Special effects of ortho-isopropylphenyl groups. Diastereoisomerism in platinum(II) and palladium(II) complexes of helically chiral PAr₃ ligands

Abstract

The coordination chemistry of the four phosphines, P{C₆H₃(o-CH₃)(p-Z)}₃ where Z = H (1a) or OMe (1b) and P{C₆H₃(o-CHMe₂)(p-Z)}₃ Z = H (1c) or OMe (1d) with platinum(II) and palladium(II) is reported. Mononuclear complexes trans-[PdCl₂L₂] (L = 1a,b) and trans-[PtCl₂L₂] (L = 1a–c) have been prepared and the crystal structures of trans-[PdCl₂(1b)₂] and trans-[PtCl₂(1c)₂] as their dichloromethane solvates have been determined. The structures show that in these complexes, the ligands adopt g⁺g⁺a conformations. Examination of the Cambridge Structural Database confirms this to be one of only two conformer types that tri-o-tolylphosphines adopt and the only viable conformer in 4 and 6 coordinate complexes. The binuclear complexes trans-[Pd₂Cl₄L₂] (L = 1c,d) are formed even when an excess of the bulky 1c,d is used in the synthesis and the crystal structure of trans-[Pd₂Cl₄(1c)₂] as its chloroform solvate is reported. Reaction of [PtCl₂(NCBu^t)₂] with 1b–d in refluxing toluene gave the cycloplatinated species [Pt₂Cl₂(L − H)₂] where L − H is phosphine 1b–d deprotonated at one of the ortho-methyl carbon atoms. Variable temperature ³¹P and ¹H NMR spectroscopy reveals that all the complexes reported are fluxional. The processes are analysed in terms of restricted P–C and P–M rotations that give rise to diastereoisomeric rotamers because of the helically chiral orientations of the aryl substituents. For the complexes of the bulky ligands 1c,d, rotation about the P–C bond is slow on the NMR timescale even up to 75 °C. The crystal structure of the cyclometallated complex [Pt₂Cl₂(1d − H)₂] has been determined.

---

Introduction

Arylphosphines are of immense importance in coordination chemistry and catalysis and the stereoelectronic effects of ortho-substituents (especially ortho-isopropyl) on an arylphosphine can profoundly influence the catalytic properties of its complexes.¹ One significant consequence of ortho-substituents is an increase in the rigidity of the phosphine caused by restricted P–C bond rotation. Howell² and others³ have published a series of studies on ortho-substituted triarylphosphines and their stereodynamics. Here we report a study of the platinum(II) and palladium(II) coordination chemistry of the bulky tri(o-alkylphenyl)phosphines 1a–d and show that in these complexes, chirality at the triarylphosphine is manifested as conformational diastereoisomerism in bis(phosphine) species.

The conformations of triarylphosphines carrying three ortho-substituted aryl rings may be described in a number of ways. It is well known that triaryl rotor systems may exhibit helical chirality in M and P conformations (see Scheme 1) which can interconvert by a range of processes.⁴ The helical conformations can be defined by the M–P–C_ipso–C_ortho torsion angles ω_l–3 which typically (see below) take values of about ±50°, i.e. gauche (denoted g⁺ or g⁻), or ±180°, i.e. anti (denoted a). The C₃ symmetric M conformation then becomes g⁺g⁺g⁺ (which is the same as conformation A, one of the exo₃ conformations described by Howell et al.²) while the enantiomeric P conformation (= A′) is denoted g⁻g⁻g⁻. We have previously used this terminology to describe the conformations of trialkylphosphine and related species.⁵

Scheme 1 Chiral C₃-symmetric conformers in triarylphosphines viewed along the M–P bond showing the helically chiral M and P conformers (termed g⁺g⁺g⁺ and g⁻g⁻g⁻ respectively in this paper and A and A′ by Howell et al.²).

Results and discussion

The phosphines 1a–d were made by modifications of literature methods (see Experimental section).

Mononuclear dichloroplatinum and dichloropalladium complexes

Phosphines 1a and 1b react with [PtCl₂(NCBu^t)₂] or [PdCl₂(NCPh)₂] in toluene to give products for which elemental analyses and FAB mass spectrometry correspond to [MCl₂(PAr₃)₂] and the IR spectra (in the 200–400 cm⁻¹ region) were consistent with the trans-isomers 2a,b and 3a,b (see Experimental section). Complexes 2a and 3a have been previously reported^6,7 and are only sparingly soluble. However the p-methoxytolyl derivatives 2b and 3b were sufficiently soluble for detailed NMR analysis (see below).

Crystals of 3b, as a dichloromethane solvate, suitable for X-ray crystallography were obtained. Crystals of 2c, as a dichloromethane solvate, were obtained as a by-product from a cyclometallation reaction (see below) and its structure is described now since it elucidates the discussion of the NMR behaviour of all the complexes described here.

Molecules of trans-[PdCl₂(1b)₂] (3b) have crystallographic C_i symmetry (see Fig. 1 and Table 1 which lists selected bond lengths and angles). The average P–C bond length in 3b is 1.83 Å, with an average C–P–C bond angle of 106.4° and a mean Pd–P–C bond angle of 112.1°. Ligand 1b has a cone angle of 181° in this structure in which it adopts a conformation with Pd–P–C_ipso–C_ortho(Me) torsion angles of −47.3, −69.9, and 170.9°, i.e. a g⁻g⁻a conformation. The shortest Pd⋯H contact is 2.91 Å and involves the ortho-hydrogen on the aryl group that adopts the anti conformation. The shortest methyl hydrogen to Pd contacts involve hydrogens on the gauche aryl rings and are at marginally longer Pd⋯H distances (2.95 and 3.16 Å)

Table 1 Selected bond lengths (Å) and angles (°) for trans-[PdCl₂(1b)₂] (3b)·4CH₂Cl₂

Pd(1)–Cl(1)	2.3076(14)	Pd(1)–P(1)	2.3589(11)
Cl(1A)–Pd(1)–Cl(1)	180	Cl(1)–Pd(1)–P(1)	93.21(5)
Cl(1)–Pd(1)–P(1A)	86.79(5)	P(1A)–Pd(1)–P(1)	180

---

Fig. 1 X-Ray crystal structure of trans-[PdCl₂(1b)₂] (3b). Hydrogen atoms are omitted for clarity.

Crystals of trans-[PtCl₂(1c)₂] (2c)·2CH₂Cl₂ were obtained from CH₂Cl₂/toluene solution (see below). Complex 2c has crystallographic C_i symmetry in the solid state. (See Fig. 2 and Table 2 which lists selected bond lengths and angles). The average P–C bond length in 2c is 1.84 Å, with an average C–P–C bond angle of 105.3° and a mean Pt–P–C bond angle of 113.4°. Ligand 1c has a cone angle of 195° in this structure, in which it adopts a conformation at P(1) with Pt–P–C_ipso–C_ortho torsion angles of −66.0, −65.6, 177.5, i.e. a g⁻g⁻a conformation. The isopropyl groups adopt an orientation that has the Me₂CH hydrogen eclipsed to the aromatic ring and towards the phosphorus atom. All ortho-isopropylphenylphosphine fragments in the Cambridge Structural Database (CSD)⁸ adopt this conformation. The shortest Pt⋯H contact in 2c is 3.01 Å and involves the ortho-hydrogen on the aryl group that adopts the anti conformation. The Me₂CH hydrogens on the gauche aryl rings are at marginally longer Pt⋯H distances (3.07 and 3.32 Å).

Table 2 Selected bond lengths (Å) and angles (°) for trans-[PtCl₂(1c)₂] (2c)·2CH₂Cl₂

Pt(1)–Cl(1)	2.3039(10)	Pt(1)–P(1)	2.3436(11)
Cl(1A)–Pt(1)–Cl(1)	180	Cl(1)–Pt(1)–P(1)	92.29(4)
Cl(1)–Pt(1)–P(1A)	87.71(4)	P(1A)–Pt(1)–P(1)	180

---

Fig. 2 X-Ray crystal structure of trans-[PtCl₂(1c)₂] (2c). Hydrogen atoms are omitted for clarity.

The influence of ortho-substituents in triarylphosphine complexes was investigated by searching the CSD for tri-o-alkylphenyl phosphine structures. The most numerous such species are metal complexes of tri-o-tolylphosphine and for the 66 fragments in 60 crystal structures, only two conformer types are observed, corresponding to g⁺g⁺g⁺ and equivalent conformations (49 examples) and g⁺g⁺a and equivalent conformations (17 examples). All square planar and octahedral or pseudo-octahedral (piano-stool etc.) complexes are of the gga conformer type as observed in 3b and 2c (and 4c below); this is the conformation termed B (or B′) and exo₂ by Howell et al.³ The more sterically demanding umbrella-like g⁺g⁺g⁺ conformation (see Scheme 1) is seen in lower coordination number (e.g. 2 or 3) and tetrahedral metal complexes where its larger cone angle is tenable. The complete absence of other conformations (e.g. g⁺aa or g⁺g⁻a) implies that large volumes of the conformation space of the three-rotor M–P(o-tolyl)₃ system are empty, presumably because the conformers in these regions are of unacceptably high energy. Notably both the observed conformer types (g⁺g⁺g⁺ and g⁺g⁺a) are chiral (see Schemes 1 and 2).

Scheme 2 Chiral gga conformers of o-substituted triarylphosphines, as observed in 3b, 2c and 4c and in examples from the CSD.

The complexes 2b and 3b showed broad NMR signals indicating fluxionality. For example, the ³¹P signal for 2b at +23 °C in CD₂Cl₂ was a very broad singlet (w_1/2 90 Hz) at δ 15.8 which, in C₂D₂Cl₄, partially resolved into two broad singlets (w_1/2 80 Hz) at δ 16.1 and 15.1. A variable temperature NMR study was carried out (see Fig. 3).

Fig. 3 Variable temperature ³¹P NMR study of 2b; spectra a–d were measured in C₂D₂Cl₄ and spectra e–h were measured in CD₂Cl₂. The inset is an expansion of the central region of spectrum h showing the two major and two minor signals.

In C₂D₂Cl₄, the ³¹P signals for 2b coalesced at ca. +50 °C and at +100 °C a sharp (w_1/2 5 Hz) singlet at δ 15.9 with ¹J(PtP) 2582 Hz was observed. At 0 °C in CD₂Cl₂, two singlets at δ 15.5 and 14.2 were resolved (intensities ca. 1 : 2), with very similar ¹J(PtP) values (2547, 2522 Hz), consistent with both being trans mononuclear species. At −40 °C, an additional two minor signals (ca. 0.2 × intensity of the major peaks) are resolved at δ 16.1 with ¹J(PtP) 2562 Hz and 14.2 with ¹J(PtP) obscured but estimated to be ca. 2500 Hz. The intensities and chemical shifts of the four signals changed considerably with temperature (see Fig. 3); for the two major species, the ratio changes from 2 : 1 at 0 °C to 2 : 3 at −80 °C. Variable temperature ¹H NMR spectroscopy further elucidated the fluxionality of 2b (see Fig. 4).

Fig. 4 ¹H NMR spectrum of 2b at (a) −60; (b) +23; (c) +100 °C; spectrum a was measured in CD₂Cl₂ and spectra b and c were measured in C₂D₂Cl₄. The inset is an expansion of the high frequency aromatic region of spectrum a showing the two major and two minor signals.

At +23 °C in C₂D₂Cl₄, the ¹H NMR signals for 2b were all broad (w_1/2ca. 60 Hz): five resonances in the 1.5–3.2 ppm range (integrating for 18 protons) were assigned to the tolyl-CH₃ groups, a broad singlet at 3.75 ppm (integrating for 18 protons) and assigned to the OMe groups, and finally the aromatic signals which fell into two regions 7.1–7.8 (integrating for 16 protons) and 8.7–9.1 ppm (integrating for 2 protons). By +100 °C, the signals are all sharp: the alkyl signals have coalesced to singlets at 2.28 (CH₃) and 3.88 ppm (OCH₃) and the aromatic signals have coalesced to two signals at 7.81 (6H) and 6.76 (12H) ppm. At −60 °C, 6 major singlets and 6 minor singlets are observed for the tolyl-CH₃ groups and 2 major and 2 minor ‘quartets’ (which reduce to doublets in the ¹H{³¹P} NMR spectrum) are observed in the 8.5–9.5 ppm region. The variable temperature ³¹P{¹H} and ¹H NMR spectra for the palladium analogue 3b were closely similar to those for 2b (see Experimental section).

In the light of the structural and database studies, the fluxionality observed in the NMR spectra of 2b and 3b can be interpreted in terms of restricted rotation about the P–C and P–M bonds in a similar fashion to Howell's treatment of the fluxionality of [Cr(CO)₅(1a)].² At high temperatures, the spectra are consistent with rapid rotation processes rendering the o-tolyl groups equivalent on the NMR timescale. At low temperatures, the ³¹P and ¹H NMR spectra of 2b show the presence of 2 major and 2 minor trans-PtCl₂P₂ species that we associate with the presence of diastereomeric rotamers. Since no P–P coupling was observed, the trans P atoms must be equivalent in each of the species. From the database results noted above, the g⁺g⁺g⁺ conformer has not previously been observed in square planar species. Given this and that the equivalence of the P atoms would require formation of ggg/ggg and gga/gga rotamers exclusively (and none of the mixed ggg/gga rotamers), we believe it more probable, and therefore assume henceforth, that the ligands adopt only the enantiomeric g⁺g⁺a and g⁻g⁻a conformations. The two major species at low temperatures are therefore the C_i symmetric meso (g⁺g⁺a/g⁻g⁻a) rotamer and the enantiomeric pair of C₂ symmetric rac (g⁺g⁺a/g⁺g⁺a and g⁻g⁻a/g⁻g⁻a) rotamers.

Some insight into the nature of the fluxionality may be gained from examination of the crystal structures of 2c and 3b. The structure of 2c in the crystal is redrawn in Scheme 3 viewed along the P–M–P axis and the proposed process for interconversion of the diastereoisomeric rotamers is shown.

Scheme 3 (a) The structure of 2c in the crystal viewed along the P–M–P axis and its schematic representation used in (b), the proposed interconversion of meso- and rac-rotamers (with the chloride ligands not shown but lying in the plane across the page as in (a)).

The meso-rotamer is shown in a staggered (C_i) conformation as is observed in the solid state structures of 2c as well as 3b and in trans-[PtI₂(1a)₂]⁷ and trans-[PdBr₂(1a)₂].⁹ The P–C rotation process (in which the back triarylphosphine goes from g⁺g⁺a to g⁻g⁻a) shown in step i, yields the rac-rotamer but as drawn, this has C₁ symmetry making the P-atoms inequivalent. A 120° rotation around one M–P bond (step ii) would render the P atoms equivalent, as shown in Scheme 3.

From the variable temperature ¹H NMR spectra of 2b, a T_c of 323 K was measured giving a calculated barrier† of ca. 65 (±2) kJ mol⁻¹ for the g⁺g⁺a/g⁻g⁻a interconversion which is considerably higher than the barrier for the related process in [Cr(CO)₅(1a)] (barrier ca. 40 kJ mol⁻¹).² From inspection of molecular models, it is apparent that the high barrier to P–C rotation observed in the square planar complexes reported here, arises from the potential energy trough created by the bulky ortho-substituents occupying the voids above and below the square plane as well as the potential energy peaks associated with the bulky ortho-substituents passing the cis M–Cl groups.

The four isomers observed at low temperatures in the ³¹P NMR spectra of 2b and 3b (see above) may be due to arrested M–P rotation. As shown in Scheme 4, the meso/rac-isomers, could then form two rotamers (with the P atoms equivalent in each) associated with their relative orientations across the P–M–P axis: staggered (i.e. meso-A and rac-A) and eclipsed (i.e. meso-B and rac-B). In the solid state, species of this type have been shown to adopt meso-A (see above), rac-A and rac-B conformations.¹⁰ From the variable temperature ³¹P NMR spectra (T_c 263 K), the energy barrier for these M–P bond rotation processes is estimated to be ca. 55 (±2) kJ mol⁻¹. Restricted M–PR₃ rotations are common for bulky phosphines and barriers of up to 58 kJ mol⁻¹ have been previously reported for square planar complexes of the type trans-[RhCl(CO)(PR₃)₂].¹¹

Scheme 4 Proposed interconversion of meso and rac rotamers viewed along the P–M–P axis (idealised structures).

The ³¹P NMR spectrum of 2c at room temperature showed two singlets at δ 26.4 and 24.7 (intensity ratio 3 : 1) which were assigned to trans-PtP₂Cl₂ species on the basis of the ¹J(PtP) values of 2555 and 2603 Hz. The ¹H NMR spectrum of 2c was sharp and showed the signals for a 3 : 1 mixture of similar species. There were 6 major and 6 minor doublets for the CH(CH₃)₂ groups along with 3 major and 3 minor multiplets for the CH(CH₃)₂ groups showing the inequivalence of the isopropyl groups. Two deshielded signals at 9.25 and 8.74 ppm (3 : 1) were also evident. These data are consistent with the presence of rac and meso rotamers analogous to the structures discussed above for 2b but significantly, these are static on the NMR timescale at room temperature. The P–C rotation process is remarkably slow in 2c since significant broadening of the ¹H and ³¹P signals is observed only above +75 °C. From a T_c of 373 K observed in the ¹H NMR spectra, the barrier to rotation is estimated to be 80 (±2) kJ mol⁻¹. In contrast to 2b, the ³¹P and ¹H NMR spectra of 2c in CH₂Cl₂ were the same at +23 °C and −80 °C and no signals for additional rotamers of the type meso/rac-A and meso/rac-B (Scheme 4) were resolved. From consideration of the greater steric bulk in 2c than 2b, it can be understood why one rotamer (e.g. meso-A) would be preferred in solution.

Binuclear chloroplatinum and chloropalladium complexes

Treatment of [PdCl₂(NCPh)₂] with 1 equivalent of 1b in refluxing toluene gave a precipitate of the binuclear 4b (see Experimental section for characterising data).

Phosphines 1c and 1d react with [PdCl₂(NCPh)₂] to give binuclear complexes 4c and 4d regardless of the Pd : PR₃ ratio. These products were characterised by a combination of elemental analysis, FAB mass spectrometry, IR spectroscopy (see Experimental section), ³¹P and ¹H NMR spectroscopy.

The crystal structure of 4c has been determined as its chloroform solvate. Molecules of trans-[Pd₂Cl₄(1c)₂] (4c) have crystallographic C_i symmetry (see Fig. 5 and Table 3 which lists selected bond lengths and angles). The average P–C bond length in 4c is 1.83 Å, with an average C–P–C bond angle of 107.4° and mean Pt–P–C bond angle 111.4°. Ligand 1c has a cone angle of 208° in this structure in which it adopts a conformation at P(1) with Pd–P–C_ipso–C_ortho torsion angles of −54.5, −64.2, −178.69°, i.e. a g⁻g⁻a conformation. The shortest Pd⋯H contact is 2.77 Å and involves a CH(CH₃)₂ hydrogen on a gauche aryl ring; the ortho-hydrogen on the aryl group that adopts the anti conformation is at 2.81 Å from the palladium. The (CH₃)₂C–H⋯Pd contacts involving the isopropyl groups on the gauche aryl rings are at marginally longer Pd⋯H distances (2.96 and 3.04 Å). The structure, and in particular the conformation of its triarylphosphine, is similar to those of complexes in the CSD with refcodes ELENIS¹² and POBVEH¹³ which are both complexes of the form trans-[Pd₂(µ-Br)₂(aryl)(1a)₂] and are meso-diastereoisomers having tri-o-tolylphosphine ligands in g⁺g⁺a and g⁻g⁻a conformations.

Table 3 Selected bond lengths (Å) and angles (°) for trans-[Pd₂Cl₄(1c)₂] (4c)·4CHCl₃

Pd(1)–Cl(1)	2.319(2)	Pd(1)–Cl(1A)	2.403(2)
Pd(1)–Cl(2)	2.300(2)	Pd(1)–P(1)	2.244(2)
Cl(1)–Pd(1)–P(1)	94.53(8)	Cl(1A)–Pd(1)–Cl(1)	85.22(7)
Cl(1A)–Pd(1)–P(1)	177.50(9)	Cl(2)–Pd(1)–Cl(1A)	90.19(8)
Cl(2)–Pd(1)–P(1)	89.83(8)	Cl(1)–Pd(1)–Cl(2)	172.95(8)

---

Fig. 5 X-Ray crystal structure of trans-[Pd₂Cl₄(1c)₂] (4c). Hydrogen atoms are omitted for clarity.

Similar spectroscopic (IR and NMR) properties are observed for 4c and 4d. The ³¹P NMR spectrum of 4d at +23 °C showed two species are present in a ca. 1 : 1 ratio: a singlet at 28.2 and two doublets at δ 27.9 and 28.7 with a small J(PP) of 6 Hz. This suggests that the two products are similar and in one of them, the P atoms are inequivalent. The ³¹P NMR spectrum of 4c similarly showed the presence of 2 species at +23 °C, the signals for which merged to a singlet at +75 °C. The ¹H NMR (COSY) spectrum of 4c at +23 °C showed the presence of 6 major doublets and 12 minor doublets for the CH(CH₃)₂ groups; some of the doublet resonances were to low frequency of TMS and therefore highly shielded. Three multiplets in the region 9–9.5 ppm are present integrating for 1 : 2 : 1. At higher temperatures, the ¹H NMR signals broaden and by +100 °C some of the signals have coalesced. In particular, the three high frequency signals become a single broad resonance (w_1/2 45 Hz) at 9.3 ppm, the CH(CH₃)₂ signals become 2 signals at 3.8 and 2.5 ppm (intensity ratio 2 : 1), and the CH(CH₃)₂ signals become 4 broad signals at 2.2, 1.5, 1.1 and 0.0 ppm (in intensity ratios 3 : 6 : 3 : 6). It is clear that even at 100 °C, there is not rapid P–C bond rotation on the NMR timescale. This behaviour is again interpreted in terms of meso (C_i, as observed in the solid state) and putative rac (C₁) rotamers; the P atoms are equivalent in meso-4c and inequivalent in rac-4c in agreement with the NMR observations. The barrier to rotation in 4c was calculated (T_c 375 K) to be 74 (±2) kJ mol⁻¹ which is similar to that in 2d and therefore supports the idea that the main impediment to P–C rotation is the passage of the ortho-substituent past the adjacent Pt–Cl groups.

When crystals of meso-4c were dissolved in CDCl₃, a mixture of meso- and rac-4c was observed by ³¹P NMR spectroscopy. This indicates that while rac–meso interconversion is not evident on the NMR timescale at ambient temperatures, the isomers do interconvert on the timescale of minutes to form equilibrium mixtures.

Tri(o-tolylphosphine) (1a) forms cyclometallates with palladium^14,15 and platinum¹⁶ and this prompted us to attempt cyclometallations with 1b–d. Treatment of [PtCl₂(NCBu^t)₂] with 1 equiv. of 1b in refluxing toluene for 4 h gave a mixture of several products, one of which (δ −4.5 and J(PtP) 4093 Hz) was assigned to the binuclear [Pt₂Cl₄(1b)₂]. After refluxing the solution for a further 24 h, two products (ratio 2 : 1) were observed by ³¹P NMR spectroscopy and were assigned to the binuclear platinacycles trans-5b and cis-5b, principally by NMR spectroscopy (see Experimental section).

Particularly diagnostic are the very large ¹J(PtP) values of 4993 and 5046 Hz and a broad ¹H signal at 3.19 ppm for the Pt–CH₂ groups with a J(PtH) value of 108 Hz, as expected.¹⁶ The ¹H NMR spectrum (in CD₂Cl₂) shows broad signals for the tolyl-CH₃ (at 2.64 ppm, w_1/2 60 Hz) and one of the aromatic protons (at 6.95 ppm, w_1/2 30 Hz). In C₆D₅CD₃, the Pt–CH₂ resonance is shifted ca. 0.5 ppm to high frequency; at temperatures above +50 °C, the Pt–CH₂ signal is resolved into two singlets (1 : 2) at 3.74 (J(PtH) 114 Hz) and 3.67 ppm J(PtH) 108 Hz) and there are two sharp singlets for the tolyl-CH₃. This high temperature behaviour shows that cis- and trans-5b do not interconvert on the NMR timescale. At −60 °C in CD₂Cl₂, the ³¹P NMR spectrum showed the presence of 4 singlets (in the ratio of ca. 1 : 1 : 1 : 2) and the ¹H NMR spectrum showed 4 pairs of methyl singlets and complex overlapping signals for the Pt–CH₂ group. The low temperature NMR spectra imply there are four species present which we assign to rac- and meso-rotamers of cis- and trans-5b (see below).

In the 2 : 1 reaction of 1b : Pt described earlier, the only product observed was the mononuclear trans species 2b which implies that cyclometallation to 5b is inhibited by the presence of the phosphine.

As noted above, treatment of [PdCl₂(NCPh)₂] with 1b gave the binuclear 4b under conditions similar to those which produced the cycloplatinated complex 5b. This indicates that cycloplatination occurs more readily than cyclopalladation in these complexes; similar observations have been made by others.¹⁶ However, in the toluene filtrate of the reaction mixture to produce 4b, a minor species was observed which had ¹H signals consistent with a cyclopalladated analogue of 5b (see Experimental section).

The 19 cyclometallated tri-o-tolylphosphine structures in the CSD show conformations that are in a number of respects similar to the non-cyclometallated species except that the torsion angle ω₃ of the metallated aryl is perforce close to zero (typically 0–25°) and the other two are g⁺g⁺ (typically ω₁ = 30–40° and ω₂ = 60–80°), as shown in Scheme 5. Only when the metal is 6-coordinate is a g⁺a (ca. 80° and −160°) conformation observed.

Scheme 5 Chiral g⁺g⁺, g⁻g⁻, g⁺a and g⁻a conformations of cyclometallated tri-o-tolylphosphine ligands.

In any event, in such species there is helical chirality in the cyclometallated species. Therefore rac and meso forms of bis-cyclometallated species are to be expected. Indeed in the crystal structure of [Pd₂(μ-I)₂{1a − H}₂] (where 1a − H is cyclometallated 1a), the meso diastereoisomer is observed.¹⁵

No reaction was observed between [PtCl₂(NCBu^t)₂] and 1c or 1d at room temperature but in refluxing toluene, a product assigned the binuclear cyclometallated structure 6d is obtained. The structure of 6d is assigned on the basis of IR (ν(PtCl) 279, 248 cm⁻¹) and the characteristic ³¹P NMR spectrum (C₆D₅CD₃) which showed two singlets (1 : 1) at δ 14.0 and 13.9 with very large ¹J(PtP) of ca. 5275 Hz. Structure 6d contains a cyclometallated tertiary carbon atom which is rare (e.g. a few Pd¹⁷ and Rh¹⁸ metallocycles containing tertiary carbons have been reported). The two species present may be the cis/trans isomers of 6d or more likely, rac and meso isomers of trans-6d, which in this case would each contain equivalent P atoms.

Poor quality crystals of trans-6d were obtained and its crystal structure determined (see Fig. 6 and Table 4 which lists selected bond lengths and angles). The trans stereochemistry of 6d is established and the trans influence of the CMe₂ is markedly greater than the phosphine (Pt–Cl trans to C 2.480(4), Pt–Cl trans to P 2.416(4) Å). The average P–C bond length in 6d is 1.817 Å, with an average C–P–C bond angle of 105.5° and mean Pt–P–C bond angle of 113.0°. Cyclometallated ligand {1d − H} has a cone angle of 233° in this structure in which it adopts a conformation at P(1) with Pt–P–C_ipso–C_ortho torsion angles of −18.8, −46.5, and −65.8°, i.e. a g⁻g⁻ conformation. The shortest Pt⋯H contact is 2.72 Å and involves a CH(CH₃)₂ hydrogen on a gauche aryl ring.

Table 4 Selected bond lengths (Å) and angles (°) for trans-[Pt₂(μ-Cl)₂(1d − H)₂] (6d)

Pt(1)–C(11)	2.115(18)	Pt(1)–P(1)	2.177(3)
Pt(1)–Cl(1A)	2.416(4)
Cl(1A)–Pt(1)–P(1)	173.32(14)	C(11)–Pt(1)–Cl(1A)	97.6(4)
C(11)–Pt(1)–P(1)	82.5(4)	Cl(1)–Pt(1)–C(11)	167.3(4)
Cl(1A)–Pt(1)–Cl(1)	79.65(13)	Pt(1)–Cl(1)–Pt(1)a	100.35(13)

---

Fig. 6 X-Ray crystal structure of trans-[Pt₂Cl₂(1d − H)₂](6d). Hydrogen atoms are omitted for clarity.

Some features of interest emerge when the structure of 6d is compared with the related cyclometallated trimesitylphosphine species 7 (Ar = 2,4,6-trimethylphenyl, CSD refcode VANHEB).¹⁹ The Pt–CMe₂ distance (2.12(2) Å) in 6d is slightly longer than the corresponding Pt–CH₂ distance (2.035(4) Å) in 7 presumably as a consequence of the greater steric crowding in 6d. Despite this, the Pt–Cl (trans to C) distance in 6d of 2.480(4) Å is similar to the corresponding Pt–Cl distance of 2.474(1) Å in 7 which indicates that the tertiary alkyl in 6d has a similar trans influence to the primary alkyl in 7. Furthermore the Pt–P distances (2.177(3) Å in 6d, 2.210(1) in 7) and Pt–Cl trans to P distances (2.416(4) Å in 6d, 2.386(1) in 7) indicate that the phosphorus in 6d is more strongly bonded to platinum and has a higher trans influence than the phosphorus in 7. This is also consistent with the ¹J(PtP) for 6d, being ca. 150 Hz larger than for 7.¹⁹

The reaction of [PtCl₂(NCBu^t)₂] with 1c in refluxing toluene gave an off-white precipitate which was so insoluble in common organic solvents that characterisation by NMR was not possible; the product is tentatively assigned the binuclear cyclometallate structure 6c on the basis of elemental analysis, the FAB mass spectrum and the IR spectrum (see Experimental section). The ³¹P NMR spectrum of the toluene filtrate, showed the presence of the mononuclear species 2c; crystals of 2c were obtained from this solution and its structure is discussed above.

Attempts to cyclopalladate 1c or 1d under the conditions used for cycloplatination, gave only the unmetallated binuclear complexes 4c and 4d.

Conclusion

The coordination chemistry of bulky PAr₃ ligands 1a–d was investigated in the hope of shedding light on the catalytic activating effect we have observed with ortho-substituted phenyl phosphines, which is especially pronounced with ligands containing o-isopropylphenyl groups.¹ Not surprisingly, complexes of the type [MCl₂(1a–d)₂] all have trans-geometry and for the very bulky 1c,d, there is a tendency to form binuclear complexes of the type trans-[M₂Cl₂(μ-Cl)₂(1c,d)₂]. Steric congestion within the coordination sphere leads to the highly restricted P–C and P–M rotation apparent from the NMR studies. The P–C rotation is so slow in complexes of the ligands containing o-isopropylphenyl groups that diastereomeric rotamers associated with the chiral conformations of the PAr₃ groups are observed even at elevated temperatures. To our knowledge, this is the first time diastereoisomerism of this type has been observed at room temperature and above. Although P–C rotation was observed in these systems on the laboratory timescale (i.e. minutes), molecular rigidity would prevail on a reaction timescale and therefore may have consequences in catalysis. The crystal structure of the cyclometallated species 6d revealed some unusual bonding features which may also be significant in understanding the special effect of o-isopropylphenyl groups.

Experimental

Unless otherwise stated, all work was carried out under a dry nitrogen atmosphere, using standard Schlenk line techniques. Dry N₂-saturated solvents were collected from a Grubbs system²⁰ in flame and vacuum dried glassware. 4-Bromo-3-isopropylanisole,²¹ [PtCl₂(NCBu^t)₂]²² and [PdCl₂(NCPh)₂]²³ were prepared by literature methods. Tri(o-tolyl)phosphine (1a) was purchased from Strem and all other starting materials were purchased from Aldrich. Ligand 1c^2c and complexes 2a and 3a¹¹ have been previously reported. NMR spectra were measured on a Jeol Eclipse 300, Jeol Eclipse 400 or Jeol GX 400 spectrometer. Unless otherwise stated ¹H, ¹³C and ³¹P NMR spectra were recorded at 300, 100, and 121 MHz respectively at +23 °C. Mass spectra were recorded on a Fisons MD800. Infrared spectra were recorded on a Perkin Elmer Spectrum 1 spectrometer as Nujol mulls between polythene plates. Elemental analyses were carried out by the Microanalytical Laboratory of the School of Chemistry, University of Bristol.

Syntheses

Tri(p-methoxy-o-tolyl)phosphine (1b). 4-Bromo-3-methylanisole (16.02 g, 79 mmol) was added to Mg turnings (1.92 g, 79 mmol) in THF (15 cm³) over 5 min. The mixture was heated at reflux for 2 h until the Mg turnings had disappeared. The grey solution was then allowed to cool to room temperature before the dropwise addition of PCl₃ (3.62 g, 26 mmol) at 0 °C. The reaction was stirred overnight and then 2 M NH₄Cl solution (100 cm³) was added and the mixture stirred for 30 min. The mixture was then extracted with Et₂O (3 × 30 cm³) and the ethereal extracts combined and dried over MgSO₄. The solution was filtered, stirred with activated charcoal for 20 min and filtered through Florisil. The solvents were removed in vacuo to yield an off white solid, which was digested with ethanol (3 × 20 cm³) to leave a white precipitate. The supernatant ethanol was then removed with a cannula before the powder was washed with cold ethanol (2 × 20 cm³) and the product 1b was dried in vacuo to leave a fine white powder (6.00 g, 0.015 mol, 58%). Elemental analysis (calc): C, 72.6 (73.1); H, 6.45 (6.90%); MS (CI) m/z: 394 (M⁺); ³¹P{¹H} (CDCl₃): δ = −34.8; ¹H NMR (CDCl₃): δ = 6.70 (d, 1H), 6.52 (m, 2H,), 3.63 (s, 3H,), 2.25 (s, 3H); ¹³C NMR (CDCl₃): δ = 160.2 (s), 144.3 (d), 134.3 (s), 126.3 (d), 115.8 (d), 111.5 (s), 55.1 (s), 21.2 (d).

Tri-(o-isopropyl-p-methoxyphenyl)phosphine (1d). Complex 1d was prepared in a similar fashion to 1b from 4-bromo-3-isopropylanisole to give a white powder (2.14 g, 4.47 mmol, 53%). Elemental analysis (calc): C, 74.90 (75.29); H, 8.60 (8.21%); MS (CI) m/z: 478 (M⁺); ³¹P{¹H} (CDCl₃): δ = −42.6; ¹H NMR (CDCl₃): δ = 6.81 (1H, dd), 6.56 (2H, dd), 3.72 (3H, s), 3.56 (1H, septet), 1.03 (6H, d); ¹³C NMR (CDCl₃): δ = 160.3 (s), 155.1 (d), 135.5 (s), 126.6 (s), 111.6 (s), 110.8 (s), 55.0 (s), 31.2 (s), 23.9 (s).

trans-[PtCl₂(1a)₂] (2a). Tri-o-tolylphosphine (1a) (0.301 g, 0.99 mmol) and [PtCl₂(NCBu^t)₂] (0.215 g, 0.494 mmol) were dissolved in toluene (10 cm³), heated at 90 °C for 6 h and then cooled to room temperature. The solvent was removed and the remaining white solid was washed with cold toluene (2 × 10 cm³) before drying in vacuo to give 2a as a white powder (0.195 g, 0.22 mmol, 45%). Satisfactory elemental analyses were not obtained and the insolubility of the product precluded purifications by recrystallisation (calc): C, 56.51 (57.67); H, 4.87 (4.84%); IR: ν(Pt–Cl) 341 cm⁻¹ in agreement with literature value.¹¹

trans-[PtCl₂(1b)₂] (2b). Complex 2b as a yellow solid (0.303 g, 0.29 mmol, 56%) was prepared in a similar fashion to 2a from ligand 1b. Elemental analysis (calc): C, 54.73 (54.65); H, 5.37 (5.16%); MS (FAB) m/z = 1018 (M⁺ − Cl), 983 (M⁺ − 2 Cl); IR: ν(Pt–Cl) 342 cm⁻¹; ³¹P{¹H} NMR (CD₂Cl₂): δ = 15.84 (w_1/2 90 Hz, ¹J(PtP) 2570); ³¹P{¹H} NMR (CD₂Cl₂, −40 °C): δ = 16.08 (¹J(PtP) 2562), 15.49 (¹J(PtP) 2547), 14.24 (¹J(PtP) ≈2500), 14.17 (¹J(PtP) 2522); ¹H NMR (CD₂Cl₂): δ = 8.80 (br), 7.05 (br), 6.59 (br) 3.72 (br), 2.95 (br), 1.92 (br), 1.50 (br); ¹H NMR (CD₂Cl₂, −60 °C): δ = 9.23 (q), 9.08 (q), 8.85 (q), 8.55 (q), 7.20–6.51 (m), 3.81 (s), 3.76 (s), 3.75 (s), 3.74 (s), 3.66 (s), 3.21 (s), 2.99 (s), 2.72 (s), 2.57 (s), 2.26 (s), 2.20 (s), 2.07 (s), 1.80 (s), 1.69 (s), 1.49 (s), 1.46 (s), 1.43 (s).

trans-[PtCl₂(1c)₂] (2c). Complex 2c was formed as a by-product during the preparation of 6c (see below). IR: ν(Pt–Cl) 348, 337 cm⁻¹; ³¹P{¹H} NMR (CD₂Cl₂): δ = 26.43 (¹J(PtP) 2555), 24.72 (¹J(PtP) 2603); ¹H NMR (CD₂Cl₂): δ = 9.25 (q), 8.74 (q), 7.7–6.9 (m), 4.20 (m), 4.0 (m), 3.8 (m), 3.32 (m), 2.92 (quintet), 2.61 (quintet), 1.72 (d), 1.56 (d), 1.43 (d), 1.25 (d), 1.20(d), 1.08 (d), 0.89 (d), 0.25 (d), 0.06 (d), −0.15 (d), −0.24 (d), −0.30 (d).

trans-[PdCl₂(1a)₂] (3a). Tri-o-tolylphosphine (1a) (0.718 g, 2.35 mmol) and [PdCl₂(NCPh)₂] (0.450 g, 1.18 mmol) were mixed in toluene (70 cm³) and the suspension heated at reflux for 21 h and then allowed to cool to room temperature to yield a yellow precipitate. The solvent was removed and the resulting solid triturated with cold toluene, filtered off, washed with cold toluene (2 × 20 cm³) and then dried in vacuo to give 3a as a yellow powder (0.61 g, 0.776 mmol, 66%). Elemental analysis (calc): C, 64.30 (64.18); H, 5.25 (5.39%); MS (FAB) m/z = 788 (M⁺), 751 (M⁺ − Cl), 713 (M⁺ − 2 Cl); IR: ν(Pd–Cl) 354 cm⁻¹. The insolubility of 3a precluded characterization by NMR spectroscopy.¹¹

trans-[PdCl₂(1b)₂] (3b). Complex 3b as a yellow solid (0.179 g, 0.19 mmol, 46%) was prepared in a similar fashion to 3a from ligand 1b. Satisfactory elemental analyses were not obtained (calc): C, 55.10 (59.67); H, 5.53 (5.63%); IR: ν(Pd–Cl) 349 cm⁻¹; ³¹P{¹H} NMR (CD₂Cl₂): δ = 18.80 (w_1/2 85 Hz); ³¹P{¹H} NMR (CD₂Cl₂, −80 °C): δ = 19.17, 18.46, 17.64, 17.18;¹H NMR (CD₂Cl₂): δ = 8.79 (br), 7.02 (br), 6.65 (br), 3.72 (br), 2.30 (br), 1.70 (br); ¹H NMR (CD₂Cl₂, −80 °C): δ = 9.12 (q), 9.00 (q), 8.77 (q), 8.49 (q), 7.20–6.52 (m), 3.83 (s), 3.76 (s), 3.67 (s), 3.64 (d), 3.62 (d), 3.59 (d), 3.21 (s), 3.09 (s), 2.68 (s), 2.56 (s), 2.27 (s), 2.25 (d), 2.16 (s), 2.14 (s), 2.04 (s), 1.74 (m), 1.65 (s), 1.44 (s), 1.40 (s).

[Pd₂Cl₄(1b)₂] (4b). Tri(p-methoxy-o-tolyl)phosphine (1b) (0.128 g, 0.32 mmol) and [PdCl₂(NCPh)₂] (0.124 g, 0.32 mmol) were dissolved in toluene (10 cm³), heated at reflux for 18 h and then cooled to room temperature to give an orange precipitate and a yellow solution. The solution, containing a cyclopalladate analogue of 5b (see below), was separated by filtration and the precipitate was washed with toluene (2 × 1 cm³) and then dried in vacuo to give 4b as an orange powder (110 mg, 0.096 mmol, 60%). Elemental analysis (calc): C, 50.39 (50.42); H, 4.62 (4.76%); IR: ν(Pd–Cl) 346, 294, 263 cm⁻¹; ³¹P{¹H} NMR (CD₂Cl₂): δ = 24.1 and 22.4; ¹H NMR (CD₂Cl₂): δ = 9.10 (br), 7.00 (br), 6.61 (br), 3.76 (br), 3.38 (br), 1.98 (br), 1.45 (br). The toluene filtrate was reduced to dryness to give an oily yellow solid (60 mg, 0.056 mmol, 35%) which was tentatively assigned to the cyclopalladate analogue of 5b on the basis of the following NMR data. ³¹P{¹H} NMR (CD₂Cl₂): δ = 34.87, 35.48. ¹H NMR (CD₂Cl₂): δ = 7.55 (m), 7.40 (m), 7.11 (m), 6.89 (m), 6.65 (m), 3.71 (m), 3.33 (br), 2.60 (br, m), 2.32 (s), 2.25 (s).

[Pd₂Cl₄(1c)₂] (4c). Tri-o-isopropylphenylphosphine (1c) (0.051 g, 0.13 mmol) and [PdCl₂(NCPh)₂] (0.050 g, 0.13 mmol) were dissolved in toluene (5 cm³) and stirred at room temperature for 1 h to yield an orange precipitate. The product was filtered off, washed with cold toluene (2 × 20 cm³) and then dried in vacuo to give 4c as an orange powder (0.042 g, 0.04 mmol, 57%). The product can be recrystallised from CHCl₃. Elemental analysis (presence of solvent confirmed by ¹H NMR) (calc 4c·4CHCl₃): C, 43.59 (43.29); H, 4.65 (4.38%); MS (FAB) m/z = 1098 (M⁺ − Cl), 1061 (M⁺ − 2 Cl), 1025 (M⁺ − 3 Cl); IR: ν(Pd–Cl) 354, 297, 264 cm⁻¹; ³¹P{¹H} NMR (CDCl₃): δ = 33.04 (br), 32.39 (s), 32.13 (br), the broadness is interpreted as due to the small (unresolved) J(PP); ³¹P{¹H} NMR (C₂D₂Cl₄, +100 °C): δ = 32.17; ¹H NMR (CDCl₃): δ = 9.34 (dd), 9.24 (dd), 9.16 (dd), 7.75 (m), 7.60 (m), 7.39 (m), 7.18 (m), 4.10 (br), 3.96 (br, m), 3.74 (br), 3.49 (br, m), 2.60 (d), 2.48 (br, m), 2.27 (d), 1.97 (d), 1.90 (br), 1.77 (br), 1.57 (m), 1.45 (d), 1.40 (d), 1.29 (d), 1.02 (d), −0.05 (d), −0.12 (d), −0.15 (m); note that from the off-diagonal peaks in the¹H COSY NMR all 18 CH(CH₃)₂ signals (in the region 2.60 to −0.15) and all 9 CH(CH₃)₂ signals (in the region 4.10 to 2.48) were identified which were overlapping in the 1D spectrum. ¹H NMR (C₂D₂Cl₄, +100 °C): δ = 9.32 (br), 7.53 (br), 7.14 (br), 3.80 (br, d), 2.49 (br), 2.25 (br), 1.48 (br), 1.09 (br), 0.00 (br).

[Pd₂Cl₄(1d)₂] (4d). Complex 4d was prepared in a similar fashion to 4c from ligand 1d except that hexane was added to induce precipitation from toluene to give the orange solid product (0.08 g, 0.061 mmol, 62%). Elemental analysis (presence of solvent confirmed by ¹H NMR) (calc 4d·0.5C₆H₅CH₃): C, 56.75 (56.17); H, 6.06 (6.09%); MS (FAB) m/z = 1276 (M⁺ − Cl), 1240 (M⁺ − 2 Cl); IR: ν(Pd–Cl) 344, 285, 252 cm⁻¹; ³¹P{¹H} NMR (CDCl₃): δ = 28.18 (s), 28.72 (d, ⁴J(PP) = 6 Hz), 27.95 (d, ⁴J(PP) = 6 Hz); ¹H NMR (CDCl₃): δ = 9.17 (m), 7.23 (m), 7.19 (m), 6.95 (d), 6.67 (m), 3.86 (m), 3.74 (d), 3.51 (m), 2.55 (d), 2.48 (m), 2.25 (d), 1.95 (d), 1.53 (d), 1.40 (d), 1.29 (d), 1.01 (d), 0.05 (d), 0.00 (d), −0.06 (d).

[Pt₂Cl₂(1b − H)₂] (5b). Tri(p-methoxy-o-tolyl)phosphine (1b) (0.347 g, 0.879 mmol) and [PtCl₂(NCBu^t)₂] (0.380 g, 0.879 mmol) were dissolved in toluene (15 cm³), heated at reflux for 22 h and then cooled to room temperature. The dark brown solution was filtered through a 2 cm thick plug of Florasil to give a yellow solution. The solvent was then removed in vacuo to give the product 5b as a pale yellow powder (0.500 g, 0.401 mmol, 91%). Satisfactory elemental analyses were not obtained despite attempts to purify by recrystallisation (calc): C, 44.51 (46.20); H, 4.68 (4.20%); IR: ν(Pt–Cl) 278, 247 cm⁻¹. ³¹P{¹H} NMR (CD₂Cl₂): δ = 14.59 (s, ¹J(PtP) = 4993 Hz), 14.82 (s, ¹J(PtP) = 5046 Hz);³¹P{¹H} NMR (C₆D₅CD₃, +50 °C): δ = 14.95, 14.77; ³¹P{¹H} NMR (CD₂Cl₂, −40 °C): δ = 14.60, 14.13, 13.95, 13.82 (approx 1 : 1 : 1 : 2 ratio); ¹H NMR (CD₂Cl₂): δ = 7.12 (m), 6.95 (br), 6.74 (m), 6.66 (m), 3.71 (s), 3.19 (br, s, ²J(HPt) = 108 Hz), 2.64 (br), 2.48 (br);¹H NMR (C₆D₅CD₃, +50 °C): δ = 7.22 (m), 6.74 (m), 6.43 (m), 3.74 (s, ²J(HPt) 114 Hz), 3.67 (s, ²J(HPt) 108 Hz), 3.30 (m), 2.94 (s), 2.78 (s); ¹H NMR (CD₂Cl₂, −40 °C): δ = 7.15 (m), 6.98 (m), 6.80 (m), 6.59 (m), 3.74 (m), 3.16 (m), 2.81 (s), 2.70 (s), 2.68 (s), 2.65 (s), 2.55 (s), 2.46 (s), 2.36 (s), 2.35 (s); ¹³C NMR (75 MHz, CDCl₃): δ = 162.3 (s), 161.5 (s), 160.6 (d), 160.3 (d), 144.5 (m), 134.8 (d), 131.9 (d), 131.7 (d), 127.8 (d), 126.9 (d), 117.8 (d), 113.0 (d), 112.6 (d), 111.0 (m), 55.4 (s), 55.3 (s), 23.6 (br), 17.0 (s, CH₂), 15.8 (s, CH₂).

[Pt₂Cl₂(1c − H)₂] (6c). Tri-o-isopropylphenylphosphine (1c) (0.178 g, 0.458 mmol) and [PtCl₂(NCBu^t)₂] (0.100 g, 0.231 mmol) were dissolved in toluene (10 cm³), heated at reflux for 50 h and then cooled to room temperature to give a white solid and a pale yellow solution. The solution, containing 2c (see above), was removed by filtration and the resulting solid was washed with cold toluene (2 × 20 cm³) and then dried in vacuo to give 6c as a white powder (0.073 g, 0.059 mmol, 51%). Elemental analysis (calc): C, 52.37 (52.47); H, 5.25 (5.22%); MS (FAB) m/z = 1234 (M⁺), 1199 (M⁺ − Cl); IR: ν(Pt–Cl) 277, 246 cm⁻¹. The lack of solubility of 6c precluded characterisation by NMR spectroscopy.

[Pt₂Cl₂(1d − H)₂] (6d). Tri-(o-isopropyl-p-methoxyphenyl)phosphine (1d) (0.051 g, 0.107 mmol) and [PtCl₂(NCBu^t)₂] (0.023 g, 0.053 mmol) were dissolved in toluene (5 cm³), heated at reflux for 28 h and then cooled to room temperature. Hexane (20 cm³) was added and the solution was cooled in a freezer until a white precipitate appeared. The solid was filtered off, washed with cold hexane (2 × 2 cm³) and then dried in vacuo to give 1d as a white powder (0.026 g, 0.018 mmol, 69%). IR: ν(Pt–Cl) 279, 248 cm⁻¹; ³¹P{¹H} NMR (C₆D₅CD₃): δ = 14.03 (¹J(PtP) ≈ 5275 Hz), 13.90 (¹J(PtP) ≈ 5275 Hz).

X-Ray crystal structure analyses of 3b·4CH₂Cl₂, 2c·2CH₂Cl₂, 4c·4CHCl₃ and 6d

X-Ray diffraction experiments on 3b, 2c and 4c as their solvates (at −100 °C) and 6d (at room temperature) were carried out on Bruker CCD diffractometers using Mo-K_α X-radiation (λ = 0.71073 Å). Crystal and refinement data are given in Table 5. Absorption corrections were based on equivalent reflections and structures refined against all F_o² data with hydrogen atoms riding in calculated positions. Final difference maps showed no features of chemical significance with the largest features lying close to solvent molecules exhibiting unresolved disorder and/or the metal atoms.

CCDC reference numbers 254102–254105.

See http://www.rsc.org/suppdata/dt/b4/b416525j/ for crystallographic data in CIF or other electronic format.

Table 5 Crystal and refinement data for 3b·4CH₂Cl₂, 2c·2CH₂Cl₂, 4c·4CHCl₃ and 6d

Compound	3b·4CH2Cl2	2c·2CH2Cl2	4c·4CHCl3	6d
Colour, habit	Yellow block	Yellow plate	Orange block	Yellow plate
Size/mm	0.2 × 0.2 × 0.2	0.25 × 0.25 × 0.05	0.05 × 0.05 × 0.05	0.01 × 0.05 × 0.08
Formula	C52H62Cl10P2O6Pd	C56H70Cl6P2Pt	C58H70Cl16P2Pd2	C60H76Cl2P2O6Pt2
Formula weight	1305.86	1212.85	1609.08	1416.23
Crystal system	Triclinic	Triclinic	Triclinic	Triclinic
Space group (No.)	P1̄ (2)	P1̄ (2)	P1̄ (2)	P1̄ (2)
a/Å	11.035(2)	9.6403(19)	12.189(3)	11.359(3)
b/Å	12.020(2)	11.055(2)	12.331(3)	11.719(3)
c/Å	12.441(3)	14.148(3)	12.420(3)	13.130(6)
α/°	63.49(3)	97.94(3)	90.563(4)	101.28(3)
β/°	77.64(3)	95.31(3)	94.222(4)	104.34(2)
γ/°	89.59(3)	113.82(3)	111.771(4)	112.02(3)
V/Å^3	1435.2(5)	1564.7(7)	1727.5(7)	1486.5(9)
Z	1	1	1	1
μ/mm^−1	0.891	2.996	1.221	4.891
Unique data	6547	6182	7821	6304
R int	0.0494	0.0387	0.1052	0.0677
Final R1 [I > 2σ(I)]	0.0582	0.0338	0.0709	0.0654

---

Acknowledgements

We would like to thank Martin Murray of University of Bristol for helpful discussions of the NMR spectra, EPSRC for support, Johnson Matthey for an Industrial CASE award (to M. J. W.) and The Leverhulme Trust for a Research Fellowship (to P. G. P.).

References and notes

1. For example see: (a) S. J. Dossett, A. Gillon, A. G. Orpen, J. S. Fleming, P. G. Pringle, D. F. Wass and M. Jones, Chem. Commun., 2001, 699; (b) N. A. Cooley, S. M. Green, D. F. Wass, K. Heslop, A. G. Orpen and P. G. Pringle, Organometallics, 2001, 20, 4769; (c) J. Dennett, A. L. Gillon, K. Heslop, D. J. Hyett, J. S. Fleming, E. Lloyd-Jones, A. G. Orpen, J. Scutt, R. H. Weatherhead, P. G. Pringle and D. F. Wass, Organometallics, 2004, 23, 6077; (d) A. H. Roy and J. F. Hartwig, Organometallics, 2004, 23, 1533 and refs. therein.
2. (a) J. A. S. Howell, M. G. Palin, P. C. Yates, P. McArdle, D. Cunningham, Z. Goldschmidt, H. E. Gottlieb and D. Hezroni-Langerman, J. Chem. Soc., Perkin Trans. 2, 1992, 1769; (b) J. A. S. Howell, M. G. Palin, P. McArdle, D. Cunningham, Z. Goldschmidt, H. E. Gottlieb and D. Hezroni-Langerman, Inorg. Chem., 1993, 32, 3493; (c) J. A. S. Howell, P. C. Yates, M. G. Palin, P. McArdle, D. Cunningham, Z. Goldschmidt, H. E. Gottlieb and D. Hezronilangerman, J. Chem. Soc., Dalton Trans., 1993, 2775; (d) J. A. S. Howell, J. D. Lovatt, P. McArdle, D. Cunningham, E. Maimone, H. E. Gottlieb and Z. Goldschmidt, Inorg. Chem. Comm., 1998, 1, 118; (e) J. A. S. Howell, N. Fey, J. D. Lovatt, P. C. Yates, P. McArdle, D. Cunningham, E. Sadeh, H. E. Gottlieb, Z. Goldschmidt, M. B. Hursthouse and M. E. Light, J. Chem. Soc., Dalton Trans., 1999, 3015.
3. (a) S. D. Chappell and D. J. Cole-Hamilton, J. Chem. Soc., Dalton Trans., 1993, 533; (b) J. Fornies, A. Martin, R. Navarro, V. Sicilia and P. Villarroya, Organometallics, 1996, 15, 1826; (c) M. R. Whitnall, K. K. M. Hii, M. Thornton-Pett and T. P. Kee, J. Organomet. Chem., 1997, 529, 35.
4. A. K. Colter, R. J. Kurland and I. I. Schuster, J. Am. Chem. Soc., 1965, 87, 2279.
5. M. B. Smith, A. G. Orpen, P. G. Pringle and K. Worboys, J. Organomet. Chem., 1998, 550, 255.
6. M. A. Bennett and P. A. Longstaff, J. Am. Chem. Soc., 1969, 91, 6266.
7. E. C. Alyea, S. A. Dias, G. Ferguson and P. J. Roberts, J. Chem. Soc., Dalton Trans., 1979, 948.
8. F. H. Allen and O. Kennard, Chem. Des. Autom. News, 1993, 8, 1 and 31.
9. J. Vicente, J. A. Abad, E. Martinez-Viviente and M. C. R. Arellano, Private Communication to the Cambridge Structural Database, 1997.
10. We have recently determined the crystal structures of [PtCl₂(1c)] as a chloroform solvate and [PdCl₂(AsAr₃)₂] (Ar = C₆H₃(o-Prⁱ)(p-OMe)) which have rac-A and rac-B conformations respectively (see subsequent papers).
11. C. H. Bushweller, S. Hoogasian, A. D. English, J. S. Miller and M. Lourandos, Inorg, Chem., 1981, 20, 3448 and refs. therein.
12. F. Paul, J. Patt and J. F. Hartwig, J. Am. Chem. Soc., 1994, 116, 5969.
13. C. Bartolome, P. Espinet, J. M. Martin-Alvarez and F. Villafane, Eur. J. Inorg. Chem., 2003, 3127.
14. W. A. Herrmann, C. Brossmer, K. Öfele, C. P. Reisinger, T. Priermeier, M. Beller and H. Fischer, Angew. Chem. Int. Ed., 1995, 34, 1844.
15. A. L. Rheingold and W. C. Fultz, Organometallics, 1984, 3, 1414.
16. (a) S. D. Chappell and D. J. Cole-Hamilton, J. Chem. Soc., Dalton Trans., 1983, 1051; (b) J. Forniés, A. Martín, V. Sicilia and P. Villarroya, Organometallics, 1996, 15, 1826; (c) S. A. Dias and E. C. Alyea, Transition Met. Chem., 1979, 4, 205.
17. (a) F. Balegroune, D. Grandjean, D. Lakkisb and D. Matt, J. Chem. Soc., Chem. Commun., 1992, 1084; (b) P. Kocovsky, S. Vyskocil, I. Cisarova, J. Sejbal, I. Tislerova, M. Smrcina, G. C. Lloyd-Jones, S. C. Stephen, C. P. Butts, M. Murray and V. Langer, J. Am. Chem. Soc., 1999, 121, 7714; (c) P. Dotta, P. G. A. Kumar and P. S. Pregosin, Organometallics, 2003, 22, 5345.
18. (a) A. Vigalok, H.-B. Kraatz, L. Konstantinovsky and D. Milstein, Chem. Eur. J., 1997, 4, 253; (b) A. Vigalok, B. Rybtchinski, L. J. W. Shimon, Y. Ben-David and D. Milstein, Organometallics, 1999, 18, 895.
19. (a) E. C. Alyea, G. Fergusen, J. Malito and B. L. Ruhl, Organometallics, 1989, 8, 1188; (b) E. C. Alyea and J. Malito, J. Organomet. Chem., 1988, 340, 119.
20. A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen and F. J. Timmers, Organometallics, 1996, 15, 1518.
21. (a) N. J. Hales, H. Heeney, J. H. Holinshead and S. V. Ley, Tetrahedron, 1995, 51, 7741; (b) H. Konishi, K. Aritomi, T. Okano and J. Kiji, Bull. Chem. Soc. Jpn., 1989, 62, 591.
22. D. Fraccarollo, R. Bertani, M. Mozzon, U. Belluco and R. Michelin, Inorg. Chim. Acta, 1992, 201, 15.
23. J. R. Doyle, P. E. Slade and H. B. Jonassen, Inorg. Synth., 1960, 6, 218.

---

Footnote

† The barriers (ΔG^‡) to rotation given in this paper were calculated from the expression ΔG^‡ = 0.0194T_c (9.972 + log{T_c/δν}) where T_c is the coalescence temperature of signals separated by δν in the ¹H or ³¹P NMR spectra. The error given in brackets allows for an error in T_c of up to ±10 °C (see J. Sandström, Chapter 7 in Dynamic NMR Spectroscopy, Academic Press, 1982).

---