Scalable Δ-Self-Consistent-Field Calculations of Core Electron Binding Energies in Periodic and Aperiodic Systems using FHI-aims
ORAL
Abstract
In the ΔSCF (Δ-Self-Consistent-Field) method, core electron binding energies are calculated directly as the total energy difference between the ground state, and the final state with a core hole. All-electron ΔSCF calculations are routinely performed for small molecules. Previous proof-of-concept studies have shown that the ΔSCF method can also yield highly accurate core electron binding energies in solids and surface species, but limitations of current numerical implementations make such calculations cumbersome to set up, and needlessly expensive to run.
In this talk, I will present recent developments in the electronic structure code FHI-aims that integrate the routines for creating and tracking a localized core hole with the scalable Electronic Structure Infrastructure (ELSI). With the new routines, calculations of core-excited states have the same scalability as regular ground state DFT, both with regards to system size as well as the use of massively parallel computing resources. In addition, ΔSCF calculations of periodic systems now no longer require the use of wavefunction based restart files, which considerably simplifies the workflow for the user. Scalability tests of the new implementation, as well as applications to practical core electron binding energy calculations in extended systems will be presented.
In this talk, I will present recent developments in the electronic structure code FHI-aims that integrate the routines for creating and tracking a localized core hole with the scalable Electronic Structure Infrastructure (ELSI). With the new routines, calculations of core-excited states have the same scalability as regular ground state DFT, both with regards to system size as well as the use of massively parallel computing resources. In addition, ΔSCF calculations of periodic systems now no longer require the use of wavefunction based restart files, which considerably simplifies the workflow for the user. Scalability tests of the new implementation, as well as applications to practical core electron binding energy calculations in extended systems will be presented.
*This project has received funding from the European Union's Horizon Europe research and innovation programme under grant agreement no. 101131173 (BETTERXPS). Refactorization of the $\Delta$SCF code in FHI-aims was funded under the embedded CSE programme of the ARCHER2 UK National Supercomputing Service.
–
Publication: Blum et al., "Roadmap on Advancements of the FHI-aims Software Package", arXiv:2505.00125 (2025)
Presenters
-
Juhan Matthias Kahk
- University of Tartu