摘要:

DALTON PERSPECTIVE J. Chem. Soc., Dalton Trans., 1998, 3343–3349 3343 Sterically protected organophosphorus compounds in low co-ordination states Masaaki Yoshifuji Department of Chemistry, Graduate School of Science, Tohoku University, Aoba, Sendai 980-8578, Japan Received 23rd April 1998, Accepted 19th June 1998 The chemistry of sterically protected low co-ordinate organophosphorus compounds is described; these compounds comprise diphosphenes, phosphaethenes, phosphaallenes, phosphabutatrienes, phosphinidenecyclobutenes, and so forth, and are discussed in terms of syntheses, 31P NMR analyses, X-ray crystallographic structure analyses, as well as physical and chemical properties. 1 Introduction One hundred and twenty years ago, Köhler and Michaelis 1 reported on phosphobenzene ‘PhP]] PPh’, as a phosphorus analogue of azobenzene, as a reaction product of dichlorophenylphosphine and phenylphosphine. This was the first report on a compound containing the P]] P bond, but later it was shown that the compound was actually oligomers of PhP units, being a tetramer or pentamer in solution 2 and a pentamer or a hexamer in the solid state.3 Subsequently, chemists generally believed that compounds such as ‘phosphobenzene’ cannot exist as stable entities, since according to the ‘classical doublebond rule’ double bonding involving the heavier main group elements is unstable,4 due to the hypothesis that the energy gain on formation of p bonding with vast but dilute electron lobes is not large through a long bond distance; i.e.electrons in the 3s, 3p or 3d orbitals do not form a strong bond because of the long bond length and unfavourable hybridization of the sp2 configuration. Thus much eVort had been paid to thermodynamic stabilization of these multiple bonds by electronic or delocalization eVects; indeed, Dimroth and HoVmann5 reported phosphacyanins and Märkl 6 reported phosphabenzenes, both having a delocalized P]] C bond.However, in 1978, utilizing the mesityl group as a sterically protecting moiety (hereafter abbreviated to Mes), Bickelhaupt Masaaki Yoshifuji was born in Tokyo in 1941. He received a bachelor degree in 1966, master degree in 1968, and Doctor of Masaaki Yoshifuji Science in 1971 from The University of Tokyo. He was appointed as Research Associate in 1971 and was promoted to Associate Professor at Tokyo (Faculty of Science) in 1988. He has been Professor at Tohoku University (Graduate School of Science) since 1989.His research interests are unusual organic compounds including heteroatoms in terms of structure, chemical reactivities and physical properties. and co-workers 7 reported on a phosphaethene with a P]] C bond (1), and in 1981, utilizing the 2,4,6-tri-tert-butylphenyl group I reported on a diphosphene with a P]] P bond (2).8 It should be noted that, in 1981, Becker et al.9 reported a phosphaalkyne with a P]] ] C bond protected with the tert-butyl group (3), and that in the field of silicon chemistry West et al.10 reported the first isolated disilene with a Si]] Si bond (4) bearing four mesityl groups (Scheme 1).These compounds are the earliest examples of localized multiple bonding which violates the classical ‘double bond rule’. Since they involve interesting bonding and novel structures, studies on the chemistry of multiple bonds of the heavier main-group elements have been carried out in various fields of chemistry including theory, chemical reactions, photochemistry, co-ordination chemistry and electrochemistry.The chemistry thus newly created for multiply bonded phosphorus compounds was generally regarded as a renaissance of phosphorus chemistry. Indeed, there have been several review articles and books 11–21 dealing with this new chemistry, since these compounds are supposed to be highly reactive unless reasonably protected, and even when being suYciently protected to be analysed they are still chemically reactive. 2 Low co-ordinated organophosphorus compounds Thus, this article focuses on the chemistry of sterically protected organophosphorus compounds in low co-ordination states, but mainly limited to my results concerning the bonding Scheme 1 P C Me Me P P C Me Me Me C Me Me Me C Me Me Me C Me Me Me C Me Me Me C Me Me Me Me Me C Me Me C P Si Si Me Me Me Me Me Me Me Me Me Me Me Me 2 1 3 43344 J. Chem. Soc., Dalton Trans., 1998, 3343–3349 with P]] X (X = P, C, As or Si), P]] C]] X (X = P, C or N), P]] C]] C]] X (X = P or C) and C]] ] P. 2.1 Diphosphenes The sterically protected diphosphene 2 was prepared as a stable compound by dechlorination reaction of the corresponding phosphonous dichloride 5 with magnesium metal for the first time, as shown in equation (1).8 Electron transfer reagents such as lithium dihydronaphthylide can be used in place of magnesium for the dechlorination of phosphonous dichlorides. There are several other methods to prepare diphosphenes especially for unsymmetrical diphosphenes such as 6.Starting from the corresponding primary phosphine 7 and phosphonous dichloride 8, the unsymmetrical diphosphene 6 is available in the presence of a base such as triethylamine or 1,8-diazabicyclo- [5.4.0]undec-7-ene (dbu), as shown in equation (2).22 From a mixture of two diVerent phosphonous dichlorides, unsymmetrical diphosphenes can be obtained by dechlorination reaction, together with the symmetrical diphosphenes, but careful purification such as column chromatography is required, to obtain the desired unsymmetrical diphosphenes, where R, R9 = 2,4,6-But 3C6H2, tri-tert-pentylphenyl,23 2,6-dimesityl-4-methylphenyl, 24 and so on, as shown generally in equation (3).At the beginning of my research on diphosphenes a reaction mechanism was postulated which involves the corresponding phosphinidene as an intermediate,8 a phosphorus analogue of nitrene, which subsequently forms the diphosphene, a formal dimer of the phosphinidene.However, it is now known that phosphinidenes such as 2,4,6-tri-tert-butylphenylphosphinidene (A), if formed, tend intramolecularly to insert into the C]H bond of one of the methyl groups of the ortho tert-butyls of the protecting group to form a phosphaindane derivative 9,25 rather than dimerize to the diphosphene. Thus the formation of a diphosphene, through a formal dimerization of diphosphene, might be explainable by an alternative intermediate such as a ‘phosphinidenoid’, RP(Cl)MgCl.Photolysis of diphosphenes has been investigated to determine whether the products depend on the wavelength of irradiation; irradiation with a mercury lamp without a filter causes the formation of phosphaindane 9, while irradiation through a Pyrex filter 25 or 514.5 nm laser light 26 causes E/Z isomerization at low temperature [equation (4); R = 2,4,6-But 3C6H2]. The isomerized Z-2 returns to E-2 upon warming, due to enormous steric congestion between the two aryl groups in the Z isomer.However, it is not yet clear whether a biradical B is really involved in the E/Z isomerization of 2, while Bickelhaupt and co-workers 27 PCl2 But But But P P But But But But But But (1) 2 5 Mg Mes PCl2 PH2 But But But P Mes P But But But 8 base + (2) 6 7 RPCl2 R¢PCl2 (3) + Mg + + R¢P PR¢ RP PR¢ RP PR experimentally determined the activation parameters for the thermal isomerization reaction of 1-(2,4,6-tri-tert-butylphenyl)- 2-(2,4,6-triisopropylphenyl)diphosphene, as DH‡ = 29.5 ± 1.4 kcal21 and DS‡ = 38 ± 6 cal K21 mol21 (cal = 4.184 J).Although a nitrogen atom is directly attached to one of the phosphorus atoms, Niecke et al.28,29 reported preparations and X-ray analyses of Z-aminodiphosphenes, which are stable at room temperature. Very recently, utilizing 1,2-bis(2-bromo-3,5- di-tert-butylphenyl)ethane as a protecting group, an internal cis-diphosphene Z-10 of the o-cyclophane type was prepared together with 2 from a bis(diphosphene) E,E-11, as shown in equation (5),30 where the reaction might proceed intramolecularly via olefin metathesis or similar mechanism.The structure of Z-10 was confirmed by X-ray analysis of the tungsten complex of the ligand. Under certain conditions, unsymmetrical diphosphenes RP]] PR9, where R = 2,4,6-But 3C6H2, R9 = Mes,31 2,4,6-tris(trifluoromethyl)phenyl,32 pentamethylcyclopentadienyl, 33 (Me3Si)2N(Me3Si)N 28 or tert-butylamino,29 gave 2 (RP]] PR), indicating that disproportionation occurred upon irradiation or warming in solutions. Although Gaspar and coworkers 34 reported direct observation of mesitylphosphinidene by ESR spectroscopy, I could not obtain any proof to support the generation of phosphinidene species, even at cryogenic temperature during the photolysis of the phosphinidene precursors. 35 On the other hand, diphosphenes of the Group 6 metal carbonyls also caused E/Z photoisomerization upon irradiation by UV light.36 2.2 Phosphaarsenes Phosphorus compounds with P]] As bonding can be prepared by the reaction of primary phosphines with dichloroarsines in the presence of a base such as dbu.37 Using an o-diarylphenyl substituent, such as the 2,6-dimesityl-4-methylphenyl, the phosphaarsene was prepared as a stable orange solid material.24 2.3 Phosphaalkenes A sterically protected phosphaalkene, 1-mesityl-2,2-diphenyl-1- phosphaethene (1), was prepared by Bickelhaupt and coworkers 7,38 as the first example of the phosphaethene with a localized P]] C bond [equation (6)]. 2.3.1 Phosphaethenes. Since 2,4,6-But 3C6H2 turned out to be an eYcient protecting group to stabilize low co-ordinated organophosphorus compounds, various phosphaalkenes with it PH Me Me But But R P R P R P R P R P R P R P (4) hn hn Z-2 E-2 B A 9 But But P P But But But But P But But P P R P R (5) hn 2 E,E-11 Z-10 + Mes P CHPh2 Cl Mes P C Ph Ph dbu (6) 1J.Chem. Soc., Dalton Trans., 1998, 3343–3349 3345 have been prepared from the corresponding silylphosphide and aldehydes or ketones [equation (7)], by the phospha-Peterson reaction.39,40 Furthermore, photoisomerization of E-12 was attained to give E/Z isomers.40–42 This method can generally be applied to the preparation of phosphaethenes and was useful to prepare phosphaalkene 13 bearing the novel aromatic compound azulene.43 GeoVroy and co-workers 44–46 and I47 reported the formation of 1,2-, 1,3- and 1,4-bis(2-phosphaethenyl) benzene derivatives 14, prepared from the corresponding dialdehydes. 2.3.2 Diphosphinidenecyclobutenes. Appel et al.,48 Märkl 49 and I 50,51 reported the synthesis and structure of diphosphinidenecyclobutene 15 (R = 2,4,6-But 3C6H2, R9 = Ph or SiMe3), as shown in equation (8). The system seems to involve both the buta-1,3-diene and the hexa-1,3,5-triene system, within a molecule. Furthermore, the 2,4,6-But 3C6H2 groups can take both the E and Z configuration.It is of interest that the direction of photoequilibrium depends on the substituents; if R9 = SiMe3 or Ph, the E,E configuration is major, while if R9 = H the E,Z configuration is major, probably due to steric congestion between the two R groups and the substituents R9. Needless to say, the Z,Z configuration is hard to achieve because of the serious steric congestion between the two R groups. Iodine-induced E/Z isomerization was observed in the case of diphosphinidenecyclobutene, and this is a good method for preparation of E,Z-phosphinidenecyclobutenes.52 On the other hand, we have developed several protecting groups other than 2,4,6-But 3C6H2 and found that 2,4-di-tertbutyl- 6-methylphenyl is useful to stabilize the diphosphinidenecyclobutene system.In addition to the E/Z configurational isomerism, as mentioned above, conformational isomerism was observed, due to the restricted rotation around the two P]C bonds at the edges of the system.The compound 16 consisted of two rotamers, syn and anti isomers, and after addition of [M(CO)5(thf)] to this mixture, where M is a Group 6 metal, the 31P NMR of the reaction products appeared as a mixture of syn-17 and anti-17 with complete retention of conformation [equation (9)]. The structure of syn-17, where M = W, was Li P R SiMe3 R P C Ph Ph (7) Ph2CO R P C H Ph R P C Ph H PhCHO hn E-12 Z-12 C P H C P H R R C P H R 1 2 3 4 13 14 RPHC CR¢ P R¢ R¢ P R R R¢ R¢ C RP C RP P P R R R¢ R¢ P R¢ R¢ P R R 2) BrCH2CH2Br E,E-15 (8) 1) PhLi heat heat E,Z-15 hn unambiguously determined by X-ray analysis.53 On the other hand, anti-17, where M = W, was analysed by HPLC using a chiral column and a baseline separation was attained; each separated enantiomer showed a symmetrical CD spectrum.These results indicate that an asymmetric environment can be created in the diphosphinidenecyclobutene due to the restricted rotation. It is interesting that diphosphinidenecyclobutenes have various possible co-ordination modes toward transition metals.Indeed, some of them give s- and p-co-ordinated complexes with Group 6 metal carbonyls, as shown in 18 and 19.54 A similar complex was obtained in the case of 2,4,6-But 3C6H2 at the phosphorus and H at the cyclobutene ring. Although Rau and Behrens 55 reported on the formation and structural analysis of tricarbonyl(h6-dimethylenecyclobutene)chromium(0) and tricarbonyl( h6-diisopropylidenecyclobutene)chromium(0), 19 is the first example of the diphosphinidenecyclobutene–tungsten carbonyl system, where the ligand is doubly co-ordinated with tungsten metals, and of the bidentate ligation in 18. 2.3.3 Phosphasilene. Compounds with a P]] Si bond have been prepared, however they are not so stable even when suYciently protected with bulky substituents,56 probably due to a weak p bond (29 kcal mol21) as calculated by an ab initio method.57 2.4 Phosphacumulenes Phosphacumulenes, such as phosphaallenes, diphosphaallenes, phosphabutatrienes and diphosphabutatrienes, were prepared by various methods. 2.4.1 Phosphaallenes. 1-Phosphaallene 20 can be prepared by either the Peterson reaction,58 the phospha-Peterson reac- P P But Me But Me But But P P Me But But Me But But OC OC CO M OC P P Me But But Me But But P P But Me But Me But But OC OC CO M OC Ph Ph Ph Ph Ph Ph Ph Ph anti-17 anti-16 [M(CO)5(thf)] syn-17 syn-16 (9) M = Cr,Mo,W Ph Ph P P C6H2Pr i 3-2,4,6 C6H2Pr i 3-2,4,6 (OC)4W Ph Ph P P C6H2Pr i 3-2,4,6 C6H2Pr i 3-2,4,6 (OC)4W W(CO)3 19 183346 J.Chem. Soc., Dalton Trans., 1998, 3343–3349 tion,59,60 or the phosphorus version of the Doering–Moore– Skattebøl reaction (DMS reaction) 61 as shown in equation (10), R = 2,4,6-But 3C6H2. It should be mentioned here that stereochemistry is not preserved during the DMS reaction, since the dichlorophosphirane 21, obtained as a reaction intermediate, was identical either starting from E- or Z-phosphaethene.Furthermore, axial chirality exists around the P]] C]] C bond of phosphaallene 22, and the separation of enantiomers by a chiral HPLC column was successful, indicating that the racemization is not fast in the dark, while it is almost instantaneous upon irradiation with light.62 2.4.2 Diphosphaallenes. 1,3-Diphosphaallene 2463 can be prepared by various methods, as shown in equation (11), R = 2,4,6-But 3C6H2. For the first time the compound was prepared from the reaction of silylphosphide with carbon dioxide, via a phosphaketene 23 64 as an intermediate.Alternatively, it was prepared by the reaction of halogenophosphino-substituted phosphaalkenes with potassium tert-butoxide.65 Recently we have developed a phosphorus version of the DMS reaction of diphosphenes with dichlorocarbene, followed by ring opening of dihalogenodiphosphiranes with alkyllithium reagents [equation (11)].66 Thus an unsymmetrical diphosphaallene was prepared from the corresponding unsymmetrical diphosphene. 23 The structure of the 1,3-diphosphaallene 24 was determined by X-ray analysis by Karsch et al.67 It should be noted that similar axial chirality exists around the P]] C]] P bond, and the separation of enantiomers through a chiral HPLC column was successful,68 though the racemization took place on irradiation with light. 2.4.3 P]] C]] N. Phosphorus compounds with the P]] C]] N bond can be prepared by the reaction of silylphosphides with isocyanates, as shown in equation (12).59,69 This type of compound protected by the tert-butyl group was first reported by Kolodiazhnyi. 70 The reaction of the azaphosphaallene with water depends on the substituents R9 on the nitrogen.71 Phosphorus chemical shifts of azaphosphaallenes 25 appear at high field: d 2106.2 for R9 = Ph, 2101.9 for R9 = But and 2135.3 for R9 = R, probably due to the existence of a canonical structure with negative charge on the phosphorus, as depicted in Scheme 2.A similar high field shift is observed for 23.64 Li P R SiMe3 R P C C Ph Ph R P C SiMe3 Li R P C CHPh P CHPh R P CH CCl2 Ph R 20 Cl2C: Ph2C=C=O (10) Ph2C=O R¢Li 22 21 R P C PR P P CCl2 R R RP C O P C R P(SiMe3)R OLi RP C H P Cl R Cl2C: (11) R¢Li 24 CO2 RP(SiMe3)Li CH2[P(Cl)R]2 dbu KOBut 2 Li P R SiMe3 23 2.4.4 Phosphabutatrienes and diphosphabutatrienes. 4,4- Diphenyl-1-(2,4,6-tri-tert-butylphenyl)-1-phosphabuta-1,2,3- triene 26 can be prepared by either the method reported by Märkl72 (Peterson reaction) or that by the phospha-DMS reaction73 from 20, as shown in equation (13) (R = 2,4,6-But 3C6H2). On the other hand, 1,4-bis(2,4,6-tri-tert-butylphenyl)-1,4- diphosphabuta-1,2,3-triene 27 was similarly prepared; coupling reactions mediated with copper halides are useful to obtain 1,4- diphosphabutatrienes 27 [equation (14)].74–76 Furthermore, the phospha-DMS reaction, stating from 1,3-diphosphaallene 24, also gave 27 via dichloromethylenediphosphirane indicating generation of a carbene intermediate.It is noteworthy that the corresponding bromo derivative Z-29 gave the 3,4-dihydro-1- phosphanaphthalene derivative 30 upon warming, indicating a carbene intermediate.77 2.5 Diphosphabutadienes As a conjugated system containing low co-ordinated phosphorus atom(s), 2,3-dichloro-1,4-bis(2,4,6-tri-tert-butylphenyl)- 1,4-diphosphabuta-1,3-diene 31 was obtained as a coupling product of the reaction of 2-chloro-2-(2,4,6-tri-tert-butyl- Scheme 2 P C O P C N P C N P C O R P C NR¢ RP C NPh RP C NBut RP C NR RN C R¢N=C=O (12) H2O H2O H2O RP(H)C(O)NHPh HC(O)NHBut + RP(O)H2 + RP(O)H2 + RP(O)H2 Li P R SiMe3 25 P R C C C Ph P R Ph C C Ph P R Ph C CCl2 SiMe3 H P R C C Li SiMe3 C Ph P R Ph C C BunLi 2) SiMe3Cl :CCl2 BunLi 1) Ph2CO (13) 26 20 PR P R C C Cl P R Cl C P R C P Li Br P P R R C C R C P Br Br R C P R Cl C Cl :CCl2 Na[C10H8] BunLi (14) cat.CuX 27 RP(Li)H + C: P R C P R P But But Me Me C: P R heat 24 30 Z-29J. Chem.Soc., Dalton Trans., 1998, 3343–3349 3347 Table 1 The 31P NMR chemical shifts, dP, and bond lengths d(P]] X) of some organophosphorus compounds in low co-ordination states Compound E-2 Z-2 1 E-12 Z-12 20 24 23 25 26 E,E-15 E,Z-15 E-27 Z-27 31 33 3 34 R9 ———————— Ph — SiMe3 SiMe3 —————— System P]] P P]] P P]] C P]] C P]] C P]] C]] C P]] C]] P P]] C]] O P]] C]] N P]] C]] C]] C P]] C]C]] P P]] C]C]] P P]] C]] C]] P P]] C]] C]] P P]] C]C]] P P]] C]C]] P P]] ] C P]] ] C dP 492.4 368 233.0 259.3 241.6 72.1 142.7 2207.4 2106.2 156.7 162.5 197.4, 176.6 180.6 170.0 248.0 190.7 269.2 34.4 d(P]] X)/pm 203.4(2) 204.1(4) a 169.2(3) 166.0(6) 167.4(2) 162.5(4) 163.5(8), 163.0(8) — 165.1(3) 164.7(8) 167.8(6), 167.6(5) 166.5(7), 166.2(7) 165.6(6) b — 169.1(4) 168.3(5), 166.9(5) 152.0(12) c 151.6(13) Ref. 8 25, 26 7, 38 40, 41 40, 41 59, 60 63, 67 64 69 73 50, 51 50 74, 75, 76 75, 76 78 83 9 88, 89 a X-Ray data taken from ref. 30 for [{W(CO)5}2(Z-10)].b X-Ray data taken from ref. 74 for [{W(CO)5}2(E-27)]. c X-Ray data taken from ref. 86 for trans-[Mo(R9C]] ] P)2(Et2PCH2CH2PEt2)2] (R9 = adamantyl). phenyl)phosphaethenyllithium Z-28 in the presence of copper chloride and the yields are higher in the presence than in the absence of oxygen.78 In the case of the bromo derivative, a similar reaction of Z-29 occurs to give the corresponding dibromodiphosphabutadiene. However, when Z-28 was allowed to react with electrophiles, such as alkyl halides, aldehydes, ketones, isocyanates and acid chlorides, nucleophilic reactions took place to give multifunctionalized methylenephosphines in good yield [equation (15)].79,80 However, in both cases, diphosphabutatriene 27 was obtained as a by-product due to dehalogenation of the dihalogenobutadienes.Under similar reaction conditions except for the absence of copper halide, or oxygen bubbling, the formation and crystal structure of the dichlorodiphosphacyclobutanediyl 32 was reported by Niecke et al.81 On the other hand Appel et al.82 reported formation of 1,4- bis(2,4,6-tri-tert-butylphenyl)-1,4-diphosphabuta-1,3-diene by the elimination reaction of ethylenebis[chloro(2,4,6-tri-tertbutylphenyl) phosphine] with 1,5-diazabicyclo[4.3.0]non-5-ene. Furthermore, we have prepared a 4-radialene,83 the dimethylenediphosphinidenecyclobutane 33, as shown in equation (16).In C P P C Cl Cl R R R C P Li Cl R C P Cl Cl P P R R Cl Cl cat.CuX BunLi O2 (15) 31 32 Z-28 E R C P E Cl P CH2 Ph CH2Ph P R R P CHBrPh CHBrPh P R R P CHPh CHPh P R R 33 (16) contrast to the instability of the isomer E-28, as mentioned below, Z-28 gave the corresponding alkylation products by reaction with various electrophiles even at room temperature. 2.6 Phosphaalkynes A sterically protected phosphaalkyne 3 of co-ordination number 1 was reported by Becker et al.9 in 1981 for the first time, carrying tert-butyl as a protecting group.Phosphaalkynes are phosphorus analogues of nitriles and their chemistry has been developed not only as ligands of transition metal complexes, but also as a building block for further interesting phosphorus-containing heterocyclic compounds.84–87 The 2,4,6- tri-tert-butylphenyl derivative 34 has been prepared by the two diVerent methods, by the elimination of a siloxane from phosphaalkenes 88,89 and by the 1,2-aryl migration from phosphorus to carbon, which can be considered as a phosphorus version of the Fritsch–Buttenberg–Wiechell reaction (FBW reaction) of E-28, as shown in equation (17).80,90 In contrast, the isomer Z-28 did not form phosphaalkyne, indicating that some stereoelectronic eVect might be operating during this FBW reaction, as has been postulated by Köbrich and Trapp91 for the original FBW reactions. 3 Conclusion Studies on low co-ordinate organophosphorus compounds have extensively been developed since the first findings were disclosed that those kinds of compounds can be kinetically stabilized eYciently by bulky substituents, such as the 2,4,6-tritert- butylphenyl group. As a result, they have stimulated many chemists in the field of organic, inorganic as well as theoretical chemistry.As has been shown for almost two decades, various kinds of unusual compounds have been isolated and characterized involving not only phosphorus but also Group 14 and 16 elements, utilizing bulky substituents as well as special substituents with electronic eVects.92 Table 1 lists 31P chemical shifts for some selected low co-ordinated phosphorus compounds, R Cl C C H R Cl P C Li R Cl P Me3SiP R OSiMe3 RC P O C C E R Cl P P(SiMe3)3 heat RLi (17) heat E-28 E 34 at low temperature3348 J.Chem. Soc., Dalton Trans., 1998, 3343–3349 together with the observed lengths for the P]] X bonds d(P]] X) determined by X-ray analyses. The chemical shifts change sensitively depending on the substituents, as well as on the bonding systems,93 although the bond lengths are almost constant: ca. 200 pm for the P]] P bonds and 160–170 pm for the P]] C bonds. The reactivities of highly protected low co-ordinated organophosphorus compounds are to some extent similar to those of the corresponding C]] C systems rather than the N]] N or N]] C systems. Since such organophosphorus compounds involve phosphorus atom(s) in low co-ordination states with lone-pair electrons, the unsaturated systems are expected potentially to include novel and unusual characters, which are hitherto unknown and far beyond our common concept of either unsaturated carbon compounds or organophosphorus compounds.Thus a breakthrough is awaited as to the application of those compounds in the direction of materials or drugs. 4 Acknowledgements The support by the Ministry of Education, Science, Sports and Culture, Japanese Government, is greatly acknowledged.The author especially thanks Professor Ken Hirotsu at Osaka City University, who made the X-ray analyses of most of my earlier compounds described. I also thank those whose names are given in the list of references for their dedicated collaboration. 5 References 1 H. Köhler and A. Michaelis, Chem. Ber., 1877, 10, 807. 2 W. Kuchen and H. Buchwald, Chem. Ber., 1958, 91, 2296. 3 J. J. Daly and L. Maier, Nature (London), 1965, 208, 383; J. J. Daly, J. Chem. Soc., 1965, 4789; J.Chem. Soc. A, 1966, 428. 4 H. A. Staab, Einfürung in die theoretische organische Chemie, 4th edn., Verlag Chemie, Weinheim, 1964, p. 76. 5 K. Dimroth and P. HoVmann, Angew. Chem., 1964, 76, 433. 6 G. Märkl, Angew. Chem., 1966, 78, 907. 7 Th. C. Klebach, R. Lourens and F. Bickelhaupt, J. Am. Chem. Soc., 1978, 100, 4886. 8 M. Yoshifuji, I. Shima, N. Inamoto, K. Hirotsu and T. Higuchi, J. Am. Chem. Soc., 1981, 103, 4587; 1982, 104, 6167. 9 G. Becker, G. Gresser and W. Uhl, Z. Naturforsch., Teil B, 1981, 36, 16. 10 R. West, M. J. Fink and J. Michl, Science, 1981, 214, 1343. 11 M. Regitz and O. J. Scherer (Editors), Multiple Bonds and Low Coordination in Phosphorus Chemistry, Georg Thieme Verlag, Stuttgart, 1990. 12 A. H. Cowley, Acc. Chem. Res., 1984, 17, 386; Polyhedron, 1984, 3, 389. 13 J. I. G. Cadogan and P. K. G. Hodgson, Phosphorus Sulfur Relat. Elem., 1987, 30, 3. 14 L. N. Markovski, V. D. Romanenko and A. V. Ruban, Chemistry of Acyclic Compounds of Two-coordinated Phosphorus, Naukova Dumka, Kiev, 1988. 15 L. Weber, Chem. Rev., 1992, 92, 1839. 16 O. J. Scherer, Angew. Chem., Int. Ed. Engl., 1985, 24, 924. 17 G. Huttner, Pure Appl. Chem., 1986, 58, 585. 18 A.-M. Caminade, J.-P. Majoral and R. Mathieu, Chem. Rev., 1991, 91, 575. 19 M. Yoshifuji, Bull. Chem. Soc. Jpn., 1997, 70, 2881. 20 R. Appel and F. Knoll, Adv. Inorg. Chem., 1989, 33, 259. 21 M. Regitz and P. Binger, Angew. Chem., Int. Ed. Engl., 1988, 27, 1484. 22 M. Yoshifuji, K.Shibayama, N. Inamoto, T. Matsushita and K. Nishimoto, J. Am. Chem. Soc., 1983, 105, 2495. 23 M. Yoshifuji, S. Sasaki and N. Inamoto, J. Chem. Soc., Chem. Commun., 1989, 1732. 24 K. Tsuji, Y. Fujii, S. Sasaki and M. Yoshifuji, Chem. Lett., 1997, 855. 25 M. Yoshifuji, T. Sato and N. Inamoto, Chem. Lett., 1988, 1735. 26 A.-M. Caminade, M. Verrier, C. Ades, N. Paillous and M. Koenig, J. Chem. Soc., Chem. Commun., 1984, 875. 27 C. M. D. Komen, F. J. J. de Kanter, S. J. Goede and F.Bickelhaupt, J. Chem. Soc., Perkin Trans. 2, 1993, 807. 28 E. Niecke, B. Kramer and M. Nieger, Angew. Chem., Int. Ed. Engl., 1989, 28, 215. 29 E. Niecke, O. Altmeyer and M. Nieger, Angew. Chem., Int. Ed. Engl., 1991, 30, 1136. 30 M. Yoshifuji, N. Shinohara and K. Toyota, Tetrahedron Lett., 1996, 37, 7815. 31 M. Yoshifuji, T. Sato and N. Inamoto, Bull. Chem. Soc. Jpn., 1989, 62, 2394. 32 M. Yoshifuji, M. Abe, K. Toyota, K. Goto and N. Inamoto, Sci. Rep. Tohoku Univ., Ser. 1, 1991, 74, 8; Bull. Chem. Soc. Jpn., 1993, 66, 1572. 33 P. Jutzi and U. Meyer, J. Organomet. Chem., 1987, 333, C18. 34 X. Li, S. I. Weissman, T.-S. Lin, P. P. Gaspar, A. H. Cowley and A. I. Smirnov, J. Am. Chem. Soc., 1994, 116, 7899. 35 K. Tsuji, S. Sasaki and M. Yoshifuji, Heteroatom Chem., in the press. 36 M. Yoshifuji, T. Hashida, N. Inamoto, K. Hirotsu, T. Horiuchi, T. Higuchi, K. Ito and S. Nagase, Angew. Chem., Int. Ed. Engl., 1985, 24, 211. 37 A. H. Cowley, J. G. Lasch, N.C. Norman and M. Pakulski, J. Am. Chem. Soc., 1983, 105, 5506. 38 Th. A. van der Knaap, Th. C. Klebach, F. Visser, F. Bickelhaupt, P. Ros, E. J. Baerends, C. H. Stam and M. Konijn, Tetrahedron Lett., 1984, 40, 765. 39 K. Toyota and M. Yoshifuji, Rev. Heteroatom Chem., 1991, 5, 152. 40 M. Yoshifuji, K. Toyota, I. Matsuda, T. Niitsu, N. Inamoto, K. Hirotsu and T. Higuchi, Tetrahedron, 1988, 44, 1363. 41 M. Yoshifuji, K. Toyota and N. Inamoto, Tetrahedron Lett., 1985, 26, 1727. 42 M.Yoshifuji, K. Toyota, N. Inamoto, K. Hirotsu and T. Higuchi, Tetrahedron Lett., 1985, 26, 6443. 43 M. Yasunami, T. Ueno, M. Yoshifuji, A. Okamoto and K. Hirotsu, Chem. Lett., 1992, 1971. 44 A. Jouaiti, M. GeoVroy, G. Terron and G. Bernardinelli, J. Chem. Soc., Chem. Commun., 1992, 155. 45 A. Jouaiti, M. GeoVroy, G. Terron and G. Bernardinelli, J. Am. Chem. Soc., 1995, 117, 2251. 46 A. Jouaiti, M. GeoVroy and G. Bernardinelli, Chem. Commun., 1996, 437. 47 H. Kawanami, K.Toyota and M. Yoshifuji, Chem. Lett., 1996, 533. 48 R. Appel, V. Winkhaus and F. Knoch, Chem. Ber., 1987, 120, 243. 49 G. Märkl, P. Kreitmeier, H. Nöth and K. Polborn, Angew. Chem., Int. Ed. Engl., 1990, 29, 927. 50 M. Yoshifuji, K. Toyota, M. Murayama, H. Yoshimura, A. Okamoto, K. Hirotsu and S. Nagase, Chem. Lett., 1990, 2195. 51 K. Toyota, K. Tashiro, M. Yoshifuji, I. Miyahara, A. Hayashi and K. Hirotsu, J. Organomet. Chem., 1992, 431, C35. 52 K. Toyota, K. Tashiro, T.Abe and M. Yoshifuji, Heteroatom Chem., 1994, 5, 549. 53 M. Yoshifuji, Y. Ichikawa, K. Toyota, E. Kasajima and Y. Okamoto, Chem. Lett., 1997, 87. 54 M. Yoshifuji, Y. Ichikawa, N. Yamada and K. Toyota, Chem. Commun., 1998, 27. 55 D. Rau and U. Behrens, Angew. Chem., Int. Ed. Engl., 1991, 30, 870; J. Organomet. Chem., 1993, 454, 151. 56 C. N. Smit, F. M. Lock and F. Bickelhaupt, Tetrahedron Lett., 1984, 25, 3011. 57 M. W. Schmidt, Phi N. Truong and M. S. Gordon, J. Am. Chem. Soc., 1987, 109, 5217. 58 M. Yoshifuji, S. Sasaki and N. Inamoto, Tetrahedron Lett., 1989, 30, 839. 59 M. Yoshifuji, K. Toyota, K. Shibayama and N. Inamoto, Tetrahedron Lett., 1984, 25, 1809. 60 M. Yoshifuji, K. Toyota, N. Inamoto, K. Hirotsu, T. Higuchi and S. Nagase, Phosphorus Sulfur Relat. Elem., 1985, 25, 237. 61 M. Yoshifuji, H. Yoshimura and K. Toyota, Chem. Lett., 1990, 827. 62 M. Yoshifuji, K. Toyota, Y. Okamoto and T. Asakura, Tetrahedron Lett., 1990, 31, 2311. 63 M. Yoshifuji, K.Toyota and N. Inamoto, J. Chem. Soc., Chem. Commun., 1984, 689. 64 R. Appel, P. Fölling, B. Josten, M. Siray, V. Winkhaus and F. Knoll, Tetrahedron Lett., 1983, 24, 2639. 65 H. H. Karsch, F. H. Köhler and H.-U. Reisacher, Tetrahedron Lett., 1984, 25, 3687. 66 M. Yoshifuji, S. Sasaki, T. Niitsu and N. Inamoto, Tetrahedron Lett., 1989, 30, 187. 67 H. H. Karsch, H.-U. Reisacher and G. Müller, Angew. Chem., Int. Ed. Engl., 1984, 23, 618. 68 M. Yoshifuji, K. Toyota, T. Niitsu, N.Inamoto, Y. Okamoto and R. Aburatani, J. Chem. Soc., Chem. Commun., 1986, 1550. 69 M. Yoshifuji, T. Niitsu, K. Toyota, N. Inamoto, K. Hirotsu, Y. Odagaki, T. Higuchi and S. Nagase, Polyhedron, 1988, 7, 2213. 70 O. I. Kolodiazhnyi, Tetrahedron Lett., 1982, 23, 4933. 71 T. Niitsu, N. Inamoto, K. Toyota and M. Yoshifuji, Bull. Chem. Soc. Jpn., 1990, 63, 2736.J. Chem. Soc., Dalton Trans., 1998, 3343–3349 3349 72 G. Märkl, H. Sejpka, S. Dietl, B. Nuber and M. L. Ziegler, Angew.Chem., Int. Ed. Engl., 1986, 25, 1003. 73 M. Yoshifuji, K. Toyota, H. Yoshimura, K. Hirotsu and A. Okamoto, J. Chem. Soc., Chem. Commun., 1991, 124. 74 S. Ito, K. Toyota and M. Yoshifuji, J. Organomet. Chem., 1998, 553, 135. 75 G. Märkl and P. Kreitmeier, Angew. Chem., Int. Ed. Engl., 1988, 27, 1360. 76 M. Yoshifuji, K. Toyota and H. Yoshimura, Chem. Lett., 1991, 491. 77 S. Ito, K. Toyota and M. Yoshifuji, Chem. Commun., 1997, 1637. 78 S. Ito, K. Toyota and M. Yoshifuji, Chem. Lett., 1995, 747. 79 M. Yoshifuji, S. Ito, K. Toyota and M. Yasunami, Bull. Chem. Soc. Jpn., 1995, 68, 1206. 80 M. Yoshifuji, T. Niitsu and N. Inamoto, Chem. Lett., 1988, 1733. 81 E. Niecke, A. Fuchs, F. Baumeister, M. Nieger and W. W. Schoeller, Angew. Chem., Int. Ed. Engl., 1995, 34, 555. 82 R. Appel, J. Hünerbein and N. Siabalis, Angew. Chem., Int. Ed. Engl., 1987, 26, 779. 83 K. Toyota, K. Tashiro and M. Yoshifuji, Angew. Chem., Int. Ed. Engl., 1993, 32, 1163. 84 M. Regitz, Chem. Rev., 1990, 90, 191. 85 J. C. T. R. Burckett-St. Laurent, P. B. Hitchcock, H. W. Kroto and J. F. Nixon, J. Chem. Soc., Chem. Commun., 1981, 1141. 86 P. B. Hitchcock, M. J. Maah, J. F. Nixon, J. A. Zora, G. J. Leigh and M. A. Bakar, Angew. Chem., Int. Ed. Engl., 1987, 26, 474. 87 P. Binger, F. Sandmeyer, C. Krüger, J. Kuhnigk, R. Goddard and G. Erker, Angew. Chem., Int. Ed. Engl., 1994, 33, 197. 88 G. Märkl and H. Sejpka, Tetrahedron Lett., 1986, 27, 171; Angew. Chem., Int. Ed. Engl., 1986, 25, 264. 89 A. M. Arif, A. R. Barron, A. H. Cowley and S. W. Hall, J. Chem. Soc., Chem. Commun., 1988, 171. 90 M. Yoshifuji, Y. Kawai and M. Yasunami, Tetrahedron Lett., 1990, 31, 6891. 91 G. Köbrich and H. Trapp, Chem. Ber., 1966, 99, 680. 92 M. Yoshifuji, Main Group Chem. News, 1998, 6, 20. 93 L. D. Quin and J. Verkade (Editors), Phosphorus-31 NMR Spectral Properties in Compound Characterization and Structural Analysis, VCH, New York, 1994. Paper 8/03041C

ISSN:1477-9226    

DOI:10.1039/a803041c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1998, 3351–3352 3351 Complexed bridging ligand, [M(bpca)2] (M 5 Mn(II) or Fe(II); Hbpca 5 bis(2-pyridylcarbonyl)amine), as a building block for linear trinuclear complexes Takashi Kajiwara and Tasuku Ito Department of Chemistry, Graduate School of Science, Tohoku University, Aoba, Aramaki, Aoba-ku, Sendai 980-8578, Japan. E-mail: ito@agnus.chem.tohoku.ac.jp Received 24th August 1998, Accepted 25th August 1998 The antiferromagnetically coupled trimanganese complex [MnII(bpca)2{MnII(hfac)2}2] (Hbpca 5 bis(2-pyridylcarbonyl) amine; Hhfac 5 hexafluoroacetylacetone) and its iron–manganese mixed metal derivative, [FeII(bpca)2- {MnII(hfac)2}2], were synthesized by the reaction of two equivalents of [Mn(hfac)2(H2O)2] with one equivalent of [M(bpca)2], the latter acting as a bridging complexedligand.The chemistry of multi-metal-centered complexes or metal complex assemblies with highly ordered solid state structures has attracted much attention.1 In such chemistry, “complexedligands” are known to be beneficial in the construction of multi-metal complexes and the control of their properties.Monomeric complexes containing the tridentate ligand bpca2 (Hbpca = bis(2-pyridylcarbonyl)amine), [M(bpca)2] (M = Mn(II),2 Fe(II),3 Ni(II),4 Cu(II),4 Zn(II) 4) and [M(bpca)2]X (M = Fe(III); X2 = NO3 2, ClO4 2),3 could be examples of this type of complexed ligand, although no example has been reported to our knowledge.They have four free C]] O groups which may act as two sets of bidentate donors, and, upon reaction with a metal ion M9, they may give a trinuclear complex of the type M9(m-bpca)M(m-bpca)M9. It has been reported that oximato and oxamido complexes can act as a complexed ligand to give di-,5 tri-,6 and tetra-nuclear complexes.7 One of the characteristics of the present system is that {M(bpca)2} has a delocalized p-system which might mediate M–M9 interactions in redox and magnetic behaviour in a diVerent way from oxamido complexes.Here we report two examples of trinuclear complexes containing bridging {M(bpca)2}. [Mn(bpca)2]?H2O2 or [Fe(bpca)2]?H2O3 was allowed to react with two equivalents of [Mn(hfac)2(H2O)2] in CHCl3 solution (Hhfac = hexafluoroacetylacetone). Slow evaporation aVorded almost quantitatively orange and black crystals of trinuclear complexes, [Mn(bpca)2{Mn(hfac)2}2] and [Fe(bpca)2{Mn- (hfac)2}2], respectively.† Fig. 1 shows the structure of [Mn(bpca)2{Mn(hfac)2}2].‡ As expected, a {Mn(bpca)2} unit binds two {Mn(hfac)2} units as a bridging bis-bidentate complexed-ligand. The central Mn ion, Mn1, is surrounded by four pyridyl nitrogens (N1, N3, N4 and N6) with Mn–N distances of 2.221(2)–2.262(2) Å and two amide nitrogens (N2 and N5) with distances of 2.196(2) and 2.203(2) Å. The latter are slightly longer than those in the parent monomeric [Mn(bpca)2] (2.179(7) and 2.169(7) Å).2 The C]] O distances (average 1.233 Å) are definitely longer than those in the monomer, which results in the low frequency shift of C]] O stretching (1670 cm21: an intense band at 1700 cm21 for the parent monomer2).These facts suggest that the minus charge of bpca2 is delocalized on the O–C–N–C–O moiety in the trinuclear complex, whereas it is located mainly on the amide nitrogen in [Mn(bpca)2]. Terminal Mn ions, Mn2 and Mn3, are coordinated by six oxygen atoms from two hfac anions and from a {Mn(bpca)2} unit with Mn–O distances of 2.113(3)– 2.190(2) Å.These two terminal Mn ions are in a chiral environment with the combination D, L. The three Mn ions are arranged in an almost linear fashion with separations of 5.6708(6) Å for Mn1 ? ? ? Mn2 and 5.6855(6) Å for Mn1 ? ? ? Mn3, respectively. The structure of [Fe(bpca)2{Mn(hfac)2}2] was isostructural to [Mn(bpca)2{Mn(hfac)2}2].§ Overall structural features including delocalization of the O–C–N–C–O moiety are very similar to each other.Fig. 1 An ORTEP8 drawing of [Mn(bpca)2{Mn(hfac)2}2] with thermal ellipsoids at 30% probability. Hydrogen atoms and fluorine atoms are omitted for clarity. Selected bond distances (Å): Mn(1)–N(1) 2.246(2), Mn(1)–N(2) 2.196(2), Mn(1)–N(3) 2.221(2), Mn(1)–N(4) 2.262(2), Mn(1)–N(5) 2.203(2), Mn(1)–N(6) 2.242(2), Mn(2)–O(1) 2.153(2), Mn(2)–O(2) 2.176(2), Mn(2)–O(5) 2.146(3), Mn(2)–O(6) 2.163(3), Mn(2)–O(7) 2.126(3), Mn(2)–O(8) 2.113(3), Mn(3)–O(3) 2.171(2), Mn(3)–O(4) 2.190(2), Mn(3)–O(9) 2.152(2), Mn(3)–O(10) 2.147(2), Mn(3)–O(11) 2.120(2), Mn(3)–O(12) 2.122(2), O(1)–C(6) 1.221(3), O(2)–C(7) 1.220(3), O(3)–C(19) 1.251(3), O(4)–C(18) 1.241(3), Mn(1) ? ? ? Mn(2) 5.6708(6), Mn(1) ? ? ? Mn(3) 5.6855(6).3352 J.Chem. Soc., Dalton Trans., 1998, 3351–3352 The temperature dependence of the magnetic susceptibility of the compounds was measured down to 2.0 K. Fig. 2 shows the magnetic susceptibility data for [Mn(bpca)2{Mn(hfac)2}2] in the form of cmT and cm vs.T plots. The cmT value at room temperature, 12.9 cm3 K mol21, is slightly smaller than the spin-only value of 13.1 cm3 K mol21 for the dilute three-spin system with a g value of 2.00. On lowering the temperature, the cmT value gradually decreases suggesting antiferromagnetic interaction between adjoining Mn(II) ions through the delocalized p-system. Magnetic data of the trimer was analyzed by the three-spin model with exchange coupling constant J [H = 22J(SMn1?SMn2 1 SMn1?SMn3)].9 The least squares calculation yielded the best fit parameters of g = 1.98(1) and J = 20.35(1) cm21.[Fe(bpca)2{Mn(hfac)2}2] containing low-spin Fe(II) gave a temperature independent cmT value of 8.52 cm3 K mol21 above 10.0 K. The weak magnetic interaction in [Mn(bpca)2{Mn- (hfac)2}2] may be related to the Mn–N distances of 2.196(2)– 2.262(2) Å, which are longer than the M–N separations of divalent late first row transition metal ions. In fact, [Mn- (bpca)2{Mn(hfac)2}2] shows no distinct MLCT (from Mn to bpca2) band in the absorption spectrum,¶ suggesting weak dp– pp interactions.This study shows that [M(bpca)2]n1 could be a potential building block for supramolecular compounds. In fact, similar trinuclear complexes [M(bpca)2{Mn(hfac)2}2] (M = Ni(II), Cu(II)) have been isolated via similar reactions, in which {M(bpca)2} is acting as a building block.10 Such studies are now in progress in our laboratories. Fig. 2 Plots of cmT (o) and cm (x) vs.T for [Mn(bpca)2{Mn(hfac)2}2]. Solid line corresponds to the theoretical curve for which parameters are given in the text. Acknowledgements This work was supported by Grant-in-Aid for Scientific Research on Priority Areas (No. 10149102) from the Ministry of Education, Science and Culture, Japan. Notes and references † Elemental analysis: 1, found: C, 36.37; 1.62; N, 5.85. Calc. for C44H18N6O12Mn3F24: C, 36.61; H, 1.26; N, 5.82%. 2, found: C, 36.59; H, 1.53; N, 5.77.Calc. for C44H18N6O12FeMn2F24: C, 36.59; H, 1.26; N, 5.82%. ‡ Crystal data: C44H18N6O12Mn3F24, M = 1443.44, orthorhombic, space group Pna21 (no. 33), a = 17.142(9), b = 20.552(3), c = 16.233(4) Å, U = 5718(3) Å3, Z = 4, Dc = 1.676 g cm23, F(000) = 2940.00, m(Mo- Ka) = 12.30 cm21, 3953 unique reflections (I > 2.0s(I)) collected at room temperature with Mo-Ka radiation (l = 0.71069 Å) up to 2q = 55.08 on a Rigaku AFC 7S diVractometer. Final R value is 0.058 for observed data.CCDC reference number 186/1132. See http:// www.rsc.org/suppdata/dt/1998/3351/ for crystallographic files in .cif format. § Crystal data: C44H18N6O12FeMn2F24, M = 1444.35, orthorhombic, space group Pna21 (no. 33), a = 16.701(6), b = 20.134(7), c = 16.353(5) Å, U = 5498(2) Å3, Z = 4, R = 0.052. ¶ Electronic spectrum in CHCl3 solution: lmax = 290 nm (e = 60500 dm3 mol21 cm21) and ca. 420 nm (shoulder, ca. 600 dm3 mol21 cm21). 1 For example, H. O. Stumpf, L.Ouahab, Y. Pei, D. Grandjean and O. Kahn, Science, 1993, 261, 447; S. L. Suib, Chem. Rev., 1993, 93, 803. 2 D. Marcos, J.-V. Folgado and D. Beltrán-Porter, Polyhedron, 1990, 9, 2699. 3 S. Wocadlo, W. Massa and J.-V. Folgado, Inorg. Chim. Acta, 1993, 207, 199. 4 D. Marcos, R. Martinez-Mañe, J.-V. Folgado, A. Beltrán-Porter, D. Beltrán-Porter and A. Fuertes, Inorg. Chim. Acta, 1989, 159, 11. 5 F. Birkelbach, M. Winter, U. Flörke, H.-J. Haupt, C. ButzlaV, M. Lengen, E. Bill, A. X. Trautwein, K. Weighardt and P. Chaudhuri, Inorg. Chem., 1994, 33, 3990; P. Basu, S. Pal and A. Chakravorty, J. Chem. Soc., Dalton Trans., 1991, 3217; A. Escuer, R. Vicente, J. Ribas, R. Costa and X. Solans, Inorg. Chem., 1992, 31, 2627. 6 For example, S. Chattopadhyay, P. Basu, S. Pal and A. Chakravorty, J. Chem. Soc., Dalton Trans., 1990, 3829; R. Costa, A. Garcia, R. Sanchez, J. Ribas, X. Solans and V. Rodriguez, Polyhedron, 1993, 12, 2697. 7 C. Krebs, M. Winter, T. Weyhermüller, E. Bill, K. Wieghardt and P. Chaudhuri, J. Chem. Soc., Chem. Commun., 1995, 1913; F. Corazza, E. Solari, C. Floriani, A. Chiesi-Villa and C. Guastini, J. Chem. Soc., Chem. Commun., 1986, 1562. 8 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1976. 9 O. Kahn, Molecular Magnetism, VCH Publishers, Weinheim, 1993; S. Ménage, S. E. Vitols, P. Bergerat, E. Codjovi, O. Kahn, J.-J. Girerd, M. Guillot, X. Solans and T. Calvet, Inorg. Chem., 1991, 30, 2666. 10 T. Noguchi, T. Kajiwara and T. Ito, unpublished work. Communication 8/06618C

ISSN:1477-9226    

DOI:10.1039/a806618c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1998, 3353–3354 3353 An unprecedented trinucleating bridging mode (Ï3-Á1 :Á1 :Á1) of a tris(pyrazolyl)borate ligand in a trinuclear silver(I) complex Elizabeth R. Humphrey, Nicholas C. Harden, Leigh H. Rees, John C. JeVery, Jon A. McCleverty * and Michael D. Ward * School of Chemistry, University of Bristol, Cantock’s Close, Bristol, UK BS8 1TS. E-mail: mike.ward@bristol.ac.uk Received 11th August 1998, Accepted 28th August 1998 The crystal structure of [Ag3(TpAn)2][ClO4], where TpAn is tris[3-(2-methoxyphenyl)pyrazol-1-yl]hydroborate, shows a unique trinucleating bridging coordination mode for the tris(pyrazolyl)borate: each of the two ligands coordinates one pyrazolyl donor to each of the three two-coordinate Ag(I) ions, affording a triangular complex in which each face is capped by a Ï3-Á1 :Á1 :Á1-tris(pyrazolyl)borate.The tris(pyrazolyl)borates are a very well known class of ligand which almost invariably coordinate as terdentate chelates to a single metal ion, ‘capping’ one triangular face of the coordination polyhedron.1 This predictability in their mode of coordination is one of the reasons for their popularity as it has allowed the preparation of complexes with coordination geometries which can be planned in advance.This has found use in diverse areas. In the realm of bioinorganic chemistry, three pyrazoles of a chelating tris(pyrazolyl)borate ligand are a reasonable structural and electronic mimic for the tris(imidazolyl) coordination which occurs at a variety of metalloprotein active sites.1,2 Tris(pyrazolyl)borate derivatives with bulky substituents at the pyrazolyl C3 positions form four-coordinate pseudo-tetrahedral complexes in which a protective screen around the fourth coordination site allows the stabilisation of low-coordinate complexes which would otherwise be inaccessible or highly reactive.1 The only well-characterised exceptions to this rule with simple tris(pyrazolyl)borates are (i) a few 16-electron squareplanar complexes in which a tris(pyrazolyl)borate is bidentate with one pyrazolyl arm pendant;3 and (ii) some dinuclear complexes [mostly of Cu(I)] in which a tris(pyrazolyl)borate acts as a bridging ligand, acting as a bidentate donor to one metal ion and a monodentate donor to the second.4 Also, we have recently described some complexes of the hexadentate podand tris[3-(2-pyridyl)pyrazol-1-yl]hydroborate (TpPy) in which the three bidentate chelating arms each coordinate to a diVerent metal ion, however the nature of this ligand makes it a special case as it behaves completely diVerently from ‘conventional’ tris(pyrazolyl)borates.5 We describe here the crystal structure of a trinuclear Ag(I) complex of the ligand tris[3-(2-methoxyphenyl)pyrazol-1- yl]hydroborate (TpAn, where the suYx ‘An’ denotes the anisyl substituent), [Ag3(TpAn)2][ClO4], in which the ligand adopts the previously unknown trinucleating bridging mode with each pyrazolyl donor coordinated to a separate Ag(I) ion.We prepared TpAn recently to see if it would act as an N3O3-donor hexadentate podand (analogous to TpPy), but found that the methoxy groups did not coordinate in any of the complexes that we structurally characterised.6 This ligand therefore behaves like conventional tris(pyrazolyl)borates bearing a bulky substituent at the pyrazolyl C3 positions.1 Reaction of K[TpAn] with AgClO4 in thf aVorded a precipitate whose FAB mass spectrum indicated the formation of the [Ag3(TpAn)2]1 cation.† Recrystallisation from acetone–diethyl ether aVorded X-ray quality crystals of [Ag3(TpAn)2][ClO4]?1.5OCMe2?0.5Et2O; the structure of the complex cation is shown in Figs. 1 and 2.‡ The complex contains a triangular array of Ag(I) ions, with Ag ? ? ? Ag separations of 3.927(2) Å [Ag(1) ? ? ? Ag(2)], 3.936(2) Å [Ag(2) ? ? ? Ag(3)] and 4.189 Å [Ag(1) ? ? ? Ag(3)], each face of the triangle being capped by a single TpAn ligand which donates Fig. 1 Crystal structure of the cation of [Ag3(TpAn)2][ClO4]? 1.5OCMe2?0.5Et2O: thermal ellipsoids are at the 40% probability level. Significant bond lengths (Å) and angles (8): Ag(1)–N(212) 2.131(3), Ag(1)–N(152) 2.134(3), Ag(2)–N(132) 2.106(3), Ag(2)–N(252) 2.104(3), Ag(3)–N(232) 2.101(3), Ag(3)–N(112) 2.104(3); N(212)– Ag(1)–N(152) 179.10(11), N(132)–Ag(2)–N(252) 176.39(11), N(232)– Ag(3)–N(112) 177.77(11).N N N N N N B H MeO OMe OMe N N N N N N B H N N N [–] [TpAn]– [–] [TpPy]–3354 J. Chem. Soc., Dalton Trans., 1998, 3353–3354 one pyrazolyl N atom to each Ag(I) ion. Each Ag(I) ion is therefore in an approximately linear two-coordinate environment, arising from one pyrazolyl donor from each of the two TpAn ligands; this is a well-known mode of coordination for Ag(I) ions.7 All of the Ag–N bond lengths lie within the range 2.10– 2.14 Å, and the methoxy groups do not coordinate [the only remotely significant Ag ? ? ? O contact is 2.687(3) Å between Ag(1) and O(167), which is just a weak cation–dipole interaction and much too long to constitute a coordinate bond].Each ligand has an ‘inverted’ conformation which allows the three pyrazolyl donors to coordinate to separate metal ions rather than converge on the same one, and in consequence the B–H bonds are directed inwards towards each other. The six Ag ? ? ? H separations are in the range 2.5–2.8 Å, too long to be considered as agostic interactions.The structure of this complex is in interesting contrast to all other structurally characterised Ag(I) complexes of tris- (pyrazolyl)borates, in which the tris(pyrazolyl)borate ligands are coordinated in the conventional terdentate manner.8 The reason for adoption of this new coordination mode is not obvious. The substituents impose no steric barrier on formation of a conventional 1:1 complex [a 1:1 complex has already been prepared and structurally characterised with Tl(I)];6 the methoxyphenyl substituents play no significant role in coordinating to the metal ions; there are no obvious co-operative interactions (such as aromatic p-stacking) between the two ligands; and the Ag ? ? ? Ag separations are too great for the structure to be stabilised by Ag ? ? ? Ag bonding interactions.9 It was suggested a while ago on the basis of mass spectroscopic and osmometric results that some binary Ag(I) complexes with simple poly(pyrazolyl)borate ligands were oligomeric via a bridging coordination mode of the ligand,10 but no structural evidence has been available to support this possibility until now.The structure of [Ag3(TpAn)2][ClO4] suggests that a more thorough investigation of the coordination behaviour of appropriatelysubstituted tris(pyrazolyl)borates with Cu(I) and Au(I) (which also support linear two-coordinate geometries),9,11 as well as Ag(I), would be fruitful.Acknowledgements We thank the EPSRC for financial support. Notes and references † A mixture of K(TpAn) (114 mg, 0.2 mmol) and AgClO4 (62 mg, 0.3 mmol) in dry thf (20 cm3) at 215 8C under N2 was stirred for 0.5 h to give a precipitate of crude [Ag3(TpAn)2][ClO4] (brown due to the pres- Fig. 2 Alternative view of the [Ag3(TpAn)2]1 core with the methoxyphenyl groups removed for clarity (only the ipso carbon atoms are shown). ence of colloidal silver). Repeated recrystallisation from acetone– diethyl ether in the dark aVorded a small number of X-ray quality colourless crystals. The yield of crystalline material was low (ca. 10%) due to the instability of the product in solution: silver(I) complexes of poly(pyrazolyl)borates are known to decompose easily and be lightand oxygen-sensitive.8,10 FAB-MS: m/z 1386 [100%, {Ag3(TpAn)2}1], 637 [40%, {Ag(TpAn)}1] (Found: C, 48.0; H, 3.4; N, 10.9. Required for [Ag3(TpAn)2][ClO4]: C, 48.5; H, 3.8; N, 11.3%). ‡ A suitable crystal of [Ag3(TpAn)2][ClO4]?1.5OCMe2?0.5Et2O (dimensions 0.4 × 0.4 × 0.2 mm) was quickly removed from the mother-liquor and mounted on a Siemens SMART diVractometer at 2100 8C.Crystal data: C66.5H70Ag3B2ClN12O12, M = 1610.0, monoclinic, space group P21/c, a = 15.264(4), b = 15.785(4), c = 29.433(11) Å, b = 103.85(2)8, U = 6885(4) Å3, Z = 4, rcalc = 1.553 g cm23, m(Mo-Ka) = 0.953 mm21. 42588 reflections were collected to 2qmax = 558, which after merging gave 15707 unique reflections (Rint = 0.0526).Refinement (SHELXTL)12 of 911 parameters on all F2 data converged at R1 = 0.0426 [data with F > 4s(F)], wR2 = 0.0986 (all data). Apart from the trinuclear cation and the perchlorate anion, there is one well-behaved acetone molecule in the asymmetric unit and disordered overlapping molecules of acetone (50% site occupancy) and ether (50% site occupancy). Largest residual peak, hole: 10.899, 20.728 e Å23. CCDC reference number 186/1138. See http://www.rsc.org/suppdata/dt/1998/3353/ for crystallographic data in .cif format. 1 S. Trofimenko, Chem. Rev., 1993, 93, 942. 2 N. Kitajima and Y. Moro-oka, J. Chem. Soc., Dalton Trans., 1993, 2665; N. Kitajima, K. Fujisawa, M. Tanaka and Y. Moro-oka, J. Am. Chem. Soc., 1992, 114, 9232; S. Hikichi, T. Ogihara, K. Fujisawa, N. Kitajima, M. Akita and Y. Moro-oka, Inorg. Chem., 1997, 36, 4539; A. Kremer-Aach, W. Klaui, R. Bell, A. Strerath, H. Wunderlich and D. Mootz, Inorg. Chem., 1997, 36, 1552; A.Looney, G. Parkin, R. Alsfasser, M. Ruf and H. Vahrenkamp, Angew. Chem., Int. Ed. Engl., 1992, 31, 92. 3 N. G. Connelly, D. J. H. Emslie, B. Metz, A. G. Orpen and M. J. Quayle, Chem. Commun., 1996, 2289; U. E. Bucher, A. Currao, R. Nesper, H. Ruegger, L. M. Venanzi and E. Younger, Inorg. Chem., 1995, 34, 66; C. P. Brock, M. K. Das, R. P. Minton and K. Niedenzu, J. Am. Chem. Soc., 1988, 110, 817. 4 C. Mealli, C. S. Arcus, J. L. Wilkinson, T. J. Marks and J. A.Ibers, J. Am. Chem. Soc., 1976, 98, 711; J. S. Thompson, R. L. Harlow and J. F. Whitney, J. Am. Chem. Soc., 1983, 105, 3522; J. S. Thompson and J. F. Whitney, J. Am. Chem. Soc., 1983, 105, 5488; J. S. Thompson and J. F. Whitney, Acta Crystallogr., Sect. C, 1984, 40, 756; D. L. Reger, J. C. Baxter and L. Lebioda, Inorg. Chim. Acta, 1989, 165, 201; S. M. Carrier, C. E. Ruggiero, R. P. Houser and W. B. Tolman, Inorg. Chem., 1993, 32, 4889; C.-T. Lee, J.-D. Chen, L.-S. Liou and J.-C.Wang, Inorg. Chim. Acta, 1996, 249, 115. 5 A. J. Amoroso, J. C. JeVery, P. L. Jones, J. A. McCleverty, P. Thornton and M. D. Ward, Angew. Chem., 1995, 107, 1577; A. J. Amoroso, J. C. JeVery, P. L. Jones, J. A. McCleverty, E. Psillakis and M. D. Ward, J. Chem. Soc., Chem. Commun., 1995, 1475; P. L. Jones, J. C. JeVery, J. P. Maher, J. A. McCleverty, P. H. Rieger and M. D. Ward, Inorg. Chem., 1997, 36, 3088. 6 P. L. Jones, K. L. V. Mann, J. C. JeVery, J. A. McCleverty and M.D. Ward, Polyhedron, 1997, 16, 2435. 7 H. Schmidbauer, A. Mair, G. Müller, J. Lachmann and S. Gamper, Z. Naturforsch., Teil B, 1991, 46, 912; F. Meyer, A. Jacobi and L. Zsolnai, Chem. Ber./Recl., 1997, 130, 1441; H. H. Murray, R. G. Raptis and J. P. Fackler, Jr., Inorg. Chem., 1988, 27, 26. 8 H. V. R. Dias and W. Jin, J. Am. Chem. Soc., 1995, 117, 11381; H. V. R. Dias and W. Jin, Inorg. Chem., 1996, 35, 267; H. V. R. Dias, W. Jin, H.-J. Kim and H.-L. Lu, Inorg. Chem., 1996, 35, 2317; H. V. R. Dias, Z.-Y. Wang and W.-C. Jin, Inorg. Chem., 1997, 36, 6205; C. Santini, G. G. Lobbia, C. Pettinari, M. Pellei, G. Valle and S. Calogero, Inorg. Chem., 1998, 37, 890. 9 K. Singh, J. R. Long and P. Stavropoulos, J. Am. Chem. Soc., 1997, 119, 2942. 10 M. I. Bruce and J. D. Walsh, Aust. J. Chem., 1979, 32, 2753; A. M. Abu Salah, G. S. Ashby, M. I. Bruce, E. A. Pederzolli and J. D. Walsh, Aust. J. Chem., 1979, 32, 1613. 11 M. K. Ehlert, S. J. Rettig, A. Storr, R. C. Thompson and J. Trotter, Can. J. Chem., 1990, 68, 1444; R. G. Raptis and J. P. Fackler, Jr., Inorg. Chem., 1988, 27, 4179. 12 SHELXTL 5.03 program system; Siemens Analytical X-Ray Instruments, Madison, WI, 1995. Communication 8/06333H

ISSN:1477-9226    

DOI:10.1039/a806333h

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J. Chem. Soc., Dalton Trans., 1998, 3355–3357 3355 DALTON COMMUNICATION Asymmetric bis(pyrazolyl)hydroborato ligands via direct synthesis: structural characterization of thallium and zinc complexes † Prasenjit Ghosh,a Tony Hascall,a Conor Dowling b and Gerard Parkin *a a Department of Chemistry, Columbia University, New York, NY 10027, USA b Elf Atochem, North America, 900 First Ave., PO Box 61536, King of Prussia, PA 19406, USA Received 10th August 1998, Accepted 1st September 1998 Asymmetrically substituted bis(pyrazolyl)hydroborato ligands, i.e.[H2B(pzRR9)(pzBut 2)]2, in which the two pyrazolyl groups possess different substituents, have been synthesized by the direct reaction of a ca. 1:1 molar mixture of the respective pyrazoles with LiBH4. While Trofimenko’s poly(pyrazolyl)hydroborato ligand system has emerged as one of the most popular in modern coordination chemistry, the majority of studies have concentrated on the use of tris(pyrazolyl)hydroborato derivatives.1 Indeed, tris(pyrazolyl)hydroborato ligands have been subject to a large variety of modifications, which include the incorporation of sterically demanding (e.g.tert-butyl), electron withdrawing (e.g. CF3), and optically active substituents.1 By comparison, bis(pyrazolyl)hydroborato ligands have received much less attention, although we have recently used such ligands as a framework for constructing tridentate [NNO] and [NNS] donor ligands.2 In this paper, we report the syntheses of asymmetric bis(pyrazolyl)hydroborato ligands in which the two pyrazolyl groups incorporate diVerent substituents, i.e.[H2B(pzRR9)- (pzBut 2)]2. Since the first report in 1982, asymmetrically substituted bis(pyrazolyl)hydroborato ligands have been restricted to derivatives in which one of the pyrazolyl groups is methyl substituted in both the 3- and 5-positions, i.e. [H2B(pzMe2)- (pzR9R0)]2.3 The principal reason for this restriction is due to the fact that the synthetic methods employed to incorporate two diVerent pyrazolyl groups on boron have to date required the use of a pyrazole–borane reagent, (HpzRR9)(BH3), of which only the 3,5-dimethylpyrazole adduct could be synthesized (Scheme 1).4 It is, therefore, significant that we have discovered that a variety of asymmetrically substituted bis(pyrazolyl)- hydroborato ligands may be constructed straightforwardly, by the direct reaction of LiBH4 with a ca. 1 : 1 molar mixture of two diVerent pyrazoles, as illustrated in Scheme 2.For example, a mixture of pyrazole and 3,5-di-tert-butylpyrazole reacts with LiBH4 to give [H2B(pz)(pzBut 2)]Li.5 Metathesis of the latter complex with TlOAc yields its thallium derivative [H2B(pz)- (pzBut 2)]Tl, which has been structurally characterized by X-ray diVraction (Fig. 1).6 Subsequent treatment of [H2B(pz)- (pzBut 2)]Tl with ZnI2 gives [H2B(pz)(pzBut 2)]ZnI, which has been structurally characterized as the 3-tert-butyl-5-isopropylpyrazole adduct, [H2B(pz)(pzBut 2)]ZnI(HpzBut,Pri) (Fig. 2). In addition to [H2B(pz)(pzBut 2)]2, the synthetic method is also Fig. 1 Molecular structure of [H2B(pz)(pzBut 2)]Tl (only one of the crystallographically independent molecules is shown). Selected bond lengths (Å) and angles (8): Tl(1)–N(12) 2.634(4), Tl(1)–N(22) 2.677(3), N(12)–Tl(1)–N(22) 72.87(10), Tl(2)–N(112) 2.658(4), Tl(2)–N(122) 2.680(3), N(112)–Tl(2)–N(122) 75.20(11). Scheme 13356 J. Chem. Soc., Dalton Trans., 1998, 3355–3357 Table 1 Comparison of M–N bond lengths for asymmetric bis(pyrazolyl)hydroborato complexes {[H2B(pz1)(pz2)]M} [H2B(pz)(pzBut 2)]Tl [H2B(pzTrip)(pzBut 2)]ZnI [H2B(pz)(pzBut 2)]ZnI(HpzBut,Pri) c [H2B(pzMe2)(pzBut 2)]ZnI(HpzBut 2) d [H2B(pz)(pzMe2)]2Ni [H2B(pzMe2)(pzPh2)]2Zn d(M–Npz1)/Å a 2.634(4), 2.658(4) b 1.994(2) 2.017(2) 1.978(4) 1.883(3) 1.982(2), 1.982(2) d(M–Npz2)/Å a 2.677(3), 2.680(3) b 1.994(2) 2.011(2) 2.017(3) 1.894(2) 2.019(2), 2.010(2) |Dd(M–N)|/Å 0.032 0.000 0.006 0.039 0.011 0.033 Ref.This work This work This work This work 78 a pz1 and pz2 are the first and second pyrazolyl groups, respectively, in the compound formula. b Values for two crystallographically independent molecules. c d(Zn–pzBut,PriH) = 2.038(2) Å. e d(Zn–pzBut 2H) = 2.090(3) Å. applicable for [H2B(pzMe2)(pzBut 2)]2 and [H2B(pzTrip)(pzBut 2)]2 (Trip = triptycyl, 9,10-o-benzeno-9,10-dihydroanthracenyl) ligands, which have been structurally characterized as the zinc iodide derivatives, [H2B(pzMe2)(pzBut 2)]ZnI(HpzBut 2) (Fig. 3) and [H2B(pzTrip)(pzBut 2)]ZnI (Fig. 4). Other than the complexes described above, structurally authenticated asymmetric bis(pyrazolyl)hydroborato complexes are limited to the nickel and zinc complexes, [H2B(pz)- (pzMe2)]2Ni7 and [H2B(pzMe2)(pzPh2)]2Zn,8 respectively. For comparison, the M–N bond lengths for all structurally characterized asymmetric bis(pyrazolyl)hydroborato ligands are summarized in Table 1.Interestingly, despite substantial diVerences in the size of the pyrazolyl substituents, there is little variation in the two M–N bond lengths for a given complex, i.e. Dd(M–N) ª 0. Fig. 2 Molecular structure of [H2B(pz)(pzBut 2)]ZnI(HpzBut,Pri). Selected bond lengths (Å) and angles (8): Zn–N(22) 2.011(2), Zn–N(12) 2.017(2), Zn–N(31) 2.038(2), Zn–I 2.6150(3), N(22)–Zn–N(12) 98.71(8), N(22)–Zn–N(31) 117.83(8), N(12)–Zn–N(31) 113.43(8), N(22)–Zn–I 127.46(6), N(12)–Zn–I 100.62(6), N(31)–Zn–I 97.95(5).Scheme 2 In summary, a new method for the synthesis of asymmetrically substituted bis(pyrazolyl)hydroborato ligands is provided by the reactions of LiBH4 with a ca. 1 : 1 molar mixture of two diVerent pyrazoles. Ligands which have been constructed using this method include [H2B(pz)(pzBut 2)]2, [H2B(pzMe2)(pzBut 2)]2 and [H2B(pzTrip)(pzBut 2)]2. As such, this method is more convenient and potentially more general than previously reported procedures for the syntheses of asymmetrically substituted bis(pyrazolyl)hydroborato ligands.Acknowledgements We thank the National Science Foundation (CHE 96-10497) and ELF Atochem North America, Inc., for support of this research. G. P. is the recipient of a Presidential Faculty Fellowship Award (1992–1997). Fig. 3 Molecular structure of [H2B(pzMe2)(pzBut 2)]ZnI(HpzBut 2). Selected bond lengths (Å) and angles (8): Zn–N(22) 1.978(4), Zn– N(12) 2.017(3), Zn–N(32) 2.090(3), Zn–I 2.5864(6), N(22)–Zn–N(12) 105.4(2), N(22)–Zn–N(32) 101.8(2), N(12)–Zn–N(32) 115.31(13), N(22)– Zn–I 113.44(10), N(12)–Zn–I 116.31(10), N(32)–Zn–I 103.94(10).Fig. 4 Molecular structure of [H2B(pzTrip)(pzBut 2)]ZnI. Selected bond lengths (Å) and angles (8): Zn–N(22) 1.994(2), Zn–N(12) 1.994(2), Zn–H(1) 2.41(3), Zn–I 2.4651(3), N(22)–Zn–N(12) 99.92(8), N(22)– Zn–H(1) 72.4(7), N(12)–Zn–H(1) 71.2(7), N(22)–Zn–I 130.08(6), N(12)–Zn–I 129.91(6), H(1)–Zn–I 116.6(7).J. Chem.Soc., Dalton Trans., 1998, 3355–3357 3357 Notes and references † Supplementary data available: experimental details and NMR data. For direct electronic access see http://www.rsc.org/suppdata/dt/1998/ 3355/, otherwise available from BLDSC (No. SUP 57435, 9 pp.) or the RSC Library. See Instructions for Authors, 1998, Issue 1 (http:// www.rsc.org/dalton). 1 For recent reviews, see: S. Trofimenko, Chem. Rev., 1993, 93, 943; G. Parkin, Adv. Inorg. Chem., 1995, 42, 291; N. Kitajima and W.B. Tolman, Prog. Inorg. Chem., 1995, 43, 419; I. Santos and N. Marques, New. J. Chem., 1995, 19, 551; D. L. Reger, Coord. Chem. Rev., 1996, 147, 571; M. Etienne, Coord. Chem. Rev., 1997, 156, 201; P. K. Byers, A. J. Canty and R. T. Honeyman, Adv. Organomet. Chem., 1992, 34, 1. 2 P. Ghosh and G. Parkin, Chem. Commun., 1998, 413; C. Dowling and G. Parkin, Polyhedron, 1996, 15, 2463; P. Ghosh and G. Parkin, J. Chem. Soc., Dalton Trans., 1998, 2281. 3 E. Frauendorfer and G. Agrifoglio, Inorg.Chem., 1982, 21, 4122; G. Agrifoglio, Inorg. Chim. Acta, 1992, 197, 159. 4 Specifically, attempts to synthesize the borane adduct of pyrazole, 4-nitropyrazole and diphenylpyrazole, resulted in elimination of H2 and the formation of pyrazolyl bridged dimers, [H2B(pzRR9)2BH2]. See ref. 3. 5 For example, [H2B(pz)(pzBut 2)]M (M = Li and Tl) are synthesized as follows. A solution of LiBH4 in THF (35 mL of 2 M, 70.0 mmol), diluted with toluene (ca. 35 mL), was added to a mixture of pyrazole (4.8 g, 70.5 mmol) and 3,5-di-tert-butylpyrazole (13.2 g, 73.2 mmol) and stirred overnight at room temperature.The solvent was removed in vacuo, toluene (ca. 40 mL) was added, and the resulting mixture was refluxed for one day. The solution was concentrated to ca. 20 mL, at which point the mixture became turbid and white solid started to precipitate. Pentane (ca. 50 mL) was added to complete precipitation and the mixture was filtered. The filtrate was allowed to stand at room temperature overnight, over which period colorless crystals of [H2B(pz)(pzBut 2)]Li were deposited (5.5 g, 30%).[H2B(pz)(pzBut 2)]Li was treated with TlOAc (8.2 g, 31.1 mmol) and THF (ca. 60 mL) and the mixture was stirred overnight at room temperature. The mixture was filtered and the residue was further extracted with pentane (ca. 50 mL). The THF and pentane extracts were combined and the solvent removed in vacuo to give [H2B(pz)(pzBut 2)]Tl as a white solid which was washed with pentane (7.4 g, 77%).[H2B(pzMe2)(pzBut 2)]Tl is synthesized similarly, with the exception that the lithium derivative was converted to the potassium complex prior to metathesis with TlOAc. 6 [H2B(pz)(pzBut 2)]Tl is monoclinic, P21/c (no. 14), C14H24BN4Tl, M = 463.55, a = 10.2116(5), b = 26.000(1), c = 14.1494(7) Å, b = 106.872(1)8, U = 3596.7(3) Å3, Z = 8, T = 293 K, m = 8.977 mm21, 8222 independent reflections, R1 = 0.0362 [I > 2s(I)]. [H2B(pz)- (pzBut 2)]ZnI(HpzBut,Pri) is monoclinic, P21/n (no. 14), C24H42BIN6Zn, M = 617.72, a = 15.4692(8), b = 12.2119(6), c = 16.4996(8) Å, b = 110.910(1)8, U = 2911.6(3) Å3, Z = 4, T = 203 K, m = 1.926 mm21, 6641 independent reflections, R1 = 0.0321 [I > 2s(I)]. [H2B(pzTrip)- (pzBut 2)]ZnI is triclinic P1� (no. 2), C34H36BIN4Zn, M = 703.75, a = 9.9469(5), b = 11.9592(5), c = 13.9733(6) Å, a = 90.085(1), b = 96.387(1), g = 103.157(1)8, U = 1607.9(1) Å3, Z = 2, T = 203 K, m = 1.752 mm21, 6874 independent reflections, R1 = 0.0327 [I > 2s(I )]. [H2B(pzMe2)(pzBut 2)]ZnI(HpzBut 2) is monoclinic, P21/n (no. 14), C27H48BIN6Zn, M = 659.79, a = 17.701(1), b = 9.5135(6), c = 21.543(2) Å, b = 113.781(2)8, U = 3319.7(4) Å3, Z = 4, T = 293 K, m = 1.693 mm21, 5797 independent reflections, R1 = 0.0444 [I > 2s(I)]. CCDC number 186/1142. 7 H. Kokusen, Y. Sohrin, M. Matsui, Y. Hata and H. Hasegawa, J. Chem. Soc., Dalton Trans., 1996, 195. 8 M. V. Capparelli and G. Agrifoglio, J. Crystallogr. Spectrosc. Res., 1992, 22, 651. Communication 8/0629

ISSN:1477-9226    

DOI:10.1039/a806298f

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1998, 3359–3361 3359 Synthesis and characterisation of Al(O3PCH2CO2)?3H2O, a layered aluminium carboxymethylphosphonate Gary B. Hix, David S. Wragg, Paul A. Wright and Russell E. Morris * Department of Chemistry, University of St. Andrews, St. Andrews, Fife, UK KY16 9ST. E-mail: gbh1@st-and.ac.uk Received 12th August 1998, Accepted 8th September 1998 Microcrystal X-ray diffraction has enabled the structure solution of a layered carboxylate–phosphonate in which aluminium atoms in the same layer are coordinated by both the carboxylic and phosphonic functional groups. Crystalline inorganic–organic hybrid materials are attracting increased attention because of the possibilities which exist for tuning the chemistry of a solid in order to target a specific property.One class of hybrid solid that has attracted particular attention is the metal phosphonates, which are of interest in many fields of chemistry.1 Their potential has been recognised in catalysis, ion exchange, charge storage and many other areas and they show great promise for future applications.2,3 The vast majority of phosphonates known to date incorporate tetravalent metals, especially zirconium, and divalent metals such as copper and zinc.Surprisingly, it is only recently that aluminium phosphonates have been reported in the literature, with crystal structures for only a handful of aluminium-methyland -phenyl-phosphonate solids having been published.3–7 The syntheses and characterisation of carboxylate–phosphonates of zinc, 8 copper 9 and cobalt 10 have recently been reported in the literature.In these materials the carboxylic acid group of the phosphonate anion is coordinated to the metal centres contained within the material. Copper and zinc carboxyethylphosphonates 8,9 have layered structures and the phosphonate groups bridge across the interlayer region in order for the carboxylic acid group to coordinate to the metal centres in an adjacent layer.Coordination is via the carbonyl oxygen, and consequently there is a reduction in the observed stretching frequency of the carbonyl, from ª1700 cm21 to 1583 cm21. Cobalt carboxyethylphosphonate,10 however, is a three dimensional framework structure, and in this material, it is the deprotonated oxygen atoms that coordinate to the metal centres leaving the carbonyl group as non-bonding. In contrast to this work, Bujoli and co-workers have described the structure determination of a gallium hydroxy carboxyethylphosphonate where the carboxy group is not coordinated to a metal, although they also report a similar hydrated solid, characterised by infrared spectroscopy and solid state NMR, where they propose that the carboxy group coordinates to a gallium atom in an adjacent layer forming a pillared layer structure similar to the copper and zinc phosphonates described earlier.11 This paper reports the synthesis and characterisation, using microcrystal diVraction at a synchrotron X-ray source, of a novel aluminium carboxymethylphosphonate. This structure is diVerent to the carboxyethylphosphonates prepared previously in that the phosphonate unit coordinates to an aluminium and a phosphorus atom in the same layer, rather than bridging across adjacent layers.This diVerence in structure can be explained by the shorter length of the carboxymethylphosphonate ligand when compared with the carboxyethylphosphonate ligand.Aluminium carboxymethylphosphonate was hydrothermally synthesised from a gel with composition Al(OH)3 [gibbsite] (0.53 g): H2O3PCH2CO2H (0.95 g) : n-butylamine (0.498 g): H2O (15 ml) in the ratio 1:1:1:120. The gel was stirred until homogeneous and heated in a Teflon-lined stainless steel autoclave of 24 ml volume at 160 8C for 48 hours. Immediately before heating the gel had a pH of 3.5. Products were recovered by filtration, washed with distilled water and dried in air at 60 8C.Inspection of the recovered sample revealed the presence of small crystals amongst a polycrystalline phase. The crystals were too small for single crystal X-ray data collection using a standard laboratory four circle diVractometer, so diVraction data (crystal size ª 30 × 20 × 10 mm) were collected at low temperature (160 K) using a Bruker AXS SMART CCD area-detector diVractometer on the high-flux singlecrystal diVraction station 9.8 at CCLRC Daresbury Laboratory Synchrotron Radiation Source, Cheshire, UK using a wavelength of 0.6045 Å.The single crystal structure determination† revealed that the material is layered. The composition calculated from the diVraction study, Al(O3PCH2CO2)?3H2O, is consistent with the microanalysis results (measured 10.99% C, 3.56% H, calculated 11.02% C, 3.69% H). The structure of Al(O3PCH2CO2)?3H2O comprises layers made up from square sub-units formed by coordination of two Al atoms by two oxygens from each of two PO3 22 groups (Fig. 1). The third oxygen from these sub-units then coordinates to an Al atom in an adjacent square to form the basic layer structure. The P–C bonds from the two phosphonate groups in each square, are directed one above and one below the plane of the layer. The C–C bonds are then directed away from the square in which the PO3 22 group is bound and the terminal carboxylate group is then coordinated to an Al atom in an adjacent sub-unit (Fig. 1). This Al atom is the same one to which the cross-linking P–O bond is attached. The octahedral coordination about the Fig. 1 Projection, perpendicular to one of the layers, viewed in the (100) direction. The AlO6 octahedra and the CPO3 tetrahedra are shown as open and shaded polyhedra respectively. Oxygen atoms are shown as open spheres and carbon atoms as filled spheres.3360 J. Chem. Soc., Dalton Trans., 1998, 3359–3361 Al atoms is completed by two water molecules, one above and one below the plane of the layer.The unbound oxygens of the carboxylate groups are directed alternately above and below the plane of the layer. The two C–O bond lengths from each of the carboxylate groups are in the range 1.248–1.283 Å, values which are between those expected for a C]] O bond (1.16– 1.21 Å) and a single C–O bond (1.34–1.43 Å), indicating that there is no distinct carbonyl group. The acidic oxygens are fully deprotonated, to leave a carboxylate anion, which is coordinated to the aluminium through one of the oxygen atoms.This observation is consistent with the single crystal X-ray diVraction experiments reported for zinc carboxyethylphosphonates 7 and is confirmed by the IR spectrum of the material (see below). The layers are held together by hydrogen bonding between water molecules situated in the interlayer region, coordinated water molecules and carboxylate oxygen atoms (Fig. 2). Rietveld refinement of the structural model against powder X-ray diVraction data (Cu-Ka, l = 1.5406 Å) confirms the structure of Al(O3PCH2CO2)?3H2O, and verifies that the as-made sample contains only one crystalline phase (Fig. 3).‡ Thermogravimetric analysis indicates that there are four stepwise mass losses. The first two, seen at 148 8C and 202 8C are accounted for by the loss of six water molecules. The ratio of these losses is 1 : 2. This can be rationalised by considering that the lower temperature event is the loss of the two water molecules in the interlayer region, and the higher temperature event is the loss of the two water molecules coordinated to each of the Al atoms (making four in total).There is then a further Fig. 2 The layered structure viewed in the (010) direction. The interlamellar water molecules can clearly be seen. The key for atom and polyhedra types is as in Fig. 1. Fig. 3 Final observed (crosses), calculated (solid line) and diVerence plots for the Rietveld refinement against powder X-ray diVraction data.loss at 253 8C of approximately 1.93 mass%. This is tentatively assigned as a loss of 1/2O2, arising from a condensation between two carboxylate groups. The final mass loss is centred at 430 8C and corresponds to removal of the organic part of the material to leave AlPO4. X-Ray powder diVraction studies show that removal of the water molecules results in a collapse of the structural integrity of the material.This is not surprising considering that removal of the water will result in a change in coordination number from six to four. It might be expected that the extended interlayer hydrogen bonding would indicate an ability to intercalate organic molecules such as amines. Attempts were made to intercalate n-butylamine into the structure both by placing the sample in an atmosphere containing n-butylamine vapour, and by contacting the sample with an aqueous solution of the amine.The powder pattern of the samples recovered, however, showed no changes from the profile of the original sample. This seems to underline the importance of the interlayer water molecules in maintaining the structure. The IR spectrum of aluminium carboxymethylphosphonate contains clear information regarding the nature of the carboxylic acid group and its coordination of the metal centre. The broad peaks observed at 1652 and 1398 cm21 arise from the symmetric and asymmetric stretching mode of the C–O bonds of the carboxy group. This is a lower frequency than one would expect to find for such a band were it to arise from a C]] O stretch in an aliphatic carboxylic acid, typically in the region 1725–1700 cm21.This shift to lower frequency is similar to that seen for zinc carboxyethylphosphonate 8 where the carboxylate group is coordinated to the zinc through one oxygen. The broad feature in the region 3700–2800 cm21 indicates the large amount of hydrogen bonding in this material, presumably through the interlayer water molecules which are responsible for holding the structure together.The extent of hydrogen bonding is also evident by the width of the carboxylate stretching bands. The 31P MAS NMR spectrum of aluminium carboxymethylphosphonate indicates the presence of two crystallographically distinct phosphorus environments. Peaks are observed at d 21.8 and 27.4, indicating two slightly diVerent P environments, which is consistent with the single crystal data.In conclusion, this work reports the synthesis and characterisation of an aluminium phosphonate with a novel layered architecture. The structure is unique in that the carboxymethylphosphonate group bridges a phosphorus and an aluminium atom in the same layer. This is in contrast to the carboxyethylphosphonate materials previously reported where the phosphonate ligand bridges a phosphorus and an aluminium in adjacent layers, producing a three dimensional material, or, as in the case of gallium carboxyethylphosphonate, the carboxyl group is not coordinated to a metal.The length of the carbon spacer is obviously important in determining how the carboxyl functional group coordinates. Carboxyethylphosphonates, with two carbon atoms in the chain, are too long to easily bridge neighbouring phosphorus and aluminium atoms in the same layer, whereas carboxymethyl phosphonates, with only one carbon, can. Acknowledgements The authors thank the EPSRC and the Royal Society of Edinburgh (REM) for support.We would also like to thank Dr Simon Coles and Dr David Taylor for their help in collecting the single crystal X-ray diVraction data. Notes and references † Crystal structure determination of Al(O3PCH2CO2)?3H2O: M = 218.04, triclinic, space group P1� , a = 8.1200(2), b = 8.2536(2), c = 11.2848(3) , a = 89.032(1), b = 79.231(1), g = 74.733(1)8, U = 716.34(3) Å3, Z = 4, T = 160 K, 2986 reflections of which 2337 observed.Final refinement of the 97 least-squares parameters converged to wR(F2 obsd data) = 0.23, R(Fobsd data) = 0.088, S(F2 all data) = 1.226. CCDC reference number 186/1156.J. Chem. Soc., Dalton Trans., 1998, 3359–3361 3361 ‡ Rietveld refinement of Al(O3PCH2CO2)?3H2O: Stoe STADIP transmission diVractometer in capillary mode. Ge-monochromated Cu-Ka radiation (l = 1.5406 Å). Final profile refinement Rwp = 0.1351, RF = 0.1004 for 251 reflections in the range 5 < 2q < 508. 1 G. Alberti, M. Cascioli, U. Costantino and R. Vivani, Adv. Mater., 1996, 8, 291. 2 B. Zhang and A. Clearfield, J. Am. Chem. Soc., 1997, 119, 2751; C. Bhardwaj, H. Hu and A. Clearfield, Inorg. Chem., 1993, 32, 4299; E. Jaimez, A. Bortun, G. Hix, J. Garcia, J. Rodriguez and R. Slade, J. Chem. Soc., Dalton Trans., 1996, 11, 2285; G. Alberti, U. Constantino, M. Casciola, R. Vivani and A. Peraio, Solid State Ionics, 1992, 46, 61; L. Vermuelen and M. Thompson, Nature (London), 1992, 358, 656. 3 K. Maeda, J. Akimoto, Y. Kiyozumi and F. Mizukami, J. Chem. Soc., Chem. Commun., 1995, 10, 1033; K. Maeda, J. Akimoto, Y. Kiyozumi and F. Mizukami, Angew. Chem., Int. Ed. Engl., 1995, 34, 1199; K. Maeda, J. Akimoto, Y. Kiyozumi and F. Mizukami, Stud. Surf. Sci. Catal., 1997, 105, 197. 4 L.-J. Sawers, V. J. Carter, A. R. Armstrong, P. G. Bruce, P. A. Wright and B. E. Gore, J. Chem. Soc., Dalton Trans., 1996, 3159. 5 L. Raki and C. Detellier, Chem. Commun., 1996, 2475. 6 V. J. Carter, P. A. Wright, J. D. Gale, R. E. Morris, E. Sastre and J. Perez-Pariente, J. Mater. Chem., 1997, 7, 2287. 7 G. B. Hix, V. J. Carter, D. S. Wragg, R. E. Morris and P. A. Wright, J. Mater. Chem., in the press. 8 S. Drumel, P. J. Janvier, P. Barboux, M. Bujoli-DoeuV and B. Bujoli, Inorg. Chem., 1995, 34, 148. 9 S. Drumel, P. J. Janvier, M. Bujoli-DoeuV and B. Bujoli, New J. Chem., 1995, 19, 239. 10 A. Distler and S. C. Sevov, Chem. Commun., 1998, 959. 11 F. Fredoueil, D. Massiot, D. Poojary, M. Bujoli-DoueV, A. Clearfield and B. Bujoli, Chem. Commun., 1998, 175. Communication 8/0637

ISSN:1477-9226    

DOI:10.1039/a806373g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1998, 3363–3365 3363 Reaction of iodocarbonylbis(trimethylphosphine)rhodium(I) with parahydrogen leads to the observation of five characterisable H2 addition products Paul D. Morran,a Simon A. Colebrooke,a Simon B. Duckett,*a Joost A. B. Lohmanb and Richard Eisenberg*c a Department of Chemistry, University of York, Heslington, York, UK YO1 5DD b Bruker UK Limited, Coventry, UK c Department of Chemistry, University of Rochester, Rochester, NY 14627, USA Received 24th August 1998, Accepted 1st September 1998 The complex RhI(CO)(PMe3)2 reacts with dihydrogen to generate five different products that have been identified and structurally assigned through the use of parahydrogen induced polarisation.When the addition of hydrogen to a metal centre is monitored by NMR spectroscopy in conjunction with dihydrogen specially enriched in the para or anti-symmetric spin state,1 the observation of previously unseen hydride products becomes possible.2 In this approach, the spin-correlated activation of parahydrogen leads to non-Boltzmann spin populations for the product hydride nuclei and consequent enhanced absorptions and emissions in their 1H NMR signals.3 The reaction chemistry presented here concerns the addition of H2 to RhI(CO)- (PMe3)2 1, and shows the power of this method in detecting and identifying diVerent dihydride products, including very minor components, from a reaction mixture.It has previously been reported that when RhCl(CO)(PPh3)2 is warmed with parahydrogen only the binuclear product 2 is detected, whereas with the PMe3 analogue, only the hydride-bridged product 3 was seen.4 Products 2 and 3 are related by interchange of a bridging chloride for hydride.When a C6D6 solution of RhI(CO)(PMe3)2, 1, under 3 atm of para-enriched hydrogen (p-H2) is monitored by 1H NMR spectroscopy at 348 K, the spectrum shown in Fig. 1(a) is obtained. Close examination reveals the presence of 10 hydride resonances belonging to five reaction products (Table 1). We show below that three of the products, 4, 5 and 6, are isomeric analogues of 2 and 3.Each pair of product resonances exhibits anti-phase lines (emission/absorption or E/A) characteristic of parahydrogen induced polarisation (PHIP) with the sense of the anti-phase lines indicating that all of the hydride–hydride spin–spin couplings are negative. In the corresponding 1H-{31P} NMR spectrum, the resonances at d 210.00, 214.63, 214.75, 214.80, 215.83, 217.70, and 218.02 simplify into anti-phase doublets of doublets, showing that each is coupled to one 103Rh nucleus as a terminal hydride ligand, while the three resonances at d 211.18, 214.46 and 216.73 couple to two rhodium centres indicating that they correspond to bridging hydride ligands [Fig. 1(b)]. The connectivity of the hydride ligands was established by a gradient assisted 1H–1H correlation spectrum (COSY) employing a 458 starting pulse and a 908 read pulse.The resulting spectrum [Fig. 1(c)] reveals the connectivity of five pairs of hydride resonances (Table 1). In order to fully characterise the reaction products, a series of 1H–31P, 1H–13C and 1H–103Rh spectra were then recorded. At 348 K the most intense signals originate from the mutually coupled hydride resonances at d 214.46 and d 214.63. The terminal hydride resonance at d 214.63 connects with a single 103Rh nucleus, d 2210 (JRhH = 23 Hz), and two equivalent 31P nuclei (JPH = 15.5 Hz) which resonate at d 215.9 in the corresponding HMQC experiments. In addition, the 1H–31P HMQC experiment reveals that the hydride partner at d 214.46, identi- fied above as bridging, couples to two separate sets of 31P nuclei, which are themselves not coupled, at d 26.0 (JPH = 14, JRhP = 155 Hz) and 215.9 (JPH = 32, JRhP = 96 Hz) respectively.The change in JRhP from 155 Hz to 96 Hz for these resonances is consistent with an oxidation state change from I to III for the corresponding rhodium centre.The rhodium(I) centre, located in a 1H–103Rh spectrum at d 2733, couples to this hydride only. When compound 1 is specifically labelled with 13CO, the resonance due to the bridging hydride of this product possesses an additional coupling of 3 Hz and correlates with a 13C resonance at d 186.1 in the corresponding 1H–13C spectrum. The identity of the product is therefore clearly established as (H)(I)Rh- (PMe3)2(m-H)(m-I)Rh(CO)(PMe3) 4, analogous to 3.Although the iodide ligands are NMR silent, their presence is required to satisfy the rhodium oxidation state requirements. A minor reaction product, 5, with hydride resonances at d 214.80 and 216.73, possesses similar spectral features to those of 4 which indicate similar ligand spheres and the presence of Rh(I) and Rh(III) centres. However, in this isomer, the couplings from phosphorus to the bridging hydride are 13.5 and 14 Hz, while the hydride coupling to 13CO has increased to 11 Hz.The decrease in JPH, from 32 to 13.5 Hz, and increase in J(13CO)H, from 3 to 11 Hz, on moving from 4 to 5 is consistent with the argument that 4 contains a m-hydride which is trans to the PMe3 ligand of the Rh(I) centre, while in 5 the m-hydride is trans to CO. The small size of JPH associated with a trans P–Rh(I)–H arrangement is consistent with that previously observed for (H)2Rh2{P[N(CH3)2]3}4.5 At 313 K the remaining resonances become sharper and are hence easier to monitor. Four of these resonances are assigned to the products (H)2Rh(PMe3)2(m-I)2Rh(CO)(PMe3) 6, and Rh(H)2I(PMe3)3 7.The mutually coupled hydride resonances of 6 appear at d 217.70 and 218.02, and possess additional couplings due to one rhodium and two phosphorus nuclei, the resonances of which were located at d 74 and 216.4 (JRhP = 113 Hz) in the corresponding 1H–103Rh and 1H–31P correlation experiments, respectively.Evidence for the iodide bridges comes directly from the chemical shifts of the hydride ligands of 6, d 217.70 and 218.02, which require the presence of trans electronegative groups. Furthermore, the3364 J. Chem. Soc., Dalton Trans., 1998, 3363–3365 Fig. 1 1H NMR spectra showing the hydride region after warming a sample of RhI(CO)(PMe3)2 1 with p-H2 in C6D6. (a) 1H spectrum at 348 K; (b) 1H-{31P} spectrum at 333 K; (c) selected cross peaks (absolute value display) and projections in the 1H–1H correlation spectrum, using gradients and 31P decoupling showing hydride–hydride connectivity. Table 1 Selected 1H, 31P, 13C and 103Rh NMR spectral data for the reaction products of RhI(CO)(PMe3)2 with H2 (all 1H data tested by simulation) a Complex 4 1H 214.46 (ddddt, 2JHH = 23.5, 1JRhH = 19 and 30, 2JPH = 14 and 32, 2JHCO = 3) 214.63 (ddt, 2JHH = 23.5, 1JRhH = 23, 2JPH = 15.5) 31P 26.0 (1JRhP = 155) 215.9 (1JRhP = 96) 13C 186.1 (2JPC = 16.9, 1JRhC = 79.8) 103Rh 2210 2733 5 216.73 (ddddt, 2JHH = 23.5, 1JRhH = 16.7 and 31.5, 2JPH = 14 and 13.5, 2JHCO = 11) 214.80 (ddt, 2JHH = 23.5, 1JRhH = 21, 2JPH = 16) 20.6 (1JRhP = 151) 216.5 (1JRhP = 97) 183.8 — 6 217.70 (ddt, 2JHH = 29, 1JRhH = 32, 2JPH = 15.6) 218.02 (ddt, 2JHH = 29, 1JRhH = 30.7, 2JPH = 16.3) 216.4 (1JRhP = 113) — 74 7 210.0 (dddt, 2JHH = 27, 1JRhH = 14.4, 2JPH = 176, 2JPH = 17.4) 215.83 (ddq, 2JHH = 27, 1JRhH = 28.1, 2JPH = 36) 213.6 (1JRhP = 102) 231.3 (1JRhP = 94) — 2644 8 211.18 (dddt, 2JHH = 26, 1JRhH = 19 and 19, 2JPH = 99.5, 21 and 16) 214.75 (ddt, 2JHH = 26, 1JRhH = 28, 2JPH = 15 and 15) 210.34 (trans to hydride, 1JRhP = 120, 4JPP = 70, 2JPP = 40) 0.6 (1JRhP = 145, 4JPP = 70) 10.38 (1JRhP = 126, 2JPP = 40) — — ad in ppm, J in Hz magnitudes of couplings only for 2JHH.Scheme 1 similarity of these chemical shifts to those previously reported for the related complex (H)2Rh(PPh3)2(m-I)2Rh(CO)(PPh3) points to the presence of a second Rh(I) centre.4 Additionally, the rhodium chemical shift, d 74, is very similar to that of d 32 observed for the related complex (H)2Rh(PPh3)2(m-I)2- Rh(CO)(PPh3). For 7, the hydride resonances appear at d 210.00 and 215.83 and contain splittings which imply the presence of one rhodiumJ.Chem. Soc., Dalton Trans., 1998, 3363–3365 3365 and three phosphorus nuclei. The resonance at d 210.00 possesses a large P–H coupling of 176 Hz, requiring the corresponding hydride ligand to be located trans to phosphine.The higher field resonance, d 215.83, arises from the hydride ligand which is trans to the iodide group. The final product contains a terminal hydride (d 214.75) cis to two phosphorus centres and a bridging hydride ligand which resonates at d 211.18 and possesses a large phosphorus coupling, JPH = 99.5 Hz, indicative of an essentially trans H–Rh– PMe3 arrangement. Further couplings to two other phosphorus nuclei, JPH = 21 and 16 Hz, and essentially equal couplings to two rhodium centres, JRhH = 19 Hz, are observed to the bridging hydride.The terminal hydride resonance at d 214.75 does not connect with the 31P nucleus at d 0.6 in the corresponding 1H–31P HMQC spectrum which confirms its location on the second rhodium centre. In the complexes, (m-H)2Rh2- {P[N(CH3)2]3}4 and (m-H)2Rh2{P[OPri]3}4, containing rhodium( I) centres with near coplanar framework atoms P2Rh- H2RhP2, reported by Muetterties, the corresponding Hbridge–P couplings are established at around 30 Hz.5,6 The related complex 9, shown below, was also detected with P[OPri]3 and has been reported to possess hydride resonances for Ha at d 211.1 (JPH = 89 Hz, 1H) and Hb at d 214.5 (1H).6 Based on the similarities between the JPH couplings and the chemical shifts of the terminal and bridging hydride ligands observed for the new species we suggest that it is most likely to adopt the structure shown as 8.When RhI(13CO)(PMe3)2 is employed, no additional hydride–13C couplings are visible; however, the resonance line width is suYcient to hide a small cis coupling indicating that the bridging hydride in 8 is cis to the carbonyl of the second rhodium centre.The formation of products 4–8 can be envisioned to proceed according to Scheme 1 in which oxidative addition of H2 to 1 is followed by rapid CO loss. Subsequent addition of excess 1 to the unsaturated species Rh(H)2I(PMe3)2 leads to the binuclear products while addition of PMe3 generates 7.Consistent with this view, the signals due to 7 decrease substantially as the concentration of 1 is increased. Since the binuclear products 4, 5 and 6 arise from a common intermediate, it is possible to assess the relative rates of the product forming steps from the relative magnitudes of the observed polarisation. For 4 and 5 for example, the intensity diVerence is 15 : 1 at 358 K reflecting a DG‡ of 8.1 kJ mol21 in the key product forming step.6 The same analysis involving 6 is more ambiguous because facile dihydride interchange broadens the hydride resonances significantly at this temperature. In this paper we have shown that parahydrogen induced polarisation can be used to monitor complex thermal reactions that are not normally decipherable or even observable.The complexity of a reaction system, as apparently simple as H2 addition to 1, is surprising and underscores the pitfalls in drawing up catalytic reaction mechanisms in which isomeric structures of intermediates are ignored. Acknowledgements Financial support from the EPSRC and Bruker UK (Spectrometer, CASE award to SAC), the Royal Society (SBD), NATO, the National Science Foundation (RE), University of York (PDM), and discussions with Mr C. Sleigh, Professor R. N. Perutz and Dr R. J. Mawby are gratefully acknowledged. References 1 C. R. Bowers and D. P. Weitekamp, J. Am. Chem. Soc., 1987, 109, 5541; R. Eisenberg, Acc. Chem. Res., 1991, 24, 110; J. Natterer and J. Bargon, Prog. Nucl. Magn. Reson. Spectrosc., 1997, 31, 293. 2 S. B. Duckett, C. L. Newell and R. Eisenberg, J. Am. Chem. Soc., 1993, 115, 1156. 3 S. B. Duckett, R. J. Mawby and M. G. Partridge, Chem. Commun., 1996, 383; J. Barkemeyer, M. Haake and J. Bargon, J. Am. Chem. Soc., 1995, 117, 2927. 4 S. B. Duckett and R. Eisenberg, J. Am. Chem. Soc., 1993, 115, 5292; S. B. Duckett, R. Eisenberg and A. S. Goldman, J. Chem. Soc., Chem. Commun., 1990, 511. 5 E. B. Meier, R. R. Burch and E. L. Muetterties, J. Am. Chem. Soc., 1982, 104, 2661. 6 A. J. Sivak and E. L. Muetterties, J. Am. Chem. Soc., 1979, 101, 4878. Communication 8/06652C

ISSN:1477-9226    

DOI:10.1039/a806652c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3367–3372 3367 Synthesis and reactivity of cationic lanthanide metallocene complexes. Hexabromocarborane and tetraphenylborate as counter ions Zuowei Xie,* Zhixian Liu, Zhong-Yuan Zhou and Thomas C. W. Mak Department of Chemistry, The Chinese University of Hong Kong, Shatin NT, Hong Kong, China. E-mail: zxie@cuhk.edu.hk Received 8th June 1998, Accepted 3rd August 1998 Treatment of unsolvated Cp02Ln(II) [Cp0 = 1,3-(Me3Si)2C5H3] with 1 equivalent of Ag(I)Y, or reaction of [Cp02LnI]2 with 2 molar equivalents of Ag(CB11Br6H6) in pure toluene at room temperature gave “Lewis base-free” cationic lanthanide metallocene complexes [Cp02Ln]Y (Y = BPh4 2, Ln = Sm 1, Yb 2; Y = CB11Br6H6 2, Ln = Sm 3, Er 4) in good yield.They slowly undergo decomposition reaction at room temperature. The reactivity of these “Lewis base-free” cationic complexes is highly dependent upon the coordinating nature of the counter ions.Complexes 3 and 4 are much more reactive than 1 and 2. Recrystallization of 1 and 2 from 1,2-dimethoxyethane (DME) and THF yielded [Cp02Sm(DME)][BPh4] 5 and [Cp02Yb(THF)2][BPh4] 6, respectively. However, recrystallization of 3 and 4 from THF resulted in the ring-opening polymerization of THF. The THF coordinated complexes [Cp02Ln(THF)2][CB11Br6H6] (Ln = Sm 7, Er 8) were isolated via recrystallization of 3 and 4 from toluene containing a small amount of THF. The “Lewis base-free” cation “Cp02Er1” can abstract one bromine atom from the counter ion CB11Br6H6 2 or one chlorine atom from CH2Cl2 to form [Cp02ErBr]2 or [Cp02ErCl]2, respectively.Unfortunately, these cations do not exhibit reactivity towards 1-hexene at room temperature. Molecular structures of 5, 7, 8 and [Cp02ErBr]2 have been confirmed by single-crystal X-ray analyses. Introduction Cationic Group 4 metallocene complexes have been the subject of intensive research activity over the past decade because of their extraordinary characteristics as olefin polymerization catalysts.1 These studies have helped chemists to better understand the fundamental aspects of olefin oligomerization and polymerization reactions.In contrast to this Group 4 system, cationic lanthanide metallocene chemistry is much less developed.2 Only a few reports on this type of complex have appeared in the literature.3–9 All of the known cationic lanthanide metallocene complexes are those in which a lanthanide metal is strongly bound to the Lewis bases (donor solvents), [Cp2LnL2][X] [L = OC4H8, SC4H8, N2H4, DME; Ln = La,4 Ce,5 Sm,3,6 Yb;7–9 X = BPh4,3–8 Co(CO)4 9].The coordinated Lewis bases cannot be removed from the metal centre under very high vacuum on heating,3 which results in very unreactive cationic species. In fact, [Cp*2Sm(THF)2]1 (Cp* = C5Me5) has shown no reactivity towards CO, azobenzene, ethylene, phenylacetylene, epoxybutane or pyridine.3 In view of recent developments in cationic Group 4 metallocene chemistry,10 the synthesis, structure and reactivity studies of “base-free” cationic lanthanide metallocene compounds becomes a very interesting subject.The first example of such a “base-free” cationic Group 3 metallocene compound, Cp92YMe?B(C6F5)3 (Cp9 = C5H5, C5H4SiMe3), has been reported very recently.11 It is much more reactive than the Lewis base coordinated cationic Group 3 species, [Cp2LnL2][X].3–9 We are interested in the eVect of counter ions on the stability and reactivity of cationic lanthanide metallocene compounds.We report here the synthesis and reactivity of several new cationic lanthanide metallocene compounds in which the counter ions are the hexabromocarborane anion [7,8,9,10,11,12-Br6-1-closo-CB11H6]2 and tetraphenylborate BPh4 2, respectively. Results and discussion Synthesis Treatment of unsolvated Cp02Ln(II) [Cp0 = 1,3-(SiMe3)2C5H3] with 1 equivalent of AgY in toluene at room temperature and removal of the Ag precipitate gave a complex which has been formulated as [Cp02Ln]Y on the basis of elemental analyses and chemical analyses [eqn.(1), Y = BPh4 2, Ln = Sm 1, Yb 2; Cp02Ln 1 AgY toluene [Cp02Ln]Y 1 Ag (1) Y = CB11Br6H6 2, Ln = Sm 3]. Complex [Cp02Yb][CB11Br6H6] was not isolated in the pure form because of its decomposition at room temperature. For those metals, for which a divalent precursor is not available, an unsolvated trivalent complex [Cp02LnI]2 can be used instead.Reaction of [Cp02LnI]2 with two molar equivalents of Ag(CB11Br6H6) in toluene at room temperature, and removal of the AgI precipitate yielded a complex [Cp02Ln][CB11Br6H6] as shown in eqn. (2) (Ln = Sm 3, Er 4). ��� [Cp02LnI]2 1 Ag(CB11Br6H6) toluene [Cp02Ln][CB11Br6H6] 1 AgI (2) These new complexes (1–4) are extremely air and moisture sensitive, and relatively thermally unstable. They do not redissolve in arene solvents once isolated, so that recrystallization from toluene, benzene or fluorobenzene is extremely diYcult.Numerous attempts were made during the course of this study to grow single crystals of complexes 1–4 from their motherliquor suitable for X-ray analysis. However, all eVorts were hampered by the slow decomposition of these complexes at room temperature, resulting in diYculty in the characterization3368 J. Chem. Soc., Dalton Trans., 1998, 3367–3372 of these complexes. The reaction of [Cp02SmI]2 with AgSbF6 gave the fluoride abstraction product [Cp02SmF]2.12a The reaction of [Cp02SmCl]2 with Ag(CB11Br6H6) was slow and did not proceed to completion.Reactivity Complexes 1–4 are not thermally stable, and 3 and 4 are even less stable than 1 and 2. They slowly undergo decomposition at room temperature. Not surprisingly, heat can dramatically increase the thermolysis rate. Thus complex 1 decomposed into the more stable Cp03Sm in 50% isolated yield at 80 8C. The decomposition of complex 2 may be complicated.Unlike complex 1, no Cp03Yb was isolated presumably due to the smaller size of the ytterbium ion which makes the formation of Cp03Yb extremely diYcult.13 On the other hand, a small amount of benzene was detected by GC/MS from its hydrolysis products, suggesting that Cp02Yb1 may abstract one phenyl group from the counter ion BPh4 2.14 Thermal decomposition of 3 in toluene or fluorobenzene produced Cp03Sm, and some unidentified species. No [Cp02SmF]2 12 was detected by MS spectroscopy.Complex 4 in toluene or fluorobenzene decomposed to generate [Cp02ErBr]2, and some unidentified species at room temperature. No fluoride abstraction product was observed either. Reaction of Cp02Yb with 1 equivalent of Ag(CB11Br6H6) in toluene at room temperature resulted in the isolation of [Cp02YbBr]2 in 11% yield which could be increased to 41% at 40 8C. It is rational to propose that compounds [Cp02LnBr]2 resulted from the bromide abstract reaction between “Cp02Ln1” and the counter ion CB11Br6H6 2.The above results indicate that “Cp02Ln1” is an extremely strong electrophile and its electrophilicity is even higher than that of the silylium ion (R3Si1) at least with respect to the bromide abstraction reactions.15 These cationic lanthanide metallocene species can also abstract the chloride from CH2Cl2. For example, treatment of 4 with CH2Cl2 at room temperature aVorded [Cp02ErCl]2 in 65% yield; reaction of Cp02Yb with Ag(CB11Br6H6) in CH2Cl2 gave [Cp02YbCl]2 in 70% yield.“Base-free” cationic lanthanide metallocene complexes are strong Lewis acids. Recrystallization of complex 1 from DME/toluene produced a DME coordinated cationic complex [Cp02Sm(DME)][BPh4] 5. Complex 2 dissolved in THF to yield a THF coordinated ionic complex [Cp02Yb(THF)2][BPh4] 6. No obvious THF ring-opening polymerization products were observed. On the other hand, recrystallization of complex 3 or 4 from THF resulted in the formation of a gum identified as poly(tetrahydrofuran).The THF coordinated complex [Cp02Ln(THF)2][CB11Br6H6] (Ln = Sm 7, Er 8) can be isolated by recrystallization of the corresponding “Lewis base-free” complex [Cp02Ln][CB11Br6H6] from toluene containing a small amount of THF. Like other Lewis base coordinatednic lanthanide metallocene complexes,3–9 7 and 8 cannot initiate the ring-opening polymerization of THF. Attempts to prepare complexes 7 and 8 by the reaction of Cp02Ln with Ag(CB11- Br6H6) in THF, a well-established method for the preparation of [Cp2Ln(THF)2][BPh4],3–9 failed due to the formation of poly(tetrahydrofuran). These results imply that [Cp02Ln]- [CB11Br6H6] is much more electrophilic than [Cp02Ln][BPh4], suggesting that the less coordinating carborane anion can largely enhance the electrophilicity (Lewis acidity) of Cp02Ln1 cation, which is consistent with cationic Group 4 and silylium ion chemistry.1,10,15 However, neither 3 nor 4 exhibits any reactivity towards 1-hexene at room temperature.Structure Crystal structures of cationic lanthanide metallocene complexes 5, 7 and 8. The solid state structures of 5, 7 and 8 as derived from single-crystal X-ray diVraction studies all consist of well separated, alternating layers of [Cp02LnL2]1 (L2 = DME, 2THF) cations and CB11Br6H6 2 or BPh4 2 anions. The apparent voids in the structures are filled by several disordered solvent molecules.Complexes 7 and 8 are isomorphous. In each cation, the lanthanide ion is h5 bound to each of two cyclopentadienyl rings and two oxygen atoms from the coordinated THF or DME molecules in a distorted tetrahedral geometry (Figs. 1–3), which is similar in structure to the reported lanthanide metallocene cations of the type Cp2LnL2 1.3–9 Selected bond distances and angles for each Fig. 1 Perspective ORTEP16 drawing of the molecular structure of [Cp02Sm(DME)]1 in 5?0.25C7H8.All hydrogen atoms are omitted for clarity (thermal ellipsoids drawn at the 35% probability). Fig. 2 Perspective ORTEP drawing of the molecular structure of [Cp02Sm(THF)2]1 in 7?OC4H8. All hydrogen atoms are omitted for clarity (thermal ellipsoids drawn at the 35% probability). Fig. 3 Perspective ORTEP drawing of the molecular structure of [Cp02Er(THF)2]1 in 8?1.25C7H8. All hydrogen atoms are omitted for clarity (thermal ellipsoids drawn at the 35% probability).J. Chem.Soc., Dalton Trans., 1998, 3367–3372 3369 complex are summarized in Tables 1–3, respectively. Table 4 lists some structural parameters for lanthanide metallocene cations of the type Cp2LnL2 1. Both BPh4 2 and CB11Br6H6 2 anions have normal distances and angles. The two cations, [Cp02Sm(DME)]1 and [Cp02Sm(THF)2]1, have a very similar geometry in terms of average Sm–C, Sm–Cent (centroid of the cyclopentadienyl ring), Sm–O bond distances and Cent–Sm–Cent angles. These structural parameters can be compared to those found in [(C5Me5)2- Sm(THF)2]1 (Table 4).3 The Er–C distances in [Cp02Er- (THF)2]1 range from 2.563(6) to 2.694(7) Å with an average value of 2.627(6) Å.This measured value is identical to the 2.629(2) Å in [Cp02ErBr]2. It is about 0.083 Å shorter than the Table 1 Selected bond distances (Å) and angles (8) for compound 5 Sm(1)–C(1) Sm(1)–C(2) Sm(1)–C(3) Sm(1)–C(4) Sm(1)–C(5) Sm(1)–C(6) Sm(1)–Cent(1) O(1)–Sm(1)–O(2) Cent(1)–Sm(1)–Cent(2) Cent(1)–Sm(1)–O(1) 2.732(16) 2.720(17) 2.680(18) 2.719(21) 2.753(17) 2.695(15) 2.436 65.1(5) 129.4 106.4 Sm(1)–C(7) Sm(1)–C(8) Sm(1)–C(9) Sm(1)–C(10) Sm(1)–O(1) Sm(1)–O(2) Sm(1)–Cent(2) Cent(1)–Sm(1)–O(2) Cent(2)–Sm(1)–O(1) Cent(2)–Sm(1)–O(2) 2.710(16) 2.693(17) 2.695(16) 2.739(18) 2.432(17) 2.386(14) 2.427 115.0 110.4 111.5 Table 2 Selected bond distances (Å) and angles (8) for compound 7 Sm(1)–C(2) Sm(1)–C(3) Sm(1)–C(4) Sm(1)–C(5) Sm(1)–C(6) Sm(1)–C(7) Sm(1)–Cent(1) O(1)–Sm(1)–O(2) Cent(1)–Sm(1)–Cent(2) Cent(1)–Sm(1)–O(1) 2.70(2) 2.70(2) 2.77(2) 2.68(2) 2.67(2) 2.70(2) 2.420 94.1(4) 130.2 104.1 Sm(1)–C(8) Sm(1)–C(9) Sm(1)–C(10) Sm(1)–C(11) Sm(1)–O(1) Sm(1)–O(2) Sm(1)–Cent(2) Cent(1)–Sm(1)–O(2) Cent(2)–Sm(1)–O(1) Cent(2)–Sm(1)–O(2) 2.74(2) 2.74(2) 2.69(2) 2.69(2) 2.424(12) 2.401(11) 2.432 108.2 109.0 105.4 Table 3 Selected bond distances (Å) and angles (8) for compound 8 Er(1)–C(2) Er(1)–C(3) Er(1)–C(4) Er(1)–C(5) Er(1)–C(6) Er(1)–C(7) Er(1)–Cent(1) O(1)–Er(1)–O(2) Cent(1)–Er(1)–Cent(2) Cent(1)–Er(1)–O(1) 2.590(8) 2.666(8) 2.641(7) 2.608(6) 2.608(7) 2.630(8) 2.324 91.2(2) 131.4 106.4 Er(1)–C(8) Er(1)–C(9) Er(1)–C(10) Er(1)–C(11) Er(1)–O(1) Er(1)–O(2) Er(1)–Cent(2) Cent(1)–Er(1)–O(2) Cent(2)–Er(1)–O(1) Cent(2)–Er(1)–O(2) 2.655(8) 2.694(7) 2.613(6) 2.563(6) 2.335(4) 2.336(4) 2.342 107.7 108.2 104.5 average Sm–C distance of 2.71(2) Å in 7.The average Er–O distance of 2.336(4) Å is 0.075 Å shorter than that of Sm–O in 7.These diVerences can be compared to the 0.075 Å diVerence between Shannon’s ionic radii 17 of eight-coordinate Sm31 (1.079 Å) and Er31 (1.004 Å). The 131.4(1)8 Cent–Er–Cent angle is quite similar to the 130.2(3)8 in 7 and the 129.4(4)8 in 5. Crystal structure of [Cp02ErBr]2. [Cp02ErBr]2 is a centrosymmetric bromide-bridging dimer with pseudo-tetrahedral geometry around each Er atom, typical of the structures of [Cp02LnX]2 dimeric organolanthanide fluoride 12 and chloride 18a–c complexes (Fig. 4). Selected bond distances and angles are listed in Table 5. The average Er–C distance of 2.629(2) Å compared to the 2.640(9) Å in Cp02ErI(THF) 19 and the 2.59(1) Å in [(C5H5)2ErBr]2.18d The 2.875(1) Å average Er–Br distance is longer than the value of 2.820(1) Å in [(C5H5)2ErBr]2 probably due to steric reasons. The 83.2(1)8 Br(1)–Er(1)–Br(1A) angle is smaller than the 84.2(1)8 in [(C5H5)2ErBr]2. Fig. 4 Perspective ORTEP drawing of the molecular structure of [Cp02ErBr]2.All hydrogen atoms are omitted for clarity (thermal ellipsoids drawn at the 35% probability). Table 5 Selected bond distances (Å) and angles (8) for compound [Cp02ErBr]2 Er(1)–C(1) Er(1)–C(2) Er(1)–C(3) Er(1)–C(4) Er(1)–C(5) Er(1)–C(6) Er(1)–C(7) Er(1)–C(8) Br(1)–Er(1)–Br(1A) Cent(1)–Er(1)–Cent(2) Cent(1)–Er(1)–Br(1) 2.631(2) 2.644(2) 2.617(2) 2.616(2) 2.627(2) 2.659(2) 2.655(2) 2.600(2) 83.2(1) 130.3 107.2 Er(1)–C(9) Er(1)–C(10) Er(1)–Br(1) Er(1)–Br(1A) Br(1)–Er(1A) Er(1)–Cent(1) Er(1)–Cent(2) Cent(1)–Er(1)–Br(1A) Cent(2)–Er(1)–Br(1) Cent(2)–Er(1)–Br(1A) 2.586(2) 2.659(2) 2.878(1) 2.872(1) 2.872(1) 2.334 2.338 107.0 109.2 109.7 Table 4 Selected structural parameters for some lanthanide metallocene cations [Cp02Sm(DME)]1 [Cp02Sm(THF)2]1 [Cp02Er(THF)2]1 [Cp*2Sm(THF)2]1 [Cp*2Sm(N2H4)(THF)]1 [Cp02La(DME)(MeCN)]1 [Cp*2Ce(SC4H8)2]1 [(MeOCH2CH2C5H4)2Yb(THF)]1 [Cp*2Yb(THF)2]1 [(ButC5H4)2Yb(THF)2]1 ave.Ln–C/Å 2.714(14) 2.71(2) 2.627(7) 2.69(2) 2.73(2) 2.83(1) 2.74(3) 2.57(2) 2.619(4) 2.586(1) ave.Ln–O/Å 2.409(17) 2.413(12) 2.335(4) 2.46(1) 2.470(2) 2.627(8) 2.32(1) 2.341(3) 2.290(5) Cent(1)–Ln–Cent(2)/8 129.4 130.2 131.4 134.2 138.9 134.6(1) 126.0 137.2 126.4 O(1)–Ln–O(2)/8 65.1(5) 94.1(4) 91.2(2) 92.9(4) 61.5(3) 92.2(1) 86.0(3) Ref. This work This work This work 36459873370 J. Chem. Soc., Dalton Trans., 1998, 3367–3372 Conclusions The preparation and reactivity of “Lewis base-free” cationic lanthanide metallocene complexes of the type [Cp02Ln][CB11- Br6H6] and [Cp02Ln][BPh4] have been described for the first time.They can be synthesized by the reaction of unsolvated Cp02Ln(II) or [Cp02LnI]2 with silver(I) salts of the weakly coordinating anions in pure toluene. Their chemical reactivities are highly dependent upon the coordinating nature of the counter ions. The more weakly coordinating carborane anion CB11Br6H6 2 can greatly enhance the reactivity (electrophilicity) of the cation Cp02Ln1.For example, [Cp02Ln][CB11Br6H6] can initiate the ring-opening polymerization of THF, abstract one bromine atom from the counter ion CB11Br6H6 2 or one chlorine atom from the solvent CH2Cl2. Experimental General procedures All experiments were performed under an atmosphere of dry dinitrogen with the rigid exclusion of air and moisture using standard Schlenk or cannula techniques, or in a glovebox. All organic solvents were freshly distilled from sodium benzophenone or CaH2 immediately prior to use. Cp02Sm,20 Cp02Yb,21 Ag(BPh4),22 and Ag(CB11Br6H6) 23 were prepared according to the literature methods.[Cp02LnI]2 (Ln = Sm, Er) can be conveniently prepared from Cp02LnI(THF) 19 by sublimation at 200–230 8C at 1022 Torr. All other chemicals were purchased from Aldrich Chemical Company and used as received unless otherwise noted. Infrared spectra were obtained from a KBr pellet, prepared in the glovebox, on a Nicolet Magna 550 Fourier-transform spectrometer.MS spectra were recorded on a Bruker APEX FTMS spectrometer. 1H and 13C NMR spectra were recorded on a Bruker 300 MHz DPX spectrometer at 300.13 and 75.47 MHz, respectively. 11B NMR spectra were recorded on a Bruker ARX-500 spectrometer at 160.46 MHz. All chemical shifts are reported in d units with reference to the residual protons of deuteriated solvent or external SiMe4 (0.00 ppm) for proton and carbon chemical shifts, to external BF3?OEt2 (0.00 ppm) for boron chemical shifts.Complexometric metal analyses were conducted by titration with EDTA. Preparations [Cp02Sm][BPh4] 1. To a mixture of Cp02Sm (0.207 g, 0.36 mmol) and Ag(BPh4) (0.202 g, 0.47 mmol) was added toluene (30 cm3) with stirring at 0 8C. The reaction mixture was allowed to warm to room temperature and stirred for 24 h. A black precipitate formed during this period, and then was filtered oV. Removal of most of the solvent under vacuum gave a yellow solid which was washed with toluene and then hexane to aVord 1 as a yellow powder (0.207 g, 64%) (Found: C, 61.81; H, 7.18; Sm, 16.86.C46H62BSi4Sm requires C, 62.18; H, 7.03; Sm, 16.92%); IR (KBr, n/cm21): 3086w, 3052w, 2954s, 1252s, 833vs. Complex 1 does not redissolve in toluene once isolated, so the NMR data are not available. Recrystallization of 1 from hot toluene (80 8C) aVorded orange crystals identified as Cp03Sm (50% based on 1) by MS and 1H NMR spectroscopy as well as the unit cell measurements.12,20 [Cp02Sm(DME)][BPh4] 5.Recrystallization of 1 (0.20 g, 0.22 mmol) from DME/toluene (10: 1, 20 cm3) at room temperature gave complex 5 (0.18 g, 84%) as yellow crystals over a period of 2 d (Found: C, 61.12; H, 7.35. C50H72BO2Si4Sm requires C, 61.36; H, 7.42%); dH (C5D5N): 20.42 (s, Me3Si), 2.84 (br, DME), 3.06 (br, DME), 6.72 (m, C6H5), 6.90 (m, C6H5), 7.70 (br, C5H3), 9.61 (br, C5H3); dC (C5D5N): 20.87 (Me3Si), 58.14, 71.58 (DME), 121.9, 125.8, 136.8 (C5H3), 164.9, 164.2, 163.6, 162.9 (BPh4 2); dB (C5D5N): 214.6; IR (KBr, n/cm21): 3088w, 2965s, 1437m, 1255s, 1079s, 922m, 833vs.[Cp02Yb][BPh4] 2. To a mixture of Cp02Yb (0.157 g, 0.26 mmol) and Ag(BPh4) (0.131 g, 0.30 mmol) was added toluene (30 cm3) at 0 8C under stirring. The reaction mixture was allowed to warm to room temperature and stirred for 24 h. A black precipitate formed during this period, and then was filtered oV. Removal of most of the solvent under vacuum gave a red solid which was washed with toluene and then hexane to aVord 2 as a red solid (0.18 g, 74%) (Found: C, 60.12; H, 6.75; Yb, 19.25.C46H62BSi4Yb requires C, 60.60; H, 6.86; Yb, 18.99%); IR (KBr, n/cm21): 3089w, 2954s, 2899m, 1579m, 1479m, 1251s, 912s, 831vs. [Cp02Yb(THF)2][BPh4]?2THF 6?2THF. Recrystallization of 2 (0.15 g, 0.16 mmol) from THF/toluene (10 : 1, 20 cm3) at room temperature gave complex 6?2THF (0.16 g, 83%) as red crystals over a period of 2 d (Found: C, 62.33; H, 7.76.C62H94BO4Si4Yb requires C, 62.08; H, 7.90%); dH (C5D5N): 0.45 (s, Me3Si), 1.27 (m, THF), 3.34 (m, THF), 6.96 (m, C6H5), 7.23 (m, C6H5), 8.44 (br, C5H3), 8.64 (br, C5H3); dC (C5D5N): 2.49 (Me3Si), 27.25, 69.30 (THF); 124.6, 128.4, 139.5 (C5H3), 165.5, 165.0, 164.2, 163.9 (BPh4 2); dB (C5D5N): 26.90; IR (KBr, n/cm21): 3050m, 3032m, 2955s, 2898m, 1578m, 1478m, 1448m, 1250s, 1080s, 921s, 831vs, 731s. [Cp02Sm][CB11Br6H6] 3. To a suspension of Ag(CB11Br6H6) (0.193 g, 0.27 mmol) in 15 cm3 of toluene was added a toluene solution (15 cm3) of Cp02Sm (0.146 g, 0.26 mmol) with stirring at 0 8C.The reaction mixture was then allowed to warm to room temperature and stirred for 2 d, followed by the procedure similar to that used in the synthesis of 1, yielding 3 as a yellow solid (0.155 g, 51%) (Found: C, 22.87; H, 3.95; Sm, 12.55. C23H48B11Br6Si4Sm requires C, 23.30; H, 4.08; Sm, 12.68%); IR (KBr, n/cm21): 3057w, 2957m, 2603s (br), 1251m, 835s. Compound 3 can also be prepared by the reaction of [Cp02SmI]2 with 2 equivalents of Ag(CB11Br6H6) in toluene in 40% yield.Compound 3 does not redissolve in pure toluene or fluorobenzene once isolated, so that NMR data are not available. If this suspension is heated for 2 h at 45 8C, a small amount of orange crystals can be isolated and identified as Cp03Sm by MS and 1H NMR spectroscopy.12,20 Complex 3 (23 mg) was dissolved in 5 cm3 of THF with stirring at room temperature. The clear yellow solution slowly became sticky and finally became a gum after 12 h.An additional 5 cm3 of THF was added to the gum giving a very viscous solution which became a gum again after 10 h. This gum was identified as poly(tetrahydrofuran) by 1H and 13C NMR spectroscopy.24 [Cp02Sm(THF)2][CB11Br6H6] 7. Recrystallization of 3 (0.15 g) from toluene containing 5% THF (35 cm3) at room temperature gave complex 7 as yellow crystals (0.10 g, 60%) (Found: C, 28.61; H, 4.67. C31H64B11Br6O2Si4Sm requires C, 28.00; H, 4.85%); dH (C5D5N): 20.08 (s, Me3Si), 1.52 (m, THF), 3.56 (m, THF), 3.40 (s, CH of carborane), 8.20 (br, C5H3), 10.12 (br, C5H3); dC (C5D5N): 21.24 (Me3Si), 25.29, 67.31 (THF), 118.6, 128.1, 135.9 (C5H3); dB (C5D5N): 4.8 (s, 1 B), 23.6 (s, 5 B), 215.2 (d, 5 B); IR (KBr, n/cm21): 3057w, 2958m, 2896m, 2603s (br), 1450m, 1385vs, 1250m, 1077s, 953m, 834s.Reaction of Cp02Yb with Ag(CB11Br6H6). A suspension of Cp02Yb (0.320 g, 0.54 mmol) and Ag(CB11Br6H6) (0.392 g, 0.54 mmol) was mixed in toluene (40 cm3) under stirring at 0 8C.The reaction mixture was then allowed to warm to room temperature and stirred for 2 d. The black precipitate was filtered oV. Slow evaporation of the solvent gave red crystals (0.04 g, 11% based on Cp02Yb) identified as [Cp02YbBr]2. No pure [Cp02Yb][CB11Br6H6] was isolated. Repeating the above experiment at 40 8C resulted in the isolation of red crystals identified as [Cp02YbBr]2 (0.15 g, 41% based on Cp02Yb) (Found: C,J. Chem.Soc., Dalton Trans., 1998, 3367–3372 3371 39.33; H, 6.76. C44H84Br2Si8Yb2 requires C, 39.33; H, 6.30%); IR (KBr, n/cm21): 3067w, 2954s, 2898m, 1438w, 1249s, 1085s, 835vs, 753s; MS (EI, m/z): 672 (��� M1, 4%), 592 ([Cp02Yb]1, 30%), 383 ([Cp0Yb]1, 5%). If the above reaction was conducted in CH2Cl2, [Cp02YbCl]2 was isolated as red crystals in 70% yield (Found: C, 41.95; H, 6.65%. Calc. for C44H84Cl2Si8Yb2: C, 42.11; H, 6.75%) which was also identified by MS.13 If the reaction was carried out in THF, poly(tetrahydrofuran) was obtained. [Cp02Er][CB11Br6H6] 4.Ag(CB11Br6H6) (0.147 g, 0.20 mmol) and [Cp02ErI]2 (0.148 g, 0.10 mmol) were mixed in toluene (30 cm3) under stirring at 0 8C. The reaction mixture was then allowed to warm to room temperature and stirred for 2 d, the yellow precipitate was filtered oV. Concentration of the clear filtrate gave a pink solid which was washed with toluene and then hexane to yield 4 as a pink solid (0.145 g, 60%) (Found: C, 22.65; H, 4.00; Er, 13.78.C23H48B11Br6ErSi4 requires C, 22.97; H, 4.02; Er, 13.91%); IR (KBr, n/cm21): 2955s, 2900m, 2603s (br), 1251s, 835vs. Complex 4 does not redissolve in toluene. Recrystallization of 4 (0.10 g) from hot toluene (50 cm3, 60 8C) gave pink crystals identified as [Cp02ErBr]2 (Found: C, 40.01; H, 6.65. C44H84- Br2Er2Si8 requires C, 39.67; H, 6.36%); IR (KBm21): 3054w, 2955s, 2897m, 1439w, 1249s, 1078s, 834vs, 753s.Recrystallization of 4 from CH2Cl2 at room temperature resulted in the isolation of [Cp02ErCl]2 13,18a,b in 65% yield (Found: C, 42.08; H, 6.75. Calc. for C44H84Cl2Er2Si8: C, 42.50; H, 6.82%). [Cp02Er(THF)2][CB11Br6H6] 8. Recrystallization of 4 (0.13 g) from toluene containing 5% THF (35 cm3) at room temperature gave compound 8 as pink crystals (0.06 g) (Found: C, 27.35; H, 4.80. C31H64B11Br6ErO2Si4 requires C, 27.65; H, 4.79%); dB (C5D5N): 10.6 (s, 1 B), 2.85 (s, 5 B), 28.23 (d, 5 B); IR (KBr, n/cm21): 3066w, 2955s, 2900m, 2603s (br), 1436m, 1251s, 1078s, 1001s, 835vs, 799s.The 1H and 13C NMR spectra consisted of many broad, unresolved resonances. Crystallography A summary of crystal data and details of data collection and structure refinement for compounds 5?0.25C7H8, 7?OC4H8, 8?1.25C7H8 and [Cp02ErBr]2 is given in Table 6. Due to the facile loss of the solvent molecules in the crystal lattices of compounds 5?0.25C7H8, 7?OC4H8 and 8?1.25C7H8, these crystals were sealed with a drop of mother-liquor under N2 in a thinwalled glass capillary.Data were collected at 294 K on a MSC/ Rigaku RAXIS-IIC imaging plate using Mo-Ka radiation (0.71073 Å) from a Rigaku rotating-anode X-ray generator operating at 50 kV and 90 mA. An absorption correction was applied by correlation of symmetry-equivalent reflections using the ABSCOR program.25 All structures were solved by direct methods and subsequent Fourier-diVerence techniques, and refined anisotropically for all non-hydrogen atoms by fullmatrix least squares, on F2 for 7?OC4H8 and 8?1.25C7H8 using the Siemens SHELXTL V 5.03 program package (PC version), 26a and on F for 5?0.25C7H8 and [Cp02ErBr]2 using the Siemens SHELXTL V 4.1 program package (PC version).26b The hydrogen atoms were geometrically fixed using the riding model.One of the Me3Si groups (Si1) in 5?0.25C7H8 is disordered over two sets of positions with 0.60 : 0.40 occupancies. The toluene molecule in the lattice of 5?0.25C7H8 is also disordered over two sets of positions with 0.50 : 0.50 occupancies.One toluene molecule in the lattice of 8?1.25C7H8 is disordered over two sets of positions with 0.50 : 0.50 occupancies. The other toluene molecule in the lattice of 8?1.25C7H8 is highly disordered, so that only the hexagonal ring could be found. CCDC reference number 186/1116. Acknowledgements We thank the Hong Kong Research Grants Council (Earmarked Grant CUHK 306/96P) for financial support.References 1 For recent reviews, see: M. Bochmann, J. Chem. Soc., Dalton Trans., 1996, 225; W. Kaminsky, Macromol. Chem. Phys., 1996, 197, 3907; H. H. Brintzinger, D. Fisher, R. Mülhaupt, B. Rieger and R. M. Waymouth, Angew. Chem., Int. Ed. Engl., 1995, 34, 1143; A. S. Guram and R. F. Jordan, Comprehensive Organometallic Chemistry II, eds. E. W. Abel, F. G. A. Stone and G. Wilkinson, Pergamon, Oxford, 1995, vol. 4, p. 589; R. C. Möhring and N.J. Coville, J. Organomet. Chem., 1994, 479, 1; T. J. Marks, Acc. Chem. Res., 1992, 25, 57. 2 For recent reviews, see: H. Schumann, J. A. Meese-MarktscheVel and L. Esser, Chem. Rev., 1995, 95, 865; F. T. Edelmann, Comprehensive Organometallic Chemistry II, eds. E. W. Abel, F. G. A. Stone and G. Wilkinson, Pergamon, Oxford, 1995, vol. 4, p. 11. Table 6 Crystal data and summary of data collection and refinement for 5?0.25C7H8, 7?OC4H8, 8?1.25C7H8 and [Cp02ErBr]2 Formula Crystal size/mm M Crystal class Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 Z Dc/Mg m23 Radiation (l/Å) 2q Range/8 m/mm21 F(000) T/K No.observed reflections [I > Xs(I)] No. parameters refined Goodness of fit R1 wR2 5?0.25C7H8 C51.75H74BO2Si4Sm 0.12 × 0.15 × 0.40 1001.63 Monoclinic P21/n 13.281(1) 22.413(1) 19.449(1) 90.00 95.14(1) 90.00 5766(3) 4 1.154 Mo-Ka (0.71073) 3.0–55.0 1.134 2094 294 3540 (X = 3) 537 1.53 0.067 0.190 7?OC4H8 C35H72B11Br6O3Si4Sm 0.30 × 0.10 × 0.30 1402.01 Monoclinic P21/n 13.487(3) 33.769(7) 14.365(3) 90.00 99.34(3) 90.00 6456(2) 4 1.443 Mo-Ka (0.71073) 3.0–55.0 4.725 2756 294 8494 (X = 2) 573 0.89 0.061 0.168 8?1.25C7H8 C39.5H69.5B11Br6O2Si4Er 0.14 × 0.12 × 0.14 1454.44 Monoclinic P21/n 13.444(3) 33.625(7) 14.327(3) 90.00 99.17(3) 90.00 6394(2) 4 1.511 Mo-Ka (0.71073) 3.0–55.0 5.166 2864 294 5665 (X = 2) 564 0.98 0.069 0.171 [Cp02ErBr]2 C44H84Br2Si8Er2 0.12 × 0.25 × 0.40 1332.20 Triclinic P1� 10.686(2) 11.663(2) 13.233(2) 73.24(2) 83.89(2) 76.62(2) 1534.9(8) 1 1.411 Mo-Ka (0.71073) 3.0–55.0 4.198 666 294 5509 (X = 3) 256 2.65 0.037 0.1523372 J.Chem. Soc., Dalton Trans., 1998, 3367–3372 3 W. J. Evans, T. A. Ulibarri, L. R. Chamberlain, J. W. Ziller and D. Alvarez, Jr., Organometallics, 1990, 9, 2124. 4 P. N. Hazin, J. W. Bruno and G. K. Schulte, Organometallics, 1990, 9, 416. 5 H. J. Heeres, A. Meetsma and J. H. Teuben, J. Organomet. Chem., 1991, 414, 351. 6 W. J. Evans, G. Kociok-Köhn and J.W. Ziller, Angew. Chem., Int. Ed. Engl., 1992, 31, 1081. 7 F. Yuan, Q. Shen and J. Sun, J. Organomet. Chem., 1997, 538, 241. 8 H. Schumann, J. Winterfeld, M. R. Keitsch, K. Herrmann and J. Z. Demtschuk, Z. Anorg. Allg. Chem., 1996, 622, 1457. 9 D. Deng, X. Zheng, C. Qian, J. Sun, A. Dormond, D. Baudry and M. Visseaux, J. Chem. Soc., Dalton Trans., 1994, 1665. 10 L. Jia, X. Yang, C. L. Stern and T. J. Marks, Organometallics, 1997, 16, 842 and refs. therein; X. Yang, C. L. Stern and T.J. Marks, J. Am. Chem. Soc., 1994, 116, 10015; P. A. Deck and T. J. Marks, J. Am. Chem. Soc., 1995, 117, 6128; Z. Wu, R. F. Jordan and J. L. Petersen, J. Am. Chem. Soc., 1995, 117, 5867; M. Bochmann, S. J. Lancaster and O. B. Robinson, J. Chem. Soc., Chem. Commun., 1995, 2081; G. Erker, W. Ahlers and R. Frohlich, J. Am. Chem. Soc., 1995, 117, 5853. 11 X. Song, M. Thornton-Pett and M. Bochmann, Organometallics, 1998, 17, 1004. 12 (a) Z. Xie, Z. Liu, F. Xue and T. C.W. Mak, J. Organomet. Chem., 1997, 539, 127; (b) Z. Xie, K. Chui, Q. Yang, T. C. W. Mak and J. Sun, Organometallics, 1998, 17, 3937. 13 Z. Xie, K. Chui, Z. Liu, F. Xue, Z. Zhang, T. C. W. Mak and J. Sun, J. Organomet. Chem., 1997, 549, 239. 14 Phenyl transfer was observed in the isolation of [Cp2ZrMe][BPh4], see: A. D. Horton and J. H. G. Frijns, Angew. Chem., Int. Ed. Engl., 1991, 30, 1152. 15 Z. Xie, D. L. Liston, T. Jelinek, V. Mitro, R. Bau and C. A. Reed, J. Chem. Soc., Chem. Commun., 1993, 384; C. A. Reed, Z. Xie, R. Bau and A. Benesi, Science, 1993, 262, 402; Z. Xie, R. Bau, A. Benesi and C. A. Reed, Organometallics, 1995, 14, 3933; Z. Xie, J. Manning, R. W. Reed, R. Mathur, P. D. W. Boyd, A. Benesi and C. A. Reed, J. Am. Chem. Soc., 1996, 118, 2922. 16 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1976. 17 R. D. Shannon, Acta Crystallogr., Sect. A, 1976, 32, 751. 18 (a) M. F. Lappert, A. Singh, J. L. Atwood and W. E. Hunter, J. Chem. Soc., Chem. Commun., 1981, 1190; (b) M. F. Lappert and A. Singh, Inorg. Synth., 1990, 27, 168; (c) I. P. Beletskaya, A. Z. Voskoboynikov, E. B. Chuklanova, N. I. Kirillova, A. K. Shestakova, I. N. Parshina, A. I. Gusev and G. K. I. Magomedov, J. Am. Chem. Soc., 1993, 115, 3156; (d ) H. Lueken, W. Lamberts and P. Hannibal, Inorg. Chim. Acta, 1987, 132, 111. 19 Z. Xie, Z. Liu, F. Xue, Z. Zhang and T. C. W. Mak, J. Organomet. Chem., 1997, 542, 285. 20 W. J. Evans, R. A. Keyer and J. W. Ziller, J. Organomet. Chem., 1990, 394, 87. 21 P. B. Hitchcock, J. A. K. Howard, M. F. Lappert and S. Prashar, J. Organomet. Chem., 1992, 437, 177. 22 R. F. Jordan and S. F. Echols, Inorg. Chem., 1987, 26, 383. 23 Z. Xie, T. Jelinek, R. Bau and C. A. Reed, J. Am. Chem. Soc., 1994, 116, 1907. 24 S. L. Borkowsky, R. F. Jordan and G. D. Hinch, Organometallics, 1991, 10, 1268; S. J. Hrkach and K. Matyjaszewski, Macromolecules, 1990, 23, 4042; M. E. Woodhouse, F. D. Lewis and T. J. Marks, J. Am. Chem. Soc., 1982, 104, 5586. 25 T. Higashi, ABSCOR, An Empirical Absorption Correction Based on Fourier CoeYcient Fitting, Rigaku Corporation, Tokyo, 1995. 26 (a) SHELXTL V 5.03 program package, Siemens Analytical X-ray Instruments, Inc., Madison, WI, 1995; (b) SHELXTL V 4.program package, Siemens Analytical X-ray Instruments, Inc., Madison, WI, 1990. Paper 8/04311F

ISSN:1477-9226    

DOI:10.1039/a804311f

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3373–3378 3373 Oxidation of the zwitterionic tungsten amide (OC)5WNPhNPhC(OMe)Ph with Br2 and PCl5. Formation of [(PhN)W(CO)2X2]2 and (PhN)W(CO)2X2L (X 5 Br or Cl; L 5 MeCN, Me3C6H2NH2 and i-BuNH2) Yingxia He, Patrick C. McGowan, Khalil A. Abboud and Lisa McElwee-White * Department of Chemistry, University of Florida, Gainesville, FL 32611-7200, USA Received 27th May 1998, Accepted 30th July 1998 Oxidation of the zwitterionic tungsten amide (OC)5WNPhNPhC(OMe)Ph 1 with Br2 or PCl5 produced [(PhN)W(CO)2X2]2 (X = Cl or Br).These halide-bridged tungsten(IV) dimers reacted with MeCN to give the mononuclear complexes (PhN)W(CO)2X2(MeCN) (X = Cl or Br). The latter complexes can also be obtained directly in high yield by oxidizing 1 in the presence of MeCN. The acetonitrile ligand can be substituted with Me3C6H2NH2 or i-BuNH2 to yield (PhN)W(CO)2X2L (X = Cl or Br, L = Me3C6H2NH2 or i-BuNH2) or with PMe3 to produce (PhN)W(CO)Br2(PMe3)2.The oxidation of 1 with BrI resulted in the formation of [(PhN)W(CO)2Br(I)]2, which reacts with MeCN to produce (PhN)W(CO)2Br(I)(MeCN). The structures of the dimers [(PhN)W(CO)2X2]2 were determined by X-ray crystallography. Introduction Synthetic routes involving amido ligands have become important in the preparation of imido complexes.1 Although imido moieties can be generated in a variety of ways,1,2 utilization of a preexisting metal–nitrogen bond is often very convenient.This strategy has proven to be particularly successful in the preparation of low-valent imido complexes, which can be diYcult targets. As an example, abstraction of either a hydride or a proton has been used to generate low-valent imido complexes of the type Tp9(OC)2W]] NR1 [Tp9 = hydridotris(3,5-dimethylpyrazolyl) borate] from the corresponding amido complexes.3 We have previously reported use of the zwitterionic amido complex (OC)5WNPhNPhC(OMe)Ph 1 4 as a precursor to tungsten imido complexes in a variety of oxidation states.Cleavage of 1 under non-oxidative conditions generates the tungsten(0) complex (OC)5W]] NPh as a reactive intermediate,5 while oxidation of 1 with I2 produces imido complexes of WIV or WVI.6 Of particular interest is the dimeric complex [(PhN)- W(CO)2I2]2 2 [eqn. (1)] which results from reaction of 1 with 1 equivalent of I2. Iodo-bridged dimer 2 is both a stoichiometric reagent 7a and a catalyst 7b for the carbonylation of primary amines to ureas.In addition, complexes of the type (PhN)- W(CO)2I2L, which can be produced by reaction of 2 with twoelectron donor ligands L, show promise as precursors for the chemical vapor deposition of tungsten nitride thin films.8 The exploration of these applications led us to prepare derivatives containing chloride and bromide ligands. We now report the oxidation of zwitterion 1 with Br2 or PCl5 to yield the halide-bridged dimers [(PhN)W(CO)2X2]2 (X = Cl 3 or Br 4).In parallel with the chemistry of iodide complex 2, derivatives of the type (PhN)W(CO)2X2L can be prepared from dimers 3 and 4. In addition, when the oxidation of 1 is carried out in the presence of MeCN, the acetonitrile complexes (PhN)W(CO)2X2(MeCN) (X = Cl 5 or Br 6) can be obtained (OC)5W N N Ph OMe Ph Ph W I W I N N OC CO CO OC I I Ph Ph Ph OMe N Ph 2 – 3 CO 1 1 equivalent I2 + (1) 1–2 directly in more than 80% yield. The MeCN ligand can easily be substituted with amines to provide the series of complexes (PhN)W(CO)2X2L.Results and discussion Oxidation of (OC)5WNPhNPhC(OMe)Ph 1 with PCl5 Reaction of complex 1 with PCl5 in diethyl ether at 0 8C resulted in precipitation of a red solid after about 25 min. Isolation of the solid by filtration aVorded the Cl-bridged dimer 3 in 63% yield (Scheme 1). When the reaction mixture was allowed to stir for more than 3 h dimer 3 began to decompose into an insoluble blue-black solid which was not characterized.Although chlorination with PCl5 was successful, reaction of 1 with Cl2 did not result in formation of 3. It is possible that the limited solubility of both PCl5 and dimer 3 in ether at 0 8C allowed 3 to form and precipitate before further chlorination could occur. Spectroscopic data for compound 3 can be found in Tables 1–3. The carbonyl bands (2084, 2016 cm21) appear at higher wavenumbers than those of its iodide analogue 2, consistent Scheme 1 (i) PCl5, ether, 0 8C; (ii) Br2, CH2Cl2, ether, 278 8C; (iii) PCl5, MeCN, ether, 0 8C; (iv) Br2, MeCN, ether, 0 8C.(OC)5W N N Ph OMe Ph Ph W X W X N N OC CO CO OC X X Ph Ph W X NCMe N OC OC X Ph W X L N OC OC X Ph L 789 10 X = Cl; X = Cl; X = Br; X = Br; L = Me3C6H2NH2 L = i-C4H9NH2 L = Me3C6H2NH2 L = i-C4H9NH2 3 X = Cl; 4 X = Br 5 X = Cl; 6 X = Br 1 (i) or (ii) MeCN (iii) or (iv)3374 J. Chem. Soc., Dalton Trans., 1998, 3373–3378 with the poorer electron donation from the chloride ligands vs.iodide. This eVect is also evident in the 13C NMR signals for the carbonyl carbons in dimers 2 and 3, where the chloride dimer 3 appears downfield of iodide 2. Oxidation of (OC)5WNPhNPhC(OMe)Ph 1 with Br2 and PBr5 In contrast to the well behaved reactions of compound 1 with PCl5 and I2, preparation of the analogous bromide dimer 4 proved problematic. When 1 was treated with 1 equivalent of Br2–CH2Cl2 solution at 278 8C in ether containing a small amount of CH2Cl2 no reaction occurred initially.However, upon warming to 240 8C, the solution changed from black to dark red and a dark red sticky solid formed over the course of 2 h. The sticky solid contained dimer 4 and unstable material, which decomposed during the work-up. Purification of the crude product aVorded clean 4 in 13% yield (Scheme 1). The spectroscopic data were similar to those for chloride dimer 3 and are summarized in Tables 1–3. EVorts to optimize the conditions using other solvents (CH2Cl2 or hexane) and altered reaction conditions were unsuccessful.Given that the presence of an excess of I2 during oxidation of compound 1 to 2 leads to overoxidation and formation of the tungsten(VI) metallacycle I3(PhN)W(NPhCPhO), it seemed likely that the more strongly oxidizing Br2 could also be causing overoxidation. In order to ensure low concentrations of Br2, PBr5 was used as a Br2 equivalent. Reaction of 1 with PBr5 in ether at 0 8C over 2 h aVorded a red solution and a dark red sticky solid. Dimer 4 was isolated from the sticky solid in 25% yield.Once again the timing of the reaction was critical. Upon Table 1 Infrared and 13C NMR data for carbonyl ligands. Complex 2 [(PhN)W(CO)2I2]2 a 4 [(PhN)W(CO)2Br2]2 3 [(PhN)W(CO)2Cl2]2 [(PhN)W(CO)2Br(I)]2 (PhN)W(CO)2I2(MeCN)a 6 (PhN)W(CO)2Br2(MeCN) 5 (PhN)W(CO)2Cl2(MeCN) (PhN)W(CO)2Br(I)(MeCN) (PhN)W(CO)2I2(Me3C6H2NH2) a 9 (PhN)W(CO)2Br2(Me3C6H2NH2) 7 (PhN)W(CO)2Cl2(Me3C6H2NH2) 10 (PhN)W(CO)2Br2(i-C4H9NH2) 8 (PhN)W(CO)2Cl2(i-C4H9NH2) (PhN)W(CO)Br2(PMe3)2 n& (CO) (CH2Cl2)/cm21 2068, 2007 2080, 2012 2084, 2016 2080, 2017 2072, 2003 2078, 2005 2081, 2013 2087, 2011 2064, 1984 2076, 1995 2087, 2011 2076, 1988 2078, 1993 1953 13C NMR, d(CDCl3, 20 8C) 202.7 208.3 b 210.4 b 207.7 206.4, 203.0 208.9, 207.1 212.3, 210.4 b 207.0, 204.2 209.1, 207.2 211.9, 211.6 213.7, 212.4 213.8, 211.3 b 216.0, 211.6 241.9 b a Data taken from ref. 6(b). b Spectrum obtained in CD2Cl2.stirring the reaction mixture for 4 h dimer 4 decomposed into intractable material. Further attempts to optimize the reaction conditions using PBr5 were also unsuccessful. Oxidation of (OC)5WNPhNPhC(OMe)Ph 1 with BrI Upon addition of BrI–CH2Cl2 solution to an ether solution of compound 1 at room temperature immediate reaction occurred. The solution changed from black to greenish, then dark red, and red solid began to precipitate within 10 min. After 2 h of stirring a red solution containing an orange-red solid had formed.Isolation of the solid yielded the dimer [(PhN)- W(CO)2Br(I)]2 as the major product. As expected,9 formation of this mixed bromide–iodide complex predominated. However, the reaction mixtures also contained trace amounts of the iodide dimer 2 and the bromide dimer 4, which presumably arise due to the equilibrium between BrI and Br2 1 I2.10 The three dimers exhibited clear diVerences in their solubilities, with 2 being the most soluble and 4 the least.The insolubility of 4 allowed crystals to be grown from solutions in which [(PhN)W(CO)2Br(I)]2 was the major component. Crystal structures of dimers 3 and 4 The chloride- and bromide-containing complexes [(PhN)W- (CO)2X2]2 and (PhN)W(CO)2X2L are rare examples of d2 imido complexes bearing two p-acid ligands 3,6 and as such could provide insight into the problem of accommodating both strongly p-donating imido and p-acid carbonyl ligands on the same metal center.Prior studies on d2 imido complexes bearing a single p-acid ligand have been interpreted in terms of electron donation from the imido ligand into two empty d orbitals in conjunction with back donation of the two d electrons into the empty orbitals of the p acid.11 For compounds with multiple p acids, such as the [(PhN)W(CO)2X2]2 and (PhN)W(CO)2X2L series, back donation will be aVected by competition for the two d electrons. In the chloride and bromide derivatives, which will be less electron rich at the metal than iodide-containing dimer 2, the already weak back bonding should be even less potent.This eVect can be detected in the variation of carbonyl stretching frequencies as the halide is changed (Table 1). In order to see if it would also be detectable in the structures of the complexes, a crystallographic study of dimers 3 and 4 was carried out for comparison to iodide dimer 2, whose structure was published previously.6a The structures of complexes 3 and 4 appear in Figs. 1 and 2, respectively. Selected bond lengths and angles can be found in Table 4. As is also observed for iodide dimer 2, the tungsten centers are octahedral and bear triply bonded imido ligands [W–N1 1.757(4) Å for 3; 1.757(10) for 4]. Within experimental error, the W–C bond lengths are the same for dimers 2 [2.051(14), 2.022(13) Å], 3 [2.043(5), 2.040(5) Å] and 4 [2.051(13), 2.029(13) Å]. The C–O bond lengths within the carbonyls are also indistinguishable in the three dimers.There is Table 2 Proton and 13C NMR data (d, J/Hz) for phenyl groups. Complex 3 * 5 * 7846 9* 10 * (PhN)W(CO)Br2(PMe3)2* [(PhN)W(CO)2Br(I)]2 (PhN)W(CO)2Br(I)(MeCN) 1H NMR, d(CDCl3), C6H5 7.48 (t, 4 H), 7.42 (m, 2 H), 7.08 (t, 4 H) 7.59 (m, 1 H), 7.37 (d, J = 8, 2 H), 7.30 (t, 2 H) 7.27 (m, 1 H), 7.16 (t, 2 H), 6.64 (d, J = 8, 2 H) 7.34 (t, 2 H), 7.29 (m, 1 H), 7.22 (d, J = 8, 2 H) 7.58 (d, J = 8, 4 H), 7.39 (t, 4 H), 7.23 (m, 2 H) 7.42 (m, 1 H), 7.31 (d, J = 7, 2 H), 7.22 (t, 2 H) 7.35 (t, 1 H), 7.18 (t, 2 H), 6.72 (d, J = 8, 2 H) 7.44 (m, 1 H), 7.33 (t, 2 H), 7.28 (t, 2 H) 7.51 (m, 2 H), 7.29 (m, 1 H), 7.14 (t, 2 H) 7.61 (m, 4 H), 7.50 (m, 2 H), 7.37 (m, 4 H) 7.67–7.32 (m, 5 H) 13C NMR, d(CDCl3, 20 8C) 154.2, 129.4, 128.5, 125.9 153.9, 139.0, 129.4, 126.9 152.7, 138.0, 129.7, 126.6 153.4, 129.2, 128.8, 126.3 154.6, 130.2, 129.5, 127.8 153.5, 132.2, 129.2, 126.0 153.1, 138.9, 129.5, 126.3 153.9, 129.7, 129.4, 126.2 154.9, 129.3, 127.3, 124.9 153.8, 129.9, 129.3, 127.1 153.4, 129.1, 126.0, 125.6 * Spectrum obtained in CD2Cl2.J.Chem. Soc., Dalton Trans., 1998, 3373–3378 3375 Table 3 Proton and 13C NMR data (d, J/Hz) for ancillary ligands. d(CDCl3, 20 8C) Complex 5 a 7 8 69 10 a a,b c Ligand MeCN NH2C6H2Me3 MeCHCH2Me | NH2 MeCN NH2C6H2Me3 MeCHCH2Me | NH2 PMe3 MeCN 1H NMR 2.68 (s, 3 H, CH3CN) 6.81 (s, 2 H, C6H2) 6.30 (d, J = 12, 1 H, NH2) 5.57 (d, J = 12, 1 H, NH2) 2.42 (s, 6 H, CH3) 2.28 (s, 3 H, CH3) 4.23 (dd, 1 H, NH2) 3.77 (dd, 1 H, NH2) 3.33 (m, 1 H, CH) 1.64 (m, 2 H, CH2) 1.33 (t, 3 H, CH3) 0.99 (t, 3 H, CH3) 2.75 (s, 3 H, CH3CN) 6.85 (s, 2 H, C6H2) 6.32 (d, J = 11, 1 H, NH2) 5.55 (d, J = 12, 1 H, NH2) 2.42 (s, 6 H, CH3C6H2) 2.26 (s, 3 H, CH3C6H2) a 4.27 (dd, 1 H, NH2) 3.88 (dd, 1 H, NH2) 3.38 (m, 1 H, CH) 1.64 (m, 2 H, CH2) 1.32 (t, 3 H, CH3) 1.00 (t, 3 H, CH3) 1.74 (t, 18 H, PMe3) 2.78 (s, 3 H, CH3CN) 13C NMR 129.2, (CH3CN), 4.7 (CH3CN) 135.1, 128.8, 128.6, 127.1 (C6H2), 20.5, 18.0 (CH3C6H2) 56.5 (H2NCH) 31.7 (H2NCHCH3) 21.5 (H2NCHCH2) 10.0 (CHCH2CH3) 125.3 (CH3CN), 5.1 (CH3CN) 135.7, 130.1, 129.9, 129.2 (C6H2), 20.6, 18.5 (CH3C6H2) 57.8 (H2NCH) 32.1 (H2NCHCH3) 21.7 (H2NCHCH2) 10.2 (CHCH2CH3) 16.4 [t, P(CH3)3] 124.9 (CH3CN), 4.9 (CH3CN) a Spectrum obtained in CD2Cl2.b Complex (PhN)W(CO)Br2(PMe3)2: 31P NMR (CD2Cl2) d 226.0 (t, JWP = 285.0 Hz, PMe3). c (PhN)W(CO)2- Br(I)(MeCN). thus no structural evidence for variation in the back bonding to the carbonyl ligands as the halide ligands are changed.The W– X bond distances in 3 [terminal 2.4349(12); bridging 2.4991(12) Å] and 4 [terminal 2.6729(13); bridging 2.7204(11) Å] fall within the normal ranges.12 The structural diVerences in the W2X2 cores of the dimers can be attributed to the variation in W–X bond lengths as X is varied. Reactions of dimers 3, 4 and [(PhN)W(CO)2Br(I)]2 with donor ligands When dimers 3 and 4 are dissolved in MeCN dissociation and solvent co-ordination result in formation of the acetonitrile complexes (PhN)W(CO)2Cl2(MeCN) 5 and (PhN)W(CO)2Br2- Fig. 1 Thermal ellipsoid diagram of [(PhN)W(CO)2Cl2]2 3 showing the crystallographic numbering scheme. (MeCN) 6, respectively. Similar behavior has been reported for iodide dimer 2.6 Reaction of the mixed bromide–iodide dimer [(PhN)W(CO)2Br(I)]2 with MeCN overnight yielded only the mixed halide complex (PhN)W(CO)2Br(I)(Me3CN). Neither (PhN)W(CO)2I2(MeCN) nor (PhN)W(CO)2Br2(MeCN) 6 were produced from [(PhN)W(CO)2Br(I)]2. This result suggests that the bridging halides in the latter are either both iodide or both bromide.We have no experimental evidence to distinguish between the possibilities, as X-ray quality crystals could not be obtained. However, the precedent from molecules such as [Mo(m-I)Br(CO)3(PPh3)]2 9 suggests that iodide bridging would be preferred. Reaction of bromide dimer 4 with an excess of PMe3 in CH2Cl2 results in formation of the bis(PMe3) complex (PhN)- Fig. 2 Thermal ellipsoid diagram [(PhN)W(CO)2Br2]2 4 showing the crystallographic numbering scheme.3376 J.Chem. Soc., Dalton Trans., 1998, 3373–3378 W(CO)Br2(PMe3)2. As is also observed for the chemistry of iodide dimer 2,6 one of the CO ligands is replaced by the more strongly nucleophilic PMe3. The 31P NMR spectrum exhibits a signal at d 226.0 for PMe3. The signal is shifted downfield with respect to its iodide analogue (PhN)W(CO)I2(PMe3)2 (d 238.8) as expected for the more electron poor bromide complex.Oxidation of (OC)5WNPhNPhC(OMe)Ph 1 with Br2 and PCl5 in the presence of MeCN Although acetonitrile complexes 5 and 6 are accessible via reaction of dimers 3 and 4 with MeCN, poor yields during the dimer syntheses result in low overall yields for 5 and 6. Therefore, more direct routes to them were developed. Oxidation of zwitterion 1 with Br2 in the presence of MeCN at 278 8C resulted in formation of a red solution containing a red precipitate of MeCN complex 6.Simple filtration and washing of the solid aVorded 81% yield. A similar protocol involving oxidation of 1 with PCl5 in the presence of MeCN at 0 8C produced complex 5 in 88% yield. Clearly, the yields of MeCN complexes obtained upon oxidation in the presence of a co-ordinating solvent are much higher than the yields of dimer obtained in non-co-ordinating solvents. This eVect can be attributed to trapping of the unsaturated intermediate [X2(OC)2W(NPh)] before side reactions can compete with dimerization. Substitution of the MeCN ligand in (PhN)W(CO)2X2(MeCN) with Me3C6H2NH2 and i-BuNH2 Since MeCN complexes 5 and 6 were readily available we explored substitution of the acetonitrile ligand as a means of generating derivatives. The hindered arylamine 2,4,6-Me3C6- H2NH2 had been demonstrated to serve as a ligand in the iodide system, where the amine was treated with dimer 2 to generate (PhN)W(CO)2I2(Me3C6H2NH2).6b Treatment of 5 and 6 with 1 equivalent of Me3C6H2NH2 in CH2Cl2 solution results in the Table 4 Selected bond lengths (Å) and angles (8) for complexes 3 and 4.Complex 3 W–N1 W–C7 W–C8 W–Cl2 W–Cl1 W–Cl1A N1–Cl C7–O8 C8–O8 N1–W–C8 N1–W–C7 C7–W–C8 N1–W–Cl2 C8–W–Cl2 C7–W–Cl2 N1–W–Cl1 C8–W–Cl1 C7–W–Cl1 Cl2–W–Cl1 N1–W–Cl1A C8–W–Cl1A C7–W–Cl1A Cl2–W–Cl1A Cl1–W–Cl1A W–Cl1–W0A C1–N1–W N1–C1–C2 N1–C1–C6 O8–C8–W O7–C7–W 1.757(4) 2.043(5) 2.040(5) 2.4349(12) 2.4991(12) 2.5085(12) 1.387(6) 1.131(7) 1.125(7) 89.3(2) 90.0(2) 89.0(2) 170.99(13) 83.8(2) 84.2(2) 96.77(13) 94.4(2) 172.5(2) 89.54(4) 98.65(13) 170.92(14) 95.3(2) 88.72(4) 80.35(4) 99.65(4) 174.9(3) 118.5(4) 120.3(4) 178.4(5) 178.0(5) Complex 4 W–N1 W–C7 W–C8 W–Br2 W–Br1 W–Br1A N1–Cl O7–C7 O8–C8 N1–W–C8 N1–W–C7 C8–W–C7 N1–W–Br2 C8–W–Br2 C7–W–Br2 N1–W–Br1 C8–W–Br1 C7–W–Br1 Br2–W–Br1 N1–W–Br1A C8–W–Br1A C7–W–Br1A Br2–W–Br1A Br1–W–Br1A W–Br1–W0A C1–N1–W N1–C1–C2 N1–C1–C6 O7–C7–W O8–C8–W 1.757(10) 2.051(13) 2.029(13) 2.6729(13) 2.7204(11) 2.7318(11) 1.37(2) 1.12(2) 1.14(2) 91.4(5) 91.3(5) 89.3(5) 170.2(3) 81.4(4) 82.1(4) 96.8(3) 93.2(4) 171.5(4) 90.25(4) 98.3(3) 169.9(4) 93.2(4) 89.26(4) 82.92(3) 97.08(3) 177.1(9) 120.0(12) 119.2(11) 177.0(13) 177.3(13) formation of the Me3C6H2NH2 complexes 7 and 9 (Scheme 1) within 2 h.The IR spectrum of 9 exhibited two CO bands at 2076 and 1995 cm21, while the two CO bands for 7 appeared at 2087 and 2011 cm21.These values are higher than the corresponding 2064 and 1984 cm21 for the iodide derivative, consistent with the weakening of electron donation to the metal upon proceeding from iodide to bromide to chloride. Similarly, reaction of compounds 5 and 6 with 1 equivalent of i-BuNH2 in CH2Cl2 solution yielded the i-BuNH2 complexes 8 and 10. Owing to the chirality of the metal center and the amine a mixture of diastereomers was obtained. This was reflected in the 1H NMR spectra, which exhibited broad multiplets at d 1.64 for the CH2 group next to the chiral carbon.In addition the 13C NMR spectra contained two sets of carbon signals for the isobutyl groups of the diastereomers. The reported data (Tables 1–3) correspond to the major isomers. Conclusion Oxidation of the zwitterionic amide complex (OC)5WNPhNPhC( OMe)Ph 1 with Br2 and PCl5 at low temperature in ether solution results in formation of the dimeric complexes [(PhN)- W(CO)2X2]2 (X = Cl; 3 or Br 4).These dimers are congeners of the previously reported iodo-bridged dimer 2, but are more reactive than it due to the poorer bridging ability of Br and Cl. Reaction of them with two-electron donor ligands L produces complexes of the type (PhN)W(CO)2X2L. When oxidation of 1 with Br2 and PCl5 is conducted in the presence of the co-ordinating solvent MeCN, the mononuclear complexes (PhN)W(CO)2X2(MeCN) (X = Cl 5 or Br 6) are obtained directly in more than 80% yield.The MeCN ligand in complexes 5 and 6 can easily be replaced with amines such as Me3C6H2NH2 and i-BuNH2. Experimental General Standard inert atmosphere techniques were used throughout. Hexane, acetonitrile and methylene chloride were distilled from CaH2, diethyl ether and THF from Na/Ph2CO. The NMR solvents were degassed by three freeze–pump–thaw cycles and stored over 3 Å molecular sieves in a dry-box. Iodine monobromide was purchased from Aldrich as a 1.0 M solution in CH2Cl2 and used as received.All other reagents were purchased in reagent grade and used without further purification. The 1H, 13C and 31P NMR spectra were recorded on Gemini-300 or VXR-300 spectrometers, IR spectra on a Perkin-Elmer 1600 spectrometer. High resolution mass spectrometry was performed by the University of Florida analytical service. Elemental analyses were performed by Robertson Microlit Laboratories, Madison, NJ. Syntheses [(PhN)W(CO)2Cl2]2 3. Zwitterion 1 (1.3 g, 2.1 mmol) and PCl5 (432.5 mg, 2.08 mmol) were placed in a Schlenk vessel and cooled to 0 8C.Diethyl ether (25 mL) and 4 mL of CH2Cl2 were added via syringe. After 25 min a red solid began to form. The solution was stirred at 0 8C for 1.5 h. The solvent was removed by cannulation and the solid washed with ether (5 mL × 3) to yield complex 3 as an orange-red solid. Yield 524 mg, 62.7% (Found: C, 23.62; H, 1.29; N, 3.75. Calc. for C8H5Cl2NO2W: C, 23.91; H, 1.25; N, 3.48%).(PhN)W(CO)2Cl2(MeCN) 5. Zwitterion 1 (1.37 g, 2.19 mmol) and PCl5 (456 mg, 2.19 mmol) were placed in a Schlenk vessel and cooled to 0 8C. Ether (15 mL) was added, then 3 mL of MeCN were added within 5 min. A vigorous reaction occurred upon addition of MeCN, as evidenced by a changeJ. Chem. Soc., Dalton Trans., 1998, 3373–3378 3377 from black to dark red. The reaction mixture was stirred for 1.5 h at 0 8C followed by 0.5 h at room temperature, during which a purple-red solid formed.The solvent was removed by cannulation and the solid washed with ether (5 mL × 3) to yield complex 5 as a purple-red solid. Yield 852 mg, 87.8% (Found: C, 27.28; H, 1.67; N, 6.19. Calc. for C10H8Cl2N2W: C, 27.12; H, 1.82; N, 6.33%). (PhN)W(CO)2Cl2(Me3C6H2NH2) 7. Complex 5 (317 mg, 0.59 mmol) was dissolved in 8 mL of CH2Cl2 and Me3C6H2NH2 (105.5 mL, 0.59 mmol) was added via syringe. The solution was stirred at room temperature for 2 h, during which the solution changed from purple-red to bright red.The solvent was removed under vacuum and the residue extracted with 10 mL of ether. Removal of the ether via vacuum yielded pure 7 as a red powder. Yield 282 mg, 89% (Found: C, 38.31; H, 3.49; N, 5.16. Calc. for C17H18Cl2N2O2W: C, 38.02; H, 3.38; N, 5.22%). (PhN)W(CO)2Cl2(i-C4H9NH2) 8. Complex 5 (237.5 mg, 0.54 mmol) was dissolved in 8 mL of CH2Cl2 and i-C4H9NH2 (56 mL, 0.54 mmol) was added via syringe. The solution was stirred for 2 h, then the solvent was removed under vacuum.The residue was extracted with ether–hexane (5 : 1) to yield a red solution from which complex 8 was obtained as a red powder upon evaporation of the solvent. Yield 184 mg, 71.7% (Found: C, 30.57; H, 3.64; N, 6.13. Calc. for C12H16Cl2N2O2W: C, 30.34; H, 3.40; N, 5.90%). [(PhN)W(CO)2Br2]2 4. Zwitterion 1 (1.95 g, 3.12 mmol) was dissolved in 100 mL of ether and cooled to 278 8C, after which a solution of Br2 in CH2Cl2 (499 mg, 3.12 mmol in 6 mL of CH2Cl2) was added via syringe.The solution was stirred for 2 h while the temperature was allowed to rise from 278 to 220 8C. During this period some dark red sticky solid formed and the solution changed from black to red. The reaction was allowed to continue for 0.5 h at room temperature and the ether was then removed by cannulation. The sticky solid was recrystallized with CH2Cl2–ether (1 : 3 by volume). The mother liquor was separated and the solvent removed from it by evaporation.The residue was then washed with ether to yield dimer 4 as a red solid. Yield 205 mg, 13.4% (Found: C, 19.47; H, 1.03; N, 2.81. Calc. for C8H5Br2NO2W: C, 19.58; H, 1.03; N, 2.85%). (PhN)W(CO)2Br2(MeCN) 6. Zwitterion 1 (1.20 g, 1.92 mmol) was dissolved in 15 mL of ether and cooled to 278 8C. A solution of Br2 in CH2Cl2 (307 mg, 1.92 mmol in 4 mL of CH2Cl2) was added via syringe, then 2 mL of MeCN were introduced. Within 10 min the solution changed from black to red and red solid began to form.The solution was stirred for 2 h while the temperature rose from 278 8C to room temperature and then was stirred for 0.5 h at room temperature. The solvent was removed by cannulation and the solid washed with ether (2 × 5 mL) to yield complex 6 as a dark red solid. Yield 830 mg, 81.3%. HRMS (FAB): (M 2 CO)1, found 503.8535, calc. 503.8479. (PhN)W(CO)2Br2(Me3C6H2NH2) 9. Complex 6 (81 mg, 0.15 mmol) was dissolved in 8 mL of CH2Cl2 and Me3C6H2NH2 (22.5 mL, 0.15 mmol) added via syringe.The solution was stirred for 1 h, during which it changed from dark red to bright red. The solvent was removed under vacuum and the residue extracted with ether. Removal of the ether yielded complex 9 as a red powder. Yield 54 mg, 57.5% (Found: C, 32.57; H, 2.65; N, 4.53. Calc. for C17H18Br2N2O2W: C, 32.62; H, 2.90; N, 4.48%). (PhN)W(CO)2Br2(i-C4H9NH2) 10. Complex 6 (71 mg, 0.13 mmol) was dissolved in 10 mL of CH2Cl2 and i-C4H9NH2 (13.70 mL, 0.13 mmol) added via syringe. The solution was stirred at room temperature for 2 h, during which a bright red color developed.The solvent was removed under vacuum and the residue extracted with 15 mL of ether. Removal of the ether yielded complex 10 as a sticky red solid. Yield 44 mg, 60.0%. HRMS (FAB): (M 2 2CO)1, found 507.9343, calc. 507.9174. (PhN)W(CO)Br2(PMe3)2. Dimer 4 (320 mg, 0.33 mmol) was dissolved in 10 mL of CH2Cl2 and PMe3 (5 equivalents) was added via syringe.The solution was stirred at room temperature for 2 h, then the solvent was removed and the residue extracted with ether–hexane (2 : 1). Removal of the solvent yielded the complex as a purple-red solid. Yield 280 mg, 64.0%. HRMS (FAB): (M 2 CO)1, found 586.9229, calc. 586.9167. [(PhN)W(CO)2Br(I)]2. Zwitterion 1 (2.0 g, 3.20 mmol) was dissolved in 15 mL of ether at 0 8C and BrI–CH2Cl2 solution (3.20 mL, 1.0 M in CH2Cl2) was added via syringe. Reaction occurred immediately as evidenced by a change from black to red and the dimer began to precipitate as an orange red solid.The solution was stirred for 2 h and the solvent then removed by cannulation. The resulting solid was washed with ether (2 mL × 3) to yield an orange-red solid. Yield 980 mg, 57% (Found: C, 17.83; H, 1.08; N, 2.43. Calc. for C8H5BrINO2W: C, 17.87; H, 0.94; N, 2.61%). (PhN)W(CO)2Br(I)(MeCN). The above dimer (432 mg, 0.40 mmol) was dissolved in 10 mL of MeCN and the solution was stirred at room temperature overnight, during which it became dark red. The solvent was removed under vacuum and the residue recrystallized with CH2Cl2–hexane to yield the complex as a red solid.Yield 252 mg, 54.4%. HRMS (FAB): (M 2 CO)1, found, 549.8339, calc. 549.8355. X-Ray crystallography Crystals of complex 4 were obtained by slow recrystallization in CH2Cl2–toluene (60 : 1) at 220 8C. Crystals of complex 3 were obtained from CH2Cl2–toluene (20 : 1) under similar conditions.Data for both compounds were collected at 173 K on a Siemens CCD SMART PLATFORM equipped with a CCD area detector and a graphite monochromator utilizing Mo-Ka radiation (l = 0.71073 Å). Cell parameters were refined using 6121 and 5919 reflections from the bromo dimer 4 and chloro dimer 3 data sets, respectively. A hemisphere of data (1381 frames) was collected using the w-scan method (0.38 frame width). The first 50 frames were remeasured at the end of data collection to monitor instrument and crystal stability (maximum correction on I was <1%).Absorption corrections were applied based on the y scan using the entire data sets. Table 5 Crystallographic data for complexes 3 and 4 * Empirical formula Ma /Å b/Å c/Å b/8 V/Å3 Total reflections measured Unique reflections m(Mo-Ka)/mm21 Goodness of fit S R1/Reflections [I > 2s(I)] wR2/Reflections Minimum, maximum peaks in Fourier-diVerence map/ e Å23 3 C16H10Cl4N2O4W 803.76 7.1433(3) 10.1298(4) 14.9460(7) 95.139(1) 1077.15(8) 7669 2475 11.191 1.13 3.10/2334 6.75/2475 21.48, 0.98 4 C16H10Br4N2O4W2 981.60 7.1118(1) 10.4920(1) 15.7374(1) 95.079(1) 1169.66(2) 8334 2685 16.682 1.12 4.90/2423 12.53/2685 21.90, 2.67 * Details in common: monoclinic, space group P21/n; Z = 2; R1 = S( Fo| 2 |Fc )/S|Fo|; wR2 = [Sw(Fo 2 2 Fc 2)2/Sw(Fo 2)2]� �� ; S = [Sw(Fo 2 2 Fc 2)2/(n 2 p)]� �� , n = number of reflections, p = number of parameters refined; w = 1/[s2(Fo 2) 1 (0.0370p)2 1 0.31p], p = [max(Fo 2,0) 1 2Fc 2]/3.3378 J.Chem. Soc., Dalton Trans., 1998, 3373–3378 Both structures were solved by direct methods in SHELXL 97,13 and refined using full-matrix least squares on F2. The non- H atoms were refined with anisotropic thermal parameters. All of the H atoms were included in the final cycle of refinement riding on the atoms to which they are bonded. A total of 128 parameters were refined in the final cycle of refinement of each compound using 2423 and 2334 reflections for 4 and 3, respectively, with I > 2s(I) to yield R1 and wR2 of 0.049 and 0.1253 for 4 and 0.031 and 0.067 for 3.Crystallographic data are given in Table 5. CCDC number 186/1110. See http://www.rsc.org/suppdata/dt/1998/3373/ for crystallographic files in .cif format. Acknowledgements Funding for this research was provided by the OYce of Naval Research. K. A. A. wishes to acknowledge the National Science Foundation and the University of Florida for funding the purchase of the X-ray equipment.Y. H. thanks Ms. Karen Torraca and Ms. Jennifer McCusker for helpful discussion. References 1 D. E. Wigley, Prog. Inorg. Chem., 1994, 42, 239. 2 W. A. Nugent and B. L. Haymore, Coord. Chem. Rev., 1980, 31, 123; S. Cenini and G. La Monica, Inorg. Chim. Acta, 1976, 18, 279. 3 L. W. Francisco, P. S. White and J. L. Templeton, Organometallics, 1997, 16, 2547; S. G. Feng, P. S. White and J. L. Templeton, Organometallics, 1995, 14, 5184; P.J. Pérez, P. S. White, M. Brookhart and J. L Templeton, Inorg. Chem., 1994, 33, 6050; K. R. Powell, P. J. Pérez, L. Luan, S. G. Feng, P. S. White, M. Brookhart and J. L Templeton, Organometallics, 1994, 13, 1851; L. Luan, M. Brookhart and J. L. Templeton, Organometallics, 1992, 11, 1433; P. J. Pérez, L. Luan, P. S. White, M. Brookhart and J. L. Templeton, J. Am. Chem. Soc., 1992, 114, 7928; L. Luan, P. S. White, M. Brookhart and J. L. Templeton, J. Am. Chem. Soc., 1990, 112, 8190. 4 S. T. Massey, N. D. R. Barnett, K. A. Abboud and L. McElwee- White, Organometallics, 1996, 15, 4625. 5 C. T. Maxey, H. F. Sleiman, S. T. Massey and L. McElwee-White, J. Am. Chem. Soc., 1992, 114, 5153; B. A. Arndtsen, H. F. Sleiman, A. K. Chang and L. McElwee-White, J. Am. Chem. Soc., 1991, 113, 4871; H. F. Sleiman, S. Mercer and L. McElwee-White, J. Am. Chem. Soc., 1989, 111, 8007; H. F. Sleiman and L. McElwee-White, J. Am. Chem. Soc., 1988, 110, 8700. 6 (a) P. C. McGowan, S. T. Massey, K. A. Abboud and L. McElwee- White, J. Am. Chem. Soc., 1994, 116, 7419; (b) N. D. R. Barnett, S. T. Massey, P. C. McGowan, J. J. Wild, K. A. Abboud and L. McElwee- White, Organometallics, 1996, 15, 424. 7 (a) J. E. McCusker, K. A. Abboud and L. McElwee-White, Organometallics, 1997, 16, 3863; (b) J. E. McCusker, J. Logan and L. McElwee-White, Organometallics, in press. 8 S. W. Johnston, Y.-X. He, T. J. Anderson and L. McElwee-White, unpublished work. 9 P. K. Baker, K. R. Flower, H. M. Naylor and K. Voigt, Polyhedron, 1993, 12, 357. 10 F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, 5th edn., Wiley-Interscience, New York, 1988, pp. 570–572. 11 F.-M. Su, J. C. Bryan, S. Jang and J. M. Mayer, Polyhedron, 1989, 8, 1261. 12 A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G. Watson and R. Taylor, J. Chem. Soc., Dalton Trans., 1989, S1. 13 G. M. Sheldrick, SHELXL 97, Program for the refinement of crystal structures. University of Göttingen, 1997. Pa

ISSN:1477-9226    

DOI:10.1039/a803989e

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3379–3390 3379 Synthesis, NMR studies, and molecular orbital calculations on cyclohexadienyl derivatives of (Á6-arene)tris(pyrazolyl)ruthenium(II) compounds Sameer Bhambri,a Andrea Bishop,b Nikolas Kaltsoyannis *b and Derek A. Tocher *a a Department of Chemistry, University College London, 20 Gordon Street, London, UK WC1H 0AJ. E-mail: d.a.tocher@ucl.ac.uk b Centre for Theoretical and Computational Chemistry, University College London, 20 Gordon Street, London, UK WC1H 0AJ.E-mail: n.kaltsoyannis@ucl.ac.uk Received 5th June 1998, Accepted 4th September 1998 The addition of nucleophiles to the complex cations [Ru(h6-arene){k3-HB(Pz3)}]1 or [Ru(h6-arene){k3-HC(Pz3)}]21 (Pz = pyrazol-1-yl) proceeds smoothly to give [Ru(h5-cyclohexadienyl){k3-HB(Pz3)}] or [Ru(h5-cyclohexadienyl)- {k3-HC(Pz3)}]1 respectively, which have been identified by NMR spectroscopy. NMR spectroscopy reveals that each of the cyclohexadienyl derivatives exhibits significant barriers to rotation of the h5 ligand.The electronic structure of [Ru(h6-arene){k3-HB(Pz3)}]1 and the cyclohexadienyl derivatives has been probed by extended Hückel molecular orbital calculations, with emphasis on the rationalisation of the experimentally observed barriers to rotation of the carbocycles. The single crystal structure of [Ru(h5-C6H6CN){k3-HC(Pz3)}][PF6] is reported. Introduction In the past we have examined the attack of hydride and other nucleophiles on the arenes in a range of [Ru(h6-arene)(h6- [2.2]paracyclophane)]21 complexes.1,2 These reactions can be used for the synthesis of a number of cyclohexadienyl 3–7 and diene 8,9 derivatives.However, the nature of the products is highly dependent upon the details of the synthetic procedure, and changes in solvent, the hydridic reagent, and the method of work-up can radically aVect the identity of the isolated products. 2 While the properties of the [2.2]paracylcophane ligand are such that in the case of double nucleophilic attack we can guarantee that diene containing complexes are formed in preference to bis(cyclohexadienyl) containing ones we nevertheless have little control over whether the new complex contains a 1,3- or 1,4-diene.Indeed in many instances a mixture of both is present in the isolated product. While this is an interesting observation it does limit the practical application of this chemistry from the point of view of arene and diene functionalisation.As a consequence of this we set out to develop a range of compounds in which the spectator ligand could exert some steric control on the chemistry occuring at the coordinated arene. The complexes which we have focused on are a selection of [Ru(h6-arene){k3-HB(Pz3)}]1 and [Ru(h6-arene){k3-HC- (Pz3)}]21 (Pz = pyrazol-1-yl) derivatives in which one might imagine that substituents in the ‘3’ position on the pyrazolyl rings would interact with the p-bound arene to influence its reactivity.10,11 Preliminary results 12 indicated that while the interaction with the coordinated arene might be limited there is evidence of significant interaction in the cyclohexadienyl products formed by the attack of a single nucleophile.That interaction has now been extensively explored, is clearly a function of the substituents on the carbocyclic ring, and is the focus of this report. Experimental and computational details Instrumental Infrared spectra were recorded on a Nicolet-205 spectrometer between 4000 and 400 cm21 as KBr discs.NMR spectra were recorded on Varian VXR400 or Bruker 300 spectrometers and referenced internally against the residual protons of the deuteriated solvents (d6-DMSO, d6-acetone, CDCl3, and CD2Cl2). Microanalyses were carried out by the departmental service at University College London. Fast atom bombardment mass spectra (assignments based on the 102Ru isotope) were recorded by the University of London Intercollegiate Research Service (ULIRS) at the London School of Pharmacy.All manipulations were carried out under nitrogen with degassed laboratory grade solvents using conventional Schlenk-line techniques. Starting materials Ruthenium trichloride hydrate was obtained on loan from Johnson Matthey plc and was purified before use by repeated dissolution in water and boiling to dryness. The compounds [Ru(h6-arene){k3-HB(Pz)3}][PF6], [Ru(h6-C6H6){k3-HB(3,5- Me2Pz)3}][PF6], and [Ru(h6-arene){k3-HC(Pz)3}][PF6]2 were synthesised as reported previously.10,11 Preparations [Ru(Á5-C6H7){Í3-HB(Pz)3}] 1.[Ru(h6-C6H6){k3-HB(Pz)3}]- [PF6] (0.09 g, 0.16 mmol) was suspended in degassed thf (10 cm3) and treated with Na[BH4] (0.05 g, excess). After stirring for 1 h at room temperature the solution was filtered through Celite to remove any unreacted Na[BH4]. Evaporation of the resulting filtrate to dryness gave 1 as a pale yellow air sensitive residue.Yield: 0.05 g, 0.12 mmol, 75% (Found: C, 45.5; H, 4.5; N, 21.4. Calc. for C15H17N6BRu: C, 45.9; H, 4.4; N, 21.4%). Mass spectrum: m/z 393 [M 2 H]1. Infrared: n(BH) 2494, n(CHendo) 2963, n(CHexo), 2782 cm21. [Ru(Á5-C6H6CN){Í3-HB(Pz)3}] 2. [Ru(h6-C6H6){k3-HB- (Pz)3}][PF6] (0.05 g, 0.09 mmol) was suspended in degassed thf and treated with KCN (0.05 g, excess). The mixture was stirred at room temperature for 2 h and then pumped to dryness. Extraction into chloroform followed by filtering through Celite and evaporation to dryness gave 2 as an air stable yellow residue.Yield: 0.033 g, 0.079 mmol, 84% (Found: C, 46.1; H,3380 J. Chem. Soc., Dalton Trans., 1998, 3379–3390 3.8; N, 23.3. Calc. for C16H16N7BRu: C, 45.9; H, 3.9; N, 23.5%). Mass spectrum: m/z 419 [M]1, 393 [M 2 CN]1. Infrared: n(BH) 2484, n(CN) 2214, n(CHendo) 2945 cm21. [Ru(Á5-C6H6OH){HB(Pz)3}] 3. [Ru(h6-C6H6){k3-HB(Pz)3}]- [PF6] (0.050 g, 0.09 mmol) was suspended in degassed methanol (10 cm3) and treated with methanolic KOH (0.047 g, excess).The mixture was stirred at room temperature for 48 h and then evaporated to dryness. Extraction into thf followed by filtration through Celite and evaporation to dryness gave 3 as a yellow powder. Yield: 0.031 g, 0.08 mmol, 81% (Found: C, 43.7; H, 4.1; N, 20.1. Calc. for C15H17N6BORu: C, 44.0; H, 4.2; N, 20.5%). Mass spectrum: m/z 393 [M 2 OH]1. Infrared: n(BH) 2497, n(OH) 3450, n(CHendo) 2950 cm21.[Ru(Á5-C6H6D){Í3-HB(Pz)3}] 4. [Ru(h6-C6H6){k3-HB(Pz)3}]- [PF6] (0.08 g, 0.16 mmol) was treated with Na[BD4] (0.04 g, excess) and worked up as described for 1. Yield: 0.05 g, 0.12 mmol, 77%. Mass spectrum: m/z 393 [M 2 D]1. Infrared: n(BH) 2492, n(CHendo) 2933, n(CDexo) 2088 cm21. [Ru(Á5-1-iPr-4-MeC6H5){Í3-HB(Pz)3}] 5a and 5b. [Ru(h6-1- iPr-4-MeC6H4){k3-HB(Pz)3}][PF6] (0.124 g, 0.21 mmol) was suspended in degassed thf (10 cm3) and treated with Na[BH4] (0.05 g, excess).After stirring for 2 h at room temperature the solution was filtered through Celite to remove any unreacted Na[BH4]. Evaporation of the resulting filtrate to dryness led to deposition of a crude yellow product. Purification by passing a diethyl ether solution of 5 down an alumina column (mesh 100– 250) with subsequent evaporation to dryness of the eluent resulted in isolation of 5a and 5b as a mildly air sensitive solid. The two isomers were not separated. Yield: 0.076 g, 0.17 mmol, 82% (Found: C, 50.8; H, 5.5; N, 18.5.Calc. for C19H25N6BRu: C, 50.8; H, 5.6; N, 18.7%). Mass spectrum: m/z 450 [M]1. Infrared: n(BH) 2452, n(CHendo) 2922, n(CHexo) 2776 cm21. [Ru(Á5-1-iPr-4-MeC6H4CN){Í3-HB(Pz)3}] 6a and 6b. [Ru(h6- 1-iPr-4-MeC6H4){k3-HB(Pz)3}][PF6] (0.134 g, 0.23 mmol) was suspended in degassed thf (15 cm3) and treated with KCN (0.05 g, excess). The mixture was refluxed for 5 d then filtered and evaporated to dryness. Purification was carried out by repeated extraction of the residue with CHCl3.The two isomers of 6 were not separated. Yield: 0.067 g, 0.14 mmol, 63%. (Found: C, 49.6; H, 5.2; N, 20.8. Calc. for C20H24N7BRu: C, 50.0; H, 5.1; N, 20.7%). Mass spectrum: m/z 449 [M 2 CN]1. Infrared: n(BH) 2456, n(CHendo) 2931, n(CN) 2221 cm21. [Ru(Á5-1-iPr-4-MeC6H4OH){Í3-HB(Pz)3}] 7a and 7b. [Ru(h6- 1-iPr-4-MeC6H4){k3-HB(Pz)3}][PF6] (0.110 g, 0.19 mmol) was suspended in degassed thf (15 cm3) and treated with NaOH (0.05 g, excess).The mixture was refluxed for 4 d then filtered and evaporated to dryness. The purification procedure was similar to that for 5. The isomers of 7 were not separated. Yield: 0.053 g, 0.11 mmol, 61% (Found: C, 49.9; H, 5.3; N, 18.0. Calc. for C19H25N6BORu: C, 49.1; H, 5.4; N, 18.1%). Mass spectrum: m/z 449 [M 2 OH]1. Infrared: n(BH) 2460, n(CHendo) 2959, n(OH) ca. 3400 cm21. [Ru(Á5-1,4-Me2C6H5){Í3-HB(Pz)3}] 8. [Ru(h6-1,4-Me2C6H4)- {k3-HB(Pz)3}][PF6] (0.121 g, 0.23 mmol) was suspended in degassed thf (15 cm3) and treated with Na[BH4] (0.05 g, excess). The mixture was refluxed for 2 h then filtered and evaporated to dryness. The crude residue was extracted with chloroform, filtered through Celite and evaporated to dryness to give the product as a yellow powder.Yield: 0.083 g, 0.20 mmol, 88% (Found: C, 46.8; H, 5.0; N, 18.8. Calc. for C17H21N6BRu: C, 46.4; H, 4.8; N, 18.9). Mass spectrum: m/z 442 [M]1. Infrared: n(BH) 2457, n(CHendo) 2952, n(CHexo) 2778 cm21.[Ru(Á5-1,4-iPr2C6H5){Í3-HB(Pz)3}] 9. [Ru(h6-1,4-iPr2C6H4)- {k3-HB(Pz)3}][PF6] (0.132 g, 0.21 mmol) was suspended in degassed thf (15 cm3) and treated with Na[BH4] (0.05 g, excess). The mixture was refluxed for 2 h then filtered and evaporated to dryness. The crude residue was extracted with diethyl ether, filtered through Celite and evaporated to dryness. Yield: 0.083 g, 0.18 mmol, 82% (Found: C, 53.7, H, 4.6; N, 17.4. Calc. for C21H28N6BRu: C, 54.0 H, 4.1; N, 18.0%).Mass spectrum: m/z 477 [M 2 H]1. Infrared: n(BH), 2473. [Ru(Á5-1,3,5-Me3C6H4){Í3-HB(Pz)3}] 10. [Ru(h6-1,3,5-Me3- C6H3){k3-HB(Pz)3}][PF6] (0.132 g, 0.21 mmol) was suspended in degassed diethyl ether (15 cm3) and treated with Na[BH4] (0.05 g, excess). The mixture was refluxed for 2 h then filtered and evaporated to dryness. The crude residue was extracted with chloroform, filtered through Celite and evaporated to dryness. Yield: 0.071 g, 0.12 mmol, 59% (Found: C 50.1, H 5.1, N, 20.1.Calc. for C18H23N6BRu: C 49.7, H, 5.3, N, 19.3%). Mass spectrum: m/z 436 [M]1. Infrared: n(BH) 2468, n(CHendo) 2950, n(CHexo) 2785 cm21. [Ru(Á5-C6H6CN){Í3-HC(Pz)3}][PF6] 11. [Ru(h6-C6H6){k3- HC(Pz)3}][PF6]2 (0.091 g, 0.13 mmol) was suspended in degassed thf (15 cm3) and treated with KCN (0.05 g, excess). The mixture was refluxed for 3 d then filtered and evaporated to dryness. The crude residue was extracted with chloroform, filtered through Celite and evaporated to dryness.Yield: 0.056 g, 0.10 mmol, 74% (Found: C, 36.1; H, 2.6; N, 17.1. Calc. for C17H16N7BF6PRu: C, 36.2; H, 2.9; N, 17.4%). Mass spectrum: m/z 420 [M 2 PF6]1, 394 [M 2 CN 2 PF6]1. Infrared: n(CHendo) 2963 cm21. [Ru(Á5-1-iPr-4-MeC6H4CN){Í3-HC(Pz)3}][PF6] 12a and 12b. [Ru(h6-1-iPr-4-MeC6H4){k3-HC(Pz)3}][PF6]2 (0.122 g, 0.16 mmol) was suspended in degassed thf (15 cm3) and treated with KCN (0.05 g, excess). The mixture was refluxed for 3 d then filtered and evaporated to dryness.The crude residue was extracted with chloroform, filtered through Celite and evaporated to dryness. The two isomers of 12 were not separated. Yield: 0.0786 g, 0.11 mmol, 65% (Found: C 40.8; H 4.0; N 15.4. Calc. for C21H24N7F6PRu: C 40.7; H 3.9, N 15.8%). Mass spectrum: m/z 476 [M 2 PF6]1. Infrared: n(CHendo) 2915, n(CN) 2224 cm21. [Ru(Á5-1,4-Me2C6H4CN){Í3-HC(Pz)3}][PF6] 13. [Ru(h6-1,4- Me2C6H4){k3-HC(Pz)3}][PF6]2 (0.108 g, 0.15 mmol) was suspended in degassed thf (15 cm3) and treated with KCN (0.05 g, excess).The mixture was refluxed for 3 d, then filtered and evaporated to dryness. The crude residue was extracted with chloroform, filtered through Celite and evaporated to dryness. Yield: 0.067 g, 0.11 mmol, 74% (Found: C, 36.7; H, 2.9; N, 17.6. Calc. for C19H20N7F6PRu: C, 36.2; H, 2.9; N, 17.4%). Mass spectrum: m/z 395 [M 2 PF6]1. Infrared: n(CHendo) 2926, n(CN) 2218 cm21. [Ru(Á5-C6H7){Í3-HB(3,5-Me2Pz)3}] 14. [Ru(h6-C6H6){k3-HB- (3,5-Me2Pz)3}][PF6] (0.05 g, 0.08 mmol) was dissolved in thf (10 cm3) and treated with Na[BH4] (0.05 g, excess).The mixture was stirred at room temperature for 1 h. The red solution was filtered through Celite and evaporated to dryness, giving 14 as a dark residue. Purification was carried out by passing a thf solution of the residue down an alumina column and subsequently evaporating the eluate to dryness. Yield: 0.03 g, 0.06 mmol, 68% Accurate analytical data could not be obtained due to instability however NMR data are consistent with the proposed formulation.Infrared: n(BH) 2538, n(CHexo) 2798, n(CHendo) 2924 cm21. Crystallography Crystal data for [Ru(h5-C6H6CN){k3-HC(Pz)3}][PF6]?Me2CO C20H22F6N7OPRu, M = 622.49, triclinic, space group P1� ,J. Chem. Soc., Dalton Trans., 1998, 3379–3390 3381 a = 10.720(5), b = 11.186(4), c = 11.821(3) Å, a = 64.56(3), b = 71.43(3), g = 77.97(3)8, U = 1209.2(9) Å3 (by least squares refinement of diVractometer angles for 25 centred reflections in the range 17.0 < 2q < 26.88), l = 0.71073 Å, Z = 2, F(000) = 624, Dc = 1.71 g cm23, m(Mo-Ka) = 7.89 cm21, yellow crystal, 0.30 × 0.23 × 0.25 mm.The w–2q technique was used to measure 4404 reflections (4176 unique) in the range 5 < 2q < 508 at 20 8C. Data were corrected for Lorentz polarisation and absorption eVects (Y scan method). The structure was solved by conventional direct methods.13 and developed by using alternating cycles of least squares refinement and diVerence-Fourier synthesis.14 In the final stages of refinement the presence of an acetone of solvation became apparent.The solvent molecule was best modelled as having a two-fold disorder along one of the carbon– carbon bonds such that two sites in the acetone were modelled as a 50 : 50 mixture of carbon and oxygen atoms. All nonhydrogen atoms, except those involved in the disorder, were modelled anistropically. Hydrogen atoms on the complex cation were placed in idealised positions and assigned a common isotropic thermal parameter (Uiso = 0.08 Å2).The hydrogens of the disordered solvent were omitted. The final cycle of least squares refinement included 315 parameters for 4172 variables and did not shift any parameter by more than 0.001 times its standard deviation. The final R values were R = 0.0773, R9 = 0.185 (for data with I > 2s(I), refinement based on F) and R = 0.109, R9 = 0.204 (for all unique reflections, refinement based on F2), and the final diVerence-fourier was featureless with no peaks greater than 1.05 e Å23.CCDC reference number 186/1151. Computation All of the calculations reported in this work were performed at the extended Hückel level of approximation, using the computer aided composition of atomic orbitals (CACAO) package due to Mealli and Proserpio.15 These programs employ standard extended Hückel methodology,16 with the weighted Hij formula.17 Mulliken population analyses were performed.18 Bond lengths and angles were taken from crystal structure data where available, and idealised to provide the highest possible symmetry.Results and discussion Experimental studies Nucleophilic attack on coordinated benzene. Treatment of a suspension of [Ru(h6-C6H6){k3-HB(Pz)3}][PF6] in tetrahydrofuran with Na[BH4] results in formation of a yellow solution, which upon work-up leads to isolation of [Ru(h5-C6H7){k3- HB(Pz)3}] 1 as an air-sensitive, yellow solid in moderate yield.Similar reactions of [Ru(h6-C6H6){k3-HB(Pz)3}][PF6] with KCN or KOH give the cyanocyclohexadienyl and hydroxycyclohexadienyl derivatives [Ru(h5-C6H6X){k3-HB- (Pz)3}], X = CN 2, OH 3. While 2 is remarkably stable in solution the corresponding solutions of 1 and 3 change from yellow to green in a few minutes and eventually deposit intractable dark solids. The infrared spectrum of 1 displays two strong bands, at 2963 and 2782 cm21, which are assigned as n(CHendo) and n(CHexo) respectively.19,20 The latter band is absent in the spectra of 2 and 3, although additional bands are observed at 3450 and 2214 cm21 (due to n(OH) 3 and n(CN) 2). Reaction of [Ru(h6-C6H6){k3-HB(Pz)3}][PF6] with Na[B] gives the deuteriocyclohexadienyl [Ru(h5-C6H6D){k3-HB(Pz)3}] 4 which exhibits a n(CHendo) band at 2933 cm21, no band at ca. 2750– 2800 cm21, and the n(CDexo) band at 2088 cm21. These observations confirm that the entering nucleophile does so along an exo reaction pathway.21 The room temperature 1H NMR spectrum of 1 exhibits five sharp signals for the cyclohexadienyl ligand over a wide range of chemical shifts (d 5.67 (t), 4.55 (t), 2.73 (t), 2.14 (m) and 2.11 (d)).The appearance of the signals for Hexo as a widely spaced doublet (2J = 13.3 Hz) and Hendo as a multiplet is consistent with the related resonances reported for the [2.2]paracyclophane derivative [Ru(h5-C6H7)(h6-C16H16)]1.1 The 1H NMR spectra (Table 1) of 2–4 are closely similar to that of 1 except that the resonance for Hexo is absent in each of the spectra and the signals due to Hendo now appear as triplets. While there is nothing remarkable about the cyclohexadienyl signals for any of these compounds, the pyrazolyl regions of the 1H NMR spectra of 1–4 are quite intriguing.While integration of the two sets of signals confirm the 1 : 1 stoichiometry of the ligands, and is consistent with the proposed formulations, the signals for the pyrazolyl protons do not appear as three sharp resonances similar to those observed for [Ru(h6-C6H6){k3-HB(Pz)3}][PF6] (d 8.67 H3, 6.44 H4 7.87 H5).10 For instance at room temperature the spectrum of 2 exhibits only two broad pyrazolyl resonances (d 7.58 and 6.16, integral 2 : 1).However on warming the sample to 50 8C three broad signals become apparent (Fig. 1, d 7.77 H3, 7.57 H5, 6.17 H4). Conversely if the temperature of the NMR probe is lowered the broad signals begin to split into two subsets, each of three signals, at around 220 8C and further lowering of the temperature results in the sharpening of these resonances such that at 265 8C there are two well defined sets of resonances (d 8.52, 7.77, 6.44 and d 7.54, 7.46, 6.10) in integral ratio 1 : 2 (Fig. 1). It is notable that throughout this temperature range the signals due to the cyclohexadienyl ligand remain invariant. These observations are repeated for compounds 1, 3 and 4 with no significant diVerence in temperature for coalescence.The 13C-{1H} NMR spectra of 2 (Fig. 2) show a similar pattern of behaviour with the cyclohexadienyl resonances appearing sharp at all temperatures while the broad signals at around 20 8C are replaced by sharp resonances on cooling to 265 8C. At the lower temperature the signals due to carbons in the 3 and 4 positions on the pyrazolyl rings are clearly split (Table 2) while only in the case of the hydroxycyclohexadienyl 3 is the resonance for the carbon in the 5 position resolved.This is quite reasonable given the greater displacement of that carbon atom from the asymmetric ligand. If the dicationic complex [Ru(h6-C6H6){k3-HC(Pz)3}][PF6]2 is reacted with KCN a new cyclohexadienyl compound [Ru(h5- C6H6CN){k3-HC(Pz)3}][PF6] 11 is formed, which is closely analogous to 2. Variable temperature NMR studies on this compound produce spectra which closely mimic those of the pyrazolylborate derivatives.Clearly these results indicate that the five compounds described so far undergo some kind of fluxional process which in the fast exchange regime average the three pyrazolyl environments yet at lower temperatures render one of the three pyrazolyl groups in the tris(pyrazolyl)borate or tris(pyrazolyl)methane ligand as unique. At first sight it is tempting to attribute these observations to the [HX(Pz)3]n1 (X = B, n = 0; X = C, n = 1) ligand undergoing an hapticity change between k3 and k2 coordination modes, as has been observed for a number of rhodium complexes.22,23 However we do not believe that this process is occurring here, as attempts to trap a complex with a k2 coordinated ligand, by placing 1 or 11 in solution in the presence of carbon monoxide or P(OMe)3, were unsuccessful. In addition we have previously reported 11 H Y Ru N N N N N N X ===== H H CN OH D CN 1234 11 X = B, Y X = C, Y3382 J.Chem. Soc., Dalton Trans., 1998, 3379–3390 Table 1 Selected 1H NMR data on non-alkylated ruthenium(cyclohexadienyl)hydridotris(pyrazolyl)-borate and -methane compounds Pyrazolyl borate (d, J/Hz), CDCl3 3- 4- 5- Additional resonances [Ru(h5-C6H7){k3-HB(Pz)3}] 1 265 8C 8.83 (b, 1H, Pz1), 7.53 (s, 2H, Pz2) 6.45 (b, 1H, Pz1), 6.07 (dd, 2H, Pz2) 7.78 (b, 1H, Pz1), 7.46 (d, 2H, Pz2) Cyclohexadienyl: 5.71 (t, 1H, 3J = 4.24, Ha), 4.62 (t, 2H, 3J = 5.56, Hb), 2.74 (m, 1H, 2J = 12.98, 3J = 5.64, Hendo), 2.14 (t, 2H, 3J = 6.07, Hc), 2.07 (d, 1H, 3J = 13.33, Hexo).[Ru(h5-C6H6CN){k3-HB(Pz)3}] 2 265 8C 8.52 (d, 1H, J = 1.11, Pz1) 7.54 (d, 2H, J = 2.02, Pz2) 6.44 (b, 1H, Pz1) 6.10 (b, 2H, Pz2) 7.77 (d, 1H, J = 2.11, Pz1) 7.46 (d, 2H, J = 1.45, Pz2) Cyclohexadienyl: 5.87 (t, 1H, 3J = 4.57, Ha), 4.83 (t, 2H, 3J = 5.51, Hb), 3.72 (t, 1H, 3J = 6.10, Hendo), 2.39 (t, 2H, 3J = 6.05, Hc). [Ru(h5-C6H6OH){k3-HB(Pz)3}] 3 265 8C 8.60 (b, 1H, Pz1) 7.55 (d, 2H, J = 1.00, Pz2) 6.43 (b, 1H, Pz1) 6.12 (b, 2H, Pz2) 7.76 (d, 1H, J = 1.95, Pz1) 7.53 (b, 2H, Pz2) Cyclohexadienyl: 5.62 (t, 1H, 3J = 4.42, Ha), 4.80 (t, 2H, 3J = 6.00, Hb), 3.68 (t, 1H, 3J = 5.71, Hendo), 2.72 (t, 2H, 3J = 5.89, Hc).(h[Ru(h5-C6H6D){k3-HB(Pz)3}] 4 21 8C 7.56 (b, 6H) a 6.11 (b, 3H) 7.56 (b, 6H) a Cyclohexadienyl: 5.66 (t, 1H, 3J = 4.44, Ha), 4.55 (t, 2H, 3J = 5.68, Hb), 2.71 (t, 1H, 3J = 6.00, Hendo), 2.12 (t, 2H, 3J = 6.09, Hc). [Ru(h5-C6H6CN){k3-HC(Pz)3}]- [PF6] 11 21 8C 8.41 (b, 6H) a 6.64 (b, 3H) 8.41 (b, 6H) a 9.41 [s, 1H, HC(Pz)3], cyclohexadienyl: 6.16 (t, 1H, 3J = 6.9, Ha), 5.34 (t, 2H, 3J = 4.8, Hb), 4.13 (t, 1H, 3J = 6.09, Hendo), 2.80 (t, 2H, 3J = 6.06, Hc).Pz1 = unique pyrazolyl group, Pz2 = doubly degenerate pyrazolyl group, s = singlet, d = doublet, b = broad, t = triplet, m = multiplet. a Broad overlapping signals. a range of complexes of the type [Ru(h6-arene){k2-HX(Pz)3}- Cl]n1 (X = B, n = 0; X = C, n = 1) in which the presence of the k2 coordinated ligand is conclusively established by X-ray crystallography.These compounds show a quite diVerent chemical shift for the H3 proton on the unique pyrazolyl ring, reinforcing the conclusion that some other process must be operating. Interestingly the crystal structure of 11 (Fig. 3) clearly demonstrates that in the solid state the complex ion adopts a conformation in which one pyrazolyl ring is eclipsed with the projection of the sp3 hybridised carbon atom (torsion angle N12–Ru–centroid–C4 = 28) and the remaining two Fig. 1 Variable temperature 1H NMR spectra for 2. crystallographically unique rings are psuedo-eclipsed with two carbons of the cyclohexadienyl (N22–Ru–centroid–C6 = 108, N32–Ru–centroid–C2 = 14.28). Such a structure is clearly consistent with that implied by the low temperature solution NMR data and this measurement has the advantage over the previously reported structure for 2 in that there is no crystallographically imposed symmetry in 11. The bond lengths in the cation are quite normal (Table 3), with the Ru–N distances being indistinguishable from each other, 2.129(7)–2.136(7) Å, the bonds from the metal to the cyclohexadienyl ligand falling in the range 2.111(9)–2.156(10) Å, and the Ru to sp3 carbon distance being 2.7 Å.As there is no experimental evidence for Ru–N bond rupture and the establishment of a k2�k3 interconversion, an alternative explanation for the dynamic NMR behaviour must be found. The only logical alternative would seem to be that at low temperature there is restricted rotation of one of the ligands about the metal–ligand axis.Although in principle either ligand could rotate about that axis it seems more logical to suggest that it is the p-bound ligand which would be able to do this without destabilising the complex. While it would be very diYcult to establish conclusively by experiment that this process was taking place, the likelihood of such a mechanism operating can be probed computationally.We have therefore conducted a series of extended Hückel molecular orbital calculations to Fig. 2 Variable temperare 13C-{1H} NMR spectra for 2.J. Chem. Soc., Dalton Trans., 1998, 3379–3390 3383 Table 2 Selected 13C-{1H} NMR data on non-alkylated ruthenium(cyclohexadienyl)hydridotris(pyrazolyl)-borate and -methane compounds a Pyrazolyl borate (d) CDCl3 Additional resonances 3- 4- 5- Ca Cb Cc Cd CN [Ru(h5-C6H7){k3-HB(Pz)3}] 1 265 8C 143.47 (Pz1) 141.97 (Pz2) 104.87 (Pz1) 105.38 (Pz2) 134.41 88.78 67.40 26.67 27.78 — [Ru(h5-C6H6CN){k3-HB(Pz)3}] 2 265 8C 145.75 (Pz1) 141.81 (Pz2) 105.71 (Pz1) 105.13 (Pz2) 134.58 88.26 70.42 24.07 28.30 120.65 [Ru(h5-C6H6OH){k3-HB(Pz)3}] 3 265 8C 142.20 (Pz1) 141.99 (Pz2) 105.44 (Pz1) 105.13 (Pz2) 134.67 (Pz1) 134.54 (Pz2) 88.21 68.80 31.47 29.84 — [Ru(h5-C6H6CN){k3-HC(Pz)3}][PF6] 11 21 8C 134.49 89.81 109.10 76.64 72.59 28.07 27.83 147.50 a Pz1 = unique pyrazolyl group, Pz2 = doubly degenerate pyrazolyl group.Ca Cd Hendo Y Ru N N N N N N XH Cb Cc evaluate this hypothesis, and the results of these studies are discussed below. Nucleophilic attack on substituted arenes. If the cyclohexadienyl ligands generated in the reactions discussed above do undergo restricted rotation about their Ru–ligand axis then it seems reasonable to suggest that such a process be influenced by the identity of any substituent on the arene ring. To probe this feature of the chemistry we have reacted a range of [Ru(h6-arene){HB(Pz)3}][PF6] compounds (arene = 1-iPr-4- MeC6H4, 1,4-Me2C6H4, 1,4-iPr2C6H4 or 1,3,5-Me3C6H3) with Fig. 3 Crystal structure of the cation in 11. nucleophiles and subjected the products to variable temperature NMR studies (Tables 4 and 5). Reaction of [Ru(h6-1-iPr-4-MeC6H4){HB(Pz)3}][PF6] with Na[BH4] in tetrahydrofuran at room temperature results in the formation of two isomeric products [Ru(h5-1-iPr-4- Table 3 Selected bond lengths (Å) and angles (8) for [Ru(h5- C6H6CN){k3-HC(Pz)3}][PF6]?Me2CO 11 Ru(1)–C(2) Ru(1)–N(32) Ru(1)–N(12) Ru(1)–C(3) C(1)–C(6) C(2)–C(3) C(4)–C(7) C(5)–C(6) N(11)–N(12) N(11)–C(40) C(13)–C(14) N(21)–C(25) N(21)–C(40) C(23)–C(24) N(31)–C(35) N(31)–C(40) C(33)–C(34) N(32)–Ru(1)–N(22) N(22)–Ru(1)–N(12) C(2)–C(3)–C(4) C(7)–C(4)–C(5) C(5)–C(4)–C(3) C(1)–C(6)–C(5) N(12)–N(11)–C(15) C(15)–N(11)–C(40) N(11)–N(12)–Ru(1) N(12)–C(13)–C(14) N(11)–C(15)–C(14) C(25)–N(21)–C(40) C(23)–N(22)–N(21) N(21)–N(22)–Ru(1) C(25)–C(24)–C(23) C(35)–N(31)–N(32) N(32)–N(31)–C(40) C(33)–N(32)–Ru(1) N(32)–C(33)–C(34) N(31)–C(35)–C(34) N(21)–C(40)–N(11) 2.111(9) 2.136(7) 2.130(6) 2.153(9) 1.38(2) 1.39(2) 1.488(14) 1.411(14) 1.340(9) 1.446(10) 1.378(13) 1.341(11) 1.434(11) 1.383(13) 1.337(11) 1.446(10) 1.342(14) 83.6(3) 82.9(2) 118.2(9) 113.1(8) 102.7(7) 120.0(9) 111.4(7) 129.1(8) 119.3(5) 109.5(8) 106.3(8) 127.7(8) 105.1(7) 118.2(5) 105.8(8) 112.2(7) 119.6(6) 137.2(6) 111.4(8) 105.6(9) 110.1(6) Ru(1)–C(6) Ru(1)–N(22) Ru(1)–C(5) Ru(1)–C(1) C(1)–C(2) C(3)–C(4) C(4)–C(5) C(7)–N(1) N(11)–C(15) N(12)–C(13) C(14)–C(15) N(21)–N(22) N(22)–C(23) C(24)–C(25) N(31)–N(32) N(32)–C(33) C(34)–C(35) N(32)–Ru(1)–N(12) C(6)–C(1)–C(2) C(4)–C(3)–Ru(1) C(7)–C(4)–C(3) C(6)–C(5)–C(4) N(1)–C(7)–C(4) N(12)–N(11)–C(40) N(11)–N(12)–C(13) C(13)–N(12)–Ru(1) C(15)–C(14)–C(13) C(25)–N(21)–N(22) N(22)–N(21)–C(40) C(23)–N(22)–Ru(1) N(22)–C(23)–C(24) N(21)–C(25)–C(24) C(35)–N(31)–C(40) C(33)–N(32)–N(31) N(31)–N(32)–Ru(1) C(33)–C(34)–C(35) N(21)–C(40)–N(31) N(31)–C(40)–N(11) 2.119(9) 2.129(7) 2.146(9) 2.156(10) 1.44(2) 1.503(14) 1.494(13) 1.142(13) 1.341(10) 1.329(10) 1.353(14) 1.352(9) 1.317(11) 1.366(14) 1.347(9) 1.340(10) 1.366(13) 82.4(2) 117.3(9) 93.5(6) 114.2(9) 118.0(9) 178.6(12) 119.4(6) 105.9(7) 134.6(6) 106.9(7) 111.8(7) 120.4(6) 136.7(6) 111.0(9) 106.2(8) 128.2(8) 103.9(7) 118.9(5) 106.9(8) 111.9(6) 109.2(7)3384 J.Chem. Soc., Dalton Trans., 1998, 3379–3390 Table 4 Selected 1H NMR data on alkylated ruthenium(cyclohexadienyl)hydriotris(pyrazole)-borate and -methane compounds Pyrazoyl borate (d, J/Hz) CDCl3 3- 4- 5- Cyclohexadienyl resonances [Ru(h5-1-iPr-4-MeC6H5){k3-HB(Pz)3}] 5a 1 5b 265 8C Isomer a 8.89, 7.93, 7.70 6.50, 6.11, 6.04 7.99, 7.69, 7.46 Isomer a: d 5.49 and 4.56 (d, 2H, 3J = 4.4, AB), 2.68 (m, 1H, 2J = 12.4, 3J = 6.8, Hendo), 2.46 (sep, 1H, 3J = 6.8, iPr), 2.15 (d, 1H, 2J = 12.4, Hexo), 1.79 (d, 1H, 3J = 6.0, Hc), 0.93 (s, 3H, Me), 0.74 (d, 3H, 3J = 6.8, iPr), 0.32 (d, 3H, 3J = 6.8, iPr).Isomer b 8.91, 7.93, 7.71 6.53, 6.12, 6.06 7.99, 7.69, 7.64 Isomer b: 5.54 and 4.58 (d, 2H, 3J = 4.4, AB), 2.77 (m, 1H, 2J = 12.4, 3J = 6.8, Hendo), 1.77 (sep, 1H, 3J = 6.8, iPr), 2.05 (d, 1H, 2J = 12.4, Hexo), 2.04 (s, 3H, Me), 1.88 (d, 1H, 3J = 6.0, Hc), 1.34 (d, 3H, 3J = 6.8, iPr), 1.08 (d, 3H, 3J = 6.8, iPr). [Ru(h5-1-iPr-4-MeC6H4CN){k3-HB(Pz)3}] 6a 1 6bc 21 8Ca Isomer a 8.80, 7.87, 7.86 6.46, 6.10, 6.03 7.88, 7.68, 7.58 Isomer a: 5.77 and 4.78 (d, 2H, 3J = 4.8, AB), 2.60 (sep, 1H, 3J = 6.8, iPr), 1.40 (s, 3H, Me), 0.93 (d, 3H, 3J = 6.8, iPr), 0.51 (d, 3H, 3J = 6.8, iPr).Isomer b Isomer b: 5.78 and 4.73 (d, 2H, 3J = 4.8, AB), 2.18 (s, 3H, Me), 1.91 (sep, 1H, 3J = 6.8, iPr), 1.41 (d, 3H, 3J = 6.8, iPr), 1.19 (d, 3H, 3J = 6.8, iPr). [Ru(h5-1-iPr-4-MeC6H4OH){k3-HB(Pz)3}] 7a 1 7b 265 8Ca,b Isomer a Isomer b 8.90, 7.96, 7.34 6.50, 6.18, 6.11 8.07, 7.73, 7.60 Isomer a: 5.57 and 4.93 (d, 2H, 3J = 4.8, AB), 3.95 (d, 1H, 3J = 5.6, Hendo), 2.68 (sep, 1H, J = 6.8, iPr), 2.44 (d, 1H, J = 5.6, Hc), 1.31 (d, 3H, J = 6.8, iPr), 1.31 (s, 3H, Me), 1.18 (d, 3H, J = 6.8, iPr).Isomer b: 5.52 and 5.13 (d, 2H, 3J = 4.8, AB), 4.81 (d, 1H, 3J = 5.6, Hendo), 4.09 (d, 1H, J = 5.6, Hc), 2.16 (s, 3H, Me), 1.06 (sep, 1H, J = 6.8, iPr), 1.09 (d, 3H, J = 6.8, iPr), 0.67 (d, 3H, J = 6.8, iPr). [Ru(h5-1,4-Me2C6H5){k3-HB(Pz)3}] 8 280 8C 8.82 (d, 1H, J = 1.6), 7.91 (d, 1H, J = 1.6), 7.73 (d, 1H, J = 2.4) 6.54 (dd, 1H), 6.12 (dd, 1H), 6.07 (dd, 1H) 8.01 (d, 1H, J = 2.0), 7.74 (d, 1H, J = 2.4), 7.68 (d, 1H, J = 2.8) 5.47 and 4.58 (d, 2H, 3J = 4.4, AB), 2.74 (dd, 1H, 3J = 5.6, 2J = 12.6, Hendo), 2.08 (d, 1H, 2J = 12.6, Hexo), 2.05 (s, 3H, Mea), 1.85 (d, 1H, 3J = 6, Hc), 0.89 (s, Meb, 3H).[Ru(h5-1,4-iPr2C6H5){k3-HB(Pz)3}] 9 40 8Cc 8.82, 7.77, 7.56 6.43, 6.05, 6.05 7.83, 7.53, 7.48 5.60 and 4.42 (2H, 3J = 4.4, AB), 2.74 (m, 1H, 3J = 6.8, 2J = 12.6, Hendo), 2.48 (sep, 1H, 3J = 6.8, iPra), 2.18 (d, 1H, 2J = 12.8, Hexo), 1.95 (d, 1H, 3J = 6.8, Hc), 1.79 (sep, 1H, 3J = 6.8, iPrb), 1.36 (d, 3H, 3J = 6.8, iPra), 1.12 (d, 3H, 3J = 6.8, iPra), 0.77 (d, 3H, 3J = 6.8, iPrb), 0.40 (d, 3H, 3J = 6.8, iPrb). [Ru(h5-1,3,5-Me3C6H4){k3-HB(Pz)3}] 10 21 8C 8.71 (d, 1H, 3J = 1.6, Pz1), 7.74 (d, 2H, 3J = 1.6, Pz2) 6.48 (dd, 1H, Pz1), 6.08 (dd, 2H, Pz2) 7.88 (d, 1H, 3J = 2.4, Pz1), 7.58 (d, 2H, 3J = 2.4, Pz2) 4.39 (s, 2H, Ha), 2.46 (d, 1H, 2J = 12.8, 6.8, Hendo), 2.32 (s, 3H, Mea), 2.21 (d, 1H, 2J = 12.8, Hexo), 1.05 (s, 6H, Meb).[Ru(h5-1,3,5-Me3C6H3CN){k3-HB(Pz)3}] 21 8C 8.32 (d, 1H, J = 1.6, Pz1), 7.51 (d, 2H, J = 1.6, Pz2) 6.43 (dd, 1H, Pz1), 6.07 (dd, 2H, Pz2) 7.72 (d, 1H, J = 2.4, Pz1), 7.48 (d, 2H, J = 2.4, Pz2) 4.36 (s, 2H, Ha), 3.28 (s, 1H, Hendo), 2.35 (s, 3H, Mea), 1.18 (s, 6H, Meb). [Ru(h5-1-iPr-4-MeC6H4CN){k3-HC(Pz)3}]- [PF6] 12a 1 12b c 21 8C 9.27, 8.42, 8.33 6.90, 6.54, 6.48 8.59, 8.04, 8.31 9.44 [s, 1H, HC(Pz)3], Isomer a: 6.05 (d, 1H, 3J = 6.3), 5.01 (d, 1H, 3J = 4.8), 4.04 (d, 1H, 3J = 6.0, Hendo), 1.22 (s, 3H, Me2), 1.00 (d, 3H, 3J = 6.9, iPr2), 0.60 (d, 3H, 3J = 6.9, iPr2).Isomer b: 6.04 and 5.09 (d, 1H, 3J = 5.1, AB), 4.04 (d, 1H, 3J = 6.0), 1.43 (d, 3H, 3J = 6.6, iPr), 2.27 (s, 3H, Me1), 1.21 (d, 3H, 3J = 6.6, iPr1). [Ru(h5-1,4-Me2C6H4CN){k3-HC(Pz)3}][PF6] 13d 21 8C 8.39 6.54 8.37 9.40 [s, 1H, HC(Pz)3], cyclohexadienyl: 6.01 and 5.04 (d, 2H, 3J = 4.0, AB), 4.03 (d, 1H, 3J = 5.9), 2.65 (s, 1H, 3J = 6.0, Hc), 2.26 (s, 3H, Me1), 1.21 (s, 3H, Me2).[Ru(h5-C6H7){k3-HB(3,5-Me2Pz)3}] 14 265 8C 2.87 (s, 3H, Pz1), 2.27 (s, 6H, Pz2) 6.05 (s, 1H, Pz1), 5.61 (s, 2H, Pz2) 2.34 (s, 3H, Pz1), 2.19 (s, 6H, Pz2) 6.16 (t, 1H, 3J = 4.67, Ha), 5.18 (dd, 2H, 3J = 5.67, Hb), 2.79 (m, 1H, 2J = 12.82, 3J = 6.54, Hendo), 2.45 (t, 2H, 3J = 6.08, Hc), 1.51 (d, 1H, 2J = 13.41, Hexo). s = singlet, d = doublet, dd = doublet of a doublet, m = multiplet, sep = septet.Pz1 = unique pyrazoyl group, Pz2 = doubly degenerate pyrazoyl group. a Isomer b pyrazoyl signals at 21 8C appear as broad resonances in the baseline. b Only four isomer b pyrazoyl signals are clearly seen due to overlap with those of isomer a (d 8.92, 7.99, 7.68, 6.06). c All pyrazoyl signals broad and equivalent to 1H. d All pyrazoyl signals broad and equivalent to 3H.J. Chem. Soc., Dalton Trans., 1998, 3379–3390 3385 Table 5 Selected 13C NMR data on alkylated ruthenium(cyclohexadienyl)hydridotris(pyrazolyl)-borate and -methane compounds Pyrazolyl borate (d) CDCl3 3- 4- 5- Additional resonances [Ru(h5-1-iPr-4-MeC6H5){k3-HB(Pz)3}] 5a 1 5b 270 8C 144.10, 143.64, 143.40, 143.15, 142.05, 141.53 135.42, 135.19, 134.96, 134.88, 134.72, 134.53 106.41, 106.11, 105.53, 105.42, 105.33, 105.17 Cyclohexadienyl: 108.91, 99.52, 86.72, 82.90, 67.43, 64.38, 46.75, 36.88, 35.17, 34.00 CH2: 32.55, 32.12 Me and iPr: 24.17, 24.02, 23.14, 22.09, 21.95, 21.19, 20.67, 19.70 [Ru(h5-1-iPr-4-MeC6H4CN){k3-HB(Pz)3}] 6a 1 6b 221 8C 145.51, 143.97, 142.07 106.49, 105.78, 105.46 135.58, 135.25, 135.22 CN: 121.69, 120.24 Cyclohexadienyl: 108.99, 99.16, 87.09, 84.58, 70.65, 68.01, 67.43, 44.83, 35.91, 34.29 CHCN: 32.84, 32.05 Me and iPr: 24.30, 23.88, 22.76, 21.84, 21.73, 21.13, 20.96, 20.15 [Ru(h5-1,4-Me2C6H5){k3-HB(Pz)3}] 8 265 8C 143.34, 143.14, 141.59 106.44, 105.53, 105.29 135.33, 134.97, 134.74 Cyclohexadienyl: 99.41, 86.93, 66.92, 35.88, 34.82 CH2: 25.39 Me: 21.90, 22.22 [Ru(h5-1,4-iPr2C6H5){k3-HB(Pz)3}] 9 221 8C 144.47, 143.64, 142.37 106.07, 105.45, 105.19 135.33, 134.91, 135.59 Cyclohexadienyl: 109.35, 83.88, 65.35, 46.77, 34.48 CH2: 33.16 iPr: 32.73, 24.29, 23.20, 21.52, 20.91, 19.84 [Ru(h5-1,3,5-Me3C6H3CN){k3-HB(Pz)3}] 221 8C 141.53, 141.22 105.02, 104.96 134.39, 134.29 CN: 119.11 Cyclohexadienyl: 87.26, 80.06, 40.28 CH2: 33.09 Me: 20.21, 18.38 [Ru(h5-C6H7CN){k3-HB(3,5-Me2Pz)3}] 14 265 8C 154.04, 17.82 152.27, 16.04 107.92 108.42 143.45, 13.27 143.08, 13.09 Cyclohexadienyl: 81.44, 67.97, 28.20 CH2: 16.04 MeC6H5){HB(Pz)3}] 5a and 5b.The observation of two isomers is not surprising as there are two potential sites of attack on the para-cymene ligand, ortho to either a methyl or an isopropyl group. Previous work on similar derivatives containing [2.2]paracyclophane as the spectator ligand reported that isomeric form b was the more prevalent due to steric eVects.1 However in this study the products are formed in similar quantities. The reactions of [Ru(h6-1-iPr-4-MeC6H4){HB(Pz)3}][PF6] with KCN and NaOH yield the related isomeric derivatives 6a/6b and 7a/ 7b.The introduction of a nucleophile onto para-cymene renders the cyclohexadienyl product asymmetric and hence considerably increases the complexity of the room temperature NMR spectrum which is nevertheless consistent with the presence of two isomers of the cyclohexadienyl ligand (Table 4).However once again it is the NMR spectrum of the pyrazolyl protons which are the more intriguing. In the presence of the asymmetric cyclohexadienyl ligand each of the pyrazolyl ring protons is unique, hence each isomer should exhibit nine resonances. Examination of the room temperature 1H NMR spectrum of 5 reveals only nine somewhat broad signals (d 8.84, 7.86, 7.77, 7.64, 7.58, 7.55, 6.45, 6.06, and 5.98) indicating a single isomer. However when the 13C-{1H} NMR spectrum is recorded it is apparent that although there are nine relatively sharp signals there are additional very broad resonances in the base line.Closer examination of the 1H NMR spectrum reveals several very broad additional resonances which can be attributed to the second isomer. Recording the NMR spectrum of 5 at low temperature conclusively demonstrates that both isomers are present as both the 1H and 13C-{1H} spectra clearly show eighteen pyrazolyl environments (Tables 4 and 5).Our interpretation of these observations is that the two isomers have diVerent barriers to rotation. From our variable temperature NMR experiments it seems that the isomers of type b have similar coalescence temperatures to those observed for compounds 1–4 while those for isomeric form a are signifi- cantly higher. Analogous NMR behaviour has been observed for the tris(pyrazolyl)methane compound [Ru(h5-1-iPr-4- MeC6H4CN){k3-HC(Pz)3}][PF6] 12a and 12b. To attempt to substantiate this deduction we prepared and studied by variable temperature NMR the compounds [Ru(h5-1,4-Me2C6H5){k3- HB(Pz)3}] 8 and [Ru(h5-1,4-iPr2C6H5){k3-HB(Pz)3}] 9.Inspection of the room temperature 1H NMR spectrum of 8 reveals seven very broad resonances, two of which each integrate for two protons (Fig. 4) with no resolved coupling in the region d 6.5–8.9. Cooling of the NMR probe results in a sharpening of the signals down to 240 8C, at which point couplings can be clearly observed, with no further changes observed down to 280 8C.If the NMR tube is warmed to ca. 50 8C the spectrum changes such that only two broad resonances, d 7.66 and 6.12, relative integral 2 : 1, are observed. Throughout this 130 8C temperature range the signals for the cyclohexadienyl ligand remained essentially invariant. By contrast, when the room temperature 1H NMR spectrum of 9 is recorded nine sharp well defined pyrazolyl proton resonances (Fig. 4) are observed between 20 and 50 8C.Related behaviour is observed in the variable temperature 13C-{1H} NMR spectra of this pair of compounds. The conclusion to be drawn from these observations must clearly be that the barrier to rotation is higher when the site of nucleophilic atack is adjacent to the isopropyl substituent. Hence in the case of compounds 5, 6, and 7, it is isomer a which is responsible for the sharper signals in the room temperature spectra. When Na[BH4] is reacted with [Ru(h6-1,3,5-Me3C6H3){k3- HB(Pz)3}][PF6] a single product resulting from the addition of a hydride to an unsubstituted aromatic carbon atom is isolated, [Ru(h5-1,3,5-Me3C6H4){k3-HB(Pz)3}] 10.Interestingly the pyrazolyl region of the NMR spectrum displays two sets of sharp H Y Ru N N N N N N X H Y Ru N N N N N N X H H a b X = B, Y = = = X = C, Y = H CN OH CN 5a, 5b 6a, 6b 7a, 7b 12a, 12b3386 J. Chem. Soc., Dalton Trans., 1998, 3379–3390 pyrazolyl signals in integral ratio 1 : 2 (d 8.71, 7.88, 6.48 and 7.74, 7.58, 6.08). Similarly the resonances observed in the room temperature 13C-{1H} NMR spectrum (Table 5) are sharp.Clearly these observations are consistent with a high barrier to rotation and perhaps indicate that it is both the identity of the substituents on the cyclohexadienyl ligand, as previously established, and their number which contribute to the height of the rotational barrier. Finally, each of the experiments described to date has involved unsubstituted pyrazolyl rings.In a final experiment we examined the reaction of the compound [Ru(h6-C6H6){k3- HB(3,5-Me2Pz)3}][PF6], which has substituents on the pyrazolyl Fig. 4 Part of the 1H NMR spectra of 8 and 9 recorded at room temperature. H Y Ru N N N N N N X H H Ru N N N N N N B H H X = B, Y = H 8 X = C, Y = CN 13 9 H H Ru N N N N N N B H H Ru N N N N N N X H H 10 14 rings placed so as to interact with the second ligand, with Na[BH4]. This reaction was not particularly clean and good analytical data could not be obtained on the product, [Ru(h5- C6H7){k3-HB(3,5-Me2Pz)3}] 14.Nevertheless 1H and 13C-{1H} NMR studies reveal that at room temperature the pyrazolyl environments are averaged, due to rapid rotation about the metal–ligand axis, while at low temperature ‘frozen-out’ structure is observed. In the case of this compound the unique pyrazolyl signals in the 1H NMR spectrum occur at d 6.05, 2.87 and 2.34, while the doubly degenerate set occur at d 5.61, 2.27 and 2.19.One might conclude from this observation that, since the barrier appears comparable to that for 1, the barrier height is determined by the nature of the cyclohexadienyl ligand. This hypothesis is explored further in the calculations described below. Computational investigations In order to understand more fully the experimentally observed barriers to rotation of the carbocyclic ligand in 1, 8, 9 and 10, we have carried out a series of extended Hückel molecular orbital (EHMO) calculations on these molecules as well as the parent arene [Ru(h6-C6H6){k3-HB(Pz)3}]1.The electronic structure of [Ru(Á6-C6H6){Í3-HB(Pz)3}]1. A fragment MO energy level diagram for the interaction of a benzene ring with a [Ru{k3-HB(Pz)3}]1 unit is shown in Fig. 5. There are four benzene MOs which lie in the eigenvalue range shown, but only two are expected to interact with the metal fragment. These are the 1a2u p0 and 1e1g p1 orbitals, which are the p orbitals with zero and one vertical nodes respectively.24 The 1b2u and 2e2g levels which lie between the p0 and p1 in energy terms are C–C s antibonding and, in the case of the latter, C–H bonding. They are essentially unaltered on complexation to the [Ru{k3-HB(Pz)3}]1 fragment. Similarly, many of the metal moiety’s orbitals are unaVected by the presence of the benzene ring.The principal [Ru{k3-HB(Pz)3}]1-benzene interaction is in the 16e, 17e and 19e orbitals.The 16e and 17e are mixtures of the [Ru{k3-HB(Pz)3}]1 12e MOs† with the p1 Fig. 5 Fragment EHMO diagram for [Ru(h6-C6H6){k3-HB(Pz)3}]1. † The 12e [Ru{k3-HB(Pz)3}]1 MO is Ru–HB(Pz)3 bonding, with 16% metal character. The 14e [Ru{k3-HB(Pz)3}]1 HOMO is primarily Ru d-based, with some Ru-HB(Pz)3 antibonding character, and the 12a1 [Ru{k3-HB(Pz)3}]1 orbital is almost exclusively Ru dz2.J. Chem. Soc., Dalton Trans., 1998, 3379–3390 3387 orbitals of benzene. The 19e MO has contributions from the 14e HOMO of the [Ru{k3-HB(Pz)3}]1 fragment and the unfilled p2 levels of benzene; it may be regarded as RuÆ benzene backbonding.The HOMO of [Ru(h6-C6H6){k3- HB(Pz)3}]1 is the 16a1 orbital derived primarily from the 12a1 MO of [Ru{k3-HB(Pz)3}]1. It is >90% Ru dz2 in character. Rotation of the benzene ring through 1208 about the Ru– benzene centroid vector in [Ru(Á6-C6H6){Í3-HB(Pz)3}]1. Although we have no experimental measure of the rotational barrier in [Ru(h6-C6H6){k3-HB(Pz)3}]1, other workers have used a variety of techniques to study the barriers in other transition metal–arene systems.25 The rotational barrier in [Cr(h6- C6H6)(CO)3] has been measured by several methods, and estimates range from 15.5–19.7 kJ mol21 at room temperature to 27.5 kJ mol21 at 10 K.A value of 1.2 kJ mol21 for the same compound has been calculated by the extended Hückel approach.26 More recently, variable temperature solid state 2H NMR studies of [Mo(h6-C6D6)2] found that the spectra are invariant in the temperature range 160–360 K, consistent with rapid ring rotation at all accessible temperatures.27 It is therefore highly probable that the barrier in [Ru(h6-C6H6){k3- HB(Pz)3}]1 is also very small, and it is important that we are able to reproduce this computationally.Fig. 6 plots the total EHMO energy of [Ru(h6-C6H6){k3-HB(Pz)3}]1 as the benzene ring is rotated by 1208 in steps of 68, beginning (and ending) in an eclipsed conformation (Fig. 7). As the benzene fragment is rotated by 1208 the total energy passes through two minima (308 and 908) and one maximum (608), the energy of the latter being the same as at 08 and 1208 of rotation, reflecting the six-fold rotational symmetry of the benzene ring. Fig. 6 reveals that the initial, eclipsed conformation is unstable with respect to the staggered geometry obtained when the ring is rotated by 308 and 908, and the barrier on moving from staggeredÆeclipsedÆ staggered is 10.6 kJ mol21.This value is entirely consistent with the experimental data discussed above for [Cr(h6-C6H6)(CO)3] and [Mo(h6-C6D6)2], and is unsurprising in view of the high symmetry of the benzene fragment. Indeed, in treatments of the interactions of planar aromatic carbocyclic ligands with transition metal centres it is common to assume that the metal– ring centroid axis is an infinite axis to rotation.28–30 This may be thought of as the s framework of the ring rotating with the p system remaining fixed.Thus we may conclude that the EHMO calculations predict a very small energy barrier to rotation of the benzene about the Ru–benzene centroid vector, due to the interaction of the benzene p system with the Ru fragment orbitals being essentially unaltered as the ring is rotated. The electronic structure of 1. A fragment MO diagram for the interaction of C6H7 2 with [Ru{k3-HB(Pz)3}]1 is given in Fig. 8. The orbitals of the cyclohexadienyl fragment are labelled according to the irreducible representation of the Cs point group that they span.It may be seen that, in addition to the 8a9 p0, 6a0 p1b and 10a9 p1nb levels,‡ there are a further three cyclohexadienyl orbitals in the energy range shown (4a0, 5a0 and 9a9). The 9a9 and 5a0 levels are C–C and C–H s bonding, while the 4a0 orbital is mainly C–C s antibonding with a smaller C–H bonding component. None of these s levels interacts to any significant extent with the Ru atom.The principal [Ru{k3- HB(Pz)3}]1–cyclohexadienyl interaction is concentrated in the 34a9 and 20a0 orbitals although there is some [Ru{k3- HB(Pz)3}]1–cyclohexadienyl orbital mixing in other levels. The 20a0 MO is made up of a mixture of the C6H7 2 6a0 p1b level and the 13e orbital of the metal fragment, while the 34a9 orbital is a combination of C6H7 2 10a9 p1nb and several Ru AOs (dz2, dxz and px). The 15e lowest unoccupied MO (LUMO) of the metal fragment (not shown) is also a contributor to this level.The ‡ The p1b and p1nb MOs are the cyclohexadienyl p orbitals with one vertical node, which are C–C bonding and non-bonding respectively. 36a9 HOMO of 1 is largely derived from the 12a1 MO of the [Ru{k3-HB(Pz)3}]1 unit, and is therefore predominantly Ru dz2 in character. Rotation of the cyclohexadienyl ring through 1208 about the Ru–cyclohexadienyl centroid vector in 1. The cyclohexadienyl ring of 1 was rotated by 1208 in a manner analogous to the Fig. 6 Variation of the total EHMO energy of [Ru(h6-C6H6){k3- HB(Pz)3}]1 as the benzene ring is rotated by 1208 in steps of 68, beginning in an eclipsed conformation. Fig. 7 Rotational motion of the benzene ring around the Ru–ring centroid vector in [Ru(h6-C6H6){k3-HB(Pz)3}]1 (---- denotes a Pz ring and d follows one of the C atoms during the course of the rotation). Fig. 8 Fragment EHMO diagram for 1.3388 J. Chem. Soc., Dalton Trans., 1998, 3379–3390 benzene ring in [Ru(h6-C6H6){k3-HB(Pz)3}]1, and the resulting energy change is shown in Fig. 9. Clearly the energy change that accompanies rotation of the cyclohexadienyl ring is both qualitatively and quantitatively diVerent from that which occurs during the benzene ring rotation (Fig. 6). In the case of the cyclohexadienyl, 08 of rotation corresponds to the most stable geometry, and the total energy goes through a single maximum at 608. This maximum energy is 58.1 kJ mol21 above the most stable orientation, a diVerence which is significantly greater than in the benzene case. This result supports the experimental evidence from our NMR studies discussed previously.Before attempting a more detailed explanation of the origin of the much larger rotational barrier in 1, it is worth noting that the lower symmetry of cyclohexadienyl (Cs) with respect to benzene (D6h) means that there is no infinite axis to rotation in the cyclohexadienyl moiety. Put another way, the interaction of the cyclohexadienyl ring with the metal fragment is much more dependent on the orientation of the ring than is the case in the benzene molecule, resulting in greater energy changes during rotation.In extended Hückel theory, the total energy of a molecule is given by the sum of the energies of its MOs multiplied by their occupation numbers. Thus the total energy change shown in Fig. 9 is the net eVect of the energy changes of the 69 EHMOs of 1.Clearly we cannot analyse each of these in detail, but we can make progress through the observation that it is the four highest occupied orbitals which experience the greatest changes during rotation of the cyclohexadienyl ring. Indeed the sum of the changes in the energies of the four highest occupied MOs Fig. 9 Variation of the total EHMO energy of 1 as the cyclohexadienyl ring is rotated by 1208 in steps of 68, beginning in an eclipsed conformation. Fig. 10 Variation of the EHMO energies of the 23a0–36a9 levels of 1 as the cyclohexadienyl ring is rotated by 1208 in steps of 68, beginning in an eclipsed conformation. account for 63% of the rotational barrier, and Fig. 10 plots the energies of the 23a0–36a9 levels as the cyclohexadienyl ring is rotated. The 36a9 HOMO is most aVected by the ring rotation, mirroring the total energy change (Fig. 9). At the starting geometry, this orbital is largely Ru dz2. However, as the ring is rotated the composition of the 36a9 MO changes significantly, to the extent that at the least stable geometry (608) it is a mixture of the 10a9 level and the 14e and 15e orbitals of the [Ru{k3-HB(Pz)3}]1 fragment.The destablisation of the 36a9 orbital may therefore be traced to the increasing contribution Fig. 11 Fragment EHMO diagram of the highest occupied orbitals of 1 at (a) 08 and (b) 608 of rotation of the cyclohexadienyl ring about the Ru–ring centroid vector.J. Chem. Soc., Dalton Trans., 1998, 3379–3390 3389 of the 15e LUMO of the metal fragment, which is much higher in energy than the Ru dz2 orbital. Fig. 11 presents fragment MO diagrams for the highest occupied MOs of 1 with the cyclohexadienyl ring at 08 (a) and 608 (b) of rotation. As discussed above, the 36a9 HOMO is strongly destabilised by the involvement of the 15e [Ru{k3-HB(Pz)3}]1 fragment LUMO. By contrast, the energy of the 35a9 orbital remains largely unaltered as the ring is rotated, although its composition changes significantly.At 08 it is primarily derived from the 14e MO of the metal fragment, but at 608 it has a 90% contribution from the mainly Ru dz2 12a1 [Ru{k3-HB(Pz)3}]1 fragment level. Thus as the dz2 character of the 36a9 orbital decreases, that of the 35a9 increases. Furthermore, as the 36a9 MO gains [Ru{k3-HB(Pz)3}]1 15e character, that of the 34a9 decreases such that, by 608 of rotation, it is derived largely from the 14e [Ru{k3-HB(Pz)3}]1 fragment orbital.The 23a0 MO is significantly stabilised upon cyclohexadienyl rotation. At 08, it has ca. 70% [Ru{k3-HB(Pz)3}]1 14e character with only very small contributions from the cyclohexadienyl ligand. At 608, however, it acquires appreciable cyclohexadienyl 10a9 character, as well as having [Ru{k3-HB(Pz)3}]1 14e and 13e content. It becomes Ru–cyclohexadienyl bonding upon rotation, which accounts for its energetic stabilisation. Thus we can see that the rotation of the carbocycle in 1 has a much greater eVect upon the valence electronic structure and total energy than in [Ru(h6-C6H6){k3-HB(Pz)3}]1.A much larger rotational energy barrier is calculated in 1, in agreement with experimental data, and that energy barrier is largely explained by the energy and composition variations of the highest occupied MOs. By contrast, the electronic structure of [Ru- (h6-C6H6){k3-HB(Pz)3}]1 is almost unaVected by ring rotation on account of the high symmetry of the benzene fragment.Rotation of the cyclohexadienyl ring through 1208 about the Ru–cyclohexadienyl centroid vector in 8, 9 and 10. Our NMR studies indicate that the magnitude of the rotational barrier in the substituted cyclohexadienyl compounds depends on both the position and size of the substituents. We have therefore attempted to analyse this eVect computationally by repeating the ring rotation studies of 1 for 8, 9 and 10. Before discussing these results, we must highlight a problem that we encountered upon ring substitution. Replacement of a ring H with, for example, a methyl group produced unrealistically high barriers in systems which are known to have small barriers.For example, replacement of benzene by mesitylene in [Ru(h6- C6H6){k3-HB(Pz)3}]1 caused a 12-fold increase in the calculated rotational barrier, which is clearly incompatible with experimental conclusions.25 Similar increases were observed in the substituted cyclohexadienyl rings. In order to establish the origin of these very high barriers we conducted a series of calculations in which the ring substituents in [Ru(h6-C6H5R)- {k3-HB(Pz)3}]1 (R = Me, iPr) were bent out of the plane of the p system and away from the metal fragment.The barriers quickly reduced and returned to sensible values at about 108 of bending. It would appear that the cause of the artificially high barriers is an interaction between the H atoms on the carbocyclic ring substituents and those of the Pz rings of the HB(Pz)3 ligand which are directed toward the carbocycle.In all our studies of 8, 9 and 10, therefore, the Me and iPr substituents are bent by 108 out of the C5 plane. Clearly this prevents quantitative comparison with our calculated barriers for 1, but we feel that we are still justified in comparing barriers within the series 8, 9 and 10. The form of the total EHMO energy change of 8 as the cyclohexadienyl ring is rotated by 1208 about the Ru–ring centroid axis is very similar to the analogous unsubstituted compound (1), with a single maximum at 608 of rotation.The energy barrier (71.8 kJ mol21) is also quite close to that for 1 but, as we have already discussed, the imposed bending of the methyl groups in 8 precludes any direct comparison with 1. The total EHMO variation on 1208 rotation of the carbocycle in the bis isopropylated derivative, 9, is shown in Fig. 12. Clearly this graph is rather diVerent from Fig. 9, with two minima (at 308 and 908) separated by two maxima (at 0/1208 and 608). The largest energy diVerence is between 608 and 908 (steps c and d on Fig. 13) and is 118.4 kJ mol21, 1.65 times greater than the barrier in 8. Thus the calculated barrier is appreciably greater in the bis isopropylated compound than the bis methylated one, in agreement with the conclusions drawn from our NMR studies. At this point we must address two questions. The first is why the form of the plots is diVerent for bis isopropylated and bis methylated cyclohexadienyl and the second is why there is a higher rotational barrier in 9 than in 8.Addressing the second question first, a logical way to proceed is to analyse the variations in the energies of the MOs of 8 and 9 in an analogous manner to our earlier treatment of 1. Unfortunately this is not a productive approach as the variations are very similar in all three molecules 1, 8 and 9, with the combination of the highest occupied orbitals accounting for a substantial part of the barrier in each case.This is unsurprising in that it is well known that substitution of carbocyclic ring H atoms by R groups does not significantly aVect the valence electronic structure of the p system, except to produce a general raising (or lowering) of the MO energies. Given the lack of a clear cut electronic explanation, it is helpful to consider other possible causes of the diVerences in barrier heights and total energy plots for 8 and 9, in particular the steric bulk of Me vs.iPr. Fig. 13 suggests that initial rotation of the cyclohexadienyl ring in 8 and 9 might be expected to result in a stabilisation, as Me1/iPr1 moves away from its eclipsed position over one of the Pz rings. There is little or no change in the total energy of 8 for this distortion, although Fig. 12 shows that the analogous rotation for the bis isopropylated ring indeed produces a significant stabilisation. If we assume that the electronic eVects are approximately the same for both rings, we may trace the diVerence in the total energies of 8 and 9 for the first 308 of rotation to the bulk of the R substituent in the 3-position, i.e.that the relief of the iPr1/Pz eclipsing is more energetically favourable than that of the Me1/Pz. Fig. 12 Variation of the total EHMO energy of 9 as the cyclohexadienyl ring is rotated by 1208 in steps of 68, beginning in an eclipsed conformation. Fig. 13 The conformations adopted by the cyclohexadienyl ring of 8 and 9 at 08, 308, 608, 908 and 1208 of rotation about the Ru–ring centroid vector (---- denotes a Pz ring).3390 J.Chem. Soc., Dalton Trans., 1998, 3379–3390 Between 308 and 608 the total energy of both 8 and 9 becomes significantly less negative (608 is the least stable geometry in both cases). Now both the electronic component and steric component of the barrier are working together, the latter being unfavourable as Me2/iPr2 come into an eclipsed geometry with a Pz ring.From 608 to 908 the electronic and steric components again reinforce one another, but this time to stabilise the molecules. The stabilisation is greater for 9 than 8 as the relief of the iPr2/Pz eclipsing is greater than that of the Me2/Pz. Finally, from 908 to 1208 the eclipsing of Me1/Pz does not greatly aVect the total energy of 8, in contrast to 9 where the interaction of the bulkier iPr1 with the Pz ring results in a destabilisation of the molecule.The variation in the total EHMO energy of 10 as the cyclohexadienyl ring is rotated through 1208 is very similar to that of 1 (Fig. 9) and 8. However, the size of the rotational barrier is calculated to be 156.5 kJ mol21, 2.18 times greater than that of 8 and 1.32 times that of 9. This result is again entirely consistent with our NMR experiments, which found that the pyrazolyl signals were sharp at room temperature, implying that 10 has the highest rotational barrier of all the compounds studied.Extension of our previous arguments for 8 and 9 readily explains the high barrier in 10. If we once again assume that the electronic contribution is similar to that in 1, then the extra barrier height of 10 arises from the movement of the three methyl groups from staggeredÆeclipsedÆstaggered in the course of the 1208 rotation. Conclusions In this contribution we have described the results of combined experimental and theoretical studies of the cyclohexadienyl compounds that result from nucleophilic attack on [Ru(h6- arene){k3-HB(Pz3)}]1 and [Ru(h6-arene){k3-HC(Pz3)}]21.Thus, reaction of [Ru(h6-arene){k3-HB(Pz3)}][PF6] with H2, CN2 or OH2 leads to the cyclohexadienyl compounds 1, 2 and 3, and the nucleophile has been shown to enter along an exo pathway. Similarly, reaction of [Ru(h6-arene){k3-HC(Pz3)}][PF6]2 with CN2 forms 11, which has been characterised crystallographically. Variable temperature NMR experiments indicate that one of the ligands in the cyclohexadienyl compounds undergoes restricted rotation about the metal–ligand axis.This is consistent with the crystal structure of 11, which reveals that in the solid state the complex cation [Ru(h5-C6H6CN){k3-HC(Pz)3}]1 adopts a conformation in which one Pz ring is eclipsed with the projection of the sp3 hybridised C atom of the cyclohexadienyl moiety. Furthermore, variable temperature NMR studies on the cyclohexadienyl compounds that result from the attack of nucleophiles on the substituted arene compounds [Ru(h6- arene){HB(Pz)3}][PF6] (arene = 1-iPr-4-MeC6H4, 1,4-Me2C6H4, 1,4-iPr2C6H4 or 1,3,5-Me3C6H3) indicate that the magnitude of the rotational barrier in the cyclohexadienyl compounds is dependent both on the position and number of the substituents on the cyclohexadienyl ring.That it is the cyclohexadienyl ring that is undergoing the restricted rotation is indicated by variable temperature NMR studies of 14 (in which the Pz rings are substituted but the cyclohexadienyl ring is not) which displays a similar coalescence temperature (and therefore a similar rotational barrier) to 1.Extended Hückel calculations have been used to investigate the MO structure of [Ru(h6-C6H6){k3-HB(Pz)3}]1 and 1 in terms of the [Ru{k3-HB(Pz)3}]1 and carbocyclic fragments. Subsequent calculations in which the carbocycle was rotated through 1208 about the Ru–ring centroid vector revealed a very small energy barrier to rotation in the benzene compound but a significantly greater barrier in the cyclohexadienyl system.This observation is entirely consistent with experiment. The very low barrier in the benzene compound is a consequence of the high symmetry of the benzene ligand, in that the interaction of the ring with the Ru fragment is essentially unaltered as the ring is rotated. By contrast, the lower symmetry of the cyclohexadienyl ring means that its interaction with the [Ru{k3-HB(Pz)3}]1 unit is much more dependent upon their relative orientation.More specifically, the greater barrier height in 1 has been analysed in terms of the changes in the composition of the highest occupied MOs during ring rotation. The HOMO is strongly destabilised, on account of the increasing contribution of the high lying LUMO of the [Ru{k3-HB(Pz)3}]1 fragment. Calculations on the substituted cyclohexadienyl compounds 8 and 9 also support experiment in finding a greater rotational barrier for the bis isopropylated derivative 9 in comparison with the bis methylated compound 8. This diVerence in barrier height is traced to the greater steric bulk of iPr vs. Me, as the electronic eVects are found to be very similar in both cases. Finally, the calculated barrier for ring rotation in the tris methylated compound 10 is greater than any of the other systems studied, once again agreeing with the conclusions drawn from our variable temperature NMR data. The cause of this increased barrier height is also attributed to steric eVects. Acknowledgements We thank Johnson Matthey plc for generous loans of ruthenium trichloride and University College London for financial support (S. B.). References 1 M. R. J. Elsegood, J. W. Steed and D. A. Tocher, J. Chem. Soc., Dalton Trans., 1992, 1797. 2 J. W. Steed and D. A. Tocher, J. Chem. Soc., Dalton Trans., 1993, 3187. 3 S. G. Davies, M. L. H. Green and D. P. M. Mingos, Tetrahedron, 1978, 34, 3047. 4 T. S. Cameron, M. D. Clerk, A. Linden, K. C. Sturge and M. J. Zaworotko, Organometallics, 1988, 7, 2571. 5 J. F. Helling and G. G. Cash, J. Organomet. Chem., 1974, 73, C10. 6 D. R. Robertson, I. W. Robertson and T. A. Stephenson, J. Organomet. Chem., 1980, 202, 309. 7 C. C. Neto and D. A. Sweigart, J. Chem. Soc., Chem. Commun., 1990, 1703. 8 V. S. Kaganovich, A. R. Kudinov and M. I. Rybinskaya, J. Organomet. Chem., 1987, 323, 111. 9 H. LeBozec, D. Touchard and P. H. Dixneuf, Adv. Organomet. Chem., 1989, 29, 163. 10 S. Bhambri and D. A. Tocher, Polyhedron, 1996, 15, 2763. 11 S. Bhambri and D. A. Tocher, J. Chem. Soc., Dalton Trans., 1997, 3367. 12 S. Bhambri and D. A. Tocher, J. Organomet. Chem., 1996, 507, 291. 13 G. M. Sheldrick, SHELXS-86, Acta Crystallogr., Sect. A, 1990, 46, 467. 14 G. M. Sheldrick, SHELXL-93, University of Göttingen, 1993. 15 C. Mealli and D. M. Proserpio, J. Chem. Educ., 1990, 67, 399. 16 R. HoVmann and W. N. Lipscomb, J. Chem. Phys., 1962, 36, 2179. 17 R. HoVmann, J. Chem. Phys., 1963, 39, 1397. 18 R. S. Mulliken, J. Chem. Phys., 1955, 23, 1833, 1841, 2338, 2343. 19 Z. Shirin, R. Mukherjee, J. F. Richardson and R. M. Buchanan, J. Chem. Soc., Dalton Trans., 1994, 465. 20 R. T. Swann, A. W. Hanson and V. Boekelheide, J. Am. Chem. Soc., 1986, 108, 3324. 21 G. Winkhaus, L. Pratt and G. Wilkinson, J. Chem. Soc., 1961, 3807. 22 M. Keys, V. G. Young and W. B. Tolman, Organometallics, 1996, 15, 4133. 23 W. D. Jones and E. T. Hessell, Inorg. Chem., 1991, 30, 781. 24 F. A. Cotton, Chemical Applications of Group Theory, Wiley- Interscience, New York, 3rd edn., 1991. 25 D. Braga, Chem. Rev., 1992, 92, 633. 26 T. A. Albright, P. Hofman and R. HoVmann, J. Am. Chem. Soc., 1977, 99, 7546. 27 D. O’Hare, S. J. Heyes, S. Barlow and S. Mason, J. Chem. Soc., Dalton Trans., 1996, 2989. 28 K. D. Warren, Struct. Bonding (Berlin), 1976, 27, 45. 29 J. G. Brennan, G. Cooper, J. C. Green, N. Kaltsoyannis, M. A. MacDonald, M. P. Payne, C. M. Redfern and K. H. Sze, Chem. Phys., 1992, 164, 271. 30 J. C. Green, N. Kaltsoyannis, M. A. MacDonald and K. H. Sze, J. Am. Chem. Soc., 1994, 116, 1994. Paper 8/04254C

ISSN:1477-9226    

DOI:10.1039/a804254c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3391–3396 3391 Synthesis and structural characterisation of ruthenium carbonyl clusters containing hydrocarbyl ligands derived from (hydroxy)alkynes HC]] CC(H)(OH)CH3 Cindy Sze-Wai Lau and Wing-Tak Wong * Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong, P.R.China. E-mail: wtwong@hkucc.hku.hk Received 10th July 1998, Accepted 2nd September 1998 Five new ruthenium ketonic clusters [Ru4(CO)12{m4-h1 :h1 :h2 :h2-CH3C(H)CCC(H)CH2C(O)CH3}] 1 (10%), [Ru5(CO)14(m-H)2{m4-h1 :h1 :h2 :h2-CC(H)C(O)CH3}] 2 (15%), [Ru6(CO)16{m4-h1 :h1 :h2 :h2-CC(H)C(O)CH3}2] 3 (15%), [Ru7(CO)19{m4-h1 :h1 :h2 :h2-CC(H)C(O)CH3}{m5-h1 :h1 :h1 :h1 :h2-CCC(H)CH3}] 4 (8%), and [Ru5(CO)12- (m-CO){m4-h1 :h1 :h2 :h2-CC(H)C(O)CH3}(m4-h2-HCCCH2CH3)] 5 (10%) have been synthesised by reaction of but-3-yn-2-ol with triruthenium dodecacarbonyl, in cyclohexane, under reflux.All the compounds have been fully characterised by spectroscopic and X-ray diVraction methods. The structure of 1 is based on a Ru4 butterfly skeleton containing a fragment of C8 ketonic chain which arises from the coupling of two ligand molecules with the elimination of a water molecule.An interesting feature in clusters 2, 3, 4 and 5 is the formation of a metallocycloketonic ring with a m4-h1 :h1 :h2 :h2 coordination mode which is derived from the activation of the C]] ] C triple bond. Both 2 and 5 consist of a wingtip bridged butterfly core, which is also bonded with two bridged hydrides in 2 and a m4-h2 acetylide fragment in 5 respectively.Cluster 2 is also closely related to cluster 3, in that it seems to be a monomeric unit of 3 forming a six atom raft geometry with two metallocycloketonic rings. The metal core of 4 is similar to 3, except that one Ru–Ru bond in 4 is broken to form two more metal–metal bonds with an additional Ru atom to give a novel Ru7 core, which is best described as a distorted Ru4 square plane sharing an edge with an edge-bridged butterfly.Moreover, one of the metallocycloketonic rings in 3 is replaced by an allenyl CCC(H)CH3 ligand, which is coordinated to an edge-bridged butterfly in 4. Introduction The chemistry of transition metal clusters containing functionalised alkynes has received considerable attention.1–8 Early work in acetylenic alcohols, such as HC]] ] CCRR9(OH), where R = alkyl and R9 = aryl groups, respectively, have been shown to react with triruthenium and triosmium clusters leading to hydroxy-functionalised alkyne clusters with the general formula [HM3(CO)9(m3-C]] ] CCRR9OH)].9,10 The metal atoms have a great influence on the reactivity of the hydroxy function to allow reactions occuring at the side chain, such as acid-induced dehydration to give [HRu3(CO)9(m3-C]] ] CCPh]] CH2)],9 acidcatalysed isomerisation to give [HOs3(m-OH)(CO)9(m3-C]] C]]CPh2)],11 and cyclization with the formation of an oxygencontaining ‘C4O’ ring.12,13 We report here the synthesis and structural characterization of five clusters obtained in moderate yield from the reaction of [Ru3(CO)12] with but-3-yn-2-ol, HC]] ] CCH(OH)CH3.A salient feature in this reaction is that formation of these new complexes involves extensive rearrangement of ligands, namely, dehydration,14 hydrogen atom transfer from carbon to metal atom, C]] ] C triple bond activation,15,16 C–C coupling,17–21 and cyclization with metal atoms.22,23 Results and discussion The reaction of [Ru3(CO)12] with an excess of but-3-yn-2-ol in refluxing cyclohexane solution aVorded five complexes, namely, [Ru4(CO)12{m4-h1:h1:h2:h2-CH3C(H)CCC(H)CH2C(O)CH3}] 1, [Ru5(CO)14(m-H)2{m4-h1:h1:h2:h2-CC(H)C(O)CH3}] 2, [Ru6- (CO)16{m4-h1:h1:h2:h2-CC(H)C(O)CH3}2] 3, [Ru7(CO)19{m4- h1:h1:h2:h2-CC(H)C(O)CH3}{m5-h1:h1:h1:h1:h2-CCC(H)CH3}] 4, and [Ru5(CO)12(m-CO){m4-h1:h1:h2:h2-CC(H)C(O)CH3}(m4- h2-HCCCH2CH3)] 5 (Scheme 1).Purification was accomplished by preparative TLC and their respective yields were 10, 15, 15, 8 and 10% [based on Ru3(CO)12]. Crystals of 1–5 suitable for X-ray determinations were obtained by slow evaporation from CH2Cl2–n-hexane solutions at 210 8C for 2 d. They were fully characterized by FAB mass spectrometry, IR, 1H NMR and 13C NMR spectroscopies and single-crystal X-ray diVraction techniques. Spectroscopic and structural characterization of 1 The positive FAB mass spectrum of 1 displays a parent molecular ion peak at m/z 862, consistent with twelve terminal carbonyl ligands and one ligated olefinic fragment.The 1H NMR spectrum recorded in CDCl3 under ambient conditions shows one triplet at d 4.32 and a quartet at d 2.71, indicative of two olefinic protons. The doublet at d 4.29 and the singlet at d 2.18 correspond to the methylene protons and methyl protons respectively. Moreover, the IR spectrum reveals the presence of terminal carbonyl ligands only.An X-ray diVraction analysis of 1 was undertaken and an ORTEP diagram is shown in Fig. 1 together with the atomic numbering scheme. The selected bond angles and distances are presented in Table 1. The four Ru atoms adopt a butterfly arrangement with twelve terminally bonded carbonyl ligands. The dihedral angle between the two coupled ruthenium triangles is 150.098. The ligand capped on the butterfly framework is diVerent to those observed in other related Ru4 clusters such as [Ru4(CO)12(MeC]] CMe)],24 [Ru4(CO)12(m4-h2-HC2Ph)] 25 with a m4-h2 coordination mode or [Fe4(CO)12(m4-CO)]2, [Ru4H(CO)12(CCPhCHPh)] with the interstitial carbon atom bonding in a m4-fashion.26,27 The two functionalized alkyne units are coupled to yield an octa-2,4- diene ketonic chain with the loss of a water molecule.The two central carbon atoms C(16) and C(15) are s bonded to Ru(2) [Ru(2)–C(16) 2.05(3) Å] and Ru(3) [Ru(3)–C(15) 2.04(3) Å],3392 J. Chem.Soc., Dalton Trans., 1998, 3391–3396 respectively. In addition, this ligand is p bonded to Ru(1) via C(16)–C(17) [Ru–C 2.29(3) and 2.35(3) Å] and to Ru(4) via C(14)–C(15) [Ru–C 2.31(3) and 2.29(3) Å]. The bonding mode of the C4 ligand in 1 behaves in a same manner as that found in the Ru4 square planar cluster [Ru4(CO)10(m4-PPh)- (m-CO){m4-h1,h1,h2,h2-PhC(H)CCC(H)Ph}].28 There are also other examples of clusters, such as [Ru4(CO)10(m4-PPh)- (m4-h1 :h1 :h3 :h3-PhC4Ph)] and [Ru4(CO)10(m4-PC]] ] CBut)- (m4-h1 :h1 :h3 :h3-ButC4But)],29 containing an unsaturated C4 hydrocarbon backbone capped on a Ru4 square plane in a similar bonding mode.Compound 1 has an electron count of 62 in which the ligand donates six electrons toward skeletal bonding. Ru Ru Ru Ru Ru Ru Ru Ru Ru Ru Ru C Ru C C C O C Ru C Ru Ru Ru Ru CHc C C C Ru Ru Ru Ru Ru C O C C CHb 3 Ha Hd Hc OH CH3 H HC C C C CHd 3 O Ru Ru Ru Ru Ru C C C C CHe 3 Hc C C C O Hb1 Hb2 Hd 3C Ha Hb Hd Hb C CHa 3 C Ha 3C CHc 3 O C 3 4 CHe 3 1 O 2 5 Ru3(CO)12 + Ha 2C C O Hb C CHd 3 Ha Scheme 1 Table 1 Some selected bond lengths (Å) and angles (8) for cluster 1 Ru(1)–Ru(2) Ru(2)–Ru(3) Ru(3)–Ru(4) Ru(1)–C(17) Ru(3)–C(15) Ru(4)–C(15) C(13)–C(14) C(15)–C(16) C(17)–C(18) C(19)–C(20) Ru(1)–Ru(2)–Ru(4) Ru(1)–Ru(3)–Ru(2) Ru(2)–Ru(3)–Ru(4) C(14)–C(15)–C(16) C(16)–C(17)–C(18) 2.770(3) 2.848(3) 2.789(3) 2.35(3) 2.04(3) 2.29(3) 1.50(4) 1.40(4) 1.49(4) 1.47(4) 113.9(1) 57.67(8) 61.38(8) 125(2) 124(2) Ru(1)–Ru(3) Ru(2)–Ru(4) Ru(1)–C(16) Ru(2)–C(16) Ru(4)–C(14) C(19)–O(13) C(14)–C(15) C(16)–C(17) C(18)–C(19) Ru(1)–Ru(2)–Ru(3) Ru(1)–Ru(3)–Ru(4) Ru(2)–Ru(1)–Ru(3) Ru(2)–Ru(4)–Ru(3) C(15)–C(16)–C(17) 2.895(3) 2.878(3) 2.29(3) 2.05(3) 2.31(3) 1.23(3) 1.40(3) 1.41(4) 1.53(4) 62.01(8) 112.8(1) 60.31(8) 60.32(8) 125(2) Spectroscopic and structural characterization of 2 The 1H NMR spectrum of 2 shows two singlet signals at d 4.52 and 2.02 with an integral ratio of 1 : 3 which are assigned to the alkyne proton and methyl protons respectively.Besides, there are two high field singlets at d 221.31 and 228.45 in a ratio of 1:1 due to the presence of two bridging hydrides. The positive FAB mass spectrum exhibits a peak envelope at m/z 967, Fig. 1 Molecular structure of [Ru4(CO)12{m4-h1:h1:h2:h2-CH3C(H)- CCC(H)CH2C(O)CH3}] 1 with the atom numbering scheme. Fig. 2 Molecular structure of [Ru5(CO)14(m-H)2{m4-h1:h1:h2:h2- CC(H)C(O)CH3}] 2 with the atom numbering scheme. Table 2 Some selected bond lengths (Å) and angles (8) for cluster 2 Ru(1)–Ru(2) Ru(1)–Ru(4) Ru(2)–Ru(5) Ru(4)–Ru(5) Ru(1)–C(15) Ru(3)–C(15) Ru(4)–C(15) C(16)–C(17) C(17)–C(18) C(15)–C(16)–C(17) C(16)–C(17)–O(15) 2.972(1) 2.874(1) 3.056(1) 2.825(1) 2.096(9) 1.088(9) 2.135(9) 1.44(1) 1.48(1) 114.4(8) 119.4(9) Ru(1)–Ru(3) Ru(2)–Ru(4) Ru(3)–Ru(5) Ru(1)–O(15) Ru(3)–C(16) Ru(5)–C(15) C(15)–C(16) C(17)–O(15) C(16)–C(17)–C(18) O(15)–C(17)–C(18) 2.829(1) 2.773(1) 2.866(1) 2.125(6) 2.204(9) 2.143(9) 1.47(1) 1.26(1) 121.5(9) 119.1(9)J.Chem. Soc., Dalton Trans., 1998, 3391–3396 3393 consistent with an isotopic distribution of five Ru atoms. Furthermore, the IR spectrum reveals the presence of terminal carbonyl ligands only. The molecular structure of 2 is illustrated in Fig. 2 and some relevant bond parameters are collected in Table 2. The bond lengths of the unbridged Ru–Ru edges [average 2.83(3) Å] are found to be shorter than the hydride-bridged Ru(1)–Ru(2) and Ru(2)–Ru(5) [average 3.014(1) Å] bonds, which is a common observation in cluster chemistry.30 The apical Ru(2) atom lies 2.33 Å above the best plane through Ru(1), Ru(3), Ru(4) and Ru(5) but is asymmetrically located with respect to the Ru(1) and Ru(5) [Ru(1)– Ru(2) 2.972(1), Ru(2)–Ru(5) 3.056(1) Å].The alkyne ligand bonded to the cluster core adopts an unusual m4-h1:h1:h2:h2 mode via the C]] ] C triple bond activation to form a fivemembered metallocycloketonic ring involving Ru(1).The semiinterstitial carbide C(15) atom can be viewed as quadruply bridged across a distorted square base. It is however more strongly bonded to Ru(3) and Ru(4) [average Ru–C(15) 2.092 Å] than to Ru(4) and Ru(5) [average Ru–C(15) 2.139 Å], resulting in a slight displacement of this atom towards the Ru(3)– Ru(4) edge. The 13C NMR studies reveal a singlet at d 323 which is attributed to this carbido carbon. In [Ru5(m5-C3PhCH)(m- SMe)2(m-PPh2)2(CO)10], a similar resonance parameter was observed for this kind of carbido atom.31 Another interesting feature in this compound is the oxidation of the hydroxyl group of the alkyne ligand to a ketone group which is then coordinated to the Ru(1) atom.The Ru(1)–O(15) distance 2.125(6) Å is typical for the OÆRu dative bond observed in the pentanuclear clusters [Ru5(m5-C)(CO)13{C2H2(CO2Me)2}] 32 and [Ru5(CO)10- (m-Br)(m-PPh2)2{m5-CCC(O)CH2CH]] CH2}].33 A singlet at d 224 in the 13C NMR spectrum of 2 can be assigned to this keto carbon.If the alkyne ligand in 2 contributed six electrons to cluster bonding, the total electron count for the cluster would Fig. 3 Molecular structure of [Ru6(CO)16{m4-h1:h1:h2:h2-CC(H)C(O)- CH3}2] 3 with the atom numbering scheme. Table 3 Some selected bond lengths (Å) and angles (8) for cluster 3 Ru(1)–Ru(2) Ru(2)–Ru(4) Ru(3)–Ru(5) Ru(5)–Ru(6) Ru(2)–C(17) Ru(3)–C(17) Ru(4)–C(17) Ru(5)–C(21) Ru(6)–C(22) C(18)–C(19) C(19)–C(20) C(22)–C(23) C(23)–C(24) C(18)–C(19)–O(17) C(21)–C(22)–C(23) C(22)–C(23)–C(24) 2.780(2) 2.910(2) 2.820(2) 2.790(2) 2.14(2) 2.04(2) 2.11(2) 2.22(2) 2.22(2) 1.43(2) 1.48(3) 1.39(2) 1.51(3) 120(1) 114(1) 118(1) Ru(1)–Ru(3) Ru(3)–Ru(4) Ru(4)–Ru(6) Ru(1)–C(17) Ru(2)–C(18) Ru(3)–C(21) Ru(4)–C(21) Ru(6)–C(21) C(17)–C(18) C(19)–O(17) C(21)–C(22) C(23)–O(18) C(17)–C(18)–C(19) C(18)–C(19)–C(20) C(22)–C(23)–O(18) 2.810(2) 2.656(2) 2.917(2) 2.25(2) 2.20(2) 2.03(2) 2.10(2) 2.17(2) 1.43(2) 1.24(2) 1.49(2) 1.22(2) 115(1) 121(1) 123(1) be 76 CVE, which is consistent with a wingtip-bridged butterfly skeleton.34–37 Spectroscopic and structural characterization of 3 The chromatographic separation of the reaction mixture yielded the third fraction which gave dark red crystals of 3 after recrystallization. The Ru6 core geometry in 3 is best described as a twisted ladder.A perspective drawing of cluster 3 with the atomic numbering scheme is shown in Fig. 3 and some selected bond parameters are listed in Table 3.In contrast to the typical giant rhombic six-atom raft geometry with nine M–M bonds,38–41 the metal skeleton of 3 comprises two distorted square planar units with a common edge, Ru(3)–Ru(4). The Ru–Ru bond lengths lie in the range 2.656(2)–2.917(2) Å. The Ru(3)–Ru(4) distance is significantly shorter than any of the other Ru–Ru bonds which may reflect the presence of electron unsaturation 41,42 and the steric requirements of the bridging alkyne ligands.Two alkyne ligands with a bonding mode similar to that observed in 2 were coordinated to the metal plane on the opposite side in order to minimize steric repulsion. The 1H NMR spectrum of 3 shows signals that are rather similar to those observed in cluster 2 since the same bridging ligands are associated with the metal core in each case. However, no hydride signal was detected for 3. Although some examples of such rhombic geometry, [Ru6(CO)14(m2-SC2H5)2{m6-C(CH3)- CCC(CH3)}] 43 and [Os6(CO)20{C]] C(H)Ph}],44 containing a C4 or C2 alkene chain are known, structural characterization of 3 provides a rare example of hexaruthenium clusters with the two carbido-carbons quadruply bridged in a ladder plane, which may represent a particularly attractive analogue of metal surface carbides.The 13C NMR studies again reveals two singlets at d 324 and 223 which are attributed to the carbido and keto carbons, respectively. Cluster 3 possesses 92 CVE which is two less than the expected 94 CVE unless Ru(3)–Ru(4) is considered to be a double bond.Some other six-atom raft geometries, such as [Os6(CO)20{C]] C(H)Ph}],44 are known to have 92 valence electrons which are composed of four osmium triangles fused together. The structure of 3 is much more symmetrical than that of 2. It contains a noncrystallographic two-fold rotation axis passing through Ru(3) and Ru(4). Spectroscopic and structural characterization of 4 Isolation by preparative TLC aVorded the fourth fraction which gave slightly air-sensitive brown crystals of 4.The 1H NMR spectrum of 4 is very similar to those for 2 and 3 showing the same spectral pattern in each case but with diVerent chemical shifts. This is in accordance with the presence of a metallocycloketonic ring in these three clusters; besides, the quartet resonance at d 4.65 and the doublet resonance at d 1.82 correspond to the allenyl proton and methyl protons, respectively.The positive FAB spectrum exhibits a molecular ion peak at m/z 1402. A perspective drawing of 4 is depicted in Fig. 4 and selected bond parameters are given in Table 4. Complex 4 consists of an edge-bridged butterfly bonded with additional two Ru atoms via the Ru(3)–Ru(4) edge. The Ru–Ru distances fall into the range 2.733(4)–2.963(4) Å with the Ru(3)–Ru(4) separation being the shortest and the Ru(4)–Ru(5) vector the longest. The seven Ru atoms are ligated by two diVerent acetylide derivatives with two diVerent coordination modes.Like 2 and 3, a five-membered metallocycloketonic ring bonded to four Ru [Ru(1), Ru(2), Ru(3) and Ru(4)] atoms adopts an unusual m4- h1:h1:h2:h2 mode. The CH3(H)Cg]] Cb]] Ca moiety can be viewed as a allenylidene ligand 45,46 bonded to five Ru [Ru(3), Ru(4), Ru(5), Ru(6) and Ru(7)] atoms, in which the alkylated carbide atom C(24) has a strong s-type interaction with Ru(3), Ru(4), and Ru(5) [2.09(3)–2.36(3) Å].The Cb atom is s bonded to Ru(6) [Ru(6)–C(25) 2.05(3) Å], which also forms a p interaction to Ru(7) with Ca [C(24)–C(25) 1.32(4) Å]. The bond length3394 J. Chem. Soc., Dalton Trans., 1998, 3391–3396 of Cb-Cg is 1.41(4) Å which reveals the presence of a partial double bond character. Regarding the allenylidene ligand as a six-electron donor and a metallocycloketonic ring also as a sixelectron donor, the valence electron count is 106 for 4, which is in agreement with the PSEP rule.Spectroscopic and structural characterization of 5 The molecular structure of 5 was established by X-ray crystallography and is shown in Fig. 5; Table 5 collects some signifi- cant bond parameters. The structure of 5 consists of two crystallographically independent molecules in the asymmetric unit. The spectroscopic data for 5 are fully consistent with the solid-state structure. An intense molecular ion peak at m/z 922 was observed in the positive FAB mass spectrum.Its 1H NMR spectrum shows two singlets at d 7.49 and 5.27 for the two sets of methyne proton. The two methylene protons on C(29), which are magnetically non-equivalent, give rise to two sets of double quartets centred at d 3.31 and 2.85 with coupling constants J(HH) 7.5 Hz due to geminal coupling. The signal at d 1.07 due to the methyl proton on C(30) appears as a triplet due to overlap of the two doublets with J(HH) 7.5 Hz, which is coupled to the non-equivalent methylene protons. The singlet resonance at d 2.16 is assigned to the other methyl protons on C(34). The metal framework is essentially the same as that in cluster 2 and the structure consists of a metallocycloketonic ring, a bridging carbonyl ligand and a m4-h2-CCCH2CH3 ligand.The C(31) Fig. 4 Molecular structure of [Ru7(CO)19{m4-h1:h1:h2:h2-CC(H)C(O)- CH3}{m5-h1:h1:h1:h1:h2-CCC(H)CH3}] 4 with the atom numbering scheme. Table 4 Some selected bond lengths (Å) and angles (8) for cluster 4 Ru(1)–Ru(2) Ru(2)–Ru(3) Ru(3)–Ru(5) Ru(4)–Ru(5) Ru(5)–Ru(6) Ru(1)–C(20) Ru(2)–C(20) Ru(3)–C(20) Ru(4)–C(20) Ru(5)–C(24) Ru(7)–C(24) C(24)–C(25) Ru(6) ? ? ? C(24) Ru(3)–C(24)–Ru(4) Ru(7)–C(24)–C(25) Ru(6)–C(25)–Ru(7) C(20)–C(21)–C(22) C(21)–C(22)–C(23) 2.799(4) 2.809(4) 2.837(4) 2.963(4) 2.827(4) 2.15(3) 2.25(3) 2.02(3) 2.14(3) 2.36(3) 2.37(3) 1.32(4) 2.44(3) 81(1) 72(2) 123(1) 115(3) 120(3) Ru(1)–Ru(4) Ru(3)–Ru(4) Ru(3)–Ru(6) Ru(4)–Ru(7) Ru(5)–Ru(7) Ru(1)–C(21) Ru(2)–O(20) Ru(3)–C(24) Ru(4)–C(24) Ru(6)–C(25) Ru(7)–C(25) C(25)–C(26) Ru(7) ? ? ? C(26) Ru(4)–C(24)–Ru(7) Ru(7)–C(25)–C(24) Ru(6)–C(25)–C(24) O(20)–C(22)–C(23) C(24)–C(25)–C(26) 2.944(4) 2.733(4) 2.913(4) 2.878(4) 2.867(4) 2.18(3) 2.18(2) 2.09(3) 2.09(3) 2.05(3) 2.33(3) 1.41(4) 2.43(3) 80(1) 75(2) 90(2) 118(3) 120(3) atom is asymmetrically s bonded to the four Ru(1–3) atoms, and also p interacts with the hinge Ru(5) atom together with C(32).The oxygen atom O(27) in the metallocycloketonic ring is s bonded to the wingtip Ru(3) atom [Ru(3)–O(27) 2.15(1) Å].The Ru(4) and Ru(5) atoms are asymmetrically bridged by a CO ligand [Ru(4)–C(1) 1.97(2) Å; Ru(5)–C(1) 2.09(2) Å]. The acetylene ligand is s bonded to the two wingtip Ru(2) and Ru(3) atoms, and p bonded to one hinge Ru(4) atom with one bridging Ru(1) atom. If the metallocycloketonic ring and HCCCH2CH3 are considered to be a six-electron donor and a four-electron donor respectively, cluster 5 has 76 CVE which obeys the EAN rule for the Ru5 wingtip bridged butterfly.To investigate the eVect of reaction time on this reaction, heating of Ru3(CO)12 and but-3-yn-2-ol in cyclohexane for 24 hours has been carried out. We observed no significant increase in the yields for complexes 1, 2 and 5. However, much lower yields were observed for 3 and 4 together with the formation of some insoluble brown powder which cannot be further characterised. The same reaction has also been carried out in THF and CHCl3 at reflux. However, only insoluble dark brown materials were obtained.No clusters 1–5 can be isolated in significant amount. Attempts have been made to examine the relationship between cluster 2, 3 and 4. Reaction of 2 with an excess of Ru3(CO)12 does not lead to 3 nor 4. No reaction is observed between 3 and Ru3(CO)12. Hence, it is believed that the formation of complexes 1–5 follows independent reaction pathways. Conclusion In view of the importance of the acetylenic metal cluster chemistry to surface chemistry, the formation of various novel Fig. 5 Molecular structure of [Ru5(CO)12(m-CO){m4-h1:h1:h2:h2- CC(H)C(O)CH3}(m4-h2-HCCCH2CH3)] 5 with the atom numbering scheme. Table 5 Some selected bond lengths (Å) and angles (8) for cluster 5 Ru(1)–Ru(2) Ru(2)–Ru(4) Ru(3)–Ru(4) Ru(4)–Ru(5) Ru(1)–C(28) Ru(2)–C(27) Ru(3)–O(27) Ru(4)–C(27) Ru(5)–C(31) Ru(2)–Ru(1)–Ru(3) Ru(2)–Ru(5)–Ru(3) C(27)–Ru(4)–C(28) C(11)–Ru(4)–C(28) 2.775(2) 2.949(2) 2.923(2) 2.825(2) 2.36(2) 2.17(2) 2.15(1) 2.29(2) 2.06(2) 83.39(6) 80.57(6) 35.6(5) 140.9(8) Ru(1)–Ru(3) Ru(2)–Ru(5) Ru(3)–Ru(5) Ru(1)–C(27) Ru(1)–C(31) Ru(2)–C(31) Ru(3)–C(28) Ru(4)–C(28) Ru(5)–C(32) Ru(2)–Ru(4)–Ru(3) C(27)–Ru(1)–C(28) C(31)–C(32)–C(33) C(11)–Ru(5)–C(31) 2.807(2) 2.884(2) 2.859(2) 2.26(2) 2.17(2) 2.16(2) 2.09(2) 2.31(2) 2.25(2) 78.45(6) 35.4(5) 116(1) 124.9(8)J. Chem.Soc., Dalton Trans., 1998, 3391–3396 3395 Table 6 Spectroscopic data for clusters 1–5 Cluster 1 IR, n(CO)a/cm21 2094s, 2064s, 2037vs, 2014w 1H NMR,b d(J/Hz) 4.32 (1H, t, J = 3.7, Hb), 4.29 (2H, d, J = 3.7, Ha), 2.71 (1H, q, J = 5.6, Hc), 2.18 (3H, s, Hd), 1.72 (3H, d, J = 5.6, He) MSc (m/z) 862 (863) 2 2099w, 2062vs, 2048s, 2018m [C]] O (KBr) 1558m] 4.52 (1H, s, Ha), 2.02 (3H, s, Hb), 221.31 (1H, s, Hc or Hd), 228.45 (1H, s, Hd or Hc) 967 (968) 3 2070s, 2058s, 2006m [C]] O (KBr) 1541m] 3.84 (2H, s, Hb), 2.10 (6H, s, Ha) 1190 (1190) 4 2070vs, 2056vs, 2032m, 2012m 4.65 (1H, q, J = 5.8, Ha), 4.21 (1H, s, Hb), 2.09 (3H, s, Hc), 1.82 (3H, d, J = 5.8, Hd) 1402 (1403) 5 2095s, 2066s, 2043vs, 2026s, 2014s, 1981 (sh) 7.49 (1H, s, Hc), 5.27 (1H, s, Ha), 3.31 (1H, dq, J = 7.5, 15.7, Hb1 or b2), 2.85 (1H, dq, J = 7.5, 15.7, Hb1 or b2), 2.16 (3H, s, Hd), 1.07 (3H, t, J = 7.5, He) 922 (923) a In CH2Cl2.b In CDCl3. c Simulated values given in parentheses. Table 7 Summary of crystal data and data collection parameters for clusters 1–5 Cluster Empirical formula M Crystal colour, habit Crystal size/mm Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 Z Dc/g cm23 T/K F(000) m(Mo-Ka)/cm21 w-Scan width/8 No.of plates Scan time/min 2q Range collected/8 Scan speed/8 min21 No. reflections collected No. unique reflections No. observed reflections [I > 3s(I)] RR w Goodness of fit, S Maximum D/s No. of parameters Maximum, minimum density in DF map/e Å23 (close to Ru) 1 Ru4C20H10O13 852.57 Purple, block 0.14 × 0.30 × 0.32 Monoclinic P21/c (no. 14) 11.102(5) 9.987(4) 23.312(5) — 100.62(3) — 2540(1) 4 2.255 298 1640 23.97 1.73 1 0.35 tan q —— 2.0–45 16.0 3762 3550 1986 0.066 0.097 3.24 0.02 214 1.83, 21.43 2 Ru5C18H6O15 967.59 Orange, block 0.11 × 0.13 × 0.17 Monoclinic C2/c (no. 15) 25.000(2) 14.835(3) 17.302(3) — 127.44(1) — 5094(1) 8 2.523 298 3632 29.69 1.31 1 0.35 tan q —— 2.0–45 16.0 3576 3483 2540 0.036 0.038 1.64 0.01 178 0.99, 20.89 3 Ru6C24H8O18 1190.74 Red, block 0.22 × 0.22 × 0.29 Triclinic P1� (no. 2) 9.542(1) 12.192(1) 15.699(1) 93.30(1) 99.45(1) 113.25(1) 1640.1(3) 2 2.411 298 1120 27.70 — 65 5 2.0–51.2 — 10887 4253 2251 0.053 0.062 1.49 0.03 223 1.01, 20.93 4 Ru7C27.50H9O20Cl 1402.30 Brown, plate 0.10 × 0.23 × 0.26 Monoclinic C2/c (no. 15) 40.140(10) 8.889(7) 23.179(8) — 113.28(3— 7597(6) 8 2.452 298 5272 28.58 1.47 1 0.35 tan q ——2.0–45 16.0 5451 5359 2348 0.062 0.078 2.43 0.03 263 2.00, 21.80 5 Ru5C21H14O10 991.65 Red, plate 0.07 × 0.26 × 0.29 Triclinic P1� (no. 2) 9.966(1) 17.564(1) 18.051(2) 114.22(1) 100.49(1) 97.22(1) 2761.8(6) 4 2.385 298 1872 27.39 — 65 5 2.0–51.2 — 14987 6552 3425 0.050 0.052 1.77 0.02 371 0.84, 20.44 hydrocarbyl ketonic fragments on a metal cluster surface is intrinsically interesting. The results reported in this paper show that the five clusters are obtained by a combination of dehydration, hydrogen atom transfer and oxidation of a secondary alcohol to give a ketone. Cluster 1 consists of a Ru4 butterfly skeleton with a C8 ketonic chain fragment.Clusters 2– 5 are interesting examples of metal carbonyls with a metallocycloketonic ring via a m4-h1:h1:h2:h2 bonding mode and the OÆRu dative interaction due to the activation of hydroxyl group is observed. Moreover, the cluster 3 has a hexaruthenium skeleton, with seven M–M bonds, which is diVerent from those commonly observed for the hexanuclear raft compounds. Cluster 4 is one of a very rare collection of a distorted Ru4 square plane sharing an edge with an edge-bridged butterfly.Experimental All the reactions were performed under an atmosphere of high purity nitrogen using standard Schlenk techniques. Analytical grade solvents were purified by distillation over the appropriate drying agents and under an inert nitrogen atmosphere prior to use. Infrared spectra were recorded on a Bio-Rad FTS-7 spectrometer using a 0.5 mm solution cell. Positive-ion fast atom bombardment mass spectra were obtained using a Finnigan MAT 95 spectrometer, 1H NMR and 13C NMR spectra were recorded in CDCl3 on a Bruker DPX 300 NMR instrument, referenced to internal SiMe4 (d = 0).The reactions were monitored by analytical thin-layer chromatography (5735 Kieselgel 60 F254, E. Merck) and the products were separated in air on preparative thin-layer chromatographic plates coated with Merck Kieselgel 60 GF254. The compound but-3-yn-2ol obtained from Lancaster was used without further purification. Synthesis The compound [Ru3(CO)12] (0.5 g, 0.78 mmol) was refluxed in cyclohexane (60 ml) with but-3-yn-2-ol (0.5 ml) for 8 h.Infrared spectroscopy and TLC indicated complete consumption of the starting material. The solvent was removed in vacuo and the residue separated by TLC using dichloromethane–hexane (15 : 85 v/v) as eluent to aVord five bands with Rf values of 0.30,3396 J. Chem. Soc., Dalton Trans., 1998, 3391–3396 0.45, 0.55, 0.70 and 0.80 respectively. All the clusters 1–5 were isolated as solids in 10, 15, 15, 8 and 10% yields respectively [based on Ru3(CO)12] (Found for Ru4C20H10O13 1: C, 28.45; H, 1.32.Calc.: C, 28.19; H, 1.19%. Found for Ru5C18H6O15 2: C, 22.51; H, 0.50. Calc.: C, 22.34; H, 0.63%. Found for Ru6C24H8O18 3: C, 24.45; H, 0.98. Calc.: C, 24.21; H, 0.68%. Found for Ru7C27H8O20 4: C, 23.96; H, 0.75. Calc.: C, 23.85; H, 0.59%. Found for Ru5C21H14O10 5: C, 25.59; H, 1.20. Calc.: C, 25.45; H, 1.43%). Table 6 summarises the IR, 1H NMR and FAB mass spectroscopies for all the new compounds. X-Ray data collection and structural determination of complexes 1–5 Crystals of clusters 1–5 suitable for X-ray analysis were obtained by slow evaporation of their respective dichloromethane –n-hexane solution at 210 8C for 2 d. Single crystals of 1, 2, 3 and 5 were mounted on a glass fibre using epoxy resin, however, crystal 4, together with solvate of stoichiometry 0.5CH2Cl2, was sealed in a 0.3 mm Lindermann glass capillary.Data were collected at ambient temperature either on a Rigaku AFC7R diVractometer (for 1, 2 and 4) or a MAR research image plate scanner (for 3 and 5) with graphitemonochromated Mo-Ka radiation (l = 0.71073 Å) using the w–2q and w scan techniques respectively. A summary of the crystallographic data and structure refinement is listed in Table 7. All intensity data were collected for Lorentz and polarization eVects. The Y scan method was employed for semi-empirical absorption corrections for 1, 2 and 4, however, an approximation to absorption correction by inter-image scaling was applied for 3 and 5.Scattering factors were taken from ref. 47(a) and anomalous dispersion eVects 47b were included in Fc. The structures were solved by a combination of direct methods (SHELXS 8648 for 1, 2, 3 and 4; SIR 88 49 for 5) and Fourierdi Verence techniques and refined on F by full-matrix leastsquares analysis. The hydrogen atoms of the organic moieties were generated in their ideal positions (C–H 0.95 Å) while all metal hydrides were estimated by potential-energy calculations. 50 All calculations were performed on a Silicon-Graphics computer, using the program package TEXSAN.51 CCDC reference number 186/1145. References 1 F. Muller, D. I. P. Dijkhuis, G. van Koten, K. Vrieze, D. Heijdenrijk, M. A. Rotteveel, C. H. Stam and M. C. Zoutberg, Organometallics, 1989, 8, 992. 2 C. J. Adams, M. I. Bruce, B. W. Skelton and A. H. White, J.Organomet. Chem., 1991, 420, 87. 3 C. J. Adams, M. I. Bruce, B. W. Skelton and A. H. White, J. Organomet. Chem., 1992, 423, 83. 4 R. D. Adams and J. T. Tanner, Organometallics, 1989, 8, 563. 5 R. D. Adams, G. Chen, L. Chen, M. P. Pompeo and J. Yin, Organometallics, 1991, 10, 2541. 6 S. Jeannin, Y. Jeannin and C. Rosenberger, Inorg. Chim. Acta, 1993, 1, 323. 7 E. Sappa, O. Gambino, L. Milone and G. Cetini, J. Organomet. Chem., 1972, 39, 169. 8 E. Sappa, A. Tiripicchio and P.Braunstein, Chem. Rev., 1983, 83, 203. 9 S. Ermer, R. Karpelus, S. Miura, E. Rosenberg, A. Tiripicchio and A. M. M Lanfredi, J. Organomet. Chem., 1980, 187, 81. 10 S. Aime and A. J. Deeming, J. Chem. Soc., Dalton Trans., 1981, 828. 11 S. Aime and A. J. Deeming, M. B. Hursthouse and J. D. J. Backer- Dirks, J. Chem. Soc., Dalton Trans., 1982, 1625. 12 M. G. Karpov, S. P. Tunik, V. R. Denisov, G. L. Starova, A. B. Nikol’skii, F. M. Dolgushin, A. I. Yanovsky and Y. T. Struchkov, J.Organomet. Chem., 1995, 485, 219. 13 S. Aime and A. J. Deeming, J. Chem. Soc., Dalton Trans., 1983, 1807. 14 S. Aime, A. Tiripicchio, M. T. Camellini and A. J. Deeming, Inorg. Chem., 1981, 20, 2027. 15 S. Aime and L. Milone, J. Chem. Soc., Chem. Commun., 1980, 1168. 16 B. F. G. Johnson, J. Lewis, J. Lunniss, D. Braga and F. Grepioni, J. Chem. Soc., Chem. Commun., 1988, 972. 17 B. F. G. Johnson, J. Lewis, J. Lunniss, D. Braga and F. Grepioni, J. Chem. Soc., Dalton Trans., 1991, 2223. 18 V. Ferrand, C. Gambs, N. Derrien, C. Bolm, H. S. Evans and G. Süss-Fink, J. Organomet. Chem., 1997, 549, 275. 19 C. J. Adams, M. I. Bruce, B. W. Skelton and A. M. White, Chem. Commun., 1996, 2663. 20 W. Y. Wong, S. Chan and W. T. Wong, J. Chem. Soc., Dalton Trans., 1995, 1497. 21 B. F. G. Johnson, D. S. Shephard, D. Braga, F. Grepioni and S. Parsons, J. Chem. Soc., Dalton Trans., 1998, 311. 22 A. A. Koridze, N. M. Astakhova, F. M. Dolgushin, A. I. Yanovsky, Y. T.Struchkov and P. V. Petrovskii, Organometallics, 1995, 14, 2167. 23 F. Muller, G. van Koten, L. H. Polm, K. Vrieze, M. C. Zoutberg, D. Heijdenrijk, E. Kragten and C. H. Stam, Organometallics, 1989, 8, 1340. 24 P. F. Jackson, B. F. G. Johnson, J. Lewis, P. R. Raithby, G. J. Will, M. McPartlin and W. J. H. Nelson, J. Chem. Soc., Chem. Commun., 1980, 1190. 25 P. Mathur, S. Ghosh, M. M. Hossain, C. V. V. Satyanarayana, A. L. Rheingold and G. P. A. Yap, J. Organomet. Chem., 1997, 538, 57. 26 M. Manassero, M. Sansoni and G. Longoni, J. Chem. Soc., Chem. Commun., 1976, 919. 27 T. Dutton, B. F. G. Johnson, J. Lewis, S. M. Owen and P. R. Raithby, J. Chem. Soc., Chem. Commun., 1988, 1423. 28 J. F. Corrigan, N. J. Taylor and A. J. Carty, Organometallics, 1994, 13, 3778. 29 J. F. Corrigan, S. Doherty, N. J. Taylor and A. J. Carty, Organometallics, 1993, 12, 1365. 30 B. F. G. Johnson, J. Lewis, J. N. Nicholls, J. Puga, P. R. Raithby, M. J. Rosales, M. McPartlin and W.Clegg, J. Chem. Soc., Dalton Trans., 1983, 277. 31 C. J. Adams, M. I. Bruce, B. W. Skelton and A. H. White, Chem. Commun., 1996, 969. 32 C. J. Way, Y. Chi, I. J. Mavual, S. L. Wang, F. L. Liao, S. M. Peng and G. H. Lee, J. Cluster Sci., 1997, 8, 87. 33 C. J. Adams, M. I. Bruce, M. J. Liddell and B. K. Nicholson, J. Organomet. Chem., 1991, 420, 105. 34 G. Conole, M. McPartlin, H. R. Powell, T. Dutton, B. F. G. Johnson and J. Lewis, J. Organomet. Chem., 1989, 379, C1. 35 C. J. Adams, M. I. Bruce, B. W. Skelton and A. H. White, J. Organomet. Chem., 1993, 445, 199. 36 C. J. Adams, M. I. Bruce, M. J. Liddell and E. R. T. Tiekink, J. Organomet. Chem., 1993, 445, 187. 37 B. F. G. Johnson, J. Lewis, P. R. Raithby, M. J. Rosales and D. A. Welch, J. Chem. Soc., Dalton Trans., 1986, 453. 38 R. D. Adams and W. Wu, Organometallics, 1993, 12, 1248. 39 L. F. Rhodes, J. C. HuVman and K. G. Caulton, J. Am. Chem. Soc., 1983, 105, 5137. 40 J. Evans, A. C. Street and M. Webster, J. Chem. Soc., Chem. Commun., 1987, 637. 41 M. P. Diebold, S. R. Drake, B. F. G. Johnson, J. Lewis, M. McPartlin and H. Powell, J. Chem. Soc., Chem. Commun., 1988, 1358. 42 R. D. Adams, Z. Li, J. C. Lii and W. Wu, Organometallics, 1992, 11, 4001. 43 S. Jeannin, Y. Jeannin, F. Robert and C. Rosenberger, Inorg. Chem., 1994, 33, 243. 44 J. G. JeVrey, B. F. G. Johnson, J. Lewis, P. R. Raithby and D. A. Welch, J. Chem. Soc., Chem. Commun., 1986, 318. 45 P. Blenkiron, J. F. Lorrigan, N. J. Taylor, A. J. Carty, S. Doherty, M. R. J. Elsegood and W. Clegg, Organometallics, 1997, 16, 297. 46 D. Nucciarone, N. J. Taylor and A. J. Carty, Organometallics, 1986, 5, 1179. 47 D. T. Cromer and J. T. Waber, International Tables for X-Ray Crystallography, Kynoch Press, Birmingham, 1974, vol 4, (a) Table 2.2B, (b) Table 2.3.1. 48 G. M. Sheldrick, SHELXS 86, Program for Crystal Structure Solution, Acta Crystallogr., Sect. A, 1990, 46, 467. 49 SIR 88, M. C. Burla, M. Camalli, G. Cascarano, C. Giacovazzo, G. Polidori, R. Spagna and D. Viterbo, J. Appl. Crystallogr., 1989, 22, 389. 50 A. G. Orpen, J. Chem. Soc., Dalton Trans., 1980, 2509. 51 TEXSAN, Crystal Structure Analysis Package, Molecular Structure Corporation, Houston, TX, 1985 and 1992. Paper 8/05368E

ISSN:1477-9226    

DOI:10.1039/a805368e

出版商:RSC    

年代:1998

数据来源: RSC