[4-tert-Butyl-2,6-bis(diphenylmethyl)phenolato-κO]diethyl(tetrahydrofuran-κO)aluminium

The title compound has monoclinic (P21/n) symmetry with a single Al atom in the asymmetric unit. The complex possesses catalytic activity in the ring-opening polymerization of ∊-caprolactone.

The title compound, {Al[O-2,6-(Ph 2 CH) 2 -4-t BuC 6 H 2 ]Et 2 (THF)} or [Al(C 2 H 5 ) 2 -(C 36 H 33 O)(C 4 H 8 O)], was formed in the reaction between 4-tert-butyl-2,6bis(diphenylmethyl)phenol and triethylaluminum in the presence of THF (THF is tetrahydrofuran) and recrystallized from hexane. The structure has monoclinic (P2 1 /n) symmetry with a single Al atom in the asymmetric unit. The terminal C atom of one ethyl substituent is nearly equally disordered over two positions. The complex possesses catalytic activity in the ring-opening polymerization of "-caprolactone.
The obtained Al complex activated by benzyl alcohol demonstrates moderate catalytic activity in "-caprolactone polymerization in THF, with 14% conversion after 10 min and 100% after 4 h for a 1 M monomer solution (Fig. 2). However, we have found that this catalytic system is not able to catalyse the ROP of rac-lactide under the same conditions.

Structural commentary
The Al atom of the title compound, {Al[O-2,6-(Ph 2 CH) 2 -4-t BuC 6 H 2 ]Et 2 (THF)}, is in a distorted tetrahedral environment (Fig. 3). The C40 atom of one ethyl group is equally disordered over two positions with an occupancy ratio of 0.50 (2):0.50 (2). As expected, the largest Al-ligand distances correspond to Al-Et bonds [1.9732 (19) for Al-C37 and 1.970 (2) Å for Al-C39]. The shortest Al-ligand length is for the Al-O1 bond [1.7171 (12) Å ], presumably because of the presence of a negative charge at the phenoxide anion OAr À regardless of its bulkiness, whereas the Al-O THF bond is somewhat longer [1.8966 (13) Å , Al-O2]. The bond angles around the Al atom range from 100.55 (6) for O1-Al1-O2 to 116.75 (10) for C37-Al1-C39, with the O-Al-C angles lying in the middle of this range. All phenyl groups are directed away from the Al atom because of the substantial steric hindrance of the phenoxide ligand. No non-coordinating solvent molecules are present in the crystal structure, and no significant non-valence intermolecular interactions have been found.

Synthesis and crystallization
All synthetic manipulations were performed under a purified argon atmosphere, using Schlenk glassware, dry-box techniques and absolute solvents. NMR spectra were recorded with a Bruker AVANCE 400 spectrometer at 298 K. C/H elemental analysis was performed with a Perkin-Elmer 2400 Series II elemental analyzer. Gel permeation chromatography (GPC) measurements were recorded on an Agilent PL-GPC 220 chromatograph equipped with a PLgel column (eluent: THF, 1 ml/min, 313 K), using universal calibration with a polystyrene standard.

Polymerization experiments
A solution of the Al complex (69 mmol) in THF was injected into a solution of a monomer [either rac-lactide (rac-LA) or "-caprolactone ("-CL), 6.9 mmol] and PhCH 2 OH (69 mmol) in THF. The monomer concentration was 1.0 M. The reaction was carried out for 10 min and for 4 h. According to 1 H NMR (in CDCl 3 ), conversion of rac-LA was 0% in both cases. Conversion of "-CL was 14% after 10 min, and 100% after 4 h. In the latter case, the recorded 1 H NMR spectrum showed the disappearance of the CH 2 OC O resonance signal of "-CL at 4.14 ppm and the presence of the poly-"caprolactone (PCL) resonance signal at 3.98 ppm (CH 2 OC O). The polymer solution was quenched with THF containing an excess of acetic acid. The polymer solution was precipitated from Et 2 O, filtered off, reprecipitated from a THF/Et 2 O mixture at 253 K, filtered off, and dried under vacuum. The isolated PCL had a regular 1 H NMR spectrum for PCL. GPC data (THF, 313 K): M n = 1.73 Â 10 4 PDI = 1.67.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 1. The hydrogen atoms were positioned geometrically (C-H = 0.95 Å for aromatic, 0.98 Å for methyl, 0.99 Å for methylene and 1.00 Å for tertiary H atoms) and refined as riding atoms with relative isotropic displacement parameters U iso (H)= 1.5U eq (C) for methyl H atoms and 1.2U eq (C) otherwise. A rotating group model was applied for methyl groups. Reflection (0 0 2) was affected by the beam stop, and was therefore omitted from the refinement. SADI and SIMU SHELXL (Sheldrick, 2015) restraints were applied for modelling the C40A/C40B disorder.
The five highest residual electron-density peaks are located at the t-Bu group and near THF atoms C42 and C43, pointing to some minor remaining disorder. Using a set of positional and bond-parameter restraints, estimated ratios for the t-Bu rotational disorder and for the disorder in the THF molecule (atoms C42, C43) were found to be 0.939 (2):0.061 (2) and 0.904 (7):0.096 (7), respectively. However, the residual electron density was not sufficient to adequately model the mentioned disorders, which were therefore not included in the final crystallographic model.  Computer programs: APEX2 and SAINT (Bruker, 2008), SHELXS97 and SHELXTL (Sheldrick, 2008), SHELXL2017 (Sheldrick, 2015)and publCIF (Westrip, 2010).

[4-tert-Butyl-2,6-bis(diphenylmethyl)phenolato-κO]diethyl(tetrahydrofuran-κO)aluminium
Crystal data 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. 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 > 2sigma(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.