With the growth of computational resources, the scope of electronic structure simulations has increased greatly. Artificial intelligence and robust data analysis hold the promise to accelerate large-scale simulations and their analysis to hitherto unattainable scales. Machine learning is a rapidly growing field for the processing of such complex data sets. It has recently gained traction in the domain of electronic structure simulations, where density functional theory (DFT) takes the prominent role of the most widely used electronic structure method. Thus, DFT calculations represent one of the largest loads on academic high-performance computing systems across the world. Accelerating these with machine learning can reduce the resources required and enables simulations of larger systems. Hence, the combination of DFT and machine learning has the potential to rapidly advance electronic structure applications such as in silico materials discovery and the search for new chemical reaction pathways. We provide the theoretical background of both DFT and machine learning on a generally accessible level. This serves as the basis of our comprehensive review, including research articles up to December 2020 in chemistry and materials science that employ machine-learning techniques. In our analysis, we categorize the body of research into main threads and extract impactful results. We conclude our review with an outlook on exciting research directions in terms of a citation analysis.

In this progress report, we examine recent efforts towards the introduction of magnetic order into topologically nontrivial systems through magnetic proximity effects, with a particular emphasis on the methods of characterizing magnetization induced at the interface. We focus on the challenges associated with comparing magneto-transport measurements with magnetic scattering and spectroscopy techniques, considering the limitations and potential artifacts associated with topological insulator heterostructures. Taking the ${(\mathrm{Bi},\mathrm{Sb})}_2{(\mathrm{Se},\mathrm{Te})}_3$ family of three-dimensional topological insulators as an example, we discuss the results associated with a wide range of magnetically ordered reservoirs and highlight the wide discrepancies in reported magnetic proximity effect strengths detected using different characterization techniques. Finally, we discuss the outlook of magnetic proximity effects in topological insulator heterostructures as a route towards a higher-temperature quantum anomalous Hall effect.

Metallic glasses have drawn the attention of researchers due to their high strength, hardness, high corrosion, and wear resistance, and ease of thermoplastic forming compared to their counterpart crystalline alloys. While these characteristics are desirable for materials used in extreme environments that often need high corrosion resistance and mechanical strength at elevated temperatures, metallic glasses have found limited use because they crystallize at lower temperatures than the operational temperatures of these applications. Upon crystallization, these characteristics are lost. In recent years, refractory metallic glasses with higher crystallization temperatures have been synthesized, making them potential candidates for use in extreme environments, such as in space and nuclear applications. This review summarizes recently reported metallic glass systems with crystallization temperatures above 700 °C. These systems are discussed in terms of glass forming ability, mechanical properties, and other properties relevant to their use in high temperature applications. This article highlights the challenges in the development and implementation of high-temperature metallic glasses, and discusses the needs and research opportunities that exist for expanding our knowledge of these glasses, including potential routes for improving glass forming ability and mechanical properties.

Pyrochlore lattices, which are found in two important classes of materials—the $A_2{B}_2{X}_7$ pyrochlore family and the $A{B}_2{X}_4$ spinel family—are the quintessential three-dimensional frustrated lattice architecture. While historically oxides ($X=\mathrm{O}$) have played the starring role in this field, the past decade has seen materials synthesis breakthroughs that have lead to the emergence of fluoride ($X=\mathrm{F}$) and chalcogenide ($X=\mathrm{S}$, Se) pyrochlore lattice materials. In this Research Update, we summarize recent progress in understanding the magnetically frustrated ground states in three families of nonoxide pyrochlore lattice materials: (i) $3d$-transition metal fluoride pyrochlores, (ii) rare-earth chalcogenide spinels, and (iii) chromium chalcogenide spinels with a breathing pyrochlore lattice. We highlight how the change of anion can modify the single ion spin anisotropy due to crystal electric field effects, stabilize the incorporation of new magnetic elements, and dramatically alter the exchange pathways. These three factors in combination lead to a range of new magnetic ground states in nonoxide pyrochlores. We also consider a range of future directions—materials with the potential to define the next decade of research in frustrated magnetism.

Halide perovskites are promising absorber materials for photovoltaics due to their high efficiency and low-temperature solution processing, though the instabilities inherent to these systems remain a primary obstacle to their utilization. In this research update, we review the chemical, structural, and functional effects of fluoride and fluorinated organic additives in halide perovskites, especially on stability and photovoltaic performance. We detail the evidence that, when incorporated as its anion, fluoride is typically localized to perovskite surfaces and grain boundaries. Due to its high electronegativity, fluoride and fluorinated species can impart strong ionic and intermolecular bonding with perovskite materials, passivating surfaces and diminishing their susceptibility to further chemical reactivity. Additionally, fluorocarbon-based additives impart hydrophobicity to perovskite films and serve as physiochemical barriers to reactive infiltrating species. To understand how these effects operate in different contexts, herein we summarize the dual role surface defects play in both diminishing photovoltaic efficiency and introducing structural instabilities in various $AB{X}_3$ perovskite halide materials: hybrid organic/inorganic lead halide, all-inorganic cesium lead halide, and tin halide perovskites. We review three avenues of fluorine inclusion in perovskites: fluoride substitution at the anion site, partial incorporation of fluorinated organic cations at the $A$ site, and large- and small-molecule fluorocarbon-based additives. We conclude with a perspective on the prospects for fluorine-based enhancement of halide perovskites and suggestions for new analytical and synthetic chemistries that will deepen our understanding and control of these systems.

A scanning probe with an atomically sharp tip is used in atomic force microscopy. As the probe is rastered over the specimen of interest, changes in its vibration mode can be ascribed to changes in the interatomic forces between the specimen and the tip. Various modes for implementing atomic force measures exist. One of the most successful modes is frequency-modulation noncontact atomic force microscopy. Here we address the role of replicating accurate forces on the probe tip to simulate images within this mode. We examine several approaches. One involves the use of classical force fields based on interatomic potentials, where the potentials are often fit to experiment. Other approaches center on the use of forces generated from quantum-based calculations. We consider the development of a “virtual tip” approximation wherein the probe tip senses electrostatic forces generated by the specimen. This approach treats the specimen quantum mechanically, while the tip is treated as a classical object. We also consider the use of an embedding approximation to avoid direct computations of the specimen in the presence of the tip. In this method, the tip and specimen are treated quantum mechanically. However, the electronic structure of the specimen is computed in the absence of the probe tip. The electronic density of the specimen is then fixed and the electronic structure of the tip is then computed in the presence of the fixed density. As a general approach, we outline new methods for the full computation of quantum forces between the tip and specimen. We note that quantum-based forces are particularly important as they directly contain the chemical nature of the forces present without the use of any adjustable parameters. We illustrate recent algorithmic developments to computing quantum forces that can produce accurate simulations of atomic force microscopy images for large and complex molecular species. We also suggest new pathways to overcome current challenges in this rapidly evolving field.

Chemical doping of graphene is the most robust way of modifying graphene's electronic properties. We review here the results obtained so far on the electronic structure and transport properties of chemically doped graphene, focusing on the results obtained with scanning tunneling microscopy/spectroscopy (STM/S), angle-resolved photoemission spectroscopy (ARPES), and magnetoresistance measurements. The majority of the results reported have been obtained on nitrogen-doped samples, but boron-doped graphene has also been well documented. Aside from the appearance of the dopant on STM topographic images, the main questions that have been addressed are the atomic configurations of the doping and their doping efficiency (number of electrons/holes brought to the graphene lattice). Both can be addressed by a local probe such as STM/S. The doping efficiency has also been complementarily studied via direct visualization of the band structure with ARPES. The effect of the dopants on the electronic transport properties and in particular their influence on the scattering mechanisms is also presented. Finally, avenues for future research efforts are suggested.

The electronic defects in any semiconductor play a decisive role for the usability of this material in an optoelectronic device. Electronic defects determine the doping level as well as the recombination centers of a solar cell absorber. $\mathrm{Cu}\left(\mathrm{In},\mathrm{Ga}\right)\mathrm{S}{\mathrm{e}}_2$ is used in thin-film solar cells with high and stable efficiencies. The electronic defects in this class of materials have been studied experimentally by photoluminescence, admittance, and photocurrent spectroscopies for many decades now. The literature results are summarized and compared to new results by photoluminescence of deep defects. These observations are related to other experimental methods that investigate the physicochemical structure of defects. To finally assign the electronic defect signatures to actual physicochemical defects, a comparison with theoretical predictions is necessary. In recent years the accuracy of these calculations has greatly improved by the use of hybrid functionals. A comprehensive model of the electronic defects in $\mathrm{Cu}\left(\mathrm{In},\mathrm{Ga}\right)\mathrm{S}{\mathrm{e}}_2$ is proposed based on experiments and theory. The consequences for solar cell efficiency are discussed.

High-entropy alloys (HEAs) are a recently opened research area in materials science and condensed matter physics. Although $3d$-metal-based HEAs have already been the subject of many investigations, studies of HEA superconductors, which tend to be based on $4d$ metals, are relatively fewer. Here we provide a short update of the progress made in studies of superconducting HEAs. We aim to summarize their current status and describe some of the key factors that appear to influence their superconducting transition temperatures and properties, including crystal structure, atomic makeup, valence electron count, molar volume, and mixing entropy. Many opportunities and challenges remain for expanding our knowledge of HEA superconductors, finding new types of HEA superconductors, and their potential for applications; these are also briefly discussed.

Since the 1960s advances in electronic and optoelectronic device technologies have been primarily orchestrated by III-V semiconductors, which have led to an age of consumer electronic devices with unprecedented social and economic impacts. Group III-V semiconductors such as GaAs, GaN, InAs, and GaP and their solid solution alloys are not only the building blocks of modern solid-state lighting, photodetectors, sensors, and high-speed power-electronic and optoelectronic devices, they have also been actively researched and developed for over six decades to understand and innovate fundamental materials science, physics, and device engineering properties. Yet there is a widespread realization today that contemporary grand challenges of our society such as energy efficient electronics and computing, secure information processing, energy security, imaging, sensing, etc., require more advanced materials and better device integration technologies. At the same time, several important device technologies of the modern era such as thermoelectricity that converts waste heat into electrical energy, plasmonic materials, and devices that could be utilized to harvest optical energy in solar photovoltaics, solar thermophotovoltaics, photocatalysis, etc., also require materials and heterostructure metamaterials that are not possible to achieve with traditional III-V semiconductors. Scandium nitride (ScN) is a group 3 rocksalt nitride semiconductor, which can overcome some of the limitations of traditional III-V semiconductors, and could lead to novel device functionalities. However, unlike other well-known III-V semiconductors, very little attention has been devoted to understand and engineer ScN's physical properties until very recently. In this research update, we detail the progress that has taken place over the last several years to overcome the materials engineering challenges for high-quality epitaxial ScN thin-film growth, analysis of its physical properties, and epitaxial integration of ScN with other rocksalt metallic nitrides. Along with the attractive physical properties common to most transition-metal nitrides such as high hardness, large melting temperature, and chemical, thermal, and morphological stability, ScN also exhibits rocksalt crystal structure with octahedral bonding coordination, indirect band gap, preferential $n$-type and $p$-type doping, and the ability to epitaxially integrate with other metallic materials (such as TiN, ZrN, HfN, etc.) to deposit single-crystalline epitaxial metal/semiconductor multilayers and superlattices without the presence of extended defects. All of these advances could lead to ScN based materials and devices with improved efficiencies and industrial applications.

Cadmium arsenide $\left({\mathrm{Cd}}_3{\mathrm{As}}_2\right)$, a time-honored and widely explored material in solid-state physics, has recently attracted considerable attention. This was triggered by a theoretical prediction concerning the presence of three-dimensional (3D) symmetry-protected massless Dirac electrons, which could turn ${\mathrm{Cd}}_3{\mathrm{As}}_2$ into a 3D analog of graphene. Subsequent extended experimental studies have provided us with compelling experimental evidence of conical bands in this system, and revealed a number of interesting properties and phenomena. At the same time, some of the material properties remain the subject of vast discussions despite recent intensive experimental and theoretical efforts, which may hinder the progress in understanding and applications of this appealing material. In this paper, we focus on the basic material parameters and properties of ${\mathrm{Cd}}_3{\mathrm{As}}_2$, in particular those which are directly related to the conical features in the electronic band structure of this material. The outcome of experimental investigations, performed on ${\mathrm{Cd}}_3{\mathrm{As}}_2$ using various spectroscopic and transport techniques within the past 60 years, is compared with theoretical studies. These theoretical works gave us not only simplified effective models, but more recently, also the electronic band structure calculated numerically using ab initio methods.

We provide a brief discussion of “What is machine learning?” and then give a number of examples of how these methods have recently aided the design and discovery of new materials, such as new shape memory alloys, with enhanced targeted properties, such as lower hysteresis. These examples illustrate how discoveries can be made from large databases, for example, those generated by high throughput DFT calculations and also how they can be made from experimentally growing smaller databases in an active learning manner. Additionally, we discuss such advanced machine learning methods as multiobjective and multifidelity optimization that permit proposing new materials with the simultaneous optimization of more than one targeted property, such as a material with low hysteresis and high Curie temperature, and permit using fewer costly experiments and calculations by combining them with less costly ones to achieve modeling comparable to using only many costly ones. We conclude with a brief discussion of future machine learning opportunities in the context of high throughput experiment and on-the-fly adjustment of synthesis. More speculatively, we end by discussing how might we mesh materials science more fittingly with machine learning.