10 Organometallic Chemistry Part (ii) The Main-Group Elements By John D. KENNEDY Department of Inorganic and Structural Chemistry University of Leeds Leeds LS2 9JT 1 Introduction and General Considerations A very useful three-volume 3232-page reference work entitled Dictionary of Organornetallic Compounds has been published this year.* This is an extensive compilation of those organometallic compounds deemed by the Editors and their advisors to be important and for each compound details of structure physical and chemical properties reactions and selected references are given. There are some 7650 entries for main-group compounds (Li Na K Rb Cs; Be Mg Ca Sr Ba; B Al Ga In,TI; Si Ge Sn Pb; As Sb Bi; Zn Cd Hg) and the book (and its readers) are well served by four different complementary indexes.Some readers will find that a quick flick through reveals a number of gaps where they would expect to find compounds which they would consider to be quite important. However selection is always difficult and it is planned to have annual supplements each containing some 2000 entries to abstract the recent literature and to extend the field of coverage the Editors state that they are always pleased to receive suggestions for compounds or groups of compounds to be included and so the project will develop into an even more valuable ‘information resource’ as they say nowadays. The publication of these volumes together with the major work Comprehensive Organometallic Chemistry mentioned in the 1982 report,2 does however introduce a disturbing consideration that this reporter has now heard aired increasingly over the past year or so.This is that for three or four years now the mean annual amount of compilatory review literature in organometallic chemistry has approached and probably exceeded the amount of published original research in the area! With other encyclopaedic compilations also currently in preparation and with the usual run of annual surveys reviews periodical reports etc. etc. this peculiar situation seems likely to continue for some time yet. As with the choice of direction of experimental work mentioned previously,2 it is again up to the organometallic chemists themselves to consider whether the advancement of their sub-discipline would not perhaps be better served by adopting other approaches.Perhaps because of this diversion of a substantial proportion of available organometallic talent from creative experimentation to compilatory literature abstraction the character of much of the new main-group organometallic chemistry ‘ Dictionary of Organometallic Compounds’. ed. J. Buckingham Chapman and Hall 1984 (3 vols.) J. D. Kennedy Annu. Rep. hog. Chem. Sect. B 1982 79 257 and 1983 80 293. 227 J. D. Kennedy reported in 1984 is again one of shall we say steady progress and consolidation rather than one of a general burgeoning of innovative chemistry. In this regard it may be relevant to note that feelings of dijh vu professed last year by this reporter were also experienced by the corresponding reporter of a decade ago.3 Fortunately many sparks of creativity still glow however and encouraging flames are still often seen.Organosilicon chemistry in particular remains a lively area. Of the two complementary structural tools that have become much more generally available in recent years viz. single-crystal X-ray diffraction analysis and versatile multielement n.m.r. spectroscopy the latter does not yet seem to be exploited to its full potential in more general aspects of main-group organometallic chemistry and so there are probably many interesting applications still to come in the near future. There is however once more a high incidence of single-crystal X-ray work reported in both older and newer areas of the sub-discipline and cyclopentadienyl and other arene complexes of the main-group metals seem to be well represented this year.In some areas the use of X-ray diffractometry is becoming quite routine and it is being applied increasingly to examine the more subtle variations in molecular dimensions rather than to elucidate new structural types and behaviour. In this context it must however be evident that (to take but one example) the collection of sorhe 45 000 X-ray data to determine the molecular structures of Ph,MSMe (where M = C Si Ge Sn or Pb)? although excellent systematic work will bring tears to the eyes of those organometallic chemists with real structural mysteries to solve but who have limited or zero access to diffractometer facilities. 2 Group I Again the bulk of the reported work for this Group is in organolithium chemistry.There is however an interesting report that unsolvated aryl alkali-metal reagents ArM where M = Na or K as well as Li can be made to be soluble in aromatic solvents by the use of the magnesium P-alkoxyoxide [Mg(OCH2CH20Et),] so that sodiation and kaliation of organic compounds under homogeneous conditions is now possible.' This has obvisus synthetic potential. The solution species are believed to have formulations such as [Na2MgPh2(OCH2CH20Et)2]n. There is continuing interest in the solution equilibria of more straightforward organolithium reagents and in the relationship of these to reactivity and product distribution. PhLi has been examined in solution by n.m.r. chemical shift and relaxation studies and has been found to be tetrameric in cyclohexane/diethyl ether solution with the same geometry as in the solid state.6 In dilute diethyl ether however dimer and tetramer coexist and in THF the species is dimeric even at -120 "C.The isomerization of vinyllithium species continues to 'attract attention from the schools involved (ref. 7 and other refs. cited therein) the equilibrium of equation (1) and its variation with solvents substituents etc. has been examined and is of interest in connection with the use of these species for stereospecific syntheses. Various isomerization mechanisms are considered. K. Smith Annu. Rep. hog. Chem. Sect. B 1975 72 136. G. D. Andreeti G. Bocelli G. Calestani and P. Sgarabotto,J. Organomet. Chem. 1984 273 31. C. G. Screttas and M.Micha-Screttas Organometallics 1984 3 904. L. M. Jackman and L. M. Scarmoutzos J. Am. Chem. SOC.,1984 106 4627. Organometallic Chemistry -Part (ii) Main-Group Elements R' Li H\ H\ / L ,c=c\ /-c=c /\ R2 Li R2 R' N.m.r. studies now including 6Li as well as 13C data have revealed two previously undefined 'dilithium semibullvalenides' prepared in dimethyl ether or THF solution from the reduction of sernibullvalene by metallic lithium.8 These two species are identified as the achiral meso-C, and chiral (*)-D2 diastereoisomers of bis(bicyclo[3.3.0]octa-3,7-diene-2,6-diyl)tetralithium [Li4(C8H&] the first charac- terized pair of diastereoisomeric organolithium compounds [structures (1) and (2) respectively] the system also displays interesting inter- and intramolecular intercon- version processes both within and between the two isomeric forms.The use of co-ordinating ligands to increase the effective bulk about the Li atom (see later) thereby inducing lower states of aggregation in organolithium compounds has its complement in the use of bulky substituents on the organic residue. This aspect of organolithiurn chemistry is still receiving some attention. For example it has been found that [LiCH(SiMe,),] is a monomer in the gas phase at 413 K with the distance Li-C about 203pm whereas in the solid it is a polymer with a -Li-C-Li-C-Li-C-backbone with angles at both Li and C each averaging about 150" and with the distances Li-C in the range 213(9)-227(2) ~m.~ This behaviour contrasts with the isoelectronic but ostensibly less bulky [LiN(SiMe,),] which is a cyclic dimer in the gas phase and a cyclic trimer in the solid.However aggregation in the amide does not require electron-deficient bonding and the 2 x (2e 2c) Li-N-Li linkages are therefore significantly shorter than the 2e 3c Li-C-Li units. The 'simplest metallocene' [Li(C5H5)] [structure (3)] has been approached yet more closely the latest report in this area containing X-ray structural work on [(tmeda)LiC5H4(SiMe3)] [structure (4)I.l' This is made in a straightforward manner by the treatment of [Me3SiC5H5] with Bu"Li in tmeda solution. The Li-C distances are not surprisingly somewhat longer than those previously calculated for [Li(C,H,)] by those chemists (commonly known as theoretical chemists) who do quantum mechanical calculations and who thereby tend not to be constrained by what may be experimentally achievable (e.g.,ref.11);the reasons for the discrepancy R. Knorr and T. von Roman Angew. Chem. Int. Ed. Engl. 1984 23 366 and other refs. cited therein. M. J. Goldstein and T. T. Wenzel J. Chem. Soc. Chem. Commun. 1984 1655. J. L. Atwood T. Fjeldberg M. F. Lappert N. T. Luong-Thi R. Shakir and A. J. "'home J. Chem SOC. Chem. Commun. 1984 1163. M. F. Lappert A. Singh. L. M. Engelhardt and A. H. White J. Orgonomet. Chem. 1984 262. 271. 230 J. D. Kennedy n are discussed.” A comparative perusal of references such as 10 and 11 in fact indicates that this is one of several areas in which there are compatibility problems between useful working approaches adopted by experimental chemists and useful working treatments adopted by the theoreticians particularly with regard to assump- tions about the orbital availability of the metal atom and the nature of the organometallic bond.Similar considerations apply for example to some cyclo- pentadienylberyllium species and to various related six-vertex beryllaboranes.’* Another area of apparent dichotomy between theory and fact arises from the experimentally determined structure of dilithium tribenzylidenemethane-2tmeda which has a gross geometry as represented in (5).13 Although it is pointed out that the geometry differs from the threefold anticipated from calculations for the free 2-anion (6)’ it should be noted that in practice the lithium counter-ions are of course required for stability and their presence contributes to the geometry.Whether this demonstrates an essential futility of performing calculations on species that are not experimentally observable or whether it should be regarded as a valid way of examining their ‘chemistry’ precisely because they are not experimentally observable is of course a moot point. The (polymeric) ‘polylithiated methanes’ {CLi H4-,}, continue to be examined from various points of view,”’-’’ and an overview of some of the philosophy behind E. D. Jemmis and P. von R. Schleyer J. Am. Chem. SOC. 1982 104 4781. 12 J. D. Kennedy hog. Inorg. Chem. 1984 32,pp. 625-627 and refs. cited therein. l3 D. Wilhelm H. Dietrich T.Clark W. Mahdi A. J. Kos and P. von R. Schleyer J. Am. Chem. Soc. 1984 106 7279. 14 A. Maercker and M. Theis Angew. Chem. Int. Ed. Engl. 1984 23,995. j5 J. A. Gurak J. W. Chinn R. J. Lagow H. Steinfink and C. S. Yannoni Inorg. Chem. 1984 23,3717. 16 J. A. Gurak J. W. Chinn R. J. Lagow R. D. Kendrick and C. S. Yannoni Inorg. Chim. Acta 1985 % L75. ” H. Kawa J. W. Chinn and R. J. Lagow 1. Chem. Soc. Chem. Commun. 1984 1665. Organometallic Chemistry -Part (ii) Main-Group Elements 231 the work has appeared.18 Two new approaches have been established for the synthesis of these species.’ These involve an initial treatment of [C(B(OMe),),] with [Hg(OAc),] followed by NaCl to give [C(HgCl),] [equation (2)] or treatment with [HgEt(OAc)] to give [C(HgEt),] [equation (3)].Treatment of either [C(HgCl),] or [C(HgEt),] with an excess of Bu‘Li or with Li dust in diethyl ether then results in the formation of CLi,. The halide reaction gives a solid that contains LiCl whereas [C(HgEt),] in cyclohexane yields a red-brown solution from which a deep red- brown pyrophoric solid may be obtained.’ The precise nature of the polylithiated methanes in the solid state remains an intriguing mystery however and 1984 has seen the application of solid-state CP- MAS 13C n.m.r. spectroscopy to the problem for which interestingly the use of 6Li isotopomers was found to be This is in principle a very suitable probe but unfortunately in this case the results were not too informative. This did not dampen the enthusiasm of the research team involved however who are obviously so proud of the work that they tell us about it twice.The interested punter therefore has the unaccustomed luxury of being able to choose to read an account of the n.m.r. work printed either on thinner shiny paper (ref 15) or thicker matt paper (ref 16). The synthetically useful lithium organocuprates or ‘Gilman Reagents’ which are usually held to be dimeric in solution have received increased attention in 1984. Structural work has shown that the yellow crystalline species [Li2CU3Ph&[ Li,C1,(OEt2),,] has a {Li2Cu3} core structure based on a trigonal bipyramid (7),19 and that a gold species [Li2A~2(2-C6H4CH2NMe2)4], used as a ‘model for the organocuprates’ has a virtually planar {Li2Au2[ C( ipso)],} entity with C(ipso) showing interaction with both Li and Au [schematic structure (8)].,’ As with straightforward alkyllithiums the use of bulky groups on C or Li in these species tends to reduce the state of aggregation which will obviously influence the nature of their reactivity.The species [LiCu{C(SiMe,),},(thf),] has been shown by (7) l8 R. J. Lagow and J. A. Gurak in ‘Chemistry for the Future’ ed. H. Griinewald Pergamon Oxford 1984 pp. 107-1 13. 19 H. Hope D. Oram and P. P. Power J. Am. Chem. Soc. 1984 106 1149. 20 G. van Koten and J. Jastnebski J. Am. Chem Soc. 1984 106 1880. 232 J. D. Kennedy crystallography to have cationic [Li(thf),]+ and anionic [CU{C(S~M~~)~}~]-moieties,2' the compound being isomorphous with the corresponding lithium 'ate' complex [Li(thf),][ Li{C(SiMe,),},] mentioned in last year's report.Another inter- esting incidence of steric bulk is in the lithiated carbaborane [(pmdeta)LiC2B,oHloMe] obtained from Bu"Li and [C2B,oH,lMe] in hexane fol- lowed by the addition of pmdeta (N,N,N',N",N"-pentamethyldiethyl-enetriamine).22 In this compound the Li atom is bound to six-co-ordinate carbon [schematic structure (9)]. The Li-C distance is short at 218(1) pm possibly due to the high carbon s-character in the bond although the correlation of bond-order with bond-length is far from clear-cut in this area.22 3 Group I1 In Group I1 nothing in organoberyllium or organocadmium chemistry has attracted this reporter's attention in 1984 although some interesting chemistry for the other elements has been published.In organomagnesium chemistry it has been found that Mg n.m.r. spectroscopy (25 MHz at 9.4 Tesla) is a surprisingly good method of investigating Schlenk equilibria et~.,~ There is a large chemical shift range of some 200 p.p.m. which is held to facilitate clarification of controversial bonding behaviour and in the investigation of the Schlenk equilibrium the method has the advantage that all three species can be identified directly. Although large linewidths are associated with bis( a-bonded-alkyl)magnesiums,it is nevertheless found to be much more useful than 13C or 'H n.m.r. spectroscopy. In the example given [Mg(C,H,),] it is concluded that predominantly covalent bonds are present.23 Magnesium compounds involving the anthracene skeleton mentioned briefly in last year's report,2 have been the subject of further interesting work.The orange magnesium-anthracene 1:l adduct in THF has been found to be an effective electron-transfer reagent in the generation of Grignard reagents such as (10) and (thf)CIMg M gCI (t hf) MgCl(thf) 21 C. Eaborn P. B. Hitchcock J. D. Smith and A. C. Sullivan J. Organornet. Chem. 1984 263 C23. 22 W. Clegg D. A. Brown S. J. Bryan and K. Wade Polyhedron 1984 3 307. 23 R. Benn H. Lehmkuhl K. Mehler and A. Rufinska. Angew. Chem. Int. Ed. Engl. 1984 23 534. Organometallic Chemistry -Part ( ii) Main-Group Elements 233 (11) from the corresponding halides.24 These two species are formed in >90% yield when stoicheiometric quantities of [Mg(C14Hlo)] are used and are not easily accessible if at all using other forms of magne~ium.,~ Further work on the characterization of the magnesium-anthracene complex itself has been reported and the work includes a study of the kinetics of f~rmation.~’ The species [ Mg(C,,Hlo)(thf),] can be recrystallized from THF solution as orange needles which are sparingly soluble.N.m.r. spectroscopy indicates that the strongest metal interaction occurs with the 9,10-positions and it is concluded that the com- pound can be best regarded either as an ion pair with strong interaction at C-9,lO or as a covalent compound with a large polar contribution [schematic structure (l2)].,’ This structure is to be compared with that of the magnesium-aluminium compounds [ (thf),MgHAlR,( C14H,0)] the magnesium dihydroanthrylene dialkyl- hydroaluminates [schematic structure (13)] which are formed by the reaction of the relatively insoluble magnesium anthracene ( 12) with R2AlH in THF solution.26 H :‘ \ (thf)3Mg AIR2 Y \I Other work on mixed-metal species containing magnesium includes the solid-state structure of lithium magnesates such as [Li,(PhC~C)~Mg(trneda),]and [ Li(tmeda) ][ (tmeda)LiBz,MgBz,] ;27 the configuration of the latter is represented in structure (14).The solid-state structure of the first organogermylmagnesium compound [ (dme)2Mg(GeMe3)2] (1 5) has been established.28 Its preparation from [Hg(GeMe,),] with Mg in 1,2-dimethoxyethane (dme) was reported initially in 1981 previous to which organogermylmagnesium compounds had been postulated A GeMe 24 C.L Raston and G. Salem J. Chem. SOC.,Chem Commun. 1984 1702. 25 B. Bogdanovik S.-T.Liao R. Mynott K. Schlichte and U. Westeppe Chem. Ber. 1984 117 1378. 26 H. Lehmkuhl K. Mehler R. Benn A. Rufinska G. Schroth and C. Kriiger Chem. Ber. 1984,117,389. 27 B. Schubert and E. Weiss Chem Ber. 1984 117 366. 28 L. Rosch C. Kruger and A.-P. Chiang 2. Naturforwh.. Ted B. 1984. 39. 855. 234 J. D. Kennedy to exist only as unstable short-lived intermediates. The compound has an approxi- mately octahedral cis-{Ge20,} configuration about the Mg centre with the mean Ge-Mg distance 271.9(6) pm.28 The renaissance of organozinc chemistry continues and developments include a study of fluxional 2-alkenylzinc compounds by i.r.Raman 'H n.m.r. and 13C n.m.r. ~pectroscopies,~~ and the generation of [ZnBr(CH,Br)] from Zn and CH2Br2 in THF,30 which is claimed to be a useful inexpensive alternative to the more usual use of CH212in Simmons-Smith-type processes for methylene generation. The Reformatsky reaction has been the subject of some attenti~n,~'-~~ and ultrasound has been found to be useful for the generation of the Reformatsky reagents from metallic Thus for example the ultrasound-promoted reaction between ethyl bromoacetate zinc and Schiff's bases gives excellent yields of p-lactams -90% after a few hours with activated zinc granules and SO-70% with presumbably pure zinc The Reformatsky reaction first described nearly a century ago is still one of the best methods for preparing P-hydroxy acids via their esters (Scheme 1),33 but as with many very well-studied organometallic reactions (I):c=o BrCH,COOR + Zn -+ BrZnCH2COOR -:C(OH)CH,COOR -:C(OH)CH,COOH (ii) HzO Scheme 1 the lack of knowledge about the actual reagent in solution inhibits the emergence of a clear picture of the mechanism.In this regard some work on the nature of the Reformatsky reagents that are derived from the ethyl and t-butyl esters of bromoacetic acid is of interest.33 The dimeric structure (16) found in the crystal persists in solution but the solution behaviour is complicated by Schlenk-type equilibria [equations (4) and (S)] and in very polar solvents such as Me,SO the reagents are monomeric C-metallated species; the consequences of these findings for the mechan- ism of the Reformatsky reaction in commonly used solvents are also discussed in the which also serves to re-emphasize the important general point that the 2[BrZnCH2COOR] ZnBr + [Zn(CH,COOR),] (4) [BrZnCH,COOR] ZnBr + l/n[Zn(CH,COOR),] (5) (16) 29 E.G. Hoffman H. Nehl H. Lehmkuhl K. Seevogel and W. Stempfle Chem. Ber. 1984 117 1364. 30 B. Fabisch and T. N. Mitchell 1. Organornet. Chem. 1984 269 219. 3' B. H. Han and P. Boudjouk J. Org. Chem. 1982 47 5030. 32 A. K. Bose K. Gupta and M. S. Manhas J. Chem SOC.,Chem Cornmun. 1984 86. 33 J. Dekker P. H. M. Budzelaar J. Boersma G. J. M. van der Kerk and A. L. Spek Organomefalfics 1984 3 1403. Organometallic Chemistry -Part (ii) Main-Group Elements 235 aggregation state of a typical early main-group organometallic reagent has critical mechanistic implications for its reactions with organic substrates.In view of the structural diversity of Group I1 organometallic compounds the amount of crystallographic work is still rather scant and it is always of interest to see new work. Structural reports this year include an acetoximate complex [Zn,Me,( Me2CN0)4] based on a Zn tetrahedron with an oxime group bound above each Zn3 face in such a way that all the Zn atoms are different.34 Another interesting compound is the colourless species [(p-CSH5)(p-N{SiMe,),)-(Zn-o-C,H,),] [schematic structure (17)] made by the reaction between [Zn(C,H,),] and [Zn{N(SiMe,),},].The bridging C5Hs is asymmetrically bound (Zn-C ca. 210 and 250 pm) and the compound is fluxional (AG ca. 70 kJ mol-') with respect to exchange of the organic groups via scission of a Zn-p-(C5H5) link and rotation about Zn-N.35 Me3Si \\\ Pi"" Alongside interesting new chemistry such as this it is always nice to see the elucidation of old problems. One of these concerns the constitution of 'mercuretin' a compound with a long history which was first obtained in 1809 by melting mercuric acetate and which may well have been the first organometallic compound to be discovered. It has now been identified as the condensation polymer of tris(acetoxy-mercuri)acetic acid (AcOHg),CCOOH with the formulation [AcO{HgC(HgOAc)2C0.0}nH] where n -The compound has a ...OHgC.CO.OHgC.CO.OHgC.CO...backbone and hydrolysis in dilute HC1 gives (ClHg),CCOOH as the only Hg-containing product (identified by X-ray diffraction as its DMSO solvate). Treatment with AcOH yields (AcOHg),CCOOH which reverts to mercuretin by AcOH Other C-metallations involving mercury carboxylates include the quantitative mercurations of bis(dipheny1phosphino)methane with [Hg(OAc),] [equation (6) where n = 2 or 3],,' and the mercuration of 1-OMe-2-NO2-C6H4 with [Hg(OCOF3),] in CF,COOH solution.38 The former reaction is thought to be of interest in that previous examples of C-mercuration in aliphatic compounds were limited to sites with more classically acidic H atoms and the second because the product [structure 34 N.A. Bell H. M. M. Shearer and C. B. Spencer Acta Crystallogr. Sect. C 1984 40,613. 35 P. H. M. Budzelaar J. Boersma G. J. M. van der Kerk and A. L. Spek Organornerallics 1984 3 1187. 36 D. GrdeniC B. Korpar-colig and M. Sikirica J. Organornet. Chern. 1984 276 1. 37 M. Lusser and P. Peringer Organornetallics 1984 3 1916. 38 G. B. Deacon G. N. Stretton and M. J. O'Connor J. Organornet. Chern. 1984. 277 C1. J. D. Kennedy (lS)] exhibits the rather large nuclear spin-state energy difference ,J( 199Hg-’99Hg) of 2163 Hz. Ph2PCH2PPh2+ n[Hg(OAc),] + [Hg(drn~o),][CF~S0,]~ __* [H,-,(AcOH~),-,C(PP~~H~OAC)~][CF,S~~~~ + (n-1)AcOH (6) Other mercury work has been mentioned above [equations (2) (3) and structure (1511. 4 Group I11 It has long been known that gallium halides of empirical formula GaX are freely soluble in benzene and can be precipitated from solution as solids containing ‘crystal benzene’ (see also 1983’s report).2 In this area details of the bis(mesity1ene)gal- lium( I)-gallium( 111) halide complex [Ga(C,H3Me3)2][GaC1,] have been reported.39 The compound is prepared by the dissolution of GaC12 (i.e.Ga[GaCl,]) in hot mesitylene followed by cooling to precipitate the product and the synthesis has also been extended to the indium analogue [In(C,H3Me3)2][InBr4].40 The latter compound is sensitive to heat and light and readily loses mesitylene. The cations of both species have the ‘bent sandwich’ structure (19) with the distances from the metal to the centre of the {c,} plane being CQ.267 and 289 pm respectively. In the indium compound the metal is also bound to three co-planar bromine atoms at 248-250 pm forming a co-ordinate polymer whereas the [Ga(C6H3Me3)2]+ cation is more chemically discrete the interionic Ga ...Cl distances being >325 pm.39*40 More novel is the ‘naked’ mono-organogallium cation [Ga(C,Me,)]+ in [Ga(C,Me,)][GaBr,] [schematic structure (20)],4’ in which although there are again longer-range interactions with nearest-neighbour bromine atoms these are weak at >320 pm and the gallium centre is bonded principally to the hydrocarbon at 252 pm 39 H. Schmidbauer V. Thewalt and T. Zafiropoulos Chern. Ber. 1984 117 3381. 40 J. Ebenhoch G. Miiller J. Riede and H. Schmidbauer Angew. Chern. Inr. Ed.Engl. 1984 23 386. 41 H. Schmidbauer U. Thewalt. and T. Zafiropoulos Angew. Chem. Int. Ed. EngL 1984 23. 76. Organometallic Chemistry -Part (ii) Main-Group Elements above the (c6) centroid i.e. much more strongly than in the [Ga(Ar),]+ species. In this general area it may be noted that there has been a theoretical analysis of the related (neutral) species [M(C5H5)] where M = In or Tl.42 Other gallium work reported in 1984 includes the potentially interesting species K[GaH2(CH2SiMe3)2] prepared from [Ga(CH2SiMe3)2Br] and 2KH in DME at 25°C.43 The compound was synthesized in the hope that reaction with [Ga(CH2SiMe3)2Br] would yield [Ga(CH2SiMe3)2H] which would then reductively eliminate TMS to give the gallium( I) species. In the event and perhaps not surpris-ingly Ga metal [Ga(CH2SiMe3)3] and H2 were formed instead.Apart from the aluminium-magnesium-anthracenespecies mentioned in Section 3 above [structure ( 13)],26 aluminium work noted in 1984 has included the formation of an unusually stable aluminium-alkyl bond in the species [EtAl( N4C22H22)] [schematic structure (21)].44 Reaction of &Et6 in hexane with the precursor [N4C22H24] but heating [schematic structure (22)] yields initially [Al( N4C22H23)Et2] of the latter in the solid state at 100°C results in the quantitative elimination of more ethane to give the product (21) which can be crystallized from hydroxyl- containing solvents such as water. Acid conditions e.g. HCl (immediately) and PhOH (much less readily) cleave the Al-C bond as does photolysis but perchloric acid protonates the methine carbons leaving the Al-C bond intact this last may therefore be a step in the HC1 acidolysis.44 Other Al-C bonds stable to thermolysis (though not now to air and moisture) are known to result from the complexing of aluminium alkyls with oxyanions.A 1984 example is K2[A14Me,2S04] formed by the stoicheiometric reaction between 4AlMe3 and a K2S04 suspension in aromatic solvents.45 The [S(0AlMe3),l2- anion has a straightforward tetrahedral configuration about sulphur and the compound interestingly forms a p-xylene clathrate K4[Al4Mel2S0412[C6H4Me2l- 5 Group IV Once again the organometallic chemistry of Group IV is the best represented of the main-group metals organosilicon chemistry in particular reseiving most attention.Again areas of new behaviour which figure largely are those of small ‘strained’ ring compounds multiple bonds to metals the metal( 11) valency state the consequences 42 E. Canadell 0. Eisenstein and J. Rubio Organometallics 1984 3 759. 43 0. T. Beachly and R. B. Hallock Organornetaflics 1984 3 199. 44 V. L. Goedken H. Ito and T. Ito J. Chem. SOC. Chem. Commun. 1984 1453. 45 R.D. Rogers and J. L. Atwood Organometallics 1984 3. 271. J. D. Kennedy of bulky substituents on the metal and the reaction chemistry involving supposed or real reactive intermediates such as free radicals and monometallenes. The interest in these types of behaviour is of course independent of general reaction chemistry and in this context the book Carbon-Functional Organosilicon Compounds46 will be of interest to chemists who are concerned with organic syntheses based .on organosilicon chemistry.The study of multiple bonds between Group IV metals is now approaching maturity; there are two review articles in the area,47,48 a very efficient concise survey of many of the salient points? and there is continued structural The species [Ge{CH(SiMe,),},] now made in improved yield from [GeCl,(dioxane)] and the new Grignard reagent [MgC1{CH(SiMe,),}(OEt2)] is monomeric in the orange-red gas phase with Ge-C 204(2) pm and the angle C-Ge-C 107(2)" but in the solid phase it is a bright yellow dimer m.p. 182"C with a Ge-Ge distance of 234.7(2) pm.49*52 The structural characterization of the dimer now permits the com- parison of M=M 'double' bonds in four periods of Group IV (Table 1).All have Table 1 Some structural parameters for crystalline Group IV dimetallenes M2Ria M C Si Ge Sn R' Sum of angles Z at M/" Ph 360 C,H2Me3-2,4,6 355.3 CH( SiMe3)2 348.5 CH(SiMe3)2 342 Fold angle O/" Twist angle /" (C(sp2 or sp3)-M)/A M-M In tetrahedral M,/A M-MIA 0 8.4 1.494 1.356 1.545 18 5 1.88 2.160 2.352 32 0 2.00 2.347( 2) 2.445 41 0 2.28 2.764( 2) 2.810 M-M Bond shortening in M,R compared with M, % 12 8 4 2 (a)Data from ref. 49 and other refs. cited therein. the trans-folded configuration [structure (23)] and it can be seen that there is a general decrease in formal double-bond character down the sequence C + Si -+ Ge -+ Sn accompanied by an increase in the folding angle 8.It will be interesting to see structural work on the lead compound first reported as a purple solid nearly a decade ago.53 It is appropriate to note here that there is an error associated with 46 'Carbon-Functional Organosilicon Compounds' ed. V. Chvalovski and J. M. Bellama Plenum 1984. 47 A. H. Cowley Polyhedron 1984 3 389. 48 A. H. Cowley. Acc. Chem. Res. 1984 17. 386. 49 P. B. Hitchcock Michael F. Lappert S. J. Miles and A. J. Thorne J. Chem Soc. Chem. Commun. 1984 480. S. Masamune S. Murakami J. T. Snow H. Tobita and D. J. Williams Organometallics 1984 3 333. 51 M. J. Fink J. Michalczyk K. J. Haller R. West and J. Michl Organometallics 1984 3 793. 52 T. Fjeldberg A. Haaland B. E. R. Schilling H.V. Volden M. F. Lappert and A. J. Thorne J. Organomel. Chem. 1984 276 C1. 53 P. J. Davidson D. H. Harris and M. F. Lappert 1. Chem. Soc. Dalton Trans. 1976. 2268. Organometallic Chemistry -Part ( ii) Main-Group Elements work reported in this area in 1983 (see footnote 2 in ref. 49); this reporter confesses to not spotting this in last year's report and also therein of wrongly transcribing the formula [Sn{N(SiMe3)2)21.54 The dissociation of the {Ge,} and {Sn,} species to their respective monomers is mirrored somewhat in the sterically hindered but otherwise conventionally singly bonded hexamesityldisilane and he~arnesityldigermane.~~ In these the M- M bonds dissociate homolytically and reversibly between -60 and -32 "C for Si(AHdiss ca.79kJmol-') and between -12 and +53"C for Ge (AHdissca.86kJmol-'). The radicals that are generated react irreversibly for example by substitution of aromatic species or by abstracting hydrogen.,' thf O+ The chemistry of compounds with multiple bonds from silicon to other elements continues to progress steadily and a review of unsaturated Si and Ge compounds of the type R2M=C(SiR3)* and R,M=N(SiR,) has been published.56 The structure of the silaethene species [Me,Si=C( SiMe,)( SiMeBu',)(thf)] has been determined [schematic drawing (24)]. The carbon is planar but there is some pyramidalization at silicon and a zwitterionic contribution to the bonding has been postulated [structures (25)].,' Unsubstituted silaethene H2Si=CH2 has been isolated at low temperatures in an argon matrix and its interconversion with methylsilylene via a photochemically induced 1 + 2 shift [equation (7)] has been studied.'8p59 In related argon matrix work silabenzene [HSiC5H5] has also been isolated.60 Other compounds with a formal metal(I1) valency state are the metallocene species which also attract steadily progressing experimentation.Monomeric germanocene [Ge(C5HJ2] previously thought to be polymeric in the solid state has now been prepared in pure form from [Na( C'H,)] and [GeCl,( dioxane)] in THF solution in 60% yield.61 The compound is stable for weeks at -30 "C and has the open sandwich structure also exhibited by [Sn(C,H,),]. The angle between the rings at 50.4" is somewhat larger than in [SII(C,H,)~] even though the Ge atom is the smaller and 54 M.F. Lappert personal communication 1985. The reporter thanks Professor Lappert for drawing his attention to this matter. 55 W. P. Neumann K.-D. Schultz and R. Vieler J. Organornet. Chem 1984,264 179. 56 N. Wiberg J. Organornet. Chern. 1984 273 141. 57 N. Wiberg G. Wagner G. Muller and J. Riede J. Organornet. Chern. 1984 271 381. 58 G. Maier G. Mihm and H. P. Reisenauer Chem Ber. 1984 117 2351. 59 G. Maier G. Mihm H. P. Reisenauer and D. Littman Chem Ber. 1984 117 2369. 60 G. Maier G. Mihm R. 0. W. Baumgartner and H. P. Reisenauer Chem Ber. 1984 117 2337. 61 M. Grew E. Hahm W.-W. duhlont and J. Pickardt Angew. Chern. Znt. Ed. EngL 1984 23 61. J. D. Kennedy the divergence among the Ge-C distances (235-273 pm) is also more pronounced than the corresponding ones in the tin compound.61 Sterically bulky substituents on the C5 rings tend to reduce these angles; [Sn(C,Me,),] has a similar angle to [Sn(C5H5)2] but in [Sn{C5H2(SiMe3)3}2] made according to Scheme 2 the angle is reduced to 18°,62 and in [Sn(C5Ph5),] the two planes are parallel even though the (mean) Sn-C distance of 269.2(8) pm is very similar to the more open species.63 The perphenylated compound is made in 56% yield in a straightforward manner by the treatment of C5Ph5Br with BuLi followed by SnC12.63 Associated with this compound are a number of ‘Zuckerman Claims’ (without which over the years the literature of organometallic chemistry would be much the Of these ‘the first symmetrical main-group sandwich com- pound’ is of course reasonable but ‘the first example of a molecular main-group species in which the lone pair is stereochemically inert’ and ‘the first molecule to violate decisively the VSEPR theory’ will presumably run into flak.(27) (26) Scheme 3 Related to these metallocenes and to some of the lithium zinc and gallium work cited above are the closo cluster stannacarboranes of general configuration (26) which can be made by the bridge-deprotonation of the nido precursor (27) in THF followed by addition of SnC1 (Scheme 3).65*66 In some related polyhedral carborane chemistry use is made of {SiMe3} as a leaving group well known in organic synthesis to generate the nido twelve-vertex species [(Me3Si)2C4BsH10] oia the thermal elimi-nation of Me3SiH from nido-[(Me3Si)2C2B4H6].67 It is a healthy sign to see links such as this between the organometallic and polyhedral sub-disciplines of main-group chemistry and it is obviously an area of good potential for new chemistry not only with Group IV organometallics.62 A. H. Cowley P. Jutzi F. X. Kohl J. G. Lasch N. C. Norman and E. Schliiter Angew. Chem. Int. Ed. Engl. 1984 23 616. 63 M. J. Heeg C. Janiak and J. J. Zuckerman J. Am. Chem. SOC.,1984 106 4259. J. J. Zuckerman as quoted in Chem. Eng. News 1984 62,‘20. 65 A. H. Cowley P. Galow N. S. Hosmane P. Jutzi and N. C. Norman J. Chem. SOC.,Chem. Commun. 1984 1564. 66 N. S. Hosmane. N. N. Sirmokadam and R. H. Herber Organomerallics 1984 3 1665. 61 N.S. Hosmane M. Dehghan and S. Davies J. Am. Chem. Soc.. 1984 106 6435. Organometallic Chemistry -Part (ii) Main-Group Elements 24 1 The more straightforward divalent and/or multiply bonded Group IV organometallic species are often invoked as reaction intermediates. Interesting 1984 examples include an unusual silanediyl-germanediyl exchange.68 In this process photochemically generated Me2% is reported to react quantitatively with the germole (28) to give the corresponding dole (29) and polygermanes (Me,Ge), with n =4 as the major polygermane product (Scheme 4). The reaction is thought to go via an initial addition to a germole double bond followed by equilibration of an intermediate such as (30).68 A second 1984 example is the generation of the true silicones R2Si=0 (known as silanones) by thermolysis of the 6-oxa-3-silabicyclo[ 3.1 .O]hexanes (3 1) in a flow system at 460 "C (Scheme 5).69The reaction also works for the corresponding germanones.The transient silanones polymerize to cyclic siloxanes (61-70%) or by using a 1,3,2-disilaoxapentane as a trap the corresponding dioxatrisilacycloheptanes may be obtained in 41 O/O yield. Me Me Me Me \/ Ge \si/ jjph -Me,Si + ;)ph +(Me,Ge), I I Mc Me Me Me Me Me \/ \/ Gi Ph,&Ph / Ph / Me I Ph (30) Scheme4 75-80% Scheme 5 An interesting extension of this is to start with the spiro compound (32) which results in a 60% yield of the spirosiloxane of configuration (33) when the cyclo- trisiloxane (Me2SiO) is used as a trapping agent.It would be nice to think of monomeric silica O=Si=O as the intermediate but reality will probably assert that the reactive >Si=O units are generated ~tepwise.~~ 68 J. A. Hawari and D. Griller J. Chem. Soc. Chem. Commun. 1984 1160, 69 G. Manuel G. Bertrand W. P. Weber and S. A. Kazoura Organometallics 1984 3 1340. rcsi3Me J. D. Kennedy 0-Si-0 0-Si-0 / \/ \ 490°C 0-(Me,SiO) Si\ Si /Si 0-si-0' \o-si -0 (32) (33) As usual there h'as been much small-ring chemistry reported either incorporating silicon as one of the ring members or as a substituent; sometimes bulky organic groups are used for stability and sometimes the organosilicon group itself constitutes the stabilizing substituent. Examples include the first single-crystal X-ray structural analysis of a silacyclobutene (34),70the first isolated siloxetane (35),71octamethyl-spiropentasilane of configuration (36),72 and the boron-containing heterocycles (37),73(38),74 and (39).75The last species (39) is of interest because it is believed to topomerize as indicated (AG* CQ.48 kJ mol-') but calculations suggest that structures such as (40) may be the most stable and that topomerization may occur via structures such as (39) and (41) as intermediate^.^^ Somewhat in accord with this in its reaction chemistry the compound behaves as if it is of configuration (41) and undergoes a number of carbene-like reactions.77 Bu'Me,Si /SiMe,Bu' \ Me,Si SiMe, N-N \/ \/ c=c B \/ I B N I SiPh, Me,Si /\SiMe (38) (37) 70 M.Ishikawa S. Matsuzawa K. Hirotsu S. Kamitori and T. Higuchi Organometallics 1984 3 1930. 71 A. Sekiguchi and W. Ando J. Am. Chem. Soc. 1984 106 1486. 72 P. Boudjouk and R. Sooriyakumaran 1. Chem. SOC.,Chem. Commun.,1984 777. 73 U. Klingebiel Angew Chem. Znf. Ed. Engl. 1984 23 815. 74 B. Pachaly and R. West Angew. Chem. Znt. Ed. Engl. 1984 23 454. 75 H. Klusik and A. Berndt Angew. Chem. Int. Ed. Engl. 1983 22 877. ?6 B. H. M. Budzelaar P. von R. Schleyer and K. Krogh-Jespersen Angew. Chem. Znt. Ed. Engl. 1984 23 825. 77 R. Wehrmann H. Klusik and A. Berndt Angew. Chem. Znt. Ed. Engl. 1984.23 826. Organometallic Chemistry -Part ( ii) Main-Group Elements 243 A large amount of the chemistry of the silacyclopropanes and silacyclopropenes first reported in a preliminary fashion in 1977 has now been reported in some The silacyclopropenes are much more reactive than the silacyclopropanes and reactions with aldehydes ketones styrenes conjugated terminal acetylenes benzyne terminal 1,3-dienes and a conjugated imine are reported.These reactions generally give five-membered cyclic organosilicon products in which the C=O C=C CEC or C=N bonds of the organic reagents have inserted into the Si-C bond of the silirene ring. C-C insertions and acyclic products isomeric with the cyclic ones are also observed and the available evidence is taken to suggest that a radical mechanism is operative.79 Apart from the aspects mentioned abo~e,~~~~*~~*~~*~~ organogermanium chemistry is not well represented this year.A review of structural organogermanium chemistry may be noted,80 and also the start of the use of 73Ge n.m.r spectroscopy in this INEPT n.m.r. techniques are found to be helpful,82 and there is a confirmation of the linear relationship between the chemical shifts S(29Si) and 6 (73Ge) for equivalent organosilicon and organogermanium species.83 In organotin chemistry too there is little novel to note. Again some developments have been dealt with above.62d6 There is the hardy perennial interest in the auto-associaton of organotin oxides and alkoxides but this now appears to be on the wane. In this area the solid-state structure of the simple 2,2-dibutyl-1,3,2- dioxastannolane has been found to be an infinite ribbon of six-co-ordinate tin atoms in highly distorted octahedral sites.84 This result indicates that conclusions arising from Mossbauer data on the same compoundss need to be re-examined.More complete details on the initial work on the ‘Lewis acid crowns’ mentioned in last year’s report,2 have been published,86 and a potential ‘Lewis acid cryptand’ PhSn((CH,),},SnPh has been reported by the same Unfortunately in this compound the potentially cryptating cavity appears to be already full of hydrogen atoms from the polymethylene chains. Since the discovery in the early 1970s of the interesting variety of stereospecificity in reactions between organostannylalkali reagents such as Me,SnNa and organic halide substrate^,^^^^^ there has been a lot of careful experimentation in the area which has not been without controversy and the papers in the field often make interesting reading.Reference 90 is a good lead-in to this literature but it mainly deals with a meticulous examination of the reaction between Me3SnM (where M = Li Na or K) with organic bromides in various solvent systems and with the establishment of the relative contributions of electron-transfer and SN2 mechanisms 78 D. Seyferth D. P. Duncan M. L. Shannon and E. W. Goldman Organometallics 1984 3 574. 79 D. Seyferth S. C. Vick and M. L. Shannon Organometallics 1984 3 1897. 80 K. C. Molloy and J. J. Zuckerman Adu. Inorg. Chem. Radiochem. 1983 27 113. 81 I. P.Sekatsis E. Liepins I. A. Zicmane and E. Lukevits Zh. Obshch. Khim. 1983 53 2064.82 K. M. Mackay P. J. Watkinson and A. L. Wilkins J. Chem SOC. Dalton Trans. 1984 133. 83 Y. Takeuchi T. Harazono and N. Kakimoto Inorg. Chem. 1984 23,3835. 84 A. G.Davies A. J. Price H. M. Dawes and M. B. Hursthouse J. Organomet. Chem. 1984 270 C1. 85 R. H. Herber A. Shanzer and J. Libman Organometallics 1984 3 586. 86 Y. Azuma and M. Newcomb Organornetallics 1984 3 9. 87 . M. Newcomb M. T. Blanda Y. Azuma and T. J. Delord J. Chem. SOC. Chem. Commun. 1984 1159. 88 G. S. Koermer M. L. Hall and T. G. Traylor J. Am. Chem. SOC.,1972 94 7205. 89 H. G. Kuivila J. L. Considine and J. D. Kennedy J. Am. Chem. SOC., 1972 94 7206. 244 J. D. Kennedy to the reaction.” Contributions identified in other systems also include metal- halogen exchange processes and it is apparent that the balance between these three mechanisms depends intimately on the nature of the organostannylalkali species in solution (compare Sections 2 and 3 above).Preliminary multielement n.m.r. work has shown large changes of n.m.r. parameters with solution conditions,” and further n.m.r. work allied with solid-state structural investigations will be particularly valuable. Other mechanistic work includes a study of the reaction between organotin hydrides and acid chlorides to give aldehydes and esters as summarized in equation (8).92Long believed to be a free-radical process this is now thought o be a non-radical mechanism in which the initial products are RCHO and RGSnCI the remaining products being formed by subsequent reaction of the aldehyde.For example R$SnOCH,R will be formed from the aldehyde and the tin hydride and can then react further with RCOCl to give RC0.0CH2R with RCHO to give R;SnOCHROCH,R and with RiSnH to give RCH20H?2 RCOCl + RjSnH -RjSnCI + RCHO + RCO.OCH,R (8) It has been found that the reaction between SnS and Me1 in H20 yields MeSnI in 33% yield it is reasonably claimed that this may bear on the ubiquitous occurrence of methylstannanes in the environment and also that it represents a convenient one-step synthesis of MeSn13.93 The implication that previously reported syntheses are much less simple or much more inconvenient is less reasonable however. For example it is very easy to devise a reaction between Me1 and the perhaps more realistic starting material SnC12 to give an 87% yield of MeSnI in a one-pot process:4 which may be compared to the overall yield of 25% in a two-step process from the same starting materials that is implicit in the more recent report.93 Theoretical studies do not usually come within the scope of this report but organotin chemists will be interested to note that MNDO parameters are now available for tin and MNDO treatments have been applied,(with some success it is claimed) to four topics of current interest in organotin chemistry?’ This has led to satisfactory interpretations of the mechanism of hydrostannylation structures of the sandwich and half-sandwich cyclopentadienyltin compounds the possibility of multiple bonding by tin in distannene and dimethylmethylenestannene,and the geometry of the trimethylstannyl radical?’ 6 Groups V and VI Organometallic compounds of these later main-group metals and metalloids continue to be a happy hunting ground for transition-metal chemists in their insatiable search for ligands.Even the heavier elements are now no longer immune 1984 examples M. S. Alnajjar and H. G. Kuivila J. Am. Chem. SOC.,1985 107 416. 91 J. D. Kennedy and W. McFarlane J. Chem. SOC.,Chem. Commun. 1974 983; J. Chem SOC.,Dalton Trans. 1976 1219. 92 L. Lusztyk E. Lusztyk B. Maillard and K. U. Ingold J. Am. Chem. SOC.,1984 106 2923. 93 W. F. Manders G. J. Olson F. E. Brinckman and J. M. Bellama 1. Chem. SOC.,Chem. Commun. 1984 538. 94 J. D. Kennedy J. Labelled Compd. 1975 11 285.95 M. J. S. Dewar G. L. Grady D. R. Kuhn and K. M. Men. J. Am. Chem. Soc. 1984 106 6773. Organometallic Chemistry -Part (ii) Main-Group Elements including the use of Sb2Ph2 as a bridging ligand between two rhenium centres96 and a kinetic study of the oxidative addition of Te2Ar2 to the iridium(1) complex [Ir(CO)Cl(PPh3),].97 Another in the same area is the reaction of Bi,Ph4 with [co,(co)g] to give [CO(B~P~,)(CO)~].~~ The study of these metal-metal bonded species such as the distibines and dibis- muthines in a non-transition-metal context continues to be an area of progress. The molecular structure of Sb2Me2 a pale yellow liquid which freezes at +17 "C to give a bright red solid has been determined at -160"C.99 The structure consists of essentially collinear chains of Sb atoms with alternating short and long inter- antimony distances of 286.2(2) and 364.5(1) pm similar to those of the higher (bistibole) homologues mentioned in previous reports2 A new thermochromic dibismuthine 2,2',5,5'-tetramethylbibismole (42) which is red in solution but black with a greenish-blue lustre in the solid state has been made.'" The synthesis from 1,4-dilithiobutadiene is given in equations (9) and (lo) and proceeds uia the potentially aromatic bismolyl anion (43).The compound (42) is air-sensitive but can be stored at room temperature as the solid m.p. 95 "C. The crystal packing is also believed to be similar to the thermochromic bistibole mentioned in the 1982 report2 The corresponding biarsole is non-thermochromic.'Oo In these general areas the main-group o-framework permutational chemists are just as voracious as the transition-metal chemists mentioned in the first paragraph of this Section and in accord with this it appears that the derivative chemistry of these M-M species may also now be receiving increasing attenti~n.'~'-'~~ Thus for example Sb,Me4 and Sb2Et4 have been found to react with elemental Se or Te to give compounds of the general formulation R2Sb-E-SbR2 where E = Se or Te,"' and Bi2Me4 reacts with PhEEPh where E = S Se or Te to give compounds of general formula Me2BiEPh.lo2 Similarly Bi2Me4 reacts with stoicheiometric amounts of O2or sg to give essentially quantitative yields of yellow [(Me,Bi),S] and colourless explosion-prone [( Me2Bi)20],'03 and the reaction of PhLi with either Bi2Me4 or 96 1.Bernal J. D. Korp F. Calderazzo R. Poli and D. Vitali J. Chem. SOC.,Dalton Trans. 1984 1945. 97 R. T. Mehdi and J. D. Miller J. Chem. SOC.,Dalton Trans. 1984 1065. 98 F. Calderazzo R. Poli and G. Pelizzi J. Chem. Soc. Dalton Trans. 1984 2535. 99 A. J. Ashe E. G. Ludwig J. Oleksyszyn and J. C. Huffman Organometallics 1984 3 337. 100 A. J. Ashe and F. J. Drone Organometallics 1984 3 495. 101 H. J. Breunig and H. Jawad J. Organomet. Chem. 1984 277 257. 102 M. Wieber and I. Sauer 2. Naturforsch. Teil B 1984 39 1668. 103 M. Wieber and 1. Sauer Z. Naturforsch. Teil B 1984 39 887. J. D. Kennedy Me2BiBr gives good yields of BiMe2Ph as a colourless very oxygen-sensitive 1iq~id.l'~ Other insertions into the Bi-Bi bond include the reaction of Bi2Ph2 with p-benzoquinone to give pale-yellow [Ph2Bi-OC6H40-BiPh2] in 87% yield and with diazomethane to give [(Ph2Bi)2CH2] in 49% yield.lo4 In a related area yellow unstable [S(NBiBu:),] is made from BuiBiBr and K2SN2in CH3CN solution and the yellow-orange {BuiSb} and yellow {Ph2As} derivatives which are more stable are made ~imilarly.''~ In addition to these singly-bound Group V M-M species there is a continuing development in the chemistry of the multiply bound species RM=MR.This area has been and a comprehensive experimental account has also been presented.lo6 The final compound in this year's report is the dark-red mouldy-smelling orange solid As4Bui which is unstable even at -78 OC.'07 It is made from Bu'AsC12 according to equation (11) and is believed to have the butterfly 2,4-di-t-butyl-bicyclo[l.l.O]tetraarsane structure (44).It is a member of the set of compounds which also includes As,Bu\ [structure (45)] and As,Buk [structure (46)] previously reported by the same school in 1981. 2ButAsC1 + 2AsC1 + lOLiH- As,Bu + lOLiCl + 5H (11) Bu' But Bu' Bu'A s I ,AsBu' /AS.As-AS \ \As/As-As Bu' Bu' 104 F. Calderazzo R. Poli and G. Pelizzi J. Chem. SOC.,Dalton Trans. 1984 2365. 105 M. Herberhold W. Ehrenreich and K. Guldner Chem. Ber. 1984 117 1999. 106 A. H. Cowley J. E. Kilduff J. G. Lasch S. K. Mehrotra N. C. Norman M. Pakulski B. R. Whittlesey J. L. Atwood and W. E. Hunter Inorg. Chem. 1984 23 2582.107 M. Baudler and S. Wietfeldt-Haltenhoff Angew. Chem. Znt. Ed. Engl. 1984 23 379.