摘要:

DALTON PERSPECTIVE J. Chem. Soc., Dalton Trans., 1999, 1027–1037 1027 Insertion chemistry into metal–carbon bonds Akio Yamamoto Department of Applied Chemistry, Graduate School of Science and Engineering, Waseda University, Shinjuku, Tokyo, 169-8555, Japan Received 26th October 1998, Accepted 14th December 1998 Transition metal alkyls (hydrocarbyls, including aryls) occupy a central position in organometallic chemistry. Particularly, their reactions with unsaturated compounds such as olefins and carbon monoxide are relevant to various important catalytic processes.The quest for the synthesis of various alkyltransition metal complexes and examination of their reactions with unsaturated compounds has led, often serendipitously, to findings of novel reactions. The present Perspective is an account of the personal inquiries of the author related to these transition metal alkyls and the results, including the polymerization of vinyl monomers and carbonylation of various substrates.Among various transition metal complexes studied, those of palladium were found most versatile, leading to findings of various novel catalytic processes. Alkyl compounds of transition metals are now known for most of the transition metals in the Periodic Table.1 They were a rarity, however, before the nineteen fifties, when I started my research work as a graduate student, except for platinum alkyls that were regarded as exceptions.2 Only in the nineteen sixties some of the thermally stable transition metal methyl and ethyl complexes were isolated as stable complexes. As a young student I was fascinated at that time by the exciting discovery of the Ziegler catalyst 3,4 and by the subsequent development led by Natta in polymerization of various unsaturated compounds.I recall now the excitement filling the lecture halls in polymer chemistry divisions of the Chemical Society of Japan and of the Polymer Society, Japan, where people were heatedly discussing the mechanism of olefin polymerization by the mixed systems of titanium chlorides and alkylaluminium compounds.Akio Yamamoto was born in Tokyo in 1930. He received his bachelor’s degree from Waseda University and the doctor’s degree from Tokyo Institute of Technology. He was promoted to Full professor of the Research Laboratory of Resources Utilization at Tokyo Institute of Technology in 1971 and has served as director since 1988. In 1990 he was invited by Waseda University and has worked as Research Professor since then.Recipient of the Chemical Society of Japan Award, Society of Polymer Science Award, and Violet Ribbon. Akio Yamamoto For me, however, it seemed to be extremely diYcult to establish the nature of the active species in the Ziegler catalyst systems by examining the mixed catalyst systems themselves. It was Cossee 5 who first proposed a simple concept to elucidate the olefin polymerization mechanism by assuming the formation of titanium alkyls as the catalyst center generated by interaction of titanium chlorides with aluminium alkyls (Scheme 1).The initiation of polymerization was explained by subsequent p co-ordination of an olefin monomer with the titanium alkyl leading to insertion of the olefin into the metal alkyl bond followed by consecutive insertion of monomers, constituting the propagation steps in the olefin polymerization. However, verification of the mechanism was much delayed because of the paucity of transition metal alkyls suitable as models of the active site.It is now hard to imagine the diYculties involved in establishing the active site of the catalyst systems at that time if we view the situation from the standpoint where the synthesis and polymerization abilities of various single site polymerization catalysts having transition metal alkyls are established.6 Some mystery was still attached then to the catalysis by transition metals and their compounds.Many people did not believe that pure compounds could serve as the catalysts and some specific functions associated with the surface of transition metals or their compounds were believed to promote the catalysis. It was Wilke 7 who removed the myth by isolating pure low valent transition metal complexes that showed excellent catalytic activities for butadiene oligomerization and polymerization. During my stay in his group at the Max Planck Institute for Coal Research in Mülheim, Germany in 1962 to 1963, I was very much influenced by the elegance of his work and by the power of the methodology of isolating low-valent transition metal complexes and examine the behavior of the complexes toward various unsaturated complexes.Some time after my return from Germany we initiated a joint work with Professor Uchida’s group at Tokyo University attempting to isolate low valent, catalytically active transition metal compounds from mixed systems of nickel, cobalt and iron acetylacetonates and aluminium alkyls containing various ligands such as 2,29- bipyridine (bpy) and tertiary phosphines (PR3).The joint work Scheme 1 Cossee’s mechanism for polymerization of ethylene.1028 J. Chem. Soc., Dalton Trans., 1999, 1027–1037 quite unexpectedly led us to isolation of transition metal alkyls having the supporting bipyridine ligand(s) [eqns. (1)]. These alkyl complexes of late transition metals 8–10 proved to be the earliest examples of thermally stable, isolated transition metal alkyls except for platinum alkyls.At that time NMR apparatus was not readily available in our group and characterization of the metal alkyls was made mostly on the basis of chemical reactions and elemental analyses. When we could convince ourselves that the nickel and iron complexes contain alkyl groups attached to the metals on the basis of characterization by their thermolysis and treatment with protic reagents releasing the alkanes, the first reactions we attempted were with butadiene followed by reactions with various vinyl compounds.The reactions of these complexes with butadiene proved that these complexes actually serve as the catalysts for oligomerization of butadiene in the same way as with the mixed catalyst systems. On the other hand, when we subjected these transition metal alkyls to vinyl compounds we observed that the ensuing processes varied depending on the transition metals employed.Coupling of the alkyl groups from nickel dialkyls was observed on treatment of the nickel alkyls with electron-deficient olefins, such as acrylonitrile and acrolein, whereas polymerization of vinyl monomers occurred with the iron alkyls. The reactions of NiEt2(bpy) 1 with electron deficient olefins such as acrylonitrile at low temperatures were revealed to involve the co-ordination of the olefin to the nickel alkyls.11 Raising the temperature led to activation of the nickel–alkyl bonds in 1 and to coupling of the two alkyl groups by a process now termed reductive elimination 12,13 [eqn.(2)]. Ni(acac)2 + AlEt2(OEt) + bpy NiEt2(bpy) 1 Fe(acac)3 + AlEt2(OEt) + bpy FeEt2(bpy)2 2 CoEt(bpy)2 Co(acac)3 + AlEt2(OEt) + bpy (1) 3 4 NiEt2(bpy) + olefin (bpy)Ni Et Et – C4H10 (bpy)Ni(olefin) n (2) n = 1, 2 R The bipyridine-co-ordinated nickel alkyls provided us with a good model to demonstrate how the transition metal–alkyl bonds are activated on interaction with olefins as was proposed in Cossee’s original concept.It was revealed that the more back bonding is provided from the transition metal to the coordinated olefin, the stronger becomes the p bond between the metal and the co-ordinated olefin, and that the stronger the back bonding from the metal to the olefin, the stronger becomes the activation of the nickel–alkyl bond to enhance the reductive elimination. On the other hand, the bipyridine-co-ordinated iron alkyl 2 was found to initiate the polymerization of electron-deficient vinyl monomers such as acrylonitrile and methyl methacrylate.This work was the first example of polymerization of polar monomers initiated by isolated alkyl complexes of late transition metals.14 The results we obtained contained important conclusions in the light of our present knowledge regarding the mechanism of vinyl polymerization by transition metal alkyls as reiterated below.15–20 (1) The polymerization is initiated by co-ordination of an olefin to a vacant site created by partial dissociation of the bipyridine ligands.(2) Competition reactions of various monomers indicate that olefin binding more strongly with the iron alkyl center by back donation is preferentially introduced into the copolymer and the feature of the copolymerization is similar to that of anionic polymerization. (3) The initiation by insertion of the co-ordinated olefin into the iron–alkyl bond is followed by rapid successive insertions of the monomers constituting the propagation steps.(4) Termination involves b-hydrogen elimination from the polymer attached to the iron propagation site or proceeds by reductive elimination of the iron-bound growing alkyl chain with the other remaining alkyl group (Scheme 2).21 However, the iron alkyls we prepared showed a diVerent behavior from the Ziegler Natta type catalysts composed of early transition metal compounds that showed high polymerization activity for non-polar a-olefins. Thus a question has remained if there is an intrinsic diVerence between the reactivities of early and late transition metal alkyls in their ability to polymerize olefins.Recent development of the chemistry of palladium and nickel alkyls shows that a-olefins such as ethylene and propylene also can be polymerized when a suitable ligand is employed.22,23 Very recently, attainment of quite high activity for ethylene polymerization was reported by two groups using iron- and cobalt-based initiators.24,25 The results indicate that high molecular weight polymers of a-olefins are available even with late transition metal complexes such as iron and cobalt by controlling chain transfer processes. In the mean time, a study related to examination of the solvent eVect in propylene dimerization with nickel complexes in Scheme 2 Mechanism of olefin polymerization initiated by FeEt2(bpy)2.J.Chem.Soc., Dalton Trans., 1999, 1027–1037 1029 chlorobenzene led us to investigate the reaction of NiEt2(bpy) with chlorobenzene.26 The nickel complex 1 exhibited a high activity for dimerization of propylene in chlorobenzene and the mechanism consisting of propylene insertions followed by b-hydrogen elimination was proposed to account for the formation of the linear and branched dimers. The investigation of the reaction of halogenoarenes with NiEt2(bpy) 1 to get information on the marked influence of chlorobenzene on the dimerization activity brought us an unexpected finding that the reaction caused the coupling of the two ethyl groups forming butane and gave arylnickel halide complexes 5 [eqn.(3)]. Without realizing the importance of this finding we reported the result simply as the preparation of a new arylnickel complex. The importance was not overlooked by those who were prepared to apply the new results to organic synthesis. Tamao and Kumada27,28 at Kyoto University and Corriu and Masse 29 in Montpelier developed a new synthetic method to realize the coupling of aryl and alkyl groups by using aryl halides and alkyl Grignard reagents [eqn.(4)]. The process was later developed to more convenient synthetic methods utilizing palladium complexes. Further modification of the process utilizing other alkylation agents such as alkylzinc, aluminium, boron, tin, and silicon compounds in place of the Grignard reagent led to development of extremely useful processes in organic syntheses for achieving various coupling processes.30 I failed to see the applicability of the process represented in eqn.(4). At that time I did not feel much chagrin but later realized that I missed a pretty big catch because of my ignorance of the significance of the synthetic methodology. The essence of the cross coupling process can be represented by Scheme 3. The process is composed of (1) oxidative addition of aryl halides (Ar–X) to zerovalent transition metal complexes to give arylmetal halides, (2) alkyl transfer (transmetallation) to give an aryl(alkyl)transition metal complex, (3) reductive 5 (bpy)Ni + PhCl + Et–Et (3) (bpy)Ni Et Et Ph Cl Scheme 3 Mechanism of transition metal-catalysed cross coupling of aryl and alkyl groups.elimination involving C–C coupling to liberate an alkyl–aryl coupling product.†,31,32 Of the above three elementary processes in Scheme 3 the oxidative addition of aryl halides to complexes of Ni0 or Pd0 to give arylnickel and arylpalladium halide complexes had been well established. The other processes, namely alkyl transfer from a metal alkyl to arylnickel or arylpalladium halides to give alkyl(aryl)-nickel and -palladium complexes, had not been well examined.Our interest to find the reasons why some transition metal alkyls exist as thermally stable compounds, while the others do not, led us to study the thermolysis mechanisms of various diorgano-nickel and -palladium complexes.33–36 Particularly interesting results were obtained by examining the thermolysis behavior of cis- and trans-dialkylpalladium complexes having monodentate tertiary phosphine ligands.It was revealed that the trans and cis isomers of the dialkylpalladium complexes having two monodentate tertiary phosphine ligands behave diVerently. Thermolysis of cis-dialkylpalladium complexes in the absence of added ligand aVorded coupling products, ethane from cis-PdMe2(PR3)2 and butane from cis-PdEt2(PR3)2 and the thermolysis was hindered by addition of PR3 to the system, whereas the thermolysis of trans-PdEt2(PR3)2 liberated ethylene and ethane in a 1 : 1 ratio.On the other hand, thermolysis of trans-PdMe2(PR3)2 proceeded through trans to cis isomerization followed by reductive elimination. These results suggested the involvement of a three-co-ordinated, T-shaped intermediate having “cis” and “trans” configurations and existence of an energy barrier between them.Theoretical studies provided the support for the intermediacy of the T-shaped species.34 Further studies on reactions of square planar cis- and transdialkyl- nickel 37 and -palladium 38 complexes with carbon monoxide revealed intriguing behavior. In our studies on reactions of CO with dimethyl- and diethyl-palladium complexes toward carbon monoxide we found liberation of ketones, diketone, ethylene and propionaldehyde depending on the methyl and ethyl complexes, and also on the cis and trans configurations of the dialkylpalladium complexes.Generation of these diVerent products in reactions of dialkylpalladium complexes with carbon monoxide was accounted for in a consistent manner by assuming the constraint of square planar geometry in the behavior of palladium complexes and by assuming alkyl migration in the CO insertion into palladium–alkyl bond in the square plane. Particularly, examination of the reason for production of only acetone from trans-PdMe2(PR3)2 on the one hand, and generation of acetone and butane-2,3-dione on the other, in the reaction of cis-PdMe2(PR3)2 with CO suggested operation of two routes (a) and (b) in Scheme 4.The assumption prompted us to investigate the reaction of the dimethylpalladium complex with CO in the presence of diethylamine with the aim of trapping the acetyl group. The reaction aVorded the a-keto amide MeCOCONEt2 where two CO molecules were introduced.Further studies on the reactions using monomethylpalladium iodide and monophenylpalladium iodide with CO and diethylamine also showed liberation of a-keto amides MeCOCONEt2 and PhCOCONEt2 respectively together with monoamides, the single carbonylation products.39 The latter process producing the a-keto amide on reaction of trans-PdPh(I)(PR3)2 6 with CO and diethylamine [eqn. (5)] suggested the possibility of application of the finding to a catalytic process to convert aryl halides into a-keto amides, since the † In causing the reductive elimination of the alkyl–aryl coupling product, interaction of the aryl halide through the phenyl ring with the alkyl(aryl)metal complex may be involved, but this process is omitted from Scheme 3 to stress the most important concept.The mechanism is straightforward when a bidentate ligand is used but consideration of trans and cis isomers and of the isomerization processes between them is required when a monodentate tertiary phosphine is employed as the ligand.1030 J.Chem. Soc., Dalton Trans., 1999, 1027–1037 Scheme 4 Formation routes of ketone and diketone form cis- and trans-PdMe2L2. reaction (5) is expected to produce a palladium(0) complex together with NEt2H2X and such complexes are well known oxidatively to add aryl halide to yield an arylpalladium halide. In fact, the catalytic conversion of aryl halide into a-keto amide on treatment with CO and diethylamine in the presence of Pd(PPh3)4 was confirmed in the first attempt of the catalytic process [eqn.(6)].40 The lesson I learned in having missed the opportunity of devizing the cross-coupling process catalysed by a transition metal complex had some eVect. It is interesting that Tanaka independently achieved the same catalytic process starting from a diVerent question regarding the formation of ketone at about the same time.41 It often happens that when one finds something new, there is always someone else quite close to the finding.We have studied the scope of application of the double carbonylation to organic synthesis as well as the mechanism of the new process. It took us about 10 years to establish the reaction mechanism to a satisfying degree. The mechanism of a-keto amide formation from aryl halide and CO in the presence of secondary amine catalysed by a palladium complex is summarized in Scheme 5.42 The catalytic cycle is composed of the fol- Scheme 5 Mechanism of palladium-catalysed a-keto amide formation from aryl halide and CO in the presence of a secondary amine.lowing elementary processes: (1) generation of a palladium(0) species from a catalyst precursor; (2) oxidative addition of aryl halide to this species to yield arylpalladium halide; (3) insertion of CO into the palladium–aryl bond to aVord an aroylpalladium complex; (4) CO co-ordination to the palladium center; (5) nucleophilic attack of the amine on the co-ordinated CO bound with a cationic center; (6) deprotonation by a base (secondary amine itself is included) to produce an aroyl- (carbamoyl)palladium; (7) reductive elimination of the aroyl and the carbamoyl ligands to produce a-keto amide with regeneration of a palladium(0) species that drives the catalytic cycle further.The key steps in the double carbonylation lie in CO insertion into the aryl–palladium bond to give an acylpalladium species and its further reactions to give the bis-acyl type palladium(II) intermediate that reductively eliminates the a-keto amide.The possibility of the consecutive CO insertion into the aryl– palladium bond to give an a-ketoacyl intermediate was excluded by studies involving the separately prepared a-ketoacyl- palladium and -platinum complexes.43,44 Further studies to increase the selectivity for the double carbonylation revealed a somewhat surprising mechanism for formation of amide as the single carbonylation product. It was established that arylpalladium halide dissociates the halide ligand in solution under CO pressure to have the CO ligand bound with a cationic arylpalladium center.In fact trans- Pd(Ar)I(PMe3)2 provides trans-Pd(Ar)(CONEt2)(PMe3)2 on interaction with diethylamine under CO. Under certain conditions, amide ArCONEt2 can be reductively eliminated from the aryl(carbamoyl)palladium complex [eqn. (7)]. Our later study on the reactivities of benzylpalladium and phenylacetylpalladium complexes revealed involvement of another course to give the amide, the single carbonylation product, by the reaction of a phenylacetylpalladium complex with the secondary amine (route a in Scheme 6) as well as the formation of amide from the benzylpalladium species on reaction with CO and the amine 45 (route b in Scheme 6).The form-J. Chem. Soc., Dalton Trans., 1999, 1027–1037 1031 ation of the amide from the CO-co-ordinated benzylpalladium species via route b was concluded to take place in the presence of an excess of the amine to deprotonate the CO bound diethylamine to give the carbamoyl ligand prior to the CO insertion (Scheme 6).Since a-keto acids can readily be converted into biologically active compounds such as a-amino acids and a-hydroxy acids, the finding of the catalytic double carbonylation process was thought to provide a new route for synthesis of these useful derivatives. However, a-keto amides are usually resistant to hydrolysis and development of a synthetic route to more readily hydrolysable a-keto ester was desirable.Studies of our group and of Tanaka’s group realized the double carbonylation of aryl halides to a-keto esters 46 [eqn. (8)]. Detailed studies on the mechanism of double carbonylation of aryl halides to a-keto esters revealed that the catalytic cycle comprises: (1) oxidative addition of aryl halide to a palladium(0) species to give an arylpalladium halide, (2) CO insertion into the Ar–Pd bond to aVord an aroylpalladium complex, (3) attack of CO co-ordinated to a cationic palladium center by the alcohol and a base to yield aroyl(alkoxycarbonyl) palladium, (4) reductive elimination of the aroyl and the alkoxycarbonyl ligands to liberate the a-keto ester with regeneration of a palladium(0) species that carries the catalytic cycle.Formation of the ester, the single carbonylation product, was found to be derived from an aroylpalladium complex not from the arylpalladium complex.Ester was found to be formed via a route involving the intermediacy of aroyl(alkoxy)palladium complex 8 by reductive elimination [eqn. (9)]. Later we have Scheme 6 Two routes to give the amide from phenylacetylpalladium and benzylpalladium complexes. confirmed that a similar mechanism also operates in the ester formation from PhCH2COPdCl(PPh3)2 and alcohols in the presence of NEt3.45 These studies indicated that the palladium-catalysed double carbonylation of aryl halides proceeds through a common mechanism involving CO insertion into an arylpalladium center and by reductive elimination of the acyl ligand with a carbamoyl or alkoxycarbonyl ligand, although the mechanisms to give the ester or amide diVer from each other.When we reached these conclusions the time for the compulsory retirement at Tokyo Institute of Technology for me approached and I had to terminate the research there.Fortunately, I was invited by Waseda University, my alma mater, to continue my work there under somewhat diVerent research conditions in 1990 to build a new research group from scratch with one student. The first project I started was to examine the properties of monoorganopalladium complexes having trimethylphosphine ligands. A kinetic study on thermolysis of trans-PdEt(X)- (PMe3)2, where X is halide, phenoxide, and various carboxylato ligands, revealed that the anionic ligand dissociates in the ratedetermining step to generate a cationic ethylpalladium species which is unstable and readily undergoes b-hydrogen elimination with liberation of ethylene (Scheme 7).47 The result suggested involvement of a cationic monoorganopalladium species as an important reactive species in the reactions of neutral monoorganopalladium complexes with various substrates as well as in b-hydrogen elimination of monoalkylpalladium complexes.We have prepared a variety of monoorganopalladium complexes with mono- and di-tertiary phosphine ligands and examined their reactions with olefins, carbon monoxide, and alkyl isocyanides. During the eVort of building a new research group at Waseda, the recent upsurge of interest regarding the alternating copolymerization of CO and a-olefins and homopolymerization of a-olefins catalysed by cationic palladium complexes spurred intensive research eVort by a number of groups in the U.S. and Europe.48 Fundamental studies on the properties of cationic organopalladium complexes, notably those led by Brookhart 49 and the Netherlands groups,50 clarified various factors regarding the insertions of CO and olefins into the Pd–C bonds of cationic organo-palladium and -nickel complexes.Theoretical studies also contributed to support many of the observations found in the experimental studies.51 Since I have already reviewed our previous results dealing with comparison of the properties of neutral and cationic organopalladium and ruthenium complexes 52 and we are contributing another account including the later development concerning the organopalladium complexes,53 I would like to restrict the discussion here on the properties of monoorganopalladium complexes to a minimum.The studies comparing the reactivities of the neutral and cationic monoorganopalladium complexes having various supporting ligands revealed that the marked enhancement in the reactivities of the neutral monoorganopalladium complexes toward unsaturated compounds such as CO and olefins by their conversion into cationic monoorganopalladium complexes arises mainly by creation of an available co-ordination site cis Scheme 7 b-Hydrogen elimination routes from trans-PdEt(X)(PMe3)2.1032 J.Chem. Soc., Dalton Trans., 1999, 1027–1037 to the alkyl or aryl ligand to allow the subsequent insertion reactions and not from the eVect of the charge of the cationic palladium complex.54 In fact, blocking the co-ordination site of a cationic monoorganopalladium complex with a strongly coordinating ligand caused a more pronounced inhibition of the insertion reaction than with the neutral complex having ligands of less co-ordinating abilities.55–59 For example, examination of the CO insertion rates into trans-[PdMe(L)(PMe3)2]1BF4 2 revealed that the rate decreased in the order of L = acetone > NO3 > CNBut.The rate of CO insertion into the Pd–Me bond was smaller with the cationic isocyanide-co-ordinated complex than into the Pd–Me bond in the neutral methylpalladium chloride complex.The reason for the remarkable rate enhancement by addition of a silver salt to the catalyst system in the olefin arylation reactions catalysed by palladium complexes 60 may be mainly ascribed to the generation of a cationic species having a readily available vacant site by removal of the halide ligand from a neutral arylpalladium halide intermediate in the catalytic cycle.It was also revealed that addition of an excess amount of a silver salt removes one of the co-ordinated tertiary phosphine ligands from the catalyst species leading to further enhancement of the reactivity.58 Another interesting observation in these studies of the insertion reactions into monoorganopalladium complexes is the finding of consecutive alternating insertion of CO and isocyanide into the Pd–C bond. It is well known that the consecutive CO insertion into the Pd–C bond is unfavorable, whereas isocyanides readily undergo consecutive insertion into the Pd– C bond.On the other hand, alternating insertion of the CO after isocyanide or vice versa has not been reported to our knowledge (Scheme 8). Among the supporting ligands used to stabilize organometallic compounds of late transition elements tertiary phosphines have been used most extensively. Recent reports indicate that employment of phosphite ligands or combination of the phosphine and phosphite donors as a specific ligand shows remarkable influence on the insertion rate and stereoselectivities of the insertion products, notably in the CO and a-olefin copolymerization. 61 Our own examination of the eVect of trialkyl phosphite 59 showed that employment of the phosphite ligands favors the cis configuration whereas monoorganopalladium complexes having the tertiary phosphines often give trans complexes. Restriction of the research resources did not permit us further to examine the eVect of the ligands more widely including the eVect of steric bulkiness on the insertion processes.The recent reports using diimine-type ligands with large steric bulkiness seem to underline the importance of steric factors in controlling the olefin insertion and b-hydrogen elimination steps.6,48 1 Catalytic double carbonylation of aliphatic substrates Regarding the catalytic double carbonylation processes there Scheme 8 Alternating insertion of CO and isocyanide into a Pd–C bond.remained one project unachieved before my retirement from Tokyo Institute of Technology. We could carry out the double carbonylation process to introduce two consecutive CO units into aromatic compounds but the process could not be applied to aliphatic systems as long as we use the principle of oxidative addition of alkyl halides to a palladium(0) species; alkyl halide reacts with a nucleophile quite readily and no catalytic cycle could be constructed by a similar mechanism as we discussed regarding Scheme 5.We could convert, however, alkenes into b,g-unsaturated a-keto amides by stoichiometric double carbonylation, i.e. by aminopalladation of olefin followed by carbonylations (Scheme 9).62 In our quest for realization of double carbonylation of aliphatic compounds allylic compounds appeared to be a promising candidate as substrates. The key elementary process was the insertion of CO into an allylpalladium bond.The oxidative addition of allylic compounds with a palladium(0) complex to give an h3-allylpalladium complex is well known. If the CO insertion into the allyl–palladium bond takes place to give an acylpalladium species before attack of the nucleophile at the h3- allyl ligand there seems to be a chance of accomplishing the catalytic double carbonylation of allylic compounds. Precedents of the CO insertion into the h3-allylpalladium bond had been quite limited.63 Before my move to Waseda, we observed that employment of the PMe3 helps to stabilize the CO-inserted acylpalladium complexes and CO insertion into the allyl– palladium bond can be accomplished under appropriate conditions [eqn.(10)].64 Our later studies at Waseda revealed that the CO insertion process can take place under various conditions and that the CO insertion reaction into 11 proceeded even in the presence of a small amount of a secondary amine to give the acylpalladium complex 12 [eqn.(11)]. Employment of strongly co-ordinating ligands such as PMe3 and dppe as well as the usage of a halide ligand of higher co-ordinating ability, such as chloride and bromide rather than iodide or non-co-ordinating anions, seems to favor the CO insertion reaction by preventing the reverse decarbonylation Scheme 9 Stoichiometric conversion of but-1-ene to a-keto amides.J. Chem. Soc., Dalton Trans., 1999, 1027–1037 1033 process. In fact removal of the halide from the acylpalladium complex 13 with AgBF4 at 230 8C immediately aVorded the decarbonylated h3-allylpalladium complex 14 [eqn.(12)] and the reaction of a cationic h3-allylpalladium complex having BF4 anion with CO did not give the acylpalladium complex. Examination of the reaction of the acylpalladium complex 12 with secondary amines under CO indicated that a-keto amides could be produced together with amide accompanied by formation of allylamines and oxamides.The results indicate that further CO co-ordination can take place to the acylpalladium complex 12 and the co-ordinated CO ligand can be attacked by a secondary amine to give a carbamoyl ligand. The subsequent reductive elimination of the acyl and the carbamoyl ligands gives an a-keto amide, as we proposed in the double carbonylation of aroylpalladium complexes. Application of the information obtained with these allyl- and acyl-palladium complexes to a catalytic process led us to realization of the double carbonylation of allylic chlorides catalysed by simple catalyst precursors such as PdCl2(PPh3)2.Employment of polar solvents was not suitable for production of the double carbonylation products. The catalytic double carbonylation did not proceed under 1 atm of CO but application of CO pressure gave quite high yields of a-keto amides with high selectivities [eqn. (13)]. The catalytic double carbonylation process is applicable to allyl chloride, methallyl chloride 15, and 2-chlorobut-1-ene but double carbonylation of cinnamyl chloride 16 gave only amide, the single carbonylation product, whereas 3-chloro-2- phenylpropene 17 aVorded the double carbonylation product [eqns.(14) and (15)]. The mechanism presented in Scheme 10, composed of oxidative addition of methallyl chloride to Pd0, CO insertion, CO co-ordination, nucleophilic attack on the co-ordinated CO, and reductive elimination of the acyl and carbamoyl ligands, seems reasonable on the basis of the behavior of the allylpalladium complexes described above. 2 Other catalytic carbonylation processes Another line of project I was interested in was the C–O bond cleavage promoted by transition metal complexes with an objective of realizing catalytic processes without using organic halides.65,66 The presently known transition metal-catalysed processes use organic halides as substrates to prepare organic products but a base has to be used to remove the hydrogen halide to prepare products containing no halogen.Thus the total eYciency is not high and these processes are not environmentally benign. In the course of examination of the C–O bond cleavage of allylic formats 67 we discovered that the C–O bond in allylic formate 18 can be cleaved in the presence of a palladium catalyst and it can be converted into b,g-unsaturated carboxylic acid 19 when the reaction is carried out under CO pressure [eqn. (16)].68,69 Since the starting allylic formates 18 can readily be prepared by treating the mixture of allylic alcohols and formic acid with Scheme 10 Mechanism of the palladium-catalysed double carbonylation of methallyl chloride.1034 J.Chem. Soc., Dalton Trans., 1999, 1027–1037 diphosphorus pentaoxide, the palladium-catalysed carbonylation of allylic formates provides a convenient means of synthesizing unsaturated carboxylic acids from allylic alcohols without starting from organic halides.For example, treatment of cinnamyl formate 20 with 0.01 mol equivalent of palladium( 0) complex in the presence of triphenylphosphine under 1 atm of CO at room temperature yields terminal and internal olefins, allylbenzene and propenylbenzenes. Increasing the CO pressure above 20 atm completely changes the reaction course and gives 4-phenylbut-3-enoic acid 21 in excellent yields [eqn. (17)]. The process is in essence a molecular rearrangement of allylic formate to b,g-unsaturated acid under the influence of CO pressure and provides a quite clean process to give various unsaturated acids.It can be performed in excellent yields and selectivity when carried out in non-polar solvents such as benzene and toluene, whereas use of polar solvents is unfavorable for getting the acids. The mechanism of the catalytic process can be accounted for by a cycle as shown in Scheme 11 composed of (a) the C–O bond cleavage of allylic formate to give h3-allylpalladium formate A, (b) CO insertion into the allyl–palladium bond to give the acylpalladium–formate intermediate B, (c) reductive elimination of the mixed anhydride C from B to regenerate the palladium(0) species to carry the catalytic cycle.The decarbonylation of the mixed anhydride C to give carboxylic acids is a known process. The application of CO pressure is critical in driving the catalytic cycle to yield the unsaturated carboxylic acids. Under 1 atm of CO, the insertion process to convert the allylic intermediate A into the acylpalladium B does not proceed well and the allylformatopalladium species A undergoes the decarboxylation to give allylhydridopalladium species that liberates olefins on reductive elimination.In an independent study we have established that the h3-allylpalladium formate undergoes the decarboxylation and the subsequent reductive Scheme 11 elimination to aVord olefins.67 The fact that the course of the reaction can be controlled from the one giving the reduction products of allylic formates to the carbonylation process giving the unsaturated carboxylic acids indicates the importance of the CO insertion process as a critical elementary process to determine the product selectivity.In our attempt to find application of the above CO rearrangement we have examined the palladium-catalysed reaction of the diformate of but-2-ene-1,4-diol 22 to get an adipic acid precursor. However we found that the CO insertion took place at only one of the two allylic C–O bonds and the product was penta- 2,4-dienoic acid 23 [eqn.(18)] The reaction course can be accounted for by Scheme 12 which involves the first carbonylation at one of the two allylic formate functionalities in 22 followed by the second allylic C–O bond cleavage to give an h3-allylpalladium complex. Occurrence of the b-hydrogen elimination from the methylene group adjacent to the allylic entity would produce the a,g-pentanedienoic acid 23 from the diformate 22 without giving the expected dibasic acid.Although the original intention of producing an adipic acid precursor from a butenediol derivative failed, the study revealed a new route to produce a dienoic acid from an unsaturated diol.69 Control of the cleavage of the allylic C–O bonds promoted by a palladium complex provides further opportunities of finding a new catalytic process. Although direct cleavage of the C–O bond in allylic alcohols is diYcult to accomplish with a palladium catalyst, we have found a new allylation process of nucleophiles such as secondary amines catalysed by palladium complexes under the influence of CO2 [eqn.(19)].70 The catalytic process is also applicable to allylation of various carbon nucleophiles such as acetylacetone, alkyl malonates, and b-ketoesters. Scheme 12J. Chem. Soc., Dalton Trans., 1999, 1027–1037 1035 The eVect of promotion of the allylic C–O bond cleavage by CO2 can be accounted for by assuming the equilibrium (20) to generate allyl hydrogencarbonate 25, albeit in minor quantities, from allyl alcohol 24 and CO2.The allyl hydrogencarbonate 25 is more susceptible than the parent allyl alcohol to the C–O bond cleavage on interaction with a palladium(0) species. We have further found that the allyl alcohol 24 undergoes the CO insertion to give but-3-enoic acid 26 and its isomer 27 in the presence of palladium complexes and that the reaction is accelerated by CO2 [eqn.(21)]. The palladium-catalysed carbonylation of allyl alcohol and the promotion eVect of carbon dioxide can be accounted for by Scheme 13. A zerovalent palladium species formed from a catalyst precursor is considered to react with allyl hydrogencarbonate 25 formed from 24 to give a cationic h3-allylpalladium complex having the hydrogencarbonate anion. In our previous study we have established the formation of a similar cationic h3-allylpalladium complex by the allyl–oxygen bond cleavage in allyl alkyl carbonate complex.71 Decarboxylation of Scheme 13 the hydrogencarbonate anion gives the h3-allylpalladium containing OH anion.Insertion of CO giving the acylpalladium intermediate followed by reductive elimination yields the butenoic acid 26. An isomerization process, either from the but-3-enoic acid 26 or in the intermediate acyl complex, gives the but-2-enoic acid 27 as shown in Scheme 13. Another type of substrate amenable to C–O cleavage promoted by a transition metal complex is a carboxylic anhydride.As an extension of the previous studies on the C–O bond cleavage of carboxylic anhydrides by nickel complexes,72 we examined the C–O bond cleavage of carboxylic anhydrides by a palladium complex and found their ready oxidative addition to give acyl(carboxylato)palladium complexes 28 [eqn. (22)].73 Treatment of 28 with dihydrogen liberated aldehydes and carboxylic acids. Based on the finding a new catalytic hydrogenation process of converting carboxylic anhydrides into aldehydes and carboxylic acids has been developed [eqn.(23)]. Combination of the C–O bond cleavage of acid anhydrides with olefin insertion into an aryl–palladium bond produced by decarbonylation of the acylpalladium complex led to development of a new type of olefin arylation process by DSM Research [eqn. (24)].74 Further studies in our group to utilize the concept of the C–O bond cleavage of carboxylic anhydrides promoted by palladium complexes led to discovery of a novel direct catalytic hydrogenation of carboxylic acids into aldehydes in the presence of a dehydrating agent such as pivalic anhydride [eqn.(25)].75 The process provides a quite convenient means of producing various aldehydes cleanly from a variety of carboxylic acids, including mono-, di-, and tri-basic acids. Aliphatic as well as aromatic and heteroaromatic carboxylic acids are converted into respective aldehydes in high yields and selectivities.However, since the process is not directly related with the central theme of the present Perspective, i.e. insertion processes, the account of the catalytic mechanism will not be dealt with here.76 3 Conclusion As can be seen from the present Perspective our attempts to prepare new organometallic complexes and to understand the chemistry of these new complexes often led to findings of unexpected reactions, some of which could be applied to useful catalytic processes.It is my great pleasure that I have had the opportunity of working in the emerging and fertile organo- (ArCO)2O + R1HC CR2R3 NMP, 160 °C [Pd], NaBr ArR1C CR2R3 + ArCOOH + CO (24) NMP = N –methylpyrridinone1036 J. Chem. Soc., Dalton Trans., 1999, 1027–1037 metallic field between inorganic and organic chemistry. I would like to invite younger people to join the field where they can satisfy both academic and practical interests. 4 Acknowledgements I would like to thank my co-workers at Tokyo Institute of Technology and Waseda University for their important contributions. I am particularly grateful to Professor Isao Shimizu, my colleague at Waseda University, for his help in my carrying out the research at Waseda University. The work described here was made possible by the generous donation of a chair created for me by Nippon Zeon Company. The grant support by the Ministry of Education, Science, Sports and Culture is gratefully acknowledged. 5 References 1 E. W. Abel, F. G. A. Stone and G. Wilkinson (Editors), in Comprehensive Organometallic Chemistry II, Pergamon, Oxford, 1995. 2 F. A. Cotton, Chem. Rev., 1955, 55, 551. 3 K. Ziegler, E. Holtzkamp, H. Breil and H. Martin, Angew. Chem., 1955, 67, 541. 4 G. Fink, R. Mulhaupt and H. H. Brintzinger (Editors), Ziegler Catalysts, Springer, Berlin, Heidelberg, 1995. 5 P. Cossee, J. Catal., 1964, 3, 80. 6 M. Bochmann, J. Chem. Soc., Dalton Trans., 1996, 255. 7 G. Wilke, J. Organomet. Chem., 1980, 200, 349. 8 A. Yamamoto, K. Morifuji, S. Ikeda, T. Saito, Y. Uchida and A. Misono, J. Am. Chem. Soc., 1965, 87, 4652; T. Saito, Y. Uchida, A. Misono, A. Yamamoto, K. Morifuji and S. Ikeda, J. Am. Chem. Soc., 1965, 88, 5198. 9 A. Yamamoto, K. Morifuji, S. Ikeda, T. Saito, Y. Uchida and A. Misono, J. Am. Chem. Soc., 1966, 90, 5198. 10 T. Saito, Y. Uchida, A. Misono, A. Yamamoto, K. Morifuji and S. Ikeda, J. Organomet. Chem., 1966, 6, 572. 11 A. Yamamoto and S. Ikeda, J. Am. Chem. Soc., 1967, 89, 5989. 12 T. Yamamoto, A. Yamamoto and S. Ikeda, J. Am. Chem. Soc., 1971, 93, 3350. 13 T. Yamamoto, A. Yamamoto and S. Ikeda, J. Am. Chem. Soc., 1971, 93, 3360. 14 A. S. Abu-Surrah and B. Rieger, Angew. Chem., 1996, 35, 2475. 15 A. Yamamoto, T. Shimizu and S. Ikeda, Makromol. Chem., 1970, 136, 297. 16 A. Yamamoto, T. Shimizu and S. Ikeda, Polymer J., 1970, 1, 171. 17 T. Yamamoto, A. Yamamoto and S. Ikeda, Polymer Lett., 1971, 9, 281. 18 T. Yamamoto, A. Yamamoto and S. Ikeda, Bull. Chem. Soc. Jpn., 1972, 45, 1104. 19 T. Yamamoto, A. Yamamoto and S. Ikeda, Bull. Chem. Soc. Jpn., 1972, 45, 1111. 20 T. Yamamoto, A. Yamamoto and S. Ikeda, Polymer Lett., 1972, 10, 835. 21 See A. Yamamoto, Ann. N. Y. Acad. Sci., 1974, 239, 60; A. Yamamoto and S. Ikeda, in Ionic Polymerization. Unsolved Problems. eds. J. Furukawa and O. Vogel, Marcel Dekker, New York, 1976; A. Yamamoto and S. Ikeda, Prog. Polym. Sci.Jpn., 1972, 3, 49; A. Yamamoto and T. Yamamoto, J. Polym. Sci. Macromol. Rev., 1978, 13. 22 See C. M. Killian, D. J. Tempel, L. K. Johnson and M. Brookhart, J. Am. Chem. Soc., 1996, 118, 11664. 23 A. Sen, Acc. Chem. Res., 1993, 26, 303. 24 G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, P. J. Maddox, S. J. McTavish, G. A. Solan, A. J. P. White and D. J. Williams, Chem. Commun., 1998, 845. 25 B. L. Small, M. Brookhart and A. M. A. Bennett, J. Am. Chem. Soc., 1998, 120, 4049. 26 M. Uchino, A. Yamamoto and S. Ikeda, J. Organomet. Chem., 1970, 24, C63; M. Uchino, K. Asagi, A. Yamamoto and S. Ikeda, J. Organomet. Chem., 1976, 84, 93. 27 K. Tamao, K. Sumitani and M. Kumada, J. Am. Chem. Soc., 1972, 94, 4374, 9278. 28 K. Tamao, K. Sumitani, Y. Kiso, M. Zembayashi, A. Fujioka, S. Kodama, I. Nakajima, A. Minato and M. Kumada, Bull. Chem. Soc. Jpn., 1976, 49, 1958. 29 R. J. Corriu and J. P. Masse, J. Chem. Soc., Chem. Commun., 1972, 144. 30 F. Diedrich and P.J. Stang (Editors), Metal-catalyzed Cross-coupling Reactions, Wiley-VCH, Weinheim, 1998 and refs. cited therein. 31 T. Yamamoto and M. Abla, J. Organomet. Chem., 1997, 535, 209. 32 F. Ozawa, M. Fujimori, T. Yamamoto and A. Yamamoto, Organometallics 1986, 5, 2144. 33 F. Ozawa, T. Ito and A. Yamamoto, J. Am. Chem. Soc., 1980, 102, 6457; F. Ozawa, T. Ito, Y. Nakamura and A. Yamamoto, Bull. Chem. Soc. Jpn., 1981, 54, 1868; S. Komiya, Y. Morimoto, A. Yamamoto and T. Yamamoto, Organometallics, 1982, 1, 1528; S.Komiya, Y. Abe, A. Yamamoto and T. Yamamoto, Organometallics, 1983, 2, 1446; F. Ozawa, K. Kurihara, T. Yamamoto and A. Yamamoto, Bull. Chem. Soc. Jpn., 1985, 58, 399. 34 K. Tatsumi, R. HoVmann, A. Yamamoto and J. K. Stille, Bull. Chem. Soc. Jpn., 1981, 54, 1857; K. Tatsumi, A. Nakamura, S. Komiya, A. Yamamoto and T. Yamamoto, J. Am. Chem. Soc., 1984, 106, 8181. 35 A. Yamamoto, T. Yamamoto, S. Komiya and F. Ozawa, Pure Appl. Chem., 1984, 56, 1621. 36 F. Ozawa, K. Kurihara, T. Yamamoto and A. Yamamoto, J. Organomet. Chem., 1985, 279, 233. 37 T. Yamamoto, T. Kohara and A. Yamamoto, Chem. Lett., 1976, 1217; Bull. Chem. Soc. Jpn., 1981, 54, 2161. 38 F. Ozawa and A. Yamamoto, Chem. Lett., 1981, 289. 39 F. Ozawa and A. Yamamoto, Chem. Lett., 1982, 865. 40 F. Ozawa, H. Soyama, T. Yamamoto and A. Yamamoto, Tetrahedron Lett., 1982, 23, 3383. 41 T. Kobayashi and M. Tanaka, J. Organomet. Chem., 1982, 233, C64. 42 For reviews, see ref. 35; A. Yamamoto, F. Ozawa, H. Nakazawa and T. Yamamoto, New Frontiers in Organometallic and Inorganic Chemistry, eds. Y. H. Huang, A. Yamamoto and B. K. Teo, Science Press, Beijing, 1984; A. Yamamoto, T. Yamamoto and F. Ozawa, Pure Appl. Chem., 1985, 57, 1789; A. Yamamoto, F. Ozawa and T. Yamamoto, Fundamental Research in Homogeneous Catalysis, ed. A. E. Shilov, Gordon & Breach Science Publ., London, 1986; A. Yamamoto, F. Ozawa, K. Osakada, L. Huang, T. I. Son, N. Kawasaki and M.K. Doh, Pure Appl. Chem., 1991, 63, 687. 43 F. Ozawa, T. Sugimoto, T. Yamamoto and A. Yamamoto, Organometallics, 1984, 3, 692. 44 See T.-M. Huang, Y.-J. You, C.-S. Yang, W.-H. Tseng, J.-T. Chen, M.-C. Cheng and Yu. Wang, Organometallics, 1991, 10, 1026; J.-T. Chen, Yu.-S. Yeh, G.-H. Lee and Y. Wang, J. Organomet. Chem., 1991, 414, C64; J.-T. Chen, W. H. Tzeng, F. Y. Tsai, M. C. Cheng and Y. Wang, Organometallics, 1991, 10, 3954. 45 Y.-S. Lin and A. Yamamoto, Organometallics, 1998, 17, 3466; Tetrahedron Lett., 1997, 38, 3747; Bull.Chem. Soc. Jpn., 1998, 71, 723. 46 F. Ozawa, N. Kawasaki, T. Yamamoto and A. Yamamoto, Chem. Lett., 1985, 567; M. Tanaka, T. Kobayashi, T. Sakakura, H. Itatani, K. Danno and K. Zushi, J. Mol. Catal., 1985, 32, 115; M. Tanaka, T. Kobayashi and T. Sakakura, J. Chem. Soc., Chem. Commun., 1985, 537; F. Ozawa, N. Kawasaki, H. Okamoto, T. Yamamoto and A. Yamamoto, Organometallics, 1987, 6, 1640. 47 F. Kawataka, Y. Kayaki, I.Shimizu and A. Yamamoto, Organometallics, 1994, 13, 3517. 48 See (a) E. Drent, J. A. M. van Broekhoven and P. H. M. Budzelaar, Recl. Trav. Chim. Pays-Bas, 1996, 115, 263; (b) A. Sen, Acc. Chem. Res., 1993, 303; (c) M. Sperre and G. Consiglio, Chem. Ber./Recueil, 1997, 130, 1557. 49 F. C. Rix, M. Brookhart and P. S. White, J. Am. Chem. Soc., 1996, 118, 4746. 50 See J. H. Green, J. G. P. Delis, P. W. N. M. van Leeuwen and K. Vrieze, Organometallics, 1997, 16, 68. 51 For example, see D. G. Musaev, R. D. J. Froose, M. Svenson and K. Morokuma, J. Am. Chem. Soc., 1997, 119, 367; L. Deng, M. Margl and T. Ziegler, J. Am. Chem. Soc., 1997, 119, 1094. 52 A. Yamamoto, J. Organomet. Chem., 1995, 500, 337. 53 A. Yamamoto and Y. Kayaki, Synlett., to be submitted. 54 F. Kawataka, I. Shimizu and A. Yamamoto, Bull. Chem. Soc. Jpn., 1995, 68, 654. 55 Y. Kayaki, I. Shimizu and A. Yamamoto, Chem. Lett., 1995, 1089. 56 Y. Kayaki and A. Yamamoto, J. Synth. Org. Chem. Jpn. (Japanese), 1998, 56, 96. 57 Y. Kayaki, I. Shimizu and A. Yamamoto, Bull. Chem. Soc. Jpn., 1997, 70, 917. 58 Y. Kayaki, I. Shimizu and A. Yamamoto, Bull. Chem. Soc. Jpn., 1997, 70, 1135. 59 Y. Kayaki, I. Shimizu and A. Yamamoto, Bull. Chem. Soc. Jpn., 1997, 70, 1141. 60 K. Karabelas and A. Hallberg, Tetrahedron Lett., 1985, 26, 3131; K. Karabelas, C. Westerlund and A. Hallberg, J. Org. Chem., 1985, 50, 3896; K. Karabelas and A. Hallberg, J. Org. Chem., 1986, 51,J. Chem. Soc., Dalton Trans., 1999, 1027–1037 1037 5286; M. M. Abelman, T. Oh and L. E. Overman, J. Am. Chem. Soc., 1988, 110, 2328 and other recent examples. 61 For example, K. Nozaki, N. Sato, Y. Tonomura, M. Yasutomi, H. Takaya, T. Hiyama, T. Matsubara and N. Koga, J. Am. Chem. Soc., 1997, 119, 12779. 62 F. Ozawa, M. Nakano, I. Aoyama, T. Yamamoto and A. Yamamoto, J. Chem. Soc., Chem. Commun., 1986, 382. 63 H. C. Volger and K. Vrieze, J. Organomet. Chem., 1968, 13, 495; G. Carturan and G. Nardin, J. Organomet. Chem., 1990, 390, 2647. 64 F. Ozawa, T. I. Son, K. Osakada and A. Yamamoto, J. Chem. Soc., Chem. Commun., 1989, 1067. 65 A. Yamamoto, Adv. Organomet. Chem., Vol. 34, 1992, 34, 111. 66 Y.-S. Lin and A. Yamamoto, Topics in Organometallic Chemistry, Activation of Unreactive Bonds and Organic Synthesis, ed. S. Murai, Springer, Heidelberg, 1999, to be published. 67 M. Oshima, I. Shimizu, A. Yamamoto and F. Ozawa, Organometallics, 1991, 10, 1221. 68 A. Yamamoto, Bull. Chem. Soc. Jpn., 1995, 68, 433. 69 T. Terashima, I. Shimizu and A. Yamamoto, unpublished results. 70 M. Sakamoto, I. Shimizu and A. Yamamoto, Bull. Chem. Soc. Jpn., 1996, 69, 1065. 71 F. Ozawa, T. Son, S. Ebina, K. Osakada and A. Yamamoto, Organometallics, 1992, 11, 171. 72 S. Komiya, A. Yamamoto and T. Yamamoto, Chem. Lett., 1981, 115. 73 K. Nagayama, F. Kawataka, M. Sakamoto, I. Shimizu and A. Yamamoto, Chem. Lett., 1995, 367. 74 M. S. Stephan, A. J. J. M. Teunissen, G. K. M. Verzijl and J. G. de Vries, Angew. Chem., Int. Ed. Engl., 1998, 37, 662. 75 K. Nagayama, I. Shimizu and A. Yamamoto, Chem. Lett., 1998, 1143. 76 A. Yamamoto and Y. Kayaki, Synlett, to be submitted. Paper 8/08297I

ISSN:1477-9226    

DOI:10.1039/a808297i

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 1039–1040 1039 Structural and complexation properties of diselenacrown ethers. Synthesis and crystal structure of a novel cationic palladium tetraselena complex C. Bornet,a R. Amardeil,a P. Meunier *a and J. C. Daran b a Laboratoire de Synthèse et d’Electrosynthèse Organométalliques associé au CNRS (LSEO, UMR 5632), Université de Bourgogne, 6, boulevard Gabriel, 21000 Dijon, France. E-mail: Philippe.Meunier@u-bourgogne.fr b Laboratoire de Chimie de Coordination UPR-CNRS 8241, 205, route de Narbonne, 31077 Toulouse Cedex, France Received 20th January 1999, Accepted 17th February 1999 The synthesis of novel cationic structures [PdL2][Y]2 [Y 5 NO3, PF6; L 5 diselenamacrocycle] starting from PdLCl2 and L, and the first determination of the structure of such a complex by X-ray diffraction was established. There has been much interest in the use of macrocyclic ligands for complexing transition metal ions.1 Most of the work has concentrated on the design of selective receptors for either alkali or alkaline-earth cations.Other research has been devoted to selective complexations of transition metal cations. Over the past 10 years, the main concern has certainly been the preparation of metallomacrocycles with hard and soft metals.2 The major aim of these studies is the construction of supramolecular species with well defined shapes and geometries. In particular, it has been shown that complexes with hard and soft metal cations can be used for bimetallic catalysis and activation, changing in particular the redox properties of the transition metal cation.To our knowledge, similar research has not been performed in selenium chemistry. As a preliminary result, we will show here the potential application of diselenacrown ethers whose synthesis we described earlier.3 Complexes of [16]aneSe4 with Pd(II) and Pt(II) have been reported.4 These results have shown that transition metal ions are readily inserted into the cavity of the tetraselenacrown ether [16]aneSe4 to yield complexes with a square planar Se4 donor set, such as [Pd([16]aneSe4)][PF6]2. The complex, [Pd([8]- aneSe2)2][PF6]2, constructed with two bidentate selenoether macrocycles, has been recently obtained and characterized by multinuclear NMR spectroscopy (1H, 77Se-{1H}).5 We now report the preparation of new cationic palladium complexes binding to two selenamacrocyclic ligands and, to our knowledge, the first structure of such a complex, determined by single-crystal X-ray analyses.The reaction (see Scheme 1) of L (95 mg, 0.24 mmol) with one equivalent of Li2PdCl4 (63 mg, 0.24 mmol) in methanol (10 cm3) at room temperature (RT) caused the immediate formation of the neutral species PdLCl2, in the form of an orange precipitate in high yield (80%). Recrystallisation of the product was realized by the diVusion of diethyl diether into a dimethyl Scheme 1 O Se Se O O O Se Se O O PdCl2 Li2PdCl4 MeOH, RT PdLCl2 L sulfoxide solution of the palladium complex.As the neutral compound is insoluble in most common organic solvents, its spectroscopic characterisation was done in CD3NO2 or in (CD3)2SO. 1H NMR spectroscopy (200 MHz) is an eYcient probe for the diVerentiation of the palladium complex from the macrocyclic ligand L. The most important diVerences appear in the aliphatic region. In fact, the spectrum of the palladium complex was not well resolved with several signals (for the macrocyclic ligand itself, two triplets and a multiplet were observed). The high field (500 MHz) NMR spectrum showed eight signals (ABCD system).This indicated that the two faces of the complex are anisochronous and that each proton on a given carbon is not equivalent with its neighbours. Moreover, only eight signals were obtained due to the symmetry properties of the complex. However no change in morphology occurred in the aromatic region.As expected, we saw two doublets of doublets consistent with an ortho-disubstitued phenyl group, each resonance being shifted downfield (Dd 10.5 ppm). This spectroscopic result suggests that the metal is probably not located in the cavity of the macrocyclic ligand, consistent with the metal’s square planar geometry and the dimensions of the oxygenated ring. The 77Se-{1H} NMR is also evidence for complexation. The 77Se-{1H} spectra were obtained for L and PdLCl2 in (CD3)2SO solution and showed singlets at d(Se) 1273 and 1522 respectively; for the palladium complex a large shift in d(Se) to high frequency is observed.6 This shift is typical and has been observed with many other d-block complexes with selenoether ligands.7 Elemental analytical data are in accordance with the formulation PdLCl2 (Found: C, 29.33; H, 3.47.C14H20Se2PdO3Cl2 requires C, 29.42; H, 3.53%). We have also demonstrated that the neutral palladium complex PdLCl2 is an eYcient precursor for the synthesis of a selenocationic palladium complex [PdL2][Y]2 [Y = NO3, PF6].Two diVerent experimental methods were used (see Scheme 2) to provide the new selenocationic palladium complex [PdL2]21: (a) displacement of Cl2 ligands from PdLCl2 (110 mg, 0.189 mmol) occurs in refluxing MeNO2 solution (34 cm3) in the presence of the donor ligand L (75 mg, 0.189 mmol), followed by addition of NH4PF6 (62 mg, 0.378 mmol) to yield [PdL2][PF6]2 Scheme 2 [PdL2][NO3]2 (40%) 2AgNO3, MeNO2 RT L, MeNO2 (a) (b) [PdL2][PF6]2 (75%) (1) L, MeNO2 reflux (2) NH4PF6 PdLCl2 PdL[NO3]2 PdLCl21040 J.Chem. Soc., Dalton Trans., 1999, 1039–1040 (yield: 168 mg, 75%) (Found: C, 28.81; H, 3.61. C28H40Se4O6- PdP2F6 require C, 28.39; H, 3.40%) and (b) by treatment of PdLCl2 (44 mg, 7.75 × 1022 mmol) with two equivalents of AgNO3 (22 mg, 0.15 mmol) in nitromethane (10 cm3), followed by removal of AgCl and addition of one equivalent of L (31 mg, 7.75 × 1022 mmol) (yield: 31.5 mg, 40%).The 77Se and 1H NMR spectra were obtained as previously in (CD3)2SO or CD3NO2 solution. Important diVerences could be observed in the 1H NMR spectra of the neutral and cationic complexes. In particular, in cationic species, only five separate broad bands were observed in the range d 3.17–4.21 integrating each for 1, 2, 3, 1 and 1 protons respectively. We also notice that the resonances of the aromatic protons are shifted downfield (0.1–0.2 ppm) in changing from PdLCl2 to [PdL2][Y]2. The 77Se-{1H} spectra (CD3NO2) also showed significant diVerences: d 1510 for PdLCl2 (d 1476 for PdL[NO3]2); d 1456 for [PdL2][NO3]2 and d 1453 for [PdL2][PF6]2. These compare favorably with similar data observed for Pd([8]aneSe2)Cl2 and [Pd([8]- aneSe2)2][PF6]2 whose signals appear at d 1199 and d 1164 respectively.5 Here again, we note that these compounds show shifts to high frequency on co-ordination to palladium [free L: d (77Se) 1268].In order to confirm the formulation [PdL2][Y]2 and establish the geometry of the complex formed, a single crystal structure determination was carried out. Orange single crystals were obtained by slow diVusion of Et2O vapour into a solution of [PdL2][PF6]2 in CH3NO2.† The single crystal X-ray Fig. 1 A CAMERON9 drawing of [PdL2][PF6]2. Hydrogen atoms are omitted for clarity. Selected bond distances (Å) and angles (8): Pd(1)– Se(1) 2.4191(5), Pd(1)–Se(2) 2.4183(5), Se(1)–Pd(1)–Se(2) 90.19(2), Se(1)–Pd(1)–Se(29) 89.81(2).structure determination showed a square planar arrangement of the four selenium atom donors around the central metal ion. The Pd–Se bond distances (see Fig. 1) were comparable with previous values [cf. Pd–Se, 2.423(1) and 2.432(1) Å in [Pd([16]aneSe4)][BF4]2;8 2.428(1) and 2.435(1) Å in {Pd([16]- aneSe4)][PF6]2}.4 The two oxygenated cavities in the cationic complex [PdL2][PF6]2 adopted a trans configuration compared to the plane defined by the two benzene rings and the four selenium atoms.With a view to obtaining the cis configuration for the oxygen cavities, we are currently investigating similar chemistry starting from tetraselenacrown ethers. Notes and references † Crystal data: C14H20O3Pd0.5Se2F6PC2H5O0.5, M = 629.45, monoclinic, space group C2/c (no. 15), a = 13.848(1), b = 15.917(2), c = 19.785(1) Å, b = 95.49(1)8, U = 4340.0(6) Å3, Z = 4, m = 39.071 cm21, T = 180(2) K, R = 0.0414 and Rw = 0.0489 for 2589 reflections [I > 2s(I)].CCDC reference number 186/1358. See http://www.rsc.org/suppdata/dt/1999/ 1039/ for crystallographic files in .cif format. 1 L. F. Lindoy, The Chemistry of Macrocyclic Ligand Complexes, Cambridge University Press, New York, 1989; P. K. Baker, M. C. Durrant, S. D. Harris, D. L. Hughes and R. L. Richards, J. Chem. Soc., Dalton Trans., 1997, 509; S. J. A. Pope, N. R. Champness and G. Reid, J. Chem. Soc., Dalton Trans., 1997, 1639; W.Levason, J. J. Quirk and G. Reid, J. Chem. Soc., Dalton Trans., 1997, 3719. 2 A. Varshney and G. M. Gray, Inorg. Chem., 1991, 30, 1748; A. Varshney, M. L. Webster and G. M. Gray, Inorg. Chem., 1992, 31, 2580; J. Powell, A. Lough and F. Wang, Organometallics, 1992, 11, 2289; G. M. Gray and C. H. DuVey, Organometallics, 1995, 14, 245; F. C. J. M. Van Veggel, W. Verboom and D. N. Reinhoudt, Chem. Rev., 1994, 94, 279; A. Mazouz, P. Meunier, M. M. Kubicki, B. Hanquet, R. Amardeil, C. Bornet and A. Zahidi, J. Chem. Soc., Dalton Trans., 1997, 1043. 3 A. Mazouz, J. Bodiguel, P. Meunier and B. Gautheron, Phosphorus Sulfur Silicon Rel. Elem., 1991, 61, 247. 4 N. R. Champness, P. F. Kelly, W. Levason, G. Reid, A. M. Z. Slawin and D. J. Williams, Inorg. Chem., 1995, 34, 651. 5 N. R. Champness, W. Levason, J. J. Quirk and G. Reid, Polyhedron, 1995, 14, 2753. 6 D. G. Booth, W. Levason, J. J. Quirk, G. Reid and S. M. Smith, J. Chem. Soc., Dalton Trans., 1997, 3493. 7 E. G. Hope and W. Levason, Coord. Chem. Rev., 1993, 122, 109. 8 R. J. Batchelor, F. W. B. Einstein, I. D. Gay, J. Gu, B. M. Pinto and X. Zhou, Inorg. Chem., 1996, 35, 3667. 9 D. J. Watkin, C. K. Prout and L. J. Pearce, CAMERON, Chemical Crystallography Laboratory, University of Oxford, 1996. Communication 9/00545E

ISSN:1477-9226    

DOI:10.1039/a900545e

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 1041–1042 1041 The synthesis and structure of a neutral tetranuclear zinc(II) complex [Zn4(L)4] [LH2 5 N,N-bis(2-mercaptoethyl)benzylamine] Douglas J. E. Spencer,a Alexander J. Blake,a Simon Parsons b and Martin Schröder *a a School of Chemistry, The University of Nottingham, Nottingham, UK NG7 2RD b Department of Chemistry, The University of Edinburgh, West Mains Road, Edinburgh, UK EH9 3JJ Received 2nd November 1998, Revised manuscript received 16th February 1999, Accepted 16th February 1999 The reaction of benzylamine with ethylene sulfide yields the monoamine–dithiol N,N-bis(2-mercaptoethyl)benzylamine, LH2.Reaction of Na2L with Zn(BF4)2 affords the neutral tetranuclear complex [Zn4L4], which shows an unusual Zn4S4 metallacyclic structure. There are a number of metalloenzymes that incorporate Zn(II) thiolates in their prosthetic group. These include alcohol dehydrogenases,1,2 metallothioneins,3 and zinc fingers.4 The study of thiolate complexes of Zn(II) as models for these enzymes, particularly of LADH (liver alcohol dehydrogenase),2 has therefore become an active area of investigation.5–7 Complexes of thiolates with Zn(II) show interesting co-ordination chemistry because of their tendency to form oligomers and clusters 8,9 comprising cages or aggregates of [ZnS4] units formed via S-bridging between metal centres.Other structures including binuclear 9,10 and tetrameric 11 species have been reported. Monomeric compounds featuring Zn(II) co-ordinated to an NS2-donor set have also been reported as models for the active centre of LADH.6,12,13 These are rare examples of nonpolymeric neutral Zn(II) complexes of the type Zn(SR)2 containing tetrahedrally co-ordinated Zn(II) ions bridged by RS2.As part of a study of thiolate-bridged complexes of biological significance,7,14 we report herein a novel, neutral tetranuclear Zn(II) system with bridging thiolates.The sodium salt of the ligand N,N-bis(2-mercaptoethyl)- benzylamine (LH2) 15,16 (Scheme 1) reacts with Zn(BF4)2 in a 1 : 1 molar ratio in THF to give a white solid after removal of the solvent.† Colourless crystals suitable for crystallographic studies were grown by diVusion of Et2O vapour into a solution of the complex in CHCl3. Elemental analysis, IR spectroscopy and FAB mass spectrometry confirm the product to have the stoichiometry [Zn4L4] and this was confirmed by X-ray diVraction studies on a single crystal of the diethyl ether hemi-solvate.‡ The complex contains four crystallographically independent Zn(II) centres (Fig. 1), each co-ordinated by an equivalent [NS3] donor set comprising a tertiary amine from one ligand [Zn–N 2.109(3)–2.161(4) Å], one thiolate donor from the same ligand co-ordinated terminally [Zn–Sterminal 2.261(4) Å] and a second thiolate bridging two Zn(II) ions in two diVerent [ZnL] units [Zn–Sbridging 2.313(2)–2.3572(14) Å].Therefore, an overall Zn4S4 metallacycle is formed within the structure. Most of the previously reported structures of Zn(II) with mixed amine– thiolate ligands involve the N-donors as part of a heteroaromatic ring (pyrazoles, imidazoles and pyridines).5,6,12,13,17 In this Scheme 1 N SH HS S NH2 LH2 C6H6 case we observe a rare example9,11,12 of aliphatic amine group functionality at Zn(II). Lippard and co-workers have reported 11 a tetranuclear Zn(II) complex with the ligand N,N9-dimethyl- N,N9-bis(2-mercaptoethyl)ethylenediamine (L1H2) (see below) to give [Zn4Cl4(L1)2], while Darensbourg and co-workers have prepared 9 a neutral binuclear Zn(II) complex [Zn2(L2)2] [L2H2 = N,N9-bis(2-mercaptoethyl)-1,5-diazacyclooctane] (see below).The bond distances in both of these structures 9,11 are similar to those in [Zn4(L)4] with Zn–N bond lengths of 2.079(11), 2.103(11) Å, Zn–Sbridging 2.284(4)–2.356(4) Å,11 and Zn–N 2.231(2)–2.255(2) Å, Zn–Sterminal 2.327(1) Å, Zn–Sbridging 2.394(1)–2.494(1) Å, respectively.9 Fig. 1 Complementary views of the structure of [Zn4(L)4]?0.5Et2O with numbering scheme adopted. Hydrogen atoms and solvent molecule are omitted for clarity. (a) View approximately along a axis; (b) view approximately along c axis. N N SH HS N N SH HS L2H2 L1H21042 J. Chem. Soc., Dalton Trans., 1999, 1041–1042 Interestingly, tetrahedral co-ordination at Zn(II) in [Zn4(L)4] is highly distorted with rather acute N–Zn–Sterminal and N–Zn– Sbridging angles, N1B–Zn1–S7B 91.27(10), N1C–Zn2–S7C 92.62(11), N1D–Zn3–S7D 91.97(11), N1A–Zn4–S7A 91.70(11), N1B–Zn1–S4B 88.86(11), N1C–Zn2–S4C 91.17(11), N1D–Zn3–S4D 90.35(10), N1A–Zn4–S4A 88.69(10)8, and expanded Sbridging–Zn–Sterminal angles, S7B–Zn1– S4B 125.93(6), S4C–Zn2–S7C 125.03(6), S4D–Zn3–S7D 122.29(5), S4A–Zn4–S7A 129.79(5)8.This distortion is probably due to the steric factors inherent in the formation of the tetranuclear complex and within individual [ZnL] units.The aromatic rings orientate themselves exo to the central Zn4S8 core and are arranged alternately up and down (Fig. 1b) due to the inversion of successive ligand units around the metallocyclic [Zn4S4] centre. This also reduces the steric interactions between the aromatic rings. There is no evidence of p–p stacking either within or between molecules. FAB and electrospray mass spectrometry confirm the integrity of the complex, at least in part, in solution with molecular ions observed for the monomer, dimer, trimer and tetramer.18 Current work is aimed at further developing thiolate chemistry at Zn(II) and related biologically relevant metal ions.Acknowledgements We thank the EPSRC for support and the EPSRC Centre for mass spectrometry at the University of Swansea. Notes and references † Synthesis of LH2 and Na2L. Benzylamine (5 g, 0.0467 mol) in benzene (5 cm3) was placed in a Schlenk tube flushed with N2. Ethylene sulfide (5.9 g, 0.098 mol) in benzene (5 cm3) was added dropwise and the resulting solution stirred at 65 8C.After 48 h analysis by 1H and 13C NMR spectroscopy confirmed that a mixture of starting material and mono-substituted product was present. A further two equivalents of ethylene sulfide (5.9 g, 0.098 mol) were added and the solution left stirring under N2 at 65 8C. After a further 48 h, analysis by NMR spectroscopy revealed that the reaction had gone to completion. The bulk solution was filtered and the excess solvent removed in vacuo to yield a foul-smelling yellow oil.The oil was redissolved in CH2Cl2, the solution filtered through a plug of silica to remove polymeric impurities, and the excess solvent removed in vacuo to yield a clear oil (5.32 g, 0.023 mol, 52%) which was stored under N2. IR spectroscopy nmax/cm21 (neat) 3059w, 3025w, 2962m, 2935m, 2803m, 2552w, 1600w, 1493m, 1369w, 1293w, 1260m, 1109m, 1028, 734, 698m (Found: C, 57.35; H, 7.71; N, 5.81.C11H17NS2 requires C, 58.15; H, 7.49; N, 5.81%). dH (CDCl3) 1.68 (2H, s, CH2SH), 2.64 (4H, m, NCH2CH2SH), 2.70 (4H, m, NCH2CH2SH), 3.64 (2H, s, PhCH2N) and 7.34 (5H, m, H of Ph). dC (CDCl3) 22.87 (CH2SH), 57.13 (NCH2CH2SH), 58.63 (PhCH2N), 127.28, 128.42 and 128.92 (CH of Ph) and 138.91 (ipso C). m/z (EI) 225 (M1). CAUTION: The ligand has been found to cause severe allergic reactions and contact with skin should be avoided. Na2L was prepared in quantitative yield by reaction of NaH (0.127 g, 5.29 mmol) with LH2 (0.4 g, 1.76 mmol) in THF.Preparation of [Zn4L4]. Reaction of Na2L with Zn(BF4)2 (1 : 1 molar ratio) in THF gave a white solid after removal of the solvent. The solid was dissolved in CHCl3 and the solution filtered to remove sodium salts. The solution was reduced in volume and the complex crystallised by addition of Et2O (Found: C, 44.25; H, 5.61; N, 4.44. Calc. for C46H65N4S8O0.5Zn4: C, 44.04; H, 5.42; N, 4.67%). IR (KBr)/cm21: 3025w, 2921w, 2849w, 1629s, 1494s, 1452s, 1310m, 1095m, 1003m, 825m, 721w, 668w. m/z (1ve FAB) 1156 (64Zn4L4)1, 868 (64Zn3L3 1 1)1, 579 (64Zn2L2 1 1)1 with correct isotopic distribution.m/z (1ve ES) 1161 (M1), 874 and 581. ‡ Crystal data: C44H60N4S8Zn4?0.5C4H10O, M = 1199.98, triclinic, space group P1� , a = 13.926(3), b = 14.593(3), c = 14.631(5) Å, a = 88.90(2), b = 88.70(2), g = 62.674(14)8, U = 2640.7(12) Å3, T = 220 K, Z = 2, Dc = 1.509 g cm23, l(Cu-Ka) = 1.54184 Å, m = 5.297 mm21. 9255 unique data measured and used in all calculations. A molecule of Et2O was found to be half-occupied and disordered over two sites. Final wR(F2) was 0.104, R1 = 0.0448. CCDC reference number 186/1354. See http:// www.rsc.org/suppdata/dt/1999/1041 for crystallographic files in .cif format. 1 H. Eklund and C.-I. Bränden, in Zinc Enzymes, ed. T. G. Spiro, Wiley, New York, 1983, p. 124. 2 Y. Pocker, in Metal Ions In Biological Systems, ed. H. Sigel and A. Sigel, Dekker, New York, 1989, vol. 25, p. 335; H. Eklund and C.-I. Bränden, in Active Sites of Enzymes, ed. F. A. Jurnak and A. McPherson, Biological Macromolecules and Assemblies, Wiley, New York, 1987, vol. 3, ch. 2. 3 J. H. R. Kägi, S. R. Himmelhoch, P. D. Whanger, J. L. Bethune and B. L. Vallee, J. Biol. Chem., 1973, 39, 127. 4 W. Kaim and B. Schwederski, Bioinorganic Chemistry: Inorganic Elements In the Chemistry of Life, Wiley, New York, 1991, ch. 12. 5 B. Kaptein, L. Wang-GriYn, G.Barf and R. M. Kellogg, J. Chem. Soc., Chem. Commun., 1987, 1457; B. Kaptein, G. Barf, R. M. Kellogg and F. Van Bolhuis, J. Org. Chem., 1990, 55, 1890; R. M. Kellogg and R. P. Hof, J. Chem. Soc., Perkin Trans. 1, 1996, 1651 and refs. therein. 6 C. Kimblin, T. Hascall and G. Parkin, Inorg. Chem., 1997, 36, 5680 and refs. therein. 7 A. J. Blake, A. Marin-Becerra, N. D. J. Branscombe, W.-S. Li, S. Parsons, L. Ruiz-Ramirez and M. Schröder, Chem. Commun., 1996, 2573. 8 K. S. Hagen, D.W. Stephan and R. H. Holm, Inorg. Chem., 1982, 21, 3928; A. Choy, D. Craig, I. Dance and M. Scudder, J. Chem. Soc., Chem. Commun., 1982, 1246; I. Dance, J. Chem. Soc., Chem. Commun., 1980, 818; I. Dance, J. Am. Chem. Soc., 1980, 102, 3445; J. L. Hencher, M. A. Kahn, F. F. Said and D. G. Tuck, Polyhedron, 1985, 4, 1263; R. H. Holm and M. J. O’Connor, Prog. Inorg. Chem., 1971, 14, 241; F. F. Sai and D. G. Tuck, Inorg. Chim. Acta, 1982, 59, 1. 9 T. Tuntulani, J. H. Reibenspies, P. J. Farmer and M. Y. Darensbourg, Inorg. Chem., 1992, 31, 3497. 10 D. C. Goodman, T. Tuntulani, P. J. Farmer, M. Y. Darensbourg and J. H. Reibenspies, Angew. Chem., Int. Ed. Engl., 1993, 32, 116; C. A. Grapperhaus, T. Tuntulani, J. H. Reibenspies and M. Y. Darensbourg, Inorg. Chem., 1998, 37, 4052. 11 W. J. Hu, D. Barton and S. J. Lippard, J. Am. Chem. Soc., 1973, 95, 467. 12 S. C. Shoner, K. J. Humphreys, D. Barnhart and J. A. Kovacs, Inorg. Chem., 1995, 34, 5933. 13 D. T. Corwin, Jr. and S. A. Koch, Inorg. Chem., 1988, 27, 493. 14 A. J. Atkins, A. J. Blake and M. Schröder, J. Chem. Soc., Chem. Commun., 1993, 1662; A. J. Atkins, A. J. Blake, D. Black, A. Marin- Becerra, S. Parsons, L. Ruiz-Ramirez and M. Schröder, Chem. Commun., 1996, 457. 15 G. J. Colpas, M. Kumar, R. O. Day and M. J. Maroney, Inorg. Chem., 1990, 29, 4779. 16 S. A. Mirza, M. A. Pressler, M. Kumar, R. O. Day and M. J. Maroney, Inorg. Chem., 1993, 32, 977. 17 L. F. Lindoy and D. H. Busch, J. Chem. Soc., Chem. Commun., 1972, 683. 18 C. K. Meng and J. B. Fenn, Org. Mass Spectrom., 1991, 26, 542. Communication 9/0125

ISSN:1477-9226    

DOI:10.1039/a901251f

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 1043–1044 1043 A leaving group strategy for the selective functionalisation of an imido Sn(II) cubane Belén Galán, Marta E. G. Mosquera,* Julie S. Palmer, Paul R. Raithby and Dominic S. Wright * Chemistry Department, University of Cambridge, Lensfield Road, Cambridge, UK CB2 1EW. E-mail: dsw1000@cus.cam.ac.uk Received 23rd February 1999, Accepted 24th February 1999 Controlled hydrolysis of the cubane [SnNtBu]4 with H2O in thf–MeCN provides a direct route to the oxo complex [Sn4(NtBu)3O], reaction of which with naphNHLi (naph 5 1-naphthyl) gives the heteroleptic cubane [Sn4- (NtBu)3(Nnaph)] as the sole product.In original studies by Veith,1 imido Sn(II) cubanes of the type [SnNR]4 1 were found to be readily accessible from the reaction of [{Me2SiNtBu}2Sn] with primary amines.2 These species prove to be valuable precursors for the preparation of heterometallic complexes containing Sn(II) imido and phosphinidene anions, such as [Li?thf ]4[{Sn(m-PCy)}2(m-PCy)]2 in which the four Li1 cations are coordinated by a metallacyclic [{Sn- (m-PCy)}2(m-PCy)]2 42 unit.3 However, although it has been found that the intact oxo cubane [Sn4(NtBu)3O] 2 can function as an ether-like donor ligand in the adduct [{Sn4(NtBu)3O}? AlMe3],4 nothing has so far been reported concerning the reactivity or synthetic utility of the cubane itself.A current interest of ours has involved uncovering new synthetic strategies which will allow the selective synthesis of a range of main group complexes.5 In contrast to the previous report that the oxo cubane [Sn4(NtBu)3O] 2 cannot be obtained from the hydrolysis of [SnNtBu]4 1a with H2O,5a we have now found that 2 is readily prepared in high yield from this reaction if MeCN–thf is employed as the solvent (rather than solely thf as used in the previous study).This discovery facilitates a leaving group strategy by which imido cubanes 1 can be converted regioselectively into monosubstituted, heteroleptic cubanes [Sn4(NR)3(NR9)], as exemplified by the formation of [Sn4- (NtBu)3(Nnaph)] 3 (naph = 1-naphthyl) from the controlled hydrolysis of [SnNtBu]4 with H2O followed by the substitution of the oxo group of 2 with naphNHLi.† The latter reaction relies on the greater polarity of the Sn–O bonds compared to the Sn–N bonds in 2 and is driven thermodynamically by the formation of LiOH (Scheme 1).It is noteworthy that 3 cannot be obtained from the reaction of the cubane 1 with naphNHLi under similar conditions and that the reaction of 1 with excess naphNHLi results in [Li(thf )4]1[(tBuN)Sn3(Nnaph)3Li?thf ]2 only after reflux.3 As far as we are aware, although hydrolytic substitution reactions of organo–oxo compounds of Sn(IV) with amines and alcohols have been employed in the preparation of amide and alkoxide complexes,6 the reaction of a metallated primary amine or similar species (with the elimination of LiOH) is a novel one.The low-temperature crystal structure of 3‡ shows it to consist of discrete cubane units, [Sn4(NtBu)3(Nnaph)] (Fig. 1). Despite the incorporation of one diVerent imido substituent into the cubane framework, only minor distortions in the Sn4N4 core have been introduced. The internal angles at the Sn (mean 81.18) and N (mean 98.38) centres are similar to those observed previously in [SnNtBu]4 1d and other homoleptic cubanes.2 However, although the majority of the Sn–N bonds of the core fall in a similar range to those present in [SnNtBu]4,1d the bonds to the aryl–imido group are on the whole longer [Sn(2,3,4)– N(4) 2.218(4)–2.224(4) Å; cf. 2.181(4)–2.203(4) Å for the other Sn–N bonds]. This pattern can be seen as arising from the greater electron acceptor ability of the naph group compared to the tBu groups, allowing some degree of dispersion of the negative charge on the naphN imido centre into the aromatic Fig. 1 ORTEP10 drawing of the structure of 2. Thermal ellipsoids are at the 40% probability level. Selected bond lengths (Å) and angles (8): Sn(1)–N(1) 2.197(4), Sn(1)–N(2) 2.203(4), Sn(1)–N(3) 2.199(4), Sn(2)– N(1) 2.185(4), Sn(2)–N(2) 2.187(4), Sn(2)–N(4) 2.224(4), Sn(3)–N(2) 2.199(4), Sn(3)–N(3) 2.181(4), Sn(3)–N(4) 2.218(4), Sn(4)–N(1) 2.194(4), Sn(4)–N(3) 2.191(4), Sn(4)–N(4) 2.219(4), C(4)–N(4) 1.415(7); range N–Sn–N 80.7(1)–81.7(2) (mean 81.1), range Sn–N–Sn 97.6(2)– 99.7(2) (mean 98.3).Scheme 1 (i) H2O (1 equivalent), MeCN–thf, 278 8C, 2tBuNH2; (ii) naphNHLi, thf, 2LiOH.1044 J. Chem. Soc., Dalton Trans., 1999, 1043–1044 substituent and resulting in correspondingly weaker Sn–N bonds. The synthetic strategy outlined above furnishes a potential route to a range of compounds of the type [Sn4(NR)3X] (e.g., X = NH, PH, S) not previously accessible, using the corresponding cubane precursors [SnNR]4 which are readily prepared. Complex 3 is the first example of a heteroleptic imido Sn(II) cubane to be prepared and structurally characterised. The synthetic methodology involved (a leaving group strategy which is related to that commonly employed in organic synthesis) provides a rare if not unprecedented example of the selective structural modification of an oligomeric main group cage.Acknowledgements We gratefully acknowledge the EPSRC (J. S. P., P. R. R., D. S. W.) and the EU (fellowship for M. E. G. M.) and the Spanish Government (B.G.) for financial support. Notes and references † Synthesis of 2. [SnNtBu]4 1a (1.25 mmol) was prepared by the in situ reaction of tBuNH2 (0.53 ml, 5.0 mmol) with [Sn(NMe2)2] 7 (1.03 g, 5 mmol) in thf (20 ml). To this solution at 278 8C was added dropwise a solution of H2O (0.02 ml, 1.1 mmol) in MeCN (10 ml, distilled over CaH2). After full addition the mixture was allowed to warm slowly to room temperature and stirred (2 h) before filtration to remove a white precipitate.The solvent was removed under vacuum to obtain 2 as a lemon yellow powder. Yields of up to 0.78 g (89%) were obtained using this method. 1H NMR (250 MHz, d6-benzene, 125 8C): d 1.34 (Me of tBu) (Found: C, 20.2; H, 3.9; N, 5.8. Calc. for [Sn4ON3C12H27]: C, 20.5; H, 3.8; N, 6.0%). Synthesis of 3. A solution of naphNHLi (0.71 mmol) was prepared by the addition of nBuLi (0.47 ml, 1.5 mol dm23 solution in hexanes) to a solution of naphNH2 (0.102 g, 0.71 mmol) in toluene (5 ml)–thf (5 ml).The solution was added to a solution of 2 (0.59 g, 0.71 mmol). After stirring at room temperature (30 min) the solution was filtered. The solvent was removed under vacuum and was replaced by thf (3 ml) and Et2O (4 ml). Storage at 215 8C (4 d) gave brown crystals of 3. Yield 0.21 (36%). 1H NMR (250 MHz, d8-toluene, 125 8C): d 8.23 [d, 1H, J = 8.2, C(2)–H], 7.72 [dd, 1H, J = 9.3, 1.5 Hz, C(9)–H], 7.32 [m, 5H, C(3,4,6,7,8)–H], 1.48 (s, 27H, tBu) (Found: C, 32.2; H, 4.2; N, 6.8.Calc. for [Sn4N4C22H34]: C, 31.8; H, 4.1; N, 5.8%). ‡ Crystal data for 3: C22H34N4Sn4, M = 829.29, monoclinic, space group P1� (no. 2), a = 8.670(5), b = 9.703(5), c = 16.815(9) Å, a = 75.71(3), b = 81.36(4), g = 81.72(4)8, U = 1347(1) Å3, Z = 2, rcalc. = 2.045 Mg m23, l = 0.71073 Å, T = 180(2) K, m(Mo-Ka) = 3.682 mm21, F(000) = 788. Data were collected on a Stoe AED diVractometer using an oil-coated rapidly-cooled crystal 8 of dimensions 0.40 × 0.28 × 0.16 mm by the w– q method (3.56 £ q £ 22.508).Of a total of 5293 collected reflections, 3505 were independent (Rint = 0.032). The structure was solved by direct methods and refined by full-matrix least-squares on F2 to final, values of R1[F > 4s(F)] = 0.023 and wR2 = 0.057 (all data); 9 largest peak and hole in the final diVerence map 0.715 and 20.665 e Å23. CCDC reference number 186/1362. See http://www.rsc.org/suppdata/dt/1999/1043/ for crystallographic files in .cif format. 1 (a) M. Veith, M.-L. Sommer and D. Jäger, Chem. Ber., 1979, 112, 2581; (b) M. Veith and G. Schlemmer, Chem. Ber., 1982, 115, 2141; (c) M. Veith d M. Grosser, Z. Naturforsch., Teil B, 1982, 37, 1375; (d ) M. Veith and O. Recktenwald, Z. Naturforsch., Teil B, 1983, 38, 1054. 2 For other synthetic methods see: H. Chen, R. A. Bartlett, H. V. R. Dias, M. M. Olmstead and P. P. Power, Inorg. Chem., 1991, 30, 3390; R. E. Allan, M. A. Beswick, A. J. Edwards, M. A. Paver, P. R. Raithby, M.-A. Rennie and D. S. Wright, J. Chem. Soc., Dalton Trans., 1995, 1991. 3 R. E. Allan, M. A. Beswick, N. L. Cromhout, M. A. Paver, P. R. Raithby, A. Steiner, M. Trevithick and D. S. Wright, Chem. Commun., 1996, 1501. 4 M. Veith and H. Lange, Angew. Chem., Int. Ed. Engl., 1980, 19, 401; M. Veith and W. Frank, Angew. Chem., Int. Ed. Engl., 1985, 24, 223. 5 (a) M. A. Beswick, M. E. G. Mosquera and D. S. Wright, J. Chem. Soc., Dalton Trans., 1998, 2437; (b) M. A. Beswick and D. S. Wright, Coord. Chem. Rev., 1998, 176, 1373. 6 A. G. Davies, Comprehensive Organometallic Chemistry II, eds. E. W. Abell, F. G. A. Stone and G. Wilkinson, Pergamon, Oxford, 1995, vol. 2, ch. 6, p. 217. 7 M. M. Olmstead and P. P. Power, Inorg. Chem., 1984, 23, 413. 8 D. Stalke and T. Kottke, J. Appl. Crystallogr., 1993, 25, 615. 9 G. M. Sheldrick, SHELXL 93, a package for crystal structure refinement, University of Göttingen, 1993. 10 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1976. Communication 9/01478K

ISSN:1477-9226    

DOI:10.1039/a901478k

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 1045–1046 1045 The template synthesis of triaryl functionalised 1,5,9- triphosphacyclododecane on molybdenum using organocopper reagents † David J. Jones,a Peter G. Edwards,*a Robert P. Tooze b and Thomas Albers a a Department of Chemistry, Cardiff University, Cardiff, PO Box 912, UK CF1 3TB. E-mail: edwardspg@cardiff.ac.uk b ICI Acrylics PLC, PO Box 90, Wilton, Middlesbrough, Cleveland, UK TS90 8JE Received 19th January 1999, Accepted 15th February 1999 Free triphenyl substituted 1,5,9-triphosphacyclododecane has been prepared via the templated reaction of 1,5,9- trichloro-1,5,9-triphosphacyclododecane on molybdenum with phenylcopper or diphenylcuprate.Previous routes to the symmetrical 12-membered tritertiary triphosphamacrocycles ([12]aneP3R3) 1 do not lead to the successful formation of aryl functionalised macrocycles and the yields for bulky alkyl or perfluoroalkyl substituents are typically poor.2 In view of the interest in more relatively p-acidic aryl phosphines and their importance in catalysis and other applications, we have investigated alternative synthetic routes to the formation of aryl functionalised P3 macrocycles based upon the [12]aneP3 core and report preliminary observations herein.One approach to the formation of aryl substituted phosphines is via transmetallation of the chlorophosphine. Reactions to form the required P–Cl from secondary phosphines generally require forcing reaction conditions and reagents such as phosgene or phosgene equivalents.3 In addition these reactions would not be expected to work on coordinated phosphines due to the unavailability of the lone pair and coordination sites in the transition state.The chlorination of free secondary phosphines with CCl4 in the presence of Et3N has been reported; a 5-coordinate transition state was proposed. 4 Surprisingly we have found that [([12]aneP3X3)- Mo(CO)3] (where X = Cl 2 or Br 3) can be readily formed in high yield, >80%, from the reaction of [([12]aneP3H3)Mo(CO)3] 1 with the corresponding CX4 in the presence of Et3N (Scheme 1).This facile reaction is complete within 2 hours at ambient temperature and, as the 5-coordinate transition state is not available for the kinetically inert Mo(0) complex, this suggests that a free radical mechanism is involved. These rates are significantly faster than those for the free phosphines reported in the literature where reaction times of days or weeks were required.Bromination is faster than chlorination while reaction Scheme 1 Halogenation of the templated trisecondarymacrocycle by reaction with an excess of CX4. Mo P OC OC P P CO H H H P X P X P Mo CO OC OC X CX4 / Et3N 2 X = Cl 3 X = Br 1 † Supplementary data available: experimental details for complexes 2–5. For direct electronic access see http://www.rsc.org/suppdata/dt/1999/ 1045/, otherwise available from BLDSC (No.SUP 57501, 6 pp.) or the RSC Library. See Instructions for Authors, 1999, Issue 1 (http:// www.rsc.org/dalton). with CI4 is rapid at RT but a stable product could not be isolated. Reaction of 2 with PhLi or PhMgBr leads, in general, to the formation of an insoluble brown material, presumably polymeric. In one case, however, reacting 2 with PhLi in diethyl ether at 0 8C led to a 20% yield of the required product [([12]aneP3Ph3)Mo(CO)3] 4 while reaction of 2 with PhMgBr in THF at 240 8C led to a small yield of an unstable product which appeared to be fully arylated but had three inequivalent phosphorus atoms as indicated by 31P and 1H NMR.Reaction of 2 with Ph2Zn in refluxing THF gave no reaction after 24 hours, however, reaction of 2 with PhCu or Ph2CuLi in THF overnight at 30 8C or 1 hour in refluxing THF did lead to the required product 4 in high yield, >70%. These reactions are readily monitored by 31P NMR spectroscopy where a new singlet at d 10.9 due to 4 increases at the expense of the singlet at d 134 due to 2.Oxidation of 4 with I2 and decomplexation in ethanolic NaOH followed by extraction into toluene following our previously reported method 2 results in isolation of the free triphenyl-macrocycle, [12]aneP3Ph3 5, according to Scheme 2. Fig. 1 shows the crystal structure of 4 with molybdenum in a distorted octahedral environment.‡ Complex 4 crystallises with two independent molecules in the asymmetric unit diVering only in the orientation of one phenyl ring; the phenyl ring is approximately parallel to the C3v mirror plane through the phosphorus to which it is attached in one form and orthogonal in the other (Fig. 1 4b).5 There is no obvious structural reason for this orientation and it may simply be due to crystal packing forces. Five of the six b-methylene groups of the macrocycles are disordered over the two possible positions to give the pseudo-chair and -boat configurations for the adjacent P–P chelate rings with the molybdenum, e.g.P1C4C5C6P2Mo, all Scheme 2 (i) PhLi or PhMgBr–THF 240 8C to RT; (ii) 6PhCu or Ph2CuLi, THF, RT for 2 hours; (iii) I2, refluxing 1,1,2-trichloroethane, excess NaOH in MeOH, diethyl ether–H2O extraction. P Cl P Cl P Mo CO OC OC Cl P Ph P Ph P Mo CO OC OC Ph P Ph P Ph P Ph 5 4 (iii) 2 (i) or (ii)1046 J. Chem. Soc., Dalton Trans., 1999, 1045–1046 have been drawn in the chair configuration with greatest occupancy. The exception is the methylene (C38) lying in the C3v plane opposite P4 which is found only in the chair configuration.The metrical parameters are otherwise comparable to the 2-propyl analogue, [([12]aneP3Pri 3)Mo(CO)3].1d The methylene protons on both the a- and b-ring carbons are diastereotopic and could be expected to have distinct environments while structural evidence indicates they could exist in either the boat or chair conformation. In solution, a rapid equilibrium between these conformations would be expected to broaden their NMR resonances, the disorder of the b-ring carbons in the solid state structure for 4 presumably reflects a low energy diVerence between the two conformations. This expected behaviour is seen for the new complexes and for the free ligand where small diVerences in d (NMR) are observed for the diastereotopic a-methylene protons in 4 § and the a- and b-methylene protons appear as broad multiplets.These resonances are shifted significantly upfield in the free macrocycle 5 as compared to 4 especially the b-methylene peaks dH 2.05 to 1.73 and 1.57.It is interesting to note that long range phosphorus coupling is seen for the meta-carbon on the phenyl ring for 4 where a doublet of doublets is seen, this is absent for the free ligand 5 where only a doublet is observed. In conclusion we have developed a new, facile route to P–Cl bond formation for coordinated secondary phosphines. This simple transformation allows for alternative routes to functionalised macrocycles.One of these routes, notably aryl/alkylation using organocopper reagents, oVers the possibility of introducing a broad range of substituents due to the extensive range of organocopper reagents available and including options for which alternative methods have failed. The chemistry of these new ligands is currently being examined and this new class of triaryl-substituted 1,5,9-triphosphacyclododecanes should also allow a more extensive examination of structure/reactivity relationships in their metal complexes.This study is currently under investigation. Fig. 1 ZORTEP plot of 4 (50% probability ellipsoids), showing the two crystallographically independent molecules 4a and 4b. Selected average bond distances (Å) and angles (8): P–C (ring) 1.817(13), Mo–P 2.493(4), Mo–C 1.933(13), C–O 1.174(13), non-bonded P–P 3.536(6); P–Mo–P 90.37(14), C–Mo–C 90.27(5), C–Mo–P 177.38(4), R–P–Mo 116.3(4). Acknowledgements We would like to thank ICI Acrylics PLC for financial support of this project (Strategic Research Fund).Notes and references ‡ Crystal data for complex 4, C30H33MoO3P3, M = 630.41, T = 20 8C, monoclinic, space group P21/n (no. 14), a = 9.399(4), b = 17.59(3), c = 34.22(2) Å, b = 90.15(3)8, V = 5656(10) Å3, Z = 8, D = 1.481 g cm23, m(Mo-Ka) = 0.664 mm21, independent reflections = 14926 (Rint = 0.0968), 2q range for data collection 3.6 < 2q < 39.68, R1 (I > 2s) = 0.0484, wR2 (all data) = 0.0924.Equivalent 1,2- and 1,3-distances involving the disordered atoms were restrained to be equal. Rigid-bond and similarity restraints were used as well. Pseudo-orthorhombic twinning was encountered (a9 = 2a, b9 = b, c9 = c), contribution of minor component 0.1954(15). CCDC reference number 186/1351. See http:// www.rsc.org/suppdata/dt/1999/1045/ for crystallographic files in .cif format. § See supplementary data for experimental details: Selected data for complex 2: dP 134.0; dH PCH2 2.33 and 2.14, PCH2CH2 2.22 and 1.57; n(CO) cm21 1970 and 1879.Complex 3: dP 119.0; dH PCH2 2.55 and 2.27, PCH2CH2 2.21 and 1.74; n(CO) cm21 1961 and 1876. Complex 4: dP 10.9; dH PCH2 2.06 and 1.90, PCH2CH2 2.05; dC 128.26 (dd, m-Ph, 3JPC = 6, 5JPC = 3 Hz); n(CO) cm21 1917 and 1809. Compound 5, dP 234.6; dH PCH2 2.16 and 1.93, PCH2CH2 1.73 and 1.57. Analytical data: 2, found: C, 28.74; H, 3.41. Calc.for C12H18O3P3Cl3Mo: C, 28.51; H, 3.59. 3, found: C, 23.10; H, 2.78. Calc. for C12H18O3P3Br3Mo: C, 22.56; H, 2.84. 4, found: C, 56.98; H, 5.32. Calc. for C30H33O3P3Mo: C, 57.15; H, 5.25. 5, found: C, 71.99; H, 7.38. Calc. for C27H33P3: C, 71.99; H, 7.33%. Satisfactory mass spectroscopic data were obtained. 1 (a) P. G. Edwards, J. S. Fleming and S. S. Liyanage, J. Chem. Soc., Dalton Trans., 1997, 193; (b) P. G. Edwards, J. S. Fleming, S. S. Liyanage, S. J. Coles and M. B. Hursthouse, J.Chem. Soc., Dalton Trans., 1996, 1801; (c) S. J. Coles, P. G. Edwards, J. S. Fleming and M. B. Hursthouse, J. Chem. Soc., Dalton Trans., 1995, 4091; (d ) S. J. Coles, P. G. Edwards, J. S. Fleming and M. B. Hursthouse, J. Chem. Soc., Dalton Trans., 1995, 1139. 2 P. G. Edwards, J. S. Fleming and S. S. Liyanage, Inorg. Chem., 1996, 35, 4563. 3 R. Rabinowitz and J. Pellon, J. Org. Chem., 1961, 26, 4623; W. A. Henderson and S. A. Buckler, J. Am. Chem. Soc., 1960, 82, 5794; E. Steininger, Chem. Ber., 1963, 96, 3184; A. N. Pudovik, G. V. Romanov and V. M. Pozhidaev, IZV. Akad. Nauk. SSSR, Ser. Khim., 1977, 9, 2172. 4 Y. A. Veits, E. G. Neganova, M. V. Filippov, A. A. Borisenko and V. L. Foss, J. Gen. Chem. USSR (Engl. Transl.), 1961, 61, 114; Y. A. Veits, E. G. Neganova, M. V. Filippova, A. A. Borisenko and V. L. Foss, Zh. Obshch. Khim., 1991, 61, 130. 5 G. M. Sheldrick, SHELXS-94, Acta Crystallogr., Sect. A, 1990, 46, 467; G. M. Sheldrick, SHELXL-97, Program for the refinement of Crystal Structures, Universität Göttingen, 1997; L. Zsolnai and G. Huttner, ZORTEP, Universität Heidelberg, 1994; L. Guoguang, PATTERN, Program for Drawing DiVraction Patterns, Karolinska Institute, 1994. Communication 9/00509I

ISSN:1477-9226    

DOI:10.1039/a900509i

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 1047–1048 1047 A metal-organic molecular box obtained from self-assembling around uranyl ions Pierre Thuéry,*a Martine Nierlich,a Bruce W. Baldwin,b Nobuko Komatsuzaki c and Takuji Hirose d a CEA/Saclay, DRECAM/SCM (CNRS URA 331), Bât. 125, 91191 Gif-sur-Yvette, France. E-mail: thuery@drecam.cea.fr b Department of Chemistry, Spring Arbor College, Spring Arbor, Michigan 49283, USA c National Institute of Materials and Chemical Research, 1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan d Department of Applied Chemistry, Faculty of Engineering, Saitama University, 255 Shimo-Ohkubo, Urawa, Saitama 338-8570, Japan Received 1st February 1999, Accepted 18th February 1999 We used the peculiar coordination geometry of uranyl ions to build a three-dimensional superstructure of unprecedented architecture. The complexation of uranyl ions by a monoester derivative of the cis,trans regioisomer of Kemp’s triacid and Ï-peroxo bridges leads to a cagelike molecule with a roughly parallelepipedic inner cavity of ca. 700 Å3 able to host four medium-sized organic molecules. The search for nanometer-sized molecular or supramolecular containers with recognition properties is a subject of wide interest and the design of molecules defining inner cavities able to host organic molecules or ions has received a new impulse in the last years due to the use of self-assembling from noncovalent interactions, in an attempt to mimic biological processes. 1,2 Among these interactions, hydrogen bonding has been used to build some remarkable supramolecules such as Rebek’s ‘tennis ball’ 3 or Atwood’s spherical assembly.4 The other most useful non-covalent interaction is the metal ion coordination bond, with its specific geometrical requirements, such as the widely used square planar arrangement.2 Among the various architectures resulting from self-assembling around metal ions, the box shape is particularly appealing.5 We show herein how the unique coordination properties of the uranyl ion UO2 21 can be used to build nanometer-sized molecular boxes of remarkable shape and size.It is well known that the highly anisotropic linear uranyl ion requires a nearly planar equatorial environment of four to six donor atoms. We recently began to investigate the complexes of uranyl with ligands derived from Kemp’s triacid (cis,cis-1,3,5- trimethylcyclohexane-1,3,5-tricarboxylic acid) or its cis,trans regioisomer.The ligand used in this work is t-5-(4-tert-butylbenzyloxycarbonyl)- 1,3,5-trimethylcyclohexane-r-1,c-3-dicarboxylic acid, noted H2L.6 Its single-step reaction with uranyl nitrate hexahydrate in the presence of triethylamine and atmospheric oxygen led to complex 1 whose structure has been determined by X-ray crystallography.† As shown in Fig. 1, the ligand assumes a chair conformation in the complex, with the two acid groups in equatorial positions.The anionic complex core (Scheme 1 and Fig. 2) appears to be built from two macrocycles consisting of four uranyl ions and four L22 molecules each, bonded to each other by four m-peroxo O2 22 ions. The coordination geometry around the uranyl ion comprises six oxygen donor atoms in the equatorial plane, as is usual for small bite bidentate ligands. Each L22 ligand is found, as expected, to be bridging two uranyl ions by its two acid functions [mean U–O distance 2.47(5) Å]. Four structures with uranyl ions bridged by m-peroxo ions have been reported 7 and a reaction mechanism proposed.7b The O–O [mean value 1.48(5) Å] and U–O [2.33(3) Å] distances in 1 are in perfect agreement with the values already reported.The most striking point in this structure is the remarkable ability of the ligand L22 to give rise to a four-membered ring of nearly perfect rectangular shape. This can only be achieved because the angle between the two binding ‘pincers’ is not far from right angle [mean value 83(5)8].The most striking examples of metal–organic rings with right angles up to now are based on the combination of linear ligands and ‘protected’ metal ions such as PtIIen or PdIIen (where en is ethylenediamine) which require square planar environments.2 The present work illustrates the possibility of another methodology, with the metal ion defining the plane of the parallelepiped faces and the ligand providing the right angle. An example of such a methodology using trans coordination of PdII ions to build large square arrays of porphyrins has recently been reported.8 The volume of the inner cavity in 1 can be estimated from the distance between two facing uranyl oxygen atoms, 10.8(3) and 7.6(5) Å in the smaller and larger dimensions respectively, and, in the third direction, which is largely open to the outside, the length of the rigid ‘channel’ defined by the ligands without their ‘tail’, which is about 12.6 Å.The resulting value, when taking into account the oxygen ionic radius, is about 700 Å3, which is a low estimation since the distance between facing oxygen atoms defines the ‘bottleneck’ of the channel. Even more than its size, the internal organization of the cavity is notable since it displays eight bonding sites, comprised of the uranyl oxygen atoms pointing inwards. The propensity of uranyl ions to behave as hydrogen bond acceptors is well documented and has been used in the stereognostic coordination concept,9 which consists in designing uranyl-specific ligands able to provide the required Fig. 1 View of the ligand L22 in the uranyl complex 1. Oxygen atoms in black.1048 J. Chem. Soc., Dalton Trans., 1999, 1047–1048 equatorial electron pair donor atoms array plus a hydrogen bonding site. Molecule 1 illustrates the reverse approach: the eight uranyl ions provide inner hydrogen bonding sites able, in principle, to recognize any donor moiety of suitable size and assume in this way a double role, structural and functional.The synthetic procedure adopted led to the inclusion of two triethyl- Scheme 1 Schematic representation of the anionic complex core in 1. Fig. 2 Two orthogonal views of the structure of 1. All counter ions and solvent molecules omitted other than the two triethylammonium ions included in the cavity. Hydrogen bonds between nitrogen and uranyl oxygen atoms in dashed lines. Top view parallel to the larger face defined by a (UO2L)4 macroring. Lower view showing the coordination geometry around uranyl ions and the bridging m-peroxo ions.ammonium ions which occupy a central position in the cavity, each of them hydrogen bonded to two uranyl oxygen atoms {mean value of N ? ? ? O distances 2.9(1) Å, slightly larger than those previously reported for the same ions [mean value 2.76(6) Å] 10}. the presence of these triethylammonium ions is necessary for the complex to be formed, which is an infrequent phenomenon referred to as ‘guest-induced organization’ or ‘induced fit molecular recognition’.11 These two cations do not fill all the available space and two chloroform molecules (not represented on the drawing for clarity) are also included in the cavity, near its openings.The inclusion of multiple guests is obviously a requisite if one wants to make chemical use of such container molecules, for example in catalysis or to study endohedral micro-environmental chemical processes.Notes and references † Preparation of 1. When a solution of uranyl nitrate hexahydrate (0.5 mmol) in 10 ml of methanol–chloroform (1 : 1) is added to a solution of H2L (0.5 mmol) in 15 ml of methanol, no reaction occurs and crystals of H2L are deposited. When the same procedure is followed by addition of a large excess of triethylamine (1 ml), the solution, initially light yellow, becomes intensely yellow and, under convenient slow evaporation conditions, yields crystals of 1 suitable for X-ray crystallography.ES-MS. Spectra recorded on a QUATTRO II system (Micromass, UK). m/z 1479, [(L22)2(UO2 21)2(O2 22)(HNEt3 1)]2, [(L22)4- (UO2 21)4(O2 22)2(HNEt3 1)2]22 and {[(L22)8(UO2 21)8(O2 22)4(HNEt3 1)2]? [(HNEt3 1)2]}42; m/z 1093, {[(L22)8(UO2 21)8(O2 22)4(HNEt3 1)2(CHCl3)2] (NEt3)6}62; m/z 1095, [(L22)8(UO2 21)8(O2 22)3 1 H1]52. Crystal data: [(UO2 21)8(L22)8(O2 22)4(HNEt3 1)8]?5CHCl3?16H2O? 6CH3OH, C243H429Cl15N8O94U8, M = 7402.93, monoclinic, space group P2(1)/n, a = 28.2861(13), b = 40.881(3), c = 29.318(2) Å, b = 96.882(4)8, V = 33657(4) Å3, Z = 4, Dc = 1.461 g cm23, m = 4.027 mm21, F(000) = 14744, T = 123(2) K.Data collected on a Nonius Kappa-CCD area-detector diVractometer with Mo-Ka radiation. Absorption eVects empirically corrected. Structure solved by direct methods. Hydrogen atoms not included. Owing to the presence of 368 non-hydrogen atoms in the asymmetric unit and many disordered counter ions and solvent molecules, many restraints and constraints on geometrical and displacement parameters had to be applied. Refinement of 1393 parameters by full-matrix least-squares on F2 based on 29482 unique reflections (out of 91745 measured reflections), R1 = 0.146 (wR2 = 0.282).CCDC reference number 186/1361. See http:// www.rsc.org/suppdata/dt/1999/1047 for crystallographic files in .cif format. 1 P. N. W. Baxter, in Comprehensive Supramolecular Chemistry, eds.J. L. Atwood, J. E. D. Davies, D. D. MacNicol and F. Vögtle, Pergamon, Oxford, 1996, vol. 9, ch. 5. 2 M. Fujita, ibid., ch. 7; M. Fujita and K. Ogura, Coord. Chem. Rev., 1996, 148, 249; Bull. Chem. Soc. Jpn., 1996, 69, 1471; M. Fujita, Chem. Soc. Rev., 1998, 27, 417. 3 R. Wyler, J. de Mendoza and J. Rebek, Jr., Angew. Chem., Int. Ed. Engl., 1993, 32, 1699; N. Branda, R. Wyler and J. Rebek, Jr., Science, 1994, 263, 1267. 4 L. R. MacGillivray and J. L. Atwood, Nature (London), 1997, 389, 469. 5 C. A. Hunter, Angew. Chem., Int. Ed. Engl., 1995, 34, 1079. 6 B. W. Baldwin, T. Hirose, Z. H. Wang, T. Uchimaru and A. Yliniemelä, Bull. Chem. Soc. Jpn., 1997, 70, 1895. 7 (a) R. Haegele and J. C. A. Boeyens, J. Chem. Soc., Dalton Trans., 1977, 648; (b) P. Charpin, G. Folcher, M. Lance, M. Nierlich and D. Vigner, Acta Crystallogr., Sect. C, 1985, 41, 1302; (c) G. A. Doyle, D. M. L. Goodgame, A. Sinden and D. J. Williams, J. Chem. Soc., Chem. Commun., 1993, 1170; (d) D. Rose, Y. D. Chang, Q. Chen and J. Zubieta, Inorg. Chem., 1994, 33, 5167. 8 C. M. Drain, F. Nifiatis, A. Vasenko and J. D. Batteas, Angew. Chem., Int. Ed. Engl., 1998, 37, 2344. 9 T. S. Franczyk, K. R. Czerwinski and K. N. Raymond, J. Am. Chem. Soc., 1992, 114, 8138. 10 P. Thuéry, N. Keller, M. Lance, J. D. Vigner and M. Nierlich, Acta Crystallogr., Sect. C, 1995, 51, 1570; New. J. Chem., 1995, 19, 619; P. Thuéry and M. Nierlich, J. Incl. Phenom., 1997, 27, 13; J. Chem. Soc., Dalton Trans., 1997, 1481; P. Thuéry, M. Nierlich, M. I. Ogden and J. M. Harrowfield, Supramol. Chem., 1998, 9, 297. 11 M. Fujita, S. Nagao and K. Ogura, J. Am. Chem. Soc., 1995, 117, 1649. Communication 9/00845D

ISSN:1477-9226    

DOI:10.1039/a900845d

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1999, 1049–1059 1049 Synthesis of heterobinuclear metallocenes containing bridging ansa-bis-Á-cyclopentadienyl ligands † Malcolm L. H. Green * and Neil H. Popham Inorganic Chemistry Laboratory, South Parks Road, Oxford, UK OX1 3QR. E-mail: malcolm.green@chemistry.ox.ac.uk Received 22nd December 1998, Accepted 5th February 1999 Binuclear metallocene derivatives have been prepared, exemplified by [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}- Rh(CO)2], [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}Rh(PPh3)2], [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}- Mn(CO)3], [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}Ru(CO)(h-C4H7)], [Fe{m-(h-C5H4)CMe2(h-C9H6)Zr(h- C5H5)Cl2], [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}Co(h-C5Me5)], [Cl2(h-C5H5)Zr{m-(h-C5H4)C(CH2)5(h-C9H6)}- Zr(h-C5H5)Cl2], [Cl2(h-C5H5)Hf{m-(h-C5H4)(CH2)5C(h-C9H6)}Hf(h-C5H5)Cl2]. The tris-h-cyclopentadienyl compounds [Zr{Me2C(h-C5H4)2(h-C5H4Me)Cl}] and the new h-indenyl compound [Zr{(4-ButC6H9)(h-C5H4)- (h2-C9H6)}(h-C5H5)Cl] are also described.The activity of a selection of the binuclear compounds as co-catalysts for ethylene polymerisation has been demonstrated. Metallocenes of the Group IV elements, especially of zirconocenes, are of considerable interest as homogeneous cocatalysts for the polymerisation of a-olefins.1,2 The use of ringsubstituted h-cyclopentadienyl ligands in metallocenes has given rise to industrially useful isospecificities. For example, the ring substituents can control whether the polymerisation of propene gives iso- or syndio-tactic polymers.2 There are claims that ring substituents can control the activity of substituted metallocene polymers.3 It is true that alkyl and other substituents on cyclopentadienyl rings of metallocenes can modify the activity of the metallocene catalyst and also they can alter the specificity of the mechanism for olefin polymerisation.However, there is a subtle interplay between the steric and electronic eVects of ring substituents that at present does not allow accurate a priori prediction of ring-substitution eVects and these can only be determined by experiment.Therefore, there is active research into a variety of ring-substituted cyclopentadienyl metallocenes in order to identify optimum compounds for use as catalyst precursors. In this context we have set out to synthesize and study the catalytic properties of binuclear metallocenes, as described below.Previous studies on the synthesis of binuclear metallocenes were initiated by Petersen 4 who reported that the reaction of the lithium salts Li2[X(h-C5H4)2] (where X = CH2 or SiMe2) with [Zr(h-C5H5)Cl3]?2THF or [Zr(h-C5Me5)Cl3] gave the binuclear compounds [{Zr(h-C5R5)Cl2}2{X(h-C5H4)2}] (where R = H or Me). Reaction of this tetrachloride compound with LiAl- (OBut)3H yield the hydride-bridged complexes [{Zr(h-C5R5)- Cl}2{X(h-C5H4)2}(m-H)2].5 The related sulfide bridged compound [{Zr(h-C5R5)}2{X(h-C5H4)2}(m-S)2] has been described.6 Binuclear zirconocenes with phenyl bridged ansa-ligands have been synthesized.7 The thallium salt Tl2[Me2Si(h-C5H4)2] with TiCl4 yields the binuclear complex [(TiCl3)2{Me2Si(h-C5- H4)2}].8,9 Nifant’ev et al.10 reported the ligand exchange reaction between [TiCl2{Me2X(h-C5H4)2}] (where X = C or Si) and TiCl4 to give [(TiCl3)2{Me2X(h-C5H4)2}].Related compounds in which the bridging system between the two cyclopentadienyl rings is (SiMe2)2 have been described.8,9,11 † Supplementary data available: full analytical and spectroscopic data.For direct electronic access see http://www.rsc.org/suppdata/dt/1999/ 1049/, otherwise available from BLDSC (No. SUP 57500, 12 pp.) or the RSC Library. See Instructions for Authors, 1999, Issue 1 (http:// www.rsc.org/dalton). Nifant’ev et al.12 reported the monotransmetallation reaction of the ansa-bridged dimethyltin complex [SnMe2{Me2Si- (h-C5H4)2}] with [Zr(h-C5H5)Cl3]?2THF or [Ti(h-C5H5)Cl3] to give [Cl2(h-C5H5)M{(h-C5H4)SiMe2(h-C5H4)}SnMe2Cl] (M = Ti or Zr).It has been shown that the reaction of the dilithium salt [Li2{Me2C(C5H4)(C9H6)}] with one equivalent of [Zr(h-C5H5)- Cl3]?DME produces a deep red mononuclear complex involving an indenyl ligand with an unprecedented h2 mode of coordination confirmed by crystal structure determination.13,14 This complex reacts with [Hf(h-C5H5)Cl3]?2THF in refluxing toluene to aVord the heterobinuclear complex [Cl2(h-C5H5)Zr- {(h-C5H4)CMe2(C9H6)}Hf(h-C5H5Cl2)] in good yield (80%).15 Nifant’ev et al.12 reported that the compound [Cl2(h-C5- H5)Ti{(h-C5H4)SiMe2(h-C5H4)}SnMe2Cl] with [{Rh(COD)- Cl}]2] gave the early–late heterobimetallic complex [Cl2(h- C5H5)Ti{(h-C5H4)SiMe2(h-C5H4)}Rh(COD)].Werner and co-workers have also reported a number of homo- and hetero-bimetallic complexes of the late transition metals using the ligand anion [(C5H4)CH2(C5H4)]22 to give the complexes [M(CO)2{(h-C5H4)CH2(h-C5H5)}] (M = Rh or Ir) 16,17 and the heterobimetallic complexes [(OC)2M{(h-C5H4)- CH2(h-C5H4)}M*(CO)2] (where M* = Rh or Co) and [(OC)2M{(h-C5H4)CH2(h-C5H4)}Ti(h-C5H5Cl2)].17,18 Preliminary reports of part of the present work have been published.13,15 Results and discussion Treatment of the previously described compound [Zr{(h-C5- H4)CMe2(h2-C9H6)}(h-C5H5)Cl]13–15 with [{Rh(CO)2Cl}2] in diethyl ether gave orange crystalline [Cl2(h-C5H5)Zr{m-(h- C5H4)CMe2(h-C9H6)}Rh(CO)2] 1 in 60% yield.This complex is slightly air- and moisture-sensitive, fairly soluble in diethyl ether and soluble in toluene, THF and dichloromethane. The analytical and spectroscopic data for 1 and for all the other new compounds described in this work have been deposited (SUP) and selected illustrative data are given in Table 1. All these data will not be further discussed except where the interpretation is not straightforward. In many cases the assignments of the NMR data were assisted by COSY, heteronuclear decoupling and related experiments.When such experiments were carried out this is indicated either in the text or in SUP 57500. A fuller description of the data is available elsewhere.191050 J. Chem. Soc., Dalton Trans., 1999, 1049–1059 Table 1 Selected analytical and spectroscopic data Compound and analysis a 3 [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2- (h-C9H6)}Rh(CO)2] Yellow C, 41.5 (41.6) H, 3.0 (3.1) Cl, 10.4 (10.2) Spectroscopic data b 1H:c 7.34 [1 H, m, Hc or f], 7.25 [1 H, m, Hc or f], 7.11 [2 H, m, Hd and e], 6.58 [1 H, pseudo q, C5H4], 6.42 [1 H, pseudo q, C5H4], 6.36 [1 H, pseudo q, C5H4, partially obscured], 6.35 [5 H, s, C5H5], 6.10 [1 H, pseudo q, C5H4], 5.72 [1 H, t, J(HH) = J(RhH) 3, Ha], 5.59 [1 H, d, J(HH) 3, Hb], 2.08 [3 H, s, Me], 1.95 [3 H, s, Me] 13C-{1H}:d 190.3 [d, J(RhC) 83, CO], 137.8 [s, Cipso], 125.0 [s, Cd or e], 124.6 [s, Cd or e], 119.9 [s, Cc or f ], 119.5 [s, Cc or f ], 117.6 [s, Cg, h or i], 116.9 [s, C5H4], 116.7 [s, Cg, h or i ], 115.4 [s, C5H4], 115.0 [s, C5H5], 113.8 [s, C5H4], 113.6 [s, Cg, h or i], 108.9 [s, C5H4], 97.8 [d, J(RhC) 6, Ca], 71.0 [s, Cb], 37.5 [s, CMe2], 30.6 [s, Me], 30.3 [s, Me] Selected IR data; (CsI disc): n(CO) 2037s, 1976s 7 [Cl2(h-C5H5Zr){m-(h-C5H4)CMe2- (h-C9H6)}Rh(PPh3)2] Orange C, 64.3 (64.9) H, 4.6 (4.7) Cl, 6.9 (6.6) 1H:d 7.46 [1 H, partially obscured, Hc or f], 7.43 [12 H, m, PPh3], 7.02 [1 H, t, Hd or e], 6.95 [18 H, m, PPh3], 6.81 [1 H, t, J(HH) 8, Hd or e], 6.21 [1 H, d, J(HH) 8, Hc or f], 6.46 [1 H, pseudo q, C5H4], 6.45 [1 H, pseudo q, C5H4], 6.02 [1 H, pseudo q, C5H4], 5.96 [1 H, t, J(RhH) = J(HH) 2.5, Ha], 5.84 [5 H, s, C5H5], 5.54 [1 H, pseudo q, C5H4], 4.25 [1 H, d, J(HH) 2.5, Hb], 2.21 [3 H, s, Me], 2.18 [3 H, s, Me] 13C-{1H} (125.7 MHz):c 141.2 [s, Cipsio C5H4], 123.1 [s, Cd or e], 122.7 [s, Cd or e], 118.8 [s, Cc or f], 118.3 [s, Cc or f], 117.3 [s, C5H4], 116.8 [s, 2 × C5H4], 116.2 [s, C5H4], 116.0 [s, C5H5], 110.5 [s, C5H4], 98.4 [m, Ca], 76.7 [m, Cb], 31.8 [s, Me], 31.2 [s, Me] 31P-{1H} (200 MHz, C6D6): 47.5 [d, J(RhP) 200] 10 [Cl2(h-C5H5)Zr{m-(h-C5H4)- CMe2(h-C9H6)}Ru(h-C4H7)(CO)] Yellow C, 50.7 (51.3) H, 3.9 (4.5) 1H: (500 MHz): d 6.96 [1 H, d, J(HH) 8, Hc or f], 6.64 [1 H, d, J(HH) 8, Hc or f], 6.43 [1 H, t, J(HH) 8, Hd or e], 6.41 [1 H, t, J(HH) 8, Hd or e], 6.38 [1 H, pseudo q, C5H4], 6.28 [1 H, pseudo q, C5H4], 5.88 [5 H, s, C5H5], 5.81 [1 H, pseudo q, C5H4], 5.57 [1 H, pseudo q, C5H4], 5.49 [1 H, d, J(HH) 3, Ha], 4.92 [1 H, d, J(HH) 3, Hb], 3.47 [1 H, m, Hsyn, (CH2)2CMe], 3.40 [1 H, m, Hsyn, (CH2)2CMe], 2.07 [3 H, s, Me], 1.97 [3 H, s, Me], 1.66 [3 H, s, (CH2)2CMe], 20.29 [1 H, s, Hanti, (CH2)2CMe], 20.31 [1 H, s, Hanti, (CH2)2CMe] 13C-{1H} (125.7 MHz): c 204.3 [br s, CO], 139.7 [s, Cipso], 124.8 [s, Cd or e], 124.6 [s, Cc or f], 124.5 [s, Cd or e], 124.3 [s, Cc or f], 117.4 [s, C5H4], 116.9 [s, C5H4], 115.9 [s, C5H5], 113.8 [s, C5H4], 112.0 [s, Cg, h or i], 110.9 [s, C5H4], 110.5 [s, Cg, h or i], 106.3 [s, Cg, h or i], 86.4 [s, Ca], 64.8 [s, Cb], 49.8 [s, (CH2)2CMe], 49.1 [s, (CH2)2CMe], 37.7 [s, CMe2 or (CH2)2CMe], 37.4 [s, CMe2 or (CH2)2CMe], 31.3 [s, Me], 31.1 [s, Me], 24.6 [s, (CH2)cCMe] Selected IR data (CsI disc): n(CO) 1912s 13 [Cl2(h-C5H5)Zr{m-(h-C5H4)- C(CH2)5(h-C9H6)}Zr(h-C5H5)Cl2] Yellow C, 50.2 (50.4) H, 4.2 (4.2) Cl, 20.0 (19.8) 1H (500 MHz):c 7.90 [1 H, d, J(HH) 8, Hc or f], 7.56 [1 H, d, J(HH) 8, Hc or f], 7.31 [1 H, t, J(HH) 8, Hd or e], 7.19 [1 H, t, J(HH) 8, Hd or e], 7.04 [1 H, d, J(HH) 3, Ha], 6.87 [1 H, pseudo q, J(HH) 3, C5H4], 6.77 [1 H, d, J(HH) 3, Hb], 6.71 [1 H, pseudo q, J(HH) 3, C5H4], 6.50 [1 H, pseudo q, J(HH) 3, C5H4], 6.36 [5 H, s, ZrCp], 6.35 [1 H, pseudo q, J(HH) 3, C5H4], 5.87 [5 H, s, Cp of Zr(C9H6)], 2.94 [1 H, m, cyclohexyl], 2.71 [2 H, m, cyclohexyl], 2.35 [1 H, m, cyclohexyl], 1.78 [1 H, m, cyclohexyl], 1.68 [1 H, m, cyclohexyl], 1.47 [2 H, m, cyclohexyl], 1.40 [1 H, m, cyclohexyl], 0.84 [1 H, m, cyclohexyl] 13C-{1H} (125.7 MHz): d 143.8 [s, Cipso of C5H4], 127.1 [s, Cg, h or i], 126.9 [s, Cc or f], 126.4 [s, Cd or e], 126.0 [s, Cc or f], 125.7 [s, Ca], 124.9 [s, Cd or e], 123.9 [s, Cg, h or i], 117.7 [s, Cp of Zr(C9H6) and C5H4], 117.3 [s, C5H4], 116.7 [s, ZrCp], 116.2 [s, C5H4], 112.0 [s, C5H4], 103.3 [s, Cb], 95.3 [s, Cg, h or i], 42.9 [s, C(CH2)5], 36.6 [s, cyclohexyl], 35.8 [s, cyclohexyl],25.5 [s, cyclohexyl], 23.7 [s, cyclohexyl], 22.7 [s, cyclohexyl] 19 [Cl(h-C5H5)Hf{m-(h-C5H4)- CMe2(h-C9H6)}(m-O)Zr(h-C5H5)Cl] White C, 46.3 (45.9) H, 3.6 (3.7) Cl, 8.7 (10.0) 1H:d 8.08 [1 H, d, J(HH) 9, Hc or f], 7.91 [1 H, d, J(HH) 9, Hc or f], 7.90 [1 H, d, J(HH) 9, Hc or f], 7.70 [1 H, d, J(HH) 9, Hc or f], 7.34 [1 H, t, J(HH) 9, Hd or e], 7.29 [1 H, m, Hd or e], 7.28 [1 H, m, Hd or e], 7.15 [1 H, d, J(HH) 9, Hd or e], 6.77 [1 H, d, J(HH) 2.5, Ha], 6.75 [1 H, pseudo q, C5H4], 6.56 [1 H, d, J(HH) 2.5, Ha], 6.54 [1 H, pseudo q, C5H4], 6.45 [1 H, pseudo q, C5H4], 6.41 [1 H, pseudo q, C5H4], 6.39 [1 H, pseudo q, C5H4], 6.33 [1 H, d, J(HH) 3, Hb], 6.31 [5 H, s, Cp], 6.26 [5 H, s, Cp], 6.22 [1 H, m, C5H4], 6.21 [1 H, m, Hb], 6.18 [5 H, s, Cp], 5.93 [1 H, m, C5H4], 5.92 [1 H, m, C5H4], 5.64 [5 H, s, Cp], 1.81 [3 H, s, Me], 1.80 [3 H, s, Me], 1.62 [3 H, s, Me], 1.58 [3 H, s, Me] 13C-{1H} (125.7 MHz): d 126.3 [s, Cc or f], 126.2 [s, Cc or f], 124.5 [s, 2x Cc or f], 125.2 [s, Cd or e], 125.1 [s, Cd or e], 124.9 [s, 2x Cd or e], 122.7 [s, Ca], 122.2 [s, Ca], 117.6 [s, C5H4], 116.4 [s, Cp], 113.6 [s, Cp], 113.2 [s, Cp], 113.1 [s, Cp], 113.1 [s, C5H4], 110.9 [s, C5H4], 109.7 [s, C5H4], 108.0 [s, C5H4], 106.5 [s, C5H4], 106.4 [s, C5H4], 104.0 [s, C5H4], 98.9 [s, Cb], 95.3 [s, Cb], 32.9 [s, 2x CMe2], 31.8 [s, CMe2], 3.06 [s, CMe2] Selected IR data: (M–O–M) 768s, 748s; bridging ligand 808s a Analytical data given as: found (required) %.b 1H NMR data given for 300 MHz, 13C NMR at 72.5 MHz unless otherwise stated.All at room temperature. Data given as: chemical shift (d) [relative intensity, multiplicity J in Hz, assignment]. Cp indicates h-C5H5. c In CD2Cl2. d In C6H6. The 1H NMR spectrum of compound 1 shows two resonances at d 2.08 and 1.95 assignable to the two inequivalent methyl groups of the bridging ligand. The infrared spectrum shows the expected two strong bands at 2038 and 1985 cm21 in good agreement with those at 2050 and 1989 cm21 observed for [Rh(h-C9H7)(CO)2].20 The reaction of [{Rh(CO)2Cl}2] with [Zr{(4-But-C6H9)(h-C5- H4)(h2-C9H6)}(h-C5H5)Cl] (see below) gives [Cl2(h-C5H5)- Zr{m-(h-C5H4)(4-But-C6H9)(h-C9H6)}Rh(CO)2] 2 as an orange solid in 62% yield.The compound is slightly air- and moisturesensitive. Treatment of [Hf{(h-C5H4)CMe2(h2-C9H6)}(h-C5H5)- Cl] with [{Rh(CO)2Cl}2] in diethyl ether gave a yellow precipitate of [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}Rh- (CO)2] 3 in an overall yield of 72%.Similarly, addition of [{Rh(CO)2Cl}2] in toluene to [Zr{Me2C(h-C5H4)2}(h-C5H5)Cl] in toluene gave dark orange [Cl2(h-C5H5)Zr{m-(h-C5H4)- CMe2(h-C5H4)}Rh(CO)2] 4 in 76% yield. A suspension of [{Rh(CO)2Cl}]2] in diethyl ether was added to an ether solution of [Zr{(h-C5H4)CMe2(h3-C13H8)}(h-C5- H5)Cl] 13 at room temperature. The solution gradually darkened over several days and a brown precipitate appeared.The reaction mixture was filtered and the precipitate dried in vacuo. This material was shown to be the complex [Zr{(h-C5H4)CMe2- (C13H9)}(h-C5H5)Cl2] by comparison of the 1H NMR spectrum with that of an authentic sample.13 The compound [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}- Rh(CO)2] 3 in dichloromethane was treated with a suspension of two equivalents of KNCS, also in dichloromethane at 0 8C. This gave orange microcrystalline [(SCN)2(h-C5H5)(Hf){m- (h-C5H4)CMe2(h-C9H6)}Rh(CO)2] 5 in 67% yield (Scheme 1).The IR spectrum shows three bands near 2000 cm21, at 2040, 2002 and 1974 cm21. The starting compound 3 has two carbonyl bands at 2037 and 1976 cm21, and the values for the n(CN) stretch in dichloromethane solution for the complexJ. Chem. Soc., Dalton Trans., 1999, 1049–1059 1051 [Hf(h-C5H5)2(NCS)2] are 2049 and 2011 cm21.21 It thus appears that the IR spectrum can be understood by the superimposition of these two spectra, assuming there is an overlap of the two highest wavenumber bands.Compound 3 in THF was treated with iodine to give black microcrystalline [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}- Rh(CO)I2] 6 in 51% yield. The presence of the carbonyl ligand in 6 was confirmed by a single band in the IR spectrum, n(CO) 2063 cm21. A toluene solution of [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2- (h-C9H6)}Rh(CO)2] 1 was treated with two equivalents of PPh3 at room temperature to give orange [Cl2(h-C5H5)Zr{m- (h-C5H4)CMe2(h-C9H6)}Rh(PPh3)2] 7 in 54% yield.The 31P NMR spectrum showed a doublet (J = 200 Hz) at d 47.5 due to coupling with the 103Rh nucleus. As expected, the IR spectrum showed no bands assignable to the presence of CO ligands. The compound [Mn(CO)5Cl] in THF was added dropwise to a THF solution of [Zr{Me2C(h-C5H4)(h2-C9H6)}(h-C5H5)Cl] 13 at room temperature to give orange-yellow [Cl2(h-C5H5)Zr- {m-(h-C5H4)CMe2(h-C9H6)}Mn(CO)3] 8 in 66% yield (Scheme 2). The compound 8 is slightly air and moisture sensitive.The infrared spectrum showed strong bands assignable to n(CO) at 2012, 1935 and 1920 cm21. This indicates that the two e symmetry vibrations are not degenerate in this molecule, contrary to the expectation for local C3v molecular symmetry. This implies that there is a degree of asymmetry in the indenyl–metal bonding, in a similar fashion to that found for some rhodium compounds [Rh(h-C9H7)L2]. Three carbonyl stretching frequencies are also observed for the compound [Mn(h-C9H7)- (CO)3] ,22 at 2023, 1950 and 1930 cm21, and they are all between 10 and 15 cm21 higher than those for compound 8, suggesting that in the bimetallic system there is greater metal–ligand back donation into the antibonding CO orbitals.When the hafnium complex [Hf{(h-C5H4)CMe2(h2-C9H6)}- (h-C5H5)Cl] 13 was treated with [Mn(CO)5Cl] in THF the yellow compound [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}Mn- (CO)3] 9 was formed. Addition of [Ru(h-C4H7)(CO)3Cl] in THF to [Zr{(h-C5H4)- CMe2(h2-C9H6)}(h-C5H5)Cl] yielded the yellow binuclear compound [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}Ru(CO)- (h-CH2CMeCH2)] 10 in 39% yield.The NMR assignments were confirmed by a COSY spectrum, and a 13C–1H heteronuclear shift correlation spectrum. These data are all indicative of there being only one isomer formed in this reaction. In the case of both [Ru(h-C5H5)(h-C3H5)(CO)] and its methylallyl analogue, the exo isomer was characterised by a CO stretching Scheme 1 i, KCNS in dichloromethane at room temperature (r.t.) and for 24 h, yield 67%; ii, PPh3 in toluene at r.t.for 24 h, 54%; iii, I2 in THF at r.t. for 3 h, 51%. frequency in hexane solution of around 1955–1960 cm21, whilst the endo-isomer was characterised by one of around 1930–1935 cm21.23 The IR spectrum of 10 as a CsI disc shows a peak at 1912 cm21 indicative of an endo isomer. In order to attempt to confirm this assignment a NOESY spectrum was obtained.This allowed assignment of the methyl resonance of the allyl ligand at d 1.66, and the magnitude of the correlations con- firmed the assignment of the syn- and anti- protons of the allyl ligand. However, no correlations were observed between any of the signals of the indenyl ligand and those of the allyl ligand. Although not definitive evidence for the endo isomer, a close proximity of the methyl group of the allyl ligand to the indenyl ring in the exo isomer would be expected to result in the appearance of a correlation in a NOESY spectrum.The compounds [Zr{(h-C5H4)CMe2(h2-C9H6)}(h-C5H5)Cl] 13 and FeCl2?1.5THF in THF gave brown microcrystalline [Fe- {m-(h-C9H6)CMe2(h-C5H4)Zr(h-C5H5)Cl2}2] 11 in 42% yield. The compound 11 appears to decompose slowly in solution at room temperature. The 1H NMR spectrum showed that only one isomer was present and a COSY spectrum allowed the assignment of all the resonances but not the determination of which isomer was present.Addition of [Zr{(h-C5H4)CMe2(h2- C9H6)}(h-C5H5)Cl] 13 in THF at 278 8C to a solution of one equivalent of [{Co(h-C5Me5)Cl}2] in THF gave black microcrystals of [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}Co(h- C5Me5)] 12 in 58% yield. The ESR spectrum of a solid sample of this compound at room temperature shows an isotropic lineshape, though with a small degree of asymmetry, which, by comparison with the position of the diphenylpicrylhydrazyl standard gives giso = 2.01.Cobaltocene 24 and decamethylcobaltocene 25 gave no ESR spectra at room temperature and this was attributed to rapid relaxation from the degenerate electronic ground state e4a2e1. The observation of a spectrum at room temperature for 12 may reflect the low symmetry of the indenyl ligand with a consequential lifting of the degeneracy of the ground state. The ESR spectrum of 12 in toluene solution at room temperature had a lineshape characteristic of an axially Scheme 2 i, [Mn(CO)5Cl] in THF at 60 8C for 5 h, yield 66%; ii, [Ru(h-C4H7)(CO)3Cl] in THF at 60 8C for 3 h, 39%; iii, FeCl2?1.5THF, reflux for 3 h, 42%; iv, [{Co(h-C5Me5)Cl}2] in THF at 278 8C, warm to r.t. for 4 h, 58%.1052 J.Chem. Soc., Dalton Trans., 1999, 1049–1059 symmetric system;25 g1 = 2.131 and g2 = 1.915. The spectrum was complicated by poorly resolved hyperfine coupling, with a coupling constant, A = 50 G. The dilithium salt Li2[(h-C5H4)C(CH2)5(h-C9H6)]?0.8Et2O14 when added to two equivalents of [Zr(h-C5H5)Cl3]?DME gave the yellow bimetallic complex [Cl2(h-C5H5)Zr{m-(h-C5H4)- C(CH2)5(h-C9H6)}Zr(h-C5H5)Cl2] 13 in 62% yield (Scheme 3).Addition of toluene to a mixture of the dilithium salt Li2[(C5H4)C(CH2)5(C9H6)]?0.8Et2O and slightly less than two equivalents of [Hf(h-C5H5)Cl3]?2THF at 278 8C gave pale yellow [Cl2(h-C5H5)Hf{m-(h-C5H4)C(CH2)5(h-C9H6)}HfCl2(h- C5H5)] 14 in 68% yield. The 1H NMR spectrum is extremely similar to that of the dizirconium complex 13.The low field region of the 13C NMR spectrum is also similar to that of the dizirconium analogue: the resonances assigned to the C5H5 ligands occur at d 116.3 and 115.8. The C5H4 carbons give rise to bands at d 115.4, 115.0, 114.6 and 110.8, with those of the C5 ring of the indenyl ligand being at d 125.6 and 114.8. The bands due to the carbons of the C6 ring of the indenyl ligand lie at 127.1, 126.5, 125.9 and 124.7: these signals were assigned with the aid of a 13C–1H heteronuclear shift correlation spectrum.In an attempted preparation of [Cl2(h-C5H5)Hf{(h-C5H4)- C(CH2)5(h-C9H6)}Hf(h-C5H5Cl2)] toluene was added to a mixture of dilithium salt Li2[(C5H4)C(CH2)5(C9H6)]?0.8Et2O and [Hf(h-C5H5)Cl3]?2THF at room temperature. The resulting mixture was heated to 120 8C for 16 h by which time a pale yellow solution was formed with a pale precipitate. The 1H Scheme 3 i, For complex 13, [Zr(h-C5H5)Cl3]?DME at 278 8C, add toluene at 278 8C, warm to r.t., heat to 105 8C for 24 h, yield 62%; ii, [Hf(h-C5H5)Cl3]?2THF at r.t., then 105 8C for 24 h, 58%; iii, [Zr(h-C5H5)Cl3]?DME at 278 8C, add toluene at 278 8C, warm to r.t.for 14 h, 86%; iv, [Hf(h-C5H5)Cl3]?2THF in toluene at r.t., then heat to 120 8C for 16 h, ca. 20%. NMR spectrum of the yellow product was diVerent from that found for the expected bimetallic complex. It was complicated by the presence of a small amount of [Hf(h-C5H5)Cl3]?2THF. Nonetheless the most obvious feature was the presence of two roofed doublets at d 3.62 and 3.32.These signals were very closely similar to those observed from the product of the reaction between [Zr{(h-C5H4)CMe2(h2-C9H6)}(h-C5H5)Cl] and [{Y(h-C5H5)2}2] namely the compound [Zr{(h-C5H4)- CMe2(h1-C9H6)}(h-C5H5)Cl]. In this reaction a hydrogen has been transferred from the bonding CH group of the h2-indenyl ligand to the second and adjacent CH group. This results in the formation of a CH2 group and an h1-vinylic system.26 The spectrum of complex 15 shows the presence of a band assignable to the h-C5H5 ligand at d 6.26 and four signals assigned to the C5H4 ligand at d 6.43, 6.20, 6.08 and 5.72, the latter being at rather high field for such a proton.The resonances assignable to the C6 ring of the indenyl ligand are a pair of doublets at d 7.68 and 7.39 and a pair of triplets at d 7.23 and 7.14. In the high field region of the spectrum the resonances consisted of two broad multiplets at d 1.42 and 1.67, two doublets at d 1.96 and 2.25 and two triplets at d 2.38 and 2.54. The structure proposed for the compound [Hf{(h-C5H4)C(CH2)5(h1-C9H6)}(h-C5H5)Cl] 15 is illustrated in Scheme 3.The complex [Zr{(h-C5H4)C(CH2)5(h2-C9H6)}(h-C5H5)Cl] 15 was treated with one equivalent of [Hf(h-C5H5)Cl3]?2THF to give the yellow heterobimetallic compound [Cl2(h-C5H5)Zr{m- (h-C5H4)C(CH2)5(h-C9H6)}Hf(h-C5H5)Cl2] 16 in 58% yield. The 1H NMR spectrum is similar to those of the homobimetallic complexes 13 and 14.In particular, the resonance assignable to the C5H5 ligand bonded to the zirconium atom occurs at d 6.36, identical to the low field C5H5 resonance of the dizirconium complex 13; and the resonance assignable to the h-C5H5 ligand bonded to the hafnium atom occurs at d 5.71, virtually identical to that of the high field resonance observed for the dihafnium complex 14. This is strong evidence that compound 16 is indeed the heterobimetallic analogue of 13 and 14, with the zirconium exclusively bonded to the C5H4 ring of the bridging ligand and the hafnium bonded exclusively to the indenyl ligand.The other resonances in the 1H NMR spectrum also correspond closely to those of 13 and 14. The 13C NMR spectrum was also very similar to those of the homobimetallic complexes described earlier; the resonances assigned to the C5H5 ligands are at d 116.8, almost identical to the value of the C5H4 ring bound to the zirconium in 13, and d 116.3, identical to that found for the indenyl ring bound to the hafnium in 14.This further confirms the analogous nature of compounds 13, 14 and 16. It was decided to investigate the eVect of increasing bulk of the bridging system on the polymerisation behaviour of these bimetallic species. The dilithium salt Li2[(h-C5H4)(4-But- C6H10)(h-C9H6)]?0.8Et2O was prepared and added to the compound [Zr(h-C5H5)Cl3]?DME at 278 8C. Yellow crystals of [Cl2(h-C5H5)Zr{m-(h-C5H4)(4-But-C6H9)(h-C9H6)}Zr(h- C5H5)Cl2] 17 was formed in 74% yield.A solution of two equivalents of LiAl(OBut)3H in THF was slowly added to a solution of [Cl2(h-C5H5)Hf{m-(h-C5H4)- CMe2(h-C9H6)}Zr(h-C5H5)Cl2]15 also in THF, at room temperature giving the pale binuclear dihydride [Cl(h-C5H5)- Hf{m-(h-C5H4)CMe2(h-C9H6)}(m-H)2Zr(h-C5H5)Cl] 18. The 1H NMR spectrum of 18 shows that only one of the possible isomers of this compound is present (see Scheme 4).The IR spectrum shows a broad band centred around 1490 cm21 which can be assigned to vibrations of the hydride ligands. A similar band centred at 1390 cm21 is observed for the species [{Zr(h- C5H5)2H(Cl)]x.27 A solution of [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}- Zr(h-C5H5)Cl2] 15 in dichloromethane was treated, at room temperature, with an equimolar quantity of water followed by a further equimolar quantity of aniline. The addition of theJ. Chem. Soc., Dalton Trans., 1999, 1049–1059 1053 aniline gave an immediate white precipitate of [Cl(h-C5H5)- Hf{m-(h-C5H4)CMe2(h-C9H6)}(m-O)Zr(h-C5H5Cl)] 19.The 1H NMR spectrum shows the presence of four peaks assignable to h-C5H5 ligands and this is indicative of the presence of two isomers of the complex. The proposed isomers are shown in Scheme 4. The 13C NMR spectrum is complicated by the presence of two isomers. Four signals assigned to cyclopentadienyl rings are observed, at d 116.4, 113.6, 113.2 and 113.1.Examination of the 13C–1H heteronuclear shift correlation spectrum shows that the low field resonance of these four is the one which correlates to the high field cyclopentadienyl resonance in the 1H spectrum. The IR spectrum shows bands at 768 and 748 cm21 assignable to the vibrations of the bridging oxo-ligand.28 The ansa-bridged zirconocene [Zr{Me2C(h-C5H4)2}Cl2] was treated with a suspension of K(C5H4Me) in THF to give the desired tris-h-cyclopentadienyl compound [Zr{Me2C(h- C5H4)2}(h-C5H4Me)Cl] 20, as a very pale yellow solid.The assignments for 20 were confirmed by a COSY NMR spectrum, and 500 MHz inverse-detection 1H–13C heteronuclear shift correlation spectrum in C6D6 at r.t. Variable temperature 1H and 13C NMR spectra in solution revealed no slowing of any fluxional processes down to 290 8C. The dilithium salt of the ansa-ligand Li2[(4-But-C6H9)(C5H4)(C9H6)] was prepared and one equivalent was mixed with the pure compound [Zr(h- C5H5)Cl3]?DME.Toluene was added at 278 8C to give a deep red reaction mixture from which air- and moisture-sensitive red crystals of [Zr{(4-But-C6H9)(h-C5H4)(h2-C9H6)}(h-C5H5)Cl] 21 were obtained. The 1H NMR spectrum showed that only one diastereomer was present. The 13C–1H heteronuclear shift correlation spectrum allows the assignment of the signals of the bridging cyclohexyl system and the 13C DEPT spectrum shows clearly the presence of the carbon atoms of the four CH2 groups of the cyclohexyl ring and allows easy assignment of the carbon atom to which the But group is bonded; this resonance is at d 48.3.The 13C–1H heteronuclear shift correlation spectrum shows a large correlation between the signals assigned to the But ligand, four carbon resonances with correlations to two proton resonances and a carbon resonance with a single correlation in the proton spectrum. Scheme 4 i, LiAl(OBut)3H in THF at r.t. for 3 h, yield 25%, two possible isomers are shown; ii, in dichloromethane, add degassed water and aniline; stir for 1 h at r.t., 77%, two possible isomers are shown.Polymerisation studies Eleven of the new metallocenes described above have been examined as catalyst precursors for ethylene polymerisation. The polymerisation experiments were carried out under the conditions described by Kaminsky et al.29 namely, using 2 bar monomer pressure at 30 8C in toluene solvent and 6.25 × 1026 mol of catalyst with a [(MeAlO)n] (MAO): metallocene ratio of 830 :1. They were performed at least twice and found to be reproducible.The data show that the relative activities of the rhodium and manganese carbonyl–containing complexes [Table 2(a)] are all lower than those of both the dizirconium complexes and very much lower than the 60900 kg PE mol21 h21 Cmon 21 found by Kaminsky et al.29 for the [Zr(h-C5H5)2Cl2]–MAO system under the same conditions. There was no appreciable deactivation of the catalytic systems over the 1 h period of each experiment.Table 2(c) shows that the early–late binuclear complexes 11 and 12 have activities approaching those of the [Zr(h-C5- H5)2Cl2]–MAO system. Indeed, the polymerisation was so rapid that the experiment had to be abandoned after a short time since the quantity of polymer produced caused the stirring of the reaction to stop. The viscous nature of the solution may also cause a slowing of the diVusion of the monomer to the catalytic centres and a reduction of the magnitude and speed of dissolution of ethene in the toluene, and the activities for these very active catalysts are likely to be underestimates. The data in Table 2(d) also show that replacing the CMe2 bridging system by a cyclohexyl system causes a rise in the polymerisation activity.In the case of the dihafnium and zirconium–hafnium binuclear compounds the increase in the activity is between 40 and 50%, whereas the increase in the dizirconium system is almost 100%.In addition, the dizirconium compound with a 4-But-C6H9 bridging system is more active than the cyclohexyl complex, with a relative activity of 3530 kg PE mol21 h21 Cmon 21. This is more than twice that observed for the mononuclear ansa-bridged complex with the same bridging ligand, namely [Zr{Me2C(h-C5H4)(h-C9H6)}- Cl2], with a relative activity of 1550 kg PE mol21 h21 Cmon 21.29 The higher activity of the zirconium with respect to the hafnium-containing species is interpreted in terms of the increased bond enthalpy of a Hf–C (ª306 kJ mol21) compared with a Zr-C bond (ª284 kJ mol21), which results in the insertion being more diYcult for hafnium catalysts.30 Propene.Polymerisation conditions were identical to those for ethene (210 cm3 toluene solvent, 30 8C, 2 bar monomer pressure). Previous studies on binuclear catalysts of this type for the polymerisation of propene revealed that only a very small quantity of polymer was produced when a similar quantity of catalyst and MAO were used (6.25 × 1026 mol and 0.3 g respectively). 14 To improve the yield of polymer, eight times this amount of catalyst was used. This increase, whilst allowing direct comparison with the previously studied binuclear catalysts, does make direct comparison with the results of Kaminsky’s study somewhat less valid, although cautious comparisons can still be made.29 The yields and activities of the carbonyl-containing complexes for the polymerisation of propene are given in Table 3.The data show that the relative activities exhibited by the new complexes are very low. The relative activity for propene polymerisation of the [Zr(h-C5H5)2Cl2]–MAO system is 140 kg PP mol21 h21 Cmon 21, and values of over 1000 kg PP mol21 h21 Cmon 21 have been obtained for some of the highly stereospecific mononuclear ansa-zirconocene complexes.29 The new complexes listed in Table 3 all show a similar activity with the exception of [Cl2(h-C5H5)Zr{m-(h-C5H4)- CMe2(h-C5H4)}Rh(CO)2] which shows a surprisingly higher activity for the polymerisation of propene, the reason for which is not clear.1054 J.Chem. Soc., Dalton Trans., 1999, 1049–1059 Table 2 Polymerisation of ethylene Compound X Ring Yield/g Activity/ kg PE mol21 h21 Relative activity/ kg PE mol21 h21 Cmon 21 (a) [Cl2(h-C5H5)Zr(h-C5H4)X(Ring)}Rh(CO)2] 124 CMe2 tBuC6H9 CMe2 C9H6 C9H6 h-C5H5 0.85 0.28 0.75 136 45 120 580 191 512 (b) [Cl2(h-C5H5){(h-C5H4)CMe2(C9H6)}X] 8 a 10 a a Mn(CO)3 Ru(C4H7)(CO) ZrCl2(h-C5H5) H 1.37 1.55 2.14 4.94 219 248 342 790 935 1060 1460 3360 (c) [{Cl2(h-C5H5)Zr[(h-C5H4)CMe2(C9H6)]}nX] 11 b 12 b b CoCp2 b Fe Co(h-C5Me5) H n = 2 n = 1 n = 1 1.76 1.52 4.94 0.03 3137 4168 790 5 13350 17730 3360 20 (d) [Cl2(h-C5H5)Zr{(h-C5H4)X(C9H6)}Zr(h-C5H5)Cl2] a 13 a 17 a a 14 a a 16 a Zr Zr Zr Zr Zr Zr Hf Hf Hf Hf Zr Hf Zr Hf CMe2 (CH2)5 tBuC6H9 CMe2 (CH2)5 CMe2 (CH2)5 2.14 4,20 5.18 0.59 0.96 1.69 2.27 342 671 828 94 153 270 363 1460 2870 3530 400 655 1150 1550 a At 2 bar absolute monomer pressure, 30 8C, 210 cm3 toluene solution, 1 h, 6.25 × 1026 mol of compound, 0.3 g MAO, Cmon = 0.235 mol dm23.b Ar 2 bar absolute pressure, 30 8C, 210 cm3 toluene solvent, 1 h, except 5 min for compound 17 and 31– 4 min for compound 18, 6.25 × 1026 mol of compound, 0.3 g MAO, Cmon = 0.235 mol dm23. Table 3 Polymerisation of propylene Compound X Ring Yield/g Activity/ kg PP mol21 h21 Relative activity/ kg PP mol21 h21 Cmon 21 (a) [Cl2(h-C5H5)Zr{(h-C5H4)X(Ring)}Rh(CO)2] 1a 2 a 4 a CMe2 1BuC6H9 CMe2 C9H6 C9H6 Cp 0.22 0.42 3.9 1.1 2.1 19.5 0.9 1.7 15.5 (b) [(h-C5H5)ZrCl2{(h-C5H4)CMe2(C9H6)}X] 8 a a a Mn(CO)3 Zr(h-C5H5)Cl2 H 0.26 2.22 0.36* 1.3 11.1 7.2 1 8.8 5.7 (c) [(h-C5H5)ZrCl2{(h-C5H4)CMe2(C9H6)}n] 11 a 12 a a Fe Co(h-C5Me5) H n = 2 n = 1 n = 1 43.3 38.4 0.36* 216 192 7.2 171 152 5.7 (d) [Cl2(h-C5H5)M{(h-C5H4)X(C9H6)}M*(h-C5H5)Cl2] b 13 b 17 b b 14 b b 16 b M M* Zr Zr Zr Zr Zr Zr Hf Hf Hf Hf ZrHf Zr Hf X CMe2 (CH2)5 tBuC6H9 CMe2 (CH2)5 CMe2 (CH2)5 2.22 2.74 9.25 1.33 2.03 0.42 2.28 11.1 13.7 46.3 6.7 10.2 2.1 11.4 8.8 10.9 36.7 5.3 8.1 1.7 9.0 a At 2 bar monomer pressure, 30 8C, 210 cm3 toluene solvent, 4 h (* 1 h), 5 × 105 mol of compound, 2.4 g MAO, Cmon = 1.26 mol dm23.b At 2 bar monomer pressure, 30 8C, "10 cm3 toluene solvent, 4 h, 5 × 1025 mol of compound, 2.4 g MAO, Cmon = 1.26 mol dm23. The polymerisation behaviour of the early–late transition metal metallocenes towards propene is shown in Table 3(c).These catalysts are far more active than the other multimetallic complexes, as was the case for ethene polymerisation. The ratio of the polymerisation activity of complexes 11 and 12 with respect to the mononuclear zirconocene [Zr{m-(h-C5H4)- CMe2(h-C9H7)}(h-C5H5)Cl2] is greater for propene polymerisation than for ethene. The polymerisation behaviour of the binuclear Group IV metallocene derivatives is given in Table 3(d).The pattern of activity is essentially similar to that for the polymerisation of ethene. For example, the cyclohexyl bridged compounds haveJ. Chem. Soc., Dalton Trans., 1999, 1049–1059 1055 higher activities than the corresponding Me2C bridged complexes and the compound with a 4-ButC6H9 bridge is more active still. Several workers have noted interesting relationships between the activity of propene polymerisation and the stereoregularity and molecular weight of the polymer produced.29,31–34 In view of this, the nature of the polypropene samples produced above was investigated.The polypropenes produced by the new compounds described above were viscous liquids or gummy solids soluble in toluene. The 13C NMR spectra were recorded in 1,2,4- trichlorobenzene–d6-benzene (80 : 20 v/v) at 130 8C. The spectra also show the chain-end structures of the polymer indicative of low molecular weights.35–37 The methyl region of the spectrum between d 23 and 19 shows a number of features typical for atactic polypropenes.37,38 The protons of the vinylidene chainend group can also be seen in the 1H NMR spectrum of the polymer.The 13C NMR spectrum shows that the signals of the two chain-end groups occur in approximately equal intensity. The main chain transfer process is likely to involve b-hydride elimination from a monomer which has inserted in a regioregular manner, with the new polymer chain being initiated from a metal–hydrogen bond, from which a primary insertion will yield the n-propyl group.The 13C NMR spectra of the polypropene samples formed from the zirconium compounds in Table 3 were all very similar, the main diVerences being in the intensity of the chain-end signals with respect to those of the main chain. Polymers from the hafnium-containing binuclear complexes were of higher molecular weight. Some polymer data are shown in the Table 4(a) and 4(b), and further discussion may be found in ref. 19. Molecular weight determinations and the values of Mw, Mn and the ratio Mw/Mn are shown in Table 5. In most cases the number-average molecular weight of the polymer, Mn, was slightly higher than the corresponding value obtained from NMR spectroscopy. It can be seen that use of the early-late heterobimetallic metallocene complexes 11 and 12 gave the smallest value of the ratio Mw/Mn indicating that these provide the most uniform polymers. In contrast, the values of this ratio from the hafniumcontaining Group IV bimetallic complexes are the largest. The molecular weight distribution curves show two components.From the dihafnium complex 14 there is a small region to Table 4 Determination of Mn (a) For early–late bimetallic compounds from NMR spectroscopy Compound (metals) 1 (Zr, Rh) 2 (Zr, Rh) 4 (Zr, Rh) 8 (Zr, Mn) 11 (Zr, Fe) 12 (Zr, Co) Chain length 22.0 42.7 37.3 28.9 39.6 29.9 Molecular weight, Mn 920 1797 1571 1216 1666 1258 (b) For the polymers produced by binuclear Group IV catalysts, together with dyad functions Compound 13 17 14 16 Bridge Cyclohexyl tBuC6H9 Cyclohexyl Cyclohexyl [m] 0.50 0.51 0.67 0.55 [r] 0.50 0.49 0.33 0.45 Chain length 45.2 39.3 136.1 57.5 Molecular weight, Mn 1900 1650 5730 2420 the low molecular weight side of the main peak, whereas the zirconium–hafnium complex 16 resulted in a small region to the high side of the main peak of the envelope.This is consistent with the notion that hafnium centres yield higher molecular weight polymers than zirconium ones,29,33–36 and that the polymerisation activity of the metal centre at the cyclopentadienyl site is larger than that of the metal at the indenyl site.In addition, the zirconium–rhodium complex 2, with a bulky 4-But- C6H9 bridging ligand, resulted in a molecular weight envelope with a considerably flatter peak than the others. In conclusion the polymerisation experiments show that the early–late heterometallic bis(metallocene) complexes [Cl2Zr(h- C5H5){(h-C5H4)CMe2(h-C9H6)}Co(h-C5Me5)] 12 and Cl2Zr- [Fe{m-(h-C9H6)CMe2(h-C8H4)Zr(h-C5H5)Cl2}2] 11 are much more active catalysts than previously described compounds of this type, giving activities approaching those obtained for [Zr(h-C5H5)2Cl2]–MAO. However, early–late heterobimetallic complexes involving carbonyl ligands have been shown to be poor catalysts.This investigation showed that the bulkier bridging systems resulted in greater polymerisation activity, with higher molecular weight polymer being produced, although no enhancement of the stereoregularity of the polypropene was observed.Experimental All manipulations, with the exception of the preparation of purely organic chemicals, were carried out in an inert atmosphere using either a dual vacuum/nitrogen line and standard Schlenk techniques, or in an inert atmosphere dry-box under dinitrogen. The nitrogen was purified by passage over 4 Å molecular sieves and either MnO, for the Schlenk line, or BASF catalyst, for the dry-box.Solvents and solutions were transferred, using a positive pressure of nitrogen, through stainless steel cannulae. Filtrations were performed using stainless steel cannulae fitted with glass fibre filter discs. All glassware and cannulae were dried overnight in an oven at over 150 8C before use. Solvents were pre-dried by standing over 4 Å molecular sieves and then refluxed and distilled, under an inert atmosphere, from the appropriate drying agent: sodium–potassium alloy (1 : 3 w/w) [n-pentane, light petroleum (bp 40–60 8C) and diethyl ether]; sodium [light petroleum (bp 100–120 8C), 1,2- dimethoxyethane, toluene]; potassium (THF, benzene); or calcium hydride (dichloromethane).All solvents were degassed by bubbling nitrogen or by repeated evacuation followed by admission of nitrogen. Solvents for polymerisation experiments were additionally stored over a potassium mirror.Deuteriated solvents for NMR studies were stored in Young’s ampoules over 4 Å molecular sieves under a nitrogen atmosphere. The were transferred using a teat pipette in a dry-box and the tubes either Table 5 Determination of Mw and Mn from GPC analysis Compound (metals) 2 (Zr, Rh) 4 (Zr, Rh) 11 (Zr, Fe) 12 (Zr, Co) 13 (Zr, Zr) 17 (Zr, Zr) 14 (Hf, Hf) 16 (Zr, Hf) Mw 5178 3355 2611 2170 4800 3719 35601 7690 Mn 2052 1707 1554 1340 2083 1697 10329 2082 Mw/Mn 2.5 2.0 1.7 1.6 2.3 2.2 3.5 3.71056 J.Chem. Soc., Dalton Trans., 1999, 1049–1059 sealed under vacuum or capped and the caps wrapped with Labfilm and Teflon tape. The NMR spectra were recorded on either a Bruker AM300 (1H, 300 MHz; 13C, 75.43 MHz) or a Varian UnityPlus spectrometer (1H, 500 MHz; 31P, 200 MHz; 13C, 125.7 MHz). Spectra were referenced internally using the residual protio solvent (1H) and solvent (13C) signals relative to tetramethylsilane (d 0), or externally using trimethyl phosphate in D2O (31P).The IR spectra were recorded on either a Perkin-Elmer 1710 or a Mattson Polaris FTIR spectrometer. ESR spectra on an X-band Varian E109 spectrometer with an operating field of 3300 G referenced externally to 1,1-diphenyl-2-picrylhydrazyl and with samples prepared in high-purity Spectrosil quartz tubes sealed with a Young’s tap. Gel-permeation chromatography (GPC) experiments were performed by Dr Lilge of BASF AG. Elemental analyses were performed by the Microanalytical Department of the Inorganic Chemistry Laboratory.The complexes [M(h-C5H5)Cl3]?DME (where M = Zr or Hf) and [Zr(h-C5Me5)Cl3]?2THF were prepared according to the literature methods and were recrystallised from tetrahydrofuran. 39,40 The co-catalyst MAO was prepared by following the literature preparations of Kaminsky et al.39 and Giannetti et al.,40 or purchased as a 30% w/v solution in toluene from Witco PLC. The compound [Hf{Me2C(h-C5H4)2}(h-C5H5)Cl] was synthesized as described.41 Preparations Li2[{(4-ButC6H9)(C5H4)(C9H6)}].A solution of 6,6-(4-tertbutylcyclohexylene) fulvene (1.32 g, 6.5 mmol) in diethyl ether (50 cm3) was added to Li(C9H7) (0.797 g, 6.5 mmol) in diethyl ether (100 cm3) at 0 8C, allowed to warm to room temperature and stirred for 24 h. The solution was cooled to 0 8C and n-BuLi (7.65 cm3 of a 1.7 M solution in hexane) was added in small portions. The solution was again allowed to warm to room temperature and a white precipitate formed.This suspension was stirred for 14 h at room temperature. The precipitate was then allowed to settle and isolated via filtration. The resulting white solid was washed with diethyl ether (20 cm3) and dried in vacuo. Yield 2.4 g (95%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)CMe2(Á-C9H6)}Rh(CO)2] 2. A solution of [Zr{Me2C(h-C5H4)(h2-C9H6)}(h-C5H5)Cl] (0.285 g, 0.7 mmol) 13 in diethyl ether (50 cm3) at room temperature was treated dropwise with a solution of [{Rh(CO)2Cl}2] (0.135 g, 0.35 mmol) 7 in diethyl ether (30 cm3).The solution changed to orange immediately. Stirring was continued for 2 h. The mixture was filtered to remove a small quantity of dark material and concentrated to 20 cm3. Cooling this solution to 220 8C aVorded an orange-yellow solid. Yield 0.25 g (60%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)(4-ButC6H9)(Á-C9H6)}Rh(CO)2] 2. A solution of [Zr{(4-ButC6H9)(h-C5H4)(1-h2-C9H6)}(h- C5H5)Cl] (0.250 g, 0.5 mmol) in diethyl ether (50 cm3), at room temperature, was treated dropwise with a solution of [{Rh(CO)2Cl}]2 (0.1 g, 0.25 mmol) in diethyl ether (30 cm3).The solution changed to orange immediately. Stirring was continued for 2 h to ensure complete reaction. The solution was filtered to remove a small quantity of dark material and the filtrate concentrated to 25 cm3. Cooling this to 220 8C aVorded an orange-yellow solid. The supernatant was decanted from this solid, and the solid dried in vacuo.Yield, 0.22 g (62%). [Cl2(Á-C5H5)Hf{Ï-(Á-C5H4)CMe2(Á-C9H6)}Rh(CO)2] 3. A solution of [Hf{Me2C(h-C5H4)(h2-C9H6)}(h-C5H5)Cl] (0.500 g, 1 mmol) in diethyl ether (50 cm3) 6 at room temperature was treated with a solution of [{Rh(CO)2Cl}2] (0.200 g, 0.5 mmol) in diethyl ether (20 cm3) in a dropwise manner. The solution lightened immediately. Stirring was continued for 10 min, during which time a yellow precipitate was obtained. The solid was isolated via filtration and dried in vacuo as the pure product.Yield 0.45 g. A further small quantity of product could be obtained by concentration of the filtrate under reduced pressure to 20 cm3 and cooling this solution to 220 8C. Combined yield 0.50 g (72%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)CMe2(Á-C5H4)}Rh(CO)2] 4. A suspension of [Zr{Me2C(h-C5H4)2}(h-C5H5)Cl] (0.205 g, 0.57 mmol) in diethyl ether (80 cm3) at room temperature was treated with a solution of [{Rh(CO)2Cl}2] (0.110 g, 0.28 mmol) in diethyl ether (30 cm3), in several small portions. The suspension was stirred at room temperature.As the reaction progressed the starting material gradually dissolved. After 3 h the solution was yellow, with only a very small amount of dark precipitate. The solution was filtered and concentrated to 40 cm3. Cooling to 280 8C gave an orange solid. Yield 0.22 g (70%). The reaction in toluene proceeded within 1 h and the yield was 76%. [(SCN)2(Á-C5H5)(Hf){Ï-(Á-C5H4)CMe2(Á-C9H6)}Rh(CO)2] 5.A solution of [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}- Rh(CO)2] (0.15 g, 0.22 mmol) in dichloromethane (30 cm3) was cooled to 0 8C. To this was added, dropwise, a solution of KCNS (0.042 g, 0.44 mmol), also in dichloromethane (40 cm3). The solution was allowed to warm to room temperature and stirred for 24 h, during which it changed from yellow to orange. The solvent was removed under reduced pressure and the residue extracted with toluene (30 cm3), leaving a small quantity of pale residue. After filtration the filtrate was concentrated to 20 cm3 under reduced pressure and cooled to 220 8C, aVording a dark orange powder.The supernatant was decanted, and the resulting solid washed with light petroleum (bp 40–60 8C) (5 cm3) and dried in vacuo. Yield 0.11 g (67%). [Cl2(Á-C5H5)Hf{Ï-(Á-C5H4)CMe2(Á-C9H6)}Rh(CO)I2] 6. A solution of [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}Rh- (CO)2] 3 (0.200 g, 0.29 mmol) in THF (20 cm3) was added to a solution of I2 (0.073 g, 0.29 mmol), also in THF (20 cm3), at room temperature.The solution was stirred at room temperature for 3 h, changing from purple to black. The solvent was removed under reduced pressure and the black residue washed with toluene (30 cm3), yielding a purplish solution and a black solid. The supernatant was decanted and the resulting solid dried in vacuo. Yield 0.13 g (51%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)CMe2(Á-C9H6)}Rh(PPh3)2] 7. A solution of [Cl2(h-C5H5)Zr{m-(h-C5H4)CMe2(h-C9H6)}Rh- (CO)2] 4 (0.200 g, 0.33 mmol) in toluene (20 cm3) was treated with PPh3 (0.17 g, 0.66 mmol) in toluene (20 cm3) at room temperature in a dropwise manner.The solution lightened slightly and was stirred for 24 h. It was concentrated under reduced pressure to 10 cm3 and cooled to 220 8C. After several weeks an orange solid separated. The solution was decanted and the solid dried in vacuo. Yield 0.19 g (54%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)CMe2(Á-C9H6)}Mn(CO)3] 8. A mixture of [Zr{(h-C5H4)CMe2(h2-C9H6)}(h-C5H5)Cl] 13 (0.500 g, 1.2 mmol) and [Mn(CO)5Cl] (0.280 g, 1.2 mmol) 8 was treated with THF (50 cm3).The reaction vessel was evacuated and heated to 60 8C for 5 h. During this time the reaction mixture became orange. The solvent was removed under reduced pressure, yielding an orange oily solid. This was washed with n-pentane (70 cm3) and extracted with toluene (30 cm3). After filtration the filtrate was concentrated to 15 cm3 and light petroleum (bp 40–60 8C) added (15 cm3).The solution was cooled to 220 8C giving a yellow solid. Yield 0.47 g (66%). [Cl2(Á-C5H5)Hf{Ï-(Á-C5H4)CMe2(Á-C9H6)}Mn(CO)3] 9. A mixture of [Hf{(h-C5H4)CMe2(h2-C9H6)}(h-C5H5)Cl] 13 (0.325J. Chem. Soc., Dalton Trans., 1999, 1049–1059 1057 g, 0.65 mmol) and [Mn(CO)5Cl] (0.150 g, 0.65 mmol) was treated with THF (50 cm3). The reaction vessel was evacuated and heated to 60 8C for 5 h. During this time the reaction mixture became orange. The solvent was removed under reduced pressure yielding an orange oily solid.This was washed with n-pentane (50 cm3) and extracted with toluene (50 cm3). The filtrate was concentrated to 10 cm3 and light petroleum (bp 40– 60 8C) added (15 cm3). The solution was cooled to 220 8C, aVording the product as a yellow solid. Yield 0.19 g (43%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)CMe2(Á-C9H6)}Ru(Á-C4H7)(CO)] 10. A mixture of [Zr{Me2C(h-C5H4)(h2-C9H6)}(h-C5H5)Cl] (0.450 g, 1.1 mmol) and [Ru(h-C4H7)(CO)3Cl] (0.300 g, 1.1 mmol)9 was treated with THF (70 cm3).The reaction vessel was evacuated and heated at 60 8C for 3 h.The reaction mixture changed to bright orange. The solvent was removed under reduced pressure, yielding an orange oily solid. This was washed with n-pentane (40 cm3) and extracted with toluene (30 cm3). The solution was cooled to 220 8C, yielding a yellow powder. This was shown to be a mixture of complexes. Further recrystallisation from diethyl ether (40 cm3) aVorded the pure product as a yellow solid.Yield 0.27 g (39%). [Fe{Ï-(Á-C9H6)CMe2(Á-C5H4)Zr(Á-C5H5)Cl2}2] 11. A mixture of [Zr{(h-C5H4)CMe2(h2-C9H6)}(h-C5H5)Cl] 13 (0.412 g, 1 mmol) and FeCl2?1.5THF (0.117 g, 0.5 mmol) were treated with THF (50 cm3) and the ampoule evacuated. The solution was then heated to reflux for 3 h. During this time the reaction mixture changed to red-brown. The solvent was removed under reduced pressure to yield a brown oily solid. The solid was washed with n-pentane (30 cm3) and extracted with toluene (50 cm3).The filtrate was concentrated to 35 cm3 and the solution cooled to 220 8C, aVording the product as a brown solid. Yield 0.20 g (42%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)CMe2(Á-C9H6)}Co(Á-C5Me5)] 12. A solution of [Zr{Me2C(h-C5H4)(h2-C9H6)}(h-C5H5)Cl] 13 (0.412 g, 1 mmol) in THF (40 cm3) was cooled to 278 8C and treated with [{Co(h-C5Me5)Cl}2] (0.230 g, 0.5 mmol) in THF (30 cm3), in a dropwise manner over 30 min. The reaction mixture was allowed to warm to room temperature and stirred for 4 h.Although it remained black when viewed with reflected light, the solution changed from deep purple to green when viewed with transmitted light. The solvent was removed under reduced pressure, yielding a green-black residue. The residue was washed with n-pentane (50 cm3) to give a dark solution and black residue. The solution was decanted and the residue extracted with toluene (30 cm3). After filtration the filtrate was concentrated to 15 cm3 and cooled to 220 8C giving a black solid.The supernatant was decanted and the solid washed with cold n-pentane (5 cm3) and dried in vacuo. Yield 0.37 g (58%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H4)C(CH2)5(Á-C9H6)}ZrCl2(Á-C5H5)] 13. A mixture of Li2[{(CH2)5C(C5H4)(C9H6)}]?0.8 Et2O (0.5 g, 1.5 mmol) and [Zr(h-C5H5)Cl3]?DME (1 g, 2.9 mmol) was cooled to 278 8C. Toluene (150 cm3) at 70 8C was added to give an orange suspension. The reaction mixture was allowed to warm to room temperature, during which time it gradually darkened to deep red.When the solution had reached room temperature the reaction vessel was partially evacuated and heated to 105 8C for 24 h. After this time the reaction mixture was light red. It was allowed to cool and filtered leaving a pale residue. The light red filtrate was concentrated under reduced pressure to 60 cm3 and cooled to 220 8C yielding a bright yellow crystalline solid. An analytically pure sample was obtained by recrystallisation from dichloromethane (40 cm3).The supernatant was decanted and the solid dried in vacuo. Further concentration of the filtrate followed by cooling aVorded more product, but this was contaminated with large red crystals of the mononuclear complex [Zr{(CH2)5C(h-C5H4)(h2-C9H6)}- (h-C5H5)Cl]. Yield 0.63 g (62%). [Cl2(Á-C5H5)Hf{Ï-(Á-C5H4)C(CH2)5(Á-C9H6)}Hf(Á-C5H5)- Cl2] 14. A mixture of Li2{(CH2)5C(C5H4)(C9H6)}?0.8 Et2O (0.5 g, 1.5 mmol) and [Hf(h-C5H5)Cl3]?2THF (1.45 g, 2.9 mmol) was cooled to 278 8C.Toluene (150 cm3) at 70 8C was added to give an orange suspension. The reaction mixture was allowed to warm to room temperature and became dark orange. When the solution had reached room temperature the reaction vessel was partially evacuated and heated to 105 8C for 24 h. The reaction mixture became yellow. It was cooled and filtered from a pale residue. The yellow filtrate was concentrated to 50 cm3 under reduced pressure and cooled to 220 8C, yielding a pale yellow crystalline solid. An analytically pure sample, oV-white, was obtained by recrystallisation from dichloromethane (40 cm3).Yield 0.86 g (68%). [Hf{(Á-C5H4)C(CH2)5(Á1-C9H6)}(Á-C5H5)Cl] 15. The reaction was undertaken in a similar manner to that described for the synthesis of complex 14 using identical quantities of chemicals. In this case, however, the solvent was added to the reagents at room temperature and the reaction mixture heated to 120 8C for 16 h.A yellow solution was obtained, from which an oV-white solid was obtained after concentrating the toluene solution to 50 cm3 and cooling to 220 8C. Yield ca. 20%. [Cl2(Á-C5H5)Zr{Ï-(Á-C5H5)C(CH2)5(Á-C9H6)}Hf(Á-C5H5)- Cl2] 16. (i) Preparation of [Zr {(h-C5H4)C(CH2)5(h2-C9H6) }- (h-C5H5)Cl] This complex was synthesized in a modification of the literature preparation.13 A mixture of Li2{(CH2)5C- (C5H4)(C9H6)}?0.8 Et2O (0.5 g, 1.5 mmol) and [Zr(h-C5H5)- Cl3]?DME (0.53 g, 1.5 mmol) at 278 8C was treated with toluene (150 cm3) 278 8C to give an orange suspension.The reaction mixture was allowed to warm to room temperature and became deep red. It was stirred for 14 h then filtered, leaving a pale residue. The filtrate was concentrated under reduced pressure to 20 cm3 and cooled to 220 8C. Red crystals of the toluene solvate were obtained, from which the supernatant was decanted and the solid dried in vacuo. Yield 0.64 g (86%).(ii) Reaction of [Zr {(h-C5H4)C(CH2)5(h2-C9H6) }(h-C5H5)- Cl] with [Hf(h-C5H5)Cl3?2THF]. A mixture of [Zr{C(CH2)5- (h-C5H4)(h2-C9H6)}(h-C5H5)Cl]?0.5 C6H5CH3 (0.23 g, 0.46 mmol) and [Hf(h-C5H5)Cl3]?2THF (0.228 g, 0.46 mmol) was treated with toluene (100 cm3) at room temperature to give a deep red solution. The mixture was heated to 105 8C and stirred for 24 h and became light red. The cooled mixture was filtered from a very small quantity of pale solid and the filtrate concentrated to 40 cm3. Cooling to 220 8C aVorded a yellow solid, which was isolated via filtration, washed with diethyl ether (10 cm3) and dried in vacuo.Yield 0.215 g (58%). [Cl2(Á-C5H5)Zr{Ï-(Á-C5H)(4-ButC6H9)(Á-C9H6)}Zr(Á-C5H5)- Cl2] 17. A mixture of L2[{(4-ButC6H9)4(C5H4)(C9H6)}]?0.8 Et2O (0.5 g, 1.28 mmol) and [Zr(h-C5H5)Cl3]?DME (0.88 g, 2.5 mmol) at 278 8C was treated with toluene (100 cm3) at 78 8C to give an orange suspension. The mixture was allowed to warm to room temperature and became deep red.It was heated to 105 8C for 24 h becoming light red and was cooled and filtered, leaving a pale residue. The light red filtrate was concentrated under reduced pressure to 20 cm3 and left to stand at room temperature, yielding a bright yellow crystalline solid. The supernatant was decanted and the solid dried in vacuo. Yield 1.43 g (74%). [Cl(Á-C5H5)Hf{Ï-(Á-C5H4)CMe2(Á-C9H6)}(Ï-H)2Zr(Á-C5H5)- Cl] 18. A stirred solution of LiAl(OBut)3H (0.167 g, 0.66 mmol) in THF (30 cm3) was added dropwise over 1 h to a solution of [Cl2(h-C5H5)Hf{m-(h-C5H4)CMe2(h-C9H6)}Zr(h-C5H5)Cl2] (0.250 g, 0.33 mmol), in THF (50 cm3) at room temperature.1058 J.Chem. Soc., Dalton Trans., 1999, 1049–1059 The mixture was stirred at room temperature for 3 h and became dark yellow. The solvent was removed under reduced pressure and 30 cm3 light petroleum (bp 40–60 8C) were added. This gave a yellow solution and a pale residue. The pale residue was extracted with toluene (30 cm3) to yield a very pale yellow solution and a very small amount of pale residue.After filtration the toluene solution was concentrated under reduced pressure to 20 cm3 and cooling to 220 8C yielded a slightly grey solid. The supernatant was decanted from the precipitate and the solid dried in vacuo. Yield 0.057 g (25%). [Cl(Á-C5H5)Hf{Ï-(Á-C5H4)CMe2(Á-C9H6)}(Ï-O)Zr(Á-C5H5)- Cl] 19. A stirred solution of [Cl2(h-C5H5)Hf{m-(h-C5H4)- CMe2(h-C9H6)}Zr(h-C5H5)Cl2] (0.25 g, 0.33 mmol) in dichloromethane (30 cm3) was treated with degassed water (6 ml, 0.33 mmol) via a microlitre syringe.Aniline (60 ml, 0.66 mmol) was added giving an immediate white precipitate. The solution was stirred for 1 h at room temperature. It was then filtered and concentrated under reduced pressure to 15 cm3. Pentane (20 cm3) was added giving a white precipitate which was isolated via filtration and dried in vacuo. Yield 0.18 g (77%). [Zr{Me2C(Á-C5H4)2}(Á-C5H4Me)Cl}] 20.A solution of [Zr{Me2C(h-C5H4)2}Cl2] (0.50 g, 1.5 mmol) in THF (75 cm3) was cooled to 278 8C and a solution of K(C5H4Me) (0.18 g, 1.5 mmol) in THF (50 cm3) added dropwise. The mixture was allowed to warm to room temperature and stirred for 12 h. The solvent was removed under reduced pressure and the residue extracted with toluene (2 × 30 cm3). The filtrate was concentrated under reduced pressure to 30 cm3 and cooled to 220 8C. A pale yellow solid separated and was isolated via filtration and dried in vacuo.Yield 0.35 g (62%). [Zr{(4-ButC6H9)(Á-C5H4)(Á2-C9H6)}(Á-C5H5)Cl] 21. Toluene (100 cm3) was added to a stirred mixture of Li2[{(4-ButC6- H9)(C5H4)(C9H6)}]?0.8 Et2O (0.50 g, 1.28 mmol) and [Zr(h- C5H5)Cl3]?DME (0.45 g, 1.28 mmol) at 278 8C. The solution was allowed to warm to room temperature, during which time it darkened, through orange to deep red. The reaction mixture was then stirred at room temperature for 16 h and filtered. Concentration of the solution under reduced pressure, to 15 cm3, and cooling to 220 8C gave red crystals.Yield 0.57 g (56%). Polymerisation studies All manipulations of catalysts and co-catalysts were carried out under an inert atmosphere, using Schlenk techniques or a drybox. Pure grade ethene and propene were further purified by passage through a column of 4 Å molecular sieves and then over finely divided potassium, which was supported on glass wool. Polymerisation reactions were carried out in a Fischer– Porter reactor, stirred using a magnetic stirrer and maintained at the required temperature by the use of a thermostatically heated bath.Of ethene. The conditions used were those as close as possible to the ones employed by Kaminsky et al.29 A Fischer– Porter reactor was loaded with MAO (0.250 g) and then connected to the computer-controlled gas supply system via a flexible steel hose. The hose was then repeatedly evacuated and filled with ethene.Toluene (200 cm3) was then added to the reactor under nitrogen, the reactor was evacuated and filled with ethene three times. The ethene pressure was then increased to 2 bar and the solution stirred at 30 8C. The gas supply was switched on so as to allow the solution to become saturated with ethene at the required temperature and pressure. Standard solutions of the catalysts were prepared by dissolving 1.125 × 1024 mol of catalyst (18 times the quantity used in each ethene polymerisation experiment) in 90 cm3 of toluene.A 5 cm3 portion of this solution was diluted with 5 cm3 of toluene, added to a Schlenk tube containing MAO (0.05 g), and this mixture was stirred for 15 min to pre-activate the catalyst. After the solution in the Fischer–Porter reactor had become saturated with ethene, illustrated by the computer showing no further pressure loss, stirring was stopped so as to minimise the loss of ethene when the pressure was released to allow the catalyst/MAO to be added quickly to the contents of the reactor via a cannula.Care was taken to avoid the introduction of excessive amounts of nitrogen whilst adding the catalyst by inserting the cannula into the catalyst solution immediately on releasing the pressure, and withdrawing the cannula from the reactor immediately after the addition was complete. The ethene pressure in the reactor was then increased to 2 bar again, stirring was recommenced and the experiment started.After exactly 1 h, the polymerisation was quenched by venting the ethene, followed by the addition of a small amount of ethanol. The reactor contents were then transferred to a conical flask containing a solution of concentrated HCl in ethanol (300 cm3 of a 30% v/v solution) and stirred overnight. The polymer was then collected on a sintered glass funnel using a Buchner flask and a water aspirator. The polymer was washed with deionised water, ethanol and finally diethyl ether.At least three 50 cm3 portions of each solvent were used. The polymer was then dried in vacuo at 50 8C to constant weight. Of propene. A similar procedure was used for the polymerisation of propene. Eight times the quantity of catalyst and co-catalyst were used. The pre-activation of the catalyst was performed using 80 cm3 of toluene and 130 cm3 of toluene were initially introduced into the Fischer–Porter reactor. The polymerisation was allowed to proceed for 4 h at 30 8C under a monomer pressure of 2 bar.After 4 h, the polymerisation was quenched by venting the propene and adding a small quantity of ethanol. The contents were transferred to a conical flask containing a solution of concentrated HCl in ethanol (300 cm3 of a 30% v/v solution) and stirred overnight. All the polypropenes produced by the catalysts were soluble in toluene, so the toluene layer was separated, washed well with water (5 × 100 cm3) and stirred over MgSO4.The solution was then filtered and the toluene removed leaving the polypropene as, typically, a viscous liquid. The polymer was dried in vacuo at 50 8C to constant weight. Acknowledgements We thank the EPSRC for financial support (to N. H. P.). References 1 M. Arndt, W. Kaminsky, A. M. Schauwienold and U. Weingarten, J. Macromol. Sci., Rev. Macromol. Chem. Phys., 1998, 199, 1135; W. Kaminsky, J. Chem. Soc., Dalton Trans., 1998, 1413, Pure Appl. Chem., 1998, 70, 1229, Polimery, 1997, 42, 587; O.Olabisi, M. Atiqullah and W. Kaminsky, J. Macromol. Sci., Rev. Macromol. Chem. Phys., 1997, C37, 519; W. Kaminsky, H. Sinn and H. G. Zachmann, Makromol Chem. Rapid Commun., 1982, 3, 327; W. Kaminsky, M. Miri, H. Sinn and R. Woldt, Makromol. Chem. Rapid Commun., 1983, 4, 417. 2 J. A. Ewen, J. Am. Chem. Soc., 1984, 106, 6355; J. A. Ewen, R. L. Jones, J. A. Razavi and J. D. Ferrara, J. Am. Chem. Soc., 1988, 110, 6255; J. A. Ewen, J. Mol. Cat. A. Chem., 1998, 128, 103; M.K. Leclerc and R. M. Waymouth, Angew. Chem., Int. Ed. Engl., 1998, 37, 922; A. L. McKnight and R. M. Waymouth, Chem. Rev., 1998, 98, 2587; G. M. Diamond, R. P. Jordan and J. L Petersen, Organometallics, 1996, 15, 4045; Z. Y. Guo, D. C. Swenson and R. F. Jordan, Organometallics, 1994, 13, 1424; X. J. Song, M. Thornton-Pett and M. Bochmann, Organometallics, 1998, 17, 1004; M. Bochmann, J. Chem. Soc., Dalton Trans., 1996, 255; M. Bochmann and S. J. Lancaster, J.Organomet. Chem., 1995, 497, 55; G. G. Hlatky, H. W. Turner and R. R. Eckman, J. Am. Chem. Soc., 1989, 111, 2728. 3 H. Welborn, Jr. and J. A. Ewen, US Pat., 5 324 800, 28th Jun. 1994. 4 K. P. Reddy and J. L. Petersen, Organometallics, 1989, 8, 2107.J. Chem. Soc., Dalton Trans., 1999, 1049–1059 1059 5 K. P. Reddy and J. L. Petersen, Organometallics, 1989, 8, 547. 6 J. Cacciola, K. P. Reddy and J. L. Petersen, Organometallics, 1992, 11, 665. 7 S. Jungling, R. Mulhaupt and H. Plenio, J. Organomet. Chem., 1993, 460, 191. 8 S. Ciruelos, T. Cuenca, J. C. Flores, R. Gomez, P. Gomez-Sal and P. Royo, Organometallics, 1993, 12, 944. 9 R. Gomez, T. Cuenca, P. Royo, W. A. Herrmann and E. Herdtweck, J. Organomet. Chem., 1990, 382, 103. 10 I. E. Nifant’ev, K. A. Butakov, Z. G. Aliev and I. F. Urazovskii, Organomet. Chem. USSR, 1991, 4, 622. 11 U. Siemeling and P. Jutzi, Chem. Ber., 1992, 125, 31. 12 I. E. Nifant’ev, M. V. Borsov, A. V. Churakov, S. G. Mkoyan and L. O. Atovmyan, Organometallics, 1992, 11, 3942. 13 G. M. Diamond, M. L. H. Green, P. Mountford, N. A. Popham and A. N. Chernega, J. Chem. Soc., Chem. Commun., 1994, 103. 14 G. M. Diamond, A. N. Chernega, P. Mountford and M. L. H. Green, J. Chem. Soc., Dalton Trans., 1996, 921. 15 G. M. Diamond, M. L. H. Green, N. A. Popham and A. N. Chernega, J. Chem. Soc., Chem. Commun., 1994, 727. 16 H. Werner, M. Treiber, A. Nessel, F. Lippert, P. Betz and C. Kruger, Chem. Ber., 1992, 125, 337. 17 H. Werner, D. Schneider and M. Schulz, Chem. Ber., 1992, 125, 1017. 18 D. Schneider and H. Werner, Organometallics, 1993, 12, 4420. 19 N. A. Popham, D.Phil. Thesis, Oxford, 1995. 20 H. Eshtiagh-Hosseini and J. F. Nixon, J. Less-Common Met., 1978, 61, 107. 21 J. L. Burmeister, E. A. DeardorV, A. Jensen and V. H. Christiansen, Inorg. Chem., 1970, 9, 58. 22 R. B. King and A. Efraty, J. Organomet. Chem., 1970, 23, 527. 23 D. H. Gibson, W.-L. Hsu, A. L. Steinmetz and B. V. Johnson, J. Organomet. Chem., 1981, 208, 89. 24 J. H. Ammeter, J. Magn. Reson., 1978, 30, 299. 25 J. L. Robbins, N. Edelstein, B. Spencer and J. C. Smart, J. Am. Chem. Soc., 1982, 104, 1882; F. F. Kneubuhl, J. Chem. Phys., 1960, 33, 1074. 26 V. C. Williams and M. L. H. Green, unpublished work. 27 B. C. Wailes and H. Weigold, Inorg. Synth., 1990, 28, 259. 28 E. Samuel and R. Setton, C. R. Hebd. Seances Acad. Sci., 1963, 256, 443. 29 W. Kaminsky, R. Engehausen, R. Zoumis, W. Spalek and J. Rohrmann, Makromol. Chem., 1992, 193, 1643. 30 L. E. Schock and T. J. Marks, J. Am. Chem. Soc., 1988, 110, 7701. 31 W. Spalek, M. Antberg, V. Dolle, R. Klein, J. Rohrmann and A. Winter, New J. Chem., 1990, 14, 499. 32 J. A. Ewen, L. Haspeslagh, J. L. Attwood and H. Zhang, J. Am. Chem. Soc., 1987, 109, 6544. 33 J. A. Ewen, M. J. Elder, R. L. Jones, L. Haspeslagh, J. L. Attwood, S. G. Bott and K. Robinson, Makromol. Chem., Makromol. Symp., 1991, 48/49, 253. 34 M. L. H. Green and N. Ishihara, J. Chem. Soc., Dalton Trans., 1994, 657. 35 H. N. Cheng and D. A. Smith, Macromolecules, 1986, 19, 2065. 36 H. N. Cheng and J. A. Ewen, Makromol. Chem., 1989, 190, 1931. 37 T. Tsutsui, A. Mizuno and N. Kashiwa, Polymer, 1989, 30, 428. 38 A. Zambelli, P. Localetti, G. Bajo and F. A. Bovey, Macromolecules, 1975, 8, 687. 39 W. Kaminsky, M. Miri, H. Sinn and R. Woldt, Makromol. Chem., Rapid Commun., 1983, 4, 417. 40 E. Giannetti, G. M. Nicoletti and R. Mazzochi, J. Poly. Sci., Polym. Chem. Ed., 1985, 23, 2117. 41 I. E. Nifant’ev, P. V. Ivchenko and M. V. Borzov, J. Chem. Res. (S), 1992, 162. Paper 8/09944H

ISSN:1477-9226    

DOI:10.1039/a809944h

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1999, 1061–1066 1061 Adducts of the Lewis acid [B(C6F5)3] with transition metal oxo compounds Georgina Barrado,a Linda Doerrer,b Malcolm L. H. Green a and Michael A. Leech b a Inorganic Chemistry Laboratory, South Parks Road, Oxford, UK OX1 3QR b Chemical Crystallography Laboratory, 9 Parks Road, Oxford, UK OX1 3PD Received 7th December 1998, Accepted 16th February 1999 The oxo anions [WO4]22 and [ReO4]2 react with the Lewis acid molecule [B(C6F5)3] to give the tris-adduct [WO{OB(C6F5)3}3]22 or the mono adduct [ReO3{OB(C6F5)3}]2 respectively.The crystal structure of the salt [nPr4N]2[WO{OB(C6F5)3}3] has been determined. The h-cyclopentadienyl-oxo compounds [Re(h-C5R5)O3] (where R = H or Me) react with [B(C6F5)3] giving the mono adducts [(h-C5R5)ReO2{OB(C6F5)3}] and the crystal structure of the compound where R = H has been determined. The triazacyclononane compounds [LMO3] (where L = N,N9,N0-trimethyl-1,4,7-triazacyclononane and M = Mo or W) with [B(C6F5)3] give the mono adducts [LMO2{OB(C6F5)3}] and the crystal structure of the tungsten compound is reported.The compound [Re{HB(pz)3}O3] [where {HB(pz)3} = hydridotris(1-pyrazolyl) borate] and [B(C6F5)3] gives the mono adduct [HB(pz)3ReO2{OB(C6F5)3}]. The Lewis acid [B(C6F5)3] is readily available as thermally stable volatile white crystals which are soluble in toluene. The solutions react only slowly with water and oxygen. Therefore, compared with the very volatile BF3 (bp 299.9 8C), which rapidly hydrolyses giving HF, the Lewis acid [B(C6F5)3] is much more convenient for the exploration of the Lewis base properties of transition metal compounds.Adducts formed between [B(C6F5)3] and metal-alkyl 1,2,3 metal-hydrido 4 and metal-oxo compounds5–8 are now well established. In this work we describe further the Lewis base properties of some neutral and anionic transition metal-oxo compounds. Results and discussion Treatment of the oxo compound [nPr4N]2[WO4] in CH2Cl2 solution with three equivalents of [B(C6F5)3] gave the compound [nPr4N]2[WO{OB(C6F5)3}3] 1 as white, microcrystalline, air-stable crystals in good yield.The compound 1 is very soluble in CH2Cl2, slightly soluble in toluene and insoluble in light petroleum ether and pentane. It has been characterised by IR and NMR (11B, 1H, 13C, 19F) spectroscopies, elemental analysis and X-ray diVraction. The analytical and selected spectroscopic data are given in Table 1 for compound 1 and the other new compounds 2–10.Crystals of 1 suitable for X-ray diVraction studies were grown from CH2Cl2/light petroleum ether (bp 30–40 8C). Selected distances and angles are given in Table 2 and the structure of the anion is shown in Fig. 1. The oxygen atoms attached to the tungsten have a slightly distorted tetrahedral geometry. The W–O–B units adopt a nearly linear disposition with angles 174.5(2), 174.4(1) and 170.1(1)8.The four W–O distances are within the range found for W–O bonds in the compound [tBuNH3][WO4] (1.60(6)–1.81(2) Å),9 but the three W–O distances of the atoms bonded to boron (1.781(2), 1.785(1), and 1.786(1) Å) are 0.07 Å longer than the distance to the uncoordinated oxygen atom (1.714(2) Å). The lengthening of the M–O bond in the W–O–B systems is of the same order as those reported for the M–O–B systems in the compounds [Ti{OB(C6F5)3}(acac)2], [V{OB(C6F5)3}(acac)2], [MoO{OB- (C6F5)3}(acac)2],6 [V(OBPh3)(C22H22N4)] 10 and [(C9H21N3)- WO2(OBPh3)].11 The O–B distances in 1 are 1.491(3), 1.508(3) and 1.494(3) Å and they lie in the range of O–B distances for other M–O–B adducts (from 1.460(6) in [(Me5Cp)2ZrOB- (C6F5)3] 5 to 1.59(3) Å in [(C9H21N3)WO2(OBPh3)] 11,12).The 11B NMR spectrum of 1 shows a very broad peak at d = 1, well within the range of a tetra-coordinated boron species and in the region of the other known M–O–B adducts.5,6,13 It was not possible to identify the W]] O absorption band in the IR spectrum because [B(C6F5)3] absorbs strongly in the same region as M]] O bonds.When [Pr4N]2[WO4] was reacted with one, two or four equivalents of [B(C6F5)3] the 11B NMR spectra of these solutions show the broad peak around 1 ppm indicating that a metal-oxo Lewis acid adduct has formed. The reaction of [nPr4N]2[WO4] with one equivalent of the Lewis acid led to the formation of a colourless oil that could not be fully characterised, but the reaction with 2–4 equivalents of [B(C6F5)3] yields only [nPr4N]2[WO{OB(C6F5)3}3] 1, as white crystals.Fig. 1 Crystal structure of 1, [nPr4N]2[WO{OB(C6F5)3}3], with a view along the W]] O bond and fluorine atoms removed for clarity.1062 J. Chem. Soc., Dalton Trans., 1999, 1061–1066 Table 1 Analytical and spectrscopic data Compound and analysis a [nPr4N]2[WO{OB(C6F5)3}3]?0.5CH2Cl2 1 White C (42.9) 42.9 H (2.6) 2.4 N (1.3) 1.3 B (1.5) 1.4 W (8.4) 8.2 NMR data b (ppm) 1H: 0.96 [t (7), 24H, NCH2CH2CH3], 1.63 [m, 16H, NCH2CH2CH3], 2.99 [m, 16H, NCH2CH2CH3] 13C: 148.0 [d (240), C6F5], 139.3 [d (246), C6F5], 136.8 [d (246), C6F5], 121.7 [s, C6F5], 60.8 [s, NCH2- CH2CH3], 15.5 [s, NCH2CH2CH3], 10.1 [s, NCH2CH2CH3] 19F: 2140.5 [d (19), C6F5], 2167.7 [t (19), C6F5], 2172.9 [dd (19, 19), C6F5] 11B: 1.0 very broad [nBu4N]2[WO{OB(C6F5)3}3] 2 Colorless oil C (45.5) 45.5 H (3.2) 3.35 N (1.2) 1.2 B (1.4) 1.4 W (8.1) 9.8 1H: 0.96 [t (7), 24H, NCH2CH2CH2CH3], 1.35 [m, 16H, NCH2CH2CH2CH3], 1.58 [m, 16H, NCH2- CH2CH2CH3], 2.99 [m, 16H, NCH2CH2CH2CH3] 13C: 146.6 [d (251), C6F5], 139.0 [d (254), C6F5], 136.4 [d (250), C6F5], 58.7 [s, NCH2CH2CH2CH3], 23.5 [s, NCH2CH2CH2CH3], 19.3 [s, NCH2CH2CH2CH3], 12.6 [s, NCH2CH2CH2CH3] 11B: 0.4 broad [nPr4N][ReO3{OB(C6F5)3}] 3 White C (38.0) 37.4 H (3.0) 3.7 N (1.5) 1.65 B (1.1) 0.9 Re (19.6) 20.5 1H: 1.04 [t (7), 12H, NCH2CH2CH3], 1.70 [m, 8H, NCH2CH2CH3], 3.09 [m, 8H, NCH2CH2CH3] 13C: 147.7 [d (234), C6F5], 140.0 [d (204), C6F5], 137.1 [d (234), C6F5], 60.5 [s, NCH2CH2CH3], 15.5 [s, NCH2CH2CH3], 10.1 [s, NCH2CH2CH3] 19F: 2137.8 [dd (24, 9) C6F5], 2162.9 [t (21), C6F5], 2168.7 [ddd (24, 21, 7), C6F5 11B: 3.1 [PhCH2Ph3P][ReO3{OB(C6F5)3}] 4 White C (46.3) 46.8 H (2.0) 1.73 B (0.97) 0.8 Re (16.7) 18.5 1H: 7.84–6.48 [m, 20H, Ph], 4.47 [d (14), 2H, CH2] 13C: 148.0 [d (242), C6F5], 138.9 [d (189), C6F5], 136.8 [d (261), C6F5], 135.8–116.4 [m, Ph], 31.8 [d (45), PhCH2P] 19F: 2134.5 [dd (24, 9), C6F5], 2159.8 [t (20), C6F5], 2165.4 [ddd (24, 20, 8), C6F5] 11B: 3.0 s 31P: 19.1 s [Ph4P][ReO3{OB(C6F5)3}] 5 White C (45.8) 45.9 H (1.8) 1.4 B (1.0) 0.8 Re (16.9) 19.0 1H: 7.91–6.58 [m, 20H, Ph] 13C: 148.0 [d (242), C6F5], 139.7 [d (246), C6F5], 137.3 [d (219), C6F5], 135.8–117.3 [m, Ph] 19F: 2.7 s 31P: 20.5 s 11B: 3.3 s [Re(h-C5H5)O2{OB(C6F5)3}] 6 Yellow C (34.05) 34.2 H (0.6) 0.4 B (1.3) 1.0 Re (22.95) 22.9 1H: 7.12 [s, 5H, C5H5] 13C: 149.2 [s, C6F5], 141.8 [s, C6F5], 138.7 [s, C6F5], 117.6 [s, C6F5], 116.1 [s, C5H5] 19F: 2145.7 [s, C6F5], 2157.8 [s, C6F5], 2164.9 [s, C6F5] 11B: 5.2 s [Re(h-C5Me5)O2{OB(C6F5)3}] 7 Orange C (38.2) 38.3 H (1.7) 2.0 B (1.2) 1.0 Re (21.1) 21.0 1H: 2.21 [s, 15H, C5Me5] 13C: 148.0 [d (234), C6F5], 139.2 [d (364), C6F5], 136.2 [d (85), C6F5], 122.9 [s, C5Me5], 119.7 [s, C6F5], 10.3 [s, C5Me5] 19F: 2133.6 [s, C6F5], 2156.5 [s, C6F5], 2165.1 [s, C6F5] 11B: 22.0 s [LMoO2{OB(C6F5)3}]?0.5CH2Cl2 8 White C (38.0) 37.4 H (2.55) 2.8 B (1.2) 1.4 Mo (11.0) 11.55 N (4.8) 4.1 1H: 3.15 [s, 9H, CH3], 2.91 [s, 12H, CH2] 19F: 2135.1 [d (25), C6F5], 2164.2 [t (19), C6F5], 2169.3 [dd (25, 19),C6F5] 11B: 1.1 s [LWO2{OB(C6F5)3}]?0.5CH2Cl2 9 White C (34.5) 34.4 H (2.3) 2.3 N (4.4) 4.7 B (1.1) 1.2 W (17.7) 23.0 1H: 3.27 [m, 9H, CH3], 2.99 [m, 12H, CH2] 13C: 148.2 [s, C6F5], 139.4 [s, C6F5], 137.1 [s, C6F5] 19F: 2131.6 [d (23), C6F5], 2161.0 [t (21), C6F5], 2166.1 [dd (23, 21), C6F5] 11B: 0.3 s [{HB(pz)3}ReO2{OB(C6F5)3}] 10 Yellow C (33.8) 33.9 H (1.05) 1.6 N (8.8) 8.2 B (2.25) 1.9 Re (19.4) 19.2 1H: 8.17 [s, 1H, CH], 7.79 [s, 1H, CH], 6.35 [s, 1H, CH] 19F: 2134.9 [s, C6F5], 2156.6 [s, C6F5], 2164.0 [s, C6F5] 11B: 24.9 s a Given as (found) calc.%. b All NMR spectra were measured in CD2Cl2. The treatment of [nBu4N]2[WO4] with one, two, three, or four equivalents of [B(C6F5)3] gives oils. After many attempts a few crystals were obtained and the elemental analysis was consistent with the formula [nBu4N]2[WO{O3B(C6F5)3}3] 2. The 11B NMR spectrum of 2 shows a broad peak at d 0.4 which is consistent with the formation of a tris-adduct.The salts of the perrhenate anion C1[ReO4]2, where C = nPr4N, (PhCH2)Ph3P or Ph4P, react with [B(C6F5)3] to give the mono adducts [C][ReO3{OB(C6F5)3}], where C = nPr4N 3, (PhCH2)Ph3P 4 or Ph4P 5, as white air-stable crystalline solids. These salts are very soluble in CH2Cl2, but insoluble in light petroleum ether, pentane or toluene.Despite many attemptsJ. Chem. Soc., Dalton Trans., 1999, 1061–1066 1063 it was not possible to grow crystals of 3–5 suitable for single-crystal X-ray diVraction. The NMR and IR spectra and elemental analyses of 3–5 support the proposed formulation of the products as the mono adducts. The 11B chemical shifts (3.1, 3.0 and 3.3 ppm respectively) are in the region expected for tetra-coordinated boron and inside the range for M–O–B adducts.5,6,13 The presence of the [B(C6F5)3] group is confirmed by 19F and 13C NMR spectra and by an IR spectrum.Treatment of the mono anion [ReO4]2 with an excess of [B(C6F5)3] gives only the mono adduct. Treatment of the oxo-complexes [Re(h-C5R5)O3] (where R = H or Me) with [B(C6F5)3] gives the mono adducts [(h-C5R5)- ReO2{OB(C6F5)3}] (where R = H 6 or R = Me 7) as air stable yellow and orange crystalline solids respectively. Both 6 and 7 are very soluble in CH2Cl2; 6 is insoluble in light petroleum ether, but 7 is slightly soluble.The compounds [Re(h-C5R5)O3] with an excess of [B(C6F5)3] give only the mono adducts. The spectroscopic data and elemental analysis support the proposed formulation and the crystal structure of 6 has been determined. The structure of 6 is shown in Fig. 2 and selected distances and angles are given in Table 3. The Re]] O distances of the terminal Re]] O groups are 1.708(4) and 1.705(3) Å and these are similar to the values for the Re]] O bonds in the starting compound 14 and also the related compounds [Re(h-C5H4Me)O3] 15 and [Re(h-C5Me4Et)O3].16 The Re–O distance in the Re–O–B system is 0.07 Å longer, as expected.The O–B distance is in the range for other known M–O–B adducts. 5,6,9,11,13 The W–O–B units in 1 show a linear disposition (174.5(2), 174.4(1) and 170.1(1)8), but in [(h-C5R5)- ReO2{OB(C6F5)3}] the unit has a bent arrangement (149.4(2)8). The M–O–B groups in the related compounds are also linear 6,12,17 or bent.5,10,11 Treatment of the triazacyclononane compounds [LMO3] (where L = N,N9,N0-trimethyl-1,4,7-triazacyclononane and M = Mo or W) with [B(C6F5)3] gives the mono adducts [LMO2- {OB(C6F5)3}], (where M = Mo 8, W 9) in good yields as white crystalline, air-stable solids.The compounds 8 and 9 are slightly Fig. 2 Crystal structure of 6, [Re(h-C5H5)O2{OB(C6F5)3}], with fluorine atoms removed for clarity. Table 2 Selected distances (Å) and angles (8) for the compound [nPr4N]2[WO{OB(C6F5)3}3], 1 W(1)–O(1) W(1)–O(101) W(1)–O(201) W(1)–O(301) O(101)–B(101) O(201)–B(201) O(301)–B(301) 1.714(2) 1.786(1) 1.785(1) 1.781(2) 1.491(3) 1.508(3) 1.494(3) O(1)–W(1)–O(10) O(101)–W(1)–O(201) O(1)–W(1)–O(201) O(101)–W(1)–O(301) O(1)–W(1)–O(301) O(201)–W(1)–O(301) W(1)–O(101)–B(101) W(1)–O(201)–B(201) W(1)–O(301)–B(301) 110.06(8) 111.16(7) 108.12(8) 109.29(7) 108.90(7) 109.28(7) 170.1(1) 174.4(1) 174.5(2) soluble in CH2Cl2, or hot toluene and insoluble in light petroleum ether or pentane.Reactions with an excess of [B(C6F5)3] yield only the mono adducts. The compounds 8 and 9 have been characterised by NMR and IR spectroscopy and elemental analysis. Due to their low solubility it was not possible to obtain satisfactory 13C NMR spectra. The 11B NMR spectra show resonances in the region expected for tetra-coordinated boron at 1.1 ppm for 8 and at 0.25 ppm for 9. The crystal structure of 9 has been determined.There are two molecules in the asymmetric unit, but the distances and angles of each are nearly identical. The structure of one molecule is shown in Fig. 3 and selected distances and angles are given in Table 4. The W–O distances of the uncoordinated oxygen atoms (average 1.721 Å) are 0.12 Å longer than the W–O distance of the coordinated oxygen (average 1.852 Å), in agreement with the expected lengthening of a metal oxygen bond upon coordination of the oxygen to the boron.6,10,11 The B–O distance (average 1.506 Å) lies in the range of the other known M–O–B moieties5,11 and the M–O–B unit adopts a bent disposition (average 141.18).The closely related compound [LWO2(OBPh3)] 11 also has a bent disposition (154.2(10)8). The W–O distance of the coordinated oxygen is slightly longer (0.07 Å) in 9 than in [LWO2(OBPh3)] and the O–B distance is slightly shorter (0.08 Å). This can be attributed to [B(C6F5)3] being a stronger Lewis acid than BPh3.Fig. 3 Crystal structure of one molecule of 9, [LWO2{OB(C6F5)3}], from the asymmetric unit, with fluorine atoms removed for clarity. Table 3 Selected distances (Å) and angles (8) for [Re(h-C5H5)O2- {OB(C6F5)3}], 6 Re(1)–O(1) Re(1)–O(2) Re(1)–O(3) O(3)–B(4) 1.708(4) 1.705(3) 1.775(3) 1.568(5) O(1)–Re(1)–O(2) O(2)–Re(1)–O(3) O(1)–Re(1)–O(3) Re(1)–O(3)–B(4) 104.9(2) 105.4(2) 104.4(2) 149.4(2) Table 4 Selected distances (Å) and angles (8) for the compound [LWO2{OB(C6F5)3}], 9 Molecule 1 in asymmetric unit Molecule 2 in asymmetric unit W(1)–O(2) W(1)–O(37) W(1)–O(38) W(1)–N(97) W(1)–N(101) W(1)–N(104) O(2)–B(3) O(2)–W(1)–O(37) O(2)–W(1)–O(38) O(37)–W(1)–O(38) W(1)–O(2)–B(3) 1.850(3) 1.708(3) 1.719(3) 2.357(4) 2.334(4) 2.384(4) 1.499(5) 107.6(1) 104.6(1) 106.7(2) 140.6(3) W(39)–O(40) W(39)–O(75) W(39)–O(76) W(39)–N(83) W(39)–N(87) W(39)–N(90) O(40)–B(41) O(40)–W(39)–O(76) O(75)–W(39)–O(76) O(40)–W(39)–O(75) W(39)–O(40)–B(41) 1.853(3) 1.720(3) 1.770(3) 2.298(4) 2.329(4) 2.290(3) 1.513(5) 104.3(1) 103.9(2) 103.9(1) 141.5(3)1064 J.Chem. Soc., Dalton Trans., 1999, 1061–1066 When the colourless compounds [Re{HB(pz)3}O3] (where {HB(pz)3} = hydridotris(1-pyrazolyl)borate) and [B(C6F5)3] are mixed in CH2Cl2 the solution immediately becomes yellow and the compound [HB(pz)3ReO2{OB(C6F5)3}], 10, can be isolated as yellow crystals. It has been characterised by NMR studies and elemental analysis. Instead of the two expected peaks, one for the pyrazolylborate ligand and the other for the Lewis acid, the 11B NMR spectrum shows a single broad peak at 24.9 ppm due to both ligands.The peak is rather broad so it seems that the two signals overlap, since the presence of the {HB(pz)3} ligand is confirmed by the 1H NMR spectrum which shows three singlets at 8.17, 7.79 and 6.35 ppm. The 19F NMR spectrum clearly shows the three resonances with chemical shifts of 2134.9, 2156.6 and 2164.0 ppm typical of [B(C6F5)3] adducts.In addition the elemental analysis supports the proposed formula. Due to the low stability of the compound it was not possible to record the 13C NMR spectrum or grow crystals of good quality for X-ray diVraction. In about 2 hours the solutions of 10 became colourless even when kept at 220 8C. In the solid state under nitrogen, it decomposed in a week. In conclusion, the synthesis and the structures proposed for the new compounds are shown in Scheme 1. The greater basicity of the dianion [WO4]22 compared to the mono anion [ReO4]2 is manifested by formation of the tris-adduct by the tungstate anion compared to the mono-adduct formed with the perrhenate anion.The formation of these adducts with the metal-oxo anions suggests there can be an extensive chemistry of related oxo-anion adducts. Experimental All manipulations were carried out under an N2 atmosphere by using standard Schlenk or dry-box techniques. Light petroleum ether (bp 40–60 8C), pentane, toluene, and dichloromethane were dried over suitable reagents and distilled under N2.The compounds [nBu4N]2[WO4], [Re(h-C5H5)O3], [Re(h- C5Me5)O3], [LWO3] (where L = N,N9,N0-trimethyl-1,4,7-triazacyclononane, M = Mo, W), [Re{HB(pz)3}O3] [{HB(pz)3} = hydridotris(1-pyrazolyl)borate] and [B(C6F5)3] were prepared as described.14,18–22 NMR spectra were recorded using a Bruker WM 300 spectrometer for 1H, 13C, 11B and 31P at 300, 75.5, 96 and 121.5 MHz respectively or a Varian AM 500 for 1H, 13C, 11B and 19F at 500, 125.7, 160 and 470 MHz respectively.The 1H and 13C chemical shifts are reported with respect to SiMe4, 11B to BF3?OEt3, 19F to CHF3 and 31P to trimethyl phosphate in D2O. Chemical shifts are given in ppm, a positive sign indicates a down field shift relative to the standard, and coupling constants in Hz. Infrared spectra were recorded as KBr discs in a Perkin-Elmer FT 1710 spectrometer, mass spectra by the EPSRC National Mass Spectrometry Service Centre.Elemental analyses were obtained by the analytical department of this laboratory. Syntheses [nPr4N]2[WO{OB(C6F5)3}3] 1. The compound [nPr4N]2[WO4] was prepared as described for [nBu4N]2[WO4] 18 using nPr4NOH instead of nBu4NOH. A mixture of [nPr4N]2[WO4] (200 mg, 0.32 mmol) and [B(C6F5)3] (495 mg, 0.97 mmol) was stirred in CH2Cl2 (20 ml) for 1 h. The colourless solution was filtered and the filtrate was concentrated to ca. 10 ml and then layered with light petroleum ether (10 ml) and cooled to 220 8C.Small colourless crystals were obtained after 48 h. Yield: 503 mg, 72%. [nBu4N]2[WO{OB(C6F5)3}3] 2. A mixture of [nBu4N]2[WO4] 18 (200 mg, 0.27 mmol) and [B(C6F5)3] (420 mg, 0.82 mmol) was stirred in CH2Cl2 (20 ml) for 1 h. The solution was filtered and the solvent removed in vacuum to give a colourless oil. [C][ReO4] (C 5nPr4N, (PhCH2)Ph3P, Ph4P). The compounds [C][ReO4] (C = nPr4N, (PhCH2)Ph3P, Ph4P) were prepared by addition of a solution of nPr4NOH, PhCH2Ph3PCl or Ph4PCl in water to a solution of NH4ReO4 in water.In each case there was an immediate formation of a white precipitate. These precipitates were collected by filtration, washed with water and dried in vacuum. These products were used without further purification or characterisation. Scheme 1 (i) In CH2Cl2 at room temperature for 1 h. (ii) In CH2Cl2 at room temperature for 1.5 h for 8 and reflux in CH2Cl2 for 4 h for 9. (iii) In CH2Cl2 at room temperature for 15 min.J.Chem. Soc., Dalton Trans., 1999, 1061–1066 1065 Table 5 Crystallographic collection and processing parameters for compounds 1, 6 and 9 Molecular formula Formula weight Crystal system Space group a/Å b/Å c/Å b/8 Unit-cell volume/Å3 Formula units per cell, Z rcalc/g cm23 m(Mo-Ka)/mm21 T/K Crystal size/mm qmax/8 No. of reflections: total unique in refinement [I > 3s(I)] No. of variables Residual electron density minimum maximum RR w [nPr4N]2[WO{OB(C6F5)3}3] 1 C78H56WO4B3F45N2 2156.51 Monoclinic P21/c 16.157(1) 21.204(1) 23.912(1) 92.731(2) 8182.78 4 1.75 1.59 100 0.21 × 0.23 × 0.32 26.69 58107 16848 14377 1198 20.42 0.98 0.0303 0.0301 [(h-C5H5)ReO2{OB(C6F5)3}] 6 C23H5ReO3BF15 811.27 Monoclinic P21/n 11.300(1) 9.585(1) 20.596(1) 90.648(2) 2230.62 4 2.42 5.67 100 0.25 × 0.15 × 0.05 26.72 14286 4469 3989 388 21.80 1.80 0.0346 0.0362 [LWO2{OB(C6F5)3}] 9 C56H46W2O6B2F30Cl4N6 2000.10 Monoclinic P21 13.289(1) 14.028(1) 18.270(1) 110.256(3) 3195.22 2 2.08 3.98 100 0.20 × 0.15 × 0.10 26.69 22644 6588 6496 956 20.92 0.73 0.0189 0.0220 [nPr4N][ReO3{OB(C6F5)3}] 3.A mixture of [nPr4N][ReO4] (200 mg, 0.2 mmol) and [B(C6F5)3] (234 mg, 0.4 mmol) was stirred in CH2Cl2 (20 ml) for 1 h. The colourless solution was filtered, the solvent removed in vacuum giving a white solid which was washed with light petroleum ether (2 × 10 ml) to yield a white solid. Yield: 365 mg, 81%. [(PhCH2)Ph3P][ReO3{OB(C6F5)3}] 4.A mixture of [(PhCH2)- Ph3P][ReO4] (300 mg, 0.5 mmol) and [B(C6F5)3] (254 mg, 0.5 mmol) was stirred in CH2Cl2 (25 ml) for 1 h. The solution was filtered and the solvent removed from the filtrate under reduced pressure. The white residue was washed with pentane (2 × 10 ml) to yield a white solid. Yield: 402 mg, 73%. [Ph4P][ReO3{OB(C6F5)3}] 5. [Ph4P][ReO3{OB(C6F5)3}] was prepared as a white solid as described for 4 from [Ph4P][ReO4] (200 mg, 0.39 mmol) and [B(C6F5)3] (174 mg, 0.34 mmol).Yield: 287 mg, 78%. [(Á-C5H5)ReO2{OB(C6F5)3}] 6. A solution of [B(C6F5)3] (171 mg, 0.33 mmol) in CH2Cl2 (5 ml) was added with stirring to a solution of [Re(h-C5H5)O3] (50 mg, 0.17 mmol) in CH2Cl2 (10 ml). The reaction mixture was stirred for 1 h. The resulting yellow-orange solution was filtered and concentrated to ca. 5 ml, then the solution was layered with light petroleum ether (5 ml) and stored at 220 8C overnight giving yellow crystals. Yield: 102 mg, 76%.[(Á-C5Me5)ReO2{OB(C6F5)3}] 7. The compound was prepared as described for [(h-C5H5)ReO2{OB(C6F5)3}] 6 using [Re(h-C5Me5)O3] (50 mg, 0.14 mmol) and [B(C6F5)3] (138 mg, 0.28 mmol), after three days at 220 8C red-orange crystals were obtained. Concentration of the mother liquor aVorded another crop. Yield: 87 mg, 73%. [LMoO2{OB(C6F5)3}] 8. A solution of [B(C6F5)3] (244 mg, 0.48 mmol) in CH2Cl2 (10 ml) was added with stirring to a suspension of [LMoO3] (150 mg, 0.48 mmol) in CH2Cl2 (10 ml).The reaction mixture was stirred for 1.5 h at room temperature. The resulting colourless solution was filtered and the solvent was removed under reduced pressure giving an oily solid. This was washed with light petroleum ether (2 × 5 ml) to give a white solid. Yield: 290 mg, 74%. [LWO2{OB(C6F5)3}]?0.5CH2Cl2 9. A solution of [B(C6F5)3] (190 mg, 0.37 mmol) in CH2Cl2 (5 ml) was added with stirring to a suspension of [LWO3] (150 mg, 0.37 mmol) in CH2Cl2 (10 ml) at room temperature.The reaction mixture was then refluxed for 4 h. The resulting colourless solution was filtered and the solvent was removed using reduced pressure giving an oily solid. The residue was washed with light petroleum ether (2 × 5 ml) giving a white solid. Yield: 220 mg, 65%. [{HB(pz)3}ReO2{OB(C6F5)3}] 10. A mixture of [{HB(pz)3}- ReO3] (50 mg, 0.11 mmol) and [B(C6F5)3] (115 mg, 0.22 mmol) was dissolved in CH2Cl2 (15 ml) and stirred for 15 min, then the yellow solution was filtered and concentrated to half volume, light petroleum ether (10 ml) was added and the solution was concentrated under reduced pressure until a crystalline yellow solid was formed.The solid was filtered, washed with light petroleum ether (2 × 5ml) and dried in vacuum. Yield: 77 mg, 72%. Crystal structure determination Crystals of the compounds [nPr4N]2[WO{OB(C6F5)3}3], 1, [(h-C5H5)ReO2{OB(C6F5)3}], 6, and [LWO2{OB(C6F5)3}], 9, were grown by slow diVusion of light petroleum ether into saturated solutions of the complexes in CH2Cl2 at 220 8C.The selected crystals were mounted on a nylon fibre using a drop of perfluoropolyether oil. They were rapidly cooled to 100 K in a flow of cold nitrogen using an Oxford Cryosystems CRYOSTREAM cooling system. The crystal data are given in Table 5. The data were collected on an Enraf–Nonius DIP2020 imageplate diVractometer using graphite monochromated Mo-Ka radiation (l = 0.7107 Å). The images were processed using the DENZO and SCALEPACK suite of programs.23 Data were corrected for Lorentz and polarisation eVects and a partial absorption correction applied by multi-frame scaling of the image-plate data using equivalent reflections. Structures were solved by direct methods, SIR92,24 giving all non-hydrogen atom positions, and refined using full-matrix least-squares procedures with anisotropic thermal parameters for all non-hydrogen atoms.The hydrogen atoms were placed in calculated positions during the final cycles of refinement.A three parameter Chebychev weighting scheme 25 and corrections for anomalous dispersion were applied to all data. All1066 J. Chem. Soc., Dalton Trans., 1999, 1061–1066 crystallographic calculations were carried out using CRYSTALS26 on a PC/AT computer. Neutral atom scattering factors were taken from reference 27. CCDC reference number 186/1355. See http://www.rsc.org/suppdata/dt/1999/1061/ for crystallographic files in .cif format.Acknowledgements We thank the Spanish Government for financial support (to G. B.) and St. John’s College, University of Oxford, for a Junior Research Fellowship (L. D.). G. B. thanks the European Commission for a Marie Curie Fellowship (contract no. ERBFMBICT 950343). References 1 X. Yang. C. L. Stern and T. J. Marks, J. Am. Chem. Soc., 1991, 113, 3623. 2 M. Bochmann, Angew. Chem., Int. Ed. Engl., 1992, 31, 1181. 3 R. Gomez, M. L. H. Green and J.L. Haggitt, J. Chem. Soc., Dalton Trans., 1996, 939. 4 A. J. Graham, D.Phil. Thesis, Oxford, 1998. 5 A. R. Siedle, R. A. Newmark, W. M. Lamanna and J. C. HuVman, Organometallics, 1993, 12, 1491. 6 J. R. Galsworthy, M. L. H. Green, M. Müller and K. Prout, J. Chem. Soc., Dalton Trans., 1997, 1309. 7 J. R. Galsworthy, J. C. Green, M. L. H. Green and M. Müller, J. Chem. Soc., Dalton Trans., 1998, 15; L. H. Doerrer, J. R. Galsworthy, M. L. H. Green and M. A. Leech, J. Chem. Soc., Dalton Trans., 1998, 2483. 8 L. H. Doerrer, J. R. Galsworthy, M. L. H. Green, M. A. Leech and M. Müller, J. Chem. Soc., Dalton Trans., 1998, 3191. 9 A. Thiele and J. Fuchs, Z. Naturforsch., Teil B, 1979, 34, 145. 10 C. H. Young, J. A. Laad and V. L. Goedken, J. Coord. Chem., 1988, 18, 317. 11 P. Schreiber, K. Wieghardt, B. Nuber and J. Weiss, Z. Naturforsch., Teil B, 1990, 45, 619. 12 J. Fischer, J. Kress, J. A. Osborn, L. Ricard and M. Wesolek, Polyhedron, 1987, 6, 1839; J. Kress, M. Wesolek, J.-P. Le Ny and J. A. Osborn, J. Chem. Soc., Chem. Commun., 1982, 1039. 13 G. S. Hill, L. Manojlovic-muir, K. W. Muir and R. J. Puddephatt, Organometallics, 1997, 16, 525. 14 F. E. Kühn, W. A. Herrmann, R. Hahn, M. Elison, J. Blumel and E. Herdtweck, Organometallics, 1994, 13, 1601. 15 W. A. Herrmann, M. Taillefer, C. M. Bellefon and J. Behm, Inorg. Chem., 1991, 30, 3247. 16 W. A. Herrmann, E. Herdtweck, M. Flöel, J. Kulpe, U. Küsthardt and J. Okuda, Polyhedron, 1987, 6, 1165. 17 B. Cashin, D. Cunningham, J. F. Gallagher and P. Mcardle, Polyhedron, 1989, 8, 1753. 18 T. M. Che, V. W. Day, L. C. Francesconi, M. F. Fredrich, W. G. Klemperer and W. Shim, Inorg. Chem., 1985, 24, 4055. 19 W. A. Herrmann and J. Okuda, J. Mol. Catal., 1987, 41, 109. 20 P. S. Roy and K. Wieghardt, Inorg. Chem., 1987, 26, 1885. 21 I. A. Degnan, W. A. Herrmann and E. Hedtweck, Chem. Ber., 1990, 123, 1347. 22 A. N. Chernega, A. J. Graham, M. L. H. Green, J. Haggitt, J. Lloyd, C. P. Mehnert, N. Metzler and J. Souter, J. Chem. Soc., Dalton Trans., 1997, 2293. 23 Z. Otwinowski and W. Minor, Methods Enzymol., 1976, 276. 24 A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M. C. Burla, G. Polidori and M. Camalli, J. Appl. Crystallogr., 1994, 27, 435. 25 J. R. Carruthers and D. J. Watkin, Acta Crystallogr., Sect. A, 1979, 35, 698. 26 D. J. Watkin, C. K. Prout, R. J. Carruthers and P. Betteridge, CRYSTALS, Chemical Crystallography Laboratory, Oxford, UK, 1996. 27 International Tables for X-ray crystallography, Vol. IV, 1974, Table 2.2B. Paper 8/09519A

ISSN:1477-9226    

DOI:10.1039/a809519a

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1999, 1067–1075 1067 Novel Cu–Se clusters with Se–layer structures: [Cu32Se7(SenBu)18- (PiPr3)6], [Cu50Se20(SetBu)10(PiPr3)10], [Cu73Se35(SePh)3(PiPr3)21], [Cu140Se70(PEt3)34] and [Cu140Se70(PEt3)36] Nianyong Zhu and Dieter Fenske * Institut für Anorganische Chemie, Universität Karlsruhe, Engesserstr. Geb. Nr. 30.45, 76128 Karlsruhe, Germany. E-mail: dieter.fenske@chemie.uni-karlsruhe.de Received 4th January 1999, Accepted 8th February 1999 The clusters [Cu32Se7(SenBu)18(PiPr3)6], [Cu50Se20(SetBu)10(PiPr3)10], [Cu73Se35(SePh)3(PiPr3)21], [Cu140Se70(PEt3)34] and [Cu140Se70(PEt3)36] were prepared from the reactions of CuCl, RSeSiMe3 and Se(SiMe3)2 (R = Ph, nBu or tBu) in the presence of phosphine ligands and their molecular structures were determined.The Se–Bu bond was cleaved more easily than the Se–Ph bond under the reaction conditions studied. Introduction The synthesis of transition-metal-containing complexes of selenium and tellurium has received1 less attention than that of the lighter chalcogen elements oxygen and sulfur. Over the past five years there has been increasing 2 interest in this area, partly due to the discovery of selenium in some enzymes,3 and the realisation that Se- and Te-containing compounds may have important applications, such as precursors for low-bandgap semiconductors, nanomaterials and window materials in solar cells.4 We are interested 2a in metal chalcogenide clusters both as potential nanoparticles and as intermediate structures between small molecules and bulk materials.We have previously developed 5–12 a general route to copper– selenide–phosphine clusters: CuX 1 nPR3 1 ��� Se(SiMe3)2 X = Cl or OAc [Cu2x 2 ySex(PR3)m] 1 Me3SiX (1) The driving force of the reaction is the elimination of Me3SiX. The formation of the cluster complexes depends very much on the nature of the phosphine. The reaction conditions, the solvents used for the synthesis and the ratio of Cu :PR3 also play a significant role in governing the structures of the products isolated, which crystallize under these conditions. A large number of complexes have been prepared with this method that cover a range of nuclearities.In order of increasing size, these include low nuclearity clusters such as [Cu12Se6(PR3)8] (R3 = Et2Ph5 or nPr3 or Cy3 6), ranging to [Cu20Se13(PEt3)12],7 [Cu26Se13(PCy3)10],6 [Cu26Se13(PEt2Ph)14],6 [Cu29Se15(PiPr3)12],8 [Cu30Se15(PR3)12] (R3 = iPr3 8 or tBu2Me9), [Cu31Se15(SeSiMe3)(PtBu2Me)12],10 [Cu32Se16(PPh3)12],11 [Cu36- Se18(PtBu3)12],8 [Cu44Se22(PEt2Ph)18],12 [Cu44Se22(PnButBu2)12],12 [Cu48Se24(PMe2Ph)20],10 [Cu52Se26(PPh3)16] 11 and [Cu59Se30- (PCy3)15],6 and finally to very large macromolecules such as [Cu70Se35(PEt3)22],7 [Cu70Se35(PiPr2Me)21],10 [Cu70Se35(PtBu2- Me)21],10 [Cu72Se36(PPh3)20] 11 and [Cu146Se73(PPh3)30].13 The last of these is one of the largest clusters characterised so far.We observe a systematic colour change from red (Cu12) through brown (Cu36) and dark brown (Cu70) to black (Cu146). Preliminary studies also show a relationship between size and conductivity.14 The above synthetic method has recently been modified 15,16 to create a route to mixed seleno–selenato and telluro–tellurato clusters by employing RESiMe3 or mixtures of E(SiMe3)2 and RESiMe3 (E = Se, Te). In this paper we describe the synthesis and characterisation of five novel clusters synthesized by the reaction of CuX (X = Cl or SCN) with mixtures of RSeSiMe3 (R = Ph or nBu or tBu) and Se(SiMe3)2 in the presence of alkyl phosphines PR3 (R3 = Et or iPr).Experimental Standard Schlenk-line techniques were employed throughout on a double-manifold vacuum line with high purity dry nitrogen. Solvents for reactions were distilled under nitrogen from appropriate drying agents prior to use. The compounds CuCl and CuSCN were purchased and CuCl was purified prior to use.The phosphines 17 and PhSeSiMe3 18 were prepared according to standard literature procedures. Syntheses RSeSiMe3 (1a R 5 nBu and 1b tBu). In an adaptation of the preparation of nBuTeSiMe3,16a grey selenium (63.0 g, 0.80 mol) was suspended in THF (700 mL) in a three-necked roundbottom flask (2 L) equipped with a water cooled condenser and dropping funnel. nBuLi (500 mL, 1.6 M in hexane) or tBuLi (500 mL, 1.6 M in pentane) was added dropwise to the rapidly stirred suspension at 0 8C, during which time the black Se gradually dissolved with the formation of a dark brown solution.After the addition was complete, stirring was continued for one hour to give a pale yellow solution. After cooling the solution to 0 8C freshly distilled ClSiMe3 (100 mL, 0.80 mol) was added dropwise with vigorous stirring, during which time a white precipitate (LiCl) was formed. After complete addition, the mixture was stirred for one day at room temperature.The solution was filtered to remove LiCl and the solvent removed by distillation. Fractional vacuum distillation aVorded 1a; 80 g, 48%, bp 75–80 8C at 20 mmHg or 1b; 94 g, 56%; bp 77–80 8C at 20 mmHg. Both yellow liquids contained about 15% each of Se(SiMe3)2 and Bu2Se according to 1H NMR. Attempts to remove these impurities by further fractional distillation were unsuccessful. 1H NMR(C6D6): 1a d 2.38 (t, a-CH2), 1.55 (m, b-CH2), 1.29 (m, g-CH2), 0.80 (t, CH3), 0.30 [s, Si(CH3)3]; 1b d 1.44 (s, 3CH3), 0.37 [s, Si(CH3)3].[Cu32Se7(SenBu)18(PiPr3)6] 2. nBuSeSiMe3 (0.70 mL, 3.0 mmol) was added to a suspension of CuCl (0.25 g, 2.5 mmol) and PiPr3 (0.25 mL, 1.33 mmol) in Et2O (25 mL) at 0 8C with vigorous stirring. After two hours a clear dark red solution was1068 J. Chem. Soc., Dalton Trans., 1999, 1067–1075 Table 1 Crystallographic data Compound Empirical formula MF (000) Crystallized from Crystal size/mm Crystal system Space group Za /Å b/Å c/Å b/8 V/Å3 r (calcd.)/g cm23 T/K m(Mo-Ka)/mm21 q range/8 Index range Reflections collected Independent reflections Reflectopms observed Parameters refined Reflections refined R indices (obs.refl.) Goodness of fit 2 C126H288Cu32P6Se25 5996.66 5824 Diethyl ether 0.5 × 0.4 × 0.4 Monoclinic P2(1)/n 2 16.628(3) 29.937(6) 19.847(4) 101.79(3) 9671(3) 2.059 203(2) 8.222 2.25 to 25.99 216 < h < 19 236 < k < 19 216 < l < 24 11978 11283 (Rint = 0.12) 8792 (I > 2s(I)) 871 11278 R1 = 0.074, wR2 = 0.20 1.055 3 C69H160Cu25OP5Se15 3933.72 7568 Diethyl ether 0.4 × 0.4 × 0.3 Monoclinic C2/m 4 3399.8(7) 2016.0(4) 1992.3(4) 119.44(3) 11892(4) 2.197 203(2) 9.052 2.33 to 25.96 238 < h < 39 224 < k < 24 218 < l < 24 28803 11157 (Rint = 0.068) 8990 (I > 22s(I)) 430 11156 R1 = 0.068, wR2 = 0.19 1.034 4 C207H441Cu73P21Se38 11234.99 21688 Diethyl ether 0.4 × 0.3 × 0.3 Monoclinic P2(1)/c 4 38.460(8) 25.720(5) 36.310(7) 99.61(3) 35414(12) 2.107 203(2) 8.309 1.58 to 22.62 241 < h < 33 227 < k < 27 232 < l < 38 99014 43379 (Rint = 0.17) 12845 (I > 2s(I)) 1870 40709 R1 = 0.095, wR2 = 0.22 1.166 5 C204H510Cu140P34Se70 18439.90 34736 Diethyl ether 0.4 × 0.1 × 0.1 Monoclinic C2/c 4 34.116(7) 48.433(10) 30.942(6) 110.47(3) 47898(17) 2.557 203(2) 11.520 1.83 to 24.20 225 < h < 39 253 < k < 55 234 < l < 18 48996 29499 (Rint = 0.13) 8293 (I > 2s(I)) 1303 19169 R1 = 0.11, wR2 = 0.27 1.903 6 C226H540Cu140P36Se70 18796.30 177632 Pentane 0.3 × 0.2 × 0.2 Monoclinic P2(1)/m 2 31.650(6) 22.550(5) 37.840(8) 97.92(3) 26749(9) 2.334 203(2) 10.276 1.81 to 24.08 235 < h < 32 225 < k < 25 243 < l < 41 110121 42703 (Rint = 0.16) 12326 (I > 2s(I)) 1371 42694 R1 = 0.086, wR2 = 0.22 1.440 formed, which was allowed to stand at room temperature.Orange crystals grew within three days. Yield 80%. [Cu52Se20(SetBu)10(PiPr3)10] 3.tBuSeSiMe3 (0.62 mL, 2.6 mmol) was added to a solution of CuCl (0.25 g, 2.5 mmol) and PiPr3 (0.47 mL, 2.5 mmol) in Et2O (25 mL) forming a dark brown solution. The mixture was allowed to stand at room temperature. After one week a brown precipitate was filtered oV. A small amount of brown crystals grew within two weeks from the fil Yield 10%. [Cu73Se35(SePh)3(PiPr3)21] 4. PhSeSiMe3 (0.32 mL, 1.25 mmol) was added to a solution of CuCl (0.25 g, 2.5 mmol) and PiPr3 (0.94 mL, 5.00 mmol) in Et2O (25 mL) to give a clear yellow solution.This was cooled to 270 8C and Se(SiMe3)2 (0.16 mL, 0.64 mmol) was added. The mixture was then allowed to warm slowly to room temperature, during which time the colour changed from yellow, through brown to black. On standing at room temperature black crystals grew within several weeks. Yield 40%. [Cu140Se70(PEt3)34] 5. Method A. The procedure is the same as for 4, except PEt3 (0.50 mL, 3.2 mmol) was used. Black needle crystals were formed in one week together with a brown precipitate.Yield 30%. Method B. nBuSeSiMe3 (0.70 mL, 3.0 mmol) was added to a solution of CuSCN (0.32 g, 2.6 mmol) and PEt3 (0.50 mL, 3.2 mmol) in Et2O (50 mL) to give a brown solution. This was allowed to stand at room temperature. Black crystals grew within one week. Yield 90%. [Cu140Se70(PEt3)36] 6. nBuSeSiMe3 (0.62 mL, 2.6 mmol) was added to a solution of CuCl (0.25 g, 2.5 mmol) and PEt3 (0.82 mL, 5.2 mmol) in pentane (25 mL) forming a yellow solution.The solution was allowed to stand at room temperature. A small amount of black crystals grew within several weeks. Yield 10%. Crystal structure analyses Single crystal X-ray structural analysis of compounds 2 to 6 were performed using a Stoe-IPDS diVractometer (Mo-Karadiation) equipped with an imaging plate area detector and a rotating anode. Structure solution and refinement were carried out with SHELXS-8619 and SHELXL-9320 software by direct methods techniques.The weighting scheme applied was of the form w = 1/[s2(Fo 2) 1 (aP)2 1 bP] [a,b = refined variables, P = 1/3 max. (Fo 2,0) 1 2/3 Fc 2]. All calculations were performed on a Silicon Graphics INDY computer. Molecular diagrams were prepared using the SCHAKAL 97 program.21 Table 1 lists the summary crystallographic data for 2 to 6. CCDC reference number 186/1348. See http://www.rsc.org/suppdata/dt/1999/1067/ for crystallographic files in .cif format.Compound 2. Data were collected at 203 K in the q range 2.25 to 25.998 with a detector distance of 70 mm and 15 min radiation time per exposure; the f went step by step per exposure from 08 to 808 with the scan-step Df value of 0.58; 11978 reflections were measured of which 11283 were independent and 8792 were considered observed with I > 2s(I). The structure was solved by direct methods and refined by full-matrix leastsquares on F 2. All Se, Cu P and C atoms were refined anisotropically.Hydrogen atoms (except for one disordered nBu group) were placed in calculated positions. R = 0.074 with a goodness of fit value of 1.055. Parameters refined = 871. Compound 3. Data were collected at 203 K in the q range 2.33 to 25.968 with a detector distance of 70 mm with 12 min radiation time per exposure; the f went step by step per exposure from 08 to 72.58 with a scan-step Df value of 0.58; 28803 reflections were measured of which 11157 were independent and 8990 were considered observed with I > 2s (I).The structure was solved by direct methods and refined by full-matrix least-squares on F 2. All Se, Cu and P atoms were refined anisotropically, C atoms were refined isotropically. R = 0.068 with a goodness of fit value of 1.034. Parameters refined = 430. The asymmetric unit contains one quarter of a molecule. Within this unit six iPr groups of two phosphines and one tBu group are disordered. The disorder of the tBu group has been modelled using two sites related by rotation about the Se–C bond.A similar procedure involving rotation about the Cu–PJ. Chem. Soc., Dalton Trans., 1999, 1067–1075 1069 bond has been used to describe the disorder of the phosphine groups. This is a good model for P(1), but is less satisfactory for P(3). Compound 4. Data were collected at 203 K in the q range 1.58 to 22.628 with a detector distance of 90 mm with 20 min radiation time per exposure; the f went step by step per exposure from 08 to 1058 with a scan-step Df value of 0.38; 99014 reflections were measured of which 43379 were independent and 12845 were considered observed with I > 2s (I).The structure was solved by direct methods and refined by full-matrix least-squares on F 2. All Se, Cu and P atoms were refined anisotropically, C atoms were refined isotropically. R = 0.095 with a goodness of fit value of 1.166. Parameters refined = 1870. Three Cu atoms in the middle of the three face-shared octahedral holes formed by Se atoms (see structure description below) were considered to have a total population of 2 and each has a population of 0.67.Because the iPr groups are quite disordered, many of the C atoms have large thermal parameters and some of them could not be found. Compound 5. Data were collected at 203 K in the q range 1.83 to 24.208 with a detector distance of 80 mm and 30 min radiation time per exposure; the f went step by step per exposure from 08 to 738 with a scan-step Df value of 0.38; 48996 reflections were measured of which 29499 were independent and 8293 were considered observed with I > 2s(I).The structure was solved by direct methods and refined by full-matrix least-squares on F 2. All Se, Cu and P atoms were refined anisotropically, C atoms were refined isotropically. R = 0.11 with a goodness of fit value of 1.903. Parameters refined = 1303. Many Et groups of PEt3 and the solvent molecules could not be found.Compound 6. Data were collected at 203 K in the q range 1.81 to 24.088 with a detector distance of 80 mm and 20 min radiation time per exposure; the f went step by step per exposure from 08 to 1108 with a scan-step Df value of 0.38; 110121 reflections were measured of which 42703 were independent and 12326 were considered observed with I > 2s(I). The structure was solved by direct methods and refined by full-matrix least-squares on F 2. All Se, Cu and P atoms were refined anisotropically, C atoms were refined isotropically.R = 0.086 with a goodness of fit value of 1.440. Parameters refined = 1371. Many Et groups of PEt3 and the solvent molecules could not be found. Results and discussion Synthesis Reactions among CuCl, PhSeSiMe3 and bidentate phosphine to produce some copper–seleno–phenylselenato clusters have been reported.14 However, the reactions of equimolar quantities of CuCl, PR3 (R3 = Et3, iPr3, tBu3) and PhSeSiMe3 in pentane produce only yellow solutions that remain unchanged over a period of a few months.This colour suggests that only lownuclearity copper–selenium molecules are present. Evidence for this was the isolation of [Cu2(SePh)2(PtBu3)2],22 which contains doubly-bridged phenylselenolate ligands and two terminallycoordinated PtBu3. As there was clearly little hope of isolating large clusters from these solutions, this line of investigation was not continued. We have observed that high nuclearity Cu–Se clusters often contain structural features similar to those found in copper selenide minerals.In order to facilitate the formation of similar motifs we have investigated reactions using a mixture of selenide sources. A combination of PhSeSiMe3 and Se(SiMe3)2 in a ratio of 2 : 1 has been used with a quantity of CuX chosen so that equal amounts of X and SiMe3 moieties were present. In order to avoid the formation of products containing only Se22, i.e. without RSe2, we first allowed the phosphine–copper salt mixture to react with PhSeSiMe3 before adding Se(SiMe3)2.Despite this most of the products we have characterised contain few or no PhSe2 fragments. The reaction performed in the presence of PtBu3 yielded the cluster [Cu36Se18(PtBu3)12],8 which was previously isolated from reactions employing Se- (SiMe3)2. When PEt3 was used we obtained [Cu140Se70(PEt3)34] 5. Although this compound contains no PhSe2 groups, it is diVerent from the material obtained from similar reactions in which Se(SiMe3)2 was the only selenium source, [Cu70Se35- (PEt3)22].7 From the reactions in which PiPr3 is present we obtain [Cu73Se35(SePh)3(PiPr3)21] 4, a markedly larger cluster than that obtained from only Se(SiMe3)2, [Cu30Se15(PiPr3)12]. The presence of a relatively small number of PhSe2 groups and the diVerence of these products from those obtained when only Se(SiMe3)2 is employed suggests that Se–C bond scission may accompany cluster growth.For a more comprehensive insight into this, we have synthesised the silylated reagents nBuSeSiMe3 and tBuSeSiMe3 from which we have been unable to remove the impurities Se(SiMe3)2 and SeBu2. We have used these reagents in a study of the CuX–PR3–BuSeSiMe3 system. The reaction of CuCl, PEt3 and nBuSeSiMe3 in diethyl ether produces the known compound [Cu70Se35(PEt3)22]. Replacement of CuCl by CuSCN leads to the new cluster 5 in high yield. It is possible that [Cu70Se35(PEt3)22] is an intermediate in the formation of 5 as it can be recognised as a fragment of the larger structure (see below).When pentane is used as the solvent for the reaction of CuCl, PEt3 and nBuSeSiMe3, cluster [Cu140Se70(PEt3)36] 6 is produced, which diVers from 5 only in the inclusion of two extra phosphine ligands. The smaller [Cu70Se35(PEt3)22] cluster may again be an intermediate. The similar reactions with the use of tBuSeSiMe3 lead invariably to [Cu70Se35(PEt3)22]. These reactions illustrate the profound influence of counter ion and solvent.The Se–Bu bond was cleaved in the reactions. For reactions in the presence of PiPr3 we observe a product dependance on the Cu :PR3 ratio in addition to the dependence on the RSeSiMe3 reagent used. From the reactions of nBuSe- SiMe3 the cluster [Cu32Se7(SenBu)18(PiPr3)6] 2 is obtained when half a molar equivalent of phosphine is used. For larger amounts of phosphine no crystalline product could be isolated.Reactions with tBuSeSiMe3 additionally show diVerent behaviour depending on which copper salt is used. For CuCl we obtain cluster [Cu50Se20(SetBu)10(PiPr3)10] 3 when one equivalent of phosphine is used, and cluster [Cu70Se35(PiPr3)21] when 1.5 to 4 equivalents of phosphine are used. With CuSCN as the starting material we isolate [Cu70Se35(PiPr3)21] when 1 to 3 equivalents of phosphine are employed. When this is increased to 4 equivalents of PiPr3 the known cluster [Cu30Se15(PiPr3)12] can be isolated.Structure description Compound [Cu32Se7(SenBu)18(PiPr3)6] 2 crystallizes from a diethyl ether solution in the monoclinic space group P2(1)/n and contains an inversion center at Se(10). Fig. 1 shows the molecular structure which consists of CuI, Se22, SenBu2 and PiPr3 ligands. The heavy atoms Cu and Se form the Cu32Se25 framework of the molecule. The six phosphine ligands are terminally linked to copper atoms [Cu(4), Cu(8), Cu(14) and their symmetry equivalents], each of which is doubly bridged by two SenBu2 ligands to give a trigonal-planar coordination. The corresponding Cu–Se bond lengths of Cu(4), Cu(8) and Cu(14) are similar to the other Cu–SenBu bonds, within a narrow range from 2.439(2) to 2.500(2) Å. 24 Cu atoms have trigonal-planar coordination geometry, of which six copper atoms [Cu(10), Cu(13) and Cu(15) and their symmetry equivalents] are each coordinated by three SenBu21070 J. Chem. Soc., Dalton Trans., 1999, 1067–1075 ligands, six Cu atoms [Cu(1), Cu(2) and Cu(9) and symmetry equivalents] by two SenBu2 and one Se22 ligand and the other twelve by one SenBu2 and two Se22 ligands.The related Cu–Se bond lengths for Se22 and SenBu2 ligands are in the range 2.381(2)–2.489(2) Å. The remaining two Cu atoms [Cu(12) and its symmetry equivalent] have ideal tetrahedral coordination formed by four selenide ligands. Accordingly the Cu–Se bond lengths are relatively long [2.600(2)–2.622(2) Å] in comparison to the other Cu–Se bonds.In contrast to the 18 SenBu2 ligands that are m3-bridging, the seven selenide ligands have greater connectivity. The selenide ligand Se(10) on the inversion centre coordinates to eight copper atoms with two long and six short Cu–Se bonds. The other six selenide ligands act as m5-bridges. The 25 Se atoms together form a regular triple-layer structure which corresponds to a cubic close packed arrangement as shown in Fig. 2.The middle layer has 9 Se atoms in a 3 × 3 rhombohedral arrangement. The other two layers each have one Se atom less. The location of the copper atoms can be explained with reference to this selenium grid. The six phosphine-bound copper atoms bind to two Se atoms of SenBu2 ligands and are located Fig. 1 Molecular structure of Cu32Se7(SenBu)18(PiPr3)6 2 (Cu: purple; Se22: dark red; Se–nBu2: light red; P green); Cu–Cu contacts and carbon atoms are omitted for clarity. Selected bond lengths (Å) in 2: Cu(4)–Se(1) 2.439(2), Cu(8)–Se(5) 2.500(2); Cu(9)–Se(1) 2.388(2), Cu(1)–Se(1) 2.489(2), Cu(1)–Se(13) 2.381(2); Cu(12)–Se(10) 2.600(2), Cu(12)–Se(11) 2.600(2), Cu(12)–Se(12) 2.622(2), Cu(12)–Se(13) 2.605(2), average Se–C 1.98(2); average Cu–P 2.240(4).Fig. 2 The Se packing of 2 (Se22: dark red; Se–nBu2: light red); nonbonding Se ? ? ? Se contacts are shown within 5.5 Å and the lines between Se atoms in the middle Se layer are drawn in violet (Se ? ? ? Se contacts between layers are omitted for clarity). on the periphery.Six copper atoms [Cu(2), Cu(15) and Cu(16) and their symmetry equivalents] are located approximately within the Se-layers and have trigonal-planar coordination geometry with normal Cu–Se bond lengths of 2.406(2)– 2.483(2) Å. The other 20 Cu atoms are located between the Se layers. Two of these [Cu(12) and its symmetry equivalent] have ideal tetrahedral geometry, while the others are shifted away from the centre of tetrahedral holes.Hence each of them has only three normal Cu–Se bonds with the fourth Cu–Se distance being relatively long (3.18 Å to 3.35 Å). The structure of the Se lattice in 2 is similar to that in the bulk material Cu2Se;23 however, most Se atoms in 2 come from SenBu2 ligands. Despite this, the similarity between the structure of 2 and the non-molecular Cu2Se suggests that a fragment of a bulk structure may be stabilised to nanosize by coating the surface with suitable groups.This tendency is also observed in three silver–seleno–selenolate clusters,24 [Ag112Se48(SenBu)32- (PtBu3)12], [Ag114Se46(SenBu)34(PtBu3)14] and [Ag172Se40- (SenBu)92{Ph2P(CH2)3PPh2}4]. Compound [Cu50Se20(SetBu)10(PiPr3)10] 3 (Fig. 3) crystallises in the monoclinic space group C2/m. The molecule contains an inversion centre and a mirror plane, so that the asymmetric unit consists of only a quarter of the cluster. Fig. 3 Molecular structure of Cu50Se20(SetBu)10(PiPr3)10 3 (Cu: purple; Se22: dark red; Se–tBu2: light red; P: green); Cu–Cu contacts and carbon atoms are omitted for clarity.Selected bond lengths (Å) and bond angles (8) in 3: Cu(4)–Se(5) 2.347(2), Cu(4)–Se(9) 2.370(2); Cu(10)–Se(8) 2.394(2); Cu(8)–Se(1) 2.386(2), Cu(2)–Se(2) 2.517(2), Cu(2)–Se(6) 2.353(2), Cu(2)–Se(10) 2.379(2); Cu(6)–Se(1) 2.573(2), Cu(1)–Se(2) 2.875(2), Cu(9)–Se(4) 2.718(2), Cu(12)–Se(4) 2.595(1), Cu(15)–Se(7) 2.715(2); average Se–C 2.03(2); average Cu–P 2.230(4); Se(5)–Cu(4)–Se(9) 136.67(7).Fig. 4 The Se packing of 3 (Se22: dark red; Se–tBu2: light red); non-bonding Se ? ? ? Se contacts are shown within 5.5 Å and Se ? ? ? Se contacts between layers are omitted for clarity.J. Chem. Soc., Dalton Trans., 1999, 1067–1075 1071 Fig. 5 Molecular structure of Cu73Se35(SePh)3(PiPr3)21 4 (Cu: purple; Se22: dark red; Se–Ph2: light red; P: green); Cu–Cu contacts and carbon atoms are omitted for clarity. Selected bond lengths (Å) in 4: Cu(8)–Se(6) 2.783(7), Cu(8)–Se(10) 2.628(7), Cu(8)–Se(32) 2.338(8), Cu(8)–Se(33) 2.674(8); Cu(51)–Se(6) 2.726(7), Cu(51)–Se(29) 2.458(6), Cu(51)–Se(34) 2.489(6); Cu(54)–Se(20) 2.480(6), Cu(54)–Se(27) 2.396(6), Cu(54)–Se(28) 2.473(5); Cu(73)–Se(6) 2.407(6), Cu(73)–Se(9) 2.498(6), Cu(73)–Se(33) 2.391(5); Se(36)–Cu(14) 2.401(8), Se(36)–Cu(16) 2.372(5), Se(37)–Cu(3) 2.398(9), Se(37)–Cu(19) 2.378(6), Se(38)–Cu(66) 2.357(6), Se(38)–Cu(70) 2.390(10); average Se–C 1.91(2).On the mirror plane lie ten Se atoms [Se(2), Se(3), Se(4), Se(6), Se(10) and their symmetry equivalents], ten Cu atoms [Cu(1), Cu(2), Cu(3), Cu(9), Cu(15) and their symmetry equivalents] and two P atoms [P(3) and its symmetry equivalent].The other Cu, Se and P atoms each have three symmetry-equivalent atoms within the molecule. The ten phosphine-bound Cu atoms [Cu(13), Cu(14) and Cu(15) and equivalents] are located on the periphery of the Cu50Se30 cluster core with two diVerent types of coordination.Cu(13), Cu(14) and their symmetry equivalents have a trigonalplanar coordination mode, and Cu(15) and its symmetry equivalent are tetrahedrally coordinated. As expeted, the Cu–Se bond lengths for Cu(13), Cu(14) and their symmetry equivalents [2.435(2)–2.584(2) Å] are significantly shorter than those for Cu(15) and its equivalent [2.630(2) and 2.715(2) Å], but the Cu–P bonds have a very small range [2.227(2)–2.233(2) Å]. The other 40 Cu atoms are coordinated exclusively by Se atoms with two types of coordination geometries.Four Cu atoms [Cu(4) and its symmetry equivalents] are bound by two Se atoms, one Se22 [Se(5)] and one SetBu2 [Se(9)]. They have short Cu–Se bond lengths [2.347(2) and 2.370(2) Å] and a bent Se–Cu–Se bond angle [136.67(7)8]. The remaining 36 Cu atoms are located in a distorted trigonal-planar coordination geometry. Four of them [Cu(10) and its equivalents] bind purely to SetBu2 ligands, ten [Cu(2), Cu(8), Cu(11) and their symmetry equivalents] are each coordinated by two Se22 and one SetBu2 ligand, and the other 22 Cu atoms [Cu(1), Cu(3), Cu(5), Cu(6), Cu(7), Cu(9), Cu(12) and their symmetry equivalents] are bound only to Se22 ligands.The Cu–Se bond lengths for Se22 and SetBu2 are similar and lie within the range of 2.353(2) to 2.595(2) Å. One exception is the bond Cu(1)–Se(2) 2.875(2) Å, which is relatively long and indicates that Cu(1) tends to be only doubly coordinated by Se atoms in a similar fashion to Cu(4).The ten SetBu2 ligands [Se(8), Se(9) and Se(10) and their symmetry equivalents] are located on the sides of the molecule and act as m3-bridging ligands, while the 20 Se22 ligands have1072 J. Chem. Soc., Dalton Trans., 1999, 1067–1075 Fig. 6 Molecular structure of Cu140Se70(PEt3)34 5 (Cu: purple; Se22: dark red); Cu–Cu contacts and carbon atoms are omitted for clarity. Selected bond lengths (Å) in 5: Cu(60)–Se(24) 2.426(6), Cu(60)–Se(36) 2.383(5); Cu(10)–Se(5) 2.456(6), Cu(10)–Se(7) 2.380(8), Cu(10)–Se(21) 2.756(7), Cu(10)–Se(25) 2.739(10); Cu(26)–Se(7) 2.577(8), Cu(26)–Se(9) 2.467(6), Cu(26)–Se(25) 2.577(8); Cu(59)–Se(1) 2.485, Cu(59)–Se(9) 2.462(5), Cu(59)– Se(10) 2.379(7); Cu(63)–Se(2) 2.489(9), Cu(63)–Se(15) 2.464(5), Cu(63)–Se(25) 2.310(9); average Cu–P 2.20(1).more neighbouring Cu atoms. 14 Se22 ligands [Se(1), Se(5), Se(6) and Se(7) and their symmetry equivalents] are each bound to five copper atoms, four [Se(3) and Se(4) and their symmetry equivalents] bridge between five Cu atoms and two Se atoms [Se(2) and its equivalent] act as m7 ligands.This variation in the bonding ability of the Se22 ligand is also demonstrated by the series of Cu–Se–cluster complexes mentioned above. The 30 Se atoms in 3 form approximately three layers (Fig. 4). The middle layer lies on the mirror plane and consists of ten Se atoms. The other two layers are symmetry equivalent and also contain ten Se atoms each.In contrast to the central layer these atoms do not form a strictly planar array. Not considering the eight out-of-plane Se atoms [Se(7) and Se(8) and their equivalents], the other 22 Se atoms show hexagonal close packing. This type of packing has been found in other Cu–Se macromolecules. The Cu atoms are positioned in the tetrahedral holes or sites relating to the octahedral holes of the Se lattice. Eight Cu atoms [Cu(5) and Cu(12) and their equivalents] are located in the tetrahedral holes with long Cu–Se distances, while ten Cu atoms [Cu(3), Cu(4) and Cu(7) and their equivalents] are not in the octahedral holes, but shifted strongly from the centre of the holes with only three Cu–Se bonds.This fragment of the structure is present in larger Cu–Se macromolecules indicating that 3 could be an intermediate in the formation of [Cu70Se35- (PiPr3)21],25 which is obtained by changing the ratio of Cu to PiPr3. Compound [Cu73Se35(SePh)3(PiPr3)21] 4 crystallizes in the monoclinic space group P2(1)/c with four molecules in the unit cell.Fig. 5 shows the molecular structure and the Se38-lattice, which is based on [Cu70Se35(PiPr3)21] with three additional CuSePh groups on the three corners. The structure of the [Cu70Se35(PiPr3)21] fragment is almost the same as that seen in the other Cu70 clusters.7,9,10 The structure of the new [Cu70- Se35(PiPr3)21] compound, also synthesised in this work, has not yet been completely solved.The Cu70 clusters have the common feature that the 35 Se atoms build a layer type structure. It contains three Se layers with 10, 15 and 10 Se atoms, respectively that form a hexagonal close packing. The copper atoms are positioned according to the Se lattice. The detailed structure of 4 is not given here due to the similarity to the known Cu70 clusters.7,10 Two of the three CuSePh groups are located on the same edge of the molecule, so that no symmetry is present.The linkage of the three CuSePh groups at corners indicates that the corners of the Cu70 clusters are capable of being altered. Another type of corner structure has been found in the very large cluster Cu146,13 which has four Se and six Cu atoms at each corner.J. Chem. Soc., Dalton Trans., 1999, 1067–1075 1073 Fig. 7 Molecular structure of Cu140Se70(PEt3)36 6 (Cu: purple; Se22: dark red); Cu–Cu contacts and carbon atoms are omitted for clarity. Selected bond lengths (Å) in 6: Cu(43)–Se(13) 2.588(9), Cu(43)–Se(33) 2.632(9), Cu(43)–Se(36) 2.617(9); Cu(60)–Se(13) 2.419(4), Cu(60)–Se(36) 2.435(4), Cu(60)–Se(41) 2.380(4); Cu(15)–Se(18) 2.585(4), Cu(15)–Se(22) 2.468(4), Cu(15)–Se(36) 2.531(4); Cu(62)–Se(22) 2.464(4), Cu(62)–Se(23) 2.457(4), Cu(62)–Se(41) 2.390(4); average Cu–P 2.20(1).The structures of [Cu140Se70(PEt3)34] 5 and [Cu140Se70(PEt3)36] 6 are very similar and diVer only by the presence of two extra phosphine ligands at the periphery (Figs. 6 and 7). In 6, all three corners have the same [Cu5Se2(PEt3)4] fragment. The corner with four PEt3 has approximate twofold symmetry as found in [Cu70Se35(PEt3)22].7 In 5 two of the corners only have three phosphine ligands. Owing to the similarities of the structures, only 5 is discussed below. The 70 Se atoms in 5 build a layer-type structure with 21, 28 and 21 Se atoms in the three layers. They are arranged in a hexagonal close packing structure (Fig. 8), which is characteristic for the Cu–Se macromolecules previously characterised.Except for the corner Se atoms, 36 Se atoms of both outer Se layers construct 36 tetrahedral holes together with the Se atoms of the middle layer. 25 Se atoms of the middle Se layer, excepting the three corner Se atoms, build 50 tetrahedral holes with the Se atoms of the outer ones. Some of these holes are open at the edges and become completed with the help of phosphine ligands. Thus, there are 86 tetrahedral holes in the Se70 lattice and 30 octahedral holes.One finds 15 pairs of octahedral holes sharing an Se triangular face. Besides the 15 Cu atoms at the three corners, the remaining 125 copper atoms are positioned according to the 86 tetrahedral holes and 15 pairs of octahedral holes of the Se lattice. The 86 tetrahedral holes are each found to be occupied by one Cu atom, although many copper atoms deviate from the center of the holes. However, the two positions in the middle of the molecule are actually found to be only partly occupied.These two positions are treated as one disordered copper atom [Cu(43)]. Hence there are 85 Cu atoms in the tetrahedral holes. The remaining 40 Cu atoms are found in the 15 pairs of octahedral holes. Six of these 15 pairs are bi-capped by two CuPEt3 groups and have the third Cu atom in the middle. The other nine pairs contain two or three Cu atoms. The structure of [Cu70Se35(PEt3)22] 7 can be seen as a substructure of 5 and they feature similar structural properties.The structure of 5 is also very similar to [Cu146Se73(PPh3)30],13 the diVerence between them being found in the corner regions. Conclusion This paper presents the synthesis and crystal structures of novel Cu–Se and Cu–Se–SeR clusters that are synthesized by employing mixtures of RSeSiMe3 (R = Ph or nBu or tBu) and Se(Si- Me3)2 in the presence of alkyl phosphines PR3 (R3 = Et or iPr). The formation of the products is strongly influenced by the1074 J.Chem. Soc., Dalton Trans., 1999, 1067–1075 Fig. 8 The selenium lattice of 5 (first layer: light red; second layer and the related Se–Se lines in the same layer: green; third layer: dark red); nonbonding Se ? ? ? Se contacts are shown within 5.5 Å. reaction conditions and the starting materials, mainly by the phosphine ligand itself. However, for the same phosphine ligand, products of diVerent cluster size and structure can be isolated through the variation of other conditions. It is not yet possible to predict the structure of the possible products when using a new phosphine ligand.The clusters presented herein show a layer-type structure of Se atoms as their common structural feature, which is often found in the non-molecular binary bulk materials. The Se substructure in 2 is cubic close packed, while the Se lattices in 3–6 are hexagonal close packed. The synthetic method employed in this work provides a route by which fragments of a bulk structure may be stabilised by coating the surface with suitable groups. References 1 I.Dance and K. Fischer, Prog. Inorg. Chem., 1994, 41, 637; L. C. Roof and J. W. Kolis, Chem. Rev., 1993, 93, 1037; J. Arnold, Prog. Inorg. Chem., 1995, 43, 353. 2 (a) G. Schmid (Editor), Clusters and Colloids, From Theory to Applications, VCH, Weinheim, 1994; (b) L. J. de Jongh (Editor), Physics and Chemistry of Metal Cluster Compounds, Kluwer, Dordrecht, 1994. 3 T. C. Standtman, Annu. Rev. Biochem., 1990, 59, 111. 4 I. Dance and G. Lee, Spec. Publ. R. Soc. Chem., 1993, 131; W. Hirpo, S. Dhingra, A. C. Sutorik and M. G. Kanatzidis, J. Am. Chem. Soc., 1993, 115, 1357; H. B. Singh and N. Sudha, Polyhedron, 1996, 15, 745; N. Herron, in Handbook of Nanophase Materials, ed. A. Goldstein, Marcel Dekker Inc., New York, 1997, p. 221; W. S. Chen, J. M. Stewart and R. A. Mickelsen, Appl. Phys. Lett., 1985, 46, 1095; H. Okimura, T. Matsumae and R.Makabe, Thin Solid Films, 1980, 71, 53; R. H. Bube, Annu. Rev. Mater. Sci., 1990, 20, 19. 5 S. Dehnen, A. Schäfer, D. Fenske and R. Ahlrichs, Angew. Chem., Int. Ed. Engl., 1994, 33, 764. 6 A. Deveson, S. Dehnen and D. Fenske, J. Chem. Soc., Dalton Trans., 1997, 4491. 7 D. Fenske and H. Krautscheid, Angew. Chem., Int. Ed. Engl., 1990, 29, 1452. 8 D. Fenske, H. Krautscheid and S. Balter, Angew. Chem., Int. Ed. Engl., 1990, 29, 796. 9 D. Fenske, A. C. Deveson and S. Dehnen, J. Cluster Sci., 1996, 7, 351. 10 S. Dehnen and D. Fenske, Chem. Eur. J., 1996, 2, 1407. 11 A. Eichhöfer and D. Fenske, J. Chem. Soc., Dalton Trans., 1998, 2969. 12 S. Dehnen and D. Fenske, Angew. Chem., Int. Ed. Engl., 1994, 33, 2287. 13 H. Krautscheid, D. Fenske, G. Baum and M. Semmelmann, Angew. Chem., Int. Ed. Engl., 1993, 32, 1303. 14 D. van der Putten, D. Olevano, R. Zanoni, H. Krautscheid and D. Fenske, J. Electron. Spectrosc. Relat. Phenom., 1995, 76, 207. 15 M. Semmelmann, Dissertation, University of Karlsruhe, 1997; M. Semmelmann, D. Fenske and J. F. Corrigan, J. Chem. Soc., Dalton Trans., 1998, 2541. 16 (a) J. F. Corrigan, S. Balter and D. Fenske, J. Chem. Soc., Dalton Trans., 1996, 729; (b) J. F. Corrigan and D. Fenske, Angew. Chem., Int. Ed. Engl., 1997, 36, 1176.J. Chem. Soc., Dalton Trans., 1999, 1067–1075 1075 17 H. D. Kaesz and F. G. A. Stone, J. Org. Chem., 1959, 24, 635; A. H. Cowley and J. L. Mills, J. Am. Chem. Soc., 1969, 91, 2915. 18 N. Miyoshi, H. Ishii, K. Kondo, S. Mural and N. Sonoda, Synthesis, 1979, 300. 19 G. Sheldrick, SHELXS 86, Program for the solution of Crystal Structures, University of Göttingen, 1986. 20 G. Sheldrick, SHELXL 93, Program for Crystal Structure Determination, University of Göttingen, 1993. 21 E. Keller, SCHAKAL 97, A Computer Program for the Graphic Representation of Molecular and Crystallographic Models, Universität Freiburg, 1997. 22 N. Zhu and D. Fenske, unpublished work. Cell constants: a = 17.663(4), b = 12.265(3), c = 19.338(4) Å, b = 106.37(3)8, U = 4019(1) Å3, R = 0.039. 23 R. McLaren Murray and R. D. Heyding, Can. J. Chem., 1975, 53, 878. 24 D. Fenske, N. Zhu and T. Langetepe, Angew. Chem., Int. Ed., 1998, 37, 2639. 25 N. Zhu and D. Fenske, unpublished work. The structure is not yet solved due to disorder of copper atoms at the corners. Paper 9/00021F

ISSN:1477-9226    

DOI:10.1039/a900021f

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1999, 1077–1083 1077 Synthesis, spectroscopic and structural studies on transition metal carbonyl complexes of cyclic di- and tetra-selenoether ligands Maxwell K. Davies,a Marcus C. Durrant,b William Levason,a Gillian Reid a and Raymond L. Richards b a Department of Chemistry, University of Southampton, Highfield, Southampton, UK SO17 1BJ b Nitrogen Fixation Laboratory, John Innes Centre, Colney Lane, Norwich, UK NR4 7UH Received 18th December 1998, Accepted 3rd February 1999 Reaction of [M(CO)4(nbd)] (M = Cr or Mo, nbd = norbornadiene) or [W(CO)4(TMPA)] (TMPA = N,N,N9,N9- tetramethyl-1,3-propanediamine) with [8]aneSe2 (1,5-diselenacyclooctane) yielded the cis-disubstituted tetracarbonyl species [M(CO)4([8]aneSe2)] (M = Cr, Mo or W).The complexes [M9X(CO)5] (M9 = Mn, X = Cl, Br or I; M9 = Re, X = Cl or Br) reacted similarly with [8]aneSe2 to give fac-[M9X(CO)3([8]aneSe2)] in high yield. Infrared and multinuclear NMR spectroscopic studies confirmed these assignments and indicated a single species in solution; d(55Mn) lies in the same range as observed for other fac-[MnX(CO)3(diselenoether)] complexes, while for all of the compounds d(77Se) is to low frequency of [8]aneSe2 itself (d 137).Crystal structures of [W(CO)4([8]aneSe2)], [MnBr(CO)3([8]aneSe2)] and [ReBr(CO)3([8]aneSe2)] show the cyclic diselenoether chelating and adopting a chair-boat conformation. The compounds [16]aneSe4 (1,5,9,13-tetraselenacyclohexadecane) and L (1,6-diselena-3,4-benzocyclononane) reacted with the metal(II) species [{MoBr2(CO)4)}2] or [MI2(CO)3(NCMe)2] in CH2Cl2 solution to give seven-co-ordinate [{MoX2(CO)3}2([16]aneSe4)], [WI2(CO)3([16]aneSe4)], [MoX2(CO)3(L)] and [WI2(CO)3(L)], although these species decompose rapidly in co-ordinating solvents.Reaction of [16]aneSe4 with two molar equivalents of [M9Cl(CO)5] yielded the dinuclear complex [{MnCl(CO)3}2([16]aneSe4)] (in which the tetraselenoether is thought to bind in a bidentate manner to each Mn) and the mononuclear complex [ReCl(CO)3([16]aneSe4)] (which is thought to involve bidentate ligation to [16]aneSe4 with two free Se donors).The cationic species fac-[Mn(CO)3(h3-[16]aneSe4)]CF3SO3 was generated by treatment of [MnCl(CO)5] with AgCF3SO3 in Me2CO followed by addition of [16]aneSe4. Introduction Although the co-ordination chemistry of cyclic selenoether ligands has been studied in some detail over the past 10 years, this area is heavily dominated by complexes with metals from Groups 8–11 and there are no examples of low valent complexes involving Group 6 or 7 metal centres.1–11 We have become interested in investigating the chemistry of Group 6 and 7 metal carbonyl species with cyclic selenoethers since the carbonyl ligands may oVer a useful route into organometallic derivatives. A few years ago Yoshida and co-workers 12 showed that the tetrathia macrocyclic complex [MoBr2(Me8[16]aneS4)] (Me8- [16]aneS4 = 3,3,7,7,11,11,15,15-octamethyl-1,5,9,13-tetrathiacyclohexadecane) is reduced by Zn/Hg under a dinitrogen atmosphere to give the molybdenum(0) bis(dinitrogen) species trans-[Mo(N2)2(Me8[16]aneS4)], the first thioether complex to contain a dinitrogen ligand. Subsequently the electrochemistry and reactions of these species have been investigated in some detail.13 In view of the ability of the tetrathia macrocycle to facilitate formation of a N2 adduct, we were interested to determine whether the tetraselena macrocycle [16]aneSe4 (1,5,9,13-tetraselenacyclohexadecane) would accommodate a molybdenum centre within the cavity and subsequently bind N2 under appropriate conditions.With regard to Group 7 species, we have recently reported studies on manganese(I) carbonyl complexes involving bidentate selenoethers and shown that 55Mn NMR spectroscopy can yield important information regarding the Mn–Se interaction.14 In this paper we describe the results of our initial investigations on the preparation and characterisation of a range of complexes of Cr, Mo, W, Mn and Re involving [8]aneSe2 (1,5-diselenacyclooctane), L (1,6- diselena-3,4-benzocyclononane) and [16]aneSe4.Crystal structures of [MnBr(CO)3([8]aneSe2)], [ReBr(CO)3([8]aneSe2)] and [W(CO)4([8]aneSe2)] are also reported and these species are the first organometallic and early transition metal complexes incorporating cyclic selenoether ligands.Results and discussion The complexes [Cr(CO)4(nbd)], [Mo(CO)4(nbd)] (nbd = norbornadiene) and [W(CO)4(TMPA)] (TMPA = N,N,N9,N9-tetramethyl- 1,3-diaminopropane) react with one molar equivalent of [8]aneSe2 in refluxing toluene to give the zerovalent complexes [M(CO)4([8]aneSe2)] in high yield. These reactions were monitored by solution IR spectroscopy which showed gradual disappearance of the CO bands associated with the tetracarbonyl precursor and the growth of new bands associated with the product (Table 1).The FAB or electrospray mass spectra of the complexes typically show peaks with the correct isotopic distributions for [M(CO)4([8]aneSe2)]1. Further peaks due to fragmentation associated with loss of CO are also observed. Microanalyses and 1H NMR spectroscopic studies are consistent with the [M(CO)4([8]aneSe2)] assignment. The seven-co-ordinate molybdenum(II) species [{MoBr2- (CO)3}2([16]aneSe4)], [{MoI2(CO)3}2([16]aneSe4)], [MoBr2- (CO)3L] and [MoI2(CO)3L] are readily formed in high yield as brown solids by treatment of [16]aneSe4 or L with [{MoBr2-1078 J.Chem. Soc., Dalton Trans., 1999, 1077–1083 Table 1 IR (CO region), 13C-{1H}, 77Se-{1H}, 55Mn and 95Mo NMR spectroscopic data Compound [Cr(CO)4([8]aneSe2)] [Mo(CO)4([8]aneSe2)] [W(CO)4([8]aneSe2)] [MnCl(CO)3([8]aneSe2)] [MnBr(CO)3([8]aneSe2)] [MnI(CO)3([8]aneSe2)] [ReCl(CO)3([8]aneSe2)] [ReBr(CO)3([8]aneSe2)] [{MnCl(CO)3}2([16]aneSe4)] [ReCl(CO)3([16]aneSe4)] [Mn(CO)3([16]aneSe4)]CF3SO3 [{MoBr2(CO)3}2([16]aneSe4)] d [{MoI2(CO)3}2([16]aneSe4)] d [WI2(CO)3([16]aneSe4)] d [MoBr2(CO)3L] d [MoI2(CO)3L] d [WI2(CO)3L] d n& (CO)a/cm21 2005, 1901, 1882, 1859 2018, 1911, 1897, 1864 2012, 1935, 1892, 1855 2035, 1961, 1907 2033, 1959, 1910 2028, 1957, 1908 2037, 1947, 1896 2036, 1948, 1902 2033, 1956, 1913 2034, 1941, 1901 2032, 1947 2033, 1959, 1927 2018, 1947, 1883 2016, 1935, 1911 2039, 1981, 1913 2025, 1973, 1917 2019, 1958, 1902 d(13C-{1H}) 23.8, 21.6, 225.9, 220.6 25.0, 23.9, 216.5, 210.2 25.5, 25.0, 206.5, 204.7 22.6, 21.3, 19.9, 19.0, 216–221.5 22.5, 21.7, 21.2, 19.2, 217.0–223.0 23.8, 22.0, 21.7, 19.5, 216.4–220.5 26.5, 22.7, 20.7, 19.6, 186.0–192.0 26.4, 22.9, 20.8, 20.6, 190.7, 188.5 29.5–32.3, 14.4–25.4, 215.5–222.0 n.o. 24.6, 32.3, 216–222 n.o. n.o. n.o. n.o. n.o. n.o. d(77Se-{1H}) 134 119 84 79 70 60 30 19 See text n.o. 146.3 sh, 146.8 c n.o. n.o. n.o. n.o. n.o. n.o. d(55Mn)b ——— 2215 (1800) 2282 (1100) 2448 (1800) —— 2210 (2100) — 2499 (5000) —————— d(95Mo) — 21424 —————— — n.o.n.o.— n.o. n.o.— n.o. = Not obtained. a Spectra recorded in CHCl3 solution. b w2� 1 /Hz in parentheses. c At 200 K. d Spectra recorded as KBr discs. (CO)4}2] or [MoI2(CO)3(NCMe)2] as appropriate in CH2Cl2 solution. Similarly, the orange tungsten(II) species [WI2(CO)3- ([16]aneSe4)] and [WI2(CO)3L] are generated by treatment of [WI2(CO)3(NCMe)2] with one molar equivalent of [16]aneSe4 or L, respectively, in CH2Cl2 solution.Compound L was used in these reactions rather than [8]aneSe2 in an eVort to improve the solubilities of the products. Microanalyses and conductivity measurements are consistent with the formulations given. The products are moderately air-sensitive powders and are very poorly soluble in chlorocarbons and hydrocarbons, and while they do dissolve in co-ordinating solvents such as MeCN, dmf or dmso this results in rapid decomposition via displacement of the selenoether ligand.This was confirmed by 1H and 13C-{1H} NMR spectroscopic studies on the compounds in CD3CN and d6-dmso and precluded further NMR spectroscopic investigations on these species. Only [WI2(CO)3([16]aneSe4)] was suf- ficiently soluble in CD2Cl2 to able a 1H NMR spectrum to be obtained. This shows a set of broad multiplets in the methylene region associated with co-ordinated [16]aneSe4. While it was not possible to obtain FAB mass spectrometry data due to the poor solubilities of these complexes in the 3-nitrobenzyl alcohol matrix, atmospheric pressure chemical ionisation (APCI) mass spectra were recorded from freshly prepared MeCN solutions.In each case the highest mass peaks were consistent with either [16]aneSe4 or L as appropriate, again indicating very facile demetallation in this solvent. The IR spectra were recorded as pressed KBr discs, and each shows three CO stretching vibrations at frequencies comparable with those for the analogous seven-co-ordinate thioether macrocyclic derivatives, e.g.[{MoI2(CO)3}2([n]aneS4)] (n = 12, 14 or 16) and [{WI2(CO)3}2([14]aneS4)].15 Thus, the dinuclear molybdenum( II) complexes of [16]aneSe4 can be formulated as [Mo2X4(CO)6(m-[16]aneSe4-Se,Se9,Se0,Se-)] (X = Br or I). In contrast, the tungsten(II) complex of [16]aneSe4 is mononuclear with two co-ordinated Se donors and the other two free. Reaction of [WI2(CO)3(h2-[16]aneSe4)] with another molar equivalent of [WI2(CO)3(NCMe)2] does not permit formation of the dinuclear species, instead the reagents are recovered from the reaction mixture.In principle the seven-co-ordinate complexes of MoII and WII involving L can adopt two structures; either monomeric with the diselenoether chelating, or polymeric with L bridging between metal centres. We have examined these possibilities by means of molecular mechanics calculations. The conformational behaviour of L in solution has been explored by multinuclear NMR,16 and the ground state conformation in solution was deduced to be the same as that of the sulfur analogue.17 In order to determine the co-ordination possiblities which are available to this compound as a ligand, we have carried out a complete conformational search using the molecular modelling program CHEM-X.18 The standard forcefield for this program contains no parameters for selenium. However, the structures of a number of macrocyclic polyselenoethers have been determined by X-ray diVraction,1,19 and these were used together with IR data for simple dialkyl selenoethers 20 to generate a minimum parameter set for this class of compound (see Experimental section).The parameter set was able to reproduce the published crystal structure conformations with good accuracy. The conformational search was carried out using established ring searching principles,21 and the results are summarised in Table 2. The spread of energies for each conformer reflects the fact that diVerent input geometries converge to slightly diVerent versions of the same conformer, whilst the number of times each conformer was located gives an indication of the completeness of the search.The geometry of the global minimum was in excellent agreement with that of its sulfur analogue, as modelled by Rys et al.16 on the basis of NMR data. Applying our results to the behaviour of L as a ligand, it is evident that although the ground state conformation A is predisposed towards bridging co-ordination, there are reasonably low energy alternatives which could chelate to a single metal (C and D, Fig. 1). Conformer C has an energy approximately 5 kJ mol21 above the ground state and appears well suited for chelation to give a monomeric complex. Possible polymeric and monomeric complex structures were both successfully modelled by coupling conformers A and C respectively to trans- and cis-[WI2(CO)3Se2] cores produced from the crystal structures of [WI2(CO)3(PEt3)2] 22 and [WI2(CO)3({4-MeC6H4SCH2}2)] 23 respectively (Scheme 1).Which structure is obtained in practice probably depends on kinetic eVects; the low solubilities of the complexes of L in this work suggest that they may have the polymeric structure shown in Scheme 1a. Table 2 Summary of the results of the conformational analysis of 1,6-diselena-3,4-benzocyclononane. The letter designations of the conformers correspond to those in Fig. 1 a Conformer AB C DEF G Energy range/kJ 0.00–5.23 3.35–9.25 4.90–8.83 6.11–14.1 10.5–19.8 13.6–23.1 30.3–40.1 Number of times located 15 15 23 36 11 22 8 a Four conformers were discarded as high energy forms of other structures.J. Chem. Soc., Dalton Trans., 1999, 1077–1083 1079 The neutral fac-tricarbonyl species [MnX(CO)3([8]aneSe2)] (X = Cl, Br or I) and [ReX(CO)3([8]aneSe2)] (X = Cl or Br) are readily formed by reaction of [MnX(CO)5] or [ReX(CO)5] with [8]aneSe2 in refluxing CHCl3.The IR spectra of the products are consistent with Cs local symmetry (three CO stretching vibrations, 2a9 1 a0), although this alone does not distinguish mer from fac geometries. However, the frequencies of the bands show very good agreement with those for fac-[MnX(CO)3- (L–L)] (L–L = diselenoether ligand) the structures of which were confirmed by X-ray crystallography,14 hence indicating a fac arrangement for the [8]aneSe2 compounds. FAB Mass spectrometry typically shows peaks with the correct isotopic Fig. 1 Conformers of 1,6-diselena-3,4-benzocyclononane. Scheme 1 arrangement for [MX(CO)3([8]aneSe2)]1 and [M(CO)3([8]ane- Se2)]1. Reaction of [16]aneSe4 with two molar equivalents of [MCl(CO)5] in refluxing CHCl3 aVords the dinuclear 2 : 1 species [{MnCl(CO)3}2([16]aneSe4)] as an orange solid or the mononuclear species [ReCl(CO)3([16]aneSe4)] as a cream solid. The IR spectra of these products are very similar to those of the complexes of MnI and ReI of [8]aneSe2, showing three strong n(CO) bands at very similar frequencies.FAB Mass spectra do not show the parent ions, but do exhibit peaks attributed to [M(CO)3([16]aneSe4)]1. The bright orange cationic maganese(I) species fac-[Mn(CO)3(h3-[16]aneSe4)]CF3SO3 was generated by treatment of [Mn(CO)3(Me2CO)3]CF3SO3 (formed in situ by treatment of [MnCl(CO)5] with AgCF3SO3 in refluxing acetone) 24 with one molar equivalent of [16]aneSe4 in acetone solution.Solution IR spectroscopy on this highly soluble species shows two CO bands (2032, 1947 cm21), consistent with approximate C3v symmetry at MnI (a1 1 e). The frequencies are comparable with those for fac-[Mn(CO)3- {MeC(CH2SeMe)3}]1, the structure of which has been con- firmed crystallographically (2039, 1962 cm21).25 The electrospray mass spectrum shows an intense cluster of peaks consistent with [Mn(CO)3([16]aneSe4)]1, as well as fragments at lower m/z values. Treatment of this complex with Me3NO in CH2Cl2 at reflux leads to loss of the CO stretches associated with the fac-tricarbonyl species and the appearance of a new band at 1945 cm21.The electrospray mass spectrum of the product of this reaction shows no evidence for the tricarbonyl precursor, the highest intensity and highest mass peaks occurring at m/z = 597, consistent with [Mn(CO)2([16]aneSe4)]1. These data suggest that the Me3NO removes one of the CO ligands, allowing tetradentate co-ordination of [16]aneSe4 possibly giving the trans-dicarbonyl species [Mn(CO)2([16]ane- Se4)]CF3SO3.However, we have been unable to isolate a pure sample of this rather unstable compound to allow full characterisation. 77Se-{1H} NMR studies were restricted to the complexes of Cr0, Mo0, W0, MnI and ReI with [8]aneSe2, since those of MoII and WII show very limited solubility. The data are presented in Table 1, together with 55Mn and 95Mo NMR data where appropriate. For the [8]aneSe2 complexes we observe a progressive shift of d(77Se) to low frequency according to the series Cr æÆ Mo æÆ W, Mn æÆ Re and Cl æÆ Br æÆ I.In all cases d(77Se) is to low frequency of that for [8]aneSe2 itself (d 137). This contrasts with the situation for late transition metal halide derivatives of this ligand where high frequency shifts are observed.7,10 Unlike the analogous complexes with acyclic diselenoethers which show invertomers, when [8]aneSe2 acts as a bidentate ligand only one configuration is possible, hence one resonance is observed by 77Se-{1H} NMR spectroscopy.Despite the moderately high quadrupole moment associated with 55Mn (0.55 × 10228 m2), 55Mn NMR spectroscopy is a valuable technique for characterising Mn-containing species. In addition to d(55Mn) being sensitive to factors such as the donor set and oxidation state, we have shown that for fac-[MnX(CO)3- (L–L)] (L–L = dithio-, diseleno- or ditelluro-ether) the linewidths are typically <3000 Hz and hence individual invertomers are easily observed.14,26 Since only one invertomer is possible for bidentate co-ordinated [8]aneSe2, a single resonance is observed.The d(55Mn) values for these compounds show very good agreement with those for fac-[MnX(CO)3(diselenoether)] [diselenoether = MeSe(CH2)nSeMe (n = 2 or 3), PhSe(CH2)2- SePh or o-C6H4(SeMe)2],14 and similarly show a shift to low frequency along the series X = Cl æÆ Br æÆ I. The 55Mn NMR spectrum of the dinuclear [{MnCl(CO)3}2([16]aneSe4)] shows a single broad resonance at d 2210, indicative of the same donor set at the Mn as in the diselenoether species and hence bidentate co-ordination of the tetraselenoether macrocycle to each MnI.At 300 K the 77Se-{1H} NMR spectrum of this species shows a weak broad resonance at d 152, and cooling this solution to 220 K gives rise to several resonances with1080 J. Chem. Soc., Dalton Trans., 1999, 1077–1083 Fig. 2 View of the structure of [MnBr(CO)3([8]aneSe2)] with numbering scheme adopted.Ellipsoids are drawn at 40% probability. Fig. 3 View of the structure of [ReBr(CO)3([8]aneSe2)]. Details as in Fig. 2. Table 3 Selected bond lengths (Å) and angles (8) of [MnBr(CO)3- ([8]aneSe2)] Br(1)–Mn(1) Se(1)–C(1) Se(2)–Mn(1) Se(2)–C(4) Mn(1)–C(8) O(1)–C(7) O(3)–C(9) C(2)–C(3) C(5)–C(6) Mn(1)–Se(1)–C(1) C(1)–Se(1)–C(6) Mn(1)–Se(2)–C(4) Br(1)–Mn(1)–Se(1) Br(1)–Mn(1)–C(7) Br(1)–Mn(1)–C(9) Se(1)–Mn(1)–C(7) Se(1)–Mn(1)–C(9) Se(2)–Mn(1)–C(8) C(7)–Mn(1)–C(8) C(8)–Mn(1)–C(9) 2.534(1) 1.982(6) 2.480(1) 1.956(7) 1.814(7) 1.158(8) 1.160(8) 1.535(10) 1.512(9) 110.2(2) 97.1(3) 108.1(2) 88.47(4) 86.8(2) 172.1(2) 175.1(2) 97.8(2) 176.7(2) 90.1(3) 88.4(3) Se(1)–Mn(1) Se(1)–C(6) Se(2)–C(3) Mn(1)–C(7) Mn(1)–C(9) O(2)–C(8) C(1)–C(2) C(4)–C(5) Mn(1)–Se(1)–C(6) Mn(1)–Se(2)–C(3) C(3)–Se(2)–C(4) Br(1)–Mn(1)–Se(2) Br(1)–Mn(1)–C(8) Se(1)–Mn(1)–Se(2) Se(1)–Mn(1)–C(8) Se(2)–Mn(1)–C(7) Se(2)–Mn(1)–C(9) C(7)–Mn(1)–C(9) 2.488(1) 1.980(7) 1.974(7) 1.805(7) 1.788(7) 1.134(8) 1.517(10) 1.535(10) 108.1(2) 108.7(2) 97.5(3) 92.41(4) 86.6(2) 86.23(4) 90.6(2) 92.9(2) 92.9(2) 87.0(3) major peaks at d 72.0, 119.6, 144.5 and 147.0.The complex [ReCl(CO)3([16]aneSe4)] was not suYciently soluble to enable 13C-{1H} and 77Se-{1H} NMR spectra to be obtained. The 55Mn NMR data were also recorded for the cationic manganese(I) complexes of [16]aneSe4. Thus fac-[Mn(CO)3- (h3-[16]aneSe4)]CF3SO3 shows a single broad peak at d 2499.This compares with d 2721 (syn) and 2672 (anti) for fac- [Mn(CO)3{MeC(CH2SeMe)3}]CF3SO3 and d 2560 for fac- [Mn(CO)3{MeSe(CH2)3Se(CH2)3SeMe}]CF3SO3.25 Single crystals of [MnBr(CO)3([8]aneSe2)] and [ReBr(CO)3- ([8]aneSe2)] were obtained by addition of hexane to solutions of the compounds in CHCl3 and cooling to ca. 218 8C. The crystal structures show (Figs. 2 and 3, Tables 3 and 4) that these species are isostructural, each displaying a distorted octahedral arrangement at the metal centre comprising three mutually fac CO ligands, a bidentate [8]aneSe2 and a Br2 ligand, Mn–Se 2.488(1), 2.480(1); Re–Se 2.607(2), 2.611(2) Å.The Se(1)–M– Se(2) angles are 86.23(4) and 83.77(6)8 respectively for M = Mn and Re, and the co-ordinated diselenoether adopts a chair-boat conformation. The M–Se distances for these species compare well with those in other reported manganese(I) and rhenium(I) selenoether species, e.g.fac-[MnCl(CO)3{MeSe(CH2)nSeMe}] (n = 2 or 3) 14 and fac-[ReI(CO)3{MeSe(CH2)3SeMe}].27 Crystals of [W(CO)4([8]aneSe2)] were also obtained by cooling a CHCl3–hexane solution of the compound to 218 8C. The crystal structure of this compound shows (Fig. 4, Table 5) the [8]aneSe2 ligand occupying two mutually cis co-ordination sites with the CO ligands completing the distorted octahedral Table 4 Selected bond lengths (Å) and angles (8) of [ReBr(CO)3- ([8]aneSe2)] Re(1)–Br(1) Re(1)–Se(2) Re(1)–C(2) Se(1)–C(4) Se(2)–C(6) O(1)–C(1) O(3)–C(3) C(5)–C(6) C(8)–C(9) Br(1)–Re(1)–Se(1) Br(1)–Re(1)–C(1) Br(1)–Re(1)–C(3) Se(1)–Re(1)–C(1) Se(1)–Re(1)–C(3) Se(2)–Re(1)–C(2) C(1)–Re(1)–C(2) C(2)–Re(1)–C(3) Re(1)–Se(1)–C(9) Re(1)–Se(2)–C(6) C(6)–Se(2)–C(7) 2.633(2) 2.611(2) 1.92(2) 1.97(2) 2.00(2) 1.18(2) 1.15(2) 1.53(2) 1.55(2) 90.41(6) 175.7(5) 89.4(5) 92.9(5) 94.9(5) 91.4(5) 88.6(7) 89.9(7) 107.7(5) 107.8(5) 97.1(7) Re(1)–Se(1) Re(1)–C(1) Re(1)–C(3) Se(1)–C(9) Se(2)–C(7) O(2)–C(2) C(4)–C(5) C(7)–C(8) Br(1)–Re(1)–Se(2) Br(1)–Re(1)–C(2) Se(1)–Re(1)–Se(2) Se(1)–Re(1)–C(2) Se(2)–Re(1)–C(1) Se(2)–Re(1)–C(3) C(1)–Re(1)–C(3) Re(1)–Se(1)–C(4) C(4)–Se(1)–C(9) Re(1)–Se(2)–C(7) 2.607(2) 1.88(2) 1.92(2) 1.98(2) 1.99(2) 1.16(2) 1.50(3) 1.53(2) 86.72(6) 88.4(5) 83.77(6) 175.1(5) 96.5(5) 175.9(5) 87.5(7) 108.1(6) 98.1(7) 110.2(5) Table 5 Selected bond lengths (Å) and angles (8) of [W(CO)4- ([8]aneSe2)] W(1)–Se(1) W(1)–C(1) W(1)–C(2) Se(1)–C(4) O(1)–C(1) O(3)–C(3) C(6)–C(7) Se(1)–W(1)–Se(1) Se(1)–W(1)–C(2) Se(1)–W(1)–C(3) Se(1)–W(1)–C(2) Se(1)–W(1)–C(3) C(1)–W(1)–C(2) C(2)–W(1)–C(2) C(2)–W(1)–C(3) W(1)–Se(1)–C(7) 2.650(1) 2.08(2) 1.96(1) 1.98(2) 1.13(2) 1.12(2) 1.49(2) 82.62(6) 176.0(4) 90.8(4) 93.9(4) 90.8(4) 86.4(5) 89.5(8) 87.2(5) 111.4(4) W(1)–Se(1) W(1)–C(2) W(1)–C(3) Se(1)–C(7) O(2)–C(2) C(4)–C(5) C(6)–C(7) Se(1)–W(1)–C(1) Se(1)–W(1)–C(2) Se(1)–W(1)–C(1) Se(1)–W(1)–C(2) C(1)–W(1)–C(2) C(1)–W(1)–C(3) C(2)–W(1)–C(3) W(1)–Se(1)–C(4) C(4)–Se(1)–C(7) 2.650(1) 1.96(1) 2.07(2) 2.01(1) 1.17(1) 1.53(2) 1.49(2) 95.9(4) 93.9(4) 95.9(4) 176.0(4) 86.4(5) 171.0(7) 87.2(5) 105.3(4) 98.4(6)J.Chem. Soc., Dalton Trans., 1999, 1077–1083 1081 geometry, W–Se(1) 2.650(1), W–C(1) 2.08(2), W–C(2) 1.96(1), W–C(3) 2.07(2) Å, thus the W–CO distance trans to Se is significantly shorter than those trans to CO. Similar trends have been observed for [Mo(CO)4(PR3)2].28 Experimental Infrared spectra were measured as CsI or KBr discs using Perkin-Elmer 983 (200–4000 cm21) or Shimadzu FTIR-8300 (400–4000 cm21) spectrometers, or in solution using NaCl plates on a Perkin-Elmer 1600 FTIR spectrometer.Mass spectra were run by fast-atom bombardment (FAB) using 3-nitrobenzyl alcohol as matrix on a VG Analytical 70-250-Se normal geometry double focusing mass spectrometer or by electrospray or APCI (MeCN solution) using a Micromass Platform quadrupole mass analyser (m/z was calculated using 80Se).The 1H NMR spectra were recorded in CDCl3 at 300 MHz unless otherwise stated using a Bruker AM300 spectrometer, 13C-{1H}, 55Mn, 77Se-{1H} and 95Mo NMR spectra using a Bruker AM360 spectrometer operating at 90.1, 89.27, 68.68 or 23.4 MHz respectively and referenced to Me4Si, external saturated, aqueous K[MnO4], external neat Me2Se and external aqueous Na2[MoO4] respectively (d 0); [Cr(acac)3] was added to the NMR solutions prior to recording 13C-{1H} and 77Se-{1H} spectra and a pulse delay of 2 s was employed for the 13C-{1H} spectra to overcome the long relaxation times.Solution conductivities were obtained using ca. 1023 M solutions and a Portland Electronic conductivity meter. Microanalyses were determined by the University of Strathclyde and the University of East Anglia microanalytical laboratories. The compounds [MnX(CO)5] (X = Cl, Br or I),29 [ReX(CO)5] (X = Cl or Br),30 [Mo(CO)4(nbd)],31 [W(CO)4(TMPA)],32 [{MoBr2(CO)4}2],33 [ MI2(CO)3(NCMe)2],34 L,10 [8]aneSe2 and [16]aneSe4 1 were prepared according to literature procedures and [Cr(CO)4(nbd)] via a modified synthesis in refluxing xylene.35 Calculations The molecular mechanics calculations were carried out using CHEM-X software.18 The standard forcefield was supplemented with the following parameters to permit calculations on selenoethers: C–Se bond length I0 = 1.956 Å, k = 180; C–Se-C q0 = 1.716 rad, k = 40, cos q = 20.144, cos k = 90.All other selenium parameters were copied from the default set for sulfur.Fig. 4 View of the structure of [W(CO)4([8]aneSe2)]. Details as in Fig. 2. The conformational analysis of L was carried out as follows. The molecule was first constructed in an arbitrary conformation and energy minimised. The structure was then modified to the ring-opened form shown in Scheme 2, where Du is a dummy atom. The angles f1 and f2–f5 were varied systematically over 1808 in 37 steps and 3608 in 48 steps respectively, giving a total of 196411392 input conformations.These were filtered by skipping those exhibiting Corey–Pauling–Koltun (CPK) contacts, and by use of the following acceptance criteria; C1 ? ? ? Du 1.93– 1.98 Å, C1 ? ? ? Du–C 95.5–101.18, C–C1 ? ? ? Du 108.5–110.58. This gave 134 accepted conformations. Each of these was then reconstructed to the original compound, energy minimised, and compared to those already processed. Enantiomeric forms were treated as one conformer. This gave a final set of seven conformers (Table 2).Syntheses All reactions were carried out under an atmosphere of dry nitrogen using standard Schlenk techniques and one representative synthesis of each type is presented in detail. [Cr(CO)4([8]aneSe2)]. To a solution of [Cr(CO)4(nbd)] (0.077 g, 0.30 mmol) in chloroform (10 cm3) a solution of [8]aneSe2 (0.073 g, 0.30 mmol) in chloroform (2 cm3) was added. This dark yellow reaction mixture was refluxed overnight. Solution IR confirmed that the reaction had gone to completion, by the disappearance of bands associated with starting material.The resulting solution was filtered (Celite) and solvent volume reduced in vacuo to ca. 4 cm3 before precipitating the product by the addition of cold hexane (8 cm3). The bright yellow solid was isolated by filtration and dried under vacuum. Yield 0.072 g, 59% (Found: C, 29.2; H, 2.5. Calc. for C10H12CrO4Se2: C, 29.6; H, 3.0%). APCI mass spectrum (MeCN): found m/z = 379, 323; calculated for [Cr(CO)3([8]aneSe2)]1 m/z = 380, [Cr(CO)([8]- aneSe2)]1 m/z = 324. 1H NMR spectrum (CDCl3): d 2.55–2.90 (br, m, SeCH2, 8 H) and 2.05 (br, CH2CH2CH2, 4 H). [Mo(CO)4([8]aneSe2)]. Yield 74% (Found: C, 27.0; H, 2.9. Calc. for C10H12MoO4Se2: C, 26.7; H, 2.7%). APCI mass spectrum (MeCN): found m/z = 244; calculated for ([8]aneSe2)1 m/z = 244. 1H NMR spectrum (CDCl3): d 2.80–2.95 (br, m, SeCH2, 8 H) and 2.28 (m, CH2CH2CH2, 4 H). [W(CO)4([8]aneSe2)]. Yield 53% (Found: C, 22.0; H, 2.2.Calc. for C10H12O4Se2W: C, 22.3; H, 2.2%). FAB mass spectrum: found m/z = 538; calculated for [184W(CO)4([8]aneSe2)]1 m/z = 540. 1H NMR spectrum (CDCl3): d 2.75–2.90 (m, SeCH2, 8 H) and 2.15–2.35 (m, CH2CH2CH2, 4 H). [MnCl(CO)3([8]aneSe2)]. To a solution of [MnCl(CO)5] (0.069 g, 0.30 mmol) in chloroform (10 cm3) a solution of [8]aneSe2 (0.073 g, 0.30 mmol) in chloroform (2 cm3) was added via syringe. The reaction mixture was refluxed and monitored by solution IR until there was an absence of bands associated with starting material (ca. 4 h). The resulting orange solution was filtered (Celite) and the solvent volume reduced in vacuo to ca. 4 cm3 before inducing precipitation of the product by the addition of cold hexane (8 cm3). The orange solid was isolated by filtration and dried under vacuum. Yield 0.072 g, 58% (Found: C, 25.3; H, 3.0. Calc. for C9H12ClMnO3Se2: C, 25.9; H, Scheme 21082 J. Chem. Soc., Dalton Trans., 1999, 1077–1083 2.9%).FAB mass spectrum: found m/z = 418, 383, 334; calculated for [Mn35Cl(CO)3([8]aneSe2)]1 m/z = 418; [Mn(CO)3([8]- aneSe2)]1 m/z = 383, [Mn35Cl([8]aneSe2)]1 m/z = 334. 1H NMR spectrum (CDCl3): d 2.1–4.0 (br, m, CH2). [MnBr(CO)3([8]aneSe2)]. Orange crystals of product were obtained from a solvent mixture of chloroform and hexane at 218 8C. Yield 55% (Found: C, 23.5; H, 2.5. Calc. for C9H12- BrMnO3Se2: C, 23.4; H, 2.6%). FAB mass spectrum: found m/z = 462, 378; calculated for [Mn79Br(CO)3([8]aneSe2)]1 m/z = 462; [MnBr([8]aneSe2)]1 m/z = 378. 1H NMR spectrum (CDCl3): d 1.8–4.1 (br, m, CH2). [MnI(CO)3([8]aneSe2)]. Yield 51% (Found: C, 21.5; H, 2.5. Calc. for C9H12IMnO3Se2: C, 21.3, H, 2.4%). 1H NMR spectrum (CDCl3): d 1.8–4.1 (br, m, CH2). [ReCl(CO)3([8]aneSe2)]. To a solution of [ReCl(CO)5] (0.090 g, 0.25 mmol) in chloroform (10 cm3) a solution of [8]aneSe2 (0.061 g, 0.25 mmol) in chloroform (2 cm3) was added via a syringe. This light yellow reaction mixture was refluxed overnight.Solution IR confirmed that the reaction had gone to completion, by the disappearance of bands associated with starting material. The resulting solution was filtered (Celite) and solvent volume reduced in vacuo to ca. 4 cm3 before inducing precipitation of the product by the addition of cold hexane (8 cm3). The pale cream solid was isolated by filtration, dried under vacuum and retained for analysis. Yield 0.076 g, 54% (Found: C, 18.9; H, 2.4.Calc. for C9H12ClO3ReSe2: C, 19.7; H, 2.2%). FAB mass spectrum: found m/z = 548, 513; calculated for [187Re35Cl(CO)3([8]aneSe2)]1 m/z = 550; [Re(CO)3- ([8]aneSe2)]1 m/z = 515. 1H NMR spectrum (CDCl3): d 3.9, 1.9– 3.1 (m, CH2). [ReBr(CO)3([8]aneSe2)]. Colourless crystals of product were obtained from a solvent mixture of chloroform and hexane at 218 8C. Yield 54% (Found: C, 18.3; H, 1.9. Calc. for C9H12- BrO3ReSe2: C, 18.2; H, 2.0%). FAB mass spectrum: found m/z = 594, 566, 538, 513; calculated for [187Re79Br(CO)3([8]- aneSe2)]1 m/z = 594; [187Re79Br(CO)2([8]aneSe2)]1 m/z = 566; [187Re79Br(CO)([8]aneSe2)]1 m/z = 538, [Re(CO)3([8]aneSe2)]1 m/z = 515. 1H NMR spectrum (CDCl3): d 4.05, 2.0–3.1 (m, CH2).[{MnCl(CO)3}2([16]aneSe4)]. To a solution of [MnCl(CO)5] (0.096 g, 0.41 mmol ) in chloroform (20 cm3) a solution of [16]aneSe4 (0.100 g, 0.21 mmol ) in chloroform (3 cm3) was added. This orange mixture was refluxed overnight. Solution IR showed an absence of bands associated with starting material.The orange solution was filtered before reducing the solvent volume in vacuo to ca. 7 cm3 and inducing precipitation by the addition of cold hexane (10 cm3). This gave a bright orange solid which was isolated by filtration, recrystallised from CH2Cl2 and dried under vacuum. Yield 0.066 g, 49% (Found: C, 24.6; H, 3.2. Calc. for C18H24Cl2Mn2O6Se4?CH2Cl2: C, 24.8; H, 2.8%). Electrospray mass spectrum (MeCN): found m/z = 622, 594; calculated for [Mn(CO)3([16]aneSe4)]1 m/z = 627; [Mn- (CO)2([16]aneSe4)]1 m/z = 599. 1H NMR spectrum (CDCl3): d 0.9–3.7 (br, m, CH2). [ReCl(CO)3([16]aneSe4)]. The method employed was essentially the same as above, except [ReCl(CO)5] (0.149 g 0.41 mmol) was used. A poorly soluble, pale cream solid was produced, which was isolated by filtration and dried under vacuum. Yield 0.106 g, 47% (Found: C, 22.1; H, 2.5. Calc. for C15H24ClO3ReSe4: C, 22.8; H, 3.0%). Electrospray mass spectrum (MeCN): found m/z = 756; calculated for [187Re(CO)3([16]- aneSe4)]1 m/z = 759. 1H NMR spectrum (CDCl3): d 4.0, 2.0–3.0 (m, CH2). [Mn(CO)3([16]aneSe4)]CF3SO3. To a solution of [MnBr- (CO)5] (0.103 g, 0.38 mmol) in acetone, AgCF3SO3 (0.096 g, 0.38 mmol) was added. This mixture was refluxed in darkness for 3 h. The reaction was allowed to cool to room temperature before removing the precipitated AgBr by filtration (Celite). The conversion was assumed to be 80%. To the resulting light orange solution [16]aneSe4 (0.147 g, 0.30 mmol) was added and the mixture stirred at room temperature overnight.The bright orange product was precipitated by the addition of cold hexanes, recrystallised from CH2Cl2–EtOH, isolated by filtration and vacuum dried. Yield 0.150 g, 52% (Found: C, 24.3; H, 2.8. Calc. for C16H24F3MnO6SSe4: C, 24.9; H, 3.1%). Electrospray mass spectrum (MeCN): found m/z = 625, 595, 541; calculated for [Mn(CO)3([16]aneSe4)]1 m/z = 627; [Mn(CO)2([16]aneSe4)]1 m/z = 599; [Mn([16]aneSe4)]1 m/z = 543. 1H NMR spectrum (CDCl3): d 2.1–3.6 (br, m, CH2). [{MoBr2(CO)3}2([16]aneSe4)]. Dichloromethane (20 cm3) was added to a solid mixture of [{MoBr2(CO)4}2] (0.190 g, 0.26 mmol) and [16]aneSe4 (0.130 g, 0.27 mmol). The mixture was stirred at room temperature for 1.5 h and then filtered. The resulting pale brown solid was washed with CH2Cl2 and dried in vacuo. Yield 0.27 g, 82% (Found: C, 18.0; H, 2.0. Calc. for C18H24Br4Mo2O6Se4?CH2Cl2: C, 18.3; H, 2.1%). LM/S cm2 mol21: 39 (dmf).APCI mass spectrum (MeCN): found m/z = 487; calculated for ([16]aneSe4)1 m/z = 488. [{MoI2(CO)3}2([16]aneSe4)]. Yield 46%. (Found: C, 16.2; H, 1.9. Calc. for C9H12I2MoO3Se2: C, 16.0; H, 1.8%). LM/S cm2 mol21: 74 (dmf). APCI mass spectrum (MeCN): found m/z = 487; calculated for ([16]aneSe4)1 m/z = 488. [WI2(CO)3([16]aneSe4)]. Yield 97% (Found: C, 18.0; H, 2.3. Calc. for C15H24I2O3Se4W: C, 17.9; H, 2.4%). LM/S cm2 mol21: 21 (dmf). 1H NMR spectrum (CD2Cl2): d 3.5–1.8 (br, m).APCI mass spectrum (MeCN): found m/z = 489; calculated for ([16]aneSe4)1 m/z = 488. [MoBr2(CO)3L]. Yield 82% (Found: C, 25.1; H, 2.0. Calc. for C14H14Br2MoO3Se2?0.5CH2Cl2: C, 25.4; H, 2.2%). LM/S cm2 mol21: 8 (dmf). APCI mass spectrum (MeCN): found m/z = 306; calculated for [L]1 m/z = 306. [MoI2(CO)3L]. Yield 85% (Found: C, 22.8; H, 1.8. Calc. for C14H14I2MoO3Se2: C, 22.8; H, 1.9%). LM/S cm2 mol21: 11 (dmf). APCI mass spectrum (MeCN): found m/z = 304; calculated for [L]1 m/z = 306.[WI2(CO)3L]. Yield 90% (Found: C, 20.4; H, 1.6. Calc. for C14H14I2O3Se2W: C, 20.4; H, 1.7%). LM/S cm2 mol21: 19 (dmf). APCI mass spectrum (MeCN): found m/z = 304; calculated for [L]1 m/z = 306. Crystal structures of [MnBr(CO)3([8]aneSe2)], [ReBr(CO)3- ([8]aneSe2)] and [W(CO)4([8]aneSe2)] Details of the crystallographic data collection and refinement parameters are given in Table 6. Data collection used a Rigaku AFC7S four-circle diVractometer operating at 150 K and graphite-monochromated Mo-Ka X-radiation (l = 0.71073 Å).No significant crystal decay or movement was observed. The data were corrected for absorption using y-scans. The structures were solved by heavy atom methods36 and developed by iterative cycles of full-matrix least-squares refinement and Fourier-diVerence syntheses.37 For [MnBr(CO)3([8]aneSe2)] and [W(CO)4([8]aneSe2)] all non-H atoms were refined anisotropically, while for [ReBr(CO)3([8]aneSe2)] the dominance of the rhenium scattering prevented refinement of the C-atom anisotropic displacement parameters. The H atoms were placed in fixed, calculated positions with d(C–H) = 0.96 Å.The FlackJ. Chem. Soc., Dalton Trans., 1999, 1077–1083 1083 Table 6 Crystallographic data collection and refinement parameters Formula M Space group Crystal system a/Å b/Å c/Å U/Å3 Z Dc/g cm23 m(Mo-Ka)/cm21 Maximum and minimum transmission factors Unique obs. reflections Obs. reflections with [Io > 2s(Io)] No parameters RR 9 [MnBr(CO)3([8]aneSe2)] C9H12BrMnO3Se2 480.95 P212121 Orthorhombic 12.274(1) 12.684(1) 8.611(1) 1340.6(2) 4 2.284 92.66 1.000, 0.887 1392 1231 145 0.021 0.022 [ReBr(CO)3([8]aneSe2)] C9H12BrO3ReSe2 592.22 P212121 Orthorhombic 12.302(2) 12.726(1) 8.656(2) 1355.1(4) 4 2.903 173.21 1.000, 0.641 1409 1215 100 0.032 0.043 [W(CO)4([8]aneSe2)] C10H12O4Se2W 537.97 Pnma Orthorhombic 12.861(2) 9.978(1) 10.715(2) 1375.0(3) 4 2.599 137.05 1.000, 0.328 1421 956 88 0.035 0.040 parameter indicated the correct enantiomorph for the manganese and rhenium structures.38 The weighting scheme w21 = s2(F) gave satisfactory agreement analyses.CCDC reference number 186/1341. See http://www.rsc.org/suppdata/dt/1999/1077/ for crystallographic files in .cif format. Acknowledgements We thank the Leverhulme Trust (M. K. D.) and the BBSRC for support. References 1 R. J. Batchelor, F. W. B. Einstein, I. D. Gay, J.-H. Gu, B. D. Johnston and B.M. Pinto, J. Am. Chem. Soc., 1989, 111, 6852. 2 R. J. Batchelor, F. W. B. Einstein, I. D. Gay, J.-H. Gu, B. M. Pinto and X.-M. Zhou, Inorg. Chem., 1996, 35, 3667. 3 R. J. Batchelor, F. W. B. Einstein, I. D. Gay, J.-H. Gu, B. M. Pinto and X.-M. Zhou, J. Am. Chem. Soc., 1990, 112, 3706. 4 R. J. Batchelor, F. W. B. Einstein, I. D. Gay, J.-H. Gu, B. M. Pinto and X.-M. Zhou, J. Organomet. Chem., 1991, 411, 147. 5 W. Levason, J. J. Quirk and G. Reid, J. Chem. Soc., Dalton Trans., 1996, 3713; P.F. Kelly, W. Levason, G. Reid and D. J. Williams, J. Chem. Soc., Chem. Commun., 1993, 1716. 6 N. R. Champness, P. F. Kelly, W. Levason, G. Reid, A. M. Z. Slawin and D. J. Williams, Inorg. Chem., 1995, 34, 651; C. S. Frampton, W. Levason, J. J. Quirk and G. Reid, Inorg. Chem., 1994, 33, 3120. 7 N. R. Champness, J. J. Quirk, W. Levason, G. Reid and C. S. Frampton, Polyhedron, 1995, 14, 2753. 8 W. Levason, G. Reid and S. M. Smith, Polyhedron, 1997, 16, 4253. 9 W. Levason, J. J. Quirk, G. Reid and S. M. Smith, J. Chem. Soc., Dalton Trans., 1997, 3719. 10 D. G. Booth, W. Levason, J. J. Quirk, G. Reid and S. M. Smith, J. Chem. Soc., Dalton Trans., 1997, 3493. 11 M. K. Davies, W. Levason and G. Reid, J. Chem. Soc., Dalton Trans., 1998, 2185. 12 T. Yoshida, T. Adachi, T. Ueda, M. Watanabe, M. Kaminaka and T. Higuchi, Angew. Chem., Int. Ed. Engl., 1987, 26, 1171; T. Yoshida, T. Adachi, M. Kaminaka and T. Ueda, J. Am. Chem. Soc., 1988, 110, 4872; T.Adachi, N. Sasaki, T. Ueda, M. Kaminaka and T. Yoshida, J. Chem. Soc., Chem. Commun., 1989, 1320. 13 T. Adachi, M. C. Durrant, D. L. Hughes, C. J. Pickett, R. L. Richards, J. Talarmin and T. Yoshida, J. Chem. Soc., Chem. Commun., 1992, 1464. 14 J. Connolly, M. K. Davies and G. Reid, J. Chem. Soc., Dalton Trans., 1998, 3833. 15 P. K. Baker, M. C. Durrant, S. D. Davies, D. L. Hughes and R. L. Richards, J. Chem. Soc., Dalton Trans., 1997, 509. 16 B. Rys, H. Duddeck and M.Hiegemann, J. Heterocycl. Chem., 1992, 29, 967. 17 B. Rys, H. Duddeck and M. Hiegemann, Tetrahedron, 1991, 47, 1417. 18 CHEM-X, Chemical Design Ltd., Oxon, July 1996 version. 19 R. D. Adams and K. T. McBride, Chem. Commun., 1997, 525; G. W. Bushnell and R. H. Mitchell, Can. J. Chem., 1982, 60, 362; T. Okajima, Z. H. Wang and Y. Fukazawa, Tetrahedron Lett., 1989, 30, 1551; M. Iwaoka, H. Komatsu and S. Tomoda, Bull. Chem. Soc. Jpn., 1996, 69, 1825. 20 K. Ohno, H. Matsuura and H. Murata, J. Mol. Struct., 1980, 66, 45; J. R. Alkins and P. J. Hendra, Spectrochim. Acta, 1966, 22, 2075. 21 W. C. Still and I. Galynker, Tetrahedron, 1981, 37, 3981; Y. Fukazawa, S. Usui and Y. Uchio, Tetrahedron Lett., 1986, 27, 1825. 22 P. K. Baker, A. I. Clark, M. M. Meehan, E. E. Parker, A. E. Underhill, M. G. B. Drew, M. C. Durrant and R. L. Richards, Transition Met. Chem., 1998, 23, 155. 23 P. K. Baker, S. D. Harris, M. C. Durrant, D. L. Hughes and R. L. Richards, J. Chem. Soc., Dalton Trans., 1994, 1401. 24 R. Uson, V. Riera, J. Gimeno, M. Laguna and M. P. Gamasa, J. Chem. Soc., Dalton Trans., 1979, 996. 25 J. Connolly, W. Levason, S. D. Orchard, S. J. A. Pope and G. Reid, unpublished work. 26 J. Connolly, G. W. Goodban, G. Reid and A. M. Z. Slawin, J. Chem. Soc., Dalton Trans., 1998, 2225; W. Levason, S. D. Orchard and G. Reid, Organometallics, in the press. 27 E. W. Abel, S. K. Bhargava, M. M. Bhatti, K. Kite, M. A. Mazid, K. G. Orrell, V. Sik, B. L. Williams, M. B. Hursthouse and K. M. A. Malik, J. Chem. Soc., Dalton Trans., 1982, 2065. 28 F. A. Cotton, D. J. Darensbourg, S. Klein and B. W. S. Kilthammer, Inorg. Chem., 1982, 21, 2661. 29 K. J. Reimer and A. Shaver, Inorg. Synth., 1979, 19, 159. 30 S. P. Schmidt, W. C. Trogler and F. Basolo, Inorg. Synth., 1990, 28, 160. 31 J. J. Eisch and R. B. King, Organomet. Synth., 1965, 1, 123. 32 G. R. Dobson and G. C. Faber, Inorg. Chim. Acta, 1987, 4, 87. 33 J. A. Broomhead, J. Budge and W. Grumley, Inorg. Synth., 1990, 28, 145. 34 P. K. Baker, S. G. Fraser and E. M. Keys, J. Organomet. Chem., 1986, 309, 319. 35 H. Werner and R. Prinz, Chem. Ber., 1967, 100, 265. 36 PATTY, The DIRDIF Program System, P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, S. Garcia-Granda, R. O. Gould, J. M. M. Smits and C. Smykalla, Technical Report of the Crystallography Laboratory, University of Nijmegen, 1992. 37 TEXSAN, Crystal Structure Analysis Package, Molecular Structure Corporation, Houston, TX, 1995. 38 H. D. Flack, Acta Crystallogr., Sect. A, 1983, 39, 876. Paper 8/09853K

ISSN:1477-9226    

DOI:10.1039/a809853k

出版商:RSC    

年代:1999

数据来源: RSC