Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Solution H₂S Chemistry of Pd-bis(Diphenylphosphino)methane (DPM) complexes; catalytic conversion of H₂S… Wong, Terrance Yu Hung 1996

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

Item Metadata

Download

Media
831-ubc_1996-148572.pdf [ 10.8MB ]
Metadata
JSON: 831-1.0059521.json
JSON-LD: 831-1.0059521-ld.json
RDF/XML (Pretty): 831-1.0059521-rdf.xml
RDF/JSON: 831-1.0059521-rdf.json
Turtle: 831-1.0059521-turtle.txt
N-Triples: 831-1.0059521-rdf-ntriples.txt
Original Record: 831-1.0059521-source.json
Full Text
831-1.0059521-fulltext.txt
Citation
831-1.0059521.ris

Full Text

S O L U T I O N H 2 S C H E M I S T R Y O F Pd-BIS(DIPHENYLPHOSPHINO)METHANE (DPM) C O M P L E X E S ; C A T A L Y T I C C O N V E R S I O N O F H 2 S T O H 2 By T E R R A N C E Y U H U N G W O N G B.Sc, The University of British Columbia, 1990 A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F PHILOSOPHY in T H E F A C U L T Y O F G R A D U A T E STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH C O L U M B I A June 1996 © Terrance Yu Hung Wong, 1996 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. Department of CH£M($7KY The University of British Columbia Vancouver, Canada Date 5BPT 4. /796 DE-6 (2/88) A B S T R A C T This thesis describes studies on the interaction of H 2S with mono- and dinuclear Pd - dpm (bis(diphenylphosphino)methane) complexes with the ultimate goal being to catalyze the conversion of H 2S to H 2 and elemental sulfur (Sg). Solution kinetic and mechanistic studies were performed on the abstraction of sulfur from Pd2X2(u-S)(n-dpm)2 [X = Cl (2a), Br (2b), I (2c)] using dpm to give, respectively, Pd2X2(dpm)2 [X = Cl (la), Br (lb), I (lc)] and dpm(S) (process 2 -> 1): H 2 S ^ —*> H 2 P-" ~-P P-^~~~-P I I I s I Scheme 1 X — P d P d — X v ^ P d - " T d I I X I I ' P P P P 1 \ y 2 dpm(S) -«a dpm The reaction is first-order in both 2 and dpm with the reactivity trend in CHCI3 being X = Cl > Br > I. The activation parameters for the chloride system are AH* = 41 ± 3 kJ mol"1 and AS* = -127 + 10 J K"1 mol"1, and for the bromide, AH* = 38 ± 1 kJ mol"1 and AS* = -144 + 4 JK"1 mol"1, showing that the entropy term is dominant in governing reactivity. No intermediates were seen in low temperature NMR studies. 2b and lb undergo rapid diphosphine ligand exchange with dpm-d2; with 2b, prior to any S-abstraction, Pd2Br2(p-S)(dpm)(dpm-d2) and Pd2Br2(p-S)(dpm-d2)2 are present, while Pd2Br2(dpm)(dpm-d2) and Pd2Br2(dpm-d2)2 are formed, as well as lb and dpm(S)-d2, in the abstraction reaction of 2b with dpm-d2. The distribution of products is close to statistical based on the stoichiometries of the reactants. Reaction of 2b with l,T-bis(diphenylphosphino)ethane (dpmMe) proceeds in an analogous way generating lb, Pd2Br2(dpm)(dprnMe), Pd2Br2(dpmMe)2, and the monosulfides dpm(S) and dpmMe(S). lb undergoes slow diphosphine ligand exchange, and 2b undergoes no ligand exchange, with dpmMe prior to S-abstraction. The findings are rationalized in terms of intermediates and transition states that include formulations with three equivalent diphosphines. No sulfur abstraction occurs on treating 2b with PPh3, PPh2Me, or Ph2P(CH2)3PPh2. Some studies were carried out on the catalyzed reaction H2S + dpm —» H 2 + dpm(S), which is the first reported, homogeneously catalyzed conversion of H 2S to H 2 . The removal of the bridged S atom from 2 using halogens (process 2 -» 9) was investigated, and kinetic and mechanistic studies on the X = I system in CHCI3 reveal that the reaction proceeds via oxidative addition of I2 to give Pd2l-4(dpm)2 (10c), which then undergoes unimolecular decomposition to give Pdl2(dpm) (9c); the sulfur is recovered in its elemental form (forward reaction - Scheme 2). The oxidative addition reaction is in the stopped-flow time regime, and is first-order in both 2c and I2; the activation parameters are AH* = 32 + 1 kJ mol"1 and AS* = -91 ± 3 J K"1 mol"1. The decomposition of 10c is slower and is simply first-order in 10c; the activation parameters are AH* = 80 ± 1 kJ mol"1 and AS* = -26 ± 3 J K"1 mol"1. Depending on reaction conditions, the by-product PdI2(dpm(S)) (11c) also forms via reaction of 9c with an Sn species where n < 8; 11c and the chloro analogue were synthesized by other routes and were characterized by X-ray analyses which reveal the envelope configuration of the five-membered pk-PPh2CH2P(S)Ph2 chelate ring. iii * * sulfur P- P Scheme 2 X ^•Pd P- p X X 2 2 H X - * H 2 S 9 The dinucleation reaction (9 -> 2) (Scheme 2) was studied in DMSO and on alumina in CHCI3, and proceeds rapidly to form 2 and HX. In DMSO, the reaction is established as an equilibrium; however, a side-product, probably Pd(SH)2(dpm), also forms. In CHCI3, the 9 -> 2 reaction proceeds completely in the presence of an alumina catalyst which functions to activate the H 2S prior to the dinucleation process; 2 is thought to form via the coupling of the undetected, intermediate PdX(SH)(dpm). With important catalytic implications seen for the X = I system, some photodecomposition studies of HI in the presence of alumina were performed; formation of I2 occurs, while the concomitantly formed H atoms react with CHCI3 to generate C H 2 X 2 species where X = CI and/or I. iv T A B L E O F C O N T E N T S Abstract ii List of Tables xii List of Figures xiv List of Non-standard/Less Common Abbreviations xxiii Numerical Key to Palladium Complexes xxvi Acknowledgements xxvii 1 INTRODUCTION 1 1.1 Properties, occurrences, and uses of H 2 S 2 1.2 Recovery of both H 2 and elemental sulfur from H 2 S 6 1.2.1 Thermodecomposition of H 2 S 7 1.2.2 Photodecomposition of H 2 S 12 1.2.3 Electrochemical decomposition of H 2 S 18 1.2.4 Plasmachemical decomposition of H 2 S 22 1.2.5 Radiolytic decomposition of H 2S 23 1.3 Homogeneous catalytic conversion of H 2 S to H 2 24 1.4 Transition metal - H 2 S chemistry 28 1.5 The chemistry of transition metal - dpm complexes 33 1.6 Statement of the problem 36 1.7 References for Chapter 1 37 2 GENERAL EXPERIMENTAL PROCEDURES 46 v 2.1 Materials 47 2.1.1 Dibenzylidenacetone (dba) 50 2.1.2 l,l'-bis(diphenylphosphino)ethane (dpmMe) 50 2.1.3 Triphenylphosphine sulfide (Ph3PS) 51 2.1.4 Bis(diphenylphosphino)methane monosulfide (dpm(S)) 51 2.1.5 rra«5-dichlorobis(benzonitrile)palladium(II) (PdCl2(PhCN)2) 52 2.1.6 Tris(diben2ylideneacetone)dipalladium(0)-chloroforrn solvate (Pd2(dba)3*CHCl3) 52 2.1.7 Dichlorobis-p-[bis(diphenylphosphino)methane]dipalladium(I) (Pd2Cl2(dpm)2) 53 2.1.8 Dibromobis-p-[bis(diphenylphosphino)methane]dipalladium(I) (Pd2Br2(dpm)2) 53 2.1.9 Diiodobis-p-[bis(diphenylphosphino)methane]dipalladium(I) dichloromethane solvate (Pd 2I 2(dpm) 2»CH 2Cl 2) 54 2.1.10 Dichloro-p-sulfidobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2Cl2(u-S)(dpm)2) 54 2.1.11 Dibromo-p-sulfidobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2Br2(p-S)(dpm)2) 55 2.1.12 Diiodo-p-sulfidobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2I2(p-S)(dpm)2) 55 2.1.13 Dichloro[bis(diphenylphosphino)methane]palladium(II) (PdCl2(dpm)) 56 2.1.14 Dibromo[bis(diphenylphosphino)methane]palladium(II) (PdBr2(dpm)) 56 2.1.15 Diiodo[bis(diphenylphosphino)methane]palladium(II) (Pdl2(dpm)) 57 2.1.16 Dichloro-p.-sulfoxobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2Cl2(p-SO)(dpm)2) 57 2.1.17 Diiodo-p-sulfoxobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2I2(p-SO)(dpm)2) 58 vi 2.1.18 Dimercapto-(x-sulfidobis-u,-[bis(diphenylphosphino)methane] (Pd2(SH)2(u-S)(dpm)2) 58 2.2 Instrumentation 59 2.2.1 Nuclear magnetic resonance spectroscopy 59 2.2.2 Electronic absorption spectroscopy 60 2.2.3 Stopped-flow electronic absorption spectroscopy 60 2.2.4 Thin-layered chromatography 60 2.2.5 Gas chromatography 61 2.2.6 Photochemistry 61 2.2.7 Electron spin resonance spectroscopy • 62 2.2.8 Elemental, mass, and X-ray crystallographic and photoelectron analyses 62 2.3 Treatment of kinetic data 62 2.4 References for Chapter 2 65 3 KINETIC AND MECHANISTIC ASPECTS OF SULFUR ABSTRACTION F R O M Pd2X2(u-S)(dpm)2 USING dpm OR dpmMe and C A T A L Y T I C CONVERSION OF H 2 S TO H 2 67 3.1 Introduction 68 3.2 Results 69 3.3 Discussion 98 3.4 Experimental section 123 3.4.1 Preparation of dpm-d2 123 3.4.2 Preparation of dpm(S)-d2 124 3.4.3 Preparation of dpmMe(S) 124 vii 3.4.4 Preparation of dpm(S)2, dpe(S), dpe(S)2, dpp(S), and dpp(S)2 125 3.4.5 Preparation of H 2S 126 3.4.6 Preparation of Pd2Br2(dpm-d2)2'2H20 (4b) 126 3.4.7 Preparation of Pd2Br2(n-S)(dpm-d2)2 (8b) 128 3.4.8 Kinetic measurements 128 3.4.9 Mechanistic studies 129 3.4.10 Catalytic conversion of H 2S 131 3.5 References for Chapter 3 133 4 KINETIC AND MECHANISTIC ASPECTS OF SULFUR ABSTRACTION F R O M Pd2X2(u-S)(dpm)2 USING HALOGENS 135 4.1 Introduction 136 4.2 Results 140 4.3 Discussion 165 4.4 Experimental section 181 4.4.1 Preparation of PdCl2(dpm(S)) (11a) 181 4.4.2 Preparation of PdBr2(dpm(S)) (lib) 182 4.4.3 Preparation of PdI2(dpm(S)) (11c) 182 4.4.4 Preparation of PdCl2(dpm(S)2) 183 4.4.5 Kinetic studies 183 4.4.6 Mechanistic studies 184 4.4.7 X-ray crystallographic analysis of PdI2(dpm(S)>0.5CH2Cl2 188 4.5 References for Chapter 4 189 viii 5 PRELIMINARY STUDIES ON THE REACTION OF PdX2(dpm) WITH H 2 S IN DIMETHYLSULFOXIDE; FORMATION OF Pd2X2(u-S)(dpm)2 194 5.1 Introduction 195 5.2 Results 196 5.3 Discussion 220 5.4 Experimental section 229 5.4.1 Preparation of gaseous HI 229 5.4.2 Reaction of PdX2(dpm) (9) with H 2S 229 5.4.3 Reaction of PdX2(dpm) (9) with elemental sulfur 231 5.4.4 Reaction of Pd2I2(u-S)(dpm)2 (2c) with I2 231 5.4.5 Reaction of Pd2X2(u-S)(dpm)2 (2) with H X 231 5.4.6 Reaction of Pd2(SH)2(p.-S)(dpm)2 with I2 232 5.4.7 Reaction of PdX2(dpm) (9) with NaSH 233 5.4.8 Photodecomposition studies of HI, HBr, and H 2 S 233 5.5 References for Chapter 5 234 6 REACTION OF PdX2(dpm) WITH H 2 S IN THE PRESENCE OF ALUMINA; C A T A L Y Z E D FORMATION OF Pd2X2(u-S)(dpm)2 236 6.1 Introduction 237 6.2 Results 239 6.3 Discussion 257 6.4 Experimental section 268 ix 6.4.1 NMR spectroscopic studies 268 6.4.2 Surface studies 270 6.4.3 Synthetic-scale studies 271 6.4.4 Photodecomposition of HI 273 6.5 References for Chapter 6 274 7 MISCELLANEOUS REACTIONS 276 7.1 Reaction of Pd2X2(u-S)(dpm)2 with alkyl halides 277 7.2 Reaction of PdX2(dpm) (9) with H 2 S in the presence of silica gel or aluminosilicate; catalyzed formation of Pd2X2(u-S)(dpm)2 (2) 281 7.3 Reaction of Pd2X2(u-S)(dpm)2 (2) with CO 282 7.4 References for Chapter 7 285 8 GENERAL CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE STUDIES 286 APPENDIX 291 Appendix I - Free energy of decomposition of H 2S, AGa 292 Appendix II - Kinetic data from the study of the reaction: Pd2X2(p.-S)(dpm)2 + dpm 294 Appendix III - Computer program to simulate the outcome of the reaction between Pd2Br2(p.-S)(dpm)2 and n mole equivalent dpm-d2 295 Appendix IV - Determination of the concentration of H 2 S in chloroform 299 Appendix V - Kinetic data from the study of the reaction: Pd2I2(pt-S)(dpm)2 + 12 301 Appendix VI - X-ray crystal structure data for PdI2(dpm(S)>0.5CH2C12 302 Appendix VII - Synthesis and characterization of [Pd(dpm(S))2]Cl2,0.5 CH 2C1 2, trans-b\s-r[-P, S-(bis(diphenylphosphino)methane monosulfide)palladium(II) chloride 0.5 CH 2C1 2 solvate 311 xi LIST O F T A B L E S 2.1 Some materials used and their sources. 48 2.2 Palladium complexes used and their respective reference(s) for synthetic procedures. 49 3.1 NMR data for the dinuclear palladium complexes, diphosphines, and the diphosphine monosulfides. 70 3.2 Temperature dependence for the bimolecular rate constant of the reaction of Pd2Cl2(u-S)(dpm)2 (2a) (kCi) and Pd2Br2(u-S)(dpm)2 (2b) (kB r) with dpm in CHC13. 75 3.3 Conditions used in the study of the catalysis of reaction 3.1.3 at R.T. using Pd2Br2(dpm)2 (lb) or Pd2Br2(u-S)(dpm)2 (2b). 93 4.1 NMR data for the palladium complexes. 141 4.2 Temperature dependence for the bimolecular (ki) and unimolecular (kD) rate constants for the formation and decomposition of the intermediate, Pd2I4(dpm)2 (10c), respectively, in CDC13. 150 4.3 Reaction of Pd2X2(u-S)(dpm)2 (2) with X 2 under various conditions; yield of by-product PdX2(dpm(S)) (11). 155 4.4 Activation parameters for some halogenation studies of dinuclear metal complexes ' and unsaturated systems. 169 4.5 Polarity (Q) of selected solvents at 25 °C (measured by (e-l)/(2e+l) where e is the dielectric constant). 169 5.1 N M R data for complexes and other compounds discussed in Chapter 5. 197 5.2 Results from studies on the reaction of PdX2(dpm) (9a, 1.8 x 10"2 M ; 9b, 1.5 x 10"2 M ; 9c, 1.3 x 10"2 M) with H 2 S at R.T. in DMSO-d 6 ; 1 = Pd2X2(dpm)2, 2 = Pd2X2(u-S)(dpm)2,11 = PdX2(dpm(S)), Y = probably Pd(SH)2(dpm). 199 5.3 Results from studies on the reaction of PdX2(dpm) (9a, 3.6 x 10"2 M ; 9b, 3.0 x 10"2 M ; 9c, 2.7 x 10"2 M) with 1 mole equivalent of elemental sulfur in DMSO-d 6 . 206 5.4 Results from studies on the reaction of Pd2X2(p-S)(dpm)2 (2a, 1.8 x 10"2 M ; 2b, 1.7 x 10"2 M ; 2c, 1.5 x 10"2 M) with HX(g). 208 xu 5.5 Results from photodecomposition studies of FA, HBr, and H 2 S. 216 6.1 Summary of the content make-up in Expts. 1 - 18; 9a = PdCl2(dpm). 240 1.1 Absolute free energies values, G , for H 2S, H 2 , S2(g), Sg(g), and Sg(s,l), and free energy of decomposition for H 2S, AGd(H2S), in the temperature range 298 to 1800K. 293 IV. 1 Dependence of the vapour pressure of CHC13 on temperature in the range T = 215.15 to 334.45 K. 299 VI. 1 Atomic coordinates of PdI2(dpm(S)>0.5CH2Cl2 (11c). 306 VI.2 Bond lengths (A) in the complex PdI2(dpm(S)>0.5CH2Cl2 (11c). 308 VI. 3 Bond angles (°) in the complex PdI2(dpm(S)>0.5CH2Cl2 (11c). 309 VII. 1 Atomic coordinates of [Pd(dpm(S))2]Cl2'0.5CH2Cl2. 317 VII.2 Bond lengths (A) of [Pd(dpm(S))2]Cl2»0.5CH2Cl2. 319 VII.3 Bond angles (°) of [Pd(dpm(S))2]Cl2'0.5CH2Cl2. 320 VTI.4 C - H — C l interactions in the complex [Pd(dpm(S))2]Cl2'0.5CH2Cl2. 322 xiii LIST O F FIGURES 3.1a Visible absorption spectral changes (350 - 550 nm region) as a function of time for a CHC13 solution of Pd2Br2(p-S)(p-dpm)2 (2b) (6.52 x 10"5 M) on addition of dpm (1.96 x 10"2M)at25 °C. 72 3.1b A rate-plot analyzed for a pseudo-first-order dependence on 2b; A t and Am represent the absorption at 428 nm at times t and co, respectively. 72 3.2a The dependence of the pseudo-first-order rate constants, k o b s , on [dpm] at [2b] = 6.52 x 10"5M, inCHCl 3 . 74 3.2b Eyring plot of the temperature dependence of the bimolecular rate constant, kB r. 74 3.3 a The dependence of the pseudo-first-order rate constants, k o b s , on [dpm] at [2a] = 6.52 x 10"5 M , in C H C I 3 at 30 °C. 76 3.3b Eyring plot of the temperature dependence of the bimolecular rate constant, kci. 76 3.4 The *H N M R (300 MHz) spectrum of the isolated Pd products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD 2C1 2 at R.T. 80 3.5 The 3 lP {Tij NMR (121 MHz) spectrum of the isolated Pd products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD 2C1 2 at R.T. 81 3.6 The lH NMR (300 MHz) spectrum of the isolated phosphine monosulfide products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD 2C1 2 at R.T. 82 3.7 The 31P{'H} NMR (121 MHz) spectrum of the isolated phosphine monosulfide products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD 2C1 2 at R T . 83 3.8 The 3 !p{ lH] NMR spectrum (121 MHz) of the completed in situ reaction between 2b and dpm-d2 (1:5 mole ratio) in CDC13 at R.T. (72 h); X = unknown. 84 xiv 3.9 The *H NMR (300 MHz) spectrum of the completed in situ reaction between 2b and dpm-d2 (1:5 mole ratio) in CDC13 at R.T. (72 h). 85 3.10 The ! H NMR spectrum (300 MHz) of the isolated Pd products from the reaction 2b and dpmMe (1:1 mole ratio) in CD 2C1 2 at R T . 87 3.11 The 31P{ 'H} NMR (121 MHz) spectrum of 2b on addition of 1 mole equivalent of dpm-d2; the spectrum was recorded at R.T. immediately after sample preparation in CDC13. 89 3.12 The ' H NMR spectrum (300 MHz) recorded at R.T. in CDC13 of lb on addition of 1 mole equivalent of dpmMe (72 h). 90 3.13 The ^ Pl'H} NMR spectrum (121 MHz) recorded at R T . in CDC13 of lb on addition of 1 mole equivalent of dpmMe (72 h). 91 3.14 The UV-vis spectrum of the isolated catalytically inactive species (in CHC13). 94 3.15 The ' H NMR spectrum (300 MHz) recorded at R.T. in CDC13 of the isolated catalytically inactive species. 95 3.16 The 31P{'H} NMR spectrum (121 MHz) recorded at R.T. in CDC13 of the isolated catalytically inactive species. 96 3.17 The UV-vis spectrum obtained from studies of the catalyzed dpm + H 2 S reaction using lb in C H C I 3 after 24 h (Expt. 2). 97 3.18 The ! H NMR spectrum (300 MHz) recorded at R.T. in CDC13 of 2b on addition of 1 mole equivalent of dpe (72 h). 105 3.19 The 31P{'H} NMR spectrum (121 MHz) recorded at R.T. in CDC13 of 2b on addition of 1 mole equivalent of dpe (72 h). 106 3.20 The *H (300 MHz) and 31P{1H} (121 MHz) NMR spectra recorded at R.T. in C D C I 3 of 2b on addition of 1 mole equivalent of dpp (72 h) (species W). 107 XV 3.21 The ' H NMR spectrum (300 MHz) recorded at R.T. in CDC13 of species W on addition of 3 mole equivalent of dpp (72 h). 109 3.22 The 31P{'H} NMR spectrum (121 MHz) recorded at R.T. in CDC13 of species W on addition of 3 mole equivalent of dpp (72 h). 110 3.23 The ' H (300 MHz) and 31*P{lH} (121 MHz) NMR spectra recorded at R.T. in CDCI3 of 2b on addition of 1 mole equivalent of dpm(S) (72 h) (species Y). 115 3.24 Mass spectrum (FAB, matrix = thioglycerol + CHCI3) obtained of species Y. 116 3.25 The UV-vis spectra of species isolated from reactions of Pd2X2(p-S)(dpm)2 (2) with 1 mole equivalent of dpm(S) in CHCI3 (X = Cl (species X), Br (species Y), I (species Z)). 117 3.26 Mass spectrum (FAB, matrix = thioglycerol + CHCI3) of the catalytically inactive species V H with the region from 1400 to 1650 m/z expanded. 122 3.27 Schematic diagram showing experimental setup for preparation of H 2 S from reaction of CaS with aq. HCl. 127 4.1 Preliminary ! H (3 00 MHz) and 3 *P {]H} (121 MHz) NMR study of the reaction of Pd2I2(p-S)(dpm)2 (2c, 1.6 x 10"2 M) with 1 mole equivalent I2 in CDC13 at R.T.; 9c = Pdl2(dpm), 10c = Pd2I4(dpm)2, 11c = PdI2(dpm(S)). 143 4.2 Mass spectrum (electron impact) of the yellow precipitate isolated from the synthetic-scale reaction of Pd2I2(u-S)(dpm)2 (2c) with I2 in CHC13 at R.T. 145 4.3 Visible absorption spectral changes (320 - 650 nm region) as a function of time (AT = 120 s) for a CHC13 solution of Pd2I2(u-S)(dpm)2 (2c, 9.4 x 10"5 M) on addition of I2 (3.8 x 10"4) at 25 °C. Inset: a rate-plot analyzed for a first-order dependence on Pd2I4(dpm)2 (10c); A» and A*, represent the absorption at 396 nm at times t and 00, respectively. 146 xvi 4.4 Electronic absorption spectra of Sg (cone. = 1.25 x 10"3 M ; X,, nm (E, M" 1 cm"1) = 244 (828), 266 (910)) and I2 (cone. = 3.94 x 10"4 M ; X, nm (s, M" 1 cm"1) = 512 (879) in CHC13 at R.T. 148 4.5 The dependence of the first-order rate constant, kobs, on [I2] for the decomposition of the intermediate Pd2I4(dpm)2 (10c) (formed from reaction with 9.4 x 10"5 M Pd2I2(u-S)(dpm)2 (2c)) in CHC13 at 25 °C. 149 4.6 Eyring plot of the temperature dependence for the unimolecular rate constants, kD, for the decomposition of Pd2I4(dpm)2 (10c). 149 4.7 Absorption spectral changes at X = 510 nm as a function of time for the formation of Pd2I4(dpm)2 (10c) in CHC13 at 25 °C from reaction of Pd2I2(p.-S)(dpm)2 (2c) (9.9 x 10"5M) with I2 (4.9 x 10"4M). 151 4.8 Rate-plot analyzed for a pseudo-first-order dependence on Pd2I2(u,-S)(dpm)2 (2c) (initial cone. = 9.9 x 10"5 M) during reaction with I2 (initial cone. = 4.9 x 10"4 M) in C H C I 3 at 25 °C to form the intermediate Pd2I4(dpm)2 (10c); A and A„ represent the absorption at 510 nm at times t and 00, respectively. 151 4.9 The dependence of the pseudo-first-order rate constants, kobs, for the formation of Pd2I4(dpm)2 (10c) on [I2] from reaction with Pd2I2(u-S)(dpm)2 (2c, 9.9 x 10"5 M) in CHCI3. 153 4.10 Eyring plot of the temperature dependence for the bimolecular rate constants, ki, for the formation of Pd2I4(dpm)2 (10c) from reaction of Pd2I2(p.-S)(dpm)2 (2c) with I2 in CHCI3. 153 xvii 4.11 ORTEP drawing and stereoview of PdI2(dpm(S)) (11c). Hydrogen atoms are omitted. 159 4.12 ORTEP and PLUTO drawings of PdCl2(dpm(S)) (1 la). Hydrogen atoms are omitted. 160 4.13 ! H (300 MHz) and "PfjH} NMR spectra of the "completed" reaction of Pd2Br2(u-S)(dpm)2 (2b) with 1 mole equivalent of Br 2 in CDC13 at R.T.; 9b = PdBr2(dpm), 1 lb = PdBr2(dpm(S)). 162 4.14a Visible absorption spectral changes (320-600 nm region) as a function of time for a C H 3 C N solution of Pd2I2(u-S)(dpm)2 (2c) (9.4 x 10"5 M) on addition of I2 (9.4 x 10" 5 M)at23 .5°C;9c = PdI2(dpm). 164 4.14b A rate-plot analyzed for an approximate second-order dependence on Pd2I4(dpm)2 (10c); At and A* represent the absorption at 360 nm at times t and oo, respectively. 164 5.1 Representative *H (200 MHz) and3*P{Ti} (81 MHz) N M R spectra showing results from the immediate reaction at R.T. in DMSO-d 6 of Pdl2(dpm) (9c, 1.3 x 10"2 M) with 0.5 mole equivalent H 2S in the absence of light (Schlenk techniques were employed); lc = Pd2I2(dpm)2, 2c = Pd2I2(p;-S)(dpm)2, Y = probably Pd(SH)2(dpm). 200 5.2 Representative *H (300 MHz) and 31P{1H} (121 MHz) N M R spectra showing results from the immediate reaction at R.T. in D M S O -d6 of PdBr2(dpm) (9b, 1.5 x 10"2 M) with 0.5 mole equivalent H 2 S in the absence of light (Schlenk techniques were employed); 2b = Pd2Br2(u-S)(dpm)2, Y = probably Pd(SH)2(dpm). 201 xviii 5.3 Representative T4 (3 00 MHz) and 3 !P {Ti} (121 MHz) NMR spectra showing results from the immediate reaction at R T . in D M S O -d6 of PdCl2(dpm) (9a, 1.8 x 10"2 M) with 0.5 mole equivalent H 2 S in the absence of light (Schlenk techniques were employed); 2a = Pd2Cl2(u-S)(dpm)2, Y = probably Pd(SH)2(dpm). 202 5.4 Representative *H (200 MHz) and 31P{1H} (81 MHz) NMR spectra showing results from the exposure of the iodide system (reaction of Pdl2(dpm) (9c, 1.3 x 10"2 M) with 0.5 mole equivalent H 2S in the absence of light using Schlenk techniques) to laboratory light for 5 d; 11c = PdI2(dpm(S)). 204 5.5 *H (200 MHz) and 3IP{T4} (81 MHz) NMR spectra showing results from the reaction of Pd2Br2(u-S)(dpm)2 (2b, 1.7 x 10"2 M) with 2 mole equivalent HCl at R.T. in DMSOd- 6 ; spectra recorded 1 h after sample preparation. 210 5.6 *H (200 MHz) and 31P{ Ti} (81 MHz) NMR spectra showing results from the reaction of Pd2Br2(u-S)(dpm)2 (2b, 1.7 x 10"2 M) with 2 mole equivalent HCl at R.T. in DMSO-d 6 ; spectra recorded after exposure of the system to laboratory light for 1 d; 9a = PdCl2(dpm), 9b = PdBr2(dpm), 11a = PdCl2(dpm(S)), l i b = PdBr2(dpm(S)). 211 5.7 ! H (300 MHz) and 31P{1H} (121 MHz) NMR spectra of the completed reaction between Pd2(SH)2(u-S)(dpm)2 (1.9 x 10"2 M) and 1 mole equivalent I2 at R.T. in DMSO-d 6 ; 2c = Pd2I2(u-S)(dpm)2. 213 5.8 3*P{Ti} NMR spectra (121 MHz) showing the partial formation of Pd2X2(u-S)(dpm)2 (2) from the reaction of PdX2(dpm) (9) with 0.5 mole equivalent NaSH at R.T. in DMSO-d 6 ; Y = probably Pd(SH)2(dpm). 214 5.9 Electronic absorption spectrum of I2 in chloroform from the photodecomposition of HI(g) (0.040 mmol) (amount of I2 not quantified). 217 xix 5.10 Electronic absorption spectrum of Nal 3 in DMSO-d 6 (amount of Nal 3 not quantified). 218 5.11 Electronic absorption spectra of an authentic sample of Br 2 in D M S O -d6 and Br 2 from photodecomposition of HBr (10 pL 8.83 M HBr(aq.) dissolved in 0.5 mL DMSO -d6; 12 h exposure to laboratory light) (amount of Br 2 not quantified). 219 6.1 Representative ' H NMR spectrum (300 MHz) of samples 1 or 2 containing y-alumina powder (15 mg) in CDC13 with and without the presence of H 2 S (110 pL at STP), respectively. 242 6.2 ! H NMR spectra (300 MHz) showing the extent of the reaction at R.T. of PdCl2(dpm) (9a) with H 2S in CDC13 in the presence of y-alumina under various experimental conditions; 2a = Pd2Cl2(p-S)(dpm)2. 243 6.3 ! H NMR spectra (300 MHz) showing the progress of the reaction at R.T. of PdCl2(dpm) (9a, 8.9 x 10"3 M) with 2.5 mole equivalent H 2 S in CDC13 in the presence of a-alumina (15 mg); 2a = Pd2Cl2(p-S)(dpm)2. 245 6.4 X-ray photoelectron spectra of y-alumina: a) experimental; b) literature, taken from ref. 4, pertaining to a surface cleaned by bombardment with Ar. 247 6.5 X-ray photoelectron spectrum (Mg Koc, R.T.) of sample 13 consisting of PdCl2(dpm) (9a) adsorbed on y-alumina. 248 6.6 X-ray photoelectron spectra (Mg Ka, R.T.) showing the Pd 3d photoelectron signals for samples 13, 15, and 16. 248 XX 6.7 X-ray photoelectron spectra (Mg Ka, R.T.) showing the CI 2p and S 2p photoelectron signals of samples 13, 14, 15, and 16. 249 6.8 X-ray photoelectron spectra (Mg Ka, R.T.) showing the S 2p and CI 2p photoelectron signals of samples 17 and 18, respectively. 249 6.9a Electronic spectrum (250 to 650 nm) of I3" (MeOH) isolated from synthetic-scale studies. 253 6.9b Electronic spectrum of an authentic sample if I3" in MeOH (prepared from I2 + Nal) (s292 = 2.89 x 104 M" 1 cm-1, e36o = 1.66 x 104 M" 1 cm4). 253 6.10 Electronic spectrum (250 to 700 nm) of I2 in MeOH (e292 = 2540 M 1 cm -1, e36o = 1380 M" 1 cm"1, £442 = 930 M" 1 cm"1). 254 6.11 ! H NMR spectrum (300 MHz) showing the results from the photodecomposition of HI(aq.) (~2 pL, 7.58 M) at R.T. in CDC13 in the presence of a-alumina (15 mg); X = CI and/or I. 254 7.1 ! H (200 MHz) and 3 !P {T4} (81 MHz) NMR spectra showing the progress of the reaction in CDC13 at R.T. of Pd2l2(n-S)(dpm)2 (2c, 1.6 x 10"2 M) with a 10-fold excess of Mel after 3 d; lc = Pd2I2(dpm)2, 9c = Pdl2(dpm), 11c = PdI2(dpm(S)). [The spectra of the various Pd species are discussed in Sections 3.2 (lc and 2c) and 4.2 (9c and 11c).] 278 7.2 ! H (200 MHz) and 3 !P {JH} (81 MHz) NMR spectra showing the progress of the reaction in CDC13 at R.T. of Pd2I2(u-S)(dpm)2 (2c, 1.6 x 10"2 M) with a 10-fold excess of EtI after 5 d; lc = Pd2I2(dpm)2, 9c = Pdl2(dpm), 11c = PdI2(dpm(S)). 279 xxi 7.3 31P{ 'Hj NMR spectrum (81 MHz) recorded for the reaction of Pd2Cl2(u-S)(dpm)2 (2a) with 700 psi CO at R.T. in CDC13 after 24 h; la = Pd2Cl2(dpm)2, 9a = PdCl2(dpm), 11a = PdCl2(dpm(S)), X = unknown. [The spectra of the various Pd species are discussed in Sections 3.2 (la) and 4.2 (9a and 11a).] 283 7.4 31P{ TT} NMR spectrum (81 MHz) recorded for the reaction of Pd2I2(u-S)(dpm)2 (2c) with 1 atm CO at R.T. in CDC13 after 11 d; lc = Pd2I2(dpm)2, 9c Pdl2(dpm), 1 lc = PdI2(dpm(S)), Y, Z = unknown. 283 1.1 Dependence of the free energy of decomposition of H 2S, AGa(H2S), on temperature in the range 298 to 1800 K. 292 IV. 1 Plot of ln(p(CHCl3)) versus 1/T. 299 VI. 1 Unit cell of the complex PdI2(dpm(S)>0.5CH2Cl2. 305 VII. 1 ORTEP drawing of [Pd(dpm(S))2]Cl2. Hydrogen atoms are omitted. 314 VJJ.2 Stereoview of [Pd(dpm(S))2]Cl2. Hydrogen atoms are omitted. 315 VII. 3 Stereoview of the unit cell of [Pd(dpm(S))2]Cl2. 315 VII.4 PLUTO drawing showing 1/2 of [Pd(dpm(S))2]Cl2 with interactions with a dichloromethane solvate molecule. 316 xxii LIST O F NON-STANDARD/LESS C O M M O N ABBREVIATIONS Ao, A , Aa, absorbancc at times zero, t, and infinite time b = path length (spectroscopy) B D H = British Drug Houses Chemicals Ltd. br = broad c r L = Cambridge Isotope Laboratories Cp = cyclopentadienyl 5 = chemical shift (NMR) d = doublet (NMR), or deuterium, or day dba = dibenzylideneacetone DMSO = dimethylsulfoxide dpe = l,2-bis(diphenylphosphino)ethane dpe(S) = 1,2-bis(diphenylphosphino)ethane monosulfide dpe(S)2 = 1,2-bis(diphenylphosphino)ethane disulfide dpm = bis(diphenylphosphino)methane dpm(S) = bis(diphenylphosphino)methane monosulfide dpm(S)2 = bis(diphenylphosphino)methane disulfide dpmMe = 1,1 -bis(diphenylphosphino)ethane dpmMe(S) = 1,1 -bis(diphenylphosphino)ethane monosulfide dpp - l,3-bis(diphenylphosphino)propane dpp(S) - 1,3-bis(diphenylphosphino)propane monosulfide dpp(S)2 l,3-bis(diphenylphosphino)propane disulfide xxiii dq = doublet of quartets dqn = doublet of quintets dt = doublet of triplets s = extinction coefficient, or dielectric constant FAB = fast atom bombardment (mass spectrometry) AGa(H2S) = free energy of decomposition of H 2S GC = gas chromatography HDS = hydrodesulfurization k = rate constant kB = Boltzmann's constant (1.38066 x 10-23 J K"1) kobs = observed pseudo-first order rate constant p. = bridging, or micro m = multiplet (NMR) M = metal, or unit of molarity M C B = Matheson, Coleman, and Bell MSD = Merck, Sharp, and Dohme m/z = mass to charge ratio n = stoichiometric equivalent p = pressure, or pseudo (NMR) Ph = phenyl q = quartet qn = quintet R = alkyl or aryl group, or reagent, or gas constant (8.31442 J K _ 1 mol-1) xxiv s = singlet (NMR), or second sh = shoulder t = triplet (NMR), or time T = temperature T L C = thin-layered chromatography TMS = tetramethylsilane V = volume XXV N U M E R I C A L K E Y T O P A L L A D I U M C O M P L E X E S 1 Pd2X2(dpm)2 10c Pd2I4(dpm)2 la Pd2Cl2(dpm)2 11 PdX2(dpm(S)) lb Pd2Br2(dpm)2 11a PdCl2(dpm(S)) lc Pd2I2(dpm)2 l ib PdBr2(dpm(S)) 2 Pd2X2(u-S)(dpm)2 11c PdI2(dpm(S)) 2a Pd2Cl2(u-S)(dpm)2 2b Pd2Br2(u-S)(dpm)2 2c Pd2I2(u-S)(dprn)2 3b Pd2Br2(dpm)(dpm-d2) 4b Pd2Br2(dpm-d2)2 5b Pd2Br2(dpm)(dpmMe) 6b Pd2Br2(dpmMe)2 7b Pd2Br2(u-S)(dpm)(dpm-d2) 8b Pd2Br2(u-S)(dpm-d2)2 9 PdX2(dpm) 9a PdCl2(dpm) 9b PdBr2(dpm) 9c Pdl2(dpm) 10 Pd2X4(dpm)2 10a Pd2Cl4(dpm)2 10b Pd2Br4(dpm)2 xxvi With a great sigh of relief, I finally set my pen down knowing full well that arriving at the end means that I really just begun. A C K N O W L E D G M E N T S I would like to thank Professor Brian James for his patience and guidance during my PhD work. I also would like to thank past and present members of the James' group, particularly Mr. Jeffrey Posakony and Mr. Stephen Cheng for their many valuable discussions. The services of the department are greatly appreciated. A big thanks goes to Mr. Kevin Lindstrom for his help in searching the Chemical Abstracts. I am indebted to Dr. Yoshikata Koga for first providing me an opportunity to explore the realm of chemistry ~ to Mrs. Koga, thanks for the great dinners you've prepared for the Koga group over the years. I thank my family for their love and support; their emphasis on the importance of education has taught me much. Last but not least, I thank my lovely wife Christina for her unfailing love, and it is to her that I dedicated this thesis. T Y H W xxvii CHAPTER 1 Introduction 1 Chapter 1 Chapter 1 Introduction Some research in this laboratory focuses on the development of homogeneous catalytic systems for the recovery of H 2 from H 2 S . In particular, Pd complexes are employed, following the initial discovery (reaction ly* that quantitative sulfur abstraction from H 2 S with concomitant H 2 evolution readily occurs in solution under ambient conditions.1 Detailed aspects of this reaction and its implications are presented in this Introduction, as well as an account of other methods developed for the decomposition of H 2S, following a brief description of the properties, occurrences, and uses of this compound. (1) Pd2X2(p-dpm)2 + H 2S -> Pd2X2(u-S)(u-dpm)2 + H 2 X = Cl, Br, I 1.1 Properties, occurrences, and uses of H 2S 2 ' 3 ' 4 The discovery of H 2S is credited to the Swedish chemist, Carl Wilhelm Scheele, who published the first systematic study of the gas in 1777.5 Although he named it Schwefelluji meaning sulfur air, H 2S was more prosaically and appropriately referred to as stinkende; the readily identifiable odor resembling rotten eggs cannot be mistaken by the average person. Hydrogen sulfide is a colourless gas with melting and boiling points at, respectively, -82.9 and -61.8 °C. It is denser than air, and at -20 °C, the specific gravity is 1.19 times that of air. Solubility occurs in a variety of polar and non-polar solvents including alcohols, ethers, glycerol, hydrocarbons, water, and amine solutions. - In aqueous solution, two acid dissociation constants exist at pK a ~7 and 12. H 2S is the only thermodynamically stable binary sulfur-hydrogen The (j.-symbol for the bridging diphosphine ligand(s) will be omitted for convenience throughout the text. 2 References on page 37 Chapter 1 compound occurring in nature; the standard enthalpy and free energy of formations are -20.64 and -33.02 kJ mol"1, respectively. Above 500 °C, the equilibrium with its elemental constituents becomes apparent because of kinetic and thermodynamic control, and at temperatures > 1500 °C, the decomposition of H2S even becomes thermodynamically favourable (see Section 1.21 and Appendix I). The toxicity of H2S is comparable to that of HCN in terms of rapidity of action and concentration from which death will result. At low concentrations between 0.014 and 140 mg m"3 (0.01 and 100 ppm), the characteristic odor is instantly recognized; however, at H 2S concentrations above 210 mg m - 3, paralysis of the olfactory apparatus occurs, accompanied by rapid loss of consciousness and subsequent respiratory failure. Often, individuals who are revived following immediate medical attention do not recall ever detecting the noxious odor. Interestingly, H 2S is normally found in the human breath and is probably the cause of halitosis. Hydrogen sulfide occurs naturally in coal, natural gas, oil, volcanic gases, sulfur springs and lakes, soil, and to lesser extents swamps and marshes. It results from either the action of steam on mineral sulfides and sulfates at elevated temperatures or the bacteriologic decomposition of mineral sulfates or S-containing organic matter. With the exception of the fossil fuel reserves, these biogenic sources contribute to the natural global sulfur cycle in which an atmosphere concentration of 3 to 30 pg nf3 (2.1 to 21 ppb) H 2S is maintained. In the troposphere, H 2S is oxidized by the ozone present to sulfur oxides which are then taken up by plants and metabolized. The cycle is completed by herbivorous consumption by animals and subsequent bacterial action on these animals (or plants) at death. Apart from natural occurrences, anthropogenic sources of H 2 S are becoming increasingly important, and there is cause for concern as consequential hazardous effects on the environment are becoming more pronounced (acid rain, smog, plant deterioration, 3 References on page 37 Chapter 1 and adverse effects on aquatic life). The two main industrial sources are the Kraft wood pulping process6 and the hydrodesulfurization7 in the refinement of petroleum. In the Kraft process, NaSH is used in the cooking liquor to enhance the strength of the paper produced; H2S is formed when the spent cooking liquor is evaporated and burned to recover the chemicals. In hydrodesulfurization, petroleum is treated with H2 over Mo-Co or W-Ni sulfide catalysts to remove the sulfur components normally present including mercaptans, sulfides, disulfides, and sulfur heterocyclics (eqs. 1.1.1 -1.1.4). This process must be carried out in order to prevent catalyst poisoning during the subsequent cracking and reforming stages; furthermore, sulfur emissions into the air will be reduced when the processed fuels are used. The H 2S, which has to be recovered, is then either recycled, for example, to regenerate NaSH for the Kraft process, or oxidized to elemental sulfur (see below). (1.1.1) RSH + H 2 - » RH + H 2S (1.1.2) RSR + 2 H 2 -> 2 RH + H 2S (1.1.3) RSSR + 3 H 2 -> 2 RH + 2 H 2S (1.1.4) C4H4S (thiophene) + 2 H 2 -> CH2=CH-CH=CH2 + H 2S Despite the negative qualities, H 2S surprisingly has many industrial uses2'3'4 including: (1) the preparation of inorganic and organic compounds; (2) the preparation of extreme pressure lubricants and cutting oils; (3) the purification of hydrochloric and sulfuric acids; (4) the hair removal process in the leather industry; and (5) the production of heavy water.8 Most of the recovered H 2S is used in the manufacture of sulfuric acid, more of which is produced than any other chemical.9 For convenience of shipping and storage, H 2S is usually oxidized to elemental sulfur, and there exists several technologies, the most important being the Claus process (eqs. 4 References on page 37 Chapter 1 1.1.5, 1.1.6).10 Essentially, a portion of the H2S is burned to SO2, which is then combined with the remaining H2S in the presence of a catalyst to produce elemental sulfur. (1.1.5) H 2S + 1.5 0 2 -> S0 2 + H 2 0 (1.1.6) S0 2 + 2 H 2S 2 H 2 0 + 3 'S' 1 For convenience, 'S ' is written for elemental sulfur instead of the correct 1/8 S 8 formulation. 5 References on page 37 Chapter 1 1.2 Recovery of both H 2 and elemental sulfur from H 2 S Decreasing fossil fuel reserves has prompted an intensive search for alternative methods of obtaining energy. The focus is on hydrogen as there is much interest in a hydrogen economy,11 and the lack of an adequate H 2 supply has made the search for possible sources a main research goal. Hydrogen is not only important as an energy source but also as a raw material in many chemical processing industries among which the refining of petroleum is a major consumer. Because H 2 is not found in nature, the current source comes from the steam reforming of C H 4 . 1 2 In view of the great demand for H 2 by both the chemical and energy sectors, researchers are looking to the abundantly available H 2S as a possible source. Because of the inherent toxicity of H 2S, high priority is placed on its removal by the industries that produce it. In the Claus process, the sulfur is recovered but the H 2 value is lost. Much research is devoted to finding and developing methods for decomposing H 2S into its elements, and a process which recovers both H 2 and sulfur could be of considerable worth. The following subsections outline such methods which can be classified under four main categories: (1) thermodecomposition, (2) photodecomposition, (3) electrochemical decomposition, and (4) plasmachemical decomposition. A fifth category, radiolytic decomposition of H 2S, is included for the sake of completeness but the emphasis in this area is not on recovery purposes. Decompositions of H 2S in homogeneous solution systems with formation of H 2 are described later (see Sections 1.3 and 1.4); no literature can be found concerning recovery of both H 2 and elemental S in solution systems. 6 References on page 37 Chapter 1 1.2.1 Thermodecomposition of H 2S Hydrogen sulfide can be thermally decomposed at high temperatures into H 2 and S (elemental sulfur). Historically, the first such study was of a fundamental nature reported Taylor and Pickett13 in 1927 where high temperatures were achieved by electrically heating a Pt filament. Similar fundamental studies14'15'16 have followed, providing a basic understanding of the kinetics and mechanism of the reaction, with Kaloidas and Papayannakos17 recently proposing a free-radical 18 mechanism. The first industrial approach to the study is documented in 1958 when Massey patented a process for recovering both H 2 and 'S' by decomposing H 2S over glowing W filaments. Few groups have since studied such direct routes19"23 as the high temperatures required (-1000 °C) make the process uneconomical; however, the use of concentrated solar energy alone or in conjugation with a catalyst (see below) has been reported.24-27 In the interest of economics, methods employing lower temperatures are sought after and for this reason catalysis has received increasing attention since Weiner and Leggett28 in 1958 first proposed a two-step process involving Fe, Co, and Ni sulfide catalysts. Development in this thermocatalytic decomposition area, in general, involved transition metal oxides and sulfides. (The introduction of a catalyst, of course, serves only to accelerate the rate at which the equilibrium, H 2S H 2 + 'S', is established; the equilibrium concentrations of each species are unaffected. However, if either or both products are continually removed, higher conversions of H 2S are attained as the reaction is forced to the right.) Optimal temperatures ranged from 400 to 800 °C and two types of catalytic processes are seen. The first type is described as being truly catalytic, the composition of the catalyst being unchanged after the reaction as evidenced by X-ray diffraction measurements. Examples include MoS2, WS 2, and Cr 2S3 and studies have shown that their 7 References on page 37 Chapter 1 catalytic activities depend inversely on temperature over this 400 - 800 °C range.29"32 Mechanistic studies with M0S2 have suggested that H2S decomposes via oxidative addition at an active site on the surface, followed by elimination of H 2 and subsequently 'S ' . 3 3 ' 3 4 The second type of catalytic process involves two steps:28 M x o r M x S y + zH 2 S - » M x S z o r M x S y + z + z H 2 M XS Z or M x S y + z -> M x or M x S y + z S In the first step, H 2S reacts with a metal, M x , or metal sulfide, M x S y , to generate H 2 and a metal sulfide, MXSZ, or a higher metal sulfide, M x S y + z . The M X S Z or M x S y + z is then thermally decomposed in the second step liberating 'S' and regenerating the metal or lower metal sulfide. The mono- and disulfides of Fe, Co, and Ni are well studied1 2 , 3 1'3 2'3 5'3 6 but are not effective catalysts due to the formation of non-stbichiometric forms as a result of varying degrees of sulfidation and/or thermal decomposition.12'31 Copper sulfides, C U 2 S , CU9S5, and CuS, are also ineffective for the same reasons.31 However, the polysulfides, Co 9 S 8 3 6 and Ni 3 S 2 , 3 6 are effective but require longer de-sulfurization times at higher temperatures particularly for Ni 3S2. 3 7 The vanadium sulfide, V 2 S 3 , was found to be the most effective polysulfide catalyst, and VS 3 8 and V S 4 3 9 were proposed as intermediates. Silver is an effective M x , with the Ag2S produced easily de-sulfurized.36 Mixed metal sulfides, Ni-Mo sulfide,25b Co-Mo sulfide,25b Fe 7S g/MoS2, 4 0 FevSg/NiS^o,40 V 2 S 3 /Cu 9 S 5 , 3 8 V 2 S 3 /Fe 7 S g , 3 8 and V 2 S 3 /ZnS, 3 8 were examined for cooperative effects and had activity comparable to that of M0S2. The metal oxides, V2O5-M0O3 4 1 and Mn nodules,42 do catalyze the thermodecomposition of H 2S but the active catalysts are the sulfides formed in the initial sulfidation step. In the case of Mn nodules, which were studied more for their sulfur absorbent properties, the activity was attributed to the high Fe content. Chivers and Lau studied the alkali metal sulfides M 2 S (M = Li, Na, and K) and polysulfides M 2 S X (x = 2-4; M = Na and K), and discovered that 8 References on page 37 Chapter 1 Li 2S was the only active species and, within the above two-step scheme, Li 2 S 2 was formed.43 The aluminas, a-Al203 and V-AI2O3, were often used as supports with increased catalytic activity resulting in some cases.30 They themselves were tested in the range 500 - 1000 °C and, whereas a-alumina did not catalyze the decomposition,173 y-alumina was effective as a physically active medium for H 2S adsorption and subsequent decomposition.44 Activation energies for the catalyzed H 2S thermodecomposition range from 14 to 27 kcal mol"1 compared to values of 42 to 50 kcal mol"1 for the uncatalyzed reaction. 1 5 ' 1 7 a ' 2 5 b ' 3 0 ' 3 9 By continuous or intermittent removal of either or both H 2 and 'S', conversions of more than 95% of H 2S have been achieved.30 A few different approaches to the thermodecomposition of H 2S have also been explored. Raymont19 proposed an open loop system where the thermodynamically unfavourable decomposition of H2S is coupled with the favourable oxidation of CO via a two-step process: 2 H 2S + 2 CO -> 2 H 2 + 2 COS 2 COS + S0 2 -> 2 C 0 2 + 3/2 S 2 where the SO2 required is produced in a manner similar to the Claus process. The overall reaction produces H 2 , 'S', and CO2. Alternatively, the COS formed in the first step could be thermally decomposed at high temperatures (i.e. 830 °C) giving 'S' and CO, and by rapid cooling the 'S' could be separated from CO which is then recycled back to the first step.45'46'47 Towler et al. 4 8 and Bowman46 extended similar chemistry using CO2: H 2S + C 0 2 -> 1/2 S2 + CO + H 2 0 The sulfur is cooled and removed, and the remaining gases undergo the water gas shift reaction: CO + H 2 0 -> H 2 + C 0 2 COS and CS 2 are side-products that invariably form, for example via the following reactions: CO + S ^ COS 9 References on page 37 Chapter 1 2 COS ^ C 0 2 + CS 2 but after separation of sulfur, the residual COS and CS 2 can be reduced at high temperatures via the reaction: COS + CS 2 + C 0 2 - » 3 CO + 3/2 S 2 with the resulting CO subjected to the water gas shift reaction.46 In the preceeding examples of both open and closed loop systems, conversion of H 2S to H 2 and elemental sulfur is greater than when H 2S is heated alone, but the rate-limiting step is still the thermal dissociation of H 2S for which transition metal sulfide catalysts are used particularly MoS2. Raymont19 also studied another open loop system based on the oxidation of hydrocarbons,49'50 as shown by the following scheme: xH 2 S + H y C z -> z C S 2 + (x+z)/2H2 z CS 2 + z S0 2 -> z C 0 2 + 3z/2 S 2 z/2 S2 + z 0 2 -> z S0 2 The overall reactions produce H 2 from H 2S and hydrocarbons, using the oxidation of the carbon present in hydrocarbons as the driving force: xH 2 S - > x H 2 + x/2 S 2 H y C z + z 0 2 -> y/2 H 2 + z C 0 2 In a different approach, Chen51 investigated the reaction of H 2S with I2 in aqueous solution, and found that HI was formed while sulfur precipitated. The elemental'S' was removed and the aqueous HI solution was treated to obtain anhydrous HI which was decomposed at high temperatures and pressures over a high surface area catalyst. H 2S + I2 (aq.) -> 'S' + HI (aq.) ffl(aq) -> ffl(g) 2HI (g) H 2 + I2 10 References on page 37 Chapter 1 Plummer52 studied the interaction of H 2S with quinones, and described a three-step process involving two catalysts with emphasis on the importance of the reaction solvent. He identified N-methyl-2-pyrrolidone (NMP) as an effective polar medium: NMP + H 2S -> N M P H + S H H 2S reacts with NMP to form the quaternary ion complex which then reacts with the quinone, t-butyl anthraquinone, to yield sulfur and the corresponding hydroquinone: NMPH+HS" + ' S ' + N M P O H After removal of the sulfur, the hydroquinone is catafytically dehydrogenated over CrC>3 at high temperatures, and H 2 is recovered. O H O H + H 7 Despite unfavourable thermodynamics, the thermodecomposition of H 2S to H 2 and S can be achieved using open or closed loop systems. Combining the use of catalysts and the intermittent or continuous removal of products, high conversions of nearly 100% are possible. 0 11 References on page 37 Chapter 1 1.2.2 Photodecomposition of H 2S The photodissociation of H 2S has been extensively studied at a fundamental level, and references can be traced through Okabe's review.53 By means of flash photolysis, the primary process following absorption of light in the ultraviolet has been shown to be the production of H atoms and S atoms or SH radicals: H 2S -> H + SH ^<320nm H 2S -> H + H + S X < 200 nm At X > 200 nm, the fates of H and SH radicals are as illustrated in the secondary reactions: H + H 2S -> H 2 + SH 2 SH S 2 + H 2 , H 2S + S, HS 2 + H, orH 2 S 2 The main final products though are H 2 and elemental sulfur. For industrial application, the use of ultraviolet radiation is not practical because of the energy expenditure; however, work with laser radiation of noble gas halide compounds (low energy consumption) has been reported.54 Currently, researchers are seeking to exploit solar energy either as solely a heat source (see Section 1.2.1) or as a cheap and plentiful source of photons. The energy available from visible light, however, is not sufficient to affect dissociation of the H-SH bond, but when certain semiconductor particulates are present, redox processes take place on the surfaces resulting in solar-driven efficient production of H 2 and elemental sulfur. Research in visible light-induced cleavage of H 2S using semiconductor materials stemmed from investigations of microheterogeneous and colloidal systems that decompose water into H 2 and 0 2 . 5 5 ' 5 6 Semiconductor materials, such as CdS and Ti0 2 , possess several attractive properties as light harvesting units: a) high extinction coefficients; b) fast carrier diffusion (migration) to solid-12 References on page 37 Chapter 1 surface interface; and c) suitable positioning of conduction and valence bands.57 The band gaps of these materials, in particular, correspond to energy differences coinciding with those of visible light. For instance, CdS has a band gap of 2.4 eV corresponding to an absorption edge of -520 nm. As such, these materials have been shown to decompose water upon illumination via generation of charge carriers, i.e. conduction band electrons, ecB - , and valence band holes, hvB + -Co-deposits of noble metal catalysts, Pt and Ru02, prevent recombination of ecB~hvB + by facilitating separation through trapping and promotion of these charge carriers in interfacial transfer to acceptor molecules: H 2 0 0.5 0 2 + 2H+ Thus, water is reduced or oxidized by ecB or h v B + , respectively, to yield H2 and O2, and the net reaction is accomplished by two quanta of light: CdS or Ti0 2 + 2 hv -> CdS or Ti0 2 (2 eC B\ 2 h v B + ) 2 H 2 0 + 2 eCB" -> H 2 + 2 OH" H 2 0 + 2 h v B + -> 0.5 0 2 + 2 H + Net: H 2 0 + 2hv -> H 2 + 0.5 0 2 In the case of CdS, R.UO2 also serves to prevent photocorrosion, for example, via the following reaction, by effectively scavenging the valence band holes as they form: CdS + 2hvB+ -> C d 2 + + S 13 References on page 37 Chapter 1 The analogous chemistry has been applied to H 2S using CdS colloidal dispersions.57"71 Whereas both Pt and Ru0 2 are required to photodecompose water, these noble metals need not be present as photocleavage of H 2S is possible on bare CdS. The rate of hydrogen production (r(H2)), however, significantly increases when Ru0 2 is present, the enhancement being attributed to effective h v B + separation and transfer to the solution. At the solid-solution interface, the holes oxidize H 2S, SH", or S2" depending on the pH: H 2S + 2 h V B + -> 2 H + + S; S H + 2 h V B + -> H + + S ; S2" + 2 h v B + S Higher r(H2) were seen at higher pH, with a sigmoidal dependence of r(H2) on pH, the effective S2" concentration being governed by the pK a . 5 8 The oxidation process is envisaged as adsorbed SH" or S2" species scavenging the holes trapped by the noble metal deposit. The presence of Pt proved superfluous as no improvement in H 2 yields was seen. Apparently, adsorbed S H or S2" induces a cathodic shift in the flat band potential and thereby greatly increasing the driving force for the 58 70 71 reduction of H 2 0 to H 2 making the intervention of Pt unnecessary; however, recent reports ' describe platinized CdS to be quite effective, with H 2 yields even higher than those achieved with Ru0 2 deposited-CdS. These discrepancies were not addressed although earlier reports57 emphasized that the method of preparing semiconductor colloids is crucial and that usage of different salts (i.e. Cd(N03)2, Cd(S04), and CdCl2) results in remarkable differences in r(H2), for example, in the order NO3" > SO4 2" > CI". The rationale given ascribed differences to modification of active sites for H 2 0 reduction and/or increase in the extent of charge separation. Attempts to improve photocatalytic activity were studied by employing different semiconductors or different metal deposits, or by adding chemical sinks. One problem inherent in 14 References on page 37 Chapter 1 photodecomposing H2S is the formation of polysulfides, Sn2", resulting from the reaction of S, once formed from oxidation of SH" or S2", with S2" or other S„2" species: S + S2" S22" ; S + Sn 2 —> SN+1 2 Once the solution is saturated with S„2", elemental sulfur begins to precipitate out. Neither S nor polysulfide interferes with the actual redox processes but the polysulfides absorb in the blue-green region of the spectrum (-500 nm) and catalytic activity is therefore progressively reduced. If, however, the solution is kept shallow and continuously flowing over an inclined bed of immobilized semiconductor particulates, then reduction in activity is eliminated.70'71 A further advantage in this approach is the resultant separation of products (H 2 gas percolates countercurrently); the issue of product separation from solutions containing colloidal dispersions remains to be resolved.71 Addition of SO3 2 " was explored as a chemical sink as it reacts with S to form thiosulfate which is colourless:59'60 S + SO3 2" -> S203 2" A significant increase in r(H2) was observed; however, as the concentration ratio of SO3 2 " to S2" is increased, a retarding effect is observed because of interference from S2O32" competing with SH" or S2" for active sites. 6 0 Ti0 2 particles were then found to be capable of efficiently photoreducing thiosulfate back to sulfide and sulfite,63 and illumination by visible light produced ecB*hvB+ pairs that respectively reduce and oxidize S2O32" according to the following reactions: 2 eCB"(Ti02) + S2O32" -> S2- + S032" 2hvB+(Ti02) + 0.5 S203 2" + 1.5 H 2 0 -> S032" + 3 H + When coupled with CdS/Ru02, the system could potentially give higher H 2 yields, but S2" would be catalytically recycled. In actuality, the system performs poorly, giving lower r(H2) than with 15 References on page 37 Chapter 1 CdS/Ru02 alone.64 Astonishingly high r(H2) were observed, on the other hand, when Ru0 2 was deposited on Ti0 2 instead of CdS; in fact, the system is superior to that of CdS/Ru02. The results were attributed to effective charge carrier separation; an inter-particle ecB~ transfer to T i 0 2 vectorially displaces the charges and thereby preventing electron-hole recombination. Thereafter, the e" migrates and is trapped by Ru0 2 which subsequently promotes an interfacial transfer to H 2 0 . Holes remaining on CdS do not transfer to Ti0 2 , perhaps being inhibited for energetic reasons, but migrate to the surface to oxidize sulfide. Replacement of Ru0 2 by Pt on T i 0 2 results in suppression of H 2 production, and no catalytic activity is seen when Ti0 2 /Ru0 2 is used alone.64 A combination of CdS/Ru02 + Ti0 2/Ru0 2 was studied for which 100% photodecomposition of H 2 S under direct sunlight irradiation was reported, the method of colloid preparation being emphasized.66 Metal deposits and semiconductor dispersions have been investigated. For example, vanadium sulfide or oxide dispersions loaded with Ru0 2 give results comparable to those of the CdS/Ru02 system.67 Furthermore, a mixture of V2Os and V 2 S 3 semiconductor dispersions loaded with Pt or Ru0 2 give higher H 2 yields due to additional vectoral displacement of charges via inter-particle e" transfer from V 2 0 5 to V 2 S 3 (cf. the CdS + Ti0 2/Ru0 2 system). Pt on V 2 S 3 facilitates interfacial e" transfer more efficiently than Ru0 2 on Ti0 2 , a phenomenon yet to be explained.69 Substitution of Ru0 2 by RuS2 on CdS dispersions improves photocleavage of H 2S several-fold.65 Finally, thin films of CdSe have been studied in the photoelectrolysis of H 2S where CdSe is utilized as a photoanode connected to a Pt cathode by Cu wire.68 The chemistry is the same; sulfide is oxidized to elemental S, and H 2 evolves at the cathode. An advantage to this system is that electricity can also be generated. 16 References on page 37 Chapter 1 Of note, an indirect method of photodecomposing H2S using oxidants such as I2 has been patented.72 Iodine reacts with H 2S in water to form aq. HI and elemental sulfur. After the sulfur is removed, the aq. HI is fed into a photoelectrolyzer, and H 2 and I2 are produced at the cathode and photoanode, respectively; the I2 is subsequently recycled. This method parallels those reported earlier where aq. HI is generated, but is either treated to give anhydrous HI which is then thermally decomposed (see Section 1.2.1) or directly electrolyzed (see Section 1.2.3). Alternatively, the HI formed in solution could be directly photodecomposed without prior treatment of any kind (see Chapter 5). 17 References on page 37 Chapter 1 1.2.3 Electrochemical decomposition of H 2S The electrochemical decomposition of H 2S has been studied as a means of recovering H 2 and elemental sulfur. The methods reported in the literature can be classified into two groups, direct and indirect. Direct conversion is simply accomplished by passage of current through an electrolyte solution containing H 2S. Indirect conversion involves reaction with an oxidant to produce elemental sulfur which is then separated prior to electrochemical treatment of the spent oxidant for regeneration and simultaneous production of H 2 . The required electronics are usually determined both experimentally and by application of Nernstian fundamentals, and vary with factors such as temperature, concentration, choice of electrode and electrolyte materials, and usage of ion-selective membranes separating the anodic and cathodic compartments. In general, higher potentials are needed for indirect conversion but anode passivation is avoided as sulfur is not electrochemically produced as in the direct processes; however, the sulfur is usually in a sticky, plastic form requiring further treatment and purification to obtain the crystalline form that is produced in direct conversion methods. Along with recovery of valuable commodities, direct electrochemical conversion of H 2S has also been explored in the areas of metal refining73 and coal and natural gas processing.74 In these processes, electrolysis of H 2S is carried out in basic solvents or solutions or in molten alkali sulfide electrolytes. One problem inherent in the direct method is the rapid passivation of the anode once oxidation takes place to form elemental sulfur.75'76 The sulfur coats the electrode and the electrochemical process terminates as the current decreases to zero. To circumvent the problem, basic solvents or solutions are used, most commonly aqueous NaOH. 7 7 ' 8 3 The respective anodic and cathodic reactions are as follows: 18 References on page 37 Chapter 1 (1.2.3.1) HS" + OH" —> S + H 2 0 + 2 e" (1.2.3.2) 2 H 2 0 + 2e" -> H 2 + 2 OH" As previously noted (Section 1.2.2), elemental sulfur in basic media is unstable and reacts with S2" to form soluble polysulfides Sn2". By forced mechanical convection, anode passivation is minimized and elemental sulfur precipitates from the bulk solution once saturation with polysulfides is reached. Immiscible organic solvents such as toluene have also been employed to extract the sulfur as it forms.80 In a different approach, the electrochemical decomposition of H 2S in molten alkali sulfides has been studied as a means to process sour hydrocarbon feeds.74'84 High temperatures (600 - 1000 °C) are used to mimick thermal conditions normally encountered in industry. Binary mixtures of M 2 S (M = Li, Na, K) are identified as appropriate electrolyte materials in view of the sulfidic nature of H 2S. As a gaseous stream containing H 2 S enters the cathodic compartment, reduction occurs and the resultant H 2 is subsequently carried off with the now-sweetened stream: (1.2.3.3) H 2S + 2e" -> H 2 + S2" The sulfide produced then migrates through the electrolyte suspended in an inert matrix to the anode, and gaseous S 2 forms via oxidation: (1.2.3.4) S2" -> 0.5 S 2 + 2e" An inert gas such as N 2 is used in the anodic compartment to sweep away the gaseous sulfur for condensation elsewhere. Anode passivation is not an issue here as high temperatures are present. Anode passivation is not seen in indirect electrochemical conversion of H 2S as the sulfur is formed via chemical reaction with an oxidant and is separated prior to electrochemical treatment of the spent oxidant. Reactions 1.2.3.5 and 1.2.3.6 represent the process in general where ox and red are oxidized and reduced species, respectively: 19 References on page 37 Chapter 1 (1.2.3.5) ox + H 2S -> red + S chemical reaction (1.2.3.6) red —> ox + H 2 electrochemical reaction The various oxidants studied can be classified into three groups: (1) Fe(in) compounds; (2) halogens; and (3) quinones. Of the iron(IU) compounds investigated, Fem(CN)63" complexes were the first to be studied in indirect electrochemical decomposition of H 2 S. 8 5 The chemical reaction with H 2S is performed in basic aqueous systems (reaction 1.2.3.7). Thereafter, the spent oxidant is regenerated according to reactions 1.2.3.8 and 1.2.3.9 with production of H 2 . (1.2.3.7) 2Fem(CN)6 3" + H 2S + 2 OFF - » 2 Fen(CN)64" + S + 2 H 2 0 (1.2.3.8) 2Fen(CN)6 4" -> 2Feffl(CN)63" + 2 e" (1.2.3.9) 2 H 2 0 + 2e" -> H 2 + 2 OH" Similarly, Fe(JJJ.)Cl3 salts have been studied, and in aqueous acidic systems, the reactions are as follows: (1.2.3.10) 2Fe 3 + + H 2S -> 2 Fe 2 + + S + 2 H + (1.2.3.11) 2 Fe 2 + -> 2 Fe 3 + + 2 e" (1.2.3.12) 2 H + + 2e" -> H 2 Numerous patents exists for this process,86 and successful bench-scale development has been reported.87 With halogens, Br2 and I2 are typically used, and employed in aqueous acidic conditions such that solubilization occurs forming the corresponding trihalides, Br3" and I 3 " . 8 8 - 9 1 Reaction with H 2S proceeds according to reaction 1.2.3.13 where X = halide: (1.2.3.13) X 2(asX 3") + H 2S - • S + 2 H + + 2 X' After the precipitated sulfur is removed, the remaining acidic solution is electrolyzed, regenerating X 2 with accompanying H 2 production: (1.2.3.14) 2X" -> X 2 + 2e" 20 References on page 37 Chapter 1 (1.2.3.15) 2 H + + 2e" -> H 2 Finally, indirect electrochemical decomposition of H 2S has also been examined using quinones (see also Section 1.2.1), and the process is briefly depicted as follows.92 O O H + S 21 References on page 37 Chapter 1 1.2.4 Plasmachemical decomposition of H 2S Currently, there is rapid development of a novel technology first introduced in the Soviet Union during the mid-1980s.93'94 Plasmachemical decomposition of H2S into H2 and elemental sulfur using microwave or radio frequency radiation has been achieved.95'96 The process is economically attractive due to low energy expenditures, and the small scale industry level has been reached,96 competing successfully with traditional H 2S removal processes such as the Claus method (see Section 1.1). The principle involves the generation of a conducting plasma (cations and electrons) which in the presence of a fluctuating magnetic field, produced by the applied radiation, flows in a closed annular path: Through cationic or electronic impact, atomization of the entering gas stream occurs and subsequent product formation follows as the stream exits. High temperatures are often present (ohmic heating), and thermodecomposition can also take place. The formed sulfur is removed via condensation, and the remaining H 2 gas is recycled to decompose unconverted H 2 S; overall conversion efficiencies in excess of 99% are typically seen. Plasma decomposition of H2S by passage through an electrical discharge has also been explored.97"100 The methodology is not new, 1 0 1' 1 0 2' 1 0 3 and recent resurgence is the result of the radiation induction coil gas flow 22 References on page 37 Chapter 1 ongoing search for economic as well as ecologically sound ways to process H2S. As in microwave or radio frequency plasmachemical decomposition, the mechanistic aspects are not well understood, but mechanisms resembling those in photodecomposition have been proposed for both types of technologies (see Section 1.2.2).94'98'104 1.2.5 Radiolytic decomposition of H 2 S Radiolytic decomposition of H 2S has been investigated105'106 but the potential use for recovering H 2 and elemental sulfur is unrealistic because of the radioactive nature of the a-, p-, and y-particles employed. Particle energies are of the order 106 eV (cf. HS-H and S-H bond strengths ' are -360 kJ mol"1 or3.7eV per molecule), and a variety of ions, excited molecules, atoms, and free radicals are produced during the primary and secondary processes. The chemistry resembles the radical mechanisms encountered in the photodecomposition of gaseous H 2 S. H 2 and elemental sulfur are the main products along with small amounts of sulfanes, H2S„. . 23 References on page 37 Chapter 1 1.3 Homogeneous catalytic conversion of H 2S to H 2 Research in this laboratory on the separation of components in coal gasification streams using coordination compounds109"112 has stimulated an interest in transition metal - H 2S chemistry. The well-known Pd2X2(dpm)2 complexes113 (X = halide, NCO) show excellent selectivity for CO in the presence of C0 2 , N 2 , 0 2, H 2 , C2H4, and C 2 H 2 , with rapid reversible binding observed: P h 2 P - ~ ^ P P h 2 P h 2 P - - ^ P P h 2 I I I . i ' . . I (1.3.1) X—Pd Pd—X + CO ^ v ^ - P d ' P d ^ , v I I X I I X Ph2P-^ PPh2 Ph2P PPh 2 Tests were made for H 2S, an impurity often present in such streams, and resulted in the first homogeneous solution reaction showing a quantitative 1:1 H 2S uptake : H 2 evolution stoichiometry at a metal center:1 PI12P T P I 1 2 Ph2P" "PPI12 I I I ..S.„ I (1.3.2) X - P d P d - X + H2S *- + FT PI12P PPh 2 P h 2 P ~-_^ -PPh 2 Deuterium labelling studies114 demonstrated that the H 2 evolved came exclusively from H 2S. Related studies showing H 2 generation from homogeneous reactions with H 2S can be found in the literature (see Section 1.4) but involve 'simple' S-abstraction by metal centres. Detailed kinetic and mechanistic investigations114'115 revealed that reaction 1.3.2 proceeds via oxidative addition that generates a hydrido(mercapto) intermediate (I); H 2 production is then envisaged as deprotonation of SH followed by protonation of the hydride ligand, with concomitant formation of the p-S species, a so-called 'A-frame' structure.116 24 References on page 37 Chapter 1 P h 2 P - ~~PPh2 I , H | „„SH . P d " . P d ' | X ^ | P h 2 P - ^ ^ - P P h 2 I Prior adduct formation with H 2S en route to the intermediate, although not detected, was not ruled out;114 indeed, in this respect, a few transition metal - H 2S complexes have been characterized117" 1 O O 1 1 -jn including H NMR spectroscopic evidence for the species Pt(PPH 3) 2(SH 2r u en route to formation of a more stable hydrido(mercapto) complex, as well as recent X-ray crystallographic characterization of Ru(SH2)(PPh3)('S4') ('S4' = tetradentate 2,2'-(ethylenedithio)bis(thiophenolate)dianion)121 and Ru(SH2)Cl2(P-N)(PR3) (P-N = o-diphenylphosphino-N,N-dimethylaniline, R = /?-tolyl).122 The rate-deterrnining step for reaction 1.3.2 was shown not to be the subsequent conversion of the intermediate I, and the observed reactivity trend X = CI > Br > I was rationalized by the reactivity being governed by the off-rates (i.e. I H 2S adduct Pd2X2(dpm)2 + H2S). The same reactivity trend for the CO systems is definitely governed by off-rates.110 Unlike reaction 1.3.1, reaction 1.3.2 was shown to be irreversible even under 10 arm H 2 . U 1 4 A search for a possible kinetic role of the bridging dpm ligands in the forward reaction was complicated by the abstraction of sulfur from the bridged-sulfide complex by added dpm. This observation has led to further kinetic and mechanistic studies (see Chapter 3) and the first reported homogeneous catalytic conversion ofH 2S to H 2 , illustrated below: 25 References on page 37 Chapter 1 H,S — ^ *• H dpm(S) dpm(S) = bis(diphenylphosphino)methane monosulfide Alternatively, the sulfur could be eliminated as SO2 from Pd2X2(p-S)(dpm)2 following oxidation by H202 and/or m-chloroperbenzoic acid to the Pd2X2(u-S02)(dpm)2115 species which spontaneously loses S0 2 , 1 2 4 and the net reaction 1.3.3 could be affected catalytically using the Pd2X2(dpm)2 1 1,11s complexes. (1.3.3) H 2S + 2'0' -> H 2 + S02 The possibility of immobilizing Pd2X2(dpm)2 on solid supports,111 for instance via the methylene group(s) of the dpm ligand(s), has resulted in the syntheses of several complexes of the types Pd2X2(dpm)(dpmMe),111 syn- and anft'-Pd2X2(dprnMe)2,11U25 and Pd2X2(dpmMe)(Ph2Ppy)ul (dpmMe = l,l-bis(diphenylphosphino)ethane; Ph2Ppy = 2-(diphenylphosphino)pyridine). These were tested for reactivity toward CO and/or H 2 S: only Pd2X2(dpm)(dpmMe) and 5yn-Pd2X2(dpmMe)2, where the Me groups are on the same side with respect to the Pd-C-Pd plane, showed reactivity and this was slow compared to that of Pd2X2(dpm)2, reflecting steric impositions within the A-frames in accommodating the Me group(s).111'125 With H 2S, the corresponding u-S complex and concomitant H 2 formation were observed; kinetics and mechanisms were not studied, but a similar oxidative addition as in reaction 1.3.2 via intermediates akin to I is readily visualized. The incorporation of Ph2Ppy bridging ligands was a matter of synthetic convenience, and in separate studies a series of Pd 2X 2(PPh npy3. n) 2 (n > 1) 26 References on page 37 Chapter 1 complexes have been prepared with the goal of imparting solubility in aqueous systems;126 the chemistry with H2S has not been explored yet. Of note, in extending the chemistry of H2 recovery from hydrides of group 16 elements, the Pd2X2(dpm)2 complexes also react with H 2 Se 1 2 7 and H 2 O 1 2 7 but the reactions are not clean and side-products, including, as yet, uncharacterized species, are seen. With H2Se, Pd2Cl2(p-Se)(dpm)2 is formed along with H 2 but Pd2(dpm)2(p-Se)Cl(SeH) and Pd2(dpm)2(p-Se)(SeH)2 are also produced due to the greater acidity of H2Se. With H 2 0 , the reaction is complex and NMR spectroscopic data reveal numerous unidentified products; however, separate studies showed that Pd2Cl2(dpm)2(p-0) can be generated at low temperatures, for example by treating Pd2Cl2(dpm)2 with m-chloroperbenzoic acid at -50 °C; warming above -50 °C causes decomposition, again to unidentified products. Clearly, these discoveries together with other metal/H2X (X = Se, O) systems128'129 demonstrate that H 2 recovery is not necessarily limited to H 2S systems. 27 References on page 37 Chapter 1 1.4 Transition metal - H 2 S chemistry There is considerable interest in transition metal - H 2S chemistry because of its relevance in biological cycles, the formation of ores, the hydrodesulfurization catalytic process, and in the potential use of H 2S as a source of H 2 and organosulfur compounds. Literature dealing with the subject is plentiful and can be traced through recent references.130-133 In view of this laboratory's efforts to study the recovery of H 2 from H 2S (see Section 1.3), it seems appropriate to review briefly the literature regarding similar chemistry, particularly in homogeneous systems. The area of recovering H 2 from H 2S in heterogeneous systems has been described earlier (see Section 1.2). Outside of this laboratory's current work with Pd, the interaction of H 2S has been explored mainly with complexes of Ru, Rh, and Ir. Kuehn and Taube first reported117 a system generating H 2 from H 2S and tentatively suggested reaction 1.4.1, but stated that [Ru(NH3)5(H2S)]2+ was not obtained in pure state. (1.4.1) 2 [Ru(NH3)5(H2S)]2+ -> 2 [Ru(NH3)5(SH)]2+ + H 2 Pignolet's group was the first to initiate solution studies focusing on the recovery of H 2 from H 2S, and examined Rh and Ir phosphine complexes: hydrido(mercapto) species were formed but no F£2 production was seen (reactions 1.4.2 and 1.4.3).134 The formation of H 2 , however, was observed in later studies using Ir hydrido precursors (reaction 1.4.4).135 H H I Cl (1.4.2) 2 RhCl(PPh3)3 + 2 H 2S • P h 3 P ^ p ] n ^ S ^ T j n ' ' P P h 3 + 2 PPh3 P h 3 P ^ | | " P P h 3 Cl | H H 28 References on page 37 Chapter 1 PPh3 O f l I . S H (1.4.3) Ir(PPh3)2(CO)(Cl) + H 2S • ^ I r ^ cr | H PPh3 (1.4.4) 2 [Ir(H)2(Me2CO)2(PPh3)2]+ + 2H 2 S - » [fr2(p-SH)2(p-H)2(H)(PPh3)4]+ + H 2 + H + + 2Me 2 CO Similarly, Osakada et al. had studied earlier the reaction of H 2S with Ru(H) hydrido complexes (reactions 1.4.5 and 1.4.6).136'137 Quantitative evolution of H 2 was reported, but its origin was not addressed. (1.4.5) Ru(H)2(PPh3)4 + H 2S Ru(H)(SH)(PPh3)3 + H 2 + PPh3 (1.4.6) 2Ru(H)2(PMe2Ph)4 + 4H 2 S (PhMe2P)3Ru(p-SH)3Ru(SH)(PMe2Ph)2 + 4 H 2 + 3 PMe2Ph James et al. in their work 1 3 1 , 1 3 8 ' 1 3 9 demonstrated oxidative addition of H 2S to Ru(0) complexes, with the resulting Ru(II)-hydrido(mercapto) species reacting with further H 2S to generate H 2 (reaction 1.4.7); other Ru(II)-dihydrides (cf. reactions 1.4.5 and 1.4.6) also generated H 2 on reaction with H 2S via an assumed protonation of coordinated hydride (reactions 1.4.8 and 1.4.9). Two mechanisms to account for the formation of H 2 were proposed: (1) following dissociation of H 2S to H + and SH", protonation of the metal hydride liberates H 2 , and the vacant site thus generated is coordinated by SH"; and (2) coordination of H 2S at a vacated site, resulting from the loss of a labile phosphine ligand, followed by protonation of the hydride ligand, elimination of H 2 , and subsequent re-coordination of the phosphine. The possibility of the first mechanism seems remote as H 2S is a very weak acid in T H F , 1 3 0 the solvent in which the reactions were studied. 29 References on page 3 7 Chapter 1 H S H S (1.4.7) Ru(CO)2(PPh3)3 — ^ — • Ru(H)(SH)(CO)2(PPh3)2 2 > Ru(SH)2(CO)2(PPh3)2 - PPh 3 - H 2 H S H S (1.4.8) Ru(H)2(CO)2(PPh3)2 —2—+ Ru(H)(SH)(CO)2(PPh3)2 Ru(SH)2(CO)2(PPh3)2 - H 2 - H 2 H S H S (1.4.9) Ru(H)2(dpm)2 — — • Ru(H)(SH)(dpm)2 — - — • Ru(SH)2(dpm)2 - H 2 - H 2 Homo- and hetero-bimetallic complexes of Ir, Rh, and Re have been used by Cowie et al. in reactions with H 2S to ascertain further (cf. Section 1.3) the roles of adjacent metals in the activation of S-H bonds (reactions 1.4.10 - 1.4.12).130,133 NMR spectroscopy was invaluable in providing evidence for the illustrated intermediates (those not in brackets) that lead to the products. The findings show how two metals in proximity can provide means for double-activation (i.e. successive oxidation additions) as in reaction 1.4.11. Reaction 1.4.11 shows reversible H 2 elimination, a phenomenon that has been previously seen by Bianchini et al. within a characterized doubly-bridged mercapto binuclear Rh complex formed from reaction of (triphos)RhCl(C2H4) with H 2S (reaction 1.4.13).140 I O C ^ I H 2 S (1.4.10) O C — R h - * — - R e — C O — — I O C " I H 2 S P ^ - P P " - p O C . I O C ^ I - . R h * * - — : R e — C O *\ oc' I p p p - - p oc. I oc^  I . R h - * R e — C O H S I * H T O O P —P O C V P -I 'Rh I P -O C . •P I' . C O C O + H 2 30 References on page 37 Chapter 1 P P I O C ^ I H 2 S (1.4.11) O C — Ir I r — C O » I I P - . - P — CO C O p-o c v | H — I r -I P-"P - I r — C O I -P O C -p-I •Ir-I P-: Ir: CO • H 2 + H 2 ^ O C -P-I Ir-I P-•P I •Ir' I -P • CO (1.4.12) O C -I oc^  I Rh-I P -R h - C O I -P O C - I •Rh. I P -- P I - R h ' I - P CO + H 2 (1.4.13) H I P ^ V ' R h R h J 2 H , In other studies, Bottomley et al. and Howard and Parkin have examined the interactions of H 2S with metallocenes of Ti and Zr (reactions 1.4.14-1.4.17). Reaction 1.4.14 is complicated and the stoichiometry has not been unequivocally substantiated.144 The H 2 is presumably formed via protonation of intermediate hydrides. 31 References on page 37 Chapter 1 (1.4.14) 10Cp2Ti(CO)2 + 12H2S -> 2Cp 5 Ti 5 S 6 + 7 H 2 + 20 CO + 10C 6 H 5 (1.4.15) Cp* 2Ti(CO) 2 + 2H 2 S -> Cp*2Ti(SH)2 + H 2 + 2 CO (1.4.16) 2 Cp2Zr(CO)2 + 2H 2 S -> [Cp2Zr(u-S)]2 + 2 H 2 + 4 CO (1.4.17) 2 Cp*2Zr(CO)2 + 3H 2 S [Cp*2Zr(SH)(p-S)]2 + 2 H 2 + 4 CO Rabinovich and Parkin have investigated the interaction of H 2S with W phosphine complexes143 in attempts to unravel the role the metal plays in hydrodesulfurization;7 quantitative evolution of H 2 was observed for reaction 1.4.18, which is the first to demonstrate formation of a terminal sulfide ligand from an H 2S reaction. That facile elimination of H 2 from H 2S occurs to give W(PMe3)4(S)2 is of considerable interest in view of the proposal that hydrogenation of organic substrates during HDS may involve hydrogen transfer from an -SH group.144 - P M e 3 (1.4.18) W(PMe3)4(q2-CH2PMe2)H + 2 H 2S • W(PMe3)4(H)2(SH)2 • rrans-W(PMe3)4(S)2 + 2 H 2 32 References on page 37 Chapter 1 1.5 The chemistry of transition metal - dpm complexes The expectation that two metal atoms in proximity can lead to reactivity patterns different to those of a single metal has resulted in broad development of the chemistry of binuclear complexes. In this respect, bis(diphenylphosphino)methane serves as a versatile ligand that is able to lock two metals in close vicinity forming closed (metal-metal bonded) or "open" five-membered rings: P h 2 P ' ^ V " P P h 2 I I M M The ligand essentially prevents dissociation of dimer to monomer and can promote bridging by other groups as well as promoting reactions involving formation and cleavage of metal - metal bonds. Since the first dpm-bridged dinuclear complex, [RhCl(CO)(dpm)]2, was prepared by Mague and Mitchener in 1969,145 many more have emerged and examples can be found for most of the transition metals. Because of the enormity of the subject, no attempt will be made here to review the chemistry, and the present short discussion is confined to more notable examples of reactivity found with the intensively studied dpm-bridged dinuclear complexes of group 10 elements of which Pd and Pt have drawn the most attention. There are several excellent reviews on transition metal - dpm complexes that include a description of mononuclear as well as heterobimetallic species containing dpm also as a chelating, non-bridging ligand. 1 4 6 - 1 4 9 Of note, the dpm-chemistry of Ni is relatively new and little developed. Kubiak's group reported the first dpm-bridged dinickel complex, [Ni2(p-CNMe)(CNMe)3(p-dpm)2]PF6, which is asymmetrical and contains a dative metal-metal bond.150 33 References on page 37 Chapter 1 Doubly-bridged M2X2(p-dpm)2 complexes of Pd and Pt display several types of reactivities. The first is the displacement of, or insertion into, the terminal ligands by anionic or neutral ligands as exemplified by the following, where X = Br, I, NCO, NCS, N 3 , N0 3 , or PPh 151-156 Ph2P I (1.5.1) C I — M -Ph2P^ PPh2 I - M - C l + 2 X o r 2 X " I „PPh2 ~~10 or 2+ PI12P PPh2 I I X - M M - X + 2C1-I I Ph2P^ .PPI12 PI12P PPh2 PI12P PPh2 Ph 2P PPh2 I I + SnCb I I I I (1.5.2) a — M M - C l *• a — M M — SnCb or CbSn — M M —SnCb I I I I I I P h 2 P ^ ^ P P h 2 P h 2 P ^ ^ P P h 2 PI12P. ^ P P h 2 Second, addition of small molecules (or atoms) across the metal-metal bond can occur with generation of A-frame complexes: PI12P PPh2 I I (1.5.3) X - M M - X + B I I Ph2P^ -PPI12 PI12P PPh2 I „ B . „ I i1 'i M M X ' PfuP, I .PPh2 B includes C H 2 , S0 2, PhN 2 + , CO, H + , RNC, S, SO, O, Se, and SeO, and double atom bridges such as RC^CR and CS2Z^ 1 3 - 1 1 5 ' 1 2 4 ' 1 2 7 ' 1 4 6 ' 1 4 7 xhe auxiliary ligands can be simple inorganic anions such as those stated above or predominantly non-ionic organic groups like CeF 5 , 1 5 7 C 6 C I 5 , 1 5 7 and C 6 H 5 S . 1 4 6 With these A-frame complexes, addition of small molecules in the 'pocket' position does not take place147 (but see Chapter 4) but insertion reactions into the terminal ligands can occur in some cases:158 34 References on page 37 Chapter 1 (1.5.4) Ph2P PPh2 Ph2P PPh2 I x I - _ - _ or Pd' 'Pd MeK I I Me M e ^ I I Ph2P^ .PPh2 Ph2P^ ^PPh2 I x -i1' 'i Pd Pd Ph2P PPra CO I , , X M I * - Pd Pd MeOC^ I Ph2P^ 1 + I ^COMe PPh2 X = Br, I or Ph2P MeOC' Pd ' I PraP. I x. I I , 1 - I , I 'PPh2 ~l+ Pd ,PPh2 The Pd2X2(p-dpm)2 complexes have been shown to exhibit catalytic activity. For example, the cyclotrimerization of alkynes is catalyzed by Pd2Cl2(u.-dpm)2 or Pd2I2(p-dpm)2 (reaction 1.5.5).159 Preliminary mechanistic studies indicated that the reaction proceeds via a metallocyclic alkene, A-frame intermediate (see above). The second example involves work from this laboratory that demonstrated that the Pd2X2(dpm)2 complexes catalyze the conversion of H 2S and dpm to H 2 and dpm(S) (see Section 1.3 and Chapter 3). 123,160 (1.5.5) 3 Me02C—C=C—CC^vle R = -COsMe 35 References on page 37 Chapter 1 1.6 Statement of the problem From the foregoing discussion, it is clear that an interest in transition metal - H 2S chemistry is the impetus for the present thesis work. Further exploration of the sulfur abstraction from Pd2X2(u-S)(dpm)2 by dpm (process 2 -> 1, R = dpm, in the figure below) was conducted (Chapter 3) in order to answer pertinent questions associated with the reaction, such as the role of the bridging dpm ligands, the effect of the auxiliary ligands, the nature of reaction intermediates, and the reasons for the unsuccessful sulfur abstraction by other phosphines (i.e. dpe, PPI13, and PPh2Me).1 6 0 In addition, further studies concentrated on unearthing other reagents for effective removal of sulfur from 2 to regenerate 1; possible candidates included CO (Chapter 7) as well as the halogens, X 2 (Chapter 4). Chapter 2 outlines general experimental procedures while Chapters 5 and 6 describe the reaction of PdX2(dpm) with H 2S to generate 2. Ph^P" •PPh2 Ph^P" PPh 2 X — P d P d — X ^ - P d ' , . -S- , X 1 PhijP- PPh 2 Ph^P-reagent (R) 2 36 References on page 37 Chapter 1 1.7 References for Chapter 1 1. Lee, C-L . ; Besenyei, G . ; James, B.R.; Nelson, D . A . ; Lilga, M . A . J. Chern. Soc, Chern. Commun. 1985, 1175. 2. Subcommittee on Hydrogen Sulfide, Committee on Medical and Biologic Effects of Environmental Pollutants, Division of Medical Services, Assembly of Life Sciences, National Research Council in 'Hydrogen Sulfide,' University Park Press, Baltimore, 1979. 3. Canada, Environmental Protection Programs Directorate, Technical Services Branch in 'Hydrogen Sulfide - Technical Services Branch, Environmental Protection Programs Directorate, Environmental Protection Service,' Gov. Doc. No. En48-10/17-1984E, The Service, Ottawa, 1984. 4. Hydrogen Sulfide, vol. 22, p. 114. In H.F. Mark, J.J. Mcketta, Jr., and D F . Othmer. Kirk-Othmer Encyclopedia of Chemical Technology (3rd ed.), Toronto: John Wiley & Sons, Inc., 1983. 5. Scheele, K.W. Ann. de Chim. Mil, 25, 233; through ref. 2. 6. Kraft Chemical Pulping, vol. 19, p. 395. In H.F. Mark, J.J. Mcketta, Jr., and D.F. Othmer. Kirk-Othmer Encyclopedia of Chemical Technology (3rd ed.), Toronto: John Wiley & Sons, Inc., 1983. 7. (a) Gates, B .C. ; Katzer, JR.; Schuit, G.C.A. in 'Chemistry of Catalytic Processes,' McGraw-Hill, New York, 1979, ch. 5. (b) Sanchez-Delgado, R.A. J. Mol. Catalysis 1994, 86, 287. 8. Deuterium and Tritium, vol. 7, p. 544. In H.F. Mark, J.J. Mcketta, Jr., and D.F. Othmer. Kirk-Othmer Encyclopedia of Chemical Technology (3rd ed.) Toronto: John Wiley & Sons, Inc., 1983. 9. Chern. Eng. News 1995, 73, 38. 10. Sulfur Recovery, vol. 22, p. 267. In H.F. Mark, J.J. Mcketta, Jr., and D.F. Othmer. Kirk-Othmer Encyclopedia of Chemical Technology (3rd ed.) Toronto: John Wiley & Sons, Inc., 1983. 11. Hydrogen Energy Progress-IV; Veziroglu, T . N . ; Van Vorst, W.D. ; Kelly, J.H., Eds.; Advances in Hydrogen Energy 3; Pergamon Press: Toronto, 1982. 12. Berk, D. ; Heidemann, R.A.; Svrcek, W.Y.; Behie, L .A. Can. J. Chern. Eng. 1991, 69, 944. 13. Taylor, H.A. ; Pickett, C F . J. Phys. Chern. 1927, 31, 1212. 37 Chapter 1 14. Kingman, F.E.T. Trans. Faraday Soc. 1936, 32, 903. 15. Darwent, B. de B.; Roberts, R. Proc. Roy. Soc. London, Ser A 1953, 216, 344. 16. Raymont, M . E . D . Ph.D. Dissertation, University of Calgary, Alberta, Canada, 1974. 17. (a) Kaloidas, V . E . ; Papayannakos, N.G. Chem. Eng. Sci. 1989, 44. 2493. (b) Kaloidas, V . E . ; Papayannakos, N.G. Int. J. Hydrogen Energy 1987, 12, 403. 18. Massey, L . G . US Patent 2 984 548 (1958); Chem. Abstr. 55, 21510a (1961). 19. Raymont, M . E . D . Hydrocarbon Processing 1975, 139. 20. Bogossian, A. FR Patent 2 238 668 (1975); Chem. Abstr. 83, 134380 (1975). 21. Daley, W.D. ; Young, R.D. US Patent 4 461 755 (1984); Chem. Abstr. 101, 154222 (1984). 22. Hellmer, L . ; Keunecke, G. ; Pachaly, R.; Lell, R ; Al-Muddarris, G.R.; Stauffer, A . ; Vangala, V.R. Chem. Abstr. 94, 33047 (1981). 23. Takeda, T. JP Patent 89 270 502 (1989); Chem. Abstr. 112, 237702 (1990). 24. Borton, D . N . ; Rogers, W.E. Proc. Intersoc. Energy Convers. Eng. Conf. 1990, 4, 315. 25. (a) Salman, O.A. ; Bishara, A . ; Marafi, A . ; Abu-Khalifeh, H . ; Al-Sayegh, A. Adv. Hydrogen Energy 8 (Hydrogen Energy Prog. 8, vol. 2) 1990, 539. (b) Bishara, A . ; Salman, O.A. ; Khraishi, N . ; Marafi, A. Int. J. Hydrogen Energy 1987, 12, 679. 26. Kappauf, T . A . Diss. Abstr. Int. B 1989, 50, 651. 27. Fletcher, E . A . ; Noring, J.E.; Murray, J.P. Int. J. Hydrogen Energy 1984, 9, 587. 28. Weiner, J .G.; Leggett, C.W. US Patent 2 979 384 (1958); Chem. Abstr. 55, 21510d (1961). 29. Kotera, Y . K . ; Todo, N T . ; Fukuda, K . N . US Patent 3 962 409 (1976); Chem. Abstr. 79, 33200 (1973). 30. Fukuda, K . N . ; Dokiya, M . ; Kameyama, T. ; Kotera, Y. Ind. Eng. Chem. Fundam. 1978, 17, 243. 31. Hirota, K. ; Aizawa, K. ; Nishizawa, T. Z. Naturforsch. 1979, 34a, 43. 32. Chivers, T . ; Hyne, J.B.; Lau, C. Int. J. Hydrogen Energy 1980, 5, 499. 33. Katsumoto, M . ; Fueki, K. ; Mukaibo, T. Bull. Chem. Soc. Jpn. 1973, 46, 3641. 34. Sugioka, M . ; Aomura, K. Int. J. Hydrogen Energy 1984, 11, 891. 35. Knoche, K.F . ; Cremer, H . ; Steinbora, G.;Schneider, W. Int. J. Hydrogen Energy 1977, 2, 269. 38 Chapter 1 36. Kiuchi, H . ; Nakamura, T. ; Funaki, K . ; Tanaka, T. Int. J. Hydrogen Energy 1982, 7, 477. 37. Kiuchi, H . ; Funaki, K. ; Tanaka, T. Metallurgical Trans. B 1983, 14, 347. 38. Chivers, T . ; Lau, C. Int. J. Hydrogen Energy 1987,12, 235. 39. Al-Shamma, L . M . ; Naman, S.A. Int. J. Hydrogen Energy 1987, 14, 173. 40. Chivers, T . ; Lau, C. Int. J. Hydrogen Energy 1987,12, 561. 41. Zhang, Y. ; Teng, Y. ; Zang, X. ; Wang, Z. HuadongHuagongXueyuanXuebao 1991, 17, 586. 42. Yumura, M . ; Furimsky, E. Appl. Catalysis 1985, 16, 157. 43. Chivers, T . ; Lau, C. Int. J. Hydrogen Energy 1985, 10, 21. 44. Bandermann, F.; Harder, K.B. Int. J. Hydrogen Energy 1985,10, 21. 45. Dokiya, M . ; Fukuda, K. ; Yokokawa, H . ; Kameyama, T. Bull. Chern. Soc. Jpn. 1978, 51, 150. 46. Bowman, M . G . US Patent 4 999 178 (1991); Chern. Abstr. 115, 117319 (1991). 47. Rendall, W.A. ; Moir, M . E . ; Szarka, J. GB Patent 2 248 444; Chern. Abstr. 117, 51829 (1992). 48. (a) Towler, G.P.; Lynn, S. lnd. Eng. Chern. Res. 1993, 32, 2800. (b) Towler, G.P.; Lynn, S. Ind. Eng. Chern. Res. 1993, 32, 2812. 49. Schuman, S.C. US Patent 3 388 971 (1968); Chern. Abstr. 69, 44920 (1968). 50. Nixon, I.G. Chern. Brit. 1974, 179. 51. Chen, W.C. US Patent 4 066 739 (1978); Chern. Abstr. 88, 107468 (1978). 52. Plummer, M . A . Hydrocarbon Processing 1987, 38. 53. Okabe, H . in 'Photochemistry of Small Molecules,' Wiley-Interscience Publication, Wiley & Sons, Toronto, 1978, ch. 6, p. 204, and references therein. 54. Fujii, T. JP Patent 63 267 420; Chern. Abstr. 110, 157677 (1989). 55. Kalyanasundaram, K. ; Borgarello, E . ; Gratzel, M . Helv. Chim. Acta 1981, 64, 362. 56. Kalyanasundaram, K. ; Borgarello, E . ; Duonghong, D . ; Gratzel, M . Angew. Chern. Int. Ed. Engl. 1981, 20, 987. 57. Serpone, N . ; Borgarello, E . ; Barbeni, M . ; Pelizzetti, E. Inorg. Chim. Acta 1984, 90, 191. 58. Borgarello, E . ; Kalyanasundaram, K. ; Gratzel, M . ; Pelizzetti, E. Helv. Chim. Acta 1982, 65, 243. 39 Chapter 1 59. Borgarello, E . ; Erbs, W.; Gratzel, M . Nouv. J. Chim. 1983, 7, 195. 60. Thewissen, D.H.M.W. ; Tinnemans, A . H . A . ; Eeuwhorst-Reinten, M . ; Timmer, K . ; Mackor, A. Nouv. J. Chim. 1983, 7, 191. 61. Borgarello, E . ; Gratzel, M . ; Kalyanasundaram, K. ; Pelizzetti, E. Chemtech. 1983,13, 118. 62. Gratzel, M . (Ed.) in 'Energy Resources Through Photochemistry and Catalysis,' Academic Press, New York, 1983, ch. 3, p. 95. 63. Borgarello, E . ; Desilvestro, J.; Gratzel, M ; Pelizzetti, E. Helv. Chim. Acta 1983, 66, 1827. 64. Serpone, N . ; Borgarello, E . ; Gratzel, M . J. Chem. Soc, Chem. Commun. 1984, 342. 65. Thewissen, D . H . M . W . ; Van der Zouwen-Assink, E .A. ; Timmer, K. ; Tinnemans, A . H . A . ; Mackor, A. J. Chem. Soc, Chem. Commun. 1984, 941. 66. Barbeni, M . ; Pelizzetti, E . ; Borgarello, E . ; Serpone, N. ; Gratzel. M . ; Balducci, L . ; Visca, M . Int. J. Hydrogen Energy 1985, 10, 249. 67. Naman, S.A.; Aliwi, S .M. ; Al-Emara, K. Int. J. Hydrogen Energy 1986,11, 33. 68. Kainthla, R.C. ; Bockris, J. O ' M . Int. J. Hydrogen Energy 1987, 12, 23. 69. Naman, S.A.; Al-Emara, K. Int. J. Hydrogen Energy 1987, 12, 629. 70. Linkous, C . A . ; Muradov, N.Z. ; Ramser, S.M.; Mingo, T . E . ; Zidan, R.A. Front. Sci. Ser. 1993, 7 (New Energy Systems and Conversions), 173. 71. Linkous, C . A . ; Mingo, T . E . ; Muradov, N.Z. Int. J. Hydrogen Energy 1994, 19, 203. 72. Osawa, Y . ; Murata, S.; Tanaka, K. JP Patent 87 12 601; Chem. Abstr. 106, 198662 (1987). 73. Lee, R.; Grove, E. US Patent 2 839 381 (1958); Chem. Abstr. 52, 14503f (1958). 74. Lim, H.S.; Winnick, J. J. Electrochem. Soc. 1984,131, 562. 75. Farooque, M . ; Fahidy, T.Z. J. Electrochem. Soc 1977, 124, 1191. 76. Grishaeva, G.A. ; Shalepova, N. Ya. ; Tret'yak, A . N . ; Novikov, G.I. Vesti Akad. Navak BSSR, Ser. Fiz.-Energ. Navuk 1982, 2, 85; Chem. Abstr. 97, 100542 (1982). 77. Bolmer, P.W. US Patent 3 409 520 (1968); Chem. Abstr. 70, 39514 (1969). 78. Gregory, T . D . ; Feke, D.L . ; Angus, J.C.; Brosilow, C.B.; Landau, U. J. Applied Electrochem. 1980, « 40 Chapter 1 10, 405. 79. Dandapani, B.; Scharifker, B.R.; Bockris, J. O ' M . Proc. Intersoc. Energy Convers. Eng. Conf. 1986, 1, 262. 80. Shih, Y.S. ; Lee, J.L. Ind. Eng. Chem. Process Des. Dev. 1986, 25, 834. 81. Nygren, K. ; Atanasoski, R.; Smyrl, W.H. ; Fletcher, E .A. Energy 1989, 14, 323, and references therein. 82. Anani, A . A . ; Mao, Z. ; White, R.E. ; Srinivasan, S.; Appleby, A.J. J. Electrochem. Soc. 1990, 137, 2703. 83. Mao, Z. ; Anani, A . A . ; White, R.E. ; Srinivasan, S.; Appleby, A.J. J. Electrochem. Soc. 1991, 137, 1299. 84. Alexander, SR. ; Winnick, J. AIChE Journal 1994, 40, 613, and references therein. 85. Fischer, F. US Patent 1 891 974 (1932); Chem. Abstr. 22, 4676 (1928). 86. (a) Honna, K. ; Noguchi, H . ; Iida, H . ; Goto, M . JP Patent 94 209 153 (1993); Chem. Abstr. 121, 208919 (1994). (b) Honna, K. ; Iida, H. JP Patent 04 238 803 (1992); Chem. Abstr. 117, 236655 (1992). (c) Honna, K. ; Iida, H. JP Patent 91 136 661 (1992); Chem. Abstr. 118, 215837 (1993). (d) Honna, K . ; Kikuchi, O. JP Patent 90 409 679 (1992); Chem. Abstr. 117, 254356 (1992). (e) Kikuchi, T . ; Komori, N . ; Iida, H. JP Patent 90 169 709 (1992); Chem. Abstr. 116, 109500 (1992). (f) Noguchi, H . ; Isshiki, S. JP Patent 88 160 445 (1990); Chem. Abstr. 112, 121612 (1990). (g) Noguchi, H . ; Itsushiki, S. JP Patent 87 216 133 (1989); Chem. Abstr. I l l , 136916 (1989). (h) Konishi, T . ; Kikuchi, O. JP Patent 86 58 060 (1987); Chem. Abstr. 107, 239271 (1987). (I) Iida, H . ; Misune, K. JP Patent 95 228 914 (1987): Chem. Abstr. 107, 224985 (1987). (j) Agency of Industrial Sciences and Technology, Japan JP Patent 82 64 962 (1984); Chem. Abstr. 100, 23986 (1984). 87. Mizuta, S.; Kondo, W.; Fujii, K. ; Iida, H . ; Isshiki, S.; Noguchi, H . ; Kikuchi, T . ; Sue, H . ; Sakai, K. Ind. Eng. Chem. Res. 1991, 30, 1601. 88. Keller, H.F. US Patent 3 401 101 (1968); Chem. Abstr. 69, 88621 (1968). 89. Deem, C.K. US Patent 4 220 505 (1981); Chem. Abstr. 94, 32975 (1981). 41 Chapter 1 90. Schuetz, G. Eur. Pat. Appl. EP 168 022; Chern. Abstr. 104, 112212 (1986). 91. (a) Kalina, D.W.; Maas, Jr., E.T. Int. J. Hydrogen Energy 1985, 10, 157. (b) Kalina, D.W. ; Maas, Jr., E.T. Int. J. Hydrogen Energy 1985,10, 163. 92. Patton, R.A. US Patent 2 819 950 (1958); Chern. Abstr. 52, 7636c (1958). 93. Asisov, R.J.; Vakur, A . K . ; Gutsol, A.F . ; Givotov, V . K . ; Krasheninnikov, E . G . ; Krotov, M.F . ; Rusanov, V . D . ; Fridman, A . A . ; Sholin, G.V. Int. J. Hydrogen Energy 1985, 10, 475. 94. Krashenninnikov, E . G . ; Rusanov, V . D . ; Sanyuk, S.V.; Fridman, A . A . Sov. Phys. Tech. Phys. 1986, 31, 645. 95. (a) Harkness, J.B.L.; Gorski, A.J. ; Daniels, E.J. Proc. Intersoc. Energy Convers. Eng. Conf. 1990, 6, 197. (b) Harkness, J.B.L.; Gorski, A.J. ; Daniels, E.J. US Patent 5 211 923 (1991); Chern. Abstr. 119, 75745 (1993). 96. Bagautdinov, A . Z . ; Jivotov, V . K . ; Eremenko, J.J.; Kalachev, I.A.; Musinov, S.A.; Pampushka, A . M . ; Rusanov, V . D . ; Zoller, V . A . Front. Sci. Ser. 1993, 7 (New Energy Systems and Conversions), 123. 97. Haas, L . A . ; Khalafalla, S.E. US Natl. Tech. Inform. Service, Bureau of Mines, Report of Investigations No. 7780 (1973). 98. Traus, I.; Suhr, H. Plasma Chern. Plasma Process. 1992, 12, 275. 99. (a) Jorgensen, P.; Czeraichowski, A . ; Chapelle, J.; Meguernes, K. FR Patent 2 620 436 (1989); Chern. Abstr. I l l , 107383 (1989). (b) Czerhnichowski, A . ; Jorgensen, P.; Lesueur, H . ; Chapelle, J.; Meguernes, K. Colloq. Phys. 1990, (C5, Eur. Congr. Therm. Plasma Processes Mater. Behav. High Temp., 1990), C5-65/C5-71. 100. Khrikulov, V . V . ; Ismailov, Kh.I. Gazov. Prom-st. 1994, 4, 38; Chern. Abstr. 122, 221442 (1995). 101. Schwarz, R.; Kunzer, W. Z. Anorg. Allg. Chern. 1929,183, 287. 102. Ferrero, P.; Corbaz, J. Helv. Chim. Acta 1931,14, 912. 103. Kolokina, L. J. Phys. Chern. (USSR) 1935, 6, 428. 104. Nicholas, J.E.; Amodio, C . A . ; Baker, M.J. J. Chern. Soc, Faraday Trans. 1 1979, 75, 1868. 105. Boyd, A .W. ; Willis, C ; Miller, O A . Can. J. Chern. 1973, 51, 1228, and references therein. 42 Chapter 1 106. Forys, M ; Jezierska, K. Radiochem. Radioanal. Lett. 1969, 1, 237, and references therein. 107. Kerr, J.A. Chem. Rev. 1966, 66, 465. 108. Ginter, D.S. ; Ginter, M L . ; Innes, K.K. Astrophys. J. 1964,139, 365. 109. Nelson, D . A . ; Hallen, R.T.; Lee, C-L. ; James, B.R. in 'Recent Developments in Separation Science,' N . L i and J.M. Calo, Eds., CRC, Cleveland, Ohio, 1986, vol. 9, p. 1. 110. Lee, C-L . ; James, B.R.; Nelson, D . A . ; Hallen, R.T. Organometallics 1984, 3, 1360. 111. Lee, C-L . ; Yang, Y.P.; Rettig, S.J.; James, B.R.; Nelson, D . A . ; Lilga, M A . Organometallics 1986, 5, 2220. 112. Lyke, S.E.; Lilga, M . A . ; Ozanich, R . M . ; Nelson, D A . Ind. Eng. Chem. Prod. Res. Dev. 1986, 25, 517. 113. Benner, L.S.; Balch, A . L . J. Am. Chem. Soc. 1978,100, 6099. 114. Barnabas, A.F . ; Sallin, D . ; James, B.R. Can. J. Chem. 1989, 67, 2009. 115. Besenyei, G . ; Lee, C-L. ; Gulinski, J.; Rettig, S.J.; James, B.R.; Nelson, D . A . ; Lilga, M . A . Inorg. Chem. 1987, 26, 3622. 116. Kubiak, C P . ; Eisenberg, R. J. Am. Chem. Soc. 1977, 99, 6129. 117. Kuehn, G.C. ; Taube, H . J. Am. Chem. Soc. 1976, 98, 689. 118. Herberhold, M . ; Suss, G. Angew. Chem. Int. Ed. Engl. 1976, 75, 366; J. Chem. Res. Synop. 1977, 246. 119. Weiss, A . ; Plass, R ; Weiss, A. Z. Anorg. Allg. Chem. 1956, 283, 390. 120. Ugo, R.; LaMonica, G . ; Cenini, S.; Segre, A . ; Conti, F. J. Chem. Soc. A 1971, 522. 121. Sellman, D . ; Lechner, P.; Knoch, F.; Moll, M . Angew. Chem. Int. Ed. Engl. 1991, 30, 552. 122. Mudalige, D . C . Ph.D. Dissertation, University of British Columbia, Vancouver, 1994. 123. Wong, T . Y . H . ; Barnabas, A.F . ; Sallin, D . ; James, B.R. Inorg. Chem. 1995, 34, 2278. 124. Balch, A . L . ; Benner, L.S.; Olmstead, M . M . Inorg. Chem. 1979, 18, 2996. 125. Besenyei, G . ; Lee, C-L. ; Xie, Y. ; James, B.R. Inorg. Chem. 1991, 30, 2446. 126. Xie, Y. ; Lee, C-L . ; Yang, Y. ; Rettig, S.J.; James, B.R. Can. J. Chem. 1992, 70, 751. 43 Chapter 1 127. Besenyei, G . ; Lee, C-L . ; James, B.R. J. Chern. Soc, Chern. Commun. 1986, 1750. 128. Bottomley, F.; Chin, T .T . ; Egharevba, G O . ; Kane, L . M . ; Pataki, D . A . ; White, P.S. Organometallics 1988, 7, 1214. 129. Hillhouse, G.L. ; Bercaw, J.E. J. Am. Chern. Soc 1984, 106, 5472. 130. McDonald, R.; Cowie, M . Inorg. Chern. 1993, 32, 1671. 131. Jessop, P.G. ; Lee, C-L . ; Rastar, G. ; James, B.R.; Lock, C.J.L.; Faggiani, R. Inorg. Chern. 1992, 31, 4601. 132. Shih, K - Y . ; Fanwick, P.E.; Walton, R.A. Inorg. Chern: 1992, 31, 3663. 133. Antonelli, D . M . ; Cowie, M . Inorg. Chern. 1990, 29, 3339. 134. Mueting, A . M . ; Boyle, P.D.; Pignolet, L . H . Inorg. Chern. 1984, 23, 44, and references therein. 135. Mueting, A . M . ; Boyle, P.D.; Wagner, R.; Pignolet, L . H . Inorg. Chern. 1988, 27, 271. 136. Osakada, K. ; Yamamoto, T. ; Yamamoto, A. Inorg. Chim. Acta 1984, 90, L5. 137. Osakada, K . ; Yamamoto, T . ; Yamamoto, A . ; Takenaka, A . ; Sasada, Y. Inorg. Chim. Acta 1985, 105, L9. 138. Lee, C-L . ; Chisholm, J.; James, B.R.; Nelson, D . A . ; Lilga, M . A . Inorg. Chim. Acta 1986, 121, L7. 139. a) Jessop, P.G. ; Rettig, S.J.; Lee, C-L. ; James, B.R. Inorg. Chern. 1991, 30, 4617. b) Jessop, P.G. ; Rastar, G . ; James, B.R. Inorg. Chim. Acta, in press. 140. Bianchini, C ; Mealli, C ; Meli, A . ; Sabat, M . Inorg. Chern. 1986, 25, 4618. 141. Bottomley, F.; Drummond, D.F. ; Egharevba, G.O. ; White, P.S. Organometallics 1986, 5, 1620. 142. Howard, W.A. ; Parkin, G. Organometallics 1993, 12, 2363. 143. Rabinovich, D. ; Parkin, G. J. Am. Chern. Soc. 1991, 113, 5904. 144. Angelici, R.J. Acc. Chern. Res. 1988, 21, 387. 145. Mague, J.T.; Mitchener, J.P. Inorg. Chern. 1969, 8, 119. 146. Balch, A . L . in 'Catalytic Aspects of Metal Phosphine Complexes,' E .C. Alyea and D.W. Meek, eds., A C S Adv. Chern. Ser., 1982, v. 196, p. 243. 147. Puddephatt, R.J. Chern. Rev. 1983, 99, and references therein. 44 Chapter 1 148. Shaw, B.L. Proc. Indian Natn. Sci. Acad. 1986, 52A, 744. 149. Chaudret, B.; Delavaux, B.; Poilblanc, R. Coord. Chern. Rev. 1988, 86, 191. 150. DeLaet, D .L . ; Powell, D.R.; Kubiak, C P . Organometallics 1985, 4, 954. 151. Brown, M.P. ; Puddephatt, R.J.; Rashidi, M . ; Seddon, K.R. J. Chern. Soc, Dalton Trans. 1977, 951. 152. Pringle, P.G. ; Shaw, B.L. J. Chern. Soc, Chern. Commun. 1982, 81. 153. Olmstead, M M . ; Benner, L.S. ; Hope, H . ; Balch, A . L . Inorg. Chim. Acta 1979, 32, 193. 154. Brown, M.P. ; Franklin, S.J.; Puddephatt, R.J.; Thomson, M . A . ; Seddon, K.R. J. Organomet. Chern. 1979, 178, 281. 155. Grossel, M . C . ; Moulding, R.P.; Seddon, K.R. Inorg. Chim. Acta Lett. 1982, 64, L275. 156. Weng, W.Z.; James, B.R. To be published. 157. Espinet, P.; Fornies, J.; Fortuno, C ; Hidalgo, G. ; Martines, F.; Tomas, M . ; Welch, A.J. J. Organomet. Chern. 1986, 317, 105. 158. Lee, C-L . ; Hunt, C.T. ; Balch, A . L . Organometallics 1982,1, 824. 159. Lee, C-L. ; Hunt, C.T. ; Balch, A . L . Inorg. Chern. 1981, 20, 2498. 160. Barnabas, A.F . M.Sc. Dissertation, University of British Columbia, Vancouver, 1989. 45 CHAPTER 2 General Experimental Procedures 46 Chapter 2 Chapter 2 General Experimental Procedures A general description of the materials, techniques, and instrumentation used during the course of the thesis work will be presented in this chapter. Detailed information on experiments and synthetic routes to novel compounds can be found in subsequent chapters. 2.1 Materials Tables 2.1 and 2.2 list the materials and compounds used with reference to commercial sources and/or literature procedures. Commercially available chemicals were pure and were used as received. The synthetic routes for the Pd complexes listed in Table 2.2 are also given; in each case, the spectroscopic data listed agree closely with those given in the literature. Unless otherwise stated, all syntheses were carried out using standard Schlenk techniques and under an inert atmosphere of N 2 ; purities were assessed by NMR spectroscopy and/or elemental analyses. All non-deuterated solvents were obtained from MCB, BDH, Aldrich, Eastman, Fisher, or Mallinckrodt Chemical Co., and were freshly distilled under N 2 over their respective drying agents prior to use (i.e. benzene, diethyl ether, hexanes, THF, and toluene using Nafoenzophenone; MeOH and EtOH using Mg/I2; CH2C12 and CHC13 using CaH 2; and acetone using K 2 C0 3 ) . Deuterated solvents (CD2C12, CDC13, CD 3 CN, C 6 D 6 , DMSO-de, CH 3 OD, and D 2 0) were obtained from CIL, Isotec Inc., or MSD Isotopes, and were de-oxygenated or de-gassed and dried (except D 2 0) over activated molecular sieves (Fisher, type 4A, grade 514, 8 -12 mesh). Purified N 2 (U.S.P.), Ar (H.P.), and H 2 (U.S.P.) were obtained from Union Carbide Canada Ltd., H 2S (CP.) 47 References on page 65 Chapter 2 Table 2.1 Some materials used and their sources. Chemical Source y-alumina (neutral) Fisher aluminum hydroxide (Al(OH)3) Aldrich aluminum oxide (a-Al203, corundum) Alfa Products benzonitrile (PhCN) Aldrich bis(diphenylphosphino)ethane (dpe) Strem 1,1 '-bis(diphenylphosphino)ethane (dpmMe) ref. 1 bis(diphenylphosphino)methane (dpm) Aldrich bis(diphenylphosphino)methane monosulfide (dpm(S)) ref. 2 bis(diphenylphosphino)propane (dpp) Strem bromine (Br2) BDH calcium sulfide (CaS) Alfa Products dibenzylideneacetone (dba) ref. 3 ethyl iodide (CH3CH2I) BDH hydrobromic acid (HBr (48% aq.)) Fisher hydriodic acid (HI (59% aq.)) Fisher hydrochloric acid (HCl (37% aq.)) Fisher iodine (I2) BDH methyl iodide (CH3I) Fisher methyl lithium (MeLi (1.40 M in Et20)) Aldrich palladium(II) chloride (PdCl2) Johnson Matthey Ltd. sodium acetate (NaOAc) Mallinckrodt sodium bromide (NaBr) Anachemia sodium hydrosulfide (NaSH) Aldrich sodium iodide (Nal) Fisher sulfur Aldrich tetrabutylammonium iodide ((CH3(CH2)3)4NI) Eastman tetrapropylammonium bromide ((CH3(CH2)2)4NBr) Kodak triphenylphosphine (PPh3) Strem 48 References on page 65 Chapter 2 Table 2.2 Palladium complexes used and their respective reference(s) for synthetic procedures. Compound Source rrara-Pd(PhCN) 2Cl 2 refs. 4, 5, 6 Pd 2(dba) 3 'CHCl 3 refs. 4, 7 Pd2Cl2(dpm)2 (la) refs. 4, 8 Pd2Br2(dpm)2 (lb) refs. 4, 9 Pd2I2(dpm)2«CH2Cl2 (lc) refs. 4, 9 Pd2Cl2(u-S)(dpm)2 (2a) refs. 4, 10 Pd2Br2(u-S)(dpm)2 (2b) refs. 4, 10 Pd2I2(n-S)(dpm)2 (2c) refs. 4, 10 PdCl2(dpm) (9a) refs. 4, 11, 12 PdBr2(dpm) (9b) refs. 4, 11 Pdl2(dpm) (9c) -Pd2Cl2(u-SO)(dpm)2 ref. 10 Pd2I2(u-SO)(dpm)2 -Pd2(SH)2(u-S)(dpm)2 ref. 10 49 References on page 65 Chapter 2 and anhydrous HCl from Matheson Gas Co., anhydrous HBr from Linde, O2 (U.S.P.) and CO (CP.) from Praxair, and COS from Aldrich. All gases were used without further purification. 2.1.1 Dibenzylideneacetone (dba) The dba ligand was readily available in this laboratory. Synthesis can be done according to a published method.3 Acetone (1.45 g, 25 mmol) is condensed with PhCHO (5.3 g, 50 mmol) in a solution of NaOH in aq. EtOH (90 mL). The resulting yellow product is filtered off, washed with water (2 x 20 mL), and dried in vacuo. Recrystallization is done using hot ethyl acetate. 2.1.2 l,l'-Bis(diphenylphosphino)ethane (dpmMe) The dpmMe ligand was readily available in this laboratory. Synthesis can be done according to a published method.1 A solution of MeLi (18.4 mL, 28.8 mmol) is added dropwise to PI12PH 1 3 (5.35 g, 28.7 mmol) dissolved in benzene (50 mL). The resulting yellow solution is stirred for 3 h before 1,1-dichloroethane (1.41 g, 14.3 mmol) is added. The mixture is stirred for another 3 h prior to filtration through Celite. The filtrate is concentrated under vacuum to give a yellow oil; EtOH (300 mL) is added to yield white crystals that are purified by dissolution in 30 mL of CH2C12 followed by reprecipitation via addition of 100 mL of EtOH. 50 References on page 65 Chapter 2 2.1.3 Triphenylphosphine sulfide (Ph3PS) This compound14 was synthesized as a precursor to dpm(S). PPh3 (40.0 g, 0.15 mol) and sulfur (5 g, 0.16 mol) were suspended in toluene (70 mL), and the mixture was refluxed for 2 h. The resulting yellow clear solution was filtered through a column (2.5 x 8.0 cm) of neutral dumina to remove the excess sulfur. CH2C12 (30 mL) was used to elute any product remaining in the column. The colourless filtrate was rotovaped to remove the solvents and the white crystalline product subsequently appeared. Yield: 43.9 g (98%). 2.1.4 Bis(diphenylphosphino)methane monosulfide (dpm(S)) This compound was synthesized according to a published method.2 To a stirred suspension of Ph3PS (20.6 g, 0.0700 mol) in a mixture of THF (110 mL) and Et20 (85 mL) was added MeLi (50 mL, 1.40 M in Et20) over a 30 min period. The resulting red solution was stirred for 1 h before a solution of Ph 2 PCl 1 5 (15.4 g, 0.0700 mol) in Et20 (70 mL) was slowly added over a 2 h period. The colour slowly changed to yellow with accompanying precipitation of LiCl. After the solution was stirred overnight, the solvents were removed by rotary evaporation. The resulting red-yellow oil was dissolved in CH2C12 (100 mL), and the solution was washed with distilled H 20 (3 x 100 mL) and dried over M g S C u . The CH2C12 was rotovaped off and the resulting viscous yellow oil was dissolved in minimal hot ethanol. The solution was cooled and crystallization of the white product occurred. Yield: 17.1 g (59 %). ! H NMR (20 ° C , CDC13): 5 7.0 - 8.0 m (20H, Ph), 5 3.36 dd (2H, C H 2 , J H H = 12.6 Hz, J P H = 1.1 Hz). "Pf^H} NMR (20 ° C , CDC13): 5 -28.1, 40.1 51 References on page 65 Chapter 2 AB pattern (JPP = 75.9 Hz). Anal. Calcd for C 25H2 2P 2S: C, 72.10; H, 5.32. Found: C, 71.89; H, 5.28. 2.1.5 7W///v-dichloiobis(benzonitrile)palladiuni(II) (P(l(PhCN):CI2) This compound was synthesized according to a published method.4"6 PdCl 2 (2.0 g, 11.3 mmol) was suspended in benzonitrile (50 mL) and the mixture heated to and kept at 100 °C for 8 h. The greater part of PdCl 2 dissolved to give a red solution which was filtered while still warm. The filtrate was poured into hexanes (300 mL) and a light yellow precipitate immediately formed. The yellow product was filtered off, washed with hexanes and dried in vacuo. Yield: 4.0 g (92%). 2.1.6 Tris(dibenzylideneacetone)dipalladium(0)-chIoroform solvate (Pd2(dba)3 'CHCl 3) This compound was synthesized according to a published method.4'7 PdCl 2 (1.05 g, 5.9 mmol) was added to a hot (~ 50 °C) MeOH solution (150 mL) containing dba (4.6 g, 19.6 mmol) and NaOAc (3.9 g, 47.5 mmol). The mixture was stirred at 40 °C to give a red-purple precipitate. After being cooled to complete precipitation, the mixture was filtered, and the solid was then washed successively with H 20 and acetone and dried in vacuo. The precipitate was dissolved in hot CHCI3 (300 mL), and the mixture was filtered to give a deep violet solution. Slow addition of Et20 (400 mL) afforded deep purple needles which were filtered off, washed with Et20, and dried in vacuo. Yield: 2.2 g (75%). The crystal structure of this compound has been reported.7 52 References on page 65 Chapter 2 2.1.7 Dichlorobis-p-[bis(diphenylphosphino)methanejdipalladium(I) (Pd2Cl2(dpm)2) This compound was synthesized according to a published method.4'8 Pd(PhCN)2Cl2 (0.41 g, 1.01 mmol), Pd2(dba)3.CHCl3 (0.55 g, 0.53 mmol), and dpm (0.82 g, 2.1 mmol) were dissolved in CH2CI2 (50 mL). The mixture was refluxed for 30 min and the resulting red solution was then cooled to R.T. The solution was filtered to remove any insoluble materials, and the filtrate was concentrated under vacuum to ~10 mL. Et 20 (20 mL) was added and a yellow-orange precipitate formed. The product was then filtered off, washed with acetone (2x10 mL) to remove any palladium(II) monomer, and dried in vacuo. Yield: 1.0 g (90 %). ! H NMR (20 °C, CDC13): 6 7.0 - 8.0 m (40H, Ph), 5 4.17 qn (4H, CH 2 , J P H = 4 Hz). ^Pf/H} NMR (20 °C, CDC13): 6 -5.5 s. 2.1.8 Dibromobis-p-|bis(diphenylphosphino)methaneJdipalladium(I) (Pd2Br2(dpm)2) This compound was synthesized according to a published method.4'9 Pd2Cl2(dpm)2 (0.23 g, 0.22 mmol) was dissolved in CH2C12 (10 mL), and a solution of NaBr (0.20 g, 2 mmol) in aq. MeOH (10 mL MeOH : 1 mL H 20) was added. The resulting solution was stirred for 1 h before being filtered and concentrated under vacuum until an orange precipitate was formed. Aqueous MeOH was added to complete the precipitation and the product was filtered off. The product was recrystallized using CH2Cl2/aq. MeOH and was dried in vacuo. Yield: 0.24 g (97 %). ! H NMR (20 °C, CDC13): 5 7.0 - 8.0 m (40H, Ph), 6 4.26 qn (4H, CH 2 , J P H = 4 Hz). ^Pj'H} NMR (20 °C, CDC13): 6-6.15 s. The crystal structure of this compound has been reported.16 53 References on page 65 Chapter 2 2.1.9 Diiodobis-p-[bis(diphenylphosphino)methane]dipalladium(I) dichloromethane solvate (Pd2I2(dpm)2.CH2Cl2) This compound was synthesized according to a published method.4'9 Pd2Cl2(dpm)2 (0.23 g, 0.22 mmol) was dissolved in CH2C12 (10 mL) and a solution of Nal (0.15 g, 1 mmol) in aq. MeOH (10 mL) was added. The resulting solution was stirred for 1 h before being filtered and concentrated under vacuum until a purple precipitate was formed. Aqueous MeOH was added to complete the precipitation; the product was filtered off, washed with acetone (2x10 mL), and dried in vacuo. Yield: 0.26 g (90 %). } H NMR (20 °C, CDC13): 6 7.0 - 8.0 m (40H, Ph), 5 4.23 qn (4H, C H 2 , J P H = 4 Hz). 31P{'H} NMR (20 °C, CDC13): 8 -11.3 s. 2.1.10 Dichloro-p-sulfidobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2Cl2(u-S)(dpm)2) This compound was synthesized according to a published method.4'10 Pd2Cl2(dpm)2 (0.50 g, 0.48 mmol) was dissolved in CH2C12 (50 mL) and H 2S gas was bubbled through the solution for 20 min at R.T. The colour changed from orange-red to brown with accompanying precipitation of a brown solid. Et 20 (50 mL) was slowly added to complete the precipitation; the product was filtered off, washed successively with acetone (2x10 mL) and Et 20 (10 mL), and dried in vacuo. Yield: 0.50 g (97 %). ! H NMR (20 °C, CDCI3): 5 7.0 - 8.0 m (40H, Ph), 8 4.70 d qn (2H, CH, J H H = 13 Hz, J P H = 6 Hz), 8 2.79 d qn (2H, CH, J H H = 13 Hz, J P H = 4 Hz). 31P{1H} NMR (20 °C, CDC13): 8 5.5 s. The crystal structure of this compound has been reported.17 54 References on page 65 Chapter 2 2.1.11 Dibromo-p-sulfidobis-p-[bis(diphenylphosphino)rnethane]dipalladium(II) (Pd 2Br 2(p-S)(dpm) 2) This compound was synthesized according to a published method.4'10 Pd2Br2(dpm)2 (0.50 g, 0.44 mmol) was dissolved in CH2C12 (50 mL) placed in a Schlenk tube stoppered with a rubber septum. H 2S (50 mL, 1 arm at 25 °C) was injected, and the reaction mixture was allowed to react for 3 h. The colour changed from orange-red to brown with accompanying precipitation of a brown solid. Et20 (50 mL) was added to complete the precipitation; the solid was filtered off, washed successively with acetone (2x10 mL) and Et20 (10 mL), and dried in vacuo. Yield: 0.51 g (98 %). ' H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (40H, Ph), 5 4.83 d qn (2H, CH, J H H = 12.8 Hz, J P H = 7.6 Hz), 5 2.88 d qn (2H, CH, J H H = 12.8 Hz, J P H = 3.2 Hz). 31P{ !H} NMR (20 °C, CDC13): 8 5.96 s. 2.1.12 Diiodo-p-sulfidobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2I2(u-S)(dpm)2) This compound was synthesized according to a published method.4'10 Pd2Cl2(p-S)(dpm)2 (0.25 g, 0.23 mmol) was dissolved in CH2C12 (25 mL) and a solution of Nal (0.30 g, 2 mmol) in aq. MeOH (10 mL) was added. The mixture was stirred for 1 h before being concentrated under vacuum to -10 mL. Precipitation of a brown solid was completed by addition of MeOH (50 mL). The product was filtered off, washed successively with aq. MeOH (2x10 mL) and Et20 (10 mL), and dried in vacuo. Yield: 0.32 g (90%). [ H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (40H, Ph), 8 55 References on page 65 Chapter 2 4.95 d qn (2H, CH, JHH = 14 Hz, J P H = 6 Hz), 5 3.06 d qn (2H, CH, JHH = 14 Hz, J P H = 3 Hz). ^PjTT} NMR (20 °C, CDC13): 5 6.1 s. 2.1.13 Dichloro|bis(diphenylphosphino)methane]palladium(II) (PdCl2(dpm)) This compound was synthesized according to a published method.4'11'12 A solution of Pd(PhCN)2Cl2 (0.30 g, 0.77 mmol) in CH2C12 (10 mL) was added with stirring to a CH 2C1 2 (10 mL) solution containing dpm (0.30 g, 0.78 mmol). The reaction mixture was stirred for 2 h and concentrated under vacuum to -10 mL before Et20 (50 mL) was added to the yellow solution affording a pale yellow solid that was filtered off and reprecipitated twice from CH2C12/Et20. The product was finally washed with Et20 (2x10 mL) and dried in vacuo. Yield: 0.39 g (90%). ! H NMR (20 °C, CDCI3): 6 7.0 - 8.0 m (20H, Ph), 8 4.211 (2H, CH 2 , J P H = 10.8 Hz). 31P{'H} NMR (20 °C, CDCI3): 6 -54.7 s. The crystal structure of this compound has been reported.12 2.1.14 Dibromo[bis(diphenylphosphino)methane]palladium(II) (PdBr2(dpm)) This compound was synthesized according to a published method.4'11 PdCl2(dpm) (0.25 g, 0.45 mmol) was dissolved in CH2C12 (10 mL) and a solution of NaBr (0.20 g, 2 mmol) in aq. MeOH (10 mL) was added. The mixture was stirred for 1 h before being filtered and concentrated under vacuum to - 10 mL. Et20 (25 mL) was added to complete the precipitation of a yellow solid. The product was filtered off, washed successively with aq. MeOH (2x10 mL) and Et20 (10 mL), and dried in vacuo. Yield: 0.26 g (90%). 1 HNMR(20 °C, CDC13): 5 7.0 - 8.0 m (20H, 56 References on page 65 Chapter 2 Ph), 5 4.371 (2H, C H 2 , J P H = 10.5 Hz). ^Pf/H} NMR (20 °C, CDC13): 8 -56.2 s. The crystal structure of this compound has been reported.18 2.1.15 Diiodo[bis(diphenylphoshino)methane]palladium(II) (Pdl2(dpm)) PdCl2(dpm) (0.10 g, 0.18 mmol) was placed in CH2C12 (20 mL) and a solution of Nal (0.40 g, 2.7 mmol) in aq. MeOH (10 mL) was added. The resulting orange solution was stirred for 1 h before being reduced in volume under vacuum to -10 mL. Et20 (30 mL) was added to precipitate an orange solid that was filtered off, washed successively with aq. MeOH (2x10 mL) and Et20 (10 mL), and dried in vacuo. Yield: 0.13 g (97%). ! H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (20H, Ph), 8 4.42 t (2H, CH 2 , J P H = 10.0 Hz). 31P{'H} NMR (20 °C, CDC13): 8 -63.2 s. UV/vis (20 °C, CHC13): X, nm (e, M 1 cm"1): 430 (6455). Anal. Calcd for C 2 5H 2 2I 2P 2Pd: C, 40.33; H, 2.98. Found: C, 40.34; H, 2.97. The crystal structure of this compound has been reported.19 2.1.16 Dichloro-p-sulfoxobis-p-[bis(diphenylphosphino)methane]dipalladium(II) (Pd2Cl2(p-SO)(dpm)2) This compound was synthesized according to a published method.10 Pd2Cl2(u-S)(dpm)2 (0.30 g, 0.28 mmol) was dissolved in CH2C12 (40 mL) and a methanolic solution of H 2 0 2 (40 mL 30% H 20 2 in 10 mL MeOH) was added. The mixture was stirred at R T . for 2 h, after which the orange CH 2C1 2 layer was removed and concentrated under vacuum to 10 mL. Addition of 50 mL of MeOH precipitated an orange solid that was filtered off, washed with MeOH (2x10 mL), and dried in vacuo. Yield: 0.28 g (90%). •H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (40H, Ph), 8 4.78 d t 57 References on page 65 Chapter 2 (1H, CFJ.2, J H H = 13.5 Hz, J P H = 10.5 Hz), 8 3.89 d t (1H, C H 2 , J H H = 13.5 Hz, J P H = 7.5 Hz), 8 2.53 m (1H, CH 2), 8 2.27 m (1H, CH2). ^Pj'H} NMR (20 °C, CDC13): 8 20 to -5 A A ' B B ' pattern. The crystal structure of this compound has been reported.10 2.1.17 Diiodo-p-sulfoxobis-p-[bis(diphenylphosphino)methaneldipalladium(II) (Pd2I2(p-SO)(dpm)2) Pd2I2(dpm)2(u.-S) (0.20 g, 0.16 mmol) was dissolved in CH2C12 (10 mL) and a methanolic solution of H 2 0 2 (30%, 1 mL in 5 mL MeOH) was added. The colour immediately changed from brown to red-brown. The solution was stirred for 1 h before MeOH (25 mL) was added to precipitate a red-brown solid that was filtered off, washed with MeOH (2x10 mL), and dried in vacuo. Yield: 0.16 g (80%). ! H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (40H, Ph), 8 4.98 d t (1H, C H 2 , J H H = 12 Hz, J P H = 13.5 Hz), 8 4.18 d t (1H, CH 2 , J H H = 15 Hz, J P H = 9 Hz), 8 2.58 m (1H, CH 2), 8 2.26 m (1H, CH2). NMR (20 °C, CDC13): 8 19 to -7 A A ' B B ' pattern. A satisfactory elemental analysis could not be obtained. 2.1.18 Dimercapto-p-sulfidobis-p-[bis(diphenylphosphino)methane|dipalladium(II) (Pd2(SH)2(p-S)(dpm)2) This compound was synthesized from a procedure slightly modified from that reported.10 PdCl2(dpm) (0.31 g, 0.55 mmol) was suspended in CH2C12 (10 mL) and a solution of NaSH (0.62 g, 11 mmol) in aq. MeOH (5 mL H 2 0 : 5 mL MeOH) was added dropwise. The solution became clear within minutes with accompanying colour changed from yellow to red-brown. The solution 58 References on page 65 Chapter 2 was then washed with distilled H 20 (to extract the excess NaSH) until the aqueous layer was at neutral pH. The organic layer was reduced in volume under vacuum to ~5 mL before Et20 (30 mL) was added to precipitate a brown solid that was filtered off, washed successively with MeOH (3 x 10 mL) and Et20 (2x10 mL), and dried in vacuo. Yield: 0.14 g (47 %). lH NMR (20 °C, CDC13): 5 7.0 - 8.0 m (40H, Ph), 8 4.71 d qn (2H, C H 2 , JHH = 13.5 Hz, J P H = 5.7 Hz), 8 3.03 d qn (2H, C H 2 , JHH = 13.2 Hz, J P H = 3.0 Hz), 8 -1.54 qn (2H, SH, J P H = 5.7 Hz). 31P{'H} NMR (20 °C, CDCI3): 8 14.8 s. 2.2 Instrumentation 2.2.1 Nuclear magnetic resonance spectroscopy NMR spectra were recorded on a Bruker AC200E or Varian X L 300 NMR spectrometer using the residual protonated species in the deuterated solvents as internal references for *H NMR (7.24 ppm for CDCI3, 7.18 ppm for C 6 D 6 , 5.32 ppm for CD2C12, 2.49 ppm for DMSO-de, and 1.93 ppm for CD3CN; all shifts are relative to TMS) or 85% H3PO4 as an external reference for 31P{1H} NMR with the downfield shifts being positive. Samples were prepared in 5 mm NMR tubes fitted with poly(propylene) caps, rubber septa, or poly(tetrafluoroethylene) J . Young valves (Aldrich). Unless otherwise stated, all samples were prepared under 1 atm of N 2 . Low temperature NMR studies were performed using the Varian X L 300 NMR spectrometer. 59 References on page 65 Chapter 2 2.2.2 Electronic absorption spectroscopy UV-vis spectroscopic measurements were performed using 1 cm quartz cells on a Hewlett Packard 8452A diode-array or a Perkin-Elmer 552A spectrophotometer both equipped with a temperature regulating unit. The accompanying UV-vis quantification software (HP 89532A, HP 89532K, and HP 89532Q) for the diode-array instrument was used for both general spectral acquisition and kinetic experiments (see Section 2.3 for treatment of kinetic data). 2.2.3 Stopped-flow electronic absorption spectroscopy Kinetically fast reactions were monitored by stopped-flow measurements using a path length of 1 cm on the Applied Photophysics Stopped-Flow spectrometer (Model SF. 17MV) equipped with a 150 W Xenon arc lamp, temperature regulating unit, and the SF.17MV kinetic software. The temperature regulating unit houses the dual syringe sample inlet system driven by a pneumatic pump. In a typical kinetic setup, the two syringes were filled with the appropriate solutions and thermostated at the appropriate temperature. Data were automatically collected by the computer after manually initiating the pump (see Section 2.3 for treatment of kinetic data). 2.2.4 Thin-layered chromatography Analytical thin-layered chromatography was carried out using Merck silica gel on plastic sheets (Mesh 60, layer thickness 0.2 mm) with F254 fluorescent indicator. Species identification was done by comparison of degrees of retention with those of authentic samples. Qualitative 60 References on page 65 Chapter 2 assessment of the status of a reaction, in general, was made by noting the disappearance of a reactant (usually a phosphine species). 2.2.5 Gas chromatography Gas chromatographic analyses were performed on a temperature-programmable Hewlett Packard 5 890A instrument equipped with a thermal conductivity detector (TCD). Retention times were established using authentic samples. A 10 ft packed molecular sieve column was used for qualitative analysis of H 2 with He as the carrier gas (see Chapter 3). Instrument settings were as follows: oven temp. = 75 °C; injector temp. = 90 °C; detector temp. = 200 °C; and column head pressure = 40 kPa. Retention time: tR(H2) =1.65 min (inverted peak20). 2.2.6 Photochemistry Photochemical experiments were carried out at R.T. using either an 18 W TLC U V lamp (Mineralight®, model UVG-11) or a 100 W medium pressure Hg vapour lamp (Ace Glass Inc.). Samples were usually prepared in either a septum-sealed NMR tube or an NMR tube fitted with a PTFE J. Young valve. 61 References on page 65 Chapter 2 2.2.1 Electron spin resonance spectroscopy ESR spectra were obtained at X-band frequencies using a Bruker ECS-106 spectrometer. Samples were prepared in the frozen-state at liquid N2 temperatures (-196 °C) in 4 mm ESR tubes (see Chapter 4). Analyses were performed by Dr. F.G. Herring of this department. 2.2.8 Elemental, mass, and X-ray crystallographic and photoelectron spectroscopic analyses Analyses of these types were performed by the technical staff at this department (elemental analyses by Mr. P. Borda, mass spectroscopy by Dr. G. Eigendorf and his staff, and X-ray crystallography by Dr. S.J. Rettig). XPS (or ESCA) was carried out by Dr. P.C. Wong in Dr. K.A.R. Mitchell's surface science laboratories. 2.3 Treatment of kinetic data Throughout the course of this thesis work, the kinetics of several reactions were monitored spectrophotometrically, and thus it seems appropriate to discuss briefly the treatment of kinetic data. Either by conventional or stopped-flow spectrophotometry, the solution electronic absorption of a system was measured as a function of time under pseudo-first-order conditions. From these absorbances (A) vs. time (t) measurements, semilog plots of lnfAc -At) vs. t were obtained (where Ac and A t are the absorbances at t = 0 0 and t, respectively) from which the observed rate constants, kobs, were evaluated from the slopes. The justification of such plots is presented below. 62 References on page 65 Chapter 2 In the reactions studied kinetically, the systems are well-behaved (i.e.there is in every case isosbesticity in spectrophotometric measurements and the absence of observable intermediates as determined by variable temperature NMR measurements) and are characteristic of one absorbing species going to another, e.g. X —» Y. Firstly, under pseudo-first-order conditions (i.e. with the non-absorbing reactant species in excess), the disappearance of X follows a first-order rate decay as expressed by eq. 2.3.1, where [X]t and [X]0 are the concentrations of X at time = t and 0, respectively. (2.3.1) [X]t=[X]0exp(-kobst) The rate of Y appearing is as described in eq. 2.3.2. (2.3.2) [Y]t = [X]0 - [X]0exp(-kobst) The absorbances according to the Beer-Lambert law at time = t for X and Y are given by (2.3.3) A t x = s x b [ X ] t and (2.3.4) A t Y = s Y b[Y], (where b is the path length and e x and sY are the molar absorptivities of X and Y, respectively), and the total absorbance of the system is given by (2.3.5) A t = A t x + A t Y = sxb[X]t + s Y b[Y] t From eq. 2.3.2, [Y]t = [X]0 - [X]t, and hence, (2.3.6) A t = ex b [X]t + eY b ([X]0 - [X],) When the reaction is complete (i.e. at t = oo), the absorbance of the system is given by (2.3.7) A o o = A M Y = 8 Y b [ Y ] t o = 8 Y b [ X ] 0 Subtraction of eq. 2.3.6 from eq. 2.3.7 yields (2.3.8) A o o - A t = sY b [X]0 - e x b [X]t - E y b ([X]0 - [X]t) = (eY b - s x b )[X]t 63 References on page 65 Chapter 2 Eq. 2.3.8 states that - A t is proportional to [X] at time = t which justifies the plotting of ln(A<„ -A t) vs. t for a first-order analysis. Furthermore, substitution of eq. 2.3.8 into eq. 2.3.1 gives (2.3.9) (A^ - At)/(sy b - e x b) = (AJex b)exp(-kobst) where A,, is the initial absorbance of the system (i.e. that of X at [X]0). Expressed in natural logarithms, eq. 2.3.9 upon rearrangement becomes (2.3.10) ln (Aoo - At) = ln[(eY - ex)A0/sx] - kobs t Analyses of the various kinetic data showed excellent correlation between the observed y-intercept and ln[(sy - sx)AJex]- The kinetic quantification software function-fits the A vs. t data into the general equation (2.3.11) A, = K 1exp(-K 2t) + K 3 where Ki = A„ - A*, (Ao is the initial absorbance at t = 0), K 2 = kobs, and K 3 = A*,. It is noted here that both semilog plots and function-fitting were used to obtain observed rate constant values. The rate constants for the reaction, k, were then evaluated from the kobs values either directly, knowing the concentrations of the non-absorbing species, or from the slopes of kobs vs. [non-absorbing species] plots as (2.3.12) kobs = k [non-absorbing species] The k values were obtained at other temperatures, and activation parameters, AH* and AS*, were obtained from usual Eyring plots (In k/T vs. 1/T) as (2.3.13) k = kBT/h exp(AS*/R) exp(-AHTRT) where k& is the Boltzmann's constant (1.38066 x 10"23 J K"1), h = Planck's constant (6.62618 x 10"34 J s), and R = gas constant (8.31442 J K"1 mol"1). 64 References on page 65 Chapter 2 2.4 References for Chapter 2 1. Lee, C-L . ; Yang, Y-P.; Rettig, S.J.; James, B.R.; Nelson, D . A . ; Lilga, M . A . Organometallics 1986, 5, 2220. 2. (a) Grim, S.O.; Mitchell, J.D. Syn. React. Inorg. Metal-Org. Chem. 1974, 4, 221. (b) Grim, S O . ; Walton, E .D. Inorg. Chem. 1980, 19, 1982. 3. Conard, C.R.; Dolliver, M . A . Org. Syn., Coll. Vol. 2, John Wiley, Toronto, p. 167. 4. Barnabas, A.F . M.Sc. Dissertation, University of British Columbia, Vancouver, 1989. 5. Kharasch, M.S. ; Seyler, R C ; Mayo, F.R. J. Am. Chem. Soc. 1941, 63, 2088. 6. Doyle, J.R.; Slade, P.E.; Jonassen, H.B. Inorg. Synth. 1960, 6, 216. 7. Ukai, T . ; Kawazura, H . ; Ishii, Y. ; Bonnet, J.; Ibers, J.A. J. Organomet. Chem. 1974, 65, 253. 8. Balch, A . L . ; Benner, L.S. Inorg. Synth. 1982, 21, 47. 9. Benner, L.S. ; Balch, A . L . J. Am. Chem. Soc. 1978, 700, 6099. 10. Besenyei, G . ; Lee, C-L . ; Gulinski, J.; Rettig, S.J.; James, B.R.; Nelson, D . A . ; Lilga, M . A . Inorg. Chem. 1987, 26, 3622. 11. Jenkins, J .M. ; Verkade, J.G. Inorg. Synth. 1968, 77, 108. 12. Steffen, W.L. ; Palenik, G.J Inorg. Chem. 1976, 75, 2432. 13. Bianco, V . D . ; Doronzo, S. Inorg. Synth. 1976, 16, 161. 14. Olah, G.A. ; Berrier, A . ; Ohannesian, L. Nouv. J. Chim. 1986, 70, 253. 15. Hideyuki, I.; Yukitaka, U . ; Masayuki, U. JP Patent 03 52,894 [91 52,894] (1989); Chem. Abstr. 115, 49982, 1991. 16. Holloway, R .G. ; Penfold, B.R. J. Chem. Soc, Chem. Commun. 1976, 485. 17. Balch, A . L . in 'Catalytic Aspects of Metal Phosphine Complexes,' E .C. Alyea and D.W. Meek, eds., A C S Advances in Chemistry Series, 1982, v. 196, p. 243. 18. Peters, K . ; Peters, E . - M . ; Von Schnering, H . G . ; Abicht, H.-P.Z. Z. Kristallogr. 1984,168, 149. 19. Davies, J.A.; Pinkerton, A . A . ; Syed, R.; Vilmer, M . J. Chem. Soc, Chem. Commun. 1988, 47. 65 References on page 65 Chapter 2 Only H 2 has thermal conductivity greater than He. See HP 5890A Gas Chromatograph Reference Manual, Hewlett-Packard Company, 2086, vol. I, p. 11-39. Depending on the concentration, the H 2 peak may appear as positive (low cone), negative (high cone), or a split peak (intermediate cone.) when using He as the carrier gas. 66 References on page 65 CHAPTER 3 Kinetic and Mechanistic Aspects of Sulfur Abstraction from Pd2X2(p-S)(dpm)2 Using dpm or dpmMe and Catalytic Conversion of H 2S to H 2 . 67 Chapter 3 3.1 Introduction The discovery in this laboratory of the stoichiometric abstraction of sulfur from H 2 S using Pd2X2(dpm)2 complexes [X = CI (la), Br (lb), I (lc)] (reaction 3.1.1)1 had prompted subsequent kinetic and mechanistic studies on this reaction.2'3 In the course of these investigations, deuterium labelling and low temperature NMR spectroscopic experiments demonstrated that H 2S oxidatively adds across the metal-metal bond to generate hydrido mercapto intermediates. Product formation occurs via deprotonation of the coordinated SH and protonation of the coordinated hydride to give the bridged sulfide complex Pd2X2(|x-S)(dpm)2 [X = CI (2a), Br (2b), I (2c)] with concomitant H 2 evolution. Attempts to unravel the role of the bridging diphosphine, for example by adding excess dpm, had been thwarted by the unexpected quantitative abstraction of the bridging S atom (reaction 3.1.2), where dpm(S) is the monosulfide.2 (3.1.1) Pd2X2(dpm)2 (1) + H 2S Pd2X2(u-S)(dpm)2 (2) + H 2 (3.1.2) Pd2X2(u-S)(dpm)2 (2) + dpm Pd2X2(dpm)2 (1) + dpm(S) It was then of interest to examine closely reaction 3.1.2 via kinetic and mechanistic studies and these are reported in this chapter. Reactions 3.1.1 and 3.1.2 imply a catalytic cycle, the net reaction 3.1.3, and some experiments were also carried out to substantiate the catalysis. (3.1.3) H 2S + dpm H 2 + dpm(S) Reaction 3.1.3 appears to be the first reported, homogeneously catalyzed conversion of H 2S to H 2 . 4 Some rationale for the previously studied, unsuccessful sulfur abstraction from 2 by other phosphines (i.e. PPh3, PPh2Me, Ph2P(CH2)3PPh2)5 is also presented. 68 References on page 133 Chapter 3 3.2 Results Reaction 3.1.2 and its stoichiometry were readily demonstrated by NMR studies (Table 3.1). Complexes 2a - 2c reveal a singlet at 5 5.5 - 6.1 in the 31P{1H} NMR spectrum at ambient conditions in CH2CI2 for the four equivalent P atoms,3 while the corresponding singlets for la - lc are in the 8 -5.5 to -11.3 region.1'2 The 31P{1H} NMR spectrum of the monosulfide dpm(S) consists of an AB doublet (8 -28.1, 40.1, J P P 75.9 Hz), the lower field signal being that of the P(V) center.6 In the lH NMR spectra of 2a - 2c, the C H 2 resonances appear as AB doublets with additional coupling to the four P atoms, while in la - lc the equivalent C H 2 protons show a characteristic quintet pattern. Table 3.1 summarizes the NMR data measured in the present work for all the Pd complexes and diphosphine compounds; kinetic data can be found in Appendix n. The rates of the reconversion reaction 2 -> 1 (eq. 3.1.2) were noticed qualitatively during preliminary kinetic and mechanistic studies, and were found to be dependent on the nature of the auxiliary ligand X and on the concentration of dpm used. The reactivity trend in CHC13 observed was X = Cl > Br > I, and, of these, the bromide system was chosen for a detailed, classical spectrophotometric study. Only limited kinetic data for the chloride system and qualitative observations for the iodide system were obtained because of complexities based on the 2 3 7 photosensitivity of la and lc, respectively. ' ' The kinetics of the reconversion reaction (2 -> 1) were studied using solution electronic spectroscopy. Species 2 are brown and have absorption maxima in the 325 - 350 and 470 - 485 nm regions, while the products (1) are reddish-orange (Cl and Br), or purple (I), and have two or three absorption maxima in the 340 - 590 nm region.2'3 Figure 3.1 shows spectral changes on treating a CHCI3 solution of the bromide 2b with dpm at 25 °C; rates for the bromide system are 69 References on page 133 Chapter 3 Table 3.1. NMR Data for the Dinuclear Palladium Complexes, Diphosphines, and the Diphosphine Monosulfides. Compound ° 8(1H)b 5 (31P) c Pd 2Cl 2(dpm) 2 (la) 4.17^(4.0) -5.5 Pd 2Br 2(dpm) 2 (lb) 4.19 d (4.0) -6.15 4.24" -5.5 e Pd 2I 2(dpm) 2 (lc) 4.23 d (4.0) -11.3 Pd2Cl2(n-S)(dpm)2 (2a) 2.79 /(12.6, 3.5) 4.73 (12.6, 6.1) 5.52 Pd2Br2(n-S)(dpm)2 (2b) 2.88 /(12.8, 3.2) 5.96 (2.90 e) 6.14 ' 4.83 (12.8, 7.6) Pd 2I 2(^-S)(dpm) 2 (2c) 3.06 /(14.0, 3.0) 4.95 (14.0, 6.0) 6.08 Pd2Br2(dpm)(dpm-d2) (3b) 4.19* -6.30 * 4.24' -5.7 e Pd 2Br 2(dpm-d 2) 2 (4b) /' -6.35 Pd2Br2(dpm)(dpmMe) (5b) 4.98• / >, 4.55*, 3.72* 1.00* (Me) / Pd 2Br 2(dpmMe) 2 (6b) 4.98 m e 1 . 0 0 ( M e ) 16.1 e Pd2Br2(n-S)(dpm)(dpm-d2) (7b) 2.88 4.83 g 5.95 K e Pd2Br2(n-S)(dpm-d2)2 (8b) /' 6.00' Pd 2I 4(dpm) 2 5 .14 d e (3.6) -1.52 * dpm 2.81 °'e (1.6) -22.5 ' dpm-d 2 / -23.0 ' dpm(S) 3.36 v (12.6, 1.1) 3.35 e -28.1,40.1 (75.9)* dpm(S)-d2 i -28.3, 39.9 (76.4) * dpmMe 3.20 r (7.2) 0.99 * (Me) (7.5, 10.2) -6.94 dpmMe(S) 3.64 s ' e 1.16 * e (Me) (7.4, 10.2) -13,4,51.9(94.3)* 70 References on page 133 Chapter 3 a The n-symbol for the bridging diphosphine ligand(s) is omitted for convenience throughout this Table and the text. * In C D 2 C 1 2 , unless stated otherwise, at 20 °C with respect to T M S ; J H H and/or J p H values in Hertz are given in parentheses; signals for C H 2 protons unless indicated otherwise. c Singlets in C D 2 C 1 2 at 20 °C with respect to 85% H 3 P 0 4 , downfield being positive. d Quintet. e I n C D C l 3 . ^Doublets of quintets for each of 2 sets of C H 2 protons. 8 Assumed triplet. A ' T i g h t ' A B quartet. ''No C H 2 protons observed. J'Triplet of quartets. * Doublet of triplets. ' A A ' B B ' pattern (see Fig. 3.13). m Quartet of quintets. " Doublet of quintets. ° Triplet, P Doublet of doublets. 9 A B pattern, J P P values in Hertz given in parentheses. r Quartet. s Multiplet. 71 References on page 133 Chapter 3 Fig. 3.1. a) Visible absorption spectral changes (350 - 550 nm region) as a function of time for a CHCI3 solution of Pd2Br2(p-S)(p-dpm)2 (2b) (6.52 x 10"5 M) on addition of dpm (1.96 x 10"2 M) at 25 ° C . b) A rate-plot analyzed for a pseudo-first-order dependence on 2b ; A T and A Q Q represent the absorption at 428 nm at times t and 0 0 , respectively. a wavelength (nm) 1000 t (sec) 2000 3000 4 0 0 0 72 References on page 133 Chapter 3 followed by observing the increasing absorption at 428 nm as a function of time, and the observed isosbestic point shows a well-behaved system. The pseudo-first-order rate constants, kobs, obtained from the excellently linear semi-log plots (Fig. 3.1b), are strictly first-order in [dpm] (Fig. 3.2) and are independent of the [2b] from (0.81 -13.0) x 10"5 M. Thus, the rate law takes the simple form Rate = -^-[2b] = kobs[2b] = kBr[dpm][2b] (3.2.1) dt where kB r is the bimolecular rate constant for the reaction. The temperature dependence data for ksr (Table 3.2) are obtained similarly from the first-order plots at other temperatures (Fig. 3.2a); an Eyring plot of the data from 20 - 35 °C gives an excellent straight line and the activation parameters AFf = 38 ± 1 kJ mol"1 and AS* = -144 ± 4 J K"1 mol"1 (Fig. 3.2b). The corresponding studies on the faster chloride system were performed, the increasing absorbance being monitored at 416 nm where spectral changes are greatest. Because of the photosensitivity of la, specifically at the later stages of the reaction where isosbesticity is lost, more limited kinetic data were obtained. The reconversion reaction 2a —> la was monitored when the solution was kept in the dark, with the optical density being recorded by a sampling rather than a continuous monitoring method. Analysis of the earlier visible spectral changes (~ 2 half-lives), using an A^ value from the known absorption spectrum of la, also gives results that are strictly first-order in [dpm] (Fig. 3.3a). The pseudo-first-order rate constant is also found to be independent of the [2a] from (3.26 - 13.0) x 10"5 M. Thus, the rate law for the chloride system also takes the same form as eq. 3.2.1, but with a bimolecular rate constant kci. The temperature dependence data for kCi were obtained at a single [dpm] of 1.96 x 10"2 M , and with [2a] = 6.52 x 73 References on page 133 Chapter 3 Fig. 3.2. a) The dependence of the pseudo-first-order rate constants, kobs, on [dpm] at [2b] = 6.52 x 10"5 M , in CHCI3. b) Eyring plot of the temperature dependence of the bimolecular rate constant, k B r . 74 References on page 133 Chapter 3 Table 3.2. Temperature dependence for the bimolecular rate constant of the reaction of Pd2Cl2(u-S)(dpm)2 (2a) (ka) and Pd2Br2(u-S)(dpm)2 (2b) (kBr) with dpm in CHC13. Temp, K ka, M-V 1 kB r, M V 1 293 0.0667 A 0.0229 298 0.0898 0.0308 303 0.125 A 0.0396 308 0.157" 0.0524 a Obtained at a single [dpm] = 1.96 x 10 2 M . 75 References on page 133 Chapter 3 Fig. 3.3. a) The dependence of the pseudo-first-order rate constants, kobs, pn [dpm] at [2a] = 6.52 x 10'5 M , in CHC13 at 30 °C. b) Eyring plot of the temperature dependence of the bimolecular rate constant, kc i . a 0.0035 -, 0 0.005 0.01 0.015 0.02 0.025 0.03 [dpm], M 76 References on page 133 Chapter 3 10"5 M (Table 3.2); an Eyring plot is reasonably linear and gives the activation parameters AH* = 41 ± 3 kJ mol"1 and AS* = -127 ± 10 J K"1 mol"1 (Fig. 3.3b). Only qualitative observations could be made for the iodide system due to its inherent photosensitive nature; the light source (either laboratory light or that of the spectrophotometer) readily induces formation of the monomer Pdl2(dpm) 2 from lc, the product of the reconversion reaction. Under the same conditions used to study the other two systems, the iodide reaction is extremely slow, even at higher temperatures (e.g. 35 °C), but the thermal reaction between 2c and dpm gives the 'expected' products, lc and dpm(S). Again, analysis of the early visible spectral changes, using an A*, value from the known absorption spectrum of lc, gives kobs ~ 7 x 10"5 s"1 at 25 °C for a reaction with [dpm] = 6.53 x 10"3 M corresponding to a bimolecular ki rate constant of ~ 1 x 10"2 M" 1 s"1. The limited data show that the rate constant is about three times lower than k B r and about nine times lower than kci. Low temperature NMR studies were performed on all three systems but no intermediates were observed en route to formation of 1 from 2, even at temperatures as low as -80 °C (in CD2CI2). For systems in CDCI3, the probe temperature was raised in 20° increments, and new signals were first observed around -40 °C in each case and which corresponded to those of the products, 1 and dpm(S). NMR samples in CD2CI2 were made up and sometimes left at -78 °C for 72 h before analyses were done in an attempt to allow sufficient time for formation of intermediates; again intermediates were not seen, although some 1 and dpm(S) were observed for both the chloride and bromide systems. For the iodide system, however, no products were formed, this being consistent with the kinetic studies which show the iodide reaction to be much slower. These NMR studies show that the reconversion reaction is kinetically possible at -80° C (at least for the chloride and bromide systems), when dpm is present in a five-fold excess at ~ 85 mM. 77 References on page 133 Chapter 3 Extrapolation of the k values of Table 3.1 to -78° C gives kCi and kB r values of 9.4 x 10"6 and 5.4 x 10"6 M" 1 s"1, respectively; use of these values at the noted conditions suggests conversions to la and lb of -20 and ~11%, respectively, while the observed values are about 25 and 15%. (Note: although no extensive kinetic studies were performed in CH2C12, a single experiment at R.T. qualitatively shows that kB r of the reaction in CH2C12 is approx. twice that in CHCI3. The use of 20 and 11% values is conservative, and an upper limit would be 40 and 22% conversion to la and lb, respectively.) The kinetics of the reaction 2b —> lb were also studied in the presence of a bromide salt in order to find a possible kinetic role for an ionic species; addition of a 10- or 100-fold excess of tetrapropylammonium bromide to 2b during a reaction with dpm at 25° C produced no change in the observed rate constant. Also, in reactions between 2b and dpm-d2, and between Pd2J3r2(p-S)(dpm-d2)2 (8b) and dpm or dpm-d2, no kinetic isotope effect was seen in the observed rate constant at 30 °C. During the reaction of 2b with either deuterated dpm (dpm-d2) or methylated dpm (dpmMe), the colour changed from the brown of 2b to reddish, as with dpm itself, and sulfur was again completely abstracted to give diphosphine monosulfides (see below). Complementary in situ NMR experiments (even when the phosphines were added in excess) showed that the products formed were the same as those obtained in the larger-scale synthetic experiments. Although no kinetic measurements were done on the reaction of 2b with dpmMe, in situ NMR experiments indicate that the reaction is some 30 times slower than that of 2b with dpm (or dpm-d2). In the reaction of 2b with dpm-d2 at a 1 : 1 mole ratio, the lH spectrum of the Pd product(s) reveals the expected, less informative multiplet between 8 7.0 and 8.0 due to the phenyl rings of the phosphine ligands; also seen is the characteristic quintet at 8 4.19 due to the dpm C H 2 78 References on page 133 Chapter 3 protons of the product lb (Fig. 3.4). However, the ratio of the integrated areas of the multiplet to the quintet is -15:1, not the 10:1 expected for the 40 phenyl protons and 4 C H 2 protons of lb, and moreover the 31P{1H} NMR spectrum shows not only the singlet of lb at 8 -6.15 but also a 'tight' AB quartet at 8 -6.30 (Fig. 3.5). As discussed below, this quartet is due to the mono-dpm-d2 substituted complex, Pd2Br2(dpm)(dpm-d2) (3b), and this product is formed with lb in the ratio of probably about 1:1. The C H 2 resonance of 3b is assumed to be a triplet that is buried under the 8 4.19 quintet; and indeed the 'quintet' resonances are not of the classical 1:4:6:4:1 intensities because of the presence of the underlying triplet. The *H NMR spectrum of the phosphine monosulfide products reveals a doublet of doublets of reduced integration (1:15, relative to that of the phenyl protons) at 8 3.36 (Fig. 3.6), and in the 31P{1H} NMR spectrum, a second set of doublets slightly upfield to those at 8 40.1 and -28.1 that are due to dpm(S) (Fig. 3.7). The second set of signals, as also discussed below, is attributed to the deuterated monosulfide product, dpm(S)-d2, and which is formed with dpm(S) in the ratio of 1:2. Of interest, in an in situ NMR experiment in CDCI3 where 2b is reacted with a 5-fold excess of dpm-d2, the same final products are observed; however, dpm(S)-d2 is formed in a slightly greater than 1:1 mole ratio to dpm(S), judging from a qualitative analysis of the 31P{1H} NMR spectrum (Fig. 3.8). Although the signals due to the diphosphine monosulfide products are resolved, the signals of the Pd products appear as one broad singlet, as do those of dpm and dpm-d2; the singlet at 8 - -6 is thought to result from a mixture of lb, 3b, and the 'di-substituted' Pd2(dpm-d2)2Br2 (4b) resulting from a ligand exchange of lb and/or 2b with dpm-d2, as discussed later. The *H NMR spectrum shows the expected Pd products at 8 4.24, the dpm(S) product at 8 3.35, and the dpm signal at 8 2.81 (Fig. 3.9). Analysis of the C H 2 region signals shows that the ratios of the integrated peak areas of the Pd products (lb + 3b + 4b) to dpm(S), dpm to dpm(S), and dpm to the Pd products are (to witliin ±0.3) 2.3, 2.9, and 1.3 to 1, 79 References on page 133 Chapter 3 Fig. 3.4. The ! H NMR (300 MHz) spectrum of the isolated Pd products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD2CI2 at R.T. 80 References on page 133 Chapter 3 Fig. 3.5. The ^ Pj'H} NMR (121 MHz) spectrum of the isolated Pd products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD2CI2 at R T . I 11 11 I 11 I 11 I j I I I » I I I 11 j M I T 1111| I I I I I 11 11 I j I I I ) 111 11| H 11 I I I 11 I I I - 4 - 5 - 6 - 7 -8 -9 -10 PPM •»ir*'» [ 1 1 1 ) 1 11 11 1 1 1 1 1 1 1 1 11 | 1 1 1 1 1 1 n T j i r r i ' 50 40 30 20 10 I 1 1 1 1 1 -10 -20 P P M 81 References on page 133 Chapter 3 F i g . 3.6. The ' H NMR (300 MHz) spectrum of the isolated phosphine monosulfide products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD2CI2 at R T . P P M 82 References on page 133 Chapter 3 Fig. 3.7. The "Pf/H} NMR (121 MHz) spectrum of the isolated phosphirie monosulfide products from the reaction of 2b with dpm-d2 at a 1:1 mole ratio in CD2CI2 at R T . dpm(S) dpm(S)-d2 ! M | l l l l | l l l l | i n i | l l l ! | l l l l | l l l l | l l l l | l l 41 40 39 38 P P M dpm(S) dpm(S)-d2 <7 i i ) i | i i i i | i i i i | i i i i | m i [ i i i i [ i i i i | i n i | i i -28 -29 -30 -3 P P M JL T—1—1—1—1—1—1—r—1—1 1 1—1—1—1—1—1—1—1—1 1 1 1—1—1—1—1—1 1 1—1—1—1—1—1—j—1—1—1—1—1—1—r 40 30 20 10 0 P P M -10 -20 -30 83 References on page 133 Chapter 3 Fig. 3.8. The *P{ ^H} NMR spectrum (121 MHz) of the completed in situ reaction between 2b and dpm-d2 (1:5 mole ratio) in CDCI3 at R T (72 h); X = unknown. dpm(S) 1 dpm(S)-d2 I i 1 1 1 1 / l b + 3b + 4b dpm-d2 dpm dpm(S)-d2 T - T - r - r 1 - 1 'P ' 1 ' 1 ' 1 1 1 1 1 1 i 1 1 1 1 | 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ' 1 1 1 1 ' 40 30 1 20 10 -10 -20 -30 P P M 84 References on page 133 Chapter 3 Fig. 3.9. The X H NMR (300 MHz) spectrum of the completed in situ reaction between 2b and dpm-d2 (1:5 mole ratio) in CDC13 at R T . (72 h). dpm i—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—r 4.50 4.00 3.50 3.00 2.50 P P M 85 References on page 133 Chapter 3 respectively. These results again reflect the combination of reaction 3.1.2 and the accompanying effects of diphosphine exchange of lb and 2b with dpm-d2 (see below). Reaction of 2b with dpmMe (1:1) after a few hours gave two immediately identifiable isolable Pd products, lb and Pd2Br2(dpm)(dpmMe) (5b),8 with the latter dominant; furthermore, some Pd2Br2(dpmMe)2 (6b)8 is also probably present (Fig. 3.10). Pd2Br2(dpm)(dpmMe) (5b) is characterized by four multiplets in the *H NMR: 8 a triplet of quartets centered at 5 4.98 (CH), a doublet of triplets at 8 1.00 (Me), and a doublet of triplets for each of the two C H 2 protons at 8 3.72 and at 4.55. Pd2Br2(dpmMe)2 (6b) is characterized in the *H NMR spectrum8 by a quartet of quintets at 8 4.98 (CH) and a doublet of quintets at 8 1.00 (Me); these multiplets overlap those due to 5b, and the presence of a small amount (-10%) of 6b is suspected because of the slightly greater than expected intensities of the 8 4.98 and 1.00 signals. The 31P{!H} NMR spectrum shows the singlet at 8-6.15 for lb and a complex AA'BB' pattern for 5b in the 8 +18 to -10 region, analogous to that recorded for Pd2Cl2(dpm)(dpmMe);8 the broad singlet expected for 6b at 8 16.18 could be buried within the AA'BB' pattern. The 31P{1H} NMR spectrum of the isolated diphosphine products shows just two sets of AB doublets: one set due to dpm(S) and the other, downfield at 8 51.9 and -13.4, due to dpmMe(S); the ratio of dpm(S) to dpmMe(S) is - 3:1, this result being consistent with the presence of more 5b than lb in the isolated Pd products. In terms of quantifying the extent of S-extraction via originally coordinated and added diphosphine, it was necessary to study diphosphine exchange with lb and 2b. Thus lb was reacted with 1 mole equivalent of dpm-d2 in CDCI3; a ! H NMR triplet at 8 2.81 due to the C H 2 protons of dpm is seen 'immediately', the integrated area being about half that of the quintet at 8 4.24 due to the C H 2 protons of the various Pd species present (lb, 3b, and possibly 4b). The 31P{1H} NMR spectrum shows the singlet of lb at 8 -5.5 (which could also correspond to some 86 References on page 133 Chapter 3 Fig. 3.10. The NMR spectrum (300 MHz) of the isolated Pd products from the reaction 2b and dpmMe (1:1 mole ratio) in CD2CI2 at R.T. lb 5h + 6h AJULA T " 5.0 ~1— 4.5 PPM 4.0 5b + 6b + dpmMe 1 1 1 1 1 • 1 1 1 « > 1 1 t < 1 7 1 1 1 1 1 4 P P M T 0 87 References on page 133 Chapter 3 4b), the tight AB quartet of 3b (5 -5.7), and a singlet at 6 -22.5 due to dpm. The spectra were invariant with time. Similarly, the ' H NMR spectrum of 2b in the presence of 1 mole equivalent of dpm-d2 shows (prior to any S-extraction) immediate generation of a triplet at 8 2.81 due to dpm, which overlaps the 5 2.90 signal of 2b. As well, new peaks appear in the 31P{1H} NMR spectrum: an AB quartet 8 5.95, attributed to Pd2Br2(p-S)(dpm)(dpm-d2) (7b), overlaps the singlet of 2b at 8 6.14 and the singlet of Pd2Br2(p-S)(dpm-d2)2 (8b) at 8 6.00, and a singlet at 8 -22.5 due to dpm is seen beside the singlet of dpm-d2 at 8 -23.0 (Fig. 3.11). Diphosphine exchange of lb with dpmMe was not studied in detail but was shown to occur, at a much slower rate in contrast to the rapidly established equilibrium of lb or 2b with dpm-d2. In a single in situ NMR experiment where lb was reacted with 1 mole equivalent of dpmMe, the final ' H and 31P{IH} NMR spectra (which are then invariant with time after ~3 days) indicate that significant amounts of Pd2Br2(dpm)(dpmMe) (5b), Pd2Br2(dpmMe)2 (6b), and dpm are formed (with some lb and dpmMe remaining) (Figs. 3.12 and 3.13). The suggestion that 6b is present is again based on 'increased intensities' of the 8H 4.98 and 1.00 signals (Fig. 3.12) and a possible buried 5P 16.1 singlet (Fig. 3.13) (see above). Analysis of the various non-phenyl protons signals in the ! H NMR spectrum gives the following absolute integrated values for lb, dpmMe, 5b, dpm, and 6b as 0.98, 0.91, 2.55, 2.27, and 0.58, respectively (with respect to the 60 phenyl protons of lb + dpmMe). Also, in contrast to the rapidly established equilibrium of 2b with dpm-d2, 2b does not undergo detectable diphosphine exchange with dpmMe (prior to any S-abstraction) as evidenced by NMR spectra. The catalysis of reaction 3.1.3 was examined at R T . using lb and 2b. Gas chromatography and NMR spectroscopy were used to analyze for the products, H2 and dpm(S), respectively. In preliminary studies, contrasting results were obtained where products were 88 References on page 133 Chapter 3 Fig. 3.11. The 31P{ !H} NMR (121 MHz) spectrum of 2b on addition of 1 mole equivalent of dpm-d2; the spectrum was recorded at R.T. immediately after sample preparation in CDCI3. P P M 89 References on page 133 Chapter 3 Fig. 3.12. The *H NMR spectrum (300 MHz) recorded at R.T: in CDC13 of lb on addition of 1 mole equivalent of dpmMe (72 h). dnm 5b + 6b + dDmMe lb 5b + 6b 5b AJ 5b dpmMe i — | — : — i — i — i — | — r — i — i — i — i — i — i — i — i — i i i — i — i — i — : — i — r 5.0 4.5 4.0 3.5 3.0 1 I ' ' 1.0 P P M 90 References on page 133 Chapter 3 Fig. 3.13. The 3lV{lH) NMR spectrum (121 MHz) recorded at R T . in CDC13 of lb on addition of 1 mole equivalent of dpmMe (72 h). lb dpm 91 References on page 133 Chapter 3 observed in some studies but not in others. A series of experiments were then carried out where [lb] ranged from (8.76 - 263) x 10"5 M, [dpm] from (3.64 - 52.0) x 10-3 M, and [H2S] = 0.013 M or 1.0 M (Table 3.3). The reaction (H2S + dpm —» H 2 + dpm(S)) was allowed to proceed for up to 72 h prior to analysis. In experiments 1 and 3 to 9, very little or no dpm(S) or H 2 products were seen. Moreover, UV-vis, lH and 31P{!H} NMR spectroscopic analyses revealed in all cases the presence of Pd species characterized by: three absorption maxima at 358, 422, and 522 nm in the UV-vis spectrum (Fig. 3.14); a multiplet, triplet, and singlet in a 10:1:2 integration ratio at 5 7.0-8.0, 4.72 and 3.37, respectively, in the ' H NMR spectrum; and a singlet at 6 -4.71 and a series of resonances resembling an AA'BB' pattern from 8 22 to 0 in the 31P{1H} NMR spectrum (Figs. 3.15 and 3.16). The same results were seen when non-cylinder, synthesized H 2S was used (expt. 8) or when 2b was used as the catalyst (expt. 10). [In the latter study, although dpm(S) was observed in a 5% yield, no H 2 was detected.] The unknown Pd species was subsequently isolated and tested for catalytic activity; under similar experimental conditions (expt. 11), no products were found. In expt. 2 where a lower concentration of dpm was used, H 2 was detected and dpm(S) was found to be formed in -47% yield after 24 h. A UV-vis spectrum was obtained at the end of the experiment and revealed absorption maxima at 340 and 426 nm (Fig. 3.17). This experiment (on the same scale) was repeated in an NMR tube fitted with a PTFE J. Young valve. After 24 h, ' H and 31P{TT} NMR spectra showed the presence of H 2 (at 8 4.64) and dpm(S) in about 15 and 20% yields, respectively, lb and 2b were the only Pd species seen, but their relative amounts were not determined because of the extremely low concentration conditions used. After 72 h, the yields of these products increased to approx. 40 and 50%, respectively. No more changes occurred in the system thereafter, and 31P{1H} NMR spectra revealed that decomposition of the catalyst had taken place. 92 References on page 133 Chapter 3 Table 3.3. Conditions used in the study of the catalysis of reaction 3.1.3 in CHCb at R.T. using Pd2Br2(dpm)2 (lb) or Pd2Br2(u-S)(dpm)2 (2b). Experiment Amount of catalyst (mM) Amount of dpm (M) [ H 2 S ] (M) Time period (h) Product formation dpm(S)a H 2 1 0.088 lb 0.026 1.0 24 trace no 2 0.35 lb 0.0036 1.0 24 -47% yes 3 0.35 lb 0.026 1.0 24 trace no 4 0.088 lb 0.013 1.0 24 trace no 5 0.88 lb 0.026 1.0 72 no no 6 1.75 lb 0.052 0.013 24 no no 7 2.63 lb 0.052 1.0 72 no no 8 2.63 lb 0.052 1.0* 72 trace no 9 0.66 lb 0.026 1.0 5 no no 10 2.56 2b 0.052 0.013 24 ~5%c no 11 d 0.052 1.0 24 no no " Based on dpm reactant. * Non-cylinder H 2 S, prepared from CaS + HCl(aq). c Stoichiometric (see text). d 10 mg of an unknown Pd species isolated from expts. 5 to 10. 93 References on page 133 Chapter 3 Fig. 3.14. The UV-vis spectrum of the isolated catalytically inactive species (in CHCI3). 300 400 500 600 wavelength (nm) 94 References on page 133 Chapter 3 Fig. 3.15. The ! H NMR spectrum (300 MHz) recorded at R.T. in CDC13 of the isolated catalytically inactive species. P P M 95 References on page 133 Chapter 3 Fig. 3.16. The 31P{'H} NMR spectrum (121 MHz) recorded at R T . in CDC13 of the isolated catalytically inactive species. I i i i I | I I I I | I i i I | II i i | i i i i | i i I I | I i I . | i i i i | i i i , ; i i i i | I i I I | I i i i | i i i i | i i i i '| i i i i | i I i ' | 40 30 20 10 0 -10 -20 -30 -40 P P M 96 References on page 133 Chapter 3 Fig. 3.17. The UV-vis spectrum obtained from catalytic studies of dpm + H2S using lb in CHCI3 after 24 h (Expt. 2). wavelength (nm) 97 References on page 133 Chapter 3 3.3 Discussion Kinetic results show that, at least for the chloride and bromide systems, the rate of the reconversion reaction (2->l) has a first-order dependence on both 2 and dpm. No intermediates were seen in low temperature NMR studies and this result, along with the observed isosbesticity in UV/vis spectroscopic studies, show that reaction 3.3.1 can be represented as shown, making no distinction in the identity of the dpm ligands: (3.3.1) Pd2X2(p-S)(dpm)2 (2) + dpm — ^ Pd2X2(dpm)2 (1) + dpm(S) Labeling Experiments and Statistical Analysis. Labeling experiments with dpm-d2 have demonstrated that diphosphine exchange takes place with both 2 and 1, with rapid equilibria being established. The systems are complex because of the numerous rapidly established reactions occurring; the distribution of products, as a consequence, is considered to be statistical according to the stoichiometries of the reactants. For example, for the reaction of 2b with dpm-d2, prior to S-abstraction, the following rapid equilibria are established: (3.3.2) Pd2Br2(u-S)(dpm)2 (2b) + dpm-d2 Pd2Br2(u-S)(dpm)(dpm-d2) (7b) + dpm (3.3.3) Pd2Br2(u-S)(dpm)(dpm-d2) (7b) + dpm-d2 Pd2Br2(p-S)(dpm-d2)2 (8b) + dpm For equilibrium 3.3.2, for an initial 1:1 ratio of 2b and dpm-d2, the following distribution of species will be present at equilibrium: 98 References on page 133 Chapter 3 fPd2Br2(u-S)(dpm)(dpm-d2) + |dpm-d2 + 4/Pd2Br2(p-S)(dpm)2 + f dpm Similarly for equilibrium 3.3.3 at an initial 1:1 ratio of 7b and dpm-d2, the distribution of species will be: f Pd2Br2(u-S)(dpm)(dpm-d2) + |dpm + |Pd2Br2(u-S)(dpm-d2)2 + f dpm-d2 The S-abstraction then occurs via the slower reactions 3.3.1, and 3.3.4 - 3.3.8: (3.3.4) Pd2Br2(p-S)(dpm)2 (2b) + dpm-d2 > }Pd2Br2(dpm)(dpm-d2) (3b) + idpm(S)-d2 + |Pd2Br2(dpm)2 (lb) + fdpm(S) (3.3.5) Pd2Br2(u-S)(dpm)(dpm-d2) (7b) + dpm > f Pd2Br2(dpm)(dpm-d2) (3b) + |dpm(S)-d2 + ^Pd2Br2(dpm)2 (lb) + fdpm(S) (3.3.6) Pd2Br2(u-S)(dpm)(dpm-d2) (7b) + dpm-d2 > }Pd2Br2(dpm)(dpm-d2) (3b) + |dpm(S) + ^-Pd2Br2(dpm-d2)2 (4b) + f dpm(S)-d2 (3.3.7) Pd2Br2(u-S)(dpm-d2)2 (8b) + dpm > f Pd2Br2(dpm)(dpm-d2) (3b) + ^-dpm(S) + |Pd2Br2(dpm-d2)2 (4b) + fdpm(S)-d2 99 References on page 133 Chapter 3 (3.3.8) Pd2Br2(p-S)(dpm-d2)2 (8b) + dpm-d2 ——> Pd2Br2(dpm-d2)2 (4b) + dpm(S)-d2 and then the rapid equilibria 3.3.9 and 3.3.10 are established within lb type species: (3.3.9) Pd2Br2(dpm)2 (lb) + dpm-d2 5=^ Pd2Br2(dpm)(dpm-d2) (3b) + dpm (3.3.10) Pd2Br2(dpm)(dpm-d2) (3b) + dpm-d2 Pd2Br2(dpm-d2)2 (4b) + dpm with the following distribution of species present at equilibrium for reaction 3.3.9 for an initial 1:1 ratio of lb and dpm-d2: fPd2Br2(dpm)(dpm-d2) + ^-dpm-d2 + |Pd2Br2(dpm)2 + f dpm and similarly for equilibrium 3.3.10: f Pd2Br2(dpm)(dpm-d2) + jdpm + |Pd2Br2(dpm-d2)2 + fdpm-d 2 To contend with the many reactions, each set of rapid equilibria, 3.3.2 and 3.3.3, and 3.3.9 and 3.3.10, occurring essentially with equal probability, and to compare with the experimental results, a computer program (see Appendix HI) was written and tested to simulate the reconversion reaction with dpm-d2; the program displays the different statistical outcomes resulting from using different reactant ratios of 2b and dpm-d2. The program is based on the premise that the diphosphine 100 References on page 133 Chapter 3 exchange and S-abstraction reactions proceed via transition states of a certain configuration where three diphosphates ligands are equivalent. To consider one possibility, I would be a reasonable and easily visualized transition state for the exchange and the S-abstraction reaction within 2b, although the much slower S-abstraction (vs. exchange) requires, of course, differences in the transition states for the two processes; II seems reasonable for the exchange within lb type species. (I and II are chosen just as models to introduce the statistical nature of the reactions; more realistic chemical formulations are considered below). Br —Pd — S — Pd — B r Br —Pd Pd—Br I II (The phenyl groups are omitted for clarity; the * implies -CD2-) The simple overall second-order process for reaction 3.3.1 could proceed, for example, via I, containing three equivalent diphosphine bridges, with the reactant diphosphine initially coordinating at a vacant axial site of the square planar Pd centers within the A-frame complex 2b. Product formation is then envisaged as sulfur abstraction by any one of the bridging phosphines with equal probability; this picture leads to statistical product distributions that correspond closely to those observed. For example, modeling of the S-abstraction within a 1:1 reaction between 2b and dpm-d2 gives the following predicted overall stoichiometry: 101 References on page 133 Chapter 3 (3.3.11) Pd2Br2(u-S)(dpm)2 + dpm-d2 -> fPd2Br2(dpm)2 + fPd2Br2(dpm)(dpm-d2) (2b) (lb) (3b) + }Pd2Br2(dpm-d2)2 + f dpm(S) + |dpm(S)-d2 (4b) The experimental results are consistent with this prediction. The 2:1 ratio of dpm(S) to dpm(S)-d2 is readily seen (Fig. 3.7), but the ratios of the Pd species (lb:3b:4b) are less well-defined. If a ] H NMR spectrum for the mixture of Pd products of eq. 3.3.11 were constructed, the integration of the phenyl proton signals to those of the C H 2 protons would result in a 15:1 ratio, which is exactly that found experimentally (Fig. 3.4). Both lb and 3b are definitely seen in the ' H andMP{'H} NMR spectra and appear to be present in comparable amounts (Figs. 3.4 and 3.5), while 4b is assumed to be present in a smaller amount giving rise to a singlet within the 8P resonances around -6 (Fig. 3.5). Similarly, a constructed ! H NMR spectrum for the mixture of the phosphine monosulfide products will give a 15:1 ratio of the integration of the phenyl proton signals to those of the C H 2 protons, the experimental results being in excellent agreement with this value (Fig. 3.6). Further support for the statistical product distribution via transition states containing three equivalent diphosphines comes from the analysis of the in situ reaction of 2b with a 5-fold excess of dpm-d2 (Figs. 3.8 and 3.9). Rapid exchange within 2b, followed by slower S-abstraction, and then rapid exchange within lb gives the following statistical outcome: (3.3.12) 2b + 5dpm-d2 -> 0.068 1b + 0.382 3b + 0.547 4b + 0.433 dpm(S) + 0.564 dpm(S)-d2 + 1.048 dpm + 2.955 dpm-d2 102 References on page 133 Chapter 3 Thus, in a simulated ! H NMR spectrum, the theoretical ratios of the integrated peak areas (CH 2 protons) of the Pd products to dpm(S), dpm to dpm(S), and dpm to the Pd products are, respectively, 1.2, 2.4, and 2.0 to 1. There is reasonable experimental agreement for the last two ratios, but a discrepancy is seen for the first; however, more importantly, the correct trends are observed: the integrated peak area of the Pd products is greater than that of dpm(S) but less than that of dpm. The 31P{1H} NMR spectrum (Fig. 3.8), on the other hand, clearly shows the approximately 1:1 ratio of dpm(S) to dpm(S)-d2, thus, providing stronger support for such transition states. Finally, in diphosphine exchange studies of type lb species with dpm-d2, the proposal of a transition state such as II is reasonable in that the expected statistical outcome of the exchange between lb and dpm-d2 (1:1) is as follows: (3.3.13) lb + dpm-d2 -> f lb + ± 3b + £ 4b + f dpm + | dpm-d2 and the theoretical ratio of the C H 2 protons of the Pd products to those of dpm is 2:1; experimentally, this ratio is found, thus providing very strong support for a transition state with equivalent diphosphines. NMR spectra have shown that dpmMe also undergoes diphosphine exchange with lb at a 1:1 ratio, and that the exchange is slow, taking ~3 d for equilibrium to be established; the equilibrium favors the formation of Pd2Br2(dpm)(dpmMe) (5b): K, (3.3.14) lb + dpmMe ^=r^ Pd2Br2(dpm)(dpmMe) (5b) + dpm 103 References on page 133 Chapter 3 K 2 (3.3.15) Pd2Br2(dpm)(dpmMe) + dpmMe Pd2Br2(dpmMe)2 (6b) + dpm The equilibrium constants at room temperature are calculated from the spectral integration values (Fig. 3.12) and are found to be Ki « 6 and K 2 « 0.6. NMR results show that 2b does not undergo rapid diphosphine exchange with dpmMe as it does with dpm-d2, but it is uncertain whether or not some slower exchange does occur on the timescale of the exchange of lb with dpmMe, which is also that of the S-abstraction by this phosphine. Nevertheless, whether or not exchange does happen, the resulting product distributions of the reaction between 2b and dpmMe are expected to differ from that of the dpm case because the phosphines are different. An approximate 3 : 1 ratio of dpm(S) to dpmMe(S) is seen in the 31P{'H} NMR spectrum, complementing the result that greater amounts of 5b and 6b than lb are observed. The preliminary experimental findings suggest that, as with dpm, the S-abstraction reaction with dpmMe probably proceed via a transition state with close to equivalent diphosphine ligands. Transition States and Intermediates. More insight into the nature of the transition state and intermediates is gleaned on consideration of reactivity of 2b with other phosphines. Several other tertiary phosphines, including PPh3,5 PPh2Me,5 and Ph2P(CH2)3PPh2 (dpp), were reacted with 2b, but surprisingly no sulfur abstraction was observed, i.e. no phosphine sulfides were formed, and no lb was generated. With Ph2P(CH2)2PPh2 (dpe), however, small amounts of lb and dpm(S) were formed, but the major Pd species generated are as yet unidentified as the spectra are complicated; no dpe(S) was seen (Figs. 3.18 and 3.19). In situ treatment of 2b with dpp (1:1) at R.T. resulted in a slow reaction (over ~3 d) and complex 31P{1H} and ' H NMR spectra, attributable to an as yet uncharacterized species W (or mixture of species) (Fig. 3.20). (The formation of lb is 104 References on page 133 Chapter 3 Fig. 3.18. The *H NMR spectrum (300 MHz) recorded at R T . in CDC13 of 2b on addition of 1 mole equivalent of dpe (72 h). -T 1 1 l j f i • T 1 1 r- ~I 1 1 1 1 1 1 r 1 r-P P M 105 References on page 133 Chapter 3 Fig. 3.19. The "Pf/H} NMR spectrum (121 MHz) recorded at R T . in CDC13 of 2b on addition of 1 mole equivalent of dpe (72 h). lb dpm(S) dpm(S) i ' ' ' ' | ' I i i i i i i i | i i i i i i i i i | i i i i i i i i i | i i i i i i i i i ( i i i i i i i i i | 60 40 20 0 -20 -40 P P M 106 References on page 133 Chapter 3 Fig. 3.20. The lH (300 MHz) and 31P{'H} (121 MHz) NMR spectra recorded at R T . in CDCI3 of 2b on addition of 1 mole equivalent of dpp (72 h) (species W). PPM 107 References on page 133 Chapter 3 questionable because, although the l H NMR spectrum shows a prominent quintet at 5 4.2 attributable to lb, the corresponding singlet at 8 -6.15 is not seen in the 31P{ TI} NMR spectrum. Instead, a distinctive singlet at 8 -7.20 is seen; accuracy in 31P{1H} NMR spectra is usually within ±0.05 ppm.) Remarkably, treatment of this solution with ~2 mole equivalent of dpm resulted in slow but quantitative formation of dpm(S). The NMR spectra are still complex; however, addition of ~3 mole equivalent more of dpm yielded clean spectra with an A A ' B B ' pattern seen in the 31P{1H} attributed to dimeric Pd species with bridging dpm and dpp ligands (Figs. 3.21 and 3.22). The 'W solution' could contain species with monodentate, 'dangling' dpp, but no 8P resonances in the region for free dpp (8P ~ -18)9 were seen; di- or polynuclear species formed by bridging dpp seem plausible. The uniqueness of dpm (and dpmMe) in quantitatively abstracting the sulfur could be satisfyingly pictured via a transition state such as I; dpm (and presumbly dpmMe) could generate this state in which the S atom is in a trigonal prismatic arrangement of P atoms, all of these being forced closer to the sulfur than the P atoms within 2b. Presumably, if the longer chain dpp bridges the two Pd atoms, it does so in a configuration not very different to the ground state of 2b and which gives no sulfur abstraction. There could be sufficient flexibility in dpm and dpmMe to generate transition states such as II, the supposed pathway for diphosphine ligand exchange with lb. This flexibility could exist for dpe, but presumably this ligand prefers to form a 5-membered chelate ring at a single Pd site; this could induce conversion of a p-dpm to an q'-dpm with subsequent abstraction of S by the latter (see below). It should be noted that complexes of the type Pd2Cl4(p-But2P(CH2)nPBut2)2 (n=7,10) are known;10 these contain square-planar, d8 Pd(U) moieties, and no metal-metal bond is involved. Species containing q'-diphosphine ligands are well documented,11"13 including q'-dpm in complexes of the type [Pt2(L)(p-dpm)2(q1-dpm)]+, L = alkyl or H . 1 3 The possibility of dpm (and 108 References on page 133 Chapter 3 Fig. 3.21. The *H NMR spectrum (300 MHz) recorded at R T . in CDC13 of species W on addition of 3 mole equivalent of dpp (72 h). 109 References on page 133 Chapter 3 Fig. 3.22. The 31P{'H} NMR spectrum (121 MHz) recorded at R T . in CDCh of species W on addition of 3 mole equivalent of dpp (72 h). dpm 1 1 I ! 1 1 1 I I I I I | ! M I | | | | | | | | | | | | | | | | i | | | | | | | | | . | | | | | | | I | I I I I | I I I I | ! I I I | II I I | I I I I | I I I I | 40 30 20 10 0 -10 -20 -30 -40 P P M 110 References on page 133 Chapter 3 dpmMe) reacting with 2b by binding in an n'-fashion at one Pd center with the 'free end' then abstracting sulfur seems the most obvious pathway for formation of the diphosphine monosulfide which would involve incipient formation of a 5-membered ring; the findings of the product distribution using dpm-d2 and dpmMe could be accommodated via rapid fluxionality between the n'- and p-diphosphines prior to S-abstraction. Such fluxionality offers the obvious pathway to diphosphine exchange with rapid on- and off-rates for the n'-ligands (see below). That other phosphines do not abstract the sulfur implies the requirement of a 5-membered ring for this process. Although I was offered as an easily visualized transition state for the S-abstraction, a more realistic chemical picture is shown in III; if the diphosphines are indistinguishable (essentially so if dpm-d2 is used as the entering diphosphine), the transition state is effectively one of 3-fold degeneracy and the "averaged" transition state picture approximates to that shown in I. I l l Consideration of the various findings reveals that dpm reacts relatively rapidly with 2b (to give exchange and S-abstraction) and with lb (exchange), and yet dpmMe and dpp (and dpm(S) (see below)) react more slowly. Presumably, all the chelating phosphines must first react at a metal center as n'-nucleophiles; the difference in apparent rates cannot be due to electronic or steric effects (indeed, regarding the latter, dpm might be more 'encumbered' than dpp), and thus 111 References on page 133 Chapter 3 some rationale is needed. The series of steps shown in eqs. 3.3.16 and 3.3.17 can account for the observations: (3.3.16) Pd2Br2(u-S)(u-dpm)2 + (P-P) Pd2Br2(p-S)(ii1-(P-P))(p-dpm)2 (TV) or Pd2Br2(u-S)(r| '-(P-P))(p-dpm)(ri '-dpm) (3.3.17) IV 5j=p Pd2Br2(p-dpm)(p-(P-P))(Ti1-dpm(S)) (V) — p r o d u c t s III would be a reasonable transition state en route from IV to V. It seems reasonable that for all the P-P systems, the ki values are comparable; the findings require that for dpm or dpm-d2, the ki/k-i equilibrium is established rapidly but the k2 step can compete sufficiently with the k.! step to result in S-abstraction, i.e. the bite of the dpm ligand makes it highly favorable for forming the 5-membered Pd-S-P-CH 2 -P ring. Diphosphine exchange with dpm (or dpm-d2) within type 2b species is pictured as proceeding via reaction 3.3.16, with an originally coordinated diphosphine having the same probability of leaving as the incoming diphosphine. The S-abstraction occurs via eq. 3.3.17 as discussed above; all three dpm ligands within JTV (because of the rapidly established ki/Li equilibrium) have an equal probability of abstracting the S. The discussion necessitates that species TV and V are genuine intermediates. A steady state treatment for TV gives the following rate expression for S-abstraction: (3.3.18) Rate = ^ [2b] [P-P] k_, +k2 112 References on page 133 Chapter 3 consistent with our experimental findings, e.g. with kB r = kik2(k_i + k2)"\ For the other phosphines, the k2 step cannot compete effectively with the k-i step, and the L i value for these phosphines must also be greater than the L i value for dpm/dpm-d2 because the exchange (e.g. with dpmMe) is established much more slowly. The rate law for S-abstraction using other phosphines thus approximates to (ki/Li)k2[2b][P-P]. The experimental findings show that the dpm/dpm-d2 exchange for 2b at R T . is established in less than seconds for solutions typically ~ 10"2 M in both complex and phosphine, while the S-abstraction occurs with a rate constant of ~ 0.03 M " V \ Qualitatively, typical, somewhat arbitrary values that would satisfy the above analysis are ki ~ 103 M V 1 , t i ~ 1 0 2 s"1, and k2 ~ 10"3 s"1; the S-abstraction via TV is ~ 10"5 times less probable than loss of the ri'-dpm. Within type lb species, the rapid exchange involving dpm-d2 can be accounted for by an equilibrium (eq. 3.3.19) analogous to that giving the exchange within type 2b species, loss of n'-dpm from VI giving the exchange process. (3.3.19) Pd2Br2(u-dpm)2 + dpm-d2 Pd2Br2(ri1-dpm-d2)(p-dpm)2 (VI) or Pd2Br2(n 1-dpm-d2)(r| 1-dpm)(p-dpm) The transition state for this exchange could be akin to that shown in II but with one Pd-P bond at each metal being stretched towards an ^ /-(P-P) configuration en route to VI. This picture satisfactorily explains the observations for dpm-d2. The slow exchange of dpmMe with lb (as with 2b) implies that loss of an u'-dpmMe from a species such as VI (as with TV) occurs much more readily than loss of n'-dpm, possibly because of steric factors. 113 References on page 133 Chapter 3 Of note also is that in situ treatment of 2b with dpm(S) (1:1) in CDCI3 yields over several days an uncharacterized species Y (Fig. 3.23) that on subsequent reaction with dpm (1:1) slowly generates lb and two mole equivalents of dpm(S). [Note: the NMR spectra of isolated Y have the same characteristics as those of in situ generated Y.] Y could be a species such as Pd2Br2(p-S)(p,-dpm)2(q1-dpm(S)) or Pd2Br2(p-S)(qI-dpm)(p-dpm)(q1-dpm(S)), while subsequent reaction with dpm could slowly replace the coordinated dpm(S) and lead then to an intermediate such as TV with P-P = dpm. No reaction occurs between 2b and dpm(S)2. Support for species Y being a single compound comes from elemental and mass spectrometric analyses of an isolated species. The FAB mass spectrum (Fig. 3.24) shows a prominent mass ion peak at 1429 m/z attributed to a species with loss of two bromine atoms, e.g. Pd2(p-S)(p-dpm)2(q1-dpm(S))+, while elemental analysis shows excellent agreement between theoretical and experimental values. Attempts to grow crystals of species Y (via solvent layering techniques) for structural information were unsuccessful due to facile formation of oils even at low temperatures (-40 °C). Of interest, it was found that 2a and 2c react with dpm(S) in an analogous way giving species X and Z, respectively; the ' H and 31P{1H} NMR spectra are exactly the same as those seen in the reaction of 2b with dpm(S) (Fig. 3.23). [Note: attempts to analyze the complex NMR spectra using, for example, a spin simulation program, were unsuccessful.] Furthermore, the UV-vis spectra of all three species (isolated from their respective reactions) show only one absorption maximum at 464 nm in the range 300 to 600 nm (Fig. 3.25). These results suggest that species X, Y, and Z may be ionic in nature; further studies to elucidate the structures are required. Of interest, we find that 2c, Pd2I2(u-S)(dpm)2, reacts with I2 in CDCI3 at R.T. according to eq. 3.3.20 (see Chapter 4); the NMR data (Table 3.1) of the Pd 2 intermediate correspond well to 114 References on page 133 Chapter 3 Fig. 3.23. The T i (300 MHz) and ^ {Tl } (121 MHz) NMR spectra recorded at R T . in CDCb of 2b on addition of 1 mole equivalent of dpm(S) (72 h) (species Y). PPM 115 References on page 133 Chapter 3 Fig. 3.24. Mass spectrum (FAB, matrix = thioglycerol + CHCI3) obtained of species Y. 100-M a v Pd2(dpm)(dpm(S))+ > o_i -rr -r 905 900 969 T T 1013 1 1049 1000 1100 1 1 1 1 j 1 r~ 1200 m/z 1 1 1 1 1429 -rVrr 1 "1 i 1300 1400 1500 1600 116 References on page 133 Chapter 3 Fig. 3.25. The UV-vis spectra of species isolated from reactions of Pd2X2(p-S)(dpm)2 (2) with 1 mole equivalent of dpm(S) in CHC13 (X = CI (species X), Br (species Y), I (species Z)). 1——• • • • 1 •——>- • • 1 •*- • • • 1 300 400 5O0 600 wavelength (nm) i _ , , , , , , , , 1 , , , , , 30O 400 500 BOO wavelength (nm) 117 References on page 133 Chapter 3 those of the tetraiodo species, characterized previously by Hunt and Balch14 in a study of the reaction of I2 with Pd2J-2(dpm)2 (lc). p-"" P (3.3.20) 2c + I2 > id^1 p ' d / ! + *S' 1 —' 1 1 — I p--_--p I 2 Pdl2(dpm) That transannular addition can occur across an A-frame complex, reaction 3.3.20, albeit with destruction of the A-frame, offers some support for reactivity via a transition state such as I or III, the diphosphine addition resembling in the broadest sense the addition of two iodide ligands. Thus, in eq. 3.3.20 the bridged sulfide is formally oxidized by I2 to elemental sulfur, with the electron transfer presumably proceeding from the (p-S) through the metal atoms toward the attacking I2 reagent, and a transition state akin to I or III but with the P-P replaced by I-I. Indeed, the conversion of the 2 to 1 species could, in principle, involve incipient formation of a S-atom (cf. eq. 3.3.20) that is then scavenged as a phosphine monosulfide. The reactivity trend for reaction 3.3.1, X = Cl > Br > I, can be rationalized in terms of transition state I. The activation parameters for the chloro and bromo systems reveal that the dominant factor governing reactivity is the entropy term (-127 and -144 for the chloro and bromo systems, respectively). A large, unfavorable AS* value is reasonable for the addition of the dpm ligand to 2b to generate I or III and this is expected to be more difficult when the auxiliary ligand is the bulkier bromide (vs. chloride), because of interactions with the phenyl groups. For the iodo 118 References on page 133 Chapter 3 system, AS* is predicted to be still more negative. The essentially constant A H * value for the chloro and bromo systems tends to imply that the energy required to form a transition state such as I or III is similar in the two systems; any contribution from the trans-effect of the halides (Br > CI)15 must be small. Catalytic Conversion of H2S. As noted in the Introduction, the demonstration of the conversion of 1 to 2, and the reconversion using dpm, constitutes a catalytic cycle for reaction 3.1.3 which could be catalyzed by complex 1 or 2. Some experiments were carried out confirming the existence of the catalysis. Difficulties, however, were encountered in these studies due to one (or several) catalytically inactive species being formed under conditions of high [dpm]. In almost all studies using lb as the catalyst, little or no product formation was seen (Table 3.3, expts. 1, 3 to 9). When 2b was used, dpm(S) was formed in -5% yield (stoichiometric); however, no H 2 was detected (expt. 10); the monosulfide is likely formed during initial reaction of 2b with dpm (i.e. the reconversion reaction). Indeed, with [2b] = 2.6 x 10"3 M and [dpm] = 0.052 M , in principle, the concentration of dpm(S) formed via the reconversion reaction is -2.6 x 10"3 M which corresponds to 5% dpm conversion. Product formation was seen in one study (expt. 2) where H 2 was detected and dpm(S) was found to be formed in -47% yield. Here, [dpm] = 3.6 x 10"3 M initially, and 1.7 x 10"3 M was converted to dpm(S) after 24 h; [lb] = 3.5 x 10"4 M and hence, true catalysis is observed. Turn over numbers based on this study is calculated to be -5 d"1. (Previous studies in CH2C12 have noted that a 10"3 M solution of lb at R.T. under 1 atm H 2S converted 0.02 M dpm completely to dpm(S) in a few hours; total turn over numbers were -20.19) The presence of H 2 was unequivocally established by corresponding NMR experiments; here, H 2 and dpm(S) were formed in -40 and 50% yields, respectively, after 3 d (why there is a difference in reaction 119 References on page 133 Chapter 3 timescales is unclear, although there was a difference of ~5 0 in the room temperature recorded (23 vs. 18 °C)). UV-vis spectrophotometry was used for a qualitative assessment of the nature of the species in solution; as confirmed by NMR spectroscopy, the UV-vis spectrum (Fig. 3.17) is characteristic of a system containing both lb and 2b (cf. Fig. 3.1). The catalytically inactive species was seen in previous studies by 31P{1H} NMR and was thought to be the active species during the catalysis.5 In the present thesis work, the inactive species was isolated and characterized using UV-vis spectrophotometry and NMR spectroscopy. The UV-vis spectrum shows three well-defined absorption maxima (Fig. 3.14); NMR spectra are not complex (Figs. 3.15 and 3.16). In the ' H NMR spectrum, two resonances, a triplet and a singlet, are seen in the range 5 3 to 5 with an integration ratio of 1:2 and are indicative of protons of CHb groups. Furthermore, the phenyl protons integrate for 10 H with respect to both the triplet and singlet resonances. In the 31P{'H} NMR spectrum, a singlet and a series of resonances resembling an A A ' B B ' pattern are observed in a 1:2 integration ratio. The implication is that multiple numbers of diphosphine ligands are present, possibly multiples of three. A complex which satisfactorily meets these requirements could be a dimeric Pd complex bridged also by the sulfur atom of a dpm(S), species VII: S' p Pd P • Br (the phenyl groups are omitted for clarity) 120 References on page 133 Chapter 3 Elemental analysis for a species formulated as VII shows excellent agreement between theoretical and experimental results. Crystals of this species were grown (via solvent layering techniques) but the poor crystallinity precluded an X-ray crystallographic study; however, mass spectrometric analysis indicates a parent ion peak at 1643 m/z (Fig. 3.26) corresponding to a species that can be formulated as VII with 1 C H 2 C I 2 solvate molecule. Because the catalytically inactive Pd species is invariably produced under conditions where the [dpm] is high, formation of VII perhaps proceeds via reaction of dpm with the hydrido(mercapto) intermediate2 formed from lb with H 2S. For example, (3.3.21) Pd2Br2(dpm)2 (lb) + H 2S Pd2Br2(H)(SH)(dpm)2 (3.3.22) Pd2Br2(H)(SH)(dpm)2 + dpm VII + ' H 2 ' H 2 is included in eq. 3.3.22 by virtue of mass balance; failure to detect it may be due to its low concentration. Although further studies are warranted, the conclusion is that catalysis can proceed simply via coupling of reactions 3.1.1 and 3.1.2. 121 References on page 133 Chapter 3 Fig. 3.26. Mass spectrum (FAJ3, matrix = thioglycerol + CHCU) of the catalytically inactive species VII with the region from 1400 to 1650 m/z expanded. Vi ft 1 1611 14 M - dpm(S) 1227 T— r M = Pd 2 Br 2 (dpm(S))(dpm) 2 «CH 2 Cl 2 1200 1400 "I * T * I 1 1 1 1 1 1600 1800 2000 m/z 1611 M - S M - Pd - S 1504 M - P d 1536 M j.'iiiiiiij.i jiilHIHIjIlii j |..iilllll|Hlihi.i| j ll'l'm 1 1 Illl I1IH|I I . I 1643 Hilii| |Hlllll"i 1400 1450 1500 1550 1600 1650 m/z 122 References on page 133 Chapter 3 3.4 Experimental section The materials used, synthetic procedures for the ligands and complexes, and instrumentation used for 3 1 P and ' H NMR, and UV/vis spectra have been described in Chapter 2. The dpm (Aldrich), l,2-bis(diphenylphosphino)ethane (dpe) (Strem), 1,3-bis(diphenylphosphino)propane (dpp) (Strem), and sulfur (Aldrich) were pure and were used as received. Dpm(S), dpmMe, and the complexes PdCl2(PhCN)2, Pd2(dba)3CHCl3, la - lc, and 2a -2c were synthesized using published methods as outlined in Chapter 2. All experiments were performed under N 2 unless otherwise specified. 3.4.1 Preparation of dpm-d2 Lithium diphenylphosphide, LiPPh2, (22.6 g, 0.118 mol, in 250 mL THF) 1 6 was slowly added dropwise to a solution of CD2C12 (5 g, 0.059 mol, in 50 mL THF) over a period of ~2 h. The solvent of the resulting yellow mixture was rotovaped off before 200 mL of CH 2C1 2 was added to re-dissolve the crude product. The CH2C12 solution was washed with 4 x 100 mL of H 2 0 , and then dried over CaS04. Removal of the solvent gave a light yellow solid which was dissolved in 50 mL of a CH2Cl2-hexane (1:2) mixture and chromatographed through a column (8 x 40 cm) of silica gel with the same solvent as eluent. The first -250 mL of eluate was discarded and then the eluate was collected as 4 x 20 mL fractions and analyzed using TLC. The required product was the last substance to emerge; -500 mL of eluate was then collected. The solvents were removed to give a crystalline white solid; yield -5 g (22%). The compound was characterized using ' H andJ'P{'H} 123 References on page 133 Chapter 3 NMR spectroscopy (Table 3.1). Anal. Calcd for C25H20D2P2: C, 77.72; H, 5.74. Found: C, 77.70; H, 5.77. 3.4.2 Preparation of dpm(S)-d2 Dpm-d2 (100 mg, 0.259 mmol) and S8 (8.30 mg, 0.032 mmol) were dissolved in hot hexanes (-10 mL) and the solution was refluxed overnight. The solution was then cooled down to room temperature (R.T.; ~20°C) and white crystals subsequently formed (within several hours). The solid was collected and identified as dpm(S)-d2 using NMR spectroscopy (Table 3.1); yield ~50mg(46%). Anal. Calcd for C25H20D2P2S: C, 71.76; H, 5.30. Found: C, 71.71; H, 5.38. 3.4.3 Preparation of dpmMe(S) DpmMe (100 mg, 0.251 mmol) and Ss (8.00 mg, 0.031 mmol) were dissolved in hot ethanol (~ 5 mL) and the solution was refluxed briefly (-10 min). The solution was then cooled to R.T. and white crystals subsequently formed (within 1 h). The solvent was removed leaving a white solid above an oily, white film. The solid was gently scraped off, dried in vacuo, and identified as dpmMe(S) using NMR spectroscopy (Table 3.1); yield - 30 mg (28%). Anal. Calcd for C26H24P2S: C, 72.54; H, 5.62. Found: C, 71.89; H, 5.54. 124 References on page 133 Chapter 3 3.4.4 Preparation of dpm(S)2, dpe(S), dpe(S)2, dpp(S), and dpp(S)2 Syntheses of these disphosphine mono- and di-sulfides were carried out by refluxing the appropriate diphosphine with sulfur in hexanes. For example, disulfides were prepared by refluxing dpm (100 mg, 0.26 mmol), dpe (100 mg, 0.25 mmol), or dpp (100 mg, 0.24 mmol) and sulfur (16 mg, 0.52 mmol) in hexanes (10 mL) for 2 h. Monosulfides were similarly prepared using 8 mg (0.025 mmol) sulfur; a mixture of products (mono- and di-sulfides) mcluding unreacted diphosphine was typically seen. Removal of solvent by rotary evaporation gave white solids which were dried in vacuo prior to thin-layered chromatographic analyses (eluant: 1:2 = CH2CI2 : hexanes by volume) and NMR spectroscopic analyses in CDCI3. Dpm(S)2: ' H NMR (20 °C, CDCI3): 5 7.0 - 8.0 m (20H, Ph), 5 3.98 t (2H, CH 2 , J P H = 14.0 Hz); 3lV{lU} NMR (20 °C, CDCI3): 8 35.3 s. Dpe: ! H NMR (20 °C, CDCI3): 8 7.0 - 8.0 m (20H, Ph), 8 2.09 m (4H, CH 2); 31P{ !H} NMR (20 °C, CDCI3): 8 -12.6 s. Dpe(S): J H NMR (20 °C, CDCI3): 8 7.0 - 8.0 m (20H, Ph), 8 2.47, 2.28 m (4H, CH 2); ^Pf/H} NMR (20 °C, CDCI3): 8 44.3, -12.9 AB pattern (JPP = 47.5 Hz). Dpe(S)2: ! H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (20H, Ph), 8 2.71 p d (4H, CH 2); ^Pf/H} NMR (20 °C, CDCI3): 8 44.3 s. Dpp: ' H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (20H, Ph), 8 2.18 m (4H, CH 2), 8 1.59 m (2H, CH 2); ^ P J ' H } NMR (20 °C, CDC13): 8 -17.5 s. Dpp(S): ! H NMR (20 °C, CDCI3): 8 7.0 - 8.0 m (20H, Ph), 8 2.56 m (2H, CH 2), 8 1.99 m (2H, CH 2), 8 1.76 m (2H, CH 2); ^ Pj'H} NMR (20 °C, CDC13): 8 41.8 s, 8 -18.9 s. Dpp(S)2: *H NMR (20 °C, CDCI3): 8 7.0 - 8.0 m (20H, Ph), 8 2.56 m (4H, CH 2), 8 2.00 m (2H, CH 2); NMR (20 °C, CDCl 3):8 41.8s. 125 References on page 133 Chapter 3 3.4.5 Preparation of H2S H 2S was separately prepared for studies of reaction 3.1.3 (see Section 3.4.10). The experimental setup is schematically shown in Fig. 3.27. To a stirred slurry of CaS (2.0 g, 0.028 mol) in distilled H 20 (10 mL) was added aq. HCl (6.0 M , 9.2 mL, 0.056 mol) dropwise via syringe over a period of 10 min. The resulting, instantly formed H 2S was dried using CaS04/P2C»5 prior to bubbling through CHCI3. 3.4.6 Preparation of Pd2Br2(dpm-d2)2-2H20 (4b) PdCl2(PhCN)2 (0.41 g, 1.01 mmol), Pd2(dba)3 CHCI3 (0.55 g, 0.53 mmol), and dpm-d2 (0.82 g, 2.1 mmol) were refluxed in CDCI3 (50 mL) for 30 min. After being cooled, the resulting red solution was filtered to remove any insoluble materials, and the filtrate reduced in volume to -10 mL. The yellow-orange product, which precipitated after the addition of Et20 (25 mL), was then filtered, washed with acetone (2x10 mL) to remove any Pd(JJ) monomer, and dried in vacuo. The yield of the product, Pd2Cl2(dpm-d2)2 is -1.0 g (90%). Some of this complex (0.25 g, 0.24 mmol) was re-dissolved in -10 mL of CDCI3 and a solution of NaBr (0.25 g, 2.4 mmol) in aqueous methanol added (-5 mL CH 3 OD : -0.5 mL D20). The resulting solution was filtered, and concentrated under vacuum until red crystals formed. Et20 (40 mL) was added to complete the precipitation. The solid was filtered and quickly washed with aqueous methanol (2x10 mL) and Et20 (2x10 mL) and then dried in vacuo; yield 0.26 g (96%). 4b was characterized using NMR spectroscopy (Table 3.1). Anal. Calcd for C5oH40Br2D4P4Pd2-2H20: C, 50.84; H, 4.06. Found: C, 50.75; H, 3.83. (The presence of H 2 0 was evidenced by *H NMR). 126 References on page 133 Chapter 3 Fig. 3.27. Schematic diagram showing experimental setup for preparation of H2S from reaction of CaS with aq. HCl. syringe containing 9.2 mL 6.0 M H C l slurry of 2.0 g CaS in 10 mL H 2 0 rubber septum-sealed Schlenk tube cannula drying tube containing C a S O ^ O s magnetic stirrer and stir bar Erlenmeyler flask containing CHCI3 127 References on page 133 Chapter 3 3.4.7 Preparation of Pd2Br2(p-S)(dpm-d2)2 (8b) Pd2Cl2(dpm-d2)2 (0.50 g, 0.48 mmol) (see above) was dissolved in CDC13 (50 mL) and H 2S gas was bubbled through the solution for 20 min at R.T.; the colour changed from orange-red to brown with accompanying precipitation of a brown solid that was completed by gradual addition of Et20 (50 mL). The product, Pd2Cl2(u-S)(dpm-d2)2, was filtered, washed successively with acetone (2 x 10 mL), and Et20 (10 mL), and then dried in vacuo; yield 0.50 g (97%). The solid was re-dissolved in CDCI3 (50 mL) and a solution of NaBr (0.5 g, ~ 5 mmol) in aqueous methanol (10 mL CH3OD : 1 mL D20) was added. The resulting mixture was filtered, and concentrated under vacuum until brown crystals formed. Et20 (30 mL) was added to complete the precipitation. The solid was filtered, washed with aqueous methanol (2x10 mL) and Et20 (2x10 mL), and then dried in vacuo; yield 0.52 g (95%). 8b was characterized by NMR spectroscopy (Table 3.1). Anal. Calcd for C5oH4oBr2D4P4Pd2S: C, 50.95; H, 3.77. Found: C, 50.27; H, 3.80. 3.4.8 Kinetic measurements The kinetics of the reconversion reaction (2 -> 1) in CHCI3 (for X = CI, Br, I) were monitored spectrophotometrically in a thermostated Perkin-Elmer 552A instrument using quartz cells of path lengths of 1.0 cm fitted with a rubber septum. A 2.00 mL solution of dpm of appropriate concentration was placed in the cell and thermostated at the required temperature (20 -35 °C); the complex 2 was then injected as a solution (0.125 mL). The cell was shaken to ensure complete mixing prior to monitoring optical density changes at some appropriate, fixed 128 References on page 133 Chapter 3 wavelength. The concentration of 2 ranged from (0.81 -13.0) x 10"5 M, and that of dpm from (6.5 - 26.1) x 10"3 M ; thus pseudo-first-order conditions were maintained and standard log(absorbance difference) vs. time plots gave excellent linearity for at least 2.5-3 half-lives, from which the pseudo-first-order rate constants, kobs, were readily evaluated. 3.4.9 Mechanistic studies a) NMR-scale. Variable temperature and equilibrium studies were performed on the NMR-scale by reacting complexes with appropriate amounts of diphosphine. Thus, for example, lb (10 mg, 0.009 mmol) or 2b (10 mg, 0.0085 mmol) was reacted at R T . in CDCh or CD 2C1 2 (0.5 mL) with dpm-d2 or dpmMe in either a 1:lor a 1:5 mole ratio (in 0.1 mL solution). The samples were analyzed using ! H and 31P{1H} NMR spectroscopy immediately and after periods of time up to 72 h. Under low temperature conditions, accurately weighed out samples of 2a, 2b, or 2c were dissolved in the appropriate solvent (~ 0.5 mL) in a septum-sealed NMR tube which was placed in a dry ice/acetone slush bath (-78 °C). A 0.1 mL solution of dpm (5 mol. equiv. excess) was then injected and the shaken sample was analyzed immediately and after periods of time up to 72 h. The temperature was then raised in 20° increments to a final 20 °C and the spectra recorded at each temperature. b) Synthetic-scale. The bromide 2b (100 mg, 0.085 mmol) was reacted with either dpm-d2 (32.9 mg, 0.085 mmol) or dpmMe (34.0 mg, 0.085 mmol) in 20 mL of CH2C12 or CHC13, the reaction being monitored by thin-layer chromatography and/or UV-vis spectroscopy. Employing Schlenk techniques, the experiments were conducted on a scale such that products could be isolated in 129 References on page 133 Chapter 3 sufficient amounts for analyses of both the Pd complexes and the dpm compounds. Upon completion of the synthesis (approx. 2 h), the solution was reduced in volume to ~ 5 mL before 20 mL of Et 20 was added to precipitate the palladium product(s). The filtrate was passed through a column of silica to remove all traces of metal compounds with the eluate collected. The solvent was rotovaped off to give a colourless oil, which was dissolved in minimal hot ethanol; cooling in a dry ice/acetone bath induced precipitation of a white solid, the diphosphine monosulfide. The ethanol solvent was slowly removed under vacuum and at low temperatures to avoid reformation of the oil. All products were dried in vacuo prior to analyses using ! H and 31P{1H} NMR spectroscopy in CDCI3. c) Reaction of 2b with dpe, dpp, dpm(S), and dpm(S)2. Reaction of 2b with other diphosphines was explored for possible sulfur abstraction. In a typical experiment, 2b (10 mg, 0.0085 mmol) and the appropriate amount of diphosphine {i.e. 3.4 mg dpe, 3.5 mg dpp, 3.5 mg dpm(S), 3.8 mg dpm(S)2; all diphosphines 0.0085 mmol) were placed in an NMR tube fitted with a PTFE J. Young valve, and CDCI3 (-0.5 mL) was vacuum transferred in. All samples were analyzed using NMR spectroscopy immediately and after periods of time up to 72 h. For reactions with dpm(S), 2a (9.2 mg, 0.0085 mmol) and 2c (11 mg, 0.0085 mmol) were also used with analyses being carried out analogously. In the above reactions with dpm(S), the solvent was removed by rotary evaporation (after completion of the reaction as evidenced by the disappearance of 2 and dpm(S) signals from NMR measurements) and the resulting residue was analyzed using NMR and UV-vis spectroscopy and mass spectrometry. Microanalysis was performed for the residue isolated from the 2b + dpm(S) reaction. Calcd for Pd2Br2(u-S)(dpm)2»dpm(S), C 7 5H66Br 2P 6S 2Pd 2: C, 56.66; H, 4.18. Found: C, 130 References on page 133 Chapter 3 56.93; H, 4.40. Attempts to grow crystals of this species for X-ray crystallographic analysis, for example by diffusion of hexanes (10 mL) or Et20 (10 mL) into a 6 mL CH2CI2 solution of 10 mg of the complex at R.T. or -40 °C, were unsuccessful. 3.4.10 Catalytic conversion of H2S The catalysis of reaction 3.1.3 was briefly studied at R T . using lb and 2b. The conditions used are outlined in Table 3.3; the concentrations of H 2S used were determined from knowledge of Henry's constant (1.3 M/atm)17 and vapour pressure data (see Appendix IV). 1 8 In a typical experiment, appropriate amounts of lb or 2b and dpm were placed in a rubber septum sealed Schlenk tube (volume = 37.0 or 165.0 mL) and CHCI3 was vacuum transferred using liquid N 2 . The reaction mixture was thawed before H 2S was introduced. For studies at 1 atm, [H2S] = 1.0 M. Low concentrations of H 2S were achieved by injecting the appropriate volume (see Appendix IV for sample calculations). Concentrations of lb ranged from (8.76 - 263) x 10"5 M, that of 2b, 2.56 x 10"3 M , and those of dpm (3.64 - 52.0) x 10"3 M. The reaction was allowed to proceed for periods of 5, 24, or 72 h before the contents were analysed. The head space was analysed for hydrogen using gas chromatography. UV-vis spectra were obtained for qualitative assessment of Pd species. For reaction mixtures with high catalyst concentrations (i.e. > 8.76 x 10"4 M), Et^O (20 mL) was added to precipitate the Pd species which were then filtered, washed with Et20 (2 x 10 mL), dried in vacuo, and subsequently analysed using NMR and UV-vis spectroscopy and microanalysis. (Calcd for Pd2Br2(dpm)2(u-dpm(S)) '2Cft2C\2, C75H66Br2P6SPd2*2CH2Cl2: C, 53.53; H, 4.08. Found: C, 53.53; H, 4.00 (the presence of CH2C12 was evidenced by ' H NMR spectroscopy). Crystals of this species, obtained by diffusion of hexanes (10 mL) into a 6 mL 131 References on page 133 Chapter 3 CH2CI2 solution of 10 mg of the complex at R.T., were subjected to X-ray analysis but the structure could not be deterrnined.) The filtrate was passed through a column of silica to remove all traces of metal compounds with the eluate collected. The solvent was removed by rotory evaporation giving a white solid and/or colourless oil; analyses of these phosphine products were done using NMR spectroscopy. Percent conversion of dpm to dpm(S) was deterrnined by comparison of the respective integrated areas. A single in situ NMR experiment was carried out using an NMR tube (volume = 3.7 mL) fitted with a PTFE J. Young valve. Complex lb (0.0014 g, 0.0012 mmol) and dpm (0.0049 g, 0.013 mmol) were placed in the NMR tube and CDCI3 (3.5 mL) was vacuum transferred using liquid N2. H2S was subsequently introduced to achieve a total pressure of 1 arm. The mixture was then left at R T . for periods of time up to 72 h before being analyzed using NMR spectroscopy. 132 References on page 133 Chapter 3 3.5 References for Chapter 3 (1) Lee, C-L . ; Besenyei, G . ; James, B. R.; Nelson, D. A . ; Lilga, M . A. J. Chern. Soc, Chern.Commun. 1985, 1175. (2) Barnabas, A. F.; Sallin, D . ; James, B. R. Can. J. Chern. 1989, 64, 2009. (3) Besenyei, G . ; Lee, C-L . ; Gulinski, J.; Rettig, S. J.; James, B. R.; Nelson, D. A . ; Lilga, M . A. Inorg. Chern. 1987, 26, 3622. (4) Wong, T . Y . H . ; Barnabas, A.F . ; Sallin, D . ; James, B.R. Inorg. Chern. 1995, 34, 2278. (5) Barnabas, A. F., M . Sc. Dissertation, University of British Columbia, Vancouver, 1989. (6) (a) Grim, S. O. ; Mitchell, J. D. Syn. React. Inorg. Metal-Org. Chern. 1974, 4,221. (b) Grim, S O . ; Walton, E. D. Inorg. Chern. 1980, 19, 1982. (7) Lyke, S. E . ; Lilga, M . A . ; Ozanich, R. M . ; Nelson, D. A . ; James, B. R.; Lee, C-L. Ind. Eng. Chern. Prod. Res. Dev. 1986, 25, 517. (8) Lee, C-L . ; Yang, Y. P.; Rettig, S. J.; James, B. R.; Nelson, D. A . ; Lilga, M . A. Organometallics 1986, 5, 2220. (9) James, B. R.; Mahajan, D. Can. J. Chern. 1979, 57, 180. (10) (a) Al-Salem, N. A . ; Empall, H. D. ; Markham, R.; Shaw, B. L ; Weeks, B. J. Chern. Soc, Dalton Trans. 1979, 1972. (b) Shaw, B. L. In Catalytic Aspects of Metal Phosphine Complexes, Adv. in Chern. Series: Alyea, C , Meek, D.W., Eds.; Am. Chern. Soc: Washington, 1982; Vol . 196, Ch. 6. (11) Ball, R. G . ; Domazetis, G. ; Dolphin, D . ; James, B. R.; Trotter, J. Inorg. Chern. 1981, 20, 1556. (12) Keiter, R. L . ; Rheingold, A. L . ; Hamerski, J. J.; Castle, C. K. Organometallics 1983, 2, 1635. (13) (a) Brown, M . P.; Yavari, A . ; Hill, R. H . ; Puddephatt, R. J. J. Chern. Soc, Dalton Trans. 1985, 2421. (b) Azam, K. A . ; Brown, M . P.; Hill , R. H . ; Puddephatt, R. J.; Yavari, A . Organometallics 1984, 3, 697. (14) Hunt, C. T. ; Balch, A. L. Inorg. Chern. 1981, 20, 2267. 133 Chapter 3 (15) Appleton, T. G . ; Clark, H . L . ; Manzer, L. E. Coord. Chem. Rev. 1973, 10, 335. (16) Luther III, G. W.; Beyerle, G. Inorg. Synth. 1977, 77, 186. (17) Fogg, P.G.T. ; Young, C L . (Eds.). Solubility data series. Vol. 32. Pergamon Press, Oxford. 1988. p.279. (18) Handbook of Chemistry and Physics. 56th ed. Chemical Rubber Co., Cleveland, O H . 1979. p. D -191. (19) Sallin, D . ; Barnabas, A.F . ; James, B.R. Unpublished data. 134 CHAPTER 4 Kinetic and Mechanistic Aspects of Sulfur Abstraction from Pd2X2(p-S)(dpm)2 Using Halogens 135 Chapter 4 4.1 Introduction During the course of this thesis work, it was discovered that the bridged sulfur atom of 2 could be removed effectively using halogens (eq. 4.1.1).1 The chemistry parallels that of reactions between 1 and X 2 previously studied by Hunt and Balch2 in that transannular oxidative addition occurs with the formation of tetrahalodipalladium(IJ) intermediates en route to production of mononuclear species (eq. 4.1.2). Kinetic and mechanistic details of reaction 4.1.1 were performed and are summarized in this chapter; there is a correlation of the results with the limited semi-quantitative kinetic data reported for reaction 4.I.2.2 (4.1.1) Pd2X2(dpm)2(u-S) (2) + X 2 -> Pd2X4(dpm)2 (10) + sulfur i 2 PdX2(dpm) (9) (4.1.2) Pd2X2(dpm)2 (1) + X 2 Pd2X4(dpm)2 -> 2 PdX2(dpm) Numerous examples of electrophilic additions of halogens to transition metal complexes can be found in the literature including studies detailing kinetic and mechanistic investigations. For example, the mono-, di-, and trinuclear transition metal carbonyl complexes of group 6, 7, and 8 elements have been shown to undergo facile oxidative addition reactions. (4.1.3) M(CO) 5 + X 2 -> M(CO)4X 2 + CO M = Fe, Ru, Os refs. 3, 4 (4.1.4) 2 M(CO) 6 + 2 X 2 -> [M(CO)4X(u-X)]2 + 4 CO M = Mo, W ref. 5 (4.1.5) M 2(CO)io + X 2 -> 2 M(CO) 5X M = Mn, Tc, Re refs. 6, 7, 8 (4.1.6) Fe3(CO)i2 % Fe2(CO)8I2 2Fe(CO)42 ref. 9 (4.1.7) Ru3(CO)i2 S Ru(CO)4X2 + Ru 2(CO) 8X 4 + Ru3(CO)i2X6 ref. 10 (4.1.8) Os3(CO)i2 + X 2 -> Os3(CO)i2X2 ref. 11 136 References on page 189 Chapter 4 With the di- and trinuclear complexes, transannular oxidative addition takes place such that the products contain equal numbers of halide ligands at each of the reacted metal centers. There is increasing metal-metal bond stability towards oxidation on descending the transition metal group, Fe, Ru, and Os, and di- and trinuclear halide products are prominent with the heavier metals. An X-ray crystallographic study of 0s 3(C0)i 2l2 has revealed a linear Os3 entity as illustrated and thus demonstrating ring opening upon oxidative addition across two Os centers.12 \ < l \ / O C — O s - O s - O s - C O Kinetic and mechanistic studies have been carried out for reactions 4.1.3 1 3 and 4.1.5 1 4 including studies on a series of substituted dinuclear species (eqs. 4.1.9 and 4.1.10). (4.1.9) M 2(CO)io- xLx + X 2 -> 2 M(CO)(1o-x)/2Lx/2X ref. 15 M , M = Mn,Mn; Mn,Re; Re,Re L = phosphine; x =1 ,2 (4.1.10) M 2 (CO ) 8 L + I2 -> L[M(CO)4l] 2 ref. 16 1 I I 1 M , M = Mn,Mn; Mn,Re; Re,Re L = Me 2 AsOCAsMe 2 CF 2 CF 2 , Ph 2 POCPPh 2 CF 2 CF 2 Similarly, halogenation of substituted dinuclear Fe and trinuclear Ru carbonyl complexes have been examined kinetically (eqs. 4.1.11 and 4.1.12). (4.1.11) {Fe(7t-RC5H4)(CO)2}2 + X 2 -> 2 Fe(7i-RC5H4)(CO)2X ref. 17 (4.1.12) Ru 3(CO) 9L 3 + 3 X 2 -> 3 Ru(CO) 3LX 2 L = phosphine ref. 18 Reports on halogenation reactions involving main group metals can also be found and these center on the use of Sn (eqs. 4.1.13 and 4.1.14). (4.1.13) (aryl)SnR3 + I2 -> (aryl)I + ISnR3 ref. 19 137 References on page 189 Chapter 4 (4.1.14) M(Cp)m(CO)„(SnMe3) + X 2 -> M(Cp)m(CO)„X + XSnMe3 ref. 20 M = Cr, Mo, W;m=l;n=3 M = Fe;m=l;n=2 M = Mn;m = 0;n=5 Within transition metal - dpm complexes, halogenations have been studied mostly with Rh complexes containing one or two bridging dpm ligands (eqs. 4.1.15 to 4.1.22); however, to date no kinetic reports can be found concerning such reactions of transition metal - dpm complexes. In general, metal - metal bond formation accompanies transannular oxidative addition; in the case of reaction 4.1.18 addition of I2 occurs at one Rh center, and this is attributed to the larger steric bulk of the T ligand, compared to Br" (eq. 4.1.17), in accommodating the steric requirements of the dpm ligand. (4.1.15) Rh2X2(CO)2(u-dpm or u-dam)2 + X 2 -> Rh2X4(CO)2(p-dpm or p-dam)2 dam = bis(diphenylarsino)methane ref. 21 (4.1.16) Rh2(CNR)4(u-dpm)2 + X> -> R h ^ C N R ^ ^ p - d p m ^ ref. 21a (4.1.17) Rh2(Cp)2(p-CO)(u-dpm) + 2 Br2 -> Rh2(Cp)2Br4(u-dpm) + CO ref. 22 (4.1.18) Rh2(Cp)2(u-CO)(u-dpm) + I2 -> Rh2(Cp)2I2(CO)(Mpm) ref. 22 (4.1.19) Rh2(p-dmpz)(CO)2(p-dpm)2+ + I2 -> Rh2(n-dmpz)I2(CO)2(p-dpm)2+ ref. 23 dmpz = 3,5-dimethylpyrazolate (4.1.20) Co2(CO)4(p-CO)2(u-dpm) + h. + dpm [Co2(CO)2(p-I)(p-CO)(p-dpm)2]I + 3 CO ref. 24 (4.1.21) Pd2(p-dpm)3 + I2 Pd2I2(p.-dpm)2 + dpm ref. 2 h (4.1.22) Ru3(CO)8(p-dpm)2 Ru2I2(CO)4(p-dpm) + RuI2(CO)2(dpm) ref. 25 Other examples of halogen addition to one metal center in dinuclear complexes include Rh compounds containing bridging bis(diphenylphosphino)propane (dpp) and bis(diphenylphosphino)butane (dpb) ligands (eqs. 4.1.23 and 4.1.24). Addition across the two 138 References on page 189 Chapter 4 metal centers cannot take place as they are too far apart (4 to 6 A). The products have not been definitively characterized and may be polymers; infrared spectroscopy was used to establish the nature of the reaction as the change in the vCo was consistent with oxidative addition at one metal center (the greater change in oxidation state (I to in vs. I to U) results in a lesser degree of backbonding to the TI* molecular orbitals of CO). (4.1.23) Rh2Cl2(CO)2(p-dpp)2 + h -> [RhCl(CO)I2(dpp)]n n = 3-4 ref. 21b (4.1.24) Rh2Cl2(CO)2(u-dpb)2 + h -*• [RhCl(CO)I2(dpb)]n ref. 21b Double transannular oxidative addition can also take place, and in reaction 4.1.25 consecutive metal - metal bond forming and breaking occurs. (4.1.25) Au2(p-(CH2)2P(CH3)2)2 + X 2 Au2X2(p-(CH2)2P(CH3)2)2 ref. 26 •i X 2 Au2X4(p-(CH2)2P(CH3)2)2 There is only one report describing the oxidative addition of iodine to a metal sulfide complex with concomitant formation of elemental sulfur. Again, as with reactions 4.1.23 through 4.1.25, no kinetic or mechanistic studies have been reported. (4.1.26) Os(n2-S2)(CO)2(PPh3)2 + I2 -> OsI2(CO)2(PPh3)2 + sulfur ref. 27 139 References on page 189 Chapter 4 4.2 Results Complexes Pd2X2(dpm)2 (1), Pd2X2(u-S)(dpm)2 (2), and PdX2(dpm) (9) are well-characterized. NMR spectroscopic data were given in earlier chapters, but for the convenience of the reader, the data are again summarized in Table 4.1. Kinetic data can be found in Appendix V. A preliminary *H and 31P{ *H} NMR study of reaction 4.1.1 (X = I) in CDC13 at R T . revealed, in general, a well-behaved system (Fig. 4.1) (the formation of small amounts of PdI2(dpm(S)) (11c) will be discussed later (p. 152)). On addition of a purple iodine solution to an equimolar brown solution of 2c, the 31P{1H} NMR singlet of 2c rapidly decreased in intensity while a new 31P{1H} NMR singlet, corresponding to an observed green-black intermediate, at 8 -1.2, concomitantly increased rapidly in intensity. This new singlet subsequently decreased slowly in intensity while the 3IP{1H} NMR singlet of 9c was observed to form and grow in intensity. Completion of the reaction (~lh) was evidenced by the eventual disappearance of the intermediate and the unchanging 9c singlet. The intermediate is also characterized by a 1:4:6:4:1 quintet at 8 5.14 in the ! H NMR spectrum and is identified as the tetraiododipalladium(IJ) (10c) species previously seen in the reaction between Pd2I2(dpm)2 (lc) and I2.2 A repeated NMR study of this reaction served to establish firmly the NMR chemical shifts of the 10c intermediate; the JPH coupling was observed to be 3.7 Hz. The final orange-red solution of 9c contained a very small amount of yellow particulate matter (~0.1 mg), attributed to elemental sulfur from thin-layered chromatographic studies (CHCI3 eluent), and which prompted a synthetic-scale study to search for and substantiate the production of elemental sulfur. Reaction of 2c and I2 on the synthetic-scale yielded immediately an initial green-black solution that slowly changed colour to a final orange-red (~3h). A pale yellow precipitate was formed which was collected (~1 mg) and analyzed using 140 References on page 189 Chapter 4 Table 4.1. NMR data for the palladium complexes. Compound" 8 ("P^H})" Pd 2Cl 2(dpm) 2 (la) 4.17 (4.0) -5.5 d Pd 2Br 2(dpm) 2 (lb) 4.19 ^ (4.0) -6.15 d 4.24' -5.5 Pd2I2(dpm)2 (lc) 4.23 d ' e (4.0) -11.3 d Pd2Cl2(u.-S)(dpm)2 (2a) 2.79 ^(12.6, 3.5) 5.52* 4.73 f (12.6, 6.1) Pd2Br2(u-S)(dpm)2 (2b) 2.88 d f (12.8, 3.2) (2.90) 5.96 d 4.83 /(12.8, 7.6) 6.14 Pd2I2(n-S)(dpm)2 (2c) 3.06 d f (14.0, 3.0) 6.08 d 4.95 f (14.0, 6.0) PdCl2(dpm) (9a) 4.21 g (10.8) -54.7 PdBr2(dpm) (9b) 4.37 s (10.5) -56.2 Pdl2(dpm) (9c) 4.42 * (10.0) -63.2 Pd 2Cl 4(dpm) 2 (10a) 4.55 (4.6) 8.5 ** Pd 2Br 4(dpm) 2 (10b) 4 .6** 4.90 d M Pd2I4(dpm)2 (10c) 5.14e (3.7) -1.52 5.18 ^ (4.0) -0.73 d M PdCl2(dpm(S)) (11a) 4.07' 54.9, 30.9 (18.3)' PdBr2(dpm(S)) (lib) 4.02 ' 56.6, 32.1 (20.4)' PdI2(dpm(S)) (11c) 3.71' 61.1,31.2(25.5)-' Pd2I2(^-SO)(dpm)2 4.98* (12, 13.5) 19 to -7 A A ' B B ' 4.18* (15, 9) 2.58' 2.26' PdCl2(dpm(S)2) 5.54 s (14) 37.3 ° The u-symbol for the bridging diphosphine ligand(s) is omitted for convenience throughout this Table and the text. * In CDC1 3 , unless stated otherwise, at 20 °C with respect to T M S ; JHH and/or J P H values in Hertz are given in parentheses; signals for C H 2 protons. 0 Unless otherwise stated, singlets in CDC1 3 at 20 °C with respect to 85% H3PO4, downfield being positive. ' ' i n C D ^ . 'Quintet. 'Doublets of quintets 141 References on page 189 Chapter 4 for each of 2 sets of C H 2 protons. 8 Triplet. * Taken from ref. 2. ' Pseudo-triplet. 3 A B pattern, J P P values in Hertz given in parentheses. * Doublets of tripets for each of 2 sets of C H proton. ' Multiplet for each of 2 sets of C H proton. 142 References on page 189 Chapter 4 Fig. 4.1. Preliminary ! H (300 MHz) and 31P{!H} (121 MHz) NMR study of the reaction of Pd2I2(u-S)(dpm)2 (2c, 1.6 x 10"2 M) with 1 mole equivalent I2 in CDC13 at R.T.; 9c = Pdl2(dpm), 10c = Pd2l4(dpm)2,11c = PdI2(dpm(S)). t = 0 min 2c A JL 2c 9c 10c = 5 min 9c 11c .1 . 1 1 t = 10 min 10c t = 20 min JL 11c t = 40 min L i . I I ' > 5 I I I I I I — 4 3 P P M t = 30 min J L t = 50 min Ho J _ L 20 -20 -60 P P M 143 References on page 189 Chapter 4 TLC, UV-vis spectroscopy, and mass spectrometry; the results were compared with those of an authentic sample of sulfur. Degrees of retention (Rf-values) as well as electronic absorption spectral characteristics were found to be identical; the mass spectrum of the yellow solid excellently illustrates an ionization pattern, indicative only of sulfur, with the major parent peak located at 256 m/z and the other major peaks all separated by 32 units of mass (Fig. 4.2). The identity of the yellow solid was further proven to be sulfur by its reaction with Pd2l2(dpm)2 (lc) which gave only 2c as evidenced by ! H and 31P{1H} N M R spectroscopy. Finally, with the identity known, the yield of elemental sulfur from the synthetic reaction was determined to be -80 %. Kinetic studies, (a) Conventional UV-vis spectrophotometric measurements. The kinetics of the slow decomposition of Pd2l4(dpm)2 (10c) were measured by means of conventional UV-vis absorption spectroscopy; the rates were independent of the initially added I2. Species 2c is brown and has absorption maxima at 368 and 485 nm, while 9c is orange and has a single absorption maximum at 429 nm in the visible region. Treatment of a CHCI3 solution of 2c with a solution of I2 immediately gave a green-black solution which gradually changed to a final yellow-orange. An illustrative example of accompanying spectral changes is shown in Fig. 4.3 where the spectrum recorded immediately after the two reactants have mixed (t = 30 s, dashed line) shows an absorption maximum at 362 nm with a shoulder at 484 nm. Below 420 nm and above 484 nm, the absorbance rapidly decreases while simultaneously increasing rapidly in the region between such that the next spectrum recorded (t = 150 s) corresponds to the stage of the reaction where 10c has fully formed and is now decomposing. The spectral changes accompanying decomposition of 10c are also characterized by two isosbestic points at 420 and 510 nm. Optical density was monitored as a function of time at 396 nm where the total spectral change is -1 absorbance unit. Iodine and 144 References on page 189 Chapter 4 Fig. 4.2. Mass spectrum (electron impact) of the yellow precipitate isolated from the synthetic-scale reaction of Pd2I2(u-S)(dpm)2 (2c) with I2 in CHC13 at R T . 2 2 4 S ' 256 S 8 p-iti111 r11 n111111'I'1111111111[ 111111111111111111111111111111111111 n 111111111111111 u 1111| iI'ITT 220 260 300 340 380 m/z 64 S 2 32 S 160 S 5 96 S 3 128 S 4 192 S 6 1111 I'I'I'I u i ' ITI 111'l'ivi i irp'i 11 ITITI 1111 I'I 11111 n i l l i i i T - [ m n i rl i j 11 11111111 n 111111 40 80 120 160 200 l i i i i i m i m m/z 145 References on page 189 Chapter 4 References on page 189 Chapter 4 sulfur have no absorptivities at this wavelength (Fig. 4.4). The first-order rate constants, kobs, obtained were found to be independent of both [I2] from (3.8-7.5)xl0"4 M (Fig. 4.5) and [2c] from (1.6-9.9)xl0"s M. Noting that 2c is rapidly converted fully to 10c, the rate law takes the simple form (4.2.1) rate = _ d P d 2Udpm) 2 ] = k ^ r p ^ ^ ] = kD[Pd 2I 4(dpm) 2] dt where ko is the unimolecular rate constant for the decomposition of the intermediate. The temperature dependence data for kD were obtained at a single [I2] of 7.5xl0"4 M and [2c] = 9.4 x 10"5 M ; an Eyring plot of the data (Table 4.2) from 20-35 °C gives an excellent straight line and the activation parameters A H D * = 80 ± 1 kJ mol"1 and ASD* = -26 + 3 J mol"1 K"1 (Fig. 4.6). (b) Stopped-flow spectrophotometric measurements. The kinetics of the initial rapid reaction between 2c and I2 in forming the intermediate 10c were measured in CHCI3 using stopped-flow spectroscopy under pseudo-first-order conditions. Inspection of the spectral results from conventional UV-vis measurements (for example, see Fig. 4.3) led to the choice of 510 nm as the appropriate wavelength to monitor the reaction as the decrease in absorbance here eventually ceased (isosbesticity was then observed) indicating complete formation of 10c. Optical density changes were monitored as a function of time, and a representative plot of absorbance vs. time is shown in Fig. 4.7 with the corresponding pseudo-first-order rate-plot shown in Fig. 4.8. The observed pseudo-first-order rate constants, kobs, were found to be strictly first-order with respect to 147 References on page 189 Chapter 4 Fig. 4.4. Electronic absorption spectra of Sg (cone. = 1.25 x 10"3 M ; X, nm (e, M" 1 cm'1) = 244 (828), 266 (910)) and I2 (cone. = 3.94 x l O ^ M ; X, nm (e, M" 1 cm"1) = 512 (879)) in CHC13 at R T . 1.6 190 240 290 340 390 440 Wavelength, nm 0.4 Wavelength, nm 148 References on page 189 Chapter 4 Fig. 4.5. The dependence of the first-order rate constant, kobs, on [I2] for the decomposition of the intermediate Pd2J4(dpm)2 (10c) (formed from reaction with 9.4 x 10"5 M Pd2l2(u,-S)(dpm)2 (2c))inCHCl3at25 °C. t 0.0025 T 0.002 + v 0.0015 4-0.001 + 0.0005 + 0.0003 0.0004 0.0005 0.0006 ft], mol L 1 1 0.0007 0.0008 Fig. 4.6. Eyring plot of the temperature dependence for the unimolecular rate constants, kD, for the decomposition of Pd2L,(dpm)2 (10c). t Note: use of [I2] = 3.8 x 10"4 M gave kinetics which analyzes excellently for first-order; the second-order rate-plot is non-linear. 149 References on page 189 Chapter 4 Table 4.2. Temperature dependence for the bimolecular (ki) and unimolecular (kD) rate constants for the formation and decomposition of the intermediate, Pd2l4(dpm)2 (10c), respectively, in CDCb." Temperature, K k ^ M V kD , s-1 293 179 2.2 x 10'3 298 235 (240)* 4.0 x 10"3 c 303 290 7.8 x 10"3 c 308 370 13 .1xl0" 3 c ° initial [2c] = 9.4 x 10"5 M . * Five-fold excess of "BojNI present. c Obtained at a single [I2] = 7.5 x 10"4 M . 150 References on page 189 Chapter 4 Fig. 4.7. Absorption spectral changes atA. = 510nmasa function of time for the formation of Pd2l4(dpm)2 (10c) in CHC13 at 25 °C from reaction of Pd2I2(u-S)(dpm)2 (2c) (9.9 x 10"5 M) with I2 (4.9 x If/4 M). 1.25 T Time, s Fig. 4.8. Rate-plot analyzed for a pseudo-first-order dependence on Pd2I2(p-S)(dpm)2 (2c) (initial cone. = 9.9 x 10'5 M) during reaction with I2 (initial cone. = 4.9 x lO^lVQinCHCh.at 25 °C to form the intermediate Pd2Lt(dpm)2 (10c); A t and A» represent the absorption at 510 nm at times t and oo, respectively. Time, s 0 5 10 15 20 25 30 35 0 i 1 1 1 1 1 1 1 151 References on page 189 Chapter 4 h ((3.8-7.5)xl0"4 M) and independent of [2c] from (1.6-9.9)xl0'5 M. Thus, the rate law is as follows (4.2.2) rate = - ^ 1 = k o b s [ 2 a ] = kt[I2][2a] dt where ki is the bimolecular rate constant obtained from a plot of the dependence of k^s on [I2] (Fig. 4.9). The temperature dependence data for ki were obtained similarly from the plots at other temperatures; an Eyring plot of the data (Table 4.2) from 20-35 °C gives an excellent straight line and the activation parameters A H / = 32 ± 1 kJ mol"1 and A S i * = -91 ± 3 J mol'1 K"1 (Fig. 4.10). No ionic role for T was found; for example, addition of a five-fold excess of "BiuNI gave an unchanged kj. value at 25 °C (Table 4.2). The kinetics measured here correlate well with observations from low temperature NMR studies (see below). Using the activation parameters, the rate constant for the formation of the intermediate at -42 °C is calculated to be 4.12 M 1 s"1 and the corresponding calculated ti/2 is 30.1 s. The intermediate was observed to be completely formed in ~3 min and this time period equates to six half-lives corresponding to -98 % conversion. F o r m a t i o n o f PdI 2 (dpm(S)) (11c). (a) V a r i a b l e t e m p e r a t u r e N M R - s c a l e studies. Closer examination of the preliminary NMR findings of the reaction between 2c and I2 at R T . reveals the presence of additional ! H and 3lV{lH} NMR resonances that are "unaccounted for" (Fig. 4.1). Specifically, a distinctive 1:2:1 triplet at 8 3.71 is observed in the ! H NMR spectrum with corresponding doublets at 8 61.1 and 8 31.3 in the 3 1 P { 1 H } NMR spectrum. These signals were 152 References on page 189 Chapter 4 Fig. 4.9. The dependence of the pseudo-first-order rate constants, kobs, for the formation of Pd2l4(dpm)2 (10c) on [I2] from reaction with Pd2I2(p-S)(dpm)2 (2c, 9.9 x 10"5 M) in CHC13. Fig. 4.10. Eyring plot of the temperature dependence for the bimolecular rate constants, ki, for the formation of Pd2Lt(dpm)2 (10c) from reaction of Pd2I2(p-S)(dpm)2 (2c)with I2 in CHC13. 153 References on page 189 Chapter 4 later identified as belonging to the mononuclear complex 11c which was separately synthesized and characterized. The ' H and 31P{1H} NMR spectra accordingly show the same resonances corresponding to the C H 2 protons and the two P atoms, respectively. As the two P atoms are inequivalent resulting in an AB pattern, the C H 2 proton resonance is more accurately described as a pseudo-triplet. This by-product was observed consistently in relatively small amounts; Table 4.3 summarizes the work carried out in part to determine the factors that govern the formation of 11c and to elucidate the nature of its origin. Similar NMR-scale experiments at R.T. under anaerobic and aerobic (1 arm air or 1 arm 02) conditions gave the same product distribution, 11c being present to -17 %. At this point, it is noted that re-analysis of the products from the reaction on the synthetic-scale revealed that 11c was also formed but to an extent of -5 %. Furthermore, when 2c and I2 were reacted at concentrations used in the UV-vis or stopped-flow spectroscopic studies (i.e. 10"4 M), 11c was not detected (after a work-up procedure). Variable temperature NMR-scale studies were performed, and at temperatures including 0, 40, and 80 °C, 11c was observed to form in 15, 25, and 21% yield, respectively. At -42 °C, however, 11c was not formed (both 2c and I2 initially at -42 °C) and its absence prompted mechanistic studies to determine the fate of the sulfur atom when initially released from 2c (see next section). At this temperature, the intermediate 10c was observed to be formed completely within 3 min, and no subsequent decomposition was seen; furthermore, no other observable intermediates were detected. When the temperature of the same sample was raised to -15 °C, the intermediate began to decompose to 9c, and again no other detectable species were seen in the *H and 31P{'H} NMR spectra. Finally, the temperature was raised to R.T., and 9c was again the only product seen in the NMR spectra; a yellow solid (-0.1 mg) seen in the NMR sample was identified using UV-vis spectroscopy and thin-layered 154 References on page 189 Chapter 4 Table 4.3. Reaction of Pd2X2(u-S)(dpm)2 (2) with X 2 under various conditions; yield of by-product PdX2(dpm(S)) (11). Reaction " Conditions b PdX2(dpm(S)) yield (%) Pd2I2(n-S)(dpm)2 + I2 NMR-scale; chloroform 1.6 x 10" 2 M; 22 °C; 1 a tmN 2 1.6 x 10~ 2M; 22 °C; 1 atmair 17, 18, 21 17 1.6 x U T 2 M ; 22 °C; 1 a t m 0 2 17 1.6 x 10" 2 M; -42 °C; 1 a U n N 2 not detected 1.6 x 1 0 " 2 M ; 0 ° C ; l a t m N 2 15 1.6 x 10" 2 M; 40 °C; 1 a tmN 2 25 1 . 6 x l O " 2 M ; 80 °C; l a t m N 2 21 Pd2I2((i-S)(dpm)2 + 0.5 I2 c NMR-scale; chloroform 1.6 x 10"2 M ; 22 °C; 1 atm N 2 25 d Pd2I2(n-S)(dpm)2 + 2 I2 NMR-scale; chloroform 1 . 6 x l O ' 2 M ; 22 °C; l a t m N 2 not detected Pd2I2(n-S)(dpm)2 + I2 NMR-scale; acetonitrile 8 x 10"4 M ; 22 °C; 1 atm N 2 32 Pd2I2(n-S)(dpm)2 + I2 synthetic-scale; chloroform 2.0 x 10"3 M ; 22 °C; 1 atm N 2 5 Pd2I2(^-S)(dpm)2 + I2 kinetic studies-scale; chloroform 9.9 x 10"4 M ; 22 °C; 1 atm N 2 not detected Pd2I2(^-SO)(dpm)2 + I2 NMR-scale; chloroform 1.6 x 10" 2 M; 22 °C; l a t m N 2 35 Pd2Br2(u-S)(dpm)2 + Br 2 c NMR-scale; chloroform 1 . 7 x l O " 2 M ; 22 °C; l a t m N 2 33 d " Unless otherwise stated, the reactions are 100% complete, and were studied after 24 h. * Concentrations refer to the Pd complex. 0 Reaction was incomplete but, once the halogen was consumed, there were no further changes. d Yield relative to PdBr2(dpm) (9b) or Pdl2(dpm) (9c). 155 References on page 189 Chapter 4 chromatography as elemental sulfur. No 11c was formed if this 'farming up" procedure was monitored from -42 to -22 °C. To determine the ease with which 11c forms during the reaction of 2c with I2, NMR samples were prepared by the simultaneous injection of equal volumes of equimolar CDCb solution of 2c and I2 initially at R.T. into NMR tubes held at -42 or -78 °C. The low temperatures either stopped or slowed down the reaction, which was examined by ' H and "Pf/H} NMR spectroscopy at the respective temperatures; in both samples, 9c and 11c were seen immediately in very small amounts (< -1 %). There was no further change in spectral characteristics for the "solidified" sample analyzed at -78 °C; the sample analyzed at -42 °C, however, continued to show slow formation of the intermediate 10c with completion within -3 min. No further changes in the NMR resonances of 9c and 11c were seen. (b) Mechanistic studies. The various Pd species present in the system during the reaction of 2c with I2 were each tested for possible reactivity with sulfur to help determine how the Pdl2(dpm(S)) (11c) was formed. Reaction of Pd2l2(dpm)2 (lc) with Sg in CDCI3 at R T . after 24 h gave only the p-S adduct 2c as evidenced by ' H and 31P{1H) NMR spectroscopy. With 9c and 10c, no reactivity was observed under purely thermal conditions. Specifically, in NMR-scale experiments in CDCI3, ten-fold excesses of sulfur were employed with 9c from 20 to 90 °C; ! H and 31P{1H} NMR studies after 24 h revealed no new signals. Using photolytic conditions, however, 11c was quantitatively formed at R.T. after -2 d of irradiation with a medium pressure Hg vapour lamp (450 W). Under the same conditions, 9c in the absence of Sg was unchanged. Possible reactivity of 10c with sulfur was studied at -42 °C where 10c, formed by reaction of lc with I2, is stable at this temperature (see above); addition of an equimolar solution of Sg to 10c in CDCI3 produced no changed as indicated 156 References on page 189 Chapter 4 by ' H and 31P{'H} NMR data after 24 h. The NMR sample was subsequently placed at R T . to allow 10c to decompose; only 9c was formed. NMR-scale reactions in CDCI3 at R.T. of 2c with I2 using different stoichiometric ratios were also carried out (Table 4.3). In the reaction of 2c with 0.5 mole equivalent of I2, 9c and 11c, in addition to unreacted 2c, were observed after 24 h; the yield of 11c was 25% relative to 9c. Reaction of 2c with 2 mole equivalents of I2, however, yielded after 24 h 9c only; 11c was not observed. Corresponding TLC studies revealed the presence of elemental sulfur in the former reaction and surprisingly its absence in the latter. A single R.T. NMR-scale reaction of Pd2I2(p,-SO)(dpm)2 with 1 mole equivalent of I2 in CDCI3 was performed. The ' H and 31P{!H} NMR spectra revealed the presence of both 9c and 11c in about 65 and 35% yields, respectively; unreacted Pd2I2(p-SO)(dpm)2 was not observed. A corresponding TLC study revealed the presence of elemental sulfur. Because of the ease with which 11c forms during the early stages of the reaction of 2c with I2 (see above), low temperature ESR experiments were performed (in attempts to detect, for example, Si and/or S 2 radicals) by injecting equal volumes of equimolar CDCI3 solutions of 2c and I2 initially at R.T. into an ESR tube held in liquid N 2 . The ESR spectrum of the mixture, which solidified immediately (~1 s), was then recorded at 104 K and did not reveal the presence of any signals. The sample was briefly warmed to liquefy the mixture and was immediately placed back in liquid N 2 afterwards (~1 min total time); the recorded ESR spectrum again revealed no signals. Crystal structure of PdI2(dpm(S))«0.5 CH2CI2 (llc»0.5 CH2CI2). Crystal data, information relating to data collection, refinement details, bond distances and angles, and atomic coordinates for PdI2(dpm(S))»0.5 CH2C12 are given in Appendix VI. 157 References on page 189 Chapter 4 The molecular structure of Pdl2(dpm(S)), ignoring the greatly disordered solvent molecule, is shown in Fig. 4.11. The Pd atom in the complex is located in an approximately square-planar coordination environment; there is slight tetrahedral distortion around the metal centre with 1(1) and P(l) lying slightly above the plane [0.0079(7) and 0.075(2) A, respectively] and 1(2) and S slightly below the plane [0.0085(7) and 0.084(2) A, respectively]. The Ph2PCH2P(S)Ph2 ligand coordinates via the phosphorus and the S atom so as to form a five-membered chelate ring. This chelate ligand lies in a twisted conformation so that P(2) is above the coordination plane, 0.698 A. The P(l)-C(l) and P(2)-C(l) bond distances and P(l)-C(l)-P(2) bond angle are 1.860(8) A, 1.814(8) A, and 107.8(4)°, respectively, which are within the normal ranges typically seen for P-C single bonds28 ,29 and P-C-P angles.29'30 The P(2)-S and Pd-S bond distances are 2.007(3) and 2.331(2) A, respectively, and are not unusual.29"31 The Pd-I(l) bond distance trans to the Pd-P(l) bond is 2.6419(9) A, while the Pd-I(2) bond distance trans to the Pd-S(l) bond is much shorter (2.6006(9) A). The difference is in keeping with the apparently greater trans influence of phosphorus versus sulfur.31"33 The poor crystallinity of PdCl2(dpm(S)) precluded an accurate X-ray crystallographic analysis; however, refinement of data on this compound to R = 0.18 revealed an approximately square-planar coordination of the ligands around the Pd atom and an envelope configuration of the five-membered dpm(S) chelate ring (Fig. 4.12). Reaction of Pd2Br2(p-S)(dpm)2 (2b) and Br 2. Corresponding kinetic and mechanistic studies in CHCI3 were attempted on the bromide system. However, complications arose due to the reactivity of Br 2 with the solvent34 and irreproducible results were obtained. UV-vis spectral characteristics of Br2 in CHCI3 at R.T. changed dramatically on the same time scale as the reaction 158 References on page 189 Chapter 4 Fig. 4.11. ORTEP drawing and stereoview of Pdl2(dpm(S)) (11c). Hydrogen atoms are omitted. 159 References on page 189 Chapter 4 Fig. 4.12. ORTEP and PLUTO olrawings of PdCl2(dpm(S)) (11a). Hydrogen atoms are omitted. c n 160 References on page 189 Chapter 4 between 2b and Br2; no change was observed for I2 in CHCI3. Nevertheless, a single qualitative NMR experiment revealed products analogous to those of the iodide system; namely, PdBr2(dpm) (9b) and PdBr2(dpm(S)) (lib) were observed to form in approximately 67 and 33% yields (Fig. 4.13). However, the reaction was incomplete as starting material 2b was still seen even though a 1:1 mole ratio of reactants was used. Yellow particulate matter attributed to elemental sulfur was not visually observed; however, TLC studies were able to establish the presence of this product. Solvent effects. Kinetic and mechanistic investigations of the reaction of 2c with I2 in various solvents were attempted. Of the solvents tested, none was found suitable for comparative studies. Either insolubility of 2c was encountered or reactivity (adduct formation?) with I2 was observed (i.e. acetone, acetonitrile, alcohol, benzene, carbon tetrachloride, dimethylacetamide, dimethylformamide, ethyl acetate, and toluene). Acetonitrile was used, however, for a qualitative NMR study because of its relatively slow reactivity toward I2 (compared with the other solvents). In a single NMR study in which 2c and I2 were reacted in a 1:1 mole ratio at R.T. in CD3CN, 10c was observed to form immediately (in seconds) and disappear rapidly (few minutes) with concomitant formation of 9c and 11c in approximately 68 and 32 % yields, respectively. The presence of 10c was evidenced by the 1:4:6:4:1 quintet at 6 5.08 in the *H NMR spectrum and the singlet at 5 1.54 in the 31P{1H} NMR spectrum; 2c was not seen. Again, 9c and 11c were characterized by a 1:2:1 triplet (5 4.74) and a 1:2:1 pseudo-triplet (5 4.15), respectively, in the lH NMR spectrum. In the 31P{'H} NMR spectrum, only the singlet of 9c at 5 -55.7 was seen. The PdI2(dpm(S)) species was probably too low in concentration to be detected (2c has limited solubility in CD 3 CN: ~8 x 10"4 M from 0.5 mg in -0.5 mL). TLC studies carried out on the NMR sample established the presence of elemental sulfur. Corresponding conventional UV-vis studies at 161 References on page 189 Chapter 4 Fig. 4.13. ' H (300 MHz) and "Pf/H} NMR spectra of the "completed" reaction (24 h) of Pd2Br2(p-S)(dpm)2 (2b) with 1 mole equivalent of Br2 in CDC13 at R.T.; 9b = PdBr2(dpm), l i b = PdBr2(dpm(S)). "Pf/H} l i b 9b i l l l i b 2b 2b 8.0 7.0 6.0 5.0 4.0 3.0 PPM 9b 60 40 20 0 PPM -20 -40 -60 162 References on page 189 Chapter 4 R T . were carried out using only equimolar quantities of 2c (9.4 x 10"5 M) and I2, as excess I2 used for pseudo-first-order conditions led to rapid decomposition of 9c in this solvent, as evidenced by subsequent NMR studies. Addition of a light brown CH 3 CN solution of I2 to a brown CH3CN solution of 2c immediately produced a yellow solution (presumably containing 10c) characterized by two absorption maxima at X = 294 and 360 nm (Fig. 4.14a, absorption maximum at X = 294 nm not shown). Thereafter, the absorbances of these bands decreased over time, and an isosbestic point at X = 414 nm was observed. The reaction was complete after -20 min, and the final UV-vis spectrum was invariant with time. The solution was subsequently analyzed using NMR spectroscopy which revealed the presence of only 9c and 11c in about 70 and 30% yields, respectively (cf. NMR-scale study done at -8 x 10"4 M 2c). The kinetic data collected at X = 360 nm where s 9 c = 8 n c = 2.0 x 103 M" 1 cm"1 (sulfur has no absorptivity here) surprisingly analyze for approximately second-order with k^s -4 x 10"3 M V 1 for the decomposition of 10c (Fig. 4.14b). Kinetic studies of the rapid formation of the 10c from 2c and I2 were attempted, but the reaction occurred too quickly to be amenable by stopped-flow spectrophotometry. Moreover, complications due to the reactivity of I2 with the solvent interfered with the experimental setup. 163 References on page 189 Chapter 4 Fig. 4.14. a) Visible absorption spectral changes (320-600 nm region) as a function of time for a C H 3 C N solution of Pd2I2(p-S)(dpm)2 (2c) (9.4 x 10"5 M) on addition of I2 (9.4 x 10"5 M) at 23.5 °C; 9c = Pdl2(dpm). b) A rate-plot analyzed for an approximate second-order dependence on Pd2l4(dpm)2 (10c); A t and An represent the absorption at 360 nm at times t and oo, respectively. « 0 500 600 Wavelength, nm b 1 -0 -I , . 1 . 1 . 0 200 400 600 800 1000 1200 Time, s 164 References on page 189 Chapter 4 4.3 Discussion Reaction of Pd2l2(p-S)(dpm)2 (2c) with I2 proceeds in two kinetically observable stages as described in eqs. 4.3.1 and 4.3.2 to yield Pdl2(dpm) (9c) and elemental sulfur. Transannular oxidative addition via electrophilic attack of I2 occurs rapidly with formation of Pd2Lt(dpm)2 (10c) and presumably monatomic sulfur. Thereafter, slower unimolecular decomposition of 10c generates the monomeric product 9c. The elemental sulfur product, Sg, was visually detected and definitively identified using mass spectrometry, thin-layered chromatography, UV-vis spectrophotometry, and by its reaction with Pd2l2(dpm)2 (lc) to give only 2c (see below). (4.3.1) Pd2l2(u-S)(dpm)2 + I2 >^ Pd2l4(dpm)2 + 'S ' 'fast' (4.3.2) Pd2l4(dpm)2 % 2 Pdl2(dpm) 'slow' The chemistry parallels in part that observed by Hunt and Balch for the reaction of lc with I2 which also proceeds via 10c (eq. 4.1.2).2 Although no quantitative kinetic studies were carried out, a qualitative ti/2 value for decomposition of 10c was determined from NMR data (see below).2 Other similar chemistry can be seen in the halogenation of the substituted metal carbonyls cited in the Introduction:15'18,20 rapid pre-equilibria occur with the formation of adducts which then undergo slower intramolecular conversion to halogen-containing products. fast slow complex + nX 2 w complex»nX2 • products In the present system, depending on experimental conditions, the by-product PdI2(dpm(S)) (11c) could also form (eq. 4.3.3) (see below) and as a consequence the yield of elemental sulfur could be affected. For example, in NMR- and synthetic-scale studies where concentrations are 2 3 about 10" and 10" M, respectively, 11c was found to form in -17 and -5% yields, respectively. The yield of sulfur was -80% in each case; the unaccounted 10% in the synthetic-scale reaction is 165 References on page 189 Chapter 4 probably due to incomplete recovery from the sides of the reaction vessel as the remaining sulfur there was difficult to remove. In UV-vis spectroscopic studies, however, where concentrations are of the order of 10"4 M , 11c does not form, and thus lowering concentrations permits more complete formation of elemental sulfur. (4.3.3) Pdl2(dpm) + Sn -> PdI2(dpm(S)) n * 8 Formation of the intermediate, Pd2l4(dpm)2 (10c). The kinetics of reaction 4.3.1, studied using stopped-flow spectroscopy, revealed a first-order dependence on both 2c and I2. The absence of any observable intermediates during low temperature NMR studies and the excellent linearity of the Eyring plot 1 5 a' 2 0 c' 3 5 are good indications of an uncomplicated system. Formation of 10c can be conveniently pictured as proceeding via a transition state such as I. Electron density is visualized as proceeding from the bridged S atom through the metal centers to the attacking I2; the net result is the oxidation of the sulfide ligand to sulfur and the accompanying reduction of I2 to iodide. The associative mechanism implied here is commonly seen for square planar, d 8 metal complexes.36 I 1 I There is ample precedence for involvement of ionic species in transannular oxidative addition of halogens.16c'17 For example, in reaction 4.1.10, conceivable intermediates such as the following are suggested.16c ~(CO)4M-I-M(CO)4 - Ii -(CCO4M4- rM(co)4 j 1 L 166 References on page 189 Chapter 4 Likewise, in reaction 4.1.11, the intermediate formed following electrophilic attack at one Fe center is thought to be a halogenonium species, which rearranges to give an isolated bridging halide species that subsequently undergoes nucleophilic attack by X" to give the product 17 dienyl O dienyl Fe- -Fe OC \\ CO O x 2 dienyl ft \ / C ^ / F e - — - F e O C x II \ o X dienyl CO dienyl ft \ dienyl / Fe^ ^Fe OC \ ft / CO x -2 Fe(dienyl)(CO)2X dienyP + O C ^ F | e CO Fe-I CO - c o x -The kinetic data and spectroscopic data do not rule out such possibilities in the Pd-system studied here. Rather than simultaneous electrophilic attack of two iodine atoms at the two Pd centers, an initial electrophilic attack at one metal center is more probable,14'20 and a transition state such as II is therefore not unreasonable. Associative substitution could then give formation of intermediate III, and then nucleophilic attack of I at the second Pd center, perhaps via transition state rv, could lead to 10c. 1—i II in p — ~ - p Pd Pd rv p p P d - 1 P d ' 1 + 'S 10c The theoretical rate equation derived for the series II through 10c, for example, ki Pd2I2(u-S)(dpm)2 + I2 - III + T k-i III + r — - — • Pd 2I 4(dpm) 2 + 'S' 167 References on page 189 Chapter 4 with steady state treatment for III, gives rate = (k1k2/(L1+k2))[2c][I2] which is also consistent with the experimental findings; the measured rates were independent of added iodide, but this clearly does not rule out the possibility of ionic intermediates. The activation parameters, A H / = 32 kJ mol'1 and ASj* = -91 J mol'1 K"1, are typical of those seen for oxidative addition at either one or two metal centres.37"40 Furthermore, these values are comparable to those reported for halogenation studies of some dinuclear complexes, as well as halogenation of alkenes and alkynes41'42 which are viewed as proceeding via halogenonium intermediates (Table 4.4). In the present study, it is not certain whether die parameters refer to reaction via transition states I or II. Decomposition of the intermediate, Pd2l4(dpm)2. The kinetics of the decomposition of 10c, reaction 4.3.2, studied using conventional UV-vis spectroscopy, revealed first- and zero-order dependences on 10c and I2, respectively. The observed isosbesticity in these measurements, the absence of any detectable intermediate species in low temperature NMR studies, and the excellent linearity in the Eyring plot indicate a well-behaved system. These results complement those of Hunt and Balch who have studied the reaction of 1 with X 2 . 2 Specifically, reaction of lc with I2 rapidly formed 10c which underwent slow unimolecular decomposition to give 9c. From a qualitative assessment of the NMR data, they reported ty2 = 360 s at 15 °C in CD 2C1 2 which compares to ti/2 = 850 s in CHCI3 in the present study, determined using the rate constant kp = 8.14 x 10'4 s"1 obtained from extrapolation of the ko values in Table 4.2 to 15 °C. As noted previously in Chapter 3, the rates of reactions in these solvents commonly differ by a factor of 2 168 References on page 189 Chapter 4 Table 4.4. Activation parameters for some halogenation studies of dinuclear metal complexes and unsaturated systems. System (ref) A H * (kJ mol_1) AS* (J mol 1 K"1) Reaction 4.1.9 (15) 21 -43 -78 to -120 Reaction 4.1.10 (16c) 28-40 -46 to -58 Reaction 4.1.13 (19) 48-61 -61 to -122 Alkenes, alkynes (41,42) 20-60 -120 to -160 Table 4.5. Polarity (Q) of selected solvents at 25 °C (measured by (e-l)/(2s+l) where e is the dielectric constant200) (see text, p. 172). Solvent Dielectric constant, e Polarity, Q CCL, 2.24 0.226 CHC1 3 4.81 0.359 CH 2 C1 2 9.08 0.422 C H 3 C N 36.2 0.480 169 References on page 189 Chapter 4 with reactions occurring faster in the more polar CH2CI2. Thus, the findings are in good agreement with those of Hunt and Balch (see later). Decomposition of 10c can be envisaged as proceeding via transition state V: V It should be noted that structure V is a realistic one for 10c, but the face-to-face dimeric structure is preferred because of the rarity of five-coordinate Pd(II).2 Substitution reactions of square planar, d 8 metal complexes are generally associative rather than dissociative and, as Hunt and Balch have suggested, it is unlikely therefore that 10c simply fragments by direct Pd-P bond rupture but rather, halide bridges form to give trigonal bipyramidal geometry at the Pd centres prior to Pd-P bond breaking.2 Furthermore, the presence of 7i-bonding phosphines should contribute positively to the transition state as removal of electron density from the metal results in a more positive metal that could more readily accept the entering, fifth ligand.43 That halide bridges could form from 10c is supported by several examples.2 For instance, the complexes Rh2Cl2(CO)2(u-dpm or u.-dam)2 have face-to-face dimeric structures but the Rh coordination planes are so tipped that both Cf ligands are folded in towards the adjacent Rh atoms, and molecular models indicate that the formation of halo-bridged structures is entirely reasonable.44'45 r I P h 2 E E P h 2 OO P h 2 E , E P h 2 E = P or As 170 References on page 189 Chapter 4 Of note, decomposition of these Rh complexes to monomeric species does not occur; their stabilities are attributed to the weak metal-metal bonding interactions that exist in related Rh(I) dimeric complexes213 and which presumably do not exist in the Pd(II) dimer. The activation parameters, A H D * = 80 kJ mol"1 and A S D * = -26 J mol"1 K"1, obtained in the present study are consistent with the proposed mechanism. The small entropic change seems reasonable for a unimolecular re-arrangement of 10c to V as the iodide ligands are already in "close vicinity", while the relatively large enthalpy term probably reflects the energy required to overcome the proximity and strain effects in forming the halide bridges. The proposed mechanism probably also applies to Pd2Br4(dpm)2 (10b) and Pd2CU(dpm)2 (10a) but no kinetic measurements for their decompositions are available. Of note, 10a and 10b were also observed as intermediates in the reactions of la and lb with HCl and HBr, respectively.46 Kinetic studies of the reaction of 2b with Br2, presumably proceeding via 10b, were attempted in the present study but reactivity of Br 2 with the solvent (presumably alkane halogenation)34 interfered with reproducibility of the results. A qualitative NMR study, however, revealed that the bromide system reacts faster than the iodide system; as in many systems, the greater reactivity is commonly attributed to Br2 being a less selective electrophile than I2,2 0 c and this is reflected in a smaller entropy of activation, for example, for the formation of 10b.20c Although solvent iodination does not occur,47 it seems unlikely that iodine atoms have a role in the system as reaction 4.3.1 also takes place in the absence of light and/or at low temperatures. Solvent effects. The reaction of 2c with I2 was briefly examined in acetonitrile, where again rapid formation and subsequent slow decomposition of the intermediate 10c, to give the products 9c and elemental sulfur, were observed. The first stage was not kinetically observable, the formation of 171 References on page 189 Chapter 4 10c occurring too quickly to be amenable for study even by stopped-flow spectroscopy. The second stage, however, was studied using conventional UV-vis spectroscopy, and although isosbesticity was observed indicating a well-behaved system, the rate of decomposition was approximately second-order in 10c. This finding is not easily rationalized using the limited data available, although reactivity of I2 with the solvent, as noted previously, could be a contributing factor. Nevertheless, the observed rapidity of the reaction of 2c with I2 in acetonitrile in comparison to the rate in CHC13 (and CH2C12) may be due to the increase in solvent polarity (Table 4.5). In general, increase in the polarity of a solvent causes reaction rates to accelerate by stabilizing the charge separation or the development of charges in the transition state (e.g. for transition states II, IV, and v).20c'48'49 Presumably, the rate of formation of PdI2(dpm(S)) is similarly affected as this by-product was formed in -32% yield (cf. -17% in CHCb) (see later). Formation of Pd2I2(u.-S)(dpm)2 (2c); reaction of Pd2I2(dpm)2 (lc) with Sg. Reaction of 2c with I2 produces elemental sulfur, one identification being made by its reaction with lc to form only 2c, as evidenced by NMR spectroscopy (eq. 4.3.4). (4.3.4) Pd2I2(dpm)2 + 1/8 S8 -> Pd2I2(u-S)(dpm)2 Although extensive studies on this particular reaction were not carried out, there are numerous reports on reactions of transition metal complexes with elemental sulfur50 including several studies describing similar chemistry. For example, Balch et al. have studied reaction 4.3.5 for which an X-ray crystal structure of Pd2Cl2(u-S)(dpm)2 was obtained.51 (4.3.5) Pd2Cl2(dpm)2 + 1/8 S8 Pd2Cl2(u-S)(dpm)2 Espenson et al. investigated the A-frame formation reactions of [Pt2(PPh3)2(dpm)2]2+ and Pt2X2(dpm)2 (reactions 4.3.6 and 4.3.7) for which some kinetic and mechanistic studies were 172 References on page 189 Chapter 4 52 53 done. ' They demonstrated that all or nearly all the sulfur atoms of Sg can be utilized, and suggested that the two reactions have different reaction profiles, with reaction 4.3.6 proceeding via an intermediate containing a monodentate "dangling" dpm ligand and an intact Pt-Pt bond, and reaction 4.3.7 via a transition state with opposite characteristics. (4.3.6) [Pt2(PPh3)2(dpm)2]2+ + 1/8 S8 -> [Pt2(PPh3)2(u-S)(dpm)2]2+ (4.3.7) Pt2X2(dpm)2 + 1/8 S 8 -> Pt2X2(u-S)(dpm)2 X = halide The issue of how Sg reacts remains unclear although Espenson et al. suggested that sulfur fragments smaller than S8 (metastable molecules that have been identified (see below)) could also undergo similar insertive reactions; in view of the suggestion that metal - metal bond cleavage occurs in the transition state (i.e. there is electron pair transfer out of the metal - metal bond53), the following is a plausible mechanistic scheme with n < 8 for reactions 4.3.4, 4.3.5, and 4.3.7. The resulting Sn.i fragment could undergo yet another insertive reaction. Formation of PdI2(dpm(S)) (11c). The fate of the sulfur atom produced from the reaction of 2 with X 2 (i.e. reaction 4.3.1) is of interest because of the formation of the by-product PdX2(dpm(S)) observed only in synthetic- and NMR-scale studies. Table 4.3 summarizes the work carried out in part to determine the factors that govern the formation of the by-product. Mass spectrometry was used in part to identify positively the elemental sulfur produced, the mass spectrum excellently illustrating the ionization pattern with the parent mass peak observed 173 References on page 189 Chapter 4 at 256 m/z and all other peaks at 32 mass units apart (Fig. 4.2). Although the exact mechanism is not known, Sg is commonly pictured to form via a series of rapid concatenation reactions, possibly through Si atoms as recently suggested by Williams and Harpp54 in their study of the MeO" induced decomposition of methoxycarbonylsulfenic 4-toluehethiosulfonic anhydride in 2,3-dimethyl-1,3-butadiene (eq. 4.3.8). (4 3 8) M e ^ Q V - S - S - S - C - O M e + NaOMe M e O - C - O M e + Me — / Q V - S - S - N a + 1/8 & In these studies, diatomic sulfur could not be detected and thus concatenation of monatomic sulfur was proposed; however, frozen-state, ESR spectroscopic investigations in the present study not only failed to detect S2 but also Si or any paramagnetic species. Si and S2 are species which can be detected by ESR spectroscopy, for example, in the gas phase.56'57 Nevertheless, as will be seen later, the yield of by-product 11c results from competition of two reactions, the concatenation reaction to form Sg, perhaps via monatomic S (eq. 4.3.9), and the reaction of an S„ species (n < 8) with PdX2(dpm) to form PdX2(dpm(S)) (eq. 4.3.10). (4.3.9) Sn + S -> S n + i (4.3.10) Sn + PdX2(dpm) -> PdX2(dpm(S)) + S„., It is noted that in addition to Si, S 2, and S 8, sulfur species S3 through S26, including allotropic forms, have been identified and, whereas Si through S5 are stable only in the vapour phase, S6, S 7 , and S9 through S26 are metastable cyclic species in solution that eventually (over "considerable" time) convert to the thermodynamically stable Sg form.5 7'5 8 In the solid phase, however, So through S26 are known to have kinetic stabilities comparable to that of Sg. Although it is uncertain whether S6 or S7 are also present in the 2c +12 system, it is unlikely that sulfur species higher than Sg are also formed. 174 References on page 189 Chapter 4 The identity of the by-product was established by the reaction of PdCl2(PhCN)2 with dpm(S). Using a 1:1 mole ratio, ligand exchange afforded PdCl2(dpm(S)) (11a) from which PdBr2(dpm(S)) (lib) and Pdl2(dpm(S)) (11c) were then obtained using the appropriate halide source. Of note, use of dpm(S) in excess resulted in formation of the ionic species [Pd(dpm(S))2]Cl2 for which an X-ray crystal structure was obtained (see Appendix VU). As with [Pd(dpm(S))2]Cl2, the X-ray crystal structures of 11a and 11c show coordination of dpm(S) through the phosphorus and sulfur atoms with the resulting five-membered ring in an envelope configuration (Figs. 4.11 and 4.12). A comparison was made (see Results, p. 158) with the recently reported structure of RhI(CO)(dpm(S)), which is the first literature X-ray crystal structure containing a dpm(S) ligand,29 although the PdCl2(dpm(S)) structure was completed in this thesis work in 1993. There exist examples in the literature of non-structured transition metal complexes containing chelating dpm(S), as well as chelating R2P(Y)CH2PR2 ligands where one or more phenyl groups are substituted with other alkyl groups, and where Y = O, S, or Se.59 Reaction of Pd2l2(p-S)(dpm)2 (2c) with I2 was explored under various conditions as summarized in Table 4.3, and the yield of the by-product Pdl2(dpm(S)) (11c) can be satisfactorily explained in terms of the two competing reactions 4.3.9 and 4.3.10. From the non-reactivity of Sg with 2c, 9c (under purely thermal conditions), or 10c, it follows that an Sn species where n < 8 must be responsible. Reaction of 2c with I2 at -42 °C where 10c is stable did not yield 11c after 24 h; 11c was not observed even when the temperature was raised to R T . thereafter. Thus, the conclusion here is that Sn does not react with 10c. Sn also does not react with 2c, as the by-product yield from reaction with excess 2c is comparable to that from reactions with equimolar quantities of 2c and I2. The possibility left from this process of elimination is that of reaction Sn with 9c forms 11c. Strong support for this suggestion comes from reaction of Sg with 9c under photolytic 175 References on page 189 Chapter 4 conditions where 11c can be quantitatively formed (reaction 4 . 3 . 1 1 ) . In the absence of S g , 9c was found to be unchanged. hv ( 4 . 3 . 1 1 ) Pdl2(dpm) + 1/8 S 8 PdI2(dpm(S)) ~2d In the present 2c +1 2 system, some 11c forms even in the absence of light, and because the Pd-P bond ( w 4 0 0 kJ mol"1)60 is approx. 2 0 0 kJ mol"1 stronger than the S-S bond,57 reaction 4 . 3 . 1 1 presumably proceeds via ring opening50j of Sg to form a reactive S n species which then reacts with 9c. The strong impetus to form S 8 in the competing rapid concatenation reaction 4 . 3 . 9 is in keeping with observations that 11c forms only during the early reaction stages and ceases to form once reaction 4 . 3 . 9 is complete even though 9c continues to form from the decomposition of 10c. The results in Table 4 . 3 can be rationalized semi-quantitatively by considering rates of reaction. Assume that reaction 4 . 3 . 9 proceeds via monatomic sulfur: k , S + S - > S 2 k 2 s2 + s S 3 k 7 s7 + s s8 Reaction 4 . 3 . 1 0 is rewritten to include all possibilities: K S + 9c -> 11c k b S 2 + 9c -> 11c + S kc S 3 + 9c 11c + S 2 \ S 7 + 9c -> 11c + S 6 A full rate derivation would be extremely complicated; however, two further assumptions can be made. As previously discussed, spectroscopic data reveal elemental sulfur forms very quickly and 176 References on page 189 Chapter 4 intermediates species such as S i and S 2 are not observable; furthermore, 11c forms only during the early stages (until Sg is fully formed) and only in minor amounts relative to, for example, elemental sulfur. Thus, it is reasonable to assume that k i , k 2 , k 7 » ka, k b , k g , and that the intermediate species Si through S7 are very reactive and are therefore present at very small concentrations such that the steady state approximation can be applied giving the expression [Sn] = ki[S]/k„ (n = 2 - 7).* The rate of formation of Sg can then be shown to have the following form: (4.3.12) rate#l = ki[S]2 The rate of formation of 11c would be as follows: (4.3.13) rate#2 = [9c](ka[S] + kb[S2] + kc[S3] +... + kg[S7]) Application of [S„] = ki[S]/kn on eq. 4.3.13 gives (4.3.14) rate#2 = [9c](ka[S] + (kbki/k2)[S] + (kcki/k3)[S] + ... + (kgki/k7)[S]) = k[9c][S], where k = ka + k bki/k 2 + kcki/k3 + ... + k gki/k 7 Eq. 4.3.14 is quite general and easily accommodates any unrealistic possibility (i.e. where the rate constant k varies). Species 9c forms from the unimolecular decomposition of 10c, and thus reaction 4.3.14 can be written as follows: (4.3.15) rate#2 = k[S](2[10c]„)(l-exp(-kDt)) where [10c]o is the [10c] when it is fully formed (before any appreciable decomposition to 9c). On consideration of initial rates, eqs. 4.3.14 and 4.3.15 become: (4.3.16) rate#l„ = ki[S] 0 2 (4.3.17) rate#20 = 2k[S]o[10c]o t The back reactions (with k_i, k . 2 , k . 7 ) are not considered. For further explanation of the steady state approximation as it applies here, see Frost, A . A . ; Pearson, R.G. "Kinetics and Mechanism, 2nd Ed." , John Wiley & Sons, London, 1963, ch. 8, p. 172. 177 References on page 189 Chapter 4 where [S]0 is the initial (total) concentration of monatomic S. Eliminating [S]„ from eqs. 4.3.16 and 4.3.17 gives the following ratio (R). (4.3.18) R = [(rate#lo)1/2]/[rate#20] = k,1/2/(2k[10c]o) Hence, on using lower concentration conditions (i.e. proceeding from synthetic- to NMR-to kinetic-scale studies), R increases as [10c]o decreases, and lower yields of by-product are observed (with correspondingly higher yields of elemental sulfur). For the bromide system, the higher by-product yield could be explained by higher k values (i.e. reaction 4.3.10 proceeds at a greater rate), although the greater decomposition rate of 10b (relative to that of 10c) could be a contributing factor. In acetonitrile, the rates of both reactions 4.3.9 and 4.3.10 are perhaps similarly affected (i.e. increased) but because of the square root factor, R decreases overall and the yield of 11c increases. Finally, in the reaction of Pd2l2(u-SO)(dpm)2 with I2, which also presumably proceeds via the intermediate 10c, R decreases perhaps because the rate of reaction 4.3.9 decreases because of the slower decomposition of the free SO species (which gives elemental sulfur and dioxygen61). It is interesting to note that while the presence of O2 has no effect on the yield of 11c, excess amounts of I2 result in its lack of formation. For example, in an NMR-scale reaction of 2c with 2 mole equivalents of I2, He was not observed to form and the only Pd species seen was the 9c product. Iodine exhibits no reactivity with 11c or with Sg, and thus the absence of 11c can be attributed to a scavenging process by I2 for an S n species to form, for example, the known 62 iodosulfanes, SJ2 where n > 2. Presumably, this scavenging process also takes place (in addition to reactivity with the solvent) in the bromide system and probably at a greater rate, as unreacted 2b is observed despite equimolar conditions. 178 References on page 189 Chapter 4 Although speculative, the mechanism for the formation of 11c could resemble that for reaction 4.3.4, the reaction proceeding via initial nucleophilic attack of an Sn species at the Pd centre and the resulting S„-i species undergoing yet another insertive reaction. It is unlikely that 11c forms simply by insertion of monatomic sulfur into the Pd-P bond as reactions of this type usually require excited sulfur atoms, S(*D), commonly generated under photolytic conditions (e.g. by photolysis of COS). 6 3 S(*D) insertion reactions are known; for example, the gas- or liquid-phase photolysis of COS in the presence of alkanes produces thiol products.64 COS -> CO + S('D) R-H + S(TJ) -> R-SH In the present study, ground state sulfur atoms, S(3P), are probably formed, but these have been shown to be inert toward such insertive reactions.63 Rather, it is surmised that reaction 4.3.10 proceeds as a result of the torsional strain that cyclic S„ species are known to have,57 although involvement of chain-like Sn reactive species cannot be discounted.65 Why a second insertion reaction does not occur to form, for example, Pdl2(dpm(S)2) is unclear. Perhaps the energetics favour a five-membered ring as opposed to a six-membered ring that might contain axial phenyl group interactions; in the X-ray crystal structure of 11c (Fig. 4.11), the axial phenyl groups are seen to be positioned away from one another: 179 References on page 189 Chapter 4 Ph pPh Ph P h ' N * s - ^ 1 / P = = S - ^ Ph Synthesis of PdCl2(dpm(S)2), by reaction of PdCl2(PhCN)2 with dpm(S)2, was attempted but because of thermal instability of the product, its identity has not been unequivocally established. Of note, transition metal complexes containing chelating dpm(S)2 as well as dpm(Se)2 are known.590'72 180 References on page 189 Chapter 4 A.A Experimental section The materials used, synthetic procedures for the ligands and complexes, and instrumentation used for ! H and 31P{!H} NMR, ESR and UV/vis spectra, were described in Chapter 2. Dpm(S), dpm(S)2, and the complexes PdCl2(PhCN)2, Pd2(dba)3CHCl3, Pd2I2(u-SO)(dpm)2, la - lc, 2a - 2c, and 9a - 9c were synthesized using published methods as outlined in Chapter 2. All experiments were performed under N 2 unless otherwise specified. 4.4.1 Preparation of PdCl2(dpm(S)) (11a) PdCl2(PhCN)2 (0.10 g, 0.26 mmol) and dpm(S) (0.11 g, 0.26 mmol) were stirred in CH 2C1 2 (20 mL) at room temperature (R.T.). Within several minutes, the colour changed from an initial yellow to a final orange with accompanying precipitation of a yellow solid. Stirring was continued for 1 h before the volume was reduced to ~ 10 mL to complete the precipitation. The yellow solid was filtered off, washed with Et 20 (2x10 mL), and dried in vacuo; yield 0.15 g (97%). *H NMR (20 °C, CDC13): 5 7.0 - 8.0 m (20H, Ph), 5 4.07 p.t (2H, CH 2). 31P{ !H} NMR (20 °C, CDC13): 8 54.9, 30.9 (AB pattern, J P P = 18.3 Hz). UV/vis: X : 372 (1500). Anal. Calcd for C 2 5 H 2 2 Cl 2 P 2 PdS: C, 50.57; H, 3.73. Found: C, 50.56; H, 3.74. Crystals of the complex were grown via diffusion of hexanes (10 mL) or Et 20 (10 mL) into a 6 mL CH 2 CI 2 solution of 10 mg 11a at R.T., and subjected to X-ray crystallographic analysis. 181 References on page 189 Chapter 4 4A.2 Preparation of PdBr2(dpm(S)) (lib) PdCl2(dpm(S)) (0.10 g, 0.17 mmol) was dissolved in -20 mL CH 2C1 2 and a solution of "Pr4NBr (0.67 g, 2.52 mmol) in CH2C12 added. The solution which instantly turned yellow-red was stirred at R.T. for 1 h before the volume was reduced to -10 mL. Et20 (2x10 mL) was added to precipitate a yellow-red solid which was filtered off and washed with a Et20/CH2C12 mixture (2x10 mL; 60:40 by volume Et20:CH2Cl2) followed by Et20 (2x10 mL). The resulting yellow solid was then dried in vacuo; yield 0.110 g (96%). ! H NMR (20 °C, CDCh): 5 7.0 - 8.0 m (20H, Ph), 8 4.02 pt (2H, CH2). ^P^H} NMR (20 °C, CDCh): 5 56.6, 32.1 (AB pattern, J P P = 20.4 Hz). UV/vis: X : 396(2180), 374 sh (2095). Anal. Calcd for C 2 5H 2 2Br 2P 2PdS»CH 2Cl 2 : C, 40.68; H, 3.15. Found: C, 41.23; H, 3.11. Despite repeated drying procedures, the CH 2C1 2 solvent as evidenced by *H NMR spectroscopy was not removed. 4.4.3 Preparation of PdI2(dpm(S)) (11c) PdCl2(dpm(S)) (0.10 g, 0.17 mmol) was dissolved in -20 mL CH2C12 and a solution of "BmNI (0.93 g, 2.52 mmol) in CH2C12 added. The remainder of the procedure is as for lib. Yield 0.13 g (97%). ! H NMR (20 °C, CDC13): 8 7.0 - 8.0 m (20H, Ph), 8 3.71 p.t (2H, CH 2). 31P{'H} NMR (20 °C, CDC13): 8 61.1, 31.3 (AB pattern, J P P = 25.5 Hz). UV/vis: X : 478(3000), 420 sh (1735). Anal. Calcd for C 2 5H 2 2I 2P 2PdS»0.5CH 2Cl 2: C, 37.39; H, 2.83. Found: C, 36.88; H, 2.73. A brown crystal of the complex, obtained by diffusion of hexanes (10 mL) into a 6 mL CH 2C1 2 solution of 10 mg 11c at R.T., was analyzed using X-ray crystallography. 182 References on page 189 Chapter 4 AAA Preparation of PdCl2(dpm(S)2) PdCl2(PhCN)2 (0.10 g, 0.26 rnmol) and dpm(S)2 (0.12 g, 0.26 mmol) were stirred in CH 2C1 2 (10 mL) at R.T. The colour of the solution immediately changed from an initial yellow to a final orange-red. After a few minutes; the initially clear solution turned cloudy with accompanying precipitation of an orange-red solid. The reaction mixture was stirred for 1 h before Et 20 (20 mL) was added to complete the precipitation. The solid was filtered off, washed with Et 20 (2x10 mL), and dried in vacuo, yield 0.13 g (80%). This compound is not thermally stable; attempts to dry it, for example at 40 °C, resulted in its decomposition as evidenced by NMR measurements. Furthermore, the compound has limited solubility in CH2C12 and CHCI3 but is soluble in DMSO. ! H NMR (20 °C, DMSO-dg): 5 7.0 - 8.0 (20H, Ph), 8 5.541 (2H, C H 2 , J P H = 14 Hz). 31P{ !H} NMR (20 °C, DMSO-de): 8 37.3 s. UV/vis (20 °C, CH2C12): X: 418 (1.0), 322 (13.5) (relative intensities). 4.4.5 Kinetic studies a) Conventional UV-vis measurements. The kinetics of the slow decomposition of the intermediate, Pd2l4(dpm)2 (10c) (formed from 2c, see below), in CHCI3 were spectrophotometrically monitored in a stoppered thermostated quartz cell (volume 1.5 mL). A 1.00 mL solution of 2c of appropriate concentration was placed in the cell and thermostated at the required temperature (20 - 35 °C). A 0.050 mL solution of iodine was injected and the cell was briefly shaken to ensure complete mixing prior to monitoring optical density changes at some appropriate, fixed wavelength (396 nm). The concentration of 2c ranged from (1.6 - 9.9) x 10"5 M , 183 References on page 189 Chapter 4 and that of I2 from (3.8 - 7.5) x 10"4 M ; pseudo-first-order conditions were maintained and standard log (absorbance difference) vs. time plots gave excellent linearity for at least 2.5-3 half-lives. The pseudo-first-order rate constants, kobs, were readily evaluated from the semi-log plots or were automatically calculated by means of a direct mathematical function fit of the actual data by the quantification software provided (see Chapter 2). Corresponding kinetic measurements in acetonitrile were conducted at 23.5 °C using a 1:1 mole ratio of 2c (9.4 x 10"5 M) to I2 (9.4 x 10"5 M). Optical density changes were monitored at 360 nm. b) Stopped-flow measurements. The kinetics of the rapid formation of the intermediate, Pd2Lt(dpm)2 (10c), in CHCI3 were measured spectrophotometrically using a stopped-flow apparatus. Two 1-mL loading syringes were filled with solutions of the appropriate concentrations of 2c and I2, respectively (as noted above in (a)), and thermostated at the required temperature (20 - 35 °C). Optical density changes were monitored at a fixed wavelength of 510 nm (an isosbestic point only with respect to the decomposition of the intermediate). Reported pseudo-first-order rate constants were deterrnined from the average of four replicate experiments using a standard non-linear regression dgorithmic fitting equation provided by the accompanying kinetic software (see Chapter 2). 4.4.6 Mechanistic studies a) NMR-scale. The sulfur abstraction reaction (2c -> 9c, eq. 4.1.1) in CDCI3 was monitored at R T . using ! H and "Pf/H} NMR spectroscopy. A 0.10 mL solution of I2 (0.002 g, 0.0079 mmol) 184 References on page 189 Chapter 4 was injected into a septum-sealed NMR tube with a 0.40 mL solution of 2c (0.010 g, 0.0079 mmol). The sample was briefly shaken to ensure complete mixing before being placed in the NMR probe; spectra were recorded approx. every 10 min for 1 h. Variable temperature NMR-scale experiments also were performed at -42, 0, 20, 40, and 80 °C under anaerobic (i.e. vacuum) and aerobic (i.e. 1 arm air or 1 arm O2) conditions. In a typical experiment here, 2c (0.010 g, 0.0079 mmol) and I2 (0.0020 g, 0.0079 mmol) were placed in an NMR tube fitted with a PTFE (poly(tetrafluoroethylene)) J. Young valve (Aldrich) and -0.5 mL CDCI3 was vacuum transferred using liquid N 2 . The sample was then briefly placed in a dry ice / acetonitrile bath (-42 °C) 6 6 to ensure complete mixing of reactants before being placed in the appropriate constant temperature bath for 24 h prior NMR analysis. The 02/air dependence was examined only at R.T. The reaction was also examined at low temperatures; the samples were prepared by the simultaneous injection of equal volumes (0.25 mL) of equimolar solutions of 2c and I2 at R.T. into NMR or ESR tubes held at -42, -78, or -196 °C. Spectroscopic analyses at the respective temperatures were performed "immediately" and (for NMR spectroscopic analyses) after periods of approx. 2 min for a total of -30 min. Finally, the reaction was investigated using different stoichiometric ratios of 2c to I2. Specifically, 1:2 2c (0.010 g, 0.0079 mmol) to I2 (0.0040 g, 0.016 mmol) and 1:0.5 2c (0.010 g, 0.0079 mmol) to I2 (0.010 g, 0.0040 mmol) ratios of reactants were studied in CDC13 (-0.5 mL) at R.T. with NMR spectroscopic analyses performed after 24 h. Corresponding NMR studies in acetonitrile were conducted at R T . using a 1:1 mole ratio. Because 2c has limited solubility in CD 3 CN, at best -8 x 10"4 M concentrations can be achieved (e.g. 0.5 mg (3.9 x 10'3 mmol) in -0.5 mL CD 3CN). A 0.10 mL solution of I2 (0.1 mg, 3.9 x 10"3 mmol) was injected, and the sample was analyzed "immediately" and after periods of approx. 5 min for a total time of -30 min. 185 References on page 189 Chapter 4 The bromide system (2b -» 9b) was briefly examined at R.T. in CDCI3. To a 0.50 mL solution of 2b (0.010 g, 0.0085 mmol) placed in an NMR tube fitted with a PTFE J. Young valve was added Br2 (0.44 uL, 0.0085 mmol) via vacuum transfer. The sample was analyzed after 1 h. TLC studies of each NMR sample were carried out using CHCI3 eluent to analyze for elemental sulfur (see below). b) Synthetic-scale. Schlenk techniques were employed, the iodo complex 2c (0.050 g, 0.039 mmol) being reacted with I2 (0.010 g, 0.039 mmol) in 20 mL of CHCI3; the reaction was monitored by UV-vis spectroscopy. After 24 h, the supernatant red liquid was decanted off leaving a pale yellow solid. The solid was washed with acetone (2x10 mL) and dried in vacuo before being collected (yield -0.001 g), and then analyzed using mass spectrometry (electron impact), UV-vis spectroscopy (Et20), and thin-layered chromatography (CHC13 eluent). The results were compared with those of an authentic sample of elemental sulfur. The solvent was removed under vacuum from the supernatant and the resulting red-orange solid (9c) was dried in vacuo prior to ' H and 31P{1H} NMR analyses in CDCI3. A similar experiment was also performed where the concentrations of 2c and I2 were the same as those used in the UV-vis or stopped-flow studies; specifically, 2c (0.005g, 0.0039 mmol) and I2 (O.OOlg, 0.0039 mmol) were reacted in 40 mL of CHCI3. c) Reaction of Pd2I2(dpm)2 (lc) with sulfur. Reaction of lc with elemental sulfur was carried out in CDCI3 on the NMR-scale whereby a 0.20 mL solution of sulfur (0.13 mg, 0.0040 mmol) was injected into a septum-sealed NMR tube containing a 0.30 mL solution of lc (5.0 mg, 0.0040 mmol). The sample was analyzed "immediately", 1 h, and 24 h after preparation by NMR 186 References on page 189 Chapter 4 spectroscopy. Reaction of lc with the yellow solid obtained from (b) above was carried out in an analogous way except the solid (~1 mg) was directly added to a 0.50 mL solution of lc. d) NMR-scale examination of the reaction of Pdl2(dpm) (9c) with sulfur. Possible reaction between PdLXdpm) (9c) and sulfur was studied in CDCI3 at 20, 50, and 90 °C under purely thermal or photolytic conditions. Samples were prepared under vacuum in NMR tubes fitted with J. Young valves: 9c (0.0050 g, 0.0067 mmol) was placed in the NMR tube with a 10-fold excess of elemental sulfur (0.0022 g), and -0.5 mL of CDCI3 was vacuum transferred using liquid N2. The samples were then left at the appropriate temperatures for 24 h prior to NMR analyses. Photolysis was carried out only at 20 °C where the samples were irradiated for periods up to 2 d using a medium pressure Hg vapour lamp (450 W). e) NMR-scale examination of the reaction of Pd2l4(dpm)2 (10c), with sulfur. A 0.30 mL CDCI3 solution of Pd2l2(dpm)2 (la) (0.0050 g, 0.0040 mmol) was prepared in a septum-sealed NMR tube at -42 °C and a 0.20 mL CDC13 solution of I2(0.0010 g, 0.0040 mmol) was injected in. The sample was shaken and maintained at -42 °C for 15 h to permit complete conversion to 10c (as monitored by NMR spectroscopy). A 0.050 mL CDC13 solution of sulfur (0.0001 g, 0.004 mmol) was then injected and the mixture shaken and held at -42 °C for another 24 h prior to NMR analysis. The sample was then placed at R.T. for 24 h, and re-analyzed. f) NMR-scale examination of the reaction of Pd2l2(n-SO)(dpm)2 with I2. Reaction of Pd2l2(p--SO)(dpm)2 with I2 was carried out in CDC13 whereby a 0.20 mL solution of I2 (0.0020 g, 0.0078 mmol) was injected into a septum-sealed NMR tube containing a 0.30 mL solution of Pd2l2(p-187 References on page 189 Chapter 4 SO)(dpm)2 (0.010 g, 0.0078 mmol). The sample was analyzed "immediately", 1 h, and 24 h after preparation by NMR spectroscopy. TLC studies were carried out thereafter to analyze for elemental sulfur. 4.4.7 X-ray crystallographic analysis of PdI 2(dpm(S))»0.5CH 2CI 2 A single crystal having approx. dimensions of 0.05 x 0.20 x 0.30 mm was mounted in a glass capillary. All intensity measurements were made on a Rigaku AFC6S diffractometer with graphite monochromated Mo-Ka radiation (k = 0.71069 A). Accurate cell constants along with orientation matrices were obtained from a least squares refinement of the setting angles of 25 reflections in the range 14.2 < 26 < 24.7°. Intensity data were collected at 21 °C using the ©-26 scan technique. The intensities of 3 representative reflections were measured after every 200 reflections and no decay correction was applied. The data were corrected for Lorentz and polarization effects. The structure was solved by direct methods67 and expanded using Fourier techniques.68 The CH2C12 solvent is disordered about a centre of symmetry - the disorder was modeled by a single 0.5 occupancy carbon atom and 3 partially occupied Cl sites. The solvent C atom and 2 of the Cl atoms were refined with isotropic thermal parameters. Hydrogen atoms associated with the metal complex were fixed in calculated positions with C-H = 0.98 A. Neutral atom scattering factors69 and anomalous dispersion terms70 were taken from the usual sources. All 71 calculations were performed using TEXSAN. 188 References on page 189 Chapter 4 4.5 References for Chapter 4 1. Wong, T . Y . H . ; Barnabas, A.F. ; Sallin, D . ; James, B.R. Inorg. Chem. 1995, 34, 2278. 2. Hunt, C.T. ; Balch, A . L . Inorg. Chem. 1981, 20, 2267. 3. a) Noack, K. Helv. Chim. Acta 1962, 45, 1847. b) Mittasch, A. Angew. Chem. 1928, 41, 827. 4. Calderazzo, F.; L'Eplattenier, F. Inorg. Chem. 1967, 6, 1220. 5. a) Colton, R.; Tomkins, LB. Austral. J. Chem. 1966,19, 1143. b) Ankes, M . W . ; Colton, R.; Tomkins, LB. Austral. J. Chem. 1967, 20, 9. 6. a) Abel, E.W.; Wilkinson, G. J. Chem. Soc. 1959, 1501. b) Brimm, E .O. ; Lynch, M . A . ; Sesny, W.J. J. Am. Chem. Soc. 1954, 76, 3831. 7. Hileman, J.C.; Huggins, D . K . ; Kaesz, H.D. Inorg. Chem. 1962, 1, 933. 8. Hieber, W.; Schuk, R.; Fuchs, H.Z. ^worg. Chem. 1941, 245, 243. 9. Cotton, F .A. ; Johnson, B.F.G. Inorg. Chem. 1967, 6, 2113. 10. Johnson, B .F .G. ; Johnston, R.D.; Josty, P.L.; Lewis, J.; Williams, I.G. Nature 1967, 213, 901. 11. Johnson, B .F .G. ; Lewis, J. J. Chem. Soc. (A) 1968, 2859. 12. Cook, N . ; Smart, L . ; Woodward, P. J. Chem. Soc, Dalton Trans. 1977, 1744. 13. a) Noack, K.J. Organomet. Chem. 1968,13, 411. b) Farona, M.F . ; Camp, G.R. Inorg. Chim. Acta 1969, 3, 395. c) Dobson, G.R.; Jernigan, R.T.; Chang, P.-T. J. Organomet. Chem. 1973, 54, C33. 14. a) Haines, L.I.B.; Hopgood, D. ; Poe, A.J. J. Chem. Soc. (A) 1968, 421. b) Haines, L.I.B.; Poe, A.J. J. Chem. Soc. (A) 1969, 2826. 15. a) Kramer, G . ; Patterson, J.; Poe, A.J. J. Chem. Soc, Dalton Trans. 1979, 1165. b) Kramer, G . ; Patterson, J.; Poe, A.J . ; Ng, L. Inorg. Chem. 1980,19, 1161. 16. a) Crow, J.P.; Cullen, W.R.; Hou, F.L. Chem. Commun. 1971, 1229. b) Crow, J.P.; Cullen, W.R.; Hou, F.L. Inorg. Chem. 1972, 11, 2125. c) Cullen, W.R.; Hou, F.L. Inorg. Chem. 1975, 14, 3121. 17. Haines, R.J.; du Preez, A . L . J. Chem. Soc. (A) 1970, 2341. 18. Kramer, G . ; Poe, A.J. ; Amer, S. Inorg. Chem. 1981, 20. 1362. 189 Chapter 4 19. Bott, R.W.; Eaborn, C ; Waters, J.A. J. Chern. Soc. 1963, 681. 20. a) Chipperfield, JR.; Ford, I.; Webster, D.E . J. Chern. Soc, Dalton Trans. 1975, 2042. b) Chipperfield, J.R.; Hayter, A . C . ; Webster, D.E . J. Chern. Soc, Dalton Trans. 1975, 2048. c) Chipperfield, J.R.; Ford, J.; Hayter, A . C . ; Lee, D.J.; Webster, D.E . / . Chern. Soc, Dalton Trans. 1976, 1024. 21. a) Balch, A . L . J. Am. Chern. Soc. 1976, 98, 8049. b) Balch, A . L . ; Tulyathan, B. Inorg. Chern. 1977, 16, 2840. 22. Faraone, F.; Bruno, G . ; Schiavo, S.L.; Bombieni, G. J. Chern. Soc, Dalton Trans. 1984, 533. 23. Oro, L . A . ; Carmona, D. ; Perez; P.L; Esteban, M . ; Tiripicchio, A . ; Tiripicchio-Camellini, M . J. Chern. Soc, Dalton Trans. 1985, 973. 24. Hanson, B.E. ; Fanwick, P.E.; Mancini, J.S. Inorg. Chern. 1982, 21, 3811. 25. Colombie, A . ; Lavigne, G. ; Bonnet, J.-J. J. Chern. Soc, Dalton Trans. 1986, 899. 26. Schmidbauer, H . ; Franke, R. Inorg. Chim. Acta 1975,13, 85. 27. Farrar, D . H . ; Grundy, K.R. ; Payne, N . C . ; Roper, W.R.; Walker, A . J. Am. Chern. Soc 1979, 101, 6577. 28. Corbridge, D.E .C . "The Structure Chemistry of Phosphorus", Elsevier, Amsterdam, 1974, p. 7. 29 Baker, M.J. ; Giles, M.F . ; Orpen, A . G . ; Taylor, M.J. ; Watt, R J . J. Chern. Soc, Chern. Commun. 1995, 197. 30. Berry, D . E . ; Browning, J.; Dixon, K.R. ; Hitts, R.W. Can. J. Chern. 1988, 66, 1272. 31. Balakrishna, M.S. ; Klein, R.; Uhlenbrock, S.; Pinkerton, A . A . ; Cavell, R.G. Inorg. Chern. 1993, 32, 5676. 32. Appleton, T . G . ; Clark, H . C . ; Manzer, L .E . Coord. Chern. Rev. 1973,10, 335. 33. Zumdahl, S.S.; Drago, R.S. J. Am. Chern. Soc. 1968, 90, 6669. 34. See for example: a) Streitwieser, Jr., A . ; Heathcock, C H . "Introduction to Organic Chemistry, 3rd Ed." , Macmillan, New York, 1985, ch. 6. b) Morrison, R.T.; Boyd, R.N. "Organic Chemistry, 3rd Ed." , Allyn and Bacon, Boston, 1973, ch. 3. 190 Chapter 4 35. Permutter-Hayman, B. Progr. Inorg. Chem. 1976, 20, 229. 36. Basolo, F.; Pearson, R.B. "Mechanisms of Inorganic Reactions", Wiley, New York, 1967, pp. 351-453. 37. Vaska, L. Acc. Chem. Res. 1968,1, 335. 38. Collman, J.P.; Roper, W.R. Adv. Organomet. Chem. 1968, 7, 54. 39. Halpern, J. Pure Appl. Chem. 1969, 20, 59. 40. a) Barnabas, A.F . ; Sallin, D . ; James, B.R. Can. J. Chem. 1989, 67, 2009. b) Xie, L .Y . ; James, B.R. Inorg. Chim. Acta 1994, 217, 209. 41. Pincock, J.A.; Yates, K. Can. J. Chem. 1970, 48, 3332. 42. Pincock, J.A.; Yates, K. Can. J. Chem. 1970, 48, 2944. 43. a) Chart, J.; Duncanson, L . A . ; Venanzi, L . M . J. Chem. Soc. 1955, 4456. b) Orgel, L .E . J. Inorg. Nucl. Chem. 1956, 2, 137. 44. Cowie, M.;Dwight, S.K.Inorg. Chem. 1980, 19, 3500. 45. Mague, J.T. Inorg. Chem. 1969, 8, 1975. 46. Barnabas, A.F. M.Sc. Dissertation, University of British Columbia, Vancouver, 1989. 47. Meislich, H . ; Nechamkin, H . ; Sharefkin, J. "Theory and Problems of Organic Chemistry, 2nd Ed . " , McGraw-Hill, Toronto, 1991, ch. 4, p. 57. 48. Carey, F .A . ; Sundberg, R.J. "Advanced Organic Chemistry, 3rd Ed. Part A : Structure and Mechanisms", Plenum Press, New York, 1993, ch. 4, p. 232. 49. Koppel, I.A.; Palm, V . A . "Advances in Linear Free Energy Relationships", N.B. Chapman and J. Shorter, Eds., Plenum Press, London, 1972, ch. 5. 50. See for example: a) Piper, T.S. ; Wilkinson, G. J. Am. Chem. Soc. 1956, 78, 900. b) Khattab, S.A.; Marko, L . ; Bor, G . ; Marko, B. J. Organomet. Chem. 1964,1, 373. c) Schunn, R.A. ; Fritchie, Jr., C.J.; Prewitt, C.T. Inorg. Chem. 1966, 5, 892. d) Schrauzer, G . N . ; Mayweg, V.P. ; Finck, H.W.; Heinrich, W. J. Am. Chem. Soc. 1966, 88, 4604. e) Fackler, Jr., J.P.; Coucouvanis, D . ; Fetchin, J.A.; Seidel, W.C. J. Am. Chem. Soc. 1968, 90, 2784. f) Balch, A . L . J. Am. Chem. Soc. 1969, 91, 6962. g) 191 Chapter 4 Epstein, E.F. ; Bernal, I. J. Organomet. Chern. 1971, 26, 229. h) Ginsberg, A.P. ; Lindsell, W.E. Chern. Commun. 1971, 232. i) Giannotti, C ; Fontaine, C. ; Septe, B.; Doue, D. J. Organomet. Chern. 1972, 39, C74. j) Giannotti, C ; Ducourant, A . M . ; Chanaud, H . ; Chiaroni, A . ; Riche, C. J. Organomet. Chern. 1977, 140, 289, and references therein. 51. Balch, A . L . ; Benner, L.S. ; Olmstead, M . M . Inorg. Chern. 1979, 78, 2996. 52. Muralidharan, S.; Espenson, J.H. J. Am. Chern. Soc. 1984,106, 8104. 53. Muralidharan, S.; Espenson, J.H.; Ross, S.A. Inorg. Chern. 1986, 25, 2557. 54. Williams, C.R.; Harpp, D . N . Sulfur Lett. 1993, 16, 63. 55. Brown, R.L. J. Chern. Phys. 1966, 44, 2827. 56. Wayne, F .D. ; Davies, P.B.; Thrush, B.A. Mol. Phys. 1974, 28, 989. 57. Meyer, B. Chern. Rev. 1976, 76, 367. 58. Laitinen, R.S.; Pekonen, P.; Suontamo, R.J. Coord. Chern. Rev. 1994,130, 1. 59. See for example: a) Grim, S O . ; Walton, E.D. Inorg. Chern. 1980, 19, 1982. b) Lusser, M . ; Peringer, P. Inorg. Chim. Acta 1987,127, 151. c) Bond, A . M . ; Colton, R.; Ebner, J.; Ellis, S.R. Inorg. Chern. 1989, 28, 4509, and references therein. 60. Although a value for the Pd-P bond strength cannot be found, the bond energy is probably comparable to that of the Pt-P bond (417 kJ mol"1). See, for example, Smoes, S.; Huguet, R.; Drowart, J. Z. Naturforsch TeilA, 1971, 26, 1934. 61. Meyer, B.; Jensen, D. ; Oommen, T. "Sulfur in Organic and Inorganic Chemistry, Vol. 2", A . Senning, Ed., Marcel Dekker, New York, 1972, ch. 12. 62. Johnson, D . A . "Sulfur in Organic and Inorganic Chemistry, Vol. 2", A. Senning, Ed., Marcel Dekker, New York, 1972, ch. 13. 63. Strausz, O.P. "Sulfur in Organic and Inorganic Chemistry, Vol. 2", A. Senning, Ed. , Marcel Dekker, New York, 1972, ch. 11. 64. Gunning, H.E . ; Strausz, O.P. Adv. Photochem. 1966, 4, 143. 65. Schumann, H. "Sulfur in Organic and Inorganic Chemistry, Vol. 3", A. Senning, Ed., Marcel 192 Chapter 4 Dekker, New York, 1972, ch. 21. 66. Gordon, A.J. ; Ford, R A . "The Chemist's Companion", John Wiley & Sons, Toronto, 1972, p. 451. 67. STR92: Altomare, A . ; Burla, M . C . ; Camalli, M ; Cascarano, M . ; Giacovazzo, C ; Guagliardi, A . ; Polidori, G. J. Appl. Cryst., in preparation. 68. DIRDIF92: Beurskens, P.T.; Admiraal, G. ; Beurskens, G. ; Bosman, W.P.; Garcia-Granda, S.; Gould, R.O. ; Smits, J . M . M ; Smykall, C. "The DIFDIP program system, Technical Report of the Crystallography Laboratory", University of Nijmegen, The Netherlands. 69. Cromer, D.T . ; Waber, J.T. "International Tables for X-ray Crystallography, Vol. 4", The Kynoch Press, Birmingham, England, 1974, Table 2.2A. 70. Ibers, J.A.; Hamilton, W.C. Acta Crystallogr. 1964, 17, 781. 71. T E X S A N : Crystal Structure Analysis Package, Molecular Structure Corporation (1985 & 1992). 72. Peringer, P.; Schwald, J. J. Chem. Soc, Chem. Commun. 1986, 1625. 193 CHAPTER 5 Preliminary Studies on the Reaction of PdX2(dpm) with H 2S in Dimethylsulfoxide; Formation of Pd2X2(u,-S)(dpm)2 194 Chapter 5 5.1 Introduction Previously in this laboratory, during investigations of the sulfur abstraction from H2S with the Pd2X2(dpm)2 complexes (1) (X = halide) (reaction 5.1.1), the mononuclear, Pd(H) complexes PdX2(dpm) (9) were tested for reactivity toward H2S. It was discovered that in benzene or in CH 2Cl2, reaction of PdCl2(dpm) (9a) with H 2S proceeded only in the presence of a base, for example Et3N, to form the dinuclear complex Pd2(SH)2(p-S)(dpm)2 (reaction 5.1.2).1 (5.1.1) Pd2X2(dpm)2 (1) + H 2S -> Pd2X2(u-S)(dpm)2 (2) + H 2 E t 3 N (5.1.2) 2PdCl2(dpm) + 3H 2 S Pd2(SH)2(u.-S)(dpm)2 + 4 HCl (as Et 3 NH + Cl j In the present thesis work, reaction 5.1.2 was explored further using DMSO as solvent in order to enhance the nucleophilicity of SH" without the need for basic conditions,2'3 and reaction 5.1.3 was discovered. The important implications are seen with the X = I system; photodecomposition of HI, for example, with solar irradiation, produces H 2 and the I2, which can react with 2 via reaction 5.1.4 (see Chapter 4). The net reaction would be the overall homogeneous, photocatalytic decomposition of H2S to give H2 and sulfur (reaction 5.1.5). Preliminary work was carried out to substantiate the catalytic nature of reaction 5.1.5, and the results are summarized in this chapter. (5.1.3) 2 PdX2(dpm) (9) + H 2S -> Pd2X2(p-S)(dpm)2 (2) + 2 H X (5.1.4) Pd2X2(p-S)(dpm)2 + X 2 -> 2PdX2(dpm) + sulfur (5.1.5) H 2S — H 2 + sulfur 195 References on page 234 Chapter 5 5.2 Results NMR spectroscopic data for the complexes and other compounds relating to the studies described in this chapter are given in Table 5.1. Reaction of PdX2(dpm) (9) with H2S. Reaction of PdX2(dpm) (9a, 1.8 x 10"2 M ; 9b, 1.5 x 10"2 M ; 9c, 1.3 x 10"2 M) with 0.5, 0.75, 1.0, or 1.5 mole equivalent H 2S was studied at R.T. in CDCK and in DMSO-d6 in both the absence and presence of light (either laboratory light or sunlight). In CDCI3, 9 does not exhibit reactivity with H 2S, remairiing unchanged even after 27 d exposure to light. Supplementary GC and TLC (CHC13 eluent) studies did not reveal the presence of H 2 or elemental sulfur, respectively. In DMSO-d6, however, 9 reacts immediately with H 2S to form, depending on conditions, some Pd2X2(p-S)(dpm)2 (2) and HX(solv.); Table 5.2 summarizes the experimental results. Upon addition of H 2S, the orange solution of 9c or the yellow solutions of 9b or 9a rapidly changed colour to brown. ! H and 31P{1H} NMR spectroscopic analyses revealed varying quantities of Pd2X2(dpm)2 (1), Pd2X2(u-S)(dpm)2 (2), unreacted 9, and an unknown species Y. The ' H NMR singlet of H 2S at 8 1.99 is not seen. A broad singlet, however, is observed at various positions ranging from 8 3.79 to 6.14 in the iodide system and at 8 3.72 and 3.70 in the bromide and chloride systems, respectively; the residual H 2 0 peak invariably present in the DMSO-d6 used is no longer seen at its usual position of 8 3.42 and is considered to be part of the broad singlet. Figures 5.1-5.3 show representative ' H and 31P{1H} NMR spectra for each of the three systems. The unknown species Y is halogen-independent and is characterized by a 31P{1H} singlet at 8 -40.4 and an unresolved broad lH triplet at 8 5.10. Furthermore, this unknown species in the iodide system is most prominent (45 to 80% with respect to all 31P{1H} NMR-196 References on page 234 Chapter 5 Table 5.1. NMR data for Complexes and Other Compounds Discussed in Chapter 5. Compound" 8 (31P{1H})e Pd 2I 2(dpm) 2 (lc) 4.23 d e (4.0) -11.3 d 4.39 e / -I0.6f Pd2Cl2(n-S)(dpm)2 (2a) 2.79 d g (12.6, 3.5) 5.52 d 4.73 (12.6, 6.1) 3 .35 , / g 4.63 5.90 7 Pd2Br2((x-S)(dpm)2 (2b) 2.88 (12.8, 3.2) 5.96 d # 4.83 (12.8, 7.6) 6.14 • 3.36, / g 4.72 6.36' Pd2I2(n-S)(dpm)2 (2c) 3.06 d g (14.0, 3.0) 6.08 d 4.95 (14.0, 6.0) 3 .49 , / g 4.87 6.52f PdCl2(dpm) (9a) 4.21 * (10.8) -54.7 4.98 -51.7 f PdBr2(dpm) (9b) 4.37* (10.5) -56.2 5.04 ^ -51.5 ' Pdl2(dpm) (9c) 4.42* (10.0) -63.2 5.17 f-h -54.6 f PdCl2(dpm(S)) (11a) 4.07 ' 54.9, 30.9 (18.3)J 4.97 56.3, 34.0 / l 7 PdBr2(dpm(S)) (lib) 4.02 ' 56.6, 32.1 (20.4)' 4.97 ^ 57.6, 36.5 f J PdI2(dpm(S)) (11c) 3.71 ' 61.1,31.2(25.5)' 4 . 86 " 61.1, 3&.9fj Y (possibly Pd(SH)2(dpm)) 5.10/fc -40.4' Pd2(Cl)(Br)(u-S)(dpm)2 see foomote / see footnote / Pd 2I 2(SH) 2(dpm) 2 5.16 e / -1 .2 ' dpm(0)2 4.05 ^ 24 .1 ' H 2 S 0.82, 1 .99 / m H 2 0 3.42 f H 2 4.64 ' 197 References on page 234 Chapter 5 " The (i-symbol for the bridging diphosphine ligand(s) is omitted for convenience throughout this Table and the text. * In C D C I 3 , unless stated otherwise, at 20 °C with respect to TMS; JHH (and J P H ) values in Hertz are given in parentheses; signals for CH 2 protons unless indicated otherwise. c Singlets in CDC13 at 20 °C with respect to 85% H3PO4, downfield being positive. ' inCDjCh. 'Quintet. / InDMSO-d 6 . g Doublets of quintets for each of 2 sets of CH 2 protons. * Triplet.' Pseudo-triplet. 3 AB pattern, J P P values in Hertz given in parentheses. * Broad triplet. ' See Figure 5.5. m Measured using 1 atm H2S. 198 References on page 234 Chapter 5 Table 5.2. Results from studies on the reaction of PdX2(dpm) (9a, 1.8 x 10'2 M ; 9b, 1.5 x 10"2 M ; 9c, 1.3 x 10"2 M) with H 2S at R.T. in DMSO-de; 1 = Pd2X2(dpm)2, 2 = Pd2X2(p-S)(dpm)2,11 = PdX2(dpm(S)), Y = probably Pd(SH)2(dpm). System 1 2 9 11 Y H 2 " sulfur* 9c + 1.5 H2S 0 25^  0 0 75 - -NMR-technique17 9c + 1 H2S 0 20 0 0 80 - -NMR-technique 9c + 0.75 H2S 2 0 89 0 9 - -Schlenk-technique6 +5 d absence of light 2 0 89 0 9 no -9c + 0.5 H2S 0 10 45 0 45 - -NMR-technique +2 d exposure to light 0 0 68 10 22 - yes 9c + 0.5 H2S 6 22 63 0 9 - -Schlenk-technique +5 d exposure to light 0 0 63 37 0 - -+27 d exposure to light 0 0 42 58 0 yes(?/ -9b + 0.5 H2S 0 23 20 0 57 - -Schlenk-technique +8 d exposure to light 0 0 74 26 0 no yes 9a + 0.5 H2S 0 24 39 0 37 - -Schlenk-technique yes(?/ +5 d exposure to light 0 0 50 20 30 -9a + 1 H2S 0 25 25 29 21 yes(?/ -NMR-technique " Detection of H 2 using gas chromatography. * Detection of elemental sulfur using thin-layered chromatography (CHC13 eluent).c NMR-techmque refers to injection of H2S via syringe into DMSO-d6 solution of 9 held in an NMR tube. d Estimated amounts (%) of all the Pd complexes determined from 31P{'H} NMR spectra.' Schlenk-technique refers to injection of H2S via syringe into DMSO-d6 solution of 9 held in a Schlenk tube of 165.0 mL volume. / rThe presence of H 2 was not positively established (see text). 199 References on page 234 Chapter 5 Fig. 5.1. Representative *H (200 MHz) and ^ Pj'H} (81 MHz) NMR spectra showing results from the immediate reaction at R.T. in DMSO-de of Pdl2(dpm) (9c, 1.3 x 10"2 M) with 0.5 mole equivalent H 2S in the absence of light (Schlenk techniques were employed); lc = Pd2I2(dpm)2, 2c = Pd2I2(p-S)(dpm)2, Y = probably Pd(SH)2(dpm). I • ' ' ' l 9.0 8.0 9c, Y DMSO i i i i i l i • • i l i i i i l i i i — i | i i 3.0 2.0 ^ — ' 7.0 6.0 5.0 4.0 PPM 2c lc 9c i i i I 0 -I—.—•—•—i—|—i—i—i—.—i— -10 -20 -30 PPM • I I | 1 -60 10 -40 -50 200 References on page 234 Chapter 5 Fig. 5.2. Representative ! H (300 MHz) and "Pf'H} (121 MHz) NMR spectra showing results from the immediate reaction at R T . in DMSO-d6 of PdBr2(dpm) (9b, 1.5 x 10"2 M) with 0.5 mole equivalent H 2S in the absence of light (Schlenk techniques were employed); 2b = Pd2Br2(p-S)(dpm)2, Y = probably Pd(SH)2(dpm). 201 References on page 234 Chapter 5 Fig. 5.3. Representative lU (300 MHz) and 31P{'H} (121 MHz) NMR spectra showing results from the immediate reaction at R.T. in DMSO-d6 of PdCl2(dpm) (9a, 1.8 x 10"2 M) with 0.5 mole equivalent H 2S in the absence of light; Schlenk techniques were employed; 2a : Pd2Cl2(u.-S)(dpm)2, Y = probably Pd(SH)2(dpm). DMSO - 1 — i — i — i — i — 1 r-4.0 9.0 8.0 7.0 6.0 5.0 PPM 3.0 2.0 "Pl'H} 2a 9a 10 -10 -20 -30 -40 -50 -60 PPM 202 References on page 234 Chapter 5 detectable species) when H 2S is introduced into a solution of 9c held in an NMR tube (typically of 2-3 mL volume). On the other hand, the yield of species Y is significantly reduced (9%) if H 2 S is adrninistered into a Schlenk tube of much greater volume (i.e. 165.0 mL) containing the same solution volume (0.50 mL) and concentration (1.3 x 10"2 M) of 9c. In the bromide and chloride systems, species Y forms in larger yields of 57 and 37%, respectively, despite the use of similar Schlenk techniques. For systems where less than 1 mole equivalent H 2S is used, unreacted 9 ranges from 45 to 89% in the iodide system and 20 and 39% in the bromide and chloride systems, respectively. 9 reacts completely when 1 or more mole equivalent of H 2S is used, for example as shown for the iodide systems where 1 or 1.5 mole equivalent is introduced. Interestingly, Pd2I2(dpm)2 (lc) forms in two iodide systems in which Schlenk techniques are employed; the corresponding bromide and chloride analogues are not observed in the respective systems. Exposure to light over a few days (either laboratory light or sunlight) results in the disappearance of 1 and 2, the disappearance or reduction of species Y, an increase in 9, and the appearance of PdX2(dpm(S)) (11) (see Chapter 4). J H NMR spectra also reveal that in most systems the broad singlet mentioned above has reduced in intensity and shifted to higher fields; in some cases, the signal has completely disappeared and that of H 2 0 reappears. Figure 5.4 shows representative TF and "Pf^H} NMR spectra of the iodide system (reaction of Pdl2(dpm) (9c, 1.3 x 10'2 M) with 0.5 mole equivalent H 2S using Schlenk techniques) after exposure to laboratory light for 5 d. A control study of the reaction of 9c with 0.75 mole equivalent H 2S in the absence of light for 5 d (3rd entry in Table 5.2) revealed the necessity of light to affect further reaction. A few GC and TLC studies were carried out and, while the presence of elemental sulfur was always definitively identified after exposure to light by using TLC (CHCI3 eluent) and comparing the degree of retention to that of an authentic sample of elemental sulfur, the presence of H 2 remained uncertain although in the gas 203 References on page 234 Chapter 5 Fig. 5.4. Representative *H (200 MHz) and 31P{!H} (81 MHz) NMR spectra showing results from the exposure of the iodide system (reaction of Pdtydpm) (9c, 1.3 x 10"2 M) with 0.5 mole equivalent H 2S in the absence of light using Schlenk techniques) to laboratory light for 5 d; 11c: PdI2(dpm(S)). H 20 9c 11c DMSO - 1 — 3.0 - 1 — 1 2.0 8.0 7.0 6.0 5.0 4.0 PPM "Pf'H} 11c 9c 60 • , - i _ r 20 i - T - i --40 40 0 PPM -20 -60 204 References on page 234 Chapter 5 chromatogram, a positive peak, with a retention time of 1.6 min and attributable to H 2 , was sometimes seen. Titration studies were carried out for the chloride system to substantiate and quantify the amount of HCl produced in reaction 5.1.3. 9a (9.9 x 10"3 M) was reacted in DMSO (4.5 mL) at R.T. with 0.5 mole equivalent H 2S using Schlenk techniques. After 4 h, H 2 0 was added to precipitate the Pd species, which were subsequently filtered off. The filtrate was then titrated with a standardized NaOH solution (0.010 M); the number of moles of HCl was deterrnined to be 1.8 x 10"5, which corresponds to a 41 % yield with respect to 9a. Reaction of PdX2(dpm) (9) with elemental sulfur. Reaction of PdX2(dpm) (9a, 3 6 x 10 2 M ; 9b, 3.0 x 10"2 M ; 9c, 2.7 x 10"2 M) with 1 mole equivalent S8 was studied on the NMR-scale at R T . and at 50 °C in DMSO-d6, in the absence or presence of light (either laboratory light or sunlight), for periods up to 29 d. ! H and 31P{1H} NMR spectroscopic analyses revealed the formation of one product, PdX2(dpm(S)) (11), and Table 5.3 summarizes the results of these studies. Reaction of 9 with Sg is seen to take place under purely thermal conditions, and in the iodide system, the presence of light has an effect, increasing the yields of 11c 10 to 15%. The yields of 11 decrease proceeding from the iodide to the bromide to the chloride system. For example, at R T . in the absence of light, 11c is present in 20% yield after 5 d, l i b , 15% yield after 9 d, and 11a, 10% yield after 10 d. The yields are significantly increased when the reactions are carried out at 50 °C where 11c is formed in 45% yield after 18 h, l i b , 30% yield after 1 d, and 11a, 25% yield after 1 d. 205 References on page 234 Chapter 5 Table 5.3. Results from studies on the reaction of PdX2(dpm) (9a, 3.6 x 10'2 M ; 9b, 3.0 x 10"2 M ; 9c, 2.7 x 10"2 M) with 1 mole equivalent of elemental sulfur in DMSO-d6. System Conditions" Unreacted PdX2(dpm(S)) yield PdX2(dpm) (9) (%)* (11) (%)* Pdl2(dpm) (9c) + 1/8S8 5 d; absence; R.T. 80 20 5 d; presence; R.T. 70 30 12 d; absence; R.T. 80 20 12 d; presence; R.T. 65 35 29 d; presence; R.T. 40 60 18 h; absence; 50 °C 55 45 2 d; absence; 50 °C 50 50 PdBr2(dpm) (9b) + 1/8S8 9 d; absence; R.T. 85 15 1 d; absence; 50 °C 70 30 PdCl2(dpm) (9a) + 1/8S8 1 d; absence; R.T. 100 trace 10 d; absence; R.T. 90 10 1 d; absence; 50 °C 75 25 " Absence and presence refer to experiments performed in the absence or presence of light (either laboratory- or sunlight), respectively. * The amounts of PdX2(dpm) and PdX2(dpm(S)) are determined from the relative integration values of the lU NMR triplet and pseudo-triplet signals, respectively, for the iodide system, and from the relative integration values of the 31P{'H} NMR singlet and set of AB doublets, respectively, for the bromide and chloride systems. 206 References on page 234 Chapter 5 Reaction of Pd2I2(u-S)(dpm)2 (2c) with I2. Reaction of Pd2I2(p-S)(dpm)2 (2c) (1.6 x 10"2 M) with 1 mole equivalent I2 was studied at R.T. in DMSO-d6. Addition of a red solution of I2 to a brown solution of 2c immediately produced a green-black solution which slowly changed to a final red-brown colour after -20 min. The sample was analyzed after 24 h using NMR spectroscopy and thin-layered chromatography (CHCI3 eluent). ! H and 31P{1H} NMR spectra revealed the presence of two species, 9c and 11c in 51 and 49% yields; TLC analysis revealed no sulfur present. Reaction of Pd2X2(p,-S)(dpm)2 (2) with HX. Reaction of Pd2X2(p-S)(dpm)2 (2a, 1.8 x 10"2 M ; '2b, 1.7 x 10"2 M ; 2c, 1.5 x 10"2 M) with HX was studied at R.T. in CDC13 and in DMSO-de. The results of these studies are summarized in Table 5.4. In CDCI3, 2 mole equivalents of HCl(g) and HBr(g) were added to solutions of 2a and 2b, respectively, and the samples were analyzed after 24 h using NMR spectroscopy. ! H and 31P{1H} NMR spectra revealed that 2a and 2b have completely reacted forming 9a and 9b, respectively, and H 2S. H 2S is characterized by a ! H singlet signal at 5 0.82 and by its characteristic odor resembling rotten eggs. Exposure to light (laboratory light or sunlight) for 1 or 5 d had no effect on the reaction; corresponding GC and TLC studies revealed no H 2 or elemental sulfur. Similar studies were also carried out using DMSO-d6. For the chloride system, NMR spectroscopic analyses after 24 h revealed, in addition to unreacted 2a, the presence of 9a, 11a, and a non-phosphorus-containing species characterized by a broad ! H singlet at 5 6.26. For the bromide system, similar findings were obtained; 9b, lib, and a species characterized by a broad lH singlet at 6 6.14 were observed, in addition to unreacted 2b. Exposure of the DMSO systems to light (from an 18 W TLC Hg vapour lamp) for 17 h resulted in the complete disappearance of 2a and 2b; 9a, 9b, 11a, and l ib were still observed and in the same relative yields. Furthermore, the 207 References on page 234 Chapter 5 Table 5.4. Results from studies on the reaction of Pd2X2(|x-S)(dpm)2 (2a, 1.8 x 10"2 M ; 2b, 1.7 x 10"2 M ; 2c, 1.5 x 10"2 M) with HX(g). System Conditions" Observed species* 2a + 2 HCl CDC13; R.T. CDC13; R.T.; 1 and 5 d exposure DMSO-ds; R.T. DMSO-dg; R.T.; 17 h exposure1* 9a, H2S 9a, H2S 2a (14%), 9a (41%), 11a (45%), b.s at 5 6.26c 9a (49%), 11a (51%), sulfur, b.s. at 8 4.67c 2b + 2 HBr CDC13; RT . CDC13; R.T.; 1 and 5 d exposure DMSO-dg; RT. DMSO-de; R.T.; 17 h exposure'' 9b, H2S 9b, H2S 2b (58%), 9b (21%), l i b (21%), b.s. at 5 6.14c 9b (50%), l i b (50%), sulfur, b.s. at 8 4.09c 2c + ffl" DMSO-dg; R.T/ 9c (19%), 11c (41%), dpm(0)2 (40%), sulfur, H 2 g 2b + 2 HCl DMSO-dg; R.T. DMSO-d6; R.T.; 1 d exposure Pd2(Br)(Cl)((4.-S)(dpm)2, 2a (?), 2b(?), b.s. at 8 8.69c Pd2(Br)(Cl)(^-S)(dpm)2, 2a(?), 2b(?) > Pd(Br)(Cl)(dpm), 9a(?), 9b(?) > Pd(Br)(Cl)(dpm(S)), lla(?), llb(?), b.s. at8 5.86c " Exposure refers to exposure of the sample to light (laboratory light or sunlight). * Observed species from 'H and 31P{'H} NMR spectroscopic and thin4ayered and gas chromatographic analyses. Values in parentheses refer to approximate relative yields determined from 31P{'H} NMR integration values. c Broad singlet observed in the ] H NMR spectrum. d Exposure to light from an 18 W TLC Hg vapour lamp. " Hydrogen iodide (0.15 M) was introduced as HI(aq.) or HI(g). f Subsequent exposure to light for 5 d gave no further changes. g H 2 was detected only in reaction of 2c with HI(g). 208 References on page 234 Chapter 5 broad ! H singlets at 8 6.26 and 6.14 were replaced by ones at 8 4.67 and 4.09 in the chloride and bromide systems, respectively. Corresponding GC and TLC studies revealed the absence of H 2 and presence of elemental sulfur, respectively, in both systems. Reaction of 2c with HI was studied only in DMSO-oV Due to the corrosive nature of aq. HL 10 mole equivalents (10 uL 7.58 M aq. HI) were used (see Experimental Section). Use of aq. HI or HI(g) (prepared from aq. HI, see Experimental Section) gave similar results. ' H and 31P{1H} NMR analyses after 24 h revealed that 2c had completely reacted, and 9c, 11c, and dpm(0)2 had formed in 19, 41, and 40% relative yields, respectively. Exposure of this system to light for 5 d had no effect on the NMR spectral characteristics. Subsequent TLC studies revealed the presence of elemental sulfur. GC studies indicated the presence of H 2 only for the reaction of 2c with HI(g); it was noted that when HI(g) was introduced into the system (i.e. an NMR tube containing a solution of 2c) via vacuum transfer using liquid N 2 , a purple solid (attributable to I2) was seen condensing on the inside walls of the NMR tube. Finally, reaction of 2b with 2 mole equivalents of HCl(g) was investigated, and NMR spectroscopic analyses after 24 h revealed the formation of a species attributable to Pd2(Cl)(Br)(p-S)(dpm)2 which is characterized by multiplets at 8 3.37 and 4.65 in the ! H NMR spectrum and an A A ' B B ' pattern from 8 7.0 to 3.0 in the31P{1H} NMR spectrum (Fig. 5.5). In addition, a broad singlet at 8 8.69 in the ! H NMR spectrum is observed. Whether 2a and 2b are also present cannot be ascertained. NMR analyses after 1 d exposure to light, however, revealed in addition to Pd2(Cl)(Br)(p.-S)(dpm)2, the presence of a mixture consisting of perhaps Pd(Cl)(Br)(dpm), 9a, 9b, 11a, lib, and Pd(Cl)(Br)(dpm(S)) (Fig. 5.6). The broad singlet in the *H NMR spectrum at 8 8.69 has disappeared and one at 8 5.86 has appeared. In all of the systems studied in DMSO-d6, the odor of H 2S was not detected nor was the H 2S singlet seen at 8 1.99 in ' H NMR spectra. 209 References on page 234 Chapter 5 Fig. 5.5. *H (200 MHz) and 31P{!H} (81 MHz) NMR spectra showing results from the reaction of Pd2Br2(p-S)(dpm)2 (2b, 1.7 x 10"2 M) with 2 mole equivalent HCl at R.T. in DMSOd-6; spectra recorded 1 h after sample preparation. Pd2(Cl)(Br)(n-S)(dpm)2 A DMSO 9.0 8.0 7.0 6.0 5.0 PPM 4.0 3.0 2.0 31ml Pd2(Cl)(Br)(n-S)(dpm)2 8.0 6.0 4.0 PPM 210 References on page 234 Chapter 5 Fig. 5.6. ! H (200 MHz) and ^ Pj^H} (81 MHz) NMR spectra showing results from the reaction of Pd2Br2(p.-S)(dpm)2 (2b, 1.7 x 10"2 M) with 2 mole equivalent HCl at R T . in DMSO-de; spectra recorded after exposure of the system to laboratory light for 1 d; 9a = PdCl2(dpm), 9b = PdBr2(dpm), 11a = PdCl2(dpm(S)), l i b = PdBr2(dpm(S)). Pd(Cl)(Br)(dpm), Pd(Cl)(Br)(dpm(S)), 9a(?), 9b(?), lla(?), llb(?) HCI(aq.), HBr(aq.), H2S(aq.) DMSO Pd2(Cl)(Br)(u-S)(dpm)2 9.0 8.0 7.0 6.0 5.0 4.0 PPM 3.0 2.0 Pd(Cl)(Br)(dpm(S)), lla(?), llb(?) Pd2(Ci)(Br)(^-S)(dpm)2 Pd(Cl)(Br)(dpm), 9a(?),9b(?) J L 60 40 20 -20 -40 PPM 211 References on page 234 Chapter 5 Reaction of Pd2(SH)2(u-S)(dpm)2 with I2. Reaction of Pd2(SH)2(p-S)(dpm)2 (1.9 x 10"2 M) with 1 mole equivalent I2 was studied at R T . in CDCI3 and in DMSO-d6. Addition of a purple (CDCI3) or red (DMSO-de) solution of I2 to a brown solution of Pd2(SH)2(p-S)(dpm)2 immediately produced a brown-black solution which slowly changed colour to a final brown after ~30 min. In CDCI3, ! H and 31P{1H} NMR spectroscopic analyses revealed the formation of two detectable products, Pd2I2(p-S)(dpm)2 (2c) and H 2S, the latter being characterized by smell and the *H singlet at 5 0.82. In DMSO-de, 2c was formed; the 5(H2S) signal at 1.99 was not seen in the ! H NMR spectrum, but a broad singlet at 5 5.68 was observed (Fig. 5.7). Also, the characteristic odor of H 2S was not detected. Unreacted Pd2(SH)2(u.-S)(dpm)2 was not seen. TLC analyses (CHCI3 eluent) of all samples revealed the presence of elemental sulfur. Reaction of Pd2(SH)2(p-S)(dpm)2 with I2 in CDCI3 was also studied at -60 °C using NMR spectroscopy. "Immediate" NMR spectra revealed that Pd2(SH)2(p-S)(dpm)2 had completely reacted to form 2c, H 2S, and an intermediate species characterized by a 1:4:6:4:1 *H quintet at 8 5.16 and a 31P{1H} singlet at 8 -1.2. There was no change in the spectra when temperature was raised to -45 °C; at 10 °C, the intensities of the NMR signals were reduced (with respect to 2c), and at 22 °C the intermediate had completely disappeared. Corresponding TLC studies again revealed the presence of elemental sulfur. Reaction of PdX2(dpm) (9) with NaSH. Reaction of PdX2(dpm) (9a, 3.6 x 10"2 M ; 9b, 3.0 x 10"2 M ; 9c, 2.7 x 10'2 M) with 0.5 mole equivalent NaSH was studied at R.T. in DMSO-de. After 24 h, ! H and 31P{'H} NMR spectra revealed in addition to unreacted 9, the presence of small amounts of Pd2X2(p-S)(dpm)2 (2) and the unknown Y (Fig. 5.8). 212 References on page 234 Chapter 5 Fig. 5.7. ' H (300 MHz) and3 T^H} (121 MHz) NMR spectra of the completed reaction between Pd2(SH)2(p,-S)(dpm)2 (1.9 x 10"2 M) and 1 mole equivalent I2 at R.T. in DMSO-de; 2c = Pd2I2(p-S)(dpm)2. — , — • — i — i — i — i — 9.0 8.0 i i i i i i -r 7.0 1—i—1 6.0 - i — i — r DMSO 5.0 PPM 4.0 3.0 2.0 1.0 3 1 p { l H j 2c r-= 1———^—»—, 60 50 40 30 20 10 0 -10 -20 -30 -40 -50 PPM 213 References on page 234 Chapter S Fig. 5.8. 3IP{XH} NMR spectra (121 MHz) showing the partial formation of Pd2X2(p-S)(dpm)2 (2) from the reaction of PdX2(dpm) (9) with 0.5 mole equivalent NaSH at R T . in DMSO-dg; Y = probably Pd(SH)2(dpm). PdCl2(dpm) (9a) Pd2Cl2(n-S)(dpm)2 (2a) , Pd2Br2(n-S)(dpm)2 (2b) PdBr2(dpm) (9b) Y Pd2I2(u-S)(dpm)2 (9c) Pdl2(dpm) (9c) 20 10 -10 -20 PPM 214 -30 -40 -50 References on page 234 Chapter 5 Photodecomposition studies of HI, HBr, and H2S. Qualitative photodecomposition studies were carried out, and Table 5.5 summarizes the results. UV-vis spectroscopy was used to analyze for Br2,12, and elemental sulfur (see Fig. 4.4, page 148), and gas chromatography for H 2 . In CDCI3, within 0.5 h after the introduction of aq. HL the system acquired a purple colour indicative of I2; the UV-vis spectrum shows an absorption maximum at 512 nm (Fig. 5.9). Exposure to laboratory light for 5 d resulted in an increase in the intensity of the purple colour and an increase in absorbance at 512 nm. GC analyses of the liquid and head space, however, revealed no detectable H2. Similar results were obtained in analogous studies of aq. HI alone or dissolved in DMSO-d6 except a dark red colour was observed instead; UV-vis spectroscopic studies in DMSO-d6 revealed absorption maxima at 294 and 364 nm characteristic of I3" (compared with an authentic sample of Nal3 prepared from Nal +12) (Fig. 5.10). There were again increases in intensities of colour and absorption after exposure to light. Exposure of FfJ(g) to sunlight for 2 d also produced I2 but no detectable H 2 ; irradiation with light from a 450 W medium pressure Hg lamp, however, resulted in complete decomposition to H 2 and I2. H 2 was positively detected and the absorbance at 512 nm corresponded to -0.020 mmol I2, in excellent agreement with the theoretical yield of 0.020 mmol. Exposure of H2S(g) to sunlight for 5 d produced no detectable H 2 or elemental sulfur. In a single study in which aq. HBr was dissolved in DMSO-de with the sample exposed to laboratory light for 12 h, the formation of Br 2 was evidenced by the appearance of a yellow colour and by comparison of the electronic absorption spectrum with that of an authentic sample of Br2 (Fig. 5.11). Both spectra show an absorption maximum at 418 nm. H2 was not detected in GC studies. 215 References on page 234 Chapter 5 Table 5.5. Results from photodecomposition studies of HI, HBr, and H 2S. System" Conditions* Decomposition products' Aq. HI (10 uL) in 0.50 mL CDC13 laboratory light; 0.5 h laboratory light; 5 d I2 I2 Aq. HI (10 uL) in 0.50 mL DMSO-de laboratory light; 5 d I 2(asl 3) Aq. HI (0.50 mL) sunlight; 1 d laboratory light; 5 d I 2(asl 3) I2 (asI3) HI(g) (prepared from 10 uL aq. mfe sunlight; 2 d 450 WHg lamp; 20 h I2 H 2 ,1 2 Aq. HBr (10 uL) in 0.50 mL DMSO-dg laboratory light; 0.5 d Br2 H2S(g)(1.0mLatSTP)e sunlight; 5 d none detected 0 Samples were prepared in NMR tubes fitted with a PTFE J. Young valves. * Exposure to light for specified time period. c Decomposition products were detected using UV/vis spectroscopy (Br2,12) and gas chromatography (H2). d See Experimental Section. e Gaseous samples were used. 216 References on page 234 Chapter 5 Fig. 5.9. Electronic absorption spectrum of I2 in chloroform from the photodecomposition of aq. HI (0.040 mmol) (amount of I2 not quantified). 217 References on page 234 Chapter 5 218 References on page 234 Chapter 5 Fig. 5.11. Electronic absorption spectra of an authentic sample of Br2 in DMSO-d6 and Br2 from photodecomposition of HBr (10 uL 8.83 M HBr(aq.) dissolved in 0.5 mL DMSO-de; 12 h exposure to laboratory light) (amount of Br2 not quantified). 0.000 ~| i i i i | i i i i | " i I I ' i | I I ' 1 | 350 400 450 500 550 600 wavelength (nm) 219 References on page 234 Chapter 5 5.3 Discussion Preliminary investigations of the reaction of PdX2(dpm) (9) with H 2S were carried out, and while 9 exhibits no reactivity toward H 2S in CDCI3, rapid reaction occurs in DMSO-d6 fonriing Pd2X2(p-S)(dpm)2 (2), a species Y which is probably Pd(SH)2(dpm) (see below), and H X (reaction 5.3.1). H S (5.3.1) PdX2(dpm)(9) 2 * Pd2X2(u-S)(dpm)2 (2) + Y + HX DMSO-de Reaction 5.3.1 was studied using NMR- and Schlenk-experimental techniques which refer to the addition of H 2S to solutions of 9 held in NMR and Schlenk tubes, respectively. The yields of 2, Y, HX, and unreacted 9 were dependent on these experimental conditions; Tables 5.1 and 5.2 summarize NMR data and the results of these studies, respectively. The presence of HX was established by titration studies on the chloride system; a 41% yield was observed (initial [9a] = 9.9 3 2 x 10" M) compared to 61% seen in an in situ NMR study where the initial [9a] = 1.8x10" M (see Table 5.2, entry 7). In all three CI", Br", and Y systems, *H NMR spectroscopic analyses showed broad singlets which vary in position from experiment to experiment even within duplicate runs. Furthermore, the DMSO-d6 solvent used, despite extensive drying, invariably contained residual H 2 0 whose *H NMR signal disappeared after addition of H 2S. It is inferred that the broad singlets observed in reaction 5.3.1 belong to aquated HX (i.e. H 30 +X"); however, as will be discussed later, H 2S could also be present in the aquated form, H 3 0 + SH", its ! H NMR signal being part of the broad singlet. Of note, ! H NMR analyses of aq. HI in DMSO-d6, for example, showed a broad singlet which also varies in position from sample to sample, and the *H NMR signal of H 2 0 was no longer seen. 220 References on page 234 Chapter 5 Species Y appears to be halide-independent as it possesses the same NMR characteristics in the three chloride, bromide, and iodide systems. Its yield, however, depended on the system and on the experimental technique used to study reaction 5.3.1. For example, in the iodide system, Y formed in higher yields when NMR-techniques were employed, while the use of Schlenk techniques greatly reduced the yield of Y but unreacted 9 was largely seen. In the chloride and bromide systems, Schlenk-techniques gave yields of Y that were comparable to those in the iodide system using NMR-techniques. Species Y is thought to be the mononuclear complex, Pd(SH)2(dpm), for which strong support comes from the reaction of 9 with 0.5 mole equivalent NaSH: this resulted in the formation of 2 and species Y, reaction 5.3.2, and the spectral data of Y are analogues to those of the PdX2(dpm) species (see Table 5.1). It is difficult to rationalize the limited results in terms of an intimate mechanism for reaction 5.3.1; however, analogies can be drawn from the previous study of the reaction of PdCl2(dpm) (9a) with H 2S under basic conditions in C&k or CH2C12 to form the dinuclear complex Pd2(SH)2(p-S)(dpm)2 (reaction 5.3.3).1 The presence of the Et3N base serves to react with H 2S to provide an S H source, and indeed, reaction of 9a with excess NaSH also forms Pd2(SH)2(p-S)(dpm)2.1 It was suggested that product formation proceeded via nucleophilic substitution of the Cl" ligands to form initially species Y, Pd(SH)2(dpm), which rapidly underwent self-association with elimination of H 2S, reaction 5.3.4.1 Species Y was not detected, and formation of Pd2(SH)2(u.-S)(dpm)2 presumably results from the tendency of dpm to bridge;4 there is ample precedence for similar dinucleation processes forming bridge sulfide complexes from mononuclear species containing terminal SH" ligands.5 Complexes of the type M(SH)2(dpe) where M = Ni, Pd, and dpe = bis(diphenylphosphino)ethane, as well as the complex Pt(SH)2(PPli3)2, are stable with respect to dinucleation.6 In the present study, Pd2(SH)2(p-S)(dpm)2 was not observed and, if Y is 221 References on page 234 Chapter 5 Pd(SH)2(dpm), reaction 5.3.4 does not occur in DMSO; this is a more polar and more coordinating solvent than CH2CI2 or CeE!*,7'8 and would appear to prevent the dinucleation. DMSO-dg (5.3.2) PdX2(dpm)(9) + 0.5 NaSH • Pd2X2(p-S)(dpm)2 (2) + Y (not balanced) E t 3 N (5.3.3) 2 PdCl2(dpm) (9a) + 3 H 2S > Pd2(SH)2(n-S)(dpm)2 + 4 HCl(as Et3NH+Cl") (5.3.4) 2 Pd(SH)2(dpm) (Y) • Pd2(SH)2(p-S)(dpm)2 + H 2S Formation of Pd2X2(p,-S)(dpm)2 (2) in the NaSH reaction is explained by invoking a similar mechanism. Nucleophilic substitution of one X" ligand would give PdX(SH)(dpm) which could then self-associate or react with 9 with elimination of H2S or HX (Scheme). [Note: it is not clear how Pd2l2(dpm)2 (lc) was also formed in some studies (see Table 5.2), but a single experiment demonstrated that a DMS0-d6 solution of 2c kept at 50 °C for 1 h generates lc and several as yet unidentifiable species.] Ph,P X <Q Pd + SH-Ph 2P X Ph 2P ,SH / \ / < Pd + X-Ph 2P X Scheme Ph 2P / S H / Ph 2P x < X ° r < X Ph 2P X \ Ph 2P X Ph 2P PPh 2 I . a . I : — " p , d - - x + H2S (orHX) P h 2 P . -PPhj It is difficult to conceive how Y (Pd(SH)2(dpm)) could be involved in the formation of 2, and Y is perhaps just a side-product that forms in higher yields under higher local solution 222 References on page 234 Chapter 5 concentrations of H 2S (NMR- vs. Schlenk-techniques). In addition, the formation of Y is probably also governed by the nucleophilicity of SH" versus T, Br", and Cl" with the trend being SH">r>Br" >C1";2 in the chloride and bromide systems, more Y was formed despite the use of Schlenk-techniques. Although PdX(SH)(dpm) was not detected, indirect support for its existence comes from studies of the reaction of 9 with 0.5 mole equivalent NaSH that generates Pd2X2(u.-S)(dpm)2 (2) (reaction 5.3.2); a mechanistic scheme similar to that above is envisaged with the coupling of PdX(SH)(dpm) proceeding via species such as I or II: Ph 2P ^ P P h 2 Ph 2P ~~PPh 2 J. SH J .—-SH „L—-X I SH Pd , , Pd . P d — Pd X I X I X I X I P h 2 P ^ -PPh 2 P h 2 P - ^ ^ ^ P P h 2 i n H 2 S or HX is then pictured to form via a deprotonation/protonation process,9 where deprotonation of an SH" ligand promotes sulfide bridge formation. The expected H 2S could not be detected within reaction 5.3.2 by smell or by NMR spectra as the reaction was <2% complete after 1 d, probably because of limited solubility of NaSH in DMSO; complex 2 could be seen only in the 31P{'H} NMR spectra, the *H NMR spectra showing only unreacted 9. H 2S, however, was detected both in NMR spectra and by its odor in closely related studies. For example, reaction of Pd2(SH)2(p-S)(dpm)2 with 1 mole equivalent I2 was studied in CDCI3 and in DMSO-d6: (5.3.5) Pd2(SH)2(p-S)(dpm)2 + I2 Pd2I2(p-S)(dpm)2 (2c) + H 2S + sulfur Oxidative addition of I2 occurs, and in low temperature NMR studies in CDCI3, an intermediate was detected, characterized by a 1:4:6:4:1 quintet in the *H NMR spectrum and a singlet in the 31P{1H} NMR spectrum. Similar spectral characteristics were seen for the intermediate 223 References on page 234 Chapter 5 Pd2l4(dpm)2 in reactions of 1 or 2 with I2 (see Chapter 4), and it is therefore reasonable to attribute the intermediate in reaction 5.3.5 to a species such as I; however, a 8(Pd-SH) signal, typically at 5 between 1.5 and -1.5,1 was not observed. Of note, a 5(Pd-SH) signal was also not seen in the'H NMR spectrum of the structurally similar intermediate Pd2X2(H)(SH)(dpm)2 (III) formed during reaction of 1 with H 2S at low temperature,10 and here it was suggested that either the 5(Pd-SH) signal was buried under the H 2S signal, or the acidic SH proton was undergoing rapid exchange withH2S protons.10'11 Ph 2 P- -PPh, _ P d - " " P ' d ^ 8 " I X I P h 2 P ^ ^ - P P h 2 m Reaction 5.3.5 is visualized to proceed in the same way that 2 reacts with X 2 (see Chapter 4), with the sulfur atoms produced during transannular oxidative addition undergoing concatenation to form elemental sulfur; elimination of H 2S, via a deprotonation/protonation involving the SH ligands, leads to formation of 2. Interestingly, the expected H 2S signal at 5 1.99 in the lH NMR spectrum of the completed reaction in DMSO-d6 was not observed, and the H 2 S smell was not evident; instead, a broad singlet was seen at 8 5.68 and the H 2 0 singlet signal was absent. The broad singlet must be due to H 30 +SH". Important implications of reaction 5.3.1 are seen with the iodide system; photodecomposition of H I 1 2 (bond strength is -295 kJ mol"1; AGf°(HI(g)) = 1.3 kJ mol'1, AGf°(HI(aq.)) = -55.9 kJ mol"1), for example, produces H 2 and I2, and the latter can react with 2c 224 References on page 234 Chapter 5 to give elemental sulfur, and complete the catalytic cycle with re-formation of 9c (reaction 5.3.6, see section 5.1). hv (5.3.6) H 2S • H 2 + sulfur 9c A few experiments were carried out in attempts to substantiate the catalysis implied in eq. 5.3.6, and Table 5.2 summarizes the results of photodecomposition studies. After the introduction of H 2S to DMSO-d6 solutions of 9a, 9b, or 9c (which generated 2, Y and HX), the samples were placed in the presence of laboratory light or sunlight for up to 27 d. ! H and 31P{ !H} NMR spectroscopic analyses revealed, in general, the disappearance of 2, the disappearance or reduction of species Y, the disappearance of a species attributable aq. HX, an increase in 9, and the appearance of PdX2(dpm(S)) (11). The 11 complexes were previously seen in the study of the reaction of 2 with X 2 in CDCI3 (see Chapter 4), and their formation was attributed to side-reactions of 9 with an S„ species (n<8); reaction of 9 with Sg in CDCI3 occurs only under photolytic conditions. In DMSO, 9 does slowly react with Sg under purely thermal conditions, and the presence of light has little effect (Table 5.3); the observed reactivity trend is I>Br>Cl, which is the relative trans effect of the halide ligands. The disappearance of both 2 and the species attributable to aq. HX is consistent with photodecomposition of HX followed by reaction of 2 with the resulting X 2 . GC analyses for H 2 , however, remain inconclusive, a small positive peak in the chromatogram with a retention time corresponding to that of H 2 sometimes being seen. TLC analyses confirmed the presence of elemental sulfur. Reaction of 2c with I2 in DMSO-Oe shows that transannular oxidative addition of I2 occurs in this solvent to give after 1 d complete conversion to 9c and 11c (in a 1:1 ratio); interestingly, elemental sulfur does not form (in contrast to that observed in CHCI3 - see Chapter 4), and considering that this reaction proceeds via the Pd2l4(dpm)2 intermediate (as 225 References on page 234 Chapter 5 evidenced by an observed green-black species - see Chapter 4), formation of 11c occurs probably because of the S„ species (n<8) being more stabilized in DMSO (than in CHCI3) (see above). Reactions of 2 with HX in CDCI3 and in DMSO-d6 were also investigated (Table 5.4). In CDCI3, reaction of 2a or 2b with 2 mole equivalents of HCl or HBr, respectively, under purely thermal conditions produces H 2S and 9a or 9b, respectively: CDCI3 (5.3.7) Pd2X2(u-S)(dpm)2 + 2 HX • 2 PdX2(dpm) + H 2S X = Br ,Cl Subsequent exposure to light yields no further changes in either system. As HCl and HBr do not exhibit acidic properties in CDC13,8 reaction 5.3.7 is pictured as proceeding via a series of HX additions across the Pd-S bonds with perhaps intermediate formation of Pd2X3(SH)(dpm)2 (II) and then Pd2X4(dpm)2; 9a and 9b would be subsequently formed by decomposition of Pd2X4(dpm)2. Analogous reactions in DMSO-d6 of HX with the chloride and bromide systems also proceed giving a mixture consisting of 9, 11, unreacted 2, and non-Pd-containing species characterized by broad singlets in the *H NMR spectra. These results imply that reaction 5.3.1 is more accurately described as an equilibrium: (5.3.8) 2PdX2(dpm) + H 2S Pd2X2(p-S)(dpm)2 + 2 H X The observed ! H NMR broad singlets could be due to a mixture of aquated (or solvated) HX and H 2S. The presence of 11 is difficult to explain but it is noted that during *H and 31P{1H} NMR spectroscopic analyses, the samples were unavoidably exposed to laboratory light (the samples were submitted for analyses using the Bruker AC200E spectrometer that is equipped with an automatic sample changer). In separate photodecomposition studies, when aq. HBr in DMSO-d6 was exposed to laboratory light for 12 h, Br2 was formed (Table 5.5, Fig. 5.11). Thus, the formation of l ib in the bromide system could result from reaction of 2b with Br 2 (see Chapter 4); presumably, a similar reaction could take place in the chloride system to form 11a. Exposure of 226 References on page 234 Chapter 5 the type 2 complexes and HX to a Hg vapour lamp leads to the disappearance of 2a, 2b, and the species attributed to a mixture of HX(aq.) and H2S(aq.) (replaced by new broad singlet - see Table 5.4); 9a, 9b and the corresponding 11a and l ib complexes were still observed (Table 5.4). GC did not reveal the presence of H 2 but elemental sulfur was seen by TLC. It is not clear why H 2 was not detected (by GC or NMR analyses) in these studies as well as in the separate photodecomposition studies of aq. HI and aq. HBr (Table 5.5), although in the latter studies sufficient amounts may not have been generated. Of note, in a single experiment where Pd2Cl2(dpm)2 (la) was reacted with H 2S in DMSO-d6 on a similar NMR reaction scale, H 2 was positively identified by both GC and NMR analyses. In the iodide systems, however, H 2 was detected (by GC) in the reaction of 2c with HI(g) (Table 5.4) but its presence is a result of decomposition of HI(g) during sample preparation because I2 was observed to form on the inside wall of the NMR tube, well above the DMSO-ck solution of 2c; 9c, 11c, sulfur, and dpm(0)2 were also observed, but of interest a 5(H2) signal in the *H NMR spectrum was not seen. The phosphine dioxide dpm(0)2 is probably a consequence of excess I2 (as well as trace 02) in the system, as a single experiment in air in which 9c was placed in the presence of 5 mole equivalents of I2 revealed immediate decomposition to Pdl2(?) and dpm(0)2; there is no reaction of 11c with 5 mole equivalents of I2 in air. Assuming that dpm(0)2 results solely from Pdl2(dpm), the original yield of 9c would be 59% with respect to 11c (Table 5.4). Reaction of 2c with aq. HI gave similar results but no H 2 was detected. In terms of possible mechanisms, reaction of 2 with HX in DMSO-Oe could proceed as in reaction 5.3.7 in CDCI3 but, in view of the acidic nature of HX in dimethylsulfoxide, via a series of nucleophilic additions of X" to give again intermediates such as Pd2X3(SH)(dpm)2 (II) and Pd2X4(dpm)2, with the latter subsequently decomposing to 9. However, having made this suggestion, it is not clear why reaction of 2b with HCl immediately 227 References on page 234 Chapter 5 forms exclusively a species thought to be Pd2(Br)(Cl)(p-S)(dpm)2 which slowly reacts further to give only trace amounts of Pd(Br)(Cl)(dpm), 9a(?), 9b(?), Pd(Br)(Cl)(dpm(S)) (cis or trans), lla(?), and llb(?). The presence and identities of the mononuclear species have not been unequivocally established but 31P{'H} NMR signals can be seen in regions indicative of chelating dpm and dpm(S) Pd complexes (Fig. 5.6). Clearly, further studies are warranted, and if the existence of H 2 can be established, then reaction 5.3.6 could be the first homogeneously catalyzed decomposition of H 2S in solution to H 2 and elemental sulfur. For a workable system, however, means must be determined to limit or prevent the formation of species Y and by-product PdX2(dpm(S)). 228 References on page 234 Chapter 5 5.4 Experimental Section The materials used, synthetic procedures for the ligands and complexes, and instrumentation used for 3 1 P and *H NMR spectroscopy, UV/vis spectrophotometry, and gas chromatography were described in Chapter 2. The complexes Pd2(SH)2(u-S)(dpm)2, 2a - 2c, and 9a - 9c were synthesized using published methods as outlined in Chapter 2. All experiments were performed under N 2 unless otherwise specified. 5.4.1 Preparation of gaseous HI Hydrogen iodide gas was prepared from hydriodic acid (57% by weight, 7.58 M) using a procedure modified from that reported.12a In a typical setup, P2Os (~5 mg) and HI (aq.) (10 uL, 0.078 mmol) were separately placed in an NMR tube fitted with a PTFE J. Young valve, and the sample was then evacuated and left in the dark for ~3 d. Thereafter, the NMR tube was immersed in an ice-bath, and the HI gas (m.p. -50.8 °C, b.p. -35.4 °C) was vacuum transferred using liquid N 2 for use. Note: due to the corrosive nature of aq. HI, 10 uL were used as this quantity was convenient to handle by a 250 uL syringe; use of a 10 uL syringe to transfer smaller amounts resulted in immediate blockage in the needle. 5.4.2 Reaction of PdX2(dpm) (9) with H2S Reaction of PdX2(dpm) (9) with 0.5, 0.75,1.0, or 1.5 mole equivalent H 2S was studied at R T . on an NMR-scale in CDCI3 and DMSO-d6. To a rubber septum-sealed NMR or Schlenk 229 References on page 234 Chapter 5 (165.0 mL volume) tube, or an NMR tube fitted with a PTFE J. Young valve, was placed a 0.50 mL solution of 9 (9a, 5 mg, 0.0089 mmol; 9b, 5 mg, 0.0076 mmol; 9c, 5 mg, 0.0067 mmol). H 2 S (9a: 108 uL at STP, 0.0045 mmol; 9b: 92 uL at STP, 0.0038 mmol; 9c: 81 uL (0.0034 mmol), 122 uL (0.0051 mmol), 162 uL (0.0068 mmol), and 243 uL (0.0102 mmol), all at STP) was slowly administered either by injection via syringe (in increments of 50 uL/min when using Schlenk-techniques) or by vacuum transfer. The sample was briefly shaken or continuously stirred, and was permitted to react in the absence of light for 1, 2, or 24 h prior to ! H and 31P{1H} NMR spectroscopic analyses. The sample was then placed under laboratory light or sunlight for periods of 1, 2, 5, 8, or 27 d; thereafter, the contents were analyzed using NMR and UV/vis spectroscopy and thin-layered (CHCb) and gas chromatography. Titration studies were carried out for the chloride system. In a Schlenk tube (165.0 mL volume) fitted with a rubber septum was placed 9a (25 mg, 0.045 mmol) dissolved in DMSO (4.5 mL). The solution was rapidly stirred, and H 2S (545 uL at STP, 0.022 mmol) was injected in increments of 100 uL/min. The mixture was continuously stirred for 2 h during which the initially yellow solution gradually turned red-brown. H 2 0 (15 mL) was then added, and a fine orange precipitate formed. This mixture was then de-gassed briefly to remove H 2S and washed with CH 2C1 2 (2x10 mL) before being filtered through a column of Celite 545 (2 cm x 10 cm). The column was subsequently washed with H 2 0 (2x5 mL), and the washings were combined with the colourless, clear filtrate, which was then titrated with a standardized solution of NaOH (0.010 M) using phenolphthalein as indicator. For a blank titration, the above procedure was repeated but with no H 2S added. 230 References on page 234 Chapter 5 5.4.3 Reaction of PdX2(dpm) (9) with elemental sulfur Reaction of PdX2(dpm) (9) with 1 mole equivalent Sg was studied at R.T. and at 50 °C on the NMR-scale in DMSO-de. PdX2(dpm) (9a, 10 mg, 0.0178 mmol; 9b, 10 mg, 0.0152 mmol; 9c, 10 mg, 0.0134 mmol) and S8 (9a: 0.57 mg, 0.0178 mmol; 9b: 0.49 mg, 0.0152 mmol; 9c, 0.43 mg, 0.0134 mmol) were placed in an NMR tube fitted with a rubber septum, and 0.50 mL DMSO-de was introduced. The sample was briefly shaken to ensure complete mixing, and was then placed in the dark or under laboratory light or sunlight for periods of 1, 2, 5, 9, 10, or 29 d prior to NMR analyses. 5.4.4 Reaction of Pd2I2(u-S)(dpm)2 (2c) with I2 Reaction of Pd2I2(p-S)(dpm)2 (2c) with 1 mole equivalent I2 was studied on an NMR-scale at R T . in DMSO-de. To a 0.40 mL solution of 2c (10 mg, 0.0078 mmol) in a rubber septum-sealed NMR tube was injected a 0.10 mL solution of I2 (2.0 mg, 0.0078 mmol). The sample was analyzed after 24 h using NMR spectroscopy and thin-layered chromatography (CHCb eluent). 5.4.5 Reaction of Pd2X2(p-S)(dpm)2 (2) with HX Reaction of Pd2X2(p-S)(dpm)2 (2) with HX was studied on an NMR-scale at R T . in CDCI3 and DMSO-de. A 0.50 mL solution of 2 (2a, 10 mg, 0.0092 mmol; 2b, 10 mg, 0.0085 mmol; 2c, 10 mg, 0.0078 mmol) was placed in a NMR tube fitted with a rubber septum or a PTFE J. Young valve, and HX (HCl(g), 446 uL at STP, 0.0184 mmol; HBr(g) at STP, 412 uL, 0.0170 231 References on page 234 Chapter 5 mmol; Ffl(aq. or g), 0.078 mmol) was injected or vacuum transferred in. The sample was then analyzed "immediately" and after 24 h using NMR spectroscopy. Subsequently, the sample was placed under laboratory light, sunlight, or light from an 18 W TLC lamp for 1 or 5 d. Thereafter, the sample was again analyzed using NMR spectroscopy and thin-layered (CHCI3 eluent) and gas chromatography. Reaction of 2b with HCl was similarly studied. To a 0.50 mL DMSO-d6 solution of 2b (10 mg, 0.0085 mmol) placed in a rubber septum-sealed NMR tube was added HCl(g) (412 uL, 0.0170 mmol) via syringe. The remainder of the procedure is as described above. 5.4.6 Reaction of Pd2(SH)2(u-S)(dpm)2 with I2 Reaction of Pd2(SH)2(u,-S)(dpm)2 with 1 mole equivalent I2 was studied on an NMR- scale at R.T. in CDC13 and DMSO-de and at low temperatures in CDC13. At R.T., a 0.40 mL solution of Pd2(SH)2(p-S)(dpm)2 (10 mg, 0.0093 mmol) was placed in an NMR tube fitted with a rubber septum, and a 0.10 mL solution of I2 (2.4 mg, 0.0093 mmol) was injected. The sample was analyzed after 24 h using NMR spectroscopy. To study the reaction at low temperatures, the sample was similarly prepared but at -60 °C (liq. N2/chloroform slush bath).13 NMR analyses were then carried out "immediately" at -60 °C; thereafter, the temperature was raised (-45, 10, and 20 °C) and NMR spectra were recorded at each new temperature. Corresponding TLC studies (CHCI3 eluent) were subsequently performed for each sample. 232 References on page 234 Chapter 5 5.4.7 Reaction of PdX2(dpm) (9) with NaSH Reaction of PdX2(dpm) (9) with 0.5 NaSH was studied on an NMR-scale at R T . in DMSO-de. PdX2(dpm) (9a, 10 mg, 0.0178 mmol; 9b, 10 mg, 0.0152 mmol; 9c, 10 mg, 0.0134 mmol) was placed in an NMR tube fitted with a rubber septum, and a 0.50 mL suspension of NaSH (9a: 0.50 mg, 0.0089 mmol; 9b: 0.43 mg, 0.0076 mmol; 9c: 0.38 mg, 0.0067 mmol) was injected. The samples were analyzed after 24 h using NMR spectroscopy. 5.4.8 Photodecomposition studies of HI, HBr, and H 2S HL HBr, and H 2S were irradiated using laboratory light, sunlight, or intense light from a 450 W medium pressure Hg vapour lamp in order to determine the extent of photodecomposition to H 2 and I2, Br2, or elemental sulfur, respectively. Specifically, aq. HI (10 uL, 0.078 mmol) or HBr (10 uL, 0.088 mmol) was dissolved in 0.50 mL CDC13 or DMSO-de held in a rubber septum-sealed NMR tube or an NMR tube fitted with a PTFE J. Young valve, and the sample was placed in the appropriate light source for periods of 0.5,1, 2, or 5 d. Thereafter, the liquid and head space of the sample were analyzed for H 2 using gas chromatography. Similar studies were also performed for H 2 S (1.0 mL at STP, 0.041 mmol) and aqueous and gaseous HI (prepared from 10 uL HI (aq.), see above) without prior dissolution into a solvent. UV/vis spectroscopy was also used to analyze all samples for I2 and elemental sulfur. For the liquid samples, analyses were carried out in CHCI3 using ~5 uL (droplet transferred using a glass pasteur pipette) of the liquid. For the gaseous samples, the NMR tube was briefly evacuated before CHCI3 was used to dissolve any deposit (I2, Sg) found on the inside walls of the tube. 233 References on page 234 Chapter 5 5.5 References for Chapter 5 1. Besenyei, G.; Lee, C-L.; Gulinski, J.; Rettig, S.J.; James, B.R.; Nelson, D.A.; Lilga, M.A. Inorg. Chem. 1987, 26, 3622. 2. Carey, F.A.; Sundberg, R.J. "Advanced Organic Chemistry, 3rd Ed. Part A: Structure and Mechanisms", Plenum Press, New York, 1993, ch. 5, p. 284. 3. Support for the dissociation of H2S in DMSO to give solvated FC and SH" ionic species comes from the similarities in solvent properties DMSO displays compared with other polar, aprotic solvents. See also: a) Plummer, M.A. Hydrocarbon Proc. 1987, 38. b) Collman, J.P.; Finke, R.G.; Cawse, J.N.; Brauman, J.I. J. Am. Chem. Soc. 1977, 99, 2515. 4. Puddephatt, R.J. Chem. Soc. Rev. 1983, 99. 5. Shaver, A.; Lai, R.D.; Bird, P.; Wickramasinghe, W. Can. J. Chem. 1985, 63, 2555, and references therein. 6. Schmidt, M. ; Hoffmann, G.G.; Holler, R. Inorg. Chim. Acta 1979, 32, L19. 7. See for example: Carey, F.A.; Sundberg, R.J. "Advanced Organic Chemistry, 3rd Ed. Part A: Structure and Mechanisms", Plenum Press, New York, 1993, ch. 4, p. 232. 8. Huheey, J.E. "Inorganic Chemistry, 3rd Ed. Principles of Structure and Reactivity", Harper and Row, New York, 1983, ch. 8, p. 328. 9. a) Jessop, P.G.; Lee, C-L.; Rastar, G.; James, B.R.; Lock, C.J.L.; Faggiani, R. Inorg. Chem. 1992, 31, 4601. b) Lee, C-L.; Chisholm, J.; James, B.R.; Nelson, D.A.; Lilga, M.A. Inorg. Chim. Acta 1986, 121, hi. c) Jessop, P.G.; Rettig, S.J.; Lee, C-L.; James, B.R. Inorg. Chem. 1991, 30, 4617. 10. Barnabas, A.F.; Sallin, D.; James, B.R. Can. J. Chem. 1989, 67, 2009. 11. Jennings, M.C.; Payne, N.C.; Puddephatt, R.J. J. Chem. Soc, Chem. Commun. 1986, 1809. 12. Although much has been published on the photodecomposition of HI, HBr, HCl, and H2S in the gas phase, similar studies in solution appear to be lacking. See for example: a) Martin, R.M.; Willard, J.E. J. Chem. Phys. 1964, 40, 2999. b) Huebert, B.J.; Martin, R.M. J. Phys. Chem. 1968, 72, 3046. 234 Chapter S c) Cadman, P.; Polanyi, J.C. J. Phys. Chem. 1968, 72, 3715. d) Compton, L.E.; Martin, R.M. J. Phys. Chem. 1969, 73, 3474. e) Oldershaw, G.A.; Porter, D.A.; Smith, A. J. Chem. Soc. Faraday J 1972, 68, 2218. f) Clear, R.D.; Riley, S.J.; Wilson, K.R J. Chem. Phys. 1975, 63, 1340. g) Calvert, J.G.; Pitts, Jr., J.N. "Photochemistry", John Wiley & Sons, London, 1967, ch. 3, p. 196. Gordon, A.J.; Ford, R.A. "The Chemist's Companion", John Wiley & Sons, Toronto, 1972, p. 451. 235 CHAPTER 6 Reaction of PdX2(dpm) with H 2S in the Presence of Alumina; Catalyzed Formation ofPd2X2(n-S)(dpm)2 236 Chapter 6 6.1 Introduction The discovery of the homogeneous conversion of H2S to H2 in solution using complexes of the type Pd2X2(dpm)2 (1) (reaction 6.1.1)1 led to some earlier attempts to heterogenize the reaction. Syntheses of Pd2X2(dpm)(dpmMe) complexes where one of the bridging dpm ligands of 1 is replaced with dpmMe (1,1 '-bis(diphenylphosphino)ethane), for example, were carried out with the aim of immobilizing the Pd complex via the methyl group to a solid support such as polystyrene.2 With the fading that Pd2X2(dpm)(dpmMe) can also react with H 2S to form quantitatively H 2 and the bridged-sulfide complex Pd2X2(|x-S)(dpm)(dpmMe) 2 and that H 2S can be catalytically converted to H2 (see Section 1.3, p. 24),3 the prospect of heterogenization for the ease of product separation seemed even more favourable. (6.1.1) H 2S + Pd2X2(dpm)2 (1) • H 2 + Pd2X2(p-S)(dpm)2 (2) Kinetic and mechanistic studies on the S-abstraction process 2 - » 1 using dpm (see Chapter 3), however, revealed that immobilization may not be appropriate or feasible because the bridging phosphine ligands of 2 are also involved in the S-abstraction; for example, reaction of 2 with added dpm-d2 (or dpmMe) gives dpm(S) (or dpmMe(S)) products with distributions that are close to statistical regarding the availability of 3 chelating (P-P) moieties (see Section 3.3).3 Work on heterogenization was continued during the course of this thesis, and reaction 6.1.2 was discovered, occurring in the presence of alumina. alumina (6.1.2) 2 PdX2(dpm) (9) + H 2S • Pd2X2(p-S)(dpm)2 (2) + 2 HX Reaction of 9 with H 2S was described in Chapter 5 and can take place in dimethylsulfoxide without the need for alumina; however, the formation of side-products, especially Pd(SH)2(dpm), was sometimes seen. Reaction 6.1.2 was studied in CHCI3, and the presence of alumina was 237 References on page 274 Chapter 6 shown to be necessary. The reaction appears to proceed cleanly, in contrast to that when DMSO solvent was utilized. Important implications are again seen with the X = I system; photodecomposition of HI, for instance, produces H2 and the I2, and the latter can complete the catalytic cycle via reaction 6.1.3 (see Chapter 4).3 (6.1.3) Pd2X2(p-S)(dpm)2 + X 2 • 2 PdX2(dpm) + sulfur This chapter summarizes preliminary work on reaction 6.1.2; results of surface studies on the alumina used are also discussed. 238 References on page 274 Chapter 6 6.2 Results Reaction 6.1.2 was studied at R.T. in CDCI3 on the NMR-scale, and NMR and X-ray photoelectron spectroscopies were used to analyze the solution and the alumina surface, respectively. Various experiments were carried out for the chloride system as outlined in the Experimental Section; Table 6.1 summarizes the content make-up for each type of experiment (hereafter also referred to as "sample"). The alumina used is of the y-type, commonly used as chromatographic absorbents (powder (av. mesh size 200) or TLC plate forms (mesh size 60)). Samples 1-5 were prepared for NMR spectroscopic analyses only. For sample 1, there were no detectable changes in the colour of the solvent (colourless) or the alumina (white) after 24 h; the *H NMR spectrum recorded 24 h after sample preparation reveals in addition to the sharp singlets of residual CHCI3 and H2O at 8 7.24 and 1.54, respectively, a broad unresolved signal (possibly a doublet or triplet) at 5 1.2 (Fig. 6.1). These observations were also made for sample 2, the FL.S singlet at 8 0.82 (observed under the same conditions in the absence of alumina) not being seen. For samples 3 and 4 (prior to addition of H2S), addition of 9a yielded immediately a yellow solution, while the white alumina gradually turned orange after ~1 h. When a stoichiometric amount of H2S (see eq. 6.1.2, H2S : 9 = 0.5) was introduced for sample 4, the alumina immediately turned orange-brown, and the solution at the same time changed from yellow to brown-yellow. ! H and 31P{1H} NMR spectroscopic analyses of sample 3 revealed the presence of 9a, with no other Pd species seen (Fig. 6.2a); for sample 4, trace amounts of Pd2Cl2(p-S)(dpm)2 (2a) were observed (Fig. 6.2b). The order of adding 9a and H2S was reversed for sample 5 compared to that for sample 4; addition of a stoichiometric amount of H 2S first again produced no detectable changes in the system. Subsequent addition of 9a (~1 h after the addition of H2S) yielded an immediate colour 239 References on page 274 Chapter 6 Table 6.1. Summary of the content make-up in Expts. 1 -18; 9a = PdCkCdpm). Sample Contents 1 CDC13 (0.7 mL), y-alumina powder (15 mg) 2 CDCI3 (0.7 mL), y-alumina powder (15 mg), H2S (110 \xL at STP, 0.0045 mmol) 3 CDCI3 (0.7 mL), y-alumina powder (15 mg), 9a " 4 CDCU (0.7 mL), y-alumina powder (15 mg), 9a, H2S (110 yL at STP, 0.0045 mmol) 5 CDCI3 (0.7 mL), Y-alumina powder (15 mg), H2S (110 uL at STP, 0.0045 mmol), 9a 6 CDCI3 (0.7 mL), y-alumina powder (15 mg, pre-dried at 150 °C for 24 h), H2S (110 ^L at STP, 0.0045 mmol), 9a 7 CDCI3 (1.0 mL), Y-alumina plate,* H2S (0.5 mL at STP, 0.021 mmol), 9a 8 CDCI3 (1.0 mL), Y-alumina powder (11.5 mg) from 1.0 cm x 1.0 cm plate, 9a, H2S (0.5 mL at STP, 0.021 mmol) 9 CDCI3 (1.0 mL), Y-alumina powder (15 mg, finely ground), 9a, H2S (0.5 mL at STP, 0.021 mmol) 10 CDCI3 (1.0 mL), a-alumina powder (corundum, 15 mg), 9a, H2S (0.5 mL at STP, 0.021 mmol) 11 10 mL CDCI3 solution of 9a (10 mg, 0.018 mmol), 10 mL MeOH solution of NaSH (1.0 mg, 0.018 mmol) 240 References on page 274 Chapter 6 Table 6.1. (cont.) 12 CDC13 (1.0 mL), y-alumina plate 13 CDCI3 (1.0 mL), y-alumina plate, 9a 14 CDCI3 (1.0 mL), y-alumina plate, H2S (0.5 mL at STP, 0.021 mmol) 15 CDCI3 (1.0 mL), y-alumina plate, 9a, H2S (0.5 mL at STP, 0.021 mmol) 16 CDCI3 (1.0 mL), y-alumina plate, H2S (0.5 mL at STP, 0.021 mmol), 9a 17 CDCI3 (1.0 mL), y-alumina plate, S8 (12 mg, 0.047 mmol) 18 CDCI3 (1.0 mL), y-alumina plate, HCl (0.5 mL at STP, 0.021 mmol) 0 5 mg of 9a (0.0089 mmol) was always used. * The y-alumina plates were all 1.0 cm x 1.0 cm. 241 References on page 274 Chapter 6 Fig. 6.1. Representative *H NMR spectrum (300 MHz) of samples 1 or 2 containing y-alumina powder (15 mg) in CDCI3 with and without the presence of H 2S (110 uL at STP), respectively. CHCI3 j H 2 0 I I I I I I I I I I I I I I I I I I I I I II I I I I I I I I I I I I I I I M I I I I I I I I M I I 1 m I 1 1 1 1 10.0 8.0 6.0 4.0 2.0 0.0 PPM 242 References on page 274 Chapter 6 Fig. 6.2. *H NMR spectra (300 MHz) showing the extent of the reaction at R T . of PdCi2(dpm) (9a) with H2S in CDCI3 in the presence of y-alumina under various experimental conditions; 2a = Pd2Cl2(u-S)(dpm)2. T r 5.0 4.0 P P M 243 Sample 3 Sample 4 Sample 5 Sample 6 Samples 8, 9 Samples 15, 16 3.0 References on page 274 Chapter 6 change of both the white alumina and the colourless solution to orange-brown; qualitatively, there were no further visible changes over time. As observed for sample 4, l H and 31P{1H} NMR analyses for sample 5 also revealed the formation of 2a but in more significant amounts, -10% yield relative to 9a (determined from the relative integrated areas of the respective C H 2 proton signals) (Fig. 6.2c). Sample 6 was prepared in the same way as sample 5 except that the dumina was pre-dried for 24 h at 150 instead of 75 °C. NMR analyses again revealed the formation of 2a but in 60% yield (Fig. 6.2d). In these experiments (4 - 6), H 2S was again not seen in the *H NMR spectra. Experiment 7 was performed to determine if removal of H 2S pre-treated alumina from an H 2S (2.5 mole excess) solution, with subsequent drying of the alumina (under an Ar atmosphere) and its placement in a CDCU solution containing only 9a, could affect the dinucleation reaction. As with sample 3, the alumina slowly turned orange, and NMR analysis of the solution revealed only the starting material 9a. It should be noted that in samples 3 - 7, undissolved 9a (0.5-1 mg) was seen. Samples 8-10 were prepared using alurnina powder scraped from a TLC plate, finely ground y-alumina powder, and a-alurnina, respectively; H 2S (2.5 mole excess) was then introduced into the system immediately after the mixture of 9a and alumina in CDCI3 was prepared. In experiments 8 and 9, the system immediately turned orange from an initial yellow, and NMR analyses after 4 h revealed that 9a has completely reacted with 2a as the only product detected (Fig. 6.2e). For experiment 10, however, the alurnina turned yellow but NMR analyses after 4 h revealed 9a and trace amounts of 2a (Fig. 6.3); after 24 h, a 10% yield of 2a was seen and, after 72 h, 50%. Experiment 11 was conducted to determine if 2a could be formed from reaction of 9a with 1 mole equivalent NaSH under conditions in which NaSH is slowly added to a solution of 9a such that the local concentration of NaSH is kept low. During this experiment, the odor of H 2S was detected; NMR analyses again revealed that 9a has completely reacted to form 244 References on page 274 Chapter 6 Fig. 6.3. ' i i NMR spectra (300 MHz) showing the progress of the reaction at R.T. of PdCl2(dpm) (9a, 8.9 x 10"3 M) with 2.5 mole equivalent H 2S in CDC13 in the presence of a-alurnina (15 mg); 2a = Pd2Cl2(p-S)(dpm)2. 9a P P M 245 References on page 274 Chapter 6 2a Samples 12-18 were prepared mainly for surface studies; NMR spectroscopic analyses were carried out for samples 15 and 16. Samples 12-14 were control samples in which alumina plates (1.0 cm x 1.0 cm) were placed in CDCU in the absence or presence of 9a or H2S (5 mole excess) and left at R T . for ~4 h before analysis. For samples 12 and 14, there were no visible changes in the systems during this time period; the colour of the alumina plates remained white and the solvent colourless. For sample 13, however, the presence of 9a yielded a yellow solution, and the initially white alumina plate slowly became orange over ~1 h; the presence of undissolved 9a (-0.5 mg) was also observed. The plates were removed ~4 h after sample preparation and were analyzed using XPS with results shown in Figs. 6.4 to 6.7. For sample 12, the XPS spectrum (Fig. 6.4a) is characteristic of dumina4 (Fig. 6.4b) and shows peaks arising from the Al 2p, Al 2s, and O Is photoelectrons with binding energies of 74.5, 119.8, and 531.8 eV, respectively; the XPS spectrum of the alumina powder used in samples 1 - 5 is identical to that shown in Fig. 6.4a. Also seen are the signals arising from the C Is and N Is photoelectrons with corresponding binding energies of 285.0 and 399.3 eV, respectively, as well as signals from the Auger electrons of O and C with kinetic energies of 507.5 and 260.7 eV, respectively. These spectral characteristics were also observed in all other samples (13 -18) studied. For sample 13, the XPS spectrum (Fig. 6.5) shows additional signals arising from the CI 2p, Pd 3ds/2, and Pd 3d3/2 photoelectrons with corresponding binding energies of 198.2, 337.0, and 342.5, respectively; the Pd 3ds/2, Pd 3d3/2, and CI 2p photoelectron signals are shown in more detail in Figs. 6.6 and 6.7. For sample 14, a S 2p photoelectron signal is observed with a binding energy of 169.1 eV (Fig. 6.7). Samples 15 and 16, as for samples 4 and 5, differ in the order of addition of 9a and H2S (2.5 mole excess). For sample 15, introduction of 9a yielded a yellow solution and an initially white alumina plate which slowly changed to orange over a period of ~1 h; some undissolved 9a (-0.5 mg) was still seen. When 246 References on page 274 Chapter 6 Fig. 6.4. X-ray photoelectron spectra of y-alumina: a) experimental; b) literature, taken from ref. 4, pertaining to a surface cleaned by bombardment with Ar. 247 References on page 274 Chapter 6 Fig. 6.5. X-ray photoelectron spectrum (Mg Kct, RT.) of sample 13 consisting of PdCl2(dpm) (9a) adsorbed on y-alumina. vt fl Ols Pd3d Cls 1 \ — ~ M A12s C12p A J 2 P 1000 800 600 400 binding energy, eV 200 Fig. 6.6. X-ray photoelectron spectra (Mg Kct, R.T.) showing the Pd 3d photoelectron signals for samples 13, 15, and 16. Pd 3d s / 2 337.0 eV t (A s . i ^ sample 13 350 345 340 335 binding energy, eV 330 325 248 References on page 274 Chapter 6 Fig. 6.7. X-ray photoelectron spectra (Mg Ka, R.T.) showing the CI 2p and S 2p photoelectron signals of samples 13,14,15, and 16. S2p 161.9 eV 169.1 eV j a m sample 16 ^ ^ A sample 15 210 " 200 190 180 170 160 binding energy, eV Fig. 6.8. X-ray photoelectron spectra (Mg Ka, R.T.) showing the S 2p and CI 2p photoelectron signals of samples 17 and 18, respectively. jC C , 2 P 3 t ft 0) 199.5 eV i > i ^ S2p 169.1 eV 210 200 190 180 170 160 binding energy, eV 249 References on page 274 Chapter 6 H 2S was added next (-2 h after the introduction of 9a), the alurnina plate immediately turned orange-brown and the solution slowly changed to a final brown colour. The sample was permitted to react for 2 h and then analyzed; undissolved 9a was no longer seen. The ' H NMR spectrum reveals the formation of 2a in >90% yield relative to unreacted 9a (Fig. 6.2f); trace amounts of H 2 S at 8 0.82 were also detected. Surface studies on the alumina plate revealed some findings similar to those for samples 13 and 14. Signals arise, for example, from the photoelectrons of the Pd 3d, Cl 2p, and S 2p subshells (Figs. 6.6 and 6.7); however, anew S 2p signal is seen at 161.9 eV (Fig. 6.7). For sample 16, the order of introducing 9a and H 2S into the system was reversed (compared to that in sample 15), and addition of H 2S produced no visual changes in the system. When 9a was added ~2 h later, the colour of the alurnina plate immediately turned orange-brown; the solution had a yellow colour and undissolved 9a (-0.5 mg) was seen. The system was then reacted for 2 h and then analyzed; the solution during the reaction period changed colour to a final brown and undissolved 9a was no longer seen. NMR and surface analyses gave the same findings as observed for sample 15 (Fig. 6.2f). Samples 17 and 18 were prepared as control samples in which alumina was tested for absorption of Sg and HCl(g), respectively; yellow and white plates resulted, respectively. The XPS spectrum for sample 17 shows an S 2p signal at 169.1 eV and that of sample 18 a Cl 2p signal at 199.5 eV (Fig. 6.8). Low temperature NMR analysis of reaction 6.1.2 was performed when H 2S (~1 mole equivalent) was added to a CDCI3 mixture of 9a (1.8 x 10"2 M) and y-alumina (15 mg) held at -42 °C. The solution immediately changed colour from pale yellow to orange-yellow. Analysis at -50 °C 1 h after sample preparation revealed the presence of the starting material 9a and the product 2a in -50% yield; no other species were detected. Re-analysis of the sample after being placed at R.T. for 24 h revealed the presence of only 2a. 250 References on page 274 Chapter 6 To substantiate and quantify the HX produced in reaction 6.1.2, a series of synthetic-scale experiments were performed in which PdX2(dpm) (9a, 1.8 x 10"2 M ; 9b, 1.5 x 10"2 M ; 9c, 1.3 x 10"2M)andH2S (1.0 M) were reacted at R.T. in 2.5 mL CHCU containing 75 mg finely ground y-alumina. Addition of H 2S to a rapidly stirred CHCI3 mixture of 9 and alumina immediately (within seconds) caused a colour change from yellow (chloride or bromide system) or orange (iodide system) to a final brown. The mixture was reacted for ~4 h before the alumina was filtered off, washed, and dried. The filtrate and the washings were combined, and the solvents were removed by rotary evaporation with the resulting residue dissolved in CDC13 and analyzed using NMR spectroscopy. The lH and 31P{!H} NMR spectra showed the presence of only Pd2X2(p-S)(dpm)2 (2); the yields of this product (2a - c) were deterrnined to be -100%. The isolated alurnina was placed in H 2 0 and the slurry was titrated with a standardized solution of NaOH (0.010 M). Two blank titrations were also carried out using unused alumina and alumina isolated from a synthetic-scale experiment using 9a but no H 2 S; both titrations gave similar results (1.1 x 10"5 mol titrated acid). The amounts of titrated HX found were 5.2, 3.4, and 3.4 x 10"5 mol for the chloride, bromide, and iodide systems, respectively, all in good agreement with theoretical values within experimental error (4.5, 3.8, and 3.4 x 10"5 mol, respectively). The synthetic-scale experiment for the iodide system was repeated twice. In the first re-trial, the isolated dumina (containing HI) was placed in -1 mL CDCI3 (in an NMR tube fitted with a PTFE J. Young valve) with the resulting mixture placed in the presence of light (laboratory and/or sunlight) for 4 d to allow for possible photodecomposition of HI. During this time, the alumina gradually became yellow but the solvent remained colourless. The mixture was subsequently irradiated using a TLC Hg vapour lamp (18 W), and within 4 h the alumina acquired an orange colour and the solvent a purple tinge; further exposure caused no further changes. UV-vis spectroscopic analysis of the solvent revealed an 251 References on page 274 Chapter 6 absorption band at 512 nm mdicating the presence of I2 (~1 x 10'7 mol) (see Chapter 4, Fig. 4.4). GC and NMR analyses of the head space and solvent, respectively, revealed no detectable H2. The orange alumina was filtered out and washed with MeOH which led to elution of an orange substance; the alumina re-acquired a white colour. The orange methanolic eluate was analyzed using UV-vis spectroscopy, and the spectrum shows the presence of two absorption bands at 292 and 360 nm (Fig. 6.9a). These bands were identified as belonging to I3" (n(T) = 3 x n(I3") = 5.4 x 10"6 mol); the UV-vis spectrum of Nal 3 (formed by reaction of I2 with NaT) is shown in Fig. 6.9b. Of interest, the UV-vis spectrum of I2 in MeOH is shown in Fig. 6.10, and consists of three absorption bands at 292, 360, and 442 nm. The filtered alurnina was titrated with NaOH solution to determine the amount of remaining HI; n(HJ) = 2.84 x 10"5 mol. The total number of moles of T (or HI) is therefore 3.38 x 10"5 mol, and this result agrees excellently with the theoretical value of 3.36 x 10"5 mol. In the second re-trial, the isolated alumina was placed in CHCI3 and a solution of 12 in CHCI3 (2.7xlO" 3M) was added to determine whether I2 reacts with the chemisorbed T. During this addition, the alurnina changed from white to orange (1.1 mL was added, 2.9 x 10"6 mol I2); at this stage, the colour of the alurnina remained constant and further I2 addition only caused the solvent to acquire a purple colour (due to unreacted I2). The durnina was filtered off and washed with CHCI3 before MeOH was used to elute an orange substance which analyzed for I3". Photodecomposition of HI(57% aq.) was qualitatively studied at R T . in CDCI3 under laboratory light in the presence of a- or y-alumina powder; gas chromatography and NMR and UV-vis spectroscopies were used to determine the extent of H 2 and I2 production, respectively. Four samples were prepared and placed in the presence of light (laboratory and/or sunlight) for periods of time up to 4 d analysis. For the first sample, addition of HI(aq.) (~2 uL) to CDCI3 (0.5 mL) already containing y-alumina immediately resulted in a colour change on the alumina from 252 References on page 274 Chapter 6 Fig. 6.9. a) Electronic spectrum (250 to 650 nm) of I3" (MeOH) isolated from synthetic-scale studies, b) Electronic spectrum of an authentic sample of I3" in MeOH (prepared from I2 + NaT) (e292 = 2.89 x 104 M" 1 cm"1, £360 = 1.66 x 104 M 1 cm"1). Wavelength (nm) Wavelength (nm) 253 References on page 274 Chapter 6 Fig. 6.10. Electronic spectrum (250 to 700 nm) of I2 in MeOH (s292 = 2540 M 1 cm"1, e 3 6 0 = 1380 M" 1 cm"1, 8442 = 930 M 1 cm"1). 2.65 at u I 1.33 0.K O.OQ^  i i | i i i i | i i i i | i i i i | i 1 1 300 M 500 800 700 Wavelength (nm) Fig. 6.11. lH NMR spectrum (300 MHz) showing the results of the photodecomposition of HI(aq.) (~2 uL, 7.58 M) at R T . in CDC13 in the presence of a-alumina (15 mg); X = CI and/or I. CHCU CDHX 2 8.0 7.0 6.0 P P M 5.0 4.0 254 References on page 274 Chapter 6 white to light yellow; the solvent remained colourless. Within minutes, the alumina turned orange-yellow and the solvent became light purple. After 30 min, the durnina was orange and the solvent, a definite purple. The sample was analyzed 24 h after preparation at which time the alumina was still orange-yellow but the solvent now an intense purple; GC analysis revealed no detectable H 2 while UV-vis spectroscopic analysis revealed the presence of I2 as evidenced by an absorption band at 512 nm. For the second and third samples, HI(aq.) (~2 uL) was added to dry y - or a -alumina powder, respectively, and analyses for H 2 were made (immediately) before CDCb was added; the gas chromatograms again revealed no H 2 peaks. The dumina of both samples immediately turned yellow and remdned this colour when CDCI3 was added. Within minutes, the a-dumina of the third sample turned orange and the solvent purple. In the second sample, however, there were no observable changes until after ~2 h when the solvent acquired a light purple tinge and the y-dumina a yellow-orange. No more observable changes occurred for this sample, and andyses after 24 h reveded the presence of I2 and still the absence of detectable H 2 . For the third sample, the solvent had an intense purple colour after 24 h and the a-dumina an orange colour. The solvent continued to darken in colour and became black after 4 d. ! H NMR andysis reveded in addition to the CHCI3 residud solvent peak at 8 7.24 a broad singlet at 8 5.2 (Fig. 6.11); H 2 was not observed as there was no signd at 8 4.64. GC andysis dso reveded no detectable H 2 . The fourth sample was prepared using a-dumina as described for the third except that CDCI3 was added after 4 d; GC andysis before solvent addition again reveded no detectable H 2 . The resulting intense purple solution was andyzed using NMR and UV-vis spectroscopies, and an absorption band at 512 nm was seen which did not change over time; no H 2 signd was seen in the ! H NMR spectrum. The solvent remained purple and did not darken as observed for the third sample. The dumina, which acquired a purple-brown colour, was filtered off and washed 255 References on page 274 Chapter 6 with CHCI3, eluting the remaining physisorbed h.. MeOH was then used to wash the resulting orange-yellow alumina, and an orange-yellow eluate emerged which was analyzed by UV-vis spectroscopy (the alumina turned white). The UV-vis spectrum revealed two absorption bands at 292 and 360 nm indicating the presence of I3" (see above). Control samples consisting of I2 in CDCI3 over a- or y-alumina were prepared and were placed under laboratory light for 24 h. There were no observable changes in these systems; the aluminas remained white, and the I2 512 nm bands constant in intensity. Finally, in blank experiments where samples consisting of CDCI3 with and without Y-alumina (20 mg) were exposed to light (laboratory, sunlight, or light from a TLC Hg vapour lamp (18 W)) for periods of time up to 4 d, ! H NMR measurements revealed that no changes occurred in these systems (i.e. no H 2 was seen). 256 References on page 274 Chapter 6 6.3 Discussion Aluminas are commonly employed as chromatographic absorbents, catalysts, and catalyst supports; these uses as well as the methods of preparation and physical properties of these materials have been reviewed.5"8 Reaction 6.1.2 was discovered during the course of this thesis work, the dinucleation process being catalyzed by alumina, and preliminary studies were carried out. Reaction of PdX2(dpm) (9) with H2S can take place rapidly to form Pd2X2(p-S)(dpm)2 (2) without the need for alumina but a highly polar solvent such as DMSO must be used (see Chapter 5); furthermore, a side-product believed to be Pd(SH)2(dpm) also forms. In the present study, 9 does not react with H 2S in CHC13 in the absence of durriina (see Chapter 5). Previously in this laboratory, however, it was shown that in the presence of a base (e.g. EtsN), PdCl2(dpm) (9a) can react with H 2S but the dinuclear product Pd2(SH)2((x-S)(dpm)2 forms exclusively.9 This same product was also obtained (exclusively) if excess NaSH was used in place of H 2 S and EtjN. 9 The role of alumina in the present study was ascertained from solution NMR and X-ray photoelectron spectroscopic studies; y-alumina, commonly employed as laboratory chromatographic absorbents (powder or TLC plate form), was used. From NMR studies, reaction 6.1.2 appears to proceed rapidly and cleanly with 2 as the only Pd-containing product formed; Pd2(SH)2(p-S)(dpm)2 and side-products such as Pd(SH)2(dpm) were not observed. The extent of the reaction appears to depend on the surface area of the alumina, and as will be discussed later, the reaction goes to completion quickly (within minutes) when finely ground dumina is used. The free HX product was not detected in solution NMR studies, and indeed, in the free state, HX reacts rapidly with 2 to form 9 and H 2 S (see 257 References on page 274 Chapter 6 Chapter 5). The HX chemisorbs on the alurnina surface, and titration studies showed quantitative formation of this species (see later). Reaction of 9 with H 2S probably proceeds via initial activation of H 2S on the alumina surface. From a series of experiments on the chloride system in which emphasis was placed on the order of introducing reactants, when 9a was added to alurnina pre-treated with H 2S, an instant colour change took place in the system (dumina and the solution), and NMR analyses showed the formation of 2a. Treatment of alumina with a solution of 9a alone, in contrast, yielded a slow colour change on the dumina but there was no reaction (i.e. dinucleation to form, for example, Pd2Br2(dpm)2) as evidenced by NMR measurements. Subsequent addition of H 2 S afforded an immediate colour change in the system, and NMR andyses agdn reveded the formation of 2a. Worth noting is that dumina that has been pre-treated with H 2S and then removed and exposed to an inert atmosphere of Ar for a few minutes shows no reaction with 9a (sample 7). As H 2S is adsorbed by dumina (as shown by NMR measurements - sample 2), the interaction may be physicd in nature (i.e. physisorption) and that subjection to an Ar atmosphere dlows desorption to take place. Surface studies were initiated for the chloride system to andyze the dumina in an attempt to determine the nature of the interaction (physisorption vs. chemisorption) of the various species involved in the reaction of 9a with H 2S. The photoelectron spectrum of the dumina used (powder or TLC plate forms) (Fig. 6.4) compares well with that reported;4 the binding energies of the Al and 0 photoelectrons and the kinetic energies of the 0 Auger electrons are in excellent agreement with the literature vdues. Signds arising from the photoelectrons of C and N are dso seen; their presence is probably due to adsorption of C0 2 1 0 and N 2 from the dr - the C binding energy dso agrees with the literature vdue (see Fig. 6.4b). The photoelectron spectra of dumina with 258 References on page 274 Chapter 6 adsorbed 9a (Figs. 6.5 - 6.7) show in addition the photoelectron signals of Pd and CI and the Pd Auger electron signal; the Pd 3d3/2, Pd 3ds/2, and CI 2p photoelectron signals are shown in detail in Figs. 6.6 and 6.7, respectively. The Pd 3ds/2 energy of 337.0 eV compares to those of Pd(TJ)X2(phospliine)2 complexes (337 - 338 eV),1 1 while that of CI (198.2 eV) is in the metal chloride range (197 - 199 eV). 1 2 These results are consistent with physisorption of 9a. Although physisorption and facile desorption of H 2S were alluded to above, surface analysis of H2S-treated alumina (under ultra-high vacuum conditions) revealed the presence of a weak S 2p photoelectron signal at 169.1 eV which corresponds to S in an SO/'-like environment (Fig. 6.7)13 and which was identified in separate studies as deriving from elemental sulfur. A similar weak signal was also seen for dumina treated with elemental sulfur (Fig. 6.8) (note: the S 2p binding energy of free Sg is approx. 164 eV 13). Although the origin of the 169.1 eV signal cannot be easily explained (decomposition of EfeS at R.T. to give Sg?), it is noted that diirnina is used as a catdyst for the high temperature C»2-oxidation of H 2S to H 2 0 and Sg as in the Claus process14 or the high temperature (500 - 1000 °C) decomposition to H2 and S 8 1 5 (note: as evidenced from separate studies, the presence of Sg does not influence or affect reaction 6.1.2). As Sg can be eluted off dumina using an appropriate solvent (see Section 4.4, for example), the Sg interaction is one of physisorption. The weak S 2p photoelectron signd (B.E. = 169.1 eV) is dso observed in the spectra of dumina samples after reaction of 9a with H 2S (Fig. 6.7); however, there is an additiond S 2p photoelectron signd at 161.9 eV, which compares with that of S in an S2'-like environment (the S 2p binding energy of metd sulfides are in the range 160- 162 eV),1 3 and is a good indication of the presence of physisorbed 2a (note: this species can be completely eluted off dumina using MeOH (see Experimentd section). The presence of the HCl product in reaction 6.1.2 is not easily ascertained, for example, from the CI 2p photoelectron signd. As a signd at 199.5 eV can be seen for the 259 References on page 274 Chapter 6 control sample (prepared using a 2.5 mole excess of HCl(g)) (Fig. 6.8), the Cl 2p signal of the HCl product must be part of the broad signal seen at 198.2 eV. Synthetic-scale studies at R.T. were performed (with X = Cl, Br, and I) using finely ground Y-alumina to substantiate and quantify the HX product in reaction 6.1.2. The alumina was isolated and washed with MeOH which eluted all physisorbed 2. The chemisorbed HX was then titrated using a standardized NaOH solution, and the amount of acid determined agrees excellently with the theoretical values. This finding along with that of species 2 being recovered in -100% yields clearly shows that reaction 6.1.2 proceeds quantitatively with respect to species 9. The mechanism of reaction 6.1.2 can only be speculated upon at this point. As mentioned, reaction of 9 with H 2S probably proceeds via initial activation of H 2S on the surface of the alumina. The effect of surface area was clearly demonstrated by comparison of the extent of the reaction using non-ground (10% complete immediately - sample 5) versus finely ground dumina (100% completion of reaction immediately - sample 9, for instance). The surface area of non-ground y-alumina is typically in the range 100 - 200 m 2 g" 1 5 and that of finely ground dumina certainly higher (a vdue could not be found in the literature). The grinding process presumably dlows more "active" sites to be exposed-for interaction with H 2S. The y-dirrnina used here belongs to a group of "transition duminas" having high surface areas resulting from the cdcination of Al(OH)3 at relatively low temperatures (200 - 600 °C), 5 when a three-dimensiond network of Al-O-Al linkages is formed via the elimination of H 2 0. Chromatographic active sites are dso created, and these are characterized by four main types6 shown below: [note: the surface structure of dumina is not well understood, and the following active site structures are suggestions based on adsorption,6 IR,6'16 and surface model17 studies] 260 References on page 274 Chapter 6 0 OH OH OH 1 I I / O ^ | ,A1 Al Al Al Al Al ^ O ^ ^ ^ O ^ O ^ O ^ v H-bonding acid-base pair sites OH OH OH O" OH I + I I I I ,A1 Alt Al Al Al Al x o" ^cf \ ^ ^of ^cT v Lewis acidic sites basic sites Interaction of H 2S with some or all of these active sites may lead to S-H bond activation that could be followed by nucleophilic substitution of an X" ligand on 9 by SH" to form a reactive species such as PdX(SH)(dpm): 8+ 8"^H Y PdX2(dpm) H X y \ , * I j I + < Pd / / / > / / / ** / / / > / / / P S S H Interaction with active sites is supported by observations that when non-ground alumina pre-dried at a higher temperature (150 vs. 75 °C) was used, reaction 6.1.2 with X = CI was -60% complete immediately (vs. 10% - sample 5) (Fig. 6.2d - sample 6). y-Alumina can have a surface completely covered with a monolayer of H 2 0 molecules, and drying at high temperatures (>100 °C) permits exposure of more active sites;18 catalytic activity can depend largely upon the extent to which alumina is dried.16 Further support for interaction of H 2S with active sites comes from observations that when 9a was added to non-ground alumina prior to the introduction of H 2S, the dinucleation reaction scarcely proceeded (sample 4); 9a perhaps interacts with active sites and 261 References on page 274 Chapter 6 blocks access to the H 2S. The reaction, however, proceeded to completion when TLC plates were used and the same experimental conditions employed (9a being added first - sample 15). This finding is rationalized by considering the mesh size of the dumina particles of the plate (60) compared to that of the non-ground dumina (av. 200). For the same amount of materid used, the lower mesh will have a higher totd surface area and more active sites than the higher mesh size dumina, assuming that the density of active sites is the same for the two types of materids (a density vdue could not be found in the literature). The dinucleation reaction to form Pd2X2(u.-S)(dpm)2 (2) could be a result of the coupling of PdX(SH)(dpm) with itself, or with 9 via a deprotonation/protonation process (see Chapter 5): Scheme Considering that HX in the free state reacts quickly with 2 (to form 9 and H2S), the driving force for the dinucleation reaction would be the chemisorption of HX so that the reverse reaction cannot take place. 262 References on page 274 Chapter 6 Support for S-H bond activation comes also from the observed reaction of PdCkCdpm) (9a) with H2S using a-alumina (corundum), this giving 2a as the only Pd product (sample 10). a-Alumina is both chemically inert7 and chromatographically inactive;6 it is formed from the calcination of Al(OH)3 or transition duminas at temperatures > 1100 °C such that water is completely eliminated, resulting in a material with a very low surface area (~1 m 2 g 1). 6 The complete absence of "active" sites is in keeping with observations that reaction 6.1.2 proceeds very slowly, being -10% complete after 1 d and -50% after 3 d (Fig. 6.3). As a-dumina is composed of a three-dimensiond network of Al-O-Al linkages shown below, 1 A l ^M—o o I U+ I 5 + — A l I 5" ls+ 1 Al Al A k 0-—" o o-- A 1~-o o I I I I I - A I ^ Q A 1 ^ Q ^ A 1 ^ . Q - A I ^ Q ^ A U S-H bond activation probably occurs because of the polarity of the Al-0 bond. Of note, adsorption studies of dcohols on y-dumina have shown that ROH can interact with unsaturated durninum ions (Lewis acidic sites) with O-H bond activation leading to chemisorption.19 In the present system, a similar chemisorption process is readily envisaged with H2S but, considering that reaction 6.1.2 does not proceed when F^S-pretreated dumina is removed from the system, dried under Ar, and subsequently placed back in the system (with no H 2S present in solution) (see above), this suggests that other types of active sites (for physisorption) may be responsible for S-H bond activation prior to the dinucleation reaction. Moreover, reaction 6.1.2 does not proceed to 263 References on page 274 Chapter 6 completion when a stoichiometric amount of H2S is used, and this observation could be interpreted as support for the role of other types of active sites. Although the proposed reaction intermediate PdX(SH)(dpm) was not observed, for example by NMR studies for the chloride system at -50 °C, its existence is supported by reaction studies of PdCl2(dpm) (9a) with 1 mole equivalent NaSH which show that 2a forms according to reaction 6.3.1. As noted earlier, previous studies in this laboratory9 demonstrated that reaction of 9a with excess NaSH gives exclusively Pd2(SH)2(p-S)(dpm)2; nucleophilic substitution of both Cl" ligands of 9a supposedly forms the reactive intermediate, Pd(SH)2(dpm), which couples to form the product with concomitant elimination of H2S. (6.3.1) 2 PdCl2(dpm) (9a) + 2 NaSH • Pd2Cl2(p-S)(dpm)2 (2a) + 2 NaCl + H 2 S In the present study, reaction of 9a with 1 mole equivalent NaSH under conditions in which NaSH was slowly added to a solution of 9a gave 2a as the only Pd product; Pd2(SH)2(p-S)(dpm)2 was not formed. It seems reasonable that, under these conditions in which the local concentration of NaSH is kept low, double nucleophilic substitution does not take place and PdCl(SH)(dpm) forms which either reacts with itself or with 9a (Scheme) (the HCl formed will react with 2a to give 9a andH2S). Important implications of reaction 6.1.2 are seen with the X = I system; photodecomposition of HI, for instance, can produce H 2 and I2, the latter then reacting with Pd2I2(p-S)(dpm)2 (2c) to form elemental sulfur and Pdl2(dpm) (9c). The catalytic cycle to generate H 2 and 1/8 Sg from H 2S is then completed via reaction 6.3.2 (see Chapter 4). (6.3.2) Pd2I2(p-S)(dpm)2 (2c) + I2 • 2 Pdl2(dpm) (9c) + sulfur Some qualitative studies were carried out to determine the extent of photodecomposition by light (laboratory and/or sunlight) of HI in CDCI3 in the presence of y- and a-alumina. Transition 264 References on page 274 Chapter 6 aluminas chemisorb hydrohalic acids,6 and this type of interaction may hinder such photodecomposition. Chemisorbed Ffl on y-alumina isolated from a synthetic-scale experiment was shown to undergo partial photodecomposition but intense light (TLC Hg vapour lamp (18W)) was needed. The amount of I2 produced after 4 h was small (~1 x 10"7 mol), however, and no H 2 was detected in NMR and GC analyses; exposure times longer than 4 h had no effect on the system (i.e. the I2 512 nm band intensity remained unchanged and H 2 was still not detected). Interestingly, the alumina itself acquired an orange colour which was deterrriined to be due to physisorbed L". Elution of this substance using MeOH produced a white alumina that was subsequently titrated using a standardized NaOH solution. This titration showed the presence of much chemisorbed HI remaining; as noted in the Results, all the T can be accounted for. Although it is not known how HI is chemisorbed onto alurnina or which active sites are responsible for reaction 6.1.2, formation of I2 probably results from the coupling of photoproduced I atoms, and the formation of I3" comes from reaction of I2 with chemisorbed T (as evidenced from separate studies). The latter reaction could be responsible for the partial photodecomposition observed as I3" exhibits strong absorption (au* -> o"g*, nu* -> CTg*) in the ultraviolet region (Fig. 6.9b). I2 formation occurs until sufficient I3" forms, this absorbing most or all the light and terrninating further photodecomposition. As noted above, titrations studies showed that much chemisorbed Ffl remains (i.e. 2.8 x 10"5 mol, ~85%) and the I3~ formed accounts for the 15% difference. Hence, -10% of chemisorbed HI was photodecomposed to form I2, most of which reacted with T to generate I3' with a negligible amount appearing in solution. Why H 2 was not observed is not clear (but see below) although the theoretical amount produced (1.7 x 10"6 mol) based on the amount of I3" formed should make the H 2 easily detectable, for example by NMR spectroscopy (-3.4 x 10'3 M). *H NMR analyses in CDCI3 showed no H 2 signal nor any other signals except for that of residual CHC13. Photodecomposition tests with 265 References on page 274 Chapter 6 HI(57% aq.) on y-alumina revealed the formation of I2 but the H 2 product was not detected (e.g. by gas chromatography). Exposure of Hi-treated a-alumina in CDCI3 to light resulted in an initially purple solution (the formation of I2 being evidenced by UV-vis spectroscopy) that gradually darkened to a black colour. NMR analyses revealed a broad signal at 8 5.2 indicative of C D H X 2 species (X = CI and/or I) (Fig. 6.11). It is possible that photochemically produced H atoms react with the solvent to give, for example, HCl and a CDC12* radical which could subsequently react with H or I atoms to give species such as CDHC12 or CDC12I (I atoms do not react with CHCI3 or alkanes in general (see Chapter 4)20). As a-alumina is chemically inert, only physisorption of H X can take place (by dipole-dipole interactions), and any H atoms produced in the photolysis probably react with the surrounding solvent species. Exposure of Hi-treated a-alumina to light without the presence of solvent also resulted in the formation of I2 but surprisingly no H 2 was detected; the presence of I2 was evidenced later by UV-vis spectroscopy when the alumina was subsequently placed in CDCI3 and a purple solution resulted. The colour did not change and the UV-vis spectral characteristics were invariant with time; it is not clear why H 2 was not observed here. A single catalytic study at R.T. in CDCI3 demonstrated that reaction of 9c with H 2 S (1 atm) in the presence of finely ground y-alumina proceeded to completion to give 2c but subsequent exposure to light (laboratory and/or sunlight) for 10 d resulted in decomposition giving unidentifiable products; no H 2 was detected. Although the following illustrates an attractive scheme for the catalytic decomposition of H 2S to give H 2 and elemental sulfur using 9c and alumina as catalyst and co-catalyst, respectively, more studies need to be carried out to determine the fate of the elusive H 2 and to determine the best set of conditions for catalysis to operate: 266 References on page 274 Chapter 6 Net: H2S *- H 2 + sulfur As a final note, it has been emphasized in the literature that dumina is a general term referring to a flexible material that cannot be defined by a single parameter as its properties can vary widely.7 And although future studies are suggested, it is recommended that care should be taken when selecting an appropriate material for these catalytic studies. 267 References on page 274 Chapter 6 6.4 Experimental section The materials used, synthetic procedures for the ligands and complexes, and instrumentation used for and 31P{1H} NMR and UV-vis spectroscopy, gas chromatography, and ESCA (electron spectroscopy for chemical analysis) were described in Chapter 2. Complexes 2a - c and 9a - c were synthesized using published methods as outlined in Chapter 2. All experiments were performed under Ar unless otherwise specified. Table 6.1 summarizes the content make-up of each experiment; the alumina powder and plates used were pre-dried at 75 °C for 24 h. Finely ground dumina was prepared via a manual grinding process using a mortar and pestle. 6.4.1 NMR spectroscopic studies Expt. 1: Alurnina powder (15 mg) was placed in an NMR tube fitted with a PTFE J. Young valve, and the tube was evacuated before CDCI3 (-0.7 mL) was vacuum transferred in using liquid N 2 . The system was left at R.T. for 24 h before analysis using NMR spectroscopy. Expt. 2: As described for Expt. 1. H 2S (110 uL at STP, 0.0045 mmol) was then vacuum transferred in using liquid N 2 . Expt. 3: As described for Expt. 1 but PdCl2(dpm) (9a, 5 mg, 0.0089 mmol) was also used. Expt. 4: As described for Expt. 3. H 2S (110 uL at STP, 0.0045 mmol) was vacuum transferred in using liquid N 2 after -1 h. Expt. 5: AJumina powder (15 mg) and CDCI3 (0.3 mL) were placed in an NMR tube fitted with a rubber septum. H 2S (110 uL at STP, 0.0045 mmol) was then injected, and the NMR tube was 268 References on page 274 Chapter 6 briefly shaken to ensure complete mixing. A 0.4 mL solution of 9a (5 mg, 0.0089 mmol) was injected after ~1 h. The system was permitted to react at R.T. for 24 h before analysis using NMR spectroscopy. Expt. 6: As described for Expt. 5 except the alumina used was pre-dried at 150 °C instead of 75 °C. Expt. 7: An alumina plate (1.0 cm x 1.0 cm) and CDCI3 (1.0 mL) were placed in a Schlenk tube sealed with a rubber septum. H2S (0.5 mL at STP, 0.021 mmol) was injected, and the system was left at R.T. for ~4 h. The plate was then removed from the solution, dried under a flow of Ar and then placed in a Schlenk tube containing a 1.0 mL CDCI3 solution of 9a (5 mg, 0.0089 mmol). The system was permitted to react at R.T. for 4 h before analysis. Expt. 8: The alumina (11.5 mg) was scraped off a plate (1.0 cm x 1.0 cm) and placed in a Schlenk tube containing a 1.0 mL CDC13 solution of 9a (5 mg, 0.0089 mmol). H 2S (0.5 mL at STP, 0.021 mmol) was then injected, and the system was permitted to react at R.T. for 4 h before analysis using NMR spectroscopy. Expt. 9: As described for Expt. 8 except finely ground alumina (15 mg) was used. Expt. 10: As described for Expt. 8 except a-alumina (corundum, 15 mg) was used. The sample was again analyzed after 72 h. Expt. 11: 9a (10 mg, 0.018 mmol) was dissolved in CHC13 (10 mL) and a MeOH solution (10 mL) of NaSH (1.0 mg, 0.018 mmol) was added dropwise over aperiod of 30 min: The solvents of the resulting yellow solution were removed, and the remaining yellow residue was dried in vacuo. CDCI3 (0.5 mL) was used to dissolve the residue, and the sample was analyzed using ! H and 31P{ !H} NMR spectroscopy. 269 References on page 274 Chapter 6 A single, low temperature NMR study at -42 °C was carried out in which H 2S (0.25 mL at STP, 0.010 mmol) was injected in a rubber septum-sealed NMR tube containing a 0.5 mL CDCI3 solution of 9a (5 mg, 0.0089 mmol) and finely ground alumina (15 mg) in an acetonitrile/liquid N 2 slush-bath. The sample was analyzed at -50 °C 1 h later using X H and "Pj/H} NMR spectroscopy. The sample was then placed at R T . for 24 h, and re-analyzed thereafter. 6.4.2 Surface studies Expt. 12: An dumina plate (1.0 cm x 1.0 cm) and CDCI3 (1.0 mL) were placed in a Schlenk tube sealed with a rubber septum. The sample was then left at R T . for ~4 h before analysis. Expt. 13: As described for Expt. 12 but 9a (5 mg, 0.0089 mmol) was also placed in the Schlenk tube. Expt. 14: As described for Expt. 12 but H 2S (0.5 mL at STP, 0.021 mmol) was also injected into the Schlenk tube. Expt. 15: An alumina plate (1.0 cm x 1.0 cm), 9a (5 mg, 0.0089 mmol), and CDC13 (1.0 mL) were placed in a Schlenk tube fitted with a rubber septum. The sample was left at R.T. for ~2 h before H 2 S (0.5 mL at STP, 0.021 mmol) was injected. The reaction mixture was then left at R.T. for another 2 h before analysis. Expt. 16: An alumina plate (1.0 cm x 1.0 cm) and CDCI3 (0.7 mL) were placed in a Schlenk tube fitted with a rubber septum, and H 2S (0.5 mL at STP, 0.021 mmol) was injected. The sample was left at R.T. for ~2 h before a 0.3 mL solution of 9a (5 mg, 0.0089 mmol) was introduced. The reaction mixture was then left at R T . for another 2 h before analysis. 270 References on page 274 Chapter 6 Expt. 17: An alumina plate (1.0 cm x 1.0 cm), Ss (12 mg, 0.047 mmol), and CDCU (1.0 mL) were placed in a Schlenk tube, and the system was left at R.T. for ~2 h prior to analysis. Expt. 18: As described for Expt. 17 but HCl (0.5 mL, 0.021 mmol) was placed (injected) in the Schlenk tube instead of sulfur. The liquid phases of Expts. 12-16 were analyzed using lH and 31P{1H} NMR spectroscopy. The alurnina plates of all experiments were taken out and dried at R T . under Ar (few minutes) before they were submitted for surface analyses using ESCA. 6.4.3 Synthetic-scale studies In a Schlenk tube fitted with a rubber septum were placed 25.0 mg PdX2(dpm) (9a, 0.045 mmol; 9b, 0.038 mmol; 9c, 0.034 mmol), 75.0 mg finely ground y-alumina, and 2.5 mL CHCI3. The mixture was rapidly stirred and H 2S (5 mL at STP, 0.20 mmol) was injected. The initial yellow (9a and 9b) or orange (9c) solution immediately (within seconds) turned brown. The mixture was continuously stirred for ~4 h before being reduced in volume (to -0.5 mL) with evacuation of the H 2S. CH2C12 (20 mL) was then added to dissolve all Pd species, and the mixture was filtered to separate out the alumina, which was subsequently washed with CH 2C1 2 (2x10 mL), MeOH (10 mL), and CH2C12 (10 mL), and then dried in vacuo. (Note: MeOH was used to elute all adsorbed Pd species off the alumina; indeed, when this solvent was added, the initially brown-coloured alumina turned white). The washings were combined with the filtrate, and the solvents were removed by rotary evaporation. The resulting residue was then dissolved in CDC13 (-0.5 mL) and analyzed using NMR spectroscopy. Yield of Pd2X2(u,-S)(dpm)2: 2a, 24.0 mg (100%); 2b, 22.0 mg (98%); 2c, 21.0 mg (99%). The isolated alumina was placed in an 271 References on page 274 Chapter 6 Erlenmeyer flask, and 10 mL H2O and a few drops of phenolphthalein indicator were added. The mixture was then titrated with a standardized solution of NaOH (0.010 M). The above procedure was repeated for the iodide system except the isolated alumina was placed in an NMR tube fitted with a PTFE J. Young valve with CDC13 (-0.7 mL) added. The resulting mixture was then placed in the presence of light (laboratory, sunlight, or light from a TLC Hg vapour lamp (18W)) for up to 4 d, and analyzed thereafter for H 2 (using GC chromatography) and I2 (using UV-vis spectroscopy). Two blank titrations were performed using unused alumina, and alumina isolated from the above procedure for the chloride system when H 2S was not used. Possible reactivity of 9a with Sg in the presence of y-dumina was investigated. 9a (25.0 mg, 0.045 mmol), finely ground alumina (75.0 mg), Sg (14 mg, 0.45 mmol), and CHCI3 (2.5 mL) were placed in a Schlenk tube, and the mixture was rapidly stirred for -4 h. During this period, the initial white dumina gradudly turned orange yet the colour of the supernatant was unchanged (yellow). CH2CI2 (20 mL) was then added to dissolve dl Pd species, and the mixture was filtered to separate out the dumina, which was subsequently washed with CH2CI2 (2x10 mL), MeOH (10 mL), and CH2CI2 (10 mL). The washings were combined with the filtrate, and the solvents were removed by rotary evaporation. The resulting residue was then dissolved in CDCI3 (~0.5 mL) and andyzed using NMR spectroscopy, which reveded only 9a. Possible decomposition of H2S on y-dumina was investigated. In a Schlenk tube were placed a y-dumina plate (1.0 cm x 10.0 cm), CHCI3 (20 mL), and 1 atm H 2S. The mixture was stirred at R T . for -4 h before the plate was taken out and dried in dr. The dumina was then scraped off and placed in a glass pasteur pipet. CHCI3 (-5 mL) was then used to wash the dumina, and the washings were collected and andyzed using UV-vis spectroscopy, which reveded 272 References on page 274 Chapter 6 the presence of elemental sulfur (e.g. see Fig. 4.4). A control experiment was performed using alumina isolated from the above procedure when H 2S was not used. 6.4.4 Photodecomposition of HI Photodecomposition of HI(aq.) was qualitatively studied in CDC13 in the presence of y- and a-alumina powder. Three experiments were performed as follows. In the first experiment, y-alumina (15 mg) and CDC13 (0.5 mL) were placed in an NMR tube, and HI(aq.) (~2 uL, 0.016 mmol) was added via a glass pasteur pipet. The sample was briefly shaken to ensure complete mixing before being placed at R.T. in the presence of laboratory light for 24 h. Thereafter, the system was analyzed for H 2 (using gas chromatography and NMR spectroscopy) and I2 (UV-vis spectroscopy). In the second experiment, a- or y-alumina (15 mg) and HI(aq.) (~2 uL, 0.016 mmol) were first placed in the NMR tube, and the sample was briefly shaken before it was "immediately" analyzed for H 2 . CDCI3 (0.5 mL) was then added, and the sample was again shaken briefly before being placed at R.T. in the presence of laboratory light for 24 h. Thereafter, the sample was analyzed for H 2 and I2. In the third experiment, a-alumina (15 mg) and HI(aq.) (~2 uL, 0.016 mmol) were placed in an NMR tube, and the sample was briefly shaken before being left at R.T. in the presence of light for 4 d. The sample was then analyzed for H 2 before CDCI3 was added for analysis of I2. A control sample consisting of I2 (0.01 mg, 0.052 pmol), alumina (15 mg), and CDC13 (0.5 mL) was prepared and was left at R.T. for 24 h in the presence of laboratory light. 273 References on page 274 Chapter 6 6.5 References f o r C h a p t e r 6 1. Lee, C.-L.; Besenyei, G.; James, B.R.; Nelson, D.A.; Lilga, M . A / . Chem. Soc, Chem. Commun. 1985, 1175. 2. Lee, C.-L.; Yang, Y.P.; Rettig, S.J.; James, B.R.; Nelson, D.A.; Lilga, M.A. Organometallics 1986, 5, 2220. 3. Wong, T.Y.H.; Barnabas, A.F.; Sallin, D.; James, B.R. Inorg. Chem. 1995, 34, 2278. 4. Wagner, C D . ; Riggs, W.M.; Davis, L.E.; Moulder, J.F. "Handbook of X-ray Photoelectron Spectroscopy", G.E. Muilenberg (Ed.), Perkin-Elmer Corporation, Physical Electronics Division, Eden Prairie, Minnesota, 1979, p. 43. 5. Snyder, L.R. "Principles of Adsorption Chromatography", Marcel Dekker, New York, 1968, ch. 7, p. 163. 6. Knozinger, H.; Ratnasamy, P. Catal. Rev.-Sci. Eng. 1978,17, 31. 7. Oberlander, R.K. "Applied Industrial Catalysis, Vol. 3", B.E. Leach (Ed.), Academic Press, Toronto, 1984, ch. 4, p. 63. 8. Lamb, H.H.; Gates, B.C.; Knozinger, H. Angew. Chem. Int. Ed Engl. 1988, 27, 1127. 9. Besenyei, G.; Lee, C.-L.; Gulinski, J.; Rettig, S.J.; James, B.R.; Nelson, D.A.; Lilga, M.A. Inorg. Chem. 1987, 26, 3622. 10. Parkyns, N.D. J. Chem. Soc. (A) 1969, 410. 11. Ref. 4, p. 110. 12. Ref. 4, p. 58. 13. Ref. 4, p. 56. 14. Sulfur Recovery, "Kirk-Othmer Encyclopedia of Chemical Technology (3rd Ed.), Vol. 22", H.F. Mark, J.J. Mcketta, Jr., D.F. Othmer, John Wiley & Sons, Toronto, 1983, p. 267. 15. Bandermann, F.; Harder, K.B. Int. J. Hydrogen Energy 1985,10, 21. 16. Peri, J.B.; Hannan, R.B. J. Phys. Chem. 1960, 64, 1526. 274 Chapter 6 17. Peri, J.B. J. Phys. Chern. 1965, 69, 220. 18. Peri, J.B. J. Phys. Chern. 1965, 69, 211. 19. Knozinger, H.; Stubner, B. J. Phys. Chern. 1978, 82, 1526. 20. See for example: a) Streitwieser, Jr., A.; Heathcock, C H . "Introduction to Organic Chemistry, 3rd Ed.", Macmillan, New York, 1985, ch. 6. b) Morrison, R.T.; Boyd, R.N. "Organic Chemistry, 3rd Ed.", Allyn and Bacon, Boston, 1973, ch. 3. 275 CHAPTER 7 Miscellaneous Reactions 276 Chapter 7 7.1 Reaction of Pd2X2(n-S)(dpm)2 with alkyl halides While the oxidative addition of halogens to Pd2X2(p-S)(dpm)2 (2) was being studied (see Chapter 4), tests for S-removal were made using alkyl halides; a slow, complex reaction occurs and eq. 7.1 describes the products observed: (7.1) Pd2l2(M.-S)(dpm)2 (2c) + RI Pd2I2(dpm)2 (lc) + Pdl2(dpm) (9c) CHCI3 R = Me, Et + PdI2(dpm(S) (11c) + R-S-R In a single in situ NMR experiment at R.T., 2c (10 mg, 0.0079 mmol) was reacted with a 10-fold excess of Mel in CDC13 (0.5 mL) in air. lU and 31V{lU) NMR analyses after 3 d showed, in addition to unreacted 2c and Mel, the presence of PdL^dpm) (9c), Pd2l2(dpm)2 (lc), PdI2(dpm(S)) (11c), Me 2 S0 2 , and Me2S (Fig. 7.1). Reaction of 2c (10 mg, 0.0079 mmol) with a 10-fold excess of Etl in CDCI3 (0.5 mL) under air was slower, and trace amounts of lc, 9c, and 11c were observed after 5 d (Fig. 7.2); the ' H NMR signals for the expected Et2S product are probably buried under those of Et2l (compare the ' H NMR signal of Me2S with that of Mel). In both of these reactions, addition across the Pd-S bond is readily envisaged to account for the 9c and R 2S products (reaction 7.2). 277 References on page 285 Chapter 7 Fig. 7.1. lR (200 MHz) and 31P{XH} (81 MHz) NMR spectra showing the progress of the reaction in CDC13 at R T . of Pd2l2(p-S)(dpm)2 (2c, 1.6 x 10"2 M) with a 10-fold excess of Mel after 3 d; lc = Pd2l2(dpm)2, 9c = Pdl2(dpm), 11c = PdI2(dpm(S)). [The spectra of the various Pd species are discussed in Sections 3.2 (lc and 2c) and 4.2 (9c and 11c).] (Me) 2S0 2 Mel H 2 0 ^ ^ M e - S - M e ^ 5.0 4.0 3.0 P P M 2.0 3 1 p { l H j 1 lie \ 2c lc 9c ~ i — -20 — i — -40 — i — -60 —1~ 60 40 20 0 P P M 278 References on page 285 Chapter 7 Fig. 7.2. *H (200 MHz) and 31P{!H} (81 MHz) NMR spectra showing the progress of the reaction in CDC13 at R T . of Pd2I2(p-S)(dpm)2 (2c, 1.6 x 10"2 M) with a 10-fold excess of EtI after 5 d; lc = Pd2I2(dpm)2, 9c = Pdl2(dpm), 11c = PdI2(dpm(S)). EtI 2c -» 1— 5.0 9c lc 11c 1 i ~ . -*\r i t H 2 0 4.0 3.0 P P M 2.0 3 1 p | l H j 11c 60 —1— 40 2c lc —1— 20 0 P P M —1— -20 — 1 — -40 9c -60 279 References on page 285 Chapter 7 (7.2) P h 2 P — ^ P P h 2 I- -Pd-I Ph7P--Pd. -I + 2 PJ -PPh, Ph2P- ^PPh 2 I A I / I Pd Pd I I I"" I Ph2P~, PPh2 + R-S-R 2c 10c Rl Rl Ph 2P- -PPh 2 1 . A | / S R ' I K | Ph2P^ ~PPh2 (see Chapter 4) Ph2P .1 < Pd > ' \ Ph,P I 9c Abstraction of the bridged S-atom by R* radicals1 could lead to establishment of the Pd-Pd bond and formation of R2S and lc (reaction 7.3). (7.3) R-I -> R» + !• PhzP-- ^PPh 2 Ph2P --PPh2 I c I I I T ^ P d ^ 5 ~ ~ - P d - ^ . T + 2R« *" I—Pd* 'Pd—I + R-S-R 1 I I 1 I I Ph2P -PPh2 Ph2P-> - P P h 2 lc The I atoms could form I2 which would react with lc or 2c to give 9c (and presumably elemental sulfur) and 11c (from reaction of I2 with 2c, see Chapter 4). In the 2c + Mel system, the presence of the siilfone* must result from oxidation of Me2S (reaction 7.1 was carried out under 1 atm air) and, considering that Me2S does not exhibit reactivity towards 0 2 as evidenced from separate x The presence of Me 2S0 2 was established by comparison of the ! H NMR signal with that of an authentic sample. 280 References on page 285 Chapter 7 NMR studies,* this oxidation is perhaps catalyzed by a Pd species; why an intermediate sulfoxide species is not observed is also not clear. Finally, the difference in reaction rates between the Mel and EtI systems is presumably due to steric reasons. 7.2 Reaction of PdX2(dpm) (9) with H 2 S in the presence of silica gel or aluminosilicate; catalyzed formation of Pd2X2(p-S)(dpm)2 (2) The use of alumina in the dinucleation reaction PdX2(dpm) (9) - » Pd2X2(u,-S)(dpm)2 (2) (Chapter 6) led to some studies utilizing other metal oxides, namely silica gel and duminosilicates (e.g. molecular sieves). In a single in situ NMR experiment at R.T., reaction of PdCl2(dpm) (9a, 5 mg, 0.0089 mmol) with 1 mole equivalent H 2S (216 uL at STP) in the presence of silica gel (15 mg) was carried out in CDCb (0.5 mL). X H and 31P{1H} NMR analyses immediately after sample preparation revealed, in addition to unreacted 9a and H 2S, the presence of 2a in -10% yield; there were no further changes in the system. The mechanism of this reaction is probably similar to that when alurnina is used (see Chapter 6). The same reaction, when attempted in the presence of molecular sieves (15 mg, type 4A, 8-12 mesh, beads, Fisher), proceeded only in the presence of light (18 W TLC Hg vapour lamp) (reaction 7.4). For the chloride system (5 mg, 0.0089 mmol), the reaction was complete after ~2 h, while the bromide (5 mg, 0.0077 mmol) and iodide (5 mg, 0.0067 mmol) systems were only -10 and -1% complete, respectively, after 2 h. The FIX product, as in the alurnina system, is likely chemisorbed on the metal oxide. Interestingly, subsequent exposure of the chloride system to sunlight for -2 d resulted in the complete disappearance of 2c ' Me2S (250 uL, 3.4 mmol) was dissolved in CDC13 (0.5 mL) held in an NMR tube. The sample, prepared under 1 atm air, was analyzed after 10 d using 'H NMR spectroscopy. 281 References on page 285 Chapter 7 and re-formation of 9c as evidenced by NMR measurements (whether the S-containing co-product was elemental sulfur or H2S was not determined). hv (7.4) 2 PdX2(dpm) (9) + H 2S • Pd2X2(n-S)(dpm)2 (2) + '2 H X ' molecular sieve X = CI, Br, I 7.3 Reaction of Pd2X2(p-S)(dpm)2 (2) with CO Reaction of 2 with CO was briefly explored for the possibility of abstracting the bridged S atom as COS (reaction 7.5): (7.5) Pd2X2(u-S)(dpm)2 (2) + CO • Pd2X2(dpm)2 (1) + COS In a high pressure autoclave, Pd2Cl2(p-S)(dpm)2 (2a, 10 mg, 0.0092 mmol) was dissolved in CDCI3 (0.5 mL) and subjected to 700 psi CO at R T . for 24 h before the system was analyzed (the sample in the autoclave was transferred to an NMR tube using a glass pasteur pipet). The 31P{ !H} NMR spectrum shows the absence of 2a and the presence of PdCl2(dpm) (9a), Pd2Cl2(dpm)2 (la), Pd2Cl2(p-CO)(dpm)2,2 PdCl2(dpm(S)) (11a), and an unknown species X characterized by a tight AB pattern (Fig. 7.3). Reaction under an atm of CO does not occur for the chloride system but for the iodide system, a slow reaction was observed. In a single in situ NMR experiment at R.T., Pd2I2(p-S)(dpm)2 (lc, 10 mg, 0.0079 mmol) was reacted under an atm of CO in CDC13 (0.5 mL). NMR analysis after 1 d revealed essentially no reaction, but after l i d analysis revealed, in addition to unreacted lc, the presence of Pdl2(dpm) (9c), Pd2I2(dpm)2 (lc), Pd2I2(p.-CO)(dpm)2,2 PdI2(dpm(S)) (11c), and perhaps two unknown species Y and Z (Fig. 7.4) (note: the reaction was carried out in the absence of light). Species Y, as with species X, is characterized by a tight AB 282 References on page 285 Chapter 7 Fig. 7.3. "P^H} NMR spectrum (81 MHz) recorded for the reaction of Pd2Cl2(p-S)(dpm)2 (2a) with 700 psi CO at R T . in CDC13 after 24 h; la = Pd2CI2(dpm)2) 9a = PdCl2(dpm), 11a = PdCl2(dpm(S)), X = unknown. [The spectra of the various Pd species are discussed in Sections 3.2 (la) and 4.2 (9a and 11a).] Pd2Cl2(u.-CO)(dpm)2 l a 9a -10 -30 -50 P P M Fig. 7.4. 3IP{ !H} NMR spectrum (81 MHz) recorded for the reaction of Pd2I2(p-S)(dpm)2 (2c) with 1 arm CO at R.T. in CDC13 after 11 d; lc = Pd2I2(dpm)2, 9c = Pdl2(dpm), 11c = PdI2(dpm(S)), Y, Z = unknown. 283 References on page 285 Chapter 7 pattern in the 31P{1H} NMR spectrum and species Z by a multiplet at 8 -3. Species X and Y could be dinuclear Pd complexes with bridging COS moieties as shown below; dinuclear metal complexes with such bridging COS modes are known.3 Of note, no reaction of 1 (la, 10 mg, 0.0095 mmol; lc, 10 mg, 0.0081 mmol) with 1 atm COS in CDC13 (-0.5 mL) at R T . after 5 d was seen as evidenced by NMR measurements. x ^ p d P d ^ x P d p d ^ x p - ^ p PC ^ p 284 References on page 285 Chapter 1.4 References for Chapter 7 1. Collman, J.P.; Finke, R.G.; Cawse, J.N.; Brauman, J.I. J. Am. Chern. Soc. 1977, 99, 2515. 2. a) Benner, L.S.; Balch, A.L. J. Am. Chern. Soc. 1978, 700, 6099. b) Olmstead, M.M. ; Hope, H.; Benner, L.S.; Balch, A.L. J. Am. Chern. Soc. 1977, 99, 5502. c) Lee, C.-L.; James, B.R.; Nelson, D.A.; Hallen, R.T. Organomet. 1984, 3, 1360. 3. Pandey, K.K. Coord. Chern. Rev. 1995,140, 37. 285 CHAPTER 8 General Conclusions and Recommendations for Future Studies 286 Chapter 8 In this thesis work, the interaction of H2S with mono- and dinuclear Pd-dpm complexes was investigated with the main goal being to develop catalytic processes for the generation of H2 and elemental sulfur from H 2S. Several systems were explored and these are summarized in the following scheme. X — P d alumina/CHC ,^ or DMSO Process 1 -> 2 was previously studied by others of this laboratory and was shown to proceed via hydrido mercapto intermediates. Kinetic and mechanistic studies were carried out on the re-conversion process 2 -> 1 in the present work with the finding that the bridging diphosphine ligands are also involved in the S-abstraction such that the distribution of monosulfide products is close to statistical; no intermediates were observed. The net reaction is shown in eq. 8.1 for which some studies were done on the X = Br system substantiating the catalytic nature of 1 and 2; eq. 8.1 is the first reported homogeneously catalyzed conversion of H2S to Ffc. lor 2 (8.1) dpm + H 2S -> dpm(S) + H 2 Depending on conditions, a catalytically inactive species with the formulation Pd2Br2(dpm(S))(dpm)2 also forms, its origin and identity not yet unequivocally established. The 287 Chapter 8 poor crystallinity of this species precluded an X-ray crystallographic analysis, and it is suggested that in future studies good crystal growth may be accomplished using low temperature conditions. The removal of the bridged S atom from 2 by halogens (process 2 —> 9) was discovered, and kinetic and mechanistic studies on the X = I system revealed that the reaction proceeds via oxidative addition thereby generating an intermediate species, Pd2X4(dpm)2, which undergoes unimolecular decomposition to give 9. The recovery of sulfur in its elemental form is attractive in terms of possible commercialization of an H 2S -> H 2 + 1/8 Sg process. The dinucleation process 9 -> 2 was examined in detail in both DMSO and CHCI3 solvents. In the former solvent, an equilibrium reaction of 9 with H 2S proceeds to form 2 and HX; however, a side-product thought to be Pd(SH)2(dpm) also forms. In CHCI3 on the other hand, the reaction of 9 with H 2S proceeds completely and cleanly, with alumina necessarily present and functioning as a heterogeneous catalyst for the activation of H 2S prior to the dinucleation process. Ligand substitution to form the intermediate species PdX(SH)(dpm) is postulated based on reaction studies of 9 with NaSH; subsequent reaction of this intermediate with itself or 9 to form 2 is proposed. The HX product (chemisorbed on the alurnina in the CHCI3 system) was quantified by titration studies. The overall process 2 —> 9 -> 2 with X = I constitutes a catalytic cycle as photodecomposition of HI would produce H 2 and the necessary I2 to continue the cycle; the net reaction would be the photocatalytic decomposition of H 2S to H 2 and elemental sulfur. Some catalytic studies were performed but the generation of H 2 remains to be substantiated; future studies should concentrate on detennining the fate of the elusive H 2 . During the course of this thesis work, some interesting observations, as summarized in Chapter 7 and in the following scheme, were made and these deserve further investigation: 288 Chapter X X v PH CO OH <N X N £ ^ ifi I 0=0 \ PH p; 01 PH 01 PH \ PH X X I T J P H — g — P H OH T3 PH 3 PH PH I PH X X X PH PH PH •13 V /H=J PH \ / P H X X 2 PH PH X / PH 43 •PH ^ .a s « CO PH T) PH X T CO X X PH PH PH oi X X X V PH GO 0 1 / / -PH (S 289 Chapter 8 Future studies should concentrate on unraveling further the nature of each of the three reactions shown. For example, a radical process may be involved in the reaction of 2 with RX, and addition of a radical scavenger, e.g. TEMPO,* would probably yield informative data. In the reaction of 2 with CO, the identity of species thought to be Pd2X2(p-COS)(dpm)2 should be established as this may be an intermediate en route to formation of 1; furthermore, the issue of how species 9 and 11 are formed should also be addressed. Finally, in the dinucleation reaction of 9 with H 2S to form 2 using silica gel or molecular sieve/hv, the mechanism may be similar to that when alumina was used (Chapter 6); in future studies, the question of how light is involved in the molecular sieve system should also be answered. TEMPO = 2,2,6,6-tetramethyl-l-piperidinyloxy, free radical 290 APPENDIX 291 Appendix I Appendix I - Free energy of decomposition of H 2 S, A G d The free energy of decomposition of H 2S, AGd, at a particular temperature is determined from the absolute free energies of H 2 , sulfur (S8(solid, liquid, gas) or S2(gas)), and H 2 S using the following general equation: AGd(H2S) = [G(H2) + G(sulfur)] - G(H2S) The G(H2S), G(H2), and G(sulfur) values are obtained from thermodynamic data tables found in ref. 1 (Table 1.1). The equilibria that exist at certain temparature ranges are as follows:2 T < 718 K Ff2S H 2 + 1/8 S8(solid, liquid) T = 718 - 900 K H 2S 5 = ^ H 2 + 1/8 S8(gas) T > 900 K H 2S 5==^ H 2 + 1/2 S2(gas) The dependence of AGd(H2S) on temperature is shown in Table 1.1 and is illustrated in Fig. 1.1. 50-45-404 35-\ 30-E 3 25-•* < 20-15 -10-j 5-200 400 600 800 1000 1200 1400 1600 1800 Temperature, K 2000 Fig. 1.1. Dependence of the free energy of decomposition of H 2S, AGd(H2S), on temperature in the range 298 to 1800 K. 292 Appendix I Table 1.1. Absolute free energies values, G, for H 2S, H 2 , S2(g), Sg(g), and S8(s,l), and free energy of decomposition for H 2S, AGa(H2S), in the temperature range 298 to 1800 K. T ( K ) G(H 2) (kcal mol"1) G(S2(g)) (kcal mol') G(S„(g)) (kcal mol"1) G(S8(s, 1)) (kcal mol"1) G(H2S) (kcal mol"1) AG d(H 2S) (kcal mol"1) AdcHzS) (kJmor 1) 298 -9.31 14.59 -6.46 -2.28 -19.56 7.98 33.38 300 -9.36 14.49 -6.65 -2.29 -19.65 8.00 33.46 400 -12.59 8.91 -17.53 -3.16 -24.69 8.94 37.40 500 -16.00 3.13 -19.42 -4.33 -29.95 9.63 40.28 600 -19.54 -2.82 -42.14 -5.68 -35.39 10.17 42.56 700 -23.20 -8.91 -55.56 -7.16 ^0.98 10.62 44.43 800 -26.96 -15.12 -69.58 - -46.71 11.05 46.23 900 -30.82 -21.44 -84.14 - -52.57 11.23 47.00 1000 -34.75 -27.86 -99.18 - -58.54 9.85 41.23 1100 -38.76 -34.37 -114.65 - -64.61 8.67 36.28 1200 ^2.83 ^10.95 -130.51 - -70.79 7.48 31.30 1300 -16.97 -47.61 -146.73 - -77.06 6.29 26.31 1400 -51.16 -54.34 -163.28 - -83.42 5.09 21.31 1500 -55.41 -61.13 -180.14 - -89.87 3.90 16.31 1600 -59.71 -67.98 -197.29 - -96.41 2.71 11.33 1700 -64.06 -74.89 -214.72 - -103.02 1.52 6.37 1800 -68.45 -81.85 -232.40 - -109.71 0.34 1.42 References 1. Fukuda, K.N.; Dokiya, M. ; Kameyama, T.; Kotera, Y. Ind. Eng. Chem. Fundam. 1978,17, 243. 2. Barin, I.; Knacke, O. "Thermochemical Properties of Inorganic Substances", Springer-Verlag, Berlin, 1975, pp 316, 317, 325, 648-651. 293 Appendix II Appendix II - Kinetic data from the study of the reaction: Pd2X2(p.-S)(dpm)2 (2) + dpm 121 (M) rdpml (M) T (°C) kobs (s'1) 2b, 6.52 x 10"5 6.52 x 10"3 20 1.27 x 10 - 4 2b, 6.52 x irr5 1.30 x 10"2 20 2.76 x lO" 4 2b, 6.52 x lO - 5 1.96 x 10"2 20 4.44 x 10"4 2b, 6.52 x 10'5 2.61 x 10"2 20 5.63 x 10"4 2b, 6.52 x 10"5 6.52 x 10"3 25 1.99 x 10"4 2b, 6.52 x 10"5 1.30 x l O - 2 25 4.14 x 10"4 2b, 6.52 x 10"3 1.96 x 10"2 25 5.99 x 10"4 2b, 6.52 x 10"5 2.61 x 10"2 25 8.07 x lO - 4 2b, 6.52 x 1(T5 6.52 x 10"3 30 2.77 x l O " 4 2b, 6.52 x 10"5 1.30 x l O " 2 30 5.27 x 10"4 2b, 6.52 x 1(T5 1.96 x l O ' 2 30 7.73 x 10"4 2b, 6.52 x 10'5 2.61 x 10"2 30 1.05 x l O " 3 2b, 6.52 x lO"5 6.52 x 10° 35 3.57 x 10"4 2b, 6.52 x lO"5 1.30 x 10"2 35 7.06 x 10"4 2b, 6.52 x 10"5 1.96 x 10"2 35 1.04 x 10"3 2b, 6.52 x 10"5 2.61 x 10"2 35 1.38 x 10- 3 2b, 3.26 x l O " 5 1.96 x 10"2 25 5.99 x lO" 4 2b, 1.63 x 10"5 1.96 x 10"2 25 5.32 x 10' 4 2b, 8.15 x i r j 6 1.96 x l O " 2 25 5.95 x lO - 4 2b, 1.30 x lO - 4 1.96 x 10"2 25 5.60 x 10"4 2b, 6.52 x lO' 5 ; (nPr)4NBr, 6.52 x 10'3 1.92 x l O " 2 25 5.88 x 10"4 2a, 6.52 x 10"5 6.52 x 10"3 30 8.18 x 10"4 2a, 6.52 x 10"5 1.30 x 10"2 30 1.60 x 10"3 2a, 6.52 x 10"5 1.96 x 10 2 30 2.36 x 10"3 2a, 6.52 x 1 0 5 2.61 x 10 - 2 30 3.29 x 10"3 2a, 6.52 x 10"5 1.96 x 10"2 20 1.31 x 10' 3 2a, 6.52 x 10"5 1.96 x 10'2 25 1.70 x 10"3 2a, 6.52 x 10"5 1.96 x l O " 2 25 1.81 x lO - 3 2a, 6.52 x 10"5 1.96 x l O " 2 35 3.07 x 10"3 2a, 3.25 x 10"5 1.96 x 10"2 30 2.38 x 10 - 3 2a, 1.30x10-" 1.96 x l O " 2 30 2.30 x 10"3 2c, 6.52 x 10'5 6.52 x 10'3 25 7.2 x 10"5 294 Appendix III Appendix III - Computer program to simulate the outcome of the reaction between Pd2Br2(p>S)(dpm)2 and n mole equivalent dpm-d2 'Program written in qbasic to determine the statistical outcome of the reaction: Pd2(dpm)2(b-S)Br2 + ndpm-d2 'written by Terrance Y. H. Wong, May 31, 1994. DIM A(5000) DIM B(5000) 5 CLS PRINT "TYPE IN NUMBER OF PD2(DPM)2(B-S)BR2 ="; : INPUT PD% PRINT "TYPE IN NUMBER OF DPM-D2 ="; : INPUT DPM% F O R X = 1 T O P D % A(X) = 0 'OR 1 FOR PD2(DPM)2BR2 NEXT X FOR Y = 1 TO DPM% B(Y)=1 NEXT Y N = 0 DPMS = 0 DPMSD2 = 0 10 'CLS LOCATE 22, 1: COLOR 7: PRINT "<ESC> - STOP ", COLOR 5: PRINT "<SPACE> - CONTINUE Q - QUIT P - PRINT": COLOR 7 DO RANDOMIZE TIMER X% = INT(RND * PD%) + 1 Y% = INT(RND * DPM%) + 1 GOSUB LIGEX 11N = N+ 1 LOCATE 3, 1: PRINT "PROCESSING: N = ", N LOOP UNTIL INKEYS = CHR$(27) 12 LOCATE 22, 1: COLOR 7: PRINT "<ESC> - STOP ", COLOR 5: PRINT "<SPACE> - CONTINUE Q - QUIT P - PRINT": COLOR 7 295 Appendix III DO RANDOMIZE TIMER X% = INT(RND * PD%) + 1 Y% = INT(RND * DPM%) + 1 RN = INT(RND * 10) + 1 IF RN <> 1 THEN GOSUB LIGEX: GOTO 14 IF A(X%) = 0 AND B(Y%) = 1 THEN GOSUB REACTION: GOTO 14 IF A(X%) = 0 AND B(Y%) = 0 THEN A(X%) = 1: B(Y%) = 2: DPMS = DPMS + 1: GOTO 14 IF A(X%) = 4 AND B(Y%) = 0 THEN GOSUB REACTION: GOTO 14 IF A(X%) = 4 AND B(Y%) = 1 THEN GOSUB REACTION4: GOTO 14 EF A(X%) = 5 AND B(Y%) = 0 THEN GOSUB REACTION4: GOTO 14 IF A(X%) = 5 AND B(Y%) = 0 THEN A(X%) = 3: B(Y%) = 2: DPMSD2 = DPMSD2 + 1: GOTO 14 14N = N+ 1 LOCATE 3, 1: PRINT "PROCESSING: N = ", N LOOP UNTIL INKEYS = CHR$(27) UU = 0 'NO. OF UNREACTED PD2(DPM)2(B-S)BR2 ZZ = 0 'NO. OF UNSUBSTITUTED PD2 SPECIES ZO = 0 'NO. OF MONOSUBSTITUTED PD2 SPECIES OO = 0 'NO. OF DISUBSTITUTED PD2 SPECIES DPM = 0 DPMD2 = 0 FOR X = 1 TO PD% IF A(X) = 0 THEN UU = UU + 1 IF A(X) = 4 THEN UU = UU + 1 IF A(X) = 5 THEN UU = UU + 1 IF A(X) = 1 THEN ZZ = ZZ + 1 IF A(X) = 2 THEN ZO = ZO + 1 IF A(X) = 3 THEN OO = OO + 1 NEXT X F O R Y = l T O D P M % IF B(Y) = 0 THEN DPM = DPM + 1 IF B(Y) = 1 THEN DPMD2 = DPMD2 + 1 N E X T Y PRINT "NO. OF UNREACTED PD2 SPECIES = ", UU PRINT "NO. OF UNSUBSTITUTED PD2 SPECIES....= ", ZZ PRINT "NO. OF MONOSUBSTITUTED PD2 SPECIES..= ", ZO PRINT "NO. OF DISUBSTITUTED PD2 SPECIES....= ", OO PRINT "NO. OF DPM = ", DPM PRINT "NO. OF DPMD2 = ", DPMD2 PRINT "NO. OF DPMS = ", DPMS PRINT "NO. OF DPMSD2 = ", DPMSD2 LOCATE 22, 1: COLOR 5: PRINT "<ESC> - STOP ", 296 COLOR 7: PRINT "<SPACE> - CONTINUE Q - QUIT P - PRINT" 15 C$ = INKEY$ IF C$ = " " THEN 12 IF C$ = "Q" THEN 16 IF C$ = "P" THEN GOSUB PRTNTSUB IF C$ = "" THEN 15 GOTO 15 16 END REACTION: RANDOMIZE TIMER RN = INT(RND * 3) + 1 UF RN = 1 THEN A(X%) IF RN = 2 THEN A(X%) IF RN = 3 THEN A(X%) B(Y%) = 2 RETURN 1:DPMSD2 = DPMSD2 + 1 2: DPMS = DPMS+ 1 2:DPMS=DPMS + 1 REACTION4: RANDOMIZE TIMER RN = INT(RND * 3) + 1 IF RN = 1 THEN A(X%) IF RN = 2 THEN A(X%) IF RN = 3 THEN A(X%) B(Y%) = 2 RETURN 3: DPMS = DPMS+ 1 2: DPMSD2 = DPMSD2 + 1 2: DPMSD2 = DPMSD2 + 1 LIGEX: IF A(X%) = 0 AND B(Y%) = 1 THEN A(X%) = 4: B(Y%) = 0: GOTO 20 IF A(X%) = 4 AND B(Y%) = 0 THEN GOSUB SUBST3: GOTO 20 IF A(X%) = 4 AND B(Y%) = 1 THEN GOSUB SUBST4: GOTO 20 IF A(X%) = 5 AND B(Y%) = 0 THEN A(X%) = 4: B(Y%) = 1: GOTO 20 IF A(X%) = 1 AND B(Y%) = 1 THEN A(X%) = 2: B(Y%) = 0: GOTO 20 IF A(X%) = 2 AND B(Y%) = 0 THEN GOSUB SUBST1: GOTO 20 IF A(X%) = 2 AND B(Y%) = 1 THEN GOSUB SUBST2: GOTO 20 IF A(X%) = 3 AND B(Y%) = 0 THEN A(X%) = 2: B(Y%) = 1: GOTO 20 20 RETURN SUBST1: RANDOMIZE TIMER RN = INT(RND * 2) + 1 IF RN = 2 THEN A(X%) = 1: B(Y%) = 1 RETURN SUBST2: RANDOMIZE TIMER RN = INT(RND * 2) + 1 IF RN = 2 THEN A(X%) = 3: B(Y%) = 0 RETURN SUBST3: RANDOMIZE TIMER 297 RN = INT(RND * 2) + 1 TF RN = 2 THEN A(X%) = 0: B(Y%) = 1 RETURN SUBST4: RANDOMIZE TIMER RN = INT(RND * 2) + 1 IF RN = 2 THEN A(X%) = 5: B(Y%) = 0 RETURN PRINTSUB: LPRTNT "NO. OF PD2(DPM)2(B-S)BR2 (INITIAL)..= ", PD% LPPJNT "NO. OF DPM-D2 (INITIAL) = ", DPM% LPRINT "NO. OF PROCESSES = ", N LPPJNT LPRINT "NO. OF UNREACTED PD2 SPECIES = ", UU LPRTNT "NO. OF UNSUBSTITUTED PD2 SPECIES....= ", ZZ LPRTNT "NO. OF MONOSUBSTITUTED PD2 SPECIES. .= ", ZO LPRINT "NO. OF DISUBSTITUTED PD2 SPECIES....= ", OO LPRTNT "NO. OF DPM = ", DPM LPRTNT "NO. OF DPMD2 = ", DPMD2 LPRTNT "NO. OF DPMS = ", DPMS LPRTNT "NO. OF DPMSD2 = ", DPMSD2 RETURN 298 Appendix IV Appendix IV - Determination of the concentration of H 2S in chloroform The concentration of H 2S in CHCI3 at R T . and 1 arm total pressure is deterrriined from knowledge of Henry's constant (K = 1.3 M arm'1)1 and vapour pressure data for C H C I 3 . 2 The dependence of vapour pressure of CHC13 on temperature is shown in Table IV. 1 and is illustrated in Fig. IV. 1. At R T . (295.15 K), the vapour pressure of CHC13 is 176 mm Hg or 0.23 arm. Thus, when the total pressure of the system is 1 arm, the partial pressure of H 2S (p(H2S)) is 0.77 arm. The corresponding concentration of H 2S in CHC13 is therefore 1.0 M (cone. = K x p(H2S) = 1.3 x 0.77). Table r v . l . Dependence of the vapour pressure of CHCI3 on temperature in the range T = 215.15 to 334.45 K T(K) p(CHClj) (mm Hg) 215.15 1 243.45 10 266.05 40 283.55 100 315.85 400 334.45 760 Fig. r v . l . Plot of ln(p(CHCl3)) versus 1/T 7 6 5 2* . O 4 X o 5 3 c 2 1 0 -0.0025 0.003 0.0035 0.004 0.0045 1/T(K) 0.005 299 Appendix IV To determine the volume of H 2S (introduced under STP conditions by injection via syringe) needed for a particular concentration in CHCI3, knowledge of the volumes of the solution (Vs) and the head space (Vh) above it are also required. The total number of moles of H 2S (ni) is equal to the number of moles of H 2S in the head space (nh) plus that in solution (ns): (TV.l) n T = nh + n8 The partial pressure of H 2S (p(H2S)) is calculated as follows: (TV.2) p(H2S) = n h R T / V h Substitution into Henry's equation gives the following: (IV.3) cone. = Kp(H2S) = n/Vt = Kn hRT/V h Finally, the volume of H 2S needed is deterrnined by application of the gas law with T = 295.15 K, p = 1 atm, and R = 0.08206 L atm. (TV.4) V = nTRT/p References 1. Fogg, P.G.T.; Young, C L . (Eds.) "Solubility data series. Vol. 32" Permagon Press, Oxford, 1988, p. 279. 2. Handbook for Chemistry and Physics. 56th ed. Chemical Rubber Co., Cleveland, OH. 1979, p. D-191. 300 Appendix V Appendix V - Kinetic data from the study of the reaction: Pd2I2(u.-S)(dpm)2 (2c) + I2 Kinetic data from conventional UV-vis spectroscopic studies P e l (M) fcl (M) T(°C) kobs (s"1) 9.39 x l O - 5 3.76 x IO"4 25 2.34 x IO"3 9.39 x IO - 5 3.76 x IO"4 25 2.35 x IO"3 9.39 x 10"5 5.63 x IO"4 25 2.27 x IO"3 9.39 x 10"5 7.51 x IO"4 25 2.18 x 10' 3 9.39 x IO'5 7.51 x IO"4 25 1.97 x IO"3 9.39 x IO"5 7.51 x IO - 4 19 1.31 x IO"3 9.39 x IO"5 7.51 x IO"4 29.5 4.05 x IO"3 9.39 x 10"5 7.51 x IO'4 35 7.87 x IO"3 Kinetic data from stopped-flow spectroscopic studies [2cl = 9 . 9 x l O - 5 M T = 19.6 °C T = 25.0 °C T = 30.1 °C T = 35.1 °C Pal kobs (s'1) kobs (s'1) k„bs (s"1) kobs (s"1) 3.9 x 10 " 4 M 0.06748 0.06932 0.06840 0.08922 0.08865 0.1090 0.1080 0.1448 0.1413 4.9 x 1 0 - 4 M 0.08804 0.09037 0.08945 0.1115 0.1140 0.1400 0.1448 0.1744 0.1751 5.9 x l O ^ M 0.1035 0.09888 0.1335 0.1320 0.1529 0.1546 7 . 9 x l O - 4 M 0.1416 0.1419 0.1913 0.1904 0.2292 0.2361 0.2905 0.3014 T=19.6 °C T2cl = 3.3 x 10"5 M [2cl = 1.6 x 10 " 5 M kobs (s"1) kobs (s"1) 7.9 x l O ^ M 0.1537 0.1530 0.1763 0.1790 301 Appendix VI - X-ray crystal structure data for PdI2(dpm(S)>0.5CH2Cl2 Experimental details A. Crystal data Empirical formula Fomula weight Crystal color, Habit Crystal dimensions Crystal system Lattice type No. of reflections used of unit cell determination (20 range) Omega scan peak width at half-height Lattice parameters Space group Z value Dcalc Fooo |i(MoKa) B . Intensity measurements C25.5oH23ClI2P2PdS 819.13 brown, plate 0.05 x 0.20 x 0.30 mm monoclinic primitive 25 (14.1 - 24.7°) 0.37° a = 9.378(2) A b= 14.231(3) A c = 21.049(2) A P = 91.60(1)° V = 2808.1(9) A3 P2i/n (#14) 4 1.937 g/cm3 1564 31.57 cm"1 302 Diffractometer Radiation Take-off angle Detector aperture Crystal to detector distance Temperature Scan type Scan rate Scan width 20max No. of reflections measured Corrections C. Structure solution and n Structure solution Refinement Function minimized Least squares weights p-factor Anomalous dispersions Rigaku AFC6S MoKa(X = 0.71069 A) graphite monochromated 6.0° 6.0 mm horizontal 6.0 mm vertical 285 mm 21 °C co-20 167min (in co) (up to 9 scans) (1.15+ 0.35 tan 9)° 55° Total: 7107 Unique: 6709 (R^ = 0.055) Lorentz-polarization Absorption (trans, factors: 0.529 - 1.000) Direct methods (SIR92) Full-matrix least-squares Era(|Fo| - |Fc|)2 1 4Fo2 ^(Fo) _ ^ (Fo 2 ) 0.000 All non-hydrogen atoms 303 No. observations (I>3o(I)) 2465 No. variables 301 Reflection/parameter ratio 8.19 Residuals. R; Rw 0.035; 0.028 Goodness of fit indicator 1.32 Max shift/error in final cycle 0.006 Maximum peak in final diff. map 0.51 e"/A3 Minimum peak in final diff. map -0.50 eV A 3 304 Appendix VI 305 Table VI.1. Atomic coordinates of PdI2(dpm(S)>0.5CH2Cl2 (11c). atom X y z 1(0 -0.17339(7) -0.10480(5) 0.25511(3) 1(2) 0.00641(7) -0.01346(5) 0.40701(3) P d © -0.02793(7) 0.04611(5) 0.29079(3) Cl(l) 0.450(2) 0.4007(9) 0.4752(6) Cl(2) 0.496(2) 0.413(2) 0.4951(9) Cl(3) 0.359(2) 0.442(2) 0.498(1) S(l) -0.0686(2) 0.1006(2) 0.1874(1) P(l) 0.1029(2) 0.1717(2) 0.3199(1) P(2) 0.0895(2) 0.1945(2) 0.1797(1) C(l) 0.0976(8) 0.2589(5) 0.2541(4) C(2) 0.0415(8) 0.2393(6) 0.3870(4) C(3) -0.0907(9) 0.2843(6) 0.3822(4) C(4) -0.1421(9) 0.3337(7) 0.4311(5) C(5) -0.066(1) 0.3411(7) 0.4870(5) C(6) 0.064(1) 0.2963(8) 0.4934(4) C(7) 0.1176(10) 0.2457(7) 0.4435(4) C(8) 0.2911(8) 0.1491(6) 0.3320(4) C(9) 0.3440(9) 0.0598(6) 0.3250(4) C(10) 0.490(1) 0.0422(7) 0.3313(4) C( l l ) 0.5820(10) 0.1147(9) 0.3441(5) C(12) 0.531(1) 0.2041(8) 0.3529(5) C(13) 0.3854(9) 0.2225(6) 0.3464(4) C(14) 0.0496(8) 0.2725(6) 0.1143(4) C(15) -0.0881(9) 0.3025(6) 0.1050(4) 306 Table VI.1. Atomic coordinates of PdI2(dpm(S)>0.5CH2Cl2 (11c) (cont). atom X y z C(16) -0.1245(10) 0.3597(7) 0.0539(5) C(17) -0.023(1) 0.3837(7) 0.0104(4) C(18) 0.116(1) 0.3552(7) 0.0198(5) C(19) 0.1536(9) 0.2982(7) 0.0716(4) C(20) 0.2614(8) 0.1411(6) 0.1694(4) C(21) 0.2693(9) 0.0487(6) 0.1513(4) C(22) 0.401(1) 0.0070(7) 0.1426(5) C(23) 0.5237(10) 0.0569(8) 0.1538(4) C(24) 0.5169(9) 0.1484(7) 0.1709(4) C(25) 0.3870(9) 0.1916(6) 0.1787(4) C(26) 0.573(4) 0.503(3) 0.482(2) 307 Appendix VI Table VI.2. Bond Lengths(A) in the complex atom atom distance 1(1) P d © 2.6419(9) P d © S © 2.331(2) SO) P(2) 2.007(3) p © C(2) 1.815(8) P(2) C © 1.814(8) P(2) C(20) 1.802(8) C(2) C(7) 1.37(1) C(4) C(5) 1.36(1) C(6) C(7) 1.38(1) C(8) C(13) 1.40(1) C(10) C( l l ) 1.37(1) C(12) C(13) 1.40(1) C(14) C(19) 1.39(1) C(16) C(17) 1.38(1) C(18) C(19) 1.40(1) C(20) C(25) 1.39(1) C(22) C(23) 1.37(1) C(24) C(25) 1.38(1) PdI2(dpm(S)>0.5CH2Cl2 (11c). atom atom distance 1(2) P d © 2.6006(9) P d © P © 2.243(2) P © C © 1.860(8) P © C(8) 1.805(8) P(2) C(14) 1.799(8) C(2) C(3) 1.40(1) C(3) C(4) 1.35(1) C(5) C(6) 1.38(1) C(8) C(9) 1.37(1) C(9) C(10) 1.39(1) C(l l ) C(12) 1.37(1) C(14) C(15) 1.370(10) C(15) C(16) 1.38(1) C(17) C(18) 1.37(1) C(20) C(21) 1.37(1) C(21) C(22) 1.38(1) C(23) C(24) 1.35(1) 308 Appendix VI Table VI.3. Bond Angles(°) in the complex PdI2(dpm(S)>0.5CH2Cl2 (11c). atom atom atom angle atom atom atom angle 1(1) Pd(l) 1(2) 92.97(3) 1(0 Pd(l) S(l) 86.28(6) IO) Pd(l) P(l) 177.90(7) 1(2) Pd(l) S(l) 177.64(6) 1(2) Pd(l) P(l) 87.14(6) S(l) Pd(l) P(l) 93.69(8) Pd(l) S(l) P(2) 101.3(1) Pd(l) P(l) C(l) 109.0(3) P d © P(l) C(2) 116.9(3) Pd(l) P(l) C(8) 115.0(3) CO) P(l) C(2) 102.9(4) C(l) P(l) C(8) 103.3(4) C(2) P(l) C(8) 108.3(4) S(l) P(2), C(l) 106.3(3) S(l) P(2) C(14) 109.6(3) S(l) P(2) C(20) 113.2(3) C(l) P(2) C(14) 110.6(4) C(l) P(2) C(20) 107.5(4) C(14) P(2) C(20) 109.6(4) P(l) C(l) P(2) 107.8(4) P(l) C(2) C(3) 119.1(6) P(l) C(2) C(7) 122.7(7) C(3) C(2) C(7) 118.1(8) C(2) C(3) C(4) 121.3(8) C(3) C(4) C(5) 120.7(9) C(4) C(5) C(6) 119.2(9) C(5) C(6) C(7) 120.5(9) C(2) C(7) C(6) 120.2(9) P(l) C(8) C(9) 120.3(7) P(l) C(8) C(13) 120.5(7) C(9) C(8) C(13) 119.2(7) C(8) C(9) C(10) 120.8(8) C(9) C(10) C(l l ) 119.8(10) C(10) C(l l ) C(12) 120.4(9) C( l l ) C(12) C(13) 120.2(9) C(8) C(13) C(12) 119.6(9) P(2) C(14) C(15) 118.5(6) P(2) C(14) C(19) 121.4(6) C(15) C(14) C(19) 120.0(8) C(14) C(15) C(16) 120.4(8) C(15) C(16) C(17) 120.0(9) C(16) C(17) C(18) 120.0(9) C(17) C(18) C(19) 120.1(9) C(14) C(19) C(18) 119.3(8) P(2) C(20) C(21) 119.5(7) P(2) C(20) C(25) 121.5(7) C(21) C(20) C(25) 118.9(8) C(20) C(21) C(22) 120.2(8) 309 Appendix Table VL3. Bond Angles(°) in the complex PdI2(dpm(S)>0.5CH2Cl2 (11c) (cont). atom atom atom angle atom atom atom angle C(21) C(22) C(23) 120.3(9) C(22) C(23) C(24) 119.9(9) C(23) C(24) C(25) 120.6(9) C(20) C(25) C(24) 120.1(8) 310 Appendix VII Appendix VII - Synthesis and characterization of [Pd(dpm(S))2]Cl2, fra«s-bis-r|-P,S-(bis(diphenyIphosphino)methane monosulfide)palladium(II) chloride Synthesis Dpm(S) (0.27 g, 0.65 mmol) was dissolved in 5 mL CH 2C1 2, and a solution of PdCl2(PhCN)2 (0.050 g, 0.15 mmol) in 5 mL CH 2C1 2 was added. The resulting orange-red solution was stirred at R.T. for 1 h before it was reduced in volume to ~5 mL. Et 2 0 (20 mL) was then added precipitating an orange solid which was filtered off, washed with Et 2 0 (2x10 mL), and dried in vacuo; yield: 0.12 g (91%). ! H NMR (20 °C, DMSO-d 6): 8 7.0 - 8.0 m (40H, Ph), 8 5.66 pt (4H, CH 2). ^Pf/H) NMR (20 °C, DMSO-d 6): 8 49.6, 37.6 (AB pattern, J P P = 23 Hz). A yellow crystal, obtained by diffusion of hexanes (10 mL) into a CH 2C1 2 (6 mL) solution of 10 mg of the complex, was analyzed using X-ray crystallography. X-ray Crystal Structure Data Experimental details A. Crystal data Empirical formula Fomula weight Crystal color, Habit Crystal dimensions Crystal system Lattice type C52H4gCl6P4PdS2 1180.08 yellow, prism 0.15 x 0.25 x 0.35 mm triclinic primitive 311 No. of reflections used of unit cell determination (26 range) Omega scan peak width at half-height Lattice parameters Space group Z value D c a i c Fooo p.(MoKa) B. Intensity measurements Diffractometer Radiation Take-off angle Detector aperture Crystal to detector distance Temperature Scan type Scan rate 25 (27.8 - 32.9°) 0.36° a= 11.612(1) A b= 11.813(1) A c = 10.3623(7) A a = 97.898(6)° p = 96.011(7)° y = 80.174(8)° V = 1382.4(2) A3 Pl"(#2) 1 1.417 g/cm3 600 8.50 cm"1 Rigaku AFC6S MoKa(X = 0.71069 A) graphite monochromated 6.0° 6.0 mm horizontal 6.0 mm vertical 285 mm 21 °C ©-20 327min (in ©) (up to 9 scans) 312 Scan width 29„ No. of reflections measured Corrections (1.26+ 0.35 tan 9)° 60° Total: 8441 Unique: 8069 (R^ = 0.028) Lorentz-polarization Absorption (trans, factors: 0.967 - 1.000) Decay (6.51% decline) C. Structure solution and refinement Structure solution Refinement Function minimized Least squares weights p-factor Anomalous dispersions No. observations (I>3o(I)) No. variables Reflection/parameter ratio Residuals: R; Rw Goodness of fit indicator Direct methods (SIR92) Full-matrix least-squares Sco(|Fo| - |Fc|)2 1 4Fo2 o2(Fo) "cf^Fo 2 ) 0.000 All non-hydrogen atoms 4663 295 15.81 0.036; 0.033 1.77 Max shift/error in final cycle 0.0003 Maximum peak in final diff. map 0.61 e"/A3 Minimum peak in final diff. map -0.49 eV A3 313 Appendix VII Fig. VTJ.1. ORTEP drawing of [Pd(dpm(S))2]Cl2. Hydrogen atoms are omitted. 314 Fig. VTI.2. Stereoview of [Pd(dpm(S))2]Cl2. Hydrogen atoms are omitted. 315 Appendix Fig. VTJ.4. PLUTO drawing showing 1/2 of [Pd(dpm(S))2]Cl2 with interactions with a dichloromethane solvate molecule. H15 H10 316 Table VH.1. Atomic coordinates of [Pd(dpm(S))2]Cl2*0.5CH2Cl2. atom X Pd(l) 0.50000 c i © 0.59790(9) Cl(2) 0.8889(2) Cl(3) 0.7815(2) SO) 0.37946(8) P(0 0.54085(7) P(2) 0.33265(7) CO) 0.4639(3) C(2) 0.6963(3) C(3) 0.7436(3) C(4) 0.8627(4) C(5) 0.9352(3) C(6) 0.8891(4) C(7) 0.7695(3) C(8) 0.4989(3) C(9) 0.4493(3) C(10) 0.4198(4) C(H) 0.4397(4) C(12) 0.4867(4) C(13) 0.5163(3) C(14) 0.2606(3) C(15) 0.2830(3) C(16) 0.2263(3) C(17) 0.1487(3) y z 0.50000 0.50000 0.04112(7) 0.70599(8) 0.1085(2) 0.9711(2) 0.3319(2) 0.9288(3) 0.44301(7) 0.63642(8) 0.31222(6) 0.39609(7) 0.30243(6) 0.52318(8) 0.2152(2) 0.4669(3) 0.2568(2) 0.4182(3) 0.2096(3) 0.5308(4) 0.1750(4) 0.5499(5) 0.1876(4) 0.4591(5) 0.2353(4) 0.3476(4) 0.2701(3) 0.3264(3) 0.2938(2) 0.2223(3) 0.3876(3) 0.1569(3) 0.3717(3) 0.0239(3) 0.2632(4) -0.0448(3) 0.1699(3) 0.0189(3) 0.1836(3) 0.1520(3) 0.2274(2) 0.6217(3) 0.1089(3) 0.6177(4) 0.0562(3) 0.6973(4) 0.1189(3) 0.7794(4) 317 Table VII.1. Atomic coordinates of [Pd(dpm(S))2]Cl2»0.5CH2Cl2 (cont). atom X y z C(18) 0.1252(3) 0.2369(3) 0.7831(4) C(19) 0.1816(3) 0.2911(3) 0.7051(3) C(20) 0.2360(3) 0.3385(3) 0.3825(3) C(21) 0.1796(4) 0.4508(3) 0.3751(4) C(22) 0.1063(5) 0.4757(4) 0.2650(6) C(23) 0.0912(4) 0.3909(5) 0.1651(5) C(24) 0.1473(4) 0.2805(4) 0.1710(4) C(25) 0.2195(3) 0.2530(3) 0.2802(3) C(26) 0.7631(5) 0.1971(6) 0.9356(6) 318 Appendix Table VH.2. Bond Lengths (A) in the complex [Pd(dpm(S))2]Cl2-0.5CH2Cl2. atom atom distance atom atom distance Pd(l) S(l) 2.3256(8) Pd(l) P(l) 2.3229(7) Cl(2) C(26) 1.684(6) Cl(3) C(26) 1.654(7) S(l) P(2) 2.017(1) P(l) C(l) 1.835(3) P(l) C(2) 1.814(3) P(l) C(8) 1.809(3) P(2) C(l) 1.791(3) P(2) C(14) 1.791(3) P(2) C(20) 1.797(3) C(2) C(3) 1.384(4) C(2) C(7) 1.382(4) C(3) C(4) 1.375(5) C(4) C(5) 1.365(6) C(5) C(6) 1.373(6) C(6) C(7) 1.383(5) C(8) C(9) 1.387(4) C(8) C(13) 1.395(4) C(9) C(10) 1.380(4) C(10) C(l l ) 1.373(5) C(H) C(12) 1.367(5) C(12) C(13) 1.379(4) C(14) C(15) 1.376(4) C(14) C(19) 1.380(4) C(15) C(16) 1.377(4) C(16) C(17) 1.357(5) C(17) C(18) 1.371(5) C(18) C(19) 1.374(4) C(20) C(21) 1.384(5) C(20) C(25) 1.382(4) C(21) C(22) 1.384(6) C(22) C(23) 1.357(7) C(23) C(24) 1.361(6) C(24) C(25) 1.379(5) 319 Appendix VII Table VTI.3. Bond Angles (°) in the complex [Pd(dpm(S))2]Cl2'0.5CH2Cl2. atom atom atom angle atom atom atom angle S(l) Pd(l) S(l)* 180.0 S(l) Pd(l) P(l) 91.21(3) SO) Pd(l) P(l)* 88.79(3) P(l) Pd(l) P(l)* 180.0 Pd(l) S(l) P(2) 99.84(4) Pd(l) P(l) C(l) 110.46(9) Pd(l) P(l) C(2) 110.71(9) Pd(l) P(l) C(8) 115.52(10) C(l) P(l) C(2) 106.8(1) C(l) P(l) C(8) 105.7(1) C(2) P(l) C(8) 107.1(1) S(l) P(2) C(l) 107.48(10) S(l) P(2) C(14) 107.60(10) S(l) P(2) C(20) 112.8(1) C(l) P(2) C(14) 111.6(1) C(l) P(2) C(20) 108.1(1) C(14) P(2) C(20) 109.3(1) P(l) C(l) P(2) 106.1(1) P(l) C(2) C(3) 120.7(2) P(l) C(2) C(7) 119.5(3) C(3) C(2) C(7) 119.6(3) C(2) C(3) C(4) 119.9(3) C(3) C(4) C(5) 120.6(4) C(4) C(5) C(6) 119.9(4) C(5) C(6) C(7) 120.4(4) C(2) C(7) C(6) 119.6(3) P(l) C(8) C(9) 121.2(2) P(l) C(8) C(13) 119.8(2) C(9) C(8) C(13) 119.0(3) C(8) C(9) C(10) 120.2(3) C(9) C(10) C( l l ) 120.3(3) C(10) C(l l ) C(12) 120.0(3) C( l l ) C(12) C(13) 120.6(3) C(8) C(13) C(12) 119.8(3) P(2) C(14) C(15) 121.8(2) P(2) C(14) C(19) 118.7(2) C(15) C(14) C(19) 119.4(3) C(14) C(15) C(16) 119.2(3) C(15) C(16) C(17) 121.4(3) C(16) C(17) C(18) 119.7(3) C(17) C(18) C(19) 119.8(3) C(14) C(19) C(18) 120.5(3) P(2) C(20) C(21) 120.7(3) P(2) C(20) C(25) 119.3(3) C(21) C(20) C(25) 120.0(3) C(20) C(21) C(22) 119.2(4) C(21) C(22) C(23) 120.2(4) C(22) C(23) C(24) 121.0(4) 320 Appendix Table VH.3. Bond Angles (°) in the complex [Pd(dpm(S))2]Ci2*0.5CH2Ci2 (cont.). atom atom atom angle atom atom atom angle C(23) C(24) C(25) 120.1(4) C(20) C(25) C(24) 119.6(4) Cl(2) C(26) Cl(3) 113.4(3) * Symmetry operation: 1-x, 1-y, 1-z. 321 Appendix VII Table VTI.4. C-H»«C1 interactions* in the complex [Pd(dpm(S))2]Cl2-0.5CH2Cl2. A H B A-H CO) HO) Cl(l)a 0.98 CO) H(2) CIO) 0-98 C(3) H(3) CIO) 0-98 C(26) H(23) CIO) 0.98 H . B A...B A - H . .B 2.50 3.446(3) 162 2.57 3.505(3) 160 2.75 3.592(4) 144 2.53 3.432(5) 153 * Symmetry operation: (a) 1-x, -y, l-z. 322 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0059521/manifest

Comment

Related Items