GFN2-xTB-An Accurate and Broadly Parametrized Self-Consistent Tight-Binding Quantum Chemical Method with Multipole Electrostatics and Density-Dependent Dispersion Contributions.
By Christoph Bannwarth, Sebastian Ehlert, Stefan Grimme. 2018
3,244 citations recorded by the source corpus.
An extended semiempirical tight-binding model is presented, which is primarily designed for the fast calculation of structures and noncovalent interaction energies for molecular systems with roughly 1000 atoms. The essential novelty in this so-called GFN2-xTB method is the inclusion of anisotropic second order density fluctuation effects via short-range damped interactions of cumulative atomic multipole moments. Without noticeable increase in the computational demands, this results in a less empirical and overall more physically sound method, which does not require any classical halogen or hydrogen bonding corrections and which relies solely on global and element-specific parameters (available up to radon, Z = 86). Moreover, the atomic partial charge dependent D4 London dispersion model is incorporated self-consistently, which can be naturally obtained in a tight-binding picture from second order density fluctuations. Fully analytical and numerically precise gradients (nuclear forces) are implemented. The accuracy of the method is benchmarked for a wide variety of systems and compared with other semiempirical methods. Along with excellent performance for the "target" properties, we also find lower errors for "off-target" properties such as barrier heights and molecular dipole moments. High computational efficiency along with the improved physics compared to its precursor GFN-xTB makes this method well-suited to explore the conformational space of molecular systems. Significant improvements are furthermore observed for various benchmark sets, which are prototypical for biomolecular systems in aqueous solution.
Subfields: Atomic and Molecular Physics, and Optics, Physical and Theoretical Chemistry
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Atomic and Molecular Physics, and Optics
- Christoph Bannwarth
- Stefan Grimme
- Andreas Hansen
- Sebastian Ehlert
- Alán Aspuru-Guzik
- K. Müller
- Pascal Friederich
- Florian Hase
In Physical and Theoretical Chemistry
- Stefan Grimme
- Christoph Bannwarth
- Andreas Hansen
- Alán Aspuru-Guzik
- Michael Gastegger
- Kristof T. Schütt
- Florian Hase
- Pascal Friederich
Citations in the corpus
- Leveraging large language models for predictive chemistry (2024, positive): “This HOMO–LUMO gap can be predicted accurately using semi-empirical quantum mechanics (GFN2-xTB 46 ), which is computationally affordable enough for us to compute for all generated molecules (Supplementary Note 77).”
- Machine-learned potentials for next-generation matter simulations (2021, background): “Computationally less costly semi-empirical and tight-binding methods 119–121 , as well as quantum mechanics/molecular mechanics (QM/MM) models 115,122–124 , are a promising approach to speed up calculations that we will discuss later on in the context of ML potentials.”
- SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects (2021, methodology): “Data sets used in ablation studies For the ablation studies in Section IV A, energies, forces, and dipoles for three new data sets were computed at the semi-empirical GFN2-xTB level of theory.(96) Both the Ag3 /Ag − 3 (see Fig.”
- Extended tight‐binding quantum chemistry methods (2020, background): “The GFN1- and GFN2-xTB methods already proved to be fast and robust semiempirical optimization tools for computing reasonably accurate molecular structures with elements up to radon.(49,53,106,107) To substantiate this statement, we summarize their performance for well-established and recently publi”
- Generating Molecular Conformer Fields (2023, methodology): “To elaborate, we calculate the following characteristics using xTB (as documented by (Bannwarth et al., 2019)): energy ( E ), dipole moment ( µ ), the gap between HOMO and LUMO ( ∆ ϵ ), and the lowest possible energy, denoted as E min .”