Combining Pseudopotential and All Electron Density Functional Theory for the Efficient Calculation of Core Spectra using a Multiresolution Approach
arXiv:1811.08397 · doi:10.1021/acs.jpca.8b11310
Abstract
Broadly speaking, the calculation of core spectra such as electron energy loss spectra (EELS) at the level of density functional theory (DFT) usually relies one of two approaches: conceptually more complex but computationally efficient projector augmented wave based approaches, or more straightforward but computationally more intensive all election (AE) based approaches. In this work we present an alternative method, which aims to find a middle ground between the two. Specifically, we have implemented an approach in the multi-wavelet MADNESS molecular DFT code which permits a combination of atoms treated at the AE and pseudopotential (PSP) level. Atoms for which one wishes to calculate the core edges are thus treated at an AE level, while the remainder can be treated at the PSP level. This is made possible thanks to the multiresolution approach of MADNESS, which permits accurate and efficient calculations at both the AE and PSP level. Through examples of a small molecule and a carbon nanotube we demonstrate the potential applications of our approach.
References in corpus (9)
- Separable Dual Space Gaussian Pseudo-potentials
- Relativistic separable dual-space Gaussian Pseudopotentials from H to Rn
- Precision and efficiency in solid-state pseudopotential calculations
- Libxc: a library of exchange and correlation functionals for density functional theory
- Daubechies wavelets as a basis set for density functional pseudopotential calculations
- Norm-conserving pseudopotentials with chemical accuracy compared to all-electron calculations
- The Elephant in the Room of Density Functional Theory Calculations
- MADNESS: A Multiresolution, Adaptive Numerical Environment for Scientific Simulation
- Calculating optical absorption spectra for large systems using linear-scaling density-functional theory