Nanoplasmonics simulations at the basis set limit through completeness-optimized, local numerical basis sets

TL;DR

This paper introduces a method to generate highly accurate local numerical basis sets for nanoplasmonics simulations within time-dependent density functional theory, enabling precise modeling of metallic nanoparticles with computational efficiency.

Contribution

The authors develop a completeness-optimization scheme to create local basis sets that approach the complete basis set limit, improving accuracy in first-principles nanoplasmonics simulations.

Findings

Basis sets achieve near-complete accuracy for metal nanoparticles.

Performance transfers effectively to larger nanoparticles and nanoalloys.

Method is compatible with various exchange-correlation functionals.

Abstract

We present an approach for generating local numerical basis sets of improving accuracy for first-principles nanoplasmonics simulations within time-dependent density functional theory. The method is demonstrated for copper, silver, and gold nanoparticles that are of experimental interest but computationally demanding due to the semi-core d-electrons that affect their plasmonic response. The basis sets are constructed by augmenting numerical atomic orbital basis sets by truncated Gaussian-type orbitals generated by the completeness-optimization scheme, which is applied to the photoabsorption spectra of homoatomic metal atom dimers. We obtain basis sets of improving accuracy up to the complete basis set limit and demonstrate that the performance of the basis sets transfers to simulations of larger nanoparticles and nanoalloys as well as to calculations with various exchange-correlation functionals. This work promotes the use of the local basis set approach of controllable accuracy in first-principles nanoplasmonics simulations and beyond.