Identifying Electronic Modes by Fourier Transform from δ-Kick Time-Evolution TDDFT Calculations

Rajarshi Sinha-Roy, Pablo García-González, Xóchitl López Lozano, Robert L. Whetten, Hans Christian Weissker

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

Time-dependent density-functional theory (TDDFT) is widely used for calculating electron excitations in clusters and large molecules. For optical excitations, TDDFT is customarily applied in two distinct approaches: transition-based linear-response TDDFT (LR-TDDFT) and the real-time formalism (RT-TDDFT). The former directly provides the energies and transition densities of the excitations, but it requires the calculation of a large number of empty electron states, which makes it cumbersome for large systems. By contrast, RT-TDDFT circumvents the evaluation of empty orbitals, which is especially advantageous when dealing with large systems. A drawback of the procedure is that information about the nature of individual spectral features is not automatically obtained, although it is of course contained in the time-dependent induced density. Fourier transform of the induced density has been used in some simple cases, but the method is, surprisingly, not widely used to complement the RT-TDDFT calculations; although the reliability of RT-TDDFT spectra is now widely accepted, a critical assessment for the corresponding transition densities and a demonstration of the technical feasibility of the Fourier-transform evaluation for general cases is still lacking. In the present work, we show that the transition densities of the optically allowed excitations can be efficiently extracted from a single δ-kick time-evolution calculation even in complex systems like noble metals. We assess the results by comparison with the corresponding LR-TDDFT ones and also with the induced densities arising from RT-TDDFT simulations of the excitation process.

Original languageEnglish (US)
Pages (from-to)6417-6426
Number of pages10
JournalJournal of Chemical Theory and Computation
Volume14
Issue number12
DOIs
StatePublished - Dec 11 2018
Externally publishedYes

ASJC Scopus subject areas

  • Computer Science Applications
  • Physical and Theoretical Chemistry

Fingerprint

Dive into the research topics of 'Identifying Electronic Modes by Fourier Transform from δ-Kick Time-Evolution TDDFT Calculations'. Together they form a unique fingerprint.

Cite this