An assessment of numerical behaviors of the r<sup>2</sup>SCAN exchange-correlation functional
ORAL
Abstract
The meta-GGA exchange-correlation functional, r2SCAN, [1] has received considerable attention. Accessments of its use in generalized density functional simulations of a variety of materials report typically improved agreement with experimental measurements compared with other functionals (Ref. [2], for example). However, numerical challenges due to the functional form of r2SCAN have also been reported [3,4], which perhaps has slowed its incorporation into several of the public-domain electronic structure codes. We report a systematic study of some of the numerical issues associated with using the r2SCAN exchange-correlation functional within generalized density functional computations.
1. J. W. Furness, et al., J. Phys. Chem. Lett. 11, 8208 (2020).
2. M. Kothakonda, et al., ACS Mater. Au 3, 102 (2023).
3. N. A. W. Holzwarth, M. Torrent, et al., Phys. Rev. Mat. 105, 125124 (2022).
4. S. Lehtola, J.Chem. Theory Comput. 19, 2502 (2023).
1. J. W. Furness, et al., J. Phys. Chem. Lett. 11, 8208 (2020).
2. M. Kothakonda, et al., ACS Mater. Au 3, 102 (2023).
3. N. A. W. Holzwarth, M. Torrent, et al., Phys. Rev. Mat. 105, 125124 (2022).
4. S. Lehtola, J.Chem. Theory Comput. 19, 2502 (2023).
*Support: NSF grant DMR-2242959 and Wake Forest University High Performance Computing Facility DOI: 10.57682/G13Z-2362. Also note M. Torrent's additional affiliation: LMCE, Paris-Saclay University in France.
–
Presenters
-
Natalie W Holzwarth
- Wake Forest University