rac-[2-(Dicyclohexylphosphanyl)phenyl](phenyl)phosphinic diisopropylamide–borane hemihydrate

In the title compound, C30H48BNOP2·0.5H2O, the water molecule is disordered about an inversion centre. Both phosphorus atoms shows distortions in their tetrahedral environments with the cyclohexyl substituents disordered over two orientations in a 0.851 (3):0.149 (3) occupancy ratio. The crystal structure is assembled via O—H⋯O interactions between pairs of phosphininc amide molecules and water molecules, creating hydrogen-bonded dimers with graph-set R 2 4(8) along [001]. Weak C—H⋯O interactions are also observed.

In the title compound, C 30 H 48 BNOP 2 Á0.5H 2 O, the water molecule is disordered about an inversion centre. Both phosphorus atoms shows distortions in their tetrahedral environments with the cyclohexyl substituents disordered over two orientations in a 0.851 (3):0.149 (3) occupancy ratio. The crystal structure is assembled via O-HÁ Á ÁO interactions between pairs of phosphininc amide molecules and water molecules, creating hydrogen-bonded dimers with graph-set R 2 4 (8) along [001]. Weak C-HÁ Á ÁO interactions are also observed.
The University of the Free State is thanked for the use of their diffractometer. Financial assistance from Sasol, THRIP and the Research Fund of the University of Johannesburg is gratefully acknowledged. Snieckus, V. (1990). Chem. Rev. 90, 879-933. Tolman, C. A. (1977. Chem. Rev. 77, 313-348. Westrip, S. P. (2010). J. Appl. Cryst. 43, 920-925. Williams, D. B. G., Evans, S. J., De Bod, H., Mokhadinyana, M. S. & Hughes, T. (2009). Synthesis, 18, 3106-3112. affords arylphosphine ligands resistant to oxidation and hydrolysis while maintaining high catalyst activity. The synthesis rests strongly on DoM technology (Snieckus, 1990) making use of a directing group that is highly underrepresented in this type of chemistry. We envisioned that the use of phosphinic amides as directing groups, together with phosphinous chloride (Cy 2 PCl) electrophiles would allow the synthesis of sterically hindered phosphines that are stable to hydrolysis and oxidation. The ortho-deprotonation of phosphinic amides with sec-butyl-lithium and quenching with dicyclohexylphosphinous chloride (Cy 2 PCl) allowed isolation of the desired ligand in good yields (45-60% yield), which are stable to air, liquid-liquid extraction, and chromatography without special exclusion of oxygen.
The title compound ( Fig. 1)  The most common method used for determining the steric behaviour of a phosphane ligand is the Tolman cone angle (Tolman, 1977). We used the geometry from the title compound and adjusted the P═O and P-B distances to 2.28 Å (the average Ni-P distance used in the original Tolman model) to cancel the bias this may have on the calculated cone angle value. In this way we obtain effective cone angle (Otto, 2001) values of 231 and 181° for P1 and P2 respectively.
The structure is stabilized by strong intermolecular O-H···O hydrogen bonds formed between the phosphinic oxygen atom and the oxygen atom of the water molecule, creating head-to-head dimeric structures with the phosphinic amide molecules (Fig. 2) The graph set notation for this interaction is R 2 4 (8) (Bernstein et al., 1995) Additional weak C-H···O interactions are also observed and summarized in Table 1.

Experimental
Cyclohexylchloride (1 mL, 8.42 mmol) was added to a solution of diethyl ether (10 ml) and magnesium turnings (1.0 eq., 204 mg, 8.42 mmol) along with one crystal of iodine as an initiator and the mixture was heated under reflux until all the magnesium had been consumed. In a separate flask, PCl 3 (3.24 mmol, 0.38 eq., 283 µL) was dissolved in diethyl ether (40 mL) and the solution cooled to -40 °C. The cyclohexylmagnesium chloride solution was added dropwise over 10 minutes and the solution was allowed to warm to room temperature over three hours. Once the reaction was complete the salts that formed were filtered through a pad of Celite. The resultant product was approximately 70% pure (as determined by 31 P NMR spectroscopy) and was used without further manipulation, the reaction producing (2.27 mmol) of chloro-di- N,N-Diisopropyldiphenylphosphinic amide (569 mg, 1.89 mmol) was weighed out in a Schlenk flask and THF (10 mL) was added. The solution was then cooled to -60 °C and sec-BuLi (1.1 eq., 1M) was added. The solution was allowed to stir for three hours between -40 and -70 °C after which it was cooled to -78 °C and the electrophile (1.2 eq.) dissolved in a small amount of THF was added. The reaction mixture was allowed to warm to room temperature over four hours and was stirred at room temperature overnight. All solvents were then removed in vacuo and the residue was extracted with EtOAc and H 2 O. The product was purified by column chromatography on flash silica.
Protection of the phosphine occurred by first dissolving the phosphine in THF (10 mL) cooling the mixture to 0 °C and adding an excess of BH 3 in THF and the reaction stirred at room temperature for 5 h. All solvents were then removed in vacuo and the resulted residue was the desired product in 100% yield. Crystals were grown by dissolving the ligand in a minimal amount of DCM and then layering an excess of hexane on the DCM and allowing to stand in a refrigerator until the crystals were formed.
Yield: 51% (White solid). 1.2U eq (C) for the aromatic, methine and methylene H and U iso (H) = 1.5U eq (C) for the methyl and B-H respectively.
Locations of the methyl hydrogen atoms were initially obtained from a Fourier difference map and refined as a fixed rotor. Refinement of the oxygen atom of the water molecule showed large thermal vibration, and in subsequent refinement cycles the occupancy thereof was freed. This refined to nearly 50% and in the final refinement cycles the occupancy value was constrained to half. The hydrogen atoms of the water molecule were located from a Fourier difference map. Both of the cyclohexcyl substituents showed somewhat large thermal ellipsoids and were subsequently refined as disordered over two positions. Their geometries and ellipsoid sizes were kept reasonable by restraining with the appropriate refinement commands (SAME, SADI and SIMU). The occupancies were refined with a free variable that added to unity and a final ratio of 85:15 was obtained between the two components. Discrepant reflection 001 was removed in the final stages of refinement.  A view of the title complex, showing the atom-numbering scheme and 50% probability displacement ellipsoids.

Computing details
Hydrogen atoms (except for the water solvate) as well as the minor part of the disorder omitted for clarity.

Figure 2
Packing diagram showing the O-H···O hydrogen bonding interactions (indicated by green dashed lines). Special details Experimental. The intensity data was collected on a Bruker X8 APEXII 4 K KappaCCD diffractometer using an exposure time of 20 s/frame. A total of 2529 frames were collected with a frame width of 0.5° covering up to θ = 28.33° with 99.6% completeness accomplished. Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.