During the past year work on this project in our laboratory has focused on the generation of alumoxanes (ArOAlO)_n and a preliminary screening of various nitrogen sources for nitride formation. The majority of the work was performed by two talented undergraduate (sophomore) students.

Originally, it was attempted to generate compounds such as ArOAl(H)OSiHMe₂ and to study their decomposition to alumoxanes (ArOAlO)_n and concomitant H₂SiMe₂ as a model system for the reaction of H₃AláNMe₃ with (Me₂HSi)₂O which affords aluminumoxyhydride HAlO. Since these attempts led to inseparable mixtures and eventually the disiloxy substituted species 2,6-t-Bu₂-4-Me-C₆H₂OAl(OSiHMe₂)₂áNMe₃, DMSO was investigated as an alternative oxygen donor. From the reactions of four different aryloxy aluminum hydrides with DMSO at low temperatures products could be isolated in two of the cases in low to moderate yields: The reaction of 2,6-t-Bu₂-4-Me-C₆H₂OAlH₂áNMe₃ with 2 equiv DMSO in Et₂O gave the trimeric alumoxane (2,6-t-Bu₂-4-Me-C₆H₂OAlOáOSMe₂)₃ as a crystalline solid which readily desolvates into a an amorphous powder. The structure was confirmed by a low-quality X-ray structure and ¹H and ¹³C{¹H} NMR spectroscopy. Recrystallization from benzene solution afforded a different compound which is believed to be the dimeric species (2,6-t-Bu₂-4-Me-C₆H₂OAlOáOSMe₂)₂ based on NMR spectroscopic and mass spectrometric results. Again, the crystals were subject to facile desolvation upon isolation, which prevented X-ray structure analysis to date. Similarly, the sterically encumbered aluminum hydride Dipp*OAlH₂áNMe₃ (Dipp* = 2,6-Dipp₂C₆H₃-; Dipp = 2,6-i-Pr₂C₆H₃-) reacted with DMSO to generate the dimer (Dipp*OAlOáOSMe₂)₂ based on NMR data. It should also be mentioned that HAlO can be prepared using DMSO instead of (Me₂HSi)₂O. While the DMSO reaction is significantly faster, the resulting product is less clean and possesses a lower hydride activity.

The use of biguanide to prepare acac-like complexes Al{HN=C(NH₂)NC(NH₂)=NH}₃ showed some initial promises, but this part of the project was hampered by unexpected difficulties in the synthesis of biguanide in our hands. Application of the commercially available dicyandiamide H₂NC(=NH)NHCN as a ligand has not yet led to a defined product. On the other hand, we have been successful in the isolation of a stable aluminum azide hydride HAl(N₃)₂á2NMe₃ from the reaction of H₃AláNMe₃ with Me₃SiN₃. Its structure consists of a penta-coordinated aluminum with the anionic ligands in the equatorial positions. The colorless crystalline solid melts without decomposition at 47 °C. Smooth gas evolution commences thereafter, and a color change to brown occurred at 240 °C.