trans-Carbonylchloridobis(triethylphosphane-κP)platinum(II) tetrafluoridoborate

Crystallographic details are given of an unexpected product − a platinum(II) carbonyl complex, in which the carbonyl oxygen atom has remarkably been extracted from Pyrex glassware by tetrafluoroethylene. Preliminary details were reported previously.


Structure description
A low-yield product in the reaction of trans-PtHCl(P(C 2 H 5 ) 3 ) 2 with C 2 F 4 in the absence of air was originally postulated to be a five-coordinate platinum complex, PtHCl(-C 2 F 4 )(P(C 2 H 5 ) 3 ) 2 (Clark & Tsang, 1967), and the crystal-structure determination was undertaken at that time in view of the then current interest in five-coordination and ofcomplexes. As described in , the preliminary crystal-structure model showed no evidence of five-coordination, nor of the presence of a -bonded tetrafluoroethylene group. Instead, a four-coordinated, cationic Pt II complex was indicated, with a carbonyl group as the fourth ligand, isoelectronic with Vaska's compound, IrCl(CO)(PR 3 ) 2 (Vaska & DiLuzio, 1961). The presence of a carbonyl group was completely unexpected, as the reaction had been carried out in a vacuum line, in the absence of oxygen. This was the first reported molecular structure of a platinum carbonyl at the time, according to our database analysis below. The strong carbonyl vibrational band in the infrared spectrum was mistaken for the anticipated Pt-H band. Evidently, the carbonyl oxygen atom had been extracted from the Pyrex glassware by the tetrafluoroethylene reagent. That reaction vessels are not always as inert as they are expected to be is the subject of a recent review by Nielsen & Pedersen (2022) in which formation of data reports the title compound in this paper is one of several examples of fluorine compounds reacting with glassware.
The crystal structure refinement based on the original X-ray intensity data recorded in 1967 is now presented here, because no atomic coordinates were given in the original report  or deposited with the Cambridge Structural Database (CSD; Groom et al., 2016). The square-planar platinum(II) cation and a tetrafluoridoborate anion are shown in Fig. 1. As can be seen, the cation has an approximate mirror plane of symmetry that extends to the conformations of the ethyl groups. The Pt-CO bond length is 1.812 (17) Å , Pt-Cl is 2.301 (4) Å , and the Pt-P bond lengths are 2.341 (5) and 2.348 (5) Å . The P-Pt-CO angles average 92.9 (8) while the Cl-Pt-P angles average 87.2 (2) . The trans angles P-Pt-P and Cl-Pt-C are 174.10 (17) and 177.0 (12) , respectively, with the slight distortions from linearity tending towards a flattened tetrahedron rather than a flattened square pyramid. Each of the triethylphosphine groups has one ethyl group in the trans conformation and two in the gauche conformation.
Packing diagrams showing views down the b and c axes are shown in Fig. 2a and 2b. There are close contacts between each tetrafluoridoborate anion and the ethyl groups of three neighboring cations with putative C-HÁ Á ÁF hydrogen bonds, as listed in Table 1. The Hirshfeld d norm plot for the BF 4 À anion shown in Fig. 3 was produced with CrystalExplorer (Spackman et al., 2021) and indicates a close contact near F2, probably due to the C13-H13Á Á ÁF2 hydrogen bond, which seems to be the strongest C-HÁ Á ÁF bond. The chlorido and carbonyl ligands do not have close intermolecular contacts, perhaps because they are shielded by the gauche conformations of the neighboring ethyl groups. A putative weak C-HÁ Á ÁO hydrogen bond is listed in Table 1 and shown in red in Fig. 3. The hydrogen bonds listed join cations and anions into thick wavy (010) sheets, as can be seen in Fig. 2b.
Database analysis From the time the preliminary structure of this compound was published in 1967, crystal and molecular structures of a Projections of the structure down the b axis (a) and c axis (b), with arbitrary sphere sizes for the atoms. The reference cation and anion have Pt and B atoms identified. Putative C-HÁ Á ÁO and C-HÁ Á ÁF hydrogen bonds are shown as red and green dashed lines, respectively. Table 1 Hydrogen-bond geometry (Å , ). Symmetry codes: (i) x À 1 2 ; Ày þ 1; z; (ii) Àx þ 1 2 ; y; z þ 1 2 ; (iii) Àx; Ày þ 1; z þ 1 2 .

Figure 3
Hirshfeld d norm surface for the BF 4 À anion, showing the red area that indicates close contacts for F2.

Figure 1
View of the molecular entities showing the atomic numbering and displacement ellipsoids at the 50% probability level. wide variety of platinum carbonyl complexes have been reported, ranging from metal clusters through monomeric complexes as in this case. All 662 structures found with the 'PtCO' search fragment in the CSD database, with all filters removed except for 'single-crystal structure', except the present one (TEPPTC) are dated 1968 or after. All but 20 of these structures have only one CO group coordinating to the Pt II atom while the rest have just two coordinating carbonyl groups except for the [Pt(CO) 4 ] 2+ cation reported by Willner et al. (2001) in entry QEZTEU. The mean Pt-CO distance for the 603 structures with coordinates given is 1.860 Å , with a wide range of 1.680 to 2.095 Å . It is interesting that the presence of phosphine ligands tends to lead to longer Pt-CO distances, while the presence of a Cl ligand to shorter Pt-CO distances. Thus, in the 35 entries in the above structures that have two PR 3 groups attached to the Pt II atom as well as the CO group, the mean Pt-C distance is 1.910 Å , with a narrow range of 1.855-1.965 Å , while for the 36 entries that have a Cl as well as a carbonyl ligand, the mean Pt-CO distance is 1.837 Å with a range of 1.753 to 1.901 Å . In the latter case, the Pt-CO distance seems insensitive to whether the Cl atom is cis or trans to the CO group. These tendencies must oppose each other in the present structure, leading to the Pt-CO distance of 1.812 (17) Å . Entry GEYBOB (Rusakov et al., 1988) has the same cation as in the present structure, but the anion is BF 3 Cl À and there is a solvent molecule in the crystal. The shape of the cation is very similar to that of the present structure, with similar distortions of the angles from 90 and a Pt-CO bond length of 1.846 Å .

Synthesis and crystallization
A sample supplied by Dr H. C. Clark had been synthesized as described in Clark & Tsang (1967). Crystals suitable for X-ray analysis were obtained by recrystallization of the sample from methyl acetate.

Refinement
With the early automatic diffractometer that was used to collect the original X-ray intensity data in 1967, it was not customary to obtain a set of Friedel pairs of reflections in the case of a non-centrosymmetric structure. In this case, however, due to the polar space group and the poor scattering by the small crystal, data were collected over the whole sphere of reflection up to = 20 ; in addition, data were recollected over four quadrants for the weaker reflections at higher angles. Initial absorption corrections using a Gaussian grid were inconclusive -perhaps due to a programming error -, so for the final refinements an overall absorption correction using the tensor analysis in XABS2 (Parkin et al., 1995) was used. Hydrogen atoms were constrained, with C-H distances of 0.97 Å and 0.96 Å for CH 2 and CH 3 groups, respectively, and U iso (H) = 1.5U eq (C). Anisotropic temperature factors for the carbonyl CO atoms required tight restraints. While the displacement ellipsoids for the fluorine atoms are large, probably indicating some disorder for the BF 4 À anion (Fig. 1), initial refinements of a disordered model were not successful and the disordered model was not pursued. There is indeed some residual electron density in the neighborhood of the BF 4 À anion, but only one of the 20 highest electron density peaks in the final difference-Fourier map is near this group. Crystal data, data collection and structure refinement details are summarized in Table 2.

data-1
IUCrData (  Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.