Received 24 February 2011
The crystal structure of the title compound, trans-[PtCl2(C16H23P)2], has been determined at 100 K. The Pt atom is located on a twofold axis and adopts a distorted square-planar coordination geometry. The structure is only the second example of a coordination complex containing a derivative of the 4,8-dimethyl-2-phosphabicyclo[3.3.1]nonane (Lim) phosphine ligand family. The ligand contains four chiral C atoms, with the stereochemistry at three of these fixed during synthesis, therefore resulting in two possible ligand stereoisomers. The compound crystallizes in the chiral space group P43212 but is racemic, comprising an equimolar mixture of both stereoisomers disordered on a single ligand site. The effective cone angles for both isomers are the same at 146°.
Lim ligands (2-Q-4,8-dimethyl-2-phosphabicyclo[3.3.1]nonane) are derived from the radical addition reaction of the monoterpene R-(+)-limonene with a QPH2 molecule, where Q = H or some other suitable monoanionic group such as alkyl or aryl, resulting in a racemic mixture of ligands being obtained. Although the phosphine Lim backbone contains four chiral C atoms, the stereochemistry at three of these sites is fixed, viz. C1 (R), C5 (R) and C8 (S), while C4 can have either an R or S configuration. This stereochemistry is a consequence of performing the synthesis with the optically pure terpene and the mechanism of addition to the P atom (Robertson et al., 2001).
Chiral phosphine ligands are of general interest in coordination chemistry and catalysis, and ligands of the Lim family have been shown to display exceptional qualities in the modified cobalt hydroformylation of alkenes to give alcohols directly (Steynberg et al., 2002; Crause et al., 2003; Dwyer et al., 2004). The only other crystal structure in the open literature of a coordination compound containing a member of the Lim ligand family, [Co(CO)3(Lim-C18)]2 (Lim-C18 is the 4R isomer of 2-octadecyl-4,8-dimethyl-2-phosphabicyclo[3.3.1]nonane), was obtained during such a study (Polas et al., 2003).
In order to investigate further the coordination mode of these ligands, we prepared [PtCl2(Lim-Ph)2] by reaction of [PtCl2(COD)] (COD is cis,cis-cycloocta-1,5-diene) with two molar equivalents of a solution containing a mixture of both Lim-Ph isomers. Recrystallization as described in the Experimental section resulted in crystals of (I) being obtained.
Compound (I) crystallizes with a distorted square-planar coordination geometry, with a twofold rotation axis passing through the Pt metal centre and bisecting the P2-Pt1-P2i and Cl1-Pt1-Cl1i angles [symmetry code: (i) y, x, -z] (Fig. 1). The Lim-Ph ligands adopt a trans orientation, suggesting significant steric bulk, although cis isomers have been observed in solution using 31P NMR (vide infra). The coordination geometry deviates significantly from ideal square planar, with P2-Pt1-P2i and Cl1-Pt1-Cl1i angles of 170.97 (9) and 175.49 (8)°, respectively. Interestingly, the C11 methyl groups, which contribute significantly to the overall steric bulk of the Lim-Ph ligands, occupy the same side of the equatorial plane, with a closest contact of only 3.655 (13) Å between C11 and C11i. This interaction manifests itself in the deviation of the P atoms below the equatorial plane. In addition, the presence of these two methyl substituents effectively blocks one apical position of the Pt atom, with Pt1C11 contacts of only 3.563 (6) Å. The Pt1-P2 bond distance of 2.3088 (14) Å is within the expected range, while the Pt1-Cl1 distance of 2.3320 (13) Å is quite long. This elongation is probably a consequence of the steric repulsion of the two bulky phosphine ligands and the resulting distortion from square planarity. The deviations in the bond angles from the ideal value of 180° would impact negatively on the efficiency of the relevant orbital overlap between the atoms involved. Table 1 presents a comparison with related structures, also containing bulky ligands, taken from the open literature, to illustrate this effect.
The Lim-Ph ligand exhibits disorder in the orientation of the C10 methyl group, with components A and B corresponding to the 4R and 4S isomers, respectively (Fig. 1). Refinement of the site occupancies for C10A and C10B yielded values that did not differ significantly from 0.5 and the occupancies were therefore constrained to 0.5 for subsequent refinement, corresponding to a true racemic mixture. Short intermolecular contacts [C10AC10Aii = 2.502 (16) Å; symmetry code: (ii) y - 1, x + 1, -z] preclude the simultaneous presence of C10A in neighbouring molecules, but there are no constraints on the presence of C10B.
Describing the steric demand of phosphine ligands has been the topic of many studies and a variety of models have been developed (Bunten et al., 2002). In practice, the Tolman cone angle (Tolman, 1977) is still the most commonly used model, due to its simplicity and ease of calculation. This principle has been further developed (Otto, 2001) into the concept of the `effective cone angle', where the crystallographically determined metal-P bond length is used in the calculations. Using the Pt1-P2 bond distance obtained in this study and calculating the cone to the outermost H atoms (H11A, H19A and H25A) on C11, C19 and C25 results in a value of 146°. In addition, the cone angle is independent of the orientation of C10.
31P NMR analysis of the reaction mixture indicated a number of species in solution corresponding to Pt complexes of both cis and trans geometry, as well as containing combinations of the different ligand isomers, i.e. (4R,4R), (4S,4S) and (4R,4S). Aside from the constraint observed for the intermolecular contacts involving C10A, the refined 50% disorder in the orientation of the C10 Me group is consistent with any of these combinations. Redissolving some of the single crystals obtained and recollecting the 31P NMR spectrum confirmed that mixtures of this nature are indeed present in both the solid and solution states.
Based on high-pressure NMR experiments, it was previously shown that the 4R isomer coordinates preferentially during modified Co hydroformylation (Polas et al., 2003; Dwyer et al., 2004), and this observation was supported by modelling studies (Crause et al., 2003). Considering, however, that the two isomers are electronically and sterically (as shown here) very similar, this behaviour is currently not well understood and may warrant further investigation.
| || Figure 1 |
The molecular structure of (I), showing the atom-numbering scheme. Displacement ellipsoids are drawn at the 50% probability level and H atoms have been omitted for clarity. [Symmetry code: (i) y, x, -z.]
The Lim-Ph ligand (mixture of isomers) was prepared by adapting methods described previously (Bungu & Otto, 2007). All manipulations involving the free ligand were performed using degassed solvents and working under a positive argon atmosphere to prevent oxidation. PtCl2(COD) (COD is cis,cis-cycloocta-1,5-diene) (200 mg, 0.53 mmol) was dissolved in dichloromethane (10 ml) and a dichloromethane solution of the ligand mixture (1.49 ml, 753 mM, 1.12 mmol) was subsequently added. The resulting reaction mixture was stirred overnight and a portion was subjected to 31P NMR analysis. The spectra were quite complex, with both cis and trans PtII complexes present as mixtures of the two ligand isomers. Crystals of compound (I) suitable for single-crystal diffraction studies were obtained by addition of acetone to the dichloromethane reaction mixture followed by slow evaporation.
31P (CDCl3): trans-[PtCl2(4R-Lim-Ph)2] -8.82 p.p.m. (t, 1JPt-P = 2378 Hz); trans-[PtCl2(4R-Lim-Ph)(4S-Lim-Ph)] -9.83 (4R, t, 1JPt-P = 2378 Hz) and -12.42 p.p.m. (4S, t, 1JPt-P = 2383 Hz); trans-[PtCl2(4S-Lim-Ph)2] -13.58 p.p.m. (t, 1JPt-P = 2384 Hz).
The disorder of the methyl substituent on C4 of the Lim-Ph ligand was modelled as two orientations with occupancies summing to unity. Occupancies of 0.493 (18) and 0.507 (18) were obtained for C10A and C10B, respectively. Since these values do not differ significantly from 0.5, they were constrained to 0.5 for further refinement. The C4-C10A and C4-C10B distances were tightly restrained to 1.530 (5) Å. H atoms were placed geometrically with C-H distances of 1.00 Å for CH (alkyl), 0.95 Å for CH (aryl), 0.99 Å for CH2 and 0.98 Å for CH3, and constrained to ride on their parent atoms, with Uiso(H) = 1.2Ueq(C) for CH and CH2 or 1.5Ueq(C) for CH3. For the methyl groups, rotation was permitted about the C-C bond.
Data collection: APEX2 (Bruker, 2008); cell refinement: SAINT-Plus (Bruker, 2004); data reduction: SAINT-Plus and XPREP (Bruker, 2004); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: DIAMOND (Brandenburg, 2001); software used to prepare material for publication: SHELXL97.
Supplementary data for this paper are available from the IUCr electronic archives (Reference: BI3013 ). Services for accessing these data are described at the back of the journal.
Cytec is thanked for generously supplying the Lim-H ligand precursor. Financial support from Sasol Technology Research and Development and from the research fund of the University of the Free State is gratefully acknowledged. Part of this material is based on work supported by the South African National Research Foundation (NRF) under grant No. GUN 2053397. Any opinion, finding and conclusions or recommendations in this material are those of the authors and do not necessarily reflect the views of the NRF.
Brandenburg, K. (2001). DIAMOND. Crystal Impact GbR, Bonn, Germany.
Bruker (2004). SAINT-Plus and XPREP. Bruker AXS Inc., Madison, Wisconsin, USA.
Bruker (2008). SADABS and APEX2. Bruker AXS Inc., Madison, Wisconsin, USA.
Bungu, P. N. & Otto, S. (2007). J. Organomet. Chem. 692, 3370-3379.
Bunten, K. A., Chen, L., Fernandez, A. L. & Poe, A. J. (2002). Coord. Chem. Rev. 233-234, 41-51.
Carreira, M., Charernsuk, M., Eberhard, M., Fey, N., Van Ginkel, R., Hamilton, A., Mul, W. P., Orpen, A. G., Phetmung, H. & Pringle, P. G. (2009). J. Am. Chem. Soc. 131, 3078-3094.
Crause, C., Bennie, L., Damoense, L., Dwyer, C. L., Grove, C., Grimmer, N., Janse van Rensburg, W., Kirk, M. M., Mokheseng, K. M., Otto, S. & Steynberg, P. J. (2003). Dalton Trans. pp. 2036-2042.
Del Pra, A. & Zanotti, G. (1980). Inorg. Chim. Acta, 39, 137-141.
Doherty, R., Haddow, M. F., Harrison, Z. A., Orpen, A. G., Pringle, P. G., Turner, A. & Wingad, R. L. (2006). Dalton Trans. pp. 4310-4320.
Dwyer, C., Assumption, H., Coetzee, J., Crause, C., Damoense, L. & Kirk, M. (2004). Coord. Chem. Rev. 248, 653-670.
Flack, H. D. (1983). Acta Cryst. A39, 876-881.
Johansson, M. H. & Otto, S. (2000). Acta Cryst. C56, e12-e15.
Johansson, M. H., Otto, S. & Oskarsson, Å. (2002). Acta Cryst. B58, 244-250.
Messmer, G. G. & Amma, E. L. (1966). Inorg. Chem. 5, 1775-1781.
Otto, S. (2001). Acta Cryst. C57, 793-795.
Otto, S. & Roodt, A. (1997). Acta Cryst. C53, 1414-1416.
Polas, A., Wilton-Ely, J. D. E. T., Slawin, A. M. Z., Foster, D. F., Steynberg, P. J., Green, M. J. & Cole-Hamilton, D. J. (2003). Dalton Trans. pp. 4669-4677.
Robertson, A., Bradaric, C., Frampton, C. S., McNulty, J. & Capretta, A. (2001). Tetrahedron Lett. 42, 2609-2612.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.
Steynberg, J. P., Govender, K. & Steynberg, P. J. (2002). Int. Patent WO2002014248.
Tolman, C. A. (1977). Chem. Rev. 77, 313-348.