Analytic energy gradients for the self-consistent direct random phase approximation

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Materials Science

Background:

• Accurate prediction of molecular geometries and vibrational frequencies is crucial for understanding chemical reactions and material properties.
• Existing methods like non-self-consistent dRPA, DFT, and MP2 have limitations in accuracy and applicability.
• The development of variational methods is key to improving computational chemistry predictions.

Purpose of the Study:

• To derive and implement analytic energy gradients for the self-consistent direct random phase approximation (sc-dRPA) method.
• To assess the accuracy and robustness of sc-dRPA for predicting molecular properties.
• To compare sc-dRPA performance against established computational methods.

Main Methods:

• Derivation and implementation of analytic energy gradients for sc-dRPA.
• Validation of analytic gradients against numerical calculations.
• Calculation of equilibrium geometries and vibrational frequencies for diverse molecular systems.
• Comparison with Møller-Plesset perturbation theory (MP2), density-functional theory (DFT), and coupled cluster methods.

Main Results:

• Analytic energy gradients for sc-dRPA were successfully implemented and validated.
• sc-dRPA accurately predicts equilibrium geometries and vibrational frequencies for various molecules, including weakly bonded dimers and transition metal compounds.
• sc-dRPA demonstrates superior accuracy and robustness compared to MP2 and DFT for challenging systems.
• Coupled cluster methods (CCSD) showed inferior performance in this study.

Conclusions:

• The sc-dRPA method with analytic gradients provides a highly accurate and robust approach for molecular property calculations.
• sc-dRPA offers a reliable alternative to conventional methods, especially for systems where other methods fail.
• This advancement in sc-dRPA facilitates more precise predictions in quantum chemistry and materials science.