Quantifying chemical short-range order in metallic alloys.

Metallic alloys often form phases-known as solid solutions-in which chemical elements are spread out on the same crystal lattice in an almost random manner. The tendency of certain chemical motifs to be more common than others is known as chemical short-range order (SRO), and it has received substantial consideration in alloys with multiple chemical elements present in large concentrations due to their extreme configurational complexity (e.g., high-entropy alloys). SRO renders solid solutions "slightly less random than completely random," which is a physically intuitive picture, but not easily quantifiable due to the sheer number of possible chemical motifs and their subtle spatial distribution on the lattice. Here, we present a multiscale method to predict and quantify the SRO state of an alloy with atomic resolution, incorporating machine learning techniques to bridge the gap between electronic-structure calculations and the characteristic length scale of SRO. The result is an approach capable of predicting SRO length scale in agreement with experimental measurements while comprehensively correlating SRO with fundamental quantities such as local lattice distortions. This work advances the quantitative understanding of solid-solution phases, paving the way for the rigorous incorporation of SRO length scales into predictive mechanical and thermodynamic models.

Machine learning magnetism classifiers from atomic coordinates.

The determination of magnetic structure poses a long-standing challenge in condensed matter physics and materials science. Experimental techniques such as neutron diffraction are resource-limited and require complex structure refinement protocols, while computational approaches such as first-principles density functional theory (DFT) need additional semi-empirical correction, and reliable prediction is still largely limited to collinear magnetism. Here, we present a machine learning model that aims to classify the magnetic structure by inputting atomic coordinates containing transition metal and rare earth elements. By building a Euclidean equivariant neural network that preserves the crystallographic symmetry, the magnetic structure (ferromagnetic, antiferromagnetic, and non-magnetic) and magnetic propagation vector (zero or non-zero) can be predicted with an average accuracy of 77.8% and 73.6%. In particular, a 91% accuracy is reached when predicting no magnetic ordering even if the structure contains magnetic element(s). Our work represents one step forward to solving the grand challenge of full magnetic structure determination.

E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials.

This work presents Neural Equivariant Interatomic Potentials (NequIP), an E(3)-equivariant neural network approach for learning interatomic potentials from ab-initio calculations for molecular dynamics simulations. While most contemporary symmetry-aware models use invariant convolutions and only act on scalars, NequIP employs E(3)-equivariant convolutions for interactions of geometric tensors, resulting in a more information-rich and faithful representation of atomic environments. The method achieves state-of-the-art accuracy on a challenging and diverse set of molecules and materials while exhibiting remarkable data efficiency. NequIP outperforms existing models with up to three orders of magnitude fewer training data, challenging the widely held belief that deep neural networks require massive training sets. The high data efficiency of the method allows for the construction of accurate potentials using high-order quantum chemical level of theory as reference and enables high-fidelity molecular dynamics simulations over long time scales.

Anisotropic 2D excitons unveiled in organic-inorganic quantum wells.

Two-dimensional (2D) excitons arise from electron-hole confinement along one spatial dimension. Such excitations are often described in terms of Frenkel or Wannier limits according to the degree of exciton spatial localization and the surrounding dielectric environment. In hybrid material systems, such as the 2D perovskites, the complex underlying interactions lead to excitons of an intermediate nature, whose description lies somewhere between the two limits, and a better physical description is needed. Here, we explore the photophysics of a tuneable materials platform where covalently bonded metal-chalcogenide layers are spaced by organic ligands that provide confinement barriers for charge carriers in the inorganic layer. We consider self-assembled, layered bulk silver benzeneselenolate, [AgSePh]∞, and use a combination of transient absorption spectroscopy and ab initio GW plus Bethe-Salpeter equation calculations. We demonstrate that in this non-polar dielectric environment, strongly anisotropic excitons dominate the optical transitions of [AgSePh]∞. We find that the transient absorption measurements at room temperature can be understood in terms of low-lying excitons confined to the AgSe planes with in-plane anisotropy, featuring anisotropic absorption and emission. Finally, we present a pathway to control the exciton behaviour by changing the chalcogen in the material lattice. Our studies unveil unexpected excitonic anisotropies in an unexplored class of tuneable, yet air-stable, hybrid quantum wells, offering design principles for the engineering of an ordered, yet complex dielectric environment and its effect on the excitonic phenomena in such emerging materials.

Direct Prediction of Phonon Density of States With Euclidean Neural Networks.

Machine learning has demonstrated great power in materials design, discovery, and property prediction. However, despite the success of machine learning in predicting discrete properties, challenges remain for continuous property prediction. The challenge is aggravated in crystalline solids due to crystallographic symmetry considerations and data scarcity. Here, the direct prediction of phonon density-of-states (DOS) is demonstrated using only atomic species and positions as input. Euclidean neural networks are applied, which by construction are equivariant to 3D rotations, translations, and inversion and thereby capture full crystal symmetry, and achieve high-quality prediction using a small training set of ≈103 examples with over 64 atom types. The predictive model reproduces key features of experimental data and even generalizes to materials with unseen elements, and is naturally suited to efficiently predict alloy systems without additional computational cost. The potential of the network is demonstrated by predicting a broad number of high phononic specific heat capacity materials. The work indicates an efficient approach to explore materials' phonon structure, and can further enable rapid screening for high-performance thermal storage materials and phonon-mediated superconductors.

An automatically curated first-principles database of ferroelectrics.

Ferroelectric materials have technological applications in information storage and electronic devices. The ferroelectric polar phase can be controlled with external fields, chemical substitution and size-effects in bulk and ultrathin film form, providing a platform for future technologies and for exploratory research. In this work, we integrate spin-polarized density functional theory (DFT) calculations, crystal structure databases, symmetry tools, workflow software, and a custom analysis toolkit to build a library of known, previously-proposed, and newly-proposed ferroelectric materials. With our automated workflow, we screen over 67,000 candidate materials from the Materials Project database to generate a dataset of 255 ferroelectric candidates, and propose 126 new ferroelectric materials. We benchmark our results against experimental data and previous first-principles results. The data provided includes atomic structures, output files, and DFT values of band gaps, energies, and the spontaneous polarization for each ferroelectric candidate. We contribute our workflow and analysis code to the open-source python packages atomate and pymatgen so others can conduct analogous symmetry driven searches for ferroelectrics and related phenomena.

Realization of a three-dimensional spin-anisotropic harmonic honeycomb iridate.

Spin and orbital quantum numbers play a key role in the physics of Mott insulators, but in most systems they are connected only indirectly--via the Pauli exclusion principle and the Coulomb interaction. Iridium-based oxides (iridates) introduce strong spin-orbit coupling directly, such that these numbers become entwined together and the Mott physics attains a strong orbital character. In the layered honeycomb iridates this is thought to generate highly spin-anisotropic magnetic interactions, coupling the spin to a given spatial direction of exchange and leading to strongly frustrated magnetism. Here we report a new iridate structure that has the same local connectivity as the layered honeycomb and exhibits striking evidence for highly spin-anisotropic exchange. The basic structural units of this material suggest that a new family of three-dimensional structures could exist, the 'harmonic honeycomb' iridates, of which the present compound is the first example.