Seven-coordinate halocarbonyl complexes of the type [MXY(CO)3(NCMe)2] (M = Mo W; X Y = halide pseudo halide) as highly versatile starting materials Paul K. Baker Department of Chemistry University of Wales Bangor Gwynedd UK LL57 2UW E-mail CHS018@BANGOR.AC.UK ABCDEF}2 dimer The fortuitous discovery of the seven-coordinate complexes [MI2(CO)3(NCMe)2] (M = Mo or W) is described; the wideranging chemistry of these compounds is discussed together with the use of the effective atomic number rule in predicting their reactions. The four structural types of seven-coordinate halocarbonyl complexes derived from [MI2(CO)3(NCMe)2] are described. The applications of the [MXY(CO)3(NCMe)2] (X = Y = Br I; X = Cl Br Y = I; X = Cl Y = GeCl3 SnCl3) system in both alkene metathesis catalysis and nitrogen fixation are also discussed.The synthesis of organometallic complexes containing six different ligands together with the preparation and stereochemical properties of the first structurally characterised {M [Cl(CO)(PPh3)(h2-MeC2Me)Mo(m-NCS)(m-SCN)MoCl- (CO)(PPh3)(h2-MeC2Me)] are described. 1 Introduction The availability of suitable starting materials has always been of considerable importance to the synthetic chemist. Since the first halocarbonyl donor ligand complex of molybdenum(ii) namely [MoI2(CO)2(diars)] {diars = C6H4(AsMe2)2-1,2} was prepared by Nigam and Nyholm in 1957 using reaction (1),1 six- and Paul K. Baker (born Isleworth Middlesex 1954) obtained his BSc (1976) and PhD (1979) from the University of Bristol.His thesis entitled ‘The Synthesis and Reactions of Cationic Organotransition-metal Complexes’ was obtained under the guidance of Neil G. Connelly and Michael Green. He held temporary Lectureships at the University of Strathclyde (1980) and the University of Manchester (1981) followed by a Lectureship in Chemistry at Thames Polytechnic (now the University of Greenwich). In 1984 he was appointed to a Lectureship in Chemistry at the University of Wales Bangor and was promoted to Senior Lecturer in 1989 and subsequently to a Readership in 1994. He is co-author of over 180 research publications mainly in the field of the organometallic chemistry of molybdenum and tungsten. He has given lectures on his research in over 40 UK Universities and Industrial Companies and he has lectured extensively in Australia Austria Belgium France Germany Holland Hong Kong Hungary Italy Poland Portugal South Africa Spain and the USA on various aspects of his research work.He is a Fellow of the Royal Society of Chemistry. In July 1997 the author was awarded the Teaching Fellowship for the Faculty of Science at the University of Wales Bangor. seven-coordinate halocarbonyl complexes have received considerable attention.2 [Mo(CO)4(diars)] + I2?[MoI2(CO)2(diars)] + 2CO (1) In the early days the two most commonly used routes to seven-coordinate halocarbonyl complexes were (i) the oxidation of complexes of molybdenum(0) or tungsten(0) containing strong field ligands and (ii) reactions of the reactive halobridged dimers [{M(m-X)X(CO)4}2] (M = Mo W; X = Cl Br I) with strong field ligands.An early example of type (i) is given in reaction (2).3 [M(CO)4(2,5-dithiahexane)] + X2? [MX2(CO)3(2,5-dithiahexane)] + CO (2) M = Mo W; X = Br I In 1966 Colton and Tomkins4 described the synthesis of the chloro-bridged dimer [{Mo(m-Cl)Cl(CO)4}2] [reaction (3)]. 278 °C 2[Mo(CO)6] + 2Cl2 ——–? [{Mo(m-Cl)Cl(CO)4}2] + 4CO (3) From 1966 to 1971 Colton et al. investigated the chemistry 4}2] (M = Mo W; X = Cl Br I) with of [{M(m-X)X(CO) strong field ligands to give a range of six- and seven-coordinate complexes of molybdenum(ii) and tungsten(ii) and some of their results are published in a review article.5 The molecular structure of the tungsten bromo-bridged dimer [{W(m- Br)Br(CO)4}2] has been crystallographically determined by Cotton et al.,6 and shows each tungsten atom to be in a capped octahedral environment with a carbonyl group capping an octahedral face (Fig.1). In 1972 Westland and Muriithi7 described the reactions of the halo-bridged dimers [{Mo(m- X)X(CO)4}2] (X = Cl Br) with weaker field ligands such as pyridine tetrahydrofuran and acetonitrile which eventually gave the non-carbonyl containing products [MoX3L3] and zerovalent [Mo(CO)6] via disproportionation of the six-coordinate molybdenum(ii) intermediates [MoX2(CO)2L2]. They were apparently unable to isolate these intermediates in a pure state. In 1962 Tate Knipple and Augl8 published the seminal paper on the synthesis of the important starting materials fac- [M(CO)3(NCMe)3] (M = Cr Mo or W) by refluxing [M(CO)6] Chemical Society Reviews 1998 volume 27 Fig 1.Molecular structure of [{W(m-Br)Br(CO)4}2]6 125 in acetonitrile for various periods of time. These zero-valent tris(acetonitrile) complexes have been used as starting materials for the preparation of a wide range of organotransition-metal complexes including p-allyl halocarbonyl complexes.2 In their paper,8 Tate Knipple and Augl showed that the reaction of [W(CO)3(NCMe)3] with iodine in methanol evolved three moles of gas and gave a non-carbonyl containing product. They did not isolate the seven-coordinate bis(acetonitrile) complex [WI2(CO)3(NCMe)2]. In view of these observations we decided to reinvestigate the halogen oxidation of fac- [M(CO)3(NCMe)3] (M = Mo or W) by reacting them in acetonitrile rather than methanol in the hope of obtaining the seven-coordinate complexes [MX2(CO)3(NCMe)2] with two 3(NCMe)2] were prepared.9 More recently,10 we have potentially labile acetonitrile ligands.These reactions (4) first carried out in November 1984 have been extremely successful and quantitative yields of the diiodo-complexes [MI2- (CO) developed a better route to the dibromo-complexes [MBr2(CO)3(NCMe)2] by carrying out the reaction of fac- [M(CO)3(NCMe)3] (prepared in situ) with Br2 at 278 °C. 0 °C fac-[M(CO)3(NCMe)] + X2 —–? [MX2(CO)3(NCMe)2] + NCMe (4) M = Mo or W; X = Br or I We have also studied the oxidation of fac-[Mo- (CO)3(NCMe)3] (prepared in situ8) with a range of other oxidizing agents XY (XY = ICl,11 IBr,12 GeCl4,13 SnCl4 14) to give [MoXY(CO)3(NCMe)2] (X = Cl Y = I; X = Br Y = I; X = Cl Y = GeCl3; X = Cl Y = SnCl3).During the past 12 years we have prepared and characterised over 1900 new organometallic complexes derived from the highly versatile starting materials [MXY(CO)3(NCMe)2].2 This article will describe the versatility of the [MI2- (CO)3(NCMe)2] complexes in Section 2 and discusses how the effective atomic number rule can be used to predict the products of these reactions. Section 3 will give an overview on the structures of seven-coordinate complexes derived from [MI- 2(CO)3(NCMe)2] and Section 4 discusses the applications of the [MXY(CO)3(NCMe)2] system in alkene metathesis and nitrogen fixation research.Section 5 discusses the synthesis of complexes of molybdenum(ii) containing six-different monodentate ligands. The conclusion (Section 6) discusses current and future developments of the [MXY(CO)3(NCMe)2] system. 2 Investigation of the chemistry of the highly versatile complexes [MI2(CO)3(NCMe)2] Although the chemistry of the [MXY(CO)3(NCMe)2] (M = Mo or W; X = Y = Br I; for M = Mo only X = Cl Y = I; X = Br Y = I; X = Cl Y = GeCl3 SnCl3) has been extensively investigated,2 and differences in reactivity are apparent as discussed in Section 5 this section will concentrate on the diiodo-complexes [MI2(CO)3(NCMe)2] which have received the most attention. The complexes [MI2(CO)3(NCMe)2] (M = Mo or W) have proved ideal starting materials for a wide range of reactions.Although they are air-sensitive in both the solid state and solution they can be weighed in air and standard vacuum/Schlenk line techniques are very suitable for carrying out all manipulations with this system. These starting materials can be easily prepared in multigram quantities in a pure state. We have modified Tate Knipple and Augl’s8 early report for the synthesis of fac-[M(CO)3(NCMe)3] complexes by reacting [Mo(CO)6] under reflux in NCMe for 24 h and [W(CO)6] under reflux in NCMe for 72 h. It is essential to have very careful and thorough degassing of the [M(CO)6]–NCMe mixture before refluxing as shown on the BBC Open University video made in Bangor in 1988,15 for the Third Level Inorganic Chemistry Course CHEM 777 S343.This video together with an associated student book15 uses the [MXY(CO)3(NCMe)2] system to illustrate Schlenk line techniques spectroscopic and Chemical Society Reviews 1998 volume 27 126 X-ray techniques and various other techniques used in modern day synthetic inorganic chemistry. Scheme 1 illustrates the diverse chemistry of [MI2(CO)3- (NCMe)2]. Although Scheme 1 is reasonably complete there are many subtle reactions which are not given in the scheme. These reactions depending on the electronic and steric effects of the ligands added can give a diverse range of different products. The top half of the scheme from [MI2(CO)3(NCMe)2] describes the reactions with neutral and anionic ligands whereas the bottom half concentrates on the alkyne derivatives of [MI2(CO)3(NCMe)2].We have found that the complexes [MI2(CO)3(NCMe)2] react with almost any potential ligand in the current chemical catalogues including poor donor ligands for example BiPh3.16 The starting materials [MI2- (CO)3(NCMe)2] and all the reaction products shown in Scheme 1 obey the effective atomic number rule. Although over 1900 new complexes have been made from the [MXY(CO)3(NCMe)2] system which has provided a good training for research students in the synthesis and characterisation of moderately air-sensitive complexes; the main significance is in the catalytic activity of many of these complexes (see Sections 4 and 6). Alkyne ligands can donate two or four electrons to a transition-metal centre,17 and in the early transition-metal complexes alkynes generally utilize both their filled pporbitals and donate four electrons to the metal.It is interesting to note that in the bis(alkyne) complexes [WI2(CO)(NCMe)( h2-RC2RA)2]18 shown in Scheme 1 each alkyne must donate an average of three electrons to the metal for the complexes to obey the effective atomic number rule. The 13C NMR chemical shifts for the alkyne carbon atoms in the complexes [WI2(CO)(NCMe)(h2-RC2RA)2] are generally in the region of 150 ppm which agrees with Templeton and Ward’s17 correlation indicating that the alkynes are donating an average of three electrons to the metal in these systems. The reaction products of [WI2(CO)(NCMe)(h2-RC2RA)2] with both phosphine19 and phosphite20 donor ligands give mono(alkyne) products [WI2(CO)L2(h2-RC2RA)] which have alkyne carbon chemical shifts above 200 ppm which indicates17 they are donating four electrons to the metal thus enabling them to obey the effective atomic number rule.Very recently,21 we have described the reactions of [WI2(CO)3L2] {L = PEt3 PPh3; L2 = Ph2P(CH2)2PPh2} with Na2[S÷S] {S÷S = C3S5 (C3S5 = 4,5-disulfanyl-1,3-dithiole-2-thionate) or maleonitriledithiolate} or H2bdt (bdt = benzene-1,2-dithiol) in acetonitrile and ethanol to give the six-coordinate complexes [W(S÷S)(CO)2L2] which do not obey the effective atomic number rule. Hence although the vast majority of our reactions are predicted by the effective atomic number rule there are several exceptions to the rule.Research workers in other countries have also used the [MI2(CO)3(NCMe)2] complexes as starting materials in their research. For example Krishnamurthy et al.22 have reported the reactions of the diphosphazane ligands RN{P(OPh)2}2 (R = Me Ph) with [MI2(CO)3(NCMe)2] to afford the acetonitrile replaced seven-coordinate complexes [MI2(CO)3- ({P(OPh)2}2NR)] which has been crystallographically characterised for M = W R = Ph. Very recently Cano et al.23 have described the reactions of [MoI2(CO)3(NCMe)2] with the hydrotris(3,5-dimethylpyrazol-1-yl)borate ligand to yield a range of products depending upon the solvent and the initial concentration of the starting materials including [Mo4O4(m3- O)2(m2-O)2(m2-OH)2(HpzMe2)6]I2·4NCMe which has been crystallographically characterised.3 Structures of seven-coordinate complexes derived from [MI2(CO)3(NCMe)2] Seven-coordination is an unusual geometry in transition-metal chemistry and there are four main types of structure exhibited by seven-coordinate complexes; these are (i) capped octahedral, Scheme 1 M = Mo or W; for the reaction of [MI2(CO)3(NCMe)2] with donor ligands described in the top half of the page L LA and LB represent a wide range of neutral carbon nitrogen phosphorus arsenic antimony bismuth oxygen and sulfur donor ligands. L÷L represents a neutral bidentate phosphine donor ligand N÷N represents a neutral bidentate nitrogen donor ligand such as 2,2A-bipy 1,10-phen etc.For the top and bottom half of the scheme S2CX2 (X = NR2 R = Me Et CH2Ph etc.; X = OEt SEt etc.). For the reactions of the bis(alkyne) complexes [WI2(CO)(NCMe)(h2-RC2RA)2] (R RA = Me Ph etc.) L L÷L and L÷L÷L represent mono- bi- and tri-dentate phosphine ligands respectively and X2; X = Cl Br I NO2 NO3 NCS OH SR (R = Et But Ph CH2Ph). 2- (ii) capped trigonal prismatic (iii) the so-called ‘4 3 geometry’ and (iv) pentagonal bipyramidal geometry. Several important reviews concerned with the structures of seven-coordinate complexes have been published,2,24,25 which also discuss the structures of the [MX2(CO)3L2] type complexes. In reality some of the structures we have found derived from the [MI2(CO)3(NCMe)2] system do not exhibit one of the four perfect geometries described above but crystallise in some intermediate structure often in between two structural types.It has been shown that very few coordination polyhedra have regular structures and the majority lie on the reaction coordinates connecting the ideals. The majority of the structures which have been determined of seven-coordinate complexes derived from [MI2(CO)3(NCMe)2] are capped octahedral or distorted capped octahedral although all four principal geometries have been observed. In order to clarify the main geometries of seven-coordinate complexes derived from [MI (CO)3(NCMe)2] accurate coloured diagrams of crystal structures together with simple model pictures of complexes for all four representative structures are shown in Figs.2–5. Fig. 2 (a) Molecular structure of [WI2(CO)3(NCMe)2].26 (b) Simple representation of the capped octahedral structure of [WI2(CO)3(NCMe)2]. The molecular structure of the tungsten starting material [WI2(CO)3(NCMe)2] has been crystallographically determined 127 Chemical Society Reviews 1998 volume 27 Fig. 3 (a) Molecular structure of [WI2(CO)3(NCMe)(SbPh3)].10 (b) Simple representation of the capped trigonal prismatic geometry of [WI2- (CO)3(NCMe)(SbPh3)]. Fig. 4 (a) Molecular structure of the cation [WI(CO)3([9]aneS3)]+.28 (b) Simple representation of the 4 3 geometry or piano-stool geometry of [WI(CO)3([9]aneS3)]+. (Fig. 2),26 and has distorted capped octahedral geometry with a carbonyl ligand in the unique capping position which is capping an octahedral face containing two carbonyl ligands and one of the iodine atoms as shown in Fig.2(b). The large iodide ligands are trans to each other in this complex. Colton and Fig. 5 (a) Molecular structure of [WI(acac)(CO)2(PEt3)2].29 (b) Simple representation of the pentagonal bipyramidal geometry of [WI(acac)(CO)2- (PEt3)2]. Chemical Society Reviews 1998 volume 27 128 Kevekordes27 have shown how the solution state 13C NMR spectra of capped octahedral complexes containing a capping carbonyl ligand can be correlated with the crystallographically determined solid-state structure. For example the low temperature 13C NMR spectrum (270 °C 250 MHz CD2Cl2) of [WI2(CO)3(NCMe)2],26 has two carbonyl resonances at d = 228.48 and 202.36 with an intensity ratio of 1 2.The lower field resonance is due to the capping carbonyl group.27 The resonance at d = 202.36 is in the typical range for octahedral carbonyl ligands. The room temperature 13C NMR spectrum (+25 °C 250 MHz CD2Cl2) of [WI2(CO)3(NCMe)2] has a single carbonyl resonance which as expected suggests the complex is fluxional due to the small amount of energy required for these seven-coordinate complexes to rearrange. Many seven-coordinate halocarbonyl complexes exhibit this fluxional behaviour.2,27 The other three less commonly observed geometries of complexes derived from [MI2(CO)3(NCMe)2] are exemplified by (a) [WI2(CO)3(NCMe)(SbPh3)],10 which has a distorted capped trigonal prismatic geometry with one triangular face having three CO groups and the other having I(2) Sb and N with I(1) capping the rectangular face defined by C(1) I(2) N and C(3) {see Fig.3(a) and (b)}; (b) the S3-bonded macrocyclic complex [WI(CO)3([9]aneS3)][BPh4]28 which can be considered to have the so called ‘4 3 geometry’ with the three sulfur atoms forming the seat and the three carbonyl ligands and an iodide group forming the four legs of this often called ‘piano stool’ geometry {refer to Fig. 4(a) and (b)}; and (c) the least commonly observed structure in this system which is the pentagonal bipyramidal geometry and which has only been found in very recently29 observed structures of [MI(acac)- (CO)2(PEt3)2] (M = Mo W) {for M = W see Fig. 5(a) and (b)}. It should be noted that many seven-coordinate lanthanide and actinide complexes do exhibit the pentagonal bipyramidal geometry.24 There is obviously a complex interplay between steric and electronic effects and multidenticity of the ligands attached to the metal in these systems which governs the geometry of these complexes.It is important to note that the geometry observed in the solid state appears to be preserved in solution as evidenced by 13C NMR spectroscopy.27 4 Applications of the complexes [MXY(CO)3(NCMe)2] and their derivatives Throughout the course of our work with the [MXY- (CO)3(NCMe)2] system we have been conscious of the importance of the potential applications of these complexes in other areas of research. Two of the main potential applications of the complexes [MXY(CO)3(NCMe)2] and their derivatives we have been exploring are in (i) alkene metathesis catalysis and (ii) nitrogen fixation research.Our interest in developing the chemistry of [MXY- (CO)3(NCMe)2] with a very wide range of ligands results partly from a paper published by Bencze and Kraut-Vass30 in 1985 which describes the catalytic activity of the halocarbonyl complexes [MX2(CO)3L2] (M = Mo or W; X = Cl or Br; L = PPh3 or AsPh3). Complexes of this type were found to be single component catalysts for the ring opening polymerisation of norbornene and norbornadiene.30 The best catalyst was found to be [WCl2(CO)3(AsPh3)2] which polymerises norbornadiene to give a vinylcyclopentene polymer with a high cis-content. Bencze and Kraut-Vass also found30 that the mechanism of these reactions (5) involved loss of L in the rate-determining step followed by coordination of the alkene (6).RDS (5) [MX2(CO)3L2] [| [MX2(CO)3L] + L [MX2(CO)3L] + alkene [| [MX2(CO)3L(h2-alkene)] (6) Further studies by Bencze et al.31 showed that the next step in the mechanism involves a 1,2-hydrogen shift on the coordinated alkene to give a carbene intermediate. In view of these observations we have been studying the catalytic activity of [MXY(CO)3(NCMe)2] and many of their derivatives shown in Scheme 1 towards the metathesis of a wide range of acyclic and cyclic alkenes with considerable success. Since Bencze and Kraut-Vass30 showed the rate-determining step in these catalytic reactions involved loss of donor ligand {see reaction (5)} it is significant that in our system the complexes [MXY- (CO)3(NCMe)2] and [MXY(CO)3(NCMe)L] have at least one labile acetonitrile ligand and many of the complexes in Scheme 1 including the alkyne complexes are highly active alkene metathesis catalysts.32 For example the complexes [MBr2(CO)3(NCMe)2] (M = Mo or W) and [WBr2(CO)3(NCMe)( EPh3)] (E = P As Sb) all rapidly initiate the room temperature polymerisation of norbornadiene.32 Since the nature of L can be varied inexhaustibly in this system the stereochemistry of the products can in principle be controlled in these reactions.The chemical industry puts a high emphasis on water soluble homogeneous catalysts and during the course of our work in conjunction with Dr Aidan J.Lavery ZENECA Specialties we have discovered the cheap and readily available water solubilising substituted pyridine ligands such as sodium [pyridine- 4-carboxylate] which forms completely water-soluble complexes such as [WI2(CO)3(4-NaO2CC5H4N)2].33 This and related water-soluble seven-coordinate complexes are being explored as potential biphasic catalysts. Over the years much research has been directed to synthesising simple models that mimic the active site of the nitrogenase enzyme which contains molybdenum iron and sulfur in the active site. Recently the X-ray structure analysis of the three forms of the active site of the FeMo protein which was isolated from Azobacter vinelandii and Clostridium pasteurianum has been described,34 and is shown in Fig.6. From our point of view CH2CO2 – S S S Fe Fe O C CH2CH2CO2 – Mo CysFe O C S S Fe S Fe O His Fe Fe S 2- S ABCDEF] complexes [MoCl(GeCl3)- S Fig. 6 Structure of the active site of FeMoco from an X-ray structural analysis34 it has a molybdenum centre with three attached sulfur atoms together with a bidentate anionic oxygen donor ligand and a histidine group. In conjunction with Dr Raymond L. Richards et al.28,29 at the Nitrogen Fixation Laboratory we have been investigating the [MXY(CO)3(NCMe)2] system in nitrogen fixation research by firstly investigating the chemistry of [MXY(CO)3(NCMe)2] and derivatives with a wide variety of neutral and anionic oxygen and sulfur donor ligands and with iron–sulfur clusters such as [Fe4Cp4S6] in order to attempt to mimic the active site of nitrogenase shown in Fig.6. In order to get three sulfur atoms attached to molybdenum or tungsten we have used28 the reactions of equimolar amounts of [MI (CO)3(NCMe)2] with for example the macrocycle [9]aneS3 in the presence of Na[BPh4] to give the cationic complexes [MI(CO)3([9]aneS3)][BPh4]. This has been crystallographically characterised for M = W shown in Fig. 4(a). We are also investigating the reduction of the sulfur donor ligand complexes in the presence of dinitrogen in order to form dinitrogen complexes containing sulfur donor ligands. 5 Complexes containing six different monodentate ligands derived from [MXY(CO)3(NCMe)2] Very few octahedral transition-metal complexes containing six different monodentate ligands have been described.An early example of a [MABCDEF] complex is [PtCl(Br)I- (NO2)(NH3)(py)] reported in 1958.35 In order to illustrate the subtle differences in the chemistry of [MI2(CO)3(NCMe)2] and other [MXY(CO)3(NCMe)2] type complexes this section describes the strategic synthesis of [MABCDEF] molybdenum complexes and the X-ray crystal structure of a dinuclear complex which has two identical MABCDEF centres. The complete synthetic scheme for the synthesis of the [MABCDEF] complexes [MoCl(GeCl3)(CO)(NCMe)(PPh3)(h2- RC2R)] (R = Me or Ph) is shown in Scheme 2.36 It is interesting to note that reaction of [MoXY(CO)3(NCMe)2] with an equimolar amount of PPh3 in NCMe for X = Y = I gives the acetonitrile displaced product [MoI2(CO)3(NCMe)(PPh3)] whereas for X = Cl Y = GeCl3 the carbon monoxide displaced product [MoCl(GeCl3)(CO)2(NCMe)2(PPh3)] is obtained.This is likely to be due to the combined higher electronegativity of the Cl and GeCl3 groups which weakens the M–C bond due to decreased back-bonding and strengthens the Mo–N bond due to increased s-bonding compared with the two iodo ligands. As previously discussed,17 alkynes can act as ‘four-electron donors’ and hence the reaction of [MoCl- (GeCl3)(CO)2(NCMe)2(PPh3)] with one equivalent of RC2R yielded the chiral [M Chemical Society Reviews 1998 volume 27 129 Scheme 2 (CO)(NCMe)(PPh3)(h2-RC2R)] via displacement of a CO and an NCMe ligand. There are 15 possible simple geometrical isomers of an [MABCDEF] octahedral complex and each one will have its own diastereoisomer.Over the years many hitherto unsuccessful attempts have been made to grow suitable single crystals for X-ray analysis of [M 2Me)] with an equimolar amount of [NBun 4][NCS] ABCDEF] complexes derived from [MoCl(GeCl3)(CO)(NCMe)(PPh3)(h2-RC2R)]. However treatment of [MoCl(GeCl3)(CO)(NCMe)(PPh3)(h2- MeC gave the anionic complex [NBun (CO)(PPh 4][MoCl(GeCl3)(NCS)- 3)(h2-MeC2Me)] which reacted in acetone with a slight trace of water to afford the dimeric {MABCDEF}2 complex 3)(h2-MeC2Me)Mo(m-NCS)(m-SCN)MoCl(CO)- 3)(h2-MeC2Me)]. This is the first {MABCDEF}2 complex to [Cl(CO)(PPh (PPh be structurally characterised (Fig. 7).37 Each molybdenum atom has terminal chloride carbon monoxide triphenylphosphine and but-2-yne ligands and a nitrogen and sulfur of two bridged thiocyanate groups.One of the reasons for preparing mononuclear MABCDEF complexes was to synthesise and characterise new chiral catalysts. However the dinuclear complex above is very interesting from a stereochemical point of view which was not discussed in the original communication.37 The dimer (Fig. 7) is not a chiral molecule as there is a centre of symmetry located at the centre of the eight-membered ring and hence the dimer is the achiral meso form of the molecule (point group Ci). The two Mo atoms are both stereocentres but have opposite absolute configurations which can be classified as C and A using the Cahn–Ingold–Prelog priority rules by analogy with the well known R/S nomenclature system for tetrahedral stereocentres.6 Conclusions and future developments of the [MXY(CO)3(NCMe)2] system The main purpose of this article has been to illustrate how the easily prepared seven-coordinate complexes [MI2- (CO)3(NCMe)2] have proved to be highly versatile starting materials for a wide range of reaction types as shown in Scheme 1. The diversity of this research is also highlighted by the use of a number of other oxidizing agents in place of I2 such as XY (XY = ICl,11 IBr,12 GeCl4 13 or SnCl4 14) which when reacted with fac-[M(CO)3(NCMe)3] (prepared in situ8) give the products [MXY(CO)3(NCMe)2]. The [MXY(CO)3(NCMe)2] Chemical Society Reviews 1998 volume 27 130 Fig.7 Molecular structure of [Cl(CO)(PPh3)(h2-MeC2Me)Mo(m-NCS)(m- SCN)MoCl(CO)(PPh3)(h2-MeC2Me)]37 series of complexes have proved to be extremely useful as starting materials for a wide range of chemistry. I believe the most significant current and future potential uses 3(NCMe)2] system are in homogeneous 3(NCMe)2] and derivatives in alkene metathesis 2(CO)- 2- of the [MXY(CO) catalysis. Apart from the previously described,32 uses of [MXY(CO) (Section 4) which we are continuing to develop; a very recent result38 has shown that the bis(phosphite) complex [MoI {P(OPri)3}2(h2-MeC2Ph)] catalyses the trimerisation of 1-phenylpropyne to give 1,2,4-trimethyl-3,5,6-triphenylbenzene. We are currently exploring the applications of [MoI (CO){P(OPri)3}2(h2-MeC2Ph)] and related complexes for the trimerisation of other alkynes.The uses of the organometallic phosphines such as [WI2(CO){PhP(CH2CH2PPh2)2-P,PA}(h2- MeC2Me)]39 which exists as two diastereoisomers in a single crystal as shown in {Fig. 8(a) and (b)} for the synthesis and applications of mixed-metal catalysts are also being investigated. 7 Acknowledgements I would like to thank all my co-workers who have been involved with the [MXY(CO)3(NCMe)2] system for their skilful experimental work and intellectual contributions. I would also like to thank in particular Dr Michael G. B. Drew University of Reading for his outstanding collaboration over many years and for producing the colour diagrams used in this review. I also thank Professor Michael B.Hursthouse and his collaborators on the EPSRC Crystallographic Service (University of Wales Cardiff) and Dr David L. Hughes (Nitrogen Fixation Laboratory Norwich) for their crystallographic work; Dr Raymond L. Richards (Nitrogen Fixation Laboratory Norwich) and Dr Aidan J. Lavery (ZENECA Specialties) for their valuable collaboration and support of our work over the years. I would like to thank the EPSRC AFRC Royal Society British Council Institute of Terrestrial Ecology and ZENECA Specialties for supporting our research work in this field over the past twelve years. Finally I would like to thank Barbara Kinsella and Fig. 8 Molecular structure of the two diastereoisomers of [WI (CO){PhP(CH2CH2PPh2)2-P,PA}(h2-MeC2Me)]39 2- Caroline Naylor for their cheerful and efficient preparation of this article.8 References 1 H. L. Nigam and R. S. Nyholm Proc. Chem. Soc. 1957 321. 2 P. K. Baker Adv. Organomet. Chem. 1996 40 45 and references therein. 3 H. C. E. Mannerskantz and G. Wilkinson J. Chem. Soc. 1962 4454. 4 R. Colton and I. B. Tomkins Aust. J. Chem. 1966 19 1143. 5 R. Colton Coord. Chem. Rev. 1971 6 269 and references therein. 6 F. A. Cotton L. R. Falvello and J. H. Meadows Inorg. Chem. 1985 24 514. 7 A. D. Westland and N. Muriithi Inorg. Chem. 1972 11 2971. 8 D. P. Tate W. R. Knipple and J. M. Augl Inorg. Chem. 1962 1 433. 9 P. K. Baker S. G. Fraser and E. M. Keys J. Organomet. Chem. 1986 309 319. 10 P. K. Baker M. B. Hursthouse A. I. Karaulov A.J. Lavery K. M. A. Malik D. J. Muldoon and A. Shawcross J. Chem. Soc. Dalton Trans. 1994 3493. 11 P. K. Baker T. Birkbeck S. Br�ase A. Bury and H. M. Naylor Transition-Met. Chem. 1992 17 401. 12 P. K. Baker K. R. Flower H. M. Naylor and K. Voigt Polyhedron 1993 12 357. 13 P. K. Baker and D. ap Kendrick J. Organomet. Chem. 1994 466 139. 14 P. K. Baker and A. Bury J. Organomet. Chem. 1989 359 189. 15 Open University S343 Laboratory Techniques of Inorganic Chemistry 1990 4. 16 P. K. Baker S. G. Fraser and T. M. Matthews Inorg. Chim. Acta 1988 150 217. 17 J. L. Templeton and B. C. Ward J. Am. Chem. Soc. 1980 102 3288. 18 E. M. Armstrong P. K. Baker and M. G. B. Drew Organometallics 1988 7 319. 19 E. M. Armstrong P. K. Baker M. E.Harman and M. B. Hursthouse J. Chem. Soc. Dalton Trans. 1989 295. 20 P. K. Baker E. M. Armstrong and M. G. B. Drew Inorg. Chem. 1989 28 2406. 21 P. K. Baker M. G. B. Drew E. E. Parker N. Robertson and A. E. Underhill J. Chem. Soc. Dalton Trans. 1997 1429. 22 M. S. Balakrishna S. S. Krishnamurthy and H. Manohar Organometallics 1991 10 2522. 23 M. Cano J. A. Campo J. V. Heras E. Pinilla and A. Monge Polyhedron 1996 15 1705. 24 M. G. B. Drew Prog. Inorg. Chem. 1977 23 67 and references therein. 25 M. Meln�ýk and P. Sharrock Coord. Chem. Rev. 1985 65 49 and references therein. 26 M. G. B. Drew P. K. Baker E. M. Armstrong S. G. Fraser D. J. Muldoon A. J. Lavery and A. Shawcross Polyhedron 1995 14 617. 27 R. Colton and J. Kevekordes Aust. J. Chem. 1982 35 895 and references therein. 28 P. K. Baker S. J. Coles M. C. Durrant S. D. Harris D. L. Hughes M. B. Hursthouse and R. L. Richards J. Chem. Soc. Dalton Trans. 1996 4003. 29 P. K. Baker A. I. Clark M. G. B. Drew M. C. Durrant and R. L. Richards J. Organomet. Chem. 1997 549 193. 30 L. Bencze and A. Kraut-Vass J. Mol. Catal. 1985 28 369. 31 L. Bencze A. Kraut-Vass and L. Pr�okai J. Chem. Soc. Chem. Commun. 1985 911. 32 D. J. Muldoon PhD Thesis University of Wales Bangor 1994. 33 P. K. Baker A. E. Jenkins A. J. Lavery D. J. Muldoon and A. Shawcross J. Chem. Soc. Dalton Trans. 1995 1525. 34 D. Sellmann Angew. Chem. Int. Ed. Engl. 1993 32 64 and references therein. 35 L. N. Essen F. A. Zakharova and A. D. Gel’man Zh. Neorg. Khim. 1958 3 2654. 36 P. K. Baker and D. ap Kendrick J. Chem. Soc. Dalton Trans. 1993 1039. 37 P. K. Baker M. E. Harman D. ap Kendrick and M. B. Hursthouse Inorg. Chem. 1993 32 3395. 38 P. K. Baker M. G. B. Drew and D. S. Evans unpublished results. 39 P. K. Baker S. J. Coles D. E. Hibbs M. M. Meehan and M. B. Hursthouse J. Chem. Soc. Dalton Trans. 1996 3995. Received 28th July 1997 Accepted 15th October 1997 131 Chemical Society Reviews 1lume 27