UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Some reactions of 1,2,3-trithia-[3]-ferrocenophanes Talaba, Ana 1991

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

Item Metadata


831-UBC_1991_A6_7 T34.pdf [ 4.38MB ]
JSON: 831-1.0059744.json
JSON-LD: 831-1.0059744-ld.json
RDF/XML (Pretty): 831-1.0059744-rdf.xml
RDF/JSON: 831-1.0059744-rdf.json
Turtle: 831-1.0059744-turtle.txt
N-Triples: 831-1.0059744-rdf-ntriples.txt
Original Record: 831-1.0059744-source.json
Full Text

Full Text

SOME REACTIONS OF l,2,3-TRITHIA-[3]-FERROCENOPHANES by ANA TALAB A B.Sc, THE POLYTECHNIC INSTITUTE OF IASSY, ROMANIA  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CHEMISTRY  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA JUNE 1991 © Ana Talaba, 1991  In presenting  this thesis in partial fulfilment of the requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may department or by  his or her  representatives.  be granted by the head of  It is understood that copying or  publication of this thesis for financial gain shall not be allowed without my permission.  Department of The University of British Columbia Vancouver, Canada Date  DE-6 (2/88)  my  written  ABSTRACT The ability of l,2,3-trithia-[3]-ferrocenophane and (N,N-dimethylaminomethyl)l,2,3-trithia-[3]-ferrocenophane to function as ligands towards transition metals of groups 7 and 8 was explored. Thermal reactions of l,2,3-rjithia-[3]-ferrocenophane with ruthenium and osmium carbonyls produced 18-electron ferrocenedithiolato ruthenium and osmium carbonyl complexes isolated in low (5-10%) or moderate yields (25-35%). The ruthenium and osmium complexes, Fe(r) -C5H4S)2Ru2(CO)6 and 5  Fe(T] 5  C5H4S)2(M-3-S)(p:-CO)20s4(CO)9, were characterized by X-ray crystallographic analysis. The crystal structure of the ruthenium complex reveals a very distorted octahedral coordination about the ruthenium atoms and a Ru-Ru separation of 2.6812(7) A which is comparable to the lengths found in other ruthenium(I) dimers. The structure of the osmium cluster compound reveals a spiked triangular configuration of osmium atoms with a sulfido ligand triply bridging the osmium triangle and the C5H4S ligands doubly bridging the spike Os-Os bond. Both structures have eclipsed and nearly parallel cyclopentadienyl rings. Reactions of l,2,3-trithia-[3]-ferrocenophane  with cobalt and manganese  carbonyls produced unidentified insoluble solids, in high yields. Two low yield soluble cobalt derivatives Fe(r| -C5H4S)2Co2(CO)6 and Fe(rj -C5H4)2Co4(CO)i2, and a 5  5  paramagnetic manganese derivative were also isolated from these reactions. Reactions of (N,N-dimemylammomethyl)-l,2,3-trithia-[3]-ferrocenophane with ruthenium and iron carbonyls, and some rhodium and palladium complexes afforded oils and solids which were not identified. These products are insoluble in common organic solvents and are produced in yields varying from 20 to 90 %. Microanalytical data indicate that the products are not analytically pure compounds. A soluble iron derivative isolated in  ii  trace quantity, was identified from the fragments exhibited by the low-resolution spectrum as rFe(C5H4S2)(C5H CH2NMe2)]2Fe2(CO)6. 3  TABLE OF CONTENTS  Abstract  ii  List of Figures  vii  List of Tables  viii  List of Schemes  ix  List of Abbreviations  x  Acknowledgements  xii  I. Introduction 1.1. An Introduction to Ferrocenophanes. A Brief Historical Development.  1  1.1.1. [l]-Ferrocenophanes.  3  1.1.2. [2]-Ferrocenophanes.  4  1.1.3. [3]-Ferrocenophanes.  5  1.1.4. [4]-Ferrocenophanes and Larger Systems.  8  1.2. Derivatives of [3]-Ferrocenophanes.  10  1.3. Structure and Bonding.  17  1.4. Applications of Ferrocenophanes.  23  1.5. Scope of the Work Presented in this Thesis.  25  n. Experimental. 2.1. General Information.  26  2.2. Preparation of l,2,3-Trithia-[3]-Ferrocenophane.  27  A. The Isolation of the Adduct of l,l'-Dilithioferrocene with N,N,N',N'-Tetramethylethylenediamine.  27  B. The Reaction of the l.l'-Dilithioferrocene /TMEDA Adduct with Sulphur.  27 iv  2.3. Reactions of l,2,3-Trithia-[3]-Ferrocenophane. 2.3.1. The Reaction with Ru (CO) . 3  28 27  1 2  A. In Hexanes.  28  B In Cyclohexanes.  29  2.3.2. The Reaction with Os (CO) . 3  30  12  A. In Hexanes.  30  B. In Cyclohexane.  30  C. In Toluene.  31  2.3.3. The Reaction with Co (CO)g.  32  2  A. In Hexanes.  32  B. In THF.  33  2.3.4. The Reaction with Mn (CO) . 2  34  1 2  2.4. Preparations of (N,N-Dimethylaminomethyl)- and Diphenylphosphino)-l,2,3-Trithia-[3]-Ferrocenophanes.  35  2.4.1. (N,N-Dimethylarninomethyl)-1,2,3-Trithia-[3]-Ferrocenophane.  35  2.4.2. (Diphenylphosphino)-l,2,3-Trithia-[3]-Ferrocenophane.  36  A. Preparation of (Diphenylphosphino)ferrocene.  36  B. The Lithiation of (Diphenylphosphino)ferrocene and Reaction of the Lithiated Product with Sulphur.  36  2.5. Reactions of the (N,N-EHmethylaminomethyl)-l,2,3-Trithia- [3]Ferrocenophane.  37  2.5.1. The Reaction with Ru (CO) . 3  38  12  A. In Hexanes.  38  B. In cyclohexane.  38  C. In THF.  39  2.5.2. The Reaction with Fe (CO) . 2  39  9  2.5.3. The Reaction with [Rh(COD)Cl] . 2  40 v  2.5.4. The Reaction with [Rh(COE) Cl] .  41  2.5.5. The Reaction with Pd(COD)Cl .  41  2  2  2  i n . Results and Discussion. 3.1. Preparation of l,2,3-Trithia-[3]-Ferrocenophane.  42  3.2. Reactions of l,2,3-Trithia-[3]-Ferrocenophane with Metal Carbonyls.  43  3.2.1 The Reaction with Ru (CO) . 3  12  43  3.2.2. Spectroscopic Properties and Molecular Structure of the Ferrocenedithiolato Derivative of Ruthenium Carbonyl. 3.2.3. The Reaction with Os (CO) . 3  12  44 50  3.2.4. Ferrocenedithiolato Derivatives of Osmium Carbonyl. Spectroscopic Properties.  53  3.2.5. Molecular Structure of (l,l'-Ferrocenedithiolato-S,S') Sulfido-Tetraosmium Carbonyl.  55  3.2.6. Cobalt and Manganese Derivatives of 1,2,3-Trithia- [3]-Ferrocenophane.  61  3.3. Preparations of (N,N-Dimethylaminomethyl)- and Diphenylphosphino)-l,2,3-Trithia-[3]-Ferrocenophanes.  64  A. (N,N-Dimemylammomemyl)-l,2,3-Trithia-[3]-Ferrocenophane.  64  B. Diphenylphosphino-Sulfido-l,2,3-Trithia-[3]-Ferrocenophane.  65  3.4. Reactions of (N,N-Dimethylaminomethyl)-l,2,3-Trithia-[3]Ferrocenophane.  67  Summary.  70  Future Studies.  73  Bibliography.  74  vi  LIST OF FIGURES  Figures  1.1.  Page  Bridge reversal of [3]-ferrocenophanes and inversion of six-membered rings.  7  1.2.  Distortions in ring-tilted ferrocenophanes.  18  1.3.  An ORTEP view of l,23-trithia-[3]-ferrocenophane.  19  1.4.  A perspective view of l,2-sulphur-3-selena-[3]-ferrocenophane.  21  1.5.  A perspective view of 26.  21  1.6.  A perspective view of 31b.  22  3.1.  An ORTEP view of 43.  45  3.2.  The proposed structures of ferrocenedithiolato derivatives of osmium carbonyl 44-48.  52  3.3.  The geometry of sulfido ligand in sulfido-osmium carbonyl clusters.  55  3.4.  An ORTEP view of 47.  57  vii  LIST OF TABLES  Tables  Page  3.1.  Crystal Data and Experimental Details for 43.  3.2.  Selected Bond Lengths (A) with Estimated Standard Deviations for 43.  3.3.  46  47  Selected Bond Angles (deg) with Estimated Standards Deviations for 43.  48  3.4.  Selected Crystal Data and Experimental Details for 47.  58  3.5.  Selected Bond Lengths (A) Estimated Standards Deviations for 47.  3.6.  Selected Bond Angles (deg) with Estimated Standards Deviations for 47.  3.7.  3.8.  59  60  Reactions of 1,2,3-Trithia-[3]-Ferrocenophane with Cobalt and Manganese Carbonyls.  63  Reactions of DMAMTF.  68  viii  LIST OF SCHEMES  Schemes  1.1.  Page  Preparation of (1.1 '-ferrocenediy^diphenylsilane and bis(l.r-ferrocenediyl)silane.  3  1.2.  Preparation of l,l,2,2-tetramethyl-[2]-ferrocenophane.  5  1.3.  Preparation of tetramethylene-[4]-ferrocenophane.  8  1.4.  Oxidative addition reaction of transition metal species.  12  1.5.  Synthesis of (l.l'-ferrocenedithiolato-S.S , Fe) 1  (triphenylphosphine)Pd(n). 1.6.  Synthesis of (1,1 '-metallocenedichalcogenato-) monoor bisphosphines complexes of Pt (II), Pd (II) and Ni (II).  1.7.  13  15  Synthesis of cationic complexes of 1,l'-bis[(alkyl- or phenyl) chalcogeno] ferrocenes.  16  3.1.  Preparation of 1,2,3-trithia- [3] -ferrocenophane.  42  3.2.  Synthesis of (l,l'-ferrocened^miolato-S,SOdiruthenium hexacarbonyl.  3.3.  43  Synthesis of (l,l'-ferrocene(nthiolato-S,S')sulfido-diosmium hexacarbonyl.  51  3.4.  Preparative route for DMAMTF.  65  3.5.  Preparative route for DPPTF.  66  LIST O F ABBREVIATIONS  Anal.  -Analysis  br.  -broad (IR)  Calcd.  -calculated  CD2CI2  -dichloromethane-d2  CD3CI  -monochloromethane-d3  cm"  -wavenumbers  COD  -cyclooctadiene  COE  -cyclooctene  deg  -degrees  DMAMTF  -dimemylanunomethyl-Mmia-ferrocenophane  DMSO  -dimethylsulfoxide  DPPF  -diphenylphosphinoferrocene  DPPTF  -(uphenylphospMno-trithia-ferrocenophane  EI  -Electron Ionization  FAB  -Fast Atom Bombardment  fdma  -ferrocenedimethylarsine  fdpa  -ferrocenediphenylphosphine  1  -proton h  -hours  IR  -infrared  R7PAC  -International Union of Pure and Applied Chemistry  m  -multiplet (NMR)  min.  -minutes x  mol  -mole  m/z  -mass-to-charge ratio  NMR  -nuclear magnetic resonance  p.d.  -pseudo doublet (NMR)  p. t.  -pseudo triplet (NMR)  RT  -room temperature  s  -singlet (NMR)  s  -strong (IR)  THF  -tetrahydrofuran  TMEDA  -N,N,N\N'-terramethyleutylene<uanune  TLC  -thin layer chromatography  w  -weak  8  -chemical shift  xi  ACKNOWLEDGEMENTS  I would like to express my sincere thanks to my research supervisor, Professor W.R. Cullen, for his expert advice, encouragement, and patience. I wish to thank the members of my research group, both past and present, for all their help. Special thanks go to Tu-cai Zheng and Paul Wood for their invaluable advice and assistance, and to Angela Tsang for her humor and friendship. I also wish to thank the technical staff of this Department for their assistance and useful discussions. Finally, but certainly not least, I thank Mihai, my husband, and Theo and Codrin, my sons, for their support and help with housekeeping during the course of this work.  xii  I INTRODUCTION  1.1. AN INTRODUCTION  TO FERROCENOPHANES.  A BRIEF HISTORICAL  DEVELOPMENT.  Ferrocenophanes represent special cases of heteroanular disubstitution of ferrocenes in which both rings are linked from position n to n' (n=l-5) by one or more 1to m-membered bridging groups (bridges). There exist two main classes of 1  ferrocenophanes: the mononuclear complexes with one (1) or several (2) bridging groups and the multinuclear systems such as 3 - 5 in which the rings in two or more ferrocene nuclei are linked by two or more bridges. ' Heteroatomic bridges are found in a large 2 3  number of ferrocenophanes.  1  2  Ferrocenophanes in which the cyclopentadienyl rings are linked by one bridging group 1 are known as [m]-ferrocenophanes, where m denotes the numbers of atoms in the bridge. Although not provided in the IUPAC rules, the ferrocenophanes nomenclature 4  2 5-7  proposed by Smith has been adopted. '  According to this nomenclature the bridge  lengths are indicated by numbers enclosed in square brackets while the relative location of the bridges is given by numbers in round brackets. This notation appears to be the choice  1  of the majority of researchers in this area, although there are many examples in which the IUPAC rules are applied and ferrocenophanes are referred to as ferrocenes. For example 2 is unambiguously named according to the IUPAC rules as 1,1'; 2,2'; 4,4' rris(trimemylene)ferrocene while its name accenting to the ferrocenophane nomenclature is [3] (1,1 ')[3] (2,2')[3] (4,4')ferrocenophane.  5  Although some ferrocenophanes have been known for close to thirty years, this class of compounds has been the subject of intensive study only in the last fifteen years.  2  Most of the present literature is still concerned with the syntheses and structural characterization of these compounds. However, there are investigations which include detailed studies of the fluxional behavior of the bridge, the possible interaction between a transition metal atom and the iron of the ferrocene moiety, comparisons between [m]ferrocenophanes and other [m]-metallocenophanes, the effects of the bridging chains on the possible distortion of the cyclopentadienyl rings from the preferred, parallel and eclipsed orientation.  1.1.1. [l]-Ferrocenophanes.  The first [l]-ferrocenophanes 6 (l,r-ferrocenediyl)diphenylsilane and bis(l,l'ferrocenediyl)silane, were prepared in 1975 by the reaction of l,r-dilithioferrocene with diphenyldichlorosilane and silicon tetrachloride respectively (Scheme 1.1.). Since then,  6b  Scheme 1.1.  3  other group 14 bridged ferrocenophanes have been prepared. " Such silicon-bridged [1]ferrocenophanes have been found to be considerably more reactive than the usual 9  organosilicon compounds with four Si-C bonds. It seemed probable that heteroatom groups other than R2Si and R2Ge could be introduced as bridges into [l]-ferrocenophanes, and the preparation of some [1]10 12 13 ferrocenophanes with group 15 atom bridges, i.e. RP and RAs, was undertaken. ' ' These ferrocenophanes react at the bridging atom with sulphur and with reactive metal carbonyl species without disruption of the ferrocenophane system while with organolithium reagents the reaction proceeds via nucleophilic attack at the bridging atom 12 13 with opening of the ferrocenophane system. ' This is a very effective method of activating the second ring for further substitution and was used in the preparation of the 14 unsymmetrically disubstituted ferrocenes. 1.1.2.  [2]-Ferrocenophanes. The first [2]-ferrocenophane, l,l,2,2-tetramethyl[2]-ferrocenophane  9 was  prepared by reacting 6,6-dimethylfulvene 7 with sodium to give the disodium salt 8 followed by addition of FeCl2. This reaction (Scheme 1.2.), first reported by Rienehart et al.  15  is typical of the approach by which many [m]-metallocenophanes have been  prepared. [2]-Ferrocenophanes linked by a 2-carbon bridge are accessible only by direct synthesis and cannot be prepared from ferrocene itself, presumably due to the gross distortion from parallel planes enforced upon the ferrocene ring by the C2 bridge. The 2  rings in 9 are tilted with respect to each other, a structure that corresponds to the one of protonated ferrocene.  4  7  8  9  Scheme 1.2.  1.1.3.  [3]-Ferrocenophanes. [3]-Ferrocenophanes are by far the best known of the [m]-ferrocenophanes, partly  due to the fact that these compounds are accessible by several routes. Carbon-bridged [3]ferrocenophanes were among the first derivatives of the parent substance to be prepared. Of the several general methods available for the synthesis of such substances the internal Friedel-Craft acylation of a suitably chosen ferrocenylalkanecarboxylic acid was initially employed for the preparation of a-keto-l.T-trimethyleneferrocene 10. The method appeared to be confined to the synthesis of derivatives bridged by a three carbon unit since a-ferrocenylbutyric acid was reported to give products of homoanular cyclization.  16  Elaboration of this procedure resulted in the preparation of the first doubly and triply 17 bridged ferrocenophanes  and of several new carbon-bridged ferrocenophanes such as  11 and 12. The C3-bridge causes little distortion in the system and although it is too short to span the distance between the cyclopentadienyl ligands in ferrocene without any strain, the construction of multibridged systems proved possible and [3]- ferrocenophanes with one, two three or four bridges have been prepared. 19  20  5  12  12a R=H; 12bR=C2H ; 12cR=CH 5  3  In spite of the increasing number of reports on the preparation and properties of ferrocenophanes, reports concerned with [3]-ferrocenophanes with bridging atoms other 21  22  than carbon are relatively few in number. Davison and co workers • synthesized 1,2,3trithia-[3]-ferrocenophane as the major product from the reaction of l.T-dilithioferrocene with sulphur. The range of [3]-ferrocenophanes with group 16 elements in a bridging position was extended by including selenium and tellurium as part of the bridge, thus making available a range of compounds containing a heterotrichalcogen chain which is an 23  uncommon structural grouping.  A series of [3]-ferrocenophanes with the symmetrical  trichalcogen chains -Se3-, -SSeS-, -STeS-, -SeSSe-, -SeTeSe-, as bridging groups has been synthesized. 6  The crystal structure of the -SSeS-bridged [3]-ferrocenophane gives information about the conformation of the cyclopentadienyl rings in relation to the replacement of the 24  central S by Se. The fluxional behavior of [3]-ferrocenophanes has been the subject of considerable 18 23 25 29  interest.  ' ' "  Variable temperature NMR studies of [3]- ferrocenophanes with  symmetrical trichalcogen chains as bridging groups indicated that the low temperature limiting spectra arise from an ABCD spin system, while the high temperature limiting spectra have a pair of unsymmetrical triplets. These spectral properties are consistent with the [3]-ferrocenophanes undergoing a bridge-reversal fluxional motion (Fig. 1.1.).  Fig. 1.1. Bridge reversal of [3]-ferrocenophanes and inversion of six-membered rings.  Detailed dynamic NMR studies of the fluxional process in these molecules indicated the order of activation energies for these motion: -S3- > -SSeS- > -SeSSe- > -Se3- >-STeS-> -SeTeSeDynamic NMR studies also indicated that the bridge reversal fluxion of [3]ferrocenophanes whilst appearing very analogous to the chair - to -chair reversal of 7  corresponding 6-membered heterocyclic rings, is a process of a much higher energy. Compounds such 21 - 23 obtained via l,l'-disubstituted  30-31  ferrocenes  (C5H4L)2Fe, where the substituents are potential donor groups (L=PR.2, AsR.2, S" or SH) are also viewed as [3]-ferrocenophanes and are referred to in Section 1.2.  1.1.4. [4]-Ferrocenophanes and Larger Systems.  The [4]-ferrocenophane with a C4-bridge 14 prepared by Luttringhaus et al. in 32  1958  , was the first example of a bridged ferrocene. The reaction of disodium salt 13  with ferric and ferrous chloride was used in the synthesis of 14 as shown in Scheme 1.3. In an analogous manner a C5-bridged-[5]-ferrocenophane was prepared.  32  13  1  4  Scheme 1.3.  [m]-Ferrocenophanes with extremely long carbon bridges have been prepared by a 33  Swedish group,  but during the past decade 'crown' ferrocenophanes have been of  greater interest for many research groups. The preparation and physico-chemical properties of several kinds of polyoxa-[n]-ferrocenophanes 15, " polythia-[n]38 39 40 ferrocenophanes 16, l,n-dithia-[n]-ferrocenophanes 17, ' polyoxathia-[n]ferrocenophanes 18 " and polyoxadithia-[n]-.ferrocenophanes 1 9 have been 34  41  43  37  44  reported.  8  Polyoxa- and polythia-[n]-ferrocenophanes are crown ether-like compounds that include a ferrocene subunit as a ring member. It seems that the interest in 'crown' ferrocenophanes arose from the steadily increasing interest in 'crown' ethers. The latter are macrocyclic polyethers containing the -O-CH2-CH2-O- group, that differ in conformation and interatomic distances. These factors together with the number of electron donating atoms and dimension of the macrocycle determine the complexation selectivity towards different cations. It was expected that considerable changes in the 45  geometry and the arrangement of the O atoms could occur when a ferrocene subunit was introduced into the macrocycle, hence the change in coordination ability of the new crown ether-like compounds preoccupied the research group mentioned above. " ' " 34  38  41  44  Crownlike ferrocenophanes have also been studied with respect to the possible interaction between the iron atom and a complexed metal ion; however, detailed stractural  9  O  Fe  <3>-s ^  n  19 mformation is, so far, very limited. Much effort has been made by several groups " to synthesize intramolecularly 48  53  multibridged ferrocenophanes wherein two cyclopentadienyl rings of ferrocene are linked with polymethylene chains. Several tribridged ferrocenophanes have been synthesized ' 48  50  prior to 1978 but terra- and penta-bridged ferrocenophanes remained  unknown. The preparation of four-bridged ([44](l,2,3,4)ferrocenophane) and fullybridged ([45](l,2,3,4,5)ferrocenophane) ferrocenophanes with tetramethylene chains have been reported by Hisatome et al. in 1990. 54  The multinuclear ferrocenophanes such as 3 - 5 are beyond the scope of this introduction.  1.2. DERIVATIVES OF  [3]-FERROCENOPHANES.  It was first reported by Bishop and Davison that ferrocene derivatives with donor atoms at l,l'-position can coordinate to some transition metals through donor atoms. The l.l'-disubstituted ferrocenes (CsrLjL^Fe, (where the substituents are potential donor groups L=PR.2, AsR.2), function as ligands towards the metals of group 6 and 8 in the formation of the coordination complexes (fdma)M(CO)4 and (fdpa)M(CO)4 (M=Mo, Cr, W) and (fdma)MX 21 2  56  20  55  (M=Pd, Pt; X=C1, Br, I). These complexes can be  viewed as [3]-ferrocenophanes. They contain the donor atoms in the 1- and 3-positions and the metal in the 2-position of the bridge.  10  After this discovery many such complexes were prepared. The isolation of 1,2,3tnthia-[3]-ferrocenophane and its quantitative conversion to ferrocene-l,l'-dithiol  22  provided a potential bidentate chelating dithiolate. Thus, a series of l,3-dithia-[3]ferrocenophanes of group 13, 14, 15 and 16 elements 22 have been readily prepared from ferrocene l,l'-dithiol and the appropriate halides REX2 or R2EX2 (R=CH3, G2H5, n - C H , Ph) 4  9  2 6  '  5 7  -  6 0  22  23  E=BR, CR2, GeR , SnR , PR, AsR, Se, Te 2  2  The l,3-dithia-2E-[3]-ferrocenophanes have been characterized analytically and spectroscopically. ' 26  60,61  Variable temperature NMR studies of these compounds  26 indicated that they are also fluxional molecules.  11  Although many l,3-dithia-[3]-ferrocenophanes containing carbon, silicon, germanium, selenium and tellurium have been prepared, no complexes of this ligand system were known with transition metals before 1983. The chemistry of l,2,3-trithia-[3]-ferrocenophane functioning as ligand towards a transition metal was first explored by Seyferth and Hames.  They reported that the  complex 23 can be prepared by the reaction of l,2,3-trithia-[3]-ferrocenophane with Fe3(CO)i2 in refluxing hexane. 63 Another research group  sought 23 from the reaction of ferrocene and  Fe3(CO)i2 with elemental sulphur in refluxing benzene. No ferrocenyl-sulphur hexacarbonyl diiron complexes were produced but some (p>S )Fe (CO)6 and (jj.32  2  S)2Fe3(CO)9 and a mixture of sulphur- containing ferrocenes resulted. The insertion of low-valent metal species into the S-S bonds of acyclic and cyclic disulfides (Scheme 1.4.) is a process that has been known since 1964. " Such 64  70  reactions of low-valent, coordinatively unsaturated species of type L 2 M (M=Ni, Pd, Pt; L=tertiary phosphine) with organic disulfides is a characteristic oxidative addition reaction of transition metal species. " 71  L2M  +  74  RS - SR  •  LJvl  Scheme 1.4.  l,2,3-trithia-[3]-ferrocenophane 24 was considered as substrate for such an  75 insertion reaction by Seyferth et al.  in  attempting the synthesis of (1,1'-  ferrocenedithiolato-S,S')bis (triphenylphosphine)palladium(II) 25 (Scheme 1.5). They obtained,  however,  a new  type  of  complex,  (l,l'-ferrocenedithiolato12  S,S\Fe)(triphenylphosphine)palladium(II) 26, instead of the expected complex 25. The *H NMR spectrum and X-ray analysis of 26 suggest the presence of a dative bond between the iron atom of the ferrocene nucleus and the palladium atom located at the 2position of the bridge.  26 Scheme  1.5.  The possible coordination of the nonbonding d electrons of the iron atom in the ferrocenophane nucleus to the vacant orbitals of another metal atom was described for the first time by Watanabe et al.  in [2](l,r)ferrocenophane-metal halide. The large Q.S.  value of the [2]-ferrocenophane.3HgCl2 adduct compared with that of ferrocene.7HgCl2 and that of [2]-ferrocenophane itself was taken as evidence for a strong interaction between Fe-Hg atoms in the ferrocenophane adduct. However, Butler et al.,  Whiteside et al.  and McCulloch et al.  reported no  evidence for any interaction between the iron atom and the metal atoms in compounds 27,  13  28a and 28b respectively.  27  27: M=Pd, X=S(i-Bu), R=C1 28a: M=Pt, X=PPh , R=n-Bu 2  28b: M=Pd, Ni; X=PPh , R=C1 or Br 2  79 80  Akabori et al.  '  reported on a systematic synthetic study of (1,1'-  metallocenedichalcogenato-X,X,M)(triphenylphosphine)platinum(II) and palladium(II) ,  complexes (X=S or Se; M=Fe, Ru or Ni) by using a modified procedure of that used by 75  Seyferth et al.  (Scheme 1.6). They determined that the reaction of 1,2,3 - trichalcogena-  [3](l,r)metallocenophane  (M=Fe  or  Ru,  X=S  or  Se)  with  tetrakis(triphenylphosphine)platinum(0) gave a monophosphine complex 31 or a bis(phosphine) complex 32 depending on the reaction temperature and the solvent used. The X-ray analysis of complex 31b suggests the presence of a dative bond from the iron atom of the ferrocene moiety to the platinum atom located at the 2-position of the bridge. The complex 31b is isostructural with the mono(phosphine)palladium complex 26 prepared by Seyferth et al. Akabory et al. determined that the kind of chalcogen atom present influenced the complex formation in the ferrocene series, while no effect was observed in the ruthenocene series. The fact that the ferrocene complex 31e and 3If could not be obtained could be explained by the following: the C-Se and M-Se (M=Pd or Pt) bonds are 14  XH M  y M*— pph  3  31 29  THF M'(PPh  3)  29'  M  y  \  s M  /PPh,  M*  X-/  ^PPhs  30 32 Scheme 1.6.  29&30a: M=Fe, X=S  29'a: M'=Pd  31&32a: M=Fe, M'=Pd, X=S  29&30b: M=Ru, X=S  29'b: M'=Pt  31&32b: M=Fe, M'=Pt, X=S  29&30c: M=Fe, X=Se  29'c: M'=Ni  31&32c: M=Ru, M'=Pd, X=S  29&30d: M=Ru, X=Se  31&32d: M=Ru, M*=Pt, X=S 31&32e: M=Fe, M'=Pd, X=Se 31&32f: M=Fe, M'=Pt, X=Se 31&32g: M=Ru, M'=Pd, X=Se 31&32h: M=Ru, M'=Pt, X=Se 31&32i: M=Fe, M'=Ni, X=S 31&32J: M=Ru, M'=Ni, X=S  15  a little longer than those of the C-S and M-S (M=Pd or Pt) bonds, respectively, because of the covalent radii of Se and S atoms, which are 1.17 A and 1.02 A  8 1  As a result, the Fe-  M (M=Pd or Pt) distance will become too long to provide a stable complex. in the Ru complexes 31g and 31h the lengthening of the Ru-M (M=Pd or Pt) distance resulting from the long Se-C and Se-M distances would be compensated by the 79  expanded 4d orbital of a Ru atom compared with the 3d orbital of an Fe atom. The ferrocene complexes 31i and 32i could not be obtained. The fact that the mono(phosphine) complex of ferrocene could not be isolated indicates that the ruthenium atom of the metallocene nucleus plays an important role in the mono(phosphine) complex formation in the nickel series. l,l'-bis[(alkyl- or phenyl)chalcogeno]ferrocenes form 1:1 cationic complexes 33 with (MeCN)4Pd(BF )2 and (MeCN) Pt(BF )2 (Scheme 1.7). The spectral data indicate 4  4  4  dative Fe-Pd and Fe-Pt bonds in these complexes. '  82 83  Trithia- and tetrathia-[n]-ferrocenophanes also form 1:1 cationic complexes 34 and 35 with (MeCN) Pd(BF )2, and a Fe-Pd bonding interaction is likely. 4  83  4  R SR Fe  + (MeCN) M(BF ) 4  4  2  + p h P  Fe  M  PPh  3  3  33 Scheme  1.7.  33a:R=Me, 33b: R=i-Pr, 33c: R=i-Bu, 33d: R=n-Bu, 33e: R=Ph M=Pt or Pd 16  1.3. STRUCTURE AND BONDING.  As it has been shown in Section 1.1. there are two basic structural types of ferrocenophanes: 1) mononuclear complexes in which the two rings of a ferrocene unit are connected by a bridge or by several bridging groups such 1 and 2, and 2) multinuclear complexes in which two or more ferrocene units are brought together into one molecule by bridging groups, 3-5. The ligands in ferrocene are planar and lie in parallel planes. According to the 85 87  early, but also later crystal structures the rings are staggered. In ferrocenes and some of their simple derivatives the cyclopentadienyl ligands lie in strictly parallel planes and can rotate easily about the symmetry axis. This is no longer true when the two ligands are joined by a bridge: the dihedral angle between the planes of the rings will depend to a large degree on the length of the bridge(s), and, any rotation will be restricted. The inter-ring separation in ferrocene is 3.32 A  8 5  and thus bridges which are  shorter than this would be expected to cause molecular distortion by tilting the planar cyclopentadienyl rings. Compounds containing bridges of less than four atoms are expected to show ring tilting. Strain in [m]-ferrocenophanes is also observed in bond 17  angle distortion at the bridging atoms and deviation of the exocyclic bonds from the planes of the cyclopentadienyl rings. 89 98 From ten molecular structures of [2]- and [3]-ferrocenophanes  the following  observations can be made: a) in all the compounds the cyclopentadienyl rings retain their regular pentagonal symmetry and planarity; b) there is a linear relationship between a (the dihedral angle) and B (Fig. 1.2.); the mean C-C bond length in the cyclopentadienyl rings is approximately constant (average value 1.42 A) and is not related to a; d) the mean Fe-C (1,1') distance decreases with increasing a ; e) the mean Fe-C (3,4,3',4') distance decreases slightly with increasing a; f) the mean Fe-Cs ring distance is almost constant at about 1.64 A.  B = (B  1+  B )/2 2  Fig. 1.2. Distortions in ring-tilted ferrocenophanes.  On the basis of these observations and assuming an eclipsed symmetry for the ferrocene part of the molecule, [l]-ferrocenophane structures were predicted to be much more distorted than any [2]-ferrocenophane, with the distortion increasing as C(l)99 102 bridging atom bond length decreases.  Furthermore, if [l]-ferrocenophanes followed  the structural trends of [2]- and [3]-ferrocenophanes large ring tilts of between 36° to 39° 18  would be expected. However, the observed ring tilt (19.2 ) in the [l]-ferrocenophane,  89  (l,r-ferrocenediyl)diphenylsilane was much smaller.  o The value of 26.7 , found in  (l,r-ferrocenediyl)phenylphosphine, is the largest inclination of the rings yet reported for  88 a [m]-ferrocenophane. The l,2,3-trithia-[3]-ferrocenophane was found to consist of two nearly parallel eclipsed cyclopentadienyl rings linked by a trisulfide chain (Fig. 1.3.). While retaining 90  parallel Cs-rings (the dihedral angle is 2.85°) and the same Fe-C distances as in ferrocene, the two rings have slipped with respect to one another. At the same time, the C-S and S-S distances and the C-S-S and S-S-S angles are close to those values normally found in simple organo-sulphur compounds. ' 103  104  The cyclopentadienyl rings are coplanar with  the sulphur atoms in the 1 and 3 positions. S3  Fig. 1.3. An O R T E P view of l,2,3-trithia-[3]-ferrocenophane.  An increase in size of the bridging atoms might lead to tilting of the rings away from the bridge, or otherwise alter the molecular conformation. The crystal structure of l,3-dithia-2-selena-[3]-ferrocenophane (Fig.1.4.) shows that the changes in molecular 26  19  geometry and conformation brought about by substitution of S by Se at the bridgehead are quite small. The expansion of the bridgehead bonds from S-S 2.049 A  9 0  to S-Se 2.195 A  introduces additional strain into the molecule, and this is relieved in two ways. Firstly, there is a reduction in the valence angle at the bridgehead , from 103.9° for S-S-S to 100.5° for S-Se-S, and, secondly the two sulphur atoms which are coplanar with the cyclopentadienyl rings planes in the trithiane, are each displaced outwards from these two planes by 0.04 A. The two C-S-Se angles do not differ significantly from the C-S-S angles reported for the trithiane. As a result, the mutual disposition of the two ring planes in the two compounds is hardly changed, with the rings retaining an eclipsed conformation. Molecular structures of several 2-metalla-[3]-ferrocenophanes 26 and 31 show that in these heterobinuclear Fe-M (M=Pt, Pd or Ni) complexes the metal centers are held together by the cyclopentadienethiolato groups (SC5H4), and what appears to be a dative F e - M bond. Both SC5H4 ligands are ri -bound to iron, as in ferrocene and its derivatives 5  and are o bound to M through the sulphur atoms. The coordination about Pd in 26 is a slightly distorted square plane in which the sulphur atoms are mutually trans as are the PPI13 group and the Fe atom of the ferrocenyl moiety (Fig. 1.5.). Both Pd-S distances are at the short end of the range observed in 75  some mono- and bis(thiolato)palladium(II) complexes. ' 105  106  Similarly, the Pd-P distance 107 108  is one of the shortest observed in typical Pd-PPh3 complexes.  '  These short Pd-S  and Pd-P distances are possibly a consequence of the coordinative unsaturation at Pd that results in these electron donating ligands being tightly bound to the electron deficient metal center. The Fe-Pd distance is significantly longer than those observed in clusters containing Fe-Pd single bonds, but might correspond to a weak dative Fe~Pd bond. This bond is necessary to give Pd a favorable 16-electron configuration; without it a very 75  unsaturated and reactive 14-electron configuration would result.  20  Fig. 1.4. A perspective view of l,2-sulphur-3-seIena-[3] ferrocenophane.  Fig. 1.5. A perspective view of 26.  In 31b (Fig. 1.6.) the ferrocene moiety is very similar to that of other ferrocenophanes. The bridge chain is held in the position at which the distance between Fe 79  and Pt can be the shortest and the Pt(H) atom has a distorted tetrahedral configuration. The Fe-Pt distance in 31b is a little longer than the Fe-Pd distance in 26, which suggests that the Fe-Pt bond in 31b is weaker than the Fe-Pd bond in 26, and the unsaturation of the Pt atom in 31b may be increased compared with that of Pd atom in 26. This suggestion is supported by the decreased bond distances (Pt-S and Pt-P) around the Pt atom.  Fig. 1.6. A perspective view of 31b.  The Cp rings take an eclipsed conformation to each other in [2]-metalla-[3]ferrocenophanes. The rings are slightly tilted in (fdma)2MX2 21 and in the Pd and Pt complexes 27 and 28, the dihedral angle being almost 0 ° and 1.9°. The dihedral angle between the rings increases to 19.6° in 26 and 21.0° in 31. Still, this is a relatively small 75 79  tilt compared with that of dicyclopentadienyl complexes Cp2MLn. * From the little X-ray structural data available for 'crown' ferrocenophanes it 22  appears that the two cyclopentadienyl rings exhibit conformations that are close to eclipsed with mean twist angles varying between 0.6° to 7.8°. Long bridging chains appear to favor attainment of the eclipsed conformation. The rings are eclipsed and nearly parallel, with dihedral anglesfrom0.3° to 3.4° and it appears not unexpectedly that longer bridging chains allow for closer approach to a parallel arrangement. '  46 47  1.4. APPLICATIONS OF FERROCENOPHANES.  The class of ferrocenophanes embraces a great variety of compounds with interesting structural features and many research groups see in them a fascinating source of potentially useful materials. [l]-Ferrocenophanes have been used in the preparation of low and high molecular weight polymers such as 36 - 39 which are air and thermally stable to temperatures up to 350°.  1 3 a  '  1 3 b  These oligomers and polymers have the ability to coordinate to transition  metal complexes through the tertiary phosphine units in their backbone. Ph P -  Ph CI  OH  P -  Fe  Fe  Li  H-i x 36  —I  x  37  [m]-Ferrocenophanes (m=3, 7 or 9) have been used in the preparation of heterobimetallic complexes that are expected to exhibit high catalytic activity for selective 80 83 cross-coupling, selective hydroformulation and asymmetric hydrogenation reactions. '  23  Similar heterobimetallic complexes obtained from ferrocene derivatives with potential 109-113  donor atoms at the 1,1'-positions have proved useful in such catalytic reactions Ph  O Ph  P -  OH  Fe  Fe  H  -i  Ph  H —i  38  _ n  x  39  Sulphur - bridged ferrocenophanes may be used as starting materials for transition metal-sulfide clusters, whose chemistry is expanding at a rapid rate commensurate with the structural, spectroscopic and chemical novelty of these compounds. " Recognized 114  118  goals in this field include the synthesis of models for important biological  118  and  119  industrial catalysts and the preparation of materials for energy storage and . 120 conversion. Crown ferrocenophanes have been steadily gaining interest in the past decade due to their novel structural features and properties. They have been used as ligands to form complexes with alkali-metal cations, and the interaction between the iron atom of ferrocene nucleus and the complexed cation have been of particularly interest. '  34 37  The few uses mentioned above indicate that the chemistry of ferrocenophanes is not only interesting and instructive, but also useful. Yet, only the first steps have been taken in this direction and many opportunities await to be explored.  1.5. SCOPE OF THE WORK PRESENTED IN THIS THESIS.  Reports of the preparation, structural characterization and properties of 24  ferrocenophanes are now very numerous, but these have been concerned very largely with carbon atom bridged species. While in the past decade the interest of various research groups has been focused towards 'crown' ferrocenophanes little has been done with ferrocenophanes with sulphur atoms in the bridge. The scope of the present investigation is to study the reactivity of S3-bridged ferrocenophanes towards metal carbonyls and complexes of the group 7 and 8 metals. l,2,3-trithia-[3]-ferrocenophane  and  (dimethylaminomethyl)l,2,3-trithia-[3]-  ferrocenophane were employed as ligands and particular emphasis was placed upon the use of ruthenium and osmium carbonyls as substrates. The interest in employing sulphur-bridged ferrocenophanes in the synthesis of transition metal-sulphur complexes was stimulated by two factors: the observation that catalytic activity of metal carbonyls is influenced, sometimes for the better, by sulphur impurities,  114  and the broad development of the structural chemistry of the polynuclear  metal-sulphur complexes. ' 117  118  25  II EXPERIMENTAL  2.1. GENERAL INFORMATION  Reactions were performed in a nitrogen atmosphere by using conventional Schlenk procedures. Freshly distilled and purified solvents were used in the reactions. Diethyl 122  ether, hexanes, cyclohexane and toluene were dried and deoxygenated by distillation from sodium benzophenone ketyl under nitrogen. Tetrahydrofuran was pre-dried by refluxing over CaH2 and distilled from sodium benzophenone ketyl under nitrogen. Commercially available (Aldrich) n-butyllithium, N,N'- dimethylaminomethylferrocene  and  chlorodiphenyl-phosphine were used. Ruthenium carbonyl was prepared by carbonylation of methanol solutions of  RUCI3.X H 2 O  at 125° C and 55 - 60° atm., and recycling the 123  mother - liquors with fresh  RUCI3.X  H 0. 2  Commercially available osmium,  manganese, cobalt and iron carbonyls were used. Column chromatography was carried out by using silica gel, 230 - 400 mesh. Preparative TLC separations were carried out by using commercial silica gel coated glass plates (0.5 mm thickness). Reagent grade solvents were used for routine column chromatography. Product mixtures were monitored routinely throughout all experimental procedures by using TLC on aluminium-backed silica gel coated plates (0.2 mm thickness). H NMR spectra were recorded by using Brucker WH200, Varian XL-300 and  l  Brucker WH400 spectrometers. All chemical shifts are reported in ppm and quoted relative to tetramethylsilane (TMS) as an external standard. Low resulution mass spectra, Electron Ionisation (EI) and Fast Atom Bombardment (FAB), were recorded on a Kratos MS 50 and an AEIMS-9 spectrometers by the staff of the Mass Spectrometry Laboratory of this department. The probe 26  temperature in EI was 120-150° C. Infrared spectra were determined for solution samples by using a Perkin-Elmer 598 infrared spectrophotometer. X - ray structural determinations were performed by Mr. Steve Rettig, and elemental analyses were performed by Mr. Peter Borda of this department. Melting points were measured on a Mettler apparatus and are uncorrected.  2.2. PREPARATION OF l,2,3-TRITHIA-[3]-FERROCENOPHANE  (A) The isolation of the adduct of 1 ,l'-dilithioferrocene with  NJNJN'JN'-  22  tetramethylethylenediamine. Ferrocene (5 g, 26.8 rrimdl) was placed In a 250 ml Schlenk tube. After the tube was evacuated and refilled with N , hexane (30 ml), and a solution of a 1.6 M n2  butyllithium in hexane (37 ml, 58.96 mmol, 10% excess) were added. The suspension was rapidly stirred and freshly distilled N,N,N\N'-tetramethylethylenediamine (TMEDA) (4.5 ml, 29.9 mmol, 12% excess) was added dropwise. The reaction was exothermic and the ferrocene slurry reacted to give a deep cheery red solution. The reaction mixture was allowed to stir for 5 h during which time a fine orange precipitate deposited. This was filtered through a medium Schlenk filter and washed with warm, dry oxygen-free hexane (2 x 30 ml). Drying in vacuo (12h) yielded a fine orange powder. This compound can be stored for long periods under N at room temperature. 2  22 (B) The reaction of 1 ,l'-dilithioferrocenelTMEDA adduct with sulphur. Dry oxygen-free 1,2-dimethoxyethane (75 ml) was added to the 1,1'dilithioferrocene.TMEDA adduct and the clear, deep orange red solution was obtained. The addition of sulphur (2.31 g, 72.3 mmol) to the rapidly stirred solution caused it to warm and darken. The Schlenk tube containing the reaction mixture was equipped with a 27  reflux condenser and the mixture was refluxed for 12 h, cooled and then filtered through a bed of "celite". The filtrate was treated with diethyl ether (75 ml) and washed with 10% aqueous sodium hydroxide (5 x 30 ml). The combined aqueous fractions were washed with diethyl ether (3 x 100 ml). The combined ether fractions were dried (MgSO-4) and the solvent removed by vacuum distillation to give a dark semi-solid. Chromatographic examination, TLC, showed two major products and traces of three other compounds. The analytically pure ferrocenophane was separated by column chromatography on silica with a petroleum ether/dichloromethane (80/20) mixture as eluent. Removal of solvent afforded the l,2,3-trithia-[3]-ferrocenophane as an orange crystalline mass (yield 45 %). The other major band eluted down the column afforded diferrocenyl sulfide as an yellow crystalline mass (yield ~20 %).  Fe(C H S) S: H NMR l  5  4  2  (CD3CI): 8 4.53  (m, 2), 4.43 (m, 2), 4.37 (m, 2), 3.84  (m, 2). EI mass spectrum, selected (relative intensity): m/z 280 (100, M+), 248 (8.20), 216 (26.3), 184 (58.4). Anal. Calcd.: C, 42.86; H, 2.88. Found: C, 42.72; H, 2.94.  [Fe(C H5)(C5H4)2]2S: ^ NMR (CD3CI): 8 4.30 (s, 8), 4.16 (s, 10). EI mass 5  spectrum, selected (relative intensity): m/z 402 (100, M ) , 370 (4.7), 306 (39.1). +  2.3. REACTIONS OF l,2,3-TRITHIA-[3]-FERROCENOPHANE.  2.3.1. The reaction with Ru3(CO)i2-  (A) In hexanes. A 100 ml three-necked flask equipped with a magnetic stir-bar and a reflux condenser topped with a nitrogen inlet tube was charged with 0.550 g (1.96 mmol) of 28  Fe(C5H4)S3, 2.500 g, (3.92 mmol) of Ru3(CO)i2 and 50 ml hexane. The resulting suspension was stirred, heated at reflux under nitrogen for 18 h, then allowed to cool to room temperature, and filtered by using a Buchner funnel. The filtrate was evaporated at reduced pressure to give a brown residue. The solid product remaining in the Buchner funnel was extracted with dichloromethane (30 ml) and the extract was combined with the brown residue. Part of the solvent was removed under vacuum and the remaining solution was transferred onto a silica chromatographic column. The reaction product and Ru3(CO)i2 (~ 60 % total) were isolated as a mixture with petroleum ether as eluent. Compound separation was achieved by preparative TLC (petroleum ether or petroleum ether/dichloromethane, 80/20 , as eluent). Orange/red crystals were obtained after solvent removal and crystallization from hexane/dichloromethane (25/75) (yield 35 %).  Fe(C5H4S)2Ru2(CO)6: *H NMR (CD3CI): 8 4.34 (p.t., 4), 4.03 (p.t., 4). IR (CH2C12): DCO (cm" ) 2090 (s), 2060 (s), 2040 (w), 2015 (s, br). EI mass spectrum, 1  selected (relative intensity): m/z 620 (100, M ) , 592 (34.3), 564 (23.6), 536 (26.1), 508 +  (28.3), 480 (48.5), 452 (85.7, M - 6 CO). +  (B) In cyclohexane. The reaction of l,2,3-trithia-[3]-ferrocenophane with Ru3(CO)i2 was also carried out in cyclohexane following the same work-up as for the reaction in hexane. The orange /red crystals were obtained in lower yield (~10%). A brown band which did not elute with a variety of solvents (petroleum ether, petroleum ether/dichloromethane, dichloromethane, ethylacetate, methanol) represented ~75% of the mixture transferred onto a silica gel column.  29  2.3.2. The reaction with Os3(CO)i  2  (A) In hexane A mixture of FeCCsrLiSteS (0.028 g, 0.1 mmol) and Os3(CO)i2 (0.100 g, 0.11 mmol, 10% excess) in a three-necked flask was stirred and heated in hexanes (40 ml) at reflux under nitrogen for 18 h. The reaction mixture was monitored by using TLC and petroleum ether/dichloromethane (80/20) as eluent. TLC examination showed no reaction after 18 h.  (B) In cyclohexane A mixture of Fe(C H4S) S (0.028 g, 0.10 mmol) and Os (CO)i (0.100 g, 0.11 5  2  3  2  mmol) was stirred and heated in cyclohexane (40 ml) at reflux for 44 h. A dark brown suspension was produced. Chromatographic examination (TLC) indicated the presence of four main compounds (among them the ligand and Os3(CO)i represented ~60%) and 2  traces of another four. Solvent was separated from the suspension and the solid was extracted with pentane, then with dichloromethane. The extracts were combined with the previously separated reaction solution, and the solvent was removed under reduced pressure. Compound separation was achieved by preparative TLC (pentahe/dichloromethane 50/50). Only one product was obtained in a reasonable yield (~10 %). Attempted crystallization yielded yellow nodules. Crystals could  not be obtained  from  a range  of dichloromethane/hexane and  dichloromethane/pentane mixtures, toluene and dichloromethane/methanol (layer technique).  Fe(C H S) (S)Os2(CO)6: H NMR (CD C1): 5 4.57 (p.t., 4), 4.28 (p.t., 4). IR l  5  4  2  3  (CH C1 ): DCO (cm- ) 2095 (s), 2075 (w), 2040 (w), 2005 (s, br.). EI mass spectrum, 1  2  2  30  selected (relative intensity): m/z 830 (35.0, M+), 802 (9.4), 774 (12.1), 746 (10.0), 718 (30.4), 690 (22.0), 662 (43.8, M - 6 CO), 630 (5.1). +  (C)In toluene The reaction in toluene was carried out using the same procedure described in 2.3.2. (A) and (B). The reaction mixture was stirred and heated at reflux for 18 h. The color changed from yellow/orange to orange then to red. A red solution with a brown precipitate was finally obtained. The solvent was removed under vacuum to yield a brown/red residue that was almost completely soluble  in dichloromethane.  Chromatographic examination (TLC) indicated the presence of five main compounds and traces of another four. The brown/red mixture was slurried in dichloromethane with silica gel and after solvent removal, the gel was applied onto the top of a silica gel column. Elution with petroleum ether/dichloromethane (6/1) gave five bands (yellow, orange, yellow/orange, and a yellow and a brown band which did not elute). The yellow and orange fractions contained three and two compounds, respectively. Separation was achieved by collecting small fractions. Slow evaporation of the solvent afforded yellow/orange crystals of a compound A (yield 10 %), orange/yellow crystals (needles) of a compound B (yield 10%), dark red crystals of a compound C (yield 25%) and orange crystals (stars) of a compound D (yield 8%). Compounds A and B could not be recrystalized from toluene  and a range of dichlofomethane/pentane  and  dichloromethane/hexane mixtures. Compound D resisted attempts at recrystallization from a range of dichloromethane/pentane mixtures.  Fe(C H4S)20s2(CO) (A): l H NMR (CD C1): 8 4.31 (p.t., 4), 4.04 (p.t., 4). IR 5  8  3  (CH Cl2): 0>CO (cm- ) 2090 (w), 2060 (s), 2040 (w), 1990 (s). FAB mass spectrum, 1  2  selected: m/z 854 (M+), 826, 798, 770, 742, 714, 686, 658, 630 (M+ - 8 CO), 532, 476. Anal. Calcd.: C, 25.29; H, 0.94. Found: C, 25.70, H, 1.12. 31  Fe(C5H4S) Os3(CO)io (B): H NMR (CD3CI): 8 4.64 (p.t., 4), 4.34 (p.t., 4). IR l  2  (CH C1 ): DCO (cm- ) 2075 (w), 2050 (s), 2020 (s, br.), 1995 (s, br.). EI mass 1  2  2  spectrum, selected (relative intensity): m/z 1100 (16.4, M ), 1072 (6.5), 1044 (8.9), +  1016 (14.8), 988 (13.5), 960 (8.4), 932 (18.6), 904 (13.8), 876 (12.0), 848 (10.0), 820 (25.4, M+ - 10 CO), 798 (16.4, Os (CO) S+), 788 (8.00), 602 (4.9, Os S+). 3  7  3  Fe(C H S)20s4(S)(CO)ii (C): B. NMR (CD3CI): 8 4.88 (m, 2), 4.34 (m, 2), !  5  4  4.23 (m, 2), 4.07 (m, 2). IR (CH C1 ): x>co (cnv ) 2100 (s), 2080 (s), 2055 (s), 2010 1  2  2  (w, br.), 1720 (s, br.). EI mass spectrum, selected (relative intensity): m/z 1350 (25.8, M+),  1322 (16.9), 1238 (9.1), 1210 (26.1), 1182 (17.1), 1154 (14.2), 1126 (11.9),  1098 (13.9), 1070 (14.5), 1042 (36.6, M+ - 11 CO), 1010 (6.4),798 (7.7, Os (CO) S+). 3  7  Anal. Calcd.: C, 18.66; H, 0.59. Found: C, 19.00; H, 0.80.  Fe(C H S) Os4(S)2(CO)io (P): 5  4  NMR (CD C1): 8 5.09 (m, 2), 4.68 (m, 2),  2  3  4.35 (m, 2), 4.28 (m, 2). IR (CH C1 ): VQO (cm") 2110 (w), 2085 (s), 2060 (w), 2030 1  2  2  (s, br.), 2100 (w). EI mass spectrum, selected (relative intensity): m/z 1354 (8.7, M ), +  1242 (5.4), 1186 (5.7), 1074 (8.2, M - 10 CO), 1070 (7.1). +  2.3.3. The reaction with Co2(CO)  8  (A) In hexane. A solution of Co (CO) (0.044 g, 0.13 mmol) and Fe(C H4S)2S (0.028 g, 0.10 2  8  5  mmol) in hexane was stirred for 20 h at room temperature under a nitrogen atmosphere. The solvent was removed from the black suspension which was produced to give a black/dark green powder. Extraction with pentane, chloroform and dichloromethane gave a yellow/green solution which contained principally the ligand and another compound in trace quantity. This compound did not elute with petroleum ether / diethyl ether (50/50) 32  on a silica gel column. The remainder of the black/dark green reaction residue (~60% yield) was insoluble in common organic solvents (chloroform, dichloromethane, acetone, methanol). In another experiment the reaction mixture was stirred 6h at reflux and a similar black suspension was produced. The black reaction residue was extracted with pentane and dichloromethane and the weight of the remaining solid corresponded to a yield of ~80 %. The organic extract were concentrated in vacuo, then chromatographed on silica gel by using petroleum ether/dichloromethane as eluent. The first two fractions (of which one was the ligand) were rather diffuse and were not collected. The third fraction (an orange band) was collected and concentrated in vacuo to give an orange/red oil (yield ~10%) which resisted attempts at crystallization from a range of dichloromethane/hexane and dichloromethane/ethanol mixtures.  Fe(C H4S)2Co2(CO) , red oil. K NMR (CD C1): 8 4.32 (p.t., 4), 4.20 (p.t., 4). l  5  6  3  FAB mass spectrum: 535 (M ), and successive loss of 6 CO's. +  (B) In THF. Co2(CO)g (0.051 g, 0.15 mmol) was dissolved in THF (50 ml) in a two-necked round bottom flask, then l,2,3-trithia-[3]-ferrocenophane  (0.042 g, 0.15 mmol) was  added. The reaction mixture was allowed to stir for 2 h at room temperature, then for 8 h at 50° C. A black suspension was produced. The solvent was removed under vacuum to give principally black needles and a small amount of black powder. The black solids were extracted with chloroform, then with dichloromethane. The combined extracts contained four compounds, of which two were the starting materials , one was present in trace quantity  and another  one  in low  yield. Preparative plate  separation  (pentane/dichloromethane 50/50, eluent) afforded a yellow/brown powder (very low yield, ~l-2 %) and a brown-greenish powder (yield ~10%), which resisted attempts at 33  crystallization from dichloromethane, dichloromethane/ethanol (50/50) and a range of dichloromethane/hexane mixtures. The  black needles (yield ~60%) were insoluble in hexane, diethyl ether,  dichloromethane and acetone. In another experiment in THF the reaction mixture was stirred 5 h at 50° C, 1 h at reflux and overnight at room temperature. The black suspension afforded a black foil (yield ~80 %) and no other isolable products. The foil was insoluble in hexane, chloroform, dichloromethane and acetone.  Black needles. Anal. Found: C, 19,71; H, 2.77; S, 12.37. Fe(C H4S)2Co4(CO)i2, brown-greenish powder. H NMR (CD C1): 8 5.15 (m, l  5  3  2), 4.98 (m, 2), 4.93 (m, 2), 4.90 (m, 2). FAB mass spectrum: 820 (M+) and successive loss of CO's. Anal. Calcd. C, 39.28; H , 1.19; S, 9.52. Found. C, 38.57; H, 2.77; S, 12.37. Fe(C5H4S)2Co2(CO)8, yellow-brown powder. FAB mass spectrum: m/z 590 (M+) with successive loss of CO's. TR (CH C1 ): "OQO (cm" ) 2080 (w), 2050 (w), 2035 1  2  2  (w), 2010 (w), 1750 (w, br.). Black foil. Anal. Found: C, 26.17; H, 3.31; S, 10.91.  2.3.4. The reaction with Mn2(CO)io.  A solution of Mn (CO)io (0.047 g, 0.12 mmol) and Fe(C H4S) S (0.028 g, 0.10 2  5  2  mmol) was allowed to stir for 20 h at room temperature under a nitrogen atmosphere. No change in the initial color was observed. The reaction mixture was then stirred for 5 h at 80° C and 2 h at reflux to produce a brown yellow suspension. The solvent was removed and the brown yellow residue was extracted with pentane, chloroform and dichloromethane leaving a brown powder (yield ~55 % ). Preparative TLC separation of 34  the combined extracts afforded a colorless fraction in low yield (~5%), a yellow fraction (the ligand, ~20%) and an orange fraction (-20%). Brown powder. FAB mass spectrum: 391 (Fe(C5H4S) Mn(CO)2S ). Anal. +  2  Found: C, 14.43; H , 2.48. Orange powder. lH NMR (CD C1): 8 4.50 (br), 3.85 (br). EI mass spectrum: 335 3  (Fe(C H4S)2MnS ) +  5  2.4. PREPARATIONS OF (NJV-DIMETHYLAMINOMETHYL)- AND (DIPHENYLPHOSPHINO)-l,2,3-TRITHIA-[3]-FERROCENOPHANE.  2.4.1. (N,N-Dimethylaminomethyl)-l,2,3-Trithia-[3]-Ferrocenophane (DMAMTF).  To a solution of (N,N-dimethylaminomethyl)ferrocene (2.00 g, 8.23 mmol) in diethyl ether (25 ml) was added 5.10 ml (8.16 mmol) of a 1.6 M solution of nbutyllithium. The reaction solution was stirred for 2 h and a further 5.50 ml (98.8 mmol) of n-butyllithium solution and TMED (1.00 g, 8,62 mmol) were added. The reaction solution was stirred overnight. Sublimed sulphur (10.00 g, 3,90 mmol) was then added carefully and the reaction mixture was stirred for a further 2 h before it was hydrolyzed with a saturated solution of sodium bicarbonate (25 ml). The organic fraction was separated and the aqueous phase was extracted with diethyl ether ( 3 x 4 ml). The combined ether fractions were evaporated to dryness under reduce pressure to obtain a brown-redish oil. DMAMTF was isolated from this oil by column chromatography on silica using diethyl ether with 5% m^thylarnine as eluent. Removal of the solvent yielded a greenish-yellow powder (yield 40%).  Fe(C H3(N(CH3)2CH2)S)(C5H4S ). *H NMR (CD3CI): 8 4.62 (m, 1), 4.54 (m, 5  2  35  1) , 4.45 (m, 1), 4.40 (m, 1), 4.36 (m, 5), 4.30 (m, 1), 4.11 (m, 1), 3.81 (m, 1), 3.76 (m, 1), 3.39 (m, 1), (Fe(-C H ); 3.23 (p.d., 2), 2.98 (p. d., 2), (-CH ); 2.22 (s, 6), 5  4  2  2.14 (s, 6), (-N(CH3)2). EI mass spectrum, selected (relative intensity): m/z 338 (19.10, (M+-H)+), 337 (100, M+), 293 (14.50), 261 (9.20), 241 (31.30), 230 (12.40), 198 (14.50). Anal. Calcd. C, 46.29; H, 4.45; N, 4.15. Found: C, 45.47; H, 4.46; N, 3.76.  2.4.2. (Diphenylphosphino)-l,2,3-trithia-[3]-ferrocenophane  (DPPTF).  147  (A) Preparation of (Diphenylphosphino)ferrocene. (Dppf) n-Butyllithium 1.6 M (32 ml, 0.05 mol) was added to ferrocene (13.00 g, 0.07 mol) in diethyl ether (75 ml). The solution was stirred (60 h) at room temperature after which it was cooled in a dry ice/acetone bath and PI12PCI (9 g, 0.04 mol) added dropwise. The solution was allowed to warm to room temperature with stirring and left for 2 h after whichtimewater (50 ml) was added. The diethyl ether fraction was isolated, reduced in volume on a rotary evaporator, and applied to a silica chromatographic column. Unreacted ferrocene was eluted with a petroleum ether/diethyl ether (20/1) mixture. Dppf was eluted with a petroleum ether/diethyl ether (1/1) mixture and evaporated to dryness under reduced pressure to give an yellow orange solid (yield 60 %). Dppf. !H NMR (CD3CI): 8 7.35 - 7.33 (m, 10), (-C H ); 4.33 (m, 2), 4.12 (m, 6  5  2) , 4.08 (s, 5), (Fe(-C5H4)). EI mass spectrum, selected (relative intensity): m/z 370 (100, M+), 293 (25.7).  (B) The lithiation of (Diphenylphosphinojferrocene and reaction of the lithiated product 148  with sulphur.  n-Butyllithium in hexane (2.1 mol equiv. of a 1.55 M solution, 1.87 ml, 2.89 mmol) was added to a rapidly stirred solution of Dppf (0.50 g, 1.38 mmol) in diethyl 36  ether (40 ml). TMEDA (1 mol equiv. with respect to n-BuLi, 0.43 ml, 2.89 mmol) was subsequently added. The reaction mixture was stirred for 22 h and the resulting reaction suspension was treated with S$ (0.099 g, 3.09 mmol). The reaction mixture was stirred for 6 h, then hydrolyzed with aqueous sodium bicarbonate (25 ml, saturated solution). The organic fraction was separated. Concentration of this fraction gave an orange oil. Chromatographic examination, TLC, showed the presence of five main products and traces of another four. Compound separation was achieved by column chromatography and preparative TLC (hexane/diethyl ether (60/40) eluent). Five fractions were collected. After solvent removal under reduced pressure two oils and three powder compounds were obtained (the yields ranged from 5 to 20% ).  Fe(C H3(P(S)Ph2)S-l,3)(C H4S2)(yield 8%). l H NMR (CD C1): 8 7.80 - 7.90 5  5  3  (m, 4), 7.50 - 7.60 (m, 6) (-C6H5). Major isomer 4.83 (m, 1), 4.79 (m, 1), 4.70 (m. 1), 4.48 (m, 1), 4.45 (m, 1), 4.05 (m, 1), 3.83 (m, 1), (Fe(-CsH )). Minor isomer 4.92 (m, 4  1), 4.83 (m, 1), 4.63 (m, 1), 4.51 (m, 1), 4.27 (m, 1), 4.01 (m, 1), 3.72 (m, 1). Ratio of major: minor = ~60 : 40. EI mass spectrum, selected (relative intensity): m/z 4.98 (8.99), 497 (11.82), 496 (41.85, M+), 432 (7.62), 402 (36.87), 400 (17.56), 368 (6.29, 337 (37.03), 260 (5.60).  Fe(C5H3(P(S)Ph2)S-l,2)(C5H4S2) (yield 3%). EI mass spectrum, selected (relative intensity): m/z 498 (21.6), 497 (27.7), 496 (100, M+), 432 (33.4), 402 (12.2), 401 (17,2), 400 (76.6), 368 (13.0), 260 (10.8).  [Fe(C5H4P(S)Ph2)(C5H4)]2S (yield 20 %). EI mass spectrum, selected (relative intensity): m/z 834 (100, M+), 433 (33.7), 402 (17.2).  [Fe(C5H4PPh2)(C5H4)] S (yield - 1 0 %). EI mass spectrum, selected (relative 2  37  intensity): m/z 770 (84.20, M+), 401 (25.10), 370 (22.0).  2.5. REACTIONS OF THE (N.N-DIMETHYLAMINOMETHYL)-1,2,3-TRITHIA-[3]FERROCENOPHANE.  2.5.1. The reaction with R u ( C O ) . 3  12  (A) In hexane. Ru3(CO)i2 (0.430 g, 0.68 mmol) was added to a rapidly stirred solution of DMAMTF (0.229 g, 0.68 mmol) in hexane. The reaction mixture was allowed to stir for 22 h at reflux. The color of the reaction mixture changed from red to dark brown. TLC examination indicated that the starting materials were still present in addition to two new compounds. The solvent was removed under reduced pressure to give a dark brown oily residue, part of which was slurried in dichloromethane and chromatographed on a silica gel column. Although dichloromethane/petroleum ether (2/1) mixture, dichloromethane, acetone and methanol were used as eluents most of the reaction mixture transferred onto the column (~80%) did not elute. In another experiment the dark brown reaction residue was extracted with dichloromethane to give a yellow solution that contained the starting materials and a colorless compound in a trace quantity. Further extraction of the residual brown powder with more polar solvents indicated that no soluble compounds were present. Brown oil. U NMR (d6-DMSO): 8 3.40 (br). EI and FAB mass spectrum: no l  ionization.  (B) In cyclohexane. A procedure similar to that described in 2.5.1.A was followed. The reaction mixture was stirred at reflux for 20 h. The color changed from reddish to dark red, then 38  red/brown and finally to dark brown. Removal of solvent afforded a dark brown oily residue that could not be separated by column chromatography or preparative TLC.  (C) In THF. To a rapidly stirred, orange solution of DMAMTF (0.034 g, 0.1 mmol) and Ru3(CO)i2 (0.640 g, 0.1 mmol) in THF, (PPh ) NCl (5.00 mg) was added. The reaction 3  2  mixture was stirred for 18 h at room temperature. During this time a brown solid precipitated. TLC examination of the reaction mixture showed that no starting materials remained. Removal of the solvent under vacuum afforded a brown solid which was extracted with dichloromethane to afford a light brown solution which contained only the catalyst. The residual brown powder (yield ~90 %) was insoluble even in more polar solvents such as THF, acetonitrile and methanol. In a separate experiment the reaction mixture was stirred only for 4 h. The brown reaction residue was extracted with dichloromethane and the resulting orange solution contained the starting materials and the catalyst as judged by TLC. The remainder of reaction residue was a brown powder (yield -40 %).  Brown powder (yield -90 %). K NMR (d6-DMSO): 8 3.4 (br). IR (DMSO): l  \) c o  (cm" ) 1  2050  (w,  br).  FAB  mass  spectrum:  538  [Fe(C H3(CH2N(Me)2)S)(C5H4S)Ru(CO)+]. Anal. Found: C, 35.56; H, 3.00; N, 1.70. 5  Brown powder (yield -40 %). Anal. Found: C, 34.82; H, 3.05; N, 1.75.  2.5.2. The reaction with F e ( C O ) . 2  9  DMAMTF (0,034 g, 0.1 mmol) was added to a rapidly stirred solution of Fe2(CO)9 (0.039 g, 0.1 mmol) in THF. The reaction mixture was stirred for 20 h at room temperature. The solvent was removed from the dark brown suspension to give a dark 39  brown oily residue. This was extracted with dichloromethane to produce a brown solution and a dark brown powder (~20%). After solvent removal the extract gave a brown oil that contained five compounds of which one was the ligand, another one in low yield (~15 %) and three in very low yield (total~15%). Compound separation was attempted by preparative TLC using diethyl ether with 5% triethylamine as eluent. A trace amount of an orange/brown oily compound was isolated. The only compound that was present in a reasonable yield could not be extracted from the preparative plate with dichloromethane or acetone.  Brown powder. FAB mass spectrum: m/z 443. Brown oil. *H NMR (CD3CI): no ferrocenyl protons. Orange/brown oil. EI mass spectrum, selected (relative intensity): m/z 943 (6.7) with successive loss of 6 CO's; 649 (6.9).  2.5.3. The reaction with [Rh(COD)Cl]  2  [Rh(COD)Cl]2 (0.025 g, 0.05 mmol) was added to a rapidly stirred solution of DMAMTF (0.017 g, 0.05 mmol) in THF. The reaction mixture was allowed to stir for 21 h at room temperature. During thistimethe initially yellow solution turned light brown, but TLC examination showed that the starting materials were the major components of the reaction mixture. No change in color was observed after 4 h at 40 - 45° C. The reaction mixture was then allowed to stir at room temperature overnight. During this time the color turned dark brown. After the solvent removal the solid reaction residue was extracted with pentane, chloroform and dichloromethane. The residue was a black powder (~25%). Removal of solvent from the extract afforded a brown solution of starting materials and two compounds in trace quantity.  40  Black powder. MS and FAB mass spectrum: no ionization. Anal. Found: C, 25.76; H , 3.76; N, 1.66.  2.5.4. The reaction with [Rh(COE) Cl] 2  2  A solution of DMAMTF (0.017 g, 0.05 mmol) and [Rh(COE) Cl] (0.036 g, 2  2  0.05 mmol) in THF was stirred for 75 min. at room temperature. The light brown/greenish reaction mixture turned brown/black after 20 min. After 75 min. there was no unreacted ligand in the reaction mixture. The solvent was removed to give a black residue, which was extracted with diethyl ether, chloroform and dichloromethane. The residue consisted of a fine black powder (yield ~ 90 %) which was insoluble in common organic solvents.  Black powder. EI and FAB mass spectrum: no ionization. Anal. Found: C, 26.41; H, 3.62; S, 11.22; N, 1.94.  2.5.5. The reaction with Pd(COD)Cl  2  A mixture of DMAMTF (0.034 g, 0.1 mmol) and Pd(COD)Cl2 (0.029 g, 0.1 mmol) in THF was stirred for 20 h at room temperature. The initially yellow solution turned brown. Removal of solvent under vacuum afforded a brown/black oily residue which was extracted with chloroform and dichloromethane to give an orange/brown solution and a brown/black powder (yield -30%) that was insoluble in common organic solvents. The extract contained the starting materials and one compound in very low yield. Brown/black powder. EI and FAB mass spectrum: no ionization.  41  I l l RESULTS AND DISCUSSION  3.1. PREPARATION OF 1,2,3-TRITHIA-[3]-FERROCENOPHANE  l,2,3-trithia-[3]-ferrocenophane was first reported by Bishop and co-workers.  22  The synthesis (Scheme 3.1.) involved the well studied dilithio ferrocene. - " 22  124  129  Scheme 3.1.  In the present investigation the l,r-dilithioferrocene.TMEDA adduct was isolated and then allowed to react with sulfur. This procedure proved to be superior to the in situ use of a slurry of the dilithioferrocene because any unreacted ferrocene remaining in the slurry, made the separation of the analytically pure material more difficult. Compound separation is better accomplished by column chromatography than in the literature procedure of extraction with acetone and recrystallization from benzene. Diferrocenyl sulphide, [Fe(Tj -C5H5)(rj -C5H4)]2S, a known compound, was 5  5  obtained as a byproduct in the preparation of 1,2,3- trithia-[3]-ferrocenophane. Its yield was higher (~30 %) when the stepwise procedure was used.  42  3.2. REACTIONS OF l,2,3-TRITHIA-[3]-FERROCENOPHANE WITH METAL CARBONYLS.  3.2.1. The reaction with Ru3(CO)i2.  The thermal reaction of l,2,3-trithia-[3]-ferrocenophane with Ru3(CO)i2 in refluxing hexane, using a 1:1 molar ratio of reactants afforded the ruthenium binuclear complex 43 (Scheme 3.2.). ^  (CO)  3  43  Scheme  3.2.  The reaction involves the opening of the ferrocenophane system, the loss of a sulphur atom and breakdown of the ruthenium cluster. At the time the present investigation was initiated the only example of a ferrocenedithiolato derivative of a metal carbonyl was  Fe(r|-C5H4S)2Fe2(CO)6 24, 5  which was reported and characterized by *H NMR, IR and elemental analysis by Seyferth and Hames.  62  The formulation of the complex 43 as a dinuclear electron precise complex is supported by its H NMR, IR, mass spectra, and confirmed by single crystal X-ray l  analysis. 43  3.2.2. Spectroscopic Properties and Molecular Structure of the Ferrocenedithiolato Derivative of Ruthenium Carbonyl.  (l.l'-ferrocenedithiolato-S.SOdiruthenium hexacarbonyl 43, is an orange/red solid, soluble in common organic solvents such as hexanes, benzene, dichloromethane and THF. It is air stable in the solid state, and, for short periods of time no noticeable decomposition was observed in solution. The *H NMR spectrum consists of two pseudo triplets at 8 (ppm) 4.03 and 4.34 corresponding to the a- and P- protons of the Cp ring. The chemical shift difference of 0.31 ppm between the two pseudo triplets is comparable with the chemical shift difference (0.36 ppm) in the isoelectronic and supposedly isomorphous ferrocenedithiolato iron derivative 24. The electron impact mass spectrum shows the parent peak (M ) and stepwise loss +  of six CO groups. The IR spectrum of the complex in CH2CI2 solution exhibits strong absorption in the region 2005-2075 cm attributable to the stretching vibration of the -1  terminal CO's. The ruthenium complex was characterized by a single-crystal X-ray diffraction study. The compound crystalizes in the monoclinic crystal system, space group P21/n and contains four independent molecules per unit cell. The ORTEP view of the molecular structure is presented in Fig.3.1. Crystal data and experimental details, and selected bond lengths and bond angles are presented in Tables 3.1., 3.2. and 3.3. The X-ray crystal structure depicted in Fig. 3.1. shows that the cyclopentadienethiolato groups ( C 5 H 4 S ) bridge symmetrically the two Ru(CO)3 units. Overall the 130  133  compound has the sawhorse arrangement observed in other ruthenium (I) dimers but one significant difference should be noted: the coordination about the ruthenium atoms is a very distorted octahedron with angles Ru(2) - Ru(l) - C(ll) (151.9(1)°) and Ru(l)44  Table 3.1. Selected Crystal Data and Experimental Details for 43.  Empirical Formula  C H Fe0 Ru S  Formula Weight  618.34  Crystal Color, Habit  orange, plate  Crystal Dimensions (mm)  0.050 x 0.350 x 0.400  Crystal System  monoclinic  16  8  6  2  2  No.Reflections Used for Unit Cell Determination Lattice Parameters:  25 (42.7 - 49.1°) a = 7.685 (3) A b = 13.866 (4) A c = 18.225 (2) A B = 96.43 (2)° V = 1929.8 (8) A  Space Group  P2i/n  Z value  4  Dcaic  2.128 g/cm  F  000  1192  Diffractometer  Rigaku AFC 6S  Radiation  MoKa  |j,(MoKa)  24.96 cm"  Scan Rate  32.0°/min  No. of Reflections Measured  Total: 6248  3  3  1  Unique: 5856 (R  int  = 0.039)  Structure Solution  Patterson Method  Refinement  Full-matrix least-squares  46  Table 3.2. Selected Bond Lengths (A) with Estimated Standard Deviations for 43.  atom  atom  distance  atom  atom  distance  Ru(l)  Ru(2)  2.6812(7)  Fed)  Cp(l)  1 .641(2)  Ru(l)  Sd)  2.398(1)  Fed)  Cp(2)  1 .639(2)  Ru(l)  S(2)  2.403(1)  S(l)  C(l)  1 .765(3)  Ru(l)  Cdl)  1.935(5)  S(2)  C(6)  1 .769(3)  Ru(l)  Cd2)  1.900(5)  0(1)  C(ll)  1 .126(5)  Ru(l)  C(13)  1.902(4)  0(2)  C(12)  1 .124(6)  Ru(2)  Sd)  2.3969(9)  0(3)  C(13)  1 .136(5)  Ru(2)  S(2)  2.402(1)  0(4)  C(14)  1 .136(5)  Ru( 2)  Cd4)  1.926(4)  0(5)  C(15)  1 .131(5)  Ru( 2)  C(15)  1.909(5)  0(6)  C(16)  1 .128(5)  Ru( 2)  C(16)  1.906(4)  C(.l)  C(2)  1 .428(5)  Fe(l)  Cd)  2.010(3)  C(l)  C(5)  1 .417(5)  Fe(l)  C(2)  2.032(4)  C(2)  C(3)  1 .422(6)  Fed)  C(3)  2.047(4)  C(3)  C(4)  1 .388(7 )  Fed)  C(4)  2.051(4)  C(4)  C(5)  1 .426(5)  Fed)  C(5)  2.033(4)  C(6)  C(7)  1 .419(5)  Fed)  C(6)  2.005(3)  C(6)  C(10)  1 .411(5)  Fed)  C(7)  2.024(4)  C(7)  C(8)  1 .425(6)  Fed)  C(8)  2.052(4)  C(8)  C(9)  1 .384(7)  Fed)  C(9)  2.048(4)  C(9)  C(10)  1 .429(6)  Fed)  CdO)  2.038(4)  47  Tab. 3.3. Selected Bond Angles (deg) with Estimated Standard Deviations for 43.  atom  atom  atom  angle  atom  atom  atom  Ru(2)  Ru(l)  S(l)  55 .99(3)  S(2)  Rut 2)  C(16)  155 .6(1)  Ru(2)  Ru(l)  S(2)  56 .07(3)  C(14)  Ru(2)  C(15)  99 .2(2)  Ru(2)  Ru(l)  C(ll)  151 .9(1)  C(14)  Ru( 2)  C(16)  98 .2(2)  Ru(2)  Ru(l)  C(12)  100 .0(2)  C(15)  Ru(2)  C(16)  90 .3(2)  Ru(2)  Ru(l)  C(13)  101 .9(1)  Cp(l)  red)  Cp(2)  177 .66(9)  S(l)  Ru(l)  S(2)  Ru(l)  S(l)  Ru(2)  68 .01(3)  S(l)  Ru(l)  C(ll)  108 .1(2)  Ru(l)  S(l)  C(l)  116.6(1)  S(l)  Ru(l)  C(12)  153 .8(1)  Rut 2)  S(l)  C(l)  115 .6(1)  S(l)  Ru(l)  C(13)  85 .0(1)  Ru(l)  S(2)  Rut 2)  S(2)  Ru(l)  C(ll)  103 .8(1)  Ru(l)  S(2)  C(6)  116 .1(1)  S(2)  Ru(l)  C(12)  87 .1(2)  Rut 2)  S(2)  C(6)  115 .6(1)  S(2)  Ru(l)  C(13)  157 .0(1)  S(l)  C(l)  C(2)  125 .2(3)  C(ll)  Ru(l)  C(12)  98 .0(2)  S(l)  C(l)  C(5)  126 .7(3)  C(ll)  Ru(l)  C(13)  99 .2(2)  C(2)  C(l)  C(5)  107 .8(3)  C(12)  Ru(l)  C(13)  91 .0(2)  C(l)  C(2)  C(3)  107 .3(4)  Ru(l)  Ru(2)  S(l)  56 .01(3)  C(2)  C(3)  C(4)  108 .8(4)  Ru(l)  Ru(2)  S(2)  56 .10(3)  C(3)  C(4)  C(5)  108 .4(4)  Ru(l)  Ru(2)  C(14)  149 .6(1)  C(l)  C(5)  C(4)  107 .7(4)  Ru(l)  Ru(2)  C(15)  103 .4(2)  S(2)  C(6)  C(7)  125 .6(3)  Ru(l)  Ru(2)  C(16)  101 .7(1)  S(2)  C(6)  C(10)  125 .8(3)  S(l)  Ru(2)  S(2)  C(7)  C(6)  C(10)  108 .2(3)  S(l)  Ru(2)  C(14)  102 .4(1)  C(6)  C(7)  C(8)  107 .6(4)  S(l)  Rut 2)  C(15)  158 .4(2)  C(7)  C(8)  C(9)  108 .1(4)  S(l)  Rut 2)  C(16)  88 .0(1)  C(8)  C(9)  C(10)  109 .1(4)  S(2)  Rut 2)  C(14)  106 .2(1)  C(6)  C(10)  C(9)  107 .0(4)  S(2)  Ru(2)  C(15)  86 .0(1)  Ru(l)  C(ll)  0(1)  177 .8(5)  86 .70(3)  86 .74(3)  angle  67 .83(3)  Ru(2)-C(14) (149.6(1)°) which differ by 28.1 and 30.4° respectively, from the ideal 180°. This distorted octahedral coordination is the result of the strain imposed by the bridging ligands, with angles Ru(l)-S(2)-Ru(2) (67.83(3)°),  Ru(l)-S(l)-Ru(2)  (68.01(3)°) and S(l)-Ru(l)-S(2) (86.70(3)°, S(l)-Ru(2)-S(2) (86.74(3)°) deviating by 41.5° and 3.30° from the ideal 109.5° and 90°, respectively. The Ru-Ru separation of 2.6812(7) A is similar to the lengths found in other  A. " A shorter Ru-Ru distances of 2.5788(3) A was 133  ruthenium dimers (average 2.70  136  137  found in Ru2(|i-RC02)2(CO)4L (L=P-donor ligands; R=alkyl, aryl) 2  which also has a  distorted octahedral arrangement of the ligand around theratheniumatoms. Each sulphur atom is equally bonded to the two ruthenium atoms (Ru(l)S(l)=2.398(l) A and Ru(2)-S(l)=2.3969(9) A ); Ru(l)-S(2)=2.403(l) A and Ru(2)S(2)=2.402(l) A). These values are very close to those reported (average 2.401 A) for Ru (CO) [li-SC(NMe2)CMe] . There are no differences in the Ru-Ru-S (55.99(3)° 134  2  6  and 56.07(3)°), S-Ru-S (86.70(3) and 86.74(3)°), S-Ru-C and Ru-Ru-C bond angles on the two ruthenium atoms. The Ru-C (carbonyl) distances range between 1.900 and 1.935 A, and are comparable with those reported for diruthenium complexes (average 1.914 A ) nuclearityratheniumclusters that contain sulfide ligands (average 1.921 A )  138  1 3 4  and high  . Of the six  Ru-C distances the Ru(l)-C(l 1) and Ru(l)-C(14), trans to the Ru-Ru bond are the longest (1.926(4) and 1.935(5) A, respectively. Both SC5H4 ligands are T| -bound in a pseudo-trans configuration to iron, much 5  as in ferrocene and its derivatives, and are o bound toratheniumthrough the sulphur atoms. The geometry about iron is surprisingly close to that of ferrocene; the two 85  cyclopentadienyl rings are nearly parallel, the dihedral between the rings angle being only 5.38°. The small tilt is close to that observed in (fdpp)MX2 (M = Pd, Pt; X = CI) ( 1 . 9 ° ) , and significantly smaller than that found in some 2-metalla-[3]-ferocenophanes 78  such as (l.l'-ferrocenedithiolato-S.S'Fe) triphenylphosphine Pt(II) (21.0°). 49  Theringsexhibit an eclipsed conformation. The two Cp rings are virtually planar, the largest deviation of any carbon from the plane is 0.055 A. While in l,2,3-trithia-[3]ferrocenophane the cyclopentadienyl rings are coplanar with the sulphur atoms in the 1 and 3 positions, in the (CsH4S)2Ru2(CO)6 molecule the two sulphur atoms are displaced outwards from these planes by 0.139 A and 0.155 A, respectively. The Fe-ring C distances range from 2.005(6) to 2.052(8) A (mean 2.034 A), and these distances compare well with the values reported for l,2,3-trithia-[3]-ferrocenophane (mean 2.044 A)  90  and (l,l-ferrocenedithiolato-S,S'Fe)(diphenylphosphine)Pt(II) (mean 2.06 A ) . ,  79  The distance of the iron from the planes of the two rings (1.639 and 1.641 A) are comparable with those found in ferrocene (1.66 A ) and 1.2.3-trithia-[3]-ferrocenophane 85  (1.662 and 1.634 A)*>. The C-C bond distances in the Cp rings ranged from 1.384(7) A to 1.429(6) A (average 1.414 A). Average values of 1.42 A and 1.41 A were found for l,2,3-trithia-[3]ferrocenophane and (1,l'-ferrocenedithiolato-S,S',Fe)(diphenylphosphine)Pt(II). The bond angles of the Cp rings are normal, ranging from 107.0(4)° to 109.1(4)° and their average value is 108.0°, which is also the average value of the C-C-C angle found in 1,2,3-trithia-[3]-ferrocenophane. The C-S distances of 1.765 and 1.769 A appear normal for thiolato 75,79,90  groups.  3.2.3. The reaction with O s ( C O ) i . 3  2  Investigations of the thermal reaction of l,2,3-trithia-[3]-ferrocenophane with Os3(CO)i2 in different solvents (hexanes, cyclohexane and toluene) indicates that a higher temperature than for the reaction with Ru3(CO)i2 is required to initiate the reaction. This is an indication of the increased difficulty of cluster breakdown due to the increase in the strength of metal-metal bonds on increasing atomic mass in the transition metal triad.  50  l,2,3-trithia-[3]-ferrocenophane does not react with Os3(CO)i2 in refluxing hexanes, but yields Fe(r| -C5H4S)2(S)Os2(CO)6 44 in low yield in refluxing 5  cyclohexane (Scheme 3.3.) The reaction in refluxing toluene affords one dinuclear Fe(T| -C H4S)20s (CO) 45 and one trinuclear complex FeO^-CsfyS^Os^CCOiQ 46 in 5  5  2  8  low yield, and two tetranuclear complexes Fe(Tj -C5H4S)2(|i3-S)(^-CO) Os4(CO)9 47 5  2  and Fe(r| -C5H4S) (iX3-S)(|i-S)Os4(CO) o 48 in moderate and low yield, respectively. 5  2  1  S  / (CO) Os^  \ Os(CO)  3  /  S  +  Os (CO) 3  oo  cyclohexane  3  to  12  44 Scheme 3.3.  The formation of the ferrocenedithiolato derivatives of osmium carbonyl occurs with the breakdown of the triosmium cluster for 44 and 45, with opening of the cluster for 46, with retention of the triosmium cluster and attachment of one Os(CO) unit for 47 3  and 48. The formulation of the above ferrocenedithiolato derivatives (Figure 3.2.) is based on H NMR, IR, mass spectral and analytical data (for 45 and 47). The molecular l  structure of 47 was confirmed by a single-crystal X- ray analysis. Two more compounds in trace quantities Fe(ri C5H S)2(S)Os (CO)i2 and 5_  4  3  Fe(rj -C5H4S)2(S)Os5(CO)i4 were identified from of the parent ions at m/z 1190 and 5  1626 in their low-resolution mass spectra.  51  (C0) Os  45  Fig. 3.2. The  4  46  proposed structures of ferrocenedithiolato derivatives  of osmium carbonyl 45 - 48.  52  3.2.4. Ferrocenedithiolato Derivatives of Osmium Carbonyl. Spectroscopic Properties.  The H NMR spectra of complexes 44,45 and 46 exhibit two multiplets of equal l  intensity, corresponding to the cc-and pVprotons of the Cp ring. This indicates an AA'BB' spin system arising from a symmetrical structure. The chemical shift difference of 0.29, 0.27 and 0.30 ppm between the two pseudo triplets in 44,45 and 46 is comparable with that observed in the ferrocenedithiolato derivatives of ruthenium and iron carbonyl 43 and 24. The H NMR spectra of complexes 47 and 48 show four multiplets of equal intensity l  in the Cp region arising from an ABCD spin system and indicating an asymmetrical structure. The IR spectra of complexes 44, 45, 46 and 48 show carbonyl stretching vibrational frequency in the terminal u(C-O) region, but no absorption in the bridging v(C-O) region, so the complexes must have a structure with only terminal carbonyls. A structure with bridging carbonyls is expected for 47 since the IR spectrum of the complex shows a broad, very strong absorption in the bridging v(C-O) region. Electron impact mass spectroscopic and FAB analysis proved to be very useful for the formulation of these osmium derivatives. All the compounds show parent ions and peaks associated with sequential loss of the carbonyl ligands. Smaller peaks are associated with loss of sulphur, fragmentation of the ligand and loss of the pendant osmium atom. Compound 44 exhibits the fragment ion Os2(C5H4S)2 after the loss of six carbonyls and +  one sulphur atom, and 45 exhibits the same ion after the loss of eight carbonyls. The parent ion of 48 indicates that this compound has one less CO group and one more sulphur atom than 47. The two compounds exhibit  Os4(S)Fe(CsH4S)2 and +  Os4(S)2Fe(C5H4S)2 fragments after the loss of eleven and ten carbonyls, respectively. +  Both compounds exhibit Os (S)(CO)8 , and 47 is the only derivative whose spectrum +  4  53  shows the fragment Os3S(CO)7 , originating from the loss of the pendant osmium. +  Only two of the osmium derivatives 45 and 47 could be obtained in an analytically pure form. The microanalytical data for 44, 46 and 48 was not good in spite of repeated attempts to further purify the compounds. The poor analytical data of 46 and 48 reflects the difficulty of separating these products in analytically pure form by chromatography and fractional crystallization since they posses similar solubility and chromatographic properties with 45 and 47. In the suggested structures 45 - 48 the thiolato groups act as one- or threeelectron donor towards the osmium atoms while the sulfido ligand serves as a two- or four-electron donor. For example, compound 45 has the Os atoms bound to the ferrocenedithiolato ligand, with each S acting as one-electron donor. Compound 46 has an OS3S2 unit. There are various degree of M-M bonding among complexes containing a M3S2 unit,  114  and each sulphur of the C5H4S groups may  act as a doubly or triply bridging ligand. A structure with two triply bridging sulphur atoms, acting as three-electron donors is unlikely in this case, since it would require a bridging CO to satisfy the 18-electron rule, which is usually obeyed by OS3 clusters.  139  The proposed structure 46 has two doubly bridging sulphur atoms acting as three electron donors. To satisfy the 18-electron rule the triosmium cluster must be open. Compounds 47 and 48 are clusters containing four Os atoms. Each compound contains a sulfido ligand which serves as a bridge to three metal atoms and acts as a fourelectron donor. In addition, in each compound the SC5H4 groups serve as bridges to two metal atoms and act as three-electron donors. Compound 48 has a sulfido ligand absent in 47 which acts as a two electron-donor and bridges one Os-Os bond of the osmium triangle. The sulfido ligand proves to be very flexible in these osmium carbonyl derivatives as it does in high nuclearity sulfido-osmium carbonyl clusters in which it acts as a two-, four- or six-electron donor while serving as a doubly, triply or quadruply bridging ligand 54  (Fig. 3.3.).  138  d  e  Fig. 3.3. The geometry of sulfido ligand in sulfido-osmium carbonyl clusters.  3.2.5. Molecular Structure of (l,l'-ferrocenedithiolato-S,S')sulfidotetraosmium carbonyl.  The molecular structure of 47 was confirm  by a single-crystal X-ray analysis.  The ORTEP view of the molecular structure is presented in Fig 3.4. Selected crystal data and experimental details are given in Table 3.4., and selected bond lengths and bond angles in Tables 3.5. and 3.6., respectively. The structure reveals a spiked triangular configuration of osmium atoms. A sulfido ligand triply bridges the osmium triangle while each S atom of the C5H4S ligands bridges the two osmium atoms Os(l) and Os(4), which form the spike bond. The bridges 55  involving the two sulphur atoms of the C5H4S ligands are essential symmetrical. The distances S-Os are equal [S(2)-Os(l)=2.428(3) A and S(2)-Os(4)=2.410(3) A; S(3)Os(l)=2.425(3) A and S(3)-Os(4)=2.413(3) A] and there are no differences in the Os-OsS [55.48(6)° and 56.13(6)°], and S-Os-S [86.0(3) and 86.7(1)°] bond angles on the two osmium atoms Os(l) and Os(4). However there is a small difference in the Os-S-C [C(l) or C(6)] [113.5(4) and 116.8(4)°] bond angle on the two osmium atoms. The osmium triangle is of unequal lengths with two osmium atoms, Os(2) and Os(3), bound to three terminal carbonyl groups. There are two weak semibridging interactions in the molecule, that involve the two carbonyls attached to (Os(l)), and the other two osmium atoms of the triangle. The geometry around Os(2), Os(3) and Os(4) is distorted octahedral with deviations varying between 18.0 and 31.1° from the ideal 180°. The unique Os(l) atom is eight coordinated. The triply bridging sulfide shortens significantly the Os-Os bonds of the triangle (average 2.730 A) relative to the value found in the parent carbonyl, Os3(CO)i2 (average 2.877 A ) .  140  The non-carbonyl-bridged Os-Os bond (Os(2)-Os(3)=2.7257(9) A is shorter  than the carbonyl-bridged Os-Os bonds (Os(l)-Os(3)=2.8563(7) and Os(l)Os(2)=2.8576(7) A). The spike Os-Os bond (Os(l)-Os(4)=2.7192(7) A) is shorter than any of the Os-Os bonds of the triangle. The Os- (Os(l)-, Os(2)-, Os(3)-) S(l) distances (average 2.354 A) are shorter than the average values of 2.405 and 2.390 A found respectively in the open cluster compound Os(u. _S)2(CO)9 3  141  and the closed cluster compound Os (U3-S)(u-H) (CO)9. 3  142  2  The Os-S-Os angles are fairly acute: Os(l)-S(l)-Os(2)=74.41(8); Os(l)-S(l)Os(3)=74.42(8); Os(2)-S-Os(3)=71.26(8) and Os(l)-S(2)-Os(4)=68.40(7)°. However, they lie in the range 65-75° reported for M-S-M angles in cluster compounds.  143  The Os-C (terminal carbonyl) distances he in the range 1.87(1)-1.91(1) A (average 1.89) while the C-0 distances average 1.15 A .Similar values are found in other osmium 56  Fig. 3.4. The ORTEP view of the molecular structure of 47.  Table 3.4. Selected Crystal Data and Experimental Details for 47.  Empirical Formula  C H Fe0 0s S3  Formula Weight  1349.11  Crystal Color, Habit  green, prism  Crystal Dimensions (mm)  0.180 x 0.240 x 0.290  Crystal System  monoclinic  21  8  11  4  No. Reflections Used for Unit Cell Determination Lattice Parameters:  25 (42.3 - 48.7°) a =11.872 (2) A b = 14.230 (3) A c = 16.543 (4) A B= 101.20(2)° V = 2741.4 (9) A  Space Group  P2!/n  Z value  4  Dcaic  3.269 g/cm  F  000  2400  Diffractometer  Rigaku AFC 6S  Radiation  MoKa  H(MoKa)  192.86 cm"  Scan Rate  32.0°/min  No. of Reflections Measured  Total: 8661  3  3  1  Unique: 8301 (R  int  = 0.078)  Structure Solution  Patterson Method  Refinement  Full-matrix least-squares  Table 3.5.  Selected Bonds Lengths  (A) with Estimated Standard  Deviations for 47.  atom  atom ,  distance  atom  atom  distance  Os(l)  Os(2)  2.8576(7)  Fe(l)  C(3)  2.05(1)  Os(l)  Os(3)  2.8563(7)  Fe(l)  C(4)  2.06(1)  Os(l)  Os(4)  2.7192(7)  Fed)  C(5)  2.04(1)  Os(l)  S(l)  2.385(3)  Fed)  C(6)  2.01(1)  Os(l)  S(2)  2.428(3)  Fed)  C(7)  2.00(1)  Os(l)  S(3)  2.425(3)  Fe(l)  C(8)  2.05(1)  Os(l)  C(ll)  1.91(1)  Fed)  C(9)  2.06(1)  Os(l)  C(12)  1.91(1)  Fe(l)  C(10)  2.04(1)  Os(2)  Os(3)  2.7257(9)  Fed)  Cp(l)  1.645(6)  Os(2)  S(l)  2.341(3)  Fe(l)  Cp(2)  1.641(7)  Os(2)  C(12)  2.74(1)  S(2)  C(l)  1.76(1)  Os(2)  C(13)  1.87(1)  S(3)  C(6)  1.76(1)  Os(2)  C(14)  1.90(1)  0(1)  C(ll)  1.13(1)  Os( 2)  C(15)  1.91(1)  0(2)  C(12)  1.13(1)  Os(3)  S(l)  2.338(3)  0(3)  C(13)  1.15(1)  Os(3)  C(ll)  2.77(1)  0(4)  C(14)  1.16(2)  Os( 3)  C(16)  1.89(1)  0(5)  C(15)  1.13(1)  Os(3)  C(17)  1.90(1)  0(6)  C(16)  1.17(1)  Os(3)  C(18)  1.88(1)  0(7)  C(17)  1.16(2)  Os(4)  S(2)  2.410(3)  0(8)  C(18)  1.14(1)  Os( 4)  S(3)  2.413(3)  0(9)  C(19)  1.14(1)  Os(4)  C(19)  1.91(1)  0(10)  C(20)  1.18(1)  Os( 4 )  C(20)  1.89(1)  0(11)  C(21)  1.16(1)  59  Table 3.6. Selected Bond Angles (deg) with Estimated Standard Deviations for 47.  atom  atom  a ton  angle  atom  a ton  atom  06(2)  06(1)  06(3)  56.96(2)  S(3)  06(1)  C(12)  145.9(4)  06(2) 06(1) 08(4)  151.45(3)  C(ll)  06(1)  C(12)  87.8(5)  06(2)  06(1)  8(1)  52.10(7)  06(1)  0e(2)  Os(3)  61.48(2)  06(2)  06(1)  S(2)  101.25(7)  06(1) 06(2)  Sd)  53.49(6)  06(2)  06(1)  S(3)  147.43(7)  06(1) 06(2)  C(12)  39.9(2)  06(2)  06(1)  can  105.0(3)  06(1)  06(2)  C(13)  151.5(4)  06(2)  06(1)  C(12)  66.6(4)  06(1)  08(2)  C(14)  99.2(4)  06(3)  06(1)  06(4)  151.47(3)  05(1)  06(2)  C(15)  108.2(4)  06(3)  06(1)  S(l)  52.04(7)  06(3)  06(2)  S(l)  54.32(7)  06(3)  06(1)  S(2)  145.75(7)  06(3)  06(2)  C(12)  90.3(3)  06(3)  06(1)  S(3)  100.00(7)  06(3)  06(2)  C(13)  95.7(4)  06(3)  06(1)  C(ll)  67.9(3)  06(3)  05(2)  C(14)  152.2(4)  0e(3)  06(1)  C(12)  106.9(3)  0&(3)  Oe(2)  C(15)  102.7(4)  06(4)  06(1)  S(l)  136.12(6)  S(l)  0s(2)  C(12)  92.2(2)  06(4)  06(1)  S(2)  55.48(6)  S(l)  0&(2)  C(13)  100.1(4)  06(4)  06(1)  S(3)  55.59(7)  S(l)  0s(2)  C(14)  98.5(4)  06(4)  06(1)  C(ll)  92.6(3)'  S(l)  06(2)  C(15)  154.5(4)  06(4)  06(1)  C(12)  92.2(3)  C(12)  Oe(2)  C(13)  167.6(4)  S(l)  06(1)  S(2)  93.9(1)  C(12)  05(2)  C(14)  84.7(5)  S(l)  06(1)  S(3)  96.1(1)  C(12) 0&(2)  C(15)  75.5(5)  S(l)  06(1)  C(ll)  118.9(4)  C(13) 05(2)  C(14)  94.8(6)  Ed)  06(1)  C(12)  116.9(4)  C(13)  06(2)  C(15)  92.5(5)  S(2)  06(1)  S(3)  C(14) 0s(2)  C(15)  102.4(6)  S(2)  05(1)  C(ll)  146.2(3)  06(1)  06(3)  06(2)  61.53(2)  S(2)  06(1)  C(12)  83.4(4)  06(1)  06(3)  S(l)  53.54(6)  S(3)  06(1)  C(ll)  83.2(4)  06(1) 06(3)  C(ll)  39.6(3)  86.03(9)  angle  60  clusters. The deviation by 0.02-0.06 A of the Os-C bond lengths from the mean value of 1.93 A in Os3(CO)i2 are consistent with the assumption that bonds trans to S are shortened by ca. 0.025A. Such a trend was recognized in other osmium cluster 142  structures. Both C5H4S ligands are T| -bound to iron and are o bound to osmium through the 5  sulphur atoms. The two cyclopentadienyl rings are eclipsed and nearly parallel, the dihedral angle between the two rings being only 5.91°. This small tilt is close to that observed in the ruthenium complex 43 (5.38°). The two Cp rings are virtually planar, while the two sulphur atoms (S(2) and S(3)) are displaced outwards from the Cp ring planes by 0.1440 and 0.1546 A, respectively. The Fe-C ring and C-C bond distances, and the bond angles of the Cp rings are very similar to those found in the ruthenium complex 43. The C-S distances (C(l)S(2)=C(6)-S(3)=1.76(1) A) are also very similar to those found in 43 and are shorter 0  145  than a single C-S bond length (i.e. 1.81 A).  3.2.6. Cobalt and Manganese Derivatives of l,2,3-trithia-[3]ferrocenophane.  The reactions of l,2,3-trithia-[3]-ferrocenophane with stoichiometric amounts of Co2(CO)8 and Mn2(CO)io produced some unexpected results. The details of the reactions, description of products and estimated yields are summarized in Table 3.7. The reaction with Co2(CO)8 in hexanes and THF under a variety of conditions afforded insoluble, black solids (powder, needles or a foil) 51,53, 54 and 57 in yields varying from 60 to 80 %. The insoluble solids did not give mass spectra in either EI or FAB modes. Elemental analysis data for 54 and 57 indicate a C/H/S ratio which does not correspond to 61  any reasonable single product. Two low yield soluble compounds  Fe(r| -C5H4S)2Co2(CO)6 and Fe(r|-C5H45  5  S) Co4(CO)i2 (52 and 55)were isolated and identified from their *H NMR and EI or 2  FAB spectra. Compound 55 proved to be unstable in the solid state; its brown-greenish color changed to black in several days. It could not be obtained in an analytically pure form. Another soluble compound Fe(C5H4S)2Co (CO)8 56 isolated in trace quantity, 2  was identified from the parent ion in FAB spectrum. The IR spectrum of 56 shows absorption in the bridging and terminal D(C-O) region, thus 56 was formulated as shown below.  56  52  The reaction with Mn2(CO)io in THF gave an insoluble powder 58 in high yield (~55%) and a soluble compound 59 in moderate yield which produced two broad peaks in the *H NMR spectrum. The broad peaks suggests this compound is paramagnetic. The highest peaks observed in the EI and FAB spectra of 58 and 59 correspond to (C5H S)2Mn(S)(CO) + and (C5H S)2Mn+, and are likely to be fragments of this 4  2  4  paramagnetic compound.  62  Table 3.7. Reactions of l,2,3-Trithia-[3]-Ferrocenophane with Co and Mn Carbonyls.  Reagent  Reaction Conditions Solvent  Co2(CO)  8  Hexane  Time  Temp.  (h)  (°C)  20  RT  Product Description, MS and 1H NMR Data,  Black-greenish powder 51 (yield~60 %); insoluble.  1 compound 52 in trace quantity. (Fe(C5H4S) Co2(CO)6 2  C02(C0)  8  Reflux  Hexane  Black-greenish powder 53 (yield -80 %); insoluble.  Red oil 52 (yield-10%) Fe(C H4S)2C02(C0) 5  Co2(CO)  8  THF  6  2  RT  Black needles 54 (yield -60 %);  8  50  insoluble; m.p. > 373° C  Brown-greenish powder 55 (yield -10 %); unstable Fe(C H4S) C04(C0)i2 5  2  Yellow-brown powder 56 63  Table 3.7. contd.  C02(CO)  Mn (CO)i 2  THF  8  0  THF  5  50  1  Reflux  14  RT  20  RT  Black foil 57 (yield-85 %); insoluble.  Brown fine powder 58 (yield -55%); insoluble.  1 orange compound 59 (yield -20 %).  3.3. PREPARATIONS OF ( NjN-DIMETHYLAMINOMETHYL)- AND (DIPHENYLPHOSPHINO)-l,2,3-TRITHIA-[3]-FERROCENOPHANES.  (A) NjN-Dimethylaminomethyl-1 il ,3-trithia-[3J-ferrocenophane (DMAMTF). 146  D M A M T F 61 was first prepared by Butler in these laboratories.  The  preparative route involves the reaction of the dilithio derivative of (N,N(Umethylaminomethyl)ferrocene with Ss (Scheme 3.4.). Previous studies * 129  146  showed that lithiation of an aminosubstituted ferrocene is  almost exclusively directed to the ring position adjacent to the substituent, while the lithiation of an unsubstituted cyclopentadienyl ring is random. The reaction of the dilithiated derivative of dimethylaminomethylferrocene 60 with sulphur afforded DMAMTF which was isolated from an oily mixture by chromatography in a reasonable yield (40%). Extraction (with petroleum ether and methylene chloride) and crystallization as indicated in the previous work  146  afforded only a modest yield (-20%). 64  CH NMe 2  Fe  -  CH NMe 2  2  n-BuLi/TMEDA  2  Fe ^—IifTMEDA)  60  61  Scheme 3.4.  (B) Diphenylphosphino-sulfido-1,2,3-trithia-[3]-ferrocenophane (DPPTF). DPPTF wasfirstprepared in these laboratories from the reaction of the dilithiated 147  derivative of diphenylphosphinoferrocene 62 and sulphur as outlined in Scheme 3.5. The two isomers 63a and 63b were obtained in a ratio 63a:63b=~3:l. Both isomers showed clear parent ions in their mass spectra, and each isomer exists in the form of two diastereomers in solution as shown by the *H NMR spectrum of 63a. In the present investigation the yield was very low (a total of the two isomers of ~10-12%). This was due to the formation of ferrocene coupled products such as 64 (M > +  834), 65 ( M , 770) and compounds which contain more than two ferrocenyl groups, +  presumably obtained from the coupling of the dilithiated derivatives of diphenylphosphinoferrocene with sulphur. The formation of 64 and 65 was favored by the monolithioferrocene obtained in higher yield than reported in previous \0f\ 10il  investigations  *  when a twofold excess of n-butyllithium is used in the presence of  an equivalent amount of TMEDA. Due to the low yield, DPPTF was not employed as ligand towards metal carbonyls.  65  Scheme 3.5.  66  3.4. REACTIONS OF  (NJN-DIMETHYLAMINOMETHYL)-I,2,3-TRITHIA-[3]-  FERROCENOPHANE.  Reactions of DMAMTF with ruthenium and iron carbonyls, and cyclooctadiene and cyclooctene rhodium and palladium chloride complexes were investigated. The reactions were carried out using thermal or catalytical procedures. The details of the reaction conditions, description of products and estimated yields are summarized in Table 3.8. The reactions afforded oils or solids insoluble in common organic solvents. Their yields varied from 20 to 90%. No compounds could be isolated from these oils and solids. With only one exception (68) the solids did not ionize in EI or FAB, thus no mass spectral data could be obtained. The brown solid 68 and the oil 66 were soluble in DMSO. Their H NMR spectra in d6-DMSO showed a broad peak which suggests that 66 l  and 66 are paramagnetic. The peak at 538 exhibited by the FAB spectrum of 68 corresponds to Fe(C5H4S2)(C5H3CH2NMe2)Ru(CO) , a fragment which indicates the +  formation of a ruthenium carbonyl derivative of DMAMTF. It could not be established if this derivative is a major component of 68 or only a compound in a trace amount. The analytical data of 68,69,73 and 74 do not correspond to any reasonable single product. A soluble iron derivative isolated in trace quantity was identified from the fragments  exhibited  by  the  EI  mass  spectrum  as  [Fe(C5H4S2)(C5H CH2NMe2)]2Fe2(CO) . 3  6  There are no reports on derivatives of ring substituted S3-ferrocenophanes so far, thus the reactivity study conducted could not be related to any previous results.  67  Table 3.8. Reactions of D M A M T F .  Reagent  Reaction Conditions Solvent  Ru3(CO)i2  Hexanes  Product Description  Time  Temp.  (h)  (°C)  22  reflux  Cat.  Dark brown oil 66 (yield -80 %); soluble in DMSO, no isolable.  Ru3(CO)i2  Cyclohexane 20  reflux  Dark brown oil 67, no isolable.  Ru (CO)i 3  2  THF  18  RT  (PPh ) NCl Brown fine powder 68, 3  2  (yield -90 %); soluble in DMSO.  Ru3(CO)i2  THF  4  RT  (PPh3)2NCl Brown fine powder 69 (yield -40 %); no isolable.  Fe2(CO)9  THF  20  RT  Dark-brown powder 70 (yield - 20%); insoluble.  Brown oil 71, partly isolable. [Fe(C5H4S )(C5H3CH NMe2)]2Fe2(CO)6  +  2  2  (72) 68  Table 3 8. contd.  [Rh(COD)Cl] THF 2  [Rh(COE)Cl ]2 THF 2  35  RT  Black powder 73 (yield  4  40  -25 %); insoluble.  75  RT  Black fine powder 74 , (yield ~ 90 %); insoluble.  (min.)  Pd(COD)Cl  2  THF  20  RT  Dark-brown powder 75, (yield -30 %); insoluble.  69  SUMMARY  The number of previously known (ferrocenedithiolato-S,S') transition metal complexes was extended in this work to include ruthenium and osmium derivatives. Thermal reaction of l,2,3-trithia-[3]-ferrocenophane with Ru3(CO)i2 in refluxing hexanes affords a ruthenium dinuclear complex FeOi^Cs^S^Ri^tCCOg in moderate yield (35 %). l,2,3-trithia-[3]-ferrocenophane does not react with Os3(CO)i in refluxing hexanes, 2  but yields Fe(r| -C5H4S)2(S)Os2(CO)6 in low yield (10 %) in refluxing cyclohexane.and 5  four products in moderate (25 %) to low (8 %) yields in refluxing toluene. These compounds were identified as the dinuclear species FeCn^-CsF^S^G^tCO^ the trinuclear species Fe(r) -C5H4S)20s (CO)io and the tetranuclear species Fe(r| 5  5  3  C H4S)2(a3-S)(ii-CO)20s (CO)9 and Fe(Ti -C H S) (li3-S)(M--S)Os4(CO) o. 5  5  4  5  4  2  1  The mechanistic pathway towards the formation of the ruthenium and osmium derivatives involves the opening of the ferrocenophane system. In addition, the formation of (l,l'-ferrocenedithiolato-S,S')(iiruthenium hexacarbonyl involves the loss of a sulphur atom and breakdown of the ruthenium cluster, while the formation of the ferrocenedithiolato derivatives of osmium carbonyl occurs with either the breakdown of the triosmium cluster, the opening of the cluster or retention of the triosmium cluster and attachment of one Os(CO)3 unit. The formation of the ferrocenedithiolato derivatives also involves the loss of a S atom or the gain of one or two sulphur atoms. The last case leads to the formation of Fe(Ti -C H S) (H3-S)(p:-CO)20s4(CO)9 and FeCn-C5HS)2(p:5  5  5  4  2  4  3  S)((p>S)Os4(CO) 1Q which contain sulfido ligands in addition to the SC5H4 ligands. l,2,3-Trithia-[3]-ferrocenophane was employed as an unusual source of sulphur in the preparation of these sulfido-osmium cluster compounds which may have catalytic activity for reactions such as olefin hydrogenation and hydrogenolysis of molecules with carbonsulphur, carbon-oxygen and carbon-nitrogen bonds. It is known that a major problem in 70  the use of cluster compounds as catalysts is their tendency to undergo degradative fragmentation when subjected to forcing reaction conditions. The sulfido bridges in addition to the SC5H4 bridges of the above ferrocenedithiolato osmium complexes can play an important role in stabilizing them against fragmentation processes and thus enhance their catalytic applications. The formulation of the ferrocenedithiolato ruthenium and osmium carbonyls as coordinatively 18-electron complexes is supported by their spectroscopic properties. In addition, this formulation was confirmed by a single-crystal X-ray crystallographic analysis for Fe(r| -C5H S)2Ru2(CO) and Fe(Ti -C H4S)2(U3-S)(u\-CO)20s (CO)9. The 5  5  4  6  5  4  two structures have eclipsed and nearly parallel cyclopentadienyl rings, the dihedral angles being only 5.38° in the ruthenium derivative and 5.91° in the osmium derivative. Both SC5H4  ligands are T] -bound in a pseudo-trans configuration to iron, much as in ferrocene 5  and its derivatives and are a bound to mthenium or osmium through the sulphur atoms. In each of the two compounds characterized by X-ray analysis, the SC5H4 groups bridge two metal atoms and each S atom acts as a three-electron donor. Compound Fe(r| 5  C H4S)2(U3-S)(u.-CO)20s4(CO) has a sulfido ligand absent in Fe(Ti -C5H4S) Ru (CO) 5  5  9  2  2  6  which serves as a bridge to three metal atoms and acts as a four-electron donor. The strain imposed by the bridging SC5H4 ligands in FeCn^CsIfyS^R^CCCOg results in a very distorted octahedral coordination about the ruthenium atoms with angles which differ by 28.1 and 30.4° from the ideal 180°. In  (l,r-ferroceneditholato-S,S')tetraosmium  carbonyl one osmium atom is eight coordinated while the geometry around the other three osmium atoms is distorted octahedral with deviations varying between 18.0 and 31.1° from the ideal 180°. One more osmium derivative, Fe(rj C5H4S)2(M-3-S)(p:-S)Os4(CO)io. contains the 5-  sulfido ligand which shows the ability to serve as a multicoordinate, multielectron bridging ligand. This versatile coordination behavior has been utilized in the development of synthetic routes to metal-sulphur cluster compounds with varied structural  71  features. 16,117.138 1  Reactions of 1,2,3-trithia [3]-ferrocenophane with Co2(CO)g and Mn (CO)io 2  affords insoluble solids that were not identified, in high yields varying from 55 to 80 %. Elemental analysis data for these solids indicate a C/H or C/H/S ratio which does not correspond to any reasonable single product. Two low yield soluble cobalt derivatives Fe(Ti -C H4S)2Co2(CO)6 and Fe0l -C H4S) Co4(CO) , and a paramagnetic manganese 5  5  5  5  2  12  derivative were also isolated from these reactions. The ability of (N,N- dimethylaminomethyl)-l,2,3-trithia-[3]-ferrocenophane (DMAMTF) to react with ruthenium and iron carbonyls, and cyclooctadiene and cyclooctene chloride rhodium and palladium complexes was also explored in this work. The reaction chemistry of DMAMTF appears to be a rewarding area for study since a ring substituent in addition to the sulphur atoms is available for coordination. Contrary to the expectation, the reaction of DMAMTF with Ru (CO) , Fe (CO) , [Rh(COD)Cl] , 3  12  2  9  2  [Rh(COE) Q] and Pd(COD)Cl under a variety of conditions affords oils and solids, 2  2  2  which were not identified, in yields varying from 20 to 90 %. These products are insoluble in common organic solvents, and the microanalytical data indicate that the products are not analytically pure compounds. A soluble iron derivative isolated in trace quantity was identified from the fragments exhibited by the EI mass spectrum, as  [Fe(C H4S2)(C H CH NMe2)]2Fe2(CO) . 5  5  3  2  6  72  FUTURE STUDIES  As it has been shown l,2,3-trithia-[3]-ferrocenophane reacts readily with Ru3(CO)!2 and Os3(CO)i2 to give soluble ferrocenedithiolato derivatives, while with Co2(CO)g and Mn (CO) o affords insoluble solids (needles, powder or foil) which were 2  1  not identified, in high yields. The identity of these products remains to be ascertained. It is hoped that the ruthenium and osmium compounds prepared in this work may have catalytic activity for reactions such as olefin hydrogenation. Particularly, ferrocenedithiolato osmium derivatives with one or more sulfido ligands in addition to the SC5H4 groups are expected to display catalytic activity for hydrogenations of unsaturated and aromatic hydrocarbons, hydrogenation of N=N bonds in azo compounds and hydrogenolyses of molecules with carbon-sulphur, carbon-oxygen, and carbon-nitrogen bonds. There are numerous reports on sulfido-transition metal complexes used as catalysts of the above reactions.  118  Investigations to determine whether ruthenium and osmium ferrocenedithiolato derivatives isolated and characterized in this work exhibit such catalytic activity are of particular interest.  73  BIBLIOGRAPHY  1.  Schloegl, K.; Falk, H. Method. Chim. 1976, 8, 469.  2.  Watts, W.E. Organomet. Chem. Rev. 1967, 2, 231.  3.  Mueller-Westerhoff, U.T. Angew. Chem. Int. Ed. Engl. 1986, 25, 702  4.  Smith, H.B. Bridged Aromatic Compounds, Academic Press, New York, 1964.  5.  Watts, W.E. / . Am. Chem. Soc. 1966, 88, 855.  6.  Watts, W.E. / . Organomet. Chem. 1967, 10, 191.  7.  Katz, T.J.; Acton, N.; Martin, J. J. Am. Chem. Soc. 1969, 91, 2804.  8.  Osborne, A.G.; Whiteley, R.H. / . Organomet. Chem. 1975, 101, C27.  9.  Fisher, A.B.; Kinney, J.B.; Staley, R.H.; Wrigton, M.S. / . Am. Chem. Soc. 1979, 101, 6501.  10.  Osborne, A.G.; Whiteley, R.H.; Meads, R.E. / . Organomet. Chem. 1980, 193, 345.  11.  Stoeckly-Evans, H.; Osborne, A.G.; Whiteley, R.H. Helv. Chim. Acta 1976, 59, 2402.  12.  Seyferth, D.; Withers, H.P. / . Organomet. Chem. 1980, 185, CI.  13a.  Seyferth, D.; Withers, H.P. Organometallics 1982,1,1275.  13b.  Withers, H.P.; Seyferth, D.; Felmann, J.D. Organometallics 1982,1, 1283.  14.  Butler, I.R.; Cullen, W.R.; Kim, T.J.; Rettig, S.T.; Trotter, J. Organometallics 1985, 4, 972.  15.  Rinehart, Jr., K.L.; Fredrichs, A.K.; Kittle, P.A.; Westman, L.F.; McMahon, J.E. J. Am. Chem. Soc. 1960, 82, 4111.  16.  Rinehart, K.L.; Curby, R.J.; Sokol, P.E. / . Am. Chem. Soc. 1957, 99, 3420.  17.  Schlogl, K.; Seiler, H. Tetrahedron Lett. 1960,7,4.  74  18.  Rosenblum, M.; Banerjee, A.K.; Danieli, N.; Fish, R.V.; Schlatter, V. / . Am. Chem. Soc. 1963, 85, 316.  19.  Hisatome, M.; Watanabe, N.; Sakamoto, T.; Yama-Kawa, K. / . Organomet. Chem. 1977, 125, 79.  20.  Hillman, M.; Fujita, I.; Dauplaise, H.; Kerber, R.C. Organometallics 1984, 3, 1170.  21.  Davison, A.; Smart, J.C. J. Organomet. Chem. 1969, 19, P7.  22.  Bishop, J.J.; Davison, A.; Klatcher, M.L.; Lichtenberg, D.W.; Merrill, R.E.; Smart, J.C. J. Organomet. Chem. 1971, 27, 241.  23.  Hollands, R.E.; Osborne, A.G.; Townsend, I. Inorg. Chim. Acta 1979, 37, L541.  24.  Osborne, A.G.; Hollands, R.E.; Howard, J.A.K. J. Organomet. Chem. 1981, 205, 395.  25.  Barr, T.H.; Watts, W.E. Tetrahedron 1968, 24, 6111.  26  Davison, A.; Smart, J.C. J. Organomet. Chem. 1979, 174, 321.  27.  Abel, E.W.; Booth, M.; Orrel, K.G. J. Organomet. Chem. 1980, 186, C37.  28.  Abel, E.W.; Booth, M.; Orrel, K.G. J. Organomet. Chem. 1981, 208, 213.  29.  Abel, E.W.; Booth, M.; Brown, C.A.; Orrel. K.G.; Woodford, R.L. / . Organomet. Chem. 1981, 214, 93.  30.  Abel, E.W.; Booth, M.; Orrel, K.G. J. Organomet. Chem. 1978, 160, 75.  31.  Abel, E.W.; Booth, M.; Orrel, K.G. J. Chem. Soc. Dalton 1980, 1587.  32.  Luttringhaus, A.; Kullich, W. Angew. Chem. 1958, 70, 438  33.  Tanner, D.; Wennerstroem, O. Acta Chem. Scand. 1980, 4, 529.  34.  Akabori, S.; Habata, Y.; Sato, M.; Ebine, S. Bull. Chem. Soc. Jpn. 1983, 56, 537.  35.  Akabori, S.; Fakuda, H.; Habata, Y.; Sato.M.; Ebine, S. Chem. Lett. 1982, 1393. 75  36.  Biernat, J.F.; Wilczewski, T. Tetrahedron 1980, 36, 2521.  37.  Akabory, S.; Habata, Y.; Sato, M. Bull. Chem. Soc. Jpn. 1984, 57, 68.  38.  Sato, M.; Watanabe, H.; Ebine, S.; Akabory, S. Chem. Lett. 1982, 1753.  39.  Sato, M.; Tanaka, S.; Morinaga, K.; Akabori, S. / . Organomet. Chem. 1985, 282, 247.  40.  Vondrak, T.; Sato, M. / . Organomet. Chem. 1989, 364, 207.  41.  Sato, M.; Kubo, M.; Ebine, S.; Akabory, S. Tetrahadron Lett. 1982, 23, 2717.  42.  Akabori, S.; Habata, Y.; Sato, M.; Ebine, S. Bull. Chem. Soc. Jpn. 1983, 56, 1459.  43.  Akabori, S.; Shibahara, S.; Habata, Y.; Sato, M. Bull. Chem. Soc. Jpn. 1984, 57, 63.  44.  Bellon, P.; Demartin, F.; Scatturin, V.; Czech, B. / . Organomet. Chem. 1984, 265, 65.  45.  Lehn, J.M. Structure and Bonding 1973,16, 1.  46.  Bernal, I.; Raabe, E.; Reisner, G.M.; Holwerda, R.R.; Czech, B.P.; Huang, Z. Organometallics 1988, 7,247.  47.  Bernal, L.; Reisner, G.M.; Hokwerda, R.A.; Czech, B.P. Organometallics 1988, 7, 253.  48.  Rinehart, K.J.; Bublitz, D.E.; Gustafson, D.H. / . Am. Chem. Soc 1980, 97, 1347  49.  Brown, A.D.; Winstead, J. A. / . Org. Chem. 1971, 36, 2832.  50.  Hisatome, M.; Watanabe, N.; Saksmoto, T.; Yamakawa, K. / . Organomet. Chem. 1977, 125, 79.  51.  Hillman, M,; Gordon, B.; Weiss, A.G.; Guzikowski, A.P. / . Organomet. Chem. 1978, 155, 77.  52.  Hisatome, M.; Hilman, M. / . Organomet. Chem. 1981, 212, 217.  76  53.  Hisatome, M.; Watanabe, J.; Yamakawa, K.; Iitaka, Y. / . Am. Chem. Soc. 1986, 108, 1333.  54.  Hisatome, M.; Watanabe, J.; Kawajiri, Y.; Yamakawa, K.; Iitaka, Y, Organometallics 1990, 9,497.  55.  Bishop, J.J.; Davidson, A. Inorg. Chem. 1971, 10, 826.  56.  Bishop, J.J.; Davidson, A. Inorg. Chem. 1971, 10, 832.  57.  Osborne, A.G.; Hollands, R.E.; Bryan, R.F.; Lockhardi, S.S. / . Organomet. Chem. 1985, 288, 207.  58.  Sato, M.; Tanaka, S.; Ebine, S.; Marinaga, K.; Akabori, S. Bull. Chem. Soc. Jpn. 1985, 58, 1915.  59.  Seyferth, D.; Hames, B.W.; Rucker, T . C ; Cowie, M.; Dickson, R.S. / . Organomet. Chem. 1987, 326, 269.  60.  Fest, D.; Habben, C D . / . Organomet. Chem. 1990, 390, 339.  61.  Osborne, A.G.; Hollands, R.E.; Nagy, A.G. / . Organomet. Chem. 1989, 373, 229.  62.  Seyferth, D.; Hames, B. Inorg. Chim. Acta 1983, 77, LI.  63.  Nametkin, N.S.; Tyurin, V.D.; Sleptosova, S.A.; Sideridu, A.Ya. Izv. Akad. Nauk SSSR, Ser. Khim. 1982, 955.  64.  Hayter, R.G.; Humiec, F.S. J. Inorg. Nucl. Chem. 1969, 26, 807.  65.  Davison, A.; Edelstein, N.; Holm, R.H.; Maki, A.H. / . Organomet. Chem. 1964, 3, 817.  66.  Zanella, R.; Ros, R.; Graziani, M. Inorg. Chem. 1973, 12, 2737.  67.  Lam, C.T.; Senoff, C V . Can. J. Chem. 1972, 50, 1868.  68.  Markham, S.J.; Chungh, Y.L.; Blake, D.M. / . Organomet. Chem. 1976, 107, 121.  69.  Teo, B.K.; Windl,F.; Marshal, J.H.; Kruger, R. / . Am. Chem. Soc. 1977, 99, 2349. 77  70.  Gal, A.W.; Gosselink, J.W.; Vollenbroek, F.A. Inorg. Chim. Acta 1979, 32, 235.  71.  Seyferth, D.; Henderson, R.S. / . Am. Chem. Soc. 1979, 101, 508.  72.  Seyferth, D.; Henderson, R.S.; Long, L.S.  73.  Morandini, F.; Consiglio, G.; Wenziger, F. Helv. Chim. Acta 1979, 62, 59.  74.  Seyferth, D.; Henderson, R.S.; Gallager, M. K. / . Organomet. Chem. 1980,  Am. Chem. Soc. 1980, 192, CI.  193, C75. 75.  Seyferth, D.; Hames, B.W.; Rucker, T.G. Organometallics 1983, 2,472.  76.  Watanabe, M.; Ichikawa, H.; Motoyama, I.; Sano, H. Bull. Chem. Soc. Jpn. 1983, 56, 3291.  77.  Whitesides, G.; Gaash, J.F.; Stedronski, E.S. / . Am. Chem. Soc. 1972, 94, 5258.  78.  McCulloch, B.M.; Ward, D.L.; Woolins, J.D.; Birnbak, C H . Organometallics 1985, 4, 1425.  79.  Akabori, S.; Kumagai, T.; Shirahige, T. Organometallics 1987, 6, 526.  80.  Akabori, S.; Kumagai, T ; Shirahige„T ; Sato, S.; Kowazoe, K.; Tamura, S.; Sato, M. Organometallics 1987,6,2105.  81.  Pauling, M . The Nature of the Chemical Bond; Cornwell University Press: New York, 1960.  82.  Sato, M; Sekino, M.; Akabori.S. / . Organomet. Chem. 1988, 344, C31.  83.  Sato, M.; Sekino, M.; Katada, M.; Akabori, S. / . Organom. Chem. 1989, 377, 327.  84.  Sato, M.; Suzuki, K.; Akabori.S. Chem. Lett. 1987, 2239.  85.  Dunitz, J.D.; Orgel, L.E.; Rich, A. Acta Cryst. 1950, 9, 373.  86.  Takusagawa, F; Koetzle, T.F. Acta Cryst. Sect. B. 1979, B35, 1074.  87.  Seiler, P.; Dunitz, J.D. Acta Cryst. Sect. B 1979, B35, 1068.  78  88.  Stoeckli-Evans, H.; Osborne, A.G.; Whiteley, R.H. / . Organomet. Chem. 1980, 194, 91.  89.  Yasufuku, K.; Aoki, K.; Yamazaki, H. Inorg. Chem. 1977, 16, 624.  90.  Davis, B.R.; BernalJ J. Cryst. Mol Struct. 1972, 2, 107.  91.  Pierpoint, C.G.; Eisenberg, R. Inorg. Chem. 1972, 11, 828.  92.  Jones, N.D.; Marsh, R.E.; Richards, J.H. Acta Cryst. 1965, 19, 330.  93.  Lecomte, C.; Dusausov, Y.; Moise, C.; Tirouflet, J. Acta CrysB 1973, 29, 488.  94.  Batail, P.; Grandjeau, D.; Astruc, D.; Dabard, R. / . Organomet. Chem. 1975, 14, 3085.  95.  Lecomte, C.; Dusausov, Protas, J.; Moise, C. Acta CrysB 1973, 29, 1177.  96.  Abramovitch, R. A.; Atwood, J. L.; Good, M . L. Inorg. Chem. 1975, 14, 3085.  97.  Laining, M. B.; Trueblood, K. N. Lecomte, G ; Dusausov, Y.; Moise, C.; Tirouflet, J. Acta Cryst. 1965, 19, 373.  98.  Churchill, M. R.; Liu, K. G. Inorg. Chem. 1973, 12, 2274.  99.  Daly, J. J. / . Chem. Soc. 1964, 3799.  100.  Weaver, J.; Woodward, P. / . Chem. Soc. Dalton.1973, 1439.  101.  Paton, W. F.; Corey, E. R.; Gluck, M . D.; Mislow, K. Acta Crys.B. 1977, 33, 268.  102.  Karipides, A.; Haller, D.A. Acta CrysB. 1972, 28, 2889.  103.  Ricci, J. S.; Bernal, I. / . Am. Chem. Soc. 1969, 91, 4078.  104.  Ricci, J. S.; Bernal, I. J. Chem. Soc. 1971, 1928.  105.  Bailey, P. M.; Taylor, S. H.; Maitlis, P. M. Ricci, J. S.; Bernal, I. / . Am. Chem. Soc. 1978, 100, 471.  106.  Roundhill, D. M.; Beaulieu, W. B.; Bagchi, U. Inorg. Chem. 1980, 19, 3365.  107.  Kai, Y.; Yasuoka, N.; Kasai, N. Bull. Chem. Soc. Jpn. 1979, 52, 737.  108.  Del Piero, G.; Cesari, M. Acta Cryst. Sect. B 1979, B35, 2411.  109.  Hayashi, T.; Konishi, N.; Kumada, M. Tetrahedron Lett. 1979, 21, 1871. 79  110.  Cullen, W.R.; Woolins, J.D. Coord. Chem. Rev. 1981,39.  111.  Hayashi, T.; Konish, M.; Kumada, M.; Higachi.T.; Hirotsu, K. / . Am. Chem. Soc. 1984, 106, 158.  112.  Hayashi, T.; Kabeta, K.; Kumada, M. Tetrahedron Lett. 1984, 25, 1499.  113.  McCulloch, B.; Ward, D.L.; Woolins, J.D.; Brubaker, C H . Organometallics 1985, 4, 1425.  114.  Vakrenkamp, H. Angew. Chem. Int. Ed. 1975, 14, 322.  115.  Day, V.W.; Lesch, D.A.; Rauchfuss, T.B. / . Am. Chem. Soc. 1982, 104, 1290.  116.  Muller, A. Polyhedron 1986,5,323.  117.  Wachter, J. / . Coord. Chem. 1987, 15, 219.  118.  Dubois, M.R. Chem. Rev. 1989, 89, 1.  119.  Ziemmerman, R.; Munch, E.; Brill, W.J.; Shah, V.K.; Henzl, M.T.; OrmJohnson, W.H. Biochim. Biophys. Acta 1978, 537, 185.  120.  Weisser, O.; Landa, S. Sulfide Catalysts, Their Properties and Applications; Pergamon. Press: New york, 1973.  121.  Wrighton, M.S. J. Am. Chem. Soc. 1980, 102, 6898.  122.  Shriver, D.F. The Manipulations of Air Sensitive Compounds; McGraw-Hill: New York, 1969.  123.  Bruce, M. I.; Matisons, J. G.; Wallis, R. C ; Patrick, J. M.; Skelton, B. W.;, White, A. H. / . Chem. Soc. Dalton Trans. 1983, 2365.  124.  Raush, M.D.; Ciappenelli, D.J. / . Organomet. Chem. 1967, 10, 127.  125.  Slocum, D.W.; Sugarman, D.I. In Polyamine-Chelated Alkali Metal Compounds; Langer, A.W. Ed.; ACS Publications 130; Am. Chem. Soc: Washington, D.C., 1974, p.222.  126.  Raush, M.D.; Moser, G.A.; Meade, C.F. J. Organomet. Chem. 1973, 51, 1.  127.  Walczak, M.; Walczak, K.; Mink, R.; Raush, M.D.; Stucky, G. / . Am. Chem. Soc. 1978, 100, 6382.  128.  Butler, I.; Cullen, W.R.; Ni, J.J.; Rettig, S.J. Organometallics 1985, 4, 2196.  129.  Butler, I.; Cullen, W.R.; Herring, F.G. Jagannathan, N.R. Can. J. Chem. 1986, 64, 667.  130.  Sherlock, S.J.; Cowwie, M.; Singleton,E.; Steyn, M.M. / . Organomet. Chem. 1989, 361, 353.  131.  Bianchi, M.; Menchi, G.; Francalanci, F.; Piacenti, F. J. Organomet. Chem. 1980, 188, 109.  132.  Jeanin, S.; Jeanin, Y.; Lavigne, G. Trans. Met. Chem. 1976, 1, 186.  133.  Colombie, A.; Savigne, G.; Bonnett, J.J. / . Chem. Soc. Dalton Trans. 1986, 899.  134.  Adams, R.D.; Chen, G.; Tanner, G.T.; Jianguo, Y. Organometallics 1990,9, 597.  135.  Davies, D.L.; Knox, S.A.; Mead, K.A.; Morris, M. J.; Woodward, PJ. / . Chem. Soc, Dalton. Trans. 1984, 2293.  136.  Vrieze, K. J. Organomet. Chem. 1986, 300, 307.  137.  Andrew, P.L.; Cabeza, J.A.; Riera, V.; Robert, F.; Jeannin, Y. / . Organomet. Chem. 1989, 372, C15.  138.  Adams, R.D. Polyhedron 1985, 4, 2003.  139.  Deeming, A.J. Advances in Organometallic Chemistry, 26,1.  140.  Churchill, M.R.; De Boer, B.G. Inorg. Chem. 1977, 16, 878.  141.  Adams, R.D.; Horvath, I.; SEgmuller, B.E.; Yang, L-W. Organometallics 1983, 2, 145.  142.  Johnson, B.F.G.; Lewis, J.; Pippard, D.; Rouse, D.K. / . Chem. Soc, Dalton. Trans. 1979, 61.  143.  Adams, R.D.; Foust, D.F. Organometallics 1983,2,323. 81  144.  And, H.G.; Kwick, W.L.; Leong, W.K.; Johnson, BF.G.; Lewis, G.; Raithby, P.R. / . Organomet. Chem. 1990, 396, C43.  145.  Adams, R.D. Polyhedron 1985,4,2003.  146.  Butler, I. R.; Cullen, W.R. Can. J. Chem. 1989, 67, 1851.  147.  Sollot, G.; Mertway, H.E.; Portnoy, S.; Sneed, J.L. / . Organomet. Chem. 1963, 28, 1090.  148.  Butler, I.R.; Cullen, W.R.; Organometallics 1986, 5,2537.  82  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items