.3(3)

N1−K1 2.673(3) N2−B1−C1 119.0(4)

N2−K2 2.709(3) C7−N1−B1 121.6(3)

K1−Centroid 2.905(3) C7−N1−K1 118.8(2)

K2−Centroid 2.834(3) B1−N1−K1 115.4(3)

K1−O1 2.773(4) C19−N2−B1 134.2(3)

K2−O2 2.704(4) C19−N2−K2 101.1(2)

K2−O3 2.655(4) B1−N2−K2 119.8(2)

102

Although three molecules of THF are incorporated into the structure of 4.8 during crystallization, the 1H NMR spectrum of the crude product suggests that the potassium cations are not solvated; however, attempts to crystallize K2bam in the absence of THF to confirm this were unsuccessful. A similar observation has been made previously in the

i preparation of [K][N(H)Trip], (Trip = 2,6−(2,4,6− Pr3C6H2)2C6H3), where although the reaction was carried out in THF, no solvation of the potassium cation was observed in the crude material (by 1H NMR and elemental analysis);172 this potassium amide also was unable to be crystallized in the absence of a coordinating solvent (Et2O).

In the solid state, dipotassiated bam 4.8 forms an extended structure resulting from arene-metal interactions. The potassium cation is coordinated in an η1-amide-η6- arene fashion, similar to that observed for potassium formamidinates

173 174 [K][HC(NMes2){HC(NMes2)H}] and {K[HC(NDipp)2]2K(THF)2}n (Mes = 2,4,6-

175 Me3C6H2) as well as for the potassium amidinate [K][PhC(NSiMe3)(NDipp)]. The potassium-nitrogen bond distances are comparable to that found in [K][HC(NMes2)]

(2.719(1) Å).173 The K(1)−centroid and K(2)−centroid distances of 2.905(3) and 2.834(3)

Å, respectively, are in agreement with the values found for

[K][HC(NMes2){HC(NMes2)H}] and [K][PhC(NSiMe3)(NDipp)].

The distorted trigonal planar geometry about boron is similar to that of the boron

104 atom in Li2bam(THF)3. However, since the nitrogen atoms only interact with a single potassium cation, their geometry differs greatly from that of the dilithiated analog, in which the nitrogen atoms are a part of two puckered four-membered rings with the boron centre and a lithium atom.104 Both nitrogen atoms in 4.8 have bond angles that deviate

103 substantially from a trigonal geometry, ranging from ca. 115°-121° for N1 and 101°-134° for N2. The advantages of using this dipotassiated bam will be exploited in Section 5.4.

4.6 Conclusions

In this chapter, synthetic routes to several new boraamidinate reagents were developed. Three new (diaminoboryl)ferrocenes were prepared, however, difficulties in deprotonation to produce anionic bam species precluded further investigations into these systems as potential precursors to radicals and biradicals. Employing a different π- connecting group, namely –C6H4–, allowed for the successful isolation of the first example of a tetra-lithiated bis-bam. This new reagent is extremely air sensitive and is easily oxidized to a radical species, as indicated by intense pink colouration upon exposure to air or iodine and by EPR spectroscopy. Metathetical reactions of bis-bamLi4 with main group dihalides did not proceed cleanly and new complexes of this bam were unable to be isolated; instead, the tetra-protio bis-bam was typically the major species accounted for by 1H NMR spectroscopy. Nevertheless, the new reagent 4.4 has the potential to be used in condensation reactions with main group element halides to add to the collection of para-phenylene diborane reagents.176,177

Mono-metallated bam ligands have also been presented. Lithiation of bamH2 with one equivalent of nBuLi produces [Li][PhB{N(H)Dipp}(NDipp)], which crystallizes in a monodentate fashion. Although not characterized crystallographically, spectroscopic evidence supports the formation of the mono-potassium bam,

[K][PhB{N(H)Dipp}(NDipp)], from the reaction of bamH2 and one equivalent of

104 benzylpotassium. The utility of these mono-metallated boraamidinates will be demonstrated in Section 5.6.3. Additionally, these new monoanionic bams are not only isoelectronic to the extensively studied amidinate ligand (see Section 1.4.2 and references

11-14) but also have the same overall charge. Consequently, a new avenue for evaluating the influence that replacing carbon by boron has on coordination chemistry is opened.

Finally, a new dianionic bam reagent, [K2][PhB(NDipp)2] was prepared and fully characterized both spectroscopically and in the solid state. X-ray crystallography revealed an extended solid state structure resulting from arene-metal interactions, which has been previously observed for potassium amidinates and formamidinates. The benefits of using a potassiated starting material will be demonstrated in Section 5.4.

4.7 Experimental Section

Reagents, general procedures and instrumentation are described in Appendix 1.

t Preparation of FcB[N(H) Bu]2 (4.1). FcBBr2 (1.37 g, 3.85 mmol) in toluene (15 mL)

t N(H)tBu was added to a stirred slurry of [Li][N(H) Bu] (0.609 g, 7.70 B mmol) in toluene (15 mL) at ca. −80 °C. After 1 h, the cold t Fe N(H) Bu bath was removed and the reaction mixture was allowed to warm to room temperature and stirred for 4 h. After filtration to remove LiBr, volatiles were taken off in vacuo yielding an orange solid (1.005 g, 2.96 mmol, 76%). Orange, X- ray quality crystals of 4.1 were grown from a concentrated hexane/toluene solution at 5 o 1 C. HRMS: Calculated m/z for [C18H29BN2]: 340.1773. Found: 340.1757. H NMR

105

3 3 (C6D6, 25°C): δ 4.28 (t, 2 H, C5H4, JH-H = 1.59 Hz), 4.12 (t, 2 H, C5H4, JH-H = 1.59 Hz),

11 4.06 (s, 5 H, C5H5), 3.35 (2 H, br s, −NH), 1.21 (s, 18 H, −C(CH3)3). B NMR (C6D6,

13 25°C): δ 29.2. C NMR (C6D6, 25°C): δ 74.4 (C5H4), 69.4 (C5H4), 69.2 (C5H5), 49.0

(−C(CH3)3), 33.6 (−C(CH3)3).

Preparation of FcB[N(H)Dipp]2 (4.2). A solution of FcBBr2 (0.720 g, 2.02 mmol) in

toluene (10 mL) was added to a stirred slurry of N(H)Dipp B [Li][N(H)Dipp] (0.742 g, 4.05 mmol) in toluene (15 mL) at ca. Fe N(H)Dipp −80 °C. After 1 h the cold bath was removed and the reaction

was allowed to warm to room temperature and stirred for 18 h.

The reaction mixture was filtered to remove LiBr and volatiles were taken off in vacuo yielding an orange solid (0.746 g, 1.36 mmol, 67%). Orange, X-ray quality crystals of

4.2 were grown from a concentrated toluene solution at 5 oC. Anal. Calcd. for

1 C34H45BN2Fe: C, 74.46; H, 8.27; N, 5.11. Found: C, 74.31; H, 8.05; N, 5.06. H NMR

(C6D6, 25°C): δ 7.21-7.03 (m, 6 H, Dipp), 4.28 (br, 2H, −CH(CH3)2), 4.07 (s, 5 H, C5H5),

3 3 i 4.01 (t, 2 H, C5H4, JH-H = 1.67 Hz), 3.87 (t, C5H4, JH-H = 1.67 Hz), second Pr septet

11 under resonance at 3.87 ppm, 3.41 (2 H, br s, −NH), 1.37-1.04 (br, 24 H, −CH(CH3)2. B

13 NMR (C6D6, 25°C): δ 30.3. C NMR (C6D6, 25°C): δ 147.6 (Dipp), 146.1 (Dipp), 123.6

(Dipp), 73.9 (C5H4), 71.0 (C5H4), 68.9 (C5H5), 28.9 (−CH(CH3)2), 25.7(−CH(CH3)2),

24.7(−CH(CH3)2), 23.7(−CH(CH3)2), 23.1 (−CH(CH3)2).

106

t Preparation of Fc{B[N(H) Bu]2}2 (4.3). A solution of Fc(BBr2)2 (0.268 g, 0.51 mmol) in

toluene (10 mL) was added to a stirred slurry of N(H)tBu B LiN(H)tBu (0.161 g, 2.03 mmol) in toluene (10 mL) N(H)tBu tBu(H)N Fe at ca. −80 °C. After 1 h the cold bath was removed B tBu(H)N and the reaction mixture was allowed to warm to room temperature and stirred for 18 h. The reaction mixture was filtered to remove LiBr and volatiles were taken off in vacuo yielding an orange solid (0.136 g, 0.28 mmol, 54%).

Orange, X-ray quality crystals of 4.3 were grown from a concentrated toluene solution at

o 1 3 5 C. H NMR (C6D6, 25°C): δ 4.28 (t, 4 H, C5H4, JH-H = 1.73 Hz), 4.21 (t, 4 H, C5H4,

3 11 JH-H = 1.73 Hz), 3.47 (4 H, br s, −NH), 1.25 (s, 36 H, −C(CH3)3). B NMR (C6D6,

13 25°C): δ 29.2. C NMR (C6D6, 25°C): δ 75.2 (C5H4), 69.8 (C5H 4), 49.1 (−C(CH3)3),

33.7 (−C(CH3)3).

t t Preparation of [N(H) Bu]2B-C6H4-B[N(H) Bu]2 (4.4). A solution of Br2B−C6H4−BBr2 tBu(H)N N(H)tBu (0.858 g, 2.05 mmol) in toluene (15 mL) was added to B B a slurry of [Li][N(H)tBu] (0.898 g, 10.0 mmol) in tBu(H)N N(H)tBu toluene (10 mL) at ca. −80 °C. After 30 min the reaction mixture was allowed to warm to room temperature and stirred for 18 h. Volatiles were removed in vacuo and the product was extracted in hexanes. After filtration to remove LiBr and excess [Li][N(H)tBu], volatiles were removed in vacuo resulting in an oily solid that completely solidified overnight at room temperature to give 4.4 as a colourless solid (0.597 g, 75%). HRMS:

1 Calculated m/z for [C22H44N2B]: 386.3752. Found: 386.3752 H NMR (C6D6, 25°C): δ

107

1 7.66 (s, 4 H, −C6H4−), 2.88 (br s, 4 H, −NH), 1.16 (s, 36H, −C(CH3)3). H NMR (THF-d8,

11 25°C): δ 7.22 (s, 4H, −C6H4−), 2.93 (br s, 4 H, −NH), 1.13 (s, 36H, −C(CH3)3). B

13 NMR (C6D6, 25°C): δ 30.0. C NMR (C6D6, 25°C): δ ipso carbon not observed, 132.0

(−C6H4−), 49.3 (−C(CH3)3), 33.6 (−C(CH3)3).

t t Preparation of [{Li(THF)}4(µ-THF)2][(N Bu)2B-C6H4-B(N Bu)2], bis-bamLi4 (4.5). A

n 4- 2.5 M solution of BuLi in hexanes (2.23 mL, 5.6 tBuN NtBu

+ B B 4 [Li] mmol) was added to a solution of 4.4 (0.54 g, 1.4 tBuN NtBu mmol) in hexane (10 mL) at 0°C. The reaction was warmed to room temperature, stirred for 18 h and volatiles were removed in vacuo giving a colourless solid. Purification by recrystallization from THF gave 4.5 (0.409 g, 35 %).

X-ray quality crystals were grown from a solution of 4.5 in a THF/Et2O solution. Suitable analysis could not be obtained owing to the highly air-sensitive nature of 4.5. 1H NMR

(THF-d8, 25°C): δ 6.98 (s, 4 H, −C6H4−), 3.61 (−OCH2CH2), 1.77 (−OCH2CH2), 0.93 (s,

11 7 36H, −C(CH3)3). B NMR (THF-d8, 25°C): δ 33.6. Li NMR (THF-d8, 25°C): δ 1.22.

13 C NMR (C6D6, 25°C): δ ipso carbon not observed, 132.3 (−C6H4−), 68.4 (−OCH2CH2),

51.4 (−C(CH3)3), 38.8 (−C(CH3)3), 26.5 (−OCH2CH2).

t t Attempted Preparation of [Mg2][(N Bu)2B-C6H4-B(N Bu)2], bis-bamMg2. In an NMR tube, bis-bamLi4 (0.022 g, 0.026 mmol) and MgCl2 (0.005 g, 0.052 mmol) were heated to

1 at 60 °C in THF-d8 for 2.5 h. After this time, the H NMR spectrum showed a new major product (δ 7.48 ppm, 1.00 ppm) and a minor impurity. When carried out on a

108

1 preparative scale in protio-THF, only bis-bamH4 is present in the H NMR spectrum of the resulting solid.

Preparation of [Li][PhB{N(H)Dipp}(NDipp)], [Li][bam(H)] (4.6). A 2.5 M solution

Ph of nBuLi in hexanes (2.5 mL, 2.5 mmol) was added to a solution

B Dipp Dipp of PhB[N(H)Dipp] (1.01 g, 2.5 mmol) in hexane (15 mL) at ca. N N 2 H −30°C. The reaction was warmed to room temperature over ca. Li 20 minutes and stirred for an additional hour and a half. Removal of volatiles in vacuo gave 4.6 as a colourless solid (0.991 g, 89%). Anal. Calcd. for C30H40BN2Li: C, 80.72;

1 H, 9.03; N, 6.28. Found: C, 79.36; H, 8.98; N, 6.18. H NMR (THF-d8, 25°C): δ 7.30

(m, 2 H, aryl protons), 7.0-6.96 (m, 5 H, aryl protons), 6.8-6.74 (m, 3 H, aryl proton),

3 6.58 (t, 1 H, aryl proton), 3.91 (sept., 2 H, −CH(CH3)2, JH-H = 6.87 Hz), 3.58 (−NH,

3 under residual solvent peak of THF), 3.41 (sept., 2 H, −CH(CH3)2, JH-H = 6.72 Hz), 1.29

3 3 (d, 6 H, −CH(CH3)2, JH-H = 6.87 Hz), 1.13 (d, 6 H, −CH(CH3)2, JH-H = 6.87 Hz), 0.81

3 7 11 (d, 12 H, −CH(CH3)2, JH-H = 6.72 Hz). Li NMR (THF-d8, 25°C): δ 0.23 B NMR

13 (THF-d8, 25°C): δ 26.4. C NMR (THF-d8, 25°C): 156.6 (aryl), 144.0 (aryl), 143.0

(aryl), 141.5 (aryl), 134.2 (aryl), 127.9 (aryl), 126.9 (aryl), 123.1 (aryl), 123.0 (aryl),

122.9 (aryl), 117.7 (aryl), 28.8 (−C(CH3)2), 28.3 (−C(CH3)2), 25.3 (−C(CH3)2), 25.0

(−C(CH3)2), 24.3 (−C(CH3)2).

109

Preparation of [K][PhB{N(H)Dipp}(NDipp)], [K][bam(H)] (4.7). A solution of

171 Ph [K][CH2Ph] (0.166 g, 1.27 mmol) in THF (10 mL) was added

B Dipp Dipp to a solution of PhB[N(H)Dipp]2 (0.562 g, 1.27 mmol) in THF N N H (10 mL) at room temperature. After 10 minutes, volatiles were K removed in vacuo and the resulting product was treated with 10 mL of hexanes. Removal of volatiles in vacuo gave a 4.7 as a cream-coloured solid (0.578 g, 82%, based on

1 4.7·THF). H NMR (THF-d8, 25°C): δ 7.34 (m, 2 H, aryl protons of Ph group), 6.95-6.78

(m, 8 H, aryl protons), 6.32 (t, 1 H, aryl proton), 3.79 (overlapping sept., 4 H,

−CH(CH3)2), 3.62 (m, −OCH2CH2), the −NH is observed at 3.62 ppm when the reaction

3 is carried out in THF-d8, 1.78 (m, −OCH2CH2), 1.10 (d, 6 H, −CH(CH3)2, JH-H = 6.12

3 11 13 Hz), 1.03 (d, 18 H, −CH(CH3)2, JH-H = 5.96 Hz) B NMR (THF-d8, 25°C): δ 24.9. C

NMR (THF-d8, 25°C): 145.7 (aryl), 138.8 (aryl), 134.4 (aryl), 127.3 (aryl), 126.1 (aryl),

123.1 (aryl), 122.7 (aryl), 114.3 (aryl), 68.4 (−OCH2CH2), 28.9 (−C(CH3)2), 28.0 (-

C(CH3)2), 26.5 (−OCH2CH2), 24.9 (−C(CH3)2), 24.3 (−C(CH3)2).

171 Preparation of [K2][PhB(NDipp)2], K2bam (4.8). A solution of [K][CH2Ph] (0.385

Ph g, 2.96 mmol) in THF (10 mL) was added to a solution of

B Dipp Dipp PhB[N(H)Dipp] (0.650 g, 1.48 mmol) in THF (10 mL) at room N N 2 temperature. After 20 minutes volatiles were removed in vacuo K K and the resulting product was treated with 10 mL of hexanes. Removal of volatiles in vacuo gave a 4.8 as a yellow solid (0.705 g, 93%). Anal. Calcd. for C30H39BN2K2: C,

1 69.74; H, 7.61; N, 5.42. Found: C, 68.19; H, 7.63; N, 5.22. H NMR (THF-d8, 25°C): δ

110

7.52 (d, 2 H, ortho- aryl protons of Ph group), 6.89 (m, 2 H, meta- aryl protons of the Ph group), 6.77 (t, 1 H, para- aryl proton of Ph group), 6.64 (d, 4 H, meta- protons of Dipp

3 groups), 6.01 (t, 2 H, para- protons of Dipp groups), 4.04 (sept., 4 H, −CH(CH3)2, JH-H =

3 6.17 Hz), 3.62 (m, −OCH2CH2), 1.78 (m, −OCH2CH2), 1.08 (d, 12 H, −CH(CH3)2, JH-H =

3 11 6.17 Hz), 0.92 (d, 12 H, −CH(CH3)2, JH-H = 6.17 Hz). B NMR (THF-d8, 25°C): δ 24.2.

13 C NMR (THF-d8, 25°C): 162.9 (aryl), 139.2 (aryl), 135.1 (aryl), 126.8 (aryl), 124.0

(aryl), 122.3 (aryl), 109.3 (aryl), 68.4 (−OCH2CH2), 27.8 (−C(CH3)2), 26.8 (−OCH2CH2),

25.2 (−C(CH3)2), 24.8 (−C(CH3)2).

111

Chapter Five: Lanthanide Complexes of a Boraamidinate Ligand

5.1 Introduction

Although monoanionic ams and guans are well known to form complexes with the lanthanides and actinides,11,53 the chemistry of bams with f-block metals remains unexplored.ii However, numerous dianionic ligands have been incorporated into lanthanide complexes and structurally characterized, including the following: the cyclooctatetraene dianion;178 the 1,4-cyclohexa-2,5-dienyl (benzene 1,4-dianion) ligand;179,180 the antiaromatic 2,2’-bipyridine dianion (bipy2-);181 tetradentate, dianionic

Schiff bases;182-192 chelating diamides;193-197 binaphtholates198 and bridged cyclopentadienyl ligands.198 Most recently, lanthanide complexes containing an indenyl- pyrrolyl ligand linked by a methylene spacer (5.1),199 a β-ketoiminate possessing an N- aryloxo substituent (5.2),200 1,2-bis[(2,6-diisopropylphenyl)imino]acenaphthene dianion

(5.3),201,202 derivatives of β-diketiminato with pendant amide donors (5.4),203 and a bis(amido)cyclodiphospha(III)azane (5.5)204 have been reported.

ii The group of Prof. Dr Sjoerd Harder was concurrently working on lanthanide complexes of the 2- boraamidinate ligand [HB(NDipp)2] while the work in this thesis was being completed.

112

O Ar N

O N N N

Ar 5.1 5.2 5.3

R' R R' N N N HN N Dipp P P N N R R 5.4 5.5

As mentioned in Section 1.4.4, there is growing interest in using amidinate and guanidinate complexes of the lanthanides as catalysts for various processes such as ring- opening polymerization of cyclic esters55,59,60,205-207 and olefin polymerization.71,208

Hybrid bam/am organomagnesium complexes have shown promise as initiators for the polymerization of rac-lactide.105 Homoleptic guanidinate lanthanide complexes are also being investigated as potential precursors for thin films of rare-earth oxides.82 It is therefore of interest to pursue analogous lanthanide(III) complexes containing bams to examine the structural and electronic differences introduced by this dianionic ligand.

Specifically, a more open coordination environment around the Ln3+ centre, provided by replacing two monoanionic ligands with a dianionic one, may lead to increased catalytic activity.209

113

Samarium(II) bam complexes are also novel synthetic targets. The divalent

210,211 212 samarium species SmI2 and Cp2Sm have found extensive use in bond-forming processes in both organic and organometallic chemistry.213-216 In recent years, increased attention has been given to the replacement of the organic Cp groups with other anionic ligands.175,217-221 In 2005, the first Sm(II) bis-formamidinate complex,

218 [HC(NDipp)2]2Sm(THF)2, was prepared and structurally characterized; two years later,

219 the related guanidinate complex, [Cy2NC(NDipp)2]2Sm, was reported. In the former example, the samarium centre is six-coordinate, while the latter case represented the first example of a planar, four-coordinate lanthanide(II) centre. The bis-guan was unreactive towards many unsaturated substrates, however, it reacted with CS2, to produce the

2 2 samarium(III) dimer [Cy2NC(NDipp)2]2Sm(µ-η -:η -S2CSCS)Sm[Cy2NC(NDipp)2]2

(5.6), through reductive coupling and C−S bond formation.221

Cy2N NCy2 C Dipp C N Dipp Dipp N S Dipp N N Sm S Sm Dipp Dipp N N N S S N C Dipp C Dipp Cy N 2 NCy2 5.6

Unsymmetrical benzamidinate complexes of Sm(II) have been shown to act as one-electron reductants towards iodine, diphenyl dichalchogenides and carbodiimides.175

Bis(phosphinimino)methanides have also been used as ligands in Sm(II) chemistry217,220 and heteroleptic complexes of this ligand have been shown to initiate the polymerization of ε-caprolactone.220 Concurrent to this work, the group of Harder synthesized a Sm(II)

114

2- 222 complex of the boraamidinate ligand [HB(NDipp)2] ; the preparation of the analogous

2- species containing the [PhB(NDipp)2] ligand would allow for the examination of the effects of the substituent on boron.

This chapter begins with the synthesis and structural characterization of the

2- dimeric “ate” complexes {[Li(THF)4][bamLnCl2(THF)]}2, (bam =[PhB(NDipp)2] , Ln =

Y, Pr, Nd, Sm, Ho, Er, Yb), representing the first use of the bam ligand in f-block

2- chemistry. The N-Dipp substituted ligand [PhB(NDipp)2] was chosen in preference to

t 2- the more widely studied N-tert-butyl analogue [PhB(N Bu)2] in an effort to provide better steric protection of low-coordinate Ln centres.85 Reactions performed in attempts to replace the chloride for an alkyl or amino group are also discussed.

Subsequently, the synthesis and structural characterization of the neutral dimers

{bamLnCl(THF)2}2 (Ln = Y, Sm), is described and reactions of these complexes are detailed. The preparation of neutral bamLnI complexes is then presented and halide substitution reactions are examined, followed by a brief investigation into 2:1 bam:metal complexes. Finally, the attempted preparation of a samarium(II) bam complex by metathesis, reduction or protonolysis is discussed.

5.2 Synthesis, Spectroscopic Characterization and X-ray Structures of

2- {[Li(THF)4][bamLnCl2(THF)]}2, (bam = [PhB(NDipp)2] , Ln = Y, Pr, Nd, Sm, Ho,

Er, Yb

2- Lanthanide(III) complexes of [PhB(NDipp)2] were first targeted in the pursuit of introducing the bam ligand into f-block chemistry. The reaction of [Li2][PhB(NDipp)2],

115

104 Li2bam, with an equimolar amount of LnCl3 (Ln = Y, Sm, Pr, Nd, Ho, Er, Yb) in THF produces the markedly air- and moisture-sensitive lanthanide “ate” complexes 5.7-5.13

(Eq. 5.1) in yields ranging from 56 to 84 %. The complexes are readily soluble in THF, but are insoluble in Et2O and hexane, precluding the removal of the LiCl by-product from the bulk solid.

THF, -80oC LnCl + Li bam 1/2 {[Li(THF) ][bamLnCl (THF)]} .(THF) (5.1) 3 2 - LiCl 4 2 2 2

Ln = Y (5.7), Pr (5.8), Nd (5.9), Sm (5.10), Ho (5.11), Er (5.12), Yb (5.13)

Crystals of all complexes, suitable for X-ray diffraction, were grown from a THF solution layered with hexanes at ca. −18°C. In the solid state, 5.7-5.13 exist as isostructural, ion-separated complexes in which the anions dimerize via chloride bridges.

The lithium cations are solvated by four THF molecules and two additional molecules of this solvent are present in the crystal lattice. A representative molecular structure is shown in Figure 5.1 and selected bond lengths and angles are presented in Tables 5.1 and

5.2, respectively.

116

Table 5.1. Selected bond lengths (Å) for 5.7-5.13.

5.7 (Y) 5.8 (Pr) 5.9 (Nd) 5.10 (Sm) 5.11(Ho) 5.12(Er) 5.13(Yb) B1−N1 1.435(6) 1.427(7) 1.438(5) 1.440(6) 1.438(6) 1.435(10) 1.447(7)

B1−N2 1.437(6) 1.436(7) 1.451(5) 1.439(6) 1.437(6) 1.448(10) 1.441(7)

Ln−N1 2.245(3) 2.316(4) 2.324(4) 2.297(3) 2.245(3) 2.232(6) 2.208(4)

Ln−N2 2.225(3) 2.335(4) 2.304(3) 2.283(3) 2.223(4) 2.215(6) 2.203(4)

Ln−Cl2 2.717(1) 2.828(2) 2.808(1) 2.779(1) 2.708(1) 2.699(2) 2.686(2)

117

Ln−Cl2A 2.764(1) 2.871(2) 2.859(1) 2.826(1) 2.761(1) 2.755(2) 2.727(2)

Ln−Cl1 2.602(2) 2.713(2) 2.696(2) 2.665(2) 2.596(2) 2.586(2) 2.572(2) Table 5.2. Selected bond angles (o) for 5.7-5.13

5.7 (Y) 5.8 (Pr) 5.9 (Nd) 5.10(Sm) 5.11(Ho) 5.12(Er) 5.13(Yb) N1−B1−N2 113.0(4) 113.9(5) 112.6(3) 113.3(4) 112.4(4) 111.9(7) 112.2(5)

N1−B1−C1 123.7(4) 123.8(5) 124.1(3) 123.9(4) 123.9(4) 125.1(7) 124.2(5)

N2−B1−C1 123.3(4) 122.3(5) 123.1(3) 122.8(4) 123.7(4) 123.0(7) 123.6(4)

C7−N1−B1 129.4(3) 130.9(5) 129.8(3) 130.0(4) 129.1(4) 128.5(6) 128.4(4)

C7−N1−Ln 138.1(3) 135.4(3) 136.1(2) 136.6(3) 137.9(3) 138.4(5) 138.8(3)

B1−N1−Ln 90.6(2) 91.6(3) 92.0(2) 91.2(2) 90.9(3) 91.2(4) 90.6(3)

B1−N2−Ln 91.4(2) 92.1(3) 92.5(2) 91.8(2) 91.8(3) 91.6(4) 91.0(3)

C19−N2−B1 125.4(3) 127.5(4) 126.6(3) 127.0(3) 125.8(4) 125.5(6) 125.4(4)

C19−N2−Ln 142.3(3) 139.4(3) 139.8(2) 140.0(3) 141.4(3) 142.1(5) 142.6(3)

N1−Ln−N2 64.8(1) 62.1(1) 62.6(1) 63.3(1) 64.7(1) 65.0(2) 65.8(1)

N1−Ln−O1 88.2(1) 87.1(1) 87.1(1) 87.4(1) 87.9(1) 88.3(2) 88.2(2)

N2−Ln−O1 152.8(1) 148.9(1) 149.4(1) 150.5(1) 152.4(1) 153.1(2) 153.8(2)

N1−Ln−Cl1 100.8(1) 101.4(1) 101.3(1) 101.3(1) 100.7(1) 100.5(2) 100.6(1)

N2−Ln−Cl1 103.1(1) 105.2(1) 104.8(1) 104.3(1) 102.8(1) 102.8(2) 102.4(1)

O1−Ln−Cl1 84.5(1) 83.9(1) 84.0(1) 84.0(1) 84.3(1) 84.3(2) 84.5(1)

N1−Ln−Cl2 99.3(1) 99.8(1) 99.8(1) 99.4(1) 99.5(1) 99.4(2) 99.2(1)

N2−Ln−Cl2 97.4(1) 99.6(1) 99.5(1) 98.9(1) 98.0(1) 97.7(2) 97.1(1)

O1−Ln−Cl2 82.9(1) 80.4(1) 80.7(1) 81.4(1) 82.9(1) 83.1(1) 83.7(1)

Cl1−Ln−Cl2 155.9(1) 152.9(1) 153.2(4) 154.1(1) 155.7(1) 156.0(1) 156.6(1)

118

In all complexes, the B−N distances are equal within experimental error and have values that are typical for N(2p)→B(2p) π-bonding in boraamidinate complexes.85 The geometry about boron is planar but distorted from trigonal, with the N1−B1−N2 angle falling in the range of ca. 112-114°, while the N1−B1−C1 and N2−B1−C1 bond angles are between 122-125°. The nitrogen atoms are substantially distorted from a trigonal planar geometry with small B−N−Ln angles of ca. 91-92° and large C−N−B and

C−N−Ln angles in the ranges of ca. 125-131° and 135-142°, respectively. The

N1−Ln−N2 angle increases going from left to right across the lanthanides series, as a result of the decreasing size of the metal centre; this trend holds for complex 5.7, as the radius of yttrium falls between that of holmium and erbium.

A search of the Cambridge Structural Database reveals a wide range of Ln−N distances; the values of the Ln−N bonds in 5.7-5.13 fall on the shorter end of this range, as expected for a chelating dianionic ligand. For instance, the Ln−N distances in 5.7 and

5.10 are nearly identical to those in the lanthanide bam complexes

[HB(NDipp)2]Ln(DMA), (Ln = Y, Sm; DMA = 2-(dimethylamino)benzyl) recently reported by the Harder group.223 Additionally, 5.7 and 5.10-5.12 have comparable Ln−N distances to those reported by Roesky and co-workers who used a dianionic bis(amido)cyclodiphosph(III)azane ligand 5.5 (R, Rʹ = tBu) to form a series of lanthanide complexes, including “ate” complexes with yttrium, samarium, holmium and erbium.204

For example, the Y−N1 and Y−N2 bond lengths of 2.287(5) Å and 2.264(5) Å in the yttrium cyclodiphospha(III)azane “ate” complex204 are similar to the corresponding distances in 5.7 (2.225(3) Å and 2.245(3) Å), and fall on the shorter end of the value

119 expected for an yttrium-nitrogen bond. Similarly, the Ho−N distances of 2.223(3) Å and

2.245(4) Å in 5.11 are consistent with those in the analogous complex prepared by the

Roesky group (2.275(3) Å and 2.263(2) Å),204 and again are at the lower end of the range of holmium-nitrogen bonds.

For complexes 5.7-5.13, the Ln−Cl2 and Ln−Cl2A bond lengths are typical for chlorides bridging the respective lanthanide centres. In all complexes, the bridging is nearly symmetrical, with differences in Ln−Cl2 and Ln−Cl2A lengths ranging from 0.041

Å to 0.056 Å. Overall, these two distances decrease by approximately 0.14 Å from praseodymium to ytterbium. In each complex, the terminal Ln−Cl1 bond lengths are slightly shorter than those typically found in the Cambridge Structural Database and going across the series of lanthanides decrease by a total of approximately 0.14 Å.

The geometry at the lanthanide centre in 5.7-5.13 is distorted octahedral, with bond angles ranging from approximately 62° to 156°. There is no notable difference in the N2−Ln−Cl1 angles in 5.7−5.13, however, both N2−Ln−Cl1 and N2−Ln−Cl2 decrease slightly across the series by a total of ca. 2.8° and 2.5°, respectively. The remaining bond angles about the metal centre increase moving from praseodymium to ytterbium.

Unlike 5.8-5.13, the yttrium boraamidinate complex 5.7 is diamagnetic and easily characterized by multinuclear NMR (1H, 7Li, 11B, 13C). The proton NMR spectrum of 5.7

(Figure 5.2) shows the expected multiplets in the aryl region for the phenyl substituent on boron and Dipp groups on nitrogen, which integrate to the appropriate relative intensities, as well as a septet (4 H) and two doublets (12 H), indicative of diastereotopic methyl protons of the Dipp substituents. One of these signals overlaps with the −CH3 resonance

120 of residual Et2O, present from washing, resulting in a larger than expected integration value. Multiplets resulting from the THF molecules are also observed at 3.61

7 11 (−OCH2CH2) and 1.78 (−OCH2CH2) ppm. The Li NMR and B NMR spectra show singlets at 0.11 and 32.6 ppm, respectively; the former is consistent with a solvated Li+ cation, while the latter is in the typical range for bam complexes.85

Despite the paramagnetic nature of complexes 5.8, 5.9 and 5.10, 1H NMR spectra were obtained in a THF-d8 solution; a summary of chemical shifts is presented in Table

5.3. As shown in Figures 5.3-5.5, the resonances for the paramagnetic complexes are substantially broadened and in the cases of the 5.8 and 5.9 there is no resolvable 1H−1H coupling. Additionally, a lanthanide-induced shift224 of the aryl and alkyl protons is observed, with the former resonances being deshielded while the latter are shielded.

When compared to the diamagnetic yttrium complex, the greatest shifts are observed for praseodymium and neodymium analogues. The most deshielded aryl resonance differs by ca. 5.5 and 4.6 ppm, respectively, the methylene protons are shielded by approximately

6.7 ppm in 5.8 and 5.3 ppm in 5.9, and the methyl resonances are found 8.6 ppm (5.8) and 7.0 ppm (5.9) upfield from those in diamagnetic 5.7. In all complexes, integration of the proton NMR spectrum gives values consistent with the solid-state structure: 11 aryl protons, 4 methyne protons, 24 methyl protons (two resonances, 12 H each) and THF.

The shifts observed for the samarium complex 5.10 are less pronounced; aryl resonances are deshielded by ≤2 ppm while aliphatic protons are shielded by <1 ppm. In contrast, NMR spectra were unable to be obtained for 5.11-5.13, as the paramagnetic nature of these samples even prevented the instrument from locking. The 11B NMR resonance for the three-coordinate boron in the bam ligand is normally quite broad and

121 attempts to collect 11B NMR spectra for 5.8-5.13 were unsuccessful, likely a result of further broadening caused by the paramagnetic metal centre.

1 Table 5.3. H NMR chemical shifts (in ppm) for complexes 5.7-5.10 (in THF-d8)

5.7 (Y) 5.8 (Pr) 5.9 (Nd) 5.10 (Sm) Aryl 6.87, 6.73, 12.35, 10.95, 11.47, 10.25, 8.98, 7.31, 7.27, 6.61, 10.55, 8.39, 7.15 8.93, 7.11, 6.34 6.63, 6.54 6.55, 6.47

−CH(CH3)2 3.98 -2.76 -1.32 3.31

−CH(CH3)2 1.12, 0.71 0.95, -7.93 1.12, -6.35 1.05, -0.06

1 Figure 5.2. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamYCl2(THF)]}2 (5.7)

122

1 Figure 5.3. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamPrCl2(THF)]}2 (5.8)

1 Figure 5.4. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamNdCl2(THF)]}2 (5.9)

123

1 Figure 5.5. H NMR spectrum (in THF-d8) of {[Li(THF)4][bamSmCl2(THF)]}2 (5.10)

5.3 Synthesis, Spectroscopic Characterization and X-ray Structures of

{bamLnCl(THF)2}2 (Ln = Y, Sm)

With novel lanthanide “ate” complexes of the bam ligand readily available, attention was turned to preparing alkyl and amide derivatives. Since the samarium complex 5.10 demonstrates the least amount of broadening in the 1H NMR spectrum,

(when compared to the praseodymium and neodymium analogues), it was selected as a starting point in this pursuit, along with diamagnetic yttrium “ate” 5.7. The treatment of

5.7 or 5.10 with either MeLi or [Li][HMDS] (HMDS = hexamethyldisilylazide) in a 1:1 stoichiometry resulted in complex reaction mixtures. Increasing the amount of alkyllithium or lithium amide in attempts to replace both chloride ligands also produced more than one species, indicated by 1H NMR spectroscopy.

124

Heating lanthanide “ate” complexes containing a dianionic bis(amido)cyclodiphospha(III)azane ligand in toluene has been shown to force the elimination of LiCl to generate neutral species.204 It was proposed that neutral bamLnCl complexes could be produced in this manner and then used for subsequent metathetical reactions. However, boraamidinate lanthanide complexes 5.7-5.13 display very limited

1 solubility in toluene-d8, and the H NMR spectrum of these solutions (for 5.7-5.10) displays multiple sets of broad resonances that cannot easily be assigned, both at room temperature and after heating.

The neutral complexes {bamLnCl(THF)2}2 (5.14: Ln = Y, 5.15: Ln = Sm) were instead prepared by the reaction of 5.7 or 5.10 with an excess of trimethylsilyl trifluoromethanesulfonate (TMSOTf) in THF (Eq. 5.2); the formation of 5.14 and 5.15 does not proceed to completion when a stoichiometric amount of TMSOTf is employed.

Trimethylsilyl trifluoromethanesulfonate has been used in metathetical reactions for anion exchange;225 however, to the best of our knowledge, this is the first example of removal of LiCl by this triflate reagent. Interestingly, the reaction of 5.10 with silver trifluoromethanesulfonate (AgOTf) gives multiple products by 1H NMR, none of which correspond to the neutral complex.

THF (5.2) {[Li(THF)4][bamLnCl2(THF)]}2 + 4 TMSOTf {bamLnCl(THF)2}2 + 2 TMSOTf - 2 TMSCl - 2 [Li][OTf] 5.14, Ln = Y 5.15, Ln = Sm

Colourless and deep orange X-ray quality crystals of 5.14 and 5.15, respectively, were grown from a THF solution layered with hexanes. The molecular structure of 5.15,

125 which is also representative of 5.14, is shown in Figure 5.6 and selected bond lengths and bond angles are listed in Table 5.4.

Figure 5.6. Thermal ellipsoid plot (30% probability) of 5.15 (representative of 5.14). For clarity, hydrogen atoms and selected THF carbon atoms have been omitted and only the α-carbon atoms of Dipp groups are shown. Symmetry elements used to generate equivalent atoms: #1 −x,−y+3,−z+1.

126

o Table 5.4. Selected bond lengths (Å) and bond angles ( ) for {bamLnCl(THF)2}2.

5.14 (Ln = Y) 5.15 (Ln = Sm) B1−N1 1.447(6) 1.443(8) B1−N2 1.451(6) 1.468(8) N1−Ln 2.218(4) 2.266(5) N2−Ln 2.204(4) 2.250(5) Cl1−Ln 2.684(1) 2.749(2) Cl1A−Ln 2.781(1) 2.846(2) N1−B1−N2 112.8(4) 112.4(5) N1−B1−C1 124.0(4) 124.6(5) N2−B1−C1 123.1(4) 123.0(6) C7−N1−B1 125.0(4) 124.7(5) C7−N1−Ln 144.6(3) 143.7(4) B1−N1−Ln 90.3(3) 91.4(3) C19−N2−B1 125.9(4) 126.1(5) C19−N2−Ln 140.3(3) 139.1(4) B1−N2−Ln 90.7(3) 91.4(3) N2−Ln−N1 66.2(1) 64.7(2) N2−Ln−O2 99.6(1) 101.2(2) N1−Ln−O2 98.7(1) 98.4(2) N2−Ln−O1 155.2(1) 153.3(2) N1−Ln−O1 89.8(1) 89.1(2) O2−Ln−O1 77.4(1) 76.0(2) N2−Ln−Cl1 106.2(1) 107.1(1) N1−Ln−Cl1 109.4(1) 110.6(1) O2−Ln−Cl1 147.6(1) 146.0(1) O1−Ln−Cl1 86.7(1) 86.7(1)

127

The values of the B−N distances in both 5.14 and 5.15 fall in the range expected for N(2p)→B(2p) π-bonding and, within experimental error, are identical to those of the corresponding “ate” complexes. Similar to 5.7 and 5.10, the sum of the bond angles about boron in 5.14 and 5.15 is 360° and the geometry is distorted from trigonal. Additionally, the N1−Ln−N2 angle is slightly larger in both cases (by ca. 1.5°) when compared to 5.7 and 5.10. The geometry about the lanthanide centre is greatly distorted from octahedral, as observed for the “ate” complexes, with bond angles ranging from approximately 65° to

155°.

In both 5.14 and 5.15 the Ln−Cl1 bond length is slightly shorter and Ln−Cl1A is slightly longer than the corresponding values in 5.7 and 5.10; however, both distances still fall in the expected range for a bridging Y−Cl and Sm−Cl bond. The asymmetry in the chloride bridging is more pronounced in 5.14 and 5.15 than in the “ate” complexes, with the difference in the two Ln−Cl bond lengths being around 0.1 Å.

Yttrium complex 5.14 was characterized by multinuclear NMR spectroscopy (1H,

11B, 13C), and displayed no significant changes in chemical shifts compared to “ate” complex 5.7. Like samarium “ate” 5.10, the 1H NMR spectrum of neutral 5.15 can be obtained, despite it being paramagnetic. A summary of the 1H NMR chemical shifts for both 5.14 and 5.15 can be found in Table 5.5, along with those of the corresponding “ate” complexes, for comparison. A similar degree of deshielding of aryl resonances and shielding of alkyl resonances to that noticed for 5.10 is found for 5.15 and all signals are significantly broadened to the extent that no resolvable 1H-1H coupling is observed, a consequence of the paramagnetic samarium(III) centre.

128

Subsequent reactions of 5.15 with lithium or potassium bis(trimethylsilyl)amide and with lithium diisopropylamide did not proceed cleanly. Regardless, the possibility of salt by-product incorporation remained, especially in the cases where lithium reagents were employed. In order to avoid this, focus was shifted towards the use of iodide starting materials for metathetical reactions with the newly prepared K2bam reagent

(4.8).226,227

5.4 Synthesis and Spectroscopic Characterization of {bamLnI(THF)n}m

In Section 4.1, the advantages of using a potassiated starting material instead of a lithiated reagent in metathetical reactions were described; then, in Section 4.5, the first example of a dipotassiated boraamidinate was presented. The reaction of K2bam 4.8 with an equimolar amount of either YI3(THF)3.5 or SmI3(THF)3.5 in THF (Eq. 5.3) results in the precipitation of KI and 1H NMR spectra that support the formation of

1 {bamYI(THF)n}m (5.16) and {bamSmI(THF)n}m (5.17). A summary of relevant H

NMR chemical shifts for 5.16 and 5.17 is found in Table 5.5 along with those of the corresponding data for the respective “ate” complexes and neutral bamLnCl(THF)2 analogues. By comparison to chloro-complexes 5.14 and 5.15, neutral complexes 5.16 and 5.17 are likely dimers in the solid state; however, attempts to grow X-ray quality crystals to confirm the molecular structures of these new products were unsuccessful.

THF LnI3(THF)3.5 + K2bam {bamLnI(THF)n}m (5.3) - 2 KI 5.16: Ln = Y 5.17: Ln = Sm

129

Table 5.5. 1H NMR chemical shifts (in ppm) for the isopropyl groups in the neutral

bamLnX(THF)2 complexes 5.15-5.17, and “ate” complexes 5.7 and 5.10 (in THF-d8)

5.7 5.14 5.16 5.10 5.15 5.17 Y “ate” Ln = Y Ln = Y Sm “ate” Ln = Sm Ln = Sm X = Cl X = I X = Cl X = I −CH(CH3)2 3.98 3.85 3.72 3.31 3.04 2.51

−CH(CH3)2 1.12, 1.12, 1.12, 1.05, 0.97, 0.62, 0.71 0.72 0.66 −0.06 0.03 0.06

Complexes 5.16 and 5.17 are isolated as pale yellow and burnt orange solids, respectively, and are soluble in THF but insoluble in Et2O and hexanes. Metathesis

t reactions of 5.16 or 5.17 with either [K][HMDS], [K][CH2C6H5], or [K][O Bu] were carried out in attempts to prepare derivatives of the neutral lanthanide bam complexes. In the case of 5.16, all reactions gave a colourless precipitate upon mixing of the two starting materials in THF-d8. While the alkoxide reaction produced a complex mixture

(by 1H NMR spectroscopy), the alkyl and amide reactions gave a new major product, with some impurities. The reaction with benzyl potassium showed the most promise, exhibiting the expected 2 H doublet in the 1H NMR spectrum, corresponding to the methylene protons of the benzyl group coupling to the spin-½ yttrium nucleus. However, when carried out on a preparative scale, the reaction gave rise to several products, none of which could be separated successfully. The analogous reactions in THF-d8 using the neutral samarium complex 5.17 also produced a colourless precipitate, however, complex mixtures were observed in all cases.

130

5.5 Synthesis and Spectroscopic Characterization of [Li2][bam2LnX] (Ln = La, X =

I; Ln = Y, Sm, X = Cl)

The final lanthanide(III) complexes of the bam ligand that were targeted were the

2:1 (ligand:metal) products, [Li][{PhB(NDipp)2}2Ln], with the intention of comparing the one electron oxidation of these species with the previously reported group 13 radicals

t • 110 [M{PhB(N Bu)2}] (M = Al, Ga) (see Eq 1.1, Section 1.5.4). Given that lanthanum(III) and yttrium(III) are diamagnetic and that 1H NMR spectra can be collected on samarium(III) bam species, these three metals were selected for the 2:1 reactions, which were carried out in THF (Scheme 5.1). Although the reaction between Li2bam and LaCl3 did not proceed, even after extended periods of time, a new product (5.18) was produced when the reagent LaI3(THF)3.5 was employed (Scheme 5.1).

Dipp 2 + o Dipp LaI3(THF)3.5 2 Li2bam THF, -80 C N X N - 2 LiI Ph B Ln B Ph 2 [Li] N N

Dipp Dipp - 2 LiCl

o LnCl3 + 2 Li2bam THF, -80 C 5.18: Ln = La, X = I 5.19: Ln = Y, X = Cl 5.20: Ln = Sm, X = Cl

Scheme 5.1

X-ray quality crystals of the lanthanum complex were grown from a solution of

5.18 in THF, layered with hexanes. The molecular structure is illustrated in Figure 5.7 and selected structural parameters are found in Table 5.6. Complex 5.18 exhibits

131

2− equivalent bam ligands in the dianion, [(PhB{NDipp}2)2LaI] , with the Li−I unit on a crystallographic symmetry centre. An equivalent of the by-product LiI is incorporated into the structure resulting in a 2− charge for the anion, which is balanced by two tetra-

+ solvated lithium cations, [Li(THF)4] . Although a coordination number of 5 is uncommon for the lanthanide elements,5 the steric bulk of the Dipp substituents likely prevents further coordination at the metal centre.

132

Table 5.6. Selected bond lengths (Å) and bond angles (o) for 5.18.

N1−B1 1.437(6) N1−B1−N2 116.1(4) N1−La1−N2 59.6(1)

N1−C7 1.378(6) N1−B1−C1 120.6(4) N1−La1−I1 114.4(1)

N1−La1 2.417(4) N2−B1−C1 123.2(4) N2−La1−I1 105.5(1)

N2−B1 1.432(7) B1−N1−La1 92.6(3) B1−N2−La1 90.3(3)

N2−C19 1.402(6) C7−N1−B1 123.6(4) C19−N2−B1 125.4(4)

N2−La1 2.476(3) C7−N1−La1 143.8(3) C19−N2−La1 141.7(3)

B1−C1 1.609(7)

La1−I1 3.195(1)

As expected, the B−N bond length in 5.18 is intermediate between a single and double bond. The La−N distances of 2.417(4) Å and 2.476(5) Å in 5.18 are shorter than the corresponding distance in the lanthanum bis-formamidinate complex

{HCN(Dipp)2}2LaF(THF), reported by Junk and coworkers (which ranged from ca.

2.538 to 2.578 Å).228 This is presumably a consequence of the dianionic charge in the

t t bam ligand, as the analogous distance in the complex [(BuNP)2( BuN)2LaCl(THF)]2, which contains a dianionic ligand, averages to ca. 2.38 Å.204 The La−I bond length of

3.195(1) Å is comparable to that observed for the heteroleptic complexes

[(Me3Si)2{Me2(Me2N)Si}C]2LaI (3.145(1) Å) and [(Me3Si)2{Me2(MeO)Si}C]2LaI(THF)

(3.173(1) Å)229 and falls in the range of a typical La−I bond.

The geometry about the boron and nitrogen atoms is planar but distorted from trigonal; this distortion is small at the boron centre, but is greatly pronounced at nitrogen

133 where the bond angles range from ca. 93-144°. The lanthanum centre in the bis-

228 formamidinate complex {HCN(Dipp)2}2LaF(THF) is six-coordinate, owing to solvation by one THF molecule; however, the metal centre in 5.18 is five-coordinate, which is a less common coordination number for the lanthanide elements.5 This is likely a consequence of the phenyl substituent on boron, instead of a hydrogen as in the formamidinate, which increases the steric interactions with the Dipp substituents, resulting in the isopropyl groups preventing additional coordination at the metal centre.

The analogous 2:1 reactions between Li2bam and YCl3 or SmCl3 (Scheme 5.1) proceeded to give 5.19 and 5.20, respectively, assigned on the basis of 1H NMR spectra; relevant data are summarized in Table 5.7.

Table 5.7. 1H NMR chemical shifts (in ppm) for the isopropyl groups 5.18-5.20

(in THF-d8).

5.18 (La) 5.19 (Y) 5.20 (Sm)

−CH(CH3)2 3.55 3.91, 3.40 3.92, 3.38

−CH(CH3)2 0.88, 0.63 1.28, 1.12, 0.81 1.28, 1.13, 0.81

The 1H NMR spectra of 5.18-5.20 show the expected aryl resonances for the Dipp and Ph substituents, as well as septets and doublets for the isoporpyl groups, which integrate to the appropriate relative intensities. The lanthanum complex 5.18 exhibits two

1 doublets for the −CH3 groups in the H NMR spectrum, indicative of diastereotopic isopropyl substituents, while the yttrium and samarium species demonstrate three doublet resonances with relative intensities of 2:1:1 suggesting disparity in the orientation of the

Dipp units between the two complexes. In addition, resonances for THF molecules are

134 observed for the solvated Li+ cations and a single resonance is found in the 7Li NMR spectrum of 5.18-5.20.

In all three cases, the 1H NMR spectra are significantly different than those of the products obtained using the corresponding 1:1 reactions; this supports the assignments of

5.19 and 5.20 as the 2:1 products, where the molecular structure was unable to be confirmed by X-ray analysis.

The one electron oxidation of 5.18-5.20 with half an equivalent of molecular iodine was performed in Et2O at ca. −80°C. In contrast to the reactivity of the group 13 species,110 no colour change characteristic of a radical species was observed. However, halide coordination to the lanthanide centre results in this not being a direct comparison to the group 13 complexes and removal of the halide ligand would be necessary for a more accurate evaluation.

5.6 Attempted Synthesis of a bamSmII Complex

5.6.1 Metathesis Reactions

The first approach used in efforts to prepare a bamSmII complex was the

104 metathesis reaction between Li2bam and an equimolar amount of SmI2 in THF. The

1H NMR spectrum of the dark brown product obtained was found to be consistent with that of 5.20, i.e. the product observed in the 2:1 reaction of Li2bam + SmCl3. This suggested that a 2:1 complex had formed, leaving behind one half equivalent of unreacted

SmI2. In support of this hypothesis, the intentional 2:1 reaction of Li2bam and SmI2 gave the same product by 1H NMR spectroscopy. Additionally, results obtained by the Harder

135 group are in agreement with the proposition of 2:1 product formation, as crystals of

SmI2(THF)n were isolated from equimolar metathetical reactions of SmI2 and

230 [K2][HB(NDipp)2]. Further, a similar observation has been made by Lee and coworkers in the preparation of unsymmetrical benzamidinate complexes of samarium(II), where even the 1:1 reaction yielded a 2:1 (ligand:metal) product, which was characterized by X- ray crystallography.175

The reaction of the dipotassiated bam reagent 4.8 with SmI2 also gave a new product (by 1H NMR spectroscopy), which formed whether the reaction was carried out in 1:1 or 2:1 stoichiometry. Notably, this product did not give the same 1H NMR chemical shift values as those obtained from the reaction of SmI2 and the dilithio reagent; however, it was also obtained from the 2:1 reaction of K2bam and SmI3(THF)3.5. A possible explanation for these observations is salt inclusion, in the cases where Li2bam is used, versus the formation of a “salt-free” product when the dipotassiated reagent is employed (Figure 5.8).

Dipp Dipp n Dipp Dipp n N X N N N Ph B Sm B Ph n [Li] Ph B Sm B Ph n [K] N N N N

Dipp Dipp Dipp Dipp

Sm(II): n = 3 Sm(II): n = 2 Sm(III): n = 2 Sm(III): n = 1

Figure 5.8. Proposed structures of the products obtained from the reactions of SmI2 or

SmI3 and Li2bam (left) or K2bam (right).

136

In order to gain further evidence in support of this hypothesis, the reaction

1 SmI3(THF)3.5 + 2 Li2bam was carried out, with the expectation of a H NMR spectrum consistent with the salt-incorporation product [Li2(THF)8][{PhB(NDipp)2}2SmI] (5.21).

This indeed was observed. Orange, X-ray quality crystals of 5.21 were grown from a solution of the crude reaction mixture in THF layered with hexanes; the molecular structure of 5.21 is shown in Figure 5.9 and selected structural parameters are found in

Table 5.8. The formation and structural characterization of 5.21 also lends further credibility to the assignment of 5.20 (Section 5.5) as the 2:1 product

1 [Li2(THF)8][{PhB(NDipp)2}2SmCl], as the H NMR spectra of the two samarium complexes are nearly identical.

137

Table 5.8. Selected bond lengths (Å) and bond angles (o) for 5.21.

N1−B1 1.442(6) N1−B1−N2 114.0(4) N3−B2−N4 115.3(4)

N1−C7 1.401(5) N1−B1−C1 121.9(4) N3−B2−C31 122.8(4)

N1−Sm1 2.311(3) N2−B1−C1 123.9(4) N4−B2−C31 121.8(4)

N2−B1 1.427(6) B1−N1−Sm1 91.9(3) B2−N3−Sm1 89.9(2)

N2−C19 1.407(5) C7−N1−B1 126.3(4) C37−N3−B2 125.8(3)

N2−Sm1 2.358(2) C7−N1−Sm1 141.8(3) C37−N3−Sm1 142.3(3)

B1−C1 1.599(6) B1−N2−Sm1 90.4(2) B2−N4−Sm1 90.8(2)

N3−B2 1.420(6) C19−N2−B1 124.8(4) C49−N4−B2 126.1(3)

N3−C37 1.402(5) C19−N2−Sm1 143.1(3) C49−N4−Sm1 143.0(3)

N3−Sm1 2.367(3) N1−Sm1−N2 62.1(1) N1−Sm1−I1 112.4(1)

N4−B2 1.452(6) N3−Sm1−N4 62.3(1) N2−Sm1−I1 104.9(1)

N4−C49 1.394(5) N1−Sm1−N4 134.3(1) N3−Sm1−I1 105.0(1)

N4−Sm1 2.324(3) N2−Sm1−N4 105.4(1) N4−Sm1−I1 113.4(1)

B2−C31 1.614(6) N1−Sm1−N3 105.5(1)

Sm1−I1 3.098(1) N2−Sm1−N3 150.1(1)

The bond lengths and angles at the boron and nitrogen atoms in 5.21 are comparable to those in the analogous lanthanum complex 5.18. The average Sm−N bond length of ca. 2.34 Å in 5.21 is somewhat elongated compared to related Sm(III) complexes [Li(THF)4][(PhB{NDipp}2)SmCl2(THF)] and (PhB{NDipp}2)SmCl(THF)2

231 (ca. 2.29 Å and 2.26 Å, respectively) as well as [(HB{NDipp}2)Sm(DMA)]2 (ca. 2.30

Å),222 however, it still falls into the typical range for this type of bond. The N1−Sm1−N2

138 and N3−Sm1−N4 bond angles are slightly narrowed when compared to the samarium complexes 5.10 and 5.15; these angles are larger than those in lanthanum complex 5.18,

(by approximately 2°), as a result of the smaller size of the Sm(III) metal centre.

Although crystals of the analogous samarium(II) species were unable to be grown, the above results suggested that the formation of a 1:1 product by metathesis was unlikely and focus turned to alternative synthetic approaches.

5.6.2 Reduction Reactions

The second approach investigated in the pursuit of bamSmII was the one electron reduction of [PhB(NDipp)2]SmI(THF)2 (5.17). A variety of reducing agents were employed (summarized with their reduction potentials in Table 5.9) to reduce the Sm3+ centre to Sm2+ (reduction potential −1.55 V).

Table 5.9 Formal reduction potentials (V vs. FcH)232 of the reducing agents used in the attempted one electron reduction of 5.17

Reducing Agent Eºʹ

Cp2Co −1.33 (in CH2Cl2)

* Cp 2Co −1.94 (in CH2Cl2)

Na/naphthalene −3.10 (in THF)

Na/benzophenone −2.30 (in THF) K/benzophenone −2.30 (in THF) 0 K −2.38 (in NH3)

139

Preliminary attempts at reductions were performed using bis(cyclopentadienyl)cobalt(II) (cobaltocene) and the pentamethyl derivative Cp*2Co. In both instances, no reaction occurred even after extended times or at elevated temperatures. Consequently, the much stronger reducing reagent sodium naphthalenide was employed. Addition of the dark green solution of the reductant to 5.17 at ca. −80°C did not produce a colour change; however, upon warming to room temperature the solution turned a pale yellow-brown colour. The 1H NMR spectrum of the resulting caramel-coloured oil, obtained after removal of solvent, showed a very complex reaction mixture, indicating that the naphthalenide radical was perhaps too strong of a reducing agent.

Given that its reduction potential is intermediate between the cobaltocene and sodium naphthalenide, the anion radical diphenylketyl, which can be produced by mixing sodium or potassium and benzophenone, was selected as the reducing agent. The addition of the deep blue solution of sodium and benzophenone to the burnt orange solution of

5.17 at ca. −80°C did not produce an immediate colour change. Upon warming to room temperature, the solution became very dark in colour and a small amount of colourless precipitate, presumed to be NaI, formed. Although the 1H NMR spectrum of the dark, sticky solid, isolated after removal of volatiles, showed the presence of more than one species, this route appeared promising as the distinct dark colour of the product was indicative of the presence of samarium(II). Extraction into diethyl ether to remove the remaining NaI by-product yielded a material that was lighter in colour and gave a 1H

NMR spectrum that still demonstrated multiple products.

140

The reduction was repeated using potassium metal and benzophenone but these conditions gave similar results, including a loss of the dark colour upon extraction with diethyl ether. It is possible that bamSmII was produced in both of the reactions with the diphenylketyl anion radical but then reacted further with another species in solution, such as the benzophenone that remained after the reduction. When only potassium metal was used, no colour change indicative of the formation of a samarium(II) species was observed and attention was turned to the third synthetic approach: protonolysis.

5.6.3 Protonolysis (Amine or Alkane Elimination)

The final method examined for the preparation of a samarium(II) complex of the bam ligand was amine or alkane elimination from the reaction of PhB[N(H)Dipp]2 with

233 222 either Sm[N(SiMe3)2]2(THF)2 or Sm(DMAT)2(THF)2 (DMAT = 2-Me2N-α-Me3Si- benzyl), respectively. Amine elimination has been successful in the preparation of the

218 samarium(II) formamidinate [HC(NDipp)2]2Sm(THF)2, while alkane elimination has been used to form trivalent223 and divalent222 lanthanide complexes of the bam ligand

2- [HB(NDipp)2] .

The equimolar reaction of PhB[N(H)Dipp]2 and Sm[N(SiMe3)2]2(THF)2, proceeds to give a new bam-containing product by 1H NMR spectroscopy (Figure 5.10). In addition to the signals for the ligand, there are two trimethylsilyl resonances, one corresponding to the by-product, HN(SiMe3)2 (0.04 ppm), and one at −0.19 ppm that is very close in chemical shift to the samarium(II) starting material (−0.18 ppm). The bis(trimethylsilyl)amine was unable to be removed by washing with hexanes, as the dark-

141 coloured product was also soluble in this solvent; several washings with cold pentane (ca.

−60°C) did not remove a significant amount of the by-product.

Although the TMS resonance at −0.19 ppm could be assigned to unreacted starting material, leaving the major product to be assigned as the targeted bam complex

5.22, the possibility also existed that only one of the amide ligands was being protonated by the bam reagent producing complex 5.23 (Scheme 5.2). While the integration value suggests the former scenario, (i.e. the −N(SiMe3)2 substituent in 5.23 would have a relative integration value of 18 H), the addition of an excess of PhB[N(H)Dipp]2 did not significantly reduce the amount of this species present. This, coupled with the fact that the HN(SiMe3)2 by-product could not be completely removed, precluded characterization by CHN analysis.

Dipp

Ph B Sm(THF)n N

Dipp - 2 HN(SiMe3)2 PhB[N(H)Dipp]2 + Sm[N(SiMe3)2]2(THF)2 5.22

Dipp - HN(SiMe3)2 NH

Ph B Sm N(SiMe3)2 N

Dipp 5.23

Scheme 5.2

142

1 Figure 5.10. H NMR spectrum (in THF-d8) of the dark product obtained from the amine elimination reaction bamH2 + Sm[N(SiMe3)2]2(THF)2.

Several efforts were made to establish the identity of the product obtained from this protonolysis reaction. First, the mono-potassiated bam reagent

[K][PhB{N(H)Dipp}(NDipp)] 4.7 was reacted with SmI2, in anticipation of a product containing a neutral nitrogen atom coordinated to the samarium centre. Both the 1:1 and

1 2:1 reactions of 4.7 and SmI2 gave the same major product by H NMR, similar to the metathesis reactions described in Section 5.6.1, which seemingly reacted in a 2:1 stoichiometry even when equimolar amounts of reagents were used (Scheme 5.3).

143

Dipp Dipp 2 Ph NH N B THF 2 [K] + SmI [K2] Dipp N N Dipp 2 Ph B Sm B Ph H - 2 KI N HN 4.7 Dipp Dipp

Scheme 5.3

An immediate difference in the product of the 4.7 and SmI2 reaction compared to that of the protonolysis reaction was observed: the former was isolated as a distinctly purple solid while the latter product was consistently obtained as a dark brown solid.

Additionally, the 1H NMR spectrum of the metathesis reaction contained very broad signals, with numerous overlapping resonances in the methyl and methyne regions. This suggests that a product containing an −NH functionality coordinated to the metal centre

1 might give a more complex H NMR spectrum than that observed in Figure 5.11.

233 The mono-potassiated reagent 4.7 was also reacted with Sm(I)(N{SiMe3}2) to intentionally generate 5.23 (Scheme 5.4). A colourless precipitate of KI was observed and the resulting 1H NMR spectrum contained the same major product as obtained in the amine-elimination reaction. Although initially appearing to support 5.23 as the Sm(II) product from the reaction described in Scheme 5.2, this observation can also be explained by protonation of the amide substituent in addition to salt metathesis, which would then yield 5.22.

144

Dipp

Ph B Sm N(SiMe3)2 N

Ph THF Dipp - KI B [K] 5.23 + Sm[I][N(SiMe ) ] Dipp N N Dipp 3 2 H THF Dipp 4.7 - KI - HN(SiMe3)2 N

Ph B Sm(THF)n N

Dipp

5.22

Scheme 5.4

The final reaction carried out in attempts to elucidate the nature of the Sm(II) species obtained from the amine-elimination reaction was the one electron oxidation of this product with one half an equivalent of iodine. It was anticipated that if the targeted bamSmII complex 5.22 was indeed the product of protonolysis, then bamSmI (5.17) would be produced from oxidation (Scheme 5.5). Lee and co-workers previously used an oxidation reaction for the generation of an ytterbium(III) iodide species from the corresponding ytterbium(II) bis-benzamidinate complex, although when the reaction was performed on the analogous samarium complex, the authors isolated only an intractable oil.175

145

Dipp Dipp N N

Ph B Sm(THF)n + 0.5 I2 Ph B Sm I N N

Dipp Dipp

5.22 5.17

Scheme 5.5

When carried out in Et2O, the one electron oxidation of the solid obtained from protonolysis gave a colour change from dark brown to orange-yellow and a precipitate formed. Similar colour changes were observed in THF, but the resulting product was an oily substance. In both cases, however, several species were present (by 1H NMR spectroscopy), with the predominant one being PhB[N(H)Dipp]2.

While this research was being performed, the Harder group was concurrently investigating the reaction of the diprotio bam HB[N(H)Dipp]2 with

Sm[N(SiMe3)2]2(THF)2, but the desired samarium(II) bam species was unable to be isolated from the reaction mixtures, despite observing colour changes supporting its

222 formation. This was proposed to potentially be a consequence of the HB[N(H)Dipp]2 not being acidic enough to drive the reaction fully to the product side.222 The phenyl substituent on boron should render PhB[N(H)Dipp]2 more acidic than HB[N(H)Dipp]2, which would account for the fact that the protonolysis reaction presented here proceeded to give a single major product, although the identity of this product was unable to conclusively be determined.

In light of the results obtained from the amine-elimination reaction, the alkane elimination reaction between PhB[N(H)Dipp]2 and Sm(DMAT)2(THF)2 was carried out

146 in THF. A dark brown solid was isolated, however, the 1H NMR spectrum of this product was not the same as that of the reaction between PhB[N(H)Dipp]2 and

Sm[N(SiMe3)2]2(THF)2. When performed under the conditions used by the Harder group

222 to isolate crystals of [HB(NDipp)2]Sm(THF)4, illustrated in Scheme 5.6, the reaction gave an oily product; although the major species was the same as that observed in the previous alkane elimination attempts in THF, it was less clean overall and a single product was unable to be isolated from the mixture.

H Benzene, ! HB[N(H)Dipp]2 + (DMAT)2Sm(THF)2 B - 2 DMAT(H) Dipp N N Dipp SiMe3 Sm H (THF)4 DMAT =

NMe2

Scheme 5.6

5.7 Conclusions

In this chapter, the first examples of incorporating the bam ligand into f-block chemistry were presented. Lanthanide “ate” complexes {[Li(THF)4][bamLnCl2(THF)]}2,

(Ln = Y, Pr, Nd, Sm, Ho, Er, Yb) were prepared from the reaction of Li2bam and the corresponding lanthanide trichloride. All seven complexes are isostructural, forming dimers in the solid state through bridging chlorides. The proton NMR spectra of paramagnetic complexes Ln = Pr, Nd, Sm were able to be obtained and showed a lanthanide-induced shift in both the aryl (deshielded) and alkyl (shielded) regions.

147

Although halide substitution reactions with metal alkyls or amides did not proceed cleanly, treatment of {[Li(THF)4][bamLnCl2(THF)]}2 (Ln = Y, Sm) with an excess of TMSOTf produced complexes of the form bamLnCl(THF)2, which are also dimeric in the solid state. Investigations into the reactions of these neutral species with lithium or potassium bis(trimethylsilyl)amide and with lithium diisopropylamide did not proceed cleanly, which prompted the preparation of analogous iodide complexes.

The equimolar reaction of the newly prepared K2bam reagent with YI3(THF)3.5 or

SmI3(THF)3.5 resulted in precipitation of KI and formation of new products that have been assigned as bamLnI(THF)2 (Ln = Y, Sm). Both of these new species primarily gave complex reaction mixtures when reacted with selected potassium alkyls, amides and alkoxides.

Lanthanide complexes containing two bam ligands of the form

[Li2(THF)8][{PhB(NDipp)2}2LnX] (Ln = La, X = I; Ln = Sm, X = Cl) were also prepared and two were structurally characterized. The one electron oxidation of these species was attempted but stable radical species were not observed.

Finally, numerous efforts were made to isolate a samarium(II) boraamidinate complex. Metathetical reactions appeared to proceed in a 2:1 stoichiometry, despite using equimolar amounts of starting materials. Reduction of bamSmI(THF)2 using a variety of reducing agents also did not yield the desired bamSmII complex. These observations paralled those obtained by the Harder group, who were only able to isolate a Sm(II) bam complex by alkane elimination. Protonolysis reactions with PhB[N(H)Dipp]2 produced new bam-containing species (by 1H NMR spectroscopy), however the molecular structure was not confirmed.

148

The incorporation of the dianionic bam ligand into lanthanide chemistry, particularly in the formation of the neutral lanthanide(III) complexes, may open new avenues of research in the areas of catalysis and materials chemistry as the utility of these new complexes could be compared with that of the well-known amidinate and guanidinate species.

5.8 Experimental Section

Reagents, general procedures and instrumentation are described in Appendix 1.

General synthetic procedure for {[Li(THF)4][bamLnCl2(THF)]}2 (Ln = Y, Pr, Nd,

Sm, Ho, Er, Yb). A solution of [Li2][PhB(NDipp)2] in THF (10 mL) was added to a

Dipp Dipp stirred slurry of LnCl3 in THF (10 mL) at ca. Cl THF N N Cl −80°C (Ln = Y, Sm, Ho, Er, Yb) or at room Ph B Ln Ln B Ph Cl N N THF Cl temperature (Ln = Pr, Nd). The reaction was kept Dipp Dipp + cold for approximately 30 min then warmed to Li (THF)4 room temperature and stirred for an additional 1−2 h (Ln = Y, Sm, Ho, Er, Yb) or heated to 65°C for 1 h (Ln = Pr, Nd). Volatiles were removed in vacuo and the reaction mixture was treated with hexane (10 mL) and solvent was removed under vacuum. Complex 5.6 was additionally washed with diethyl ether (5 mL). NMR data for all complexes were collected using the initially formed product, which contained LiCl. Yields (%) are estimated based on the mass of this product; they were calculated by including the five molecules of THF in the complexes and assuming the inclusion of an equimolar amount of LiCl. X-ray quality crystals of all complexes were obtained from concentrated THF

149 solutions layered with hexanes at −18°C. Satisfactory CHN analyses were unable to be obtained for even the recrystallized samples; however, the 1H NMR spectra of the initially formed bulk samples of 5.7 -5.10, demonstrate their purity. The reactivity of the complexes towards acetonitrile precluded the determination of electrospray ionization mass spectra.

5.7: YCl3 (0.100 g, 0.512 mmol), [Li2][PhB(NDipp)2] (0.231 g, 0.511 mmol); Yield:

1 0.384 g pale yellow powder (75%). H NMR (THF-d8, 25°C): δ 6.87 (2 H, m), 6.73 (4

3 H, m), 6.61 (1 H, m), 6.55 (2 H, m) 6.47 (2 H, m), 3.98 (4 H, sept, JH-H = 6.78 Hz), 3.61

3 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 1.12 (12 H, d, JH-H = 6.78 Hz), 0.71

3 11 7 (12 H, d, JH-H = 6.78 Hz). B NMR (THF-d8, 25°C): δ 32.6 (br, s). Li NMR (THF-d8,

13 25°C): δ 0.11. C NMR (THF-d8, 25°C): δ 155.3, 151.0, 141.3, 136.3, 126.3, 126.0,

122.6, 117.8, 68.4 (−OCH2CH2, THF), 28.2, 26.5 (−OCH2CH2, THF), 25.6, 24.6, 15.8.

5.8: PrCl3 (0.111 g, 0.449 mmol), [Li2][PhB(NDipp)2] (0.204 g, 0.451 mmol); Yield:

1 0.328 g, pale yellow powder (69 %). H NMR (THF-d8, 25°C): δ 12.37 (2 H, br s), 10.95

(1 H, br s), 10.59 (4 H, br s), 8.39 (2 H, s), 7.16-7.11 (2H, br m), 3.61 (m, −OCH2CH2,

THF), 1.75 (m, −OCH2CH2, THF), 0.94 (12 H, br s), -2.80 (4 H, br s), -7.99 (12 H, br s).

7 Li NMR (THF-d8, 25°C): δ -0.43.

5.9: NdCl3 (0.109 g, 0.435 mmol), [Li2][PhB(NDipp)2] (0.198 g, 0.438 mmol); Yield:

1 0.260 g, green powder (56 %). H NMR (THF-d8, 25°C): δ 11.46 (2 H, br s), 10.25 (1 H,

150 br s), 8.93 (4 H, br s), 7.11 (2 H, br s), 6.34 (2 H, br s) 3.61 (m, −OCH2CH2, THF), 1.75

7 (m, −OCH2CH2, THF), 1.12 (12 H, br s), -1.32 (4 H, br s), -6.35 (12 H, br s). Li NMR

(THF-d8, 25°C): δ -0.04.

5.10: SmCl3 (0.285 g, 1.11 mmol), [Li2][PhB(NDipp)2] (0.503 g, 1.11 mmol); Yield: 1.00

1 g, orange powder (84 %). H NMR (THF-d8, 25°C): δ 8.99 (2 H, d), 7.36-7.27 (3 H, m),

6.63 (2 H, d), 6.54 (4 H, t), 3.63 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 3.30

3 3 7 (4 H, br, septet), 1.05 (12 H, d, JH-H = 6.2 Hz ), -0.06 (12 H, d, JH-H = 6.2 Hz). Li NMR

(THF-d8, 25°C): δ 0.23.

5.11: HoCl3 (0.091 g, 0.332 mmol), [Li2][PhB(NDipp)2] (0.150 g, 0.332 mmol); Yield:

0.223 g, cream coloured powder (61 %). NMR spectra were unable to be obtained.

5.12: ErCl3 (0.091 g, 0.333 mmol), [Li2][PhB(NDipp)2] (0.151 g, 0.334 mmol); Yield:

0.220 g, light pink powder (61 %). NMR spectra were unable to be obtained.

5.13: YbCl3 (0.105 g, 0.376 mmol), [Li2][PhB(NDipp)2] (0.171 g, 0.378 mmol); Yield:

0.280 g, dark red powder (78 %). NMR spectra were unable to be obtained.

Preparation of {bamYCl(THF)2}2 (5.14). A solution of trimethylsilyl triflate (0.092 g,

Dipp Dipp 0.414 mmol) in THF (2 mL) was added to a THF THF N Cl N solution of 5.7 (0.198 g, 0.103 mmol) in THF (7 Ph B Y Y B Ph Cl N N THF THF Dipp Dipp 151 mL). The resulting solution was stirred for 45 min, volatiles were removed in vacuo and the resulting sticky solid was washed with ca. 7 mL of hexane. Removal of volatiles in vacuo gave a colourless solid, which was a mixture of LiOTf and 5.14 (0.166 g, 92%).

Colourless, X-ray quality crystals of 5.14 were grown from a THF solution layered with hexanes. Suitable CHN analyses were unable to be obtained for even the recrystallized samples; however, the 1H NMR spectrum of the initially formed bulk sample

1 demonstrates purity. H NMR (THF-d8, 25°C): δ 6.82-6.52 (11 H, m), 3.84 (4 H, br,

3 sept), 3.62 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 1.12 (12 H, d, JH-H =

3 11 6.71 Hz), 0.72 (12 H, d, JH-H = 6.71 Hz). B NMR (THF-d8, 25°C): δ 31.8 (br, s).

Preparation of {bamSmCl(THF)2}2 (5.15). A solution of trimethylsilyl triflate (0.096 g,

Dipp Dipp 0.432 mmol) in THF (2 mL) was added to a THF THF N N Cl solution of 5.10 (0.222 g, 0.108 mmol) in THF (7 Ph B Sm Sm B Ph Cl N N THF THF mL). The resulting orange solution was stirred for Dipp Dipp 15 min, volatiles were removed in vacuo and the resulting sticky solid was washed with ca. 7 mL of hexane. Removal of volatiles in vacuo gave 5.15 as a mixture of an orange solid and colourless [Li][OTf] (0.201 g). Orange, X- ray quality crystals of 5.15 were grown from a THF solution layered with hexanes (24% yield). Suitable CHN analyses were unable to be obtained for even the recrystallized samples; however, the 1H NMR spectrum of the initially formed bulk sample

1 demonstrates purity. H NMR (THF-d8, 25°C): δ 8.97 (2 H, m), 7.44-6.74 (9 H, m), 3.04

(4 H, br, sept), 3.62 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 0.97 (12 H, br, d), 0.03 (12 H, br, d).

152

Preparation of {bamYI(THF)n}m (5.16). A solution of [K2][PhB(NDipp)2] (0.416 g,

Dipp 0.449 mmol) in THF (10 mL) was added to a stirred slurry of N YI3(THF)3.5 (0.323 g, 0.447 mmol) in THF (10 mL) at room Ph B Y(I)(THF)n N temperature. The reaction was stirred for 15 min and the Dipp m lemon yellow solution was filtered to remove the colourless precipitate (KI). Volatiles were removed in vacuo and the reaction mixture was treated with hexanes (10 mL) and solvent was removed in vacuo, yielding 5.16 as a pale yellow

1 powder (0.260 g, 0.728 mmol, 73%). H NMR (THF-d8, 25°C): δ 6.84-6.42 (11 H, m),

3.72 (4 H, br, overlapping sept), 3.62 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF),

3 3 11 1.12 (12 H, d, JH-H = 6.82 Hz), 0.66 (12 H, d, JH-H = 6.72 Hz). B NMR (THF-d8,

25°C): δ 31.8 (br, s).

Preparation of {bamSmI(THF)n}m (5.17). A solution of [K2][PhB(NDipp)2] (0.330 g,

0.639 mmol) in THF (10 mL) was added to a stirred slurry of Dipp

N SmI3(THF)3.5 (0.500 g, 0.638 mmol) in THF (10 mL) at room Ph B Sm(I)(THF)n N temperature. The reaction was stirred for 15 min and the burnt

Dipp m orange solution was filtered to remove the colourless precipitate (KI). Volatiles were removed in vacuo and the reaction mixture was treated with hexane (10 mL). Removal of solvent in vacuo gave 5.17 as a burnt orange powder

1 (0.461 g, 0.536 mmol, 84%). H NMR (THF-d8, 25°C): δ 8.99 (2 H, m), 7.33-7.11 (9 H, m), 2.50 (4 H, br), 3.62 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 0.62 (12 H, br), 0.06 (12 H, br).

153

Attempted Synthesis of bamY(Bz). A solution of [K][CH2C6H5] (0.003 g, 0.023 mmol) in THF-d8 (0.5 mL) was added to an NMR tube containing 5.16 in THF-d8 (0.5 mL). The deep red colour was immediately consumed and a colourless precipitate formed. 1H NMR

3 (THF-d8, 25°C): δ 6.80-6.12 (m, aryl), 3.82 (4 H, sept, JH-H = 6.81 Hz), 3.62 (m,

3 −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 1.56 (2 H, d, JY−H = 2.46 Hz), 1.14 (12

3 3 H, d, JH-H = 6.81 Hz), 0.77 (12 H, d, JH-H = 6.81 Hz). Performing this reaction on a preparative scale in THF gave multiple products by 1H NMR, none of which had chemical shifts corresponding to the product described above.

Preparation of [Li2(THF)8][{PhB(NDipp)2}2LaI] (5.18). A solution of

2 Dipp Dipp [Li2][PhB(NDipp)2] (0.225 g, 0.497 mmol) in Et2O

N I N (20 mL) was added to a stirred slurry of LaI3(THF)3.5 Ph B La B Ph 2 [Li]

N N (0.200 g, 0.248 mmol) in Et2O (5 mL) at room Dipp Dipp temperature. The reaction was stirred for approximately 20 min and filtered to remove LiI. Volatiles were removed in vacuo and the reaction mixture was treated with hexane (5 mL). Solvent was removed to give 5.18 as a pale yellow solid (0.304 g, 0.175 mmol, 71%). Anal. Calcd. for

1 C96H150N4B2LaILi2O9: C, 63.82; H, 8.37; N, 3.10. Found: C, 60.02; H, 7.76; N, 3.02. H

3 NMR (THF-d8, 25°C): δ 6.78-6.34 (22 H, m, aryl), 3.56 (4 H, sept, JH-H = 6.86 Hz) 3.62

3 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 0.87 (24 H, d, JH-H = 6.86 Hz), 0.64

3 7 11 (24 H, d, JH-H = 6.86 Hz). Li NMR (THF-d8, 25°C): δ 0.47. B NMR (THF-d8, 25°C):

δ 32.9 (br, s).

154

Preparation of [Li (THF) ][{PhB(NDipp) } YCl] 2 2 8 2 2 Dipp Dipp

N Cl N (5.19). A solution of [Li2][PhB(NDipp)2] (0.137 g, Ph B Y B Ph 2 [Li] 0.303 mmol) in THF (5 mL) was added to a stirred N N

Dipp Dipp slurry of YCl3 (0.029 g, 0.149 mmol) in THF (5 mL) at ca. −80°C. The reaction was kept cold for approximately 20 min, warmed to room temperature and stirred for an additional hour. Volatiles were removed in vacuo and the reaction mixture was treated with hexane (5 mL) and extracted with Et2O. Removal of solvent gave 5.19 as a peach coloured solid (0.104 g, 0.065 mmol, 44%). 1H NMR (THF- d8, 25°C): δ 7.30 (4 H, m, aryl) 7.00-6.92 (10 H, m, aryl), 6.80-6.71 (6 H, m, aryl), 6.58

3 3 (2 H, m, aryl), 3.91 (4 H, sept, JH-H = 6.86 Hz), 3.40 (4 H, sept, JH-H = 6.72 Hz) 3.62 (m,

3 −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 1.28 (12 H, d, JH-H = 6.72 Hz), 1.13 (12

3 3 7 H, d, JH-H = 6.72 Hz), 0.81 (24 H, d, JH-H = 6.86 Hz). Li NMR (THF-d8, 25°C): δ 0.35.

11 B NMR (THF-d8, 25°C): δ 31.4 (br, s).

Preparation of [Li2(THF)8][{PhB(NDipp)2}2SmCl] (5.20). A solution of

[Li2][PhB(NDipp)2] (0.215 g, 0.475 mmol) in THF 2 Dipp Dipp

N Cl N (10 mL) was added to a stirred slurry of SmCl3 Ph B Sm B Ph 2 [Li] (0.061 g, 0.238 mmol) in THF (10 mL) at ca. −80°C. N N Dipp Dipp The reaction was kept cold for approximately 1 h, warmed to room temperature and stirred for an additional hour. Volatiles were removed in vacuo and the reaction mixture was treated with hexane (5 mL) and extracted in Et2O.

Solvent was removed yielding 5.20 as golden yellow solid (0.323 g, 0.195 mmol, 82%).

155

1 H NMR (THF-d8, 25°C): δ 9.81 (2 H, m, aryl) 7.66-6.23 (20 H, m, aryl), 3.91 (4 H, sept,

3 3 JH-H = 6.86 Hz), 3.39 (4 H, sept, JH-H = 6.72 Hz) 3.62 (m, −OCH2CH2, THF), 1.78 (m,

3 3 −OCH2CH2, THF), 1.28 (12 H, d, JH-H = 6.72 Hz), 1.13 (12 H, d, JH-H = 6.72 Hz), 0.81

3 7 (24 H, d, JH-H = 6.86 Hz). Li NMR (THF-d8, 25°C): δ −0.15.

II Attempted Synthesis of bamSm by Metathesis with Li2bam. A solution of

[Li2][PhB(NDipp)2] (0.126 g, 0.279 mmol) in THF (8 mL) was added to a solution of

SmI2 (0.112 g, 0.277 mmol) in THF (8 mL) at ca. −80°C. The reaction was kept cold for approximately 30 min, warmed to room temperature and stirred for an additional 30 min.

Volatiles were removed in vacuo and the reaction mixture was treated with hexanes (ca.

5 mL). Removal of solvent yielded a very dark coloured solid, proposed to contain the

2:1 product [Li3(THF)12][{PhB(NDipp)2}2SmI] in addition to half-an-equivalent of unreacted SmI2. This was supported by the observation that the same major product was obtained (by 1H NMR spectroscopy) when the reaction was carried out in a 2:1 molar ratio. Additionally, the chemical shifts of the product are comparable to those found for

1 5.20 and 5.21. H NMR (THF-d8, 25°C): δ 9.98 (2 H, m, aryl) 7.76-5.84 (20 H, m, aryl),

3 3 3.91 (4 H, sept, JH-H = 6.76 Hz), 3.39 (4 H, sept, JH-H = 6.83 Hz) 3.62 (m, −OCH2CH2,

3 3 THF), 1.78 (m, −OCH2CH2, THF), 1.28 (12 H, d, JH-H = 6.76 Hz), 1.13 (12 H, d, JH-H =

3 6.76 Hz), 0.81 (24 H, d, JH-H = 6.83 Hz).

II Attempted Synthesis of bamSm by Metathesis with K2bam. A solution of

[K2][PhB(NDipp)2] (0.102 g, 0.197 mmol) in THF (7 mL) was added to a solution of

156

SmI2 (0.080 g, 0.198 mmol) in THF (7 mL) at ca. −80°C. The reaction was warmed to room temperature and stirred for 30 min. Volatiles were removed in vacuo and the reaction mixture was treated with hexanes (5 mL). Removal of solvent yielded a very dark coloured solid, proposed to contain the 2:1 product

[K2(THF)x][{PhB(NDipp)2}2Sm] in addition to half-an-equivalent of unreacted SmI2.

This was supported by the observation that the same major product was obtained (by 1H

NMR spectroscopy) when the reaction was carried out in a 2:1 molar ratio. 1H NMR

(THF-d8, 25°C): major product δ 9.98 (4 H, m, aryl) 7.80-7.64 (8 H, m, aryl), 6.67-6.30

3.91 (10 H, m, aryl), septet not observed, likely under the THF resonance at 3.62, 3.62

(m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 0.89 (24 H, no observable coupling),

-1.63 (24 H, no observable coupling).

Preparation of [Li2(THF)8][{PhB(NDipp)2}2SmI] (5.21). A solution of

[Li2][PhB(NDipp)2] (0.014 g, 0.031 mmol) in THF-d8 (0.5mL) was added to a slurry of

1 SmI3(THF)3.5 (0.012 g, 0.015 mmol) in THF-d8 (0.5mL) at room temperature. H NMR

(THF-d8, 25°C): major product δ 9.81 (2 H, m, aryl) 7.66-6.23 (20 H, m, aryl), 3.89 (4 H, br, sept), 3.38 (4 H, br, sept) 3.62 (m, −OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF),

3 3 1.28 (12 H, br), 1.12 (12 H, d, JH-H = 6.72 Hz), 0.81 (24 H, d, JH-H = 6.86 Hz). When performed on a preparative scale, carrying out the reaction in Et2O gave a cleaner crude product than when THF was used (68 % yield). Suitable anlayses could not be obtained, likely a because of THF loss.

157

Attempted Synthesis of [K][{PhB(NDipp)2}2Sm]. A solution of [K2][PhB(NDipp)2]

(0.013 g, 0.025 mmol) in THF-d8 (0.5 mL) was added to a solution of SmI3(THF)3.5

(0.010 g, 0.013 mmol) in THF-d8 (0.5 mL) at room temperature yielding an orange solution, for which 1H NMR data was obtained, and a colourless precipitate (KI). 1H

NMR (THF-d8, 25°C): δ 9.98 (4 H, m, aryl) 7.80-7.64 (8 H, m, aryl), 6.67-6.30 3.91 (10

H, m, aryl), septet not observed, likely under the THF resonance at 3.62, 3.62 (m,

−OCH2CH2, THF), 1.78 (m, −OCH2CH2, THF), 0.89 (24 H, no observable coupling), -

1.63 (24 H, no observable coupling).

Attempted Synthesis of bamSmII by Reduction of bamSmI. Sodium (0.030 g, 0.130 mmol) and benzophenone (0.238 g, 1.30 mmol) were added to THF (10.0 mL) in a

Schlenk tube and stirred for ca. 1 h. A 1.0 mL aliquot of the resultant dark blue solution was added to a burnt orange solution of bamSmI(THF)2 (0.112 g, 0.130 mmol) in THF (7 mL) cooled to ca. −30°C. The reaction was warmed to room temperature at which time the solution was still very dark in colour and a colourless precipitate formed. Volatiles were removed in vacuo yielding a very dark sticky solid. The 1H NMR spectrum of this product in THF-d8 demonstrated a complex reaction mixture. When performed using potassium (0.034g, 0.869 mmol), benzophenone (0.159 g, 0.868 mmol) and bamSmI(THF)2 (0.075 g, 0.087 mmol) under the same reaction conditions, similar observations were made and the 1H NMR spectrum again showed numerous species in solution.

158

II Attempted Synthesis of bamSm by Protonation of Sm[N(SiMe3)2]2(THF)2. A solution of PhB[N(H)Dipp]2 (0.108 g, 0.245 mmol) in THF (3 mL) was added to a solution of Sm[N(SiMe3)2]2(THF)2 (0.150 g, 0.244 mmol) in THF (7 mL) at room temperature. The reaction was stirred for 10 min and volatiles were removed in vacuo, giving a dark sticky product. Treatment with pentane (2 x 8 mL) and removal of volatiles yielded a dark brown solid (0.150 g). Numerous cold pentane washes were unable to

1 remove all of the HN(SiMe3)2 by-product. H NMR (THF-d8, 25°C): δ 7.28-6.46 (11 H,

3 3 aryl), 3.98 (2 H, sept, JH-H = 6.6 Hz), 3.50 (2 H, overlapping sept), 1.29 (6 H, d, JH-H =

3 3 6.6 Hz), 1.11 (6 H, d, JH-H = 6.6 Hz), 0.84 (12 H, d, JH-H = 6.3 Hz), 0.04 (s,

HN(SiMe3)2), −0.19 (s).

159

Chapter Six: Conclusions and Future Work

6.1 Concluding Remarks

Prior to the work described in this thesis, only dianionic boraamidinates with alkyl or aryl substituents on the boron were known and the chemistry of this ligand was limited to complexes of the s-, p- and d-block elements.iii As described in Chapter One, the objectives of this work were to expand the scope of boraamidinate chemistry in the following ways: (i) preparing new nitrogen- and boron-containing ligands and examining their metathetical reactions with main group halides; (ii) modifying known bam ligands by either installing a bridging group on boron or by mono-metallating the backbone; (iii) using a known bam ligand to form complexes of the f-block elements. As detailed in

Chapters Two through Five, these goals have been achieved.

Chapter Two presented the synthesis and structural characterization of the first

i example of a stable dilithio boraguanidinate, {[Li2][ Pr2NB(NDipp)2]}2, Li2bog (2.4), prepared by various synthetic approaches. In the process, two new halo(diamino)boranes

(2.5 and 2.6) were also isolated, in addition to a tris(amino)borane (2.7).

iii As previously mentioned, the Harder group recently reported a bam ligand with hydrogen as the substituent on boron as well as lanthanide complexes of this ligand (see references 222 and 223).

160

i i N Pr2 N Pr2

Dipp B Dipp Dipp B Dipp N N N N

H H Li Li

bogH2, 2.3 Li2bog, 2.4

H Dipp X N

Dipp B Dipp Dipp B Dipp N N N N

H H H H

2.5, X = Cl 2.7 2.6, X = Br

The bog ligand was found to be more strongly reducing than both boraamidinates and guanidinates. Although evidence of complex formation with main group and lanthanide elements was observed in selected cases (by 1H and 11B NMR spectroscopy) attempts to separate these complexes from the salt by-product by either extraction or crystallization resulted in isolation of the diprotonated ligand, bogH2. Alkane-elimination reactions, performed to circumvent reactions with the reducing dianionic ligand, did not proceed to give new species. Consequently, replacement of the diisopropylamino substituent with amino groups containing at least one electron-withdrawing substituent was carried out. The new bog ligand [Li2][Ph(Me)NB(NDipp)2] was prepared, however, metathetical reactions yielded similar results to those of Li2bog.

While examining various synthetic routes to Li2bog, the novel seven-membered heterocycle 3.2a was isolated from the reaction of BCl3 and two equivalents of

[Li][N(H)Dipp] in THF. Chapter Three described the intentional preparation and structural characterization of this 1,3,2-oxazaborepane, formed by ring-opening of THF.

By varying the stoichiometry of the reaction, it was determined that the dichloroborane

161

1 11 Cl2B[N(H)Dipp] likely initiated the ring-opening process and NMR evidence ( H, B) supported the formation of a second heterocyclic product with a chloride substituent on boron (3.2b).

H2 H2 C C H2C CH2 H2C CH2

O CH2 O CH2 B N B N

Dipp(H)N Dipp Cl Dipp 3.2a 3.2b

In Chapter Four, the boraamidinate ligand was modified in two different ways: (i) by introducing a bridging group as the substituent on boron to form bis-bams and (ii) by mono-metallating neutral bamH2 to give the first examples of monoanionic bams.

Additionally, a new dianionic bam reagent, [K2][PhB(NDipp)2] was prepared.

New ferrocenyl (diamino)boranes 4.1-4.3 were also prepared and structurally characterized; however, the isolation of ferrocenyl boraamidinates was precluded, presumably by competition between metallation of the Cp ring and deprotonation of the

−NH functionality. The first example of a tetralithio bis-bam ligand (4.5) was prepared and characterized by single crystal X-ray diffraction. This new type of bam is extremely air-sensitive, turning pink upon exposure to trace amounts of oxygen. The one-electron oxidation using half an equivalent of I2 gave a complex EPR spectrum with more than one radical species present and the identity of these species could not be conclusively determined.

162

N(H)tBu

BH[N(H)R]2 B N(H)tBu Fe tBu(H)N Fe

tBu(H)N 4.1, R = tBu 4.3 4.2, R = Dipp

4- tBuN NtBu

+ B B 4 [Li]

tBuN NtBu 4.5

Two monoanionic bams, [M][PhB{N(H)Dipp}(NDipp)] (4.6: M = Li, 4.7: M =

n K) were prepared by deprotonation of bamH2 with a single equivalent of either BuLi or

[K][CH2Ph]. In the solid state, the monolithiated ligand demonstrates the less common bonding mode of a monodentate structure. Benzyl potassium was also used to generate a new bam reagent, [K2][PhB(NDipp)2] (4.8), which forms an extended structure in the solid state via η1-amide-η6-arene coordination. This potassiated ligand is useful for avoiding salt by-product incorporation into complexes.

THF Ph Ph - Ph 2- Li B Dipp Dipp B Dipp [K+] Dipp B Dipp 2 [K+] N N N N N N THF H H Dipp

4.6 4.7 4.8

Finally, Chapter Five focused on preparing boraamidinate complexes of the lanthanide elements. Seven “ate” complexes 5.7-5.13 were synthesized and structurally

163 characterized, representing the first time that the bam ligand has been used in f-block chemistry. Neutral yttrium and samarium complexes (5.14 and 5.15, respectively) were obtained from the reaction of the corresponding “ate” species with an excess of trimethylsilyl triflate.

Li+(THF) Dipp 4 Dipp Dipp Dipp Cl THF THF THF N Cl N N Cl N Ph B Ln Ln B Ph Ph B Ln Ln B Ph Cl Cl N N N N THF Cl THF THF Dipp Dipp Dipp Dipp

+ Li (THF)4

Ln = Y (5.7), Pr (5.8), Nd (5.9), Sm (5.10), Ln = Y (5.14), Sm (5.15) Ho (5.11), Er (5.12), Yb (5.13)

Halide replacement reactions did not proceed cleanly, which prompted the synthesis of iodide derivatives 5.16 and 5.17 using the newly formed K2bam reagent.

However, the solid-state structures of both the neutral iodide complexes as well as the products formed from metathetical reactions were unable to be determined conclusively by X-ray analysis.

Dipp

Ph B Ln(I)(THF)n N

Dipp m Ln = Y (5.16), Sm (5.17)

Spirocyclic complexes of lanthanum, yttrium and samarium (5.18-5.21) were prepared from the 2:1 reaction of Li2bam and the corresponding lanthanide trihalide.

Unlike the analogous group 13 species, these lanthanide complexes did not form stable

164 radicals upon one-electron oxidation with iodine, but still represent a new class of bam complexes of the f-block elements. However, in order to get a direct comparison with the group 13 systems, removal of the halide ligand would be necessary.

2 Dipp Dipp N X N Ph B Ln B Ph 2 [Li] N N

Dipp Dipp

5.18, Ln = La, X = I 5.19, Ln = Y, X = Cl 5.20, Ln = Sm, X = Cl 5.21, Ln = Sm, X = I

Numerous attempts were made to prepare a samarium(II) complex of a bam ligand. Metathetical reactions between SmI2 and either Li2bam or K2bam appeared to proceed in a 2:1 stoichiometry, even when reactions were performed in equimolar amounts. The two different bam reagents produced unique products, proposed to be a result of salt incorporation in the case where Li2bam was used but not when K2bam was employed (see A and B, respectively).

Dipp Dipp 3 Dipp Dipp 2

N I N N N Ph B Sm B Ph 3 [Li] Ph B Sm B Ph 2 [K] N N N N

Dipp Dipp Dipp Dipp A B

165

Reduction of [PhB(NDipp)2]SmI(THF)2 did not yield the targeted Sm(II) bam, despite characteristic colour changes and formation of a colourless precipitate (presumed to be the salt by-product) when the reducing agent was benzophenone and either sodium or potassium. The protonolysis reaction between bamH2 and Sm[N(SiMe3)2]2(THF)2 gave a new, highly soluble species by 1H NMR and numerous efforts were made in attempts to determine if this product was the targeted bamSmII complex (C) or if only one −NH group had been deprotonated (D).

Dipp Dipp

N NH Ph B Sm N(SiMe ) Ph B Sm(THF)n 3 2 N N

Dipp Dipp C D

6.2 Suggested Future Work

The work presented in this thesis opens several new avenues of research. For example, the tris(amino)borane B[N(H)Dipp]3 (2.7) prepared herein contains three acidic protons. Deprotonation of these sites would yield the nitrogen analogue of the orthoborate trianion. The first example of a complex containing such a ligand was reported in 2002 by the Russell group who structurally characterized

n {[Li4B(NR)3(THF)][B Bu4](toluene)}2 from the trilithiation of B[N(H)R]3 (R = 2-

n 234 MeOC6H4) with BuLi; however, the free trianion was not prepared. That same year,

Nöth and co-workers attempted the trilithiation of B[N(H)R]3 and isolated the trianionic

115 species as its [Li][N(H)R] adduct, (R = 2-pyridyl). In the case of B[N(H)Dipp]3, the

166 bulky Dipp substituents on nitrogen may facilitate triple deprotonation to give a discrete

3- ligand. It would be appealing to explore the redox chemistry of the trianion, as the

t 2- 235 t 2- 236 related dianions {[S(N Bu)3] } and {[Se(N Bu)3] } have shown interesting

235,237,238 3- oxidation chemistry; the Dipp substituents in [B(NDipp)3] may support the formation of a dianion radical through both steric and electron stabilization.

The bis(amino)haloboranes 2.5 and 2.6 prepared in Chapter 2 possess a halide site that can be further functionalized. For example, reaction with secondary phosphides,

[Li][PR2] (R = alkyl, aryl), would yield R2PB[N(H)Dipp]2. Subsequent dilithiation would result in the phosphorus analogue of boraguanidinate 2.4 (Scheme 6.1); since any multiple bonding between phosphorus and boron is minimized when nitrogen is also bonded to the boron atom,239 this new type of ligand is anticipated to have electronic properties intermediate between those of bams and bogs. Observing the electronic effects of the less electron-donating phosphine substituent (compared to an amine) would be of interest in the formation of main group element complexes.

2- X PR2 PR2 2 nBuLi + B + [Li][PR2] B B 2 [Li] Dipp(H)N N(H)Dipp -LiX Dipp(H)N N(H)Dipp - 2 DippN NDipp

2.5: X = Cl R = alkyl, aryl E F 2.6: X = Br

Scheme 6.1

These neutral (E) and dianionic (F) species would also be the boron analogues of phospha(III)-guanidines and guanidinates, respectively, which have been shown to have novel coordination chemistry and can form multimetallic complexes containing main

167 group and transition metals via coordination of the phosphorus lone pair.240-249 As such,

2- the preparation of analogous complexes continaing [R2PB(NDipp)2] is of interest for examining the effects imposed by the boron atom in the backbone. Complex formation by protonolysis reactions between R2PB[N(H)Dipp]2 and metal alkyls or amides may also be possible as the −NH protons would be expected to be more acidic than those in the bog ligands; this has been found to be a feasible route to phosphaguanidainte complexes.243,248

A further extension of this work would involve installing a −CH2PR2 (R = Ph,

Me) group on boron, by the reaction of XB[N(H)Dipp]2 and [Li][CH2PR2], which upon dilithiation would produce a dianionic, tripodal ligand containing hard and soft donors.

Ligands of this type are of growing interest for the preparation of metal complexes with novel reactivity.250

Protonolysis reactions of bis-bamH4 4.4 described in Chapter 4 (used to make tetraanion 4.5) may be a feasible route to complexes of this novel ligand (Eq. 6.1). The tert-butyl groups on the nitrogen atoms could also be replaced by Dipp or Ph substituents, should the alkyl groups decrease the acidity of the −NH proton to such an extent that alkane- or amine-elimination does not proceed.

tBu tBu tBu(H)N N(H)tBu N N 2 MR2 or MR2X B B M B B M - 4 alkane or amine t t N N Bu(H)N N(H) Bu t t Bu Bu (6.1) 4.4 R = alkyl or amide or tBu tBu N N X M B B M X N N tBu tBu

168

The synthesis of mono-metallated bams, also presented in Chapter 4, opens a new facet of bam research as the monoanionic ligand could be reacted in equimolar amounts with main group(I) halides or in a 2:1 stoichiometry with main group dihalides (Scheme

6.2). Since amidinate complexes of main group metals in low oxidation states, e.g. indium(I),32 thallium(I),32 germanium(II),251,252 tin(II),251 and lead(II)253 and a unique magnesium(I)29 guanidinate complex have been reported, preparation of the analogous

[bam(H)]- species would allow for structural and reactivity comparisons.

Dipp N M' M'X Ph B Ph N(H)Dipp - MX B Dipp(H)N NDipp Dipp M - MX N N(H)Dipp

M = Li, K 0.5 Ph B M' B Ph 0.5 M'X2 Dipp(H)N N Dipp

Scheme 6.2

For instance, Mg(I) dimers stabilized by the β-diketiminate ligand have been used as reducing agents to generate a digermanium(0) complex254 and, as mentioned above, the analogous Mg(I) dimer of the guan ligand is known; this makes the [bam(H)]- derivative an interesting target. Additionally, DFT calculations have shown that boron(I) guanidinates may be isolable;255 however, reduction of guan boron dichlorides has not yet been successful in this pursuit.255 The difference in the electronics of a [bam(H)]- species may make reduction possible. Furthermore, the mono-lithiation of

i boraguanidinate Pr2NB[N(H)Dipp]2 (2.3) and subsequent complex formation would also

169 be a worthwhile pursuit in this regard as the resulting ligand would be the boron- derivative of a guan ligand.

Finally, since stable radicals could not be produced from the one-electron oxidation of [Li2][{PhB(NDipp)2}2LnI], preparing the salt-free derivatives

[Li][{PhB(NDipp)2}2Ln] is desirable in the quest for neutral radicals. The 2:1 reaction of

K2bam and lanthanide triiodides should afford the targeted spirocyclic products, which could then be oxidized by half an equivalent of molecular iodine (Scheme 6.3) to form radical species analogous to the group 13 complexes previously reported.110

Dipp Dipp Dipp Dipp

N N N N 0.5 I2 LnI3(THF)3.5 + 2 K2bam Ph B Ln B Ph [K] Ph B Ln B Ph - 3 KI - KI N N N N Dipp Dipp Dipp Dipp

Scheme 6.3

170

References

(1) Rodgers, G. E. Descriptive Inorganic, Coordination, and Solid-State

Chemistry; 2nd ed.; Brooks/Cole, Thomson Learning: Toronto, 2002.

(2) Rosenberg, B.; Vancamp, L.; Trosko, J. E.; Mansour, V. H. Nature 1969,

222, 385.

(3) Henderson, W. Main Group Chemistry; Wiley-Interscience

Royal Society of Chemistry: Bristol, 2002.

(4) Massey, A. G. Main Group Chemistry; 2nd ed.; Wiley: West Sussex,

2000.

(5) Cotton, S. Lanthanide and Actinide Chemistry; John Wiley & Sons, Ltd:

West Sussex, 2006.

(6) Pyykkö, P. Chem. Rev. 1988, 88, 563.

(7) Housecroft, C. E.; Sharpe, A. G. Inorganic Chemistry; Prentice Hall:

Harlow, 2001.

(8) Kauffman, G. B. Inorganic Coordination Compounds; Heyden: London,

1981.

(9) Crabtree, R. H. The Organometallic Chemistry of the Transition Metals;

4th ed.; Wiley-Interscience, 2005.

(10) Pearson, R. G. Science 1966, 151, 172.

(11) Edelmann, F. T. Adv. Organomet. Chem. 2008, 57, 183.

(12) Bailey, P. J.; Pace, S. Coord. Chem. Rev. 2001, 214, 91.

(13) Edelmann, F. T. Coord. Chem. Rev. 1994, 137, 403.

(14) Barker, J.; Kilner, M. Coord. Chem. Rev. 1994, 133, 219.

171

(15) Sanger, A. R. Inorg. Nucl. Chem. Lett. 1973, 9, 351.

(16) Boeré, R. T.; Oakley, R. T.; Reed, R. W. J. Organomet. Chem. 1987, 331,

161.

(17) Chandra, G.; Jenkins, A. D.; Lappert, M. F.; Srivastava, R. C. J. Chem.

Soc. A 1970, 2550.

(18) Junk, P. C.; Cole, M. L. Chem. Commun. 2007, 1579.

(19) Knapp, C.; Lork, E.; Watson, P. G.; Mews, R. Inorg. Chem. 2002, 41,

2014.

(20) Cragghine, I.; Davidson, M. G.; Mair, F. S.; Raithby, P. R.; Snaith, R. J.

Chem. Soc. Dalton Trans. 1993, 2423.

(21) Averbuj, C.; Tish, E.; Eisen, M. S. J. Am. Chem. Soc. 1998, 120, 8640.

(22) Barker, J.; Barr, D.; Barnett, N. D. R.; Clegg, W.; Cragg-Hine, I.;

Davidson, M. G.; Davies, R. P.; Hodgson, S. M.; Howard, J. A. K.; Kilner, M.;

Lehmann, C. W.; Lopez-Solera, I.; Mulvey, R. E.; Raithby, P. R.; Snaith, R. J.

Chem. Soc. Dalton Trans. 1997, 951.

(23) Pietryga, J. M.; Jones, J. N.; Macdonald, C. L. B.; Moore, J. A.; Cowley,

A. H. Polyhedron 2006, 25, 259.

(24) Schmidt, J. A. R.; Arnold, J. Chem. Commun. 1999, 2149.

(25) Schmidt, J. A. R.; Arnold, J. J. Chem. Soc. Dalton Trans. 2002, 2890.

(26) Baker, R. J.; Jones, C. J. Organomet. Chem. 2006, 691, 65.

(27) Aharonovich, S.; Botoshanski, M.; Rabinovich, Z.; Waymouth, R. M.;

Eisen, M. S. Inorg. Chem. 2010, 49, 1220.

(28) Niemeyer, M.; Power, P. P. Inorg. Chem. 1997, 36, 4688.

172

(29) Green, S. P.; Jones, C.; Stasch, A. Science 2007, 318, 1754.

(30) Jones, C.; Junk, P. C.; Platts, J. A.; Stascht, A. J. Am. Chem. Soc. 2006,

128, 2206.

(31) Green, S. P.; Jones, C.; Stasch, A. Inorg. Chem. 2007, 46, 11.

(32) Jones, C.; Junk, P. C.; Platts, J. A.; Rathmann, D.; Stasch, A. Dalton

Trans. 2005, 2497.

(33) Jones, C.; Aldridge, S.; Gans-Eichler, T.; Stasch, A. Dalton Trans. 2006,

5357.

(34) Baker, R. J.; Jones, C.; Junk, P. C.; Kloth, M. Angew. Chem. Int. Ed. 2004,

43, 3852.

(35) Clobanu, O.; Emeljanenko, D.; Kaifer, E.; Mautz, J.; Himmel, H. J. Inorg.

Chem. 2008, 47, 4774.

(36) Green, S. P.; Jones, C.; Junk, P. C.; Lippert, K. A.; Stasch, A. Chem.

Commun. 2006, 3978.

(37) Green, S. P.; Jones, C.; Jin, G. X.; Stasch, A. Inorg. Chem. 2007, 46, 8.

(38) Bailey, P. J.; Gould, R. O.; Harmer, C. N.; Pace, S.; Steiner, A.; Wright,

D. S. Chem. Commun. 1997, 1161.

(39) Brym, M.; Forsyth, C. M.; Jones, C.; Junk, P. C.; Rose, R. P.; Stasch, A.;

Turner, D. R. Dalton Trans. 2007, 3282.

(40) Lyhs, B.; Schulz, S.; Westphal, U.; Blaser, D.; Boese, R.; Bolte, M. Eur. J.

Inorg. Chem. 2009, 2247.

(41) Dawson, D. Y.; Arnold, J. Organometallics 1997, 16, 1111.

(42) Mullins, S. M.; Bergman, R. G.; Arnold, J. Dalton Trans. 2006, 203.

173

(43) Hao, S. K.; Berno, P.; Minhas, R. K.; Gambarotta, S. Inorg. Chim. Acta

1996, 244, 37.

(44) Zou, G.; Ren, T. Inorg. Chim. Acta 2000, 304, 305.

(45) Yamaguchi, Y.; Ozaki, S.; Hinago, H.; Kobayashi, K.; Ito, T. Inorg. Chim.

Acta 2005, 358, 2363.

(46) Rose, R. P.; Jones, C.; Schulten, C.; Aldridge, S.; Stasch, A. Chem. Eur. J.

2008, 14, 8477.

(47) Abdou, H. E.; Mohamed, A. A.; Fackler, J. P. Inorg. Chem. 2007, 46,

9692.

(48) Cotton, F. A.; Daniels, L. M.; Murillo, C. A.; Timmons, D. J.; Wilkinson,

C. C. J. Am. Chem. Soc. 2002, 124, 9249.

(49) Jones, C.; Schulten, C.; Rose, R. P.; Stasch, A.; Aldridge, S.; Woodul, W.

D.; Murray, K. S.; Moubaraki, B.; Brynda, M.; La Macchia, G.; Gagliardi, L.

Angew. Chem. Int. Ed. 2009, 48, 7406.

(50) Rohde, J. U.; Lee, W. T. J. Am. Chem. Soc. 2009, 131, 9162.

(51) Smolensky, E.; Eisen, M. S. Dalton Trans. 2007, 5623.

(52) Nagashima, H.; Kondo, H.; Hayashida, T.; Yamaguchi, Y.; Gondo, M.;

Masuda, S.; Miyazaki, K.; Matsubara, K.; Kirchner, K. Coord. Chem. Rev. 2003,

245, 177.

(53) Edelmann, F. T. Chem. Soc. Rev. 2009, 38, 2253.

(54) Aubrecht, K. B.; Chang, K.; Hillmyer, M. A.; Tolman, W. B. J. Polym.

Sci. A: Polym. Chem. 2001, 39, 284.

174

(55) Luo, Y. J.; Yao, Y. M.; Shen, Q.; Sun, J.; Weng, L. H. J. Organomet.

Chem. 2002, 662, 144.

(56) Wang, J. F.; Sun, H. M.; Yao, Y. M.; Zhang, Y.; Shen, Q. Polyhedron

2008, 27, 1977.

(57) Wang, J. F.; Yao, Y. M.; Zhang, Y.; Shen, Q. Inorg. Chem. 2009, 48, 744.

(58) Yao, Y. M.; Luo, Y. J.; Chen, J. L.; Zhang, Z. Q.; Zhang, Y.; Shen, Q. J.

Organomet. Chem. 2003, 679, 229.

(59) Luo, Y. J.; Yao, Y. M.; Shen, Q.; Yu, K. B.; Weng, L. H. Eur. J. Inorg.

Chem. 2003, 318.

(60) Ajellal, N.; Lyubov, D. M.; Sinenkov, M. A.; Fukin, G. K.; Cherkasov, A.

V.; Thomas, C. M.; Carpentier, J. F.; Trifonov, A. A. Chem. Eur. J. 2008, 14,

5440.

(61) Luo, Y. J.; Wang, X. L.; Chen, J.; Luo, C. C.; Zhang, Y.; Yao, Y. M. J.

Organomet. Chem. 2009, 694, 1289.

(62) Duchateau, R.; Meetsma, A.; Teuben, J. H. Chem. Commun. 1996, 223.

(63) Coles, M. P.; Jordan, R. F. J. Am. Chem. Soc. 1997, 119, 8125.

(64) Dagorne, S.; Guzei, I. A.; Coles, M. P.; Jordan, R. F. J. Am. Chem. Soc.

2000, 122, 274.

(65) Decker, J. M.; Geib, S. J.; Meyer, T. Y. Organometallics 1999, 18, 4417.

(66) Brussee, E. A. C.; Meetsma, A.; Hessen, B.; Teuben, J. H. Chem.

Commun. 2000, 497.

(67) Brandsma, M. J. R.; Brussee, E. A. C.; Meetsma, A.; Hessen, B.; Teuben,

J. H. Eur. J. Inorg. Chem. 1998, 1867.

175

(68) Nelkenbaum, E.; Kapon, M.; Eisen, M. S. Organometallics 2005, 24,

2645.

(69) Bambirra, S.; Bouwkamp, M. W.; Meetsma, A.; Hessen, B. J. Am. Chem.

Soc. 2004, 126, 9182.

(70) Trifonov, A. A.; Fedorova, E. A.; Fukin, G. K.; Bochkarev, M. N. Eur. J.

Inorg. Chem. 2004, 4396.

(71) Trifonov, A. A.; Skvortsov, G. G.; Lyubov, D. M.; Skorodumova, N. A.;

Fukin, G. K.; Baranov, E. V.; Glushakova, V. N. Chem. Eur. J. 2006, 12, 5320.

(72) Brazeau, A. L.; Wang, Z. H.; Rowley, C. N.; Barry, S. T. Inorg. Chem.

2006, 45, 2276.

(73) Brazeau, A. L.; Barry, S. T. Chem. Mater. 2008, 20, 7287.

(74) Li, Z. W.; Barry, S. T.; Gordon, R. G. Inorg. Chem. 2005, 44, 1728.

(75) Coyle, J. P.; Monillas, W. H.; Yap, G. P. A.; Barry, S. T. Inorg. Chem.

2008, 47, 683.

(76) Carmalt, C. J.; Newport, A. C.; O'Neill, S. A.; Parkin, I. P.; White, A. J.

P.; Williams, D. J. Inorg. Chem. 2005, 44, 615.

(77) Potts, S. E.; Carmalt, C. J.; Blackman, C. S.; Abou-Chahine, F.; Pugh, D.;

Davies, H. O. Organometallics 2009, 28, 1838.

(78) Lim, B. S.; Rahtu, A.; Park, J. S.; Gordon, R. G. Inorg. Chem. 2003, 42,

7951.

(79) Li, X. G.; Li, Z. W.; Li, H. Z.; Gordon, R. G. Eur. J. Inorg. Chem. 2007,

1135.

176

(80) Wiedmann, M. K.; Heeg, M. J.; Winter, C. H. Inorg. Chem. 2009, 48,

5382.

(81) Paivasaari, J.; Dezelah, C. L.; Back, D.; El-Kaderi, H. M.; Heeg, M.;

Putkonen, M.; Niinisto, L.; Winter, C. H. J. Mater. Chem. 2005, 15, 4224.

(82) Milanov, A. P.; Fischer, R. A.; Devi, A. Inorg. Chem. 2008, 47, 11405.

(83) Milanov, A. P.; Toader, T.; Parala, H.; Barreca, D.; Gasparotto, A.; Bock,

C.; Becker, H. W.; Ngwashi, D. K.; Cross, R.; Paul, S.; Kunze, U.; Fischer, R. A.;

Devi, A. Chem. Mater. 2009, 21, 5443.

(84) Milanov, A. P.; Thiede, T. B.; Devi, A.; Fischer, R. A. J. Am. Chem. Soc.

2009, 131, 17062.

(85) Fedorchuk, C.; Copsey, M.; Chivers, T. Coord. Chem. Rev. 2007, 251,

897.

(86) Manke, D. R.; Loh, Z. H.; Nocera, D. G. Inorg. Chem. 2004, 43, 3618.

(87) Fuβstetter, H.; Nöth, H. Chem. Ber. Recl. 1979, 112, 3672.

(88) Gudat, D.; Niecke, E.; Nieger, M.; Paetzold, P. Chem. Ber. Recl. 1988,

121, 565.

(89) Fest, D.; Habben, C. D.; Meller, A.; Sheldrick, G. M.; Stalke, D.; Pauer, F.

Chem. Ber. 1990, 123, 703.

(90) Albrecht, T.; Elter, G.; Noltemeyer, M.; Meller, A. Z. Anorg. Allg. Chem.

1998, 624, 1514.

(91) Luthin, W.; Stratmann, J.-G.; Elter, G.; Meller, A.; Heine, A.; Gornitzka,

H. Z. Anorg. Allg. Chem. 1995, 621, 1995.

177

(92) Chivers, T.; Gao, X. L.; Parvez, M. Angew. Chem. Int. Ed. Engl. 1995, 34,

2549.

(93) Geschwentner, M.; Noltemeyer, M.; Elter, G.; Meller, A. Z. Anorg. Allg.

Chem. 1994, 620, 1403.

(94) Koch, H. J.; Roesky, H. W.; Besser, S.; Herbst-Irmer, R. Chem. Ber. Recl.

1993, 126, 571.

(95) Paetzold, P.; Hahnfeld, D.; Englert, U.; Wojnowski, W.; Dreczewski, B.;

Pawelec, Z.; Walz, L. Chem. Ber. Recl. 1992, 125, 1073.

(96) Koch, H. J.; Roesky, H. W.; Bohra, R.; Noltemeyer, M.; Schmidt, H. G.

Angew. Chem. Int. Ed. Engl. 1992, 31, 598.

(97) Habben, C. D.; Heine, A.; Sheldrick, G. M.; Stalke, D. Z. Naturforsch. B.

1992, 47, 1367.

(98) Habben, C. D.; Herbstirmer, R.; Noltemeyer, M. Z. Naturforsch. B. 1991,

46, 625.

(99) Habben, C. D.; Heine, A.; Sheldrick, G. M.; Stalke, D.; Buhl, M.;

Schleyer, P. V. Chem. Ber. 1991, 124, 47.

(100) Heine, A.; Fest, D.; Stalke, D.; Habben, C. D.; Meller, A.; Sheldrick, G.

M. J. Chem. Soc. Chem. Commun. 1990, 742.

(101) Brask, J. K.; Chivers, T.; Schatte, G. Chem. Commun. 2000, 1805.

(102) Chivers, T.; Fedorchuk, C.; Schatte, G.; Brask, J. K. Can. J. Chem. 2002,

80, 821.

(103) Chivers, T.; Fedorchuk, C.; Schatte, G.; Parvez, M. Inorg. Chem. 2003,

42, 2084.

178

(104) Chivers, T.; Fedorchuk, C.; Parvez, M. Inorg. Chem. 2004, 43, 2643.

(105) Chivers, T.; Fedorchuk, C.; Parvez, M. Organometallics 2005, 24, 580.

(106) Konu, J.; Balakrishna, M. S.; Chivers, T.; Swaddle, T. W. Inorg. Chem.

2007, 46, 2627.

(107) Manke, D. R.; Nocera, D. G. Polyhedron 2006, 25, 493.

(108) Manke, D. R.; Nocera, D. G. Inorg. Chim. Acta 2003, 345, 235.

(109) Manke, D. R.; Nocera, D. G. Inorg. Chem. 2003, 42, 4431.

(110) Chivers, T.; Eisler, D. J.; Fedorchuk, C.; Schatte, G.; Tuononen, H. M.;

Boeré, R. T. Chem. Commun. 2005, 3930.

(111) Chivers, T.; Eisler, D. J.; Fedorchuk, C.; Schatte, G.; Tuononen, H. M.;

Boeré, R. T. Inorg. Chem. 2006, 45, 2119.

(112) Konu, J.; Tuononen, H. M.; Chivers, T. Can. J. Chem. 2009, 87, 461.

(113) Konu, J.; Tuononen, H. M.; Chivers, T.; Corrente, A. M.; Boeré, R. T.;

Roemmele, T. L. Inorg. Chem. 2008, 47, 3823.

(114) Ishida, Y.; Donnadieu, B.; Bertrand, G. Proc. Natl. Acad. Sci. U S A 2006,

103, 13585.

(115) Braun, U.; Habereder, T.; Nöth, H.; Piotrowski, H.; Warchhold, M. Eur. J.

Inorg. Chem. 2002, 1132.

(116) Zhang, C.-J.; Zhan, C.-G. Phosphorus, Sulfur, and Silicon, 1997; Vol.

126.

(117) Weber, L.; Rausch, A.; Wartig, H. B.; Stammler, H. G.; Neumann, B. Eur.

J. Inorg. Chem. 2002, 2438.

(118) Jones, C. Coord. Chem. Rev. 2010, 254, 1273.

179

(119) Lo, H. C.; Han, H. N.; D'Souza, L. J.; Sinha, S. C.; Keinan, E. J. Am.

Chem. Soc. 2007, 129, 1246.

(120) Beweries, T.; Jager-Fiedler, U.; Bach, M. A.; Burlakov, V. V.; Arndt, P.;

Baumann, W.; Spannenberg, A.; Rosenthal, U. Organometallics 2007, 26, 3000.

(121) Mommertz, A.; Leo, R.; Massa, W.; Harms, K.; Dehnicke, K. Z. Anorg.

Allg. Chem. 1998, 624, 1647.

(122) Polamo, M.; Mutikainen, I.; Leskela, M. Acta Crystallogr C 1997, 53,

1036.

(123) Breen, T. L.; Stephan, D. W. Inorg. Chem. 1992, 31, 4019.

(124) Guo, Z. Y.; Bradley, P. K.; Jordan, R. F. Organometallics 1992, 11, 2690.

(125) Evans, W. J.; Leman, J. T.; Ziller, J. W.; Khan, S. I. Inorg. Chem. 1996,

35, 4283.

(126) Avens, L. R.; Barnhart, D. M.; Burns, C. J.; McKee, S. D. Inorg. Chem.

1996, 35, 537.

(127) Boisson, C.; Berthet, J. C.; Lance, M.; Nierlich, M.; Ephritikhine, M.

Chem. Commun. 1996, 2129.

(128) Campello, M. P. C.; Domingos, A.; Santos, I. J. Organomet. Chem. 1994,

484, 37.

(129) Campbell, J. P.; Gladfelter, W. L. Inorg. Chem. 1997, 36, 4094.

(130) Kunnari, S. M.; Oilunkaniemi, R.; Laitinen, R. S.; Ahlgren, M. J. Chem.

Soc. Dalton Trans. 2001, 3417.

(131) Chivers, T.; Schatte, G. Eur. J. Inorg. Chem. 2003, 3314.

(132) Welch, G. C.; Masuda, J. D.; Stephan, D. W. Inorg. Chem. 2006, 45, 478.

180

(133) Welch, G. C.; Prieto, R.; Dureen, M. A.; Lough, A. J.; Labeodan, O. A.;

Holtrichter-Rossmann, T.; Stephan, D. W. Dalton Trans. 2009, 1559.

(134) Geier, S. J.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 3476.

(135) Holschumacher, D.; Bannenberg, T.; Hrib, C. G.; Jones, P. G.; Tamm, M.

Angew. Chem. Int. Ed. 2008, 47, 7428.

(136) Matsuo, Y.; Iwashita, A.; Oyama, H.; Nakamura, E. Tetrahedron Lett.

2009, 50, 3411.

(137) Nakata, N.; Izumi, R.; Lee, V. Y.; Ichinohe, M.; Sekiguchi, A. Chem. Lett.

2005, 34, 582.

(138) Guindon, Y.; Yoakim, C.; Morton, H. E. Tetrahedron Lett. 1983, 24,

2969.

(139) Guindon, Y.; Therien, M.; Girard, Y.; Yoakim, C. J. Org. Chem. 1987, 52,

1680.

(140) Cross, W. I.; Lightfoot, M. P.; Mair, F. S.; Pritchard, R. G. Inorg. Chem.

2000, 39, 2690.

(141) Höpfl, H.; Sánchez, M.; Farfán, N.; Barba, V. Can. J. Chem. 1998, 76,

1352.

(142) von Steuber, E.; Elter, G.; Noltemeyer, M.; Schmidt, H. G.; Meller, A.

Organometallics 2000, 19, 5083.

(143) Wrackmeyer, B.; Klimkina, E. V.; Milius, W.; Tok, O. L.; Herberhold, M.

Inorg. Chim. Acta 2005, 358, 1420.

(144) Stepanenko, V.; Ortiz-Marciales, M.; Barnes, C. E.; Garcia, C.

Tetrahedron Lett. 2006, 47, 7603.

181

(145) Kuznetsov, V. V.; Brusilovskii, Y. E.; Mazepa, A. V. Russ. J. Gen. Chem.

2001, 71, 817.

(146) Hirao, A.; Itsuno, S.; Nakahama, S.; Yamazaki, N. J. Chem. Soc. Chem.

Commun. 1981, 315.

(147) Corey, E. J. Angew. Chem. Int. Ed. 2002, 41, 1650.

(148) Paetzold, P. Adv. Inorg. Chem. 1987, 31, 123.

(149) Rivard, E.; Merrill, W. A.; Fettinger, J. C.; Wolf, R.; Spikes, G. H.;

Power, P. P. Inorg. Chem. 2007, 46, 2971.

(150) Hicks, R. G. Org. Biomol. Chem. 2007, 5, 1321.

(151) Power, P. P. Chem. Rev. 2003, 103, 789.

(152) Konu, J.; Chivers, T. In Stable Radicals: Fundamental and Applied

Aspects of Odd-Electron Compounds; Hicks, R. G., Ed.; John Wiley & Sons, Ltd.:

Chicester, UK, 2010, p 381.

(153) Breher, F. Coord. Chem. Rev. 2007, 251, 1007.

(154) Grützmacher, H.; Breher, F. Angew. Chem. Int. Ed. 2002, 41, 4006.

(155) Ferrocenes; Togni, A.; Hayashi, T., Eds.; VCH: Weinheim, 1995.

(156) Ruf, W.; Fueller, M.; Siebert, W. J. Organomet. Chem. 1974, 64, C45.

(157) Renk, T.; Ruf, W.; Siebert, W. J. Organomet. Chem. 1976, 120, 1.

(158) Ruf, W.; Renk, T.; Siebert, W. Z. Naturforsch. B. 1976, 31, 1028.

(159) Kotz, J. C.; Post, E. W. Inorg. Chem. 1970, 9, 1661.

(160) Kotz, J. C.; Post, E. W. J. Am. Chem. Soc. 1968, 90, 4503.

(161) Wrackmeyer, B.; Dorfler, U.; Herberhold, M. Z. Naturforsch. B. 1993, 48,

121.

182

(162) Braunschweig, H.; Breitling, F. M.; Kraft, K.; Kraft, M.; Seeler, F.;

Stellwag, S.; Radacki, K. Z. Anorg. Allg. Chem. 2006, 632, 269.

(163) Appel, A.; Jäkle, F.; Priermeier, T.; Schmid, R.; Wagner, M.

Organometallics 1996, 15, 1188.

(164) Wrackmeyer, B.; Dorfler, U.; Milius, W.; Herberhold, M. Polyhedron

1995, 14, 1425.

(165) Guillaneux, D.; Kagan, H. B. J. Org. Chem. 1995, 60, 2502.

(166) Morris, J. J.; Noll, B. C.; Honeyman, G. W.; O'Hara, C. T.; Kennedy, A.

R.; Mulvey, R. E.; Henderson, K. W. Chem. Eur. J. 2007, 13, 4418.

(167) Haberecht, M. C.; Heilmann, J. B.; Haghiri, A.; Bolte, M.; Bats, J. W.;

Lerner, H. W.; Holthausen, M. C.; Wagner, M. Z. Anorg. Allg. Chem. 2004, 630,

904.

(168) Cole, M. L.; Davies, A. J.; Jones, C.; Junk, P. C. J. Organomet. Chem.

2004, 689, 3093.

(169) Boeré, R. T.; Cole, M. L.; Junk, P. C. New J. Chem. 2005, 29, 128.

(170) Cole, M. L.; Davies, A. J.; Jones, C.; Junk, P. C. New J. Chem. 2005, 29,

1404.

(171) Bailey, P. L.; Coxall, R. A.; Dick, C. M.; Fabre, S.; Henderson, L. C.;

Herber, C.; Liddle, S. T.; Lorono-Gonzalez, D.; Parkin, A.; Parsons, S. Chem.

Eur. J. 2003, 9, 4820.

(172) Arnold, P. L.; Liddle, S. T. C.R. Chim 2008, 11, 603.

(173) Baldamus, J.; Berghof, C.; Cole, M. L.; Evans, D. J.; Hey-Hawkins, E.;

Junk, P. C. J. Chem. Soc. Dalton Trans. 2002, 2802.

183

(174) Cole, M. L.; Junk, P. C. J. Organomet. Chem. 2003, 666, 55.

(175) Yao, S.; Chan, H. S.; Lam, C. K.; Lee, H. K. Inorg. Chem. 2009, 48, 9936.

(176) Rodriguez, A.; Fuks, G.; Bourg, J. B.; Bourissou, D.; Tham, F. S.;

Bertrand, G. Dalton Trans. 2008, 4482.

(177) Rodriguez, A.; Tham, F. S.; Schoeller, W. W.; Bertrand, G. Angew. Chem.

Int. Ed. 2004, 43, 4876.

(178) Hodgson, K. O.; Raymond, K. N. Inorg. Chem. 1972, 11, 171.

(179) Cassani, M. C.; Gunko, Y. K.; Hitchcock, P. B.; Lappert, M. F. Chem.

Commun. 1996, 1987.

(180) Cassani, M. C.; Gun'ko, Y. K.; Hitchcock, P. B.; Lappert, M. F.; Laschi, F.

Organometallics 1999, 18, 5539.

(181) Fedushkin, I. L.; Petrovskaya, T. V.; Girgsdies, F.; Kohn, R. D.;

Bochkarev, M. N.; Schumann, H. Angew. Chem. Int. Ed. 1999, 38, 2262.

(182) Liu, Q. C.; Ding, M. X.; Lin, Y. G.; Xing, Y. J. Organomet. Chem. 1997,

548, 139.

(183) Dube, T.; Gambarotta, S.; Yap, G. Organometallics 1998, 17, 3967.

(184) Liu, Q. C.; Ding, M. X.; Lin, Y. H.; Xing, Y. Polyhedron 1998, 17, 2327.

(185) Liu, Q. C.; Ding, M. X.; Lin, Y. H.; Xing, Y. Polyhedron 1998, 17, 555.

(186) Liu, Q. C.; Ding, M. X. J. Organomet. Chem. 1998, 553, 179.

(187) Evans, W. J.; Fujimoto, C. H.; Ziller, J. W. Chem. Commun. 1999, 311.

(188) Schuetz, S. A.; Day, V. W.; Sommer, R. D.; Rheingold, A. L.; Belot, J. A.

Inorg. Chem. 2001, 40, 5292.

(189) Evans, W. J.; Fujimoto, C. H.; Ziller, J. W. Polyhedron 2002, 21, 1683.

184

(190) Schuetz, S. A.; Day, V. W.; Clark, J. L.; Belot, J. A. Inorg. Chem.

Commun. 2002, 5, 706.

(191) Schuetz, S. A.; Day, V. W.; Rheingold, A. L.; Belot, J. A. Dalton Trans.

2003, 4303.

(192) Schuetz, S. A.; Silvernail, C. M.; Incarvito, C. D.; Rheingold, A. L.; Clark,

J. L.; Day, V. W.; Belot, J. A. Inorg. Chem. 2004, 43, 6203.

(193) Roesky, P. W. Organometallics 2002, 21, 4756.

(194) Sugiyama, H.; Korobkov, I.; Gambarotta, S.; Moller, A.; Budzelaar, P. H.

M. Inorg. Chem. 2004, 43, 5771.

(195) Lorenz, V.; Gorls, H.; Thiele, S. K. H.; Scholz, J. Organometallics 2005,

24, 797.

(196) Collin, J.; Daran, J. C.; Jacquet, O.; Schulz, E.; Trifonov, A. Chem. Eur. J.

2005, 11, 3455.

(197) Li-Ying, Z.; Hong-Ting, S.; Ying-Ming, Y.; Yong, Z.; Qi, S. J.

Organomet. Chem. 2007, 692, 2990.

(198) Aspinall, H. C. Chem. Rev. 2002, 102, 1807.

(199) Pi, C. F.; Zhang, Z. X.; Liu, R. T.; Weng, L. H.; Chen, Z. X.; Zhou, X. G.

Organometallics 2006, 25, 5165.

(200) Peng, H. M.; Zhang, Z. Q.; Qi, R. P.; Yao, Y. M.; Zhang, Y.; Shen, Q.;

Cheng, Y. X. Inorg. Chem. 2008, 47, 9828.

(201) Vasudevan, K.; Cowley, A. H. Chem. Commun. 2007, 3464.

(202) Fedushkin, I. L.; Maslova, O. V.; Baranov, E. V.; Shavyrin, A. S. Inorg.

Chem. 2009, 48, 2355.

185

(203) Lu, E. L.; Gan, W.; Chen, Y. F. Organometallics 2009, 28, 2318.

(204) Rastätter, M.; Muterle, R. B.; Roesky, P. W.; Thiele, S. K. H. Chem. Eur.

J. 2009, 15, 474.

(205) Zhou, L. Y.; Sun, H. M.; Chen, J. G.; Yao, Y. M.; Shen, Q. J. Polym. Sci.

A: Polym. Chem. 2005, 43, 1778.

(206) Li, C.; Wang, Y. R.; Zhou, L. Y.; Sun, H. M.; Shen, Q. J. Appl. Polym.

Sci. 2006, 102, 22.

(207) Stanlake, L. J. E.; Beard, J. D.; Schafer, L. L. Inorg. Chem. 2008, 47,

8062.

(208) Yang, Y.; Wang, Q. Y.; Cui, D. M. J. Polym. Sci. A: Polym. Chem. 2008,

46, 5251.

(209) Dehnicke, K.; Greiner, A. Angew. Chem. Int. Ed. 2003, 42, 1340.

(210) Namy, J. L.; Girard, P.; Kagan, H. B. Nouv. J. Chim. 1977, 1, 5.

(211) Namy, J. L.; Girard, P.; Kagan, H. B.; Caro, P. E. Nouv. J. Chim. 1981, 5,

479.

(212) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc.

1981, 103, 6507.

(213) Nicolaou, K. C.; Ellery, S. P.; Chen, J. S. Angew. Chem. Int. Ed. 2009, 48,

7140.

(214) Evans, W. J.; Davis, B. L. Chem. Rev. 2002, 102, 2119.

(215) Evans, W. J. Coord. Chem. Rev. 2000, 206, 263.

(216) Molander, G. A.; Harris, C. R. Chem. Rev. 1996, 96, 307.

(217) Hill, M. S.; Hitchcock, P. B. Dalton Trans 2003, 4570.

186

(218) Cole, M. L.; Junk, P. C. Chem. Commun. 2005, 2695.

(219) Heitmann, D.; Jones, C.; Junk, P. C.; Lippert, K. A.; Stasch, A. Dalton

Trans. 2007, 187.

(220) Wiecko, M.; Roesky, P. W.; Burlakov, V. V.; Spannenberg, A. Eur. J.

Inorg. Chem. 2007, 876.

(221) Heitmann, D.; Jones, C.; Mills, D. P.; Stasch, A. Dalton Trans. 2010, 39,

1877.

(222) Harder, S.; Naglav, D. Eur. J. Inorg. Chem. 2010, 2836.

(223) Harder, S. Dalton Trans 2010, 39, 6677.

(224) Peters, J. A.; Huskens, J.; Raber, D. J. Prog. Nucl. Magn. Reson.

Spectrosc. 1996, 28, 283.

(225) Dutton, J. L.; Sutrisno, A.; Schurko, R. W.; Ragogna, P. J. Dalton Trans.

2008, 3470.

(226) Geoffrey, F.; Cloke, N.; Elvidge, B. R.; Hitchcock, P. B.; Lamarche, V.

M. E. J. Chem. Soc. Dalton Trans. 2002, 2413.

(227) Izod, K.; Liddle, S. T.; Clegg, W. Inorg. Chem. 2004, 43, 214.

(228) Cole, M. L.; Deacon, G. B.; Junk, P. C.; Konstas, K. Chem. Commun.

2005, 1581.

(229) Bowman, L. J.; Izod, K.; Clegg, W.; Harrington, R. W. Organometallics

2006, 25, 2999.

(230) Harder, S., Personal Communication.

(231) Corrente, A. M.; Chivers, T. Inorg. Chem. 2010, 49, 2457.

(232) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877.

187

(233) Evans, W. J.; Drummond, D. K.; Zhang, H. M.; Atwood, J. L. Inorg.

Chem. 1988, 27, 575.

(234) Jeffery, J. C.; Leedham, A. P.; Russell, C. A. Polyhedron 2002, 21, 549.

(235) Fleischer, R.; Freitag, S.; Pauer, F.; Stalke, D. Angew. Chem. Int. Ed.

Engl. 1996, 35, 204.

(236) Chivers, T.; Parvez, M.; Schatte, G. Inorg. Chem. 1996, 35, 4094.

(237) Brask, J. K.; Chivers, T.; McGarvey, B.; Schatte, G.; Sung, R.; Boeré, R.

T. Inorg. Chem. 1998, 37, 4633.

(238) Fleischer, R.; Freitag, S.; Stalke, D. J. Chem. Soc. Dalton Trans. 1998,

193.

(239) Power, P. P. Angew. Chem. Int. Ed. Engl. 1990, 29, 449.

(240) Coles, M. P.; Hitchcock, P. B. Chem. Commun. 2002, 2794.

(241) Grundy, J.; Coles, M. P.; Hitchcock, P. B. Dalton Trans. 2003, 2573.

(242) Grundy, J.; Coles, M. P.; Avent, A. G.; Hitchcock, P. B. Chem. Commun.

2004, 2410.

(243) Mansfield, N. E.; Coles, M. P.; Hitchcock, P. B. Dalton Trans. 2005,

2833.

(244) Iravani, E.; Neumuller, B.; Grunenberg, J. Z. Anorg. Allg. Chem. 2006,

632, 739.

(245) Mansfield, N. E.; Coles, M. P.; Hitchcock, P. B. Dalton Trans. 2006,

2052.

(246) Mansfield, N. E.; Grundy, J.; Coles, M. P.; Avent, A. G.; Hitchcock, P. B.

J. Am. Chem. Soc. 2006, 128, 13879.

188

(247) Zhang, W. X.; Nishiura, M.; Hou, Z. M. Chem. Commun. 2006, 3812.

(248) Mansfield, N. E.; Coles, M. P.; Avent, A. G.; Hitchcock, P. B.

Organometallics 2006, 25, 2470.

(249) Grundy, J.; Mansfield, N. E.; Coles, M. P.; Hitchcock, P. B. Inorg. Chem.

2008, 47, 2258.

(250) Liang, L. C. Coord. Chem. Rev. 2006, 250, 1152.

(251) Foley, S. R.; Bensimon, C.; Richeson, D. S. J. Am. Chem. Soc. 1997, 119,

10359.

(252) Foley, S. R.; Yap, G. P. A.; Richeson, D. S. J. Chem. Soc. Dalton Trans.

2000, 10, 1663.

(253) Stasch, A.; Forsyth, C. M.; Jones, C.; Junk, P. C. New J. Chem. 2008, 32,

829.

(254) Sidiropoulos, A.; Jones, C.; Stasch, A.; Klein, S.; Frenking, G. Angew

Chem Int Ed Engl 2009, 48, 9701.

(255) Findlater, M.; Hill, N. J.; Cowley, A. H. Dalton Trans. 2008, 4419.

(256) Schade, C.; Bauer, W.; Schleyer, P. V. J. Organomet. Chem. 1985, 295,

C25.

(257) Sheldrick, G. M.; University of Göttingen, Germany: SHELXS-97,

Program for solution of crystal structures, 1997.

(258) Sheldrick, G. M.; University of Göttingen, Germany: SHELXL-97,

Program for refinement of crystal structures, 1997.

(259) van der Sluis, P.; Spek, A. L. Acta Crystallogr. A 1990, 46, 194.

189

APPENDIX A: REAGENTS, PROCEDURES AND INSTRUMENTATION

A.1. Reagents and General Procedures

All reactions and the manipulation of moisture- and/or air-sensitive reagents or products were carried out under an atmosphere of argon using standard Schlenk line techniques or in an inert-atmosphere glove box. Solvents were dried with appropriate drying agents, distilled under argon before use, and stored over molecular sieves. Prior to use, all glassware was carefully dried. Deuterated solvents were purchased from

Cambridge Isotope Laboratories, dried over 4 Å molecular sieves for at least one week and degassed using at least three cycles of the freeze-pump-thaw method.

The anhydrous reagents InCl (99.995%), CdCl2, PrCl3, NdCl3, HoCl3, ErCl3 and

YbCl3 were purchased from Strem, TeI4 (99%) and TlCl (99.99%) were purchased from

Alfa Aesar, dicyclopentadienyl iron (Cp2Fe) was purchased from Eastman Organic

Chemical Company and all other chemicals were purchased from Aldrich Chemical

Company. All chemicals were used as received, with the following exceptions: 2,6- diisopropylaniline (97%, Aldrich) was purified by distillation (at approximately 100°C

-2 and 10 Torr) and anhydrous ZnCl2 was prepared by heating ZnCl2·H2O under vacuum

(150°C, 10-2 Torr, 18 h).

i The reagents [Li][N(H)Dipp], [Li][N Pr2], [Li][N(Me)Ph], [Li][NPh2] and

[Li][N(H)tBu] were prepared by the addition of nBuLi (2.5 M in hexane) in an equimolar amount to the corresponding amine in n-hexanes and purity was checked by 1H NMR

104 156,158 161 spectroscopy. The reagents PhB[N(H)Dipp]2, FcBBr2, 1,1′-Fc(BBr2)2, 1,4-

167 171 256 Br2BC6H4BBr2, benzylpotassium, and n-butylsodium were prepared by literature methods.

190

A.2. Instrumentation

All NMR spectra were acquired at room temperature using a Bruker DRX 400 spectrometer. All chemical shifts are reported in parts per million (ppm) with higher frequency taken as positive. Chemical shifts for 1H and 13C{1H} NMR spectra are reported with respect to tetramethylsilane and were calibrated based on the signal of the residual solvent peak. A solution of 1.0 M LiCl in D2O was used as the external standard for 7Li NMR spectra, 11B{1H} NMR chemical shifts are reported with respect to a

31 1 solution of BF3•OEt2 in C6D6, and P{ H} chemical shifts are relative to 85% H3PO4 in

D2O.

Elemental and mass spectrometry analyses (using a Waters GCT Premier spectrometer) were performed by the Analytical Services Laboratory of the Department of Chemistry, University of Calgary. EPR spectra were collected using a Bruker EMX

113 spectrometer operating at X-band frequencies.

A.3. X-Ray Structure Determinations

A single crystal suitable for X-ray analysis was covered with Paratone oil and mounted on a glass fibre in a stream of N2 at 173 K on a Nonius KappaCCD

diffractometer (MoKα radiation, λ = 0.71073 Å) using COLLECT (Nonius, B.V. 1998) software. The unit cell parameters were calculated and refined from the full data set. All crystal cell refinement and data reduction was carried out using the Nonius DENZO package. After reduction, the data were corrected for absorption based on equivalent reflections using SCALEPACK (Nonius, B.V. 1998). The structures were solved by direct or Patterson methods with SHELXS-97257 and refinement was carried out on F2

191 against all independent reflections by the full-matrix least-squares method by using the

SHELXL-97258 program. All non-hydrogen atoms were refined with anisotropic thermal parameters. The hydrogen atoms were calculated geometrically and were riding on their respective atoms. Special considerations and changes to the above procedures for certain structures are listed below.

Compound 3.2a: Two carbon atoms in the seven-membered ring were disordered and modelled isotropically in a 60:40 ratio with minor geometric restraints.

Compound 4.3: The hydrogen atoms on nitrogen were located and refined isotropically.

Compound 4.5: A disordered diethyl ether molecule was present in the lattice and no suitable model could be found. The electron density associated with this solvent molecule was removed from the reflection data using the program SQUEEZE (PLATON)259 leaving a void of 355.7 Å3.

Compound 4.8: A disordered hexane molecule was present in the lattice and no suitable model could be found. The electron density associated with this solvent molecule was removed from the reflection data using the program SQUEEZE (PLATON),259 leaving a void of 327.4 Å3.

Compound 5.14: A disordered hexane molecule was present in the lattice and no suitable model could be found. The electron density associated with this solvent molecule was

192 removed from the reflection data using the program SQUEEZE (PLATON),259 leaving a void of 509.4 Å3.

Compound 5.15: A disordered hexane molecule was present in the lattice and no suitable model could be found. The electron density associated with this solvent molecule was removed from the reflection data using the program SQUEEZE (PLATON),259 leaving a void of 496.2 Å3.

Compound 5.18: Dr. Jari Konu performed the initial structural solution.

Compound 5.21: Two disordered THF molecules were present in the lattice and no suitable model could be found. The electron density associated with these solvent molecules was removed from the reflection data using the program SQUEEZE

(PLATON),259 leaving a void of 870.1 Å3.

193

APPENDIX B: CRYSTALLOGRAPHIC DATA

For all data: λ(MoKα) = 0.71073 Å, T = 173(2) K; R1 = Σ⏐⏐Fo⏐-⏐Fc⏐⏐/Σ⏐Fo⏐; wR2 =

2 2 2 4 ½ [Σw(Fo -Fc ) /ΣwFo ] .

Table B.1 Crystallographic data for 2.2-2.4.

2.2 2.3 2.4 empirical formula C30H39BClN2P C30H50BN3 C33H55BLi2N3 fw 504.86 463.54 518.49 cryst. system Monoclinic Orthorhombic Orthorhombic space group P2(1)/c Pbca Pbcn a, Å 15.876(3) 18.7744(4) 14.103(3) b, Å 12.416(3) 15.687(3) 21.061(4) c, Å 15.917(3) 20.218(4) 21.701(4) β, deg. 115.22(3) 90 90 V, Å3 2839(1) 5954(2) 6446(2) Z 4 8 8 3 ρcalcd, g/cm 1.181 1.034 1.069 µ(Mo Kα), mm-1 0.212 0.059 0.060 crystal size, mm3 0.10 x 0.08 x 0.06 0.40 x 0.36 x 0.28 0.40x0.16 x 0.08 F(000) 1080 2048 0.080.080.082280 Θ range, deg 3.27-25.03 3.38-25.03 2.56-25.03 reflns collected 64200 87203 34800 unique reflns 4807 5242 5663

Rint 0.0539 0.0234 0.0502

R1 [I>2σ(I)] 0.0429 0.0468 0.0627 wR2 (all data) 0.1087 0.1238 0.1749 GOF on F2 1.016 1.033 1.030 completeness 0.961 0.996 0.994

194

Table B.2 Crystallographic data for 2.5, 2.7 and 3.1.

2.6 2.7 3.1 empirical formula C24H36BBrN2 C36H54BN3 C28H43BN2O fw 443.27 539.63 434.45 cryst. system Monoclinic Monoclinic Monoclinic space group P2(1)/n P2(1)/n P2(1)/n a, Å 19.626(4) 10.744(2) 10.839(2) b, Å 6.426(1) 28.035(6) 20.360(4) c, Å 20.311(4) 11.875(2) 12.049(2) β, deg. 103.98(3) 112.08(3) 97.85(3) V, Å3 2486(1) 3314(1) 2634(1) Z 4 4 4 3 ρcalcd, g/cm 1.184 1.081 1.096 µ(Mo Kα), mm-1 1.665 0.062 0.065 crystal size, mm3 0.08x0.08x0.04 0.28x0.28x0.16 0.28x0.28x0.24 F(000) 936 1184 952 Θ range, deg 3.35-25.03 2.99-25.03 2.57-25.03 reflns collected 46407 42938 26111 unique reflns 4379 5830 4634

Rint 0.1371 0.0393 0.0278

R1 [I>2σ(I)] 0.0643 0.0583 0.0597 wR2 (all data) 0.1566 0.1507 0.1671 GOF on F2 1.039 1.032 1.057 completeness 0.995 0.996 0.995

195

Table B.3 Crystallographic data for 4.1-4.3.

4.1 4.2 4.3 empirical formula C18H49BFeN2 C34H45BFeN2 C26H48B2FeN4 fw 340.09 548.38 494.15 cryst. system Orthorhombic Monoclinic Triclinic space group Pna2(1) C2/c P-1 a, Å 17.365(4) 21.451(4) 10.265(2) b, Å 5.996(1) 10.016(2) 12.189(2) c, Å 35.480(7) 28.699(6) 13.101(3) α, deg. 90 90 109.70(3) β, deg. 90 99.88(3) 102.12(3) γ, deg. 90 90 103.19(3) V, Å3 3670(1) 6075(2) 1427(1) Z 8 8 2 3 ρcalcd, g/cm 1.231 1.199 1.150 µ(Mo Kα), mm-1 0.820 0.521 0.548 crystal size, mm3 0.28 x 0.08 x 0.04 0.20x0.16 x 0.04 0.40x0.40x0.32 F(000) 1456 2352 536 Θ range, deg 3.28-25.02 2.25-27.46 2.97-25.03 reflns collected 22275 30049 24390 unique reflns 5629 6921 5021 a Rint 0.0 0.0583 0.0200

R1 [I>2σ(I)] 0.0428 0.0525 0.0366 wR2 (all data) 0.1121 0.1380 0.0966 GOF on F2 1.126 1.081 1.043 completeness 0.943 0.995 0.996 a Flack parameter: 0.50(2).

196

Table B.4 Crystallographic data for 4.5, 4.6 and 4.8.

4.5 4.6 4.8 empirical formula C46H88B2Li4N4O6 C38H56BLiN2O2 C42H63BK2N2O3 fw 842.58 590.60 732.95 cryst. system Monoclinic Monoclinic Monoclinic space group P2(1)/c P2(1)/n P2(1)/c a, Å 10.003(2) 14.028(3) 10.456(2) b, Å 14.712(3) 16.356(3) 15.323(3) c, Å 19.859(4) 16.606(3) 27.854(6) β, deg. 91.29(3) 103.66(3) 90.91(3) V, Å3 2922(1) 3702(1) 4462(2) Z 2 4 4 3 ρcalcd, g/cm 0.958 1.060 1.091 µ(Mo Kα)/ mm-1 0.060 0.063 0.248 crystal size/ mm3 0.20 x 0.20 x 0.20 0.14x0.14 x 0.12 0.20x0.12x0.08 F(000) 924 0.080.080.081288 1584 Θ range, deg 2.65-25.03 2.79-25.03 2.57-25.03 reflns collected 32486 56077 37588 unique reflns 5107 6541 7805

Rint 0.0188 0.0666 0.0417

R1 [I>2σ(I)] 0.0679 0.0740 0.0859 wR2 (all data) 0.2254 0.1958 0.2078 GOF on F2 1.074 1.059 1.093 completeness 0.990 0.998 0.991

197

Table B.5 Crystallographic data for 5.7-5.9.

5.7 5.8 5.9 empirical formula C58H95BCl2LiN2O7Y C58H95BCl2LiN2O7Pr C58H95BCl2LiN2O7Nd fw 1109.92 1161.92 1165.25 cryst. system Monoclinic Monoclinic Monoclinic space group P2(1)/n P2(1)/n P2(1)/n a, Å 13.385(3) 13.320(3) 13.347(3) b, Å 21.503(4) 21.595(4) 21.586(4) c, Å 22.053(4) 22.190(4) 22.178(4) β, deg. 102.88(3) 103.04(3) 103.04(3) V, Å3 6188(2) 6218(1) 6225(2) Z 4 4 4 3 ρcalcd, g/cm 1.191 1.241 1.243 µ(Mo Kα)/ mm-1 1.077 0.917 0.968 crystal size/ mm3 0.36 x 0.32 x 0.24 0.16 x 0.16 x 0.12 0.24 x 0.24 x 0.16 F(000) 2376 0.080.080.082456 2460 Θ range, deg 3.00-25.03 2.75-25.03 3.13-25.03 reflns collected 81171 77592 86571 unique reflns 10864 10951 10965

Rint 0.0679 0.0477 0.0337

R1 [I>2σ(I)] 0.0644 0.0620 0.0433 wR2 (all data) 0.1567 0.1296 0.1043 GOF on F2 1.085 1.103 1.059 completeness 0.994 0.996 0.996

198

Table B.6 Crystallographic data for 5.10-5.12.

5.10 5.11 5.12 empirical C58H95BCl2LiN2O7Sm C58H95BCl2LiN2O7Ho C58H95BCl2LiN2O7Er formulafw 1171.36 1185.94 1188.27 cryst. system Monoclinic Monoclinic Monoclinic space group P2(1)/n P2(1)/n P2(1)/n a, Å 13.356(3) 13.370(3) 13.368(3) b, Å 21.563(4) 21.484(4) 21.474(4) c, Å 22.123(4) 22.045(4) 22.046(4) β, deg. 102.99(3) 102.91 (3) 102.91(3) V, Å3 6208(2) 6172(1) 6169(2) Z 4 4 4 3 ρcalcd, g/cm 1.253 1.276 1.279 µ(Mo Kα)/ mm-1 1.080 1.417 1.495 crystal size/ mm3 0.20 x 0.20 x 0.16 0.60 x 0.48 x 0.16 0.16 x 0.12 x 0.08 F(000) 2468 0.080.080.082488 2492 Θ range, deg 2.67-25.03 2.68-25.03 2.68-25.03 reflns collected 65850 57847 44171 unique reflns 10907 10805 10542

Rint 0.0249 0.0350 0.0536 b R1 [I>2σ(I)] 0.0440 0.0445 0.0677 c wR2 (all data) 0.1152 0.1155 0.1366 GOF on F2 1.073 1.065 1.124 completeness 0.994 0.991 0.967

199

Table B.7 Crystallographic data for 5.13-5.15.

5.13 5.14 5.15 empirical formula C58H95BCl2LiN2O7Yb C38H55BClN2O2Y C38H55BClN2O2Sm fw 1194.05 707.01 768.45 cryst. system Monoclinic Monoclinic Monoclinic space group P2(1)/n P2(1)/n P2(1)/n a, Å 13.394 (3) 13.690(3) 13.825(3) b, Å 21.458(4) 15.867(3) 15.955(4) c, Å 22.062(4) 18.516(4) 18.451(4) β, deg. 102.86(3) 93.24(3) 94.06(3) V, Å3 6182(2) 4016(1) 4056(2) Z 4 4 4 3 ρcalcd, g/cm 1.283 1.169 1.257 µ(Mo Kα)/ mm-1 1.647 1.550 1.543 crystal size/ mm3 0.28 x 0.20 x 0.16 0.16 x 0.14 x 0.08 0.20 x 0.16 x 0.10 F(000) 2500 0.080.080.081496 1588 Θ range, deg 2.68-25.03 2.55-25.03 2.56-25.03 reflns collected 54758 51912 54928 unique reflns 10686 7069 7149

Rint 0.0366 0.1061 0.1056 b R1 [I>2σ(I)] 0.0453 0.0661 0.0565 c wR2 (all data) 0.1136 0.1401 0.1184 GOF on F2 1.073 1.077 1.060 completeness 0.979 0.997 0.998

200

Table B.8 Crystallographic data for 5.18 and 5.21.

5.18 5.21 empirical formula C96H150B2ILaLi2N4O9 C96H150B2ISmLi2N4O9 fw 1805.51 1816.95 cryst. system Orthorhomic Triclinic space group P2(1)2(1)2(1) P-1 a, Å 17.519(4) 15.970(3) b, Å 22.534(5) 16.058(3) c, Å 14.394(3) 21.356(4) α, deg. 90 85.43(3) β, deg. 90 89.49(3) γ, deg. 90 85.47(3) V, Å3 5682(2) 5422(2) Z 2 2 3 ρcalcd, g/cm 1.055 1.109 µ(Mo Kα)/ mm-1 0.694 0.872 crystal size/ mm3 0.32 x 0.16 x 0.08 0.28 x 0.24 x 0.20 F(000) 1904 0.080.080.081914 Θ range, deg 2.04-25.03 2.55-25.03 reflns collected 38671 64743 unique reflns 9451 18346

Rint 0.0374 0.0290 b R1 [I>2σ(I)] 0.0445 0.0508 c wR2 (all data) 0.1359 0.1279 GOF on F2 1.149 1.044 completeness 0.959 0.955