Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Reversible binding of dihydrogen to dinuclear ruthenium complexes containing chelating diphosphines Chau, Daniel Elliot Kwok-Yue 1992

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

Item Metadata

Download

Media
831-ubc_1993_spring_chau_daniel.pdf [ 5.42MB ]
Metadata
JSON: 831-1.0061778.json
JSON-LD: 831-1.0061778-ld.json
RDF/XML (Pretty): 831-1.0061778-rdf.xml
RDF/JSON: 831-1.0061778-rdf.json
Turtle: 831-1.0061778-turtle.txt
N-Triples: 831-1.0061778-rdf-ntriples.txt
Original Record: 831-1.0061778-source.json
Full Text
831-1.0061778-fulltext.txt
Citation
831-1.0061778.ris

Full Text

REVERSIBLE BINDING OF DIHYDROGEN TO DINUCLEAR RUTHENIUM COMPLEXES CONTAINING CHELATING DIPHOSPHINES BY DANIEL ELLIOT KWOK-YUE CHAU  B. Sc., McMaster University, 1990  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CHEMISTRY  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF/ BRITISH COLUMBIA December, 1992 © Daniel Elliot Kwok-Yue Chau 1992  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of  Chemistry  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  \P9-1•1 (A Pi-PT^, f 9 9.  ABSTRACT The preparation of dinuclear mixed-valence ruthenium complexes of general formula Ru2C15(P-P)2, previously reported for a range of chelating ditertiary phosphines including chiral systems, has been extended to include compounds of the diphosphine 1,4bis(dicyclohexylphosphino)butane (P-P = DCYPB). The complexes are generally prepared by the reaction of RuC13(PR3)2(DMA)•DMA, R = phenyl or p-tolyl, DMA = N,N-dimethylacetamide, with one equilvalent of the appropriate diphosphine. The H2reduction of Ru2C15(P-P)2 complexes in DMA, or in toluene in the presence of an added base, generally gives the dinuclear Ru(H,II) complexes [RuCl(P-P)(11-C1)12, but the DCYPB species was not formed; the procedures with DCYPB (including studies in CH2C12) led to poorly characterised complexes which may be polynuclear polyhydrides and appear to contain an r1 2 -H2 ligand. 1 H and 31 P{ 1 H}NMR spectroscopy was used extensively to study the systems. The previously reported reaction of the dinuclear ruthenium(H) complex, [RuCl(DPPB)(t-C1)]2 (DPPB = 1,4-bis(diphenylphosphino)butane), with 1 atm H2 at ambient temperature to give the molecular hydrogen complex (71 2 -H2)Ru(DPPB)(11C1)3RuCl(DPPB), has been extend to the 1,3-bis(diphenylphosphino)propane (DPPP) system; a corresponding ti 2 -H2 species with the DCYPB ligand has been detected following reaction of Ru2C15(DCYPB)2 with H2. The i1 2 -H2 ligands in the DPPP and DCYPB systems have internuclear distances of 0.86  A as estimated by 1 H NMR variable  temperature spin-lattice relaxation data (T 1 (min) = 12 ms at 300 MHz). With the DPPP system, a Vim of 29.4 Hz was found for the i1 2 -HDisotopomer, and the i 2 -H2 ligand is replaceable by N2. The reversible equilibrium conversion under H2 of [RuCl(DPPB)2(.t-Cl)]2, 9, to (T1 2 -H2)Ru(DPPB)(11.-C1)3RuC1(DPPB), 9a , in CH2C12 has been studied at 0-25°C by ll  UV-vis and 31 13 ( 1 H)NMR spectroscopy, and by stopped-flow kinetics; the equilibrium constant and the corresponding thermodynamic parameters, and the forward and reverse rate constants have been determined. The equilibrium is close to being thermally neutral (AH° – zero), while a AS° value of –60 J/mol K indicates the involvement of solvent molecules. A rough estimate of -60 kJ/mol is made for AH° of the process Run + H2 RuII(1 2 -H2). Preliminary studies were made on the reaction of [RuCl(DPPB)2(11.-C1)]2, 9, with 02, which may give a peroxide, and the subsequent reaction of this product with H2; the latter reaction results in the formation of water and regeneration of a mixture of 9 and 9a.  iii  TABLE OF CONTENTS Page  ABSTRACT^ TABLE OF CONTENTS ^  iv  LIST OF TABLES ^  viii  LIST OF FIGURES ^  ix  LIST OF ABBREVIATIONS ^  xii  ACKNOWLEDGMENTS ^  xv  CHAPTER 1  Introduction ^  1  General Introduction ^ 1 Homogeneous Hydrogenation ^ 2 Dihydrogen Activation ^ 6 Molecular Hydrogen Complexes: a Brief Review ^ 9 1.4.1^Discovery of Molecular Hydrogen Complexes ^ 9 1.4.2^Further Examples of Molecular Hydrogen Complexes ^ 10 1.4.3^Properties of Molecular Hydrogen Complexes ^ 13 1.4.4^Bonding and Stability of Molecular Hydrogen Complexes ^ 14 1.4.5^Methods Used in Recognition of Molecular Hydrogen Complexes ^ 17 1.5 Scope of This Thesis^ 22 1.1 1.2 1.3 1.4  CHAPTER 2  2.1  Experimental Procedures ^  23 23 23 23 24  Materials ^ 2.1.1^Solvents ^ 2.1.2^Gases ^ 2.1.3^Phosphines ^ 2.1.3.1^Modified Preparation of 1,4bis(dicyclohexylphosphino)butane, DCYPB ^ 24 2.1.4^Other Materials ^ 25 iv  2.2 Instrumentation ^ 25 2.3 Synthesis and Characterisation of Ruthenium Complexes ^ 27 2.3.1 Ruthenium Precursors, RuC13(PAr3)2(DMA)•DMA ^ 28 2.3.1.1 Trichlorobis(triphenylphosphine)(DMA)ruthenium (III)-DMA solvate, RuC13(PPh3)2(DMA)•DMA ^ 28 2.3.1.2 Trichlorobis(tri-p-tolylphosphine)(DMA)ruthenium(III).DMA solvate, RuC13(P(p-toly1)3)2(DMA)•DMA ^ 28 2.3.2 Dichloro-trilt-chloro-bis(bidentate phosphine)diruthenium (II, III) Complexes, [(P-P)C1Ru(4-C1)3RuC1(P-P)] or Ru 2 C1 5 (P-P) 2 ^ 29 2.3.2.1 P-P = DPPP: Ru2C15(DPPP)2, 1 ^ 29 2.3.2.2 P-P = DPPB: Ru2C15(DPPB)2, 2 ^ 29 2.3.2.3 P-P = DCYPB: [Ru2C15(DCYPB)2•H2O] n, 3 ^ 30 2.3.2.4 P-P = DPPPt: Ru2C15(DPPPt)2, 5^ 30 2.3.2.5 P-P = DPPH: Ru2C15(DPPH)2, 6 ^ 30 2.3.3 Dichloro-dilt-chloro-bis(bidentate phosphine)diruthenium(II) Complexes, [RuCl(P-P)(11-C1)12 ^ 31 2.3.3.1 P-P = DPPP: [RuC1(DPPP)(1.-C1)12, 7^ 32 2.3.3.2 P-P = DPPB: [RuCl(DPPB)(11- C1 )12, 9 ^ 32 2.3.3.3 P-P = DPPPt: Attempted Synthesis of [RuCl(DPPPt) (p-C1)}2, 10 ^ 33 2.3.4 P-P = DCYPB: Attempted Synthesis of [RuCl(DCYPB) 33 (11-CI)]2 ^ CHAPTER 3^Synthesis and Characterisation of Chloro-Bridged Diruthenium Complexes Containing Chelating Ditertiary Phosphines and their Role in Activation of Hydrogen ^ 35 3.1 Introduction - Review of [RuC1(P-P)(11-C1)]2 Complexes ^ 35 3.2 Present Work ^ 41 3.3 Ru2C15(P-P)2 Complexes ^ 41 3.4 [RuCl(P-P)(g-C1)]2 Complexes ^ 43 3.4.1 [RuCl(DPPP)(1.1-C1)]2•H20^ 44  3.4.1.1 Molecular Structure of the Ru3C17(PPh3)3(DPPP)(H20) Cluster, 8 ^ 46 3.4.2^[RuCl(DPPPt)(1-C1)12 ^ 49 3.5 Interaction of [RuCl(DPPP)(A-C1)]2, 7, with H2 and N2 ^ 50 3.5.1^Formation of a Molecular Hydrogen Complex by Interaction of 7 with H2 ^ 50 3.5.2^Formation of a Dinitrogen Complex by Interaction of 7 with N2 ^ 57 3.6 Interaction of Ru2C15(DCYPB)2, 3, with H2 ^ 61 3.6.1 Reaction of Ru2C15(DCYPB)2, 3, with H2 in DMA ^ 61 3.6.1 Reaction of Ru2C15(DCYPB)2, 3, with H2 in CD2C12 ^ 63 CHAPTER 4  Thermodynamic and Kinetic Studies of the Interaction of [RuCI(DPPB)(p.-C1)]2 with H2 ^ 70 4.1 Introduction ^ 70 4.2 Solubility of Hydrogen in CD2C12(CH2C12) ^ 71 4.3 Determination of the Equilibrium Constant by UV-Visible Spectroscopy ^ 72 4.4 Determination of the Equilibrium Constant by NMR Spectroscopy ^ 74 4.4.1 31 P NMR T 1 Measurements ^ 75 4.4.2 Thermodynamic Parameters ^ 77 4.5 Stopped-Flow Experiments- Determination of k1 and k..1 ^ 79 4.5.1^Sample Handling ^ 79 4.5.2 Data Treatment and Results ^ 80 4.5.2.1 Ratio of 9 to H2 = 1.0 : 1.3 ^ 80 4.5.2.2 Ratio of 9 to H2 = 2.2 : 1.0 ^ 86 4.5.3^Analysis of the Stopped-Flow Data ^ 89  CHAPTER 5 Reaction of [RuCI(DPPB)(11-C1)}2 with 02 ^ 91 5.1 Introduction ^ 91 5.2 Interaction of [RuCl(DPPB)(,t-C1)]2 with 02 and subsequent reaction of the product with H2 ^ 91  vi  CHAPTER 6  CHAPTER 7  General Conclusions and Recommendations for Future Work ^  96  References and Footnotes ^  98  APPENDIX  109 A-1 X-ray Crystallographic Analysis of the Ru3a7(PPh3)3(DPPP)(H20) Cluster ^ A-2 Computer Programs in BASIC used in Section 4.5 ^  110 122  A-3 Derivation for Equation 4.22 ^  128  vii  ^  LIST OF TABLES Table^  Title^  1.1^Examples of molecular hydrogen complexes ^ 1.2^Correlation of dihydrogen versus dihydride complex formation with VTJrJ of analogous dinitrogen complexes M(N2)L5, and Vp for some dihydrogen complexes M(H2)L5 ^  Page  12  17  1.3^Selected 1 H NMR HD coupling constants (Hz) and T 1(um) values for some molecular hydrogen and hydride complexes ^ 19 3.1^Selected bond lengths (A) for (Ru3C17(PPh3)3(DPPP)(H20)], 8, with estimated standard deviations in parentheses. ^ 47 3.2^Selected bond angles (°) for [Ru3C17(PPh3)3(DPPP)(H20)], 8, with estimated standard deviations in parentheses ^ 48 3.3^31P(1H) NMR data (121.42 MHz, 20°C) for the complexes [RuCl(P-P)(tCl)]2 (P-P = DPPB, 9; DPPP, 7), and [(L)Ru(P-P)(.t-C1)3RuCl(P-P)] (L H2, a; cr-N2, b; P-P = DPPB, DPPP) ^ 54 3.4^Temperature dependence of the 1H NMR T1 relaxation time for the 01 2-H2) moiety in (1)2-H2)Ru(DPPP)(p-C1)3RuCl(DPPP), 7a, at 8 -10.90 ppm (300 MHz, CD2C12) ^ 55 3.5^Temperature dependence of the 1H NMR T1 relaxation time for the (1) 2-H2) moiety in complex 13, at 8 -12.05 ppm (300 MHz, CD2C12). ^ 66 3.6^1H and 31 P( 1 H) NMR data (CD2C12, 20°C) for the Complexes 13 and 13a ^ 69 4.1^Summary of Ti values for the phosphorus nuclei in [RuCl(DPPB)(1.1.-C1)]2, 9, and [(i 2-H2)Ru(DPPB)(4-C1)3RuCl(DPPB)], 9a (121.42 MHz, 25°C) ^ 75 4.2 Summary of K values determined at various temperatures for the conversion of [RuCl(DPPB)(1-C1)12, 9, with H2 to [(r1 2-H2)Ru(DPPB)(p-C1)3RuCl(DPPB)], 9a, in CD2C12. 77 4.3^Summary of data to determine k1 from Equation 4.21 ^ 86  viii  ^  LIST OF FIGURES Figure  ^  Tide^  Page  1.1^Some chiral diphosphines used in catalytic asymmetric synthesis ^ 5 1.2^Proposed transition states during a) homolytic and b) heterolytic activation of dihydrogen by transition metal complexes ^ 7 1.3^The three early examples of molecular hydrogen complexes ^ 10 1.4^Bonding scheme of M-(1 2 -H2) ^ 14 2.1^UV-Visible anaerobic spectral cell ^ 26 3.1^Diphosphines used successfully in the preparation of Ru2C15(P-P)2 complexes ^ 38 3.2^Suggested geometry for [RuCl(P-P)(11-C1))2 complexes with a) P-P = DPPP, DPPB, DPPPt, DIOP, CHIRAPHOS, S,S,-BDPP; and b) P-P = BINAP ^ 39 3.3^31 P{ 1 H) NMR spectrum of (RuCl(DPPP)(1-C1)12, 7, at a) 20°C; b) -80°C and c) -90°C (121.42 MHz, CD2C12) ^ 45 3.4^Molecular structure of [Ru3C17(PPh3)3(DPPP)(H20)], 8, with the atom numbering scheme used^ 47 3.5^1 H NMR spectrum of [(11 2 -H2)(DPPP)Ru(.1-C1)3RuCl(DPPP)), 7a, formed in situ by reacting [RuCl(DPPP)(.t-Cl)]2, 7, under 1 atm H2 pressure (300 MHz, CD2C12, 20°C) ^ 51 3.6^31 P{ 1 H) NMR spectrum of [(11 2 -H2)(DPPP)Ru(p.-C1)3RuC1(DPPP)], 7a, formed in situ by reacting [RuCI(DPPP)(11-C1)12, 7, under 1 atm H2 pressure (121.42 MHz, CD2C12, 20°C) ^ 52 3.7^Plot of temperature dependence of Ti for the molecular hydrogen moiety in (11 2 H2)Ru(DPPP)([1.-C1)3RuCl(DPPP), 7a. ^ 55 3.8^1 H NMR spectrum of [(1 2 -HD)(DPPP)Ru(p.-C1)3RuCl(DPPP)], formed in situ by stirring [RuCl(DPPP)(p-C1)]2, 7, under —1.1 atm of H2 and —1.5 atm of D2 (300 MHz, CD2C12, 20°C). ^ 56 3.9^Suggested intermediate for 7-catalysed H2/D2 ^HD isotope exchange^ 57 3.10 1 H NMR spectrum of the dinitrogen complex 7b formed in situ from [RuCl(DPPP)(p-Cl)]2, 7, under —1 atm N2 pressure (300 MHz, CD2C12, 20°C) 58 3.11 31 P( 1 H)NMR spectrum of the dinitrogen complex 7b formed in situ from [RuCl(DPPP)(1.1-C1)]2, 7, under —1 atm N2 pressure (121.42 MHz, CD2C12, 20°C) ^ 59 ix  ^  3.12 Infra-red spectrum of a CH2C12 solution of [RuCl(DPPP)(11-C1)]2, 7, saturated with N2 (solvent spectrum subtracted). ^ 60 3.13 1 H NMR spectrum of [(r1 2 -H2)(DCYPB)Ru(A-C1)3RuCl(DCYPB)], 13, under vacuum (300 MHz, CD2C12, 20°C) ^ 64 3.14 31 P{ 1 H) NMR spectrum of [(r1 2 -H2)(DCYPB)Ru(g-C1)3RuC1(DCYPB)], 13, under vacuum (121.42 MHz, CD2C12, 20°C) ^ 65 3.15 Plot of temperature dependence of Ti for the molecular hydrogen moiety of 13 at 8 -12.05 ppm ^ 67 3.16 1 H NMR spectrum of 13a, formed from the reaction of [(r1 2 -H2)Ru(DCYPB)(1^1C1)3RuCl(DCYPB)], 13, with 1 atm H2 (300 MHz, CD2C12, 20°C) ^ 68 3.17 31 P{ 1 H) NMR spectrum of 13a, formed from the reaction of [(1) 2 H2)Ru(DCYPB)(1-C1)3RuCl(DCYPB)], 13, with 1 atm H2 (121.42 MHz, CD2C12, 20°C) ^ 68 3.18 Suggested geometries for species 13a ^ 69 4.1^Plot of dissolved H2 concentration in CD2C12 versus H2 pressure at 25°C ^ 72 4.2^(a) Changes in the UV-visible spectrum with increase in concentration of H2 for a CH2C12 solution of [RuCl(DPPB)2(p-Cl)]2, 9; [Ru2] = (5.4 ± 0.3) x 10 -5 M. c' = absorbance / total [Ru2]. (b) Plot of changes in absorbance at X = 320 nm versus H2 concentration; data taken from (a). The residual plot indicates the difference between the observed value in absorbance and the calculated value using the rectangular hyperbola fit. Temp = 25°C ^ 73 4.3^31 P ( 1 H) NMR stacked plots for the measurements of T1 relaxation times for the resonances of [RuCl(DPPB)(11-C1)12, 9, and K1 2 -H2)(DPPB)Ru(µC1)3RuC1(DPPB)1, 9a (121.42 MHz, CD2C12, 25°C) ^ 76 4.4^A fitted stopped flow trace for the change in absorbance over 200 s, at X = 481 nm, for the reaction (at a 1.0 : 1.3 ratio) between [RuCl(DPPB)(.t-Cl)]2, 9 ((4.5 ± 0.5) x 10 -4 M) and H2 ((5.9 ± 0.1) x 10 -4 M) in CH2C12, temp = 25°C; the concentrations given are for the mixed solutions. The residual plot shows the difference between observed and fitted data. ^ 82 4.5^Plots of appearance of [(11 2 -H2)Ru(DPPB)(1.t-C1)3RuCI(DPPB)], 9a, versus time using data collected from Figure 4.4 and analysis using Equations 4.21 and 4.22. [91 = (4.5 ± 0.5) x 10 -4 M; [H2] = (5.9 ± 0.1) x 10 -4 M; temp = 25°C. (a) K = (1.3 ± 0.1) x 10 3 M -1 and (b) K = (7.5 ± 0.7) x 10 2 M-1 ^ 85  ^ ^  4.6^Fitted stopped flow traces of the changes in absorbance at X = 481 nm, for a 2.2:1.0 reaction ratio between [RuCl(DPPB)(11-C1)]2, 9 ((1.3 ± 0.1) x 10-3 M) and H2 ((5.9 ± 0.1) x 10 -4 M) in CH2C12, temp = 25°C. (a) raw data before subtracting the blank and (b) after background correction. ^ 87 4.7^Plots of appearance of [(r1 2-H2)Ru(DPPB)(11,-C1)3RuC1(DPPB)], 9a, versus time using data collected from Figure 4.6b and analysis using Equations 4.21 and 4.22. [9] = (1.3 ± 0.1) x 10-3 M; [1-12] = (5.9 ± 0.1) x 10-4 M; temp = 25°C. (a) K = (1.3 ± 0.1) x 103 M-1 and (b) K = (7.5 ± 0.7) x 102 M-1 ^ 88 5.1^IR spectra (Nujol mull) of (a) the black solid isolated from the reaction of [RuCl(DPPB)2(g-C1)}2, 9, with 02 and (b) pure 9 ^ 93 5.2 Visible spectral changes of oxygen-reacted [RuC1(DPPB)20-i-CIA2, 9 in CH2C12 on exposure to 1 atm H2. [9] = 2.7 x 10-3 M, T = 25°C; a = initial spectrum, b = after 10 min, c = after 15 min, d = after 20 min, e = after 30 min 94  xi  LIST OF ABBREVIATIONS The following list of abbreviations, most of which are commonly used in the chemical literature, will be employed in this thesis: At^absorbance at time t (UV-Vis) atm^atmosphere; 1 atm = 760 mm Hg (S,S)-BDPP^(2S,4S)-bis(diphenylphosphino)pentane; also known as SKEWPHOS BINAP^(R)- or (S)-2,2'-bis(diphenylphosphino)-1,1'-binaphthyl BIPHEMP^2,2'-dimethy1-6,6'-bis(diphenylphosphino)biphenyl BFFPA^1,1'-bis(diphenylphosphino)-2'(1-N,N-adimethylaminoethyl)ferrocene br^broad Bu^butyl, CH2(CH2)2CH3 cat.^catalyst S,S-CHIRAPHOS (2S,3S)-bis(diphenylphosphino)butane Cp^cyclopentadienyl, T1 5 -05H5 Cy^cyclohexyl COSY^homonuclear correlation spectroscopy d^doublet (NMR) DCYPB^1,4-bis(clicyclohexylphosphino)butane DIOP^(2R,3R) or (2S,3S)-0-isopropylidene-2,3-dihydroxy-1,4bis(diphenylphosphino)butane R,R-DiPAMP^(1R,2R)-bis(o-anisylphenylphosphino)ethane DMA^N,N-dimethylacetamide, CH3(C=O)N(CH3)2 DMS O^dimethylsulphoxide DPCYCP^rac-(±)-1,2-bis(dicyclohexylphosphino)cyclopentane DPPB^1,4-bis(diphenylphosphino)butane DPPCP^rac-(±)-1,2-bis(diphenylphosphino)cyclopentane DPPE^1,2-bis(diphenylphosphino)ethane, also known as DIPHOS DPPH^1,6-bis(diphenylphosphino)hexane DPPP^1,3-bis(diphenylphosphino)propane xii  DPPPt^1,5-bis(diphenylphosphino)pentane e.e.^enantiomeric excess ipr^iso-propyl, (CH3)2CH J^coupling constant, in Hz K^equilibrium constant k^rate constant L^ligand litre configuration relative to L-glyceraldehyde 1^path length M^central metal atom in a complex molarity, moles per litre m^multiplet (NMR) moderate intensity (IR) Me^methyl, CH3 NORPHOS^(R ,R)- (+2- exo -3 - endobis(diphenylphosphino)bicy cloP, .2 .11heptane obs^observed P-P^ditertiary phosphine Ph^phenyl, C6H5 PHENOP^the chiral aminophosphinephosphinite ligand Ph2PN(EOCH(CH2Ph)CH2OPh2 PROPHOS^1,2-bis(diphenylphosphino)propane q^quartet (NMR) absolute configuration (Latin: rectus; right) (R)RT^room temperature S^solvent or substrate (S)absolute configuration (Latin: sinister; left) s^singlet (NMR) strong (IR) T1^spin-lattice (or longitudinal) relaxation time (NMR) t^time; triplet (NMR) tert^tertiary TMS^tetramethylsilane  tolyl^C6H4CH3 w^weak intensity (IR) w1/2^linewidth at half height X^anionic ligand A^heat 8^chemical shift E^extinction coefficient (in M - lcm -1 ) *^chiral centre  ACKNOWLEDGMENTS  I would like to thank Professor B. R. James for his advice and guidance throughout the duration of this work, and members of the group for their support. I would also like to give special thanks to: the proof-readers Mr. Christopher Alexander, Mr. Jeff Debad, Mr. Kenneth MacFarlane, Ms. Golnar Rastar and Mr. Kevin Ross for their patience and helpful suggestions (Thanks guys!); Dr. Andrew Pacheco for his help on the stopped-flow system and his designing of the computer programs; and support staff from various departmental services for their assistance. I am also very much indebted to Mr. Andy Lee for the loan of his computer and his incredible patience in helping me with the computer programs. I also wish to express my deepest appreciation to Mr. Samson Chan and Mr. George Wang for their computer expertise. Finally, I would like to thank my family for believing in me, and their unfailing love, support and encouragement throughout the years.  December, 1992^  xv  DEKC  CHAPTER 1 Introduction 1.1 General Introduction  Homogeneous catalytic conversion of unsaturated organic substrates mediated by transition metal catalysts has been studied extensively over the past three decades. Homogeneous catalysts exhibit certain advantages over heterogeneous catalystsl, including higher activity, selectivity, stereospecificity, reproducibility, and more importantly, the fact that mechanistic details are generally attainable. Moreover, by simply changing the ligands and reaction conditions, properties of the catalyst may be varied. These catalysts also allow incorporation of auxiliary chirality for asymmetric syntheses, in which enantiomeric products could be produced in unequal amounts. This is particularly important in the pharmaceutical industry where usually only one of the enantiomeric products is desired because the other enantiomer may be inactive or even harmful to the body ? Homogeneous catalysis systems do suffer from limited commercial interest, mainly due to difficulties in separation and recovery of the catalysts from the products at the end of the catalytic processes and, in the case of phosphine-containing catalysts under reducing conditions, phosphine degradation. 3 Moreover, sensitivities of the catalyst toward air and moisture remain as major obstacles for such homogeneous systems to be employed industrially. 4 In recent years, supported or "heterogenized homogeneous" catalysis, in which homogeneous catalysts are "anchored" on insoluble inorganic supports (e.g. silica, alumina and zeolites), or on organic polymer supports (e.g. polystyrene-divinylbenzene 4 and polyamides 5 ) have been used in order to circumvent the separation problem. Many 1  important industrial processes such as the Wacker and the Oxo processes, and the ZieglerNatta systems, are now currently employing homogeneous catalysts. 6  1.2. Homogeneous Hydrogenation  Hydrogenation of unsaturated organic substrates still remains as one of the most intensively studied areas in homogeneous catalysis. Vast amounts of literature reports on the topic, including comprehensive reviews 7-9 and books, 1,6,1°-14 have been published. Dihydrogen is often used as the major source of hydrogen, although the transfer of hydrogen from other solvents (e.g. alcohols, glycols, aldehydes, amides, amines, ethers and alkylbenzenes) or donor organic molecules (e.g. silanes) provides an alternative route to H2-hydrogenation? Most of the literature on homogeneous hydrogenation deals with substrates bearing carbon-carbon multiple bonds. 15 A number of systems have been studied in detail and have had their mechanisms deduced. One such well known example is the hydrogenation of olefins by the so-called Wilkinson's catalyst, RhCl(PPh3)3, which is routinely used for the hydrogenation of a range of olefins under relatively mild conditions (25-100°C, 1-3 atm H2, Equation 1.1) 6 Mechanistic details for this and related systems were also considered. 10 . 16 In recent years, however, a considerable amount of attention has been focused on the reduction of carbon-nitrogen 10. 17 3 8 and carbon-oxygen double bonds. 10,15,17,19,20 Hn 4  RhC1(13Ph3 )3  25 - 1000C 1 - 3 atm  2  Perhaps the most sophisticated application of homogeneous hydrogenation catalysis is the synthesis of optically active organic compounds from nonchiral starting materials. A tremendous amount of research has been focused on designing such catalysts in recent years. To achieve asymmetric hydrogenation, three conditions are required: (i) the substrate molecule possesses prochiral faces; (ii) the catalyst incorporates a chiral site; and (iii) the substrate and hydrogen are held in close proximity to each other within the diastereomeric transition state which produces the new chiral center. 21 Product with high optical purity is obtained only when the prochiral site is bound preferentially in one conformation to a vacant site on the catalyst. A number of excellent reviews on the subject have appeared in the literature. 19 . 22-27 One commercial application of such a system is the Monsanto process, in which a Wilkinson-type rhodium(I) catalyst with a chiral diphosphine ligand (P*-P*), is used for the large scale production of L-Dopa (Equation 1.2), an amino acid drug used for treating Parkinson's disease. 26  CO 2R NHCOR  1) [RhOCI3* )(S)2]+  9.),*. R, R -DiPAMP S = Me0H  + H2  2) H30+  AcO^OCH3 R, R-DiPAMP is illustrated below in Figure 1.1  (1.2) OH  ^  L - Dopa  One of the quantitative measures of the efficiency of an asymmetric synthesis is the enantiomeric excess, which is defined as: - [S]1 ^ x 100 % enantiomeric excess (% e. e.) = 1[R] [R]^[ + S]  3  (1.3)  As the use of chiral ligands on a metal results in a preferred conformation of substrate binding, metal complexes modified with chiral auxiliary ligands can result in products having a high % e.e. in asymmetric hydrogenation. This provided the impetus for designing chiral ligands that complex a metal and provide high stereoselectivity. In 1968, chiral monophosphines of type P * PhR 1 R2 , in which a phosphorus atom bears chirality, were first prepared. 28 Following that development, the synthesis of the first chiral diphosphine DIOP (Figure 1.1), a bidentate ligand prepared from naturally occurring tartaric acid, with chiral carbons joining the two phosphorus atoms, was reported by Kagan's group in 1971. 29 A wide variety of chiral ligands was soon prepared, including non-phosphine chiral ligands such as Schiff bases, amines, amides, and sulphoxides. 7 . 17 However, these ligand systems in general could produce only moderate enantioselectivity (-50% e.e.). Subsequently, attention turned even more to the preparation of chiral phosphines which were tested as ligands. 23 .25 b Some representative examples of such phosphines are depicted in Figure 1.1. Of particular interest are: a) CHIRAPHOS 3 °a, which contains two asymmetric centres in the carbon backbone, b) ferrocene-based BPPFA 2Th, which exhibits planar as well as carbon chirality, c) DiPAMP which is used in the Monsanto process for L-DOPA outlined above, 23b and d) BINAP, which exhibits just a planar chirality. 19 Before the 1980s, the most prominent catalysts for asymmetric hydrogenation were soluble rhodium complexes which contain a chiral phosphine ligand, with generally mild conditions being required to hydrogenate a variety of prochiral substrates. 21 . 31 Recently, however, certain ruthenium complexes bearing chiral diphosphines have shown high enantioselectivities for asymmetric hydrogenation of a variety of organic substrates, and developments in this area are well documented in the literature.  4  19  1-1 CH3 (....„4_ PPh22 PPh2  R1^PPh2 H H R2 PPh2  N(CH3 )2  We  ( PPh2  '  DIOP29  R 1 = H, R2 = Me: PROPHOS 3 ob R 1 = R2 = Me: CHIRAPHOS 3 oa  R, R -DT AMp 26  BIN AP 1 9  PPh2 BPPFA27 b  BIPHEMP201  FIGURE 1.1 : Some chiral diphosphines used in catalytic asymmetric synthesis  While an immense amount of literature has been devoted to the synthetic and potential catalytic aspects of asymmetric hydrogenation, it was quantitative kinetic studies which provided understanding of the catalytic mechanism and allowed better systems to be designed. One such contribution in this area came from Halpern's group, who studied and elucidated the mechanism for asymmetric hydrogenation of a-amino acids catalysed by Rh complexes using CHIRAPHOS and DiPAMP as ligands. 32 (cf. equation 1.2) In order for a transition metal complex to be catalytically active towards hydrogenation, three fundamental processes in the mechanistic cycle have been characterised. 812 a) activation of both substrate and hydrogen b) transfer of hydrogen to the substrate c) release of product, and regeneration of catalyst.  5  The minimum and essential requirement is that H2 must be activated and, because this thesis focuses on this topic, activation of dihydrogen is outlined below in more detail.  1.3 Dihydrogen Activation  Activation of dihydrogen using metal complexes in homogeneous systems remains the key step towards both stoichiometric and catalytic hydrogenation of unsaturated substrates, although activation of the substrate is also important. Direct addition of dihydrogen to an unsaturated substrate is thermodynamically favourable in the ground state, but is a symmetry forbidden process. 33 One role of a transition metal complex is simply to provide a template to circumvent the symmetry restrictions. The reaction proceeds through a series of symmetry allowed reaction steps involving a metal hydride intermediate. Hydrogenation involves activation of both dihydrogen and the substrate at the metal centre; some systems operate with the dihydrogen being added oxidatively to the metal centre to form a dihydride, followed by stepwise transfer of a hydrogen atom to the substrate and subsequent reductive elimination of the product. 7,8,12 Coordinative unsaturation at the metal centre is generally a prerequisite for activation of dihydrogen, 8,12 but coordinatively saturated complexes can become active toward dihydrogen by dissociation of a labile ligand in solution. Complexes with group  8 to  10 metals in low  oxidation states (0, I, II) provide the most common systems for dihydrogen activation. Electronic and steric properties of the auxiliary ligands can also influence critically the nature of the interaction between a metal complex and dihydrogen. 12 In general, activation of hydrogen can be achieved in two basic ways: (a) oxidative addition at a metal centre which proceeds via initial coordination of the dihydrogen molecule, presumbly through a three-centre, two-electron transition state (Figure 1.2a), the addition resulting in a homolytic cleavage of the H-H bond, and (b) heterolytic 6  cleavage of the H-H bond, through a four-centre transition state (Figure 1.2b), which generates a metal hydride and a proton. These different processes concerning H2 activation have been discussed in great length in a number of review articles on hydrogenation. 7 ' 8 Ha:^H8+  H^H  M8+^ X8a  Figure 1.2 Proposed transition states during a) homolytic and b) heterolytic activation of dihydrogen by transition metal complexes (a) Homolytic cleavage Oxidative addition of H2 to a metal centre ultimately results in the formation of a dihydride (or polyhydride if hydride ligands are already present) with an increase in the oxidation state of the metal by two. Depending on the type of metal complex being used, there are generally three ways to represent such addition (where n is the formal oxidation state of the metal): n+2 (H)2 m H2 Mn +^______) 2M n + H2^2HMn+1 Mt n + H2^2HM I1+1 ---)  (1.4a) (1.4b) (1.4c)  (b) Heterolytic cleavage Dihydrogen activation is accomplished by a net substitution of a hydride for another ligand without changing the oxidation state of the metal (Equation 1.5a). A base  7  is usually required to promote an overall splitting of the H-H bond. In a similar fashion, hydrogenolysis or metathesis of an M-R bond (R = C, Si, Ge) with H-H also generates monohydrides in an analogous manner (Equation 1.5b), with the leaving group essentially behaving as a base. M n X + H2 + B MnH + BH +  +  (1.5a)  R—M n + H2 --> M n H + R—H (1.5b)  Whether hydrogen activation is achieved homolytically or heterolytically, these interactions are of great importance for both stoichiometric and catalytic reactions. The energetic aspects of hydrogen activation are also a very important consideration in a catalytic cycle. While a strong metal-hydrogen bond may slow down or even prevent hydrogen transfer from the metal to the coordinated substrate, an interaction that is too weak may result in a M-H concentration that is too low for any substantial hydrogen transfer reaction to take place. A new insight in the modes of hydrogen activation has emerged with the discovery of the first isolable molecular hydrogen complex by Kubas' group in 1984.  34  Since then, a  vast amount of research has been done in this area, and a number of reviews have been published in the literature. 35-38 The next section is devoted to various aspects of these novel complexes.  8  1.4 Molecular Hydrogen Complexes: a Brief Review  1.4.1 Discovery of Molecular Hydrogen Complexes  The formation of hydride complexes has always been thought to play a key role in most catalytic hydrogenation systems. 38 Speculation of the possible interaction of a dihydrogen molecule with a metal centre during the transition state or as an intermediate has been raised periodically, and reports of possible H2 binding appeared sporadically in the literature, but all lacked sufficient proof. In 1976, Ashworth and Singleton reported indirect evidence for existence of dihydrogen binding. 39 A series of reactions with the complexes RuH4(PPh3)3 and RuH3(DPPE)2+ was observed to proceed by initial loss of H2, and from this behaviour, similiar to that of an analogous complex containing a reversibly bound dioxygen ligand, they proposed that RuH4(PPh3)3 was better formulated as Ru(H2)H2(PPh3)3, whereby these complexes could be considered as H2 adducts of Ru(II). However, no direct spectroscopic or structural evidence was presented at that time. Kubas, in a 1980 report, 40 based on the infrared data of the complexes written as M(H2)(CO)3(PR3)2 (M = Mo, W; R = Pri, Cy) and the lability of dihydrogen in these complexes, suggested that bonding of H2 in these complexes "may be novel". No direct evidence for the structure of a dihydrogen complex was found until 1983 when the complex W(C0)3(PPr 13)2(r 2 -H2) was studied by diffraction methods (Figure 1.3a). This complex contained a side on (1 2 -) bonded H2 ligand with the H-H distance determined to be 0.75(16) A by X-ray and 0.84  A  by neutron diffraction analysis. 34 Moreover,  observation of a 1 : 1 : 1 triplet in the 1 H NMR spectrum of the HD isotopomer with a large HD coupling constant of 33.5 Hz 34 . 38 (cf. 1 40 = 43.2 Hz for gaseous HD 41 ) also proved to be diagnostic of a coordinated 11 2 -HD molecule with somewhat reduced H-D bond  9  order. These findings together with the initial IR data thus provided conclusive evidence for the first example of a complex containing a 'nonclassical' dihydrogen ligand. 38 1.4.2. Further Examples of Molecular Hydrogen Complexes  Soon after Kubas' discovery, Crabtree and Lavin reported in 1985 that a polyhydride complex, [IrE1(1 2 -H2)(bq)(PPh3)2]+ (Figure 1.3b; bq = 7,8-benzoquinolinate), formed by displacement of an H2O ligand by H2, also showed non-classical behaviour by 1 H NMR T1 measurements. 42 The relevance of T1 measurements on molecular hydrogen complexes will be discussed later. Next, Morris et al. found that [MH(11 2 -H2)(DPPE)21+ (M = Fe, Ru), formed by protonating MH2(DPPE)2, also contained bound dihydrogen, as evidenced by X-ray diffraction (H-H = 0.89(11) A) and 1H  NMR spectroscopy ( 1 JHD = 32 Hz) data, 43 concordant with the original speculation  by Ashworth and Singleton about the Ru complex (Figure 1.3c). 39  PR3 oc„, I ,,,CO • ^H o c- I PR3 H M = Mo, W R = Pri, Cy a  H—H IP)  , , IPPh3  is  Ph3P 0^I^'1115/ 1-1  PN = 7,8-benzoquinolinate  P  M= Fe, Ru 1C-1) = DPPE  b  Figure 1.3 The three early examples of molecular hydrogen complexes With the remarkable discovery of molecular hydrogen complexes, many new complexes soon followed in quick succession, as evidenced by the large number of y ars.44-52 There are about 80 known 12-H2 literature reports appearing in the last sevene 10  complexes, 60 of which are stable at room temperature. 35 The first such dinuclear species (a Rut complex) was reported from work at this University. 53 A number of review articles have also emerged, 35-38,54 and theoretical aspects of the subject are also being considered. 55,56 Interestingly enough, several polyhydrides previously considered to have unusually high oxidation states are now being reformulated as containing H2 ligaridS. 46M 8d . 57 For example, through NMR T1 studies and neutron diffraction data, the complex RuH4(PPh3)3 described previously was reformulated as Ru(r) 2 -H2)H2(PPh3)3. 46b ,48d Meanwhile, the dinuclear complex originally formulated [RuH2C1(PPh3)2]2 has been shown to be [(11 2 -H2)(PPh3)2Ru(g-H)(11.-C1)2RuH(PPh3)2] by NMR and structural data. 57 With the exceptions of V, Ta, Tc, Ag and Au, molecular hydrogen complexes have thus far been reported for all transition elements from Group 5 to 11, and Table 1.1 provides some representative examples from each group. 38 Many new complexes were derivatives of Group 8 metals, with ruthenium being the most prominent. Of interest, dihydrogen has also been found in the matrix-isolated "ligand free" species including the complex Pd(r1 2 -H2), 58 small cluster complexes Cu2H2(11 2 -H2) x and Cu3(r1 2 -H2) 59 and stepped Ni(100) surfaces. 60 Lanthanide metal Eu 61 and some osmium-porphyrin systems also appear to contain axially bound 1 2 -H2 ligands. 62  11  ^  Table 1.1 Examples of molecular hydrogen complexes Group^Electron Element^ number^configuration  Example(s)^Ref.  5^Nb^d2^NbCP211(r12-112)^54 Cr(C0)50.1 2 -H2)a 6^Cr^d6 Cr(C0)4(PCY3)2012-H2r  38,54  d6^M(C0)3(PR3)2(r12-H2), Mo,W^ R = Cy, Pri d4^M(H)(112-H2)Cp(C0)2  38,54  Mo^do^[MoH4(712-H2)(DPPE)2] 2+ b^54 7^Re^d2  ReH5(r12-H2)(PPh3)2,^63 ReH5(712-H2)(DPPE)  8^Fe, Ru, Os^d6^[MH(712-H2)(DPPE)21+ ^54 8^Ru^d6^Ru(H)(r12-H2)(I)(PCy3)2^51e 8^Os^d4^[0s(H)3(r12-H2)(PPh3)31+ ^37 9^Co^  d8^CoR(T12-H2)(C0)3,a R = H, CH3 38,54 d9^Co(C0)2(N0)(12-H2)a^  9^Rh^d8^{Rh[P(CH2CH2PPh2)3]012-H2))+ ^54 [IrH(71 2 -H2)(bq)(PPh3)2]+ 9^Ir^d8^(bq = 7,8-benzoquinolinate) ^37,54 [ 11E2(r1 2-112)2(PCY3)2] + 10^Ni^d10^Ni(C0)3 (r12412)a^ 64 10^Pd^d10^pdcr12_H2)a^  58  10^Pt^d10^Pt(DPPE)(112-H2)13^54 H2) x ,^59 11^Cu^d10^Cu2H2012-^Cu3(112-H2)a a Thermally unstable species. b Detected as transient intermediates under ambient conditions.  12  1.43 Properties of Molecular Hydrogen Complexes  As depicted in Table 1.1, molecular hydrogen complexes have been found to span a range of even numbered do configurations, with formal oxidation states between 0-VI and coordination numbers from four to nine, counting 11 2 -H2 as occupying two sites. All complexes are either neutral or cationic, and nearly all the known stable dihydrogen complexes are d 6 in electronic configuration, forming 18-valence electron species upon H2 coordination with the exception of Ru(H)(1 2 -H2)(I)(PCy3)2, which is a stable 16electron species, counting r) 2 -H2 as a two-electron donor. 51 e The first d 2 example was ReH5(r1 2 -H2)L2 (L2 = DPPE or 2PPh3) while [Os(H) 3(112-H2)(PPh3)3]-1- is a d 4 complex. The d 0 and d 10 species have not been isolated thus far, and the d° species has been seen only as a transient intermediate. Some d 10 examples like Ni(C0)3(r1 2 -H2) are known from matrix work only. 64 Most d 8 , 16-electron species (e.g. of 11. 1 and RhI) tend to bind H2 oxidatively, forming d 6 , 18-electron species with an octahedral geometry that have a particularly strong ligand field stabilization. Except for the matrix stabilised complexes, 58,59,64 almost all dihydrogen complexes contain at least one or more n-acceptor ligands. There are three general routes for synthesizing H2 complexes: i) protonation of metal hydrides to give cationic species, ii) addition of H2 to unsaturated 16-electron precursors, particularly those having an agostic C—H interaction, and iii) photolytic displacement of CO under H2, which usually results in the formation of thermally unstable complexes.38  13  1.4.4. Bonding and Stability of Molecular Hydrogen Complexes  The bonding scheme of dihydrogen complexes can be best described by the Dewar-Chatt-Duncanson model (Figure 1.4). 36-38  Figure 1.4 Bonding scheme of M-(1 2 -H2) The bonding in M-(11 2 -H2) consists mainly of two components. The first component is donation of the H—H a-bonding electron density to the empty M(d 6) orbital. This weakens but does not break the H—H bond, because the resulting threecentre molecular orbital (MO) is bonding over all three centres. The second component is the electron donation from the filled M(d ir ) orbital back to the empty H—H a * -orbital. If this back component is too strong, a rupture of H—H bond will result because the H—H 0* -orbital iis being filled, resulting in an oxidative addition of H2 to give dihydrides.  36-38  Two-electron, three-centre bonds are common in boron-hydride chemistry and, more recently, in agostic type interactions where a C—H bond is essentially chelated to the metal complex. Such interactions appear to form most readily when at least one of the three centres is H. 37 This is perhaps because of the small size of the hydrogen atom which allows close proximity of the H-containing ligand to the metal; the absence of lone pairs of electrons on the H2 molecule prevent its "direct" binding. Stable complexes with weak 14  donors like H—H are formed only with non-d 0 transition metals, suggesting that some back-bonding is necessary. Thus, Lewis acids without d o electrons (BF3), and d 0 transition metals, do not generally seem to bind H2. 38 In considering what effect electronic and steric influences of auxiliary ligands have on stabilizing dihydrogen versus dihydride formation, Kubas 38 demonstrated that increasing the basicity of the metal centre in H2 complexes leads to H—H bond cleavage due to higher M(dic ) —) H—H(e) donation. Ab initio calculations illustrate that within the series of complexes Mo(H2)(C0) xP5_ x (P = PH3), the strong n-acceptor CO ligand stabilizes 11 2 -H2 coordination (i.e. when x > 0). When trans to H2, Tc-acceptors interact with the same metal d orbital which donates to the H—H(6 * ) orbital, reducing the amount of back-bonding to the H2, and so mixtures of a-donor and strong 7E-acceptor co-ligands favor dihydrogen binding. 65 Indeed, the crystal structure of W(C0)3(PPri3)2(71 2 -H2) 34 shows that the H2 ligand is oriented in a way that such n-interaction with the carbonyl ligands is possible. However, complexes like Mo(C0)5(r1 2 -H2) are thermally unstable because the amount of back-bonding is too small. In contrast, Mo(PMe3)5H2, with no itacceptor ligand, is a 7-coordinate dihydride. 38 This suggests that some back-bonding is required for stable molecular hydrogen complexes. Of note, even though trans NH(71 2 -  -  H2)(DPPE)2r species (M = Fe, Ru) contain no n-acceptor ligand, the stability of H2 coordination arises from the low basicity of the metal, the high trans effect of the hydride, and the positive charge that decreases back donation from the meta1. 66 Within the series Mo(CO)(R2PCH2CH2PR2)2H2, where R is varied to alter the donor strength of the chelating phosphines, Kubas demonstrated that an r1 2 -H2 ligand was found with R = Ph whereas a dihydride was observed with R = Et. 38 With R = Bui, where the size of PBui3 --- PPh3 but basicity resembles that of the Et complex, a dihydride complex was observed. This suggests that electronic factors are more important for stable dihydrogen complexes than steric influences. Meanwhile, the effects of chelating ring size in dihydrogen formation has been illustrated with the series CpRuH(PPh2(CH2) nPPh2) 15  studied by Conroy-Lewis and Simpson. 46" Protonation of the complexes results in dihydrogen ligand formation with n = 1, dihydride with n = 3, and a mixture of both tautomers with n = 2. Using the dinitrogen derivatives as a tool to measure the electronic environment of the binding sites of metal complexes, Morris et al. 47 a have suggested that the stabilities of molecular hydrogen complexes can be predicted by merely noting the stretching frequencies (VNN) of the related N2 complexes; they proposed that upon replacement of N2, stable molecular hydrogen complexes should be obtained if 2060 < V NN < 2160 cm -1 versus dihydride formation when VNN < 2060 cm -1 . With increased ic-basicity at the metal centre, both the N2 7E * and H2 a* orbitals are populated, resulting for example in a lower VNN, which indirectly suggests a higher probability of cleaving the H—H bond. In a recent paper, 67a Morris provided a theoretical approach for predicting stable dihydrogen complexes, by estimating electrochemical potentials of dinitrogen complexes using Lever's ligand additivity method. 67 b,c Table 1.2 lists some examples correlating VNN with the type of d 6 complexes formed when N2 is replaced by H2, and the observed VHH for some dihydrogen complexes. Of note, the Ru-porphyrin systems Ru(TMP)(L)(N2), 68 where TMP = dianion of 5,10,15,20-tetramesitylporphyrin and L = THF, DMF and Et3N, do not form molecular H2 complexes upon replacement of N2 with dihydrogen, even though the VNN reported falls within the range of 2060 to 2160 cm -1 .  16  Table 1.2 Correlation of dihydrogen versus dihydride complex formation with VNN of analogous dinitrogen complexes M(N2)L5 a , and Vim for some dihydrogen complexes M(H2)L5 . Formula (ML5) Cr(CO)5 Cr(C0)4(C2H4) Ru(1.1.-C1)3(DPPB)c Mo(C0)3(PCy3)2 [OsH(DPPE)2] -4Mo(CO)(DPPE)2 W(C0)3(PCY3)2 [FeH(DPPE)2]+ FeH2(PEtPh2)3 Mo(PMe3)5 ReCl(PMePh2)4 Ru(TMP)(L)d  VNN (cm 1 ) -  2237 2223 2175 2159 2136 2120 2120 2120 2058 1950 1925 2108-2147  H2 vs dihydridesb VHH (cm 1 ) -  unstable H2^3030 unstable H2 labile H2 labile H2^2690 stable H2 labile H2^2650 labile H2^— 2950 stable H2 labile H2^2380 dihydride dihydride no reaction  a Data taken from ref. 37, 38, 67a, 68. b Unstable means unstable at ambient temperature; labile means the complex is fully formed only under 1 atm H2. C Part of Ru(DPPB)(11-C1)3RuC1(DPPB). 69 d TMP = 5, 10,15,20-tetramesitylporphyrin dianion; L = THF, DMF, Et3N.  1.4.5 Methods Used in Recognition of Molecular Hydrogen Complexes  The most direct proof for the existence of a non-classical hydride is undoubtedly provided by crystal structure determination, particularly using neutron diffraction analysis. Use of neutron rather than X-ray diffraction is more reliable in locating the 71 2 -H2 moiety, but the need for relatively large crystals (— 10 mm 3 ) limits it usage. Even then, problems with physical disorder and chemical instability of the crystal pose further difficulties. 36-38 Use of infrared and Raman spectroscopy data can be unreliable, because the absorption 17  mode V(MH2) for 1 2 -H2 ligand is often either masked by other modes or is too weak to be observed. Nevertheless, broad IR bands in the range of 2400 to 3100 cm -1 are generally assigned to the H—H stretching mode (Table 1.2). 38 By far the most useful diagnostic tool for distinguishing dihydrides from molecular hydrogen complexes is 1 H NMR spectroscopy. Two methods routinely employed are: (a) Measurement of H—D coupling constants Formation of HI) isotopomers can be accomplished by using ID instead of H2 during synthesis, or by deuteration of an 1 2 -H2 complex. Coupling between hydrogen and deuterium is commonly found to be at least an order higher in magnitude in HD isotopomers (28-34 Hz) than in a hydrido (deuterido) species (-=-. .1 Hz), 36-38 and should be comparable to that for HD gas (43.2 Hz 41 ) (see Table 1.3). However, complexes exhibiting 1 JHD values as low as 13 Hz have been reported, 38 indicating stronger interactions of the HD molecule with the metal centre. Moreover, fluxional behaviour in polyhydride complexes and unresolved coupling with metal and phosphine ligands often result in loss of coupling information. (b) Measurement of proton relaxation times, T1 This method was first developed by Crabtree et al. in characterising the complex fIrH2(H2)(PCy3)21+, 7 ° and established a significant alternative NMR criterion for identifying molecular H2 coordination. The T1 (also known as longitudinal relaxation time) of any resonance can be measured by a standard inversion-recovery (180°-T-90°) pulse sequence." The nuclear spins are first inverted by a 180° pulse, and the rate of their return to their initial equilibrium state along the applied magnetic field is measured as (T1) -1 . 36,37 For protons less than 2  A apart, the dipole-dipole relaxation mechanism  [R(DD)] is the major contributor to such relaxation. Therefore the relaxation rate can be related to the internuclear distance by equation (1.6) depicted below: 18  Table 1.3 Selected 1 H NMR HD coupling constants (Hz) and Ti( . .^valuesa for some molecular hydrogen and hydride complexes. Complexes Molecular Hydrogen Complexes: W(C0)3(PPr13)2(T1 2 -H2) ReH5(71 2 -H2)(PPh3)2 ReH5(11 2 -H2)(DPPE) [FeH(11 2 -H2)(DPPE)21 + RuH2(71 2 -H2)(PPh3)3 [CpRu(11 2 -H2)(C0)(PCy3)1+ Ru(1 2 -H2)(DPPB)(1.1.-C1)3 RuCl(DPPB) [0s(11 2 -H2)H3(PPh3)3]+ [IrH(i 2 -H2)(bq)(PPh3)2J+  vb (Hz)  )  (ms)  0(,min )  rHH (calcd)  rHH (obsd)  (K)  (A)  (A)  28  <5 110f 67 <8.5 30 4  193 200 222 210 266 205  <0.96Cd 1.20C 1.25C <1.05C 1.16C 0.93C  29  12  277  1.08C  35 8  220 200  1.16C 1.04C  165 179 <820  250 240 <193  1.72C 1.74C <2.22C  34  30  Hydride complexes: MoH4(PMePh2)4 WH5(PMePh2) IrH5(PCy3)2  0.84e  0.89(11)g 0.97h  0.92h  a Data taken from references 38, 63, 69; rHH calculated from Equation 1.6. b For analogous HD complexes. C At 250 MHz. d C = 1.0 for hydrides and nonrotating H2, 0.79 for fast rotating H2. See text for details. e From neutron diffraction data. f At 500 MHz. g From X-ray diffraction data. h From solid-state NMR data.  4;  R(DD) = (T1 (DD)) -1 = 0.3h 2 yti rai { tc2 2 + 1+ CO ti e 1+ 40.) 2 T! 3  (1.6)  where, h = Planck's constant / 2n (1.055e-34 J s) y H = gyromagnetic ratio for proton rHH =^internuclear  ti c =  distance (cm)  rotational correlation time (s / rad) = 0.63 / co at 0 (  = Larmor frequency (s / rad) 19  i)  ; 0 = temperature (K)  The dependency of T1 on r6 indicates its sensitivity towards the presence of two protons that are close together, as in the case of an T1 2 -H2 complex. Further, differentiating equation 1.6 shows that T1 will pass through a minimum when the Brownian motion, measured as the rotational correlation time, ;, best matches with the Larmor frequency, co. This results in the relation: ; = 0.63/w at Ti( min), and the internuclear distance can then be estimated. This T1 (i) also allows data from different solvents to be compared. T1 usually lies between 4-100 ms for dihydrogen complexes, with r = 0.8 - 1.2 A, whereas the observed T1 for a typical hydride moiety is 1 s with r > 1.6 A35-37 (see Table 1.3 for selected samples). Using Ti(min) often yields systematically larger r values than crystallographic data for dihydrogen complexes, and Morris' group showed that the discrepancy is likely due to H2 rotation. 72 When the H—H molecule is rotating at a faster speed about the M-(H2) bond than the molecule rotates as a whole, then the H—H bond will be relaxed less efficiently because both H atoms will not have a rotation frequency close to the Larmor frequency. As a result, Morris and coworkers refined equation 1.6 by inserting a correction factor C to the internuclear distance calculated; C is given a value of 0.79 for dihydrogen complexes having fast H2 rotation, and this increases to 1.0 for systems with nonrotating H2. Certain polyhydride complexes often exhibit borderline T1 values causing the distinction between classical and non-classical hydrides to be difficult (eg. ReH5(11 2 H2)(PPh3)2 and ReH5(i 2 -H2)(DPPE) in Table 1.3). In fact, some reports have advised caution against such classification. 38,73 .74 One such report from Desrosiers et al. 74 notes that many earlier attempts to distinguish between classical and non-classical polyhyrides failed to take into account interactions of the hydride ligands with other nuclei, and assigned dubious T1 values to non-classical hydrides even though the classical cases were also consistent with the observed T1 values. Through a series of studies of polyhydrides 20  using internuclear distance derived from X-ray and neutron diffraction data, these workers concluded that Ti( mi n ) can be calculated to within a few percent if all dipole-dipole interactions between ligand and metal nuclei are included. Further, some polyhydrides show T1 values (e.g. 67 and 110 ms for ReH5(1 2 -H2)(L2), L2 = DPPE and 2PPh3, see Table 1.3) consistent with both classical and non-classical structures. This represents a major limitation to T1 studies, namely, "while structure uniquely defines Ti ( min), Ti( min ) does not uniquely define structure". 74 Even in the case of classical hydrides, dipole-dipole interaction between hydrogens are still the main contributors to hydrogen relaxation. With the developments in molecular hydrogen chemistry, it appears that initial complete rupture of H—H bond at a metal centre may not be necessary in catalytic hydrogenation, and that catalysts like RhCl(PPh3)3, which are known to form dihydrides with H2, may effect hydrogenation through direct transfer of H from an T 2 -H2-containing intermediate or equilibrium species. 38 Indeed, direct evidence for transfer of a 1 2 -H2 to an olefin has been presented for a dinuclear Ru2 system. 53  21  1.5 Scope of This Thesis  Previous work from this laboratory revealed that the reaction of the chlorobridged, dinuclear Ru(II) complexes, [RuCl(P-P)(µ-C1)]2, (P-P = DPPB), with H2 yielded the in situ formation of the molecular hydrogen complex, [(11 2 -H2)(P-P)Ru(g-C1)3RuC1(PP)]. 69 This in turn, in the presence of an added base, formed the trinuclear ruthenium(II)hydride cluster, [RuHC1(P-P)] 3 . 17 .75 The dichloro-bridged, dinuclear Ru(n) complex catalyzed the hydrogenation of alkenes under 1 atm H2, with the molecular hydrogen derivative believed to be the catalytically active species. 45 It was of importance to determine whether analogous dihydrogen complexes would form using different diphosphine ligands, and how such complexes behaved under catalytic conditions. Chapter 2 details all the experimental procedures for the present study. Chapter 3 describes the synthesis of other [RuCl(P-P)(11.-C1)]2 dimers containing different diphosphine ligands and their reactivity under a hydrogen atmosphere. Also, continued investigations of the interaction of [RuCl(DPPB)(11.-C1)]2 with H2, including thermodynamic parameters and kinetic studies, are described in Chapter 4. Chapter 5 contains some preliminary results on the reaction of [RuCl(P-P)(11-C1)12 with 02. Chapter 6 will summarize the results found in the present study with conclusions and suggestions for further work. Finally, Chapter 7 lists all references and footnotes.  22  CHAPTER 2 Experimental Procedures 2.1 Materials  2.1.1 Solvents  Spectral or reagent grade solvents were obtained from Fisher, BDH, MCB, Eastman or Aldrich Chemical Co.. Benzene, diethyl ether, hexanes, tetrahydrofuran(THF) and toluene were distilled from sodium/benzophenone under an atmosphere of nitrogen. Acetone was distilled from anhydrous K2CO3, dichloromethane from P2O5, while both methanol and ethanol were distilled from Mg/I2. 76 N,N-Dimethylacetamide (DMA) was refluxed over CaH2 for at least 24 hours, vacuum distilled at 35-40 0 C, and stored under argon in the dark. All solvents were deoxygenated by saturating with argon prior to use. The deuterated solvents (CDC13, C6D6, C7D8 and CD2C12), used in NMR spectroscopy, were obtained from MSD Isotopes and the CIC Chemical Co.. All deuterated solvents were dried over activated molecular sieves (Fisher: Type 3 A, 8-12 mesh), and deoxygenated by freeze-pump-thaw cycles.  2.1.2 Gases  Purified argon (H.P.), nitrogen (U.S.P.) and hydrogen (U.S.P.) were obtained from Union Carbide Canada Ltd.. Nitrogen was used without purification. Argon was passed through columns of activated molecular sieves and Ridox® to remove traces of water and oxygen. Hydrogen was passed through an Engelhard Deoxo catalytic hydrogen purifier to remove traces of oxygen. A lecture bottle of deuterium gas was obtained from MSD Isotopes Ltd. and was used without further purification. 23  2.13 Phosphines Reagent grade triphenylphosphine (PPh3), tri-(p-tolyl)phosphine (P(p-tolyl)3), dicyclohexylphosphine (HPCy2) and the chelating phosphines Ph2P(CH2) nPPh2 (n = 3, DPPP; 4, DPPB; 5, DPPPt; 6, DPPH) were supplied by Strem Chemicals and Aldrich Chemical Co. Inc., and were used as supplied, unless specified otherwise. The 1,4bis(dicyclohexylphosphino)butane ligand, DCYPB, was prepared by a procedure that was based on a literature method (Section 2.1.3.1). 77 The purity of all phosphine ligands was checked by both 31 P{ 1 H} and 1 H NMR spectroscopy.  2.1.3.1 Modified Preparation of 1,4-bis(dicyclohexylphosphino)butane, DCYPB 77  The precursor dicyclohexylphosphine (HPCy2) was first distilled under vacuum (b.p. = 129°C @ 8 mm Hg) and its purity was ascertained by 31 P{ 1 H) NMR [8 -29.4 ppm (s, C6D6, 20°C) 78 a ]. To a solution of this phosphine (10 mL, 45.6 mmol) in THE (100 mL) was added dropwise at -20°C with stirring exactly one equivalent of n-butyl lithium (1.05M in hexanes, 43.4 mL, 45.6 mmol). The resulting yellow mixture was allowed to warm to room temperature, and then stirred for one hour to give a milky yellow suspension. 1,4-Dibromobutane (2.7 mL, 45.2 mmol) was then added dropwise to this solution of LiPCy2 at -20 0 C , and the reaction mixture allowed to stir for another hour at room temperature. The volume of the resulting clear yellow solution was reduced to 5 mL in vacuo, after which boiling hexanes (100 mL) were added and the suspension reheated to boiling for 5 minutes. The insoluble lithium bromide salt was separated by filtration, and the filtrate was allowed to cool slowly to precipitate the diphosphine. The resulting white product was recrystallized from ethanol/THF (70:30) to give needle-like crystals [9.2 g (90%)]. Anal. Calcd for C28H52P2: C, 74.63; H, 11.63. Found: C, 74.45; 24  H, 11.89. The 1 H NMR spectrum showed two broad unresolved peaks in the region of 12 ppm and was uninformative. 31 P( 1 H) NMR (20°C): 8 -5.7 ppm (s, C6D6); 8 -4.0 ppm (s, CDC13). Low resolution MS: (m/z(I)) = 450[M]+(1); 367[M-Cy]+(100). Literature data77 give 31 P{ 1 H) NMR (20°C): 8 -5.8 ppm (s, d 8 -THF) and low resolution MS: (m/z(I)) = 450[M]+(1); 367[M-Cy]+(100). Modifications of the literature preparation 77 included: a) use of HPCy2 and n-BuLi to afford in situ LiPCy2 and direct consumption of this lithiated salt, and b) use of ethanol in the recrystallisation procedure instead of both ethanol and methanol.  2.1.4 Other Materials  Polyvinylpyridine (PVP) was used as supplied by Aldrich Chemical Co. Inc.. nButyl lithium (Aldrich) was supplied as a hexanes solution, and its concentration was determined by titration with 1,3-dipheny1-2-propanone tosylhydrazone in THE prior to use. 79 1,4-Dibromobutane was supplied by Eastman Co., and was deoxygenated by several freeze-pump-thaw cycles before use.  2.2 Instrumentation  Infrared spectra were recorded on a Nicolet 5DX FT-IR instrument, in which spectra were obtained as Nujol mulls between KBr plates, unless otherwise stated. UVvisible spectra were recorded on a Perkin-Elmer 552A spectrophotometer with thermostated cell compartments, using an anaerobic spectral quartz cell of 1.0 cm path length (Figure 2.1). 80 In most cases the sample was weighed and contained in a small glass bucket and inserted into the sidearm flask while the cell was in a glove bag. The solvent was deoxygenated by a freeze-pump-thaw cycle in another Schlenk flask and subsequently vacuum transferred to the cell. 25  Sidearm—. Flask  Figure 2.1 UV-Visible anaerobic spectral cell  All stopped-flow experiments were carried out on an Applied Photophysics model SF.17MV spectrophotometer. Two 2 5 mL drive syringes were used, and a pressure of 650 kPa was employed to drive the syringes. A Grant LTD 6 constant-temperature bath was connected to the stopped-flow sample handling unit via Tygon® tubing, and a constant temperature of 25.0 ± 0.1 ° C was established. The absorbance change with time was monitored across a 1.00 cm path length cell. A 150-W Xenon arc lamp was used as the light source. Solution nuclear magnetic resonance (NMR) spectra were recorded on a Bruker AC200 (200.1 MHz for 1 H, 50.3 MHz for 13C and 81.0 MHz for 31 P), a Varian XL300 (300.0 MHz for 1 H and 121.42 MHz for 31P), or a Bruker AMX500 (500.0 MHz for 1H) FT-NMR spectrometer, with tetramethylsilane (TMS) at 5 0.0 and PPh3 (ca. -6 ppm  26  w.r.t. 85% H3PO4) 78b as external standards. All 31 P NMR shifts are reported relative to 85% H3PO4, with downfield shifts taken as positive. Variable temperature NMR experiments were conducted on the Varian XL300 spectrometer. 1 H and 31 P NMR longitudinal relaxation time (T1) measurements were recorded on the Varian XL300 spectrometer, employing a standard (180 ° -T-90 ° ) pulse sequence. 71 Solid-state magnetic susceptibility was recorded on a Johnson-Matthey magnetic susceptibility balance. Elemental analyses were performed by Mr. P. Borda of this department. Single crystal X-ray diffraction studies were conducted by Dr. Steven J. Rettig of the Departmental Crystallographic Service. Mass spectra were recorded on the Kratos MS 50 spectrometer using an EI source (70 eV), while FAB spectra were recorded on the AEI MS 9 spectrometer using 3-nitrobenzyl alcohol as matrix.  2.3 Synthesis and Characterisation of Ruthenium Complexes  Ruthenium was supplied on loan from Johnson, Matthey Ltd. as RuC13•xH20 (41.50% Ru). All reactions were carried out under an atmosphere of argon employing Schlenk techniques, 81 as most of the ruthenium complexes prepared in the course of this work were air-sensitive in solution and, to a certain extent, in the solid state. The yields reported in the following syntheses are generally the average of several preparations.  27  2.3.1 Ruthenium Precursors, RuCI3(PAr3)2(DMA)•DMA  2.3.1.1 Trichlorobis(triphenylphosphine)(DMA)ruthenium(III)•DMA solvate, RuC13(PPh3)2(DMA)•DMA 82,83 RuC13•xH20 (2.0 g, 8.2 mmol Ru) was stirred with 2 equivalents of PPh3 (4.3 g, 16 4 mmol) in DMA (30 mL) for 24 h at room temperature. The resulting air-stable green product was collected, washed with DMA (2 x 5 mL) and hexanes (2 x 10 mL) and vacuum dried overnight. Yield: 5.4 g (73%). Anal. Calcd for C44H481\1202C13P2Ru: C, 58.32; H, 5.34; N, 3.09. Found: C, 58.14; H, 5.30; N, 3.11. IR (Nujol, cm -1 ): v(C=0) at 1633 (m, uncoordinated DMA); v(c = 0) at 1595 (m, coordinated DMA). The IR data correspond to those in the literature. 82 . 83 A small amount of a crystalline dark red complex (--0.5 g) was isolated from the DMA filtrate after —2 months. Unit cell measurements on one of this crystal by Dr. S. J. Rettig of this Department revealed that complex had the same dimensions as the Ru(III) complex, mer-RuC13(PPh3)(DMA)2•DMA solvate, whose X-ray crystal structure had been determined earlier. 45 IR (Nujol, cm -1 ): v(c = 0) at 1634 (m, uncoordinated DMA); v(c = 0) at 1600, 1587 (m, coordinated DMA). The IR data agreed with those reported earlier. 45 However, a satisifactory elemental analysis was not obtained.  2.3.1.2. Trichlorobis(tri-p-tolylphosphine)(DMA)ruthenium(III)•DMA solvate, RuCl3(P(p-tolyl)3)2(DMA)•DMA 82 . 83 The synthetic procedure used to prepare the title complex was the same as for the triphenylphosphine complex, but used instead two equivalents of P(p-tolyl)3 (5.0 g, 16.4 mmol). 82 . 83 This procedure yielded a bright green solid: 5.32 g (65%). Anal. Calcd for C50H60N2O2C13P2Ru: C, 60.64; H, 6.11; N, 2.83; Cl, 10.74. Found: C, 60.51; H, 6.10; N, 2.83; Cl, 10.88. IR (Nujol, cm -1 ): v(c = 0) at 1644 (m, uncoordinated DMA); v(c = 0) at 1601 (m, coordinated DMA). The IR data agreed with the literature data. 82,83 28  2.3.2. Dichloro-trigt-chloro-bis(bidentate phosphine)diruthenium(II, III) Complexes, [(P-P)C1Ru(.1-C1)3RuCl(P-P)] or Ru2Cl5(p_p)2  84,85  P-P = DPPP, DPPB, DCYPB, DPPPt, DPPH:  A suspension of RuC13(PAr3)2(DMA)•(DMA) solvate (1.0 g, —1 mmol) and one equivalent of the appropriate bidentate phosphine (Ar = phenyl, for P-P = DPPP, DPPB, DCYPB, DPPPt, or DPPH, 1.1 mmol; Ar = p-tolyl for P-P = DPPP, 1.1 mmol) was  refluxed in 150 mL of hexanes under argon for 24 h to 3 days. The resulting brown precipitate in each case was collected on a filter, washed with warm hexanes (20 x 20 mL) to remove any monodentate phosphines and their oxides, and dried under vacuum. Except in the case where P-P = DCYPB, the crude products were subsequently dissolved and washed through the filter as solutions with CH2C12 (-30-300 mL) to remove any insoluble impurities. Concentration of the filtrate to —10 mi., followed by addition of diethyl ether (-40 mL) reprecipitated the products as air-stable, reddish brown powders. In most cases several reprecipitation procedures were generally required to obtain pure products. 2.3.2.1 P P = DPPP: Ru2C15(DPPP)2, 184,85 -  Yield: 0.45 g (74%). Anal. Calcd for C54H52C15P4Ru2: C, 53.86; H, 4.35; Cl, 14.72. Found: C, 53.60; H, 4.39; Cl, 14.95. 2.3.2.2 P P = DPPB: Ru2C15(DPPB)2, 2 84,85 -  Yield: 0.50 g (74%). Anal. Calcd for C561-156C15P4Ru2: C, 54.58; H, 4.58; Cl, 14.37. Found: C, 54.81; H, 4.63; Cl, 14.17.  29  2.3.23 P-P = DCYPB: [Ru2C15(DCYPB)2•H2O] n , 3 In contrast to the other mixed valence dimers, this red powder was insoluble in almost all solvents, and only sparingly soluble in CH2C12 . Purification of this product was accomplished by just washing with CH2C12 (-150 mL). Yield: 0.38 g (54%). Anal. Calcd for C56H104C15P4Ru2•H20: C, 51.79; H, 8.23; Cl, 13.65. Found: C, 51.90; H, 8.21; Cl, 13.56. MS(-FAB): m/z = 1281±2 for Ru2C15(DCYPB)2. A solid-state magnetic susceptibility measurement of this complex yielded a Leff value of 4.3 B.M. Discussion on these data is given in Section 3.3. Concentration of the filtrate to —5 mL followed by addition of 20 mL diethyl ether afforded a brownish yellow precipitate, which was collected on a frit, washed with 2 x 5 mL ether and vacuum dried. Yield: —0.1 g. IR (Nujol, cm -1 ): v(c =0) at 1600 (m, coordinated DMA). Anal. Calcd for an empirical formula of Ru4C19(PPh3)4(DCYPB)(DMA), 4, C104H121C19P6NORu4: C, 54.07; H, 5.28; N, 0.61; Cl, 13.81. Found: C, 54.06; H, 5.00; N, 0.48; Cl, 14.95. 2.3.2.4 P-P = DPPPt: Ru2C15(DPPPt)2, 5 85 Yield: 0.27 g (38%). Anal. Calcd for C58H60C15P4Ru2: C, 55.27; H, 4.80. Found: C, 55.73; H, 4.57. 2.3.2.5 P-P = DPPH: Ru2C15(DPPH)2, 6 85 Yield: 0.14 g (20%). A satisfactory elemental analysis was not obtained; e.g. Anal. Calcd for C601164C15P4Ru2: C, 55.93; H, 5.01. Found: C, 57.38; H, 4.67.  30  2.3.3 Dichloro-di-wchloro-bis(bidentate phosphine)diruthenium(II) Complexes, [RuCl(P-P)(µ-C1)]2  The dinuclear complexes, [RuCl(P-P)2(1-C1)]2, were prepared following the general procedures developed by I. S. Thorburn and A. M. Joshi in this laboratory in which the Ru(II, III)-bidentate phosphine complexes, Ru2C15(P-P)2 (Section 2.3.2) were reduced by H2 in the presence of an appropriate base (DMA or polyvinylpyridine).  85  The  complexes prepared were air-sensitive in both solution and solid state, as evidenced by the orange-brown colour turning greenish black upon exposure of the complex to air (within a few minutes in solution and after an hour in the solid state). As a result, these complexes were stored under argon in Schlenk tubes. (a) Syntheses using DMA for P-P = DPPP, DPPB, DPPPt Ru2C15(P-P)2 (0.5 g, —0.5 mmol) was stirred in 20 mL DMA under 1 atm H2 for 24 h. The resulting dark brown solution was concentrated to 5 mL, dry methanol (40 mL) was added, and the resulting orange yellow suspension stirred overnight under H2. The product was filtered off, washed with 2 x 5 mL methanol and 2 x 5 mL diethyl ether, and vacuum dried. In most cases, several reprecipitation procedures were generally required in efforts to obtain pure products. (b) Synthesis with polyvinylpyridine for P-P = DPPP Ru2C15(P-P)2 (0.2 g, — 0.2 mmol) was added to a rigorously deoxygenated benzene suspension (20 mL) of PVP (1.0 g) and the mixture stirred under 1 atm H2 for 24 h. The brown suspension was filtered to remove the insoluble pyridine polymer as well as  unreacted Ru2C15(P-P)2 precursor, and the resulting orange-brown filtrate was concentrated to —1 mL. Addition of dry hexanes (20 mL) followed by stirring under 1 atm  31  H2 for a few hours yielded a brown product which was filtered, washed with 2 x 5 mL hexanes and vacuum dried. 2.3.3.1 P P = DPPP: [RuCl(DPPP)(,u-C1)]2, 7 85 (half-scale) -  In synthesizing complex 7 from the DMA/H2 route it was often necessary to use small amounts of Me0H (5 mL) and low temperature (-30°C) to bring about precipitation of the product from the DMA mother liquor. The complex is hydroscopic in the solid state ( 1 H NMR evidence). Yield: 0.13 g (65%). Anal. Calcd for Ru2C14(DPPP)2•1120, C54H52C14P4Ru•H20: C, 54.65; H, 4.59. Found C, 54.17; H, 4.72; N, 0.49. The N content arises from the DMA "impurity". Attempts to synthesize Ru2C14(DPPP)2 from Ru2C15(DPPP)2, 1 and H2 using PVP as an alternative base proved to be unsuccessful, due to the insolubility of the precursor, 1, in benzene. Most of the starting material remained undissolved even with 100 mL of benzene used, and a very low yield (< 5%) was recorded. Consequently, the DMA/H2 route was preferred. After one synthesis of 7, a small amount (— 0.1 g) of a previously unknown, crystalline black compound was isolated from the DMA/Me0H/Et20 filtrate after — 2 months. A single crystal X-ray diffraction study by Dr. S. J. Rettig of this Department revealed the complex to be an unsymmetrical trinuclear Ru cluster, Ru3C17(PPh3)3(DPPP)(H20), 8. The crystal structural data, including tables of all bond lengths and bond angles, and the ORTEP stereoview of the molecular structure, are listed in Appendix (A-1). Discussion on this complex is given in Section 3.4.1.1. 2.3.3.2 P P = DPPB: [RuCl(DPPB)(1 C1)]2, 9 85 -  -  Yield: 0.36 g (72%). Anal. Calcd for C56H56C14P4Ru2: C, 56.20; H, 4.72; Cl, 11.85. Found: C, 56.27; H, 4.93; Cl, 11.70.  32  2.3.3.3 P P = DPPPt: Attempted synthesis of [RuCl(DPPPt)(t-C1)]2, 10 85 -  The procedure attempted for the reduction of Ru2C15(DPPPt)2, 5, to the Ru(II, II) dimer was the same as for the DPPB complex. However, five attempts to repeat the preparation failed to yield the desired dimer. Rationalization for the failed synthesis is given in Chapter 3. 2.3.4 P P = DCYPB: Attempted synthesis of [RuCl(DCYPB)(1.1,-C1)12 -  (a) in DMA solvent Due to the insolubility of the starting material in most organic solvents, attempts to synthesise the title complex were unsuccessful. Stirring of the mixed valence precursor, 3, in DMA (50 mL) under 1 atm of H2 for 10 days yielded a pale orange solution with the precursor remaining essentially undissolved. The mixture was filtered to remove the solids and the filtrate was concentrated to —1 mL. Addition of 10 nil, methanol did not precipitate any solid products, but turned the solution from brown to green. Evaporation of the solution to dryness regenerated a reddish brown residue, which was dissolved in —40 mL of CH2C12. Concentration of the solution to —1 mL and addition of 10 mL diethyl ether precipitated a reddish brown solid. The product was subsequently washed with small amounts of diethyl ether (5 mL total) and vacuum dried. Yield 0.007 g. IR. (Nujol, cm - 1 ): v(c = 0) at 1641 (m, uncoordinated DMA) and 1939 (m) suggested the possible presence of a Ru-H bond. Elemental analysis is consistent with the formula Ru4C110(DCYPB)3•DMA solvate, 11, C881-1165C110P6NORu4: C, 48.09; H, 7.57; N, 0.64. Found: C, 48.05; H, 7.67; N, 0.61. Some NMR data and discussion on the possible nature of complex 11 are given in Section 3.6.1. (b) in CH2C12 solvent The starting mixed valence precursor, 3, was stirred in CH2C12 (100 mL) under 1 atm of H2 for 1 week to yield a dark brown suspension. The mixture was filtered to give 33  an orange-brown solid, and a dark brown filtrate. The solid was washed with 2 x 10 mL CH2C12 and vacuum dried. Yield: — 0.1 g. Elemental analysis is consistent with the formula Ru4C110(DCYPB)3 12 r..._84_156aloP6Ru4: C, 47.80; H, 7.45; Cl, 16.80. ,  ,  Found: C, 48.02; H, 7.33; Cl, 17.08. The filtrate was concentrated to — 1 mL, and addition of 5 mL diethyl ether precipitated out a dark brown solid, 13, which was washed with 2 x 10 mL Et20 and vacuum dried. Yield: — 0.01 g. However, a satisfactory elemental analysis of this compound corresponding to the dimer formulation Ru2C14(DCYPB)2 was not obtained (e.g. Anal. Calcd for C56H104C14P4Ru2: C, 53.93; H, 8.57; Cl, 11.85. Found: C, 50.90; H, 8.05). Some NMR data and discussion of these results are given in Section 3.6.2.  34  CHAPTER 3 Synthesis and Characterisation of Chloro-Bridged Diruthenium Complexes Containing Chelating Ditertiary Phosphines and their Role in Activation of Hydrogen  3.1 Introduction - Review of [RuCl(P-P)(µ-C1)]2 Complexes  Coordination chemistry involving ruthenium has been developed to a great extent over the last 30 years. The synthesis, reactivity, and catalytic applications of ruthenium complexes containing tertiary phosphines are well documented in the literature.  86-88  One  such intriguing example is the ruthenium hydrido complex RuHC1(PPh3)3; its ability to catalytically hydrogenate olefins is often comparable to that of the prominent Wilkinson catalyst, RhCl(PPh3)3. 8,10 Use of chiral phosphine ligands, which are essential in asymmetric syntheses, has also been considered. 19,22-24 Work on ruthenium complexes containing chelating diphosphines, which often provide better stereochemical control than the monodentate analogues, was first reported by Chatt and Hayter. 89 A series of complexes with general formulae Ru 11X2(P-P)2 and Ru 11XY(P-P)2 (X, Y = halides, pseudohalides, hydrides, or a-bonded alkyl and aryl groups; P-P = chelating diphosphines) was later synthesised by simple displacement of monodentate phosphine from RuIIXY(PR3)3 precursors, with chelating diphosphines. 88 However, this route only yielded stable six-coordinate complexes RuXY (P-P)288 or fivecoordinate mixed phosphine complexes of the form RuXY(P-P)(PR3). 85 Mechanistic studies on asymmetric hydrogenation of prochiral alkenes by ruthenium hydrides containing chelating phosphines (e.g. trans-RuHC1(BINAP)2 9 ° and trans35  RuHC1(DIOP)2 91 .92 ) have suggested that the catalytically active species probably have the form "RuHC1(P-P)", with one diphosphine per Ru. Generation of coordinatively unsaturated "Rull(P-P)" species with one diphosphine per Ru is desirable in order to understand better the above catalytic systems. The synthesis of coordinatively unsaturated d6 , Run complexes is also a preferred prerequisite for activation of dihydrogen.  38  In order to synthesize the target "RuHC1(P-P)" species, Thorburn, previously of this laboratory, developed a general scheme for the formation of dimeric [Ru 11 C12(P-P)]2 complexes 84 .85 (Scheme 3.1). These complexes should, in principle, yield the corresponding, required monohydride by reaction with H2. DMA 20 °C, 24 h R = phenyl or p-tolyl DMA = N,N-ditrethylacetamide  RuC13 .xH2 O + 2PR3  [RuCl(P -P)(µ--C1)} 2  RuC13 (PR3 ) 2DMA.DMA solvate 1 eq. P-P / Rum hexanes, 24 h, reflux  1 atm H2, 24 h, 20 0 C  44^  ,Cli^Cl . 0^i„ ..  Cl ,  PII. ""  BaseH+Cr^Base  Cl  \  C17  ^'fin  Base = DMA for P-P = DPPP, DPPB and DPPPt Base = polyvinylpyridine for P-P = CHIRAPHOS, DIOP, BINAP and S,S-BDPP Scheme 3.1: Synthetic route to dinuclear ruthenium(II) complexes with one chelating diphosphine per Ru. Through early attempts to synthesize RuCl3(P-P) complexes by phosphine displacement from RuCl3(PR3)2, Thorburn discovered the new, formally mixed-valence dinuclear Ru(II,III) compounds with the formulation Ru2C15(p_p\-)284(Equation 3.1).  36  4 RuC13(PR3)2 + 4 P-P  hexanestrace , 1120 reflux  2 Ru2C15(P-P)2 + 7 PR 3 + R3P=0 + 2 HC1  R = phenyl or p-tolyl^(3.1) P-P = DPPP, DPPB, DPPPt, DPPH, DIOP, CHIRAPHOS, NORPHOS, BINAP, DPPCP, DPCYCP, PHENOP, S,S-BDPP The X-ray diffraction analysis of one of the mixed-valence dimers (P-P = S,SCHIRAPHOS) revealed the complex to be a highly symmetric trichloro-bridged species, formulated as [(P-P)C1Ru(1.1-C1)3RuCl(P-P)], with somewhat irregular octahedral geometry about each ruthenium centre (see Scheme 3.1). 84 Reduction of the Ru2C15(PP)2 complexes with H2 in the presence of an appropriate base then generated the desired "RAP-Pr compounds as dimers. Thorburn's work, which was later elaborated on by Joshi, also from this laboratory, forms the basic synthetic route of the present study. Scheme 3.1 summarises the synthetic procedure leading to the dimeric [RuCl(P-P)(µ-C1)]2 species. It should be noted that these "Rull(P-P)" complexes have been generated also via Rull(arene), Ruil(diene) or Run(T 3 -ally1) precursors. 85 Prior to this thesis work, there had been twelve Ru2C15(P-P)2 complexes prepared containing different diphosphine ligands (Figure 3.1, except DCYPB). 85 These complexes had been isolated and characterised by various spectroscopic and analytical techniques. 84,85 Only seven reduced products had been isolated as the [RuCl(P-P)(11-C1)]2 species (Scheme 3.1, Figure 3.2). Thorburn noted that while the CHIRAPHOS complex was successfully synthesised by this route, the analogous nonchiral complex with DPPE as ligand could not be synthesised. 84 This finding is somewhat anomalous as the only difference between the two ligands is the hydrogens instead of methyl groups on the carbon backbone.  37  PPh2 PPh2  Ph2  H Mel*  PPh2  PPh2  Ph2 CHIRAPHOS NORPHOS  DIOP  PhCH PPh2 PPh2 BINAP Me  \O I Ph2P^pph2 Ar = phenyl, DPPCP;^PHENOP Ar = cyclohexyl, DPCYCP  r  ^aPo aPt  Ph2P PPh2 S,S- BDPP^DCYPB  Ph2P (CH2 )nPPh2 n = 3, DPPP; 4, DPPB; 5, DPPPt; 6, DPPH  Figure 3.1 Diphosphines used successfully in the preparation of Ru2C15(P-P)2 complexes The structure of the [RuCl(P-P)(µ-C1)]2 complexes (Figure 3.2) resembles that of the analogous complexes with monodentate phosphines (PPh3 and P(p-toly1) 3),82,93,94 which consist of two five-coordinate ruthenium (II) centres in square pyramidal geometry with two edge-bridging chlorides. Except for P-P = BINAP and CHIRAPHOS, the 31 1){ 1 H}  NMR spectra (121.42 MHz, CD2C12, 20°C) of the [RuCl(P-P)(1.1.-C1))2  complexes 85 show a single AB pattern, suggesting the dimers possess C2 symmetry. Such symmetry makes P A (Rul) and P A (Ru 2 ) chemically and magnetically equivalent, and this holds also for P B (Figure 3.2a). The BINAP complexes possibly have a different geometry in which one chelating phosphine occupies apical/basal positions (PA/PB), while the other takes up two cis-positions in the basal plane (P c/PD , Figure 3.2b), giving rise to 38  two AB quartets of equal integral intensity in the  31 P{ 1 H)  NMR spectrum. The  assignment of structure b in Figure 3.2 to the BINAP system is based on chemical shift data. 85 The CHIRAPHOS complex also exhibits two AB patterns with equal integral intensity because of the chirality in the CHIRAPHOS backbone. However, the S- and RDIOP complexes show only one AB pattern, even with the chiral backbone, which may be due to coincidental degeneracy. 85  Cl C I^ A.Th C1)„,^io'Cl"..^I^ Cl„... '' 1 ...„0.0C1".....,,iti....0.1111 '''' 13 ^Rti^le/ 2 s Rii,, ■ sippi p^ 1^..%(21 ... 211 '^ Cl^BP^1  B^...........,"  A  ...........•PA  a  b  Figure 3.2 Suggested geometry for [RuCl(P-P)(1.1.-C1)]2 complexes with a) P-P = DPPP, DPPB, DPPPt, DIOP, CHIRAPHOS, S,S,-BDPP; and b) P-P = BINAP  Of particular interest, the [RuCl(P-P)(µ-C1)]2 complexes (P-P = CHIRAPHOS and DIOP) are efficient catalysts for asymmetric hydrogenation of prochiral olefinic acids under mild conditions (30°C, DMA solvent, 1 atm H2). 75 . 85 Further, [RuCl(DPPB)(p.Cl)]2 also catalyses the hydrogenation of various alkenes, ketones and imines under the same condition S. 45 In the presence of a coordinating ligand, L, [RuCl(DPPB)(.t-Cl)]2 forms adducts of the type Ru2C14(P-P)2(L), where L coordinates to one of the metal centres to form a triply chloro-bridged diruthenium species (Equation 3.2). The X-ray crystal structure of the Ru2C14(DPPB)2(DMS0) complex 85 reveals a geometry similar to that of the mixed valence complex Ru2C15(CHIRAPHOS)2, the only difference being the unsymmetrical positioning of the DPPB ligands in the DMSO complex (cf. Scheme 3.1).  39  L^,C1,, ... ,... 0 '^' i..„ .. ill......ti—ma '="--:Ru ^ iffilpF li -L^,^sC1 2 a L  ...  (3.2) P-P = DPPB; L = (CH3)2CO, CO, (CH3)2S0, PhCN, Mel, DMA, NEt 3 , NHBu 2 , 71 2-H2 , a-N2 In an attempt to generate a hydrido-chloro complex from reaction of [RuC1(DPPB)(11-C1)}2, 9, with 1 atm of H2, Joshi discovered that, without an added base, a reversible in situ formation of a molecular hydrogen complex (r) 2 -H2)Ru(DPPB)(1.C1)3RuCl(DPPB), 69 9a, occurs (Equation 3.3). Complex 9 also reacts reversibly with 1 atm of N2 to give the corresponding dinitrogen complex, 9b, with the N2 ligand bound in the usual end-on fashion. 69 Only four other dinuclear r12-H2 complexes,48d,53,57,82 all containing ruthenium, have been reported in the literature.  Pik-"---% Clt ,^oiCli.,^I^. ,op^H2 1 1 , ..^1 • ...I' s^"s..^.."^B ^ 1f* Id -H2 1313^1 'Cl^2^% Cl P-P = DPPB ./IIA 9^  01  „ C l„. ..  R11 i ip ,../ 2Nsi F /^ ..C1^a  1 l l I , ...... Rd I  ■IMIN  C  11.NENIM.  9a  ........  . I • ......  (3.3)  It should be pointed out that in the presence of NEt3 as an added base, the reaction of the [RuCl(P-P)(11-C1)]2 complexes (P-P = DPPB or CHIRAPHOS) with H2 generates a trinuclear Ru(II) hydride complex [RuHC1(P-P)]3. 45 '75 The formation of such trimers are thought to proceed via deprotonation of the initially formed molecular hydrogen complexes (e.g. 9a) by NEt3. Subsequent reactions of the intermediate with H2 40  and further deprotonation then afford the trinuclear species. Interestingly, when under a hydrogen atmosphere, [RuHC1(DPPB)]3, converts to another molecular hydrogen species [( .1 2 -H2)Ru(DPPB)(11-H) (1-C1)2RuHC1(DPPB)]  .  45  3.2 Present Work  The preparation of [RuCl(P-P)(µ-C1)12 complexes, and their subsequent reactions with H2 in some cases, have been studied using DPPP, DPPB, DPPPt, DPPH, and DCYPB ligands. Section 3.3 first discusses the preparation of the precursor Ru2C15(PP)2 complexes, while Sections 3.4 and 3.6 discuss the successful and unsuccessful preparations of [RuCl(P-P)(1.1-C1)]2 complexes. Section 3.5 describes the interaction of [RuCl(DPPP)(.t-Cl)]2, 7, with H2 and N2.  3.3 Ru2C15(P P)2 Complexes -  The formally mixed valence dinuclear complexes, Ru2C15(P-P)2, can be prepared by refluxing a hexanes suspension of RuC13(PR3)2DMA (R = phenyl or p-tolyl) with an equivalent amount of the appropriate diphosphine. 85 The resulting Ru2C15(P-P)2 complexes (DPPP, 1; DPPB, 2; DCYPB, 3; DPPPt, 5; DPPH, 6) are all air-stable, reddish-brown powders. In the case of complexes 1, 2, 5 and 6, the presence of free monophosphines and DMA often poses difficulties in obtaining analytically pure products. Large volumes of warm hexanes are needed to remove the phosphines and several reprecipitations with CH2C12/Et20 are generally required to eliminate DMA impurities. Before purification of complex 1, mixed phosphine complexes, possibly in the form of Ru2C15(DPPP)(PR3)2 (R = phenyl or p-tolyl), were present (— 20%). This 41  contamination is evidenced by the presence of a sharp singlet (8 = 2.38 ppm, CDC13) in the 1 H NMR spectrum, indicative of the methyl resonance of coordinated P(p-tolyl)3, while no free phosphine signal was detected in any of the  31 P{ 1 H)NMR spectra (CDC13:  8 = -5.6 ppm, s, free PPh3; 8 = -7.8 ppm, s, free P(p-tolyl)3 ; 8 = -17.7 ppm, s, free  DPPP). Prolonged refluxing (3 days) did not result in complete reaction. Consequently, analytically pure product was sometimes difficult to obtain. Use of CH3I as a phosphine sponge to remove dissociated monodentate phosphine may assist in driving the exchange reaction to completion, and this should be investigated. The reaction of RuC13(PPh3)2DMA with DCYPB in hexanes results in the production of a bright reddish brown solid which is very insoluble in all organic solvents tried. Characterisation is limited to solid state measurements, due to its low solubility. While elemental analysis suggests the empirical formula Ru2C15(DCYPB)2•H20, 3, mass spectroscopy (- FAB) indicates the presence of Ru2C15(DCYPB)2 (Section 2.3.2.3). In either case, a dinuclear species is implied. However, the solid-state magnetic susceptibility of this complex yields a value of 4.3 B.M. for a Ru2C15(DCYPB)2 formulation, suggesting three unpaired electrons per complex. The data perhaps indicate a multinuclear, or polymeric species with the empirical formula Ru2C15(DCYPB)2. Increasing the steric congestion about the phosphorus centre may induce oligomerisation or polymerisation via .t2-diphosphines, rather than the diphosphine chelating to the same Ru centre. This might explain the insolubility of the product. Further evidence is required in order to fully characterise this complex. A minor product from this reaction was isolated as a yellowish-brown precipitate from the CH2C12 liquor following the purification procedure (Section 2.3.2.3). The  mixed phosphine formulation of Ru4C19(PPh3)4(DCYPB)(DMA), 4, was supported by the presence of broad phenyl (8 — 7 - 8 ppm), methylene (8 — 0.5 - 3.5 ppm) and DMA resonances (8 = 2.05, 2.92, 2.98 ppm, s) in the 1 H NMR spectrum (CDC13, 20°C), and by elemental analysis for C, H, N, and Cl. The integrated intensity ratios (— 1 : 1) between 42  the phenyl and methylene plus DMA resonances (60 : 61) is equal to that expected for formulation 4. The ratio of Ru : P is 1 : 1.5 instead of 1 : 2 as in the case of the mixed valence dimers, and the average Ru oxidation state is 2.25 instead of 2.5. The isolation of 4 again supports the tendency of complexes with the DCYPB ligand to form clusters  rather than dinuclear compounds. A relatively large volume (>200 mL) of dichloromethane was needed during the reprecipitation procedure for complexes 5 and 6 due to their low solubilities in CH2C12 and other common solvents. Such low solubility is perhaps a result of the increased alkyl chain length of the phosphine backbone, which may favor the formation of bridged phosphine complexes where the P-P ligands are coordinated to two different Ru centres. Even after several reprecipitation procedures, the products obtained were often analytically impure; some marginally acceptable analytical data were obtained for 5 but not 6 (Sections 2.3.2.4 and 2.3.2.5). Attempts were made to purify these complexes by  column chromatography using silica gel and CH2C12JMeOH but without success. Previous workers from this laboratory obtained 'marginal' analytical data for both 5 and 6. 85  3.4 [RuCl(P P)(.t Cl)]2 Complexes -  -  The reaction of the mixed valence Ru2C15(P-P)2 complexes in DMA with H2 generates the ionic species [Ru2C15(P-P)2] - DMAH+ (Equation 3.4). 75,84 From these, the neutral [RuCl(P-P)(11-C1)12 complexes are isolated by precipitation with Me0H, which causes displacement of DMA•HC1 and precipitation of the product (P-P = DPPP, 75,85 7; DPPB, 75,85 9; DPPPt, 85 10). The neutral dimers can also be prepared by reacting Ru2C15(P-P)2 with H2 in benzene or toluene, using polyvinylpyridine (PVP) as the base. Both PVP and the generated PVP•HC1 salt are insoluble and can be removed by filtration, 43  allowing the required product to be isolated from the filtrate easily (Section 2.3.3). All the complexes produced are hygroscopic and air-sensitive in both solution and solid-state, turning from orange-brown to a green color quite readily when exposed to air. " DMA Ru 2 a 5 (P-P) 2 + 0.5 1 12 [Ru2C15(P-P)2J— DMAH + (3.4) -  3.4.1 [RuCI(DPPP)(11-C1)]2•H20  The high solubility of [RuCl(DPPP)(g-C1))2, 7, in methanol makes it difficult to prepare it from the DMA/H2 route, although it has been made successfully by this route by previous workers. 85 The use of small volumes of the solvent at low temperature (-30° C) is often necessary to bring about precipitation. Even so, in this work the product is generally obtained in low yield (— 50%) and is often analytically impure, with DMA and [Ru2C15(DPPP)2] -DMAH+ as the major impurities (for the latter, in the  31 P{ 1 H)  NMR  spectrum, 8 = 47.6 ppm, s, C6D6 or 40.7 ppm, s, CD2C12; cf. data for the analogous RuDPPB anion: 50.9 ppm, s, C7D8). 95 The alternative PVP/H2 route seemed promising, because the by-product PVP•HC1 can be easily removed, methanol is not required for precipitation and no DMA is used. However, the Ru2C15(DPPP)2 precursor is only sparingly soluble in benzene or toluene, and subsequently even lower yields were obtained for this alternate synthetic route. The 31 P{ 1 H) NMR spectrum of 7 in CD2C12 or C6D6 at ambient temperature shows a very broad signal centred at 56 ppm, in contrast to the AB quartet characteristic of such [RuCl(P-P)(11-C1)]2 complexes. When the solution is cooled to -90°C, however, the spectrum exhibits an AB pattern for the major species (Figure 3.3c, Table 3.3 on p.54), with chemical shifts and 2JAB coupling constant in agreement with the values reported. 85 The room temperature spectrum can be best explained by invoking a rapid 44  Ji11 1 1111  ; 11.1 1 11 ;;IIIIII , Ill i . ,1 4 1 m1 1111 i;m1) 1 11111 , 41111 , 1111. i milli,TiNJI01,11,1JJ1.11J111:tifil li 60^58^56^54^52^50^48^46^44^42 PPM 40 Figure 3.3 31 p( 1 H) NMR spectrum of [RuCl(DPPP)(g-C1)]2, 7, at a) 20°C; b) -80°C and c) -90°C (121.42 MHz, CD2Cl2). The resonances marked with asterisks in (b) are tentatively assigned to f(DMA)Ru(DPPP)(1-C1)3RuC1(DPPP)] complex. 45  equilibrium between the dichloro-bridged complex 7 and a triply chloro-bridged Ru2CI4(DPPP)2(DMA) species formed with DMA coordinated to either one of the ruthenium centres (cf. Equation 3.2). This ligand association/dissociation process at 20°C must be fast on the NMR time-scale, and must involve both ruthenium centres in complex 7, which results in a scrambling of all four phosphorus centres. If the exchange involves  only one ruthenium centre (Ru 1 ), then scrambling will only occur on the Pc and PD resonances, assuming that the amide can re-coordinate at any one of the three nonbridging sites, but this will not affect the PE and PF resonances. In fact at -80 ° C, the phosphorus spectrum shows in addition to the AB pattern from [RuCl(DPPP)(.4,-C1)}2, two other AB patterns (8c = 59.4 ppm, OD = 49.6 ppm, 2Jcp, = 56.5 Hz; SE = 53.0 ppm, 8F = 51.9 ppm, 2JEF = 59.5 Hz, cf. Equation 3.2) which are tentatively assigned to the triply chloro-bridged species RDMA)Ru(DPPP)(1-C1)3RuC1(DPPP)]; the data for the corresponding DPPB species have been reported. 85  3.4.1.1 Molecular Structure of the Ru3C17(PPh3)3(DPPP)(H20) Cluster, li  The single crystal X-ray diffraction study of 8, which slowly formed in the filtrate from one of the syntheses of 7, was carried out by Dr. S. J. Rettig of this Department, and revealed the complex to be the unsymmetrical trinuclear Ru cluster, Ru3C17(PPh3)3(DPPP)(H20), in which a formal oxidation states of II, II and III are assigned to the ruthenium centres (Figure 3.4). The structural parameters are listed in the Appendix (A-1), and some selected bond distances and bond angles are listed in Tables 3.1 and 3.2 respectively.  46  C4  Figure 3.4 Molecular structure of [Ru3C17(PPh3)3(DPPP)(H20)], 8, with the atom numbering scheme used. All the hydrogen atoms and all atoms of the phenyl rings except the a-carbon have been omitted for clarity. Table 3.1 Selected bond lengths (A) for [Ru3C17(PPh3)3(DPPP)(H20)], 8, with estimated standard deviations in parentheses. Bond  Length (A)  Bond  Length (A)  Ru(1)—C1(1) Ru(1)--C1(2) Ru(1)--C1(3) Ru(1)—C1(5) Ru(2)—Cl(1) Ru(2)—C1(2) Ru(2)—C1(3) Ru(2)--C1(4) Ru(3)--C1(1)  2.566(2) 2.489(2) 2.433(2) 2.400(2) 2.512(2) 2.480(2) 2.412(2) 2.400(2) 2.591(2)  Ru(3)--C1(4) Ru(3)--C1(6) Ru(3)—Cl(7) Ru(3)—O(1) Ru(1)—P(1) Ru(1)—P(2) Ru(2)—P(3) Ru(2)—P(4) Ru(3)—P(5)  2.364(2) 2.291(2) 2.316(2) 2.126(4) 2.241(2) 2.269(2) 2.319(2) 2.319(2) 2.280(2)  47  Table 3.2 Selected bond angles (°) for [Ru3C17(PPh3)3(DPPP)(H20)], 8, with estimated standard deviations in parentheses. Bonds C1(1)--Ru(1)--C1(3) C1(2)--Ru(1)--C1(3) C1(3)--Ru(1)--C1(5) C1(3)—Ru(1)—P(1) C1(3)--Ru(1)—P(2) C1(1 )--Ru (2)--C1(2) C1(1)--Ru(2)--C1(3) C1(1)--Ru(2)--C1(4) C1(1)--Ru(2)—P(3)  Angle (°)  Bonds C1(2)--Ru(2)--C1(3) C1(2)—Ru(2)----P(4) C1(3)--Ru(2)—C1(4) C1(3)--Ru(2)---P(4) C1(7)--Ru(3) --0(1) Ru(1)--C1(1)--Ru(2) Ru(1)—C1(1)--Ru(3) Ru(2)--C1(4)--Ru(3) C1(4)--Ru(3)--C1(7)  79.12(5) 78.78(5) 164.19(6) 100.70(6) 101.77(6) 81.34(5) 80.59(5) 82.63(6) 166.76(6)  Angle (°) 79.37(5) 168.50(6) 162.22(6) 103.87(6) 82.4(1) 81.95(5) 127.61(7) 101.92(6) 168.76(7)  The details of the formation of 8 are uncertain, but it probably involves the combination of the mixed phosphine complex [(Ph3P)2C1Ru(2)(p.-Cl)2Ru(1)C1(DPPP)], produced in a side-reaction during the reduction procedure of an impure sample of 1 with DMA/H2, with a residual RuIII(3) monomer. As mentioned in Section 3.3, the mixed phosphine complexes [(PR3)2C1Ru(p.-C1)3RuC1(P-P)] possibly form during the synthesis of the mixed valence Ru2C15(P-P)2 complexes, and separation of the two complexes can sometimes be difficult. Presumably the sample of 1 in one case contained some RuC13(PPh3)2(DMA) precursor. The structure of 8 contains three Ru centres each with distorted octahedral geometry. There are three doubly bridging chlorides between Ru(1) and Ru(2), while C1(4) just bridges Ru(2) and Ru(3), and CI(1) triply bridges all three Ru atoms. The bridging chlorides between all Ru atoms are roughly symmetric with the Ru—CI distances ranging from 2.364(2) to 2.591(2) A. The three relatively short Ru(3)—Cl distances (2.291(2) to 2.364(2) A) lead to the assignment of a formal oxidation state of III to Ru(3),  48  while the other two Ru centres are assigned formal oxidation states of II. The Ru—P distances (average = 2.29 A) are comparable to those in other Ru(II) and Ru(II,III)phosphine complexes. 75,84,85 The Ru--C1 distances for the bridging chlorides which are trans  to phosphorus (average = 2.53 A) are longer than the Ru(1)--C1(3) (2.433(2) A)  and Ru(3)--C1(4) bond lengths (2.364(2) A) of the chlorides  trans  to the terminal  chlorides. This illustrates the stronger trans influence of phosphine versus chloride, resulting in a somewhat weaker, longer bond. Of interest, Cl(2) and Cl(5) are also hydrogen-bonded to the hydrate molecule on Ru(3) (see Figure 3.4).  3.4.2 [RuCl(DPPP0(.1 C1)]2 -  Previous work from this laboratory has reported on the synthesis of [RuC1(DPPPt)(11-C1)12, 10, using the same procedure as for the DPPP and DPPB analogues. 85 Several attempts were made throughout the present work to prepare this complex but without success. The 31 P{ 1 H} NMR spectrum (CD2C12, 20°C) of the product isolated from the reduction of 5 in DMA/H2 shows a very broad signal centred at 45 ppm, which remains unresolved even down to -80°C. Both the 1 H NMR spectrum of the same solution, and elemental analysis of the isolated product, indicate a significant amount of DMA present. Thus, scrambling of all phosphorus signals by DMA exchange may perhaps explain the appearance of the 31 P spectrum (see Section 3.4.1). It should be pointed out that difficulties in obtaining the precursor 5 analytically pure may also effect the clean reduction with DMA/H2.  49  3.5 Interaction of [RuCl(DPPP)(.t-Cl)]2,  2, with H2 and N2  3.5.1 Formation of a Molecular Hydrogen Complex by Interaction of  I with H2  Without an added base, fRuCl(DPPP)(11-C1)J2, 7, reacts reversibly with 1 atm H2 at ambient temperature in benzene, toluene, or methylene chloride solutions to form the molecular hydrogen complex 7a (Equation 3.5), identified by 1 H and 31 P{ 1 H) NMR spectroscopy (Figures 3.5 and 3.6). [RUCKDPPP)(11- CI)]2 + H2^111' [012 -H2XDPPP)RU(11-0)3RUCKDPPPA (3.5)  7^  7a  The 1 H and 31 P NMR spectra suggest ca. 95% equilibrium conversion of 7 to 7a under one atmosphere of H2 (CD2C12, 20°C) . The 1 H NMR spectrum shows a broad resonance at -10.90 ppm (Figure 3.5), which disappears when the H2 atmosphere is removed by evacuation, or replaced by argon. The T1 relaxation time of 19 ms (300 MHz, CD2C12, 20°C, w1/2 = 36 Hz) for this resonance is short, and is typical for an 71 2 31 P { 1 H) NMR spectrum (Figure 3.6) displays two independent AB-H2moiety.Th  patterns with equal integral intensity which are assigned to the T1 2 -H2 complex, 7a. The resonance signal for 7 appears as a very broad peak centred at 55.5 ppm due to scrambling of the phosphorus nuclei of 7 by DMA impurity (Section 3.4.1). The singlet at 40.7 ppm, which shifts to 47.6 ppm in C6D6 solvent, is assigned to the ionic species [Ru2C15(DPPP)2] - DMAH+ present initially in 7 (Section 3.4.1.). The two AB patterns, along with characteristic chemical shifts and 2Jpp coupling constants, imply a geometry similar to that of 9a (cf. Equation 3.3); the geometry is consistent with that of other related triply chloro-bridged complexes of the form [(L)Ru(DPPB)(4-C1)3RuC1(DPPB)] 50  4^2^0^—2^—4^—6^—8^—10 PPM-12 Figure 3.5 1 H NMR spectrum of [(11 2 H2)(DPPP)Ru(.1 C1)3RuCI(DPPP)], 7a, formed in situ by reacting [RuCl(DPPP)(it C1)12, 7, -  -  under 1 atm H2 pressure (300 MHz, CD2Cl2, 20°C).  -  56^54^52^50^48^46^44^42^40 PPM 38 Figure 3.6 31 P( 1 H ) NMR spectrum of [(t1 2 -H2)(DPPP)Ru(p,-C1)3RuC1(DPPP)], 7a, formed in situ by reacting [RuCl(DPPP)(i-C1)12, 7, under 1 atm H2 pressure (121.42 MHz, CD2Q2, 20°C); the broad signal at — 56 ppm is that of 7.  (cf. Equation 3.2). 85 A comparison of the spectroscopic data between 9a and 7a is  summarised in Table 3.3. The presence of two sharp AB patterns in the 31 P{ 1 H) NMR spectrum of 7a indicates that exchange between bound and free H2 is relatively slow on the NMR time scale. Assuming no intra- or inter-molecular hydrogen exchange, the broadness of the r1 2 H2 resonance of 7a is likely attributable to rapid internal rotation of the 1 2 -H2 lig and . 35,37,38,72 Th e T1 2-H 2  signal broadens further with decrease in temperature (Table  3.4), which implies a decrease of the internal rotation of the 1 2 -H2 ligand consistent with the behaviour of many other molecular hydrogen complexes. 35,37 .38 .72 A variable temperature T1 relaxation experiment was conducted for the r1 2 -H2 resonance of 7a from 193 to 296 K (Table 3.4). The plot of In T1 versus 1/0 generates a sharp V-shaped curve (Figure 3.7) as predicted by Equation 1.6 (p.19), and yields a minimum T1 value of 12 ms at 233 K (0( mi n)). From this, an internuclear H—H distance of 1.08 x C A is calculated, where C is a correction factor for the r1 2 -H2 rotation (see section 1.4.5b). Assuming the 1 2 -H2 rotation is rapid, a value of C = 0.794 is used as suggested by Morris's group. 72 Thus, an H—H distance of 0.86 ± 0.03 A is estimated. This value is exactly the same as the one calculated for 7a, 69 and is also comparable with that of other dinuclear and mononuclear Ru molecular hydrogen (cf. Table 1.3, p.20).  53  COMPleXeS 35 . 3738 •53 '5732  Table 3.3 31 P{ 1 H} NMR data (121.42 MHz, 20°C) for the complexes {RuCl(P-P)(1.1C1)]2 (P-P = DPPB, 9; DPPP, 7), and [(L)Ru(P-P)(1-C1)3RuCl(P-P)) (L = H2, a; a-N2, b; P-P = DPPB, DPPP).a Complex  Solvent  6Av ppm  8B, ppm  2JAB, Hz  Ref.  DPPB, 9  CD2C12 C6D6  62.0 64.0  53.8 54.9  46.6 47.3  DPPP, 7  twe 85  CD2C12b C6D6  57.2 59.0  49.8 51.0  57.0 57.0  i 2 -H2, 9a  tw 85  CD2C12  53.0 54.6 53.7 53.8 48.7 46.6 50.3 46.4  52.5 39.0 53.2 38.3 46.4 42.9 47.4 42.7  44.7 34.2 44.4 33.8 54.9 43.6 55.6 42.5  tw  54.4 46.6 50.1 39.2  53.5 36.8 48.1 36.3  45.1 32.1 54.1 39.9  C6D6 11 2 - H2, 7a  CD2C12 C6D6  a-N2, 9b  C6D6  a-N2, 7b  CD2C12  85 tw tw 85 tw  The complexes were prepared in situ by introducing 1 atm of the respective gas (H2 or N2) into solutions of 7 and 9. Discussion of the dinitrogen complexes 9b and 7b is provided in Section 3.5.2. b At -90°C. c This work. a  To gain further evidence for the existence of an r1 2 -H2 moiety in complex 7a, a H2/D2 isotope exchange experiment was performed. The 1 H NMR spectrum of 7 under - 1.1 atm of H2 and - 1.5 atm of D2 in CD2C12 shows a 1:1:1 triplet ( 1 ./HD = 29.4 Hz, w1/2 = 4 Hz) of 1:2:1 triplets (cis, 2.1Hp = 7.6 Hz) centred at 8 = -10.86 ppm, corresponding to an 71 2 -HD ligand (Figure 3.8; see Section 4.2 for preparation details). Both 1 Jp and 2./Hp coupling constants are the same as those for the DPPB complex 54  (29.4 and 7.5 Hz, respectively). 69 The resonance from free HD in CD2C12 also appears as a 1:1:1 triplet at 4.67 ppm with 1J11 = 42.9 Hz, comparable to the value reported for gaseous HD (43.2 Hz). 41 The 1 JHD value is within the range of values (18-34 Hz) reported for other 1 2 HD complexes, 35,3738 and resolved 2./Hp coupling provides further -  evidence of the slow exchange between bound and free H2 on the NMR time-scale. 35,37,38 Table 3.4 Temperature dependence of the 1 H NMR T1 relaxation time for the (71 2 2 -H2)Ru(DPPP)(1.1-C1)3RuC1(DPPP), 7a, at 8 -10.90 ppm -H2)moietyn(71 (300 MHz, CD2C12). Temperature, 0, K  T1, msa  Linewidth, w 112 , Hz  T2*, msb  296 274 255 233 213 193  19±1 16±1 14±1 12±1 13±1 16±2  36 39 45 61 69 79  8.8 8.2 7.1 6.2 4.6 4.0  Obtained by the inversion-recovery method (180°-t-90° pulse sequence). 71 b T2* values are calculated from the relation: linewidth = (nT2*) -1 . a  3.8  -4  -4.2  -4A  -4.6 0.0032  ^  0.0036^0.004^0.0044  ^  0.0348  ^  0.0052  ^  0.0056  1 / 0^(K)  Figure 3.7 Plot of temperature dependence of Ti for the molecular hydrogen moiety in (71 2 -H2)Ru(DPPP)(1.1,-C1)3RuCl(DPPP), 7a.  55  2./Hp  IT11711,11111111111111)IIIIIIIIIII  ON  -16.7^-16.8^-101.9 PPM -ILO  9^8^7^6^5^4^3^2^1 PPM Figure 3.8 1 H NMR spectrum of [(11 2 -HD)(DPPP)Ru(p.-C1)3RuCl(DPPP)], formed in situ by stirring [RuCl(DPPP)(.1-C1)12, 7, under -- 1.1 atm of H2 and - 1.5 atm of 132 (300 MHz, CD2C12, 20°C). The high field region is shown in the inset.  Molecular hydrogen complexes are known to catalyse the isotope exchange reaction to give HD. 46,56 c One such example is the Kubas complex, W(C0)3(PPr 13)2(1 2 H2), 46 a which catalyses H2/D2 exchange even in the solid state. The mechanism for the formation of HD from H2/D2 in the presence of 7, suggested by Joshi for the DPPB complex, 9, may involve an intermediate where both H2 and D2 coordinate to the same Ru centre, giving rise to a (1 2 -H2)(11 2 -D2) complex (Figure 3.9). 45  Ru....  ... ....  "" I  PF  CI^ CI  Figure 3.9 Suggested intermediate for 7-catalysed H2/D2 HD isotope exchange.  3.5.2 Formation of a Dinitrogen Complex by Interaction of  I with N2  Interaction of complex 7 with 1 atm of N2 in methylene chloride solution leads to the rapid, reversible in situ formation of a dinitrogen complex 7b (Equation 3.6, Figures 3.10 and 3.11). The  31 P  NMR spectrum of the dinitrogen complex 7b indicate ca. 70%  equilibrium conversion of 7 to 7b at 20°C. [(6-N2)(DPPP)Ru(g-C1)3RuCI(DPPP)]  [RUCI(DPPP)01-C1)]2 + N2 7  (3.6) 0. ,  (Pei  d  i.^  ?  PF 'C1 7b  57  9.,  V CH3 CON (C113) 2  9^8^7^6^5^4^3^2^i PPM^0 Figure 3.10 1 H NMR spectrum of the dinitrogen complex 7b formed in situ from [RuCI(DPPP)(.1-C1)12, 7, under —1 atm N2 pressure (300 MI-h, CD2C12, 20°C). The sharp peaks are due to the presence of DMA impurity in 7.  kr,  I  F  r^r^riiiiiri^III 45 40^35 PPM^30 Figure 3.11 IIP( IH)NMR spectrum of the dinitrogen complex 7b formed in situ from [RuCl(DPPP)(ii-C1)12, 7, under —1 atm N2  60  I^  55^50  ►^rii  pressure (121.42 MHz, CD2Q2, 20°C); the broad signal at — 56 ppm is that of 7.  The 31 P NMR spectrum of 7b (Figure 3.11) closely resembles that of the DPPB analogue, 9b, and those of other triply chloro-bridged dinuclear Ru species, [(L)Ru(DPPB)(1.1-C1)3RuC1(DPPB)), having similar chemical shifts and 2./pp coupling constants (see Table 3.3). 85 The sharp signals in the  31 P [ 1 H) NMR spectrum also implies  slow exchange between bound and free N2 on the NMR time-scale. The presence of an IR band at 2172 cm -1 (m, CI-12C12 solution in 0.1 mm anaerobic NaC1 cell, Figure 3.12) indicative of a N=N stretch, provides further evidence for the presence of an a-N2 ligand in 7b (cf. 2175 cm -1 for 9b). 69 This value of 2172 cm -1 is in agreement with one of the criteria set by Morris 47a.67a in predicting formation of molecular hydrogen complexes (see Section 1.4.4.). Of note, the IR band at 1630 cm -1 in Figure 3.12 arises from the(.) stretch for the uncoordinated DMA impurity.  VNN = 2172 cm -1  = 1630 cm -1 ^0000. 0 2102.  a  MM. 0 1042.7 11071L 0 1000.^1707.0 10011. 7 2000. VAVEM ^111:10-1)  Figure 3.12 Infra-red spectrum of a CH2a2 solution of [RuCl(DPPP)(i-a)}2, saturated with N2 (solvent spectrum subtracted).  7,  3.6 Interaction of Ru2C15(DCYPB)2, 1, with H2 3.6.1 Reaction of Ru2C15(DCYPB)2, 2, with H2 in DMA  The insolubility of 3 in most solvents caused difficulties in the reduction of this compound. After the DMA suspension of 3 was left stirring under H2 for over 10 days, a very pale orange solution was obtained with most of the precursor 3 remaining unreacted. A reddish brown solid 11 was isolated from the DMA liquor (after treatment with methanol) in very low yield (< 1%; refer to Section 2.3.4 for details). The proton NMR spectrum of this air-sensitive product shows a broad signal at -12.05 ppm in the absence of a H2 atmosphere, with a T1 value of 22 ms for the high field resonance (300 MHz, CD2C12, 20°C), characteristic of a i 2 -H2 moiety. In addition, signals for the ionic species DMAH+ are observed (6 = 2.56 (CH3C0), 3.18 (N(CH3)2) ppm, s).  95  The  31 P{ 1 H)NMR spectrum of the same solution consists mainly of two AB patterns (6A =  59.2 ppm, 8 B = 42.0 ppm, 2./AB = 39.2 Hz; Sc = 65.6 ppm, 8 D = 46.6 ppm, 2.ICD = 25 . 7 Hz) of equal integral intensity, possibly due to a dimeric species, and a singlet at 48.1 ppm, indicating a mixture of products. According to the reduction scheme (Equation 3.4), reaction of 3 under H2 in DMA should afford the ionic species [Ru2C15(DCYPB)2] - DMA11+ as the initial product, and the neutral Ru2C14(DCYPB)2 species should be isolable upon the addition of methanol. The singlet at 48.1 ppm in the phosphorus spectrum is thus assigned to the ionic species [Ru2C15(DCYPB)2] - DMATI+ based on a similar observation with the analogous DPPB (50.9 ppm, s, C7D8), 95 and DPPP complexes (40.7 ppm, s, CD2C12, Section 3.4.1). The Ru2C15(P-P)2 complexes are known to disproportionate in coordinating solvents like DMA and CH3CN to give two dimeric [RuC12(P-P)]2 and [RuC13(P-P)]2 compounds, with the possible involvement of a tetranuclear intermediate [Ru4C1 10(13-13 )4] .84 Even though elemental analysis of 11 suggests a tetranuclear 61  formulation of Ru4C110(DCYPB)3(DMA) (see Section 2.3.4), the complicated 31 P{ 1 H} NMR spectrum suggests the elemental data may be rather meaningless because of the indication that more than one species are present. Of note, an IR band at 1939 cm -1 possibly a Ru—H stretch is observed for this product mixture. Together with the existence of an 1 2 -H2 ligand, the major product isolated from this reaction may be a polyhydride. The assignment of the AB patterns in the 31 P{ 1 H) NMR spectrum is not possible at this stage. Interestingly enough, when the NMR solution of 11 was exposed to an atmosphere of H2, the original light brown solution slowly turned green and red crystals formed after – 3 weeks. Some of these crystals obtained were carefully taken out and submitted for Xray analysis.t The proton NMR spectrum of the remaining filtrate exhibits, in addition to the original r1 2 -H2 signal, a second 1 2 -H2 signal centred at -12.34 ppm, with a T1 of 33 ms (300 MHz, CD2C12, 20°C). The 31 P ( 1 H ) NMR spectrum of the same solution shows two new AB patterns (5 A = 67.9 ppm, 5B = 47.1 ppm, 2JAB = 25.6 Hz; Sc = 67.8 ppm, 5D = 39.0 ppm, 2Jcp = 32.7 Hz), along with the disappearance of the singlet at 48.1 ppm. The complexity of this reaction and the more basic nature of the DCYPB (relative to DPPB) suggest that activation of H2 by 3 may be achieved even without an added base, and that the presence of the coordinating solvent may complicate the reaction by forming multinuclear species. As a result of this reasoning, the reaction of 3 with H2 was then carried out in an non-coordinating solvent (next section).  t The X-ray analysis has not yet been undertaken.  62  3.6.2 Reaction of Ru2C15(DCYPB)2, 2, with H2 in CH2C12  The reaction of 3 in CH2C12 under 1 atm of H2 for 1 week yields two major products: a) an insoluble orange-brown solid, 12, which turns greenish black upon exposure to air, and b) a dark brown filtrate which precipitates out a brown solid, 13, upon concentration of the solution volume to – 1 mL and the addition of diethyl ether (see Section 2.3.4 for details). The insolubility of 12 in most organic solvents again poses difficulties in its characterisation. The elemental analysis gives an empirical formulation of Ru4C1 10(DCYPB)3• The IR spectrum contains no bands attributable to Ru—H stretches, but even so, the presence of an 1 2 H2 ligand cannot be ruled out. The tetranuclear, mixed -  valence formulation of 12 has a Ru : P-P ratio of 4 : 3, one bis(phosphine) short of the intermediate [Ru4C110(P-P)4] proposed for the disproportionation of Ru2C15(P-P)2 complexes in coordinating solvents. 84 The bulkiness of the three DCYPB ligands may hinder the coordination of a fourth diphosphine ligand, but ligands like H2 or 02 may be small enough to occupy the vacant sites. This might explain the air-sensitivity of 12, and the high basicities on the Ru centres exerted by the DCYPB ligands may also contribute to the air-sensitivity. The 1 H NMR spectrum of complex 13 in solution under vacuum (Figure 3.13) exhibits a broad signal at 8 -12.05 ppm, similar to that observed from the same reaction in DMA (previous section), with a T1 value of 23 ms (300 MHz, CD2C12, 20°C). The 31p{lin NMR spectrum of the same solution shows only two AB quartets with equal  integral intensity (Figure 3.14). The two AB patterns have the same chemical shifts and 2.ipp coupling constants as the one observed for 11 formed from the reaction in DMA, and  the spectrum differs only in the position of the 'extra' singlet being at 49.1 instead of 48.1 ppm. The presence of the singlet is possibly due to the anionic species [Ru2C15(DCYPB)21 - by analogy with the previous data for the reaction in DMA, but as 63  ........■■  6^4^ 0^—2^—4^ 0^2 PPM Figure 3.13 1 H NMR spectrum of [(t1 2 -t-i2)(DCYPB)Ru(p.-C1)3RuCI(DCYPB)), 13, under vacuum (300 MHz, CD2C12, 20°C).  75^70^65^60^55^50^45^40 PPM Figure 3.14 31 P { 1 11) NMR spectrum of [(i 2-142)(DCYPB)Ru(g-C1)3RuC1(DCYP13)1, 13, under vacuum (121.42 MHz, CD2C12, 20°C)..  there is no added base present, the corresponding accompanying cation has yet to be identified. The basic nature of the DCYPB ligand may be sufficient to remove any HC1 produced, similar to an analogous observation made for the monodentate PCy3 ligand in some iridium complexes. 96 The two independent AB quartet patterns of equal integrated intensity in the 31 P { 1 H) NMR spectrum of 13 are tentatively assigned to the unsymmetric, triply chloro-bridged dinuclear structure (i 2 -H2)Ru(DCYPB)(4C1)3RuCl(DCYPB), similar to complexes 9a and 7a (cf. Equation 3.3). Further evidence, including better analytical results, is needed to provide a more conclusive structure for 13. A temperature dependent T1 relaxation experiment was performed for the r1 2H2 resonance of 13 from 223 to 293 K (Table 3.5), and the plot of In T1 versus lie results in a V-shaped curve (Figure 3.15, from Equation 1.6), although the minimum T1 value is less well defined due to broadness of the dihydrogen signal at low temperature. Nevertheless, a T1(min) value of 13 ms is estimated at 233 K (0( min)), corresponding to an internuclear H—H distance of 0.86 ± 0.04 A (see Section 1.4.5b and 3.3.1 for details). The T1(min) value is consistent with that found for other Ru(II) molecular hydrogen complexes discussed previously. Table 3.5 Temperature dependence of the 1 H NMR T1 relaxation time for the (r1 2 H2) moiety in complex 13, at 6 -12.05 ppm (300 MHz, CD2C12). Temperature, 8, K  Tl, msa  Linewidth, wt /2 Hz  T2*, msb  293 273 253 233 223  23±2 19±2 16±2 13±3 16±3  26 37 43 — 69 c  12 8.6 7.4 — 4.6  ,  Obtained by the inversion-recovery method (180°-t-90° pulse sequence). 7 ' b T2* values are calculated from the relation: linewidth = (itT2*) -1 . Signal too broad to give reasonable linewidth. a  66  -3.0  n T1  -3.4  -3.8  -4.2  I^  ^ +  t  -4.8  -6.0 32  3.8^4.0  ^  4.4  ^  48  1/ 0^(103K)  Figure 3.15 Plot of temperature dependence of Ti for the molecular hydrogen moiety of 13 at 8 -12.05 ppm.  When complex 13 is exposed to 1 atm of H2 in CD2C12 in a NMR tube, a gradual change of colour from light brown to green is observed. The resulting 1H NMR spectrum shows a partially formed species 13a (Figure 3.16), which exhibits an 1 2-H2 signal at -12.33 ppm (T1 = 33 ms, 300 MHz, CD2C12, 20°C), the 'same' result as obtained for the reaction of 13 with H2 in DMA (previous section). Removal of H2 under vacuum gives only partial reversal back to 13. The 31P{ 1 }1) NMR spectrum shows a further AB pattern attributed to 13a (SA = 67.8 ppm, h  =  47.2 ppm, 2JAB = 27.2 Hz), and a new  singlet at 55.4 ppm (Figure 3.17). The observed phosphorus NMR spectral data are consistent with a complex with a geometry as shown in Figure 3.18a, but such a structure would have equivalent dihydrogen signals because of the C2 symmetry across the two bridging chlorides. A complex with geometry as in Figure 3.18b would account for the two dihydrogen signals from 1H NMR, but two AB patterns would be expected in the 31P OH) NMR spectrum. The 'missing' AB pattern may result from coincidental  67  4^2^0^—2^—4^—6^—8^—1 0^ 12 PPM Figure 3.16 1 H NMR spectrum of 13a, formed from the reaction of [(r1 2 -H2)Ru(DCYPB)(1.1C1)3RuCl(DCYPB)), 13, with 1 atm H2 (300 MHz, CD2C12, 20°C). Resonances marked with —  asterisks are those from solvent impurities.  1^1^iiiI1111^1^1 1^1^I^II^1^1^1^1^1^1,^1^1 1^11^I^III^1 7 0^65^60^5 5^50^45 PPM^40 Figure 3.17 31 P ( 1 H} NMR spectrum of 13a, from the reaction of Ri 2 -H2)Ru(DCYPB)(u-C1)3RuC1(DCYPB)], 13, under 1 atm H2 (121.42 MHz, CD202, 20°C) 68  degeneracy but this is considered unlikely. The spectral data are also consistent with 13a being a mononuclear species containing one T1 2 -H2 ligand (Figure 3.18c). It is very unclear why a second T1 2 -H2 is observed when complex 13 is exposed to H2, particularly as 13 is formed from 3 under the same H2 atmosphere conditions. Further exploration in this area is needed before more definitive conclusions can be made. Table 3.6 summarises the spectroscopic data obtained for 13 and 13a.  C 1„ 1...  ^ .1211 1. BPS II A H  p(-A  Cl, H T H Ck„ 1PiTh 2 „op t, '"Rii'''''^ ^ R a ° . . ^H Bpie,^---,c i_ ■ 1 \?, tl H  ...,,oCti„,,..h.....„,01PB T'^ ..  Ppl.  CI HIC1 H  4  1  a  c P-P = DCYPB Figure 3.18 Suggested geometries for species 13a.  Table 3.6 1 H and 31 P{ 1 H) NMR data (CD2C12, 20°C) for 13 and 13a Complex  1 H (300 MHz),^31 P{ 1 H) (121.42 MHz),  8 (I)Pm)^8 (PPm)  2JA13, Hz  13^-12.05^8A= 59.2, 8,3 = 42.0^39.2 c 65.6, 8 D = 46.6^25.7 13a^-12.33^8,k= 67.8, 8 B = 47.2^27.2  s  69  .  CHAPTER 4 Thermodynamic and Kinetic Studies of the Interaction of [RuCI(DPPB)2(11-C1)]2 with H2  4.1 Introduction In the absence of an added base, the dichloro-bridged, dinuclear complex [RuCl(DPPB)2(4-C1)]2, 9, had been found earlier to react with 1 atm 112 to give the molecular hydrogen complex (1 2 -H2)Ru(DPPB)(p-C1)3RuCl(DPPB), 9a. 69 Complex 9 also catalytically hydrogenates alkenes, ketones, imines and nitriles under relatively mild conditions (1-12 atm H2, 30-100°C), with 9a thought to be a catalytically important species. 45 Kinetic investigations by Joshi, 45 using H2-uptake measurements to follow the hydrogenation of styrene to ethylbenzene catalyzed by 9, produced values of 93 M -1 s -1  and 2.9 M -1 s -1 for ki and k2, respectively, the rate constants from Equations 4.1 and 4.2 (DMA, 30°C).  [RuCl(DPPB)(t-C1)] 2 + H2  k_^ 9^  R►12-H2)(DPPB)Ru(11-C1)3RuC1(DPPB)] (4.1) 9a  k2 [012 -H2)(DPPB)Ru(1-0)3RuCl(DPPB)1^[RuCl(DPPB)01-0012 C6H5CH =CH2  ^  70  (4.2) C6H5CH2CH3  From these investigations, a value of 0.060 ± 0.005 s -1 was computed for k_1, and thus a value of 1.5 x 10 3 ± 150 M-1 for the equilibrium constant K = k1/k_1 at 30°C was calculated for the styrene system. This chapter presents a series of studies aimed at assessing the rate constants k 1 , k_1 and the equilibrium constant K independently from direct measurements on equilibrium 4.1 in the absence of any added substrate. Measurements of the corresponding activation and thermodynamic parameters are also included.  4.2. Solubility of Hydrogen in CD2C12(CH2C12)  The solubility of hydrogen in CD2C12 was determined using P(p-tolyl)3 as a calibration at 25°C. All the samples were prepared by first introducing the H2 gas at a known pressure into the NMR tubes, which contained the phosphine in CD2C12, at liquid nitrogen temperature (77 K). After the samples were sealed at 77 K, the total H2 pressure inside the NMR tube at 25°C was estimated by calculating the pressure at 298 K. The T1 value for the dissolved H2 signal in CD2C12 (I5 4.60 ppm, s) was found to be 2.0 ± 0.1 s, whereas the T1 values for all the proton resonances from P(p-tolyl)3 were less than 1.2 s. Thus a pulse sequence with a time duration of 12 seconds (5 s acquisition time + 7 s delay) was used. The H2 solubility obeys Henry's Law at least up to 1.25 atm pressure of hydrogen, with the plots of molar solubility against partial pressure of H2 being linear. The Henry's Law constant KH was determined to be 3.64 ± 0.04 x 10 -6 M mmHg -1 at 25 °C, corresponding to a H2 solubility of 2.77 ± 0.03 x 10 -3 M at one atm pressure (Figure  4.1).  71  0 0^200.0^400.0^600.0  ^  800.0  ^  1000.0  P (mmHg) Figure 4.1 Plot of dissolved H2 concentration in CD2C12 versus H2 pressure at 25°C.  4.3 Determination of the Equilibrium Constant by UV-Visible Spectroscopy  The equilibrium constant K for the conversion of 9 to 9a is expressed by the following equation:  [9a] K = [9] [H2]  (4.3)  and can be determined by monitoring the changes in absorbance observed at 320 nm in the UV-visible spectra as the H2 pressure is increased (Figure 4.2a). Equation 4.3 can be expressed alternatively as  [9a] K =^ [M T - 9a] [H2 ]  (4.4)  where MT = total metal complex concentration. Rearranging Equation 4.4 gives  [9a] = [M,T] [H2]^(4.5) +[H 2 ] 72  300^350^400^450^500  ^  550  ^  600  ^  Wavelength (nm)  650  (a)  0.07 0.06 0.05 *t 0.04 Fc' 0.03 0.02 0.01 0  0  0.5^1^1.5  [H7] x 10' M  2  2.5  3  (b) Figure 4.2 (a) Changes in the UV-visible spectrum with increase in concentration of H2 for a CH2C12 solution of [RuC1(DPPB)2(.1.-C1)]2, 9; [Ru2] = (5.4 ± 0.3) x 10 -5 M. e' = absorbance / total [Ru2]. (b) Plot of changes in absorbance at X = 320 nm versus H2 concentration; data taken from (a). The residual plot indicates the difference between the observed value in absorbance and the calculated value using the rectangular hyperbola fit. Temp = 25°C.  73  The Beer-Lambert expression for the equilibrium system can be written as A = (e[9] + eq9a])•1^  (4.6)  where A = absorbance; e and e* = extinction coefficients for 9 and 9a respectively; 1 = absorbance path length. Combining Equations 4.5 and 4.6 and rearranging gives Ao - Ae =  —Ae' [H 2 ] [M T ] K -1 +11-121  (4.7)  where A o = initial absorbance of 9; A e = equilibrium absorbance at any chosen [H2]; and Ae' = (e - e*)/ 1. Equation 4.7 is in the form of a rectangular hyperbolic expression: Y  ax  + c .^A non-linear least squares fitted program, supplied for the b+x  Archimedes workstation for the stopped-flow instrument (Section 4.5.2), gives a value of 750 ± 70 M -1 for the equilibrium constant K at 25°C in CH2C12, which is about a factor of two smaller than the value calculated from the styrene-hydrogenation system mentioned earlier (see Section 4.1). The analysis also leads to a e* value of — 5300 M - lcm -1 at 320 nm; e is measured directly as 7100 M -1 cm -1 . Considering the differences in the solvent used between the present system (CH2C12) and the one used in the styrene-hydrogenation system (DMA), the K values are remarkably close.  4.4 Determination of the Equilibrium Constant by NMR Spectroscopy The measurement of the equilibrium constant by NMR spectroscopy is accomplished by a) estimating the solubility of hydrogen in CD2C12 by 1 H NMR and b) using the integral intensity of the 31 P signals attributable to species 9a and 9 to determine their ratio at some known [H2]. Since the determination of K is based on integral intensity ratios, T1 values were measured for all the resonances needed for the calculation of K, both phosphorus and proton signals. Pulse sequences with duration based on five times  74  ^  the largest T1 values found were then employed to ensure quantitative values for all the integral intensity ratios.  4.4.1 31 P NMR T1 Measurements  From the T1 experiment on the 31 P ( 1 H) spectrum, relaxation times of all phosphorus nuclei were determined to be in the range of 1.7 to 2.3 seconds (Table 4.1; Figure 4.3). An equilibration time of 13 seconds (3 s acquisition time + 10 s delay) was used in the pulse sequence in order to obtain quantitatively reliable integration values. Table 4.1 Summary of T1 values for the phosphorus nuclei in fRuCl(DPPB)(p,-C1)12, 9, and [( -1 2-H2)Ru(DPPB)(12-C1)3RuCl(DPPB)], 9a (121.42 MHz, 25°C).a ^ ^ Complex Chemical Shifts T1 (s) (1)Pm) ^  ^ 8A = 63.1 1.8 ± 0.2 ^ S B = 54.9 1.7 ± 0.1 ^ ^ 6 c = 53.0 1.9 ± 0.1 9a ^ 1.8 ± 0.1 = 52.5 ^ 1.8 ± 0.1 SE = 54.6 ^ 8 F, = 39.0 2.3 ± 0.2 9  a  Obtained by the inversion-recovery method71 (180°-T-90° pulse sequence).  The integral intensity ratio of 9a and 9 from the 31 P NMR spectrum at 25°C, for example, was found to be 3.76 at 2.76 x 10 -3 M [H2] (Table 4.2). Combining with the H2 solubility data, the equilibrium constant K for the conversion of 9 to 9a (Equation 4.1) was calculated to be 1.4 ± 0.2 x 103 M-1 at 25°C (see Equation 4.3). This value is  75  4.0 s  0.25 s  0.0625 s  I^I^I^i^i^1^I^I - r--i^1 65^60^55^50^45^40 PPM Figure 4.3 31P ( 1 H) NMR stacked plots for the measurements of Ti relaxation times for the resonances of (RuCl(DPPB)(g-C1))2, 9, and [(1 2-H2)(DPPB)Ru(ii-C1)3RuC1(DPPB)], 9a (121.42 MHz, CD2C12, 25°C). Resonances marked with asterisks are those from 9.  reasonably close to the one calculated from the styrene-hydrogenation system using DMA as the solvent (see Section 4.1).  4.4.2 Thermodynamic Parameters  The thermodynamic parameters AH° and AS° for the reaction shown in Equation 4.1 can be derived from Gibb's Law: AG° = -RT lnK^ In K = -AH RT  °^AS°  (4.8) (4.9)  To obtain these parameters, measurements of the equilibrium constant were carried out at various temperatures (0 to 25°C) and pressures (0.5 to 1.5 atm H2) using the NMR spectroscopy method described previously; the results are summarised in Table 4.2. Table 4.2 Summary of K values determined at various temperatures for the conversion of [RuCl(DPPB)(4.-C1)]2, 9, with H2 to [(1 2 -H2)Ru(DPPB)(4-C1)3RuC1(DPPB)], 9a, in CD 2 C1 2 . Temp (K)  [H2] x 10 3 M a  [9a] / [9]  298  1.39 2.76 4.15 1.48 2.70 1.50 2.60 3.63  1.72 3.76 6.29 1.89 4.49 1.72 4.73 5.33  288 273  b  K x 10 -3 M -1  In K  1.24 1.36 1.52 1.28 1.66 1.15 1.82 1.47  7.12 7.21 7.32 7.15 7.41 7.04 7.51 7.29  Solubility of H2 at various temperature was determined by calibration using P(ptolyl)3 as the standard. Refer to Section 4.2 for details. b Ratio determined by taking integrated intensity values of resonance signals from 9 and 9a in the 31 P 1 H) NMR spectra (121.42 MHz). a  77  Examination of the data obtained from the above experiments indicates that K is essentially independent of temperature over the range 273 to 298 K. This suggests that AH° 0 for the reaction, while the corresponding AS° is – 60 ± 2 J/mol K. The AH° value, ignoring effects of solvent, is governed by changes in the relevant bond energies; in this reaction (Equation 4.1), the H—H bond is lengthened (an endothermic contribution), a Ru-(7) 2 -H2) bond is formed (an exothermic term) and a terminal Ru—Cl bond is 'replaced' by a bridged Ru—Cl—Ru moiety. Bond energies for a bridged chloride are typically about 75% that of the terminal chloride, 97 and thus the terminal to bridged rearrangement will be an endothermic process (probably about 60 kJ/mo1 97 ). Such data could be used to estimate that for the process: Ru II + H 2  Run -(r1 2 -H2) (4.10)  AH° is – -60 kJ/mol; i.e. the formation of the i12-H2 complex is favoured, ignoring other changes in the metal coordination spheres (in this case terminal chloride 112-C1, and changes in solvation). The latter, however, appears to be very important as judged by the measured positive AS° value (– 60 J/mol K). Reaction 4.10, as well as the chloride rearrangement, would be expected to be unfavourable from an entropic point of view. A rationale for the positive AS° value is that the reactant 'five-coordinate' dimer in solution contains a coordinated CH2C12 (CD2C12) at each Ru; reaction 4.1 would thus also generate two molecules of solvent per reactant dimer (Equation 4.11). [RuCl(DPPB)(g-C1)(S)] 2 + H2 [(112-H2)(DPPB)Ru(g-C1)3RuCl(DPPB)] + 2S S = CH2 C12 (CD2 C12 )  (4.11)  Dichloromethane can coordinate at metal centres. 98 The bonding is expected to be weak, and the basic conclusions regarding the enthalpy terms should remain valid; the estimated molar enthalpy of – -60 kJ/mol would then be the sum of the AH° for Equation 4.10 plus two times the bond energy for the Ru-Cl solvate bond (Ru•••C1CH2C1).  78  4.5. Stopped Flow Experiments - Determination of k1 and k_1  4.5.1 Sample Handling All experiments were carried out under anaerobic conditions with special precautions taken to exclude oxygen and water. The procedures for general anaerobic precautions are listed in the operating manual for the stopped-flow instrument. 99 Typically, the closed circuit thermostat bath and the complete flow circuit, including reagent reservoirs, drive syringes, stopping syringe and flow lines, were filled with a deoxygenated pH 8.0 Trizmae•HC1 buffer (5.0 x 10 -2 M). A deoxygenated solution of sodium dithionite (5.0 x 10 -2 M) containing buffer was then added to remove any dissolved oxygen in the stopped-flow system, and the whole instrument was left standing for at least 12 hours under a continued argon purge. After that, more sodium dithionite was added to the thermostat bath, while all the flow lines and syringes were flushed thoroughly (at least four times) with deoxygenated buffer to remove all traces of sodium dithionite. The flow lines and the syringes were then purged four times with deoxygenated ethanol, and then with deoxygenated acetone, before the solvent used for the study (CH2C12) was introduced. For any given experiment, one of the drive syringes was filled with a CH2C12 solution of fRuCl(DPPB)2(1.1-C1)]2, 9, saturated with argon, and the other with CH2C12 saturated with H2 at 1 atm total pressure. Due to its reactive nature, the CH2C12 solution of 9 was prepared immediately before the stopped-flow experiments and was used immediately upon its preparation in order to avoid possible side-reactions. One experiment was taken to be the average of at least four stopped-flow runs done at constant concentration of all reagents. The concentration of 9 was varied from experiment to experiment, while the concentration of H2 was held constant. Blank runs in which both syringes were filled with neat CH2C12 were carried out before each experiment to check for normal operation of the instrument. 79  4.5.2 Data Treatment and Results  Raw data from the stopped-flow experiments were analyzed on an Archimedes workstation using a non-linear least squares fitting program, supplied with the stoppedflow instrumentation, which implements the Levenberg-Marquardt algorithm. 100 All other data analyses, including the numerical solution of differential equations, were carried out on a PC using customised programs written in BASIC by A. Pacheco from this laboratory. The complete customised programs are listed in Appendix A-3. The equilibrium shown in Equation 4.1 was studied in CH2C12 at 25°C using the stopped-flow technique. Changes in the visible spectrum upon reaction of 9 with H2 were monitored at a wavelength of 481 nm. The relatively small changes in the visible spectra (see Figure 4.2a, p.73), and the possibility of side-reactions with 02, required that the concentration of 9 be kept high [ (1 - 3) x 10 -3 M]. This constraint, together with the [H2] of 1.2 x 10 -3 M in CH2C12 (the saturating concentration of H2), represent the concentration limits that can be used for the two reagents. In fact, the concentrations of 9 and H2 will only be half of the values mentioned above after allowing for the dilution factor of two during the mixing within the stopped-flow experiments. The above constraints limited the range over which experiments could be carried out; in practice, only two experiments were performed, with ratios of 1.0:1.3 and 2.2:1.0 between the concentrations of 9 and H2.  4.5.2.1 Ratio of 9 to H2 = 1.0 : 1.3  Assuming a pseudo-first order reaction with respect to H2, the absorbance changes as a function of time over the first 150 s can be described by a simple decay equationlol (Equation 4.12). 80  A = A e + ae(-kobst)^  (4.12)  where A = absorbance at time t  ^  A e = absorbance at t = 00 (equilibrium)  (4.13)  ^  (4.14)  a = Ao - A e^(4.15) = total absorbance change observed for a given reaction ^ kobs = observed rate constant (4.16) Under pseudo-first order conditions where [H2] is kept essentially constant, the observed rate constant can be expressed as: kobs =^1- k-i,^where k i t = k 1 [H2]^(4.17) Rearrangement and substitution of  k k  with K gives  kl kobs = k 1 [H2] + —  (4.18)  which gives =  'cobs  (4.19)  11-1 2 1+ K -1  The absorbance versus time trace is depicted in Figure 4.4. Even though all precautions were taken to exclude oxygen contamination in the stopped-flow system, sidereactions occurred, as evidenced from the increase in absorbance after --120 s (Figure 4.4a). Side-reactions involving ligand dissociation (for example, by dilution) were ruled out: Measurement of the extinction coefficient £ at X = 481 rim for the reactant dimer 9 gave a consistent value of –800 M- lcm-1 over the concentration range of 10 -5 to 10-3 M. The other investigations of the equilibrium reaction (Equation 4.1) described previously (by NMR and UV-visible measurements) were done under conditions in which oxygen contamination was easily avoided. In this kinetic study, even though preparation of the  81  solution was done under argon and experiments were performed as soon as the solution was made up, residual oxygen in the system still appeared to be a problem. Despite the presence of the side-reaction, Equation 4.12 still fits the data in Figure 4.4 reasonably well. The side-reaction could be an oxidation reaction (see Chapter 5), or possibly reaction with N2 . 69 0.01 A = Ae+a e(-1(.1.0  Ae = -2.1E-3 8  g 0.005  a = 0.010 k obs = 0.050  0  0  -0.005  ^ ^ 50^100 150 200 Time (seconds)  Figure 4.4 A fitted stopped flow trace for the change in absorbance over 200 s, at X = 481 nm, for the reaction (at a 1.0 : 1.3 ratio) between fRuCl(DPPB)(4-C1)]2, 9 ((4.5 ± 0.5) x 10-4 M) and H2 ((5.9 ± 0.1) x 10-4 M) in CH2C12, temp = 25°C; the concentrations given are for the mixed solutions. The residual plot shows the difference between observed and calculated data.  From the calculated value of k o b s (0.050 s-1 in Figure 4.4), values of 26 and 20 M- ls-1 were calculated for k1 using K = 1.4 x 103 and 7.5 x 102 M-1 respectively. This preliminary value of k1 (23 ± 3 M - ls -1 ) is comparable with the one determined previously from the styrene-hydrogenation system (93 M -1 s-1 ). Considering differences in the 82  solvents used (CH2C12 vs. DMA) and the quite different types of kinetic studies, there is remarkably good agreement between the values. It should be emphasized that, for the system described here, pseudo-first order conditions are assumed, although the initial [H2] is only about 1.3 times that of the initial [9]. A second method can be used to determine k1 without assuming pseudo-first order conditions by examining the rate expression for the appearance of 9a: d[9a] — k 1 [9] [H 2 ] - k_ 1 [9a]^(4.20) dt The above equation can be rewritten as:  d[9a]  = ki([MT — 9a][H T — 9a]- K - 1[9a])^(4.21) dt^  where MT = total metal complex concentration HT = total dihydrogen concentration  Equation 4.21 is a non-linear differential equation, and cannot be solved directly by analytical methods. However, except for the unknown constant k1, all the other parameters (MT, HT and K) in Equation 4.21 are known prior to calculation. Therefore, Equation 4.21 can be integrated numerically using approximate methods. 10 ° A computer program (Appendix A-2.2) was used to generate [9a] versus t plots obtained for various trial k1 values. These plots were compared graphically to the [9a] versus t plots obtained from the experimental stopped-flow data, until a k1 value was found which gave good agreement between the theoretical and experimental plots. To transform the initial absorbance versus time data to concentration versus time, one can use the equilibrium expression (Equation 4.4) and the Beer-Lambert expression (Equation 4.6) to solve for 9a, which gives [9a] = (  A  A° )[9a] e A'  83  (4.22)  where^A = the absorbance at time t Ao = absorbance at t = 0 A' = Ae - A o Ae = absorbance at equilibrium [9a] e = [9a] at equilibrium  Details of the derivation of Equation 4.22 are presented in Appendix A-3. With the use of the two different K values determined previously (see Section 4.2 and 4.3), the concentration of 9a versus time plots generated by Equation 4.22 are shown in Figure 4.5, along with the theoretical plots generated using Equation 4.21. Values of k1 of 32 and 22 M 1 s-1 were obtained, respectively, for the two cases; the numerical values are -  summarised in Table 4.3. From the residual plots in Figure 4.5 between observed and calculated data, the uncertainties from the numberical approximation are found to be about 20%, which could be attributed to side-reactions occurring during the experiments. The calculated k1 values are in good agreement with those obtained using the pseudo-first order approximation.  84  m4re-iik 4401 1.5 k1 = 32 M -1 1 1 using K = 1.4 E3 W 1  7:7  0.5 ^ ^ 50 100 150 Time (seconds) Ts, 4.0E-05 LA iiii  Ali 1 6.1  (a) 1.5  1 0.5  0 ^ ^ ^ 0 50 100 150 Time (seconds)  1-,•3.0E-05  (b) Figure 4.5 Plots of appearance of [(71 2 -H2)Ru(DPPB)(11-C1)3RuC1(DPPB)], 9a, versus time using data collected from Figure 4.4 and analysis using Equations 4.21 and 4.22. [9] = (4.5 ± 0.5) x 10 -4 M; [H2] = (5.9 ± 0.1) x 10 -4 M; temp = 25 ° C. (a) K = (1.3 ± 0.1) x 10 3 M-1 and (b)K = (7.5 ± 0.7) x 102M-1.  85  4.5.2.2 Ratio of 2 to H2 = 2.2 : 1.0  The initial absorbance versus time trace for a 2.2 : 1.0 reaction ratio between 9 and H2 on a 200 s timescale produced a decay curve which appeared to be hampered severely by some side-reactions, possibly oxidation (Figure 4.6a). Blank runs carried out before and immediately after the experiment, in which 9 in CH2C12 was mixed with a CH2C12 solution containing only argon, gave traces where the curvature between 80 to 200 s fits well with the absorbance data obtained for the experiment (see Figure 4.6a). Subtraction of this blank run from the raw data in the experiment produces a decay trace which flattens out beyond 100 s (Figure 4.6b) and this was used to determine k1. Only the second method mentioned in the previous section was used to determine k1. The concentration versus time plots obtained by transforming the absorbance versus time data using the two equilibrium K values and subsequent numerical solution to Equation 4.21 are shown in Figure 4.7. The two k1 values are estimated to be 17 and 13 M-1 s'l (Table 4.3). The uncertainties from the numerical approximation are found to be about 20% from the residual plots in Figure 4.7 between observed and calculated data. Table 4.3 Summary of data to determine k1 from Equation 4.21 [RuC1(DPPB)(11-C1)]2, 9^[H2]^K^k1 (x 10 -3 M)^(10 M)^(vi-1)^(M-1 s -1)  0.45 ± 0.05 0.45 ± 0.05 1.3 ± 0.1 1.3 ± 0.1  (1.4 ± 0.2) x 10 3 (7.5 ± 0.7) x 10 2 (1.4 ± 0.2) x 10 3 (7.5 ± 0.7) x 10 2  5.9 ± 0.1 5.9 ± 0.1 5.9 ± 0.1 5.9 ± 0.1  86  32 22 17 13  0.02 Ai‘ raw data  0.01  8  0  -e -0.01 -0.02 -0.03 -0.04 -0.05 -0.06 0  400  100^200^300 Time (seconds) (a)  500  0.08 0.07  8 0  0.06 0.05 0.04 0.03  c  0.02 0.01 0 -0.01  ALA^sitti 'JILL/war Lia  0^50^100 Time (seconds)  r  .1 J. 1  ILA  150  Jill 200  (b) Figure 4.6 Fitted stopped flow traces of the changes in absorbance at = 481 nm, for a 2.2:1.0 reaction ratio between [RuCl(DPPB)(.4.-C1)]2, 9 ((1.3 ± 0.1) x 10 -3 M) and 112 ((5.9 ± 0.1) x 10 M) in CH2C12, temp = 25°C. (a) raw data before subtracting the blank and (b) after background correction.  87  4  3  0  50^100 Time (seconds)  150  200  150  200  4.0E-05 ,riri/fr t9  1 -5.0E-05  (a) 3 2.5 2 1.5 1 0.5 0  0  I  50^100 Time (seconds)  3.0E-05  4.0E-05  -  (b) Figure 4.7 Plots of appearance of [(11 2 -H2)Ru(DPPB)(g-C1)3RuC1(DPPB)], 9a, versus time using data collected from Figure 4.6b and analysis using Equations 4.21 and 4.22. [9] = (1.3 ± 0.1) x 10 -3 M; [H2] = (5.9 ± 0.1) x 10 -4 M; temp = 25°C. (a) K = (1.3 ± 0.1) x 10 3 M -1 and (b) K = (7.5 ± 0.7) x 102M-1.  88  4.5.3 Analysis of the Stopped-Flow Data  The average rate constant k1 value (overall 21 ± 8 M -1 s -1 ) is determined as 24 ± 8 or 18 ± 6 M -1 s -1 for the two sets of experiments described, using K values of 1.4 x 10 3 or 7.5 x 102 M -1 , respectively, in the calculation. These k1 values can be used to determine k_1 by the relation K = ki/k_i. Thus, k_1 values of 1.7 x 10 -2 and 2.4 x 10 -2 s 1 are calculated at 25°C. The k_1 values obtained from two sets of experiments with -  different ratios between 9 and H2 and the two different K values are in reasonable agreement and show that the data are internally consistent. Another similar hydrogenation system involving a dinuclear Ru-(1 2 -H2) intermediate also gives comparable values. For example, in the hydrogenation of 1-hexene catalyzed by the dinuclear complex RT1 2 H2)(PPh3)2Ru(4-H)(1.1-C1)2RuH(PPh3)2] in DMA, values of k1 = 39 M - ls -1 , k_1 = 4.9 x 10 -2 s - land K = 7.9 x 10 3 M -1 at 30°C have been determined. 102 It should be pointed out that the k1 and k..1 values determined in the present study are subject to large errors, especially given the radically distorted baseline in some of the experiments caused by side-reactions. More accurate k1 values could be obtained if such deviations could be eliminated. Even though large uncertainties are associated with k1, the estimated value is considered to be in very good agreement with the one obtained from the styrene-hydrogenation system 45 (93 M -1 s -1 ), which is obtained indirectly from catalytic hydrogenation rate dependencies on H2 and styrene concentrations. The estimated k1 value is about a factor of 4 different from the corresponding value determined from the styrene hydrogenation system, whereas the calculated k_1 value differs by a factor of about 3. 45 Two factors should be considered when comparison is made between the two systems. The first is the fact that the styrene hydrogenation system was conducted at a temperature of 30°C, while the present analysis was performed at 25° C; this temperature difference could probably account for a factor of up to about two between the rate constants. The second, and perhaps the more important factor, is the 89  solvent differences between the two systems (CH2C12 versus DMA). With a more strongly coordinating solvent like DMA, an alternate equilibrium could occur where the solvent coordinates to [RuCl(DPPB)(g-C1)12, 9, to give RDMA)Ru(DPPB)(11C1)3RuCl(DPPB)} (Equation 4.23). The conversion from 9 to the DMA-coordinated complex has been shown to be over 95 % in CHC13 containing ca. 100-fold excess of DMA at 20°C, based on NMR data. 45 As a result, in the styrene hydrogenation system, two other equilibria may be established competitive with Equation 4.1; the first one is the equilibrium between 9 and [(DMA)Ru(DPPB)(1.1.-C1)3RuC1(DPPB)], while the other is the equilibrium between the DMA adduct and 9a (Equations 4.23 and 4.24). Thus the value of 93 M 4 s 4 could represent an "overall rate constant" for the reaction between 9 and H2 to give 9a via an intermediate DMA adduct. Effects on solvent coordination to 9 would play an important role in determining the measured k1 values, and further studies are of merit in this area. Use of DMA as the solvent at 30 ° C in the stopped-flow experiments would provide a more conclusive answer. The possible role of CH2C12 as a coordinating solvent has already been mentioned (Section 4.4). [RuCl(DPPB)(p-Cl)] 2 +^f(DMA)(DPPB)Ru(g-C1)3RuC1(DPPB)] (4.23) 9 RDMAXDPPB)Ru(g-C1)3RuCI(DPPB)] —1` [(112 -112)(DPPB)Ru(g-C1)3RuCl(DPPB)] 9a^(4.24) + H2^ + DMA  90  CHAPTER 5 Reaction of [RuCI(DPPB)2(11-0)]2 with 0 2  5.1 Introduction  It was mentioned in Chapters 2 and 3 that dinuclear complexes of the form [RuCl(P-P)2(11-C1)J2 (P-P = DPPP, DPPB, DIOP, BINAP, etc.) are air-sensitive in both solution and solid state. When exposed to air, these complexes generally turn from yellowish-brown to green in the solid state within an hour, or within minutes in solution. The green oxidation products could not be detected by NMR spectroscopy, showing them to be paramagnetic. This chapter summarises some of the observations from the interaction of [RuCl(DPPB)2(2-C1)]2, 9, with 02, and subsequent reaction of the product with H2.  5.2 Interaction of [RuCI(DPPB)2(11-C1)12 with 02, and subsequent reaction of the product with H2  When a solution of fRuCl(DPPB)2(g-C1)]2, 9 (0.05 g, 0.04 mmol), is exposed to 1 atm of 02 in DMA (10 mL), the color of the solution immediately turns from brown to green. The intensity of the green color increased and eventually a black solution is observed after about one hour. Concentration of this brown solution and addition of hexanes (10 mL) precipitated a black solid. The 31 P { 1 H} NMR spectrum of the solid (CDC13, 20°C) shows only a singlet at 31.7 ppm, the resonance for free phosphine oxide 91  0=PPh2(CH2)4PPh2=0, 103 which is the only diamagnetic phosphine-containing species  present. The IR spectrum of this black solid shows a medium band at 803 cm -1 , possibly indicative of a metal-peroxo (0 2 ) stretch (Figure 5.1). This black complex is perhaps a -  mixture of Ru(IV)-peroxo phosphine oxide species. No reaction was observed when the black solid is left stirring in CH2C12 under one atmosphere of H2 at 20°C. Interestingly, after the color of solutions of 9 in CD2C12 used for NMR changed from brown to green upon reaction with 'residual' 02, addition of 1 atm of H2 regenerated the brown color. The 31 P{ 1 H) NMR spectrum of the brown solution exhibits the one AB pattern of the starting material 9 and the two AB patterns for the (7.1 2 -H2)-complex, 9a but no resonances from the phosphine oxide. The 1 H NMR spectrum also indicates a significant amount of H2O (— 0.7 mole per mole 9) produced from the reaction with H2. It is worth noting that some triphenylphosphine complexes of Pt, Ir, Rh and Ru catalyse the reaction of molecular hydrogen and oxygen to give water.'° 4 The interaction of the oxygen-reacted [RuCl(DPPB)2(1-C1)]2 complex with H2 is also evidenced by changes in the visible spectra (Figure 5.2). The broad band centered at 675 nm, which does not appear in the spectrum of pure complex 9, slowly disappears after about 0.5 hour when 1 atm of H2 was introduced, and the solution turns from green to yellow-brown. The final solution contains a mixture of 9 and the i1 2 -H2 species 9a (see Figure 4.2, page 73). Removal of H2 converts the 7 .1 2 -H2 species to 9.  92  VO43 = 803 cm -1  4  I^v^ ^4^I^II^+^I I DOD. 0 11111117. • 1075. O 0132. SO 050. 00 757. SO SVS. OD Sta. MO 41IX>. OD  V AVENV4U  ^CCM- 1)  (a)  IMOD. O 1187. S 1071. 0 8152. SO SSD. 00 757. $0 125. 00 13 5. SO 400. 00 WAVIINUMOC11  (b) Figure 5.1 IR spectra (Nujol mull) of (a) the black solid isolated from the reaction of [RuCl(DPPB)2(.t-Cl)]2, 9, with 02 and (b) pure 9.  93  1030  COD 0 a400  40:1^500^600^700  ^  Wavelength (nm)  800  ^  sco  Figure 5.2 Visible spectral changes of oxygen-reacted [RuCI(DPPB)2(g C1)12, 9, in CH2C12 on exposure to 1 atm H2. [9] = 2.7 x 10 -3 M, T = 25°C; a = initial spectrum, b = after 10 min, c = after 15 min, d = after 20 min, e = after 30 min. -  On consideration of all the above observations, the following, highly speculative reaction scheme for the reaction of [RuCl(DPPB)2(p CI)J2, 9, with 02 and H2 is -  proposed (Scheme 5.1):  94  IrTh oCk,^0...p  P/Rii'  CV Rti  o 0 \ pi ...  02  (  01.11  4  Cl  ,  P-P = DPPB  9^(brown)  -H211 H2  (F H2 0 + 02 H202 •  9a  H2  C\  ra,^NR.111,/1) "  Cl •  ..I^  "I  CY "  ....  (green)  Cl  02  0=rP=0 + RuIVC 1202' (black mixture) Scheme 5.1 Proposed tentative scheme for reaction of [RuCl(DPPB)2(1-C1)]2, 9, with 02 and H2. It is yet to be determined whether H202 or H2O is the initial product when H2 is allowed to react with the oxygen-coordinated dimer; the stoichiometry is consistent with formation of H202 and its subsequent decomposition to H2O. The existence of the oxygen-bound Run dimer and the Ruin peroxo species proposed in Scheme 5.1 remains to be proven. Further work in this area, particularly some low temperature experiments, is needed in order to characterise the intermediates proposed.  95  CHAPTER 6 General Conclusions and Recommendations for Future Work The objectives of the work described in this thesis were to extend the synthetic work developed in this laboratory on ruthenium(II) complexes containing one chelating diphosphine per ruthenium centre, and to elucidate further their role in activation of dihydrogen. Some of these complexes had been shown previously to be efficient catalysts for hydrogenation of unsaturated substrates. Use of chiral diphosphines in such species allows these complexes to effect catalytic asymmetric hydrogenation. The work demonstrating formation of the molecular hydrogen complex (i 2 H2)Ru(P-P)(1.1.-C1)3RuCl(P-P) (P-P = DPPB), discovered earlier in our group, was extended to include the diphosphines DPPP and DCYPB. The presence of the ri 2 -H2 moiety for the DPPP and DCYPB complexes was indicated by the short 1 H NMR T1 relaxation times of 12 ms (300 MHz) and for the DPPP system, the large H-D coupling constant for the 71 2 -HD isotopomer ( 1 Jp = 29.4 Hz). The H—H internuclear distances were estimated to be 0.86  A based on the T1 data for both DPPP and DCYPB systems.  Increase in the basicity of the diphosphines appears to favor the irreversible formation of a stable molecular hydrogen complex, as in the case with the DCYPB system, at least in comparison with both the DPPP and DPPB systems which show reversible binding of H2. It was reported from earlier work45 that the [RuCl(CHIRAPHOS)2(p.-C1)]2 complex, in which the diphosphine forms a five-membered ring with the metal, did not give a molecular hydrogen species upon reaction with H2. Thus together with the present work, the six- or seven-membered ring seems to be the optimum size for the chelating diphosphine to afford molecular hydrogen species. Studies should be continued to incorporate other basic diphosphines to give analogous Ru(II) dimers which also give sixor seven-membered rings.  96  Previous work from this laboratory has demonstrated the ability of [RuCl(DPPB)2(11-C1)]2 to hydrogenate catalytically various unsaturated substrates in the presence of H2. It would be interesting to see if such catalytic ability is retained using DPPP as ligand. This would also give an estimation of the effect of ring size on catalytic hydrogenation. Involvement of molecular hydrogen species in the catalytic cycle has been strongly suggested from previous work. Isolation of such a molecular hydrogen complex could help provide conclusive evidence for its role in hydrogen transfer to the substrate in the catalytic cycle. Studies on the thermodynamic parameters for the equilibrium conversion of fRuCl(DPPB)2(1.-C1)12 with H2 to (i 2 -H2)Ru(DPPB)(1a-C1)3RuC1(DPPB) in CH2C12 indicate a temperature independence of the equilibrium constant (K -- 1.4 x 10 M 4 ) over the temperature range studied (0 - 25°C). This suggests that the reaction enthalpy (AH°) for the equilibrium is approximately zero, and this leads to a rough estimate of the enthalpy change for the H2-binding at the Ru site; a positive AS° value strongly implies coordination of the solvent in the precursor dimer. Studies on the temperature dependence of the equilibrium constant using other methods like hydrogen uptake measurements could verify the results found in the present study. The rate constant for the forward reaction of the equilibrium has been measured by the stopped-flow technique (k1 20 M -1 s -1 at 25°C), and the value was comparable to that determined previously from a catalytic styrene-hydrogenation system. Preliminary studies were made on the reaction of [RuC1(DPPB)2(p.-C1))2 with 02 and the subsequent reaction of the product with H2. The latter results in the formation of water, it would be interesting to see if hydrogen peroxide is the initial product, and whether such a reaction system could be made catalytic. Further examination of the oxygen reaction alone is needed; the findings may provide some insight onto the existence of dioxygen adducts of ruthenium-phosphine complexes.'° 5  97  CHAPTER 7 References and Footnotes 1.  Nakamura, A.; Tsutsui, M. Principles and Applications of Homogeneous Catalysis; Wiley-Interscience: New York, 1980.  2.  (a) Borman, S. Chem. & Eng. News 1990, July 9, 9. (b) Nugent, W.; Parshall, G. W. Chem. Tech. 1988, 18, March, 184.  3.  Garrou, P. G. Chem. Rev. 1985, 85, 171.  4.  Hartley, F. R. Supported Metal Complexes, Catalysis by Metal Complexes Ser.; James, B. R., Ugo, R., Eds.; D. Reidel Publishing Co.: Boston, 1985.  5.  Michalska, Z. M.; Ostaszewski, B. J. Organomet. Chem. 1986, 299, 259.  6.  Parshall, G. W. Homogeneous Catalysis; Wiley-Interscience: New York, 1980.  7.  James, B. R. Adv. Organomet. Chem. 1979, 17, 319.  8.  James, B. R. in Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, 1982; Vol. 8, Chapter 51.  9.  (a) Crabtree, R. H. Acc. Chem. Res. 1979, 12, 331. (b) Brown, J. M. Angew. Chem., Int. Ed. Engl. 1987, 26, 190.  10.  Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987; Chapter 10.  11^Freifelder, M. Practical Catalytic Hydrogenation; Wiley-Interscience: New York, 1971.  98  12.  James, B. R. Homogeneous Hydrogenation; Wiley: New York, 1973.  13.  Rylander, P. N. Catalytic Hydrogenation in Organic Synthesis; Academic Press: New York, 1979.  14.  Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; Wiley-Interscience: New York, 5th Edn., 1990; Chapter 28.  15.  Mestroni, G.; Camus, A.; Zassinovich, G. in Aspects of Homogeneous Catalysis; Ugo, R., Eds.; D. Reidel Publishing Co.: Dordrecht, 1981; Vol.4, p 71.  16.  (a) Dawans, F.; Morel, D. J. Mol. Catal. 1977-78, 3, 403. (b) Halpern, J.; Okamoto, T.; Zakhariev, A. J. Mol. Catal. 1976, 2, 65.  17.  James, B. R.; Joshi. A. M.; Kvintovics, P.; Morris, R. H.; Thorburn, I. S. in Catalysis of Organic Reactions; Blackburn, D. W., Ed.; Marcel Dekker; New York, 1990; Chapter 2; and references therein.  18.^(a) Wang, G.-Z.; Backvall, J.-E. J. Chem. Soc., Chem. Commun. 1992, 980. (b) Burk, M. J.; Feaster, J. E; J. Am. Chem. Soc. 1992, 114, 6266. (c) Bacalski, A. G.; Cullen W. R.; Fryzuk, M. D.; James, B. R.; Kang, G.-J.; Rettig, S. J. Inorg. Chem. 1991, 30, 5002. (d) Bakos, J.; Orosz, A.; Heil, B.; Laghmari, M.; Lhoste, P.; Sinou, S. J. Chem. Soc., Chem. Commun. 1991, 1684. (e) Chan, Y. N. C.; Meyer, D.; Osborn, J. A. J. Chem. Soc., Chem. Commun. 1990, 869.  19.^(a) Noyori, R. Science 1990, 248, 1194; and reference therein. (b) Noyori, R. Chem. Soc. Rev. 1989, 18, 187. 20.^(a) Mezzetti, A.; Consiglio, G. J. Chem. Soc., Chem. Commun. 1991, 1675. (b) Chan, A. S. C.; Landis, C. R.; J. Mol. Catal. 1989, 49, 165. (c) Sanchez-Delgado, R. A.; Valencia, N.; Marquez-Silva, R.-L.; Andriollo, A.; Medina, M. Inorg Chem. 1986, 25 ,1106.  99  21.  Dickson, R. S. Homogeneous Catalysis with Compounds of Rhodium and Iridium, Catalysis by Metal Complexes Ser.; D. Reidel Publishing Co.: Dordrecht, 1985.  22.  Kagan, H. B. in Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, 1982; Vol. 8, Chapter 53; and references therein.  23.  (a) Brunner, H. Synthesis 1988, 645. (b) Brunner, H. J. Organomet. Chem. 1986, 300, 39.  24.  (a) Bosnich, B.; Fryzuk, M. D. in Topics in Inorganic and Organometallic  Stereochemistry; Geoffrey, G. L., Ed.; Wiley-Interscience: New York, 1981, p.119. (b) Bosnich, B. Asymmetric Catalysis, Martinus Nijhoff, Dordrecht, 1986. 25.  (a) Blystone, S. L. Chem. Rev. 1989, 89, 1663. (b) Caplar, V.; Comisso, G.; Sunjic, V. Synthesis 1981, 85.  26.  Knowles, W. S. Acc. Chem. Res. 1983, 16, 106.  27.  (a) Pino, P.; Consiglio, G. in Fundamental Research in Homogeneous Catalysis; Tsutsui, M., Ed.; Plenum: New York, 1979; Vol. 3, p 519. (b) Hayashi, T.; Kumada, M. in Fundamental Research in Homogeneous Catalysis; Ishii, Y.; Tsutsui, M., Eds.; 1978; Vol. 2, p 159.  28.  (a) Knowles, W. S.; Sabacky, M. J. J. Chem. Soc., Chem. Commun. 1968, 1445. (b) Homer, L.; Siegel, H.; Buthe, H. Angew. Chem., Int. Ed. Engl. 1968, 7, 941.  29.  (a) Dang, T. P.; Kagan, H. B. J. Chem. Soc., Chem. Commun. 1971, 481. (b) Kagan, H. B.; Dang, T. P. J. Am. Chem. Soc. 1972, 94, 6429.  30.  (a) Fryzuk, M. D.; Bosnich, B. J. Am. Chem. Soc. 1977, 99, 6262. (b) Fryzuk, M. D.; Bosnich, B. J. Am. Chem. Soc. 1978, 100, 5491.  31.^(a) Koenig, K. E. in Catalysis of Organic Reactions; Kosak, J. R., Ed.; Marcel Dekker: New York, 1984; Chapter 3. 100  (b) Marko, L.; Bakos, J. in Aspects of Homogeneous Catalysis; Ugo, R., Eds.; D. Reidel Publishing Co.: Dordrecht, 1981; Vol.4, p 145. 32.  (a) Landis, C. R.; Halpern, J. J. Am. Chem. Soc. 1987, 109, 1746, and references therein. (b) Halpern, J. in Asymmetric Synthesis; Morrision, J. D. Ed.; Academic Press: New York, 1985: Vol 5, Chapter 2.  33.  Pearson, R. G. Acc. Chem. Res. 1971, 4, 152.  34.  Kubas, G. J.; Ryan, R. R.; Swanson, B. I.; Vergamini, P. J.; Wassweman, H. J. J. Am. Chem. Soc. 1984, 106, 451.  35.  Jessop, P. G.; Morris, R. H. Coord. Chem. Rev. 1992, 121, 155.  36^Crabtree, R. H. Acc. Chem. Res. 1990, 23, 95; and references therein. 37.  Crabtree, R. H.; Hamilton D. G. Adv.. Organomet. Chem. 1988, 28, 299; and references therein.  38.  (a) Kubas, G. J. Acc. Chem. Res. 1988, 21, 120; and references therein. (b) Kubas, G. J. Comments Inorg. Chem. 1988, 7, 17; and references therein.  39.  Ashworth, T. V.; Singleton, E. J. Chem. Soc., Chem. Commun. 1976, 705.  40.  Kubas, G. J. J. Chem. Soc., Chem. Commun. 1980, 61.  41.  Nageswara Rao, B. D.; Anders, L. R. Phys. Rev. 1965, 140, A112.  42.  Crabtree, R. H.; Lavin, M. J. Chem. Soc., Chem. Commun. 1985, 1661.  43.  Morris R. H.; Sawyer, J. F.; Shiralian, M.; Zubkowski, J. D. J. Am. Chem. Soc. 1985, 107, 5581.  44.  A recent Chemical Abstracts computer search (August 1992) on molecular hydrogen complexes gave a listing of more than 300 references. Some 101  representative examples are cited in references 46 through 52, with reference 46 to 50 cited from reference 45. 45.  Joshi, A. M. Ph.D Dissertation, The University of British Columbia, Vancouver, Canada, 1990.  46.  (a) Kubas, G. J.; Unkefer, C. J.; Swanson, B. I.; Fukushima, E. J. Am. Chem. Soc. 1986, 108, 7000. (b) Crabtree, R. H.; Hamilton D. G. J. Am. Chem. Soc. 1986, 108, 3124. (c) Wasserman, H. J.; Kubas, G. J.; Ryan, R. R. J. Am. Chem. Soc. 1986, 108, 2294. (d) Kubas, G. J.; Ryan, R. R.; Wrobleski, D. A. J. Am. Chem. Soc. 1986, 108, 1339. (d) Conroy-Lewis, F. M.; Simpson, S. J. J. Chem. Soc., Chem. Commun. 1986, 506.  47.^(a) Morris, R. H.; Earl, K. A.; Luck, R. L.; Lazarowych, N. J.; Sella, A. Inorg Chem. 1987, 26, 2674. (b) Kubas, G. J.; Ryan, R. R.; Unkefer, C. J. J. Am. Chem. Soc. 1987, 109, 8133. (c) Chinn, M. S.; Heinekey, D. M. J. Am. Chem. Soc. 1987, 109, 5865. (d) Bianchini, C.; Mealli, C.; Peruzzini, M.; Zanobini, F. J. Am. Chem. Soc. 1987, 109, 5548. 48.^(a) Bautista, M. T.; Earl, K. A.; Maltby P. A.; Morris R. H. J. Am. Chem. Soc. 1988, 110, 4056. (b) Bianchini, C.; Peruzzini, M.; Zanobini, F. J. Organomet. Chem. 1988, 354, C19. (c) Fontaine, X. L. R.; Fowles, E. H.; Shaw, B. L. J. Chem. Soc., Chem. Commun. 1988, 482. (d) Arliguie, T.; Chaudret, B.; Morris, R. H.; Sella, A. Inorg Chem. 1988, 27, 598. 49.^(a) Jia, G.; Meek, D. W. J. Am. Chem. Soc. 1989, 111,757 . (b) Arliguie, T.; Chaudret, B. J. Chem. Soc., Chem. Commun. 1989, 155. (c) Chinn, M. S.; Heinekey, D. M.; Payne, N. G.; Sofield, C. D. Organometallics 1989, 8, 1824. 102  ^  (d) Cappellani, E. P.; Maltby, P. A.; Morris, R. H.; Schweitzer, C. T.; Steele, M. R. Inorg Chem. 1989, 28, 4437. (e) Howdle, S. M.; Poliakoff, M. J. Chem. Soc., Chem. Commun. 1989, 1099. 50.^(a) Chinn, M. S.; Heinekey, D. M. J. Am. Chem. Soc. 1990, 112, 5116. (b) Van Der Sluys, L. S.; Eckert, J.; Eisenstein, 0.; Hall, J. H.; Huffman, J. C.; Jackson, S. A.; Koetzle, T. F.; Kubas, G. J.; Vergamini, P. J.; Caulton, K. G. J. Am. Chem. Soc. 1990, 112, 4381. (c) Howdle, S. M.; Healy, M. A.; Poliakoff, M. J. Am. Chem. Soc. 1990, 112, 4804. (d) Harman, W. D.; Taube, H. J. Am. Chem. Soc. 1990, 112, 2261. (e) Jia, G.; Morris, R. H. Inorg Chem. 1990, 29, 581. (f) Amendola, P.; Antoniutti, S.; Albertin, G.; Bordignon, E. Inorg Chem. 1990, 29,318. (g) Mediatti, M.; Tachibana, G. N.; Jenson, C. M. Inorg Chem. 1990, 29, 3. 51.^(a) Li, Z.-W.; Taube, H. .1. Am. Chem. Soc. 1991, 113, 8946. (b) Bautista, M. T.; Cappellani, E. P.; Drouin, S. D.; Morris R. H.; Schweitzer C. T.; Sella A.; Zubkowski, J. J. Am. Chem. Soc. 1991, 113, 4876. (c) Earl, K. A.; Jia, G.; Maltby P. A.; Morris R. H. J. Am. Chem. Soc. 1991, 113, 3027. (d) Van Der Sluys, L. S.; Miller, M. M.; Kubas, G. J.; Caulton, K. G. J. Am. Chem. Soc. 1991, 113, 2513. (e) Chaudret, B.; Chung, G.; Eisenstein, O.; Jackson, S. A.; Lahoz, F. J.; Lopaz, J. A. J. Am. Chem. Soc. 1991, 113, 2314. (f) Jia, G.; Morris, R. H. J. Am. Chem. Soc. 1991, 113, 875. 52.^(a) Michos, D.; Luo, X.-L.; Crabtree, R. H. Inorg Chem. 1992, 31, 4245. (b) Mediati, M; Tachibana, G. N.; Jenson, C. M. Inorg Chem. 1992, 31, 1827. (c) Gusev, D. G.; Vymenits, A. B.; Bakhmutov, V. I. Inorg Chem. 1992, 31,1. (d) Bianchini, C.; Meli, A.; Peruzzini, M.; Frediani, P.; Bohanna, C.; Esteruelas, M. A.; Oro, L. A. Organometallics 1992, 11, 138. (e) Jia, G.; Lough, A.J.; Morris, R. H. Organometallics 1992, 11, 161. 53.^Hampton, C. R. S. M.; Cullen W. R.; James, B. R.; Charland, J-. P. J. Am. Chem. Soc. 1988, 110, 6918. 103  54.  Henderson, R. A. Transition Met. Chem. 1988, 13, 474.  55.  (a) Maseras, F.; Duran, M.; LledOs, A; Bertran J. Am. Chem. Soc. 1992, 114, 2922. (b) Maseras, F.; Duran, M.; Lled6s, A; Bertnin in Theoretical and Computional Models for Organic Chemistry; Formosinho, S. J., Csizmadia, I. G., Arnaut, L. G., Eds.; Kluwer Academic Publishers: Netherlands, 1991; Vol. 339, p375-396.  56.^(a) Lin, Z.; Hall, M. B. J. Am. Chem. Soc. 1992, 114, 6102. (b) Lin, Z.; Hall, M. B. Inorg Chem. 1992, 31, 4262. (c) Eckert, J.; Albinati, A.; White, R.P.; Bianchini, C.; Peruzzini, M. Inorg Chem. 1992, 31, 4241. (d) Riehl, J.-F.; Pelissier, M.; Eisenstein, 0. Inorg Chem. 1992, 31, 3345. (e) Pacchioni, G. J. Am. Chem. Soc. 1990, 112, 80. (0 Burden, J. K.; Pourian, M. R. Inorg Chem. 1988, 27, 4445. 57.^(a) Hampton, C. R. S. M.; Dekleva, T. W.; James, B. R.; Cullen, W. R. Inorg. Chim. Acta 1988, 145, 165. (b) Hampton, C. R. S. M.; Butler, I. R.; Cullen W. R.; James, B. R.; Charland, J.P.; Simpson, J. Inorg. Chem. in press. 58.  Ozin, G. A.; Garcia-Prieto, J. J. Am. Chem. Soc. 1986, 108, 3099.  59.  Hauge, R. H.; Kafafi, Z. H.; Margrave, J. L. Proceedings, The Physics and Chemistry of Small Clusters, Richmond, VA, Oct. 1986; through reference 38.  60.  Martensson, A.-S.; Nyberg, C.; Andersson, S. Phys. Rev. Lett. 1986, 57, 2045.  61.  Nolan, S. P.; Marks, T. J. J. Am. Chem. Soc. 1989, 111, 8538.  62.  Collman, J. P.; Wagenknecht, P. S.; Hembre, R. T.; Lewis, N. S. J. Am. Chem. Soc. 1990, 112, 1294.  63.  Hamilton D. G.; Crabtree, R. H. J. Am. Chem. Soc. 1988, 110, 4126.  104  64.  Sweaney, R. L.; J. Am. Chem. Soc. 1985, 107, 2374.  65.  Burdett, J. K.; Philips, J. R.; Pourian, M. R.; Poliakoff, M.; Turner, J. J.; Upmacis, R. Inorg. Chem. 1987, 26, 3054.  66.^Jean, Y.; Eisenstein, O.; Volatron, F.; Maouche, B.; Sefta, F. J. Am. Chem. Soc. 1986, 108, 6587. 67.  (a) Morris, R. H. Inorg Chem. 1992, 31, 1471. (b) Lever, A. B. P Inorg Chem. 1991, 30, 1980. (c) Lever, A. B. P Inorg Chem. 1990, 29, 1271.  68.  Camenzind, M. J.; James, B. R.; Dolphin, D.; Sparapany, J. W.; Ibers, J. A. Inorg Chem. 1988, 27, 3054.  69.  Joshi, A. M.; James, B. R. J. Chem. Soc., Chem. Commun. 1989, 1785.  70.  Crabtree, R. H., Lavin, M.; Bonneviot, L. J. Am. Chem. Soc. 1986, 108, 4032.  71.  (a) Sanders, J. K. M.; Hunter, B. K. Modern NMR Spectroscopy: A Guide for Chemists; Oxford University Press: Oxford, 1987; p 61-65. (b) Delrome, A. E. in Modern NMR Techniques for Chemistry Research; Pergamon Press: Oxford, 1987; p 164-167.  72.  Bautista, M. T.; Earl, K. A.; Maltby P. A.; Morris R. H. J. Am. Chem. Soc. 1988, 110, 7031.  73.  Cotton, F. A.; Luck, R. L. Inorg. Chem. 1989, 28, 6.  74.  Desrosiers, P. J.; Cai, L.; Lin, Z.; Richards, R.; Halpern, J. J. Am. Chem. Soc. 1991, 113, 4173.  75.  James, B. R.; Pacheco, A.; Rettig, S. J.; Thorburn, I. S.; Ball, R. C.; Ibers, J. A. J. Mol. Catal. 1987, 41, 147.  105  76.  Perrin, D. D.; Armarego, W.L.F.; Perrin, D.R. Purification of Laboratory Chemicals; Pergamon Press: Oxford, 2nd Edn., 1980.  77.  Priemer, H., Ph.D Dissertation, der Ruhr-Universitat Bochum, 1987.  78.^Kosolapoff, G. M.; Maier, L. Organic Phosphorus Compounds; WileyInterscience: New York, 2nd Edn., 1972; Vol. 1. (a) p.115. (b) p.130. 79.^Lipton, M. F.; Sorensen, C. M.; Sadler, A. C; Shapiro, R. H. J. Organomet. Chem. 1980, 186, 155. 80.  Joshi, A. M., M.Sc. Dissertation, The University of British Columbia, Vancouver, Canada, 1986; Chapter 2.  81.  Shriver, D. F.; Drezdon, M. A. The Manipulation of Air-Sensitive Compounds; Wiley: New York, 2nd Edn., 1986.  82.  Dekleva, T. W.; Thorburn, I. S.; James, B. R. Inorg. Chim. Acta 1985, /00, 49.  83.  Stephenson, T. A.; Wilkinson, G. J. Inorg. Nucl. Chem. 1966, 28, 945.  84.  Thorburn, I. S.; Rettig, S. J.; James, B. R. Inorg Chem. 1986, 25, 234.  85.  Joshi, A. M.; Thorburn, I. S.; Rettig, S. J.; James, B. R. Inorg. Chim. Acta 1992,  198-200, 283. 86.  Seddon, E. A.; Seddon, K. R. The Chemistry of Ruthenium; Elsevier: Amsterdam, 1984.  87.  Bruce, M. I. in Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, 1982; Vol. 4, Chapter 32.  106  88.  Schroder, M.; Stephenson, T. A. in Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D., McCleverty, J. A., Eds.; Pergamon Press: Oxford, 1987; Vol. 4, Chapter 45, and references therein.  89.  (a) Chatt, J.; Hayter, R. G. J. Chem. Soc. 1961, 772; 896; 5507. (b) Chatt, J.; Hayter, R. G. J. Chem. Soc. 1963, 6017.  90.  Kawano, H.; Ikariya, T.; Ishii, Y.; Saburi, M.; Yoshikawa, S.; Uchida, Y.; Kumobayashi, H. J. Chem. Soc., Perkin Trans. 1989, 1571.  91.  James, B. R.; McMillan, R. S.; Morris, R. H.; Wang, D. K. W. Adv. Chem. Ser. 1978, /67, 122.  92.  James, B. R.; Wang, D. K. W. Can. J. Chem. 1980, 58, 245.  93.  James, B. R.; Thompson, L. K.; Wang, D. K. W. Inorg. Chim. Acta 1978, 29, L237.  94.  Hoffman, P. R.; Caulton, K. G. J. Am. Chem. Soc. 1975, 97, 4221.  95.  Thorburn, 1. S., Ph.D Dissertation, The University of British Columbia, Vancouver, Canada, 1985.  96.  (a) James, B. R.; Preece, M.; Robinson, S. D. Adv. Chem. Ser. 1982, 196, 145. (b) James, B. R.; Preece, M.; Robinson, S. D. Inorg. Chim. Acta 1979, 34, L219.  97.  (a) Schafer, H.; Wiese, U.; Rinke, K.; Brendel, K. Angew. Chem., Int. Ed. Engl. 1967, 6, 253. (b) Dell'Amico, D. B.; Calderazzo, F.; Veracini, C. A.; Zandona, N. Inorg. Chem. 1984, 23, 3030.  98.^(a) Bown, M.; Waters, J. M. J. Am. Chem. Soc. 1990, 112, 2442. (b) Zanardo, A.; Michelin, R. A.; Pinna, F.; Strukul, G. Inorg. Chem. 1989, 28, 1648.  107  99.  Applied Photophysics Ltd. Stopped-Flow spectrometer (model SF.17MV) instruction manual; August 1989, p.16-17.  100.  (a) Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T. Numerical Recipes (the Art of Scientific Computing); Cambridge University Press: New York, 1987; Chapter 14. (b) Sprott, J. C. Numerical Recipes in Basic (Routines and Examples); Cambridge University Press; New York, 1991.  101.  Laidler, K. J . Chemical Kinetics; Harper and Row Publishers: New York, 3rd Edition, 1987; Chapter 2.  102.  Hampton, C.R.S.M. Ph.D Dissertation, The University of British Columbia, Vancouver, Canada, 1989.  103.  MacFarlane, K. S., private communication.  104.  Vaska, L.; Tadros, M. E. J. Am. Chem. Soc. 1971, 93, 7099.  105. James. B.R.; Markham, L.D.; Rattray, A.D.; Wang, D.K.W. Inorg. Chim. Acta 1976, 20, L25.  108  APPENDICES  109  APPENDIX A-1 X-ray Crystallographic Analysis of the Ru3C17(PPh3)3(DPPP)(H20) Cluster  C7 C CIO ..) C12^C19 C11 .  _s`  Cu6 CS1^CS AS^C63 C36^CS  Molecular Structure  110  C4  111  Experimental Details  A. Crystal Data  Empirical Formula^  C 87 H 88 C1 7 0 2.5 P 5 Ru 3  Formula Weight^  1879.90  Crystal Color, Habit^ black, prism Crystal Dimensions (mm)^0.250 X 0.250 X 0.400 Crystal System^  monoclinic  No. Reflections Used for Unit Cell Determination (28 range) ^25 ( 23.3 - 29.4°) Omega Scan Peak Width at Half-height^ Lattice Parameters:  0.38  ^ a ^19.153 17.534 b ^ c ^25.786 97.582 0  (2)A (4)A (2)A (9)6  V^8584 (2)A 3 Space Group^  P2 1 /c (#14)  Z value^  4  D  1.454 g/cm 3  calc^ ^ F 000 ^ P (MoKa)  3824  8.68 cm -1  B. Intensity Measurements  Diffractometer^  Rigaku AFC6S  Radiation^  MoKa (X 1. 0.71069 A)  Temperature^  21°C  Take-off Angle^  6.0°  Detector Aperture^  6.0 mm horizontal 6.0 mm vertical  112  Crystal to Detector Distance ^285 mm Scan Type Scan Rate^  32.0°/min (in omega) (8 rescans)  Scan Width^  (0.98 + 0.35 tanel)°  29 max^  5 50.0°  No. of Reflections Measured^Total: 16193 Unique: 15674 ( Rint  058)  Lorentz-polarization Absorption (trans. factors:^0.95 - 1.00). Secondary Extinction (coefficient:^0.70(3) E-07)  Corrections  C. Structure Solution and Refinement Structure Solution Refinement  ^  ^  Function Minimized  Full-matrix least-squares  ^  Least-squares Weights p-factor  Patterson Method  I w (IFol - iFc1)  ^  4Fo 2 /a 2 (Fo 2 )  ^  Anomalous Dispersion  2  0.00  ^  All non-hydrogen atoms  No.^Observations^(I>3.00o(I)) No.^Variables Reflection/Parameter Ratio  7499 965 7.77  Residuals:^R;^R w  0.038; 0.036  Goodness of Fit Indicator  1.52  Max Shift/Error in Final Cycle  0.16  ^ 0.34 e - /A! Maximum Peak in Final Diff. Map Minimum Peak in Final Diff. Map^ 0.70 e-/A' -  113  Table A-1.1: Selected Bond Lengths (A) with estimated standard deviations in  parentheses.  atom  atom  distance  atom  atom  distance  Ru(1)  C1(1)  2.566(2)  Ru(3)  0(1)  2.126(4)  Ru(1)  C1(2)  2.489(2)  P(1)  C(1)  1.830(7)  Ru(1)  C1(3)  2.433(2)  P(1)  C(4)  1.842(7)  Ru(1)  C1(5)  2.400(2)  P(1)  C(10)  1.841(7)  Ru(1)  P(1)  2.241(2)  P(2)  C(3)  1.848(7)  Ru(1)  P(2)  2.269(2)  P(2)  C(16)  1.833(7)  Ru(2)  C1(1)  2.512(2)  P(2)  C(22)  1.846(8)  Ru(2)  C1(2)  2.480(2)  P(3)  C(28)  1.863(7)  Ru(2)  C1(3)  2.412(2)  P(3)  C(34)  1.835(7)  Ru(2)  C1(4)  2.400(2)  P(3)  C(40)  1.839(7)  Ru(2)  P(3)  2.319(2)  P(4)  C(46)  1.841(7)  Ru(2)  P(4)  2.319(2)  P(4)  C(52)  1.850(7)  Ru(3)  C1(1)  2.591(2)  P(4)  C(58)  1.842(7)  Ru(3)  C1(4)  2.364(2)  P(5)  C(64)  1.826(8)  Ru(3)  C1(6)  2.291(2)  P(5)  C(70)  1.827(7)  Ru(3)  C1(7)  2.316(2)  P(5)  C(76)  1.817(7)  Ru(3)  P(5)  2.280(2)  114  Table A 1.2: Selected Bond Angles (°) with estimated standard deviations in parentheses. -  atom  atom  atom  angle  atom  atom  atom  angle  C1(1)  Ru(1)  C1(2)  80.11(5)  C1(6)  Ru(3)  0(1)  175.4(1)  C1(1)  Ru(_)  C1(3)  79.12(5)  C1(7)  Ru(3)  P(5)  91.74(7)  C1(1)  Ru(1)  C1(5)  90.22(6)  C1(7)  Ru(3)  0(1)  82.4(1)  C1(1)  Ru(1)  P(1)  173.46(6)  P(5)  Ru(3)  0(1)  93.7(1)  C1(1)  Ru(1)  P(2)  96.47(6)  Ru(1)  C1(1)  Ru(2)  81.95(5)  C1(2)  Ru(1)  C1(3)  78.78(5)  Ru(1)  C1(1)  Ru(3)  127.61(7)  C1(2)  Ru(1)  C1(5)  87.96(6)  Ru(2)  C1(1)  Ru(3)  92.93(5)  C1(2)  Ru(1)  P(1)  93.42(6)  Ru(1)  C1(2)  Ru(2)  84.15(5)  C1(2)  Ru(1)  P(2)  176.40(6)  Ru(1)  C1(3)  Ru(2)  86.83(5)  C1(3)  Ru(1)  C1(5)  164.19(6)  Ru(2)  C1(4)  Ru(3)  101.92(6)  C1(3)  Ru(1)  P(1)  100.70(6)  Ru(1)  P(1)  C(1)  113.4(2)  C1(3)  Ru(1)  P(2)  101.77(6)  Ru(1)  P(1)  C(4)  122.5(2)  C1(5)  Ru(1)  P(1)  88.56(7)  Ru(1)  P(1)  C(10)  114.0(2)  C1(5)  Ru(1)  P(2)  90.92(7)  C(1)  P(1)  C(4)  100.5(3)  P(1)  Ru(1)  P(2)  89.97(7)  C(1)  P(1)  C(10)  103.1(3)  C1(1)  Ru(2)  C1(2)  81.34(5)  C(4)  P(1)  C(10)  100.8(3)  C1(1)  Ru(2)  C1(3)  80.59(5)  Ru(1)  P(2)  C(3)  112.8(2)  C1(1)  Ru(2)  C1(4)  82.63(6)  Ru(1)  P(2)  C(16)  120.0(2)  C1(1)  Ru(2)  P(3)  166.76(6)  Ru(1)  P(2)  C(22)  116.8(2)  C1(1)  Ru(2)  P(4)  93.10(6)  C(3)  P(2)  C(16)  103.2(3)  C1(2)  Ru(2)  C1(3)  79.37(5)  C(3)  P(2)  C(22)  100.3(3)  C1(2)  Ru(2)  C1(4)  92.37(6)  C(16)  P(2)  C(22)  100.9(3)  C1(2)  Ru(2)  P(3)  87.19(6)  Ru(2)  P(3)  C(28)  114.3(2)  C1(2)  Ru(2)  P(4)  168.50(6)  Ru(2)  P(3)  C(34)  115.5(2)  C1(3)  Ru(2)  C1(4)  162.22(6)  Ru(2)  P(3)  C(40)  118.4(2)  C1(3)  Ru(2)  P(3)  103.87(6)  C(28)  P(3)  C(34)  101.2(3)  C1(3)  Ru(2)  F(4)  89.83(6)  C(28)  P(3)  C(40)  98.7(3)  115  C1(4)  Ru(2)  P(3)  91.29(7)  C(34)  P(3)  C(40)  106.2(4)  C1(4)  Ru(2)  P(4)  96.91(6)  Ru(2)  P(4)  C(46)  117.1(2)  P(3)  Ru(2)  P(4)  99.33(6)  Ru(2)  P(4)  C(52)  113.1(2)  C1(1)  Ru(3)  C1(4)  81.65(6)  Ru(2)  P(4)  C(58)  121.3(2)  C1(1)  Ru(3)  C1(6)  91.04(6)  C(46)  P(4)  C(52)  100.5(3)  C1(1)  Ru(3)  C1(7)  92.33(6)  C(46)  P(4)  C(58)  100.8(3)  C1(1)  Ru(3)  P(5)  175.30(7)  C(52)  P(4)  C(58)  100.9(3)  C1(1)  Ru(3)  0(1)  84.5(1)  Ru(3)  P(5)  C(64)  114.3(2)  C1(4)  Ru(3)  C1(6)  93.11(7)  Ru(3)  P(5)  C(70)  119.3(3)  C1(4)  Ru(3)  C1(7)  168.76(7)  Ru(3)  P(5)  C(76)  111.3(2)  C1(4)  Ru(3)  P(5)  93.94(7)  C(64)  P(5)  C(70)  101.2(3)  C1(4)  Ru(3)  0(1)  87.6(1)  C(64)  P(5)  C(76)  106.4(4)  C1(6)  Ru(3)  C1(7)  96.50(7)  C(70)  P(5)  C(76)  102.9(3)  C1(6)  Ru(3)  P(5)  90.83(7)  116  Table A-1.3: Final Atomic Coordinates (fractional) and B eg atom  x  (A2)* z  B  eg  Ru(1)  0.11941(3)  0.14018(3)  0.16202(2)  2.50(2)  Ru(2)  0.28224(3)  0.08938(3)  0.14464(2)  2.46(2)  Ru(3)  0.23149(3)  -0.08885(3)  0.20682(2)  3.22(3)  C1(1)  0.21111(8)  0.0562(1)  0.21667(6)  2.94(7)  C1(2)  0.16664(8)  0.06887(9)  0.09076(6)  2.93(7)  C1(3)  0.22688(8)  0.21108(9)  0.15465(6)  2.97(7)  C1(4)  0.30766(8)  -0.0447(1)  0.14827(7)  3.37(8)  C1(5)  0.03286(9)  0.0414(1)  0.16304(7)  3.76(9)  C1(6)  0.3186(1)  -0.0878(1)  0.27685(7)  4.6(1)  C1(7)  0.1400(1)  -0.1192(1)  0.25362(8)  5.0(1)  F(1)  0.04171(9)  0.2056(1)  0.10638(7)  3.03(8)  P(2)  0.0792(1)  0.1994(1)  0.23052(7)  3.24(9)  P(3)  0.3296(1)  0.1015(1)  0.06682(7)  3.27(9)  P(4)  0.37870(9)  0.1280(1)  0.20266(7)  2.98(8)  P(5)  0.2527(1)  -0.2143(1)  0.19178(8)  3.6(1)  0(1)  0.1479(2)  -0.0812(2)  0.1439(2)  4.0(2)  C(1)  -0.0479(3)  0.2030(4)  0.1237(3)  4.1(4)  C(2)  -0.0550(3)  0.2437(4)  0.1755(3)  4.5(4)  C(3)  -0.0179(4)  0.2034(4)  0.2241(3)  4.2(4)  C(4)  0.0513(4)  0.3083(4)  0.0942(3)  3.4(3)  C(5)  0.1043(4)  0.3507(4)  0.1203(3)  4.5(4)  C(6)  0.1065(5)  0.4294(5)  0.1110(4)  6.8(5)  C(7:)  0.0557(6)  0.4638(5)  0.0765(4)  7.8(6)  C(8)  0.0046(6)  0.4207(6)  0.0509(4)  8.1(6)  C(9)  0.0006(4)  0.3440(5)  0.0593(3)  5.6(5)  117  O CC.  Table A 1.3: Final Atomic Coordinates (fractional) and B eg (A2 )* (cont.) -  atom  x  y  z  B eq  C(10)  0.0307(4)  0.1684(4)  0.0390(3)  3.2(3)  C(11)  0.0752(4)  0.1943(4)  0.0045(3)  4.5(4)  C(12)  0.0707(5)  0.1622(5)  -0.0458(3)  5.6(5)  C(13)  0.0212(6)  0.1066(5)  -0.0602(3)  6.1(5)  C(14)  -0.0220(5)  0.0818(5)  -0.0264(4)  5.5(5)  C(15)  -0.0170(4)  0.1117(4)  0.0238(3)  4.3(4)  C(16)  0.1056(4)  0.2979(4)  0.2470(3)  3.4(3)  C(17)  0.1771(4)  0.3172(4)  0.2514(3)  4.2(4)  C(18)  0.2001(4)  0.3909(5)  0.2616(3)  5.6(5)  C(19)  0.1516(5)  0.4471(5)  0.2674(4)  6.3(5)  C(20)  0.0820(5)  0.4297(5)  0.2643(4)  6.4(5)  C(21)  0.0589(4)  0.3560(5)  0.2543(3)  5.0(4)  C(22)  0.0987(4)  0.1530(5)  0.2950(3)  4.1(4)  C(23)  0.1183(4)  0.1929(5)  0.3407(3)  5.3(4)  C(24)  0.1291(5)  0.1565(7)  0.3878(3)  7.1(6)  C(25)  0.1201(5)  0.0803(7)  0.3907(4)  7.6(6)  C(26)  0.1000(5)  0.0381(5)  0.3463(4)  7.1(6)  C(27)  0.0904(4)  0.0746(5)  0.2985(3)  5.1(5)  C(28)  0.2788(3)  0.0516(5)  0.0102(3)  3.8(4)  C(29)  0.2653(4)  0.0854(5)  -0.0382(3)  4.8(4)  C(30)  0.2253(5)  0.0504(6)  -0.0805(3)  6.4(5)  C(31)  0.1994(5)  -0.0200(7)  -0.0740(4)  6.9(6)  C(32)  0.2130(4)  -0.0568(5)  -0.0270(4)  6.1(5)  C(33)  0.2521(4)  -0.0216(5)  0.0155(3)  4.7(4)  C(34)  0.4175(4)  0.0598(5)  0.0665(3)  4.2(4)  C(35)  0.4252(4)  -0.0163(5)  0.0563(3)  5.1(4)  118  0CC.  Table A 1.3: Final Atomic Coordinates (fractional) and B eg (A 2 )* (cont.) -  atom  x  Y  z  B eg  C(36)  0.4918(6)  -0.0514(6)  0.0611(4)  6.9(6)  C(37)  0.5499(5)  -0.0068(8)  0.0757(4)  7.9(7)  C(38)  0.5450(5)  0.0692(7)  0.0865(4)  7.2(6)  C(39)  0.4776(4)  0.1031(5)  0.0823(3)  5.6(5)  C(40)  0.3331(4)  0.1969(4)  0.0378(3)  3.7(4)  C(41)  0.3875(4)  0.2230(5)  0.0122(3)  5.6(5)  C(42)  0.3833(6)  0.2948(6)  -0.0103(4)  7.6(6)  C(43)  0.3252(8)  0.3402(6)  -0.0069(4)  8.1(7)  C(44)  0.2703(5)  0.3136(5)  0.0162(4)  6.7(6)  C(45)  0.2754(4)  0.2425(5)  0.0391(3)  4.5(4)  C(46)  0.4344(3)  0.2037(4)  0.1805(3)  3.1(3)  C(47)  0.4029(3)  0.2621(4)  0.1500(3)  3.7(4)  C(48)  0.4428(4)  0.3203(4)  0.1331(3)  4.8(4)  C(49)  0.5156(5)  0.3206(5)  0.1471(3)  5.7(5)  C(50)  0.5473(4)  0.2620(5)  0.1769(3)  5.5(5)  C(51)  0.5071(4)  0.2048(4)  0.1936(3)  4.2(4)  C(52)  0.3540(3)  0.1719(4)  0.2629(3)  3.2(3)  C(53)  0.3278(4)  0.1268(4)  0.3001(3)  4.0(4)  C(54)  0.3092(4)  0.1577(5)  0.3452(3)  5.2(5)  C(55)  0.3174(4)  0.2344(6)  0.3546(3)  5.5(5)  C(56)  0.3434(4)  0.2797(5)  0.3192(3)  5.0(4)  C(57)  0.3617(3)  0.2495(4)  0.2728(3)  3.9(4)  C(58)  0.4462(3)  0.0593(4)  0.2305(3)  3.3(3)  C(59)  0.4802(4)  0.0657(4)  0.2812(3)  5.0(4)  C(60)  0.5353(5)  0.0167(6)  0.2992(3)  6.5(5)  C(61)  0.5565(4)  -0.0376(5)  0.2672(4)  6.0(5)  119  0CC.  Table A 1.3: Final Atomic Coordinates (fractional) and B eg (A 2)* (cont.) x y z atom -  B eg  O CC .  C(62)  0.5243(4)  -0.0435(5)  0.2172(4)  5.6(5)  C(63)  0.4689(4)  0.0042(4)  0.1985(3)  4.4(4)  C(64)  0.3295(4)  -0.2314(4)  0.1582(3)  4.3(4)  C(65)  0.3269(5)  -0.2746(6)  0.1139(4)  8.3(6)  C(66)  0.3892(7)  -0.2897(8)  0.0925(5)  C(67)  0.4532(6)  -0.2663(7)  0.1179(5)  9.4(8)  C(68)  0.4537(5)  -0.2253(6)  0.1613(5)  7.1(6)  C(69)  0.3942(4)  -0.2063(5)  0.1818(3)  5.3(4)  C(70)  0.2699(3)  -0.2796(4)  0.2472(3)  3.8(4)  C(71)  0.2487(4)  -0.2652(5)  0.2947(3)  5.1(4)  C(72)  0.2578(4)  -0.3188(6)  0.3344(4)  6.5(5)  C(73)  0.2888(5)  -0.3879(6)  0.3260(4)  6.5(6)  C(74)  0.3111(5)  -0.4021(5)  0.2791(4)  6.8(5)  C(75)  0.3027(4)  -0.3494(5)  0.2401(3)  5.4(4)  C(76)  0.1783(4)  -0.2585(4)  0.1519(3)  3.9(4)  C(77)  0.1593(4)  -0.2328(4)  0.1002(4)  5.6(5)  C(78)  0.1005(5)  -0.2631(5)  0.0702(4)  6.4(5)  C(79)  0.0603(5)  -0.3164(6)  0.0902(5)  6.8(6)  C(80)  0.0776(5)  -0.3408(5)  0.1398(4)  6.2(5)  C(81)  0.1375(4)  -0.3134(4)  0.1709(3)  4.7(4)  0(2)  0.745(1)  0.070(1)  0.480(1)  27(2)  0.90  C(82)  0.655(4)  0.281(4)  0.521(3)  21(3)  0.32  C(83)  0.707(2)  0.001(3)  0.395(2)  24(4)  0.52  C(84)  0.754(5)  0.125(4)  0.587(2)  20(2)  0.41  C(85)  0.746(3)  0.195(3)  0.542(2)  20(2)  0.47  C(86)  0.620(4)  0.242(4)  0.541(3)  19(2)  0.30  120  12(1)  .  Table A-1.3: Final Atomic Coordinates (fractional) and B (A 2 )* (cont.) atom  x  y  z  B  eg  OCC.  C(87)  0.795(2)  0.133(3)  0.564(3)  20(2)  0.56  C(88)  0.784(2)  0.098(2)  0.521(1)  21(3)  0.74  C(89)  0.719(4)  0.071(4)  0.576(3)  21(3)  0.34  C(90)  0.694(3)  0.183(3)  0.559(2)  20(2)  0.44  C(91)  0.734(2)  0.029(2)  0.428(1)  26(3).  0.90  C(92)  0.438(3)  0.008(2)  0.4252(8)  24(3)  0.83  C(93)  0.472(2)  -0.009(3)  0.493(3)  22(4)  0.72  C(94)  0.438(4)  -0.041(4)  0.469(4)  20(3)  0.43  C(95)  0.494(3)  0.012(4)  0.455(3)  17(2)  0.52  *B  eg^  2 (8/3)n ZEU..a.*a.3*(a..a.3) 13  APPENDIX A-2 Computer Programs in BASIC used in Section 4.5 ABSTOCON - This is the program that converts the initial absorbance versus time data into [9a] versus time data. CLS INPUT "How many data pairs do you wish to input?", NPT INPUT "What is the name of the Absorbance file?", DUM$ OPEN DUM$ FOR INPUT AS #1 LINE INPUT #1, DUM$ INPUT #1, NPT LINE INPUT #1, DUM$ LINE INPUT #1, DUM$ DIM TIME(NPT), AB(NPT), CONC(NPT), SIG(NPT) FOR J = 1 TO NPT INPUT #1, TIME(J), AB(J) NEXT J INPUT "What is the value of Mt? ", Mt INPUT "What is the value of Ht? ", Ht INPUT "What is the value of K? ", K INPUT "What is the value of Ao? ", INAB INPUT "What is the uncertainty in Ao? ", DINAB INPUT "What is the value of Ae? ", AEQ INPUT "What is the uncertainty in Ae? ", DAEQ B = -(Mt + Ht + 1 / K) C = Mt * Ht Q = -.5 * (B + SGN(B) * (B A 2 - 4 C) (.5)) R1 =Q R2 = C / Q IF R1 < Mt AND R1 < Ht THEN R = R1 ELSE R = R2 AMP = AEQ - INAB DAMP = DINAB + DAEQ FOR I = 1 TO NPT CONC(I) = (AB(I) - INAB) * R / AMP SIG(I) = (((DAMP / (AB(I) - INAB)) 2 + (DAMP / AMP) 2) (.5)) * CONC(I) NEXT I INPUT "Under what name do you wish to save the converted file?", DUM$ OPEN DUM$ FOR OUTPUT AS #2 PRINT #2, "NPT=" WRITE #2, NPT A  122  PRINT #2, "" PRINT #2, "X Y" FOR J = 1 TO NPT WRITE #2, TIME(J), CONC(J), SIG(J) NEXT J CLOSE #1 CLOSE #2 END  123  MODSPLOT This is the program that generates the calculated [9a] vs. time plots which can be used to compare graphically with the observed [9a] versus time data. -  DECLARE SUB RKPLOT (YSTART!(), YCOPY!(), xl!, x2!, XDAT!(), YDAT!(), SIG!(), NPT, NVAR!) DECLARE SUB PLOT (XDAT!(), YDAT!(), X!0, Y!0, NDATA!, NTHEOR!, SIG!0) DECLARE SUB ODEINT (YSTART!(), NVAR!, xl!, x2!, EPS!, Hi!, HMIN!, NOK!, NBAD!, DUM1!, DUM2!) COMMON SHARED KMAX, KOUNT, DXSAV, XP(), YP() COMMON SHARED Mt, Ht, Keq, kl 'This is the main module DO CLS DIM XP(200), YP(10, 200) NVAR = 1 DIM YSTART(NVAR) DIM YCOPY(NVAR) INPUT "What is the initial time in seconds?", xl INPUT "What is the final time in seconds?", x2 INPUT "What is the initial dimer concentration?", Mt INPUT "What is the initial dihydrogen concentration?", Ht INPUT "What is the value of K?", Keq YSTART(1) = 0 YCOPY(1) = YSTART(1) PRINT "Input a filename containing raw data for comparison" INPUT DUM$ OPEN DUM$ FOR INPUT AS #1 LINE INPUT #1, DUM$ INPUT #1, NPT DIM XDAT(NPT), YDAT(NPT), SIG(NPT) LINE INPUT #1, DUM$ LINE INPUT #1, DUM$ FOR I = 1 TO NPT INPUT #1, XDAT(I), YDAT(I), SIG(I) NEXT I DO CALL RKPLOT(YSTARTO, YCOPY(), xl, x2, XDAT(), YDAT(), SIGO, NPT, NVAR) PRINT "Do you want to do further analysis on this data set (y/n)?" INPUT DUM$ IF DUM$ = "n" THEN EXIT DO LOOP 124  PRINT "Any other data you wish to analyse (y/n)?" INPUT DUM$ IF DUM$ = "n" THEN END ERASE XP, YP, XDAT, YDAT, SIG, YSTART, YCOPY CLOSE #1 LOOP END  SUB DERIVS (X, YO, DYDXO) DYDX(1) = kl * ((Mt - Y(1)) * (Ht - Y(1)) - Y(1) / Keq) END SUB  SUB PLOT (XDATO, YDATO, X(), YO, NDATA, NTHEOR, SIGO) DO SCREEN 2 CLS 2 VIEW PRINT 1 TO 4 LOCATE 1, 1 PRINT "Enter X1,X2 (X1=X2 to stop)" INPUT xl, x2 IF xl = x2 THEN EXIT SUB PRINT "Enter Y1,Y2" INPUT Yl, Y2 CLS VIEW (50, 35)-(550, 180) WINDOW (0, 0)-(500, 145) LINE (0, 0)-(500, 145), B DX = (x2 - xl) / 500 'X Units per pixel DY = (Y2 - Y1) / 145 'Y Units per pixel FOR K = 1 TO NDATA SX = INT((XDAT(K) - xl) / DX) SY = INT((YDAT(K) - Y1) / DY) ERRY = INT(SIG(K) / DY) CIRCLE (SX, SY), 1 LINE (SX, SY + ERRY)-(SX, SY - ERRY) NEXT K FOR L = 1 TO NTHEOR C1X = INT((X(L) - xl) / DX) C1Y = INT((Y(L) - Y1) / DY) IF L < > 1 THEN LINE (C2X, C2Y)-(C1X, C1Y) 125  C2X = ClX C2Y = ClY NEXT L LOOP END SUB SUB RKPLOT (YSTARTO, YCOPYO, xl, x2, XDATO, YDATO, SIGO, NPT, NVAR) DO DO INPUT "Estimate kl", kl EPS = .0001 INPUT "Estimate the required stepsize", H1 HMIN = 0! KMAX = 200 DXSAV = (x2 - xl) / (x2 - xl) CALL ODEINT(YSTARTO, NVAR, xi, x2, EPS, H1, HMIN, NOK, NBAD, DUM, RKQC) PRINT "Successful steps:^"; NOK PRINT "Bad steps:^"; NBAD PRINT "Stored intermediate values:"; KOUNT PRINT "Press return to continue" INPUT DUM$ PRINT "^t^[MH2]^ti FOR I = 1 TO KOUNT PRINT USING "####.# "; XP(I); PRINT USING "##.####' "; YP(1, 1) NEXT I FOR J = 1 TO NVAR YSTART(J) = YCOPY(J) NEXT J PRINT "Do you wish to try another value of kl (y/n)?" INPUT FLAG$ IF FLAG$ = "n" THEN EXIT DO LOOP PRINT "Do you wish to see a plot of the data (y/n)?" INPUT FLAG$ DO IF FLAG$ = "n" THEN EXIT DO VAR# = 1 DIM YTHEOR(KOUNT) FOR I = 1 TO KOUNT YTHEOR(I) = YP(VAR#, I) NEXT I CALL PLOT(XDATO, YDATO, XPO, YTHEORO, NPT, KOUNT, SIG()) -  126  SCREEN 0 PRINT "Do you wish to save the calculated data set (y/n)?" INPUT FLAG$ IF FLAG$ = "y" THEN INPUT "Document to be saved:", DUM$ OPEN DUM$ FOR OUTPUT AS #2 FOR I = 1 TO KOUNT WRITE #2, XP(I), YTHEOR(I) NEXT I CLOSE #2 END IF PRINT "Do you wish to try another variable? (y/n)" INPUT FLAG$ ERASE YTHEOR LOOP PRINT "Do you wish to plot another value of kl (y/n)?" INPUT FLA G$ IF FLAG$ = "n" THEN EXIT DO LOOP END SUB The actual integrator programs, RK4, RKQC and ODEINT, were taken directly from numerical recipes without any modification.  127  APPENDIX A-3 Derivation of Equation 4.22 The equilibrium expression for the conversion of [RuCl(DPPB)2(11-C1)12, 9, with H2 to (71 2-H2)Ru(DPPB)(1-C1)3RuCl(DPPB), 9a, can be expressed as:  [RuCl(DPPB)(11-C1)) 2 + H2 N. lc ^[(► 2-H2 )(DPPB)Ru(p,-C1) 3 RuC1(DPPB)] (A.1) !  1  9a  9^  The rate expressions for the appearance of 9a can be written as d[9a]  = k i [9][H 2 ]—k_ i [9a]^ (A.2) dt^ (A.3) Let^x = [9a] at time t^ xe = [9a] in the equilibrium system^(A.4) MT = total metal concentration (A.5) = [9] at time t = 0^ (A.6) HT = [ H2] at time t = 0^  Then MT x = [9] at time t^ HT x = [112] at time t^ MT x e= [9] at equilibrium^ HT xe= [H2] at equilibrium^ -  -  -  -  (A.7) (A.8) (A.9) (A.10)  The equilibrium expression can be written as: K—^ [MT x] [HT]  (A.11)  and equation A.2 can be rewritten as: dx^rri  dt  —xp--vr --x ) -  128  (A.12)  In terms of absorbance: A = e' 9 [9] + e'9 a [9a] ei9 [MT x] + e 9a x^ where e' 9 = extinction coefficient for 9 / path length  (A.13)  t  = e 9 /1^ & e l 9a = E9a /1^  (A.14) (A.15)  If A o = 8' 9 M T^(A.16) then A = A o + (e' 9 a —e 9 )x^ -A or x — AAe ^ '0  (A.17) (A.18)  where Ac' = 9 a —e9 To eliminate De', one can write from A.13: Ae = e l 9 [MT — xe]+ e l 9a xe (A.19)  = Ao + AE' xe Ae Or^De' = ^ xe  (A.20)  -  Combining A.18 and A.20 gives: (  x=  A-A s )xe A e Ao  (A.21)  —  From the equilibrium expression xe [MT - x e ] [HT —x e ]  K= To solve for xe:  u ,2— 1/4 .T _^ xe e^= .s  nut j  -I-  1.1.1' I  116.  pt  (A.22)  Let^B = — (M T + HT + K-1 ) and C = M TH T^(A.23)  129  Also let  ^ ^q = — 1 [B + (sign of B)I7--(A.24) —4C]  The two roots from equation A.22 will be R2= 2=C  R1 = q  A. —  g  (A.25)  With B < 0, C > 0 and B 2 > 4C, there will be two positive roots. Choose the one with x e 5 MT and HT •••  R = root with x e 5 MT and HT  Then equation A.21 becomes x = ( = (  AA ° )R^ Ae — A c3,  (A.26)  A-A ° )R^where A' = Ae — Ao^(A.27) A'  In Section 4.5.2.1 (Equation 4.22, p.84), x is written as [9a] and R is [9a] e .  130  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0061778/manifest

Comment

Related Items