Si- ■ -H distances

One of the most conspicuous differences between computational results is in the degree to which a normal H—Si chemical bond is formed. In the local-density pseudopotential calculations, the Si—H separation is about 1.6 A. This is much larger than the predictions of MNDO, Hartree-Fock, or PRDDO calculations, which are much closer to the molecular Si—H distance. It is not clear at this point whether the H—Si bond is, in fact, weaker than a conventional bond when in this configuration and therefore is overestimated by the Hartree-Fock-like calculations, or whether the strength is being underestimated in the local-density calculations. 

One study (DeLeo et al., 1988 Fowler et al., 1989) has found that neutral H at the B site in Si has a tendency to preferentially bind to one of the two Si neighbors, leading to an asymmetric configuration, with Si—H distances of 1.48 A and 1.77 A respectively. This tendency was interpreted in terms of a pseudo-Jahn-Teller distortion. However, the potential barrier that leads to the asymmetric position is so low (< 0.2 eV) that it can readily be surmounted by zero-point motion of the proton. Experimental observation of such an asymmetry is therefore unlikely, except maybe through an isotope shift measurement in an infrared experiment (DeLeo et al., 1988). None of the other theoretical approaches has produced this type of asymmetry. 

Calculations of vibrational frequencies in a three-center bond as a function of Si—Si separation were performed by Zacher et al. (1986), using linear-combination-of-atomic-orbital/self-consistent field calculations on defect molecules (H3Si—H—SiH3). The value of Van de Walle et al. for H+ at a bond center in crystalline Si agrees well with the value predicted by Zacher et al. for a Si—H distance of 1.59 A. 

This behavior must be related to the Si-H distance, 2.73 A which is closer to the sum of the van der Waals radii, 3.1 A, and corresponds to almost independent ligands. 

Cowie and Bennett explain the short Si-H distance in the manganese complex in terms of steric hindrance however, the chemical behavior of this complex is not consistent with this interpretation. 

Fig. 12. Potential energy diagram for the unimolecular decomposition of SiH4 to SiHj and H, showing the activation energy barrier . The sketch above is a geometrical representation of the events (1 to r) along the reaction coordinate, which is the Si-H distance.

To summarize, the Si-H distance can serve as a criterion of the Si-H interaction when it does not differ much from the normal Si-H bond and is determined by an accurate method such as ND and high-level quantum mechanic calculations. The discussion of other structural parameters requires the proper choice of a reference system. In systems with elongated Si-H interactions a justified conclusion can be made only on the basis of a combined application of several independent structural, spectroscopic, and computational methods. 

Hz found for 11. These values correlate well with the X-ray determined Si-H distance of 1.61(4) A in 27, which is shorter than the range 1.75(4)-1.802(5) A determined in 11. The stronger Si-H interaction in 27 is apparently the result of a weaker donor ability of the arene ligand compared with the cyclopentadienyl ligand, which leads to a weaker backdonation from the chromium center. A similar correlation between the r(Si-H) and the /(Si-H) in complexes 12 has been discussed above. 

The silane o-complexes 11 have been intensively studied as discussed above and in Refs. 2, 4, and 12-14. No examples of such complexes are known for technetium, whereas the related rhenium complex [Re(H)(SiR3)(CO)2Cp] has been concluded to be classical on the grounds of a long-estimated Si-H distance of 2.2 A. Schubert favors a classical description of the latter compound, whereas Kubas noted that this distance may correspond to a stretched a-complex on the verge of oxidative... 

X-ray structure analysis of 49 revealed that agostic bonding causes a severe distortion of the carbene ligand, so that the W-C-Si bond angle is reduced to 87.8(6)°, compared with 113.1(4)° in the starting complex. The hydride atom was, however, found closer to the silicon atom, the W-H and Si-H distances being 2.1(1) and 1.5(1) A, respectively. 

The silanol complex 57 exhibits a Si H M agostic interaction characterized by a /(Si-H) of 41 Hz and a Si-H distance of 1.70(7) It would be incautious to interpret such a low value of the Si-H coupling in terms of a significant Si-H bond activation, because the Si-H bond forms rather acute angles with the Si-C and Si-Si bonds (about 82 and 101°, respectively) and thus must have a considerable p character on silicon, which should contribute to the decrease of /(Si-H). The silanol ligand is -coordinate to ruthenium and the Ru-Si bond of 2.441(3) A is not exceptional, but the Si(SiMe3)3 deviates from the silanol plane by 19.0°, probably as a result of the Si-H interaction. Deprotonation of 57 by strong bases affords a neutral ruthenocene-like product. 

---