The calculation of accurate core electron binding energy shifts and absolute core electron binding energies using the SCAN exchange-correlation functional
ORAL
Abstract
Core electron X-ray Photoelectron Spectroscopy (XPS) is often used to characterize the elemental as well as the chemical composition of surfaces. However, as experimental practice moves towards the study of increasingly complex systems, the analysis of recorded spectra becomes ever more challenging. First principles calculations of core electron binding energies can substantially alleviate the problems that are commonly encountered in “peak assignment”.
Recently, we have found that the SCAN density functional is particularly well suited for the calculation of core electron binding energies using the ΔSCF method. In comparison to gas phase reference data, SCAN yields absolute core electron binding energies that are accurate to within ~0.2 eV for the 1s and 2p core levels of C-F and Si-Cl respectively, without any empirical adjustments.
For calculations of adsorbates on metal surfaces, we have devised a two-step approach in which the geometry of the adsorbate is first relaxed using a slab model of the surface, and then a ΔSCF calculation is performed on a cluster cut from the slab. Analysis of the results for various cluster sizes shows that finite size effects can be minimized to the point where they are smaller than the error due to the DFT functional.
Recently, we have found that the SCAN density functional is particularly well suited for the calculation of core electron binding energies using the ΔSCF method. In comparison to gas phase reference data, SCAN yields absolute core electron binding energies that are accurate to within ~0.2 eV for the 1s and 2p core levels of C-F and Si-Cl respectively, without any empirical adjustments.
For calculations of adsorbates on metal surfaces, we have devised a two-step approach in which the geometry of the adsorbate is first relaxed using a slab model of the surface, and then a ΔSCF calculation is performed on a cluster cut from the slab. Analysis of the results for various cluster sizes shows that finite size effects can be minimized to the point where they are smaller than the error due to the DFT functional.
–
Presenters
-
Juhan Matthias Kahk
Department of Materials, Imperial College London
Authors
-
Juhan Matthias Kahk
Department of Materials, Imperial College London
-
Johannes Lischner
Imperial College London, Department of Materials and Department of Physics, Imperial College London