Mario G. Zauchner
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mario G. Zauchner has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Andrew P. Horsfield
Imperial College London
3 shared publications - Tchavdar N. Todorov
Queen’s University Belfast
2 shared publications Andrew Horsfield
1 shared publicationGábor Cśanyi
1 shared publicationJohannes Lischner
1 shared publication- Johannes Lischner
Imperial College London
1 shared publication
Latest Publications
- Open Access
- 7 October 2023
- Published by Springer Nature in npj Computational Materials
- Vol. 9 (1) , 1-9
Accelerating GW calculations through machine-learned dielectric matrices
The GW approach produces highly accurate quasiparticle energies, but its application to large systems is computationally challenging due to the difficulty in computing the inverse dielectric matrix. To address this challenge, we develop a machine learning approach to efficiently predict density–density response functions (DDRF) in materials. An atomic decomposition of the DDRF is introduced, as well as the neighborhood density–matrix descriptor, both of which transform in the same way under rotations. The resulting DDRFs are then used to evaluate quasiparticle energies via the GW approach. To assess the accuracy of this method, we apply it to hydrogenated silicon clusters and find that it reliably reproduces HOMO–LUMO gaps and quasiparticle energy levels. The accuracy of the predictions deteriorates when the approach is applied to larger clusters than those in the training set. These advances pave the way for GW calculations of complex systems, such as disordered materials, liquids, interfaces, and nanoparticles.
- 7 October 2023
- Published by Springer Nature in npj Computational Materials
- Vol. 9 (1) , 1-9
Open AccessAccelerating GW calculations through machine-learned dielectric matrices
The GW approach produces highly accurate quasiparticle energies, but its application to large systems is computationally challenging due to the difficulty in computing the inverse dielectric matrix. To address this challenge, we develop a machine learning approach to efficiently predict density–density response functions (DDRF) in materials. An atomic decomposition of the DDRF is introduced, as well as the neighborhood density–matrix descriptor, both of which transform in the same way under rotations. The resulting DDRFs are then used to evaluate quasiparticle energies via the GW approach. To assess the accuracy of this method, we apply it to hydrogenated silicon clusters and find that it reliably reproduces HOMO–LUMO gaps and quasiparticle energy levels. The accuracy of the predictions deteriorates when the approach is applied to larger clusters than those in the training set. These advances pave the way for GW calculations of complex systems, such as disordered materials, liquids, interfaces, and nanoparticles.
- Open Access
- 22 October 2021
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 2 (4) , 045029
Predicting polarizabilities of silicon clusters using local chemical environments
- 22 October 2021
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 2 (4) , 045029
Open AccessPredicting polarizabilities of silicon clusters using local chemical environments
- 12 January 2018
- Published by American Physical Society (APS) in Physical Review B
- Vol. 97 (4) , 045116
Efficient electron open boundaries for simulating electrochemical cells
Nonequilibrium electrochemistry raises new challenges for atomistic simulation: we need to perform molecular dynamics for the nuclear degrees of freedom with an explicit description of the electrons, which in turn must be free to enter and leave the computational cell. Here we present a limiting form for electron open boundaries that we expect to apply when the magnitude of the electric current is determined by the drift and diffusion of ions in a solution and which is sufficiently computationally efficient to be used with molecular dynamics. We present tight-binding simulations of a parallel-plate capacitor with nothing, a dimer, or an atomic wire situated in the space between the plates. These simulations demonstrate that this scheme can be used to perform molecular dynamics simulations when there is an applied bias between two metal plates with, at most, weak electronic coupling between them. This simple system captures some of the essential features of an electrochemical cell, suggesting this approach might be suitable for simulations of electrochemical cells out of equilibrium.
- 12 January 2018
- Published by American Physical Society (APS) in Physical Review B
- Vol. 97 (4) , 045116
Efficient electron open boundaries for simulating electrochemical cells
Nonequilibrium electrochemistry raises new challenges for atomistic simulation: we need to perform molecular dynamics for the nuclear degrees of freedom with an explicit description of the electrons, which in turn must be free to enter and leave the computational cell. Here we present a limiting form for electron open boundaries that we expect to apply when the magnitude of the electric current is determined by the drift and diffusion of ions in a solution and which is sufficiently computationally efficient to be used with molecular dynamics. We present tight-binding simulations of a parallel-plate capacitor with nothing, a dimer, or an atomic wire situated in the space between the plates. These simulations demonstrate that this scheme can be used to perform molecular dynamics simulations when there is an applied bias between two metal plates with, at most, weak electronic coupling between them. This simple system captures some of the essential features of an electrochemical cell, suggesting this approach might be suitable for simulations of electrochemical cells out of equilibrium.
- Open Access
Efficient electron open boundaries for simulating electrochemical cells
Non-equilibrium electrochemistry raises new challenges for atomistic simulation: we need to perform molecular dynamics for the nuclear degrees of freedom with an explicit description of the electrons, which in turn must be free to enter and leave the computational cell. Here we present a limiting form for electron open boundaries that applies when the magnitude of the electric current is determined by the drift and diffusion of ions in solution, and which is sufficiently computationally efficient to be used with molecular dynamics. We demonstrate its capabilities by way of tight binding simulations of a parallel plate capacitor with and without a dimer situated in the space between the plates.
Open AccessEfficient electron open boundaries for simulating electrochemical cells
Non-equilibrium electrochemistry raises new challenges for atomistic simulation: we need to perform molecular dynamics for the nuclear degrees of freedom with an explicit description of the electrons, which in turn must be free to enter and leave the computational cell. Here we present a limiting form for electron open boundaries that applies when the magnitude of the electric current is determined by the drift and diffusion of ions in solution, and which is sufficiently computationally efficient to be used with molecular dynamics. We demonstrate its capabilities by way of tight binding simulations of a parallel plate capacitor with and without a dimer situated in the space between the plates.