Track synthesis, materials and catalysis as it is published.
Search papers, labs, and topics across Lattice.
This paper addresses the computational bottleneck in variational rovibrational configuration interaction (RVCI) calculations by introducing a basis-pruning algorithm. The method selects vibrational basis states based on their influence on rovibrational energies within a specified spectral window, analogous to configuration-selective VCI. Results on formaldehyde and thioketene demonstrate the algorithm's ability to achieve accurate spectral predictions with significantly reduced basis sizes compared to traditional polyad schemes.
Rovibrational spectra can now be computed with unprecedented efficiency by pruning the vibrational basis using approximate wavefunctions, outperforming traditional polyad schemes.
The variational calculation of rovibrational spectra without model Hamiltonians remains challenging, with the size and composition of the vibrational basis constituting a major bottleneck. Preceding vibrational configuration interaction (VCI) calculations reduce this basis, but leave the question of which vibrational state functions to include. In close analogy to configuration-selective VCI theory, we propose a basis-pruning algorithm in which candidate vibrational states are represented by inexpensive approximate wavefunctions and selected according to their impact on rovibrational energies in a spectral window of interest. For formaldehyde, comparisons with widely used polyad schemes controlling the vibrational basis are provided. Calculations on thioketene benchmark the accuracy and efficiency of the pruning approach for different threshold values.