Hydride oxidation from a titanium–aluminum bimetallic complex: insertion, thermal and electrochemical reactivity

We report the synthesis and reactivity of paramagnetic heterometallic complexes containing a Ti(iii)-μ-H-Al(iii) moiety.


S3
analysis. For an estimation of the purity of these compounds, see the NMR data reported in section B. X-ray structural determinations were performed at CHEXRAY, University of California, Berkeley on a APEX II QUAZAR or SMART APEX Diffractometer or at the Advanced Light Source (ALS) station 11.3.1 using a silicon-monochromated beam of 16 keV (0.7749 Å) synchrotron radiation. In both cases, diffraction data was collected, integrated, and corrected for absorption using Bruker APEX3 software and its associated modules (SAINT, SADABS).
Structural solutions and refinements (on F 2 ) were carried out using SHELXT and SHELXL-2014 in WinGX. Ellipsoid plots and figures were made using Mercury. 3 (1): Synthesized according to a modification of the LiH 3 AlC(TMS) 3 ·(THF) 2 procedure. 5 (TMS) 3 CH (7.25 mL, 0.026 mol) was dissolved in 30 mL of THF and methyllithium (1.6 M in Et 2 O, 17 mL, 0.027 mol) was added via syringe. The solution was heated to reflux for 16 h after which it was cooled to room temperature and added to LiAlH 4 (1.2 g, 0.032 mol) in 5 mL of THF. The solution immediately became cloudy white. This mixture was heated to 60 °C for 2 h after which the solvent was removed in vacuo. The solid was triturated twice with 5 mL of toluene before extraction into 100 mL of toluene. This solution was added to KHMDS (5.16 g, 0.026 mol) in 10 mL of toluene and stirred overnight. The solvent was removed in vacuo and the solid triturated thoroughly with 5 mL of toluene before it was redissolved in 100 mL of toluene.

KH 3 AlC(TMS)
The cloudy suspension was filtered through Celite ® to give a colorless solution. The solvent was removed in vacuo and the resulting solid was suspended in hexane and collected on a Schlenk frit. Yield: 5.55 g (71%). 1  S4 colorless solution. The solvent was removed in vacuo and the resulting solid was suspended in hexane and collected on a Schlenk frit. Yield: 2.7 g (69%). Characterization is identical to that reported for 1 but the IR stretch at 1646 cm -1 was absent. (See Figure S30) Cp 2 TiH 3 AlC(TMS) 3 (2): Cp 2 TiCl (0.485 g, 2.27 mmol) and KH 3 AlC(TMS) 3 (1.41 g, 4.69 mmol) were dissolved in 20 mL of Et 2 O cooled to -78 °C. The solution immediately turned bright purple.
The mixture was allowed to warm to room temperature and stirred for 30 min. Solvent was removed in vacuo and the solid was triturated with 5 mL of hexane. The solid was then suspended in 50 mL of hexane and stirred for 3 h, over which time the solution became blue-purple. The mixture was concentrated to 20 mL and filtered through Celite ® . The solvent was removed in vacuo to yield blue-purple crystals. Yield: 826 mg (83%). Cp 2 TiH 3 AlC(TMS) 3 obtained contained approximately 5% KH 3 AlC(TMS) 3 which was not found to impede further reactivity studies. Xray quality and analytically pure crystals were grown by storage of a concentrated hexane solution at -40 °C (Approximate recovery: 40%). 1  to -78 °C. The solution immediately turned bright purple. The mixture was allowed to warm to room temperature and stirred for 30 min. Solvent was removed in vacuo and the solid was triturated with 5 mL of hexane. The solid was then suspended in 30 mL of hexane and stirred for 3 h, over which time the solution became blue-purple. The mixture was concentrated to 10 mL and filtered through Celite ® . The solvent was removed in vacuo to yield blue-purple crystals. Characterization is identical to that reported for 2 but the IR stretch at 1834 cm -1 was absent. (See Figure S31)      Minor decomposition (*) was observed immediately and did not change during the experiment

F. Computational Details
All calculations were carried out with the Gaussian 09 program (G09) 8 , employing the B3LYP 9 functional with standard 6-31G+(d,p) 10 basis set to fully optimize the geometries of the complexes. All resultant stationary points were subsequently characterized by vibrational analyses.
Since a test calculation between two intermediates with the C(TMS) 3 and CH 3 ligand showed the same relative energies with the full and model system, the abbreviated methyl model ligand was used for all further calculations. Additionally, we studied a unimolecular pathway rather than a bimolecular pathway. Since the same types of bonds are activated in a unimolecular or bimolecular process, the trends in energies determined for a monomeric system are expected to apply to potential bimolecular pathways while simplifying comparison between systems and pathways. While trends in the energies of the transition states are expected to remain the same comparing a unimolecular or bimolecular reaction mechanism, the absolute barriers to activation may not remain the same. For this reason, while the calculated barriers are quite high, the different pathways may still be compared.
For the thermal reactivity of 2', two intermediates were calculated. The intermediate corresponding to reductive elimination (A) of two hydrides is 14.3 kcal/mol higher in energy than that corresponding to elimination via s-bond metathesis (B). However, the transition states for formation of each intermediate are at similar energies (∆∆G ‡ = 7.5 kcal/mol). The small energy difference favors s-bond metathesis, but it is not significant enough to disregard either pathway given the high reaction temperature (100 °C). The transition state for the oxidative addition of the Cp C-H bond between the two intermediates was also calculated and was found to be 73.3 kcal/mol higher energy than the starting material, suggesting that the oxidative addition of this bond in a unimolecular system is thermally inaccessible ( Figure S43). Despite the tendency for early metals to react via s-bond metathesis rather than reductive elimination, the transition states for reductive elimination and s-bond metathesis in this system are at similar energies and cannot be differentiated with DFT calculations.