UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Activation of carbon monoxide by ruthenium carbonyl complexes in solution Plackett, David Victor 1977

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

Item Metadata

Download

Media
831-UBC_1977_A1 P43.pdf [ 11.77MB ]
Metadata
JSON: 831-1.0059302.json
JSON-LD: 831-1.0059302-ld.json
RDF/XML (Pretty): 831-1.0059302-rdf.xml
RDF/JSON: 831-1.0059302-rdf.json
Turtle: 831-1.0059302-turtle.txt
N-Triples: 831-1.0059302-rdf-ntriples.txt
Original Record: 831-1.0059302-source.json
Full Text
831-1.0059302-fulltext.txt
Citation
831-1.0059302.ris

Full Text

ACTIVATION OF CARBON MONOXIDE BY RUTHENIUM CARBONYL COMPLEXES IN SOLUTION BY DAVID VICTOR PLACKETT B.Sc. (Hons.) University of Sussex, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Chemistry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1977 © David Victor Plackett, 1977 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of Brit ish 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 representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of Brit ish Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 i ABSTRACT The thesis describes some aspects of the aqueous solution chemistry of chlororuthenate(III) and chlorocarbonylruthenate(III or II) complexes including their reactivity toward carbon monoxide. This led to the synthesis and characterisation of a polymeric complex [HRu(CO),j]n, which i s formally a Ru(I) derivative. The use of these ruthenium complexes for activating CO catalytically was studied, especially for the carbonylation of amines. The [HRu(CO) 3J N polymer was characterised by microanalysis, infra-red and high-field "''H n.m.r. , and i t s chemistry In donor solvents in which i t was soluble. The polymer may be formed by reductive carbonylation of chloro complexes of Ru"^ "^, Ru"'"'^  (CO), Ru I I(CO), Ru I I(C0)2 and Ru'^CO).^ and stoichiometric evidence suggests processes such as: III + 2Ru + 10 CO + 3H20 >2HRu(C0) 3 + 4C02 + 6H (1) ITT + 2Ru (CO) + 8C0 + 3H20 >2HRu(CO)3 + 4C02 + 6H (2) 2RuI:E(C0) + 7C0 + 3H20 > 2HRu(C0)3 + 3C02 + 4H + (3) 2Ru I i :(C0) 2 + 5C0 + 3H20 >2HRu(C0) 3 + 3C02 + 4H + (4) Increasing acidity and chloride concentration inhibit the reductive carbonylation process, which l i k e l y requires simultaneous co-ordination of cis CO and OH ligands. Reactions (1) - (4) are i i accompanied by formation of smaller amounts of low valent ruthenium complexes inc l u d i n g Ru 3(CO) l 2, which could r e s u l t from a reductive carbonylation process such as Ru I i :(CO) 3 + 2C0 + H 20 5> Ru°(CO) 4 + C0 2 + 2H + (5) or v i a 'combination' of Ru"*" and Ru * species. Evidence i s presented for reaction (5) s t a r t i n g with CsRu(CO) 3C£ 3. Reductive carbonylation 2-of Ru(CO) 2C£^ (reaction (4)) shows a u t o c a t a l y t i c gas uptake p l o t s , i n d i c a t i n g c a t a l y s i s of the reaction v i a a Ru(0) or Ru(I) intermediate. The k i n e t i c s for the carbonylation of p i p e r i d i n e to N-formyl pi p e r i d i n e catalysed by each of the complexes [HRu(CO) 3] n, [Ru(CO) 2(OAc)(pip)] 2, and CsRu(CO) 3C£ 3, have been studied under mild conditions. Mechanisms are proposed to explain the observed k i n e t i c s and i n each case a t r i c a r b o n y l monomeric species appears to be the a c t i v e c a t a l y s t . A CO i n s e r t i o n reaction i n a Ru(CO) (pip) i n t e r -j X mediate must be involved. A scheme such as (6) e.g. (C0) 9Ru —N( )> g> (C0) ?Ru-C0-N^ *) =3* (CO)9Ru+0HC-r/ *) (6) CO H H - , . (or pipCO) requires a hydride s h i f t , l i k e l y metal activated. A l t e r n a t i v e l y , p i p e r i d i n e could behave as a proton acceptor with the reaction proceeding v i a a carbamoyl intermediate (reaction (7)). (C0) 9Ru N / / P 1 P > (C0)„Ru—N(~> > ( C 0 ) o R u - C 0 — < > z I / — I 1 , (7) CO H CO p i p H + 0HC-N<^~) + (CO) 2Ru(pip) (or pipCO) i i i Both [HRu(CO>3]n and CsRu(C0)3C£3 carbonylate piperidine in a stoichiometric reaction in the absence of CO, and in the case of the cesium salt evidence suggests the following reactions: CsRu(CO) 3C£ 3 + 2 pip > CsC£ + pipH+C£~ + Ru(C0)3C£(pip~) (8) Ru(CO)3C£(pip~) + pip > Ru(CO)2C£(pip~) + pip(CO) (9) (or pipH +) (or pip) Only secondary amines were carbonylated effectively. Attempts to isolate and characterise ruthenium complexes via the reactions of [HRu(CO)3Jn or CsRu(CO)3C£3 with piperidine proved frustrating, although one complex isolated from the polymer reaction i s thought to be H 2Ru2(CO)^(pip) 3, and an oxygenated solution of [HRu(CO) 3] n in piperidine yielded a complex which analysed well for [HRu(CO) 9(pip)] 9•0 iv TABLE OF CONTENTS Page ABSTRACT i TABLE OF CONTENTS iv LIST OF TABLES Vl ' i i LIST OF FIGURES x i i ABBREVIATIONS x ix ACKNOWLEDGEMENTS XXi Chapter 1. THE USES OF RUTHENIUM CARBONYLS IN HOMOGENEOUS CATALYSIS . 1 1.1. General Introduction 1 1.2. Hydrogenation 4 1.3. Hydroformylation 10 1.4. Other miscellaneous reactions 12 1.5. Amine carbonylation 14 Chapter 2. APPARATUS AND EXPERIMENTAL PROCEDURE 23 2.1. Materials 23 2.1.1. Ruthenium compounds 23 2.1.2. Gases 25 2.1.3. Substrates and other materials 25 2.2. Apparatus for constant pressure gas-uptake measure-ments 29 2.3. Procedure for a typical gas-uptake experiment 31 V Page 2.4. Gas s o l u b i l i t y measurements 34 2.5. Spectrophotometry k i n e t i c measurements 37 2.6. Analysis of gaseous reaction products 37 2.7. Analysis of l i q u i d and s o l i d r eaction products 38 2.8. Analysis of chl o r i d e content i n s o l i d s 38 2.9. Instrumentation 39 Chapter 3. SYNTHESIS AND CHARACTERISATION OF [HRu(CO) 3] n 41 3.1. The synthesis of [HRu(CO) 3] n 41 3.2. Characterisation of [HRu(CO) 3l n 42 3.3. The f i n a l r eaction s o l u t i o n 51 3.4. Benzene-soluble products 63 3.5 [HRu(CO) 3] n s o l u t i o n chemistry i n pyridine 70 3.6 Chemistry of [HRu(C0) 3] n i n other solvents 85 Chapter 4. AN INVESTIGATION OF THE SYNTHETIC ROUTE TO [HRu(C0) 3] n -THE INITIAL STAGES 103 4.1. Introduction 103 4.2. Preparation of Ru(III) compounds and some comments on the "blue ruthenium(II)" solutions 104 4.3. U.V./visible and i . r . studies upon the carbonylation of "RuC£ 3•3H 20 M 109 4.4. Aqueous s o l u t i o n chemistry of ruthenium(III) chloro species 113 vi Page 4.5. Carbonylation of ruthenium(III) species i n aqueous sol u t i o n 125 Chapter 5. AQUEOUS SOLUTION CHEMISTRY OF CHLOROCARBONYLS OF RUTHENIUM(II) AND RUTHENIUM(III) 139 5.1. Introduction 139 5.2. Preparation of Ru 1 1 1(CO) species 139 5.3. Aqueous sol u t i o n chemistry and carbonylation of Cs 2Ru(CO)CJl 5 143 5.4. Preparation of Cs2Ru(CO) (H 20)C£^ 150 5.5. Aqueous solution chemistry and carbonylation of Cs 2Ru(CO)C£ 4(H 20) 151 5.6. Preparation of Cs 2Ru(CO) 2C£ 4 154 5.7. Aqueous s o l u t i o n chemistry and carbonylation of Cs 2Ru(CO) 2C£ 4 154 5.8. Preparation of CsRu(C0) 3C£ 3 181 5.9. Aqueous sol u t i o n chemistry and carbonylation of CsRu(CO) 3C£ 3 182 Chapter 6. CARBONYLATION OF AMINES USING [HRu(C0) 3l n 200 6.1. K i n e t i c s and mechanism of p i p e r i d i n e carbonylation i n the presence of [HRu(C0) 3] n 200 6.2. The addition of phosphines to the amine systems ... 215 6.3. Carbonylation of other amines using [HRu(C0) 3] n ... 221 6.4. The [Ru(C0) 2(0Ac)] n system 226 v i i Page Chapter 7. AMINE CARBONYLATION USING OTHER RUTHENIUM CARBONYLS 234 7.1. Introduction 234 7.2. The carbonylation of piperidine using CsRu(C0)3CJl3 235 7.3. Ruthenium carbonyl species from the CsRu(C0)3C&3/ piperidine/CO system 241 7.4. Kinetics and mechanism of piperidine carbonylation in the presence of CsRu(C0) 3C£ 3 245 7.5. Synthesis of [RuCCO^CJ^^ a n d a piperidine derivative 259 7.6. The carbonylation of piperazine using Ru^CCO).^ 264 Chapter 8. SOME RELATED RUTHENIUM CARBONYLPHOSPHINE CHEMISTRY.. 274 8.1. Introduction 274 8.2. Reactions of [HRu(C0)„] with PPh„ 277 3 n 3 Chapter 9. GENERAL CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK 280 9.1. General conclusions 280 9.2. Recommendations for future work 287 REFERENCES 289 vi i i LIST OF TABLES Page Chapter 1 I Chronology of metal carbonyl syntheses 3 II O l e f i n disproportionation a c t i v i t y of Mo and Ru c a t a l y s t s 15 Chapter 2 III Uptake of carbon dioxide by soda-lime absorbent 35 Chapter 3 IV I.R. data for chlorocarbonyl complexes of ruthenium 52 V Possible formulae and analyses for Compound(I) 56 VI Possible formulae and analyses for Compound (IV) 61 VII Hydrido and deutero-carbonyls of ruthenium 66 VIII [HRu(CO) 3] n i n pyridine - r e s u l t s of gas evolution studies. 75 IX Microanalyses of samples of the green s o l i d and possible formulae 84 Chapter 4 X K i n e t i c data for aquation of hexachlororuthenate(III) 117 XI K i n e t i c data for carbonylation of ruthenium(III) i n aqueous s o l u t i o n 128 XII Stoichiometries and i n i t i a l rates of carbonylation f or futhenium(III) species inc l u d i n g data i n the presence of soda lime 131 ix Chapter 5 Page XIII Spectroscopic data from 5M.HC£ solutions of Cs 2Ru(CO)C£ 5 142 XIV CO uptake stoichiometrics f o r aqueous solutions of Cs 2Ru(CO) 2C£ 4 158 XV E f f e c t of varying [Ru] on the maximum rate of carbonylation of Cs 2Ru(CO) 2C£^ i n aqueous so l u t i o n 164 XVI E f f e c t of varying [CO] on the maximum rate of carbonylation of Cs 2Ru(CO) 2C£^ i n aqueous so l u t i o n 167 XVII E f f e c t of varying pH and [CH ] on the carbonylation of Cs 2Ru(CO) 2C&^ i n aqueous s o l u t i o n 172 XVIII Data from the carbonylation of Cs 2Ru(CO)^Cl^ i n the presence of excess [CI ] 177 XIX V a r i a t i o n of pH and [C£ ] i n an aqueous s o l u t i o n of CsRu(CO) 3C£ 3 at 20°C i n a i r 183 XX E f f e c t of varying [Ru] on the i n i t i a l rate of carbonylation of CsRu(CO) 3C£ 3 190 XXI Further data from the carbonylation of CsRu(CO) 3C£ 3 i n aqueous s o l u t i o n 191 Chapter 6 XXII Temperature depend ence f o r [HRu(C0)„] catalysed 3 n carbonylation of p i p e r i d i n e 206 X Page XXIII Carbonylation of p i p e r i d i n e using [HRu(CO) 3] n or [DRu(CO) 3] n i n the presence of various added reagents 212 XXIV E f f e c t of added phosphine on n-butylamine carbonylation 216 XXV E f f e c t of added PPh 3 on the [HRu(CO) 3] n catalysed carbonylation of p i p e r i d i n e 218 XXVI Carbonylation of various amines using [HRu(CO) 3] n ... 222 XXVII Carbonylation of amines using [Ru(CO) 2 (OAc) ] or [Ru(CO) 2(OAc)(pip).] 2 229 Chapter 7 XXVIII Possible formulae f o r the compound (1) from the CsRu(CO) 3C£ 3/piperidine/CO system 244 XXIX E f f e c t of varying [Ru] upon the maximum rate of carbonylation of p i p e r i d i n e i n the presence of CsRu(CO) 3C£ 3 246 XXX E f f e c t of varying [CO] upon the maximum rate of carbonylation of p i p e r i d i n e i n the presence of CsRu(CO) 3C£ 3 248 XXXI E f f e c t of varying [piperidine] on the maximum rate of carbonylation of p i p e r i d i n e i n the presence of CsRu(CO) 3C£ 3 251 XXXII E f f e c t of temperature v a r i a t i o n on the maximum rate of carbonylation of p i p e r i d i n e i n the presence of CsRu(CO)^CH^ 254 xi XXXIII A comparison of various piperdine carbonylation catalysts 257 XXXIV Pyridine derivatives of [Ru(CO) 3C£ 2l 2 261 XXXV Thin-layer chromatography results from an experiment with Ru 3(CO)^ 2 and piperazine 267 XXXVI Effect of varying [Ru] upon the i n i t i a l rate of carbonylation of piperazine in the presence of Ru 3(CO)^ 2 269 XXXVII Carbonylation of piperazine derivatives 272 Chapter 9 XXXVIII Gas uptake stoichiometries in the carbonylation of Ru"1"1 and Ru"*""*""1  species in aqueous solutions 286 x i i LIST OF FIGURES Figure Page Chapter 2 1. Infra-red spectrum (nujol mull) of [Ru(C0)2(OAc)(pip)] 2 i n the carbonyl st r e t c h i n g region 26 2. Infra-red spectrum (nujol mull) of [Ru(CO) 2(OAc)(pip)] 2 i n the region 500-1400 cm ^ 27 3. 1H n.m.r. spectrum of [Ru(C0) 2 (OAc) (pip) ] 2 i n C^D^ 28 4. Apparatus for constant pressure gas-uptake or evolution measurements 30 5. C0 2 uptake by soda lime 36 Chapter 3 6. Infra-red spectrum (KBr disc) of [HRu(C0) 3] n i n the carbonyl stretching region 44 7. So l i d - s t a t e v i s i b l e r e f l e c t a n c e spectrum of [HRu(C0) 3] n 46 8. n.m.r. spectrum of [HRu(C0) 3] n i n methyldiphenyl-phosphine 48 9. Infra-red spectrum (nujol mull) of p r e c i p i t a t e (I) i n the carbonyl stretching region 54 10. Infra-red spectrum (nujol mull) of p r e c i p i t a t e (III) i n the carbonyl stretching region 55 11. Infra-red spectrum (nujol mull) of p r e c i p i t a t e (II) i n the carbonyl stretching region 58 x i i i Page 12. Infra-red spectrum (nujol mull) of p r e c i p i t a t e (IV) i n the carbonyl stretching region 59 13. Infra-red spectrum (nujol mull) of p r e c i p i t a t e (IV) i n the region 250-950 cm 1 62 14. Solution i n f r a - r e d spectrum of compound (B) i n n-hexane 67 15. Solution i n f r a - r e d spectrum of compound (C) i n dichloromethane 68 16. Gas evolution experiments, [HRu(C0) 3] n i n pyridine 74 17. Gas uptake by [HRu(C0) 3] n i n pyridine under oxygen 78 18. V i s i b l e spectroscopic changes i n a pyridine s o l u t i o n of [HRu(C0) 3] n under oxygen 79 19. Vapour pressure osmometer c a l i b r a t i o n for molecular weight determinations 83 20. Infra-red spectrum (nujol mull) i n the carbonyl stretching region of the product derived from [HRu(C0) 3] n and p i p e r i d i n e 87 21. Infra-red spectrum (nujol mull) i n the carbonyl stretching region of the product derived from [HRu(C0) 3] n and pi p e r i d i n e (by an improved i s o l a t i o n technique) 88 22. n.m.r. spectrum of the product derived from [HRu(C0) 3] n and p i p e r i d i n e i n CDC£ 3 91 23. n.m.r. spectrum of p i p e r i d i n e hydrochloride i n CDC&3 .. 92 24. Oxygen uptake by [HRu(C0) 3] n i n pi p e r i d i n e 93 x iv Page 25. V i s i b l e spectroscopic changes i n a p i p e r i d i n e s o l u t i o n of [HRu(CO) 3] n under oxygen 95 26. Infra-red spectrum (nujol mull) i n the carbonyl stretching region of the product derived from [HRu(CO) 3] n i n p i p e r i d i n e under oxygen 96 27. ''"H n.m.r. spectrum of the product derived from [HRu(CO) 3] n i n p i p e r i d i n e under oxygen ( i n CDCl^) 97 28. Solution i n f r a - r e d spectra of [HRu(CO) 3] n i n DMA 99 29. V i s i b l e spectrum of [HRu(CO) q] T i i n DMA 101 Chapter 4 30. V i s i b l e spectroscopic changes i n an aqueous s o l u t i o n of K„RuC£, 3 6 31. CO uptake by an aqueous s o l u t i o n of K 0RuC£ £ at [Ru]=2.2xl0~^M., 80°C. and 1 atm. t o t a l pressure 127 32. CO uptake by an aqueous s o l u t i o n of K 3RuC£g at [Ru] = 10.6x10 M., 80°C. and 1 atm. t o t a l pressure 129 33. [Ru] dependence of the i n i t i a l rate of CO uptake by K 3RuC£g i n aqueous s o l u t i o n 130 34. CO uptake by an aqueous s o l u t i o n of K 3RuC£^ at [Ru] = 14.1xl0~ 2M. , 80°C. and 560 mm Hg t o t a l pressure 133 35. CO uptake by an aqueous s o l u t i o n of K 3RuC£^ at [Ru] = _2 3.13x10 M., 80°C. and 1 atm. t o t a l pressure i n the presence of soda-lime 1^5 XV Page Chapter 5 36. V i s i b l e spectroscopic changes i n an aqueous s o l u t i o n of Cs 2Ru(CO)CJ£ 5 at 20°C 144 37. Plots of -log 1 0(A t-A o o), (A t-Aj and ( A ^ A J - 1 against time for an aqueous s o l u t i o n of Cs 2Ru(CO)C£^ 146 38. V i s i b l e spectroscopic changes i n an aqueous s o l u t i o n of Cs 2Ru(CO)C£ 5 at 80°C 147 39. CO uptake by an aqueous s o l u t i o n of Cs 2Ru(CO)C£^ at -2 [Ru] = 14.1x10 M. , 80°C. and 1 atm. t o t a l pressure 149 40. V i s i b l e spectroscopic changes i n an aqueous s o l u t i o n of Cs 2Ru(C0)(H 20)C£ 4 153 41. V i s i b l e spectroscopic changes i n an aqueous s o l u t i o n of Cs 2Ru(CO) 2C£ 4 156 42. CO uptake by an aqueous s o l u t i o n of Cs 2Ru(CO) 2C£ 4 at [Ru] = 14.1x10 M. , 80°C, and 1 atm. t o t a l pressure 160 43. [Ru] dependence of the maximum rate of CO uptake by Cs 2Ru(CO) 2C£ 4 i n aqueous s o l u t i o n 165 2 44. [Ru/2] dependence of the maximum rate of CO uptake by Cs 2Ru(CO) 2C£ 4 i n aqueous s o l u t i o n 166 45. [CO] dependence of the maximum rate of CO uptake by Cs 2Ru(CO) 2C£^ i n aqueous s o l u t i o n 168 46. CO uptake by an aqueous s o l u t i o n of Cs 2Ru(CO) 2C£ 4 at -2 [Ru] = 14.1x10 M., 80°C. and 660 mm Hg t o t a l pressure 170 47. E f f e c t of temperature v a r i a t i o n on CO uptake by Cs 2Ru(CO) 2C£ 4 _2 at [Ru] = 14.0x10 M and 1 atm. t o t a l pressure 171 xvi Page 48. CO uptake by aqueous solutions of Cs2Ru(CO)2C£ 4 at [Ru] = _2 14.0 x 10 M., 80°C. and 1 atm. t o t a l pressure i n the presence of (A) pH=7.0 buffer, (C) 3.3 M.KC£ and (D) 11.0 M.CsC£ 173 49. Infra-red spectra i n n-hexane of products from the [HRu(C0) 3] n synthesis and a number of experiments i n v o l v i n g carbonylation of Cs2Ru(C0)2C£ 4 180 50. CO uptake by aqueous solutions of CsRu(C0).jC£3 under a v a r i e t y of conditions 188 51. CO uptake by an aqueous s o l u t i o n of CsRu(CO).^^ at [Ru] = 14.3xl0~ 2M. , 80°C. and 1 atm. t o t a l pressure 192 52. CO uptake by aqueous solutions of CsRuCCO).^^ a t [Ru] = _2 1.30x10 M. , 6 0 ° C , 1 atm. and % atm. t o t a l pressure 194 53. Infra-red spectra i n n-hexane of products from a number of experiments i n v o l v i n g carbonylation of CsRu(CO) 3C£ 3 196 Chapter 6 54. CO uptake p l o t for the c a t a l y t i c carbonylation of neat p i p e r i d i n e at 71°C. and 1 atm. t o t a l pressure using [HRu(C0) 3] n, [Ru] = 2.2xlO" 2M 201 55. [HRu(C0) 3] n - catalysed carbonylation of p i p e r i d i n e . Amine dependence 202 56. [HRuCCO)^]^ - catalysed carbonylation of p i p e r i d i n e . CO dependence 204 57. [HRu(C0) 3] n - catalysed carbonylation of p i p e r i d i n e . Ru dependence 205 XVTI Page 58. Plot of ( k " ) _ 1 vs. [ C O ] - 1 from data i n F i g . 56 208 59. Arrhenius p l o t f or the [HRu(C0) 3] n catalysed carbonylation of p i p e r i d i n e 210 60. Solution i n f r a - r e d spectrum i n p i p e r i d i n e of the f i l t r a t e from the [HRu(C0) 3] n catalysed carbonylation of p i p e r i d i n e i n the presence of triphenylphosphine 220 61. "^H n.m.r. spectrum of azetidine i n C,D, s o l u t i o n 224 D o 62. Adenine 227 63. CO uptake p l o t f o r the c a t a l y t i c carbonylation of neat p i p e r i d i n e at 75°C. and 1 atm. t o t a l pressure using [Ru(C0) 2(0Ac) ( p i p ) ] 2 , [Ru] = 3.0xl0~ 2M 228 64. [Ru(CO) 2(OAc)(pip)] 2 - catalysed carbonylation of p i p e r i d i n e . CO dependence 231 65. [Ru(CO) 2(OAc)(pip)] 2 - catalysed carbonylation of p i p e r i d i n e . Ru dependence 232 Chapter 7 66. I n i t i a l CO uptake p l o t for the c a t a l y t i c carbonylation of neat p i p e r i d i n e at 60°C. and 1 atm. t o t a l pressure using CsRu(CO) 3C£ 3, [Ru]=2.10x10~2M 236 67. CO uptake p l o t s for the c a t a l y t i c carbonylation of neat p i p e r i d i n e at 6 0 ° C , 650 mm Hg t o t a l pressure using CsRu(CO) 3C£ 3, [Ru] = 2.10xl0 _ 2M . 237 68. Infra-red spectrum (nujol mull) i n the carbonyl s t r e t c h i n g region of the compound (1) derived from CsRu(C0) 3C& 3 i n piperdine under CO 243 x v i i i Page 69. CsRu(CO)„C£„ - catalysed carbonylation of p i p e r i d i n e . Ru dependence , 247 70. CsRu(C0) 3C£ 3 - catalysed carbonylation of p i p e r i d i n e . CO dependence 249 71. CsRu(CO) 3C£ 3 - catalysed carbonylation of p i p e r i d i n e . Amine dependence 252 72. CO uptake pl o t s for the c a t a l y t i c carbonylation of neat p i p e r i d i n e at 65°C., 580 mm Hg t o t a l pressure using CsRu(CO) C£ , [Ru] = 2.20xl0~ 2M. Pre-e q u i l i b r a t e d under vacuum for 12 hours 253 73. CsRu(CO) 3C£ 3 - catalysed carbonylation of p i p e r i d i n e . Log^Q(maximum rate) vs. 1/T from data i n Table XXXII 258 74. [Ru(CO) 3C£2l2 geometrical isomers 260 75. CO uptake p l o t for the c a t a l y t i c carbonylation of _2 piperazine at [Ru]=7.8xl0 M., [piperazine]=1.45M. at 80°C. and 1 atm. t o t a l pressure i n toluene 265 76. Ru 3(CO)^2 ~ catalysed carbonylation of piperazine. [Ru] dependence 270 77. CO uptake p l o t for the c a t a l y t i c carbonylation of neat 1-methyl piperazine at 80°C. and 1 atm. t o t a l pressure using Ru^(C0) 1 9, [Ru] = 5.4xl0~ 2M 273 xix ABBREVIATIONS The following l i s t of abbreviations, most of which are commonly adopted in chemical research literature, w i l l be employed in this thesis. A l l temperatures are in °C unless specifically denoted °K. acac atm. Bu S u (Me, Et used similarly for simple alkyl groups) tBu.N+ 4 Et 4N Me3S + Calcd. DMA e.s.r. i . r. L log In m M 0" 0 I II acetylacetonate, CH3C=CHCCH3 atmosphere butyl, C 4H g t-butyl, -C(CH 3) 3 M n+ tetrabutylammonium cation tetraethylammonium cation trimethylsulphonium cation calculated N,N-dimethylacetamide, CHACON(CH3)2 electron spin resonance infrared ligand common logarithm natural logarithm moles molar, mole l i t r e metal atom X X min minutes nm nanometres n.m.r. nuclear magnetic resonance OAc acetate, CILjCOCf Ph phenyl, PPh^ triphenylphosphine, (C^R^)^P Ph^As + tetraphenylarsonium cati o n , (CgH^^As V - 1 N t o t a l pressure ( t o t a l ) I? (CO) CO p a r t i a l pressure pip p i p e r i d i n e , C^ H^ -^ N p i p H + piperidinium cation, Cj-H^N*" py pyridine, C^H^N pyH + pyridinium cation, C^H^N+ s,sec seconds THF tetrahydrofuran t . l . c . t h i n - l a y e r chromatography u.v. u l t r a v i o l e t v.p.c. vapour phase chromatography e molar e x t i n c t i o n c o e f f i c i e n t A„ molar conductance, ohm "'"mole "*"cm2 M v frequency, cm "*". ACKNOWLEDGEMENTS I wish to express my sincere appreciation to Professor B. R. James, who has been a source of advice, guidance, and inspiration throughout the course of this work. I am indebted to Mrs. Anna Wong for her excellent typing services. I also wish to thank the National Research Council of Canada for the award of a Postgraduate Scholarship. 1 Chapter 1 THE USES OF RUTHENIUM' CARBONYLS IN HOMOGENEOUS' CATALYSIS " 1.1 General introduction The past fifteen years have seen increased research interest in homogeneous catalysis, particularly by transition metal complexes in solution. Predictions were made that many heterogeneous processes would be replaced by more efficient homogeneous ones. Studies in homogen-ous catalysis have been largely prompted by the comparative ease with which the active catalytic centre may be defined. Homogeneous catalytic systems are more efficient than heterogeneous systems in terms of available active sites and so have an economic advantage for industry, particularly i f they can be conducted under mild reaction conditions when product selectivity i s usually higher. The only shortcoming l i e s in the possibility of a more d i f f i c u l t product separation stage at the end of the reaction. This thesis investigates the chemistry of several ruthenium carbonyl complexes with special reference to their activity as catalysts in amine carbonylation reactions. There i s an extensive literature covering the 1 2 3 use of transition metal carbonyls as catalysts in organic reactions, ' ' 2 but, apparently, no review devoted s o l e l y to c a t a l y s i s by carbonyl complexes of ruthenium. By way of i n t r o d u c t i o n to the m a t e r i a l under i n v e s t i g a t i o n , a survey of previous research w i l l be presented. An i n i t i a l c o n s i d e r a t i o n of the h i s t o r y of t r a n s i t i o n metal 4 carbonyls r e v e a l s the chronology shown i n Table 1. Since the f i r s t r eport of Ru(CO)^ i n the l i t e r a t u r e ^ many mononuclear and pol y n u c l e a r s u b s t i t u t e d ruthenium carbonyls have been prepared. However, the only other d e f i n i t e l y c h a r a c t e r i s e d non-substituted complex i s the t r i -ruthenium dodecacarbonyl s p e c i e s , Ru^CCO)^ which was o r i g i n a l l y d escribed as Ru2(C0)g. In general, the wide range and nature of ruthenium carbonyl d e r v i a t i v e s can be explained i n terms of the accepted p i c t u r e of " s y n e r g i s t i c " metal carbonyl bonding^ and the intermediate p o s i t i o n of ruthenium i n the " s o f t - h a r d a c i d and base" g scheme . Since almost every other group i s thought to have poorer II-9 acceptor a b i l i t y than carbon monoxide , r e p l a c i n g carbonyl groups w i t h s u b s t i t u e n t s should enhance the metal-carbon bonding of those carbonyl groups that remain. Molecular o r b i t a l theory using extended Huckel c a l c u l a t i o n s ^ ^ confirms experimental determination of metal-carbon bond strengths f o r a s e r i e s of chromium carbonyls and p r e d i c t s that carbonyl groups t r a n s - t o other s u b s t i t u e n t s ' should be more s t r o n g l y bound than carbonyl groups that are t r a n s - t o one another. Although carbonyls of the Group V I I I metals have been known f o r some eighty years, i t i s only r e c e n t l y that comprehensive reviews of 11 t h e i r chemistry have appeared . Ruthenium, i n p a r t i c u l a r , has a wide range of p o s s i b l e o x i d a t i o n s t a t e s (0-VIII) and a great tendency to form polynuclear species. These c h a r a c t e r i s t i c s are, i n many ways, 3 Table I Chronology of metal carbonyl syntheses (from ref.4) 1890-1900 : N i ( C 0 ) 4 , Fe(C0> 5. 1900-1910 : Co 2(C0) g, Mo(C0) 6 > Ru(C0) 5. 1920-1930 : C r ( C 0 ) 6 , W(C0) 6-1940-1950 : I r 2 ( C 0 ) g , Re 2(C0) 1 ( ), 0s(C0> 5, Rh 2(C0) g. 1954 : Mn 2(C0) 1 ( ). 1959 : V(C0) 6. 4 compatible with those required for catalytic activity in a soluble transition metal complex. The main prerequisites of a catalyst are that (a) i t should not react in an irreversible way with the substrate, (b) i t should provide a low-energy path for the formation of the inter-mediates, and (c) the intermediates should not be kinetically too stable. A consideration of available experimental data for the metal carbonyl derivatives of a vertical triad shows that the reactivity of the second-row transition metals i s substantially higher than that of the two other metals. This is particularly true for reactions involving the la b i l i s a t i o n of a metal-carbon bond, such as the exchange reaction between M(CO)^ and 14 12 CO (activation energies for the exchange are 39 kcal/mole for Cr(CO)^, 30.8 kcal/mole for Mo(C0)6, and 40.4 kcal/mole for W(C0)6), and the carbon 13 monoxide displacement from M(C^Hcj) (C0) 2 for which the order of reactivity is Co<Rh^Ir. In view of this finding, one might expect carbonyl complexes of ruthenium to be quite active species in reactions requiring loss of co-ordinated carbon monoxide (e.g. carbonylation and hydroformylation processes). Our present knowledge of the nature of carbon monoxide binding suggests that low oxidation states of the metal w i l l be stabilised by carbon monoxide and other strong II-acceptor ligands, hence hydrogenation processes are less l i k e l y to result in reduction of the catalyst to metal when these ligands are present. 1.2 Hydrogenation The catalytic activity of ruthenium carbonyl complexes in hydro-14 genations has not been extensively studied. A patent describes the use of HRuC£(C0)(PPh 3)3 and H2Ru(C0)(PPh2Me)3 as catalysts for the reduction 5 of keto-, formyl and n i t r i l e groups, as well as non-aromatic ^C=C^ and -C=C- groups. The reaction appears to require pressures of hydrogen i n the 10-100 atm. range at temperatures from 20°C. to 130°C. i n hydrocarbon, ether or alcohol solvents and the substrates reduced include hexenes, mesityl oxide, methyl i s o b u t y l ketone, acetophenone, propion-aldehyde and b e n z o n i t r i l e . The HRuC£(CO)(PPh^) 3 complex has been shown to give p a r t i a l reduction of both acetylene and ethylene under ambient 15 16 conditions ' i n the presence of hydrogen; hydrogen-deuterium exchange of t h i s complex i s thought to occur v i a oxidative a d d i t i o n to give an eight co-ordinate ruthenium(IV) intermediate as i l l u s t r a t e d below (eq.1.1) HRuC£(C0) (PPh3)3+D2===== [HRuC£(C0) (PPh-j) 3 D 2 ] DRuC£ (CO) (PPh^+HD (1.1) Eight co-ordinate intermediates may be involved i n the c a t a l y t i c o l e f i n hydrogenation reactions. However, the work of Schunn"^ on studies of deuterium exchange i n the presence of C^D^, 1-butene and c e r t a i n metal complexes reveals that HRuC£(CO)(PPh 3) 3 produces amounts of H 2 and HD i n excess of t h e o r e t i c a l values calculated f o r complete exchange of the o r i g i n a l Ru-H bond. This fact i s explained by proposing a ligand-to-metal hydrogen trans f e r process i n v o l v i n g oxidative a d d i t i o n of an ortho-phenyl carbon-hydrogen bond to the metal. Such ligand d i s s o c i a t i o n s are l i k e l y to be important e s p e c i a l l y at higher temperatures, and the higher co-ordination number for the RuHD2 intermediate i s not e s s e n t i a l . In the 18 event of t h i s orthometallation mechanism operating, a stoichiometric hydrogenation of a substrate should be possible i n the absence of mole-19 cular hydrogen, as observed f o r the HRuC£(PPh 3) 3 complex 20 The complex RuC£ 2(CO)(PPh 3) 2(DMF) i s useful f o r hydrogenation of alk-l-enes i n methanol under mild conditions, although the c a t a l y s i s occurs 6 only a f t e r i n i t i a l hydride formation using borohydride. The c a t a l y t i c a c t i v i t y of chlorocarbonyl complexes of 21 22 23 ruthenium i n DMA s o l u t i o n has been investigated. ' ' The I I species Ru (CO), obtained by treatment of Ru solutions i n DMA with carbon monoxide, w i l l catalyse the slow hydrogenation of maleic acid. The trend i n c a t a l y t i c a c t i v i t y , Ru^ ~> Ru''" (C0)> Ru''" (CO) 2 i s also exhibited by the corresponding ruthenium (II) complexes, and a s i m i l a r trend has been observed i n aqueous hydrochloric a c i d 24 solutions. The trend towards decreasing a c t i v i t y with increasing number of carbonyl groups i s expected since t h i s w i l l tend to increase .25 the "promotion energy1 (P) for the hydrogen a c t i v a t i o n process, i f t h i s involves oxidative addition. i . e . H„ + ML ========i H„ML (1.2) 2 n ^ 2 n In t h i s process, c a t a l y t i c a c t i v i t y requires that 2E(M-H)r* E(H-H) + P, the presence of strong IT-acceptor ligands, such as carbon monoxide, withdrawing electron density from the metal w i l l reduce the a b i l i t y of the species to a c t i v a t e molecular hydrogen. A c t i v a t i o n of hydrogen i n aqueous acid s o l u t i o n by the species 2-Ru(CO)C£^ occurs i n an a u t o c a t a l y t i c manner. The proposed mechanism fo r the o v e r a l l reaction (eq. 1.3) involves a c t i v a t i o n of H 2 by a Ru(II) complex as i l l u s t r a t e d below (eqs. 1.4 and 1.5). 2Ru(CO)C£ 2~ + H 2 + 2H 20 2Ru(C0)(H 20)C£ 2~ + 2H + + 2C£~ (1.3) Ru I : E(CO) + H 2< Ru I : t(CO)H + H + (1.4) Ru I : i :(CO)(H) + 2Ru i : C I(C0) £ a S t > 3Ru I ] :(C0) + H + (1.5) 7 Isotopic exchange of deuterium with water is catalysed by the 2- 2-2^ Ru(CO) (H20)CJ14 species, but not by Ru(CO) 2C£ 4 . 27 28 Fahey ' has found a number of ruthenium complexes to be active in hydrogenation of l,5,9-cyclododecatriene(CDT) to cyclo-dodecene (CDE), an industrially important process. The most useful complex, RuC£ 2(CO) 2(PPh 3) 2, gives a 98-99% yield of CDE in the presence of certain solvents or added Lewis bases, e.g. N,N-dimethyl-formamide or RPh^. This high degree of selectivity i s very desirable since separation of CDE from the other possible hydrogenation products, cyclododecadiene (CDD) and cyclododecane (CDA) by d i s t i l l a t i o n would be d i f f i c u l t owing to the close proximity of their boiling points. The prior synthesis of the catalyst can be avoided by preparing i t 'in situ' from RuC£^, PPh^ and CO. Lower-valent complexes of ruthenium, such as Ru(CO) 3(PPh 3) 2 and Ru^CCO)^, have been found to be active catalysts for this reaction at pressures of approximately 200 p.s.i. of hydrogen and temperatures in the range 100-200°C. The mechanism of the reaction i s thought to parallel that of other ruthenium hydride 29 30 24 complex hydrogenations ', ' and follow the course shown in eqs. (1.6, 1.7 and 1.8) where the hydrogenation steps are rate-determining. K l k l RuH + CDT ^ -» Ru(dienyl) => RuH + CDD (1.6) H 2 k2 RuH + CDD ^ ~~* Ru(alkenyl) r f r > RuH + CDE (1.7) H 2 k3 RuH + CDE ^ " ~- Ru(alkyl) rr => R UH + CDA (1.8) 2 31 A further report discusses the use of RuC& 2(C0) 2(PPh 3) 2 in the hydro-0 genation of a variety of alkenes and dienes. The presence of added — 2 K0 3 8 PPh^ was found to be necessary and an act i v e intermediate with the composition RuHC£(CO) 2(PPh 3) 2 was proposed; t h i s hydride has also been 32 synthesised. 33 L'Eplattenier and co-workers discovered that Ru.j(CO)^ 2, Ru(CO),. and Ru(acac) 3 were active f o r the conversion of nitrobenzene to a n i l i n e i n reasonable y i e l d s (60-70%). The rea c t i o n conditions required 100 atm. p a r t i a l pressure of both CO and tL, and temperatures i n the range 140-160°C. Small q u a n t i t i e s of 2,2'-diphenylurea are produced and the amount of t h i s side product tends to increase as the C0:H2 r a t i o i s increased. Stoichiometry i s as indicated i n eq. 1.9.. C,H,N0o + 2C0 + H„ ^ C-H,-NH0 + 2C0 o (1.9) 6 5 2 2 6 5 2 2 The formation of a phenylnitrene intermediate, s t a b i l i s e d by co-ordination to ruthenium, was proposed as one possible mechanism to account for the experimental evidence. i . e . C 6H 5N0 2 + 2C0 > ( C g H ^ ) + 2C0 2 (1.10) (C6H5N=) + H 2 ~> C 6 H 5 N H 2 ( L U ) (C6H5N=) + 2C0 >' C6H5NCO (1.12) C-H-NH,, + C,HcNC0 3* C,Hr.NHCONHC,Hr- (1.13) 6 5 2 6 5 6 5 6 5 The capture of nitrenes by metal carbonyls was f i r s t reported i n 1967 by 3 A Dekker and Knox, who i s o l a t e d a bis-phenylnitrene of formula F e 2 (C0)g(NCgH,_)2. This compound was assigned structure (I) as shown below. (0C)„Fe Fe(CO) J r-r \ J 9 In solution (I) is spontaneously converted into a urea-type of complex (II), the structure of which has been elucidated by X-ray 35,36 0 diffraction.' C ,C,HC 6 5 H5 C6 A (OC) 3Fe- -Fe(CO) (II) 33 The replacement of Ru^CO)^ by Fe(CO),. was found to result in a lower percentage of aniline as product and about 50% of the nitro-benzene remained unreactive. One possible explanation for this fact regards the reduction of nitrobenzene as involving ruthenium analogues of complexes "(I) and (II), which can undergo hydrogenation to aniline or 2,2'diphenylurea respectively. The lower catalytic activity of iron was ascribed to a less labile iron-nitrogen bond in (I) relative to the ruthenium-nitrogen bond in an analogous species. Osmium is thought to form more stable metal-hydrogen bonds than ruthenium (for 37 example OsK^CO)^ is stable, while the corresponding hydridocarbonyl 38 of ruthenium is unstable and decomposes very readily at room temper-ature) . If hydrogen transfer from a metal centre is a key step in the reaction, then i t might be expected that osmium carbonyls would be less active than similar ruthenium compounds. Metal carbonyl clusters have generally been found to be stable 39 only under limited conditions of temperature and pressure, thereby 10 possibly discouraging i n v e s t i g a t i o n of t h e i r c a t a l y t i c uses. On the other hand, c l u s t e r s s u f f i c i e n t l y robust to survive t y p i c a l conditions for c a t a l y t i c processes have often proved unreactive. The fact that ruthenium shows a great a f f i n i t y f o r n i t r o s y l ligands and that n i t r o s y l complexes are often more r e a c t i v e than the i s o -40 e l e c t r o n i c carbonyl complexes has led various workers to i n v e s t i -gate the c a t a l y t i c p o t e n t i a l of n i t r o s y l substituted carbonyl c l u s t e r s 41 42 43 of ruthenium '. The complex RU 3(CO)^Q(NO)2 , f i r s t prepared by Collman i n 1969, shows a c t i v i t y f o r the hydrogenation of 1-hexene to n-hexane at 25°C. and 4 atm. H 2, although the major reaction appears to involve isomerisation to i n t e r n a l hexenes. The same complex w i l l catalyse the t r i m e r i s a t i o n of dimethylacetylenedicarboxylate to 43 hexakis(carbomethoxy)benzene. This n i t r o s y l complex, prepared from Ru 3(C0)^2 a n < i NO, i s only one of a vast number of d e r i v a t i v e s that can be prepared from dodecacarbonyl triruthenium as s t a r t i n g 4 4 4 5 46 4 7 material ' \ ' although the c a t a l y t i c p o t e n t i a l of such d e r i -vatives has received l i t t l e a t tention. 1.3 Hydroformylation In the past, studies on the hydroformylation process have l a i d emphasis upon the use of cobalt and rhodium complexes as c a t a l y s t s . The a v a i l a b l e information suggests that ruthenium complexes are not very 48 49 50 e f f i c i e n t , although examples do e x i s t i n the patent l i t e r a t u r e .' The most e f f e c t i v e complex observed so f a r i s RuCCO^CPPh^)!^; at 100°C. and 100 atm. pressure of a 1:1 C0/H2 mixture, the complex gives 80% 11 conversion of pent-l-ene to a mixture of corresponding s t r a i g h t and branehed-chain aldehydes i n 16 hours. The co-ordination of hydrogen and carbon monoxide to the same metal atom endows i t with c a t a l y t i c p o t e n t i a l f o r both hydroformylation and hydrogenation processes, and a l i k e l y intermediate H 2Ru(CO) 2 (PPh 3) 2 > has been made by t r e a t i n g 52 Ru(CO) 3(PPh 3) 2 i n THF with 120 atm. of hydrogen at 130°C. Valentine 53 54 and Collman ' had reported e a r l i e r that trans-Ru(CO) 3(PPh 3) 2 was unreactive toward hydrogen at temperatures up to 150°C., although the hydrogen pressure and solvent were not s p e c i f i e d i n t h e i r p u b l i c a t i o n . However, the same workers did observe a reaction with hydrogen upon i r r a d i a t i o n of the complex at 365 nm. and the formation of an unstable photoproduct capable of hydrogenating cyclohexene. The following reaction scheme was envisaged (eq. 1.14). hv H Ru(C0) 3(PPh 3) 2 n m > Ru(CO) 2(PPh 3) 2^ ^>[H 2Ru(CO) 2(PPh 3) 2] (1.14) (active species) Photochemical a c t i v a t i o n of metal carbonyl c a t a l y s t s has been reviewed by Strohmeier^^, however, no mention i s made here of uses of ruthenium carbonyl complexes. 32 56 James and Markham ' have reported that the complex HRuC£(C0) 2(PPh 3) 2 i s a c t i v e , i n terms of gas uptake, under hydro-formylation conditions at 80°C. and 1 atm. t o t a l pressure i n the presence of alkene substrates. However, the process involved may have been simply hydrogenation of the substrate. The use of Ru(C0) 3(PMe 2Ph) 2 for the hydroformylation of terminal o l e f i n s to aldehydes has been patented by 12 Lawrenson and Green.^ Other ruthenium carbonyl complexes of unspecified 58 stoichiometry are sa i d to require more vigorous conditions than those required for cobalt c a t a l y s t s . Several groups have worked on the use of Ru-j(CO)^2 i n hydro-formylation reactions. Conversion of dienes and formaldehyde to d i o l s i n the presence of Ru.j(CO)^2 requires severe conditions of temperature 59 and pressure. Hydroformylation of simple o l e f i n s using t h i s c a t a l y s t 60 r e s u l t s i n some production of p a r a f f i n s and shows lower a c t i v i t y than eit h e r C o 2(C0) g or Rh^ ( C O ) ^ ! 6 1 The cobalt and rhodium systems have been extensively studied and are thought to involve hydrido-carbonyl 62 1 intermediates .' C a t a l y s i s of h] probably involves s i m i l a r species. intermediates .' C a t a l y s i s of hydroformylation by ruthenium complexes 1.4 Other miscellaneous reactions Ruthenium carbonyl complexes have been used to catalyse a v a r i e t y of other types of reaction. For instance, Ru^CCO)^ i s act i v e as a 63 c a t a l y s t i n the carbonylation of acetylene to give hydroquinone ; a 60% y i e l d i s obtained i n THF solu t i o n at 200°C. using 12 atm. CO and 10 atm. H 2. i . e . 2C 2H 2 + 2C0 + H 2 >H0-"T~jK)H (1.15) Water or alcohol could also be used as the hydrogen source and s i m i l a r y i e l d s of hydroquinone are obtained. i . e . 2C 2H 2 + 3C0 + H 20 » H 0 - ^ ^-OH + C0 2 (1.16) The complex R u g ( C 0 ) l g 6 ^ ' 6 5 c a n also be used to convert acetylene to hydro-quinone i n 58% y i e l d under s i m i l a r conditions. 13 66 67 In a hydrocarboxylatlon reaction ' dodecacarbonyl t r i -ruthenium was used i n methanol at 140°C. under high pressures of carbon monoxide to convert a l l e n e to methyl methacrylate i n 50% y i e l d (eq. 1.17). CO C\H3 C H 2 = C = C H 2 C i ^ C H 2 = C - C ° 2 C H 3 ( 1 - 1 7 ) When the reaction was c a r r i e d out f i r s t at 140° and then completed at 190°, dimethyl-a,a-dimethyl-a'-methyleneglutarate(I) was obtained i n 23% y i e l d together with methyl methacrylate i n 18% y i e l d (eq. 1.18). Several mechanisms were proposed for the generat ion of the e s t e r ( I ) , but none would f u l l y account f o r the experimental evidence. CH. CH, CH„ co | I 3 II 2 CH 2 = C = CH 2 C R Q H > CH 2 = C-C0 2CH 3 + CH^C-C-CH^C-CO^^ (1.18) 3 CH 3 (I) Unspecified ruthenium hydrocarbon-carbonyl complexes have been 68 proposed as intermediates i n a polymethylene synthesis. In t h i s r eaction ruthenium tetroxide or ruthenium dioxide decomposes upon i r r a d i a t i o n i n hydrocarbon solvents i n the presence of CO and H 2 and at temperatures l e s s than 140°C. to y i e l d s t r a i g h t chain, oxygen-free p a r a f f i n hydrocarbons of molecular weighty 108,000. Ruthenium complexes such as HRuBr(CO)(PEt 2Ph)^ a n d 69 HRuC£(Et 2PC 2H 4PEt 2) have been used to catalyse the decomposition of formic acid i n t o C0 2 and H 2, and the l i b e r a t e d H 2 may be used to convert saturated aldehydes into alcohols, but the c a t a l y t i c a c t i v i t y i s not as good as with c e r t a i n i r i d i u m complexes. The species [Ru(0C0Rp) 2 (CO) (PPh^),, ], 14 where Rp = CF^, C^F^, or C^F^, catalyse the reverse reaction, the dehydrogenation of alcohols to aldehydes or ketones.^ A number of ruthenium species have been used for the hydro-dimerisation (hydrogenation during dimerisation) of a c r y l o n i t r i l e . 7 1 The only carbonyl complex studied was RuC£ 2 ( C 0 ) 2 (py) 2 > a n d this species appears to inhibit a l l reactions of acrylonitrile, which was rationalised in terms of the cis-complex containing strongly co-ordinated CO and pyridine ligands. At the other extreme the species Ru (NO) (PPh.^ (CO) ( X ) 7 2 , where X = OH,CN,NCS or CI, contain a very labile CO group and the complexes w i l l , for example, bind oxygen irreversibly. Ru(NO)(PPh3)2(CO)X + 0 2 > Ru(NO)(PPh3)202X + CO (1.19) The oxygen complexes have been used as catalysts under 1 atm. 0 2 to oxidise tertiary phosphines and arsines to the corresponding oxides. 74 One brief account of the use of ruthenium carbonyl complexes for olefin disproportionation indicates that they show l i t t l e activity when compared with some iso-electronic molybdenum nitrosyl compounds (Table II). Both of the nitrosyl containing ruthenium derivatives show some double-bond isomerisation activity with 1-2% 1-pentene being produced. 1.5 Amine carbonylation The main subject matter of this thesis concerns the nature and mode of action of a number of ruthenium species capable of carrying 15 Table I I . Comparison of o l e f i n disproportionatlon a c t i v i t y  for Mo and Ru c a t a l y s t s (from ref.74) Complex % disproportionation of 2-pentene i n 1 hour M o C £ 2 ( N O ) 2 ( P P h 3 ) 2 50 MoC£ 2(NO) 2(py) 2 50 MoC£ 2(NO) 2(bipy) 2 50 RuC£ 2(CO) 2(PPh 3) 2 trace RuC£(C0)(NO)(PPh 3) 2 trace RuC£ 3(N0)(PPh 3) 2 trace ( a - conversion to 2-butene and 3-hexene) 16 out one particular carbonylation, that i s , the conversion of various amines to amides or substituted ureas. Cyclic secondary amines are particularly interesting substrates in view of their relatively high basicity 7"' and stereochemical r i g i d i t y . R2NH + CO 5* R2NCHO (1.20) 2R2NH + CO 3- R2NC0NR2 + H 2 (1.21) A high degree of selectivity, resulting in good yields of the N-formylamines, is desirable since these products are not only useful in industry^*, but are also interesting synthetic intermediates in organic chemistry. In an early reference 7 7 dicobalt octacarbonyl was employed in benzene under 200 atm. of CO and at 200-220°C. to convert piperidine to N-formylpiperidinein 78% yield. After evaporation of solvent and subsequent d i s t i l l a t i o n of the residue under reduced pressure, the N-formyl product was characterised by formation of a characteristic 78 complex with mercuric chloride. The reaction of dicobalt octa-carbonyl with dimethylamine was examined in more d e t a i l . 7 7 Evidence from analytical and infrared spectroscopic data suggested the homo-molecular disproportionation shown in equation (1.22). Subsequent formation of dimethylformamide took place without liberation [Co(C0) 4] 2 + (CH3)2NH — ^ [ (CH3)2NHCo(CO)4] + [Co(CO) 4] _ (1.22) of carbon monoxide which led these workers 7 7 to suggest a mechanism such as that shown in equation 1.23. 17 CH o CH„ 3 / 3 N: H B:CO=C = 0 CH- ,CH-11+ 3 \ / 3 B: + =Co + N: H I C 0: CH- CH-3 \ / 3 Ni (1.23) H-C=0: (B = dimethylamine) 79 In the 1960's Saegusa and co-workers discovered that both cupric chloride and cupric cyanide would catalyse the insertion of isocyanides into amine (N-H) bonds to give amidines (eq. 1.24), and 80 shortly afterwards the same group reported the excellent activity of various copper compounds for insertion of CO into N-H bonds in both aliphatic and aromatic amines. N-H + R"-N=C Cu compound R \ > ,N-C-H III N (1.24) These copper catalysts seemed to give good selectivity as compared to 81 82 Pd or Mn systems. ' Even under the severe conditions of 60-80 atm. and 140°C. only a few percent conversion of the amine to a substituted urea was observed. Interestingly, the addition of water was found to accelerate the reaction rate, a finding which suggests l i k e l y carbon monoxide co-ordination to the metal as one intermediate step, since co-ordination of carbon monoxide to cuprous complexes i s known to be 80 favoured by the presence of water. Further, simultaneous CO and 18 83 amine co-ordination to Cu(I) i s well-known . Thus, i t was suggested that Cu(I) complexes might be the true a c t i v e species i n such r e -actions. Later k i n e t i c studies on amine carbonylation by a Cu(I)-Cu(II) system using carbon monoxide under ambient conditions have 84 been conducted by Brackman , although i n t h i s case the substituted urea was produced i n 95% y i e l d and the experimental r e s u l t s were interpreted i n terms of a c t i v e cuprous species. f , Cu(I),Cu(II) ( N-H + CO + JsC ^ / V c - < N ' * \ — / II V i n p i p e r i d i n e 0 (1.25) + H 20 Interest i n t h i s system i s enhanced by the s u r p r i s i n g l y mild conditions under which i t operates. The use of s a l t s of other Group IB and IIB metals has also been 80 mentioned , generally, the conversion of p i p e r i d i n e to N-formyl p i p e r i -dine gave lower y i e l d s than copper s a l t s under the same conditions. In 85 a more recent report , s i l v e r acetate has been used under 1 atm. of CO and at room temperature to convert primary amines to substituted ureas and secondary amines to N,N,N',N' t e t r a - a l k y l oxamides (eqs. 1.26 and 1.27). RNH„ CO/AgOAc > M . C _ M ( 1 > 2 6 )  2 II 0 CO/AgOAc R„NH 3> R„NC-C-NR„ (1.27) 2 2 IMI 2 0 0 19 A mechanism was proposed i n which CO inserted into t r a n s i e n t l y formed Ag-N bonds to give carbamoylsilver intermediates (eqs. 1.28, 1.29 and 1.30). [" R 2 N y e.g. 2R2NH + 2AgOAc R, 2C0 -> 2 0 = C i A g -0 II -OC-CH, 0=C / < -H ! i 0 « I II Ag 0C-CH3 2HOCOCH3 + 2R2NC-Ag (1.28) (1.29) 2 R0NC-Ag H> R„NC-C-NR„ 2 II II 2 0 0 + 2Ag (1.30) Reactions of Group VIII metal carbonyls with amines have been 8 6 known for some time. Durand and Lassau employed the dimeric complex [Rh 2(C0) 4C£ 2] as a c a t a l y s t f o r the conversion of n-butylamine to N-butylformamide and N,N'-dibutylurea. The addition of c e r t a i n phosphines (P(Me) 3, PPh 3, P ( 0 E t ) 3 , etc.) resulted i n a marked increase i n the 87 88 percentage of amide product. Edgell and co-workers ' have studied the reactions of ir o n pentacarbonyl with amines i n an attempt to char-a c t e r i s e species containing both co-ordinated CO and amine. The species are p a r t i c u l a r l y i n t e r e s t i n g since amine carbonylation occurs upon heating to 60°C. Investigation by n.m.r. and i . r . spectroscopic techniques indicated attack of amine at the co-ordinated CO group, but 20 from these results i t was not clear whether direct attack or a prior co-ordination of amine to the metal atom was involved. Interest in ruthenium carbonylamine species as potential catalytic intermediates 89 stems from the work of Allen et. a l . on the synthesis of ruthenium(II) carbonyl ammines and, particularly, the interesting work of Hieber and 90 Heusinger. The latter prepared the polymeric complex [Ru(CO) 2I 2] n and carried out reactions with complex-forming ligands(X) having free electron pairs to give species of general formula Ru ( 0 0 ) 2 X 2 1 2 ' The ammine complex was found to decompose above -30°C. to give free and co-ordinated formamide. >-30°C. [Ru(CO) 2I 2]- >Ru(C0) 2(NH 3) 2I 2 > 0 II Ru(NH0) (H0NCH). 3 n 2 4-n I 2 (1.31) It seems very l i k e l y that carbamoyl species (M-C0NR2) are intermediates in amine carbonylation process, and carbamoyl complexes of metals such as Cu, Mn and Pd are known to exist. Synthesis is usually achieved by employing a cationic metal carbonyl, and predictions of carbamoyl sta b i l i t y have generally been made by using correlations with C-0 u- . 9 1 stretching frequencies. L MCEO + 2 HNRR" > L M C ^ ™ - + H„NRR (1.32) n n ^NRR 2 87 88 As previously mentioned ' , the only neutral metal carbonyl to be used specifically for carbamoyl preparation has been Fe(CO)^, although other - 92 species such as Cr (CO) ,-CON(C2H,-)2 have been isolated and there is - 93 94 spectroscopic evidence for Ni(CO) 3CON(CH 3)2 . One kinetic study on reactions of primary amines with cationic manganese and rhenium carbonyls was interpreted as involving base-catalysed nucleophilic attack of the amine at the carbonyl carbon atom resulting in generation of the carbamoyl products in 21 a concerted process (eq. 1.33). An increase i n the bulkiness of the ligand was found to r e s u l t i n rates slow enough to measure by following changes i n low frequency carbonyl stretching absorptions. This trend has also been observed i n the r e a c t i o n of organic esters with amines . , 9 5 to give amides. CO OC \ 1 / M/ \ CO CO + + 2RNH„ 0^ . .NHR OC \ M i \ "CO CO + RNH 3 + (1.33) The importance of s t e r i c e f f e c t s seems to i n d i c a t e that s t e r i c crowding around the metal i s responsible for the slowness of the reaction, and 96 l e s s s t e r i c a l l y crowded metal carbonyls are c e r t a i n l y more reac t i v e . 97 A n g e l i c i has also prepared a series of ruthenium carbamoyl d e r i v a t i v e s containing the cyclopentadienyl group (C^H,.); h i s studies on these species of general formula (C,-Hc)Ru(C0)o (CONHR) showed very low terminal C=0 stretching frequencies, and he has suggested by comparison with (CCJH,-)RU(C0)2C£ that the carbamoyl group i s a better a^-donor than the c h l o r i d e ligand. C y c l i c secondary amines such as p i p e r i d i n e and p y r r o l i d i n e gave a mixture of two products (indicated by t . l . c ) , one probably being (C^H^)Ru(00)2(CONR2), however neither product was i s o l a t e d . In the l a s t few years work at the U n i v e r s i t y of B r i t i s h Columbia and the Un i v e r s i t y of Waterloo has shown several ruthenium carbonyls to 22 be us e f u l c a t a l y s t s f o r amine carbonylation. > y y » - L U U T h e c o n t e n t of t h i s work w i l l be summarised i n l a t e r chapters and the chemistry of one p a r t i c u l a r c a t a l y s t w i l l be examined i n more d e t a i l . 23 Chapter 2 APPARATUS-AND- EXPERIMENTAL PROCEDURE 2.1. Materials 2.1.1. Ruthenium compounds Ruthenium was obtained as the t r i h y d r a t e RuC^.SlL^O, which i s a c t u a l l y a mixture of Ru(III) and Ru(IV) species"'"^''"> 1^2,103^ f r o m Johnson Matthey Co. Tri-ruthenium dodecacarbonyl was prepared under 104 mild conditions by the method of Mantovani and Cenini. Ruthenium t r i c h l o r i d e (3.5 gms.) was added to 60 mis. of f r e s h l y d i s t i l l e d 2-ethoxyethanol and the mixture refluxed f or about 6 hours under a rapid stream of CO. The r e s u l t i n g lemon-yellow s o l u t i o n , which i s thought to contain species of formula Ru(CO) C£ , was reduced to the desired Ru(0) n m x ' species by the addition of 50 mis. ethanol and 4 gms. of granular zinc followed by heating to 85°C. with vigorous s t i r r i n g for 6 hours. The rapid stream of CO was maintained during t h i s second stage of the pre-paration and an orange c r y s t a l l i n e s o l i d gradually accumulated i n the 24 reaction f l a s k . The product could be separated from the remaining zinc by decanting the l i q u i d and repeatedly washing the metal with methanol. R e c r y s t a l l i s a t i o n from e i t h e r benzene or toluene and subsequent drying under vacuum gave 1.37 gms (16% y i e l d ) of t r i -ruthenium dodecacarbonyl (Calculated for Ru^C-^O^2 ;C, 22.49;Ru,47.54%. Found: C,22.97;Ru,47.75%). The i n f r a - r e d spectrum of the dodecacarbonyl, which has sharp bands at 2060(s), 2030(s) and 2010(m) cm 1 i n n-hexane, was used to assess the p u r i t y of the compound. The bridged acetate dicarbonyl polymer [Ru(CO) 2(OAc)] has been 99 prepared previously by treatment of a c e t i c acid - acetate solutions of the t r i c h l o r i d e i n propanol with 1 atm. CO at 80°C. Aft e r about 12 hours a yellow s o l u t i o n and some s o l i d products were produced. F i l t r a t i o n of t h i s mixture and subsequent removal of solvent from the f i l t r a t e i s reported to give the desired complex. Crooks and co-workers have obtained the same compound by t r e a t i n g Ru 3(CO)^ 2 with g l a c i a l a c e t i c 47 acid. The former of these two methods was employed i n t h i s work; however, extended pumping under vacuum on the reaction product and r i n s i n g the s o l i d with acetone and benzene f a i l e d to remove a l l of the ac e t i c acid. Since the polymeric product [Ru(CO) 2(OAc)] n proved d i f f i c u l t to p u r i f y , the species [Ru(CO) 2(OAc)(pip)] 2 was prepared by d i s s o l v i n g the impure polymer i n p i p e r i d i n e at 4 0 ° C , p r e c i p i t a t i n g the amine d e r i v a t i v e by the addition of excess water and c o l l e c t i n g the p r e c i p i t a t e by f i l t r a t i o n i n a i r . The yellow p r e c i p i t a t e was rinsed many times with water, dried i n a i r , and then dried under vacuum for 24 hours. Characterisation of the complex was accomplished by micro-25 analysis (Calculated for RuCgH^C^N:C,35. 90;H,4. 65;N,4. 65%. Found: C,35.54;H,4.80;N,4.61%), solid state (nujol mull) infra-red spectroscopy (Fig. 1 shows the carbonyl stretching region and Fig. 2 the region from 1400 to 500 cm and the "''H n.m.r. spectrum in C^ D^  (Fig. 3). The synthesis of the polymeric hydridocarbonyl catalyst [HRu(C0) 3] n 3- 2- 2- 2-and of the anionic species RuC£ 6 , Ru(C0)C£5 , Ru(C0)C£ 4(H 20) , Ru(CO) 2C£ 4 and Ru(CO) 3C£ 3 w i l l be discussed in later chapters. The species [Ru(CO) 3C£ 2] 2 was prepared by the method of Cleare and Griffith"*"^, as described in Chapter 7 and was used in the synthesis of certain amine-substituted derivatives. 2.1.2. Gases Carbon monoxide was obtained as CP. grade product from Matheson Co. Hydrogen, also from Matheson Co., was pre-purified by passage through a Deoxo-catalytic purifier before use to remove traces of oxygen. Nitrogen (L grade), oxygen, argon and carbon dioxide were obtained from the Canadian Liquid Air Company. 2.1.3 Substrates and other materials CP. grade piperidine and pyridine were purchased from Fisher Scientific Co. Both solvents were r e - d i s t i l l e d after drying over potassium hydroxide pellets in a nitrogen atmosphere, and these were also stored under argon in suitable flasks. N-formylpiperidine and piperidine hydrochloride were obtained from the Aldrich Chemical Company and not subject to further purification; N-ethylpiperidine was obtained from Reilly Ltd. Azetidine, obtained from Eastman Organic Chemicals, was 26 1200 1000 .800 800 WAVEN U M B E R ( C M ) r e f i o n ' s O O ^ ^ 29 dried over calcium hydride and purified by a bulb-to-bulb d i s t i l l a t i o n . Anhydrous piperazine and 1-methylpiperazine were also obtained from Eastman. 1-Formylpiperazine was prepared by the method of Horrom, 106 Freifelder and Stone in which methyl formate (30 gms, 0.5 moles) was mixed with anhydrous piperazine (43 gms., 0.5 moles) in a flask equipped with a reflux condenser. The temperature rose to 85°C. in one to two minutes and was then maintained at this level for five hours after which time the mixture was d i s t i l l e d . The fraction boiling at 94-97°C. (under 0.3-0.4 mm.Hg pressure) was collected. 1,4-Bis-formyl piperazine was prepared by a similar technique"'"^ involving the addition of methyl formate (60 gms., 1 mole) to anhydrous piperazine (43 gms., 0.5 moles) and d i s t i l l a t i o n of the product boiling at 151-159°C. under 0.1-0.2 mm.Hg pressure immediately after the i n i t i a l vigorous reaction has subsided. Propyleneimine, ethyleneimine, pyridine-2-aldehyde, pyridine-3-aldehyde and pyridine-4-aldehyde were a l l K. and K. products. Certified N,N-dimethylacetamide, obtained from Fisher Scientific Co., was purified by d i s t i l l a t i o n from calcium hydride under vacuum and stored under vacuum. Toluene, also obtained from Fisher Scientific Co., was purified by d i s t i l l a t i o n over calcium hydride under nitrogen and stored under argon. A l l other chemicals were reagent grade; d i s t i l l e d water was used in a l l experiments. 2.2. Apparatus for constant pressure gas-uptake measurements. Gas-uptake measurements were performed using the constant pressure apparatus shown in Fig. 4. A flexible glass spiral tube connected a Apparatus f or constant pressure gas-uptake or evolution measurements. 3 1 c a p i l l a r y manometer D at tap C to the pyrex reaction f l a s k A; two-necked reaction f l a s k s could also be used, allowing attachment of a sampling tube for c o l l e c t i o n of gaseous re a c t i o n products. The f l a s k was im-mersed i n a thermostated bath B containing s i l i c o n e o i l (Dow Corning 550), and clipped to a piston rod and wheel driven by a Welch v a r i a b l e speed e l e c t r i c motor for shaking purposes. The c a p i l l a r y manometer D containing n-butyl phthalate (a l i q u i d of n e g l i g i b l e vapour pressure) was connected to the gas-measuring burette c o n s i s t i n g of a mercury r e s e r v o i r E and a precision-bored tube N of known diameter. The gas burette was i n turn connected, v i a an Edwards high vacuum needle valve M, to the gas-handling part of the apparatus; t h i s consisted of a mercury manometer F, gas i n l e t Y, and connections to the Welch Duo-Seal rotary vacuum pump G. The c a p i l l a r y manometer and gas burette were contained i n a transparent perspex water tank thermostated at 25°C. The s i l i c o n e o i l bath consisted of a f o u r - l i t r e glass beaker insulated by polystyrene foam on a l l sides and enclosed byawooden box, with the top also covered by polystyrene. Both thermostat units were con t r o l l e d by Jumo thermo-regulators and e l e c t r i c a l r elay control c i r c u i t s , with heating provided by a 40 watt elongated l i g h t bulb. With mechanical s t i r r i n g and good i n s u l a t i o n , the temperature could be maintained to within ±0.05°C. A vertically-mounted cathetometer was used to follow the gas-uptake, and time was recorded during k i n e t i c experiments using a Lab-Chron 1400 timer. 2 - 3 P r o c e d u r e f o r a t y p i c a l g a s - u p t a k e e x p e r i m e n t F o r e a c h e x p e r i m e n t , known amounts o f t h e r e a c t a n t s ( e . g . r u t h e n i u m 32 complex, substrate) were added to the solvent i n the reaction f l a s k A. The c a t a l y s t was usually added i n a glass bucket and l i q u i d substrates with a syringe. S o l i d substrates, such as piperazine, were also u s u a l l y added i n a glass bucket. The solvent was always degassed before the addition of c a t a l y s t , e i t h e r by repeated freezing followed by warming under vacuum or simply by shaking under vacuum at room temperature and pumping (the l a t t e r method being used for a solvent of very low vapor pressure, such as DMA). The reaction f l a s k and s p i r a l were f i l l e d with the reactant gas, at a pressure somewhat l e s s than that required f o r the experiment, by connection to the gas-handling part of the apparatus at 0. The taps C and P were then closed and the reaction f l a s k compete with s p i r a l was disconnected from 0 and attached to the motor driven shaker I. The whole system up to tap C was then pumped down with taps H,K,L,J and M open. Reactant gas was admitted to the rest of the apparatus beyond C to a pressure s l i g h t l y l e s s than that desired f o r reaction. Tap C was then opened and pressure adjusted to the desired r e a c t i o n pressure by introduction of gas through gas i n l e t Y. Vapour pressure measurements on piperidine-toluene mixtures showed that Raoult's Law was obeyed approximately, and the p a r t i a l CO pressures over piperidine-toluene mixtures could be r e a d i l y estimated. For p r a c t i c a l purposes, the p a r t i a l pressure of piperidine-toluene mixtures, 2.0-10.1M (neat) i n p i p e r i d i n e could be taken as that of pure p i p e r i d i n e . The reaction was started by turning the q u i c k - f i t B14 cone on the f l a s k side-arm R thus allowing the bucket and i t s contents, an accurately weighed amount of c a t a l y s t , to f a l l into the known volume 33 of so l u t i o n . At the same time the taps K and L were closed and both shaker and timer were started. Gas-uptake was followed by r e s u l t i n g d i f f e r e n c e s i n the o i l manometer D. A s l i g h t over-pressure of gas was put into the apparatus through Y up to the needle valve M and the o i l l e v e l s equalised once again by admitting gas through needle valve M and tap J . Thus, the reaction was followed by measuring changes i n the mercury l e v e l i n N with the aid of a cathetometer. Since the diameter of the tube N was known, the volume of gas consumed could be calculated and expressed as moles of gas uptake per l i t r e of solu t i o n . A rapid shaking rate together with the use of small volumes of sol u t i o n (2-5 mis.) i n a r e l a t i v e l y large indented reaction f l a s k ( 30 mis.) provided k i n e t i c r e s u l t s which were free of d i f f u s i o n c o n t r o l . For gas evolution experiments tap L was closed p r i o r to the f i n a l pressure adjustment, r e s u l t i n g i n a r i s e i n the l e v e l of mercury i n N. The reaction was started with taps K,J,L and needle valve M closed. Changes i n the o i l l e v e l s i n D were compensated by opening J and M and having a s l i g h t l y lower pressure of gas on the r i g h t hand side of valve M. A measurement of the drop i n the mercury l e v e l i n N against time could then be converted to moles of gas evolved. An a l t e r n a t i v e v a r i e t y of reaction f l a s k was used for c e r t a i n experiments. The side arm R was replaced by a side arm with a B7 socket attachment which could be connected to a sample bulb with stopcock. At the completion of a reaction, admission of gas through t h i s stopcock to the evacuated sample bulb enabled the gaseous reaction 34 products to be submitted for mass spectrographs or gas chromatographic an a l y s i s . Carbon dioxide could also be i d e n t i f i e d by absorption on soda lime which would be loo s e l y packed between glass wool i n one neck of a double-necked absorption f l a s k . Experiments on various samples of freshly-opened and old batches of i n d i c a t o r grade soda lime (Fisher S c i e n t i f i c Co. and Matheson, Coleman and B e l l Ltd.) revealed that ab-sorption of carbon dioxide was by no means instantaneous even at 1 atm. pressure. The experimental data shown i n Table I I I and depicted i n Fig.5 reveal that the absorption of C O ^ by soda-lime i s pressure dependent, although not i n any very obvious fashion, and that the presence of I^O i n the f l a s k causes a d e f i n i t e enhancement i n the rate of absorption. Thus, any experiment inv o l v i n g t y p i c a l amounts of ruthenium complexes ( i . e . nxlO ~\noles) with associated stoichiometric amounts of CC^ would l i k e l y exhibit very slow absorption of CC^ by the soda lime. As a r e s u l t , only f i n a l stoichiometries a f t e r long periods of time are l i k e l y to be meaningful i n the presence of t h i s absorbent. 2.4. Gas s o l u b i l i t y measurements The s o l u b i l i t y of gas i n solvents (e.g. carbon monoxide i n pi p e r i d i n e ) under s p e c i f i c temperature and pressure conditions could be determined using the gas-uptake apparatus and a reaction f l a s k containing a stopcock i n i t s neck. The solvent was degassed i n the usual manner and then the ent i r e system was brought under vacuum at room temperature. The tap on the f l a s k was then closed, and the f l a s k was placed i n the o i l bath at the desired temperature. The system was then evacuated to the f l a s k tap 3 5 Table I I I . Uptake of carbon dioxide by soda-lime absorbent Origin of sample Mattheson, Coleman and B e l l Fisher S c i e n t i f i c Co. Fisher S c i e n t i f i c Co. Mattheson, Coleman and B e l l 162.5 mm Hg. 3.77x10 moles/sec. -7 , * Mattheson, Coleman and B e l l 12.5 mm Hg. 1.71x10 moles/sec. (* 5.0 mis. of R^O i n the re a c t i o n f l a s k at 65°C. In the presence of K^O the rate of CO2 uptake r a p i d l y becomes too f a s t to measure as the C0 9 p a r t i a l pressure approaches % atm.) Pressure of CO^ I n i t i a l CO^ uptake rate 370 mm Hg. 6.86x10 7 moles/sec, 370 mm Hg. 1.53x10 7 moles/sec. 760 mm Hg. 5.16x10 7 moles/sec. / and f i l l e d with gas to the approximate pressure desired. The f l a s k tap was then opened and the pressure adjusted immediately to that required. Taps K and L were closed, the shaker started and the immediate uptake measured as described i n the previous section, allowing c a l c u l a t i o n of the gas s o l u b i l i t y . 2.5 Spectrophotometric k i n e t i c measurements Spectrophotometric measurements i n the u l t r a - v i o l e t and v i s i b l e range were used to study a number of reactions i n both aqueous and amine solutions. In order to study these reactions under i n e r t atmosphere conditions or carbon monoxide atmosphere a round-bottomed f l a s k with a spectrophotometric c e l l attached on a side-arm was em-ployed. De-gassing of the accurately measured volume of solvent was conducted by the usual "freeze-pump-thaw" technique. The s o l i d , accurately weighed, had been tipped into the f l a s k side-arm p r i o r to evacuation and the experiment was i n i t i a t e d by allowing solvent to run into the side-arm under the appropriate atmosphere. This technique required that the s o l i d was r e a d i l y soluble i n the solvent i n question, or that i t would dissolv e r e a d i l y upon gentle warming of the f l a s k (e.g. [HRu(CO) 3] n i n p i p e r i d i n e or p y r i d i n e ) . 2.6. Analysis of gaseous re a c t i o n Products For c o l l e c t i o n of gaseous reaction products a double-necked reaction f l a s k was employed, one neck being connected to an evacuated sample bulb having a stopcock. At the completion of the reaction, the stopcock on the bulb was opened momentarily and a gas-sample c o l l e c t e d . The sample was then subjected to either gas chromatographic or mass spectrographic 38 analysis. 2.7. Analysis of liquid and solid reaction products Products of the reactions in aqueous solution were studied by measuring the pH, chloride ion concentration, and conductivity of the filtered reaction solution. The solid products were investigated by microanalysis, including chloride content; infra-red spectroscopy i n an appropriate solvent and n.m.r. spectroscopy. Solid products from the amine carbonylation reactions were studied by the same techniques. The amide products from the carbonylation reactions were qualitatively and quantitatively examined by gas chromatographic analysis. 2.8. Analysis of chloride content in sol i d s "^ 7 The percentage of chloride in a sample was determined by a technique involving combustion of the solid in an oxygen atmosphere. An accurately weighed amount of solid was added to a small quantity of sucrose and then wrapped in a piece of ashless f i l t e r paper. A 10 ml. volume of 0.1 M. sodium hydroxide solution and a few drops of hydrogen peroxide were put into a 500 ml. conical flask equipped with a B24 quickfit socket. The stopper for this flask was designed so as to incorporate a platinum wire holder into which the sample and f i l t e r paper could be placed. The flask was flushed thoroughly with oxygen, the f i l t e r paper ignited and the stopper inserted while gradually tipping the flask upside-down in such a manner as to prevent any loss of solution or dampening of the sample. The sample was allowed to burn 39 completely and then the contents of the f l a s k were shaken u n t i l any cloudiness had disappeared. A f t e r 15 minutes the stopper was removed and the s o l u t i o n was bo i l e d for about 1 minute i n order to ensure removal of excess hydrogen peroxide. The s o l u t i o n was made s l i g h t l y a c i d i c using d i l u t e n i t r i c acid and the c h l o r i d e content was t i t r a t e d p otentiometrically using standardised s i l v e r n i t r a t e s o l u t i o n . The method seemed to give r e l i a b l e and reproducible r e s u l t s ; for example, S-benzyl thiazonium ch l o r i d e was treated i n t h i s manner using 0.01015 M. s i l v e r n i t r a t e s o l u t i o n (Calculated: CI, 17.49%. Found: OL, 17.47%). 2.9. Instrumentation V i s i b l e and u l t r a v i o l e t absorption spectra were recorded using ei t h e r a Perkin-Elmer 202 or a Cary 14 spectrophotometer, f i t t e d when necessary with a thermostated c e l l compartment. The s o l i d state r e -flectance spectrum of [HRu(C0)„] was measured on a Bausch and Lomb 600 3 n instrument. Infra-red spectra were recorded on a Perkin-Elmer 457 grating instrument, s o l u t i o n spectra using 0.1 mm. NaC£ c e l l s , and mull spectra on C s l plates. Nujol was used as the mulling medium. Laser Raman spectra were measured either on a Cary 81 instrument at U.B.C. or on a s i m i l a r machine belonging to Professor H. Kaesz at the U n i v e r s i t y of C a l i f o r n i a , Los Angeles. E.S.R. spectra were recorded on a Varian E-3 machine. Proton n.m.r. spectra were obtained using e i t h e r a Varian T-60 or XL-100 machine, and mass spectra were recorded on an Associated 40 Electrical Industries MS9 mass spectrometer. A Beckmann GC-2A unit with thermal conductivity type detector was used for gas chromatographic analysis with a wide variety of columns including carbowax and chromosorb. Detection of reaction products in piperidine and in reactions with other amines was also conducted using a Varian Aerograph-model 90P instrument and the Pennwalt 223/Gas Chrom R or Chromosorb 103 columns. Conductivity measurements were carried out using a Thomas Serfass conductivity bridge and dip-type conductivity c e l l . Measure-ments of pH were made using either a Corning model 12 or Orion 701 pH meter and a Fisher combination glass electrode. The determination of chloride ion concentration in solution was carried out with the same instruments, an Orion chloride ion activity electrode (model 94-17), and a reference electrode containing saturated KNO^  solution. Melting points were determined on a Fisher-Johns apparatus; t . l . c . experiments were performed using Eastman polythene-backed s i l i c a or alumina sheets, and handling of air-sensitive solutions was conducted on a typical double-manifold vacuum line with appro-priate glassware. Molecular weight determinations and analyses for ruthenium were carried out by Galbraith Laboratories Inc. of Knoxville, Tennessee, U.S.A. 41 Chapter 3 SYNTHESIS AND CHARACTERISATION OF •'[HRu(GO) 3.1. The synthesis of [HRu(CO) ] n The complex formulated as the hydridocarbonyl polymer 98 [HRu(CO)3]n was previously prepared by James and Rempel , although at that time i t was poorly characterised and tentatively described as [H xRu(CO) 3] n. In this work i t was synthesised by treating an aqueous solution of RuC^.SH^O with carbon monoxide over a period of several days. In order to carry the reaction to completion in as short a time as possible, carbon monoxide was bubbled through the solution maintained at approximately 80°C. in a flask equipped with a reflux condenser. The colour of the solu-tion changed from an i n i t i a l deep red to a pink-red within the f i r s t 24 hours, then changed to a yellow colour with accompanying precipitation of solids in the succeeding 48 hours. In every pre-paration that was attempted, orange-coloured solids collected on the walls of the flask and the lower part of the reflux condenser as well as in the bulk of the flask contents in association with 42 the brown polymeric product. It was discovered that the yield of polymer could be increased after 72 hours at 80°C. by a series of cooling and heating cycles in which the products were cooled to room temperature and re-heated to 80°C. for a short period of time. At the end of this procedure, f i l t r a t i o n of the reaction mixture yielded a nearly colourless f i l t r a t e and a precipitate of [HRu(CO)3]n which was washed successively with water and benzene. The benzene washings contained the orange-coloured by-products of this synthesis; these were generally produced in only milligram quantities when the usual ruthenium concentration of approximately 0.07M (1.0 gms. RuCA^ .SIL^ O in 50 mis.1^0) was employed. The remaining precipitate was dried in air and under vacuum to give a very powdery purple-brown solid in a maximum yield of about 80% recorded over the course of many preparations Some batches of this polymer were observed to change colour from purple-brown to black on standing, but this process could be avoided i f the compound was stored under argon in a v i a l wrapped with aluminium f o i l to protect the contents from light. The synthetic procedure appeared to generate [HRu(C0) 3] n at a slower rate at lower temperatures, and the optimum value of the ruthenium concentration was found to be about 0.1M. 3.2. Characterisation of |"HRu(C0)o] .. • J n The general lack of solubility of [HRuCCO)^]^ in most solvents strongly suggested i t s polymeric nature; also, when heated in air or vacuum i t showed no tendency to melt but became quite black in colour 43 in the 150-200°C. range. A typical analysis of the polymer for carbon and hydrogen gave the following results (Calculated for HRuC303:C,19.3;H,0.54%. Found:C,19.3;H,0.50%). A sample sent to Galbraith Laboratories for Ru analysis in particular, showed Ru=54.97% which is very close to the calculated value of Ru=54.31%. 108-110 Laser Raman spectroscopy has been used recently to reveal information regarding the structure of transition metal hydridocarbonyls, and in view of this fact a sample of [HRu(C0) 3] n was dispatched to Professor H. Kaesz at the University of California, Los Angeles. Unfortunately, exposure of a [HRu(C0) 3] n sample to the red, yellow and blue lasing lines of a Krypton-ion laser with power at the sample equal to approximately 50 mW. resulted in fluorescence of the polymer lasting for several hours. The use of an A r + laser and a high frequency purple line (same colour as the compound) might have solved this problem, but this type of laser was not immediately avail-able. Infra-red spectroscopic studies on KBr pellets of the polymer have consistently shown weak bands at the following approximate fre-quencies 1415,1270,1085,945,835,592,508 and 390 cm"1, as well as a broad band at 2000 cm ^ in the region normally associated with CO stretching frequencies (Fig. 6). Sometimes, a very weak, sharp band was observed as a shoulder on the high frequency side of this broad carbonyl band (Fig. 6). The detection of bridging hydrogen by observation of vibrational modes in the infra-red spectrum is 108 d i f f i c u l t owing to the weakness of such features. However, 45 Johnson, Lewis and Williams report a broad band centred at 1284 44 . 2 5 0 0 2 0 0 0 ^ 1 8 0 0 W A V E N U M B E R C C M ) Fig. 6. Infra-red spectrum (KBr disc) of [HRu(CO) ] in the carbonyl. stretching region. ' n . " 45 (AVj~40)cm 1 in the i . r . spectrum of H.Ru,(CO)10 which shifted to H H 1Z 902(Av^~20)cm_1 in the i . r . spectrum of D^Ru^(CO)12-The deuterated compound [DRu(CO)3]n was synthesised by an analogous procedure using D20 in the hope that the vibrational spectrum of this material could be used to assign the hydride vibrations. However, samples of this deuterated product gave infra-red spectra almost identical to those of the hydrido product. It is possible that the species was undergoing some exchange with non-deuterated solvent during the work-up procedure; such behaviour has been observed with the species D o0s o(C0) i n and D„Re„(CO) 0. 1 1 1 z 3 1U z Z o Nevertheless, a sample of [DRu(CO)3]n that had not come into contact with non-deuterated solvents at any stage of the synthesis or work-up gave the same result. This strongly suggests that the hydride and deuteride vibrations are not observed in the infra-red spectra of o these compounds. Exposure of a sample of [DRu(CO)3] n to the 6328A line of a He-Ne laser in a Cary 81 Laser Raman spectrometer resulted in decomposition to a black material, even at a reduced laser power of 15 mW. Measurement of the spectrum of this material at various machine sensitivities failed to reveal any deuteride modes. Samples of [DRu(CO)3] n did not appear susceptible to fluorescence. A solid-state reflectance spectrum of [HRu(CO) 3] n in the visible wavelength region showed one band at X =487 nm. (Fig. 7). This band ° max ° was not observed in any visible spectroscopic studies of [HRu(CO)3]n in those donor solvents in which i t could be dissolved (see sections 3.5 and 3.6). 46 47 Mass spectrometric examination of the polymer was conducted on several occasions but to no a v a i l . Peaks at m/e values of 28 and 44 which could be due to ISL,, CO and C0 2 were not r e a d i l y distinguished from the same peaks i n the background spectrum and no parent peak close to 186 (expected for the monomer) was observed. The low v o l a t i l i t y of the polymer and apparent decomposition at elevated temperatures have made i d e n t i f i c a t i o n by means of mass spectrometry impossible. The development of f i e l d desorption mass 112 spectrometry as a t o o l f or i n v e s t i g a t i n g i n v o l a t i l e compounds may s h o r t l y solve such problems. It should be noted at t h i s point that a number of hydridocarbonyl species of ruthenium are now well established and well characterised compounds. These include compounds of formula H^Ru^(C0)^ 2,H 2Ru^(CO).^, H 2Ru^(C0)^g as well as the species formed i n the c a t a l y t i c polymethylene 113 synthesis (see section 3.4) and t e n t a t i v e l y proposed to have the formulae H^Ru^CO)^ and H 2Ru 3 (CO).^. These species appear to be generally soluble i n organic solvents and do not resemble the h y d r i -docarbonyl polymer i s o l a t e d i n t h i s work. Evidence for the presence of hydride i n the polymer was hard to obtain. Repeated i n v e s t i g a t i o n of saturated solutions of [HRu(CO) i n amine solvents f a i l e d to give any high f i e l d "*"H n.m.r. sig n a l s , however hydrogen exchange could be a problem. A deep orange-coloured solu t i o n , produced by warming the polymer i n methyldiphenylphosphine under a nitrogen atmosphere, was transferred to an n.m.r. tube under argon and a search for high f i e l d "Hi n.m.r. signals gave two m u l t i p l e t s near T=30 (Fig. 8). This complex pattern l i k e l y r e s u l t s from coupling 49 of the hydride with the phosphine which is presumably co-ordinated to ruthenium; coupling to ruthenium is possible (isotopes with 99 nuclear spin are with natural abundance = 12.72% and 1=5/2, and also ^Ru"^ 1 with natural abundance = 17.07% and 1=5/2)11^, but the sensitivity of the ruthenium isotopes in question is some 1000 times less than that of "'"H and to our knowledge coupling of to Ru has not been reported. The concentration of ruthenium in this solution, calculated in terms of monomer [HRu(C0)3], was approximately 7.0 x 10"2M. The addition of aqueous or alcoholic acid solutions to metal hydrides frequently results in complete replacement of the hydride ligands1"'"^ as hydrogen (eq. 3.1), and this reaction can be used to determine the number of hydride ligands by measurement of the evolved hydrogen. In an attempt to use a similar technique, the polymer [HRu(C0)J was added to 3 n MX H + H + > MX+ + EL (3.1) n n I a DMA solution of p-toluenesulphonic acid (source of H +), and the resulting gas evolution was studied using the technique described in section 2.3. The experiment was conducted under a nitrogen atmosphere at 75°C. The [HRu(C0) 3] n monomer concentration was -3 -2 9.44 x 10 M. in the presence of a 1.87 x 10 M. solution of p-toluenesulphonic acid in neat DMA solvent. Although gas evolution was observed, i t occurred to the extent of 1 mole of gas evolved per monomer mole of [HRu(C0)„] within 500 seconds and was followed by 3 n 50 further slow evolution of gas (1.6:1 evolution per monomer mole of [HRu(C0) 3J n a f t e r 1 hour). A s i m i l a r experiment was conducted under argon i n order to c o l l e c t a sample of the gas above the reaction mixture a f t e r 18 hours. The sample was submitted f o r high r e s o l u t i o n mass spectrometry and c l e a r l y showed an m/e=2 peak at l e a s t 2.5 times higher than the same peak i n the background spectrum. The unexpectedly large volume of evolution observed i n the previous experiment po s s i b l y resulted from loss of co-ordinated CO when [HRu(C0) 3] n was dissolved i n DMA. However, the mass spectrum showed no enhancement of the m/e=28 peak over that observed i n the background spectrum and addition of [HRu(C0) 3] n alone to DMA under nitrogen resulted i n no observable gas evolution. This l a s t r e s u l t suggests that some species HRu(C0) 3(DMA) n i s i n i t i a l l y formed i n so l u t i o n . The a b i l i t y of DMA to act as a co-ordinating ligand i s well-established"'""'"^ and the chemistry of [HRu(CO) 1 i n DMA w i l l be 3 n discussed i n section 3.6. The use of ruthenium tribromide or t r i i o d i d e as a l t e r n a t i v e s t a r t i n g materials appeared to o f f e r no p a r t i c u l a r advantage for the synthesis of [HRu(C0) 3J n. The deep-brown insoluble powder l e f t at the end of the experiments, judging by the i . r . spectrum and micro-analysis, seemed to be the same compound as that produced from ruthenium t r i c h l o r i d e . For example, the compound prepared from ruthenium tribromide gave the following data; calculated for HRuC303:C,19.30%. Found:C,20.64%. The r e s u l t i s s l i g h t l y high, but compares with some r e s u l t s obtained when using the t r i c h l o r i d e 51 and could possibly be due to r e s i d u a l benzene i n the s o l i d . 3.3. The f i n a l r eaction s o l u t i o n The aqueous f i l t r a t e remaining a f t e r c o l l e c t i o n of s o l i d products, including [HRu(CO)g] n> varied i n colour from pale yellow to nearly c o l o u r l e s s . If repeated heating and cooling cycles were used to increase the y i e l d of [HRu(CO)gl n» at the end of the pre-paration ( i . e . a f t e r 3 or 4 days) the f i l t r a t e tended to have l e s s colour than i f no heating and cooling cycles were applied. Various s o l i d s could be p r e c i p i t a t e d from t h i s f i l t r a t e , which was reasonably a i r - s t a b l e , by the addition of i o n i c s a l t s containing large cations (e.g. Ph 4As +C£ , tBu 4N +I~, Me 3S +I , e t c . ) . The most s a t i s f a c t o r y technique appeared to be the a d d i t i o n of s o l i d tetraphenylarsonium ch l o r i d e (for example) to the s t i r r e d aqueous f i l t r a t e u n t i l the f i r s t signs of cloudiness occurred. In t h i s way a f i n e yellow p r e c i p i t a t e was obtained i n a 0.1-0.2 gm. y i e l d (maximum=0.21 gms.) when s t a r t i n g with 1.0 gms. ruthenium t r i c h l o r i d e . This p r e c i p i t a t e (I) was f i l t e r e d and subjected to further s c r u t i n y by analysis and i n v e s t i g a t i o n of i t s i . r . spectrum. The remaining f i l t r a t e frequently deposited more s o l i d (II) upon standing. A comparison of the n u j o l mull i . r . spectra of t y p i c a l species (I) and (II) with data recorded by Cleare and G r i f f i t h s " * " ^ (Table IV) for a range of chlorocarbonyl ruthenium complexes proved u s e f u l . The species ( I ) , p r e c i p i t a t e d from the f i l t r a t e by Ph 4AsC£ a f t e r a reaction time of nearly 3 days, revealed two peaks (see footnote a_ i n Table IV) i n the carbonyl 52 Table IV. I.R. data f or chlorocarbonyl complexes of ruthenium (from r e f . 105) Complex Stretching frequencies (cm "*") Cs 2Ru(C0)C£ 5 Cs 2Ru(C0)(H 20)C£ 4 Cs 2Ru ( C O ^ C ^ CsRu(C0) 3C£ 3 i 2042(s),2015(vs),545(m),478(vw),320(vs), 272(m). 1951(vs),649(mb),578(m),567(m),520(w,br), 311(s),289(sh,m). 2036(vs),1935(vs),642(m),572(m),541(w), 500(m),312(s),301(s),268(m),259(m). 2137 (s) , (2074 (s) , 2061 (s) , 2047 (s)),616 (m), 575(m),475(m),472(w),315(s),282(s). (* nujol mull spectra) a. The three peaks at 2074,2061 and 2047 cm x i n the spectrum of CsRu(C0) 3C£ 3 r e s u l t from s o l i d - s t a t e s p l i t t i n g e f f e c t s and were also observed i n a sample of CsRu(C0) 3C£ 3 prepared i n t h i s work (section 5.8). I.R. spectra of CsRu(C0),C£- i n aqueous solutions -1 gave peaks at 2145 (s) and 2075 (s) cm i n close agreement with those reported by Cleare and G r i f f i t h 1 0 5 at 2141(s) and 2077(s) cm"1. The s o l i d - s t a t e s p l i t t i n g e f f e c t s and exact p o s i t i o n s of the carbonyl peaks are l i k e l y to be cation dependent; for example, the CO 2 6 —1 stretching frequencies are reported at about 2060 and 1990 cm for (NH 4) 2Ru(C0) 2C£ 4(cf. the CO stretching frequencies of the Ph 4As + s a l t shown i n F i g . 10). 53 stretching region that suggested the presence of Ru(CO) 3C£ 3 (Fig. 9). The remaining peaks in the "fingerprint" region of the spectrum can be assigned to various stretching and bending modes of the Ph^ As"*" group."'""'"7 The addition of Ph^AsCJi to the reaction mixture after only 2 days resulted in a mustard-yellow precipitate (III). The i . r . spectrum of (III) in the carbonyl stretching region (Fig. 10) showed two peaks in the 1900-2100 cm 1 range which can be attributed to the 2-species cis-Ru (CO^CJ^ (see footnote a in Table IV) and a third peak of low intensity above 2100 cm 1 that i s probably due to Ru(CO) 3C£ 3 . Analysis of compound (I) gave the following data (Found: C,46.87; H,2.87%) and comparison with a number of possible formulae suggested that the compound was largely Ph^As.Ru^O^CJi^. (Table V). The analysis of compound (I) is slightly low in % carbon when compared with the formula (Ph^As)Ru(CO)-jCJ^; this could be due to some traces of water, although no evidence for the presence of IL^ O could be found in the i . r . spectrum. It should be noted that on. one occasion a precipitate gave a very good analysis for the tricarbonyl species (Found:C,47.96; H,2.83%), but peaks in the infra-red spectrum (nujol mull) at 2123(s), 2049 (s) and 2041 cm "*"(s) suggested a mixture of products. The preparative method of Cleare and G r i f f i t h 1 0 5 for CsRu(CO) 3C£ 3 (section 5.8) involves precipitation of the cesium salt by the addition of cesium chloride to the reaction mixture after lengthy refluxing of a solution of ruthenium trichloride (2 gms.) in concentrated hydro-chloric acid (15 mis.) and 90% formic acid (20 mis.). Following this procedure, but using Ph^AsCA instead of CsC£, gave a yellow precipitate 54 F i g . 9. Infra-red spectrum (nujol mull) -of p r e c i p i t a t e (I) i n carbonyl stretching region. • . -the 55 Fig. 10. Infra-red spectrum (nujol mull) of precipitate(III) in the carbonyl stretching region. f 56 Table V. Formula Ph 4As.Ru(CO) 3C£ 3 (Ph 4As) 2Ru(CO) 2C£ 4 (Ph4As)2Ru(CO)(H20)C£ Possible formulae and analyses for  Compound (I) %C %H 48.03 2.96 56.33 3.76 56.81 4.06 57 with an analysis very s i m i l a r to that of compound ( I ) . (Found: C,46.42;H,3.30%). The compound ( I I ) , which resulted from the f i l t r a t e acquired i n c o l l e c t i n g compound ( I ) , showed an i . r . spectrum (Fig. 11) that was quite complicated i n the carbonyl stretching region and was probably due to a'mixture of species 2- 2-containing ions such as Ru(C0) 2C£ 4 ,Ru(CO)(H 20)C£ 4 and Ru(C0) 3C£ 3 . The a d d i t i o n of trimethylsulphonium iodide to the f i l t r a t e remaining at the end of the [HRu(CO) 3] n preparation deposited a yellow powder (IV), which was characterised by i t s i . r . spectrum i n the carbonyl region (Fig. 12) as containing the Ru^ 1(CO) 3 e n t i t y . These t r i c a r b o n y l species are generally distinguished by the presence of only two C-0 s t r e t c h i n g modes i n the i . r . spectrum, although s p l i t t i n g of the peaks i s possible i n s o l i d state spectra (see Table IV). A c i s - d i c a r b o n y l species also shows two C-0 stretching modes, but the symmetric and asymmetric stretches occur at lower frequencies than those of the corresponding t r i c a r b o n y l s , i n which each carbonyl group w i l l have a lower percentage of the a v a i l a b l e metal t 2 electrons. The necessity to minimise competition by the carbonyl groups for the t 2 electrons probably r e s u l t s i n the formation of the c i s - and fac- forms of the d i - and t r i - c a r b o n y l s r e s p e c t i v e l y . Species 2- 2-containing the Ru(C0)C£,- and Ru(CO)(H 20)C& 4 moieties generally exhibit the expected s i n g l e C-0 s t r e t c h or a peak i n the appropriate region that i s s p l i t into a very close doublet. Hence, i n our present case (Fig. 12) the r e s u l t i n g spectrum with two peaks i n the 2000-2100 cm range indicates a Ru(CO)^ species. The micro-analysis of the yellow powder 58 F i g . 11. Infra-red spectrum (nujol mull) of p r e c i p i t a t e ( I I ) i n the carbonyl stretching"'region... _ . ; 59 Fig. .12..-' Infra-red "spectram:"(nujol - mull) of precipitate (IV) in the carbonyl stretching region. 60 (IV) gave the following data (Found: C,11.73;H,1.65%) and comparison with a number of possible formulae (Table VI) suggests that the compound i s almost c e r t a i n l y Me^S.Ru^O)^!^. A further examination of the nujol mull i . r . spectrum of compound (IV) (Fig. 13) f a i l e d to 2 -reveal the c h a r a c t e r i s t i c Ru-C-0 bending mode of RutCO^I^ at 613 cm \ but did show the Ru-C-0 bending mode at460 cm "*"(m), the Ru-I stretching mode at 590 cm "*"(s) and the Ru-C stretching mode at 552 cm 1(m) which are c h a r a c t e r i s t i c of R u ( C 0 ) 3 I 3 s p e c i e s . 1 0 5 The a d d i t i o n of tetrabutylammonium iodide to the aqueous f i l t r a t e remaining at the completion of the [HRu(CO) ] ^ synthesis resulted i n p r e c i p i t a t i o n of a tan-coloured powder. The conditions of t h i s preparation involved the use of 1 gm. of ruthenium t r i c h l o r i d e i n 50 mis. of H 20, generating 0.58 gms. (82% y i e l d ) of [HRu(C0) 3] n and producing 0.18 gms. of p r e c i p i t a t e (5.8% y i e l d based on^u^NRuCCO^I^) upon addition o f t f i u 4 N + I to the aqueous f i l t r a t e . Unfortunately, the t e t r a b u t y l ammonium s a l t could not be characterised owing to i t s very hygroscopic nature. The tan p r e c i p i t a t e very r a p i d l y becomes an orange-coloured o i l upon standing i n a i r . This c h a r a c t e r i s t i c of the 118 t e t r a b u t y l ammonium s a l t was observed previously by James and Rempel. In another experiment the [HRu(C0) 3] n synthesis was continued for 70 hours and the reaction mixture was cooled and f i l t e r e d to pro-duce a yellow s o l u t i o n as f i l t r a t e . The f i l t r a t e was reduced to a small volume (several mis.) by pumping under vacuum and cesium c h l o r i d e was added to the l i q u i d u n t i l p r e c i p i t a t i o n occurred. The s o l i d had an i . r . spectrum i n n u j o l mull which indicated that Cs 2Ru(C0) 2C£ 4 had been produced. 61 Table VI. Possible formulae and analyses for Compound (IV) Formula %C %H Me 3S.Ru(CO) 3C£ 3 16.32 2.04 Me 3S.Ru(CO) 3C£ 2I 15.65 1.96 Me 3S.Ru(CO) 3C£I 2 13.05 1.63 Me 3S.Ru(C0) 3I 3 11.19 1.40 (Me 3S) 2.Ru(C0) 2I 4 11.71 2.19 62 63 Thus, i t would appear that the yellow solutions c h a r a c t e r i s t i c of the l a t e r stages i n the synthesis of [HRu(CO) 1 contain a mixture of ruthenium (II) chlorocarbonyls ( i . e . Ru(C0)2 and Ru(CO) 3 moieties). When conditions are optimal for maximum y i e l d of [HRu^O)^]^, that i s , 3 to 5 days at 80°C. under CO followed by successive heating and cooling cycles, the aqueous f i l t r a t e contains mostly ruthenium t r i c a r b o n y l species and l i t t l e dicarbonyl can be detected i n any product p r e c i p i t a t e d by the use of a large counter-cation. The examination of these p r e c i p i t a t e s by any further means was l i m i t e d by t h e i r general i n s o l u b i l i t y i n water. The Ph^AsRu^O^CJ^ complex appeared to dissolv e slowly i n hot concentrated hydrochloric acid, but addition of excess cesium chloride resulted i n r e - p r e c i p i t a t i o n of the o r i g i n a l complex, presumably as the r e s u l t of a common-ion 119 e f f e c t . Hence, the so l u t i o n i n concentrated acid did not provide an a l t e r n a t i v e route to the synthesis of the water-soluble CsRu(CO).jC£ complex, (see section 5.8). 3.4. Benzene-soluble products The preparat ion of the polymer [HRu(CO) 3]^ as described also resulted i n some orange-coloured water-insoluble s o l i d s (A) , p a r t i c u l a r l y i f the synthesis was stopped a f t e r approximately 36 hours. At t h i s stage, cooling the cl e a r yellow s o l u t i o n under CO produced t h i s apparent mixture of s o l i d s . When CO was replaced by N 2 or H 2 t n e c o o l i n g process appeared to generate much smaller y i e l d s of the s o l i d products. If the synthesis was conducted over longer periods of time i n order to maximise 64 the y i e l d of [HRu(CO) 3] n > the y i e l d of (A)was only a few milligrams from 1 gm. of ruthenium t r i c h l o r i d e . This mixture of s o l i d s (A) was extracted into benzene and then freeze-dried from that solvent to y i e l d a f i n e powder which varied i n colour from orange to brick-red (maximum y i e l d = 0.4 gms. from 1.0 gms. ruthenium t r i c h l o r i d e ) . 98 O r i g i n a l l y , James and Rempel" , considered product (A) to be a mixture of the known compounds Ru^tCO)^ a n < i ^^u^^co^2' T h i s conclusion was based on i . r . spectroscopic evidence. Kaesz and co-120 workers have also prepared such a mixture by means of the previously 44 reported l i t e r a t u r e preparation for a compound which was thought ed to be H^Ru^(CO).^ • The s o l u t i o n i . r . spectrum of the product produc 120 by Kaesz revealed carbonyl bands at 2081 (s), 2067(vs), 2063(s), 2030(s), 2024 (s) and 2009(m) cm "*"s and chromatography on s i l i c a gel with n-hexane indicated about a 1:1 mixture of Ru-^CO)^ a n < i a-H^Ru^CO)^- Pure a-H^Ru^ (CO)^^ c a n be synthesised by hydrogenation of Ru-^CO)^ i n r e f l u x i n g octane"'""'"0, and t h i s l a t e r report discounts e a r l i e r suggestions 44 by Johnson et. a l . that a 3-isomer of the hydridocarbonyl can be pre-pared. We followed the preparative route of Kaesz"'""'"0, f o r H^Ru^ (CO).^» using 0.36 mmoles. of R u ^ C O ) ^ (prepared as i n section 2.1.1) i n 80 mis. of n-octane which had been previously dried over molecular sieve 5A under an argon atmosphere. Upon heating the mixture to r e f l u x under a rapid stream of hydrogen, the i n i t i a l orange suspension became an orange s o l u t i o n changing to a deep yellow s o l u t i o n within 15 minutes. The change i n colour appeared to have stopped a f t e r 30-40 minutes at which time sampling of the s o l u t i o n indicated the los s of carbonyl peaks i n 65 the i . r . due to Ru,(CO) , p a r t i c u l a r l y the absorption at 2063 cm X , Iz and the production of f i v e maxima due to H^Ru^CO)^- The s o l u t i o n was l e f t to cool overnight under a steady stream of and, on the following morning, a deep yellow p r e c i p i t a t e was evident. The p r e c i p i t a t e was c o l l e c t e d and r e c r y s t a l l i s e d from a 1:1 mixture of n-hexane and dichloromethane. Microanalysis of the product gave s a t i s f a c t o r y r e s u l t s (Calculated for H J l u ^ C ^ O ^ : C,19. 34 ;H, 0.54%. Found:C,19.59;H,0.60%) and the i . r . spectrum i n cyclohexane s o l u t i o n showed maxima at 2080(s), 2065 (vs), 2030(m), 2025(s) and 2010(s) cm "'"in agreement with values reported i n the literature."'"''" 0 Subsequent examination of the by-products (A) from the [HRu(CO) ]^ synthesis suggested that (A) was a mixture, but that i t did not consist of Ru^CCO).^ a n d H^Ru^CO)^ only. At t h i s point i t i s worth considering the properties of the presently known carbonyl hydrides of ruthenium. Table VII summarises some important data. The observations made on mixture (A) are summarised below. F i r s t , the orange product which was obtained many times consisted of a v a r i a b l e mixture of hexane-soluble (B) and hexane-insoluble (C) s o l i d s . The s o l u t i o n i . r . spectra of (B) i n hexane (Fig. 14) and (C) i n d i c h l o r o -methane (Fig. 15) c o n s i s t e n t l y reveal a complex pattern of peaks i n the carbonyl stretching region. The presence of maxima above 2100 cm 1 and at approximately 1950 cm "*" (Fig. 15 only) would seem to r u l e out (C) being any of the compounds l i s t e d i n Table VII. Samples of compound (C) from several d i f f e r e n t syntheses gave carbon analyses of 17.80-18.00%, and f a i l e d to give any high f i e l d "*"H n.m.r. signals even when employing saturated solutions i n C^D^. Two separate samples of compound (C)were 66 Table VII. Hydrido and deutero-carbonyls of ruthenium Compound vCO(cm _ 1) n.m.r.data Analyses: C% H% Ru% a-H.Ru.(CO) 2081s,2067s,2030m,3 x=27.98(s) 19.4 0.5 54.3 2024s,2009w.(ref.110) (ref.110) B-H,Ru,(C0) l 2 2080s,2068s,2056s, 3 T=18.6(S) 19.4 0.5 54.3 2034m,2027s,2008w (ref.44) (ref.44) D Ru,(C0) 2 2079s,2067s,2029m,a - 19.2 - 54.2 2023s,2008w(ref.110) H R u , ( C 0 ) n 2083s,2068s,2056s, a T=19.1(S) 20.3 0.3 52.5 2033m,2026s,2008w, (ref.44) 1880w (ref.44) D Ru,(CO) 2079s,2063s,2056s, a - 20.2 - 52.4 2033m,2019s,2003w, 1880w (ref.44) H 0Ru,(C0)* 2060s,2054s,2008mb - 19.6 0.18 55.1 2 6 1 8 (ref.121) H 4Ru 3(CO) 1 0 (ref.H3-no d e t a i l s given) 20.4 0.68 51.6 H 2 R u 3 ( C 0 ) n 21.5 0.33 49.5 Ru (CO) „ 2060s,2030s,2010(m) C - 22.5 - 47.5 (ref.105) I.R. spectra were recorded i n a: cyclohexane, b: unspecified, c: n-hexane. A l l n.m.r. were recorded i n CDCJi^. * D 2Rug(C0)^g has been made by the same workers. The i . r . spectrum shows l i t t l e d i f f e r e n c e from that of the hydride complexes but can be distinguished by the mass spectrum which shows a molecular ion with an m/e value two u n i t s higher than H oRu,(C0) 1 o. I O l o 67 F i g . 14. - S o l u t i o n - i n f r a - r e d -..spectrum. ;of compound (B) i n n-hexane. C a l i b r a t i o n was conducted by using' the free C0" v i b r a t ion^a t.2147 cm-1.- A grating" change occurs at 2003 cm' 68 2 1 4 7 2003 2029 1 9 5 0 Fig-.; 15. Solution ^infra-red spectrum of: compound' (C) i n dichloromethane. C a l i b r a t i o n was conducted by using the free CO v i b r a t i o n at -2147 cm-!'. - -A • grating- change occurs at 2003 cm - 1. 69 submitted to Galbraith Laboratories f o r analysis and gave reasonably consistent r e s u l t s of 35.40 and 36.00% ruthenium; a molecular weight of 440 was determined i n benzene so l u t i o n . When a sample of the mixture (A) was heated to 60°C. under vacuum i n a sublimation tube, a small amount of orange c r y s t a l l i n e material c o l l e c t e d on the cold finger and the remaining material became black. The sublimate was soluble i n heptane giving the c h a r a c t e r i s t i c i . r . spectrum of Ru^CCO)^ a n c ^ a sample of (A) submitted for mass spectrometry showed a parent peak at m/e=640 corresponding to Ru^CCO)^"1" with peaks r e s u l t i n g from s i n g l y charged t r i n u c l e a r species down to Ru^ + also being observed. No peaks at higher m/e value were apparent, at l e a s t up to m/e=1000. This evidence ind i c a t e s that Ru^CCO)^ i - s a minor component of the mixture (A) and of (B) from the i . r . evidence. Attempts to carry out a separation of the components of (A) by chromatography using a number of solvent mixtures on both s i l i c a and alumina t . l . c . plates were unsuccessful (e.g. CH^CJ!^ eluent on alumina gave one spot with R =0.80, acetone eluent on s i l i c a gave one spot with R =0,71). The replacement of H„0 by D„0 i n the r Z Z synthetic procedure enabled us to prepare the deuterated equivalent of (A), however no dif f e r e n c e s i n the carbonyl stretching region of the so l u t i o n i . r . spectrum were apparent, which may i n d i c a t e the absence of hydride (deuteride) ligands (cf. footnote i n Table VII). Since measurements on f i n a l aqueous solutions (see also section 4.1) ind i c a t e that a l l of the ch l o r i d e ion i n the s t a r t i n g material remains i n s o l u t i o n during the synthesis of [ H R u ( C O ) , i t might appear 70 that the mixture (A) contains no c h l o r i d e . However, since the y i e l d of (A) i s small, any c h l o r i d e bound i n (A) may not s i g n i f i c a n t l y a f f e c t the concentration observed i n the f i n a l s o l u t i o n (chloride analyses were v a r i a b l e and inconsistent giving r e s u l t s from 5 to 10%). A product obtained from ruthenium tribromide did give a some-what lower carbon analysis (16.01%) which i s consistent with the presence of some halogen i n the mixture. In summary, the evidence suggests that (A) consists of R u ^ C O ) ^ i n small amounts and one or more new ruthenium carbonyl species. The presence of the known h y d r i -docarbonyls (Table VII) seems u n l i k e l y on the basis of the i . r . , mass spectrometric and ruthenium analysis data. 3.5. [HRu(C0) 3] n s o l u t i o n chemistry i n pyridine In an attempt to obtain more information concerning the nature of [HRu(C0) 3] n, solutions of t h i s polymer i n various donor solvents were studied. The ruthenium atom i n [HRu(C0).j] n i s formally i n oxidation state (+1). A few such complexes are known i n the s o l i d state, and these are diamagnetic with metal-metal bonding (e.g. species of general formula 47 [Ru(00)2(RCOO)L]2 ); however, r e l a t i v e l y l i t t l e i s known about ruthenium (I) i n s o l u t i o n . The most notable example would appear to be the brown a i r - s e n s i t i v e solutions obtained by hydrogen reduction of blue ruthenium 21 (II) solutions i n DMA. These brown solutions undergo stoichiometric oxidation to ruthenium (III) and react with carbon monoxide to give a Ru 1(C0)2 species which can be oxidised s t o i c h i o m e t r i c a l l y to Ru^ 1(CO)^ (see equations 3.2,3.3,3.4 and 3.5). An incomplete c h a r a c t e r i s a t i o n of a complex Ru^C^(HDMA), where (HDMA) i s believed to be protonated DMA, 71 y.Kn11 + ^ ^ 2 R U 1 + 2H + (3.2) 2H + + Ru 1 + h02 > Ru11Z+ H 20 (3.3) Ru 1 + 2C0 5> Ru I(CO) 2 (3.4) H + + Ru I(CO) 2 + %02* > R u I : E ( C O ) 2 + %H20 (3.5) isolated from the brown DMA solutions suggested that the ruthenium (I) species was dimeric with metal-metal bonding. The polymer [HRu(CO).j]n w i l l dissolve in pyridine to a limited extent; for instance, at 60°C. a saturated solution (concentration approximately 0.04 M. in ruthenium monomer) can be obtained by shaking the solid with pyridine under vacuum or an inert atmosphere. At concentrations approaching saturation point, an orange-yellow solution was produced within one hour. Several attempts were made to isolate a solid from these pyridine solutions. The addition of excess oxygen-free, anhydrous ether under an inert atmosphere was found to result in precipitation of a yellow solid. Unfortunately, this solid proved d i f f i c u l t to collect and attempts using conventional Schlenk-type apparatus only resulted in "glue-like" products upon f i l t r a t i o n of the mixtures under reduced pressure. In an alternative procedure, pyridine was removed by pumping under vacuum and ether was added to the resulting brown, oi l y material. Surprisingly, the material appeared to be almost completely soluble in ether and upon removal of the solvent under vacuum gave a brown powder. The infra-red spectrum (nujol mull) of this powder showed a very broad carbonyl peak in the region 1900-2000cm"1 and peaks at 1750(m,br), 1220(w), 1070(w), 760(m) and 700(m) 72 cm 1 which are presumably due to pyridine, although several important bands reported i n the i . r . spectra of both free and co-ordinated -1 122 pyridine (notably those at 3000, 1600 and 600 cm ) appeared to be absent. Although the usual spectroscopic techniques were employed i n an attempt to characterise the pyridine solutions, d i f f i c u l t i e s p e r s i s t e d . -3 5 A 4.56 x 10 M. s o l u t i o n (based on monomer [HRu(C0) 3]) i n pyridine-d f a i l e d to show any high f i e l d metal hydride resonance i n the n.m.r. spectrum even when a Fourier Transform spectrum was recorded on a 100 MHz machine. The low concentration of ruthenium should not have presented any problems under such conditions. Solution i n f r a - r e d studies i n pyridine were hampered by the presence of several strong bands i n the 1850-2100 cm 1 region. These bands were observed i n spectroscopic grade solvent both before and a f t e r d i s t i l l a t i o n but 123 are not observed i n previously reported p y r i d i n e spectra. The solutions of [HRu(C0) 3] n i n pyridine under i n e r t atmospheres (argon or nitrogen) gave a continuum i n the u . v . / v i s i b l e absorption spectra down to the pyridine solvent cut-off point at 305 nm. The addition of tetraphenylarsonium c h l o r i d e to the pyridine s o l u t i o n of [HRu(C0) 3] n gave no p r e c i p i t a t i o n . A p r e c i p i t a t e d s a l t might have been expected had the abstraction of hydrogen from the metal (as shown i n general i n equation 3.6)^^ been taking place. i . e . base + MH > base H + + M~ (3.6) (e.g. py) (e.g. pyH +) Several experiments were conducted i n order to f i n d out i f any 73 gas was evolved upon dissolving [HRu(CO)3]n in pyridine under inert atmospheres. Despite repeated de-gassing of the solvent by the freeze-pump-thaw technique (section 2.3) and temperature e q u i l i -bration of the solvent by prolonged shaking of the reaction flask at the operating temperature of the o i l bath, the addition of the solid [HRu(CO) 3] n to the pyridine solvent was always accompanied by a sudden apparent evolution of gas. This evolution effect had essentially ceased within 600 seconds, even though complete solution of the polymer was known to require longer time periods at the tem-peratures of 60-70°C. employed in these experiments. Similar blank experiments were then conducted using pyridine alone. The de-gassing and temperature equilibration processes were the same as before and, upon addition of the empty glass bucket, which previously had held the solid material, an almost identical sudden evolution was observed (Fig. 16). The data in Table VIII indicate that l i t t l e distinction can be made between experiments in which [HRu(C0) 3] n is present or absent in terms of the amount of gas evolution detected. This evolution effect seems to be a characteristic of certain solvents, also being observed with piperidine, and in some way must be related to a temperature change or a "nucleation effect"(the evolution of inert gas molecules absorbed on the glass or polymer complex surface) which occurs when the glass bucket i s dropped into the solvent. It seems unlikely that any small temperature changes would account for the observed evolution (in terms of the solubility of nitrogen or argon in pyridine). In any event, after the f i r s t 100 seconds this effect i ~T 1 1 1 1 1 — — i r F i g . 16.; -.- Gas evolution experiments (•(!) i n Table V I I I ) ; ( A ) A.0x10 moles [HRu(CO) 3] n i n 10 mis. pyridine at 65°C, (• )' 10 mis. pyridine at 65°C. 75 Table VIII. [HRu(CO) 3J n i n pyridine - r e s u l t s of gas evolution studies Experiment [HRu(CO) 3] nxl0 5m. T°C. P^.mm Hg Observed gas evolution x 10 m. (1) Solution 4.00 65 613 3.52 Neat pyridine - 65 658 3.50 (2) Solution 6.21 66 661 3.54 Neat pyridine - 66 668 2.75 (3) Solution 5.87 66 667 4.28 Neat pyridine - 66 670 2.29 76 had v i r t u a l l y ceased and observations for the succeeding 3000 seconds, during which s o l u t i o n of the complex occurs, f a i l e d to show any further evolution of gas. Thus, i t appears that the polymer [HRu(C0) 3] n d i s s o l v e s i n pyridine at 60-70°C. under argon with the retention of a l l three carbonyl groups within the co-ordination sphere of each ruthenium atom, and the orange-yellow pyridine solutions l i k e l y contain species such as [HRu(C0) 3(py) x] n or [ ( R u ( C 0 ) 3 ( p y ) x ) " ] n . The [ HRu (CO) 3 ^ / p y r i d i n e system was thought to have p o t e n t i a l for the production of pyridine aldehydes by a d i r e c t formylation process. Pyridine-2-aldehyde (picolinaldehyde) has been prepared by formylation of pyridine using N-methyl formanilide i n the presence of 124 phosphoryl chloride, but only low y i e l d s were obtained. Pyridine _? solutions of [HRu(C0)„] at [Ru] = 2 x 10 M., T=75°C. and t o t a l 3 n pressure = 1 atmosphere (carbon monoxide), however, showed no evidence _2 of any carbonylation. In one experiment at [Ru] = 0.93 x 10 M., T=60°C. and t o t a l pressure = 685 mm. Hg., the uptake of carbon monoxide corres--3 -1 ponded to 3.06 x 10 M. atm., even l e s s than the measured s o l u b i l i t y -3 -1 of CO i n pure pyridine which was found to be 3.66 x 10 M. atm at 60°C. These negative r e s u l t s were supported by a v.p.c. experiment using a chromosorb column, which indicated that none of the three possible pyridine-aldehydes had been generated i n s o l u t i o n . Prolonged exposure of [HRu(C0) 3] n solutions i n pyridine (made up under vacuum or i n e r t atmosphere) to carbon monoxide f a i l e d to show any change i n colour and the v i s i b l e spectra remained e s s e n t i a l l y unchanged. Pyridine solutions of [HRu(C0)„] (prepared under vacuum or i n e r t 77 atmosphere) were air-sensitive. At room temperature the solutions changed colour from orange-yellow to deep-green within a matter of hours, and at 70°C the change appeared to be complete within 400 seconds. The results of uptake experiments under oxygen (e.g. Fig. 17) showed fi n a l stoichiometries of % mole of gas taken up permole of ruthenium with concomitant spectroscopic changes in the vis i b l e region as exemplified by Fig. 18. Neither the gas uptake nor the spectros-copic data analyse for a pseudo 1st order rate process. There are several possible explanations for this reactivity toward 0^, (a) pyridine being converted to pyridine-N-oxide, (b) oxidation of co-ordinated CO to C O 2 , (c) formation of a dioxygen complex or (d) con-version of ruthenium (I) to some higher oxidation state, presumably Ru 1 1 1. The f i n a l stoichiometry of the reaction appears to rule out (a) since i f such ligand oxidation occurred, then a catalytic process would almost certainly result. It seems unlikely that pyridine-N-oxide would remain co-ordinated to ruthenium in the presence of excess pyridine. Assuming that "ruthenium i s in an oxidation state which can possess significant crystal f i e l d stabilisation then complexation with the N-donor ligand should be preferred over complexation with pyridine-N-oxide in the absence of back-bonding from the metal to the ligand. The location of the v(M-K)) modes in the spectra of pyridine-N-oxide complex-126 127 es ' has indicated l i t t l e or no II-bonding effects and gives some justi f i c a t i o n to a comparison of pyridine and pyridine-N-oxide co-ordination on the grounds of crystal f i e l d stabilisation. The oxidation of co-ordinated carbon monoxide according to 7 . 0 III a. 2 . 0 i — i — s — r ] — i — i — i — i — j — i i i — i — r J i i i i i i i i i J i i i i 2.0 TIMEx-10 . s e c . Fig..17.'.Gas.uptake by [HRu(C0) 3] n in-pyridine under oxygen, 21.0x10. 5moles n i n pyridine at 60°C. under P 10.32x10" 5moles. Pyridine volume = 5-mis! f ^ ^ O h J n  p ( t o t a l ) = 7 0 0 mm Hg.. F i n a l gas uptake 79 'l I i I 1 I I I I » I 1 1 l _ L _ L _ i - J . 6 0 0 6 5 0 7 0 0 WAVELENGTHinm. F i g . 18V -Vi s i b l e spectroscopic.changes i n . a 5.4x10- M . pyridine solution of [HRu(C0) 3] n under oxygen at 35°C. (1) t=0 sec. under argon,- (2) .t=250 ' sec. , (3) t=650 sec. , (4)-±=1200:. sec..,. (5) t=1800 s e c , (6)'t=2600 sec. , (7) t=3800 s e c , (8).t=12 hours. 80 reactions such as eq. (3.7) or eq. (3.8) would not r e s u l t i n net uptake of gas and can be discounted, unless the CO2 remains co-ordinated. Such a reaction has not yet been observed but i s not 128 impossible since such CO^ complexes have been documented. Insertion of the CO2 moiety into the Ru-H bond to give a formate 129 i s also a p o s s i b i l i t y . Processes (c) and (d) are also f e a s i b l e and are next considered. 2HRu(CO) 3(py) x + 0 2 > 2HRu(CO) 2 (py) + 2C0 2 (3.7) HRu(C0) 3(py) x + 0 2 > HRu(C0) 2(py) x_ 1 + py-0 + C0 2 (3.8) A s o l i d product was p r e c i p i t a t e d from the oxidised pyridine solutions by removal of solvent under reduced pressure and addition of anhydrous ether. The s o l i d was i n s o l u b l e i n non-polar solvents such as hexane or benzene, p a r t i a l l y soluble i n water and completely soluble i n methanol and chloroform. Solution i n methanol, followed by removal of solvent under reduced pressure appeared to o f f e r the best method of obtaining the s o l i d i n a u s e f u l , dry powder form which could be r e -dissolved i n pyridine to regenerate a greensolution with X = 640 nm. max i n the v i s i b l e spectrum. The i n f r a - r e d spectrum of t h i s s o l i d (KBr disc) d i d show a weak band i n the 800-900 cm 1 range normally associated with 130 dioxygen complexes of Group VIII metals ; however, the stoichiometry of h mole of oxygen per mole of ruthenium would be hard to r a t i o n a l i s e i n terms of dioxygen complex formation under such conditions. The s o l i d -state i n f r a - r e d spectrum of separate samples of t h i s green s o l i d obtained by prolonged pumping on the o r i g i n a l material always gave e s s e n t i a l l y the 81 same r e s u l t s ( i . e . 2000(w), 1915(m,br), 1590(m,br), 1480(vs), 1440(m), 1260(w), 1210(w) , 1155(w), 1065(w), 835(w), 765(m) and 695 (m) cm "*"). The spectra were i d e n t i c a l i n both n u j o l mull and KBr di s c form and a very broad band of low i n t e n s i t y i n the 3200-3400 cm 1 region, possibly due to remaining methanol, was evident. The absence of a symmetrical N-H stretching frequency, which occurs near 3200 cm 1 i n non-hydrogen bonded s a l t s and near 2800 cm 1 i n 122 hydrogen bonded s a l t s , appeared to r u l e out the p o s s i b i l i t y that the green s o l i d was a pyridinium s a l t . However, such bands are 131 frequently broad and d i f f i c u l t to detect. Wilkinson and co-workers have reported the p r e c i p i t a t i o n of [Ph^As][Ru(C0)C£ 4py] by add i t i o n of a concentrated, s l i g h t l y acid, aqueous s o l u t i o n of tetraphenylarsonium c h l o r i d e to an aqueous s o l u t i o n of [pyH][Ru(C0)C£ 4py]. The add i t i o n of an aqueous methanolic so l u t i o n of tetraphenylarsonium ch l o r i d e to a methanolic s o l u t i o n of the green s o l i d f a i l e d to generate any s i m i l a r kind of p r e c i p i t a t e . A good method for attempted p u r i f i c a t i o n of the s o l i d involved conducting column chromatography on methanolic solutions 132 using f l o r i s i l , a useful adsorbent for basic substances that might s t i l l be adhering to the s o l i d . The r e s u l t was rapid e l u t i o n by methanol of a sharp yellow-green band of material down the column. The phy s i c a l c h a r a c t e r i s t i c s of the s o l i d obtained by t h i s method (e.g. U.V./visible, i . r . spectra) were i n d i s t i n g u i s h a b l e from those of the s t a r t i n g green material. A vapour pressure osmometer was used to compare a methanolic s o l u t i o n of the p u r i f i e d green s o l i d with standard b e n z i l solutions 82 i n methanol ( for c a l i b r a t i o n of the osmometer see F i g . 19). A molecular weight of approximately 200 was estimated by t h i s technique (A mass spectrum of the same green s o l i d showed weak peaks at m/e=233 and 234, and peaks at m/e values of 80 and l e s s i n d i c a t i n g the presence of p y r i d i n e ) . The conductivity of methanolic solutions was found to be -1 -1 2 approximately 10 moles ohms cm., which i s far below the range of 70-90 moles 1 ohms 1 cm. expected for a 1:1 e l e c t r o l y t e i n t h i s s o l v e n t . 1 3 3 However, d i s s o c i a t i o n of t h i s compound i n methanol seems l i k e l y on the basis of the low measured molecular weight. E.S.R. experiments on solutions prepared by d i s s o l v i n g [HRuCCO)^]^ i n pyridine under vacuum or i n a i r , f a i l e d to show any signals at room temperature or l i q u i d nitrogen temperature. These observations i n d i c a t e that the ruthenium species i n these solutions i s not i n oxidation state + l ( d 7 ) , unless coupling of electrons i n some dimeric Ru(I) species occurs i n which case no e.s.r. s i g n a l would be expected. The green s o l i d obtained from solutions of [HRuCCO)^]^ i n pyridine seemed quite soluble i n chloroform, but no high f i e l d ^H n.m.r. signals could be detected i n CDC£ 3 s o l u t i o n . The oxygen uptake by pyridine solutions of [HRu(C0) 3l n was observed to proceed unhindered i n the absence of l i g h t and was not reversed by prolonged exposure of the green py r i d i n e solutions to hydrogen. Three separate microanalyses of the green s o l i d gave the r e s u l t s indicated i n Table IX i n which the data are compared with possible mono-meric formulae. Both species i n Table IX would s a t i s f y the observed gas uptake and give analyses that are s a t i s f a c t o r y except for the s l i g h t l y low percentage of hydrogen. 8 3 F i g . 19. Vapour pressure osmometer c a l i b r a t i o n for molecular weight determinations using b e n z i l s o l u t i o n s i n methanol (instrument readin vs. [ B e n z i l ] ) . 84 Table IX. Microanalyses of samples of the green solid and possible formulae C% H% N% Found (a) 43.40 4.63 8.69 " (b) 42.77 5.10 7.60 " (c) 43.04 3.80 8.92 HRu(CO) 2(C0 2)(py) 2 43.32 3.05 7.88 [Ru(CO) 3(py) 2] 2-0 2 43.44 2.78 7.80 85 3.6. Chemistry of [HRu(CO) 3J n i n other solvents The [HRuCCO)^]^ polymer was dissolved i n pi p e r i d i n e by warming to a temperature of 60-70°C. under vacuum or i n e r t atmosphere to y i e l d yellow coloured solutions. The s o l u b i l i t y i n p i p e r i d i n e was s i m i l a r to that i n pyridine at those temperatures and solutions with ruthenium concentrations of the order of 0.03M. were r e a d i l y achieved. Under a carbon monoxide atmosphere the carbonylation of p i p e r i d i n e to N-formyl p i p e r i d i n e could be conducted (as discussed i n section 6.1). Experiments designed to detect evolution of gas upon d i s s o l v i n g the hydridocarbonyl i n p i p e r i d i n e under argon or nitrogen showed no such evolution, while -2 v.p.c. analysis of a 9.67 x 10 M. s o l u t i o n a f t e r 7 hours at 50°C under nitrogen indicated that exactly 1.0 moles of N-formyl p i p e r i d i n e was formed per mole of ruthenium. This suggested that one mole of CO was consumed i n a stoichiometric reaction and that the pi p e r i d i n e product was a dicarbonyl. For example, HRu(C0) 3 p i p e r i d i n e > (co)^(pip)^ + pip.CO (3.9) The i s o l a t i o n of a carbonylamine complex from p i p e r i d i n e solutions of [HRu(C0) 3] n was conducted by a method s i m i l a r to that used i n the preparation of [Ru(C0) 2(OAc)(pip)]^ (section 2.1.1). In an i n i t i a l experiment, a saturated s o l u t i o n of [HRu(C0) 3] n i n p i p e r i d i n e was obtained by warming to 75°C. under nitrogen f o r 2 hours. When the r e s u l t i n g orange s o l u t i o n was cooled, no p r e c i p i t a t i o n occurred but, upon addition of an excess of de-gassed, d i s t i l l e d water, a l i g h t brown p r e c i p i t a t e was produced. This p r e c i p i t a t e was c o l l e c t e d by f i l t r a t i o n 86 under nitrogen and dried in vacuo for several hours. The infra-red spectrum in nujol mull (Fig. 20) showed three peaks in the CO region at 2020,1965 and 1930 cm 1 remarkably similar to those exhibited by [Ru(C0) 2(OAc)(pip)] 2- No bands attributable to Ho0 or OH were evident; however, a band was detected at 1730 cm 1 in the region associated with non-aromatic amide stretching frequencies and the microanalysis was incompatible with [HRu(C0) 2(pip)] 2 as product (Calculated for RuC7H12N02:C,34.55;H,4.94;N,5.76%. Found: C,48.61; H,7.15;N,5.34%). The presence of N-formyl piperidine as contaminant seemed l i k e l y and in subsequent experiments a more thorough procedure of washing the sample many times with water was adopted. In addition, the solid was dried under vacuum for up to 12 hours in order to remove the rather involatile N-formylpiperidine (b.pt.=222°C. at 760 mm. Hg.). With this improved technique, the product obtained gave no amide CO stretching frequency in the infra-red spectrum (Fig. 21) and also a very different microanalysis (Found: C,41.77;H,5.60;N,7.56%). The solid was observed to be stable in air since i t s infra-red spectrum did not change at a l l and a repeat microanalysis on a sample exposed to air for several days showed l i t t l e change (Found: C,42.28;H,5.72; N,7.17%). In addition, the solid was washed several more times with water and dried under vacuum for a further 12 hours to give a product with essentially the same analysis (Found: C,41.59;H,6.00;N,6.96%). In later experiments an even better method involving removal of piperidine solvent under vacuum and addition of water to the resulting red-brown o i l was successful in producing a fine light-brown powder. Microanalyses 87 F i g . 2 0 . Infra-red spectrum (nujol mull) i n the carbonyl stretching region of the product.derived from [HRu ( C 0 ) 3 ] and-piperidine.- - ; -88 F i g . 21. Infra-red. spectrum (nujol,mull) i n the carbonyl stretching region of the product derived from [HRu(C0)3] n and p i p e r i d i n e (by an improved i s o l a t i o n technique). • 89 performed on two separate batches of a compound prepared i n t h i s way were consistent with the previous r e s u l t s ((I) Found: C,41.10;H,5.77; N,7.40% and (2) Found: C,40.63;H,6.00;N,7.21%). A reasonable f i t for the a n a l y t i c a l data i s provided by a compound of empirical formula R u 2 H 2 ( C 0 ) 4 ( p i p ) 3 which requires: C,39.91;H,6.12;N,7.35%. This species could be formed by loss of a p i p e r i d i n e ligand from an i n i t i a l l y formed dimeric species i n s o l u t i o n (eq. 3.10), giving a 3:2 r a t i o of co-ordinated p i p e r i d i n e to ruthenium, which i s c e r t a i n l y indicated by the a n a l y t i c a l 47 data. A s i m i l a r d i s s o c i a t i o n has been proposed previously to explain P|P pip °C I ^ CO OC >UCT ^ R l C > ";Ru— H - ^ R U - CO + pip + H + (3.10) OC^ ^ H - ^ ^CO (XT / \ / P!P pip / pip p pip CD (ID low molecular weight r e s u l t s obtained for various ruthenium complexes i n s o l u t i o n . e.g. Ru 2(CO) 4(CH 3COO) 2(py) 2^^Ru 2(CO) 4(CH 3COO) 2(py) + py (3.11) In equation 3.10, the t h i r d p i p e r i d i n e ligand could be e i t h e r bridging (as depicted), i n which case i t may be described as a pip. or p i p e r i d i n o group, or i t may be co-ordinated to only one of the ruthenium 134 atoms. Triply-bridged dimeric ruthenium (II) species have been reported 135 but none involve bridging amino ligands. Sappa and Milone have d i s -cussed the structure of a HRu„(CO) 1„HNPh complex i n terms of a n i l i n e 90 bridging two of the ruthenium atoms and a number of derivatives of Os^CCO)^ containing bridging aniline have been synthesised by 136 Deeming and Yin. However, piperidine should not be such a good bridging ligand as aniline since charge on the nitrogen atom cannot be stabilised by delocalisation as in an aromatic system. The structures (I) or (II) would be d i f f i c u l t to distinguish by infra-red spectroscopy; the great similarity of the carbonyl stretching region with that recorded for [Ru (00 )2(OAc)(pip)]^ supports a structure such as II. The absence of strong absorption at 3000 cm 1 due to N-H stretching vibrations rules out the possibility of a piperi-dinium salt. The compound was found to be insoluble in hexane, but readily soluble in chloroform, acetone and methanol; the ionic character could be consistent with a formula such as H +[II] . A.-.conductivity -1 -1 2 measurement in methanol gave A M = 12.5 moles ohms cm (assuming a molecular weight of 500), which i s well below the expected range of -1 -1 2 70-90 moles ohms cm for 1:1 electrolytes in this solvent, but i s indicative of an appreciable concentration of ions. The "^H n.m.r. spectrum in CDCA^ (Fig. 22) i s close to that of free piperidine and is unlike that of a piperidinium salt (e.g. Fig. 23); unfortunately, no metal hydride resonance could be detected in the region upfield from T.M.S. The purity of the compound was established by t . l . c . on s i l i c a gel using chloroform as solvent and eluting agent, which gave one spot at an RF value of 0.66. V.P.C. analyses of chloroform solutions of this compound using the Pennwalt 223 column showed no traces of piperidine or N-formyl piperidine. 91 F i g . 22 . H n.m.r. spectrum of the product derived from [HRu(CO) 3] n and p i p e r i d i n e i n CDC£~. 30.0 ® 20.0 ® 10.0-F i g . 24 pip e r i d i n e of oxygen Y I U f E x 1 © w ,sess. c Oxygen uptake by [HRu(CO) 3] n i n piperidi n e , 12. OxlO.-. moles. [HRu (CO) ql i n . ••at 60 C. under P ( t o t a l ) = 690 mm Hg ( A ). ( . ) represents the s o l u b i l i t y xn p i p e r i d i n e under the same conditions. 94 When exposed to oxygen atmospheres, the p i p e r i d i n e solutions of [HRu(CO) 3] n show very d i f f e r e n t behaviour from that of the pyridine solutions. A t y p i c a l experiment inv o l v i n g oxygen uptake by an orange-yellow s o l u t i o n at ruthenium concentration = 0.024M. made up under vacuum at 60°C. i s i l l u s t r a t e d i n Figure 24. Judging by the s t o i c h i o -metry, the process involved i s not simply oxidation of ruthenium. This experiment reveals that the oxygen uptake exceeds a 2.5:1 value of O^Ru within 2^ hours at 60°C; when the r e s u l t i n g red s o l u t i o n was cooled overnight and then re-subjected to oxygen, the re a c t i o n appeared to continue. At room temperature the bubbling of oxygen through the s o l u t i o n of [HRu(CO) 3] n i n p i p e r i d i n e r e s u l t s i n a gradual colour change from orange-yellow to red wi t h i n the space of one hour and a corresponding change i n the v i s i b l e spectrum from a continuous absorption to an absorption with a band at approx. 530 nm. (Fig. 25). I s o l a t i o n of a s o l i d from the reaction mixture of the experiment i l l u s t r a t e d i n F i g . 24 was accomplished by removal of p i p e r i d i n e under vacuum and addition of water to the remaining o i l . The r e s u l t i n g brown powder gave an i . r . spectrum (Fig. 26) with a s i m i l a r carbonyl s t r e t c h i n g region to that obtained from the product i n the absence of oxygen (Fig. 20). However, a sharp strong peak at 1265 cm 1 and a broad, strong peak at 805 cm 1 are now evident. The "*"H n.m.r. spectrum (Fig. 27) i s very s i m i l a r to that of the product obtained i n the absence of oxygen. The a n a l y t i c a l data for t h i s compound are consistent with a species such as [HRu(00)2(pip)]2•0 (Calculated for Ru 2C 1 4H 22N 20 6: C,32.55;H,4.26;N,5.42%. Found: C,32.94; H,4.03;N,5.57%), i n which 0 o could be co-ordinated to one ruthenium 95 J I L 1 J ' ' I J I L i J I I L 5 0 0 5 5 0 6 0 0 WAVELENGTH, nm. 7 F i g . 25. V i s i b l e spectroscopic changes i n a 1.58x10 "M.- pi p e r i d i n e so l u t i o n of [HRu(C0) 3] n under oxygen at 25°C. (1) t=0 sec. under argon, (2) t = l % hours, (3) t=3 hours, (4) t=4 hours. 96 97 98 or could act as a bridging ligand. The stoichiometry of gas uptake c e r t a i n l y exceeds the requirement f o r formation of such a complex and/or oxidation to higher valent Ru, and suggests that some c a t a l y t i c oxidation of p i p e r i d i n e may be taking place. The polymer [HRu(CO) 3J n was found to be soluble i n DMA upon heating to 60-70°C. f o r several hours, giving orange-yellow solutions -2 with concentrations up to approximately 2.8 x 10 M. i n ruthenium as monomer (see section 3.2 for experiments on detection of the metal hydride i n DMA). DMA showed no spurious gas evolution e f f e c t s (cf. pyridine or piperidine) when attempting to determine the e f f e c t of adding [HRu(C0).j] n to the neat solvent under i n e r t atmosphere. Neither the use of pure DMA with an empty glass bucket, nor the addi t i o n of [HRu(C0) 3J n to DMA solvent i n the same bucket resulted i n any gas evolution e f f e c t at 70°C. over a period of 2-3 hours. The conclusion then i s that the polymer remains as a t r i c a r b o n y l species upon d i s s o l u t i o n i n neat DMA. The v i s i b l e spectra of such solutions showed no i n t e r e s t i n g features above the DMA cut-off point (270 nm) and no changes were observed upon prolonged exposure to carbon monoxide atmosphere at room temperature. -2 An experiment inv o l v i n g treatment of a 2.77 x 10 M s o l u t i o n with carbon monoxide at 25°C. over a period of several days indicated that the so l u t i o n i n f r a - r e d spectrum i n the carbonyl s t r e t c h i n g region had changed from an i n i t i a l sharp peak at approximately 2000 cm 1 to a f i n a l spectrum showing two absorptions on either side of t h i s o r i g i n a l absorption (Fig. 28). The use of a v a r i a b l e path-lenth s o l u t i o n i n f r a - r e d c e l l was quite successful for i n v e s t i g a t i n g the metal carbonyl stretching region i n DMA s o l u t i o n . 99 2 5 0 0 2 0 0 0 1800 wAVEi^ ugyiBER(c§y§7 F i g . 28. Solution i n f r a - r e d spectra i n D M A of (i) [HRu(C0 ) o ] n dissolved i n D M A under argon at [Ru]=2.7X10~2M. ( s o l i d l i n e ) , ( i i ) the same so l u t i o n a f t e r 3 days under CO at 1 atm. t o t a l pressure (broken l i n e ) 100 However, gas-uptake experiments f a i l e d to provide evidence for carbon monoxide uptake by these solutions at higher temperatures (e.g. 60-7 0 ° C ) . DMA appears to co-ordinate very well and as the l a t e r d i s -cussion w i l l i n d i c a t e (section 6.1) competes e f f e c t i v e l y f o r co-ordination to ruthenium when i n the presence of amines such as pi p e r i d i n e . E.S.R. studies on DMA solutions of [HRuCCO)^]^ gave no signals at room temperature nor at l i q u i d nitrogen temperature and attempts to extract a s o l i d by removal of the solvent under reduced pressure gave red-coloured o i l s . Attempts to obtain c r y s t a l s by ad d i t i o n of dry, oxygen-free ether to these o i l s gave no s a t i s f a c t o r y r e s u l t s , although some small quantities of a red powder could be p r e c i p i t a t e d by t h i s method. Exposure of the solutions to oxygen seemed to r e s u l t i n only very s l i g h t and gradual changes i n colour, and no i n t e r e s t i n g features were generated in- the v i s i b l e spectrum (Fig. 29, to be compared with the refl e c t a n c e spectrum of [HRu(C0) 3] n i n F i g . 7). The chemistry of [HRu(C0) 3] n i n l i q u i d phosphine solvents was not extensively investigated. Although solvents such as MePt^P and Me2PhP appeared to o f f e r advantages over amine solvents i n terms of the ruthenium concentration that could be attained (at l e a s t 0.05M; heating to 40-50°C. for prolonged periods under i n e r t atmosphere gave yellow solutions i n these solvents), there was only one case (MePh^P) i n which a high f i e l d "*"H n.m.r. si g n a l was detected (see section 3.2). -2 In one experiment employing a 5.78 x 10 M. s o l u t i o n i n the more v o l a t i l e Me„PhP solvent, removal of solvent under reduced pressure gave a yellow 101 4 5 0 5 0 0 5 5 0 W A V E L E I ^ G ™ , n m F i g . 29. V i s i b l e spectrum of a 1.15x10 M. s o l u t i o n of. [HRu(CO) ] i n DMA under oxygen. 102 "glue" and a few small crystals. Attempts to obtain complete c r y s t a l l i -sation by addition of solvents such as n-heptane under inert atmosphere, followed by warming and subsequent refrigeration, were unsuccessful. Triphenylphosphine, when heated to i t s melting point (110°C.) under nitrogen did not dissolve the hydridocarbonyl, even after s t i r r i n g at this temperature for five hours. The reactions of the hydridocar-bonyl polymer with triphenylphosphine in ethanol w i l l be discussed in section 8.2. The polymer [HRu(C0) 3] n had only limited solubility in DMSO, certainly less than 0.013M.when heated to 60°C. under nitrogen atmosphere. 6 13 A sample of such a dilute solution in DMSO- d was submitted for C F.T. n.m.r., but no signals could be detected. The addition of a paramagnetic transition metal ion frequently prevents the long re-13 laxation times normally observed in the C n.m.r. of metal carbonyl 137 solutions and this modification might have assisted in detecting a signal from our DMSO solution. However, the solution concentrations 13 were probably too low and our inab i l i t y to detect C n.m.r. signals from solutions of other ruthenium carbonyls (e.g. Ru^CO).^ i n CDCS-^ ), even in the presence of a paramagnetic shiftless relaxation agent, {Cr(acac) 3}, mitigated against any further work in ,this direction. The solution of [HRu(C0)„] in DMSO - d also failed to show any high 3 n fi e l d n.m.r. signals. 103 Chapter 4 AN INVESTIGATION OF THE SYNTHETIC ROUTE TO [HRu(CO) ]. - THE INITIAL STAGES 4.1. Introduction The method of production df the polymer [HRu(CO) 3] n by carbonylation of aqueous solutions of ruthenium t r i c h l o r i d e not only gives products of c a t a l y t i c i n t e r e s t but also involves some i n t e r e s t i n g s o l u t i o n chemistry, i n terms of reductive carbonylation processes. Such chemistry of platinum metals has been studied 138 139 f a i r l y extensively for some Rh(III) and I r ( I I I ) systems, but not for Ru, and i s of i n t e r e s t i n terms of p o t e n t i a l synthetic routes to low valent carbonyl complexes. An examination of the aqueous f i l t r a t e remaining at the end of one week's continuous CO bubbling through the RuCJ^ s o l u t i o n at 8 0 ° C , including the heating and cooling procedures previously mentioned (section 3.1), reveals the following d e t a i l s . If the i n i t i a l material i s assumed to be "RuC&3.3H20", then s t a r t i n g with 0.40x10 moles of the hydrate -2 - -2 -( i . e . 1.20x10 moles of C£ ) produces 1.14x10 moles CH i n the f i n a l 104 s o l u t i o n . This s o l u t i o n i s c o l o u r l e s s , with a pH of 1.8 ( c a l i b r a t i o n -2 + corresponds to 1.2x10 moles H ) and contains no side-products such 140 as formic acid (generated for example from CO and H^ O ). These r e s u l t s imply that a l l of the ch l o r i d e ion remains i n s o l u t i o n and that there are 3 moles of H + produced per "mole" of i n i t i a l Ru (known to be a mixture of Ru(III) and Ru(IV) - Section 2.1.1). Considering the nature of the s t a r t i n g material these r e s u l t s , although s t o i c h i o -m e t r i c a l l y pleasing, are somewhat ambiguous and so the reactions were investigated with a pure Ru(III) compound. 4.2. Preparation of Ru(III) compounds and some comments on the  "blue ruthenium(II)" solutions. Potassium pentachloroaquoruthenate(III) was prepared by the 141 method of Mercer and Buckley. A s o l u t i o n of 1.39 gms. of ruthenium t r i c h l o r i d e i n 125 mis. of cone, hydrochloric acid was refluxed for 2 hours and then a stoichiometric amount of potassium chl o r i d e (0.8 gms.) was added. The r e s u l t i n g mixture was cooled and s t i r r e d with excess mercury u n t i l no more mercurous chl o r i d e was p r e c i p i t a t e d . This procedure had the e f f e c t of reducing a l l Ru(IV) to Ru(III), and upon f i l t r a t i o n the dark red s o l u t i o n was evaporated to approximately 20 mis. which y i e l d e d a red c r y s t a l l i n e s o l i d on cooling. This s o l i d was f i l t e r e d , washed with ethanol and drie d i n a i r . R e c r y s t a l l i s a t i o n was achieved by d i s s o l v i n g t h i s s o l i d i n the minimum of 6M.HC£, cooling the s o l u t i o n slowly and f i l t e r i n g the r e s u l t i n g dark brown c r y s t a l l i n e s o l i d . An analysis 105 of this solid (Calculated for K 2RuC£ 5.H 20: Ru,26.95;C£,47.40%. Found: Ru,25.61;C£,46.80%) suggests that a small amount of K~RuC£, may also be present (Calculated for K»RuC£., :Ru,23. 45; J o J O 142 C£,49.41%). Fine has reported that a solution of K^RuCJ^.I^O in 6M-HC£ equilibrates within a few minutes to one containing roughly equal percentages of the hexachloro- and pentachloroaquo-ruthenate(III) species. However, the product reported above had a u.v./visible spectrum in 6M.HC£ with one maximum at 384 nm (e=3000M "'"cm "*") 142 and no sign of the maxima at 307 nm and 347 nm that are reportedly 3-characteristic of solutions containing RuC£, (e,.„-,=2400 and £„, =3300 6 307 34/ M "'"cm 1 ) . In addition, there was no evidence for the presence of Ru(IV) -1 -1 which is characterised by a band at 485 nm in 6M.HC£(e4g,-=2500M cm for a solution of (NH 4) 2Ru(0H)C£ 5). 143 Potassium hexachlororuthenate(III) was prepared by refluxing 1.0 gm. of ruthenium trichloride in methanol under an atmosphere of hydrogen for several hours. The time required varied from 2 to 5 hours, presumably depending upon the percentage of ruthenium(III) and ruthenium(IV) in any particular batch of the trichloride. At the end of this time the solution had begun to turn green, a consequence of the f i r s t stages of reduction from ruthenium(III) to ruthenium(II). Then, 0.90 gms. of potassium chloride (KC£:Ru~3) was added. The mixture was refluxed in air and slow dissolution of the potassium chloride was accompanied by the formation of a brown precipitate. This brown solid'was filtered from the colourless supernatant liquid, washed with methanol, and then recrystallised from 12M.HC£ to give a 106 y i e l d of K^RuCAg usually between 40 and 50%, a f t e r the product had been dried under vacuum ( t y p i c a l l y 0.70 gms.). The absorption spectrum of the s o l i d i n 12M.HC£ showed two maxima at 314 nm (e=2215M "'"cm "*") and 350 nm (e=2800M "'"cm "*"),in reasonable agreement 142-145 with previous studies. The i . r . of the s o l i d showed a strong i n t e n s i t y band at 310 cm 1 with two moderate i n t e n s i t y shoulders at 280 and 330 cm 1 . Chemical analysis gave s a t i s f a c t o r y r e s u l t s (Calculated f o r K^RuCJig :Ru,23.45;K,27 .19%. Found: Ru,23.87; K,26.93%). On two occasions the E^-reduction process a c c i d e n t a l l y proceeded beyond the required stage and a green-blue s o l u t i o n was obtained. The add i t i o n of the same quantity of potassium c h l o r i d e to t h i s s o l u t i o n , followed by r e f l u x i n g i n a i r , resulted i n pro-duction of a green p r e c i p i t a t e . F i l t r a t i o n and drying i n a i r gave a large quantity (1.50 gms. from 1.0 gms. RuCJl^• 31^0) of a green powder (1) which showed a rather broad band i n the s o l i d state i n f r a -red spectrum (nujol mull) at approximately 300cm ^, i n the region 146 . associated with terminal Ru-C£ stretching frequencies. A sample (0.018 gms.) of the s o l i d dissolved i n 5 mis. of water f a i l e d to absorb CO at 75°C under a t o t a l pressure of 1 atmosphere, at l e a s t 26 within the f i r s t hour. Halpern, James and Kemp have shown that solutions of ruthenium(II) i n 3M.HC£ take up one mole of CO per 2-mole of Ru(II) quite r a p i d l y to give RuC£^(CO)(E^O) , followed by 2-slower uptake of a further mole of CO to give Ru^O^CJi^ . The fact that the green p r e c i p i t a t e does not react with CO i n aqueous s o l u t i o n argues somewhat against the s o l i d being a ruthenium(II) species, i n 107 addition the solid was obtained from an air refluxing procedure. 147 Moreover, Rose and Wilkinson have isolated compounds of ruthenium(II) by adding salts containing large cations such as Cs +, Ph^ As"*" and Et^N + to blue methanolic solutions under nitrogen atmosphere; these workers analysed the resulting precipitates and suggested that a cluster anion 2-(Ru^C£^2 ) w a s present. The blue solutions (in either water or methanol) apparently became green very rapidly on air oxidation and addition of + 147 Ph^As gave a green precipitate which Rose and Wilkinson suggested was the trans-isomer of [Ph^As] [ R u 1 1 * ^ (H 20) 2] {Found :C,43. 3 ;H,3.5; C£, 21.4%.C 2 4H 2 4As 5C£ 40 2Ru requires C,43.6;H,3.6;C£,21.3%}. The green solid obtained in the present work could be dissolved in aqueous methanol and made to produce a further green precipitate by addition of a similar concentrated solution of Ph 4As +C£ . The analysis of this solid (Found: C,40.44;H,3.20%) suggests a ruthenium(III) salt with a 1:1 ratio of Ph 4As + to Ru(III), and the i n i t i a l green compound(1) l i k e l y contains mainly the potassium salt of the RuC& 4(H 20) 2 anion. The blue solutions produced by reducing Ru(III) chloride solutions 148 in HC£ remain a source of continuing controversy. Mercer and Dumas have produced such solutions by electrolysis of K2RuC£,-.H20 in 0.01M. acid. By employing a cation-exchange column designed for use under inert atmosphere conditions they were able to separate their blue solutions into three components. Determination of the charge on each of these species and also chloride analyses suggested that mixed oxi-dation state complexes of ruthenium(II,III) were present. However, + 2+ the electronic spectra of their species Ru2C£,-,Ru2C£4 , Ru 2C£^ and the corresponding Ru(III) oxidation products in acidic aqueous solutions 108 showed no evidence of the maximum at 330 nm which was recorded i n the u. v . / v i s i b l e spectrum of the green s o l i d (1) immediately a f t e r d i s -s o l u t i o n i n water; t h i s p a r t i c u l a r spectrum was also characterised by a shoulder at 430 nm and a broad maximum of low i n t e n s i t y at 665 nm. Compound (1) was also found to change colour to dark brown gradually upon standing i n a i r and aqueous solutions of the r e s u l t i n g s o l i d gave u. v. / v i s i b l e spectra which were very s i m i l a r to those obtained with compound (L)except for the disappearance of the band at 665 nm. It seems l i k e l y that (1) contained a small amount of ruthenium(II), slow oxidation leading to the observed colour change i n the s o l i d . Experiments on the aqueous "blue s o l u t i o n s " i n the present work were l i m i t e d to a study of carbon monoxide uptake i n the presence and absence of a soda-lime attachment, which absorbs carbon dioxide. This reaction i s known^^ to give Ru^^CO^Cil^ i n acid s o l u t i o n by a process which, i n the absence of soda lime, involves gas absorption corresponding to 2 moles of gas per mole of ruthenium. If the blue solutions contain only ruthenium(II), one would expect no carbon dioxide to be generated (eq. 4.1), but i f ruthenium (II,III) species are present some reductive carbonylation and carbon dioxide generation must occur (eq. 4.2). i . e . Either R u 1 1 ^ > R u^CO) — — R u I I ( C 0 ) 2 (4.1) or 0 , v II III +4C0^ , . I I , III +CO',+Ho0 II II +4C0 __ I I , r n , ( l n . 2(Ru -Ru ) > 2 Ru -<Ru - — 2 — 2 Ru -Ru 2Ru (C0) o (4.2) I I -CO -2H + I I CO CO 2, n CO CO The expected t o t a l stoichiometry i s 2:1 f o r both schemes i n the absence of 109 soda lime. In the presence of soda lime; the reactions depicted i n eq. 4.1 give the same 2:1 stoichiometry, while those shown i n eq. 4.2 should give a 9:4 stoichiometry. Therefore, i f ruthenium(III) i s present at a l l and the scheme represented i n eq. 4.2 operates, then observations on the amount of CO taken up by the s o l u t i o n should d i f f e r i n the presence and absence of soda lime. The "blue-solutions" were prepared by the addition of an excess of a 2M. s o l u t i o n of titanium(III) c h l o r i d e i n 3M.HCJI to a roughly 0.035M. s o l u t i o n of "ruthenium t r i c h l o r i d e " i n 3M.HC& under a nitrogen atmosphere. The gas uptake experiments were conducted at 80°C. under one atmosphere of carbon monoxide i n the presence (a) and absence (b) of a soda lime attachment. An i n i t i a l gas uptake of approximately one mole of CO per mole of ruthenium took place i n approximately 4000 seconds i n both experiments (a) and (b). The r a t i o of the f i n a l amount of gas taken up i n experiment (a) to the f i n a l amount taken up i n experiment (b) was 1.19:1, which i s f a i r l y close to the r a t i o of 1.13:1 expected from the scheme shown i n equation 4.2. The r e s u l t of t h i s work supports 148 previous suggestions that these-:blue solutions of ruthenium chloro species contain mixed oxidation state complexes of ruthenium(II) and ruthenium(III). V.P.C. detection of CO2 (by sampling the gas above the r e a c t i o n mixture as described i n section 2.6) was unsuccessful, possibly a r e s u l t of the low concentrations involved. 4.3. U.V./visible and i . r . studies upon the carbonylation of "RuCJJ.^.SH^O" The synthesis of [HRu(C0)^] T i by carbonylation of ruthenium t r i c h l o r i d e 110 i n aqueous sol u t i o n may be followed by the use of u . v . / v i s i b l e spectroscopy. An examination of the solutions i n the i n i t i a l stages of the reaction and the aqueous f i l t r a t e from the l a t e r stages of the preparation aids i n understanding which species may be present i n solution. Aqueous solutions of ruthenium(III) c h l o r i d e , containing 1 to 7M.HC£, are known to absorb carbon monoxide at 2 6 conveniently measurable rates i n the temperature range 65-80°C. However, there seems to be no evidence of any reductive carbonylation and a 1:1 uptake of CO per mole of ruthenium(III) i s achieved within TTT ?-15,000 seconds at 80°C. The f i n a l deep red solutions of Ru (C0)C£^ are characterised by a band at 450 nm i n the v i s i b l e spectrum. This band was not detected i n our preparative experiments using purely aqueous solutions. Instead the experiments revaled the production of spectroscopic features at lower wavelengths. For instance, a f t e r 48 hours at 80°C. the f i l t r a t e from the reaction medium gave a single band 2-at 365 nm, which i s i n the correct region for a hydrolysed Ru(CO) 2C£ 4 149 species (reported X i n 3M.HC£ at 390 nm, but t h i s s h i f t s to lower max ' wavelength upon hydrolysis (section 5.7)). In another preparative experiment the f i l t r a t e from the reaction medium showed a maximum at 364 nm a f t e r 4 days, when the experiment was continued f or a further 5 days the band at 364 nm gradually s h i f t e d to lower wavelength and eventually was replaced by an absorption con-tinuum. This change i n spectroscopic features i s consistent with II II gradual formation of hydrolysed Ru (CO) 2 and Ru (CO)^ species i n so l u t i o n (sections 5.7 and 5.9). I l l Berch and D a v i s o n 1 5 0 have studied the carbonylation of ruthenium t r i c h l o r i d e i n r e f l u x i n g ethanol by the use of s o l u t i o n i n f r a - r e d spectroscopy. A peak at 2041 cm 1 i s r a p i d l y produced, followed by peaks of gradually increasing i n t e n s i t y at 2069 and 1998 cm ". Af t e r 42 hours the 2041 cm 1 peak i s no more than a shoulder on the 2069 cm 1 peak and sho r t l y thereafter becomes no longer d i s c e r n i b l e , although the so l u t i o n remains red i n colour. A f t e r 2 to 4 days, the s o l u t i o n lightens to orange and then quickly becomes yellow. No further change i s observed i n the i . r . spectrum a f t e r the disappearance of the peak at 2041 cm \ This study was repeated and found to be s u b s t a n t i a l l y c o r r e c t , although the peak at 2041 cm 1 seemed to disappear completely within 24 hours and a weak peak at 2123 cm 1 was apparent a f t e r the reaction had been conducted for 3 days. The experiment employed a 0.013 M. s o l u t i o n of "ruthenium t r i c h l o r i d e " i n absolute ethanol; a 0.1 mm sample s o l u t i o n i n f r a - r e d c e l l and a v a r i a b l e path length c e l l containing solvent was placed i n the reference beam of the i n f r a - r e d spectrometer. Berch and D a v i s o n 1 5 0 suggest that the i n i t i a l peak at 2041 cm 1 i s due to some species "Ru(C0)C£ 3", as previously proposed by Colton and F a r t h i n g 1 5 1 from t h e i r work on ruthenium carbonyl halides i n formic acid-hydrohalic 2-acid mixtures. However, Ru(C0)C£,- i n s l i g h t l y wet and a c i d i c ethanol (for s o l u t i o n purposes) shows one peak at 2036 cm 1 1 5 ° , and the i n i t i a l peak observed i n the carbonylation reaction may be due to one of several species depending upon chloride ion a v a i l a b i l i t y i n s o l u t i o n and the p o s i t i o n of e q u i l i b r i a such as those depicted i n eq. 4.3. In any case -1 III the 2041 cm peak i s a t t r i b u t e d to a Ru (CO) species. Ru(C0)C£ 3 + 2 C £ ~ ^ ^ Ru(C0)C£ 4" + C £ _ ^ = ^ Ru(C0)C£ 5 2" (4.3) 112 Berch and D a v i s o n 1 5 0 a t t r i b u t e d the peaks at 2069 and 1998 cm"1 to "Ru(CO) 2C£ 2" (or a solvated form). Previous w o r k e r s 2 6 ' 1 5 1 have reported such spectra a f t e r I s o l a t i o n of t h i s compound as a 2-dimer by evaporation of a c i d i c solutions containing [Ru(CO) 2C£ 4 ] ions. The weak band at 2131 cm 1 (measured at 2123 cm 1 i n the present work) which occurs during the l a t e r stages of the reaction was a t t r i b u t e d by Berch and D a v i s o n 1 5 0 to the fac-Ru(C0) 3C£ 3 species (see section 5.8). Attempts to study the carbonylation of a pure rutheniurn(III) compound, s p e c i f i c a l l y potassium hexachlororuthenate(III), i n a l c o h o l i c solvents were thwarted by s o l u b i l i t y problems. Af t e r 3 hours i n r e f l u x -ing ethanol the reaction mixture showed weak peaks at 2045 and 1993 cm 1 i n the metal carbonyl stretching region, but at 5 hours the mixture was s t i l l a suspension and the experiment was abandoned. The use of s i l v e r c h l o r i d e i . r . plates enabled us to study the carbonylation of ruthenium t r i c h l o r i d e (1 gm.) i n aqueous s o l u t i o n (50 mis). The reaction was conducted at 80°C. and was characterised by the appearance of a very weak band at 1962 cm 1 i n the f i r s t few hours of the reaction. From 10 to 30 hours peaks at 2073 and 2009 cm 1 increased i n i n t e n s i t y and the peak at 1962 cm 1 gradually decreased i n i n t e n s i t y . From 30 hours onwards there was no further change i n the i n f r a - r e d 151 spectrum i n t h i s region. Colton and Farthing a t t r i b u t e absorption at ~1 2-2070 and 2003 cm to the species Ru(CO) 2C£ 4 i n aqueous HC£, and absorption at 2140 and 2073 cm"1 to the species Ru(CO) 3C£ 3~ i n aqueous HC£. (The preparation of CsRu(CO) 3C£ 3 as outlined i n section 5.8 r e s u l t s i n a 113 compound with peaks at 2145 (s) and 2075 (s) cm i n the aqueous so l u t i o n i . r . spectrum). Therefore, the evidence from i . r . spectroscopic studies on the l a t e r stages of the carbonylation of ruthenium t r i c h l o r i d e i s consistent with the presence of a mixture of dicarbonyl and t r i c a r b o n y l species (see also section 3.3). The peak at 1962 cm observed i n the early stages of the reaction i s l i k e l y due to some mono-carbonyl species of ruthenium(II) containing co-ordinated water. For example, Colton and F a r t h i n g 1 5 1 suggest that the species 2- -1 Ru(CO)(H 20)C£ 4 absorbs at 1950 cm i n aqueous HC£, and c l e a r l y the p o s i t i o n of t h i s peak w i l l depend upon the a v a i l a b i l i t y of chl o r i d e ion i n the s o l u t i o n . Their work made no mention of the aqueous acid concentration employed i n the experiments. A further -2 experiment using a 3.7 x 10 M. s o l u t i o n of potassium hexachloro-ruthenate(III) i n water at 70-80°C. was found to generate peaks i n the i n f r a - r e d spectrum at 2073, 2009 and 1958 cm 1 over a period of 2 days. The absorption at 1958 cm 1 was very weak a f t e r 3 days which indicates that there i s very l i t t l e remaining Ru(II) or Ru(III) monocarbonyl species i n s o l u t i o n . The presence of some s o l i d products (notably R u ^ C O ) ^ indicated by r i n s i n g the s o l i d products with hexane and measuring the so l u t i o n i . r . spectrum) shows that a process of reductive carbonylation (see section 4.5) i s occurring. 4.4. Aqueous so l u t i o n chemistry of ruthenium(III) chloro species The aqueous s o l u t i o n chemistry of Group VIII platinum metal 114 hexachloro- and pentachloro-aquo species containing the metal in oxidation state(III) has been the subject of several detailed 152 153 152 153 studies ' • Harris's ' group have investigated the kinetics of equilibration of a l l the aquochlororhodium(III) complexes (by u t i l i s i n g spectrophotometric and isotropic tracer techniques). For example, for the potassium hexachlororhodium(III) and potassium pentachloroaquorhodium(III) species in aqueous acid solutions (eq.4.4), RhC£^" + Ho0 —1—RhC£cHo02" + C£~ (4.4) o z -s: ; 5 z k2 studies in perchloric-hydrochloric acid media of constant ionic strength showed that the aquation or anation processes could be described by: -d[RhC£^"] b = k [RhCAg ] - k 2[RhC£ 5(H 20) ][CA ] (4.5) These workers obtained the values of k^=0.11 min 1 and k2=0.013 M "'"min 1 at 25°C. from an analysis of their spectrophotometric data. The observed rate constant was found by measuring the gradient of the straight line plot of ^(A^-A^) against time, where A is the absorbance of the solution at any given time and A^ i s the optical density at equilibrium. The reactions were conducted in the presence of a 4M. concentration of chloride ion, using either hydrochloric acid or lithium chloride, and -3 -3 using concentrations of rhodium(III) varying from 1.0x10 M. to 5.0x10 M. Under such conditions the reaction depicted in equation 4.4 can be des-cribed by, 1 = M ^ r - J = (k,+k2b)t (4.6) 115 where a = i n i t i a l concentration of hexachlororhodium(III) complex, (a-x) = concentration at time t and b = c h l o r i d e ion concentration ( e s s e n t i a l l y constant). Thus, a p l o t of k, , against t J * > r (observed) ° enabled k^ and k 2 to be evaluated from the intercept and the slope r e s p e c t i v e l y . Preliminary experiments indicated that the rate of the second aquation step (eq. 4.7) was l e s s than 10% of the rate of the f i r s t step and i t s contribution to the absorbance data was ignored. This was thought to be p a r t i c u l a r l y true under conditions of excess chl o r i d e ion when the equilibrium concentration df the tetrachlorodiaquo species i s l i k e l y to be quite low. RhC£ cH„0 2~ + TT 0 ^ Tihr.9 ( H O ) " + Ci~ (4.7) 5 z z 4 z z An /examination of k, , values showed that the aquation r a t e was (observed) independent of the i n i t i a l concentration of the hexachlororhodium(III) species and increased i n a l i n e a r fashion with increase i n c h l o r i d e ion concentration at constant i o n i c strength. 154 Poulsen and Garner conducted s i m i l a r work with the analogous 3_ ir i d i u m ( I I I ) system. The changes i n the v i s i b l e spectrum of IrC£, o upon aquation were found to be too small to be us e f u l i n determining 3- 2-the reaction rate. However, solutions containing IrC£^ and I r C ^ ^ O 2-were s u c c e s s f u l l y oxidised to IrC£^ and IrC£,-H20 r e s p e c t i v e l y by chlorine and the spectra of these oxidised products were s u f f i c i e n t l y d i f f e r e n t to be of use i n analysing the reaction progress. In addition, a t i t r i m e t r i c determination of chloride ion released i n aquation of 3-IrC£g i n the range 60-110% aquation of the f i r s t c h l o r i d e ligand allowed 11.6 the f i r s t - o r d e r rate constant (for the r e a c t i o n shown i n eq. 4.8) to be estimated. Once again, the rate constant for aquation of I r C £ 5 ( H 2 0 ) 2 ~ + I L O ^ i I r C £ 4 ( H 2 0 ) 2 ~ + C£~ (4.8) the second c h l o r i d e ligand was found to be l e s s than 10% of the rate 154 constant f o r aquation of the f i r s t c h l o r i d e l i g a n d . Adamson 1 5 5 has studied the k i n e t i c s of the isotope exchange reactions of anionic ruthenium(II),(III) and (IV) chloro-complexes i n 3 6 concentrated aqueous K C£ s o l u t i o n s . A c a l c u l a t i o n of the apparent rate constant (kg,."*) for aquation of hexachloro-ruthenate(III) (using 142 spectroscopic data included i n Fine's t h e s i s ) revealed an inverse dependence upon hydrochloric a c i d concentration (Table X ) . 1 5 5 It was 153 suggested that the previous work of Harris and Robb on the analogous rhodium(III) system had f a i l e d to reveal t h i s a c i d dependence because of the comparatively narrow range of t o t a l a c i d concentrations that had been studied. The r a t i o k„ ':H„0 did not remain constant as the hydro-65 a. 2 J c h l o r i c acid,concentration decreased and hence the a c t i v i t y of the s o l u t i o n was not the only factor involved i n the decrease of kgcj'> arid primary s a l t e f f e c t s were assumed to be important. The e f f e c t of the 2-aquation of RuC£^ H 20 was ignored, since at 25°C. the value of the rate constant ( k ^ ^ f o r t h i s process i s known to be a factor of 40 156 l e s s than k, * i n IM. hydrochloric a c i d . 65 J X^i 2 Fine has studied changes i n the absorption spectra of K2RuC£,_ H^O i n 0.10 M.HC£ as a function of time. The known spectra of lower anionic, n e u t r a l and c a t i o n i c ruthenium(III) species, previously separated by i o n -117 X. The f i r s t order rate constant 0*-^^') for aquation of hexachlororuthenate(III) as a function of t o t a l hyd r o c h l o r i c a c i d concentration and the a c t i v i t y of water at 25°C. (from r e f . 155) [HC£]M. a(H 20) k 6 5 " ( m i n 1 ) k 6 5 / a H 2 0 ( m i n } 11.75 0.20 0.101 0.505 11.30 0.23 0.107 0.467 9.25 0.35 0.134 0.383 7.80 0.46 0.176 0.384 6.00 0.62 0.277 0.446 1.00* 0.96 4.0 4.04 ( A l l expe riments conducted using [Ru'"""'"1]=0 .050M., except = 2x10 4M.) 118 exchange chromatography, were used to evaluate changes i n aqueous acid solutions of K^ RuCfl.,. 1^0. The apparent sequence of events indicated by the spectra was thought to be: (1) rapid l o s s of a 2-chl o r i d e from RuC&^Cl^O) to form RuCJ^Cl^O^ , mainly i n one isomeric form; (2) l o s s of a chloride from RuC£.(ELO),. to form a 4 2 2 RuCJlgO^O)^ isomer; (3) p a r t i a l isomerisation of RuC^Cl^O)^ and loss of a chloride to form a R u C ^ C ^ O ) ^ isomer; (4) p a r t i a l isomerisation of RuCJ^ ( H 2 ^ 4 + ' a n < ^ l o s s °f a chloride to form 2+ RuCilClLjO),- . The l a t t e r stages of t h i s process appeared to take place at a very slow rate, the production of c a t i o n i c species re q u i r i n g from 8 to 70 days at room temperature. The decreasing l a b i l i t y of complexes i n the aquation process as c h l o r i d e ions are 153 removed has been observed previously with rhodium(III). This trend can be explained, at l e a s t q u a l i t a t i v e l y , by c r y s t a l - f i e l d theory which predicts that the e l e c t r o s t a t i c f a c t o r i n the aquation w i l l increase as water molecules replace chlorides since the c r y s t a l -f i e l d s p l i t t i n g i s greater f or water than for chloride ion."'" 5 7 15 8 Shukla has studied the e f f e c t of ageing on solutions of potassium pentachloroaquoruthenate(III) i n 0.1M. hydrochloric acid by c o r r e l a t i n g spectrophotometric changes with the r e s u l t s of paper electrophoresis studies. The conclusions were much the same as those 142 of Fine f or experiments over 70 days, but a f t e r 105 days a raspberry-red s o l u t i o n , which generated one c a t i o n i c band on the electrophoresis paper, had been produced and i t was suggested that the raspberry-red species contained ruthenium(IV) formed through oxidation of ruthenium(III) by a i r during ageing. 119 Studies in the present work were concerned with the solution chemistry of ruthenium(III) chloro-species in purely aqueous media containing no added hydrochloric acid or chloride ion. Clearly, the situation in solution w i l l be complicated by the change in pH i f ionisation of co-ordinated water should occur. In electrostatic terms the production of H + in solution by this means becomes more lik e l y as the charge on the complex becomes more positive (i.e. the (n—3)— pK of co-ordinated water in the series [RuC£ (Ho0) ] decreases a n v 2 B-n as n decreases). The recrystallisation of samples of potassium hexachlororuthenate 143 (III), prepared by the method of James and McMillan , from 6M.HC5, gave a dark brown crystalline solid. When this solid was dissolved in water a measurement of the conductivity within two minutes of dissolution 159 -1 -1 2 indicated the presence of three ions in solution (A^=278 ohm mole cm -3 at [Ru]=10 M.) This suggested that the starting material was in fact the pentachloro-aquo salt. The value of the molar conductivity increased to 460 within 12 hours and after 4 days was calculated as 1025, indicative of the gradual loss of chloride ion from the species in solution. A -3 5.4x10 M. solution of the same compound in water was stirred for 4 days prior to a pH measurement which indicated 1.0 H + per ruthenium had been produced in solution. A sample of potassium hexachlororuthenate(III) that had been recrystallised from 12M.HC£ gave a conductivity, measured im--1 -1 2 -3 mediately after dissolution, of AM=388 ohm mole cm at 10 M. This is consistent with the presence of 5 ions in solution, as shown in equation 160 4.9, which is expected since previous workers have noted that the loss 120 K RuC£ g + H20 >3K + + RuC£ 5(H 20) 2 + C£ (4.9) 3- 2-of chloride from RuC£, to RuC£ c (solvent) takes place in a matter of D D seconds in aqueous solutions. In fact, upon dissolving the hexachloro-ruthenate(III) in water at 20°C. the visible spectrum of the pentachloro-aquo species was always observed no matter how quickly the spectrum was -4 measured. The changes in spectrum with time for a 3.55x10 M. solution of K^RuCAg in water are illustrated in Fig. 30. The i n i t i a l maximum at 332 nm decreases in intensity and dr i f t s to lower wavelength at approxi-mately 329 nm. This process seems to take between 800 and 1000 seconds and is accompanied by the generation of isosbestic points at 292 nm and 308 nm. Subsequent spectroscopic changes are slow and involve further decreasing absorbance at 329 nm with concomitant loss of the isosbestic points and gradual production of a band a t ^ m a x = 2 9 8 nm which could correspond to the production of RuC£ 3(H 20) 3 and RuC£ 2(H 20) 4 + species as previously 142 observed in 0.10M.HC£ by Fine. The hexachlororuthenate(III) salt also showed very gradual increasing acidity in aqueous solution. For -3 -5 + instance, a 1.05x10 M. solution had produced 1.48x10 M. H (pH=4.8) -4 + within 130 seconds and 1.59x10 M. H after 4 hours, indicating very l i t t l e hydrolysis of RuC£ 4(H 20) 2 at pH»5. Concentrations of H + were evaluated by measuring the pH and determining the corresponding con-centration from a calibration graph. The calibration graph was constructed by measuring the pH of solutions of hydrochloric acid of known concentration. The generation of the isosbestic points in the u.v./visible experi-ments i s li k e l y due to slow loss of chloride from the pentachloro-aquo species i n i t i a l l y formed in solution by an equilibrium process as indicated 121 — q. F i g . 30. V i s i b l e spectroscopic changes of a 3.55x10 M. aqueous solu t i o n of K 3RuC£ 6 i n a i r at 20°C. 122 by equation 4.10. RuC£„H„0 2- RuC£ 4(H 20) 2 + C£ (4.10) (expected as cis-form) An estimate of the forward rate constant (k n) was made by analysing evaluating rate constants of f i r s t order reactions involves plotting log (A '-A ) against time (t), where k^' and A are absorbances at times t" and t respectively, in order to obtain a straight line of gradient -k/2.303. An application of this method to the data of Figure 30 for t up to 1200 seconds gave a good straight line plot and -2 . -1 provided avalue of k^ equal to 6.68x10 min Very similar results were obtained when a sample of potassium -4 pentachloroaquoruthenate(III) was dissolved in water to make a 5.43x10 M. solution and a comparison of molar extinction coefficients after 200 seconds in this and the previous experiment confirmed that Beer's law was obeyed. A specific chloride ion electrode was used to study the variation -3 of free chloride ion concentration in these solutions. A 2.42x10 M. solution of potassium pentachloroaquoruthenate(III) was found to generate 1.0 chloride per ruthenium after 800 seconds at 20°C. in a purely aqueous solution. This corresponds to the rapid i n i t i a l changes observed in the visible spectrum and the spectrum at this stage could 142 correspond to either of those reported by Fine for the two isomeric forms of RuC£.(H„0)„ in aqueous acid solution. Continued observations the data illustrated in Figure 30. The Guggenheim method 161 for 1 2 3 of the changing chloride concentration in solution confirmed that - 3 the second aquation step takes place slowly. The 2.42x10 M. solution of potassium pentachloroaquo-ruthenate(III) had generated _3 3.03x10 M. of free chloride ion after 1% hours. A study of the aquation of the hexachlororuthenate(III) potassium -4 salt at [Ru]=3xl0 M. in the presence of a five-fold excess of potassium chloride showed molar extinction coefficients at 330 nm of 2170 and 1825 after 450 seconds and 1650 seconds respectively. A similar study in the absence of added chloride ion gave values of 2225 and 1800 respectively at these times (see Fig. 30). The use of a pR=4.0 buffer, made from 50 mis of 1M. sodium acetate solution mixed with 40 mis of 1M. hydro-chloric acid and diluted to 250 mis (i.e. f i n a l solution i s 0.19 M. in chloride), gave a changing spectrum on a similar time scale to the previous two experiments. Clearly, at these concentrations of ruthenium -4 (III){[Ru]=3-5xl0 M.} the concentration of added acid or chloride i s insufficient to markedly influence the aquation of the hexachlororuthenate (III) or pentachloroaquoruthenate(III) species at 20°C (at least to the tetrachloro stage). The next point to be considered was the effect of heating ruthenium -4 (III) chloro species to a temperature of 80°C. in water. A 3.8x10 M. solution of the pentachloroaquo salt in water was heated to 80°C in air for several minutes. The resulting solution had a pH of 4.04 (corres-ponding to [H+] = 5.4x10 5M. - from a calibration graph) and use of the chloride ion electrode indicated that a l l of the chloride ion had been ionised from the i n i t i a l complex. A spectroscopic study at a similar 124 ruthenium(III) concentration (4.57x10 HM.) revealed that at 80°C a band persisted i n the v i s i b l e spectrum at approximately 325 nm for about 700 seconds. Between 700 and 900 seconds a f t e r d i s s o l u t i o n the band s h i f t e d to a broad maximum i n the region 290-300 nm and subsequently decreased i n i n t e n s i t y . In t h i s way the i n i t i a l red-brown s o l u t i o n became grey-green i n colour and when l e f t overnight i n a i r at 80°C produced a black f l o c c u l e n t p r e c i p i t a t e with a colourless f i l t r a t e . This black s o l i d was separated by f i l t r a t i o n and the r e s u l t i n g f i l t r a t e was found to contain 4 free c h l o r i d e ions per i n i t i a l ruthenium and a 2.5 to 1 r a t i o of H + generated per -4 ruthenium. An analogous experiment using [Ru]=3.85x10 M. exhibited s i m i l a r spectroscopic changes and the values' of [H +]/[Ru]= 3:1 and [C£~]/[Ru]= 4.7:1 a f t e r being l e f t overnight at 80°C. The black s o l i d was thought to be a polymeric ruthenium(IV) species generated by a i r oxidation of ruthenium(III) i n s o l u t i o n . The s o l i d was found to be generally i n s o l u b l e i n organic solvents and showed no evidence of Ru-C£ or -OH groups i n the s o l i d state (nujol mull) i n f r a - r e d —1 158 spectrum (at le a s t down to 250 cm ). In addition, Shukla has reported the eventual oxidation of ruthenium(III) to the species 4+ Rui^O)^ upon ageing aqueous solutions of K^RuC^^O for several months at room temperature. This oxidation and any tendency to poly-merise v i a hydrolysis and hydroxy bridges (olation) are l i k e l y to be greatly enhanced at 80°C compared to room temperature. -2 Inte r e s t i n g l y , a 1.8x10 M. so l u t i o n of K^RuC£^ that had been made up and stored at 80°C. under argon for an overnight period showed 125 no precipitation of solid. The resulting grey-green solution had a pH of 1.50 (i.e. [H+]=3xl0~2M and [H+]/[Ru]=l.66:1) and - -3 [CI ]=8.5x10 M. (i.e. [C£ ]/[Ru]=0.5:l). Upon re-heating this solution in air for a further 24 hours precipitation of the black solid occurred. The pH of the f i n a l f i l t r a t e was 1.26, indicating + -2 + [H ]=6.5x10 M. or an approximately 3:1 ratio of H released per Ru in the starting material. The same f i l t r a t e showed a chloride ion -3 -2 concentration of 8.8x10 M. A repeat experiment at [Ru]=l.12x10 M. under nitrogen also failed to generate any solid material within 24 hours. In this case the experiment was conducted in the gas-uptake apparatus and no gas evolution was observed, indicating that an oxidation such as that shown in eq. 4.13 is not occurring. Ru I V + e ^ ^ Ru 1 1 1 E =+0.86v. (4.11) ^ o 2Ho0 + 2 e — = ^ 2 0 H " + H „ E=-0.828v. (4.12) I ^ 2 o 2RuI]CI+2Ho0 — = ^ 2RuIV+20H~+Ho E=-1.688v. (4.13) 1S2. 163 The electrode potentials for the half-reactions ' , (4.11) and (4.12), combine to give a potential for reaction 4.13 which suggests that the oxidation of Ru(III) by water would not take place at 25°C. 4.5 Carbonylation of ruthenium(III) species in aqueous solution The behaviour of ruthenium(III) chloro species in aqueous solution at 80°C. under carbon monoxide w i l l be important in the f i r s t stage of the synthesis of [HRu(C0) 3] n. Experiments were conducted in the gas-uptake apparatus with either potassium hexachlororuthenate(III) or pentachloroaquoruthenate(III) as starting material. I n i t i a l l y , the 126 carbonylation studies were conducted at r e l a t i v e l y low ruthenium(III) concentrations. For instance, Figure 31 i l l u s t r a t e s a t y p i c a l p l o t . -2 of CO uptake against time. At a ruthenium concentration of 2.19x10 M. and t o t a l pressure of 760 mm Hg, gas uptake corresponded to a 1:1 stage a f t e r approximately 8 hours. The rates of gas uptake appeared to decrease with time and i n any case l a t e r observations w i l l be complicated by changes i n pH and c h l o r i d e ion concentration i n the s o l u t i o n . Observations at higher ruthenium concentrations (Table XI) were easier to i n t e r p r e t , giving an apparent l i n e a r i n i t i a l rate over the f i r s t few thousand seconds (Fig. 32) and a f i r s t order dependence on [Ru]. Figure 33 shows the f i r s t order dependence upon [Ru] and also includes points from various experiments conducted at lower values of [Ru]. The increase i n concentration of ruthenium(III) appeared to have no s i g n i f i c a n t e f f e c t upon the i n i t i a l induction period. A 0.14 M. s o l u t i o n of ruthenium(III) that had been aged i n a i r at room temperature for 24 hours showed a s l i g h t l y f a s t e r i n i t i a l rate of carbonylation at 80°C (18.9x10 ^M.s "*") , although the induction period and stoichiometries of uptake a f t e r 3000 and 500 seconds were very s i m i l a r to those observed when using a f r e s h l y prepared s o l u t i o n (Table XII). Table XII also includes data from experiments conducted at 0.14M. ruthenium concentration i n the presence of a soda lime tube. The r e s u l t s of such studies on the i n i t i a l carbonylation were somewhat v a r i a b l e and gave inconsistent rates and stoichiometries within the f i r s t 5000 seconds. The uptake of C0 9 by soda-lime i s probably somewhat 15.0 1.0 2.0 3.0 ct4 4-0 T I M E X I G T s s e c s . F i g . 31. CO uptake by an aqueous sol u t i o n of K3RuC£ 6(10.9xl0~ 5moles) at 80°C. ([Ru]=2.2xl0 ZM.) and 1 atm. t o t a l pressure. 128 Table XI. I n i t i a l rates of carbonylation of ruthenium(III) i n aqueous s o l u t i o n at 80°C and 1 atmosphere t o t a l pressure [Ru(III)]xlQ 2, M. 7.40 10.62 14.40 I n i t i a l rate x 10 ,M.s 8.60 11.80 16.00 o 5.0 10.0 2- 1 5 0 2 0 0 F i g . 33. Dependence of i n i t i a l rate of CO uptake by K^RuCAg i n aqueous solution on [Ru]. Experiments at 80°C. and 1 atm. t o t a l pressure. Table XII. Stoichiometries and i n i t i a l rates of carbonylation for ruthenium(III) species, including data in the presence of soda lime 1) Solution aged for 24 hours Stoichiometry at 3000s=0.32:l Stoichlometry at 5000s=0.A8:l I n i t i a l rate=18.9xl0 5M.s 1 2) Freshly prepared solution Stoichiometry at 3000s=0.26:l Stoichiometry at 5000s=*0.A3:1 I n i t i a l rate=16.0x10 6M.s 1 3) Freshly prepared + soda lime Stoichiometry at 3000s=0.53:l Stoichiometry at 5000s=0.80:l (scatter) ^ k) Freshly prepared + soda lime Stoichiometry at 3000s=0.45:l Stoichiometry at 5000s=0.65:l I n i t i a l rate=13.3xl0 6M.6 1 5) Freshly prepared + soda lime Stoichiometry at 3000s=0.29:l Stoichiometry at 5000s=0.49:1 I n i t i a l rate=13.8x10 6M.s 1 ( A l l experiments at [ R u 1 1 1 ] = 0.1AM., T=80°C., Total pressure = 1 atm.) O J I — 1 132 slow and v a r i a b l e (see section 2.3); the soda-lime experiments were discarded i n terms of determining k i n e t i c s (e.g. for the f i r s t carbony-l a t i o n step) but were used to estimate the f i n a l stoichiometry of the reaction. Experiments conducted at lower CO pressures gave the same problems as those noted for experiments with low ruthenium concentrations, and a great deal of scatte r was observed on p l o t t i n g CO uptake against time. Even at 0.14M. ruthenium concentration, the p l o t t i n g of an i n i t i a l rate was d i f f i c u l t . For instance, (Fig. 34) an experiment conducted at a t o t a l pressure of 560 mm Hg, when the p a r t i a l pressure of CO i s one-ha l f that at a t o t a l pressure of 760 mm Hg, and at T=80°C gave an —6 —1 i n i t i a l rate estimated as (18.40±2.00)xl0 M.s and c l e a r l y the CO dependence would be d i f f i c u l t to e s t a b l i s h . Measurements of the pH of the 0.14M.[Ru] solutions a f t e r 5000 seconds under CO co n s i s t e n t l y revealed a 1:1 r a t i o of H + generated per ruthenium, and t h i s r a t i o appeared unchanged when apparent gas uptake of one mole per mole of ruthenium was reached (no soda lime); t h i s indicates formation of R u 1 1 by a process such as that depicted i n equation 4.14. 2 R U 1 1 1 + CO + H 20 > 2 R U 1 1 + C0 2 + 2H + (4.14) The re a c t i o n solutions a f t e r 5000 seconds were generally deep red i n colour and showed between 0.4 and 0.5 moles of net uptake per mole of ruthenium. At the 1:1 uptake stage (8 hours i n the absence of soda lime) i n an experiment with K 2RuC£^H 20 the r e s u l t i n g s o l u t i o n was yellow .and showed [ H + ] / [ R u ] = l and [CSi ]/[Ru]»3 (suggesting some chlorocarbonyl 134 ruthenium(II) species i n s o l u t i o n e.g. "Ru (CO)C£ 2"). This evidence shows that the i n i t i a l process, to the 1:1 stage, i s not simply formation of a Ru 1 1 1(CO) species. When the "gas-uptake" reaction was followed to completion the reaction products appeared to be the same as those found i n preparative experiments (see section 4.1), a colourless f i l t r a t e together with the brown polymeric material and orange coloured s o l i d material adhering to the walls of the f l a s k ; the o v e r a l l gas uptake stoichiometry was 3 moles per mole of ruthenium. Experiments conducted at ruthenium -2 concentrations i n the range 2.5-3.0x10 M. showed a gas uptake s t o i -chiometry of 2 moles per ruthenium a f t e r 40 hours, and between 2 and 3 moles per ruthenium a f t e r 3 to 4 days. An experiment conducted at -2 a [Ru]=6.1xl0 M. showed a 2:1 uptake a f t e r 13 hours that had reached a plateau at 3:1 uptake a f t e r about 16 hours. No further uptake of CO was observed i n the succeeding 2 or 3 hours, even though there appeared to be very l i t t l e orange coloured s o l i d material and no polymer present i n the r e s u l t i n g yellow sol u t i o n . This implies that l a t e r stages of the reaction involve reductive carbonylation to y i e l d the s o l i d products with no net gas uptake (e.g. eq. 4.I5). M n + + H 20 + CO > C0 2 + 2H + + M ( n _ 2 ) + (4.15) The CO uptake by a sol u t i o n of ruthenium(III) at 80°C under one atmosphere t o t a l pressure i n the presence of soda lime was followed -2 for 80 hours (Fig. 35). At the ruthenium concentration of 3.13x10 M. the stoichiometry was observed to reach 6(±0.5) moles of CO per ruthenium V 2 0 , 4 0 . 6 0 T I M E , h o u r s F i g . 35. CO uptake by an aqueous sol u t i o n of K 3RuCil6(6.26xlO 5moles ([Ru]=3.13xlO~?M,) and -1 atm. t o t a l pressure i n the presence of soda 136 i n t h i s time. The pH of the f i l t r a t e at the end of the reaction c o n s i s t e n t l y gave a [H +]:[Ru] value of 3 i n agreement with r e s u l t s from preparative experiments. In addition, use of the c h l o r i d e ion electrode showed that the free c h l o r i d e ion concentration was equal to that i n the s t a r t i n g material. The r e s u l t s suggested a scheme such as that shown below. Colour of reaction mixture: Brown Red Yellow Yellow Yellow (using [Ru] = 2.5x10 M.) "Apparent stoichiometry" : -0.30:1 0.50:1 1:1 2:1 3:1 (in the absence of soda lime) Time : l hr. 2 hrs. 8 hrs. 40 hrs. Species present : 1,2,3 1,2,3 3 4 5 R u 1 1 1 l l R u I I I ( C 0 ) " R u 1 1 Ru i : E(C0) R U 1 1 (CO) Ru I : C(C0) 3 (4.16) Ruthenium(II) species then undergo reductive carbonylation to y i e l d s o l i d products. The net reactions to 3, 4 and 5 are thus considered to be: 137 2 R U 1 1 1 + 3C0 H 2 ° ^ 2 R u I ] : ( C 0 ) + CO,, + 2 H + 3 2CO I X (4.17) TT ?ro 2 Ru (C0) 3 2 Ru" ( C O ) 2 Subsequent reductive carbonylation according to equation (4.18) would account 2RuI]"(C0) + 3C0 + 3H20 =>2HRu(C0)3 + 3C0 2 + 4H + (4.18) for the observed uptake stoichiometrics i n the absence and presence of soda lime, and the f i n a l proton concentration. It i s i n t e r e s t i n g to compare the r e s u l t s of these experiments 2 6 with r e s u l t s previously reported i n ~3M.HC£ s o l u t i o n . The ruthenium (III) solutions then took up CO quite r a p i d l y to give deep red solutions (A =250, 310 and 440 nm) characterised by i s o l a t i o n of the NH.+ s a l t max 3 4 2-of Ru(C0)C£^ . It was only possible to generate more CO uptake by f i r s t reducing the solutions to Ru 1 1(C0) by shaking under hydrogen. Subsequently, 2-carbonylation resulted i n formation of Ru(C0) 2C£ 4 which was characterised by evaporation of the f i n a l yellow s o l u t i o n and analysis of the r e s u l t i n g 2-s o l i d . The i n h i b i t i o n of the aquation of Ru(C0)C£^ by the presence of high i n i t i a l HC& concentrations l i k e l y prevents simultaneous co-ordination of CO and H 20 thereby making reductive carbonylation (eq. 4.19) impossible. (H o0)Ru I I ]" + C O — > Rt^H-CO >Ru I l : t-CO + H + > Ru I ] C I-C00H+H + 1 I I ,0V 0 -2e (4.19) V v. / \ / H H H R u I + C 0 2 + 2 H + 138 James and Rempel 1 6^ have stated that Rh^ 1 1 complexes containing both CO and IL^ O are u n l i k e l y to e x i s t , and the same could well be true f o r R u 1 1 1 . Aquocarbonyl complexes of platinum metals i n oxidation state I II or greater have not been synthesised. The Ru(II) species 2— 2 6 [Ru(CO)C£ 4(H^O) ] i s known t i n d i c a t i n g that the tendency for re-ductive carbonylation i s reduced for lower valent species. This could i n part be due to the lower a c i d i t y of the co-ordinated H^ O (see eq. 4.19). The data i n section 4.5 show that reductive carbonylation at II 26 Ru i s very slow. The CO uptake p l o t s i n ~3M.HC£ solutions were e a s i l y measured, even at low ruthenium concentrations and low CO III -2 pressures, and, for example at 80°C, [Ru ]=2.5xl0 M, p(C0)=161 mm Hg, the k value for the reaction depicted i n equation(4.20]was 0.66M 1 . RuC£ 5(H 20) + CO • 3 U- n L j X' > Ru(C0)C&j: + H 20 (4.20) In purely aqueous s t a r t i n g solutions, the v a r i a t i o n i n pH which r e s u l t s from reductive carbonylation (there i s no evidence f or H + production from hydrolysis of co-ordinated H 20) and the accompanying changes i n [C£ ] and i o n i c strength with t h e i r possible e f f e c t upon the s o l u b i l i t y of CO could well account f o r the observed scatter i n the i n i t i a l gas uptake measurements. 139 Chapter 5 AQUEOUS SOLUTION CHEMISTRY OF CHLOROCMBONYLS OF RUTHENIUM(II) AND RUTHENIUM(III) 5.1. Introduction In Chapter 4 the aqueous solution chemistry of ruthenium(III) chloro species in the presence and absence of carbon monoxide was discussed. The colour and stoichiometry changes observed in the i n i t i a l stages of the carbonylation of ruthenium(III) in aqueous solution suggest the formation of Ru I I I(C0) with subsequent reductive carbonylation to Ru"*"1. In acidic chloride solutions supposedly con-taining ruthenium(II), a deep blue colour is observed 2*^' 1 4^. A number of processes must occur simultaneously in the later stages of the carbonylation of ruthenium(III), but i n i t i a l l y the formation of Ru 1 1 1(CO) w i l l be an important step, and hence the aqueous solution chemistry of Ru 1 1 1(CO) was investigated. 5.2. Preparation of Ru^^CO) species The preparation of Cs2Ru(CO)C£5 was conducted according to the procedure of Cleare and G r i f f i t h . 1 0 5 "Ruthenium trichloride"(1.0 gms. was added to a mixture of concentrated hydrochloric acid (10 mis) and 140 90% formic acid (10 mis). The mixture was refluxed in air for 1 hour, producing a deep red solution. The solution was fi l t e r e d and the resulting f i l t r a t e evaporated to a low bulk. Upon addition of a slight excess of CsC£ to the red f i l t r a t e a deep red-brown precipitate(1) resulted. This precipitate(1) was washed with absolute ethanol and then dried in air. The solid state (nujol mull) infra-red spectrum gave a strong peak at 320 cm 1 due to the Ru-C£ stretching mode, but the absorption in the carbonyl stretching region was weak in intensity and did not compare well with the literature v a l u e . 1 0 5 The results of microanalysis (Calculated for Cs 2Ru(C0)C£ 5: 0,2.10; C£,31.0%. Found: C,0.63; C£,32.4%) suggested that a number of other compounds, such as Cs^RuC^ and CsC£, might also be present. The use of this method of preparation for Cs2Ru(C0)C£ seems to depend upon the particular batch of "ruthenium trichloride" that i s used and so an alternative procedure was sought. In the second procedure "ruthenium trichloride" (1.0 gms.) was dissolved in 25 mis. of 5M.HC£ and brought to 80°C. while hydrogen was bubbled through the solution. The reduction of ruthenium(IV) was followed by sampling the solution at various time intervals and ob-serving the disappearance of a band at 485 nm in the visible spectrum (cf. section 4.2). When this process was completed, hydrogen was replaced by carbon monoxide and heating was continued for a further 12 hours. The result was a deep red solution with a visible spectrum containing maxima at 320,444 and 530 nm, comparable in position to 149 those reported by Kemp (see Table XIII). The solution was reduced 141 to half-volume by evaporation under reduced pressure and upon addition of a stoichiometric amount of CsC£ (1.28 gms) a deep red p r e c i p i t a t e (2) was produced. The p r e c i p i t a t e (2) was f i l t e r e d , washed with absolute ethanol and then dried i n a i r . Although the i n f r a - r e d spectrum (KBr disc) was characterised by stretching frequencies at 2042(s), 2015(vs), 540(m), 475(w), 320(s) and 270(w) cm"1, very s i m i l a r to those reported by Cleare and G r i f f i t h 1 0 5 , the r e s u l t of analysis was once again somewhat low i n carbon (Found: 0,1.84%) and molar e x t i n c t i o n c o e f f i c i e n t s of 5M.HC£ solutions at maxima i n the 149 v i s i b l e spectrum d i f f e r e d considerably from those reported by Kemp (see Table X I I I ) . The t h i r d and most successful method for preparing C S2RU (C0)C£,-involved the use of a pure ruthenium(III) source as s t a r t i n g material. Carbon monoxide was bubbled through a so l u t i o n of K^RuCSL^ (0.10 gms) i n 5M.HC£(10 mis) at 80°C. for a period of 20 hours. The r e s u l t i n g red sol u t i o n was evaporated under reduced pressure to a volume of approxi-mately 1 ml and the addition of a stoichiometric quantity of CsC£(0.08 gms) produced a brick-red p r e c i p i t a t e (3). The p r e c i p i t a t e (3) was f i l t e r e d , washed with absolute ethanol and dried i n a i r to give a y i e l d of 0.11 gms. (86.4% based on the formula Cs2Ru(C0)C£,-) . The s o l i d gave a s a t i s f a c t o r y analysis (Found: C,2.02%) and molar e x t i n c t i o n c o e f f i c i e n t s of 5M.HC£ solutions i n the v i s i b l e spectrum were considered to be con-149 s i s t e n t with those found by Kemp (Table XIII). Samples of Cs 2Ru(C0)C£ 5 prepared by t h i s route were employed f o r aqueous s o l u t i o n studies and CO uptake experiments. Table XIII. Data f or 5M.HC£ solutions of Cs 9Ru(CO)C£ * m a x ( n m ) eM cm (ref.149) eM cm (present work-product(3)) 257 6760 5600 316 1520 1360 440 3560 3400 525 448 370 143 5.3. Aqueous s o l u t i o n chemistry and carbonylation of Cs 2Ru (C0)C£,. According to previous workers 2^»!49 s o i u t £ o n s Q f t h e 2-ammonium s a l t of Ru(CO)C£^ i n 5M.HC£ are stable i n a i r for periods of up to 2 months. Experiments conducted i n purely aqueous solutions containing no added hydrochloric acid revealed that t h i s was no longer the case. Instead, the prominent maxima at 440 and 316 nm i n the v i s i b l e region were observed to decrease i n i n t e n s i t y to produce very l i g h t - g r e y coloured solutions over a period of a few hours at 20°C i n a i r . Figure 36 i l l u s t r a t e s the changes observed i n the v i s i b l e region of the spectrum. Experiments conducted under argon showed the same spectroscopic changes. The disappearance of the bands at 316 and 440 nm was most e a s i l y followed, but the band at 528 nm also decreased i n i n t e n s i t y and the spectrum measured a f t e r 4 hours revealed an absorption continuum. The fa c t that these changes occurred i n a i r at 20°C suggested that l o s s of c h l o r i d e ion by aquation was probably responsible and that reduction to Ru(II) was not occurring. I n i t i a l i s o s b e s t i c points were formed at 330, 375 and 560 nm over Ih hours, but were then observed to d r i f t s l i g h t l y and l a t e r spectra did not pass through these points. This behaviour i s s i m i l a r to that observed III 3-for the Ru C£^ species i n water (Fig. 30) and could be explained by gradual los s of ch l o r i d e ions from the s t a r t i n g complex.. In agreement with t h i s suggestion, the colourless s o l u t i o n produced by running these experiments overnight was found to contain a 5:1 r a t i o of c h l o r i d e ion per ruthenium and a [H +]:[Ru] value of approximately 0.5:1 for experiments III -4 -4 conducted at [Ru (CO)]=4.16x10 M. and 3.90x10 M. Attempts to analyse 144 F i g . 36. V i s i b l e spectroscopic changes of a 2.65x10 M. aqueous solu t i o n of Cs2Ru(C0)C£5 i n a i r at 20°C. 145 the k i n e t i c s involved were unsuccessful. Figure 37 shows plo t s of A vs t,-log(A -A ) vs t, and (1/A -A ) vs t, where A^ = absorbance at ' t 0 0 t 0 0 t 440 nm at time t and A = f i n a l absorbance. Indeed, the p l o t s of A,_ 00 • t against t were e s s e n t i a l l y l i n e a r . The production of acid indicates considerable hydrolysis of some co-ordinated water. Thus, a c h l o r i d e aquation reaction accompanied by some i o n i s a t i o n of co-ordinated H^ O must give r i s e to the l i n e a r p l o t . The changes i n the v i s i b l e spectrum were greatly accelerated by heating the s o l u t i o n to 80°C i n a i r (Fig. 38). Within 300 seconds the s o l u t i o n had changed from deep yellow to pale yellow and a f t e r 500 seconds i t had become completely c o l o u r l e s s . No further changes were observed and measurements on the s o l u t i o n a f t e r 1100 seconds gave [H +]:[Ru]=1.0 and [C£~]:[Ru]=5.0. A re-examination of the c h l o r i d e ion concentration i n s o l u t i o n II I -4 at [Ru (C0)]=7xl0 M. showed that l o s s of the f i r s t c h l o r i d e ion took place i n 8000 seconds at 20°C, and further replacement of c h l o r i d e by water took place at a slower rate. Thus, the spectroscopic changes shown i n Figure 36 are probably due to l o s s of the f i r s t c h l o r i d e ion to give hydroxy- and aquo-chloro species, with low molar e x t i n c t i o n c o e f f i c i e n t s i n t h i s region of the spectrum, rather than some reduction process. The co-ordinated carbon monoxide group would seem to remain on the complex. If l o s s of CO was occurring one would expect some ruthenium(III) aquo- or hydroxy-chloro species to be generated. However, the f i n a l spectra i n the experiments with Cs2Ru(C0)C£^ reveal none of the d e t a i l s exhibited by the u. v. or v i s i b l e spectra T l ^ E x I o f s e c . _ r F i g . 37. Plots of - l o g ^ o(A t_A c o) ( • ), (A^A^,) (-• ) and (A t-A 0 0) ( T ) against time for a ,3..75xlO~^M. aqueous, sol u t i o n .of. Cs 2Ru(CO)C£ 5 i n a i r . a t _2.Q°C. 147 1 4 8 of aqueous solutions of ruthenium(III) chloro species at s i m i l a r - 4 concentrations ( i . e . 3x10 M.)(see section 4.4). The complicated changes involved i n these solutions mitigated against any further study i n t h i s d i r e c t i o n . However, i t i s worth pointing out that III 2-the lack of s t a b i l i t y of Ru (CO)Cl^ i n water would explain why no band at 440 nm i s detected when heating solutions of ruthenium (III) under carbon monoxide (e.g. during the [HRu(C0) 3] n preparation). The i n i t i a l gas uptake experiments on aqueous solutions of Cs2Ru(C0)C£^ were conducted at [Ru 1 1 1(CO)]» 0.01M., at 80°C, under one atmosphere of carbon monoxide. No gas uptake was detected within 10,000 seconds. I l l -2 An experiment at [Ru (CO)] = 4.30x10 M. showed slow gas uptake, req u i r i n g approximately 12,000 seconds to reach the 1:1 stage; sub-sequent carbonylation was even slower. In contrast, an experiment conducted at a s i m i l a r concentration i n the presence of soda lime showed no s i g n i f i c a n t decrease i n rate a f t e r the 1:1 stage. This would suggest that reductive carbonylation to Ru 1 1(C0) occurs, involving i n i t i a l generation of 1 mole of CO2 per mole of CO absorbed. Instant-aneous absorption of CO2 by soda lime i s u n l i k e l y (see section 2.3) but would be r e f l e c t e d by a d i f f e r e n c e i n the observed net uptake rate. In addition, no simple dicarbonyl species of ruthenium(III) has been i s o l a t e d , and i s considered u n l i k e l y on the basis of previous discussion (section 1.1). Experiments conducted at higher [Ru 1 1 1(CO)] ( i . e . 0.14M.) also gave scattered data (Fig. 39). In the absence of soda lime a s t o i c h i o -metry of 1.5:1 was reached i n 20 hours and i n the presence of soda lime 15J0-F i g . 39. CO uptake by an aqueous solution of Cs 2Ru(CO)C£ 5(14.4xlO moles) at 80°C. ([Ru]=14.lxlO~ 2M.) and 1 atm. t o t a l pressure. 150 a stoichiometry of 2.5:1 was achieved at that stage. The f i n a l stoichiometry i n the presence of soda lime appeared to be 3:1 a f t e r 60 hours; and i n both experiments the reaction products were much the same as those observed when commencing with ruthenium(III) chloro species. Evidence up to t h i s point again indicates that the reactions shown i n equations 5.1-5.4 represent the f i r s t steps i n the process leading to the production of the polymer [HRu(C0).j] n by carbonylation of Ru(C0)C£ 2" (cf. eq. 4.16). 2 Ru i : [ I(C0)+ CO + H 20 > 2 Ru I 3 :(C0) + C0 2 + 2H + (5.1) 2 Ru I : E(CO) + 2C0 > 2 Ru i : [(C0) 2 (5.2) 2 Ru I : C(C0) 2 + 2C0 > 2 Ru I 3"(CO) 3 (5.3) Ove r a l l : 2 Ru l : C I(C0) + 5C0 + H 20 3> 2 Ru l : C(C0) 3 + C0 2 + 2H + (5.4) I f , as seems l i k e l y , processes 5.1 to 5.3 overlap to a c e r t a i n extent, the i n t e r p r e t a t i o n of gas uptake measurements becomes very d i f f i c u l t . The i n t e r p r e t a t i o n of f i n a l stoichiometries f o r the R u 1 1 1 and Ru 1 1 1(CO) species w i l l be discussed at greater length l a t e r i n t h i s chapter. C l e a r l y , processes 5.2 and 5.3 w i l l be important and the remainder of t h i s chapter i s concerned with these reactions. 5.4. Preparation of C s ^ u 1 1 (CO) (E20)CI^ III Since the carbonylation of Cs 2Ru (CO)CH^ appeared to r e s u l t i n some Ru I I(C0) species i n sol u t i o n i t was decided that the next step II should be the synthesis of a sui t a b l e Ru (CO) species. Cs 2Ru(C0) (H^CJt^ 151 was made as a by-product i n the synthesis of CsRuCCCO^CJl^ by the method of Cleare and G r i f f i t h . " R u t h e n i u m t r i c h l o r i d e " (2.0 gms.) was r e -fluxed i n a mixture of concentrated hydrochloric acid (15 mis) and 90% formic acid (20 mis) for a period of 24 hours. The addition of a stoichiometric quantity of CsCH to the yellow s o l u t i o n f a i l e d to pro-duce any p r e c i p i t a t e even a f t e r standing overnight (despite the reported p r e c i p i t a t i o n of CsRu(C0) 3C£ 3 at t h i s stage 1 0 5),Excess CsCZ was then added u n t i l the f i r s t signs of cloudiness appeared i n the s o l u t i o n and the mixture was then s t i r r e d to give a pale orange p r e c i p i t a t e (mostly Cs 2Ru(C0) 2CJl 4 by i . r . i d e n t i f i c a t i o n ) . A f t e r f i l t r a t i o n , the supernatant l i q u i d was found to p r e c i p i t a t e a mixture of red (probably CspRutCO)^,.) and green c r y s t a l s on standing. This mixture of s o l i d s i n the yellow f i l t r a t e was refluxed i n a i r for approximately 4 hours and by t h i s means t o t a l conversion to a green coloured p r e c i p i t a t e was achieved. The p r e c i p i t a t e was f i l t e r e d and dried i n a i r to give a green powder. The i n f r a - r e d spectrum (nujol mull) compared well with that reported i n the literature"*" 0 5 for Cs 2Ru(C0)(H 20)C£ 4 and the product gave a good r e s u l t from microanalysis (Calculated for C s 2 R u C 0 2 H 2 C £ 4 : C,2.10;H,0.40%. Found: C, 2.16;H,0.41%.) In addition the v i s i b l e spectrum of t h i s compound i n 5M.HC& had maxima at 315 and 376 nm which 2 6 are comparable with those reported previously. 5.5. Aqueous s o l u t i o n chemistry and carbonylation of Gs 2Ru(C0)(H 20)C£ The green so l u t i o n of Cs 2Ru(C0)(H 20)C& 4 i n water shows very slow and small changes i n colour and v i s i b l e spectrum compared to those ex-152 h i b i t e d by the R u 1 1 1 and Ru 1 1 1(CO) species (Figs. 30 and 36 r e s p e c t i v e l y ) . -3 Figure 40 i l l u s t r a t e s these spectroscopic changes observed i n a 1.96x10 M. so l u t i o n over a period of 14 hours. The f i n a l spectrum e x h i b i t s maxima at 310 and 355 nm, a s h i f t that Is consistent with replacement of c h l o r i d e ion by water molecules. No more changes are observed from 12 to 14 hours a f t e r making up the s o l u t i o n , and the f i n a l s o l u t i o n i s e s s e n t i a l l y colour-- -3 + -5 l e s s with [C£ ]=1.9xl0 M. and [H ]=3.8xl0 M. Therefore, the s o l u t i o n would appear to contain Ru(C0)C£ 3 ( H 2 0 ) 2 with very l i t t l e production of hydroxy species as indicated by the comparatively high pH of 4.32 i n the f i n a l s o l u t i o n (eq. 5.5). Ru(CO)(H 20)C£ 2 _ + H 20 > Ru(CO)(H^O) 2C£~ + C£~ (5.5) Carbonylation of solutions of Ru 1 1(C0) took place slowly. At a concentration of 4.15x10 M i n water at 80°C under one atmosphere of carbon monoxide, l e s s than 0.5 moles net gas uptake had occurred i n the f i r s t 8000 seconds. Over a period of 20 hours under these conditions the uptake had increased to nearly one mole per mole of ruthenium, and at the end of t h i s time the reaction mixture already appeared to contain some of the f i n a l polymeric products. Once again, t h i s indicates that several steps i n the o v e r a l l process are occurring simultaneously. The measured pH of the s o l u t i o n a f t e r 20 hours was equal to 1.70, corresponding to [H +]=2.1xlO~ 2M or [H +]:[Ru 1 1(CO)]=0.5. Since complete reduction of Ru 1 1(C0) to lower valent ruthenium products would generate at l e a s t a 1:1 r a t i o of H + to i n i t i a l Ru 1 1(CO), t h i s i s another i n d i c a t i o n that the reaction has not gone to completion, assuming that a l l H + generated 153 F i g . 40. V i s i b l e spectroscopic changes of a 1.96x10 M. aqueous s o l u t i o n of Cs 2Ru(C0) (H 20)CJ1 4 i n a i r at 20°C. 154 remains i n s o l u t i o n (that i s , no H + i s consumed i n metal hydride formation). Experiments at higher [Ru 1 1(CO)] were l i m i t e d owing to the r e l a t i v e i n s o l u b i l i t y of Cs 2Ru(CO)(H 20)C£ 4 i n water. A 0.14 M. solut i o n could not be achieved, even a f t e r shaking at 80°C f o r 3000 seconds and CO uptake appeared to occur very slowly. Experiments on d i - and t r i - c a r b o n y l species of ruthenium(II) were then investigated. 5.6. Preparation of Cs 2Ru(CO) 2C£ 4 The general procecure of Cleare and G r i f f i t h 1 0 5 was followed. "Ruthenium t r i c h l o r i d e " (2.0 gm.) was refluxed i n a mixture of con-centrated hydrochloric acid (20 mis) and 90% formic acid (20 mis) u n t i l the s o l u t i o n became golden-yellow i n colour. This process took between 5 and 8 hours, and at the end of t h i s time a stoichiometric amount of cesium ch l o r i d e was added (2.9 gms). The so l u t i o n was evaporated to low bulk to give Cs 2Ru(C0) 2C£ 4 (1.25 gms or 30% y i e l d ) as a yellow p r e c i p i t a t e . The s o l i d was f i l t e r e d , washed with absolute ethanol and dried i n a i r to give a yellow powder which gave an i n f r a - r e d (nujol mull) spectrum which agreed with the l i t e r a t u r e 1 0 5 and a correct microanalysis (Calculated for Cs 2RuC 20 2C£ 4:C,4.25;C£,25.13%. Found: C,4.36;C£,25.22%). In addition, the v i s i b l e spectrum i n 5M.HC£ sol u t i o n showed one band with X =380 nm r max -1 -1 -1 -1 with a molar e x t i n c t i o n c o e f f i c i e n t of 450 M cm (cf.e=400M cm recorded 149 by Kemp for (NH 4) 2Ru(CO) 2C£ 4). 5.7. Aqueous so l u t i o n chemistry and carbonylation of Cs 2Ru(C0) 2C£ 4 Z3 ~ -—• A 1.4x10 M so l u t i o n of Cs 2Ru(CO) 2C£ 4 i n water at 20°C. showed only 155 very slow and small changes i n the v i s i b l e spectrum (Fig. 41). The i n i t i a l band at 376 nm (e =454) gradually s h i f t e d to 371 nm (e =408) max ' b J M A X over a period of 12 hours. At the end of t h i s time the pH was 4.61 ( i . e . [H +]=1.3xl0 5 M.) and use of the s p e c i f i c c h l o r i d e ion electrode gave - -3 [C£ ]=1.2xl0 M. Therefore, i t seems l i k e l y that the process represented i n equation 5.6 occurs with very l i t t l e subsequent production of hydroxo complexes. R u I I ( C 0 ) 2 C £ 2 ~ + H 20 & R u I : i : ( C O ) 2 C £ 3 ( H 2 0 ) ~ + C£~ (5.6) An experiment conducted at 80°C gave s i m i l a r spectroscopic changes within 1 minute, and over a period of 12 hours the band maximum moved to 353 nm corresponding to further hydrolysis and los s of remaining chloride ligands. The conductivity of a f r e s h l y prepared s o l u t i o n of Cs 2Ru(C0) 2C£ 4 (AM=316 at 25°C) also serves to i l l u s t r a t e the slowness with which the i n i t i a l c h l o r i d e aquation occurs; t h i s conductivity value i s i n the correct 159 range for 3 ions i n s o l u t i o n , showing a subsequent slow increase i n value as hydrolysis (equation 5.6) takes place. The slow rate of aquation II I 3- III 2-observed with the Ru(II) species when compared with Ru C^ -g and Ru (C0)C£<. species may be at l e a s t p a r t i a l l y understood i n terms of c r y s t a l f i e l d theory, although the ligands are not i d e n t i c a l i n each case. Aquation of Ru(II)(with d electron configuration t 2 ) whether proceeding by S^l or S^2 mechanism, w i l l i n e v i t a b l y lead to a considerable los s of c r y s t a l f i e l d s t a b i l i s a t i o n energy and hence w i l l be slow. In discussing rates i n t h i s manner i t should be borne i n mind that c r y s t a l f i e l d energies are only a small part of the bonding energy i n any system. Large contributions to the a c t i v a t i o n energy 156 F i g . 41. V i s i b l e spectroscopic changes of a 1.40x10 M. aqueous so l u t i o n of Cs 2Ru(CO) 2C£ 4 i n a i r at 20°C. 157 because of changes In metal-ligand a t t r a c t i o n , ligand-ligand repulsions, 3- 2- 2-etc. , p a r t i c u l a r l y i n a series such as RuC£^ , Ru(CO)C£,- and RuO^O^CJl^ , must also be considered. An increasing number of CO groups co-ordinated to the k i n e t i c a l l y i n e r t ruthenium(II) may lead to enhanced rates of aquation by a c t i v a t i o n of chlo r i d e ligands trans to carbon monoxide. C h a t t ^ 6 5 and O r g e l 1 6 6 have proposed a theory based on Il-bonding s t a b i l i -sation of the activated complex to account for the large " t r a n s - e f f e c t " exhibited by I I-acceptor ligands such as C^^-^, PR3 and CO. Carbonylation of Cs2Ru(CO)2C£ 4 i n water was i n i t i a l l y studied at low ruthenium concentrations (Table XIV). The slow rate of gas uptake made rates d i f f i c u l t to measure, but the net uptake approached 1:1 over extended periods of time and the contents of each reaction f l a s k appeared s i m i l a r to the products from the synthesis of [HRu(CO) 3] n (section 3.1). The f i n a l aqueous l i q u i d was colourless and contained the insolu b l e poly-meric material. The remaining orange reaction products appeared to have mostly sublimed on to the walls of the reaction f l a s k . In each experiment (1),(2) and (3)(Table XIV) the f i n a l s o l u t i o n was f i l t e r e d and t i t r a t e d p otentiometrically with standardised AgNO^ so l u t i o n . The r e s u l t s gave good p l o t s and i t was evident that a l l of the i n i t i a l C£ i n the complex remained i n the f i n a l c olourless s o l u t i o n . pH measurements on the f i l t r a t e indicated an [H +]: [Ru] value of 2 a f t e r a period of 2 days. In each experiment the s o l i d r eaction products were at l e a s t p a r t i a l l y soluble i n n-hexane to give yellow solutions with peaks i n the carbonyl stretching region of the i n f r a - r e d spectrum at 2114(vw),2078(s),2071(s),2063(m), 2032(s),2013(w) and 2000(vw) cm"1 (cf. section 3.4). 158 Table XIV. CO uptake stoichiometries for Cs„Ru(CO)0C£ [Ru]xlO M T°C. Pco(mmHg) Time(hours) Stoichiometry(net uptake/Ru) (1) 1.04 50 288 0.06 0.05 0.82 0.12 19.0 1.0 (2) 0.77 70 146 3.3 0.38 4.8 0.56 22.0 1.00 (3) 3.1 80 405 1.28 0.2 2.40 1.0 159 Studies on the carbonylation of Cs2Ru(CO)2C£4 at higher [Ru] were more informative. When the [Ru] was ^ 6x10 M., S-shaped CO uptake p l o t s were obtained (Fig. 42). As the [Ru] was increased, the i n i t i a l period of time before achieving the maximum rate decreased. Thus, at 1 atm t o t a l pressure, 80°C and [Ru]=0.14M. the maximum rate was reached a f t e r only 750 seconds, whereas i n an experiment at [Ru]=0.06M. approximately 1700 seconds was required to reach the equivalent p o s i t i o n i n the CO uptake p l o t . In each case the stoichiometry of the uptake per ruthenium eventually approached 1:1. There appeared to be an i n i t i a l stage of reaction under CO during which no uptake of gas occurred; at [Ru]=0.14M., T=80°C and 1 atm t o t a l pressure t h i s i n t e r v a l was approximately 300 seconds. Upon increasing [Ru] to 0.17M and 0.2M t h i s induction period was reduced to approximately 200 and 100 seconds r e s p e c t i v e l y . Moreover, when • Cs2Ru(C0) was added to water under vacuum, shaken f o r 1000 seconds at 80°C and then exposed to the CO atmosphere, the induction period persisted. In a separate 2-experiment i t was shown that the R u ^ O ^ C ^ species under CO at 80°C loses a l l co-ordinated CH within 1000 seconds. Thus, i t seems that the induction period associated with t h i s r eaction must involve some i n t e r a c t i o n between CO and ruthenium(II) species i n s o l u t i o n rather than some so l u t i o n e q u i l i -bration of the complex. When the reaction at [Ru]=0.14M was stopped a f t e r 300 seconds the contents of the reaction f l a s k consisted of a deep yellow s o l u t i o n and minute amounts of orange s o l i d . This evidence favours some II i n i t i a l reductive carbonylation of Ru (00)2 presumably to some neutral species, as well as co-ordination of CO to give Ru 1 1(C0) 3 species. The formation of small amounts of a reduced species by reductive carbonylation 30.0 TIME x 1 0 s sec F i g . 42. CO uptake by an aqueous solution of Cs2Ru(CO) 2C£4(28.0xlO moles at 80°C. ([Ru]=14.1xlO"2M.) and 1 atm. t o t a l pressure. \ 161 would involve no apparent consumption of CO (see equations 5.7, 5.8 and 5.9). TT H 2 ° I + 2 Ru (C0) 2 + CO — > 2 Ru (C0) 2 + C0 2 + 2H (5.7) HO 2 Ru (C0) 2 + CO —>, 2 Ru°(CO) 2 + C0 2 + 2H (5.8) TT 2S° n + 2 Ru (CO) 2 + 2C0 =—> 2 Ru (CO) 2 + 2C0 2 + 4H (5.9) The production of Ru° species by combination of Ru 1 and Ru 1 species (obtained by further reductive carbonylation) i s also conceivable. The use of a soda-lime tube while i n v e s t i g a t i n g CO uptake during the i n i t i a l stages of the reac t i o n made very l i t t l e d i f f e r e n c e to eit h e r the induction period or the maximum rate of gas uptake at low ruthenium concentrations. However, at [Ru]=0.14M, T=80°C and 1 atm .tot a l pressure the gas uptake per ruthenium reached 1:1 a f t e r 4000 seconds, 1.5:1 a f t e r 13,000 seconds and 2:1 a f t e r 36 hours. Beyond t h i s stage l i t t l e further uptake was observed and a f t e r 60 hours (2.2:1 uptake) the reac t i o n products appeared to consist mainly of the ins o l u b l e polymer [HRu(C0) 3] n, i d e n t i f i e d as such by i n f r a - r e d spectroscopy, and also a colo u r l e s s s o l u t i o n . Benzene-soluble products appeared to be present i n only trace amounts and measurements on the f i n a l aqueous s o l u t i o n revealed [CI ]:[Ru]=4.0 and [H +]:[Ru]=2.0. The r e s u l t s are consistent with a scheme such as that proposed i n equations 5.7 and 5.8 followed by reactions such as those depicted below. 2 Ru°(C0) 2 + CO + H 20 > 2 R u - 1 ( C 0 ) 2 + C0 2 + 2H + (5.10) (or Ru 1(C0) 2 + CO + H 20 > R u _ 1 ( C 0 ) 2 + C0 2 + 2H +) (5.11) 162 2 Ru (CO) 2 + 2H > 2HRu(CO) 2 (5.12) 2HRu(CO) 2 + 2C0 •> 2HRu(CO)3 (5.13) The o v e r a l l equation obtained by combining equations 5.7, 5.8, 5.10, 5.12 and 5.13 i s Equation (5.14) explains the observed [H1"] : [Ru] value i n the f i n a l aqueous sol u t i o n , the expected 1:1 stoichiometry of CO uptake i n the absence of soda lime and also predicts a 2.5:1 CO uptake i n the presence of soda lime. The fact that the stoichiometry never quite reaches 2.5:1 i n the presence of soda lime could be a t t r i b u t e d to the very slow nature of the reaction i n i t s f i n a l stages, i n addition, any Ru° product w i l l r e s u l t i n a lower t o t a l CO uptake. Equation (5.14) contains the species HRu(C0) 3 which can presumably polymerise to give the [HRu(C0) 3] n product. The formation of [HRu(C0) 3] n must require p r i o r reduction of some species i n s o l u t i o n to the Ru(-l) oxidation state, since the only s i g n i f i c a n t source of co-ordinated hydride i s protons generated i n s o l u t i o n by the process of reductive carbonylation, and so equation 5.12 i s postulated. The o r i g i n of the benzene-soluble products remains i n doubt but presumably the presence of R u ( - l ) , Ru(l) and Ru(II) species i n s o l u t i o n could lead to the formation of c l u s t e r complexes. The i n f r a - r e d spectra of these products c o n s i s t e n t l y resembled those of previously reported polynuclear species (see Chapter 3. Table VII), p a r t i c u l a r l y i n the number of peaks reported, but the observed stretching frequencies never quite matched those of any reported complex. + (5.14) An attempt was made to study the k i n e t i c s of t h i s system using 163 maximum rates obtained from- the S-shaped uptake p l o t s . The data from v a r i a t i o n of [Ru] are presented i n Table XV; [Ru] r e f e r s to the i n i t i a l 2-Ru^O^CA^ concentration. The plo t of maximum rate against [Ru] i s shown i n Figure 43. At ruthenium concentrations l e s s than 0.17M a roughly f i r s t - o r d e r dependence on [Ru] i s apparent, but at [Ru]^0.2M the reaction rate i s higher than expected on a f i r s t - o r d e r basis. Of 2 i n t e r e s t , a p l o t of maximum rate against [Ru/2] gave a reasonable straight l i n e p l o t (Fig. 44). Such a k i n e t i c dependence might be expected at the 50% reaction point i f the reaction i s catalysed by the products of reductive carbonylation and the rate law i s dominated by a term such as [Ru 1 1][Ru 1] or [Ru 1 1][Ru°]. Previous workers h a v e 1 ^ 8 ' 1 6 7 obtained s i m i l a r gas uptake p l o t s from the carbonylation of RhBr^O^O^ i n aqueous acid s o l u t i o n , both i n i t i a l and maximum rates being analysed to give a rate law dominated by the term [ R h 1 1 1 ] [ R h 1 ] at the 50% reaction point. In the case of R u ^ ^ O ^ reduction, the i n i t i a l gas uptake rates were too slow to be of assistance i n e s t a b l i s h i n g a complete rate expression. It i s possible that the p r e c i p i t a t i o n of s o l i d s , -which occurred e s p e c i a l l y during the maximum rate portion of the CO uptake, might r e s u l t i n occlusion of CO within the reaction mixture and give r i s e to some uncertainty i n the rate measurements. Problems i n i n t e r p r e t i n g k i n e t i c data were also encountered when the e f f e c t of pressure v a r i a t i o n was studied, although i t appeared that there was a zero-order dependence on [CO] at p a r t i a l pressures above 300 mm Hg and a f i r s t order dependence at pressures below t h i s value (Table XVI and F i g . 45). 164 Table XV. E f f e c t of varying [Ru] on the maximum rate of carbonylation of Cs 2Ru(CO) 2C£ 4 [Ru]xl0 2M [^y] 2xl0 3M 2 Max.rate x loV s 1 6.1 10.0 14.0 14.6 17.0 20.0 21.6 14.1^ 0.9 2.5 4.9 7.23 10.0 12.1 0.62 1.32 1.53 1.64 2.30 3.40 3.64 1.61 A l l experiments conducted at 80°C under 1 atmosphere t o t a l pressure. ( - soda lime tube employed; T - experiment conducted i n DO) 3.0 [jRuli M. F i g . 43. Dependence of maximum rate of CO uptake by "Cs"2'Rvi'(CO)2CA'2iin aqueous sol u t i o n on [Ru]. Experiments at 80°C. and 1 atm. t o t a l pressure. 167 Table XVI. E f f e c t of varying [CO] on maximum rate of carbonylation of Cs2Ru(CO)2C£ 4 4 -1 P (total)™11 H g P ( C 0 ) m m H g M a X ' r a t e x 1 0 M > s 760 405 1.53 660 305 1.54 610 255 1.32 560 205 0.83±0.05 355 100 0.58±0.08 ( A l l experiments at [Ru]=0.14M and T=80°C) 168 F i g . 45. Dependence of maximum rate of CO uptake by Cs 2Ru (CO^CJ!,^ i n aqueous s o l u t i o n on the p a r t i a l pressure of CO. Experiments at [Ru]=0.14M. and T=80°C. 169 The decrease i n p a r t i a l pressure of CO had no great e f f e c t upon the general shape of the gas uptake p l o t (Fig. 46). The experiments conducted at the lower CO pressures gave rather scattered uptake p l o t s . 2-The complexity of the k i n e t i c s i n the Ru^O^CA^ system was emphasised by the r e s u l t s of a temperature v a r i a t i o n study (Fig. 47). When the rea c t i o n was conducted at [Ru]=0.14M and 1 atm. t o t a l pressure the i n i t i a l part of the reaction was found to be remark-ably temperature dependent. An increase i n reaction rate took place a f t e r 450 secondsat 80°C, 650 seconds at 75°C and 2500 seconds at 70°C. In each experiment the maximum rate obtained from the l i n e a r portion -4 -1 of the CO uptake p l o t was approximately the same (i.e.»1.5x10 M.s ). At the lower temperatures the contents of the reaction f l a s k at the 0.5:1 stage appeared the same as i n the equivalent stage at 80°C. The sol u t i o n remained yellow i n colour, i n d i c a t i n g remaining Ru^^CO^ species and the orange s o l i d products obtained from reduction of the s t a r t i n g complex were also present. The e f f e c t s of a l t e r i n g the pH and chl o r i d e ion content of the i n i t i a l reaction s o l u t i o n are shown i n Table XVII and the r e s u l t s of experiments (A)', (C) and (D) are shown i n Figure 48. The gas uptake plo t from experiment (A) gave a very sudden increase i n rate a f t e r about 400 seconds, but the rate decreased quite r a p i d l y thereafter and 1:1 stoichiometry was reached a f t e r 2500 seconds (cf. F i g . 42). The increase i n the maximum CO uptake rate i n the presence of pH=7.0 buffer i s reasonable i f processes such as those shown i n equations 5.7 and 5.8 represent the i n i t i a l stages of the reaction, although c l e a r l y t h i s w i l l F i g . 47. E f f e c t of temperature v a r i a t i o n on CO uptake by Cs 2Ru(C0) ?C£ 4 (28.0xl0~ 5moles) i n aqueous s o l u t i o n , [Ru]=14.0xl0 2M., under 1 atm. t o t a l pressure. Insert has the same units as the major axes. 172 Table XVII. E f f e c t of varying pH and [CI ] on the carbonylation of Cs 2Ru(CO) 2C£ 4. 4 -1 Experiment Nature of reaction s o l u t i o n Rate x 10 M.s (A) pH=7.0(phosphate buffer) (B) 5M HC£ (C) 3.3M KC£ (D) 11.0M CsC£ (E) Saturated KC£(6.85M) (F) 2.0M KC£ •'. 3.32 1.38 0.69 Stoichiometry 1:1 i n 3000 second 0.14:1 i n 3000 second 0.8 :1 i n 1500 second 0.66:1 i n 2000 second 0.68:1 i n 2700 second 0.89:1 i n 2700 second ( A l l experiments at [Ru]=0.14M, T=80°C and P, , N=760 mm Hg) 30.0 ®2Q.Q fi.10.0 (A)-28-0x10 moles Ru -5 (0-14-0x10 moles Rjs B - • ~ (D)-14.0x10~ 5moBes 2.0 see F i g . 48. CO uptake by aqueous solutions of"cs 2Ru"(CO) 2C£4([Ru]=14.0x10 2M.) i n the presence of (A) PH=7.0 buffer, (C) 3.3M.KC£ and (D) ll.OM.CsCA at 80°C. and 1 atm. t o t a l pressure (see Table XVII). F =>=.ui.c v ^ 174 depend upon the p a r t i c u l a r mechanism involved (e.g. reductive carbon-y l a t i o n requires co-ordinated or free hydroxide). The gas uptake pl o t s obtained from reactions i n the presence of excess C£ were somewhat d i f f e r e n t i n appearance (Fig. 48). The i n i t i a l induction period remained the same, but a f t e r the f i r s t few hundred seconds a l i n e a r gas uptake was observed with no evidence of any autocatalysis and the reaction appeared to stop abruptly at the stoichiometries indicated i n Table XVII. Furthermore, the product of reactions (C) and (D) was a yellow s o l u t i o n containing no s o l i d products. When the reaction s o l u t i o n from experiment (D)was cooled under CO over a period of several hours a white c r y s t a l l i n e p r e c i p i t a t e was slowly formed. The p r e c i p i t a t e was f i l t e r e d from the remaining yellow s o l u t i o n and dried under vacuum. An i n f r a - r e d spectrum (nujol mull) of t h i s s o l i d revealed peaks i n the carbonyl stretching frequency region at/ 2133(s), 2071(s), 2057(s) and 2043(s) cm"1 which showed, by comparison with data from Cleare and G r i f f i t h " * " 0 5 , that CsRu(CO) 3C£ 3 had been produced. The pH of the f i l t e r e d yellow s o l u t i o n was 1.53, equi-+ -2 valent to [H ]=2.7xlO M. Presumably, t h i s acid concentration a r i s e s from the formation of some hydroxy species i n s o l u t i o n , since the process of reductive carbonylation does not appear to occur under these conditions. The evidence strongly suggests that i n the presence of excess 2-chloride the hydrolysis of Ru(CO) 2C£ 4 i s not favoured, and under CO an equilibrium such as that shown i n equation 5.15 e x i s t s i n s o l u t i o n . The absence of any reduction under such circumstances seems reasonable, 2- k1 - -Ru(CO) 2C£ 4 + CO 5======^ Ru(CO) 3C£ 3 + C£ (5.15) 175 since the reduction of metal complexes by CO i s thought to involve at some stage an " i n s e r t i o n r e a c t i o n " of the CO molecule between the lack of any evidence f o r autoc a t a l y s i s i s consistent with equation 5.15 and probably indicates tha"t Ru(I) or Ru(0) i s causing the auto-c a t a l y t i c e f f e c t i n the absence of added c h l o r i d e . The observed induction period i s approximately 300 seconds ( i . e . before any gas uptake occurs) i n a l l of the experiments at 80°C. The f a i r l y sudden cessation of the reaction, the extent of which seems dependent on the amount of added chl o r i d e , could r e s u l t from an equilibrium such as that depicted i n equation 5.15 i n an environment i n which neither the dicarbonyl nor the t r i c a r b o n y l species can be reduced. Furthermore, the i s o l a t i o n of CsRu(CO) 3C£ 3 as a p r e c i p i t a t e from experiment (D) i s consistent with t h i s i n t e r p r e t a t i o n . The adverse e f f e c t on the carbonylation rate produced by addi t i o n of a very large excess of chl o r i d e i n experiment (D) can be r a t i o n a l i s e d i n terms of the k i n e t i c s expected for such a system. If k^ and ^ a r e the rate constants for the forward and back reactions r e s p e c t i v e l y ( i n equilibrium 5.15), the rate of reaction may be expressed by: Thus, adding excess ch l o r i d e ion to the system w i l l i n e v i t a b l y lead to the metal and co-ordinated water or hydroxide 168-170 In addition, (5.16) a reduction i n the o v e r a l l rate expressed as -d[C0] dt In addition, the equilibrium constant(K) may be expressed by K = [Ru(CO) 3C£ 3 ][C£ ] (5.17) 176 An increase i n the t o t a l concentration of free c h l o r i d e ion i n s o l u t i o n w i l l i n e v i t a b l y lead to a reduction i n the equilibrium - 2-value of [Ru(CO) 3C£ 3 ]:[Ru(CO) 2C& 4 ] i n order to maintain K as a constant, and hence w i l l lead to a lower f i n a l stoichiometry of CO uptake per ruthenium. When the concentration of added KC£ was s u f f i c i e n t to exceed the s o l u b i l i t y of that compound i n water at 80°C(i.e. 0.51 gms/ml) a plot of CO uptake against time gave much scat t e r , supporting previous suggestions that the presence of any s o l i d material might influence the v a l i d i t y of any k i n e t i c i n t e r -pretations. The f i n a l stoichiometry for experiment (E) under such conditions ( i . e . a saturated s o l u t i o n of KCl at 80°C, equivalent to 6.85M. KC£) was 0.68:1 a f t e r 2700 seconds and showed no further change. The r e s u l t of a further experiment (F) was a f i n a l s t o i c h i o -metry of 0.89:1 i n the presence of 2.0M KCl. Assuming a 1:1 s t o i c h i o -metry for reaction 5.15 and that the concentration of CO i n these 26 solutions i s equal to that i n 3MHC& , the concentrations of the various species at f i n a l equilibrium from experiments (C), (E) and (F) and the corresponding K values (eq. 5.17), can be calculated (Table XVIII). The mean value of K=4.39±0.40x10 c l e a r l y favours the formation of the t r i c a r b o n y l species i n s o l u t i o n . 2-The apparent lack of any s i g n i f i c a n t carbonylation of RutCO^CJ!^ i n the presence of excess hydrochloric a c i d (experiment (B)) i s consistent 2 6 with the observations of previous workers. The yellow s o l u t i o n i n 5M.HC£ obtained a f t e r 3000 seconds under carbon monoxide (experiment(B) i n Table XVII) p r e c i p i t a t e s the cesium s a l t of Ru(GO) (H.-.O)^ , by l o s s 177 Table XVIII. Data from the carbonylation of Cs 2Ru(CO) 2C£^ i n the presence of excess [C£ ] Experiment [Ru(CO) 2C£ 4 ]M [Ru(CO) 3C£ 3 ]M [CO]xlO 4M [C£ ]M KxlO 4 (C) 0.028 0.112 3.4 3.34 3.93 (E) 0.045 0.095 3.4 6.85 4.25 (F) 0.015 0.125 3.4 2.03 4.98 178 of CO, on standing i n a i r over a period of several days. Since rea c t i o n 5.15 as w r i t t e n involves no a c i d dependence, the observed acid i n h i b i t i o n i n d i c a t e s reactions v i a hydroxy intermediates, f o r example Ru(CO) 2C£ 2" + H 20 ^_ ^.Ru(CO) 2C)l 3(H 20)~ + Cl~ (5.18) - - ro 2- + OH + Ru(CO) 3C£ 3 ; = = = Ru(C0) 2C£ 3(0H) + H V reductive carbonylation The process i s c l e a r l y not as simple as written i n equation 5.15 since t h i s would lead to an o v e r a l l pseudo f i r s t order CO uptake p l o t . An examination of the benzene-soluble products obtained from 2-the reductive carbonylation of Ru(CO) 2C£ 4 at [Ru]=0.14M under varying conditions gave remarkably consistent r e s u l t s . In Chapter 3 (section 3.4) i t was suggested that the by-products from the synthesis of [HRu(C0) 3] n represented a mixture of polynuclear ruthenium carbonyl species and the evidence from mass spectrometry and i n f r a - r e d spectro-scopy c e r t a i n l y supported t h i s idea. In Figure 49 the i n f r a - r e d spectra of several products i n n-hexaneare compared; (A) i s the spectrum obtained from the by-products of an [HRu(CO) 3] n synthesis conducted over a period of 32 hours, (B), (C) and (D) are the spectra 2-of hexane-soluble products from the carbonylation of Ru(CO) 2C& 4 under the conditions noted. The spectra are s t r i k i n g l y s i m i l a r and appear to have the peaks i n common at 2115, 2089, 2079, 2071, 2063, 2033 and 179 2010 cm 1 . The 2115 cm 1 peak i s very weak i n a l l of the spectra, the peak at 2089 cm 1 i s strong i n i n t e n s i t y except f or the r e s u l t (A) from the preparative experiment, the peak at 2079 cm 1 i s strong i n a l l of the spectra. The peak at 2071 cm 1 v a r i e s somewhat i n i n t e n s i t y and i s strongest at the end of the preparative experiment, the same i s true f o r the peak at 2063 cm 1 and seems to argue against the presence of an appreciable amount of Ru^CCO)^ production i n the 2-RuCCO^CJt^ carbonylation (see Table VII i n Chapter 3 and discussion of sublimation experiments i n section 3.4). The peak at 2033 cm 1 i s usually the strongest and shows evidence of a shoulder at lower f r e -quency, although t h i s was never completely resolved. The peak at 2010 cm 1 i s present i n a l l of the spectra and i s c o n s i s t e n t l y of medium i n t e n s i t y . In experiments conducted at lower [Ru], the CO uptake was studied over longer periods of time and i n f r a - r e d spectra of n-hexane extracts from these experiments gave s i m i l a r r e s u l t s , except that the peaks at 2114 and 2089 cm 1 were no longer v i s i b l e and the peak at 2071 cm 1 appeared to be much more intense. The decrease i n i n t e n s i t y of the 2089 cm 1 peak and the increase i n the i n t e n s i t y of the peak at 2071 cm 1 seem c h a r a c t e r i s t i c of the l a t e r stages of these carbony-l a t i o n reactions. The spectrum (E) from the n-hexane extract of the experiment conducted i n pH=7.0 buffer reveals s i m i l a r d e t a i l s . Once again the product at the 1:1 stage shows a strong peak at 2089 cm 1 and the peak at 2071 cm 1 i s barely v i s i b l e as a shoulder on the peak at 2079 cm The r e s u l t of a s i m i l a r extraction of the products from 2-a carbonylation of Ru^O^CA^ i n D20(F) revealed an e s s e n t i a l l y i d e n t i c a l 180 A F i g . 49. Infra-red spectra i n n-hexane. (A) n-hexane extract from [HRu(C0)3] n synthesis a f t e r 32 hours; (B), (C) and (D) n-hexane extracts from Cs2Ru(CO)2C£^ carbonylation at 0.5'1, 0.9<1 and 0.95:1 CO uptake stages r e s p e c t i v e l y . (E) n-hexane extract from Cs2Ru(CO)2C£4 carbonylation i n pH=7.0 buffer and (F) i n D2O. Numbers r e f e r to wavenumbers (cm--'-) for each peak. 181 spectrum to those recorded i n experiments (C) and (D). The 2115 cm 1 peak was not evident i n t h i s case (F), but was previously observed 45 as being of very weak i n t e n s i t y , and i n any case Johnson and others have shown that the replacement of hydrogen by deuterium i n a poly-nuclear hydridocarbonyl structure need not ne c e s s a r i l y lead to very great changes i n the 2200-1800 cm 1 region of the i n f r a - r e d spectrum. Thus, the evidence from experiment (F) does not r u l e out the presence of a hydridocarbonyl i n the n-hexane solutions. 5.8. Preparation of CsRu(CO)-}C£3 CsRu(00)^0^2 preparations were attempted by the method of Cleare and G r i f f i t h . 1 0 5 This technique ( b r i e f l y mentioned i n section 3.3) involves r e f l u x i n g "ruthenium t r i c h l o r i d e " (2.0 gms) i n a mixture of concentrated hydrochloric acid (15 mis) and 90% formic acid (20 mis) for a period of 24 hours, and subsequent p r e c i p i t a t i o n of the desired product by the addition of a stoichiometric amount of cesium c h l o r i d e (1.30 gms). However, no s o l i d s p r e c i p i t a t e d , even a f t e r extended r e f r i g e r -ation, unless an excess of cesium ch l o r i d e was added. In addition, the ..solid thus p r e c i p i t a t e d could be i d e n t i f i e d by i n f r a - r e d spectroscopy 2- 2-as c o n s i s t i n g l a r g e l y of the cesium s a l t s of R u ^ O ^ C ^ and Ru(CO) ( ^ 0 ) 0 ^ However, increasing the r e f l u x time to 2 or 3 days gave a yellow s o l u t i o n from which the product CsRu(CO) 3C£ 3 could be p r e c i p i t a t e d as a white c r y s t a l l i n e powder by the ad d i t i o n of an excess of cesium c h l o r i d e . This material was then r e c r y s t a l l i s e d from a 1:1 mixture of acetone and l i g h t 182 petroleum ether to give approximately 0.80 gms (25% y i e l d ) of CsRu(CO)^CJl^. The i n f r a - r e d spectrum (nujol mull) compared well with the l i t e r a t u r e 1 0 5 , peaks occurring at 2140(s), 2078(s), 2060(s), 2045(s), 620(m), 580(m), 483(m), 477(m), 320(m) and 285(m) cm ". The three peaks i n the 2040-2080 cm"1 region of the carbonyl stretching frequency range represent the e f f e c t of s o l i d - s t a t e s p l i t t i n g on the asymmetric C-0 s t r e t c h and were also previously reported by Cleare and G r i f f i t h . 1 0 5 ( s e e also Table IV i n section 3.3). The compound was s u f f i c i e n t l y soluble i n water to enable an i n f r a - r e d spectrum i n t h i s solvent to be obtained with carbonyl stretching bands at 2145(s) and 2075(s) cm 1 . In addition, a so l u t i o n obtained by warming 0.0155 gms (3.65x10 5 moles) of the compound i n 3 mis of absolute ethanol showed peaks at 2123(s) and 2055(vs) cm ". Microanalysis also gave good data (Calculated for CsRuC 30 3C£ 3:C,8.47;C£,25.05;Ru,23.88%. Found: C,8.51;C£,25.22; -4 Ru,24.16%). Aqueous solutions of CsRu(C0) 3C£ 3 from [Ru]=10 to _2 [Ru] = 10 M lacked any c h a r a c t e r i s t i c bands i n the u.v. or v i s i b l e spectrum. 5.9. Aqueous so l u t i o n chemistry and carbonylation of CsRu(CO) 3C£ 3 Measurements of [H +] and^ [C£ ] i n aqueous solutions of CsRu(CO) 3C£ 3 were made under various conditions. The r e s u l t s of a study at 20°C are indicated i n Table XIX. The experiment was conducted by immersing the appropriate electrodes i n a constantly s t i r r e d , f r e s h l y prepared s o l u t i o n of CsRu(CO) 3C£ 3 i n water. The pH'and chl o r i d e ion p o t e n t i a l 183 Table XIX. V a r i a t i o n of pH and |C£ J i n an aqueous s o l u t i o n of CsRu(CO),C£, at 20°C i n a i r Time(sees) pH [H ] M Time(sees) Potential(mv) [CI ]M 0 5.77 9x10~ 7 0 -9.6 -275 2.53 2.1xl0" 3 176 -106.5 4.3x10" •3 525 2.49 2.3xl0~ 3 370 -120.0 7.2x10" •3 1035 2.37 3.0xl0" 3 700 -124.0 8.4x10" -3 3600 2.37 3.0xl0~ 3 1320 -125.7 9.0x10" •3 (* [H +] obtained from pH by means of a c a l i b r a t i o n graph. [Ru]=3.0xl0 3M) 184 readings were made at i n t e r v a l s of every few hundred seconds with the s t i r r i n g temporarily halted. The r e s u l t s show a rapid i n i t i a l l o s s of C£ and increasing a c i d i t y of the s o l u t i o n , a f t e r 1 hour [H +]:[Ru]=l.0 and [CiT]:[Ru]=3.0. Thus, i n the f i n a l s o l u t i o n a species with formula Ru (CO)^ ( I ^ O ^ (0H) + would appear to pre-dominate. A l t e r n a t i v e l y , the i n i t i a l species could be undergoing reductive carbonylation through the i n t e r a c t i o n of co-ordinated H^ O or OH and co-ordinated C0(cf. eq. 4.19). Such a process would explain the increased a c i d i t y of the s o l u t i o n but seems u n l i k e l y i n the presence of a i r and also s o l i d products would l i k e l y be formed. Simple los s of co-ordinated CO to give Ru 1 1(CO)^ species which have c h a r a c t e r i s t i c maxima i n the u.v. spectrum (see section 5.7) i s also considered u n l i k e l y since the u . v . / v i s i b l e spectrum of CsRu(CO)^C^ l n -3 ^ 0 shows no maxima at any time. An experiment at [Ru]=5.4xl0 M gave s i m i l a r changes i n the [H +]:[Ru] and [C£ ]:[Ru] values. An experiment _3 conducted at 80°C and [Ru]=2.75x10 M showed the same behaviour. Conductivity measurements on aqueous solutions of CsRu(C0).jC£3 i n a i r at 25°C always gave high values of A^ . (molar conductivity) when compared with Cs2Ru(CO)2C£ 4 (section 5.7) e.g. (a) [Ru] = 1.20xl0~ 3M I n i t i a l i s 644 (b) [Ru] = 1.30x10"3M I n i t i a l A., i s 614 M at t=300s i s 650 A„ at t=600s i s 656 M The r e s u l t s r e f l e c t the much greater ease with which Ru(CO) 3C£ 3 loses c h l o r i d e ion and undergoes basic hydrolysis when compared with the 185 Ru(CO) 2C£ 4 species. When CsRu(CO) 3C£ 3 was dissolved i n water (5 mis) at 80°C -2 under nitrogen i n the gas-uptake apparatus at [Ru]=1.7xl0 M, the i n i t i a l l y c o lourless s o l u t i o n became pale yellow a f t e r 2400 seconds, at which point the experiment was stopped. ( I n i t i a l gas evolution was always observed i n experiments employing H^ O alone. The s i z e of t h i s e f f e c t (cf. pyridine and p i p e r i d i n e i n section 3.5) was -4 equivalent to 10 moles of gas evolved by 5 mis. of EI^ O within 100 seconds at 60°C, thus i n t e r f e r i n g with attempts to detect CO loss upon d i s s o l u t i o n of CsRu(CO) 3C£ 3 i n water). The s o l u t i o n from t h i s experiment was l e f t to cool under N 2 overnight and within 12 hours had become colourless. Chloride ion and pH measurements on t h i s c o l o u r l e s s s o l u t i o n gave [H +]:[Ru]=l.0 and [CI ]:[Ru]=3.0. At 20°C quite d i f f e r e n t r e s u l t s were obtained under nitrogen. A _2 2.2x10 M sol u t i o n was i n i t i a l l y c o l o u r l e s s , but became cloudy a f t e r 40 minutes and within 2 hours had p r e c i p i t a t e d some s o l i d . The s o l i d was extracted with n-hexane and found to have peaks at 2089(m), 2079(m), 2071(w), 2034(s) and 2001(w) i n the i n f r a - r e d spectrum. T.L.C. of t h i s s o l u t i o n on s i l i c a gel using acetone as eluent gave one spot with R =0.77. r The observations discussed to t h i s point i n section 5.9 suggest that CsRu(C0) 3C£ 3, i n aqueous s o l u t i o n under nitrogen atmosphere at 20°C, undergoes a process of reductive carbonylation (cf. eq. 4.19) in v o l v i n g co-ordinated CO. This process i s not evident at 80°C under nitrogen nor i n the presence of a i r at either 20°C or 80°C. Thus, the reductive carbonylation of Ru(II) to lower valent species appears to be 186 favoured at lower temperatures (as supported by observations i n the synthesis of [HRu(CO) 3l n discussed i n section 3.1). I n i t i a l i n v e s t i g a t i o n s of the carbonylation of CsRu(CO) 3C£ 3 were c a r r i e d out at [Ru]=0.1M. and at room temperature. CO was passed through the aqueous solutions and cloudiness was noticed within 30 minutes, a f t e r 24 hours a brown p r e c i p i t a t e was c o l l e c t e d by f i l t r a t i o n from the remaining pale yellow s o l u t i o n . The s o l i d was found to have an i n f r a - r e d spectrum (nujol mull) with one strong peak at 2036 cm 1 and peaks of low i n t e n s i t y at 2075 and 2120 cm In addition an absorption occurred at 377 (m) cm 1 i n the Ru-CJ£ stretching region. The r e s u l t s are consistent with the formation of trans- Ru(CO)^Gl^ > since a more complex CO stretching pattern and more than one Ru-C£ stretching frequency would be expected for the c i s - i s o m e r . T h e r e s u l t s of the microanalyses from two separate experiments were s a t i s f a c t o r y (Calculated f o r RuC^O^CJ!^: C,16.87% Found: (a) C,16.56%. (b) C,16.26%). Yie l d s of t h i s s o l i d were low but when CO was once again passed through the f i l t r a t e more s o l i d was produced. When water was replaced by 0.3M.HC£(aq.) as the solvent medium no s o l i d could be produced by t h i s means within 36 hours, presumably as a r e s u l t of the i n h i b i t i o n of i o n i s a t i o n of CSl~ from Ru(CO) 3C£ 3~ i n 0.3M.HCJI. 172 Previous workers have reported the preparation of the c i s -isomer of Ru(CO) 4C& 2 by treatment of a CH 2Br 2 s o l u t i o n of [Ru(C0) 3C£ 2] 2 with CO at 80-100 atm. i n the temperature range 20 to 120°C. However, although i d e n t i f i e d by the i . r . spectrum i n so l u t i o n (v(C0) at 2182(w), 2132 (s),2113(m) and 2080(s) cm "*") the compound was not i s o l a t e d , since 187 i n the absence of carbon monoxide i t was found to r a p i d l y reconvert to [Ru(CO) 3C£ 2] 2. T n e highest conversion into Ru(CCO^C^ w a s found, by use of a high-pressure i . r . c e l l , to occur at room temperature 173 and high carbon monoxide pressure (eq. 5.19). [Ru(CO) 3C£ 2] 2 + 2 0 0 ^ = = ^ 2Ru(CO) 4C£ 2 (5.19) 174 Johnson and co-workers have claimed that the cis-form of Ru(CO) 4C£ 2 can be prepared by reacting c h l o r i n e with Ru„(C0) i n ether at -80°C, X.Z-however no compound was i s o l a t e d . The assignment of the product i n the present work to the trans-isomer of Ru(CO) 4C& 2 seems consistent with i . r . and a n a l y t i c a l data although somewhat s u r p r i s i n g since P e a r s o n ^ 5 has stated, "two soft ligands i n mutual trans p o s i t i o n s w i l l have a d e s t a b i l i z i n g e f f e c t on each other when attached to cla s s b metal atoms". Our product, unlike the c i s isomer, does not readily lose CO. CO uptake experiments using CsRu(C0) 3C£ 3 were c a r r i e d out using a v a r i e t y of conditions. A number of t y p i c a l gas uptake p l o t s are shown i n Figure 50 (solids are p r e c i p i t a t e d during the gas uptake). Reactions were conducted using ruthenium concentrations ranging from 0.01M. to 0.14M., at temperatures from 30°C to 80°C using CO pressures from 400 to 600 mm Hg. An induction period of 100 to 200 seconds was always observed before any gas uptake occurred. I n i t i a l gas uptake was nearly l i n e a r with time, showing a gradually decreasing rate and approaching 1:1 stoichiometry a f t e r several thousand seconds. In order to minimise e f f e c t s due to changes i n pH and [C£ ], i n i t i a l rates of gas uptake were compared. The r e s u l t s from experiments at 60°C and 1 atm t o t a l 189 pressure are shown i n Table XX. The data of experiments (2), (5) and (7) in d i c a t e a f i r s t order dependence on [Ru] but c l e a r l y no fi r m con-clusions can be drawn concerning the k i n e t i c s . Experiments conducted i n the presence of soda lime (e.g. experiments (1) and (6)) showed a 2:1 f i n a l stoichiometry. I t i s noteworthy that a so l u t i o n p r e - e q u i l i b r a t e d i n a i r at 25°C for 48 hours (experiment (3)), presumably containing Ru(CO) 3 (t^O^OH"1", showed no s i g n i f i c a n t l y d i f f e r e n t behaviour. Results from other gas-uptake experiments performed using aqueous solutions of CsRu(CO) 3C£ 3 under CO are shown i n Table XXI. The rates over a range of temperatures showed l i t t l e f l u c t u a t i o n . Experiment 12 conducted i n the presence of [H +]:[Ru] equal to 1.0 (using HC£0 4),was designed to mimic the conditions that might e x i s t during the preparation of [HRu(C0) 3] n. That i s , i f III II reduction from Ru to Ru (CO)^ occurs i n a stepwise fashion, then a 1:1 r a t i o of [H +] to [Ru] i s expected ( i . e . i n i t i a l formation of III I I I Ru (CO) from Ru followed by the reactions depicted i n equations 5.1, 5.2 and 5.3 r e s p e c t i v e l y (section 5.4)). In f a c t , the addition of acid seems to have very l i t t l e e f f e c t upon the reaction rate, at l e a s t at the concentrations employed. Experiment (17) i n the presence of buffer at pH=7.0 showed a s l i g h t l y enhanced i n i t i a l rate, but the stoichiometry a f t e r 2000 seconds was 0.67:1 and the reaction showed no gas uptake beyond t h i s point. The reaction conducted at [Ru]=0.l4M (experiment(23))showed a very rapid i n i t i a l rate but slowed down appreciably a f t e r the f i r s t 2000 seconds and the l a t e r stages of the reaction appeared to be taking place very slowly (Fig. 51). The 190 Table XX. E f f e c t of varying [Ru] on the i n i t i a l rate of carbonylation of CsRu(CO) 3C£ 3. [Ru]xl0 3M I n i t i a l rate x loS t . s " 1 1 8.26 11.10 2 8.74 9.30 3+ 9.67 8.27 4 10.00 15.00 5 12.00 13.20 6* 13.74 12.40 7 16.08 18.40 8 23.20 16.10 9 58.00 21.70 ( A l l experiments at 60°C under t o t a l pressure of 1 atm. - soda lime tube employed. 1 - s o l u t i o n p r e - e q u i l i b r a t e d at 25°C for 48 hours i n a i r ) 191 Table XXI. Further data from the carbonylation of CsRu(CO) 3C£ 3 i n aqueous s o l u t i o n [Ru]xl0 3M. T°C ^ t o t a l ) ^ P ( c o ; ™ i H g I n i t i a l rate x (10) 17.12 60 660 510 12.5 (11) 16.60 60 675 525 . 11.0 ** (12) 10.00 70 760 530 7.7 (13) 11.76 . 70 660 430 7.6 (14) 14.26 70 670 440 11.7 (15) 16.28 70 760 530 11.3 (16) (17,++ 21.60 70 670 440 14.1 9.83 70 760 530 9.2 (18) 10.00 50 760 668 5.71 (19) 10.00 50 760 668 6.26 (20) 17.60 50 655 560 6.40 (21) 15.9 40 668 610 4.70 (22) 21.1 30 614 580 -(23) 140.0 80 760 405 70.9 (** experiment conducted i n the presence of 1x10 M. RCZO^ i . e . i n i t i a l [H +]:[Ru]=1.0. =f4 experiment conducted i n pH=7.0 b u f f e r ) . F i g . 51. CO uptake by an aqueous solution of CsRu(CO) 3C£ 3(28.6x10 moles, [Ru]=14.3x10 M.) at 80°C. and 1 atm. t o t a l pressure. 193 stoichiometry of CO uptake per ruthenium had reached only 0.33 within 4000 seconds, compared to a value of 0.8 within 2000 seconds in experiment (2) at 60°C. Although data at 1 atm.total pressure were readily recorded, reduction in pressure caused fluctuations that made kinetic inter-pretations impossible. The rate was clearly dependent on the partial pressure of CO but the exact dependence could not be established (see Fig. 52). Variations in the chloride ion concentration at various stages in the carbonylation of CsRu(CO) 3C£ 3 were determined by f i l t e r i n g the resulting mixture and immersing the chloride ion electrode in the f i l t r a t e . At the 1:1 gas uptake point a l l chloride ion was found to be.in solution in experiments at both 60 and 70°C. In an _2 experiment at 60°C, [Ru]=lxl0 M and 1 atm. total pressure gas uptake began after about 200 seconds and the reaction mixture at this stage showed [C£ ]:[Ru]=2.0. The pH was found to decrease in these experi-ments giving a value of [H+] : [Ru] equal to 2.0 at the 1:1 gas uptake stage and showing no change subsequently. Extraction of the reaction products at various stages with n-hexane gave results similar to those reported for Cs2Ru(C0)2C£4 (section 5.7). However, several notable differences were evident. Experiment (17) (Table XXI) yielded an orange solid strongly resembling Ru^CO)^ and having i . r . bands in n-hexane at 2063(s),2031(m) and 2009(w) cm 1 that are characteristic of this complex (see Table VII in section 3.4). One possible explanation is that the excess of buffer serves to greatly reduce the yield of hydridocarbonyl species by limiting the availability F i g . 52. CO uptake by aqueous solutions of .CsRu(CO) 3C£ 3 (6.5x10 moles, [Ru]=l.30x10 M.) at 60°C. and (•) 1 atm. t o t a l pressure; ( A ) and (• ) at h atm. t o t a l pressure. 195 of H + i n solution and as a r e s u l t only Rug(CO) 2 1 S detected. Figure 53 shows t y p i c a l i . r . spectra of n-hexane extracts from various experiments. (A) i s the product at the 0.3:1 stage i n experiment (23), characterised by peaks at 2114(w),2087(s),2077(s), 2071(w),2063(w),2035(s) and 2013(m) cm"1 i n the carbonyl stretching region; (B) i s the product at the 0.16:1 stage i n an analogous -2 experiment at [Ru]=lxl0 M, at a t o t a l pressure of 1 atmosphere and at 80°C; the i . r . spectrum reveals peaks at 2111(w),2078(s), 2071(s),2059(m),2031(s) and 2013 cm"1. (C) i s the product at the -2 0.6:1 stage i n an experiment at [Ru]=lxl0 M, at a t o t a l pressure of 1 atmosphere and at 70°C; the i . r . spectrum reveals peaks at 2111(vw), 2077(s),2071(s),2059(w),2033(s) and 2012w(cm~1). The observations on 2-Ru(CO)^Ci^ solutions were also d i f f e r e n t from those on Ru(C0) 2C£ 4 solutions i n that much larger amounts of s o l i d products were observed i n the f i r s t few hundred seconds of the reaction, p r i o r to gas uptake. _2 For instance, at 70°C, [Ru]=lxl0 M and t o t a l pressure of 1 atmosphere, the r e s u l t of a hexane extract before CO uptake gave spectrum (D), which appears i d e n t i c a l to spectrum (C). When the temperature was _2 raised to 80°C and ruthenium concentration raised to 6x10 M, the i n i t i a l products before CO uptake yielded spectrum (E) i n which a peak at 2089 cm 1 i s evident, but the peak at 2071 cm 1 appears to be quite weak. In summary, the peak at 2089 cm 1 could only be -2 detected at high concentrations of ruthenium (5x10 M and higher) and was only of low i n t e n s i t y i n i t i a l l y . In ad d i t i o n the i . r . spectra of the hexane solutions gave no evidence of any co-ordinated water or hydroxide ions. 196 c CM F i g . 53.Infra-red spectra i n n-hexane. (A),(B) and (C) are n-hexane extracts from CsRii (CO) 365,3 carbony-l a t i o n at.0.3:1, 0.16:1 CO uptake stages re s p e c t i v e l y , (D) n-hexane extract before CO uptake at 70°C and [Ru]=1.0xlO"2M., (E) n-hexane extract before CO uptake at 80°C and [Ru]= Numbers r e f e r 6.0xlO~2M to wavenumbers i n cm ,-1 E 197 The i . r . spectra of species from experiments using Ru(CO) 3C£ 3 2-and RuCCO^CJ!^ species are remarkably s i m i l a r and argue for a s i m i l a r reaction path. The generation of the i n i t i a l products from the t r i c a r b o n y l species by reductive carbonylation v i a co-ordinated CO seems necessary to explain the apparent generation of s o l i d s before CO uptake commences. The r e s u l t s of the experiment at 20°C under nitrogen c l e a r l y indicated that reductive carbonylation could take place i n the absence of a carbon monoxide atmosphere, but a temperature increase to 80°C prevented t h i s process occurring. This seems consistent with observations i n the [HRu(CO) 3J n preparative experiments where reduction to the desired product was favoured by repeated heating and cooling cycles of the reaction mixture (see section 3.1). The present state of knowledge concerning the orange product obtained by reductive carbonylation of CsRu(CO) 3C£ 3 i s far from complete. A comparison of the observed i . r . spectra with those for the reported hydridocarbonyl complexes of ruthenium (Table VII i n section 3.4) reveals the d i f f i c u l t i e s involved. A l l known hydrido-carbonyl complexes of ruthenium have strong peaks i n the 2056-2063 cm 1 region of the i . r . spectrum (Table VII) which would suggest that the products obtained from carbonylation of CsRu(CO) 3C£ 3 contain new ruthenium carbonyl complexes. The observations made during CO uptake studies on aqueous solutions of CsRu(CO) 3C£ 3 ( i . e . 1:1 uptake i n the absence of soda lime and 2:1 uptake i n the presence of soda lime) are not consistent with a process such as 198 2Ru I ] :(CO) 3 + 3C0 + 3H20 2HRu(CO)3 + 3C02 + 4H + (5.20) In particular, experiments conducted in the absence of soda lime should result in no net gas uptake i f reaction 5.20 was occurring. A l i k e l y explanation seems to involve the formation of a Ru°(C0) 4 species, with subsequent further reductive carbonylation to give a mixture of pro-ducts (eq. 5.21). Ru I I(C0) 3 + 2C0 + H20 > Ru°(C0) 4 + C0 2 + 2H+ (5.21) Reaction 5.21 explains the observed stoichiometries and also the observed [H+]:[Ru] value at the point of 1:1 gas uptake. In addition, a larger scale carbonylation experiment using 0.30 gms of CsRu(C0)3C&3 in aqueous solution at 70°C, resulted in only 0.03 gms of [HRu(C0) 3] n (25% yield based on the monomeric formula) after 3 hours. Since 1:1 uptake is complete within this time at 70°C (from gas uptake data) this lends some support to the idea that formation of Ru(0) species i s important in the carbonylation of CsRu(CO) 3C£ 3. Despite similar infra-red spectra recorded for n-hexane extracts from reaction mixtures after carbonylation of CsRu(CO) 2C£ 4 (Fig. 49) and CsRu(C0) 3C£ 3 (Fig. 53) the different appearance of the gas uptake plots (Figs. 42 and 50 respectively) argues against immediate loss of CO from Ru 1 1(C0) 3 to give a Ru 1 1(C0) 2 species which would then undergo carbonylation. The much faster aquation and hydrolysis of Ru(CO) 3C£ 3 2-(Table XIX) when compared with Ru(CO) 2C£ 4 (section 5.7) would f a c i l i t a t e the process of reductive carbonylation (eq. 4.19), which requires co-ordinated water or hydroxide ion, and might explain the more rapid i n i t i a l 199 reductive carbonylation (in the f i r s t few hundred seconds) i n the case of the t r i c a r b o n y l species. 200 Chapter 6 CARBONYLATION OF AMINES USING [HRu(CO).J J n 6.1. K i n e t i c s and mechanism of p i p e r i d i n e carbonylation i n the presence of [HRu(C0) 3] n. A part of Chapter 3 discussed the chemistry of [HRuCCO)^]^ i n p i p e r i d i n e , and the stoichiometric carbonylation of p i p e r i d i n e i n the absence of carbon monoxide with production of a ruthenium dicarbonyl complex. In the presence of carbon monoxide, conversion of p i p e r i d i n e to N-formyl p i p e r i d i n e takes place continuously and gives CO uptake plot s (Fig. 54) that are l i n e a r i n i t i a l l y followed by a gradually decreasing rate over a period of several hours. This decreasing rate of carbonylation i s a t t r i b u t e d i n part to poisoning of the c a t a l y s t by N-formyl p i p e r i d i n e which i s the only product of the reaction, since addition of t h i s (or other amides such as N,N-dimethylacetamide) slowed the gas uptake. The k i n e t i c s of the carbonylation system were i n v e s t i -gated using the i n i t i a l rates. The [HRu(C0).j]n polymer was soluble i n piperidine-toluene mixtures and showed a f i r s t order dependence of maximum rate on [piperidine] (Fig. 55) at both high and low CO pressures. o F i g . 55. [HRu(CO) 3] n - catalysed carbonylation of pipe r i d i n e . Amine dependence at 7 5 ° C , 2.2xlO"2M. Ru i n toluene-piperid'ine solutions. ( • ) , 490 mm CO; ( • ), 100 mm CO. 203 The CO dependence was f i r s t order at low pressures, but i t approached zero-order at partial pressures above 0.5 atmospheres (Fig. 56). The -2 Ru dependence was f i r s t order at least up to 3x10 M. in conditions of both zero-and first-order CO dependence (Fig. 57). The carbonylation rates were also studied from 50-80°C. at 1 atmosphere total pressure where the reaction is independent of CO pressure (Table XXII). The rate law for the catalysis of piperidine carbonylation by [HR u(C0) 3] n can be written as - d[C0] = k'[Ru T][pip], where k' is a pseudo-second order rate constant which includes the CO dependence. A mechanism which satisfies the observed kinetics considers the active species to be monomeric in solution and can be written as, k l Scheme A HRu(CO)2 (pip) x + pip _ , _ . k _ ^ HRu(C0)2 ( p i p ) ^ (6.1) k2 HRu(C0) 2(pip) x + 1 + CO >HRu(C0) 3(pip) x + 1 (6.2) •f o o f HRu(C0) 3(pip) x + 1 — > HRu(C0) 2(pip) x + pipCO (6.3) Assuming a steady state treatment for [HRu(C0) 2(pip) x +^] gives a rate law the form k'[Ru T][pip], where k' = k^k^CO] (6.4) k_^+k2[CO] Rearranging gives: 0.02 . 0.04 F i g . 57. [HRu(C0)3] n - catalysed carbonylation of pipe r i d i n e . Ru dependenc 75°C. i n neat amine. ( A ) , 490 mm CO; (A) , ' 100 mm CO. 206 Table XXII. Temperature dependence for [HRuCCO)^]^ catalysed carbonylation of p i p e r i d i n e Temperature, °C Max.rate x lO^M.s k/ x lO^M.^s 50 1.17 0.05 60 3.52 0.16 65 4.16 0.19 71 7.02 0.32 75 10.40 0.48 80 12.10 0.55 _2 ([Ru]=2.2xl0 M., 1 atm. t o t a l pressure, neat p i p e r i d i n e (10.IM), = pseudo - second order rate constant) 207 Figure 58 shows a p l o t of (k') 1 vs. [CO] 1 for the two sets of data i n Figure 56. Good l i n e a r p l o t s r e s u l t , and the intercept and slope,for the data at 75°C. give ^=5.3X10~ 4MT 1S~ 1 and k g / k ^ = 500 M - 1. An a l t e r n a t i v e mechanism which accounts equally w e l l for the k i n e t i c s involves a pre-equilibrium with CO: Scheme B HRu(CO)„(pip) + CO K HRu(CO) (pip) (6.6) HRu(C0) 3(pip) x + pip > H R u ( C 0 ) 3 ( p i p ) x + 1 (6.7) H R u ( C 0 ) 3 ( p i p ) x + 1 — > HRu(C0) 2(pip) x + pipCO (6.8) This mechanism gives the rate law of eq. 6.9. by a non-steady-state de r i v a t i o n . - d [ C 0 ] _ kK[Ru T][pip][CO] (6.9) dt (1 + K[C0] ) i . e . k' kK[C0] (1 + K[C0]) (6.10) -4 -1 -1 The same analysis as discussed above gives the value of 5.3x10 M s to k and 500M 1 to K at 75°C. The K value, however, implies that r e a c t i o n 6.6 should be observable, but no evidence for the pre-equilibrium could be found. In p a r t i c u l a r , exposing solutions of the polymer [HRu(C0) 3] n i n p i p e r i d i n e to increasing pressures of CO should favour formation of the t r i c a r b o n y l species. However, on subjecting a 3x10 M s o l u t i o n of the polymer to p a r t i a l pressures of 125 and 430 mm Hg of CO at 80°C. no changes i n the carbonyl region of the l i q u i d i n f r a - r e d spectrum could be 209 detected. Further, no i n i t i a l rapid CO uptake by the polymer solutions was detected, and thus conversion of dicarbonyl to t r i c a r b o n y l was not observed. The former mechanism outlined i n equations 6.1-6.3 i s then favoured. An Arrhenius p l o t (Fig. 59) of the temperature v a r i a t i o n data (Table XXII), obtained under conditions of zero-order dependence on CO, gives a good s t r a i g h t l i n e and y i e l d s the a c t i v a t i o n parameters but the a c t i v a t i o n energy appears to be somewhat high, possibly a r e s u l t of s t e r i c problems caused by the incoming p i p e r i d i n e ligand. possibly by means of a metal-assisted hydride s h i f t (e.g. isomerisation 176 of o l e f i n s catalysed by ir o n carbonyls i s thought to involve metal-assisted hydrogen migration v i a a hydride addition-elimination mechanism.) Of i n t e r e s t , N-ethyl p i p e r i d i n e could not be carbonylated under s i m i l a r conditions to those employed with p i p e r i d i n e and t h i s shows the importance of an ac t i v e hydrogen. A further possible hydride s h i f t r eaction involves p i p e r i d i n e i n the r o l e of a proton acceptor, for example (writing R 0 for The formation of an Ru-CO-N moiety at some stage seems e s s e n t i a l , HRu(C0) 2(R 2NH) + R2NH HRu(C0) 2(R 2N) + R_NH2 + (6.11) (cf. eq. 6.1) HRu(C0) 2R 2N + CO ^ HRu(CO) 3(R 2N) (6.12) HRU(CO)3R2N' fast •> HRu(CO)2-CONR2 (6.13) i I I i I l I I I I 1 1 I i I i i I : 2.85 2.90 2.95 3.00 (TT1x id3, V 1 F i g . 59. [HRu(C0) 3] catalysed carbonylation of p i p e r i d i n e . L og 1 0(k') vs. 1/T from data i n Table (k'=pseudo - second order rate constant). 211 HRu(CO)2CONR2 + R 2NH 2 + f a S t > HRu(CO) 2 + R NCHO + R2NH (6.14) There i s ample precedent for the formation of carbamoyl species by reactions of amines with metal carbonyls, e s p e c i a l l y c a t i o n i c ones and t h i s topic has been discussed i n section 1.5. The mechanism i s of the type outlined by equations 6.1-6.3 (expected rate law shown i n equation 6.4), but i f the species R 2NH 2 + e x i s t s i n an equilibrium such as that shown i n equation 6.11, then the addit i o n of a piperidinium s a l t to the reaction mixture should cause a marked decrease i n the reaction rate (a d i r e c t inverse dependence i s predicted at low CO pressures) Table XXIII summarises the r e s u l t s of experiments conducted i n the presence of p i p e r i -dine hydrochloride. Experiments were also conducted i n the presence of li t h i u m c h l o r i d e and l i t h i u m n i t r a t e i n order to observe any i o n i c e f f e c t s that might be inf l u e n c i n g the reaction. The r e s u l t s shown i n Table XXIII indi c a t e that [DRu(C0) o] and 3 n [HRu(CO) 3] n are equally e f f e c t i v e i n carrying out the carbonylation reaction. Experiments (1) and (4) show that an increase i n [pipHC£] alone r e s u l t s i n a s l i g h t rate increase. A rate enhancement was observed upon adding l i t h i u m c h l o r i d e (experiments (5) and (6)) and the e f f e c t of increasing [pipH +] under conditions of constant t o t a l [C£ ] was a decrease i n rate (experiments (8) and (9)). These r e s u l t s lend some support to the rol e of p i p H + i n a reaction mechanism as depicted i n equations 6.11-6.14. Experiments (10) and (11) which were c a r r i e d out i n the presence of a trace of water (0.02 mis), added by syringe to the pi p e r i d i n e solvent (5 mis), gave greater maximum rates (by a fac t o r of 2) than experiments 212 Table XXIII. Carbonylation of p i p e r i d i n e using [HRuCCO)^]^ or [DRu(CO) ] i n the presence of various added reagents [pipHCi> ]xl0 2M. [LiC£]xl0 2M. [LiNO 3]xl0 2M. Max.rate(M. s. 1)xlO (1) * 1.37 - - 1.06 (2) * - - - 1.38 (3) - - - 1.42 (4) 1.73 - - 2.25 (5) - 3.44 - 4.65 (6) - 7.60 - 5.73 (7) - - 3.72 3.19 (8) 1.21 2.12 - 3.61 (9) 2.41 1.03 - 1.94 (10) ^  - - - 3.22 (11) 1" - - 2.89 A l l experiments conducted at [Ru]=1.0xl0 M., T=75°C. and p a r t i a l pressure of C0=100 mm Hg. i n neat p i p e r i d i n e . (* indicates [HRuCCO)^]^, a l l other experiments employed [DRu(C0) 3__ =f Experiments conducted i n the presence of a 0.22M. concentration of added H,.0 213 (2) and (3) which were conducted i n dry p i p e r i d i n e . This e f f e c t could possibly explain the rate enhancement produced by inorganic s a l t s such as LiC£ and LiNO^ which r e a d i l y pick up water from the atmosphere. At higher concentrations of added water (e.g. 1 ml. H^ O + 5 mis. p i p e r i d i n e = 9.16 M. i n K^O) the uptake of CO was slower than i n the experiments discussed above and gave scattered p l o t s of CO uptake against time. The measured AH^ value, f or the carbonylation of p i p e r i d i n e using [HRu(CO) 3] n as c a t a l y s t , i s compatible with a rate determining reaction such.as that shown i n equation 6.11 and such i o n i s a t i o n 177 reactions do have large negative entropies of a c t i v a t i o n i n the same range as our measured value of -22.8 e.u. existence of a dimeric Ru(I) species, which was i s o l a t e d i n the s o l i d state by addition of water to p i p e r i d i n e solutions of [HRu(CO),j]n. Mechanisms invo l v i n g dimeric hydridocarbonyl species i n s o l u t i o n can be written, f o r example: Evidence presented i n Chapter 3 (section 3.6) suggests the Scheme C H 2 R u 2 ( C O ) 4 ( p i p ) 3 + pip (I) HRu(CO) 2(pip) 2 + CO — 2HRu(C0) 2(pip) 2 (6.15) ( I D > HRu(CO) 3(pip) 2 (6.16) ( I D HRu(CO) 3(pip) 2 fast > pipCO + HRu(CO) 2(pip) (6.17) HRu(CO) 2(pip) + pip i fast -> HRu(CO) 2(pip) 2 (6.18) ( I I ) 214 A steady-state treatment of species II gives: k 1 [ I ] [ p i p ] = k_±[II]2 + k 2[II][CO] (6.19) -d[CO] = k 2[II][CO] = k^k 2[I][pip][CO] (6.20) d t k_ 1[II]+k 2[C0] Such a rate law accommodates a l l the observed dependences except the f i r s t order [Ru] dependence measured at low CO pressures, i . e . when k_^[II]>k 2[CO], and the rate becomes h a l f order i n [Ru]. Another p o s s i b i l i t y i s an i n i t i a l l i g a n d d i s s o c i a t i o n mechanism as depicted i n Scheme (D). This mechanism would also explain the i s o l a t i o n of a complex that apparently contains three p i p e r i d i n e groups for every two ruthenium atoms (see section 3.6). Crooks and co-workers^ 7 have suggested such a d i s s o c i a t i o n mechanism to explain low molecular weight measurements i n s o l u t i o n for complexes of the type Ru 2 (CO)^ (RC0 2) 2T_ 2, although no d e t a i l s were reported. e.g. Ru 2(CO) 4(OAc) 2(MeCN) 2-^=^Ru 2(CO) 4(OAc) 2(MeCN) + MeCN (6.21) Scheme D pxp pip pip pip CQ^ | ^ H ^ | ^ C O OC^ | ^ H . | CO Ru Ru k ^ Ru — H Ru — CO + pip CO^ | ^ F K I ^ C O ^ 1 OC^ ^ p i p ^ pip pip k_ (I) 1 CO Products (ID k 2 (6.22) Once again, applying the steady-state treatment to species II i n equation 215 6.22 gives k x [ i ] = k ^ E p i p H i i ] + k 2 [ c o ] [ i i ] (6.23) r and -d[CO] = k ^ i c o i t i ] (6.24) dt k-JpipJ+k^CO] Equation 6.24 shows an unsatisfactory inverse dependence on [pip] at low CO pressures and zero-order dependence on [pip] at high CO pressures. Hence, scheme D i s also an u n l i k e l y mechanism for the carbonylation of amide formation and also increased t o t a l y i e l d of product when con-ducting the rhodium catalysed carbonylation of n-butylamine i n the presence of increasing amounts of various phosphines (see section 1.5 and Table XXIV). At 1:1 P/Rh the percentage conversion i s lower than i n the presence of the carbonyl c a t a l y s t alone; however an increase i n the P/Rh r a t i o r e s u l t s i n enhanced s e l e c t i v i t y and percentage conversion. 8 6 Presumably, the rapid evolution of gas observed by these workers at the beginning of each experiment represents displacement of CO by a phosphine to give a more s t e r i c a l l y crowded a c t i v e species. The formation of a urea as product could require simultaneous co-ordination of two amines to each rhodium while formation of amide would require only one amine per rhodium and hence the l a t t e r process could be favoured p i p e r i d i n e catalysed by [HRu(CO),] 3 n 6.2. The addition of phosphines to the amine systems 8 6 Durand and Lassau have observed increased s e l e c t i v i t y towards 216 Table XXIV. E f f e c t of added phosphine on n-butylamine carbonylation S e l e c t i v i t y Catalyst P/Rh Conversion % Amide urea (A)Rh 2C£ 2(CO) 4 - 92 35 65 (A)+P(CH 3) 3 1 74.5 62 38 2 94.5 94 6 4 98.5 98 2 6 100 100 (A)+P(C 6H 5) 3 2 46.5 70 30 (* From r e f . 86 - reactions conducted i n benzene, t y p i c a l l y [Rh]=0.16M., P=60 atm. under CO, T=160°C f or 4 hours i n the presence of an appropriate amount of phosphine) 217 i n a more crowded molecular environment. Although, the c a t a l y s t [HRu(C0) 3_ n appeared to be 100% s e l e c t i v e towards conversion of p i p e r i d i n e to the N-formyl product, as judged by v.p.c. analysis of reaction products, the e f f e c t of added phosphine on t h i s system was studied. Experiments employing a PPh^/Ru r a t i o (based on [HRu(C0) 3l n monomeric units) of about 1:1 were characterised by an i n i t i a l period of no reaction followed by a slower rate of carbonylation (Table XXV) than was observed i n previous experiments (Table XXIII). The addition of PPh^, p a r t i c u l a r l y at 80°C., decreased the i n i t i a l rate quite remarkably. The s l i g h t increase i n rate with time indicated by the three measurements i n experiment (A) was observed i n a duplicate run which gave the same data and was probably due to a slower rate of d i s s o l u t i o n of the polymer i n p i p e r i d i n e when PPh^ was present. The slower rates l i k e l y r e s u l t from competition of the phosphine for c a t a l y s t co-ordination s i t e s . In an experiment conducted under s i m i l a r conditions to those employed i n (A) the reaction was stopped a f t e r 13,000 seconds ( 0.27 mole % N-formyl product at t h i s point as calculated from gas uptake data) to give a l i g h t yellow s o l u t i o n . Upon cooling the system to room temperature under CO a yellow p r e c i p i t a t e resulted which was subsequently separated from the remaining yellow s o l u t i o n by f i l t r a t i o n under nitrogen and then washed with water. Af t e r pumping on t h i s s o l i d under vacuum for several hours an i n f r a - r e d spectrum (nujol mull) of the s o l i d was found to show one sharp peak i n the CO region at 1910 cm 1 . The spectrum c l o s e l y resembled that of Ru(C0) 3(PPh 3)2 prepared by the 178 method of Levison and Robinson (see also section 8.1) and the micro-218 Table XXV. E f f e c t of added PPh 3 on the [HRu(CO) 3] n catalysed carbonylation of p i p e r i d i n e 7 2 5 - 1 [Ru]xlO M. [PPh 3]xlO M. T°C. Pco(mm.Hg) Maximum ratexlO M.s. at time(t) (A) 2.08 1.84 70 534 t=5000s. , Rate=0.48 t=13000s., Rate=0.72 t=44000s., Rate=0.72 (B) 1.97 - 70 534 t=200s. , Rate=3.60 (C) 1.70 1.55 80 432 t=1000s., Rate=2.08 t=30 hrs., Rate=1.45 t=40 hrs., Rate=1.28 (D) 2.20 - 80 432 t=500s., Rate=12.1 219 analysis was reasonable for such a complex (Calculated for RuC^gH^QO.^: C,66.00;H,4.25%. Found: C,66.33;H,5.20;N,1.57%). The small percentage of nitrogen argued f o r the presence of co-ordinated p i p e r i d i n e i n a complex such as [Ru(CO) 3 (PPh 3) 2 ( C ^ Q N ) ] (Calculated f or R U C ^ H ^ Q N O ^ : C,68.48;H,4.76;N,1.66%) but, the carbon analysis here i s too high and a more complex pattern of CO stretching frequencies would be expected i n the i n f r a - r e d spectrum. The percentage of nitrogen i s probably due to remaining free p i p e r i d i n e , which, although notdetected i n the i . r . spectrum, could r e s u l t from inadequate r i n s i n g of the s o l i d with water before drying under vacuum. The y i e l d of t h i s s o l i d from 0.18 gms (9.60 x 10~ 5 monomer moles) of [HRu(C0) 3J n was 0.22 gms. or 3.1 x 10~ 5 moles based on Ru(C0) 3(PPh 3) 2. Since the i n i t i a l PPh 3 to Ru r a t i o was 1:1 the expected y i e l d of Ru(CO) 3(PPh 3) 2 would be 4.8 x 10 _ 5moles. The l i q u i d i . r . spectrum of the f i l t r a t e could not be distinguished from that of previously prepared solutions of [HRu(C0) 3] n i n p i p e r i d i n e alone (Fig. 60) suggesting that the same a c t i v e species i s involved i n each case. The reduced rate i n the presence of PPh 3 i s probably due to formation of Ru(CO) 3(PPh 3) 2 and hence a reduced concentration of the active species [ HRu(C0) 2(pip) x] n- A reduction i n rate by a factor of 5 (see Table XXV) suggests that 20% of the t o t a l ruthenium concentration i s s t i l l a c t i v e . A sample of Ru(CO) 3(PPh 3) 2 prepared 178 by the method of Levison and Robinson (section 8.1) f a i l e d to d i s s o l v e i n p i p e r i d i n e at 50°C. under CO within f i v e hours (8.98 x 10 5 moles Ru(CO) 3(PPh 3) 2 added to 5 mis. piperidineE[Ru]=l.8x10 M.). C l e a r l y , the r e s u l t s obtained at 70 and 80°C. show that the slow carbonylation 220 11 2003 2029 F i g . 60. Solution i n f r a - r e d spectrum i n p i p e r i d i n e of the f i l t r a t e from the [HRu(C0) 3] n catalysed carbonylation of pi p e r i d i n e i n the presence of triphenylphosphine. C a l i b r a t i o n was conducted using the free CO v i b r a t i o n at 2147 cm - 1. A grating change occurs at 2003 cm - 1. 221 rates at these temperatures did not j u s t i f y a further i n v e s t i g a t i o n of Ru(CO) 3(PPh 3)2 as a c a t a l y s t under these conditions. The s o l u t i o n from experiment (A) a f t e r 66 hours was cooled under CO to give the same p r e c i p i t a t e of Ru(C0) 3(PPh^^• The mixture was then f i l t e r e d under nitrogen and a sample of the f i l t r a t e was stored i n a sealed tube under vacuum p r i o r to v.p.c. analysis using a Pennwalt 223 column (section 2.9). Only 9.02 mole % of N-formyl p i p e r i d i n e was detected i n the sample, and there was no evidence for formation of the urea, l , l ' - c a r b o n y l d i p i p e r i d i n e . This represents approximately one-third of the N-formyl product y i e l d obtained a f t e r 55 hours i n the presence of [HRu(C0) 3_ n alone (Table XXVI). 6.3. Carbonylation of other amines using [HRu(C0) 3] n The [HRu(C0) o] polymer has been used p r e v i o u s l y 1 0 0 to convert j n a number of other amines to t h e i r N-formyl products; t h i s work i s summarised i n Table XXVI. Primary amines react with CO only very slowly and pyridine does not react at a l l (see section 3.5). The low s o l u b i l i t y of [HRu(C0) 3] n i s c l e a r l y a factor i n determining some 84 of the r e a c t i o n rates but, as Brackman has discovered with copper-catalysed systems (section 1.5), the b a s i c i t y and stereochemistry of the amine w i l l play an important r o l e i n determining the reaction rate. The polymer was sparingly soluble i n the primary and s t r a i g h t chain secondary amines, r e s u l t i n g i n very poor conversion to N-formyl product. The carbonylation rates of the secondary amines seemed to roughly p a r a l l e l 179 t h e i r pK a values at 25°C. as depicted i n Table XXVI. 222 Table XXVI. Carbonylation of various amines using [HRuCCO)^]^ Amine % Conversion to N-formylamine Maximum ratexlO^M.s. 1 Primary Amines a n i l i n e cyclohexylamine 1(10 hrs.) 1.57 n-octylamine 0.52 Secondary Amines diethylamine 0.30 di-n-butylamine 0.24 dibenzylamine 0.04 hexamethyleneimine (pKa=11.07) 33(45 hrs.) 5.80 morpholine ( " =8.39) 14(48 hrs.) 1.30 pip e r i d i n e ( " =11.12) 32(55 hrs.) 10.4 p y r r o l i d i n e ( " =11.27) 34(48 hrs.) 11.5 T e r t i a r y Amines pyridine (* from r e f . 100 - a l l experiments at [Ru]=2.2xl0 M., T=75°C., 1 atms. t o t a l pressure and 5 ml. amine) 223 It was anticipated that the use of smaller r i n g compounds containing the secondary amine group might r e s u l t i n enhanced carbonylation rates owing to the generally greater r e a c t i v i t y of such compounds compared to the larger heterocycles. Ethyleneimine (C2H1-N) was dried over calcium hydride and -2 then stored under vacuum p r i o r to use. A 2 x 10 M. s o l u t i o n of [HRuCCO)^]^ (2.0x10 5 moles) was obtained by shaking the polymer i n 1 ml. of the neat amine at 30°C. under CO with a t o t a l pressure of 1 atmosphere. The low temperature was necessary because of the low b o i l i n g point (56°C at 760 mm Hg. pressure) of ethyleneimine. At the end of 20 hours the polymer had dissolved to give an orange-red s o l u t i o n , but instead of CO uptake a small amount ( 7.5 x 10 5 moles) of gas evolution was observed. The evolution could be due to CO from a polymer + amine reaction, or possibly to some ethyleneimine since t h i s amine w i l l r e a d i l y decompose by r i n g opening i n the presence of various added reagents, p a r t i c u l a r l y i n the presence of traces of a c i d . 1 8 ^ The low pK of ethyleneimine (8.01) may be a f a c t o r i n the non-carbonylation (cf. Table XXVI). The four-membered r i n g i n the s e r i e s , azetidine, has a reported 179 pK of 11.29 at 25°C. and t h i s seemed a promising candidate for carbonylation. Azetidine (Eastman Kodak Ltd.) was p u r i f i e d by bulb-to-bulb d i s t i l l a t i o n with intermittent pumping under vacuum and the p u r i t y checked by ^H n.m.r. (Fig. 61) which gave a t r i p l e t at 6=3.40, a quintet at 6=2.20 and a s i n g l e t at 6=2.25 i n C^D^ s o l u t i o n . The t r i p l e t may be assigned to protons on carbons (1) and (3), adjacent 224 L_™i-~JL-_J-~~L JLJL J__! L F i g . 61. H n.m.r. spectrum of azetidine i n C._D, s o l u t i o n . o 6 225 to the amine group and the quintet assigned to protons on carbon atom (2). The sharp s i n g l e t at 6=2.25 may be at t r i b u t e d to the proton attached to nitrogen. No water was detected i n the n.m.r. spectrum. Analysis by v.p.c. on a carbowax column at an oven temperature of 160°C. and helium c a r r i e r gas pressure of 35 p . s . i . gave a sharp peak with a retention time of approximately 1 minute and no water or organic impurities could be detected. Azetidine i s 181 made by a rather lengthy synthetic procedure and as a r e s u l t only very small quantities were ever a v a i l a b l e . Thus, carbonylation of the material was conducted i n toluene s o l u t i o n . In one experiment 8.82 x 10 5 monomer moles of [HRu(C0)„] was warmed to 50°C. i n 5 mis. 3 n -3 of toluene containing 50 pis ( «10 moles) of azetidine under a p a r t i a l pressure of CO equal to 670 mm Hg. CO uptake was observed -4 to occur at a very slow rate r e s u l t i n g i n 10 moles of CO consumed i n 72 hours. This i s equivalent to 10% of the azet i d i n e undergoing reaction to carbonylated products. The i . r . spectrum of the r e s u l t i n g orange s o l u t i o n showed a very broad absorption at 2029 cm 1 and a very weak band at 1978 cm In t e r e s t i n g l y , at s i m i l a r concentrations of ruthenium i n the absence of added amine and at 60-80°C. no d i s s o l u t i o n of [HRu(C0) 3] n or [DRu(C0) 3] n i n toluene could be achieved a f t e r 12 hours. This was true regardless of whether ^ or CO was present above the reaction mixture and i s i n d i c a t i v e of some reaction between the polymer and added amine. The lack of a v a i l a b l e amine prevented further studies on t h i s promising reaction at higher azetidine concentrations; an increase i n reaction temperature was not p r a c t i c a l owing to the high 226 v o l a t i l i t y of t h i s material (b.pt=63°C. at 748 mm Hg). The carbonylation of adenine (structure i n F i g . 62) with [HRu(CO) 3] n as c a t a l y s t was attempted i n both DMA and toluene solutions. At 70°C. and 1 atmosphere t o t a l pressure both adenine and [HRu(CO) 3] n dissolved i n DMA to give a lemon-yellow s o l u t i o n -2 -3 {adenine=4.27x10 M. and [Ru]=8.9xl0 M.} within 4 hours, but no gas uptake was observed; DMA, although u s e f u l for s o l u b i l i t y purposes, i s also known to 'poison' the c a t a l y s t (see section 6.1). In toluene, under s i m i l a r conditions of temperature and pressure, and [adenine]/ [HRu(C0) 3] ]S5 5, the reaction was prevented by the l i m i t e d s o l u b i l i t y of both [HRu(CO),,] and adenine. The reaction mixture a f t e r 24 hours J n was f i l t e r e d to give a pale yellow s o l u t i o n which showed peaks i n the carbonyl s t r e t c h i n g region of the i n f r a - r e d spectrum at 2062(m)cm 1 and 2030(w)cm 1 . This suggests that some reaction between [HRu(C0).j] n and adenine must be occurring. 6.4. The [Ru(C0) 2(OAc)] system The acetate polymer [Ru(CO) 2(OAc)] r and the dimer [Ru(CO) 2(OAc)(pip)] may be dissolved i n amines to give yellow solutions which are r e a c t i v e towards CO. The t y p i c a l gas uptake p l o t s show an i n i t i a l a u t o c a t a l y t i c region (e.g. F i g . 63) at 75°C. under 1 atm. of CO. Maximum rates were obtained a f t e r about 20 hours and the sole products were the N-formyl d e r i v a t i v e s (Table XXVII). The dimeric complex [Ru(C0) 2(OAc)(pip)] 2 could be used at the appropriate ruthenium m o l a r i t i e s to reproduce exactly the uptake pl o t F i g . 62. Adenine. -4.0 CO F i g . 63. CO uptake plot f o r the c a t a l y t i c carbonylation of neat piperidine at 75°C..and 1 atm. t o t a l pressure using [Ru(C0) 2(OAc)(pip)] 2, [Ru]=3.OxlO"2M. 229 Table XXVII. Carbonylation of amines using [Ru(CO) ?(OAc)] ^ (TT) * / n or [ R u ( C 0 ) 2 ( O A c ) ( p i p ) ] 2 V J - L ; Amine % conversion to N-f ormylamine Maximum ratexlO^M. s. Primary amines a n i l i n e Secondary amines morpholine (I) 6(20 hrs) 0.40 pi p e r i d i n e (I or II) 15(30 hrs) 1.0 p y r r o l i d i n e (I) 35(35 hrs) 7.4 T e r t i a r y amines pyridine (* Data from r e f . 100) 230 observed with the polymeric catalyst; solution i . r . measurements in the region of maximum activity gave the carbonyl stretching pattern (three peaks in the range 2030-1940 cm "*") expected for the amine dimers (see section 2.1.1) (e.s.r. experiments failed to detect the presence of Ru(I) monomers in such solutions). Precipitation of the dimer could be achieved at any stage of the reaction by addition of excess water to the solvent. In contrast to the hydridocarbonyl system no stoichiometric carbonylation could be detected in the absence of carbon monoxide and quite different kinetics were observed when the maximum rates were analysed. The f i r s t order dependence on CO (Fig. 64) and half-order dependence on [Ru](Fig. 65) may be explained by the scheme outlined below in equations 6.25-6.28. k l J_[Ru(C0)2(0Ac) (pip)] 2 + pip^- kZ_-^Ru(C0) 2(0Ac) (pip) 2 (6.25) (a) (b) k Ru(C0)2(0Ac) (pip) 2 + CO —>Ru(C0) 3(0Ac) (pip) x (6.26) (b) f O Q f -Ru(C0)3(0Ac) (pip) — > pipCO + Ru(C0)2(0Ac) (pip) ^(6.27) f p o t-Ru(C0)2(0Ac) (Pip) ! + Pip —=^-=> Ru(CO)2 (OAc) (pip) (6.28) (b) The i . r . spectroscopic evidence indicates that species (b) can only be present at low concentrations in the region of maximum activity (i.e. k_^ > k^) and so a steady-state treatment gives: h kn[Dimer] [pip] = k_1[Monomer] + k9[CO][Monomer] (6.29) 233 or k - j a j ^ p i p ] = k_ x[b] + k 2[CO][b] (6.30) [b] = k ^ a l ^ t p i p ] (6.31) k^+k^CO] - d[C0] =k„[CO][b] = k l k 2 [ a ^ [ p l p ] [ C 0 ] (6.32) dt k^+k^CO] The observed f i r s t order dependence on [CO] requires k_^>k 2[CO], g i v i n g -d[C0] = k l k 2 [a]'^[pip] [CO] (6.33) dt k_± The a u t o c a t a l y t i c region could be associated with the build-up of a steady-state .concentration of the a c t i v e monomer (b) from the dimer (a). Writing \ [Ru T] K , = z— and [a] = — r — where [Ru m] = concentration of ruthenium 1 k_^ 2 T calculated as monomer, gives -d[CO] = \ h k [Ru ] %[pip][C0] (6.34) dt 2 The observed k i n e t i c s are consistent with t h i s rate law although the p i p e r i -dine dependence was not established. The data of Figures 64 and 65 give an k —3 —3/2 —1 average 2 k 2 value of 2.7x10 M s at 75°C. An.undetectable implies a -3 -1 -1 -1 value <10 M and thus k 2 must be >0.1 M s 234 Chapter 7 AMINE CARBONYLATION USING OTHER RUTHENIUM CARBONYLS 7.1. Introduction In Chapters 4 and 5 a serie s of Ru(III) and Ru(II) compounds with varying carbonyl content ( i . e . K^RuC^ to CsRu^O ^ C J l g ) w ^ s made and aqueous s o l u t i o n studies were conducted. The p o s s i b i l i t y that some of these compounds might be useful carbonylation c a t a l y s t s was also investigated. In the i n i t i a l experiments, each compound was added to neat p i p e r i d i n e , i n quantities s u f f i c i e n t to generate -2 a 1.0x10 M. so l u t i o n at 60°C and 1 atm. t o t a l pressure under CO. Neither K„RuC£, nor Cs„Ru(C0)C£_ would d i s s o l v e i n the neat amine J O Z. D under these conditions. However, Cs2Ru(C0)2C£ 4 did d i s s o l v e to a -2 c e r t a i n extent, and at [Ru]=1.93x10 M, an i n i t i a l induction period of some 900 seconds was followed by a very slow l i n e a r uptake of gas. The i n i t i a l rate was 2.93x10 ^M.s 1 increasing to 3.55x10 ^M.s 1 a f t e r 6 hours, at which time the reaction was stopped. The rea c t i o n f l a s k contained a pale yellow l i q u i d and a white p r e c i p i t a t e which were subsequently separated by f i l t r a t i o n under i n e r t atmosphere. The y i e l d of p r e c i p i t a t e (see section 7.2) a f t e r several hours drying 235 under vacuum was 0.031 gms from 0.055 gms (9.66x10 moles) of Cs2Ru(CO)2C£ 4. The pale yellow f i l t r a t e showed three peaks i n the 1920-2100 cm 1 region of the i n f r a - r e d spectrum and a peak at 1700 cm 1 c h a r a c t e r i s t i c of an amide or urea reaction product. Further i n v e s t i g a t i o n of t h i s system was not conducted owing to the extreme slowness of the carbonylation and also i t s s i m i l a r i t y to the f a s t e r reaction catalysed by CsRu^O^CA^ (see section 7.2). 7.2. The carbonylation of p i p e r i d i n e using CsRu^O^CJl^ A t y p i c a l CO uptake p l o t f o r CsRu(CO)^CJ^ i n p i p e r i d i n e under CO i s shown i n Figure 66. An i n i t i a l period of approximately 300 seconds followed the addition of s o l i d to the neat amine before any gas uptake was observed. The rate of gas uptake gradually increased and was apparently l i n e a r a f t e r 4500 seconds under the conditions noted i n F i g . 66; however, l a t e r observations i n subsequent experiments revealed that the rate was s t i l l increasing slowly even a f t e r the f i r s t 4500 seconds, and for k i n e t i c purposes rate measurements were made at i n t e r v a l s over extended periods of time under constant pressure i n order to determine the maximum rate of CO uptake (Fig. 67). Vapour phase chromatography and mass spectrometry were used to show that the sole product from each reaction was N-formyl p i p e r i d i n e . The mass spectrum of the l i q u i d f i l t e r e d from the reaction mixture showed a parent peak at m/e=113, corresponding to N-formyl p i p e r i d i n e , but no sign of any s i g n i f i c a n t peaks at higher m/e values that might r e s u l t from the urea, 1,1^-carbonyldipiperidine. By comparison with peaks F i g . 66. CsRu(C0)3C£3 - catalysed carbonylation of pipe r i d i n e . 10.5x10 moles c a t a l y s t i n 5 mis. amine, [Ru]=2.10xl0"2M. at 60°C., 1 atm. t o t a l pressure. N 3 F i g . 67. CsRuCCO^CJ^ - catalysed carbonylation of piperidine, 10.5x10 moles ca t a l y s t i n 5 mis. amine, [Ru]=2.10xl0 - 2M. at 60°C, 650 mm Hg t o t a l pressure. Rates measured at (•) 2800 sees., (A) 4100 sees., (•) 6000 sees, and (•) 7800 sees. r e s p e c t i v e l y 238 obtained from pure p i p e r i d i n e and pure N-formyl p i p e r i d i n e , v.p.c. -2 indicated that an experiment at 60°C and [Ru]=1.0xl0 M. under one atmosphere of carbon monoxide resulted i n a 64% y i e l d of N-formyl-p i p e r i d i n e over 44 hours. This compared very favourably with y i e l d s obtained when using other c a t a l y s t s under s i m i l a r conditions (see Table XXVI and XXVII i n section 6.3). The reactions were generally followed f o r several hours i n order to e s t a b l i s h maximum rates, at the end of t h i s time the contents of the reaction f l a s k always consisted of a white p r e c i p i t a t e i n a pale yellow l i q u i d . The p r e c i p i t a t e , a f t e r drying under vacuum, had an i n f r a - r e d spectrum (nujol mull) which c l o s e l y resembled that of a pure sample of p i p e r i d i n e hydrochloride. However, a sample submitted for microanalysis gave C, H and N percentages of only about one-third the expected values (Calculated for Ci-H^NCJl: C ,49. 38; H,9.88;N,11.52%. Found: C,16.81;H,3.30;N,3.90%). This r e s u l t , and the fact that a residue remained a f t e r combustion for microanalysis, strongly suggested that cesium ch l o r i d e was also present and that the mixture consisted of one-third p i p e r i d i n e hydrochloride and two-thirds cesium ch l o r i d e by weight. A flame test on a sample of the s o l i d gave the l i l a c colour c h r a c t e r i s t i c of the cesium ion. These r e s u l t s could be explained by an i n i t i a l r e action such as that depicted i n equation 7.1, i n which the o r i g i n a l complex di s s o l v e s and the mixture of cesium chl o r i d e and p i p e r i d i n e hydrochloride i s p r e c i p i t a t e d . CsRu(C0) 3C£ 3 + 2 pip > CsC£ + pip.HC£ + Ru(CO) 3C£(pip") (7.1) The absence of i n f r a - r e d evidence f o r the presence of H ?0 or OH i n the 239 reaction f i l t r a t e seems to support the idea that a pip e r i d i n o (pip ) species i s involved and that species such as Ru(CO).jC£(pip) (OH) are not s i g n i f i c a n t l y involved. Thus, we seem to have evidence f or pi p e r i d i n e acting as a proton acceptor i n t h i s reaction, as previously discussed i n section 6.1. The reaction postulated i n equation 7.1 would explain the observation that incomplete s o l u t i o n seems to r e s u l t when CsRu(CO),jC£3 i s added to p i p e r i d i n e under argon or carbon monoxide. The gradual d i s s o l u t i o n of the s t a r t i n g complex i s presumably accompanied by con-comitant p r e c i p i t a t i o n of the CsC£/pip.HC£ mixture. The maximum y i e l d -4 of p r e c i p i t a t e from 0.046 gms.(1.08x10 moles) s t a r t i n g material was approximately 0.03 gms. i n a number of experiments. If the reaction proceeds according to equation (7.1) the expected y i e l d of CsC£ and -4 pip.HC£ would be 1.08x10 moles of each or 0.018 gms. of CsC£ and 0.013 gms. of pip.HC£. Thus, the observed and expected y i e l d s of s o l i d material are i n good agreement. The fa c t that the actual y i e l d of s o l i d s never quite equals the expected value of 0.031 gms. may be explained by the s l i g h t s o l u b i l i t y of pip.HC£ i n p i p e r i d i n e . Dropwise add i t i o n of anhydrous d i e t h y l ether to the reaction f i l t r a t e r esulted i n the p r e c i p i t a t i o n of a further white s o l i d , which was confirmed by i . r . spectroscopy to be p i p e r i d i n e hydrochloride. -2 -5 An experiment conducted at [Ru]=2.15x10 M ( i . e . 10.76x10 moles i n 5 mis piperidine) at 60°C under 1 atm.of argon resulted i n an apparent gas evolution of 1.85x10 ^moles within 20 seconds with no subsequent evolution of gas. The s l i g h t evolution i s much too low for any loss of co-ordinated CO from the complex and i s probably a t t r i b u t a b l e to the 240 solvent e f f e c t that has been observed previously with p i p e r i d i n e (see section 3.5). Thus, a t r i c a r b o n y l species i n v o l v i n g co-ordinated pip , as shown i n eq. 7.1, i s l i k e l y formed i n i t i a l l y ; t h i s could be 5 co-ordinate or 6 co-ordinate with a further p i p e r i d i n e l i g a n d . Equation 7.1 does not consider any subsequent carbonylation reaction (see below). When CsRu(CO) 3C£ 3 (0.05 gms) was s t i r r e d i n p i p e r i d i n e (5 mis) at 65°C under N 2 for a period of 12 hours, the mixture remained cloudy throughout the r e a c t i o n and could be separated eventually into a white p r e c i p i t a t e and a pale yellow l i q u i d . The l i q u i d i . r . spectrum showed peaks at 1922(s),1975(m) and 2020(s) cm 1 comparable with those reported for [Ru(C0) 2(OAc)(pip)] 2 (Fig- 1 i n section 2.1.1). N-formylpiperidine could neither be detected i n the i . r . spectrum nor found by v.p.c. a n a l y s i s . However, the l i m i t of detection of N-formyl-p i p e r i d i n e by v.p.c. was calculated to be i n the range of 10-20x10 5 moles at highest machine s e n s i t i v i t y and favourable operating conditions. So, t h i s means of detection might not be r e l i a b l e at very low concentra-t i o n s . In f a c t , concentration of the reaction mixture f i l t r a t e to approximately half-volume under vacuum gave a l i q u i d which showed the presence of N-formyl product by a peak at 1720 cm"1 i n the i . r . spectrum. Thus, stoichiometric carbonylation seemed to be occurring i n the absence of carbon monoxide according to a process such as eq. 7.2 which follows that described by eq. 7.1. i . e . Ru(C0) 3C£(pip~) E i P - ^ R u(C0) 2C£(pip~) + pipCO (7.2) 241 7.3. Ruthenium carbonyl species from the CsRu(CO) 3C£ 3/piperidine/CO system The separation of products from p i p e r i d i n e solutions when em-ploying [HRuCCO)^]^ and [Ru(CO) 2 (OAc)(pip)] 2 a s c a t a l y s t s was discussed i n e a r l i e r chapters (sections 2.1.1 and 3.6). A s i m i l a r approach was taken f o r t h i s CsRu(CO) 3C& 3 system. An experiment was set up under -2 the following conditions, [Ru]=2.18x10 M., t o t a l pressure = 650 mm Hg under CO, T=60°C and volume of p i p e r i d i n e = 5 mis. Frequent observations of the carbonylation rate revealed a maximum value of 4.79x10 5M.s 1 a f t e r approximately 5000 seconds. Thereafter, the rate decreased and the reaction was stopped a f t e r 8000 seconds, at which time the CO uptake was occurring at a rate of 4.41x10 5M.s The white pre-c i p i t a t e of cesium c h l o r i d e and p i p e r i d i n e hydrochloride was removed by f i l t r a t i o n under argon and the yellow f i l t r a t e concentrated by pumping under vacuum. Several hours of treatment i n t h i s manner was considered s u f f i c i e n t to remove p i p e r i d i n e and the very small amount of N-formyl product (possibly 3 mole % by rough c a l c u l a t i o n from observed rates of CO uptake, neglecting stoichiometric carbonylation). Nevertheless the remaining material was o i l y i n appearance. A continuation of the method previously adopted i n the other systems, i n v o l v i n g dropwise addition of de-gassed water to the o i l and vigorous s t i r r i n g , succeeded i n converting the o i l to a brown powder (1). The compound (1) was f i l t e r e d under argon and dried under vacuum for several hours. This s o l i d gave the following peaks i n the s o l i d state (nujol mull) i n f r a -red spectrum, 2040(m),2015(m),1965(m,br),1920(m,br),1263(w),1090(w,br), 242 1030(w,br),880(w),810(m),720(w) and 660(w) cm 1 . The carbonyl stretching region of the spectrum i s shown i n Figure 68. This region i s i d e n t i c a l to that obtained i n the l i q u i d i . r . spectrum of the reaction mixture and also c l o s e l y resembles that recorded for [Ru(CO) 2(OAc)(pip)] 2 i n Figure 1. D i s s o l u t i o n i n n-heptane or a s i m i l a r non-polar solvent could not be achieved; however, the s o l i d was r e a d i l y soluble i n C H 2 C £ 2 to give a s o l u t i o n with peaks i n the carbonyl region at 2043(m),2015(m),1975(m,br) and 1935(m,br) cm The absence of an N-H s t r e t c h i n g v i b r a t i o n near 3200 cm 1 seemed to i n d i c a t e that there was no free p i p e r i d i n e present. Thin layer chromatography was used to check the p u r i t y of t h i s compound. When dichloromethane was used as solvent and methanol as eluent on s i l i c a gel the compound gave one spot with R =0.77. A r sample of (1) submitted for analysis gave the following r e s u l t s , 0=37.61%, H=6.06% and N=6.29%. If we assume that the f i n a l product has retained two CO groups and one c h l o r i d e , then a number of species of formula [Ru(CO) 2CJl(pip)__]_ are conceivable. The calculated a n a l y t i c a l r e s u l t s for a number of formulae of t h i s general form and also some t r i c a r b o n y l species are shown i n Table XXVIII. An examination of the figures i n Table XXVIII indicates that several of these formulae are nearly correct for the compound obtained, but a l l of them show some discrepancies, p a r t i c u l a r l y i n the percentage of carbon. In order to consider t h i s problem further a d i f f e r e n t synthetic approach was adopted (see section 7.5). 243 244 Table XXVIII. Possible formulae f or the compound (1) from the CsRu(CO) C£ /piperidine/CO system Formula C% H% N% Reaction Product (1) 37.61 6.06 6.29 Ru(CO) 2C£(pip) 30.26 3.96 5.04 Ru(CO) 2C£(pip) 2 39.71 6.07 7.72 Ru 2(CO) 4C£ 2(pip) 3 35.61 5.15 6.56 Ru(CO) 3C£(pip) 31.41 3.60 4.58 Ru(CO) 3C£(pip) 2 39.94 5.63 7.17 Ru 2(CO) 6C£ 2(pip) 3 36.94 4.83 6.16 Ru(CO) 2C£ 2(pip) 2 36.17 5.52 7.03 245 7.4. K i n e t i c s and mechanism of p i p e r i d i n e carbonylation In the presence of CsRu(CO) 3C& 3 The e f f e c t of varying [Ru] upon the maximum rate of the reaction was studied at 60°C and 1 atm.total pressure under CO. Each reaction was i n i t i a t e d by dropping a glass bucket containing CsRu(CO)3C&-. ^- n t o n e a t p i p e r i d i n e i n the re a c t i o n f l a s k and rate measurements were made at frequent i n t e r v a l s u n t i l a maximum rate of gas uptake had been achieved. The t y p i c a l r e s u l t of such an experiment i s shown i n Figure, 67. The pressure was maintained at 1 atm. throughout each experiment by frequent a d d i t i o n of gas to the e n t i r e apparatus. The r e s u l t s from these experiments are shown i n Table XXIX. The data from Table XXIX are plo t t e d i n Figure 69, which shows a reasonable f i r s t order dependence of maximum rate on [Ru]. The dependence of the maximum rate upon p a r t i a l pressure of CO was studied at [Ru]=2.2xl0 M and T=60°C, the data are shown i n Table XXX. A p l o t of maximum rate against p a r t i a l pressure of CO gives a reasonable str a i g h t l i n e (Fig. 70), i n d i c a t i n g a f i r s t order dependence of maximum rate upon [CO], at l e a s t at t h i s ruthenium concentration. Varying the CO pressure did not change the shape of the uptake p l o t , except that a reduction i n the CO pressure resulted i n a longer time elapsing before the maximum rate was achieved. For instance, the experiment at P(CO)=607 mm Hg reached a maximum rate a f t e r 4200 seconds, whereas the experiment at P(C0)=47 mm Hg was observed to reach a maximum rate a f t e r 12,000 seconds. The dependence of maximum rate on p i p e r i d i n e concentration was 246 Table XXIX. E f f e c t of varying [Ru] upon the maximum rate of carbony-l a t i o n of p i p e r i d i n e i n the presence of CsRu(CO) C£„. ? [Ru] x 10 M. 2.18 2.15 1.50 1.04 0.40 Maximum Rate x 10 M.s 5.81 5.36 3.90 3.47 1.05 ( A l l experiments at T=60°C and t o t a l pressure = 1 atm.) 247 Fig. 69. CsRu(00)3(^3 - catalysed carbonylation of piperidine. Ruthenium dependence at 1 atm. total pressure, 60°C, neat amine. 248 Table XXX. E f f e c t of varying the p a r t i a l pressure of carbon monoxide upon the maximum rate of carbonylation of p i p e r i d i n e i n the presence of CsRu(CO) C£„ P a r t i a l pressure of CO(mm Hg.) Maximum Rate x 10~*M.s 1 607 5.81 497 4.78 397 3.44 97 1.10 47 0.34 (Vapour pressure of p i p e r i d i n e at 60°C = 153 mm Hg. -3 -1 S o l u b i l i t y of CO i n p i p e r i d i n e at 60°C = 5.9 x 10 M atm . _2 A l l experiments at [Ru] = 2.2 x 10 M and T = 60°C.) 249 F i g . 70. CsRu(CO)3C&3 - catalysed carbonylation of p i p e r i d i n e CO dependence for neat amine solutions at 60°C. and [Ru]=2.2xl0 250 determined by using toluene-piperidine mixtures. Since the s o l u b i l i t i e s of CO i n toluene and p i p e r i d i n e were very s i m i l a r 1 ^ , at l e a s t at 75°C, and the vapour pressures of these two solvents d i f f e r e d by only about 15-20 mm between 50 and 75°C, toluene was a s u i t a b l e solvent with which to d i l u t e the amine substrate. The e f f e c t of varying [piperidine] i s shown i n Table XXXI. Once again a f i r s t order dependence was observed (Fig. 71), and an examination of the i n f r a - r e d spectrum of the pre-c i p i t a t e remaining at the end of each experiment (Yield=0.030 gms, c f . section 7.2) showed peaks due to p i p e r i d i n e hydrochloride and no evidence of any metal carbonyl peaks. This indicated that the reduction i n rate upon lowering [piperidine] was not due to a lower s o l u b i l i t y of the i n i t i a l complex, since no CsRu(C0).jC£3 appeared to be present i n s o l i d form at the end of the r e a c t i o n . When CsRu(CO) 3C£ 3 was e q u i l i -brated with p i p e r i d i n e at 65°C under vacuum for 12 hours and then exposed to CO the induction period and a u t o c a t a l y t i c region were no longer observed. Instead, a rapid i n i t i a l gas uptake occurred (Fig. 72) followed by a gradual decrease i n the reaction rate. The i n i t i a l rate of gas uptake under such conditions was estimated as 29.8x10 ^M.s compared to 3.15x10 5M.s 1 (see Table XXXII) when the r e a c t i o n was c a r r i e d out without p r e - e q u i l i b r a t i o n . This observation with a mixture which has been previously e q u i l i b r a t e d under vacuum seems to r u l e out any mechanism inv o l v i n g a pre-equilibrium with CO, and the i n i t i a l slow, gas uptake could be j u s t a r e s u l t of gradual d i s s o l u t i o n of CsRu(CO) 3C£2 i n p i p e r i d i n e . The mechanism of the carbonylation must explain the k i n e t i c s 251 e XXXI. E f f e c t of varying [piperidine] on maximum rate carbonylation of p i p e r i d i n e i n the presence of CsRu(CO) 3C£ 3 [piperidine]M. Maximum rate x lO^M.s 1 10.1 5.81 8.0 4.25 5.0 3.40 « 3.0 1.50 (Experiments conducted at [Ru]=2.2 x 10 M, T=60°C and 1 atm.total pressure) K5 1-0 20 T I M E X10"^sec _5 ; F i g . 72. CsRu(CO)3C£3 - catalysed carbonylation of pipe r i d i n e . 11.0x10 moles of c a t a l y s t i n 5 mis. X amine, [Ru]=2.20x10"2M. at 65°C. and 580 mm Hg t o t a l pressure. - Pre-equilbrated under vacuum for .12 hours. (a) i n i t i a l CO uptake,, subsequent uptake measured: commencing' at (b) 650 s"ecs.,r( c) 1500 sees, (d) 2400 sees. 254 Table XXXII. E f f e c t of temperature v a r i a t i o n on maximum rate of carbonylation of pi p e r i d i n e i n the presence of CsRu(CO) 3C£ 3. Temperature (°C) P ( t o t a l ) 1 1 1 1 1 1 H ^ ^ ( C O ) mm Hg Maximum Rate xlO^M.s -1 60 550 397 3.08 65 584 397 3.15 70 624 397 3.79 80 730 397 6.80 _2 ( A l l experiments at [Ru]=2.2xl0 M i n neat pip e r i d i n e ) 255 and observations discussed i n t h i s section and be consistent with rate law 7.3. -d[CO] = k[Ru][pip][CO] (7.3) dt One possible mechanism giving the correct rate law i s expressed i n Scheme (1). The mechanism involves an i n i t i a l slowly established pre-equilibrium with CO to form the ac t i v e species which then undergoes CO i n s e r t i o n into a Ru-N bond, followed by displacement of the product and regeneration of the s t a r t i n g species. Thus, i f the reaction commences v i a the process described i n equation 7.1 the carbonylation might continue as - k l Scheme 1 CO + Ru(CO) C£(pip ) . . ^Ru(CO) ( 1C£(pip ) n —± 1 n+± (I) k 3 Y k 2 pip (7.4) pipCO + Ru(CO) nC£(pip ) < f a g t R u ( C Q ^ + 1 C & ( p i p )(pip) (I) (III) The gas uptake pl o t s show an i n i t i a l a u t o c a t a l y t i c region which could represent the slow accumulation of species (II) or a l t e r n a t i v e l y the i n i t i a l r e action 7.1 could be slow. A steady-state treatment applied to species (II) i n Scheme (1) gives -d[CO] = W C O ] [ p i p ] [ I ] (7.5) dt k ^ + k ^ p i p ] which gives a rate law of the form shown i n equation 7.3 i f k ^>>k 2[pip]. However, the evidence from the experiment i n which CsRu(CO) 3C£ 3 was pre-256 e q u i l i b r a t e d with p i p e r i d i n e under vacuum argues against the pre-equilibrium shown i n Scheme (1). An a l t e r n a t i v e mechanism (Scheme (2)) involves an i n i t i a l e q uilibrium with p i p e r i d i n e . Applying a steady-state treatment to species (II) gives the expression (7.7) and rate law (7.8). k -: ^Ru(CO)_C£(pi P ) ( P i p ) (7.6) Scheme 2 pip + Ru(CO) C£(pip )^-p - i ( i ) ( I D CO pipCO + Ru(CO)_C£(pip ) < ^ s f c — Ru(CO) n + 1C£(pip ) (pip) (I) (III) k-JpipHl] - k ^ t i i . + k ^ c o n m . d [ c o ] _ k A [ P i p ] [ I ] [ C O ] dt k^+k^CO] (7.7) (7.8) Therefore, providing k_^>>k2[C0] a rate expression of the desired form i s obtained and an observed rate constant of the form shown i n equation 7.9 should r e s u l t . k = k l k 2  k - l (7.9) Experiments conducted i n order to determine the e f f e c t of temper-ature v a r i a t i o n upon the k i n e t i c s are summarised i n Table XXXII. Between 60 and 70°C there appeared to be no dramatic change i n the shape of the CO uptake p l o t . The t o t a l pressure was adjusted so as to maintain a p a r t i a l pressure of 397 mm Hg of CO i n each experiment. The s o l u b i l i t y of CO i n p i p e r i d i n e v a r i e s l i t t l e with temperature, Table XXXIII. A comparison of various p i p e r i d i n e carbonylation c a t a l y s t s 2 5 - 1 Catalyst [RujxlO M. T°C Maximum Rate x 10 M.s %Conversion [HRu(C0) o] j n 2.2 60 3.52 -[HRu(CO)_] j n 2.2 75 10.40 32(55 hrs.) Ru 3(CO) 1 2 3.0 75 7.20 45(65 hrs.) [Ru(CO) 2(OAc)(pip)] 2 3.0 75 1.00 30(70 hrs.) CsRu(tO) 3C£ 3 2.18 60 5.81 -CsRu(CO) 3C£ 3 1.04 60 3.47 64(44 hrs.) ( A l l experiments conducted at 1 atm.total pressure under CO) F i g . 7 3 . CsRu(CO)3C£3 - catalysed carbonylation of p i p e r i d i n e Log 1 ^(maximum rate) vs. 1/T from data i n Table XXXII. 259 being 5.8x10 3M. atm 1 at 75°C and 6.5x10 3M. atm 1 at 21°C. Thus, for the purposes of our study the s o l u b i l i t y of CO i n p i p e r i d i n e can be viewed as nearly constant over the temperature range 60 to 70°C. If the rate of reaction can be expressed i n the form of equation 7.3 where k i s a simple rate constant, then a p l o t of log (maximum rate) against 1/T °K 1 should give a st r a i g h t l i n e Arrhenius p l o t . In f a c t , -3 -1 i f [CO] i s assumed constant at 5.9x10 M atm for the temperature range from 60 to 70°C, then the log p l o t from the data i n Table •XXXII i s of a d i s t i n c t l y non-linear form (Fig. 73) and t h i s emphasises the d i f f i c u l t i e s involved i n i n t e r p r e t i n g temperature v a r i a t i o n data i n the presence of the s o l i d r eaction products. In summary, CsRu(C0)3C£-. appears to be a u s e f u l c a t a l y s t for the s e l e c t i v e conversion of p i p e r i d i n e to N-formyl p i p e r i d i n e under mild conditions. The percentage y i e l d s of the formyl product and maximum rates of CO uptake compare very favourably with those obtained when using other ruthenium carbonyl c a t a l y s t s (Table XXXIII). 7.5. Synthesis of [ R u ^ O ^ C J ^ ^ a n c * a p i p e r i d i n e d e r i v a t i v e The method of Cleare and G r i f f i t h 1 0 5 was used to prepare the dimer [Ru(CO)^Cl^]^. Ruthenium t r i c h l o r i d e (2.0 gms), 90% formic acid (20 mis) and concentrated hydrochloric acid (15 mis) were mixed and refluxed f or 72 hours i n order to complete conversion to Ru(C0),jC£3 i n s o l u t i o n . The so l u t i o n was then cooled and the solvent was removed under reduced pressure to give a yellow residue. This material was p u r i f i e d by sublimation under vacuum at 140°C to give a pale yellow sublimate (Calculated for Ru„C,0,C£.: C,14.06%. Found: C,13.54%) with 260 i n f r a - r e d spectra i n n u j o l mull and chloroform (dissolves slowly on warming) which were consistent:;with those previously reported by Bruce and Stone 182 P u r i f i c a t i o n by sublimation i s thought to y i e l d l a r g e l y the trans-isomer of [ R u ^ O ^ C J ^ ^ * although being of the same symmetry as the cis-form and possessing the same number of a c t i v e modes i n the carbonyl i n f r a - r e d , i t i s not easy to d i s t i n g u i s h the geometrical isomers (Fig. 74) by t h i s means. The compound [Ru (CO) -jCJ^ 3 2 has been used previously to make der i v a t i v e s with various donor ligands such as pyridine, THF and phosphines, i n which the chlorine bridges F i g . 74 C£ CO OC / oc ca / \ Ru Ru \ / CI / \ CO CO CO CI CO CO CI \ Ru C£. / \ A OC Ru / \ CO CI CO CO c i s - C 2^ symmetry I.R. a c t i v e - A +2B , u u/ trans - C2^ symmetry I.R. a c t i v e - A U+2B U 172 are e a s i l y broken. In p a r t i c u l a r , complexes of the form c i s -Ru (00)^0^2 (py) and cis-Ru (CO) 2 ^ 2 ^ 7 ) 2 have been made and characterised by t h e i r i . r . spectra and melting points (Table XXXIV) In an attempt to prepare an analogous compound of formula Ru(CO)3C£,2 (pip) the synthetic technique employed by Benedetti and 172 co-workers to make pyridine d e r i v a t i v e s was considered. The 261 Table XXXIV. Pyridine derivatives of [Ru(C0) 3C£ 2_ 2 Compound m.pt.(°C) v(C=0) cm 1 fac-Ru(CO)3C£2(py) 170-172 2136(s),2075(s),2051(s) cis-Ru(CO) 2C£ 2(py) 2 264-266 2070(s),2006(s)* (* in CH 2C£ 2, data from ref. 172) 262 compound [Ru(C0) 3C£ 2] 2 (0.10 gms., 1.95xl0 _ i tmoles) was added to C H 2 C £ 2 (5 mis) under nitrogen at room temperature and p i p e r i d i n e ( 0.04 mis, -4 4.04x10 moles) was added with constant s t i r r i n g . A [pip]:[Ru(CO) 3Cft 2_ 2 value of 2 was used i n order to avoid formation of Ru(C0) 2C£ 2(pip) 2 which should occur i n the presence of excess ligand. The s o l u t i o n was i n i t i a l l y lemon yellow, changing to golden yellow over a period of 20 minutes and subsequently showing no further changes i n colour. A f t e r 2 hours the s t i r r i n g was stopped and the solvent was removed under reduced pressure.- A deep yellow o i l was produced and slowly s o l i d i f i e d under vacuum to give a bright orange powder. This powder was then p u r i f i e d by s o l u t i o n i n C H 2 C £ 2 and r e p r e c i p i t a t e d using heptane. The i n f r a - r e d spectrum (nujol mull) of t h i s material showed three sharp peaks i n the v(CH0) region at 2040(s),1975(s) and 1965(s) cm"1, (cf.[Ru(CO) 3C£ 2] 2 i n 172 -1 nu j o l mull at 2140(s),2092(s) and 2066(s) cm ), and also peaks at 3230(w,sh.),3110(m),3050(m),1278(m),1198(m),1100(m),1080(w),1035(m), 1015(m),924(m),888(s),863(s),818(m),730(w,br),650(m),635(m),585(s), 545(m),499(s),431(m),305(s) and 285(m,br). Previous d e r i v a t i v e s con-t a i n i n g co-ordinated p i p e r i d i n e (sections 2.1.1 and 3.6) showed no i n d i c a t i o n of N-H st r e t c h i n g v i b r a t i o n s i n t h e i r i n f r a - r e d spectra which would seem to suggest that the underlined frequencies represent the presence of free p i p e r i d i n e despite the manner i n which t h i s s o l i d was treated. Thin layer chromatography using C H 2 C £ 2 as solvent and CH^OH as eluent on s i l i c a gel gave one spot with Rp=0.77. This was i d e n t i c a l to the r e s u l t obtained from the compound described as Ru(C0) 2C£(pip) x i n section 7.3. However, the r e s u l t of microanalysis 263 did not correspond to that of the previous compound, nor was i t close to that expected for Ru(CO) 3C£ 2(pip) • (Calculated for RuC g0 3H i ; LC£ 2N:C,28.14;H,3.23;N,4.10%. Found: C,32.67;H,5.30;N,6.00%). A sample submitted for mass spectrometry gave a parent c l u s t e r con-t a i n i n g 13 peaks i n the range of m/e values from 390 to 402 and with s i m i l a r c l u s t e r s from 362 to 374 and 334 to 346 suggesting successive l o s s of two CO groups. The p o s i t i o n of the parent peak suggested the presence of the species Ru(CO) 2C£ 2 ( p i p ^ (molecular weight=398.1) although a rather d i f f e r e n t microanalysis should r e s u l t i f t h i s was the case (see Table XXVIII). C l e a r l y , t h i s synthesis has resulted i n a complex, which although apparently pure, i s rather i l l - d e f i n e d . Of i n t e r e s t , the species 183 [Ru(CO) 3C£ 2_ 2 n a s recently been used i n reactions with primary amines (eq. 7.10) to give chlorocarbamoyl complexes. However, according CHC£ [Ru(CO) 3C£ 2] 2 + 8RNH2- ^2Ru(CO) 2(RNH 2) 2C£(CONHR) + 2RNH2.HCi> (7.10) 183 to t h i s report , secondary amines did not give r e a d i l y characterisable products. Thus, the reaction of [Ru(CO) 3C£ 2_ 2 with p i p e r i d i n e as conducted i n t h i s work did not a s s i s t i n determining products derived from CsRu(CO) 3C£ 3 i n p i p e r i d i n e under carbon monoxide (section 7.3). At the present time the evidence indicates that the ac t i v e species involved i n the c a t a l y t i c conversion of pi p e r i d i n e to N-formyl p i p e r i d i n e , i n the presence of CsRu(CO) 3C£ 3, i s of the form [Ru(CO) 2C£(pip)_J although a compound pr e c i p i t a t e d from s o l u t i o n during the period of maximum gas uptake rate (section 7.3) has yet to be su c c e s s f u l l y i d e n t i f i e d . 264 7.6. The carbonylation of piperazine using Ru-(CO) „ ~J 1^ . The use of Ru^CCO)^ f ° r the carbonylation of p i p e r i d i n e has been reported p r e v i o u s l y 1 ^ (also see Table XXXIII i n section 7.4). Gas uptake p l o t s c l o s e l y resembled those obtained when using [HRu(CO) 3] n as ca t a l y s t with s i m i l a r maximum rates and percentage conversions to the N-formyl product. Attempts to i d e n t i f y the i n -organic reaction products were l a r g e l y unsuccessful. When R u ^ C O ) ^ i s warmed with excess p i p e r i d i n e i n toluene solvent under nitrogen atmosphere a deep red o i l i s r a p i d l y produced i n a red s o l u t i o n . 135 Previous workers have characterised several d i f f e r e n t compounds with varying degrees of s u b s t i t u t i o n when R u ^ C O ) ^ i s refluxed with a n i l i n e or nitrobenzene (e.g. RU 3(C0)^QNP1I and Ru,j(C0)g(NPh)2 from nitrobenzene and HRU^CCO^QHNPII from a n i l i n e ) . Although experimental d e t a i l s were somewhat l i m i t e d , i t appeared that the tri-ruthenium c l u s t e r was preserved, as was also observed i n reactions with phos-184 phines. Given the previous d i f f i c u l t i e s experienced i n i s o l a t i n g p i p e r i d i n e complexes i n t h i s work i t seemed inadvisable to devote much time to t h i s aspect of the work. Instead the u t i l i t y of R u ^ C O ) ^ was extended to another system i n v o l v i n g piperazine (a s o l i d substrate) and d e r i v a t i v e s of piperazine. Reactions were conducted at 75-80°C i n toluene using a range of CO pressures and ruthenium concentrations. A t y p i c a l CO uptake p l o t with piperazine i s shown i n Figure 75. An i n i t i a l l i n e a r gas uptake rate slows down quite r a p i d l y ; f o r example, under the conditions noted i n Figure 75 the i n i t i a l rate of 8.40 M.s 1 has decreased to 2.86x10 5M.s 1 a f t e r only 1 hour. 0-5 -j.O TIME x 1 0 " * sec F i g . 75. R u 3 ( C 0 ) 1 2 - catalysed carbonylation of piperazine. 15.6xl0" 5moles Ru i n 2 mis. toluene, [Ru]=7.8xlO"2M. [piperazine]=l.45M. at 80°C. and 1 atm. t o t a l pressure. 266 The possible products of carbonylation, 1-formylpiperazine and 1,4-bis-formylpiperazine were prepared and characterised according to the methods described i n Chapter 2 (section 2.1.1). -3 A preparative experiment was established using [Ru]=7.80x10 M, [piperazine]=1.42M. i n toluene at 80°C under an atmosphere of CO. Af t e r 48 hours the reaction was stopped, the solvent removed under vacuum and the residue dissolved i n methanol. A sample of the r e s u l t i n g methanolic s o l u t i o n was transferred to a Carius tube and sealed under argon u n t i l such time as v.p.c. analysis was possible. Similar tubes were prepared containing piperazine i n methanol and 1-formyl piperazine i n methanol. The use of a Pennwalt 223 column enabled v.p.c. analysis of such amine containing solutions to be conducted. The reaction product s o l u t i o n was shown to contain un-reacted piperazine and a very small amount (less than 1% conversion of piperazine) of the 1-formyl product. Since 1,4-bis-formylpiperazine has a melting point of 126-129°C. i t would not elute under the v.p.c. analysis conditions. In addition, t h i n layer chromatography was applied to the product s o l u t i o n using s i l i c a gel plates and chloroform as eluent. The r e s u l t s are shown i n Table XXXV. No 1-formylpiperazine i s detected and although the 1,4-bis-formyl product cannot be separated from piperazine by t h i s means i t i s presumably the major product. The accumulation of the amide product l i k e l y r e s u l t s i n the decreasing amine carbonylation rate (cf. section 6.1). The observed spot at R =0.13±0.02 obtained with the reaction product i s probably due to F remaining piperazine. The r e s u l t s of k i n e t i c studies i n which [Ru] was varied are shown 267 Table XXXV. Thin-layer chromatography r e s u l t s from an experiment with Ru^CCO)^ a n d piperazine Compound R values r piperazine 0.15±0.02 1,4-bis-formylpiperazine 0.15±0.02 1-formylpiperazine 0.37 Reaction products 0.13±0.02; 0.60 (thought to be inorganic) (* methanolic solutions using s i l i c a g el plates and CHC£ as eluent) 268 i n Table XXXVI and a plot of i n i t i a l rate against [Ru](Fig. 76) shows a f i r s t order dependence on [Ru]. In order to compare the reaction with the carbonylation of neat p i p e r i d i n e using the same ca t a l y s t an experiment was conducted at 75°C, -2 [Ru]=3xl0 M. and [piperazine]=10.14M. i n toluene. An i n i t i a l rate of 10.5x10 ^M.s 1 was obtained, which compares with an i n i t i a l rate of 7.2xl0~ 5M.s _ 1 (Table XXXIII section 7.4) for the Ru^CO).^ catalysed carbonylation of neat p i p e r i d i n e ( i . e . 10.14M) under the same conditions (which also shows a rate decrease with time). The i n i t i a l rate observed with piperazine i s somewhat lower than that observed with p i p e r i d i n e i n terms of the number of r e a c t i v e amine s i t e s ; however, t h i s can be ex-179 plained by the r e l a t i v e l y low b a s i c i t y of piperazine (pK^=9.81, pK2=5.55 at 25°C) compared to that of p i p e r i d i n e (pK =11.12 at 25°C) and the 3. generally observed decrease i n r e a c t i v i t y towards carbonylation with decreasing b a s i c i t y of the amine (see section 6.3). In general, the r e s u l t s indicated that the t o t a l percentage conversion to carbonylated products was too low to make the carbony-l a t i o n of piperazine using Ru-^CO)^ of p a r t i c u l a r commercial i n t e r e s t -2 (e.g. If [Ru]=9.0xl0 M, [piperazine]=1.5M i n toluene, p ( t o t a i ) = x a t m at 80°C, conversion to product estimated from gas uptake data a f t e r 20 hours = 8%). Ru 3(CO)^2 w a s also used to catalyse the carbonylation of 1-formyl piperazine. The i n i t i a l gas uptake was much slower than that of piperazine under s i m i l a r conditions. 1,4-Bis-formylpiperazine i n toluene showed no CO uptake i n the presence of Ru^CO)^- The l i q u i d 1-methyl piperazine 269 Table XXXVI. E f f e c t of varying [Ru] upon the i n i t i a l rate of carbonylation of piperazine i n the presence of R u 3(CO) 1 2 2 5 - 1 [Ru] x 10 M. I n i t i a l rate x 10 M.s 1.5 1.20 2.25 1.80 2.94 2.40 3.42 3.35 5.04 4.80 5.70 5.50 7.80 8.40 ( A l l experiments at 80°C, t o t a l pressure = i n toluene) 1 atm, [piperazine]=1.45M 270 F i g . 76. Ru3(CO) 12 - catalysed carbonylation of piperazine. Ruthenium dependence at 1 atm. t o t a l pressure, 80°C. and [piperazine]=1.45M. i n toluene. 271 was carbonylated i n the presence of both Ru^CCO)^ and [HRuCCO)^]^ (see Table XXXVII). The use of 1-methyl piperazine as a substrate appeared promising although the fate of carbonylation decreased quite - 2 , r a p i d l y , for example, the experiment employing R u ^ C O ) ^ at [Ru]=5.4xl0 M. with an i n i t i a l rate of 8.47x10 5M.s 1 showed a rate decrease to 5.12x10 5 M.s 1 within 1000 seconds(Fig. 77). In summary, Ru 3(CO)^ 2 has proven to be a u s e f u l c a t a l y s t for carbonylation of piperazine and d e r i v a t i v e s , although i n h i b i t i o n of the reaction by the amide products seems to be even more of a problem than i n the p i p e r i d i n e system and an i d e n t i f i c a t i o n of the r e s u l t i n g inorganic species has not been achieved. The t . l . c . r e s u l t s (Table XXXV) indica t e the presence of only one inorganic product with Rp=0.60 under the conditions noted. Since, no i n i t i a l evolution of CO was observed (by gas evolution measurements) when adding Ru 3(C0)^ 2 t o toluene i n the presence of these substrates, the l i k e l y product would probably be of the form Ru 3 ( C 0 ) 1 2 (pip)__,Ru 3 (C0)±1 (pipC0)__ or HRu 3 ( C 0 ) 1 2 (pip)__ i n which an uncertain number (x) of amine ligands are present and hydridic compounds are possibly produced by hydrogen abstraction from the amine. 135 A comparison with products derived' from nitrobenzene or a n i l i n e and Ru 3(C0)^ 2 indicates that the amide products from reactions of piperazine or p i p e r i d i n e with Ru 3(C0)^ 2 could d i s p l a y a wide v a r i e t y of structures. Table XXXVII. Carbonylation of piperazine derivatives Substrate Catalyst [Ru]xl0 2M T°C P , ..mm He (total) B Max.rate x '. 1-formyl piperazine (neat) Ru 3(CO) 1 2 5.40 80 760 0. 93 1,4—bis-formyl piperazine(0.66M i n toluene) R u 3 ( C O ) 1 2 6.39 80 760 -1-methyl piperazine (neat) Ru 3(CO) 1 2 5.40 80 760 8. 47 1-methyl piperazine (neat) R u 3 ( C 0 ) 1 2 5.82 80 760 10. 64 1-methyl piperazine (neat) [HRu(C0) 3] n 2.30 80 760 1. 91 piperazine (10.14M i n toluene) R u 3 ( C O ) 1 2 3.00 75 760 10. 50 6.0 in "o E O4.0h fit (ft < 2.0 C5 TIMFx10*?sec 1.0 l ± f \ l h R u 3 ( c ° ) l 2 - catalysed carbonylation of 1-methyl piperazine 5.4x10 moles Ru, 1 ml. neat amine, [RU]=5.4X10"2M. at 80°C. and 1 atm t o t a l pressure. 274 Chapter 8 SOME RELATED -RUTHENIUM CARBONYLPHOS-PHINE CHEMISTRY 8.1. Introduction Several carbonylphosphines of ruthenium were discussed in Chapter 1 as part of the general review of ruthenium carbonyl chemistry. In view of some studies on the interaction of our [HRuCCO)^]^ polymer with PPh^ (section 8.2), i t i s worth con-sidering some of the known carbonylphosphine complexes in more detail. The complex Ru(CO)3(PPh.j)2 has found application as a hydroformylation c a t a l y s t 5 1 and as an important precursor for a range of ruthenium species. A single stage preparation reported 178 by Levison and Robinson provides a convenient method of synthesis by successive addition of "ruthenium trichloride", aqueous formal-dehyde and sodium borohydride to a solution of triphenylphosphine in boiling 2-methoxyethanol.' After 1 hour this solution can be cooled to give high yields ( 80%) of a crystalline solid that analyses well for the desired species (Calculated for RuC Q QH q n0,P o: C,66.0; 275 H,4.25%. Found i n our work: C,66.16;H,4.10%)• This method seems more convenient than previous synthetic procedures, such as the 18 5 reaction of R u 3 ( C O ) 1 2 with PPh 3 or the reduction of RuC& 2(C0) 2(PPh 3) 2 186 with zinc dust i n the presence of carbon monoxide. The presence of only one peak i n the v(C=0) region of the i n f r a - r e d spectrum strongly suggests a t r i g o n a l bipyramidal structure with 3 equatorial CO groups. 18 7 There i s only one report of the complex Ru(C0) 2(PPh 3) 3 which was made by using methoxide ion to deprotonate the species [RuH(C0) 2(PPh 3) 3] Once again a s i n g l e carbonyl s t r e t c h was observed i n the s o l u t i o n i . r . at 1905 cm 1 which i s very close to values reported for the t r i c a r b o n y l _ 1178 . i 1 8 6 species (1900 cm or 1895 cm ' ) . However, the product i s d i s -tinguished from Ru(C0) 3(PPh 3) 2 by the a b i l i t y of the dicarbonyl to r a p i d l y add H 2J^H^,PhC=CPh and 0 2 i n s o l u t i o n to give compounds of the general formula RuL(C0) 2(PPh 3) 2 where L i s the adding ligand. A d i s s o c i a t i o n of Ru(C0) 2(PPh 3) 3 i n s o l u t i o n to give Ru(C0) 2(PPh 3) 2, 187 followed by rapid addition of ligand, has been proposed. The tetracarbonylphosphine complex Ru(C0) 4(PPh 3) may be prepared by carbonylation (at 150°C and 80 atm.)of R u 3 ( C 0 ) g ( P P h 3 ) 3 , a d e r i v a t i v e 184 188 189 obtained from Ru 3(C0)^ 2 and PPh-j. ' ' A l t e r n a t i v e l y , u.v. i r -r a d i a t i o n of the v o l a t i l e ruthenium pentacarbonyl i n the presence of + 52 PPh 3 may be used, or photolysis of Ru 3(CO)^ 2 i n the presenc  of PPb 3 may be u t i l i s e d to give a 2:1 mixture of Ru(CO)^(PPh 3) and Ru(CO) 3(PPh 3) 2 The carbonylation of Ru(CO) 3(PPh 3) 2 at 100°C and 80 atmospheres pressure 191 also r e s u l t s i n quantitative conversion to the tetracarbonyl species. 192 Like Fe(CO)^(PPh 3) , which decomposes slowly over long periods i n a i r , 190 276 the tetracarbonylphosphine d e r i v a t i v e of ruthenium (0) forms a black surface deposit on standing i n the dark over a period of weeks. In addition, attempts to p u r i f y Ru(C0) 4(PPh^) by sublimation under 192 vacuum at 125 °C have resulted i n the formation of a deep red s o l i d . The v i s i b l e spectrum of t h i s s o l i d i n acetone shows a band with A =510 max nm, i d e n t i c a l with an authentic sample of [Ru 3(CO)g(PPh 3) 3] , while the i . r . spectrum i n CH^C^ shows peaks i n the carbonyl stretching region 184 a t t r i b u t a b l e to t h i s complex and p y r o l y s i s products. The l i t e r a t u r e appears to contain no references to the use of Ru(C0) 4(PPh^) i n homo-geneous c a t a l y s i s , although the complex loses CO f a r more r e a d i l y than the analogous i r o n compound and might be a useful CO source f o r carbony-l a t i o n reactions. 193 Whyman has presented i n f r a - r e d spectroscopic evidence that H 2Ru(C0) 3(PPh 3) i s i n r e v e r s i b l e equilibrium with Ru(CO)^(PPh3> and H 2 under high pressure. The osmium complex H 2Os(CO) 3(PPh 3) i s well 52 characterised with a very s i m i l a r i . r . spectrum to that of the species formed from Ru(CO)^(PPh 3) and H 2 under high pressure (eq. 8.1). In n-heptane Ru(C0) 4(PPh 3) shows peaks at 2061(s),1982(ms),1955(vs) and Ru(CO) 4(PPh 3) + H 2 ^ ~-H 2Ru(C0) 3(PPh 3) + CO (8.1) 193 1919(vw). The reaction product from Ru(CO)^(PPh 3) under H 2 at 550 atm. and 100°C shows peaks at 2089(vw),2080(s),2061(m),2021(vs),2010(s), 193 52 1987(mw),1955(s) and 1908(w) and H 20s(CO) 3(PPh 3) has peaks reported at 2079(vs),2027(vs),2018(vs),1959(w) and 1922(w). The species 193 H 2Ru(CO) 3(PPh 3) has proven d i f f i c u l t to i s o l a t e . 277 S i m i l a r l y , H 2Ru(CO) 2(PPh 3) 2 may be made from Ru(CO) 3(PPh 3) 2 52 i n THF by using 120 atm. hydrogen at 130°C , the p o t e n t i a l c a t a l y t i c u t i l i t y of t h i s species was mentioned previously (section 1.3). In 178 addition, Levison and Robinson report the f a c i l e synthesis of [H 2Ru(C0)(PPh 3) 3], by a v a r i a t i o n of the route previously discussed. CO" with i n f r a - r e d s t r e t c h i n g frequencies reported at v =1940 cm \ v =1960 and 1900 cm \ In the same work the i n f r a - r e d spectrum Ru-H of H-Ru(CO) 2(PPh 3) 2 was reported as v c o=2011 and 1974 cm 1 , v R u _ H = 1 8 7 8 and 1823 cm 1 although the pure complex could not be i s o l a t e d . 8.2. Reactions of [HRu(C0) o] with PPh 0  3 n o The a b i l i t y of the polymer [HRu(C0) 3] n to d i s s o l v e i n donor solvents such as pyridine and p i p e r i d i n e (sections 3.5 and 3.6), n a t u r a l l y suggested that phosphines might also be useful materials i n t h i s respect. In f a c t , as discussed i n section 3.6, solutions of [HRu(C0) 3]_ i n Me(Ph) 2P have provided the only "4l n.m.r. evidence to date for the presence of the co-ordinated hydride. [HRu(C0) 3] n could not be dissolved to any s i g n i f i c a n t extent i n molten triphenylphosphine, but r e f l u x i n g the polymer with t h i s phosphine i n ethanol under nitrogen did o f f e r some promising r e s u l t s . In an i n i t i a l experiment ( I ) , [HRu(C0) 3J n and PPh 3 (PPh3:Ru=3) were mixed i n ethanol and refluxed for 4 days under nitrogen. A f t e r a further 24 hours at room temperature under nitrogen the reaction f l a s k contents consisted of a mustard-yellow p r e c i p i t a t e i n a deep orange s o l u t i o n . The p r e c i p i t a t e was c o l l e c t e d by f i l t r a t i o n of the mixture under argon and 278 found to give one peak i n the carbonyl s t r e t c h i n g region of the i . r . spectrum (nujol mull) at 1905 cm Thus, the compound could be e i t h e r Ru (CO) 3 (PPh 3) 2 or Ru(C0) 2 (PPh.^. However, the r e s u l t of microanalysis on a sample of t h i s material, a f t e r thorough washing with ethanol and drying under vacuum,' was much closer to that expected for the t r i c a r b o n y l species (Calculated f or RUC 3^H 3Q0 3P 2 C,66.0; H,4.25%. Found: C,64.68;H,4.29%) than that expected for the dicarbonyl species (Calculated f o r R u C ^ H ^ O ^ : C,61.04;H,4.77%). The y i e l d of -4 material was 0.146 gms. or 2.06x10 moles based on Ru(CO) 3(PPh 3) 2, -4 from 5.0x10 monomer moles of [HRu(C0) 3] n > In a subsequent experi-ment (II),using the same conditions and the same Ru to PPh 3 r a t i o , r e f l u x i n g f o r a period of 4 days produced a deep red s o l u t i o n which only gave a yellow p r e c i p i t a t e a f t e r evaporation at room temperature under N,,. The s o l i d was obtained by f i l t r a t i o n under argon and was found to be soluble i n chloroform and acetone, but not i n heptane, ethanol or water. An examination of the i n f r a - r e d spectrum (nujol mull) of t h i s material showed peaks at 2018(s),1973(m),1940(s),1932(m),1900(m), 1096(m),745(m),705(m) and 695(m) cm" 1,which suggested that Ru(CO) 3(PPh 3) 2 was accompanied by another product i n t h i s preparation. The peaks at or near 1900,1096,745,705 and 695 (cm "*") are a l l present i n the spectrum of Ru(CO) 3(PPh 3) 2. However, the r e s u l t s of microanalysis were quite d i f f e r e n t from the expected values f o r the t r i c a r b o n y l species (Found: C,62.51;H,4.41%). The peaks at 2018,1973 and 1940 cm"1 strongly r e -sembled the patterns obtained i n t h i s region f or the dimeric p i p e r i d i n e d e r i v a t i v e s of [HRu(C0) 3] n and [Ru(C0) 2(OAc)] n {see sections 3.6 and 2.1.1} and the presence of a species such as [HRu(C0) 2(PPh 3)] 2 (Calculated: 279 194 C,57.1;H,3.80%) might explain the carbon analysis r e s u l t . Rempel has also suggested the formation of such a compound, but attempts to prove i t s existence by t h i n layer chromatography of the reaction product using chloroform as solvent and eluent on s i l i c a g e l , as well as n.m.r. of the reaction product i n CDCZ^, have given no p o s i t i v e r e s u l t s . An experiment (III) which involved a PPh^tRu r a t i o of 1:1 gave a tan-coloured s o l i d i n ethanolic s o l u t i o n a f t e r 5 days r e f l u x i n g under argon. Despite the s l i g h t d i f f e r e n c e i n colour the s o l i d had an i . r . spectrum i n n u j o l mull which was e s s e n t i a l l y i d e n t i c a l to that obtained i n experiment (II) using a three-fold excess of PPh^ over [HRu(CO) 3] n. In both experiments the y i e l d of s o l i d was close to -4 10 mgms. for every 1.0x10 monomer moles of [HRu(C0) 3] n. Since none of the i . r . spectra reported i n section 8.1 appear to coincide with that of the product from [HRu(C0) 3_ n and PPh-j i n experiment ( I I ) , the best explanation at the present time would seem to involve a compound such as [HRu(C0) 2(PPh 3)] 2 i n a mixture with Ru(CO) 3(PPh 3) 2 < Interpretation could be complicated by the s l i g h t s o l u b i l i t y of [HRu(C0) 3]_ i n ethanol alone (<0.01M). Refluxing [HRu(C0) 3] n i n ethanol under N 2 overnight was s u f f i c i e n t to generate the formation of some orange s o l u t i o n . F i l t r a t i o n and removal of solvent yielded an orange s o l i d which dissolved i n chloroform to give a s o l u t i o n with peaks at 2080(w,sh), 2068(m) and 2020(s) cm 1 i n the i n f r a - r e d spectrum. This s o l i d was not characterised. 280 Chapter 9 GENERAL CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE.WORK 9.1. General conclusions The purpose at the outset of t h i s work was to elucidate the mode of synthesis of the [HRuCCO)^]^ polymer, to characterise t h i s complex and to extend the c a t a l y t i c a p p l i c a t i o n s of t h i s and rela t e d species. The synthesis must involve reductive carbonylation processes, and these have not previously been studied i n any d e t a i l f o r ruthenium complexes. Such processes (e.g. eq. 9.1) are now of p a r t i c u l a r s i g n i f i c a n c e CO + H 20 + M 1 1- 3> C0 2 + 2H + + M 1 (9.1) i n view of t h e i r r e l a t e d chemistry to systems which catalyse the water-195 gas s h i f t - r e a c t i o n (eq. 9.2). This reaction i s of importance i n the commercial preparation of hydrogen. CO + H 20 ^ ~- C0 2 + H 2 (9.2) An i n t e r e s t i n g feature of the [HRu(C0),j] n polymer synthesis i s the requirement for the heating and cooling cycles (section 3.1) i n r 281 order to obtain improved y i e l d s . Information presented i n section 3.3 suggests that the carbonylation of ruthenium (III) i n aqueous II II solut i o n at 80°C r e s u l t s i n a mixture of Ru (00)2 and Ru (CO) 3 species, and c e r t a i n l y f o r the t r i c a r b o n y l species (Table XIX and XX) an increase i n subsequent carbonylation rate occurs on lowering the temperature from 70°C to 60°C. In addition, observations on aqueous solutions of CsRu(CO) 3C£ 3 under nitrogen (section 5.9) suggest a process of reductive carbonylation v i a co-ordinated CO that i s favoured at lower temperatures. Spectroscopic evidence f o r co-ordinated hydride i s l i m i t e d to the high f i e l d n.m.r. si g n a l (Fig. 8) detected i n methyldiphenyl-phosphine solvent. [DRu(CO)3__ was synthesised (section 3.2) but no s h i f t of v i b r a t i o n a l modes was recorded i n the i . r . spectrum. The presence of the hydride i s also indicated by the small but consistent percentage of hydrogen detected by microanalysis on many d i f f e r e n t samples of [HRu(CO) 3_ n. Examination of the reaction s o l u t i o n obtained during the pre-paration of [HRu(CO) 3] n was conducted by i . r . spectroscopy and by p r e c i p i t a t i o n of various s o l i d s (section 3.3). The r e s u l t s favour an i n t e r p r e t a t i o n i n which both Ru^^CO^ and R u 1 1 ( C O ) 3 species are present a f t e r 2 days at 80°C, and i n which Ru^^CO^ predominates a f t e r 3 days or more. Other studies (sections 5.7 and 5.9) revealed that both Cs2Ru(CO)2C£ 4 and CsRu(CO) 3C& 3 underwent reductive carbonylation quite r a p i d l y when used as s t a r t i n g species but that the addition of C£ or H + could slow these reactions or stop them completely. Hence, 282 the a c i d i t y and c h l o r i d e ion concentration (section 4.1) generated i n s o l u t i o n during the synthesis of [HRu(CO) 3] n could be responsible for the slow production of lower valent species from Ru^^CO^ and R u 1 1 (CO) ^ during the l a t e r stages of the r e a c t i o n . The other minor products of reductive carbonylation, described i n section 3.4, have defied complete c h a r a c t e r i s a t i o n . Sublimation followed by i n v e s t i g a t i o n using mass spectrometry c e r t a i n l y i ndicates that R u ^ C O ) ^ i s present, although not i n large q u a n t i t i e s (section 3.4), and examination of these orange products by i . r . spectroscopy c o n s i s t e n t l y reveals d e t a i l s that are incompatible with the known compounds l i s t e d 121 i n Table VII. The comparatively recent i d e n t i f i c a t i o n of K^RugCCO)^ 113 (Table VII) and a report of the incompletely characterised H^Ru^CO)^ and H^Ru-jtCO)^ species suggests that further new hydridocarbonyls may have been prepared i n t h i s work. [HRu(CO) 3] n dissolved i n p y r i d i n e appears to give t r i c a r b o n y l species of the form [HRu(CO)„(py) ] or [Ru(CO) (py) ] n i n s o l u t i o n (section 3.5). The discussion i n sections 3.6 and 6.3 shows that such solutions are not susceptible to carbonylation. Oxygenation of the pyridine solutions l i k e l y leads to ruthenium (III) species (section 3.5). The i n v e s t i g a t i o n of the chemistry of [HRu(CO) 3l n i n p i p e r i d i n e (section 3.6) i s relevant to the c a t a l y t i c amine carbonylation discussed i n Chapter 6. Analysis by v.p.c. has confirmed that stoichiometric carbonylation of p i p e r i d i n e occurs i n the absence of CO to y i e l d solutions presumably containing HRu(CO) 2(pip) x species (equation 3.9). A s o l i d product has been i s o l a t e d from these solutions by a method 283 s i m i l a r to that used i n the preparation of [Ru(CO)^(OAc)(pip)]^ (section 2.1.1). The product has a very s i m i l a r i . r . spectrum i n the carbonyl stretching region to that of [Ru(C0)2(OAc)(pip)]^, but analyses for a species with 3 p i p e r i d i n e s for every 2 ruthenium atoms i n the complex (species II i n equation 3.10). Chapter 4 discussed the examination of chlororuthenium (III) species as an aid i n understanding the synthesis of [HRu(C0) 3l n. Experiments with Ru(III) i n purely aqueous s t a r t i n g solutions are to 2 6 be compared with work in v o l v i n g solutions i n 1.0 to 7.0 M.HC£ i n which reductive carbonylation does not occur. Presumably, the required simultaneous co-ordination of IL^ O (or OH ) and CO (equation 4.19) i s not favoured under such conditions. The studies of Ru11"1" carbonylation (section 4.3) using i . r . spectroscopy support the conclusions reached by other means (section 3.3) regarding the presence of both Ru^^CO^ and Ru1'!"(C0)3 species i n s o l u t i o n at various stages of the r e a c t i o n . Evidence presented i n section 4.5 suggests that CO uptake by aqueous solutions of ruthenium (III) proceeds at l e a s t to some extent by reductive carbonylation to ruthenium (II) (based on pH measurements) although, as l a t e r work indicated (section 5 . 3 ) , Ru 1 1 1(CO) may not be e a s i l y detected i n aqueous s o l u t i o n at 80°C. The induction period observed i n CO uptake studies with Ru^ 1 1 species i s presumably re l a t e d to the production of a s u f f i c i e n t concentration of hydrolysed species i n s o l u t i o n . The scheme (shown i n equations 4.17 and 4.18) s a t i s f i e s the pH, [CH ] and stoichiometry of gas uptake observed when using aqueous solutions of chlororuthenium (III) species. 284 In Chapter 5 the preparation and aqueous s o l u t i o n chemistry of chlorocarbonyls of Ru(II) and Ru(III) was discussed. The i n t e r e s t i n g changes i n the v i s i b l e spectrum of Cs2Ru(C0)C£P. upon solu t i o n i n water are best explained by l o s s of chloride ion and replacement with water (Fig. 36). The gradual l o s s of i s o s b e s t i c points i n the spectrum, as time proceeds, i s consistent with aquation accompanied by i o n i s a t i o n of some co-ordinated K^O. The i n t e r p r e t a t i o n of CO uptake r e s u l t s f o r solutions of Cs2Ru(C0)C£,- was d i f f i c u l t because of the very slow rate of such reactions. Ruthenium (II) chlorocarbonyl species hydrolysed very slowly at 20°C (sections 5.5 and 5.7), Ru(CO)C£ 4(H 20) 2~ and Ru(CO) 2C£ 4~ g i v i n g the species Ru(CO) (H^O^CJL^ (equation 5.5) and Ru(C0)2 (H^O)^^ (equation 5.6) re s p e c t i v e l y . The corresponding spectroscopic changes are shown i n Figures 40 and 41. The i n t e r e s t i n g a u t o c a t a l y t i c behaviour exhibited i n CO uptake by aqueous solutions of CS2RU(CO)2^4 (Fig. 41) i s reminiscent of the dependence on [ R h 1 1 1 ] [ R h 1 ] shown i n the c a r b o n y l a t i o n 1 3 ^ ' 1 6 ^ of RhBr^ (K^O^ . Assuming that Ru° and lower-valent species would be II I p r e c i p i t a t e d as s o l i d s then a k i n e t i c dependence on [Ru ][Ru ] seems to be indicated i n t h i s case and the autocatalysis may r e s u l t from slow production of Ru 1 i n so l u t i o n . This suggestion i s supported by the i n i t i a l induction period (Fig. 42) which i s reduced as [Ru 1 1] increases and which p e r s i s t s even a f t e r p r i o r e q u i l i b r a t i o n of an aqueous s o l u t i o n of Cs2Ru(CO)2C£4 under vacuum before exposure to a CO atmosphere. The induction period i s probably caused by slow i n i t i a l formation of Ru 1 which then catalyses the reductive carbonylation of 285 Ru^CCO^. The temperature v a r i a t i o n study (Fig. 47) indicates that the i n i t i a l reduction to Ru"*" i s very temperature s e n s i t i v e . The i n t e r p r e t a t i o n of CO uptake by Cs2Ru(CO)2C£ 4 i n the presence of added chloride ion was discussed i n section 5.7. Table XXXVIII shows a serie s of reactions which explain the formation of HRu(CO) 3 from a number of Ru"*"1 and R u 1 1 1 species by processes of reductive carbonylation. With one exception (Ru 1 1(CO)^ i n the absence of soda lime) the calculated and experimentally determined stoichiometries of gas uptake are reasonably consistent, given the long reaction times and the fac t that Ru° products are not included i n the scheme. Since some R u ^ C O ) ^ c a n he extracted from reaction products (section 3.4) i t i s clear that f i n a l stoichiometries w i l l be s l i g h t l y d i f f e r e n t from those shown i n Table XXXVIII. Equation 5.21 (section 5.9) i n which R u 1 1( C O ) 3 undergoes reductive carbonylation to Ru°(CO) 4 w i l l explain the observed stoichiometries of gas uptake for the t r i c a r b o n y l species and presumably [HRu(CO) 3_ n could be formed from Ru°(CO) 4 by further reductive carbonylation. The u t i l i s a t i o n of CsRu^O^CJ^ as a c a t a l y s t f o r the carbony-l a t i o n df p i p e r i d i n e i s consistent with the previously o b s e r v e d 1 0 0 r e l a t i v e a c t i v i t i e s of [HRu(C0) o] a Ru„(CO), 0>[Ru(CO)„(OAc)] for 3 n 3 12 2 n t h i s reaction (also see Table XXXIII). Tricarbonyl species have been proposed as the a c t i v e c a t a l y s t s f o r a l l of these s y s t e m s . 1 0 0 Once again stoichiometric carbonylation was detected and a reaction scheme commencing by generation of Ru(CO).jC£(pip ) (equation 7.1) was suggested. Attempts to i d e n t i f y and i s o l a t e a carbonylamine from t h i s r e a c t i o n (section 7.2) met with d i f f i c u l t i e s and the same was true of work i n -Table XXXVIII. Gas uptake stoichiometries i n the carbonylation of R u 1 1 and Ru"*"1'1" species i n  aqueous solutions Reaction Scheme F i n a l gas uptake stoichiometry F i n a l gas uptake stoichiometry (Calculated) (Found)  (soda lime) (no soda lime) (soda lime) (no soda lime) 2RuII]:+10CCH-3H20 2HRu(CO) 3+4C0 2+6H + 2 R u m (C0)+8C0+3H20 -^>2HRu (CO) 3+4C0 2+6H + 5:1 4:1 3:1 2:1 6±0.5:1 3:1 3:1 1.5:1 00 ON 2RuI:C(CO)+7CO+3H20 -^>2HRu(CO)3+3C02+4H+ 3.5:1 2:1 not measured 1:1 2RU 1 1(CO) 2+5C0+3H 20-> 2HRu(CO)3+3C02+4H+ 2.5:1 1:1 2.2:1 1:1 2RU11(CO)3+3C0+3H20 ->2HRu(CO)3+3C02+4H+ 1.5:1 zero 2:1 1:1 R u 1 1 (CO) 3+2CO+H20 > R u ° (CO) 4+C0 2+2H + 2:1 1:1 2:1 1:1 (F i n a l stoichiometries r e f e r to moles net gas uptake per mole of ruthenium) 287 v o l v i n g [RuCCCOgCJ!^^ a n c ^ p i p e r i d i n e (section 7.5). The carbonylation of piperazine and d e r i v a t i v e s using Ru^CCO)^ as c a t a l y s t i s discussed i n section 7.6. The reaction i s rapid i n i t i a l l y (Table XXXVI) but decelerates very quickly. The evidence from v.p.c. analysis suggests that 1,4 -bis-formylpiperazine i s the predominant reaction product and the accumulation of t h i s species with two s i t e s a v a i l a b l e for co-ordination to the metal l i k e l y r e s u l t s i n the observed decrease i n reaction rate. The extreme s t r u c t u r a l v a r i a b i l i t y of 196 t r i n u c l e a r metal carbonyl molecules w i l l be a factor involved i n i s o l a t i n g p i p e r i d i n e and piperazine d e r i v a t i v e s of Ru3(CO)^2> p a r t i -c u l a r l y since piperazine may be involved as a unidentate or bidentate 197 198 ligand. ' If piperazine takes a boat conformation a d e r i v a t i v e such as Ru 3(CO)^Q(pipz.) i s j u s t one possible product. An analogous 199 compound Ru^(C0)^Q(l,2-diazine) has recently been reported and t h i s represents the f i r s t d e r i v a t i v e of t h i s s t r u c t u r a l type. 9.2. Recommendations for future work Since the polymer [HRu(CO) 3] n contains the elements of both hydrogen and carbon monoxide, i t should be a s u i t a b l e c a t a l y s t for hydrogenation and hydroformylation of o l e f i n s . However, preliminary experiments indicated that the generally low s o l u b i l i t y of [HRu(CO) 3l n _3 i n solvents such as 1-hexene (410 M.) was a major obstacle to such studies. In order to overcome t h i s problem, the formation of amine and phosphine de r i v a t i v e s of [HRu(CO) 3] n, as discussed i n Chapters 3 and 8, could be followed by an examination of the c a t a l y t i c p o t e n t i a l 288 of such d e r i v a t i v e s (e.g. H^R^ (CO)^(pip)^) . The benzene-soluble side products obtained i n the [HRu(CO) 3] n synthesis seem to consist of at l e a s t one new polynuclear ruthenium carbonyl and further e f f o r t s to i d e n t i f y these products should prove i n t e r e s t i n g . The use of polynuclear metal complexes immobilised on s o l i d supports may represent the ultimate i n homogeneous c a t a l y s i s ^ 0 0 201 and Ugo has pointed out the relevance of such complexes to the understanding of heterogeneous c a t a l y t i c processes. The carbonylation of C ^ R u t C O ^ C ^ i n aqueous sol u t i o n provided some useful information i n support of the r e a c t i o n schemes shown i n Table XXXVIII for production of [HRu(C0) 3_ n; auto c a t a l y s i s i n v o l v i n g Ru^ or Ru° was c e r t a i n l y indicated, however k i n e t i c i n t e r p r e t a t i o n s were hindered by the p r e c i p i t a t i o n of s o l i d s during the course of the reaction. This problem was not encountered when Cs2Ru(C0)2C& 4 was dissolved i n solutions containing excess c h l o r i d e , since reductive carbonylation did not then take place, and t h i s system should give k i n e t i c s that are easier to i n t e r p r e t . The use of CsRu(CO) 3C£ 3 as a c a t a l y s t for the carbonylation of p i p e r i d i n e gave good y i e l d s of N-formyl p i p e r i d i n e (based on gas-uptake studies) and t h i s system seems worthy of further study. In p a r t i c u l a r , the i s o l a t i o n and c h a r a c t e r i s a t i o n of chlorocarbamoyl species from p i p e r i d i n e solutions (as attempted i n section 7.3) should a s s i s t i n i d e n t i f y i n g more p r e c i s e l y the a c t i v e species involved i n the CsRu(CO) 3C& 3 catalysed carbonylation system. 289 REFERENCES 1. I. Wender and P. Pino, Eds., "Organic Synthesis v i a Metal Carbonyls", Vol. I, Interscience, New York, 1968. 2. J. Falbe, "Carbon Monoxide i n Organic Synthesis", Springer-Verlag, New York, 1970. 3. C. Bird, " T r a n s i t i o n Metal Intermediates i n Organic Synthesis", Acad. Press, London, 1967. 4. H.E. Podall , J. Chem. Ed., 38, 187 (1961). 5. W. Manchot and W.J. Manchot, Z. Anorg. Allgem. Chem., 226, 385 (1936). 6. E.R. Corey and L.F. Dahl, J. Amer. Chem. S o c , 83, 2203 (1961). 7. F.A. Cotton and G. Wilkinson, "Advanced Inorganic Chemistry", 3rd ed., Interscience, New York, 1972, p.684. 8. S. Ahrland, J . Chatt and N.R. Davies, Q. Rev. (Chem. S o c , London), 12, 265 (1965). 9. T.A. Manuel, Adv. Organomet. Chem., _3, 191 (1965). 10. D.W. Smith, J. Chem. Soc. (Dalton), 834 (1976). 11. S.C. T r i p a t h i , S.C. Srivastava, R.P. Mani and A.K. Shrimal, Inorg. Chim. Acta, 15_, 249 (1975) and references therein. 12. F. Calderazzo, R. E r c o l i and G. Natta i n "Organic Synthesis v i a Metal Carbonyls", I. Wender and P. Pino, Eds., Vol. I, Interscience, New York, 1968, p.175. 13. F. Basolo and R. Pearson, "Mechanisms of Inorganic Reactions", 2nd ed., John Wiley and Sons Inc., New York, 1967, p.576. 14. K.C. Dewhurst, U.S. Pat. 3,454,644 from Chem. Abstr., 71_, 83317 (1969). 290 15. L. Vaska, "Proc. 8th Intern. Conf. Co-ord. Chem.", V. Gutmann, Ed., Springer-Verlag, New York, 1964, p.99. 16. L. Vaska, Proc. 1st Intern. Conf. Organomet. Chem., Madison, Wisconsin, 1965, p.79. 17. R.A. Schunn, Inorg. Chem., _9, 2567 (1970). 18. G.W. P a r s h a l l , Acc. Chem. Res., 8_, 113 (1975). 19. B.R. James, L.D. Markham and D.K.W. Wang, Chem. Comm. (London), 439 (1974). 20. B.R. James, "Homogeneous Hydrogenation", John Wiley and Sons Inc., New York, 1973, r e f . 1424, p.94. 21. B. Hui and B.R. James, Chem. Comm. (London), 198 (1969). 22. B. Hui, Ph.D. d i s s e r t a t i o n , U n i v e r s i t y of B r i t i s h Columbia (1969). 23. B. Hui and B.R. James, Can. J. Chem., 48, 3613 (1970). 24. B.R. James, Inorg. Chim. Acta. Rev., 4_, 73 (1970). 25. J.A. Osborn, F.H. Jardine, J.F. Young and G. Wilkinson, J. Chem. Soc. (A), 1711 (1966). 26. J. Halpern, B.R. James and A.L.W. Kemp, J. Amer. Chem. S o c , 88, 5142 (1966). 27. D. Fahey, J. Org. Chem., 38, 80 (1973). 28. D. Fahey i n "Cat a l y s i s i n Organic Syntheses", P.N. Rylander and H. Greenfield, Eds., Acad. Press, 1976, p.287. 29. P.S. Hallman, B.R. McGarvey and G. Wilkinson, J. Chem. Soc. (A), 3143 (1968). 30. D. Rose, J.D. G i l b e r t , R.P. Richardson and G. Wilkinson, J. Chem. Soc. (A), 2610 (1969). 291 31. =D. Fahey, J . Org. Chem., 38, 3343 (1973). 32. B.R. James and L.D. Markham, Inorg. Nucl. Chem. L e t t . , ]_, 373 (1971). 33. F.L'Eplattenier, P. Matthys and F. Calderazzo, Inorg. Chem., % 342 (1970). 34. M. Dekker and G. Knox, Chem. Comm. (London), 1243 (1967). 35. J.A.J. J a r v i s , B.E. Job, B.T. Kilbourn, R.H.B. Mais, P.G. Owston and P.F. Todd, Chem. Comm. (London), 1149 (1967). 36. J. Piron, P. P i r e t and M. Van Meersche, B u l l . Soc. Chim., Beiges, 76, 505 (1967). 37. F.L'Eplattenier and F. Calderazzo, Inorg. Chem., 6, 2092 (1967). 38. M.I. Bruce and F.G.A. Stone, Angew. Chem. (Intern, ed.), 80, 460 (1968). 39. P. Chini , Inorg. Chim. Acta Rev., _2, 31 (1968). 40. B.F.G. Johnson and J.A. McCleverty, Prog. Inorg. Chem., ]_, 277 (1966). 41. J.P. Collman, N.W. Hoffman and D.E. Morris, J. Amer. Chem. S o c , 91, 5659 (1969). 42. J.P. Candlin and W.H. Janes, J . Chem. Soc. (C), 1856 (1968). 43. J. Norton, D. Valentine and J.P. Collman, J. Amer. Chem. S o c , 91, 7537 (1969). 44. B.F.G. Johnson, R.D. Johnston, J. Lewis, B.H. Robinson and G. Wilkinson, J. Chem. Soc. (A), 2856 (1968). 45. B.F.G. Johnson, J. Lewis and I.G. Williams, J . Chem. Soc. (A), 901 (1970). 292 46. J. Knight and M.J. Mays, J. Chem. Soc. (A), 711 (1970). 47. G.R. Crooks, B.F.G. Johnson, J. Lewis, I.G. Williams and G. Gamlen, J. Chem. Soc. (A), 2761 (1969). 48. P. Pino, G. Braca, F. Piacenti, G. Sbrana, M. Bianchi and E. Benedetti, "New Aspects Chem. Met. Carbonyl Derivs.", 1st Intern. Symp., Venice, 1968, Abstract E2. 49. P. Smith and H. Jaeger, Ger. Pat. 1, 159, 926 from Chem. Abstr., 60, 14389 (1964). 50. G.F. Cox and G.H. Whitfield, Brit. Pat. 999, 461 from Chem. Abstr. , 63, 9811 (1965). 51. D. Evans, J.A. Osborn and G. Wilkinson, J. Chem. Soc. (A), 3133 (1968). 52. F.L'Eplattenier and F. Calderazzo, Inorg. Chem., 1_, 1290 (1968). 53. D. Valentine and J.P. Collman, unpublished results quoted in ref. 54. 54. J.P. Collman, Acc. Chem. Res., 1, 136 (1968). 55. W. Strohmeier, J. Organomet. Chem., 94, 273 (1975). 56. L.D. Markham, Ph.D. dissertation, University of British Columbia (1973). 57. M.J. Lawrenseon and M. Green, Ger. Pat. 2,026,926 from Chem. Abstr. 74, 124, 822 (1971). 58. For example, Brit. Pat. 966,482(1964); Ger. Pat. 1,159,826 (1963); U.S. Pat. 3,239,566(1966) and U.S. Pat. 3,239,570(1966). 59. T. Anderson and R.V. Lindsey Jr., U.S. Pat. 3,081,357 from Chem. Abstr. 59, 8594 (1963). 293 60. G. Braca, G. Sbrana, F. P i a c e n t i and P. Pino, Chim. e Ind. (Milan), 52, 1091 (1970). 61. P. Pino, F. P i a c e n t i , M. Bianchi and L. Lazzaroni, Chim. e Ind. (Milan), 50, 106 (1968). 62. A.J. Chalk and J.F. Harrod, Adv. Organomet. Chem., 6_, 119 (1968). 63. P. Pino, G. Braca, G. Sbrana and A. Cuccini, Chem. and Ind. (London), 1732 (1968). 64. P. Pino, F. P i a c e n t i and M. Bianchi, B r i t . Pat. 1,167,687 from Chem. Abstr.., 72, P36330h (1970). 65. F. P i a c e n t i , M. Bianchi and E. Benedetti, Chem. Comm. (London), 775 (1967). 66. R.E. Benson, U.S. Pat. 2,871,262 from Chem. Abstr. 5J3, 14008 (1959). 67. T.J. Kealy and R.E. Benson, J. Org. Chem. 26, 3126 (1961). 68. H. P i c h l e r and B. Firnhaber, Brenstoff-Chem., 44, 33 (1963) from Chem. Abstr. 59., 4037b (1963). 69. R.S. Coffey, Chem. Comm. (London), 923 (1967). 70. A. Dobson and S.D. Robinson, Inorg. Chem., 16, 137 (1977). 71. A. Misono, Y. Uchida, M. Hidai and I. Inomota, Chem. Comm. (London), 704 (1968). 72. B.W. Graham, K.R. Laing, C.J. O'Connor and W.R. Roper, J. Chem. Soc. (Dalton), 1237 (1972). 73. K.R. Laing and W.R. Roper, Chem. Comm. (London), 1556 (1968). 74. W.B. Hughes, Advances i n Chemistry, Vol. 132, American Chemical Society, Washington, D.C., 1974, p.202. 75. S. Searles, M. Tamres, F. Block and L.A. Quarterman, J . Amer. Chem. S o c , 78^ , 4917 (1956). 76. M.J. A s t l e , " I n d u s t r i a l Organic Nitrogen Compounds", Reinhold 294 Pub. Corp., New York, 1961, p.221. 77. H. Sternberg, I. Wender, R. F r i e d e l and M. Orchin, J. Amer. Chem. Soc., 75, 3148 (1953). 78. M. Farlow and H. Adkins, J. Amer. Chem. S o c , 57, 2222 (1975). 79. T. Saegusa, Y. Ito, S. Kobayashi, K. Hirota and H. Yoshioka, Tet. L e t t e r s , 6121 (1966). 80. T. Saegusa, S. Kobayashi, K. Hirota and Y. Ito, B u l l . Chem. So c Jap., 42, 2610 (1969). 81. J. T s u j i and N. Iwamoto, Chem. Comm. (London), 380 (1966). 82. F. Calderazzo, Inorg. Chem., _4, 293 (1965). 83. "Gmelins Handbuch der Anorganischen Chemie", System Nummer 60, Verlag ChemieGMBH, Weinheim, p.240. 84. W. Brackman, Disc. Faraday S o c , 46, 122 (1968). 85. T. Tsuda, Y. Isegawa and T. Saegusa, J. Org. Chem., 37_, 2670 (1972). 86. D. Durand and C. Lassau, Tet. L e t t e r s , 28, 2329 (1969). 87. W.F. Edgell, M.T. Yang, B.J. Bulkin, R. Bayer and N. Korzum, J. Amer. Chem. S o c , 87_, 3080 (1965). 88. W.F. Edgell and B.J. Bulkin, J. Amer. Chem. S o c , 88, 4839 (1966). 89. A.D. A l l e n , T. Eliades, R.O. Harris and P. Reinsalu, Can. J. Chem., 47, 1605 (1969). 90. W. Hieber and H. Heusinger, J. Inorg. Nucl. Chem., 4-, 179 (1957). 91. R.J. A n g e l i c i , Acc. Chem. Res., _5, 335 (1972). 92. E.O. Fischer and H.J. Kollmeier, Angew. Chem. (Intern, ed.), j), 309 (1970). 93. S. Fukuoka, M. Ryang and S. Tsutsui, J . Org. Chem., 33, 2973 (1968). 295 94. R.J. A n g e l i c i and R.W. Brink, Inorg. Chem., 12, 1067 (1973). 95. E. McC. Arnett, J.G. M i l l e r and A.R. Day, J. Amer. Chem. Soc., 72, 5635 (1950). 96. R.J. A n g e l i c i and L.J. Blacik, Inorg. Chem., 11, 1754 (1972). 97. A.E. Kruse and R.J. A n g e l i c i , J. Organomet. Chem., 2A_, 231 (1970) . 98. B.R. James and G.L. Rempel, Chem. and Ind. (London), 1036 (1971) . 99. J . J . Byerley, G.L. Rempel, N. Takebe and B.R. James, Chem. Comm. (London), 1482 (1971). 100. G.L. Rempel, W.K. Teo, B.R. James and D.V. Plackett, Advances i n Chemistry, V ol. 132, American Chemical Society, Washington, D.C., 1974, p.166. 101. J.F. Harrod, S. Cicconi and J. Halpern, Can. J. Chem., _39_, 1372 (1961). 102. B.R. James and J. Louie, Inorg. Chim. Acta, _3, 568 (1969). 103. J. Louie, M.Sc. Thesis, U n i v e r s i t y of B r i t i s h Columbia (1968). 104. A. Mantovani and S. Cehini i n "Inorganic Syntheses", Vol. XVI, McGraw-Hill, New York, 1976, p.47. 105. M.J. Cleare and W.P. G r i f f i t h , J. Chem. Soc. (A), 372 (1969). 106. B.W. Horrom, M. F r e i f e l d e r and G.R. Stone, J. Amer. Chem. S o c , 77, 753 (1955). 107. P. Borda, personal communication. 108. B.I. Swanson, J . J . Rafalko, D.F. Shriver, J . San F i l i p p o J r . and T.G. Spiro, Inorg. Chem., 14, 1737 (1975). 296 109. S.A.R. Knox and H.D. Kaesz, J . Amer. Chem. Soc., 93, 4594 (1971). 110. S.A.R. Knox, J.W. Koepke, M.A. Andrews and H.D. Kaesz, J. Amer. Chem. S o c , 91_, 3942 (1975). 111. H.D. Kaesz, personal communication. 112. H.D. Beckey and H.-R. Schulten, Angew. Chem. (English ed.), 14, 403 (1975). 113. H. P i c h l e r , H. Meier, W. Gobler, R. Gaertner and D. Kioussis, Brenstoff-Chem. , 48_, 266 (1967). 114. "Handbook of Chemistry and Physics", 51st ed., Chemical Rubber Co., Cleveland, Ohio, 1970, pp. B307-308. 115. R.A. Schunn i n "Systematics of Tr a n s i t i o n Metal Hydride Chemistry", E.L. Muetterties, Ed., Marcell Dekker, New York, 1971, p.243. 116. W.E. B u l l , S.K. Madan and J.E. W i l l i s , Inorg. Chem., 2, 303 (1963). 117. For example, "The A l d r i c h L i b r a r y of Infrared Spectra", 2nd ed., C.J. Pouchert, A l d r i c h Chem. Co. Inc., Milwaukee, Wisconsin, 1975. 118. B.R. James and G.L. Rempel, personal communication. 119. W.L. Masterton and E.J. Slowinski, "Chemical P r i n c i p l e s " , W.B. Saunders Co., Philade l p h i a , 1969, p.406. 120. H.D. Kaesz, S.A.R. Knox, J.W. Koepke and R.B. S a i l l a n t , Chem. Comm. (London), 477 (1971). 121. M.R. C h u r c h i l l , J. Wormald, J . Knight and M.J. Mays, Chem. Comm. (London), 458 (1970). 122. N.S. G i l l , R.H. N u t a l l , D.E. Scaife and D.W.A.. Sharp, J. Inorg. Nucl. Chem., 18, 79 (1961). 123. R.A. Barnes, "Properties and Reactions of Pyridine and i t s 297 Hydrogenated Derivatives", Chapter 1 i n "Pyridine and Derivatives", Part One, E. Klingsberg, Ed., Interscience, New York, 1961, p.11. 124. Q. Mingoia and P.C. F e r r i e r a , Anais fac. farm, e odontol, Univ. Sao Paulo, 11, 9 (1953) from Chem. Abstr., 49, 7566a (1955). 125. E.N. Shaw, "Pyridine N-Oxides", Chapter IV i n "Pyridine and Derivatives", Part Two, E. Klingsberg, Ed., Interscience, New York, 1961, p.119. 126. D.M. Adams, "Metal-Ligand and Related Vi b r a t i o n s " , Edward-Arnold, London, 1967, p.246. 127. S.I. Shupack and M. Orchin, Inorg. Chem., 2> 374 (1964). 128. M.E. Vol'pin, Pure Appl. Chem., 30, 607 (1972). 129. J. T s u j i , "Organic Synthesis by Means of T r a n s i t i o n Metal Complexes", Springer-Verlag, New York, 1975, p.140. 130. L. Vaska, Acc. Chem. Res. % 175 (1976). 131. T.A. Stephenson and G. Wilkinson, J . Inorg. Nucl. Chem., 28, 945 (1966). 132. A. Asatoor and C E . Dal g l e i s h , J. Chem. S o c , 1717 (1958). 133. R.S. Drago and K.F. P u r c e l l i n "Non-Aqueous Solvent Systems", T.C. Waddington, Ed., Acad. Press, New York, 1965, p.240. 134. R.A. Head and J.F. Nixon, Chem. Comm. (London), 135 (1975) and references therein. 135. E. Sappa and L. Milone, J . Organomet. Chem., ^1_, 383 (1973). 136. C C . Yin and A.J. Deeming, J. Chem. Soc. (Dalton), 1013 (1974). 137. P.S. Braterman, "Metal Carbonyl Spectra", Acad. Press, London, 1975, p.240. 138. B.R. James and CN. Rosenberg, Can. J . Chem., 54, 313 (1976). 298 139. D. Forster and G.R. Beck, Chem. Comm. (London), 1072 (1971). 140. N. von Kutepow, W. Himmele and H. Hohenschutz, Chem. Ingr-Tech., 37, 383 (1965). 141. E.E. Mercer and R.R. Buckley, Inorg. Chem., 12, 1692 (1965). 142. D. Fine, Ph.D. Thesis, U n i v e r s i t y of C a l i f o r n i a Lawrence Radiation Laboratory, Report UCRL-9059, Feb. 2, 1960, p.57. 143. B.R. James and R.S. McMillan, Inorg. Nucl. Chem. L e t t . , 11, 837 (1975). 144. C.K. Jorgensen, Acta. Chem. Scand., 10, 518 (1956). 145. C.K. Jorgensen, Moi. Phys., _2, 309 (1959). 146. J.R. Ferraro, "Low Frequency Vibrations of Inorganic and Co-ordination Compounds", Plenum Press, New York, 1971, p.111. 147. D. Rose and G. Wilkinson, J. Chem. Soc. (A), 1791 (1970). 148. E.E. Mercer and P.E. Dumas, Inorg. Chem., 10, 2755 (1971). 149. A.L.W. Kemp, Ph.D. Thesis, U n i v e r s i t y of Chicago, I l l i n o i s , U.S.A. (1964). 150. M.L. Berch and A. Davison, J. Inorg. Nucl. Chem. , _3_5, 3763 (1973). 151. R. Colton and R.H. Farthing, Aust. J. Chem., 24, 903 (1971). 152. D.A. Palmer and G.M. Har r i s , Inorg. Chem., 14, 1316 (1975). 153. W. Robb and G.M. Harris, J . Amer. Chem. S o c , 87_, 4472 (1965). 154. I.A. Poulsen and C.S. Garner, J. Amer. Chem. S o c , 84_, 2032 (1962). 155. M.G. Adamson, J . Chem. Soc. (A), 1370 (1968). 156. M.G. Adamson and R.E. Connick, Abstract No. 009, D i v i s i o n of 299 Inorganic Chemistry, 150th Meeting, American Chemical Society, A t l a n t i c C i t y , Sept. 13-17, 1965. 157. F. Basolo and R.G. Pearson, "Mechanisms of Inorganic Reactions", John Wiley and Sons Inc., 2nd ed., New York, 1967, p.67. 158. S.K. Shukla, Journal of Chromatography, 8_, 96 (1962). 159. M. Sneed and J . Maynard, "General Inorganic Chemistry", Van Nostrand, New York, 1942, p.813. 160. R.E. Connick i n "Advances i n the Chemistry of the Co-ordination Compounds", S. Kirschner, Ed., MacMillan, New York, 1961, p.15. 161. E.A. Guggenheim, P h i l . Mag., _2, 538 (1926). 162. W.M. Latimer, "The Oxidation States of the Elements and t h e i r P o t e n t i a l s i n Aqueous Solution", 2nd ed., New York, Prentice-H a l l , 1952, p.228. 163. T.D. Avtokratova, " A n a l y t i c a l Chemistry of Ruthenium", Ann Arbor-Humphrey Science Pub. Inc., 1969, p.107. 164. B.R. James and G.L. Rempel, J . Chem. Soc. (A), 79 (1969). 165. J . Chatt, L.A. Duncanson and L.M. Venanzi, J . Chem. S o c , 4456 (1955). 166. L.E. Orgel, J. Inorg. Nucl. Chem., 2 , 137 (1956). 167. G.N. Rosenberg, Ph.D. D i s s e r t a t i o n , U n i v e r s i t y of B r i t i s h Columbia (1974). 168. D. Forster, Inorg. Chem., 8_, 2556 (1969). 169. J.V. Rund, F. Basolo and R.G. Pearson, Inorg. Chem., 3^, 658 (1964). 170. B.R. James and G.L. Rempel, Can. J. Chem., 44_, 233 (1966). 300 171. D.M. Adams, "Metal-Llgand and Related Vibrations", Edward-Arnold, London, 1967, p.99. 172. E. Benedetti, G. Braca, G. Sbrana, F. S a l v e t t i and B. Gra s s i , J. Organomet. Chem., 37_, 361 (1972). 173. M. Bianchi, E. Benedetti and F. P i a c e n t i , Chim. e Ind. (Milan), 51, 613 (1969). 174. B.F.G. Johnson, R.D. Johnston and J. Lewis, J. Chem. Soc. (A), 792 (1969). 175. R.G. Pearson, Inorg. Chem., 12, 712 (1973). 176. T.A. Manuel, J. Org. Chem. , 27_, 3941 (1962). 177. A.A. Frost and R.G. Pearson, " K i n e t i c s and Mechanism", Wiley, New York, 1963, p.137. 178. J . J . Levison and S.D. Robinson, J. Chem. Soc. (A), 2947 (1970). 179. D.D. Pe r r i n , " D i s s o c i a t i o n Constants of Organic Bases i n Aqueous Solution", Butterworths, London, 1965, p.132. 180. O.C. Dermer and G.E. Ham, "Ethyleneimine and other A z i r i d i n e s " , Acad. Press, New York, 1969, p. 205. 181. A.A. Burg and CD. Good, J. Inorg. Nucl. Chem., 2_, 237 (1956). 182. M.I. Bruce and F.G.A. Stone, J . Chem. Soc. (A), 1238 (1967). 183. S. Cenini, M. P i z z o t t i , F. Porta and G. La Monica, J . Organomet. Chem., 125, 95 (1977). 184. M.I. Bruce, G. Shaw and F.G.A. Stone, J. Chem. Soc. (Dalton), 2094 (1972). 185. F. P i a c e n t i , M. Bianchi, E. Benedetti and G. Sbrana, J. Inorg. Nucl. Chem., 2_9, 1389 (1967). 301 186. J . Collman and W.R. Roper, J . Amer. Chem. S o c , 87, 4008 (1965). 187. B.E. C a v i t t , K.R. Grundy and W.R. Roper, Chem. Comm. (London), 60 (1972). 188. J.P. Candlin and A.C. Shortland, J . Organomet. Chem., 16, 289 (1969). ' 189. M.I. Bruce and F.G.A. Stone, J . Chem. Soc. (Dalton), 2094 (1972). 190. B.F.G. Johnson, J . Lewis and M.V. Twigg, J . Organomet. Chem., 67, C75 (1974). 191. B.F.G. Johnson, J . Lewis and M.V. Twigg, J . Chem. Soc. (Dalton), 1876 (1975). 192. A.E. C l i f f o r d and A.K. Mukherjie, Inorganic Synthesis, 8_, 185 (1966). 193. R. Whyman, J . Organomet. Chem., 56, 339 (1973). 194. G.L. Rempel, personal communication. 195. R.M. Laine, R.G. Rinker and P.C. Ford, J . Amer. Chem. S o c , 99_, 252 (1977). 196. F.A. Cotton, Prog. Inorg. Chem., 21, 1 (1976). 197. P.J. Hendra and D.B. Powell, J . Chem. S o c , 5105 (1960). 198. D. Fowles and D.K. Jenkins, Inorg. Chem., _3> 2 5 7 (1964). 199. F.A. Cotton and J.D. Jamerson, J . Amer. Chem. S o c , 98_, 5396 (1976). 200. A.L. Robinson, Science, 194, 1261 (1976). 201. R. Ugo, Cat. Rev.-Sci. Eng., 11, 225 (1975). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059302/manifest

Comment

Related Items