A trust-region augmented Hessian implementation for state-specific and state-averaged CASSCF wave functions

TL;DR

This paper introduces a robust second-order convergence algorithm for state-specific and state-averaged CASSCF wave functions using a trust-region augmented Hessian approach, enabling reliable calculations on large molecules.

Contribution

The work extends the trust-region augmented Hessian method to both state-specific and state-averaged CASSCF, improving convergence reliability and efficiency for large molecular systems.

Findings

Achieves robust convergence with the TRAH algorithm.

Enables calculations on large molecules like a Ni(II) complex with 231 atoms.

Performance is competitive with other second-order methods, with reliable convergence even when first-order methods fail.

Abstract

In this work, we present a one-step second-order converger for state-specific (SS) and state-averaged (SA) complete active space self-consistent field (CASSCF) wave functions. Robust convergence is achieved through step restrictions using a trust-region augmented Hessian (TRAH) algorithm. To avoid numerical instabilities, an exponential parametrization of variational configuration parameters is employed, which works with a nonredundant orthogonal complement basis. This is a common approach for SS-CASSCF and is extended to SA-CASSCF wave functions, in this work. Our implementation is integral direct and based on intermediates that are formulated either in the sparse atomic-orbital or small active molecular-orbital basis. Thus, it benefits from a combination with efficient integral decomposition techniques, such as the resolution-of-the-identity or the chain-of-spheres for exchange approximations. This facilitates calculations on large molecules such as a Ni(II) complex with 231 atoms and 5154 basis functions. The runtime performance of TRAH-CASSCF is competitive with other state-of-the-art implementations of approximate and full second-order algorithms. In comparison with a sophisticated first-order converger, TRAH-CASSCF calculations usually take more iterations to reach convergence and, thus, have longer runtimes. However, TRAH-CASSCF calculations still converge reliably to a true minimum even if the first-order algorithm fails.