transition metal complexes and main group frustrated lewis pairs for stoichiometric ... ·...
TRANSCRIPT
Transition Metal Complexes and Main Group Frustrated
Lewis Pairs for Stoichiometric and Catalytic P-P and H-H
Bond Activation
by
Stephen Joseph Geier
A thesis submitted in conformity with the requirements
for the degree of Doctor of Philosophy
Department of Chemistry
University of Toronto
© Copyright by Stephen Joseph Geier 2010
ii
Transition Metal and Main Group Frustrated Lewis Pairs for Stoichiometric and Catalytic P-P
and H-H Bond Activation
Stephen Joseph Geier
Doctor of Philosophy
Department of Chemistry
University of Toronto
2010
Abstract
Stoichiometric and catalytic small molecule activation reactions are vital for the synthesis
of new materials. The activation of phosphorus-hydrogen or phosphorus-phosphorus bonds
allows for the facile synthesis of new phosphorus-containing molecules for a wide variety of
applications.1
An investigation of the P-H dehydrocoupling reaction was undertaken utilizing two
rhodium(I) based catalysts. Over the course of this investigation it was found that the Rh(I)
systems were also active catalysts for the reverse reaction: phosphorus-phosphorus bond
hydrogenation (and hydrosilylation). This reaction was exploited for the synthesis of novel
phosphines from P-P bound species. Molecules with P-P bonds were reacted in a stoichiometric
fashion with the catalyst precursor, producing a variety of novel species with interesting bonding
features which shed some light on the reaction mechanism.
iii
Following the discovery in 2006 that a linked phosphine-borane system could reversibly
activate hydrogen2 a tremendous effort has been put forth to understand and expand this
unprecedented reactivity.3,4
This new archetype for metal-free small molecule activation,
containing a bulky Lewis acid and Lewis base which are unable to bond directly due to steric
repulsion, has been termed a “frustrated Lewis pair” (FLP).3,4
The FLP concept is expanded to include bulky P-P bound species, pyridines and P-O
bound Lewis bases as partners for B(C6F5)3. In some cases small molecule activation produced
ion pairs or zwitterions related to those found for reactions with tertiary phosphines,3,4
but in
others novel reaction pathways were discovered including phosphorus-phosphorus bond
cleavage, catalytic hydrogenations and the formation of novel intramolecular FLPs. An
unexpected situation was observed for the pair of 2,6-lutidine with B(C6F5)3, where adduct
formation was observed along with free Lewis acid and base, but H2 activation by the FLP
proceeded smoothly.
Covalently bound phosphinoboranes of the general formula R2PB(C6F5)2 are synthesized.
While systems with small R groups dimerized, monomers existed for cases with bulkier R
groups. These monomers were found to exhibit extraordinarily short phosphorus-boron bonds
yet were still capable of H2 activation analogous to bimolecular phosphine-borane systems.
These systems also showed unique reactivity with Lewis acids and Lewis bases.
This work further demonstrates the broad and general utility of the FLP concept in the
synthesis of new materials and in catalytic transformations.
iv
All synthetic and characterization work was performed by the candidate, with the
valuable assistance of others (see Acknowledgements), with the exception of Appendix A, a
significant portion of which was performed by Eva Ouyang (an undergraduate student working
under the candidate’s guidance) and portions of the work presented in section 5.3.6 which was
part of a collaboration with Jonathan Webb, Veronique Laberge and Dr. Cathleen Crudden of
Queen’s University.
Portions of this work have been previously discussed in the following publications:
Chapter 2: Geier, S.J.; Stephan, D.W. Chem. Commun. 2008, 99. Geier, S.J.; Stephan, D.W.
Chem. Commun. 2008, 2779.
Chapter 3: Geier, S.J.; Dureen, M.A.; Ouyang, E.Y.; Stephan, D.W. Chem.-Eur. J. 2010, 16, 988.
Geier, S.J.; Stephan, D.W. Chem. Commun. 2010, 1026.
Chapter 4: Geier, S.J.; Gilbert, T.M.; Stephan, D.W. J. Am. Chem. Soc. 2008, 130, 12632.
Chapter 5: Geier, S.J.; Stephan, D.W. J. Am. Chem. Soc. 2009, 131, 3476. Geier, S.J.; Gille,
A.L.; Gilbert, T.M.; Stephan, D.W. Inorg. Chem. 2009, 48, 10466. Webb, J.D.; Laberge, V.S.;
Stephan, D.W.; Crudden, C.M. Chem-Eur. J. 2010, 16, 4895. Geier, S.J.; Chase, P.A.; Stephan,
D.W. Chem. Commun. 2010, In Press. Birkmann, B.; Voss, T.; Geier, S.J.; Ullrich, M.; Kehr,
G.; Erker, G.; Stephan, D.W. Organometallics. 2010, In Press.
Appendix A: Neu, R.C.; Ouyang, E.Y.; Geier, S.J.; Zhao, X.; Ramos, A.; Stephan, D.W. Dalton
Trans. 2010, 39, 4285.
v
Acknowledgments
I would like to take this opportunity to thank the many people who have helped me in my
time at the University of Windsor and at the University of Toronto. In particular I would like to
thank Jenny McCahill, Greg Welch and Preston Chase who were invaluable mentors and friends
during my “formative years” in graduate school and also for assistance with the preparation of
this thesis. I would also like to thank Dr. Charles MacDonald, Dr. Samuel Johnson, Dr. Ulrich
Fekl and Dr. Robert Morris for serving on my committees and offering helpful advice. Next, I
would like to thank Mike Fuerth and Dr. Robert Schurko for setting up several 2D NMR
experiments and for help with the analysis of some complex NMR spectra. Special thanks to
Alan Lough for invaluable assistance with X-ray crystallography. Thanks to Eva Ouyang for her
contributions to the phosphite/phosphinite work presented in Appendix A.
I would like to thank Megan Mitton for the support throughout my graduate studies,
feigning interest in my work and particularly for keeping me ground and not letting my
frustration get the best of me.
Thanks to Steve Westcott, who deserves much of the credit (or blame) for starting me
down this path.
Finally, I would like to thank Doug Stephan for guidance, encouragement, financial
support and offering the freedom to pursue different avenues of research during my 4+ years of
graduate studies.
vi
Table of Contents
Acknowledgments ........................................................................................................................... v
Table of Contents ........................................................................................................................... vi
List of Tables ................................................................................................................................. xi
List of Figures .............................................................................................................................. xiii
List of Abbreviations, Nomenclature and Symbols ................................................................ xix
Chapter 1: Introduction ................................................................................................................... 1
1.1: Phosphines in Coordination and Materials Chemistry ........................................................ 1
1.2: P-P Bond Activation ........................................................................................................... 3
1.3: Lewis Acid/Lewis Base Chemistry ..................................................................................... 3
1.4: Frustrated Lewis Pairs: Reactivity ...................................................................................... 5
1.4.1: Nucleophilic Aromatic Substitution ......................................................................... 5
1.4.2: THF Ring-Opening ................................................................................................... 6
1.4.3: H2 Activation by FLPs .............................................................................................. 6
1.4.4: Addition to Alkenes and Alkynes ............................................................................. 8
1.4.5: Other Small Molecules ........................................................................................... 10
1.4.6: Catalytic Hydrogenations ....................................................................................... 10
Chapter 2: Stoichiometric and Catalytic P-P Bond Activation by Rh(I) Complexes ................... 12
2.1: Introduction ....................................................................................................................... 12
2.2: Experimental ..................................................................................................................... 15
2.2.1: General Considerations ........................................................................................... 15
2.2.2: Synthesis of Rh(Ph2PPPh2) Complexes .................................................................. 15
2.2.3: General Catalytic Procedures .................................................................................. 16
2.2.4: Characterization of Related Species ....................................................................... 17
2.2.5: Synthesis of RhNacNac(P5R5) Complexes ............................................................. 18
vii
2.2.6: X-Ray Data Collection, Reduction, Solution and Refinement ............................... 19
2.3: Results and Discussion ...................................................................................................... 23
2.3.1: Catalyst Selection and Initial Screening ................................................................. 23
2.3.2: Stoichiometric Reactions of Catalyst Precursors with Ph2PH and Ph2PPPh2 ......... 24
2.3.3: Catalytic Hydrogenation and Hydrosilylation Reactions of Ph2PPPh2 .................. 26
2.3.4: Mechanistic Insight into the Catalytic Activation of P-P Bonds ............................ 28
2.3.5: Heterodehydrocoupling of Silanes with Diphenylphosphine ................................. 31
2.3.6: Reactions of Catalyst Precursors with Additional Phosphines and Biphosphines . 31
2.3.7: Reactions with Cyclic Polyphosphines ................................................................... 33
2.4: Conclusions ................................................................................................................ 39
Chapter 3: Frustrated Lewis Pair Reactivity of Bulky Catena-Polyphosphines with B(C6F5)3 ... 40
3.1: Introduction ....................................................................................................................... 40
3.2: Experimental ..................................................................................................................... 43
3.2.1: General Considerations ........................................................................................... 43
3.2.2: Generation of a Phosphonium Borate Zwitterion through Nucleophilic
Aromatic Substitution ........................................................................................... 43
3.2.3: Synthesis of Alkenyl-bridged Phosphonium Borate Zwitterions via Activation
of Terminal Alkynes ............................................................................................. 44
3.2.4: Synthesis of a Phosphonium Borate Ion Pair via H2 Activation ............................. 45
3.2.5: Hydrogenation and Hydrosilylation of P5Ph5 ......................................................... 45
3.2.6: X-Ray Data Collection, Reduction, Solution and Refinement .............................. 48
3.3: Results and Discussion ...................................................................................................... 51
3.3.1: Stoichiometric Reactions of Bulky Polyphosphines with B(C6F5)3 ....................... 51
3.3.2: Nucleophilic Aromatic Substitution (NAS) Reactions ........................................... 51
3.3.3: Synthesis of Alkenyl-Bridged Phosphonium Borate Zwitterions by Terminal
Alkyne Activation ................................................................................................. 54
3.3.4: Activation of H-H and Si-H Bonds by Polyphosphosphine/Borane FLPs ............. 56
viii
3.3.5: Scope of Reactivity in Terms of Bulk at Silicon and Lewis Acidity of the
Borane ................................................................................................................... 62
3.4: Conclusions ....................................................................................................................... 63
Chapter 4: Frustrated Lewis Pairs: Synthesis and Reactivity of Covalently-Bound
Phosphinoboranes .................................................................................................................... 64
4.1: Introduction ....................................................................................................................... 64
4.2: Experimental ..................................................................................................................... 66
4.2.1: General Considerations ........................................................................................... 66
4.2.2: Synthesis of Phosphinoboranes R2PB(C6F5)2 ......................................................... 66
4.2.3: Synthesis of Secondary Phosphine Adducts of HB(C6F5)2 .................................... 68
4.2.4: Reactions of R2PB(C6F5)2 with 4-tert-butylpyridine .............................................. 71
4.2.5: Synthesis of Dimers (R2PBCl2)2 and ClB(C6F5)2 by Reaction of BCl3 with
R2PB(C6F5)2 .......................................................................................................... 72
4.2.6: X-Ray Data Collection, Reduction, Solution and Refinement ............................... 72
4.3: Results and Discussion ...................................................................................................... 77
4.3.1: Synthesis and Characterization of Phosphinoboranes R2PB(C6F5)2 ....................... 77
4.3.2: Reactions of Phosphinoboranes with H2 and Independent Synthesis of
Phosphine-Borane Adducts R2(H)PB(H)C6F5)2 ................................................... 82
4.3.3: Reactions of Phosphinoboranes with Lewis Acids and Lewis Bases ..................... 86
4.4: Conclusions ....................................................................................................................... 92
Chapter 5: Frustrated Lewis Pairs: Reactions of Pyridines and Other Nitrogen-Containing
Heterocycles with B(C6F5)3 ...................................................................................................... 93
5.1: Introduction ....................................................................................................................... 93
5.2: Experimental Section ........................................................................................................ 95
5.2.1: General Considerations ........................................................................................... 95
5.2.2: Synthesis of Pyridine-B(C6F5)3 adducts: ................................................................ 95
5.2.3 Synthesis of Pyridinium Borate Ion Pairs through H2 Activation by Pyridine-
Borane FLPs .......................................................................................................... 99
5.2.4: Synthesis of a Pyridinium Borate Zwitterion via THF Ring-Opening ................. 101
ix
5.2.5: Reactions of Diethyl-2,6-dimethyl-3,5-pyridinedicarboxylate with B(C6F5)3 ..... 101
5.2.6: Reactions of Bulky Substituted Quinolines with B(C6F5)3 ................................... 102
5.2.7: Metal-Free Catalytic Hydrogenations ................................................................... 103
5.2.8: Reaction of Aminopyridines with Fluoroarylboranes .......................................... 103
5.2.9: X-Ray Data Collection, Reduction, Solution and Refinement ............................. 107
5.3: Results and Discussion .................................................................................................... 114
5.3.1: Reactions of Alkyl or Aryl-Substituted Pyridines with B(C6F5)3 ......................... 114
5.3.2: Frustrated Lewis Pairs of Pyridines with B(C6F5)3 ............................................... 117
5.3.4: Small Molecule Activation by Pyridine-Borane FLPs ......................................... 119
5.3.5: FLPs of Bulky Pyridines with Other Lewis Acids ............................................... 122
5.3.6: Reaction of Diethyl-2,6-dimethyl-3,5-pyridinedicarboxylate with B(C6F5)3 ....... 123
5.3.7: Reactions of Substituted Quinolines with B(C6F5)3 ............................................. 125
5.3.8: Reactions of 2-Aminopyridines with Fluoroarylboranes ...................................... 129
5.4: Conclusions ..................................................................................................................... 134
Chapter 6: Summary and Conclusions ........................................................................................ 136
Appendix A: Frustrated Lewis Pairs Derived From P(OR)nR3-n and B(C6F5)3 .......................... 138
A.1: Introduction .................................................................................................................... 138
A.2: Experimental .................................................................................................................. 139
A.2.1: General Considerations ........................................................................................ 139
A.2.2: Lewis Acid-Base Adducts of Phosphites with B(C6F5)3 ...................................... 139
A.2.3: Synthesis of Phosphinites tBu2POR ..................................................................... 140
A.2.4: Generation of a Phosphine-Oxide Adduct of B(C6F5)3 ........................................ 141
A.2.5: Generation of Phosphonium Borate Ion Pairs by H2 Activation ......................... 141
A.2.6: X-Ray Data Collection, Reduction, Solution and Refinement ............................ 142
A.3: Results and Discussion ................................................................................................... 144
A.3.1: Reactions of Phosphites with B(C6F5)3 ................................................................ 144
x
A.3.2: Reactions of RnP(OtBu)3-n with B(C6F5)3 ............................................................ 147
A.3.3: FLP Reactions of tBu2POAr with B(C6F5)3 ......................................................... 149
A.4: Conclusions .................................................................................................................... 151
References ................................................................................................................................... 152
xi
List of Tables
Table 2.1: Selected crystallographic data for compounds 2-3, 2-4 and 2-11 ............................... 21
Table 2.2: Selected crystallographic data for compounds 2-12, 2-13 and 2-14 ........................... 22
Table 2.3: Silylation of Ph2PPPh2 using 10 mol% 2-1 at 100°C in toluene ................................. 28
Table 2.4: Results of heterodehydrocoupling reactions of Ph2PH with silanes using 5 mol% 2-1
at 50°C in toluene.......................................................................................................................... 29
Table 3.1: Selected crystallographic data for compounds 3-1, 3-2, 3-3 and 3-9 .......................... 50
Table 3.2: Selected NMR spectroscopic data and yields for adducts 3-6 – 3-11 ......................... 60
Table 4.1: Selected crystallographic data for compounds 4-1, 4-2 and 4-3 ................................. 73
Table 4.2: Selected crystallographic data for compounds 4-4, 4-6 and 4-7 ................................. 74
Table 4.3: Selected crystallographic data for compounds 4-8, 4-12 and 4-13 ............................. 75
Table 4.4: Selected crystallographic data for compound 4-14 ..................................................... 76
Table 4.5: Selected NMR spectroscopic data for compounds 4-1 to 4-5 ..................................... 80
Table 4.6: Selected metrical parameters for monomeric phosphinoboranes ................................ 82
Table 4.7: Hydrogenation of phosphinoboranes 4-3 and 4-4 ....................................................... 83
Table 4.8: Selected NMR spectroscopic data for phosphine-borane adducts 4-6 to 4-10 ............ 84
Table 5.1: Selected crystallographic data for compounds 5-1, 5-2 and 5-3 ............................... 108
Table 5.2: Selected crystallographic data for compounds 5-4, 5-5 and 5-6 ............................... 109
Table 5.3: Selected crystallographic data for compounds 5-7, 5-8 and 5-9 ............................... 110
Table 5.4: Selected crystallographic data for compounds 5-13, 5-14 and 5-19 ......................... 111
xii
Table 5.5: Selected crystallographic data for compounds 5-25, 5-26 and 5-27 ......................... 112
Table 5.6: Selected crystallographic data for compounds 5-29 and 5-30 ................................... 113
Table 5.7: Selected NMR spectroscopic and X-ray crystallographic data obtained for pyridine-
borane adducts 5-1 to 5-8 ............................................................................................................ 116
Table 5.8: Catalytic hydrogenation of quinolines (reactions were conducted under 4 atm H2 in
toluene in a sealed Teflon-capped Schlenk bomb) ..................................................................... 128
Table A.1: Selected crystallographic data for compounds A-1, A-6 and A-7 ............................ 143
Table A.2: Comparison of Me3P and (MeO)3P and their B(C6F5)3 adducts ............................... 145
Table A.3: Cone angles and basicities of phosphorus bases and reactivity with B(C6F5)3 ........ 146
xiii
List of Figures
Figure 1.1: Wilkinson’s Catalyst (left) and Grubbs’ Catalyst (right) ............................................. 1
Figure 1.2: Generation of the catalytically active species from Wilkinson’s catalyst (R=Ph) ....... 1
Figure 1.3: Formation of metallacyclobutane intermediate in olefin metathesis ............................ 2
Figure 1.4: Examples of phosphorus-containing polymers: polyphosphazenes (left) and
polyphosphinoboranes (right). ........................................................................................................ 2
Figure 1.5: Catalytic dehydrocoupling of PhPH2 to generate P5Ph5 ............................................... 2
Figure 1.6: Formation of a Lewis acid-base adduct (B(C6F5)3 – Lewis acid, PMe3 – Lewis base) 3
Figure 1.7: Rankings of pyridine basicity (top, pKa’s in nitrobenzene) and nucleophilicity
(bottom) ........................................................................................................................................... 5
Figure 1.8: Resonance forms of B(C6F5)3 and resulting para-NAS by a phosphine ...................... 6
Figure 1.9: Para-NAS by a bulky phosphine on trityl cation ([B(C6F5)4]- is the counterion) ........ 6
Figure 1.10: THF ring-opening by a phosphine-borane pair .......................................................... 6
Figure 1.11: Reversible H2 activation by a linked phosphine-borane ............................................ 7
Figure 1.12: H2 activation mechanisms: Top: Papai and co-workers; Bottom: Grimme and co-
workers ............................................................................................................................................ 8
Figure 1.13: Alkene activation by a frustrated Lewis pair .............................................................. 9
Figure 1.14: Terminal alkyne activation by frustrated Lewis pairs (left-deprotonation, right-
addition) .......................................................................................................................................... 9
Figure 1.15: Addition of a linked phosphine-borane to norbornene ............................................... 9
Figure 1.16: Activation of N2O and CO2 by FLPs ........................................................................ 10
xiv
Figure 1.17: Imine hydrogenation catalyzed by B(C6F5)3 ............................................................ 11
Figure 2.1: Scheme for homonuclear catalytic dehydrocoupling reactions .................................. 12
Figure 2.2: Mechanism for the catalytic dehydrocoupling of HPPh2 by Rh(I)-based catalysts.
Left: Cp*Rh (Brookhart and Bohm), right: (dippe)Rh (Tilley and Han) ..................................... 13
Figure 2.3: Rh(I) Catalyst Precursors 2-1: RhNacNac(COE)(N2) (Ar=2,6-di-iso-propylphenyl)
and 2-2: (n5-C9H7)Rh(COE)2 ........................................................................................................ 23
Figure 2.4: Equilibria involving mono-, bis- and tris-diphenylphosphine-rhodium complexes
(L=NacNac or n5-C9H7) ................................................................................................................ 24
Figure 2.5: POV-Ray depictions of 2-3 (left) and 2-4 (right) ....................................................... 25
Figure 2.6: Formation of Rh-phosphide dimer 2-3 ....................................................................... 26
Figure 2.7: Rh-catalyzed hydrosilylation of P2Ph4. ...................................................................... 28
Figure 2.8: Structure of 2-10 ......................................................................................................... 30
Figure 2.9: Proposed catalytic cycles for hydrogenation (left) and hydrosilylation (right) of
Ph2PPPh2 ....................................................................................................................................... 30
Figure 2.10: Proposed mechanism for dehydrocoupling of Ph2PH with silanes .......................... 31
Figure 2.11: POV-Ray depiction of 2-11 (left) and 2-12 (right) .................................................. 33
Figure 2.12: Cyclic Polyphosphines: A) P4R4 (R=Cy, tBu) B) P5R5 (R=Ph, Et) ........................ 34
Figure 2.13: 31
P NMR Spectrum (top, left) and POV-Ray depiction of 2-13-0.5 C6H14 ............. 35
Figure 2.14: 31
P NMR Resonances (top) and POV-Ray depiction (bottom) of 2-14 ................... 38
Figure 3.1: Nucleophilic aromatic substitution (NAS) at the para-position of a C6F5 ring by
phosphines on B(C6F5)3 ................................................................................................................ 40
Figure 3.2: Activation of alkenes (left) and alkynes (right) by a frustrated Lewis pair: PR3 +
B(C6F5)3 ........................................................................................................................................ 41
xv
Figure 3.3: Multinuclear NMR Spectra for 3-1 in CD2Cl2: A: 31
P (resonance for cationic
phosphorus centre), B: 31
P (other 4 phosphorus resonances), C: 19
F, D: 11
B ............................... 52
Figure 3.4: POV-Ray depiction of 3-1-C6H6 ................................................................................ 53
Figure 3.5: Formation of alkyne addition products 3-2 and 3-3 ................................................... 54
Figure 3.6: 31
P{1H} NMR spectrum (left) and POV-Ray depiction of 3-2 (right) ....................... 55
Figure 3.7: POV-Ray depiction of 3-3 .......................................................................................... 56
Figure 3.8: Formation of the phosphonium borate ion pair 3-4 .................................................... 56
Figure 3.9: Multinuclear NMR spectra for 3-4 in CD2Cl2: 1H – showing PH and BH peaks (top
left), 11
B (top right), 19
F (bottom left), 31
P[1H} (bottom right) ..................................................... 57
Figure 3.10: Formation of 3-5 from the reaction of P5Ph5 with B(C6F5)3 and H2. ....................... 58
Figure 3.11: Nucleophilic attack by PMe3 on the cationic phosphorus centre of P4Cy4Me+ ....... 58
Figure 3.12: Hydrogen activation by the frustrated Lewis pair P5Ph5/B(C6F5)3 and subsequent
rearrangement to form (Ph)H2P-B(C6F5)3 ..................................................................................... 59
Figure 3.13: Mechanism for B(C6F5)3 catalyzed hydrosilylation of imines ................................. 59
Figure 3.14: Lewis acid promoted hydrosilylation of P5Ph5. ........................................................ 60
Figure 3.15: POV-Ray depiction of 3-9 ........................................................................................ 61
Figure 3.16: Synthesis of 3-11 ...................................................................................................... 62
Figure 4.1: Dehydrogenation of ammonia borane by a nickel(0)carbene catalyst ....................... 64
Figure 4.2: Potential Reactivity of R2PB(C6F5)2: dimerization (top), H2 activation (bottom) ..... 65
Figure 4.3: Synthesis of 4-1 to 4-5 (LiCl is removed upon workup) ............................................ 77
Figure 4.4: POV-Ray depictions of phosphinoborane dimers 4-1 (left) and 4-2–CH2Cl2 (right).
....................................................................................................................................................... 78
xvi
Figure 4.5: Multinuclear NMR spectra for 4-3. A: 1H, B:
11B, C:
19F, D:
31P{
1H} ..................... 79
Figure 4.6: POV-Ray depictions of phosphinoboranes 4-3 (left) and 4-4 (right) ......................... 81
Figure 4.7: Resonance forms of phosphinoboranes 4-3 to 4-5 ..................................................... 82
Figure 4.8: Synthesis of 4-8 to 4-10 through H2 activation (left) or Lewis acid-base adduct
formation (right) ............................................................................................................................ 83
Figure 4.9: POV-Ray depictions of 4-6 and 4-7 ........................................................................... 85
Figure 4.10: POV-Ray depiction of 4-8 ........................................................................................ 85
Figure 4.11: Newman projection along the B-P bond of 4-6 (R-Et) and 4-7 (R=Ph) (left); and 4-8
(right) as determined by X-ray crystallography ............................................................................ 86
Figure 4.12: Formation of adducts 4-11 and 4-12 ........................................................................ 87
Figure 4.13: POV-Ray depiction of 4-12-C7H8 ............................................................................ 87
Figure 4.14: POV-Ray depiction of 4-13 ...................................................................................... 89
Figure 4.15: Proposed formation of 4-13 from 4-11 (L=4-(tBu)C5H4N) ...................................... 90
Figure 4.16: Formation of dimers 4-14 and 4-15 .......................................................................... 90
Figure 4.17: POV-Ray depiction of 4-14 ...................................................................................... 91
Figure 5.1: Brown’s observation of a surprising lack of reactivity between 2,6-lutidine and BMe3
....................................................................................................................................................... 93
Figure 5.2: Amine-based reducing agents: Hantzsch’s Ester (A), acridan (B) and
tetrahydroquinoline (C) ................................................................................................................. 94
Figure 5.3: Proposed scheme for catalytic transfer hydrogenation through Hantzsch’s ester. ..... 94
Figure 5.4: Synthesis of Lewis acid-base adducts 5-2 – 5-7; R=Me (5-2), Et (5-3), R=N(H)(2-
C5H4N) (5-4), Ph (5-5), 2-C5H4N (5-6) ...................................................................................... 114
xvii
Figure 5.5: POV-Ray Depictions of 5-3 (left) and 5-4 (right) .................................................... 115
Figure 5.6: POV-Ray Depictions of 5-5-0.5 C7H8 (left) and 5-7 (right) .................................... 115
Figure 5.7: Equilibrium observed in solution between FLP and Lewis acid-base adduct 5-8 ... 117
Figure 5.8: POV-Ray depiction of 5-8 ........................................................................................ 118
Figure 5.9: Synthesis of ion pairs 5-9 (R=R1=Me), 5-10 (R=
tBu, R
1=H) and 5-11 (R=R
1=Ph) 119
Figure 5.10: Multinuclear NMR spectra for 5-9 in CD2Cl2: A: 1H, B:
11B and C:
19F. .............. 120
Figure 5.11: POV-Ray depiction of 5-9 ...................................................................................... 120
Figure 5.12: H2 loss from ion pairs 5-9 – 5-11 ........................................................................... 121
Figure 5.13: Formation of zwitterion 5-13 by THF ring-opening .............................................. 121
Figure 5.14: POV-Ray depiction of 5-13 .................................................................................... 122
Figure 5.15: POV-Ray depiction of 5-14 .................................................................................... 124
Figure 5.16: Partial hydrogenation of adduct 5-14 ..................................................................... 125
Figure 5.17: Generation of 5-15 and 5-16 by hydride abstraction from Hantzsch’s Ester. ........ 125
Figure 5.18: Equilibria involving the formation of adducts 5-17 and 5-18 ................................ 126
Figure 5.19: POV-Ray depiction 5-19 (one of two crystallographically independent molecules)
..................................................................................................................................................... 127
Figure 5.20: Formation of zwitterion 5-25 ................................................................................. 129
Figure 5.21: POV-Ray depictions of 5-25 and 5-26 (one of two crystallographically independent
molecules) ................................................................................................................................... 130
Figure 5.22: POV-Ray depiction of 5-27 .................................................................................... 130
Figure 5.23: Synthesis of linked pyridine-borane 5-28 .............................................................. 131
xviii
Figure 5.24: Multinuclear NMR spectra for 5-28 in CDCl3. A: 1H, B:
11B, C:
19F ................... 132
Figure 5.25: POV-Ray depictions of 5-29 and 5-30 ................................................................... 133
Figure 5.26: Formation of linked pyridine-borane 5-31 ............................................................. 134
Figure 5.27: Formation of Lewis acid-base adducts 5-32, 5-33 and 5-34 .................................. 134
Figure A.1: Formation of phosphite-borane adducts A-1 and A-2 ............................................. 144
Figure A.2: POV-Ray depiction of A-1 ...................................................................................... 144
Figure A.3: Phosphite-based FLPs and attempted hydrogen activation ..................................... 145
Figure A.4: POV-Ray depiction of A-6 ...................................................................................... 148
Figure A.5: 1H NMR spectrum showing formation of iso-butene, along with adduct A-9 ........ 148
Figure A.6: Proposed mechanism for the formation of A-9 and iso-butene from reaction of A-3
with B(C6F5)3 .............................................................................................................................. 149
Figure A.7: FLP reactivity of phosphinites A-5 and A-6 with B(C6F5)3 .................................... 149
Figure A.8: Multinuclear NMR spectra for A-7 in CD2Cl2. A: 1H, B:
11B, C:
19F and D:
31P ... 150
Figure A.9: POV-Ray depiction of A-8 ...................................................................................... 150
xix
List of Abbreviations, Nomenclature and Symbols
Å Angstrom
atm atmospheres
br m broad multiplet
Bu n-butyl (C4H9)
calcd calculated
CCD charge coupled device
COD cyclooctadiene (C8H12)
COE cis-cyclooctene (C8H14)
Cp cyclopentadienyl (C5H5)
Cp* pentamethylcyclopentadienyl (C5(CH3)5)
Cy cyclohexyl (C6H11)
C degrees Celsius
Dcalc calculated density
d doublet
dippe 1,2-bis-(di-iso-propylphosphino)ethane
eq equivalents
Et ethyl (C2H5)
FLP frustrated Lewis pair
xx
g grams
GOF goodness of fit
HOMO Highest Occupied Molecular Orbital
hr hour
Hz Hertz
iPr iso-propyl (CH(CH3)2)
IR infrared
J coupling constant
K degrees Kelvin
kcal kilocalories
kJ kilojoules
L liter
LUMO Lowest Unoccupied Molecular Orbital
m multiplet
M mol L-1
m meta
Me methyl (CH3)
Mes mesityl (2,4,6-(CH3)3C6H2)
mg milligram
MHz megahertz
xxi
min minute
mL milliliter
mmol millimole
NacNac HC{CN(2,6-iPr2C6H3)}2
NAS nucleophilic aromatic substitution
NMR nuclear magnetic resonance
o ortho
ORTEP Oak Ridge thermal ellipsoid plot
p para
Ph phenyl (C6H5)
POV-Ray Persistence of Vision Raytracer
ppm parts per million
R residual
Rw weighted residual
RT room temperature
s singlet, seconds
sept. septet
t triplet
tBu tert-butyl (C(CH3)3)
THF tetrahydrofuran (C4H8O)
xxii
TMS trimethylsilyl (Si(CH3)3)
wt % weight percent
μmol micromole
1
Chapter 1: Introduction
1.1: Phosphines in Coordination and Materials Chemistry
Over the past century phosphines have become ubiquitous as ligands in inorganic
chemistry, owing to their tunability and their strong and generally predictable coordination
mode.5 Some of the most commercially important transition metal catalysts, including
Wilkinson’s Catalyst and Grubbs’ Catalyst, incorporate phosphines as ligands (Figure 1.1).5
Figure 1.1: Wilkinson’s Catalyst (left) and Grubbs’ Catalyst (right)
The tunability of phosphines is key to the mechanism of activation for both of these
species. One of the most important mechanisms for the hydrogenation of olefins utilizing
Wilkinson’s catalyst involves initial oxidative addition of dihydrogen (H2). The oxidative
addition of H2 to the rhodium centre generates an octahedral species with a phosphine ligand
trans to a hydride. As a result of the strong trans-effect of the hydride ligand, this phosphine can
dissociate from the metal centre, allowing for the coordination of olefin (Figure 1.2) and
subsequent hydrogenation.6 The PEt3 analogue of Wilkinson’s catalyst was found to be inactive
for the hydrogenation catalysis but the octahedral Rh(III) dihydride species could be isolated.5
The smaller size and stronger electron-donating ability of PEt3 compared to PPh3 makes
dissociation of the phosphine much less favourable. These results illustrate the advantage of
being able to easily tune the steric and electronic properties of phosphine ligands.
Figure 1.2: Generation of the catalytically active species from Wilkinson’s catalyst (R=Ph)
2
Phosphine dissociation is also a vital step for the activation of Grubbs’ Catalyst. The size
and strong trans-effect of the PCy3 groups results in facile loss of one phosphine substituent
upon coordination of the olefin, generating the metallacyclobutane intermediate (Figure 1.3).7
Again, the easy tunability of phosphines allowed for selection of a ligand of appropriate size and
electron-donating ability.
Figure 1.3: Formation of metallacyclobutane intermediate in olefin metathesis.7
Recently there has been much work on the incorporation of phosphorus into polymeric
materials, due to their ability to render materials resistant to fire, their optical electronics
properties and their ability to provide Lewis basic sites for the potential coordination of transition
metals (Figure 1.4).8,9
One approach to synthesizing new materials containing phosphorus is the
dehydrocoupling of P-H bonds, either with other P-H bonds (homodehydrocoupling), or with
other E-H bonds (heterodehydrocoupling).10
Figure 1.4: Examples of phosphorus-containing polymers: polyphosphazenes11
(left) and
polyphosphinoboranes10
(right).
The homodehydrocoupling reaction has been used to create P-P bound monomers and
macrocycles:12-14
however, long reaction times and harsh conditions are required to attain
reasonable yields (an example is shown in Figure 1.5).
Figure 1.5: Catalytic dehydrocoupling of PhPH2 to generate P5Ph512
3
1.2: P-P Bond Activation
The hydrogenation of P-P bonds, the reverse reaction of P-P dehydrocoupling, is an
important process, as current industrial synthesis of organophosphines involves the reaction of
white phosphorus, P4, with chlorine gas to form PCl3, which is subsequently reacted further to
yield organophosphines.15
Circumventing the use of chlorine gas is highly desirable on health,
environmental and economic grounds. 16-26
There has been much work done on the activation of white phosphorus, P4, involving
stoichiometric use of carbenes,25,26
or transition metals18,20
with the primary goal of fragmenting
P4 cleanly into P1 units thus avoiding the use of chlorine gas in the synthesis of
organophosphines.
While other P-P bound species, such as P5Ph5, P4Cy4 and P2Ph4, are well known in the
literature,27
little work has been dedicated to the possibility of controlled fragmentation of these
molecules to P1 units. Such pathways may allow for the synthesis of previously inaccessible
molecules while also providing the opportunity to discover new modes of reactivity which may
be useful in reactions of P4.
1.3: Lewis Acid/Lewis Base Chemistry
In addition to their ability to coordinate to transition metals, phosphines (and other
molecules with pairs of available valence electrons) have long been known to coordinate to main
group electron acceptors. The resulting compounds, referred to as Lewis adducts or donor-
acceptor adducts, were first explained by G.N. Lewis in 1923.28
Lewis defined acids as
compounds which can accept a pair of electrons and bases as compounds which can donate
electrons. Thus, when these compounds are added to one another, the Lewis base can donate a
pair of electrons to the Lewis acid, forming a new dative bond between the molecules, as shown
in Figure 1.6.
Figure 1.6: Formation of a Lewis acid-base adduct (B(C6F5)3 – Lewis acid, PMe3 – Lewis base)29
4
Owing to its strong Lewis acidity and relatively strong B-C bonds, B(C6F5)3 is a very
commonly used species for Lewis acid catalyzed organic transformations and is used as a co-
catalyst for α-olefin polymerization.30,31
Halogenated boranes such as BF3 and BCl3 exhibit
comparable Lewis acidity to B(C6F5)3, however, they suffer from instability due to redistribution
of the halides and loss of HX when reacted with protic acids, while trialkyl boranes are also
unstable and are significantly less Lewis acidic than B(C6F5)3.30
B(C6F5)3 is also known to form
adducts with a wide variety of Lewis bases including phosphines, amines, imines, pyridines,
nitriles, aldehydes, ketones and esters. 30,31
Recently, the Stephan group has focused on systems which have been termed “frustrated
Lewis pairs” (FLPs).3,4
A FLP is defined as a Lewis acid and base pair which does not form a
Lewis acid-base adduct due to steric conflict. While this chemistry has only recently garnered
much attention in the literature, the first reported observation of a frustrated Lewis pair dates
back to 1942 when H.C. Brown and co-workers noted that 2,6-lutidine failed to form a Lewis
adduct with BMe3.32
In addition to this observation, Brown cited a number of results in reactions
of Lewis acids and Lewis bases which proceeded not to the electronically favoured product, but
to the sterically favoured product. Many of these results were compiled in an article titled
“Chemical Effects of Steric Strains.”33
For example, Brown noted that while the basicity of a
series pyridines increased from pyridine to 2-methylpyridine to 2-tert-butylpyridine to 2,6-
lutidine, the series was nearly reversed in terms of nucleophilicity (pyridine>2-
methylpyridine>2,6-lutidine>2-tert-butylpyridine). This trend was attributed to the steric effects
of the substituents of the pyridines (Figure 1.7).
5
Figure 1.7: Rankings of pyridine basicity (top, pKa’s in nitrobenzene) and nucleophilicity
(bottom) 33
While Brown did observe a lack of interaction between Lewis acid and Lewis base in
solution, the potential reactivity of these systems was not explored. However, these FLPs seen
by Brown include weak Lewis acids that may not have been capable of the small molecule
activation reactions recently discovered with FLPs, pioneered by the Stephan group.
1.4: Frustrated Lewis Pairs: Reactivity
1.4.1: Nucleophilic Aromatic Substitution
Among the first alternate reactivity attributed to FLPs is the para-nucleophilic aromatic
substitution (NAS) by large Lewis bases on aromatic rings of the Lewis acids B(C6F5)3 and
[CPh3]+.2,34-38
This reaction occurs due to steric bulk preventing the reaction of the Lewis base
with the most Lewis acidic centre (boron for B(C6F5)3 and the central cationic carbon for
[CPh3]+). For the smallest phosphines, reaction with B(C6F5)3 results in adduct formation, while
no reaction takes place for the largest phosphines. The reaction of some intermediate-sized
phosphines (PCy3 for example) with B(C6F5)3 occurs at an alternate Lewis acidic centre in the
molecule: the para-carbon.36
While these phosphines are too bulky to interact directly with the
boron centre, they are still nucleophilic enough to attack the molecule at a more accessible site.
Resonance forms for these electrophiles show that both ortho- and para-carbons of the aromatic
rings are also electron-deficient. As the para-carbons are less sterically hindered, substitution
occurs at this site (Figure 1.8).
6
Figure 1.8: Resonance forms of B(C6F5)3 and resulting para-NAS by a phosphine
In the case of B(C6F5)3, fluoride transfer to boron is always observed, while for trityl
cation both the intermediate cyclohexadienyl compound (Figure 1.9 A) and the final para-NAS
(Figure 1.9 B) product can be isolated.35
Figure 1.9: Para-NAS by a bulky phosphine on trityl cation ([B(C6F5)4]- is the counterion)
1.4.2: THF Ring-Opening
While the THF adduct of B(C6F5)3 is well-known, the bound THF molecule is generally
known to be displaced by a stronger base.30
When the stronger base is too sterically encumbered
to form a traditional adduct with B(C6F5)3, nucleophilic ring-opening of THF may occur.39,40
Coordination of the oxygen of THF to B(C6F5)3 renders the α–carbons electrophilic and
susceptible to nucleophilic attack by the Lewis base (Figure 1.10).
Figure 1.10: THF ring-opening by a phosphine-borane pair
1.4.3: H2 Activation by FLPs
FLP chemistry took a giant leap when a report by the Stephan group showed that a linked
phosphine-borane system could reversibly activate H2 (Figure 1.11),2 a reaction which is
common for transition metals but virtually unprecedented for a main group system.41,42
The
7
reaction proceeds via heterolytic cleavage of the dihydrogen molecule, with H+ going to the
Lewis basic phosphorus centre and H- to the Lewis acidic boron centre.
Figure 1.11: Reversible H2 activation by a linked phosphine-borane
Subsequent studies showed that simple bimolecular phosphine-borane Lewis pairs such
as tBu3P or Mes3P and B(C6F5)3 were capable of activating H2.
43 The Stephan group and others
have since shown that pairs of carbenes,44,45
imines46
or amines46,47
with B(C6F5)3 are capable of
activating H2. While several other Lewis acids have been utilized, the scope for hydrogen
activation seems to be limited to B(C6F5)3 and related fluoroaryl boranes3,4,48-50
(while BPh3 was
shown to activate H2 with tBu3P, poor yields were obtained).
38 Interestingly, neither Lewis acid
nor Lewis base alone were found to react with H2 in solution. Several other related linked
systems have also emerged in literature, containing a bulky Lewis basic centre and a 3-
coordinate fluoroaryl boron centre which are unable to coordinate to one another due to steric
factors.51,52
A detailed kinetic study of the hydrogen activation reaction has proven to be very
difficult due to a lack of control over the concentration of H2 in solution. Studies of the reverse
reaction (hydrogen loss) have been hampered by the more favourable nature of the H2 activation
reaction, leading to inconsistent and unclear results.53
The activation of H2 by FLPs has undergone several computational studies. As
termolecular reactions are entropically very unlikely to occur, two of the molecules (Lewis acid,
Lewis base and H2) must interact prior to reaction with the third component. Interaction of
borane or phosphine with H2 has computational precedent in the literature: both BH354-57
and
phosphines58,59
have been calculated to interact independently with H2. While either of these
intermediates seem reasonable, recent computational studies on FLPs suggest that it is the Lewis
acid and base that come together prior to activation of H2.
8
In theoretical studies, Papai and co-workers suggested that the borane and phosphine
form an “encounter complex” where the 2 molecules are held together by secondary interactions
(for example weak hydrogen bonding between alkyl groups and fluorine atoms).60-62
H2 is then
proposed to enter the space between the phosphorus and boron centres and is subsequently
cleaved by interaction of the phosphorus lone pair with the LUMO of H2 and of the vacant p-
orbital at boron with the HOMO of H2 (Figure 1.12, top). Similar transition states were proposed
for other related systems involving a carbene/B(C6F5)3 pair45
and amine-borane FLPs.47,52
Grimme and co-workers have since proposed that the proximity of the Lewis acid and
Lewis base creates an electric field which cleaves H2 in a nonlinear fashion, with B-H bond
formation slightly preceding P-H bond formation.63
This mechanism suggests that a molecular
orbital argument in terms of the FLP is not necessary to explain the H2 activation reaction, but
the mere presence of Lewis acid and base creates an electric field of sufficient strength to cleave
the H-H bond (Figure 1.12, bottom). Previous theoretical work has also suggested that H2
activation by an electric field is indeed possible.64-66
Figure 1.12: H2 activation mechanisms: Top: Papai and co-workers;60-62
Bottom: Grimme and
co-workers63
1.4.4: Addition to Alkenes and Alkynes
Concurrent with the discovery of the H2 activation reaction, work in the Stephan lab
showed that FLPs could also add to alkenes, creating alkyl-bridged phosphonium borate
zwitterions (Figure 1.13).67
Again neither Lewis acid nor Lewis base were found to react with
olefins independently.
9
Figure 1.13: Alkene activation by a frustrated Lewis pair
Terminal alkynes could also be activated by FLPs, forming the addition product and/or an
ion pair in ratios dependant on the strength of the base used (Figure 1.14).68
Stronger bases tend
to deprotonate the alkyne rather than adding to it. In cases of intermediate base strength,
mixtures are observed.
Figure 1.14: Terminal alkyne activation by frustrated Lewis pairs (left-deprotonation, right-
addition)
A recent experimental and computational study by Grimme and co-workers has
suggested that the activation of alkenes by the linked phosphine-borane FLP
Mes2PCH2CH2B(C6F5)2 proceeds via an essentially concerted mechanism, with B-C bond
formation slightly preceding P-C bond formation.69
This concerted pathway is supported
experimentally by the fact that addition of this FLP to norbornene produces only one new
species, resulting from endo-cis-addition across the double bond (Figure 1.15).
Figure 1.15: Addition of a linked phosphine-borane to norbornene
If B-C bond formation occurred significantly prior to P-C bond formation a mixture of
products would be expected from this reaction as a result of a series of rapid Wagner-Meerwein
rearrangements and hydride shifts of the norbornyl cation intermediate. Such products are not
10
observed in the reaction, thus the addition of the phosphine is rapid enough to preclude this
pathway.
1.4.5: Other Small Molecules
CO270
and N2O71
can also be activated by FLPs (Figure 1.16). In the case of CO2 this has
been extended to afford the sub-stoichiometric conversion of CO2 to methanol.72
This work
suggests that catalytic transformations of these materials may be possible. A catalytic cycle
converting these problematic greenhouse gases to useful organic precursors would be of
tremendous importance.
Figure 1.16: Activation of N2O and CO2 by FLPs
1.4.6: Catalytic Hydrogenations
Following the discovery of the hydrogen activation reaction, FLPs were studied for
catalytic hydrogenation activity, traditionally an area dominated by transition metals.42,73
The
common use of stoichiometric borohydride reducing agents suggested this should be possible.42
Addition of hydrogen by a FLP creates a borohydride reagent in situ, which could then transfer
hydride to the substrate. Indeed, the linked system 4-(Mes2PH)-C6F4-B(H)(C6F5)2 was found to
be effective for the hydrogenation of imines.73
Subsequent work showed that B(C6F5)3 alone
was effective as a catalyst for this reaction (the imine itself acts as the Lewis base in the FLP
activation of H2, Figure 1.17).46
In this manner, protected nitriles could be hydrogenated and
aziridines could be ring-opened.46
Other systems have since been shown to effect the catalytic
hydrogenations of imines,52,74
enamines74,75
and silyl enol ethers.76
11
Figure 1.17: Imine hydrogenation catalyzed by B(C6F5)346
These results demonstrate that certain combinations of main group elements can be
exploited for chemistry that has been traditionally confined to transition metals and furthermore,
can be used for completely novel reactivity. Studies of the relatively unexplored P-P bond
activation reaction utilizing both transition metals and FLPs will be presented in this thesis. A
further exploration of the FLP concept, including the synthesis and use of novel FLPs for small
molecule activation reaction, will also be presented.
12
Chapter 2: Stoichiometric and Catalytic P-P Bond Activation by Rh(I) Complexes
2.1: Introduction
There has been much recent interest in the field of “inorganometallics,” which involves
the melding of the chemistry of main group elements, other than carbon and hydrogen, with
transition metals.77
Among the most active fields is the study of inorganic polymers, which
includes polymers containing main group and/or transition metals.78,79
Many inorganic
oligomers and polymers can be synthesized by using a transition-metal catalyzed
dehydrocoupling reaction, which involves the formation of a E-E bond through loss of
dihydrogen (H2) from 2 E-H bonds (E=B, Si, Ge, Sn, P, Sb),10
shown in Figure 2.1.
Heterodehydrocoupling reactions can also be achieved by using 2 different element-hydrogen
bonds. This synthetic methodology has been used perhaps most elegantly by Manners and co-
workers in the heterodehydrocoupling of boranes with amines or phosphines to form polymers
and oligomers containing N-B or P-B bonds.80-97
This reactivity has also been used for the
homocoupling of B-H,98,99
P-H,12-14,100-102
Ge-H,103,104
Si-H,105-109
Sn-H110,111
and Sb-H112
bonds.
The homocoupling of two P-H bonds to form a new P-P bond is of utmost interest given
the high natural abundance of phosphorus and its common use in coordination chemistry.113
While P-P bonds can also be formed by reduction of phosphine chlorides or through salt
metathesis,27
these methods are not atom-economical and generate large amounts of salt by-
products.
Previous work in the Stephan group has shown that the catalytic dehydrocoupling of
primary phosphines can be accomplished using a zirconocene trihydride species to form P5R5
rings.12
Work by the groups of Tilley13
and Brookhart14
has shown that the dehydrocoupling of
secondary phosphines can be accomplished using a Rh(I) catalyst precursor, forming the
biphosphines R2PPR2.
Figure 2.1: Scheme for homonuclear catalytic dehydrocoupling reactions
13
While Brookhart’s proposed catalytic cycle involved dual P-H bond activation resulting
in a Rh(V) intermediate (Figure 2.2, left), Tilley’s catalytic cycle invoked the Rh(III) oxidation
state (Figure 2.2, right). Logically, as both cycles require oxidative addition of at least one P-H
bond followed by elimination of P2R4, these reactions should work best with bulky, electron-rich
catalyst precursors. The steric bulk should hinder aggregation often seen with PR2 fragments
and encourage dissociation of the product, while the increased electron density facilitates
oxidative addition at the metal by making the higher oxidation state more readily accessible.
Figure 2.2: Mechanism for the catalytic dehydrocoupling of HPPh2 by Rh(I)-based catalysts.
Left: Cp*Rh (Brookhart and Bohm)14
, right: (dippe)Rh (Tilley and Han)13
Rhodium(I) is an obvious candidate for this type of reaction as oxidative addition of P-H
bonds and the +3 oxidation state are well-known. In addition, the +5 oxidation state has also
been previously invoked for Rh-based catalysts (for example see Figure 2.2, left). Given the
importance of new phosphorus-containing materials, we sought to gain some insight into the
mechanism of the dehydrocoupling reaction and improve upon current catalyst options. To date,
Rh(I) systems for catalytic dehydrocoupling reactions have shown high activities for only a few
substrates and only under forcing conditions (extended reaction times and high
temperatures).13,14
14
15
2.2: Experimental
2.2.1: General Considerations
All preparations were done under an atmosphere of dry, O2-free N2 employing both Schlenk line
techniques and an Innovative Technologies or Vacuum Atmospheres inert atmosphere glove box.
Solvents (pentanes, hexanes, toluene, and methylene chloride) were purified employing a
Grubbs’ type column systems manufactured by Innovative Technology and stored over
molecular sieves (4 Å). Molecular sieves (4 Å) were purchased from Aldrich Chemical Company
and dried at 140 ºC under vacuum for 24 hours prior to use. Deuterated solvents were dried over
Na/benzophenone (C6D6, C7D8). All common organic reagents were purified by conventional
methods unless otherwise noted. 1H,
13C,
29Si, and
31P nuclear magnetic resonance (NMR)
spectroscopy spectra were recorded on a Bruker Avance-300 spectrometer at 300K unless
otherwise noted. 1H,
13C and
29Si NMR spectra are referenced to SiMe4 using the residual solvent
peak impurity of the given solvent. 31
P NMR experiments were referenced to 85% H3PO4.
Chemical shifts are reported in ppm and coupling constants in Hz as absolute values.
Combustion analyses were performed in house employing a Perkin Elmer CHN Analyzer.
Et2PH, Cy2PH, Ph2PH and Ph2PPPh2 were purchased from Aldrich Chemical Company and used
as received. Silanes were purchased from Strem Chemicals and used as received. tBu2PLi, and
Ph2PLi were prepared by treating the corresponding phosphine with 1 equivalent of tBuLi in
toluene and collecting the precipitate. Et2PPEt2 was prepared by reaction of the lithium
phosphide and the phosphine chloride. RhNacNac(COE)N2 (2-1)114
and Rh(n5-C9H7)(COE)2
(COE=cis-cyclooctene) (2-2),115
P5Ph5 and P5Et527
were prepared as previously reported.
2.2.2: Synthesis of Rh(Ph2PPPh2) Complexes
[Rh(η5-C9H7)( μ-PPh2)]2 (2-3) - Tetraphenyl biphosphine (Ph2PPPh2) (22 mg, 0.061 mmol) was
added to a solution of Rh(η5-C9H7)(COE)2 (50 mg, 0.11 mmol) in toluene (5 mL). The solution
was transferred to a Schlenk bomb and was heated at 80 ºC for two hours. The solvent was
removed in vacuo and the resulting solid was washed with hexanes (2 mL) leaving a dark solid
Yield: 32 mg (73%). Anal. Calcd. for RhP2C41H34 (%) C: 62.55, H: 4.25; found: C: 62.91, H:
4.71.
16
1H NMR (C6D6) δ: 5.19 (m, 4H), 5.49 (m, 2H), 6.88-7.08 (m, 28H).
31P NMR (C6D6) δ: 152.9
(t, 1JP-Rh=153 Hz).
13C{
1H} NMR (C6D6) δ: 78.7, 86.2, 111.4, 120.2, 122.3, 126.8-128.6 (m,
obscured by C6D6), 133.8.
RhNacNac(Ph2PPPh2) (2-4) - To a solution of RhNacNac(COE)N2 (50 mg, 0.076 mmol) in
toluene (5 mL) was added a solution of Ph2PPPh2 (30 mg, 0.081 mmol) in toluene (5 mL). The
mixture was allowed to stir overnight after which the solvent was removed in vacuo. The dark
red residue was taken up in 10 mL cold pentane (-35°C) and filtered through Celite. Cooling
overnight at -35ºC gave the product as a red solid. Yield: 35 mg (50%). X-ray quality crystals
were grown by slow evaporation from a pentane solution. Anal. Calcd. for RhP2N2C53H61 (%)
C: 71.21, H: 7.22, N: 3.13; found: C: 70.71, H: 7.22, N: 2.72.
1H NMR (C6D6) δ: 0.84 (d,
3JH-H=7 Hz, 6H), 1.20 (d,
3JH-H=7 Hz, 6H), 1.71 (s, 6H), 4.45 (4H,
sept, 3JH-H=7 Hz), 5.15 (1H, s), 6.73-7.19 (26 H, m).
31P NMR (C6D6) δ: -51.43 (d,
1JP-Rh=140
Hz). 13
C NMR (C6D6) δ: 24.0, 24.4, 28.6, 98.0, 124.0, 124.1, 127.5-128.6 (m, obscured by
C6D6), 128.9, 134.9 (app. t, J=8 Hz), 157.7, 159.6.
2.2.3: General Catalytic Procedures
Hydrogenation of Ph2PPPh2 – RhNacNac(COE)N2 (3.5 mg, 0.0055 mmol, 10 mol%) was
added to a solution of 20 mg Ph2PPPh2 (20 mg, 0.055 mmol) in toluene-d8 (0.75 mL). The
solution was transferred to a J. Young’s NMR Tube, subjected to 3 freeze-pump-thaw cycles and
placed under a H2 atmosphere at 77 K. The tube was allowed to thaw to room temperature and
was subsequently heated at 323 K for 12 hours.
Silylation of Ph2PPPh2 – A procedure similar to that of the hydrogenation was carried out,
however five equivalents of silane was added instead of 4 atm hydrogen (See Table 2.3 for
reaction conditions).
General procedure for the heterodehydrocoupling of silanes and phosphines – 20 mg of
phosphine in toluene-d8 (0.75 mL) was added to 5 equivalents of silane and RhNacNac(COE)N2
(1.7 mg, 0.0027 mmol, 5 mol% relative to phosphine). The reaction was heated to 50°C or
100°C and monitored by 31
P NMR spectroscopy (see Table 2.4 for reaction conditions).
17
NMR Data for silyl phosphines:
Data for Ph2PSi(H)Ph2 (2-5),116
Ph2PSiPh3 (2-8),117
and Ph2PSiEt3 (2-9)118
were as previously
reported.
Ph2MeSiPPh2 (2-6) - 1H NMR (C6D6) δ: 0.62 (3H, d,
3JH-P=3 Hz), 6.93-7.16, 7.32-7.53 (m, 20
H). 31
P NMR (C6D6) δ: -59.2 (s). 29
Si NMR (C6D6) δ: -8.2 (d, 1JSi-P=25 Hz).
PhMe2SiPPh2 (2-7) - 1H NMR (C6D6) δ: 0.35 (6H,
3JH-P=4 Hz), 6.95-7.17, 7.33-7.56 (m, 15 H).
31P NMR (C6D6) δ: -58.2 (s).
29Si NMR (C6D6) δ: -3.8 (d,
1JSi-P=23 Hz)
2.2.4: Characterization of Related Species
RhNacNac(H)(HPPh2)(HSiPh2) (2-10) - To a solution of RhNacNac(COE)N2 (20 mg, 0.030
mmol) in toluene (2 mL) was added Ph2SiH2 (12 mg, 0.064 mmol). The solution was cooled to
-35ºC upon which Ph2PH (6 mg, 0.03 mmol) was added. The solution was allowed to warm to
room temperature, the solvent was removed in vacuo and the residue was washed with pentane
(2 x 2 mL), leaving crude 2-10. Yield: 12 mg (43%). Rapid decomposition precluded elemental
analysis.
1H NMR (C6D6) δ: -13.5 (dd,
2JH-P=28 Hz,
1JH-Rh=15 Hz, 1H), 0.23 (d,
3JH-H=7 Hz, 3H), 0.62
(3d, 3JH-H=7 Hz, 3H), 0.71 (d,
3JH-H=7 Hz, 3H), 1.04 (d,
3JH-H=7 Hz, 3H), 1.06 (d,
3JH-H=7 Hz,
3H), 1.11 (d, 3JH-H=7 Hz, 3H), 1.43 (d,
3JH-H=7 Hz, 3H), 1.47 (d,
3JH-H=7 Hz, 3H), 1.65 (s, 3H),
1.88 (s, 3H), 2.76 (sept, 3JH-H=7Hz, 1H), 2.82 (sept,
3JH-H=7 Hz, 1H), 4.07-4.16 (ov sept,
3JH-H=7
Hz, 2H), 5.05 (d, 1JP-H=361 Hz, 1H), 5.00 (d,
2JH-Rh=36 Hz, 1H), 5.34 (s, 1H), 6.35-7.61 (26 H,
ov m). 31
P NMR (C6D6) δ: 47.4 (dd, 1JP-H = 361 Hz,
1JP-Rh=137 Hz).
29Si NMR ( C6D6) δ: 21.4
(dd, J=24 Hz, J=34 Hz).
RhNacNac(P(H)Cy2)(N2) (2-11) - To a solution of RhNacNac(COE)N2 (100 mg, 0.152 mmol) at
-35°C in toluene (5 mL) was added Cy2PH (30 mg, 0.15 mmol). The mixture was allowed to stir
overnight, and the solvent was removed in vacuo. Pentane (5 mL) was added and the solution
was cooled overnight to -35ºC and decanted to give the product as a dark red crystalline solid.
Yield: 90 mg (77%). A second recrystallization from the filtrate yielded X-ray quality crystals.
Anal. Calcd. for RhPN4C41H54 (%) C: 65.94, H: 8.64, N: 7.50; found: C: 66.35, H: 8.24, N:
7.12.
18
1H NMR (C6D6)δ: 1.10-1.84 (m, 20 H, broad peaks obscured by iso-propyl signals), 1.23 (d,
3JH-
H=6 Hz, 6H), 1.36 (d, 3JH-HJ=7 Hz, 6H), 1.59 (d,
3JH-H=7 Hz, 6H), 1.66 (d,
3JH-HJ=7 Hz, 6H), 1.70
(s, 3H), 1.91 (s, 3H), 2.20 (br m, 2H), 3.00 (dm, 1JH-Rh=333 Hz), 3.91 (sept,
3JH-HJ=7 Hz, 2H),
4.07 (sept, 3JH-H=7 Hz, 2H), 5.22 (s, 1H), 7.17-7.32 (m, 12 H).
31P NMR (C6D6) δ: 44.9 (
1JP-
H=333 Hz, 3JP-RhJ=158 Hz).
13C NMR ( C6D6) δ: 23.6, 24.1, 24.2, 24.5, 24.7, 25.4, 26.2, 27.3
(J=9 Hz), 27.7 (d, J-10 Hz), 27.9, 28.4, 30.9, 32.7, 33.2 (d, J-22 Hz), 97.0, 123.4, 125.2, 125.4,
141.5, 141.7, 151.4, 157.9, 158.6.
RhNacNac(Et2PPEt2) (2-12) - To a solution of RhNacNac(COE)N2 (85 mg, 0.13 mmol) in 5 mL
of pentane was added a solution of Et2PPEt2 (23 mg, 0.13 mmol) in 5 mL of pentane. The
mixture was allowed to stir overnight, filtered through Celite and cooled overnight at -35ºC.
Solvent was decanted to give the product as an orange crystalline solid. Yield: 30 mg (37%).
Recrystallization of the pentane wash at -35ºC yielded X-ray quality crystals. Anal. Calcd. for
RhP2N2C37H55 (%) C: 63.32, H: 9.19, N: 3.99; found: C: 63.55, H: 9.24, N: 4.12.
1H NMR (C6D6) δ: 0.74 (m,
3JH-H=8 Hz, 8H), 1.26 (td,
3JH-H=7 Hz,
3JH-P=3 Hz, 12H), 1.35 (d,
3JH-H=7 Hz, 12H), 1.63 (d,
3JH-H=7 Hz, 12H), 1.84 (s, 6H), 4.19 (sept.
3JH-H=7 Hz, 4H), 5.19 (1H,
s), 7.15-7.29 (m, 6H). 31
P NMR (C6D6) δ: -64.51 (d, 1JP-Rh=127 Hz).
13C NMR ( C6D6) δ: 9.8,
12.9, 22.4, 23.9 (d, J=10 Hz), 28.3, 97.0, 123.2, 123.8, 127.3-130.5 (m, obscured by C6D6),
140.3, 156.7, 159.4.
2.2.5: Synthesis of RhNacNac(P5R5) Complexes
RhNacNac(P5Ph5) (2-13) - P5Ph5 (86 mg, 0.16 mmol) was added to a solution of
RhNacNac(COE)N2 (100mg, 0.152 mmol) in toluene (5 mL). The mixture was allowed to stir
overnight at room temperature. Volatiles were removed and cold pentane (5 mL, -35 °C) was
added. The solution was filtered and 2-13 was isolated as a red powder. Yield: 120 mg (73%).
Anal. Calcd. for RhP5N2C59H66 (%) C: 66.79, H: 6.27, N: 2.64; found: C: 66.65, H: 6.47, N:
2.42.
1H NMR (C6D6) δ: 0.49 (d,
3JH-H=7 Hz, 3H), 1.01 (d,
3JH-H=7 Hz, 3H), 1.16 (d,
3JH-H=7 Hz, 3H),
1.35 (d, 3JH-H=7 Hz, 3H), 1.40 (d,
3JH-H=7 Hz, 3H), 1.51 (s, 3H), 1.55 (s, 3H), 1.78 (d,
3JH-H=7
Hz, 3H), 2.18 (d, 3JH-H=7 Hz, 3H), 2.82 (sept.,
3JH-H=7 Hz, 1H), 3.76 (sept.,
3JH-H=7 Hz, 1H),
4.46 (sept., 3JH-H=7 Hz, 1H), 4.83 (sept.,
3JH-H=7 Hz, 1H), 5.04 (s, 1H), 6.20 (d, J=8 Hz, 1H),
19
6.29 (t, J=8 Hz, 2H), 6.33 (t, J=7 Hz, 2H), 6.41 (t, J=7 Hz, 1H), 6.45-6.51 (m, 2H), 6.60 (m, 3H),
6.74-6.85 (m, 6H), 6.90-7.03 (m, 3H), 7.11 (1d, J=7 Hz, 1H), 7.21 (d, J=8 Hz, 1H), 7.24 (t, J=7
Hz, 1H), 7.36 (t, J=7 Hz, 2H), 7.69 (t, J=9 Hz, 4H), 9.51 (t, J=8 Hz, 2H). 31
P NMR (C6D6) δ: -
16.1 (ddddd, J=201 Hz, J=213 Hz, J=18 Hz, J=12 Hz, J=6 Hz), -1.4 (ddddd, J=364, J=265 Hz,
J=9 Hz, J=18 Hz, J=9 Hz), 15.5 (ddddd, J=265 Hz, J=198 Hz, J=12 Hz, J=9 Hz, J=2 Hz), 31.2
(ddddd, J=201 Hz, J=198 Hz, J=150 Hz, J=9 Hz, J=3 Hz), 49.3 (ddddd, J=364 Hz, J=213 Hz,
J=157 Hz, J=9 Hz, J=3 Hz).
RhNacNac(P5Et5) (2-14) - P5Et5 (28 mg, 0.079 mmol) was added to a solution of
RhNacNac(COE)N2 (50 mg, 0.076 mmol) in toluene (5 mL). The mixture was allowed to stir
for two weeks. Volatiles were removed in vacuo and pentane (0.5 mL) was added to the residue.
This solution was stored at -35 °C for three days and the solvent was decanted, leaving the
product as a dark red powder. Yield: 28 mg (43%). Anal. Calcd. for RhP5N2C39H66 (%) C:
57.07, H: 8.11, N: 3.41; found: C: 57.31, H: 8.47, N: 3.18.
1H NMR (C6D6) δ: 0.38 (m, 2H), 0.52 (dt, J=17 Hz, J=7 Hz, 3H), 0.96-1.03 (m, 6H), 1.06 (d,
3JH-H=7 Hz, 3H), 1.10 (d,
3JH-H=7 Hz, 3H), 1.11-1.35 (m, 12H), 1.32 (d,
3JH-H=7 Hz, 3H), 1.37 (d,
3JH-H=7 Hz, 3H), 1.39 (d,
3JH-H=7 Hz, 3H), 1.70 (d,
3JH-H=7 Hz, 3H), 1.72 (s, 3H), 1.73 (d,
3JH-
H=7 Hz, 3H), 1.78 (s, 3H), 1.81 (d, 3JH-H=7 Hz, 3H), 2.25 (m, 1H), 2.48 (m, 1H), 3.12 (sept.,
3JH-
H=7 Hz, 1H), 3.57 (sept., 3JH-H=7 Hz, 1H), 3.61 (sept.,
3JH-H=7 Hz, 1H), 3.65 (sept.,
3JH-H=7 Hz,
1H), 4.94 (s, 1H), 7.02 (dd, J=7 Hz, J=1 Hz, 1H), 7.12 (t, J=8 Hz, 1H), 7.15-7.21 (obscured by
C6D6, 2H), 7.30 (dd, J=8 Hz, J=2 Hz, 1H). 31
P NMR (C6D6) δ: -110.0 (dddd, J=259 Hz, J=174
Hz, J=85 Hz, J=36 Hz), -15.9 (dd, J=147 Hz, J=75 Hz), 18.5 (dd, J=212 Hz, J=167 Hz), 20.9 (dd,
J=174 Hz, J=147 Hz), 79.7 (dddd, J=168 Hz, J=61 Hz, J=39 Hz, J=7 Hz).
2.2.6: X-Ray Data Collection, Reduction, Solution and Refinement
Single crystals were mounted in thin-walled capillaries either under an atmosphere of dry N2 in a
glove box and flame sealed or coated in paratone-N oil. The data were collected using the
SMART software package on a Siemens SMART System CCD diffractometer using a graphite
monochromator with Mo Κα radiation (λ = 0.71073 Å). A hemisphere of data was collected in
1448 frames with 10 second exposure times unless otherwise noted. Data reductions were
performed using the SAINT software package and absorption corrections were applied using
SADABS. The structures were solved by direct methods using XS and refined by full-matrix
20
least-squares on F2 using XL as implemented in the SHELXTL suite of programs. All non-H
atoms were refined anisotropically. Carbon-bound hydrogen atoms were placed in calculated
positions using an appropriate riding model and coupled isotropic temperature factors.
Phosphorus-bound hydrogen atoms were located in the electron difference map and their
positions refined isotropically. Single crystal X-ray structures were obtained for 2-3, 2-4, 2-11,
2-12, 2-13 and 2-14. Selected crystallographic data are included in Tables 2.1 and 2.2.
Diagrams and selected bond lengths and angles are provided in Figures 2.5, 2.11, 2.13 and 2.14.
21
Table 2.1: Selected crystallographic data for compounds 2-3, 2-4 and 2-11
Crystal 2-3 2-4 2-11
Formula C53H61N2P2Rh C42H34P2Rh2 C41H64N4PRh
Formula weight 890.89 806.45 746.85
Crystal system Monoclinic Orthorhombic Monoclinic
Space group P21/n Pbca C2/c
a(Å) 10.8591(11) 10.454(4) 23.1939(5)
b(Å) 35.296(4) 18.159(6) 11.7596(4)
c(Å) 12.6312(13) 35.832(13) 29.7579(9)
(o) 90.0 90.00 90.00
( o) 94.769(2) 90.00 93.5886(17)
( o) 90.0 90.00 90.00
V (Å3) 4824.6(9) 6802(4) 8100.6(4)
Z 4 8 8
d(calc) g cm-1
1.227 1.575 1.225
Abs coeff, , cm-1
0.456 1.094 0.493
Data collected 8472 4882 9146
Data Fo2>3(Fo
2) 6281 4203 5189
Variables 525 415 428
Ra 0.0484 0.0446 0.0547
Rwb 0.1099 0.1097 0.1159
Goodness of Fit 1.053 1.124 1.002
These data were collected at 293 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
22
Table 2.2: Selected crystallographic data for compounds 2-12, 2-13 and 2-14
Crystal 2-12 2-13-(0.5 C6H14) 2-14
Formula C37H61N2P2Rh C62H73N2P5Rh C39H66N2P5Rh
Formula weight 698.73 1103.98 820.70
Crystal system Triclinic Monoclinic Monoclinic
Space group P-1 P21/c P21/n
a(Å) 10.566(3) 12.652(2) 12.7417(15)
b(Å) 11.676(4) 23.285(4) 17.733(2)
c(Å) 17.636(6) 20.121(4) 19.684(2)
(o) 102.412(4) 90.0 90.0
( o) 93.045(4) 105.382(2) 103.2450(10)
( o) 114.125(4) 90.0 90.0
V (Å3) 1914.7(11) 5715.3(18) 4329.3
Z 2 4 4
d(calc) g cm-1
1.212 1.283 1.259
Abs coeff, , cm-1
0.555 0.479 0.607
Data collected 6726 13164 7614
Data Fo2>3(Fo
2) 6106 7128 6478
Variables 381 631 424
Ra 0.0459 0.0568 0.0579
Rwb 0.1233 0.1090 0.1480
Goodness of Fit 0.969 1.022 1.020
These data were collected at 293 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
23
2.3: Results and Discussion
2.3.1: Catalyst Selection and Initial Screening
To explore the mechanism and scope of the dehydrocoupling of P-H bonds, the
secondary phosphine Ph2PH was used as the test substrate. The dehydrocoupling activity with
the catalyst precursors RhNacNac(COE)(N2) (2-1) (NacNac=HC{CN(2,6-iPr2C6H3)}2, COE=cis-
cyclooctene) and (n5-C9H7)Rh(COE)2 (2-2) (Figure 2.3) was examined. These catalysts were
chosen according to the criteria previously mentioned: steric bulk and an electron-donating
anionic ligand. Both Rh species present unique possibilities for reactivity as compared to the
previously used systems. 2-1 is a 16-electron precursor with two labile ligands, while 2-2 is an
18-electron precursor with 2 labile ligands, which can allow for facile coordination or oxidative
addition of 2 equivalents of HPPh2. NacNac ligands such as that seen in 2-1 have attracted much
recent attention due to their facile synthesis and ease of tunability (both sterically and
electronically).119
Additionally, the indenyl ligand of 2-2 presents the possibility of ring slippage
from η5 to η
3 to open up an additional coordination site.
120,121
Figure 2.3: Rh(I) Catalyst Precursors 2-1: RhNacNac(COE)(N2) (Ar=2,6-di-iso-propylphenyl)
and 2-2: (n5-C9H7)Rh(COE)2
Initial dehydrocoupling reactions with Ph2PH using both Rh species showed promise as
the product, Ph2PPPh2, was observable in 35% yield after 24 hours. However, additional
reaction time, increasing reaction temperature to 100°C, and use of coordinating solvents such as
THF all failed to improve the yield significantly.
24
2.3.2: Stoichiometric Reactions of Catalyst Precursors with Ph2PH and Ph2PPPh2
In light of these results, stoichiometric reactions of the catalyst and substrate (Ph2PH)
were performed. The addition of two equivalents of Ph2PH to each catalyst precursor resulted in
equilibria between mono-, bis- and tris-diphenylphosphine Rh species (Figure 2.4) and the
formation of small amounts of the dehydrocoupled product: Ph2PPPh2. In each case the bis-
coordinated complexes were favoured. It should be noted that Brookhart and co-workers
observed a similar equilibrium in the reaction of Ph2PH with their Cp*Rh(I) based catalyst
precursor.14
Figure 2.4: Equilibria involving mono-, bis- and tris-diphenylphosphine-rhodium complexes
(L=NacNac or n5-C9H7)
The addition of one equivalent of Ph2PH to each precursor resulted in significantly
different outcomes. In the case of 2-1, equilibrium between starting material and mono- and bis-
diphenyl phosphine species was observed by 31
P NMR spectroscopy. In contrast, when 2-2 was
employed quantitative formation of a new species was observed after 24 hours. This new
compound exhibited a unique signal in the 31
P{1H} NMR spectrum, a triplet appearing at 152.9
ppm with 1
JP-Rh=153 Hz. The 31
P NMR resonance showed no large 1-bond coupling to
hydrogen, thus it was proposed that the large coupling constant is due to coupling to two
equivalent Rh nuclei. The species was formulated as [(n5-C9H7)Rh(µ-PPh2)]2. This connectivity
was further confirmed by X-ray crystallography (Figure 2.5, left). The molecular structure of 2-
3 exhibits a butterfly-type conformation, with two formally Rh(II) centres bound to each other
(as the compound remains diamagnetic). This arrangement is fairly typical of other phosphide-
bridged rhodium complexes known in the literature.122-130
While these related species are
known, most are made by reacting the transition metal chloride with the corresponding lithium
phosphide. It appears that this is the first time such a compound has been made through P-H
bond activation and loss of H2. The Rh-Rh bond length is fairly short at 2.7272(8) Ǻ,
significantly shorter than the related compound (Cp*)Rh(µ-PPh2)(µ-PMe2)Rh(Cp*) at 2.7952(4)
25
Ǻ,129
presumably a result of reduced steric repulsion for 2-3, and similar to [(CO)3Fe(µ-CO)(µ-
PPh2)Rh(µ-PPh2)]2 at 2.723(2) Ǻ.124
The analogous Cp* compound has been synthesized by
reaction of Cp*RhCl2 with HPPh2, followed by treatment with diethylamine, although silica gel
column chromatography was required to separate this product from (Cp*Rh)2(µ-PPh2)(µ-Cl).129
Figure 2.5: POV-Ray depictions of 2-3 (left) and 2-4 (right). Carbon: black, Nitrogen: blue,
Rhodium: pink, Phosphorus: orange. Hydrogen atoms are omitted for clarity. Selected metrical
parameters (Distances: Å, Angles: °): 2-3: Rh1-Rh1a 2.7272(8), Rh1-P1 2.2295(16), Rh1-P1a
2.2326(16), Rh1a-P1a 2.2224(16), Rh1a-P1 2.2292(17), P1-Rh1-P1a 83.65(6), P1-Rh1a-P1a
83.89(5), Rh1-P1-Rh1a 75.42(5), Rh1-P1a-Rh1a 75.49(5). 2-4: Rh1-N1 2.065(3), Rh1-N2
2.054(3), Rh1-P1 2.2300(10), Rh1-P2 2.2424(10), P1-P2 2.1390(14), N1-Rh1-N2 90.52(11), N1-
Rh1-P2 162.38(9), N1-Rh1-P1 107.38(9), N2-Rh1-P1 160.62(8), N2-Rh1-P2 106.23(8), P1-Rh1-
P2 57.14(14).
Compound 2-3 was found to be an active catalyst for the dehydrocoupling reaction of
Ph2PH, thus its potential formation during the dehydrocoupling reaction is not responsible for the
low activity observed. Some other factor must be at play in the dehydrocoupling reaction,
resulting in the low activity observed.
Finally, reactions of the catalyst precursors 2-1 and 2-2 with Ph2PPPh2, the product of the
dehydrocoupling reaction, were attempted. Though the dehydrocoupling reactions had been
previously conducted using Rh(I) precursors, to our knowledge the stoichiometric reaction with
Ph2PPPh2 has not been previously examined.
26
Reaction of 2-2 with one equivalent of Ph2PPPh2 overnight at room temperature did not
show any observable reaction. However, heating at 80 °C for 2 hours produced 2-3, consuming
only 0.5 equivalents of Ph2PPPh2 (Figure 2.6).
Figure 2.6: Formation of Rh-phosphide dimer 2-3
Reaction of 2-1 with one equivalent of Ph2PPPh2 yielded a new product 2-4, showing a
doublet resonance in the 31
P{1H} NMR spectrum at -51.4 (
1JP-Rh=140 Hz). Given the nature of
the reactants and the observed NMR data, species 2-4 was identified as RhNacNac(Ph2PPPh2).
X-ray crystallography showed that the P-P bond remained intact (Figure 2.5, right). As might be
imagined this compound shows a rather unique pseudo-square planar geometry, with an
extraordinarily tight bite angle of 57.14(4)° for the P-P ligand on rhodium. The P-P bond length
of 2.1389(14) Ǻ is reasonable for the intact P-P bond and is in fact substantially shorter than the
bond length determined for the starting material, Ph2PPPh2 at 2.217(1) Ǻ131
and other simple
P2R4 species (R=Me, 2.212(1) Ǻ;132
R=Cy, 2.215(3) Ǻ;133
R=Mes, 2.260(1) Ǻ134
). The Rh-P2
plane is twisted 14.8° from the Rh-N2 plane to minimize crowding between the β-diketiminate
ligand and the phenyl groups of the Ph2PPPh2 ligand. Compound 2-4 is also unique in that
previous reactions of rhodium complexes with biphosphines resulted in either monodentate
complexes,125,127,135-137
or bridging species127,135,138-141
in which the P2 fragment remains intact. It
is perhaps not surprising that bridging compounds were not formed given the bulk of the NacNac
ligand, while a monodentate species is also unlikely due to the highly labile nature of both the
cis-cyclooctene and N2 ligands.
2.3.3: Catalytic Hydrogenation and Hydrosilylation Reactions of Ph2PPPh2
Given the facile formation of 2-3 and 2-4, transformations involving the P2 fragment
were examined. The first reaction attempted was the catalytic hydrogenation of Ph2PPPh2 to see
if this reverse reaction was a possible explanation for the low yields observed in the initial
27
dehydrocoupling reaction (section 2.3.1). Indeed, addition of hydrogen to a solution of
Ph2PPPh2 and 10 mol % of the catalyst precursor 2-1 or 2-2 resulted in hydrogenation of
Ph2PPPh2 to yield two equivalents of the secondary phosphine, Ph2PH. Using 2-1, 95%
conversion was observed over 24 hours at 50°C. Repeating the reaction using 2-2 as the catalyst
precursor showed significantly slower conversion, reaching a maximum of 82% conversion over
36 hours under identical conditions. This is possibly due to the reduced steric bulk of 2-2, which
does not prompt the loss of Ph2PH as efficiently as 2-1. Precursor 2-1 was chosen for further
investigation due to its increased activity. Over the course of the reaction, the peak at -13 ppm
(Ph2PPPh2) in the 31
P NMR spectrum disappears and the peak at -39 ppm (HPPh2) grows in. The
initial Rh species in the reaction (by 31
P NMR spectroscopy) is 2-4 and as the reaction proceeds
more RhNacNac(HPPh2)2 becomes visible by 31
P NMR spectroscopy.
Given the promising results of the hydrogenation of P-P bonds and the analogy of Si-H
and H-H bonds,142
similar reactions were attempted with silanes. Reaction of Ph2PPPh2 with two
equivalents of H2SiPh2 in the presence of 5 mol% 2-1 generated the silyl phosphine
Ph2(H)SiPPh2 in 74% yield, along with the byproduct Ph2PH in 23% yield. Yields of this
relatively slow reaction could be improved significantly by using 5 equivalents of the silane and
heating to 100°C for 48 hours which ensures that any decomposition of the silane by
homodehydrocoupling or reaction with trace moisture is not a factor in the rate of reaction.
Analogous reactions were performed with a series of secondary and tertiary silanes, forming the
corresponding silylphosphines 2-5 - 2-9, with the general trend being that smaller, electron-
deficient silanes provided the best results and that trialkyl silanes proved to be very poor
reactants (Table 2.3). This is likely due to the increased polarization of the Si-H bond in electron
poor silanes, which can facilitate oxidative addition. In all cases, the dominant Rh species in
solution appears to be RhNacNac(HPPh2)2. Notably, uncatalyzed reactions of Ph2PPPh2 with
any of the tested silanes did not proceed.
28
Table 2.3: Silylation of Ph2PPPh2 using 10 mol% 2-1 at 100°C in toluene
Silane (5 equivalents) Time (hours) Yield P-Si (%) Yield P-H (%)
Ph2SiH2 48 98 (2-5) <1
PhMe2SiH 48 95 (2-6) 5
Ph2MeSiH 48 76 (2-7) 17
Ph3SiH 48 72 (2-8) 25
Et3SiH 24 16 (2-9) 44
iPr3SiH 24 - 29
% Yields determined using 31
P NMR spectroscopy
2.3.4: Mechanistic Insight into the Catalytic Activation of P-P Bonds
The silylation of Ph2PPPh2 fragments is proposed to proceed in two steps. The initial
step would provide equal amounts of Ph2PH and Ph2(H)SiPPh2 and the second step must convert
Ph2PH to H2SiPPh2 through reaction with a second equivalent of silane (Figure 2.7).
Figure 2.7: Rh-catalyzed hydrosilylation of P2Ph4
The second step of this reaction, the heterodehydrocoupling of secondary phosphines
with silanes is much faster than the first since only trace amounts of Ph2PH are observed over the
course of the reaction. The dehydrocoupling of diphenylphosphine with a series of silanes was
examined and results are summarized in Table 2.4. Similar reactions involving the
dehydrocoupling of Ph2PH with silanes have been previously reported by Harrod and co-workers
with titanocene catalyst precursors.116
29
Table 2.4: Results of heterodehydrocoupling reactions of Ph2PH with silanes using 5 mol % 2-1
at 50°C in toluene
Silane (5 eq) Time (hours) P-Si P-P
Ph2SiH2 18 99 (2-5) -
PhMe2SiH 18 84 (2-6) -
Ph2MeSiH* 18 85 (2-7) <1
Ph3SiH* 18 40 (2-8) 10
Et3SiH* 18 <1 (2-9) 29
iPr3SiH* 18 - 29
*Reactions were performed at 100°C as reactions at 50°C had very low conversions
In an effort to garner mechanistic insight, further stoichiometric reactions were
undertaken. Addition of hydrogen and excess silane to 2-4 did not show any reaction by
multinuclear NMR spectroscopy. Based on this result it was proposed that the oxidative addition
to Rh(III) must be reversible and that in the absence of excess substrate the equilibrium lies in
favour of 2-4. A similar intermediate can be proposed for the hetero-dehydrocoupling of
phosphines and silanes. The stoichiometric reaction of 2-1 with one equivalent of Ph2PH and
then one equivalent of H2SiPh2 (all conducted at -35°C to slow any potential dehydrocoupling)
produced a new rhodium complex assigned as RhNacNac(Si(H)Ph2)(H)(P(H)Ph2) (2-10) based
on multinuclear NMR spectroscopic data (Figure 2.8). Compound 2-10 was a short-lived
species, decomposing within 2 hours in solution, thus further analysis was not possible. The
NMR data was quite conclusive as a Rh-hydride resonance was observed at -13.5 ppm (2JH-P=28
Hz, 1JH-Rh=15 Hz) and a P-H resonance was observed at 5.09 ppm (
1JP-H=361 Hz) in the
1H NMR
spectrum. Resonances split into doublets of doublets are also observed in the 29
Si{1H} and
31P
NMR spectra. This compound is reminiscent of NacNacIr(H)2(PR3) species reported by Chirik
and co-workers143
and also closely related to Cp*Rh(H)(Si(H)Ph2)(PR3) species synthesized by
Marder and co-workers.144
30
Figure 2.8: Structure of 2-10
From these results, a catalytic cycle was proposed for P2 bond activation which entails
partial dissociation of the P2 ligand, opening up the metal centre for oxidative addition of the H-
H or Si-H bond. This Rh(III) intermediate eliminates Ph2PH, generating a 14-electron Rh(III)
species which rapidly eliminates Ph2PH (or Ph2PSi(H)Ph2) in favour of another P2 unit, restarting
the catalytic cycle (Figure 2.9). A cycle involving a Rh(V) intermediate should not be dismissed
as Rh(V) was proposed as an intermediate in Brookhart’s catalytic cycle (Figure 2.2) and the
Ir(I) analogue of 2-1 has been shown to undergo oxidative addition of two equivalents of H2 by
Chirik and coworkers.145
Figure 2.9: Proposed catalytic cycles for hydrogenation (left) and hydrosilylation (right) of
Ph2PPPh2
31
2.3.5: Heterodehydrocoupling of Silanes with Diphenylphosphine
The reactions of secondary phosphines (the second part of the P2 activation reaction) with
H2 or Si-H bonds is proposed to proceed in a very similar fashion as the P-P bond activation.
Here, the initial loss of a Ph2PH molecule from NacNacRh(PHPh2)2 is required for the
subsequent H-H or Si-H oxidative addition at Rh(I). Elimination of H2, possibly through a
Rh(V) intermediate, leads to a 14-electron Rh(III) species which rapidly picks up two Ph2PH
molecules and reductively eliminates the silyl-phosphine product (Figure 2.10).
Figure 2.10: Proposed mechanism for dehydrocoupling of Ph2PH with silanes
In a related reaction, a solution of Ph2PH and 5 mol% 2-1 was reacted with excess D2,
achieving 85% conversion to Ph2PD in 24 hours at room temperature. This reaction, providing
easy access to Ph2PD without the use of LiAlD4, is expected to proceed in an analogous fashion
to the catalytic cycle shown in Figure 2.10 for the heterodehydrocoupling reaction with silanes.
2.3.6: Reactions of Catalyst Precursors with Additional Phosphines and Biphosphines
In efforts to expand the scope of the reaction, efforts were made to carry out catalytic
reactions with Cy2PH and P2Et4. Unfortunately these substrates were not found to be as reactive
as Ph2PH or Ph2PPPh2. Heterodehydrocoupling reactions of Cy2PH with silanes produced small
amounts of product but conversion was only slightly above stoichiometric in Rh for Ph2SiH2 and
32
PhMe2SiH (21% conversion) and sub-stoichiometric for other silanes. It should also be noted
that heating reactions to 100°C was required to achieve even this minimal conversion, where
heating to only 50°C provided excellent yields using Ph2PH. The low catalytic activity in these
cases is further support for the proposed mechanism since the dissociation of these more basic
phosphines and biphosphines is less facile. While catalytic reactions with Cy2PH were not
successful, stoichiometric reaction provided some insight into reactions of Ph2PH with 2-1, as
reaction of one equivalent of phosphine with 2-1 resulted in formation of the
RhNacNac(P(H)Cy2)(N2) (2-11). The X-ray crystal structure of this species confirmed
connectivity (Figure 2.11, left). The Rh-N2 bond length at 1.951(4) Å is similar to that of the
starting material 2-1, 1.943(4) Å. Surprisingly, the N-N bond length observed in 2-11 of
1.029(4) Å, is substantially shorter than that of 2-1, 1.091(6) Å. The relative stability of this
species compared to the Ph2PH analogue is due to increased basicity and bulk of
dicyclohexylphosphine which reduces the chances of dissociation and bis-coordination,
respectively.
Stoichiometric reaction of Et2PPEt2 with 2-1 resulted in rapid formation of 2-12, a
species which looked spectroscopically quite similar to 2-4, with a doublet resonance seen in the
31P NMR spectrum at -64.5 ppm (
1JP-Rh=127 Hz). X-Ray crystallography confirmed that 2-12
possessed a structure very similar to 2-4 (Figure 2.11, right). 2-12 shows a P-P bond length of
2.1254(14) Ǻ, which results in a Rh-P-P bite angle of 56.77(4)° at rhodium, even tighter than
that observed in 2-4. A combination of reduced steric bulk and increased electron-donating
ability oppose the partial dissociation of the P2Et4 ligand believed to be required for the catalytic
cycle.
33
Figure 2.11: POV-Ray depiction of 2-11 (left) and 2-12 (right). Carbon: black, Nitrogen: blue,
Rhodium: pink, Phosphorus: orange, Hydrogen: white. Carbon-bound hydrogen atoms are
omitted for clarity. Selected metrical parameters (Distances: Å, Angles: °): 2-11: Rh1-N1
1.951(4), Rh1-N3 2.041(3), Rh1-N4 2.0.87(3), Rh1-P1 2.2705(11), N1-N2 1.029(4), N3-Rh1-N4
89.36(12), N1-Rh1-N3 176.83(13), N1-Rh1-N4 90.97(13), N4-Rh1-P1 175.68(9), Rh1-N1-N2
178.6(4). 2-12: Rh1-N1 2.068(2), Rh1-N2 2.064(2), Rh1-P1 2.2465(10), Rh1-P2 2.2231(11),
P1-P2 2.1250(14), N1-Rh1-N2 90.30(9), N1-Rh1-P1 108.03(7), N2-Rh1-P1 160.10(7), N2-Rh1-
P2 105.36(7), P1-Rh1-P2 56.77(4).
2.3.7: Reactions with Cyclic Polyphosphines
Cyclic polyphosphines of the general formula P4R4 and P5R5 are well-known in the
literature27
and are also bulky species containing P-P bonds and may be suitable for P-P bond
activation by 2-1. Thus, polyphosphines A and B (Figure 2.12) were reacted with 2-1. Once
again, while catalytic reactions were unsuccessful, investigations of the stoichiometric reactions
proved to be very interesting. The P4R4 fragments showed no reaction with 2-1, presumably due
to the large bulk or the trans-R groups on adjacent phosphorus centres blocking access of any
lone pair to the Rh centre. In contrast, the reactions of 2-1 with P5Ph5 and P5Et5, resulted in
products which showed very clean, unique patterns in the 31
P NMR spectra after 1 day and 2
weeks at room temperature, respectively.
34
Figure 2.12: Cyclic Polyphosphines: A) P4R4 (R=Cy, tBu) B) P5R5 (R=Ph, Et)
The product of the reaction of 2-1 with P5Ph5 showed 5 remarkably well-resolved
phosphorus resonances from 60 to -20 ppm in the 31
P NMR spectrum (Figure 2.13). Relative
integration of the 1H NMR spectrum confirmed a 1:1 reaction between the P5 fragment and 2-1.
This 1:1 metal:polyphosphine ratio is not common for the P5R5 unit, as most reactions show
either fragmentation of the P5 ring or coordination to multiple metal centres.146,147
Further
examination of the 31
P NMR spectrum (including the use of a 31
P-31
P NMR correlation
spectroscopy) and NUMMRIT148
NMR simulation performed using Spinworks149
revealed that
each resonance was coupled to the other 4 phosphorus centres, as well as to rhodium, with
coupling constants ranging from 2-364 Hz. The 1H NMR spectrum showed 8 inequivalent iso-
propyl-methyl resonances, 2 inequivalent backbone methyl resonances and 4 inequivalent iso-
propyl-methyne resonances, suggesting that 2-13 is asymmetric above and below the RhN2 plane
and each side of the NacNac ligand is also inequivalent. This suggests that the P5 unit is bound
in an unsymmetrical and likely polydentate fashion and there is restricted rotation within the
molecule.
35
The structure of 2-13 was determined crystallographically (Figure 2.13) and this
connectivity is consistent with spectroscopic data obtained in solution. The P5Ph5 ring has not
-15-10-555 50 45 40 35 30 25 20 15 10 5 0 ppm
Figure 2.13: 31
P NMR Spectrum (top, left) and POV-Ray depiction of 2-13-0.5 C6H14. Carbon: black,
Nitrogen: blue, Rhodium: pink, Phosphorus: orange. Solvent and hydrogen atoms are omitted for
clarity. Selected metrical parameters (Distances: Å, Angles: °): Rh1–P1 2.2488(12), Rh1–P2
2.2739(12), Rh1–N2 2.079(3), Rh1–N1 2.105(4), P1–P3 2.1791(16), P1–P5 2.2554(17), P2–P3
2.2088(17), P2–P4 2.2452(17), P4–P5 2.2160(18); N2–Rh1–N1 89.99(14), N2–Rh1–P1 98.58(10),
N1–Rh1–P1 168.65(11), N2–Rh1–P2 166.80(10), N1–Rh1–P2 100.91(11), P3–P1–Rh1 93.39(6),
Rh1–P1–P5 110.23(6), P1–Rh1–P2 71.69(4), P3–P2–Rh1 91.92(5), P4–P2–Rh1 108.21(6), P3–P1–P5
102.32(7), P3–P2–P4 106.99(7), P1–P3–P2 74.26(6), P5–P4–P2 94.17(6), P4–P5–P1 96.65(6).
a
a b c d e
b
c d
e
36
only coordinated to the Rh(I) centre in a bidentate fashion, but also one phosphorus atom has
inverted (in the product, 4 Ph groups are on the same side of the P5 ring, whereas only 3 are cis
in the starting material). While the inversion barrier of phosphorus is generally considered rather
high, experimental studies on cyclic phosphines have shown that the barrier to inversion can be
dramatically lower than barriers for inversion of acyclic phosphines.27,150
It should also be noted
that metal-phosphide complexes27,151-155
and compounds containing Lewis acids proximate to
phosphorus156
have also been shown to have lowered barriers to inversion. This suggests another
possibility: oxidative addition, followed by inversion at the phosphide centre, and then finally
reductive elimination to give 2-13.
The crystal structure provides insight into the nature of the 1H NMR signal observed at
9.51 ppm, integrating for 2 hydrogen atoms. At first glance, this appears to be a simple triplet.
However, in a 1H{
31P} NMR spectrum the resonance is simplified to a doublet, thus these atoms
are coupled to phosphorus with a coupling constant of 9 Hz, characteristic of ortho positions on a
phenyl ring bound to phosphorus. The crystal structure of 2-13 reveals an ortho-hydrogen of the
phenyl ring on P5 lies only 2.780 Ǻ from rhodium. Previous studies have noted similar 1H NMR
chemical shifts based on proximity to d8 metals.157
In solution, rotation about the P-C bond must
be very rapid as even at -80°C this signal did not resolve into two separate peaks. A 31
P-1H
HETCOR experiment was conducted and revealed that resonance “c” in Figure 2.13 was coupled
to these hydrogen atoms. This assignment was supported by comparison of coupling constants
and bond lengths for P4 and P5 (“bottom” P’s) which also suggested that resonance “c” was P5.
Having associated one signal in the 31
P NMR spectrum with the atom in the crystal structure, the
remaining resonances could be assigned using coupling constants (a:P2; b:P1; c:P5; d:P4; e:P3).
Interestingly, the 31
P NMR spectrum of the reaction between P5Et5 and 2-1 after 2 weeks
appeared significantly different from that of 2-13. Again, 5 inequivalent phosphorus centres are
observed, this time spread over a much wider range (80 to -120 ppm, see Figure 2.14) and the 1H
NMR spectrum reveals a lack of symmetry among the iso-propyl groups of the NacNac ligand.
31P NMR spectroscopy of the reaction in progress showed the initial appearance of a species with
resonances very similar to 2-13. This seems to suggest that the reaction proceeds through an
intermediate analogous to 2-13 but continues to a different product. Again, X-ray
crystallography was necessary to elucidate the connectivity of 2-14 (Figure 2.14).
37
In the case of 2-14, oxidative addition of a P-P bond has taken place and the resulting
complex is a five coordinate pseudo-square pyramidal Rh(III) species. Two coordination sites
are occupied by phosphide donors, another is occupied by a neutral phosphine donor and the
other two are occupied by the NacNac ligand. The Rh-phosphide bond lengths are 2.2993(12) Ǻ
for Rh-P1 and 2.2721(11) Ǻ for Rh-P5, while the Rh-phosphine bond length is only slightly
longer, at 2.3189(12) Ǻ. A combination of less steric requirements and stronger donor ability of
the P5Et5 fragment must contribute to the formation and stabilization of the 5-coordinate Rh(III)
complex.
38
Figure 2.14: 31
P NMR resonances (top) and POV-Ray depiction (bottom) of 2-14. Carbon: black,
Nitrogen: blue, Rhodium: pink, Phosphorus: orange. Hydrogen atoms are omitted for clarity.
Selected metrical parameters (Distances: Å, Angles: °): Rh1–P1 2.2993(12) Rh1–P3 2.2721(11),
Rh1–P5 2.3189(12), Rh1–N2 2.152(3), Rh1–N1 2.140(3), P1–P2 2.2226(16), P2–P3 2.1890(16),
P3–P4 2.1858(15), P4–P5 2.2243(15); N2–Rh1–N1 89.93(13), N1–Rh1–P3 68.16(9), N2–Rh1–
P3 101.93(9), N1–Rh1–P1 92.26(9), N2–Rh1–P1 115.67(10), P3–Rh1–P1 78.85(4), N1–Rh1–P5
95.22(9), N2–Rh1–P5 152.56(10), P3–Rh1–P5 77.36(4), P1–Rh1–P5 91.31(4), N1–Rh1–P3
168.16(9), P2–P1–Rh1 97.05(5), P3–P2–P1 82.30(6), P4–P3–P2 106.23(6), P3–P4–P5 81.17(5),
P4–P5–Rh1 99.39(5).
39
2.4: Conclusions
While the present catalyst precursors were relatively ineffective for P-H dehydrocoupling
reactions, a novel catalytic P-P bond activation pathway was discovered allowing for facile,
atom-economical synthesis of novel silyl phosphines. Stoichiometric reactions helped elucidate
a potential reaction pathway involving oxidative addition of Si-H bonds to the Rh(I) centre.
Reaction of the bulky Rh(I) complex 2-1 with cyclic P5R5 (R=Ph, Et) species provided clean
products in a rare 1:1 ratio. These products showed dramatically different connectivity as 2-13
shows simple coordination at 2 phosphorus centres accompanied by inversion at another
phosphorus centre, while 2-14 shows oxidative addition resulting in a tridentate P5R5 dianion.
40
Chapter 3: Frustrated Lewis Pair Reactivity of Bulky Catena-Polyphosphines with B(C6F5)3
3.1: Introduction
As introduced in Chapter 1, Frustrated Lewis Pairs (FLPs) are combinations of Lewis
acids and bases which do not form conventional donor-acceptor Lewis adducts due to steric
constraints. The ability for strong Lewis acids and bases to co-exist without quenching one
another through adduct formation has been exploited in a wide variety of small molecule
activations, including H2,43
olefins,67
THF,39,40
acetylenes,68
N2O71
and CO2.69
To date, this
reactivity has largely focused on combinations of bulky Lewis basic phosphines and the Lewis
acid B(C6F5)3. In addition to small molecule activation reactions in combination with B(C6F5)3,
phosphines have also been shown to carry out nucleophilic aromatic substitution at the para-
position of a C6F5 ring on B(C6F5)3 (Figure 3.1).2,36,37
As discussed in section 1.4.2, these
reactions can be explained in terms of FLP chemistry as the phosphine attacks at a para-carbon
rather than the more Lewis acidic, but more sterically hindered, boron centre. Phosphines have
proven to be particularly good Lewis bases for FLP chemistry as they are good nucleophiles and
sterically bulky derivatives are available commercially at low cost. In particular FLPs of
B(C6F5)3 in combination with the phosphines tBu3P and Mes3P are used extensively in small
molecule activation reactions.3,4
Figure 3.1: Nucleophilic aromatic substitution (NAS) at the para-position of a C6F5 ring by
phosphines on B(C6F5)3
The para-nucleophilic aromatic substitution reactions have even been extended to even
include some smaller phosphines (e.g. Et3P and Cy2PH) which form adducts with B(C6F5)3 at
room temperature.36
At elevated temperatures, these can dissociate in solution, generating free
41
phosphine and borane, and subsequently execute nucleophilic aromatic substitution on the Lewis
acid (Figure 3.1, bottom), forming zwitterionic phosphonium borates.
To date, only mono-phosphines and 1-8-bis(diphenyphosphino)naphthalene,76
a bis-
phosphine, have been utilized in FLP chemistry. The use of catena-polyphosphines represents a
new class of Lewis base used in these reactions, and could allow for unique reactivity based on
relatively weak P-P bonds.27
Despite the reduced basicity of these polyphosphines compared to
most tertiary phosphines, their large steric bulk makes them promising candidates for the
generation of FLPs.
The synthesis of catena-cyclopolyphosphinophosphonium cations, species in which a
formally cationic phosphorus centre is bound to another phosphorus centre, has drawn much
recent attention.158-171
These systems are of fundamental interest due to the diagonal relationship
between carbon and phosphorus and the ubiquitous nature of carbon-carbon bonds in organic
chemistry.172
While an extensive assortment of these cations has been synthesized, the vast
majority of these species have an additional alkyl or aryl group on a PxRx ring or P2R4 moiety.
Catena-polyphosphinophosphonium cations with a pendant borane or borate moiety are
attractive synthetic targets. Such materials may possess interesting structural properties and the
potential for further reactivity unavailable with the current library of catena-
polyphosphinophosphonium cations, allowing for further derivatization. For example, alkene or
alkyne activation reactions, which have been previously examined with tertiary phosphines
(Figure 3.2), could generate alkyl- or alkenyl-bridged polyphosphinophosphonium borates.
Figure 3.2: Activation of alkenes (left) and alkynes (right) by a frustrated Lewis pair: PR3 +
B(C6F5)3
The functionalization of white phosphorus (P4) has attracted recent interest, largely
focusing on the possibility of converting P4 cleanly into small organophosphines. Currently, the
vast majority of organophosphines, which are important in catalysis, pharmaceuticals, materials
and other applications, stem from PCl3, which is derived from P4 and chlorine gas.15
A method
42
for the synthesis of organophosphines avoiding the use of highly corrosive and toxic chlorine gas
is highly desirable.16-23
The functionalization of polyphosphines via FLP reactivity would
provide insight into potential pathways for the functionalization of P4 and the synthesis of
organophosphines from P-P bound species.
43
3.2: Experimental
3.2.1: General Considerations
All preparations were done under an atmosphere of dry, O2-free N2 employing both Schlenk line
techniques and an Innovative Technologies or Vacuum Atmospheres inert atmosphere glove box.
Solvents (pentane, hexanes, toluene, and methylene chloride) were purified employing a Grubbs’
type column systems manufactured by Innovative Technology and stored over molecular sieves
(4 Å). Molecular sieves (4 Å) were purchased from Aldrich Chemical Company and dried at 140
ºC under vacuum for 24 hours prior to use. Deuterated solvents were dried over
Na/benzophenone (C6D6, C7D8) or CaH2 (CD2Cl2, CDCl3) and distilled prior to use. All common
organic reagents were purified by conventional methods unless otherwise noted. 1H,
13C,
11B,
19F
and 31
P nuclear magnetic resonance (NMR) spectroscopy spectra were recorded on a Bruker
Avance-400 spectrometer at 300K unless otherwise noted. 1H and
13C NMR spectra are
referenced to SiMe4 using the residual solvent peak impurity. 11
B and 19
F NMR experiments
were referenced to 15% BF3-Et2O in CDCl3 and 31
P NMR experiments were referenced to 85%
H3PO4. Chemical shifts are reported in ppm and coupling constants in Hz as absolute values.
Combustion analyses were performed in house employing a Perkin Elmer CHN Analyzer.
Silanes were purchased from Strem Chemicals and used as received. Phenyl acetylene was
purchased from Aldrich Chemicals and used as received. H2 was passed through a dririte
column prior to use. tBu2PLi was prepared by treating the corresponding phosphine with 1
equivalent of tBuLi in toluene and collecting the precipitate by vacuum filtration. R2PPR2 was
prepared by reaction of the corresponding lithium phosphide and phosphine chloride. P5Ph527
and P4Cy427
were prepared as previously reported. B(C6F5)3 was generously donated by Nova
Chemicals.
3.2.2: Generation of a Phosphonium Borate Zwitterion through Nucleophilic Aromatic Substitution
P5Ph5-C6F4-B(F)(C6F5)2 (3-1) – P5Ph5 (300 mg, 0.55 mmol) was added to a solution of B(C6F5)3
(283 mg, 0.55 mmol) in toluene (20 mL) in a teflon-capped reaction bomb. The solution was
heated at 120°C for 6 days. Volatiles were removed in vacuo and the residue was washed with
hexanes (2 x 5 mL). Yield: 554 mg (95%). X-Ray quality crystals were grown from a layered
44
solution of CH2Cl2/C6H6/pentane. Anal. Calcd. for C48H35BF15P5 (%) C: 54.78, H: 2.39; found:
C: 55.16; H: 3.02.
-1H NMR (CD2Cl2) δ: 7.12-7.55 (m, 18H), 7.63 (t, J=7 Hz, 1H), 7.74 (t, J=8 Hz, 2H), 7.87 (t,
J=7 Hz, 2H), 7.94 (t, J=8 Hz, 2H). 19
F NMR (CD2Cl2) δ: -127.6 (d, 3
JF-P=68 Hz, 2F, C6F4), -
130.3 (s, 2F, C6F4), -135.4 (m, 4F, o-C6F5), -161.9 (t, 3JF-F=20 Hz, 2F, p-C6F5), -166.9 (m, 4F, m-
C6F5), -193.1 (br s, 1F, B-F). 31
P NMR (CD2Cl2) δ: 13.3 (tm, 1JP-P=359 Hz, 1P), -19.3 (ddm,
1JP-
P=359 Hz, 1JP-P=78 Hz, 1P), -28.8 (dd,
1JP-P=343 Hz,
1JP-P=107 Hz, 1P), -37.8 (m, 2P).
11B NMR
(CD2Cl2) δ: -2.7 (d, 1JB-F=52 Hz).
3.2.3: Synthesis of Alkenyl-bridged Phosphonium Borate Zwitterions via Activation of Terminal Alkynes
E-P3Cy3(PCy)(Ph)C=C(H)(B(C6F5)3) (3-2) - To a cold (-35°C) solution of B(C6F5)3 (25 mg,
0.049 mmol) and P4Cy4 (22 mg, 0.048 mmol) in CH2Cl2 (5 mL) was added phenyl acetylene (20
mg, 0.20 mmol) dropwise. The pale yellow solution was allowed to stir overnight, the solvent
was removed in vacuo and the residue was washed with pentane (2 x 2 mL), leaving a yellow
powder. Yield: 41 mg (78%). Anal. Calcd. for C50H50BF15P4 (%): C, 56.09%; H, 4.71% ; found:
C, 56.26; H, 4.94.
1H NMR (CDCl3) δ: 1.00 (m, 2H), 1.10-1.37 (m, 18H), 1.55-1.95 (m, 21H), 2.05 (m, 1H), 2.34
(m, 2H), 6.88 (d, 3JH-H=7 Hz, 2H, o-C6H5), 7.11 (t,
3JH-H=7 Hz, 2H, m-C6H5), 7.19 (tm,
3JH-H=7
Hz, 1H, p-C6H5), 8.30 (d, 3JP-H=37 Hz, C=C-H.
19F NMR (CDCl3) δ: -130.6 (d,
3JF-F=24 Hz, o-
C6F5), -161.7 (t, 3JF-F=22 Hz, p-C6F5), -165.9 (t,
3JF-F=23 Hz, m-C6F5).
31P NMR (CDCl3) δ:
20.4 (t, 1JP-P=247 Hz, 1P), -47.2 (dd,
1JP-P=247 Hz,
1JP-P=123 Hz, 2P), -59.4 (t,
1JP-P=123 Hz, 1P).
11B NMR (CDCl3) δ: -15.9 (br s).
E-P4Ph4(PPh)(Ph)C=C(H)(B(C6F5)3) (3-3) - To a solution of B(C6F5)3 (55 mg, 0.11 mmol) and
P5Ph5 (50 mg, 0.093 mmol) in CH2Cl2 (5 mL) was added phenyl acetylene (20 mg, 0.20 mmol).
The pale yellow solution was allowed to stir overnight, the solvent was removed in vacuo and
the residue was washed with pentane (2 x 2 mL), leaving an off-white powder. Yield: 105 mg
(91%). Anal. Calcd. for C56H31BF15P5 (%): C, 58.26; H, 2.71; found: C, 58.37; H, 2.94. X-Ray
quality crystals were grown from a layered solution of CDCl3/pentane.
45
1H NMR (CDCl3) δ: 5.98 (d,
3JH-H=9 Hz, 2H, o-C6H5), 6.51 (t,
3JH-H=8 Hz, 2H, m-C6H5), 6.76 (t,
3JH-H=8 Hz, 1H, p-C6H5), 7.01-7.74 (m, 25H, C6H5), 8.59 (dd,
3JP-H=42 Hz,
4JP-H=4 Hz, 1H,
C=CH). 19
F NMR (CDCl3) δ: -131.0 (d, 3
JF-F=24 Hz, 6F, o-C6F5), -163.1 (t, 3JF-F=22 Hz, 3F, p-
C6F5), -167.3 (t, 3JF-F=19 Hz, 6F, m-C6F5).
31P NMR (CDCl3) δ: 20.9 (t,
1JP-P=300 Hz, 1P), -22.5
(m, 1P), -30.8 (m, 1P), -35.6- -38.9 (m, 2P). 11
B NMR (CDCl3) δ: -15.5 (br s).
3.2.4: Synthesis of a Phosphonium Borate Ion Pair via H2 Activation
[tBu2PP(H)
tBu2]
+[HB(C6F5)3]
- (3-4) - B(C6F5)3 (50 mg, 0.20 mmol) was added to tetra-tert-
butylbiphosphine (28 mg, 0.20 mmol) in toluene (2 mL). The pink solution was placed in a
Schlenk bomb sealed with a teflon cap, subjected to 3 freeze-pump-thaw cycles and backfilled
with H2 at 77 K (generates ~4 atm at room temperature). The solution was allowed to stir
overnight and the solvent was removed in vacuo. The solid was washed with pentane (2 x 2
mL) and all volatiles were removed in vacuo to give a white solid. Yield: 64 mg (74%). Anal.
Calcd for C34H38BF15P2: C, 50.77%; H, 4.76%. Found: C, 50.34%; H, 4.68%. X-Ray quality
crystals were grown from a layered solution of CDCl3/pentane
1H NMR (CDCl3) δ: 1.46 (dd,
3JP-H=14 Hz,
4JP-H=2 Hz, 18H, C(CH3)3), 1.57 (dd,
3JP-H=15 Hz,
4JP-H=1 Hz, 18H, C(CH3)3), 3.67 (q,
1JB-H=91 Hz, 1H, BH), 5.27 (dd,
1JP-H=395 Hz,
2JP-H=8 Hz,
1H, PH). 19
F NMR (CDCl3) δ: -133.1 (br d, 3JF-F=23 Hz, 6F, o-C6F5), -164.0 (t,
3JF-F=20 Hz, 3F,
p-C6F5), -166.9 (tm, 3JF-F=23 Hz, 6F, m-C6F5).
31P{
1H} NMR (CDCl3) δ: 35.0 (d,
1JP-P=464 Hz),
69.7 (d, 1JP-P=464 Hz)
11B NMR (CDCl3) δ: -25.3 (d,
1JB-H=91 Hz).
13C NMR (CDCl3) δ: 31.4
(dd, JP-C=15 Hz, JP-C=6 Hz, C(CH3)3), 32.6 (dd, JP-C=15 Hz, JP-C=7 Hz, C(CH3)3), 36.8 (dd, JP-
C=7 Hz, JP-C=7 Hz, C(CH3)3), 37.2 (dd, JP-C=11 Hz, JP-C=7 Hz, C(CH3)3), 125.2 (br m, BC),
136.3 (dm, 1JF-C=240 Hz, C-F), 137.7 (dm,
1JF-C=240 Hz, CF), 148.3 (dm,
1JF-C=240 Hz, CF).
3.2.5: Hydrogenation and Hydrosilylation of P5Ph5
H2(Ph)P-B(C6F5)3 (3-5) – To a solution of P5Ph5 (22 mg, 0.041 mmol) in toluene (4 mL) was
added B(C6F5)3 (100 mg, 0.20 mmol). This solution was placed in a Schlenk bomb, subjected to
3 freeze-pump-thaw cycles and exposed to 1 atm of H2 at 77 K (generates ~4 atm at room
temperature). The mixture was allowed to stir overnight whereupon the solvent was removed in
vacuo and the residue was recrystallized from hexanes (2 mL) at -35°C. Analytical data matched
that previously published.173
Yield: 110 mg (91%)
46
Ph(Et2(H)Si)(H)P-B(C6F5)3 (3-6) – To a solution of P5Ph5 (11 mg, 0.020 mmol) in CH2Cl2 (4
mL) was added B(C6F5)3 (50 mg, 0.10 mmol). To this solution was added Et2SiH2 (36 mg, 0.20
mmol). The mixture was allowed to stir overnight, whereupon the solvent was removed in
vacuo. The residue was washed with pentane and all solvent was removed in vacuo to give a
white solid. Yield: 57 mg (82%). Anal. Calcd. for C28H17BF15PSi (%): C, 47.48; H, 2.42; found:
C, 47.22; H, 2.82.
1H NMR (C6D6) δ: 0.30 (m, 4H, CH2CH3), 0.51 (t,
3JH-H=8 Hz, 6H, CH2CH3), 3.83 (d,
2JP-H=29
Hz, 1H, SiH), 4.73 (d, 1JP-H =357 Hz, 1H, PH), 6.55 (td,
3JH-H=8 Hz,
4JP-H=2 Hz, 2H, m-C6H5),
6.69 (m, 3H, o-C6H5, p-C6H5). 19
F NMR (C6D6) δ: -129.8 (br s, 6F, o-C6F5), -156.2 (t, 3JF-F=21
Hz, 3F, p-C6F5), -163.5 (td, 3JF-F=21 Hz,
4JF-F=5 Hz, 6F, m-C6F5).
31P NMR (C6D6) δ: -53.7 (br
s); 11
B NMR (C6D6) δ: -12.8 (br s). 13
C{1H} NMR (C6D6) partial δ: 2.5 (d,
2JP-C =4 Hz), 3.3 (d,
J=6 Hz), 7.3 (d, J=3 Hz), 7.7 (d, J=3 Hz), 129.7 (d, 2JC-P=10 Hz, o-PC6H5), 131.1 (d,
4JC-P=3 Hz,
p-PC6H5), 133.4 (d, 3JC-P=7 Hz, m-PC6H5).
Ph(Ph2(H)Si)(H)P-B(C6F5)3 (3-7) – To a solution of P5Ph5 (22 mg, 0.041 mmol) in CH2Cl2 (4
mL) was added B(C6F5)3 (100 mg, 0.20 mmol). To this solution was added Ph2SiH2 (75 mg,
0.41 mmol). The mixture was allowed to stir overnight whereupon the solvent was removed in
vacuo and the residue was recrystallized from pentane at -35°C over one week to give a white
solid. Yield: 147 mg (94%). Anal. Calcd. for C36H17BF15PSi (%): C, 53.75; H, 2.13; found: C,
54.20; H, 2.81.
1H NMR (C6D6) δ: 5.46 (dd,
2JP-H=27 Hz,
3JH-H=5 Hz, 1H, SiH), 5.63 (1H, dd,
1JP-H=363 Hz,
3JH-
H=5 Hz, 1H, PH), 6.81 (td, 3JH-H=8 Hz, J=2 Hz, 2H), 6.95 (m, 2H), 7.09 (t, J=7 Hz, 2H), 7.29 (m,
8H), 7.70 (dd, 3JH-H=8 Hz, J=2 Hz, 2H), 7.75 (1H, dd,
3JH-H=8 Hz, J=2 Hz, 1H), 7.82 (dd,
3JH-H=8
Hz, J=2 Hz, 1H). 19
F NMR (C6D6) δ: -129.4 (d, 3JF-F=21 Hz, 6F, o-C6F5), -156.4 (br s, 3F, p-
C6F5), -163.3 (br s, 6F, m-C6F5). 31
P NMR (C6D6) δ: -47.1 (br s). 11
B NMR (C6D6) δ: -12.8 (br
s). 13
C{1H} NMR (CD2Cl2)partial δ: 128.9 (d,
2JC-P=10 Hz, o-PC6H5), 131.1 (d,
4JC-P=3 Hz, p-
PC6H5), 133.8 (d, 3JC-P=7 Hz, m-PC6H5), 135.5 (d, J=17 Hz, o-SiC6H5), 135.5 (d, J=17 Hz, o-
SiC6H5).
Ph(PhMe(H)Si)(H)P-B(C6F5)3 (3-8) – To a solution of P5Ph5 (11 mg, 0.020 mmol) in CH2Cl2 (4
mL) was added B(C6F5)3 (50 mg, 0.10 mmol). To this solution was added PhMeSiH2 (20 mg,
0.16 mmol). The mixture was allowed to stir overnight whereupon the solvent was removed and
47
the residue was washed with pentane (2 x 2 mL). Yield: 64 mg (96%). Anal. Calcd. for
C31H15BF15PSi (%): C, 50.16; H, 2.04; found: C, 50.18; H, 2.26.
1H NMR (CDCl3) δ: 0.50 (dd,
3JP-H=7 Hz,
2JH-H=4 Hz, 3H, CH3-major), 0.67 (dd,
3JP-H=6 Hz,
2JH-H
=4 Hz, 3H, CH3-minor), 4.86 (d, 2JP-H=29 Hz, 1H, Si-H), 5.03 (d,
1JP-H =357 Hz, 1H, P-Hminor),
5.13 (d, 1JP-H=363 Hz, P-Hmajor), 6.92 (dd,
3JP-H=11 Hz, J=8 Hz, 1H, o-PC6H5), 7.07 (dd,
3JP-H
=11 Hz, J=8 Hz, 1H), 7.12-7.48 (m, 8 H). 19
F NMR (CDCl3) δ: -130.0 (br s, 6F, o-C6F5), -156.6
(t, 3JF-F=27 Hz, 3F, p-C6F5 - minor), -156.7 (t,
3JF-F=27 Hz, 3F p-C6F5 - minor), -163.5 (br d,
3JF-F=17
Hz, 6F, m-C6F5). 31
P NMR (C6D6) δ: -40.9 (minor), -41.6 (major). 11
B NMR (CDCl3) δ: -13.1
(br s). 13
C{1H} NMR (CDCl3) partial δ: 128.9 (d, J=21 Hz), 129.1 (d, J=10 Hz), 129.5 (d, J=10
Hz), 131.3 (d, J=3 Hz), 131.6 (d, J=3 Hz), 131.7 (d, J=1 Hz), 131.9 (d, J=1 Hz), 135.0 (d, J=10
Hz).
Ph(Et3Si)(H)P-B(C6F5)3 (3-9) – To a solution of P5Ph5 (22 mg, 0.041 mmol) in CH2Cl2 (4 mL)
was added B(C6F5)3 (100 mg, 0.20 mmol). To this solution was added Et3SiH (50 mg, 0.43
mmol). The mixture was allowed to stir overnight whereupon the solvent was removed. X-Ray
quality crystals were grown from hexanes at -35°C. Yield: 142 mg (99%). Anal. Calcd. for
C30H21BF15PSi (%): C, 48.93; H, 2.87; found: C, 49.07; H, 3.02.
1H NMR (C6D6) δ: 0.54 (m, 15 H, CH2CH3), 4.72 (d,
1JP-H=345 Hz, 1H, PH), 6.66 (td,
3JH-H=7
Hz, 4JP-H=2 Hz, 2H, m- C6H5), 6.77 (td,
3JH-H=7 Hz, J=2 Hz, 1H, p-C6H5), 6.85 (ddd,
3JH-P=10
Hz, 3JH-H=7 Hz,
4JH-H=2 Hz, 2H, o-C6H5).
19F NMR (C6D6) δ: -129.8 (br s, 6F, o-C6F5), -156.5
(br s, 3F, p-C6F5), -163.7 (br s, 6F, m-C6F5). 31
P NMR (C6D6) δ: -46.6 (br s). 11
B NMR (C6D6)
δ: -12.3 (br s). 13
C{1H} NMR (C6D6) partial δ: 4.4 (d,
2JC-P=8 Hz, Si-CH2CH3), 6.8 (d,
3JC-P=3
Hz, Si-CH2CH3), 128.9 (d, 2JC-P=9 Hz, o-C6H5), 130.7 (d,
4JC-P=3 Hz, p-C6H5), 133.6 (d,
2JC-P=7
Hz, m-C6H5).
(Ph3Si)(Ph)HPB(C6F5)3 (3-10) – To a solution of P5Ph5 (22 mg, 0.041 mmol) in CH2Cl2 (4 mL)
was added B(C6F5)3 (100 mg, 0.20 mmol). To this solution was added 1.05 equivalents of
Ph3SiH (51 mg, 0.21 mmol). The mixture was allowed to stir overnight whereupon the solvent
was removed and the residue was washed with pentane (2 x 2 mL). Yield: 164 mg (95%). Anal.
Calcd. for C42H21BF15PSi (%): C, 57.29; H, 2.40; found: C, 57.34; H, 2.57.
48
1H NMR (C6D6) δ: 5.59 (d,
1JP-H =347 Hz, 1H, PH), 6.49 (td,
3JH-H=8 Hz, J=2 Hz, 2H, o- C6H5),
6.66 (td, 3JH-H=7 Hz, J=2 Hz, 1H, p-CH), 6.78 (ddd,
3JH-H=8 Hz,
3JH-H=10 Hz, J=2 Hz, 2H, m-
CH), 6.94 (t, 3JH-H=7 Hz, 6H, m-C6H5), 7.05 (t,
3JH-H=7 Hz, 3H, p-C6H5), 7.40 (dd,
3JH-H=8 Hz,
4JP-H=1 Hz, 6H, o-C6H5).
19F NMR (C6D6) δ: -129.2 (br s, 6F, o-C6F5), -155.1 (br s, 3F, p-C6F5),
-163.0 (br s, 6F, m-C6F5). 31
P NMR (C6D6) δ: -45.3 (br s). 11
B NMR (C6D6) δ: -11.7 (br s).
13C{
1H} NMR (C6D6) partial δ: 122.2 (d, J=40 Hz), 128.4, 130.6 (d, J=3 Hz), 131.3, 134.6 (d,
J=6 Hz), 136.4 (d, J=1 Hz).
1,4-[SiMe2(Ph)(H)PB(C6F5)3]2C6H4 (3-11) – To a solution of P5Ph5 (22 mg, 0.041 mmol) in
CH2Cl2 (4 mL) was added B(C6F5)3 (100 mg, 0.20 mmol). To this solution was added 1,4-
(Me2SiH)2C6H4 (19 mg, 0.41 mmol). The mixture was allowed to stir overnight whereupon the
solvent was removed and the residue was washed with pentane (2 x 2 mL). Yield: 134 mg
(96%). Anal. Calcd. for C58H28B2F30P2Si2 (%): C, 48.56; H, 1.97; found: C, 48.46; H, 2.18.
1H NMR (CDCl3) δ: 0.44 (d,
3JP-H=6 Hz, 6H, CH3), 0.50 (d,
3JP-H=6 Hz, 6H, CH3), 4.88 (2H, d,
1JP-H=354 Hz, 2H, PH), 6.87 (m, 4H), 7.12 (t, J=7 Hz, 4H), 7.18 (d, J=2 Hz, 4H), 7.30 (t, J=7 Hz,
2H). 19
F NMR (CDCl3) δ: -130.0 (br s, 6F, o-C6F5), -156.7 (t, 3JF-F=22 Hz, 3F, p-C6F5), -163.5
(t, 3
JF-F=21 Hz, 6F, m-C6F5). 31
P NMR (CDCl3) δ: -37.0 (br s). 11
B NMR (CDCl3) δ: -15.3 (br
s). 13
C{1H} NMR (CDCl3) partial δ: 0.1 (d, J=11 Hz), 0.7 (d, J=10 Hz), 0.8 (d, J=10 Hz), 125.1
(d, J=5 Hz), 124.6 (d, J=5 Hz), 132.4 (d, J=10 Hz), 134.5 (d, J=3 Hz), 136.9, 137.1 (d, J=6 Hz),
140.3 (dm, 1JCF=247 Hz, CF), 151.3 (dm,
1JCF=244 Hz, CF).
3.2.6: X-Ray Data Collection, Reduction, Solution and Refinement
Single crystals were mounted in paratone-n oil on a Teflon-tipped fiber. The data were collected
using the SMART software package on a Bruker SMART Apex II System CCD diffractometer
using a graphite monochromator with Mo Κα radiation (λ = 0.71073 Å). Data collection
strategies were determined using Bruker Apex software and optimized to provide >99.5%
complete data to a 2θ value of at least 55°. 10 second exposure times were used unless otherwise
noted. Data reductions were performed using the SAINT software package and absorption
corrections were applied using SADABS. The structures were solved by direct methods using
XS and refined by full-matrix least-squares on F2 using XL as implemented in the SHELXTL
suite of programs. All non-H atoms were refined anisotropically. Carbon-bound hydrogen atoms
were placed in calculated positions using an appropriate riding model and coupled isotropic
49
temperature factors. Phosphorus-bound hydrogen atoms were located in the electron difference
map and their positions refined isotropically. Single crystal X-ray structures were obtained for 3-
1, 3-2, 3-3, 3-9. Selected crystallographic data are included in Table 3.1. Diagrams and selected
bond lengths and angles are provided in Figures 3.4, 3.6, 3.7 and 3.15.
50
Table 3.1: Selected crystallographic data for compounds 3-1, 3-2, 3-3 and 3-9
Crystal 3-1-0.25 C6H6c 3-3 3-2 3-9
Formula C49.5H26.5BF15P5 C50H50BF15P4 C56H31BF15P5 C30H21BF15PSi
Formula weight 1071.87 1070.59 1154.47 736.34
Crystal system Monoclinic Monoclinic Triclinic Triclinic
Space group C2/c P21/n P-1 P-1
a(Å) 35.645(4) 15.6548(11) 12.8543(9) 9.5286(7)
b(Å) 15.6446(15) 16.7208(10) 12.9762(10) 9.8618(7)
c(Å) 25.300(3) 19.0352(12) 18.0053(12) 17.2444(13)
(o) 90.00 90.00 74.201(4) 95.866(3)
( o) 134.5620(10) 100.341(2) 76.183(4) 99.447(3)
( o) 90.00 90.00 61.367(4) 105.895(2)
V (Å3) 10052.2(17) 4901.7(5) 2515.5(3) 1518.74(19)
Z 8 4 2 2
d(calc) g cm-1
1.417 1.451 1.524 1.610
Abs coeff, , cm-1
0.272 0.247 0.278 0.245
Data collected 8858 11237 11491 5223
Data Fo2>3(Fo
2) 4672 4840 7793 3831
Variables 622 631 694 438
Ra 0.0765 0.0719 0.0449 0.0675
Rwb 0.2695 0.1375 0.1072 0.1828
Goodness of Fit 1.047 0.968 1.024 1.070
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
c These data were collected at 293 K with Mo Kα radiation (λ = 0.71069 Å).
51
3.3: Results and Discussion
3.3.1: Stoichiometric Reactions of Bulky Polyphosphines with B(C6F5)3
The bulky polyphosphines P2tBu4, P5Ph5, P4Cy4 and P4
tBu4 showed no sign of reaction
with B(C6F5)3 after 24 hours stirring at room temperature in a toluene solution. The 11
B, 19
F and
31P NMR spectra obtained were consistent with the presence of only starting materials, thus these
combinations are considered FLPs. The mixture of P2tBu4 and B(C6F5)3 was pink, such colour
change was previously observed for the mixture of PMes3 and B(C6F5)3 and has been attributed
to weak charge transfer from the Lewis base to the Lewis acid, facilitated by close approach of
the two species without adduct formation.43
3.3.2: Nucleophilic Aromatic Substitution (NAS) Reactions
Under harsh conditions and extended reaction times (up to 7 days at 120 °C) , no
evidence of NAS was observed to occur for solutions of P2tBu4, P4
tBu4 or P4Cy4 with B(C6F5)3.
This lack of reactivity can likely be attributed to the extreme bulk and rigidity of these bases.
However, over 6 days at 120°C in toluene, P5Ph5 proved to be capable of quantitative
nucleophilic aromatic substitution at the para-carbon of one of the aromatic rings of B(C6F5)3.
This reaction quantitatively produced the zwitterionic phosphonium borate 1-(P5Ph5)-(C6F4)-4-
B(F)(C6F5)2 (3-1) over 6 days at 120°C in toluene, isolated in 95% yield. The 11
B and 19
F NMR
data for this product were typical of related compounds,36
most notably the signal in the 11
B
NMR spectrum is a doublet at -2.7 ppm (1JB-F=52 Hz) and a broad signal at -193.1 ppm observed
in the 19
F NMR spectrum, characteristic of the B-F resonance (Figure 3.3). The 31
P NMR
spectrum showed a triplet resonance with very large 1 bond P-P coupling at 13.3 ppm (1JP-P=359
Hz) (Figure 3.3). This signal is typical of the resonance arising from the cationic phosphorus
centre in related [P5Ph5R+] cations.
169 3-1 was additionally characterized by X-ray
crystallography (Figure 3.4).
52
Figure 3.3: Multinuclear NMR Spectra for 3-1 in CD2Cl2: A: 31
P (resonance for cationic
phosphorus centre), B: 31
P (other 4 phosphorus resonances), C: 19
F, D: 11
B
C6F4
o-C6F5
p-C6F5
m-C6F5
BF
A B
C D 1JB-F=52 Hz
53
Figure 3.4: POV-Ray depiction of 3-1-C6H6. Carbon: black, Boron: yellow-green, Fluorine:
deep pink, Phosphorus: orange. Solvent and hydrogen atoms are omitted for clarity. Selected
metrical parameters (distances: Å, angles: °):P1-P2 2.216(2); P2-P3 2.242(2); P3-P4 2.238(2);
P4-P5 2.228(2); P5-P1 2.200(2); P1-C16 1.808(6); P1-C19 1.780(6); P2-C25 1.831(7), P3-C31
1.826(6), P4-C37 1.838(7), P5-C43 1.834(6), C16-P1-C19 107.7(3); P5-P1-P2 103.45(9); P1-P2-
P3 98.42(9); P2-P3-P4 101.85(9); P3-P4-P5 92.00(8); P4-P5-P1 89.16(8).
Structural parameters determined for the borate moiety were analogous to those of related
zwitterionic phosphonium borates.36
The catena-polyphosphinophosphonium fragment of the
molecule exhibited parameters similar to other [P5Ph5R]+ cations.
169 The ring adopts a twist
conformation in the solid state with the cationic phosphorus centre (P1) as the first of three co-
planar atoms (P1-P2-P3). The newly formed P-C bond length (P1-C16=1.808(5) Å) is similar to
those observed for the para-nucleophilic aromatic substitution of tertiary and secondary
phosphines on B(C6F5)3.36
This bond length is also similar to P-C bond lengths observed for
previously studied [P5Ph5R]+ cations. One notable difference is that the bond between the
cationic phosphorus centre and the ipso-carbon of the phenyl ring (P1-C19=1.780(6) Å) has
shortened quite dramatically in comparison to the P-C bonds for the other four phosphorus
centres of 3-1 and cationic phosphorus centres of other related compounds.169
This observation
is consistent with the increased electron-withdrawing ability of the fluoroaryl borate substituent
versus aryl or alkyl groups.
54
3.3.3: Synthesis of Alkenyl-Bridged Phosphonium Borate Zwitterions by Terminal Alkyne Activation
Due to the fact that P4Cy4 and P4tBu4 do not react with B(C6F5)3, and that zwitterion 3-1
formed only under forcing conditions it is conceivable that these Lewis bases could participate in
other reactions typical of FLPs. Accordingly, to probe further reactivity, phenyl acetylene was
added to each FLP. While the use of the P4tBu4\B(C6F5)3 FLP showed only decomposition of the
borane, the use of FLPs of both P4Cy4 and P5Ph5 with B(C6F5)3 produced quantitative
conversions to products 3-2 and 3-3, respectively. Multinuclear NMR spectroscopy indicated
that 3-2 and 3-3 were alkyne addition products (Figure 3.5).
Figure 3.5: Formation of alkyne addition products 3-2 and 3-3
The 19
F NMR spectrum showed meta-para gaps of 4.2 ppm for both 3-2 and 3-3,
characteristic of a borate anion, while the 11
B NMR spectra showed resonances at -15.9 and -15.5
ppm, respectively. The 1H NMR spectrum showed characteristic alkene C-H resonances, which
exhibit coupling to phosphorus, observed at 8.30 and 8.59 ppm, respectively. These data are
consistent with those found for the products of alkyne addition by tertiary phosphines.34
The 31
P
NMR spectra exhibited resonances for the formally cationic phosphorus centres at 20.4 and 20.9
ppm, respectively, appearing as triplets, typical of related systems.169-171
The 31
P NMR spectrum
for 3-2 exhibited three well-resolved resonances in a 1:2:1 ratio (Figure 3.6), while the spectrum
for 3-3 was much more complex due to the lack of symmetry and severe second-order effects for
all neutral phosphorus centres.
Compounds 3-2 and 3-3 were characterized crystallographically. Metrical parameters for
both were similar to those of FLP alkyne addition products34
and of [PxRxR′]+ cations.
169-171 In
3-3, the P5 ring adopts an envelope conformation, with P2 at the vertex (Figure 3.7). As is the
case with addition of other FLPs to alkynes, the FLP adds to produce the E-isomer, with the
Lewis base adding to the more substituted alkynyl carbon, which is attributed to the orientation
55
of the proposed transition state where, by analogy with the alkene activation reaction, B-C bond
formation occurs prior to P-C bond formation. Thus, B-C bond formation occurs at the more
sterically accessible terminal alkynyl carbon.69
Figure 3.6: 31
P{1H} NMR spectrum (left) and POV-Ray depiction of 3-2 (right). Carbon: black,
Boron: yellow-green, Fluorine: deep pink, Phosphorus: orange. Hydrogen atoms are omitted for
clarity. Selected metrical parameters (distances: Å, angles: °): P1-P2 2.2137(8); P2-P3 2.2190(8),
P3-P4 2.2227(9), P4-P5 2.22188(8), P5-P1 2.2077(8), P1-C20 1.821(2), P1-C27 1.801(2), P2-
C33 1.836(2); P3-C39 1.844(2); P4-C45 1.850(2); P5-C51 1.830(2); C20-P1-C27 108.32(10),
P5-P1-P2 106.83(3); P1-P2-P3 97.17(3); P2-P3-P4 103.42(3); P3-P4-P5 107.68; P4-P5-P1
101.19(3).
56
Figure 3.7: POV-Ray depiction of 3-3. Carbon: black, Boron: yellow-green, Fluorine: deep pink,
Phosphorus: orange. Hydrogen atoms are omitted for clarity. Selected metrical parameters
(distances: Å, angles: °): P1-P2 2.2106(15); P2-P4 2.2331(16); P1-P3 2.1990(15); P3-P4
2.2266(16); P1-C20 1.818(4); P1-C27 1.830(4); P2-C39 1.861(4); P3-C33 1.864(4); P4-C45
1.869(4); C20-P1-C27 111.90(19); P3-P1-P2 87.84(6); P1-P2-P4 84.39(6); P1-P3-P4 84.81(6);
P2-P4-P3 86.60(6).
3.3.4: Activation of H-H and Si-H Bonds by Polyphosphosphine/Borane FLPs
The heterolytic cleavage of H2 was investigated using FLPs based on bulky
polyphosphines with B(C6F5)3. In a typical reaction, a toluene solution of P2tBu4 and B(C6F5)3
was pressurized to 4 atm H2. Stirring overnight at room temperature resulted in quantitative
conversion to the ion pair [HP2tBu4][HB(C6F5)3] (3-4) (Figure 3.8) as determined by
multinuclear NMR spectroscopy. The 1H NMR spectrum is particularly diagnostic with 2
doublet resonances for the tert-butyl groups, a P-H resonance at 5.27 ppm which showed
coupling to both phosphorus centres (1JP-H=395 Hz,
2JP-H=8 Hz), and a quartet B-H resonance at
3.67 ppm (1JB-H=91 Hz), attributed to the HB(C6F5)3 anion.
Figure 3.8: Formation of the phosphonium borate ion pair 3-4
57
Figure 3.9: Multinuclear NMR spectra for 3-4 in CD2Cl2: 1H – showing PH and BH peaks (top
left), 11
B (top right), 19
F (bottom left), 31
P[1H} (bottom right)
The analogous reaction was conducted with the P5Ph5/B(C6F5)3 pair. Surprisingly, this
generated one equivalent of (Ph)H2P-B(C6F5)3 (3-5) and left 0.8 equivalents of the P5Ph5
polyphosphine ring in solution. This observation suggests that either only 1/5 of the P5Ph5
molecules were completely consumed, or that one phosphorus centre was abstracted from each
P5Ph5 molecule and the P4Ph4 product rapidly rearranged to P5Ph5. This was indicative of a new
reaction pathway being followed. To determine if this transformation could fully consume the
P5Ph5, the reaction was repeated using 5 equivalents of B(C6F5)3. The known primary phosphine
adduct 3-5 was formed quantitatively in situ and subsequently isolated in 91% yield (Figure
3.10).
A B 1JB-H=91Hz
PH
BH
C D
o-C6F5
p-C6F5 m-C6F5
[tBu2PP(H)
tBu2]
+
1JP-P=464 Hz
[tBu2PP(H)
tBu2]
+
1JP-P=464 Hz
58
Figure 3.10: Formation of 3-5 from the reaction of P5Ph5 with B(C6F5)3 and H2.
This unique reaction can be rationalized through a multi-step mechanistic pathway, with
literature precedent for related reactions at each stage. The H2 activation by the FLP has been
extensively demonstrated by the Stephan group3 and is assumed to be the first step. Nucleophilic
attack, a possible pathway for abstracting the cationic phosphorus centre from a catena-
cyclopolyphosphinophosphonium cation, has been demonstrated by Burford and co-workers with
PMe3 and [P4Cy4Me][OTf], (Figure 3.11).174
As the P5Ph5H+ cation is unprecedented in the
literature, even though P5Ph5 is generated in the presence of excess HCl, this cation is likely
unstable and could potentially decompose in this fashion.
Figure 3.11: Nucleophilic attack by PMe3 on the cationic phosphorus centre of P4Cy4Me+.174
Proposed steps for this reaction are shown in Figure 3.12. The reaction begins with H2
activation by the P5Ph5/B(C6F5)3 FLP, generating the catena-cyclopolyphosphinophosphonium
borohydride. The borohydride then acts as a nucleophile, abstracting the cationic centre from the
ring. One notable difference from the work by Burford and co-workers is that in the current
chemistry, the remaining cyclopolyphosphine either rearranges back to the thermodynamically
preferred 5-membered ring, or the resulting P4 ring reacts until it is consumed. As P5Ph5,
B(C6F5)3 and H2 remain present in solution the reaction can continue until one of the reagents is
completely consumed.
59
Figure 3.12: Hydrogen activation by the frustrated Lewis pair P5Ph5/B(C6F5)3 and subsequent
rearrangement to form (Ph)H2P-B(C6F5)3. As P4Ph4 is not observed in solution it must either
rearrange to form P5Ph5 or rapidly react further until it is consumed.
Given the success of the H2 activation reaction, other E-H bond activation reactions were
pursued. B(C6F5)3 is well known to catalyze the hydrosilylation of ketones,175-177
and imines178
through a proposed interaction between the borane and Si-H bond, which promotes attack by the
imine nitrogen (or carbonyl oxygen) at silicon. The resulting borohydride then attacks the alpha-
carbon, resulting in net hydrosilylation (Figure 3.13). This mechanism for the hydrosilylation
reaction is very similar to the current hydrogen activation reaction and thus a similar reaction
could be imagined with P5Ph5 as the nucleophile attacking the silane, instead of oxygen or
nitrogen. The resulting ion pair may then undergo the nucleophilic attack resulting in
phosphonium abstraction, as with the H2 reaction.
Figure 3.13: Mechanism for B(C6F5)3 catalyzed hydrosilylation of imines178
Initial reactions of the P5Ph5/B(C6F5)3 with silanes were probed using the commercially
available HSiEt3. The results observed were analogous to the reaction of the FLP with H2 as
formation of the silyl phosphine (Et3Si)(H)PhP-B(C6F5)3 (3-9) observed. When a stoichiometric
amount of borane is used, one equivalent of silylphosphine-borane adduct is formed, while the
conversion is quantitative when five equivalents of borane are used (Figure 3.14). A series of
silanes were screened, and silylphoshine borane adducts 3-6 to 3-11 were isolated in excellent
yields (Table 3.2).
60
Figure 3.14: Lewis acid promoted hydrosilylation of P5Ph5
Table 3.2: Selected NMR spectroscopic data and yields for adducts 3-6 – 3-11
Product P-H (
1JP-H)
1H
NMR δ:
11B
NMR δ:
31P NMR
δ:
Δm-p 19
F
NMR δ:a
Isolated
Yield (%)
(H)(Et2HSi)PhP-
B(C6F5)3 (3-6) 4.73 (357 Hz) -12.8 -53.7 7.3 82
(H)(Ph2HSi)PhP-
B(C6F5)3 (3-7) 5.63 (363 Hz) -12.8 -47.1 6.9 94
(H)(MePhHSi)PhP-
B(C6F5)3 (3-8)
5.03 (357 Hz)
5.13 (363 Hz)
-13.1 -40.9
-41.6
6.9
5.9
96
(H)(Et3Si)PhP-
B(C6F5)3 (3-9) 4.72 (345 Hz) -12.3 -46.6 7.2 99
(H)(Ph3Si)PhP-
B(C6F5)3 (3-10) 5.59 (347 Hz) -11.7 -45.3 7.9 95
1,4-[(C6F5)3B-
PPh(H)]2-C6H4 (3-11) 4.88 (354 Hz) -15.3 -37.0 6.8 96
aThis value is the difference in chemical shift between the meta and para-fluorines, and has been
noted to be characteristic for different bonding environments at boron in fluoroarylboranes (the
shortest m-p gaps are generally found for 4 coordinate fluoroaryl borates, while the largest m-p
gaps are found for neutral 3 coordinate fluoroaryl boranes).179,180
Adduct 3-9 was characterized crystallographically (Figure 3.15). The metrical
parameters were as expected, with the P-B34,38
and Si-P181-185
bond lengths comparable to those
observed in secondary phosphine-B(C6F5)3 adducts and silyl phosphines, respectively.
61
Worthy of note is that these reactions create a new chiral centre at phosphorus. While no
enantioselectivity could be promoted in the present reactions, the proposed mechanism indicates
that the use of a chiral Lewis acid may allow for enantioselective reactions. This would produce
chiral phosphines which could be extremely useful for asymmetric catalysis.186-190
For adduct 3-
8 where a chiral centre is created at silicon and phosphorus, slight diastereoselective
enhancement is observed at room temperature, as the 2 products are observed in a 5:4 ratio by
1H,
19F and
31P NMR spectroscopy.
Figure 3.15: POV-Ray depiction of 3-9. Carbon: black, Hydrogen: white, Boron: yellow-green,
Fluorine: deep pink, Phosphorus: orange, Silicon: pink. Carbon-bound hydrogen atoms are
omitted for clarity. Selected metrical parameters (distances: Å, angles: °): P1-B1 2.093(6); P1-
Si1 2.333(2).
Adducts 3-6 to 3-8 retain one Si-H bond; however this Si-H bond is not capable of further
reaction to form a silyl-bisphosphine. This would have provided an attractive route towards a
new family of silyl-bridged phosphines which could be used as bidentate ligands. To
demonstrate multifunctional reactivity in these systems, the reaction with 1,4(Me2SiH)2C6H4
with 2 equivalents of B(C6F5)3 and 2/5 P5Ph5 produces the bisborane adduct of the 1,4-bis-
silylphosphine (3-11) (Figure 3.16). This suggests that while this reactivity is dependent on
steric bulk, the methodology can be used in the synthesis of new bisphosphines and potentially
new phosphorus-containing macrocycles or oligomers (or potentially even polymers) through the
use of bis-silanes and bis-boranes in a similar reaction.
62
Figure 3.16: Synthesis of 3-11
3.3.5: Scope of Reactivity in Terms of Bulk at Silicon and Lewis Acidity of the Borane
The B(C6F5)3-promoted oxidative addition to phosphorus(I) is fundamentally interesting,
however catalytic transformation of this polyphosphine would be especially useful, as this would
provide an efficient method for the synthesis of novel phosphines which could be used in a wide
variety of applications. The major obstacle to achieving catalytic reactivity is the strong Lewis
acid-base adduct formed between the secondary silylphosphine and B(C6F5)3. In efforts to
circumvent this issue, bulkier silanes iPr3SiH and
tBu2SiH2 were used with the hope that the
increased bulk would prevent or at least hinder the ability of the silylphosphine product to
coordinate to B(C6F5)3. This could allow for the free borane to turn over catalytically. Studies
employing these silanes showed no reaction with P5Ph5 and B(C6F5)3 over 24 hours at room
temperature as their steric bulk must prevent this reactivity. This further supports the notion that
adducts 3-6 – 3-8 will not react with a second equivalent of phosphine and borane due to steric
congestion. Another potential avenue to hinder the silylphosphine-borane adduct formation is to
diminish the Lewis acidity at boron. This was investigated by employing BEt3, BPh3 and BMes3
as the Lewis acid. These boranes showed no interaction at room temperature in toluene with
P5Ph5 by multinuclear NMR spectroscopy, however no reaction was observed upon addition of
H2 or Et3SiH to these FLPs. This observation suggests that these boranes are not Lewis acidic
enough to activate the H-H or Si-H bond, thus preventing reactivity.
63
3.4: Conclusions
Frustrated Lewis pair chemistry utilizing polyphosphines can be exploited to synthesize
novel catena-polyphosphinophosphonium borate zwitterions by known reactions of nucleophilic
aromatic substitution, by addition to alkynes and by H2 activation. In addition, a new class of
reactivity for these species has been uncovered: the controlled fragmentation of P5Ph5 to primary
phosphine and silyl phosphine-B(C6F5)3 adducts. The zwitterionic phosphonium borates 3-1 – 3-
3 possess similar properties to the observed moieties in related species, however, they
incorporate B-F (3-1) or alkenyl (3-2 – 3-3) functional groups, previously unavailable in catena-
cyclopolyphosphinophosphonium cations. The adducts 3-5 – 3-11 were isolated in excellent
yields through the straightforward fragmentation pathway involving P5Ph5, 5 equivalents of
B(C6F5)3 and excess silane (or H2). The use of a chiral borane may allow for the enantioselective
synthesis of these chiral secondary silyl phosphines which would otherwise be difficult to access.
The unique fragmentation pathway observed here may be useful in related chemistry, possibly
with white phosphorus.
64
Chapter 4: Frustrated Lewis Pairs: Synthesis and Reactivity of Covalently-Bound Phosphinoboranes
4.1: Introduction
Combinations of group 13 Lewis acids and group 15 Lewis bases are of significant
interest. In addition to unique multiple bonding modes, their potential use as hydrogen storage
materials has become increasingly attractive in light of current issues with fossil fuels and the
quest for alternative energies. Much of this research has been focused on ammonia borane, H3N-
BH3, due to its high hydrogen content by mass, room temperature stability and commercial
availability.191
Ammonia-borane and other related amine-borane or phosphine-borane adducts
have been shown to liberate varying amounts of H2 under thermal duress192
or catalytically,
using transition metal,10,80-97,193-196
Lewis acid191
or Lewis base catalysts197
(see Figure 4.1 for an
example).
Figure 4.1: Dehydrogenation of ammonia borane by a nickel(0)carbene catalyst196
One of the main barriers to the use of such compounds in hydrogen storage is the
thermodynamically downhill pathway for the loss of H2,191
meaning that the hydrogenation of
these products would be strongly endothermic. Thus, the hydrogenation of dehydrogenated
ammonia borane has proven to be challenging and while several reports exist illustrating
potential steps in the reaction,191,198-200
none can hydrogenate these spent materials efficiently
and cost-effectively. Catalytic hydrogenation of these materials could potentially introduce
detrimental impurities into the hydrogen storage material in addition to increased costs.
While these problems are difficult to surmount for the addition of H2 to dehydrogenated
ammonia borane, perhaps judicious modification of the hydrogen storage material could yield a
65
compound which, after dehydrogenated, could readily re-add hydrogen. The un-catalyzed
addition of H2 to a main group compound has been previously observed for germynes, giving a
mixture of products including primary germanes.201
As discussed in section 1.4.2, main group frustrated Lewis pairs have been shown to
activate hydrogen and even act as catalysts in catalytic hydrogenation reactions (section 1.4.7).
In light of this unprecedented reactivity, other main group element systems were envisioned to
continue the fundamental study of these systems in an effort to effect novel metal free bond
activation and potential hydrogen storage systems. Previous computational and experimental
studies on directly bound phosphinoboranes suggest that there is limited interaction between the
lone pair at phosphorus and the vacant p-orbital on boron due to incompatible orbital energy and,
in some cases, geometry.202,203
The lack of reaction between the Lewis acidic and Lewis basic
sites is a key element in the reactivity of FLPs.4 A covalently-bound phosphine-borane system
would require some similarity to these systems, in that substituents at P and B would have to be
large enough to prevent dimerization or the formation of larger aggregates by Lewis acid-base
adduct formation. This system would have to also be “electronically frustrated” in order to
activate H2. In other words, the lone pair on phosphorus and the vacant p-orbital cannot fully
quench each other intramolecularly through π-bonding (Figure 4.2). This P-B π-bonding
interaction should be relatively weak due to the relatively poor orbital overlap between 2nd
and
3rd
row elements discussed above.
Figure 4.2: Potential Reactivity of R2PB(C6F5)2: dimerization (top), H2 activation (bottom)
66
4.2: Experimental
4.2.1: General Considerations
All preparations were done under an atmosphere of dry, O2-free N2 employing both Schlenk line
techniques and an Innovative Technologies or Vacuum Atmospheres inert atmosphere glove box.
Solvents (pentane, hexanes, toluene, and methylene chloride) were purified employing a Grubbs’
type column systems manufactured by Innovative Technology and stored over molecular sieves
(4 Å). Molecular sieves (4 Å) were purchased from Aldrich Chemical Company and dried at 140
ºC under vacuum for 24 hours prior to use. Deuterated solvents were dried over
Na/benzophenone (C6D6, C7D8) or CaH2 (CD2Cl2, CDCl3) and distilled prior to use. All common
organic reagents were purified by conventional methods unless otherwise noted. 1H,
13C,
11B,
19F
and 31
P nuclear magnetic resonance (NMR) spectroscopy spectra were recorded on a Bruker
Avance-300 or Avance-400 spectrometer at 300K unless otherwise noted. 1H and
13C NMR
spectra are referenced to SiMe4 using the residual solvent peak impurity of the given solvent. 11
B
and 19
F NMR experiments were referenced to 15% BF3-Et2O in CDCl3 and 31
P NMR
experiments were referenced to 85% H3PO4. Chemical shifts are reported in ppm and coupling
constants in Hz as absolute values. Combustion analyses were performed in house employing a
Perkin Elmer CHN Analyzer. Silanes were purchased from Strem Chemicals and used as
received. Phenyl acetylene was purchased from Aldrich Chemicals and used as received. H2
was passed through a dririte column prior to use. R2PLi (R=Et, Ph, tBu, Cy, Mes) were prepared
by treating the corresponding phosphine with 1.1 equivalents of tBuLi in toluene and collecting
the precipitate.
4.2.2: Synthesis of Phosphinoboranes R2PB(C6F5)2
[Et2PB(C6F5)2]2- (4-1) - To a slurry of Et2PLi (51 mg, 0.53 mmol) in toluene (5 mL) was added a
solution of (C6F5)2BCl (200 mg, 0.53 mmol) in toluene (5 mL) at -35°C. The mixture was
allowed to stir overnight and was then run through celite. The filtrate was concentrated to ~2 mL
and stored at -35 °C overnight. The solution was dried in vacuo and washed with cold pentane
(2 x 2 mL). Et2PB(C6F5)2 was isolated as a colourless polycrystalline solid. Yield: 192 mg
(84%). Anal. Calcd. for C16H10BF10P: C, 44.28; H, 2.32; Found: C, 44.66; H, 2.64. Crystals
were grown from a pentane solution at room temperature.
67
1H NMR (CD2Cl2) δ: 1.07 (6H, dt,
3JP-H=16 Hz,
3JH-H=8 Hz, CH3), 2.16 (4H, m, CH2).
19F NMR
(CD2Cl2) δ: -125.0 (d, 3JF-F=20 Hz, 4F, o-C6F5), -153.5 (t,
3JF-F=23 Hz, 2F, p-C6F5), -160.3 (t,
3JF-
F=20 Hz, 4F, m-C6F5). 31
P NMR (CD2Cl2) δ: -23.4 (br m). 11
B NMR (CD2Cl2) δ:-12.9 (t, 1J P-B
=72 Hz). 13
C{1H} NMR (CD2Cl2) partial δ: 8.3 (CP), 16.2 (m, CH3), 137.4 (dm,
1JC-F=248 Hz,
CF), 140.4 (dm, 1JC-F =209 Hz, CF), 147.3 (dm,
1JC-F =227 Hz, CF).
[Ph2PB(C6F5)2]2 (4-2) - To a slurry of Ph2PLi (101 mg, 0.53 mmol) in toluene (5 mL) was added
a solution of (C6F5)2BCl (200 mg, 0.53 mmol) in toluene (5 mL) at -35°C. The mixture was
allowed to stir for 3 hours and was then run through celite. The filtrate was concentrated to ~2
mL and 2 mL of hexanes were added. The solution was decanted and the precipitate was dried
in vacuo. Yield: 184 mg (65%). Anal. Calcd. for C24H10BF10P: C, 54.38; H, 1.90; Found: C,
55.32; H, 2.30. Crystals were grown by slow evaporation from a 1:1 dichloromethane:pentane
solution.
1H NMR (CD2Cl2) δ: 7.26 (4H, t,
3JH-H=8 Hz, m-C6H5), 7.41 (6H, m, o-C6H5, p-C6H5).
19F
NMR (CD2Cl2) δ: -121.3 (s, 4F, o-C6F5) -156.3 (t, 3JF-F=20 Hz, 2F, p-C6F5), -164.2 (m, 4F, m-
C6F5). 31
P NMR (CD2Cl2) δ: -0.8 (s). 11
B NMR (CD2Cl2) δ:-2.2 (t, 1
J P-B=66 Hz). 13
C{1H}
NMR (CD2Cl2) partial δ: 127.8 (m, C6H5), 129.2 (d, 1JP-C=32 Hz, PC), 130.8 (C6H5), 134.3
(C6H5), 137.0 (dm, 1JC-F =228 Hz, CF), 140.3 (dm,
1JC-F =242 Hz, CF), 146.8 (dm,
1JC-F=239 Hz,
CF).
tBu2PB(C6F5)2 (4-3) - To a slurry of
tBu2PLi (80 mg, 0.53 mmol) in toluene (5 mL) was added a
solution of (C6F5)2BCl (200 mg, 0.53 mmol) in toluene (5 mL). The mixture was allowed to stir
for 3 hours and was then run through celite. The filtrate was stored at -35 °C overnight. The
solution was decanted and dried in vacuo. Yield: 160 mg (61%). Anal. Calcd. for C20H18BF10P:
C, 49.01; H,3.70; Found: C, 48.16; H, 3.54. Crystals were grown from hexanes at -35°C.
1H NMR (CD2Cl2) δ: 1.40 (d,
3JP-H=15 Hz).
19F NMR (CD2Cl2) δ: -130.7 (s, 4F, o-C6F5), -156.0
(t, JF-F=23 Hz, 2F, p-C6F5), -163.4 (m, 4F, m-C6F5). 31
P NMR (CD2Cl2) δ: 120.7 (br m). 11
B
NMR (CD2Cl2) δ: 41.8 (d, 1J P-B=150 Hz).
13C{
1H} NMR (CD2Cl2) partial δ: 32.9 (CH3), 39.6
(d, 1JP-C=23 Hz, PC), 115.9 (m, BC), 137.5 (dm,
1JC-F =253 Hz, CF), 140.9 (dm,
1JC-F =253 Hz,
CF), 143.0 (dm, 1JC-F =239 Hz, CF).
68
Cy2PB(C6F5)2 (4-4) - To a slurry of Cy2PLi (107 mg, 0.53 mmol) in toluene (5 mL) was added a
solution of (C6F5)2BCl (200 mg, 0.53 mmol) in toluene (5 mL). The mixture was allowed to stir
for 18 hours and was then run through celite. The solution was decanted and the solid was dried
in vacuo and washed with cold hexanes (-35°C, 2 mL). Yield: 235 mg (83%). Anal. Calcd. for
C24H22BF10P: C, 53.16; H, 4.09; Found: C, 52.28; H, 4.20. Crystals were grown from hexanes at
-35°C.
1H NMR (CD2Cl2) δ: 1.15 (tt, J=13, J=3, 2H), 1.25 (m, J=13 Hz, J=3 Hz, 4H), 1.49 (m, J=13 Hz,
JP-H=5 Hz, J=3 Hz, 4H), 1.64 (d, J=13 Hz, 2H), 1.75 (dd, J=13 Hz, JP-H=3 Hz, 4H), 1.99 (d, J=13
Hz, 4H), 2.33 (dtt, J=13 Hz, JP-H=9 Hz,, J=3 Hz, 2H). 19
F NMR (CD2Cl2) δ: -130.78 (s, 4F, o-
C6F5), -155.46 (t, 3JF-F=20 Hz, 2F, p-C6F5), -163.51 (m, 4F, m-C6F5).
31P NMR (CD2Cl2) δ: 92.1
(br m). 11
B NMR (CD2Cl2) δ: 39.5 (d, 1JB-P=142 Hz).
13C{
1H} NMR (CD2Cl2) δ: 25.4 (C6H11),
26.8 (d, 2JC-P=34 Hz, C6H11), 33.7 (d,
3JC-P=4 Hz, C6H11), 35.0 (d,
1JC-P=27 Hz, PC), 113.1 (BC),
137.6 (dm, 1JC-F =260 Hz, CF), 141.0 (dm,
1JC-F=264 Hz, CF), 145.2 (dm,
1JC-F =247 Hz, CF).
Mes2PB(C6F5)2 (4-5) - To a slurry of Mes2PLi (145 mg, 0.53 mmol) in toluene (5 mL) was added
a solution of (C6F5)2BCl (200 mg, 0.53 mmol) in toluene (5 mL). The mixture was allowed to
stir for 18 hours and was then run through celite. The solution was dried in vacuo leaving a
yellow oil which subsequently crystallized. The yellow crystals were washed with cold hexanes
(-35°C, 2 mL). Yield: 220 mg (68%). Anal. Calcd. for C30H22BF10P: C, 58.66; H, 3.61; Found:
C, 57.51; H, 3.53.
1H NMR (CD2Cl2) δ: 2.25 (s, 6H, p-CH3), 2.29 (s, 12H, o-CH3), 6.89 (d,
4JP-H=6 Hz, 4H, CH).
19F NMR (CD2Cl2)δ: -131.2 (s, 4F, o-C6F5), -154.6 (t,
3JF-F=20 Hz, 2F, p-C6F5), -163.5 (m, 4F,
m-C6F5). 31
P NMR (CD2Cl2, 121 MHz) δ: 29.3 (br m). 11
B NMR (CD2Cl2) δ: 40.1 (br m).
13C{
1H} NMR (CD2Cl2) partial δ: 20.9 (p-CH3), 22.6 (d,
3JC-P =7.7 Hz, o-CH3), 123.1 (d, JC-P
=72 Hz, PC), 129.3 (d,3 JC-P =11 Hz, CH), 137.2 (
1JC-F=246 Hz, CF), 141.2 (d,
1JC-F=252 Hz,
CF), 141.4 (d, 4JC-P =3 Hz, p-CCH3), 143.6 (d,
2JC-P=7 Hz, o-CCH3), 146.0 (d,
1JC-F=248 Hz, CF).
4.2.3: Synthesis of Secondary Phosphine Adducts of HB(C6F5)2
Et2(H)PB(H)(C6F5)2 (4-6) - A solution of Et2PH (12 mg, 0.15 mmol) in toluene (1 mL) was
added to (C6F5)2BH (50 mg, 0.15 mmol) in hexanes (2 mL). The mixture was stirred for 1 hour,
then stored at -35°C for 2 hours. The solution was decanted and the white precipitate was dried
69
in vacuo. Yield: 48 mg (77%). Anal. Calcd. for C16H12BF10P: C, 44.07; H, 2.77; Found: C,
43.80; H, 2.73. Crystals were grown from hexanes at -35°C.
1H NMR (CD2Cl2) δ: 1.17 (dt,
3JH-P=17 Hz,
2JH-H =8 Hz, 6H, CH3), 1.84 (dq,
2JP-H =24 Hz,
2JP-
H=8 Hz, 4H, CH2), 3.43 (br m, 1H, BH), 4.95 (dm, 1JP-H=388 Hz, 1H, PH).
19F NMR (CD2Cl2)
δ: -131.7 (s, 4F, o-C6F5), -159.2 (t, 3JF-F=20 Hz, 2F, p-C6F5), -164.7 (m, 4F, m-C6F5).
31P NMR
(CD2Cl2) δ: -4.6 (br m). 11
B NMR (CD2Cl2) δ: -30.0 (d, 1JB-P=65 Hz).
13C{
1H} NMR (CD2Cl2)
partial δ: 8.5 (d, 1JC-P =6 Hz, CP), 10.6 (d,
2JC-P =38.5 Hz, CH3), 137.1 (dm,
1JC-F =208 Hz, CF),
140.3 (dm, 1JC-F =235 Hz, CF) 148.0 (dm,
1JC-F=233 Hz, CF).
Ph2(H)PB(H)(C6F5)2 (4-7) - A solution of Ph2PH (28 mg, 0.15 mmol) in toluene (1 mL) was
added to (C6F5)2BH (50 mg, 0.15 mmol) in toluene (1 mL). The mixture was stirred for 1 hour,
upon which hexanes (2 mL) was added and the solution was decanted. The white precipitate was
washed with hexanes (2 x 2 mL) and the remaining solid was dried in vacuo. Yield: 58 mg
(74%). Anal. Calcd. for C24H12BF10P: C, 54.17; H, 2.27; Found: C, 53.82; H, 2.25. Crystals
were grown from 1:1 dichloromethane:hexanes at -35°C.
1H NMR (CD2Cl2) δ: 3.94 (br m, 1H, BH), 6.81 ( ddm,
1JP-H=409 Hz,
3JH-H=15 Hz, 1H, PH),
7.44 (ddd, 3JH-H=8 Hz,
3JH-H=6 Hz,
4JH-P =1 Hz, 4H, m-C6H5), 7.54 (tt,
3JH-H=6 Hz,
4JH-H =2 Hz,
2H, p-C6H5), 7.60 (ddd, 3JH-P=12 Hz,
3JH-H =8 Hz,
4JH-H =2 Hz, 3H, o-C6H5).
19F NMR (CD2Cl2)
δ: -131.2 (s, 4F, o-C6F5), -158.8 (t, 2F, 3JF-F J=20 Hz, p-C6F5), -164.7 (m, 4F, m-C6F5).
31P
NMR (CD2Cl2,) δ: -1.5 (br m). 11
B NMR (CD2Cl2) δ: -28.5 (d, 1JP-B=71 Hz).
13C{
1H} NMR
(CD2Cl2) partial δ: 122.9 (d, 1JC-P=65 Hz, CH), 129.5 (dm,
2JC-P=164 Hz, CH), 133.7(CH), 135.3
(CH) 136.9 (dm, 1
JC-F=248 Hz, CF), 140.7 (dm, 1JC-F=254 Hz, CF), 148.0 (dm,
1JC-F=235 Hz,
CF).
tBu2(H)PB(H)(C6F5)2 (4-8) - A solution of
tBu2PH (22 mg, 0.15 mmol) in hexanes (2 mL) was
added to (C6F5)2BH (50 mg, 0.15 mmol) in hexanes (2 mL). The mixture was stirred for 1 hour,
upon which the solution was concentrated to 2 mL and stored at -35°C overnight. The solution
was decanted and the remaining solid was dried in vacuo. Yield: 48 mg (65%). Anal. Calcd.
for C20H20BF10P: C, 48.81; H, 4.10; Found: C, 48.42; H, 3.90. Crystals were grown from the
hexane wash layer.
70
Alternate synthesis A: An J Young’s tube was charged with tBu2PB(C6F5)2 (20 mg, 0.041 mmol)
and tol-d8 (0.75 mL) the solution was subjected to 3 freeze-pump-thaw cycles and 1 atm of H2
was added at 77 K (~4 atm at room temperature). 80% conversion to tBu2(H)P-B(H)(C6F5)2 was
achieved in four weeks at 25°C, while quantitative conversion was achieved over 48 hours at
60°C.
Alternate synthesis B: An NMR tube was charged with of tBu2PB(C6F5)2 (20 mg, 0.041 mmol),
Me2NH-BH3 (2 mg, 0.04 mmol) and CD2Cl2 (0.75 mL) After 15 minutes at 25°C, 1H,
11B,
31P
and 19
F NMR spectroscopy revealed quantitative conversion to tBu2(H)P-B(H)(C6F5)2 and 0.5
(Me2N-BH2)2.
1H NMR (CD2Cl2) δ: 1.27 (d,
3JH-P =14 Hz, 18H, CH3), 3.48 (br m, 1H, BH), 4.84 (dd,
1JH-P=375
Hz, 3JH-H=11 Hz, 1H, PH).
19F NMR (CD2Cl2) δ: -129.7 (s, 4F, o-C6F5), -159.4 (t,
3JF-F=20 Hz,
2F, p-C6F5), -164.7 (m, 4F, m-C6F5). 31
P NMR (CD2Cl2) δ: 32.0 (br m). 11
B NMR (CD2Cl2) δ:
-30.0 (d, 1JP-B=48 Hz).
13C{
1H} NMR (CD2Cl2) δ: 29.2 (CH3), 33.0 (d,
1JC-P =29 Hz, CP), 117.9
(br m, BC), 137.1 (dm, 1JC-F =255 Hz, CF), 139.6 (dm,
1JC-F=250 Hz, CF), 148.0 (dm,
1JC-F =239
Hz, CF).
Cy2(H)PB(H)(C6F5)2 (4-9) - A solution of Cy2PH (29 mg, 0.15 mmol) in hexanes (2 mL) was
added to (C6F5)2BH (50 mg, 0.15 mmol) in hexanes (2 mL). The reaction was stirred for 1 hour,
upon which the solution was decanted and the white precipitate was washed with hexanes (2 x 2
mL) and the remaining solid was dried in vacuo. Yield: 61 mg (77%). Anal. Calcd. for
C24H24BF10P: C, 52.97; H, 4.45; Found: C, 52.50; H, 4.56. Crystals were grown from hexanes
at -35°C.
Alternate synthesis A: An J Young’s tube was charged with 20 mg Cy2PB(C6F5)2 (20 mg, 0.037
mmol) and tol-d8 (0.75 mL), the solution was subjected to 3 freeze-pump-thaw cycles and 1 atm
of H2 was added at 77 K (~4 atm at room temperature). Quantitative conversion to Cy2(H)P-
B(H)(C6F5)2 was achieved in two weeks at 25°C or 48 hours at 60°C.
Alternate synthesis B: An NMR tube was charged with Cy2PB(C6F5)2 (20 mg, 0.037 mmol),
Me2NH-BH3 (2 mg, 0.04 mmol) and 0.75 mL CD2Cl2 (0.75 mL). After 15 minutes at 25°C, 1H,
11B,
31P and
19F NMR spectroscopy revealed quantitative conversion to Cy2(H)P-B(H)(C6F5)2
and 0.5 (Me2N-BH2)2.
71
1H NMR (CD2Cl2) δ: 1.15 (m, 2H, PC6H11), 1.20-1.29 (br m, 6H, PC6H11), 1.37 (m, 2H,
PC6H11), 1.68 (br d, 2JH-H=13 Hz, 2H, PC6H11), 1.75-1.84 (br m, 6H, PC6H11), 1.89 (m, 2H,
PC6H11), 2.00 (m, 2H, PC6H11), 3.33 (1H, br m, BH), 4.78 (ddm, 1JP-H=381 Hz,
3JH-H=13 Hz,1H,
PH). 19
F NMR (CD2Cl2) δ: -131.0 (s, 4F, o-C6F5), -159.4 (t, 3JF-F=20 Hz, 2F, p-C6F5), -164.7 (t,
3JF-F=17 Hz, 4F, m-C6F5).
31P NMR (CD2Cl2) δ: 7.1 (br m).
11B NMR (CD2Cl2) δ: -28.1 (d,
1JP-
B=68 Hz). 13
C{1H} NMR (CD2Cl2) partial δ: 25.4(CH2), 26.7 (m, CH2), 29.2, (d, J=17 Hz, CH2),
29.7 (d, J=35 Hz, CH), 136.6 (dm, 1JC-F=185 Hz, CF), 148.0 (dm,
1JC-F=237 Hz, CF).
Mes2(H)PB(H)(C6F5)2 (4-10) - A solution of Mes2PH (40 mg, 0.15 mmol) in hexanes (2 mL)
was added to (C6F5)2BH (50 mg, 0.15 mmol) in hexanes (2 mL). The reaction was stirred for 1
hour, upon which the solution was concentrated to 2 mL and stored at -35°C overnight. The
solution was decanted and the remaining solid was dried in vacuo. Yield: 70 mg (78%). Anal.
Calcd. for C30H24BF10P: C, 58.47; H, 3.93; C, 57.82; H, 3.91
1H NMR (CD2Cl2) δ: 2.24 (s, 12H, o-CH3), 2.26 (s, 6H, p-CH3), 6.88 (s, 4H, CH), 7.06 (dd,
1JH-P
=402 Hz, 3JH-H=13 Hz, 1H, PH).
19F NMR (CD2Cl2) δ: -131.4 (s, 4F, o-C6F5), -158.7 (t,
3JF-F=20
Hz, 2F, p-C6F5), -164.8 (m, 4F, m-C6F5). 31
P NMR (CD2Cl2) δ: -39.0 (br m). 11
B NMR
(CD2Cl2) δ: -25.4 (d, 1JP-B =48 Hz).
13C{
1H} NMR (CD2Cl2) partial δ: 20.7 (CH3), 21.6 (CH3),
118.0 (d, 1JC-P =58 Hz, CP), 130.5 (d, J=9 Hz, CH), 136.9 (dm,
1JC-F=245 Hz, CF), 142.4 (p-C-
CH3), 143.0 (d, J=8 Hz, o-C-CH3), 148.4 (dm, 1JC-F=240 Hz, CF).
4.2.4: Reactions of R2PB(C6F5)2 with 4-tert-butylpyridine
A solution of R2PB(C6F5)2 (25 mg) in CDCl3 (0.75 mL) was added to 4-tert-butylpyridine (one
equivalent). The solution was monitored by multinuclear NMR. Using toluene (1 mL) as the
solvent gave X-ray quality crystals for R=Cy. Rapid decomposition precluded further
characterization by elemental analysis and 13
C NMR spectroscopy. While 1H,
19F and
31P NMR
spectra for R=tBu indicated that little reaction had taken place, the clear emergence of a new
peak at -1.3 ppm in the 11
B NMR spectrum indicated the presence of B-N adduct 4-11 in at least
trace amounts. For R=Mes, no clear evidence of adduct formation was observed and the reaction
showed numerous unidentifiable products after several days at room temperature.
72
1H NMR (CDCl3) δ: 0.41-2.00 (m, 22H, C6H11), 1.41 (s, 9H, C(CH3)3), 7.32 (d,
3JH-H=7 Hz, 2H,
CH), 9.05 (br s, 2H, CH). 19
F NMR (CDCl3) δ: -128.0 (br s, 4F, o-C6F5), -157.2 (br s, 2F, p-
C6F5), -162.2 (m, m-C6F5). 31
P NMR (CDCl3) δ: -28.3 (br s). 11
B NMR (CDCl3) δ: -1.3.
4.2.5: Synthesis of Dimers (R2PBCl2)2 and ClB(C6F5)2 by Reaction of BCl3 with R2PB(C6F5)2
BCl3 (one equivalent) is added to a solution of R2PB(C6F5)2 (25 mg) in toluene. The solution
was allowed to stir for 3 hours. Volatiles were then removed in vacuo and the residue was taken
up in CDCl3 for NMR spectroscopy. Resonances attributed to ClB(C6F5)2 corresponded to those
previously reported in the literature.204,205
In the case of 4-13, trace evidence of monomer
tBu2PBCl2 was observed in the
11B NMR spectrum.
4-13 - 1H NMR (CDCl3) δ: 1.33 (d,
3JP-H=15 Hz, 9H, C(CH3)3).
31P NMR (CDCl3) δ: -14.8
(sept, 1JP-B=90 Hz).
11B NMR (CDCl3) δ: 4.2 (t,
1JP-B=90 Hz, (
tBu2PBCl2)2), 2.7 (d,
1JP-B=135
Hz, tBu2PBCl2). 4-14 -
1H NMR (CDCl3) δ: 1.11-2.58 (m, 22H, Cy).
31P NMR (CDCl3) δ: -14.5
(sept, 1JP-B=99 Hz).
11B NMR (CDCl3) δ: -0.5 (t,
1JP-B=99 Hz).
4.2.6: X-Ray Data Collection, Reduction, Solution and Refinement
Single crystals were mounted in thin-walled capillaries either under an atmosphere of dry N2 in a
glove box and flame sealed or coated in paratone-N oil. The data were collected using the
SMART software package on a Siemens SMART System CCD diffractometer using a graphite
monochromator with Mo Κα radiation (λ = 0.71073 Å). A hemisphere of data was collected in
1448 frames with 10 second exposure times unless otherwise noted. Data reductions were
performed using the SAINT software package and absorption corrections were applied using
SADABS. The structures were solved by direct methods using XS and refined by full-matrix
least-squares on F2 using XL as implemented in the SHELXTL suite of programs. All non-H
atoms were refined anisotropically. Carbon-bound hydrogen atoms were placed in calculated
positions using an appropriate riding model and coupled isotropic temperature factors.
Phosphorus-bound hydrogen atoms were located in the electron difference map and their
positions refined isotropically. Single crystal X-ray structures were obtained for 4-1, 4-2, 4-3, 4-
4, 4-6, 4-7, 4-8, 4-12, 4-13 and 4-14 Selected crystallographic data are included in Tables 4.1 to
4.4 . Diagrams and selected bond lengths and angles are provided in Figures 4.5, 4.7, 4.10, 4.11,
4.14, 4.15 and 4.18.
73
Table 4.1: Selected crystallographic data for compounds 4-1, 4-2 and 4-3
Crystal 4-1 4-2-CH2Cl2 4-3
Formula C32H20P2F20B2 C50H22P2F20B2Cl2 C20H18BF10P
Formula weight 868.04 1131.10 490.12
Crystal system Triclinic Triclinic Orthorhombic
Space group P-1 P-1 Pbca
a(Å) 9.699(2) 10.3197(16) 12.194(9)
b(Å) 9.703(2) 12.0982(19) 18.331(13)
c(Å) 10.404 20.926(3) 19.774(14)
(o) 67.072(2) 74.492(2) 90.00
( o) 80.770(3) 76.700(2) 90.00
( o) 67.852(2) 68.458(2) 90.00
V (Å3) 835.1(3) 2316.1(6) 4420(5)
Z 1 2 8
d(calc) g cm-1
1.726 1.622 1.473
Abs coeff, , cm-1
0.269 0.327 0.212
Data collected 2933 8128 3891
Data Fo2>3(Fo
2) 2274 3801 2640
Variables 253 676 289
Ra 0.0390 0.0573 0.0553
Rwb 0.0942 0.1347 0.1889
Goodness of Fit 1.035 0.919 1.059
These data were collected at 293 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
74
Table 4.2: Selected crystallographic data for compounds 4-4, 4-6 and 4-7
Crystal 4-4 4-6 4-7
Formula C24H22BF10P C16H12BF10P C24H22BF15P
Formula weight 542.20 436.04 542.20
Crystal system Triclinic Triclinic Triclinic
Space group P-1 P-1 P-1
a(Å) 9.4503(13) 9.198(6) 8.9820(6)
b(Å) 10.4530(14) 12.989(8) 10.3690(6)
c(Å) 13.7545(19) 15.681(9) 12.4150(6)
(o) 110.6870(10) 83.854(6) 106.381(3)
( o) 99.036(2) 87.955(7) 94.690(3)
( o) 95.566(2) 75.462(6) 96.226(3)
V (Å3) 1238.1(3) 1803.0(19) 1095.11(1)
Z 2 4 2
d(calc) g cm-1
1.454 1.606 1.644
Abs coeff, , cm-1
0.197 0.249 0.223
Data collected 4354 6329 4947
Data Fo2>3(Fo
2) 2870 5186 3199
Variables 325 521 333
Ra 0.0631 0.0493 0.0573
Rwb 0.2154 0.1449 0.1834
Goodness of Fit 1.043 1.050 1.078
These data were collected at 293 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
75
Table 4.3: Selected crystallographic data for compounds 4-8, 4-12 and 4-13
Crystal 4-8c 4-12 4-13
Formula C20H20BF10P C40H43BF10NP C29H31BF10NP
Formula weight 492.14 769.53 625.33
Crystal system Triclinic Triclinic Monoclinic
Space group P-1 P-1 Cc
a(Å) 9.5420(17) 9.8554(5) 19.0354(6)
b(Å) 12.872(2) 12.3799(6) 11.9222(4)
c(Å) 19.447(3) 16.4730(8) 27.8012(12)
(o) 96.656(2) 72.596(2) 90.00
( o) 95.583(2) 78.921(2) 106.466(2)
( o) 109.144(2) 86.032(2) 90.00
V (Å3) 2217.9(2) 1881.95(16) 6050.6(4)
Z 4 2 8
d(calc) g cm-1
1.474 1.358 1.373
Abs coeff, , cm-1
0.212 0.153 0.172
Data collected 7778 19836 13781
Data Fo2>3(Fo
2) 2777 13490 7979
Variables 594 478 775
Ra 0.0949 0.0441 0.0498
Rwb 0.2839 0.1333 0.1195
Goodness of Fit 0.952 1.038 1.001
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
cThese data were collected at 293 K with Mo Kα radiation (λ = 0.71069 Å).
76
Table 4.4: Selected crystallographic data for compound 4-14
Crystal 4-14
Formula C16H36B2Cl4P2
Formula weight 453.81
Crystal system Monoclinic
Space group P21/n
a(Å) 8.8289(3)
b(Å) 14.3784(5)
c(Å) 9.0538(3)
(o) 90.00
( o) 92.991(1)
( o) 90.00
V (Å3) 1147.77(7)
Z 2
d(calc) g cm-1
1.313
Abs coeff, , cm-1
0.654
Data collected 2633
Data Fo2>3(Fo
2) 2416
Variables 115
Ra 0.0277
Rwb 0.0770
Goodness of Fit 1.050
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
77
4.3: Results and Discussion
4.3.1: Synthesis and Characterization of Phosphinoboranes R2PB(C6F5)2
A series of phosphinoboranes R2PB(C6F5)2 (R=Et, Ph, tBu, Cy, Mes) were synthesized
from (C6F5)2BCl and the corresponding lithium phosphide. The smaller derivatives: R=Et (4-1)
and R=Ph (4-2), exist as dimers [R2P-B(C6F5)2]2, while the R=tBu (4-3), R=Cy (4-4) and R=Mes
(4-5) phosphinoboranes are monomers in solution (Figure 4.3).
Figure 4.3: Synthesis of 4-1 to 4-5 (LiCl is removed upon workup)
Dimers 4-1 and 4-2 are characterized by triplet resonances at -12.9 (1JP-B=72 Hz) and -2.2
(1JP-B=68 Hz) ppm respectively in the
11B NMR spectra. Further NMR spectroscopic evidence
supported these assignments, including gaps of 6.85 and 7.82 ppm between the meta- and para-
fluorine signals in the 19
F NMR spectra for 4-1 and 4-2 respectively, which are typical of neutral,
4 coordinate boron centres.179,180
These structural assignments were additionally supported by
X-ray crystallography (Figure 4.4).
78
Figure 4.4: POV-Ray depictions of phosphinoborane dimers 4-1 (left) and 4-2–CH2Cl2 (right).
Carbon: black, Boron: yellow-green, Fluorine: deep pink, Phosphorus: orange. Solvent and
hydrogen atoms are omitted for clarity. Selected metrical parameters (distances: Å, angles: °)
4-1: B1-P1 2.056(3), B1-P1a 2.058(3), B1-P1-B1a 92.30(10), P1-B1-P2 87.70(10). 4-2–
CH2Cl2: P1-B1 2.096(5), P1-B2 2.088(5), P2-B1 2.022(5), P2-B2 2.080(5), B1-P1-B2
92.65(19), B1-P2-B2 95.05(18), P1-B1-P2 86.57(18), P1-B2-P2 85.32(18)
The crystal structure of 4-1 reveals a symmetric dimer with only one half of the dimer in
the asymmetric unit. The P-B distances are 2.056(3) and 2.058(3) Ǻ, while the B-P-B and P-B-P
bond angles are 92.30(10)° and 87.70(10)° respectively, the remaining metrical parameters are
unexceptional. The structure of the diphenylphosphino analog 4-2 shows similar metrical
parameters, with average P-B bond lengths of 2.072 Ǻ, average P-B-P angle of 85.95° and
average B-P-B angle of 93.87°. The data for these dimers are consistent for those reported for
the related species [(Et2P)2B(µ-PEt2)]2,206
[Et2B(µ-PtBu2)]2
207 and 1,1
1-ferrocene[B(CH3)(µ-
PPh2)]2208
with the exception of P-B bond lengths. The bond lengths are slightly longer for the
compounds reported here, presumably due to crowding resulting from the larger substituents at
boron.
Reaction of the more sterically demanding lithium phosphides (R=tBu, Cy and Mes) with
(C6F5)2BCl resulted in the generation of phosphinoborane monomers of the general form
R2PB(C6F5)2 (Figure 4.3). While related compounds of the general form R2PBMes2 have been
previously synthesized,202,209
these novel compounds were expected to provide different
reactivity and structural features due to the highly electron-withdrawing nature of the fluoroaryl
groups on boron.
79
The NMR spectroscopic data for compounds R2PB(C6F5)2 (4-3, R=tBu; 4-4, R=Cy; 4-5,
R=Mes) were consistent with the proposed formulations and are summarized in Table 4.5, while
spectra for 4-3 are shown in Figure 4.5. The gaps between para- and meta-fluorine resonances
lie closer to the range for typical 4-coordinate boranes, suggesting that there is substantial
electron density being donated into the vacant p-orbital on boron from the lone pair at
phosphorus. The 11
B NMR spectra shows an upfield doublet resonance, typical of 3-coordinate
boranes, which exhibits coupling to phosphorus (1JB-P~150 Hz), while the
31P NMR resonances
are also upfield and significantly broadened due to coupling to the quadrupolar boron centre.
Figure 4.5: Multinuclear NMR spectra for 4-3. A: 1H, B:
11B, C:
19F, D:
31P{
1H}
C
B A
D
o-C6F5
p-C6F5
m-C6F5
80
Table 4.5: Selected NMR spectroscopic data for compounds 4-1 to 4-5
Compound 19
F NMR δ (o, p, m): 19
F NMR
Δm-pa
31P NMR
δ:
11B NMR δ:
[(C6F5)2B(PEt2)]2 (4-1) -125.0, -153.5, -160.3 6.8 -12.9 -23.4
[(C6F5)2B(PPh2)]2 (4-2) -121.3, -156.3, -164.2 7.9 -0.8 -2.2
(C6F5)2B(PtBu2) (4-3) -130.7, -156.0, -163.4 7.4 120.7 41.8
(C6F5)2B(PCy2) (4-4) -130.8, -155.5, -163.5 8.0 92.1 39.5
(C6F5)2B(PMes2) (4-5) -131.2, -154.6, -163.5 8.9 29.3 40.1
a This value is the difference in chemical shift between the meta and para-fluorines, and has been
noted to be characteristic for different bonding environments at boron in fluoroarylboranes (the
shortest m-p gaps are generally found for 4 coordinate fluoroaryl borates, while the largest m-p
gaps are found for neutral 3 coordinate fluoroaryl boranes).179,180
The structures of 4-3 and 4-4 were confirmed by X-Ray crystallography (Figure 4.6).
The crystal structure of compound 4-3 appeared as anticipated from other analytical data.
Though both phosphorus and boron centres are planar, with sums of angles at 359.07° and
360.0°, respectively, they are not coplanar. This geometry is a result of steric repulsion between
the bulky tert-butyl groups on phosphorus and C6F5 groups on boron. This repulsion manifests
itself in C-P-B-C torsion angles of 21.6° and 7.4°. Compound 4-3 does, however, show a
significantly shorter P-B bond length (1.786(4) Å) than the sterically similar compound
Mes2BPtBu2 (1.841 Ǻ –average of 2 molecules in asymmetric unit) suggesting the electron-
withdrawing effect of the C6F5 group results in a contraction of the P-B bond length.
Compound 4-4 shows essentially coplanar phosphorus and boron centres and a very
short P-B bond length of 1.762(4) Ǻ, significantly shorter than even that observed for 4-3. The
shorter bond length of 4-4 is the result of a stronger P-B π–bond, due to the better overlap on
account of the smaller substituents on phosphorus which allow the phosphorus and boron atoms
to be essentially coplanar. The sum of angles at phosphorus and boron are 359.85° and 359.97°
respectively, while the C-P-B-C torsion angles are 1.1° and 5.6°.
81
Figure 4.6: POV-Ray depictions of phosphinoboranes 4-3 (left) and 4-4 (right). Carbon: black,
Boron: yellow-green, Fluorine: deep pink, Phosphorus: orange. Hydrogen atoms are omitted for
clarity. Selected metrical parameters (distances: Å, angles: °). 4-3: P1-B1 1.786(4), P1-C13
1.862(3), P1-C17 1.865(3), B1-C1 1.580(4), B1-C7 1.590(4), C13-P1-C17 117.08, B1-P1-C13
120.85(14), B1-P1-C17 121.14(15), C1-B1-C7 113.3(2), C1-B1-P1 124.0(2), C7-B1-P1
122.7(2). 4-4: P1-B1 1.762-(4), P1-C13 1.820(4), P1-C19 1.835(4), B1-C1 1.585(5), B1-C7
1.596(5), C13-P1-C19 112.6(2), B1-P1-C13 126.4(2), B1-P1-C19 120.88(19), C1-B1-C7
117.0(3), C1-B1-P1 119.9(3), C7-B1-P1 123.1(3).
The bond lengths observed for 4-3 and 4-4, along with the torsion angles, suggest a
significant π-bonding interaction between the lone pair at phosphorus and the vacant p-orbital on
boron (Figure 4.7). These bonds are much shorter than those observed for the analogous
R2PBMes2 compounds (Table 4.6). This interaction is aided by the bulky, electron-donating R-
groups at phosphorus and the bulky electron withdrawing groups at boron, which both serve to
enforce planarity and encourage electron donation from phosphorus to boron. In fact, values
suggested for P-B double bonds in the literature are 1.79-1.84 Å, while P-B single bonds range
from 1.90-2.00 Å.210
Using these values as a guideline, compounds 4-3 and 4-4 can be
considered as having significant P-B multiple bonding character (in fact, the bond lengths are
shorter than the range attributed to P-B double bonds).
82
A DFT study of 4-3 undertaken in collaboration with Thomas M. Gilbert at the
University of Northern Illinois showed that the π-bonding HOMO was highly polarized, with
74% of this molecular orbital derived from the phosphorus atom, with only 26% from the boron
atom.211
Figure 4.7: Resonance forms of phosphinoboranes 4-3 to 4-5
Table 4.6: Selected metrical parameters for monomeric phosphinoboranes
Compound P-B (Ǻ) Torsion Angles
(C-P-B-C)
Sum of Angles
(B)
Sum of Angles
(P)
4-3 1.786(4) 21.6, 7.4 359.07 360.0
4-4 1.762(4) 1.1, 5.4 359.97 359.85
Mes2BPtBu2
202 1.839(8)
1.843(8)
2.2, 41.2
13.8, 29.8
359.4
359.8
352.0
359.2
Mes2BPMes2202
1.839(8) 4.7, 4.7 360.0 360.0
Mes2BPPh2209
1.859(3) 33.3, 28.5 359.3 339.4
4.3.2: Reactions of Phosphinoboranes with H2 and Independent Synthesis of Phosphine-Borane Adducts R2(H)PB(H)C6F5)2
Though the experimental data for compounds 4-3 to 4-5 suggests significant interaction
between the lone pair at phosphorus and the vacant orbital at boron, the computational data
suggests that they may be highly polarisable. This may allow them to participate in Lewis
acid/Lewis base chemistry. The analogous nature of these species and bimolecular FLPs of
83
phosphines and boranes, suggests that activation of H2 may be possible for these compounds. As
such, compounds 4-1 to 4-5 were exposed to 4 atm H2 at both room temperature (25°C) and
60°C. Indeed, hydrogenation of compounds 4-3 and 4-4 was observed (these results are
summarized in Table 4.6). Compounds 4-1 and 4-2 showed no signs of reaction while 4-5 only
showed trace amounts of reactions after several weeks. The hydrogenations of 4-3 and 4-4 at
60°C are significantly accelerated versus those at room temperature (Table 4.7), suggesting that
while the reaction of related (CF3)2BPR2 species with H2 has been calculated as being
exothermic,203
there may be a relatively large kinetic barrier in the present case. These reactions
produce the compounds R2(H)P-B(H)(C6F5)2, which are secondary phosphine adducts of the
known borane HB(C6F5)2.204,205
These compounds were synthesized independently from
reactions of the phosphines R2PH and the borane HB(C6F5)2 (Figure 4.9). NMR spectroscopic
data for the compounds R2(H)PB(H)(C6F5)2 (4-6, R=Et; 4-7, R=Ph; 4-8, R=tBu; 4-9, R=Cy; 4-
10, R=Mes) are summarized in Table 4.8.
Figure 4.8: Synthesis of 4-8 to 4-10 through H2 activation (left) or Lewis acid-base adduct
formation (right)
Table 4.7: Hydrogenation of phosphinoboranes 4-3 and 4-4
Compound Temperature Time (days) Conversion (by 19
F NMR)
4-3 25 14 100
4-3 60 2 100
4-4 25 28 80
4-4 60 2 100
84
Table 4.8: Selected NMR spectroscopic data for phosphine-borane adducts 4-6 to 4-10
Compound 19
F NMR δ (o, p, m): 31
P NMR δ: 11
B NMR δ:
(C6F5)2HBPHEt2 (4-6) -131.7, -159.2, -164.7 -4.6 -30.0
(C6F5)2HBPHPh2 (4-7) -131.2, -158.8, -164.7 -1.5 -28.5
(C6F5)2HBPHtBu2 (4-8) -129.7, -159.4, -164.7 32.0 -30.0
(C6F4)2HBPHCy2 (4-9) -131.0, -159.4, -164.7 7.1 -28.1
(C6F5)2HBPHMes2 (4-10) -131.4, -158.7, -164.8 -39.0 -25.4
Crystal structures were obtained for compounds 4-6, 4-7 (Figure 4.9) and 4-8 (Figure
4.10). Species 4-6 has the shortest P-B bond length (1.950(3) Å), while 4-7 and 4-8 show
identical P-B bond lengths (1.966(3) Å and 1.966(9) Å, respectively). These lengths are
significantly shorter than those for the R2(H)P-B(C6F5)3 relatives: R=Cy (2.0270(14) Ǻ),212
R=cyclopentyl (2.0243(3) Ǻ), R=Et (2.036(8)Ǻ) and R=Ph (2.098(3)Ǻ),34
as a result of reduced
steric bulk of HB(C6F5)2 compared to B(C6F5)3.
85
Figure 4.9: POV-Ray depictions of 4-6 and 4-7. Carbon: black, Hydrogen: white; Boron:
yellow-green, Fluorine: deep pink, Phosphorus: orange. Carbon-bound hydrogen atoms are
omitted for clarity. Selected metrical parameters (distances: Å, angles: °). 4-6 (one of two
crystallographically independent molecules): P1-B1 1.950(3), H1-B1-P1-H2 176.76. 4-7: P1-B1
1.966(3), H1-B1-P1-H2 178.98.
Figure 4.10: POV-Ray depiction of 4-8. Carbon: black, Hydrogen: white, Boron: yellow-green,
Fluorine: deep pink, Phosphorus: orange. Carbon-bound hydrogen atoms are omitted for clarity.
Selected metrical parameters (distances: Å, angles: °). 4-8: P1-B1 1.966(9); H1-B1-P1-H2
166.55.
86
In compounds 4-6 and 4-7 the P-H and B-H hydrogen atoms are trans to one another, in a
typical staggered conformation (H-P-B-H=176.76° and 178.98°, respectively), while 4-8 shows
more twisting in the solid state from the staggered conformation (H-P-B-H=166.55°). This
twisting (Figure 4.11) appears to be a solid-state effect as there is no evidence of inequivalent
C6F5 or tert-butyl groups by solution NMR spectroscopy.
Figure 4.11: Newman projection along the B-P bond of 4-6 (R-Et) and 4-7 (R=Ph) (left); and 4-8
(right) as determined by X-ray crystallography
Efforts were made to initiate the loss of H2 from these compounds, however, heating 4-6
to 4-10 (2d, 140°C), even in the presence of a smaller base, which has been shown to accelerate
the loss of H2 from related systems,213
showed no loss of H2. These reactions often showed signs
of dissociation and subsequent decomposition, likely due to trace moisture in solution. These
observations are in contrast to other phosphine borane adducts which can lose hydrogen under
thermal duress,91,195
likely due to the presence of electron-withdrawing groups at boron in the
present case.
A DFT study of this system by Thomas M. Gilbert showed that H2 activation by 4-3 is
exothermic (-43 kcal/mol),211
while the limiting step for the reaction was found to be the first
step: attack of H2 at boron, with a barrier of 22 kcal/mol. The highly exothermic nature of the H2
activation makes the loss of H2 very unfavourable, consistent with the experimental observations.
4.3.3: Reactions of Phosphinoboranes with Lewis Acids and Lewis Bases
To demonstrate the ambiphilic nature of these monomeric P-B compounds, further
reactions were conducted with the goal to coordinate various Lewis acids and bases to the Lewis
basic and acidic sites, respectively, within the monomeric P-B complexes.
These species do react with relatively small Lewis bases, such as 4-tert-butylpyridine,
showing evidence for the formation of Lewis acid-base adducts 4-11 and 4-12, within 20 minutes
87
in CDCl3 at room temperature (Figure 4.12). In the case of 4-12, adduct formation is
quantitative, while only a trace amount of 4-11 is formed initially by multinuclear NMR. Both
reactions show the appearance of multiple products over 24 hours at room temperature in CDCl3.
The molecular structure of 4-12-C7H8 was determined by X-ray crystallography (Figure 4.13).
Figure 4.12: Formation of adducts 4-11 and 4-12
Figure 4.13: POV-Ray depiction of 4-12-C7H8. Carbon: black, Boron: yellow-green, Fluorine:
deep pink, Nitrogen: blue, Phosphorus: orange. Hydrogen and solvent atoms are omitted for
clarity. Selected metrical parameters (distances: Å, angles: °). B1-N1 1.6332(11), B1-P1
2.0329(9), N1-B1-P1 106.04, N1-B1-C1 101.81(6), N1-B1-C7 109.06(6), C1-B1-C7 113.53(7),
P1-B1-C1 115.87(5), P1-B1-C7 109.77(5), B1-P1-C13 101.27(4), B1-P1-C19 109.43(4), C13-
P1-C19 104.20(4).
The crystal structure reveals a dramatic lengthening of the P-B bond to 2.0329(9) Ǻ as a
result of the loss of the π-bonding interaction between the formerly vacant orbital on boron and
88
the lone pair on phosphorus. The B-N bond of 1.6332(11) Ǻ is slightly longer than that reported
for the analogous B(C6F5)3 adduct of 4-tert-butyl pyridine (5-1, 1.618(2) Ǻ). This longer bond
length is a result of diminished Lewis acidity at boron and increased steric crowding caused by
the adjacent phosphorus centre. As expected, the boron centre has become tetrahedral, with
bond angles varying no more than 8 degrees from the idealized value of 109.5°. The N-B-C
angles average 105.44° while the P-B-C angles average 112.82°, indicative of greater steric
repulsion on the B(C6F5)2 fragment from the PCy2 unit than from the coordination of the
pyridine. The phosphorus centre has also pyramidalized with the sum of angles now totalling
314.9°. This is somewhat surprising since the primary reason given for the planarity at
phosphorus in related systems was steric repulsion from the groups on boron.202
The
pyramidalization at phosphorus in 4-12 despite a similar steric environment suggests that the
planarity observed in 4-4 is also aided by a significant electronic factor.
In the case of the reaction of 4-3 with 4-tert-butylpyridine after one day, two broad
upfield peaks typical of B-F resonances were observed in the 19
F NMR spectrum, while 1H,
11B
and 31
P NMR spectra showed a complex mixture of products. The 19
F NMR resonances suggest
that nucleophilic aromatic substitution (NAS) has occurred at one of the C6F5 rings. 4-13, one
component of this complex mixture of products, was characterized crystallographically (Figure
4.14).
89
Figure 4.14: POV-Ray depiction of 4-13. Carbon: black, Boron: yellow-green, Fluorine: deep
pink, Nitrogen: blue, Phosphorus: orange. Hydrogen atoms are omitted for clarity. Selected
metrical parameters (distances: Å, angles: °): B1-N1 1.619(5), B1-C1 1.637(5), B1-C7 1.662(5),
B1-F12 1.395(4), P1-C12 1.854(4), N1-B1-F12 105.1(3), C1-B1-F12 107.6(3), N1-B1-C1
109.8(3), N1-B1-C7 107.0(3), C12-P1-C22 102.27(18), C12-P1-C26 107.59(17), C22-P1-C26
111.27(19). -
4-13 proved to be the product of an unexpected ortho-nucleophilic aromatic substitution
reaction. Such reactions have not been previously reported for fluoroarylboranes and could
potentially be used to generate novel linked FLPs. Optimization of this reaction by tuning of the
size and strength of the Lewis base, could provide cleaner formation of species related to 4-13,
which could transformed into ortho-linked FLPs by abstraction of the base (pyridine in this case)
by a stronger Lewis acid. The formation of 4-13 likely involves initial formation of 4-12, the
loss of π–bonding in 4-12 weakens the P-B bond and makes the phosphine much more
nucleophilic than in 4-3, resulting in rearrangement to 4-13 (Figure 4.15). The P-B bonding in
the starting material places the phosphorus centre close to the ortho-carbons and thus more likely
to attack there than at the para-carbon. While a bimolecular reaction is possible, it would likely
form the more commonly observed para-nucleophilic aromatic substitution product.
90
Figure 4.15: Proposed formation of 4-13 from 4-11 (L=4-(tBu)C5H4N)
Reactions of phosphinoboranes 4-3 and 4-4 with the small Lewis acid BCl3 over 30
minutes in toluene produced dimers 4-14 and 4-15 respectively, [R2PBCl2]2, along with
ClB(C6F5)2 (Figure 4.17). For 4-14 a trace amount of monomer was observed by 11
B NMR
spectroscopy, this observation is consistent with equilibria observed for the related [tBu2PBMe2]2
and [((CH3)3Si)2PBMe2]2 dimers and their respective monomers.214
4-5 did not react with BCl3
under similar conditions, suggesting that the phosphorus centre is not nucleophilic enough to
coordinate to the incoming Lewis acid (the mesityl groups of 4-5 are both bulkier and less
electron-donating than the tert-butyl or cyclohexyl groups of 4-3 and 4-4, resulting in a less
nucleophilic phosphorus centre).
The formation of these species suggests that coordination of BCl3 at phosphorus leads to
a sufficient increase in the Lewis acidity at the fluoroarylborane centre to cause chloride transfer
from BCl3 to the B(C6F5)2 boron (Figure 4.16). Dissociation of ClB(C6F5)2 from the very bulky
phosphine allows dimerization of transient phosphinoborane monomer R2PBCl2. This reaction is
thermodynamically favourable as it results in the formation of 2 P-B bonds while only breaking
one such bond. 4-14 was characterized crystallographically (Figure 4.17).
Figure 4.16: Formation of dimers 4-14 and 4-15
91
Figure 4.17: POV-Ray depiction of 4-14. Carbon: black, Boron: yellow-green, Chlorine:
aquamarine, Phosphorus: orange. Phosphorus: orange; Fluorine: deep pink; Boron: yellow-
green; Chlorine: aquamarine; Carbon: black. Hydrogen atoms are omitted for clarity. Selected
metrical parameters (distances: Å, angles: °): P1-B1 2.0552(15), P1-B1a 2.0557(15), B1-Cl1
1.8409(15), B1-Cl2 1.8453(15), B1-P1-B1a 89.01(6), P1-B1-P1a 90.99(6).
Metrical parameters for 4-14 were similar to those of 4-1 and 4-2, with the P-B bond
lengths again falling towards the long end of the range for related P-B dimers206-208
due to the
bulky groups at phosphorus.
In an effort to react compounds 4-3 and 4-4 with small Lewis acids and bases
simultaneously and possibly trap intermediates in the known Lewis acid catalyzed
dehydrogenation of these species,191
reactions with H3NBH3 and Me2(H)NBH3 were conducted.
While no adduct formation was observed, compounds 4-3 and 4-4 abstracted H2 from the amine-
borane adducts to form compounds 4-8 and 4-9, respectively. The nature of the byproducts from
the reaction with H3NBH3 was not clear due to poor solubility and broad peaks in solvents
compatible with 4-3 and 4-4, while reaction with Me2(H)NBH3 produced the well-known dimer
(Me2N-BH2)2.83
These experiments demonstrate the greater affinity for H2 shown by compounds
4-3 and 4-4 when compared to amine-borane adducts. This is due to enhanced Lewis acidity at
boron in 4-3 and 4-4 compared to the boron centres of Me2(H)NBH3. The proton transfer from
nitrogen to phosphorous is aided by the enthalpically favourable formation of B=N double bond,
while the sterically small B=N species subsequently oligomerize.
92
4.4: Conclusions
Monomeric phosphinoboranes of the general formula R2PB(C6F5)2 can be synthesized,
provided the R groups are large enough to prevent dimerization (tBu, Cy and Mes for example).
These phosphinoboranes do exhibit significant π–bonding as the P-B bond lengths are extremely
short. This bond, however, is sufficiently polarisable for these species to react with H2, or a
variety of small Lewis acids or bases. Lewis acid-base adducts of these ambiphilic
phosphinoboranes are rather unstable due to the steric repulsion and weakening of the P-B bond
caused by the loss of the π-bonding component of the interaction. These reactions could
potentially be optimized to allow synthesis of novel ortho-substituted FLPs related to 4-13.
These results also illustrate the potential of covalently bound group 13-group 15 species
to add hydrogen under mild conditions without a catalyst. This knowledge could aid in the
design of hydrogen storage materials which can be readily hydrogenated following loss of H2.
93
Chapter 5: Frustrated Lewis Pairs: Reactions of Pyridines and Other Nitrogen-Containing
Heterocycles with B(C6F5)3
5.1: Introduction
Hydrogen activation is a tremendously important aspect of modern chemistry. From the
Haber-Bosch process to recent asymmetric hydrogenation reactions, hydrogen activation
chemistry has been dominated by transition metals.215
The area is generally confined to
transition metals due to their ability to undergo oxidative addition reactions with H2.215
As
discussed in section 1.4.3, in recent years heterolytic H2 activation has been found to be possible
utilizing main group frustrated Lewis pairs (FLPs).2 With much recent work focusing on this
reaction,3 efforts are being made to make use of cheaper, lighter Lewis bases in order to expand
the chemistry to more commercially viable applications. Of the currently unexplored options
available, pyridines are a particularly interesting group of compounds. There are a wide variety
of readily available monodentate and bidentate pyridines and potential catalytic hydrogenation
could yield dihydropyridines, which are extremely useful as organic reducing agents.216
Knowing that imines46,73
and amines46,47
are capable of H2 activation with B(C6F5)3, the similar
basicity of pyridines suggests this chemistry should be possible. While pyridine is well known to
form a stable adduct with B(C6F5)3,217
pyridines substituted at the 2 or 6 position should offer the
opportunity for FLP chemistry due to crowding at nitrogen. Also promising is the report by
Brown et al. that 2,6-lutidene does not form an adduct with BMe3 (Figure 5.1).32
Figure 5.1: Brown’s observation of a surprising lack of reactivity between 2,6-lutidine and
BMe332
An interesting aspect of this chemistry is the potential to catalytically generate a
Hantzsch Ester, a powerful pyridine-based source of H2 utilized in a number of organic
reductions,216
from the spent starting material – which is a bulky pyridine. Related compounds
94
such as acridan and tetrahydroquinolines have also been shown to act as stoichiometric reducing
agents (Figure 5.2) 218-223
and again could potentially be regenerated by catalytic hydrogenation
of the corresponding quinoline.
Figure 5.2: Amine-based reducing agents: Hantzsch’s Ester (A), acridan (B) and
tetrahydroquinoline (C)
The regeneration of the spent Hantzsch’s Ester or acridan would require initial H2
activation by the FLP, followed by hydride transfer from boron to the para-carbon of the
pyridine (Figure 5.3). This catalytic hydrogenation pathway is identical to that seen in the
B(C6F5)3 catalyzed hydrogenation of imines,46
except here the hydride attack would take place at
the 4-position of the nitrogen-containing heterocycle. This route could allow for reductions not
possible directly with B(C6F5)3, while maintaining the generally reduced cost and toxicity
compared to transition-metal catalysts.
Figure 5.3: Proposed scheme for catalytic transfer hydrogenation through Hantzsch’s ester.
95
5.2: Experimental Section
5.2.1: General Considerations
All preparations were done under an atmosphere of dry, O2-free N2 employing both Schlenk line
techniques and an Innovative Technologies or Vacuum Atmospheres inert atmosphere glove box.
Solvents (pentane, hexanes, toluene, and methylene chloride) were purified employing a Grubbs’
type column systems manufactured by Innovative Technology and stored over molecular sieves
(4 Å). Molecular sieves (4 Å) were purchased from Aldrich Chemical Company and dried at 140
ºC under vacuum for 24 hours prior to use. Deuterated solvents were dried over CaH2 (CD2Cl2,
CDCl3) and vacuum distilled prior to use. All common organic reagents were purified by
conventional methods unless otherwise noted. All liquid pyridines were stored over 4 Å
molecular sieves. 1H,
13C,
11B, and
19F nuclear magnetic resonance (NMR) spectroscopy spectra
were recorded on a Bruker Avance-400 spectrometer at 300K unless otherwise noted. 1H and
13C
NMR spectra are referenced to SiMe4 using the residual solvent peak impurity of the given
solvent. 11
B and 19
F NMR experiments were referenced to 15% BF3-Et2O in CDCl3.Chemical
shifts are reported in ppm and coupling constants in Hz as absolute values. Combustion analyses
were performed in house employing a Perkin Elmer CHN Analyzer. B(C6F5)3 was generously
donated by NOVA Chemicals Corporation.
5.2.2: Synthesis of Pyridine-B(C6F5)3 adducts:
(4-tBu)C5H4NB(C6F5)3 (5-1), (2-Me)C5H4NB(C6F5)3 (5-2), (2-Et)C5H4NB(C6F5)3 (5-3), (2-
C5H4N)NH(2-C5H4N)B(C6F5)3 (5-4) (2-Ph)C5H4NB(C6F5)3 (5-5), (2-C5H4N)C5H4NB(C6F5)3 (5-
6), C9H7NB(C6F5)3 (5-7) - These compounds were prepared in a similar fashion and thus only
one preparation is detailed. B(C6F5)3 (100 mg, 0.20 mmol) was added to a solution of 4-tert-
butylpyridine (26 mg, 0.20 mmol) in toluene (2 mL). The solution was stirred for 4 hours,
hexanes (2 mL) was added and the solution was stored at -35 °C overnight. The solution was
decanted from the resulting white precipitate. The precipitate was washed with hexanes (2 mL)
and dried in vacuo.
5-1 - Yield: 98 mg (78%). Anal. Calcd for C27H13BF15N: C, 50.11%; H, 2.02%; N, 2.16%.
Found: C, 50.27%; H, 2.08%; N, 2.16%. Crystals were grown from the hexane wash layer at -
35°C. 1H NMR (CD2Cl2) δ: 1.40 (s, 9H, CH3), 7.64 (d,
3JH-H = 7 Hz, 2H), 8.46 (d,
3JH-H = 7 Hz,
96
2H). 19
F NMR (CD2Cl2) δ: -132.2 (d, 3JF-F = 19 Hz, 6F, o-C6F5), -158.2 (t,
3JF-F = 20 Hz, 3F, p-
C6F5), -164.6 (dd, 3JF-F = 19 Hz,
3JF-F = 20 Hz, 6F, m-C6F5).
11B NMR (CD2Cl2) δ: -4.1 (br s).
13C NMR (CD2Cl2) δ: 30.0 (CH3), 36.1 (C-CH3), 123.0, 137.4 (dm,
1JC-F = 241 Hz, CF), 140.3
(dm, 1JC-F = 260 Hz, CF), 146.3, 148.1 (dm,
1JC-F = 248 Hz, CF), 168.8.
5-2 - Yield: 89%. Anal. Calcd for C24H7BF15N: C, 47.64%; H, 1.17%; N, 2.31%. Found: C,
48.05%; H, 1.38%; N, 2.26%. Crystals were grown from a layered solution of CDCl3/pentane at
-35°C. 1H NMR (CDCl3) δ: 2.51 (s, 3H, CH3), 7.43 (m, 2H), 7.99 (td,
3JH-H = 7 Hz,
4JH-H = 2 Hz,
1H), 8.62 (m, J = 6 Hz, 1H). 19
F NMR (CDCl3) δ: -126.3 (t, 3JF-F = 22 Hz, 1F, o-C6F5), -128.9
(m, 1F, o-C6F5), -132.4 (d, 3JF-F = 22 Hz, 1F, o-C6F5), -133.2 (m, 1F, o-C6F5), -133.4 (m, 1F, o-
C6F5), -137.7 (m, 1F, o-C6F5), -155.6 (t, 3JF-F = 22 Hz, 1F, p-C6F5), -156.2 (t,
3JF-F = 22 Hz, 1F,
p-C6F5), -157.7 (t, 3JF-F = 22 Hz, 1F, p-C6F5), -161.9 (td,
3JF-F = 21 Hz,
4JF-F = 9 Hz, 1F, m-C6F5),
-162.9 (td, 3JF-F = 22 Hz,
4JF-F = 10 Hz, 1F, m-C6F5), -163.8 (td,
3JF-F = 22 Hz,
4JF-F = 9 Hz, 1F,
m-C6F5), -163.9 (td, 3JF-F = 21 Hz,
4JF-F = 9 Hz, 1F, m-C6F5), -164.2 (td,
3JF-F = 22 Hz,
4JF-F = 9
Hz, 1F, m-C6F5), -164.5 (td, 3JF-F = 22 Hz,
4JF-F = 8 Hz, 1F, m-C6F5).
11B NMR (CDCl3) δ: -3.6.
13C NMR (CDCl3) (partial) δ: 14.3, 122.6, 129.3, 142.3, 147.9, 159.8.
5-3 - Yield: 88%. Anal. Calcd. for C25H9BF15N: C, 48.50%; H, 1.47%; N, 2.26%. Found: C,
48.25%; H, 1.58%; N, 2.26%. Crystals were grown from the pentane wash layer at room
temperature. 1H NMR (CD2Cl2) δ: 0.80 (t,
3JH-H = 8 Hz, 3H, CH2CH3), 2.99 (dq,
2JH-H = 23 Hz,
3JH-H = 8 Hz, 1H, CH2CH3), 3.05 (dq,
2JH-H = 23 Hz,
3JH-H = 8 Hz, 1H, CH2CH3), 7.51 (t,
3JH-H =
7 Hz, 1H), 7.63 (d, 3JH-H = 8 Hz, 1H), 8.15 (td,
3JH-H = 8 Hz,
4JH-H = 1 Hz, 1H), 8.67 (q, J = 6 Hz,
1H). 19
F NMR (CD2Cl2) δ: -126.5 (t, 3JF-F = 22 Hz, 1F, o-C6F5), -129.6 (m, 1F, o-C6F5), -132.4
(d, 3JF-F = 22 Hz, 1F, o-C6F5), -133.7 (m, 1F, o-C6F5), -134.7 (m, 1F, o-C6F5), -137.3 (td,
3JF-F =
24 Hz, 4JF-F = 9 Hz, 1F, o-C6F5), -157.0 (t,
3JF-F = 21 Hz, 1F, p-C6F5), -157.3 (t,
3JF-F = 20 Hz, 1F,
p-C6F5), -159.2 (t, 3JF-F = 20 Hz, 1F, p-C6F5), -163.2 (td,
3JF-F = 22 Hz,
4JF-F = 9 Hz, 1F, m-C6F5),
-164.0 (td, 3JF-F = 22 Hz,
4JF-F = 10 Hz, 1F, m-C6F5), -164.8 (td,
3JF-F = 22 Hz,
4JF-F = 9 Hz, 1F,
m-C6F5), -165.1 (td, 3JF-F = 21 Hz,
3JF-F = 9 Hz, 1F, m-C6F5), -165.4 (td,
3JF-F = 22 Hz,
4JF-F = 9
Hz, 1F, m-C6F5), -165.6 (td, 3JF-F = 22 Hz,
3JF-F = 8 Hz, 1F, m-C6F5).
11B NMR (CD2Cl2) δ: -3.6.
13C NMR (CD2Cl2) (partial) δ: 13.2, 27.4, 122.6, 127.6, 142.8, 165.5.
5-4 - Yield: 86%. Anal. Calcd for C28H9BF15N3: C, 49.23%; H, 1.33%; N, 6.15%. Found: C,
49.59%; H, 1.69%; N, 6.13%. X-ray quality crystals were grown by slow evaporation from
97
CD2Cl2. 1H NMR (CD2Cl2) δ: 6.44 (d,
3JH-H = 8 Hz, 1H), 6.96 (dd,
3JH-H = 7 Hz,
3JH-H = 5 Hz,
2H), 7.02 (td, 3JH-H = 7 Hz,
4JH-H = 1 Hz, 1H), 7.55 (td,
3JH-H = 8 Hz,
4JH-H = 2 Hz, 1H), 7.67 (br s,
NH), 7.91 (ddd, 3JH-H = 9 Hz,
3JH-H = 7 Hz,
4JH-H = 2 Hz, 1H), 8.21 (m, 2H), 8.58 (d,
3JH-H = 9 Hz,
1H). 19
F NMR (CD2Cl2) δ: -127.0 (m, 1F, o-C6F5), -128.1 (m, 1F, o-C6F5), -131.6 (d, 3JF-F = 23
Hz, 1F, o-C6F5), -133.0 (m, 1F, o-C6F5); -135.7 (m, 1F, o-C6F5), -137.1 (m, 1F, o-C6F5), -157.0
(t, 3JF-F = 20 Hz, 3F, p-C6F5), -163.0 (td,
3JF-F = 22 Hz,
4JF-F = 8 Hz, 1F, m-C6F5), -163.9 (tt,
3JF-F
= 22 Hz, 4JF-F = 9 Hz, 2F, m-C6F5), -164.0 (td-,
3JF-F = 21 Hz,
4JF-F = 7 Hz, 1F, m-C6F5), -164.2
(td, 3JF-F = 22 Hz,
4JF-F = 8 Hz, 1F, m-C6F5), -165.1 (td,
3JF-F = 22 Hz,
4JF-F = 8 Hz, 1F, m-C6F5).
11B NMR (CD2Cl2) δ: -5.1.
13C NMR (CD2Cl2) (partial) δ: 139.0, 142.8, 144.1 (m), 148.2,
151.2, 152.4 (m).
5-5 - Yield: 85%. Anal. Calcd. for C29H9BF15N: C, 52.21%; H, 1.36%; N, 2.10%. Found: C,
51.77%; H, 1.71%; N, 2.32%. Crystals were grown from toluene at room temperature. 1H NMR
(CDCl3) δ: 7.05 (br s, 2H), 7.25 (t, 3JH-H = 8 Hz, 1H), 7.38 (d,
3JH-H = 8 Hz, 1H), 7.40 (br s, 1H),
7.65 (t, 3JH-H = 8 Hz, 1H), 7.85 (br s, 1H), 8.13 (t,
3JH-H = 8 Hz, 1H), 8.93 (s, 1H).
19F NMR
(CDCl3) δ: -125.4 (br s, 1F, o-C6F5), -128.6 (br s, 1F, o-C6F5), -131.2 (br s, 1F, o-C6F5), -131.9
(d, 3JF-F = 18 Hz, 1F, o-C6F5), -133.9 (br s, 2F, o-C6F5), -155.4 (t,
3JF-F = 20 Hz, 1F, p-C6F5), -
157.7 (br s, 2F, p-C6F5), -162.0 (t, 3JF-F = 23 Hz, 1F, m-C6F5), -162.9 (t,
3JF-F = 23 Hz, 1F, m-
C6F5), -164.5 (br s, 2F, m-C6F5), -165.1 (br m, 1F, m-C6F5), -165.6 (br s, 1F, m-C6F5). 11
B NMR
(CDCl3) δ: -2.9. 13
C NMR (CDCl3) (partial) δ: 124.3, 128.0, 128.5, 129.8, 131.6, 142.3, 148.5.
5-6 - Yield: 79%. Anal. Calcd for C28H8BF15N2: C, 50.33%; H, 1.21%; N, 4.19%. Found: C,
49.87%; H, 1.44%; N, 4.32%. Crystals were grown from toluene at -35°C. 1H NMR (CD2Cl2) δ:
6.68 (d, 3JH-H = 8 Hz, 1H), 7.16 (ddd,
3JH-H = 8 Hz,
3JH-H = 5 Hz,
4JH-H = 1 Hz, 1H), 7.43-7.48 (ov
m, 2H), 7.72 (ddd, 3JH-H = 8 Hz,
3JH-H = 6 Hz,
4JH-H = 2 Hz, 1H), 8.17 (ddd,
3JH-H = 5 Hz,
4JH-H =
2 Hz, 4JH-H = 1 Hz, 1H), 8.23 (td,
3JH-H = 8 Hz,
4JH-H = 2 Hz, 1H), 8.82 (br s, 1H).
19F NMR
(CD2Cl2) δ: : -125.2 (m, 1F, o-C6F5), -130.9 (m, 1F, o-C6F5), -131.5 (m, 1F, o-C6F5), -133.1 (m,
2F, o-C6F5), -135.6 (d, 3JF-F=21 Hz, 1F, o-C6F5), -156.7 (t,
3JF-F = 19 Hz, 1F, p-C6F5), -158.3 (m,
1F, p-C6F5), -160.0 (t, 3JF-F = 21 Hz, 1F, p-C6F5), -160.5 (m, 1F, m-C6F5), -163.3 (t,
3JF-F = 21
Hz, 1F, m-C6F5), -163.8 (t, 3JF-F = 21 Hz, 1F, m-C6F5), -166.0 (t,
3JF-F = 20 Hz, 1F, m-C6F5), -
166.5 (m, 1F, m-C6F5), -167.6 (m, 1F, m-C6F5). 11
B NMR (CD2Cl2) δ: -2.7. 13
C NMR (CD2Cl2)
(partial) δ: 123.7, 124.0, 125.0, 130.8, 136.4, 142.8, 148.5, 149.3, 153.6, 158.9.
98
5-7 - Yield: 96%. Anal. Calcd. for C27H7BF15N: C, 50.58%; H, 1.10%; N, 2.18%. Found: C,
50.23%; H, 0.98%; N, 2.35%. 1H NMR (CDCl3) δ: 7.77 (m, 2H), 7.84 (m, 1H), 8.09 (dd,
3JH-H =
8 Hz, 4JH-H = 2 Hz, 1H), 8.51 (d,
3JH-H = 9 Hz, 1H), 8.72 (d,
3JH-H = 8 Hz, 1H), 9.19 (q,
3JH-H = 5
Hz, 1H). 19
F NMR (CDCl3) δ: -126.6 (t, 3JF-F = 27 Hz, 1F, o-C6F5), -128.8 (br m, 1F, o-C6F5), -
131.9 (br m, 1F, o-C6F5), -132.9 (br d, 3JF-F = 36 Hz, 1F, o-C6F5), -133.3 (br m, 1F, o-C6F5), -
133.7 (m, 1F, o-C6F5), -155.1 (tt, 3JF-F = 20 Hz,
4JF-F = 4 Hz, 1F, p-C6F5), -156.2 (tt,
3JF-F = 20
Hz, 4JF-F = 3 Hz, 1F, p-C6F5), -157.2 (tt,
3JF-F = 20 Hz,
4JF-F = 3 Hz, 1F, p-C6F5), -161.2 (td,
3JF-F
= 21 Hz, 4JF-F = 8 Hz, 1F, m-C6F5), -162.3 (td,
3JF-F = 23 Hz,
4JF-F = 10 Hz, 1F, m-C6F5), -163.2
(td, 3JF-F = 22 Hz,
4JF-F = 8 Hz, 1F, m-C6F5), -163.7 (td,
3JF-F = 22 Hz,
4JF-F = 9 Hz, 1F, m-C6F5), -
163.8 (m, peaks overlapping, 1F, m-C6F5), -163.9 (m, peaks overlapping, 1F, m-C6F5). 11
B
NMR (CDCl3) δ: -3.2. 13
C NMR (CDCl3) (partial) δ: 120.2, 122.4, 128.6, 129.6, 130.1, 133.1,
142.6, 145.0, 150.4.
(2,6-Me2C5H3N)B(C6F5)3 (5-8) - B(C6F5)3 (100 mg, 0.20 mmol) was added to 2,6-lutidine (21
mg, 0.20 mmol) in 2 mL of toluene. The solution was allowed to stir for 4 h and pentane (2 mL)
was added. The solution was stored at -35°C. X-ray quality crystals precipitated from solution
and were washed with pentane (2 x 2 mL) and dried in vacuo. Yield: 60 mg (51%). NMR data
were acquired at -10°C.
1H NMR (CD2Cl2) δ: 2.58 (s, CH3), 7.36 (d, 2H,
3JH-H = 8 Hz, m-CH), 7.89 (t, 1H,
3JH-H = 8 Hz,
p-H). 19
F NMR (CD2Cl2) δ: -131.4 (br s, 2F, o-C6F5), -132.4 (br s, 2F, o-C6F5), -133.0 (d, 2F,
3JF-F = 18 Hz, o-C6F5), -157.6 (t, 1F,
3JF-F = 20 Hz, p-C6F5), -158.7 (t, 2F,
3JF-F = 20 Hz, p-C6F5),
-164.4 (t, 2F, 3JF-F = 21 Hz, m-C6F5), -165.2 (m, 4F, m-C6F5).
11B NMR (CD2Cl2) δ: -3.9.
99
van’t Hoff Plot for the Equilibrium Between Adduct 5-8 and the Separated Lewis Acid and
Lewis Base (values for 2 separate reactions are plotted, K was calculated based on 1H and
19F
NMR spectroscopy)
ΔG= ΔH - TΔS= -RTlnK
lnK= - ΔH/(RT) + ΔS/R (y = mx + b)
ΔH0= -mR= -(5062.4)(8.31451 J/(mol*K))= -42(1) kJ/mol
ΔS0= bR= (-15.809) (8.31451 J/(mol*K))= -131(5) J/(mol*K)
ΔG0= ΔH
0-TΔS
0= (-42.1 kJ/mol)-(298.15 K)(8.31451 J/(mol*K))= -2.9 kJ/mol
5.2.3 Synthesis of Pyridinium Borate Ion Pairs through H2 Activation by Pyridine-Borane FLPs
[2,6-Me2C5H3NH][HB(C6F5)3] (5-9), [(2,6-Ph2)C5H3NH][HB(C6F5)3] (5-10), [(2-
tBu)C5H4NH][HB(C6F5)3] (5-11) - These compounds were prepared in a similar fashion and thus
only one preparation is detailed. B(C6F5)3 (100 mg, 0.20 mmol) was added to 2,6-lutidine (21
mg, 0.20 mmol) in toluene (10 mL). The solution was subjected to 3 freeze-pump-thaw cycles
and backfilled with H2 at 77 K (~4 atm). The solution was allowed to stir overnight at room
temperature and dried in vacuo. The solid was washed with pentane (2 x 2 mL) and again dried
in vacuo. 5-9 - Yield: 105 mg (87%). Anal. Calcd for C26H11BF15N: C, 48.34%; H, 1.78%; N,
2.25%. Found: C, 48.49%; H, 2.06%; N, 2.43%. X-Ray quality crystals were grown by slow
evaporation of a toluene solution. 1H NMR (CD2Cl2) δ: 2.61 (s, 6H, CH3), 3.55 (q,
1JB-H = 88
y = 5062.4x - 15.809R² = 0.9824
0
1
2
3
4
5
6
7
8
0.0032 0.0034 0.0036 0.0038 0.004 0.0042 0.0044 0.0046
1/T (K-1)
lnK
100
Hz, B-H), 7.53 (d, 3JH-H = 8 Hz, 2H, m-CH), 8.22 (t, 1H,
3JH-H = 8 Hz, p-CH), 12.01 (br s, 1H, N-
H). 19
F NMR (CD2Cl2) δ: -136.8 (br d, 6F, 3JF-F = 18 Hz, o-C6F5), -165.8 (t, 3F
3JF-F = 20 Hz, p-
C6F5), -169.3 (br t, 6F, 3JF-F = 20 Hz, m-C6F5).
11B NMR (CD2Cl2) δ: -24.7 (d,
1JB-H = 88 Hz).
13C NMR (CD2Cl2) (partial) δ: 19.9, 125.5, 136.7 (dm,
1JC-F = 245 Hz CF),138.4 (dm,
1JC-F =
249 Hz, CF), 147.2, 148.2, (dm, 1JC-F = 238 Hz, CF), 153.8.
5-10 - Yield: 82%. Anal. Calcd for C35H15BF15N: C, 56.40%; H, 2.03%; N, 1.88%. Found: C,
56.19%; H, 2.18%; N, 2.09%. 1H NMR (CD2Cl2) δ: 3.35 (q,
1JB-H = 92 Hz, B-H), 7.57-7.62 (m,
4H), 7.64-7.68 (m, 2H), 7.74-7.77 (m, 4H), 8.05 (d, 3JH-H = 8 Hz, 2H, m-CH), 8.51 (t,
3JH-H = 8
Hz, 1H, p-CH), 11.27 (br s, 1H, N-H). 19
F NMR (CD2Cl2) δ: -134.3 (br d, 3JF-F = 21 Hz, 6F, o-
C6F5), -164.8 (t, 3JF-F = 20 Hz, 3F, p-C6F5), -167.7 (br t,
3JF-F = 20 Hz, 6F, m-C6F5).
11B NMR
(CD2Cl2) δ: -24.6 (d, 1JB-H = 92 Hz).
13C NMR (CD2Cl2) (partial) δ: 122.6, 126.9, 127.5, 130.2,
132.4.
5-11 - Yield: 105 mg (83%). Anal. Calcd for C27H15BF15N: C, 49.95%; H, 2.33%; N, 2.16%.
Found: C, 49.76%; H, 2.22%; N, 2.06%. 1H NMR (CD2Cl2) δ: 1.51 (s, 9H, C-CH3), 3.66 (q,
1JB-
H = 88 Hz, 1H, B-H), 7.80 (t, 3JH-H = 7 Hz, 1H), 7.96 (d,
3JH-H = 8 Hz, 1H), 8.45 (dd,
3JH-H = 8
Hz, 4JH-H = 2 Hz, 1H), 8.48 (d,
3JH-H = 7 Hz, 1H), 12.13 (br s, 1H, N-H).
19F NMR (CD2Cl2) δ: -
134.7 (br d, 3JF-F = 22 Hz, 6F, o-C6F5), -163.6 (t,
3JF-F = 21 Hz, 3F, p-C6F5), -167.1 (m, 6F, m-
C6F5). 11
B NMR (CD2Cl2) δ: -24.7 (d, 1JB-H = 87 Hz).
13C NMR (CD2Cl2) (partial) δ: 28.8,
37.1, 125.2, 125.3, 136.8 (dm, 1JC-F = 254 Hz, CF), 140.7, 147.7 148.2 (dm, CF,
1JC-F = 240 Hz).
[(2,3,5,6-Me4C4N2H)][HB(C6F5)3] (5-12) – In a J. Young-type NMR tube, B(C6F5)3 (20 mg,
0.039 mmol) was added to tetramethylpyrazine (5 mg, 0.04 mmol) in toluene-d8 (0.75 mL). The
solution was subjected to 3 freeze-pump-thaw cycles and backfilled with 1 atm H2 at 77 K (~4
atm at ambient temperature). The tube was sealed and warmed to room temperature. The
solution was monitored by multinuclear NMR; the reaction proceeded to completion over 18
hours. The product decomposed to several other species upon attempted workup and could not
be isolated.
1H NMR (tol-d8) δ: 2.04 (s, 12H, CH3), 3.84 (q,
1JB-H=94 Hz, B-H), 13.98 (br s, 1H, N-H).
19F
NMR (tol-d8): -133.3 (br d, 3JF-F=22 Hz, 6F, o-C6F5), -162.0 (t,
3JF-F=19 Hz, 3F, p-C6F5), -165.8
(br t, 3JF-F=21 Hz, 6F, m-C6F5).
11B NMR (tol-d8) δ: -24.9 (d,
1JB-H=90 Hz).
101
5.2.4: Synthesis of a Pyridinium Borate Zwitterion via THF Ring-Opening
2,6-Me2C5H3N(CH2)4OB(C6F5)3 (5-13) - 2,6 lutidine (25 mg, 0.23 mmol) was added to a
solution of B(C6F5)3 (100 mg, 0.20 mmol) in THF (2 mL). The solution was stirred for 3 days.
Pentane (2 mL) was added to ensure complete precipitation of the product, the solvent was
decanted and the resulting solid washed with pentane (2 x 2 mL). Yield: 120 mg (89%). X-Ray
quality crystals were grown from CH2Cl2 at room temperature. Anal. Calcd. for C29H17BF15NO:
Calcd: C, 50.39; H, 2.48; N, 2.03. Found: C, 50.30; H, 2.66; N, 2.17. 1H NMR (THF-d8) δ: 1.62
(m, 2H, CH2), 1.90 (m, 2H, CH2), 2.74(s, 6H, CH3), 3.18 (t, 3JHH=8 Hz, 2H, CH2), 4.68 (m, 2H,
CH2), 7.65 (d, 3JHH = 8 Hz, m-CH), 8.09 (t,
3JHH = 8 Hz, p-CH).
19F NMR (THF-d8) δ:-132.5 (d,
6F, 3JFF = 21 Hz, o-C6F5), -163.3 (t, 3F,
3JFF=21 Hz, p-C6F5), -166.6 (t, 6F
3JFF = 19 Hz, m-C6F5).
11B NMR (THF-d8) δ: -6.3.
13C NMR (THF-d8) (partial) δ: 22.0, 28.7, 30.2, 55.5, 65.4, 129.8,
138.5(dm, 1JCF = 244 Hz, CF), 140.3 (dm,
1JCF = 244 Hz, CF), 145.7, 150.1 (dm,
1JCF = 241 Hz,
CF), 158.0.
5.2.5: Reactions of Diethyl-2,6-dimethyl-3,5-pyridinedicarboxylate with B(C6F5)3
2,6-(CH3)2-3,5-(COOCH2CH3)2C5HN-B(C6F5)3 (5-14) - B(C6F5)3 (100 mg, 0.20 mmol) was
added to of diethyl-2,6-dimethyl-3,5-pyridinedicarboxylate (48 mg, 0.20 mmol) in toluene (2
mL). The solution was allowed to stir for 4 hours and then dried in vacuo. The solid was
washed with pentane (2 x 2 mL) and again dried in vacuo. Yield: 110 mg (74%). X-Ray quality
crystals were grown from pentane at -35°C. Anal. Calcd for C31H17BF15NO4: C, 48.78%; H,
2.24%; N, 1.84%. Found: C, 48.59%; H, 2.17%; N, 1.85%. Cooling to -60°C resulted in only
broadening of the peaks, not in resolution.
1H NMR (CD2Cl2) δ: 1.40 (t,
3JH-H=8 Hz, CH2-CH3), 2.75 (s, C-CH3), 4.47 (q,
3JH-H=8 Hz, CH2-
CH3), 8.48 (s, CH). 19
F NMR (CD2Cl2): -131.5 (br d, 3JF-F=17 Hz, 6F, o-C6F5), -150.5 (br s, 3F,
p-C6F5), -162.9 (br s, 6F, m-C6F5). 11
B NMR (CD2Cl2) δ: 42.2. 13
C NMR (CD2Cl2) (partial) δ:
13.9, 24.4, 64.0, 122.5, 137.7 (dm, 1JC-F=252 Hz, CF), 140.3, 148.1 (dm,
1JC-F=252 Hz, CF),
162.1, 168.3 (m).
[2,6-(CH3)2-3,5-(COOCH2CH3)2C5HNH][HB(C6F5)3] (5-15) and 2,6-(CH3)2-3,5-
(COOCH2CH3)2-2-H-C5H2NH-B(C6F5)3 (5-16) - 5-14 (20 mg, 0.027 mmol) in 0.75 mL toluene-
d8 was exposed to 4 atm H2 in a J. Young tube. The reaction was monitored by multinuclear
102
NMR spectroscopy and compared to independent syntheses of the products 5-15 and 5-16
through reaction of B(C6F5)3 with Hantzsch’s Ester.224
5.2.6: Reactions of Bulky Substituted Quinolines with B(C6F5)3
C13H9N-B(C6F5)3 (5-17) – Acridine (18 mg, 0.10 mmol) was added to a solution of B(C6F5)3 (50
mg, 0.10 mmol) in CDCl3 (0.75 mL) . In addition to the signals reported for the adducts, peaks
for free B(C6F5)3 were observed in the 19
F and 11
B NMR spectra. Only one set of peaks was
observed in the 1H NMR spectrum, as the resonances for the adduct are averaged with those for
free acridine. From the 19
F NMR spectrum at 25°C in CDCl3, Keq=19.1 M-1
.
1H NMR (CDCl3) δ: 7.64 (br s, 2H), 7.81(t,
3JH-H=8 Hz, 2H), 8.13 (s, 2H), 8.44 (br s, 2H), 9.10
(v br s, 1H). 19
F NMR (CDCl3) δ: -130.4 (br s, 6F, o-C6F5), -157.0 (br s, 3F, p-C6F5), -163.6 (br
s, 6F, m-C6F5). 11
B NMR (CDCl3) δ: -3.2.
2-(CH3)C9H6N-B(C6F5)3 (5-18) – 2-methylquinoline (14 mg, 0.10 mmol) was added to a solution
of B(C6F5)3 (50 mg, 0.10 mmol) in CDCl3 (0.75 mL). In addition to the signals reported for the
adduct, peaks for free B(C6F5)3 were observed in the 19
F and 11
B NMR spectra. Only one set of
peaks was observed in the 1H NMR spectrum, as the resonances for the adduct are averaged with
those for free 2-methylquinoline. From the 19
F NMR spectrum at 25°C in CDCl3, Keq=26.7 M-1
.
1H NMR (CDCl3) δ: 2.86 (s, 3H, CH3), 7.46 (br s, 1H), 7.63 (m, 2H), 7.90 (d,
3JH-H=6 Hz, 1H),
8.30 (br s, 2H). 19
F NMR (CDCl3) δ: -129.9 (br s, 3F, o-C6F5), -130.9(br s, 3F, o-C6F5), -156.3
(br s, 1F, p-C6F5), -156.9 (br s, 2F, p-C6F5), -162.7(br s, 2F, m-C6F5), -163.8 (br s, 2F, m-C6F5).
11B NMR (CDCl3) δ: -3.2.
1,10-C12H8N2-B(C6F5)3 (5-19) – 1,10-phenanthroline (7 mg, 0.04 mmol) was added to B(C6F5)3
(20 mg, 0.039 mmol) in CH2Cl2 (2 mL). The solution was allowed to stir for 2 hours, dried in
vacuo and the resulting solid was washed with pentane (2 x 2 mL). The resulting white solid
was again dried in vacuo. X-Ray quality crystals were grown from a layered solution of
CDCl3/pentane. Yield: 26 mg (96%). Anal. Calcd, for C30H8BF15N2: C, 52.06%; H, 1.16%; N,
4.05%. Found: C, 51.78%; H, 1.19%; N, 4.23%.
1H NMR (CDCl3) δ: 7.77 (m, 2H), 7.84 (m, 1H), 8.09 (dd,
3JH-H=8 Hz,
4JH-H=2 Hz,1H), 8.51 (d,
3JH-H=9 Hz, 1H), 8.72 (d,
3JH-H=8 Hz, 1H), 9.19 (q,
3JH-H=5 Hz, 1H).
19F NMR (CDCl3) δ: -
103
124.9 (br s, 1F, o-C6F5), -130.1 (br s, 1F, o-C6F5), -131.5 (br s, 1F, o-C6F5), -132.5 (br s, 1F, o-
C6F5), -134.9 (br s, 1F, o-C6F5), -155.8 (br s, 1F, p-C6F5), -159.0 (br s, 1F, p-C6F5), -161.7 (br s,
1F, p-C6F5), -162.2 (br s, 1F, m-C6F5), -163.0 (br s, 1F, m-C6F5), -166.0 (br s, 1F, m-C6F5), -
166.3 (br s, 1F, m-C6F5), -166.8 (br s, 1F, m-C6F5), -167.7 (br s, 1F, m-C6F5). 11
B NMR
(CDCl3) δ: -3.2. 13
C NMR (CDCl3) (partial) δ: 123.2 (dm, 1JC-F =227 Hz, CF), 125.9, 130.9
(dm, 1JC-F=296 Hz, CF), 136.5, 146.0 (dm,
1JC-F=279 Hz, CF), 153.2.
5.2.7: Metal-Free Catalytic Hydrogenations
A solvent bomb charged with 100 mg of substrate and the appropriate mass of B(C6F5)3 in
toluene (5 mL) was subjected to 3 freeze-pump-thaw cycles and exposed to 1 atm H2 at 77 K.
The solution was stirred under the prescribed conditions and then allowed to cool. After cooling,
5 mL of ethyl acetate was added and the solution was run through a plug of silica gel. Volatiles
were removed in vacuo. Analytical data for products 5-20,225
5-21,18,226,227
5-22,227
5-23226
and
5-24228
matched that previously published.
5.2.8: Reaction of Aminopyridines with Fluoroarylboranes
(5-Me)C5H3NH(2-NH)B(C6F5)3 (5-25) - 2-amino-6-picoline (4.5 mg, 0.04 mmol) was added to a
solution of B(C6F5)3 (20 mg, 0.039 mmol) in CH2Cl2 (2 mL), The solution was allowed to stand
for two hours, then all volatiles were removed and the residue was washed with pentane (2 x 2
mL). The resulting white solid was dried in vacuo. Yield: 23 mg (96%). X-Ray quality crystals
were grown from a layered solution of CDCl3/pentane at room temperature. Anal. Calcd for
C24H8BF15N2: C, 46.48%; H, 1.30%; N, 4.52%. Found: C, 46.30%; H, 1.18%; N, 5.02%.
1H NMR (CDCl3) δ: 2.13 (s, 3H, CH3), 6.07 (br s, 1H, amide N-H), 6.23 (dm,
3JH-H = 7 Hz, 1H),
6.55 (br d, 3JH-H = 9 Hz, 1H), 7.38 (dd,
3JH-H = 9 Hz,
3JH-H = 7 Hz), 8.65 (br s, 1H, pyridinium N-
H). 19
F NMR (CDCl3) δ: -133.7 (d, 3JF-F = 20 Hz, 6F, o-C6F5), -157.0 (t,
3JF-F = 19 Hz, 3F, p-
C6F5), -163.0 (br s, 6F, m-C6F5). 11
B NMR (CDCl3) δ: -11.1. 13
C NMR (CDCl3) (partial) δ:
19.3, 109.5, 114.5, 137.0 (dm, 1JC-F = 256 Hz, CF), 141.8, 142.3 (dm,
1JC-F = 242 Hz, CF), 148.1
(dm, 1JC-F = 240 Hz, CF), 155.2.
(5-Me)C5H3NH(2-NH)BCl(C6F5)2 (5-26) - 2-amino-6-picoline (28 mg, 0.26 mmol) was added to
a solution of ClB(C6F5)2 (100 mg, 0.29 mmol) in CH2Cl2 (5 mL). The solution was allowed to
stir overnight, the solvent was removed in vacuo and the residue was washed with hexanes (2 x 2
104
mL). X-Ray quality crystals were grown from the hexane wash layer. Yield: 119 mg (91%).
Anal. Calcd. for C18H8BClF15N2 (%) C: 44.25, H: 1.65, N: 5.73; found C: 44.19, H: 1.94, N:
5.55.
1H NMR (CDCl3) δ: 2.39 (s, 3H, CH3), 6.38 (d,
3JH-H=7 Hz, 1H), 6.46 (br s, 1H, NH), 6.58 (d,
3JH-H=9 Hz, 1H), 7.48 (dd,
3JH-H=9 Hz,
3JH-H=7 Hz, 1H, p-CH), 11.10 (br s, 1H, NH) ;
19F NMR
(CDCl3) δ: -133.5 (dd, 3JF-F=23 Hz,
4JF-F=8 Hz 6F, o-C6F5), -155.9 (t,
3JF-F=21 Hz, 3F, p-C6F5), -
162.3 (td, 3JF-F=21 Hz,
4JF-F=8 Hz, 6F, m-C6F5).
11B NMR (CDCl3) δ: -2.8.
13C NMR (CDCl3)
partial δ: 19.6, 110.4, 114.4, 142.5, 144.8, 155.2.
(5-Me)C5H3NH(2-NH)BH(C6F5)2 (5-27) - 2-amino-6-picoline (31 mg, 0.28 mmol) was added to
a solution of HB(C6F5)2 (100 mg, 0.29 mmol) in CH2Cl2 (5 mL). The solution was allowed to
stir overnight, the solvent was removed in vacuo and the residue was washed with hexanes (2 x 2
mL). Yield: 119 mg (91%). X-Ray quality crystals were grown from the hexane wash layer.
Anal. Calcd. for C18H9BF15N2 (%) C: 47.61, H: 2.00, N: 6.17; found C: 47.26, H: 2.38, N: 6.36.
1H NMR (CDCl3) δ: 2.37 (s, 3H, CH3), 3.85 (q,
1JH-B=94 Hz, 1H, BH), 6.18 (d,
3JH-H=7 Hz, 1H),
6.26 (br s, 1H, NH), 6.47 (d, 3JH-H=9 Hz, 1H), 7.32 (dd,
3JH-H=9 Hz,
3JH-H=7 Hz, 1H, p-CH), 9.85
(br s, 1H, NH) . 19
F NMR (CDCl3) δ: -135.4 (d, 3JF-F=23 Hz, 6F, o-C6F5), -159.4 (t,
3JF-F=20
Hz, 3F, p-C6F5), -164.1 (tm, 3JF-F=20 Hz, 6F, m-C6F5).
11B NMR (CDCl3) δ: -18.3 (d,
3JH-B=94
Hz). 13
C NMR (CDCl3) partial δ: 19.9, 108.7, 114.3, 141.5, 144.1, 155.2.
(5-Me)C5H3N(2-NH)B(C6F5)2 (5-28) – Iso-propylmagnesiumchloride (0.657 mL of a 2.0 M
solution in diethyl ether, 1.31 mmol) was added dropwise to a solution of 5-26 (642 mg, 1.31
mmol) in diethyl ether (10 mL). The cloudy solution was allowed to stir for 2 hours, hexanes (10
mL) was added and the solution was filtered. The filtrate was dried in vacuo. Yield: 543 mg
(92%). Anal. Calcd. for C18H7BF10N2 (%) C: 47.82, H: 1.56, N: 6.20; found C: 47.44, H: 1.98,
N: 6.23.
1H NMR (CDCl3) δ: 2.40 (s, 3H, CH3), 6.55 (d,
3JH-H=8 Hz, 1H), 6.94 (d,
3JH-H=8 Hz, 1H), 7.46
(t, 3JH-H=8 Hz, 1H, p-CH), 7.79 (br s, 1H, NH).
19F NMR (CDCl3) δ: -131.7(m, 4F, o-C6F5), -
148.9 (t, 3JF-F=21 Hz, 1F, p-C6F5), -152.1 (t,
3JF-F=20 Hz, 1F, p-C6F5), -160.9 (td,
3JF-F=21 Hz,
4JF-F=8 Hz, 2F, m-C6F5), -161.5 (td,
3JF-F=22 Hz,
4JF-F=8 Hz, 2F, m-C6F5).
11B NMR (CDCl3) δ:
36.0 (br s). 13
C NMR (CDCl3) partial δ: 24.0, 111.0, 120.1, 138.5, 152.3, 158.1.
105
(5-CF3)C5H3NH(2-NH)B(C6F5)3 (5-29) - 2-amino-6-(trifluoromethyl)pyridine (32 mg, 0.20
mmol) was added to a solution of B(C6F5)3 (100 mg, 0.20 mmol) in CH2Cl2 (5 mL). The
solution was allowed to stir overnight, the solvent was removed in vacuo and the residue was
washed with hexanes (2 x 2 mL). Yield: 124 mg (95%). X-Ray quality crystals were grown
from the hexane wash layer. Anal. Calcd. for C24H5BF18N2 (%) C: 42.76, H: 0.75, N: 4.16;
found C: 42.73, H: 0.95, N: 4.24.
1H NMR (CDCl3) δ: 6.74 (br s, 1H, NH), 6.87 (d,
3JH-H=7 Hz, 1H, CH), 7.01 (d,
3JH-H=9 Hz, 1H,
CH), 7.64 (dd, 3JH-H=8 Hz,
3JH-H=7 Hz, 1H, CH), 8.96 (br s, 1H, NH).
19F NMR (CDCl3) δ: -
67.8 (s, 3F, CF3), -133.3 (d, 3JF-F=22 Hz, 6F, o-C6F5), -155.3 (t,
3JF-F=21 Hz, 3F, p-C6F5), -164.1
(tm, 3JF-F=22 Hz, 6F, m-C6F5).
11B NMR (CDCl3) δ: -10.9.
13C NMR (CDCl3) partial δ: 108.5,
122.0, 137.4, (dm, 1JC-F=255 Hz, CF), 140.0, 148.1, (dm,
1JC-F=239 Hz, CF), 154.6.
(5-CF3)C5H3NH(2-NH)BCl(C6F5)2 (5-30) - 2-amino-6-(trifluoromethyl)pyridine (25 mg, 0.15
mmol) was added to a solution of ClB(C6F5)2 (59 mg, 0.15 mmol) in CH2Cl2 (5 mL). The
solution was allowed to stir overnight, the solvent was removed in vacuo and the residue was
washed with hexanes (2 x 2 mL). Yield: 73 mg (87%). X-Ray quality crystals were grown
from the hexane wash layer. Anal. Calcd. for C18H9BF15N2 (%) C: 39.85, H: 0.93, N: 5.16;
found C: 40.12, H: 0.93, N: 5.00.
1H NMR (CDCl3) δ: 7.07 (br m, 3H), 7.79 (t,
3JH-H=8 Hz,
3JH-H=7 Hz, 1H, p-CH), 11.59 (br s,
1H, NH). 19
F NMR (CDCl3) δ: -66.9 (s, 3F, CF3), -133.3 (br d, 3JF-F=19 Hz, 4F, o-C6F5), -
154.7 (br s, 2F, p-C6F5), -161.9 (br s, 4F, m-C6F5). 11
B NMR (CDCl3) δ: 0.4. 13
C NMR
(CDCl3) partial δ: 109.9, 117.8, 120.6, 121.3, 137.4 (dm, 1JC-F=255 Hz, CF), 140.6 (dm,
1JC-
F=255 Hz, CF), 140.9, 147.8, (dm, 1JC-F=244 Hz, CF), 154.9.
(5-CF3)C5H3N(2-NH)B(C6F5)2 (5-31) - 2-amino-6-(trifluoromethyl)pyridine (25 mg, 0.15 mmol)
was added to a solution of HB(C6F5)2 (53 mg, 0.15 mmol) in CH2Cl2 (5 mL). The solution was
allowed to stir overnight, the solvent was removed in vacuo and the residue was washed with
hexanes (1 mL). Yield: 58 mg (76%). Anal. Calcd. for C18H4BF13N2 (%) C: 42.72, H: 0.80, N:
5.54; found C: 42.34, H: 1.05, N: 5.83.
1H NMR (CDCl3) δ: 7.02 (d,
3JH-H=8 Hz, 1H), 7.46 (d,
3JH-H=8 Hz, 1H), 7.74 (br s, 1H, NH),
7.82 (t, 3JH-H=8 Hz, 1H, p-CH).
19F NMR (CDCl3) δ: -67.9 (s, 3F, CF3), -130.9 (br s, 4F, o-
106
C6F5), -147.6 (br s, 1F, p-C6F5), -150.7 (br s, 1F, p-C6F5), -160.2 (br s, 4F, m-C6F5). 11
B NMR
(CDCl3) δ: 37.0 (br s). 13
C NMR (CDCl3) partial δ: 116.8, 117.0, 119.4, 122.1, 137.6 (dm, 1JC-F
=255 Hz, CF), 139.9, 142.8 (dm, 1JC-F=264 Hz, CF), 147.4, (dm,
1JC-F=246 Hz, CF), 153.7.
C9H6N(8-NH2)B(C6F5)3 (5-32) - 8-aminoquinoline (28 mg, 0.19 mmol) was added to a solution
of B(C6F5)3 (100 mg, 0.19 mmol) in CH2Cl2 (5 mL). The light brown solution was allowed to
stir for 3 hours, the solvent was removed in vacuo and the residue was washed with hexanes (2 x
2 mL) and again dried in vacuo. Yield: 111 mg (87%). Anal. Calcd. for C27H8BF15N2 (%) C:
49.42, H: 1.23, N: 4.27; found C: 48.92, H: 1.23, N: 4.08.
1H NMR (CDCl3) δ: 7.55 (m, 2H), 7.67 (d,
3JH-H=7 Hz, 1H), 7.82 (d,
3JH-H=8 Hz, 1H), 8.25 (d,
3JH-H=8 Hz, 1H), 8.39 (br s, 2H, NH2), 8.84 (d,
3JH-H=4 Hz, 1H).
19F NMR (CDCl3) δ: -133.0
(br s, 6F, o-C6F5), -156.4 (t, 3JF-F=20 Hz, 3F, p-C6F5), -163.1 (t,
3JF-F=20 Hz, 6F, m-C6F5).
11B
NMR (CDCl3) δ: -5.8. 13
C NMR (CDCl3) partial δ: 122.4, 122.8, 126.4, 128.0, 131.4, 136.7,
149.9.
C9H6N(8-NH2)-BCl(C6F5)2 (5-33) - 8-aminoquinoline (25 mg, 0.17 mmol) was added to a
solution of ClB(C6F5)2 (66 mg, 0.17 mmol) in CH2Cl2 (5 mL). White precipitate was
immediately visible, the bright orange solution was allowed to stir for 3 hours, the solvent was
removed in vacuo and the residue was washed with hexanes (2 x 2 mL) and again dried in vacuo.
Yield: 78 mg (86%). Anal. Calcd. for C21H8BClF10N2 (%) C: 48.08, H: 1.54, N: 5.34; found C:
47.91, H: 1.35, N: 5.08.
1H NMR (CDCl3) δ: 6.76 (d,
3JH-H=7 Hz, 1H), 6.98 (d,
3JH-H=7 Hz, 1H), 7.50 (t,
3JH-H=8 Hz, 1H),
7.65 (dd, 3JH-H=7 Hz,
3JH-H=5 Hz , 1H), 8.43 (d,
3JH-H=8 Hz, 1H), 8.88 (d,
3JH-H=5 Hz, 1H).
19F
NMR (CDCl3) δ: -134.9 (br s, 6F, o-C6F5), -156.3 (br s, 3F, p-C6F5), -163.1 (br s, 6F, m-C6F5).
11B NMR (CDCl3) δ: 2.7 (br s).
13C NMR (CDCl3) was not obtained due to poor solubility.
C9H6N(8-NH)B(C6F5)2 (5-34) - 8-aminoquinoline (25 mg, 0.17 mmol) was added to a solution of
HB(C6F5)2 (60 mg, 0.17 mmol) in CH2Cl2 (5 mL). The bright red solution was allowed to stir
for 5 days, the solvent was removed in vacuo and the residue was washed with cold hexanes (2
mL) and again dried in vacuo. Yield: 64 mg (76%). Anal. Calcd. for C21H7BF10N2 (%) C:
51.68, H: 1.45, N: 5.74; found C: 51.44, H: 1.68, N: 5.60.
107
1H NMR (CDCl3) δ: 4.85 (br s, 1H, NH), 6.74 (d,
3JH-H=7 Hz, 1H), 6.97 (d,
3JH-H=8 Hz, 1H),
7.65 (t, 3JH-H=8 Hz, 1H), 8.42 (d,
3JH-H=8 Hz, 1H), 8.87 (d,
3JH-H=5 Hz, 1H).
19F NMR (CDCl3)
δ: -134.9 (dd, 3JF-F=21 Hz,
4JF-F=8 Hz, 6F, o-C6F5), -155.9 (t,
3JF-F=21 Hz, 3F, p-C6F5), -162.0
(tm, 3JF-F=21 Hz,
4JF-F=8 Hz, 6F, m-C6F5).
11B NMR (CDCl3) δ: 2.4 (br s).
13C NMR (CDCl3)
partial δ: 105.8, 108.9, 122.4, 129.1, 133.2, 140.4, 142.2, 147.9.
5.2.9: X-Ray Data Collection, Reduction, Solution and Refinement
Single crystals were mounted in thin-walled capillaries either under an atmosphere of dry N2 in a
glove box and flame sealed or coated in paratone-N oil. The data were collected using the
SMART software package on a Siemens SMART System CCD diffractometer using a graphite
monochromator with Mo Κα radiation (λ = 0.71073 Å). A hemisphere of data was collected in
1448 frames with 10 second exposure times unless otherwise noted. Data reductions were
performed using the SAINT software package and absorption corrections were applied using
SADABS. The structures were solved by direct methods using XS and refined by full-matrix
least-squares on F2 using XL as implemented in the SHELXTL suite of programs. All non-H
atoms were refined anisotropically. Carbon-bound hydrogen atoms were placed in calculated
positions using an appropriate riding model and coupled isotropic temperature factors.
Phosphorus-bound hydrogen atoms were located in the electron difference map and their
positions refined isotropically. Single crystal X-ray structures were obtained for 5-1, 5-2, 5-3, 5-
4, 5-5, 5-6, 5-7, 5-8, 5-9, 5-13, 5-14, 5-19, 5-25, 5-26, 5-27, 5-29 and 5-30 Selected
crystallographic data are included in Tables 5.1-5.6. Diagrams and selected bond lengths and
angles are provided in Table 5.7 and Figures 5.5, 5.6, 5.8, 5.11, 5.14, 5.15, 5.19, 5.21, 5.22 and
5.25.
108
Table 5.1: Selected crystallographic data for compounds 5-1, 5-2 and 5-3
Crystal 5-1 5-2-0.5 CHCl3 5-3
Formula C27H13BF15N C24.5H7.5BF15N C25H9BF15N
Formula weight 647.19 664.80 619.14
Crystal system Monoclinic Monoclinic Triclinic
Space group P21/n P21/n P-1
a(Å) 11.5466(4) 9.2940(8) 9.7020(9)
b(Å) 13.3174(5) 14.3404(14) 11.5301(11)
c(Å) 16.5172(6) 18.4081(18) 12.1581(11)
α(o) 90.00 90.00 105.149(5)
β( o) 92.0820(10) 98.874(4) 94.518(5)
γ( o) 90.00 90.00 113.003(4)
V (Å3) 2538.18(16) 2424.1(4) 1183.10(19)
Z 4 4 2
d(calc) g cm-1
1.694 1.822 1.738
Abs coeff, μ, cm-1
0.176 0.347 0.185
Data collected 6669 5542 7401
Data Fo2>3(Fo
2) 4924 3159 5857
Variables 397 401 379
Ra 0.0392 0.0528 0.0427
Rwb 0.0938 0.1270 0.1238
Goodness of Fit 1.020 1.007 1.093
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
109
Table 5.2: Selected crystallographic data for compounds 5-4, 5-5 and 5-6
Crystal 5-4 5-5-0.5 C7H8 5-6
Formula C28H9BF15N3 C32.5H13BF15N C28H8BF15N2
Formula weight 763.27 713.25 668.17
Crystal system Triclinic Monoclinic Monoclinic
Space group P-1 P21/n P21/c
a(Å) 9.6501(10) 9.0823(18) 10.706(2)
b(Å) 9.7264(8) 15.684(3) 16.332(3)
c(Å) 14.9925(11) 20.242(4) 14.368(3)
γ(o) 107.830(3) 90.00 90.00
β( o) 102.077(4) 101.05(3) 91.27(3)
γ( o) 98.166(3) 90.00 90.00
V (Å3) 1277.46(19) 2830.0(10) 2511.8(9)
Z 2 4 4
d(calc) g cm-1
1.776 1.674 1.767
Abs coeff, μ, cm-1
0.182 0.167 0.182
Data collected 6580 6440 4389
Data Fo2>3(Fo
2) 4246 4163 2499
Variables 424 466 381
Ra 0.0427 0.0522 0.0504
Rwb 0.1017 0.1506 0.1409
Goodness of Fit 1.020 1.045 1.006
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
110
Table 5.3: Selected crystallographic data for compounds 5-7, 5-8 and 5-9
Crystal 5-7 5-8 5-9
Formula C27H7BF15N C25H9BF15N C25H11BF15N
Formula weight 641.15 619.14 621.16
Crystal system Triclinic Monoclinic Monoclinic
Space group P-1 P21/n P21/c
a(Å) 9.9848(9) 12.8108(4) 17.8525(12)
b(Å) 10.9162(10) 13.4663(4) 9.8407(7)
c(Å) 11.6538(11) 13.3937(4) 15.2883(10)
α(o) 107.046(5) 90.00 90.0
β( o) 94.178(5) 100.156(2) 115.010(30)
γ( o) 101.061(5) 90.00 90.0
V (Å3) 1180.52(19) 2274.40(12) 2430.8(3)
Z 2 4 4
d(calc) g cm-1
1.804 1.808 1.697
Abs coeff, μ, cm-1
0.189 0.192 0.180
Data collected 5413 7942 8206
Data Fo2>3(Fo
2) 4093 6118 4499
Variables 397 379 383
Ra 0.0549 0.0494 0.0516
Rwb 0.0903 0.1475 0.1680
Goodness of Fit 1.017 1.055 1.005
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
111
Table 5.4: Selected crystallographic data for compounds 5-13, 5-14 and 5-19
Crystal 5-13 5-14 5-19
Formula C29H17BF15NO C31H17BF15NO4 C30H8BF15N2
Formula weight 691.25 763.27 692.19
Crystal system Orthorhombic Monoclinic Triclinic
Space group Pna21 P21/c P-1
a(Å) 17.4334(9) 11.3933(4) 12.4522(10)
b(Å) 10.6793(6) 18.7049(8) 12.8361(10)
c(Å) 14.4679(8) 15.0463(6) 16.6764(13)
α(o) 90.00 90.00 79.106(4)
β( o) 90.00 108.309(2) 79.436(4)
γ( o) 90.00 90.00 85.941(4)
V (Å3) 2693.6(3) 3044.2(2) 2571.3(4)
Z 4 4 4
d(calc) g cm-1
1.705 1.665 1.788
Abs coeff, μ, cm-1
0.175 0.170 0.182
Data collected 13352 8515 11857
Data Fo2>3(Fo
2) 10991 6553 7640
Variables 424 469 865
Ra 0.0383 0.0375 0.0444
Rwb 0.1146 0.1030 0.1166
Goodness of Fit 1.046 1.026 1.037
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
112
Table 5.5: Selected crystallographic data for compounds 5-25, 5-26 and 5-27
Crystal 5-25 5-26 5-27
Formula C24H8BF15N2 C18H8BClF10N2 C18H9BF10N2
Formula weight 620.13 488.52 454.08
Crystal system Triclinic Monoclinic Triclinic
Space group P-1 P21/c P-1
a(Å) 9.8535(13) 24.9904(18) 8.4351(16)
b(Å) 11.1355(16) 13.1642(11) 10.3150(19)
c(Å) 11.5291(16) 11.2738(8) 10.496(2)
α(o) 74.467(8) 90.00 100.424(9)
β( o) 73.964(7) 91.412(4) 91.554(9)
γ( o) 80.673(7) 90.00 109.413(8)
V (Å3) 1166.1(3) 3707.7(5) 843.3(3)
Z 2 8 2
d(calc) g cm-1
1.697 1.750 1.788
Abs coeff, μ, cm-1
0.189 0.313 0.183
Data collected 12524 8539 7306
Data Fo2>3(Fo
2) 7552 3644 5014
Variables 388 595 293
Ra 0.0416 0.0585 0.0396
Rwb 0.1383 0.1254 0.1210
Goodness of Fit 1.007 0.945 1.067
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
113
Table 5.6: Selected crystallographic data for compounds 5-29 and 5-30
Crystal 5-29 5-30
Formula C24H5BF18N2 C18H5BClF13N2
Formula weight 674.11 542.50
Crystal system Triclinic Monoclinic
Space group P-1 P21/n
a(Å) 9.8877(10) 10.2702(3)
b(Å) 11.1457(12) 31.6155(10)
c(Å) 12.4221(12) 12.3047(4)
α(o) 74.787(5) 90.00
β( o) 70.681(6) 92.436(2)
γ( o) 77.611(5) 90.00
V (Å3) 1234.3(2) 3991.7(2)
Z 2 8
d(calc) g cm-1
1.814 1.805
Abs coeff, μ, cm-1
0.203 0.321
Data collected 5606 9116
Data Fo2>3(Fo
2) 2902 6762
Variables 414 647
Ra 0.0448 0.0413
Rwb 0.1082 0.1024
Goodness of Fit 0.958 1.024
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
114
5.3: Results and Discussion
5.3.1: Reactions of Alkyl or Aryl-Substituted Pyridines with B(C6F5)3
While pyridine forms a strong adduct with B(C6F5)3, a detailed survey of the effects of
increased steric bulk of pyridine on its reactions with B(C6F5)3 has not been undertaken. Such a
study should reveal a limit between adduct formation and the generation of frustrated Lewis pairs
(FLPs). A series of pyridines of varying steric and electronic properties were added to B(C6F5)3.
Pyridine-borane adducts of 4-tert-butylpyridine (5-1), 2-methylpyridine (5-2), 2-ethylpyridine
(5-3), 2,2′-dipyridylamine (5-4), 2-phenylpyridine (5-5), 2,2′-dipyridyl (5-6) and quinoline (5-7)
were formed rapidly and quantitatively within 4 hours in CH2Cl2 (Figure 5.4). The adducts with
a substituent at the 2-position showed 15 different resonances in the 19
F NMR spectrum,
indicative of restricted rotation of the N-B and B-C bonds, as a result of crowding and/or
intramolecular H-F interactions which can also serve to restrict rotation.212
Crystal structures
were obtained for adducts 5-1 – 5-7. Selected metrical parameters are compared in Table 5.7 and
depictions of selected structures are presented in Figures 5.5 and 5.6. Remote substitution shows
mainly electronic effects on B-N bond length as the B(C6F5)3 adducts of 4-(Me2N)C5H4N and 4-
(CH3)3C-C5H4N (5-1) exhibited shorter B-N bond lengths than analogous adduct of the
unsubstituted pyridine.
Figure 5.4: Synthesis of Lewis acid-base adducts 5-2 – 5-7; R=Me (5-2), Et (5-3), R=N(H)(2-
C5H4N) (5-4), Ph (5-5), 2-C5H4N (5-6)
115
Figure 5.5: POV-Ray Depictions of 5-3 (left) and 5-4 (right). Carbon: black, Boron: yellow-
green, Fluorine: deep pink, Nitrogen: blue. Hydrogen atoms are omitted for clarity. Selected
metrical parameters (distances: Å, angles: °): 5-3: B1-N1 1.638(2), B1-C1 1.642(2), B1-C7
1.648(2), B1-C13 1.643(2), N1-B1-C1 102.91(11), N1-B1-C7 110.97(11), C1-B1-C7
112.54(12), C1-B1-C13 116.30(12), C13-B1-C7 103.00(11). 5-4: B1-N1 1.629(2), B1-C1
1.651(3), B1-C7 1.650(3), B1-C13 1.651(3), N1-B1-C1 112.04(13), N1-B1-C7 110.77(14), N1-
B1-C13 102.54(14), C1-B1-C7 102.95(14), C1-B1-C13 112.06(15), C7-B1-C13 116.80(14).
Figure 5.6: POV-Ray Depictions of 5-5-0.5 C7H8 (left) and 5-7 (right). Carbon: black, Boron:
yellow-green, Fluorine: deep pink, Nitrogen: blue. Solvent and hydrogen atoms are omitted for
clarity. Selected metrical parameters (distances: Å, angles: °) 5-5: N1-B1 1.650(3), C1-B1
1.659(3), B1-C7 1.641(3), B1-C13 1.646(3), N1-B1-C1 111.01(16), N1-B1-C7 112.23(17), N1-
116
B1-C13 103.23(16), C1-B1-C7 101.78(17), C1-B1-C13 113.60(17), C7-B1-C13 115.30(18). 5-6:
N1-B1 1.641(2), B1-C1 1.646(2), B1-C7 1.651(2), B1-C13 1.651(2), N1-B1-C1 111.79(12), N1-
B1-C7 110.28(13), N1-B1-C13 102.70(12), C1-B1-C7 102.56(12), C1-B1-C13 116.39(13), C7-
B1-C13 113.35(12).
Table 5.7: Selected NMR spectroscopic and X-ray crystallographic data obtained for pyridine-
borane adducts 5-1 to 5-8
Pyridine Substituent(s) 11
B NMR
δ:
Δ m-p 19
F
NMR δ:a
N-B (Å) Σ C-B-C (°)
None31,229
-3.6 6.4 1.628(2) 333.84
4-NMe2230
-5.3 5.9 1.602(6) 333.4
4-tert-butyl (5-1) -4.1 6.4 1.618(2) 334.43
2-methyl (5-2) -3.6 7.0 1.639(4) 332.1
2-ethyl (5-3) -3.6 6.8 1.638(2) 331.84
2,2′-pyridylamine (5-4) -5.1 7.0 1.629(2) 331.81
2-phenyl (5-5) -2.9 7.2 1.651(4) 330.7
2,2′-pyridine (5-6) -2.7 6.3 1.649(5) 330.9
quinoline (5-7) -3.2 6.8 1.641(2) 332.30
2,6-dimethyl (5-8) -3.9 6.7 1.661(2) 328.30
a This value is the difference in chemical shift between the meta and para-fluorines, and has been
noted to be characteristic for different bonding environments at boron in fluoroarylboranes (the
shortest m-p gaps are generally found for 4 coordinate fluoroaryl borates, while the largest m-p
gaps are found for neutral 3 coordinate fluoroaryl boranes).179,180
Substitution at the 2-position, however, shows a substantial effect on B-N bond lengths.
The addition of a single small inductive donor at this position, such as methyl or ethyl, results in
noticeable elongation in B-N bond length compared to pyridine-B(C6F5)3, suggesting that in this
117
case the steric effect dominates the electronic effect. The B-N bond length of the 5-4 is slightly
shorter than that of pyridine-B(C5F5)3 reflecting slightly increased nucleophilicity of the pyridyl
nitrogen due to electron donation from the amine nitrogen to the pyridine ring. The flexibility of
the substituent allows it to point away from the C6F5 groups, minimizing steric conflict. The
presence of electron-withdrawing substituents at the 2-position results in a more dramatic
elongation of the B-N bond as exemplified by adducts 5-5 - 5-7.
As shown in Table 5.7, 11
B NMR chemicals shift and meta-para gaps in the 19
F NMR
spectrum are relatively constant throughout the series of adducts, while B-N bond lengths and
the sum of C-B-C bond angles vary in a fairly consistent fashion. As discussed above, the B-N
bond lengths can be rationalized in terms of bulk and basicity, while the sum of C-B-C bond
angles decreases with increasing bulk of the pyridine. Bulkier pyridines force the C6F5 groups
closer together, resulting in a more idealized tetrahedral geometry at boron (tetrahedral geometry
would have the sum of C-B-C angles at 328.5, which is nearly exactly what is observed for 5-8).
5.3.2: Frustrated Lewis Pairs of Pyridines with B(C6F5)3
The reaction of 2,6-lutidine with B(C6F5)3 presented a unique scenario. At room
temperature the solution combination of 2,6-lutidine with B(C6F5)3 gave a 19
F NMR spectrum
with extremely broad peaks, including signals characteristic of B(C6F5)3, suggesting that an
equilibrium exists between adduct and the free Lewis acid and Lewis base (Figure 5.7). The 1H
NMR spectrum showed resonances corresponding to 2,6-lutidine as well as a new set of
resonances attributed to a Lewis acid-base adduct, confirming that there is an equilibrium
between adduct 5-8 formation and free Lewis acid and Lewis base at room temperature (Keq=3.3
M-1
in CD2Cl2). A crystal structure was obtained for the adduct 5-8 confirming the anticipated
connectivity (Figure 5.8).
Figure 5.7: Equilibrium observed in solution between the FLP and Lewis acid-base adduct 5-8
118
Figure 5.8: POV-Ray depiction of 5-8. Carbon: black, Boron: yellow-green, Fluorine: deep pink,
Nitrogen: blue. Hydrogen atoms are omitted for clarity. Selected metrical parameters
(distances: Å, angles: °) N1-B1 1.661(2), B1-C1 1.660(2), B1-C7 1.643(2), B1-C13 1.651(2),
N1-B1-C1 111.11(12), N1-B1-C7 104.00(11), N1-B1-C13 113.84(12), C1-B1-C7 115.59(13),
C1-B1-C13 99.30(12), C7-B1-C13 113.41(12).
Monitoring the equilibrium by variable temperature NMR spectroscopy yielded
thermodynamic data for the process: ΔH = -42(1) kJ/mol; ΔS = -131(5) J/(mol*K); ΔG° = -3
kJ/mol (see experimental section for van’t Hoff plot). These data indicate that the reaction is
only slightly exothermic at room temperature. The crystal structure of 5-8 supports the
experimental data suggesting that formation of adduct 5-8 is not as favourable as formation of
the adducts formed with smaller pyridines. 5-8 showed the longest B-N bond length of all
crystallized pyridine-B(C6F5)3 adducts.31
The pyridines that either did not react with B(C6F5)3 or only showed weak interaction
were those substituted at both the 2- and 6-positions (2,6-lutidine and 2,6-diphenylpyridine) or
with a single very bulky substituent at the 2-position (2-tert-butylpyridine).
These results are consistent with the trends in nucleophilicity observed by Brown and co-
workers, who concluded that while the order of basicity of a series of substituted pyridines was
2,6-lutidine > 2-methylpyridine > 2-tert-butylpyridine > pyridine; the order of nucleophilicity
was pyridine > 2-methylpyridine > 2,6-lutidine > 2-tert-butylpyridine (see Figure 1.3).33
119
5.3.4: Small Molecule Activation by Pyridine-Borane FLPs
To explore the reactivity of the novel FLPs, the activation of H2 at room temperature was
attempted. The pyridine-borane FLPs were exposed to 4 atm H2 and stirred overnight.
Multinuclear NMR spectroscopy revealed quantitative formation of the novel ion pairs 5-9, 5-10,
and 5-11 (Figure 5.9, see Figure 5.10 for NMR spectra of 5-9).
Figure 5.9: Synthesis of ion pairs 5-9 (R=R1=Me), 5-10 (R=
tBu, R
1=H) and 5-11 (R=R
1=Ph)
A closer study of the activation of H2 revealed that, in all cases, quantitative reaction
occurred in 2 hours at 1 atm H2. Ion pair 5-9 was further characterized by X-ray crystallography
(Figure 5.11). The structure revealed a short N-H - - - H-B contact of 1.862 Å. This contact is
substantially shorter than those observed in related phosphonium borate species.38,43
This type of
short N-H - - - H-B contact has been noted in other related species, primarily amine-borane
adducts, and has been proposed to contribute to facile loss of H2 in these species.231
120
Figure 5.10: Multinuclear NMR spectra for 5-9 in CD2Cl2: A: 1H, B:
11B and C:
19F.
Figure 5.11: POV-Ray depiction of 5-9. Carbon: black, Hydrogen: white, Boron: yellow-green,
Fluorine: deep pink, Nitrogen: blue. Carbon-bound hydrogen atoms are omitted for clarity.
Selected metrical parameters (distances: Å, angles: °) B1-C1 1.640(3), B1-C7 1.643(3), B1-C13
1.637(3), H1 - - H1a 1.862, C1-B1-C7 112.59(16), C1-B1-C13 116.46(16), C7-B1-C13
107.71(16), N1-H1a-H1 169.09, B1-H1-H1a 150.48.
A B
NH
p-CH
m-CH
BH
CH3
o-C6F5
C
1JB-H=92 Hz
p-C6F5
m-C6F5
121
The donor 2,3,5,6-tetramethylpyrazine also showed no reaction with B(C6F5)3 at room
temperature and this FLP was capable of H2 activation, however, the resulting ion pair 5-12
could not be isolated. This species showed decomposition to multiple products. 1,2-
hydrogenation, resulting from borohydride attack at the 2-position of the pyridinium cation of 5-
12, is one suspected reaction pathway, however this could not be determined conclusively.
In an effort to promote loss of H2, ion pairs 5-9 to 5-11 were subjected to controlled
heating. After 6 hours at 80°C, ~35% loss of H2 was observed in all cases. The reaction stopped
at this point as presumably the opposite reaction, the activation of H2, prevents further loss of H2
from solution. Addition of 1 equivalent of a smaller base, pyridine in this case, in order to
quench the free borane after loss of H2 to prevent the hydrogen activation reaction,213
results in
quantitative loss of H2 over 10 hours at 80°C in toluene (Figure 5.12).
Figure 5.12: H2 loss from ion pairs 5-9 – 5-11
To probe the utility of 2,6-lutidine as a nucleophile in FLP chemistry, B(C6F5)3 and 2,6-
lutidine were stirred in a THF solution. The FLP was able to effect THF ring-opening,
analogous to that previously observed with phosphines and B(C6F5)340
(Figure 5.13). The
product, 5-13, was fully characterized, including by X-ray crystallography (Figure 5.14). The
newly-formed B-O bond length of 1.4584(14) Å is similar to the phosphonium borate zwitterions
formed by ring-opening THF while the new N-C bond at 1.4843(16) Å, as expected, is
significantly shorter than the related P-C bonds found in the analogous phosphonium borate
zwitterions.40
Figure 5.13: Formation of zwitterion 5-13 by THF ring-opening
122
Figure 5.14: POV-Ray depiction of 5-13. Carbon: black, Boron: yellow-green, Fluorine: deep
pink, Nitrogen: blue, Oxygen: red. Hydrogen atoms are omitted for clarity. Selected metrical
parameters (distances: Å, angles: °) B1-O1 1.4584(14), B1-C1 1.6711(16), B1-C7 1.6575(15),
B1-C13 1.6670(15), N1-C22 1.4843(16), O1-C19 1.4122(13). O1-B1-C1 111.57(9), O1-B1-C7
108.53(8), O1-B1-C13 106.65(9), C1-B1-C7 102.68(8), C1-B1-C13 114.30(8), C7-B1-C13
113.07(8), C24-N1-C28 122.05(10), C24-N1-C22 118.86(10), C28-N1-C22 119.09(10), C19-
O1-B1 119.01(8).
5.3.5: FLPs of Bulky Pyridines with Other Lewis Acids
In an effort to expand the scope of the H2 activation reactions, other Lewis acids were
coupled with the bulky pyridines. Based on the lack of reaction of 2,6-lutidine with BMe3,32
BEt3 was used as it is sterically and electronically similar to BMe3 but exists as a liquid at room
temperature, while BMe3 is a highly reactive gas. As expected, there was no adduct formation
with 2,6-lutidine or 2-tert-butylpyridine. However, charging solutions of these FLPs with 4 atm
H2 resulted in no reaction. Presumably, the Lewis acidity of the trialkyl borane is insufficient to
allow the cleavage of H2 by the pair, however, an equilibrium favouring H2 loss cannot be ruled
out. The notion that this pair lacks the potential for hydrogen activation was later confirmed
computationally by Papai and co-workers.50
123
5.3.6: Reaction of Diethyl-2,6-dimethyl-3,5-pyridinedicarboxylate with B(C6F5)3
Diethyl-2,6-dimethyl-3,5-pyridinedicarboxylate, the product of loss of H2 from
Hantzsch’s Ester, is very sterically similar to 2,6-lutidine at the pyridine nitrogen but exhibited
dramatically different reactivity with B(C6F5)3. The new species formed in the reaction, 5-14,
showed 19
F NMR resonances with a meta-para gap of 12.4 ppm, a relatively large value for a
Lewis acid-base adduct, suggesting the dative interaction is fairly weak.179,180
The high
symmetry of this spectrum, where only one three resonances were observed, suggests that the
bonding does not take place at nitrogen (analogous with 5-8). Based on steric factors, the most
likely site of adduct formation is a carbonyl oxygen atom of one of the ester groups. The 1H
NMR spectrum showed equivalent methyl and ethyl groups suggesting that there is a rapid
equilibrium, if the boron centre is oxygen-bound, with the borane switching between ester
groups. X-ray analysis confirmed that the borane does indeed bind a carbonyl group (Figure
5.15). The B-O bond length of the 5-14 is comparable to other bond lengths of carbonyl-
B(C6F5)3 adducts.30,232
For example, the B-O bond length of 1.5877(16) Ǻ observed in the
crystallographically-determined structure for 5-14 is slightly shorter than that found in the very
similar ester-bound adduct C6H5COOEt-B(C6F5)3 at 1.594(6) Ǻ.232
124
Figure 5.15: POV-Ray depiction of 5-14. Carbon: black, Boron: yellow-green, Fluorine: deep
pink, Nitrogen: blue, Oxygen: red. Hydrogen atoms are omitted for clarity. Selected metrical
parameters (distances: Å, angles: °) O1-B1 1.5885(16), O1-C25 1.2473(15), B1-C1 1.6441(18),
B1-C7 1.6331(18), B1-C13 1.6281(18), C25-O1-B1 140.80(10), O1-B1-C1 101.20(9), O1-B1-
C7 110.35(10), O1-B1-C13 106.31(9), C1-B1-C7 114.70(10), C1-B1-C13 109.02(11), C7-B1-
C13 114.17(10).
Adduct 5-14 was exposed to 4 atm H2 at room temperature in toluene-d8. Over several
days a mixture of two products and the starting material was observed by multinuclear NMR
(Figure 5.16). The major product, at 51%, showed several 1H NMR resonances characteristic of
1,2-hydrogenation product (in particular a doublet at 1.10 ppm for CH-CH3) 5-16, while 9%
showed the characteristic NMR signals for ion pair 5-15 (particularly the NH peak at 13.05 ppm)
and the remainder was unreacted starting material. This reactivity suggests that H2 activation by
5-14 is not favourable, likely due to electron-withdrawing substituents on the pyridine and
competing adduct formation at the ester groups. In addition, the strong electron-withdrawing
groups also make the pyridyl ring quite electrophilic, which prompts hydride transfer from
B(C6F5)3.
125
Figure 5.16: Partial hydrogenation of adduct 5-14
To further explore the reactivity of ion pair 5-15, another approach to its synthesis was
undertaken. Hydride abstraction by B(C6F5)3 on Hantzsch’s Ester showed predominantly
formation of 5-15 at -20°C, which rapidly rearranged to 5-16 at room temperature (Figure
5.17).224
Figure 5.17: Generation of 5-15 and 5-16 by hydride abstraction from Hantzsch’s Ester
The observed instability of the 5-15 at room temperature led to attempts at conducting the
H2 activation at lower temperatures. At -30°C, addition of 4 atm H2 to 5-14 showed no
reactivity, while at -15°C a similar product distribution was observed after several weeks. These
observations suggest that the activation energy required for H2 activation by the FLP is higher
than the energy required for hydride insertion, resulting in very little of 5-15 being observed in
solution from the H2 activation reaction. No evidence of 1,4-hydrogenation was observed under
any condition, suggesting that the 2-position is more electrophilic than the 4-position and the
methyl group is not sufficiently bulky to hinder hydride transfer to this carbon.
5.3.7: Reactions of Substituted Quinolines with B(C6F5)3
Next, substituted quinolines were examined for FLP reactivity. These substrates were
chosen for a number of reasons. For example, hydrogenated quinolines are of great interest in
natural product synthesis233
and can be used, similarly to Hantzsch’s Ester, as stoichiometric
126
sources of H2.218-223
Quinolines have been hydrogenated by a variety of methods including Birch
reduction,225,234
transition metal catalyzed hydrogenation,18,235-238
transfer hydrogenation,239,240
stoichiometric reduction using a borohydride source241,242
and H2 transfer from Hantsch’s
Ester.243,244
As quinoline was found to form a strong adduct with B(C6F5)3 (5-7), which did not
react with H2, the steric bulk at nitrogen must be increased. This could be accomplished with
substitution at the 2 or 8 positions. 8-methylquinoline, 2-methylquinoline, 2-phenylquinoline,
1,10-phenanthroline and acridine were chosen due to their steric bulk and commercial
availability.
Stoichiometric reaction of these species with B(C6F5)3 showed a range of reactivity. The
donors 8-methylquinoline and 2-phenylquinoline showed no interaction with B(C6F5)3 while 2-
methylquinoline and acridine showed evidence of adduct formation (5-17 and 5-18, respectively,
Figure 5.18) and free B(C6F5)3 characteristic of equilibrium between adduct and frustrated Lewis
pair (Keq~26.7 M-1
and 19.1 M-1
, respectively, by 19
F NMR spectroscopy at room temperature in
CDCl3). These equilibrium constants are significantly higher than that observed in the case of
2,6-lutidine, suggesting that reduced steric hindrance in the quinoline derivatives results in more
favourable B-N bonding. 1H NMR spectra for both of these reactions showed only one set of
very broad peaks suggesting rapid exchange between adduct and free Lewis acid and base.
Figure 5.18: Equilibria involving the formation of adducts 5-17 and 5-18
Reaction of B(C6F5)3 with 1,10-phenanthroline in CH2Cl2 after 4 hours showed
quantitative formation of the Lewis acid-base adduct 5-19. In this case 15 inequivalent
resonances were observed in the 19
F NMR spectrum, typical of asymmetrically substituted
127
pyridines. Adduct 5-19 was also characterized crystallographically (Figure 5.19) and revealed an
extraordinarily long B-N bond length (1.691(3) and 1.692(3) Å for two independent molecules),
significantly longer than that observed in 5-9, indicative of decreased nucleophilicity of 1,10-
phenanthroline versus 2,6-lutidine.
Figure 5.19: POV-Ray depiction 5-19 (one of two crystallographically independent molecules).
Carbon: black, Boron: yellow-green, Fluorine: deep pink, Nitrogen: blue. Hydrogen atoms are
omitted for clarity. Selected metrical parameters (distances: Å, angles: °): N1-B1 1.692(3), B1-
C1 1.663(3), B1-C7 1.626(3), B1-C13 1.636(3), C1-B1-C7 100.24(17), C1-B1-C13 112.25(17),
C7 B1 C13 118.89(17).
These pairs of bulky quinolines with B(C6F5)3 were exposed to H2 and, surprisingly,
these did not exhibit typical NMR spectroscopic data for clean pyridinium borate ion pairs (such
as 5-8 to 5-11). Rather, in all cases, some degree of hydrogenation of the quinoline backbone
was seen. As free B(C6F5)3 was generally observed in solution during the reaction, catalytic
hydrogenations were attempted.
Optimal conditions were found for the catalytic hydrogenations of acridine, 2-
methylquinoline, 2-phenylquinoline, 8-methylquinoline and 1,10-phenanthroline; and good
yields of products 5-20 to 5-24 were easily isolated by flash chromatography eluting with 1:1
toluene:ethyl acetate (Table 5.8).
128
Table 5.8: Catalytic hydrogenation of quinolines (reactions were conducted under 4 atm H2 in
toluene in a sealed Teflon-capped Schlenk bomb)
Substrate Mol %
B(C6F5)3
Time
(h)
Temp.
(°C) Product
% yield
(isolated)
Acridine
5 2 25
80
2-Methylquinoline
5 16 50
74
2-phenylquinoline
5 4 25
80
8-methylquinoline
10 6 50
88
1,10-
phenanthroline
5 3 80
84
Acridine was reduced under very mild conditions, with quantitative conversion to 5-20
seen in only 2 hours at room temperature in toluene under 4 atm H2. 1,10-phenanthroline
required the highest temperature for conversion, perhaps due to the stronger adduct formation
observed in the stoichiometric reaction with B(C6F5)3. Interestingly, only one of the pyridyl
rings of 1,10-phenanthroline could be reduced by this method, even under harsh conditions
(toluene, 4 atm H2, 100°C, 7 days).
129
5.3.8: Reactions of 2-Aminopyridines with Fluoroarylboranes
2-amino-6-picoline is an intriguing Lewis base for reactions with boranes. While the
steric bulk at the more basic (pyridyl) nitrogen is almost identical to that of 2,6-lutidine, the more
accessible nitrogen centre is the significantly less basic (arylamino) nitrogen centre. This raises
the question of whether the Lewis acid will react with the more basic (pyridyl) or more
accessible (arylamino) nitrogen centre.
Reaction of 2-amino-6-picoline with B(C6F5)3 showed quantitative formation of a new
product, 5-25. The 1H NMR spectrum for 5-25 proved to be particularly diagnostic as 2 separate
N-H peaks were observed at 6.07 and 6.85 ppm. The 11
B and 19
F NMR spectra were
characteristic of a 4 coordinate anionic borate. Together, these data suggest that while the initial
coordination has occurred at the amine nitrogen, the more basic pyridyl nitrogen subsequently
deprotonated the amine centre (Figure 5.20). X-ray crystallography confirmed the anticipated
connectivity of the molecule (Figure 5.21, left).
Figure 5.20: Formation of zwitterion 5-25
Based on the facile formation of 5-25, other reactions were envisioned between 2-amino-
6-picoline and fluoroaryl boranes. Reaction of 2-amino-6-picoline with ClB(C6F5)2 and
HB(C6F5)2 cleanly produced the zwitterions 5-26 and 5-27, respectively (Figures 5.21 and 5.22).
Connectivity of these species was also confirmed by X-ray crystallography. Compound 5-27
shows a N-H - - - H-B contact of 2.095 Å, which is shorter than those observed for phosphonium
borates38,43
but longer than that observed for 5-9.
B-N bond lengths for compounds 5-25 – 5-27 range from 1.536(6) to 1.5596(12) Å,
similar to those observed for other amido-fluroarylborate anions44,245,246
and significantly shorter
than those seen in amine-borane adducts.31
130
Figure 5.21: POV-Ray depictions of 5-25 and 5-26 (one of two crystallographically independent
molecules). Carbon: black, Hydrogen: white, Boron: yellow-green, Chlorine: aquamarine,
Fluorine: deep pink, Nitrogen: blue. Carbon-bound hydrogen atoms are omitted for clarity.
Selected metrical parameters (distances: Å, angles: °). 5-25: B1-N2 1.5596(12), N1-C19
1.3601(13), C19-N2 1.3240(13), B1-N2-C19 127.24(8), N1-N2 119.71(8). 5-26: N2-B1
1.536(6), B1-Cl1 1.945(4), H1 - - - Cl1 2.553, B1-N2-C13 129.6(4), N2-C13-N1 120.0(4), B1-
Cl1-H1 70.18, N1-H1-Cl1 131.82.
Figure 5.22: POV-Ray depiction of 5-27. Carbon: black, Hydrogen: white, Boron: yellow-green,
Fluorine: deep pink, Nitrogen: blue. Carbon-bound hydrogen atoms are omitted for clarity.
Selected metrical parameters (distances: Å, angles: °). B1-N2 1.5524(13), N2-C13 1.3572(13),
C13-N1 1.3572(12), H1- - - H1a 2.084, B1-H1-H1a 96.13, N1-H1a-H1 123.84, B1-N2-C13
123.72(8), N1-C13-N2 118.82(8).
131
Compound 5-27 is a linked pyridinium borohydride, which could be the product of
hydrogen activation by a linked pyridine-borane FLP. To determine if 5-27 could lose H2, to
produce a linked pyridine-borane FLP, 5-27 was heated to 80°C for 3 hours. Unfortunately, a
complex mixture of products was observed.
Another approach was taken to obtain the desired linked FLP. Adduct 5-27 was reacted
with one equivalent of iPrMgCl. This reaction showed clean and quantitative loss of HCl from
5-27, forming the desired linked pyridine-borane 5-28 (Figure 5.23).
Figure 5.23: Synthesis of linked pyridine-borane 5-28
The multinuclear NMR spectra were diagnostic for this species (Figure 5.24), particularly
the 19
F NMR spectrum which showed a large meta-para gap of 10.7 ppm typical of a weak
Lewis acid-base adduct (donation by the lone pair of N reduces the meta-para gap),179,180
along
with separate resonances for meta and para fluorines from each C6F5 ring, indicative of restricted
rotation about the B-N bond. This restricted rotation is expected due to the ability of the amide
nitrogen to donate electron density to the vacant p-orbital of the boron centre.
132
Figure 5.24: Multinuclear NMR spectra for 5-28 in CDCl3. A: 1H, B:
11B, C:
19F
In an effort to reduce the basicity at the pyridyl nitrogen and possibly allow for direct H2
loss upon reaction of the pyridylamine with HB(C6F5)3, reactions of the base 2-NH2-6-
(CF3)C5H3N with fluoroarylboranes were examined.
In reactivity analogous to that observed with 2-amino-6-picoline, reaction of 2-NH2-6-
(CF3)C5H3N with B(C6F5)3 and ClB(C6F5)2 produced zwitterions 5-29 and 5-30, respectively
(Figure 5.25). The B-N bond lengths of 5-29 and 5-30 at 1.564(3) and 1.534(3) Å, respectively
are very similar to those observed for the 2-amino-6-picoline analogs 5-25 and 5-26.
A B
C
NH
CH
CH3
o-C6F5
p-C6F5
m-C6F5
133
Figure 5.25: POV-Ray depictions of 5-29 and 5-30. Carbon: black, Hydrogen: white, Boron:
yellow-green, Chlorine: aquamarine, Fluorine: deep pink, Nitrogen: blue. Carbon-bound
hydrogen atoms are omitted for clarity. Selected metrical parameters (distances: Å, angles: °).
5-29: B1-N2 1.564(3), N2-C19 1.318(3), N1-C19 1.359(3), C19-N2-B1 127.4(2), N1-C19-N2
119.4(2). 5-30: B1-N2 1.534(3), B1-Cl1 1.929(2), N2-C13 1.325(3), N1-C13 1.355(3), Cl1 - - -
H1 2.457, N2-B1-Cl1 105.45(14), B1-N2-C13 125.82(18), N2-C13-N1 119.62(19), B1-Cl1-H1
70.52, N1-H1-Cl1 134.77.
Reaction of 2-NH2-6-(CF3)C5H3N with HB(C6F5)3 over the course of 24 hours in CH2Cl2
showed clean conversion to the linked pyridine-borane 5-31. The intermediate zwitterionic
pyridinium borate is observed by multinuclear NMR spectroscopy over the course of the reaction
(Figure 5.26). As is the case with 5-28, inequivalent C6F5 rings are observed in the 19
F NMR
spectrum of 5-31 due to restricted rotation about the N-B bond.
This reaction is reminiscent of work by Piers and co-workers, who demonstrated that the
ortho-substituted ammonium borate 1-(Ph2HN)-2-(BH(C6F5)2)C6H4 rapidly loses H2 to form the
linked amine-borane 1-(Ph2N)-2-(B(C6F5)2)C6H4.247
134
Figure 5.26: Formation of linked pyridine-borane 5-31
8-aminoquinoline was reacted with B(C6F5)3 and ClB(C6F5)2. Interestingly, in these
reactions, the proton transfer reaction observed in the formation of compounds 5-25, 5-26, 5-29
and 5-30 was not seen. Instead, 8-aminoquinoline formed the classical Lewis acid-base adducts
5-32 and 5-33 (Figure 5.27). Reaction of 8-aminoquinoline with HB(C6F5)2 over 2 days in
CH2Cl2 gave a bright red solution with NMR data consistent with loss of H2 and the formation of
intramolecular Lewis acid-base adduct 5-34 (Figure 5.27). This reaction likely occurs due to the
thermodynamically favourable formation of the 5-membered ring. The analogous N-B-O species
has been previously synthesized by reaction of 8-hydroxyquinoline with ClB(C6F5)2 with
concomitant loss of HCl.248
Figure 5.27: Formation of Lewis acid-base adducts 5-32, 5-33 and 5-34
5.4: Conclusions
Mixtures of pyridines with B(C6F5)3 exhibit a range of reactivity from simple adduct
formation, through equilibrium between adduct and FLP, to classical FLPs (where no interaction
between Lewis acid and Lewis base is observed). Despite the reduced basicity of pyridines
compared to most amines, as a result of the sp2 hybridization at nitrogen, FLP-type hydrogen
activation is possible. THF ring-opening has also been seen with a pyridine borane FLP,
producing a zwitterionic pyridinium borate. Substituted quinolines are capable of hydrogen
activation; however, the resulting ion pairs react further to reduce the pyridyl ring and catalytic
hydrogenation can be performed with a number of these species. These represent the first metal-
free catalytic hydrogenations which involve the addition of two equivalents of H2 to an
135
unprotected substrate and could be of considerable interest commercially as tetrahydroquinolines
are important intermediates in natural product synthesis.
Reaction of bulky amino-pyridines with B(C6F5)3 or ClB(C6F5)2 can result in
coordination at the amine nitrogen and proton transfer to the pyridyl nitrogen. The analogous
reaction with HB(C6F5)2 can result in the same proton transfer or loss of H2, depending on the
basicity of the pyridyl nitrogen. This creates a linked pyridinium borate or pyridine borane in a
single step, offering an easy synthesis of linked frustrated Lewis pairs. The proximity of the
nitrogen and boron atoms to one another may allow for more facile activation of small
molecules. This close relationship could also encourage loss of H2 from the pyridinium borate
which could facilitate the catalytic hydrogenation of less polar substrates than those previously
hydrogenated using FLPs.
136
Chapter 6: Summary and Conclusions
A series of small molecule activations have been conducted utilizing both transition metal
catalysis and novel frustrated Lewis pairs. These reactions have been studied in detail and a
number of new reaction modes have been discovered. This has allowed for the discovery of
novel synthetic routes to potentially valuable compounds.
A study of the Rh-catalyzed dehydrocoupling of P-H bonds revealed a new pathway
involving P-P bond activation. This reaction has been exploited in the synthesis of new silyl
phosphines. Reaction of P5Ph5 or P5Et5 with NacNacRhCOE(N2) resulted in complexes
involving bidentate and tridentate coordination modes of the P5 fragment. As is often the case
with phosphorus-based ligands, the substituents at phosphorus profoundly affect the nature of the
product.
The work presented in this thesis demonstrates that the FLP concept is not confined to
tertiary phosphine-borane pairs. The concept is in fact quite general and applies to a wide range
of nucleophiles of varying size and basicity. Bimolecular FLPs of polyphosphines and pyridines
and with B(C6F5)3 can be utilized to effect small molecule activation reactions. Perhaps most
interesting is the observation that covalently-bound phosphinoboranes are also capable of the
activation of H2.
FLPs composed of catena-polyphosphines and B(C6F5)3 have been shown to activate H2
and undergo para-nucleophilic aromatic substitution in a similar fashion to tertiary
phosphine/B(C6F5)3 pairs, yielding zwitterionic phosphonium borates. The FLP P5Ph5/B(C6F5)3
showed new reactivity with H-H and Si-H bonds, forming adducts of the general form
H(E)(Ph)P-B(C6F5)3 (E=H, SiR3) through P-P bond cleavage. This methodology could
potentially be exploited in the synthesis of chiral phosphines and in the activation of white
phosphorus.
Phosphinoborane monomers of the general form R2PB(C6F5)2 can be synthesized,
provided the substituents are bulky enough to prevent dimerization. The monomers show very
short P-B bonds, yet are found to react with H2, Lewis acids and Lewis bases.
137
Bulky pyridines and other nitrogen-containing heterocycles can act as partners for
B(C6F5)3 in FLPs. For the 2,6-lutidine/B(C6F5)3 pair, both adduct and free borane and pyridine
were observable in solution. This represents the border between classical Lewis acid-base
adduct and FLP chemistry. This pair was found to activate H2 and ring-open THF. Other bulky
alkyl- or aryl-substituted pyridine-borane FLPs were found to activate H2 in a fashion analogous
to the previously reported phosphine-borane systems. Several bulky quinolines were found to
undergo B(C6F5)3-catalyzed hydrogenations under mild conditions. The reaction of amine-
substituted pyridines with boranes was found to provide a facile synthesis for linked pyridine-
borane FLPs.
The small molecule activation pathways presented here, in particular the catalytic
hydrogenation reactions, could find utility in various academic and industrial applications. Rh
catalyzed P-P bond cleavage opens a new route to phosphine synthesis. This work serves to
expand the FLP concept as a whole, exploring novel metal-free small molecule activations, and
serves to demonstrate the potential of more environmentally benign main-group systems to
perform chemistry previously confined to transition metals.
138
Appendix A: Frustrated Lewis Pairs Derived From P(OR)nR3-n and B(C6F5)3
A.1: Introduction
While frustrated Lewis pairs derived from trialkyl and triaryl phosphines in combination
with B(C6F5)3 have been examined extensively,3 a detailed study of the cone angles and
basicities needed for hydrogen activation has not been undertaken. The phosphonium borates
derived from these reactions release H2 only under prolonged heating in the presence of a smaller
base.213
The use of less basic phosphorus centres, such as phosphites (which are less basic due to
the adjacent highly electronegative oxygen atoms), will allow testing of the limits of basicity and
cone angles required for the FLP hydrogen activation reaction. A better understanding of these
limits would help target bases which are basic enough to activate H2 with B(C6F5)3, but not so
basic that the reaction is irreversible. This knowledge will be helpful for the targeting of
potential hydrogenation catalysts.
139
A.2: Experimental
A.2.1: General Considerations
All preparations were done under an atmosphere of dry, O2-free N2 employing both Schlenk line
techniques and an Innovative Technologies or Vacuum Atmospheres inert atmosphere glove box.
Solvents (pentane, hexanes, toluene, and methylene chloride) were purified employing a Grubbs’
type column systems manufactured by Innovative Technology and stored over molecular sieves
(4 Å). Molecular sieves (4 Å) were purchased from Aldrich Chemical Company and dried at 140
ºC under vacuum for 24 hours prior to use. Deuterated solvents were dried over
Na/benzophenone (C6D6, C7D8) or CaH2 (CD2Cl2, CDCl3). All common organic reagents were
purified by conventional methods unless otherwise noted. 1H,
13C,
11B,
19F and
31P nuclear
magnetic resonance (NMR) spectroscopy spectra were recorded on a Bruker Avance-400
spectrometer at 300K unless otherwise noted. 1H and
13C NMR spectra are referenced to SiMe4
using the residual solvent peak impurity of the given solvent. 31
P NMR experiments were
referenced to 85% H3PO4, while 19
F and 11
B NMR experiments were referenced to 85% BF3-
Et2O in CDCl3. Chemical shifts are reported in ppm and coupling constants in Hz as absolute
values. Combustion analyses were performed in house employing a Perkin Elmer CHN
Analyzer. B(C6F5)3 was generously donated by NOVA Chemicals Corporation. tBu2PCl,
P(OCH3)3 and P(OPh)3 were purchased from Aldrich Chemicals, P(O-2,4-tBu2C6H3)3 was
purchased from Strem Chemicals and used as received. P(O-2,6-Me2C6H3)3 was prepared as
previously reported. 249
A.2.2: Lewis Acid-Base Adducts of Phosphites with B(C6F5)3
(RO)3PB(C6F5)3 (R = Me, A-1; R = Ph, A-2) - These compounds were prepared in a similar
fashion and thus only one preparation is reported. A clear solution of B(C6F5)3 (0.100 g, 0.19
mmol) and P(OCH3)3 (0.024 g, 0.19 mmol) in CH2Cl2 (2 mL) was stirred for one hour at room
temperature. The solvent was removed in vacuo. To the remnants was added pentane (2 mL)
resulting in a clear supernatant and a white solid layer. The supernatant was removed by pipette
and the solid was dried in vacuo.
A-1 - Yield: 96 mg (77 %). Anal. Calcd. for C21H9F15BPO3: C, 39.65; H, 1.43. Found: C, 39.71;
H, 1.24. X-Ray quality crystals were grown by slow evaporation from a CH2Cl2 solution. 1H
140
NMR (CDCl3) δ: 3.80 (d, 3JP-H=10 Hz, 9H, CH3).
19F NMR (CDCl3): -131.1 (d,
3JF-F=22 Hz,
6F, o-C6F5), -156.6 (t, 3JF-F=20 Hz, 3F, p-C6F5), -164.3 (td,
3JF-F=20 Hz,
4JF-F=6 Hz, 6F, m-C6F5).
31P NMR (CDCl3) δ: 75.5 (br m).
11B NMR (CDCl3) δ: -15.5 (d,
1JP-B=141 Hz).
13C{
1H} NMR
(CDCl3) partial δ: 55.7 (d, JP-C=12 Hz).
A-2 - Yield: 75 mg (93 %). Anal. Calcd. for C36H15F15BPO3: C, 52.58; H, 1.84. Found: C, 52.08;
H, 2.07. 1H NMR (CD2Cl2) δ: 7.23 (t,
3JH-H=7 Hz, 6H, o-C6H5), 7.20 (t,
3JH-H=7 Hz, 3H, p-
C6H5), 6.90 (dd, 3JH-H=8 Hz, J=1 Hz, 6H, m-C6H6).
19F NMR (CD2Cl2) δ: -129.8 (s, 6F, o-C6F5),
-156.2 (s, 3F, p-C6F5), -164.3 (s, 6F, m-C6F5). 31
P NMR (CD2Cl2) δ: 62.0 (br s). 11
B NMR
(CD2Cl2) δ: -8.9 (br s). 13
C{1H} NMR (CD2Cl2) partial δ: 150.9 (d, JP-C=14 Hz), 148.8 (d,
1JF-
C=248 Hz, o-C6F5), 140.7 (d, 1JF-C=256 Hz, p-C6F5), 137.3 (d,
1JF-C=252 Hz, m-C6F5), 130.2 (d,
JP-C=1 Hz), 126.2 (d, JP-C=1 Hz), 120.3 (d, JP-C=4 Hz).
A.2.3: Synthesis of Phosphinites tBu2POR
tBu2POR (R =
tBu, A-3; Ph, A-4; 2,6-Me2C6H3, A-5) - These compounds were prepared by
reaction of tBu2PCl with the appropriate potassium alkoxide or aryloxide (generated in situ by
reaction of the corresponding alcohol with KH) over 4 h in THF, followed by removal of the
solvent, extraction into pentane and filtration through celite (similar to the method previously
described for A-4).250
Careful removal of the solvent in vacuo left the products as colourless
oils. If excess alcohol remained, the product was redissolved in pentane and run through a plug
of alumina.
A-3 - Yield: 76 %. 1H NMR (C6D6) δ: 1.14 (d,
3JP-H=11 Hz, 18H,
tBu), 1.26 (d,
4JP-H=1 Hz,
OtBu, 9H).
31P{
1H} NMR (C6D6) δ: 136.7 (s).
13C{
1H} NMR (C6D6) δ: 28.7 (d, JP-C=16 Hz, P-
C(CH3)3), 31.4 (d, JP-C=7 Hz, OC(CH3)3), 34.6 (d, JP-C = 27 Hz, PCMe3), 75.5 (d, JP-C=11 Hz,
OCMe3).
A-4 - Yield: 83 %. 1H NMR (C6D6) δ: 1.08 (d,
3JP-H=12 Hz, 18H,
tBu), 6.78 (t,
3JH-H=7 Hz, 1H,
p-C6H5), 7.07 (t, 3JH-H=8 Hz, 2H, m-C6H5), 7.25 (dm,
3JH-H=8 Hz, 2H, o-C6H5);
31P{
1H} NMR
(C6D6) δ: 153.5 (s); 13
C{1H} NMR (C6D6) δ: 27.4 (d, JP-C=16 Hz), 35.6 (d, JP-C=26 Hz), 118.6 (d,
JP-C=11 Hz), 121.6 (d, JP-C=1 Hz), 129.6 (d, JP-C=1 Hz), 160.3 (d, JP-C=10 Hz).
141
A-5 - Yield: 78 %. 1H NMR (C6D6) δ: 1.08 (d,
3JP-H=11 Hz, 18 H,
tBu), 2.29 (s, 3H, CH3), 2.59
(s, 3H, CH3), 6.76 (t, 3JH-H=7 Hz, 1H, p-C6H3), 6.88 (d,
3JH-H=7 Hz, 2H, m-C6H3).
31P NMR
(C6D6) δ: 162.3 (s). 13
C{1H} NMR (C6D6) δ: 27.5 (d, JP-C=16 Hz), 36.4 (d, JP-C = 33 Hz), 122.0,
154.9 (d, JP-C = 2 Hz).
A.2.4: Generation of a Phosphine-Oxide Adduct of B(C6F5)3
tBu2(H)POB(C6F5)3 (A-6) - A clear solution of B(C6F5)3 (0.050 g, 0.10 mmol) and A-3 (0.030 g,
0.10 mmol) in CH2Cl2 (2 mL) was prepared. The solvent was removed in vacuo. To the remnants
was added pentane (2 mL) resulting in a clear supernatant and a white solid layer. The
supernatant was removed by pipette and the solid was dried in vacuo. Yield: 55 mg (82%).
Anal. Calcd. for C26H19BF15OP (%) C: 46.32, H: 2.84; found: C: 46.23, H: 2.97. 1H NMR
(C6D6) δ: 0.57 (d, 3JP-H=16 Hz, 18H, PC(CH3)3), 5.33 (d,
1JP-H=452 Hz, 1H, PH).
19F NMR
(C6D6) δ: -132.3 (d, 3JF-F=23 Hz, 6F, o-C6F5), -157.2 (t,
3JF-F=21 Hz, 3F, p-C6F5), -163.8 (tm,
3JF-
F = 23 Hz, 6F, m-C6F5). 31
P{1H} NMR (C6D6) δ: 77.0 (s).
11B{
1H} NMR (C6D6) δ: -0.30 (br s).
13C{
1H} NMR (C6D6) partial: 24.6 (br s, C(CH3)3), 33.7 (d, JP-C=57 Hz, C(CH3)3), 137.5 (dm, JC-
F=256 Hz, CF), 14 8.2 (dm, JC-F=242 Hz, CF).
A.2.5: Generation of Phosphonium Borate Ion Pairs by H2 Activation
[tBu2PH(OAr)][HB(C6F5)3] (Ar=Ph, A-7; 2,6-Me2C6H3, A-8) – Compounds A-7 and A-8 were
prepared in a similar fashion, thus only one preparation is described. A-4 (47 mg, 0.20 mmol)
was added to a solution of B(C6F5)3 (100 mg, 0.20 mmol) in toluene (5 mL) in a sealed Teflon-
capped reaction vessel. The solution was subjected to three freeze-pump-thaw cycles and
subsequently exposed to an atmosphere of H2 at 77 K. On warming to room temperature this is
equivalent to approximately 4 atm H2. The solution was allowed to stir overnight at room
temperature. The solution was removed and the bomb was washed with CH2Cl2 (2 x 2 mL).
Volatiles were removed in vacuo and pentane was added to the resulting thick oil. The product
crystallized over 3 days, producing X-Ray quality crystals.
A-7 - Yield: 142 mg (97%). Anal. Calcd. for C32H25BF15PO (%) C: 51.09, H: 3.35; found: C:
51.21; H: 3.59. 1H NMR (CD2Cl2) δ: 1.43 (d,
3JP-H=19 Hz, 18 H, C(CH3)3), 3.55 (q,
1JB-H=93 Hz,
1H, BH), 6.98 (d, 1JP-H=474 Hz, 1H, PH), 7.08 (d, J=8 Hz, 2H, CH), 7.26 (br s, 1H, CH), 7.37 (t,
J=8 Hz, 2H, CH). 19
F NMR (CD2Cl2) δ: -134.2 (d, 3JF-F=24 Hz, o-C6F5), -164.8 (t,
3JF-F=23 Hz,
142
p-C6F5), -167.8 (tm, 3JF-F=24 Hz, m-C6F5).
31P NMR (CD2Cl2) δ: 98.5 (dm,
1JP-H=474 Hz,
3JP-
H=19 Hz); 11
B NMR (CD2Cl2) δ: -25.6 (d, 1JB-H=93 Hz);
13C{
1H} NMR (CD2Cl2) partial δ: 25.4
(d, J=2 Hz), 36.4 (d, J=42 Hz), 115.4, 117.6 (d, J=6 Hz), 131.5.
A-8 - Yield: 147 mg (96%). Anal. Calcd. for C34H29BF15PO (%) C: 52.33, H: 3.75; found: C:
52.33; H: 3.98. 1H NMR (CD2Cl2) δ: 1.44 (d,
3JP-H=20 Hz, 18H, C(CH3)3), 2.28 (s, 6H, Ar-
CH3), 3.51 (q, 1JB-H=93 Hz, BH), 7.02 (d,
1JP-H=483 Hz, 1H, PH), 7.02-7.13 (m, 3H, CH);
19F
NMR (CD2Cl2) δ: -134.3 (d, 3JF-F=23 Hz, o-C6F5), -165.0 (t,
3JF-F=20 Hz, p-C6F5), -167.9 (tm,
3JF-F=20 Hz, m-C6F5).
31P NMR (CD2Cl2) δ: 97.5 (d,
1JP-H=483 Hz,
3JP-H=20 Hz);
11B NMR
(CD2Cl2) δ: -25.3 (d, 1JB-H=93 Hz);
13C{
1H} NMR (CD2Cl2) partial δ: 8.3 (CP), 16.2 (m, CH3),
137.4 (dm, 1JC-F=248 Hz, CF), 140.4 (dm,
1JC-F =209 Hz, CF), 147.3 (dm,
1JC-F =227 Hz, CF).
A.2.6: X-Ray Data Collection, Reduction, Solution and Refinement
Single crystals were mounted in thin-walled capillaries either under an atmosphere of dry N2 in a
glove box and flame sealed or coated in paratone-N oil. The data were collected using the
SMART software package on a Bruker SMART Apex II System CCD diffractometer using a
graphite monochromator with Mo Κα radiation (λ = 0.71073 Å). Data collection strategies were
determined using Bruker Apex software and optimized to provide >99.5% complete data to a 2θ
value of at least 55°. 10 second exposure times were used unless otherwise noted. Data
reductions were performed using the SAINT software package and absorption corrections were
applied using SADABS. The structures were solved by direct methods using XS and refined by
full-matrix least-squares on F2 using XL as implemented in the SHELXTL suite of programs. All
non-H atoms were refined anisotropically. Carbon-bound hydrogen atoms were placed in
calculated positions using an appropriate riding model and coupled isotropic temperature factors.
Phosphorus-bound hydrogen atoms were located in the electron difference map and their
positions refined isotropically. Single crystal X-ray structures were obtained for A-1, A-6 and A-
7. Selected crystallographic data are included in Table A.1. Diagrams and selected bond lengths
and angles are provided in Figures A.2, A.4 and A.9.
143
Table A.1: Selected crystallographic data for compounds A-1, A-6 and A-7
Crystal A-1 A-6 A-7
Formula C21H9BF15O3P C26H19BF15OP C34H29BF15OP
Formula weight 636.06 674.19 780.35
Crystal system Triclinic Monoclinic Orthorhombic
Space group P-1 Pc Pbca
a(Å) 10.1164(4) 10.6660(5) 18.3248(13)
b(Å) 10.9283(4) 9.8862(4) 18.6811(13)
c(Å) 10.9799(4) 16.6408(6) 19.3253(15)
(o) 70.753(2) 90.00 90.00
( o) 87.539(2) 129.863(2) 90.00
( o) 81.346(2) 90.00 90.00
V (Å3) 1132.96(8) 1346.88(10) 6615.6(8)
Z 2 2 8
d(calc) g cm-1
1.865 1.662 1.567
Abs coeff, , cm-1
0.271 0.347 0.197
Data collected 9906 5947 7578
Data Fo2>3(Fo
2) 8039 4386 5442
Variables 370 407 477
Ra 0.0362 0.0325 0.0393
Rwb 0.1120 0.0537 0.1045
Goodness of Fit 1.020 0.791 1.010
These data were collected at 150 K with Mo Kα radiation (λ = 0.71069 Å).
aR=Σ(Fo-Fc)/ΣFo
bRw=(Σ[w(Fo
2-Fc
2 )
2] /Σ[w(Fo)
2])
½
144
A.3: Results and Discussion
A.3.1: Reactions of Phosphites with B(C6F5)3
Stoichiometric reactions of several phosphites with B(C6F5)3 were initially investigated.
The relatively smaller phosphites, such as trimethyl- and triphenylphosphite form strong adducts
with B(C6F5)3 (A-1 and A-2, respectively).
Figure A.1: Formation of phosphite-borane adducts A-1 and A-2
A-1 was also characterized by single-crystal X-ray diffraction (Figure A.2). Of particular
interest in this case is the extremely short P-B bond length of 2.0209(11) Å. This bond length is
even shorter than that of the adduct formed between Me3P and B(C6F5)3 (2.061(4)Å). The
shorter bond length of the phosphite compared to the phosphine is likely caused by the reduced
cone angle of the phosphite as the phosphine is considerably more basic (Table A.2).
Figure A.2: POV-Ray depiction of A-1. Carbon: black, Boron: yellow-green, Fluorine: deep
pink, Oxygen: red, Phosphorus: orange. Hydrogen atoms are omitted for clarity. Selected
metrical parameters (Distances: Å, Angles: °): P1-B1 2.0209(11).
145
Table A.2: Comparison of Me3P and (MeO)3P and their B(C6F5)3 adducts
Base pKa251
cone angle ()252
dP-B (Å)
(MeO)3P 2.60 107 2.0209(11)
Me3P 8.65 118 2.061(4)
29
The relatively larger phosphites, tris(2,6-dimethylphenyl)phosphite and tris(2,4-di-tert-
butylphenyl)phosphite show no reaction with B(C6F5)3 by multinuclear NMR spectroscopy.
Based on this observation, solutions of these phosphites with B(C6F5)3 are FLPs. Unfortunately
these FLPs showed no reactivity upon addition of H2 (Figure A.3), THF or olefins to solutions of
these acid/base pairs. These results would seem to suggest that, while these phosphites have
sufficiently large cone angles to prevent adduct formation, their low basicity prevents them from
acting as bases or nucleophiles in any of the present FLP small molecule activations (see Table
A.3 for a comparison of phosphines and phosphites).
Figure A.3: Phosphite-based FLPs and attempted hydrogen activation
146
Table A.3: Cone angles and basicities of phosphorus bases and reactivity with B(C6F5)3
Base Cone Angle (°)252
pKa (conjugate
acid)253
Product of reaction with
B(C6F5)3 at 25°C213
(MeO)3P 107 2.6 Adduct (A-1)
Me3P 118 8.6 Adduct
(PhO)3P 128 -2.0 Adduct (A-2)
Bu3P 132 8.7 Adduct
Ph3P 145 2.7 Adduct
iPr3P 160 9.3 Para-NAS
Cy3P 170 9.7 Para-NAS
(tBuO)3P 172 4.5 Multiple Products
[2,6-(CH3)2C6H3-O]3P 182 -0.4 No reaction
tBu3P 182 11.4 No reaction
[2,4-tBu2C6H3-O]3P
190
254 N/A No reaction
[o-(CH3)C6H4)3P 194 3.1 No reaction
Mes3P 212 N/A No reaction
The reactivity of B(C6F5)3 with phosphites found in this study fit the trend of the
previously obtained data for tertiary phosphines. Bases with cone angles of 182° or larger do not
react with B(C6F5)3 under ambient conditions, while the weakest base found to activate H2 in a
FLP with B(C6F5)3 remains tri(ortho-tolyl)phosphine with a pKa of 3.1. These results stress the
147
importance of not only utilizing a large base (i.e. poor nucleophile) to prevent adduct formation,
but also that there is a threshold of base strength required for activation of H2 with B(C6F5)3.
A.3.2: Reactions of RnP(OtBu)3-n with B(C6F5)3
As [o-(CH3)C6H4]3P (cone angle 194°, pKa=3.1) is capable of hydrogen activation in
tandem with B(C6F5)3,49
electronically, tri-tert-butylphosphite (cone angle 172°, pKa=4.5)
should have been capable of similar reactivity. However, based on the similar cone angle to
tricyclohexylphosphine (cone angle 170°) para-nucleophilic aromatic substitution on a C6F5 ring
of the borane is also a possibility.38
Unfortunately, the reaction of (tBu3O)3P with B(C6F5)3
produced several products by multinuclear NMR spectroscopy, none of which could be
conclusively identified.
To probe the possible reactions caused by interaction of the OtBu substituent with
B(C6F5)3, the phosphinite tBuOP
tBu2 (A-3) was synthesized cleanly by reaction of
tBu2PCl with
KOtBu (KCl is removed upon workup). Interestingly, A-3 reacts with B(C6F5)3, and following
workup the peak in the 1H NMR spectrum corresponding to the oxygen-bound tert-butyl group
had disappeared. This observation, along with data from the 11
B, 19
F and 31
P NMR spectra,
suggested the formation of the adduct tBu2(H)P=O-B(C6F5)3 (A-9). This assignment was
confirmed by X-ray crystallography (Figure A.4). Monitoring the reaction by 1H NMR
spectroscopy revealed that the oxygen-bound tert-butyl group is lost as iso-butene upon reaction
with B(C6F5)3 (Figure A.5).
148
Figure A.4: POV-Ray depiction of A-6. Carbon: black, Hydrogen: white, Boron: yellow-green,
Fluorine: deep pink, Oxygen: red, Phosphorus: orange. Carbon-bound hydrogen atoms are
omitted for clarity. Selected metrical parameters (Distances: Å, Angles: °): P1-O1 1.5340(14);
O1-B1 1.524(2); P1-O1-B1 139.55(13).
Figure A.5: 1H NMR spectrum showing formation of iso-butene, along with adduct A-9
The proposed mechanism for this reaction, illustrated in Figure A.6, involves initial co-
ordination of the Lewis acid to the oxygen centre. This initial reaction makes the tert-butoxy
hydrogen atoms acidic, which makes them susceptible to abstraction by the nearby basic
phosphorus centre. The reaction is presumably driven by the formation of strong P=O and C=C
bonds (~500 kJ/mol, and ~733 kJ/mol, respectively).255
PH 1JP-H=452 Hz
CH2=C(CH3)2
CH2=C(CH3)2 PC(CH3)3 3JP-H=16 Hz
149
Figure A.6: Proposed mechanism for the formation of A-9 and iso-butene from reaction of A-3
with B(C6F5)3
A.3.3: FLP Reactions of tBu2POAr with B(C6F5)3
In order to avoid the potential loss of alkene from these species, the phosphinites
tBu2POPh and
tBu2PO(2,6-Me2C6H3), A-4 and A-5 respectively, were synthesized. Loss of the
aryl group in this fashion would have to be accompanied by loss of a benzyne derivative, which
is highly unlikely. These bases did not form adducts with B(C6F5)3 at room temperature. Thus,
FLP reactivity was examined. Both phosphinites proved to be capable of H2 activation in
tandem with B(C6F5)3, quantitatively forming ion pairs A-7 and A-8, respectively, by NMR
spectroscopy (Figure A.7, A.8). The structure of A-8 was confirmed through X-ray
crystallography (Figure A.9).
Figure A.7: FLP reactivity of phosphinites A-5 and A-6 with B(C6F5)3
150
Figure A.8: Multinuclear NMR spectra for A-7 in CD2Cl2. A: 1H, B:
11B, C:
19F and D:
31P
Figure A.9: POV-Ray depiction of A-8. Carbon: black, Hydrogen: white, Boron: yellow-green,
Fluorine: deep pink, Oxygen: red, Phosphorus: orange. Hydrogen atoms are omitted for clarity.
Selected metrical parameters (Distances: Å, Angles: °): B1-C1 1.638(3), B1-C7 1.645(3), P1-O1
1.5757(13), P1-C20 1.8290(19), P1-C24 1.827(2), P-H - - - H-B 2.348, C1-B1-C7 114.07(15),
C1-B1-C13 115.27(15), O1-P1-C20 113.75(8), O1-P1-C24 120.32 (9), P1-O1-C28 134.12(12).
PC(CH3)3 3JP-H=19 Hz
C6H5
PH 1JP-H=474 Hz
BH 1JB-H=93 Hz
1JB-H=93 Hz
A B
C D
o-C6F5
p-C6F5
m-C6F5
PH 1JP-H=474 Hz
151
A.4: Conclusions
In summary, the use of phosphites as donors in FLPs with B(C6F5)3 proved to be
unsuccessful for H2 activation, however, new reaction pathways were discovered. The bulky
phosphites such as P(O-2,6-Me2C6H3)3 and P(O-2,4-tBu2C6H3)3 proved to not be basic enough to
activate H2 in combination with B(C6F5)3. This observation is consistent with the computational
study by Papai and co-workers that suggests that a cumulative acid/base strength is required for a
FLP to activate H2.50
The tert-butoxy substituent of tBuOP
tBu2 was found to react with B(C6F5)3
to form iso-butene and the corresponding phosphine oxide. The aryl-oxy functional group could
be incorporated into the base component of FLP’s capable of H2 activation through the use of
tBu2P(OAr) as the Lewis base.
152
References
(1) Greenberg, S.; Stephan, D. W. Chem. Soc. Rev. 2008, 37, 1482.
(2) Welch, G. C.; Juan, R. R. S.; Masuda, J. D.; Stephan, D. W. Science 2006, 314, 1124.
(3) Stephan, D. W.; Erker, G. Angew. Chem. Int. Ed. 2010, 49, 46.
(4) Stephan, D. W. Org Biomol Chem 2008, 6, 1535.
(5) Crabtree, R. H. The organometallic chemistry of the transition metals; 4th ed.; Wiley-
Interscience: Hoboken, N.J., 2005.
(6) Meakin, P.; Tolman, C. A.; Jesson, J. P. J. Am. Chem. Soc. 1972, 94, 3240.
(7) Dias, E. L.; Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc. 1997, 119, 3887.
(8) Gates, D. P. Top. Curr. Chem. 2005, 250, 107.
(9) Baumgartner, T.; Reau, R. Chem. Rev. 2006, 106, 4681.
(10) Clark, T. J.; Lee, K.; Manners, I. Chem.-Eur. J. 2006, 12, 8634.
(11) Allcock, H. R. Chemistry and applications of polyphosphazenes; Wiley-Interscience:
Hoboken, N.J., 2003.
(12) Fermin, M. C.; Stephan, D. W. J. Am. Chem. Soc. 1995, 117, 12645.
(13) Han, L. B.; Tilley, T. D. J. Am. Chem. Soc. 2006, 128, 13698.
(14) Bohm, V. P. W.; Brookhart, M. Angew. Chem. Int. Ed. 2001, 40, 4694.
(15) Thompson, R.; Royal Society of Chemistry (Great Britain). Inorganic Chemicals Group.;
Royal Society of Chemistry (Great Britain). Fine Chemicals and Medicinals Group.; Royal
Society of Chemistry (Great Britain). Industrial Division. North West Region. Speciality
inorganic chemicals : the proceedings of a symposium organised by the Inorganic Chemicals
Group, the Fine Chemicals and Medicinals Group, and the N.W. Region of the Industrial
Division of the Royal Society of Chemistry, in association with the Dalton Division, University of
Salford, September 10th-12th 1980; The Royal Society of Chemistry: London, 1981.
(16) Cossairt, B. M.; Cummins, C. C. Angew. Chem. Int. Ed. 2008, 47, 8863.
(17) Piro, N. A.; Cummins, C. C. J. Am. Chem. Soc. 2008, 130, 9524.
(18) Fox, A. R.; Clough, C. R.; Piro, N. A.; Cummins, C. C. Angew. Chem. Int. Ed. 2007, 46,
973.
(19) Figueroa, J. S.; Cummins, C. C. Dalton Trans. 2006, 2161.
153
(20) Piro, N. A.; Figueroa, J. S.; McKellar, J. T.; Cummins, C. C. Science 2006, 313, 1276.
(21) Stephens, F. H.; Johnson, M. J. A.; Cummins, C. C.; Kryatov, O. P.; Kryatov, S. V.;
Rybak-Akimova, E. V.; McDonough, J. E.; Hoff, C. D. J. Am. Chem. Soc. 2005, 127, 15191.
(22) Back, O.; Kuchenbeiser, G.; Donnadieu, B.; Bertrand, G. Angew. Chem. Int. Ed. 2009.
(23) Masuda, J. D.; Schoeller, W. W.; Donnadieu, B.; Bertrand, G. J. Am. Chem. Soc. 2007,
129, 14180.
(24) Cummins, C. C. Angew. Chem. Int. Ed. 2006, 45, 862.
(25) Masuda, J. D.; Schoeller, W. W.; Donnadieu, B.; Bertrand, G. J. Am. Chem. Soc. 2007,
129, 14180.
(26) Masuda, J. D.; Schoeller, W. W.; Donnadieu, B.; Bertrand, G. Angew. Chem. Int. Ed.
2007, 46, 7052.
(27) Baudler, M.; Glinka, K. Chem. Rev. 1993, 93, 1623.
(28) Lewis, G. N. Valence and the Structure of Atoms and Molecules; Chemical Catalogue
Company, Inc.: New York, 1923.
(29) Chase, P. A.; Parvez, M.; Piers, W. E. Acta Crystallogr. E 2006, 62, O5181.
(30) Piers, W. E. Adv. Organomet. Chem. 2005, 52, 1.
(31) Focante, F.; Mercandelli, P.; Sironi, A.; Resconi, L. Coord. Chem. Rev. 2006, 250, 170.
(32) Brown, H. C.; Schlesinger, H. I.; Cardon, S. Z. J. Am. Chem. Soc. 1942, 64, 325.
(33) Brown, H. C. J. Chem. Soc. 1956, 1248.
(34) Welch, G. C.; Prieto, R.; Dureen, M. A.; Lough, A. J.; Labeodan, O. A.; Holtrichter-
Rossmann, T.; Stephan, D. W. Dalton Trans. 2009, 1559.
(35) Cabrera, L.; Welch, G. C.; Masuda, J. D.; Wei, P.; Stephan, D. W. Inorg. Chim. Acta
2006, 359 3066.
(36) Welch, G. C.; Holtrichter-Roessmann, T.; Stephan, D. W. Inorg. Chem. 2008, 47, 1904.
(37) Doring, S.; Erker, G.; Frohlich, R.; Meyer, O.; Bergander, K. Organometallics 1998, 17,
2183.
(38) Welch, G. C.; Cabrera, L.; Chase, P. A.; Hollink, E.; Masuda, J. D.; Wei, P. R.; Stephan,
D. W. Dalton Trans. 2007, 3407.
(39) Chivers, T.; Schatte, G. Eur. J. Inorg. Chem. 2003, 3314.
(40) Welch, G. C.; Masuda, J. D.; Stephan, D. W. Inorg. Chem. 2006, 45, 478.
154
(41) Kubas, G. J. Science 2006, 314, 1096.
(42) de Vries, J. G.; Elsevier, C. J. The Handbook of Homogeneous Hydrogenation; Wiley-
VCH: Weinheim, 2007.
(43) Welch, G. C.; Stephan, D. W. J. Am. Chem. Soc. 2007, 129, 1880.
(44) Chase, P. A.; Stephan, D. W. Angew. Chem. Int. Ed. 2008, 47, 7433.
(45) Holschumacher, D.; Bannenberg, T.; Hrib Cristian, G.; Jones Peter, G.; Tamm, M.
Angew. Chem. Int. Ed. 2008, 47, 7428.
(46) Chase, P. A.; Jurca, T.; Stephan, D. W. Chem. Commun. 2008, 1701.
(47) Sumerin, V.; Schulz, F.; Nieger, M.; Leskela, M.; Repo, T.; Rieger, B. Angew. Chem. Int.
Ed. 2008, 47, 6001.
(48) Neu, R. C.; Ouyang, E. Y.; Geier, S. J.; Zhao, X.; Ramos, A.; Stephan, D. W. Dalton
Trans. 2010, 39.
(49) Ullrich, M.; Lough, A. J.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 52.
(50) Rokob, T. A.; Hamza, A.; Papai, I. J. Am. Chem. Soc. 2009, 131, 10701.
(51) Spies, P.; Erker, G.; Kehr, G.; Bergander, K.; Froehlich, R.; Grimme, S.; Stephan, D. W.
Chem. Commun. 2007, 5072.
(52) Sumerin, V.; Schulz, F.; Atsumi, M.; Wang, C.; Nieger, M.; Leskela, M.; Repo, T.;
Pyykko, P.; Rieger, B. J. Am. Chem. Soc. 2008, 130, 14117.
(53) Otten, E.; Stephan, D. W. Unpublished Results.
(54) Tague, T. J.; Andrews, L. J. Am. Chem. Soc. 1994, 116, 4970.
(55) Jursic, B. S. J. Mol. Struct. 1999, 492, 97.
(56) Watts, J. D.; Bartlett, R. J. J. Am. Chem. Soc. 1995, 117, 825.
(57) Schreiner, P. R.; Schaefer, H. F.; Schleyer, P. V. J. Chem. Phys. 1994, 101, 7625.
(58) Moroz, A.; Sweany, R. L.; Whittenburg, S. L. J. Phys. Chem. 1990, 94, 1352.
(59) Moroz, A.; Sweany, R. L. Inorg. Chem. 1992, 31, 5236.
(60) Rokob, T. A.; Hamza, A.; Stirling, A.; Soos, T.; Papai, I. Angew. Chem. Int. Ed. 2008,
47, 2435.
(61) Hamza, A.; Stirling, A.; Rokob, T. A.; Papai, I. Int. J. Quantum Chem 2009, 109, 2416.
(62) Rokob, T. A.; Hamza, A.; Stirling, A.; Papai, I. J. Am. Chem. Soc. 2009, 131, 2029.
155
(63) Grimme, S.; Kruse, H.; Goerigk, L.; Erker, G. Angew. Chem. Int. Ed. 2010, 49, 1402.
(64) Rozas, I.; Alkorta, I.; Elguero, J. Chem. Phys. Lett. 1997, 275, 423.
(65) Hugas, D.; Simon, S.; Duran, M.; Guerra, C. F.; Bickelhaupt, F. M. Chem-Eur J 2009,
15, 5814.
(66) Meir, R.; Chen, H.; Lai, W. Z.; Shaik, S. Chemphyschem 2010, 11, 301.
(67) McCahill, J. S. J.; Welch, G. C.; Stephan, D. W. Angew. Chem. Int. Ed. 2007, 46, 4968.
(68) Dureen, M. A.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 8396.
(69) Momming, C. M.; Fromel, S.; Kehr, G.; Frohlich, R.; Grimme, S.; Erker, G. J. Am.
Chem. Soc. 2009, 131, 12280.
(70) Momming, C. M.; Otten, E.; Kehr, G.; Frohlich, R.; Grimme, S.; Stephan, D. W.; Erker,
G. Angew. Chem. Int. Ed. 2009, 48, 6643.
(71) Otten, E.; Neu, R. C.; Stephan, D. W. J. Am. Chem. Soc. 2009, 131, 9918.
(72) Menard, G.; Stephan, D. W. J. Am. Chem. Soc. 2010, 132, 1796.
(73) Chase, P. A.; Welch, G. C.; Jurca, T.; Stephan, D. W. Angew. Chem. Int. Ed. 2007, 46,
8050.
(74) Spies, P.; Schwendemann, S.; Lange, S.; Kehr, G.; Fröhlich, R.; Erker, G. Angew. Chem.
Int. Ed. 2008, 47, 7543.
(75) Axenov, K. V.; Kehr, G.; Fröhlich, R.; Erker, G. J. Am. Chem. Soc. 2009, 131, 3454.
(76) Wang, H. D.; Frohlich, R.; Kehr, G.; Erker, G. Chem. Commun. 2008, 5966.
(77) Fehlner, T. P. Inorganometallics; Plenum Press: New York, 1992.
(78) Chandrasekhar, V. Inorganic and organometallic polymers; Springer: Berlin ; New York,
NY, 2005.
(79) Mark, J. E.; Allcock, H. R.; West, R. Inorganic polymers; 2nd ed.; Oxford University
Press: New York, 2005.
(80) Dorn, H.; Singh, R. A.; Massey, J. A.; Lough, A. J.; Manners, I. Angew. Chem. Int. Ed.
1999, 38, 3321.
(81) Dorn, H.; Jaska, C. A.; Singh, R. A.; Lough, A. J.; Manners, I. Chem. Commun. 2000,
1041.
(82) Dorn, H.; Singh, R. A.; Massey, J. A.; Nelson, J. M.; Jaska, C. A.; Lough, A. J.; Manners,
I. J. Am. Chem. Soc. 2000, 122, 6669.
156
(83) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. Chem. Commun. 2001, 962.
(84) Jaska, C. A.; Lough, A. J.; Manners, I. Inorg Chem 2004, 43, 1090.
(85) Jaska, C. A.; Bartole-Scott, A.; Manners, I. Dalton Trans. 2003, 4015.
(86) Jaska, C. A.; Dorn, H.; Lough, A. J.; Manners, I. Chem-Eur J 2003, 9, 271.
(87) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 1334.
(88) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. Phosphorus Sulfur and Silicon and
the Related Elements 2004, 179, 733.
(89) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 2698.
(90) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 9776.
(91) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 2698.
(92) Bartole-Scott, A.; Jaska, C. A.; Manners, I. Pure Appl. Chem. 2005, 77, 1991.
(93) Clark, T. J.; Rodezno, J. M.; Clendenning, S. B.; Aouba, S.; Brodersen, P. M.; Lough, A.
J.; Ruda, H. E.; Manners, I. Chem.-Eur. J. 2005, 11, 4526.
(94) Jaska, C. A.; Clark, T. J.; Clendenning, S. B.; Grozea, D.; Turak, A.; Lu, Z.-H.; Manners,
I. J. Am. Chem. Soc. 2005, 127, 5116.
(95) Jaska, C. A.; Clark, T. J.; Clendenning, S. B.; Grozea, D.; Turak, A.; Lu, Z. H.; Manners,
I. J. Am. Chem. Soc. 2005, 127, 5116.
(96) Clark, T. J.; Russell, C. A.; Manners, I. J. Am. Chem. Soc. 2006, 128, 9582.
(97) Clark, T. J.; Jaska, C. A.; Turak, A.; Lough, A. J.; Lu, Z. H.; Manners, I. Inorg Chem
2007, 46, 7394.
(98) Corcoran, E. W., Jr.; Sneddon, L. G. J. Am. Chem. Soc. 1985, 107, 7446.
(99) Corcoran, E. W., Jr.; Sneddon, L. G. J. Am. Chem. Soc. 1984, 106, 7793.
(100) Hoskin, A. J.; Stephan, D. W. Angew. Chem. Int. Ed. 2001, 40, 1865.
(101) Etkin, N.; Fermin, M. C.; Stephan, D. W. J. Am. Chem. Soc. 1997, 119, 2954.
(102) Roering, A. J.; MacMillan, S. N.; Tanski, J. M.; Waterman, R. Inorg. Chem. 2007, 46,
6855.
(103) Aitken, C.; Harrod, J. F.; Malek, A.; Samuel, E. J. Organomet. Chem. 1988, 349, 285.
(104) Choi, N.; Tanaka, M. J. Organomet. Chem. 1998, 564, 81.
157
(105) Aitken, C.; Harrod, J. F.; Samuel, E. J. Organomet. Chem. 1985, 279, C11.
(106) Aitken, C. T.; Harrod, J. F.; Samuel, E. J. Am. Chem. Soc. 1986, 108, 4059.
(107) Tilley, T. D. Acc. Chem. Res. 1993, 26, 22.
(108) Corey, J. Y.; Zhu, X. H.; Bedard, T. C.; Lange, L. D. Organometallics 1991, 10, 924.
(109) Rosenberg, L.; Davis, C. W.; Yao, J. J. Am. Chem. Soc. 2001, 123, 5120.
(110) Babcock, J. R.; Sita, L. R. J. Am. Chem. Soc. 1996, 118, 12481.
(111) Neale, N. R.; Tilley, T. D. J. Am. Chem. Soc. 2002, 124, 3802.
(112) Waterman, R.; Tilley, T. D. Angew. Chem. Int. Ed. 2006, 45, 2926.
(113) Emsley, J. The 13th element : the sordid tale of murder, fire, and phosphorus; John
Wiley & Sons, Inc.: Chichester ; New York, 2000.
(114) Masuda, J. D.; Stephan, D. W. Can. J. Chem. 2005, 83, 324.
(115) Garon, C. M.; McIsaac, D. I.; Vogels, C. M.; Decken, A.; Williams, I. D.; Kleeberg, C.;
Marder, T. B.; Westcott, S. A. Dalton Trans. 2009, 1624.
(116) Shu, R.; Hao, L.; Harrod, J. F.; Woo, H.-G.; Samuel, E. J. Am. Chem. Soc. 1998, 120,
12988.
(117) Antoniadis, A.; Kunze, U. Z. Anorg. Allg. Chem. 1979, 34, 116.
(118) Hayashi, M.; Matsuura, Y.; Watanabe, Y. Tetrahedron Lett. 2004, 45, 1409.
(119) Bourget-Merle, L.; Lappert, M. F.; Severn, J. R. Chem. Rev. 2002, 102, 3031.
(120) Foo, T.; Bergman, R. G. Organometallics 1992, 11, 1811.
(121) Rerek, M. E.; Basolo, F. J. Am. Chem. Soc. 1984, 106, 5908.
(122) Arif, A. M.; Jones, R. A.; Seeberger, M. H.; Whittlesey, B. R.; Wright, T. C. Inorg.
Chem. 1986, 25, 3943.
(123) Haines, R. J.; Steen, N. D. C.; English, R. B. J. Chem. Soc., Dalton Trans. 1983, 1607.
(124) Haines, R. J.; Steen, N. D. C. T.; English, R. B. J. Chem. Soc., Chem. Commun. 1981,
587.
(125) Hieber, W.; Kummer, R. Chem. Ber. 1967, 100, 148.
(126) Klingert, B.; Werner, H. J. Organomet. Chem. 1987, 333, 119.
(127) Klingert, B.; Werner, H. J. Organomet. Chem. 1983, 252, C47.
158
(128) Tejel, C.; Sommovigo, M.; Ciriano, M. A.; Lopez, J. A.; Lahoz, F. J.; Oro, L. A. Angew.
Chem. Int. Ed. 2000, 39, 2336.
(129) Werner, H.; Klingert, B.; Rheingold, A. L. Organometallics 1988, 7, 911.
(130) Yasufuku, K.; Yamazaki, H. Bull. Chem. Soc. Jpn. 1973, 46, 1502.
(131) Dashti-Mommertz, A.; Neumuller, B. Z. Anorg. Allg. Chem. 1999, 625, 954.
(132) Mundt, O.; Riffel, H.; Becker, G.; Simon, A. Z. Anorg. Allg. Chem. 1988, 43, 952.
(133) Richter, R.; Kaiser, J.; Sieler, J.; Hartung, H.; Peter, C. Acta Crystallogr., Sect. B 1977,
B33, 1887.
(134) Baxter, S. G.; Cowley, A. H.; Davis, R. E.; Riley, P. E. J. Am. Chem. Soc. 1981, 103,
1699.
(135) Werner, H.; Klingert, B. J. Organomet. Chem. 1981, 218, 395.
(136) Klingert, B.; Werner, H. Chem. Ber. 1983, 116, 1450.
(137) Zhang, S. Y.; Yemul, S.; Kagan, H. B.; Stern, R.; Commereuc, D.; Chauvin, Y.
Tetrahedron Lett. 1981, 22, 3955.
(138) Bitterwolf, T. E.; Spink, W. C.; Rausch, M. D. J. Organomet. Chem. 1989, 363, 189.
(139) Crumbliss, A. L.; Topping, R. J.; Szewczyk, J.; McPhail, A. T.; Quin, L. D. J. Chem.
Soc., Dalton Trans. 1986, 1895.
(140) Thompson, D. T.; (Imperial Chemical Industries Ltd.). Application: GB
GB, 1969, p 4 pp.
(141) Werner, H.; Klingert, B.; Zolk, R.; Thometzek, P. J. Organomet. Chem. 1984, 266, 97.
(142) Butts, M. D.; Bryan, J. C.; Luo, X. L.; Kubas, G. J. Inorg. Chem. 1997, 36, 3341.
(143) Bernskoetter, W. H.; Lobkovsky, E.; Chirik, P. J. Organometallics 2005, 24, 4367.
(144) Campian, M. V.; Harris, J. L.; Jasim, N.; Perutz, R. N.; Marder, T. B.; Whitwood, A. C.
Organometallics 2006, 25, 5093.
(145) Bernskoetter, W. H.; Lobkovsky, E.; Chirik, P. J. Chem. Commun. 2004, 764.
(146) Ang, H.-G.; Ang, S.-G.; Zhang, Q. J. Chem. Soc., Dalton Trans. 1996, 3843.
(147) Bai, G.; Wei, P.; Das, A. K.; Stephan, D. W. Dalton Trans. 2006, 1141.
(148) Quirt, A. R.; Martin, J. S. J Magn Reson 1971, 5, 318.
159
(149) Marat, K. Spinworks 3.1, University of Manitoba 2009.
(150) Kesanli, B.; Mattamana, S. P.; Danis, J.; Eichhorn, B. Inorg. Chim. Acta 2005, 358, 3145.
(151) Charles, S.; Danis, J. A.; Fettinger, J. C.; Eichhorn, B. W. Inorg. Chem. 1997, 36, 3772.
(152) Couret, C.; Escudie, J.; Saint-Roch, B.; Andriamizaka, J. D.; Satge, J. J. Organomet.
Chem. 1982, 224, 247.
(153) Dixon, D. A.; Arduengo, A. J.; Fukunaga, T. J. Am. Chem. Soc. 1986, 108, 2461.
(154) Rogers, J. R.; Wagner, T. P. S.; Marynick, D. S. Inorg. Chem. 1994, 33, 3104.
(155) Wicht, D. K.; Kovacik, I.; Glueck, D. S.; Liable-Sands, L. M.; Incarvito, C. D.;
Rheingold, A. L. Organometallics 1999, 18, 5141.
(156) Fryzuk, M. D.; Giesbrecht, G. R.; Rettig, S. J. Inorg. Chem. 1998, 37, 6928.
(157) Yao, W. B.; Eisenstein, O.; Crabtree, R. H. Inorg. Chim. Acta 1997, 254, 105.
(158) Burford, N.; Losier, P.; Mason, S.; Royan, B. W.; Spence, R. E. v. H.; Bakshi, P. K.;
Borecka, B.; Cameron, T. S.; Richardson, J. F.; Rogers, R. D. Phosphorus Sulfur and Silicon and
the Related Elements 1993, 76, 277.
(159) Burford, N.; Losier, P.; Bakshi, P. K.; Cameron, T. S. J. Chem. Soc., Dalton Trans. 1993,
201.
(160) Burford, N.; Losier, P.; Macdonald, C.; Kyrimis, V.; Bakshi, P. K.; Cameron, T. S. Inorg
Chem 1994, 33, 1434.
(161) Burford, N.; Cameron, T. S.; Ragogna, P. J.; Ocando-Mavarez, E.; Gee, M.; McDonald,
R.; Wasylishen, R. E. J. Am. Chem. Soc. 2001, 123, 7947.
(162) Burford, N.; Ragogna, P. J. J. Chem. Soc., Dalton Trans. 2002, 4307.
(163) Burford, N.; Ragogna, P. J.; McDonald, R.; Ferguson, M. J. J. Am. Chem. Soc. 2003, 125,
14404.
(164) Burford, N.; Ragogna, P. J.; McDonald, R.; Ferguson, M. J. Chem. Commun. 2003, 2066.
(165) Burford, N.; Dyker, C. A.; Decken, A. Angew. Chem., Int. Ed. 2005, 44, 2364.
(166) Burford, N.; Phillips, A. D.; Spinney, H. A.; Lumsden, M.; Werner-Zwanziger, U.;
Ferguson, M. J.; McDonald, R. J. Am. Chem. Soc. 2005, 127, 3921.
(167) Dyker, C. A.; Burford, N.; Lumsden, M. D.; Decken, A. J. Am. Chem. Soc. 2006, 128,
9632.
(168) Weigand, J. J.; Burford, N.; Decken, A.; Schulz, A. European Journal of Inorganic
Chemistry 2007, 4868.
160
(169) Dyker, C. A.; Riegel, S. D.; Burford, N.; Lumsden, M. D.; Decken, A. J. Am. Chem. Soc.
2007, 129, 7464.
(170) Dyker, C. A.; Burford, N.; Menard, G.; Lumsden, M. D.; Decken, A. Inorg Chem 2007,
46, 4277.
(171) Dyker, C. A.; Burford, N. Chem.-Asian J. 2008, 3, 28.
(172) Dillon, K. B.; Mathey, F.; Nixon, J. F. Phosphorus : the carbon copy : from
organophosphorus to phospha-organic chemistry; Wiley: Chichester ; New York, 1998.
(173) Denis, J.-M.; Forintos, H.; Szelke, H.; Toupet, L.; Pham, T.-N.; Madec, P.-J.; Gaumont,
A.-C. Chem. Commun. 2003, 54.
(174) Burford, N.; Dyker, C. A.; Lumsden, M.; Decken, A. Angew. Chem. Int. Ed. 2005, 44,
6196.
(175) Blackwell, J. M.; Morrison, D. J.; Piers, W. E. Tetrahedron 2002, 58, 8247.
(176) Parks, D. J.; Blackwell, J. M.; Piers, W. E. J. Org. Chem. 2000, 65, 3090.
(177) Parks, D. J.; Piers, W. E. J. Am. Chem. Soc. 1996, 118, 9440.
(178) Blackwell, J. M.; Sonmor, E. R.; Scoccitti, T.; Piers, W. E. Org. Lett. 2000, 2, 3921.
(179) Blackwell, J. M.; Piers, W. E.; Parvez, M. Org. Lett. 2000, 2, 695.
(180) Horton, A. D.; deWith, J. Organometallics 1997, 16, 5424.
(181) Boere, R. T.; Masuda, J. D. Can. J. Chem. 2002, 80, 1607.
(182) Okazaki, M.; Jung, K. A.; Tobita, H. Organometallics 2005, 24, 659.
(183) Petrie, M. A.; Power, P. P. J. Chem. Soc., Dalton Trans. 1993, 1737.
(184) Tardif, O.; Hou, Z. M.; Nishiura, M.; Koizumi, T.; Wakatsuki, Y. Organometallics 2001,
20, 4565.
(185) Tardif, O.; Nishiura, M.; Hou, Z. M. Tetrahedron 2003, 59, 10525.
(186) Hayashi, T. J. Organomet. Chem. 2002, 653, 41.
(187) Marinetti, A.; Voituriez, A. Synlett 2010, 174.
(188) Ohff, M.; Holz, J.; Quirmbach, M.; Boerner, A. Synthesis 1998, 1391.
(189) Pietrusiewicz, K. M.; Zablocka, M. Chem. Rev. 1994, 94, 1375.
(190) Vedejs, E.; Daugulis, O.; MacKay, J. A.; Rozners, E. Synlett 2001, 1499.
161
(191) Stephens, F. H.; Pons, V.; Baker, R. T. Dalton Trans. 2007, 2613.
(192) Niedenzu, K.; Dawson, J. W. Boron-Nitrogen Compounds; Academic Press, Inc: New
York, 1965.
(193) Clark, T. L.; Rodezno, J. M.; Clendenning, S. B.; Aouba, S.; Brodersen, P. M.; Lough, A.
J.; Ruda, H. E.; Manners, I. Chem-Eur J 2005, 11, 4526.
(194) Jaska, C. A.; Bartole-Scott, A.; Manners, I. Phosphorus Sulfur and Silicon and the
Related Elements 2004, 179, 685.
(195) Dorn, H.; Vejzovic, E.; Lough, A. J.; Manners, I. Inorg Chem 2001, 40, 4327.
(196) Keaton, R. J.; Blacquiere, J. M.; Baker, R. T. J. Am. Chem. Soc. 2007, 129, 1844.
(197) Himmelberger, D. W.; Yoon, C. W.; Bluhm, M. E.; Carroll, P. J.; Sneddon, L. G. J. Am.
Chem. Soc. 2009, 131, 14101.
(198) Hausdorf, S.; Baitalow, F.; Wolf, G.; Mertens, F. O. R. L. Int. J. Hydrogen Energy 2008,
33, 608.
(199) Ramachandran, P. V.; Gagare, P. D. Inorg. Chem. 2007, 46, 7810.
(200) Davis, B. L.; Dixon, D. A.; Garner, E. B.; Gordon, J. C.; Matus, M. H.; Scott, B.;
Stephens, F. H. Angew. Chem. Int. Ed. 2009, 48, 6812.
(201) Spikes, G. H.; Fettinger, J. C.; Power, P. P. J. Am. Chem. Soc. 2005, 127, 12232.
(202) Pestana, D. C.; Power, P. P. J. Am. Chem. Soc. 1991, 113, 8426.
(203) Gilbert, T. M.; Bachrach, S. M. Organometallics 2007, 26, 2672.
(204) Parks, D. J.; Spence, R. E. V. H.; Piers, W. E. Angew. Chem. Int. Ed. 1995, 34, 809.
(205) Parks, D. J.; Piers, W. E.; Yap, G. P. A. Organometallics 1998, 17, 5492.
(206) Noeth, H. Z. Anorg. Allg. Chem. 1987, 555, 79.
(207) Karsch, H. H.; Hanika, G.; Huber, B.; Riede, J.; Mueller, G. J. Organomet. Chem. 1989,
361, C25.
(208) Jakle, F.; Mattner, M.; Priermeier, T.; Wagner, M. J. Organomet. Chem. 1995, 502, 123.
(209) Feng, X.; Olmstead, M. M.; Power, P. P. Inorg. Chem. 1986, 25, 4615.
(210) Paine, R. T.; Noth, H. Chem. Rev. 1995, 95, 343.
(211) Geier, S. J.; Gilbert, T. M.; Stephan, D. W. J. Am. Chem. Soc. 2008, 130, 12632.
162
(212) Lancaster, S. J.; Mountford, A. J.; Hughes, D. L.; Schormann, M.; Bochmann, M. J.
Organomet. Chem. 2003, 680, 193.
(213) Welch, G. C., University of Windsor, 2008.
(214) Groshens, T. J.; Higa, K. T.; Nissan, R.; Butcher, R. J.; Freyer, A. J. Organometallics
1993, 12, 2904.
(215) Clarke, M. L.; Roff, G. J.; Vries, J. G. d., Elsevier, C. J., Eds.; Wiley-VCH: Weinheim,
2007, p 413.
(216) You, S. L. Chem.-Asian J. 2007, 2, 820.
(217) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1964, 2, 245.
(218) Singh, S.; Chhina, S.; Sharma, V. K.; Sachdev, S. S. J. Chem. Soc., Chem. Commun.
1982, 453.
(219) Sridharan, V.; Avendano, C.; Carlos Menendez, J. Tetrahedron 2009, 65, 2087.
(220) Sahin, A.; Cakmak, O.; Demirtas, I.; Okten, S.; Tutar, A. Tetrahedron 2008, 64, 10068.
(221) Fadel, F.; Titouani, S. L.; Soufiaoui, M.; Ajamay, H.; Mazzah, A. Tetrahedron Lett.
2004, 45, 5905.
(222) Yunnikova, L. P.; Tigina, O. V. Zh. Org. Khim. 1993, 29, 651.
(223) Friebolin, H.; Ruchardt, C. Liebigs Ann 1995, 1339.
(224) Webb, J. D.; Laberge, V. S.; Geier, S. J.; Crudden, C. M.; Stephan, D. W. Chem.-Eur. J.
2010, 16, 8.
(225) Nandi, P.; Dye, J. L.; Jackson, J. E. J. Org. Chem. 2009, 74, 5790.
(226) Murahashi, S.; Imada, Y.; Hirai, Y. Bull. Chem. Soc. Jpn. 1989, 62, 2968.
(227) Pitts, M. R.; Harrison, J. R.; Moody, C. J. J. Chem. Soc., Perkin Trans. 1 2001, 955.
(228) Bianchini, C.; Dal Santo, V.; Meli, A.; Moneti, S.; Psaro, R.; Sordelli, L.; Vizza, F. Inorg.
Chim. Acta 2008, 361, 3677.
(229) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1966, 5, 218.
(230) Lesley, M. J. G.; Woodward, A.; Taylor, N. J.; Marder, T. B.; Cazenobe, I.; Ledoux, I.;
Zyss, J.; Thornton, A.; Bruce, D. W.; Kakkar, A. K. Chem. Mater. 1998, 10, 1355.
(231) Klooster, W. T.; Koetzle, T. F.; Siegbahn, P. E. M.; Richardson, T. B.; Crabtree, R. H. J.
Am. Chem. Soc. 1999, 121, 6337.
163
(232) Parks, D. J.; Piers, W. E.; Parvez, M.; Atencio, R.; Zaworotko, M. J. Organometallics
1998, 17, 1369.
(233) Katritzky, A. R.; Rachwal, S.; Rachwal, B. Tetrahedron 1996, 52, 15031.
(234) Rabideau, P. W.; Marcinow, Z. Organic Reactions (Hoboken, NJ, United States) 1992,
42, No pp given.
(235) Radivoy, G.; Alonso, F.; Yus, M. Tetrahedron 1999, 55, 14479.
(236) Rosales, M.; Vallejo, R.; Bastidas, L. J.; Gonzalez, B.; Gonzalez, A. React. Kinet. Catal.
Lett. 2007, 92, 99.
(237) Rosales, M.; Castillo, J.; Gonzalez, A.; Gonzalez, L.; Molina, K.; Navarro, J.; Pacheco,
I.; Perez, H. Transition Met. Chem. 2004, 29, 221.
(238) Hoenel, M.; Vierhapper, F. W. J. Chem. Soc., Perkin Trans. 1 1980, 1933.
(239) Voutchkova, A. M.; Gnanamgari, D.; Jakobsche, C. E.; Butler, C.; Miller, S. J.; Parr, J.;
Crabtree, R. H. J. Organomet. Chem. 2008, 693, 1815.
(240) Fujita, K.-i.; Kitatsuji, C.; Furukawa, S.; Yamaguchi, R. Tetrahedron Lett. 2004, 45,
3215.
(241) Srikrishna, A.; Reddy, T. J.; Viswajanani, R. Tetrahedron 1996, 52, 1631.
(242) Ranu, B. C.; Jana, U.; Sarkar, A. Synth. Commun. 1998, 28, 485.
(243) Rueping, M.; Theissmann, T.; Antonchick, A. P. Catalysts for Fine Chemical Synthesis
2007, 5, 170.
(244) Rueping, M.; Antonchick, A. P.; Theissmann, T. Angew. Chem. Int. Ed. 2006, 45, 6751.
(245) Mountford, A. J.; Clegg, W.; Coles, S. J.; Harrington, R. W.; Horton, P. N.; Humphrey,
S. M.; Hursthouse, M. B.; Wright, J. A.; Lancaster, S. J. Chem.-Eur. J. 2007, 13, 4535.
(246) Kehr, G.; Roesmann, R.; Frohlich, R.; Holst, C.; Erker, G. Eur. J. Inorg. Chem. 2001,
535.
(247) Roesler, R.; Piers, W. E.; Parvez, M. J. Organomet. Chem. 2003, 680, 218.
(248) Ugolotti, J.; Hellstrom, S.; Britovsek, G. J.; Jones, T. S.; Hunt, P.; White, A. J. Dalton
Trans. 2007, 1425.
(249) Naumann, D.; Butler, H.; Gnann, R. Z. Anorg. Allg. Chem. 1992, 618.
(250) Stewart, A. P.; Trippett, S. J. Chem. Soc., C 1970, 1263.
(251) Rahman, M. M.; Liu, H. Y.; Prock, A.; Giering, W. P. Organometallics 1987, 6, 650.
164
(252) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(253) Rahman, M. M.; Liu, H. Y.; Eriks, K.; Prock, A.; Giering, W. P. Organometallics 1989,
8, 1.
(254) Steyl, G.; Roodt, A. Acta Crystallogr. C 2004, 60, m324.
(255) Lide, D. R.; Frederikse, H. P. R. CRC handbook of chemistry and physics : a ready-
reference book of chemical and physical data; 76th ed / ed.; CRC: Boca Raton, Fla. ; London,
1995.