Files
Abstract
Ab initio quantum chemistry enhances experimental chemistry by predicting new phenomena and improving our fundamental understanding of the natural world. In this work, we perform highly accurate quantum chemistry techniques on a variety of molecular systems. These studies motivate the search for optimal coordinate representations that reduce the cost of ab initio chemistry. First, we investigate the competition between halogen and hydrogen bonding in the very first theoretical examination of the HOX --- SO2 (X =F, Cl, Br, I) binary complexes, molecules that are prominent in the atmosphere. We then utilize rigorous ab initio quantum chemistry methods to investigate the tin and lead analogs of formaldehyde. By characterizing the potential energy surface and computing highly accurate fundamental frequencies, we provide the theoretical fingerprint for molecules that have thus far eluded experimental detection. Challenged by the vast amount of computing resources required to perform a vibrational analysis, we improve the accuracy of the Concordant Mode Approach for molecular vibrations with the inclusion of off-diagonal force constants. We provide a balance between cost and accuracy by selectively increasing the coupling force constants until the method converges to the exact target frequencies. Finally, we present MolSym, a Python package for handling nonabelian symmetry in molecular quantum chemistry. MolSym seeks to provide a standard yet highly sophisticated approach for common symmetry techniques in chemistry calculations. A symmetry-adapted Hartree--Fock benchmark demonstrates the remarkable computational speedups to be gained by using the full molecular point group, which most electronic structure package fail to do.