Low-scaling analytical gradients for the direct random phase approximation using an atomic orbital formalism

Area of Science:

• Quantum Chemistry
• Computational Physics

Background:

• Calculating first-order properties in molecules is computationally intensive.
• Existing methods face challenges with large systems and basis set redundancies.

Purpose of the Study:

• To develop an efficient and accurate atomic orbital formalism for calculating analytical gradients.
• To reduce computational complexity in electronic structure calculations.

Main Methods:

• Utilizing an atomic orbital formalism for analytical gradients within the random phase approximation.
• Exploiting sparsity in electronic structure.
• Employing Cholesky decomposed densities to manage basis set redundancies.

Main Results:

• Demonstrated the validity and accuracy of the new approach and its approximations.
• Showcased the method's efficiency by computing nuclear gradients for large systems (up to 600 atoms).

Conclusions:

• The developed method offers a competitive alternative to canonical theories, especially for small molecules.
• The formalism is general and extensible to other correlation methods, paving the way for broader applications in computational chemistry.