Diffusion Monte Carlo evaluation of disiloxane linearization barrier

Abstract

The disiloxane molecule is a prime example of silicate compounds containing the Si-O-Si bridge, which is of great interest within the field of quantum chemistry, due to the difficulty in theoretically predicting its properties. The linearization barrier of disiloxane is investigated by {\it ab initio} quantum Monte Carlo (QMC), which is currently the most reliable first-principles calculation method in accounting for electron correlation. Density functional theory (DFT) and coupled cluster single double and perturbative triple (CCSD(T)) calculations are also carried out alongside QMC as points of comparison. Various basis sets are also used to investigate the dependence of calculation results, most notably of the Pople split valence and the correlation consistent (cc-) family of basis sets. We find that QMC successfully predicts the disiloxane linearization barrier with less dependence on the completeness of basis sets than either DFT or CCSD(T), showing its viability in this subject.