PALLADIUM AND PLATINUM COMPLEXES FOR THE OXIDATION OF SMALL MOLECULES
A DISSERTATION SUBMITTED TO THE DEPARTMENT OF CHEMISTRY AND THE COMMITTEE ON GRADUATE STUDIES OF STANFORD UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
David M. Pearson October 2010
© 2011 by David Michael Pearson. All Rights Reserved. Re-distributed by Stanford University under license with the author.
This work is licensed under a Creative Commons Attribution- Noncommercial 3.0 United States License. http://creativecommons.org/licenses/by-nc/3.0/us/
This dissertation is online at: http://purl.stanford.edu/zn293zr6223
ii I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.
Robert Waymouth, Primary Adviser
I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.
Eric Kool
I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.
Barry Trost
Approved for the Stanford University Committee on Graduate Studies. Patricia J. Gumport, Vice Provost Graduate Education
This signature page was generated electronically upon submission of this dissertation in electronic format. An original signed hard copy of the signature page is on file in University Archives.
iii Abstract
Palladium is a versatile metal used prominently in synthesis. The rapid growth of palladium oxidation chemistry in the past decade has spurred a number of new processes that allow for the selective oxidation of substrates under mild conditions.
This growth, specifically in the field of aerobic alcohol oxidation, attracted our attention, with the hope that it would provide new catalysts for use in direct methanol fuel cells. Toward that goal we developed a new cationic palladium complex,
[(neocuproine)Pd(OAc)]2[OTf]2, which shows unprecedented initial turnover frequencies for aerobic alcohol oxidation at room temperature. However, catalyst lifetimes are limited due to the generation of reactive partially reduced oxygen species that promote oxidation of the ligand and deactivation of the catalyst.
The use of milder oxidant like benzoquinone extended catalyst lifetimes prompting a further exploration of the substrate scope. Oxidation of glycerol proceeds exclusively at the secondary alcohol to yield dihydroxyacetone. Other 1,2-diols also favors oxidation of the secondary alcohol. Methanol oxidation proceeds at a much slower rate, but yields methyl formate with selectivities greater than 90 %. The mechanism for this process was probed through the use of model studies and isotopic labels suggesting the transient generation and oxidation of methyl hemiformal. In a parallel effort toward the extension of catalyst lifetimes, a number of new oxidatively resistant ligands were synthesized and used toward the formation of palladium catalysts that exhibit extended catalyst lifetimes. Additional efforts have focused on the use of similar catalyst for the oxidative carbonylation of 1,2- and 1,3-diols to yield
iv 5-membered and 6-membered cyclic carbonates in the presence of N- chlorosuccinimide, iodosylbenzene, or dichloroisocyanuric acid.
Finally, a new platinum dimer containing a trimethylplatinum unit and bridging amides was synthesized and characterized. Its reactivity toward the reductive elimination of ethane of methanol derivatives was explored. Additional work with known platinum dimers suggests these species, upon addition of silver triflate, facilitates the C-H activation of a number of unactivated arenes and promotes the C-H functionalization of propargyl aryl ethers.
v Acknowledgements
First and foremost, I want to thank my parents, Michael and Patricia. They have shaped my very being, and have supported me through this process one calls life. They’ve taught me much of what I know outside of chemistry and I don’t think I could have gotten through graduate school without them. They have always been available to listen and to offer advice on many problems that have arisen; giving me a different and often times wiser prospective.
So many have inspired me to pursue chemistry. My first chemistry teacher Mr.
Lawson would talk frequently about his many exploits in the field before he settled down to teach. The life he lived seemed so foreign and exciting that I couldn’t help but be intrigued. The trail continued through AP chemistry where I first began to realize that the field came naturally to me eventually landing me in Professor Robert
Sheridan’s Organic Chemistry Lecture at which time I made the decision to pursue a
Bachelor’s Degree in Chemistry at the University of Nevada, Reno (UNR). I look back with fondness on my days spent in Professor David A. Lightner’s classes on
Spectroscopy and Structure Determination. They were no less than extraordinary. It was at UNR that I was introduced to inorganic and organometallic chemistry thanks to the wonderful classes taught by Professor Brain J. Frost and Professor Jason Shearer.
Finally, it was through the guidance of Professor Benjamin T. King that I first experienced chemical research in a laboratory setting. Being away from home, the
King lab soon became my second family and I have met many lifelong friends there.
vi I’d like to acknowledge my fellow labmates who showed me the joys of experimental chemistry; Cameron Hilton, Jason Ormsby, Charlie Robertson, Justin Korinek, Jeremy
Crowfoot, Pawel Rempala, and Lisa Gotari.
When I first came to Stanford I couldn’t help but feel that I was coming home in a sense. My grandparents once had a fruit orchard off Maude and Highway 237 in
Mountain View and my grandmother still lives nearby. We visited the area with some frequency as I grew up and while we had toured Stanford in the past, I never really expected to be here. The five years I have spent here have been better than I could have imagined.
I have learned so much from my teachers here at Stanford. The lectures taught by Professor Dimitry Yandalov, Professor Bernhard Breit, Professor Barry Trost,
Professor Paul Wender, Professor Robert Waymouth, and Dr. Stephen Lynch have provided me with a comprehensive background in organic and organometallic chemistry for which I am in their debt.
I would like to thank all my former and current labmates. Darcy Culkin, Yasu,
Stefan Benson, Bridgett Payne, Nahrain Kamber, Wonhee Jeong, Marc Scholton,
Nicole Kirn, Sang-Jin Jeon, Binia and Jorge, Nick Conley, Soren Randall, Karen Son,
Elizabeth Kiesewetter, Matt Kiesewetter, Ron Painter, Eun-Ji Shin, Kristen Brownell,
Justin “Junior” Edward, Hayley Brown, Anthony De Crisci, and Jeff Simon. It’s been a pleasure working with all of you.
I’ve had the pleasure of collaborating with many people through my graduate school career. I’d like to thank Nicholas Conley, Charles McCrory, Karen Son, Sang-
Jin Jeon, Anando Devadoss, Anthony De Crisci, and Ali Hosseini. While not all of
vii these collaborations have been successful, they have all been stimulating and informative. Most importantly, they have all been fun.
I have made many wonderful friends here at Stanford; DJ Kleinbaum, Douglas
Cordeau, Patrick Morris, Nate Cardin, Scott Tabakman, Charles McCrory, Anando
Devadoss, Jonathan Prange, Ali Hosseini, John Parkhill, Michelle Bond, Phillip Walls,
Ashley, Matt Cordeau, Jill Porter, Josh Rosenberg, the Wetter’s of Kiese, and many more.
The chemistry department here at Stanford has a wonderful support staff. I want to thank our CAS representative Dewi Fernandez. She’s been crucial in all the
Waymouth lab’s contact with the outside world; from sample analysis to ordering and without her, nothing would get done. Steven Lynch has been a good friend, teacher, and last but not least a great NMR laboratory manager. Alice Chen has done a great job with the mass spec facility, and has always been helpful in establishing new analysis methods and getting results. Brian Palermo has made dealing with our computers and networks painless, which is a tremendous feat. Lastly I’d like to thank
Todd Eberspacher. All the things that I can’t fathom can be done, he does! I’m still in amazement.
I would like to thank my advisor, Professor Robert M. Waymouth. Throughout my five years here I think Bob has given me a great appreciation for establishing research programs that are meaningful. The GCEP collaboration between the Stack,
Chidsey, and Waymouth labs is nothing short of ambitious. I’m proud to have been a part of the collaboration, and fortunate that we’ve found a few nice stories along the
viii path to better understanding the problems and challenges of electrocatalysis. He’s been a great advisor and friend.
Lastly, I would like to thank my wife, Amanda. I don’t think either of us could have imagined we’d be here after we first met in an art supply store, shopping for the introductory painting class for which we’d both registered. We’ve had a wonderful seven years together. She has been my foundation throughout these years. She has been incredibly patient, understanding, and supportive. There have been many evenings working late, missing dinners or weekend activities. I hope that all those sacrifices will prove worthwhile as we proceed into the next stage of our lives together.
ix Preface
“I do not write books, I write pages.”
-Dan Fante
x Table of Contents
Abstract iv
Acknowledgements vi
Preface x
Table of Contents xi
List of Tables xviii
List of Figures xx
List of Schemes xxvii
List of Abbreviations xxix
Chapter 1. Palladium Oxidation Catalysis 1
1.1 Overview of Palladium Catalysis 1
1.2 Palladium Oxidation Catalysis 3
1.2.1 General mechanisms in palladium oxidation catalysis 3
xi 1.2.2 Palladium oxidase catalysis 4
1.2.3 Oxidants other than oxygen 5
1.3 Palladium Catalyzed Oxidative Processes 8
1.3.1 Wacker oxidation 8
1.3.2 Diamination of olefins 9
1.3.3 C-H activation - oxidation 11
1.3.4 Oxidation of carbon monoxide, the water gas shift reaction and related
carbonylation chemistry 11
1.4 Palladium Catalyzed Alcohol Oxidation 12
1.4.1 History of alcohol oxidation 12
1.4.2 Mechanistic analysis of alcohol oxidation 14
1.5 References 17!
Chapter 2. Aerobic Alcohol Oxidation with Cationic Palladium Complexes:
Insights into Catalyst Design and Decomposition 31
2.0 Preface 31
2.1 Introduction 31
2.2 Results and Discussion 33
2.2.1 Synthesis and characterization of cationic palladium - neocuproine
complexes 33
2.2.2 Catalyst screening in the aerobic oxidation of 2-heptanol 37
xii 2.2.3 Identification of catalyst decomposition products 40
2.2.4 Proposal of the catalytic cycle for 1 through mechanistic interpretation 42
2.2.5 The use of alternative oxidants in alcohol oxidation 45
2.2.6 Substrate scope of alcohol oxidation 47
2.3 Experimental Section 49
2.3.1 Synthesis of compounds 49
2.3.2 Alcohol oxidation and analysis 54
2.3.3 Thermodynamic data for [(neocuproine)Pd(OAc)]2[OTf]2 56
2.3.4 Electrochemical detection of hydrogen peroxide 57
2.3.5 Conversion of µ-hydroxo Pd dimer 4 to 1 using acetic acid 62
2.3.6 Conversion of 1 to µ-hydroxo Pd dimer 4 using water 64
2.4 References 66
Chapter 3. Mechanistic Studies of the Oxidative Dehydrogenation of Methanol using a Cationic Palladium Complex 77
3.0 Preface 77
3.1 Introduction 77
3.2 Results and Discussion 79
3.3 Experimental 88
3.3.1 Methyl formate selectivity dependence of benzoquinone concentration. 89
3.3.2 Formic acid – methanol esterification 91
xiii 3.3.3 Carbon monoxide studies – oxidation of methanol under a CO atmosphere 93
3.3.4 Oxidation of CD3OH 96
3.3.5 Crossover experiments with isotopically labeled methanols 97
3.3.6 Evidence of a hemiformal intermediate 101
3.4 References 102
Chapter 4. Oxidatively-Resistant Ligands for Palladium Mediated Aerobic
Alcohol Oxidation 108
4.0 Preface 108
4.1 Introduction 108
4.2 Results and Discussion 110
4.2.1 Synthesis of oxidatively resistant ligands 110
4.2.2 Synthesis of phenanthroline palladium complexes 112
4.2.3 Catalytic aerobic alcohol oxidation 120
4.3 Conclusion 124
4.4 Preliminary Results for Future Directions 125
4.5 Experimental Section 127
4.5.1 Synthesis of compounds 128
4.5.2. Complexation attempts between Pd and bis-2,9-(trifluoromethyl)-1,10-
phenanthroline (btfm-phen) 141
xiv 4.5.3 Comparison of bond lengths and angles between complexes 8 and 1 145
4.5.4 General screening strategy for catalyst activity in the oxidation of 2-heptanol
by in-situ generation of catalysts 146
4.6 References 150
Chapter 5. Palladium Catalyzed Carbonylation of Diols to Cyclic Carbonates 159
5.0 Preface 159
5.1 Introduction 159
5.2 Results and Discussion 161
5.2.1 Catalyst screening studies 161
5.2.2 Carbonylation of 1,2-diols to yield 5-membered carbonates using N-
chlorosuccinimide as the oxidant 162
5.2.3 Carbonylation of 1,3-diols to 6-membered carbonates using hypervalent iodine
oxidants. 165
5.2.4 Carbonylation of 1,2- and 1,3-diols using sodium dichloroisocyanuric acid 169
5.2.5 Proposed mechanism for oxidative carbonylation of diols 173
5.3 Conclusion 175
5.4 Experimental Procedures 175
5.4.1 General information 175
5.4.2 Preparation of palladium complexes 176
5.4.3 Carbonylation reaction conditions 176
xv 5.4.4 Characterization of compounds 178
5.5 References 181
Chapter 6. Synthesis and Reactivity of Pt-Pt Dimers 190
6.0 Preface 190
6.1 Introduction 190
6.2 Synthetic Strategies Toward the Formation of Pt(III)-Pt(III) Dimers 196
6.3 Synthesis of Platinum(II) Monomers; Potential Synthons to Pt-Pt dimers 198
6.4 Synthesis of Pt(IV) and Pt(II) Complexes for use with Pt(II) monomers 203
6.5 Synthesis and Characterization of Platinum Dimers 205
6.5.1 Synthesis and characterization of trimethylplatinum dimers. 205
6.5.2 Synthesis of other amide bridged dimers 214
6.5.3 Attempted synthesis of dimers bearing amidine or guanidine bridges 215
6.6 Computational Analysis of Pt-Pt Dimers 217
6.7 Reactivity Studies 227
6.7.1 Reactivity of trimethyl platinum dimer 26a 227
6.7.2 Reactivity of tetrabromide platinum dimer 29 229
6.8 Modification of Carbon Electrodes and the Synthesis of Clickable Ligands to support Pt Dimers 232
xvi 6.9 Conclusion 236
6.10 Experimental 236
6.10.1 Instrumentation 236
6.10.2 Materials 237
6.10.3 Computation details 237
6.10.4 General procedures for surface functionalization and attachment 238
6.10.5 Synthesis of compounds 238
6.10.6 Thermolysis experiments 243
6.10.7 Dynamic NMR experiments 243
6.11 References 245
Appendix A. XYZ Coordinates for Geometry Optimized Pt Dimers 261
xvii List of Tables
Table 2.1 Selective oxidation of glycerol to 1,3-dihydroxyacetone using 1a 48
Table 2.2 Monomer and dimer concentrations determined by 1H NMR in a 2.56 mM
solution of 1 in acetonitrile-d3 at various temperatures. 56
Table 2.3 Calculated values for K, !G, and !S at various temperatures. 57
Table 3.1 Catalytic oxidation of 2-heptanol and methanol.a 80
Table 4.1 Efficiency of various cationic palladium complexes for the aerobic
oxidation of 2-heptanol at room temperature 124
Table 4.2 Bond lengths and angles for complexes 8 and 1 145
Table 5.1 Catalyst screening for the carbonylation of 1-phenyl-ethane-1,2-diola 162
Table 5.2 Scope of 1,2-Diol Carbonylation 163
Table 5.3 Scope of 1,2-Diol Carbonylation 164
Table 5.4 Scope of 1,3- and 1,4-diol carbonylation.a 166
Table 5.5 Elemental analyses of iodosobenzene batchesa prepared from iodobenzene
diacetate and theoretical elemental compositions for selected hypervalent iodide
species. 168
Table 5.6 Initial screening experiments with 1,3-propanediol 170
Table 5.7 Scope of 1,2-diol carbonylation 172
1 2 Table 6.1 H-NMR chemical shift data and JPt-H coupling constants for 26a 208
Table 6.2 Comparison of bond lengths between the crystal structure and geometry
optimized structures of 26. 218
xviii Table 6.3 C-H Funcationalization of propagryl aryl ether 34a 231
Table 6.4 Numerical values for the intensities of equatorial methyl resonances and
delay times !2 245
xix List of Figures
Figure 1.1 Examples of the many possible reaction catalyzed by palladium 2
Figure 1.2 Oxidation cycles for (a) Pd(II)/Pd(0) and (b) Pd(II)/Pd(IV) processes 3
Figure 1.3 Simplified oxidation cycles for metals 4
Figure 1.4 Mediator strategy for reoxidation of palladium by molecular oxygen 6
Figure 1.5 Mechanism for reoxidation of Pd(0) by benzoquinone 7
Figure 1.6 General mechanism of the Wacker oxidation. 9
Figure 1.7 Mechanisms for the catalytic diamination of olefins 10
Figure 1.8 General mechanism for CO oxidation to (a) CO2 or (b) to carbonylated
products 12
Figure 1.9 An abridged timeline of palladium-catalyzed aerobic alcohol oxidation. 13
Figure 2.1 Solid state structure of 1 with ellipsoids drawn at the 50% probability level.36
Figure 2.2 Solid-state structure of 4 on a crystallographic two-fold axis with ellipsoids
drawn at the 50% probability level. 37
Figure 2.3 Reaction progress of air oxidation (1 atm) of 2-heptanol (0.5 M) with
complexes 1, 2, 3, and 4 (3 mol % Pd) in acetonitrile at room temperature. 38
Figure 2.4 Solid-state structure of 5 with ellipsoids drawn at the 50% probability
level. 41
Figure 2.5 Oxidants screened for alcohol oxidation with
[(neocuproine)Pd(OAc)]2[OTf]2 46
xx Figure 2.6 van’t Hoff plot obtained from the monomer:dimer ratio at various
temperatures as determined by 1H NMR. 57
Figure 2.7 Cyclic voltammograms of 9.7 mM and 19.4 mM hydrogen peroxide in the
presence of 2-heptanol. 59
Figure 2.8 Monitoring of H2O2 evolution from 1 during 2-heptanol oxidation by
chronoamperometry. 60
Figure 2.9 Cyclic voltammograms of the solution before addition of 2-heptanol, 1200
s after addition of 2-heptanol, and after addition of hydrogen peroxide 61
Figure 2.10 Electrochemical detection of H2O2 using a rotating disk electrode held at
2100 mV vs. NHE and its decomposition in the presence of 1. 62
Figure 2.11 Conversion of 4 to 1 by addition of acetic acid. 63
Figure 2.12 Formation of 4 by addition of H2O to 1. 65
Figure 3.1 Ligand oxidation of cationic palladium complex 1 during the aerobic
oxidation of alcohols. 79
Figure 3.2. Pathways for the formation of methyl formate from methanol. 82
Figure 3.3 Isotopic differences as a consequence of reaction pathway in the oxidation
of CD3OH 83
1 13 Figure 3.4 H-NMR of the oxidation of CH3OH and CD3OH mixture using 1 (2.5
mol %) and benzoquinone after 100 minutes at 50 °C. 85
Figure 3.5 Mechanisms of the (a) Tishchenko and (b) Cannizzaro disproportionation
reactions. 85
Figure 3.6 Proposed mechanism for the oxidation of methanol to methyl formate by 1
and benzoquinone. 87
xxi 13 Figure 3.7 Oxidation of CH3OH using 1 and (a) 0.5 equiv (0.1 M), (b) 1 equiv (0.2
M), (c) 2 equiv (0.4M) of benzoquinone under N2(g). 90
Figure 3.8 1H-NMR spectrum of esterification reaction at t = 0. 91
Figure 3.9 1H-NMR spectrum of esterification reaction after 9 days. 92
Figure 3.10 13C-NMR spectrum of esterification reaction after 9 days. 92
Figure 3.11 1H-NMR of reaction after exposure to 1 atm of CO during the oxidation
of 13C-MeOH. 94
Figure 3.12 13C-NMR of reaction after exposure to 1 atm of CO during the oxidation
of 13C-MeOH 94
Figure 3.13 13C-NMR of methanol oxidation after 360 minutes at 50°C. 95
Figure 3.14 The postulated role of CO in oxidation of methanol to methyl formate. 96
2 Figure 3.15 H-NMR of the oxidation of CD3OH after 48 hours. 97
Figure 3.16 Possible products formed during the crossover experiment between
13 CD3OH and CH3OH. 98
13 13 Figure 3.17 Proton decoupled C-NMR of CH3OH and CD3OH mixture after 1440
minutes at 50 °C. 99
Figure 3.18 Mass spectrum of methyl formate produced in the oxidation of equimolar
mixtures of CH3OH and CD3OH at 50°C after (a) 2h and (b) 24. 100
13 13 Figure 3.19 C-NMR of CH3OH oxidation by 1 (5 mol %) and benzoquinone at
50°C after 65 minutes. 101
Figure 4.1 Substituted phenanthroline ligands 110
Figure 4.2 X-ray crystal structure of [(phen)Pd(OAc)]2[OTf]2 (8) with ellipsoids
drawn at 50% probability. 114
xxii Figure 4.3 X-ray crystal structure of [(tfmm-phen)Pd(CH3CN)2][OTf]2 (10) with
ellipsoids drawn at 50% probability. 115
Figure 4.4 X-ray crystal structure of 12 with ellipsoids drawn at 50% probability.
Hydrogen atoms have been omitted for clarity. 117
Figure 4.5 X-ray crystal structure of 15 with ellipsoids drawn at 50% probability. 119
Figure 4.6 Reaction progress for the palladium-catalyzed aerobic oxidation of 2-
heptanol to 2-heptanone using 1, 8, 9, 10, 9/10, 11/14 in acetonitrile. 121
Figure 4.7 Screening of additional ligands for the aerobic oxidation of 2-heptanol
using in situ generated palladium catalysts. 126
1 Figure 4.8 H-NMR (CDCl3) of the crude reaction mixture from attempt 2 (A2). 143
1 Figure 4.9 H-NMR of the attempts at complexation of (CH3CN)4Pd(OTf)2 with btfm-
phen in CD3CN. 144
1 Figure 4.10 H-NMR overlay of the aromatic region of (neocuproine)Pd(OAc)2,
[(neocuproine)Pd(NCCH3)2][OTf]2, and 8 147
Figure 4.11 1H-NMR overlay of the aromatic region of 9, 10, and 9/10 148
Figure 4.12 1H-NMR overlay of the aromatic region of 11, 14, 15, and 11/14 149
Figure 5.1 Proposed mechanism for carbonylation of diols with
(neocuproine)Pd(OAc)2. 174
Figure 6.1 (a) the Shilov process for methane oxidation to methanol and (b) the
proposed catalytic cycle for the reaction. 192
Figure 6.2 Platinum complexes that facilitate the oxidation of methane to methanol192
Figure 6.3 Trimethylplatinum complexes used to study the reductive elimination of
ethane or methanol derivatives. 193
xxiii 4+ Figure 6.4 Cyclic voltammogram of [(NH3)4Pt2(NHC(O)C(Me3)3)2] (8) (0.8 mM) in
aqueous KNO3 (0.1 M, pH 1) at 25 mV/s. 195
Figure 6.5 Select examples of the structural variation in platinum dimers. 195
Figure 6.6 Retrosynthetic analysis for the synthesis of platinum dimers. 197
Figure 6.7 Possible imine configurations in platinum amidine complexes 199
Figure 6.8 X-ray crystal structure of [[(CH3)3CCNH(NHCH3)]2Pt(NH3)2][ClO4]2, 20,
with ellipsoids drawn at 50% probability. 202
Figure 6.9 Bridge accepting platinum complexes used in this study. 203
Figure 6.10 Synthesis of cationic trimethyl platinum dimers 205
Figure 6.11 X-ray crystal structure of [(CH3)3Pt((CH3)3CCNH(NHCH3)]2Pt(NH3)2]
[PF6], 26b. 206
1 Figure 6.12 H-NMR of 26a in d6-benzene at 25°C. 207
Figure 6.13 Axial-equatorial exchange in 26a at 50 °C measured by 1H-NMR through
selective inversion of the axial methyl group. 209
195 Figure 6.14 Pt-NMR of 26a in d6-benzene at 25°C. 210
Figure 6.15 UV-visible spectrum of 26a in CH2Cl2 and THF 212
Figure 6.16 Molecular weight mass loss as a function of temperature for 26a. 213
Figure 6.17 Cyclic voltammogram of
[(CH3)3Pt((CH3)3CCNH(NHCH3)]2Pt(NH3)2][OTf], 26a. 213
Figure 6.18 Comparison of structure of 26 obtained from (a) the crystal structure and
(b) the computationally derived geometry optimized structure using M06-L
SDD/6-31g(d,p). 218
xxiv Figure 6.19 Simplified orbital diagram for dimers of square planar complexes of (a)
d8-d8 and (b) d7-d7 complexes. 220
Figure 6.20 Surface contour diagrams calculated by DFT (0.04 isodensity) showing
the three highest occupied and the three lowest unoccupied molecular orbitals of
26a 221
Figure 6.21 Surface contour diagrams calculated by DFT (0.04 isodensity) showing
the three highest occupied and the three lowest unoccupied molecular orbitals of
30. 222
Figure 6.22 Surface contour diagrams calculated by DFT (0.04 isodensity) showing
the three highest occupied and the three lowest unoccupied molecular orbitals of
31 224
Figure 6.23 Surface contour diagrams calculated by DFT (0.04 isodensity) showing
the three highest occupied and the three lowest unoccupied molecular orbitals of
32 225
Figure 6.24 Surface contour diagrams calculated by DFT (0.04 isodensity) showing
the three highest occupied and the three lowest unoccupied molecular orbitals of
29. 226
Figure 6.25 Comparison of structure between an activated species, generated through
the combination of 29 with silver salts, and complexes used previously for C-H
activation. 229
Figure 6.26 C-H activation of arenes by 29/AgOTf 230
xxv Figure 6.27 XPS data of the N(1s) region of a pyrolyzed photo-resist after clicking C
and the Pt(4f) region after exposure to a pH 3 solution of cis-(NH3)2Pt(OH2)2
(4mM) for 20 h. 235
xxvi List of Schemes
Scheme 2.1 Preparation of Complex 1 34
Scheme 2.2 Monomer-Dimer Equilibrium of Cationic Palladium Complex 1 35
Scheme 2.3 Generation of µ-OH dimer 4 36
Scheme 2.4 Reaction of Catalyst 1 with Hydrogen Peroxide in Acetonitrile 42
Scheme 2.5 Proposed Mechanism for Aerobic Alcohol Oxidation with Catalyst 1 43
Scheme 4.1 Ligand oxidation resulting from oxygen reduction in aerobic alcohol
oxidation 109
Scheme 4.2 Synthesis of 4-methyl-2-(trifluoromethyl)-1,10-phenanthroline (6) 111
Scheme 4.3 Abridged synthetic route to odfp-phen 7 112
Scheme 4.4 General procedure for the formation of cationic palladium complexes in
situ. 112
Scheme 4.5 Proposed mechanism for palladium-mediated fluoride substitution by
adventitious water 118
Scheme 4.6. Synthesis of 14 and 15 118
Scheme 4.7. Formation of µ-hydroxo-bridged dimers 121
Scheme 5.1 Pd-catalyzed carbonylation of 1-phenylethane-1,2-diol with CO 160
Scheme 6.1 Consecutive double nucleophilic attack on an olefin using a
Pt(III)-dimer 194
Scheme 6.2 Matsumoto’s synthesis of platinum complexes possessing amide briding
ligands. 198
xxvii Scheme 6.3 Synthesis of amidine and guanidine platinum bridging dimers. 200
Scheme 6.4 Synthesis of potential bridge acceptors 24 and 25. 204
Scheme 6.5 Synthesis of mixed dimer 29 214
Scheme 6.6 Synthesis of ESPN 233
xxviii List of Abbreviations
Anal. Calcd calculated elemental analysis aq. aqueous bathocuproine 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline BQ benzoquinone btfm-phen 2,9-bis(trifluoromethyl)-1,10-phenanthroline Bu butyl CO carbon monoxide DFT density functional theory dichlor sodium dichloroisocyanuric acid diphos 1.2-bis-(diphenylphosphino)ethane dmso dimethylsulfoxide EA elemental analysis ECP effective core potential eq. equivalents ESI electrospray ionization esp 3,3'-(1,3-phenylene)bis(2,2-dimethylpropanoic acid) ESPN 3,3'-(1,3-phenylene)bis(2,2-dimethylpropanamide) Et ethyl eV electron volt FID flame ionization detector GC gas chromatography gem geminal HOMO highest occupied molecular orbital IMes 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene IS internal standard KIE kinetic isotope effect
xxix LUMO lowest unoccupied molecular orbital m/z mass-to-charge ratio Me methyl MF methyl formate MS mass spectrometry NaHMDS sodium hexamethyldisilazide NCS N-chlorosuccinimide neocuproine 2,9-dimethyl-1,10-phenanthroline NHC N-heterocyclic carbene NHE normal hydrogen electrode NMR nuclear magnetic resonance odfp-phen 2-(2´,6´-difluoro-phenyl)-1,10-phenanthroline OPiv trimethylacetate OTf trifluoromethylsulfonate phen 1,10-phenanthroline PhIO iodosobenzene PPF pyrolyzed photo-resist ppm parts per million PROS partially reduced oxygen species RF response factor SAM self-assembled monolayer TBA tetrabutylammonium TCE 1,1,2,2-tetrachloroethane TFA trifluoroacetic acid TFE trifluoroethanol tfmm-phen 2-(trifluoro)-4-methyl-1,10-phenanthroline THF tetrahydrofuran TLC thin layer chromatography TOF turnover frequency
TOFi initial turnover frequency
xxx TON turnover number trichlor trichloroisocyanuric acid triflate trifluoromethylsulfonate TTMA (triethyl-2,2",2""-(4,4",4""- nitrilotris(methylene)tris(1H-1,2,3-triazole-4,1- diyl))triacetate UV ultraviolet XPS x-ray photoelectron spectroscopy HH head to head HT head to tail
xxxi Chapter 1.
Palladium Oxidation Catalysis
1.1 Overview of Palladium Catalysis
Palladium is a group 10 metal with an estimated terrestrial abundance of 0.015 ppm.1 It is most commonly found in the oxidation states 0, +2, and +4.2 Oxidation states of +1 and +3 are rare but are increasingly found in bimetallic complexes where the odd electron can participate in metal-metal bonding.3-7 Palladium participates in a number of reactions with organic substrates and is one of the most versatile metals for use in organic synthesis.8
Palladium is most well known for its use with olefinic and aryl substrates in coupling reactions. Many useful coupling reactions have been reported (e.g. Suzuki,9
Stille,10 Heck,11 Tsuji-Trost,12,13 Sonogashira,14 Negishi,15 etc16) establishing palladium catalysis as a powerful tool in the synthesis of carbon-carbon bonds. The scope of reactions possible with palladium has expanded dramatically allowing for the catalytic formation of carbon-oxygen,17 carbon-nitrogen,18,19 carbon-halogen,20 carbon-boron,21 carbon-phosphorus,22 and carbon-sulfur23 bonds among many others.
1
Ar R N
X O R R' Ar-X N ox Ph-M R R H2O OH O Nu Ar-X Pd R R' Nu R R' Nu OR Nu Nu CO Ar-X R O Nu Nu Nu R
Figure 1.1 Examples of the many possible reaction catalyzed by palladium
This chapter will focus on the oxidative chemistry of palladium. A brief discussion of general mechanisms for oxidative transformations using palladium will be provided along with examples of several major oxidative transformations. Finally, the field of alcohol oxidation will be discussed with a focus on its mechanism.
2 1.2 Palladium Oxidation Catalysis
1.2.1 General mechanisms in palladium oxidation catalysis
Palladium oxidation chemistry typically proceeds either through a (II)/(0) or a
(II)/(IV) cycle. Typical Pd (II)/(0) cycles proceed through oxidation of a substrate by Pd
(II) producing a reduced Pd(0) species which must be reoxidized either by an exogenous oxidant or an oxidized substrate. The reoxidation of Pd(0) is of paramount importance due to the inherent instability of Pd(0) toward aggregation in the form of palladium black
(!H " -378 kJ/mol).24 a.) b.) (II) Sub Pd(0) Pd SubOX Ox SubOX
Pd-Sub
Reductive SubH Red Ox 2 Elimination (IV) Pd(II) Pd
Figure 1.2 Oxidation cycles for (a) Pd(II)/Pd(0) and (b) Pd(II)/Pd(IV) processes
Pd(II) / Pd(IV) cycles are often invoked in the case where strong oxidants are used and the reductive elimination of an intermediate is known to be slow or non-existent from Pd(II) complexes.25,26 Oxidation to a Pd(IV) intermediate often produces a more reactive species from which reductive elimination can proceed.26 For instance, this strategy has been used to promote reductive elimination to form carbon-oxygen,2 carbon- fluorine,27 and methyl-methyl28 bonds. Palladium (II)/(III) cycles have also been invoked, and recent reports suggest that this pathway may actually occur in several reactions previously thought to proceed through (II)/(IV) cycles.4,5
3 1.2.2 Palladium oxidase catalysis
The use of molecular oxygen as a terminal oxidant in oxidation chemistry has many practical advantages.29 Besides the obvious economic benefits, the biproduct from oxygen reduction, hydrogen peroxide or water, are environmentally benign. There are two major cycles by which oxygen can be employed in catalysis. These cycles differ in the role of oxygen in the cycle. Oxygenase pathways use the oxygen to facilitate oxygen- atom transfer to the desired substrate. In contrast, oxidase pathways use oxygen exclusively to reoxidize the metal catalyst, and the oxygen used in oxidation is not transferred to the substrate. Uncoupling catalyst reoxidation from substrate oxidation provides additional benefits. For example, substrate oxidation need not explicitly result in C-O bond formation but can incorporate other functionalities such as heteroatom or halogens into the product. Aerobic palladium oxidation chemistry typically proceeds through oxidase pathways.
O a.) b.) M(n+2)+ M(n+2)+ H2O H2O Sub SubH2 (H2O2)
+ + O2 + 2H Sub(O) 1/2 O2 + 2H SubOX - + 2e n+ (O ) n+ M 2 M
Figure 1.3. Simplified oxidation cycles for metals: (a) oxygenase pathway whereby the metal facilitates oxygen-atom transfer to the substrates and (b) oxidase pathway where oxygen serves solely as a reoxidant for the metal.
The reoxidation of Pd by O2 has been the subject of numerous studies. Pd(0) complexes have been reported to react with O2 to produce Pd-O2 complexes, several of
4 which have been structurally characterized.30 These complexes display long O-O bond lengths similar to that of hydrogen peroxide and can be described as peroxopalladium(II) complexes. Addition of Brønsted acids to these complexes results in the formation of hydrogen peroxide and a Pd(II) complex.31,32 Another possible route for palladium reoxidation by dioxygen does not formally proceed through a Pd(0) intermediate. Instead oxygen can insert directly into Pd-H complexes, formed either through #-H elimination or by protonation of Pd(0) complexes.33 Computational33-36 and experimental31,32,37 work suggests that both pathways are feasible and that the exact pathway is likely to be highly dependent on the experimental conditions.
Independent of the process employed for Pd reoxidation, hydrogen peroxide intermediates are a biproduct of aerobic oxidation chemistry mediated by palladium. In most cases, H2O2 produced during oxidation reactions undergoes disproportionation to yield H2O and O2. The mechanism for H2O2 disproportionation by palladium catalysts is currently unknown.30
1.2.3 Oxidants other than oxygen
While the direct use of molecular oxygen has several benefits, it may oftentimes be impractical for a given system, necessitating the use of other oxidants or oxidant/mediators. This has classically been the case in the Wacker oxidation (see
Section 1.3.1). Due to the kinetic sluggishness of oxygen to facilitate the reoxidation of the palladium species used, a copper oxidant/mediator is often employed. While Cu(II) salts can be used stoichiometrically to reoxidize Pd(0) to Pd(II), it is also possible in many cases to use a mediator strategy whereby catalytic amounts of copper reoxidize
5 palladium and in turn are reoxidized by molecular oxygen (Figure 1.4).38 Other stoichiometric oxidants for palladium like benzoquinone39 or iodine40 can participate in this mediator strategy.
n Mox PdII H2O SubH2 (H2O2)
+ 1/2 O2 + 2H SubOX (O ) 2 n Mred Pd0 Figure 1.4 Mediator strategy for reoxidation of palladium by molecular oxygen
Electron deficient alkenes are known to bind to and stabilize Pd(0) species due to metal to ligand $-backbonding interactions. In select cases, the addition of protons to these species can result in a reduction of the alkene with concomitant generation of
Pd(II). Benzoquinone is the most common alkene for this type of reoxidation, but other electron deficient alkenes41-43 and azo compounds44 have been reported to behave in an analogous fashion. A general mechanism for reoxidation with benzoquinone is presented in Figure 1.5. Our group has employed benzoquinone as a terminal oxidant in the oxidation of methanol, propane diol, glycerol, and a number of other polyols (see Section
2.2.5 and Chapter 3).
6 O OH OH OH HX HX 0 0 II + (Ln)Pd (Ln)Pd (Ln)Pd Pd(II)X2
O O O OH
OH
II (Ln)Pd O OH II (Ln)Pd O Figure 1.5 Mechanism for reoxidation of Pd(0) by benzoquinone
Several other oxidants have proved effective for the oxidation of Pd(0) to Pd(II), including one-electron oxidants: Ag(I),45,46 ferrocene,47 and Th(III)48 and two electron oxidants: chlorobenzene,49 N-halosuccinimides,50,51 halogens,52 potassium persulfate
53 54 55,56 (K2S2O8), potassium peroxymonosulfate (KHSO5), peroxides, and iodine (III) reagents.25 Many of these oxidants, are also competent for the oxidation of Pd(II) to
Pd(IV) specifically in the fields of diamination and C-H functionalization.
Palladium can also be oxidized electrochemically. For example, Pd(PPh3)4 is
57 oxidized at 0.298 V (NHE, 2 mM in DMF, N(Bu)4BF4 0.3M). The oxidation of the Pd-
57 H, generated by addition of strong acids to Pd(PPh3)4, is significantly more positive.
While the majority of electrochemical studies on Pd have focused on Pd(0) and Pd(II) redox processes, electrochemical oxidation of Pd(II) to Pd(IV) species is known.58 Use of the electrode as a terminal oxidant for palladium mediated alcohol oxidation has also been demonstrated using benzoquinone as a mediator.59
7 1.3 Palladium Catalyzed Oxidative Processes
Several oxidative processes are possible with palladium. This chapter will only touch upon a few major processes with a focus on mechanism and the oxidants used.
Oxidation chemistry of palladium is the subject of several great reviews.29,60,61
1.3.1 Wacker oxidation
The Wacker oxidation is a major industrial process for the conversion of ethylene to acetaldehyde. It is one of the few industrial processes that employ a homogeneous catalysts.62 A general scheme for the process is given in Figure 1.6. The process has been the subject of numerous mechanistic studies, often times providing confusing or contradictory information about the mechanism.63 Keith has recently provided a review on the subject in hopes of clarifying mechanistic details.64 Much of the confusion stems from a failure to recognize that the solvent, its pH, the presence of ligands, and their concentrations can all affect the mechanism; leading not only to different reaction pathways but at times to different products (e.g. chlorohydrins).64
8
Figure 1.6 General mechanism of the Wacker oxidation.
The Wacker oxidation goes through a Pd(II)/(0) cycle and typically relies on a Cu mediator to couple the reaction to the terminal oxidant, oxygen. Recent advances in the palladium oxidase chemistry have provided several methods for the use of molecular oxygen directly in Wacker-type processes.65-67
1.3.2 Diamination of olefins
Vicinal diamines have a privileged role in chemistry. They are found in a variety of biologically active compounds and in the framework of many chelating ligands in metal based catalysis.68 Palladium has had a prominent role in the field of catalytic diamination.69 Bäckvall reported the first stoichiometric palladium promoted diamination reaction in 1978.70 The reaction required addition of an oxidant and was believed to
9 proceed via a Pd(II)/Pd(IV) cycle relying on bromine and other strong oxidants to oxidize the Pd(II) intermediate in the reaction. Bäckvall’s system was made catalytic by Muñiz through the use of an intramolecular cyclization of a tethered urea and PhI(OAc)2 as the terminal oxidant.71
Figure 1.7 Mechanisms for the catalytic diamination of olefins
A similar strategy has been employed in an intermolecular fashion by Booker-
Milburn and co-workers in this case invoking a Pd(II)/Pd(0) cycle allowing for the use of
72 benzoquinone or O2 oxidants. This system required the use of dienes, in contrast to the
Muñiz system with unconjugated terminal olefins. The use of a diene allows for the formation of electrophilic Pd-allyl species which can undergo nucleophilic attack by amines without oxidation of the Pd(II) to Pd(IV).73 Shi and co-workers employed a similar strategy utilizing dienes and a Pd(II)/Pd(0) catalytic cycle.74 This system is unique in that instead of using an amine and an oxidant, the two were combined in the form of diaziridinone.
10 1.3.3 C-H activation - oxidation
Palladium is known to perform C-H activation-oxidation chemistry.48,75 However, selectivity has often been a problem.76 While several early examples of directed C-H activation using palladium and other metals were known,75 subsequent functionalization of these substrates was scarce.77 Sanford and coworkers demonstrated that these metallocycles could be induced to further react with substrates to yield functionalized
25 products. These reactions commonly employ strong oxidants like PhI(OAc)2 suggesting that Pd(III) or Pd(IV) species are crucial to catalytic functionalization. (Figure 1.2b)4,78
Following the initial reports, this field has expanded to provide a number of new strategies for C-H activation functionalization opening a route to the formation of C-C,
C-N, C-O, and C-X bonds.60
1.3.4 Oxidation of carbon monoxide, the water gas shift reaction and related carbonylation chemistry
Palladium is employed throughout the world in automobile catalytic converters for the oxidation of carbon monoxide,79,80 and it shows some activity for the analogous water-gas shift reaction.81 Homogenous palladium catalysts for CO oxidation have also been reported, however, these systems commonly employ oxidatively-prone phosphines ligands that limit the lifetime of such catalysts.82 From these initial reports, researchers at
Eni-chem were able to create more robust catalysts by using phenanthrolines.83-85 With
83,84 these complexes they used CO as a reductant for O2 to generate hydrogen peroxide.
11 H -Ox a.) H2-Ox b.) 2 Pd(II) Pd(II) + 2H+ + Ox 2H + Ox
Pd(0) Pd CO Pd(0) Pd CO
O H O CO O 2 O Nu 2 Nu Nu H + Pd Pd + H + OH + + H Nu H NuH H+
Figure 1.8 General mechanism for CO oxidation to (a) CO2 or (b) to carbonylated products
A general mechanism for CO oxidation by palladium is proposed in Figure 1.8a.
First CO binds to palladium to generate the carbonyl complex. Nucleophilic attack by water generates a Pd-acyl, which can undergo decarboxylation to form a Pd-H. A closer look at this mechanism reveals that the nucleophile may not need to be water and that other nucleophilics (e.g. alcohols or amines) should be acceptable in the oxidation of CO
(Figure 1.8b).86 In this case, completion of the cycle requires the inclusion of additional nucleophile or electrophile (e.g. an alkene). The realization that Pd can participate in this second pathway has lead to a series of useful carbonylation reactions to produce ureas, carbamates, carbonates, and carboxylic acid derivatives.29,87,88 Initial reports of
86 oxidative carbonylations with palladium relied on CuCl2. Since then, a number of
89 oxidants have been employed: such as halogens and O2. Our efforts toward the oxidative carbonylation of diols to yield cyclic carbonates is described in Chapter 5.
1.4 Palladium Catalyzed Alcohol Oxidation
1.4.1 History of alcohol oxidation
12 In 1828, Berzelius firsts observed that the heating of palladium(II) salts in the presence of alcohols results in the formation of a stoichiometric quantity of metallic palladium and the corresponding aldehyde, ketone, or carboxylic acid.90 Since these initial reports, much work has focused on the reoxidation of Pd(0) species, allowing for the development of catalytic strategies for alcohol oxidation. The first successful catalytic oxidation was reported in 1967 by Lloyd who employed Cu(II) salts to reoxidize
Pd(0).43
Figure 1.9. An abridged timeline of palladium-catalyzed aerobic alcohol oxidation.
Almost 10 years laters Schwartz and Blackburn reported the first catalytic aerobic
91 alcohol oxidation in the absence of cocatalyst. This report employed a mixture of PdCl2 and NaOAc in ethylene carbonate solvent. Finally, in the late 1980’s Moiseev and
13 coworkers reported formation of well defined giant Pd clusters which also catalyzed the aerobic oxidation of alcohols.92,93
A renaissance of palladium catalyzed alcohol oxidation occurred in the late
1990’s. Almost simultaneously the groups of Larock94 and Uemura95 published what would be the two prominent systems for alcohol oxidation. Subsequently Sheldon reported a water soluble palladium catalyst for Wacker oxidation, as part of a series of papers detailing the groups efforts in aerobic oxidation chemistry,65 which was later was reported to promote aerobic alcohol oxidation.96 Additional bidentate ligands for palladium were subsequently discovered. Moberg reported a C,N-palladium metallocycle which exhibited modest activities for the oxidation of 1-phenylethanol.97 In near succession, the Sigman and Stoltz group independently reported the use of sparteine with palladium salts toward enantioselective alcohol oxidation.98,99 The growing popularity of N-heterocyclic carbenes (NHC), and its many appealing properties in metal based catalysis, prompted the Sigman group to explore their use with Pd in alcohol oxidation resulting in highly active catalysts under mild conditions.100,101 Finally, our group entered the field in 2007 with the report of a cationic analogue of the Sheldon system that displays the highest reported initial turnover frequency at room temperature to date. Further analysis of this system, the use of other oxidants, and ligand analogues are the subjects of Chapters 2-4.
1.4.2 Mechanistic analysis of alcohol oxidation
There are several crucial steps involved in palladium catalyzed alcohol oxidation.
These can roughly be divided into 6 sections (Figure 1.10):
14 (1) generation of an open coordination site for alcohol coordination
(2) alcohol coordination
(3) deprotonation of the palladium bound alcohol
(4) generation of an open coordination site for #-H elimination
(5) !-H elimination to yield the oxidized product and Pd-H or Pd(0)
(6) reoxidation of Pd and ligand exchange to regenerate the active species
It may often be necessary to take into account additional steps relating to palladium speciation and ligand coordination depending on the system considered. Steps 1-5 generally dominate the observed reaction rates for many of the known palladium systems.
Step 6 was believed to be rate limiting in the Pd/DMSO system,102 but this mechanistic
103 interpretation was later revised and attributed to the low solubility of O2 in DMSO.
The reoxidation of Pd by benzoquinone (Step 6) has recently been found to be the rate limiting step in the oxidation of diols in DMSO using [(neocuproine)Pd(OAc)]2[OTf]2.
To the best of our knowledge it is the only case of rate limiting reoxidation of palladium in catalytic oxidations in the presence of an abundance of oxidant.
The use of molecular oxygen is prevalent in palladium alcohol oxidation chemistry. The direct use of oxygen necessitates the use of ligands that stabilize Pd(0) and promote its reoxidation by oxygen vida supra. This poses an interesting problem for catalysts development, as the generation of open coordination sites is also crucial for several steps in alcohol oxidation. This problem was expressed by Stahl and coworkers
104 in their study of Uemura’s pyridine/Pd(OAc)2 system. While lowering the pyridine concentration lead to more active catalysts due to the relative acceleration of steps 1 and
4, catalyst lifetimes were greatly diminished due to limitations in the reoxidation step 6.
15
Figure 1.10 Crucial steps in palladium mediated alcohol oxidation.
Sigman and coworkers demonstrated that through the use of NHC ligands, which form strong bonds to palladium, the need for additional ligands to stabilize palladium was unnecessary.100 Furthermore, as a monodentate ligand, an open coordination site for alcohol oxidation and #-H elimination should be readily accessible. In alignment with this hypothesis, Sigman’s NHC-based catalyst exhibited impressive activity and is one of only a few systems known to operate at room temperature.101
Bidentate ligand systems figure prominently in alcohol oxidation catalysis vida supra. Unlike the NHC system described earlier, the requirement for ligand dissociation can sometimes be problematic in these systems. Several strategies to provide ready
16 access to open coordination sites in palladium systems with bidentate ligands have been reported; either through the use of anionic ligands, as in Moberg’s systems, 97 or through the generation of cationic complexes, which is the subject of Chapters 2-4. In water,
1,10-phenanthroline complexes speciate as µ-hydroxy dimers which appear to be resting states for the catalyst. As a result, kinetics studies of these systems display a half-order dependence on catalyst. This process can fit by analogy into Step 1 of Figure 1.10 with L
= (HO)Pd(N-N). Attempts to increase the rate of Step 1 (Figure 1.10) by destabilizing the
µ-hydroxy dimers have been marginally successful, and have largely relied on the incorporation of steric bulk in the 2 and 9 positions of 1,10-phenanthrolines.105 However, large increases in the steric bulk can promote palladium dissociation and decomposition.84 Due to these factors 2,9-dimethyl-1,10-phenanthroline (neocuproine) typically displayed the highest activity of the ligands studied by Sheldon and coworkers.105
1.5 References
[1] Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; 2nd ed.;
Butterworth-Heinemann: Woburn, MA, 1997; Vol. 27.
[2] Hartley, F. R. The Chemistry of Platinum and Palladium; Applied Science
Publishers: London, 1973.
[3] Kullberg, M. C.; Kubiak, C. P. Organometallic Chemistry in Aqueous Solution:
Water Soluble Bis(dimethylphophino)methane-Bridged Dipalladium Complexes.
Organometallics 1984, 3, 632-634.
17 [4] Powers, D. C.; Ritter, T. Bimetallic Pd(III) complexes in palladium-catalysed
carbon-heteroatom bond formation. Nature Chemistry 2009, 1, 302-309.
[5] Powers, D. C.; Geibel, M. A. L.; Klein, J. E. M. N.; Ritter, T. Bimetallic
Palladium Catalysis: Direct Observation of Pd(III)-Pd(III) Intermediates. J. Am.
Chem. Soc. 2009, 131, 17050-17051.
[6] Cotton, F. A.; Koshevoy, I. O.; Lahuerta, P.; Murillo, C. A.; Sanaú High Yield
6+ Syntheses of Stable, Singly Bonded Pd2 Compounds. J. Am. Chem. Soc. 2006,
126, 13674-13675.
[7] Berry, J. F.; Bill, E.; Borthe, E.; Cotton, F. A.; Dalal, N. S.; Ibragimov, S. A.;
Kaur, N.; Liu, C. Y.; Murillo, C. A.; Nellutla, S.; North, J. M.; Villagrán, S. A
Fractional Bond Order of 1/2 in Pd2.5+ Formamidinate Species; The Value of Very
High-Field EPR Spectra. J. Am. Chem. Soc. 2007, 129, 1393-1401.
[8] Palladium in Organic Synthsis; Tsuji, J., Ed.; Springer: Berlin, 2005.
[9] Miyaura, N.; Suzuki, A. Palladium-Catalyzed Cross-Coupling Reactions of
Organoboron Compounds. Chem. Rev. 1995, 95, 2457-2483.
[10] Stille, J. K. The Palladium-Catalyzed Cross-Coupling Reactions of Organotin
Reagents with Organic Electrophiles. Angew. Chem., Int. Ed. 1986, 25, 508-524.
[11] Heck, R. F. Palladium-Catalyzed Vinylation of Organic Halides. Organic
Reactions 1982, 27, 345-390.
[12] Tsuji, J. Carbon-Carbon Bond Formation via Palladium Complexes. Acc. Chem.
Res. 1969, 2, 144-152.
18 [13] Godleski, S. A. Nucleophiles with Allyl-Metal Complexes. In Comprehensive
Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford,
1991; Vol. 4, Chapter 3.3.
[14] Sonogashira, K. Cross-Coupling Reactions to sp Carbon Atoms. In Metal-
catalyzed Cross-coupling Reactions; Diederich, F., Stang, P. J., Eds.; Wiley-
VCH: Weinheim, 1998; Vol. 1, Chapter 5.
[15] Negishi, E. Palladium- or Nickel-Catalyzed Cross Coupling. A New Selective
Method for Carbon-Carbon Bond Formation. Acc. Chem. Res. 1982, 15, 340-348.
[16] Metal-catalyzed Cross-coupling Reactions; Diedrich, F.; Stang, P. J., Eds.; Wiley-
VCH: Weinheim, 1998.
[17] Desai, L. V.; Hull, K. L.; Sanford, M. S. Palladium-Catalyzed Oxygenation of
Unactivated sp3 C-H Bonds. J. Am. Chem. Soc. 2004, 126, 9542-9543.
[18] Hartwig, J. F. Carbon-Heteroatom Bond-Forming reductive Eliminations of
Amines, Ethers, and Sulfides. Acc. Chem. Res. 1998, 31, 852-860.
[19] Wolfe, J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Rational Development
of Practical Catalysts for Aromatic Carbon-Nitrogen Bond Formation. Acc. Chem.
Res. 1998, 31, 805-818.
[20] Vigalok, A. Metal-Mediated Formation of Carbon-Halogen Bonds. Chem.--Eur.
J. 2008, 14, 5102-5108.
[21] Ishiyama, T.; Murata, M.; Miyaura, N. Palladium(0)-Catalyzed Cross-Coupling
Reaction of Alkoxydiboron with Haloarenes: A Direct Procedure for Arylboronic
Esters. J. Org. Chem. 1995, 60, 7508-7510.
19 [22] Schwan, A. L. Palladium Catalyzed Cross-Coupling Reactions for Phosphorus-
Carbon Bond Formation. Chem. Soc. Rev. 2004, 33, 218-224.
[23] Kondo, T.; Mitsudo, T. Metal-Catalyzed Carbon-Sulfur Bond Formation. Chem.
Rev. 2000, 100, 3205-3220.
[24] ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Catalytic Conversion in
Water. Part 21: Mechanistic Investigations on the Palladium-Catalysed Aerobic
Oxidation of Alcohols in Water. Adv. Synth. Catal. 2002, 344, 355-369.
[25] Dick, A. R.; Hull, K. L.; Sanford, M. S. A Highly Selective Catalytic Method for
the Oxidative Functionalization of C-H Bonds. J. Am. Chem. Soc. 2004, 126,
2300-2301.
[26] Sehnai, P.; Taylor, R. J. K.; Fairlamb, I. J. S. Emergence of Palladium(IV)
Chemistry in Synthesis and Catalysis. Chem. Rev. 2010, 110, 824-889.
[27] Furuya, T.; Benitez, D.; Tkatchouk, E.; Strom, A. E.; Tang, P.; Goddard III, W.
A.; Ritter, T. Mechanism of C-F Reductive Elimination from Palladium(IV)
Fluorides. J. Am. Chem. Soc. 2010, 132, 3793-3807.
[28] Lanci, M. P.; Remy, M. S.; Kaminsky, W.; Mayer, J. M.; Sanford, M. S.
t Oxidatively Induced Reductive Elimination from ( Bu2bpy)Pd(Me)2:
Palladium(IV) Intermediates in a One-Electron Oxidation Reaction J. Am. Chem.
Soc. 2009, 131, 15618-15620.
[29] Stahl, S. S. Palladium Oxidase Catalysis: Selective Oxidation of Organic
Chemicals by Direct Dioxygen-Coupled Turnover. Angew. Chem., Int. Ed. 2004,
43, 3400-3420.
20 [30] Popp, B. V.; Stahl, S. S. Palladium-Catalyzed Oxidation Reactions: Comparison
of Benzoquinone and Molecular Oxygen as Stoichiometric Oxidants. Top.
Organomet. Chem. 2007, 22, 149-189.
[31] Konnick, M. M.; Guzei, I. A.; Stahl, S. S. Characterization of Peroxo and
Hydroperoxo Intermediates in the Aerobic Oxidation of N-Heterocyclic-Carbene-
Coordinated Palladium(0). J. Am. Chem. Soc. 2004, 126, 10212.
[32] Stahl, S. S.; Thorman, J. L.; Nelson, R. C.; Kozee, M. A. Oxygenation of
Nitrogen-Coordinated Palladium(0): Synthetic, Structural, and Mechanistic
Studies and Implications for Aerobic Oxidation Catalysis. J. Am. Chem. Soc.
2001, 123, 7188.
[33] Denney, M. C.; Smythe, N. A.; Cetto, K. L.; Kemp, R. A.; Goldberg, K. I.
Insertion of Molecular Oxygen into a Palladium(II) Hydride Bond. J. Am. Chem.
Soc. 2006, 128, 2508-2509.
[34] Popp, B. V.; Stahl, S. S. Insertion of Molecular Oxygen into a Palladium-Hydride
Bond; Computation Evidence for Two Nearly Isoenergetic Pathways. J. Am.
Chem. Soc. 2007, 129, 4410-4422.
[35] Keith, J. M.; A., G. I. W. Mechanism for Activation of Molecular Oxygen by cis-
and trans-(Pyridine)2Pd(OAc)H: Pd0 versus Direct Insertion. J. Am. Chem. Soc.
2009, 131, 1416-1425.
[36] Keith, J. M.; Goddard III, W. A.; Oxgaard, J. Pd-Mediated Activation of
Molecular Oxygen: Pd(0) versus Direct Insertion. J. Am. Chem. Soc. 2007, 129,
10361-10369.
21 [37] Konnick, M. M.; Gandhi, B. A.; Guzei, I. A.; Stahl, S. S. Reaction of Molecular
Oxygen with a PdII-Hydride to Produce a PdII-Hydroperoxide: Acid Catalysis and
Implications for Pd-Catalyzed Aerobic Oxidation Reactions. Angew. Chem., Int.
Ed. 2006, 45, 2904-2907.
[38] Piera, J.; Bäckvall, J. E. Catalytic Oxidation of Organic Substrates by Molecular
Oxygen and Hydrogen Peroxide by Multistep Electron Transfer - A Biomimetic
Approach. Angew. Chem., Int. Ed. 2008, 47, 3506-3523.
[39] Baeckvall, J. E.; Hopkinds, R. B.; Grennberg, H.; Mader, M.; Awasthi, A. K.
Multistep Electron Transfer in Palladium-catalyzed Aerobic Oxidations via a
Metal Macrocycle Quinone System. J. Am. Chem. Soc. 1990, 112, 5160-5166.
[40] Gabriele, B.; Costa, M.; Salerno, G.; Chiusoli, G. P. An Efficient and Selective
Palladium-catalysed Oxidation Dicarbonylation of Alkynes to Alkyl- or Aryl-
maleic Esters. J. Chem. Soc., Perkin Trans. 1 1994, 83-87.
[41] Tsuchiya, Y.; Hamashima, Y.; Sodeoka, M. A New Entry to Pd-H Chemistry:
Catalytic Asymmetric Conjugate Reduction of Enones with EtOH and a Higher
Enantioselective Synthesis of Warfarin. Org. Lett. 2006, 8, 4851-4854.
[42] Monguchi, D.; Beemelmann, C.; Hashizume, D.; Hamashima, Y.; Sodeoka, M.
Catalytic asymmetric conjugate reduction with ethanol: A more reactive system
Pd(II)-iPr-DUPHOS complex with molecular sieves 4Å. J. Organomet. Chem.
2008, 693, 867-873.
[43] Lloyd, W. G. Homogeneous Oxidation of Alcohols with Palladium(II) Salts. J.
Org. Chem. 1967, 32, 2816-2819.
22 [44] Muñiz, K.; Nieger, M. Catalytic Activation of N-N Multiple Bonds: A
Homogeneous Palladium Catalyst for Mechanistically Unprecedented Reduction
of Azo Compounds. Angew. Chem., Int. Ed. 2006, 46, 2305-2308.
[45] Takahashi, M.; Masui, K.; Sekiguchi, H.; Kobayashi, N.; Mori, A.; Funahashi,
M.; Tamaoki, N. Palladium-Catalyzed C-H Homocoupling of Bromothiophene
Derivatives and Synthetic Application to Well-Defined Oligothiophenes. J. Am.
Chem. Soc. 2006, 128, 10930-10933.
[46] Hull, K. L.; Sanford, M. S. Catalytic and High Regioselective Cross-Coupling of
Aromatic C-H Substrates. J. Am. Chem. Soc. 2007, 129, 11904-11905.
[47] Lanci, M. P.; Remy, M. S.; Kaminsky, W.; Mayer, J. M.; Sanford, M. S.
Oxidatively Induced Reductive Elimination from (tBu2bpy)Pd(Me)2:
Palladium(IV) Intermediates in a One-Electron Oxidation Reaction. J. Am. Chem.
Soc. 2009, 131, 15618-15620.
[48] Henry, P. M. Palladium Catalyzed Oxidation of Hydrocarbons; Reidel:
Dordrecht, 1980.
[49] Berini, C.; Brayton, D. F.; Mocka, C.; Navarro, O. Homogeneous, Anaerobic (N-
Heterocyclic Carbene) - Pd or -Ni Catalyzed Oxidation of Secondary Alcohols at
Mild Temperatures. Org. Lett. 2009, 11, 4244-4247.
[50] Whitfield, S. R.; Sanford, M. S. Reactivity of Pd(II) Complexes with Electrophilic
Chlorinating Reagents: Isolation of Pd(IV) Products and Observation of C-Cl
Bond-Forming Reductive Elimination. J. Am. Chem. Soc. 2007, 129, 15142-
15143.
23 [51] Kalyani, D.; Dick, A. R.; Anani, W. Q.; Sanford, M. S. A Simple Catalytic
Method for the Regioselective Halogenation of Arenes. Org. Lett. 2006, 8, 2523-
2526.
[52] Pri-Bar, I.; Schwartz, J. I2-Promoted Palladium-Catalyzed Carbonylation of
Amines. J. Org. Chem. 1995, 60, 8124-8125.
[53] Muehlhofer, M.; Strassner, T.; Herrmann, W. A. New Catalyst Systems for the
Catalytic Conversion of Methane into Methanol. Angew. Chem., Int. Ed. 2002, 41,
1745-1747.
[54] Hull, K. L.; Lanni, E. L.; Sanford, M. S. Highly Regioselective Catalytic
Oxidative Coupling Reactions: Synthetic and Mechanistic Investigations. J. Am.
Chem. Soc. 2006, 128, 14047-14049.
[55] Yin, G.; Liu, G. Palladium-Catalyzed Oxidative Cyclization of Enynes with
Hydrogen Peroxide as the Oxidant. Angew. Chem. 2008, 120, 5522-5525.
[56] Cornell, C. N.; Sigman, M. S. Discovery and Mechanistic Insight into a Ligand-
Modulated Palladium-Catalyzed Wacker Oxidation of Styrenes using TBHP. J.
Am. Chem. Soc. 2005, 127, 2796-2797.
[57] Amatore, C.; Jutand, A.; Meyer, G.; Carelli, I.; Chiarotto, I. Reversible Formation
+ of a Cationic Palladium(II) Hydride [HPd(PPh3)2] in the Oxidative Addition of
Acetic or Formic Acid to Palladium(0) in DMF. Eur. J. Inorg. Chem. 2000, 1855-
1859.
[58] Cámpora, J.; Palma, P.; del Río, D.; López, J. A.; Alvarez, E.; Connelly, N. G.
Redox Behavior of an Organometallic Palladium(II)/Palladium(IV) System. A
24 New Method for the Synthesis of Cationic Palladium(IV) Complexes.
Organometallics 2005, 24, 3624-3628.
[59] Amatore, C.; Cammoun, C.; Jutand, A. Palladium/Benzoquinone-Catalyzed
Electrochemical Oxidation of Alcohols Under Anaerobic Conditions. Synlett
2007, 14, 2173-2178.
[60] Dick, A. R.; Sanford, M. S. Transition Metal Catalyzed Oxidative
Functionalization of Carbon-Hydrogen Bonds. Tetrahedron 2006, 62, 2439-2463.
[61] Gligorich, K. M.; Sigman, M. S. Recent advancements and challenges of
palladium(II)-catalyzed oxidation reactions with molecular oxygen as the sole
oxidant. Chem. Commun. 2009, 3854-3867.
[62] Hagen, J. Industrial Catalysis. In Industrial Catalysis; Wiley, VCH: Weinheim,
1999, Chapter 3: Homogeneously Catalyzed Industrial Processes.
[63] Keith, J. A.; Nielsen, R. J.; Oxgaard, J.; Goddard III, W. A.; Henry, P. M.
Commont on "Mechanism and kinetics of the Wacker Process: A Quantum
Mechanical Approach". Organometallics 2009, 28, 1618-1619.
[64] Keith, J. A.; Henry, P. M. The Mechanism of the Wacker Reaction: A Tale of
Two Hydroxypalladations. Angew. Chem., Int. Ed. 2009, 48, 9038-9049.
[65] ten Brink, G.-J.; Arends, I. W. C. E.; Papdogianakis, G.; Sheldon, R. A. Catalytic
Conversion in Water. Part 10. Aerobic Oxidation of Terminal Olefins to Methyl
Ketones Catalysed by Water Soluble Palladium Complexes. Chem. Commun.
1998, 2359-2360.
[66] Cornell, C. N.; Sigman, M. S. Discovery of a Practical Direct O2-Coupled Wacker
Oxidation with Pd[(-)-sparteine]Cl2. Org. Lett. 2006, 8, 4117-4120.
25 [67] Cornell, C. N.; Sigman, M. S. Recent Progress in Wacker Oxidations: Moving
toward Molecular Oxygen as the Sole Oxidant. Inorg. Chem. 2007, 46, 1903-
1909.
[68] Lucet, D.; Le Gall, T.; Mioskowski, C. The Chemistry of Vicinal Diamines.
Angew. Chem., Int. Ed. 1998, 37, 2580-2627.
[69] de Figueriredo, R. M. Transition-Metal-Catalyzed Diamination of Olefins.
Angew. Chem., Int. Ed. 2009, 38.
[70] Bäckvall, J. E. Stereospecific palladium-promoted vicinal diamination of olefins.
Tetrahedron Lett. 1978, 19, 163.
[71] Steuff, J.; Hõvelmann, C. H.; Nieger, M.; Muñiz, K. Palladium(II)-Catalyzed
Intramolecular Diamination of Unfunctionalized Alkenes. J. Am. Chem. Soc.
2005, 127, 14586-14587.
[72] Bar, G. L. J.; Lloyd-Jones, G. C.; Booker-Milburn, K. I. Pd(II)-Catalyzed
Intermolecular 1,2-Diamination of Conjugated Dienes. J. Am. Chem. Soc. 2005,
127, 7308-7309.
[73] Trost, B. M.; Van Vranken, D. L. Asymmetric Transition Metal-Catalyzed Allylic
Alkylations. Chem. Rev. 1996, 96, 395-422.
[74] Xu, L.; Du, H.; Shi, Y. Diamination of Conjugated Dienes and Trienes Catalyzed
by N-Heterocyclic Carbene Pd(0) Complexes. J. Org. Chem. 2007, 72, 7038-
7041.
[75] Shilov, A. E.; Shul'pin, G. B. Activation and Catalytic Reactions of Saturated
Hydrocarbons in the Presence of Metal Complexes; Kluwer Academic Publishers:
Boston, 2000.
26 [76] Yoneyama, T.; Crabtree, R. H. Pd(II) catalyzed acetoxylation of areanes with
iodosyl acetate. J. Mol. Catal. A: Chem. 1996, 108, 35-40.
[77] Kakiuchi, F.; Murai, S. Catalytic C-H/Olefin Coupling. Acc. Chem. Res. 2002, 35,
826-834.
[78] Dick, A. R.; Kampf, J. W.; Sanford, M. S. Unusually Stable Palladium(IV)
Complexes: Detailed Mechanistic Investigation of C-O Bond-Forming Reductive
Elimination. J. Am. Chem. Soc. 2005, 127, 12790-12791.
[79] Palladium and platinum are the principle catalysts used in automobile catalytic
converters for both hydrocarbon and carbon monoxide oxidation.
[80] Environmetal Catalysis; Gerhard, E.; Knözinger, H.; Weitkamp, J., Eds.; Wiley-
VCH: Weinheim, 1999.
[81] Kullberg, M. L.; Kubiak, C. P. Homogeneous Catalysis of the Water Gas Shift by
bis-(dimethylphosphino) Methane Bridged di-Palladium Complexes: Mechanistic
Studies. C1 Mo. Chem. 1984, 1, 171-177.
[82] Zudin, V. N.; Likholobov, V. A. Catalytic Synthesis of Hydrogen-peroxide from
Oxygen and Water in the Presense of Carbon-Monoxide and Phosphine
Complexes of Palladium. Kinet. Catal. 1979, 20, 1559.
[83] Bianchi, D.; Bortolo, R.; D'Aloisio, R.; Ricci, M. Biphasic Synthesis of Hydrogen
Peroxide from Carbon Monoxide, Water, and Oxygen Catalyzed by Palladium
Complexes with Bidentate Nitrogen Ligands. Angew. Chem., Int. Ed. 1999, 38,
706-708.
27 [84] Bianchi, D.; Bortolo, R.; D'Aloisio, R.; Ricci, M. A novel palladium catalyst for
the synthesis of hydrogen peroxide from carbon monoxide, water, and oxygen. J.
Mol. Catal. A: Chem. 1999, 150, 87-94.
[85] Bortolo, R.; Bianchi, D.; D'Aloisio, R.; Querci, C.; Ricci, M. Production of
hydrogen peroxide from oxygen and alcohols, catalyzed by palladium complexes.
J. Mol. Catal. A: Chem. 2000, 153, 25-29.
[86] Tam, W. Carbonylation of beta-aminoethanols, diols, and diol amines. J. Org.
Chem. 1986, 51, 2977-2981.
[87] Kiss, G. Palladium-Catalyzed Reppe Carbonylation. Chem. Rev. 2001, 101, 3435-
3456.
[88] Gabriele, B.; Salerno, G.; Costa, M. Oxidative Carbonylations. Top. Organomet.
Chem. 2006, 18, 239-272.
[89] Gabriele, B.; Mancuso, R.; Salerno, G.; Ruffolo, G.; Costa, M.; Dibenedetto, A. A
Novel and Efficient Method for the Catalytic Direct Oxidative Carbonylation of
1,2- and 1,3-Diols to 5-Membered and 6-Membered Cyclic Carbonates.
Tetrahedron Lett. 2009, 50, 7330-7332.
[90] Berzelius, J. J. Versuche über die mit dem Platin vorkommenden Metalle, und
über das Verfahren zur Zerlegung der natürlichen Platinlegirungen oder
Platinerze. Ann. 1828, 13, 435.
[91] Blackburn, T. F.; Schwartz, J. Homogeneous Catalytic Oxidation of Secondary
Alcohols to Ketones by Molecular Oxygen under Mild Conditions. J. Chem. Soc.,
Chem. Commun. 1977, 157-158.
28 [92] Starchevsky, M. K.; Hladiy, S. L.; Pazdersky, Y. A.; Vargaftik, M. N.; Moiseev,
I. I. Giant Pd-561 clusters: onset to new catalytic properties. J. Mol. Catal. A:
Chem. 1999, 146, 229-236.
[93] Vargaftik, M. N.; Zagorodnikov, V. P.; Stolarov, I. P.; Moiseev, I. I.; Kochubey,
D. I.; Likholobov, V. A.; Chuvilin, A. L.; Zamaraev, K. I. Giant palladium
clusters as catalysts of oxidative reactions of olefins and alcohols. J. Mol. Catal.
1989, 53, 315-348.
[94] Peterson, K. P.; Larock, R. C. Palladium-Catalyzed Oxidation of Primary and
Secondary Allylic and Benzylic Alcohols. J. Org. Chem. 1998, 63, 3185-3189.
[95] Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. Pd(OAc)2-Catalyzed Oxidation of
Alcohols to Aldehydes and Ketones by Molecular Oxygen. Tetrahedron Lett.
1998, 39, 6011-6014.
[96] ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Green, Catalytic Oxidation of
Alcohols in Water. Science 2000, 287, 1636-1639.
[97] Paavola, S.; Zetterberg, K.; Privalov, T.; Csöregh, I.; Moberg, C. Aerobic
Oxidation of 1-Phenylethanol Catalyzed by Palladaheterocycles. Adv. Synth.
Catal. 2004, 34, 237-244.
[98] Ferreira, E. M.; Stoltz, B. M. The Palladium-Catalyzed Oxidative Kinetic
Resolution of Secondary Alcohols with Molecular Oxygen. J. Am. Chem. Soc.
2001, 123, 7725-7726.
[99] Jensen, D. R.; Pugsley, J. S.; Sigman, M. S. Palladium-Catalyzed Enantioselective
Oxidation of Alcohols Using Molecular Oxygen. J. Am. Chem. Soc. 2001, 123,
7475-7476.
29 [100] Jensen, D. R.; Schultz, M. J.; Mueller, J. A.; Sigman, M. S. A Well-Defined
Complex for Palladium-Catalyzed Aerobic Oxidation of Alcohols: Design,
Synthesis, and Mechanistic Considerations. Angew. Chem., Int. Ed. 2003, 42,
3710-3813.
[101] Sigman, M. S.; Jensen, D. R. Ligand-Modulated Palladium-Catalyzed Aerobic
Alcohol Oxidations. Acc. Chem. Res. 2006, 39, 221-229.
[102] Steinhoff, B. A.; Fix, S. R.; Stahl, S. S. Mechanistic Study of Alcohol Oxidation
by the Pd(OAc)2/O2/DMSO Catalyst System and Implications for the
Development of Improved Aerobic Oxidation Catalysts. J. Am. Chem. Soc. 2002,
124, 766-767.
[103] Steinhoff, B. A.; Stahl, S. S. Mechanism of Pd(OAc)2/DMSO-Catalyzed Aerobic
Alcohol Oxidation: Mass-Transfer-Limitation Effects and Catalyst
Decomposition Pathways. J. Am. Chem. Soc. 2006, 128, 4348-4355.
[104] Steinhoff, B. A.; Stahl, S. S. Ligand-Modulated Palladium Oxidation Catalysis:
Mechanistic Insights into Aerobic Alcohol Oxidation with the Pd(OAc)2/Pyridine
Catalyst System. Org. Lett. 2002, 4, 4179-4181.
[105] ten Brink, G.-J.; Arends, I. W. C. E.; Hoogenraad, M.; Verspui, G.; Sheldon, R.
A. Catalytic conversions in water. Part 23: Steric effects and increased substrate
scope in the palladium-neocuproine catalyzed aerobic oxidation of alcohols in
aqueous solvents. Adv. Synth. Catal. 2003, 345, 1341-1352.
30 Chapter 2
Aerobic Alcohol Oxidation with Cationic Palladium Complexes: Insights into Catalyst Design and Decomposition
2.0 Preface
This chapter describes previously published research: Conley, N. R.; Labios, L. A.; Pearson, D. M.; McCrory, C. C. L.; Waymouth, R. M. “Aerobic alcohol oxidation with cationic palladium complxes: insights into catalyst design and decomposition.” Organometallics 2007, 26, 5447-5453, newer research into the use of other oxidants, and initial results which expand the substrate scope tolerated by this catalyst. My entry into this project began shortly after N. R. Conley had discovered the active catalyst 1, but before it was positively assigned. In close collaboration with N. R. Conley, we studied the system and proposed the current mechanism. L. A. Labios provided definitive evidence for the decomposition pathway through the crystallization of the oxidized product from H2O2 mixtures as well as the hydroxyl bridge dimer. C. C. L. McCrory and myself probed the solution concentration of H2O2 in solution electrochemically and verified the catalyst’s competency for H2O2 disproportionation. The exploration of new oxidants was carried out by N. R. Conley and I. I carried out initial screening of methanol and glycerol. R. M. Painter, S. Banik, and I have further explored the oxidation of glycerol and other polyols which is briefly discussed in Section 2.2.6.
2.1 Introduction
The selective, catalytic oxidation of organic molecules remains one of the central challenges in chemical synthesis.1 Air is an attractive terminal oxidant as it is abundant and, if reduced completely to water, generates environmentally benign coproducts. 31 Nevertheless, the high overpotential of dioxygen for most oxidative transformations of interest, coupled with its kinetic stability and tendency to generate highly reactive partially reduced oxidizing species (PROS) once activated2 are the major chemical challenges in taming this readily available oxidant for selective oxidation reactions. It is instructive to consider that aerobic organisms have not completely solved this problem in that a significant component of physical aging is the inability of aerobic organisms to mitigate deleterious free-radical oxidation reactions. To quote accomplished biochemist
Bruce Ames, “We’re all going rancid.”3
In an effort to develop highly active oxidation catalysts for potential use in low temperature fuel cells, we sought active catalysts systems that would mediate oxidative transformations, preferably at ambient temperatures. To this end, we were attracted to recent work which suggested that appropriately ligated Pd(II) complexes could mediate selective oxidative transformations with air as a terminal oxidant.4,5 In particular, the palladium-catalyzed aerobic oxidation of alcohols to aldehydes and ketones has been the subject of several excellent reviews,4,6,7 and its mechanism has been studied extensively.8-
25 Sigman and coworkers reported that the Pd N-heterocyclic carbene complex
26 (IiPr)Pd(OPiv)2 catalyzes the oxidation of alcohols at room temperature in air. Since the first report of ligand-accelerated catalysis using Pd(OAc)2 and pyridine by Uemura and coworkers,27 a variety of ligands have been employed, including triethylamine,20,26,28 substituted phenanthrolines,11,12,29-31 N-heterocyclic carbenes,16,26,32,33 and sterically encumbered pyridines.34,35 While much attention has been given to the effect of ligand modulation on catalyst activity, there are only a few examples of effective catalyst
11,26,32,36,37 21 systems that use palladium salts other than Pd(OAc)2. Sigman and Jensen
32 attribute the success of acetate to its dual role as an anionic ligand and as a base for intramolecular deprotonation of the palladium-bound alcohol species. However, the use of non-coordinating ions to make cationic palladium(II) complexes with an open coordination site has also been shown to dramatically increase the rate of aerobic oxidations.38 We reasoned that a palladium(II) complex with both a basic, coordinating acetate ion and a non-coordinating triflate counterion could yield a more active alcohol oxidation catalysts.
Herein, we report that cationic neocuproine palladium acetate complexes exhibit very fast initial rates of alcohol oxidation at room temperature in air, but undergo competitive ligand oxidation from hydroperoxide intermediates generated from the partial reduction of O2. Mechanistic studies reveal that the presence of an open coordination site and an internal acetate base are key features that enable the rapid dehydrogenation of alcohols.
2.2 Results and Discussion
2.2.1 Synthesis and characterization of cationic palladium - neocuproine complexes
On the basis of Sheldon's report that substituted (neocuproine)Pd(OAc)2 complexes were active alcohol oxidation catalysts under 30 bar air at 80 °C,30,31 we sought a synthetic route to cationic (neocuproine)Pd(OAc)(OTf) complexes. We found
30 that conproportionation of (neocuproine)Pd(OAc)2 (2) and the ditriflate analogue
39 (neocuproine)Pd(NCCH3)2(OTf)2 (3) (prepared by treating 2 with triflic acid) in
33 acetonitrile afforded the dimeric acetate-bridged compound 1, as shown in Scheme 2.1.
1 Investigations by H NMR spectroscopy in acetonitrile-d3 as a function of concentration reveal that this complex speciates as a dimer and monomer in fast equilibrium (Scheme
2.2) with a dissociation constant K =
2 [(neocuproine)Pd(OAc)(L)] /[((neocuproine)Pd(OAc))2(OTf)2] = 5 mM at 295 K. The nature of the ligand L in the monomeric structure (L = OTf or CH3CN) could not be unambiguously established. Thermodynamic parameters for this equilibrium were obtained by variable temperature 1H NMR spectra from 295 K to 345 K (!H° = 32.1 kJ•mol-1, !S° = 65.1 J•mol-1•K-1). The solid-state structure of 1 (Figure 2.1) reveals a stacked dimeric structure spanned by bridging acetates, analogous to the recently reported cationic palladium bis-carbene dimer40 and the neutral palladacycle dimers.41,42
2 2
2 ! OTf 2 ! OTf N CH3CN N N N + N N N N Pd Pd ppt. w/ ether PdPd O AcO OAc N N O O O
2 3 1 Scheme 2.1. Preparation of Complex 1
Addition of water to an acetonitrile-d3 solution of 1 generates an equilibrium mixture of [(neocuproine)Pd(µ-OH)]2(OTf)2 (4), acetic acid, and 1, as shown in Scheme
11,43 2.3. Similar equilibria have been reported for other L2PdX2 complexes. The
µ-hydroxo Pd dimer 4 is only sparingly soluble in acetonitrile and precipitates from solution. The X-ray crystal structure of 4, shown in Figure 2.2, indicates that 4 is isostructural with the previously reported platinum analogue44 and adopts a butterfly
34 structure to minimize non-bonded interactions between the methyl groups of the neocuproine ligands. The structure reveals a hydrogen bond between bridging OH groups and the triflate counterions. Addition of an excess of acetic acid to an isolated sample of 4 in acetonitrile resulted in its clean conversion back to 1. As the µ-hydroxo
Pd dimer 4 is analogous to the resting state of the catalyst system reported by Sheldon et al. for the aerobic oxidation of alcohols,11,30,31,45 it was also examined as a catalyst precursor for aerobic alcohol oxidation.
2 n
2 ! OTf CH3CN N N N N N N PdPd Pd O L O O O O O
- L = CH3CN, n = 1, OTf L = OTf, n = 0 Scheme 2.2. Monomer-Dimer Equilibrium of Cationic Palladium Complex 1
35
Figure 2.1. Solid state structure of 1 with ellipsoids drawn at the 50% probability level. Hydrogen atoms and triflate anions are omitted for clarity. Selected bond lengths (Å) and angles (deg): Pd1-Pd2, 2.9582(4); Pd1-O1, 2.016(2); Pd1-O2, 2.015(2); Pd1-N2, 2.024(2); Pd1-N1, 2.019(3); Pd2-O4, 2.026(2); Pd2-O3, 2.0105(19); Pd2-N4, 2.056(2); Pd2-N3, 2.038(2); O4-C37, 1.272(3); O1-C37, 1.256(4); O3-C45, 1.271(3); O2-C45, 1.262(3); O1-Pd1-O2, 84.17(9); O4-Pd2-O3, 81.85(8); O4-Pd2-N3, 173.46(9); O3-Pd2- N4, 176.69(9); N3-Pd2-N4, 82.77(9).
2 2
2 ! OTf CH CN, H O 2 ! OTf N 3 2 N N N N N + 2 CH3CO2H PdPd Pd O HO 2 O O O
1 4
Scheme 2.3. Generation of µ-OH dimer 4
36
Figure 2.2. Solid-state structure of 4 on a crystallographic two-fold axis with ellipsoids drawn at the 50% probability level. Hydrogen atoms, except those corresponding to the µ-OH groups, and one triflate anion are omitted for clarity. Selected bond lengths (Å) and angles (deg): Pd1-N1, 2.038(8); Pd1-N2, 2.047(9); Pd1-O4, 2.014(9); Pd1-O4_2, 2.006(9); O4-Pd1-O4_2, 79.3(3); N1-Pd1-N2, 82.1(3); Pd1-O4_2-Pd1_2, 95.8(3); C14- C13-N2-Pd1, 14.4(3); C1-C2-N1-Pd1, 12.5(5). Nonbonded distances (Å): Pd1-Pd1_2, 2.982(2); O4-O4_2, 2.564(9); C(methyl)-Pd1, 3.44(8) (average); C(methyl)-(µ-O), 2.89(3) (average); O3-O4, 2.74(9).
2.2.2 Catalyst screening in the aerobic oxidation of 2-heptanol
Complexes 1-4 were investigated for the catalytic oxidation of 2-heptanol to 2- heptanone in acetonitrile46 at room temperature under an ambient pressure of air. The reaction progress for each complex is plotted in Figure 2.3. Under these mild conditions,
-1 -1 complex 1 (3 mol %) exhibits a fast initial TOF (TOFi = 78 Pd atom •hr ) for the oxidation of 2-heptanol, but slows rapidly to afford a 36% yield of 2-heptanone after 24 h, which corresponds to a turnover number (TON) of 12 Pd atom-1. The initial rates are the same in the presence of either air or dioxygen, implying that reoxidation of palladium is not rate limiting.
37
Figure 2.3. Reaction progress of air oxidation (1 atm) of 2-heptanol (0.5 M) with complexes 1 (!), 2 ("), 3 (#), and 4 (!) (3 mol % Pd) in acetonitrile (acetonitrile/dichloromethane, 1:1 for 2; acetonitrile/dimethylsulfoxide, 1:1 for 4) at room temperature.
To assess the role of charge and ligand substitution on catalytic alcohol oxidation, complexes 2, 3, and 4 were investigated for alcohol oxidation under similar conditions
(Figure 2.3). The µ-hydroxo Pd dimer 4 showed similar behavior to that of the mixed
-1 -1 acetate/triflate Pd dimer 1, but exhibited lower initial rates (TOFi = 2.0 Pd atom •hr )
47 and yield. In contrast, under these conditions, both the (neocuproine)Pd(OAc)2 2 and
- the (neocuproine)Pd(NCCH3)2(OTf)2 3 were ineffective catalysts (TOFi = 0.24 Pd atom
1•hr-1 and 0.16 Pd atom-1•hr-1, respectively).
The high initial rates for the aerobic oxidation of 2-heptanol by the palladium acetate triflate dimer 1 under ambient air are remarkable. Of the few reported systems for alcohol oxidation with ambient air,26,29,34,41,48,49 only the N-heterocyclic carbene system 38 described by Sigman and coworkers26 proceeds at room temperature. The high activity of this catalyst under such mild conditions50 was proposed to be due to the low coordination number of the (IMes)Pd(OAc)2 that enables facile binding of alcohol, and deprotonation of the bound alcohol by the internal acetate base21 to yield the Pd alkoxide intermediate.26,33 We propose that the presence of both triflate and acetate counterions in complex 1 allows for fast alcohol binding and fast deprotonation of the palladium-bound alcohol species, followed by !-hydride elimination.51 The slow rate observed with the diacetate 2 suggests that an open coordination site for the alcohol is a key feature leading to fast oxidation with 1.52 The slow rate observed with the ditriflate 3 implies that the presence of a suitable base to deprotonate the palladium-bound alcohol is also important.14 We attribute the slower rate of 4 (relative to 1) to the slow dissociation of the dimer53 in the absence of acetic acid. These data suggest that a weakly coordinating ligand and an intramolecular base are necessary to achieve high TOFs with the neocuproine/palladium(II) system under mild conditions.
Despite the fast initial rates of alcohol oxidation with complexes 1 and 4, the rates decrease rapidly with conversion, implicating that these catalysts are deactivated, even under these mild reaction conditions. The time course of the reaction reveals that the rate of alcohol oxidation decreases linearly with percent yield of 2-heptanone. Palladium black was not observed in the catalysis with 1 during the time in which the reaction was monitored, and therefore, cannot account for this loss of activity. Furthermore, control experiments indicated that neither water, a byproduct of the oxidation,54 nor 2-heptanone inhibited the catalyst. The addition of molecular sieves23 leads to lower initial rates and conversion. While hydrogen peroxide, a possible byproduct of the oxidation,31 was
39 found to inactivate the catalyst, it could not be detected electrochemically in the reaction mixture during the course of the reaction,55 and the addition of a variety of hydrogen peroxide disproportionation catalysts into the reaction mixture did not prolong the life of the catalyst 1.56 These observations are inconsistent with reversible catalyst inhibition, and implicate a competitive process involving catalyst decomposition to a soluble, inactive palladium complex.
2.2.3 Identification of catalyst decomposition products
To identify the fate of catalyst 1, the reaction mixture was precipitated with diethyl ether after 24 h of reaction. Analysis of the palladium-containing products revealed the presence of the cationic palladium carboxylate 5 (see Figure 2.4), indicating that one of the methyl groups of the 2,9-dimethyl phenanthroline ligands is oxidized to the carboxylate. This complex was inactive as catalyst for alcohol oxidation. Complex
5 was also formed in the aerobic oxidation of 2-heptanol with µ-hydroxo Pd dimer 4.
The solid-state structure of 5 reveals a square-planar coordination geometry with an intramolecularly bound carboxylate derived from one of the ligand methyl groups and a coordinated acetonitrile ligand.
The formation of the inactive complex 5 during the course of the reaction implies that either oxygen, or a PROS, degrades the ligand in competition with alcohol oxidation.
Given that the dimer 1 is stable to air in the absence of alcohol, we suspected that ligand oxidation was mediated by hydroperoxide intermediates generated during the course of the catalytic cycle. To gain insight into the pathway through which 5 was formed, an acetonitrile solution of dimer 1 was treated with aqueous hydrogen peroxide. Three
40
Figure 2.4. Solid-state structure of 5 with ellipsoids drawn at the 50% probability level. Only one of the three identical cations composing the asymmetric unit is shown, with hydrogen atoms and triflate anions omitted for clarity. Selected bond lengths (Å) and angles (deg): Pd1-N1, 1.917(3); Pd1-N2, 2.095(2); Pd1-N3, 2.021(3); Pd1-O1, 2.020(2); C1-O1, 1.308(4); C1-O2, 1.207(4); N2-Pd1-N1, 81.24(10); N1-Pd1-O1, 80.22(10). products were identified from this reaction: the µ-hydroxo palladium dimer 4, alkoxide 6, and carboxylate 5 (Scheme 2.4). The formation of the µ-hydroxo dimer 4 is likely due to the presence of water from the aqueous hydrogen peroxide. The formation of the alkoxide 6 and the carboxylate 5 are proposed to derive from the hydroperoxide intermediate 7. The reaction of palladium(II) complexes with hydrogen peroxide is known to give palladium hydroperoxides.57 An isolated sample of the alkoxide 6 was converted quantitatively to 5 by stirring in acetonitrile in the presence of air, with no requirement for hydrogen peroxide. In the absence of air, a yellow dimethylsulfoxide-d6 solution of alkoxide 6 decomposes to give a deep red solution, the 1H NMR spectrum of which exhibited an aldehyde peak at 11.0 ppm. These data suggest that in the absence of air, alkoxide 6 undergoes "-hydride elimination to generate the ligand-derived aldehyde.
We tentatively assign the palladium product as a neocuproine bound Pd(0) 8, as we could not observe a characteristic palladium hydride chemical shift around -18 ppm in the 1H
41 NMR spectrum,25 and a similar red color is observed when tris(di- benzylideneacetone)dipalladium(0) was complexed to neocuproine in dichloromethane.
The formation of 5 both in aerobic alcohol oxidation with 1, and in treatment of 1 with hydrogen peroxide in acetonitrile, implicates the intermediacy of hydroperoxide 7 in the aerobic oxidation of alcohols mediated by 1.
Scheme 2.4. Reaction of Catalyst 1 with Hydrogen Peroxide in Acetonitrile
2.2.4 Proposal of the catalytic cycle for 1 through mechanistic interpretation
On the basis of our mechanistic studies and other studies of aerobic alcohol oxidation, 8-25,27 we propose the mechanism illustrated in Scheme 2.5 (solid line arrows) for the oxidation of 2-heptanol with the dimeric palladium complex 1. We propose that the alcohol binds to the monomeric form of 1 followed by intramolecular deprotonation to generate a cationic palladium alkoxide (neocuproine)PdOR. "-Hydrogen elimination
42 generates the ketone and cationic (neocuproine)Pd-H 8, which is subsequently oxidized to the hydroperoxide 7.25,58,59
Scheme 2.5. Proposed Mechanism for Aerobic Alcohol Oxidation with Catalyst 1
Several mechanisms have been proposed for the reoxidation of reduced palladium species, such as 8, in aerobic alcohol oxidation reactions.4 Uemura and coworkers proposed that insertion of molecular oxygen into a Pd(II) hydride yields Pd(II) hydroperoxides directly (Scheme 2.5, pathway A),60 a transformation which was later shown to be feasible by Stahl and Goldberg.25,58,59 However, a Pd(0)/Pd(II) pathway may also be operative based on Stahl and coworkers’ finding that molecular oxygen reacts
43 readily with a (bathocuproine)Pd(0) complex to give a Pd(II) peroxo complex that can be protonated by two equivalents of acetic acid to give hydrogen peroxide and
61 (bathocuproine)Pd(OAc)2. Although not reported, addition of a single equivalent of acetic acid to the Pd(II) peroxo complex should yield palladium hydroperoxide 7
(Scheme 2.5, pathway B). Finally, Sheldon and coworkers have proposed a bimolecular pathway that involves reaction of a Pd(II) peroxo complex with a Pd(0) complex and two equivalents of Bronsted acid to give µ-hydroxo Pd(II) dimer 4.29
We favor a variation of the bimolecular pathway whereby the hydroperoxide 7, formed through pathway B, conproportionates with either the cationic (neocuproine)Pd-H
8 or (neocuproine)Pd(0) 9 / H+ to generate the µ-hydroxo Pd dimer 4, which reacts either with acetic acid to generate 1 (pathway C) or directly with the alcohol to regenerate the cationic (neocuproine)PdOR (pathway D). We propose that the rapid drop-off in rate and low turnover numbers observed is due to competitive reactions of hydroperoxide intermediate 7. If intramolecular ligand oxidation of 7 is competitive with regeneration of the active catalyst species, then after several cycles all the palladium would speciate to the inactive palladium carboxylate 5. Evidence for the mechanism shown in Scheme 5 includes the following: (a) no hydrogen peroxide can be detected in the reaction mixture,55 (b) µ-hydroxo Pd dimer 4 is observed in the aerobic alcohol oxidation reaction mixture,62 (c) µ-hydroxo Pd dimer 4 can be cleanly converted to 1 by action of acetic acid, and (d) a deep red-colored species that does not exhibit the characteristic palladium hydride 1H chemical shift25 accumulates in the aerobic alcohol oxidation reaction mixture under oxygen-starved conditions.
44 These observations illustrate one of the potential liabilities of air as a terminal oxidant. Air is certainly a convenient terminal oxidant, but its large overpotential and tendency to generate highly oxidizing species (such as 7) upon partial reduction illustrates the liabilities of utilizing more powerful terminal oxidants than necessary to carry out oxidative transformations of interest.63 Nevertheless, the fast initial rates observed suggest that catalysts with readily accessible coordination sites and coordinated internal bases are promising leads for further catalyst development.
2.2.5 The use of alternative oxidants in alcohol oxidation
The catalyst degradation observed during the use of complex 1 under aerobic conditions (see Section 2.2.3), lead us to explore the use of other oxidants (Figure 2.5).
Electron deficient conjugated alkenes have previously been reported to undergo reduction in the presence of Pd-hydrides.64-66 Similarly, Muniz and coworkers reported the reduction of azo-complexes like diethyl azodicarboxylate by [(neocuproine)Pd(dba)] in the prescence of acetic acid.67 Our attempts to use hydride acceptors like cyclopentenone, ethyl cinnamate, dibenzylideneacetone (dba), or diethyl azodicarboxylate all resulted in the consumption of only stoichiometric (relative to catalyst) quantities of the alcohol. At the same time, no palladium black was observed, suggesting that while initial insertion of the olefin or azo-compound into the palladium hydride was facile, protonation of the subsequent palladium-alkyl or palladium-amide to regenerate the active catalyst was not under these conditions. The use of sacrificial aldehydes, commonly employed to generate metal-oxo complexes from O2, as a
45
Figure 2.5 Oxidants screened for alcohol oxidation with [(neocuproine)Pd(OAc)]2[OTf]2 scavenger for reactive PROS resulted in a slower rate of alcohol oxidation but similar turnover numbers to those in the observed in the absence of the aldehyde.
Benzoquinone has commonly been employed as a two electron oxidant for
Pd(II)/Pd(0) cycles (See Section 1.2.3). Initial screenings with benzoquinone and 1 in the oxidation of 2-propanol showed clean first order behavior in alcohol conversion to acetone, suggesting that catalyst decomposition was not significant using benzoquinone.
Rates of oxidation with benzoquinone, however, were slower than those observed with oxygen.
Oxygen mediators such as Fe(II)-phthalocyanine68 or oxidants like manganese dioxide69 have been previously reported to promote oxidation of hydroquinone to 46 benzoquinone Conceivably these systems could be employed to oxidize alcohols using only catalytic loadings of benzoquinone. Our attempts to use these and other oxygen mediators such as Cu(II) bis-(1,10-phenanthroline) ditriflate70 were unsuccessful. In the case of the benzoquinone and Fe or Mn mediators, little improvement was seen with their use compared to reactions conducted in their absence with air, suggesting that oxygen is still a competitive coordinator of Pd(0) in the presence of benzoquinone, and that the degradative oxidative pathways previously observed still occurs under these conditions.
Cu(II) salts generally were not compatible with our system. We attribute this incompatibility to ligand exchange reactions between Cu and Pd.
2.2.6 Substrate scope of alcohol oxidation
Methanol
Methanol may serve as an intermediate energy carrier in the rapidly changing energy landscape.71 Methanol is particularly attractive in the development of fuel cells due to its high energy density (relative to hydrogen) and relative ease of transportation.
Our studies on methanol oxidation are covered in Chapter 3.
Glycerol
Glycerol is an attractive and versatile feedstock as it is nontoxic, edible, and biodegradable. Novel and selective transformations must be developed to effectively use glycerol as a building block for value-added chemicals and to allow for the development of new applications for glycerol-derived products.72-76
47 Initial screenings for glycerol oxidation using 1 (5 mol % Pd) and benzoquinone in a mixture of 7:1 CD3CN/D2O yielded 1,3-dihydroxyacetone over 2 h in high selectivity as observed by 1H-NMR. Substitution of the terminal oxidant benzoquinone for air resulted in a slower reaction, which reached 47% conversion after 24 h with 80 % selectivity for dihydroxyacetone. Analysis of the ligand resonances of the catalyst suggest that ligand oxidation to 5 had occurred, limiting the lifetime of aerobic oxidation.
Almost total conversion of glycerol could be acheived by increasing the catalyst loading to 10 mol % Pd, as observed by 1H-NMR.
Table 2.1 Selective oxidation of glycerol to 1,3-dihydroxyacetone using 1a Entry mol % Pd oxidant (M) alcohol, time conversion, mmol % 1 5 BQ (0.425) 0.1 2 h 98 2 5 air 0.1 24 h 47 3 10 air 1 3 h >90 a reactions were performed in a 7:1 mixture of CD3CN/D2O
These preliminary results suggest a promising future for palladium catalysis in the selective oxidation of glycerol and other polyols. Further experimental results can be found in Ron Painter’s thesis or in our recently submitted publication to Angewandte
Chemie International Edition; Painter, R. M.; Pearson, D. M.; Waymouth, R. M.
“Selective catalytic oxidation of glycerol to dihydroxyacetone” 2010 In Press.
Diols and Polyols
The high selectivity for the 2° alcohol in glycerol despite the presence of two 1° alcohols, has prompted investigations into the factors affecting the selectivity of 1 toward other 1,2-diols and polyols and contrasting these observations with those found in mono- ol substrates like iso- or n-propanol. Initial screens by R. M. Painter and S. M, Banik in 48 our lab using the benzoquinone conditions developed previously for methanol and glycerol oxidation have revealed a strong preference for the oxidation of the 2° alcohol in diols. Significantly, over oxidation is not observed in many cases. Additional mechanistic experiments have been carried out by Painter, Banik and myself and will be reported shortly as part of a full paper.
2.3 Experimental Section
2.3.1 Synthesis of compounds
[(2,9-Dimethyl-1,10-phenanthroline)Pd(µ-OAc)]2(OTf)2 (1). To a 25 ml RB flask with stirbar was added 2 (0.0400 g, 0.0924 mmol), 3 (0.0642 g, 0.0924 mmol), and acetonitrile (10.0 ml). The resulting mixture was stirred until all solids dissolved and then precipitated with diethyl ether to give an orange solid. This solid was isolated by centrifugation, washed with diethyl ether, and dried under vacuum to give 1 as an orange
1 solid (0.0589 g, 0.0563 mmol, 61% yield). H NMR (300 MHz, CD3CN, satd. soln. to favor dimer): # 2.22 (s, 6H), 2.59 (s, 12H), 7.36 (d, 4H, J = 8.4 Hz), 7.66 (s, 4H), 8.24 (d,
13 4H, J = 8.4 Hz); C NMR (75 MHz, CD3CN, satd. soln. to favor dimer): # 23.77, 24.73,
- 128.02, 128.99, 129.10, 141.29, 146.92, 166.68, 188.56, CF3SO3 not observed; ESI-MS:
- + m/z (relative intensity) [ion] 374 (75%) [0.5M – CF3SO3 + H] , 314 (47%) [0.5M –
- - + - - 2+ + + CF3SO3 – CH3COO ] , 209 (100%) [0.5M – CF3SO3 – CH3COO – Pd + H ] . Anal.
Calcd for C34H30N4O10S2Pd2F6 (1045.59): C, 39.06; H, 2.89; N, 5.36; Pd, 20.36. Found:
C, 39.14; H, 2.87; N, 5.22; Pd, 20.1.
49 (2,9-Dimethyl-1,10-phenanthroline)Pd(OAc)2 (2). This compound was prepared as previously reported.30
(2,9-Dimethyl-1,10-phenanthroline)Pd(NCCH3)2(OTf)2 (3). To a slurry of 2
(0.221 g, 0.511 mmol) in acetonitrile (1.0 ml) was added a solution of triflic acid in acetonitrile (0.33 M, 3.8 ml, 2.5 eq.). The solution was stirred briefly and then precipitated with diethyl ether to give a yellow solid. This solid was isolated by centrifugation, precipitated two more times from acetonitrile using diethyl ether, and dried under vacuum to give 3 as a light yellow solid (0.090 g). Additional triflic acid
(0.33 M, 1.0 ml) was added to the original supernatant, followed by brief stirring and precipitation with diethyl ether. The resulting yellow solid was subjected to the same workup as described above to give additional 3 (0.021 g). The pure solids were
1 combined (0.111 g, 0.160 mmol, 31% yield). H NMR (300 MHz, CD3CN): # 2.98 (s,
6H), 7.77 (d, 2H, J = 8.4 Hz), 8.07 (s, 2H), 8.68 (d, 4H, J = 8.4 Hz); 1H NMR (400 MHz,
DMSO-d6): # 2.06 (s, 6H, CH3CN), 3.09 (s, 6H), 7.90 (d, 2H, J = 8.4 Hz), 8.17 (s, 2H),
13 8.80 (d, 2H, J = 8.4 Hz); C NMR (125 MHz, DMSO-d6): # 1.19, 24.85, 118.15, 120.70
- (q, J = 320.1 Hz, CF3SO3 ), 126.72, 127.42, 128.52, 140.16, 146.40, 164.65; ESI-MS: m/z
2+ - + + (relative intensity) [ion] 209 (100%) [M – Pd – 2CF3SO3 + H ] , 315 (6%) [M –
- + - 2CF3SO3 + H] , 149 (100%) [CF3SO3 ]. Anal Calcd for C20H18O6N4S2PdF6 (694.92): C,
34.57; H, 2.61; N, 8.06; Pd, 15.3. Found: C, 34.81; H, 2.68; N, 8.77; Pd, 14.6.
Synthesis of (2,9-dimethyl-1,10-phenanthroline)Pd(µ-OH)]2(OTf)2 (4) from 1 and water. A concentrated solution of 1 (100 mg, 0.095 mmol) in acetonitrile (4 ml)
50 was prepared and then the solution was diluted by addition of an equal volume of H2O, causing the orange solution to become yellow and cloudy. The solution was stirred for 1 h and the precipitate was isolated by centrifugation. The pellet was washed with acetonitrile and dried under vacuum, yielding the dimer 4 as a yellow solid (23.0 mg,
1 25% yield). H NMR (300 MHz, DMSO-d6): # 2.13 (s, 2H), 2.87 (s, 12H), 7.86 (d, 4H, J
13 = 8.1 Hz), 8.15 (s, 4H), 8.79 (d, 4H, J = 8.1 Hz); C NMR (500 MHz, DMSO-d6): #
- 23.12, 126.72, 127.92, 128.26, 139.86, 146.10, 165.14, CF3SO3 not observed; ESI-MS:
- + m/z (relative intensity) [ion] 372 (100%) [0.5M – CF3SO3 + CH3CN] , 314 (30%) [0.5M
- - + - - 2+ + + – CF3SO3 – OH] , 209 (50%) [0.5M – CF3SO3 – HO – Pd + H ] . Anal. Calcd for
C30H26N4Pd2O8S2F6 (961.51): C, 37.47; H, 2.73; N, 5.83; Pd, 22.1. Found: C, 37.27; H,
2.61; N, 5.89; Pd, 22.0.
Synthesis of (2,9-dimethyl-1,10-phenanthroline)Pd(µ-OH)]2(OTf)2 (4), (9- methyl-1,10-phenanthroline-2-carboxylate)Pd(OTf) (5), and [(9-methyl-1,10- phenanthrolin-2-yl)methanolate]Pd(OTf) (6) from 1 and 30% hydrogen peroxide.
In a test tube, 9.79 M (~30%) hydrogen peroxide (73.2 $L, 0.717 mmol, 15.0 eq.) was added to a solution of 1 (0.0500 g, 0.0478 mmol) in acetonitrile (2.0 mL) with stirring.
The solution immediately, albeit temporarily, changed color from yellow to deep red and bubbling was observed. After 15 minutes of stirring in air, the solution had become bright yellow in color and contained a light yellow precipitate. This precipitate was isolated from the reaction mixture by centrifugation, washed with acetonitrile, and dried under vacuum to give 4 as a pale yellow solid (0.016 g, 0.017 mmol). The supernatant was then filtered and precipitated with diethyl ether (~2 ml). The resulting yellow solid
51 was isolated by centrifugation, precipitated a second time from acetonitrile, washed with diethyl ether, and dried under vacuum to give 6 as a bright yellow solid (0.010 g, 0.019 mmol). The supernatants from both precipitations of 6 were combined, filtered, and evaporated to give 5 as an orange solid (0.0011 g, 0.0021 mmol).
Alternative synthesis of (9-methyl-1,10-phenanthroline-2- carboxylate)Pd(OTf) (5) by aerobic oxidation of 2-heptanol. 2-heptanol (1.14 mL,
8.04 mmol) was added to a solution of 1 (0.1256 g, 0.1201 mmol) in acetonitrile (8.0 mL). The dark orange reaction mixture was stirred vigorously in air overnight and became light yellow-orange in color. After filtration, a light orange solid was precipitated from the filtrate using diethyl ether. This solid was isolated by centrifugation, precipitated a second time from acetonitrile, washed with diethyl ether, and dried under vacuum to give 5 as a light orange solid (0.027 g, 0.051 mmol, 42% yield).
Characterization of (9-methyl-1,10-phenanthroline-2-carboxylate)
1 Pd(NCCH3)(OTf) (5). H NMR (300 MHz, CD3CN): # 2.87 (s, 3H), 7.91 (d, 1H, J = 8.4
Hz), 8.01 (d, 1H, J = 8.7 Hz), 8.12 (d, 1H, J = 9.0 Hz), 8.21 (d, 1H, J = 8.7 Hz), 8.69 (d,
13 1H, J = 8.7 Hz), 8.89 (d, 1H, J = 8.1 Hz); C NMR (300 MHz, CD3CN): # 27.08,
125.58, 126.89, 130.27, 130.41, 131.23, 132.51, 141.69, 142.29, 145.75, 148.08, 150.02,
- 166.77, 172.16, CF3SO3 not observed; ESI-MS: m/z (relative intensity) [ion] 384 (10%)
- + - + - [M – CF3SO3 ] , 345 (22%) [M – CF3SO3 – CH3CN] , 299 (100%) [M – CF3SO3 –
- + - 2+ + CH3CN – COO ] , 195 (20%) [M – CF3SO3 – CH3CN – CO2 – Pd ] . Anal. Calcd for
52 C17H12N3PdO5SF3 (533.78): C, 38.25; H, 2.27; N, 7.87; Pd, 19.9. Found: C, 38.15; H,
2.19; N, 7.77; Pd, 20.2.
Characterization of [(9-methyl-1,10-phenanthrolin-2-yl)methanolate]
1 Pd(NCCH3)][OTf] (6). H NMR (500 MHz, DMSO-d6): # 2.06 (s, 3H), 5.24 (s, 2H),
7.67 (d, 1H, J = 8.5 Hz), 7.77 (d, 1H, J = 8.5 Hz), 7.86 (d, 1H, J = 9.0 Hz), 7.88 (d, 1H, J
= 8.5 Hz), 8.47 (d, 1H, J = 8.0 Hz), 8.65 (d, 1H, J = 8.5 Hz); 1H NMR (300 MHz,
CD3CN): # 2.91 (s, 3H), 5.35 (s, 2H), 7.82 (d, 1H, J = 8.7 Hz), 7.83 (d, 1H, J = 8.7 Hz),
8.10 (d, 2H, J = 0.9 Hz), 8.61 (d, 1H, J = 8.4 Hz), 8.76 (d, 1H, J = 8.4 Hz); 13C NMR
- (500 MHz, DMSO-d6): # 1.18, 23.57, 74.63, 118.14, 120.70 (q, J = 320.6 Hz, CF3SO3 ),
123.81, 126.05, 127.54, 127.67, 128.62, 128.66, 139.21, 140.02, 142.82, 144.84, 150.91,
- + 164.59; ESI-MS: m/z (relative intensity) [ion] 329 (28%) [M – CF3SO3 – CH3CN] , 299
- + - 2+ + (100%) [M – CF3SO3 – CH3CN – CH2O] , 223 (10%) [M – CF3SO3 – CH3CN – Pd ] .
+ - + HRMS (ESI) calcd for C14H11N2PdO [M – CF3SO3 – CH3CN] , required m/z 328.9906, found, 328.9910.
Conditions for crystallization of 1, 4, and 5. Crystallization of 1 was achieved by diffusion of ether vapor into a saturated solution of 1 in acetonitrile. High quality crystals were obtained in less than 30 min. Crystallization of 4 was effected by vapor diffusion of dichloromethane into a solution of 4 in dimethylsulfoxide. Due to its asymmetry, crystallization of 5 proved to be much more difficult and a large number of vapor diffusion conditions failed. Suitable crystals were finally obtained by slow evaporation of an acetonitrile solution of 5.
53 2.3.2 Alcohol oxidation and analysis
Protocol for aerobic alcohol oxidation of 2-heptanol using 1 or 3. To a 25 ml round-bottom flask with stirbar was added acetonitrile (2.0 ml), 2-heptanol (142.2 µL,
1.000 mmol), and optionally, n-decane (100.0 µL, 0.5130 mmol, int. std.). The mixture was stirred until the n-decane dissolved completely, and an aliquot was collected at t = 0 for analysis by GC (gas chromatography). The catalyst (3 mol% Pd) was added and the reaction mixture was stirred vigorously at room temperature under a balloon of air.
During the reaction, aliquots were collected, quenched by dilution into acetonitrile, and subjected to GC analysis.
Protocol for aerobic alcohol oxidation of 2-heptanol using 2. The above procedure was followed, with the exception that acetonitrile/dichloromethane (1:1) was used as the solvent (2.0 ml).
Protocol for aerobic alcohol oxidation of 2-heptanol using 4. The above procedure was followed, with the exception that acetonitrile/dimethylsulfoxide (1:1) was used as the solvent (2.0 ml).
Conditions for GC separation. An Alltech EC-WAX GC column (30 m, 0.53 mm ID, 1.00 µm film) was employed for all separations using the following conditions: column head pressure, 100 kPa (14.5 psig) helium; initial column temp., 70 °C; initial hold time, 0 min; rate of temp. ramp, 20 °C/min; final temp., 150 °C; final hold time, 0 min; injection temp., 225 °C; detector temp. 225 °C. The effluent was combusted in a 54 H2/O2 flame and detected using an FID (flame ionization detector). Ion count data were sent to a plotter, which integrated the area under the peaks.
Determination of reaction progress. The reaction progress was determined by use of either (a) an internal standard method or (b) a ratiometric method.
a. Internal standard method
The response factor (RF) for the 2-heptanone/n-decane (internal standard) combination was determined using the following equation:
RF = (areaIS % amountketone)/(amountIS % areaketone), (Eqn. 2.1)
where IS is the internal standard. The amount of the 2-heptanone present in the reaction mixture at any given time could then be calculated using the following equation:
amountketone = (RF % amountIS % areaketone)/(areaIS) (Eqn. 2.2)
b. Ratiometric method
In the aerobic oxidation of 2-heptanol, it was found that the percent yield could also be determined using a ratiometric method, shown by the following equation:
% yield = [area2-heptanone/(area2-heptanone + area2-heptanol)] % 100. (Eqn. 2.3)
55 This equation is valid because (1) 2-heptanol is converted only to 2-heptanone and (2) equimolar amounts of 2-heptanol and 2-heptanone produce the same FID response. The validity of this equation was verified using the internal standard method. In cases where the secondary alcohol and its corresponding ketone produce different detector responses, it is necessary to account for this using a response factor. The ratiometric method is not suitable for primary alcohols, such as 1-heptanol, because such alcohols can be oxidized to carboxylic acids.45
2.3.3 Thermodynamic data for [(neocuproine)Pd(OAc)]2[OTf]2
Table 2.2. Monomer and dimer concentrations determined by 1H NMR in a 2.56 mM solution of 1 in acetonitrile-d3 at various temperatures. T (° C) Ratio of monomer:dimer [monomer] (mM) [dimer] (mM) 1H-NMR peak area 22 1 2.55 1.28 34 1.44 3.01 1.05 48 2.25 3.53 0.79 62 3.23 3.89 0.61 72 4.08 4.10 0.50
56
Figure 2.6 van’t Hoff plot obtained from the monomer:dimer ratio at various temperatures as determined by 1H NMR. The equilibrium constant K was calculated by [monomer]2/[dimer].
Table 2.3. Calculated values for K, !G, and !S at various temperatures. T (K) K (M) = [M]2/[D] !G° (J/mol) = -RTln(K) !S (J/mol) = -[(!G-!H)/T] 295 0.00510 12947 64.9 307 0.00866 12119 65.1 321 0.0159 11058 65.6 335 0.0250 10274 65.2 345 0.0336 9732 64.8
2.3.4 Electrochemical detection of hydrogen peroxide
General Experimental Details:
2-Heptanol, tetrabutylammonium triflate, and hydrogen peroxide (30%) were purchased from Sigma-Aldrich and used without further purification. Acetonitrile was
57 purchased from Fisher Chemical and used without further purification. Electrochemistry experiments were carried out with solutions of tetrabutylammonium triflate (0.1 M) in acetonitrile (10 mL) unless otherwise noted.
All electrochemical measurements were conducted with a 0.195 cm2 Pt-working electrode on a rotating disk. The electrode was cleaned by polishing with 0.05 µm alumina in water on a felt cloth and sonicating for 2 minutes in water. Ag/AgNO3(sat’d) reference and a Pt-mesh counter electrode were employed. Cyclic voltammograms were acquired using a BioAnalytical Systems CV-50W potentiostat and Pine Instrument and a
Pine Instrument Company ASR rotator.
Determining a limit of detection for hydrogen peroxide:
To estimate a limit of detection for hydrogen peroxide under conditions similar to those used in the aerobic oxidation of 2-heptanol, various amounts of hydrogen peroxide were added to standard solutions containing either 1 (3 mM) or 2-heptanol (200 mM), and cyclic voltammograms were recorded. Above a concentration of 0.5 mM hydrogen peroxide in either solution, cyclic voltammograms showed a peak at approximately 1500 mV vs. NHE. Figure 2.7 shows examples at two different hydrogen peroxide concentrations.
58
Figure 2.7 Cyclic voltammograms of 9.7 mM (black) and 19.4 mM (red) hydrogen peroxide in the presence of 2-heptanol. The vertical line denotes 1500 mV vs. NHE.
Electrochemical monitoring of aerobic 2-heptanol oxidation for hydrogen peroxide:
To test for hydrogen peroxide evolution during the aerobic oxidation of 2- heptanol, a 3 mM solution of 1 was prepared (see general experimental details). The solution was stirred continuously using the rotating disk electrode (1600 rpm) and was constantly sparged with air. At t = 0, 2-heptanol was added (200 mM final concentration) and current was monitored for 1200 s with the working electrode held at +2100 mV
(Figure 2.8). No increase in current was observed over the time course of the experiment.
Independent GC analysis showed a 2-heptanone yield of 13.5% during this interval.
Addition of hydrogen peroxide at the end of the electrochemical experiment gave the
59 characteristic peak, although the peak was shifted to 1200 mV vs. NHE (Figure 2.9). A control experiment was also conducted using a solution of 3 mM 1 without 2-heptanol present and no current was observed when the electrode was held at +2100 mV (Figure
2.8).
Figure 2.8 Monitoring of H2O2 evolution from 1 during 2-heptanol oxidation by chronoamperometry. Current was measured in an aerated solution of 1 in the absence of 2-heptanol for 600 s (red) and in the presence of 2-heptanol for 1200 s (blue). The electrode was rotated at 1600 rpm and held at 2100 mV vs. NHE.
60
Figure 2.9 Cyclic voltammograms of the solution before addition of 2-heptanol (black), 1200 s after addition of 2-heptanol (red), and after addition of hydrogen peroxide (40 mM) (blue).
Electrochemical monitoring of hydrogen peroxide decomposition by 1:
A 3 mM solution of 1 was prepared and stirred using the rotating disk electrode.
The potential was held at +2100 mV (vs. NHE) and the electrode rotated at 1600 rpm.
Hydrogen peroxide (10 mM final concentration) was added and the current was monitored for 600 s. Whereas little decrease in current was observed in the control solution with no catalyst, the catalyst solution exhibited a substantial decrease in the current (Figure 2.10) that could be fit to an exponential decay function, suggesting first order decay in H2O2. This experiment was repeated two more times with similar results.
61
Figure 2.10 Electrochemical detection of H2O2 using a rotating disk electrode held at 2100 mV vs. NHE (blue) and its decomposition in the presence of 1 (red).
2.3.5 Conversion of μ-hydroxo Pd dimer 4 to 1 using acetic acid
µ-Hydroxo Pd dimer 4 (5 mg, 0.005 mmol) was added to an NMR tube containing
1 acetonitrile-d3 (0.7 mL, sparingly soluble) and the H NMR spectrum was recorded.
Immediately afterwards, a solution of acetic acid in acetonitrile-d3 was added (1 eq. w.r.t.
Pd, 60 µL, 0.17 M). Additional equivalents of acetic acid were added (up to 8 equivalents w.r.t. Pd) and 1H NMR spectra were recorded. After addition of 8 eq., the
NMR tube was sonicated and the remaining solid reacted to give 1 (Figure 2.11).
62
Figure 2.11 Conversion of 4 to 1 by addition of acetic acid; a.) before addition, b.) after addition of HOAc (1 eq·Pd atom-1), c.) after addition of HOAc (2 eq·Pd atom-1), d.) after addition of HOAc (4 eq·Pd atom-1). Inset: enlargement of the aromatic region for a-d.
63 2.3.6 Conversion of 1 to μ-hydroxo Pd dimer 4 using water
The dimeric compound 1 (10 mg, 0.0096 mmol) was dissolved in acetonitrile-d3
(0.7mL). Two drops of water were introduced to the NMR tube followed by vigorous shaking. More concentrated solutions of 1 can also be used; in this case, 4 can be isolated by precipitation, which will occur upon addition of water (up to a 1:1 water- acetonitrile mixture).
64
Figure 2.12. Formation of 4 by addition of H2O to 1; (top) 1 in acetonitrile; (bottom) after addition of 2 drops of H2O. Peaks are labeled as specified in the legend.
65 2.4. References
[1] Jackson, N. B. Catalyst Technology Roadmap Report, Sandia National
Laboratories, 1997.
[2] Collman, J. P.; Devaraj, N. K.; Decreau, R. A.; Yang, Y.; Yan, Y. L.; Ebina, W.;
Eberspacher, T. A.; Chidsey, C. E. D. A cytochrome c oxidase model catalyzes
oxygen to water reduction under rate-limiting electron flux. Science 2007, 315,
1565-1568.
[3] Olmsted, D. Forever Young: Don't Help the Reaper. University Press
International 2005.
[4] Stahl, S. S. Palladium oxidase catalysis. Selective oxidation of organic chemicals
by direct dioxygen-coupled turnover. Angew. Chem., Int. Ed. 2004, 43, 3400-
3420.
[5] Stahl, S. S. Palladium-catalyzed oxidation of organic chemicals with O2. Science
2005, 309, 1824-1826.
[6] Muzart, J. Palladium-catalysed oxidation of primary and secondary alcohols.
Tetrahedron 2003, 59, 5789-5816.
[7] Schultz, M. J.; Sigman, M. S. Recent advances in homogeneous transition metal-
catalyzed aerobic alcohol oxidations. Tetrahedron 2006, 62, 8227-8241.
[8] Mueller, J. A.; Jensen, D. R.; Sigman, M. S. Dual Role of (-)-Sparteine in the
Palladium-Catalyzed Aerobic Oxidative Kinetic Resolution of Secondary
Alcohols. J. Am. Chem. Soc. 2002, 124, 8202-8203.
66 [9] Steinhoff, B. A.; Stahl, S. S. Ligand-Modulated Palladium Oxidation Catalysis:
Mechanistic Insights into Aerobic Alcohol Oxidation with the Pd(OAc)2/Pyridine
Catalyst System. Org. Lett. 2002, 4, 4179-4181.
[10] Steinhoff, B. A.; Fix, S. R.; Stahl, S. S. Mechanistic Study of Alcohol Oxidation
by the Pd(OAc)2/O2/DMSO Catalyst System and Implications for the
Development of Improved Aerobic Oxidation Catalysts. J. Am. Chem. Soc. 2002,
124, 766-767.
[11] ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Catalytic conversions in
water. Part 21: mechanistic investigations on the palladium-catalyzed aerobic
oxidation of alcohols in water. Adv. Synth. Catal. 2002, 344, 355-369.
[12] ten Brink, G.-J.; Arends, I. W. C. E.; Hoogenraad, M.; Verspui, G.; Sheldon, R.
A. Catalytic conversions in water. Part 22: Electronic effects in the
(diimine)palladium(II)-catalysed aerobic oxidation of alcohols. Adv. Synth. Catal.
2003, 345, 497-505.
[13] Mueller, J. A.; Sigman, M. S. Mechanistic Investigations of the Palladium-
Catalyzed Aerobic Oxidative Kinetic Resolution of Secondary Alcohols Using (-
)-Sparteine. J. Am. Chem. Soc. 2003, 125, 7005-7013.
[14] Steinhoff, B. A.; Guzei, I. A.; Stahl, S. S. Mechanistic Characterization of
Aerobic Alcohol Oxidation Catalyzed by Pd(OAc)2/Pyridine Including
Identification of the Catalyst Resting State and the Origin of Nonlinear [Catalyst]
Dependence. J. Am. Chem. Soc. 2004, 126, 11268-11278.
[15] Nielsen, R. J.; Keith, J. M.; Stoltz, B. M.; Goddard, W. A., III A Computational
Model Relating Structure and Reactivity in Enantioselective Oxidations of
67 Secondary Alcohols by (-)-Sparteine-PdII Complexes. J. Am. Chem. Soc. 2004,
126, 7967-7974.
[16] Mueller, J. A.; Goller, C. P.; Sigman, M. S. Elucidating the Significance of b-
Hydride Elimination and the Dynamic Role of Acid/Base Chemistry in a
Palladium-Catalyzed Aerobic Oxidation of Alcohols. J. Am. Chem. Soc. 2004,
126, 9724-9734.
[17] Zierkiewicz, W.; Privalov, T. A theoretical study of the essential role of DMSO as
a solvent/ligand in the Pd(OAc)2/DMSO catalyst system for aerobic oxidation.
Organometallics 2005, 24, 6019-6028.
[18] Privalov, T.; Linde, C.; Zetterberg, K.; Moberg, C. Theoretical studies of the
mechanism of aerobic alcohol oxidation with palladium catalyst systems.
Organometallics 2005, 24, 885-893.
[19] Keith, J. M.; Nielsen, R. J.; Oxgaard, J.; Goddard, W. A., III Pd-Mediated
Activation of Molecular Oxygen in a Nonpolar Medium. J. Am. Chem. Soc. 2005,
127, 13172-13179.
[20] Schultz, M. J.; Adler, R. S.; Zierkiewicz, W.; Privalov, T.; Sigman, M. S. Using
Mechanistic and Computational Studies To Explain Ligand Effects in the
Palladium-Catalyzed Aerobic Oxidation of Alcohols. J. Am. Chem. Soc. 2005,
127, 8499-8507.
[21] Sigman, M. S.; Jensen, D. R. Ligand-Modulated Palladium-Catalyzed Aerobic
Alcohol Oxidations. Acc. Chem. Res. 2006, 39, 221-229.
68 [22] Steinhoff, B. A.; Stahl, S. S. Mechanism of Pd(OAc)2/DMSO-Catalyzed Aerobic
Alcohol Oxidation: Mass-Transfer-Limitation Effects and Catalyst
Decomposition Pathways. J. Am. Chem. Soc. 2006, 128, 4348-4355.
[23] Steinhoff, B. A.; King, A. E.; Stahl, S. S. Unexpected Roles of Molecular Sieves
in Palladium-Catalyzed Aerobic Alcohol Oxidation. J. Org. Chem. 2006, 71,
1861-1868.
[24] Nielsen, R. J.; Goddard, W. A., III Mechanism of the Aerobic Oxidation of
Alcohols by Palladium Complexes of N-Heterocyclic Carbenes. J. Am. Chem.
Soc. 2006, 128, 9651-9660.
[25] Konnick, M. M.; Gandhi, B. A.; Guzei, I. A.; Stahl, S. S. Reaction of molecular
oxygen with a PdII-hydride to produce a PdII-hydroperoxide: acid catalysis and
implications for Pd-catalyzed aerobic oxidation reactions. Angew. Chem., Int. Ed.
2006, 45, 2904-2907.
[26] Schultz, M. J.; Hamilton, S. S.; Jensen, D. R.; Sigman, M. S. Development and
Comparison of the Substrate Scope of Pd-Catalysts for the Aerobic Oxidation of
Alcohols. J. Org. Chem. 2005, 70, 3343-3352.
[27] Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. Pd(OAc)2-catalyzed oxidation of
alcohols to aldehydes and ketones by molecular oxygen. Tetrahedron Lett. 1998,
39, 6011-6014.
[28] Schultz, M. J.; Park, C. C.; Sigman, M. S. A convenient palladium-catalyzed
aerobic oxidation of alcohols at room temperature. Chem. Commun. 2002, 3034-
3035.
69 [29] ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Green, catalytic oxidation of
alcohols in water. Science 2000, 287, 1636-1639.
[30] ten Brink, G.-J.; Arends, I. W. C. E.; Hoogenraad, M.; Verspui, G.; Sheldon, R.
A. Catalytic conversions in water. Part 23: Steric effects and increased substrate
scope in the palladium-neocuproine catalyzed aerobic oxidation of alcohols in
aqueous solvents. Adv. Synth. Catal. 2003, 345, 1341-1352.
[31] Arends, I. W. C. E.; ten Brink, G.-J.; Sheldon, R. A. Palladium-neocuproine
catalyzed aerobic oxidation of alcohols in aqueous solvents. J. Mol. Catal. A:
Chem. 2006, 251, 246-254.
[32] Jensen, D. R.; Sigman, M. S. Palladium Catalysts for Aerobic Oxidative Kinetic
Resolution of Secondary Alcohols Based on Mechanistic Insight. Org. Lett. 2003,
5, 63-65.
[33] Jensen, D. R.; Schultz, M. J.; Mueller, J. A.; Sigman, M. S. A well-defined
complex for palladium-catalyzed aerobic oxidation of alcohols: Design, synthesis
and mechanistic considerations. Angew. Chem., Int. Ed. 2003, 42, 3810-3813.
[34] Iwasawa, T.; Tokunaga, M.; Obora, Y.; Tsuji, Y. Homogeneous Palladium
Catalyst Suppressing Pd Black Formation in Air Oxidation of Alcohols. J. Am.
Chem. Soc. 2004, 126, 6554-6555.
[35] Komano, T.; Iwasawa, T.; Tokunaga, M.; Obora, Y.; Tsuji, Y. MALDI TOF Mass
Study on Oligomerization of Pd(OAc)2(L)2 (L = Pyridine Derivatives): Relevance
to Pd Black Formation in Pd-Catalyzed Air Oxidation of Alcohols. Org. Lett.
2005, 7, 4677-4679.
70 [36] Ferreira, E. M.; Stoltz, B. M. The Palladium-Catalyzed Oxidative Kinetic
Resolution of Secondary Alcohols with Molecular Oxygen. J. Am. Chem. Soc.
2001, 123, 7725-7726.
[37] Jensen, D. R.; Pugsley, J. S.; Sigman, M. S. Palladium-catalyzed enantioselective
oxidations of alcohols using molecular oxygen. J. Am. Chem. Soc. 2001, 123,
7475-7476.
[38] Wenzel, T. T. Cationic palladium nitro complexes as catalysts for the oxygen-
based oxidation of alkenes to ketones, and for the oxydehydrogenation of ketones
and aldehydes to the !,"-unsaturated analogs. J. Chem. Soc., Chem. Commun.
1989, 932-3.
[39] Drent, E.; Budzelaar, P. H. M. Palladium-Catalyzed Alternating
Copolymerization of Alkenes and Carbon Monoxide. Chem. Rev. 1996, 96, 663-
81.
[40] Slootweg, J. C.; Chen, P. Cationic Palladium Bis-carbene Carboxylate
Complexes. Organometallics 2006, 25, 5863-5869.
[41] Hallman, K.; Moberg, C. Palladium(II)-catalyzed oxidation of alcohols with air as
reoxidant. Adv. Synth. Catal. 2001, 343, 260-263.
[42] Paavola, S.; Zetterberg, K.; Privalov, T.; Csoeregh, I.; Moberg, C. Aerobic
oxidation of 1-phenylethanol catalyzed by palladaheterocycles. Adv. Synth. Catal.
2004, 346, 237-244.
[43] Wimmer, S.; Castan, P.; Wimmer, F. L.; Johnson, N. P. Preparation and
interconversion of dimeric di-#-hydroxo and tri-#-hydroxo complexes of
71 platinum(II) and palladium(II) with 2,2'-bipyridine and 1,10-phenanthroline. J.
Chem. Soc., Dalton Trans. 1989, 403-12.
[44] Fekl, U.; van Eldik, R.; Richardson, C.; Robinson, W. T. Bowl-Shaped and
Highly Inert Bis(#-hydroxo)platinum(II) Dimer. Inorg. Chem. 2001, 40, 3247-
3251.
[45] Sheldon, R. A.; Arends, I. W. C. E.; ten Brink, G.-J.; Dijksman, A. Green,
Catalytic Oxidations of Alcohols. Acc. Chem. Res. 2002, 35, 774-781.
[46] For solubility reasons, 2 and 4 were tested in acetonitrile/dichloromethane (1:1)
and acetonitrile/DMSO (1:1), respectively. To demonstrate that the use of
acetonitrile/DMSO (1:1) was not responsible for the large difference in activity
between 1 and 4, 1 was also tested in this solvent system and exhibited a TOFi =
52.4 Pd atom-1•hr-1.
[47] After 24 h, the rate of aerobic oxidation of 2-heptanol catalyzed by either 1 or 4
was negligible due to complete decomposition of the catalyst. The TON for 1
(~12) was higher than the TON for 4 (~9) because, in addition to decomposing to
5, complex 4 has a second decomposition pathway; 4 is unstable in solution, and
decomposes to an unidentified species on the same time scale as the reaction.
[48] Mori, K.; Yamaguchi, K.; Hara, T.; Mizugaki, T.; Ebitani, K.; Kaneda, K.
Controlled synthesis of hydroxyapatite-supported palladium complexes as highly
efficient heterogeneous catalysts. J. Am. Chem. Soc. 2002, 124, 11572-11573.
[49] Korovchenko, P.; Donze, C.; Gallezot, P.; Besson, M. Oxidation of primary
alcohols with air on carbon-supported platinum catalysts for the synthesis of
aldehydes or acids. Catal. Today 2007, 121, 13-21.
72 [50] The NHC palladium pivolate system reported by Sigman and coworkers (see ref.
26) exhibits 97% conversion in the oxidation of 2-decanol in 14 h with 1 mol %
catalyst under an ambient atmosphere of air at room temperature. Although no
-1 -1 TOFi is reported, the average TOF for this catalyst is 6.9 Pd atom ·hr .
[51] Other phenanthroline catalysts (see ref. 11) are believed to exhibit rate-limiting ß-
hydride elimination.
[52] Complex 2 has been used as an effective aerobic alcohol oxidation catalyst at 80
°C with 30 bar air (see refs. 30 and 31).
[53] At room temperature, the 1H NMR spectrum of 4 (7.4 mM in acetonitrile-
d3/dimethylsulfoxide-d6, 1:1) indicates that only a single species is present. While
we suspect that this species is the dimer 4, we have not yet ruled out its
assignment as monomeric 4.
[54] Although the presence of water does create an equilibrium between 1, 4, and
acetic acid, the equilibrium is fast. As a result, the concentration of 4 is quickly
depleted as 1 reacts with 2-heptanol to generate acetic acid and a high TOFi is still
observed.
[55] The limit of detection for hydrogen peroxide by this method was estimated to be
0.5 mM (see Section 2.3.4). .
[56] These disproportionation catalysts include catalase enzyme (water/acetonitrile
1:1), silver(I) oxide, manganese(II) oxide, and palladium/calcium carbonate. The
use of either potassium iodide or sodium thiosulfate resulted in precipitation of a
palladium complex.
73 [57] Strukul, G.; Ros, R.; Michelin, R. A. Preparation and oxygen-transfer properties
of novel palladium(II) and platinum(II) hydroperoxo and alkylperoxo complexes.
Inorg. Chem. 1982, 21, 495-500.
[58] Denney, M. C.; Smythe, N. A.; Cetto, K. L.; Kemp, R. A.; Goldberg, K. I.
Insertion of molecular oxygen into a palladium(II) hydride bond. J. Am. Chem.
Soc. 2006, 128, 2508-2509.
[59] Popp, B. V.; Wendlandt, J. E.; Landis, C. R.; Stahl, S. S. Reaction of molecular
oxygen with an NHC-coordinated Pd0 complex: computational insights and
experimental implications. Angew. Chem., Int. Ed. 2007, 46, 601-604.
[60] Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. Palladium(II)-Catalyzed
Oxidation of Alcohols to Aldehydes and Ketones by Molecular Oxygen. J. Org.
Chem. 1999, 64, 6750-6755.
[61] Stahl, S. S.; Thorman, J. L.; Nelson, R. C.; Kozee, M. A. Oxygenation of
Nitrogen-Coordinated Palladium(0): Synthetic, Structural, and Mechanistic
Studies and Implications for Aerobic Oxidation Catalysis. J. Am. Chem. Soc.
2001, 123, 7188-7189.
[62] During the course of the aerobic 2-heptanol oxidation, dimer 4 appears as a
precipitate in the reaction mixture. Dimer 4 eventually redissolves, presumably
by action of acetic acid, to regenerate 1.
[63] It is noteworthy that Sheldon, in carrying out similar oxidations with
(dmp)Pd(OAc)2 2 at higher temperatures (80 °C, 30 bar air) did not report ligand
oxidation. This may be due to the different reaction conditions (presence of
74 sodium acetate) or the faster rate of hydrogen peroxide (or palladium
hydroperoxide) disproportionation at these higher temperatures.
[64] Tsuchiya, Y.; Hamashima, Y.; Sodeoka, M. A New Entry to Pd-H Chemistry:
Catalytic Asymmetric Conjugate Reduction of Enones with EtOH and a Higher
Enantioselective Synthesis of Warfarin. Org. Lett. 2006, 8, 4851-4854.
[65] Monguchi, D.; Beemelmann, C.; Hashizume, D.; Hamashima, Y.; Sodeoka, M.
Catalytic asymmetric conjugate reduction with ethanol: A more reactive system
Pd(II)-iPr-DUPHOS complex with molecular sieves 4Å. J. Organomet. Chem.
2008, 693, 867-873.
[66] Brunel, J. M. Scope, limitations and mechanistic aspects in the selective
homogeneous palladium-catalyzed reduction of alkenes under transfer hydrogen
conditions. Tetrahedron 2007, 3899-3906.
[67] Muñiz, K.; Nieger, M. Catalytic Activation of N-N Multiple Bonds; A
Homogeneous Palladium Catalyst for Mechanistically Unprecedented Reduction
of Azo Compounds. Angew. Chem., Int. Ed. 2006, 45, 2305-2308.
[68] Baeckvall, J. E.; Hopkinds, R. B.; Grennberg, H.; Mader, M.; Awasthi, A. K.
Multistep Electron Transfer in Palladium-catalyzed Aerobic Oxidations via a
Metal Macrocycle Quinone System. J. Am. Chem. Soc. 1990, 112, 5160-5166.
[69] Baeckvall, J. E.; Bystroem, S. E.; Nordberg, R. E. Stereo- and Regioselective
Palladium-catalyzed 1,4-diacetoxylation of 1,3-dienes. J. Org. Chem. 1984, 49,
4619-4631.
[70] Cu(II) salts are commonly employed in the reoxidation of Pd(0) species and the
oxidation of bis-(phenanthroline)Cu(I) by oxygen has previously been reported.
75 See for example Zamudio, W.; Garcia, A. M.; Baraona, R. Formation Constants
and Oxidatioin of bis(1,10-Phenanthroline)copper(I) Perchlorate and bis(2,2'-
Bipyridine)copper(I) Perchlorate by Molecular Oxygen in Formamide. Trans.
Met. Chem. 1995, 20, 518-522.
[71] Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas: The Methanol
Economy; Wiley-VCH: Weinheim, 2006.
[72] Zhou, C. H. C.; Beltramini, J. N.; Fan, Y. X.; Lu, G. Q. M. Chemoselective
catalytic conversion of glycerol as a biorenewable source to valuable commodity
chemicals. Chem. Soc. Rev. 2008, 37, 527-549.
[73] Christensen, C. H.; Rass-Hansen, J.; Marsden, C. C.; Taarning, E.; Egeblad, K.
The renewable chemicals industry. Chemsuschem 2008, 1, 283-289.
[74] Behr, A.; Eilting, J.; Irawadi, K.; Leschinski, J.; Lindner, F. Improved utilisation
of renewable resources: New important derivatives of glycerol. Green Chem.
2008, 10, 13-30.
[75] Tyson, K. S.; Bozell, J.; Wallace, R.; Petersen, E.; Moens, L.; (National
Renewable Energy Laboratory NREL/TP-510-34796, Boulder , Co, June 2004,
available at www1.eere.energy.gov/biomass/pdfs/34796.pdf): 2004.
[76] Pagliaro, M.; Ciriminna, R.; Kimura, H.; Rossi, M.; Della Pina, C. From glycerol
to value-added products. Angew. Chem., Int. Ed. 2007, 46, 4434-4440.
76 Chapter 3
Mechanistic Studies of the Oxidative Dehydrogenation of Methanol using a Cationic Palladium Complex
3.0 Preface
This chapter describes previously published research: Pearson, D. M; Waymouth, R. M. Mechanistic Studies of the Oxidative Dehydrogenation of Methanol using a Cationic Palladium Complex. Organometallics 2009, 28, 3896-3900. I performed all experiments in this chapter.
3.1 Introduction
The catalytic oxidation of alcohols serves not only as a practical process to the synthetic chemist, but also as a key transformation in the methanol fuel cell.1-3 The catalytic electro-oxidation of methanol on hetereogeneous electrodes in fuel cells typically proceeds through carbon monoxide (CO) intermediates; the strong affinity of hetereogeneous electrode catalysts for CO leads to poisoning and consequently high overpotentials.1,4-6 The oxidation of methanol to CO has also been observed for homogenous Group 10 complexes.7,8 As part of our efforts to develop low-temperature alcohol oxidation catalysts,9 we sought catalyst systems that could effect methanol
77 oxidation by mechanisms which might avoid CO as an intermediate. Herein we report investigations of methanol oxidation by the cationic palladium complexes
9 [(neocuproine)Pd(OAc)]2[OTf]2 (1) and demonstrate that this complex oxidizes methanol to methyl formate. Mechanistic studies provide evidence for a two-step mechanism involving formaldehyde and formal intermediates. Significantly, CO is not an intermediate in this process.
The selective oxidation of methanol to methyl formate has been previously observed for selected homogeneous10-13 and heterogeneous catalysts;14-18 limited mechanistic studies suggest that these can proceed through a hemiformal intermediate.11,19 Methyl formate is an important chemical feedstock, which can readily provide a variety of C1 to C3 compounds of significant value.20-22 Industrially, methyl formate is produced by the carbonylation of methanol in the presence of sodium methoxide,23 but is also catalyzed by metal carbonyls24 or homogeneous Pt phosphine
25 complexes. Methyl formate can also be prepared by the catalytic reduction of CO2 in
26 methanol by Pd(diphos)2 and H2. The dehydrogenation of methanol to methyl formate with concomitant generation of H2 has been reported with heterogeneous catalysts
(typically Cu based) at high temperatures.14-18
Homogeneous Pd complexes have been shown to be competent alcohol oxidation catalysts.9,12,13,27-32 We recently reported the synthesis of the cationic palladium complex
9 [(neocuproine)Pd(OAc)]2[OTf]2 (1) and its use toward aerobic oxidation of 2-heptanol.
This catalyst showed high initial rates for alcohol oxidation at room temperature, which we attributed to the presence of an internal acetate base and a noncoordinating, anionic triflate, providing an open coordination site. However, the aerobic oxidation of 2-
78 heptanol was accompanied by the oxidation of the proximal methyl group of the ligand; the resultant tridentate complex 2 was inactive (Figure 3.1).
OTf OTf Ac N N O ROH, O2 NCCH3 Pd Pd 2 N N O
1 2 O
Figure 3.1. Ligand oxidation of cationic palladium complex 1 during the aerobic oxidation of alcohols.
3.2 Results and Discussion
Attempts to oxidize methanol with 1 under the aerobic conditions previously used for 2-heptanol9 were unsuccessful. Very slow rates were observed at room temperature; at higher temperatures (50 °C), the catalytic oxidation of methanol in CD3CN in the presence of air or oxygen as the terminal oxidant selectively generates methyl formate, albeit in low yields (Table 3.1). Under these conditions (1 atm of air, 50 °C), palladium black rapidly precipitates from solution. A slight improvement was observed if 1 atm of
O2 was used in place of air; in both cases low turnover numbers (TONs) were observed
(entries 1-2). These results suggest that under these conditions, air or oxygen are not kinetically competent oxidants for the reduced palladium species.
79 Table 3.1 Catalytic oxidation of 2-heptanol and methanol.a entry alcohol oxidant catalyst time, h conv. yield Sel (M) (mol % Pd) (%) (%) (%)e
1 MeOH Air 1 (5) 2 13 11b 85
b 2 MeOH O2 1 (5) 24 21 19 90 3 2-heptanol BQ (0.25) 1 (5) 1 94 94c - 4 MeOH BQ (0.25) 1 (5) 24 82 64b,d 78 5 MeOH BQ (0.25) 2 (5) 24 3 - - 6 MeOH BQ (0.25) 3 (5) 24 12 4b 33 a b c Reaction Conditions; alcohol (0.2 M) in CD3CN at 50 °C, yield of methyl formate, yield of 2-heptanone, d 4-hydroxyphenyl formate is formed in 6% yield, e selectivity for 2•[MF] methyl formate is defined as SelMF = , where [MF] and [MeOH] are [MeOH]i "[MeOH] the observed concentrations of methyl formate and methanol respectively and [MeOH]i is the initial concentration of methanol ! The low conversions observed under our conditions using O2 at 50 °C prompted us to investigate benzoquinone33-37 as a terminal oxidant. The oxidation of 2-heptanol at room temperature using 1 and benzoquinone exhibited slower rates than those previously observed with air, but resulted in a longer-lived catalyst. At 50 °C, catalytic oxidation of
2-heptanol with 1 in the presence of benzoquinone yielded 2-heptanone in 94% yield after 1 h (Table 3.1, entry 3).
Catalytic oxidation of methanol with 1 in the presence of benzoquinone proceeded to 82% conversion after 24 hours at 50 °C, significantly slower than that observed for 2-heptanol oxidation (Table 3.1, entries 3 and 4). The lower rate of methanol oxidation relative to that of other alcohols has been observed previously in aerobic oxidations using palladium.38 The major product was methyl formate (78% selectivity); low yields of 4-hydroxyphenyl formate were also identified by NMR. In
80 accord with our previous work on the aerobic oxidation of 2-heptanol,9
(neocuproine)Pd(OAc)2, 2, was ineffective as a catalyst under these conditions (entry 6).
The dicationic complex [(neocuproine)Pd(CH3CN)2][OTf]2, 3, was slightly more effective for methanol oxidation than 2, but showed significantly lower conversions and selectivities for the formation of methyl formate than 1 (entries 4 and 6).
The lower rates for 2-heptanol oxidation and lower selectivity for methyl formate using benzoquinone as a terminal oxidant suggest that this oxidant and/or the reduced hydroquinone may not be entirely innocent. To investigate the role of benzoquinone, the concentration of benzoquinone was varied from 0.1 to 0.4 M (0.5 to 2 equiv respectively) in the oxidation of methanol with 1 (5 mol %) at 50 °C (see Section 3.3.1, Figure 3.7). In all cases the reaction profile is characterized by an initial sharp loss in selectivity for methyl formate. The selectivity for methyl formate increases gradually with increasing conversion. This behavior becomes more pronounced as the starting concentrations of benzoquinone are increased. These observations strongly suggest the build up of an unobservable intermediate that is trapped reversibly by benzoquinone or hydroquinone.
Acid catalyzed transesterification of crude reaction mixtures with benzyl alcohol did not result in an increase in the total amount of formate (total yield of benzyl formate = initial yield of methyl + 4-hydroxyphenyl formates). We suggest that the lower selectivity in the presence of excess benzoquinone may be a consequence of condensation reactions between benzoquinone and formaldehyde (see Section 3.3.1) .
Several mechanisms for the oxidative dehydrogenation of methanol to methyl formate have been proposed (Figure 3.2).10-19 These mechanisms implicate formaldehyde as a key intermediate. Formaldehyde can react further yielding formic acid (Pathway A),
81 carbon monoxide (Pathway B), methyl formate (Pathway C), or methyl hemiformal
(Pathway D).
Pathway A O
H OH - 2e MeOH + OH- Pathway B -H2O - 2H+ CO MeOH - 2e- Pathway C O O - 2H+ Disportionation MeOH - 2e- H H H O
MeOH Pathway D - 2H+ - 2e- HO OMe
Figure 3.2. Pathways for the formation of methyl formate from methanol.
We performed several mechanistic and labeling studies with 1 in an effort to distinguish among these reaction pathways. In-situ 13C-NMR monitoring of the oxidation
13 of CH3OH with 1 in the presence of benzoquinone provided no evidence for the buildup
13 13 of CO, or labeled formic acid (H CO2H) under standard conditions (see Section 3.3.2).
13 Furthermore, when formic acid and CH3OH were subjected to the reaction conditions,
13 13 only doubly labeled methyl formate (H CO2 CH3) was observed. These studies suggest that oxidation of methanol to formic acid followed by esterification (Pathway A) is not significant under the reaction conditions employed.
The oxidation of methanol under an atmosphere of CO with 2.5 mol % 1 at room temperature led to the formation of a Pd carbonyl adduct after 5 min and complete suppression of methanol oxidation at 50 °C (see Section 3.3.3). At higher catalyst loadings (5 mol %), the formation of methyl formate was observed, but the rate was
82 13 roughly half that of oxidations done in the absence of CO. Oxidation of CH3OH under
1 atm of CO resulted in no detectable incorporation of CO into the methyl formate, nor was 13CO observed during the reaction in the absence of CO. These results reveal that exogenous CO is not incorporated into methyl formate and imply that carbon monoxide is not an intermediate (Pathway B) in the catalytic oxidation of methanol by 1.
To further probe the feasibility of pathway B, we investigated the oxidation of
CD3OH under standard reaction conditions (see Section 3.3.4). Substitution of deuterium into the methyl position resulted in noticeable decrease in rate. After 48 hours a 51% yield of methyl formate was obtained. Comparison of the deuterium resonances at the formyl and methyl positions gave a ratio of 1.1:3.0 suggesting quantitative conservation of deuterium in the formyl position. The generation of deutero-formate products disfavors a carbonylation mechanism (Pathway B) as insertion of CO into the palladium- methoxide bond and subsequent protonation by acetic acid39 or other proton sources would yield protio-formates (Figure 3.3). Consequently, these labeling experiments are most consistent with pathways C and D (Figure 3.2).
Pathway B: CO Insertion Mechanism HOAc O N OAc HOCD N OCD3 CO N CD N OAc Pd 3 Pd Pd O 3 Pd N N N N
HOAc O CD3 Pathway D: Hemiformal Oxidation Mechanism H O
HO OCD3 N OAc N O OCD N D Pd D D Pd 3 Pd N N D D N
HOAc O CD D O 3 Figure 3.3. Isotopic differences as a consequence of reaction pathway in the oxidation of CD3OH
83 To address the possibility of a Tishchenko or Cannizzaro type disproprotionation
13 mechanisms (Pathway C), the oxidation of a 1:1 mixture of CH3OH and CD3OH was carried out and monitored by 1H and 13C-NMR (see Section 3.3.5). After 100 minutes the reaction had reached 30% conversion. Three discrete methyl formate isotopomers
* * * * H CO2 CH3, H CO2CD3, and DCO2 CH3 were observed (Figure 3.4, DCO2CD3 could not be observed under these conditions) in a ratio of 38:38:12 at 30% conversion (100 min) and 37:33:14 at 67% conversion (24 hr, 50 °C). The absence of any isotope exchange into the methyl signals (CH2DOY or CD2HOY, Y = C(O)H(D) or H) argues against a
Tishchenko or Cannizzaro type disproportionation, as either disproportion mechanism
(Figure 3.5) should lead to scrambling of isotopes into the methyl groups. Similarly, oxidation of a mixture of CH3OH and CD3OH and analysis by GC/MS yielded d0-, d1-, d3-, and d4-methyl formates; the absence of a significant amounts of d2-methyl formates
(m/z = 62, !2%)40 provides further evidence against the disproportionation of formaldehyde as a mechanism for generation of methyl formate (Pathway C, Figure 3.2).
84
1 13 Figure 3.4. H-NMR of the oxidation of CH3OH and CD3OH mixture using 1 (2.5 mol %) and benzoquinone after 100 minutes at 50 °C. The ratio of H4:HD3:DH3 is 38:38:12.
a.) Tishchenko Mechanism M M O O O H H + MF H H H H H O H O H
H b.) Cannizzaro Mechanism H M H O O O MF M + O H H H H H CH3OH CH3 O H O CH3 Figure 3.5. Mechanisms of the (a) Tishchenko and (b) Cannizzaro disproportionation reactions.
These results are most consistent with a mechanism in which methanol is first oxidized to formaldehyde, followed by the formation of a methyl hemiformal (Pathway
D), which is subsequently oxidized to methyl formate (Figure 3.6).41,42 Support for the intermediacy of methyl hemiformal was obtained by monitoring the oxidation of
85 13 13 2 CH3OH at early reaction times; by C-NMR, two doublets at 54.8 and 90.7 ppm ( JC-C=
2.5 Hz) could be observed (see Section 3.3.6, Figure 3.19) which we attribute to the methyl and formal carbons of hemiformal.
The key steps in the oxidation of methanol are proposed to involve the beta- hydrogen elimination from either Pd-OCH3 or Pd-OCH2OCH3 intermediates (Figure 3.6).
The monocationic [(neocuproine)Pd(OAc)]+ likely facilitates binding of methanol (or methyl hemiformal); deprotonation of the bound methanol by the PdOAc followed by beta-hydrogen elimination generates formaldehyde.43 Condensation of formaldehyde with methanol affords the hemiformal, which upon binding to Pd and beta-hydride elimination yields methyl formate. A similar mechanism could account for the formation of 4- hydroxyphenyl formate in the later stages of the reaction as the concentration of hydroquinone increases.
Protiated methanol is oxidized more rapidly than deuterated methanol: analysis of
13 the product distribution for the oxidation of mixtures of CH3OH( CH3OH)/CD3OH yields an integrated isotope effect of approx. (HC(O)OCX3 / DC(O)OCX3, X = H or D)
3 for the formation of the methyl formate. According to the mechanism of Figure 3.6, the formation of methyl formate involves two sequential beta-hydrogen elimination steps.
Comparison of the initial rates of CD3OH and CH3OH consumption yielded a kinetic
44 isotope effect (KIE) of kH/kD = 1.4, similar to that previously reported for other Pd
31,38,45,46 catalysts. We also observed an isotope effect of kOH/kOD = 1.3 by comparison of the initial rates of oxidation of CH3OH and CH3OD, implicating deprotonation as a key step in the oxidation of methanol by 1.44
86 2 RH2COH NCCH3 Ac 2 OTf N N N O CH3CN OAc O Pd 2 2 Pd Pd O N N N N O H CH2R HQ 1
CH3OH O H+ + HOAc + BQ HO OR BQ/HQ Adducts HOAc H H
R = H N N N H Pd H Pd 0 Pd H N O N N H+ R
R= OCH3 or R = HQ O H O BQ = Benzoquinone CH3 HQ = Hydroquinone H O O OH methyl formate 4-hydroxyphenyl formate
Figure 3.6. Proposed mechanism for the oxidation of methanol to methyl formate by 1 and benzoquinone.
In summary, we report the selective oxidation of methanol to methyl formate by the cationic palladium complex [(neocuproine)Pd(OAc)]2[OTf]2 (1) in the presence of benzoquinone as a terminal oxidant. Mechanistic studies implicate the beta-hydride elimination of palladium methoxide to generate formaldehyde followed by formation of methyl hemiformal and beta-hydrogen elimination to generate methyl formate. Labeling studies indicate that carbon monoxide is not an intermediate in these oxidation reactions, illustrating that under appropriate conditions, methanol can be oxidized by mechanisms which do not involve CO as an intermediate.
87 3.3 Experimental
13 CD3CN, CH3OH, and CD3OH were purchased from Cambridge Isotope
Laboratories and were degassed over three freeze-thaw cycles and stored in an inert atmosphere glovebox. Acetonitrile was dried over CaH2, transferred under reduced pressure and stored under an inert atmosphere. Methanol was dried over activated molecular sieves and stored over an inert atmosphere. Benzoquinone was purchased from Sigma-Aldrich and recrystallized from reagent grade ethanol, dried in-vacuo, and stored in an inert atmosphere glovebox. Carbon monoxide was purchased from
Matheson Gas. Product yields and rates were determined relative to an internal standard of benzene or hexamethyldislane by NMR spectroscopy using a Varian 300, 400, or 500
MHz spectrometer. Isotopic mass data were acquired using a HP 5890/5970 GC/MS.
Oxidation reactions were carried out in sealable screw-cap NMR tubes under nitrogen unless specified otherwise. [(2,9-dimethyl-1,10-phenanthroline)Pd(µ-OAc)]2(OTf)2, 1, was prepared as previously reported.9
Kinetic isotope effects were measured by the method of initial rates. In a typical experiment, methanol (0.144mmol), benzoquinone (0173.mmol, 1.2equiv), benzene, and acetonitrile (0.7mL) were added to a NMR tube equipped with a rubber septum in a N2 drybox. The samples were removed from the drybox, injected with a stock solution of 1
(60 µL, 12mM, 0.5 mol %), and immediately transferred to an NMR spectrometer preheated to 50º C. Initial rates were measured by loss of methanol relative to benzene internal standard. For CD3OH measurements, dry CH3CN was used as solvent using added CD3CN as an internal standard.
88 3.3.1 Methyl formate selectivity dependence of benzoquinone concentration.
The concentration of benzoquinone was varied from 0.1 M to 0.4 M (0.5 to 2.0 equiv) using CH3OH (0.144 mmol) and 1 (5 mol %). The concentrations of methanol, methyl formate (MF), and 4-hydroxyphenyl formate were monitored relative to benzene internal standard. These concentrations were compared to the initial methanol concentration, [MeOH]I, to establish a mass balance for the reaction (Equation 3.1).
[MeOH]+ 2•[MF]+ [HQ ] Mass balance = Form (Equation 3.1) [MeOH]i
The starting concentration of benzoquinone has a drastic effect on the rate of MF formation (Figure! 3.7). We believe part of this dependence may be due in part to the formation of formals with hydroquinone, which inhibit further oxidation to MF or 4- hydroxyphenol. Comparison of 0.2 M and 0.4 M benzoquinone loadings show a pronounced loss of mass balance at longer reaction times while maintaining similar methanol conversions. This observation suggests that benzoquinone itself may also be responsible for the loss in selectivity for MF. Faster rates for methyl formate production were found with 0.5 equivalents of benzoquinone while slightly higher conversion was observed with 2.0 equivalents.
89
13 Figure 3.7. Oxidation of CH3OH using 1 and (a) 0.5 equiv (0.1 M), (b) 1 equiv (0.2 M), (c) 2 equiv (0.4M) of benzoquinone under N2(g). Concentrations of CH3OH( ), mass balance( ), methyl formate ( ), and 4-hydroxyphenol ( ) have been normalized for direct comparison.
Benzoquinone was not found to react directly with methanol under our reaction conditions. We employed a transesterification reaction to establish the oxidation state of the intermediate trapped by benzoquinone and hydroquinone. Oxidation of methanol using 1 (2.5 mol %) and benzoquinone (1.2 eq) (monitored using 1H-NMR, p-xylene as an internal standard) yielded methyl formate and 4-hydroxyphenyl formate in 59% and
2% yield respectively after 24 h at 50 °C. Approximately 27% of the reactants were absent from the observable mass balance. The reaction material was diluted in benzyl alcohol and a catalytic amount of triflic acid was added to induce transesterification of all formate species to benzyl formate. Analysis by GC/MS showed formation of 62% benzyl formate compared with the 61% of formate species observable by 1H-NMR. Given the results of this experiment we believe that formaldehyde intermediates react competitively with either benzoquinone or hydroquinone to yield products that could not be identified.
90 3.3.2 Formic acid – methanol esterification
A mixture of formic acid and 13C-methanol (1:1, 0.25 mmol each), HQ and BQ
(1:1 0.28 mmol each), and 1 (0.005 mmol) were added to a NMR tube containing 0.7 mL
CD3CN. Minimal conversion to methyl formate (14%) was observed over a period of 9 days at room temperature (30% total conversion). Furthermore, analysis of 13C-NMR suggests that a majority of MF produced is saturated with 13C at both carbons. This data supports that esterification is not the active mechanism for MF production. Formic acid was also consumed over the course of this reaction (Figure 3.8 and 3.9). The observable
CO2 peak in Figure 3.10 is suggestive that formic acid is also oxidized under these conditions.
Figure 3.8. 1H-NMR spectrum of esterification reaction at t = 0.
91
Figure 3.9. 1H-NMR spectrum of esterification reaction after 9 days.
Figure 3.10. 13C-NMR spectrum of esterification reaction after 9 days. Small doublets (insets) are observed in both carbons of methyl formate suggesting formic acid is not responsible for the observed conversion.
92 3.3.3 Carbon monoxide studies – oxidation of methanol under a
CO atmosphere
In a typical reaction, 1 (7.5 mg, 10 mol% Pd) and benzoquinone (32 mg, 2 equiv) were dissolved in d3-acetonitrile (0.7 mL) containing benzene as an internal standard in a
13 sealable screw cap NMR tube. C-methanol is added, the solution is frozen in N2(l), and the head space of the tube is evacuated. One atmosphere of carbon monoxide was then introduced upon thawing. Solutions were subsequently heated to 50°C and monitored by
1H and 13C-NMR.
Addition of CO to solutions containing 1 resulted in clean conversion to an unknown palladium carbonyl species (Figure 3.11). Integrals for the aromatic proton resonances of the neocuproine ligand and the labeled methyl group were approximately
6:3 suggesting that a single methoxide was bound to palladium. We were not able to positively identify the species but believe it to be either a Pd methoxide carbonyl adduct or the product of CO insertion into the palladium methoxide bond and further coordination of CO. A shift in the coordinated methanol was also observed in the 13C-
NMR (Figure 3.12).
93
Figure 3.11. 1H-NMR of reaction after exposure to 1 atm of CO during the oxidation of 13C-MeOH.
Figure 3.12. 13C-NMR of reaction after exposure to 1 atm of CO during the oxidation of 13C-MeOH
94 At lower starting concentrations of 1 (0.05 mM, 2.5 mol %), oxidation of methanol and the production of methyl formate were completely inhibited. However, when the starting concentration of 1 was increased to 0.10 mM (5 mol %), methanol oxidation and methyl formate were observed. Carbon monoxide was not incorporated into methyl formate as evidenced both by the relative intensities of the carbon resonances
13 13 13 2 in the C-NMR and the observed two bond C- C coupling ( JC-O-C = 3.2 Hz) (Figure
3.13).
Figure 3.13 13C-NMR of methanol oxidation after 360 minutes at 50°C.
These observations suggests that CO only serves to occupy the free coordination site of the catalyst (which is necessary for methanol oxidation) and is not involved directly in methyl formate production (Figure 3.14).
95 N OMe N OMe formaldehyde Pd Pd and eventually N CO N methyl formate
"Inactive" open site for !-hydride elimination
Figure 3.14. The postulated role of CO in oxidation of methanol to methyl formate.
3.3.4 Oxidation of CD3OH
Benzoquinone (18.7 mg, 0.173mmol), CD3OH (0.144 mmol), and 1 (0.35 mmol,
2.5 mol %) were added to a solution of CH3CN containing CD3CN (internal standard) in a N2 drybox. The solution was then transferred to a screw cap NMR tube, sealed and heated to 50 ºC. The reaction was periodically monitored by 2H-NMR. After 48 hours the reaction had progressed to 52% conversion (51% yield of MF). Integration of the formyl and methyl resonances in MF yielded a ratio of 1.1:3.0 (Figure 3.15).
96
2 Figure 3.15. H-NMR of the oxidation of CD3OH after 48 hours.
3.3.5 Crossover experiments with isotopically labeled methanols
13 i. . CH3OH and CD3OH
Oxidations experiments were performed by addition of 1 (3.75 mg, 0.0035 mmol,
2.5 mol %), benzoquinone (18.7 mg, 0.173 mmol, 1.25 equiv), CD3OH (2.8 µL, 0.07
13 mmol), CH3OH or CH3OH (0.07 mmol), and CD3CN (0.7 mL) to a screw cap NMR tube. Benzene was added as an internal standard. The solution was brought to 50 °C and monitored periodically by NMR.
The oxidation of CD3OH and CH3OH can lead to four products in the absence hydrogen/deuterium scrambling into the methyl groups.
97 OH O * * * C CH3 * CH H O H O 3 MF-H4 O H * Not Observed HO CH3 *C H H OH O O * * CD * C CD3 3 MF-HD3 * CH D H O H O H O 2 * H HO CH2D O O OH * * CH2D C *CH CH3 MF-DH3 D O O D O 3 D O HOCD3 C D D D OH O C CD CD3 MF-D4 D O 3 D O D
Figure 3.16. Possible products formed during the crossover experiment between CD3OH 13 and CH3OH.
Scrambling of isotopes into the methyl groups (via Tischenko-type mechanisms) increases this number significantly. Products of such shifts should be observable in the
1H-NMR spectrum. Additionally use of proton-decoupled 13C-NMR can easily identify such species if deuterium is transferred.
No evidence of deuterium or hydride transfers was observed using either technique, providing further evidence against a Tischenko disproportionation. Products
MF-H4, MF-HD3, and MF-DH3 were all observed. MF-D4 was not detectable with our experimental setup. The initial ratios of MF-H4, MF-HD3, MF-DH3 were 38:38:12, and little deviation was observed over 24 hours.
98
13 13 Figure 3.17. Proton decoupled C-NMR of CH3OH and CD3OH mixture after 1440 minutes at 50 °C.
ii. CH3OH and CD3OH
Oxidations experiments were performed by addition of 1 (11.3 mg, 0.011 mmol,
2.5 mol %), benzoquinone (56.1 mg, 0.52 mmol, 1.25 equiv), CD3OH (8.8 µL, 0.215 mmol), CH3OH (8.7 µL, 0.215 mmol), and CH3CN (2.1 mL) to a Schlenk flask.
Mesitylene was added as an internal standard. The flask was sealed and the solution was brought to 50°C and monitored periodically by GC/MS.
The reaction displayed an isotopic ratio of 32:12:41:15 for isotopes 60, 61, 63, and 64 after 2h (Figure 3.18). This ratio changed slightly after 24h to 29:12:41:18. The gradual decrease in mass 60 and increase in mass 64 correlates to the anticipated relative abundances of CH3OH and CD3OH at longer times.
99 The 62 mass isotope consistently remained below 2% of the total intensities. We attribute the observance of this isotope to incomplete deuterium labeling in our CD3OH source, estimated at 99.5% D. Using this number we anticipate the 62 m/z signal to account for roughly 1% of total intensities. Deviation from this number is small and we believe is within the error of this experiment.
Figure 3.18. Mass spectrum of methyl formate produced in the oxidation of equimolar mixtures of CH3OH and CD3OH at 50°C after (a) 2h and (b) 24.
100 3.3.6 Evidence of a hemiformal intermediate
13 At early reaction times during the oxidation CH3OH one can often observe a hemiformal intermediate in the 13C-NMR as evidenced by a pair of doublets at 54.8 and
2 90.7 ppm ( JC-C= 2.5 Hz)(Figure 3.19). As the reaction progresses this species disappears.
13 13 Figure 3.19. C-NMR of CH3OH oxidation by 1 (5 mol %) and benzoquinone at 50°C after 65 minutes. Insets highlight the transient generation of methyl hemiformal.
101 3.4 References
[1] Carrette, L.; Friedrich, K. A.; Stimming, U. Fuel cells: Principles, types, fuels,
and applications. ChemPhysChem 2000, 1, 162-193.
[2] Hogarth, M. P.; Ralph, T. R. Catalysis for Low Temperature Fuel Cells, Part III.
Challenges for the Direct Methanol Fuel Cell. Plat. Met. Rev. 2002, 46, 146-164.
[3] Olah, G. A. Beyond oil and gas: The methanol economy. Angew. Chem. Int. Ed.
2005, 44, 2636-2639.
[4] Beden, B.; Hahn, F.; Juanto, S.; Lamy, C.; Leger, J. M. Infrared Spectroscopic
Study of the Methanol Adsorbates at a Platinum-Electrode. 1. Influence of the
Bulk Concentration of Methanol upon the Nature of the Adsorbates. J.
Electroanal. Chem. 1987, 225, 215-225.
[5] Jusys, Z.; Kaiser, J.; Behm, R. J. Composition and activity of high surface area
PtRu catalysts towards adsorbed CO and methanol electrooxidation - A DEMS
study. Electrochim. Acta 2002, 47, 3693-3706.
[6] Burstein, G. T.; Barnett, C. J.; Kucernak, A. R.; Williams, K. R. Aspects of the
anodic oxidation of methanol. Catal. Today 1997, 38, 425-437.
[7] Periana, R. A.; Mironov, O.; Taube, D.; Bhalla, G.; Jones, C. J. Catalytic,
Oxidative Condensation of CH4 to CH3COOH in One Step via CH Activation.
Science 2003, 301, 814-818.
[8] Portnoy, M.; Milstein, D. A Binuclear Palladium(I) Hydride - Formation,
Reactions, and Catalysis. Organometallics 1994, 13, 600-609.
102 [9] Conley, N. R.; Labios, L. A.; Pearson, D. M.; McCrory, C.; Chidsey, C. E. D.;
Waymouth, R. M. Aerobic Alcohol Oxidation with Cationic Palladium
Complexes: Insights into Catalyst Design and Decomposition. Organometallics
2007, 26, 5447-5453.
[10] Starchevsky, M. K.; Hladiy, S. L.; Pazdersky, Y. A.; Vargaftik, M. N.; Moiseev,
I. I. Giant Pd-561 clusters: onset to new catalytic properties. J. Mol. Catal. A:
Chem. 1999, 146, 229-236.
[11] Lastovyak, Y. V.; Gladii, S. L.; Pasichnyk, P. I.; Starchevskii, M. K.; Pazderskii,
Y. A.; Vargaftik, M. N.; Moiseev, I. I. Methanol Oxidation Catalyzed with a
Giant Palladium Cluster. Kinet. Catal. 1994, 35, 512-515.
[12] Lloyd, W. G. Homogeneous oxiations of alcohols with palladium(II) salts. J. Org.
Chem. 1967, 32, 2816-2818.
[13] Nishimura, T.; Kakiuchi, N.; Onoue, T.; Ohe, K.; Uemura, S. Palladium(II)-
catalyzed oxidation of terminal alkenes to methyl ketones using molecular
oxygen. J. Chem. Soc., Perkin Trans. 1 2000, 1915-1918.
[14] Cant, N. W.; Tonner, S. P.; Trimm, D. L.; Wainwright, M. S. Isotopic labeling
studies of the mechanism of dehydrogenation of methanol to methyl formate over
copper-based catalysts. J. Catal. 1985, 91, 197-207.
[15] Iwasa, N.; Yamamoto, O.; Akazawa, T.; Ohyama, S.; Takezawa, N.
Dehydrogenation of Methanol to Methyl Formate over Palladium/Zinc Oxide
Catalysts. J. Chem. Soc., Chem. Commun. 1991, 1322-1323.
103 [16] Liu, Z.-T.; Lu, D.-S.; Guo, Z.-Y. Polyacrylonitrile coated silica as support for
copper catalyst in methanol dehydrogenation to methyl formate. Appl. Catal., A
1994, 118, 163-171.
[17] Forzatti, P.; Tronconi, E.; Busca, G.; Tittarelli, P. Oxidation of methanol to
methyl formate over V-Ti oxide catalysts. Catal. Today 1987, 1, 209-218.
[18] Miyazaki, E.; Yasumori, I. Kinetics of Catalytic Decomposition of Methanol
Formaldehyde and Methyl Formate over a Copper-Wire Surface. Bull. Chem. Soc.
Jpn. 1967, 40, 2012-2017.
[19] Pocker, Y.; Davis, B. C. Mechanism of Methanol Oxidation by Lead Tetra-
acetate. The Novel Intermediacy of Formaldehyde Hemiacetal. J. Chem. Soc.
Chem. Commun. 1974, 803-804.
[20] Ikarashi, T. New MGC Process for High-Purity Carbon Monoxide Production and
Review on Derivatives of Methyl Formate. Chem. Econ. Eng. Rev. 1980, 12, 31-
34.
[21] Lee, J. S.; Kim, J. C.; Kim, Y. G. Methyl Formate as a New Building Block in C1
Chemistry. Appl. Catal. 1990, 57, 1-30.
[22] Roeper, M. Oxygenated Base Chemicals from Synthesis Gas. Erdoel Kohle,
Erdgas, Petrochem. 1984, 37, 506-510.
[23] Aguilo, A.; Horlenko, T. Formic Acid. Hydrocarbon Process. 1980, 59, 120.
[24] Darensbourg, D. J.; Gray, R. L.; Ovalles, C.; Pala, M. Homogeneous catalysis of
methyl formate production from carbon monoxide and methanol in the presence
of metal carbonyl catalysts. J. Mol. Catal. 1985, 29, 285-290.
104 [25] Head, R. A.; Tabb, M. I. Platinum-catalysed carbonylation of methanol to methyl
formate. J. Mol. Catal. 1984, 26, 149-158.
[26] Inoue, Y.; Sasaki, Y.; Hashimoto, H. Synthesis of Formates from Alcohols,
Carbon Dioxide, and Hydrogen Catalysed by a Combination of Group VIII
Transition-Metal Complexes and Tertiary Amines. J. Chem. Soc., Chem.
Commun. 1975, 718-719.
[27] Stahl, S. Palladium Oxidase Catalysis: Selective Oxidation of Organic Chemicals
by Direct Dioxygen-Coupled Turnover. Angew. Chem. Int. Ed. 2004, 43, 3400-
3420.
[28] Stoltz, B. M. Palladium Catalyzed Aerobic Dehydrogenation: From Alcohols to
Indoles and Asymmetric Catalysis. Chem. Lett. 2004, 33, 362-367.
[29] Sigman, M. S.; Jensen, D. R. Ligand-Modulated Palladium Catalyzed Aerobic
Alcohol Oxidations. Acc. Chem. Res. 2006, 39, 221-229.
[30] Schultz, M. J.; Sigman, M. S. Recent advances in homogeneous transition metal-
catalyzed aerobic alcohol oxidations. Tetrahedron 2006, 62, 8227-8241.
[31] ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Catalytic Conversions in
Water. Part 21: Mechanistic Investigations on the Palladium-Catalysed Aerobic
Oxidation of Alcohols in Water. Adv. Synth. Catal. 2002, 344, 355-369.
[32] Arends, I.; Tenbrink, G.; Sheldon, R. Palladium-neocuproine catalyzed aerobic
oxidation of alcohols in aqueous solvents. J. Mol. Catal. A: Chem. 2006, 251,
246-254.
105 [33] Amatore, C.; Cammoun, C.; Jutand, A. Palladium/Benzoquinone-Catalyzed
Electrochemical Oxidation of Alcohols Under Anaerobic Conditions. Synlett
2007, 2173-2178.
[34] Grennberg, H.; Gogoll, A.; Baeckvall, J. Acid-induced transformation of
palladium(0)-benzoquinone complexes to palladium(II) and hydroquinone.
Organometallics 1993, 12, 1790-1793.
[35] Baeckvall, J.; Bystroem, S.; Nordberg, R. Stereo- and regioselective palladium-
catalyzed 1,4-diacetoxylation of 1,3-dienes. J. Org. Chem. 1984, 49, 4619-4631.
[36] Baeckvall, J.; Hopkins, R.; Grennberg, H.; Mader, M.; Awasthi, A. Multistep
electron transfer in palladium-catalyzed aerobic oxidations via a metal
macrocycle quinone system. J. Am. Chem. Soc. 1990, 112, 5160-5166.
[37] Popp, B. V.; Thorman, J. L.; Stahl, S. S. Similarities between the reactions of
dioxygen and alkenes with palladium(0): Relevance to the use of benzoquinone
and molecular oxygen as stoichiometric oxidants in palladium-catalyzed oxidation
reactions. J. Mol. Catal. A: Chem. 2006, 251, 2-7.
[38] Noronha, G.; Henry, P. M. Heterogenized polymetallic catalysts: Part III.
Catalytic air oxidation of alcohols by Pd(II) complexed to a polyphenylene
polymer containing [beta]-di- and tri-ketone surface ligands. J. Mol. Catal. A:
Chem. 1997, 120, 75-87.
[39] Acetic acid would be generated by methanolysis of the Pd acetate bond by
HOCD3.
[40] Products of 62 m/z remained less than 2% of the observable products over the
course of the reaction. Given the reported isotopic purity of CD3OH (99.5% D)
106 and the natural abundance of 13C in these sources, we estimate 62 m/z should
account for 1% of the total observable isotopes, and conclude that the level
observed is within our experimental error .
[41] Schultz, M. J.; Hamilton, S. S.; Jensen, D. R.; Sigman, M. S. Development and
Comparison of the Substrate Scope of Pd-Catalysts for Aerobic Oxidation of
Alcohols. J. Org. Chem. 2005, 70, 3343-3352.
[42] Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. Palladium(II)-Catalyzed
Oxidation of Alcohols to Aldehydes and Ketones by Molecular Oxygen. J. Org.
Chem. 1999, 64, 6750-6755.
[43] It is not clear if formaldehyde remains bound to Pd or if it is released into
solution.
[44] Reaction were run in parallel monitoring the initial consumption of methanol.
[45] Mueller, J. A.; Jensen, D. R.; Sigman, M. S. Dual Role of (-)-Sparteine in the
Palladium-Catalyzed Aerobic Oxidative Kinetic Resolution of Secondary
Alcohols. J. Am. Chem. Soc. 2002, 124, 8202-8203.
[46] Steinhoff, B. A.; Stahl, S. S. Ligand-Modulated Palladium Oxidation Catalysis:
Mechanistic Insights into Aerobic Alcohol Oxidation with the Pd(OAc)2/Pyridine
Catalyst System. Org. Lett. 2002, 4, 4179-4181.
107 Chapter 4
Oxidatively-Resistant Ligands for Palladium Mediated Aerobic Alcohol Oxidation
4.0 Preface
This chapter describes research to be submitted for publication in Organometallics: Pearson, D. M.; Conley, N. R.; Waymouth, R. M. Oxidatively-Resistant Ligands for Palladium Mediated Aerobic Alcohol Oxidation. Organometallics 2010 Submitted. This projected was done in close collaboration with Nicholas R. Conley. N. R. Conley and I provided initial syntheses of ligands and metal complexes, which I subsequently repeated and fully characterized. All ligands mentioned in Section 4.4 were synthesized by myself.
4.1 Introduction
Palladium-catalyzed aerobic alcohol oxidation1-6 has advanced considerably since the initial report in 1977.7 More than a decade later, it was discovered that coordinating solvents such as DMSO or pyridine could effectively stabilize Pd(0) intermediates and facilitate their reoxidation by oxygen before Pd black formation occurs.8-11
Subsequently, other ligands were used to similar effect and well-defined complexes of palladium with sparteine,12-14 phenanthroline,15-19 and N-heterocyclic carbene (NHC)2,20-
23 scaffolds have been reported for aerobic alcohol oxidation. Frequently these systems
108 rely on acetate counter-ions due to their ability to facilitate deprotonation of coordinated alcohols; while other coordinating anions are tolerated, they typically require additional base for catalysis.2,24,25
We previously reported that the 2,9-dimethyl-1,10-phenanthroline (neocuproine) complex [(neocuproine)Pd(OAc)]2[OTf]2 (1) is active for the aerobic oxidation of 2- heptanol at room temperature.26 Complex 1 exhibits a high initial turnover rate (~78 Pd-1 h-1), but degrades rapidly due to competitive oxidation of the ligand by partially reduced oxygen species (Scheme 4.1). Stahl and coworkers27 reported similar ligand oxidation after partial reduction of oxygen by Pd yielding a reactive Pd(II)-hydroperoxo species.
The use of other oxidants, such as benzoquinone, can mitigate this destructive side reaction, but the synthetic utility of oxygen as a terminal oxidant remains attractive.28
Herein we document efforts to develop more oxidatively resistant phenanthroline ligands to mitigate competitive ligand oxidation pathways in the aerobic oxidation of alcohols.
Scheme 4.1. Ligand oxidation resulting from oxygen reduction in aerobic alcohol oxidation
109 4.2 Results and Discussion
4.2.1 Synthesis of oxidatively resistant ligands
Four phenanthroline derivatives were targeted as potential ligands: 1,10- phenanthroline (phen), bis-2,9-(trifluoromethyl)-1,10-phenanthroline (btfm-phen), 4- methyl-2-(trifluoromethyl)-1,10-phenanthroline (tfmm-phen), and 2-(o-difluorophenyl)-
1,10-phenanthroline (odfp-phen). Of these, phen is commercially available and btfm- phen can be synthesized by radical chlorination of neocuproine,29 followed by halogen
30 exchange with SbF3/SbF5.
CH3
N N F3C N F3C N N N N N
CF3 F F phen btfm-phen tfm-mphen 6 odfp-phen 7 Figure 4.1. Substituted phenanthroline ligands
Several 4,5-disubstituted 2-(trifluoromethyl)-1,10-phenanthrolines are known,31,32 but 2-(trifluoromethyl)-1,10-phenanthroline (tfm-phen) has not been reported. We initially attempted the synthesis of tfm-phen by a Skraup-Doebner-Von Miller reaction using !,"-unsaturated carbonyl 4 (Scheme 4.2) and 8-aminoquinoline.33,34 However,
35 isolation of the aldehyde 4 from Et2O proved difficult due its high volatility. The less volatile ketone derivative 536 was easier to purify, and condensation of 5 with 8- aminoquinoline yielded the desired trifluoromethyl-substituted phenanthroline 6.33
110
Conditions: (a) Na2S2O4, NaHCO3, CH3CN, H2O, 10 °C, 1h.; (b) Et2O, polyphosphoric acid, r.t., 2 h (63%); and (c) 8-aminoquinoline, NaI (1 mol %), H2SO4, 110 °C, 5 h (21%). Scheme 4.2. Synthesis of 4-methyl-2-(trifluoromethyl)-1,10-phenanthroline (6)
Fluorinated aryl phenanthrolines have been proposed as oxidatively resistant ligands as reported by Sadighi et al.37 in Cu-mediated transfer of nitrenes. Related 2- polyfluoroaryl-phenanthrolines have also been reported by Deacon et al.38,39 We adapted a synthetic route to 2-phenyl-1,10-phenanthroline reported by Thummel and coworkers
(Scheme 4.3).40 Substitution of 2!,6!-difluoroacetophenone for acetophenone in the final step yielded 7 in 58% yield from 8-amino-7-quinolinecarbaldehyde. Despite being five steps, this strategy provides an expeditious route to 7, requiring no purification steps until the penultimate step. Attempts to synthesize this compound through Sauvage-type conditions41 using 2,6-difluorophenyl lithium42 and phenanthroline provided low yields of 7 which could not readily purified.
111 4 steps 64 % overall N yield N CHO NH2 O
N F F N (1 equiv)
F F KOH, EtOH Reflux,16h (58%) 7
Scheme 4.3. Abridged synthetic route to odfp-phen 7
4.2.2 Synthesis of phenanthroline palladium complexes
The synthesis of cationic (N-N)Pd acetate complexes was carried out by conproportionation of a 1:1 mixture of the diacetate (N-N)Pd(OAc)2 (A) and ditriflate
[(N-N)Pd(NCCH3)2][OTf]2 (B) complexes by a procedure previously described (Scheme
4.4).26
Scheme 4.4. General procedure for the formation of cationic palladium complexes in situ.
For the phenanthroline complexes, the diacetate complex (phen)Pd(OAc)2 was synthesized as previously reported.43 Our attempts to generate the ditriflate complex
112 44 [(phen)Pd(CH3CN)2][OTf]2 by reaction of (phen)PdCl2 with AgOTf in acetonitrile were unsuccessful and yielded at least two species by 1H NMR. Attempted ligation of phenanthroline with (CH3CN)4Pd(OTf)2 also yielded multiple unidentified species, as
1 determined by H NMR. However, treatment of (phen)Pd(OAc)2 with triflic acid followed by two recrystallizations from CH3CN and Et2O generated the desired ditriflate compound as a single species. The observed 1H-NMR chemical shifts were consistent with that previously reported.44
The addition of [(phen)Pd(CH3CN)2][OTf]2 to (phen)Pd(OAc)2 in CD3CN yielded a complex 1H-NMR spectrum that was consistent with the formation of several new species. Slow diffusion of diethyl ether into a saturated solution of the mixture yielded orange crystals of the dimeric complex [(phen)Pd(OAc)]2[OTf]2 (8) which were analyzed by X-ray diffraction (Figure 4.2).
The structure of [(phen)Pd(OAc)]2[OTf]2 (8) is similar to that previously reported for its neocuproine analog 126 and other palladium phen dimers.45-47 A comparison of selected bond lengths and angles of 8 with the neocuproine analogue 1 is provided in
Section 4.5.3. Due to the absence of flanking methyl groups, the Pd-N bond lengths of the dimeric phen complex 8 are shorter than those for the neocuproine dimer 1. The O-
Pd-O bond angles in 8 (91.26(15) and 90.67(18)) are closer to the ideal 90º for a square- planar geometry than those found in 1 (84.17(9) and 81.85(8)), and the Pd-Pd distance in
8 (2.8548(9) Å) is shorter than that of 1 by 0.1 Å and in the range of that proposed for weak Pd-Pd interactions.47-50
113
Figure 4.2. X-ray crystal structure of [(phen)Pd(OAc)]2[OTf]2 (8) with ellipsoids drawn at 50% probability. The unit cell contains two dimeric complexes. An interdimer Pd-Pd distance of 3.0996(11) was observed. The second dimeric unit, hydrogen atoms, triflate counter ions, and solvent molecules have been omitted for clarity. Select bond lengths (Å) and angles (deg): Pd1-Pd2, 2.8548(9); Pd1-O1, 2.007(3); Pd1-O3, 2.021(4); Pd1-N1, 2.014(4); Pd1-N2, 2.000(4); Pd2-O2, 2.013(4); Pd2-O4, 2.016(4); Pd2-N3, 2.034(4); Pd2-N4, 2.032(4); O3-C27, 1.286(7); C27-O4, 1.299(7); O1-C25, 1.282(6); C25-O2, 1.308(7); O1-Pd1-O3, 91.26(15); O2-Pd2-O4, 90.67(18); O1-Pd1-N1, 93.38(16); and N1-Pd1-N2, 82.21(16).
Despite a previous report describing (btfm-phen)Pd(OAc)2 as an effective alcohol oxidation catalyst at 80 °C in DMSO/water,18 our attempts to prepare (btfm- phen)Pd(OAc)2 were unsuccessful; we were unable to observe complexation of this ligand to Pd(OAc)2 in either CH3CN or toluene/CH2Cl2 (see Section 4.5.2).
The less sterically demanding 2-(trifluoromethyl)-4-methyl-1,10-phenanthroline
(tfmm-phen) 6 could be readily complexed to palladium by stirring an equimolar solution of 6 and Pd(OAc)2 in toluene/CH2Cl2 to yield (tfmm-phen)Pd(OAc)2 9 as a red solid.
Complex 9 is only sparingly soluble in chloroform, dichloromethane, and acetonitrile, but
114 exhibits higher solubility in 1,1,2,2-tetrachloroethane (TCE). 1H-NMR spectroscopy in d2-TCE showed a single species with two inequivalent acetate ions.
The dicationic analogue [(tfmm-phen)Pd(NCCH3)2][OTf]2 10 was prepared by addition of [(CH3CN)4Pd][OTf]2 to 6; preparation by addition of triflic acid to 9 did not yield 10 cleanly. Slow diffusion of Et2O into a solution of 10 in acetonitrile yielded crystals suitable for X-ray diffraction (Figure 4.3). The Pd-N1 bond (2.073(3)) is elongated compared to Pd-N2 (1.984(3)), and it is also slightly longer than previously observed Pd-N bond lengths in analogous dicationic phenanthroline-based systems.26,44
N NCCH3 Pd
N NCCH3
H3C CF3
10
Figure 4.3. X-ray crystal structure of [(tfmm-phen)Pd(CH3CN)2][OTf]2 (10) with ellipsoids drawn at 50% probability. Hydrogen atoms and triflate counter ions are omitted for clarity. Select bond lengths (Å) and angles (deg): Pd(1)-N(1), 2.073(3); Pd(1)-N(2), 1.984(3); Pd(1)-N(3), 2.000(3); Pd(1)-N(4), 1.992(3); N(2)-Pd(1)-N(4), 93.48(11); N(2)- Pd(1)-N(3), 175.77(11); N(4)-Pd(1)-N(3), 82.58(11); N(2)-Pd(1)-N(1), 81.70(10); C(1)- N(1)-Pd(1), 133.6(2); and C(14)-N(2)-Pd(1), 127.2(2).
Conproportionation of 9 and 10 in a 1:1 molar ratio produced a complex 1H-NMR spectrum consistent with the generation of several new cationic species; no residual 9 or
115 10 was detectable after mixing. The complexity in the 1H-NMR spectrum is likely a consequence of monomer-dimer equilibria26 of a variety of stereoisomeric dimers of the asymmetric phenanthroline Pd complexes.
The Pd(OAc)2 complex of o-difluorophenyl-phenanthroline (odfp-phen) was prepared by mixing Pd(OAc)2 with odfp-phen in dry toluene/CH2Cl2 to yield (odfp- phen)Pd(OAc)2 11 as a yellow solid. A half-equivalent of CH2Cl2 remained even after long periods under high vacuum, but the identity of 11 was established by 1H- and 13C-
NMR spectroscopy and combustion analysis (as the CH2Cl2-solvate).
The diacetate complex 11 was crystallized over several days from a saturated acetonitrile solution, and analysis of the X-ray structure revealed that the new phenoxide complex 12 had formed (Figure 4.4). Complex 12 was fully characterized by 1H- and
13C-NMR spectroscopy, both of which were consistent with substitution of one of the fluorine atoms with oxygen. Solutions of 11 in rigorously dried acetonitrile did not convert to 12, suggesting that the substituted oxygen in 12 arises from adventitious water.
116
Figure 4.4. X-ray crystal structure of 12 with ellipsoids drawn at 50% probability. Hydrogen atoms have been omitted for clarity. Select bond lengths (Å) and angles (deg): Pd(1)-N(2), 1.987(2); Pd(1)-N(1), 1.998(2); Pd(1)-O(1), 1.9521(19); Pd(1)-O(2), 2.0340(19); N(2)-Pd(1)-N(1), 84.05(9); O(1)-Pd(1)-N(2), 92.99(8); O(1)-Pd(1)-O(2), 91.36(8); N(1)-Pd(1)-O(2), 91.48(9); O(1)-Pd(1)-N(1), 176.45(9); N(2)-Pd(1)-O(2), 174.49(8); O(1)-C(18)-C(13), 127.0(2); O(1)-C(18)-C(17), 113.7(2); and Pd(1)-O(1)- C(18)-C(13), 4.1(4).
The formation of 12 by hydrolysis of 11 was confirmed by the addition of water to a dry acetonitrile solution of 11. After three days, crystals of 12 could be isolated in
36% yield. We propose that adventitious water displaces one of the acetate ligands to generate the Pd-hydroxo complex 13, which undergoes intramolecular nucleophilic aromatic substitution of fluoride, yielding 12 (Scheme 4.5). Alternatively, reversible dissociation of one of the acetates could generate a cationic Pd complex, which could activate one of the ortho-fluorines toward intermolecular SNAr by hydroxide. We currently disfavor the second hypothesis, as similar substitution by methoxide was not observed when methanol was used as a solvent. The transformation of 11 to 12 may have
117 implications for carbon-fluorine bond activation, such as those reported for Pt(II)51-53 and
Ni(0)54-58 complexes and is similar to substitution reactions observed by Usòn et al.59
N +H2O N SNAr N Pd N -HOAc Pd N -HF N AcO AcO AcO Pd AcO F F HO F F O F
11 13 12 Scheme 4.5. Proposed mechanism for palladium-mediated fluoride substitution by adventitious water
Initial attempts to prepare the palladium ditriflate complex of odfp-phen (14) from a mixture of (CH3CN)2PdCl2 and two equiv of AgOTf in acetonitrile led to the cationic
Pd chloride complex [(odfp-phen)Pd(NCCH3)Cl][OTf] 15 in 10-20% yield (Scheme
4.6). Crystallization of 15 from acetonitrile afforded crystals suitable for X-ray analysis
(Figure 4.5). The asymmetrically substituted complex 15 exists as a single isomer in the solid state. The coordinated chloride is trans to the nitrogen bearing the o-difluorophenyl substituent. The o-difluorophenyl substituent lies out of plane of the phenanthroline ligand by 58.8(5)º as evidenced by the N2-C12-C13-C18 torsion angle.
2 2 OTf N N (CH CN) PdCl N Ag(OTf) 3 2 2 Ligand 7 Cl Pd CH CN Pd N + 3 CH CN, 2 Ag(OTf) N F F 3 N CH3CN reflux F F C C CH3 CH3 15 14 Scheme 4.6. Synthesis of 14 and 15
118
Figure 4.5. X-ray crystal structure of 15 with ellipsoids drawn at 50% probability. Hydrogen atoms and a triflate counter ion have been omitted for clarity.
In contrast to 11, the monocationic complex 15 did not hydrolyze in wet CD3CN and showed no sign of fluoride displacement, even after heating to 70 °C for 7 days. The higher hydrolytic stability of 15 relative to 11, suggests that the presence of the ancillary acetate ligand in 11 facilitates the displacement of the fluoride to give 12.
Treatment of 15 with excess AgOTf in refluxing acetonitrile for several days afforded the dicationic bis-acetonitrile complex 14. (Scheme 4.6). The dicationic 14 was less hydrolytically stable than 15, and over the course of several days in wet acetonitrile, the complex degraded into several unidentifiable species.
Combination of 11 and 14 in a 1:1 molar ratio in dry CD3CN yielded several new species. Analysis of the 1H-NMR spectrum was complicated by the increased number of possible stereoisomers combined with the known solution equilibria;26 analogous to the mixing of 9 and 10, no residual peaks attributable to 11 or 14 were observed after mixing.
119 4.2.3 Catalytic aerobic alcohol oxidation
To assess the influence of the ligands on the activity of the Pd complexes for aerobic alcohol oxidation, we investigated the oxidation of 2-heptanol in acetonitrile in air at room temperature, as previously described (Figure 4.6 and Table 4.1).26 For these experiments, the cationic Pd complexes were either generated in situ (from 9/10 and
11/14) or introduced as the isolated Pd dimers (complexes 1 and 8) at a Pd concentration of 1.5 mM (3 mol % relative to 2-heptanol).
As previously reported, the dimeric complex 1 exhibits a fast initial rate (Figure
4.6, Table 4.1) but the activity drops off rapidly and the maximum conversion under these conditions is only 36%.26 As previously discussed, this behavior was attributed to competitive ligand oxidation, generating the inactive carboxylate 3 (Scheme 4.1).
The cationic phenanthroline dimer 8, lacking oxidizable methyl groups, was completely inactive under these conditions (Table 4.1, entry 2). We attribute the lack of activity to the formation of the stable of µ-hydroxo bridged dimer 16.26,60 Addition of water to dimer 8 leads to the formation of 16 (identified by 1H-NMR and independent synthesis60) along with several other unidentified species (Scheme 4.7). Complexes analogous to 16 have been proposed as resting states by Sheldon and coworkers16,18 in aerobic alcohol oxidations carried out at higher temperatures. It has been previously observed that increased steric bulk or higher temperatures are required with
(phen)Pd(OAc)2 to destabilize the µ-hydroxo intermediates and achieve greater alcohol- oxidation activity.18 This hypothesis provides a rationale for the poor activity of 8; if the
µ-hydroxo bridged dimer 16 is formed in the course of the catalytic cycle, this dimer would likely be inactive at the low temperatures investigated here. These results imply 120 that substitution at the 2- and/or 9- positions of phenanthroline is important to retain activity at low temperature, but these substituents must also resist competitive oxidation.26
2 2 2 OTf 2 OTf N Ac N H N O 2 • H2O O Pd 2 Pd Pd + 2 • AcOH O N N H N
8 16
Scheme 4.7. Formation of µ-hydroxo-bridged dimers
Figure 4.6. Reaction progress for the palladium-catalyzed aerobic oxidation of 2- heptanol to 2-heptanone using 1( ), 8( ), 9( ), 10( ), 9/10( ), 11/14( ) in acetonitrile; in all cases, 3 mol % of Pd was employed.
The introduction of a single trifluoromethyl group to the 2-position of phenanthroline was expected to destabilize dimers analogous to 16, while preventing the harmful degradation pathways caused during oxygen reduction. A 1:1 mixture of complexes 9 and 10 catalyzed the oxidation of 2-heptanol at room temperature to afford
121 2-heptanone in 67% yield after 24 h (entry 3). In accordance with our previous observations,26 the diacetate complex 9 or ditriflate complex 10 under identical conditions provided only 6% and 3% conversion, respectively, after 24 h. Catalysts derived from a mixture of 9/10 reached a turnover number of 22 after 24 h, which is nearly twice that previously observed with 1. Full conversion of 2-heptanol at 3 mol % catalyst loading was not obtained even at extended reaction times. Though a sub- stoichiometric amount of palladium black was observed after 24 h of reaction, a sharp decrease in the rate of alcohol oxidation preceded Pd precipitation by several hours, indicating that Pd-black formation in itself may not be responsible for the observed loss in activity. Analysis of the reaction mixture after 24 hours by ESI-MS yielded the expected mass of the ligand with no evidence of ligand oxidation. Attempts to prolong catalyst lifetime by introducing additional ligand into the reaction mixture (0.5 equiv relative to Pd) resulted in slower TOFs but similar levels of conversion (61% yield at 72 h). Thus, while the activity and longevity of catalysts derived from the tfmm-phen ligand
6 are superior to that of catalyst 1 derived from neocuproine, the modest turnover numbers observed implicate additional catalyst deactivation pathways and highlight the need for a better understanding of catalyst inactivation processes to enable further optimization.
Investigation of a 1:1 mixture of 11 and 14 for the aerobic oxidation of 2-heptanol in dry acetonitrile (entry 4) revealed initial rates of alcohol oxidation comparable to 1, but the reaction rate quickly dropped off (Figure 4.6, diamonds), and only 2.5 turnovers were achieved after 24 h. Precipitation of the resulting palladium complex was accomplished by addition of diethyl ether, and the 1H-NMR spectrum of the precipitate exhibited
122 splitting consistent with the formation of a chelated phenoxide complex, analogous to 12.
Two resonances were observed by 19F-NMR, consistent with the presence of a triflate counterion and a single fluoride on the ligand, suggesting the catalyst degradation product is the cationic complex 17. It is worth noting that fluorine substitution by 2- heptanol was not observed. The formation of tridentate complex 17 inactivates the Pd in a manner that is reminiscent of the inactivation of 1 by intramolecular ligand oxidation to tridentate complex 3,26 which exhibited negligible activity compared to its bidentate precursor. Addition of molecular sieves to the 2-heptanol-oxidation reaction mixture catalyzed by 11/14 increased the TON of the catalyst to 4.5, suggesting that water produced during oxygen reduction can be scavenged competitively with fluoride displacement.
17
123 Table 4.1. Efficiency of various cationic palladium complexes for the aerobic oxidation of 2-heptanol at room temperature
a b -1 c entry complex yield (%) TON TOFi (h ) 1 1 36 12 78 2 8 trace - - 3 9/10 67 22 21 4 11/14 8 2.5 60 a after 24 h reaction time; b TON = moles ketone/mole Pd c initial turnover frequency (mol ketone/(mol Pd•h))
4.3 Conclusion
Cationic Pd-acetate complexes ligated by neocuproine or 2-substituted phenanthroline ligands are active for the aerobic oxidation of secondary alcohols under mild conditions (room temperature, 1 atm air). The dimeric complex
[(neocuproine)Pd(OAc)]2[OTf]2 1 exhibits high initial turnover frequency for the aerobic oxidation of 2-heptanol; however, oxidative degradation of the ligand leads to rapid deactivation and low turnover numbers.26 The synthesis of two new phenanthroline ligands, bearing CF3 or 1,5-difluorophenyl substituents, was carried out in an effort to generate more oxidatively robust Pd catalysts. Pd complexes derived from these ligands were structurally characterized and their activity in the aerobic oxidation of 2-heptanol was compared to that of complex 1. The inactivity of the [(phen)Pd(OAc)]2[OTf]2 8 implies that substituents at the 2- and/or 9- positions of the phen ligand are important for catalytic activity. The higher catalyst lifetimes of Pd complexes bearing the 2-CF3-
124 substituted ligand 4-methyl-2-(trifluoromethyl)-1,10-phenanthroline (tfmm-phen) reveal that inhibiting ligand oxidation can lead to more robust catalysts for aerobic alcohol oxidation. Nevertheless, the modest turnover numbers observed (22 mol ketone/mol Pd) are less than those observed with Pd carbene complexes (approx. 100 TO).21 While we observe no evidence for ligand oxidation, other decomposition pathways may limit the lifetimes of catalysts derived from tfmm-phen. Pd complexes derived from the 2-(o- difluorophenyl)-1,10-phenanthroline (odfp-phen) ligand exhibit poor catalyst lifetimes as a consequence of facile substitution of the ortho-fluoro substituent by water (a byproduct of the reaction). These results provide insights into the structure/property relationships of Pd oxidation catalysts that may guide future design of oxidatively robust catalysts for the aerobic oxidation of alcohols.
4.4 Preliminary Results for Future Directions
Following our initial studies carried out in Section 4.2, the synthesis of several new ligands was carried out in hopes that they might also serve as oxidatively resistant ligands for palladium catalyzed aerobic alcohol oxidation. These ligands were screened by in-situ formation of the ligated palladium complex (L-L)Pd(OAc)2 in acetonitrile followed by the addition triflic acid (1 equiv relative to Pd) to generate the cationic derivative and subsequently used for the aerobic oxidation of 2-heptanol (see Section
4.5.4). A summary of preliminary results of our ligand screen is given in Figure 4.7.
125 Cl MeO CF3 N N N
N N N
TON 8 TON 0.2 TON 15.7 TOFi 77 TOFi 2.8 TOFi 2.3
Cl F OMe
N Cl N N F
N N N
TON 3.2 TON < 1 TON 2.4 TOFi 100 TOFi 0 TOFi 16
Figure 4.7 Screening of additional ligands for the aerobic oxidation of 2-heptanol using in situ generated palladium catalysts.
Of the ligands screened 2-(trifluoromethyl)-4-(methoxy)-1,10-phenanthroline showed the most promising results, which were comparable to those observed with 2-
(trifluoromethyl)-4-methyl-1,10-phenanthroline. Placement of a chloride in the 2- position resulted in an inactive catalyst providing a TON of 0.2. Our attempts to provide
more robust phenyl derivatives were generally unsuccesful. Attempts to discourage SNAr by addition of an electron donating methoxy group were unsucessful and provided similar
TON to those observed with a mixture of 11/14. The chloro-substituted analog provided rapid initial rates but became inactive after approximately 3 turnovers. Analysis by ESI-
MS suggests that deactivation occurs through the formation of a tridentate complex which could form upon oxidative addition of the aryl-chloride. Finally, the mesityl derivative proved completely inactive for alcohol oxidation.
126 Taken as a whole, these results provide key insights that will inform the design of next-generation oxidatively robust palladium catalysts for aerobic alcohol oxidation. A future area of exploration might include replacement of the 2-trifluoromethyl group in
9/10 with a difluoromethylsulfonate group (i.e., an oxidatively robust pendant counterion) to give difluoro(1,10-phenanthrolin-2-yl)methanesulfonate, which may be less susceptible to ligand dissociation. Yet another strategy could entail enlarging the N-
Pd-N bite angle of neocuproine, which is about 83°, by replacing its central benzenoid ring with a cyclopenta-1,3-diene ring; this could impart oxidative resistance to the ligand by increasing the distance between the oxidizable C-H bonds of the 2,9-dimethyl groups and the palladium-bound hydroperoxide implicated in the oxidative degradation pathway
(see Scheme 4.1). Given the substantial innovation in palladium-catalyzed aerobic oxidation reactions over the last decade, there is every reason to be optimistic that a new generation of high-TON palladium catalysts that exhibit resistance to partially reduced oxygen species is forthcoming.
4.5 Experimental Section
Materials
Unless otherwise stated, solvents were purchased from Sigma-Aldrich or Fisher
Chemical and used as received. Deuterated solvents were purchased from Cambridge
Isotopes and used as received. 7-methylquinoline was purchased from TCI America and used without further purification. 8-aminoquinoline was purchased from Sigma-Aldrich
43 26 35 36 and used without further purification. (1,10-phenanthroline)Pd(OAc)2, ,1, 4, and 5 were prepared as previously described. [(CH3CN)4Pd][OTf]2 was prepared in analogy to
127 61 [(CH3CN)4Pd][BF4]2 and was used in situ. 8-amino-7-quinolinecarbaldehyde was prepared by the route established by Thummel and coworkers.40 2-chloro-1,10- phenanthroline was prepared by the route established by Halcrow and Kermack.62 All
NMR spectra were acquired on Varian Inova 300, Mercury 400, or Inova 500 MHz spectrometers. 1H- and 13C-NMR spectra are referenced to the solvent residual peaks.
19 F-NMR spectra are referenced to CF3CO2H (-76.55 ppm) or free triflate anion (-77.6 ppm). For splitting patterns, “s” refers to single, “d” refers to doublet, “t” refers to triplet, and “q” refers to quartet.
4.5.1 Synthesis of compounds
2-(trifluoromethyl)-4-methyl-1,10-phenanthroline (6): Enone 536 (6.86 g, 49 mmol) was added over 5 h to a stirred solution of 8-aminoquinoline (4.24 g, 29 mmol) and sodium iodide (0.043 g, 0.28 mmol) in 70% sulfuric acid (10.6 mL) at 110 °C. The reaction was allowed to continue for an additional hour and was then cooled to room temperature. The dark orange-red mixture was poured into 1 M Na2CO3 (123 mL) and extracted with CH2Cl2 (3 " 100 mL). The combined organics were extracted with 12 M
HCl (5 " 50 mL). The aqueous, acidic layer was neutralized using 3 M NaOH/1 M
Na2CO3 and extracted with CH2Cl2 (3 " 100 mL). The compound was purified by silica- gel column chromatography (CH2Cl2/MeOH; polarity was slowly increased until product eluted) yielding 6 (1.6 g, 21%) in reasonable purity as a brown solid. Further purification can be achieved by crystallization (CHCl3/hexanes) or by column chromatography
1 (acetone:hexanes, 1:4) affording 6 as a white solid. H-NMR (CDCl3) #: 9.29 (dd, J =
1.7, 4.6 Hz, 1H), 8.30 (dd, J = 1.7, 8.0 Hz, 1H) 8.07 (d, J = 9.2 Hz, 1H), 7.95 (d, J = 9.2
128 13 Hz, 1H), 7.84 (s, 1H), 7.69 (dd, J = 4.6, 8.0 Hz, 1H), 2.90 (s, 3H). C-NMR (CDCl3) #:
2 4 151.3, 147.4 (q, JCF = 34.4 Hz), 146.9, 146.2, 145.4, 136.2, 129.5 (q, JCF = 1.0 Hz),
1 3 128.8, 128.5, 123.6, 122.0, 121.8 (obsd d, JCF = 275.5 Hz), 120.19 (q, JCF = 2.2 Hz),
19 19.6 F-NMR (CDCl3) #: -65.5 (s) Anal Calcd for C14H9F3N2: C, 64.12; H, 3.46; N,
10.68. Found: C, 62.88; H, 3.10; N, 10.36. HRMS-(ES+) Calcd for C14H10F3N2 [M+H]:
263.0796. Found: 263.0795.
2-(2!,6!-difluorophenyl)-1,10-phenanthroline (7): This compound was prepared by adaptation of the procedure reported by Thummel and coworkers40 for the synthesis of
2-phenyl-1,10-phenanthroline. A saturated solution of ethanolic potassium hydroxide (0.5 mL) was added dropwise to a solution of 8-amino-7-quinolinecarbaldehyde (0.231 g,
1.34 mmol) and 2!,6!-difluoroacetophenone (0.209 g, 1.34 mmol) in absolute ethanol (15 mL) under an atmosphere of N2, and the mixture was refluxed for 15 h. The solution was diluted with water (30 mL) and extracted with CH2Cl2 (3 " 30 mL). The combined organic extracts were washed with H2O (2 " 20 mL) and dried (MgSO4). Flash chromatography on alumina using CH2Cl2 as the eluent and subsequent recrystallization with EtOAc/hexane provided the title compound as an off white solid (0.228 g, 58%).
1 H-NMR (CDCl3) #: 9.22 (dd, J = 1.3, 3.8 Hz, 1H, H9), 8.34 (d, , J = 7.6 Hz, 1H, H4),
8.25(dd, J = 1.3, 7.2 Hz, 1H, H7) , 7.84(s, 2H, H5-6), 7.79 (d, J = 7.6 Hz, 1H, H3), 7.63
(dd, J = 3.8, 7.2 Hz, 1H, H8) 7.38 (m, J = 1.8, 6.7, 6.3 Hz, 1H, p-Ph), 7.04 (m, J = 4.0,
13 1 3 6.5 Hz, 2H, m-Ph); C-NMR (CDCl3) #: 161.0 (dd, JCF = 250.5 Hz, JCF = 7.1 Hz),
3 150.8, 150.1, 146.4, 146.4, 136.3, 136.1, 130.3 (t, JCF = 10.3 Hz), 129.0, 127.9, 127.2,
2 2 19 126.3, 125.4, 123.1, 119.0 (t, JCF =19.2 Hz), 111.8 (m, JCF = 24.2 Hz); F-NMR
129 (CDCl3) # -113.0 (t, J = 6.5 Hz ). ESI-MS (+): 293 [M+H] Anal Calcd for C18H10F2N2: C,
73.97; H, 3.45; N, 9.58. Found: C, 73.71; H, 3.38; N 9.39.
[(1,10-phenanthroline)Pd(NCCH3)2][OTf]2: Triflic acid (1.5 mL, 0.33 M, 2 eq.) was added dropwise to a stirred solution of (phen)Pd(OAc)2 (100 mg, 0.247 mmol) in dry acetonitrile (10 mL). Dry diethyl ether was added to induce precipitation of the resulting complex, which was recrystallized three times from CH3CN/Et2O to achieve the desired compound in high purity (22 mg, 0.033 mmol, 13%). While this compound has been previously reported,44 the 1H-NMR data indicated two species in a 1:1 mixture. We have isolated a single species that matches only one of those previously reported. 1H-NMR
(CD3CN) #: 8.96 (dd, J = 1.2, 8.4 Hz, 2H), 8.85 (dd, J = 1.2, 5.5 Hz, 2H), 8.24 (s, 2H),
13 8.04 (dd, J = 5.5, 8.4 Hz, 2H), 1.96 (s, 6H). C-NMR (CD3CN) #: 153.2, 148.3, 143.5,
1 132.5, 129.1, 127.4, 122.0 (q, JCF = 320.7Hz). Anal Calcd for C18H14F6N4O6PdS2: C,
32.42; H, 2.12; N, 8.40. Found: C, 32.31; H, 1.88; N, 8.22.
[(1,10-phenanthroline)Pd(OAc)]2[OTf]2 (8): To a solution of (phen)Pd(OAc)2
(10.7 mg, 0.026 mmol) in acetonitrile (2 mL), was added [(phen)Pd(NCCH3)2][OTf]2
(17.7 mg, 0.026 mmol) and the solution was stirred for 15 min. Removal of solvent under vacuum afforded 8 quantitatively. High quality crystals of the dimer were obtained
by slow vapor diffusion of Et2O into acetonitrile. Anal Calcd for C30H22F6N4O10Pd2S2: C,
36.42; H, 2.24; N, 5.66. Found C, 35.87; H, 1.56; N, 6.09.
130 (4-methyl-2-(trifluoromethyl)-1,10-phenanthroline)Pd(OAc)2 (9): Pd(OAc)2
(0.0856 g, 0.38 mmol) was added to a stirred solution of 6 (0.100 g, 0.38 mmol) in acetone (10 mL). The solution was allowed to stir overnight, during which time a fine, dark-red precipitate formed. The red solid was isolated by vacuum filtration and rinsed repeatedly with acetone, Et2O, and finally dried under high vacuum to give 9 (0.138 g,
1 74%). H-NMR (Cl2CDCDCl2) #: 8.62 (d, J = 6.8 Hz, 2H, H7 + H9), 8.16 (d, J = 9.1 Hz,
1H, H5), 8.11 (d, J = 9.1 Hz, 1H, H6), 8.07 (s, 1H, H3), 7.90 (m, 1H, H8), 2.98 (s, 3H,
13 Me), 2.17 (s, 3H, OAc), 1.99 (s, 3H, OAc). C-NMR (Cl2CDCDCl2) # : 178.8, 178.0,
152.4, 150.2, 147.2, 146.82, 146.78, 138.8, 130.7, 130.3, 128.7, 125.5, 124.2, 123.5, 23.2,
19 23.1, 19.7 F-NMR (Cl2CDCDCl2) # : -61.2, -61.4, -62.2, -65.5 Anal Calcd for
C18H15F3N2O4Pd: C, 44.42; H, 3.11; N, 5.76. Anal Calcd for C18H15F3N2O4Pd · (H2O): C, 42.83; H, 3.39; N, 5.55. Found: C, 42.54; H, 2.88; N, 5.88. HRMS (ES+) Calcd for
C16H12N2O2F3Pd [M-OAc]: 426.9886 Found: 426.9887.
[(4-methyl-2-(trifluoromethyl)-1,10-phenanthroline)Pd(NCCH3)2][OTf]2 (10):
Trifluoromethyl ligand 6 (0.109 g, 0.42 mmol) was added to a stirred solution of
[CH3CN)4Pd][OTf]2 (0.580 g, 1.02 mmol) in acetonitrile (4 mL), during which time the solution became orange. The solution was filtered and Et2O was carefully layered onto the filtrate (i.e. without mixing) to induce crystallization. The resulting needlelike crystals were recrystallized two times from acetonitrile/Et2O yielding pure 10 (0.11 g,
35%). The mother liquor from the two previous recrystallizations was concentrated and treated with Et2O to yield an additional 0.045 g (49% combined yield) of the title
1 compound. H-NMR (CD3CN) # : 9.48 (dd, J = 1.2, 5.6 Hz, 1H), 8.86 (dd, J = 1.2, 8.4
131 Hz, 1H), 8.37 (d J = 9.0 Hz, 1H), 8.36(m, 1H), 8.28 (d, J = 9.0 Hz, 1H), 8.01 (dd, J = 5.6,
13 8.4 Hz, 1H), 3.04 (s, 3H). C-NMR (CD3CN) # : 156.9, 153.6, 142.7, 133.0, 132.8,
19 130.3, 127.1, 125.8, 125.4, 123.2, 121.9, 120.6, 119.7, 19.9. F-NMR (CD3CN) # : -
59.83, -77.47. ESI-MS (+): 665 [M – 2CH3CN] 517 [M - 2CH3CN - OTf]. Anal Calcd for C20H15F9N4O6PdS2: C, 32.08; H, 2.02; N, 7.48. Found: C, 31.95; H, 1.76; N, 6.51
+ HRMS (ES+) Calcd for C14H12N2F3Pd [LPdH ]: 368.9831 Found: 368.9838.
Conproportionation of 9 and 10: Equal molar amounts of the diacetate 9 and the ditriflate 10 complexes were combined in acetonitrile, to generate a mixture of stereoisomers (Figure 4.11). A similar spectrum is obtained upon addition of one equivalent of triflic acid to the diacetate complex 9 in acetonitrile. The resulting solution exhibits similar catalytic activity to that of an equal molar solution prepared by the conproportionation method; the initial rate of oxidation using 9/triflic acid is slightly retarded but the final conversion is equal to that obtained by conproportionation within the limits of experimental uncertainty. The complexity associated with the solution speciation and isomers is highlighted in Figure 4.11, which shows the aromatic regions of the 1H-NMR spectra of 9, 10 and their mixture.
(2-(2!,6!-o-difluorophenyl)-1,10-phenanthroline)Pd(OAc)2 (11): A solution of palladium diacetate (0.076 g, 0.342 mmol) in CH2Cl2 (1.2 mL) was added to a stirred solution of ligand 7 (0.100 g, 0.342 mmol) in toluene (6.2 mL). The solution was allowed to stir for 24 h at room temperature, during which time a yellow solid formed.
The mixture was poured into petroleum ether (5 mL) to induce further precipitation, and
132 the yellow precipitate was isolated by vacuum filtration and dried under high vacuum
(0.168 g, 82%). When complex 11 is prepared by this method, it contains half of an
1 equivalent of CH2Cl2 even after extensive drying, as observed by H-NMR spectroscopy.
Due to its low solubility in CDCl3, chemical shifts are also reported in anhydrous
1 CD3CN. H-NMR (CDCl3) #: 8.67 (dd, J = 1.2, 5.3 Hz, 1H), 8.66 (d, J = 9.2 Hz, 1H),
8.57 (dd, J = 1.2, 8.2 Hz, 1H), 8.48 (d, J = 9.2 Hz, 1H), 8.01 (d, J = 8.8 Hz, 1H), 7.91 (d,
J = 8.8 Hz, 1H), 7.85 (dd, J = 5.3, 8.2 Hz, 1H), 7.25-7.11 (m, 2H), 6.48 (ddd, J = 1.5, 7.6,
1 14.4 Hz, 1H) H-NMR (CD3CN) #: 8.68 (dd, J = 1.3, 8.3 Hz, 1H), 8.66 (d, J = 8.3 Hz,
1H), 8.28 (dd, J = 1.3, 5.3 Hz, 1H), 8.08 (d, J = 8.8 Hz, 1H), 8.05 (d, J = 8.8 Hz, 1H),
7.76 (dt, J = 1.0, 8.5 Hz, 1H), 7.73 (dd, J = 5.3,8.2 Hz, 1H), 7.75 (m, 1H), 7.15 (t, J = 8.2
13 Hz, 2H), 1.99 (2, 3H), 1.28 (s, 3H). C-NMR (CD3CN): 177.5 (OAc), 177.4 (OAc),
1 3 161.0 (dd, JCF = 249.5 Hz, JCF = 6.23 Hz), 154.4, 150.7, 148.3, 147.3, 140.6, 140.5,
3 2 133.7 (t, JCF = 10.7 Hz), 131.3, 130.57, 130.5, 128.75, 128.3, 126.2, 115 (t, JCF = 19.2
2 19 Hz), 112.5 (m, JCF = 22.7 Hz), 23.2 (OAc), 22.1 (OAc). F-NMR (CD3CN): -111.6 (t, J
= 7.1 Hz). ESI-MS(+): 398.9 [M-2OAc+H], 442.9 [M-2OAc+O2CH], 456.9 [M-OAc].
Anal Calcd for C22H16F2N2O4Pd · !(CH2Cl2): C, 48.32; H, 3.06; N, 5.01. Found: C, 48.06; H, 3.33; N, 4.80.
Synthesis of phenoxide complex 12: Diacetate 11 (20 mg, 0.038 mmol) was dissolved in acetonitrile (5 mL). The solution was filtered into a 20-mL vial through a plug of glass fiber to remove any undissolved material. Two drops of water were added to the solution, and the vial was capped, shaken, and allowed to sit at room temperature for 3 d. Crystals became visible after 24 h but continued to grow when allowed to sit for
133 additional time. X-ray-quality, needle-shaped crystals were isolated by removal of the mother liquor with a pipette and repeated washing of the crystals with Et2O. Upon drying under high vacuum, complex 12 was obtained in 36% yield (6.3 mg). Additional crystals were obtained by concentration of the mother liquor. Complex 12 is nearly insoluble in
1 acetonitrile and DMSO but is sparingly soluble in chloroform. H-NMR (CDCl3) #: 8.74
(d, J = 5.4 Hz, 1H, H9), 8.61 (d, J = 9.1 Hz, 1H, H4), 8.55 (d, J = 8.2 Hz, 1H, H7), 8.49 (d,
J = 9.1 Hz, 1H, H3), 8.00 (d, J = 8.7 Hz, 1H, H5), 7.88 (d, J = 8.7 Hz, 1H, H6), 7.84 (dd, J
= 5.4, 8.2 Hz, 1H, H8), 7.20 (m, 1H, HAr), 7.11(d, J = 8.7 Hz, 1H, HAr), 6.46 (dd, J = 7.8,
1 14.3 Hz, 1H, HAr), 2.30 (s, 3H, OAc). H-NMR (d6-DMSO) #: 8.99 (dd, J = 1.4, 8.3 Hz,
1H), 8.91 (d, J = 9.2 Hz, 1H), 8.66 (dd, J = 1.8, 9.2 Hz, 1H), 8.62 (dd, J = 1.4, 5.2 Hz,
1H), 8.28(d, J = 8.6 Hz, 1H), 8.24 (d, J = 8.6 Hz, 1H), 8.11 (dd, J = 5.2, 8.3 Hz, 1H), 7.29
(dd, J = 8.0, 15.5 Hz, 1H), 6.90 (d, J = 8.7 Hz, 1H), 6.60 (dd, J = 8.0, 14.2 Hz, 1H),
13 1 2.04(s, 3H) C-NMR (CDCl3) #: 179.1 (OAc), 166.6, 162.5 (d, JCF = 251.5 Hz), 150.9,
2 149.6, 146.3, 145.7, 138.4, 136.0, 132.3 (d, JCF =14.8 Hz), 129.3, 127.2, 126.7, 126.5,
2 19 126.0, 124.7, 119.4, 102.7 (d, JCF = 26.0 Hz), 24.08 (OAc). F-NMR (CDCl3) #: -107.9
(dd, JHF = 14.3, 6.5 Hz) ESI-MS(+): 394.8 [M-OAc], 412.8 [M-OAc+H2O], 792.8 [2M-
2OAc+H], 836.8 [2M-OAc]. Anal Calcd for C20H13FN2O3Pd: C, 52.82; H, 2.88; N, 6.16.
Found: C, 52.47; H, 2.55; N, 5.90.
[(2-(2!,6!-o-difluorophenyl)-1,10-phenanthroline)Pd(CH3CN)2][OTf]2 (14): In a flame-dried flask equipped with a reflux condenser was added PdCl2 (57 mg, 0.32 mmol, 1.1 eq) and odfp-phen 7 (87 mg, 0.30 mmol, 1.0 eq). Dry acetonitrile (5 mL) was added to the solids and the solution was refluxed overnight under N2 atmosphere, during
134 which time a yellow solid was formed. The suspension containing the yellow solid was treated with AgOTf (206 mg, 0.80 mmol, 2.7 eq) and was refluxed in the dark for 2 d.
Additional AgOTf (103 mg, 0.40 mmol) was added to the reaction mixture, and refluxing was continued in the dark for 1 d. At this time, the reaction flask contained a mixture of the 15 and 14 in a 2:1 ratio as observed by 1H-NMR spectroscopy. The solvent was evaporated to dryness, dissolved in dry acetonitrile (5 mL), filtered to remove residual
AgCl, and AgOTf (106 mg, 0.41 mmol, 1.4 eq) was added to the solution. The reaction mixture was heated an additional 4 d in the dark at 60 °C. The solvent- evaporation/dissolution/filtration process was repeated two more times, and in this manner, the ratio of 15 to 14 changed to 1:10. Finally, the solvent was evaporated to dryness, dry acetonitrile was added, and the solution was filtered through Celite. Diethyl ether was added to the solution until it became slightly turbid, and then the solution was cooled to -25 °C. Yellow crystals formed overnight, which were isolated by cannula
1 filtration and dried under high vacuum yielding 14 (44 mg, 19%). H-NMR (CD3CN) #:
9.03 (d, J = 8.4 Hz, 1H, H4), 8.99(dd, J = 1.1, 8.3 Hz, 1H, H9), 8.85 (dd, J = 1.1, 5.5 Hz,
1H, H7), 8.28 (s, 2H, H5-6), 8.10 (d, J = 8.4 Hz, 1H, H3), 8.04 (dd, J = 5.5, 8.3 Hz, 1H,
13 1 H8), 7.78 (m, 1H, Arp), 7.35 (m, 2H, Arm) C-NMR (CD3CN) #: 160.8 (dd, JCF = 251.4
3 3 Hz, JCF = 5.6 Hz), 154.4, 154.1, 149.6, 148.8, 143.8, 143.7, 136.3 (t, JCF = 10.8
1 2 Hz),132.8, 132.1, 130.8, 129.5, 129.0, 127.3, 122.0 (q, JCF = 321.2 Hz), 116.2 (t, JCF =
2 19 18.6 Hz) , 113.5 (m, JCF = 20.5 Hz) F-NMR (CD3CN) #: -77.6, -110.7 (t, J = 7.5 Hz)
Exposure to high vacuum for prolonged periods of time resulted in partial loss of coordinated acetonitrile. Anal Calcd for C24H16F8N4O6PdS2: C, 37.01; H, 2.07; N, 7.19;
135 Found: C, 36.60; H, 1.83; N, 6.68. Anal Calcd for C22H13F8N3O6PdS2: C, 35.81; H, 1.78;
N, 5.69; Found: C, 35.56; H, 1.82; N, 6.10.
[(2-(2!,6!-o-difluorophenyl)-1,10-phenanthroline)Pd(CH3CN)Cl][OTf] (15):
Cationic-palladium precursor, (CH3CN)4Pd(OTf)2 (320 mg, 0.564 mmol), was prepared in-situ by addition of (CH3CN)2PdCl2 (0.1 g, 0.564 mmol) to AgOTf (0.289 g, 1.128 mmol) by analogy to previous reports.61 After removing AgCl by filtration, the precursor solution was added to a stirred solution of odfp-phen (148 mg, 0.508 mmol, 0.9 eq) in acetonitrile and allowed to stir 2 h, after which time the reaction mixture was concentrated and precipitated with Et2O. The resulting solid (0.161 g) contained an acetonitrile-insoluble portion (we attribute this to additional AgCl), which was removed by redissolving the precipitate in acetonitrile and filtering. The filtrate was evaporated to dryness and the complex was recrystallized twice by slow addition of Et2O to acetonitrile,
1 producing complex 15 in 10% yield (0.043 g). H-NMR (CD3CN) #: 9.43 (dd, J = 1.3
5.6 Hz, 1H, H9), 8.95 (d, J= 8.4 Hz, 1H, H4), 8.86 (dd, J = 1.3, 8.2 Hz, 1H, H7), 8.23 (d,
J = 1.5 Hz, 2H, H5-6), 8.08 (d, J = 8.4 Hz, 1H, H3), 7.99 (dd, J = 5.6, 8.2Hz, 1H, H8), 7.73
13 1 3 (m, 1H, Arp), 7.32 (m, 2H, Arm) C-NMR (CD3CN) #: 160.9 (dd, JCF = 250.6 Hz, JCF =
3 5.8 Hz), 153.1, 149.4, 148.5, 143.8, 143.6,142.4, 135.6 (t, JCF = 10.4 Hz) 132.4, 131.5,
1 2 130.5, 129.3, 128.6, 126.6, 122.0 (q, JCF = 320.9 Hz), 116.6 (t, JCF = 18.6 Hz), 113.41
2 19 (m, JCF = 23.1 Hz). F-NMR (CD3CN) #: -77.6, -111.2 (t, J = 7.2 Hz) Anal Calcd for
C21H13ClF5N3O3PdS C, 40.40; H, 2.10; Cl, 5.68; N, 6.73 Found: C, 40.29; H, 1.71; Cl, 6.07; N, 6.53.
136 Conproportionation of 11 and 14: An equal molar quantity of the ditriflate complex 14 (20 mg, 0.0257 mmol) and the diacetate complex 11 (13 mg, 0.0257 mmol) were combined in acetonitrile until all solid were dissolved. Figure 4.12 shows the aromatic regions of the 1H-NMR spectra of 11,14 and their mixture.
Phenoxide complex 17: The diacetate complex 11 (8.07 mg, 0.015 mmol) was added to a flame-dried flask with anhydrous CH3CN (1 mL), and the flask was capped with a rubber septum. To this was added a stock solution of ditriflate complex 14 (11.68 mg, 0.015 mmol) in CH3CN (0.5 mL), and the reaction mixture was stirred. Upon complete dissolution of all solids, a stock solution of 2-heptanol (0.5 mL, 0.2 M) was added through the septum, and the solution was stirred at room temperature under a balloon full of air for 24 h. The reaction mixture was filtered and Et2O was added to effect precipitation of 17. The solid was collected by centrifugation, washed with Et2O,
1 and dried briefly under high vacuum. H-NMR (CD3CN) #: 8.76 (d, J = 8.2 Hz, 1H), 8.64
(d, J = 5.2 Hz, 1H), 8.51 (dd, J = 15.7, 9.2 Hz, 1H), 8.01 (dd, J = 11.6, 8.7 Hz, 2H), 7.89
(dd, J = 5.2, 8.2 Hz, 1H), 7.18 (dd, J = 14.9, 8.1 Hz, 1H), 6.66 (d, J = 8.6 Hz, 1H), 6.50
19 (dd, J = 7.8, 14.5 Hz, 1H) F-NMR (CD3CN) #: -77.6, -105.9 (dd, J = 14.4, 6.2 Hz)
General method for aerobic oxidation of 2-heptanol using cationic palladium complexes: A solution of 2-heptanol (146 µL, 1.00 mmol) and decane (100 µL, 0.538 mmol) in acetonitrile (1.5 mL) was stirred under a balloon of air for 5 min. At t = 0, the palladium catalyst (3 mol %) was added either as a solid or as a solution in acetonitrile; in either case, the reaction mixture was adjusted such that in contained a total acetonitrile
137 volume of 2 mL. The mixture was stirred at room temperature, and aliquots were taken periodically and subjected to analysis by GC. 2-heptanone was confirmed by co- injection with an authentic sample.
2-(2!,6!-dichlorophenyl)-1,10-phenanthroline: This compound was prepared using the adapted condensation procedure described for 7 using 8-amino-7- quinolinecarbaldehyde (0.342 g, 1.98 mmol) and 2!,6! dichloroacetophenone (0.375 g,
1 1.98 mmol) to yield the target compound (0.046 g, 7%). H-NMR (CDCl3) #: 9.24 (dd, J
= 1.6, 4.4 Hz, 1H), 8.36 (d, J = 8.2 Hz, 1H), 8.35 (m, 1H), 7.89 (s, 2H), 7.66 (dd, J = 4.4,
8.0 Hz, 1H) 7.63 (d, J = 8.1 Hz, 1H), 7.38 (d, J = 8.0 Hz, 1H), 7.30 (dd, J = 7.3, 8.8 Hz,
13 2H), C- NMR (CDCl3) #: 156.0, 150.8, 146.5, 146.4, 139.3, 136.5, 136.1, 135.2, 130.0,
129.1, 128.2, 128.0, 127.2, 126.5, 124.6, 123.2.
2-(2!,4!,6!-trimethylphenyl)-1,10-phenanthroline: This compound was prepared using the adapted condensation procedure described for 7 using 8-amino-7- quinolinecarbaldehyde (0.327 g, 1.9 mmol) and 2!,4!,6!-trimethylacetophenone (0.308 g,
1.9 mmol) to yield the target compound in 19% yield (0.111 g, 0.368 mmol). 1H-NMR
(CDCl3) #: 9.20 (dd, J = 1.7, 4.2 Hz, 1H), 8.28 (d, J = 8.0 Hz, 1H), 8.27 (dd, J = 1.7, 8.0
Hz, 1H), 7.85 (d, J = 8.9 Hz, 1H), 7.81 (d, J = 8.9 Hz, 1H), 7.62 (dd, J = 4.4, 8.1 Hz, 1H),
7.55 (d, J = 8.1 Hz, 1H), 6.91 (s, 2H), 2.43 (s, 3H), 2.02 (s, 6H).
2-(2!,6!-difluoro-4!-methoxy-phenyl)-1,10-phenanthroline: This compound was prepared using the adapted condensation procedure described for 7 using 8-amino-7-
138 quinolinecarbaldehyde (0.558 g, 3.24 mmol) and 2!,6! difluoro-4!-methoxy-acetophenone
1 (0.603 g, 3.24 mmol) to yield the target compound (185 mg, 18%). H-NMR (CDCl3) #:
9.23 (dd, J = 1.6, 4.3 Hz, 1H), 8.31 (d, J = 8.3 Hz, 1H), 8.25 (dd, J = 1.7, 8.0 Hz, 1H),
7.83 (s, 2H), 7.76 (d, J = 8.3 Hz, 1H), 7.62 (dd, J = 4.4, 8.1 Hz, 1H), 6.59 (d, J = 9.5 Hz,
19 2H), 3.86 (s, 3H). F-NMR (CDCl3) #: -113.6 (d, J = 9.1 Hz).
Formation of electron-rich CF3 phen.
The synthesis of 4-methoxy-2-(trifluoromethyl)-1,10-phenanthroline was carried out in analogy to the route employed by Joullie and coworkers in three steps.31,32 The phenanthroline core was formed through a Conrad-Limpach-type condensation of 8- aminoquinoline and ethyltrifluoroacetoacetate. The subsequent phenol was chlorinated and the chloride was displaced by sodium methoxide to yield the desired product.
4-hydroxy-2-(trifluoromethyl)-1,10-phenanthroline. A mixture of 8- aminoquinoline (7.2 g, 54 mmol) and phosphoric acid (30 mL) was stirred and heated to
100-110 °C. Ethyl trifluoroacetoacetate was added to it in small portions. The mixture was then heated to 140-155 °C for 15 h, after which time the reaction mixture was cooled and allowed to stand at room temperature overnight. The dark brown solution was cooled to 0 °C and diluted with water (150 mL) and neutralized with 10% NaOH (60 mL) to pH 7. The mixture was extracted with diethyl ether until the extracts became nearly colorless and the solvent removed under vacuum. The crude material was purified by
1 column chromatography (SiO2, CH2Cl2 to EtOH) to yield 132 mg of product. H-NMR
(CDCl3) #: 9.01 (d, J = 4.3 Hz, 1H), 8.40 (dd, J = 1.6, 8.3 Hz, 1H), 8.17 (d, J = 9.0 Hz,
139 1H), 7.81 (d, J = 9.0 Hz, 1H), 7.73 (dd, J = 4.3, 8.3, 1H), 7.32(s, 1H). UV-vis: 210, 243,
261, 326. FT-IR: 1581, 1446, 1394, 1267, 1176, 1120, 1103, 982, 833, 576.
4-chloro-2-(trifluoromethyl)-1,10-phenanthroline. Phosphorus pentachloride
(0.135 g, 0.65 mmol) was added to a mixture of 4-hydroxy-2-(trifluormethyl)-1,10- phenanthroline (0.132 g, 0.5 mmol) in phosphorus oxychloride (0.152 g, 0.99 mmol).
The reaction was heated to 130 °C for 40 minutes after which time the volatiles were removed by distillation. The residue was rinsed with ice-cold water and neutralized with
1 M NaOH. The resulting white solid was isolated by filtration. The solid was extracted with Et2O and EtOAc to yield the crude product, which was immediately used in the following reaction.
4-methoxy-2-(trifluoromethyl)-1,10-phenanthroline. The chloride precursor
(93 mg, 0.329 mmol) was added to a flask containing NaOMe (68 mg, 1.25 mmol).
Methanol (11 mL) was added to the flask and the reaction was brought to reflux for 3 h.
All volatiles were then removed under high vacuum. The resulting solid was extracted with CCl4. Upon cooling the extract a light tan solid precipitated that was isolated by
1 filtration (50 mg, 0.181 mmol, 55%). H-NMR (CDCl3) #: 9.15 (dd, J = 1.8, 4.3 Hz, 1H),
8.41 (dd, J = 1.8, 8.2 Hz, 1H), 8.23 (d, J = 9.1 Hz, 1H), 7.98 (d, J = 9.1 Hz, 1H), 7.74 (dd,
J = 4.3, 8.1 Hz, 1H), 7.55 (s, 1H), 4.17 (s, 3H). 1H-NOE yielded strong coupling to 7.55
13 2 and a very weak coupling to 8.24. C-NMR (CDCl3) #: 163.8, 151.2, 149.2 (( JCF = 34.7
1 Hz), 146.6, 145.8, 136.3, 129.4, 127.6, 123.6, 121.8, 121.8 ( JCF = 276.1 Hz), 119.8, 99.5,
140 19 56.6 F-NMR (CDCl3) #: -65.5 FT-IR: 1591, 1566, 1515, 1377, 1274, 1245, 1164, 1130,
1053, 1028, 914, 848, 700
General method for aerobic oxidation of 2-heptanol using cationic palladium complexes: A solution of 2-heptanol (146 µL, 1.00 mmol) and decane (100 µL, 0.538 mmol) in acetonitrile (1.5 mL) was stirred under a balloon of air for 5 min. At t = 0, the palladium catalyst (3 mol %) was added either as a solid or as a solution in acetonitrile; in either case, the reaction mixture was adjusted such that in contained a total acetonitrile volume of 2 mL. The mixture was stirred at room temperature, and aliquots were taken periodically and subjected to analysis by GC. 2-heptanone was confirmed by co- injection with an authentic sample.
4.5.2. Complexation attempts between Pd and bis-2,9-
(trifluoromethyl)-1,10-phenanthroline (btfm-phen)
A. Attempts at producing (btfm-phen)Pd(OAc)2
Attempt 1 (A1).
To a stirred solution of Pd(OAc)2 (35.5 mg, 0.158 mmol) in acetonitrile (5 mL) was added bis-2,9-(trifluoromethyl)-1,10-phenanthroline (50 mg, 0.158 mmol). The reaction mixture was heated to 60 ºC and stirred overnight, during which time a significant amount of Pd black formed and the reaction mixture turned dark orange. The solution was filtered to remove Pd black. Our attempt to induce precipitation of the palladium complex by addition of Et2O was unsuccessful. An aliquot of the solution was
141 1 evaporated to dryness and dissolved in CDCl3. H-NMR chemical shifts were indistinguishable from those of the starting material. Elution of the reaction mixture through a short silica plug using CH2Cl2 yielded the free ligand (47 mg, 94% recovery).
Attempt 2 (A2).
The recovered ligand from A1 (47 mg, 149 mmol) was dissolved in CH2Cl2 (1 mL). To this solution was added Pd(OAc)2 in toluene (4 mL). The reaction mixture was stirred for at room temperature for 1 d, after which time an aliquot was taken. 1H-NMR analysis of the aliquot indicated that no reaction had occurred. The reaction time was extended to 3 d, but again, only the free ligand was observed by 1H-NMR. This experiment was repeated 3 additional times and similar results were obtained in all cases.
In some attempts, we isolated of a light brown solid from the mixture, but this gave spectra consistent with the free ligand.
142 1 Figure 4.8. H-NMR (CDCl3) of the crude reaction mixture from attempt 2 (A2). Inset contains blowup of desired region of the (a) crude reaction and (b) free ligand.
B. Attempt at producing [(btfm-phen)Pd(NCCH3)2][OTf]2
Following our failed attempts to isolate a palladium complex of btfm-phen, we turned to dicationic palladium species in hopes that they would be more susceptible toward ligand complexation. To this end, a solution of (CH3CN)4Pd(OTf)2 (59 mg, 0.104 mmol) in d3-acetonitrile (1 mL) was added to a solution of the ligand (33 mg, 0.104 mmol) in d3-acetonitrile (1 mL). The reaction was stirred for days at room temperature.
1H-NMR analysis of the mixture showed peaks slightly shifted from the free ligand
(Figure 4.9). Given the small magnitude of the shift from free ligand and the absence of any time dependence (The shift was observed immediately after preparation of the sample), we do not believe that complexation has occurred.
143 1 Figure 4.9. H-NMR of the attempts at complexation of (CH3CN)4Pd(OTf)2 with btfm- phen in CD3CN. The inset compares the downfield region of the spectrum (a) immediately after mixing, (b) 23 h after mixing, and (c) of pure btfm-phen in CD3CN.
144 4.5.3 Comparison of bond lengths and angles between complexes 8 and 1
Table 4.2. Bond lengths and angles for complexes 8 and 1 Complex 8 Complex 1
Pd1-Pd2 2.8548(9) Pd1-Pd2, 2.9582(4)
Pd1-O1, 2.007(3) Pd1-O1, 2.016(2)
Pd1-O3, 2.021(4) Pd1-O2, 2.015(2)
Pd1-N1, 2.014(4) Pd1-N2, 2.024(2)
Pd1-N2, 2.000(4) Pd1-N1, 2.019(3)
Pd2-O2, 2.013(4) Pd2-O4, 2.026(2)
Pd2-O4, 2.016(4) Pd2-O3, 2.0105(19)
Pd2-N3, 2.034(4) Pd2-N4, 2.056(2)
Pd2-N4, 2.032(4) Pd2-N3, 2.038(2)
O3-C27, 1.286(7) O4-C37, 1.272(3)
C27-O4, 1.299(7) O1-C37, 1.256(4)
O1-C25, 1.282(6) O3-C45, 1.271(3)
C25-O2, 1.308(7) O2-C45, 1.262(3)
O1-Pd1-O3, 91.26(15) O1-Pd1-O2, 84.17(9)
O2-Pd2-O4, 90.67(18) O4-Pd2-O3, 81.85(8)
O1-Pd1-N1, 93.38(16) O4-Pd2-N3, 173.46(9)
N1-Pd1-N2, 82.21(16) N3-Pd2-N4, 82.77(9)
145 4.5.4 General screening strategy for catalyst activity in the oxidation of 2-heptanol by in-situ generation of catalysts
Palladium (II) acetate was added to a solution of the desired ligand in acetonitrile.
The solution was stirred at room temperature for 4 hours to allow for the formation of the desired Pd complex. A stock solution of triflic acid was then added (1 equiv. HOTf relative to Pd) to generate the desired activated palladium complex. The flask was then equipped with a septum and sparged with air for 5 minutes followed by addition of 2- heptanol (142 uL, 1 mmol). Aliquots of the reaction mixture were taken after the desired time period, diluted in acetonitrile, and analyzed by gas chromatography. This method produces similar results to those obtained with isolated catalysts. The neocuproine result shown using this method had a comparable initial turnover frequency but a slightly lower turnover number than that of a isolated catalyst (TOFi = 78, TON = 12).
146 1 Figure 4.10 H-NMR overlay of the aromatic region of (neocuproine)Pd(OAc)2, [(neocuproine)Pd(NCCH3)2][OTf]2, and 8
147 Figure 4.11 1H-NMR overlay of the aromatic region of 9, 10, and 9/10
148 Figure 4.12 1H-NMR overlay of the aromatic region of 11, 14, 15, and 11/14
149 4.6 References
[1] Gligorich, K. M.; Sigman, M. S. Recent advancements and challenges of
palladium(II)-catalyzed oxidation reactions with molecular oxygen as the sole
oxidant. Chem. Commun. 2009, 3854-3867.
[2] Sigman, M. S.; Jensen, D. R. Ligand Modulated Palladium-Catalyzed Aerobic
Alcohol Oxidation. Acc. Chem. Res. 2006, 29, 221-229.
[3] Schultz, M. J.; Sigman, M. S. Recent advances in homogeneous transition metal-
catalyzed aerobic alcohol oxidations. Tetrahedron 2006, 62, 8227-8241.
[4] Stoltz, B. M. Palladium Catalyzed Aerobic Dehydrogenation: From Alcohols to
Indoles and Asymmetric Catalysis. Chem. Lett. 2004, 33, 362-367.
[5] Stahl, S. S. Palladium Oxidase Catalysis: Selective Oxidation of Organic
Chemicals by Direct Dioxygen-Coupled Turnover. Angew. Chem., Int. Ed. 2004,
43, 3400-3420.
[6] Muzart, J. Palladium-catalysed oxidation of primary and secondary alcohols.
Tetrahedron 2003, 59, 5789-5816.
[7] Blackburn, T. F.; Schwartz, J. Homogeneous Catalytic Oxidation of Secondary
Alcohols to Ketones by Molecular Oxygen under Mild Conditions. J. Chem. Soc.,
Chem. Commun. 1977, 157-158.
[8] Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. Pd(OAc)2-Catalyzed Oxidation of
Alcohols to Aldehydes and Ketones by Molecular Oxygen. Tetrahedron Lett.
1998, 39, 6011-6014.
150 [9] Peterson, K. P.; Larock, R. C. Palladium-Catalyzed Oxidation of Primary and
Secondary Allylic and Benzylic Alcohols. J. Org. Chem. 1998, 63, 3185-3189.
[10] Steinhoff, B. A.; Fix, F. R.; Stahl, S. S. Mechanistic Study of Alcohol Oxidation
by the Pd(OAc)2/O2/DMSO Catalyst System and Implications for the
Development of Improved Aerobic Oxidation Catalysts. J. Am. Chem. Soc. 2002,
124, 766-767.
[11] Steinhoff, B. A.; Stahl, S. S. Ligand-Modulated Palladium Oxidation Catalysis:
Mechanistic Insights into Aerobic Alcohol Oxidation with the Pd(OAc)2/Pyridine
Catalyst System. Org. Lett. 2002, 4, 4179-4181.
[12] Ferreira, E. M.; Stoltz, B. M. The Palladium-Catalyzed Oxidative Kinetic
Resolution of Secondary Alcohols with Molecular Oxygen. J. Am. Chem. Soc.
2001, 123, 7725-7726.
[13] Jensen, D. R.; Pugsley, J. S.; Sigman, M. S. Palladium-Catalyzed Enantioselective
Oxidations of Alcohols using Molecular Oxygen. J. Am. Chem. Soc. 2001, 123,
7475-7486.
[14] Ebner, D. C.; Trend, R. M.; Genet, C.; Mcgrath, M. J.; O'Brien, P.; Stoltz, B. M.
Palladium-Catalyzed Enantioselective Oxidation of Chiral Secondary Alcohols:
Access to Both Enantiomeric Series. Angew. Chem., Int. Ed. 2008, 47, 6367-6370.
[15] ten Brink, G.-J.; Arends, I. W. C. E.; Hoogenraad, M.; Verspui, G.; Sheldon, R.
A. Catalytic Conversion in Water. Part 23: Steric Effects and Incereased Substrate
Scope in the Palladium-Neocuproine Catalyzed Aerobic Oxidation of Alcohols in
Aqueous Solvents. Adv. Synth. Catal. 2003, 345, 1341-1352.
151 [16] ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Catalytic Conversions in
Water. Part 21: Mechanistic Investigations on the Palladium-Catalysed Aerobic
Oxidation of Alcohols in Water. Adv. Synth. Catal. 2002, 344, 355-369.
[17] ten Brink, G.-J.; Arends, I. W. C. E.; Hoogenraad, M.; Verspui, G.; Sheldon, R.
A. Catalytic Conversion in Water. Part 22: Electronic Effects in the
(Diimine)palladium(II)-Catalyzed Aerobic Oxidation of Alcohols. Adv. Synth.
Catal. 2003, 345, 497-505.
[18] Arends, I. W. C. E.; ten Brink, G.-J.; Sheldon, R. A. Palladium-neocuproine
catalyzed aerobic oxidation of alcohols in aqueous solvents. J. Mol. Catal. A:
Chem. 2006, 251, 246-254.
[19] ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Green, Catalytic Oxidation of
Alcohols in Water. Science 2000, 287, 1636-1639.
[20] Mueller, J. A.; Goller, C. P.; Sigman, M. S. Elucidating the Significance of b-
Hydride Elimination and the Dynamic Role of Acid/Base Chemistry in a
Palladium-Catalyzed Aerobic Oxidation of Alcohols. J. Am. Chem. Soc. 2004,
128, 9724-9734.
[21] Schultz, M. J.; Hamilton, S. S.; Jensen, D. R.; Sigman, M. S. Development and
Comparison of the Substrate Scope of Pd-Catalysts for the Aerobic Oxidation of
Alcohols. J. Org. Chem. 2005, 70, 3343-3352.
[22] Jensen, D. R.; Sigman, M. S. Palladium Catalysts for Aerobic Oxidative Kinetic
Resolution of Secondary Alcohols Based on Mechanistic Insight. Org. Lett. 2003,
5, 63-65.
152 [23] Jensen, D. R.; Schultz, M. J.; Mueller, J. A.; Sigman, M. S. A Well-Defined
Complex for Palladium Catalyzed Aerobic Oxidation of Alcohols: Design,
Synthesis, and Mechanistic Considerations. Angew. Chem., Int. Ed. 2003, 42,
3810-3813.
[24] Mueller, J. A.; Sigman, M. S. Mechanistic Investigations of the Palladium-
Catalyzed Aerobic Oxidative Kinetic Resolution of Secondary Alchols using (–)-
Sparteine. J. Am. Chem. Soc. 2003, 125, 7005-7013.
[25] Mueller, J. A.; Jensen, D. R.; Sigman, M. S. Dual Role of (–)-Sparteine in
Palladium-Catalyzed Aerobic Oxidative Kinetic Resolution of Secondary
Alcohols. J. Am. Chem. Soc. 2002, 124, 8202-8203.
[26] Conley, N. R.; Labios, L. A.; Pearson, D. M.; McCrory, C. C. L.; Waymouth, R.
M. Aerobic Alcohol Oxidation with Cationic Palladium Complexes: Insights into
Catalyst Design and Decomposition. Organometallics 2007, 26, 5447-5453.
[27] Konnick, M. M.; Stahl, S. S. Reaction of Molecular Oxygen with a PdII-Hydride
To Produce a PdII-Hydroperoxide: Experimental Evidence for an HX-Reductive-
Elimination Pathway. J. Am. Chem. Soc. 2008, 130, 5753-5762.
[28] Pearson, D. M.; Waymouth, R. M. Mechanistic Studies of the Oxidative
Dehydrogenation of Methanol Using a Cationic Palladium Complex.
Organometallics 2009, 28, 3896-3900.
[29] Beer, R. H.; Jimenez, J.; Drago, R. S. Syntheses of 2,9-bis(halomethyl-1,10-
phenanthrolines: potential robust ligands for metal oxidation catalysts. J. Org.
Chem. 1993, 58, 1746-1747.
153 [30] Kovalevsky, A. Y.; Gembicky, M.; Coppens, P. Cu(I)(2,9-Bis(trifluoromethyl)-
1,10-phenanthroline)2+ Complexes: Correlation between Solid-State Structure and
Photoluminescent Properties. Inorg. Chem. 2004, 43, 8282-8289.
[31] Nyquist, E. B.; Joullie, M. M. Synthesis and Properties of Fluorine-Containing
Heterocyclic Compounds V. Trifluoromethyl-1,8- and 1,10-Phenanthrolines J.
Heterocycl. Chem. 1967, 4, 539-545.
[32] Loy, M.; Joullie, M. M. Synthesis and Properties of Fluorine-Containing
Heterocyclic Compounds. VIII. $-(2-Pyridyl)- and $-(2-Piperidyl)-2-
(trifluoromethyl)-4-azaphenanthrenemethanol. J. Med. Chem. 1973, 16, 549-552.
[33] Belser, P.; Bernhard, S.; Guerig, U. Synthesis of mono- and dialkylsubstituted
1,10-phenanthrolines. Tetrahedron 1996, 52, 2937-2944.
[34] Ferretti, F.; Ragaini, F.; Lariccia, R.; Gallo, E.; Cenini, S. New Nonsymmetric
Phenanthrolines as Very Effective Ligands in the Palladium-Catalyzed
Carbonylation of Nitrobenzene. Organometallics 2010, 29, 1465-1471.
[35] Plenkiewicz, H.; Dmowski, W.; Lipinski, M. Sodium dithionite initiated reactions
of Halothane(R) with enol ethers: Facile synthesis of 3-trifluoromethyl substituted
vinyl carbonyl compounds. J. Fluorine Chem. 2001, 111, 227-232.
[36] Ogoshi, H.; Mizushima, H.; Toi, H.; Aoyama, Y. 1,1,1-Trifluoro-2-penten-4-one
as a building block of trifluoromethyl-substituted compounds. J. Org. Chem.
1986, 51, 2366-2368.
[37] Hamilton, C. W.; Laitar, D. S.; Sadighi, J. P. Oxidation-resistant, sterically
demanding phenanthrolines as supporting ligands for copper(I) nitrene transfer
catalysts. Chem. Commun. 2004, 1628-1629.
154 [38] Cookson, P. G.; Deacon, G. B. Syntheses of organonickel compounds by
decarboxylation reactions. J. Organomet. Chem. 1971, 33, C38-C40.
[39] Cookson, P. G.; Deacon, G. B. Preparations of organonickel compounds by
thermal decomposition of nickel carboxylates. Aust. J. Chem. 1972, 25, 2095-
2105.
[40] Riesgo, E. C.; Jin, X. Q.; Thummel, R. P. Introduction of benzo[h]quinoline and
1,10-phenanthroline subunits by Friedlander methodology. J. Org. Chem. 1996,
61, 3017-3022.
[41] Dietrich-Buchecker, C. O.; Marnot, P. A.; Sauvage, J. P. Direct Synthesis of
Disubstituted Aromatic Polyimine Chelates. Tetrahedron Lett. 1982, 23, 5291-
5294.
[42] 2,6-difluorophenyl lithium was prepared in situ by addition of n-BuLi at low
temperature to an etherial solution of 1-bromo-2,6-difluorobenzene.
[43] Milani, B.; Alessio, E.; Mestroni, G.; Sommazzi, A.; Garbassi, F.; Zangrando, E.;
Brescianipahor, N.; Randaccio, L. Synthesis and Characterization of
Monochelated Carboxylatopalladium(II) Complexes with Nitrogen-Donor
Chelating Ligands - Crystal-Structures of Diacetato(1,10-Phenanthroline) and
Diacetato(2,9-Dimethyl-1,10-Phenanthroline)-Palladium(II). J. Chem. Soc.,
Dalton Trans. 1994, 1903-1911.
[44] Adrian, R. A.; Broker, G. A.; Tiekink, E. R. T.; Walmsley, J. A. Palladium(II)
complexes of 1,10-phenanthroline: Synthesis and X-ray crystal structure
determination. Inorg. Chim. Acta 2008, 361, 1261-1266.
155 [45] Borriello, C.; Centore, R.; Roviello, G. Hydrolysis of Pd(II)- bound acetonitrile:
Molecular and crystal structure of the syn isomer of the dinuclear (1,10-
phenanthroline)palladium(II) complex bridged with acetamido groups and of its
acetonitrile solvate. Inorg. Chem. Commun. 2005, 8, 755-758.
[46] Adrian, R. A.; Zhu, S. R.; Powell, D. R.; Broker, G. A.; Tiekink, E. R. T.;
Walmsley, J. A. Dinuclear palladium(II) complexes with bridging amidate
ligands. Dalton Trans. 2007, 4399-4404.
[47] Bercaw, J. E.; Durrell, A. C.; Gray, H. B.; Green, J. C.; Hazari, N.; Labinger, J.
A.; Winkler, J. R. Electronic Structures of Pd-II Dimers. Inorg. Chem. 2010, 49,
1801-1810.
[48] Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.; Echeverria, J.;
Cremades, E.; Barragan, F.; Alvarez, S. Covalent radii revisited. Dalton Trans.
2008, 2832-2838.
[49] Berry, J. F.; Cotton, F. A.; Sibragimov, S. A.; Murillo, C. A.; Wang, X. Searching
for Precursors to Metal-Metal Bonded Dipalladium Species: A Study of Pd2(4+)
complexes. Inorg. Chem. 2005, 44, 6129-6137.
[50] Powers, D. C.; Ritter, T. Bimetallic Pd(III) complexes in palladium-catalysed
carbon-heteroatom bond formation. Nature Chem. 2009, 302-309.
[51] Anderson, C. M.; Crespo, M.; Ferguson, G.; Lough, A. J.; Puddephatt, R. J.
Activation of Aromatic Carbon-Fluoride Bonds by Organoplatinum Complexes.
Organometallics 1992, 11, 1177-1181.
[52] Lopez, O.; Crespo, M.; Font-Bardia, M.; Solans, X. Activation of C-F and C-H
Bonds by Platinum in Trifluorinated [C,N,N'] Ligands. Crystal Structures of
156 [PtFMe2{Me2NCH2CH2NHCH(CH2COMe)(2,4-C6H2F2)} and
[PtMe{Me2NCH2CH2NCH(2,3,4-C6HF3)}]. Organometallics 1997, 16, 1233-
1240.
[53] Crespo, M.; Martinez, M.; Sales, J. Intramolecular activation of a C-F bond at
Platinum(II) in the presence of weaker C-X bonds (X=H, Cl, and Br). J. Chem.
Soc., Chem. Commun. 1992, 822-823.
[54] Osterberg, C. E.; Richmond, T. G. Activation of carbon-fluorine bonds by
oxidative addition of low-valent transition metals. ACS Symp. Ser. 1994, 555,
392-404.
[55] Kiplinger, J. L.; Richmond, T. G.; Osterberg, C. E. Activation of Carbon Fluorine
Bonds by Metal Complexes. Chem. Rev. 1994, 94, 373-431.
[56] Wang, T.; Alfonso, B. J.; Love, J. A. Platinum(II)-Catalyzed Cross-Coupling of
Polyfluoroaryl Imines. Org. Lett. 2007, 9, 5629-5631.
[57] Wang, T.; Love, J. A. Insight into the Mechanism of Platinum-Catalyzed Cross-
Coupling of Polyfluoroaryl Imines. Organometallics 2008, 27, 3290-3296.
[58] Buckley, H. L.; Wang, T.; Tran, O.; Love, J. A. Selective Platinum-Catalyzed C-F
Bond Acitivation as a Route to Fluorinated Aryl Methyl Ethers. Organometallics
2009, 28, 2356-2359.
[59] Usón, R.; Fornies, J.; Espinet, P.; Garcia, A.; Tomas, M.; Foces-Foces, M.; Cano,
F. H. J. Organomet. Chem. 1985, 282, C35-C38.
[60] Wimmer, S.; Castan, P.; Wimmer, F. L.; Johnson, N. P. Aqueous chemistry of
Pt(II) and Pd(II) complexes of 2,2'-bipyridine and 1,10-phenanthroline: pH
dependence. Inorg. Chim. Acta 1988, 142, 13-15.
157 [61] Beletskaya, I. P.; Chuchurjukin, A. V.; Dijkstra, H. P.; van Klink, G. P. M.; van
Koten, G. Conjugated G0 metallo-dendrimers, functionalized with tridentate
'pincer'-type ligands. Tetrahedron Lett. 2000, 41, 1081-1085.
[62] Halcrow, B. E.; Kermack, W. O. Attempts to find New Antimalarials. Part XXIV.
Derivatives of o-Phenanthroline (7:8:3':2'-pyridoquinoline). J. Chem. Soc. 1946,
155-157.
158 Chapter 5
Palladium Catalyzed Carbonylation of Diols to Cyclic Carbonates
5.0 Preface
This chapter describes research to be submitted to Advanced Synthesis and Catalysis.: Pearson, D. M.; Conley, N. R.; Waymouth, R. M. “Palladium Catalyzed Carbonylation of Diols to Cyclic Carbonates.“ The project was initiated by myself and Nicholas Conley after the discovery that [(neocuproine)Pd(OAc)]2[OTf]2 selectively oxidizes glycerol to dihydroxyacetone as a possible way of producing polymerizable cyclic carbonate monomers directly from glycerol using a single catalyst. Initial 1H-NMR screening studies using (neocuproine)Pd(OAc)2 and N-chlorosuccinimide were carried out by N. R. Conley. Upon optimization of these conditions, I explored the use of other Pd-complexes with these conditions. N. R. Conley and myself looked at substrate scope of 1,2- and 1,3- diols on 1H-NMR scale. I carried out and optimized scale up procedures all compounds with the exception of 2b which was carried out by N. R. Conley.
5.1 Introduction
Cyclic carbonates are versatile intermediates, solvents1-3 and monomers.4 Diols are attractive precursors to cyclic carbonates given the ease with which they can be prepared,5,6 and in some cases, their commodity chemical status (e.g., ethylene glycol,
1,2-propylene glycol, glycerol). Preparative routes to cyclic carbonates include carbon dioxide insertion into epoxides,1-3 transesterification of diols with alkyl carbonates,1-3,7,8 condensation of diols with urea9-11 or carbon dioxide,12,13 and carbonylation of diols in
159 the presence of elemental sulfur.14 A common preparation of cyclic carbonates from diols involves the use of phosgene,1,15,16 an effective but toxic reagent that is still employed in many laboratory and industrial syntheses due to the lack of a more suitable alternative.
We describe herein an efficient, general approach to preparing cyclic carbonates from diols and CO under mild conditions.
The oxidative carbonylation of methanol to dimethylcarbonate is practiced commercially,1 but there are few reports on the oxidative carbonylation of 1,2- and 1,3- diols.17-24 Our interest in a palladium-catalyzed oxidative carbonylation of diols was stimulated by a report describing the carbonylation of 1-phenylethane-1,2-diol using
CO,17 as shown in Scheme 5.1. In this system, sodium acetate acts as a base to neutralize the hydrochloric acid that is generated, and copper(II) chloride serves as a stoichiometric oxidant for palladium. Unfortunately, when ethylene glycol was used as the substrate, no cyclic carbonate was produced, purportedly due to the interaction of CuCl2 with ethylene glycol. The stoichiometic carbonylation of ethylene glycol with PdCl2 in the absence of
CuCl2 yielded ethylene carbonate in 91% yield.
O PdCl2 (10 mol %)/CuCl2 (2 eq.) HO OH NaOAc (2 eq.) O O CO (3 atm) RT, 4 d
100% Scheme 5.1. Pd-catalyzed carbonylation of 1-phenylethane-1,2-diol with CO
Giannoccaro and coworkers23 determined the solid-state structures of alkoxycarbonyl complexes resulting from the interaction of a PN-palladium complex with CO and either 1,3-propanediol or 1,2-butanediol, in the presence of triethylamine.
Upon heating, these complexes liberated the cyclic carbonates and Pd black. The authors 160 propose that the cyclic carbonate is formed by intramolecular nucleophilic attack of the free -OH on the carbonyl carbon atom. These studies provided key mechanistic insights into palladium-catalyzed carbonylation of diols.23
Following our initial investigations, Gabriele and coworkers published a method for the catalytic carbonylation of diols using PdI2 at high temperatures and pressures of oxygen and carbon monoxide (i.e. 100 °C, 10 atm).25 Under these conditions, HI is generated during the reaction and is subsequently oxidized by oxygen to yield I2 which serves to reoxidize Pd.26
5.2 Results and Discussion
5.2.1 Catalyst screening studies
We have discovered that (neocuproine)Pd(OAc)2, previously shown to catalyze the aerobic oxidation of alcohols,27-30 catalyzes the carbonylation of 1-phenylethane-1,2- diol in the presence of carbon monoxide and a suitable oxidant such as N-
31 chlorosuccinimide (NCS). Our initial success with (neocuproine)Pd(OAc)2 prompted us to explore other common palladium based alcohol oxidation catalysts (Table 5.1). The use of Pd(OAc)2 in dsmo resulted in minimal conversion of the diol to the hydroxyketone and no cyclic carbonate was observed. Combination of Pd(OAc)2 and pyridine in acetonitrile yielded small amounts of the cyclic product. The use of the chelating diamine, (-)-sparteine, with Pd(OAc)2 was effective for this transformation resulting in a
80% yield of the five membered carbonate by 1H NMR. The use of the tridentate ligand
9-methyl-1,10-phenanthroline-2-carboxylate (derived from aerobic oxidation of
161 neocuproine, see Chapter 2) resulted in low conversions but similar yields of the cyclic carbonate as pyridine/Pd(OAc)2. While the oxidative carbonylation of 1,2- phenylethanediol has not been attempted with (1,10-phenanthroline)Pd(OAc)2 work with other diols gave significantly inferior yields of the cyclic carbonate relative to neocuproine.
Table 5.1. Catalyst screening for the carbonylation of 1-phenyl-ethane-1,2-diola entry catalyst yield/convb c 1 (neocuproine)Pd(OAc)2 89/95 d 2 dmso / Pd(OAc)2 0/10 e 3 pyridine / Pd(OAc)2 10/10 f 4 (-)-sparteine /Pd(OAc)2 80/100 c 5 (9-methyl-1,10-phenanthroline-2-carboxylate)Pd(NCCH3)(OTf) 10/20 a Reaction conditions: 1-phenyl-1,2-ethanediol (0.1 mmol), N-chlorosuccinimide (0.15 b 1 c d mmol), CD3CN (1.5 mL), CO (1 atm), 55 °C. Observed by H-NMR. 3 mol % Pd. e f Pd(OAc)2 (5 mol %). Pd(OAc)2 (5 mol %), pyridine (20 mol %). Pd(OAc)2 (5 mol %), (-)-sparteine (10 mol %)
5.2.2 Carbonylation of 1,2-diols to yield 5-membered carbonates using N-chlorosuccinimide as the oxidant
The success of the (neocuproine)Pd(OAc)2 system toward the carbonylation of 1- phenylethane-1,2-diol prompted us to explore the tolerance of other diols toward oxidative carbonylation (Table 5.2). Initial studies were conducted on small scale employing 1H-NMR spectroscopy to assess the conversion and selectivity cyclic carbonate formation. The high yield of ethylene carbonate 2a obtained from ethylene glycol 1a (entry 1) is notable, as 1a is not compatible with the previously reported
17 PdCl2/CuCl2/CO catalytic system, shown in Scheme 1. Likewise, monoalkyl-cyclic carbonates (entries 2 and 6) can be obtained from their corresponding glycols. To
162 determine the ability of the catalyst to tolerate unsaturation, vinyl-substituted ethylene glycol 1i was carbonylated to give the polymerizable 2i in 51% yield (entry 9). The catalyst tolerated the asymmetrical dimethyl-substituted ethylene glycol 1d (entry 4) slightly better than the symmetrical isomer 1c (entry 3), while the tetramethyl-substituted glycol 1e was not a competent substrate (entry 5). These results highlight the steric constraints of the catalyst, and indicate that vic-diols containing at least one primary alcohol are preferred substrates, while vic-diols with two tertiary alcohols are not tolerated. The mild conditions and high selectivity for formation of the cyclic carbonates is noteworthy, and is in contrast to existing systems that often require elevated temperatures (80-100 °C).25,32
Table 5.2. Scope of 1,2-Diol Carbonylation O (neocuproine)Pd(OAc) (3 mol %) HO OH 2 NCS (1.5 eq.), NaOAc (1 eq.) O O 1 R4 R 2 3 CO (1 atm), CH3CN, 3Å mol sieves, R R R1 R4 24 h, 55 °C R2 R3 1 2 substrate/ yield/ entry R1 R2 R3 R4 product conv.c 1 1a/2a H H H H 89/95 2 1b/2b Me H H H 83/93 3 1c/2c Me H Me H 62/66 4 1d/2d Me Me H H 72/73 5 1e/2e Me Me Me Me NR 6d 1f/2f Et H H H 97/97 7d 1g/2g Ph H H H 97e/97 8 1h/2h CH2OH H H H 88/89 9 1i/2i Vinyl H H H 51/56 10 1j/2j (CH2)2OH H H H 80/81 a Crushed molecular sieves (400 mg/mmol of 1) were employed. b For this screening, acetonitrile-d3 was used as the solvent in an amount such that the substrate concentration was 166 mM. c Yields and conversions were calculated from 1H NMR spectra using p- xylene as an internal standard. d Pd black was observed in the reaction mixture. e The product was isolated in 95% yield when performed on a 1g-scale.
163 The selectivity for five-membered cyclic carbonates was investigated using triol
1h. Under the conditions shown in Table 5.2, glycerol 1h is converted exclusively to glyceryl carbonate 2h (entry 8),18,33 a versatile chemical intermediate. The six-membered cyclic carbonate could not be detected in the reaction mixture; demonstrating high selectivity for the thermodynamically more stable, five-membered ring carbonate.
Table 5.3. Scope of 1,2-Diol Carbonylation O
HO OH Pd cat , NCS (1.5 eq) O O
NaOAc (1 eq), CO (1 atm), CH3CN, 1 1 R 3Å mol sieves, 24 h, 55 °C R 1 2 substrate/ entry R1 Cond.a yieldc product 1 1a/2a H A 53 2 B 82 3 1g/2g Ph A 81 4 B 81 5 1h/2h CH2OH A 49 6 B 43 7 1i/2i Vinyl A 28 8 B 49 a Reaction conditions (A): diol (3 mmol), (neocuproine)Pd(OAc)2 (3 mol %, 0.09 mmol), N-chlorosuccinimide (0.6 g, 1.5 eq, 4.5 mmol), NaOAc (0.246 g, 3 mmol), molecular sieves (1.2 g), CH3CN (18 mL), CO (1 atm), 55 °C, 24h. Reaction conditions (B): diol (3 mmol), Pd(OAc)2 (33.7mg, 5 mol %, 0.15 mmol), (-)-sparteine (70.3 mg, 10 mol %, 0.3 mmol) N-chlorosuccinimide (0.6 g, 1.5 eq, 4.5 mmol), NaOAc (0.246 g, 3 mmol), c molecular sieves (1.2 g), CH3CN (18 mL), CO (1 atm), 55 °C, 24h Isolated by column chromotography on SiO2.
Selected substrates were scaled up to 3 mmol scale for isolation using N- chlorosuccinimide as the oxidant and either (neocuproine)Pd(OAc)2 (conditions A, Table
5.3) or (-)-sparteine / Pd(OAc)2 (conditions B, Table 5.3) as the catalyst. All products were worked up and purified by column chromotography on silica gel. In general, the yields of isolated products were lower than those observed in our initial studies using 1H 164 NMR and trace amounts of succinimide remained in some of the purified products. We attribute the lower observed isolated yield not only due to problems of isolation,34 but also due to scaling problems with the reaction.35 Both methods gave similar results for the phenyl substituted diol 1f and glycerol. Given the prevalent use of (-)-sparteine / Pd catalysts in enantioselective alcohol oxidation, one might expect enantioselectivity in the formation of glycerol carbonate 1h. Formation of the Mosher ester yielded two distinct
1 H NMR peaks in a 1:1 ratio, indicating a racemic mixture. The (-)-sparteine / Pd(OAc)2 catalyst proved superior in the carbonylation of ethylene glycol and the vinyl diol 1h.
We attribute both of these results due to the greater catalyst loading in the (-)-sparteine /
Pd(OAc)2 system, as scaled up runs using (neocuproine)Pd(OAc)2 at 5 mol % loadings gave comparable isolated yields of 2a and 2h (74 And 57 % respectively).
5.2.3 Carbonylation of 1,3-diols to 6-membered carbonates using hypervalent iodine oxidants.
While the intramolecular competition experiments demonstrated a preference for formation of five-membered cyclic carbonates, (neocuproine)Pd(OAc)2 also proved to be a competent catalyst for the carbonylation of 1,3- and 1,4-diols resulting in the formation of six- and seven-membered cyclic carbonates, respectively, as shown in Table 5.4.
These carbonates required a higher catalyst loading (10 mol %) and greater dilution of the substrate to favor intramolecular cyclization over intermolecular alkoxylation, the latter of which produces linear carbonates. Iodosobenzene (PhIO)36 was found to be a more selective terminal oxidant than NCS, as the former afforded cyclic carbonates in higher yields and selectivities. For example, oxidative carbonylation of 1k with NCS proceeded
165 in 73% conversion to give the cyclic carbonate in 31% yield, along with a byproduct, tentatively assigned as the linear carbonate. The byproduct is present only in a trace amount when PhIO is employed as the oxidant. PhIO is ineffective for the carbonylation of 1,2-diols, however, as it mediates the competitive oxidative cleavage of the carbon- carbon glycol bond of some 1,2-diols, a known chemistry of I(III);37,38 For example, we successfully prepared 2a from 1a in 95% yield using PhIO, but 1c was converted nearly quantitatively to acetaldehyde.
Table 5.4. Scope of 1,3- and 1,4-diol carbonylation.a entry substrate product yield/conv.b
1c 86/94 1k 2k
2c 77/85 1l 2l
3 85/100
1m 2m
4 64/76
1n 2n
5c 30/74
2o 1o a Reaction conditions are as follows: substrate (56 mM), (neocuproine)Pd(OAc)2 (10 mol %), iodosylbenzene (4.5 eq.), sodium acetate (3 eq.), crushed molecular sieves (2.4 b g/mmol), and CO (1 atm) in acetonitrile-d3 at 55 °C for 24 h. Yields and conversions were calculated from 1H NMR spectra using p-xylene as an internal standard. c Pd black was observed in the reaction mixture.
166 Using these modified conditions, 1,3-propanediol 1k was converted in good yield to the six-membered cyclic carbonate 2k (Table 5.4, entry 1). Dimethyl substitution of the propanediol at the 2-position, which might be expected to facilitate cyclization on account of the gem-dialkyl effect,39-42 resulted in slightly lower yield and conversion
(entry 2). This suggests that steric effects are a stronger determinant of yield than conformational effects. Ketals such as 1m (entry 3) and benzyl esters such as 1n (entry 4) are well tolerated, affording the corresponding cyclic carbonates in yields of 85% and
64%, respectively. Expanding the scope of substrates, 1,4-diols were introduced and converted to seven-membered cyclic carbonates; 1o was subjected to the conditions in
Table 5.4 (entry 5), and afforded cyclic carbonate 2o,43 albeit in modest yield. The linear carbonate was the major product, indicating that 1,4-diols require greater dilution to favor intramolecular cyclization.
We have had some difficulty reproducing the high yields and selectivity for the six-membered cyclic carbonates (see Table 5.4) using subsequent batches of iodosobenzene;36 results were reproducible using the original batch in duplicate experiments, however. The insolubility of the polymeric iodosobenzene in non-reactive organic solvents precludes the use of conventional purification methods (i.e., column chromatography or recrystallization). Furthermore, it complicates characterization by conventional solution-phase methods (e.g., 1H NMR spectroscopy). Though elemental analysis provides some insight, the percent oxygen—perhaps the most useful figure of merit for assessing functional purity—cannot be accurately quantified in the presence of such a large percentage of iodine (see Table 5.5). In describing the preparation of iodosobenzene, Saltzman and Sharefkin36 note that the purity of the precursor,
167 iodobenzene diacetate, is a strong determinant in the purity of the iodosobenzene after workup.
All experiments used in the substrate screening (Table 5.4) were carried out using batch 1, which displayed lower carbon values than expected. In addition, the theoretical oxygen content, taken as the difference in the total experimental C, H, and I contents from 100%, was significantly larger than that expected for iodosobenzene. Batch 2, which had a similar composition, yielded higher linear to branched ratios than those observed with batch 1 and a similar result was observed with batch 3, which matched the expected elemental composition for iodosobenzene.
Table 5.5 Elemental analyses of iodosobenzene batchesa prepared from iodobenzene diacetate and theoretical elemental compositions for selected hypervalent iodide species. batch C H I O (theory)b 1 29.11 2.06 50.86 17.97 2 32.12 2.26 51.22 14.4 3 32.22 2.14 57.54 7.3 PhIO 32.76 2.29 57.68 7.27
PhI(OH)2 30.28 2.96 53.32 13.44
PhIO2 30.52 2.14 53.77 13.56 a Elemental analysis for the percent composition of C, H, and N was performed on batches 1-3. A nitrogen content lower than 0.1% was observed in all cases. b The theoretical oxygen content displayed assumes the remainder of the percent composition not observed is due to oxygen as was calculated using the following equation: O(theory) = 100 – (C + H + N).
The elemental composition of batch 1 suggests that it is not strictly iodosobenzene. The PhIO prepared in batch 1 had been placed under dynamic vacuum for several months and was stored several years before use. Iodosobenzene is known to undergo disproportionation to iodobenzene and iodoxylbenzene (PhIO2) at elevated
168 temperature and when performed at reduced pressure, iodoxylbenzene can be isolated.
The prolonged exposure of batch 1 to reduced pressure at room temperature may promote a similar shift creating an increase in percent composition of iodoxylbenzene and may be responsible for not only the lower C and higher theoretical oxygen percentages of batch
1, but also its unique reactivity.
5.2.4 Carbonylation of 1,2- and 1,3-diols using sodium dichloroisocyanuric acid
While promising results were observed using iodosobenzene as the terminal oxidant (See Section 5.2.3) in the carbonylation of 1,3-diols, the large variation between batches of the oxidant motivated the pursuit of oxidants and conditions that behave consistently. Of the many oxidants screened, the most promising results were observed with sodium dichloroisocyanuric acid.44
Sodium dichloroisocyanuric acid (dichlor) has been extensively used as a stable chlorine source for sanitation, and, more rarely, it is used in chemical synthesis.45-48
Unlike trichloroisocyanuric acid (trichlor), which is commonly used as a chemical oxidant for alcohol oxidation,49-51 dichlor appears significantly less active for alcohol oxidation under our conditions and in the absence of catalyst yielded only small quantities (~ 15%) of 3-hydroxypropyl-3-hydroxypropanoate after several days at room temperature.
We used 1,3-propanediol as a test substrate for carbonylation on larger scale, to established standard carbonylation conditions using dichlor. While carbonylation of 1,3- diols still occurred at 1 atm of CO they showed greater sensitivity to the carbon
169 monoxide concentration in solution, which we alleviated by using slightly elevated
pressure (~ 20 psig). Increasing the concentration lead to higher conversions after 18 h
but a lower selectivity for the cyclic product (Table 5.6 entries 2, 4, and 5). Lowering the
acetate concentration generally increased the selectivity for cyclic products, and no linear
products were observed when acetate was excluded (entries 1-3). A third product was
observed by 1H NMR, integrating to approximately 10-20% of the product resonance.
We have tentatively assigned this product as hydroxypropyl functionalized isocyanuric
acids produced through the ring opening of trimethylene carbonate 2k with isocyanuric
acid generated during the reaction.52
Figure 5.6 Initial screening experiments with 1,3-propanediol
O (neocuproine)Pd(OAc)2 (5 mol %)_ HO O O HO OH O O + dichlor, CH3CN, NaOAc, mol sieves HO O CO (22 psig), 55 °C, 18 h entry [diol] NaOAc dichlor conv. selectivity (mM) (equiv) (equiv) (%)a (C/L)b 1c 0.065 2 0.9 65 3:1 2 0.065 1 0.9 56 8.5:1.5 3 0.065 0 0.9 100 1:0 4 0.131 1 0.9 67 4:1 5 0.263 1 0.9 72 3.1:1 6d 0.065 0 0.9 0 -
Reaction conditions: 1,3-propane diol, dichlor, NaOAc, (neocuproine)Pd(OAc)2 (0.32 a. mmol, 55 °C, CO (22 psig), CH3CN, 18 h. determined by the integration of cyclic (“C”) and linear (“L”) carbonates versus starting material (“S”) using the following equation: C + 2L conv = . b. percentage of total carbonates. c. reaction time is 36 h. d. No C + 2L + S catalyst was added.
!
170 The increased strain present in 6-membered carbonates53 resulted in increased presence of ring opened carbonate at 55 °C prompting us to drop the temperature to 35
°C. Additionally, prolonged reaction times for 1,3-diols generally resulted in lower isolated yields of the product. Analysis of the 1H-NMR of the filtered reaction mixture did not suggest the presence of additional biproducts. These observations suggest that the six-membered carbonate undergoes nucleophilic attack with isocyanuric acid generated during the reaction creating a linear carbonate which is insoluble in the reaction mixture.52 As a result the reaction time was shortened to 9 hours for 1,3-diol substrates.54
Optimum conditions established from the experiments above were applied to the carbonylation 1,2 and 1,3-diols (Tables 5.7 and 5.8). Generally, the isolated yields for the carbonylation 1,2-diols using dichlor were lower than those using NCS.
Carbonylation of ethylene glycol 1a gave the highest yield of 5-membered carbonates.
As observed previously in our 1H NMR screens using NCS, the assymetric diol 1d was better tolerated then the symmetric isomer 1c; supporting our previous hypothesis that a steric constraint is present with this catalyst. Pinacol 1e showed no reaction under these conditions. Using phenyl substituted diol 1g, a lower yield was obtained compared to that observed with NCS.. Finally, the vinyl substituted diol 1i yielded a complicated 1H
NMR upon workup with only trace amounts of the desired product and the olefinic region was noticably bare. These results suggest competing chlorination of the olefin by dichlor. Trichlor and dichlor have higher oxidation potentials than NCS,49 and reactivity differences between trichlor and NCS have been observed by Cohen and coworkers comparing the two oxidants in select allylic chlorinations.55 N-chlorosuccinimide is
171 therefore the preferred oxidant for diols bearing groups sensitive toward chlorination like olefins.
Table 5.7. Scope of 1,2-diol carbonylation O HO OH (neocuproine)Pd(OAc)2 (5 mol %) O O 1 R4 R 2 3 CO (20 psig), CH3CN,18 h, 35 °C R R R1 R4 sodium dichloroisocyanuric acid R2 R3 1 2 substrate/ entry R1 R2 R3 R4 yieldc product 1 1a/2a H H H H 61 2 1b/2b Me H H H 46 3 1c/2c Me H Me H 12 4 1d/2d Me Me H H 47 5 1e/2e Me Me Me Me NR 6 1g/2g Ph H H H 45 7 1h/2h CH2OH H H H 49 8 1i/2i Vinyl H H H trace Reaction conditions: 1,2-diol (6.57 mmol), (neocuproine)Pd(OAc)2 (5 mol %), sodium b dichloroisocyanuric acid (1.5 g, 1.1 equiv), CH3CN (50 mL), CO (20 psig), 35 °C, 18h Isolated by column chromotography on SiO2.
172 Table 5.8 Scope of 1,3-diol carbonylationa
entry substrate product yieldb O O 1 HO OH 40 1k O 2k O HO OH O 2 57 1l O 2l HO OH O O O 3 O 38 O O O 1m 2m O OH O O 4 Ph O 66 OH Ph O O 1n O 2n a Reaction conditions: 1,3-diol (6.57 mmol), (neocuproine)Pd(OAc)2 (5 mol %), sodium b dichloroisocyanuric acid (1.5 g, 1.1 equiv), CH3CN (50 mL), CO (20 psig), 35 °C, 18h. Isolated percent yield
Carbonylation of 1,3-propanediol 1k yielded trimethylene carbonate 2k in 40% isolated yield (Table 5.8, entry 1). Dimethyl substitution of the propanediol at the 2- position resulted in higher yield and conversion. This result is in contrast to those observed with hypervalent iodine species (see Section 5.2.3). Ketals such as 1m (entry 3) and benzyl esters such as 1n (entry 4) are also tolerated, affording the corresponding cyclic carbonates in yields of 41% and 63%, respectively.
5.2.5 Proposed mechanism for oxidative carbonylation of diols
The proposed mechanism for oxidative carbonylation of diols using
(neocuproine)Pd(OAc)2 is shown in Figure 5.1. Ligand exchange with CO, followed by attack of the diol on the electrophilic carbon atom and deprotonation, generates the
173 palladium-alkoxycarbonyl complex.23 Upon deprotonation of the alcohol and intramolecular cyclization, (neocuproine)Pd0 and the cyclic carbonate are generated.
The reoxidation of the Pd0 species by sodium dichloroisocyanuric acid, followed by protonation of the Pd-bound amide regenerates the active species. Following the first turnover, the identities of the monodentate ligands (“X”) of palladium cannot be definitively etablished since both chloride and acetate ions are present in the reaction mixture.
Giannocarro and coworkers propose a similar mechanism for the formation of a palladium alkoxycarbonyl intermediate,19-24 although CO insertion into a metal-alkoxy bond has also been proposed.56-61 Cyclic carbonate formation could also proceed through prior reductive elimination of the chloroformate followed by cyclization or a
Pd(II)/Pd(IV) cycle can also be envisaged.31 While both of these reductive elimination pathways are feasible, (neocuproine)Pd(OAc)2 itself is competent for oxidative carbonylation as observed by a single turnover experiment with 1g in the absence of exogenous oxidant or chloride.
HO - CO X- OH X
N N N N N N N N Pd Pd Pd O Pd O O X X OC X HX O HX O O N H - -
+ X = AcO or Cl Cl- O HO O N O Cl HX + X- O O N N O Pd O N N N Cl Pd0 O
Figure 5.1 Proposed mechanism for carbonylation of diols with (neocuproine)Pd(OAc)2.
174 5.3 Conclusion
In summary, we report a mild and general strategy for the Pd-catalyzed oxidative carbonylation of diols to generate both five- and six-membered cyclic carbonates. The reaction proceeds under exceptionally mild conditions, especially compared to the Cu(I)- catalyzed carbonylation of glycerol with CO.18 Sterically uncongested 1,2- and 1,3-diols are acceptable substrates for the catalyst. When substrates are employed that allow for cyclization to five-, six-membered rings, the catalyst exhibits high selectivity for the more thermodynamically stable, five-membered cyclic carbonate. The oxidative carbonylation of diols represents an attractive alternative to the use of phosgene for the synthesis of cyclic carbonates.
5.4 Experimental Procedures
5.4.1 General information
Reagent Information. Neocuproine (>99%), 1,10-phenanthroline (99+%), (!)-sparteine
(99%), N-chlorosuccinimide (98+%), 3Å molecular sieves, ethylene glycol 1a (99.8%),
1,2-propanediol 1b (99.5+%), 2,3-butanediol 1c (99%), pinacol 1e (98%), 1,2-butanediol
1f(98%), 1-phenyl-1,2-ethanediol 1g (97%), glycerol 1h (!99.5 %), (±)-3-butene-1,2-diol
1i ("98.5%, Fluka), 1,2,4-butanetriol 1j (95%), 1,3-propanediol 1k (99.6+%), neopentyl glycol 1l (99%), 2,3-0-isopropylidene-D-threitol lo (99%) and dichloroisocyanuric acid, sodium salt (96%) were purchased from Sigma-Aldrich and used as received. Anhydrous sodium acetate (99.4%, J. T. Baker), palladium (II) acetate trimer (99.2%, Pressure
Chemical), carbon monoxide (99.99%, Prax Air), acetonitrile-d3 (99.6 atom % D, Acros
175 Organics), acetonitrile (99.9%, Fisher Scientific), and p-xylene (99.4%, Fisher Scientific) were also used as received. Iodosobenzene,1 2-methylpropane-1,2-diol 1d,2 2,2- dimethoxypropane-1,3-diol 1m,3 and benzyl 3-hydroxy-2-(hydroxymethyl)-2- methylpropanoate 1n4 were synthesized according to known methods and purity was verified by 1H NMR spectroscopy.
5.4.2 Preparation of palladium complexes
(neocuproine)Pd(OAc)2. To a 100-ml round-bottom flask with stir bar was added neocuproine (0.600 g, 2.88 mmol, 1.1 eq.; FW = 208.27 g mol-1), palladium(II) acetate
(0.588 g, 2.62 mmol, 1.0 eq.; FW = 224.5 g mol-1), and acetone (55 ml), and the reaction mixture was stirred overnight. The yellow precipitate was isolated by vacuum filtration, rinsed with acetone, and dried under vacuum to afford 0.87 g of (neocuproine)Pd(OAc)2
(77% yield). This complex has been previously prepared, albeit by a slightly different route, and characterized.5
5.4.3 Carbonylation reaction conditions
(neocuproine)Pd(OAc)2 with N-chlorosuccinimide (Procedure A): To an oven dried round bottom flask (100 mL) with stirbar was added (neocuproine)Pd(OAc)2 (38.9 mg, 3 mol %, 0.09 mmol), N-chlorosuccinimide (0.6 g, 4.5 mmol, 1.5 equiv), NaOAc
(0.246 g, 3 mmol), molecular sieves (1.2 g), and CH3CN (18 mL). The flask was capped with a rubber septum and equipped with a balloon of CO. The headspace was purged over 10 minutes, after which time the diol (3 mmol, 1 equiv) was added and the solution was heated to 55 °C for 24 h. In the case of the solid diol 1b, the solution was purged
176 with CO for 10 minutes, the septum briefly removed and 1b was added quickly to the solution. The septum was reequipped and the flask was purged for another 10 minutes with a CO balloon prior to heating to 55 °C for 24 h. After 24 h the reaction was cooled and the solvent was removed under vacuum followed by purification by column chromatography (SiO2).
Pd(OAc)2 / (-)-sparteine with N-chlorosuccinimide (Procedure B): To an oven dried round bottom flask (100 mL) with stirbar was added Pd(OAc)2 (34 mg, 5 mol %,
0.15 mmol), (-)-sparteine (10 mol %, 0.3 mmol), N-chlorosuccinimide (0.6 g, 4.5 mmol,
1.5 equiv), NaOAc (0.246 g, 3 mmol), molecular sieves (1.2 g), and CH3CN (18 mL).
The flask was capped with a rubber septum and equipped with a balloon of CO. The headspace was purged over 10 minutes, after which time the diol (3 mmol, 1 equiv) was added and the solution was heated to 55 °C for 24 h. In the case of the solid diol 1b, the solution was purged with CO for 10 minutes, the septum briefly removed and 1b was added quickly to the solution. The septum was reequipped and the flask was purged for another 10 minutes with a CO balloon prior to heating to 55 °C for 24h. After 24 h the reaction was cooled and the solvent was removed under vacuum followed by purification by column chromatography (SiO2).
(neocuproine)Pd(OAc)2 with sodium dichloroisocyanuric acid (Procedure C):
To an vacuum oven dried Fisher-Porter bottle was added (neocuproine)Pd(OAc)2 (142 mg, 5 mol %), sodium dichloroisocyanuric acid (1.5 g), and CH3CN (50 mL). The vessel was assembled and placed in a mesh wire sleeve. The headspace of the vessel was
177 evacuated briefly, then pressurized to 20 psig with CO and stirred 10 minutes. After 10 minutes the vessel was vented, briefly opened, and the diol (6.57 mmol) was added. The vessel was charged with CO (20 psig) and vented a total of three time. Finally, the vessel was pressurized to 20 psig and stirred at 35 °C for 9-18 h. Following this time the reaction was cooled and the solvent was removed under vacuum followed by purification by column chromatography (SiO2). The purity of all compounds prepared by this method was established by elemental analysis.
5.4.4 Characterization of compounds
All cyclic carbonates reported herein have been previously prepared by other methods and characterized by 1H NMR: ethylene carbonate 2a,6-8 4-methyl-1,3-dioxolan-
2-one 2b,6-9 4,5-dimethyl-1,3-dioxolan-2-one 2c,7 4,4-dimethyl-1,3-dioxolan-2-one
2d,104,4,5,5,-tetramethyl-1,3-dioxolan-2-one 2e,7 4-ethyl-1,3-dioxolan-2-one 2f,9 4- phenyl-1,3-dioxolan-2-one 2g,6,9,11 4-(hydroxymethyl)-1,3-dioxolan-2-one 2h,6 4-vinyl-
1,3-dioxolan-2-one 2i,124-(2-hydroxyethyl)-1,3-dioxolan-2-one 2j.14 1,3-dioxan-2-one
2k,8,13 5,5-dimethyl-1,3-dioxan-2-one 2l,8 5,5-dimethoxy-1,3-dioxan-2-one 2m,3 benzyl
5-methyl-2-oxo-1,3-dioxane-5 carboxylate 2n,4a (3aR,8aR)-2,2-dimethyl-tetrahydro-
[l,3]dioxolo[4,5-e][l,3]dioxane-6-one 2o,43
ethylene carbonate 2a: Purified by column chromatography (SiO2, 40% EtOAc/hexanes).
1 H-NMR (CDCl3) !: 4.51 (s, 4H). Anal Calcd for C3H4O3: C, 40.92; H, 4.58. Found: C,
41.04; H, 4.36.
178
: 4-methyl-1,3-dioxolan-2-one 2b Purified by column chromatography (SiO2, 40%
1 EtOAc/hexanes). H-NMR (CDCl3) !: 4.85 (m 1H), 4.55 (dd, J = 7.8, 8.5 Hz, 1H), 4.02
(dd, J = 7.3, 8.5 Hz, 1H), 1.48 (d, J = 6.4 Hz, 3H). Anal Calcd for C4H6O3: C, 47.06; H,
5.92. Found: C, 46.87; H, 5.79.
4,5-dimethyl-1,3-dioxolan-2-one 2c: Purified by column chromatography (SiO2, 30%
1 EtOAc/hexanes). H-NMR (CDCl3) !: 4.33 (m, 2H), 1.45 (m, 6H) Anal Calcd for
C5H8O3: C, 51.72; H, 6.94. Found: C, 51.49; H, 6.82.
4,4-dimethyl-1,3-dioxolan-2-one 2d: Purified by column chromatography (SiO2, 40%
1 EtOAc/hexanes). H-NMR (CDCl3) !: 4.06 (s, 2H), 1.11 (s, 6H). Anal Calcd for C5H8O3:
C, 51.72; H, 6.94. Found: C, 51.48; H, 6.91.
4-phenyl-1,3-dioxolan-2-one 2g: Purified by column chromatography (SiO2, 10%
1 EtOAc/hexanes). H-NMR (CDCl3) !: 7.44(m, 3H), 7.36 (m, 2H), 5.68 (t, J = 8.0 Hz,
1H), 4.80 (dd, J = 8.2, 8.6Hz, 1H), 4.35 (dd, J = 7.9, 8.6 Hz, 1H). Anal Calcd for C9H8O3:
C, 65.85; H, 4.91. Found: C, 65.58; H, 4.83.
4-(hydroxymethyl)-1,3-dioxolan-2-one 2h: Purified by column chromatography (SiO2,
1 6:1 CHCl3/acetone). H-NMR (CD3CN) !: 4.74 (m, 1H), 4.46 (dt, J = 0.9, 8.4 Hz, 1H),
4.29 (ddd, J = 0.9, 6.0, 8.4 Hz, 1H), 3.77 (dq, J = 2.8, 4.7, 12.7 Hz, 1H), 3.57 (dq, J = 3.7,
179 5.4, 12.7 Hz, 1H), 3.26 (t, J = 5.7 Hz, 1H). Anal Calcd for C4H6O4: C, 40.68; H, 5.12.
Found: C, 40.68; H, 4.91.
4-vinyl-1,3-dioxolan-2-one 2i: Purified by column chromatography (SiO2, 40%
1 EtOAc/hexanes). H-NMR (CDCl3) !: 5.90 (ddd, J = 6.9, 10.5, 17.2, 1H), 5.51 (d, J =
17.2 Hz, 1H), 5.45 (d, J = 10.5 Hz, 1H), 5.12 (q, J = 7.6 Hz, 1H), 4.60 (dd, J = 8.3, 8.6
Hz, 1H), 4.16 (dd, J = 7.6, 8.6Hz, 1H). Anal Calcd for C5H6O3: C, 52.63; H, 5.30. Found:
C, 52.37; H, 5.27.
1 1,3-dioxan-2-one 2k: Purified by column chromatography (SiO2, 6:1 CHCl3/acetone). H-
NMR (CDCl3) !: 4.44 (t, J = 5.6 Hz, 4H), 2.14 (p, J = 5.6 Hz, 2H). Anal Calcd for
C4H6O3: C, 47.06; H, 5.92. Found: C, 46.79; H, 5.84
5,5-dimethyl-1,3-dioxan-2-one 2l: Purified by column chromatography (SiO2, 5%
1 acetone/CHCl3). H-NMR (CDCl3) !: 4.14 (s, 4H), 1.51 (s, 6H). Anal Calcd for C6H10O3:
C, 55.37; H, 7.74. Found: C, 55.66; H, 7.86.
5,5-dimethoxy-1,3-dioxan-2-one 2m: Purified by column chromatography (SiO2, 10%
1 acetone/CHCl3). H-NMR (CDCl3) !: 4.29 (s, 3H), 3.33 (s, 6H). Anal Calcd for C6H10O5:
C, 44.45; H, 6.22. Found: C, 44.41; H, 6.18.
benzyl 5-methyl-2-oxo-1,3-dioxane-5 carboxylate 2n: Purified by column
1 chromatography (SiO2, 5% CHCl3/acetone). H-NMR (CDCl3) !: 7.36 (m, 5H), 5.22 (s,
180 2H), 4.71 (d, J = 10.8 Hz, 2H), 4.20 (d, J = 10.8 Hz, 2H), 1.34 (s, 3H). Anal Calcd for
C13H14O5: C, 62.39; H, 5.64. Found: C, 62.12; H 5.74.
5.5 References
[1] Shaikh, A. A. G.; Sivaram, S. Organic carbonates. Chem. Rev. 1996, 96, 951-976.
[2] Clements, J. H. Reactive applications of cyclic alkylene carbonates. Ind. Eng.
Chem. Res. 2003, 42, 663-674.
[3] Csihony, S.; Mika, L. T.; Vlad, G.; Barta, K.; Mehnert, C. P.; Horvath, I. T.
Oxidative carbonylation of methanol to dimethyl carbonate by chlorine-free
homogeneous and immobilized 2,2'-bipyrimidine modified copper catalyst.
Collect. Czech. Chem. Commun. 2007, 72, 1094-1106.
[4] Rokicki, G. Aliphatic Cyclic Carbonates and Spiroorthocarbonates as Monomers.
Prog. Polym. Sci. 2000, 25, 259-342.
[5] Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Catalytic Asymmetric
Dihydroxylation. Chem. Rev. 1994, 94, 2483-2547.
[6] Emmanuel, L.; Shaikh, T. M. A.; Sudalai, A. NalO4/LiBr-mediated
diastereoselective dihydroxylation of olefins: A catalytic approach to the prevost-
woodward reaction. Org. Lett. 2005, 7, 5071-5074.
[7] Carloni, S.; De Vos, D. E.; Jacobs, P. A.; Maggi, R.; Sartori, G.; Sartorio, R.
Catalytic activity of MCM-41-TBD in the selective preparation of carbamates and
unsymmetrical alkyl carbonates from diethyl carbonate. J. Catal. 2002, 205, 199-
204.
181 [8] Yu, C.; Zhou, B.; Su, W.; Xu, Z. Chemoselective synthesis of asymmetrical
carbonate from alcohol and dimethyl carbonate catalyzed by ytterbium(III)
triflate. Synth. Commun. 2007, 37, 645-651.
[9] Okutsu, M.; Kitsuki, T. Process for the Preparation of Glycerol Carbonate. US
Patent 6,495,703. 2002
[10] Yoo, J. W.; Mouloungui, Z. Catalytic carbonylation of glycerin by urea in the
presence of zinc mesoporous system for the synthesis of glycerol carbonate. Stud.
Surf. Sci. Catal. 2003, 146, 757-760.
[11] Li, Q. B.; Zhang, W. Y.; Zhao, N.; Wei, W.; Sun, Y. H. Synthesis of cyclic
carbonates from urea and diols over metal oxides. Catal. Today 2006, 115, 111-
116.
[12] Kadokawa, J.; Habu, H.; Fukamachi, S.; Karasu, M.; Tagaya, H.; Chiba, K. Direct
condensation reaction of carbon dioxide with alcohols using trisubstituted
phosphine-carbon tetrabromide-base system as a condensing agent. J. Chem. Soc.,
Perkin Trans. 1 1999, 2205-2208.
[13] Aresta, M.; Dibenedetto, A.; Nocito, F.; Pastore, C. A study on the carboxylation
of glycerol to glycerol carbonate with carbon dioxide: The role of the catalyst,
solvent and reaction conditions. J. Mol. Catal. A: Chem. 2006, 257, 149-153.
[14] Mizuno, T.; Nakamura, F.; Egashira, Y.; Nishiguchi, I.; Hirashima, T.; Ogawa,
A.; Kambe, N.; Sonoda, N. Facile Synthesis of Carbonates from Alcohols and
Carbon Monoxide Promoted by Elemental Sulfur. Synthesis 1989, 636-638.
[15] Methoden der Organischen Chemie (Houben Weyl) Georg Theime Verlag:
Stuttgart, 1952; Vol. 8/3.
182 [16] Davies, A. G.; Hua-de, P.; Hawari, J. A. A. The Reaction of 1,3,2-
Dioxastannolans with Diacyl Chlorides: Decarbonylation in the Reactions with
Oxalyl Chloride. J. Organomet. Chem. 1983, 256, 251-260.
[17] Tam, W. Carbonylation of beta-Aminoethanols, Diols, and Diol Amines. J. Org.
Chem. 1986, 51, 2977-2981.
[18] Teles, J. H.; Rieber, N.; Harder, W. Preparation of Glyceryl Carbonate. US Patent
5359094. 1994
[19] Clark, H. C.; Dixon, K. R.; Jacobs, W. J. Chemistry of Metal Hydrides. V.
+ Preparation of Platinum Hydrides from [PtX(CO)(R3P)2] and
[Pt(X(COOR)(Ph3P)2]. J. Am. Chem. Soc. 1969, 91, 1346-1350.
[20] Byrd, J. E.; Halpern, J. Kinetics of Reversible Formation of Alkoxycarbonyl-
Platinum Complexes through Reactions of Alcohols with trans-
Chlorocarbonylbis(triphenylphosphine)platinum(II) I. J. Am. Chem. Soc. 1971,
93, 1634-1636.
[21] Giannoccaro, P.; Nobile, C. F.; Latronico, M. The Reactivity of Ni(II)-tertiary
Phosphine Complexes towards Alkoxides under Carbon Monoxide: Methoxo-
Carbonyl Complexes, Synthesis and Reactivity Inorg. Chim. Acta 1990, 175, 133-
139.
[22] Gabriele, B.; Mancuso, R.; Salerno, G.; Costa, M. An improved procedure for the
palladium-catalyzed oxidative carbonylation of beta-amino alcohols to
oxazolidin-2-ones. J. Org. Chem. 2003, 68, 601-604.
[23] Giannoccaro, P.; Cornacchia, D.; Doronzo, S.; Mesto, E.; Quaranta, E.; Aresta,
M. Interaction of PdCl2-2-(#-diphenylphosphine)ethylpyridine complex with diols
183 and CO: Synthesis of new alkoxycarbonyl complexes, key intermediates to cyclic
carbonates. Organometallics 2006, 25, 2872-2879.
[24] Giannoccaro, P.; Dibenedetto, A.; Gargano, M.; Quaranta, E.; Aresta, M.
Interaction of palladium(II) complexes with amino-alcohols: Synthesis of new
amino-carbonyl complexes, key intermediates to cyclic carbamates.
Organometallics 2008, 27, 967-975.
[25] Gabriele, B.; Mancuso, R.; Salerno, G.; Ruffolo, G.; Costa, M.; Dibenedetto, A. A
Novel and Efficient Method for the Catalytic Direct Oxidative Carbonylation of
1,2- and 1,3-Diols to 5-Membered and 6-Membered Cyclic Carbonates.
Tetrahedron Lett. 2009, 50, 7330-7332.
[26] Gabriele, B.; Costa, M.; Salerno, G.; Chiusoli, G. P. An Efficient and Selective
Palladium-catalysed Oxidation Dicarbonylation of Alkynes to Alkyl- or Aryl-
maleic Esters. J. Chem. Soc., Perkin Trans. 1 1994, 83-87.
[27] Conley, N. R.; Labios, L. A.; Pearson, D. M.; McCrory, C. C. L.; Waymouth, R.
M. Aerobic alcohol oxidation with cationic palladium complexes: insights into
catalyst design and decomposition. Organometallics 2007, 26, 5447-5453.
[28] Brink, G.; Arends, I. W. C. E.; Hoogenraad, M.; Verspui, G.; Sheldon, R. A.
Catalytic Conversions in Water. Part 23: Steric Effects and Increased Substrate
Scope in the Palladium-Neocuproine Catalyzed Aerobic Oxidation of Alcohols in
Aqueous Solvents. Adv. Synth. Catal. 2003, 345, 1341-1352.
[29] Arends, I. W. C. E.; ten Brink, G. J.; Sheldon, R. A. Palladium-neocuproine
catalyzed aerobic oxidation of alcohols in aqueous solvents. J. Mol. Catal. A:
Chem. 2006, 251, 246-254.
184 [30] Stahl, S. S. Palladium oxidase catalysis: Selective oxidation of organic chemicals
by direct dioxygen-coupled turnover. Angew. Chem., Int. Ed. 2004, 43, 3400-
3420.
[31] Whitfield, S. R.; Sanford, M. S. Reactivity of Pd(II) complexes with electrophilic
chlorinating reagents: Isolation of Pd(IV) products and observation of C-Cl bond-
forming reductive elimination. J. Am. Chem. Soc. 2007, 129, 15142-15143.
[32] Amadio, E.; Cavinato, G.; Dolmella, A.; Ronchin, L.; Toniolo, L.; Vavasori, A.
New carboalkoxybis(triphenylphosphine)palladium(II) cationic complexes:
Synthesis, characterization, reactivity and role in the catalytic
hydrocarboalkoxylation of ethene. X-ray structure of trans-
[Pd(COOMe)(TsO)(PPh3)2] " 2CHCl3. J. Mol. Catal. A: Chem. 2009, 298, 103-
110.
[33] Behr, A.; Eilting, J.; Irawadi, K.; Leschinski, J.; Lindner, F. Improved utilisation
of renewable resources: New important derivatives of glycerol. Green Chem.
2008, 10, 13-30.
[34] Cyclic carbonates and succinimide have similar volatility making purification by
sublimation or distillation problematic. These products have similar solubility
profiles and elute with similar retentions on silica gel making recrystallation or
column chromatography difficult.
[35] One possible reason for the difficulty experienced in scaling these reactions may
be the decrease in surface area/volume ratio upon going to larger reaction
volumes; lengthing the time required to saturate in CO and decreasing the rate at
which CO is replenished. Typical reactions were run at 167 mM diol while the
185 solubility of CO in organonitriles is approximately 7 mM at 55 °C (or
approximately 4 mol % relative to diol). See Gjaldbaek, J. C.; Andersen, E. K.
The Solubility of Carbon Dioxide, Oxygen, Carbon Monoxide, and Nitrogen in
Polar Solvents. Acta Chem. Scand. 1954, 1398-1413. .
[36] Saltzman, H.; Sharefkin, J. G. Iodosobenzene. Org. Synth. 1973, 5, 658.
[37] Cambie, R. C.; Chamgers, D.; Rutledge, P. S.; Woodgate, P. D. Cleavage of
Vicinal Diols by Iodine Triacetate and Iodine(I) Acetate. J. Chem. Soc., Perkin
Trans. 1 1978, 1483-1485.
[38] Nicolaou, K. C.; Adsool, V. A.; Hale, C. R. H. An Expedient Procedure for the
Oxidative Cleavage of Olefinic Bonds with PhI(OAc)2, NMO, and Catalytic
OsO4. Org. Lett. 2010, 12, 1552-1555.
[39] Beesley, R. M.; Ingold, C. K.; Thorpe, J. F. The formation and stabillity of spiro-
compounds. Part I. Spiro-compounds from cyclo-hexane. J. Chem. Soc., Trans.
1915, 107, 1080-1106.
[40] Ingold, C. K. The conditions underlying the formation of unsaturated and of
cyclic compounds from halogenated open-chain derivatives Part I Products
derived from a-halogenated glutaric acids. J. Chem. Soc., Trans. 1921, 119, 305-
329.
[41] Schleyer, P. V. Thorpe-Ingold Hydpothesis of Valency Deviation - Intramoledular
Hydrogen-Bonding in 2-Substituted Propane-1,3-diols J. Am. Chem. Soc. 1961,
83, 1368-1373.
[42] Jung, M. E.; Gervay, J. gem-Dialklyl Effect in the Intramolecular Diels-Alder
Reaction of 2-Furfuryl Methyl Fumarates: The Reactive Rotamer Effect,
186 Enthalpic Basis for Acceleration, and Evidence for a Polar Transition State. J.
Am. Chem. Soc. 1991, 113, 224-232.
[43] Wu, R.; Al-Azemi, T. F.; Bisht, K. S. Functionalized Polycarbonate Derived from
Tartaric Acid: Enzymatic Ring-Opening Polymerization of a Seven-Membered
Cyclic Carbonate. Biomacromolecules 2008, 9, 2921-2928.
[44] Both sodium dichloroisocyanuric acid and its dihydrate, which is commonly sold
for pool chlorination, were tolerated in the reaction. However, the use of the
hydrate required the addition of crushed molecular sieves.
[45] Staskun, B.; van Es, T. Production of 3-benzoyl-2,1-benzisoxazoles, 2-phenyl-4-
H-3,1-benzoxazin-4-ones, and Novel Quinolinone Derivates from 2-
Phenylquinolin-4-(1H)-ones and Sodium Dichloroisocyanurate. J. Chem. Soc.,
Perkin Trans. 1 1993, 511-516.
[46] Walters, T. R.; Zajac, W. W. J.; Woods, J. M. New Reagents for the Synthesis of
gem-Halonitro Compounds from Oximes. J. Org. Chem. 1991, 56, 316-321.
[47] Zakrzewski, J. A Simple Method for the Synthesis of the Sterically Hindered
Chloroamines. Synth. Commun. 1988, 18, 2135-2140.
[48] Mathew, T.; Tonne, J.; Sedelmeier, G.; Grund, C.; Keller, M.; Hunkler, D.;
Knothe, L.; Prinzbach, H. [6+6]Photocycloadditions in Face-to-Face
Benzo/Pyridazino Substrates - En Route to Azapagodanes. Eur. J. Org. Chem.
2007, 2133-2146.
[49] Tilstam, U.; Weinmann, H. Trichloroisocyanuric Acid: A Safe and Efficient
Oxidant. Org. Process Res. Dev. 2002, 6, 384-393.
187 [50] Hiegel, G. A.; Nalbandy, M. The Oxidation of Secondary Alcohols to Ketones
with Trichloroisocyanuric Acid. Synth. Commun. 1992, 22, 1589 - 1595.
[51] Mukawa, F. Steroids. XVII. The Oxidation of Hydroxysteroids with Isocyanuric
Chloride. Nippon Kagaku Zasshi 1957, 78, 450-452.
[52] Lunczak, J.; Weglowska, E. Synthesis of Polyetherols with isocyanuric Ring.
Kinetics and Mechanism of Reactions, Part 1: Reaction between Isocyanuric Acid
and Ethylene Carbonate. Int. J. Chem. Kinet. 2009, 41, 512-522.
[53] Six-membered cyclic carbonates possess a larger ring strain relative to five-
membered cyclic carbonates. This behavior is largely analogous to that found in
lactones. See for example, Clements, J. H. Reactive Applications of Cyclic
Alkylene Carbonates. Ind. Eng. Chem. Res. 2003, 42, 663-674 and references
therein.
[54] The pressure of the reaction vessel began to stabilize after 9 hours, suggesting the
reaction was near completion.
[55] Mura, A. J. J.; Bennett, D. A.; Cohen, T. A Simple Synthesis of 1-Thiophenoxy-
3-chloroalkenes: Useful Synthons for Acrolein and Methacrolein. Tetrahedron
Lett. 1975, 16, 4433-4436.
[56] Ruiz, J.; Martinez, M. T.; Florenciano, F.; Rodriguez, V.; Lopez, G.; Perez, J.;
Chaloner, P. A.; Hitchcock, P. B. Insertion reactions into Pd-O and Pd-N bonds:
Preparation of alkoxycarbonyl, carbonato, carbamato, thiocarbamate, and
thioureide complexes of palladium(II). Inorg. Chem. 2003, 42, 3650-3661.
188 [57] Macgregor, S. A.; Neave, G. W. Theoretical study of CO migratory insertion
reactions with group 10 metal-alkyl and -alkoxide bonds. Organometallics 2003,
22, 4547-4556.
[58] Bennett, M. A.; Yoshida, T. Synthesis and Reactivity of Zerovalent Platinum
Complexes of Short-Lived Cyclic Alkynes. Formation of Monomeric Platinum(II)
Hydroxo-, Methoxo, and sigma-Carbon-Bonded Complexes by Oxidative
Addition Reactions with Water, Methanol, and Activated Hydrocarbons and
Catalysis of Hydration of Nitriles to Amides J. Am. Chem. Soc. 1978, 100, 1750-
1759.
[59] Michelin, R. A.; Napoli, M.; Ros, R. Alkoxo and Hydroxo Complexes of
Platinum(II) - Novel Insertion Reactions into Platinum-Oxygen Bonds and
Condensation with Compounds Containing Active Hydrogen-Atoms J.
Organomet. Chem. 1979, 175, 239-255.
[60] Rees, W. M.; Atwood, J. D. Carbonylation of Metal-Oxygen Bonds. Formation
of Carbalkoxydicarbonylbis(triphenylphosphine)Iridium, ROC(O)Ir(CO)2(PPh3)2,
by Carbonylation of trans-Alkoxycarbonylbis(triphenylphosphine)Iridium, trans-
ROIr(CO)(PPh3)2 (R = Me, n-Pr, Ph) Organometallics 1985, 4, 402-404.
[61] Bryndza, H. E. Mechanism of CO Insertion into Metal-Oxygen Bonds. Reaction
of (DPPE)Pt(OCH3)R (R = CH3 or OCH3) with CO Organometallics 1985, 4,
1686-1687.
189
Chapter 6
Synthesis and Reactivity of Pt-Pt Dimers
6.0 Preface
This chapter describes our work toward the synthesis and reactivity of platinum(III)- platinum(III) dimers. I performed all synthetic work toward the formation of platinum dimers and carried out all studies on their reactivity. Computational analysis of the molecular orbitals of Pt dimers was done in a close collaboration with A. G. De Crisci, and P. V. Verma. Over the years, several people have helped me carry out surface and electrochemical studies. A. Devadoss and J. Prange performed all XPS measurements on surfaces. A. Devadoss prepared pyrolyzed photo-resists (PPF) and together we performed surface modification of these surfaces. A. Hosseini prepared self-assembled monolayers on Au(111) surfaces and together we performed additional surface modifications. While not explicitly involved in this project, C. C. L. McCrory provided helpful insight into electrochemistry and surface modification. 6.1 Introduction
The placement of two or more metal centers in close proximity to one another can allow for cooperative effects between metals. These effects can provide unique properties, which can be exploited to provide new reactivity, improved efficiency, or improved selectivity.1-6 Such bimetallic constructs are prominent in biological systems7
8-10 11 12 13 14 allowing for activation of small molecules such as O2, N2, H2O, CH4, and H2,
This chapter will discuss our studies of Pt-dimers to probe key steps in the conversion of methane to ethane or methanol.
190 Methane is the simplest hydrocarbon. At standard temperature and pressure it exists in the gas phase. Methane is the principle component of natural gas, typically making up over 90 % of the mixture. Proven natural gas reserves are in excess of 187 trillion cubic meters, exceeding known oil reserves.15 However, a large fraction of known reserves are located in remote locations. To effectively transport natural gas from these locations it must first be liquefied, requiring the gas to be cooled at -162 °C. This process accounts for almost half of the total cost of the fuel. Additionally, transportation of natural gas entails significant hazards.16 As a result, a direct process for the conversion of methane to liquid fuels is highly desirable.17
Methane has exceptionally strong C-H bonds (104.9 kcal/mol) in comparison to other saturated hydrocarbons and lacks functional handles for directed activation.18 In contrast, the oxidation products of methane typically possess functionality and weaker C-
H bonds making these products more reactive toward further oxidation.19 As a consequence the activation of methane and its subsequent selective functionalization to methanol, ethane, higher hydrocarbons, or other useful feedstock chemicals remains an important fundamental challenge in chemistry.20,21
The pioneering work of Shilov and coworkers demonstrated that simple platinum
(II) salts in acidic media facilitate C-H activation of methane.22 Furthermore, through the combination of K2PtCl4 and H2PtCl6, methane could be successfully converted to methanol catalytically.23 The general mechanism for the Shilov process is depicted in
Figure 6.1.18,24,25 First methane undergoes C-H activation by a Pt(II) species. The newly formed electron rich Pt(II)-methyl complex is then oxidized by Pt(IV),26 creating an
191 electrophilic Pt(IV) intermediate, allowing for nucleophilic attack of the Pt-CH3 bond by water yielding methanol and regenerating the Pt(II) active species.27-29
CH L L 4 L II L a) b) PtII Pt L L L CH3 K2PtCl4 (cat) K2PtCl6 (1 eq) IV CH4 CH3OH Pt CH3OH H2O L H O L L PtII 2 PtIV L CH3 L Figure 6.1. (a) the Shilov process for methane oxidation to methanol and (b) the proposed catalytic cycle for the reaction.
Periana and coworkers developed platinum complexes of amine ligands such as ammonia or bipyrimidine and investigated their use in the oxidation of methane to methanol (Figure 6.2). Both catalysts exhibited fast rates and excellent selectivities to methanol products.19 Complex 1 displayed faster initial rates than 2, but exhibited short catalyst lifetimes in the reaction solvent due to irreversible protonation of the NH3 ligands. Computational analysis of these systems suggests that oxidation of the Pt(II)- methyl complex may be rate limiting.30 The increased !-donating properties of the amine ligands relative to the !-donating/"-accepting bipyrimidine, yielded a lower barrier for
Pt(II)/Pt(IV) oxidation for 1, providing a rationale for the faster initial rates observed with
1 relative to 2.
N N H3N Cl Cl Pt Pt H3N Cl N N Cl
"catalytica" cis-platin system 1 2 Figure 6.2 Platinum complexes that facilitate the oxidation of methane to methanol
192 A key feature of the Catalytica system 2 is the use of oleum, a mixture of sulfuric acid and sulfur trioxide, which serves as the oxidant: eliminating the need for stoichiometric quantities of Pt(IV) salts. Furthermore, due to the solvent media, methylbisulfate is generated rather than methanol, protecting the product from further oxidation. Similarly, several research groups have explored the use of other oxidants,
31-36 such as O2, in the oxidation Pt(II)-methyls to Pt(IV)-methyls with varying success.
Polymethyl platinum complexes have been used to study reduction elimination processes relevant to the formation of ethane (methane homologation) and methanol
(Figure 6.3). Goldberg and coworkers have shown that selectivity for either process with
3 can be achieved with careful consideration of the reaction conditions.37 Five- coordinate platinum complexes, like 4, serve as key intermediates for either process.38 In the case of methanol formation, it is reasoned that a cationic five coordinate species is need to provide an electrophilic Pt-methyl capable of accepting an oxygen nucleophile.37,39
R Me Me P Me R N Me Pt Pt P Me N Me O2CMe P-P = dppe R=t-Bu 3 4 Figure 6.3 Trimethylplatinum complexes used to study the reductive elimination of ethane or methanol derivatives.
Platinum(III) dimers display unique reactivity not seen in monometallic complexes. Due to the fluidity of electrons in the Pt-Pt bond, the metals can behave as both Pt(II) and Pt(IV) species.40 For example, Matsumoto and coworkers have reported the consecutive attack of an olefin with two nucleophiles (Scheme 6.1).41 For this
193 reaction to take place, the dimer must first coordinate the olefin (complex 6), a process which is common for Pt(II) complexes but rare for Pt(IV) species. After the first nucleophilic attack a platinum alkyl is generated (complex 7). The Pt-alkyl undergoes a second nucleophilic attack. This is analogous to the final step in the Shilov reaction described earlier and characteristic of a Pt(IV)-alkyl species.
OR OR NO 3 2+ 4+ 3+ O III NH3 O III NH3 O III NH3 OR t-Bu Pt t-Bu Pt t-Bu Pt O NH3 O NH3 HOR O NH3 HOR HN III NH HN III NH HN III NH Pt 3 Pt 3 -H+ Pt 3 -H+ + t-Bu NH t-Bu NH t-Bu NH N 3 N 3 N 3 Pt(II) complex H NO3 H H 2 5 6 7 Scheme 6.1 Consecutive double nucleophilic attack on an olefin using a Pt(III)-dimer
Additionally, platinum dimers display reversible electrochemical oxidation in
2+ acidic solutions. For example, [(NH3)4Pt2(NHC(O)C(Me3)3)2] (8) shows a two electron oxidation at 0.76 V vs NHE. Similar behavior has been observed with a variety of bridged platinum complexes.42,43 The electrooxidation of platinum dimers is typically described as two consecutive one electron oxidations and examples are known where the processes are within 30 mV of each other and thus overlap.41 In contrast, the monomeric form of the dimer cis-(NH3)2Pt(NHC(O)C(CH3)3)2 (9) shows no oxidative peaks out to
1.5 V (vs. NHE). Electron rich platinum dimethyl or diphenyl complexes are typically electrochemically oxidized irreversibly with potentials that are comparable with those found in Pt-Pt dimers. This suggests that the presence of a second metal in the dimers significantly stabilizes higher oxidation states; making platinum dimers promising candidates for oxidation chemistry.
194
4+ Figure 6.4 Cyclic voltammogram of [(NH3)4Pt2(NHC(O)C(Me3)3)2] (8) (0.8 mM) in aqueous KNO3 (0.1 M, pH 1) at 25 mV/s.
Y L X X X Z O CH3 R Pt R R Pt R Pt X X X X O CH3 X Pt X X Pt Z R'N Pt CH3 R R R R R X X X Y X N CH3 R' L Paddlewheel Tribridge Cotton / Vrieze Dimers Dimers Dimers 10 11 12 + n X O NH3 O X R Pt R Pt O NH3 O X R'N Pt NH3 R'N Pt NH3 R R N NH3 N NH3 R' R' X
Pt-blue Mixed Dimers Dimers 13 5 / 8 Figure 6.5 Select examples of the structural variation in platinum dimers.
Dimers possessing two, three, and four bridging ligands have been reported,40,44-47 and can bear a large variety of ancillary ligands. For example, Cotton and Vrieze have extensively developed neutral Pt-dimers bearing four methyl groups and two bridging
195 ligands (complex 12 in Figure 6.5). Platinum-blue dimers (of general architecture similar to complex 5/8) have been extensively studied for many years. They were first discovered in 1895 and noted for their intense colors yielding vibrant blue, green, and red solutions.45,48
The large range of possible configurations, combined with their unique reactivity and known oxidation chemistry prompted us to pursue this class of compounds. This chapter will detail our progress toward the synthesis of mixed dimers like 13 possessing methyl and/or halide substituents synthesized through combination of Pt(II) complexes bearing bridging ligands with Pt(IV) complexes capable of accepting the bridging ligands. Following their synthesis we report initial reactivity studies using mixed dimers that are relevant to methane oxidation toward ethane or methanol derivatives.
6.2 Synthetic Strategies Toward the Formation of Pt(III)-
Pt(III) Dimers
Several routes have been established for the formation of Pt(III)-Pt(III) dimers.45
The earliest platinum dimers were synthesized in a stepwise fashion.49 First, simple
Pt(II) salts were added in the presence of bridging ligands such as sulfates, acetates, amides, or phosphates. Then, the intermediate dimeric species formed were oxidized to the desired oxidation state. This method is one of the most common synthetic strategies for the formation of Pt-dimers. The incorporation of ligands such as amines or methyl groups allows for the generation of complexes similar to 12 and 8.42,44,50 A major limitation to this strategy, however, is that both Pt nuclei typically share similar, if not identical, ligand environments and to date the synthesis of Pt complexes bearing
196 dramatically differing equatorial ligands like 13 has not been demonstrated using this method.
Platinum dimers may also be synthesized through conproportionation of platinum salts. In this strategy, a Pt(II) compound can be combined with a Pt(IV) species in the presence of bridging ligands to yield the desired Pt(III)-Pt(III) dimer. The feasibility of strategy has been demonstrated by Natile and coworkers through combination of K2PtCl4
47 with K2PtCl6 with amides to yield tribridge dimers 11. This pathway was also invoked by Goldberg and coworkers to account for the fortuitous generation of a Pt-dimer possessing a trimethyl-platinum unit.39
We envisioned that the formation of mixed dimers of general structure 13 could be formed through the combination of a Pt(II) complex possessing amide, amidine, or guanidine ligands, with a coordinatively unsaturated Pt(IV) complex (Figure 6.6).
Section 6.3 describes our efforts toward the synthesis of Pt(II) complexes possessing amide, amidine, and guanidine ligands. Section 6.4 discusses our synthesis of Pt(IV) complexes and Pt(II) complexes which can easily be oxidized after complexation to the lower fragment. Finally, Section 6.5 details our efforts toward the synthesis of dimeric compounds through the combination of the Pt(II) and Pt(IV) compounds described in
Sections 6.3-6.4.
X Y X Y R Pt R X Y X Y L X + Pt R'N Pt NH3 R'N Pt NH3 R R L X N NH3 N NH3 R' X R' X
Figure 6.6 Retrosynthetic analysis for the synthesis of platinum dimers.
197
6.3 Synthesis of Platinum(II) Monomers; Potential
Synthons to Pt-Pt dimers
Matsumoto and coworkers have reported the synthesis of Pt(II) complexes of the
51 general formula (NR3)2Pt(NHC(O)R’)2. In contrast to previous syntheses of Pt-dimers whereby the two equivalent platinum compounds are brought together by addition of a bridging ligand, this strategy discretely forms a single platinum complex possessing briding amide ligands, allowing for the independent introduction of a second metal. This is particularly attractive for the synthesis of hetero-bimetallic complexes and homo- bimetallic complexes where each metal possesses a unique set of ligands (e.g. 13).
O t + Bu 2 •ClO H 2 4 t AgClO4 1M NaOH Bu N NH3 Cl NH H2O NH3 RCN N NH3 Pt Pt 3 Pt Pt t H O NH N NH Bu N NH3 Cl NH3 H2O 2 3 3 2 h, rt H tBu 1 14 O 9
Scheme 6.2. Matsumoto’s synthesis of platinum complexes possessing amide briding ligands.
The Pt(II) bis-amide complex 9 can be prepared conveniently in two steps from
51 cis-platin. Addition of AgClO4 to an aqueous solution of cis-platin facilitates substitution of the two chloride ligands to form the di-aquo platinum complex and AgCl.
Filtration followed by addition of nitriles causes gradual precipitation of the dinitrile complex (R=tBu) as a white solid. Finally, the electrophilic nature of the metal bound
198 nitrile allows for hydrolysis in basic aqueous solutions to yield the neutral diamide complex 9.
Amines and alcohols can also serve as nucleophiles to coordinated electrophilic nitriles. For example, neutral substituted amidine complexes can be formed by addition
52 of methyl amine to (RCN)2PtCl2. More recently, Bokach and coworkers have shown similar reactivity using ammonia to generate unsubstituted amidine platinum dichloride complexes.53 The further addition of excess ammonia allows for the generation of bis- amine dicationic complexes.
R2 R1 N R R2 R N Pt N Pt N R1 H H Z-confirmation E-confirmation Figure 6.7. Possible imine configurations in platinum amidine complexes
Unlike amide ligands which serve as anionic ligands, guanidines and amidines, formed as described previously, coordinate as neutral ligands. The double bond resides on the imine bound to platinum giving the possibility of two stereoisomers (Figure 6.7).
Both Z and E confirmations have been reported. Typically the Z isomer is preferred when halides are present on the metal due to intramolecular hydrogen bonding with the
53,54 amine (R1 or R2 = H). When both R1 and R2 are alkyl groups the E isomer has been observed.52,55
Our attempts to purify and isolate cis-(RCN)2PtCl2 (R = Me, NMe2) complexes by established methods generally resulted in low selectivity for the cis complex.56,57
Problems obtaining cis-selectivity prompted us to consider if the route used by
Matsumoto and coworkers could be adapted to provide cis amidine and guanidine
199 complexes in the Z-geometry. Complex 14 was synthesized as previously reported.51
Unlike 14 (R=t-Bu), which precipitates from water, the addition of the acetonitrile or dimethylcyanamide resulted in highly water soluble complexes 15 and 16 that resisted attempts at precipitation. As a result, 15 and 16 were isolated as white solids by lyophilization of the water solution.
NHR' 2 •ClO R 2 ClO 4 • 4 H Cl NH i. H O, AgClO N NH NH R, dioxane/THF R N NH Pt 3 2 4 Pt 3 2 Pt 3 Cl NH ii. RCN N NH3 CH Cl , rt, 5 h R N NH 3 2 2 H 3 R NHR' 1 14 R = tBu 17 R = tBu, R' = H 15 R = Me 18 R = Me, R' = H 16 R = NMe2 19 R = NMe2, R' = H 20 R = tBu R' = Me Scheme 6.3 Synthesis of amidine and guanidine platinum bridging dimers.
Suspension of the dicationic dinitrile complexes 14 or 15 in dioxane/ammonia solutions yielded the corresponding dicationic diamidine complex over the course of 5 h at room temperature. Complexes 17 and 18 were isolated as white solids following
1 removal of solvent under reduced pressure and repeated washings with Et2O. The H
NMR was consistent with those previously reported for the methyl complex, displaying three broad resonances in a 2:2:2 ratio for each N-H present in the unsubstituted amidine.
The dicationic diguanidine complex 19 could be accessed similarly through the suspension of the bis-(dimethylaminocyanogen) complex 16 in a solution of dioxane/ammonia. The 1H NMR obtained upon work-up of this reaction matched those previously reported for the complex possessing Cl counterions.53
Analogously, addition of a methylamine solution in THF to the dinitrile 14, yielded the corresponding dicationic methyl-substituted amidine complex 20. Slow
200 evaporation of acetone solutions of 20 produced crystals suitable for X-ray analysis
(Figure 6.8) allowing us to assign the stereochemistry of the amidine and verify retention of the cis configuration. The two methyl-amine substituents of each amidine are oriented in a Z geometry and occupy different faces of the platinum complex giving the complex a
C2 symmetry.
The previously reported neutral chloride analogue trans-
Pt(Cl)2(NHCNH(CH3)CH3)2, posseses a geometry in which methyl groups of the amine face away from the Pt-metal, allowing a hydrogen bonding interaction between the N-H of the amidine and the chloride ligand.54 This interaction is believed to be responsible for the Z geometry observed in the imine. The dicationic version presented here lacks hydrogen bond accepting chlorides, but possesses a larger substituent (tBu vs Me). In this case, it appears that sterics favors the Z orientation placing the larger substituent away from Pt. This steric interaction may also account for the relative positioning of the amine-methyl group, which is oriented away from the tBu group and toward the platinum center. The positioning of the methyl group toward the platinum center may limit the formation of platinum dimers due to unfavorable interactions as a second platinum center approaches the bridging monomer, encouraging us to use unsubstituted amidines (R’=H in Scheme 6.3) for our attempts at dimer formation (see Section 6.5).
201
Figure 6.8. X-ray crystal structure of [[(CH3)3CCNH(NHCH3)]2Pt(NH3)2][ClO4]2, 20, with ellipsoids drawn at 50% probability. The unit cell contains two isostructural complexes and an acetone molecule. A Pt-Pt distance between complexes of 6.380 Å was observed. The second complex, hydrogen atoms, perchlorate counter ions, and solvent molecules have been omitted for clarity. Select bond lengths (Å) and angles (deg): Pt1- N3, 1.980(8); Pt1-N1, 2.000(10); Pt1-N5, 2.060(9); Pt1-N6, 2.067(8); N1-C1, 1.323(13); C1-N2, 1.281(13); C1-C2, 1.542(14); N2-C6, 1.473(13); N3-C7, 1.308(10); C7-N4, 1.329(15); C7-C8, 1.577(12); N4-C12, 1.411(17); N3-Pt1-N1, 90.6(4); N3-Pt1-N5, 177.3(4); N1-Pt1-N5, 88.2(4); N3-Pt1-N6, 87.7(3); N1-Pt1-N6, 177.2(4); N5-Pt1-N6, 93.6(3); C1-N1-Pt1, 134.2(7); N2-C1-N1, 123.8(9); N2-C1-C2, 116.8(9); N1-C1-C2, 119.4(9); C1-N2-C6, 128.8(10); C7-N3-Pt1, 135.3(7); N3-C7-N4, 123.0(9); N3-C7-C8, 119.1(7); N4-C7-C8, 115.3(9); N3-Pt1-N1-C1, 62.8(11); N5-Pt1-N1-C1, -114.8(11); N6- Pt1-N1-C1, 116(8); Pt1-N1-C1-N2, 14.4(18); Pt1-N1-C1-C2, -168.7(8); N1-C1-N2-C6, 12.2(19); C2-C1-N2-C6, -164.7(10); N1-Pt1-N3-C7, 53.6(12); N5-Pt1-N3-C7, 117(7); N6-Pt1-N3-C7, -124.2(12); Pt1-N3-C7-N4, 20(2); Pt1-N3-C7-C8, -179.4(8); N3-C7-N4- C12 8(3); C8-C7-N4-C12, -153.5(18).
As an aside, bimetallic complexes starting from trans-bridging complexes have previously been studied by Lippert and coworkers.58 In the case of bridging heterocycles, introduction of a second metal to bridging trans-complexes does not yield the analogous trans-binuclear complexes. Instead, the second metal adds such that the dx2-y2 of the metals are perpendicular to one another and a dative Pt-M bond is observed between the
202 dz2 of the bridging metal and one of the equitorial sites in the incoming metal. This prefered orientation has been attributed to the small bite angle of the bridging heterocycles, as ligands possessing larger bite angles do not orient in this fashion.59
Bridging amides, amidines, or guanidines are likely to exhibit this binding preference, and while these complexes may also yield interesting reactivity, we have not currently pursued their synthesis.
6.4 Synthesis of Pt(IV) and Pt(II) Complexes for use with
Pt(II) monomers
Several coordinatively unsaturated Pt(IV) complexes were considered in an effort to generate Pt(III)-Pt(III) dimers with bis-amide complex 9 or bis-amidine/bis-guanidine complexes 17-20. [PtMe3I]4, 21 is commercially available. PtMe3X (22, X = OTf or PF6)
60 can readily accessed from [PtMe3I]4 by the addition of the corresponding silver(I) salt.
Additionally, the commercially available Pt(II) salt K2PtBr4, 23, was selected as a precursor to the formation of mixed halide dimers which have been previously reported.61
Me Me Me Me Me Pt I O ClO4 Br Br Me H I Pt Me [PtMe X] H2O Me Me O Me MeI Pt Me 3 n K2PtBr4 Pt Pt Pt H2O Me Me O Me H Me Pt I Me OH2 OH2 OH2 Me Me 22a X = OTf 22b X = PF6 23 24 25 [Pt(CH3)3I]4 21 Figure 6.9 Bridge accepting platinum complexes used in this study.
Complexes 24 and 25 were prepared as previously reported starting from K2PtCl4.
62,63 Refluxing K2PtCl4 with norbornadiene (nbd) in a mixture of HCl/HOAc/H2O gave
203 64 the desired (nbd)PtCl2 complex in 67% yield. Methylation of (nbd)PtCl2 using 2.25 equivalents of methyl lithium gave the corresponding dimethyl complex in 60% yield, which served as a common precursor for both species.65
62 Addition of pyridine to (nbd)PtMe2 gave the dipyridine complex, which, upon exposure to oxygen in methanol, yielded the corresponding Pt(IV) methoxide product.62
Finally, 24 could be generated by refluxing in dilute perchloric acid.62 Alternatively
(nbd)PtMe2 can be oxidized by addition of bromine. The strained nature of nbd as a chelator in platinum complexes promotes its lability. As a consequence, after oxidation by bromine, nbd is released in favor of bridging bromides creating an oligomeric species
62 [PtMe2Br2]n. Addition of AgOTf to the polymer in H2O/acetone yielded the desired precursor 25.63
HCl/HOAc/H2O Cl MeLi Me pyridine N Me K2PtCl4 Pt Pt Pt nbd, reflux, 45min Cl Et2O, 0°C Me Et2O, rt N Me 67% 60% 64%
Br2 MeOH CH Cl 2 2 O2 89% 34% [PtMe2Br2]n Me O N Me AgOTf Pt H2O/Aceteone N Me OH Br H Br Me O Me HClO , H O Pt Pt 4 2 Me O Me reflux, 15min H OH2 OH2 25 Me O ClO4 H O Me 2 Pt H2O Me OH2 24 Scheme 6.4 Synthesis of potential bridge acceptors 24 and 25.
204 6.5 Synthesis and Characterization of Platinum Dimers
Using the conproportionation strategy described earlier, we sought to combine the monomeric Pt(II) and Pt(IV) complexes synthesized in Sections 6.3 and 6.4 to generate
Pt(III)-Pt(III) dimers with structures similar to mixed dimer 13.
6.5.1 Synthesis and characterization of trimethylplatinum dimers.
Our attempts at the formation of the neutral trimethyl dimers by combination of the Pt(II) bis amide complex 9 with [PtMe3I]4 21 yielded multiple species, which we were unable to separate. In contrast, mixing complex 9 with the cationic trimethylplatinum analogs 22a-b yielded a single species as evidenced by 1H-NMR. This proved an effective strategy toward the formation of cationic trimethyl platinum dimers
26a-b (Figure 6.10).
X O "Pt(CH3)3X" CH3 H O III CH3 t-Bu Pt t-Bu N NH3 CH Pt O 3 III t-Bu N NH3 thf HN Pt NH3 H t-Bu N NH3 O H 26a X=OTf 9 26b X=PF6 Figure 6.10 Synthesis of cationic trimethyl platinum dimers
Complex 26a was highly soluble in both polar and non-polar solvents due to its ionic nature and the presence of the lipophilic methyl and t-butyl groups. Our attempts to crystallize 26a from a variety of solvents were unsuccessful. By changing the counter- ion from OTf to PF6 we were able to grow crystals of the complex over the course of several days from a saturated pentane solution at -35 °C.
205 The crystals obtained were analyzed by X-ray diffractometry. The refined structure contained two crystallographically independent molecules of 26b in the asymmetric unit, one of which is presented in Figure 6.11. The dimer possesses a trimethyl platinum unit that adopts an octahedral geometry, in which it is coordinated to the two bridging amides ligands and the lower platinum. The lower platinum exhibits a square pyramidal geometry and lacks a second axial ligand. The square planes of each metal are at an angle of roughly 24° to one another. As can be seen in Figure 6.11b, the equatorial ligands are skewed slightly in the two metals, as evidenced by the dihedral angle of N1-Pt1-Pt2-C1 of -18.4 (3)°. A Pt-Pt bond length of 2.7231(5) was observed,
Figure 6.11. X-ray crystal structure of [(CH3)3Pt((CH3)3CCNH(NHCH3)]2Pt(NH3)2] [PF6], 26b, shown from the side (a) and from the top along the Pt2-Pt1 bond (b) with ellipsoids drawn at the 50% probability level. The structure contains two crystallographically independent molecules in the asymmetric unit. The second molecules, hydrogen atoms, and PF6 counterions have been omitted for clarity. Selected bond lengths (Å) and angles (deg): Pt(1)-N(3) 1.995(7), Pt(1)-N(4) 2.004(7), Pt(1)-N(2) 2.044(7), Pt(1)-N(1) 2.048(7), Pt(1)-Pt(2) 2.7231(5), Pt(2)-C(2) 2.013(9), Pt(2)-C(3) 2.021(8), Pt(2)-C(1) 2.033(8), Pt(2)-O(2) 2.141(6), Pt(2)-O(1) 2.152(5), O(1)-C(4) 1.261(9), O(2)-C(9) 1.259(9), N(3)-Pt(1)-N(4) 89.2(3), N(2)-Pt(1)-N(1) 91.7(3), C(3)- Pt(2)-Pt(1) 169.3(3), O(1)-C(4)-N(3) 122.8(8), O(2)-C(9)-N(4), 122.5(8)
206 which is within the range of covalent radii for the two metals66 and similar to platinum dimers possessing alkyl substituents,39,67,68 suggesting the presence of bonding interactions between the two metals.
1 The H-NMR of 26a in d6-benzene is shown in Figure 6.12. Two broad resonances for the NH and NH3 resonances are observed at 4.66 and 3.22 ppm integrating to 2H and 6H respectively. The two methyl resonances are observed for the axial and equatorial protons at 2.30 and 1.14 ppm. Both sets of methyl groups are split by 195Pt
2 t giving similar JPt-H of 77.6 and 76.0 Hz. Finally, the Bu was observed as a singlet at
1.10 ppm.
1 Figure 6.12. H-NMR of 26a in d6-benzene at 25°C.
207 Hall and co-workers have observed a correlation between the coupling constant of
69 the Pt-Me and the basicity of ligands for a series of [PtMe3(X)3] compounds. To further probe possible solution interactions with 26a, 1H-NMR’s were collected in a variety of solvents (Table 6.1). In all cases, two distinct methyl groups were observed. The axial methyl group consistently displayed a coupling constant of approximately 78 Hz. The equatorial methyl groups also varied slightly and remained near 76 Hz. The observation that neither coupling constant changes suggests that little change in the coordination sphere of the top platinum or bottom platinum occurs upon solvation.
1 2 Table 6.1 H-NMR chemical shift data and JPt-H coupling constants for 26a
mult CDCl3 (CD3)2CO (CD3)2SO C6D6 CD3CN D2O THF NH s(br) 4.85 5.58 5.62 4.66 5.05 5.95 5.48 a NH3 s(br) 3.45 4.04 3.89 3.22 3.25 - 3.86 1.98 1.88 1.72 2.30 1.83 1.77 1.86 Me(ax) s (77.3) (78.6) (78.4) (77.6) (77.9) (77.6) (78.1) 0.94 1.00 0.90 1.14 0.92 0.99 0.97 Me(eq) s (75.9) (76.4) (76.8) (76.0) (76.5) (76.7) (76.3) t-Bu s 1.10 1.12 1.03 1.10 1.07 1.09 1.11 a proton resonance exchanges with solvent after 0.5 h
The magnitude of the 195Pt coupling constants of the equatorial methyl groups is consistent with those previously reported for PtMe3X complexes where X are oxygen ligands. Fekl and coworkers correlated the coupling constant of Pt-Me compounds with the observed Pt-Me bond length observed in the solid state.70 Using this correlation, for a coupling constant of 76 Hz, one would anticipate a bond length of approximately 2.0, which is in good agreement with the values observed for this complex (Figure 6.11).
208
Figure 6.13 Axial-equatorial exchange in 26a at 50 °C measured by 1H-NMR through selective inversion of the axial methyl group.
To assess the stability of the dimer in solution we carried out selective inversion
1 experiments by H-NMR on 26a in the presence [Pt(CH3)3OTf]n 22a in d6-benzene from
25 to 50 °C. At 50 °C transfer between the axial and equatorial methyl resonances in the
1H NMR was observed (Figure 6.13). The inversion data was fit using equations 6.1 - 6.3, as specified in Section 6.10.7, yielding a rate constant of 5.06 s-1 for this process.
Transfer between the axial and equatorial methyl groups can be rationalized by several mechanisms: (i) ligand dissociation followed by Berry psuedorotation; (ii) dissociation of
PtMe3 into solution followed by reassociation; and (iii) reversible methyl group transfer from dimer 26a to exogenous 22a. No inversion transfer was observed between exogenous 22a and the axial methyl group of 26a at either temperature. The absence of such transfers suggests that dimer 26a is stable toward both methyl and PtMe3 exchange up to 50 °C. These data also suggest that the loss of a single ligand from PtMe3, generating a 5-coordinate species capable of Berry-psuedorotation is likely the predominant process responsible for the exchange of axial and equatorial methyl group.
209
195 Figure 6.14. Pt-NMR of 26a in d6-benzene at 25°C. Chemical shifts are reported relative to K2PtCl4 (-1622ppm)
The 195Pt{1H}-NMR spectrum contained two peaks centered at -1904 and -2466
71 ppm relative to a K2PtCl4 standard referenced to -1622 ppm (Figure 6.14). Both
1 resonances display a JPt-Pt coupling constant of 2083 Hz. Matsumoto and coworkers have reported a similar Pt dimer, 27, which possess an alkyl group in the axial position.68
In this report, platinum “a” was assigned to an observed resonance at -272 ppm while
1 platinum “b” was assigned to 2253 ppm and a JPt-Pt coupling constant of 2560 Hz was observed for both resonances. Based on analogy to this complex, we are assigning the resonance at -2466 to the lower platinum “c” in our system. The top platinum, “d” is assigned to -1904 ppm, which is similar to mononuclear cationic trimethyl-platinum complexes bearing oxygen ancillary ligands.69
210 O
+ 2 CH3 + b d O NH3 O CH3 tBu Pt tBu Pt O NH O CH a 3 c 3 HN Pt NH3 HN Pt NH3 tBu tBu N NH3 N NH3 H H 27 26
Complex 26a is isolated as a light colored solid displaying a variety of colors
(yellow, purple, red, and blue). Interestingly, a loss of color is observed when coordinating species are introduced. This phenomenon is evidenced in the UV-visible spectrum of the complex in CH2Cl2 and THF (Figure 6.15). At low concentrations (0.1 mM), the UV-vis spectrum is similar in both CH2Cl2 and THF. However, at higher concentrations (1.0 mM), a noticeable deviation arises between the two solvents. The
CH2Cl2 solution contains three peaks at 421, 505, and 695 nm that are absent in the THF solution. We are currently unable to assign the source of color in these complexes. While a slightly colored impurity may likely be the source, it is also possible that these colors are the result of the formation of tetrameric platinum complexes formed through dimerization of 26 that can easily be degraded by addition of a coordinating solvent.
211
Figure 6.15 UV-visible spectrum of 26a in CH2Cl2 (black) and THF (grey) at (a) 0.1 mM and (b) 1.0 mM
Thermal gravimetric analysis was carried out of 26a up to 320 °C as shown in
Figure 6.16. The relative mass loss has been scaled such that it is related to the molecular mass of the complex. The complex appears stable in the solid state up to approximately
165 °C. Above 165 °C a gradual loss of mass is observed which we to the loss of three methyl groups. Further increases in temperature causes a steep drop in mass, which we attribute to the loss of both triflate and the amines from the complex, followed by further decomposition above 200 °C which corresponds to the sequential loss of the amide ligands. Above 320 °C, the remaining mass correlates approximately with the mass of platinum metal in the sample.
We investigated the electrochemistry of complex 26a in acetonitrile. An irreversible oxidation peak was observed at +1.5V (Figure 6.17).72 The absence of a reverse wave during oxidation suggests the loss of one or more methyl groups from the complex upon oxidation.73 No reduction peaks were observed out to -1.5 V (vs
Ag/AgNO3).
212
Figure 6.16. Molecular weight mass loss as a function of temperature for 26a (starting molecular weight = 818.67, temperature ramp =10 °C/min)
Figure 6.17 Cyclic voltammogram of [(CH3)3Pt((CH3)3CCNH(NHCH3)]2Pt(NH3)2][OTf], 26a, (solid line) and background (dashed line) in CH3CN containing 0.1 M Bu4NPF6.
213 6.5.2 Synthesis of other amide bridged dimers
Our synthesis of tetrabromide dimer, 29 was achieved as previously reported.61
Combination of bis-amide 9 with K2PtBr4 in water yielded a blue solid which could be oxidized by addition of K2S2O8 to yield 29 as a red/brown solid. While the synthesis of the chloride congener is also known, we did not attempt its synthesis due to the low yields reported. 61
O Br H O III Br t-Bu Pt t-Bu N NH3 K2PtBr4 Br Pt O III t-Bu N NH3 Na S O HN Pt NH3 H 2 2 8 t-Bu H2O N NH3 O H Br 9 29
Scheme 6.5. Synthesis of mixed dimer 29
Combination of acidic solutions of fac-[(H2O)3Pt(OMe)(Me)2][ClO4] 24 with complex 9 yielded no reaction when observed by 1H NMR. The lack of reactivity observed could be due to residual pyridine remaining from the synthesis of 24. Residual pyridine resonances typically integrated to half of the methyl resonances, suggesting one pyridine present per molecule remained. Our efforts to further remove this pyridine, either by addition of perchloric acid or by concentration to induce precipitation of the pyridinium salt, have thus far been unsuccessful.
Similarly, combination of [(Me)2Pt(Br)(µ-OH)(H2O)]2 25 with 9 did not yield any
1 new resonances by H NMR after 20 h in a 1:1 D2O/d6-acetone solution. In contrast, dissolving a small amount of [Pt(CH3)3OTf]n (22a) in water followed by addition of complex 9 immediately yielded dimer 26a. While the exact reason for this difference in
214 reactivity is unclear, a possible explanation may be the lower trans effect of the bromide in 25 relative to the methyl group found in 22a.
6.5.3 Attempted synthesis of dimers bearing amidine or guanidine bridges
Our attempts to make amidine or guanidine analogs of trimethylplatinum dimer
26 have generally been unsuccessful. Introduction of [Pt(CH3)3OTf]n (22a) to solutions
1 of bis-amidine complex 17 in d6-acetone or THF yielded no new resonances in the H-
NMR, suggesting that the dimer was not formed under these conditions. The difference in reactivity between bridging complexes 17 and 9 with [Pt(CH3)3OTf]n may be attributed to the difference in charge between the two species. The difference in charge requires the loss of two protons to form the analogous mono-cationic dimer 26. To address this problem we attempted to add Brønsted bases to 17 in an attempt to form the neutral complex. Addition of two equivalents of sodium hydride (NaH) or sodium bis(trimethylsilyl)amide (NaHMDS) to 17 in d8-THF caused an instantaneous formation of a new species as observed by 1H-NMR spectroscopy. While 17 possessed four N-H resonances in a 2:2:2:6 ratio in accord with the predicted number of inequivalencies (3 different types of amidine N-H’s, and 1 type of N-H from the bound ammonia), the addition of base lead to a single resonance at 5 ppm integrating to 10 protons. Lowering the temperature of the NMR to -40 °C caused a splitting of the resonances into two peaks.
While these peaks were not resolved, the integration of the two resonances was estimated to be 8:2. Further cooling to -80 °C did not result in additional resolution. These
215 observations suggest that all N-H resonances are of similar pKa and in rapid exchange with one another.
1 Addition of [Pt(CH3)3OTf]n to the deprotonated species yielded a complicated H-
NMR spectra suggestive of the formation of several new species along with the recovery of small amounts of 17. The characteristic Pt-Me resonances, possessing a coupling constant of approximately 70 Hz, were difficult to observe and integrated to only fractional values of those expected relative to the tBu groups of the bridge. Our attempts to perform the addition of 22a to deprotonated solutions of 17 at -78 °C resulted in slightly cleaner spectra, but were still generally complicated with lower than expected integration values for the Pt-Me resonances.
The synthesis of the desired amidine- or guanidine-bridged dimers has thus far been unsuccessful. However, recent reports by the Natile74 and Kukushkin75 groups suggest that some success may be found in the synthesis of Pt(III)-Pt(III) dimers in protic
n+ solvents. Both [(X)2Pt(HN=C(NH2)R)2] 28a(X = Cl, R = CH3, n = 0) and 28b(X =
NH3, R = Et, n = 2) form dimeric platinum(III) complexes in protic solvents under aerobic conditions. For the neutral complex 28a the reaction occurred at room temperature in water or methanol over days, while the dication 28b required heating to
70 °C for 2 days. Previous results using the bis-amide 9 and [Pt(CH3)3OTf]n in water to form dimer 26a suggest that this path could be feasible for the synthesis of amidine or guanidine bridged analogues of 26 provided the solutions are sufficiently acidic.
Together these observations suggest that future attempts toward the synthesis of amindine and guanidine linkers should be carried out in protic media.
216 6.6 Computational Analysis of Pt-Pt Dimers
While the crystal structure of the trimethyl platinum dimer 26b provides structural and bonding information, it inherently lacks information regarding the molecular orbitals underlying such observables. To analyze the molecular orbital of complex 26 we pursued computational chemistry. Computational studies on Pt(III) species are scarce in the literature.61,76 However, computational studies on similar systems (e.g. isoelectronic
Rh(II)-Rh(II) dimers77 or Pt(II)-Pt(II) dimers78) are more common.
Geometry optimization in the gas phase was carried out on cationic trimethylplatinum dimer 26 using the crystal structure from Figure 6.11 as the starting geometry. Two different computational methods were used; M06-L, SDDf/6-31G(d,p) and B3LYP, LANL2D/6-31G(d). The M06-L density-functional is a newer method that has been reported to handle electron correlation better than classical B3LYP.79 Each density-functional employed a different basis set for heavy atoms. M06-L was used in combination with the SDD basis set, while B3LYP was used with LANL2DZ. The SDD basis set has been optimized to treat relativistic metals, such as platinum.80 For both computations light atoms were treated using 6-31G(d) basis sets. The M06-L computation included an addition p-function for all hydrogen atoms.
Selected bond lengths were compared to judge the validity of the computationally-derived, geometry-optimized structure (Table 6.2). The crystal structure described in Section 6.5.1 yielded two independent complexes, which we have arbitrarily assigned as Structure 1 and Structure 2. Both computational methods exaggerated the bond-lengths between the metal and its ligands. In almost all cases, however, the M06-L
217 functional provided a better correlation with the bond lengths observed in the crystal structures. A visual comparison of the computed structure and Structure 1 from the crystal structure is provided in Figure 6.18.
We utilized computational chemistry to analyze the frontier molecular orbitals involved in several Pt-dimers in hopes that this would provide additional insight into the
Table 6.2 Comparison of bond lengths between the crystal structure and geometry optimized structures of 26. Bond Structure 1 Structure 2 M06-L B3LYP PtMe_ax 2.032 2.021 2.042 2.059 PtMe_eq 2.011 2.033 2.030 2.049 PtMe_eq2 2.027 2.013 2.034 2.050 PtO_1 2.176 2.152 2.282 2.239 PtO_2 2.160 2.141 2.246 2.240 Pt-Pt 2.734 2.723 2.811 2.925 PtNH3_1 2.075 2.048 2.138 2.139 PtNH3_2 2.071 2.044 2.144 2.138 PtN_1 2.000 1.995 2.012 2.021 PtN_2 1.988 2.004 2.018 2.022 See Section 6.10.3 for additional details
Figure 6.18 Comparison of structure of 26 obtained from (a) the crystal structure and (b) the computationally derived geometry optimized structure using M06-L SDD/6-31g(d,p).
218 unique reactivity found in platinum dimers and to provide a context for the molecular orbitals observed in our computation studies on dimer 26. In addition to dimer 26, four other systems were chosen for this study; HH-(CH3COCH2)(Br)2Pt2(NH3)2(NHCOCH3)2
30;61 Cotton’s asymmetric tetramethyl platinum (III) dimer HH-
81 (py)[(Me)2Pt(NHCOCH3)]2 31; the symmetric analogue HT-[(py)(Me)2Pt(NHCOCH3)]2
81 32; and tetrabromide complex HH-(Br)4Pt2(NH3)2(NHCOCH3)2 29. In all cases, geometry optimizations were performed using M06-L correlation function and SDD/6-
31G(d,p) basis sets due to its higher fidelity with experimentally observed bond-lengths compared to B3LYP LANL2DZ/6-31G(d). Both singlet and triplet multiplicities were considered and in all cases the optimized singlet structure provided a lower energy. In general, good agreement was found for the bond lengths between the optimized structures and those obtained from the corresponding crystal structure for all the dimers considered in this study.82 The three highest occupied molecular orbitals and the three lowest unoccupied orbitals are displayed in Figures 6.20 through 6.24.
A simplified orbital depiction of bonding in d8-d8 Pt(II) dimers and d7-d7 Pt(III) dimers is provided in Figure 6.19.a.83 In co-facial square planar dimers, an interaction occurs between the two dz2 orbitals of two metals forming both bonding and antibonding molecular orbitals. The antibonding d!* molecular orbital is further stabilized by interactions with low-lying metal s and p orbitals. As a result the metal-metal interaction in d8-d8 dimers may be slightly bonding in nature. The removal of one electron from each metal allows for a further stabilization of the metal-metal bond through depopulation of the d!* orbital (Figure 6.19.b). In both cases the combination of dxy, dyz, and dxz orbitals between the two metals provides an equal number of bonding and antibonding orbitals
219 and thus are overall nonbonding in natural. However, the metal-metal antibonding interactions derived these orbitals should occupy the higher occupied molecular orbitals in both d8-d8 and d7-d7 dimers. This simplified orbital model serves as a basis for our discussion of the molecular orbitals observed through computation
Figure 6.19 Simplified orbital diagram for dimers of square planar complexes of (a) d8-d8 and (b) d7-d7 complexes.
220
Figure 6.20 Surface contour diagrams calculated by DFT (0.04 isodensity) showing the three highest occupied and the three lowest unoccupied molecular orbitals of 26a
Complex 26 has a LUMO metal-ligand antibonding orbital (Figure 6.20). The
LUMO+1 contains the d!* antibonding orbital and is very close in energy to the LUMO.
The LUMO+1 is also strongly antibonding with respect to the axial methyl group. It is worth noting that earlier calculations using the B3LYP method reversed the order of the
LUMO/LUMO+1 displayed here. Given the similarities in energy we cannot definitely identify which molecular orbital is the true LUMO. The HOMO, HOMO-1, and HOMO-
2 of 26 all contain metal-metal antibonding orbitals mixed with metal-ligand antibonding orbitals as expected given the simplified bonding picture from Figure 6.19.b.
221
Figure 6.21 Surface contour diagrams calculated by DFT (0.04 isodensity) showing the three highest occupied and the three lowest unoccupied molecular orbitals of 30.
Complex 30 is structurally similar to dimer 26; possesses an overall positive charge, an octahedral upper Pt unit with three anionic ligands, and a square-pyramidal lower platinum unit identical to that found in 26 (Figure 6.21). Like 26, the LUMO contains metal-ligand antibonding orbitals, however, unlike 26 this interaction is now localized on the top platinum unit instead of the bottom platinum unit. The LUMO+1 contains the d!* orbital and additional mixing with antibonding orbitals on the acetone substituent. Like 26, the LUMO and LUMO+1 are very similar in energy. The Pt-C bond of complex 30 is known to homolyze over time upon exposure to light.61 As photoexcitation is expected to populate the LUMO’s, and LUMO+1 contains metal-
222 acetone antibonding orbitals, our calculators are in good agreement with this observation.
While bond homolysis in light has not been observed in 26 or for the other complexes in this study, similarities in their unoccupied orbitals suggest that photochemistry similar to that observed with 30 or other platinum dimers may be possible.84 The HOMO of the acetone complex 30 is dominated by an acetone antibonding interaction but contains a small acetone-metal antibonding interaction. Like 26, the HOMO-1 and -2 of 30 contain metal-metal and metal-ligand antibonding orbitals (Figure 6.21).
Tetramethyl complex 31 maintains the octahedral/square-pyramidal coordination geometry of 26 and 30, but is a neutral complex. In contrast to 26 and 30, the d!* of the dz2 orbitals are now in the LUMO resulting in a Pt-Pt interaction that is overall bonding in nature (Figure 6.22). Low-lying pyridine antibonding orbitals are observed in the
LUMO +1 and LUMO +2. Like the other dimers reported in this study, the HOMO,
HOMO -1, and HOMO -2 all contain metal-metal and metal-ligand antibonding orbitals.
223
Figure 6.22 Surface contour diagrams calculated by DFT (0.04 isodensity) showing the three highest occupied and the three lowest unoccupied molecular orbitals of 31
The addition of a second pyridine ligand to 31 has been observed experimentally to induce a flip in the amide ligands to yield a head to tail complex 32 (Figure 6.23).81 In complex 32, both platinum atoms occupy an octahedral coordination environment. The addition of the second pyridine causes a shift in the LUMO energy levels, such that both pyridine antibonding orbitals are now lower in energy than the antibonding d!* of the dz2 orbitals, which occupies the LUMO +2. Complex 32 displays a similar HOMO,
HOMO-1 and HOMO-2 to that found for 31 (Figure 6.22).
224
Figure 6.23 Surface contour diagrams calculated by DFT (0.04 isodensity) showing the three highest occupied and the three lowest unoccupied molecular orbitals of 32
The tetrabromide complex 29 shares the coordination number and charge of 32, but differs in the placement of neutral and anionic ligands about the two platinum centers
(Figure 6.24). The HOMO and HOMO-1 contain ligand-metal antibonding interactions similar to those observed in complex 30 (Figure 6.21). The HOMO-2 is antibonding with respect to all bromides but contains a small metal-metal bonding interaction between the dxz and dz2 of the two platinum atoms. The LUMO through LUMO+2 are similar to those
225 observed in the other systems containing the d!* and metal-ligand antibonding interactions.
Figure 6.24 Surface contour diagrams calculated by DFT (0.04 isodensity) showing the three highest occupied and the three lowest unoccupied molecular orbitals of 29.
The computed molecular orbitals for the complexes used in this study largely agree with the simplified model discussed earlier. The antibonding of the dz2 orbitals occupies one of the lower unoccupied molecular orbitals and is usually found in the
LUMO or LUMO +1. A notable exception is observed in 32, where due to the low lying antibonding orbitals found in pyridine, the d!* is found in the LUMO +2. The higher
226 occupied molecular orbitals typically contain significant metal-metal and metal-ligand antibonding orbitals.
6.7 Reactivity Studies
Our reactivity studies on platinum dimers can be categorized into two sections:
(1) reductive elimination studies using the trimethyl platinum dimer 26a and (2) C-H activation and C-H functionalization of arenes using 29. These topics are covered in
Section 6.7.1 and 6.7.2 respectively.
6.7.1 Reactivity of trimethyl platinum dimer 26a
The trimethylplatinum dimer 26a was synthesized to study reductive elimination of Pt(III)-Pt(III)-methyl species toward the formation of ethane or methanol derivatives.
The reductive elimination of two methyl groups to form ethane is a sluggish process. As a result, these reactions are commonly heated to approximately 100 °C to induce reductive elimination. To test for this reactivity, 26a was dissolved in d6-benzene and heated to 100 °C in a sealable screw-cap NMR tube. After several days, only small amounts of ethane and methane were observed, suggesting this process may be feasible from 26a. However, upon repeating this experiment with a purer batch of 26a neither product could be observed. A similar result was observed upon heating this sample in nitrobenzene, a solvent previously shown to accelerate the reductive elimination of ethane, suggesting that our initial observations may have been due to impurities in the synthesis of 26a.
227 In attempts to speed up reductive elimination of ethane while discouraging oligomerization processes we explored several additives. The dimers are stable in 1 atm of CO at room temperature but upon heating to 100°C, CO facilitates reductive elimination to form ethane as well as acetone as minor products. The addition of triphenylphosphine to 26a in d6-benzene resulted in the instantaneous formation of a white solid and concomitant release of ethane as observed by 1H NMR. The solid was isolate by filtration and redissolved in CHCl3. Spectra data of the solid is consistent with
(PPh3)2PtMeOTf. Analogous addition of PPh3 to PtMe3thf1.5PF6 rapidly yielded a white solid and ethane. The white solid gave identical spectra. While the addition of PPh3 to
PtMe3thf1.5PF6 has previously been reported to produce ethane and the observed Pt(II) salt,85 we were surprised by the speed at which this transformation takes place, especially in comparison to thermally induced reductive elimination from Pt(IV).
We chose three nucleophiles (acetate, hydroxide, and methoxide) to investigate
SN2 reductive elimination in hopes of forming methanol or methanol-like derivatives.
Acetate resulted in complicated decomposition of 26a, with no formation of methyl acetate. Upon addition of 1 equivalent of sodium hydroxide to 26a in benzene a white solid immediately formed. The solid was identified as the tetrameric cluster [PtMe3OH]4 by 1H and 195Pt-NMR.86 This again suggests some lability of the top platinum when nucleophiles are introduced.. Addition of 1 equivalent of sodium methoxide did not yield dimethyl ether and we could not identify any of the resulting platinum species from the reaction.
The lack of activity toward reductive elimination of ethane is in good agreement with predictions related to the observed coupling constants of the Pt-Me groups.70,87
228 Furthermore, the instability of 26a toward nucleophiles makes this a problematic system to study further. As both of these problems directly relate to the bridging ligands, further exploration using stronger donating bridging ligands like the amidine and guanidines is desirable.
6.7.2 Reactivity of tetrabromide platinum dimer 29
Tetrabromide platinum dimer 29 has been reported to facilitate the C-H activation
- - - - - of acetone upon activation by AgX salts (X = NO3 , BF4 , PF6 , ClO4 , or CH3C6H4SO3
).61 Compound 29 has some structural similarity to the compounds developed by Periana and Shilov for C-H activation (Figure 6.25). As such, we have begun to investigate the reactivity of 29, in combination with silver salts, toward the C-H activation of hydrocarbons.
Br O Br O Br 2 L Cl t-Bu Pt t-Bu Pt Pt O Br 2 AgX O Br L Cl HN Pt NH3 HN Pt NH3 t-Bu t-Bu N NH3 N NH3 H Br H general structure of Pt complexes used in C-H activated species activation chemistry 29 Figure 6.25. Comparison of structure between an activated species, generated through the combination of 29 with silver salts, and complexes used previously for C-H activation.
To test for C-H activation with 29, we carried out screening studies in 2,2,2- trifluoroethanol (TFE) using AgOTf as the activator avoiding exposure to light whenever possible.61 Activation of 29 with two equivalents of AgOTf in TFE generated a deep red- brown solution which was used in our screening studies. Addition of benzene to this solution generated a new species, which we have tentatively assigned as the C-H
229 1 activated product 33. Analysis of H-NMR of 33 in d6-acetone shows three broad peaks that overlap with one another over the range of 7.25-6.9 ppm. Integration of these protons and comparison with the NH and NH3 resonances of the complex correlates roughly with the expected values (5:2:7). Lowering the temperature to -15 °C caused some separation of the three peaks, which integrated approximately to 1:1:2. Analysis by
ESI-MS yielded a peak at 861 m/z, corresponding to the expected mass of the 33 ion
[M]+. Similar results were observed with toluene, m-xylene, and ethylbenzene. In the case of toluene HR-MS was performed yielding the expected exact mass (HRMS (+):
Calcd for C17H33N4O2Br2Pt2: 873.0266; Found: 873.0261) and isotopic pattern.
Unfortunately, our attempts to crystallize or isolate these products have thus far been unsuccessful.
R
R OTf Br O Br O Br t-Bu Pt t-Bu Pt O Br O Br HN Pt NH3 CF CH OH HN Pt NH3 t-Bu 3 2 t-Bu N NH3 AgOTf (2 equiv) N NH3 H Br H 29 33 Figure 6.26 C-H activation of arenes by 29/AgOTf
We were curious as to whether the reactivity observed could be transferred to a catalytic reaction. Hydroarylation reactions have recently been shown to be promoted by
Pt salts88-91 and this reaction seemed a likely candidate for the catalytic use of 29.
Propargyl ether 34 was used to test the catalytic activity of 29 and the products were analyzed by GC/FID. In the absence of 29, AgOTf acts stoichiometrically with 34 to produce the chromene, 35. Addition of 1 equivalent AgOTf to 29 showed promising
230 activity approximately doubling the yield of 35. The addition of two equivalents resulted in high conversion of 34 along with formation of 35 and the furan 36 in approximately a
2:1 ratio. The formation of the 36 was not previously observed in hydroarylation reactions carried out by Sames and coworkers using simple platinum salts and suggests that a mechanism involving C-H activation followed by insertion of the alkyne may be operative.88,92 The addition of other silver salts or the transition to 1,2-dichloroethane88 solvent generally resulted in lower yields.
Table 6.3 C-H Funcationalization of propagryl aryl ether 34a
29
O AgX O O
34 35 36
CF3CH2OH ClCH2CH2Cl
[Pt2Br4] Entry AgX (mM) (mM) Conv, % Yield 35:36 Conv % Yield 35:36 1 AgOTf (30) 0 11 5:0 - - 2 AgOTf (15) 15 38 11:2 9 5:1 3b AgOTf (30) 15 80 22:11 100 18:7
4 AgPF6 (30) 15 41 15:2 5 2:0
5 AgClO4 (30) 15 72 24:6 65 17:1 6 AgNO3 (30) 15 37 6:0 20 1:0 a reaction were carried out in air at room temperature using 300 mM 34. b reaction performed in CF3CH2OH taken at 6 h, full conversion reached before 18 h.
Our initial reactivity studies show promise for the use of dimer 29 in C-H activation and C-H functionalization chemistry. Future studies of 29 will focus on
231 investigating H/D exchange reactions in CF3CH2OD and CF3CO2D such that we can benchmark 29 against existing Pt complexes in this reaction.93
6.8 Modification of Carbon Electrodes and the Synthesis of Clickable Ligands to support Pt Dimers
As part of an on-going collaboration with the Stack and Chidsey groups we have sought to develop electrocatalysts for the oxidation of hydrocarbons and related transformations. While 26a did not show any promising electrochemistry, traditional platinum blue dimers like 5/8 typically exhibit electrochemically reversible processes and an array of interesting oxidation chemistries (See Section 6.1).44 Over the course of our collaboration, Anando Devadoss in the Chidsey group developed methods for the formation of surface bound azides on carbon surfaces.94 These azides have been demonstrated to be suitable substrates for Cu(I) catalyzed Huisgen-type cycloadditions between terminal alkynes and azides commonly known as the “click” reaction.95
We sought to support platinum dimers on graphite carbon surfaces. Toward this aim we wanted to develop more robust bridging ligands. Espino et al. had previously developed chelating bridging ligands “esp” for Rh(II)-Rh(II) dimers which are isoelectronic to Pt(III)-Pt(III) systems.96 The use of modified esp ligands may thus prove suitable toward the formation of stable platinum dimers. The chelating diamide complex
ESPN, 37, was readily achieved in two steps by nucleophilic substitution using #,#,- dichloro-m-xylene and isopropylnitrile with LDA followed by controlled hydrolysis using concentrated sulfuric acid at 60 °C (Scheme 6.6).
232 O O NC CN H2SO4 H2N NH2 60 °C, 2h H H ESPN 60 % Scheme 6.6. Synthesis of ESPN
The diamide ESPN was used toward the synthesis of Pt blue complexes. Addition of ESPN to the diaquo complex [(NH3)2Pt(OH2)2][NO3]2 in water at pH 4 resulted in a green solution which precipitated a green solid upon addition of concentrated sodium nitrate. This behavior was analogous to that previously observed in the formation of Pt blues with pivalamide, pyridinone, and other briding amides, suggesting that ESPN is an acceptable bridging ligand for the formation of Pt-blue complexes.
To take advantage of the surface supporting methodology, we developed clickable versions of the ESPN. We envisioned the molecule could be easily formed by a late stage Sonogashira coupling of a halide-substituted analogue of ESPN. Our initial route toward the formation of the clickable ESPN relied on aromatic bromide derivative of compound 39 or 40. Despite repeated attempts, we were unable to generate the desired alkyne using standard Sonogashira procedures between the dinitrile or diamide precursors and trimethylsilylacetylene.
To alleviate this problem, we targeted the iodide derivative (Scheme 6.7) Starting from 1,3-bis-bromomethyl-5-iodobenzene, 38, the alkyl bromides were nucleophilically displaced using isopropylcyanate to generate the dinitrile complex 39. Subsequent hydrolysis of the dinitrile with concentrated sulfuric acid yielded the corresponding iodide-substituted ESPN, 40. In contrast to the bromides, the iodide underwent coupling
233 with trimethylsilyl-acetylene to yield the protected alkyne 41 in excellent yield. Finally, removal of the TMS group in basic methanol provided the desired clickable ethynylated-
ESPN 42 in 25% overall yield (4 steps).
O O NC CN H SO Br Br LDA, 0 °C 2 4 H2N NH2 60 °C, 2h CN 9 h 87 % I 38 % I I 38 39 40
O O O O PdCl2(PPh3)2 (2 mol %) CuI (1 mol %) H N NH H N NH THF / Triethylamine 2 2 THF/MeOH 2 2
TMS-acetylene Na2CO3 r.t. 2 d 77 % 98 % Si H
41 42 Scheme 6.7. Synthesis of clickable ligands to support Pt dimers
Exposure of 42 to acidic solutions of cis-[Pt(NH3)2(OH2)2][NO3]2 yielded an intractable brown solution, in contrast to ESPN, where the expected blue/green precipitate formed. We postulate that the platinum dimers do form, but that they subsequently react with the free alkyne in analogy to previously reported reactivity of Pt- blues and alkynes.97
To avoid possible problems due to the alkyne we attempted to first covalently attach the ligand to an azide modified pyrolyzed photo resist (PPF) followed by assembly of the platinum dimer on the surface. Azide modification of PPF was performed by the method of Devadoss et al.94 The azide functionality is typically observed by XPS as two peaks at 399 and 403 eV in a 2:1 ratio corresponding to the two periphery nitrogen atoms and the center nitrogen atom respectively. Upon formation of the clicked-triazole the two peaks condense into a single broad peak. In accord with these observations, exposure of
234 the PPF to 42 under standard click conditions (see Section 6.9.4) yielded a single broad resonance by XPS (Figure 6.27).
Figure 6.27 XPS data of the N(1s) region of a pyrolyzed photo-resist after clicking C (left) and the Pt(4f) region after exposure to a pH 3 solution of cis-(NH3)2Pt(OH2)2 (4mM) for 20 h (right).
The ligand-modified PPF was then exposed to solutions of cis-(NH3)2Pt(OH2)2 at pH 3 for 20 h. Analysis of the Pt(4f) region yielded two peaks in a 55:45 ratio, fits 1 and
2 respectively, suggesting that at least two platinum species may be present on the surface (Figure 6.27). Electrochemical analysis of the modified surface by cyclic voltammetry yielded no observable oxidation or reduction waves over the range of 200 to
1200 mV (NHE) in aqueous KNO3 (0.1 M, pH 1) where one would typically expect reversible, one-electron process for Pt dimers. Attempts to functionalize carbon surfaces by incubating with higher Pt (0.1 M) and KNO3 (0.1 M) concentrations, directly analogous to methods used to form the dimer in solution, met with similar results as observed both by XPS and cyclic voltammetry. Recent attempts to generate the Pt dimer using an analogous procedure on Au(111)-thiol self-assembled monolayers (SAMs)
235 covalently modified with 42 have also yielded no oxidation or reduction waves by cyclic voltammetry.
6.9 Conclusion
Dimeric platinum complexes exhibit a rich tapestry of inorganic and organometallic chemistry. Our studies toward the synthesis of mixed dimers like 13 yielded the formation of a new trimethylplatinum dimer 26 allowing us to study its structure and reactivity. We postulate that due to the weak coordinating ability of the carbonyl bridging ligands, that reductive elimination to form ethane is not feasible.
Likewise, the addition of nucleophiles generally resulted in the formation of monomeric products. Together these observations make a strong case for transitioning to amidine and guanidine bridges 17-20, which is the subject of on-going study.
The previously reported tetrabromide dimer 29 appears to promote C-H activation of a number of arenes and the C-H functionalization of aryl-propargyl ethers upon activation with AgOTf. Future studies will seek to benchmark 29 against other platinum complexes for these reactions.
6.10 Experimental
6.10.1 Instrumentation
1H and 13C NMR spectra were referenced to the solvent residual peak. 195Pt NMR spectra were referenced to a saturated, aqueous solution of K2PtCl4 (-1622ppm).
Electrochemical measurements were performed under an N2 atmosphere using a Pine
Instrument Company AFCBP1 bipotentiostat with a MSR rotator, an auxiliary Pt-mesh
236 electrode and a nonaqueous Ag/AgNO3 reference electrode or, in the case of aqueous electrochemistry, a saturated Ag/AgCl reference electrode. GC analysis was carried out on a HP 5890 equipped with an FID detector using an internal standard (decane or mesitylene). All product peaks were verified by co-injection with an authentic sample.
6.10.2 Materials
Benzene, THF, and pentane were dried and distilled from Na/benzophenone,
CH2Cl2 was dried over CaH2 and stored over 4Å MS. Unless otherwise specified all reactions were done under an atmosphere of N2 using standard Schlenk-line techniques.
51 51 60 98 62 63 61 88 99 9 , 14, 22a, 22b2 24, 25, 29, , 34, 1,3-bis-bromomethyl-5-iodobenzene , and
#,#,#’,#’-tetramethyl-1,3-benzenedipropionitrile96 were prepared as previously reported.
6.10.3 Computation details
DFT calculations were performed using Gaussian 09100 with the restricted M06-L correlation functional.79 SDD atomic basis sets with effective core potentials (ECP)80 were used as implemented by Guassian 09. To improve the accuracy of the calculation, f orbitals were added to platinum as augmented to the SDD basis set with coefficients of
0.9930 from the LANL08(f) basis set.101 The double-zeta 6-31G(d,p) was used on all lighter elements. All complexes were minimized as both singlet and triplet structures, in all cases the singlet structure was lower in energy. The HOMO (highest occupied molecular orbital) and LUMO (lowest unoccurpied molecular orbital) energies were determined using minimized singlet geometries to approximate the ground state. After minimization, vibration analysis was performed at the same level of theory to determine if imaginary frequencies were present. In all cases reported no imaginary frequencies
237 were observed. Molecular orbital visualizations were created using GaussView 4 using an iso-value of 0.04. Atomic coordinates are provided in Appendix A.
6.10.4 General procedures for surface functionalization and attachment
Pyrolyzed photoresist films were prepared as previously reported by Devadoss et al. and were subsequently exposed to a solution of IN3 in acetonitrile, which was freshly prepared by addition of ICl (20 uL) to a suspension of NaN3 (200 mg) in acetonitrile (10 mL). The surface was allowed to sit in the IN3 solution for 1 h in the dark and was subsequently rinsed with methanol and water. After washing, the azide modified surface was immediately submerged in a solution of the ethynylated ligand 42 (0.5 mM), CuSO4
(1 mM), sodium ascorbate (10 mM), and TTMA (1 mM) and allowed to sit overnight in the dark.
6.10.5 Synthesis of compounds
Synthesis of cis-[(NH3)2Pt(NCCH3)2][ClO4]2 (15)
Cis-platin 1 (0.5 g, 1.66 mmol), was added to a flask containing water (3 mL).
AgClO4 (0.688 g, 3.32 mmol) was added and the solution was allowed to stir in the dark overnight. The white precipitate was filtered off and CH3CN (1 mL) was added to the solution and allowed to stir for 4 h. The solvent was frozen and placed under high vacuum until all volatiles were removed to yield the desired product 15 as a white solid
1 (0.52 g, 61 %) . H-NMR (d6-acetone) !: 4.97 (brd s, 6H), 2.75(s, 6H) .
238 Synthesis of complex cis-[(NH3)2Pt(NCN(CH3)2)2][ClO4]2 (16)
The dimethylcyanamide was prepared in an analogous fashion to 15 using 1 (0.5 g, 1.66 mmol), AgClO4 (0.688 g, 3.32 mmol) and cyanamide (0.54 mL) yielding 16 a white solid (0.882 g, 93%).
Synthesis of cis-[(NH3)2Pt(NHC(NH2)C(CH3)3)2][ClO4]2 (17)
Complex 14 (0.5 g, 0.84 mmol) was added to a flask. The flask was capped with a septum and a solution of ammonia in dioxanes (0.5 M, 30 mL) was added via syringe.
The solution was allowed to stir for 3h at room temperature. The solvent was removed under vacuum and the compound was washed extensively with diethyl ether and isolated
1 as a white solid (0.43 g, 80%). H-NMR (d6-acetone): 7.16 (s, 2H), 6.81 (s, 2H), 6.38 (s,
2H), 4.06 (s, 6H), 1.27 (s, 18H)
Synthesis of cis-[(NH3)2Pt(NHC(NH2)CH3)2][ClO4]2 (18)
Complex 18 was prepared analogously to 17 using 15 (0.1g, 0.196 mmol) and
1 ammonia in dioxanes (0.5 M, 6 mL) yielding 18 as a white solid. H-NMR (d6-acetone)
!: 7.04 (d, 4H), 6.47 (s, 2H), 4.02 (s, 6H), 2.17 (s, 6H)
Synthesis of cis-[(NH3)2Pt(NHC(NHCH3)N(CH3)2)2][ClO4]2 (19)
Complex 19 was prepared analogously to 17 using 16 (0.5 g, 0.81 mmol) and NH3 in dioxanes (0.5 M, 30 mL) yielding a white solid (0.53 g, 100%) that shows 1H-NMR spectra similar to those seen by Bokach for the complex with chloride counterions.53 1H-
NMR (d6-acetone) !: 6.04 (s, 4H), 4.40 (s, 2H), 3.96 (s, 6H), 3.01 (s, 12H)
239
Synthesis of cis-[(NH3)2Pt(NHC(NHCH3)C(CH3)3)2][ClO4]2 (20)
A solution of methylamine in THF (2 M, 0.42 mL) was added to a suspension of complex 14 (0.100 g, 0.168 mmol) in CH2Cl2 (5.7 mL) and the solution was allowed to stir overnight. Removal of the solvent under reduced pressure yielded complex 20 as a
1 white solid (70 mg, 63%). H-NMR (d6-acetone) !: 6.89 (s, 2H), 6.07 (s, 2H), 4.17 (s,
6H), 3.96 (s, 6H), 1.26 (s, 18H)
Synthesis of [Pt2(CH3)3(NH3)2(NHCOC(CH3)3)2][OTf], (26a).
A solution of [Pt2(CH3)6(thf)3][OTf]2 (22a)(0.538mmol) in THF (2mL) was added to (NH3)2Pt(NHCOC(CH3)3)2 (0.025g, 0.538mmol) and stirred for 1 h. Solvent was removed in-vacuo. The residual was extracted with pentane and immediate filtered.
1 Removal of solvent gave a light green almost white solid. H-NMR (d6-benzene) !: 4.66
3 (s, 2H, HNCO), 3.22 (s, 6H, NH3), 2.30 (s, 3H, Pt-CH3(axial) J(Pt-H)= 77.6Hz), 1.14 (s,
3 13 6H, Pt-CH3 (equatorial) J(Pt-H)=76.0Hz), 1.10 (s, 18H, t-Bu); C-NMR (d6-benzene) !:
195 1 187.55, 39.88, 28.17, 3.77, -15.18. Pt-NMR (d6-benzene) !: -1904 (s JPt-Pt= 2083Hz,
1 PtC3O2), -2466 (s JPt-Pt= 2083Hz, PtN4); ESI-MS (+ ion): 669 (100% [M-OTf]); HRMS
Calculated for C13H35N4O2Pt2 : 669.2056 Observed: 669.2042 (Tolerance 5ppm).
Synthesis of [Pt2(CH3)3(NH3)2(NHCOC(CH3)3)2][PF6], (26b)
The complex was prepared analogous to 26a using PtMe3thf1.5PF6 (22b) instead of [Pt2(CH3)6(thf)3][OTf]2.
240 Synthesis of modified esp ligand - ESPN
#,#,#’,#’-tetramethyl-1,3-benzenedipropionitrile (1.44 g, 6 mmol) was added to concentrated sulfuric acid and the solution was heated to 60 °C. After 1.5 h the solution was neutralized to pH 7 with Na2CO3 and extracted with EtOAc (3 $ 300 mL). The combined extracts were dried with MgSO4 and the solvent removed under vacuum to
1 yield the diamide, ESPN (0.993 g, 60%) as a white solid. H-NMR (d6-acetone) !:
7.12(m, 2H), 7.04 (m, 1H), 6.52 (s, 2H), 6.08 (s, 2H), 2.80 (s, 4H), 1.12 (s, 12H).
3,3'-(5-iodo-1,3-phenylene)bis(2,2-dimethylpropanenitrile), (39)
The title compound was prepared in analogy to that previously reported by Du
Bois and coworkers.96 Diisopropylamine (0.59 mL, 4.188 mmol), and THF (5 mL) were added to a flame-dried flask. The flask was cooled to 0 °C and n-BuLi (1.7 mL, 2.5 M in hexanes) was added dropwise. After stirring 20 minutes, isobutyronitrile (289 mg, 4.188 mmol) was added dropwise and the solution was stirred an additional 1.5 h at 0 °C. After the stated time, a mixture of 1,3-bis-bromomethyl-5-iodobenzene and 1-bromomethyl-3- dibromomethyl-5-iodobenzene (0.568 g, 0.8 equiv total Br) in THF (5 mL) was added slowly via cannula and the reaction was then allowed to warm to room temperature and stirred for 9 h. The reaction was quenched by addition of H2O (20 mL) and then transferred to a separatory funnel containing 30 mL of EtOAc. The organic layer was washed with brine (20 mL) and then dried over MgSO4 and concentrated under reduced pressure. Purified by column chromatography (SiO2, 6:1 Hexanes/ EtOAc) yielded the product. (38 %)
241 3,3'-(5-iodo-1,3-phenylene)-bis(2,2-dimethylpropanamide), (40)
This reaction was carried out in analogy to that reported for ESPN. The dinitrile complex (1 g, 2.73 mmol) was dissolved in neat H2SO4 and heated to 60 °C. The reaction was allowed to cool to room temperature and was neutralized to pH 7 with
Na2CO3. The aqueous solution was extracted with EtOAc (3 $ 150 mL) and the combined organic fractions were dried over MgSO4 and concentrated under reduced pressure to yield the diamide (0.957 g, 87% yield). . 1H-NMR (DMSO) !: 7.34 (d, J = 1.4
Hz, 2H), 7.05 (s, 2H), 6.93 (t, J = 1.4 Hz, 1H), 6.87 (s, 2H), 2.56 (s, 4H), 0.99 (s, 12H)
3,3'-(5-((trimethylsilyl)ethynyl)-1,3-phenylene)bis(2,2-dimethylpropanamide) (41)
In a flame dried flask under an N2 atmosphere was added the diamide (0.8 g, 1.99 mmol),
Pd(PPh3)2Cl2 (28 mg, 0.04 mmol), trimethylsilylacetylene (0.195g, 1.99 mmol), CuI (4 mg, 0.02 mmol), and THF (25 mL). After several minutes stirring, triethylamine (10 mL) was added. After 2 days at room temperature, all volatiles were removed under reduced pressure. The residue was extracted with EtOAc (200 mL), washed with H2O (3 x 50 mL) and then passed through a neutral alumina plug eluting with THF (98%). 1H-NMR
(DMSO) !: 7.06 (s, NH, 2H), 7.04 (s, Ar, 2H), 6.94 (s, Ar, 1H), 6.85 (s, NH, 2H), 2.68 (s,
4H), 0.99 (s, 12H), 0.2 (s, 9H)
3,3'-(5-ethynyl-1,3-phenylene)bis(2,2-dimethylpropanamide) (42)
The TMS-protected alkyne (100 mg, 0.268 mmol) was added to a flask containing
K2CO3 (185 mg, 1.34 mmol), THF (2 mL), and MeOH (1 mL) and stirred for 2 h. The solvent was removed under high vacuum. The crude material was purified by column
242 chromatography (SiO2, 2% MeOH in EtOAc) to yield the desired product (33 mg, 41 %).
1H-NMR (DMSO) !: 7.08 (s, 2H), 7.07(s, 2H), 6.95 (s, 1H), 6.86 (s, 2H), 4.07 (s, 1H),
2.69 (s, 4H), 0.99 (s, 12H). 13C-NMR (DMSO) !: 178.5, 138.4, 133.2, 130.9, 120.6, 84.1,
80.0, 45.0, 42.5, 25.2. Anal Calcd for C19H24N2O2!H2O: C, 67.9; H, 8.23; N, 8.80.
Found: C, 68.3; H, 7.52; N, 8.79
6.10.6 Thermolysis experiments
All thermolysis experiments were carried out in 5mm resealable screw-cap NMR tubes. Thermolyses carried at temperature higher than the boiling point of the solvent were performed in medium wall tubes submerged completely in a large oil bath. After the stated reaction time, the tube was removed from the oil bath, cooled, and analyzed by
1H-NMR.
6.10.7 Dynamic NMR experiments
Measuring axial-equatorial exchange in 2a by selective inversion of 1H-NMR resonances:
The sample was heated to 50°C in a Varian Inova 300Mhz NMR Spectrometer.
Rates of exchange were measured using the technique reported by Robinson et al.102 The
pulse sequence is "/2x(#a) - $1 - "/2±x - $2 -"/2±x, ±y – acquisition. In this pulse sequence, two frequencies (#a and #b) that exchange are selected. For our experiment, #a and #b are the axial methyl and the equatorial methyl signals in 2a respectively. The transmitter offset was set to #a and a delay period $1 was set to 1/(2|#a - #b|) (1.6 ms). The mixing time ($2) was arrayed from 2 ms to 5 s (Table 6.4). The intensities of the equatorial
243 methyl group were plotted against the delay time ($2) (Figure 6.13). The data was fit to
the equation described by Robinson et al (Equation 6.1)103 with variables % and & (Figure
a a 6.13). M0 was set to the maximum intensity recorded for the equatorial position and Me
was set to the initial
! !
A a A A Mz (t) = M0 [Me / M0 " #1 exp("$1t) " #2 exp("$2t)] (6.1)
! A B #1 + #2 +1/T1 +1/T1 k"1 = (6.2) 1+ Ke
! [B]e k1 Ke = = (6.3) [A]e k"1
! -1 value of our experiment. The experimental values of our fit were 11.25 ± 0.55 s (%1),
-1 0.671 ± 0.027 s (%2), -0.544 ± 0.013(&1), 0.547 ± 0.013(&2). These values were then put
A into Equation 6.2 along with the experimentally determined T1 values for A (T1 = 1.039
B ± 0.076 s) and B (T1 = 1.179 ± 0.060 s) at 50°C. The equilibrium constant was taken as ! unity (Ke = 1). As such k1 and k-1 via Equation 6.3 are equal, solving Equation 6.2 gives
! -1 a rate of 5.06 ± 0.28 s . It should be noted that as this process is intramolecular the T1
values recorded cannot be independent of exchange, which is undesirable, and this may
introduce additional error into our measurement.
244 Table 6.4 Numerical values for the intensities of equatorial methyl resonances and delay times $2 delay time ($2) intensity delay time ($2) intensity 0.0005 60.6 0.25 34.1 0.001 59.6 0.3 34.5 0.002 59.4 0.35 34.7 0.003 59.4 0.4 35.3 0.004 58.9 0.45 35.9 0.005 58.8 0.55 37.8 0.0075 57.9 0.65 39.7 0.01 57 0.75 41 0.015 55.5 0.8 41.5 0.02 54.2 0.95 43.5 0.025 53 1.25 46.9 0.035 50 1.4 47.9 0.045 48.7 1.6 49.7 0.055 47.1 1.8 50.7 0.065 44.8 2 51.9 0.08 42.5 2.8 54.6 0.1 40.2 3.5 56.5 0.12 38.6 4 58 0.16 36.1 5 58.3 0.2 34.7
6.11 References
[1] Broussard, M. E.; Juma, B.; Train, S. G.; Peng, W. J.; Laneman, S. A.; Stanley, G.
G. A Bimetallic Hydroformylation Catalysts - High Regioselectivity and
Reactivity through Homobimetallic Cooperativity. Science 1993, 260, 1784-1788.
[2] McCollum, D. G.; Bosnich, B. Cooperative bimetallic reactivity. Inorg. Chim.
Acta 1998, 270, 13-19.
[3] Trost, B. M.; Ito, H. A direct catalytic enantioselective aldol reaction via a novel
catalyst design. J. Am. Chem. Soc. 2000, 122, 12003-12004.
245 [4] Rosenthal, J.; Luckett, T. D.; Hodgkiss, J. M.; Nocera, D. G. Photocatalytic
oxidation of hydrocarbons by a bis-iron(III)-mu-oxo Pacman porphyrin using O2
and visible light. J. Am. Chem. Soc. 2006, 128, 6546-6547.
[5] Cook, T. R.; Surendranath, Y.; Nocera, D. G. Chlorine Photoelimination from a
Diplatinum Core: Circumventing the Back Reaction. J. Am. Chem. Soc. 2009,
131, 28-29.
[6] Rodriguez, B. A.; Delferro, M.; Marks, T. J. Bimetallic Effects for Enhanced
Polar Comonomer Enchainment Selectivity in Catalytic Ethylene Polymerization.
J. Am. Chem. Soc. 2009, 131, 5902-5919.
[7] Holm, R. H.; Kennepohl, P.; Solomon, E. I. Structural and Functional Aspects of
Metal Sites in Biology. Chem. Rev. 1996, 96, 2239-2314.
[8] Iwata, S.; Ostermeier, C.; Ludwig, B.; Michel, H. Structure at 2.8 Å resolution of
cytochrome c oxidase from Paracoccus denitrificans. Nature 1995, 376, 660-669.
[9] Solomon, E. I.; Sundaram, U. M.; Machonkin, T. E. Multicopper Oxidases and
Oxygenases. Chem. Rev. 1996, 96, 2563-2605.
[10] Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Functional Analogues of
Cytochrome c Oxidase, Myoglobin, and Hemoglobin. Chem. Rev. 2004, 104, 561-
588.
[11] Chan, M. K.; Kim, J.; Rees, D. C. The Nitrogenase FeMo-Cofactor and P-Cluster
Pair: 2.2 Å Resolution Structures. Science 1992, 260, 792-794.
[12] Yachandra, V. K.; Sauer, K.; Klein, M. P. Manganese Cluster in Photosynthesis:
Where Plants Oxidize Water to Dioxygen. Chem. Rev. 1996, 96, 2927-2950.
246 [13] Lieberman, R. L.; Rosenzweig, A. C. Crystal structure of a membrane-bound
metalloenzyme that catalyses the biological oxidation of methane. Nature 2005,
434, 177-182.
[14] Peters, J. W.; Lanzilotta, W. N.; Lemon, B. J.; Seefeldt, L. C. X-ray Crystal
Structure of the Fe-Only Hydrogenase (CpI) from Clostridium pasteurianum to
1.8 Angstrom Resolution. Science 1998, 282, 1853-1858.
[15] "BP Statistical Review of World Energy", BP, 2005, available online at
www.bp.com/statisticalreview.
[16] For example, a recent study by Sandia National Laboratory estimates that a
breach in a single tank of a commercial Liquefied Natural Gas (LNG) tanker
could cause secondary burns to individuals up to 0.8 miles away. See Luketa, A.;
Hightower, M.; Attaway, S. Breach and Safety Analysis of Spills Over Water
from Large Liquefied Natural Gas Carriers. SAND2008-3153, Sandia National
Laboratories , 2008.
[17] Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas: The Methanol
Economy; Wiley-VCH: Weinheim, 2006.
[18] Shilov, A. E.; Shul'pin, G. B. Activation and Catalytic Reactions of Saturated
Hydrocarbons in the Presence of Metal Complexes; Kluwer Academic Publishers:
Boston, 2000.
[19] Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.; Fujii, H. Platinum
Catalysts for the High-Yield Oxidation of Methane to Methanol Derivatives.
Science 1998, 280, 560-564.
247 [20] Arakawa, H.; Aresta, M.; Armor, J. N.; Barteau, M. A.; Beckman, E. J.; Bell, A.
T.; Bercaw, J. E.; Creutz, C.; Dinjus, E.; Dixon, D. A.; Domen, K.; DuBois, D. L.;
Eckert, J.; Fujita, E.; Gibson, D. H.; Goddard, W. A.; Goodman, D. W.; Keller, J.;
Kubas, G. J.; Kung, H. H.; Lyons, J. E.; Manzer, L. E.; Marks, T. J.; Morokuma,
K.; Nicholas, K. M.; Periana, R.; Que, L.; Rostrup-Nielson, J.; Sachtler, W. M.
H.; Schmidt, L. D.; Sen, A.; Somorjai, G. A.; Stair, P. C.; Stults, B. R.; Tumas,
W. Catalysis Research of Relevance to Carbon Management: Progress,
Challenges, and Opportunities. Chem. Rev. 2001, 101, 953-996.
[21] Lunsford, J. H. Catalytic conversion of methane to more useful chemicals and
fuels: a challenge for the 21st century. Catal. Today 2000, 63, 165-174.
[22] Golʹdshleger, N. F.; Tyabin, M. B.; Shilov, A. E.; Shteinman, A. A. Activation of
Saturated Hydrocarbons. Deuterium-Hydrogen Exchange in Solutions of
Transition Metal Complexes. Zh. Fiz. Khim. 1969, 43, 2174.
[23] Golʹdschleger, N. F.; Esʹkova, V. V.; Shilov, A. E.; Shteinman, A. A. Reactions
of Alkanes in Solutions of Chloride Complexes of Platinum. Zh. Fiz. Khim. 1972,
46, 1352.
[24] Shilov, A. E.; Shteinman, A. A. Activation of Saturated Hydrocarbons by Metal
Complexes in Solution. Coord. Chem. Rev. 1977, 24, 97.
[25] Goldman, A. S.; Goldberg, K. I. Organometallic C-H Bond Activation: An
Introduction. ACS Symp. Ser. 2004, 885, 1-43.
[26] Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. Oxidaiton of
2- Zeise's Salt by [PtCl6] : A Mechanistic Model for Hydrocarbon Oxidation.
Organometallics 1994, 13, 755-756.
248 [27] A reductive elimination process involving either bound water or chloride can also
be envisioned but stereochemical studies on Pt-alkyls support a nucleophilic
mechanism (see Ref's 28 and 29).
[28] Luinstra, G. A.; Labinger, J. A.; Bercaw, J. E. Mechanism and Stereochemistry
for Nucleophilic Attack at Carbon Platinum(IV) Alkyls: Model reactions for
Hydrocarbon Oxidations with Aqueous Platinum Chlorides. J. Am. Chem. Soc.
1993, 115, 3004-3005.
[29] Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. C-H
Activation by Aqueous Platinum Complexes: A Mechanistic Study. J.
Organomet. Chem. 1995, 504, 75-91.
[30] Paul, A.; Musgrave, C. B. A Detailed Theoretical Study of the Mechanism of
Energetics of Methane to Methanol Conversion by Cisplatin and Catalytica.
Organometallics 2007, 26, 793-809.
[31] Lin, M.; Shen, C.; Garcia-Zayas, E. A.; Sen, A. Catalytic Shilov Chemistry:
Platinum Chloride-Catalyzed Oxidation of Terminal Methyl groups by Dioxygen.
J. Am. Chem. Soc. 2001, 123, 1000-1001.
[32] Rostovtsev, V. V.; Labinger, J. A.; Bercaw, J. E.; Lasseter, T. L.; Goldberg, K. I.
Oxidation of Dimethylplatinum(II) Complexes with Dioxygen. Organometallics
1998, 17, 4530-4531.
[33] Rostovtsev, V. V.; Henling, L. M.; Labinger, J. A.; Bercaw, J. E. Structural and
Mechanistic Investigations of the Oxidation of Dimethylplatinum(II) Complexes
by Dioxygen. Inorg. Chem. 2002, 41, 3608-3619.
249 [34] Horváth, I. T.; Cook, R. A.; Millar, J. M.; Kiss, G. Low-Temperature Methane
Chlorination with Aqueous Platinum Chlorides in the Presense of Chlorine.
Organometallics 1993, 12, 8-10.
[35] DeVries, N.; Roe, D. C.; Thorn, D. L. Catalytic Hydroxylation using
Chloroplatinum Compounds. J. Mol. Catal. A: Chem. 2002, 189, 17-22.
[36] Geletii, Y. V.; Shilov, A. E. Catalytic Oxidation of Alkanes by Molecular
Oxidation. Oxidation of Methane in the presence of Platinum Salts and
Heteopoly Acids. Kinet. Catal. 1983, 24, 413-416.
[37] Williams, B. S.; Goldberg, K. I. Studies of Reductive Elimination Reactions to
Form Carbon-Oxygen Bonds from Pt(IV) Complexes. J. Am. Chem. Soc. 2001,
123, 2576-2587.
[38] Crumpton-Bregel, D. M.; Goldberg, K. I. Mechanisms of C-C and C-H Alkane
Reductive Eliminations from Octahedral Pt(IV): Reactions via Five-Coordinate
Intermediates or Direct Elimination. J. Am. Chem. Soc. 2003, 125, 9442-9456.
[39] Luedtke, A. T.; Goldberg, K. I. Reductive Eliminiation of Ethane from Five-
Coordinate Platinum(IV) Alkyl Complexes. Inorg. Chem. 2007, 46, 8496-8498.
[40] Natile, G.; Intini, F. P.; Pacifico, C. Diplatinum(III) Complexes: Chemical
Species More Widely Spread Than Suspected. In Cisplatin: Chemistry and
Biochemistry of a Leading Anticancer Drug; Lippert, B., Ed.; Verlag: Zürich,
1999.
[41] Lin, Y.-S.; Takeda, S.; Matsumoto, K. Consecutive Double Nucleophilic Attack
on an Olefin Promoted by a Platinum(III) Dimer Complex. Organometallics
1999, 18, 4897-4899.
250 [42] Hollis, L. S.; Lippard, S. J. Synthesis, Structure, and Electrochemical Behavior of
cis-Diammineplatinum(III) Dimers with Bridging !-Pyridonate Ligands. J. Am.
Chem. Soc. 1981, 103, 6761-6763.
[43] Matsumoto, K.; Matoba, N. Catalytic effect of cis-Diammineplatinum #-
Pyrrolidone Tan Adsorbed on a Platinum Electrode in Electrochemical Oxidation
of Water into Molecular Oxygen. Inorg. Chim. Acta 1988, 142, 59-67.
[44] Matsumoto, K. Inorganic and Organometallic Chemistry of Cisplatin-Derived
Diplatinum(III) Complexes. In Cisplatin: Chemistry and Biochemistry of a
Leading Anticancer Drug; Lippert, B., Ed.; Verlag: Zürich, 1999.
[45] Woolins, J. D.; Kelly, P. F. The Preparation and Properties of Compounds
Containing Pt(III). Coord. Chem. Rev. 1985, 65, 115-140.
[46] Matsumoto, K.; Sakai, K. Structures and Reactivities of Platinum-Blues and the
Related Amidate-Bridged Platinum(III) Compounds. Adv. Inorg. Chem. 1999, 49,
375-427.
[47] Pellicani, R. Z.; Intini, F. P.; Maresca, L.; Mesto, E.; Pacifico, C.; Natile, G. The
First Example of Dinuclear Platinum(III) Complex with Three Bridging Ligands.
Eur. J. Inorg. Chem. 2006, 1635-1642.
[48] Lippert, B. Platinum Blues: On the Way toward Unraveling a Mystery. In
Cisplatin: Chemistry and Biochemistry of a Leading Anticancer Drug; Lippert,
B., Ed.; Verlag: Zürich, 1999.
[49] Muraveiskaya, G. S.; Kukina, G. A.; Orlova, V. S.; Evtaf'eva, O. N.; Porai-
Koshits, M. A. Platinum(III) Sulfates with Metal-Metal Bond. Synthesis and
Structure. Dokl. Akad. Nauk. SSSR. 1976, 226, 596.
251 [50] Bancroft, D. P.; Cotton, F. A.; Falvello, L. R.; Schwotzer, W.
Tetramethyldiplatinum(III) (Pt-Pt) Complexes with 2-Hydroxypyridinato
Bridging Ligands. Inorg. Chem. 1986, 25, 763-770.
[51] Chen, W. Z.; Matsumoto, K. Pivalamidate-bridged dinuclear platinum and
platinum-palladium complexes: synthesis, NMR and X-ray structural
t characterization of cis-[Pt(NH3)2((NHCOBu) Bu)2]· 2H2O, and cis-
t [Pt(NH3)2((NHCOBu) Bu)2MLL']Xn (M = Pt, LL' = cod; L = L = dmso; L = Cl,
- - L' = dmso; M = Pd, L = L' = NH3; X = ClO4 , NO3 ). Inorg. Chim. Acta 2003, 342,
88-96.
[52] Belluco, U.; Benetollo, F.; Bertani, R.; Bombieri, G.; Michelin, R. A.; Mozzon,
M.; Pombeiro, A. J. L.; da Silva, F. C. G. Stereochemical Investigation of the
Addition of Primary and Secondary Aliphatics Amines to the Nitrile Complexes
cis- and trans-[PtCl2(NCMe)2]. X-ray Structures of the Amidine Complexes trans-
i i [Pt(NH2Pr )2{Z-N(H)=C(NHPr )Me}]Cl2・4H2O and trans-[PtCl2(NCMe){E-
N(H)=C(NMeBut)Me}]. Inorg. Chim. Acta 2002, 330, 229-239.
[53] Tyan, M. R.; Bokach, N. A.; Wang, M. J.; Haukka, M.; Kuznetsov, M. L.;
Kukushkin, V. Y. Facile Cyanamide-Ammonia Coupling Mediated by cis- and
II trans-[Pt L2] Centers and Giving Metal-Bound Guanidines. Dalton Trans. 2008,
5178-5188.
[54] Bertani, R.; Catanese, D.; Michelin, R. A.; Mozzon, M.; Bandoli, G.; Dolmella,
A. Reactions of Platinum(II)-Nitrile Complexes with Amines. Synthesis,
Characterization, and X-ray Structure of the Platinum(II)-amidine Complex trans-
[PtCl2{Z-N(H)=C(NHMe)Me}2]. Inorg. Chem. Commun. 2000, 3, 16-18.
252 [55] Michelin, R. A.; Bertani, R.; Mozzon, M.; Sassi, A.; Benetollo, F.; Bombieri, G.;
Pombeiro, A. J. L. Cis Addition of Dimethylamine to the Coordinated Nitriles of
cis- and trans-[PtCl2(NCMe)2]. X-ray Structure of the Amidine Complex cis-
[PtCl2{E-N(H)=C(NMe2)Me}2]・CH2Cl2. Inorg. Chem. Commun. 2001, 4, 275-
280.
[56] Fraccarollo, D.; Bertani, R.; Mozzon, M.; Belluco, U.; Michelin, R. A. Synthesis
and Spectroscopic Investigation of cis and trans Isomers of
bis(Nitrile)dichloroplatinum(II) Complexes. Inorg. Chim. Acta 1992, 201, 15-22.
[57] Bokach, N. A.; Pakhomova, T. B.; Kukushkin, V. Y.; Haukka, M.; Pombeiro, A.
J. L. Hydrolytic Metal-Mediated Coupling of Dialkylcyanamides at a Pt(IV)
Center Giving a New Family of Diimino Ligands. Inorg. Chem. 2003, 42, 7560-
7568.
[58] Krumm, M.; Lippert, B.; Randaccio, L.; Zangrando, E. Nucleobase Complexes
with Metal-Metal Dative Bonds: Mixed Pt,Pd Compounds with Bridging 1-
Methylcytosinato Ligands and Unprecedented Short Pt(II)-Pd(II) Contacts. J. Am.
Chem. Soc. 1991, 113, 5129-5130.
[59] Hutton, A. T.; Shabanzadeh, B.; Shaw, B. L. [Pt2Me2(µ-Ph2PCH2PPh2)2][BF4]2, a
Highly reactive and Co-ordinately Unsaturated Di-platinum(II) Complex. Crystal
Structures of [Pt2Me2L2(µ-Ph2PCH2PPh2)2][BF4]2, L = MeCN or CO. J. Chem.
Soc., Chem. Commun. 1983, 1053-1055.
[60] Schlecht, S.; Magull, J.; Fenske, D.; Dehnicke, K. Trimethylplatinum Triflate: A
Versatile Building Block in Coordinationg Chemistry. Angew. Chem., Int. Ed.
1997, 36, 1994-1995.
253 [61] Matsumoto, K.; Arai, S.; Ochiai, M.; Chen, W.; Nakata, A.; Nakai, H.; Kinoshita,
S. Synthesis of the Pivalamidate-Bridged Pentanuclear Platinum(II,III) Linear
Complexes with Pt-Pt Interactions. Inorg. Chem. 2005, 44, 8552-8560.
[62] Appleton, T. G.; D'Alton, J. C.; Hall, J. R.; Mathieson, M. T.; Williams, M. A.
Synthesis of dimethylplatinum(IV) compounds, [{PtMe2X2}n], [{PtMe2XY}n],
+ and, in solution, fac-[PtMe2X(H2O)3] , where X and Y are anionic ligands. Can.
J. Chem. 1996, 74.
[63] Hall, J. R.; Swile, G. A. Dimethyl(IV) Compounds. IV. Preparation of
+ [Pt(CH3)2Br(H2O)3] and its Reactions with Neutral and Uninegative Ligands. J.
Organomet. Chem. 1977, 139, 403-418.
[64] Butikofer, J. L.; Hoerter, J. M.; Peters, R. G.; Roddick, D. M. Synthesis and
Reactivity of [(C2F5)2MeP]2Pt(Me)X (X = Me, O2CCF3, OTf, OSO2F): A
Reactivity Comparison with Chelate Acceptor Analogues. Organometallics 2004,
23, 400-408.
[65] Appleton, T. G.; Hall, J. R.; Williams, M. A. Displacement of Norbornadiene
(NBD) from PtMe2(NBD) by N-donors, Dimethylsulfoxide, and Cyanide, and
Reactions of cis-PtMe2L2 with Iodomethane. J. Organomet. Chem. 1986, 303,
139-149.
[66] Cordero, B.; Gómex, V.; Platero-Prats, A. E.; Revés, M.; Echeverría, J.;
Cremades, E.; Barragán, F.; Alvarez, S. Covalent Radii Revisited. Dalton Trans.
2008, 2832-2838.
254 [67] Ochiai, M.; Fukui, K.; Iwatsuki, S.; Ishihara, K.; Matsumoto, K. Synthesis of
Aryl-Platinum Dinuclear Complexes via ortho C-H Bond Activation of Phenol
and Transmetalation of Arylboronic Acid. Organometallics 2005, 24, 5528-5536.
[68] Matsumoto, K.; Nagai, Y.; Matsunami, J.; Mizuno, K.; Abe, T.; Somazawa, R.;
Kinoshita, J.; Shimura, H. A Synthetic Route to Alkyl-PtIII Dinuclear Complexes
from Olefins and Its Implication on the Olefin Oxidation Catalyzed by Amidate-
Bridged PtIII Dinuclear Complexes. J. Am. Chem. Soc. 1998, 120, 2900-2907.
[69] Agnew, N. H.; Appleton, T. G.; Hall, J. R. Multinuclear NMR Study of Some Di-
and Tri-methylplatinum(IV) Complexes. Aust. J. Chem. 1982, 35, 881-894.
[70] Fekl, U.; Eldik, R.; Lovell, S.; Goldberg, K. I. Effects of Trifluoromethyl
Substituents in a Tris(pyrazolyl)borate Ligand: A Structural and Spectroscopic
Study of Analogous Platinum(IV) Trimethyl Complexes. Organometallics 2000,
19, 3535-3542.
[71] There are two commonly practiced referencing systems for platinum.
195 Historically, Pt-NMR chemical shifts have been reported relative to H2PtCl6
which was referenced to 0 ppm (See for example: Pregosin, P.S. Platinum-195
Nuclear Magnetic Resonance. Coord. Chem. Rev. 1982, 44, 247-291). A newer
method has been developed that references to an absolute frequency of 21.4 MHz.
Using this method the H2PtCl6 resonance is observed at 4533 ppm. We have
chosen to reference to the older method, but conversion to the newer method can
easily be acheived by adding 4533 to the reported chemical shift. .
[72] The small anodic peak at 1.0 V was also observed in the background.
255 [73] Wik, B. J.; Tilset, M. Oxidatively Induced Reactions of a Diimine Platinum(IV)
Tetramethyl Complex. J. Organomet. Chem. 2007, 692, 3223-3230.
[74] Cornacchia, D.; Pellicani, R. Z.; Intini, F. P.; Pacifico, C.; Natile, G. Solution
Behavior of Amidine Complexes: An Unexpected cis/trans Isomerization and
Formation of Di- and Trinuclear Platinum(III) and Platinum(II) Species. Inorg.
Chem. 2009, 48, 10800-10810.
[75] Bokach, N. A.; Tyran, M. R.; Aleksandrov, G. G.; Haukka, M.; Kukushkin, V. Y.
III Diplatinum(III) [Pt2 Cl2(µ2-amidinate)4] compound derived from air oxidation
1 2+ of mononuclear cis-[PtII(NH3)2(amidine-% )2] complex. Inorg. Chem. Commun.
2009, 12, 1061-1063.
[76] Stephan, E.; Blake, A. J.; Davies, E. S.; McMaster, J.; Schröder, M. The structural
characterisation and elucidation of the lectronic structure of the mononuclear
3+ Pt(III) complex [Pt([9]aneS3)2] ([9]aneS3 = 1,4,7-trithiacyclononane). Chem.
Commun. 2008, 5707-5709.
[77] Brusten, B. E.; Chisholm, M. H.; D'Acchioli, J. S. Oxalate-Bridged Dinuclear M2
Units: Dimers of Dimers, Cyclotetramers, and Extended Sheets (M = Mo, W, Tc,
Ru, Rh). Inorg. Chem. 2005, 44, 5571-5579.
[78] Novozhilova, I. V.; Volkov, A. V.; Coppens, P. Theoretical Analysis of Triplet
4- Excited States of the [Pt2(H2P2O5)4] Ion and Comparison with Time-Resolved
X-ray and Spectroscopic Results. J. Am. Chem. Soc. 2003, 125, 1079-1087.
[79] Zhao, Y.; Truhlar, D. G. Density Functionals with Broad Applicability in
Chemistry. Acc. Chem. Res. 2008, 41, 157-167.
256 [80] Dolg, M. Effective Core Potentials. In Modern Methods and Algorithms of
Quantum Chemistry; John von Neumann Institute for Computering: Jülich, 2000;
Vol. 1, p 479.
[81] Peterson, E. S.; Bancroft, D. P.; Min, D.; Cotton, F. A.; Abbott, E. H. Proton,
Carbon-13 and Platinum-195 Nuclear Magnetic resonance Spectroscopy of
Hydroxypyridinate-Bridged Dinuclear Platinum(III) Complexes. Equilibria and
Mechanism of Bridging Ligand Rearrangement. Inorg. Chem. 1990, 29, 229-232.
[82] The standard deviation of bond lengths from the crystal structure and the
calculated structure was typically 0.03 Å and never greater than 0.05 Å.
[83] Bercaw, J. E.; Durrell, A. C.; Gray, H. B.; Green, J. C.; Hazari, N.; Labinger, J.
A.; Winkler, J. R. Electronic Structures of PdII Dimers. Inorg. Chem. 2010, 49,
1801-1810.
[84] Cook, T. R.; Surendranath, Y.; Nocera, D. G. Chlorine Photoelimination form a
Diplatinum Core: Circumventing the Back Reaction. J. Am. Chem. Soc. 2008,
131, 28-29.
[85] Romero, P.; Valderrama, M.; Contreras, R.; Boys, D. C-Pt(IV) Activation in New
Trimethylplatinum(IV) Complexes: Nucleophilic Attack at Metal-Carbon Bond.
J. Organomet. Chem. 2003, 673, 102-110.
[86] Appleton, T. G.; Hall, J. R. 1H, 13C, and 195Pt N.M.R. Specra of
Trimethylplatinum(IV) Tetramers [PtMe3Z]4: Effects of Pt-Z-Pt and Long-Range
Pt-Z-Pt-C Couplings. Aust. J. Chem. 1980, 33, 2387-2403.
257 [87] De Crisci, A. G.; Kleingardner, J.; Lough, A. J.; Larsen, A.; Fekl, U. A Weakly
Coordinating Anion as a Tripodal "Br3"-ligand for Platinum(IV) - Structure of
[(closo-CB11H6Br6)PtMe3]. Can. J. Chem. 2009, 87, 95-102.
[88] Pastine, S. J.; Youn, S. W.; Sames, D. PtIV-Catalyzed Cyclization of Arene-
Alkyne Substrates via Intramolecular Electrophilic Hydroarylation. Org. Lett.
2003, 5, 1055-1058.
[89] Oyamada, J.; Kitamura, T. Pt(II)-catalyzed hydroarylation reaction of alkynes
with pyrroles and furans. Tetrahedron 2009, 65, 3842-3847.
[90] Chatani, N.; Inoue, H.; Ikeda, T.; Murai, S. Ru(II) and Pt(II)-Catalyzed
Cycloisomerization of &-Aryl-1-alkynes. Generation of Carbocationic Species
from Alkynes and Transition Metal Halides and Its Interception by an Aromatic
Ring. J. Org. Chem. 2000, 65, 4913-4918.
[91] Oyamada, J.; Hashimoto, T.; Tsugio, K. A High Hydroarylation Activity of
K2PtCl4/AgOTf Catalyst in the Reaction of Propiolic Acid with Unactivated and
Activated Arenes. J. Organomet. Chem. 2009, 694, 3626-3632.
[92] Pastine, S. J.; Youn, S. W.; Sames, D. Pt(IV)-catalyzed cyclization of arene-
alkyne substrates via C-H bond functionalization. Tetrahedron 2003, 59, 8859-
8868.
[93] Hickman, A. J.; Villalobos, J. M.; Sanford, M. S. Quantitative Asssay for the
Direct Comparison of Platinum Catalysts in Benzene H/D Exchange
Organometallics 2009, 28, 5316-5322.
258 [94] Devadoss, A.; Chidsey, C. E. D. Azide-Modified Graphitic Surfaces for Covalent
Attachment of Alkyne-Terminated Molecules by "Click" Chemistry. J. Am.
Chem. Soc. 2007, 129, 5370-5371.
[95] While Sharpless originally described a series of click reactions (see Kolb, H. C.;
Finn, M. G.; Sharpless, K. B. Click Chemistry: Diverse Chemical Function from a
Few Good Reactions. Angew. Chem. int. Ed. 2001, 40 (11), 2004-2021) the Cu(I)
catalyzed Huisgen reaction is the only one that has become synonymous with the
"click reaction".
[96] Espino, C. G.; Fiori, K. W.; Kim, M.; Du Bois, J. Expanding the Scope of C-H
Amination through Catalyst Design. J. Am. Chem. Soc. 2004, 126, 15378-15379.
[97] Saeki, N.; Nakamura, N.; Ishibashi, R.; Arime, M.; Sekiya, H.; Ishihara, K.;
Mastumoto, K. Mechanism of Ketone and Alcohol Formation from Alkenes and
Alkynes on the Head-to-head 2-Pyridonato-Bridged cis-Diammineplatinum(III)
Dinuclear Complex. J. Am. Chem. Soc. 2003, 125, 3605-3616.
[98] Baldwin, J. C.; Kaska, W. C. Interaction of Hexaphenylcarbodiphosphorane with
the Trimethylplatinum(IV) Cation. Inorg. Chem. 1979, 18, 686-691.
[99] Alberico, D.; Rudolph, A.; Lautens, M. Synthesis of Tricyclic Heterocycles via a
Tandem Aryl Alkylation/Heck Coupling Sequence. J. Org. Chem. 2006, 72, 775-
781.
[100] Frisch, M., J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
259 Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven,
T.; Montgomery Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;
Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;
Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.;
Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09, Revision A.02; Wallingford CT, 2009
[101] Roy, L. E.; Hay, P. J.; Martin, R. L. Revised Basis Sets for the LANL Effective
Core Potentials. J. Chem. Theory Comput. 2008, 4, 1029-1031.
[102] Robinson, G.; Kuchel, P. W.; Chapman, B. E.; Doddrell, D. M.; Irving, M. G. A
simple procedure for selective inversion of NMR resonances for spin transfer
enzyme kinetic measurements. J. Magn. Reson. 1985, 63, 314-319.
[103] Robinson, G.; Chapman, B. E.; Kuchel, P. W. 31P NMR spin-transfer in the
phosphoglyceromutase reaction. Eur. J. Biochem. 1984, 143, 643-649.
260 Appendix A.
XYZ Coordinates for Geometry Optimized Pt Dimers
All coordinates are listed in the order X, Y, Z.
Complex 26 H 0.89564400 4.61003800 1.46799500
H 0.91669500 5.14766300 -0.98385300
H -4.36910400 2.82835700 0.86510200
H -5.10716500 0.55707600 1.74494300
H -2.76252300 2.53236000 1.55499100
H -0.32954700 3.93133400 -1.20363800
H -3.48157000 0.24242200 2.38196300
C 0.75585400 4.08634500 -1.19050700
H -0.23273300 3.25457200 1.28302800
C -3.38309400 2.39933900 0.66635500
C 0.82383300 3.54004600 1.25234500
H 3.09425500 4.61358800 0.16058200
C -4.12061300 0.15324800 1.50036200
H 1.15283900 3.89323000 -2.19315200
H 1.34543600 2.99343700 2.04079200
H -2.93516200 2.96837200 -0.15683100
261
H -5.41504800 1.18246600 -0.68569000
H -4.24438600 -0.91094500 1.27267700
C 1.44433000 3.24369900 -0.12040600
C -3.52191200 0.91496600 0.31093900
C 2.94280300 3.55950600 -0.08856000
H 3.45281900 2.95296400 0.66252900
C -4.42987700 0.76351000 -0.90653700
H 3.41354000 3.37856100 -1.06041000
H 0.03143400 1.95922800 -1.88759200
H -4.04725800 1.30298000 -1.78079600
C -2.10913400 0.36777700 0.08922400
H -4.58904800 -0.28611100 -1.17819700
C 1.28057800 1.74293200 -0.34031600
O -1.28492000 0.52289400 1.03166400
N 0.54741800 1.27729700 -1.34561900
H -2.52266500 -0.24771300 -1.76036200
N -1.80008800 -0.25994600 -1.05113000
O 1.86461000 0.95450600 0.46830400
H -0.06563200 0.32432100 3.46128500
H 1.72145800 0.39931600 3.19105800
C 0.83134500 -0.23453600 3.18102200
Pt 0.11031600 -0.67577700 -1.54956600
262 Pt 0.55293500 -0.75273300 1.22544700
H 0.96420300 -1.11343100 3.81398700
H -1.44974100 -2.38624600 1.05724000
H 2.50812100 -0.48541600 -2.75997300
H 2.63703000 -0.42436500 -1.12285300
H -1.41770600 -2.82474500 -1.49286800
N 2.20410200 -0.93288200 -1.89874100
C -0.66998600 -2.26464100 1.82094900
N -0.42287600 -2.74739500 -1.69629700
H -1.16077500 -2.00099500 2.76263800
H 3.08648000 -1.32451600 1.27693400
H -0.26958000 -3.18158100 -2.60361700
C 2.18892400 -1.94765200 1.35718800
H 2.58298300 -1.87579300 -1.89302700
H -0.12614400 -3.20507500 1.95688100
H 0.04936800 -3.32338800 -1.00137700
H 2.22385300 -2.49958000 2.30025600
H 2.18032400 -2.68455100 0.54109500
Complex 29 Pt 1.51628000 -0.06811700 -0.11810900
Pt -1.14666600 -0.02939800 0.12795300
Br -1.49919700 -1.89943900 -1.44346200
Br 4.23281800 0.02422800 -0.23036000
263
Br -3.62036100 0.02051900 0.72730100
Br -1.50594600 1.63635800 -1.65132600
C 0.41946900 -2.01986000 3.64508200
N 1.34528500 1.78280300 0.67837800
H 2.19336600 2.33247700 0.74508100
O -0.79199100 1.56761200 1.46168300
N 1.43465900 -0.89351800 1.72655200
H 2.28926800 -0.91843900 2.26920100
C 0.30695300 2.20635200 1.37274700
O -0.80703300 -1.34903300 1.75367800
C 0.34364900 -1.37795000 2.29378400
N 1.56861500 0.77195500 -2.05555900
H 1.07008900 1.66289500 -2.03647700
H 1.07911800 0.20440600 -2.74426000
H 2.54014600 0.90685600 -2.33645400
C 0.36214800 3.51667900 2.09333900
N 1.71126200 -2.02713000 -0.88295800
H 2.50983500 -2.09609700 -1.50964400
H 0.83957700 -2.32448600 -1.34087400
H 1.88840400 -2.65717000 -0.10450300
H -0.15899400 -1.42357800 4.35421100
H -0.04907300 -3.00489600 3.60322600
264 H 1.44309100 -2.12064300 4.00966200
H 1.36640600 3.94290900 2.11838000
H -0.31410000 4.21800600 1.59822300
H -0.00288700 3.38687700 3.11363900
Complex 30 C 3.25449400 1.10857500 0.73757400
O 3.10220500 1.98642500 1.56797400
C 0.62726700 3.15823200 -2.35136500
C 0.11912800 2.08429500 -1.44720700
N -1.09759300 1.58965900 -1.59527400
H -1.63905200 2.00800100 -2.34180400
N -2.06958600 -1.16077300 -1.88220600
H -1.13913700 -1.18648100 -2.31759300
H -2.24545500 -2.11075100 -1.56114300
H -2.75170500 -0.95761400 -2.60930700
O 0.94099100 1.67747600 -0.55571800
C -0.71392800 1.54416200 2.08086500
N -2.75377200 -1.06208700 1.12382200
H -3.57628700 -1.59148500 0.84478600
H -2.01541400 -1.72977800 1.38441600
H -3.00390500 -0.55076500 1.96790400
N -1.69541100 1.60229800 1.19240700
H -2.27538800 2.43098100 1.24192700
265
O 0.17581200 0.62697300 2.08211000
Pt -1.95162800 0.27331000 -0.31579800
Pt 0.71574700 -0.25950900 0.25734700
Br 1.36255200 -1.16905100 -1.92949200
Br 0.33058900 -2.50616300 1.15827100
C -0.59816800 2.56353400 3.16430200
C 2.68731100 -0.26603500 0.99112400
H 3.19895000 -1.08827200 0.48378100
H 2.52714400 -0.45812800 2.05396200
C 4.00386200 1.30234800 -0.54617800
H 4.18702000 2.36218100 -0.72298400
H 4.96609800 0.78073200 -0.49835000
H 3.45002700 0.86210000 -1.38230200
H -1.44111200 3.25487900 3.19439800
H 0.33101200 3.11929400 3.01158800
H -0.50606500 2.06112800 4.12896900
H -0.15412900 3.61089000 -2.96308700
H 1.39021600 2.72612800 -3.00580100
H 1.11861900 3.93161000 -1.75829500
266 Complex 31
Pt 0.57202900 -0.02438400 -0.42128400
Pt -1.99612200 -0.06028800 -0.12958300
O 0.61760500 1.54134100 1.15682000
O 0.65803000 -1.62396400 1.10623400
N -1.61413600 1.86482900 0.82142300
N -1.45508000 -1.05608000 1.73785700
N 2.65756600 -0.02223700 -0.28666600
C -0.46061800 2.20717000 1.34381500
C -0.32315500 3.43859900 2.20090700
C -0.32104100 -1.68437200 1.93089200
C -0.08628200 -2.52643300 3.15686000
C 3.32398800 -1.18933600 -0.27461900
C 4.69814800 -1.23690300 -0.09627500
C 5.39947200 -0.04835700 0.07966100
C 4.69719800 1.15212500 0.08275500
C 3.32258100 1.13067500 -0.10073100
C 0.59913900 -1.41495600 -1.92084900
C 0.57952200 1.47227700 -1.82091000
C -2.40446400 -1.89352800 -0.93767700
C -2.56086100 0.80142500 -1.89375800
H -1.26918700 3.96476600 2.34762900
H 0.07900500 3.16119200 3.17878800
267
H 0.39519600 4.12766000 1.74808600
H 2.70819600 2.02505200 -0.07888100
H 5.19983000 2.10200000 0.22905500
H 6.47621800 -0.05852100 0.21924900
H 5.20187500 -2.19747400 -0.09352500
H 2.71512200 -2.07918400 -0.39583400
H -0.25613100 -1.26078600 -2.58357800
H 1.52649200 -1.31958300 -2.49890100
H 0.53176400 -2.41053500 -1.47002200
H 0.07567300 2.34320900 -1.38935100
H 1.61532500 1.73033500 -2.07485300
H 0.05634500 1.16827400 -2.73108300
H -2.40503400 -1.89850600 -2.03328700
H -1.69903300 -2.65079400 -0.57701900
H -3.40979300 -2.18207100 -0.59240800
H -2.12673800 0.29329500 -2.76366000
H -2.28805400 1.86160600 -1.93251500
H -3.65503100 0.71480600 -1.97830300
H 0.17123700 -3.54598800 2.85840900
H 0.76814200 -2.13352300 3.71417700
H -0.95325300 -2.56531900 3.82021700
H -2.36230600 2.51009900 1.04430200
268 H -2.12189000 -1.19049300 2.48776100
Complex 32
Pt -1.29098800 -0.42849600 -0.05594900
Pt 1.29063800 -0.43166900 0.05266000
O -1.17436600 1.39841000 1.24559200
O 1.17577200 1.40878600 -1.23002600
N -0.96355100 0.85606800 -1.77989700
N 0.96427300 0.83611700 1.78939300
N -3.49712800 0.03424200 -0.10306900
N 3.49748400 0.03201400 0.10189700
C -0.13705700 1.53233100 1.97263800
C -0.23689300 2.54500600 3.08585200
C 0.13907700 1.55188500 -1.95622800
C 0.24171900 2.57419000 -3.06031900
C -4.45079600 -0.89575900 0.03889200
C -5.80126500 -0.57462200 0.08859300
C -6.17404000 0.76178000 -0.00453800
C -5.18174700 1.72782500 -0.13983800
C -3.85350800 1.32540300 -0.18254600
C 3.85112300 1.32382200 0.18444800
C 5.17858000 1.72903300 0.14373700
269 C 6.17301100 0.76536200 0.00723800
C 5.80307300 -0.57160100 -0.08901100
C 4.45322200 -0.89560700 -0.04108000
C -1.57812000 -1.56111500 1.63742000
C 1.47882200 -2.12992400 1.16845800
C 1.57407800 -1.55070100 -1.65042000
C -1.47847600 -2.11517700 -1.18882500
H 0.91109800 -2.95579100 0.72729300
H 2.53625800 -2.42494400 1.21646100
H 1.11542600 -1.95003800 2.18423600
H 1.00849700 -1.09988500 -2.47390100
H 2.63839800 -1.54528000 -1.92452100
H 1.24866100 -2.59070700 -1.53469000
H -2.53497500 -2.41375600 -1.23555900
H -1.11910900 -1.92313600 -2.20387000
H -0.90527700 -2.94255700 -0.75788400
H -2.63665800 -1.52364700 1.93067800
H -1.29077600 -2.61018000 1.50246200
H -0.98329900 -1.14031700 2.45599500
H 4.10633300 -1.92333300 -0.12300400
H 6.54391300 -1.35588500 -0.20414500
H 7.21978000 1.05277200 -0.02854400
H 5.42232200 2.78428000 0.21150600
270
H 3.02382300 2.02420300 0.26533600
H -4.10147100 -1.92284500 0.11864600
H -6.54038800 -1.36068100 0.20271100
H -7.22137800 1.04692600 0.03268900
H -5.42777200 2.78267800 -0.20518400
H -3.02801600 2.02790800 -0.26356800
H -1.68594000 1.05583600 -2.46181400
H 1.68741600 1.02897700 2.47247800
H 0.51837200 3.54403100 -2.63853500
H 1.04177400 2.28911300 -3.74888400
H -0.68528300 2.69024200 -3.62743300
H -1.04118400 2.25898300 3.76900200
H 0.68891200 2.65002800 3.65703700
H -0.50577400 3.52042700 2.67188800
271