Efficient Automatized Density-Functional Tight-Binding Parametrizations: Application to Group IV Elements

Huran, Ahmad W.; Steigemann, Conrad; Frauenheim, Thomas; Aradi, Bálint; Marques, Miguel A. L.

doi:10.1021/acs.jctc.7b01269

Cited by 21 publications

(20 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…25 For this step, we used home-made high-quality Slater-Koster parameters, specifically fitted to obtain energies and forces with quasi-DFT precision. 35 The number of calculations also forced us to neglect spin-orbit coupling at this stage, although we verified a posteriori that this is acceptable (see below).…”

mentioning

confidence: 76%

Structural prediction of stabilized atomically thin tin layers

et al. 2019

Self Cite

View full text Add to dashboard Cite

The family of group IV two-dimensional materials shows a rich variety of structural, electronic and topological properties. Only graphene is stable in the honeycomb structure, while buckling and dumbbell configurations stabilize silicene and germanene. Here we investigate from first principles the lowest-energy atomic arrangements of atomically-thin tin layers. Our calculations are performed with a very efficient method for global structural prediction, combined with constrains that enforce the desired onedimensional confinement and include the effect of strain due to the substrate. We discover a series of new structures that span a large range of atomic densities and are considerably more stable than hexagonal single-or double-layer stanene, as well as dumbbell structures. The ground state, a metallic double layer with a square lattice that lies 295 meV/atom below honeycomb stanene and only 149 meV/atom above bulk α-tin, is akin to the atomic arrangement of a layer of romarchite tin oxide. Due to its enhanced stability with respect to honeycomb stanene, we propose that this structure can be easily synthesized on appropriate lattice-matched metallic substrates.

show abstract

mentioning

confidence: 76%

Structural prediction of stabilized atomically thin tin layers

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…A number of recent DFTB parameterizations have, in addition to fitting the repulsive potential, empirically adjusted parameters that define the electronic Hamiltonian during the fitting process. [53][54][55][56][57] Adjusted parameters include those that define the constraining potential and electron density cutoffs in the standard approach utilized to construct the DFTB electronic Hamiltonian from QC solutions for isolated atoms. 53,55,56 Empirical fits have also adjusted the atomic orbital energies, and the Hubbard parameters that specify electron-electron repulsion.…”

Section: Related Workmentioning

confidence: 99%

A Density Functional Tight Binding Layer for Deep Learning of Chemical Hamiltonians

Collins

Tanha

et al. 2018

J. Chem. Theory Comput.

109

104

View full text Add to dashboard Cite

Current neural networks for predictions of molecular properties use quantum chemistry only as a source of training data. This paper explores models that use quantum chemistry as an integral part of the prediction process. This is done by implementing selfconsistent-charge Density-Functional-Tight-Binding (DFTB) theory as a layer for use in deep learning models. The DFTB layer takes, as input, Hamiltonian matrix elements generated from earlier layers and produces, as output, electronic properties from self-consistent field solutions of the corresponding DFTB Hamiltonian. Backpropagation enables efficient training of the model to target electronic properties. Two types of input to the DFTB layer are explored, splines and feed-forward neural networks. Because overfitting can cause models trained on smaller molecules to perform poorly on larger molecules, regularizations are applied that penalize non-monotonic behavior and deviation of the Hamiltonian matrix elements from those of the published DFTB model used to initialize the model. The approach is evaluated on 15,700 hydrocarbons by comparing the root mean square error in energy and dipole moment, on test molecules with 8 heavy atoms, to the error from the initial DFTB model. When trained on molecules with up to 7 heavy atoms, the spline model reduces the test error in energy by 60% and in dipole moments by 42%. The neural network model performs somewhat better, with error reductions of 67% and 59% respectively. Training on molecules with up to 4 heavy atoms reduces performance, with both the spline and neural net models reducing the test error in energy by about 53% and in dipole by about 25%. arXiv:1808.04526v2 [physics.chem-ph]

show abstract

“…Huran et al. employed a pattern‐search algorithm for parameterizing the electronic and repulsive potential at the same time for Group IV elements …”

Section: Dftb Parameterizationmentioning

confidence: 99%

“…The confining radii in the electronic parameters were conventionally heuristically chosen but in the recent years they have been usually optimized to electronic band structures prior to the optimization of the repulsive potentials. However, recent developments have shown that it may be necessary to optimize both electronic and repulsive parts simultaneously to have better performance and transferability due to error compensation ,,,. Thus, we tried to optimize the confining potentials to reproduce the bonding angles between the Li cation and the selected ligands including the standard H 2 O, NH 3 molecules, and common used organic electrolyte molecules.…”

Section: Illustrative Applicationsmentioning

confidence: 99%

Development of Divide‐and‐Conquer Density‐Functional Tight‐Binding Method for Theoretical Research on Li‐Ion Battery

Chou

Sakti

Nishimura

et al. 2018

The Chemical Record

View full text Add to dashboard Cite

The density‐functional tight‐binding (DFTB) method is one of the useful quantum chemical methods, which provides a good balance between accuracy and computational efficiency. In this account, we reviewed the basis of the DFTB method, the linear‐scaling divide‐and‐conquer (DC) technique, as well as the parameterization process. We also provide some refinement, modifications, and extension of the existing parameters that can be applicable for lithium‐ion battery systems. The diffusion constants of common electrolyte molecules and LiTFSA salt in solution have been estimated using DC‐DFTB molecular dynamics simulation with our new parameters. The resulting diffusion constants have good agreement to the experimental diffusion constants.

show abstract

Efficient Automatized Density-Functional Tight-Binding Parametrizations: Application to Group IV Elements

Cited by 21 publications

References 66 publications

Structural prediction of stabilized atomically thin tin layers

Structural prediction of stabilized atomically thin tin layers

A Density Functional Tight Binding Layer for Deep Learning of Chemical Hamiltonians

Development of Divide‐and‐Conquer Density‐Functional Tight‐Binding Method for Theoretical Research on Li‐Ion Battery

Contact Info

Product

Resources

About