A monomeric aluminum hydride complex bearing substituted pyrrolyl ligands, AlH[C(4)H(3)N(CH(2)NMe(2))-2](2) (1), was synthesized and structurally characterized. To further confirm the presence of Al--H bonds, the compound AlD[C(4)H(3)N(CH(2)NMe(2))-2](2) ([D]1) was synthesized by reacting LiAlD(4) with [C(4)H(4)N(CH(2)NMe(2))-2]. Compound 1 and [D]1 react with phenyl isothiocyanate yielding Al[C(4)H(3)N(CH(2)NMe(2))-2](2)[eta(3)-SCHNPh] (2) and Al[C(4)H(3)N(CH(2)NMe(2))-2](2)[eta(3)-SCDNPh] ([D]2) by insertion. The reactions of 1 with 9-fluorenone and benzophenone generated the unusual aluminum alkoxide complexes 3 and 4, respectively, through intramolecular proton abstraction and C-C coupling. A mechanistic study shows that 9-fluorenone coordinates to [D]1 and releases one equivalent of HD followed by C-C coupling and hydride transfer to yield the final product. Reduction of benzil with 1 affords aluminum enediolate complex 5 in moderate yield. Mechanistic studies also showed that the benzil was inserted into the aluminum hydride bond of [D]1 through hydroalumination followed by proton transfer to generate the final product [D]5. All new complexes have been characterized by (1)H and (13)C NMR spectroscopy and X-ray crystallography.

Object correlations are common semantic patterns in virtual reality systems. They can be exploited for improving the effectiveness of storage caching, prefecthing, data layout, and minimization of queryresponse times. Unfortunately, this information about object correlations is unavailable at the storage system level. Previous approaches for reducing I/O access time are seldom investigated. On the other side, data mining techniques extract implicit, previously unknown and potentially useful information from the databases. This paper proposes a class of novel and efficient pattern-growth method for mining various frequent sequential traversal patterns in the virtual reality. Our pattern-growth method adopts a divideand-conquer approach to decompose both the mining tasks and the databases. Moreover, our efficient data structures are proposed to avoid expensive, repeated database scans. The frequent sequential traversal patterns are used to predict the user navigation behavior and help to reduce disk access time with proper placement patterns into disk blocks. We also define the terminologies such as paths, views and objects used in the model. We have done extensive experiments to demonstrate how these proposed techniques not only significantly cut down disk access time, but also enhance the accuracy of data prefetching.

No abstract