Scalable Molecular GW Calculations: Valence and Core Spectra

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Spectroscopy

Background:

• Accurate prediction of molecular ionization energies is crucial for understanding chemical processes.
• Existing computational methods for GW approximation can be computationally expensive.

Purpose of the Study:

• To present a scalable and efficient implementation of the GW approximation using Gaussian atomic orbitals.
• To enable accurate studies of molecular valence and core ionization spectroscopies.

Main Methods:

• Implementation of spectral decomposition and contour-deformation methods for screened-Coulomb interaction.
• Utilized variational fitting approximation for electron repulsion integrals.
• Employed MINRES solver and a heuristic in the quasiparticle equation solver for computational speed-up.

Main Results:

• Achieved a reduction in computational scaling by one order of magnitude.
• Demonstrated accurate predictions for valence and core ionization energies through benchmark tests (GW100, CORE65).
• Validated the method using the carbon 1s binding energy of ethyl trifluoroacetate.

Conclusions:

• The developed GW approximation implementation is accurate and scalable for molecular ionization spectroscopy.
• The method shows promising parallel performance and computational scaling for various molecular systems.