| Summary: | The reaction of a Ge(II) hydride compound HC{CMeArN}GeH (Ar = 2,6-iPrCH) 1 with 2,2,2-trifluoroacetophenone (CFPhCO) is theoretically investigated with density functional theory and spin-component-scaled second-order Møller–Plesset methods. This reaction easily occurs with moderate activation barrier and considerably large exothermicity, to afford a Ge(II) alkoxide 2 through a four-membered transition state. In the transition state, the charge transfer from the Ge–H σ-bonding molecular orbital (MO) to the C=O π*-antibonding MO of CFPhCO plays an important role. Acetone ((CH)CO) and benzophenone (PhCO) are not reactive for 1, because their π*-antibonding MOs exist at higher energy than that of CFPhCO. Though 2 is easily formed, the catalytic hydrogenation of CFPhCO by 1 is difficult because the reaction of 2 with a dihydrogen molecule needs a large activation energy. On the other hand, our calculations clearly show that the catalytic hydrogenation of ketone by cis-RhH(PPh)4 easily occurs, as expected. The comparison of catalytic cycle between 1 and 4 suggests that the strong Ge–O bond of 2 is the reason of the very large activation energy for the hydrogenation by 1. To overcome this defect, we investigated various reagents and found that the catalytic cycle can be completed with the use of SiFH. The product is silylether CFPhCHOSiF, which is equivalent to alcohol because it easily undergoes hydrolysis to afford CFPhCHOH. The similar catalytic cycles are also theoretically predicted for hydrosilylations of CO and imine. This is the first theoretical prediction of the full catalytic cycle with a heavier main-group element compound.
|
|---|