Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Platinum group metal complexes of pyridyl- and anilinyldiphosphine ligands Jones, Nathan David 2001

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2001-714802.pdf [ 14.02MB ]
Metadata
JSON: 831-1.0061299.json
JSON-LD: 831-1.0061299-ld.json
RDF/XML (Pretty): 831-1.0061299-rdf.xml
RDF/JSON: 831-1.0061299-rdf.json
Turtle: 831-1.0061299-turtle.txt
N-Triples: 831-1.0061299-rdf-ntriples.txt
Original Record: 831-1.0061299-source.json
Full Text
831-1.0061299-fulltext.txt
Citation
831-1.0061299.ris

Full Text

P L A T I N U M GROUP M E T A L COMPLEXES OF P Y R I D Y L - A N D ANILINYLDIPHOSPHINE LIGANDS by N A T H A N DAVID JONES B.Sc.(Hons), The University of British Columbia, 1996 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE D E G R E E OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A September, 2001 © 2001 Nathan D. Jones UBC Special Collections - Thesis Authorisation Form Page 1 of 1 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Li b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I further agree that permission f o r extensive copying of this thesis f o r sc h o l a r l y purposes may be granted by the head of my department or by h i s or her representatives. It i s understood that copying or p u b l i c a t i o n of t h i s thesis f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of CntMiSTte"} The U n i v e r s i t y of Vancouver, Canada Date Al)oq | x«oi B r i t i s h Columbia http://www.library.ubc.ca/spcoll/thesauth.html 9/26/01 Abstract This thesis describes the synthesis, structure, solution behaviour, reactivity and catalytic properties of a collection of platinum metal complexes coordinated by pyridyl- or anilinyldiphosphine ligands; the ligands have the general formula Ar 2P(Z)PAr2 (for Ar = o-C6H4NMe2, Z = C H 2 (dmapm), (CH2)2 (dmape), cyclic-CsHg (dmapcp); for Ar = o-py, Z = (CH 2 ) 2 (dpype), cycl ic-C 5 H 8 (dpypcp); py = pyridyl). dpype dpypcp The ligands are made in 33-55 % yield by the low temperature reaction of C12P(Z)PC12 in E t 2 0 with 4 equiv. of 2-lifhiopyridine or 2-lithio-Af Af-dimethylaniline generated in situ from the appropriate 2-bromoarene and "BuLi. Dmape and dmapcp are characterised crystallographically. The number of C-atoms in the "bridge" connecting the P-atoms in an anilinyldiphosphine ligand determines the coordination mode with M 1 1 halides (M = Pd, Pt): dmapcp gives solely [MX(P,P',N-dmapcp)]X; dmape establishes in CH 2 C1 2 the equilibrium MX2(P,P'-dmape) [MX(P,P',7V"-dmapcp)]X (I) (for M = Pd and X = CI, A H 0 = -11 ± 7 kJ mol"1, AS° = -60 ± 20 J mol"1 K" 1 ; for M = Pt and X = CI, A H 0 = -1 9 ± 4 k J mol"1, AS° = -100 ± 3 0 J mol"1 K" 1); dmapm gives PdX2(P,P'-dmapm) PdX2(P,iV-dmapm) (II) (for X = CI, A H 0 = -5.5 ± 0.5 kJ mol"1, AS° = -10 ± 1 J K" 1 mol"'), and solely ?tX2(P,P'-dmapm). PtCl2(P,P'-dmape), [PtCl(P,P',A^-dmapcp)]Cl and PdCl2(P,A^-dmapm) are characterised crystallographically. The 4-membered ring strain associated with the last complex is ca. 32 kJ mol"1. [MCl(P,P',iV-dmapcp)]Cl complexes are formed stereoselectively, and exhibit slow exchange of coordinated and free anilinyl N-atoms. n I II (In I-VIII, Ar = o-C 6 H 4 NMe 2 ) The PdCl2(dmape) system reacts with K O H followed by K P F 6 in CH 2 C1 2 /H 2 0 to form [PdCl(dmapeO)]2[PF6]2 (III) which is structurally characterised and possesses a Pd-Pd separation of 4.873 A. In reactions of PdX2(diolefin) (diolefm = cod, nbd) with dmapcp and dmapm, long-lived, presumed trigonal bipyramidal intermediates are formed in which the coordinated olefin has undergone nucleophilic attack by either the P- (IV) or the N-atom (V) of the anilinyldiphosphine ligand. Ill PdCl2(P,7V-dmapm) reacts with peroxides and with sulphur to form PdCl2(P,Af-dmapmO) and [PdCl(/>,./V,S'-dmapmS)]Cl (VI), respectively, and with trans-PdCl 2(PhCN) 2, K 2 P t C l 4 and Y2 [Rh(u-Cl)(CO)2]2 to form the phosphine-bridged bimetallic complexes PdCl2M(X)(Y)(dmapm) (VII) (M = Pd, X = Y = CI; M = Pt, X = Y = CI; M = Rh, X = CI, Y = CO, respectively). The Pd 2 and PtPd bimetallic complexes catalyse effectively the Heck coupling of iodobenzene and styrene under air in D M F / H 2 0 solution to give predominantly trans-stilbene; the catalyst precursor eventually ii i decomposes by phosphine oxidation and via an equilibrium-controlled splitting into monometallic species. A generally applicable index of cooperativity for polymetallic catalysts is proposed. Conproportionation of PdCi2(P,./V-dmapm) and Pd2(dba)3 gives the metal-metal bonded complex Pd2Cl2(dmapm) (VIII, M = Pd) which is structurally characterised and possesses a very short (2.527 A) Pd—Pd bond. The mixed-metal analogue (VIII, M = Pt) is made by reducing PtPdCL^dmapm) with ethanolic K O H . These complexes react with CO and diethylacetylene dicarboxylate to give bridged species which are stable in the Pd2 case, but disproportionate in the PtPd case to Pd metal and PtC^^.P'-dmapm). VI VII VIII The pyridyldiphosphine ligands P-P react with M 1 1 halo precursors to give exclusively the P,P '-bonded species MX 2 (P -P) and [M(P-P) 2 ]X 2 , depending on the M to P-P ratio. Unlike the anilinyldiphosphines which do not react, 1:1 combinations of P-P and M(PPh 3) 4 give Pt(PPh 3) 2(P-P) and Pd(P-P) 2; these complexes react with the electron deficient olefins dimethyl- and diethylmaleate and the corresponding fumarates to form Pt(n2-olefin)(P-P) (which do not catalyse olefin isomerisation), and with CDCI3 to form [PtCl(PPh3)(P-P)]+. Pt(r|2-dimethylfumarate)(dpypcp) is structurally characterised. Several water-soluble complexes are assayed as catalyst precursors for the homogeneous, aqueous-phase hydration of maleic acid to form malic acid; the most active compound, [PdCl(P,A^5-dmapmS)]Cl, gives only 8 turnovers in 24 h at 100 °C. iv Table of Contents Abstract ii Table of Contents v List of Figures xiii List of Tables xvii List of Abbreviations and Symbols xxi Acknowledgements xxiv Dedication xxv CHAPTER ONE General Introduction 1.1 Evolutionary Trends in Homogeneous Catalysis 1 1.2 Product-Catalyst Separation 2 1.3 Water-soluble Phosphine Ligands 5 1.4 Aqueous Phase Homogeneous Catalysis 7 1.5 Other Considerations 9 1.6 Atom Economy 10 1.7 Scope of this Thesis 11 1.8 References 13 CHAPTER TWO Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.1 Introduction 17 2.1.1 Philosophical approach 17 2.1.2 A brief literature survey '. 19 2.1.2.1 Polymer-attached, optically-active diphosphine ligands 19 2.1.2.2 Water-solubilisation via incorporation of sulphonate groups 21 2.1.2.3 Water-solubilisation via quaternisation 22 2.1.2.4 Post-reaction separation by aqueous acid extraction 24 2.1.2.5 Miscellaneous ligands 25 2.2 Scope 25 2.3 The 2-Anilinyldiphosphine Ligands 26 2.3.1 Historical 26 v 2.3.2 Synthesis 26 2.3.3 Structures 28 2.4 Pt" and Pd" Complexes 31 2.4.1 Dmapcp complexes 31 2.4.1.1 Platinum 31 2.4.1.2 Palladium 37 2.4.2 Dmape complexes 39 2.4.2.1 Solution equilibrium 39 2.4.2.2 Structure 41 2.5 Organometallic Pd Anilinyldiphosphine Intermediates 43 2.5.1 An introduction to 5-coordinate Pt and Pd complexes 44 2.5.2 Nucleophilic attack on coordinated olefin 45 2.5.3 Variation of diolefin and halide 46 2.5.4 Reaction between diolefin and [PdCl(P,P',iV-dmapcp)]Cl 47 2.5.5 Low temperature, in situ reaction between PdCl2(cod) and dmapcp 47 2.5.6 ' H N M R spectroscopy 47 2.5.7 Comparison between the isolated product mixtures from low and r.t. reactions 48 2.5.8 Variation of the anilinyldiphosphine ligand 51 2.6 Reactions of PdCl2(dmape) with Bases 55 2.6.1 Reaction with K O H 55 2.6.1.1 Structure of [PdCl(dmapeO)]22+ 56 2.6.1.2 Formation of [PdCl(dmapeO)]22+ 57 2.6.2 Reaction of PdCl2(dmape) with K 2 C 0 3 , KO'Bu and Pd2(dba)3-CHC13 59 2.7 Conclusions 60 2.8 Recommendations for Future Work 61 2.8.1 Preparation of Pt° and Pd° complexes 61 2.8.2 Preparation and investigation of allyl complexes 62 2.8.3 Catalytic C - C and C - N bond formations 63 2.8.4 Ruthenium complexes: novel coordination modes, and reactivity 63 2.9 Experimental 65-2.9.1 Ligand syntheses 65 2.9.1.1 l,l-bis(di(o-A/,A/-dimethylanilinyl)phosphino)methane, dmapm 65 vi 2.9.1.2 1,2-bis(di(o-A^7V-dimethylanilinyl)phosphino)cyclopentane, dmapcp . 66 2.9.1.3 1,2-bis(di(o-N, Af-dimethylanilinyl)phosphino)ethane, dmape 66 2.9.2 Syntheses of Pt and Pd complexes 67 2.9.2.1 [PtCl(P,PJV-dmapcp)]Cl-H 20 67 2.9.2.2 [PtCl(P,P',AA-dmapcp)][PF6] 68 2.9.2.3 PtCl 2(P,P '-dmape) and [PtCl(P,P ', N-dmape)]Cl 68 2.9.2.4 [PtCl(P,P JV-dmape)][PF6] 69 2.9.2.5 Pt(ox)(P,P'-dmape) 69 2.9.2.6 [PdCl(P,P',7V-dmapcp)]Cl 69 2.9.2.7 [PdI(P,P ',/V-dmapcp)]I 70 2.9.2.8 [PdCl(P,P',N-dmapcp)][PF6] 70 2.9.2.9 PdCl 2(P,P '-dmape) and [PdCl(P,P ',7V-dmape)]Cl 71 2.9.2.10 [PdCl(P,P',/V-dmape)][PF6] 71 2.9.2.11 Pd(OAc)2(P,P'-dmape) 71 2.9.2.12 [PdCl(dmapeO)]2[PF6]2 72 2.9.2.13 Pd2Cl2(dmape) 72 2.9.2.14 [Pd(n3-allyl)(P,P'-dmape)][PF6].... 73 2.9.3 Syntheses and reactions of Ru complexes 73 2.9.3.1 RuCl 2 (P,P ',N,N '-dmape) 73 2.9.3.2 [Ru(P)P',^,7V',7V",7V'"-dmape)][PF6]2 74 2.9.3.3 Reaction of [Ru(P,P',N,N\N",N"'-dmape)][PF6]2 with H 2 S 74 2.9.3.4 Reaction of [Rxx(P,P',N,N',N",N"'-dmape)][PF6]2 with H 2 0 74 2.9.4 Miscellaneous reactions 75 2.9.4.1 Reactions between dmapcp and PdCl2(cod) 75 2.9.4.2 Halide-free reaction between PdCl2(dmape) and K O H 76 2.9.4.3 Reduction of PtCl2(dmape) by Na/Hg 76 2.9.4.4 In situ generation of Pt(n2-dmm)(P,P '-dmape) 76 2.9.4.5 Reaction of PdCl2(dmape) and bases 77 2.9.5 Catalytic C - N bond formation 77 2.9.5.1 Experimental 77 2.9.5.2 GC analysis. 77 2.9.6 Determination of equilibrium constants from 3 1 P{'H} V T N M R data 78 vii 2.9.6.1 The PtCl 2(P,P '-dmape) = ^ [PtCl(P,P',7V-dmape)]+ + CI" equilibrium 78 2.9.6.2 The PdCl 2(P,P '-dmape) [PdCl(P,P\N-dmape]+ + CI" equilibrium 80 2.10 References 81 CHAPTER THREE Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.1 Introduction 85 3.2 Scope 87 3.3 Complexes Containing One Metal Centre 87 3.3.1 Solution dynamics 87 3.3.2 Four-membered ring strain 93 3.3.3 Structure and bonding 94 3.3.4 Reactions with halide and cyanide 100 3.3.5 Reactions with peroxide and Ss 100 3.3.6 PdCl(Me)(P,7V-dmapm) 101 3.4 Complexes Containing Two Metal Centres 102 3.4.1 Reactions with fra«s-PdCl 2(PhCN) 2 and K 2 P t C l 4 102 3.4.2 Reaction with N H 4 P F 6 105 3.4.3 Reaction with [Rh(u-Cl)(CO)2]2 105 3.4.4 Rh 2Cl 2(CO) 2(dmapm) 106 3.4.5 Complexes containing M - M bonds 108 3.4.5.1 Reaction of PdCl2(dmapm) with Pd2(dba)3 108 3.4.5.2 Two-electron reductions of bimetallic M(II) complexes 113 3.4.5.3 Reactions of bimetallic M 1 complexes 115 3.5 A General Introduction to Cluster Catalysis 122 3.5.1 Homogeneous vs. heterogeneous 122 3.5.2 Multi-site activation.. 123 3.5.3 Product distribution in single-site cluster catalysis 123 3.5.4 Cluster fragmentation 123 3.5.5 Indications 124 3.6 The Heck Reaction 124 3.6.1 General 124 viii 3.6.2 Heck coupling of iodobenzene and styrene using anilinylphosphine complexes 125 3.6.2.1 Compounds tested 125 3.6.2.2 Results 126 3.6.2.3 Oxidative degradation 131 3.6.2.4 Attempted isolation and identification of Heck catalytic intermediates 133 3.6.2.5 Summary of the catalysis 135 3.6.3 A cooperativity index proposal 136 3.7 Conclusions 138 3.8 Recommendations for Future Work 139 3.8.1 More accurate determination of ring strain energy 139 3.8.2 Assessment of cooperative effects in other types of homogeneous catalysis 140 3.8.2.1 Hydroformylation 140 3.8.2.2 Imine/CO copolymerisation 140 3.8.3 Exploration of the reactivity of MM'X 2(dmapm) complexes 141 3.9 Experimental 142 3.9.1 PdCl 2(P,P '-dmapm) and PdCl2(P,JV-dmapm) 142 3.9.2 PdBr2(P,P'-dmapm) and PdBr2(P,7V-dmapm) 143 3.9.3 PdI2(P,P-dmapm) and PdI2(P,7V-dmapm) 144 3.9.4 Pd(CN)2(P,iV-dmapm) 144 3.9.5 PdCl(Me)(P,Ar-dmapm) 145 3.9.6 PdCl2(P,7V"-dmapmO) 145 3.9.7 [PdCl(P,A^-dmapmS)][PF6] 146 3.9.8 [PdCl(P,A^5-dmapmS)]Cl 146 3.9.9 PtCl 2(P,P '-dmapm) 146 3.9.10 Pt(CN)2(P,P'-dmapm) and Pt(CN)2(P,iV-dmapm) 147 3.9.11 Pd2Cl4(dmapm) 147 3.9.12 Pd2l4(dmapm) 147 3.9.13 PtPdCl4(dmapm) 148 3.9.14 [Pd(u-Cl)(P,P'-dmapm)]2[PF6]2 148 3.9.15 PdRhCl3(CO)(dmapm) 149 IX 3.9.16 Rh2Cl2(CO)2(dmapm) 149 3.9.17 Pd2Cl2(dmapm) 149 3.9.18 PtPdCl2(dmapm) 150 3.9.19 Pd2Cl2(CO)2(dmapm) 151 3.9.20 Pd 2Cl 2(DEAD)(dmapm)-H 20 151 3.9.21 Attempted isolation of a Heck catalytic intermediate 151 3.9.22 Attempted preparation of PtI2(P,A^-dmapmO) 152 3.9.23 Heck reactions 152 3.10 References 154 CHAPTER FOUR Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.1 Introduction 160 4.2 Scope 161 4.3 The Pyridyldiphosphine Ligands and Derivatives 162 4.3.1 History and synthesis 162 4.3.2 Pyridyldiphosphine dioxides 164 4.3.3 Protonated pyridyldiphosphines 165 4.4 Synthesis and Characterisation of Pyridyldiphosphine Complexes 168 4.4.1 A brief literature survey 168 4.4.2 Platinum(II) and palladium(II) pyridyldiphosphine complexes 169 4.4.2.1 Monometallic platinum(II) complexes 170 4.4.2.2 "Tethered paddlewheel" complexes 175 4.4.2.3 Palladium(II) pyridyldiphosphine complexes 185 4.4.3 Pt° and Pd° pyridyldiphosphine complexes 186 4.4.3.1 Reaction of Pt(PPh3)2(P-P) with maleic and fumaric acid diesters 188 4.5 Conclusions 199 4.6 Recommendations for Future Work 200 4.6.1 Synthesis of dpypm, Pd 2 ' complexes, and the WGS reaction 200 4.7 Experimental 201 4.7.1 Racemic-dpypcp 201 4.7.2 Racemic-dpypcp(0)2 203 4.7.3 dpype(0)2 203 4.7.4 [dpypcp(H)2][PF6]2 204 4.7.5 [dpype(H)2][PF6]2 204 4.7.6 Racemic-dpypcp(S)2 205 4.7.7 PtCl2(dpypcp) 204 4.7.8 PtBr2(dpypcp) 205 4.7.9 Ptl2(dpypcp) 205 4.7.10 PtCl2(dpype) 206 4.7.11 Ptl2(dpype) 206 4.7.12 [Pt(P-P) 2]X 2 (X = CI, I) 206 4.7.13 [Pt(dpypcp)2][PF6]2-H20 206 4.7.14 [Pt(dpype)2][PF6]2 207 4.7.15 PdCl2(dpypcp) 207 4.7.16 PdBr 2(dpypcp)H 20 208 4.7.17 Pdl2(dpypcp) 208 4.7.18 [Pd(dpypcp)2][PF6]2-H20 208 4.7.19 Preparation of [Pt 2(dpype) 2Ag 4(N03)8(H 20) 2] n, 1 209 4.7.20 Preparation of [Pt2(dpypcp)2Ag6(N03)io]n, 2 209 4.7.21 [Pt2(dpype)2][N03]4-2H20 209 4.7.22 Pt(PPh3)2(dpypcp) 210 4.7.23 Pt(PPh3)2(dpype) 210 4.7.24 Pt(dpypcp)2 211 4.7.25 Pt(dpype)2 211 4.7.26 Pt(n2-dmf)(dpypcp) 211 4.7.27 Pt(n2-def)(dpypcp) 212 4.7.28 Determination of pK a values for [dpypcp(H)2]2+ 212 4.7.28.1 Materials 213 4.7.28.2 Instrumentation 213 4.7.28.3 Procedure 213 4.8 References 214 CHAPTER FIVE Attempted Catalytic Hydration of Maleic Acid 5.1 Introduction 220 5.2 Scope 221 xi 5.3 Results 222 5.4 Discussion 223 5.5 Conclusions 223 5.6 Recommendations for Future Work 224 5.7 Experimental 224 5.7.1 Preparation of catalyst I, and attempted catalytic hydration protocol 224 5.7.2 Preparation of catalyst II 224 5.7.3 Reaction of PtCl2(dmape) with K O H followed by maleic acid 225 5.7.4 General protocol for the catalytic hydration of maleic acid 225 5.7.5 Determination of extent of hydration from 1 H N M R spectra 226 5.8 References 227 APPEND FX A l General Experimental Protocols A l . 1 General Procedures 229 A l .2 Instrumentation 229 Al.2.1 Nuclear magnetic resonance (NMR) spectroscopy 229 A l .2.2 X-ray crystallography 230 Al.2.3 Elemental analysis 230 Al.2.4 Ultraviolet-visible (UV-vis) spectroscopy 230 Al.2.5 Infra-red (IR) spectroscopy 230 Al.2.6 Conductivity 230 A l .2.7 Gas chromatography (GC) 231 A1.3 Materials 231 Al.3.1 Gases 231 Al.3.2 Solvents 231 Al.3.3 Reagents 231 Al.3.3.1 Metal complexes 231 Al.3.3.2 Organic compounds 232 A1.4 References 233 APPENDICES A2-A13 Crystal Structure Data A2 Crystal Structure Data for Dmape 234 xii A3 Crystal Structure Data for Dmape 239 A4 Crystal Structure Data for [PtCl(P,P ',7V-dmapcp)]Cl-1.46 H 2 0 C H 2 C 1 2 242 A5 Crystal Structure Data for PtCl 2(P,P '-dmape)-CH2Cl2 247 A6 Crystal Structure Data for [PdCl(dmapeO)]2[PF6]2-4 CDC1 3 253 A7 Crystal Structure Data for PdCl2(P,N-dmapm) 256 A8 Crystal Structure Data for Pd2Cl2(dmapm)-2 CHC1 3 262 A9 Crystal Structure Data for PtCl 2(dpype)-CH 2Cl 2 268 A l 0 Crystal Structure Data for [Pt 2 (dpype) 2 Ag 4 (N0 3 ) 8 (H 2 0) 2 ]„ 273 A l 1 Crystal Structure Data for PtI2(dpypcp)-0.18 CH 2 C1 2 278 A12 Crystal Structure Data for [Pt 2(dpypcp) 2Ag 6(N0 3)io]„ 285 A l 3 Crystal Structure Data for Pt(r)2-dmf)(dpypcp)-2 CDC1 3 289 Colophon 294 List of Figures Figure 1.1 Schematic representing the application of the biphasic catalysis principle to industrial reactor design (adapted from ref. 28). The system illustrated here is very similar to that used in the R/RP oxo process 4 Figure 2.1 ORTEP representation of the centrosymmetric crystallographic form of dmape (50 % ellipsoids). H-atoms are omitted for clarity 29 Figure 2.2 ORTEP representation of dmapcp (50 % ellipsoids). H-atoms are omitted for clarity 30 Figure 2.3 The 5 1-4 region of the ] H{ 3 1 P} N M R (121 MHz, CDC1 3 , 300 K) spectrum of [PtCl(P,P'(7V-dmapcp)]Cl. Asterisks (*) and number symbols (#) indicate peaks due to the C / / 2 and CH protons, respectively, in the ligand "backbone". Other assignments are discussed in the text. The singlet at 1.56 ppm is due to H 2 0 32 Figure 2.4 ORTEP representation (50 % ellipsoids) of [PtC\(P,P ',7V-dmapcp)]+. H -atoms have been omitted for clarity 34 Figure 2.5 The 5 2-4 range of the 2D ' H E X S Y (300 MHz, CDC1 3 , 300 K) spectrum of [PtCl(P,P',Ar-dmapcp)]Cl 35 Figure 2.6 PLUTO representation of the molecular structure of [PdCl(P,P\N-dmapcp)]+ H-atoms are omitted for clarity 37 X l l l Figure 2.7 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) spectra of [MC1(P,P',7V-dmapcp)]Cl (M = Pt, Pd). Peaks due to 3 1 P- 1 9 5 Pt coupling fall outside the range given for the Pt spectrum 38 Figure 2.8 The 3l?{lH} N M R (121 MHz, CD 2 C1 2 , 240 K) spectrum of PtCl2(dmape). The * symbol designates peaks due to PtCl 2(P,P '-dmape) and # denotes those due to [PtCl(P,P,N'-dmape)]Cl 39 Figure 2.9 Van't Hoff plot for the MC1 2(P,P '-dmape) [MC1(P,P ',7V-dmape)]Cl equilibrium 41 Figure 2.10 ORTEP representation (50 % ellipsoids) of the molecular structure of PtCl2(dmape). Except for those in the "backbone" of the ligand, H-atoms have been omitted for clarity 42 Figure 2.11 The 3 1 P{ ] H} N M R (121 MHz, CDCI3, 300 K) spectrum of the isolated product mixture from the reaction between dmapcp and PdCl2(cod). The * symbol denotes peaks due to [PdCl(P,P',7V-dmapcp)]Cl 43 Figure 2.12 The 6 4.8-6.0 region of the ! H N M R (121 MHz, CDC1 3 , 300 K) spectrum of the isolated product mixture from the r.t. reaction between PdCl2(cod) and dmapcp.. 48 Figure 2.13 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) of the isolated product mixtures from the reaction between PdCl2(cod) and dmapcp at (i) r.t. and (ii) —40 °C recorded 0.5 h after making the samples. Spectrum (iii) shows the evolution of (ii) after 6 h at r.t. Peaks marked by asterisks are due to [PdCl(P,P',./V-dmapcp)][PF6]. Other assignments are discussed in the text 49 Figure 2.14 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) spectrum of the in situ reaction between dmapm and PdCl2(cod). The * symbol denotes peaks due to PdCl2(P,A^-dmapm) and # identifies that due to PdCl2(P,P'-dmapm). Other labels are discussed in the text.. 52 Figure 2.15 l H{ 3 1 P} (top) and *H (bottom) N M R (300 MHz, CDC1 3 , 300 K) spectra of the product mixture from the reaction between PdCl2(cod) and dmapm. Peak assignments are discussed in the text 54 Figure 2.16 ORTEP representation (50 % ellipsoids) of the molecular structure of the [PdCl(dmapeO)]22+ cation. H-atoms are omitted for clarity 56 Figure 2.17 A typical low temperature 3 1 P{'H} N M R (121 MHz, CD 2 C1 2 , 240 K) spectrum of an equilibrium mixture of PtCl2(P,P-dmape) and [PtCl(P,P',A /1dmape)]Cl. Peak assignments are discussed in the text 79 Figure 2.18 The 3 , P{ 'H} N M R (121 MHz, CD 2 C1 2 , 233 K) spectrum for the PdCl2(dmape) system. Peak assignments are discussed in the text 80 Figure 3.1 Temperature-dependence of the N C H 3 region of the ' H N M R (400 MHz, CDCI3) spectrum of an equilibrium mixture of PdCl2(P,P'-dmapm) and PdCl2(P,iV-dmapm) 90 Figure 3.2 The 2D [ H E X S Y spectrum (300 MHz, CDC1 3 , 300 K) of the N C H 3 region of an equilibrium mixture of PdCl2(P,P'-dmapm) and PdCl2(P,7V-dmapm) 92 xiv Figure 3.3 Van't Hoff plot for the PdCl2(P,P '-dmapm) PdCl2(P,N-dmapm) equilibrium ; 93 Figure 3.4 ORTEP representation of the molecular structure of PdCl2(P,iV-dmapm) (50 % ellipsoids) 94 Figure 3.5 Walsh correlation diagram (adapted from Dunne et al.46) for the deformation of a P R 3 species, showing rehybridisation of the phosphine lone pair 2ai, and 71-acceptor function 2e, on going from trigonal planar (D3h) geometry (left) to pyramidal (C 3 v , right) 97 Figure 3.6 The N C H 3 region of the *H N M R (400 MHz, CDC1 3 , 300 K) spectrum of PtPdCl4(dmapm) 104 Figure 3.7 The ^ P ^ H } N M R (162 MHz, CDC1 3, 300 K) spectrum of Rh2Cl2(CO)2(dmapm) 107 Figure 3.8 ORTEP representation of the molecular structure of Pd2Cl2(dmapm) (50 % ellipsoids). H-atoms have been omitted for clarity 110 Figure 3.9 The coordination environments of the Pd centres in Pd2Cl2(dmapm) viewed along (left) and perpendicularly to (right) the Pd—Pd axis I l l Figure 3.10 Space-filling representation of Pd2Cl2(dmapm) showing the lopsided distribution of steric bulk about the Pd—Pd bond. A l l visible, unlabelled atoms are C-atoms 112 Figure 3.11 The 3 I P{ 'H} N M R (121 MHz, CDCI3, 300 K) spectrum and proposed structure of PfPdCl2(dmapm). The small peak at 5 -30 is due to trace Pd2Cl2(dmapm) impurity. (Ar = o-C 6 H 4 NMe 2 . ) 115 Figure 3.12 The 3 1 P{ 1 H} N M R (121 MHz, CDC13) spectra of Pd 2Cl 2(DEAD)(dmapm) at 325 (top) and 220 K (bottom) 119 Figure 3.13 The 3 1 P{'H} N M R (162 MHz, CDC1 3, 253 K) spectrum of PtPdC^dmapm) under CO, 90 min after admitting the gas. * denotes PfPdCl2(dmapm) and # signifies a small Pd2Cl2(dmapm) impurity. Other assignments are discussed in the text 120 Figure 3.14 The variation of [Phi] with time for the Heck coupling of Phi and styrene using Pd2Cl4(dmapm) (4.4 x 10"4 mol L"1) as catalyst precursor (100 °C, D M F / H 2 0 solvent (3:2 by vol.), K 2 C 0 3 base). The inset shows a plot of ln([PhI]/[PhI]0) vs. time where [Phl] 0 is the initial Phi concentration 126 Figure 3.15 Variation of initial rate of the Heck coupling of Phi and styrene with total Pd concentration for reactions catalysed by PdMCl4(dmapm) (M = Pd, Pt) and PdCl 2 (PMA) under the conditions outlined in Section 3.9.23 128 Figure 3.16 Plot of ln(lnitial rate) vs. ln(Total [Pd]) for the Heck coupling of Phi and styrene catalysed by the bimetallic complexes PdMCL^dmapm) ( M = Pd, Pt) 129 Figure 3.17 PLUTO representation of the molecular structure of PtPdL^dmapm). H -atoms are omitted for clarity 132 xv Figure 3.18 PLUTO representation of the molecular structure of PtI2(P,./V-dmapmO). H-atoms are omitted for clarity 132 Figure 3.19 The 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) spectrum of the orange powder isolated from the Heck reaction catalysed by PtPdCLXdmapm) 134 Figure 4.1 TOP: pH dependence on vol. of NaOH (1.094 x 10"1 mol L"1) added for an HC1 solution (1.135 x 10"2 mol L"1) containing dissolved dpypcp (6.112 x 10"4 mol L"1) (solid line), and the HC1 solution alone (dashed line). BOTTOM: number of protons bound per dpypcp molecule (n ), as a function of pH 167 Figure 4.2 ORTEP representation of PtCl2(dpype)-CH2Cl2. Thermal ellipsoids for non-hydrogen atoms are drawn at 33 % probability 173 Figure 4.3 ORTEP representation of one of the 2 crystallographically independent molecules of PtL;(dpypcp) in the unit cell of Ptl2(dpypcp)-0.18 CH2CI2 (50 % probability ellipsoids). Except for the two bonded to the methine C-atoms in the ligand "backbone", H-atoms have been omitted for clarity 174 Figure 4.4 ORTEP representation (33 % probability) of a section of 1 showing the Ag(N0 3 ) connections between Pt"2(dpype)2 moieties. The longer Ag-0 bonds, H-atoms, and the pyridyl C-atoms not involved in bridges to metal centres have been omitted for clarity 178 Figure 4.5 ORTEP representation (33 % probability) of the "Ag(N0 3 ) bridge" of 1. "Short" Ag—O bonds are indicated by heavy lines, "medium" length bonds by double lines and "long" bonds by single lines 179 Figure 4.6 ORTEP representation (33 % probability) of the "tethered paddlewheel" Ptn2(dpype)2 moiety of 1. Also shown are two Ag-atoms, each bound to one pyridyl N -atom of each dpype ligand. H-atoms have been omitted for clarity 180 Figure 4.7 ORTEP representation (33 % probability) of the repeating unit of 2. H -atoms have been omitted for clarity 182 Figure 4.8 The 3 1 P{'H} N M R (162 MHz, C 6 D 6 , 300 K) spectrum of Pt(PPh3)2(dpypcp) 187 Figure 4.9 The 5 15-29 range of the 3 1 P{'H} N M R (121 MHz, CDCI3, 300 K) spectrum of Pt(PPh3)2(dpypcp). The peak due to PtCl2(dpypcp) is marked by an asterisk (*), and those of [Pt(PPh3)(dpypcp)]+ by the number symbol (#). Pt satellites for all peaks fall outside of this window 191 Figure 4.10 3 1 P{'H} N M R spectrum (CDC13, 202 MHz, 300 K) of Pt(n2-dmf)(dpypcp) 194 Figure 4.11 Quadrant diagram showing steric interactions between a chiral C 2 -symmetric metal-diphosphine fragment and a bound olefin, looking down the M—mp axis (mp = C=C mid-point). The grey rectangles represent steric bulk presented by the diphosphine ligand, and the black circles represent the substituents on the olefin. The left-hand diagram illustrates the sterically more favourable diastereomer 195 xvi Figure 4.12 ORTEP representation of Pt(n2-dmf)(dpypcp)-2 CDC1 3 . Except for the 2 olefinic ones, H-atoms have been omitted for clarity (50 % ellipsoids) 196 Figure 5.1 A sample ' H N M R (300 MHz, 300 K) spectrum of an acetone-d6 solution containing maleic, fumaric and malic acids, fa = fumaric acid, ma = maleic acid. H a and Hb,c refer to the protons given in the Fischer projection of malic acid 226 Figure A2.1 ORTEP representation of dmape (50 % ellipsoids) 234 Figure A3.1 ORTEP representation of dmapcp (50 % ellipsoids) 239 Figure A4.1 ORTEP representation of [PtCl(P,P',7V-dmapcp)]+ (50 % ellipsoids).242 Figure A5.1 ORTEP representation of PtCl 2(P,P '-dmape) (50 % ellipsoids) 247 Figure A6.1 ORTEP representation of [PdCl(dmapeO)]22+ (50 % ellipsoids) 253 Figure A7.1 ORTEP representation of PdCl2CP,./V-dmapm) (50 % ellipsoids) 256 Figure A8.1 ORTEP representation of Pd2Cl2(dmapm) (50 % ellipsoids) 262 Figure A9.1 ORTEP representation of PtCl2(dpype) CH 2C1 2 (50 % ellipsoids) 268 Figure A10.1 ORTEP representation of the repeating unit of [Pt2(dpype)2Ag4(N03)8(H20)2]„ (33 % ellipsoids) 273 Figure A l l . l ORTEP representation of Ptl2(dpypcp) (50 % ellipsoids) 278 Figure A12.1 ORTEP representation of the repeating unit of [Pt2(dpypcp)2Ag6(N03)io]n (33 % ellipsoids) 285 Figure A13.1 ORTEP representation of Pt(n2-dmf)(dpypcp)-2 CDC1 3 (50 % ellipsoids) 289 List of Tables Table 1.1 A selection of reactions catalysed by metal-phosphine complexes dissolved in water 7 Table 2.1 Selected bond distances (A) and angles (°) for dmape with estimated standard deviations given in parentheses 29 Table 2.2 Selected bond distances (A) and angles (°) for dmapcp with estimated standard deviations given in parentheses 30 Table 2.3 Selected bond distances (A) and angles (°) for [PtCl(P,P ',7V-dmapcp)]Cl-CH2Cl2-1.46 H 2 Owith estimated standard deviations given in parentheses 34 Table 2.4 Selected bond distances (A) and angles (°) for PtCl 2(dmape)-CH 2Cl 2 with estimated standard deviations given in parentheses 42 xvn Table 2.5 Summary of 3 1 P{'H} (121 MHz, 300 K) data for the series of reactions shown in Scheme 2.7 46 Table 2.6 Summary of the 3 1 P{'H} N M R data for the spectra shown in Figure 2.13, excluding peaks due to [PdCl(P,P;N-dmapcp)][PF6] 49 Table 2.7 Summary of 3 1 P{'H} (121 MHz, 300 K) data for a series of reactions of the type shown in Scheme 2.8 52 Table 2.8 Selected bond distances (A) and angles (°) for [PdCl(dmapeO)]2[PF6]2-4 CDCI3 with estimated standard deviations given in parentheses 56 Table 2.9 Colour changes observed for the reaction between PdCl2(dmape) and K O H in H2O/CH2CI2 (I) and in H 2 0 / C H 3 N 0 2 (II) 59 Table 2.10 3 1 P N M R data for a variety of Ru(dmape) complexes and for the in situ reactions of Ru(P,P ',N,N',N'\N'' '-dmape)] [PF 6] 2 with H 2 0 and H 2 S 64 Table 2.11 G C parameters used in the determination of C - N bond forming reaction components 78 Table 2.12 Elution times for Heck reaction components under the GC conditions given in Table 2.11 78 Table 3.1 Selected bond lengths (A) and angles (°) for PdCl^TV-dmapm) with estimated standard deviations in parentheses 95 Table 3.2 Average P—C and N — C bond distances (A) for the crystallographically characterised anilinyldiphosphine ligands and their complexes 99 Table 3.3 Selected bond distances (A) and angles (°) for Pd2Cl2(dmapm)-2 CDC1 3 with estimated standard deviations in parentheses 100 Table 3.4 Initial rates and TOFs for the Heck coupling of Phi (0.4 mol L"1) and styrene (0.4 mol L" 1) at 100 °C in D M F / H 2 0 (3:2 by vol.) with K 2 C 0 3 (0.4 mol L" 1) as base 127 Table 3.5 The 3 1 P {'H} N M R (121 MHz, C 6 D 6 , 300 K) spectral data for compounds 1,2 and 3 134 Table 3.6 Calculation of cooperativity indices for PdMCL^dmapm) catalysts 138 Table 3.7 3 1 P{'H} N M R (CDC1 3, 300 K) data for the in situ reactions between small molecules and MM'Cl 2(dmapm) 141 Table 3.8 GC parameters used in the determination of Heck reaction components 153 Table 3.9 Elution times for Heck reaction components under the GC conditions given in Table 3.8 154 Table 4.1 Spectroscopic data for pyridylphosphine ligands and their oxides (tw = this work) 165 Table 4.2 The known coordination modes of the Type IV pyridyldiphosphine ligands and their respective metal centres (tw = this work) 169 xvin Table 4.3 3 1 P{ ] H} N M R data (CDC1 3, 300 K) for Pt pyridyldiphosphine complexes 171 Table 4.4 Selected bond lengths (A) and angles (°) for PtCl 2(dpype)-CH 2Cl 2 with estimated standard deviations in parentheses 174 Table 4.5 Selected bond lengths (A) and angles (°) for PtI2(dpypcp)-0.18 CH 2 C1 2 with estimated standard deviations in parentheses 175 Table 4.6 Selected bond lengths (A) and angles (°) for 1 with estimated standard deviations in parentheses 181 Table 4.7 Selected bond lengths (A) and angles (°) for 2 with estimated standard deviations in parentheses 182 Table 4.8 Pt-Pt separations and calculated bond orders for some known paddlewheel complexes (tw = this work) 184 Table 4.9 3 1 P{'H} N M R data for Pt(P-P)(n2-olefin) complexes 189 Table 4.10 Selected bond lengths (A) and angles (°) for Pt(r|2-dmf)(dpypcp)-2 CDC1 3 with estimated standard deviations in parentheses 196 Table 4.11 Relevant parameters for the intra- and intermolecular H-bonding interactions in Pt(n2-dmf)(dpypcp)-2 CDC1 3 199 Table 5.1 Product distributions for the attempted catalytic hydration of maleic acid (0.1 mol L"1) in water at 100 °C using pyridyl- and anilinyldiphosphine complexes of platinum metals as catalyst precursors (substratexatalyst = 100:1) according to Section 5.7.4. Percent conversions, which are equivalent to TONs, are reported for 24 h reaction times. Error is ± 1 % 222 Table A l . l Spectrometer frequencies for 1 H , 3 1 P and 1 3 C for each of the N M R instruments used in the course of this work 229 Table A1.2 Procedures used to synthesise metal complex precursors 232 Table A2.1 Experimental details 235 Table A2.2 Atomic coordinates and B,- i 0 /B e ? 236 Table A2.3 Bond lengths (A) 237 Table A2.4 Bond angles (°) 238 Table A3.1 Crystal data and structure refinement 240 Table A3.2 Atomic coordinates (A x 104) and equivalent isotropic displacement parameters (A2 x 103). U(eq) is defined as one third of the trace of the orthogonalized Uy tensor 240 Table A3.3 Bond lengths (A) and angles (deg) 241 Table A4.1 Experimental details 243 Table A4.2 Atomic coordinates and Beq 244 Table A4.3 Bond lengths (A) 245 xix Table A4.4 Bond angles (°) 246 Table A5.1 Experimental details 248 Table A5.2 Atomic coordinates and Biso/Beq 249 Table A5.3 Bond lengths (A) 251 Table A5.4 Bond angles (°) 251 Table A6.1 Crystal data, data collection, and solution and refinement 254 Table A6.2 Atomic coordinates (A x 104) and equivalent isotropic displacement parameters (A2 x 103). U(eq) is defined as one third of the trace of the orthogonalized Uy tensor 254 Table A6.3 Bond lengths (A) and angles (deg) 255 Table A7.1 Experimental details 257 Table A7.2 Atomic coordinates and BiS0IBiq 258 Table A7.3 Bond lengths (A) 260 Table A7.4 Bond angles (°) 261 Table A8.1 Experimental details 263 Table A8.2 Final coordinates and equivalent isotropic displacement parameters (A2) of the non-hydrogen Atoms 264 Table A8.3 Bond distances (A) 265 Table A8.4 Bond angles (°) 266 Table A9.1 Full crystallographic data 269 Table A9.2 Atomic coordinates and Beq 270 Table A9.3 Bond lengths (A) 271 Table A9.4 Bond angles (°) 271 Table A10.1 Experimental Details 274 Table A10.2 Atomic coordinates and Beq 275 Table A l 0.3 Bond lengths (A) 276 Table A10.4 Bond angles (°) 277 Table All.1 Experimental details 279 Table A11.2 Atomic coordinates and Beq 280 Table A11.3 Bond lengths (A) 282 Table A11.4 Bond angles (°) .' 283 Table A12.1 Crystal data and structure refinement 286 xx Table A12.2 Atomic coordinates (A x 104) and equivalent isotropic displacement parameters (A2 x 103). U(eq) is defined as one third of the trace of the orthogonalized Uy tensor 286 Table A12.3 Bond lengths (A) and angles (deg) 287 Table A13.1 Experimental details 290 Table A13.2 Atomic coordinates and Beq 291 Table A13.3 Bond lengths (A) 292 Table A13.4 Bond angles (°) 293 List of Abbreviations and Symbols Anal. Analysis aq. Aqueous Ar Aromatic, aryl av. Average B D D P D S l,3-Bis(diphenylphosphino)propane disulphonate BINAP 2,2 '-bis(diphenylphosphino)-1,1 '-dinaphthyl BINAPO 2,2'-bis(diphenylphosphino)-1,1 '-dinaphthyl monooxide bipy 2,2'-Bipyridine B P M O Bis(phosphine)monooxide br Broad "Bu Normal-butyl Calcd Calculated cod 1,5-cyclooctadiene COSY Correlation N M R spectroscopy Cp Cyclopentadienyl Cp* Pentamethylcyclopentadienyl Cy Cyclohexyl d Doublet dba Dibenzylidene acetone (PhC'H=CHC(0)CH=CHPh) dcp Dicyclopentadiene dcype 1,2-bis(dicyclohexylphosphino)ethane (Cy 2P(CH 2)2PCy 2) dd Doublet of doublets D E A D Diethylacetylene dicarboxylate (EtOC(0)C=CC(0)OEt) def Diethylfumarate (£-EtOC(0)CH=CHC(0)OEt) dem Diethylmaleate (Z-MeOC(0)CH=CHC(0)OMe) demOH Diethylmalate (EtOC(0)CH 2CH(OH)C(0)OEt) depm l,l-bis(diethylphosphino)methane (Et 2PCH 2PEt 2) dil. Dilute DIOP 4,5-bis((diphenylphosphino)methyl)-2,2-dimethyl-l,3-dioxolane D M A D Dimethylacetylene dicarboxylate (MeOC(0)C=CC(0)OMe) dmapcp 1,2-bis(di(o-A', A^-dimethylanilinyl)phosphino)cyclopentane dmape 1,2-bis(di(o-7V,7V-dimethylanilinyl)phosphino)ethane dmapm 1,1 -bis(di(o-/V, A/-dimethylanilinyl)phosphino)methane dmapmO 1,1 -bis(di(o-./V,A'-dimethylanilinyl)phosphino)methane monooxide dmf Dimethylfumarate (g-MeOC(Q)CH=CHC(0)OMe) xxi D M F Dimethylformamide dmm Dimethylmaleate (Z-MeOC(0)CH=CHC(0)OMe) dmpb 1,2-bis(dimethylphosphino)benzene (Me 2P(C 6H 4)PMe2) dmpe 1,2-bis(dimethylphosphino)ethane (Me 2P(CH 2)2PMe 2) dmphen 2,9-dimethyl-1,10-phenanthroline dmpm l,l-bis(dimethylphosphino)methane (Me 2 PCH 2 PMe 2 ) DMSO Dimethyl sulphoxide (Me 2S=0) dppb l,4-bis(diphenylphosphino)butane (Ph 2P(CH 2) 4PPh 2) dppcp 1,2-bis(diphenylphosphino)cyclopentane (Ph 2PCH(CH 2)3CHPPh 2) dppe l,2-bis(diphenylphosphino)ethane (Ph 2P(CH 2) 2PPh 2) dppm l,l-bis(diphenylphosphino)methane (Ph 2PCH 2PPh 2) dppp l,3-bis(diphenylphosphino)propane (Ph 2P(CH 2) 3PPh 2) dpypcp 1,2-bis(di-o-pyridylphosphino)cyclopentane (py 2 PCH(CH 2 ) 3 CHPpy 2 ) dpypcp(0)2 l,2-bis(di-o-pyridylphosphino)cyclopentane dioxide (py 2P(0)CH(CH 2) 3CHP(0)py 2) dpype 1,2-bis(di-o-pyridylphosphino)ethane (py 2P(CH 2) 2Ppy 2) dpype(0)2 1,2-bis(di-o-pyridylphosphino)ethane dioxide (py 2P(0)(CH 2) 2P(0)py 2) dpypm l,l-bis(di-o-pyridylphosphino)methane (py 2PCH 2Ppy 2) e.e. Enantiomeric excess eq. Equation equiv. Equivalent(s) Et Ethyl E X S Y Exchange N M R spectroscopy fa Fumaric acid (£-H0 2 CCH=CHC0 2 H) FID Flame ionisation detector GC Gas chromatography H E M A 2-hydroxethyl methacrylate HETCOR Heteronuclear correlation N M R spectroscopy H.P. High purity HT Head-to-tail hxd 1,5-hexadiene i.d. Internal diameter J Coupling constant (Hz) IR Infra-red m Multiplet ma Maleic acid (Z-H0 2 CCH=CHC0 2 H) Me Methyl mol Mole Hmtpo 4,7-//-5-methyl-7-oxo[ 1,2,4]triazolo[ 1,5a]pyrimidine nbd Norbornadiene N M R Nuclear magnetic resonance NOE Nuclear Overhauser effect OAc Acetate, acetato O P N 2 Di(o-pyridyl)phenylphoshine oxide (Ph(py)2P=0) O'Bu Tertiary-butoxide OTf Trifluoromethanesulphonate, triflate (CF 3S0 3") ox Oxalate, oxalato (C 20 4 2") p Pseudo- (e.g., pd = pseudo-doublet) p. Page Ph Phenyl phen 1,10-phenanthroline P M A Diphenyl-o-AfAf-dimethylanilinylphosphine, Ph 2 P-o-C 6 H 4 NMe 2 PN) Diphenyl-o-pyridylphosphine (Ph2Ppy) P N 2 Di-o-pyridylphenylphosphine (PhPpy2) P N 3 Tri(o-pyridyl)phosphine ppm Parts per million py Pyridyl xxii q Quartet quin 2-quinaldinate, 2-quinaldinato Ref. Reference ROMP Ring-opening metathesis polymerisation R/RP Ruhrchemie/Rhone-Poulenc R V B M Resonating valence-bond theory of metals s Singlet sh Shoulder SHOP Shell higher olefin polymerisation spt Septet t Triplet THF Tetrahydrofuran TOF Turnover frequency (mol product per mol catalyst per unit time) TON Turnover number (mol product per mol catalyst) TPA l,3,5-triaza-7-phosphaadamantane TPPMS Triphenylphosphine monosulphonate TPPTS Triphenylphosphine trisulphonate tw This work UV-vis Ultraviolet-visible vol. Volume WGS Water-gas shift 8 Chemical shift (ppm) £ Extinction coefficient (L mol"' cm"1) n Hapticity k Ligating atom X Wavelength (nm) A M Molar conductivity (ohm"' mol"' cm2) u Bridging coordination mode v Wavenumber (cm"1) [ ] Concentration { } Broad-band decoupled XX111 Acknowledgements I thank Dr. Brian James whose intellectual and professional achievements have been an inspiration, whose vast knowledge of inorganic chemistry and catalysis has been an invaluable resource and whose editorial skill has prevented this thesis from reaching ridiculous proportions. I acknowledge a fruitful collaboration with Dr. Patric Meessen, and many useful discussions with Dr. Paul Cyr and Mr. Julio Reboucas. I am grateful to Drs. Kenneth MacFarlane and Richard Schutte for early advice and demonstration of laboratory techniques. I am greatly indebted to the departmental services staff. In particular, I thank Mr. Peter Borda for his careful attention to detail and for encouraging me to hold myself to a higher standard. I thank the Canadian International Development Agency (CIDA) who, through generous funding of the Zimbabwe-Canada General Training Facility, made it possible for me to pursue undergraduate studies in Canada: without this early support, I would not have aspired to graduate research. I acknowledge also the University of British Columbia for a University Graduate Fellowship and the Department of Chemistry for a Gladys Estella Laird scholarship. On a personal note, I thank Kirk for his steadfast friendship. I thank my family for their constant faith in me and Romina for her enduring love and support. xxiv This thesis is dedicated to my family and to Romina. xxv Chapter 1: General Introduction 1 General Introduction 1.1 Evolutionary Trends in Homogeneous Catalysis The case of hydrogenation serves to illustrate a general trend in the evolution of homogeneous catalysis.1'2 The first kinetically detailed report of a homogeneous catalytic hydrogenation by a platinum metal complex is due to Halpern and coworkers.3 In this system, aqueous solutions of chlororuthenate(II) were active for the hydrogenation of maleic, fumaric and acrylic acids at hydrogen pressures of up to 1 atm and at temperatures of 65-90 °C. Olefins not bearing carboxylate groups were not hydrogenated. Two years later, in 1963, Cramer et al. from Du Pont reported that a Pt-Sn complex in methanol solution was the first effective catalyst for the room temperature reduction of ethylene at 1 atm hydrogen pressure.4 In 1965, Ru-phosphine complexes were introduced by Wilkinson et al. who reported that RuCl2(PPh 3)„ (n = 3, 4) reacted with hydrogen at room temperature in ethanol/benzene to give R u C l ( H ) ( P P h 3 ) 3 which was an extremely active catalyst for the reduction of olefins and acetylenes.5 Likewise, Pt-Sn complexes containing phosphines were investigated: a wide range of olefins could be reduced by PtCl2(PPh 3) 2/SnCl2 in benzene/methanol under 40 atm of hydrogen at 90-105 °C. 6" 9 Also in 1965, what is now his eponymous catalyst, RhCl(PPh 3) 3, was reported by Wilkinson; 1 0 ' 1 1 its catalytic activity was discovered simultaneously and independently by Coffey.1 2 This complex, which is capable of effecting the reduction of a wide range internal and terminal olefins in benzene solution at 25 °C under 1 atm of hydrogen pressure,1 3 - 1 5 has been the focus of intensive study and is now accepted as the prototypical hydrogenation catalyst. In other reactions, as in hydrogenations, phosphine ligands came to be introduced. The Ruhrchemie hydroformylation process which is based on Co(H)(CO)4 was discovered by Roelen in 1938.1 6 The system has since been modified by the addition of trialkylphosphine. In the process developed by Shell, 1 7 the active catalyst, Co(H)(PBu3)(CO)3, 1 8 is less active for hydroformylation than the phosphine-free 1 References on page 13 Chapter 1: General Introduction compound, but gives higher selectivity for the linear aldehyde. In addition, the same catalyst can be used to hydrogenate the aldehyde so that long chain linear alcohols can be produced in a single reactor from terminal olefins. Because Co(H)(PBu3)(CO)3 is more stable than Co(H)(CO)4, significantly lower pressure can be employed (100 vs. 200-300 atm of H 2 /CO) and because of the enhanced stability and lower volatility of the catalyst, product-catalyst separation can be achieved by simple distillation. In addition, to the improved catalyst activity, selectivity and stability that were often achieved by the incorporation of phosphine ligands, this development greatly advanced the solution characterisation of the complexes by facilitating their study by 3 1 P N M R spectroscopy. But the evolution of catalysts from simple salts to metal-phosphine complexes also had a secondary effect: because of the water-insolubility of phosphines, water and hydrophilic liquids were gradually eliminated as solvents for homogeneous catalytic reactions. In the examples of homogeneous hydrogenation presented here, the astute reader will recognise the progression in solvent polarity from water to methanol to benzene/ethanol and finally to benzene. The impetus to reinstate water as a solvent for homogeneous catalytic reactions is derived from two sources: the need for a cheap, environmentally benign solvent and the need for facile product-catalyst separation. So strong have been these motivations that catalytic hydrogenation has now come full circle. Toth et al. have developed a Rh catalyst based on an ammonium phosphine ligand which is capable of effecting the asymmetric hydrogenation of cinnamic acid derivatives with high efficiency and excellent enantioselectivity in water solution.1 9 With the publication of dedicated books, 2 0 journal issues21 and numerous review articles, 2 2 - 2 4 catalysis in water has made an impressive return to the research spotlight. 1.2 Product-Catalyst Separation The problem of product-catalyst separation has been addressed most successfully by the development of biphasic catalysis. In this approach, the goal has been to combine the advantages of homogeneous catalysis,2 5-2 6 i.e., enhanced activity and selectivity, ease of characterisation and tailoring of the catalyst, and relatively mild operating conditions, 2 References on page 13 Chapter 1: General Introduction with the primary advantage of heterogeneous catalysis, i.e., ease of product-catalyst separation. In this sense, the active catalyst is "trapped" in or on a solid or liquid support which can be removed from the products once the reaction is complete by a simple physical process such as filtration or decanting. The "heterogenisation" of a homogeneous catalyst sees its most simple application in a system consisting in two immiscible liquid phases. The approach was first enunciated in the academic literature in 1973 by Manassen2 7 who wrote, "The use of two immiscible liquid phases, one containing the catalyst and the other containing the substrate, must be considered. The two phases can be separated by conventional means and a high degree of dispersion can be obtained through emulsification. This ease of separation may be particularly useful in situations where frequent catalyst regeneration or reactivation is required." The benefits of this approach over other types of immobilization are that even though the catalyst is heterogeneous, i.e., in a different phase from the reactants and products, the catalysis is homogeneous, i.e., goes via a discernible catalyst cycle which gives rise to reproducible reaction kinetics and which consists of molecular intermediates which are dispersed uniformly within the medium and are accessible to study by spectroscopy.28 Other means of immobilizing an intrinsically homogeneous catalyst, e.g., by attaching it to a solid polymer or other scaffold,29 are more problematic because stresses on the bond between metal and support lead over time to leaching of the metal. The first commercial application of liquid-liquid biphasic catalysis was the Shell Higher Olefin Polymerisation (SHOP) process due to Keim. 3 0> 3 1 In this system, ethylene is polymerised to give C4-C20 a-olefins at 80-120 °C and 70-140 bar. The catalyst is generated from Ni(cod)2 and Ph 2 PCH 2 COOH dissolved in 1,4-butanediol. The products, which are apolar and less dense than the solvent, are removed by decanting. Sheldon, in a 1992 article entitled "Organic Synthesis - Past, Present and Future," took the idea of Manassen one step further by saying that, "If a solvent is needed, it should preferably be water ... the development of water-soluble catalysts that work in a two phase system could be the answer in some cases."32 Biphasic catalysis in which one of the phases is water forms the basis of the Ruhrchemie/Rhone-Poulenc (R/RP) oxo process,33 which converts propylene to n-3 References on page 13 Chapter 1: General Introduction butyraldehyde on a scale of 3 x 105 metric tons per annum.34 In this process, an aqueous solution containing Rh and the sodium salt of tris(m-sulphonatophenyl)phosphine, also known as triphenylphosphinetrisulphonate or TPPTS, is used as the catalyst (the active species is [Rh(H)(CO)(TPPTS)3]9")- The product and catalyst are once again separated by decanting: olefin and aldehyde are only sparingly soluble in the aqueous phase, and the catalyst is completely insoluble in the organic fraction. A simplified schematic of the R/RP oxo process reactor design is shown in Figure 1.1. Figure 1.1 Schematic representing the application of the biphasic catalysis principle to industrial reactor design (adapted from ref. 28). The system illustrated here is very similar to that used in the R/RP oxo process. Prior to the 1980s, the field of homogeneous catalysis was almost exclusively confined to the use of organic solvents, with the exception of the Wacker process for the oxidation of olefins. The primary reasons for this were the insolubility of the catalysts in water and their chemical reactivity with it. Water was seen as detrimental, and extensive precautions were taken to exclude it even though it had been recognised in some cases to be beneficial, e.g., in solubilising Co catalyst precursors in the hydroformylation processes of B A S F 3 5 and Ruhrchemie. 3 6 ' 3 7 Also, the addition of aqueous sodium formate r e a c t a n t s p r o d u c t s r e a c t o r a q . c a t a l y s t s o l u t i o n 4 References on page 13 Chapter 1: General Introduction was found by Ruhrchemie to increase the overall yield by cleaving the initially-formed formic esters which hampered hydroformylation reactions.38 The tremendous success of the R/RP process is in large part due to the pioneering academic research of Kuntz who began investigating biphasic hydroformylation while at Rhone-Poulenc.33 His breakthrough can be attributed to a combination of three factors: the use of Rh as the catalytic centre instead of Co, the modification of the Rh-catalyst with phosphine ligands, and the choice of water as the catalyst-immobilizing phase.28 This last choice was based on the unique physiochemical properties of water and its fulfilment of several criteria posited by chemical engineers: it is a nearly ubiquitous substance which is cheaply available in large volumes of reasonable purity; it is odourless and colourless which makes detection of impurities easy; it is not flammable or combustible making it safe for scaled-up reactors; it is polar which makes it easy to separate from non-polar organic solvents, reactants and products from which it also (typically) differs in density; it has a high thermal conductivity, specific heat capacity and enthalpy of evaporation making it ideal as a solvent as well as a temperature control substance; and, finally, it is non-toxic and environmentally safe. 1.3 Water-soluble Phosphine Ligands The R/RP process constitutes an excellent example demonstrating the possibility for combination of the advantageous properties of metal-phosphine complexes with those of homogeneous catalysis in water. The prerequisite for this type of combination is a library of water-soluble phosphine ligands which can be used to solubilise and stabilise metal centres. In general, phosphine ligands can be made water-soluble by the incorporation of polar groups such as sulphonate, carboxylate, ammonium, pyridinium, phosphonium or hydroxyl. 2 4 The selection of ligands given in Chart 1.1 is not intended to be comprehensive, but rather is meant to convey an impression of the types and variety of water-soluble phosphines which have been investigated to date. (Where the ligands have names that are in common use in the literature, they are given; otherwise, they are denoted by numbers.) 5 References on page 13 Chapter 1: General Introduction 6 References on page 13 Chapter 1: General Introduction 1.4 Aqueous Phase Homogeneous Catalysis Following the success of the R/RP oxo process in 1984, aqueous phase homogeneous catalysis has received renewed attention. The potential for transferring a wide variety of catalytic reactions from organic solvents into water is the subject of ongoing research and, as mentioned previously, has been reviewed on a number of occasions. 2 2 - 2 4 Table 1.1 illustrates the general applicability of homogeneous, aqueous phase catalysis to organic synthesis by giving a selection of the reactions which can be carried out either in water or in water-organic solvent biphasic mixtures, and the metal-ligand combinations which catalyse them. Table 1.1. A selection of reactions catalysed by metal-phosphine complexes dissolved in water. Reaction type Example reaction and metal/ligand combination3 Ref. Hydroformylation o 33 Rh/TPPTS Hydrogenation C=C 53 C 0 2 R C O , R r 2 R NHCOR R NHCOR Rh/2 c = o 54-57 R Ru/TPPMS, Ru/TPA (by hydrogen transfer from sodium formate) O N 43,58, 59 Rh/BDDPos 7 References on page 13 Chapter 1: General Introduction Table 1.1 (cont.) Reaction type Example reaction and metal/ligand combination3 Ref. Hydrogenolysis A — ^ • H0X Pd/TPPMS, Pd/TPPTS, Pd/5 60 Carbonylation Reductive carbonylation NO, NH, + 3 CO, Pd/TPPTS Carboxylation 62 Br Base + H 2 0 + CO C O O H Pd/TPPTS C - C coupling Heck 63 Base Pd/TPPMS Stille 64,65 PhSnCI, Base COOH Pd/TPPMS Suzuki 63 B(OH)2 + Br Base^ —/7\— ( N Pd/TPPMS Hydrocyanation 66 + HCN C N Ni/TPPTS 8 References on page 13 Chapter 1: General Introduction Table 1.1 (cont.) Reaction type Example reaction and metal/ligand combination3 Ref. Polymerisation ROMP 47,67 N ( C H 3 ) 3 + O N ( C H 3 ) 3 + Ru/3/4 Ethylene/CO copolymerisation 68-70 < ^ + C O o Pd/1 a Throughout this thesis, * designates a chiral carbon centre. 1.5 Other Considerations The hydrophobic effect, which plays a critical role in the folding of proteins, in the formation and stabilisation of membranes and micelles, and in molecular recognition events such as those between antibody and antigen, substrate and enzyme, and receptor and hormone, also contributes significantly to organic reactions conducted in water. For example, Breslow's discovery of the dramatic acceleration of the Diels-Alder reaction in aqueous solution has been attributed to hydrophobic packing of the reagents.71-72 This has been postulated by Lubineau to be due to the high cohesive energy of water (2.3 kJ mL"1) which necessitates that every kinetically-controlled reaction between two non-polar molecules for which AV* is negative must be accelerated in water.73 In addition to the hydrophobic effect, rate enhancement in water is brought about by stabilisation of polar transition states, especially when one of the reagents is a hydrogen-bond donor or acceptor. Monte-Carlo simulations of Diels-Alder reactions involving methyl vinylketone as the dienophile reveal significant polarisation of the C=0 bond in the transition state, and consequently enhanced hydrogen-bonding. This makes a contribution to the rate enhancement equal to that of the hydrophobic effect.74 9 References on page 13 Chapter 1: General Introduction Another important factor influencing the reactivity of substances dissolved in water is pH. As an illustrative example of this point, Scheme 1.1 shows the equilibrium reactions of the dichloro-bridged diruthenium compound [RuCl2(TPPMS)2]2 with H 2 in the presence and absence of added TPPMS. Scheme 1.1 [RuCl 2(TPPMS) 2]2 + 2 H 2 - [Ru(H)Cl(TPPMS) 2] 2 + 2 H + + 2 CI" [RuCl 2(TPPMS) 2]2 + 2 H 2 + 2 TPPMS - 2 Ru(H)Cl(TPPMS) 3 + 2 H + + 2 CI" [RuCl 2(TPPMS) 2]2 + 4 H 2 + 4 TPPMS - 2 Ru(H) 2(TPPMS) 4 + 4 H + + 4 CI" Obviously, the position of these equilibria and therefore the rate of Ru/TPPMS-catalysed olefin hydrogenations, for example, will be crucially dependent upon pH. It has been determined by Joo et al. that below pH 3, the major species in solution is Ru(H)Cl(TPPMS) 3, whereas Ru(H) 2(TPPMS) 4 dominates in neutral and basic solutions; at pH 3-7, a mixture of hydrides is present.75 The implication of these findings is that kinetic data are valid only within a narrow pH range; differences in pH, therefore, may account for apparent contradictions in the literature. Since the active catalyst for olefinic hydrogenation is the monohydride "Ru(H)Cl(TPPMS)2" 7 6 while for aldehyde groups it is the dihydride Ru(H)2(TPPMS) 4 , 7 7- 7 8 the rate and product distribution of H 2 reductions of unsaturated aldehydes like prenal (2-methylpropen-2-al) are especially sensitive to pH.78,79 1.6 Atom Economy In a 1991 article in Science, Trost laid out the concept of atom economy, i.e., the maximisation of the mass of atoms in the reactants which is incorporated into the desired product.80 Two reactions can be compared: A + B —» C + D, and A + B —> C. The first type can be very wasteful indeed, as is the case in the Wittig reaction where the unwanted product, D, is triphenylphosphine oxide which has a molecular weight of 285 g mol"1 and usually dwarfs that of the desired olefinic product, C (Scheme 1.2). For example, when 10 References on page 13 Chapter 1: General Introduction the aldehyde is benzaldehyde, this reaction has an atom economy of only 28 %, and can be better regarded as a synthesis of OPPh 3 (atom economy = 75 %). Scheme 1.2 RCHO + Ph 3P=CH 2 -> RCH=CH 2 + OPPh 3 In the case of the addition reaction, A + B —> C, all of the atoms in the reactants are incorporated into the product. This is the definition of a zero-waste process. As the chemical industry faces ever more stringent environmental regulations, the more it will be pressured to devise synthetic strategies of this type. Sheldon expands the concept of atom economy to consider not only the amount of waste resulting from a reaction, but also its nature}2 In addition, he takes into account not only the reactants and products, but also the volume and nature of solvent(s). If possible, he asserts, solvents should not be used at all but, when necessary, the solvent should preferably be water. The aqueous phase catalytic hydration of olefins (Scheme 1.3) takes these concepts to their highest level: the solvent is environmentally benign, and is a reactant in an addition, zero-waste reaction. So appealing is this result that the anti-Markovnikov addition of water to terminal olefins to produce linear alcohols has been cited as one of the ten remaining challenges for homogeneous catalysis.81 In principle, prochiral olefins could also be asymmetrically hydrated to give optically pure alcohols. Scheme 1.3 RCH=CH 2 + H 2 0 -> RCH 2 CH 2 (OH) and/or RC*H(OH)CH 3 1.7 Scope of this Thesis This thesis is founded on the study of platinum metal complexes of two nitrogen-containing families of chelating diphosphine ligands. The first family is characterised by having o-./V,7V-dimethylanilinyl rings appended to each of the P-atoms, and the second by 11 References on page 13 Chapter 1: General Introduction having o-pyridyl substituents in the same positions. A l l of the ligands and their monikers are given in Chart 1.2. Prior to the beginning of this study in September 1996, the pyridyldiphosphine dpype had been reported by this group,8 2 and the other ligands, with the exception of dmapm which is new in this work, had been isolated in impure form by a previous postdoctoral fellow, Martin Smith, but had not yet been reported in the open literature. Chart 1.2 Anilinyldiphosphine Ligands N M e 2 Me 2 N dmapcp Pyridyldiphosphine Ligands dpype dpypcp The main objectives of this work were to uncover good synthetic strategies for each ligand family and their corresponding platinum metal complexes, to study the solution behaviour of the complexes, which hopefully would show enhanced water-12 References on page 13 Chapter 1: Generallntroduction solubility, and to use these as catalysts in aqueous or mixed aqueous/organic homogeneous reactions with the primary goal being the hydration of olefins. The anilinyldiphosphine ligands and M 1 1 (M = Pt, Pd) complexes of dmape and dmapcp are discussed in Chapter 2. Chapter 3 investigates M 1 1 and M 2 ' complexes of dmapm, and discusses the use of homo- and heterobimetallic complexes bridged by dmapm as catalysts for the Heck reaction in a water/organic medium. Chapter 4 summarises the coordination chemistry of the pyridyldiphosphine ligands with M 0 / " precursors and some reactions of the complexes. Finally, Chapter 5 presents a summary of the results for the attempted catalytic hydration of maleic acid in water. 1.8 References 1. James, B. R. Homogeneous Hydrogenation; John Wiley & Sons: New York, 1973. 2. Chaloner, P. A. ; Esteruelas, M . A. ; Joo, F.; Oro, L. Homogeneous Hydrogenation; Kluwer Academic Publishers: Dordrecht, 1994. 3. Halpern, J.; Harrod, J. F.; James, B. R. J. Am. Chem. Soc. 1961, 83, 753. 4. Cramer, R. D.; Jenner, E. L.; Lindsey, R. V. ; Stolberg, U . G. J. Am. Chem. Soc. 1963,55, 1691. 5. Evans, D.; Osborn, J. A . ; Jardine, F. H. ; Wilkinson, G. Nature 1965, 208, 1203. 6. Tayim, H. A. ; Bailar, J. C. Am. Chem. Soc. 1967, 89, 4330. 7. Adams, R. W.; Batley, G. E.; Bailar, J. C. J. Am. Chem. Soc. 1968, 90, 6051. 8. Adams, R. W.; Batley, G. E.; Bailar, J. C. Inorg. Nucl. Chem. Lett. 1968, 4, 455. 9. Ichinohe, Y . ; Kameda, N . ; Kujirai, M . Bull. Chem. Soc. Jap. 1969, 42, 3614. 10. Jardine, F. H. ; Osborn, J. A. ; Wilkinson, G. Chem. Ind. 1965, 560. 11. Young, J. F.; Osborn, J. A. ; Jardine, F. H. ; Wilkinson, G. Chem. Commun. 1965, 131. 12. Coffey, R. S. British Patent 1 121 642, 1965 (Imperial Chemical Industries); from ref. 1. 13. Osborn, J. A. ; Jardine, F. H.; Young, J. F.; Wilkinson, G. J. Chem. Soc. (A) 1966, 1711. 14. Jardine, F. H. ; Osborn, J. A. ; Wilkinson, G. J. Chem. Soc. (A) 1967, 1574. 13 References on page 13 Chapter 1: General Introduction 15. Montelatici, S.; van der Ent, A. ; Osborn, J. A. ; Wilkinson, G. J. Chem. Soc. (A) 1968, 1054. 16. Roelen, O. German Patent 849 548, 1938 (Ruhrchemie AG); from ref. 25. 17. Slaugh, L. H . ; Mullineaux, R. D. United States Patent 3 239 569 and 3 239 570, 1966 (Shell Oil Company); from ref. 25. 18. Ibers, J. A . J. Organomet. Chem. 1968,14, 423. 19. Toth, I.; Hanson, B. E.; Davis, M . E. J. Organomet. Chem. 1990, 396, 363. 20. Aqueous Phase Organometallic Catalysis; Cornils, B. ; Herrmann, W. A. , Eds; Wiley-VCH: Weinheim, 1997. 21. Horvath, I. Mol. Catal. (A) 1997,116. 22. Sinou, D. Top. Curr. Chem. 1999, 206, 41. 23. Joo, F.; Katho, A . J. Mol. Catal. A 1997,116, 3. 24. Herrmann, W. A. ; Kohlpainter, C. W. Angew. Chem. Int. Ed. Engl. 1993, 32, 1524. 25. Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis; 2nd ed.; John Wiley & Sons, Inc.: New York, 1992. 26. Parshall, G. W.; Nugent, W. A. Chemtech 1988, 18, 184. 27. Manassen, J. in Catalysis: Progress in Research; Basolo, F. and Burwell, R. L., Ed.; Plenum Press: New York, 1973. 28. Cornils, B. in Modern Solvents in Organic Synthesis; Knochel, P., Ed.; Springer: Berlin, 1999; Vol . 106. 29. Hartley, F. R. Supported Metal Complexes, a New Generation of Catalysts; Reidel: Dordrecht, 1985. 30. Bauer, R. S.; Orinda, H. C ; Glockner, P. W.; Keim, W. United States Patent 3 635 937, 1972 (Shell Oil Company); from ref. 28. 31. Keim, W. Chem. Ing. Tech. 1984, 56, 850. 32. Sheldon, R. Chem. Ind. 1992, 903. 33. Kuntz, E. G. French Patent 2 314 910, 2 349 562, 2 338 253 and 2 366 237, 1976 (Rhone-Poulenc); from ref. 28. 34. Cornils, B.; Herrmann, W. A.; Eckl, R. W. Mol. Catal. A 1997,116, 27. 35. Nienburg, H. German Patent 948 150, 1953 (BASF); from ref. 28. 36. Kolling, H. ; Buckner, K. ; Stiebling, E. British Patent 736 875, 1952 (Ruhrchemie AG); from ref. 28. 14 Chapter 1: General Introduction 37. Rolling, H. ; Biickner, K. ; Stiebling, E. German Patent 949 737, 1965 (Ruhrchemie AG); from ref. 28. 38. Tummes, H. ; Meis, J. United States Patent 3 462 500, 1965 (Ruhrchemie AG); from ref. 28. 39. Joo, F.; Beck, M . T. React. Kin. Catal. Lett. 1975, 2, 257. 40. Joo, F.; Toth, Z.; Beck, M . T. Inorg. Chim. Acta 1977, 25, L61. 41. Herrmann, W. A. ; Albanese, G. P.; Manetsberger, R. B.; Lappe, P.; Bahrmann, H. Angew. Chem. Int. Ed. Engl. 1995, 34, 811. 42. Bartik, T.; Bartik, B.; Hanson, B. E.; Glass, T.; Bebout, W. Inorg. Chem. 1992, 31, 2667. 43. Bakos, J.; Orosz, A. ; Heil, B.; Laghmari, M . ; Lhoste, P.; Sinou, D. J. Chem. Soc, Chem. Commun. 1991, 1684. 44. Alario, F.; Amrani, Y . ; Colleuille, Y . ; Dang, T. P.; Jenck, J.; Morel, D.; Sinou, D. J. Chem. Soc, Chem. Commun. 1986. 45. Daigle, D. J. Inorg. Synth. 1998, 32, 40. 46. Toth, I.; Hanson, B. E. Tetrahedron: Asymmetry 1990, 1, 895. 47. Lynn, D. M . ; Mohr, B.; Grubbs, R. H. Am. Chem. Soc. 1998,120, 1627. 48. Mohr, B.; Lynn, D. M . ; Grubbs, R. H. Organometallics 1996, 75, 4317. 49. Gilman, H. ; Brown, G. E. J. Am. Chem. Soc. 1945, 67, 824. 50. Renaud, E.; Russell, R. B.; Fortier, S.; Brown, S. J.; Baird, M . C. J. Organomet. Chem. 1991, 419, 403. 51. Amrani, Y . ; Sinou, D. J. Mol. Catal. 1986, 36, 319. 52. See Section 4.3.3 (p. 165) of this thesis. 53. Toth, I.; Hanson, B. E. Tetrahedron: Asymmetry 1990, 7, 913. 54. Joo, F.; Benyei, A . Organomet. Chem. 1989, 363, C19. 55. Benyei, A. ; Joo, F. Mol. Catal. 1990, 58, 151. 56. Darensbourg, D. J.; Joo, F.; Kannisto, M . ; Katho, A . Organometallics 1992, 77, 1990. 57. Darensbourg, D. J.; Joo, F.; Kannisto, M . ; Katho, A. ; Reibenspies, J. H. ; Daigle, D. J. Inorg. Chem. 1994, 33, 200. 58. Lensink, C.; Rijnberg, E.; Vries, J. G. d. J. Mol. Catal. (A) 1997, 77o~, 199. 59. Lensink, C.; Vries, J. G. d. Tetrahedron: Asymmetry 1992, 3, 235. 15 Chapter 1: General Introduction 60. Okano, T.; Moriyama, I.; Konishi, H. ; Ki j i , J. Chem. Lett. 1986, 1463. 61. Tafesh, A . M ; Beller, M . Tetrahedron Lett. 1995, 36, 9305. 62. Monteil, F.; Kalck, P. J. Organomet. Chem. 1994, 482, 45. 63. Casalnuovo, A. L.; Calabrese, J. C. J. Am. Chem. Soc. 1990,112, 4324. 64. Roshchin, A . I.; Bumagin, N . A. ; Beletskaya, I. P. Tetrahedron Lett. 1995, 36, 125. 65. Rai, R.; Aubrecht, K . B.; Collum, D. B. Tetrahedron Lett. 1995, 36, 3111. 66. Kuntz, E. United States Patent 4 087 452, 1978 (Ruhrchemie AG); from ref. 20. 67. Lynn, D. M . ; Mohr, B.; Grubbs, R. H. ; Henling, L. M . ; Day, M . W. J. Am. Chem. Soc. 2000, 122, 6601. 68. Verspui, G.; Papadogianakis, G.; Sheldon, R. A. Chem. Commun. 1998, 401. 69. Verspui, G.; Schanssema, F.; Sheldon, R. A. Appl. Catal. (A) 2000, 198, 5. 70. Verspui, G.; Schanssema, F.; Sheldon, R. A. Angew. Chem. Int. Ed. 2000, 39, 804. 71. Rideout, D. C.; Breslow, R. J. Am. Chem. Soc. 1980,102, 7816. 72. Breslow, R. Acc. Chem. Res. 1991, 24, 159. 73. Lubineau, A. J. Org. Chem. 1986, 51, 2142. 74. Furlani, T. R.; Gao, J. Org. Chem. 1996, 61, 5492. 75. Joo, F.; Kovacs, J.; Benyei, A. ; Katho, A. Angew. Chem. Int. Ed. Engl. 1998, 37, 369. 76. Toth, Z.; Joo, F.; Beck, M . T. Inorg. Chim. Acta 1980, 42, 153. 77. Fache, E.; Santini, C.; Seocq, F.; Basset, J. M . J. Mol. Catal. 1992, 72, 331. 78. Hernandez, M . ; Kalck, P. J. Mol. Catal. 1997,116,117. 79. Grosselin, J. M . ; Mercier, C ; Allmang, G.; Grass, F. Organometallics 1991, 10, 2126. 80. Trost, B. M . Science 1991, 254, 1471. 81. Haggin, J. Chem. Eng. News, May 31 1993; p 23. 82. Baird, I. R.; Smith, M . B.; James, B. R. Inorg. Chim. Acta 1995, 235, 291. 16 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2 Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.1 Introduction The following chapter deals with the three tetraanilinyldiphosphine ligands shown in Chart 2.1 and a selection of their Pt" and Pd" complexes. The ligands fall into a general class whose members incorporate at least 2 tertiary P-atoms and at least 1 amine N-atom. As these have not yet been discussed as a collective in the literature (although a significant body of work exists), the following paragraphs will provide a brief overview of the known types, their metal complexes and uses in catalysis. Chart 2.1 NMe 2 Me 2 N dmapcp Whereas the synthesis of dmapm is covered in this chapter (Section 2.9.1.1) as is a discussion of the organometallic intermediates which result from its reaction with PdCl2(cod) (Section 2.5.8), a detailed investigation of its complexes and their use in catalysis appears in Chapter 3. 2.1.1 Philosophical approach Over the last 25 years, amine functional groups have been incorporated into diphosphine ligands with a view to surmounting the most problematic aspect of homogeneous 17 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes catalysis: separation of the catalyst from the product. The most successful approach to solving this problem has been to "heterogenise" the homogeneous catalyst either by its incorporation into an insoluble polymer, or by solubilising it in a solvent (such as H2O) with which the reactants and products are only very slightly miscible, or by extracting it post-reaction with a solvent (such as aqueous acid) in which the product does not dissolve, etc. A l l of these approaches have been investigated with varying degrees of success. Prior to the 1990 pioneering work of Toth et al.,us amine N-atoms in diphosphine ligands were viewed as functional groups either for incorporation of the ligands into polymers (Achiwa, 6 Stille and coworkers;7 Section 2.1.2.1), or for the subsequent introduction of water-solubilising sulphonate moieties (Whitesides and coworkers;8 Section 2.1.2.2) and not as potential water-solubilising sites themselves (via quaternisation). Since 1990, however, most of the work using this ligand class has been with a view to water-solubilisation of metal complexes for use in aqueous catalysis via protonation or alkylation of the N-atoms. The N-atoms are typically distant from the P-centres and are not involved in coordination in the complexes. For example, in all of the ligands made by Toth et al. (see p. 23), the N-atoms are incorporated as /?-C6H 4NMe2 substituents which prohibits their coordination to the same metal centre bound by the P-atoms. No reports of this type of ligand in a bridging mode have appeared. The ligands presented in this and the following chapter (Chart 2.1) are different from those of Toth et al. in that they bear ortho- as opposed to /rara-anilinyl substituents on their P-atoms. This makes the N-atoms available for coordination to the same metal centre as the P-atoms, and in some cases renders complexes water-soluble by displacing halide from the coordination sphere. Because the N-atoms are positioned for coordination, and because of the different affinities of group 10 metals for P ("soft") and N ("hard"), complexes of the o-anilinyldiphosphine ligands exhibit fluxional behaviour in organic solution, i.e., the ligands are "hemilabile". 18 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.1.2 A brief literature survey By far the majority of the research into diphosphine ligands bearing amine groups has been directed towards the search for hydrogenation catalysts, although some work has also focussed on hydroformylations. Without exception, the catalysts that have appeared in the literature to date have been complexes of Rh 1 and Ru". 2.1.2.1 Polymer-attached optically-active diphosphine ligands The first group of ligands to combine 2 or more tertiary P-atoms and at least 1 amine N -atom is derived from prolene and is due to Achiwa. 9 Representatives of this group are illustrated in Chart 2.2 (1-4). In these ligands, the N-atom serves as an attachment point for groups such as -C(0)CH=CH2 which facilitate the incorporation of the diphosphine into polymers. The goal of this approach was to heterogenise the catalyst. Stille's group has successfully copolymerised both the S,S and R,R enantiomers of 3 with 2-hydroxyethyl methacrylate (HEMA) to give polymer 5 (Chart 2.3), and N,N-dimethylacrylamide (polymer not shown) by free-radical initiators using ethylene dimethacrylate as a crosslinking agent.7 These workers used Rh 1 complexes of these polymers in the asymmetric hydrogenation of dehydroaminoacids with e.e.s as high as 91 %. Chart 2.2 In a similar fashion, Achiwa has copolymerised S,S-4 with H E M A in the presence of azobisisobutylnitrile.6 Subsequent reaction of this polymer with [Rh((x-Cl)(cod)]2 generated an insoluble catalyst for the hydrogenation of itaconic acid with e.e.s up to 82 %, and (Z)-2-acetamidocinnamic acid with optical yields of 70 %. In order to facilitate product catalyst separation by using two immiscible liquid phases, Malmstrom and Andersson have generated the water-soluble polymer 6 (Chart 19 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.3) by acylation of S,S-1 with poly(acrylic acid) in the presence of dicyclohexylcarbodiimide, followed by neutralisation with N a H C 0 3 . , ( ) Reaction of this with [Rh(nbd)2][CF3S03] gave a polymeric water-soluble catalyst which was employed in the asymmetric hydrogenation of (Z)-2-acetamidocinnamic acid with e.e.s of up to 93 %. Chart 2.3 — T ^ " p h 2 p / s ^ _ ^ P P h 2 0.05 0.85 0.10 (6) Ligand 2 has also been successfully employed in the homogeneous asymmetric hydrogenation of itaconic acid and a-acylamino acrylic acid and its derivatives. Ojima et al. have found that the stereoselectivity of the reaction catalysed by [Rh(cod)(2)][C104] is sensitive to H 2 pressure, the presence or absence of base (NEt 3), and the temperature.11 Lemaire and coworkers have synthesised a 6,6' methylamine-substituted BINAP in 5 steps starting from S-binol (7 in Scheme 2.1). 1 2 This ligand has been copolymerised with 2,6-tolylene diisocyanate to give 8. Reaction of 8 with either of [RuCl2(CeH6)]2 or 20 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Ru(cod)(2-Me-allyl)2 gave catalysts whose exact composition was not determined but which were effective for the asymmetric hydrogenation of (3-ketoesters with e.e.s consistently ca. 99 %; the catalyst can be recovered by simple filtration as the polymer is insoluble in MeOH in which the reactions are conducted, and there is negligible loss of activity over 4 cycles. Scheme 2.1 (8) 2.1.2.2 Water-solubilisation via incorporation of sulphonate groups The use of a secondary amine incorporated in a diphosphine ligand as a site for the subsequent introduction of sulphonate groups was initiated by Whitesides's group.8 This work was based on bis(2-diphenylphosphinoethyl)amine (9, Chart 2.4). The general method for sulphonate group introduction involved the coupling of the amine and an acid chloride acid anhydride (trimellitic (10) or tricarballylic (11)) followed by opening of the acid anhydride with sodium taurinate (N^CB^CFf^SOsNa) to give 12, for example. Reaction of the sulphonated ligands with [Rh(u-Cl)(nbd)]2 in the presence of AgOTf generated catalysts (which remain uncharacterised) for the aqueous phase hydrogenation of unsaturated acids. The highest TOF was > 200 h"1, recorded for the hydrogenation of CH 2 =C(NHAc)C0 2 H by 12-Rh. 21 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Chart 2.4 (12) 2.1.2.3 Water-solubilisation via quaternisation Although water-solubilisation of phosphine ligands and their corresponding complexes by the introduction of charge via alkylation of amine groups had been accomplished earlier (e.g., in the 1982-3 work of Baird et al. on Ph2PCH 2CH 2NMe3 + and its Fe°, Mo° 1 3 and Rh 1 complexes14), it was not until 1990 that such an approach had been applied to dz'phosphine ligands, and more specifically to chiral diphosphine ligands for use in aqueous phase, asymmetric catalysis. Toth et al. reported the synthesis of a range of chiral diphosphine ligands bearing p - C 6 H 4 N M e 2 substituents on the P-atoms (Chart 2.5, 13-15, 1 7 ) . M These workers have shown that Rh 1 complexes of the ligands can be used as asymmetric hydrogenation catalysts in water once the N-atoms have been quaternised either via protonation with aq. H B F 4 or by alkylation with (CH 3 ) 3 OBF 4 . The [(15)Rh(nbd)][BF4] and [(16)Rh(nbd)][BF4] complexes could also be immobilized on Nafion and other cation exchange resins according to Scheme 2.2; the modified resins could be used for the asymmetric hydrogenation of dehydroaminoacid derivatives with e.e.s up to 76 % and could be recycled 6 times with minimal loss of catalytic activity.5 22 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Scheme 2.2 (Nafion)-CF 2 CF 2 S0 3 H + [15-Rh][BF4] -> [(Nafion)-CF 2CF 2S0 3][15-Rh] + H B F 4 Following their report of 8, Lemaire's group demonstrated that 7 (Scheme 2.1) can be protonated at the N-atoms with HBr . 1 5 Reaction of the ammonium salt with either of [RuCl2(C6Fl6)]2 or Ru(cod)(2-Me-allyl)2 gave a catalyst, again uncharacterised, which was water-soluble and effective for the biphasic reduction of ethylacetoacetate with good enantioselectivity. The catalyst could be recycled once the water-soluble alcohol product had been extracted with pentane. The recycled catalyst showed minimal loss of activity, but did suffer a loss of enantioselectivity by the 3 r d cycle. Ligand 13 was reported by Mirabelli et al. in 1987, and a derived Au 1 complex was assayed for antitumor activity.1 6 However, no synthetic or characterisation data for these compounds were included in the publication. Ligand 17 is covered by a Hoechst patent.17 Chart 2.5 / \ / \ P R 2 R 2 P I (13) (14) (15) R 2 P 2 2 P R 2 R 2 P P R 2 R 2 P P R 2 R 2 P ' (16) (17) R 2 P P R 2 R = — v v — N M e 2 < 1 8 ) Nagel and Kinzel were the first to report ligand 20 (Chart 2.6) and its Rh complex [Rh(cod)(20)][BF4].18 Treatment of the complex with (CH 3 ) 3 OBF 4 resulted in the methylation of the N-atom to give the water-soluble complex [Rh(cod)(Me-20)][BF4]2 23 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes which was employed in the asymmetric hydrogenation of a-acylaminocinnamic acid in H2O to give (5)-N-acetylphenylalanine with 90 % e.e.. Chart 2.6 R Ph 2 P " P P h 2 R = H , Me (19) (20) 2.1.2.4 Post-reaction separation by aqueous acid extraction Van Leeuwen and coworkers have synthesised a number of />-Et2NCH2C6H4-substituted diphosphine ligands (Chart 2.7). 1 9> 2 0 These N-containing analogues of bisbi (21), POP (22) and xantphos (23) have large bite angles that allow them to occupy equatorial sites in the trigonal bipyramidal Rh-hydride complex which is the crucial intermediate in the hydroformylation catalytic cycle. This geometry leads to higher proportions of the n-aldehyde product than does that in which the P-atoms of the smaller bite angle ligands are constrained to chelate in the apical and equatorial positions.21 Chart 2.7 The Rh catalysts with ligands 21-23 in the hydroformylation of oct-l-ene in toluene solution behaved similarly to their phenyl analogues, giving almost identical 24 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes activity and linearrbranched aldehyde product ratios. In addition, the catalysts could be almost quantitatively recovered using aqueous acid, but re-extraction after neutralisation by fresh toluene for subsequent catalytic runs was only partially successful and the recycled catalysts had residual activities ranging between 51-86 % . 1 9 ' 2 0 2.1.2.5 Miscellaneous ligand In a study of the hemilabile nature of phosphine ligands containing amine groups, Andrieu et al. have synthesised compound 24 (Chart 2.8) by treatment of 2 equiv. of Ph 2PCH 2CH(Ph)NH(Ph) with oxalylchloride.22 Although the ligand is well characterised, these researchers do not report any of its metal complexes. Chart 2.8 Ph Ph (24) 2.2 Scope This chapter describes the syntheses (Sections 2.9.1) and structures (Sections 2.3.3) of alkyl-bridged anilinyldiphosphine ligands of the general formula (o-Me 2 NC 6 H 4 ) 2 P(X)P(o-C 6 H 4 NMe 2 ) 2 (X = C H 2 (dmapm), (CH 2 ) 2 (dmape), cyclic-C 5 H 8 (dmapcp); Chart 2.1) and investigates their coordination chemistry with Pt" and Pd". The syntheses of the complexes are given in Section 2.9.2, and ORTEP representations of the molecular structures of [PtCl(P,P',/V-dmapcp)]+ (Figure 2.4), PtCl2(P,P'-dmape) (Figure 2.10) and [PdCl(dmapeO)]2 2 + (Figure 2.16) are presented. The behaviour of a selection of the complexes dissolved in chlorinated solvents is investigated: thermodynamic parameters for the equilibrium between P,P'- and P.P'.N-25 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes bound dmape are evaluated for MCl2(dmape) (M = Pd, Pt) (Section 2.4.2.1); exchange of free and coordinated anilinyl N-atoms is demonstrated for [MCl(P,P',./V-dmapcp)]Cl and a mechanism for the stereoselective formation of these compounds is presented (Section 2.4.1). The possible structure(s) of organometallic intermediates which result from the reaction of dmapcp (Section 2.5) and dmapm (Section 2.5.8) with PdX2(diolefm) (X = halide; diolefin = cod, nbd) en route to the formation of the expected metal-ligand complexes are discussed, and the reaction of PdCi2(dmape) with K O H to form a bis(phosphine) monooxide (BPMO) complex (Section 2.6.1), as well as with other bases and Pd° to give metal-metal bonded species (Section 2.6.2), is investigated. 2.3 The 2-Anilinyldiphosphine Ligands 2.3.1 Historical A family of 2-anilinyldiphosphine ligands of formula (o-NMe2C6H4)2P(X)P(o-NMe2C6H4)2 (X - CH2 (dmapm), ( C H ^ (dmape), cyclic-CsHg (dmapcp); Chart 2.1) was synthesised during the course of this work. The ligand family was initially conceived by Dr. Martin Smith, a postdoctoral fellow in the James laboratory during 1992-3, as a variation on the /^-anilinyldiphosphine compounds which had been made previously by Toth et al.]-4 He attempted to make dmape and dmapcp, but was not able to isolate them pure. A subsequent postdoctoral fellow, Dr. Patric Meessen (1998-9), made subtle modifications to the syntheses which resulted in the isolation of pure dmape and dmapcp. This author, using the procedure developed by Meessen, was the first to make dmapm. 2.3.2 Synthesis Sulphonated phosphines, like TPPTS, are typically made by the direct action of oleum on the corresponding neutral ligand; sulphonation occurs exclusively in the meta position.2 3 Because of the oxidising effects of the oleum, a significant proportion of the phosphine is converted to phosphine oxide which complicates the purification process, and the harsh reaction conditions preclude modification of ligands like DIOP which bear sensitive (in this case, acetal) functional groups.24 A solution to the former problem involves reduction of the phosphine oxide by trichlorosilane following conversion of the 26 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes sulphonates to sulphonic esters; hydrolysis of the esters then regenerates the water-soluble ligand. 2 5 A n important procedure by Herrmann et al. allows direct sulphonation of arylphosphines without oxidation of the P-atom.2 6 Here, the sulphonating reagent is the superacidic mixture of orthoboric and anhydrous sulphuric acid. Because free S 0 3 is absent and because the P-atom is protected by protonation, oxidation is eliminated. As an alternate approach, sulphonate groups can be introduced during the synthesis of the aryl phosphine according to Scheme 2.3. 2 7 Scheme 2.3 In a similar manner, the water-soluble soluble phosphines Ar3.„P(m-S03-C6H4)„ (n = 1-3; Ar = Ph, 2-py, 3-py) can be made by the reaction of /7-fluorobenzenesulphonate with P H 3 or primary or secondary phosphines in the "superbasic" medium K O H / D M S O . 2 8 The last two approaches offer the significant advantages of eliminating phosphorus oxidation and giving freedom over the position of the substituent on the ring. The synthesis of the anilinyldiphosphine ligands (Sections 2.9.1.1-2.9.1.3) is similar to that shown in Scheme 2.3 in that it involves coupling of a substituted arene with a phosphine, although in this case the alkali metal attends the aryl group. The procedure is the same in principle as that used for the synthesis of the pyridyldiphosphine ligands (Section 4.3.1). It involves a low-temperature, two-step, one-pot procedure in which A^7V"-dimethyl-2-lithioaniline is first generated by the reaction of "BuLi with N,N-dimethyl-2-bromoaniline, and then this is exposed to 0.25 mol equiv. of the appropriate chlorodiphosphine (see Scheme 2.4 for an example). The ligand is removed from the reaction mixture by dissolution in dil. aq. HC1 and purified after neutralisation by recrystallisation from boiling EtOH. The procedure results in an analytically pure, phosphine oxide-free product. 27 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.3.3 Structures Single crystals of dmape and dmapcp suitable for X-ray diffraction analysis were isolated by Meessen from CH2CI2 solutions onto which EtOH had been layered; the structures are reported in this thesis because of their relevance to this work and because they have not yet appeared in the literature. The former compound crystallised with 2 independent molecules in the unit cell. One of these contained an inversion centre and is represented as an ORTEP in Figure 2.1. The other is essentially the same but does not contain an inversion centre. The ORTEP representation of dmapcp is given in Figure 2.2, and selected bond lengths and angles for the two ligands appear in Table 2.1 and Table 2.2, respectively. 28 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Figure 2.1 ORTEP representation of the centrosymmetric crystallographic form of dmape (50 % ellipsoids). H-atoms are omitted for clarity. Table 2.1 Selected bond distances (A) and angles (°) for dmape with estimated standard deviations given in parentheses. C(35)—C(35*) 1.538(4) P (3 ) - -C(35) 1.849(2) P(3)—C(36) 1.844(2) P (3 ) - -C(44) 1.837(2) N(5)—C(37) 1.426(2) N(5)--C(42) 1.466(3) N(5)—C(43) 1.461(3) C(35*)—C(35)—P(3) 110.7(2) C(35> -P (3 ) - -C(36) 96.62(9) C(35)—P(3)—C(44) 101.99(8) C(36)--P (3 ) - -C(44) 101.42(8) C(37)—N(5)—C(42) 115.2(2) C(37> -N(5) --€(43) 115.6(2) C(42)—N(5)—C(43) 111.2(2) 29 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Figure 2.2 ORTEP representation of dmapcp (50 % ellipsoids). H-atoms are omitted for clarity. Table 2.2 Selected bond distances (A) and angles (°) for dmapcp with estimated standard deviations given in parentheses. C ( l ) - P ( l ) 1.882(2) C(2)-P(2) 1.879(2) C ( l ) - C ( 2 ) 1.537(3) P ( l ) -C(6 ) 1.852(2) P(2)—C(22) 1.831(2) N ( l ) - C ( 7 ) 1.441(3) N( l ) -C(12) 1.465(4) N(4)—C(31) 1.411(3) N(4)—C(36) 1.466(3) P ( l ) - C ( l ) - C ( 2 ) 108.47(16) C ( l ) - C ( 2 ) - P ( 2 ) 111.33(15) C ( l ) - P ( l ) - C ( 6 ) 99.75(10) C(2)—P(2)—C(22) 102.04(10 C(7)—N(l)—C(12) 116.2(2) C(31)—N(4)—C(37) 115.4(2) 30 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.4 Pt" and Pd" Complexes Reaction of 1 equiv. of dmape or dmapcp with PtCl2(cod) or /ra«.s-PdCl2(PhCN)2 gives the 1:1 complexes MC1 2 (P-P) or [MC1(P-P)]C1 quantitatively. The reactions proceed swiftly in CH2CI2 at r.t., except that of dmapcp with PtC^Ccod) which was successful when conducted in refluxing CH 2 C1CH 2 C1. An alternate synthesis of [FtC\(P,P',N-dmapcp)]Cl involves reaction of the ligand with cz's-PtC^MeCNh in refluxing MeCN. The complexes have been characterised by ' H , 'HI 3 1 ?} and 3 1 P{'H} N M R spectroscopies and elemental analysis and, in the Pt cases, by X-ray crystallography. The molecular structure of [PdCl(P,P',A^-dmapcp)]Cl was also determined but could not be refined sufficiently well for publication. The following sections describe in detail the coordination modes of the ligands with each metal, and the solution behaviour of the complexes. With the exception of organometallic intermediates resulting from reactions of dmapm with PdX2(diolefin) (X = halide; diolefin = cod, nbd) (Section 2.5.8), dmapm complexes are discussed in Chapter 3. 2.4.1 Dmapcp complexes 2.4.1.1 Platinum The number and relative integrations of the singlets due to the NC//3 protons in the ' H N M R spectra of anilinyldiphosphine complexes constitute an invaluable diagnostic tool for determining the overall structure of the molecule. The two anilinyl substituents on each P-atom in dmapcp are diastereotopic and are related to those on the opposite P-atom by a C2 axis, i.e., the ligand bears homotopic pairs of diastereotopic anilinyl rings and the 'Ft N M R spectrum therefore shows two singlets due to the NCrY3 protons (5 2.55 and 2.63). In the case of [PtCl(P,P',/V-dmapcp)]Cl, the C 2 symmetry is lifted and 4 sharp peaks due to NC//3 protons are apparent at 5 3.50, 3.05, 2.67 and 2.22 in a 1:1:2:2 ratio (i.e., 3:3:6:6 protons). The remaining 6 protons are manifested by a broad peak at 5 2.32 (Figure 2.3). The pattern of NC/Y 3 peaks defines a P,P',N coordination mode for the ligand; the cartoon in Chart 2.9 gives a representation of the grouping of N C H 3 groups in the [PtCl^P'./V-dmapcp)]"1" cation. On the basis of integration, the two most downfield singlets (equivalent to 3 protons each) are assigned unequivocally to methyl groups a and b which are associated with the Pt-bound N-atom: these peaks show the most dramatic 31 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes coordination shifts from their position in the free ligand (A8 = 0.91 and 0.46 ppm), and they show 3-bond H- 1 9 5 Pt coupling (with magnitudes of 17 and 22 Hz, respectively). c,d a b # # e,f| 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i i 11 [ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 4.0 3.0 2.0 ppm 1.0 Figure 2.3 The 8 1-4 region of the 'H{ 3 1 P} N M R (121 MHz, CDC1 3 , 300 K) spectrum of [PtCl(P,P',/V~-dmapcp)]Cl. Asterisks (*) and number symbols (#) indicate peaks due to the CH2 and CH protons, respectively, in the ligand "backbone". Other assignments are discussed in the text. The singlet at 8 1.56 is due to H 2 0 . Chart 2.9 5 ^ P t \ / / ~|\|"— C Me' "Me M e 2 N a,b e,f 32 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes For a structure like the one shown in Chart 2.9, 3 equal-intensity peaks due to methyl groups associated with free N-atoms are expected: c,d, e,f and g,h. Peaks due to the last two pairs should have similar chemical shifts, while that due to c,d is expected to be somewhat removed from the others. On this basis, the peak at 5 2.67 (6 protons) is due to methyl groups c and d and those at 8 2.32 (6 protons) and 2.22 (6 protons) are due to groups e-h. 3 1 1 The P{ H} N M R spectrum reveals chemically inequivalent P-atoms. A positive assignment of the peaks is possible because of the relative magnitudes of the associated 1 Jppt values, 'jppt values for P-atoms trans to CI are typically on the order of 3500 Hz, e.g., 3490 Hz for PtCl2(dpypcp) and 3480 Hz for PtCl2(dpype) (Sections 4.7.7 and 4.7.10, respectively), and are lower for P trans to N , e.g., 3200 Hz for [Pt2(dpype)2]4+ (Section 4.7.21). Thus, the peak at 5 18.9 (Vppt = 3490 Hz) is due to P A (trans to CI) while that at § 28.6 ('Jppt = 3400 Hz) is due to P B (trans to N). The structure of [PtCl(P,P',/V-dmapcp)]Cl-CH2Cl2T.46 H 2 0 was determined by X-ray crystallography (Figure 2.4) using crystals prepared by Meessen, and selected bond distances and angles are given in Table 2.3. The metal coordination sphere is approximately square planar, and in agreement with the solution N M R data, the ligand adopts a P,P '^-configuration. The absolute configurations of P(l) and the methine C-atoms C(l) and C(2) are R,S,S, respectively, and the 5-membered chelate ring is in the 8-configuration. As the space group is PI and Z = 2, the mirror image is present in the unit cell, but the diastereomeric forms R,R,R and S,S,S are absent. This is consistent with the fact that the substituents on the P-atoms of a chiral C2-symmetric ligand are presented as pseudo-axial and pseudo-equatorial pairs, and the hypothesis that only the equatorial of these are sterically available for binding to a square planar metal centre. Thus, the R,R,R and S,S,S forms of [PtCl(P,P',A^-dmapcp)]+ cannot arise. In order to show this, a 2D l H E X S Y experiment was conducted (Figure 2.5). 33 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Figure 2.4 ORTEP representation (50% ellipsoids) of [PtCl(P,P ',/V-dmapcp)]+. H-atoms have been omitted for clarity. Table 2.3 Selected bond distances (A) and angles (°) for [PtCl(P,P',/V-dmapcp)]Cl-CH2Cl2-1.46 H 2 0 with estimated standard deviations given in parentheses. Pt(l)--P( l ) 2.2012(12) Pt(l)--P(2) 2.2646(10) Pt(l)-- N ( l ) 2.184(3) Pt(l)-- C l ( l ) 2.3642(13) P ( l ) - -C(l) 1.822(4) P ( l ) - -C(6) 1.782(4) P ( l ) - -C(14) 1.814(5) P(2)--C(2) 1.848(4) P(2)--C(22) 1.819(5) P(2)--C(30) 1.829(4) N ( l ) - -C(7) 1.481(6) N ( l ) - -C(12) 1.490(6) N ( l ) - -C(13) 1.496(6) N(2)- -C(15) 1.452(6) N(2)- -C(20) 1.500(6) N(2)- -C(21) 1.449(6) P ( l ) - -Pt( l)--N(l) 84.07(11) P(2)--Pt ( l ) - - c i ( i ) 94.08(4) P ( l ) - -Pt( l)--P(2) 87.43(4) Cl( l ) --Pt ( l ) -- N ( l ) 94.53(11) C ( l ) - - P ( l ) - -C(6) 112.1(2) C ( l ) - - P ( l ) - -C(14) 109.1(2) C(6)- - P ( l ) - -C(14) 115.9(2) C(12> - N ( l > -C(13) 109.8(3) C(7)- - N ( l ) - -C(12) 108.3(4) C(7)- - N ( l ) - -C(13) 109.2(4) C(20> -N(2)--C(15) 114.6(4) C(20> -N(2> -C(21) 109.5(4) C(15> -N(2> -C(21) 109.9(4) 34 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes e,f b e,f 4, ' 1 M 0 m 0 ... W -2.5 h3.0 3.5 ppm 1 ppm 3.5 3.0 2.5 Figure 2.5 The 5 2-4 range of the 2D ] H E X S Y (300 MHz, CDC1 3 , 300 K) spectrum of [PtCl(i> P ',7V-dmapcp)]Cl. If the "pendant" and bound N-atoms are in free chemical exchange, the stereogenic P-atoms should be racemized. If, however, only pseudo-equatorial anilinyl rings can bind the square planar metal centre, then methyl groups a and b cannot be in exchange with c and d, i.e., with methyl groups associated with the other (pseudo-axial) anilinyl group attached to the same (bound) P-atom. Rather, a and b can only be in exchange with the NC//3 protons of the corresponding pseudo-equatorial anilinyl ring on the other P-atom. This is clearly demonstrated in Figure 2.5: a and b are in chemical 35 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes exchange with g,h and not with c,d or e,f. Instead, c,d and e,f are in exchange with one another. If it is assumed that an associative mechanism is more likely than a dissociative one for a d square planar metal centre, these data can be interpreted in light of the exchange pathways illustrated in Scheme 2.5 (for clarity, "bonding" anilinyl rings have been simplified to ethyl chains, while "pendant" rings have been omitted altogether. The single positive charge on each ion has also been omitted.) Scheme 2.5 Only one diastereomer of the postulated trigonal bipyramidal P,P',N,N' intermediate can lead to the exchange of P-atoms. Inspection of the crystal structure (Figure 2.4) shows no obvious reason why the "non-productive" diastereomer (I) should not also form, however. The broad appearance of the peak due to e,f is tentatively attributed to the rapid equilibrium formation of this intermediate, i.e., to the reversible binding of N(3) in Figure 2.4. The aqueous conductivity of [PtCl(P,P',/V-dmapcp)]Cl is 150 ohm"1 mol"1 cm 2, consistent with a 1:1 conductor, i.e., the second chloride ligand does not dissociate. In 36 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes addition, the CI" counter-ion is easily exchanged for PF6- by reaction with NH4PF6 in acetone. Even in the presence of a vast excess of the ammonium salt, only the ionic chloride is replaced under these conditions. 2.4.1.2 Palladium The ' H N M R spectrum of [PdCl(P,P',#-dmapcp)]Cl is similar to that of [PtCl(dmapcp)]Cl, and shows 5 chemically inequivalent NC//3 peaks in the 1:1:2:2:2 integration ratio expected for a P,P \N coordination mode of dmapcp (Section 2.4.1.1). In addition, the crystal structure (Figure 2.6) reveals a cation very similar to [PtCl(dmapcp)]+ (Figure 2.4). However, the 3 1 P{'H} spectrum shows two closely-separated singlets at 5 48.7 and 49.6 instead of the anticipated A X pattern. The 3 1 P{'H} spectra of [MCl(i',P',7V-dmapcp)]Cl (M = Pt, Pd) are shown for comparison in Figure 2.7. Figure 2.6 PLUTO representation of the molecular structure of [PdCl(P,P',7V-dmapcp)]+ H-atoms are omitted for clarity. 37 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Pd Figure 2.7 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) spectra of [MC\(P,P',N-dmapcp)]Cl (M = Pt, Pd). Peaks due to 3 1 P- 1 9 5 Pt coupling fall outside the range given for the Pt spectrum. The 2D ' H E X S Y spectrum of PdCl2(dmapcp) is similar to that of the Pt analogue, pointing to the same type of exchange between coordinated and free anilinyl N -• - a -I -I atoms. In light of these data, the P{ H} N M R spectrum is difficult to interpret: it suggests that the P-atoms are rendered chemically equivalent by the exchange. Rapid exchange between P-atoms whose pendant N-containing "arms" are either coordinated or free has been observed by Stelzer and coworkers for the compound [PdCl(P,/3 ,,/V-py(CH2)2P(Me)(CH2)3P(Me)(CH2)2py)]Cl (py = o-pyridyl), which gives rise to two 3 1 P{'H} singlets at 5 13.7 and 15.5.2 9 In the same manner that [PtC\(P,P',N-dmapcp)]+ gives the expected A X pattern, the analogous Pt complex of Stelzer gives two well separated signals for the chemically inequivalent P-atoms at 5 -6.2 and —12.9. Reaction of [PdCl(P,P',iV-dmapcp)]Cl with an excess of the mild halide-abstracting reagent NH4PF6 in acetone gives [PdCl(P,P',./V-dmapcp)][PF6]. In water, • 1 2 1 however, the molar conductivity is 200 ohm" cm mol" which perhaps indicates that the second chloride dissociates to give [Pd(OH2)(P,P',/V-dmapcp)]Cl2. 38 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.4.2 Dmape complexes 2.4.2.1 Solution equilibrium The r.t. 3 1 P{'H} N M R spectrum of PtCl2(dmape) consists of several very broad, i l l -defined peaks in the range 15-70 ppm. At less than ca. 250 K , the spectrum resolves into 3 singlets and associated Pt "satellites" (Figure 2.8). The peak at 5 46 is due to PtCl2(P,P '-dmape) and those at 8 32 and 53 correspond to the chemically inequivalent P-atoms of [PtCl(P,P')/V"-dmape)]Cl. # Figure 2.8 The 3 1 P{ ] H} N M R (121 MHz, CD 2 C1 2 , 240 K) spectrum of PtCl2(dmape). The * symbol designates peaks due to PtCl2(P,P'-dmape) and # denotes those due to [PtCl(P,P,/V -dmape)]CI. The peak assignments were confirmed by the following experiment. PtCl2(dmape) was reacted with N H 4 P F 6 in acetone to give [PtCl(P,P',/V-dmape)][PF6]. Elemental analysis confirmed that even though a large excess of the halide abstracting reagent was used, only one chloride ligand was removed. The r.t. 3 1 P{'H} N M R spectrum of this complex showed singlets at 8 31.6 and 52.7 which correspond closely to two of the three peaks observed in the low temperature spectra of PtCi2(dmape). As for [PtCl(P,P',iV-dmapcp)]+, 2-bond PP coupling was not observed. The remaining singlet must 39 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes correspond to a C2-symmetric complex which could be either FtChiP.P -dmape) or [Pt(P,P',./V,Af'-dmape)]Ci2, but the failure to generate the latter compound synthetically rules it out. The Pd analogue PdC^Cdmape) exhibited analogous behaviour. Its r.t. CD2CI2 solution 3 1 P { ' H } N M R spectrum showed a single broad peak centred at ca. 66 ppm. At lower temperatures (less than ca. 250 K), this peak diminished and two new singlets at 8 60.5 and 73.4 appeared. Reaction of PdCl2(dmape) with excess NH4PF6 in acetone gave [PdCl(P,P',/V-dmape)][PF6] with two 3 1 P{'H} singlets at 8 60.5 and 73.4. The 8 66.4 singlet is due to the chemically equivalent P-atoms of PdCi2(P,P '-dmape) while those at 8 60.5 and 73.4 belong to the inequivalent P-atoms of [PdCl(P,P',/V-dmape)]Cl ( 2 J P P is unresolved for this compound). Varying the temperature at which the spectra of MCl2(dmape) (M = Pt, Pd) were acquired reversibly altered the relative peak intensities showing that MCi2(P,P -dmape) and [MCl(P,P',/V-dmape)]Cl are in thermal equilibrium in CH2CI2 solution. With use of the protocol given in Section 2.9.6, K values for the equilibrium shown in Chart 2.10 were determined at 210-250 K by 3 1 P{'H} N M R spectroscopy; the resulting Van't Hoff plots appear in Figure 2.9. 40 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes -4 -I 1 1 , , 1 3.8e-3 4.0e-3 4.2e-3 4.4e-3 4.6e-3 4.8e-3 1/T(K_1) Figure 2.9 Van't Hoff plot for the MCl 2(P,P'-dmape) [MCl(P,P',AMmape)]Cl equilibrium. The determined thermodynamic parameters are AH° = -19 ± 4 kJ mol"1 and AS° = -100 ± 3 0 J mol"1 K" 1 for PtCl2(dmape), and AH° = -11 ± 7 kJ mol"1 and AS° = -60 ± 20 J mol"1 K" 1 for PdCl2(dmape). The entropy decrease must be the result of solvent ordering about the charged species, and the exofhermicity is due predominantly to the formation of strong M — N bonds (vs. M—CI) with that for Pt being stronger than for Pd. The large error values result from less than ideal reproducibility between separate measurements. 2.4.2.2 Structure The structure of PtCl2(dmape)-CH2Cl2 was determined by X-ray crystallography using crystals prepared by Meessen (Figure 2.10); selected bond distances and angles are given in Table 2.4. 41 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes C(14) C(15) Figure 2.10 ORTEP representation (50 % ellipsoids) of the molecular structure of PtCl2(dmape). Except for those in the "backbone" of the ligand, H-atoms have been omitted for clarity. Table 2.4 Selected bond distances (A) and angles (°) for PtCl 2(dmape)-CH 2Cl 2 with estimated standard deviations given in parentheses. Pt(l)--P( l ) 2.2317(13) Pt(l)--P(2) 2.2355(13) Pt(l)-- C l ( l ) 2.3728(12) Pt(l)--Cl(2) 2.3776(13) P ( l ) - -C(l) 1.8819(5) P ( l ) - -C(3) 1.810(4) P ( l ) - -C( l l ) 1.818(7) P(2)--C(2) 1.833(5) P(2)--C(19) 1.844(7) P(2)--C(27) 1.813(4) P ( l ) - -Pt(l)--Cl(2) 92.74(5) P(2)--Pt ( l ) --Cl( l ) 91.25(5) P ( l ) - -Pt(l)--P(2) 86.07(5) Cl(l)--Pt ( l ) --C l (2 ) 90.99(5) C ( l ) - - P ( l ) - -C(3) 109.2(2) C(3)-- P ( l ) - -C( l l ) 105.5(3) C( l l> - P ( l ) - - C ( l ) 109.2(3) C(2)--P (2 ) --C(19) 106.2(3) C(19> -P(2)--C(27) 105.7(3) C(2)--P (2 ) --C(27) 110.2(2) 42 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.5 Organometallic Pd Anilinyldiphosphine Intermediates In contrast to the reaction between dmapcp and fra«s-PdCl2(PhCN)2 which gives [PdCl(P,P,N'-dmapcp)]CI cleanly, the ligand reacts with PdCi2(cod) to give not only this product, but also another whose 3 1 P{ 1 H} N M R spectrum is characterised by an A X pattern (5A 36.5 (d), 5 X 47.8 (d), V P A px = 8.8 Hz) (Figure 2.11). Because of several 31 1 reasons: the P{ H} N M R spectrum depends both on the nature of the diolefin and the halide (Section 2.5.3); the product is a kinetic one (Section 2.5.5) and not the result of a "back-reaction" (Section 2.5.4); and finally, *H N M R peaks characteristic of coordinated olefin are seen (Section 2.5.6), the second product is proposed to be an organometallic intermediate bearing both the diolefin and the anilinyldiphosphine ligand. Based on these observations, on the limiting coordination number of 5 for Pd1 1, and on the known susceptibility of Pd-coordinated olefins to nucleophilic attack (Section 2.5.2), the structure of the intermediate is proposed to be that depicted in Chart 2.11 (three of the anilinyl rings have been omitted for clarity.) 1 1 1 1 1 1 1 1 1 1 " 50 i i | i i i > I i i i i | i i i i | i 45 I i i i i I i i i i | i i i i | i i i 40 i | i i i i | i i i i | i i i i 35 ppm F i g u r e 2.11 The 3 1 P{ ] H} N M R (121 MHz, CDC1 3, 300 K) spectrum of the isolated product mixture from the reaction between dmapcp and PdCl2(cod). The * symbol denotes peaks due to [PdCl(P,JP',7V'-dmapcp)]Cl. 43 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Chart 2.11 CI C l -2.5.1 An introduction to 5-coordinate Pt and Pd complexes Most 5-coordinate complexes of Pt and Pd are trigonal bipyramidal (rather than square pyramidal) and have the general formula M(N-N)(olefm)(X)(Y), 3 0 where N - N is a bulky bidentate N-ligand such as 2,9-dimethyl-1,10-phenanthroline (dmphen), and X and Y are either two anions (usually halides or alkyl groups) or one anion and one neutral ligand. Conditions favouring the formation of trigonal bipyramidal over square planar complexes of the group 10 metals are summarised in the following statements:31 (1) The strongest it-acceptor ligand (the olefin) lies in the equatorial plane where the greatest degree of back-bonding is expected. (2) The most hindered ligand (typically the chelating N - N ligand) occupies the two remaining equatorial sites so as to minimise its steric interactions with other ligands in the coordination sphere. Axial-equatorial dispositions of this ligand, though rare, are also known, e.g., in PtMe(dmphen)(E'-NCCH=CHCN)(CH(C0 2 Me) 2 ) . 3 1 (3) The strongest donor ligands occupy the axial positions. Of particular relevance to this work is the reaction shown in Scheme 2.6. 3 2 The Pt reactant is the product of the reaction between PtCl2(cod) and NaOMe, 3 3 " 3 5 while the product is very similar to that shown in Chart 2.11. The olefinic moiety is "held fast" because it is part of a chelating n^n 2 modified cod ligand. 44 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes The complex shown in Chart 2.11 may represent a new class of complexes. To date, the only known 5-coordinate, Pdn-phosphine-olefin complexes have the general formula [Pd(Cp)(olefin)(PR3)]+.36>37 Scheme 2.6 2.5.2 Nucleophilic attack on coordinated olefin It has been known since the 1930s that coordination of olefins to platinum group metals renders them susceptible to nucleophilic attack.38 Of particular relevance to this thesis is attack by amines3 9 and phosphines.40'41 However, there have been no reports of the particular type of reaction described here, i.e., nucleophilic attack by one donor atom of a polydentate ligand that gives rise to a complex in which at least one of the donors is bound to the metal and one has reacted with the olefin (in this case to generate an ammonium species). The 5-coordinate organometallic intermediate proposed in the reaction between dmapcp and PdCl2(cod) results from nucleophilic attack by an anilinyl N-atom at one of the coordinated olefinic C-atoms, causing a slip in the diolefin coordination mode from n 2 ,n 2 to n'/n 2; this prevents displacement of the C% moiety. Whether coordination of the metal by P precedes the nucleophilic attack at the bound olefin is unknown. The presence of multiple P- and N-atoms seems necessary for the formation of the 5-coordinate complex: for example, an in situ reaction of PdCl2(cod) with P M A gave the simple substitution product, PdCl2(PMA), and no intermediates were detected (PMA = Ph2P-o-C 6 H 4 NMe 2 ) . CI MeO 45 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.5.3 Variation of diolefin and halide Table 2.5 summarises the 3 1 P{'H} N M R data for the reaction shown in Scheme 2.7. The data for pure [PdCl(P,JP',/V-dmapcp)]Cl dissolved in CDC1 3 and CD 2 C1 2 are included for reference purposes. Scheme 2.7 PdX2(diolefm) + dmapcp ->• [PdX(P,PJV-dmapcp)]X (a) + b Table 2.5 Summary of 3 1 P{ ! H} (121 MHz, 300 K) data for the series of reactions shown in Scheme 2.7. 3 1 P{'H} N M R chemical shift (ppm) [2JpP (Hz) ] Reaction Solvent diolefin X aa bb 1 CDC1 3 cod CI 48.4 48.9 36.8 [10.3] 47.9 [10.3] 2 CDCI3 nbd CI 48.4 49.0 33.1 [13.0] 49.6 [13.0] 3 CDCI3 dcp CI 48.4 48.9 — 4 CD 2 C1 2 cod Br 47.0 50.0 34.7 [10.0] 47.5 [10.0] 5 CD 2 C1 2 cod I 39.9 50.4 — Compound Dmapcp CDCI3 -25.6 [PdCl(P,P ',7V-dmapcp)]Cl CDCI3 48.4 48.9 [PdCl(P,P ',/V-dmapcp)]Cl CD 2 C1 2 48.6 48.8 a Both peaks due to a are singlets. b Both peaks due to b are doublets. The downfield 3 l P{ l H} shifts given in Table 2.5 indicate that both P-atoms are bound to Pd. Peaks due to a are necessarily independent of the diolefin. Changing, however, from CI to Br to I causes the downfield singlet due to a to shift from 5 48.9 to 8 50.4, and the upfield singlet to shift from 5 48.4 to 5 39.9. Table 2.5 clearly shows that both the diolefin and the halide are implicated in the second product b: variation of either results in variation of the 3 1 P{ 1 H} shifts. The r.t. reaction between PdI2(cod) and dmapcp did not produce b in detectable concentration. An independent synthesis of [PdI(P,P',./V-dmapcp)]I (Section 2.9.2.7) via 46 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes halide metathesis of [PdCl(P,P',7V-dmapcp)]Cl with Nal gave a compound whose 3 1 P{ 1 H} N M R spectrum was identical to that of the product of reaction 5 in Table 2.5. The reaction of dmapcp with PdCl2(dcp) did not produce b, either (reaction 3); an in situ N M R investigation revealed that [PdCl(P,P',7^-dmapcp)]Cl was the sole product at r.t. These results indicate that the nature of the halide and olefin dramatically affect the stability of the 5-coordinate intermediate. -i i i Reaction 2 was monitored by P{ H} N M R spectroscopy over the course of 10 d at r.t. Twenty minutes after the reagents had been mixed, integration revealed the mixture to be approximately 60 % b, 40 % a. After 2 d, the mixture contained ca. 52 % b, 48 % a and after 10 d, this had dropped to 45 % b, 55 % a. In order to ensure that all of b converts to a and that no decomposition occurs, this experiment needs to be repeated in the presence of an inert internal concentration standard. 2.5.4 Reaction between diolefin and [PdCl(P,P',7V-dmapcp)]Cl In situ N M R measurements revealed that [PdCl(P,P',/V-dmapcp)]Cl did not react with 100-fold excesses of cod, nbd or hxd, demonstrating that the diolefin-containing product b is a true intermediate en route to the formation of [PdCl(P,P',/V-dmapcp)]Cl and not the result of a back-reaction between the product and the displaced diolefin. 2.5.5 Low temperature in situ reaction between PdCl2(cod) and dmapcp The reaction between dmapcp and PdCl2(cod) was followed by low temperature 3 1 P{'H} N M R spectroscopy. In this experiment, CDCI3 was added by vacuum transfer to an N M R tube containing the solid reagents held at 77 K. The solvent was allowed to melt at 230 K within the spectrometer probe. Within the time taken to make the first measurement (6 min), the intermediate b had begun to form. It was not until after about 1 h that peaks due to [PdCl(P,P',A^-dmapcp)]Cl appeared. At this point, the ligand had almost completely been consumed. 2.5.6 *H NMR spectroscopy The ' H N M R spectrum of the isolated product mixture from the reaction of PdCl2(cod) and dmapcp with or without added NH4PF6 shows several peaks in addition to those due to [PdCl(P,P',7V-dmapcp)]+ (see Section 2.9.4.1 for full ' H N M R data for the product of the reaction in the presence of NH4PF6). Most prominent are a pair of pseudoquartets that 47 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes appear at 5 5.23 and 5.65; by virtue of their integrations and chemical shifts, these can be assigned to 2 inequivalent olefinic protons of a coordinated CsHi 2 moiety (Figure 2.12). (For useful reference, the olefinic protons of PdCl(Me)(cod) give rise to multiplets at 8 5.15 and 5.90,4 2 while those of [Pd(u-Cl)(MeO-cod)]2 appear at 8 5.45 and 5.90.43) ' h ' H COSY measurements reveal that each of these peaks is coupled to two others in the aliphatic region of the spectrum, consistent with the presence of a saturated organic component. In all, 9 protons of the anticipated 12 could be clearly identified with the aid of the COSY spectrum. A multiplet at high field (8 1.05) is tentatively assigned to the proton attached to the Pd-bound C-atom. This proton is strongly coupled to another at much lower field (8 4.15) which is therefore assigned as proton bound to the C-atom which has undergone the postulated nucleophilic attack by N . In addition, the NCH3 region shows an "extra" 4 singlets at 8 2.90, 2.85, 2.21 and 2.18 in a 1:2:4:1 ratio, consistent with the number and type of NC//3 protons illustrated in Chart 2.11. Presumably, the two 1:1 singlets are due to the methyl groups bonded to the N-atom responsible for the nucleophilic attack. Because these are well separated, it is concluded that the N-atom in b is a different type from that in the thermodynamic product [PdClCP.P ',/V-dmapcp)]Cl. 11 I I 1 I I I I 1 1 I I I I I I I I I I I I I I I I I I I I 1 I I I I I I I I I I I I I I I I I I I I J I I I I I I I I I I I 6.0 5.a S.6 5.4 5.2 5.0 4.8 Figure 2.12 The 8 4.8-6.0 region of the ' H N M R (121 MHz, CDC1 3 , 300 K) spectrum of the isolated product mixture from the r.t. reaction between PdCl2(cod) and dmapcp. 2.5.7 Comparison between the isolated product mixtures from low and r.t. reactions Section 2.9.4.1 describes the isolation of product mixtures from the reactions of dmapcp with PdCl2(cod) in the presence of the mild halide-abstracting reagent N H 4 P F 6 at r.t. and 48 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes -40 °C. The ^ P l ' H } N M R spectra of these are shown in Figure 2.13; spectral data appear in Table 2.6. 1 1 1 ' | I' | I' M | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I 50 40 30 ppm Figure 2.13 31twH 'P{'H} N M R (121 MHz, CDCI3, 300 K) of the isolated product mixtures from the reaction between PdCl2(cod) and dmapcp at (i) r.t. and (ii) -40 °C recorded 0.5 h after making the samples. Spectrum (iii) shows the evolution of (ii) after 6 h at r.t. Peaks marked by asterisks are due to [PdCl(P,P',/V-dmapcp)][PF6]. Other assignments are discussed in the text. Table 2.6 Spectrum Summary of the 3 1 P{'H} N M R data for the spectra shown in Figure 2.13, excluding peaks due to [PdCl(P,P',/V-dmapcp)][PF6]. ~ J I P{ 'H} N M R chemical shift (ppm) [ V P P (Hz) ] a b c d e (i) 36.8 [10.3] — — — 47.9 [10.3] (ii) 33.8 [111] 38.6 [111] — — (iii) 36.8 [10.3] 33.8 [111] 24.7 [85.5] 26.4 [102] 47.9 [10.3] 38.6 [111] 50.3 [85.5] 41.4 [102] a A l l peaks due to b-e are doublets. 49 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes As spectrum (i) clearly demonstrates, the single organometallic product b resulting from the r.t. reaction in the presence of the halide-abstracting reagent is identical to the kinetic product of the -40 °C reaction in its absence (Section 2.5.5). The low temperature reaction in the presence of NH4PF6, however, produces a different result. Spectrum (ii) shows the two peaks due to [PdCl(P,-P',/V-dmapcp)][PF6] and an A X pattern due to another product, c. After 6 h at r.t., 5 species are seen: [PdC\(P,P',N-dmapcp)][PF6], b, c, and new complexes d and e (spectrum (iii)); b-e all show A X patterns. The most significant difference between b and c-e is that b has a small JPp value of 10.3 Hz, while c-e have corresponding values in the range 85-111 Hz. The small coupling constant observed for b is consistent with a cis disposition of P-atoms coordinated to Pd or Pt for the anilinyldiphosphine ligand family, e.g., 2JPP = 12.3 Hz for PtCl(P,P',/v~-dmapcp)]Cl (Section 2.9.2.1) and is too small to be observed for [PtCl(P,P',/V-dmape)][PF6] (Section 2.9.2.4), [PdC\(P,P',N-L)][PF6] (L = dmape (Section 2.9.2.10), or dmapcp (Section 2.9.2.8)). This, together with the NMe "fingerprint" in the *H N M R spectrum of b (Section 2.5.6), is consistent with the structure given in Chart 2.11, i.e., a structure in which both P-atoms are bound to Pd and one of the anilinyl N -atoms acts as a nucleophile to attack the originally coordinated cod. An alternate view has the anilinyldiphosphine ligand P, TV-bound to the metal and one of the P-atoms acting as the nucleophile. In this case, 3-bond PP coupling should be observed. The only comparison available for the anilinyldiphosphine ligands are the complexes [PdCl(dmapeO)]2[PF6]2 ( J?? = 66.2 Hz; see Section 2.6.1 for discussion and Section 2.9.2.12 for synthesis) and PdCl(Me)(/>,7v'-dmape) ( 3 J P P = 44.1 Hz; from the in situ reaction between PdCl(Me)(cod) and dmape). Although these couplings are on the order of lA of magnitude of those observed for c-e, they are much larger than observed for b, and may indicate structures of the type shown in Chart 2.12 (anilinyl rings not involved in bonding have been omitted for clarity, and the counterion in each case is PF<f). 50 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Chart 2.12 The structures in Chart 2.12 are depicted as unipositive ions. If these complexes were dipositive, i.e., chloride free, they could not serve as intermediates en route to the formation of [PdCl(P,P',/V-dmapcp)]+ which is the ultimate product at r.t. Clearly, the presence of a halide-abstracting reagent significantly impacts the path of the reaction at low temperature. However, the specific role of NH4PF6 is not understood, and positive assignments of c-e cannot be made. 2.5.8 Variation of the anilinyldiphosphine ligand The r.t. reaction between the CH2-bridged anilinyldiphosphine ligand dmapm and PdCl2(cod) produces not one but two olefin-bearing intermediates en route to the formation of the intended product, PdCl2(dmapm). The 3 1 P{'H} N M R spectrum of an in situ reaction given by Scheme 2.8 (X = CI) is shown in Figure 2.14; PdCl2(dmapm) exists in organic solution as an equilibrium mixture of PdCbCPP '-dmapm) and ~PdC\i{P,N-dmapm) (see Section 3.3.1). Scheme 2.8 PdX2(diolefm) + dmapm ->• PdX2(P,P'-dmapm) (f) + PdX2(P,7v~-dmapm) (g) + h + i 51 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes # II I I I I I I I I I I I I I I II I I I I I I I I I I I I I M I I I I I I I I I I I I j I I I I I II I I I I I I I I I I I I I I I I I I I I I I I I I! I I I I I I I I I I II I I I I I I I I I I I!I I I I I I I I I I I 50 40 30 20 10 0 -10 -20 -30 -40 -50 ppm Figure 2.14 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) spectrum of the in situ reaction between dmapm and PdCl2(cod). The * symbol denotes peaks due to PdCl2(P,/V-dmapm) and # identifies that due to PdCl 2(P,P '-dmapm). Other labels are discussed in the text. Table 2.7 Summary of 3]?{:H} (121 MHz, 300 K) data for a series of reactions of the type shown in Scheme 2.8.a J 1 P{'H} N M R chemical shift (ppm) [%? (Hz) ] Reaction diolefin X f gc hc i c 1 cod CI -56.8 -40.0 [108] 18.8 [83.3] 19.6 [79.4] 33.5 [108] 46.4 [83.3] 49.0 [79.4] 2 d nbd CI -56.8 -40.0 [108] 16.3 [13:7] -33.5 [108] 33.8 [13.7] 3 cod Br -58.5 -40.4 [112] 16.2 [84.5] 17.8 [80.6] 32.4 [112] 46.3 [84.5] 49.4 [80.6] Compound -36.0 dmapm PdCl2(P,/V-dmapm)e - -40.0 [108] - -33.5 [108] PdCl2(P,P'-dmapm) e -56.8 - - -a Measured in CDCI3 unless otherwise noted. b Singlets. 0 Doublets. d Measured in CD 2 C1 2 . eSee Section 3.3.1. Complexes containing P-atoms involved 4-membered metallacycles (such as f) give rise to 3 1 P{'H} N M R singlets in the high-field region, 4 4 e.g., 8 -56.8 for PdCl 2(P,P'-52 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes dmapm). As peaks due to h and i are dramatically shifted from that of free dmapm, fall in the low-field region, and are well separated from each other, one P-atom of the ligand is likely involved in a 5-membered P,/V-chelate with P d 4 4 and the other probably participates in another type of interaction. Comparing reactions 1 and 2 in Table 2.7, it is evident that this second interaction (manifested by the doublets at 8 46.4 and 33.8, respectively) is very sensitive to the nature of the diolefin. Consistent with these findings, and those for the reaction of PdCl2(cod) with dmapcp, are the structures shown in Chart 2.13 (3 anilinyl rings have been omitted for clarity in each case). No concrete assignment of the isomers h and i is possible, however. These structures differ from that proposed for intermediate b (Section 2.5.3) in that a P-atom rather than an N-atom attacks the coordinated olefin. This alleviates the 4-membered ring strain that would result i f the N-atom were the nucleophile (Chart 2.11). Moreover, there are no known 5-coordinate Pt/Pd"-olefm complexes containing 4-membered chelates. Chart 2.13 + + CI- C l -The distribution of products for reaction 1 (Table 2.7), for example, varies with time. Within the time taken to make the first N M R measurement, isomer i dominates. After 48 h, h is more abundant than i, and peaks due to PdCl2(.P,./V-dmapm) (f) and VdCh(P,P -dmapm) (g) are also marginally larger than they were initially. Shown in Figure 2.15 are the ] H and 'H{ 3 1P} N M R spectra of the mixture isolated from the reaction of PdCl2(cod) and dmapm; this spectrum is more complicated in the 8 53 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 4.0-6.0 region than that of PdCi2(cod)(dmapcp) because of the presence of not only olefinic but also methylene peaks due to the PGff 2P protons. I i i i i | i i i i | i i i i | i i i i | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i - ] — i — i — i — i — | — i — i — i — i — 6.0 5.8 5.6 5.4 52 5.0 4.S 4.6 4.4 4.2 ppm Figure 2.15 'H{ 3 1 P} (top) and ' H (bottom) N M R (300 MHz, CDC1 3 , 300 K) spectra of the product mixture from the reaction between PdCl2(cod) and dmapm. Peak assignments are discussed in the text. The peaks labelled "o" correspond to the protons due to the coordinated olefinic moiety. Those with the prefix "m" correspond to the methylene protons in the "backbone" of the dmapm ligand. In each case, there are two sets reflecting the organometallic intermediates h and i. The olefinic protons o(h) and o(i) are represented by pseudo-quartets in both the ] H and ] H { 3 1 P } spectra, i.e., these protons are not strongly coupled to P. The chemical shifts of o(h) and o(i) are in the range typical for 5-coordinate alkene complexes of Pd" 32,45 ' h ' H COSY measurements show each of the pair of o(h) and o(i) peaks to be coupled to one another and also to be coupled to several peaks in the aliphatic region of the spectrum (1-4 ppm), consistent with assignment of these to coordinated olefin. The methylene protons m(h) and m(i) are represented by doublets in the 'H{ 3 1P} spectrum and their assignment is unequivocal: each proton is coupled only to one other 54 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes proton (the large magnitude, 19 Hz, identifies it as geminal coupling) and to two inequivalent P-atoms. This gives rise to a ddd pattern for each of the m(h) and m(i) peaks in the absence of the P-decoupler. The multiplet at 5 4.15 which collapses to a pseudo-triplet on P-decoupling is tentatively assigned to the proton bound to the C-atom which has undergone nucleophilic attack. This proton should be coupled to 2 inequivalent P-atoms via V-C-H and P - C - P -H connectivities. The corresponding "formerly olefinic" proton, which is bound to the C-atom bound to Pd, is identified by its characteristic upfield shift (a multiplet at 5 1.00) and response to the P-decoupler (collapse to a pseudo-triplet). This proton is coupled to 2-inequivalent P-atoms through P-Pd-C-# and V-C-C-H linkages. 2.6 Reactions of PdCl2(dmape) with Bases 2.6.1 Reaction with KOH In an attempt to make anilinyldiphosphine analogues of the well-documented hydroxo-bridged Pd 2 compounds of the type illustrated in Chart 2.14 (left; P = a tertiary phosphine ligand), 4 6- 4 7 PdCl2(dmape) was reacted with K O H according to the protocol outlined in Section 2.9.2.12. Pale yellow single crystals of [PdCl(dmapeO)]2[PF6]2 (Chart 2.14, 6 anilinyl rings have been omitted for clarity) deposited from CDCI3 solution. The ORTEP representation is shown in Figure 2.16, and selected bond distances and angles appear in Table 2.8. Chart 2.14 Me 2 55 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.6.1.1 Structure of [PdCl(dmapeO)]2 C(16) Figure 2.16 ORTEP representation (50 % ellipsoids) of the molecular structure of the [PdCl(dmapeO)]22+ cation. H-atoms are omitted for clarity. Table 2.8 Selected bond distances (A) and angles (°) for [PdCl(dmapeO)]2[PF6]2-4 CDCI3 with estimated standard deviations given in parentheses. Pd(l> - C l ( l ) 2.296(2) Pd(l)-- N ( l ) 2.114(5) Pd(l)-- P ( l ) 2.182(2) Pd(l)-- O ( I A ) 2.124(4) P(2)- -0(1) 1.515(4) Pd(l)--Pd( l*) 4.873(2) P ( l ) - -Pd(l)—N(l) 87.17(14) C l ( l ) --Pd(l)—0(1 A) 95.71(13) Pd(l)-- P ( l ) - C ( 1 7 ) 111.2(2) P ( l ) - -C(17)—C(18) 113.8(5) C(17> -C(18)—P(2) 111.2(5) C(18> - P ( 2 ) - 0 ( l ) 108.7(3) P(2)- -0(1)—Pd(lA) 137.1(3) 56 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes The structure consists of 2 face-to-face Pd" square planes in a head-to-tail (HT) orientation (i.e., they are related by an inversion centre which lies at the mid-point of the Pd-Pd axis). The metal centres are separated by a distance of 4.873 A; each is P,N-chelated by the monooxide derivative of dmape, and P,0-bridged to the other. A 12-membered ring containing the metal centres results, and this is flanked on either side by CDCI3 molecules (not shown in Figure 2.16). The P — C — C angles in the dmapeO ligand "backbone" (113.8 and 111.2 °) are very similar to that found for dmape (110.7 °; Figure 2.1), indicating that the ligand is unstressed and that the Pd centres are too far removed from one another to interact. The Pd—P and Pd—N bond lengths (2.182 and 2.114 A) as well as the P—Pd—N angle (87.17°) are similar to those found for PdCl2(P,/VT-dmapm) (2.178 and 2.132 A, and 86.10 °, respectively; Figure 3.4, p. 94). The P—O bond length (1.515 A) is almost identical to those found recently within this laboratory for coordinated BINAPO in the complexes [RuCl(BINAPO)(N-N)][PF 6] (1.518 A for N - N = bipy and 1.515 A for N - N = phen),48 and is in the range typically found for coordinated R 3 P=0 (1.49-1.52 A).4 9 2.6.1.2 Formation of [PdCl(dmapeO)]22+ The dmapeO ligand represents an example of a bis(phosphine) monooxide (BPMO). Complexes of this ligand class have been investigated since 1996 in hydroformylation,50'51 hydrosilylation,52 hydrovinylation53 and oligomerisation and polymerisation reactions.54 Although BPMOs are now routinely made via metal-free organic synthesis (e.g., by benzylation followed by basic hydrolysis of the resulting phosphonium salt5 5) or with the aid of a Pd catalyst,56 this brief introduction will deal only with those BPMOs which result from reactions of metal-phosphine complexes and, once formed, remain bound to the metal. An excellent and more detailed introduction to the syntheses and uses of BPMOs can be found in the PhD thesis of Cyr . 4 8 While there are several examples of oxidation of "dangling" P-atoms in complexes bearing polydentate ligands (e.g., PdCl2(.P,./V-dmapmO) from PdCl2(dmapm) (Section 3.3.5), and in C o , 5 7 - 5 8 M o , 5 9 R h , 6 0 and other Pd 6 1> 6 2 systems), in only a few cases do coordinated phosphine oxides result.5 8-5 9 Even more rare is the oxidation of a 57 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes coordinated diphosphine to give a BPMO in which both the Pm-atom and the O-atom are bound. Two examples of this are the formation of [(Cp*)Ru(O2)(P,0-dppmO)][BF4] from [(Cp*)Ru(H2)(P,P'-dppm)][BF4] via a multistep process63 and the oxidation of RuCl 2(BINAP)(N-N) ( N - N = bipy, phen) in MeOH to give [RuCl(BINAPO)(N-N)]Cl in which the BINAPO ligand is coordinated not only via the P- and O-atoms but also through 2 C-atoms of one of the naphthyl rings.4 8 The reaction of PdCl2(dmape) with base represents another example of this uncommon occurrence, the reaction again probably requiring multiple steps (vide infra). Alper and coworkers have shown that in the biphasic carbonylation of A r X catalysed by PdCl 2(PR3) 2 in the presence of alkali, the active species is a Pd° phosphine complex generated by the in situ reduction of Pd 1 1 by PR3/OH". 6 4 Indeed, in many Pd-catalysed coupling reactions (e.g., the Heck, 6 5 Stille 6 6 and Suzuki 6 7 reactions, as well as animations,68-69 and diaryl ether formations70), the true Pd° active species is generated in situ by reduction of Pd". In the Heck reaction, for example, the active catalyst is often derived from a mixture of Pd(OAc) 2 and a phosphine ligand. The formation of Pd° from this combination has been studied in detail by Amatore et al.lx The basic principle of Pd 1 1 reduction by PR3/OH" is outlined in Scheme 2.9. Scheme 2.9 LnPd"—PR 3 + OH" -> LnPd0 + OPR 3 + H + The compound [PdCl(dmapeO)]2[PF6]2 probably arises in two steps. First, reduction of PdCl2(dmape) by O H - gives "Pd(dmapeO)" or "Pd(dmapeO)Cr" which then reacts with the chlorinated solvent to generate the Pd" product. Reactions of M°(phosphine) complexes (M = Pd, Pt) with chlorinated solvents to generate the corresponding M 1 1 chloro species are well documented,72 and are also investigated in this thesis (Section 4.4.3.1). In order to test this hypothesis, two concurrent experiments were conducted (Table 2.9). In the first, the reaction was performed as outlined in Section 2.9.2.12, while in the second, both the reduction and analysis of the product were done in the absence of a chlorinated solvent (Section 2.9.4.2). 58 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Table 2.9 Colour changes observed for the reaction between PdCl2(dmape) and K O H in H 2 0 / C H 2 C 1 2 (I) and in H 2 0 / C H 3 N 0 2 (II). Prior to K O H addition Upon K O H addition After 20 h Layer I II I II I II Aqueous Colourless Colourless Colourless Colourless Colourless Red Organic Yellow Yellow Orange Yellow Orange Orange The two experiments differ significantly. Analysis of the organic fraction of experiment II by acetone-d6 solution 3 1 P{ ] H} N M R spectroscopy reveals broad peaks (5 61, 73) of the starting material PdCl2(dmape). The baseline of this spectrum is undulating in the lowfield region, possibly reflecting the presence of fluxional "Pd°(dmapeO)" and/or "Pd°Cl(dmapeO)"" species. (Variable temperature N M R experiments are needed to probe this phenomenon.) Most importantly, the spectrum does not contain peaks attributable to [PdCl(dmapeO)]22+, which supports the hypothesis that this cation results from the reaction of a Pd° species with the chlorinated solvent. 2.6.2 Reaction of PdCl2(dmape) with K 2 C 0 3 , K O ' B u and Pd 2(dba) 3CHCl 3 An attempt was made to make the 0,0'-bonded maleate complex Pd(P,P-dmape)((3,<9'-maleate) because of potential implication of the compound in the Pd-catalysed hydration of water-soluble olefins (see Chapter 5). Thus, an acetone solution containing PdCl2(dmape) and an excess of maleic acid was stirred together with solid K 2 C 0 3 ; this resulted in a colour change from yellow to deep orange. Analysis of the orange product by 3 1 P{'H} spectroscopy revealed that two species were formed: one gave rise to a singlet at 5 35.2 and the other to a pair of doublets (8 70.8 and 86.2) which showed severe second-order distortion. Moreover, the same two products resulted from the reaction of PdCl2(dmape) in the absence of maleic acid (Section 2.9.4.5) which showed that the olefin had not been incorporated. Further investigations showed that not only did these products arise from the reaction of PdCl2(dmape) with K 2 C 0 3 , but that they also resulted from the reaction with KO'Bu (Section 2.9.4.5). Moreover, reaction of trans-PdCl 2(PhCN) 2, Pd2(dba)3-CHC13 and dmape, which was done with the conscious aim of producing dmape-bridged Pd 2 ' compounds, gave the same products (Section 2.9.2.13). These findings suggest that the base is acting to reduce the Pd, and that the products are likely to be Pd 2 ' complexes; those shown in Chart 2.15 would satisfy the 3 1 P{'H} N M R 59 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes data (anilinyl rings not involved in coordination have been omitted for clarity). Unlike in the reaction of PdCl2(dmape) with K O H , phosphine oxidation did not result in the analogous reactions with K2CO3 and KO'Bu, pointing to a different reduction mechanism for these bases. Chart 2.15 2.7 Conclusions An effective synthesis has been established for a family of o-anilinyldiphosphine ligands whose members differ from one another in the number and arrangement of C-atoms in the "bridge" linking the two P-atoms. The nature of this bridge is the determining factor in the mode of coordination of the ligands to Pt" and Pd". The "cyclopentyl-bridged" dmapcp binds in a P,P',N fashion, while the "ethyl-bridged" dmape forms complexes which are in equilibrium between their MX 2(P,P'-dmape) and [MX(P,P',iV-dmape)]X isomers (X = halide). The thermodynamic parameters associated with this equilibrium 31 1 have been evaluated by variable temperature P{ H} N M R spectroscopy to be: AH° = -19 ± 4 kJ mol"1, A S 0 = -100 ± 30 J mol"1 K" 1 for M = Pt, and AH° = 11 ± 7 kJ mol"1, A S 0 = -60 ± 2 0 J mol"1 K" 1 for M = Pd. Two-dimensional *H E X S Y measurements demonstrate that free and bound N-atoms in the complexes [MCl(P,P' )ALdrnapcp)]Cl are in chemical exchange via one of two diastereomeric 5-coordinate P,P',N,N' intermediates. Only enantiomers result from the combination of R,R- or ^S-dmapcp with Pd and Pt, with the absolute configuration at the stereogenic P-centre being opposite to that of the methine C-atoms in the ligand "backbone": exchange does not occur between a coordinated anilinyl ring and the free ring associated with the same, i.e., bound, P-atom. 60 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Water-soluble complexes [MCl(P,P',./V-dmapcp)]Cl result via displacement of halide from the metal coordination sphere. The reactions between dmapcp or dmapm and PdX2(diolefm) (X = halide; diolefin = cod, nbd) result in a mixture of products. The thermodynamic product is [PdX(P,P ,,7v'-dmapcp)]X, and the kinetic products are likely to be trigonal bipyramidal compounds with the general formula [PdX(r|1,ri2-diolefin)(L)]X where L = P,P '-dmapcp or PTV-dmapm. In these compounds, either a P- or an N-atom of the anilinyldiphosphine ligand participates in a nucleophilic attack on the coordinated diolefin. PdCl2(dmape) reacts with K O H in a CH2CI2/H2O mixture to form the bimetallic head-to-tail complex [PdCl(dmapeO)]22+. This compound arises in two steps: reduction of Pd" to Pd° by OH" with concomitant oxidation of phosphine to phosphine oxide, followed by oxidation of the metal by solvent to give the final product. PdCl2(dmape) also reacts with other bases ( K 2 C O 3 , KO'Bu) and with Pd° to give the dmape-bridged bimetallic complex Pd2Cl2(dmape). 2.8 Recommendations for Future Work 2.8.1 Preparation of Pt° and Pd° complexes In contrast to the 1:1 reactions of the pyridyldiphosphine ligands with M(PPh 3 ) 4 (M = Pd, Pd) which quickly generate Pt(PPh3)2(P-P) (P-P = dpype, dpypcp) or Pd(P-P) 2 (Sections 4.5.22-25), the anilinyldiphosphine ligands do not react under the conditions tested, possibly for steric reasons. Other attempts were made during the course of this work to synthesise M° complexes of the anilinyldiphosphine ligands; the most promising involved either reduction of the corresponding M 1 1 complexes using Na/Hg, or photolysis of a M"(oxalate) complex. Sodium amalgam reduction of PtCi2(dmape) (Section 2.9.4.3) produced an orange product whose 3 1 P{'H} N M R spectrum showed the product to bear equivalent P-atoms (8P 38.9, s, 1 Jppt = 3580 Hz). This product remains poorly characterised, but it could be the tetrahedral complex ~Pt(P,P',N,N'-dmape). Further work along the lines of this reaction will, in all likelihood, yield a new class of highly reactive Pt° complexes. 61 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Trogler's group pioneered the use of U V light to effect the reductive elimination of CO2 from Pt and Pd oxalate complexes (Scheme 2.10, Reaction l ) . 7 3 During this thesis work, the analogous reaction with Pt(C204)(P,P'-dmape) was attempted in the presence of dmm (Section 2.9.4.4), the function of the olefin being to trap the reactive, photochemically generated "Pt(dmape)" (Scheme 2.10, reaction 2). Scheme 2.10 Pt(C 2 0 4 )L 2 -> "PtL 2 " + 2 C 0 2 [ 1 ] Pt(C 20 4)(P,P '-dmape) + dmm -> Pt(n2-dmm)(P,P '-dmape) + 2 C 0 2 [2] As judged by the N M R data, the reaction was successful. The 3 1 P{'H} N M R spectrum of the product consists of a singlet at 5 43.8 ('jppt = 3910 Hz), and the ] H spectrum contains peaks which may be attributed to C(0)Gf/3 and CH=CH protons of coordinated dmm. A l l the NC//3 protons in the product are equivalent (giving rise to a singlet at 8 2.50), which demonstrates that there are no coordinated N-atoms. An effective general strategy for the synthesis of M° complexes of the anilinyldiphosphine ligands is a goal worthy of future attention. 2.8.2 Preparation and investigation of allyl complexes In order to determine whether the anilinyldiphosphine ligands would participate in nucleophilic attack on coordinated allyl as they do on coordinated cod, dmape was reacted with [Pd(^-Cl)(n3-allyl)]2 in the presence of N H 4 P F 6 (Section 2.9.2.14). The product of this reaction was [Pd(r|3-allyl)(PP'-dmape)] [PFe], i.e., no nucleophilic attack occurred. The 3 1 P{ ! H} N M R spectrum shows a sharp singlet at 8 44.2 which is invariant from 230-300 K. Sharp peaks, all of which may be assigned, appear in the ] H spectrum. If the reaction is performed in the absence of the halide-abstracting reagent, the 3 1 P peak now appears at 8 28.4, and the *H spectrum is broad and ill-defined. The system is further complicated, i f the anilinyldiphosphine is varied: reaction of [Pd(u.-Cl)(n3-allyl)]2 with T 1 1 dmapm gave 3 products as determined by P{ H} N M R spectroscopy, while reaction 31 with dmapcp in the presence of NH4PF6 gave a compound whose P spectrum was i l l -defined at r.t. and showed 8 peaks in the range 8 -25 to 52 at 220 K. The possibilities of equilibria between n 3 - and n 1 - allyl, coordinated and free CI", and different coordination 62 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes modes of the anilinyldiphosphine ligands, and for nucleophilic attack on the allyl moiety, deserve attention. A recent publication by Braunstein et al. describes an equilibrium between n 3 - and n 1 - allyl at a ~?&(PNP) complex.7 4 2.8.3 Catalytic C - C and C-N bond formations While the Heck reaction7 5 (C-C coupling) catalysed by Pd-dmapm complexes forms the basis of Chapter 3 of this thesis, catalysts based on dmapcp and dmape remain unexplored. The use of Pd-anilinyldiphosphine complexes in catalytic C - N bond forming reactions is also a worthy focal point for future research. Over the last few years, this reaction has been developed primarily by Buchwald and coworkers into a useful synthetic too l . 7 6 - 8 0 The reason for suggesting Pd-anilinyldiphosphine complexes as potential C - X bond forming catalysts is that it has been found, e.g., in the amination of arylbromides by Pd(OAc) 2/BINAP catalysts, that oxidative addition is rate limiting 7 9 and requires the predissociation of BINAP from Pd(BINAP) 2 . 7 7 Presumably, the corresponding "Pd°(dmape)" or "Pd°(dmapcp)" species would not suffer from this slowdown, as the labile "N-arms" should dissociate easily. In preliminary studies, PdCl2(dmape) was used as the catalyst precursor in the amination of PhBr by aniline (Section 2.9.5). At 70 °C in C 6 D 6 solution with KO'Bu as the base and a catalyst loading of 2 mol %, the conversion to Ph 2 NH was 11 % after 3 h (TON = 6). While not impressive, the result does indicate that the complexes here under study have potential as amination catalysts, and further work in this area is warranted. 2.8.4 Ruthenium complexes: novel coordination modes, and reactivity Water-soluble Ru"-phosphine complexes are attractive targets both for the homogeneous catalysis and pharmaceutical industries. In the course of this study, it was found that dmape reacted with RuCl2(PPh3)3 (Section 2.9.3.1) to give cis- and trans-RuCl2(P,P',7V;/V"-dmape) (Chart 2.16, p. 64; arcs terminated by " N " represent o-C6H4NMe2), while dmapcp did not react. High temperature and r.t. reactions give different product distributions: only the cis isomer resulted from the r.t. reaction, while a mixture of the cis and trans (ca. 9:1) was formed at 80 °C. Cz'5-RuCl2(dmape) is freely 1 1 2 soluble in H 2 0 ( A m = 211 ohm" mol" cm ) with the assumed dissociation of both chlorides to give czs-[Ru(0H2)2(P,P',Af./V'-dmape)]Cl2; it is a non-conductor in CH 2 C1 2 . 63 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes The chloro compound reacts with NH4PF6 in acetone to give a mixture of products [Ru(S)„(P2,A^-«-dmape)][PF6]2 (S = acetone, H 2 0 ; n = 1-4). The solvent ligands can be removed by heating the compound in vacuo; the hexa-coordinated product, [Ru(P,P\N,N',N",N" "-dmape)][PF6]2 is very reactive, presumably because of strain built up in the ligand: it reacts in the solid-state with trace H 2 0 in H.P. Ar, and in solution with H 2 S (the solid-state reaction with H 2 S has not yet been attempted). A summary of the P{ H} N M R data for the Ru complexes and for reactions of the halide-free compound with H 2 E (E = O, S) is given in Table 2.10. Table 2.10 3 1 P N M R data for a variety of Ru(dmape) complexes and for the in situ reactions of Ru(P,P',/V,Ar',N'',Ar''-dmape)] [PF 6] 2 with H 2 0 and H 2 S. Compound or Reaction Solvent J1P{'H} chemical shift3 b (ppm) [Vpp (Hz)] cis-RuC\2(P,P', N, N '-dmape) CDCI3 94.6 D 2 0 83. l c trans-RuC\2(P,P ',N,N -dmape) CDCI3 26.0 [Ru(P,P ',N,N',N'\N" -dmape)][PF6]2 CDCI3 95.6 [Ru(P,P ',N,N',N",N'' '-dmape)] [PF 6] 2 + CH 2 C1 2 /H 2 0 74.0, 78.1 [20]d H 2 0 75.3, 88.7 [unresolved]d 79.6, 93.9 [20]d 80.8, 83.1 [24]d 82.7d [Ru(P,P \N,N',N'\N'' '-dmape)] [PF 6] 2 + CDCI3 66.9, 97.7 [13] H 2 S 74.1, 89.7 [14] 77.1 79.5, 95.5 [20] 85.8 a Measured at r.t., and using a 121 MHz spectrometer, unless otherwise indicated. b Single numbers represent singlets; comma-separated values signify pairs of doublets. Where there is more than one product, each appears on a separate line. c Measured using an 81 M H z spectrometer. d Measured using a 202 M H z spectrometer. Unlocked acquisition. e Measured at 220 K. Chart 2.16 64 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes The reactions of [Ru(P,P',Af/V',Ar'',Af ,' ,-dmape)][PF6]2 with H 2 E yield several products because of the possibility for geometrical isomers and, assuming the P-atoms remain coordinated, the binding of up to (theoretically) 4 H2E molecules. In addition, it is not known whether the H2E ligands are deprotonated, as in the case of the reaction between H 2 S and cz's-Ru(H)2(dppm)2 which gives cz's-Ru(SH)2(dppm)2 + 2 H 2 , 8 1 or not deprotonated, as the in the reaction of RuCl2(P,/V-PMA)(PPh3) with H 2 S which gives Ru(SH2)(P,A^-PMA)(PPh 3). 8 2 ' 8 3 Further investigations in this area are worthwhile as there are no known complexes bearing more than one coordinated H2S. In addition, it would be of interest to attempt the reaction between [R\i(P,P',N,N',N",N"'-dmape)][P¥(,]2 and N 2 0 . 8 4 2.9 Experimental 2.9.1 Ligand syntheses 2.9.1.1 l,l-bis(di(o-/V,A/-dimethylanilinyl)phosphino)methane, dmapm To a solution of n-BuLi in hexanes (1.6 mol L" 1, 16.5 mL, 26.4 mmol), cooled to -40 °C on a dry ice/CH 3 CN bath, was added o-bromo-7V>7vr-dimethylaniline (5.05 g, 25.2 mmol) in Et20 (20 mL) via cannula over a period of 20 min. The yellow solution was stirred for 15 min and a white precipitate formed. The slurry was allowed to warm to r.t. and stirring was continued for 1 h. After the solution was cooled again to -40 °C, 1,1-bis(dichlorophosphino)methane (1.33 g, 6.09 mmol) in Et20 (15 mL) was added over 5 min. The resulting orange-red slurry was stirred for 15 min and then allowed to warm to r.t.. Stirring was continued for 1 h and then HC1 (ca. 1 mol L" 1 , 50 mL) was added. The organic layer was removed and K O H (ca. 2 mol L"1) was added dropwise to the aqueous layer until it was neutral. The aqueous fraction was extracted with CH 2 Cl2 (3 x 20 mL) and the combined extracts were dried over MgS04. The solvent was removed in vacuo and EtOH (15 mL) was added. The slurry was refluxed for 0.5 h, cooled to r.t. and filtered to give a white powder that was washed with cold EtOH ( 3 x 5 mL) and dried in vacuo. Yield: 1.84 g (54%). Anal. Calcd for C 3 3 H4 2 N 4 P 2 : C, 71.2; H , 7.6; N , 10.1. Found: C, 71.0; H , 7.8; N , 9.9. ! H N M R (300 MHz, CDC1 3 , 300 K) : 6 2.23 (t, 2H, CH2, 2JHP = 4.2), 2.68 (s, 24H, NC# 3 ), 7.02 (pt, 2H, Ar), 7.11 (pd, 4H, Ar), 7.25 (pt, 4H, Ar), 65 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 7.40 (pd, 4H, Ar). ^ P l ' H } N M R (121 MHz, CDC1 3, 300 K): 5 -36.0 (s). 1 3 C{ ! H} N M R (50 MHz, CDCI3, 300 K) : 8 28.0 (CH 2 , XJC? = 26.2), 45.3 (NCH 3 ), 119.7 (CH), 124.2 (CH), 128.6 (CH), 132.3 (CH), 139.2 (CN), 157.1 (CP). 2.9.1.2 1,2-bis(di(o-7V,ALdimethylanilinyI)phosphino)cyclopentane, dmapcp The synthesis of this compound corresponds to that of dmapm (Section 2.9.1.1). Thus, reaction of rc-BuLi in hexanes (1.6 mol L" 1, 23.0 mL, 36.8 mmol), o-bromo-N,N-dimethylaniline (6.83 g, 34.1 mmol) and l,2-bis(dichlorophosphino)cyclopentane (2.33 g, 8.57 mmol) at -40 °C gave 2.72 g (52 %) of a white powder. Anal. Calcd for C37H48N4P2: C 72.8; H 8.0; N 9.2. Found: C 72.9; H 8.1; N 9.1. *H N M R (300 MHz, CD 2 C1 2 , 300 K): 8 1.51 (2H, m, CH2), 1.69 (2H, m, CH2), 2.23 (2H, m, CH 2 ) , 2.55 (12H, s, N C / / 3 , obscures CH protons), 2.63 (12H, s, NC/ / 3 ) , 6.49 (2H, m, Ar), 6.71 (2H, m, Ar), 6.96-7.28 (12H, m, Ar). 1 3 C{'H} N M R (50 MHz, CDCI3, 300 K): 8 24.7 (CH 2), 29.9 (CH 2), 40.4 (CH), 45.1 (NCH 3), 120.0 (Ar), 124.2 (Ar), 128.6 (Ar), 133.1 (Ar), 137.9 (Ar), 157.7 (Ar). 3 1 P{'H} N M R (121 MHz, CDCI3, 300 K) 8 -25.6 (s). Crystals of dmapcp suitable for X-ray analysis deposited from a CH 2 C1 2 solution containing dmapcp onto which EtOH had been layered. 2.9.1.3 l,2-bis(di(o-7V,A/-dimethylanilinyl)phosphino)ethane, dmape The synthesis of this compound follows that of dmapm (Section 2.9.1.1). Thus, reaction of n-BuLi (1.6 mol L" 1 , 17.4 mL, 27.8 mmol), o-bromo-A^N-dimefhylaniline (5.56 g, 27.8 mmol) and l,2-bis(dichlorophosphino)ethane (1.61 g, 6.94 mmol) gave 2.44 g (61 %) of a white powder. Anal. Calcd for C34H 4 4N 4 P 2 : C, 71.6; H , 7.8; N , 9.8. Found: C, 71.0; H , 7.6; N , 9.5. ] H N M R (300 MHz, CD 2 C1 2 , 300 K): 8 1.86 (t, 4H, CH2,2JHP= 14), 2.62 (s, 24H, NC/ / 3 ) , 6.95 (m, 8H, Ar), 7.11 (m, 4H, Ar), 7.25 (m, 4H, Ar). , 3 C{'H} N M R (50 MHz, CDCI3, 300 K): 8 24.4 (CH 2), 45.3 (NCH 3), 113.1 (Ar), 119.9 (Ar), 128.9 (Ar), 132.5 (Ar), 137.0 (Ar), 157.8 (Ar). 3 1 P{'H} N M R (121 MHz, CD 2 C1 2 , 300 K) 8 -27.5 (s). Crystals of dmape suitable for analysis by X-ray diffraction deposited from a CH 2 C1 2 solution containing dmape onto which EtOH had been layered. 66 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.9.2 Syntheses of Pt and Pd complexes 2.9.2.1 [PtCl(P,P',/V-dmapcp)]Cl-H20 This compound could be made either from PtCl 2 via m-PtCl 2 (MeCN) 2 , or from PtCl2(cod). (a) From PtCh- To a Schlenk tube containing PtCl 2 (100 mg, 0.38 mmol) was added C H 3 C N (10 mL) and the slurry refluxed for 2 h, during which the solid dissolved to give a yellow solution. Dmapcp (230 mg, 0.38 mmol) was then added to the hot solution which became colourless. The solvent was removed in vacuo and the residue was redissolved in the minimum CH 2 C1 2 . Addition of E t 2 0 (20 mL) gave a white powder that was isolated by filtration, washed with E t 2 0 and dried in vacuo at 78 °C. Yield: 240 mg (72 %). (b) From PtCh(cod). To a Schlenk tube charged with PtCl2(cod) (97 mg, 0.26 mmol) and dmapcp (160 mg, 0.26 mmol) was added 1,2-dichloroethane (15 mL), and the resulting colourless solution refluxed for 17 h. The work up was the same as that for the PtCl 2 route. Yield: 150 mg (66 %). The routes gave identical N M R spectra and very similar elemental analyses. Several independent syntheses of [PtCl(P,P',./V-dmapcp)]Cl-H20 analysed as the monohydrate, and ' H N M R spectroscopy confirmed the presence of H 2 0 . Moreover, crystals of this compound contained 1.46 H 2 0 per unit cell (vide infra). Anal. Calcd for C 3 7 H 4 2 N 4 C l 2 P 2 P t H 2 0 : C 49.7; H 5.6; N 6.3; Found: C 50.0; H 5.7; N 6.1. ' H N M R (300 MHz, CDCI3, 300 K): 5 1.30 (m, 1H, CH2), 1.60 (s, 2H, H 2 0) , 1.70 (m, 1H, CH2), 1.90 (m, 2H, CH2), 2.22 (s, 12H, NG/7 3), 2.35 (m, 3H, Ctfand CH2), 2.67 (s, 6H, NC# 3), 3.05 (s, 3H, NCH3), 3.50 (s, 3H, NC//3), 3.72 (m, 1H, CH), 6.70 (m, 2H, Ar), 7.05 (m, 1H, Ar), 7.3-7.9 (m, 11H, Ar), 8.24 (m, 1H, Ar), 9.32 (m, 1H, Ar). 3 1 P{'H} N M R (121 MHz, CDCI3, 300 K) 5 18.9 (d, ]JP?i = 3490,V P P = 12.3), 28.6 (d, V P P t = 3400, V P P = 12.3). A M (H 2 0 , 298 K): 150. Colourless, X-ray diffraction quality crystals of [PtCl(P,P',7V-dmapcp)]Cl-CH2Cl2T.46 H 2 0 were grown over 4 d by evaporation of a CH 2 C1 2 solution containing the compound onto which EtOH had been layered (CH 2 Cl 2 :EtOH2:5by vol). 67 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.9.2.2 [PtCl(P,P',/V-dmapcp)] [PF6] To a Schlenk tube charged with [PtCl(P,P',7v~-dmapcp)]Cl (38 mg, 0.043 mmol) and NH 4PF6 (7.0 mg, 0.045 mmol) was added acetone (10 mL). The resulting white suspension was stirred at r.t. for 1 h and then filtered through Celite 545. The Celite was washed with acetone (3 x 10 mL) and the combined filtrate was reduced to ca. 1 mL in vacuo. The product was afforded as a white powder by the addition of E t 2 0 (20 mL), isolated by filtration, washed with E t 2 0 ( 3 x 3 m L ) and dried in vacuo at 100 °C. Yield: 21 mg (49 %). Anal. Calcd for C37H48N4ClF6P3Pt: C 45.1; H 4.9; N 5.7. Found: C 44.8; H 4.9; N 5.6. The CDC1 3 solution ' H N M R spectrum of [PtCl(P,P',7V-dmapcp)][PF6] was almost identical to that of [PtCl(P,P',iV-dmapcp)]Cl. 3 1 P{ ! H} N M R (121 MHz, CDC1 3, 300 K) 5 17.0 (d, Vppt = 3510, 2JPP = 13.7), 26.7 (d, ]JPPt = 3390, 2JPP = 13.7), -145 (spt, 1 J P F = 710, PF6"). 2.9.2.3 PtCl2(P,P'-dmape) and [PtCl(P,P',JV-dmape)]Cl This complex was synthesised according to the two routes detailed for the preparation of [PtCl(P,P',/V-dmapcp)]Cl (Section 2.9.2.1). Thus, reaction of PtCl 2(105 mg, 0.40 mmol) and dmape (230 mg, 0.40 mmol) yielded 130 mg (40 %) of a white powder; and reaction of PtCl2(cod) (110 mg, 0.28 mg) with the same ligand (160 mg, 0.28 mmol) gave 200 mg (85 %) of PtCl2(dmape). However, elevated temperatures were not required for the latter preparation that was conducted in CH 2 C1 2 at r.t. Anal. Calcd for C 34H 4 4N4Cl 2P 2Pt: C 48.8; H 5.3; N 6.7; Found: C 48.5; H 5.3; N 6.5. PtCl2(P,P'-dmape): 3 1 P{'H} N M R (121 MHz, CDC1 3 , 233 K) : 5 48.0 (br s, lJm - 3750). The ! H N M R spectrum of this compound is broad and uninformative, even at low temperature, and could not be resolved from that of [PtCl(P,P',7V-dmape)]Cl. The N C / / 3 protons and the methylene protons in the ligand "backbone" fall in the 2-4 ppm range, while the peaks due to aromatic protons lie between 6.5-9 ppm. [PtCl(P,PyV-dmape)]Cl: 3 1 P{'H} N M R (121MHz, CDC1 3, 233 K): 5 34.0 (br s, lJm = 3510), 55.1 (br s, lJPPt = 3460). Evaporation over a period of 3 d of a CH 2 C1 2 solution containing the compound onto which EtOH had been layered (CH 2 Cl 2 :EtOH 1:1 by vol) yielded colourless, X-ray diffraction quality crystals of PtCl 2(P,P-dmape)-CH 2Cl 2. 68 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.9.2.4 [PtCl(P,P',7V-dmape)] [PF6] This compound was prepared in the same manner as outlined for [PtCl(P,P ',N-dmape)][PF6] (Section 2.9.2.2), except that the acetone was removed at the pump and the residue taken up in CH2CI2 (10 mL) before filtration through Celite 545. This removed excess N H 4 P F 6 which is insoluble in CH2CI2. Thus, reaction of PtCLXdmape) (92 mg, 0.11 mmol) and N H 4 P F 6 (42 mg, 0.26 mmol) gave 48 mg (47%) of a white powder. Anal. Calcd for C34H44N4ClF6P3Pt: C 43.2; H 4.7; N 5.9. Found: C 43.3; H 4.9; N 5.8. ' H N M R (200 MHz, CDC1 3 , 300 K): 5 2.2-3.6 (br m, 28H, CH2 and NC//3), 7.10-7.80 (br m, 12H, Ar), 8.05 (m, 4H, Ar). 3 1 P{ ! H} N M R (81 MHz, CDC1 3 , 300 K) : 8 31.6 (s, xJm = 3540, Vp P not observed), 52.7 (s, V P p t = 3446,2JPP not observed), -145 (spt, ]JPF = 710, PF6"). 2.9.2.5 Pt(ox)(P,P '-dmape) To a CH2CI2 (10 mL) solution containing PtCLXdmape) (85.3 mmol, 0.10 mmol) was added H 2 0 (10 mL) followed by solid Na 2ox under air. The biphasic system was stirred overnight and the aqueous layer was removed. The organic fraction was washed with H2O (2x10 mL) and EtOH (5 mL) was added. The solvent volume was reduced to ca. 1 mL in vacuo and the product was afforded as a fine white precipitate by the addition of Et20 (20 mL). This was isolated by filtration, washed with Et20 ( 2 x 5 mL) and air-dried. Yield: 56 mg (65 %). Anal. Calcd for C3 6H44N404P2Pt: C, 50.6; H , 5.2; N , 6.6. Found: C, 50.0; H, 5.5; N 6.4. ! H N M R (300 MHz, CDC1 3, 300 K): 5 2.43 (s, 24H, NC# 3 ), 3.17 (m, 2H, CH2), 7.32 (pt, 4H, Ar), 7.55 (pt, 4H, Ar), 7.63 (pt, 4H, Ar), 7.85 (m, 4H, Ar). 3 1 P{ 1 H} N M R (121 MHz, CDC1 3 , 300 K): § 35.2 (s, xJ?n = 3750 Hz). 2.9.2.6 [PdCl(P,PyV-dmapcp)]Cl To a Schlenk tube containing trans-FdCl2(7hCN)2 (55 mg, 0.14 mmol) and dmapcp (86 mg, 0.14 mmol) was added CH2CI2 (10 mL), and the initially orange solution stirred for 15 min during which it became yellow. The volume was reduced under vacuum to ca. 1 mL and addition of Et20 (20 mL) gave the yellow solid product. This was isolated by filtration, washed with Et20 ( 3 x 3 mL) and dried in vacuo. Yield: 109 mg (98 %). Anal. Calcd for C ^ g N t C L J ^ P d : C 56.4; H 6.1; N 7.1. Found: C 56.1; H 6.3; N 7.0. ' H N M R 69 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes (400 MHz, CDCI3, 300 K) : 5 1.42 (m, 1H, CH2), 1.65 (br m, 1H, CH2), 1.83 (br m, 1H, CH2), 2.15 (br m, 1H, CH2), 2.29 (s, 13H, N C # 3 and CH2 obscured), 2.45 (br m, 1H, CH2), 2.73 (s, 6H, N C / / 3 ) , 2.93 (s, 3H, NC7/ 3), 3.14 (br m, 1H, CH), 3.30 (s, 3H, NC# 3 ), 4.05 (br m, 1H, CH), 6.73 (m, 1H, Ar), 6.82 (pt, 1H, Ar), 7.08 (pt, 1H, Ar), 7.58 (m, 11H, Ar), 8.21 (pt, 1H, Ar), 9.43 (m, 1H, Ar). 3 1 P{'H} N M R (121 MHz, CDC1 3 , 300 K): 5 48,7 (s), 49.6 (s). UV-vis: 334 [6100] (CH 2C1 2); 378 [3900] (H 20). A M (H 2 0, 298 K): 200. 2.9.2.7 [PdI(P,PyV-dmapcp)]I To a Schlenk tube containing solid £raws-PdCl2(PhCN)2 (71 mg, 0.19 mmol) and dmapcp (110 mg, 0.190 mmol) was added CH 2 C1 2 (5 mL), and the resulting orange solution was stirred for 0.5 h during which time it became yellow. To this was added solid Nal (97 mg, 0.65 mmol) followed by acetone (5 mL). The solution immediately became orange and turbid, and was stirred for a further 0.5 h. The solvent was removed in vacuo. The residue was taken up in CH 2 C1 2 (10 mL) and the mixture was filtered through Celite 545. The volume of the filtrate was reduced in vacuo to ca. 2 mL and the product was afforded as a fine, orange powder by the addition of E t 2 0 (20 mL). It was isolated by filtration, washed with Et 2 0 (3 x 3 mL) and dried in vacuo. Yield: 130 mg (72%). Anal. Calcd for C 3 7H48N 4I 2P 2Pd: C, 45.8; H , 5.0; N , 5.8. Found: C, 45.9; H , 5.1; N , 5.6. ' H N M R (200 MHz, CDC1 3 , 300 K): § 1.80 (br m, 2H, CH2), 2.35 (br s, 12H, NGtf 3), 2.70 (br s, 6H, NC/7 3), 3.05 (br s, 3H, NCtf 3 ), 3.70 (br s, 3H, NC/ / 3 ) , 3.95 (br m, 1H, CH), 6.80 (br m, 1H, Ar), 7.05 (br m, 2H, Ar), 7.30-8.00 (br m, 13H, Ar), 8.20 (br m, 1H, Ar), 9.40 (br m, 1H, Ar). The peaks due to the remaining CH2 protons in the ligand "backbone" are obscured by the broad NG/Y 3 peaks. 3 1 P{'H} N M R (81 MHz, CDC1 3 , 300 K) 5 39.9 (s), 50.4 (s). 2-bond PP coupling could not be resolved. 2.9.2.8 [PdCl(P,P',/V-dmapcp)] [PF6] This complex was made in the same manner as outlined for [PtCl(P,P',./V-dmape)][PF6] (Section 2.9.2.2). Thus, reaction of [PdCl(P,P',/V-dmapcp)]Cl (150 mg, 0.19 mmol) and NH4PF6 (157 mg, 0.97 mmol) gave 174 mg (91 %) of a yellow powder. Anal. Calcd for C 3 7 H48N 4 ClF 6 P 3 Pd: C 49.5; H 5.4; N 6.2. Found: C 49.7; H 5.4; N 5.8. The ' H N M R spectrum of this compound is essentially the same as that for [PdCl(P,P',7V-dmapcp)]Cl 70 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes and the 3 1 P{ 1 H} N M R spectrum is identical save for the presence of the spt (5 -145) due to PF6". 2.9.2.9 PdCl2(P,P'-dmape) and [PdCl(P,P',7V-dmape)]Cl This complex was prepared in the same manner as outlined for PtCl2(dmape) (Section 2.9.2.3). Thus, reaction of ^rans-PdCbCPhCNh (56 mg, 0.15 mmol) and dmape (83 mg, 0.15 mmol) gave 88 mg (81 %) of a yellow powder. Anal. Calcd for C 34H44N 4Cl2P2Pd: C 54.6; H 5.9; N 7.5; Found: C 54.6; H 5.9; N 7.6. The ' H N M R spectrum of an equilibrium mixture of the two compounds is uninformative throughout the temperature range 220-300 K because the peaks are broad and overlapped. PdCl2(P,P'-dmape): 3 1 P{'H} N M R (121MHz, CD 2 C1 2 , 233 K): § 66.4 (br s). [PdCI(P,P',7V-dmape)]Cl: 3 1 P{'H} N M R (121 M H z , CD 2 C1 2 , 233 K) : 73.4 (br s), 60.5 (br s). 2.9.2.10 [PdCl(P,P',Ar-dmape)] [PF6] This complex was prepared in the same manner as outlined for [FtCl(P,P',N-dmape)][PF6] (Section 2.9.2.4). Thus, reaction of PdCl2(dmape) (83 mg, 0.11 mmol) and NH4PF6 (93 mg, 0.57 mmol) gave 60 mg (63 %) of a yellow powder. Anal. Calcd for C 34H44N4ClF 6P 3Pd: C 47.6; H 5.2; N 6.5. Found: C 47.4; H 5.4; N 6.4. *H N M R (300 MHz, CDCI3, 300 K): 5 2.6 (br s, 24H, NC/ / 3 ) , 6.8-8.0 (br m, 16H, Ar). 3 1 P{'H} N M R (121 MHz, CD2CI2, 300 K) 5 73.4 (s), 60.5 (s). 2.9.2.11 Pd(OAc)2(P,P'-dmape) This complex was made in a manner similar to that outlined for [PdCl(dmapcp)]Cl (Section 2.9.2.6). Thus an overnight reaction between Pd(OAc)2 (37 mg, 0.17 mmol) and dmape (96 mg, 0.17 mmol) gave 92 mg (70 %) of a yellow powder. Acceptable elemental analysis was not obtained for this complex. ' H N M R (300 MHz, CDC1 3 , 300 K): 5 2.59 (br s, 24H, NC# 3 ) , 3.64 (br m, 4H, CH2), 7.08-7.56 (m, 12H, Ar), 8.00 (br m, 4H, Ar). 3 1 P{ 1 H} N M R (121 MHz, 300 K): 5 61 (br s), 74 (br s) [D 2 0]; 5 70 (br s) [CDC13]. 2.9.2.12 lPdCl(dmapeO)]2[PF6]2 To a CH2CI2 (10 mL) solution containing PdCl2(dmape) (30 mg, 0.041 mmol) was added an aqueous K O H solution (5 mL, 1 mol L" 1), this causing an immediate colour change in 71 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes the organic layer from yellow to orange-red. The two phase mixture was stirred for 1 h and then the aqueous layer was removed. H 2 0 (5 mL) was added, followed by K P F 6 (83 mg, 0.45 mmol), and stirring was continued for 0.5 h. The aqueous layer was removed and the CH2CI2 phase was washed with H2O ( 3 x 1 0 mL) before being filtered through Celite 545. The filtrate was reduced to dryness in vacuo and taken up in CDCI3 for analysis by solution N M R spectroscopy. 3 1 P{'H} N M R (121 MHz, C D C 1 3 , 300 K): 8 28.1 (d, 2JPp = 62.2), 29.3 (d, 2JPP = 62.2), 31.5 (s, dmape(0)2), -144 (spt, PF6", :JP¥ = 710). A small crop of X-ray diffraction quality crystals of the title complex formed by slow evaporation of the CDCI3 solution. 2.9.2.13 Pd 2Cl 2(dmape) To a combination of trans-FdCl2(VhCN)2 (35 mg, 0.091 mmol), Pd 2 (dba) 3 CHCl 3 (47 mg, 0.046 mmol) and dmape (52 mg, 0.091 mmol) was added CH 2 C1 2 (10 mL), and the resulting purple solution was stirred over night. The brown solution was filtered through Celite 545 and the volume of the filtrate was reduced in vacuo to ca. 1 mL. Addition of E t 2 0 (20 mL) gave a green-brown precipitate (57 mg). 3 1 P{ ] H} N M R spectroscopy revealed that this solid contained a symmetric (5p = 35.2) and an asymmetric (5p = 70.8, 86.2) product. The former was isolated by column chromatography using silica gel as the stationary phase (20 cm x 0.8 cm i.d.) and CH 2 Cl2 /EtOH (19:1 by vol.) as eluent. The first band (orange) was collected and reduced to ca. 1 mL in vacuo. Addition of Et20 (10 mL) gave an orange precipitate that was washed with Et20 (3 x 2 mL) and dried under vacuum at 78 °C. Yield: 15 mg (19 %). The N M R data are consistent with the title formulation but, for reasons that remain unclear, the elemental analysis is poor. Anal. Calcd for C34H44N4Cl2P2Pd2: C, 47.8; H , 5.2; N , 6.6. Found: C, 45.9; H , 5.1; N , 6.1. ' H N M R (300 MHz, C D C 1 3 , 300 K): 5 2.38 (s, 12H, NC# 3 ), 2.77 (m, 4 H , CH2), 2.87 (s, 6 H , NC// 3 ) , 3.11 (s, 6 H , N C / / 3 ) , 7.01 (pt, 2H, Ar), 7.15 (pt, 4 H , Ar), 7.29 (m, 2H, Ar), 7.42 (m, 4 H , Ar), 7.54 (m, 2H, Ar), 7.62 (m, 2H, Ar). 3 1 P{ 1 H} N M R (121 MHz, CDCI3, 300 K): 8 35.2 (s). 2.9.2.14 [PdO^-allylXPP'-dmape)] [PF6] To a combination of [Pd(u.-Cl)(n3-allyl)]2 (39 mg, 0.11 mmol), dmape (120 mg, 0.21 mmol) and N H 4 P F 6 (120 mg, 0.72 mmol) was added CH 2 C1 2 (5 mL) and acetone 72 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes (5 mL), and the resulting slurry was stirred for 0.5 h. The solvent was removed in vacuo and the work-up procedure was identical to that for [PtCl(P,P',jV-dmape)][PF6] (Section 2.9.2.4). Yield: 160 mg (86 %). Acceptable elemental analysis was not obtained for this complex. ' H N M R (300 MHz, CDC1 3, 300 K): 5 2.27 (s, 24H, N C i / 3 ) , 4.42 (d, 2H, CHE), 3.07 (d, 2H, CEH), 5.57 (qn, 1H, CH), 7.1-7.6 (m, 16H, Ar). 3 1 P{ ] H} N M R (121 MHz, CDCI3, 300 K): 5 44.2 (s), -144 (spt, ]JPF = 710, PF6"). 2.9.3 Syntheses and reactions of Ru complexes 2.9.3.1 RuCl2(P,P',AyV'-dmape) The title compound was made under two different sets of conditions. (a) CH2CI2, r.t., overnight reaction. To a combination of RuCl2(PPh 3) 3 (71 mg, 0.074 mmol) and dmape (43 mg, 0.075 mmol) was added CH2CI2 (10 mL). The brown solution was stirred magnetically overnight at r.t. The volume of the solvent was reduced in vacuo to ca. 2 mL and E t 2 0 (20 mL) and hexanes (20 mL) were added; when the solvent volume was reduced to ca. 20 mL under vacuum, the product precipitated solely as the cis isomer. Yield: 24 mg (43 %). Anal. Calcd for C 34H44N 4Cl2P 2Ru: C, 55.0; H , 6.0; N , 7.5. Found: C, 55.1; H , 6.1; N , 7.4 %. ] H N M R (200 MHz, D 2 0 , 300 K): 6 1.70 (m, 4 H , CH2), 2.50 (s, 12H, NC//3), 2.90 (s, 12H, NCH3), 7.2-8.0 (m, 16H, Ar). 3 1 P{'H} N M R (121 MHz, D 2 0 , 300 K): 8 83.1 (s). A M (298 K): 211 (H 2 0) , < 1 (CH 2C1 2). (b) CH2CICH2CI, reflux, 2.5 h reaction. To a combination of RuCl2(PPh 3)3 (230 mg, 0.24 mmol) and dmape (140 mg, 0.25 mmol) was added 1,2-dichloroethane (10 mL) and the resulting brown solution refluxed for 2.5 h. The orange solution was reduced to dryness in vacuo and Et20 (10 mL) was added. Tritulation gave a yellow powder which was isolated by filtration, washed with Et20 ( 3 x 5 mL) and dried in vacuo. Yield: 157 mg (86 %). This procedure gave predominantly the cis isomer with ca. 5-10 % of the trans. Cis-RuCh^P'^AT'-dmape): *H N M R (300 MHz, CDC1 3, 300 K): 5 2.25 (s, 6H, N C / / 3 ) , 2.45 (br m, 4 H , CH2), 2.63 (s, 6 H , NCif 3 ) , 2.93 (s, 6 H , NC/ / 3 ) , 3.36 (s, 6H, N C / / 3 ) , 7.05-7.70 (m, 16H, Ar). 3 1 P{'H} N M R (121 MHz, CDC1 3 , 300 K): 5 94.6 (s). rraHs-RuCl2(P,P',AyV'-dmape): 3 1 P{'H} N M R (121 MHz, CDC1 3 , r.t.): 5 26.0 (s). *H N M R peaks were completely obscured by those of the cis isomer and could not be identified. 73 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.9.3.2 [Ru(P,P',N,N',N' \N'' '-dmape)] [PF6]2 To a combination of RuCl2(P, P',N,N'-dmape) (71 mg, 0.096 mmol) and N H 4 P F 6 (80 mg, 0.49 mmol) was added acetone (6 mL), and the yellow suspension stirred for 1 h. The solvent was removed in vacuo and the residue taken up in CH2CI2 (10 mL) which was then filtered through Celite 545. The yellow filtrate was reduced to ca. 1 mL at the pump and the product was afforded as a pale yellow powder by the addition of Et20 (10 mL). This was isolated by filtration and washed with Et20 ( 3 x 2 mL). The yield could not be determined accurately at this point due to the product being a mixture of solvent-coordinated species, [Ru(S)„(/,i)7v^.„-dmape)][PF6]2 (S = acetone, H2O; n = 1-4), but was approximately 70 %. The product on drying thoroughly in vacuo at 78 °C changed from yellow to orange-red. Anal; Calcd for C 3 4 H 4 4 N 4 F 1 2 P 4 R U : C, 42.5; H , 4.6; N , 5.8. Found: C, 42.4; H , 4.9; N , 6.0 %. ' H N M R (300 MHz, CDC1 3, 220 K): 5 2.30 (m, 4H, CH2), 1.70 (br s, 6H, NC# 3 ) , 2.67 (br s, 12H, NC# 3 ), 3.10 (br s, 6H, NCH3), 7.0-7.8 (br m, 16H, Ar). 3 1 P{'H} N M R (121 MHz, CDC1 3 , 220 K): 6 95.6 (s). 2.9.3.3 Reaction of [Ru(P,P,,N,N',N",N",-dmape)][PF6]2 with H 2S An N M R tube containing ca. 5 mg of [Ru(S)„(P2,Ar4-n-dmape)][PF6]2 (see Section 2.9.3.2, above) and fitted with a J. Young tap was evacuated at 100 °C for 5 min. In this time, the solid reagent changed from yellow to orange-red. CDCI3 (ca. 0.5 mL) was then condensed into the tube, and 1 atm of H2S was admitted. The solution changed colour from red to yellow. The 3 1 P{'H} N M R data for this reaction are reported in Table 2.10. 2.9.3.4 Reaction of [Ru^P'^/V^/V^/V'''-dmape)] [PF6]2 with H 2 0 An N M R tube charged with [Ru(S)„(P 2,^-»-dmape)][PF6]2 (ca. 45 mg; Section 2.9.3.2) was placed uncapped in a Schlenk tube connected to a vacuum pump. The Schlenk tube was immersed in an oil-bath set at 100 °C and evacuated for 10 min. To CH2CI2 (20 mL), freshly-distilled from CaH 2 , was added H2O (20 uL) and the mixture was shaken thoroughly. Analysis by Carl-Fisher titration showed the CH2CI2 to contain 850 ppm H 2 0 . Under a flow of Ar, this solution (ca. 1.5 mL) was used to dissolve the [Ru(P,P',7V,7V',/V",/V'"-dmape)][PF6]2 generated in the N M R tube, which was then capped and sealed with Parafilm. The 3 1 P{'H} N M R data for this reaction are reported in Table 2.10. 74 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 2.9.4 Miscellaneous reactions 2.9.4.1 Reactions between dmapcp and PdCl2(cbd) The title reaction was conducted both at r.t. and at -40 °C. (a) At r.t. To a combination of solid PdCl2(cod) (44 mg, 0.15 mmol) and dmapcp (91 mg, 0.15 mmol) was added CH2C12 (8 mL). The resulting orange solution was stirred at r.t. for 1 h when it became bright yellow. NH4PF6 (126 mg, 0.77 mmol) was added followed by acetone (7 mL). The solution became cloudy instantly. Stirring was continued for an additional 1 h before the suspension was reduced to dryness in vacuo. The residue was taken up in CH2CI2 (10 mL) and the mixture was filtered through Celite 545. The filtrate was reduced to ca. 1 mL at the pump and ether (20 mL) was added to give the product mixture as a yellow powder. This was isolated by filtration, washed with Et20 and dried in vacuo. Yield: 145 mg. The product is a mixture of [PdCl(P,P',iV-dmapcp)][PF6] and an organometallic intermediate as discussed in Section 2.5.7; the ] H and 3 1 P N M R spectra contain peaks due to [PdCl(P,P',7V-dmapcp)]+ (Section 2.9.2.6) and these others (assignments followed by "?" are tentative): ' H N M R (300 MHz, CDC1 3, 300 K): 5 0.90 (m, 1H, CH2 of C 8 H 8 moiety), 1.05 (m, 1H, HC-Pd ?), 1.60 (m, 1H, CH2 of dmapcp), 1.80 (m, 1H, CH2 of C 8 H 8 moiety), 1.90 (m, 1H, CH2 of C 8 H 8 moiety), 2.05 (m, 1H, CH2 of dmapcp), 2.18 (s, 3 H , NGtf 3), 2.21 (s, 12H, N C i / 3 ) , 2.85 (s, 6 H , NC/fc), 2.90 (s, 3 H , NC# 3 ) , 3.40 (m, 1H, CH2 of C 8 H 8 moiety), 3.70 (m, 1H, CH of dmapcp), 4.15 (m, 1H, N + - C # ?), 5.23 (m, 1H, CH=CH), 5.65 (m, 1H, CH=CH). Except for a prominent multiplet at 8 8.67, peaks due to the intermediate overlap with those due to [PdCl(P,.P',./V-dmapcp)]+ in the aromatic region and could not be discerned. 3 1 P{ 1 H} N M R (121 MHz, CDC1 3 , 300 K): 8 36.5 (d, 2JPP = 8.8), 47.8 (d, 2JPP = 8.8). (b) At -40 °C (In this procedure, "cold" and "cooled" refer to -40 °C.) A CH 2 C1 2 (5 mL) slurry containing PdCl2(cod) (94 mg, 0.33 mmol) and NH4PF 6 (160 mg, 0.98 mmol) was cooled on a dry ice/CH 3 CN bath. Over the course of 5 min, a cold CH2CI2 (5 mL) solution containing dmapcp (200 mg, 0.33 mmol) was added to the slurry. After 5 min, cold acetone (5 mL) was added and stirring was continued for 0.5 h. The solvent was removed in vacuo at -40 °C, and the residue was taken up in cold CH2CI2 (10 mL) and the mixture was filtered through Celite 545. The filtrate was reduced in vacuo to ca. 1 mL and cold E t 2 0 (20 mL) was added to give the product mixture as an 75 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes off-white powder that was recovered in the same manner as outlined in (a). Yield: 160 mg. The product is a mixture of [PdCl(P,P',/v'-dmapcp)][PF6] and several organometalhc intermediates as discussed in Section 2.5.7. The P N M R (121 MHz, CDCI3, 300 K) spectrum recorded 6 hours after making the sample contains peaks due to [PdCl(JP)P',7V-dmapcp)]+ (Section 2.9.2.6) and these others: 5 33.8 (d, V P P =111), 38.6 (d, 2 J P P = 111), 24.7 (d, 2JPP = 85.5), 50.3 (d, 2 J P P = 85.5), 26.4 (d, V P P = 102), 41.4 (d, J P P = 102). The H N M R spectrum is too complicated for any definite assignments to be made, but evidence of coordinated olefin is given by a series of multiplets in the 8 4.6-6.4 range. 2.9.4.2 Halide-free reaction between PdCh(dmape) and KOH The initial step of title reaction was performed as outlined in Section 2.9.2.12, except that the PdCi2(dmape) (100 mg, 0.14 mmol) was dissolved in CH3NO2 (5 mL) instead of in CH2CI2. After 2 h of reaction, the aqueous (red) and organic (orange) phases were separated. The aqueous phase was washed with CH3NO2 ( 3 x 5 mL) and the combined organic fractions were reduced to dryness in vacuo and the residue was taken up in acetone (10 mL). The volume of the red solution was reduced at the pump to ca. 1 mL, and Et 2 0 (20 mL) was added to give the product as a clump which was broken up by sonication. Yield: 70 mg. ^ P ^ H } N M R (121 MHz, acetone-d6, 300 K) : 5 61 (br s), 73 (brs). 2.9.4.3 Reduction of PtCl2(dmape) by Na/Hg An amalgam was made by dissolving Na (20 mg) in Hg (3 mL). Onto this was layered a THF (20 mL) solution containing PtCi2(dmape) (ca. 50 mg, 0.060 mmol). The colourless organic layer immediately became brown and a precipitate formed. After 20 min, the slurry was decanted into C6H6 (15 mL) and the combined mixture was filtered through Celite 545. The red filtrate was reduced to dryness in vacuo yielding an orange solid. 3 1 P{'H} N M R (121 M H z , C 6 D 6 , 300 K): 8 38.9 (s, ' j P P t = 3580). 2.9.4.4 In situ generation of Pt(t]2-dmm)(i>,P'-dmape) To an N M R tube fitted with a J. Young screw cap was added Pt(ox)(P,P'-dmape) (7.4 mg, 0.009 mmol), dmm (5.5 uL, 0.044 mmol) and CD 2 C1 2 (0.5 mL). Three freeze-76 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes pump-thaw cycles were performed. The tube was then subjected to U V irradiation by a Hg lamp for 2.25 h when the solution changed from colourless to yellow. The following peak assignments are tentative. ] H N M R (300 MHz, CD 2 C1 2 , 300 K): 8 2.50 (NC# 3), 3.20 (CH2), 3.69 and 3.73 (CH=CH), 3.77 (C(0)OC// 3), 7.1-7.7 (Ar). 3 1 P{ ] H} N M R (121 MHz, CD 2 C1 2 , 300 K): 8 43.8 (s, ]JPPt = 3910). 2.9.4.5 Reaction of PdCl2(dmape) and bases (a) K2C03. To a mixture of PdCl2(dmape) (15 mg, 0.020 mmol) and K 2 C 0 3 (140 mg, 1.00 mmol) was added acetone (5 mL) and the resulting yellow slurry was stirred magnetically overnight when it turned orange-red. The solvent was removed in vacuo, the residue was taken up in CH 2 C1 2 and the mixture filtered through Celite 545. The filtrate was reduced to dryness under vacuum and the solid was dissolved in CDC1 3 for solution N M R analysis. 3 1 P{'H} N M R (121 MHz, CDC1 3, 300K): 8 8 35.2 (s), 70.8 (d, 2 J P P = 565), 86.2 (d, 2JPP = 565). (b) KCrBu. To a CH 2 C1 2 (5 mL) solution of PdCl2(dmape) (110 mg, 0.15 mmol) was added KO'Bu (16 mg, 0.15 mmol). The yellow solution turned orange within 5 min and was stirred under Ar for 2 d during which a yellow precipitate formed. This was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried in vacuo. The yield was not recorded. 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K): 8 35.2 (s), 70.8 (d, V P P = 565), 86.2 (d, 2 J P P = 565). 2.9.5 Catalytic C-N bond formation 2.9.5.1 Experimental To a CgH6 solution (5 mL) containing PhBr (100 uL, 0.98 mmol), aniline (90 uL, 0.99 mmol) and PdCl2(dmape) (15 mg, 0.019 mmol) in a 3-neck round-bottom flask topped with a condenser and heated to 70 °C was added KO'Bu (HOmg, 0.98 mmol) under air. This marked the beginning of the reaction. At 30 min intervals, 2 uL samples were withdrawn by syringe, 0.1 uL of each being used for GC analysis. 2.9.5.2 GC analysis GC parameters used in analysis of C - N bond forming reactions are given in Table 2.11, and elution times under these conditions for components of interest appear in Table 2.12. 77 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes Table 2.11 GC parameters used in the determination of C - N bond forming reaction components. Parameter Setting Initial oven temperature 80 °C Final oven temperature 220 °C Rate 20 °C min"1 Initial time 2 min Final time 4 min Injector temperature 220 °C Detector temperature 220 °C Column head pressure 105 kPa Table 2.12 Elution times for Heck reaction components under the GC conditions given in Table 2.11. Component Retention time (min) PhBr 2.79 Aniline 3.90 Diphenylamine 9.15 2.9.6 Determination of equilibrium constants from 31P{JH} VT NMR data 3 I P { ' H } N M R spectra were acquired for CD2CI2 solutions containing the appropriate 2 1 complex (ca. 6-9 x 10" mol L") at various temperatures in the range 213-273 K. In order to ensure accurate integrations, the delay between pulses was set to 4 s. This is reasonable in light of spin-inversion-recovery experiments which showed that the longest Ti relaxation of all the P-nuclei was approximately 0.75 s. The decoupler was set to fire only in the acquisition phase of the pulse program so as to minimise NOE perturbations. The acquisition time in all experiments was 0.4 s. An equilibration time of 0.5 h was allowed at each temperature before acquisition was begun. 2.9.6.1 The PtCl2(P,P'-dmape) [PtCl(P,P',7V-dmape)]+ + CY equilibrium A typical 3 1 P { ' H } N M R spectrum for the system is shown in Figure 2.17. The peaks due to [PtC\(P,P',N-dmape)]+ are marked by the number symbol (#). The peaks due to PtChCP.P '-dmape) are indicated by asterisks (*). Because some of the peaks are overlapped or closely juxtaposed, the 9 peaks have been divided into 7 groups (labelled 78 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 1-7), with areas A1-A7, respectively. The total concentration of Pt is P t T O T - In order to determine the concentrations of PtCl 2(P,P -dmape) and [PtCl(P,P ',JV-dmape)]+ in solution and thus an equilibrium constant, K , at each temperature, the following procedure was adopted. 6 Figure 2.17 A typical low temperature ^Pf 'H} N M R (121 MHz, CD 2 C1 2 , 240 K) spectrum of an equilibrium mixture of PtCl2(P,P'-dmape) and [PtCl(P,P',/V-dmape)]Cl. Peak assignments are discussed in the text. The area under the major peaks due to [PtCl(P,P',./V-dmape)]+ ( A P P N ) , i.e., not counting "satellites", is given by: A P P N = A 3 + A 6 - A 2 The area under the corresponding peak due to PtCl2(P,P'-dmape) (A P P ) is: A P P = A4 - A7 The concentration of [PtCl(P,P',./V-dmape)]+ in solution (PPN) is therefore: PPN = A P P N / ( A P P N + A P P ) x P t T 0 T And, likewise, the concentration of PtCl 2(P,P '-dmape) (PP) is given by: PP = A P P / ( A P P N + A P P ) x P t T 0 T For the equilibrium PtCl 2(P,P '-dmape) = ^ [PtCl(P,P ',/V-dmape)]+ + CI", 79 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes K = P P N 2 / P P 2.9.6.2 The PdCl2(P,P'-dmape) [PdCl(P,P',/V-dmape]+ + CX equilibrium The case of Pd is simplified due to the absence of "satellites" and the spectrum contains only 3 peaks, labelled 1-3, with areas A1-A3, respectively. A typical spectrum is shown in Figure 2.18. Once again, peaks due to the P,P',/V-isomer [PdCl(P,P',7V-dmape)]+ are marked with the number symbol (#) and that due to the P,P '-isomer by an asterisk (*). The calculation of equilibrium constants is performed in exactly the same manner as outlined above save for the fact that overlapping "satellites" do not have to be accounted for. As in the case of PtCl2(dmape), the total concentration of Pd is P d T O T -3 # 1—I—1—1—1—1—1—1—1—1—1—p—I I I I—j—1—1—1—1—[—1—1—|— 80 75 70 65 60 ppm Figure 2.18 The 3 1 P{'H} N M R (121 MHz, CD 2 C1 2 , 233 K) spectrum for the PdCl2(dmape) system. Peak assignments are discussed in the text. The area under the peaks due to [PdCXPP^TV-dmape)]"1" ( A P P N ) is given by: A P P N = A i + A 3 The area under the peak due to PdCl2(P,P'-dmape), A P P = A 2 . The concentration of [PdCl(P,P',N-dmape)]+ in solution (PPN) is therefore: PPN = A P P N / ( A P P N + A P P ) x P d T 0 T And, likewise, the concentration of PdCl2(P,P-dmape) (PP) is given by: 80 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes PP = Ap P / (Ap P N + ApP) X P t T 0 T For the equilibrium PdCl2(P,P'-dmape) ^== [PdCl(P,P',iV-dmape)]+ + CI", K = P P N 2 / P P 2.10 References 1. Toth, I.; Hanson, B. E.; Davis, M . E. J. Organomet. Chem. 1990, 396, 363. 2. Toth, I.; Hanson, B. E.; Davis, M . E. Catal. Lett. 1990, 5, 183. 3. Toth, I.; Hanson, B. E. Tetrahedron: Asymmetry 1990, 7, 913. 4. Toth, I.; Hanson, B. E. Tetrahedron: Asymmetry 1990,1, 895. 5. Toth, I.; Hanson, B. E.; Davis, M . E. J. Organomet. Chem. 1990, 397, 109. 6. Achiwa, K. Chem. Lett. 1978, 905. 7. Baker, G. L.; Fritschel, S. J.; Stille, J. R.; Stille, J. K. J. Org. Chem. 1981, 46, 2954. 8. Nuzzo, R. G.; Feitler, D.; Whitesides, G. M . Am. Chem. Soc. 1979,101, 3683. 9. Achiwa, K . J. Am. Chem. Soc. 1976, 98, 8265. 10. Malmstrom, T.; Andersson, C. Chem. Commun. 1996, 1135. 11. Ojima, I.; Kogure, T.; Yoda, N . J. Org. Chem. 1980, 45, All?,. 12. ter Halle, R.; Colasson, B.; Schulz, E.; Spagnol, M . ; Lemaire, M . Tetrahedron Lett. 2000, 41, 643. 13. Smith, R. T.; Baird, M . C. Inorg. Chim. Acta 1982, 62, 135. 14. Smith, R. T.; Ungar, R. K. ; Sanderson, L. J.; Baird, M . C. Organometallics 1983, 2, 1138. 15. Lamouille, T.; Saluzzo, C ; ter Halle, R.; Guyader, F. L. ; Lemaire, M . Tetrahedron Lett. 2001, 42, 663. 16. Mirabelli, C. K. ; Hi l l , D. T.; Faucette, L. F.; McCabe, F. L.; Girard, G. R.; Brian, D. B.; Sutton, B. M . ; Bartus, J. O. L.; Crooke, S. T.; Johnson, R. K. J. Med. Chem. 1987, 50,2181. 17. Regnat, D.; Kleiner, H.-J.; Bahrmann, H. German Patent 196 09 336 CI , 1996 (Hoeschst AG). 18. Nagel, U . ; Kinzel, E. Chem. Ber. 1986,119, 1731. 81 References on page 81 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 19. Buhling, A. ; Elgersma, J. W.; Nkrumah, S.; Kamer, P. C. J.; van Leeuwen, P. W. N . M . J. Chem. Soc., Dalton Trans. 1996, 2143. 20. Buhling, A. ; Kamer, P. C. J.; van Leeuwen, P. W. M . N . ; Elgersma, J. W.; Goubitz, K. ; Fraanje, J. Organometallics 1997, 16, 3027. 21. Casey, C. P.; Whiteker, G. T.; Melville, M . G.; Petrovich, L. M . ; James A. Gavney, J.; Powell, D. R. J. Am. Chem. Soc. 1992, 114, 5535. 22. Andrieu, J.; Camus, J.-M.; Dietz, J.; Richard, P.; Poli, R. Inorg. Chem. 2001, 40, 1597. 23. Bartik, T.; Bartik, B.; Hanson, B. E.; Glass, T.; Bebout, W. Inorg. Chem. 1992, 31, 2667. 24. Amrani, Y . ; Lecomte, L.; Sinou, D.; Bakos, F.; Toth, I.; Heil, B. Organometallics 1989, 5, 542. 25. Larpent, C ; Patin, H. ; Valdor, J. F. Synth. Commun. 1991, 21, 495. 26. Herrmann, W. A. ; Albanese, G. P.; Manetsberger, R. B. ; Lappe, P.; Bahrmann, H. Angew. Chem. Int. Ed. Engl. 1995, 34, 811. 27. Schmidbauer, H . Monatsh. Chem. 1965, 96, 2051. 28. Herd, O.; Langhans, K . P.; Stelzer, O.; Weferling, N . ; Sheldrick, W. S. Angew. Chem. Int. Ed. Engl. 1993, 32, 1058. 29. HeBler, A. ; Fischer, J.; Kucken, S.; Stelzer, O. Chem. Ber. 1994,127, 481. 30. Albano, V . G.; Natile, G.; Panunzi, A . Coord. Chem. Rev. 1994, 133, 67. 31. Albano, V . G.; Monari, M . ; Orabona, I.; Ruffo, F.; Vitagliano, A . Inorg. Chim. Acta 1997, 265, 35. 32. Albano, V . G.; Castellari, C ; Morelli, G.; Vitagliano, A . Gazz. Chim. Ital. 1989, 119,235. 33. Green, M . ; Hancock, R. I. J. Chem. Soc. (A) 1967, 2054. 34. Stille, J. K. ; Morgan, R. A. J. Am. Chem. Soc. 1966, 88, 5135. 35. Stille, J. K. ; Morgan, R. A. ; Whitehurst, D. D.; Doyle, J. R. J. Am. Chem. Soc. 1965, 87, 3282. 36. Miki , K. ; Yama, M . ; Kai, Y . ; Kasai, N . J. Organomet. Chem. 1982, 239, All. 37. Kurosawa, H. ; Majima, T.; Asada, N . J. Am. Chem. Soc. 1980,102, 6996. 38. Anderson, J. S. J. Chem. Soc. 1934, 971. 39. Panunzi, A. ; Renzi, A . D.; Paiaro, G. Am. Chem. Soc. 1970, 92, 3488. 82 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 40. Huang, C - C ; Duan, J.-P.; Wu, M. -Y. ; Liao, F.-L.; Wang, S.-L.; Cheng, C.-H. Organometallics 1998, 17, 676. 41. Fallis, S.; Anderson, G. K. ; Rath, N . P. Organometallics 1991,10, 3180. 42. Rulke, R. E.; Ernsting, J. M . ; Spek, A. L.; Elsevier, C. J.; van Leeuwen, P. W. N . M . ; Vrieze, K . Inorg. Chem. 1993, 32, 5769. 43. Bailey, C. T.; Lisensky, G. C. J. Chem. Educ. 1985, 62, 896. 44. Dixon, K. R. in Multinuclear NMR; Mason, J., Ed.; Plenum: New York, 1987. 45. Renzi, A . D.; Morelli, G.; Panunzi, A. ; Vitagliano, A . Gazz. Chim. Ital. 1987, 117, 445. 46. Pieri, G.; Pasquali, M . ; Leoni, P.; Englert, U . J. Organomet. Chem. 1995, 491, 27. 47. Ruiz, J.; Vicente, C ; Marti, J. M . ; Cutillas, N . ; Garcia, G.; Lopez, G. J. Organomet. Chem. 1993, 460, 241. 48. Cyr, P. W. Ruthenium Phosphine Complexes with Ancillary N-donor Ligands; Ph.D. Dissertation, University of British Columbia: Vancouver, 2001. 49. Orpen, A . G.; Brammer, L.; Allen, F. H. ; Kennard, O.; Watson, D. G. J. Chem. Soc, Dalton Trans. 1989, SI. 50. Abu-Gnim, C. O.; Amer, I. J. Organomet. Chem. 1996, 516, 235. 51. Gladiali, S.; Aberico, E.; Pulacchini, S.; Kollar, L. J. Mol. Catal. (A) 1999, 143, 155. 52. Gladiali, S.; Pulacchini, S.; Fabbri, D.; Manassero, M . ; Sansoni, M . Tetrahedron: Asymmetry 1998, P, 391. 53. Nandi, M . ; Jin, J.; RajanBabu, T. V. J. Am. Chem. Soc. 1999, 121, 9899. 54. Brassat, I.; Keim, W.; Killat, S.; Mothrath, M . ; Mastorilli, P.; Nobile, C. F.; Suranna, G. P. Mol. Catal. (A) 2000,157, 41. 55. Abatjoglou, A . G.; Kapicak, L. US Patent 4 429 161, 1984 56. Grushin, V . V . J. Am. Chem. Soc. 1999,121, 5831. 57. Bao, Q.-B.; Landon, S. J.; Rheingold, A . L.; Haller, T. M . ; Bri l l , T. B. Inorg. Chem. 1985, 24, 900. 58. Heinze, K. ; Huttner, G.; Zsolnai, L. Chem. Ber. 1997,130, 1393. 59. Tanase, T.; Igoshi, T.; Yamamoto, Y . Inorg. Chim. Acta 1997, 256, 61. 60. Siegl, W. O.; Lapporte, S. J.; Collman, J. P. Inorg. Chem. 1971,10, 2158. 61. Sevillano, P.; Habtemariam, A. ; Castineiras, A. ; Garcia, M . E.; Sadler, P. J. Polyhedron 1999, 75, 383. 83 Chapter 2: Anilinyldiphosphine Ligands and Pt and Pd Complexes 62. Pringle, P. G.; Shaw, B. L. J. Chem. Soc., Chem. Commun. 1982, 581. 63. Jia, G.; Ng, W. S.; Chu, H . S.; Wong, W.-T.; Y u , N . T.; Williams, I. D. Organometallics 1999,18, 3597. 64. Grushin, V . V. ; Alper, H. Organometallics 1993,12, 1890. 65. Ziegler, C. B.; Heck, R. F. J. Org. Chem. 1978, 43, 2941. 66. Stille, J. K. Angew. Chem. Int. Ed. Engl. 1986, 25, 508. 67. Miyaura, N . ; Suzuki, A . Chem. Rev. 1995, 95, 2457. 68. Guari, Y . ; van Strijdonck, G. P. F.; Boele, M . D. K. ; Reek, J. N . H. ; Kamer, P. C. J.; van Leeuwen, P. W. N . M . Chem. Eur. J. 2001, 7, 475. 69. Driver, M . S.; Hartwig, J. F. J. Am. Chem. Soc. 1996,118, 7217. 70. Aranyos, A. ; Old, D. W.; Kiyomori, A. ; Wolfe, J. P.; Sadighi, J. P.; Buchwald, S. L. J. Am. Chem. Soc. 1999,121, 4369. 71. Amatore, C ; Jutand, A. ; M'Barki, M . A . Organometallics 1992,11, 3009. 72. Malatesta, L.; Cariello, C. / . Chem. Soc. 1958, 2323. 73. Paonessa, R. S.; Prignano, A. L.; Trogler, W. C. Organometallics 1985, 4, 647. 74. Braunstein, P.; Naud, F.; Dedieu, A. ; Rohmer, M . - M . ; DeCian, A. ; Rettig, S. J. Organometallics 2001, ASAP. 75. Dieck, H . A. ; Heck, R. F. J. Am. Chem. Soc. 1974, 96, 1133. 76. Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 1997, 62, 6066. 77. Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem. Soc. 1996,118, 7215. 78. Marcoux, J.-F.; Wagaw, S.; Buchwald, S. L. J. Org. Chem. 1997, 62, 1568. 79. Old, D. W.; Wolfe, J. P.; Buchwald, S. L. J. Am. Chem. Soc. 1998,120, 9722. 80. Hartwig, J. F. Synlett 1997, 329. 81. Khorasani-Motlagh, M . ; Safari, N . ; Pamplin, C. B.; Patrick, B. O.; James, B. R. Inorg. Chim Acta 2001, 320, 184. 82. Ma, E. S. F. Reactivity and Coordination Chemistry of Ru(II) Aminophosphine Complexes with H2S, Thiols, H2O and Other Small Molecules; Ph.D. Dissertation, University of British Columbia: Vancouver, 1999. 83. Mudalige, D. C ; Ma, E. S.; Rettig, S. J.; James, B. R.; Cullen, W. R. Inorg. Chem. 1997, 36, 5426. 84. Pamplin, C. B. ; Ma, E. S. F.; Safari, N . ; Rettig, S. J.; James, B. R. J. Am. Chem. Soc. 2001,123, 8596. 84 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3 Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.1 Introduction The Pd-catalysed Heck reaction represents an important method for the formation of C—C bonds. Since its introduction in 1968,1 it has been the subject of intensive research,2 and mechanistic aspects3'4 have been elucidated. In addition, much work has focussed on improving the efficiency and scope of the reaction by tailoring the Pd catalyst and by varying the reaction conditions (solvent, temperature, base, additives). The first use of Pd-phosphine complexes for vinylic hydrogen substitution reactions was by Dieck and Heck in 1974.5 Although the catalysts are most frequently generated in situ (e.g. from Pd(OAc)2 and a monodentate tertiary phosphine),6"9 many preformed catalysts are now being reported. These include Pd complexes of chelating PP , 1 0 P C , 1 1 P C P , 1 2 S C S 1 3 and P N 1 4 ligands. In general, these compounds demonstrate much higher catalytic activities and lifetimes than their predecessors (which typically required a 1-5 mol % loading) and turnover numbers as high as 106 have been observed.15 The appearance of Pd catalyst precursors bearing " P N " ligands, and the fact that the Heck and related reactions have been successfully conducted in water (in the presence of a quaternary ammonium salt),1 6 or in mixtures of polar organic solvents (e.g. DMF, THF, z'-PrOH, nitriles) and water using Pd complexes of sulphonated phosphines,17 or of guanadinium phosphines,18 prompted this author to investigate the anilinyldiphosphine complexes resulting from this work as catalyst candidates for the Heck reaction. In addition, two other developments encouraged the investigation. The first was the observation by Bankston et al. that in'an intramolecular Heck cyclization of a series of crotyl ethers, the use of RhCl(PPh3)3 in combination with Pd(OAc)2 yielded a system in which the rate of reaction and the selectivity for the endocyclic form of the bicyclic ether products were significantly enhanced as compared to the use of Pd(OAc)2 alone:19 85 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction the two metals were considered to interact synergistically to mediate the intramolecular coupling, although no concrete proof or mechanism was presented. It was recognised early in this work that the strong tendency for -CH2- tethered diphosphines to bridge metal centres20 combined, in the case of dmapm (Chart 3.1, left), with the possibility of five-membered P N chelate formation, gave access to a variety of homo- and heterobimetallic complexes of the general form illustrated in Chart 3.1 (right; Ar = o-C 6 H 4 NMe 2 , X = halide). Chart 3.1 N M e 2 M e 2 N dmapm Use of these complexes as Heck catalysts allowed the assessment of the catalytic effects of a second metal centre, not supplied simply as an additive, as in Bankston's case, but rather as an integral component of a bimetallic catalyst. Cooperative effects between adjacent metal centres in complexes containing two or more bridged metal ions are now established for a variety of homogeneous, catalytic transformations and continue to attract attention. The great majority of these employ homobimetallic catalyst precursors. Representative examples include bimetallic catalysts for phosphoester hydrolysis (Co 2 , 2 1 Cu 2 , 2 2> 2 3 Zn 2 2 4 ) , nitrile hydration (Ni 2 , 2 5 > 2 6 Pd 2 2 7 ) , hydroformylation (Rh 2 , 2 8 " 3 1 Ru 2 3 2 ) , alkene33 and alkyne 3 4 hydrogenation (Ir2) and asymmetric epoxide ring-opening35 (Cr2). In addition, cooperative effects between two different metal centres have been invoked but not proven to be operative in the catalytic hydrogenation of cyclohexene36 (Rulr) and in hydroformylations30 (RhPd). The second development which motivated the Heck study was the so-called "thermomorphic" method described by Bergbreiter and coworkers.37 In this system, a Pd-centre is covalently bound to a soluble polymer which is immiscible with the liquid reactants at r.t. At elevated temperature, the polymer and reactants become miscible and s=J Ar Ar \ = , M e 2 N ~P P x N M e 2 X X X X 86 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction the catalysis occurs. Once the reaction is complete and the temperature is lowered, the product and polymer separate into two phases and the catalyst containing phase can be reused. The catalytic conditions outlined in this chapter are similar in principle: at r.t. the reaction mixture comprises two phases, whereas at 100 °C there is only one phase, and at r.t. after 3 h of reaction the product, stilbene, crystallises. 3.2 Scope This chapter examines the coordination chemistry of Pt", Pd" and Rh 1 with the C H 2 -bridged anilinyldiphosphine ligand, dmapm, (see Sections 3.9.1-3.9.16 for the syntheses of the compounds, and Sections 3.3-3.4.4 for discussion) and investigates a pair of the resulting homo- and heterobimetallic complexes as catalysts for the Heck coupling of iodobenzene and styrene paying particular attention to the possibility for cooperative effects (Section 3.6). A generally applicable cooperativity index is proposed as a kinetic complement to the Hi l l coefficient (Section 3.6.3). The solution dynamics of the precursor used for the bulk of the work, PdCl2(dmapm) (Section 3.3.1), and of the bimetallic complexes MM'CL^dmapm) (M = Pd"; M ' = Pd", Pt"; Section 3.4.1), as well as the structure and bonding of PdCl2(dmapm) (Section 3.3.3), and its reactivity with halide and cyanide (Section 3.3.4), peroxide and sulphur (Section 3.3.5) and metal precursors (Sections 3.4.1 [Pt" and Pd"], 3.4.3 [Rh1] and 3.4.5.1 [Pd0]), are examined in detail. The syntheses of metal-metal bonded homo-and heterobimetallic M 1 complexes both by conproportionation of M° and M ! I (Section 3.9.17) and by 2-electron reduction of M 1 1 dimers (Section 3.9.18) are presented. The reactivities of these bimetallic complexes toward CO and diethylacetylene dicarboxylate (DEAD) are also described (Section 3.4.5.3). 3.3 Complexes Containing One Metal Centre 3.3.1 Solution dynamics The 3 , P{ ! H} solution N M R spectrum (at 25 °C) of the product of the reaction between dmapm and rrarcs-PdC^PhCNh shows that two species are formed (Section 3.9.1). The 87 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction spectrum consists of an upfield singlet (8 -56.8) indicative of a C2-symmetric compound, in which both P-atoms are involved in a 4-membered chelate with Pd , 3 8 i.e., VdCliiP.P'-dmapm), and an A X pattern with high- and low-field doublets (8 -40.0 and 33.5, 2Jpp- = 108 Hz), consistent with a complex bearing one "dangling" P-atom (high-field) and one P-atom involved in a 5-membered chelate with Pd, i.e., PdCi2(P,./V-dmapm) (see Figure 3.4 for the molecular structure of the latter isomer). Elemental analysis of the isolated solid is consistent with the formulation PdCl2(dmapm), confirming that these are isomers. In addition, varying the temperature at which the spectrum is acquired reversibly alters the relative peak intensities, which shows that PdCl2(P,.P'-dmapm) and PdCh(P,N-dmapm) are in thermal equilibrium (Scheme 3.1; Ar = o-C6H 4NMe 2). In contrast to this is the fact that reaction of one equiv. of dmapm with PtCl2(cod) generates PtCl 2(P,P'-dmapm) exclusively (Section 3.9.9). Scheme 3.1 Me 2 N C I 7 KQ\ NMe 2 The temperature-dependence of the N C H 3 region of the ' H N M R spectrum of an equilibrium mixture of PdCi2(dmapm) isomers is given in Figure 3.1. At 300 K, five singlets due to N C / / 3 protons are apparent at 8 3.70, 3.47, 2.85, 2.49 and 2.35 with relative integrations of 1:1:2:2:3, respectively. These peaks represent all 24 N C / / 3 protons associated with PdC^CP.P'-dmapm) and 18 of the 24 corresponding protons of PdChCP./V-dmapm). The remaining 6 N C / / 3 protons of the latter are broadened into the baseline at r.t. and only become visible as the temperature approaches 240 K . These assignments are based on the following observations. When the solution is cooled from 300 to 215 K , the singlet which at 300 K appears at 8 2.35 becomes less intense, broadens and eventually disappears, while two new peaks at 8 1.61 and 2.93 grow in. A l l the other M e 2 N , 88 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction peaks in this region maintain their relative intensities, but drift slightly in chemical shift (by less than 0.1 ppm). As judged by the 3 1 P{ 1 H} spectrum at 215 K which shows a greatly diminished singlet due to PdCl2(P,P'-dmapm), the ] H spectrum at this temperature is due almost completely to PdC^PiV-dmapm). Therefore, the peak at 5 2.35 (essentially temperature-invariant) is due to all 24 chemically equivalent NCP/3 protons of the P,P '-isomer, while the singlets at 8 1.61 and 2.93 are associated with the PAMsomer. The two equal intensity peaks which at 300 K appear at 8 3.70 and 3.47 are most likely due to the two diastereotopic methyl groups directly connected to the Pd-bound N-atom as judged by the large coordination shifts from their positions in the free ligand (AS = 1.02 and 0.79, respectively). By virtue of integration, the 2 peaks which grow in at low temperature are thus due to the 2 methyl groups on the free anilinyl ring associated with the Pd-bound P-atom. The disappearance of these peaks at r.t. can be explained by assuming that PdC^PAf-dmapm) is in rapid equilibrium with the 5-coordinate complex PdChCP.Af.AT-drnapm), in which both N-atoms associated with a single P-atom are bound to Pd. Such 5-coordinate MX2(P,N,N') complexes have been proposed by Xie in the cis-trans isomerisation of PtI 2(PNi)2. 3 9 The equilibrium for the dmapm system is illustrated by the top and bottom lines of Scheme 3.2. 89 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction I 1 1 1 ' I i i i i | i i i i | i i i i | i i i i | i i M | i i i i | i i i i | i i i i | i i i i | i i i i | M i i | i n i | i I I i 4-2 4-0 3.8 3.6 3.4 3.2 3.0 2.8 2.6 2.4 2.2 2.0 1.8 1.6 Figure 3.1 Temperature-dependence of the N C H 3 region of the ' H N M R (400 MHz, CDCI3) spectrum of an equilibrium mixture of PdCl2(P,P '-dmapm) and PdCl2(P,A^-dmapm). 90 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Scheme 3.2 -N ^ C l , P , N 2 "CI P 2 N 3 N 4 P 2 N 3 N 4 vh-P \ > r 1 ;pd ;pd p 2 n 3 PiNiN 2 X I x i X I / P d \ P ^ N , ^ c i X I P 2 N 3 N 4 X I P 1 N 1 N 2 CI ;pd X I X l P 2 N 3 N 4 P 2 N 3 N 4 ;Pd P 2 N 4 ;pd P 2 N 4 PiN,N 2 X I X I X I P,N,N2 ^ C l X I X I Scheme 3.2 outlines all the possible intramolecular associative mechanisms for the interchange between ?dCh(P,P -dmapm) and PdCbCP/V-dmapm). In this scenario, the 5-coordinate intermediates are present in concentrations beneath the level of detection by N M R spectroscopy and the overall equilibrium PdCh(P,P '-dmapm) PdC^CP./V-dmapm) is slow on the N M R timescale. The top and bottom rows of the scheme correspond to stereo inversions at the P-atoms, while the left and right columns represent exchange of the P-atoms. The PdCl2(P,./V-dmapm) complexes represented at the corners of the scheme are indistinguishable in an achiral medium. 91 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction If all the equilibria are operative, all N-atoms on one dmapm ligand should be in chemical exchange with each other, even across isomers. That this is so has been shown by a 2D ' H E X S Y experiment (Figure 3.2). Clearly indicated in this spectrum are off-diagonal peaks which demonstrate chemical exchange between all NC//3 groups. f > « $ '1 * •0" * * • 0 t * •2.5 -3.0 •3.5 |_PP«I 1 r " ppm ^ — 3.5 3.0 2.5 Figure 3.2 The 2D ! H E X S Y spectrum (300 MHz, CDC1 3 , 300 K) of the N C H 3 region of an equilibrium mixture of PdCl 2(P,P -dmapm) and PdCl 2(P,N-dmapm). From the relative intensities of the peaks at 5 2.35 and 3.70 in the spectra shown in Figure 3.1, equilibrium constants at each temperature can be calculated and the van't 92 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Hoff plot shown in Figure 3.3 constructed. This yields AH° = -5.5 ± 0.5 kJ mol - 1 and AS° = -10 ± 1 J K" 1 mol"1 for the isomerisation process. 2.00 i 1.75 -1.50 -1.25 -1.00 -| 3.2e-3 3.4e-3 3.6e-3 3.8e-3 4.0e-3 4.2e-3 4.4e-3 4.6e-3 4.8e-3 1/T(K"1) Figure 3.3 Van't Hoff plot for the PdCl 2(P,P '-dmapm) PdC^P./V-dmapm) equilibrium. 3.3.2 Four-membered ring strain A direct comparison between Pd-PPh3 and Pd-py bond strengths was made by Partenheimer and Hoy 4 0 who measured the heats of reaction for the following substitution processes in CH2CI2 solution: ;ra«s-PdCl2(PhCN)2 + 2 PPh 3 -> trans-FdC\2(V?h3)2 + 2 PhCN A H = -163 kJ mol"1 /ra«s-PdCl 2(PhCN) 2 + 2 py ->• ;ra«s-PdCl 2(py) 2 + 2 PhCN A H = -109 kJ mol"1 From these experiments, the Pd—PPh 3 bond is stronger than the Pd—py bond by 27 kJ mol"1. It is therefore reasonable to infer that the small magnitude of A H for the PdCbCP.-P'-dmapm) ; s = = S : PdCUCPAf-dmapm) equilibrium arises because the energy cost of breaking the stronger Pd—P bond is almost equal in magnitude to the sum of the energy recovered from the formation of the weaker Pd—N bond and alleviation of the 4-membered ring strain. Thus, if the 5-membered P,N ring is assumed to be completely strain-free, the 4-membered ring strain energy can be estimated to be ca. 32 kJ mol"1. Nolan and coworkers have calculated the ring strain energies associated with the binding of chelating tertiary diphosphine ligands (P-P) to " (CsR^RuG" (P-P = dppm, 93 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction dppb, dppe, dppp; R - H , 4 1 Me 4 2 ) by comparison of the heats of reaction for the following substitution reactions in THF at 30 °C: (C 5R 5)RuCl(cod) + P-P -> (C 5R 5)RuCl(P-P) + cod (1) (C 5R 5)RuCl(cod) + 2 PPh 2Me -> (C 5R 5)RuCl(PPh 2Me)2 + cod (2) The ring strain energies thus calculated (AH(i) - AH( 2 )) for dppm are 42 and 56 kJ mol"1 for R = H and R = Me, respectively, in reasonable agreement with that found for PdCl 2CP,P '-dmapm). 3.3.3 Structure and bonding The crystal structure of PdCl^PAf-dmapm) is shown in Figure 3.4. Relevant bond distances and angles are given in Table 3.1. The most interesting aspect is that the structure allows a direct comparison between both bound and free N - and P-atoms of equivalent type and thus permits an assessment of the relative degrees of a- and n-bonding between these atoms and the metal centre. C(29) Cl(2) Figure 3.4 ORTEP representation of the molecular structure of PdCl2(P,Af-dmapm) (50 % ellipsoids). 94 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Table 3.1 Selected bond lengths (A) and angles (°) for PdCl2(P,/V-dmapm) with estimated standard deviations in parentheses. Pd(l)—P(l) 2.1798(6) P(2)—C(18) 1.847(2) Pd(l)—N(l) 2.132(2) P(2)—C(26) 1.845(2) Pd(l)—Cl(l) 2.3041(6) N ( l ) - C ( 6 ) 1.481(3) Pd(l)—Cl(2) 2.3812(6) N ( l ) - C ( 7 ) 1.492(4) P ( l ) - C ( l ) 1.801(2) N ( l ) - C ( 8 ) 1.490(3) P(l)—C(9) 1.806(2) N(3)—C(23) 1.425(3) P( l ) -C(17) 1.824(2) N(3)—C(24) 1.467(3) P(2)—C(17) 1.873(2) N(3)—C(25) 1.453(3) P(l)—Pd(l)—N(l) 86.10(6) P(l)—Pd(l)—Cl(l) 87.93(2) Cl(l)—Pd(l)—Cl(2) 92.63(2) N(l)—Pd(l)—Cl(2) 93.54(6) P(l)—C(17)—P(2) 108.4(1) C—P(l)—C 107.4 (av.) C—P(2)—C 100.5 (av.) The variations in P—C bond lengths on coordination of dmapm to Pd(II) can only be rationalized in light of the relatively recent hypothesis that the P-based 7i-acceptor orbitals (the LUMOs) are not purely 3d, but have significant 3p character and local a* symmetry with respect to the P—C bonds. Calculations by Xiao et al. on PH3, PMe3 and PF 3 demonstrate that such a L U M O arises from mixing of the P-based 3d and 3p orbitals (with the 3p component being largest for PF3). 4 3 The hybrid orbital is oriented such that it is able to accept 71-back donation from the metal ion. Additional calculations by Marynick even show that a qualitative understanding of Tt-back bonding can be attained without even including the d-orbitals in the basis set, and prove, by examination of the phase relationships between P, H and Cr orbitals in the complex Cr(NH3)5(PH3), that the L U M O has local o* symmetry with respect to the P—H bonds.4 4 In order to find physical evidence for these hypotheses, Orpen and Connelly have compared the M — P and P—C bond lengths and C—P—C bond angles for 24 structurally characterized redox-related pairs of transition metal phosphine and phosphite complexes and have demonstrated that in going from the lower to the higher oxidation state the following general trends hold true: the Pd—P distance increases, the P—C distance decreases and the C—P—C angle increases.45 That the P—C distance should decrease is in accord with Marynick's assertion that the phosphine L U M O is o* with 95 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction respect to the P—C bonds: on oxidation, the degree of M—>P 7i-bonding is necessarily decreased because of a lowering of electron density at the metal and thus there is a depopulation of the phosphine L U M O and consequently a strengthening of the P—C bond. This 7i-back bonding explanation for the relative geometries of redox pairs does not hold for comparison of bound and free P-atoms in PdCl2(P,/V-dmapm). One might expect that a direct analogy could be drawn between a bound P-atom and the P-atom(s) of any of the reduced species in Orpen's study, and between a free P-atom and the P-atom(s) of any of the corresponding oxidized species. The reason for this expectation is that 7r-back donation is expected to be greater for a coordinated P-atom than for a free P-atom, as it is for a P-atom coordinated to a reduced species than for a P-atom coordinated to an oxidized one. This assumption would lead to the prediction that P—C bond lengths should increase upon coordination of dmapm to Pd, whereas the opposite is observed. The P—C bond lengths involving the bound P-atom, both for P—C(aromatic) (av. 1.804 A) and P—C(aliphatic) (av. 1.824 A) are significantly shorter than those involving the free P-atom (av. 1.846 and 1.873 A, respectively). Problems such as this are tackled in another study by Orpen's group which compares 1860 unique crystallographically characterised ZPPI13 fragments where Z may be a transition metal, main group metal or non-metal.46 The study describes the molecular orbital energy levels of PPI13 qualitatively in terms of a Walsh correlation diagram which has a trigonal plane (Z>3h) and trigonal pyramid (C3v) as its extreme geometries (Figure 3.5), a model that was originally considered by Gimarc 4 7 and Albright et al.48 Orpen and coworkers propose that when the potential for 7t-back bonding exists, the PPI13 ligand responds to give a lower overall energy by adopting a more pyramidal geometry, enabling better overlap between the P 3d and P-C a* orbitals which together constitute the 71-acceptor LUMOs (2e in Figure 3.5). This leads to a compression of the C—P—C angles and a lengthening of the P—C bonds. Conversely, on coordination to a strong o-acceptor (such as H + ) the tendency is for the PPI13 to assume a more planar structure with larger C—P—C angles giving better overlap for P—C o-bonding and shorter P—C bonds. 96 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction F i g u r e 3.5 Walsh correlation diagram (adapted from Dunne et al.46) for the deformation of a PR3 species, showing rehybridisation of the phosphine lone pair 2a\, and 71-acceptor function 2e, on going from trigonal planar (D3h) geometry (left) to pyramidal ( C 3 V , right). It is therefore more reasonable to rationalize the Pd-P bonding in VdCh(P,N-dmapm) in terms of a dominant P—>M a-contribution. This view of the bonding conforms to Orpen's observation that the relative importance of 71-bonding with respect to a-bonding decreases on going from left to right across the transition series (due to metal d-97 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction electrons becoming increasingly tightly bound).46 Indeed, in addition to having a significantly shorter average P—C bond length, the bound P-atom of PdCl2(P,/V-dmapm) also has an average C—P—C angle (107.4 °) which is much larger than that found at the free atom (100.5°). In addition, the N — C bonds, both for N—-C(aromatic) (av. 1.481 A) and N — C(aliphatic) (av. 1.491 A) involving the bound N-atom are significantly longer than those involving the free (av. 1.430 and 1.463 A, respectively). Nitrogen does not have d-orbitals available for bonding and therefore the observed increase in N — C bond lengths upon coordination points to the fact that the N-based "lone pair", typically thought to be an sp3 orbital, is not entirely non-bonding but has significant N — C a character. Depopulation of this orbital on coordination would induce a lengthening of the N — C bonds. Other aspects of the structure are also worthy of mention: (1) The Pd(l)—Cl(l) bond is shorter than the Pd(l)—Cl(2) bond because of the higher trans influence of P with respect to N . 4 9 (2) The Pd(l)—P(l) bond (2.1798 A) is unusually short. As a comparison, the mean Pd—P distances for PMe3, PPI13, PPhMe 2, dppe and dppm complexes calculated by Orpen et al. are 2.287, 2.308, 2.253, 2.260 and 2.258 A, respectively.50 (3) The P(l)—C(17)—P(2) angle is a relaxed 107.4°, intermediate between the compressed P—C—P angles commonly found in 4-membered P,P'-metallacycles (93.0 ° for PdCl2(dppm)5 1) and those found in complexes containing r^-dppm (118.9 0 for trans-Ru(n1-dppm)2(^,0-quin)2, quin = 2-quinaldinate anion5 2). (4) The P(l)—Pd(l)—N(l) "bite" angle is 86.1 °. This similar to the corresponding angles found in Ru complexes of P M A. These have been determined for both 5- and 6-coordinate Ru(II) complexes to lie in the range 80-83 °. 5 3> 5 4 The average P—C and N — C bond lengths for the crystallographically characterised anilinyldiphosphine ligands and complexes discussed in this dissertation are collected in Table 3.2. 98 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Table 3.2 Average P — C and N — C bond distances (A) for the crystallographically characterised anilinyldiphosphine ligands and their complexes.3 Bond dmape dmapcp £ w o + o £ u •+-> Ph 1—1 + rs o o e £ u Ph I u T3 Ph Pf- — C a i 1.854 [2] 1.881 [2] - - - 1.873 [1] -Pf- Car 1.849 [2] 1.842 [4] - - - 1.846 [2] -Pb-— C a l - - 1.826 [2] 1.835 [2] 1.840 [ l ] b 1.824 [1] 1.839 [2] 1.820 [ l ] c Pb- Car - - 1.821 [4] 1.811 [4] 1.810 [2] b 1.804 [2] 1.819 [4] 1.813 [2] c N f —Cal 1.457 [8] 1.460 [8] 1.462 [8] 1.472 [6] 1.470 [4] 1.463 [6] 1.454 [4] N f Car 1.427 [4] 1.434 [4] 1.433 [4] 1.438 [3] 1.438 [2] 1.430 [3] 1.440 [2] N b -— C a i - - - 1.493 [2] 1.495 [2] 1.491 [2] 1.493 [4] N b - Car - - - 1.481 [1] 1.483 [1] 1.481 [1] 1.458 [2] a The number of each type of observation appears in square brackets, f = free, b = bound, ar = aromatic, al = aliphatic; data are excerpted from Sections 2.3.3, 2.4.1.1, 2.4.2.2, 2.6.1.1, 3.3.3 and 3.4.5.1. b P-atom coordinated to Pd. c P-atom bound to O. The entries in Table 3.2 suggest that the conclusions drawn about the bonding in PdCl2(P,./V-dmapm) can perhaps be extended to encompass the bonding of the anilinyldiphosphines to late transition metals and non-metals generally. From the table, it is evident that: (1) In all cases, whether the "metal" is Pd11, Pt", Pd 1 or " O " , the P b — C a r and P b — C a i bond lengths are shorter than the corresponding lengths involving the free P-atoms. This implies that the o-contribution to the bonding is more significant than the 7t-contribution. (2) In all cases, the N b — C a i and N b — C a r are longer than the corresponding lengths involving free N-atoms implying that the N "lone pair" is not purely non-bonding but has a significant N — C a component. 99 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.3.4 Reactions with halide and cyanide PdCl2(dmapm) reacts with an excess of NaBr (Section 3.9.2) or Nal (Section 3.9.3) in acetone/water or acetone to give the bromo or iodo analogue. Like the parent chloro compound, both exist as an equilibrium mixture of the P,P- and P,/V-isomers in CDCI3 solution. The ~PdX2(P,P '-dmapm) T)dX2(P,N'-dmapm) equilibrium constants at 250 K for the series are 4.1, 4.8 and 6.3 for X = CI, Br and I, respectively. Pringle and Shaw discovered that PdCl2(dppm) reacts with NaCN to give the dppm-bridged dimers [?ra«5-Pd(CN)2(jj,-dppm)]2.55 In contrast, PdCi2(dmapm) reacts with 2 equiv. of K C N to give exclusively the monometallic compound, Pd(CN)2(P,./V-dmapm) (Section 3.9.4), i.e., because the P,P'-isomer is present in CDCI3 solution at a concentration lower than the limit of detection by N M R spectroscopy, K > 100; however, PtCl 2(P,P-dmapm) reacts to give both Pt(CN)2(P,P'-dmapm) and Pt(CN)2(P,/V-dmapm) (Section 3.9.10). This result is in accord with the reaction of dmapm with PtCl2(cod) which generates PtCh(P,P '-dmapm) exclusively, showing that Pt has a higher affinity than Pd for P. Thus, the magnitude of K for the PdX2(P,P'-dmapm) PdX2(P,7V'-dmapm) equilibrium correlates with the position of X in the trans effect series (CI < Br < I « CN). Because P-donors also show a strong trans effect, PdX2(P,/V-dmapm) should be progressively more stable than PdX2(/>,P'-dmapm) as the trans effect of X increases. 3.3.5 Reactions with peroxide and S 8 PdC^fdmapm) reacts with cumene hydroperoxide in CH2CI2 or H2O2 in CH2C1 2/H 20 at r.t., and with S 8 in CH 2C1CH 2C1 heated to reflux, to give PdChCP^-dmapmO) (Section 3.9.6; the molecular structure of the analogous PtbCP/V-dmapmO) is given in Figure 3.18, p. 132) and [PdCl(P,7v;S-dmapmS)]Cl (Section 3.9.8) respectively (Chart 3.2, Ar = o-C6H4NMe2). The S-atom of the P=S fragment being a "soft" donor is a sufficiently good ligand to displace CI" from the metal coordination sphere, whereas the "harder" O-atom of the P=0 fragment is not. PdCi2CP,/V-dmapmO) is a non-electrolyte in CFi2Cl 2 solution ( A M < 1 cm 2 ohm"1 mof 1) and does not dissolve in H 2 0 , while [PdC\(P,N,S-dmapmS)]Cl dissolves in Ff 20 as a 1:1 electrolyte ( A M = 99 cm 2 ohm"1 mol" 1), 5 6 and the CI" counterion is easily exchanged for PF6" by reaction with KPF6 (Section 3.9.7). 100 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Chart 3.2 3.3.6 PdCl(Me)(/>,7V-dmapm) In contrast to PdC^Cdmapm) which exists as an equilibrium mixture of P,P'- and P,N-isomers in CDCb solution, the title complex (Section 3.9.5) exists solely as a P,yV-isomer (Chart 3.3; Ar = o-C 6 H 4 NMe 2 ) . This is easily ascertained from the 3 1 P{'H} N M R spectrum which shows the presence of both bound and free P-atoms (5 24.0 (d), -39.3 (d), 2JpP = 130 Hz), the large 2 JPP value being reminiscent of that of PdCl2(P,iV-dmapm) (108 Hz). The P,N bonding mode can be ascribed not only to the need for 4-membered ring strain (ca. 32 kJ mol"1) to be overcome in order for the P,P' isomer to form, but also to the fact that a trans arrangement of the strong field ligands P and Me is highly disfavoured. In fact, even i f 4-membered ring strain is eliminated, e.g., in the reaction between PdCl(Me)(cod) and dmape (which contains two C-atoms in its "bridge"), a mixture of the P,N- and P,P'- isomers is produced ( 3 1P{'H} N M R : PdCl(Me)(P,P'-dmape) 5 28.2 (d), 30.2 (d), 2JPP = 63 Hz. PdCl(Me)(P,/V-dmape) 5 28.0 (d), -27.9 (d), V P P = 38 Hz). 101 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Chart 3.3 ^ / / N M e 2 P P d — C I < P A r 2 M e 3.4 Complexes Containing Two Metal Centres 3.4.1 Reactions with */Ywis-PdCl2(PhCN)2 and K 2 PtCl 4 The synthesis of homo- and heterobimetallic complexes containing bridging dmapm depends both upon the presence in the monometallic precursor of a "dangling" P-atom and the on existence of the P.P'^^ P,N equilibrium. Thus, PdCl2(dmapm) reacts with ;ra«s-PdCl2(PhCN)2 in CH 2 C1 2 at r.t. (Section 3.9.11) or with K 2 P t C l 4 in EtOH/H 2 0 at 70 °C (Section 3.9.13) to form Pd2Cl4(dmapm) or PtPdCl4(dmapm), respectively (Chart 3.4; Ar = o-C6H 4NMe 2, M = Pd, Pt. The molecular structure of the iodo analogue PtPdL)(dmapm) is given in Figure 3.17, p. 132) Were it not for the requisite equilibrium, such reactions would result in a less than quantitative yield of the bimetallic product as a significant proportion of the precursor would remain unreacted. The mixed metal compound PfPdCl4(dmapm) could not be made by reaction of PdCl2(dmapm) with either of czs-PtCl 2(MeCN) 2 or PtCl2(cod). In addition, PtPdCl4(dmapm) could not be made from PtCl 2(P,P '-dmapm) and trans-Chart 3.4 CI CI CI CI 102 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction PdCl2(PhCN)2 because of the absence in the monometallic precursor of a dangling P-atom. This means that under these conditions, it is not possible to make homobimetallic dmapm-bridged Pt" complexes. However, analysis of the off-white product mixture from the reaction between Pt(CN)2(dmapm) and K 2 P t C l 4 in EtOH/H 2 0 at 70 °C by 3 1 P{'H} N M R spectroscopy revealed that Pt2Cl2(CN)2(dmapm) was generated in low yield ( 3 1P{'H} N M R (300 MHz, CDC1 3, 300K): 5 12.9 (s, ]Jm = 2730 Hz, P trans to CN), 7.3 (s, 1 Jppt = 4070 Hz, P trans to CI)). This route was not pursued as it was felt that C N -containing complexes would generally not make for good catalysts. Single crystals of Pd2Cl4(dmapm) were grown on two occasions and analysed by X-ray crystallography. Unfortunately, the molecular structure could not be refined to acceptable quality because of a twinning problem. Nevertheless, the analysis established the essential geometry of the complex to be analogous to that given in the PLUTO representation of PtPdL^dmapm) (Figure 3.17, p. 132). In contrast to the large coupling between inequivalent P-atoms in PdCUCP.A7''-dmapm) (108 Hz), no 2-bond PP coupling is observed in the 3 1 P{'H} N M R spectra of Pd2Cl4(dmapm) and PfPdCl4(dmapm). In addition, although both of the P-atoms in the bimetallic complexes are chiral, no diastereomers are observed in either the 3 1 P{'H} or ' H N M R spectra. The N C H 3 region of the 5 H N M R spectrum of e.g. PtPdCL(dmapm) contains 8 distinct peaks, each corresponding to 3 protons (Figure 3.6). These represent diastereotopic methyl groups associated with both the bound and free N-atoms on the Pt and Pd "sides" of the molecule. The peak shapes suggest that there is either slow or no exchange of coordinated and free anilinyl N-atoms, and imply thereby that diastereomers should be apparent. 103 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction c a b d T i | i i i I | i i i i | i i i — i — | — I — I — I — i — | — i — I — I — i — | — I — I — i — i — | — i — i — i — r 3.40 3.20 3.00 2.80 2.60 2.40 2.20 Figure 3.6 The N C H 3 region of the ' H N M R (400 MHz, CDC1 3 , 300 K) spectrum of The assignment of the peaks in Figure 3.6 is based on chemical shift, integration and shape (see below), and is illustrated in Chart 3.5 (Cl-atoms have been omitted for clarity). A ' r l 2D E X S Y measurement of PtPdCl4(dmapm) revealed that the free and bound N-atoms are in fact in chemical exchange. The spectrum demonstrates that the methyl group a is in exchange with c, b with d, e with h and f with g. Furthermore, the spectrum shows that there is no exchange between the N-atoms associated with opposite "sides" of the molecule, i.e., between groups a-d and e-h. The fact that no diastereomers of this compound or its dipalladium analogue are observed may be due to the fact that the exchange occurs in a concerted way that generates enantiomers only. If so, the rate of exchange at Pt and Pd must be equal, in contradiction to the well established greater PtPdCl4(dmapm). Chart 3.5 104 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction lability of Pd". In addition, peaks e-h appear broader than a-d implying that the protons they represent are in a fluxional process which occurs faster than it does in the former group. The broader peaks are likely to be due to the N C H 3 protons associated with the Pd "side" of the molecule. 3.4.2 Reaction with NH 4 PF 6 In addition to dmapm-bridged bimetallic complexes, the chloride-bridged dimer [Pd(u-Cl)(/>,P'-dmapm)]2[PF6]2 was also synthesised (Section 3.9.14). In an attempt to make a Pd(II)-Rh(I) dimer, PdCl2(dmapm) was stirred with [Rh(u-Ci)(cod)]2 in the presence of K P F 6 in a two-phase CH2CI2/H2O mixture, a standard methodology for making Rh(I) phosphine complexes.57 However, no new Rh-containing species were produced and the Pd complex was cleanly converted to a species whose 3 1 P{'H} N M R spectrum consisted of a singlet at 5 -52.0 and a septet at 5 -145.0 (PFg") and whose elemental analysis was consistent with the formulation "PdCl(dmapm)(PF6)". In addition, the ' H N M R spectrum demonstrated the chemical equivalence of all 24 NMe protons by the existence of only 31 one singlet (5 2.56) in the appropriate region. The upfield P singlet is indicative of chemically equivalent P-atoms involved in a 4-membered metallacycle, and the ' H singlet at 5 2.56 shows that none of the N-atoms are coordinated. This compound was also made successfully in a rational manner via the reaction of PdCl2(dmapm) and N H 4 P F 6 in acetone and was shown by its conductivity in C H 3 N O 2 to be a 2:1 electrolyte ( A M = 213 cm 2 ohm'1 mol" 1). 5 6 3.4.3 Reaction with [Rh(u-Cl)(CO)2]2 Although it does not react with [Rh(u-Cl)(cod)]2 under the conditions tested, PdCl2(dmapm) reacts rapidly with [Rh(u-Cl)(CO)2]2 in CH 2 C1 2 solution at r.t. (Section 3.9.15) to give a complex whose elemental analysis is consistent with the formulation PdRhCl3(CO)(dmapm). This compound is proposed, on the basis of IR and N M R data, to have.the structure illustrated in Chart 3.6 (Ar = o-C 6 H 4 NMe 2 ) . Its 3 1 P{'H} N M R spectrum comprises a singlet at 8 28.1 due to the P-atom bound to Pd, and a doublet centred at 8 41.7 due to the P-atom bound to Rh ( ! J p R h = 179 Hz). As in the case of 105 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction PtPdCl^dmapm), neither 2-bond PP coupling nor diastereomers were observed in the N M R spectra. Chart 3.6 // \\ M e 2 N X P P N M e 2 P d Rh / \ 7 \ CI CI OC CI The solid-state IR spectrum of PdRhCl3(CO)(dmapm) reveals a terminal carbonyl (oco = 1975 cm"1), assumed still to be bound to Rh as Pd has a well-established low affinity for CO, especially when it is terminal. In addition, this arrangement alleviates the necessity for ligand reorganisation on Pd during the synthesis of the compound. Finally, the CO ligand is presumably trans to N because this configuration lowers competition for electron density between the strong-field ligands P and CO. PdRhCl3(CO)(dmapm) dissolves in boiling water to give a compound whose molar conductivity is 363 cm 2 ohm"1 mol"1 (once the water has been cooled to 25 °C), consistent with the formulation [PdRh(OH2)3(CO)(dmapm)]Cl3. The neutral complex is a non-electrolyte in CH3NO2 solution, and does not react with 1 atm H 2 in CDCI3 solution. 3.4.4 Rh2Cl2(CO)2(dmapm) The ligand dmapm reacts cleanly and briskly with [Rh(CO)2(u.-Cl)]2 in CH2CI2 solution at r.t. to give Rh 2Cl 2(CO) 2(dmapm) (Section 3.9.16). The 3 1 P{'H} N M R spectrum of this complex is given in Figure 3.7. 106 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction i 1 • • • i 1 • • • i i • i i • • • • i 44 42 40 38 36 ppm Figure 3.7 The 3 1 P{'H} N M R (162 MHz, CDC1 3, 300 K) spectrum of Budzelaar et al. have reported the crystal structure of Rh2Cl2(CO)2(pyPhPCH2PPhpy) (py = 2-pyridyl), the anilinyl analogue of which is depicted in Chart 3.7 (centre).58 The N M R coupling constants for this compound are: 'jpRh = 101, 2Jp P = 67, VpRh = 9 Hz. The complex pattern is a result of magnetic inequivalence of the P-atoms. The corresponding values for Rh 2Cl 2(CO) 2(dmapm) are 180, 22 and 1 Hz, respectively, the last having been determined by simulation.5 9 The ' H N M R spectrum of Rh2Cl2(CO)2(dmapm) at 300 K is broad and uninformative. At 220 K , the N C H 3 region of the spectrum consists of 4 singlets of equal intensity which signify 2 bound and 2 free N-atoms. These data are consistent with all of the structures depicted in Chart 3.7. Rh 2Cl 2(CO) 2(dmapm). 107 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.4.5 Complexes containing M - M bonds 3.4.5.1 Reaction of PdCl2(dmapm) with Pd2(dba)3 Reaction of PdCl2(dmapm) with 0.5 equiv. of Pd 2 (dba) 3 CHCl 3 in refluxing CH 2 C1 2 generates the Pd 1 dimer Pd2Cl2(dmapm) in 70-80 % isolated yield (Section 3.9.17). This represents a minor modification to the now widely-adopted general method of Balch and Benner6 0 for the preparation of phosphine-bridged Pd 1 dimers such as Pd 2Cl 2(u-dppm) 2; 6 1 the procedure is a significant improvement over the original synthesis involving reaction of [Pd(CO)Cl]n and dppm 6 2 which has been reported to give variable yields. 6 1 Prior to the determination of the crystal structure of Pd2Cl2(dmapm), its geometry could be inferred from 3 1 P{ 1 H} and ' H N M R spectra and from known structures of 31 1 analogous complexes. The P{ H} singlet at 6 -29.9 indicates 2 chemically equivalent P-atoms in the product. The N C H 3 ' H peaks appear as singlets at 8 2.44, 2.89 and 3.07 with relative integrations of 2:1:1, respectively, i.e., 12:6:6 protons. By analogy to PdCl^P/V-dmapm), the two downfield singlets correspond to two diastereotopic sets of methyl protons associated with bound N-atoms (sets a and b in Chart 3.8) and the more upfield singlet is due to the methyl groups attached to the free N-atoms (set c in Chart 3.8). The chemically equivalent C H 2 protons appear as a triplet at § 3.72 because of coupling to 2 identical P-atoms. The crystal structure of the related complex Pd2Br2(dppm)2 has been determined, and consists of 2 interpenetrating square planes mutually twisted by 39 ° . 6 3 This structure 108 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction maintains the chemical equivalence not only of the 4 P-atoms, but also of the 4 CH2 protons (by virtue of a C2-axis which bisects the Pd—Pd bond and contains the methylene C-atoms). A similar motif seemed very likely in the structure of Pd2Cl2(dmapm), and indeed an X-ray crystallographic study confirmed the structure to be that shown in cartoon form in Chart 3.8. Chart 3.8 Molecular structure of PdjChfdmapm) Slow evaporation of a CDC1 3 solution of the .complex in an N M R tube resulted in the deposition of orange, irregular crystals of Pd2Cl2(dmapm)-2CDCl3. The molecular structure is given in Figure 3.8, and selected bond distances and angles appear in Table 3.3. 109 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction C(28) C(l l ) Figure 3.8 ORTEP representation of the molecular structure of Pd2Cl2(dmapm) (50 % ellipsoids). H-atoms have been omitted for clarity. Table 3.3 Selected bond distances (A) and angles (°) for Pd2Cl2(dmapm)-2 CDC1 3 with estimated standard deviations in parentheses. Pd(l)--Pd(2) 2.527(1) Pd(l> - C l ( l ) 2.384(4) Pd(l)-- P ( l ) 2.153(4) Pd(l)-- N ( l ) 2.23(1) Pd(2)--Cl(2) 2.364(3) Pd(2)--Cl (2) 2.184(3) Pd(2)--N(3) 2.26(1) P ( l ) - -Pd( l ) - -N(l) 85.8(3) N ( l ) - -Pd( l ) - -Pd(2) 164.4(3) P(2)- -Pd(2)--N(3) 86.3(3) N(3)--Pd(2)--Pd(l) 165.5(3) P ( l ) - -Pd( l ) - -Pd(2) 78.7(1) P(2)- -Pd(2)--Pd(l) 86.99(9) c i ( i ) --Pd( l ) - -Pd(2) 98.86(9) Cl(2)--Pd(2)--Pd( l ) 92.12(9) P ( l ) - -C(17)--P(2) 102.9(6) 110 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction At the core of Pd2Cl2(dmapm) lie two interpenetrating Pd square planes mutually canted by 59.9 °; Figure 3.9 shows perpendicular views of the metal coordination spheres. This structural motif has been described previously as "T-over-square" and is most perfectly exemplified by the homoleptic Pd 1 dimer [Pd 2(NCMe)6]2+, first prepared by Eisenberg and coworkers, 6 4 ' 6 5 in which the dihedral angle between the planes is 90 °. The Pd—Pd bond length in Pd2Cl2(dmapm) is 2.527(1) A, the shortest observed for a neutral Pd 1 dimer and the second shortest ever after [Pd2(o-Ph2PC6H4CH20(CH2)3-o-py) 2][BF 4]2-4Fl20 (2.500(1) A),66. Structurally characterised Pd 1 dimers supported by tridentate ligands are relatively rare. The other example is [Pd2(Ph2PC2H4P(Ph)C2H4PPh2)2][BF4]2 which has a Pd—Pd length of 2.617(1) A. 6 7 Pd2Cl2(dmapm) represents, to this author's knowledge, the first example of a bimetallic Pd' complex supported by a tetradentate ligand. Figure 3.9 The coordination environments of the Pd centres in Pd2Cl2(dmapm) viewed along (left) and perpendicularly to (right) the Pd—Pd axis. Figure 3.9 (left) clearly demonstrates that the P d—N bonds are not collinear with the Pd—Pd bond but are "bent back" by ca. 15 0 from this axis. In addition, the P—Pd— Pd angles are both less than 90 °; one (78.7 °) is substantially smaller than the other (86.99 °) which gives rise to unexpected asymmetry in the molecule. This tilting of the Pd—P bonds towards each other is indicative of compressive strain within the PCH 2 P "backbone" of the dmapm ligand. Indeed, the P—C—P angle found for PdCl2(P,A^-dmapm) which is assumed to represent the "strain free" state (108.4 °) is larger than that found in Pd2Cl2(dmapm) (102.9 °). I l l References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction The steric bulk presented by dmapm in this complex rests largely behind the Pd— Pd bond which, protected only by the chloride ligands, is exposed to electrophilic attack (Figure 3.10). This structural characteristic is atypical of diphosphine-bridged Pd1 dimers which usually bear two of the bridging ligands and whose metal-metal bonds are much more sterically inaccessible. This feature should give rise to differences in reactivity, e.g. with small molecules, between this complex (see Section 3.4.5.3) and the more typical Pd 2 X 2 (P-P) 2 61,68-70 o r pd 2 X 2 (P-N) 2 7 1> 7 2 compounds. Figure 3.10 Space-filling representation of Pd2Cl2(dmapm) showing the lopsided distribution of steric bulk about the Pd—Pd bond. A l l visible, unlabelled atoms are C-atoms. The P - N "bite" angle found for Pd2Cl2(dmapm) is approximately 86 °, the same as that found for PdCl2(P,A^-dmapm). The Pd—CI bond lengths found for the two complexes are the same within error for CI trans to P; the P d—N bond lengths are significantly longer in the former (2.25 A) than in the latter (2.132 A) possibly due to the high trans influence of the metal-metal bond. In the case of the Pd—P bond lengths, one Pd(i) 112 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction of those found for Pd2Cl2(dmapm) (Pd(2)—P(2), 2.184(3) A) is essentially the same as that found for PdCl2(P,7V-dmapm) (2.1798(6) A) while the other (Pd(l)—P(l), 2.153(4) A) is significantly shorter. As in the case of discrepancy between the P—Pd— Pd angles, there is no obvious reason for the asymmetry within the Pd2Cl2(dmapm) molecule, but crystal packing effects can be invoked. The absolute configurations of the P-atoms in the ORTEP shown in Figure 3.8 are R,R. Because the space group of the crystal (C2/c) contains a glide plane, the opposite enantiomer must also be present in equal abundance in the unit cell. The diastereomeric R,S form is not present, however. In addition, the 300 K 3 1 P{'H} N M R spectrum of this compound consists of a lone singlet which indicates that there is either rapid exchange between diastereomers, or the molecule is formed stereoselectively. In order to shed light on this question, two N M R experiments were conducted. In the first, 3 1 P{'H} N M R spectra were acquired at successively lower temperatures in order to "freeze out" the putative diastereomeric exchange; no splitting of the singlet was observed. This indicates either that the exchange between diastereomers is facile and occurs rapidly even at 220 K, or that there is no exchange. Because the mechanism of diastereomeric switching must proceed through exchange of free and coordinated N-atoms, a 2D ' H E X S Y experiment was also carried out at 300 K. The resulting spectrum shows no cross peaks between the NCH3 groups of the bound and the free N-atoms. This evidence definitively rules out diastereomeric exchange and proves that the Pd2Cl2(dmapm) molecule is formed stereoselectively. These results are very different from those obtained by Xie et al. for the complex Pd 2 Cl 2 (u-PN 2 ) 2 . 7 3 This compound exhibits a broad singlet in its r.t. 3 IP{ 1F£} spectrum which transforms into two closely separated sharp singlets at 253 K indicating the presence of diastereomers. In addition, the lack of exchange between free and coordinated N-atoms in Pd2Cl2(dmapm) contrasts with the opposite behaviour observed for the monometallic Pd" compound, PdCl2(dmapm) (Section 3.3.1, Scheme 3.2). 3.4.5.2 Two-electron reductions of bimetallic M(II) complexes In contrast to the reaction between PdCl2(dmapm) and Pd2(dba)3-CHCl3, PdCl2(dmapm) does not react under the conditions tested with the analogous Pt(0) starting material 113 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Pt(dba)2. Instead, the only viable route to the mixed metal Pt'-Pd 1 dimer was via a lower-yielding 2-electron reduction of the Pt ' -Pd" dimer PtPdCl4(dmapm) using hot ethanolic K O H (Scheme 3.3. Section 3.9.18). Scheme 3.3 PtPdCl4(dmapm) + K O H + C H 3 C H 2 O H -> PtPdCl2(dmapm) + KC1 + HC1 + C H 3 C H O + H 2 0 Although several reducing agents have been used in attempts to uncover convenient and reliable syntheses of M 1 dimers, conproportionation reactions between the appropriate M 1 1 and M° complexes in the presence of the bridging ligand give the best results. Pringle and Shaw tested zinc dust, formic acid, hydrazine and sodium borohydride in the synthesis of Pd 2Cl 2(dppm) 2 and the results were generally unsatisfactory.74 Overall yields were typically in the range of 40 % and the desired product was often contaminated with PdCl2(P,P'-dppm) which was difficult to remove. The mixed metal dimer PtPdCl2(dppm)2, first reported by Pringle and Shaw in 1982,75 was made in high yield by addition of the labile Pt" precursor PtCl 2(NC'Bu) 2 and dppm to the red solution resulting from the reaction between Pd(PPh 3) 4 and dppm. Mixed metal PtPd complexes containing bridging PN„ ligands have also been made by the conproportionation route. 3 9- 7 2 Reduction of platinum metal(II) salts by ethanolic K O H is not new (and, for example, is the standard route by which Pt(PPh3)4 is synthesised from K 2 P t C l 4 7 6 ) but to this author's knowledge this is the first occasion in which the method has been used to make a M - M 1 complex from the corresponding M " 2 compound. The 3 1 P { ' H } N M R spectrum of PtPdCl2(dmapm) and its proposed structure are shown in Figure 3.11. The spectrum is typical for a phosphine-bridged PtPd bimetallic compound. 3 9 ' 7 4- 7 5 The P-atoms are chemically inequivalent and give rise to doublets (Vpapb = 21.9 Hz), while P a shows a large one-bond coupling to 1 9 5 Pt ('jPapt = 4200 Hz) and Pb shows a small two-bond coupling ( JPbpt = 260 Hz). 114 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Ar A r Me,N R — P d — N M e , 2 / l 2 CI CI „ 1 . . J T „ . , 11111111| i i i i j i i i i j i i ii in n i i i i i i n n i i i i i | i i i i | i i i i i i i i i i i i i i i i i i i i i in i i m i -12 -14 -16 -18 -20 -22 -24 -26 -28 -30 -32 -34 -36 -38 4 0 -4, 10 I 11 11 II II 1111111 11111 II 111II '2 -44 -46 -48 ' I ' " ' ! " " ! -50 -52 -54 Figure 3.11 The 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) spectrum and proposed structure of PtPdCl2(dmapm). The small peak at 8 -30 is due to trace Pd2Cl2(dmapm) impurity. (Ar = o-C 6 H 4 NMe 2 . ) As for Pd2Cl2(dmapm), the mixed metal complex does not exhibit a "doubling" of N M R spectral peaks which would indicate the presence of diastereomers and, by analogy to the former complex, is most probably formed stereoselectively. The ' H N M R spectrum of PtPdCl2(dmapm) is consistent with the proposed structure. The N C H 3 region shows six sharp peaks in a 1:1:1:1:2:2 ratio at 8 3.19, 3.07, 3.02, 2.78, 2.45 and 2.36. By virtue of their similar positions in the spectrum of Pd2Cl2(dmapm), the singlets at 3.07 and 2.78, and 2.45 can be assigned to the two diastereotopic NCH3 groups of the Pd-bound N-atom and the two chemically equivalent NCH3 groups of the free N-atom associated with the Pd-bound P-atom, respectively. The remaining peaks at 8 3.19, 3.02 and 2.36 correspond to the analogous peaks on the Pt "side" of the molecule. 3.4.5.3 Reactions of bimetallic M 1 complexes Like the well known P,P'- and P,TV-bridged Pd 1 dimers, Pd2Cl2(dmapm) contains a reactive Pd—Pd bond. Although Pd1 dimers of this type typically undergo insertion reactions with small molecules such as CO, 6 1> 6 8> 7 2 ' 7 7 CNMe, 6 ] > 7 7 S 0 2 , 7 8 C S 2 , 7 9 H 2s,69,70,80 H 2 Se 8 1 etc (Chart 3.9, (a); Y = small molecule), there are ligand-dependent 115 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction differences in their reactivities. For example, whereas the P,P '-bridged complexes react with CO to form insertion products of the type Pd 2X 2(u-CO)(u-P-P) 2 (X = halide),61 the P,/v~-bridged complexes such as Pd 2 Cl 2 (u-PNi) 2 (PNi = 2-diphenylphosphinopyridine) react via displacement of the pyridyl groups to form Pd 2 Cl 2 (CO) 2 (PNi) 2 which contains terminal CO ligands and an unsupported metal-metal bond. 7 2 This difference in reactivity is probably due to rigidity in the bridging ligand. As another example, Pd 2 X 2 (u-dppm)2 is unreactive toward C S 2 8 2 but Pd 2X 2(u-dmpm) 2 reacts rapidly to give Pd 2 X 2 (u-CS2)(u-dmpm)2.79 Pd2Cl2(dmapm) and PtPdCl2(dmapm) differ fundamentally from the majority of diphosphine-bridged M 1 dimers in that they bear only one diphosphine ligand which has potentially dissociable anilinyl "arms". The doubly-bridged M 1 complexes give " A -frame" insertion products with small molecules (Chart 3.9, (a)). Because of the putative lability of the N-atoms, MM'Cl 2(dmapm) complexes should be able to form not only the analogous "A-frame" complexes (Chart 3.9, (c)), but also double insertion products (Chart 3.9, (b)). By analogy to the reaction of CO with Pd 2 Cl 2 (u-PNi) 2 , 7 2 another possible reaction pathway is displacement of the anilinyl groups without disruption of the M - M ' bond (Chart 3.9, (d)). Chart 3.9 116 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction In addition, as noted above, the geometries of Pd2Cl2(dmapm) and PtPdChCdmapm) leave the M - M ' bond significantly more exposed than in the case of MM'Cl2(u-P-P)2 (Figure 3.10). Finally, the chloride ligands in the dmapm complexes are trans to P, whereas they are trans to the M - M ' bond in the "more typical" dimers. It was therefore of interest to examine the reactivity of the new bimetallic M 1 complexes with small molecules. Pd2Cl2(dmapm): Reaction with CO. Exposure of an orange CDCI3 solution containing Pd2Cl2(dmapm) to CO (1 atm) resulted in a rapid colour change to intense purple. The in situ 3 I P{ ] H} N M R spectrum consists of a singlet at 5 43.0 and an upfield singlet corresponding to unreacted Pd2Cl2(dmapm) in a ratio of ca. 6:1. The ' H N M R spectrum indicates that all 24 N C H 3 protons of the CO-containing product are equivalent. From this one can conclude that the two coordinated N-atoms in the starting material are displaced by CO. Conducting 3 freeze-pump-thaw cycles on this solution is sufficient to regenerate quantitatively the starting material. A purple solid whose elemental analysis is consistent with the formulation Pd2Cl2(CO)2(dmapm) could be isolated from a CH 2 Cl2/Et 2 0 slurry containing Pd2Cl2(dmapm) which had been exposed to CO (1 atm) overnight (Section 3.9.19). The IR spectrum of this solid showed a single strong uco at 1798 cm"1 which corresponds to a bridging carbonyl. In contrast, Pd 2Cl 2(CO) 2(PNi)2 has uCo bands at 2019 and 1994 cm"1, clearly indicating terminal CO ligands. Thus, the ER and elemental analysis findings lead to the conclusion that the product contains two bridging CO ligands opposite one-another (Chart 3.10). In this arrangement, the symmetric CO stretch is IR invisible and only one CO band is observed. A Pd—Pd bond is included to ensure that each metal centre retains the requisite 16 electrons. Chart 3.10 M e 2 N P 6 \ \ ^ P 6 N M e 2 CI O CI 117 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction PdChfdmapm): Reaction with diethylacetylenedicarboxylate (DEAD). Interest in acetylene complexes has been fuelled by the search for hydrogenation and cyclotrimerisation catalysts.83 For bimetallic complexes, two acetylene coordination modes are known, hi the first, prevalent in carbonyl complexes of the first row transition metals, the acetylene sits perpendicular to the metal-metal axis with the acetylene C-atoms bridging (Chart 3.11, left). 8 4" 8 6 In the second, most commonly observed for phosphine complexes of the second and third row transition metals, the acetylene lies parallel to the metal-metal axis and each C=C C-atom binds one metal atom (Chart 3.11, right). 8 7" 8 9 The crystal structure of #T-Pd 2 Cl 2 ( | i -DMAD)(u-PN 3 ) 2 shows the acetylene to adopt the second "parallel" coordination mode.7 3 In addition, the kinetic aspects of the reaction of D M A D with H, r-Pt 2I 2(u-PN 3) 2 have been examined.71 It should be noted that whereas the "perpendicular" mode is not possible (without either dissociation of diphosphine or invocation of 5-coordinate Pd) for the reaction between acetylenes and Pd 2Cl 2(u-P-P) 2 , the ability of the N "arms" to dissociate makes this mode a possibility in the analogous reaction with Pd2Cl2(dmapm). Chart 3.11 M R The Pd(I) dimer Pd2Cl2(dmapm) reacts with D E A D to give the compound Pd 2Cl 2(DEAD)(dmapm) as a monohydrate (Section 3.9.20). This compound is fluxional in CDC1 3 solution, the ^ P ^ H } N M R spectra at 325 and 220 K being given in Figure 3.12. 118 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction i i : 1 1 1 1 r PP« 45 40 35 30 25 20 Figure 3.12 The 3 1 P{'H} N M R (121 MHz, CDC13) spectra of Pd 2Cl 2(DEAD)(dmapm) at 325 (top) and 220 K (bottom). One possible interpretation of the N M R spectra is according to the process illustrated in Scheme 3.4 (R = C(O)OEt). At higher temperatures, the P-atoms are rendered chemically equivalent by rapid exchange. Coordination of a "dangling" N-atom of dmapm causes the acetylene moiety to be shuttled from one Pd centre to the other with concomitant displacement of the recipient Pd centre's bound N-atom. The two extremes represented in the scheme are, of course, identical and should therefore give rise to identical spectra should the motion be frozen. In the "frozen" state, for example at 220 K, when the shuttling mechanism is slow on the N M R time-scale, the two P-atoms become inequivalent. A 2D 3 1 P E X S Y measurement at 220 K confirmed that the P-atoms of Pd2Cl2(DEAD)(dmapm) are in chemical exchange. 119 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction PtPdChfdmapm) In contrast to the reactions of CO and D E A D with Pd2Cl2(dmapm), the analogous reactions with PtPdCl2(dmapm) at r.t. resulted in disproportionation to PtCl 2(P,P'-dmapm) and Pd metal; this reaction was followed by low temperature 3 1 P{ ! H} N M R spectroscopy, after 1 atm of CO was admitted to an N M R tube containing a CDCI3 solution of PtPdCl2(dmapm) at ca. 230 K. The tube was then allowed to warm in the spectrometer probe to 300 K over 2.5 h with measurements taken at 10 K intervals. A typical spectrum is given in Figure 3.13. b I 1 1 1 1 I 1 1 ' 1 I 1 1 1 1 I 1 1 1 1 — I — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — i — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — i — 1 — r 60 40 20 0 -20 F Figure 3.13 The 3 1 P{'H} N M R (162 MHz, CDC1 3, 253 K) spectrum of PtPdCl2(dmapm) under CO, 90 min after admitting the gas. * denotes PtPdCl2(dmapm) and # signifies a small Pd2Cl2(dmapm) impurity. Other assignments are discussed in the text. 120 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Whereas meaningful data can be extracted from the 3 1 P spectra, peaks in the corresponding ' H spectra at this temperature are broad and uninformative. The following analysis is tentative. The spectrum in Figure 3.13 clearly demonstrates the presence of two new species, both containing Pt and Pd bridged by dmapm and most probably a Pt—Pd bond. The first (a) has the following spectral data: 5 -18.4 (d, V p p , = 3060, V P P = 23.3 Hz, P bonded to Pt), -8.3 (d, 2 y P P t = 300, 2J ? ? = 23.3 Hz, P bonded to Pd). By comparison to the 3 1 P singlet of Pd2Cl2(|i.-CO)2(dmapm) (6 43.0), which shows a downfield shift from that of Pd2Cl2(dmapm) (5 -29.0), species a, not manifesting such a dramatic shift from that of the starting PtPdCl2(dmapm) (5 -31.7, -23.0), may contain only one bridging CO ligand (Chart 3.12, left; Ar = o-C6H 4NMe 2), although this would necessitate breaking the Pt— Pd bond, i f the anilinyl N-atoms were to remain bound, and a quenching of the 2-bond P-Pt coupling would presumably result. Or, a may contain CO bound terminally, most probably to Pt (Chart 3.12, middle), which would maintain the 2-bond P-Pt coupling. The second major new species (b) gives the following values: 8 25.6 (d, 'jppt = 5120, 2JPp = 173 Hz, P bonded to Pt), 46.9 (d, V P P t = 1270, V P P = 173 Hz, P bonded to Pd). The large downfield shift of peaks due to b relative to those of PfPdCl2(dmapm) suggest a structure analogous to that of Pd2Cl2(CO)2(dmapm). Of note, all the coupling constants observed for b are very much larger than those of PtPdCl2(dmapm) ('7ppt = 4200,2Jppt = 260, 2Jp P = 21.9 Hz). Chart 3.12 a a b 121 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction The intermediates a and b began to appear after 30 min at 233 K. At 243 K , their concentrations increased markedly and that of PtPdCl2(dmapm) decreased accordingly. This trend continued as the probe was warmed over the next 60 min to 263 K, when the peak due to PtChCdmapm) (§ -65.2) began to appear. At 283 K, all of the starting material disappeared but a and b persisted. At 300 K, a vanished and b persisted for a brief period (ca. 10 min) before it, too, decomposed and the 3 1 P N M R spectrum indicated PtCl2(dmapm) to be the sole P-containing product. 3.5 A General Introduction to Cluster Catalysis A cluster catalyst is a complex containing two or more identical or different metal centres supported by one or more metal-metal bonds which, in addition to operating homogeneously, satisfies at the minimum the most basic criterion in an hierarchical set of conditions outlined by Rosenberg and Laine. 9 0 In order of decreasing precedence these are: (1) the catalyst must facilitate multi-site activation of organic substrates, (2) i f activation occurs at only one site on the intact cluster, the system must give different product distributions from the same reaction mediated by complexes containing a single metal centre, (3) i f (1) and (2) are not met, and i f the cluster fragments giving a monometallic active catalyst, the intact cluster must at least be a part of the catalytic cycle. 3.5.1 Homogeneous vs. heterogeneous Compared to monometallic complexes, clusters have a greater tendency to fragment and reaggregate in solution after loss of one or more ligands, and this leads to metal particle formation driven by metal-metal bonding. 9 1 ' 9 2 One effective test for the homogeneity of a reaction (especially those catalysed by complexes of group 10 metals which are known to form amalgams) is the addition of metallic Hg to the reaction mixture.9 3 This is a reasonably certain way to eliminate metal particles and therefore their heterogeneous contribution to the catalysis. In the opposite sense, addition of the rigid, tub-shaped dibenzocyclooctatetraene, which binds homogeneous catalysts irreversibly, will inhibit 122 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction the homogeneous component.94 A combination of Hg tolerance and inhibition by the tetraene is a good indicator therefore of the homogeneous nature of a reaction. 3.5.2 Multi-site activation In heterogeneously catalysed organic reactions, more than one metal centre on the surface is required to effect the transformation of reactant into product. By analogy, cluster catalysis in the strictest sense requires multi-site activation of the substrate. Only in catalysts where this criterion is fulfilled can true cooperative effects occur. However, the cluster may contain only a single active site, or multiple non-interacting sites, and these possibilities must be allowed for in a less rigid definition of cluster catalysis. In the case of a single active site, inactive metal centres and their associated ligand scaffold are conceptually equivalent to the L„ portion of the monometallic catalyst, ML„. Where there are m non-interacting active sites, the catalyst operates as the kinetic equivalent of mMLn. 3.5.3 Product distribution in single-site cluster catalysis If the product distributions given by a monometallic and a single-active-site polymetallic catalyst differ, cluster catalysis is indicated. The ring-opening cyclooligomerisation of thietane represents a good example of a reaction which is mediated both by one centre of the cluster Os 4(CO)i i[S(-CH 2-) 3](u-H) 4 9 5 and by the monometallic complex W(CO)5[S(-CH2-)3].96 However, the product distributions resulting from the two catalysts differ significantly, with the cluster producing predominantly small crown thioethers because of steric effects resulting from the extended metal-CO scaffold. 3.5.4 Cluster fragmentation Metal cluster catalysis often involves fragmentation either into smaller clusters or monometallic species,97 and the latter may be the active component. This possibility is accommodated in the least strict definition of cluster catalysis which necessitates that the intact cluster must form part of the catalytic cycle. This is the case for the Ru 3 (CO)i 2 -catalysed reduction of CO to ethylene glycol 9 8 - 9 9 in which the carbonyl is converted to [HRu3(CO)n]" and [Rul3(CO)3]" in the presence of H 2 , CO and I". The hydride donor is proposed to be [HRu(CO)4]" formed from [HRu3(CO)n]" and CO. Although the latter Ru 123 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction compound is not the active species, it is regenerated during the cycle. In contrast, the synthesis of acetic acid mediated by Ru3(CO)i2/CoI2 involves only monometallic species generated by fragmentation.100 3.5.5 Indications Fulfillment of one or'more of the following criteria indicates cluster catalysis:90 (1) The turnover frequency (TOF) increases with increasing catalyst concentration. (2) Product distributions obtained using cluster catalysts differ significantly from those obtained using monometallic precursors, or products cannot be explained using reasonable mechanisms occurring at a single metal centre (as mentioned above). (3) The use of a cluster or cluster precursor containing two or more different metals significantly enhances the reaction rate, or changes the product selectivity of a reaction catalysed by one of the metals, or allows catalysis by a metal which is normally inactive. (4) Reaction conditions are such that metal-metal bond formation is promoted. (5) A chiral metal framework results in catalytic asymmetric induction. 3.6 The Heck Reaction 3.6.1 General The Heck reaction represents the Pd-catalysed C—C bond forming reaction between an alkene and an aryl- or alkenylhalide in the presence of a base (Scheme 3.5; B = base). Scheme 3.5 B: + A r X + / Pd cat / R = / + B H + X Ar 124 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction The most widely accepted mechanism2 involves the oxidative addition of the aryl- or alkenylhalide to Pd° (most commonly generated in situ from Pd11), coordination of the alkene and insertion into the Pd—C bond, (3-hydride elimination to give the coupled product, and finally deprotonation by base to regenerate Pd°. An alternative mechanism involves a Pd'VPd™ cycle, although this is usually invoked only in the case of electron-rich Pd centres.4 3.6.2 Heck coupling of iodobenzene and styrene using anilinylphosphine complexes 3.6.2.1 Compounds tested The complexes tested as catalyst precursors for the production of cis- and trans-stiVoene by Heck coupling of iodobenzene and styrene at 100 °C in DMF/H2O with K2CO3 as base (Section 3.9.23) are shown in Chart 3.13 (M = Pt, Pd; Ar = o-C 6 H 4 NMe 2 ) . Also tested were MCI2. Catalyst loadings were on the order of 0.05-0.1 mol %. Chart 3.13 /Pd\ A X CI CI CI CI CI CI PdMCI 4(dmapm) MCI 2 (PMA) The monometallic compounds M C l 2 ( P M A ) 1 0 1 (PMA = Ph 2P-o-C6H 4NMe2) were chosen so that their steric and electronic properties would mimic closely the "halves" of the bimetallic complexes; to assess whether cooperative effects were a factor during the bimetal-catalysed reactions, a baseline activity for each of the metals was needed. This approach has been adopted previously by Stanley and coworkers in the analysis of hydroformylations catalysed by dirhodium complexes.29 125 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.6.2.2 Results PdCl 2 (PMA) and PdMCl4(dmapm) (M = Pt, Pd) were effective catalyst precursors for the Heck coupling of Phi and styrene under the conditions described in Section 3.9.23 (100 °C, D M F / H 2 0 solvent (3:2 by vol.), K 2 C 0 3 base). Although D M F and H 2 0 are miscible in all proportions, the reaction mixture consists of 2 phases at r.t. because of the presence of significant quantities of the reactants. However, at the operating temperature of 100 °C, the phases are miscible and the solution is homogeneous by visual inspection (see Homogeneity below). A typical plot of [Phi] vs. time for the coupling reaction is given in Figure 3.14. 0.40 -1 0 Time(s) Figure 3.14 The variation of [Phi] with time for the Heck coupling of Phi and styrene using Pd2Cl4(dmapm) (4.4 x 10"4 mol L"1) as catalyst precursor (100 °C, D M F / H 2 0 solvent (3:2 by vol.), K 2 C 0 3 base). The inset shows a plot of ln([PhI]/[PhI]0) vs. time where [Phl]0 is the initial Phi concentration. Figure 3.14 shows first-order loss of Phi (see inset), consistent with one step in the catalytic cycle involving oxidative addition of the aryl halide. However, Figure 3.14 is also consistent with the reaction being first-order in styrene and zero-order in Phi with the first-order decay of [Phi] resulting from the stoichiometry; further experiments in 126 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction which the [Phi]:[styrene] ratio is varied need to be done in order to rule out this possibility. Loss of styrene could not be followed accurately because the GC peak corresponding to the alkene overlapped with that of the D M F solvent. The stilbene product could be recovered pure simply by allowing the reaction mixture to cool: this resulted in precipitation of rrans-stilbene. Q?-stilbene, which is a liquid at r.t., is only produced to the extent of about 1 % under these conditions and remains in solution. The initial rates and turnover frequencies (TOFs) for mono- and bimetal-catalysed Heck coupling reactions between Phi and styrene are given in Table 3.4. (TOF = mol product per mol catalyst per s.) Table 3.4 Initial rates and TOFs for the Heck coupling of Phi (0.4 mol L"1) and styrene (0.4 mol L"1) at 100 °C in D M F / H 2 0 (3:2 by vol.) with K 2 C 0 3 (0.4 mol L"1) as base. Catalyst Concentration (molL" 1, x 104) Initial rate (mol L"1 s"],x 105) Initial TOF PtCl 2 (PMA) 7.00 0.00 0.000 PdCl 2 (PMA) 7.41 11.9 0.161 Pd2Cl4(dmapm) 1.15 5.87 0.510 PtPdCl4(dmapm) 2.14 4.85 0.227 3.80 10.1 0.266 Av. = 0.247 The initial TOFs imply both that the bimetallic catalysts are more active than would be predicted from the sum of the activities of their corresponding "halves" and that the catalysis is mediated by Pd only given that (i) Pd2Cl4(dmapm) is almost exactly twice as active as PfPdCl4(dmapm) and (ii) PtCl 2 (PMA) is completely inactive. Comparing TOFs in this concentration regime is valid because catalyst loading studies (see below) show the reaction to be first-order in Pd up to ca. 5 x 10"4 mol L" 1 . Catalyst Loading The variation of initial rate of the Heck coupling of Phi and styrene catalysed by PdMCl4(dmapm) (M = Pd, Pt) and PdCl 2 (PMA) with total Pd concentration is given in Figure 3.15. 127 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 0.0 1.0e-3 2.0e-3 3.0e-3 4.0e-3 Total [Pd] (mol L"1) Figure 3.15 Variation of initial rate of the Heck coupling of Phi and styrene with total Pd concentration for reactions catalysed by PdMCL^dmapm) (M = Pd, Pt) and PdCl2(PMA) under the conditions outlined in Section 3.9.23. If, during catalysis, the bimetallic PdMCL^dmapm) fragments completely or i f no fragmentation at all occurs, then the rate should show a first-order dependence on the catalyst concentration. If, however, an equilibrium between bimetallic and monometallic species is established and, i f the reaction is catalysed by one or both of the smaller fragments, then the rate should show a first-order dependence on catalyst concentration in the lower concentration regime (in which the bimetallic complex is almost completely fragmented into its constituent pieces) and a half-order dependence in the higher concentration regime (in which it is almost completely undissociated).90 When PdCl2(PMA) is used as the catalyst, the initial rate shows a first-order dependence on total [Pd] to a concentration of at least 1.6 x 10"3 mol L" 1; at higher concentrations, Pd metal precipitates immediately on addition of K2CO3, and the initial rate shows a corresponding non-linear dependence on catalyst loading. The dependence of the initial rate of the bimetal-catalysed reactions on total [Pd] is first-order to a limit of ca. 5 x 10"4 mol L" 1 . In this concentration range, the bimetallic 128 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction catalysts are slightly more active than PdC^CPMA) while being very similar to each other in activity. These observations suggest that both homo- and heterobimetallic catalyst precursors give rise to the same active species. The initial rates of both bimetal-catalysed reactions show almost identical non-linear dependence on total [Pd] above the 5 x 10"4 mol L" 1 limit. For example, when the concentration of PfPdCL^dmapm) is approximately doubled from 5.38 x 10"4 to 1.16 x 10"3 mol L" 1 , the initial rate increases from 1.10 to 1.56 x 10"4 mol L s"1, i.e., by a factor of 1.42, or approximately the square root of 2. At higher concentrations still, this half-order dependence is lost, pointing to catalyst degradation. (It should be noted, however, that in no reactions catalysed by bimetallic complexes was metal precipitation observed.) In addition, above the 5 x 10"4 mol L" ! concentration limit, the bimetallic catalysts are substantially less active than PdCbXPMA), pointing to a dearth of monometallic active species in solution at high total [Pd]. Shown in Figure 3.16 is a plot of ln(Initial rate) vs. ln(Total [Pd]) for the Heck coupling of Phi and styrene catalysed by the bimetallic complexes PdMCl^dmapm) (M = Pd, Pt). The plot is broken into low and high [Pd] regimes; linear regression analysis of these shows a rough first- to half-order transition. The point corresponding to the highest total [Pd] has been omitted from this analysis. -10.0 4--8.5 -8.0 -7.5 -7.0 -6.5 -6.0 InfTotal [Pd]) Figure 3.16 Plot of ln(lnitial rate) vs. ln(Total [Pd]) for the Heck coupling of Phi and styrene catalysed by the bimetallic complexes PdMCUCdmapm) (M = Pd, Pt). 129 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction These observations suggest that an equilibrium between bi- and monometallic species is established and that the catalysis is due mainly to monometallic species. For the heterobimetallic complex, PtPdCl4(dmapm), the two possible equilibria are shown in Scheme 3.6 (PdCl2(dmapm) represents an equilibrium mixture of the P,P'- and P,N-bonded forms). Scheme 3.6 PtPdCl4(dmapm) PdCl2(dmapm) + PtCl 2 I PtPdCl4(dmapm) ^ = PtCl2(P,P'-dmapm) + PdCl 2 II In order to assess which of the two equilibria predominates, PdCl2(dmapm) was used as a catalyst precursor for the Heck reaction at a concentration of 4.0 x 10"4 mol L" 1 . This gave an initial rate of 2.2 x 10"5 mol L"1 s"1 which is about 10 times less than predicted by extrapolation of the curve due to PdMCl4(dmapm) ( M = Pt, Pd) in Figure 3.15 (ca. 2 x 10"4 mol L" 1 s"1). This is consistent with equilibrium II being dominant and with the catalysis proceeding via a simple derivative of PdCl 2 . The following experiment suggests perhaps that equilibrium II dominates but that I is also operative. A mixture of D M F (1 mL) and H 2 0 (1 mL) containing equimolar PtPdCl4(dmapm) and dppe was heated to 100 °C under air for 4.5 h. The solvent was removed in vacuo and the residue analysed by N M R spectroscopy. The most abundant product was PdCl2(dppe); i f it is assumed that dppe reacts preferentially with MC1 2 , this result reflects the predominance of equilibrium II. In order of decreasing abundance, the other products were: PtCl2(dppe) which is accessed presumably by equilibrium I, unreacted PtPdCl4(dmapm), [Pt(dppe)2]Cl2, dmapm and PtCl 2(P,P -dmapm). Given that equilibrium II is likely to predominate and, assuming that PtCl 2(P,P'-dmapm) does not contribute to the catalysis (a likely prediction in light of inactivity of PtCl 2 (PMA) and PtCl 2), the overall catalysis is most probably mainly due to PdCl 2 (which may be complexed by any of DMF, H 2 0 or styrene). Of note, the Heck coupling of iodobenzene derivatives and styrene has been shown to occur in the presence of Pd(OAc) 2 without any added ancillary ligands in predominantly aqueous medium. 1 0 2 130 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction The initial TOF for a PdCl2-catalysed Heck reaction between Phi and styrene under the conditions used in this work was 4.86 x 10"2 s"1, significantly lower than would be expected from inspection of Table 3.4 - especially i f a simple derivative of PdCL. is the active species in the bimetal-catalysed reactions. In this experiment, PdCb was dissolved in the reaction mixture at 100 °C prior to addition of aqueous K2CO3. Metallic Pd deposited during the course of the reaction from the resulting blood red solution, pointing to the fact that the catalytic cycles involving the homo- and heterobimetallic complexes PdMCL^dmapm) and that involving PdC^ differ. The former is homogeneous while in the latter significant precipitation of Pd metal occurs. In addition, the bimetallic catalytic precursors are more active than simple monometallic derivatives of PdC^, perhaps because of the possibility for cluster recombination during the cycle which prevents the precipitation of metal. Homogeneity The "Hg test" was used to determine whether the Heck reaction is truly homogeneous under the conditions outlined in Section 3.9.23. Thus, 0.05 mL of Hg was added to a reaction run under the standard conditions and catalysed by PtPdCl^dmapm) (1.0 x 10" mol L"1) after the reaction had already been in progress for 30 min. GC analysis of the reaction mixture over the subsequent 30 min showed that addition of Hg had not affected the rate, indicating that the catalysis is homogeneous. 3.6.2.3 Oxidative degradation Subsequent to an unsuccessful attempt to follow the PtPdCl4(dmapm)-catalysed reaction by N M R spectroscopy, orange-brown crystals deposited from a mixture of DMF/D2O containing Phi and styrene which had been heated to 80 °C for 0.5 h in air and then left for 2 d at r.t. These were submitted for X-ray crystallographic analysis but unfortunately the structure did not refine sufficiently for publication standards. However, the molecular structures of two constituent complexes were established unequivocally. Present in the unit cell were the compounds PtPdL^dmapm) (Figure 3.17) and PtI2(.P,./V-dmapmO) (Figure 3.18), and one solvate molecule of both D M F and D2O. In these complexes, 131 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction chloride has been substituted by iodide (which is produced stoichiometrically as a by-product of the Heck reaction; see Scheme 3.5). Figure 3.17 PLUTO representation of the molecular structure of PtPdLt(dmapm). H -atoms are omitted for clarity. Figure 3.18 PLUTO representation of the molecular structure of PtI2(P,./V-dmaprnO). H-atoms are omitted for clarity. 132 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction The crystallographic results show that the bimetallic catalyst can also decompose via a redox route. Discounting equilibrium I, the paths outlined in Scheme 3.7 are plausible (X is CI or I). In the first, PtPdCl^dmapm) (or its iodo-derivative) is attacked by OH" to give a mono-oxide derivative of dmapm bound to Pt in a P,N fashion, i.e., PtX2(P,7V-dmapmO), and Pd° (which may contribute to the catalysis). The generation of Pd° from Pd"-phosphine complexes during catalytic transformations of arylhalides is well documented. 1 0 3 ' 1 0 4 In the second, less likely and more speculative path, VtCh(P,P'-dmapm), generated by equilibrium II, reacts with OH" to give an intermediate such as "Pt(dmapmO)" which is transformed in a subsequent step(s) to give PtX2(P,./V-dmapmO) (e.g., by double oxidative addition of Phi followed by reductive elimination of biphenyl). It is important to note that even under the high temperature, strong base conditions of the reaction, a significant proportion of the catalyst survived intact (based on the composition of the isolated crystals). Scheme 3.7 PtPdX 4(dmapm) + OH- - PfX 2(dmapmO) + 2 X + H + + Pd PtPdX 4(dmapm) . PfX 2(dmapm) + PdCI 2 OH-"Pt(dmapmO)" + H + + 2X-PfX 2(dmapmO) 3.6.2.4 Attempted isolation and identification of Heck catalytic intermediates Section 3.9.21 outlines an attempt to isolate an intermediate from the Heck reaction catalysed by PtPdCL^dmapm). The orange powder thus obtained gave the 3 1 P{ 1 H} N M R spectrum shown in Figure 3.19; the spectrum is obviously due to a mixture of complexes. The species considered most likely to be present were: PtPdLt(dmapm), Ptl2(P,N-133 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction dmapmO), PdI2CP,./V-dmapmO) and PtPdI2(dmapm). True organometallic intermediates are also possible. Not surprisingly, neither the parent complex PtPdCL^dmapm) nor its 2-electron reduced product, PtPdCl2(dmapm) is present because of the stoichiometric iodide produced and the known degradation pathways. Three P-containing compounds (1, 2 and 3) can be detected (Table 3.5), and in an attempt to identify them, the following experiments were performed. 1,2 I ' 1 1 1 I 1 60 i 1 1 1 40 1,2 ~i—i—|—i—i—i—r 20 -20 Figure 3.19 The 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) spectrum of the orange powder isolated from the Heck reaction catalysed by PtPdCL^dmapm). Table 3.5 The 3 1 P{'H} N M R (121 MHz, C 6 D 6 , 300 K) spectral data for compounds 1, 2 and 3. Species 5pa 5pb 2 •ZpaPb (Hz) ' / p a P t (Hz) 1 -3.5 24.3 Not observed 3680 2 -4.0 24.3 Not observed 3560 3 48.5 65.6 11.3 3370 PtPdL^dmapm) was synthesised in situ by reaction of PtPdCLXdmapm) with an excess of Nal in a mixture of acetone-d6 and CDCI3, and was characterised by 3 1 P{'H} N M R spectroscopy: 6 P a 12.1, 8Pb 23.6; 2-bond PP coupling was again not observed, while Vpapt could not be determined because of the very low solubility of the complex in this medium. The synthetic strategy given in Scheme 3.8 was employed in order to synthesise PtI2(P,A^-dmapmO); experimental details are given in Section 3.9.22. Although the 134 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction procedure did not give either PtCl2(P,/V-dmapmO) or PtI2(P,iV-dmapmO) pure, these compounds could be identified spectroscopically. The 3 1 P{'H} N M R (300 MHz, CDC1 3, 300 K) data are as follows: PtCl2(P,/V-dmapmO) 5pa —3.4, 5pb 26.1, 2 . / p a p b not observed, ' j paP t = 4110 Hz. PtI2(P,A^-dmapmO) 5 P a -5.4, § P b 26.9, 2 J P a P b not observed, ' j p a P t = 3850 Hz. Because of the similarity between these data and those given in Table 3.5, complexes 1 and 2 probably contain dmapmO P,N-bound to Pt. In addition, the downfield resonances due to 3, cannot be rationalised. Scheme 3.8 Ar A N A r ' O NMe 2 -Pd—CI CI Ar . Ar" ^ / ~ - N M e 2 Ar O Pt-I excess KCN excess Nal >]2 + K2[Pd(CN)4] NMe„ Me 2N PtCI2(cod) U--NMe2 i Ar -Pt CI + cod Ar-+ 2 NaCI 3.6.2.5 Summary of the catalysis The major observations of the catalysis study are summarised below: (1) The homo- and heterobimetallic complexes are more active catalysts than their monometallic "halves", and they also outstrip both PtCl 2 and PdCl 2 . (2) The catalyst does not remain intact during the reaction but is degraded by fragmentation and oxidation. A general scheme which accounts for these observations remains elusive. Whereas the kinetic data suggest the equilibria given in Scheme 3.6, examination of the post-reaction 135 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction residue does not conclusively reveal any of these species. In addition, although a simple derivative of PdCl 2 is suggested as the most active species, reactions catalysed by PdCl2 do not proceed as quickly as those catalysed by the bimetallic complexes. Finally, phosphine oxidation by base precludes the equilibria I and II as they are written because it prevents recombination to give bimetallic complexes of the original type; however, P,./V-bridged as opposed to P,P '-bridged compounds remain possible, resulting in equilibria of the type III shown in (M, M ' = Pt, Pd). Scheme 3.9 MM'X4(dmapmO-K-P,7V-|a-P7vT') MX2(dmapmO-K-P,A0 + M ' X 2 III 3.6.3 A cooperativity index proposal Because bi- and polymetallic catalysis research has received renewed attention and that cooperative mechanisms have been suggested and in some cases proven, I now propose a general index for the degree of cooperativity between metal centres in a polymetallic assembly. This index serves as a kinetic complement to the well-established Hil l coefficient105 for thermodynamic cooperativity. Assuming no inter-metallic interactions, the predicted total activity, Ap, of a polymetallic catalyst of n metal centres is given by equation (1), where At is the measured activity of the monometallic complex which most closely mimics the steric and electronic attributes of the z'th metal centre in the polymetallic assembly. By activity, any reasonable observable should suffice, e.g., initial rate, TOF or % conversion in a given time. (1) The average activity, A , is then given by equation (2). - A ( 2 ) A = ^ n If the observed activity of the polymetallic complex is Ao, then equation (3) defines one possibility for an index of cooperativity, a. A A ( 3 ) A 136 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction This index aligns with common sense in that impeding effects result in a values less than zero while completely non-interacting centres give a = 0, and cooperative effects yield positive a values. If a = 0, cluster catalysis is not ruled out. In this case, catalysis may proceed at only one centre in the complex, or at multiple non-interacting centres (Section 3.5.2). Conversely, the cluster may be fragmenting. In general, the index corresponds to a number of "virtual" metal centres, i.e., the number of centres of average catalytic activity the polymetallic complex appears to possess beyond its real number, n. Of note, Jacobsen and coworkers have been able to distinguish kinetically between inter- and intramolecular pathways in the catalytic asymmetric ring-opening of cyclopentene oxide by Cr(salen)-type dimers linked by tethers of varying lengths, and for each of these an effective reactive concentration of the two salen units has been determined by the ratio of A:jntra/^inter-35 This method is akin to the cooperativity index now proposed and is an exact treatment for reactions which necessarily involve two metal centres, even i f they are not held in proximity by a bridging ligand. The a index is appropriate for reactions which occur readily at one active site and which may or may not benefit by the juxtaposition of other sites. The a values for the bimetallic catalysts used here are calculated from initial TOFs. The observed activities (Ao) for PdMCL^dmapm) (M = Pt, Pd) are taken to be the initial TOFs given in Table 3.4; the calculations of a for each of the bimetallic complexes are outlined in Table 3.6. Ap values are determined by summing the appropriate initial TOFs of reactions catalysed by MC1 2(PMA). 137 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Table 3.6 Calculation of cooperativity indices for PdMCl4(dmapm) catalysts. Pd2Cl4(dmapm) PtPdCl4(dmapm) Ao 0.510 0.247 AP 0.322 0.161 A =AP/2 0.161 0.081 a = (A0-AP) 1 A 1.17 1.06 The a indices imply that each of the bimetallic catalysts acts as i f it possessed an additional metal centre of average catalytic activity. As a comparison, the a values for Stanley's bimetallic Rh hydroformylation catalysts [Rh2(nbd)2(rac-et,ph-P4)][BF4]2 and [Rh2(nbd)2(we5o-et,ph-P4)][BF4]2 are 422 and 34, respectively (see Chart 3.14, p. 140).29 It should be noted that the index a is based solely on kinetic data at a single catalyst concentration (i.e., it assumes that the reaction is first-order in catalyst) and may lead to the conclusion that a cooperative mechanism is at work whereas studies of the rate dependence on catalyst loading may prove otherwise. Notwithstanding the fact that the Heck reaction studied here may be catalysed mainly by a simple derivative of PdC^ and not by a bimetallic assembly, I feel that the a index outlined above will be of general utility. 3.7 Conclusions The dmapm ligand gives access to a wide variety of platinum metal complexes because of its ability both to bridge (P,P') and chelate (P,P' and P,N). In monometallic Pd" complexes, the ligand is in equilibrium between its P,P'- and P,7V-chelated forms, the exchange mechanism probably involving 5-coordinate intermediates of the P,P',N- and P,N,N'- variety. The ligand binds to Pd" via P predominantly through a-donation. The higher affinity of Pt" than Pd" for P is illustrated by the fact that in monometallic halo complexes of Pt", only the P,P '-chelated ligand is observed; this necessitates the surmounting of an approximately 32 kJ mol"1 4-membered ring strain. 138 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Homo- and heterobimetallic complexes supported by dmapm can be synthesised, where the ligand adopts a P,P '-bridging, P,/V-chelate coordination mode. In the 11,11 complexes MM'CL^dmapm) (M = Pd; M ' = Pt, Pd), the bound and free N-atoms are in chemical exchange, whereas in the 1,1 species MM'Cl2(dmapm), which can be accessed either through conproportionation of M° and M 1 1 or via reduction of the appropriate II,n compound, there is no exchange. The MM'Cl2(dmapm) complexes, having a very different structure from most other diphosphine-bridged M 1 complexes, show unique reactivity with EtSH and CS2 (see Section 3.8.3, p. 141). The possibility for the formation of doubly-bridged adducts with small molecules via displacement of the anilinyl "arms" exists. Fluxional compounds can also result, and the high affinity of Pt for P can lead to disproportionation in the case of the mixed-metal complexes. The homo- and heterobimetallic 11,11 compounds serve as useful catalytic precursors for the Heck reaction in aqueous media. These compounds do not survive intact under catalytic conditions but are degraded both by phosphine oxidation and by fragmentation. The latter process generates PdCl2, a simple derivative of which likely being the predominantly active species in the catalysis. A generally applicable index of cooperativity is proposed. This corresponds to the number of virtual centres of average activity a complex appears to have beyond its real number. 3.8 Recommendations for Future Work 3.8.1 More accurate determination of ring strain energy The 4-membered dmapm ring strain energy has been calculated based on the relative bond strengths of Pd—PPI13 and Pd—py which serves as the approximate energy difference between the Pd—P and Pd—N bonds in PdCi2(dmapm) (Section 3.3.2). A better value for this difference may be determined from the PdCKMeXP,/*-dmape) = J = = ± PdCl(Me)(P,./V-dmape) equilibrium. The difference in enthalpy between this and the analogous equilibrium involving PdCl2(dmapm) should give a reasonably 139 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction accurate value for the 4-membered ring strain energy, with the caveat that non-identical ligand sets are being compared. 3.8.2 Assessment of cooperative effects in other types of homogeneous catalysis 3.8.2.1 Hydroformylation Stanley and coworkers have demonstrated cooperative effects in dirhodium catalysed hydroformylations. Their highly active, regioselective catalysts are based on the electron-rich tetraphosphine ligand rac-et,ph-P4 (Chart 3.14, left) which forms complexes of the type [Pvh2(nbd)2(rac-et,ph-P4)][BF4]2 (Chart 3.14, right). Chart 3.14 Given that P,N-chelate ligands have been shown to enhance the rate of hydroformylation of styrene,106 it would be of interest to compare the rate and regioselectivity of Stanley's catalyst to those of [Rh2(nbd)2(dmapm)][BF4]2 which should be accessible. 3.8.2.2 Imine/CO copolymerisation The catalytic copolymerisation of imines and CO to form polypeptides is an attractive goal. Imine insertion into Pd—acyl bonds has been observed 1 0 7" 1 1 0 but, unlike in ethylene/CO copolymerisations,111 subsequent coordination and insertion of CO to continue chain growth are hampered by the resulting formation of a stable 5-membered palladacycle (Chart 3.15, left). 1 0 7 If the amide carbonyl could be tethered to a sacrificial metal centre in a bimetallic complex, for example in one supported by dmapm (Chart 3.15, right; Ar = o-CeEWSIM^, X = halide), CO coordination would then be possible at the other centre and chain propagation could proceed. Cationic organometallic 140 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction derivatives of dipalladium dmapm complexes warrant investigation for such an application. Chart 3.15 P 2 V " P O ^ M e P h 0 M e 9 N C O \ / 2+ Pd P h p Ar A r < N P O : \ / / P d x N X M e 0 M e 3.8.3 Exploration of the reactivity of MM'X2(dmapm) complexes For the geometrical reasons outlined in Section 3.4.5.3, MM'Cl 2(dmapm) complexes are fundamentally different from the "typical" bimetallic diphosphine-bridged M 1 complexes M 2Cl2(P-P)2- During the course of this work, preliminary investigations into the reactivity of these complexes with small molecules like H2S, EtSH, and CS2 were conducted, and some in situ characterisation data are summarised in Table 3.7. None of the products of these reactions has been isolated. Table 3.7 3 1 P{'H} N M R (CDC1 3, 300 K) data for the in situ reactions between small molecules and MM'Ci2(dmapm). a Molecule M M ' Spectral data H 2 S PtPd 5 21.2 (d), 46.8 (d), 2JP? = 47.5 Hz H 2 S PdPd 5 47.4 (s) and broad, i l l defined peaks between § 25-35 EtSH PdPd 8 48.4 (s) C S 2 b PdPd 8 9.9 (d, Vp P = 70.1 Hz), 16.4 (d, V P P = 55.7 Hz), 19.0 (d, Vp P = 65.7 Hz), 20.7 (s), 26.9 (d, 2 J P P = 63.9 Hz), 33.8 (d, V P p = 57.6 Hz), 40.3 (d, V P P = 61.4 Hz), 41.1 (d, VPP = 55.5HZ) a Spectra recorded at 121 MHz unless otherwise indicated. b Recorded at 162 MHz. 141 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction In the reactions with H 2 S, H 2 was identified as a product in the ' H N M R spectra (5 4.6). Moreover, significant effervescence occurred on introduction of H 2 S to CDCI3 solutions containing M M ' C l 2 (dmapm). This suggests that, like M 2 X 2 ( P - P ) 2 , 8 0 MM'Cl 2(dmapm) mediates the conversion of H 2 S to H 2 and 'S, ' with the metal complex product probably being a bridged sulphide species like MM'Cl 2(u-S)(dmapm) or MM'Cl 2(u-S) 2(dmapm). In the H 2 S reaction with Pd2Cl2(dmapm), the broad 5 25-35 region is not resolved at 220 K , perhaps indicating fluxionality involving the anilinyl "arms" of the complex. Peaks in the ' H N M R spectra both at 220 and 300 K are broad and uninformative. In the ' H N M R spectrum of the in situ reaction with EtSH, no hydride was detected to 5 —4. The C H 2 protons in the ligand backbone gave a triplet at 5 4.75 and the NMe protons appeared to be equivalent, giving rise to a broad singlet at 8 2.9. These data are consistent with a C2-symmetrical product resulting from substitution of the bound N -atoms of Pd2Cl2(dmapm) by EtSH. Alternatively, the similarity in the spectra of this and that of the reaction with H 2 S is consistent with a bridging-sulphide or -thiolate species. It should be noted that M 2 X 2 ( P - P ) 2 species do not react with E t S H . 1 1 2 The reaction of C S 2 with Pd2Cl2(dmapm) gives a range of products which remain unidentified. This finding is in contrast to that of the analogous reaction with Pd 2Cl 2(dmpm) 2 which yields PdCl2((J.-CS2)(dmpm)2.79 Clearly, further work is necessary to elucidate the nature of the products of the reactions outlined in Table 3.7, and to rationalise the results in terms of the structural features of the MM'Cl 2(dmapm) complexes. 3.9 Experimental 3.9.1 PdCl2(i>P'-dniapm) and PdCI2(P,7V-dmapm). To a combination of ^ra«s-PdCl 2(PhCN) 2 (31 mg, 0.081 mmol) and dmapm (48 mg, 0.086 mmol) was added CH 2 C1 2 (5 mL). The initially orange solution turned yellow within a few seconds. After the solution was stirred for 5 min, the volume was reduced in 142 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction vacuo to ca. 1 mL and E t 2 0 (20 mL) was added to give a yellow powder. This was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried under vacuum. Yield: 54 mg (92 %). Anal. Calcd for C 3 3 H 4 2 N 4 C l 2 P 2 P d : C, 54.0; H , 5.8; N , 7.6. Found: C, 53.8; H, 5.9; N , 7.4. UV-vis (CH 2C1 2): 360 [3870]. PdCl2(P,P'-dmapm): *H N M R (300 MHz, CDC1 3, 300 K): 5 2.35 (s, 24H, NC# 3 ), 5.31 (t, 2H, CH2, 2JH? = 12.1). The aromatic proton resonances of this and PdCl2(P,A/-dmapm) overlap (6 6.4-8.2) and could not be resolved. 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K) : 5 -56.8 (s). PdCl2(P,/V-dmapm) *H N M R (300 MHz, CDCI3, 215 K): 5 1.61 (s, 3H, NC# 3), 2.43 (s, 6H, NC/ / 3 ) , 2.87 (pt, 1H, CH2), 2.91 (s, 3H, NC//3), 2.97 (s, 6H, NCi / 3 ) , 3.46 (s, 3H, N G H 3 ) , 3.79 (s, 3H, NC# 3), 4.01 (pt, 1H, CH2). 3 1 P{'H} N M R (121 MHz, CDCI3, 215 K): 5 -39.8 (d, V P P = 88.9), 35.4 (d, 2JPp = 88.9). Crystals of PdCl2(P,/V~-dmapm) suitable for analysis by X-ray diffraction were grown by slow evaporation of a CDCI3 solution. 3.9.2 PdBr2(P,P'-dmapm) and PdBr2(P,/V-dmapm). To a combination of PdCl2(dmapm) (77 mg, 0.10 mmol) and NaBr (170 mg, 1.7 mmol) was added acetone (10 mL) and H 2 0 (2 mL). The yellow slurry was stirred for 2 h before reduction to dryness at the pump. The residue was taken up in CH 2 C1 2 (10 mL) and filtered through Celite 545. The volume of the filtrate was reduced in vacuo to ca. 1 mL and Et 2 0 (20 mL) was added to give a yellow powder which was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried under vacuum. Yield: 55 mg (64 %). Anal. Calcd for C 3 3H 4 2 N 4 Br 2 P 2 Pd: C, 48.2; H , 5.1; N , 6.8. Found: C, 48.1; H , 5.1; N , 6.6. UV-vis (CH 2C1 2): 378 [4310]. PdBr2(P,P'-dmapm): *H N M R (400 MHz, CDCI3, 250 K): 5 2.30 (br s, 24H, N C i / 3 ) , 5.41 (t, 2H, CH2, 2Jhp = 12.0). The aromatic proton resonances of the two isomers overlap (6 6.3-7.9) and could not be resolved. 3 1 P{'H} N M R (162 MHz, CDCI3, 250 K) : § -58.5 (s). PdBr2(P,/V-dmapm): ' H N M R (400 MHz, CDCI3, 250 K): 6 1.61 (br s, 3H, NC//3), 2.45 (s, 6H, NC# 3 ), 2.92 (s, 9H, NGtf 3), 3.01 (pt, 1H, CH2), 3.51 (s, 3H, NC/ / 3 ) , 3.82 (s, 3H, NC/ / 3 ) , 3.97 (pt, 1H, CH2). 3 l P{ 'H} N M R (162 MHz, CDC1 3, 250 K) : 5 -40.4 (d, V P P = 112), 33.3 (d, 2 J P P = 112). 143 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.9.3 PdI2(P,P-dmapm) and PdtyP^-dmapm). To a combination of PdCl 2(PhCN) 2 (91 mg, 0.24 mmol), dmapm (130 mg, 0.23 mmol) and Nal (190 mg, 1.3 mmol) was added CH 2 C1 2 (5 mL) followed after 5 min by acetone (10 mL) which caused an almost immediate colour change from yellow to deep orange. The slurry was stirred for 2 h at r.t. before reduction to dryness in vacuo. The work-up was the same as that for PdBr2(dmapm). Yield: 180 mg (83%). Anal. Calcd for C33H 42N 4I 2P 2Pd: C, 43.2; H, 4.6; N , 6.1. Found: C, 43.4; H, 4.7; N , 6.0. UV-vis (CH 2C1 2): 304 [16100], 430 [4010]. PdI2CP,P,-dmapm): ' H N M R (300 MHz, CDC1 3 , 250 K): 6 2.26 (br s, 24H, NGf/ 3 ) , 5.58 (t, 2H, CH2,2JH? = 12.4). The aromatic proton resonances of the two isomers overlap (8 6.3-8.0) and could not be resolved. 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K): 8 -65.6 (s). PdI2(P,/V-dmapm): ' H N M R (300 MHz, CDC1 3 , 250 K): 8 1.55 (br s, 3H, NGtf 3), 2.46 (s, 6H, NC/ / 3 ) , 2.80 (br s, 3H, NC/ / 3 ) , 2.88 (s, 6H, NC/ / 3 ) , 3.22 (pt, 1H, CH2), 3.55 (s, 3H, NC/ / 3 ) , 3.84 (s, 3H, NGtf 3), 3.93 (pt, 1H, CH2). 3 1 P{'H} N M R (121 MHz, CDC1 3 , 300 K) : 8 -40.8 (d, 2JPP = 98), 27.3 (d, 2JPP = 98). 3.9.4 Pd(CN)2(P,7V-dmapm). To a yellow slurry of PdCl2(dmapm) (46 mg, 0.063 mmol) and K C N (8.3 mg, 0.013 mmol) in EtOH (5 mL) was added H 2 0 (2 mL) whereupon a colourless solution formed. The solvent was removed in vacuo after 10 min and the residue was taken up in CH 2 C1 2 (10 mL). This was filtered through a mixture of Celite 545 and M g S 0 4 and reduced to ca. 1 mL. Addition of E t 2 0 (10 mL) gave the product as a white powder that was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried under vacuum. Yield: 24 mg (54 %). Anal. Calcd for C 3 5 H 4 2 N 6 P 2 P d : C, 58.8; H , 5.9; N , 11.8. Found: C, 58.9; H, 6.0; N , 11.5. ] H N M R (300 MHz, CDC1 3, 233 K): 8 1.57 (s, 3H, NC# 3 ), 2.33 (s, 6H, NC/ / 3 ) , 2.74 (s, 3H, N C / / 3 ) , 2.96 (s, 6H, NC/ / 3 ) , 3.11 (pt, 1H, CH2), 3.41 (s, 3H, NC/ / 3 ) , 3.61 (pt, 1H, CH2), 3.77 (s, 3H, NC# 3 ), 6.39 (pt, 1H, Ar), 6.66 (m, 2H, Ar), 6.96 (m, 3H, Ar), 7.15 (pt, 1H, Ar), 7.22 (pt, 1H, Ar), 6.39 (pt, 1H, Ar), 7.32 (m, 5H, Ar), 7.48 (m, 2H, Ar), 7.72 (pt, 1H, Ar). 3 1 P{ 'H} N M R (121 MHz, CDC1 3, 233 K) : 8 21.1 (d, 2J?P = 130), -41.6 (d,V P p= 130). 144 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.9.5 PdCKMeXP^-dmapm). This compound was made in the same manner as for PdCl2(dmapm). Thus, reaction of PdCl(Me)(cod) (60 mg, 0.23 mmol) and dmapm (130 mg, 0.23 mmol) in CH 2 C1 2 (10 mL) gave 86 mg (53 %) of a yellow powder. ' H N M R (300 MHz, CDC1 3 , 300 K): 5 0.19 (d, 3H, Pd-C# 3, V H p = 2.84), 2.16 (s, 6H, NC# 3 ), 2.41 (s, 6H, N C / / 3 ) , 2.41 (s, 6H, NC# 3), 2.78 (s, 6H, NC7/ 3), 2.85 (m, 1H, CH2), ca. 3.1 (m, 1H, CH2, obscured), 3.14 (s, 3H, NC/ / 3 ) , 3.31 (s, 3H, NGr73), 6.6-7.6 (m, 14H, Ar), 7.80 (m, 2H, Ar). 3 1 P{ ] H} N M R (121 MHz, CDCI3, 300 K): 8 24.0 (d, 2 J P P = 130), -39.3 (d, 2 y P P = 130). 3.9.6 PdCl2(P,A^-dmapmO). This complex could be prepared either in a single phase using cumene hydroperoxide as oxidant, or in a two-phase mixture using H 2 0 2 . (a) Cumene hydroperoxide. The precursor PdCl2(dmapm) was prepared in situ by the reaction of dmapm (140 mg, 0.25 mmol) and fra«^-PdCl 2(PhCN) 2 (94 mg, 0.25 mmol) in CH 2 C1 2 (10 mL), and to the yellow solution was added cumene hydroperoxide (75 uL, 0.41 mmol). The solution was stirred over night and then reduced in vacuo to ca. 2 mL, whereupon E t 2 0 (20 mL) was added to give the product as a yellow precipitate. This was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried under vacuum. Yield: 155 mg (84 %). (b) H20?. To a CH 2 C1 2 solution (5 mL) containing PdCl2(dmapm) (28 mg, 0.038 mmol) was added a 3 % H 2 0 2 (2 mL) solution. The two-phase mixture was stirred for 1 h and the aqueous phase was removed. The volume of CH 2 C1 2 was reduced in vacuo to ca. 1 mL, and E t 2 0 (10 mL) was added to give the product as a yellow precipitate. Yield: 25 mg (88 %). Anal. Calcd for C33H 4 2 N 4 Cl 2 OP 2 Pd: C, 52.9; H , 5.6; N , 7.5. Found: C, 53.0; H , 6.0; N , 7.0. ' H N M R (300 MHz, CDC1 3, 300 K): 5 1.7-3.3 (br, 12H, NC//3), 2.35 (s, H , NC/ / 3 ) , 2.51 (s, H , N C / / 3 ) , 3.47 (s, 3H, NC# 3 ), 3.70 (s, 3H, NC# 3 ) , 4.27 (m, 1H, CH2), 4.81 (m, 1H, CH2), 6.8-7.9 (br m, 16H, Ar). ^P j 'H} N M R (121 MHz, CDC1 3, 300 K): 5 26.2 (d, V P P = 7.0), 26.5 (d, 2 J P P = 7.0). u P 0 : 1185 (m). A M (CH 2C1 2 , 298 K): < 1. 145 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.9.7 [PdCKWS-dmapmS)][PF6]. To a mixture of PdCl2(dmapm) (44 mg, 0.060 mmol) and Ss (13 mg, 0.42 mmol) was added CH2CICH2CI (5 mL) and the resulting yellow solution heated to reflux for 3.5 h. The solvent was removed in vacuo and the residue dissolved in warm H2O (20 mL). After filtration through Celite 545 to remove the excess Sg and unreacted PdCl2(dmapm), an aqueous solution (5 mL) containing K P F 6 (90 mg, 0.49 mmol) was added. This immediately gave a yellow-orange precipitate that was isolated by filtration and washed with H 2 0 (10 mL). Yield: 30 mg (56%). Anal. Calcd for C33H42N4ClF6P3PdS: C, 45.3; H, 4.8; N , 6.4. Found: C, 45.2; H, 5.0; N , 6.2. ] H N M R (300 MHz, CDC1 3 , 300 K): 5 2.15 (s, H , NC/f 3 ) , 2.86 (s, H , NCrY 3), 4.50 (br m, 2H, C H 2 ) , 7.30 (m, 4H, Ar), 7.46 (m, 4H, Ar), 7.57 (m, 4H, Ar), 7.87 (m, 2H, Ar), 8.50 (dd, 2H, Ar, 2JHH = 7.39, V H P = 15.3). 3 1 P{'H} N M R (121 MHz, CDC1 3 , 300 K): 8 37.0 (d, V P P = 33.6), 48.6 (d, 2 J P P = 33.6), -144 (spt, ' J P F = 710, PF6"). 3.9.8 [PdCI(P,/V,S-dmapmS)]Cl. This compound was made in the same manner as for [PdCl(P,N, S-dmapmS)][PF6] except that after the reaction of PdCl2(dmapm) (130 mg, 0.18 mmol) and Sg (45 mg, 1.4 mmol), the orange solution was allowed to cool to r.t. and was then filtered through Celite 545. The volume of the yellow filtrate was reduced in vacuo to ca. 1 mL and Et20 (20 mL) was added to give the product as a beige precipitate. Yield: 100 mg (71 %). The N M R spectroscopic data for this complex are the same as for [PdCl(P,A7,5'-dmapmS)][PF6] except for the absence of peaks due to PFg". A M (H 2 0, 298 K): 99. 3.9.9 PtCl2(P,P'-dmapm). The synthesis of this compound was identical in principle to that of PdCl2(dmapm). Thus, PtCl2(cod) (58 mg, 0.15 mmol) and dmapm (86 mg, 0.16 mmol) gave 83 mg (66 %) of an off-white powder. Anal. Calcd for Q ^ ^ C L F ^ P t : C, 48.2; H , 5.2; N , 6.8. Found: C, 48.5; H , 5.4; N , 6.7. l H N M R (300 MHz, CDC1 3 , 300 K): 6 2.24 (s, 24H, N C / / 3 ) , 5.77 (t, 2H, CH2, 2JHP = 12.9 , V H P t = 60), 7.17 (pt, 4H, Ar), 7.25 (pd, 4H, Ar), 7.44 (pt, 4H, Ar), 7.17 (br m, 4H, Ar). 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K): 6 -65.6 (s, ^ = 3 0 8 4 ) . 146 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3.9.10 Pt(CN)2(P,P'-dmapm) and Pt(CN)2CP,/V-dmapm). To a slurry of PtCl 2(P,P '-dmapm) (190 mg, 0.23 mmol) in EtOH (10 mL) was added solid K C N (30 mg, 0.47 mmol) followed by H 2 0 (2 mL). The mixture was warmed to 50 °C for 0.5 h before all the solvent was removed at the pump. The work-up was identical to that for Pd(CN)2(P,7V"-dmapm). Yield: 140 mg (74%). Anal. Calcd for C 3 5H 4 2 N 6 P 2 Pt: C, 52.3; H , 5.3; N , 10.5. Found: C, 51.5; H , 5.3; N , 10.0. Pt(CN)2(P,P'-dmapm): ' H N M R (300 MHz, CDC1 3, 300 K): 5 2.34 (s, 24 H , NGF/ 3), 5.46 (t, 2H, CH2, 2JHP = 12.3, 3 J H P t = 38.3). The aromatic proton resonances of the two isomers overlap (§ 6.5-7.9) and could not be resolved. 3 1 P{'H} N M R (121 MHz, CDC1 3 , 300 K): 5 -60.7 (s, 1 Jp P t = 2230). PttCNMP^-dmapm): ' H N M R (300 MHz, CDC1 3 , 300 K): 5 2.38 (s, 6H, NC/ / 3 ) , 2.88 (s, 6H, NGf/ 3 ) , 3.19 (pt, 1H, CH2), 3.59 (s, 4H, NGf/ 3 and CH2 (obscured), 3 J H P t = 22.5), 3.84 (s, 3H, NCH3, VHpt = 22.5). 3 1 P{ ] H} N M R (121 MHz, CDC1 3 , 300 K): 5^*1.5 (d, 2 J P P = 125), 11.6 (d, 2 J P P = 125, 1 J P P t = 2760). 3.9.11 Pd2Cl4(dmapm). This compound was made in the same fashion as PdCl2(dmapm), but a 2:1 mol ratio of Pd:dmapm was used. Thus, trans-?dCl2(PhCN)2 (HOmg, 0.29 mmol) and dmapm (82 mg, 0.15 mmol) gave HOmg (78%) of a yellow powder. Anal. Calcd for C 33H4 2N 4Cl4P 2Pd 2: C, 43.5; H , 4.7; N , 6.2. Found: C, 43.8; H , 4.6; N , 6.0. UV-vis (CH 2C1 2): 354 [3050]. ] H N M R (300 MHz, CDC1 3, 300 K): 5 2.31 (s, 12H, N C i / 3 ) , 2.78 (s, 6H, NCHi), 3.28 (s, 6H, N C i / 3 ) , 4.97 (t, 2H, CH2, 2JHP =17.1), 7.40 (br m, 6H, Ar), 7.65 (br m, 4H, Ar), 7.80 (br m, 2H, Ar), 8.52 (br m, 2H, Ar), 8.74 (br m, 2H, Ar). 3 1 P{'H} N M R (121 MHz, CDC1 3 , 300 K): 5 34.8 (s). 3.9.12 Pd2I4(dmapm). The synthesis of this compound was the same in principle as that of PdBr2(dmapm), except that H 2 0 was not used. Thus, a combination of Pd2Cl4(dmapm) (42 mg, 0.046 mmol) and Nal (140 mg, 0.95 mmol) in acetone gave 20 mg (34 %) of a red powder. Anal. Calcd for C 3 3 H 4 2 N 4 l4P2Pd 2 : C, 31.0; H , 3.3; N , 4.4. Found: C, 31.0; H, 3.4; N , 4.2. ' H N M R (300 MHz, CD 2 C1 2 , 300 K): 5 1.83 (br s, 6H, NC/ / 3 ) , 2.37 (br s, 6H, NC# 3), 2.76 (br s, 6H, NC/ / 3 ) , 3.46 (br s, 6H, NC# 3 ), 3.52 (t, 2H, CH2, 2JHP = 16), 7.55 147 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction (br m, 14H, Ar), 8.12 (br m, 2H, Ar). 3 1 P{'H} N M R (121 MHz, CD 2 C1 2 , 300 K): 5 22.3 (s). 3.9.13 PtPdCl4(dmapm). A mixture of EtOH (10 mL) and H 2 0 (4 mL) containing K 2 P t C l 4 (25 mg, 0.060 mmol) and PdCl2(dmapm) (42 mg, 0.057 mmol) was heated to 70 °C for 1.5 h during which a beige precipitate formed. The solvent was removed in vacuo and the residue was dried thoroughly before being taken up in CH 2 C1 2 (10 mL). The slurry was filtered through a mixture of Celite 545 and M g S 0 4 and reduced to ca. 1 mL. Addition of E t 2 0 (10 mL) gave the product as a beige powder which was collected by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried under vacuum. Yield: 39 mg (69 %). Anal. Calcd for C 3 3 H 4 2 N 4 Cl 4 P 2 PdPt : C, 39.6; H , 4.2; N , 5.6. Found: C, 39.9; H , 4.3; N , 5.4. UV-vis (CH 2C1 2): 352 [2000]. *H N M R (400 MHz, CDC1 3, 300 K): 8 2.10 (s, 3H, NC/ / 3 ) , 2.23 (s, 3H, NCr7 3), 2.56 (s, 3H, NC/ / 3 ) , 2.69 (s, 3H, NC# 3 ), 2.70 (s, 3H, N C / / 3 ) , 2.85 (s, 3H, NC/ / 3 ) , 3.30 (s, 3H, N C / / 3 ) , 3.44 (s, 3H, NCtf 3), 4.74 (dt, 1H, CH2, V H H = 19.2, 2JHP = 27.7), 5.27 (dt, 1H, CH2, 2JHH = 19.2, 2JHP = 26.6). 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K): 8 11.8 (s, ]JPPt = 3980), 30.7 (s). Not observed were the two-bond PP coupling or the three-bond HPt coupling with either the C H 2 or N C H 3 protons. 3.9.14 [Pd(p-Cl)(P,P'-dmapm)]2[PF6]2. To a CH 2 C1 2 (4 mL) solution containing PdCl2(dmapm) (110 mg, 0.15 mmol) was added acetone (6 mL) followed by N H 4 P F 6 (220 mg, 1.3 mmol). The resulting orange slurry was stirred at r.t. for 3 h and then reduced to dryness in vacuo. The residue was taken up in CH 2 C1 2 and filtered through Celite 545. The orange filtrate was reduced to ca. 1 mL under vacuum, and EtOH (10 mL) was added to give the product as an orange precipitate. Addition of E t 2 0 (10 mL) gave more of the product that was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried in vacuo. Yield: 77 mg (61 %). Anal. Calcd for C 6 6H8 4 N 8 Cl 2 P 6 Fi 2 Pd2: C, 47.0; H , 5.0; N , 6.6. Found: C, 47.1; H , 5.0; N , 6.6. UV-vis (CH 2C1 2): 360 [14600]. ] H N M R (300 MHz, CDC1 3, 300 K): 8 2.56 (s, 24H, NC/ / 3 ) , 4.91 (t, 2H, CH2, 2JHP = 12.3 Hz), 7.32 (m, 8H, Ar), 7.57 (pt, 4H, Ar), 7.97 (m, 4H, Ar). 148 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K): 5 -52.0 (s), -145.0 (spt, P F 6 \ ' / p f = 710). A M (CH3NO2, 298 K): 213. 3.9.15 PdRhCl3(CO)(dmapm). To a CH2CI2 (5 mL) solution containing PdCi2(dmapm) (230 mg, 0.31 mmol) was added [Rh(u-Cl)(CO)2]2 (61 mg, 0.32 mmol), and this resulted in a brisk effervescence. The volume of the solution was reduced at the pump to ca. 1 mL and E t 2 0 (10 mL) was added to give the product as a beige precipitate. Yield: 250 mg (90 %). Anal. Calcd for C34H42N4Cl30P2PdRh: C, 45.4; H , 4.7; N , 6.2. Found: C, 45.5; H , 4.7; N , 6.0. UV-vis (CH2CI2): 358 [5520]. ' H N M R (300 MHz, CDCI3, 300 K): 5 2.62 (s, 3H, NC//3), 2.72 (s, 18H, NCH3), 3.31 (s, 3H, NCH3), 4.25 (m, 1H, CH2), 4.51 (m, 1H, CH2), 6.93 (m, 1H, Ar), 7.09 (m, 1H, Ar), 7.44 (m, 7H, Ar), 7.65 (m, 4H, Ar), 7.85 (pt, 1H, Ar), 8.08 (m, 1H, Ar), 8.68 (pt, 1H, Ar). 3 1 P{'H} N M R (121 MHz, CDCI3, 300 K): 5 28.1 (s), 41.7 (d, V P R h = 179). A M (298 K) : 363 (H 2 0), 12 (CH3NO2). u c o : 1975 [s]. 3.9.16 Rh2Cl2(CO)2(dmapm). To a combination of [Rh(CO)2(u-Cl)]2 (46 mg, 0.12 mmol) and dmapm (67 mg, 0.12 mmol) was added CH2CI2 (10 mL). Over the course of 2 min, the solution changed from orange to yellow and there was a brisk effervescence due to release of CO. The solution was stirred for 5 min before all but 1 mL of the solvent was removed in vacuo. Addition of E t 2 0 (10 mL) gave the product as a canary yellow powder that was isolated by filtration, washed with E t 2 0 (4 x 2 mL) and dried under vacuum. Yield: 100 mg (95 %). Anal. Calcd for C 3 5 H 4 2 N 4 C l 2 0 2 P 2 R h 2 : C, 47.3; H , 4.8; N , 6.3. Found: C, 46.7; H , 4.9; N , 6.1. *H N M R (300 MHz, CDC1 3, 220 K): 5 2.31 (s, 6H, N C / / 3 ) , 2.73 (s, 6H, NC/ / 3 ) , 2.95 (s, 6H, NC/ / 3 ) , 3.77 (s, 6H, NCH3), 4.64 (br m, 2H, CH2), 6.87 (br m, 2H, Ar), 7.08 (br m, 2H, Ar), 7.35-7.85 (br m, 12H, Ar). 3 1 P{'H} N M R (162 MHz, CDC1 3, 300 K): 5 40.4 (m, V P R h = 180, 2 J P P = 22, V P R h = 1). uCo: 2000 [s]. 3.9.17 Pd2Cl2(dmapm) To a Schlenk tube containing /ra«5-PdCl 2(PhCN) 2 (32 mg, 0.083 mmol) and dmapm (47 mg, 0.084 mmol) was added CH2CI2 (10 mL), and the resulting yellow solution was 149 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction stirred for 15 min. Solid Pd 2 (dba) 3 CHCl 3 (44 mg, 0.043 mmol) was then added and the resulting purple solution was warmed to reflux for 3 h during which it became orange. A l l but 1 mL of solvent was removed in vacuo, and E t 2 0 (20 mL) was added to give the product as an orange powder that was isolated by filtration, washed with E t 2 0 ( 3 x 5 mL) and dried under vacuum. Yield: 54 mg (76 %). Anal. Calcd for C 33H42N 4Cl 2P 2Pd 2: C, 47.2; H, 5.0; N , 6.7. Found: C, 47.5; H , 5.1; N , 6.6. ' H N M R (300 MHz, CDC1 3 , 300 K) : 5 2.44 (s, 12H, NGrY 3), 2.89 (s, 6H, NC/ / 3 ) , 3.07 (s, 6H, NCH3), 3.72 (t, 2H, CH2, 2JHP = 11.3 Hz), 6.99 (pt, 2H, Ar), 7.22 (m, 6H, Ar), 7.45 (m, 4H, Ar), 7.60 (m, 4H, Ar). 3 1 P{'H} N M R (121 MHz, CDC1 3 , 300 K): 5 -29.9 (s). 3.9.18 PtPdCl2(dmapm) To a combination of /ra«s-PdCl 2(PhCN) 2 (130 mg, 0.33 mmol) and dmapm (190 mg, 0.33 mmol) in a Schlenk tube was added CH 2 C1 2 (7 mL). The initially orange solution turned yellow within a few seconds. The solvent was removed at the pump and EtOH (10 mL) was added followed by an aqueous solution (5 mL) containing K 2 P t C l 4 (140 mg, 0.33 mmol). The orange slurry was heated to 70 °C for 1 h when it turned yellow. An ethanolic solution containing K O H (13 mL, 70 mmol L"1) was added over 3 min and the resulting brown solution stirred at 70 °C for an additional 0.5 h. The solvents were then removed in vacuo and the residue dried thoroughly overnight. The residue was partially dissolved in warm C6H6 (30 mL) and then the mixture was filtered through a plug of Celite 545/MgS0 4. The brown filtrate was shown by N M R to contain a combination of unreacted PtPdCl4(dmapm) and the desired Pt'Pd1 dimer. The solid trapped on the Celite plug was then washed through with CH 2 C1 2 (10 mL) and the brown-red filtrate was concentrated to ca. 1 mL at the pump. Addition of E t 2 0 (25 mL) gave the product as a green-brown precipitate that was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried in vacuo. Yield: 99 mg (32 %). Anal. Calcd for C 3 3 H 4 2 N 4 Cl 2 P 2 PdPt : C, 42.7; H, 4.6; N , 6.0. Found: C, 41.5; H , 4.6; N , 5.7. l H N M R (300 MHz, CDCI3, 300 K): 5 2.36 (s, 6H, NCJ73), 2.45 (s, 6H, NC# 3 ) , 2.78 (s, 3H, NC# 3), 3.02 (s, 3H, N C i / 3 ) , 3.07 (s, 3H, NC# 3), 3.19 (s, 3H, NC/ / 3 ) , 3.54 (ddd, 1H, CH2, 2JHH = 15.9 Hz, 2JHP = 7.77), 3.93 (ddd, 1H, CH2, 2 J h h = 15.9 Hz, 2JHp = 7.77), 6.94 (ddd, 1H, Ar, V H H = 6.4 Hz, 3 7 H P = 8.0 Hz, V H P = 1.3 Hz), 7.01 (pt, 1H, Ar), 7.13 (pt, 1H, Ar), 7.24 (m, 4H, Ar), 7.41 (m, 4H, Ar), 7.53 (m, 150 • References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 4H, Ar), 8.18 (dd, 1H, Ar, V H H = 7.5 Hz, V H P = 14.2 Hz). 3 , P{ 'H} N M R (121 MHz, CDC1 3, 300 K): 8 -23.0 (d, V P P = 21.9 Hz, 2 J P P t = 260 Hz, P bound to Pd), -31.7 (d, V P P = 21.9 Hz, 1 J P P t = 4200 Hz, P bound to Pt). 3.9.19 Pd2Cl2(CO)2(dmapm) To a Schlenk tube containing Pd2Cl2(dmapm) (31 mg, 0.037 mmol) was admitted CO (1 atm) followed by E t 2 0 (10 mL) and CH2CI2 (2 mL). The orange slurry was stirred overnight when it became blue-purple. The solid was removed by filtration, washed with E t 2 0 and dried in vacuo. Yield: 21 mg (64 %). Anal. Calcd for C35H42N4Cl202P2Pd2: C, 46.9; H , 4.7; N , 6.2. Found: C, 46.4; H , 4.8; N , 6.3. ' H N M R (300 MHz, CDCI3, 300 K, 1 atm CO): 5 2.67 (t, 2H, CH2, 2JHp = 12.0 Hz), 2.72 (s, 24H, NC/ / 3 ) , 7.12 (m, 8H, Ar), 7.45 (m, 8H, Ar). 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K , 1 atm CO): 5 43.0 (s). IR (KBr pellet): u c o : 1798 [s]. 3.9.20 Pd2Cl2(DEAD)(dmapm)H20 To a CH 2 C1 2 (5 mL) solution containing Pd2Cl2(dmapm) (63 mg, 0.075 mmol) was added D E A D (0.020 mL, 0.13 mmol). During 10 h, the solution changed from red to yellow, and some decomposition to Pd metal occurred. After filtration through Celite 545, the solution was reduced in vacuo to ca. 1 mL, and E t 2 0 (20 mL) was added to give the product as a fine yellow precipitate. This was isolated by filtration, washed with E t 2 0 (3 x 3 mL) and dried under vacuum. Yield: 44 mg (58 %). Anal. Calcd for C 4 iH54N 4 Cl 2 0 5 P 2 Pd 2 : C, 47.9; H , 5.3; N , 5.4. Found: C, 47.8; H , 5.4; N , 5.4. 3 1 P{'H} N M R (121 MHz, CDCI3, 325 K): 5 31.6 (s). ' H N M R (300 MHz, CDCI3, 325 K): § 1.47 (t, 6H, CU2CH3, V H H = 6.9), 1.78 (s, 2H, H 2 0 ) , 2.32 (s, 12H, NC# 3 ) , 2.84 (s, 6H, NC/7 3), 2.89 (s, 6H, NC/ / 3 ) , 3.19 (dt, 1H, PC/ / 2 P, 2JHH = 2JHP = 14.0), 3.70 (m, 5H, PC/ / 2 P and C//2CH3), 6.90 (pt, 2H, Ar), 7.05 (m, 2H, Ar), 7.35 (m, 8H, Ar), 7.50 (pt, 2H, Ar), 8.55 (m, 2H, Ar). 3.9.21 Attempted isolation of a Heck catalytic intermediate A mixture of PtPdCl4(dmapm) (52 mg, 0.052 mmol), D M F (1 mL), a D M F solution containing 1.00 mol L" 1 Phi and 1.00 mol L"1 styrene (1 mL) and an H 2 0 solution 151 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction containing 1.00 mol L"1 K2CO3 (1 mL) was heated to 100 °C for 5 min under air. The volatiles were removed in vacuo and the residue taken up in CeH 6 . The resulting slurry was filtered through Celite 545 and the filtrate was reduced to ca. 0.5 mL. Addition of hexanes (20 mL) gave the product as an orange powder that was isolated by filtration and washed with E t 2 0 ( 3 x 3 mL). The 3 1 P{ 1 H} N M R data for the mixture of complexes thus isolated are given in Table 3.5. 3.9.22 Attempted preparation of PtI2(P,Ar-dmapmO) PdCl2(dmapm) was prepared in situ from ?ra«^-PdCl2(PhCN)2 (89 mg, 0.23 mmol) and dmapm (130 mg, 0.23 mmol) in CH 2 C1 2 (5 mL). To the yellow solution was added cumene hydroperoxide (0.080 mL, 0.43 mmol). After 2.5 h, the solvent was removed in vacuo and the Schlenk warmed gently with a hot air-gun so as to ensure the removal of all traces of the peroxide. CAUTION: cumene hydroperoxide may explode violently if heated above 82 °C. Solid K C N (98 mg, 1.5 mmol), EtOH (5 mL) and H 2 0 (3 mL) were added. The yellow colour disappeared within a few seconds. After 20 min the solvents were removed and the residue taken up in CH 2 C1 2 (10 mL). The slurry was filtered through a mixture of Celite 545 and M g S 0 4 into a Schlenk tube containing PtCl2(cod) (81 mg, 0.22 mmol). An intense yellow colour developed and quickly faded. After 0.5 h, the solution was reduced in vacuo to ca. 1 mL, and E t 2 0 (20 mL) was added to give the product as an off-white powder. This was isolated by filtration, washed with E t 2 0 (3 x 3 mL) and dried under vacuum. *H N M R spectroscopy indicated that this product contained a significant proportion of PtCl2(cod). Nevertheless, the crude mixture was treated with Nal in a manner similar to that outlined in Section 3.9.12 in order to form the desired product. 3 1 P{'H} N M R data for this compound and its chloro analogue appear in Section 3.6.2.4 (p. 133). 3.9.23 Heck reactions. The catalyst precursor (ca. 1-2 umol) was dissolved in a stock D M F solution containing 1.00 mol L"1 Phi and 1.00 mol L" 1 styrene (2 mL). This was diluted by the addition of neat DMF (1 mL) and preheated to 100 °C in a 20 mL three-necked round-bottom flask fitted with a condenser. Over the course of about 30 s, a 1.00 mol L" 1 stock solution of K 2 C03 152 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction in H2O (2 mL) was added. The beginning of this addition marked zero time. The temperature was maintained at 100 °C throughout the course of the reaction on a thermostatted oil-bath. No precautions were taken to exclude O2. (If the reaction solution is allowed to cool following the catalysis (2-3 h), the stilbene product precipitates and can be isolated by simple filtration.) Aliquots of the reaction mixture (0.1 mL) were withdrawn at 10 min intervals and diluted by addition to CH 2 C1 2 (1 mL) at r.t. The CH 2 C1 2 solution (0.1 uL) was then analysed by GC in order to determine the extent of reaction. This was accomplished by comparing the relative peak areas for Phi and cis- and fr-ans-stilbene. Analysis of solutions containing known concentrations of Phi and trans-s\\\bzne indicated that a weighting factor of 2.32 needed to be applied to the area of the peak due to Phi to compensate for its lower response factor at the FID. This ensured that the use of relative peak areas gave accurate representations of the relative concentrations of the two compounds in solution. The peak due to styrene was not used in the analysis due to its overlap with the tail of the peak due to DMF. The GC parameters are summarised in Table 3.8, and the component elution times under these conditions are given in Table 3.9. Table 3.8 GC parameters used in the determination of Heck reaction components. Parameter Setting Initial oven temperature 80 °C Final oven temperature 220 °C Rate 20 °C min 1 Initial time 2 min Final time 1.5 min Injector temperature 220 °C Detector temperature 220 °C Column head pressure 105 kPa 153 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction Table 3.9 Elution times for Heck reaction components under the GC conditions given in Table 3.8. Component Retention time (min) Styrene 2.30 Iodobenzene 4.18 G's-stilbene 8.26 Trarcs-stilbene 9.74 3.10 References 1. Heck, R. F. J. Am. Chem. Soc. 1968, 90, 5518. 2. de Meijere, A. ; Meyer, F. E. Angew. Chem. Int. Ed. Engl. 1994, 33, 2379. 3. Crisp, G. T. Chem. Soc. Rev. 1998, 27, 427. 4. Shaw, B. L. New J. Chem. 1998, 77. 5. Dieck, H . A. ; Heck, R. F. J. Am. Chem. Soc. 1974, 96, 1133. 6. Ziegler, C. B.; Heck, R. F. J. Org. Chem. 1978, 43, 2941. 7. Frank, W. C ; Kim, Y . C ; Heck, R. F. J. Org. Chem. 1978, 43, 29Al. 8. Ziegler, C. B.; Heck, R. F. J. Org. Chem. 1978, 43, 2949. 9. Cortese, N . A. ; Ziegler, C. B.; Hrnjez, B. J.; Heck, R. F. J. Org. Chem. 1978, 43, 2952. 10. Shaw, B. L.; Perera, S. D. Chem. Commun. 1998, 1863. 11. Herrmann, W. A. ; Brossmer, C ; Ofele, K. ; Reisinger, C.-P.; Priermeier, T.; Beller, M . ; Fischer, H. Angew. Chem. Int. Ed. Engl. 1995, 34, 1844. 12. Ohff, M.; Ohff, A. ; van der Boom, M . E.; Milstein, D. J. Am. Chem. Soc. 1997, 119, 11687. 13. Bergbreiter, D. E.; Osburn, P. L.; Liu, Y.-S. J. Am. Chem. Soc. 1999,121, 9531. 14. Reddy, K . R.; Surekha, K. ; Lee, G.-H.; Peng, S.-M.; Liu, S.-T. Organometallics 2000, 19, 2637. 15. Bohm, V . P. W.; Herrmann, W. A. Chem. Eur. J. 2000, 6, 1017. 16. Jeffery, T. Tetrahedron Lett. 1994, 35, 3051. 17. Blart, E.; Genet, J. P.; Safi, M . ; Savignac, M . ; Sinou, D. Tetrahedron 1994, 50, 505. 18. Hessler, A. ; Stelzer, O. J. Org. Chem. 1997, 62, 2362. 154 References on page 154 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 19. Bankston, D.; Fang, F.; Huie, E.; Xie, S. J. Org. Chem. 1999, 64, 3461. 20. Puddephatt, R. J. Chem. Soc. Rev. 1983,12, 99. 21. Vance, D. H. ; Czarnik, A. W. Am. Chem. Soc. 1993,115, 12165. 22. Liu, S.; Luo, Z.; Hamilton, A. D. Angew. Chem. Int. Ed. Engl. 1997, 36, 2678. 23. Young, M . J.; Chin, J. J. Am. Chem. Soc. 1995,117, 10577. 24. Chapman, W. H. ; Breslow, R. J. Am. Chem. Soc. 1995, 117, 5462. 25. Meyer, F.; Kaifer, E.; Kircher, P.; Heinze, K.; Pritzkow, H . Chem. Eur. J. 1999, 5, 1617. 26. Meyer, F.; Hyla-Krypsin, I.; Kaifer, E.; Kircher, P. Eur. J. Inorg. Chem. 2000, 771. 27. McKenzie, C. J.; Robson, R. J. Chem. Soc, Chem. Commun. 1988, 112. 28. Matthews, R. C ; Howell, D. K.; Peng, W.-J.; Train, S. G.; Treleaven, W. D.; Stanley, G. G. Angew. Chem. Int. Ed. Engl. 1996, 35, 2253. 29. Broussard, M . E.; Juma, B.; Train, S. G.; Peng, W.-J.; Laneman, S. A. ; Stanley, G. G. Science 1993, 260, 1784. 30. Kalck, P. Polyhedron 1988, 7, 2441. 31. Suss-Fink, G. Angew. Chem. Int. Ed. Engl. 1994, 33, 67. 32. Jenck, J.; Kalck, P.; Pinelli, E.; Siani, M . ; Thorez, A . J. Chem. Soc, Chem. Commun. 1998, 1428. 33. Sola, E.; Bakhmutov, V . I.; Torres, F.; Elduque, A. ; Lopez, J. A. ; Lahoz, F. J.; Werner, H. ; Oro, L. A . Organometallics 1998,17, 683. 34. Torres, F.; Sola, E.; Elduque, A. ; Martinez, A. P.; Lahoz, F. J.; Oro, L. A . Chem. Eur. J. 2000, 6, 2120. 35. Konsler, R. G.; Karl, J.; Jacobsen, E. N . J. Am. Chem. Soc. 1998,120, 10780. 36. Esteruelas, M . A. ; Garcia, M . P.; Lopez, A. M . ; Oro, L. A . Organometallics 1991, 10, 127. 37. Bergbreiter, D. E.; Osburn, P. L.; Wilson, A. ; Sink, E. M . J. Am. Chem. Soc. 2000,122, 9058. 38. Dixon, K . R. in Multinuclear NMR; Mason, J., Ed.; Plenum: New York, 1987. 39. Xie, Y . Towards Transition Metal Catalyzed Hydration of Olefins; Aquo Ions and Pyridylphosphine-Platinum and Palladium Complexes; Ph.D. Dissertation, University of British Columbia: Vancouver, 1990. 40. Partenheimer, W.; Hoy, E. F. Inorg. Chem. 1973,12, 2805. 155 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 41. L i , C ; Cucullu, M . E.; Mclntyre, R. A. ; Stevens, E. D.; Nolan, S. P. Organometallics 1994,13, 3621. 42. Luo, L.; Zhu, N . ; Zhu, N.-J.; Stevens, E. D.; Nolan, S. P.; Fagan, P. J. Organometallics 1994, 13, 669. 43. Xiao, S.-X.; Trogler, W. C ; Ellis, D. E.; Berkovich-Yellin, Z. J. Am. Chem. Soc. 1983, 105. 44. Marynick, D. S. J. Am. Chem. Soc. 1984, 106, 4064. 45. Orpen, A. G.; Connelly, N . G. Organometallics 1990, 9, 1206. 46. Dunne, B. J.; Morris, R. B.; Orpen, A. G. J. Chem. Soc, Dalton Trans. 1991, 653. 47. Gimarc, B. M . Molecular Structure and Bonding; Academic Press: New York, 1979. 48. Albright, T. A. ; Burdett, J. K.; Whangbo, M. -H . Orbital Interactions in Chemistry; Wiley: New York, 1985. 49. Appleton, T. G.; Clark, H . C ; Manzer, L. E. Coord. Chem. Rev. 1973, 10, 335. 50. Orpen, A . G.; Brammer, L.; Allen, F. H. ; Kennard, O.; Watson, D. G. J. Chem. Soc. Dalton Trans. 1989, SI. 51. Steffen, W. L.; Palenik, G. J. Inorg. Chem. 1976,15, 2432. 52. Barral, M . C ; Jimenez-Aparicio, R.; Royer, E. C ; Saucedo, M . J.; Urbanos, F. A. Inorg. Chim. Acta 1993, 209, 105. 53. Ma, E. S. F. Reactivity and Coordination Chemistry of Ru(II) Aminophosphine Complexes with H2S, Thiols, H2O and Other Small Molecules; Ph.D. Dissertation, University of British Columbia: Vancouver, 1999. 54. Mudalige, D. C ; Ma, E. S.; Rettig, S. J.; James, B. R.; Cullen, W. R. Inorg. Chem. 1997, 36, 5426. 55. Pringle, P. G.; Shaw, B. L. J. Chem. Soc, Chem. Commun. 1982, 956. 56. Geary, W. J. Coord. Chem. Rev. 1971, 7, 81. 57. Schrock, R. R.; Osborn, J. A . J. Am. Chem. Soc. 1971, 93, 2397. 58. Budzelaar, P. H. M . ; Frijns, J. H. G.; Orpen, A . G. Organometallics 1990, 9, 1222. 59. Cobas, C ; Cruces, J.; Sardina, J. Magnetic Resonance Companion; Ver. 2.3. 60. Balch, A . L.; Benner, L. S. Inorg. Synth. 1982, 21, 47. 61. Benner, L. S.; Balch, A. L. J. Am. Chem. Soc. 1978,100, 6099. 62. Colton, R.; Farthing, R. H.; McCormick, M . J. Aust. J. Chem. 1973, 26, 2607. 156 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 63. Holloway, G. R.; Penfold, B. R. J. Chem. Soc, Chem. Commun. 1976, 485. 64. Doonan, D. J.; Balch, A . L.; Goldberg, S. Z.; Eisenberg, R.; Miller, S. J. J. Am. Chem. Soc. 1975,97, 1961. 65. Goldberg, S. Z.; Eisenberg, R. Inorg. Chem. 1976,15, 535. 66. Tani, K. ; Nakamura, S.; Yamagata, T.; Kataoka, Y . Inorg. Chem. 1993, 32, 5398. 67. DuBois, D. L.; Miedaner, A. ; Haltiwanger, R. C. J. Am. Chem. Soc. 1991, 113, 8753. 68. Lee, C.-L.; James, B. R.; Nelson, D. A. ; Haller, R. T. Organometallics 1984, 3, 1360. 69. Lee, C.-L.; Bensenyei, G.; James, B. R.; Nelson, D. A. ; Lilga, M . A. J. Chem. Soc, Chem. Commun. 1985, 1175. 70. Besenyei, G.; Lee, C.-L.; Gulinski, J.; Rettig, S. J.; James, B. R.; Nelson, D. A. ; Lilga, M . A. Inorg. Chem. 1987, 26, 3622. 71. Xie, L. Y . ; James, B. R. Inorg. Chim. Acta 1994, 217, 209. 72. Maisonnat, A. ; Fair, J. P.; Balch, A. L. Inorg. Chim. Acta 1981, 53, L217. 73. Xie, Y . ; Lee, C.-L.; Yang, Y . ; Rettig, S. J.; James, B. R. Can. J. Chem. 1992, 70, 751. 74. Pringle, P. G.; Shaw, B. L. J. Chem. Soc. Dalton Trans. 1983, 889. 75. Pringle, P. G.; Shaw, B. L. J. Chem. Soc, Chem. Commun. 1982, 81. 76. Ugo, R.; Cariati, F.; Monica, G. L. Inorg. Synth. 1968, 11, 105. 77. Olmstead, M . M . ; Hope, H. ; Benner, L. S.; Balch, A . L. J. Am. Chem. Soc. 1977, 99, 5502. 78. Benner, L. S.; Olmstead, M . M . ; Balch, A . L. J. Organomet. Chem. 1978, 153, C31. 79. Pamplin, C. B. Palladium Complexes of l,l-bis(dimethylphosphino)methane; Ph.D. Dissertation, University of British Columbia: Vancouver, 2001. 80. James, B. R. Pure Appl. Chem. 1997, 69, 2213. 81. Besenyei, G.; Parkanyi, L.; Simandi, L. I.; James, B. R. Inorg. Chem. 1995, 34, 6118. 82. Wong, T. Y . - H . Solution H2S Chemistry of Pd-Bis(diphenylphosphino)methane (DPM) Complexes; Catalytic Conversion of H2S to H2; Ph.D. Dissertation, University of British Columbia: Vancouver, 1996. 157 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 83. Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry; 1st ed.; University Science Books: M i l l Valley, 1987. 84. Muetterties, E. L. ; Pretzer, W. R.; Thomas, M . G.; Beier, B. F.; Thorn, D. L.; Day, V . W.; Anderson, A. B. J. Am. Chem. Soc. 1978,100, 2090. 85. Bennett, M . A. ; Johnson, R. N ; Robertson, G. B.; Turney, T. W.; Whimp, P. O. Inorg. Chem. 1976, 15, 97. 86. Day, V . W.; Abdel-Meguid, S. S.; Dabestani, S.; Thomas, M . G.; Pretzer, W. R.; Muetterties, E. L. J. Am. Chem. Soc. 1976, 98, 8289. 87. Mague, J. T. Inorg. Chem. 1989, 28, 2215. 88. Dickson, R. S.; Hames, B. W.; Cowie, M . A. Organometallics 1984, 3, 1879. 89. Balch, A . L.; Lee, C.-L.; Lindsay, C. H. ; Olmstead, M . M . J. Organomet. Chem. 1979, 177, C22. 90. Rosenberg, E.; Laine, R. M . in Catalysis by Di- and Polynuclear Metal Cluster Compounds; Adams, R. D. and Cotton, F. A. , Eds.; Wiley-VCH: New York, 1998. 91. Gladfelter, W. L.; Roesselet, J. in Chemistry of Metal Cluster Complexes; Shriver, D. F., Kaesz, H. D. and Adams, R. D., Eds.; V C H : New York, 1990, p 329. 92. Gates, B. C. Chem. Rev. 1995, 95, 511. 93. Whitesides, G. M . ; Hackett, M . ; Brainard, R. L.; Lavalleye, J. P. M . ; Sowinski, A. F.; Izumi, A . N . ; Moore, S. S.; Brown, D. W.; Staudt, E. M . Organometallics 1985, 4, 1819. 94. Anton, D. R.; Crabtree, R. H. Organometallics 1983, 2, 855. 95. Adams, R. D.; Falloon, S. B. Organometallics 1995,14, 4594. 96. Adams, R. D.; Falloon, S. B.; Perrin, J. L.; Queisser, J. A. ; Yamamoto, J. H. Chem. Ber. 1996, 129, 313. 97. Bradley, J. S. J. Am. Chem. Soc. 1979,101, 7419. 98. Knifton, J. F. J. Am. Chem. Soc. 1981,103, 3959. 99. Dombek, B. P. J. Am. Chem. Soc. 1981,103, 6508. 100. Knifton, J. F. Chem. Eng. News, May 5 1986; p 43. 101. Fritz, H . P.; Gordon, I. R.; Schwarzhans, K. E.; Venanzi, L. M . J. Chem. Soc. 1965, 5210. 102. Bumagin, N . A. ; Bykov, V . V. ; Sukhomlinova, L. I.; Tolstaya, T. P.; Beletskaya, I. P. J. Organomet. Chem. 1995, 486, 259. 158 Chapter 3: Late Transition Metal Complexes of dmapm, and the Heck Reaction 103. Grushin, V . V. ; Alper, H. Organometallics 1993,12, 1890. 104. Amatore, C.; Jutand, A. ; M'Barki, M . A. Organometallics 1992,11, 3009. 105. Creighton, T. E. Proteins: Structures and Molecular Properties; Second ed.; W. H. Freeman and Company: New York, 1993. 106. Abugnim, C ; Amer, I. J. Mol. Catal. 1993, 85, L275. 107. Kacker, S.; Kim, J. S.; Sen, A. Angew. Chem. Int. Ed. Engl. 1998, 37, 1251. 108. Davis, J. L.; Arndtsen, B. A. Organometallics 2000, 19, 4657. 109. Dghaym, R. A. ; Yaccato, K. J.; Arndtsen, B. A. Organometallics 1998,17, 4. 110. Lafrance, D.; Davis, J. L.; Dhawan, R.; Arndtsen, B. A . Organometallics 2001, 20, 1128. 111. Jiang, Z.; Sen, A. J. Am. Chem. Soc. 1995,117, 4455. 112. Pamplin, C. B. (personal communication). 159 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4 Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.1 Introduction In this chapter, the coordination chemistry of Pt and Pd with heteropolydentate ligands incorporating both "hard" (N) and "soft" (P) donors is extended to include the 2-pyridyldiphosphines. The use of pyridylmonophosphines (i.e., those containing only one P-atom and 1-3 pyridyl groups) as ligands for transition metals has generated an extensive body of literature, and is the subject of a 1993 review.1 In addition to a large number of monometallic complexes,2 - 6 several bimetallic species 7 - 1 1 as well as clusters 1 2 ' 1 3 and oligomers,1 4 the last three classes employing the ligand in a P,N-bridging mode, have been characterised. In addition, some homogeneous catalytic applications, including hydrogenation,1 hydroformylation,7'1 5'1 6 and carbonylatioh17 have been investigated. A widely used pyridylmonophosphine is 2-pyridyldiphenylphosphine, PPh2(py) (Type I, Chart 4.1; denoted PNi throughout this thesis), which as a ligand can adopt three different coordination modes: P-monodentate,18-22 P,/V-chelate2>4 and p/V-bridge. 7 - 1 4 Di -and tri-substituted 2-pyridylphosphines such as PPh(py)2 (PN 2), P(py)3 (PN3) and XPPh 3 . „(py)„ (X = O, S, AuCl , Au(C6F5); n = 2, 3) can take on a variety of additional coordination modes using a combination of one or more of the Group 15/16 donor centres. 2 3 - 2 7 Recent work within this laboratory has found an unprecedented P,N,N'-coordination mode for PPh(py)2 and P(py)3.28 Detailed introductions to the 2-pyridylmonophosphines, their complexes, and catalytic applications can be found in the PhD theses of X i e , 2 9 Schutte30 and LePage.31 In contrast to the vast number of reports concerning the pyridylmonophosphines, pyridyW/phosphines have been less studied and, of those documented, two distinct structural classes are apparent. In the first class, two -PR 2 moieties (R = Ph, typically) are directly bonded in the 2- and 6-positions of the pyridine ring, affording a P/VP-ligand system (Type I I ) . 3 2 - 3 5 Ligands of this type readily bridge metal centres and have been 160 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes used to stabilize both homo- and heterobimetallic complexes. Introduction of a -CH(R)-(R = H, Me) spacer between the -PPh 2 groups and the central pyridyl group (Type 111)36,37 permits extra flexibility within the iWP-tridentate ligand and, in the case of R = Me, allows for the synthesis of C2-symmetric chiral complexes, some of which have been used for the enantioselective hydrogenation of imines. 3 8 This type of ligand favours bis five-membered chelate ring formation upon coordination to a mononuclear metal fragment.39-42 In the second class, two -PR(py) (R = Ph, py) fragments are linked by an aliphatic carbon spacer of variable length, linear, branched or cyclic (Type IV). 4 3 - 4 5 Chart 4.1 Type I Type II Type III Type IV (R = Ph,py) (R = Ph, typically) (R = H, Me) (R = Ph,py) 4.2 Scope This chapter reports the synthesis and characterisation (Section 4.7.1) of a new pyridyldiphosphine ligand containing a cyclopentane ring bridge between the two P-atoms, dpypcp (Chart 4.2), which is isostructural with the previously described tetraphenyl chiral analogue, dppcp.4 6 The protonated (see Section 4.7.4 for synthesis, and Section 4.3.3 for discussion) and oxidised (Sections 4.7.2 and 4.3.2) forms of this ligand together with those of the known dpype ligand (see Sections 4.7.3 and 4.7.5 for syntheses) are also described. 161 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Chart 4.2 dpype dpypcp The synthesis and characterisation of the range of P,P '-bonded complexes MX 2 (P -P) (M = Pd, Pt; X = CI, Br, I; P-P = dpypcp, dpype) are presented (Sections 4.7.7-4.7.11 and 4.7.15^1.7.17) together with the molecular structures of PtCl2(dpype) and Ptl2(dpypcp) (Section 4.4.2). In addition, a pair of coordination polymers which bear P.P '-chelated, P,//-bridged [Pt 2(P-P) 2] 4 + "tethered paddlewheel" clusters connected by A g N 0 3 "bridges" are reported (Section 4.4.2.2). The reactions of dpypcp and dpype with M(0) precursors to give tetrahedral complexes of the type M(PPh 3) 2(P-P) (see Sections 4.7.22 and 4.7.23 for syntheses) are discussed in Section 4.4.3. Reactions of the Pt complexes with electron-deficient olefins, which led to an investigation into cis to trans isomerisation of the coordinated olefin, and the molecular structure of the olefin-containing complex, Pt(n2-dmf)(dpypcp), are presented (Section 4.4.3.1). 4.3 The Pyridyldiphosphine Ligands and Derivatives 4.3.1 History and synthesis The first report of a pyridylphosphine dates to 1944.47 Davies and Mann prepared, among other tertiary phosphines in which the P-atom was linked to 3 different alkyl or aryl groups, the compounds 2- and 3-pyridylphenyl-p-bromophenylphosphine, and also tri-2-pyridylphosphine (PN 3) using standard Grignard methods. In 1948, 2-pyridyldiphenylphosphine (PN,) and di-2-pyridylphenylphosphine (PN 2) were prepared by Mann and Watson, also by Grignard methods 4 8 These syntheses typically gave low yields (e.g., 16% for PN 2 ) and required high temperature (160-180 °C) fractional 162 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes vacuum distillation as a purification step. Although Davies and Mann had used L i in the preparation of phenyl-p-bromophenylethylphosphine,47 it was not until the introduction of 2-lithiopyridine by Wibaut et al. in 1952,4 9 and the development of a revised procedure by Plazek and Tyka in 1957,5 0 that a higher-yielding, general route into the pyridylphosphines was uncovered. Reaction of 2-lithiopyridine and the appropriate chlorophosphine remains today as the standard synthetic strategy (see, for example, Scheme 4.1). Scheme 4.1 L l . P C I 2 + / CI 2 P ,\ >)2 + 4 L i C I N-The syntheses of dpype and dpypcp represent modifications to the standard procedure for making P N 3 5 1 which has been adopted in this laboratory and described in detail by previous researchers X i e , 2 9 Schutte30 and LePage.3 1 Although dpype was first reported in a 1986 Smithkline Beckman patent52 (its Au 1 complexes effectively inhibiting the growth of animal tumor cells) and was mentioned in the open literature 1987,4 4 no synthetic or characterisation data appeared until the publication by Baird 3 of this group in 1995. The ligand was prepared according to Baird's procedure during the course of this work, and its molecular structure was reported in 1999.53 Berners-Price and coworkers devised an alternate scheme for the synthesis of dpype which involved cleavage of P N 3 by L i to generate lithium di-2-pyridylphosphide, followed by reaction with 1,2-dibromoethane.54 These workers also reported the synthesis of the 3- and 4-pyridyl analogues according to a variation of the standard route. The ligand dpypcp was first made by a Smith, a postdoctoral fellow in this laboratory during 1993-1994.55 It was characterised more fully by this author and reported together with the Pt11 complexes which constitute a significant portion of this chapter, and some Ru 1 1 compounds, in 1999.53 The molecular structure was subsequently determined by X-ray crystallography and appears in the PhD thesis of LePage.3 1 The material is synthesised as a racemic mixture of R,R and S,S enantiomers, where the 163 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes absolute configuration designations refer to the two methine C-atoms in the cyclopentane "backbone" linking the two P-atoms (Chart 4.2 and Section 4.7.1). The compound is a white solid which is freely soluble in chlorinated solvents, partially soluble in acetone, Et20 and alcohols, and sparingly soluble in H2O. It is freely soluble in dil. HC1 (Section 4.3.3). 4.3.2 Pyridyldiphosphine dioxides The syntheses of metal phosphine complexes (particularly air-sensitive materials) are often accompanied by the formation of phosphine oxides which arise because of the presence of adventitious H2O or O2, e.g., Pt(PPh3)4, which is made via reduction of [PtCU] " in the presence of excess PPI13, actually catalyses the oxidation of the phosphine.56 The pyridyldiphosphine dioxides were therefore made in order to establish their absence in the syntheses and reactions of, particularly, Pt° complexes (Sections 4.7.22-4.7.27), and for completeness of characterisation (Sections 4.7.2 and 4.7.3). The standard method for the synthesis of pyridylphosphine oxides used in this laboratory is via reaction of aq. H2O2 and an organic solution containing the P m compound in a two-phase system.30 The synthesis presented here is a modification which involves only one phase. Thus, the ligand was dissolved in dil. HC1 and peroxide was added. After the oxidation reaction was complete, K O H was added to deprotonate the oxidised ligand. At this point it was expected that the product would precipitate but, surprisingly, the pyridyldiphosphine dioxides show enhanced solubility in H2O compared to their P 1" analogues, and a CH2CI2 wash was required to remove the dissolved oxide from the water phase. The ligand dpypcp also reacts smoothly with Ss in refluxing CeH6 to give the pyridyldiphosphine disulphide (Section 4.7.6). The most useful spectroscopic "handles" for the pyridyldiphosphines and their dioxides, together with those of P N 2 and OPN2, are given in Table 4.1. 164 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Table 4.1 Spectroscopic data for pyridylphosphine ligands and their oxides (tw = this work). Compound J , P{ 'H} NMR chemical shift (ppm)a ' H NMR chemical shift of H6 protons (ppm)a'b vP=o (cm"')c Ref. P N 2 -2.6 8.72 — 30 OPN 2 17.3 8.79 1194 30 dpype -6.1 8.63 — 3 dpypcp -2.2 8.38, 8.49d — tw dpype(0)2 29.1 8.70 1206 tw dpypcp(0)2 35.8 7.90, 8.45 (D 2 0) d 1194 tw 38.7 (D 2 0) " Measured in CDCI3 solution unless otherwise noted. b These peaks appear as pseudo-doublets. c K B r pellet. d dpypcp and dpypcp(0)2 have two diastereotopic sets of H 6 protons. Apparent from Table 4.1 is that the pyridyldiphosphine ligands and their dioxides are spectroscopically very similar to each other and to their P N 2 counterparts. This is to be expected, but is somewhat surprising in light of their different reactions with acids (Section 4.3.3). Oxidation of the P-atom shifts the 3 1 P resonances dramatically downfield as expected, but has little impact on the H6 resonances. The peaks due to the H6 protons (bonded to C immediately adjacent to the pyridyl-N), easily identified by their characteristic downfield chemical shifts and multiplicity (pd),57>58 are useful in determining the number of equivalent pyridyl rings in a given metal complex (by integration) and therefore give insight into the bonding mode of the ligand. As judged by the effects of oxidation, the §(He) shifts of these peaks should not be greatly perturbed by P,P '-bonding of a pyridyldiphosphine ligand but should be affected i f N-bonding occurs. 4.3.3 Protonated pyridyldiphosphines The protonated pyridyldiphosphine [dpypcp(H)2]2+ has been prepared by LePage3 1 via reaction of triflic acid and the neutral ligand in EtOH. Such protonated ligands can also be made by adding K P F 6 to a dil. HC1 solution (pH ~ 0) containing the ligands; the insoluble [P-P(H) 2][PF 6] 2 precipitates (P-P = dpype, dpypcp) (Sections 4.7.4 and 4.7.5). Because each proton requires a high molecular weight PF6* counter-ion, elemental analysis unequivocally establishes the presence of only 2 protons {e.g., Anal. Calcd for [dpypcp(H)2][PF6]2: C, 40.9; H , 3.6; N , 7.6. Found: C, 40.9; H , 3.6; N , 7.5.). This result is 165 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes perhaps surprising in that the reactions are conducted at a pH well below the pK a of pyridinium (5.25) and each of the ligands has 6 basic sites: 4 pyridyl N-atoms and 2 P-atoms. It should be noted that Mann and Watson isolated both PN2-2HC1 and PN3-3HC1 and thereby showing that there is no steric nor electrostatic reason why the number of protons taken up by the PN„ family should not equal « 4 8 In principle, each of the ligands here under study could "take up" a minimum of 4 protons. By titration of a dil. HC1 solution containing dpypcp with NaOH, pK a values of 3.66 ± 0.02 and 4.77 ± 0.02 were determined for [dpypcp(H)2]2+ (Section 4.7.28). The dependence of pH on volume of NaOH added for an HC1 solution containing dissolved dpypcp is shown in Figure 4.1. Also shown, is the number of protons bound by the ligand, n, as a function of pH. This curve has a maximum of n = 2, in agreement with the elemental analysis data for the isolated [dpypcp(H)2][PFe]2. One significant difference between the pyridyldiphosphine ligands and the PN„ family is that in the former, the P-atom is bonded to 2 aromatic rings and an alkyl fragment whereas in the latter, it is bonded solely to aromatic rings. Protonation at one of the pyridyl rings causes a decrease in electron density at the remaining pyridyl rings via an inductive mechanism mediated by the central P-atom;4 8 thus, electronic factors at this atom are crucially important in the controlling the number of protonations. For example, the tertiary amine Npy 3 (py = 2-pyridyl) takes up only 2 protons when exposed to a saturated solution of HC1 in E tOH; 4 8 the inductive deactivating effect of a protonation on a pyridyl rings appears to be transmitted more efficiently through an N-atom than through a P-atom. In addition, O P N 3 takes up 1 equiv. of picric acid (pK a 0.38) while P N 3 takes up 2 equiv.4 8 Perhaps the substitution of one aromatic ring on the P-atom for an alkyl fragment is sufficient to enhance the ability of phosphorus to transmit the inductive effect and so reduce the propensity for multiple protonations. 166 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 12 1 Figure 4.1 TOP: pH dependence on vol. of NaOH (1.094 x 10"1 mol L"1) added for an HC1 solution (1.135 x 10"2 mol L"1) containing dissolved dpypcp (6.112 x 10"4 molL" 1) (solid line), and the HC1 solution alone (dashed line). B O T T O M : number of protons bound per dpypcp molecule (« ) , as a function of pH. As judged by their 3 1 P{'H} N M R spectra, the protonated ligands bear chemically equivalent P-atoms. A plausible picture of, e.g., [dpype(Ff)2] , is shown in Chart 4.3, 167 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes where a proton is "chelated" by both pyridyl N-atoms. This is similar to the basic action of proton sponge.59-60 Chart 4.3 2+ 4.4 Synthesis and Characterisation of Pyridyldiphosphine Complexes 4.4.1 A brief literature survey The first report of a Type IV pyridyldiphosphine-transition metal complex was that in 1987 of an Au 1 complex of dpype, although no synthetic or characterisation data were presented.44 This was followed by the 1990 description by Budzedlaar et al. of bimetallic Rh 1 complexes supported by (Ph)(py)P(X)P(py)(Ph) (X = C H 2 , C 2 H 4 , C 3 H 6 ; py = 2-pyridyl). 4 5 James's group was first to report the synthesis and characterisation of dpype,3 this 1995 report also describing some N i " complexes of the ligand. The N i work was continued by LePage who extended it to include dpypcp.31 In 1998, Berners-Price et al. reported A g 1 complexes of dpype and uncovered their solution behaviour,43 and a year later published on the analogous Au 1 compounds61 and a review covering the syntheses, structures and solution behaviour of these complexes, as well as a summary of their in vitro activity against human tumour cell lines. 6 2 In the same year, this author and other members of the James group reported the syntheses of a range of Pt" and Ru" complexes, and preliminary results for the catalytic hydrogenation of imines using the Ru compounds.53 The known coordination modes of the Type IV pyridyldiphosphine ligands with transition metal centres are illustrated in Chart 4.4, and the respective metal centres are given in Table 4.2. 168 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Chart 4.4 ( > M P ^ N ' < ) ^ P II III .M ' M Table 4.2 The known coordination modes of the Type IV pyridyldiphosphine ligands and their respective metal centres (tw = this work). Coordination Mode Metal ret I II III IV V VI Ni(0) 3 1 Ni(II)33i Pd(H) t w Ag(I)43 Pt(0)1' Pt(n)53-twAu(I)6i Ru(II)5 3 Ag(I) 43Au(I) 6 1 Rh(I)45 Pt(II) tw Rh(I) 4 5 Pt(II)[M' =Ag(I)]63- tw 4.4.2 Platinum(II) and palladium(II) pyridyldiphosphine complexes The initial work on the synthesis and characterisation of M 1 1 (M = Pd, Pt) complexes of the pyridyldiphosphine ligands dpype and dpypcp was performed by Smith in this laboratory.55 Although some of the compounds which form the basis of this chapter had thus been made previously, sufficient characterisation (assigned ' H N M R spectra, elemental analyses, etc) was lacking. In addition, no crystal structures had been obtained. 169 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes A more thorough investigation of these complexes was undertaken as the focus of my undergraduate research project, and was continued as the initial phase of this dissertation. 4.4.2.1 Monometallic platinum(II) complexes The synthesis and chemistry of Pt" (halo)pyridyldiphosphine complexes (type I, Chart 4.4) are unremarkable and closely mimic those of dppe. The precursor PtCl2(cod) reacts cleanly and quantitatively with one or two equiv. of the pyridyldiphosphine ligands at r.t. in CH 2 C1 2 solution to form complexes of the type PtCl 2(P-P) or [Pt(P-P) 2]Cl 2, respectively (P-P = dpype, dpypcp) (Sections 4.7.7 and 4.7.10). (In making PtCi2(P-P), care must be taken to administer a dilute solution of the ligand to a concentrated solution of the metal precursor to prevent the formation of [Pt(P-P)2]Cl2.) The compounds are white, microcrystalline solids, and are freely soluble in CH2CI2, partially soluble in CHCI3, slightly soluble in low molecular weight alcohols and acetone, and insoluble in hexanes and Et20. In addition, the ionic complexes are partially soluble in water at r.t., while the neutral ones are completely insoluble. At 100 °C in 0.1 mol L" 1 maleic acid (catalytic hydration conditions, Chapter 5), all the compounds show a minimum solubility of ca. 1 mg mL"1. The compounds PtX2(P-P) (X = Br, I) can be made either from the corresponding chloro compounds via halide metathesis with NaX in acetone/water (Br) or acetone (I) solution, or by reaction of the appropriate ligand with PtX2(cod) (Sections 4.7.8, 4.7.9 and 4.7.11). Like the chloro complexes, PtBr2(dpypcp) is white; the iodo complexes Pti2(P-P) are yellow. The chloride counter-ions in the ionic complexes [Pt(P-P)2]Cl2 are easily substituted by PFV by reaction with NH4PF6 in acetone (Sections 4.7.13 and 4.7.14). In all of these compounds, the ligands are exclusively P,P '-bonded as indicated by their 3 1 P{'H} N M R spectra (Table 4.3) and representative crystal structures. The spectra show singlets for all compounds (two, in the case of [Pt(dpypcp)][PF6]2 due to diastereomers - vide infra) which indicate chemical equivalence of the 2 P-atoms due to molecular C2-symmetry. In addition, large positive coordination shifts {e.g., A8 = 53.2 for 170 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes PtCbCdpype), 20.1 for PtCi2(dpypcp)) are observed, consistent with the formation of 5-membered P-containing metallacycles.64 Table 4.3 3 , P{ 'H} N M R data (CDC1 3, 300 K) for Pt pyridyldiphosphine complexes. In all cases, the peaks due to the dpype complexes appear 30—40 ppm downfield from those of their dpypcp counterparts. This phenomenon, as yet unexplained, has also been observed for a range of analogous N i complexes.31 In addition, the signals reported in Table 4.3 are generally downfield of those of the corresponding P-bonded cis- and ;rans-PtX 2(PN„) 2 complexes (X = CI, Br, I) (8 6-20). 6- 2 9 Attempts have been made to rationalise ligand basicities by comparing the 3 1 P{ 1 H} N M R chemical shifts of ligands within a family. 2 9 ' 3 1 At the heart of these lie a simple shielding/deshielding argument: the more basic the phosphine, the greater the electron density at the P-atom and the more shielded (upfield) it is, and vice-versa for a less basic phosphine. The general scheme has also been extended to assess the strength of metal-ligand bonding by comparing the chemical shifts of the resulting complexes. 2 9' 3 1 While it is true, for example, that 8P shifts for the PN„ family increase with increasing n (8 -3.28, -1.70 and -0.05, for n = 1-3, respectively), and that the scheme neatly accounts for this trend by assuming that successive substitution of Ph by py decreases the electron density at the P-atom,6 5 the diphosphines presented in this work do not fit comfortably into this paradigm. It is not obvious why dpypcp (8 -2.2) should be roughly as basic as PN2, and yet dpype (8 -6.1) should be significantly more basic than PNi , even though the P-atoms of both dpypcp and dpype are each bonded to 2 py rings and an alkyl fragment. In addition, the shielding/deshielding argument can only account for the sign of the coordination shifts (positive) experienced by these ligands, but not for their magnitudes Complex P{'H} chemical shift (ppm) [lJm (Hz)] PtCl2(dpypcp) PtBr2(dpypcp) Ptl2(dpypcp) PtCl2(dpype) Ptl2(dpype) [Pt(dpypcp)2][PF6]2 [Pt(dpype)2][PF6]2 17.9 [3490] 17.6 [3460] 12.6 [3290] 47.1 [3480] 49.8 [3280] 18.4, 18.5 [2400] 54.4 [2480] 171 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes nor for their difference in magnitudes. Even the so-called ring contribution (A R) to the coordination shift,66 aside from being arbitrary, seems of little use in that dpypcp and dpype have the same number of "backbone" C-atoms. Thus, such rationalisation warrants scepticism, and the approach throughout this thesis will be to describe chemical shifts phenomenologically. The ' j p p t values for the neutral complexes (Table 4.3) are in the range typically found for Pt complexes bearing cw-phosphine ligands,6 7 and are relatively independent of the nature of the pyridyldiphosphine ligand. These values increase with increasing field strength (CI > Br > I) of the trans halide ligand, a trend also observed for cis- and trans-PtX2(PN„)2 complexes.29 The smaller XJ??X values for the ionic compounds are in the range typically found for Pt complexes bearing trans phosphine ligands; these are ca. 70 % of the corresponding cis coupling constants.68 The complex [Pt(dpypcp)2][PF6]2 is distinct in bearing two chiral ligands. As the ligand is produced as a racemic mixture of R,R and S,S enantiomers, 4 equally probable ligand combinations on the metal are possible: R,R,R,R, S,S,S,S, R,R,S,S and S,S,R,R, where the first enantiomeric pair are diastereomers of the second identical pair. Thus, distinct peaks due to the two diastereomeric groups can be seen in the 3 1 P{'H} N M R spectrum of this complex. The separation between the two (ca. 0.1 ppm) is smaller than those observed for trans-R\\Cl2(dppcp)2 (0.5 ppm) 5 3 and #r-Pt 2Cl2(u-PN 2)2 (0.5 ppm).1 1 The molecular structures of PtCl2(dpype)-CH2Cl2 and Ptl2(dpypcp)-0.18 CH 2 C1 2 have been determined by X-ray crystallography and are shown in Figure 4.2 and 4.3, respectively. The latter compound exists as 2 crystallographically unique molecules in the unit cell; only one is shown. Relevant bond distances and angles appear in Table 4.4 and Table 4.5. As expected for both structures, the geometry about the metal centre is approximately square planar, and bond lengths and angles fall within the range of those typically observed for Pt" complexes of chelating phosphine ligands.6 9 The structure of PtCl2(dpype) is very similar to that of PtCi2(dppe),70 a minor difference being the relative magnitudes of the Pt—P bond lengths: these average 2.227(3) A for PtCl2(dppe) and 2.210(2) A for PtCi2(dpype). The shorter length in the latter may be due to the enhanced capacity for 7i-backbonding in this ligand which arises from the electron-withdrawing pyridyl substituents. Both dppe and dpype have a "bite" angle of ca. 87 °. 172 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes The chirality at the methine carbon atoms in the dpypcp ligand backbone (Figure 4.3) is S,S, and the chelate ring is in the 5-configuration; the same is true for the second crystallographically independent molecule. The R,R enantiomers must also be present in the unit cell, as racemic dpypcp was used in the preparation of Ptl2(dpypcp) which crystallized in the achiral space group P2\/c. The average Pt—P bond length (2.246(2) A) is marginally longer than that observed in PtCl2(dpype) (2.210(2) A), perhaps due to the stronger trans-influence of iodide versus chloride, although the comparison is drawn between two non-identical P-P ligands. Figure 4.2 ORTEP representation of PtCl 2(dpype)-CH 2Cl 2. Thermal ellipsoids for non-hydrogen atoms are drawn at 33 % probability. 173 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Table 4.4 Selected bond lengths (A) and angles (°) for PtCl 2(dpype)-CH 2Cl 2 with estimated standard deviations in parentheses. Pt(l)-- C l ( l ) 2.359(2) Pt(l)--Cl(2) 2.348(2) Pt(l)-- P ( l ) 2.209(2) Pt(l)--P(2) 2.211(2) Cl(l)--Pt ( l> -C l (2 ) 90.80(6) Cl( l ) --Pt( l> - P ( l ) 177.19(6) Cl(l)--Pt ( l> -P(2) 91.87(6) Cl(2)--Pt( l> - P ( l ) 91.18(6) Cl(2)--Pt ( l> -P(2) 177.25(6) P ( l ) - -Pt(l)--P(2) 86.17(6) Pt(l)-- P ( l ) - -C( l ) 108.6(2) Pt(l)-- P ( l ) - -C(3) 113.7(2) Pt(l)-- P ( l ) - -C(8) 115.4(2) C ( l ) - - P ( l ) - -C(3) 106.9(3) C ( l ) - - P ( l ) - -C(8) 105.3(3) C(3)- - P ( l ) - -C(8) 106.4(3) C(2)- - C ( l ) - -P(l) 108.5(4) C ( l ) - -C(2)--P(2) 109.5(4) C(2)- -P(2) --C(13) 105.1(3) C(2)- -P(2)--C(18) 105.7(3) Pt(l)--P(2) --C(2) 109.5(2) Pt(l)--P(2)--C(13) 115.5(2) Pt(l)--P(2) --C(18) 115.6(2) C(2)- -P(2)--C(13) 105.1(3) C(2)- -P(2) --C(18) 105.7(3) C(13> -P(2)--C(18) 104.5(3) P ( l ) - -C(8) --N(2) 114.1(5) P ( l ) - -C(3)--N(l) 112.3(5) P(2)- -C(18)--N(4) 114.0(5) P(2)- -C(13)--N(3) 114.3(5) C(23) Figure 4.3 ORTEP representation of one of the 2 crystallographically independent molecules of Ptl2(dpypcp) in the unit cell of Ptl2(dpypcp)-0.18 CH 2 C1 2 (50 % probability ellipsoids). Except for the two bonded to the methine C-atoms in the ligand "backbone", H-atoms have been omitted for clarity. 174 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Table 4.5 Selected bond lengths (A) and angles (°) for Ptl2(dpypcp)-0.18 CH 2 C1 2 with estimated standard deviations in parentheses. Pt(l> -1(1) 2.639(1) Pt(l)—1(2) 2.653(1) Pt(l> - P ( l ) 2.244(2) Pt( l )-P(2) 2.247(2) Pt(2> -1(3) 2.651(1) Pt(2)-I(4) 2.653(1) Pt(2> -P(3) 2.243(2) Pt(2)-P(4) 2.248(2) C( l ) --C(2) 1.480(1) C(26)—C(27) 1.547(1) 1(1)--Pt(l)--1(2) 91.43(2) I ( l ) - P t ( l ) - P ( l ) 176.18(5) 1(1)--Pt(l)--P(2) 92.08(5) I ( 2 ) - P t ( l ) - P ( l ) 88.98(5) 1(2)--Pt(l)--P(2) 175.07(5) P(l)—Pt(l)—P(2) 87.75(7) 1(3)--Pt(2)--1(4) 92.38(2) 1(3)—Pt(2)—P(3) 178.01(5) 1(3)--Pt(2)--P(4) 91.17(5) I(4)-Pt(2)-P(3) 89.38(5) 1(4)--Pt(2)--P(4) 176.15(5) P(3)—Pt(2)—P(4) 87.05(7) 4.4.2.2 "Tethered paddlewheel" complexes In an attempt to make halide-free Pt11 complexes of the pyridyldiphosphine ligands, Ptl2(dpype) and Ptl2(dpypcp) were treated with excess A g N 0 3 in glacial acetic acid/EtOH. Halide abstraction occurred, but the anticipated monomelic Pt(P-P)(N0 3) 2 complexes did not result. Instead, the crystalline, two-dimensional coordination polymers [Pt 2(dpype) 2Ag4(N0 3) 8(H 20)]„ (1) and [Pt 2(dpypcp) 2Ag 6(NO 3) 1 0] n (2) deposited from the reaction mixtures (Sections 4.7.19 and 4.7.20). Work on these systems has been published.63 1 and 2 contain Pt 2 face-to-face dimers bridged by A g N 0 3 clusters. The face-to-face dimers result from simultaneous P,P' chelation and bis(P,7V) bridging of the pyridyldiphosphine ligand, and because of their shape I have dubbed them "tethered paddlewheels" (as a subclass of the "paddlewheel" or "lantern" complexes which are prevalent in the literature. See Chart 4.5.) The difference between the two types is that in the tethered paddlewheel case the supporting ligand must be at least tetradentate so as to both bridge and chelate; in the paddlewheel case, bridging only is sufficient. The only other known, tethered paddlewheel complexes are Rh 2 Cl 2 (P 2 N 2 ) 2 (P 2 N 2 = (Ph)(py)P(X)P(py)(Ph); X = (CH 2 ) 2 , (CH 2 ) 3 ; py = 2-pyridyl).45 175 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Chart 4.5 "Paddlewheel" or "lantern" "Tethered paddlewheel" The short M - M contacts (2.7-3.1 A) displayed by the "paddlewheel" complexes Pt2(bridge)4n+, the fluorescence and phosphorescence properties of these compounds, 7 1 - 7 3 and their capacity to stabilize unusual oxidation states (e.g. Pt i n -Pt m 7 4 - 7 6 and Pt n-Pt n i 77>78), have drawn attention in the past two decades. The intermetallic distances in these compounds are generally shorter than those of the vast array of doubly-bridged dimers which have been synthesised by the reaction of cisplatin and its analogues with D N A bases and their derivatives.79 Various types of usually bidentate bridging ligands have been successfully employed in the synthesis of such tetrabridged dimers. The bridges have been of the N-N' type as in [Pt2(mtpo)4]-4H20 (Hmtpo = 4,7-/f-5-methyl-7-oxo[l,2,4]triazolo[l,5a]pyrimidine),80 of the 0-0 type as in Ba 2[Pt 2(C 404)4]-6H 20 ( C 4 0 4 = squarate),81 of the S-S type, as in P t 2 (S 2 CCH 3 ) 4 , 8 2 and of the P-P type, as in K4[Pt 2 (P 2 0 5 H 2 ) 4 ] . 8 3 Structure descriptions The overall structure of 1 is a two-dimensional coordination polymer, a section of which is shown in Figure 4.4; selected bond lengths and angles are given in Table 4.6. Each Pt2(dpype)2 4 + unit is connected to four others by way of a "net" two Ag2(N03) 4(H 20) 2~ bridges. The structure of such a bridge is shown in Figure 4.5 (p. 179). The two Ag-atoms are inequivalent: for the first, Ag(l), the coordination geometry could be described as a severely distorted trigonal bipyramid, with roughly apical pyridyl N-atoms (N(2) and N(4)*, one arising from each of the dpype ligands in the Pt2(dpype)24+ cluster), and three O-atoms belonging to two N0 3~ groups; the second, Ag(2), shows eight-fold coordination via the O-atoms of four chelated NO3" anions. Two short (2.35-2.45 A), three medium 176 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes length (2.54-2.64 A) and three long (2.85-3.00 A) Ag—O bonds are evidenced in this coordination environment. In addition, an H2O molecule is H-bonded to 0(1). In the Ag 2(N03)4(H 20) 2" bridge portion of the polymer, three distinct NO3" coordination modes are displayed: n2-chelating (0(7), 0(8)), p,-n2,n2-bis(chelating) and bridging (0(4), 0(5), 0(6)), and u-n ^ -chelat ing and bridging (0(1), 0(2), 0(3)). The Pt2(dpype)2 4 + cluster shown in Figure 4.6 (p. 180) consists of face-to-face Pt" square planar moieties arranged in a head-to-tail conformation, regarding the location of the P- and TV-donors. Each pyridylphosphine acts as a P,P '-chelating agent at one Pt-centre as well as a P./V-bridging ligand across the two Pt-centres. Thus, the P-atoms of each dpype ligand coordinate to one Pt-centre in a cis fashion, while two of the four o-py N-atoms (on each dpype) coordinate to the second Pt-atom forming a five-membered chelate ring containing a P-C-N bridge; the remaining two pyridyl N-atoms bond to Ag as discussed above. The structure of the repeating unit of 2 is shown in Figure 4.7; selected bond lengths and angles are given in Table 4.7. The Pt2(dpypcp)2 4 + cluster is structurally analogous to the corresponding unit in 1. The silver nitrate bridges in 2, however, differ from those in 1 by the inclusion of one further Ag(M>3) unit and by the absence of an H -bonded H 2 0 molecule. 177 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Ag(2*) Figure 4.4 ORTEP representation (33 % probability) of a section of 1 showing the Ag(N0 3 ) connections between Pt'^dpype^ moieties. The longer Ag—O bonds, H-atoms, and the pyridyl C-atoms not involved in bridges to metal centres have been omitted for clarity. 178 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Figure 4.5 ORTEP representation (33 % probability) of the "Ag(N0 3 ) bridge" of 1. "Short" Ag—O bonds are indicated by heavy lines, "medium" length bonds by double lines and "long" bonds by single lines. 179 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes one pyridyl N-atom of each dpype ligand. H-atoms have been omitted for clarity. 180 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Table 4.6 Selected bond lengths (A) and angles (°) for 1 with estimated standard deviations in parentheses. Pt(l)- -Pt( l) a 2.7690(7) Pt( l ) - -P(l) 2.252(2) Pt(l)- -P(2) 2.234(2) Pt( l ) - - N ( l ) a 2.148(8) Pt(l)- -N(3) a 2.131(7) Ag(l> - 0 ( 1 ) 2.638(11) Ag(l> - 0 ( 4 ) 2.552(10) Ag(l> - 0 ( 5 ) 2.892(9) Ag(l> -N(2 ) 2.282(8) Ag(l> —N(4)a 2.257(9) Ag(2> - 0 ( 2 ) 2.436(12) Ag(2> - 0 ( 3 ) 2.995(10) Ag(2> - 0 ( 4 ) b 2.642(11) Ag(2> - 0 ( 6 ) b 2.546(9) Ag(2> - 0 ( 7 ) 2.378(13) Ag(2> - 0 ( 8 ) 2.778(15) Ag(2> -0(10) 2.83(2) Ag(2> -0(11) 2.639(15) Pt(l) a-- P t ( l ) - - P ( l ) 85.09(6) Pt(l) a- - P t ( l ) - -P(2) 91.18(7) Pt(l) a-- P t ( l ) - - N ( l ) 93.3(2) Pt(l) a- - P t ( l ) - -N(3) 90.1(2) P ( l ) - - P t ( l ) - P(2) 84.47(9) P ( l ) - P t ( l ) - N( l ) a 173.3(2) P ( l ) - - P t ( l ) - N(3) a 94.8(2) P (2 ) - - P t ( l ) - N( l ) a 89.0(2) P(2) - - P t ( l ) - N(3) a 178.5(2) N ( l ) - - P t ( l ) - N(3) a 91.8(3) 0(1)- -Ag( l ) --0(4) 90.5(4) 0 (1) - -Ag( l ) --0(5) 95.8(3) 0(1)- -Ag( l ) --N(2) 79.3(3) 0 (1) - -Ag( l ) --N(4) a 114.3(3) 0(4)- -Ag( l ) --0(5) 45.3(3) 0 (4) - -Ag( l ) --N(2) 88.2(3) 0(4)- -Ag( l ) --N(4) a 116.1(3) 0 (5) - -Ag( l ) --N(2) 133.5(3) 0(5) - -Ag( l ) --N(4) a 73.0(3) N(2) - -Ag( l ) --N(4) a 150.8(3) 0(2) - -Ag(2)--0(3) 44.7(3) 0 (2 ) - -Ag(2)--0(4) b 110.2(4) 0(2) - -Ag(2)--0(6) b 87.5(3) 0 (2 ) - -Ag(2)--0(7) 118.8(4) 0(2) - -Ag(2)--0(8) 77.9(4) 0 (2 ) - -Ag(2)--0(10) 112.2(4) 0(2) - -Ag(2)--0(11) 95.9(4) 0 (3 ) - -Ag(2)--0(4) b 115.1(3) 0(3) - -Ag(2)--0(6) b 68.7(3) 0 (3 ) - -Ag(2)--0(7) 149.9(4) 0(3) - -Ag(2)--0(8) 122.0(3) 0 (3 ) - -Ag(2)--0(10) 70.6(3) 0 (3) - -Ag(2)--0(11) 78.1(3) 0(4) b--Ag(2)-- 0 ( 6 ) b 48.7(3) 0(4) b--Ag(2> - 0 ( 7 ) 93.6(4) 0(4) b--Ag(2)-- 0 ( 8 ) 87.7(4) 0(4) b--Ag(2> -0 (10) 114.3(3) 0(4) b--Ag(2> -0(11) 152.8(3) 0(6) b--Ag(2> - 0 ( 7 ) 141.3(4) 0(6) b--Ag(2> - 0 ( 8 ) 125.3(4) 0(6) b--Ag(2> -0 (10) 86.0(3) 0(6) b--Ag(2)--0(11) 128.3(3) 0 (7) - -Ag(2)--0(8) 46.7(4) 0 (7 ) - -Ag(2)--0(10) 106.7(4) 0 (7) - -Ag(2)--0(11) 79.8(4) 0 (8 ) - -Ag(2)--0(11) 105.7(4) 0 (8) - -Ag(2)--0(10) 148.3(4) 0(10)--Ag(2) - 0 ( 1 1 ) 45.1(3) * Superscripts refer to symmetry operations: (a) l-x, \-y, l-z (b) M2-x,y-\l2, 3/2-za. 181 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 0(95) Figure 4.7 ORTEP representation (33 % probability) of the repeating unit of 2. H -atoms have been omitted for clarity. Table 4.7 Selected bond lengths (A) and angles (°) for 2 with estimated standard deviations in parentheses. Pt(l)—Pt(l) a 2.7644(6) Pt( l ) --P(l) 2.241(2) Pt(l)-P(2) 2.255(2) Pt( l ) --N(2) a 2.144(7) Pt(l)—N(3) a 2.125(7) Ag(l> - N ( l ) 2.301(7) Ag(l)—N(4) a 2.232(8) Ag(l> -0(97) 2.518(10) Ag(2)—0(94) 2.50(2) Ag(2> -0(90) 2.537(10) Ag(2)-0(97) 2.579(10) Ag(3> -0(90) 2.513(11) Ag(3)—0(102) 2.423(9) Ag(3> -0(99) 2.362(11) Pt(l) a—Pt(l)—P(l) 89.54(6) Pt(l) a-- P t ( l ) - P ( 2 ) 86.78(6) Pt(l) a—Pt(l)—N(2) a 90.5(2) Pt(l) a-—Pt(l)—N(3)a 93.0(2) P ( l ) - P t ( l ) - P ( 2 ) 85.99(8) P ( l ) - -Pt(l)—N(2)a 177.7(2) P(l)—Pt(l)—N(3) a 88.2(2) P ( 2 ) - Pt(l)—N(2) a 91.8(2) P(2)—Pt(l)—N(3)a 174.2(2) N(2) a--Pt(l)—N(3) a 94.1(3) N(4) a —Ag(l)—N(l) 153.6(3). N(4) a-- A g ( l ) - 0 ( 9 7 ) 123.6(3) N ( l ) - A g ( l ) — 0 ( 9 7 ) 79.6(3) 0(94) —Ag(2)—0(90) 130.9(4) 0(94) —Ag(2)— 0(97) 126.6(5) 0(90) - A g ( 2 ) - 0 ( 9 7 ) 80.9(4) 0(99)—Ag(3)—0(102) 144.3(4) 0(99) —Ag(3)— 0(90) 122.7(4) 0(102)—Ag(3)—0(90) 91.0(3) Ag(l) - 0 ( 9 7 ) - A g ( 2 ) 137.7(4) Ag(2)—0(90)—Ag(3) 154.4(4) Refers to the symmetry operations: 2-x, -y, l-z. 182 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes A l l four of the P-atoms in 1 and 2, as well as all four of the methine C-atoms (C(l), C(1A), C(5), C(5A)) in the ligand backbones of 2 are chiral. The fact that 1 and 2 crystallize in a centrosymmetric space group, with the centre of inversion lying midway between the Pt-atoms, ensures that each enantiomeric centre is represented by equal populations of R and S absolute configurations. Pt-Pt bonding As the Pt-Pt separations in the P t 2 L 2 4 + cores of polymers 1 and 2 are comparable to the intermetallic distances found for some dimers which contain formal Pt-Pt single bonds, the metal centres of these clusters are definitely within range to interact. The Pt-Pt distances in 1 (2.769 A) and 2 (2.764 A) are essentially the same as those found in the relatively rare d 7-d 7 Pt'"2 dimers which are unsupported by bridging ligands, e.g. cis- and fra/w-[PtCl3(HN=C(Me)(OMe))2]2 (2.765 and 2.758 A, respectively),84 and in the d 8-d 8 Pt"2 T-over-square (ML4-ML3) dimers, e.g. [(Me)2Pt(p.-dppm)2Pt(Me)]+ (2.769 A).8 5 Face-to-face d 8-d 8 dimers of the type M 2 L 8 n + of formal a V 5 2 5 * V V 2 configuration display a net M - M bond order of 0; in addition, Cotton et al. have determined by X a M O calculations86 that, at least in the case of the model compounds M n 2 (HNCFfNH) 4 ( M = N i , Pd), no pair of ns or np orbitals on different metal centres overlaps in such a way as to provide an additional bonding orbital that can be filled at the expense of an antibonding d-based orbital. Therefore, it is often difficult to determine whether there is indeed an M - M interaction or whether the metal centres are merely held in proximity by the bridging ligands.8 7 Cotton's calculations and those of Mealli et A/. 8 8 do show, however, that a weak M - M bond arises through hybridization of the metal «d z 2 with the («+l)s and («+l)p z orbitals. In addition, calculations by Navarro et a/. 8 0 for the model compound Pt 2(HNCHN(C(=CH 2)Me)) 4 locate stabilizing charge density (of cylindrical symmetry) between the Pt-atoms equal to roughly half of that located between the Pt- and N-atoms. These things being so, a weak Pt-Pt bond probably does exist in the P t 2 L 2 4 + cores of polymers 1 and 2. Pauling's resonating valence bond theory of metals ( R V B M ) , 8 9 ' 9 0 which has been successful in correlating empirical data, relates the distance between two closely juxtaposed and interacting metal centres to the bond order between them by 183 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes D n = D 0 - f l l o g « [4.1] where D n is the intermetallic distance, Do is twice the covalent radius of the metal (2.590 A for Pt), a is an empirically-derived constant and n is the bond order. Investigations by Reis and Peterson into partially-oxidized Pt blues9 1 fix the value of a = 0.68 for Pt. For comparison with 1 and 2, the Pt-Pt distances and respective bond orders calculated by eq. 4.1 for some paddlewheel complexes are collected in Table 4.8; the data imply significant Pt-Pt interaction in 1 and 2, with bond orders of magnitude similar to those found for other tetrabridged structures. Table 4.8 Pt-Pt separations and calculated bond orders for some known paddlewheel complexes (tw = this work). Type of Bridge Compound Pt-Pt (A) Bond Order, n Ref. P-P K4[Pt 2 (P 2 0 5 H 2 ) 4 ]-2H 2 0 2.925 0.29 83 N - N Pt2(mtpo)4-4H20 2.744 0.57 80 0 - 0 Ba 2 [Pt 2 (C 4 0 4 ) 4 ]-6H 2 0 3.061 0.18 81 s-s Pt 2 (S 2 CCH 3 ) 4 2.767 0.52 82 P - N [Pt 2(dpype) 2A g 4(N0 3)8(H 20) 2]„, 1 2.769 0.52 tw P - N [Pt 2(dpypcp) 2Ag 6(NO 3) 1 0]„, 2 2.764 0.53 tw Aqueous solution chemistry Although aqueous solutions of 1 slowly precipitate Ag metal (over ca. 48 h) even when stored in the dark, the UV-vis spectra of the solutions (from 350-800 nm) are invariant over time. In addition, the *H N M R spectrum of 1 in D 2 0 shows no splitting in the pyridyl region (specifically in the doublet at 8 8.58 which is attributed to Fi6 5 7 ' 5 8 ) and thus the absence of 3-bond 1 H - 1 0 7 / 1 0 9 A g coupling. These facts, taken in conjunction with the known low formation constants of Ag-pyridine species,92 show that upon dissolution of 1 in H 2 0 the Ag 1 ions dissociate from the P t 2 L 2 4 + core and that it is this core which is responsible for the observed solution spectra. The 3 1 P{ 1 H} N M R spectrum of 1 in D 2 0 shows a basic singlet for equivalent P-atoms with Pt-satellites via one- and three-bond coupling (cf. Figure 4.6). The Pt2(dpype)2 4 + core can be synthesised independently, i.e., in the absence of A g + . This is accomplished by reacting PtI2(cod) with A g N 0 3 in EtOH/H 2 0 (which results presumably either in [Pt(S)2(cod)][N03]2 (S = H 2 0 or EtOH) or Pt(N03)2(cod)) and 184 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes removing the solid Ag l prior to addition of 1 equiv. of the dpype ligand (Section 4.7.21). In this manner, the complex [Pt2(dpype)2][N03] can be isolated as a dihydrate. This complex has solution U V and N M R spectroscopic properties identical to those of the analogous coordination polymer. 4.4.2.3 Palladium(II) pyridyldiphosphine complexes Whereas the Pt" dpypcp complexes are straightforward to make, the Pd 1 1 analogues pose a slight synthetic challenge, because Pd is generally more reactive than Pt. Thus, i f no precautions are taken to control the relative concentrations of the reactants, addition of solvent to a 1:1 mixture of £ra«s-PdCi2(PhCN)2 and dpypcp generates three products. The first is a beige precipitate, easily isolated from the other two components by filtration prior to the work-up; this powder is completely insoluble in all common deuterated solvents and could not be analysed by solution N M R spectroscopy. Elemental analysis showed, however, that it has the empirical formula Pd2Ci4(dpypcp). The formation of this precipitate could never be completely eliminated. The remaining two components are PdCl2(dpypcp) and [Pd(dpypcp)2]Cl2. Although the analogous synthesis of PtCi2(dpypcp) is sometimes complicated by the production of [Pt(dpypcp)2]Ci2, its abundance is not as great as in the case of Pd. This is attributed to two factors: PdCi2(dpypcp) is more reactive toward a second dpypcp molecule than is PtCl2(dpypcp), and Pd has a higher affinity for the pyridyl N-atoms than does Pt, which presumably causes formation of the 2:1 adduct Pd2Ci4(dpypcp). Formation of this species lowers the concentration of trans-PdCl2(PhCN)2 in solution and raises the ligand:metal ratio, so promoting the formation of [Pd(dpypcp)2]Cl2. The formation of the 2:1 adduct was retarded by appropriate choice of Pd precursor. Thus, use of PdCi2(cod) gives better results as cod is less easily displaced than PhCN (Sections 4.7.15, 4.7.16 and 4.7.17). Once again, the presence of downfield singlets in the 3 1 P{'H} N M R spectra of all the Pd" pyridyldiphosphine compounds indicates molecular Ci symmetry and exclusive P,P' bonding of the ligand in 5-membered metallacycles (see Section 4.4.2.1). Contrary to the case of Pd" complexes of the PN„ ligand family, 2 9 there is no cis/trans ambiguity here: the complexes are exclusively cis as enforced by the cyclopentane ring linking the two P-atoms of the ligand. 185 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.4.3 Pt° and Pd° pyridyldiphosphine complexes The initial work on the synthesis and characterisation of M° (M = Pd, Pt) complexes of the pyridyldiphosphine ligands dpype and dpypcp was performed by Smith. 5 5 Although several complexes were reported as being prepared in situ, they were not isolated, and characterisations were limited to 3 1 P{ ] H} N M R data. Smith ascertained that these compounds could be accessed by two basic routes: by reduction of the appropriate [M(P-P) 2]C1 2 compound with either N2H4 in DMSO or NaBH 4 in MeOH, or by reaction of a suitable M° precursor {e.g., M(PPh 3) 4) with either 1 or 2 equiv. of the ligand. The latter ligand substitution route was chosen in this work because the reduction route was not as high yielding. The pyridyldiphosphine ligands react with Pt(PPh3)4 in C6H6 solution at r.t. to form the presumably tetrahedral Pt(PPh3)2(P-P) and Pt(P-P) 2 depending on the metakligand ratio (Sections 4.7.22-4.7.25). Unlike the Pt" case, 1 equiv. of ligand reacts to form Pt(PPh3)(P-P) exclusively; no sign of the 1:2 adduct is observed, even i f no precautions are taken to control the relative concentrations of the reactants. This result contrasts with that of the analogous reaction of Pt(PPh3)4 with 1 equiv. of dppe which produces predominantly the 1:2 adduct Pt(dppe)2.55 There is no obvious reason for this difference although it may involve the lower basicity of the pyridyldiphosphines with respect to their phenyl analogues (see Section 4.4.2.1). Xie made both Pt(PN 3) 4 and Pt(PNi) 3 by methods analogous to those used in the preparation of Pt(PPh 3) 4 and has shown that in contrast to the PPh 3 compound, which undergoes significant dissociation in CeFi6 at r.t. to give Pt(PPh 3) 3 , 9 3 the PN„ complexes show no tendency to lose phosphine.18 The same observation holds for Pt(PPh3)2(P-P); these compounds give rise to sharp N M R spectra that contain no peaks due to free PPh 3 or P-P. The ^P j 'H} N M R spectrum of Pt(PPh3)2(dpypcp) (Figure 4.8) is the expected result for a non-fluxional, 4-coordinate complex in which two sets of two chemically equivalent P-atoms couple with each other and with 1 9 5 Pt to give 18 lines. By virtue of its similar chemical shift in the corresponding spectrum of Pt(PPh3)2(dpype) (8 21.3), the triplet at 8 23.1 is assigned to the P-atoms associated with the PPh 3 ligands. This assignment results in 8 14.6 and 32.0 for the P-atoms associated with the 186 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes pyridyldiphosphine ligands of Pt(PPh3)2(dpypcp) and Pt(PPh3)2(dpype), respectively, and is consistent with the fact that in all previously made pyridyldiphosphine complexes of N i " , 3 1 Pd" (Section 4.4.2.3) and Pt" (Section 4.4.2.1), the peaks due to dpype fall significantly downfield of those due to dpypcp. Accordingly, the 'jppt values for the PPh 3 ligands are significantly larger than those of the pyridyldiphosphine ligands (by 660 and 850 Hz for Pt(PPh3)2(dpypcp) and Pt(PPh3)2(dpype), respectively). This result is consistent with studies by Chatt et al.9A and Al-Ohaly and Nixon 9 5 on the series of compounds Pt(P)(P-P-P) (P = a monodentate tertiary phosphine; P-P-P = triphos). These researchers found that although the magnitude of 1 J P P t increases with increasing electronegativity of the groups directly bonded to the ligand P (e.g., ]JPPt = 5400 and 10200 Hz for P = PPh 3 and PF 3 , respectively), small P—Pt—P angles imposed by chelating ligands result in smaller than predicted 1 J P P t values. Thus, the anticipated result that the pyridyldiphosphine ligand should show larger lJPPt values than the PPh 3 ligand because of the greater electronegativity of py vs. Ph is not observed in Pt(PPh3)(P-P). In addition, the JPP values recorded for this series of complexes range from 51-95 Hz, which makes Pt(PPh3)(dpypcp) (2JPP = 52 Hz) and Pt(PPh3)2(dpype) ( V P P = 53 Hz) typical. I 35 25 —r~ 15. -Tr-io Figure 4.8 The 3 1 P{ ] H} N M R (162 MHz, C 6 D 6 , 300 K) spectrum of Pt(PPh3)2(dpypcp). 187 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Suspension of Pt(PPh3)4 in boiling EtOH gives the isolable, yellow, trigonal planar Pt(PPh 3) 3 . 9 3 Reaction of this with 1 equiv. of dpypcp gave Pt(PPh3)2(dpypcp) exclusively and not the intended 3-coordinate complex Pt(PPh3)(dpypcp). A possible explanation is that the bite angle of dpypcp constrains the P-atoms so as to prohibit the P—Pt—P angle of 120 ° required to stabilise trigonal planar M° complexes. In another attempt to make a 3-coordinate complex, Pt(PPh3)2(dpypcp) was subjected to an excess of the bulky and basic PCy 3 . It was reasoned that the steric demand of this ligand would compensate for the small bite angle of the chelating pyridyldiphosphine and stabilise the 3-coordinate substitution product. However, there was no reaction. The Pt(PPh3)2(P-P) compounds are orange, air-stable, microcrystalline solids which in solution are 02-sensitive; exposure to air of C6H6 solutions containing the complexes results in the slow formation of P-P(0)2 and OPPh 3 , presumably due the mechanism outlined by Halpern and Sen. 5 6 The 1:2 adducts Pt(P-P)2 are air-sensitive, red solids. The analogous Pd° complexes, Pd(PPh3)2(P-P), were not isolated because of the lability and reactivity of the Pd(PPh3)4 precursor, but Pd(PPh3)2(dpype) was prepared in situ. Reaction of 1 equiv. of P-P with Pd(PPh3)4 in C6H 6 at r.t. resulted in partial formation of Pd(P-P)2. To work around this problem, the same reaction was conducted on an ice-bath in the presence of a 5-fold excess of PPh 3; these conditions limited the dissociation of PPh 3 from Pd, and reaction with dpype gave Pd(PPh3)2(dpype) as the major product: 5 P 26.7 (t, V P P = 24.7 Hz), 29.1 (t, 2J?? = 24.7 Hz). The only reported Pd°-pyridylphosphines are Pd(PNi) 3, Pd(PNi)2(dba) and Pd(PNi) 2 (n 2 -DMAD). 9 6 4.4.3.1 Reaction of Pt(PPh3)2(P-P) with maleic and fumaric acid diesters Reactions between prochiral olefins and C2-symmetric chiral metal-diphosphine complexes continue to attract attention because of their relevance to asymmetric catalysis.97 The two routes to Pt°-olefin complexes are reduction of Pt" precursors in the presence of the olefin, and displacement of a labile ligand from Pt° by the olefin. Thus, for example, stilbene complexes can be made by reduction of PtCl2(PPh 3) 2 with N 2 H 4 followed by addition of the olefin; 9 8 and treatment of Pt(PPh 3) 4 with chloro-, 9 9 fluoro- 1 0 0 188 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes and cyanoolefins101 in refluxing CeH 6 gives the corresponding Pt(PPh3)2(r|2-olefin) complexes. The latter route was chosen as the synthetic method for Pt°-olefin complexes of the pyridyldiphosphines. The Pt(PPh3)2(P-P) complexes react with dimethyl- and diethylmaleate (dmm and dem, respectively) to form Pt(n2-olefin)(P-P) complexes. The fumarate compounds Pt(n2-dmf)(dpypcp) and Pt(n2-def)(dpypcp) were isolated from the corresponding reactions with dmm and dem (see Sections 4 .7 .26 and 4 .7.27, respectively, for syntheses, 1 31 and below for discussion); these compounds were characterised by H and P N M R spectroscopy and elemental analysis, and, in the case of the former complex, by X-ray crystallography. The maleate analogues were characterised in situ in CeD6 solution by 3 1 P{'H} N M R spectroscopy. A summary of the 3 1 P{ 1 H} N M R data is given in Table 4.9. Table 4.9 3 1 P{'H} N M R data for Pt(P-P)(n2-olefm) complexes. P-P Olefin •"Pl/H} NMR chemical shift (ppm) ['/pPt (Hz)] dpypcp dmf0'* Major: 33.1 [3360] def"'6 Minor: 31.8 [3400] Major: 33.6 [3360] dem c r f Minor: 32.1 [3410] 35.4 [3260] , 33.3 [3290] dpype defc 60.3 [3290] dem c 59.8 [3300] " Measured in CDC1 3 . b The fumarate diesters give two configurational diastereomers when bonded to rac-"Pt(dpypcp)". c Measured in C6D 6. d Pt(dpypcp)(r)2-dem) has chemically inequivalent P-atoms as shown in Chart 4.6. 2JPP= 13 Hz. The isolated product from the reaction of excess dmm with Pt(PPh3)2(dpypcp) 7 7 was not the expected Pt(n -dmm)(dpypcp) but rather Pt(n -dmf)(dpypcp) (Section 4.7.26); and indeed, this reaction produced crystals whose molecular structure is depicted in Figure 4 .12 (p. 196). Selected bond distances and angles are given in Table 4.10. Although this result indicated that the metal mediated a cis to trans olefin isomerisation, doubt arose about this when analysis by GC of the dmm used showed that it contained ca. 2 % dmf, a more than sufficient amount to react with all of the Pt, and the isolated yield was only 47 %. The same observations were made for the dem reaction, although the 189 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes abundance of def was approximately 10 % as determined by N M R spectroscopy. Monitoring the 1:1 reaction between Pt(PPh3)2(dpypcp) and dem in CeD 6 by 3 1 P{'H} N M R over 2 d indicated that both Pt(n2-dem)(dpypcp) and Pt(n2-def)(dpypcp) were formed initially (in the expected 9:1 ratio) and that their relative concentrations did not change over time. In addition, reaction between excess dem and Pt(PPh3)2(dpypcp) did not result in the catalytic conversion of dem to def. This result contrasts with that observed by Xie and James who determined that the initial product, Pt(PNi)2(n2-dem), of the reaction between dem and Pt(PNi) 3 converted over hours into Pt(PNi)2(r|2-def) in CDC1 3 solution at 293 K . 1 8 If the cis to trans isomerisation proceeds through an ylide transition state as shown in Scheme 4.2, 1 0 2 then the process should occur more readily in polar solvents that would tend to stabilise the charge separation. This has been proposed earlier in the substitution reaction between Pt(PPh3)2(n -stilbene) and an excess of a 1:1 mixture of cis/trans-\ ,2-d\c\\\oxo-\ ,2-difluoroethylene which gave Pt(PPh3)2(r|2-ClFC=CFCl) in a transxis ratio of 2.3:1 when the reaction was conducted in E t 2 0 and in a 2.9:1 ratio in C H 3 C N . 1 0 2 In order to determine whether the difference in the olefin isomerisation behaviour of Pt(PNi)3 in CDCI3 and Pt(PPh3)2(dpypcp) in C6D6 was due to a solvent effect, excess dem was reacted with Pt(PPft3)2(dpypcp) in CDCI3 solution and the reaction monitored at 300 K by 3 I P{'H} N M R spectroscopy. Scheme 4.2 A portion of the 300 K 3 1 P{'H} N M R spectrum of Pt(PPh3)2(dpypcp) dissolved in CDC1 3 (in the absence of dem) is shown in Figure 4.9. This spectrum is dramatically different from that observed for the same complex in CeD6 (Figure 4.8). It consists of (1) a 2nd-order pattern centred at 8 23.6 and a pseudo-triplet at 8 17.2 that together constitute 2' 190 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes an A B X pattern; (2) a broad peak at 8 -5 due to PPh 3 which is involved in a fluxional process; and (3) singlets at 5 18.5 and 29.6 due to PtCl2(dpypcp) and OPPh 3, respectively. # # # # # I I I 1 I I I I I—I I I I 28 26 24 i i i i I i i i I i i i i I i i—n—I i i i i I i i i i I—i i i i I i i i i 22 20 18 16 ppm Figure 4.9 The 5 15-29 range of the 3X?{lH} N M R (121 MHz, CDC1 3 , 300 K) spectrum of Pt(PPh3)2(dpypcp). The peak due to PtCl2(dpypcp) is marked by an asterisk (*), and those of [Pt(PPh3)(dpypcp)]+ by the number symbol (#). Pt satellites for all peaks fall outside of this window. The A B X pattern is consistent with a Pt complex in which 3 P-atoms are coordinated at the coiners of a "T", indicating a change in geometry from tetrahedral to square planar on dissolution of Pt(PPh3)2(dpypcp) in CDC1 3. One P-atom holds a cis disposition to the other two while the remaining pair are each cis to one P-atom and trans to each other; this is reflected in the 2JPP values: two are small (ca. 18 Hz), consistent with cis P-atoms giving rise to the pseudo-triplet due to their very similar magnitudes, while the other is much larger (384 Hz) and consistent with trans P-atoms. In accordance with this, two 1 Jppt values are consistent with mutually trans P-atoms (2370 and 2440 Hz), while the other lies in the range observed for cis P-atoms (3640 Hz). 6 7> 6 8 The Jppt and JPP values for the complex giving rise to the A B X pattern are very similar to those observed by Anderson and Lumetta for the broad family of cations [Pt(P-P)(P)C1]+ (P-P = dppe, dmpe, dppm; P = PEt 3, PMePh 2 , PBu 3 , PPh 3): 2995-3551 Hz for 'jppt(cis); 1867-2414 Hz for VpPt(trans); 4-17 Hz for 2/ P P(cis) (P-P = dppe, dmpe), 8-66 Hz (P-P = dppm); and 368-414 Hz for Vpp(trans).103 The findings strongly suggest 191 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes that dissolution of Pt(PPh3)2(dpypcp) in CDC1 3 gives rise to the square planar Pt11 species [PtCl(PPh3)(dpypcp)]+ resulting from initial oxidative addition of the solvent (Scheme 4.3). The identity of the species giving the A B X pattern was confirmed by the in situ reaction between PtCl2(dpypcp) and excess PPh 3 which reproduced the 3 1 P{ 1 H} N M R pattern exactly; the complex responsible is undoubtedly [PtCl(PPh3)(dpypcp)]Cl. Scheme 4.3 The oxidative addition step (eq. 4.2) may be preceded by dissociation of one PPh 3 ligand to generate the (presumably reactive) 3-coordinate species Pt(PPh3)(dpypcp). Steps such as 4.3 (the oxidative addition of C-Cl of one Pt chloromethyl complex to another, followed by reductive elimination) and 4.4 ([3-chloride elimination) have been proposed by McCrindle et al. for decomposition of a range of cis-chloro(chloromethyl)palladium complexes but no direct evidence was provided. 1 0 4 This group has studied the solution stabilities of several chloromethyl P t 1 0 5 and P d 1 0 4 complexes, and determined that complexes of the type m-PtCl(CH 2 Cl)(P-P) (P = dppe, dppp), when pure, are indefinitely stable with respect to decomposition to cw-PtCl 2(P-P) in CDC1 3 solution under vacuum and, in the case of PtCl(CH2Cl)(dppe), even in the presence of water. In this thesis work, dissolution of Pt(PPh3)2(dpypcp) in CDC1 3 rapidly gives PtCl2(dpypcp), but as noted by McCrindle et al. "small amounts of impurities may have a marked effect on the apparent stability of a transition metal compound."1 0 5 They found that the chloromethyl complexes underwent rapid decomposition in the reaction mixture used to create them, but showed enhanced stability in the same solvent after purification. In the case under study here, the system involves a dichloromethyl species, and also contains PPh 3 . In addition, the involvement of radical processes cannot be excluded. Pt(PPh3)2(dpypcp) + CDC1 3 2 PtCl(CDCl2)(dpypcp) — PtCl(CDClCDCl 2)(dpypcp) PtCl2(dpypcp) + PPh 3 — - PtCl(CDCl2)(dpypcp) + 2 PPh 3 [4.2] - PtCl2(dpypcp) + PtCl(CDClCDCl 2)(dpypcp) [4.3] — PtCl2(dpypcp) + CDC1=CDC1 [4.4] [PtCl(PPh3)(dpypcp)]Cl [4.5] 192 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes In an attempt to observe PtCl(CHCl2)(dpypcp), CHCI3 was added to a C6D 6 solution of Pt(PPh3)(dpypcp), and the system monitored by 3 1 P{ 1 H} N M R spectroscopy. After 2 h, the only species observed were the unreacted Pt° complex, and trace, equal quantities of PtCl2(dpypcp) and free PPh 3. Once again, this result points to the instability of any chloro(dichloromethyl) intermediate and also suggests perhaps that PPh 3 dissociation is necessary prior to oxidative addition of CHC1 3 . This seems reasonable in light of the fact that even in the presence of a substantial excess of CHC1 3, Pt(PPh3)2(dpypcp) reacts very slowly in CeD6 whereas the corresponding reaction in neat CDC1 3 is complete within the mixing time. The decomposition of Pt° phosphine complexes to the corresponding Pt" dichlorides is frequently observed; e.g., dissolution of Pt(PPh3)4 in CCI4 produces cis-PtCl 2 (PPh 3 ) 2 . 1 0 6 The CDC1 3 3 l P{ 'H} N M R spectrum of isolated Pt(ri2-dmf)(dpypcp) consists of two singlets (with corresponding Pt "satellites") at 5 33.1 and 31.8 in a ratio of ca. 5:1 (Figure 4.10). On the basis of a comparative study of the in situ reactions of def and dem with Pt(PPh3)2(dpypcp), these singlets have been assigned to the two configurational diastereomers107 which can result from the combination of the prochiral olefin dmf with R,R and S,S "Pt(dpypcp)". In this study, the reactions of slight excesses of def and dmf with Pt(PPh3)2(dpypcp) in C 6 D 6 were followed by 3 1 P{ ! H} N M R spectroscopy. Within the time taken to make the first measurement (4 min), reaction with dem gave three products, giving rise to two singlets at S 33.2 and 34.4 ('jppt= 3360 and 3330 Hz) and an AJ3 pattern (§ 33.3 and 35.4, lJpPt = 3290 and 3260, Vp P = 13 Hz). This spectrum can be rationalised in the following manner: the A B pattern is due to Pt(n2-dem)(dpypcp) while the singlets are due to the configurational diastereomers of Pt(n -def)(dpypcp) which arise due to the def contaminant. No diastereomers can result from the former compound because dem is not prochiral (i.e., its faces are homotopic). That the P-atoms in Pt(n2-dem)(dpypcp) should be chemically inequivalent, while those in Pt(n2-def)(dpypcp) should not, can be accounted for i f it is assumed that there is restricted rotation about the axis bisecting the Pt atom and the mid-point of the olefinic bond; Chart 4.6 shows the maleate and fumarate complexes as viewed down this axis (Pt-atoms, P—Pt bonds, P-substituents and 3 of the-5 cyclopentyl C-atoms of each dpypcp ligand have been omitted 193 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes for clarity; R = C0 2 Et) . Reaction with def gives only the singlets at 5 33.2 and 34.4 which confirms these assignments. • ' I • < • • 1 • 1 —• 1 1 • • 1 1 1 1 1 r PPm 10 35 30 25 Figure 4.10 3 1 P{'H} N M R spectrum (CDC1 3, 202 MHz, 300 K) of Pt(n2-dmf)(dpypcp). Chart 4.6 maleate fumarate Over the course of 45 min, the peak at 8 33.2 gradually disappeared while the one at 8 34.4 intensified. This represents the conversion of the kinetic to thermodynamic distribution of configurational diastereomers of Pt(n2-def)(dpypcp). A positive assignment of the two peaks could not be made in the def case, but i f it is assumed that the diastereomeric form observed in the crystal of Pt(n2-dmf)(dpypcp) represents the thermodynamically more stable isomer, then assignment of the peaks in the CDCI3 3 1 1 2 P{ H} N M R spectrum of Pt(n -dmf)(dpypcp) is possible because of the structure 194 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes determination. The larger (§ 33.1) is due to the S,S,R,R/R,R,S,S pair (only the former of these is illustrated in Figure 4.12, but the latter must be present in equal abundance as demanded by the centrosymmetric P2]/a space group); the smaller (5 31.8) is due to the R,R,R,R/S,S,S,S pair. The quadrant diagram shown in Figure 4.11 clearly illustrates that the former diastereomer is more stable than the latter based on steric interactions between the bound olefin and the metal-diphosphine fragment. The A B pattern due to Pt(n2-dem)(dpypcp) maintained its intensity throughout the course of the study. ^|S|b|W|ijEi|H • • • Figure 4.11 Quadrant diagram showing steric interactions between a chiral C 2 -symmetric metal-diphosphine fragment and a bound olefin, looking down the M—mp axis (mp = C=C mid-point). The grey rectangles represent steric bulk presented by the diphosphine ligand, and the black circles represent the substituents on the olefin. The left-hand diagram illustrates the sterically more favourable diastereomer. The reaction between excess dem (containing ca. 10 mol % def) and Pt(PPh3)2(dpype) in CeD 6 was also followed by 3 1 P{'H} N M R spectroscopy. In this case, diastereomers could not result as dpype is achiral. The reaction was complete within 0.5 h and gave two products which were manifested by singlets at 5p 60.3 and 59.8 in a 1:2 ratio. In an independent reaction between def and Pt(PPh3)2(dpype), the peak at 8 60.3 was identified as being due to Pt(PPh3)2(n2-def). Because the relative abundances did not correspond to the relative abundances of dem and def (9:1) in the reacting olefin mixture, def must react faster than dem with Pt(PPh3)2(dpype). In addition, as the relative concentrations of the products did not change over time, there is no metal-mediated cis to trans isomerisation of the olefin. 195 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Figure 4.12 ORTEP representation of Pt(n2-dmf)(dpypcp)-2 CDC1 3 . Except for the 2 olefinic ones, H-atoms have been omitted for clarity (50 % ellipsoids). Table 4.10 Selected bond lengths (A) and angles (°) for Pt(n2-dmf)(dpypcp)-2 CDC1 3 with estimated standard deviations in parentheses. Pt ( l ) --P(l) 2.2602(9) Pt(l)--P(2) 2.2543(9) Pt( l ) --C(27) 2.116(3) Pt(l)--C(28) 2.132(3) C(27)--C(28) 1.471(5) D ( l ) - -N( l ) 2.49 D ( l ) - -0(4) 2.07 H ( l ) - -0(3) 2.42 H(2)- -o(i) 2.49 P ( l ) - -Pt(l)-P(2) 86.91(3) C(27)-—Pt(l)—C(28) 40.52(12) The molecular structure of Pt(n2-dmf)(dpypcp) (Figure 4.12, Table 4.10) embodies all the features typical of known Pt°-olefm complexes.1 0 8 These are a lengthening of the C—C double bond upon coordination, a bending back of the olefinic substituents out of the plane of the double bond and away from the Pt-centre, and a 196 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes twisting by a few degrees of the olefin out of the plane containing the Pt-atom, the iigating atoms of the 2 ancillary ligands and the mid point of the double bond (Chart 4.6). In addition, and in contrast to Pt"-olefin complexes, the barrier to rotation of the olefin about the Pt—mp is high in this class of compounds. For a d 1 0 metal centre there is no difference between the ligand field stabilisation energies of tetrahedral and square planar arrangements, and so the ligands separate as much as possible to minimise steric interactions, i.e., a tetrahedron results.1 0 9 However, i f a strongly 7i-accepting ligand such as an olefin bearing electron-withdrawing groups is coordinated and the o-donating ability of this ligand is comparatively small, then the metal centre will be left with slightly less than 10 electrons and will distort according to the Jahn-Teller theorem to give a flattened tetrahedron which approximates a square plane. The most accommodating qualitative theoretical treatment of Pt°-olefin bonding, both for its merits in rationalising the gross coordination geometry at the metal centre, and for its accounts of the other experimentally observed phenomena outlined above, is the Dewar-Chatt-Duncanson model. , , 0> 1 1 1 In the specific case of olefinic bonding to "Pt°(phosphine)2", this model embodies a-donation from the olefin 7t-bond into a dp2 hybrid on the metal (formed from d x y , p x, py) and 7t-back donation from the metal dx2-y2 to the 7i* orbital on the olefin, i f the complex is oriented along the axes shown in Chart 4.7. Chart 4.7 2 Several Pt(n -olefm)(PPh3)2 complexes have been structurally characterised, and by far the majority bear one or more electron-withdrawing substituents such as F , 1 1 2 C I , 1 1 2 ' 1 1 3 C N , 1 1 3 - 1 1 6 / ? - N 0 2 C 6 H 4 1 1 7 on the olefin; however, only one report of a structurally characterised Pt° phosphine complex bearing a CC^R-substituted olefin has appeared.118 The known C—Pt—C angles in these compounds range from 38.8 ° (olefin = ;ra^-CH(p-NO 2C 6H 4)=CH0t7-NO 2C6H4) 1 1 8) to 47.1 ° (olefin = CC1 2=CC1 2 1 1 2). The corresponding angle in Pt(n2-dmf)(dpypcp) (40.5 °) is thus typical for this class of 197 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes compounds and is similar to that found for Pt(n2-CH(C02Me)=C(C02Me)2)(PPh3)2 (39 .5° ) . 1 1 8 The dihedral angles between the C—Pt—C and P—Pt—P planes of complexes in this family lie in the range 1.0-22.1 °, and in this respect, Pt(n2-dmfj(dpypcp), with its corresponding angle of 9.6 °, is also unremarkable. Not surprisingly, the P—Pt—P angles of the PPI13 complexes are all significantly larger (> 100 °) than the corresponding angle in Pt(n2-dmf)(dpypcp) (86.9 °). A fairer comparison can be made to Pt(n2-fra«s-stilbene)(chiraphos) in which this angle is restricted to 87.1 and 86.6 0 (found for two crystallographically independent forms).97 The crystal structure of dmf does not appear in the Cambridge Structure Database, but the average C=C bond distance for two independent measurements of fumaric acid is 1.324 A.1 '9,120 The corresponding distance for the coordinated dmf in Pt(n2-dmf)(dpypcp) is significantly longer (1.471 A), consistent with the Dewar-Chatt-Duncanson model. Hydrogen bonding in Pt(n2-dmf)(dpypcp)-2CDCls The most reliable indicator of a hydrogen bond X - H — Y is that the H—Y distance is less than the sum of the van der Waals radii of H and Y . 1 2 1 When the proton cannot be located and the X - H — Y linkage is linear or near-linear, an X-to-Y distance less than the sum of the van der Waals radii of X and Y is a sufficient indicator of hydrogen bonding; indeed, when this difference is > 0.3 A, the hydrogen bonding is termed strong and the interaction is better depicted as X - H - Y . 1 2 2 The structure of Pt(n2-dmf)(dpypcp)-2 CDCI3 shows 2 intramolecular C - H - 0 contacts, as well as intermolecular C-D—N and C-D—O hydrogen bonds, the latter possibly being the strongest interaction of its type ever to be observed (vide infra). The intramolecular contacts exist between each olefinic proton and the OMe O-atom of the ester group on the opposite C-atom of the olefin. The C-D—N bond lies between one of the pyridyl N-atoms and the D-atom of one of the CDCI3 solvates, while the other CDCI3 solvate is bonded to the carbonyl O-atom of one of the ester groups. The H-bonding interactions shown in Figure 4.12 are summarised in Table 4.11. 198 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Table 4.11 Relevant parameters for the intra- and intermolecular H-bonding interactions in Pt(n2-dmf)(dpypcp)-2 CDCI3. X - H / D - Y 8 A . . . Y H / D - Y X - H / D - Y d (A)c'd d' (A)d'e (A)b (A) (°) C(27)-H(l)-0(3) 2.783(5) 2.42 101.5 0.52 0.53 C(28)-H(2)-0(l) 2.845(5) 2.49 100.7 0.46 0.46 C(32)-D(l)-N(l) 3.374(5) 2.49 149.2 -0.12 0.46 C(33)-D(l)-0(4) 3.033(5) 2.07 165.8 0.27 0.88 a X - H distances are fixed at 0.98 A. b This is the straight-line X-to-Y distance. c d = r(X) + r(Y) - A . . . Y , where r(X) and r(Y) are the van der Waals radii of X and Y , respectively. d The van der Waals radii are 1.75, 1.55, 1.50 and 1.20 A for C, O, N and H, respectively.121 e d ' =r(H) + r ( Y ) - H / D - Y . To this author's knowledge, the only C-H—O H-bond shorter than the 2.07 A observed here is that of 2.045 A determined by neutron diffraction for the intermolecular interaction between 1-methylthymine molecules in the crystalline material. 1 2 3 However, as the relevant D-atom was not located in the structure of Pt(n2-dmf)(dpypcp)-2CDCl3, the D—O distance is not known accurately and therefore, a better comparison of the two H-bonds may be their d-values; these are 0.18 A for 1-methylthimine and 0.27 A for the Pt complex. Thus, the C-H/D—O interaction observed here may be stronger than that observed before. 4.5 Conclusions A new "cyclopentyl-bridged" ligand, dpypcp, was successfully synthesised and characterised. In addition, the diprotonated, [dpypcp(H)2]2+, and oxidised, dpypcp(0)2, forms were isolated, as were those of the known "ethyl-bridged" ligand, dpype. The pK as of [dpypcp(H)2]2+ are 3.66 and 4.77. Dpypcp and dpype form exclusively P,P '-bonded complexes with M n-halo precursors: MX 2 (P -P ) (for M = Pt, X = CI, Br, I when P-P = dpypcp, and X = CI, I when P-P = dpype; for M = Pd, X = CI, Br, I, P-P = dpypcp) and [M(P-P) 2][PF 6] 2 (P-P = dpypcp, dpype) have been isolated and characterised by *H and 3 1 P{ 1 H} N M R spectroscopy and elemental analysis (some of the complexes are isolated as 199 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes monohydrates), and in the cases of Pfi2(dpypcp) and PtCi2(dpype) by X-ray crystallography. Both pyridyldiphosphines serve as supports for "tethered paddlewheel" complexes that contain [Pt(P-P)] 2 4 + moieties in which the ligand adopts P,P '-chelating and bis(P,./V-bridging) coordination modes. These complexes can be made either by treatment of Pti2(P-P) with A g N 0 3 (in which case A g + ions become incorporated via coordination by the "unused" pyridyl N-atoms), or by reaction of the ligand with the halide-free precursor Pt(N03)2(cod). In these compounds, the metal centres are separated by a distances of ca. 2.76 A, corresponding to a Pt—Pt bond-order of ca. 0.52 according to Pauling's R V B M . Dpypcp and dpype react with M(PPh 3) 4 (M = Pd, Pt) to form P,P '-chelated Pt(PPh3)2(P-P) and Pd(P-P)2 complexes (Pd(PPh3)2(dpypcp) was observed in situ in the presence of a 5-fold excess of PPh 3). Pt(PPh3)2(dpypcp) reacts with CDC1 3 to form [PtCl(PPh3)(dpypcp)]+. The Pt complexes react with the electron-deficient olefins dimethyl- and diethylmaleate and fumarate to form Pt(n2-olefm)(P-P) complexes. Although complexes bearing the trans, i.e., fumaric, olefin were isolated from reactions of Pt(PPh3)2(dpypcp) with the cis, i.e., maleic, isomer, the metal does not mediate a cis to trans isomerisation. However, reactions of def and dmf with Pt(PPh3)2(dpypcp) give rise to configurational diastereomers whose relative abundances vary with time; the S,S,R,R/R,R,S,S pair is thermodynamically favoured due to steric interactions between the coordinated olefin and the anilinyldiphosphine ligand. 4.6 Recommendations for Future Work 4.6.1 Synthesis of dpypm, Pd2J complexes, and the WGS reaction A modification to the synthesis of dpypcp (Section 4.7.1) should give access to the ligand l,l-bis(di-o-pyridylphosphino)mefhane, or dpypm (Chart 4.8, (a)). Using this ligand, it should be possible to make bimetallic complexes like (b). 200 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Chart 4.8 (a) N -(b) p y 2 p fj>py2 C I — P d — P d — C I P y 2 p \ / p P y 2 (c) p y 2 p n C I — P d — P d — C I % / p p y 2 (d) 12+ p y 2 p n H ? 0 — P d — P d — O H , 2 I I 2 • N ^ / p p y 2 The pyridylmonophosphine-bridged Pd2 ! complex Pd2Ci2(u-PNi)2 (c) does not react for example with CO to give an "A-frame" insertion product in the same way that Pd2Cl2(P-P)2 (P-P = dmpm, dppm, etc.) compounds do 1 2 4 because of the inflexibility of the bridging ligand. 1 2 5 The PN3 analogue dissolves in H 2 0 to give [Pd2(H20)2(u.-PN3)2]Cl2, (d).6 It is thought that Pd2Cl2(u-dpypm)2 should both be water-soluble (with dissociation of chloride) and be able to react with small molecules in water to form "A-frame" insertion products. The only known, water-soluble Pd2* compounds that react with small molecules (CO and H2S) in water are Pd2Ci2(dmpm)2 and Pd2Cl2(depm)2; the former has been studied by Kubiak's group as an aqueous-phase WGS catalyst,1 2 6 and it would be of interest to determine whether the pyridyldiphosphine analogue has the same capacity. 4.7 Experimental 4.7.1 Racemic-dpypcp To E t 2 0 (100 mL), cooled to -77 °C in a dry ice / acetone bath, was added "BuLi (100 mL, 160 mmol) by cannula, and the mixture allowed to cool for 5 min. To this solution was added 2-bromopyridine (16 mL, 160 mmol) which caused an immediate 201 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes colour change from pale yellow to brown-red. Over 4 h the solution became deep red. rra«s-l,2-(Cl2P)2C5H 8 (7 mL, 40 mmol) in E t 2 0 (30 mL) was then added dropwise over 15 min, and stirring was continued for 2 h at -77 °C. This resulted in the formation of a brown suspension. The mixture was then allowed to warm to r.t., whereupon the slurry was extracted with H 2 S04 (2 x 100 mL, 2 mol L*1) and the red aqueous layer was removed by cannula from the yellow organic layer. The extract was neutralized by the dropwise addition of saturated NaOH (ca. 35 mL) which resulted in the formation of an "oily mass". This mixture was filtered to yield a dark orange "grease" which was re-suspended in acetone (60 mL) to give a fine, white powder and red-brown filtrate. The solid was collected, washed with acetone ( 3 x 2 mL) and then thoroughly with H2O (ca. 150 mL). Reprecipitation from acetone/hexanes yielded a cream-coloured solid which was collected, and dried under vacuum at 100 °C for 72 h. Yield: 6.28 g (35 %). Anal. Calcd for C 2 5 H 2 4 N 4 P 2 : C, 67.9; H , 5.5; N , 12.7. Found: C, 67.9; H , 5.5; N 12.6. The atom-numbering scheme for dpypcp is given in Chart 4.9. Assignments have been made on the basis of APT, ' H - ' H COSY and , 3 C - H HETCOR N M R experiments. Peaks in the C{ H} N M R spectrum of this compound generally appear as multiplets due to coupling to 3 1 P , and, although the coupling constants have not been determined, the number of lines observed in each case is given in parentheses following the peak frequency. ' H N M R (400 MHz, C 6 D 6 , 300 K): 5 1.74 (qn, 2H 9 , V H H = 5.3), 1.97 (m, 2H 8), 2.38 (m, 2H 8), 3.90 (ddd, 2H 7 , VHP = 11-47,VHH(an.i)= 7.78, VH H(gaUche)= 3.21), 6.47 (dd, 2H 5 , V H H = 4.8, V H H = 7.6), 6.51 (dd, 2H 5 , V H H = 4.8, 3 J H H = 7.6), 6.87 (m, 4H 4), 7.32 (m, 2H 3), 7.46 (m, 2H 3 . This pattern could be simulated with unassigned coupling constants of 7.7, 1.9 and 1.8 Hz), 8.38 (pd, 2H 6 , 3 J H H = 10.1), 8.49 (pd, 2H 6 , 3 J H H = 10.1). l 3 C{'H} N M R (50 MHz, CDCh, 300 K): 5 25.1 (3, C 9 ) , 30.2 (3, C 8 ) , 39.1 (4, C 7 ) , 122.5 (2), 129.3 (8), 135.2 (4, C 4 ) , 149.8 (6, C 6 ) 162.9 (1, C 2 ) . 3 1 P{ 1 H} N M R (121 MHz, CDCI3, 300 K): 5 -2.2 (s). The diphosphine was made as a mixture ofR,R- and ^S-enantiomers, the chirality designators referring to the C7-atoms of the cyclopentane backbone. 202 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes Chart 4.9 4.7.2 Racemic-dpypcp(0)2 To a 0.2 mol L"1 HC1 solution (15 mL) containing dpypcp (160 mg, 0.36 mmol) in a flask open to the atmosphere was added dropwise aq. 30 % H2O2 solution (2 mL). The colourless solution was stirred for 1 h, cooled on ice, and then made basic using a saturated K O H solution. The aqueous phase was extracted with CH2CI2 (3 x 25 mL) and the combined organic fractions dried over MgS04. This mixture was filtered and the filtrate reduced to ca. 2 mL under reduced pressure. Et20 (30 mL) was then added to afford the product as a white precipitate, which was collected, washed with Et20 (3 x 3 mL) and dried in vacuo. Yield: 91 mg (53 %). Anal. Calcd for C25H24N402P2: C, 63.3; H, 5.1; N , 11.8. Found: C, 63.1; H , 5.1; N 11.8. The atom-numbering scheme is the same as that used for dpypcp (Chart 4.9). ! H N M R (300 MHz, D 2 0 , 300 K): 5 1.55 (m, 2H, CH2), 1.70 (br m, 2H, CH2), 1.92 (br m, 2H, CH2), 3.64 (br m, 2H, CH), 7.15 (m, 2H, py), 7.32 (m, 2H, py), 7.62 (br m, 4H, py), 7.71 (br m, 4H, py), 7.90 (pd, 2H, py), 8.45 (pd, 2H, py). 1 3 C{'H} N M R (75 MHz, CDC1 3, 300 K): 5 26.2 (1, C 9 ) , 28.8 (1, C 8 ) , 35.0 (2, C 7 ) , 124.7 (2), 128.2 (5), 135.5 (2, C 4 ) , 150.0 (5, C 6 ) , 168.2 (1, C 2 ) . 3 1 P{'H} N M R (121 MHz, 300 K): 5 35.8 (s) [CDCI3]; 38.7 (s) [D 2 0]. v P 0 : 1194 (s). 4.7.3 dpype(0)2 This compound was made in the same manner as dpypcp(0)2 except that hexanes was used rather than E t 2 0 to precipitate the product. Thus, reaction between dpype (87 mg, 0.22 mmol) and a 30 % H 2 0 2 solution (3 mL) in aq. HC1 gave 38 mg (41 %) of a white powder. The elemental analysis of this compound is poor, possibly because of a small KC1 impurity. Anal. Calcd for C22H20N4O2P2: C, 60.8; H , 4.6; N , 12.9. Found: C, 59.8; H , 4.5; N , 12.4. ' H N M R (300 MHz, CDCI3, 300 K): § 2.88 (pd, 4H, CH2), 7.34 (m, 4H, py), 203 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 7.76 (m, 4H, py), 8.06 (m, 4H, py), 8.70 (pd, 4H, py). 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K): 8 29.1 (s). v P 0 : 1206 (s). 4.7.4 [dpypcp(H)2][PF6]2 Some dpypcp (82 mg, 0.19 mmol) was dissolved in aq. HC1 (20 mL, 1.2 mol L"1) to give a colourless solution. To this was added solid KPF6 (95 mg, 0.52 mmol) which immediately gave a white precipitate. The slurry was stirred for 0.5 h and then filtered to give a white powder which was washed with H 2 0 ( 2 x 5 mL), EtOH (3 mL) and Et 2 0 ( 2 x 5 mL) and dried in vacuo. Yield: 97 mg (71 %). Anal. Calcd for C 2 sH 2 6N 4 Fi 2 P 4 : C, 40.9; H, 3.6; N , 7.6. Found: C, 40.9; H , 3.6; N , 7.5. 3 1P{ ]H} N M R (121 MHz, acetone-d6, 300 K): 8 -25.5 (s), -143 (spt, PF6", ' J P F = 710). 4.7.5 [dpype(H)2][PF6]2 The compound was prepared in the same manner as for [dpypcp(H)2][PF6]2. Thus, reaction of dpype (57 mg, 0.14 mmol) and KPF6 (91 mg, 0.49 mmol) in aq. HC1 gave 33 mg (34 %) of a white powder. The elemental analysis of this compound is poor, possibly because of a small K P F 6 and/or KC1 impurity. Anal. Calcd for C 2 2 H 2 2 N 4 F i 2 P 4 : C, 38.1; H , 3.2; N , 8.1. Found: C, 37.3; H , 3.2; N , 7.7. ! H N M R (300 MHz, acetone-d6, 300 K): 8 2.54 (t, 4H, CH2, 2JHP = 5.32), 7.91 (m, 4H, py), 8.03 (m, 4H, py), 8.34 (pt, 4H, py), 9.01 (pd, 4H, py), ca. 11.8 (br, H + ) . 3 1 P{'H} N M R (121 MHz, acetone-d6, 300 K): 5 -21.4 (s), -143 (spt, PF6", 'JPF = 710). 4.7.6 Racemic-dpypcp(S)2 A C6H 6 solution (15 mL) containing dpypcp (250 mg, 0.57 mmol) and Ss (38 mg, 0.15 mmol) was brought to reflux for 3 h. The solution was allowed to cool and then reduced to approximately Vi its original volume in vacuo. An off-white solid began to deposit, and E t 2 0 (20 mL) was added to complete the precipitation. The product was isolated by filtration, washed with C6H6 ( 1 x 3 mL) and E t 2 0 ( 2 x 3 mL) and air-dried. Yield: 240 mg (84 %). ' H N M R (300 MHz, CDC1 3, 300 K): 8 1.79 (m, 2H, CH2), 2.05 (m, 2H, CH2), 2.18 (m, 2H, CH2), 4.69 (m, 2H, CH), 6.99 (m, 2H, py), 7.25 (m, 2H, py), 7.53 (m, 2H, py), 7.68 (m, 2H, py), 8.09 (pd, 2H, py), 8.23 (m, 2H, py), 8.31 (m, 2H, py), 8.70 (pd, 2H, py). 3 1 P{'H} N M R (121 MHz, CDC1 3, 300 K): 8 54.6 (s). 204 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.7.7 PtCl2(dpypcp) A CH2CI2 (10 mL) solution containing dpypcp (100 mg, 0.24 mmol) was added dropwise via cannula over 5 min to a CH2CI2 (5 mL) solution of PtCl2(cod) (93 mg, 0.25 mmol). The resulting colourless solution was stirred for 1 h, and the volume was then reduced to ca. 2 mL. The product was afforded as a fine white precipitate by the addition of Et20 (30 mL), isolated by filtration, and washed with Et20 ( 3 x 5 mL). Reprecipitation from CH 2 C1 2 and drying in vacuo at 100 °C yielded pure product. Yield: 144 mg (86 %). Anal. Calcd for C z s ^ N j C L ^ P t : C, 42.4; H , 3.5; N , 7.9. Found: C, 42.6; H , 3.4; N , 7.6. *H N M R (200 MHz, CDCI3, 300 K): S 1.6 (m, 2H, CH2), 2.1 (m, 4H, CH2), 3.6 (pt, 2H, CH), 7.3-8.8 (m, 16H, py). 31p{lH} N M R (81 MHz, CDCI3, 300 K) : 6 17.9 (s, lJPPl = 3490). 4.7.8 PtBr2(dpypcp) This complex was made in the same manner as for PtCl2(dpypcp). Thus, reaction of PtBr2(cod) (61 mg, 0.13 mmol) and dpypcp (60 mg, 0.14 mmol) gave 62 mg (59 %) of an off-white powder. Anal. Calcd for C 2 5 H 2 4 N 4 B r 2 P 2 P t : C, 37.7; H , 3.0; N , 7.0. Found: C, 38.0; H , 3.2; N , 6.9. ' H N M R (300 MHz, CD 2 C1 2 , 300 K): 5 1.56 (br m, 2H, CH2), 1.80 (br m, 2H, CH2), 2.12 (m, 2H, CH2), 3.59 (m, 2H, CH), 7.42 (m, 2H, py), 7.48 (m, 2H, py), 7.78 (m, 2H, py), 7.92 (m, 2H, py), 7.97 (m, 2H, py), 8.55 (pt, 2H, py), 8.75 (pt, 4H, py). 3 1 P{'H} N M R (121 MHz, CD 2 C1 2 , 300 K): 5 17.6 (s, 'jppt = 3460). 4.7.9 Ptl2(dpypcp) Preparation of this complex in the same manner as that outlined for PtCl2(dpypcp) yielded, in addition, a small quantity of [Pt(dpypcp)2]I2, as determined by 3 1 P{'H} N M R spectroscopy (see Section 4.7.13). This contaminant was removed by washing the product with dil. HC1 (ca. 50 mL). Thus, reaction of PtI2(cod) (100 mg, 0.18 mmol) and dpypcp (79 mg, 0.18 mmol) yielded 129 mg (80 %) of a yellow powder. Anal. Calcd for C 2 5 H 2 4 N 4 I 2 P 2 Pt : C, 33.7; H , 2.7; N , 6.3. Found: C, 34.1; H , 2.7; N , 6.0. *H N M R (300 MHz, CDCI3, 300 K) : § 1.55 (m, 2H, CH2), 1.75 (m, 2H, CH2), 2.10 (m, 2H, CH2), 3.65 (2H, m, CH), 7.3-8.8 (m, 16H, py). 3 1 P{ 1 H} N M R (121 MHz, CDC1 3, 300 K): 8 205 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 12.6 (s, ' J p p t = 3290). Crystals of Ptl2(dpypcp)-0.18 CH2CI2 were isolated after 72 h from a CH2CI2 solution onto which Et 2 0 had been layered. 4.7.10 PtCl2(dpype) This compound was prepared in the same manner as for PtCl2(dpypcp), except that the product was washed with MeOH instead of Et^O. Reaction of PtCl2(cod) (230 mg, 0.60 mmol) and dpype (240 mg, 0.60 mmol) yielded 300 mg (70 %) of a white powder. Anal. Calcd for C 22H2oN 4Cl2P2Pt: C, 39.5; H , 3.0; N , 8.4. Found: C, 39.7; H , 3.0; N , 8.1. ' H N M R (300 MHz, CDCI3, 300 K): 5 2.9 (m, 4H, CH2), 7.4-8.7 (m, 16H, pyridyl). 3 , P{ 'H} N M R (121 MHz, CDC1 3 , 300 K): 5 47.1 (s, ' j p P t = 3480). Colourless crystals of PtCl2(dpype)-CH2Cl2, suitable for study by X-ray diffraction, were isolated after 24 h from a CH2CI2 solution which had been layered with Et 2 0. 4.7.11 Ptl2(dpype) This compound was prepared in the same manner as for PtCl2(dpypcp). Thus, reaction of PtI2(cod) (71 mg, 0.13 mmol) and dpype (51 mg, 0.13 mmol) yielded 50 mg (46 %) of a yellow powder. Anal. Calcd for C ^ o ^ ^ P t : C, 31.0; H , 2.4; N , 6.6. Found: C, 31.2; H, 2.4; N , 6.4. *H N M R (300 MHz, CD 2 C1 2 , 300 K) 5 2.73 (m, 4H, CH2), 7.25-8.85 (m, 16H, py). 3 1 P{ ] H} N M R (121 MHz, CDC1 3, 300 K) 8 49.8 (s, ]J?Pt= 3400). 4.7.12 [Pt(P-P)2]X2 (X = CI, I) Preparations of the compounds [Pt(P-P) 2]X 2 (P-P = dpype, dpypcp; X = CI, I) either by reaction of two equiv. of the ligand with PtX2(cod) in CH 2 C1 2 or, in the case of the iodide salts, by metathesis of [Pt(P-P) 2]Cl 2 with Nal in acetone, were successful as determined by ' H and 3 1 P{'H} N M R spectroscopies. However, satisfactory elemental analyses were not obtained for these complexes. 4.7.13 [Pt(dpypcp)2][PF6]2H20 To a Schlenk tube charged with PtCl2(cod) (14 mg, 0.04 mmol), dpypcp (34 mg, 0.08 mmol) and NH4PF6 (13 mg, 0.08 mmol) were added CH 2 C1 2 (3 mL) and acetone (4 mL). The resulting cloudy mixture was stirred at r.t. for 75 min and then filtered through Celite 545. The volume of the colourless filtrate was reduced to ca. 2 mL under 206 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes vacuum and the product, afforded as a white precipitate by the addition of E t 2 0 (20 mL), was isolated by filtration and dried overnight in vacuo. Yield: 39 mg (73 %). Anal. Calcd for C5oH 5oN 8Fi 2OP 6Pt: C, 43.3; H , 3.8; N , 8.1. Found: C, 43.1; H , 3.6; N , 8.0. ' H N M R (400 MHz, CDCI3, 300 K): 5 1.27 (br m, 4H, CH2), 1.56 (s, 2H, H20), 1.72 (m, 4H, CH2), 2.13 (m, 4H, CH2), 3.45 (m, 4H, CH), 7.15-8.75 (m, 32H, py). 3 1 P{'H} N M R (81 MHz, CDCI3, 300 K): 5 18.40, 18.44 (s, ^ = 2400), -144 (spt, PF6", :JP¥ = 710). 4.7.14 [Pt(dpype)2][PF6]2 An impure sample of [Pt(dpype)2]Cl2 (50 mg, ca. 0.05 mmol) and NH4PF6 (16 mg, 0.01 mmol) were dissolved in acetone (20 mL) and the resulting cloudy solution was stirred for 30 min; the mixture was filtered through Celite 545 which was subsequently washed with acetone ( 3 x 5 mL) and the combined filtrate reduced to ca. 2 mL. The product was afforded as a white precipitate by the addition of E t 2 0 (30 mL), isolated by filtration, washed with E t 2 0 ( 3 x 5 mL) and dried in vacuo. Yield: 42 mg (77 %). Anal. Calcd for C44H 4oN 8Fi 2P 6Pt: C, 41.0; H , 3.1; N , 8.7. Found: C, 40.8; H , 3.0; N , 8.5. *H N M R (300 MHz, CD 3 OD, 300 K): 5 1.30 (m, 8H, CH2), 5.7-7.1 (m, 32H, py). 3 1 P{ ! H} N M R (81 MHz, C D 3 O D , 300 K): 5 54.4 (s, xJP?l = 2480), -144 (spt, P F 6 \ V P F = 710). 4.7.15 PdCl2(dpypcp) To a CH 2 C1 2 solution (10 mL) containing PdCl2(cod) (57 mg, 0.20 mmol) was added a CH 2 C1 2 (10 mL) solution containing dpypcp (87 mg, 0.20 mmol) over the course of ca. 2 min. Although the solution was stirred for 2 h, the colour change from orange to yellow was complete within 0.5 h. A fine precipitate which formed during this time was removed by filtration through Celite 545. The yellow filtrate was concentrated to ca. 2 mL and Et 2 0 (20 mL) was added to give the product as a yellow powder. This was isolated by filtration, washed with E t 2 0 ( 3 x 3 mL) and dried in vacuo. Yield: 100 mg (94 %). Anal. Calcd for C 2 5 H 2 4 N 4 C l 2 P 2 P d : C, 48.4; H , 3.9; N , 9.0. Found: C, 48.0; H , 3.8; N , 8.8. *H N M R (400 MHz, CDC1 3 , 300 K): 8 1.66 (m, 2H, CH2), 1.82 (m, 2H, CH2), 2.10 (m, 2H, CH2), 3.81 (m, 2H, CH), 7.40 (m, 4H, py), 7.73 (m, 2H, py), 7.84 (m, 2H, py), 8.14 (m, 2H, py), 8.49 (m, 2H, py), 8.69 (m, 4H, py). 3 1 P{'H} N M R (162 MHz, CDC1 3 , 300 K): § 39.3 (s). 207 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.7.16 PdBr2(dpypcp)H20 This complex was made in the same manner as for PdCl2(dpypcp). Thus, reaction of PdBr2(cod) (47 mg, 0.12 mmol) and dpypcp (55 mg, 0.12 mmol) gave 81 mg (92 %) of a yellow powder. Anal. Calcd for C 2 5 H 2 6 N 4 B r 2 O P 2 P d : C, 41.3; H , 3.6; N , 7.7. Found: C, 41.4; H , 3.5; N , 7.5. ! H N M R (300 MHz, CDC1 3, 300 K): 5 1.56 (s, 2H, H20), 1-65 (br m, 2H, CH2), 1.83 (br m, 2H, CH2), 2.10 (m, 2H, CH2), 3.85 (br m, 2H, CH), 7.38 (m, 2H, py), 7.45 (m, 2H, py), 7.74 (m, 2H, py), 7.88 (m, 2H, py), 8.08 (m, 2H, py), 8.58 (m, 2H, py), 8.67 (pd, 2H, py), 8.73 (pd, 2H, py). 3]?{]R} N M R (121 M H z , CDCI3, 300K): 8 36.9 (s). 4.7.17 Pdl2(dpypcp) This complex was made in the same manner as for PdCl2(dpypcp). Thus, reaction of PdI2(cod) (58 mg, 0.13 mmol) and dpypcp (55 mg, 0.12 mmol) gave 73 mg (73 %) of an orange powder. Anal. Calcd for C 2 5 H 2 4 N 4 I 2 P 2 P d : C, 37.4; H , 3.0; N , 7.0. Found: C, 37.6; H, 3.0; N , 6.8. j H N M R (300 MHz, CDC1 3, 300 K): § 1.57 (br m, 2H, CH2), 1.74 (br m, 2H, CH2), 2.07 (m, 2H, CH2), 3.81 (br m, 2H, CH), 7.34 (m, 2H, py), 7.45 (m, 2H, py), 7.73 (m, 2H, py), 7.88 (m, 2H, py), 8.05 (pd, 2H, py), 8.62 (m, 2H, py), 8.66 (pd, 2H, py), 8.75 (pd, 2H, py). ^ P l ' H } N M R (121 MHz, CDCI3, 300 K): S 30.0 (s). 4.7.18 [Pd(dpypcp)2][PF6]2H20 This complex was made in the same manner as for [Pt(dpypcp)2][PF6]2H20. Thus, reaction of trans-PdCl2(PhCN)2 (23 mg, 0.06 mmol), dpypcp (54 mg, 0.12 mmol) and NH4PF6 (37 mg, 0.22 mmol) gave 61 mg (80 %) of a pale yellow powder. Anal. Calcd for C 5 oH 5 oN 8 F 1 2 OP 6 Pd: C, 46.2; H , 3.9; N , 8.6. Found: C, 46.1; H , 3.8; N , 8.4. This compound exists as a 1:1 mixture of diastereomers. Where the peaks for each are resolved they are given as comma-separated pairs. ! H N M R (300 MHz, CD 2 C1 2 , 300 K): 5 1.56 (s, 2H, H20), 1.97 (br m, 4H, CH2), 1.61, 1.71 (br m, 4H, CH2), 1.98, 2.07 (br m, 4H, CH2), 3.49, 3.62 (br m, 4H, CH), 7.26 (m, 4H, py), 7.37 (m, 2H, py), 7.66 (m, 16H, py), 7.54 (m, 2H, py), 8.16, 8.54 (pd, 4H, py), 8.38, 8.76 (pd, 4H, py). ^ P l ' H } N M R (121 MHz, CD 2 C1 2 , 300 K): 5 29.6, 30.2 (s), -144 (spt, lJP? = 710). 208 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.7.19 Preparation of [Pt2(dpype)2Ag4(N03)8(H20)2]„, 1 To a mixture of glacial acetic acid (4 mL) and EtOH (15 mL) in a Schlenk tube open to the atmosphere were added Ptl2(dpype) (110 mg, 0.13 mmol) and A g N 0 3 (250 mg, 1.5 mmol), and the resulting yellow slurry was stirred for 4.5 h. Distilled H 2 0 (~ 1 mL) was then added and the mixture was stirred for a further 48 h when it became cream in colour. The suspension was then filtered through Celite 545 to afford a pale yellow solution. Yellow crystals of 1 deposited by slow evaporation of this solution in the dark over a period of 3 months. Yield (of isolated crystals, based on Pt): 12%. Anal. Calcd for C44H44N16026P4Pt2: C, 24.5; H , 2.1; N , 10.4. Found: C, 24.1; H , 2.0; N , 10.5. ] H N M R (400 MHz, D 2 0 , 300 K): 5 3.05 (m, 4H, CH2), 4.10 (m, 4H, CH2), 7.53 (pt, 4H, py), 7.53 (pt, 4H, py), 8.00 (m, 16H, py), 8.58 (d, 4H 6 , lJm = 4.53), 8.78 (d, 4H 6 ' , ]JHH = 5.05). 3 1 P{'H} N M R (81 MHz, D 2 0 , 300 K): 8 41.5 (s, ]Jm = 3200; V P P T = 69). UV-vis (H 20): 346 [6030]. 4.7.20 Preparation of [Pt2(dpypcp)2Ag6(N03)io]n, 2 In a procedure similar to that used for the synthesis of 1, addition of Ptl2(dpypcp) (ca. 5 mg, 0.006 mmol) and a large excess of A g N 0 3 to a mixture of glacial acetic acid (1 mL) and EtOH (3 mL) resulted in the formation of a yellow solution which, by slow evaporation in the dark over 3 weeks, yielded a small number of orange crystals. The yield was not determined. : H N M R (400 MHz, D 2 0 , 300 K): 8 1.30 (m, 2H, CH2), 1.46 (dq, 2H, CH2), 1.74 (m, 2H, CH2), 2.15 (m, 2H, CH2), 2.33 (m, 4H, CH2), 2.80 (m, 4H, CH), 7.43 (pt, 1H, py), 7.61 (m, 6H, py), 7.84 (m, 6H, py), 8.11 (m, 11H, py), 8.32 (d, 1H, py), 8.61 (d, 1H, py), 8.75 (pd, 2H, py), 8.90 (pd, 2H, py), 8.97 (d, 1H, py), 9.05 (pd, 1H, py). UV-vis (H 2 0): 348 [5500]. The limited amount of 2 obtained precluded elemental analysis determination and measurement of 3 1 P{ 1 H} N M R data. 4.7.21 [Pt2(dpype)2][N03]4-2H20 To a combination of PtI2(cod) (180 mg, 0.32 mmol) and A g N 0 3 (HOmg, 0.64 mmol) was added EtOH (5 mL) and H 2 0 (2 mL), and the mixture was stirred in the dark for 1 h. The solvent was removed at the pump, the residue was taken up in MeOH and the mixture filtered through Celite 545. Some dpype (130 mg, 0.32 mmol) was added to the filtrate as a solid. The pale yellow solution was stirred overnight and a yellow precipitate 209 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes formed. This was isolated by filtration and washed with MeOH ( 3 x 3 mL) in which it was slightly soluble. Yield of first crop: 31 mg (13 %). The filtrate was stirred for a further 2.5 h during which more product deposited. The slurry was reduced in vacuo about l /5 t h of its original volume. Acetone (10 mL) and E t 2 0 (15 mL) were added to complete the precipitation of the second crop which was isolated by filtration and dried in vacuo at r.t. Yield: 83 mg (36 %). Anal. Calcd for C44H44Ni2Oi4P4Pt2: C, 35.7; H , 3.0; N , 11.4. Found: C, 35.7; H , 3.0; N , 11.3. An attempt to remove the hydrates by heating the complex to 100 °C under vacuum resulted in decomposition. The N M R and UV-vis spectroscopic data were the same as those found for 1. 4.7.22 Pt(PPh3)2(dpypcp) To a combination of dpypcp (58 mg, 0.13 mmol) and Pt(PPh3)4 (160 mg, 0.13 mmol) was added CeH 6 (10 mL) and the resulting orange solution was stirred for 2 h. The solvent was removed in vacuo, and E t 2 0 (3 x 10 mL) was added to the residue which was triturated to give a finely-divided solid in an orange solution. The product was isolated by filtration, washed thoroughly with E t 2 0 (3 x 10 mL) to remove PPh 3 and dried under vacuum. Yield: 130 mg (84 %). Anal. Calcd for C 6iH 54N 4P4Pt: H , 63.1; H , 4.7; N , 4.8. Found: C, 63.1; H , 4.7; N , 4.9. ] H N M R (300 MHz, C 6 D 6 , 300 K): 5 2.24 (br m, 2H, CH2), 2.72 (br m, CH2), 3.10 (br m, 2H, CH2), 4.84 (br m, 2H, CH), 6.47 (pt, 2H, py), 6.77 (pt, 2H, py), 6.88 (pt, 2H, py), 7.00 (pd, 2H, py), 7.15 (m, 8H, py and p- Ph), 7.67 (m, 24H, o- and m- Ph), 8.28 (pd, 2H, py), 8.50 (pd, 2H, py), 8.94 (pd, 2H, py). 3 1 P{ ] H} N M R (121 MHz, C 6 D 6 , 300 K): 5 14.6 (t, 2JPP = 52, ]JPPt = 3550), 23.1 (t, 2JPP = 52, ]JPPt = 4210). 4.7.23 Pt(PPh3)2(dpype) The complex was made in the same manner as for Pt(PPh3)2(dpypcp). Thus, reaction of Pt(PPh3)4 (240 mg, 0.19 mmol) and dpype (77 mg, 0.19 mmol) gave 150 mg (69 %) of an orange powder. Anal. Calcd for C58H5oN4P4Pt: C, 62.1; H , 4.5; N , 5.0. Found: 61.6; H , 4.7; N , 4.9. *H N M R (300 MHz, C 6 D 6 , 300 K): § 3.63 (m, 4H, CH2), 6.32 (pt, 4H, py), 6.64 (pt, 4H, py), 6.88 (m, py and Ph), 7.33 (m, Ph), 8.34 (pd, 2H, py). 3 1 P{ 1 H} N M R (121 MHz, C 6 D 6 , 300 K): § 21.3 (t, V P P = 53, ]JPPt = 4180), 32.0 (t, 2JPP = 53, ]JPPt = 3330). 210 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.7.24 Pt(dpypcp)2 This compound was made in the same manner as for Pt(PPh3)2(dpypcp) except that a ligand:metal ratio of 2:1 was used. Thus, reaction of dpypcp (50 mg, 0.11 mmol) and Pt(PPh3)4 (70 mg, 0.056 mmol) gave 36 mg (59 %) of a red powder. Because of the air-sensitivity of this compound, satisfactory elemental analysis could not be obtained. The best is given by: Anal. Calcd for C5oH48N8P4Pt: C, 55.6; H , 4.5; N , 10.4. Found: C, 54.3; H, 4.5. N , 9.6. ' H N M R (300 MHz, C 6 D 6 , 300 K) : Major diastereomer: 5 1.47 (br m, 4H, CH2), 1-95 (br m, 4H, CH2), 2.07 (br m, 4H, CH2), 3.80 (m, 4H, CH), 6.47, 6.90, 7.67, 8.09, 8.37, 8.45. 3 1 P{'H} N M R (121 MHz, C 6 D 6 , 300 K): Major diastereomer: 8 20.1 (s, 'JPP, = 3570). Minor diastereomer: 8 23.5 (s, 1 J?Pt = 3580). 4.7.25 Pt(dpype)2 This compound was made in the same manner as for Pt(dpypcp)2. Thus, an overnight reaction between Pt(PPh3)4 (62 mg, 0.050 mmol) and dpype (41 mg, 0.010 mmol) gave 27 mg (56 %) of an orange powder. Once again, the air-sensitivity of this compound precluded satisfactory elemental analysis. *H N M R (300 MHz, C 6 D 6 , 300 K): 8 3.35 (m, 8H, CH2), 6.39 (pt, 8H, py), 6.72 (pt, 8H, py), 7.50 (pd, 8H, py), 8.34 (pd, 8H, py). 3 1 P{ 1 H} N M R (121 MHz, C 6 D 6 , 300 K): 8 41.2 (s, ]J?Pt = 3600). 4.7.26 Pt(ii2-dmf)(dpypcp) To an orange CeH-6 solution (5 mL) containing Pt(PPh3)2(dpypcp) (33 mg, 0.029 mmol) was added dmm (0.2 mL). The solution was stirred for 2 h when it became colourless. The volume was reduced in vacuo to < 1 mL and E t 2 0 (10 mL) and hexanes (20 mL) were added. The solution was then cooled over liquid-N 2, and a white precipitate formed. This was isolated by filtration and dried in vacuo. Yield: 11 mg (47 %). Anal. Calcd for C 3 i H 3 2 N 4 0 4 P 2 P t : C, 47.6; H , 4.1; N , 7.2. Found: C, 47.1; H , 4.1; N , 6.4. ' H ^ ' P } N M R (500 MHz, CDC1 3, 300 K) : Major diastereomer: 8 1.60 (br m, 2H, CH2), 2.18 (br m, 2H, CH2), 2.27 (m, 2H, CH2), 3.13 (m, 2H, CH), 3.30 (s, 6H, Gf73), 3.58 (s, 2H, =CH, 2JHPt = 60.7; turning off the 3 1 P decoupler gives V H p = 8.05), 7.18 (pdd, 2H, Ar, 3JHH = 4.83, V H H = 7.56), 7.29 (pdd, 2H, Ar, V H H = 4.91, V H H = 6.80), 7.60 (pdt, 2H, Ar, 3Jm = 7.65, V H H = 1-69), 7.72 (pd, 2H, Ar, V H H = 6.86), 7.76 (pdt, 2H, Ar, 3 / H H = 7.69, V H H = 1-68), 8.09 (pd, 2H, Ar, 3JHH = 7.77), 8.60 (pd, 2H, Ar, V H H = 4.72), 8.69 (pd, 2H, Ar, V H H = 211 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.84). Minor diastereomer: 5 2.05 (br m, 2H, CH2), 2.12 (br m, 2H, CH2), 3.24 (s, 6H, CH 3 ) , 7.56 (pd, 2H, Ar, V H H = 7.78), 7.65 (pdt, 2H, Ar, V H H = 7.71, V H H = 1-59), 8.12 (pd, 2H, Ar, 3 J H H = 7.52), 8.64 (pd, 2H, Ar, 3 J H H = 4.71). A l l other peaks are obscured by those of the major diastereomer. 3 , P{ 'H} N M R (202 MHz, CDC1 3 , 300 K): Major diastereomer: 5 33.1 (s, 'jppt = 3360). Minor diastereomer: 5 31.8 (s, 'jppt = 3400). 4.7.27 Pt(n2-def)(dpypcp) To solid Pt(PPh 3) 4 (99 mg, 0.071 mmol) and dpypcp (32 mg, 0.072 mmol) was added CeH6 (1 mL) and the resulting orange solution was stirred for 45 min. An Et20 (1 mL) solution containing dem (0.1 mL) was then added via cannula. This caused the immediate formation of a white precipitate which was isolated by filtration, washed with E t 2 0 (5 x 3 mL) and dried in vacuo. Yield: 32 mg (56 %). Dissolution of the product in CH2CI2, filtration through Celite 545 and reprecipitation with E t 2 0 gave the analytically pure product. Anal. Calcd for C 3 3 H 3 6 N 4 0 4 P 2 P t : C 49.0; H 4.5; N 6.7. Found: C 48.7; H 4.5; N 6.7. 'H{ 3 1P} N M R (500 MHz, CDC1 3, 300 K): Major diastereomer: 8 0.87 (t, 6H, C H 2 C / / 3 , VHH = 7.12), 1.53 (m, 2H, CH2), 2.17 (m, 2H, CH2), 2.27 (m, 2H, CH2), 3.11 (m, 2H, CH, = 38.2), 3.58 (d, 2H, =CH, 2JUPt = 61.2; turning off the 3 1 P decoupler gives 3JHp = 8.19), 3.72 (m, 2H, CH2CH2, 2JHH~ 9-10 (by simulation 1 2 7), 2JHU = 7.21), 3.82 (m, 2H, CH2CR2, 2 J H H~ 9-10 (by simulation), 3Jm = 7.21), 7.17 (ddd, 2H, Ar, 3 J H H = 7.65, 3 7 H H = 4.80, V H H = 1.16), 7.29 (ddd, 2H, Ar, V H H = 7.65, V H H = 4.80, V H H = 1.16), 7.59 (pdt, 2H, Ar, V H H = 7.69, V H H = 1.80), 7.73 (pd, 2H, Ar, V H H = 6.48), 7.75 (pdt, 2H, Ar, V H H = 7.71, V H H = 1.77), 8.19 (pd, 2H, Ar, 3 J H H = 7.76), 8.58 (pdd, 2H, Ar, V H H = 4.74, V H H = 0.81), 8.70 (pdd, 2H, Ar, 3JHH = 4.87, V H H = 0.76). Minor diastereomer: 5 0.98 (t, 6H, CH3, VHH = 7.12), 3.54 (s, 2H, CH, 7Hp t = 59.3). A l l other peaks are obscured by those of the major diastereomer. 3 1 P{'H} N M R (202 MHz, CDC1 3, 300 K): Major diastereomer: 8 33.6 (s, 'jppt = 3360). Minor diastereomer: 5 32.1 (s, 'jppt = 3410). 4.7.28 Determination of pK a values for [dpypcp(H)2]2+ The pK a determinations were carried out with the kind assistance of Dr. Song Bin, a former postdoctoral fellow in the laboratory of Dr. Chris Orvig at U B C . 212 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.7.28.1 Materials Water was deionised (Barnstead D8902 and D8904 cartridges) and distilled (Corning MP-1 Megapure still), and depleted in CO2 by boiling under Ar for 30 min. The concentration of the NaOH titrant was established to be 0.1094 mol L" 1 by titration against potassium biphthalate. 4.7.28.2 Instrumentation An automatic titration system controlled by an IBM compatible PC running "in-house" software written in QBasic, and consisting of a Metrohm 713 pH meter equipped with a Metrohm 6.0233.100 electrode and a model 665 Metrohm Dosimat autoburet (5 mL capacity, 0.005 mL accuracy), was used. Titrations were performed at 298 K in a water-jacketed vessel thermostatted by a Julabo UC circulating bath. 4.7.28.3 Procedure A stock solution of dpypcp was made by dissolving the ligand (33.8 mg, 7.64 x 10"2 mmol) in a mixture of dil. HC1 (10.0 mL, 0.1419 mol L" 1 by titration against standard NaOH) and H 2 0 (15.0 mL). Solutions for titration were made by mixing 10.0 mL of the stock solution, H 2 0 (35.0 mL) and aq. NaCl (5.0 mL, 1.60 mol L" 1). It was necessary to add between 3-3.5 mL of NaOH (0.1094 mol L"1) to this solution prior to each titration to raise the pH which was initially too low for accurate determination by the instrument. Computer-controlled titrations were performed by sequential addition via an autoburet of 0.05 mL aliquots of NaOH (0.1094 mol L"1) to this mixture. After equilibration, the electrode potential at each addition was recorded by the computer and translated into a pH value; thus a plot of pH vs. vol. NaOH added was obtained. To determine the pK a values, the plot was analysed using a Newton-Gauss non-linear least-squares curve-fitting program over the pH range 2.7-11.0. The number of protons bound per ligand, n, was calculated at each pH using the difference in vol. of NaOH added between a titration containing dpypcp and one in its absence; this number in mol divided by [dpypcp] gives n. 213 References on page 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 4.8 References 1. Newkome, G. R. Chem. Rev. 1993, 93, 2067. 2. Nicholson, T.; Hirsch-Kuchma, M . ; Schellenbarger-Jones, A. ; Davison, A. ; Jones, A. G. Inorg. Chim. Acta 1998, 267, 319. 3. Baird, I. R.; Smith, M . B.; James, B. R. Inorg. Chim. Acta 1995, 235, 291. 4. Braunstein, P.; Kelly, D. G.; Tiripicchio, A. ; Ugozzoli, F. Bull. Soc. Chim. Fr. 1995,132, 1083. 5. Arena, C. G.; Bruno, G.; Munno, G. D.; Rotondo, E.; Drommi, D.; Faraone, F. Inorg. Chem. 1993, 32, 1601. 6. Xie, Y . ; Lee, C.-L.; Yang, Y . ; Rettig, S. J.; James, B. R. Can. J. Chem. 1992, 70, 751. 7. Francio, G.; Scopelliti, R.; Arena, C. G.; Bruno, G.; Drommi, D.; Faraone, F. Organometallics 1998,17, 338. 8. Chan, W.-H.; Zhang, Z.-Z.; Mak, T. C. W.; Che, C . -M. J. Chem. Soc, Dalton Trans. 1998, 803. 9. Kuang, S.-M.; Xue, F.; Zhang, Z.-Z.; Xue, W.-M. ; Che, C . -M. ; Mak, T. C. W. J. Chem. Soc, Dalton Trans. 1997, 3409. 10. Kuang, S.-M.; Cheng, H.; Sun, L.-J.; Zhang, Z.-Z.; Zhou, Z. Y . ; Wu, B . - M . ; Mak, T. C. W. Polyhedron 1996,15, 3417. 11. Xie, L. Y . ; James, B. R. Inorg. Chim. Acta 1994, 217, 209. 12. Gobetto, R.; Arena, C. G.; Drommi, D.; Faraone, F. Inorg. Chim. Acta 1996, 248, 257. 13. Deeming, A . J.; Smith, M . B. J. Chem. Soc, Chem. Commun. 1993, 844. 14. Braunstein, P.; Knorr, M . ; Stampfer, M . ; Cian, A . D.; Fischer, J. J. Chem. Soc, Dalton Trans. 1994, 117. 15. Drommi, D.; Nicolo, F.; Arena, C. G.; Bruno, G.; Faraone, F.; Gobetto, R. Inorg. Chim. Acta 1994, 221, 109. 16. Gladiali, S.; Pinna, L.; Arena, C. G.; Rotondo, E.; Faraone, F. J. Mol. Catal. 1991, 66, 183. 17. Zhang, Z.-Z.; X i , H.-P.; Zhao, W.-J.; Jiang, K . - Y . ; Wang, R.-J.; Wang, H.-G.; Wu, Y . J. Organomet. Chem. 1993, 454, 221. 18. Xie, Y . ; James, B. R. J. Organomet. Chem. 1991, 417, 277. 214 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 19. Newkome, G. R.; Evans, D. W.; Fronczek, F. R. Inorg. Chem. 1987, 26, 3500. 20. Inoguchi, Y . ; Milewski-Mahrla, B.; Neugebauer, D.; Jones, P. G.; Schmidbaur, H. Chem. Ber. 1983,116, 1487. 21. Alcock, N . W.; Moore, P.; Lampe, P. A. J. Chem. Soc, Dalton Trans. 1982, 207. 22. Wajda, A. ; Pruchnik, F.; Lis, T. Inorg. Chim. Acta 1980, 40, 207. 23. Casares, J. A. ; Espinet, P.; Hernando, R.; Iturbe, G.; Villafafie, F.; Ellis, D. D.; Orpen, A . G. Inorg. Chem. 1997, 36, 44. 24. Casares, J.; Espinet, P. Inorg. Chem. 1997, 36, 5428. 25. Schutte, R. P.; Rettig, S. J.; James, B. R. Can. J. Chem. 1996, 74, 2064. 26. Guillaume, P.; Postel, M . Inorg. Chim. Acta 1995, 233,109. 27. Espinet, P.; Gomez-Elipe, P.; Villafane, F. J. Organomet. Chem. 1993, 450, 145. 28. Schutte, R. P.; Rettig, S. J.; Joshi, A . M . ; James, B. R. Inorg. Chem. 1997, 36, 5809. 29. Xie, Y . Towards Transition Metal Catalyzed Hydration of Olefins; Aquo Ions and Pyridylphosphine-Platinum and Palladium Complexes; Ph.D. Dissertation, University of British Columbia: Vancouver, 1990. 30. Schutte, R. P. Ruthenium(II) Complexes of 2-Pyridylphosphines; Ph.D. Dissertation, University of British Columbia: Vancouver, 1995. 31. LePage, M . D. The Synthesis, Characterization and Reactivity of Nickel 2-Pyridylphosphine Complexes; Ph.D. Dissertation, University of British Columbia: Vancouver, 2000. 32. Kuang, S.-M.; Zhang, Z.-Z.; Wang, Q.-G.; Mak, T. C. W. J. Chem. Soc, Chem. Commun. 1998,581. 33. Zhang, J.-K.; Zhang, Z . -M. ; Yu, A. ; Zhao, S.-L.; Zhang, W.-D.; Zhang, Z.-Z. Polyhedron 1996,15, 2583. 34. McFarlane, H . C. E. ; McFarlane, W.; Muir, A. S. Polyhedron 1990, 9, 1757. 35. Wood, F. E.; Hvoslef, J.; Hope, H.; Balch, A . L. Inorg. Chem. 1984, 23, 4309. 36. Jiang, Q.; Plew, D. V. ; Murtuza, S.; Zhang, X . Tetrahedron Lett. 1996, 37, 797. 37. Sablong, R.; Newton, C ; Dierkes, P.; Osborn, J. A . Tetrahedron Lett. 1996, 37, 4933. 38. Sablong, R.; Osborn, J. A. Tetrahedron Lett. 1996, 37, 4937. 39. Hahn, C ; Sieler, J.; Taube, R. Polyhedron 1998, 17, 1183. 215 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 40. Hahn, C ; Vitagliano, A. ; Giordano, F.; Taube, R. Organometallics 1998, 77, 2060. 41. Rahmouni, N . ; Osborn, J. A. ; Cian, A. D.; Fischer, J.; Ezzamarty, A . Organometallics 1998,77, 2470. 42. Jia, G.; Lee, H . M . ; Williams, I. D.; Lau, C. P.; Chen, Y . Organometallics 1997, 76,3941. 43. Berners-Price, S. J.; Bowen, R. J.; Harvey, P. J.; Healy, P. C ; Koutsantonis, G. A . J. Chem. Soc. Dalton Trans. 1998, 1743. 44. Mirabelli, C. K. ; Hi l l , D. T.; Faucette, L. F.; McCabe, F. L.; Girard, G. R.; Brian, D. B.; Sutton, B. M . ; Bartus, J. O. L.; Crooke, S. T.; Johnson, R. K . J. Med. Chem. 1987, 30, 2181. 45. Budzelaar, P. H. M . ; Frijns, J. H . G.; Orpen, A . G. Organometallics 1990, 9, 1222. 46. Allen, D. L.; Gibson, V . C ; Green, M . L. H. ; Skinner, J. F.; Bashkin, J.; Grebenik, P. D. J. Chem. Soc, Chem. Commun. 1983, 895. 47. Davies, W. C ; Mann, F. G. J. Chem. Soc. 1944, 276. 48. Mann, F. G.; Watson, J. J. Org. Chem. 1948,13, 502. 49. Wibaut, J. P.; de Jonge, A. P.; van der Voort, H . G. P.; Otto, P. P. H. L. Rec. Trav. Chim. 1951, 70, 1054. 50. Plazek, E.; Tyka, R. Zeszyty Nauk. Politech. Wroclaw., Chem. 1957, 4, 79; through Chem. Abs. 52:20156c. 51. Kurtev, K. ; Ribola, D.; Jones, R. A. ; Cole-Hamilton, D. J.; Wilkinson, G. J. Chem. Soc, Dalton Trans. 1980, 55. 52. Hi l l , D. T. European Patent 0 198 696, 1986 (Smithkline Beckman). 53. Jones, N . D.; MacFarlane, K. S.; Smith, M . B.; Schutte, R. P.; Rettig, S. J.; James, B. R. Inorg. Chem. 1999, 3956. 54. Bowen, R. J.; Garner, A . C ; Berners-Price, S. J.; Jenkins, I. D.; Sue, R. E. J. Organomet. Chem. 1998, 554, 181. 55. Smith, M . B . New Phosphorus Ligands as Water-Solubilising Agents for Late Transition Metals, 1994 (research report to B. R. James). 56. Sen, A. ; Halpern, J. J. Am. Chem. Soc. 1977, 99, 8337. 57. Jackobsen, H . J. J. Mol. Spectrosc. 1970, 34, 245. 58. Griffin, G. E.; Thomas, W. A. J. Chem. Soc. (B) 1970, 477. 59. Staab, H. A. ; Saupe, T. Angew. Chem. Int. Ed. Engl. 1988, 27, 865. 216 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 60. Wozniak, K. ; Krygowski, T. M . ; Kariuki, B.; Jones, W.; Grech, E. J. Mol. Struct. 1990, 240, 111. 61. Bemers-Price, S. J.; Bowen, R. J.; Hambley, T. W.; Healy, P. C. J. Chem. Soc, Dalton Trans. 1999, 1337. 62. Bemers-Price, S. J.; Bowen, R. J.; Galettis, P.; Healy, P. C.; McKeage, M . J. Coord. Chem. Rev. 1999, 185-186, 823. 63. Jones, N . D.; Rettig, S. J.; James, B. R. J. Cluster Sci. 1998, 9, 243. 64. Dixon, K. R. in Multinuclear NMR; Mason, J., Ed.; Plenum: New York, 1987. 65. Neri, B. P.; Wilson, G. S. Anal. Chem. 1973, 45, 442. 66. Garrou, P. E. Chem. Rev. 1981, 229. 67. Pidcock, A. ; Richards, R. W.; Venanzi, L. M . Proc Chem. Soc. 1962, 184. 68. Grim, S. O.; Keiter, R. L.; McFarlane, W. Inorg. Chem. 1967, 6, 1133. 69. Orpen, A . G.; Brammer, L.; Allen, F. H. ; Kennard, O.; Watson, D. G. J. Chem. Soc, Dalton Trans. 1989, SI. 70. Engelhardt, L. M . ; Patrick, J. M . ; Raston, C. L.; Twiss, P.; White, A . H. Aust. J. Chem. 1984, 37, 2193. 71. Fordyce, W. A. ; Brummer, J. G.; Crosby, G. A. J. Am. Chem. Soc. 1981, 103, 7061. 72. Isci, H. ; Mason, W. R. Inorg. Chem. 1985, 24, 1761. 73. Che, C . -M. ; Butler, L. G.; Gray, H. B. J. Am. Chem. Soc. 1981,103, 7796. 74. Bellitto, C ; Flamini, A. ; Gastaldi, L.; Scaramuzza, L. Inorg. Chem. 1983, 22, AAA. 75. Appleton, T. G.; Byriel, K. A.; Hall, J. R.; Kennard, C. H . L. ; Mathieson, M . T. J. Am. Chem. Soc. 1992,114, 7305. 76. El-Mehdawi, R.; Fronczek, F. R.; Roundhill, D. M . Inorg. Chem. 1986, 25. 11. Cotton, F. A. ; Matonic, J. H. ; Murillo, C. A. Inorg. Chem. 1996, 35, 498. 78. Cotton, F. A . ; Matonic, J. H . ; Murillo, C. A. Inorg. Chim. Acta 1997, 264, 61. 79. Zangrando, E.; Pichierri, F.; Randaccio, L.; Lippert, B. Coord. Chem. Rev. 1996, 156, 275. 80. Navarro, J. A. ; Romero, M . A.; Salas, J. M . ; Quiros, M . ; Bahraoui, J. E.; Molina, J. Inorg. Chem. 1996, 35. 81. Simonsen, O.; Toftlund, H. Inorg. Chem. 1981, 20, 4044. 217 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 82. Bellitto, C ; Flamini, A. ; Piovesana, O.; Zanazzi, P. F. Inorg. Chem. 1980, 19, 3632. 83. Pinto, M . A . F. dos R.; Sadler, P. J.; Neidle, S.; Sanderson, M . R.; Subbiah, A. ; Kuroda, R. J. Chem. Soc., Chem. Commun. 1980, 13. 84. Bandoli, G.; Caputo, P. A. ; M i n i , F. P.; Sivo, M . F.; Natile, G. J. Am. Chem. Soc. 1997,119, 10370. 85. Brown, M . P.; Cooper, S. J.; Frew, A . A. ; Manojlovic, L. M . ; Muir, K . W.; Puddephatt, R. J.; Seddon, K. R.; Thomson, M . A. Inorg. Chem. 1981, 20, 1500. 86. Cotton, F. A . ; Matusz, M . ; Poli, R.; Feng, X . J. Am. Chem. Soc. 1988,110, 1144. 87. Novoa, J. J.; Aullon, G.; Allemany, P.; Alvarez, S. J. Am. Chem. Soc. 1995,117. 88. Mealli, C ; Pichierri, F.; Randaccio, L.; Zangrando, E.; Krumm, M . ; Holtenrich, D.; Lippert, B. Inorg. Chem. 1995, 34, 3418. 89. ' Pauling, L. The Nature of the Chemical Bond; 3rd ed.; Cornell University Press: Ithaca, N .Y. , 1960. 90. Pauling, L. J. Am. Chem. Soc. 1947, 69, 542. 91. Reis, A . H. ; Peterson, S. W. Inorg. Chem. 1976,15, 3186. 92. Lancashire, R. I. in Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D. and McCleverty, J. A. , Eds.; Pergamon Press: Oxford, 1987; Vol . 5. 93. Ugo, R.; Cariati, F.; Monica, G. L. Inorg. Synth. 1968, 11, 105. 94. Chatt, J.; Mason, R.; Meek, D. W. J. Am. Chem. Soc. 1975, 97, 3826. 95. Al-Ohaly, A.-R.; Nixon, J. F. Inorg. Chim. Acta 1981, 47, 105. 96. Dervisi, A. ; Edwards, P. G.; Newman, P. D.; Tooze, R. P. J. Chem. Soc, Dalton Trans. 2000, 523. 97. Wicht, D. K. ; Zhuravel, M . A. ; Gregush, R. V. ; Glueck, D. S. Organometallics 1998,17, 1412. 98. Chatt, J.; Shaw, B. L.; Williams, A. A . J. Chem. Soc. 1962, 3269. 99. Bland, W. J.; Kemmitt, R. D. W. J. Chem. Soc. (A) 1968, 1278. 100. Roundhill, D. M . ; Wilkinson, G. J. Chem. Soc (A) 1968, 506. 101. Baddley, W. H. ; Venanzi, L. M . Inorg. Chem. 1966, 5, 33. 102. Ashley-Smith, J.; Green, M . ; Wood, D. C. J. Chem. Soc. (A) 1970, 1847. 103. Anderson, G. K. ; Lumetta, G. J. Inorg. Chem. 1987, 26, 1518. 218 Chapter 4: Pyridyldiphosphine Ligands, Derivatives, and Pt and Pd Complexes 104. McCrindle, R.; Ferguson, G.; McAlees, A . J.; Arsenault, G. J.; Gupta, A. ; Jannings, M . C. Organometallics 1995,14, 2741. 105. McCrindle, R.; Arsenault, G. J.; Hampden-Smith, M . J.; Rice, R. E.; McAlees, A . J. J. Chem. Soc, Dalton Trans. 1991, 949. 106. Malatesta, L.; Cariello, C. J. Chem. Soc. 1958, 2323. 107. Gladysz, J. A. ; Boone, B. J. Angew. Chem. Int. Ed. Engl. 1997, 36, 551. 108. Hartley, F. R. in Comprehensive Organometallic Chemistry; 1st ed.; Wilkinson, G., Stone, F. G. A . and Abel, E. W., Eds.; Pergamon: Oxford, 1982; Vol . 6, p 471. 109. Gillespie, R. J.; Nyholm, R. S. Quart. Rev. 1957,11, 339. 110. Dewar, M . J. S. Bull. Soc Chim. Fr. 1951, 18, C79. 111. Chatt, J.; Duncanson, L. A . J. Chem. Soc. 1953, 2939. 112. Francis, J. N . ; McAdam, A. ; Ibers, J. A . J. Organomet. Chem. 1971, 29, 131. 113. McAdam, A. ; Francis, J. N . ; Ibers, J. A . J. Organomet. Chem. 1971, 29, 149. 114. Bombieri, G.; Forsellini, E.; Panattoni, C ; Graziani, R.; Bandioli, G. J. Chem. Soc. (A)\91Q, 1313. 115. Osborne, R. B.; Lewis, H. C ; Ibers, J. A . J. Organomet. Chem 1981, 208, 125. 116. Panattoni, C ; Graziani, R.; Bandoli, G.; Clemente, D. A. ; Belluco, U . J. Chem. Soc (B) 1970,371. 117. Baraban, J. N . ; McGinnety, J. A . Inorg. Chem. 1974,13, 2868. 118. Caruso, F.; Camalli, M . ; Pellizer, G.; Asaro, F.; Lenarda, M . Inorg. Chim. Acta 1991,181, 167. 119. Brown, C. J. Acta Crystallogr. 1966, 21, 1. 120. Bednowitz, A. L.; Post, B. Acta Crystallogr. 1966, 21, 566. 121. Taylor, R.; Kennard, O. J. Am. Chem. Soc. 1982,104, 5063. 122. Emsley, J. Chem. Soc. Rev. 1980, 9, 91. 123. Kvick, A. ; Koetzle, T. F.; Thomas, R. J. Chem. Phys. 1974, 61, 2711. 124. Benner, L. S.; Balch, A. L. J. Am. Chem. Soc. 1978,100, 6099. 125. Maisonnat, A. ; Farr, J. P.; Balch, A. L. Inorg. Chim. Acta 1981, 53, L217. 126. Kullberg, M . L.; Kubiak, C. P. CI Mol. Chem. 1984, 1, 171. 127. Cobas, C ; Cruces, J.; Sardina, J. Magnetic Resonance Companion; Ver. 2.3. 219 Chapter 5: Attempted Catalytic Hydration of Maleic Acid 5 Attempted Catalytic Hydration of Maleic Acid 5.1 Introduction To reiterate the point made in the General Introduction, the aqueous phase catalytic hydration of olefins (Scheme 5.1) takes the concept of atom economy to its highest level: the solvent is environmentally benign, and is a reactant in an addition, zero-waste reaction. In fact, the anti-Markovnikov addition of water to terminal olefins to produce linear alcohols has been cited as one of the ten remaining challenges for homogeneous catalysis.1 In principle, prochiral olefins could be asymmetrically hydrated to give optically pure alcohols. Scheme 5.1 RCH=CH 2 + H 2 0 RCH 2 CH 2 (OH) and/or RC*H(OH)CH 3 There remains a dearth of transition metal complexes which are capable of bringing about this transformation. Aqueous solutions of chlororuthenate(II) catalyse the hydration of C F 2 C F 2 , 2 and Cr 1" and A l 1 " complexes effect the hydration of maleic to malic acid. 3 ' 4 The report by Jensen and Trogler, which claimed that 1-hexene could be hydrated to 1-hexanol by /ra«s-Pt(H)Cl(PMe3)2,5 was shown by Ramprasad et al. to be irreproducible.6 The literature to 1991 has been well reviewed by Xie , 7 and since then there have been few developments. Ganguly and Roundhill have found that complexes of the type [Pd(u.-OH)(L)]22+ (L = dppe, dcype) are minimally active for the hydration of diethylmaleate in H 2 0/THF solution: both give about 14 % conversion of diethylmaleate (dem) to diethylmalate (demOH) after 30 h at 140 °C, and significant quantities of diethylfumarate (17-19 %) and the hydrolysis products maleic acid (22-28 %) and fumaric acid monoethyl ester (27 %) are also formed.8 These researchers have also demonstrated that the combination of Na 2PdCLi and CuCl 2 is as active, giving 15 % conversion of dem to demOH after 30 h 220 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid at 140 °C (together with the isomerisation and hydrolysis products and, in this case, the oxidation product diethyl oxaloacetate).8 Bennett's group has observed what constitutes one step in one of several possible catalytic cycles: the insertion of an alkene into a Pt-OH bond.9 Thus, dimethylmaleate reacts with cw-(L)PtMe(OH) (L = 2PPh 3, dppe, dmpb) to give cis-(L)PtMe(CH(C0 2Me)CH(OH)(C0 2Me)) (Scheme 5.2; only the coordinating P-atoms of the diphosphine ligand are shown). Moreover, these complexes reacted with aqueous acids (HBF 4 and HPFg) to give [cw-(L)PtMe(H20)]+ and dmmOH, corresponding to an overall stoichiometric hydration of the olefin. Scheme 5.2 Important developments in the closely-related catalytic hydroamination reaction (e.g., addition of H and "NHR" across a double bond) have recently been reported by Hartwig and coworkers who used Pd(PPh3)4 for the hydroamination of styrene with aniline. 1 0" 1 3 This discovery, in addition to the similarity of calculated reaction enthalpies of olefin hydrations and hydroaminations,14 is almost certain to reinvigorate efforts in the search of the elusive olefin hydration catalyst. 5.2 Scope This chapter reports the results for the attempted homogeneous, aqueous-phase, catalytic hydration of maleic acid (Scheme 5.3) using water-soluble Pt11, Pd 1 1 and Ru 1 1 complexes of pyridyl- and anilinyldiphosphine ligands as catalyst precursors. 221 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid Scheme 5.3 H O 5.3 Results The results for the attempted catalytic hydrations of maleic acid are given in Table 5.1. Table 5.1. Product distributions for the attempted catalytic hydration of maleic acid (0.1 molL" 1) in water at 100 °C using pyridyl- and anilinyldiphosphine complexes of platinum metals as catalyst precursors (substratexatalyst = 100:1) according to Section 5.7.4. Percent conversions, which are equivalent to TONs, are reported for 24 h reaction times. Error is ± 1 %. Catalyst % maleic acid % fu marie acid % malic acid None 95 3 2 HC1(1.2 mol L"1) 78 16 6 PdCl2(dpype) a , b' c 95 3 2 PdBr2(dpype)a'c 95 3 2 PtCl2(dpypcp)a 93 4 3 [Pt(dpype)2]Cl2c 92 4 4 [Pt(dpypcp)2]Cl2c 94 3 3 PdCl2(dmape) 92 6 2 PdCl2(dmape) + SnCl 2- 2H 2 O d 90 8 2 [PdCl(P,P ',/V~-dmapcp)]Cl 88 6 6 [PdI(P,P',7V-dmapcp)]I 93 4 3 [PdCl(P,#S-dmapmS]Cl 85 8 7 [PtCl(P,P ',7V-dmapcp)]Cl 91 4 5 [PtCl(P,P',7V-dmapcp)]Cl + HC1 74 20 6 (1.2 mol L"1) RuCl 2 (P,PW, AT-dmape) 95 3 2 Ie 13 87 0 IIf 28 71 1 a Not completely soluble at r.t. but dissolves at 100 °C. b Decomposes to metal during the course of the reaction. c Complex characterised spectroscopically but not yet isolated in analytically pure form (see Section 4.7.12 for the Pt complexes). d Pd:Sn =1:1. Mixture not homogeneous. eSee Section 5.7.1. f See Section 5.7.2. 222 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid 5.4 Discussion The only complexes whch showed even marginal activity for the hydration of maleic acid were [PdCl(P,P',/V-dmapcp)]Cl and [PdCl(P,/v;S-<imapmS)]Cl, both cationic complexes of Pd containing at least a P, //-bonded ligand. Although their activities are very low, they are superior to those of [(L)Pd(u-OH)] 2 2 + which, in the hydration of dem in THF/H2O, are 2.6 and 2.9 mol malic acid/mol catalyst for L = dppe and dcype, respectively, after 24 h at 120 °C, and 0 for both systems at 100 °C. 8 Itaconic acid (CH 2 =C(C0 2 H)CH 2 C0 2 H) was also tested as a substrate with [PdCl(P,P 7V-dmapcp)]Cl as catalyst, but no hydration was observed showing that- the internal position of the double bond is not the limiting factor in these reactions. Although the addition of SnCl 2 to chloro Pt"-phosphine complexes has been shown to increase the interaction between metal and olefin, 1 5 the PdCl2(dmape)/SnCl2 combination is no more effective a catalyst than the Pd complex alone, which itself shows no activity over that of the uncatalysed reaction. In addition, the \?XC\(P,P',N-dmapcp)]Cl/HCl combination does not catalyse the hydration any more effectively than HC1 alone, and also gives a comparable rate of isomerisation. By analogy to the reaction of K O H with PdCl2(dmape) in CH2CI2/H2O (Section 2.9.2.12), catalyst I probably contains the "PdCl(P,7V-dmapcpO)" fragment. Consistent with the ^P j 'H} N M R data, II is probably [Pd(CH3CN)(P,P',Ar-dmape)][N03]2. Although their identities remain uncertain, I and II have been included in Table 5.1 because they are both effective olefin isomerisation catalysts, implying that there is an interaction between these complexes and the olefin. Once again, the active species are (likely) cationic Pd complexes containing at least a P./V-bound ligand. 5.5 Conclusions Whereas a range of Ru, Pd and Pt pyridyl- and anilinyldiphosphine complexes are soluble in water, none of them is an effective catalyst for the hydration of maleic acid. The compounds which appear to hold the most promise are cationic Pd° species containing a P,/V-bonded ligand. 223 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid 5.6 Recommendations for Future Work The C2-symmetric complex resulting from the reaction of K O H with PtCl2(dmape) (Section 5.7.3) is probably the hydroxo-bridged Pt 2" complex [Pt((j,-OH)2(P,P'-2_j_ dmape)]2 , judging by the magnetic equivalence of the P-atoms and of the NMe groups; further work is necessary to confirm this, however. This compound reacts with maleic acid in acetone to form another species whose 3 1 P{'H} N M R spectrum shows 2 inequivalent P-atoms bound to Pt, possibly indicating P,P',jV-bound dmape (Section 5.7.3). The *H N M R spectrum consists of broad peaks, and it was not possible to determine whether maleate had been inserted into the Pt—OH bond or whether it had merely protonated the bridging ligand to form a compound like [Pt(OH2)(P,P',N-dmape)]2+. Elucidation of this reaction may give some insight into the possible steps involved in a catalytic hydration cycle by "M(P,P',7V)"+" complexes. In addition, the exact nature of the olefin isomerisation catalysts I and II needs to be determined. 5.7 Experimental 5.7.1 Preparation of catalyst I, and attempted catalytic hydration protocol To [PdCl(P,P',/V-dmapcp)]Cl (54 mg, 0.069 mmol) dissolved in CH 2 C1 2 (10 mL) was added an aqueous solution of K O H (10 mL, 0.3 mol L"1) under air. The two phase system was stirred for 25 min when the organic layer turned from yellow to orange. The H2O layer was removed and the CH2CI2 fraction was washed with successive portions of H2O ( 2 x 5 mL). The CH 2 C1 2 was removed in vacuo and the residue was dissolved in H20/EtOH (10:3 by vol.). The mixture was filtered through Celite 545 into a thick-walled glass bomb containing maleic acid (400 mg, 3.44 mmol) and a magnetic stir bar. The vessel was evacuated and filled with Ar three times before being placed on an oil-bath at 95 °C for 24 h. Analysis of the reaction solution was as outlined in Section 5.7.5. 5.7.2 Preparation of catalyst II To PdCl2(cod) (99 mg, 0.35 mmol) dissolved in C H 3 C N (10 mL) was added A g N 0 3 (120 mg, 0.71 mmol); a white precipitate formed immediately. The slurry was stirred for 0.5 h and then filtered through Celite 545. To the pale yellow filtrate was added dmape 224 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid (202 mg, 0.35 mmol). After 1 h, the volume of the solution was reduced in vacuo to ca. 1 mL and the product was afforded as a pale yellow powder by the addition of Et 2 0 (10 mL). Yield: 220 mg (73 % based on [Pd(CH3CN)(P )P',/V-dmape)][N03]2). 3 , P{ 'H} N M R (81 MHz, CD 3 OD, 300 K): S 22.4 (d, 2JP? = 4.3), 24.9 (d, V P P = 4.3). 5.7.3 Reaction of PtCI2(dmape) with KOH followed by maleic acid To PtCLXdmape) (22 mg, 0.026 mmol) dissolved in CH2CI2 (10 mL) was added an aqueous K O H solution (10 mL, 1.0 mol L"1) under air. The two phase system was stirred overnight when the CH2CI2 layer became pale yellow. The H 2 0 was removed and the organic fraction was washed with successive portions of H 2 0 ( 3 x 1 0 mL) whereupon it was reduced to dryness in vacuo. The residue was taken up in CDC1 3 and analysed by N M R spectroscopy. ] H N M R (200 MHz, CDC1 3, 300 K): 8 2.30 (s, 24H, NGtf 3), 2.90 (m, 4H, CH2), 7.0-8.1 (m, 16H, Ar). 3 1 P{'H} N M R (81 MHz, CDC1 3 , 300 K): 5 31.0 (s, 1 J P P t = 3680). The product from the previous reaction was dissolved in acetone (3 mL) and maleic acid (4 mg, 0.034 mmol) was added, causing the solution to change immediately from yellow to colourless. The solution was stirred overnight and then reduced to dryness in vacuo. The residue was dissolved in CDC1 3 and analysed by N M R spectroscopy. ] H N M R (200 MHz, CDC1 3 , 300 K): 8 2.0-3.6 (br, 28H, NC7f 3 and CH2), 6.8-8.2 (br, 16H, Ar). A singlet at 8 6.1 due to maleic acid was also apparent. ^P j 'H} N M R (121 MHz, CDC1 3, 300 K): 8 35.7 (d, 2 J P P = 14, lJPPi = 3620), 47.0 (d, 2JPP = 14, }JPn = 3420). 5.7.4 General protocol for the catalytic hydration of maleic acid To a combination of maleic acid (116 mg, 1.00 mmol) and the appropriate metal complex (0.01 mmol) in a thick-walled glass bomb charged with a magnetic stir bar was added H2O (10 mL). The solution was degassed by shaking the vessel while a vacuum was applied. Once the degassing was complete, Ar (1 atm) was admitted. The glass bomb was immersed in a thermostatted oil-bath at 100 °C. After 24 h, a 3 mL aliquot was withdrawn. This was reduced to dryness on a rotatory evaporator and the residue was ground to a fine powder using a pestle and mortar. The powder (7 mg) was dissolved in acetone-d6 (0.5 mL) and analysed by : H N M R spectroscopy according to Section 5.7.5. 225 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid 5.7.5 Determination of extent of hydration from *H NMR spectra The proportions of maleic, fumaric and malic acids present in an acetone-d6 solution containing a mixture of the three were calculated using the ratios of the appropriate ' H N M R peak integrations. Shown in Figure 5.1 is a representative spectrum. The following are the ' H N M R (300 MHz, acetone-d6, 300 K) data for the 3 components.7 (Assignments for malic acid are given according to the Fischer projection in Figure 5.1) Maleic acid: 5 6.40 (s, 2H, cw-HOOCC#=C#COOH). Fumaric acid: 5 6.80 (s, 2H, fra/w-HOOCC/f=€i/COOH). Malic acid: 5 4.52 (dd, 1H, HOOCC// f l (OH)CH b H c COOH, V H a H b = 7.2, V H a H c = 4.5), 2.75 (m, 2H, HOOCC//„(OH)CH bH cCOOH, X H b = 7.2, V H a H c = 4.5, V H b H c = 15.9). COOI-I I 11 I I I I I I I I I I I I I I I I I I I 6.8 6.4 Figure 5.1 A sample ' H N M R (300 MHz, 300 K) spectrum of an acetone-d6 solution containing maleic, fumaric and malic acids, fa = fumaric acid, ma = maleic acid. H a and Hb,c refer to the protons given in the Fischer projection of malic acid. The area under the peaks due to fumaric and maleic acids, and the H a and HbjC protons of malic acid are denoted Af a, A m a , A H 3 and AHb,o respectively. As H a Hb, H c • 11 i 11 i 11 i 111 11 i 11 i | 11 i 11 i 11 1 1 11 4,6 4.2 3.0 2.5 226 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid corresponds to 1 proton and H B > C to 2, the total peak area due to the three acids is given either by: ATOT= A f a + A™ + 2 A H a , or AxOT = Afa + A m a + A H a ,b In practice, the former of the above equations was used in subsequent calculations because the H A peak always fell in a region of the spectrum which was uncomplicated by peaks due to the catalyst. The percentages of each of the acids in solution are given by: % fumaric acid = A f a / ATOT x 100 % maleic acid = A m a / A T O T x 100 % malic acid = 2 A H a / ATOT x 100 By comparison of the relative concentrations (calculated by ' i i N M R peak ratios) to the actual relative concentrations for acetone-d6 solutions containing known masses of the three acids, the error in the technique was found by Xie to be ± 1 %. 7 A l l results were compared to those obtained for "blank" reactions, i.e., those run under identical conditions in the absence of metal complex. 5.8 References 1. Haggin, J. Chem. Eng. News, May 31 1993; p 23. 2. Louie, J.; James, B. R. Inorg. Chim. Acta 1969, 3, 568. 3. Bzhasso, N . A. ; Pyatnitskii, M . P. Izv. Vyssh. Ucheb. Zaved., Pishch. Tehch. Tekhol. 1967, 5, 207; through Chem. Abs. 68:39013k. 4. Bzhasso, N . A. ; Pyatnitskii, M . P. Zh. Prikl. Zhim. 1969, 42, 1610; through Chem. Abs. 71:101239f. 5. Jensen, C. M . ; Trogler, W. C. Science 1986, 233, 1069. 6. Ramprasad, D.; Yue, H. J.; Marsella, J. A. Inorg. Chem. 1988, 27, 3151. 7. Xie, Y . Towards Transition Metal Catalyzed Hydration of Olefins; Aquo Ions and Pyridylphosphine-Platinum and Palladium Complexes; Ph.D. Dissertation, University of British Columbia: Vancouver, 1990. 8. Ganguly, S.; Roundhill, D. M . Organometallics 1993,12, 4825. . 9. Bennett, M . A. ; Jin, H . ; Lin, S.; Redina, L. M . ; Willis, A . J. Am. Chem. Soc. 1995, 117, 8335. 10. Kawatsura, M . ; Hartwig, J. F. J. Am. Chem. Soc. 2000,122, 9546. 227 References on page 227 Chapter 5: Attempted Catalytic Hydration of Maleic Acid 11. Kawatsura, M . ; Hartwig, J. F. Organometallics 2001, 20, 1960. 12. Kawatsura, M ; Hartwig, J. F. J. Am. Chem. Soc. 2000,122, 9546. 13. Lober, O.; Kawatsura, M . ; Hartwig, J. F. J. Am. Chem. Soc. 2001,123, 4366. 14. Koch, H . F.; Girard, L. A. ; Roundhill, D. M . Polyhedron 1999,18, 2275. 15. Cramer, R. D.; Jenner, E. L.; R. V . Lindsey, J.; Stolberg, U . G. J. Am. Chem. Soc. 1963,55, 1691. 228 Appendix 1: General Experimental Procedures Al General Experimental Protocols A l . l General Procedures A l l procedures were conducted using standard Schlenk techniques under an Ar or N 2 atmosphere unless otherwise noted. In some cases, an Ar-filled glovebox was used to handle particularly 0 2 - or H20-sensitive materials. A l l reactions were performed at r.t. (ca. 20-25 °C) unless otherwise specified. A1.2 Instrumentation Al.2.1 Nuclear magnetic resonance (NMR) spectroscopy Six different N M R spectrometers were used in the course of this work. Spectrometer frequencies for ' H , 3 1 P and 1 3 C for each instrument are given in Table A l . l . Residual solvent proton ( 'H, relative to external SiMe 4 5 0.00), external P(OMe) 3 ( 3 1 P{'H}, 5 141.00 vs. external 85 % aq. H3PO4), or solvent carbon ( 1 3 C, 5 77.0 relative to external SiMe4) were used as the reference. Downfield shifts were taken as positive. A l l ./-values are given in Hz; s = singlet, d = doublet, t = triplet, m = multiplet, br = broad, p = pseudo. Table A l . l Spectrometer frequencies for ' H , 3 1 P and 1 3 C for each of the N M R instruments used in the course of this work. Spectrometer >H Spectrometer Frequency (MHz) 3 1 p , 3 C Bruker AC200 200.13 81.02 50.32 Bruker AV300 300.13 121.49 75.46 Bruker AV400 400.13 161.98 100.61 Bruker AM400 400.13 161.98 100.61 Bruker AMX500 500.14 202.47 125.76 Varian XL300 299.94 121.42 75.43 229 References on page 233 Appendix 1: General Experimental Procedures For variable temperature N M R experiments, the probe temperature was calibrated using MeOH. 1 Some N M R experiments were conducted by Ms. Liane Darge, Ms. Marietta Austria, or Dr. Nick Burlinson, all of the UBC Chemistry Department. Al.2.2 X-ray crystallography X-ray crystallographic analyses were performed either by the late Dr. Steven Rettig or Dr. Brian Patrick both of the U B C Chemistry Department, or by Dr. Victor Young of the University of Minnesota. Al.2.3 Elemental analysis Elemental analyses were conducted using a Carlo Erba 1108 analyser by Mr. Peter Borda of the U B C Chemistry Department. Al.2.4 Ultraviolet-visible (UV-vis) spectroscopy UV-vis spectra of coloured complexes were recorded on a Hewlett Packard 8452A diode array spectrophotometer and are reported as A,max (± 2 nm) [s (L mol"1 cm"1)]; sh = shoulder. Al.2.5 Infra-red (IR) spectroscopy IR spectra were recorded on either an ATI Mattson Genesis series or a Bomem-Michelson MB-100 FTJJR. spectrometer, scanning 500-4000 cm"1. Samples were prepared as KBr discs. IR data are reported as v (± 4 cm"1) (relative intensity); s = strong, m = medium, w = weak. Al.2.6 Conductivity Conductivity measurements were made on approximately 10"3 mol L" 1 solutions using a Serfass conductance bridge model RCM15B1 (Arthur H . Thomas Co. Ltd.) connected to a 3404 cell (Yellow Springs Instrument Co.) and are reported as AM (± 0.5 ohm"1 mol"1 cm2). 230 References on page 233 Appendix 1: General Experimental Procedures Al.2.7 Gas chromatography (GC) A Hewlett Packard 5890A gas chromatograph equipped with a 25 m-0.32 mm HP17 column and an H2/air flame ionisation detector (FED) was used. He was used as the carrier gas. A1.3 Materials Al.3.1 Gases Gases were purchased from commercial sources and used without purification, except Ar (H.P.) and N 2 (U.S.P.) which were dried by passing through P2O5 and anhydrous CaS04, respectively. Al.3.2 Solvents Reagent grade solvents (Fisher Scientific) were either distilled from CaH 2 (CH2C1 2, CH3NO2), Na (Et 20, hexanes), Mg/I 2 (EtOH), or anhydrous K 2 C 0 3 (acetone) under N 2 . PhCN was stored over CaS04. Other solvents were used as supplied. A l l deuterated solvents were purchased from Cambridge Isotope Laboratories. CDCI3, and C$D(, were dried over activated molecular sieves (Fisher: Type 4 A, 4-8 mesh), deoxygenated, and stored under Ar. CD2CI2 was stored under vacuum over CaH 2 . Other deuterated solvents (CD3OD, D2O, acetone-d6, DIVISOR) were used as supplied. Al.3.3 Reagents Unless otherwise noted, reagents were purchased from commercial sources.Nand used without purification. Al.3.3.1 Metal complexes Metal complex precursors were synthesised according to the literature procedures given in Table A1.2. N M R spectroscopic data for the complexes matched those given in the literature in all cases. RuCl3-xH 20 was obtained on loan from Johnson Matthey, Ltd. and Colonial Metals, Inc. 231 References on page 233 Appendix 1: General Experimental Procedures Table A1.2 Procedures used to synthesise metal complex precursors. Complex Reaction Reference PtCl2(cod) K 2 P t C l 4 + cod -» PtCl2(cod) + 2 KC1 2 Pt(PPh3)4 K 2 P t C l 4 + 4 PPh 3 + 2 K O H + EtOH -> Pt(PPh 3) 4 + 3 4 KC1 + H 2 0 + C H 3 C H O Pt(PPh3)3 Pt(PPh 3) 4 -> Pt(PPh3)3 + PPh 3 3 ;ra«s-PdCl 2(PhCN) 2 PdCl 2 + 2 PhCN -> PdCl 2(PhCN) 2 4 PdCl2(cod) H 2 P d C l 4 + cod -> PdCl2(cod) + 2 HC1 2 PdCl2(nbd) PdCl 2(PhCN) 2 + nbd -» PdCl2(nbd) + 2 PhCN 5 PdCl(Me)(cod) PdCl2(cod) + SnMe 4 -» PdCl(Me)(cod) + 6 SnCl(Me) 3 [Pd(u-Cl)(MeO- 2 PdCl2(cod) + 2 NaOMe -> [Pd(u-Cl)(MeO- 7 cod)]2 cod)]2 + 2 NaCl Pd(PPh3)4 2 PdCl 2 + N 2 H 4 + 8 PPh 3 -» 2 Pd(PPh 3) 4 + 4 HC1 8 + N 2 Pd2(dba)3-CHC13 2 PdCl 2 + 3 dba + 2 MeOH + 4 NaOAc -» 9 Pd2(dba)3 + 2 H 2 C O + 4 NaCl + 4 HO Ac RuCl 2(PPh 3) 3 2 RuCl 3 + 7 PPh 3 + H 2 0 -> 2 RuCl 2(PPh 3) 3 + 10 OPPh 3 + 2 HC1 [Rh(u-Cl)(CO)2]2 2 RhCl 3 + 6 CO -> [Rh(u-Cl)(CO)2]2 + 2 C O C l 2 11 [Rh(u-Cl)(cod)]2 2 RhCl 3 + 2 EtOH + 2 cod -> [Rh(u-Cl)(cod)]2 + 2 12,13 C H 3 C H O + 4 HC1 Al.3.3.2 Organic compounds On one occasion, 2-bromopyridine was stirred with NaOH pellets overnight and distilled from CaO to remove H 2 0 and then degassed by 3 freeze-pump-thaw cycles to remove 0 2 , but this was found not to enhance significantly the yields of the pyridyldiphosphine ligand syntheses. The ligand dpype was made according to the procedure by Baird et al.u The precursor to the cyclopentane-bridged diphosphine ligands, 1,2-bis(dichlorophosphino)cyclopentane, was made by a literature procedure15 and kindly donated by previous researchers in this group Drs. Richard Schutte and Kenneth MacFarlane. The physical state of samples of 2-bromoaniline purchased from Aldrich was variable. This compound is a white crystalline solid at r.t., but on two occasions was delivered as a red viscous liquid; the pure compound was separated by sublimation. N,N-232 References on page 233 Appendix 1: General Experimental Procedures dimethyl-2-bromoaniline was made via methylation of the pure compound with dimethylsulphate according to a literature procedure.16 Dimethyl- and diethylmaleate and dimethylfumarate were degassed by 3 freeze-pump-thaw cycles prior to use. A1.4 References 1. Braun, S.; Kalinowski, H.-O.; Berger, S. 100 and More Basic NMR Experiments, 1st ed . ;VCH: Weinheim, 1996; p. 112. 2. McDermott, J. X . ; White, J. F.; Whitesides, G. M . J. Am. Chem. Soc. 1976, 98, 6521. 3. Ugo, R.; Cariati, F.; Monica, G. L. Inorg. Synth. 1968,11, 105. 4. Hartley, F. R. Organometallic Rev. (A) 1976, 6, 119. 5. Abel, E. W.; Bennett, M . A. ; Wilkinson, G. J. Chem. Soc. 1959, 3178. 6. Rulke, R. E.; Ernsting, J. M . ; Spek, A . L.; Elsevier, C. J.; van Leeuwen, P. W. N . M . ; Vrieze, K . Inorg. Chem. 1993, 32, 5769. 7. Bailey, C. T.; Lisensky, G. C. J. Chem. Educ. 1985, 62, 896. 8. Coulson, D. R. Inorg. Synth. 1972,13, 121. 9. Ukai, T.; Kawazura, H. ; Ishii, Y . ; Bonnett, J. J.; Ibers, J. A . J. Organomet. Chem. 1974, 65, 253. 10. Hallman, P. S.; Stephenson, T. A. ; Wilkinson, G. Inorg. Synth. 1970,12, 237. 11. McCleverty, J. A. ; Wilkinson, G. Inorg. Synth. 1966, 8, 211. 12. Chatt, J.; Venanzi, L. M . Chem. Soc. 1957, 4735. 13. Giordano, G.; Crabtree, R. H. Inorg. Synth. 1990, 28, 88. 14. Baird, I.