Acceleration of rovibrational spectrum calculations through sparsity techniques

Area of Science:

• Computational Chemistry
• Molecular Spectroscopy

Background:

• Variational calculations of rovibrational spectra face challenges with large memory and computation times.
• Existing methods employ sparsity and symmetry to mitigate these demands.

Purpose of the Study:

• To investigate techniques for reducing computational costs in variational calculations of microwave and high-resolution rovibrational spectra.
• To analyze the impact of sparsity and symmetry on computational efficiency and spectral accuracy.

Main Methods:

• Employed simple sparsity techniques.
• Incorporated explicit account of molecular symmetry.
• Utilized different rotational basis functions.
• Studied the dependence of eigenvector sparsity on the completeness of the Watson operator.

Main Results:

• Demonstrated reduction in computational demands without deteriorating spectral quality.
• Benchmark calculations for thioformaldehyde, thioformyl cyanide, and thiopropynal validated the approach.
• Detailed analysis of eigenvector sparsity concerning the Watson operator was performed.

Conclusions:

• Sparsity techniques, symmetry, and basis set choices effectively reduce computational burdens in rovibrational spectral calculations.
• The developed methods offer a more efficient pathway for accurate spectral predictions.