DALTON FULL PAPER J. Chem. Soc. Dalton Trans. 1999 2111–2118 2111 Evidence for cationic Group 4 zirconocene complexes with intramolecular phenyl co-ordination Linda H. Doerrer Malcolm L. H. Green Daniel Häußinger and Jörg Saßmannshausen Inorganic Chemistry Laboratory South Parks Road Oxford UK OX1 3QR Received 21st October 1998 Accepted 11th May 1999 The mono- and bis-ring substituted zirconocenes with pendant phenyl groups [Zr(h-C5H5)(h-C5H4CMe2Ph)Me2] 2 [Zr(h-C5H5)(h-C5H4CMe2C6H4Me-p)Me2] 3 [Zr(h-C5H4CMe2Ph)2Me2] 4 and [Zr(h-C5H4CMe2C6H4Me-p)2Me2] 5 have been prepared. The crystal structures of 3 and 4 have been determined. Compounds 2–5 react with methyl abstracting reagents such as B(C6F5)3 or [Ph3C]1[B(C6F5)4]2 to form cationic zirconocene complexes 6–9 as solvent separated ion pairs as shown by low temperature NMR spectroscopy.For the cationic complexes [Zr(h-C5H5)(h- C5H4CMe2Ph)Me]1[RB(C6F5)3]2 (R = Me 6a or C6F5 6b) and [Zr(h-C5H5)(h-C5H4CMe2C6H4Me-p)Me]1[RB(C6F5)3]2 (R = Me 7a or C6F5 7b) evidence for the co-ordination of a phenyl group to the zirconium centre via agostic C–H–M interaction was obtained by NMR spectroscopy. These cationic complexes can be considered as models for solvent adducts in Kaminsky catalysts. The cationic complexes [Zr(h-C5H4CMe2Ph)2Me]1[RB(C6F5)3]2 (R = Me 8a or C6F5 8b) (derived from 4) and [Zr(h-C5H4CMe2C6H4Me-p)2Me]1[RB(C6F5)3]2 (R = Me 9a or C6F5 9b) (derived from 5) respectively exhibit more complex behaviour. These observations contrast with those from the previously published benzyl congener [Zr(h-C5H4CH2Ph)2Me2] 1 which with methyl abstracting agent generates both a solvent separated cation/anion pair and a tight ion pair.Introduction There is considerable evidence that the active centres in homogeneous Kaminsky catalyst systems are cations,1–4 which may be represented by the general formula [Zr(h-C5HnR5 2 n)2R9]1 where R may be a hydrocarbyl group n = 0–4 and R9 is an alkyl group. These cations can be generated by methyl abstraction from the corresponding methyl compounds [Zr(h-C5HnR5 2 n)2- Me2] using the methyl abstracting reagent [Ph3C]1[B(C6F5)4]25,6 or the Lewis acid B(C6F5)3.7,8 The cationic Group 4 metallocene species [Zr(h-C5HnR5 2 n)2Me]1 is a strong Lewis acid and can form Lewis acid–base binuclear adducts with the neutral precursor 9,10 when activated with [Ph3C]1[B(C6F5)4]2 or with the anion 7,8 in the case of B(C6F5)3 (see Scheme 1).Since most polymerisations using the cations [Zr(h-C5Hn- R5 2 n)2R9]1 are conducted in toluene or other arenes as solvents it is possible that adducts of the general formula [Zr(h-C5Hn- R5 2 n)2R9(solv)]1 (solv = solvent) could serve as a resting state of the catalyst cycle (Scheme 1). These solvent adducts have been suggested before 1 but there is little direct evidence for such species. Related solvent adducts have been observed for ruthenium complexes 11 and for Group 4 half-sandwich complexes for example [MCp*Me2(arene)]1 (Cp* = C5Me5; Scheme 1 Cp¢ Zr Cp¢ CH3 Cp¢ Zr Cp¢ CH3 CH3 Cp¢ Zr Cp¢ CH3 C B(C6F5)3 H H H Cp' Zr Cp' CH3 Cp¢ Zr Cp¢ CH3 Cp¢ Zr Cp¢ CH3 Cp¢ Cp' Zr Me Cp¢ Zr Cp¢ CH3 solv Cp¢ Zr Cp' CH3 solv n –Cp'2ZrMe2 –MeB(C6F5)3 – + solv –solv –solv + solv ? ? Cp¢2ZrMe2 + 1/2 Ph3C+ + B(C6F5)3 C2H4 C2H4 C2H4 arene = benzene toluene m- and p-xylene anisole styrene or mesitylene; M = Ti Zr or Hf)12,13 and [M{h-C5H3(SiMe3)2}(h- C6H5Me)Me2]1 (M = Zr or Hf).14 We previously reported 15 the synthesis and reactions of [Zr(h-C5H4CH2Ph)2Me2] 1 in which benzyl groups are tethered to the cyclopentadienyl ring.It was hoped that evidence for interaction between the phenyl group and the zirconium centre in the cation [Zr(h-C5H4CH2Ph)2Me]1 would be observed. However no direct evidence was forthcoming. Therefore we have prepared the new monosubstituted zirconocenes [Zr(h-C5H5)(h-C5H4CMe2Ph)Me2] 2 and [Zr(h- C5H5)(h-C5H4CMe2C6H4Me-p)Me2] 3 and the bis-substituted zirconocenes [Zr(h-C5H4CMe2Ph)2Me2] 4 and [Zr(h-C5H4- CMe2C6H4Me-p)2Me2] 5 in the hope that the cations derived from these compounds using methyl abstracting reagents might show evidence for phenyl–zirconium interactions.Results and discussion Preparation of the metallocenes 2–5 The compound [Zr(h-C5H5)(h-C5H4CMe2Ph)Cl2] 2a was prepared by treating LiPh with 6,6-dimethylfulvene and subsequently quenching the reaction mixture with [Zr(h-C5H5)Cl3]? dme. A mixture of desired 2a and [Zr(h-C5H4CMe2Ph)2Cl2] 4a was obtained and these two compounds were separated by fractional crystallisation. Methylation of 2a was performed with MgMeBr in order to reduce any ligand scrambling and methylation of 4a was performed with LiMe (see Scheme 2). Crystals of 4 suitable for X-ray analysis were grown by slow cooling of a light petroleum solution to 280 8C.The crystal structure of 4 is shown in Fig. 1 and selected bond angles and distances are summarised in Table 1. The phenyl group is bent away from the zirconium and there is no evidence for intermolecular interactions. The compound [Zr(h-C5H5)(h-C5H4CMe2C6H4Me-p)Cl2] 3a was prepared by treating LiC6H4Me-p with 6,6-dimethylfulvene and quenching the resulting lithium salt with [Zr(h-C5H5)- Cl3]?dme. In contrast to the synthesis of 2a no ligand scrambling was observed and 3a was obtained in good yields. 2112 J. Chem. Soc. Dalton Trans. 1999 2111–2118 Methylation of 3a was performed with MgMeBr in diethyl ether to aVord [Zr(h-C5H5)(h-C5H4CMe2C6H4Me-p)Me2] 3 in high yield (see Scheme 3). Crystals of 3 suitable for X-ray analysis were grown by slow cooling of a light petroleum solution to 220 8C.The crystal structure of 3 is shown in Fig. 2 and selected bond angles and distances are summarised in Table 2. Unlike the solid state structure of 1 the phenyl groups in 3 adopt an anti conformation presumably to avoid repulsive interactions between the CMe2 and the ZrMe2 groups in a syn conformation. This is typical for substituted Group 4 metallocenes. 16 No close intermolecular contacts of phenyl groups to the zirconium centre were found. The compound [Zr(h-C5H4CMe2C6H4Me-p)2Cl2] 5a was prepared by the reaction between LiC6H4Me-p and 6,6- Fig. 1 Crystal structure of complex 4. Scheme 2 Zr CMe2Ph Me Me Zr CMe2Ph Cl Cl CMe2Ph Zr CMe2Ph CMe2Ph Cl Cl + LiPh + 2a 2 4a 4 2 LiMe + 2 MeMgBr Li+ CpZrCl3•dme Table 1 Selected bond distances [Å] and angles [8] for complex 4 Zr(1)–C(1) Zr(1)–C(2) Zr(1)–C(3) Cpcentr–Zr C(21)–Zr(1)–C(21B) Cpcentr–Zr(1)–Cpcentr C(5)–C(6)–C(11) 2.5167(19) 2.4727(19) 2.5230(19) 1.866 92.3(1) 131.12 108.80(15) Zr(1)–C(4) Zr(1)–C(5) Zr(1)–C(21) C(7)–C(6)–C(8) C(7)–C(6)–C(11) 2.5862(19) 2.6127(18) 2.2854(19) 109.32(17) 107.80(16) dimethylfulvene and subsequent quenching of the resulting lithium salt with 0.5 equivalent of ZrCl4?2thf.Methylation with LiMe gave 5 in good yield (see Scheme 3). Low temperature NMR spectroscopy reaction studies All spectroscopic data for these studies are given in Table 3 together with assignments where possible. The reaction of complex 3 with B(C6F5)3 in CD2Cl2 was investigated in detail by 2-D NMR spectroscopy (1H–13C-GHSQC (gradient selected Fig. 2 Crystal structure of complex 3. Scheme 3 CMe2C6H4Me Zr CMe2C6H4Me Me Me Zr CMe2C6H4Me Cl Cl Zr CMe2C6H4Me Me Me Zr CMe2C6H4Me Cl Cl CMe2C6H4Me CMe2C6H4Me Li+ + LiC6H4Me CpZrCl3•dme + 2 MeMgBr 3a 3 + 2 LiMe 1/2 ZrCl4 2 thf 5a 5 Table 2 Selected bond distances [Å] and angles [8] for complex 3 Zr(1)–C(1) Zr(1)–C(2) Zr(1)–C(3) Zr(1)–C(4) Zr(1)–C(5) Zr(1)–C(6) Cp9centr–Zr C(21)–Zr(1)–C(22) Cpcentr–Zr(1)–Cpcentr C(10)–C(11)–C(14) 2.516(8) 2.516(8) 2.511(8) 2.556(11) 2.551(11) 2.526(7) 1.866 93.6(2) 131.87 106.5(6) Zr(1)–C(7) Zr(1)–C(8) Zr(1)–C(9) Zr(1)–C(10) Zr(1)–C(21) Zr(1)–C(22) Cpcentr–Zr C(12)–C(11)–C(13) C(12)–C(11)–C(14) 2.498(8) 2.53(1) 2.561(9) 2.596(8) 2.291(6) 2.289(5) 1.859 107.5(6) 109.1(6) J.Chem. Soc. Dalton Trans. 1999 2111–2118 2113 Heteronuclear Single Quantum Coherence) with and without GARP (Globally optimised Alternating-phase Rectangular Pulses)-13C decoupling 1H–13C-GHMBC (Gradient selected Heteronuclear Multiple Bond Correlation) 1H–1H NOESY 1H– 1H EXSY) which allowed complete assignment of the resulting cation.At 260 8C the reaction proceeds cleanly to give the solvent separated species [Zr(h-C5H5)(h-C5H4CMe2C6H4Me-p)- Me]1[MeB(C6F5)3]2 7a. A solvent separated species is clearly indicated by the broad singlet at d 0.40 in the 1H NMR spectrum which is assigned to the free anion [MeB(C6F5)3]2. In agreement with this assignment the 13C NMR spectrum shows a broad peak at d 9.1 typical for non-co-ordinated [MeB(C6F5)3]2.12 Further the chemical shift diVerence between the m- and p-19F of the anion Dd is 2.8 ppm corresponding to a solvent separated ion pair.17 The cation of complex 7a must be chiral because in the 1H NMR spectrum eight distinct aromatic signals are observed for the C5H4CMe2C6H4Me-p ligand four resonances for the substituted cyclopentadienyl ring and four signals for the phenyl ring.Similarly the two methyl groups on the bridging carbon appear as two singlets. Accordingly 15 signals were observed for the C5H4CMe2C6H4Me-p ligand in the 13C-{1H} NMR spectrum. The proton and carbon NMR spectra suggest an interaction of the phenyl group with the zirconium centre. The 1H NMR signal for the proton labelled Ph1 (cf. Table 3) is shifted 0.95 ppm upfield to d 6.18 whereas the signals for Ph2 Ph3 and Ph4 are slightly shifted downfield compared to those of the starting material 3. A similar upfield shift is observed for Ph1 in the 13C NMR spectrum (d 114.8).In addition the 1JCH coupling constant for Ph1 (148 Hz) is significantly smaller compared to the coupling constants for Ph2 (169 Hz) Ph3 (174 Hz) and Ph4 (163 Hz). A reduced 1JCH coupling constant is a strong indication for an agostic C–H–M interaction.18 Furthermore one of the cyclopentadienyl ring protons (Cp4) is observed at d 6.93 which can be attributed to the magnetic anisotropy of the phenyl group as the NOESY spectrum indicates close proximity of Cp4 to the phenyl ring. Two possible structures of the cation A and B (see Fig. 3) Fig. 3 (Top) proposed structures A and B for interaction of an electrophilic zirconium centre with a phenyl ring. (Bottom) proposed structure of compounds 7. one with the phenyl ring co-ordinated side on the other with agostic co-ordination via one of the hydrogens have been considered.However only B is supported by NMR data as A does not explain the upfield shift of Ph1. The NOESY spectrum as well clearly favours structure B as the protons Cp4 and Zr–CH3 mainly interact with the protons Ph1 and Ph2 and only to a much smaller extent with Ph3 and Ph4. This small interaction is explained by the EXSY spectrum (mixing time 650 ms) that gives unambiguous proof for the following site exchanges Ph1 with Ph3 Ph2 with Ph4 Cp1 with Cp2 Cp3 with Cp4 and Me1 with Me2 . Owing to these exchanges small cross peaks can be detected in the NOESY spectrum between Cp4 and Ph3 Ph4 as well as Zr–CH3 and Ph3 Ph4. The NOESY/EXSY spectra further show that at 260 8C a slow co-ordination/deco-ordination process takes place in the course of which the entire h-C5H4CMe2C6H4- Me-p moiety is free to rotate around the Zr–C5H4R (centroid) axis and the chiral Zr undergoes racemisation.The reaction of complex 3 with [Ph3C]1[B(C6F5)4]2 does not lead to the analogue of the well characterised and previously observed 15 homodinuclear species [{Zr(h-C5H4CH2Ph)2Me}2- (m-Me)]1[B(C6F5)4]2 but to the formation of [Zr(h-C5H5)(h- C5H4CMe2C6H4Me-p)Me]1[B(C6F5)4]2 7b which contains the same solvent separated cation as observed in the reaction of 3 with B(C6F5)3 i.e. 7a. This behaviour with [Ph3C]1[B(C6F5)4]2 is quite unusual and has not been observed before for zirconocene complexes. It suggests that the formation of 7b is favoured over the formation of a homodinuclear species. The reaction of complex 2 with either B(C6F5)3 or [Ph3C]1[B(C6F5)4]2 at 260 8C in CD2Cl2 leads to the solvent separated complexes [Zr(h-C5H5)(h-C5H4CMe2Ph)Me]1[RB- (C6F5)3]2 (R = Me 6a or C6F5 6b) with almost identical NMR data for the cation (see Scheme 4).Although in depth 2-D NMR investigations were not performed on 6a and 6b the 1-D and 2-D NMR spectra recorded demonstrate the chiral nature of the cation in these two salts and the pronounced upfield shifts for Ph1 strongly suggest an analogous agostic C–H–M interaction as observed with 7a. Scheme 4 Zr CMe2 Me Me Zr Me CMe2 H Zr CMe2 Me Me Me Zr Me CMe2 Me H Zr CMe2C6H4Me Me Me CMe2C6H4Me Zr CMe2C6H5 Me Me CMe2C6H5 B(C6F5)3 1 2 3 4 2 1 3 4 5 2 6a 6b 3 7a 7b RB(C6F5)3 1 2 3 4 2 1 3 4 RB(C6F5)3 5 4 8a or 8b 9a or 9b or Ph3C+ B(C6F5)3 or Ph3C+ B(C6F5)3 or Ph3C+ B(C6F5)3 or Ph3C+ 2114 J. Chem.Soc. Dalton Trans. 1999 2111–2118 Table 3 NMR Data (d J/Hz) of cationic complexes a Compound 1H NMRb,c 13C NMRb,d Zr Me H C Me Me MeB(C6F5)3 – 6a 1 2 3 4 5 0.43 (br 3 H) 0.68 (s 3 H) 1.68 (s 3 H) 1.74 (s 3 H) 5.84 (s 5 H) 5.90 (q 1 H) 6.16 (q 1 H) ª6.25 (br 1 H) 6.41 (q 1 H) 6.98 (q 1 H) ª7.31 (br 1 H) 7.71 (t 1 H) ª7.84 (br 1 H) ª8.05 (br 1 H) CH3B ZrMe CCH3 CCH3 Cp Cp9 Cp9 Ph H1 Cp9 Cp9 Ph H3 Ph H5 Ph H2 Ph H4 ª10 24.9 29.7 39.3 47.0 103.0 112.7 114.1 115.3 115.9 118.6 128.4 128.8 139.3 140.9 CH3B CCH3 CCH3 CMe2 ZrMe Cp9 Cp9 Ph C1 Cp9 Cp Cp9 Ph C3 Ph C5 Ph C2 Ph C4 Zr Me H C Me Me B(C6F5)4 – 6b 1 2 3 4 5 0.69 (s 3 H) 1.70 (s 3 H) 1.76 (s 3 H) 5.84 (s 5 H) 5.89 (s 1 H) 6.15 (s 1 H) 6.26 (br 1 H) 6.42 (s 1 H) 6.99 (s 1 H) ª7.32 (br 1 H) e 7.71 (t 1 H) ª7.84 (br 1 H) e 8.05 (br 1 H) ZrMe CCH3 CCH3 Cp Cp9 Cp9 Ph H1 Cp9 Cp9 Ph H3 Ph H5 Ph H2 Ph H4 24.1 28.9 38.6 46.2 102.1 111.9 114.1 114.5 115.1 117.8 128.2 128.8 139.4 140.9 CCH3 CCH3 CMe2 ZrMe Cp9 Cp9 Ph C1 Cp9 Cp Cp9 Ph C3 Ph C5 Ph C2 Ph C4 Zr Me H C Me Me Me MeB(C6F5)3 – 7a 1 2 3 4 1 2 3 1 2 4 0.40 (s br fwhs = 8.6 3 H) 0.64 (s 3 H) 1.62 (s 3 H) 1.69 (s 3 H) 2.52 (s 3 H) 5.80 (s 5 H) 5.87 (m 1 H) 6.12 (m 1 H) 6.18 (d 1 H 3JHH = 6.7) 6.40 (m 1 H) 6.93 (m 1 H) 7.30 (d 1 H 3JHH = 7.7) 7.59 (d 1 H 3JHH = 6.7) 7.79 (d 1 H 3JHH = 7.7) CH3B ZrMe CC2H3 CC1H3 PhCH3 Cp Cp H3 Cp H1 Ph H1 Cp H2 Cp H4 Ph H3 Ph H2 Ph H4 9.1 (1JCH = 116) 20.2 (1JCH = 127) 23.7 (1JCH = 128) 28.8 (1JCH = 128) 38.1 45.0 (1JCH = 121) 102.2 (1JCH = 177) 112.4 (1JCH = 179) 114.5 (1JCH = 176) 114.8 (1JCH = 148) 115.2 (1JCH = 179) 118.3 (1JCH = 181) 129.8 (1JCH = 174) 130.0 130.5 139.9 (1JCH = 169) 140.8 141.7 (1JCH = 163) CH3B PhCH3 CC1H3 CC2H3 CMe2 ZrMe Cp C3 Cp C1 Cp C4 Ph C1 Cp Cp C2 Ph C3 i-C of Cp i-C of Ph Ph C2 i-C of PhCH3 Ph C4 Zr Me H C Me Me Me B(C6F5)4 – 7b 1 2 3 4 3 1 2 4 1 2 0.68 (s 3 H) 1.66 (s 3 H) 1.73 (s 3 H) 2.55 (s 3 H) 5.81 (s 5 H) 5.87 (d 1 H) 6.11 (d 1 H) 6.20 (d 1 H 3JHH = 6.5) 6.40 (d 1 H) 6.94 (d 1 H) 7.31 (d 1 H 3JHH = 7.8) 7.59 (d 1 H 3JHH = 6.5) 7.80 (d 1 H 3JHH = 7.8) ZrMe CC2H3 CC1H3 PhCH3 Cp Cp H3 Cp H1 Ph H1 Cp H2 Cp H4 Ph H3 Ph H2 Ph H4 20.4 24.0 29.0 38.2 45.1 101.8 111.9 114.0 114.4 114.8 117.8 129.1 129.4 130.0 139.3 140.1 141.0 PhCH3 CC1H3 CC2H3 CMe2 ZrMe Cp C3 Cp C1 Cp C4 Ph C1 Cp Cp C2 Ph C3 i-C of Cp i-C of Ph Ph C2 i-C of PhCH3 Ph C4 Zr Me CMe2 CMe2Ph MeB(C6F5)3 – 8a 0.44 (br 3 H) 0.71 (s 3 H) 1.53 (br 12 H) 5.01 (br 1 H) 5.66 (br 1 H) 5.83 (br 1 H) 7.06 (br 1 H) 6.95 (d 4 H 3JHH = 7.0) 7.45 (t 2 H 3JHH = 7.5) 7.60 (m 4 H) CH3B ZrMe CMe2 Cp9 Cp9 Cp9 Cp9 o-H of Ph p-H of Ph m-H of Ph 9.7 28.2 (br) 39.0 47.0 106.6 (br) 110.5 (br) 117.2 (br) 119.4 (br) 123.2 (br) 127.9 (br) 134.4 (br) CH3B C(CH3)2 CMe2 ZrMe Cp9 Cp9 Cp9 Cp9 o-C of Ph p-C of Ph m-C of Ph J.Chem. Soc. Dalton Trans. 1999 2111–2118 2115 Table 3 (cont’d ) Compound 1H NMRb,c 13C NMRb,d Zr Me CMe2 CMe2Ph B(C6F5)4 – 8b 0.73 (s 3 H) 1.56 (br 12 H) 4.99 (br 2 H) 5.65 (br 2 H) 5.83 (br 2 H) 7.08 e 6.98 (d 4 H JHH = 7.0) 7.46 (t 2 H 3JHH = 7.5) 7.61 (m 4 H) ZrMe CMe2 Cp9 Cp9 Cp9 Cp9 o-H of Ph p-H of Ph m-H of Ph 28.3 (br) 39.0 47.0 106.6 (br) 110.5 (br) 117.2 (br) 119.4 (br) 123.2 (br) 127.9 (br) 134.4 (br) C(CH3)2 CMe2 ZrMe Cp9 Cp9 Cp9 Cp9 o-C of Ph p-C of Ph m-C of Ph Zr Me CMe2 CMe2C6H4Me Me MeB(C6F5)3 – 9a 0.43 (br 3 H) 0.73 (s 3 H) 1.51 (br 12 H) 2.38 (s 6 H) 4.84 (br 2 H) 5.62 (br 2 H) 5.76 (br 2 H) 7.04 (br 2 H) 6.89 (m 4 H) 7.37 (m 4 H) CH3B ZrMe CMe2 PhCH3 Cp9 Cp9 Cp9 Cp9 Ph H Ph H 9.6 20.4 28.4 38.6 45.9 106.4 (br) 110.4 (br) 117.0 (br) 118.7 (br) 123.3 (br) 135.0 (br) CH3B PhCH3 C(CH3)2 CMe2 ZrMe Cp9 Cp9 Cp9 Cp9 Ph C Ph C Zr Me CMe2 CMe2C6H4Me Me B(C6F5)4 – 9b 0.75 (s 3 H) 1.53 (br 12 H) 4.92 (br 2 H) 5.64 (br 2 H) 5.79 (br 2 H) ª7.04 e 6.91 (m 4 H) 7.40 (m 4 H) ZrMe CMe2 Cp9 Cp9 Cp9 Cp9 Ph H Ph H 20.4 28.4 (br) 38.7 45.9 106.5 (br) 110.5 (br) 117.0 (br) 118.8 (br) 123.3 (br) 134.7 e PhCH3 C(CH3)2 CMe2 ZrMe Cp9 Cp9 Cp9 Cp9 Ph C Ph C a The chemical shifts for the MeB(C6F5)3 2 anion are virtually the same for all compounds and are as follows 19F NMR d 2133.6 (d 6 F o-F); 2164.0 (t 3 F p-F); 2166.8 (t 6 F o-F); 11B NMR d 215.2.b 500 MHz. c Cp Hn (n = 1–4) denotes hydrogens of the C5 ring coupled to Cp Cn (connectivity determined by C–H correlation). d 125.7 MHz. e Obscured by triphenylethane. The complexes 6a 6b and 7a 7b are stable in CD2Cl2 up to 230 8C but significant broadening of the resonances is observed. Recooling the sample to 260 8C restores the original spectrum. This can be explained by the exchange processes observed in the EXSY spectrum of 7a as discussed above. The reactions of complexes 4 and 5 with B(C6F5)3 and [Ph3C]1[B(C6F5)4]2 at 260 8C in CD2Cl2 lead to the formation of the solvent separated species [Zr(h-C5H4CMe2Ph)2- Me]1[RB(C6F5)3]2 (R = Me 8a or C6F5 8b) and [Zr(h-C5H4- CMe2C6H4Me-p)2Me]1[RB(C6F5)3]2 (R = Me 9a or C6F5 9b).However the resulting spectra are more complex than for the cases described earlier. The reaction of 5 produces a simpler spectrum compared to that of 4 and therefore will be discussed in detail. At 260 8C the signals assigned for the cyclopentadienyl ring are unusually broad compared to those for complex 6 or 7. However the signals for the phenyl ring appear as two sharp doublets. Cooling the sample to 280 8C leads to a sharpening of three of the cyclopentadienyl ring signals whilst one remains broad and one of the phenyl ring signals broadens. Spin saturation experiments at this temperature reveal that the sharp signal at d 4.84 exchanges with the broad signal at d 5.62. Further cooling to 2120 8C (in CDCl2F) leads to further broadening of all of the peaks.Heating the sample leads to a coalescence of two of the cyclopentadienyl signals (d 4.92 and 5.64) at 240 8C. The process is reversible and recooling the sample to 260 8C restores the original spectrum. This observation together with the observation of two coupled doublets for the phenyl ring (selective irradiation of one leads to the formation of a singlet in the other) indicates that more than one exchange process with diVerent reaction rates is occurring. In the absence of a fluxional process four resonances for the co-ordinated and two resonances for the freely rotating phenyl ring would be expected. At 260 8C only one set of two signals is observed however therefore the two phenyl rings must be in rapid exchange with each other on the NMR timescale.This process could occur by either rapid exchange of the phenyl rings at one co-ordination site in a screen wiper type fashion (C) or by rapid exchange of the Zr–Me group between two sites (D Scheme 5). Indeed the 1H NMR spectrum at 2120 8C indicates two diVerent phenyl rings and hence two diVerent cyclopentadienyl rings but the signal assigned for the [MeB(C6F5)3]2 anion remains unchanged. Unfortunately we were not able to lower the temperature further and freeze out this process. A second process is the exchange of the co-ordinated proton within the same ring similar to the process observed for complexes 6 and 7 which is in competition with the co-ordination of the other ring. The reaction of complex 4 or 5 with [Ph3C]1[B(C6F5)4]2 is interesting.Unlike that of 1 the formation of a homobinuclear complex could not be observed. Since the only diVerence between 1 and 4 is the substitution of the benzylic hydrogens in Scheme 5 PhMe2C CMe2 Ph Zr Me PhMe2C CMe2 Ph Zr Me PhMe2C CMe2 Ph Zr Me CMe2Ph Me2C Ph Zr Me Me2C CMe2 Ph Zr Ph Me C D 2116 J. Chem. Soc. Dalton Trans. 1999 2111–2118 1 with methyl groups this subtle change had a significant change in the chemistry. Several reasons for this behaviour can clearly be identified (a) the steric bulk of 4 is greater than that of 1 as indicated by the solid state structures; (b) substitution of the benzylic hydrogens with methyl groups on the bridging carbon enhances ring closure; (c) the methyl groups have a 1I eVect therefore the phenyl ring is more electron rich than in 1.The factors (a)–(c) in addition to the ansa eVect enhance the co-ordination of the phenyl group to the cationic metal centre unlike in 1 where only the ansa eVect is present. The introduction of a substituent in the para position of the phenyl ring has little if any influence on the chemical behaviour. However due to a simplified spin system the NMR spectra are easier to interpret. Overall the NMR data show the formation of the solvent separated species 8a 8b and 9a 9b similar to 6a 6b and 7a 7b with the possible co-ordination of the phenyl rings but unambiguous evidence for this could not be found in the collected data. Conclusion Monosubstituted zirconocene complexes with pendant phenyl rings 2 and 3 and bis-substituted zirconocene complexes 4 and 5 have been prepared and the solid state structure of one of each group has been determined.NMR Studies of the reaction of the zirconocenes 2–5 with either B(C6F5)3 or [Ph3C]1[B(C6F5)4]2 revealed the formation of discrete anions and cations 6–9. These results are in marked contrast to that observed with [Zr(h-C5H4R)2Me2] (R = H Me SiMe3 or Si(SiMe3)3 15) which do not form discrete ion pairs. In the case of R = Si(SiMe3)3 the substituent is sterically more demanding than the CMe2Ph group and should favour the formation of discrete anions and cations taking only steric eVects into account. The formation of the discrete ion pairs can be rationalised by the ability of the phenyl group to saturate the otherwise co-ordinatively unsaturated zirconium centre therefore electronic eVects are dominating in these complexes.In the case of complexes 6 and 7 the co-ordination of the phenyl ring via agostic interaction could be derived from NMR spectra. The picture for 8 and 9 is more complicated however because several competing dynamic processes occur preventing unambiguous assignment by NMR. The NMR studies of 6 and 7 show that the co-ordination of aromatic solvents such as toluene under standard polymerisation conditions is possible and likely. In addition our investigations of 8 and 9 demonstrate the labile nature of the arene; rapid exchange between co-ordinated and unco-ordinated arene occurs even at 260 8C. With respect to olefin polymerisation catalysts the arene co-ordination is labile enough to be displaced by an olefin monomer. The labile nature of the arene adduct might explain the diYculties in observing these proposed adducts.Further studies on 4 and 5 are currently being undertaken in this laboratory. Experimental All experiments were carried out under a nitrogen atmosphere by using standard Schlenk techniques. Solvents were dried over sodium (toluene low in sulfur) sodium–potassium alloy (diethyl ether; light petroleum bp 40–60 8C) sodium– benzophenone (thf) and calcium hydride (dichloromethane). NMR Solvents were dried over activated molecular sieves freeze thawed and stored in Young’s-tap sealed ampoules. NMR Spectra were recorded at a Bruker AM300 or a Varian UnityPlus 500 spectrometer and referenced to the residual protio solvent peak for 1H. Chemical shifts are quoted in ppm relative to tetramethylsilane. The 13C spectra were referenced with the solvent peak relative to TMS and were proton decoupled using a WALTZ sequence.CH Coupling constants were measured by coupled Pulsed Field Gradient-Heteronuclear Single Quantum Coherence (PFG-HSQC). 19F NMR Spectra were referenced with external C6F6 (d 163.0) and 11B NMR spectra with BF3?Et2O (d 0). Phase sensitive NOESY/ EXSY spectra were performed using a standard Time Proportional Phase Increment (TPPI) pulse sequence and a mixing time of 650 ms at 260 8C. Mass spectra were determined by the EPSRC National Mass Spectrometry Service Centre by Dr. J. A. Ballantine. The compounds [Zr(h-C5H5)Cl3]?dme19 and 6,6-dimethylfulvene 20 were prepared as described. The zirconocene dichlorides 2a–5a were characterised by NMR spectroscopy and methylated without further purification. Preparations [Zr(Á-C5H5)(Á-C5H4CMe2Ph)Cl2] 2a and [Zr(Á-C5H4CMe2- Ph)2Cl2] 4a.Iodobenzene (6.12 g 30 mmol) was added to a solution of 12 ml (30 mmol) n-butyllithium (2.5 mol l21) in 200 ml light petroleum at room temperature. A white solid precipitated and the reaction mixture was stirred for 20 min. The mixture was cooled to 0 8C and filtered. The residue was dissolved in 150 ml diethyl ether and cooled to 278 8C. Neat 6,6- dimethylfulvene (3.19 g 30 mmol) was added and the reaction mixture allowed to warm to room temperature yielding a white suspension which was stirred overnight. Tetrahydrofuran (50 ml) was added to dissolve the precipitate and the clear solution recooled to 278 8C. The compound [Zr(h-C5H5)Cl3]?dme (10.6 g 30 mmol) was added in small portions and the slurry stirred for 1 h at this temperature before being warmed to room temperature.The reaction mixture was stirred overnight to yield a yellow-orange suspension. The volatiles were removed under reduced pressure to yield a yellow solid which was extracted several times into warm (50 8C) toluene. The extract was stored at 230 8C for 4 d to yield a white solid. Yield of 2a 2.46 g 5.9 mmol (20%). The mother-liquor was concentrated and recooled to 230 8C. A second crop could be obtained which was a mixture of complexes 2a and 4a (ª2:1 by 1H NMR). Complex 2a 1H NMR (CDCl3 300 MHz 20 8C) d 1.76 (s 6 H CMe2); 6.27 (s 5 H Cp); 6.37 (“t” 2 H Cp9); 6.48 (“t” 2 H Cp9) and 7.2–7.3 (m 5 H Ph); 13C NMR (CDCl3 125.7 MHz 20 8C) d 30.5 (CCH3); 41.3 (CCH3); 117.0 (Cp9); 117.1 (Cp); 118.2 (Cp9); 127.1 (p-C of Ph); 127.3 (m-C of Ph); 129.2 (o-C of Ph); 143.4 (Cq); and 150.8 (Cq).Complex 4a 1H NMR (CDCl3 300 MHz 20 8C) d 1.78 (s 12 H CMe2); 6.02 (“t” 4 H Cp9); 6.36 (“t” 4 H Cp9) and 7.2–7.3 (m 10 H Ph); 13C NMR (CDCl3 125.7 MHz 20 8C) d 29.3 (CCH3); 40.4 (CCH3); 113.0 (Cp9); 117.2 (Cp9); 126.1 (p-C of Ph); 126.1 (m-C of Ph); 128.1 (o-C of Ph); 142.0 (Cq); and 149.8 (Cq). [Zr(Á-C5H5)(Á-C5H4CMe2C6H4Me-p)Cl2] 3a. The compound n-butyllithium (12 ml 30 mmol 2.5 mol l21 in hexanes) was added to a solution of 5.13 g (30 mmol) of 4-bromotoluene in 200 ml of diethyl ether at room temperature. The reaction mixture was stirred at 30 8C for 1 h then cooled to 278 8C and 3.19 g (30 mmol) of pure 6,6-dimethylfulvene were slowly added dropwise. The resulting yellow solution was slowly warmed to room temperature.An oV-white precipitation occurred which was dissolved by addition of 30 ml of thf. The slightly yellow solution was recooled to 278 8C and 10.6 g (30 mmol) of [Zr- (h-C5H5)Cl3(dme)] were added in several portions. The slurry was stirred for 30 min at this temperature before being allowed to warm to room temperature. The oV-white suspension was then stirred overnight. The volatiles were removed under reduced pressure and the resultant white solid was extracted into 100 ml of toluene at 50 8C. The extract was concentrated and stored at 230 8C to yield the desired compound as a white solid. Yield 5.79 g 13.6 mmol (45%). 1H NMR (CDCl3 300 MHz 20 8C) d 1.77 (s 6 H CCH3); 2.32 (s 3 H PhCH3); 6.28 J. Chem. Soc. Dalton Trans. 1999 2111–2118 2117 (s 5 H Cp); 6.37 (“t” 2 H Cp9); 6.49 (“t” 2 H Cp9); and 7.16 (“d” 4 H Ph).13C NMR (CDCl3 75.5 MHz 20 8C) d 20.8 (PhCH3); 29.5 (CCH3); 40.0 (CCH3); 114.2 (Cp9); 116.1 (Cp); 126.1 (o-C of Ph); 128.7 (m-C of Ph) and 135.8 (Cq). [Zr(Á-C5H4CMe2C6H4Me-p)2Cl2] 5a. The compound nbutyllithium (24 ml 60 mmol 2.5 mol l21 in hexanes) was added to a solution of 10.26 g (60 mmol) of 4-bromotoluene in 250 ml of diethyl ether at room temperature. The reaction mixture was stirred at 30 8C for 1 h then cooled to 278 8C and 6.38 g (60 mmol) of neat 6,6-dimethylfulvene were slowly added. The yellow solution was slowly warmed to room temperature. An oV-white precipitation occurred which was dissolved by addition of 30 ml of thf. The slightly yellow solution was recooled to 278 8C and 11.3 g (30 mmol) of ZrCl4?2thf were added in several portions.The slurry was stirred for 30 min at this temperature before being allowed to warm to room temperature. The oV-white suspension was stirred overnight. The volatiles were removed under reduced pressure and the resulting white solid was extracted into 180 ml of toluene at 50 8C. The extract was concentrated and stored at 230 8C to yield the desired compound as a white solid. Yield 5.85 g 10.5 mmol (35%). 1H NMR (CDCl3 500 MHz 20 8C) d 1.75 (s 12 H CCH3); 2.28 (s 6 H PhCH3); 6.15 (“t” 4 H Cp9); 6.34 (“t” 4 H Cp9); and 7.08 (m 8 H Ph). 13C NMR (CDCl3 125.7 MHz 20 8C) d 20.8 (PhCH3); 29.4 (CCH3); 40.1 (CCH3); 113.1 (Cp9); 117.1 (Cp9); 126.0 (m-C of Ph); 128.8 (o-C of Ph); 135.6 (Cq); 142.3 (Cq); and 146.9 (Cq). [Zr(Á-C5H5)(Á-C5H4CMe2Ph)Me2] 2. A suspension of 2.46 g (5.9 mmol) of complex 2a in 80 ml of diethyl ether at 278 8C was treated with 4.00 ml (11.9 mmol) of MgMeBr (3 mol l21 in diethyl ether) in a dropwise manner.The reaction mixture was slowly warmed to room temperature and stirred for 3 h. The volatiles were removed under reduced pressure to yield an oV- white solid which was extracted into 180 ml of light petroleum. The extract was concentrated and cooled to 230 8C to yield cushions of needles. Yield 1.27 g 3.4 mmol (58%) (Found C 67.7; H 7.1. C21H26Zr requires C 67.6; H 6.9%). 1H NMR (CDCl3 300 MHz 20 8C) d 20.29 (s 6 H ZrMe2); 1.61 (s 6 H CMe2); 5.99 (s 5 H Cp); 6.04 (m 4 H Cp9); and 7.19–7.30 (m 5 H Ph). 13C NMR (CDCl3 125.7 MHz 20 8C) d 30.0 (CCH3); 30.5 (ZrMe); 39.6 (CCH3); 108.6 (Cp9); 110.3 (Cp9); 110.6 (Cp); 125.8 (Cq); 126.0 (m-C of Ph); 128.0 (o-C of Ph); 137.6 (Cq); and 150.5 (Cq).[Zr(Á-C5H5)(Á-C5H4CMe2C6H4Me-p)Me2] 3. The preparation was carried out in a manner similar to that for complex 2. The crystals obtained were suitable for X-ray analysis. Yield 3.66 g 9.5 mmol (70%) (Found C 68.2; H 7.3. C22H28Zr requires C 68.8; H 7.4%). 1H NMR (CDCl3 300 MHz 20 8C) d 20.28 (s 6 H ZrCH3); 1.60 (s 6 H CCH3); 2.37 (s 3 H PhCH3); 6.00 (s 5 H Cp); 6.04 (m 4 H Cp9); and 7.13 (m 4 H Ph). 13C NMR (CDCl3 75.5 MHz 20 8C) d 20.8 (PhCH3); 30.1 (CCH3); 30.5 (ZrCH3); 39.4 (CCH3); 108.7 (Cp9); 110.2 (Cp9); 110.6 (Cp); 126.0 (o-C of Ph); 128.7 (m-C of Ph); 135.3 (Cq); 137.9 (Cq); and 147.6 (Cq). [Zr(Á-C5H4CMe2Ph)2Me2] 4. To a suspension of 4.32 g (10.5 mmol) of complex 4a (second fraction obtained from the preparation of 2a) in diethyl ether 14 ml (21 mmol) of LiMe (1.5 mol l21) in diethyl ether were added at 278 8C.After the addition was complete the reaction mixture was slowly warmed to room temperature and stirred for 30 min. The volatiles were removed under reduced pressure and the oV white residue was extracted into warm light petroleum. The extract was stored at 5 8C for 3 d to yield 4 as the only product. Crystals suitable for X-ray analysis were grown by slowly cooling a light petroleum solution to 280 8C. Yield 0.54 g 0.8 mmol (8%) (Found C 73.5; H 7.6. C30H36Zr requires C 73.9; H 7.4%). 1H NMR (CDCl3 300 MHz 20 8C) d 20.24 (s 6 H ZrMe); 1.58 (s 12 H CCH3); 5.90 (“t” 4 H Cp9); 5.98 (“t” 4 H Cp9); and 7.19–7.27 (m 10 H Ph). 13C NMR (CDCl3 75.5 MHz 20 8C) d 29.9 (CCH3); 31.1 (ZrMe); 39.8 (CCH3); 109.7 (Cp9); 110.8 (Cp9); 125.8 (p-C of Ph); 126.0 (m-C of Ph); 128.0 (o-C of Ph); 137.2 (Cq); and 150.6 (Cq).[Zr(Á-C5H4CMe2C6H4Me-p)2Me2] 5. The preparation was conducted in a similar manner to that of complex 4 but 5a was used as starting material. Yield 1.73 g 3.3 mmol (31%) (Found C 73.9; H 7.8. C32H40Zr requires C 73.9; H 7.85%). 1H NMR (CDCl3 300 MHz 20 8C) d 20.21 (s 6 H ZrMe); 1.56 (s 12 H CCH3); 2.31 (s 6 H PhMe); 5.91 (“t” 4 H Cp9); 5.98 (“t” 4 H Cp9); and 7.10 (“s” 8 H Ph). 13C NMR (CDCl3 75.5 MHz 20 8C) d 20.8 (PhCH3); 29.9 (CCH3); 31.1 (ZrMe); 39.4 (CCH3); 109.6 (Cp9); 110.1 (Cp9); 125.9 (m-C of Ph); 128.6 (o-C of Ph); 135.2 (Cq); 137.4 (Cq); and 147.7 (Cq). Low temperature NMR studies on cationic compounds general procedure The zirconocene complex (0.1 mmol) was dissolved in 0.25 ml of CD2Cl2 and transferred to a precooled (278 8C) NMR tube.The cation generating agent (0.11 mmol) such as B(C6F5)3 or [Ph3C]1[B(C6F5)4]2 was dissolved in 0.28 ml of CD2Cl2 and transferred to the top of the zirconocene solution in the NMR tube. The tube was sealed with a Suba Seal and shaken vigorously to ensure complete mixing. The colour changed to yellow and the sample was inserted into a precooled (260 8C) spectrometer. The 1H 13C-{1H} H–H COSY and C–H COSY spectra were recorded at 260 8C. The sample was warmed to ambient temperature in steps of 20 K and at each temperature a 1H NMR spectrum was recorded. Crystal structure determination Data collection and processing. Data were collected on an Enraf-Nonius DIP2000 image plate diVractometer with graphite monochromated Mo-Ka radiation (l = 0.71069 Å) as summarised in Table 4.The images were processed with the DENZO and SCALEPACK programs.21 Corrections for Lorentz-polarisation eVects were performed. Table 4 Crystallographic data of complexes 3 and 4 3 4 Formula M Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 V/Å3 Z Dc/g cm23 T/K m(Mo-Ka)/mm21 Transmission coeYcients F(000) Total data No. unique data No. observed data [I > 3s(I )] No. parameters RR 9 R(int) Goodness of fit Largest peak/e Å23 C22H28Zr 383.69 Triclinic P1� 7.009(3) 11.633(4) 12.838(4) 109.01(2) 94.91(2) 106.07(2) 933.0 2 1.37 110 0.58 0.68–0.75 400 5648 1144 1021 209 0.0407 0.0501 0.049 1.0860 0.31 C15H18Zr0.5 243.92 Monoclinic C2/c 18.8890(8) 6.8630(3) 19.0360(5) 101.401(3) 2419.0 8 1.34 125 0.46 0.86–0.91 1024 6627 2502 2463 141 0.0403 0.0395 0.029 0.9832 0.47 2118 J.Chem. Soc. Dalton Trans. 1999 2111–2118 Structure solution and refinement. All solution refinement and graphical calculations were performed using the CRYSTALS22 and CAMERON23 software packages. Figs. 1 and 2 were generated with ORTEP,24 Fig. 3 with CAChe.25 The crystal structure was solved by direct methods using the SIR 92 program26 and refined by full-matrix least squares procedure on F. All non-hydrogen atoms were refined with anisotropic displacement parameters. All carbon-bound hydrogen atoms were generated and allowed to ride on their corresponding carbon atoms with fixed thermal parameters. A Chebychev weighting scheme was applied as well as an empirical absorption correction.27 For compound 3 the crystal was of moderate quality thus giving a relatively low ratio of data to refined parameters.We have processed the data for two diVerent mosaicities and obtained R(int) = 0.049 at low mosaicity and 0.033 at higher mosaicity. The final R factors shift from 0.0407 (R9 = 0.0501) to 0.0448 (0.0542) for processing with higher mosaicity leading us to believe the data processed with a lower mosaicity better represent the molecular structure which is unambiguous in either case. CCDC reference number 186/1459. See http://www.rsc.org/suppdata/dt/1999/2111/ for crystallographic files in .cif format. Acknowledgements D. H. wishes to thank the Deutsche Forschungsgemeinschaft for financial support and L. H. D. thanks St. John’s College (Oxford) for a Junior Research Fellowship. References 1 M. Bochmann J. Chem.Soc. Dalton Trans. 1996 255. 2 W. Kaminsky and M. Arndt Adv. Polym. Sci. 1995 127 144. 3 H.-H. Brintzinger D. Fischer R. Mühlhaupt B. Rieger and R. Waymouth Angew. Chem. Int. Ed. Engl. 1995 34 1143. 4 G. L. Rempel and J. Huang Prog. Polym. Sci. 1995 20 459. 5 M. Bochmann and S. J. Lancaster J. Organomet. Chem. 1992 434 C1. 6 J. A. Ewan and M. J. Elder Chem. Abstr. 1991 115 136987c 136988d; Eur. Pat. Appl. 1991. 7 X. Yang C. L. Stern and T. J. Marks J. Am. Chem. Soc. 1991 113 3623. 8 X. Yang C. L. Stern and T. J. Marks J. Am. Chem. Soc. 1994 116 10015. 9 M. Bochmann and S. J. Lancaster Angew. Chem. Int. Ed. Engl. 1994 33 1634. 10 S. Beck M.-H. Prosenc H.-H. Brintzinger R. Goretzki N. Herfert and G. Fink J. Mol. Catal. A 1996 111 67. 11 D. Huang J. C. HuVman J. C. Bollinger O. Eisenstein and K.G. Caulton J. Am. Chem. Soc. 1997 119 7398. 12 D. J. Gillis M. J. Tudoret and M. C. Baird J. Am. Chem. Soc. 1993 115 2543. 13 D. J. Gillis R. Quyoum M. J. Tudoret Q. Wang D. Jeremic A. W. Roszak and M. C. Baird Organometallics 1996 15 3600. 14 S. J. Lancaster O. B. Robinson M. Bochmann S. J. Coles and M. B. Hursthouse Organometallics 1995 14 2456. 15 M. Bochmann M. L. H. Green A. K. Powell J. Saßmannshausen M. U. Triller and S. Wocaldo J. Chem. Soc. Dalton Trans. 1999 43. 16 P. C. Möhring N. Vlachakis N. E. Grimmer and N. J. Coville J. Organomet. Chem. 1994 483 159. 17 A. D. Horton J. de With A. J. v. d. Linden and H. v. d. Weg Organometallics 1996 15 2672. 18 M. Brookhart M. L. H. Green and L. Wong Prog. Inorg. Chem. 1988 36. 19 E. C. Lund and T. Livinghouse Organometallics 1990 9 2426.20 W. Freiersleben Angew. Chem. 1963 75 576. 21 Z. Otwinowski and W. Minor Processing of X-Ray DiVraction Data Collected in Oscillation Mode Academic Press New York 1996 p. 276. 22 D. J. Watkin C. K. Prout J. R. Carruthers and P. W. Bettridge CRYSTALS Oxford University 1996. 23 D. J. Watkin C. K. Prout and L. J. Pearce CAMERON Oxford University 1996. 24 C. K. Johnson ORTEP Report ORNL-5138 Oak Ridge National Laboratory Oak Ridge TN 1976. 25 CAChe Work System V. 3.8 Oxford Molecular Group 1996. 26 A. Altomare G. Cascarano G. Giacovazzo A. Guagliardi M. C. Burla G. Polidori and M. Camalli J. Appl. Crystallogr. 1994 27 435. 27 N. Walker and D. Stuart Acta Crystallogr. Sect. A 1983 39 158. Paper 8/0817