Our group works in the development and application of state-of-the-art ab-initio methods to systems of both fundamental and technological interest. Until 2014 the group was located at the Institut Lumière Matière, situated at the Université Claude Bernard Lyon 1, in Lyon, France. Since then we are at the Institut für Physik of the Martin-Luther-Universität Halle-Wittenberg. We are also members of the European Theoretical Spectroscopy Facility.

We are also sponsored by the following institutions:



Highlights (all highlights)

May 31, 2021

Atomically Thin Pythagorean Tilings in Two Dimensions

Pythagorian lattice We performied a theoretical study of an atomically thin, two-dimensional layer obtained by positioning atoms at the vertices of the classical Pythagorean tiling. This leads to an unusual geometrical pattern that is only stable for the three halogens Cl, Br, and I. In this Pythagorean structure, halogen atoms are arranged in strongly bound diatomic units that bind together by weaker electrostatic bonds. The energy of these phases is competitive with those of the low-temperature phase of the halogens and the two-dimensional layer obtained by exfoliating it. The Pythagorean layers are semiconducting, with an unusual band structure composed of very mobile holes and extremely heavy electrons. They are also soft, exhibiting small values of the elastic constants and a very low energy flexural mode. Analysis of the allowed Raman transitions reveals breathing-like modes that might be used to fingerprint, experimentally, the Pythagorean structure. Finally, we presented a series of substrates that, due to lattice matching and compatible symmetry, can be used to stabilize these peculiar two-dimensional layers. This work has just been published in the J. Phys. Chem. Lett..

November 17, 2020

Finding new crystalline compounds using chemical similarity

Finding new crystalline compounds using chemical similarity We proposed an efficient high-throughput scheme for the discovery of new stable crystalline phases. Our approach is based on the transmutation of known compounds, through the substitution of atoms in the crystal structure with chemically similar ones. The concept of similarity is defined quantitatively using a measure of chemical replaceability, extracted by data mining experimental databases. In this way we built 189981 possible crystal phases, including 18479 that are on the convex hull of stability. The resulting success rate of 9.72% is at least one order of magnitude better than the usual success rate of systematic high-throughput calculations for a specific family of materials, and comparable with speed-up factors of machine learning filtering procedures. As a first characterization of the set of 18479 new stable compounds, we calculated their electronic band gaps, magnetic moments, and hardness. Our approach, that can be used as a filter on top of any high-throughput scheme, enabled us to efficiently extract stable compounds from tremendously large initial sets, without any initial assumption on their crystal structures or chemical compositions. This work has just been accepted in NPJ Comput. Mater.. Structural data can be downloaded from here.

February 04, 2019

Special issue in honor of Eberhard K.U. Gross for his 65th birthday

Hardy Gross With this special issue of the European Physical Journal B we pay homage to the scientific career of Eberhard Kurt Ulrich (Hardy) Gross, on occasion of his 65th birthday. Hardy is one of the most influential researchers in the field of theoretical density functional theory (DFT). His significant contributions started early, already as a student of Reiner Dreizler in Frankfurt and as a post-doc with the Nobel prize laureate Walter Kohn. In those years, Hardy Gross contributed to the birth of time-dependent density functional theory (TDDFT), DFT for superconductors, ensemble DFT, etc. Later, he was interested in other topics of electronic structure theory. This issue contains original contributions with topics close to Hardy’s heart (some of them already mentioned above),and is a mixture of colloquium and research papers. This special issue has just been published in the Eur. Phys. J. B.

February 20, 2018

Local hybrid density functional for interfaces

Local hybrid density functional for interfaces Hybrid functionals in density functional theory are becoming the state-of-the-art for the calculation of electronic properties of solids. The key of their performance is the way in which an amount of Fock exchange is mixed with semi-local exchange-correlation functionals. We propose here a local mixing dependent on the density alone, extending the results of a previously reported functional [Phys. Rev. B 83, 035119 (2011)] to enable accurate calculations for interfaces and nanostructures. We verify that this hybrid functional has the potential to yield results of comparable quality as GW for band alignments and defects energy levels at interfaces, at the reduced cost of a hybrid density functional. This is possible as the form of the mixing is derived from GW theory, accounting for the electronic screening through its dependence on a density estimator of the local dielectric function. In contrast with other recent self-consistent schemes for the mixing parameter, our approach does not require to calculate the dielectric function and therefore it leads to a negligible increase of the computation time. This work has just been accepted in J. Chem. Theory Comput..

December 01, 2017

Libxc 4.0

Libxc We are happy two annouce the release of version 4.0 of libxc. This is a library of exchange-correlation functionals for density-functional theory. We are concerned with semi-local functionals (or the semi-local part of hybrid functionals), namely local-density approximations, generalized- gradient approximations, and meta-generalized-gradient approximations. Currently we include around 400 functionals for the exchange, correlation, and the kinetic energy, spanning more than 50 years of research. Moreover, libxc is by now included in more than 20 codes, not only from the atomic, molecular, and solid-state physics, but also from the quantum chemistry community. This work has just been published in Software X. The software can be downloaded here.

May 29, 2017

Predicting the stability of solids with machine learning

Predicting the stability of solids with machine learning We perform a large scale benchmark of machine learning methods for the prediction of the thermodynamical stability of solids. We start by constructing a data set that comprises density functional theory calculations of around 250000 cubic perovskite systems. This includes all possible perovskite and anti-perovskite crystals that can be generated with elements from hydrogen to bismuth, and neglecting rare gases and lanthanides. Incidentally, these calculations already reveal a large number of systems (around 500) that are thermodynamically stable, but that are not present in crystal structure databases. Moreover, some of these phases have unconventional compositions and define completely new families of perovskites. This data set is then used to train and test a series of machine learning algorithms to predict the energy distance to the convex hull of stability. In particular, we study the performance of ridge regression, random forests, extremely randomized trees (including adaptive boosting), and neural networks. We find that extremely randomized trees give the best results, achieving errors in the test set of around 120 meV/atom when trained in 20000 prediction accuracy is not uniform across the periodic table, being worse for first-row elements and elements forming magnetic compounds. Our results point to the fact that machine learning can be successfully used to guide high-throughput density functional theory calculations to speed up by at least a factor of 5 systematic searches of new materials, without any degradation of the accuracy. This work has just been accepted in Chemistry of Materials.