Diffusion coefficients of zirconocenendash;borate ion pairs studied by pulsed field-gradient NMRmdash;evidence for ion quadruples in benzene solutions

摘要

Diffusion coefficients of zirconocenendash;borate ion pairs studied by pulsed field-gradient NMRmdash;evidence for ion quadruples in benzene solutions Stefan Beck,dagger; Armin Geyer* and Hans-Herbert Brintzinger Fakultauml;t fuuml;r Chemie, Universitauml;t Konstanz, D-78457 Konstanz, Germany. E-mail: Hans.Brintzinger@uni-konstanz.de Received (in Oxford, UK) 11th October 1999, Accepted 5th November 1999 Diffusion coefficients for ion pairs containing zirconocene cations and fluorinated tetraphenyl- or methyltriphenylborate anions in benzene solutions are substantially smaller than those expected for monomeric species and indicate that ion quadruples are present in these solutions.Electron deficient metallocene alkyl cations of group 4 metals, in combination with weakly coordinating anions are considered to be reactive species in homogeneous olefin polymerization catalysis.1 Particularly well defined in this regard, in several instances even by solid-state structures,2,3a are zirconocene ion pairs which contain the fluorinated borate anions B(C6F5)42 or MeB(C6F5)32; these are obtained by reaction of a dimethyl zirconocene complex with a triorganylammonium or triphenylcarbenium salt of B(C6F5)42 or with B(C6F5)3, respectively. 4 The binding of the anion to the cationic center in these contact-ion pairs is of importance for the activity of a catalyst system, since the anion A2 has to be displaced by the olefin substrate to initiate the catalytic cycle. Exchange of these anions between zirconocene coordination sites as well as their displacement by entities such as a Lewis base, AlMe3 or another dimethyl zirconocene complex have been studied in considerable detail.1ndash;3 Rather little is known, however, about the structures of these ion pairs in solution, especially with regard to the extent to which they form dissociated, associated or aggregated ion pairs in the non-polar solvents commonly used for homogeneous olefin polymerizations.In order to clarify this point, we have determined the size of these catalytically active species in solution by measuring their translational diffusion coefficients Dt by NMR methods.The LED (longitudinal eddy current delay) experiment measures the intensity of a spinndash;echo amplitude as a consequence of the change in spatial position of a molecule during the time interval between two pulsed field gradients.5 The changing signal intensity (I) with incremented gradient strength G is related to the translational diffusion coefficient Dt by the Stejskalndash;Tanner equation I = I0exp{2(gdG)2 D 2 (d/3) Dt},6 where g is the gyromagnetic ratio, d the gradient duration and D the interval between the pulsed field gradients. The method works also for complexes and salts.7 For approximatively spherical particles which are significantly larger than the solvent molecules, the Stokesndash;Einstein equation Dt = kT/(6p hr), where k = Boltzmann constant, T = absolute temperature and h = viscosity of the solvent, is used to relate the translational self-diffusion coefficient Dt to the radius r of the dissolved particle. For molecules with van der Waals radii rW 5 Aring;, the Stokesndash;Einstein factor of 6 has to be changed to a value of 4.8 We find this so-called perfect slip condition to hold for dilute benzene solutions of the bis(cyclopentadienyl) dimethylzirconium compounds shown in Scheme 1 (Table 1).The diffusion coefficients of these neutral, monomeric zirconocene complexes decrease linearly with increasing molecular size: The inverse diffusion coefficients of these species correlate with their calculated van der Waalrsquo;s radii rW with deviations of 10 (Fig. 1). Rather unexpected results, however, pertain to a number of ion pairs derived from compounds 1 and 5: While the Dt value obtained for the methylborate ion pair 5A deviates only slightly from the calibration provided by compounds 1ndash;5, the experimentally determined Dt value of ion pair 1A amounts to only ca. 80 of the expected value. These deviations from expectation are even more pronounced for the ion pairs 1B and 5B, which contain the more weakly coordinating B(C6F5)42 anion and are, hence, presumably more polar. The Dt value obtained for ion pair 1B in particular (5.1 3 10210 m2 s21), yields an apparent particle volume of 1260 Aring;3, which corresponds to just twice the expected volume of this ion pair. Even though a number of caveats have to be heededmdash;a pronounced anisotropic molecular shape, a change from perfect slip (SE factor 4) to perfect stick conditions (SE factor 6) or an increased solvent shell might contribute to the observed decrease of Dtmdash;we propose that ion pair 1B is present in these solutions predominantly in form of its dimer, i.e.of an ion quadruple (Scheme 2). This contention is supported by the observation that 1B has practically the same diffusion coefficient as the related ion pair (C5H5)2ZrMe(m-Me)Me- Zr(C5H5)2+ B(C6F5)42 1C, for which a binuclear structure has been independently established.3a,9 From the theory of ion solutions in organic solvents it is indeed to be expected that ion quadruples are present in benzene solutions of ion pairs such as 1B at the concentrations considered here,10 which are close to the limiting concentrations at which these ion pairs begin to form separate phases.In accord with these theoretical considerations, the degrees of ion quadruple formation, indicated by the deviations of Dt from the values expected for monomeric species, are higher for the more polar ion pairs 1B and 5B than for the presumably partly covalent ion pairs 1A and 5A and higher for ion pairs 1A and 1B, which contain the rather small (C5H5)2ZrMe+ cation, than for their more voluminous analogues 5A and 5B, respectively.More detailed data on degrees of aggregation and their dependence on the nature of the zirconocene cation and its counter anion will clearly be needed for a full description of dagger; Present address: School of Chemistry, University of Leeds, Leeds, UK LS2 9JT.Scheme 1 This journal is copy; The Royal Society of Chemistry 1999 Chem. Commun., 1999, 2477ndash;2478 2477reaction systems of this type. From the data presented here it is already clear, however, that the hitherto entirely neglected formation of ion quadruples will have to be taken into account for an analysis of exchange reactions between anions and other ligands in zirconocene-based catalyst systems.Financial support of this work by BASF AG, BMBF and by funds of the University of Konstanz is gratefully acknowledged. Notes and references 1 M. Bochmann, J. Chem. Soc., Dalton Trans., 1996, 255 and references therein. 2 X. Yang, C. L. Stern and T. J. Marks, J. Am. Chem. Soc., 1994, 116, 10 015; P. A. Deck, C. L. Beswick and T. J. Marks, J. Am. Chem. Soc., 1998, 120, 1772; L. Luo and T. J. Marks, Top.Catal., 1999, 7, 97 and references therein. 3 (a) S. Beck, M. H. Prosenc, H. H. Brintzinger, R. Goretzki, N. Herfert and G. Fink, J. Mol. Catal., A: Chem., 1996, 111, 67; (b) S. Beck, M. H. Prosenc and H. H. Brintzinger, J. Mol. Catal. A: Chem., 1997, 128, 41. 4 G. G. Hlatky, H. W. Turner and R. R. Eckman, J. Am. Chem. Soc.,1989, 111, 2728; J. A. Ewen and M. J. Elder, Makromol. Chem., Macromol. Symp., 1993, 66, 179; J. C. Chien, W. M. Tsai and M. D. Rausch, J. Am. Chem. Soc., 1991, 113, 8570. 5 S.J. Gibbs and C. S. Johnson, Jr., J. Magn. Reson., 1991, 93, 395; K. F. Morris and C. S. Johnson, Jr., J. Am. Chem. Soc., 1993, 115, 4291. 6 E. O. Stejskal and J. E. Tanner, J. Chem. Phys., 1965, 42, 288. 7 A. Mayzel and Y. Cohen, J. Chem. Soc., Chem. Commun., 1994, 1901; R. E. Hoffman, E. Shabtai, M. Rabinovitz, V. S. Iyer, K. Muuml;llen, A. K. Rai, E. Bayrd and L. T. Scott, J. Chem. Soc., Perkin Trans. 2, 1998, 1659. 8 R. A. Robinson and R. H. Stokes, Elektrolyte Solutions, Butterworth, London, 2nd edn.revised, 1970; J. T. Edwards, J. Chem. Educ., 1970, 47, 261; M. Ue, J. Electrochem. Soc., 1994, 141, 3336. 9 M. Bochmann and S. J. Lancaster, J. Organomet. Chem., 1992, 434, C1; Angew. Chem., Int. Ed. Engl., 1994, 33, 1634; Y.-X. Chen. C. L. Stern, S. Yang and T. J. Marks, J. Am. Chem. Soc., 1996, 118, 12 451; I. Tritto, R. Donetti, M. C. Sacchi, P. Locatelli and G. Zannoni, Macromolecules, 1997, 30, 1247. 10 C. A. Kraus, J. Chem.Phys., 1956, 60, 129; J. E. Gordon, The Organic Chemistry of Electrolyte Solutions, Wiley, New York, 1975, p. 371. Communication 9/08203D Table 1 Diffusion coefficients Dt, measured for the neutral dimethyl zirconocene compounds 1ndash;5 and for the ion pairs 1Andash;C and 5A, B in C6D6 solution at 300 K, and van der Waals radii rW and molecular volumes VW of these zirconocene complexes Zirconocene complex a 1010 Dt b/m2 s21 rW c/Aring; VW c/Aring;3 1 (C5H5)2ZrMe2 14.9 3.6 196 2 Me4C2(C5H4)2ZrMe2 13.4 4.1 290 3 (C5H4But)2ZrMe2 12.0 4.3 335 4 rac-Me2Si(ind)2ZrMe2 11.5 4.4 354 5 rac-Me2Si(2-Me-benzeind)2ZrMe2 10.1 4.9 481 1A (C5H5)2ZrMe+middot;middot;middot;MeB(C6F5)32 8.1 5.0 513 1B (C5H5)2ZrMe+middot;middot;middot;B(C6F5)42 5.1 5.2 593 1C {(C5H5)2ZrMe}2(m-Me)+ B(C6F5)42 4.9 5.7 789 5A rac-Me2Si(2-Me-benzeind)2ZrMe+middot;middot;middot;MeB(C6F5)32 8.2 5.7 797 5B rac-Me2Si(2-Me-benzeind)2ZrMe+middot;middot;middot;B(C6F5)42 6.1 5.9 877 a Zirconocene concentrations (mmol dm23): 1, 1ndash;20; 2ndash;5, 1; 1A, 4.7; 1B, 1.6; 1C, 2; 5A, 4; 5B, 2.b Measured on a Bruker DRX 600 spectrometer with a BGU II gradient unit using ten different gradient strengths (5ndash;50 G cm21) of 1 ms duration, with diffusion times D of 0.5 s (32 scans) and relaxation delays of 20 s. Integrals and peak intensities were quantified with the Bruker T1/T2 software package. Independent Dt values, determined for individual NMR signals of each complex, gave the average values listed with mean deviations of 5. c Calculated from molecular hard-sphere volume increments: D. Ben-Amotz and K. G. Willis, J. Phys. Chem., 1993, 97, 7736. Fig. 1 Reciprocal diffusion coefficients, 1/Dt, and van der Waals radii, rW, for the neutral zirconocene complexes 1ndash;5, for the ion pairs (C5H5)2Zr- Me+middot;middot;middot;MeB(C6F5)32 1A, (C5H5)2ZrMe+middot;middot;middot;B(C6F5)42 1B, rac-Me2Si(2- Me-benzeind)2ZrMe+middot;middot;middot;MeB(C6F5)32 5A and rac-Me2Si(2-Me-benz- eind)2ZrMe+middot;middot;middot;B(C6F5)42 5B, and for the dinuclear complex {(C5H5)2ZrMe}2(m-Me)+B(C6F5)42 1C. Scheme 2 2478 Chem. Commun., 1999, 2477ndash;2478