Leonid Komissarov^1*, Toon Verstraelen^1*

1 Center for Molecular Modeling (CMM), Ghent University, Technologiepark-Zwijnaarde 46, B-9052, Ghent, Belgium

* Corresponding authors emails: leonid.komissarov@ugent.be, toon.verstraelen@ugent.be

DOI10.24435/materialscloud:14-4m [version v1]

Publication date: Sep 21, 2021

How to cite this record

Leonid Komissarov, Toon Verstraelen, Improving the silicon interactions of GFN-xTB, Materials Cloud Archive 2021.152 (2021), doi: 10.24435/materialscloud:14-4m.

Description

This record addresses inaccuracies in the widely-used GFN-xTB model when it comes to the description of organosilicon compounds. Here, an ab initio reference data set of 10000 compounds is provided alongside files needed to execute a parameter fitting scheme that improves geometries, nuclear gradients and energies predicted by the GFN-xTB Hamiltonian.

Materials Cloud sections using this data

Files

File name	Size	Description
gfn-xtb-si.tar.gz MD5md5:fbda97ff283aed0e27c264e7d438a70f	201.9 MiB	Data repository, individual hash sums, parametrization files
README.md MD5md5:e7c83282c92dc43f11ad3580aa107956	2.2 KiB	Readme with per-file descriptions and data format documentation

License

Files and data are licensed under the terms of the following license: GNU General Public License v3.0 only.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Keywords

organosilicon parameter optimization DFT density-functional theory gfn-xtb gfn1-xtb Horizon Europe Marie Curie Fellowship H2020 DFTB