Search

We investigate the behavior of small water droplets impacting fabrics of different roughness and coating type in a range of impacting conditions, and compare it to measurements of static and roll-off angles. Fabrics with low roughness display lower static contact angle, no roll-off, a slightly slower retraction rate for impacting droplets, a narrower Weber number (We) range in which rebound occurs, and a limited maximum rebound at intermediate We. Rougher fabrics exhibit slightly higher static contact angles, lower roll-off angles, a slightly faster retraction rate for impacting droplets, and a larger We range for rebound; additionally, the restitution of kinetic energy upon rebound is proportional to 1/We, indicating a near-to-constant rebound height independent of the impact velocity. As regards coating type, roll-off angle, We range for rebound, and to a lesser extent restitution coefficient, depend on coating type, with the more environmentally-friendly wax-based coating being superior to silicone and perfluorobutyl-based ones in both drop impact and roll-off experiments. Our results thus provide hope that perfluorinated coatings can be eliminated in standard textiles, and provide an extensive view of how roughness, coating and impact conditions control the fate of small droplets falling on textiles.

WO

The solar absorber is a key component in a solar cell as it captures photons and converts them into electron–hole pairs. Its efficiency is driven by the carrier lifetime and the latter is controlled by Shockley–Read–Hall non-radiative processes, which involve defects. Here, we present an ab initio high-throughput screening approach to search for new high-efficiency photovoltaic absorbers taking into account carrier lifetime and recombination through defects. We first show that our methodology can distinguish poor and highly efficient solar absorbers. We then use our approach to identify a handful of defect-tolerant, high carrier lifetime, absorbers among more than 7000 Cu-based known materials. We highlight K3Cu3P2 and Na2CuP as they combine earth-abundance and the potential for high efficiency. Further analysis of our data articulates two challenges in discovering Cu-based solar absorbers: deep anti-site defects lowering the carrier lifetime and low formation-energy copper vacancies leading to metallic behavior. The alkali copper phosphides and pnictides offer unique chemistries that tackle these two issues.

The mobility of carriers, as limited by their scattering with phonons, can now routinely be obtained from first-principles electron-phonon coupling calculations. However, so far, most computations have relied on some form of simplification of the linearized Boltzmann transport equation based on either the self-energy, the momentum- or constant relaxation time approximations. Here, we develop a high-throughput infrastructure and an automatic workflow and we compute 69 phonon-limited mobilities in semiconductors. We compare the results resorting to the approximations with the exact iterative solution. We conclude that the approximate values may deviate significantly from the exact ones and are thus not reliable. Given the minimal computational overhead, our work encourages to rely on this exact iterative solution and warns on the possible inaccuracy of earlier results reported using relaxation time approximations., 6 pages, 2 figures

WO3 is the state of the art of electrochromic oxide materials finding technological application in smart windows. In this work, a set of WO3 thin films were deposited by magnetron sputtering by varying total pressure, oxygen partial pressure, and power. On each film two properties were measured, the electrochemical reversibility and the blue color persistence of LixWO3 films in simulated ambient conditions. With the help of machine learning, prediction maps for such electrochromic properties, namely, color persistence and reversibility, were designed. High-performance WO3 films were targeted by a global score which is the product of these two properties. The combined approach of experimental measurements and machine learning led to a complete picture of electrochromic properties depending of sputtering parameters providing an efficient tool in regards to time saving.

Machine-learning models have recently encountered enormous success for predicting the properties of materials. These are often trained based on data that present various levels of accuracy, with typically much less high- than low-fidelity data. In order to extract as much information as possible from all available data, we here introduce an approach which aims to improve the quality of the data through denoising. We investigate the possibilities that it offers in the case of the prediction of the band gap using both limited experimental data and density-functional theory relying on different exchange-correlation functionals. After analyzing the raw data thoroughly, we explore different ways to combine the data into training sequences and analyze the effect of the chosen denoiser. We also study the effect of applying the denoising procedure several times until convergence. Finally, we compare our approach with various existing methods to exploit multi-fidelity data and show that it provides an interesting improvement.

By investigating the vacancy and substitutional defects in monolayer WS$_2$ with hybrid functionals, we find that there is no unique amount of Fock exchange that concurrently satisfies the generalized Koopmans' condition and reproduces the band gap and band-edge positions. Fixing the mixing parameter of Fock exchange based upon the band gap can lead to qualitatively incorrect defect physics in two-dimensional materials. Instead, excellent agreement is achieved with experiment and many-body perturbation theory within $GW$ approximation once the mixing parameters are tuned individually for the defects and the band edges. We show the departure from a unique optimized mixing parameter is inherent to two-dimensional systems as the band edges experience a reduced screening whilst the localized defects are subject to bulklike screening., Comment: Main manuscript (7 pages, 3 figures) with supplemental material (3 pages, 5 figures)

Summary The progress of Cu(In,Ga)S2 remains significantly limited mainly due to photovoltage (Voc) losses in the bulk and at the interfaces. Here, via a combination of photoluminescence, cathodoluminescence, electrical measurements, and ab initio modeling, we address the bulk and interface losses to improve ∼1.6-eV-band-gap (Eg) Cu(In,Ga)S2. The optoelectronic quality of the absorber improves upon reducing the [Cu]/[Ga+In] (CGI) ratio, as manifested by the suppression of deep defects, higher quasi-Fermi level splitting (QFLS), improved charge-carrier lifetime, and higher Voc. We identify antisite CuIn/CuGa as a major performance-limiting deep defect by comparing the formation energies of various intrinsic defects. Interface recombination is suppressed using a Zn(O,S) buffer layer in Cu-poor devices, which leads to the activation energy of recombination equal to the Eg. We demonstrate an efficiency of 15.2% with Voc of 902 mV from a H2S-free, Cd-free, and KCN-free process.

Transparent conducting materials (TCMs) are required in many applications from solar cells to transparent electronics. Developing high performance materials combining the antagonistic properties of transparency and conductivity has been challenging especially for p-type materials. Recently, high-throughput ab initio computational screening has emerged as a formidable tool for accelerating materials discovery. In this review, we discuss how this approach has been applied for identifying TCMs. We provide a brief overview of the different materials properties of importance for TCMs (e.g., dopability, effective mass, and transparency) and present the ab initio techniques available to assess them. We focus on the accuracy of the methodologies as well as their suitability for high-throughput computing. Finally, we review the different high-throughput computational studies searching for new TCMs and discuss their differences in terms of methodologies and main findings.

Combining ferroelectricity with other properties such as visible light absorption or long-range magnetic order requires the discovery of new families of ferroelectric materials. Here, through the analysis of a high-throughput database of phonon band structures, we identify a new structural family of anti-Ruddlesden-Popper phases A$_4$X$_2$O (A=Ca, Sr, Ba, Eu, X=Sb, P, As, Bi) showing ferroelectric and anti-ferroelectric behaviors. The discovered ferroelectrics belong to the new class of hyperferroelectrics which polarize even under open-circuit boundary conditions. The polar distortion involves the movement of O anions against apical A cations and is driven by geometric effects resulting from internal chemical strains. Within this new structural family, we show that Eu$_4$Sb$_2$O combines coupled ferromagnetic and ferroelectric order at the same atomic site, a very rare occurrence in materials physics., 21 pages, 4 figures with supplemental information attached

The Zintl phases, Yb14 MSb11 (M = Mn, Mg, Al, Zn), are now some of the highest thermoelectric efficiency p-type materials with stability above 873 K. Yb14MnSb11 gained prominence as the first p-type thermoelectric material to double the efficiency of SiGe alloy, the heritage material in radioisotope thermoelectric generators used to power NASA's deep space exploration. This study investigates the solid solution of Yb14Mg1-x Al x Sb11 (0 ≤ x ≤ 1), which enables a full mapping of the metal-to-semiconductor transition. Using a combined theoretical and experimental approach, we show that a second, high valley degeneracy (N v = 8) band is responsible for the groundbreaking performance of Yb14 MSb11 This multiband understanding of the properties provides insight into other thermoelectric systems (La3-x Te4, SnTe, Ag9AlSe6, and Eu9CdSb9), and the model predicts that an increase in carrier concentration can lead to zT > 1.5 in Yb14 MSb11 systems.

We have studied the growth of epitaxial films of stannate pyrochlores with a general formula A 2 Sn 2 O 7 (A = La and Y) and find that itis possible to incorporate ∼ 25% excess of the A-site constituent; in contrast, any tin excess is expelled. We unravel the defect chemistry,allowing for the incorporation of excess A-site species and the mechanism behind the tin expulsion. An A-site surplus is manifested by ashift in the film diffraction peaks, and the expulsion of tin is apparent from the surface morphology of the film. In an attempt to increaseLa 2 Sn 2 O 7 conductivity through n-type doping, substantial quantities of tin have been substituted by antimony while maintaining good filmquality. The sample remained insulating as explained by first-principles computations, showing that both the oxygen vacancy and antimony-on-tin substitutional defects are deep. Similar conclusions are drawn on Y 2 Sn 2 O 7 . An alternative n-type dopant, fluorine on oxygen, is shallowaccording to computations and more likely to lead to electrical conductivity. The bandgaps of stoichiometric La 2 Sn 2 O 7 and Y 2 Sn 2 O 7 filmswere determined by spectroscopic ellipsometry to be 4.2 eV and 4.48 eV, respectively.

As the number of novel data-driven approaches to material science continues to grow, it is crucial to perform consistent quality, reliability and applicability assessments of model performance. In this paper, we benchmark the Materials Optimal Descriptor Network (MODNet) method and architecture against the recently released MatBench v0.1, a curated test suite of materials datasets. MODNet is shown to outperform current leaders on 4 of the 13 tasks, whilst closely matching the current leaders on a further 3 tasks; MODNet performs particularly well when the number of samples is below 10,000. Attention is paid to two topics of concern when benchmarking models. First, we encourage the reporting of a more diverse set of metrics as it leads to a more comprehensive and holistic comparison of model performance. Second, an equally important task is the uncertainty assessment of a model towards a target domain. Significant variations in validation errors can be observed, depending on the imbalance and bias in the training set (i.e., similarity between training and application space). By using an ensemble MODNet model, confidence intervals can be built and the uncertainty on individual predictions can be quantified. Imbalance and bias issues are often overlooked, and yet are important for successful real-world applications of machine learning in materials science and condensed matter.

The controversial role of cation disorder in the extraordinarily low open-circuit voltage (VOC) of the Cu2ZnSnS4 (CZTS) kesterite absorber is examined through a statistical treatment of disorder within the cluster-expansion method. It is demonstrated that the extensive Cu–Zn disorder alone cannot be responsible for the large Urbach tails observed in many CZTS solar cells. While the band gap is reduced as a result of the Gaussian tails formed near the valence-band edge due to Cu clustering, band-gap fluctuations contribute only marginally to the VOC deficit, thereby excluding Cu–Zn disorder as the primary source of the low efficiency of CZTS devices. On the other hand, the extensive disorder stabilizes the formation of SnZn antisite and its defect complexes, which as nonradiative recombination and minority carrier trapping centers dominate the VOC loss in CZTS. Our analysis indicates that current CZTS devices might have already approached the maximum conversion efficiency (14%) given the limited growth conditions and the remnant cation disorder even after postannealing. In view of the improved efficiency achieved with CZTS-derived kesterite absorbers, the methodology presented in this work offers an avenue to understanding and optimizing these emerging kesterite solar devices towards higher efficiency.

International audience; Cu2O nanostructures of cubic or octahedral shape as well as decorated with gold nanoparticles were synthesized by a chemical process. The electrochemical activity of these nanostructures for the oxygen evolution reaction (OER) in alkaline media was assessed in the dark and under solar light irradiation. Electrodes modified with cubic-shaped Cu2O and Cu2O-Au nanostructures revealed both enhanced electrocatalytic OER activity over the octahedral-shaped ones, for which hydroxylation of the (111) surfaces might hinder the OER as supported by numerical computations. Illumination of Cu2O nanostructures with solar light did not enhance the electrocatalytic OER. Cu2O-Au nanostructures, however, showed improved OER activity with an overpotential of 200 mV (10 mA cm−2) and a Tafel slope of 97 mV dec−1. The enhanced OER activity was ascribed to increased light absorption due to the plasmonic properties of Cu2O-Au cubes.

We report a giant thermoelectric figure of merit ZT (up to six at 1100 K) in n-doped lanthanum oxysulphate LaSO. Thermoelectric coefficients are computed from ab initio bands within Bloch-Boltzmann theory in an energy-, chemical potential-, and temperature-dependent relaxation time approximation. The lattice thermal conductivity is estimated from a model employing the ab initio phonon and Grüneisen-parameter spectrum. The main source of the large ZT is the significant power factor which correlates with a large band complexity factor. We also suggest a possible n-type dopant for the material based on ab initio calculations., Project funded by the Région Wallonne (Programmes FEDER) and from CISM and CECI for computational support. V.F., R.F., and G.C. thank CINECA for ISCRA supercomputing grants. R.F. thanks ICMN-UCL for hospitality. V.F. is on secondment leave at the Italian Embassy to Germany; his views as expressed herein are not necessarily shared by the Italian Ministry of Foreign Affairs.

Traditionally, chemistry problems are solved by means of a deductive approach. The question to be addressed is typically related to the value of a property that is either measured experimentally, computed using quantum-chemistry software, or (more recently) predicted using a machine-learned model. In this paper, we demonstrate that an inductive approach can be adopted using End-to-End (E2E) machine learning. This approach is illustrated for tackling the following chemistry problems: (i) determine the fully coordinated (FC) and undercoordinated (UC) atoms in a molecule with one missing atom, (ii) identify the type of atom that is missing in such an incomplete molecule, and (iii) predict the direction of a reaction between two molecules according to an existing dataset. The E2E approach leads to accuracies higher than 99%, 98%, and 93% for these three problems, respectively. Finally, in order to achieve such accuracies, a descriptor for the molecules, called bag of clusters, is introduced and compared with a series previously proposed descriptors, highlighting a series of advantages.

The typical strategy to design high performance thermoelectric materials is to dope a semiconducting material until optimal properties are obtained. However, some known thermoelectric materials such as La3Te4, Mo3Sb7, Yb14MnSb11, and NbCoSb are actually gapped metals, i.e., their band structure displays a gap slightly above or below the band crossed by the Fermi level. This key feature makes these metals comparable to degenerate semiconductors and thus suitable for thermoelectric applications. In this work, we perform a computational high-throughput search for such gapped metals exhibiting attractive thermoelectric properties. Several thousands of metals are found to present this key feature, and about one thousand of them show decent thermoelectric properties as evaluated by a computed zT. We present the different chemistry of gapped metals we discovered such as clathrates, Chevrel phases, or transition metal dichalcogenides and discuss their previous studies as thermoelectric and their potential as new thermoelectric materials.

A new tool called ChemEnv, which can identify coordination environments in a fast and robust manner, is presented., Coordination or local environments have been used to describe, analyze and understand crystal structures for more than a century. Here, a new tool called ChemEnv, which can identify coordination environments in a fast and robust manner, is presented. In contrast to previous tools, the assessment of the coordination environments is not biased by small distortions of the crystal structure. Its robust and fast implementation enables the analysis of large databases of structures. The code is available open source within the pymatgen package and the software can also be used through a web app available on http://crystaltoolkit.org through the Materials Project.

The transformation from crystalline SrIrO3 to active amorphous iridium oxide electrocatalyst occurs via the lattice oxygen redox., The use of renewable electricity to prepare materials and fuels from abundant molecules offers a tantalizing opportunity to address concerns over energy and materials sustainability. The oxygen evolution reaction (OER) is integral to nearly all material and fuel electrosyntheses. However, very little is known about the structural evolution of the OER electrocatalyst, especially the amorphous layer that forms from the crystalline structure. Here, we investigate the interfacial transformation of the SrIrO3 OER electrocatalyst. The SrIrO3 amorphization is initiated by the lattice oxygen redox, a step that allows Sr2+ to diffuse and O2− to reorganize the SrIrO3 structure. This activation turns SrIrO3 into a highly disordered Ir octahedral network with Ir square-planar motif. The final SryIrOx exhibits a greater degree of disorder than IrOx made from other processing methods. Our results demonstrate that the structural reorganization facilitated by coupled ionic diffusions is essential to the disordered structure of the SrIrO3 electrocatalyst.

Boron phosphide has recently been identified as a potential high-hole-mobility transparent conducting material. This promise arises from its low hole effective masses. However, boron phosphide has a relatively small, 2 eV, indirect bandgap which will affect its transparency. In this work, we computationally study both optical absorption across the indirect gap and phonon-limited electronic transport to quantify the potential of boron phosphide as a $p$-type transparent conductor. We find that phonon-mediated indirect optical absorption is weak in the visible spectrum and that the phonon-limited hole mobility is very high (around 900 ${\mathrm{cm}}^{2}$/Vs) at room temperature. This exceptional mobility comes from the combination of a low hole effective mass and very weak scattering by polar phonon modes. We rationalize the weak scattering by the less ionic bonding in boron phosphide compared to oxides. We suggest that this could be a general advantage of nonoxides for $p$-type transparent conducting applications. Using our computed properties, we assess the transparent conductor figure of merit of boron phosphide and show that it exceeds by one order of magnitude that of established $p$-type transparent conductors, confirming the potential of this material.

We include the treatment of quadrupolar fields beyond the Fröhlich interaction in the first-principles electron-phonon vertex in semiconductors. Such quadrupolar fields induce long-range interactions that have to be taken into account for accurate physical results. We apply our formalism to Si (nonpolar), GaAs, and GaP (polar) and demonstrate that electron mobilities show large errors if dynamical quadrupoles are not properly treated., G. B., G.-M. R., and G. H. acknowledge financial support from F.R.S.-FNRS. H. P. C. M. acknowledges financial support from F.R.S.-FNRS through the PDR Grants HTBaSE (T.1071.15). M. G., M. J. V., and X. G. acknowledge financial support from F.R.S.-FNRS through the PDR Grants AIXPHO (T.0238.13) and ALPS (T.0103.19). M. J. V. thanks FNRS and ULiege for a sabbatical grant in ICN2. M. S. and M. R. acknowledge the support of Ministerio de Economia, Industria y Competitividad (MINECO-Spain) through Grants No. MAT2016-77100-C2-2-P and No. SEV-2015-0496; of Generalitat de Catalunya (Grant No. 2017 SGR1506); and of the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 724529). The present research benefited from computational resources made available on the Tier-1 supercomputer of the Federation WallonieBruxelles, infrastructure funded by the Walloon Region under Grant Agreement No. 1117545.

Paper published as part of the special topic on Electronic Structure Software, ABINIT is probably the first electronic-structure package to have been released under an open-source license about 20 years ago. It implements density functional theory, density-functional perturbation theory (DFPT), many-body perturbation theory (GW approximation and Bethe–Salpeter equation), and more specific or advanced formalisms, such as dynamical mean-field theory (DMFT) and the “temperaturedependent effective potential” approach for anharmonic effects. Relying on planewaves for the representation of wavefunctions, density, and other space-dependent quantities, with pseudopotentials or projector-augmented waves (PAWs), it is well suited for the study of periodic materials, although nanostructures and molecules can be treated with the supercell technique. The present article starts with a brief description of the project, a summary of the theories upon which ABINIT relies, and a list of the associated capabilities. It then focuses on selected capabilities that might not be present in the majority of electronic structure packages either among planewave codes or, in general, treatment of strongly correlated materials using DMFT; materials under finite electric fields; properties at nuclei (electric field gradient, Mössbauer shifts, and orbital magnetization); positron annihilation; Raman intensities and electro-optic effect; and DFPT calculations of response to strain perturbation (elastic constants and piezoelectricity), spatial dispersion (flexoelectricity), electronic mobility, temperature dependence of the gap, and spin-magnetic-field perturbation. The ABINIT DFPT implementation is very general, including systems with van der Waals interaction or with noncollinear magnetism. Community projects are also described: generation of pseudopotential and PAW datasets, high-throughput calculations (databases of phonon band structure, second-harmonic generation, and GW computations of bandgaps), and the library LIBPAW. ABINIT has strong links with many other software projects that are briefly mentioned., This work (A.H.R.) was supported by the DMREF-NSF Grant No. 1434897, National Science Foundation OAC-1740111, and U.S. Department of Energy DE-SC0016176 and DE-SC0019491 projects. N.A.P. and M.J.V. gratefully acknowledge funding from the Belgian Fonds National de la Recherche Scientifique (FNRS) under Grant No. PDR T.1077.15-1/7. M.J.V. also acknowledges a sabbatical “OUT” grant at ICN2 Barcelona as well as ULiège and the Communauté Française de Belgique (Grant No. ARC AIMED G.A. 15/19-09). X.G. and M.J.V. acknowledge funding from the FNRS under Grant No. T.0103.19-ALPS. X.G. and G.-M. R. acknowledge support from the Communauté française de Belgique through the SURFASCOPE Project (No. ARC 19/24-057). X.G. acknowledges the hospitality of the CEA DAM-DIF during the year 2017. G.H. acknowledges support from the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division under Contract No. DE-AC02-05-CH11231 (Materials Project Program No. KC23MP). The Belgian authors acknowledge computational resources from supercomputing facilities of the University of Liège, the Consortium des Equipements de Calcul Intensif (Grant No. FRS-FNRS G.A. 2.5020.11), and Zenobe/CENAERO funded by the Walloon Region under Grant No. G.A. 1117545. M.C. and O.G. acknowledge support from the Fonds de Recherche du Québec Nature et Technologie (FRQ-NT), Canada, and the Natural Sciences and Engineering Research Council of Canada (NSERC) under Grant No. RGPIN-2016-06666. The implementation of the libpaw library (M.T., T.R., and D.C.) was supported by the ANR NEWCASTLE project (Grant No. ANR-2010-COSI-005-01) of the French National Research Agency. M.R. and M.S. acknowledge funding from Ministerio de Economia, Industria y Competitividad (MINECO-Spain) (Grants Nos. MAT2016-77100-C2-2-P and SEV-2015-0496) and Generalitat de Catalunya (Grant No. 2017 SGR1506). This work has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation program (Grant Agreement No. 724529). P.G. acknowledges support from FNRS Belgium through PDR (Grant No. HiT4FiT), ULiège and the Communauté française de Belgique through the ARC project AIMED, the EU and FNRS through M.ERA.NET project SIOX, and the European Funds for Regional Developments (FEDER) and the Walloon Region in the framework of the operational program “Wallonie-2020.EU” through the project Multifunctional thin films/LoCoTED. The Flatiron Institute is a division of the Simons Foundation. A large part of the data presented in this paper is available directly from the Abinit Web page www.abinit.org. Any other data not appearing in this web page can be provided by the corresponding author upon reasonable request.

Ba2BiTaO6 is a transparent p-type oxide recently discovered and exhibiting attractive hole mobility but low carrier concentration. Using first-principles computations, we study how defects influence the carrier concentration in Ba2BiTaO6. The calculated defect formation energies confirm that K is an adequate p-type shallow extrinsic dopant but that high p-type doping is prevented by the presence of compensating, ‘‘hole-killing’’, intrinsic defects: O vacancies but also Ta on Bi anti-sites. Our work stresses the inherent difficulty in doping Ba2BiTaO6 to high carrier concentration and discusses a few avenues towards this goal.

We describe a new approach to compute the electron-phonon self-energy and carrier mobilities in semiconductors. Our implementation does not require a localized basis set to interpolate the electron-phonon matrix elements, with the advantage that computations can be easily automated. Scattering potentials are interpolated on dense q meshes using Fourier transforms and ab initio models to describe the long-range potentials generated by dipoles and quadrupoles. To reduce significantly the computational cost, we take advantage of crystal symmetries and employ the linear tetrahedron method and double-grid integration schemes, in conjunction with filtering techniques in the Brillouin zone. We report results for the electron mobility in Si, GaAs, and GaP obtained with this new methodology., G.B., G.-M.R., and G.H. acknowledge financial support from F.R.S.-FNRS. H.P.C.M. acknowledges financial support from F.R.S.-FNRS through the PDR Grants HTBaSE (T.1071.15). M.G., M.J.V., and X.G. acknowledge financial support from F.R.S.-FNRS through the PDR Grants AIXPHO (T.0238.13) and ALPS (T.0103.19). M.J.V. thanks FNRS and ULiege for a sabbatical grant in ICN2. M.S. and M.R. acknowledge the support of Ministerio de Economia, Industria y Competitividad (MINECO-Spain) through Grants No. MAT2016-77100-C2-2-P and No. SEV-2015-0496; of Generalitat de Catalunya (Grant No. 2017 SGR1506); and of the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 724529).

Abinit is a material- and nanostructure-oriented package that implements density-functional theory (DFT) and many-body perturbation theory (MBPT) to find, from first principles, numerous properties including total energy, electronic structure, vibrational and thermodynamic properties, different dielectric and non-linear optical properties, and related spectra. In the special issue to celebrate the 40th anniversary of CPC, published in 2009, a detailed account of Abinit was included [Gonze et al. (2009)], and has been amply cited. The present article comes as a follow-up to this 2009 publication. It includes an analysis of the impact that Abinit has had, through for example the bibliometric indicators of the 2009 publication. Links with several other computational materials science projects are described. This article also covers the new capabilities of Abinit that have been implemented during the last three years, complementing a recent update of the 2009 article published in 2016. Physical and technical developments inside the abinit application are covered, as well as developments provided with the Abinit package, such as the multibinit and a-tdep projects, and related Abinit organization developments such as AbiPy . The new developments are described with relevant references, input variables, tests, and tutorials. Program summary Program Title: Abinit Program Files doi: http://dx.doi.org/10.17632/csvdrr4d68.1 Licensing provisions: GPLv3 Programming language: Fortran2003, Python Journal reference of previous version: X .Gonze et al, Comput. Phys. Commun. 205 (2016) 106–131 Does the new version supersede the previous version?: Yes. The present 8.10.3 version is now the up-to-date stable version of abinit , and supercedes the 7.10.5 version. Reasons for the new version: New developments Summary of revisions: • Many new capabilities of the main abinit application, related to density-functional theory, density-functional perturbation theory, GW, the Bethe-Salpeter equation, dynamical mean-field theory, etc. • New applications in the package: multibinit (second-principles calculations)and tdep (temperature-dependent properties) Nature of problem: Computing accurately material and nanostructure properties: electronic structure, bond lengths, bond angles, primitive cell, cohesive energy, dielectric properties, vibrational properties, elastic properties, optical properties, magnetic properties, non-linear couplings, electronic and vibrational lifetimes, etc. For large-scale systems, second-principles calculations, building upon the first-principles results, are also possible. Solution method: Software application based on density-functional theory and many-body perturbation theory, pseudopotentials, with plane waves or wavelets as basis functions. Different real-time algorithms are implemented for second-principles calculations.

To improve the precision of machine-learning predictions, we investigate various techniques that combine multiple quality sources for the same property. In particular, focusing on the electronic band gap, we aim at having the lowest error by taking advantage of all available experimental measurements and density-functional theory calculations. We show that learning about the difference between high- and low-quality values, considered a correction, significantly improves the results compared to learning on the sole high-quality experimental data. As a preliminary step, we also introduce an extension of the MODNet model, which consists of using a genetic algorithm for hyperparameter optimization. Thanks to this, MODNet is shown to achieve excellent performance on the Matbench test suite.

The Pauling rules have been used for decades to rationalise the crystal structures of ionic compounds. Despite their importance, there has been no statistical assessment of the performances of these five empirical rules so far. Here, we rigorously and automatically test all five Pauling rules for a large data set of around 5000 known oxides. We discuss each Pauling rule separately, stressing their limits and range of application in terms of chemistries and structures. We conclude that only 13 % of the oxides simultaneously satisfy the last four rules, indicating a much lower predictive power than expected., Exceptions to the rules: The statistical analysis of the five Pauling rules reveals their limited predictive power. Each of the Pauling rules is tested on a set of 5000 experimentally known oxides. Their limits in terms of chemistries and structures are discussed in detail. Only 13 % of all tested structures fulfil the last four rules simultaneously.

The diffusion of large databases collecting different kind of material properties from high-throughput density functional theory calculations has opened new paths in the study of materials science thanks to data mining and machine learning techniques. Phonon calculations have already been employed successfully to predict materials properties and interpret experimental data, e.g. phase stability, ferroelectricity and Raman spectra, so their availability for a large set of materials will further increase the analytical and predictive power at hand. Moving to a larger scale with density functional perturbation calculations, however, requires the presence of a robust framework to handle this challenging task. In light of this, we automatized the phonon calculation and applied the result to the analysis of the convergence trends for several materials. This allowed to identify and tackle some common problems emerging in this kind of simulations and to lay out the basis to obtain reliable phonon band structures from high-throughput calculations, as well as optimizing the approach to standard phonon simulations.