Session B47: Computational Design and Discovery of Novel Materials I

Theory and high-throughput discovery of new electride materials

Electrides are materials with electrons localized on interstitial sites in the crystal lattice and exhibit an array of interesting properties that show promise for applications such as catalysts, electron emitters and superconductors. Electrides can be found among many different classes of materials, including elemental metals at high pressure, organic crystals, intermetallic compounds and ceramic materials.

In this work, we present a general theory that provides a unified understanding of the origin of interstitial electrons in all currently known types of electrides. Using this theoretical understanding, we have devised an automated high-throughput algorithm for identifying electrides that only relies on knowledge of the composition and crystal structure. Applying our workflow to the ~55,000 materials in the Materials Project crystal structure database, we identified ~3,000 potential (mostly new) electrides. Analysing these candidate systems with first-principles electronic structure calculations, we find that ~50% of them (~1,500 materials) indeed possess interstitial electrons.   

Computational discovery of extreme-gap semiconductors

The magnitude of the band gap has been traditionally applied as a criterion to distinguish semiconductors from insulators; materials with gaps less than 3 eV are typically semiconductors, while wider-gap materials tend to be insulators. However, the development of ultra-wide-band-gap semiconductors such as AlGaN, diamond, BN, and Ga₂O₃ has challenged this gap-based criterion for materials classification and gave rise to questions such as how far the band gap of semiconductors can increase while maintaining delocalized carriers for conductivity and what is the widest band gap semiconductor. By applying high-throughput density functional theory calculation, we develop a materials discovery strategy to identify new extreme-gap semiconductors. We discover that materials composed of light elements, crystallized in simple, densely packed structures, and having s-orbital characteristics of conduction/valence bands have large band gap (> 7 eV) but light carrier effective mass (m_e^* < 0.7 m_e m_h^* < 2 m_e) that enable shallow dopants and high mobility and suppress the formation of polarons. We subsequently perform atomistic calculation to predict dopability and mobility of candidate materials. Our analysis uncovers several known compounds with gaps as wide as 11.3 eV that can host delocalized carriers such as rs-BeO and reveals that there is no practical upper limit to the band gap of semiconductors.   

Computational discovery of semiconducting high-entropy chalcogenide alloys

High-entropy materials are formed by mixing typically five or more principal components into a single crystal structure and show improved thermodynamic stability due to the large configurational disorder. While significant progress has been made to synthesize entropy-stabilized metals and ceramics for structural applications, little attention has been paid to the discovery of new semiconducting materials using the design principle of entropy stabilization.  Here, we present a new class of entropy-stabilized semiconducting alloys based on the IV-VI binary chalcogenides, namely Ge_xSn_yPb_1–x–yS_zSe_tTe_1–z–t high-entropy chalcogenides (HECs). By utilizing high-throughput first-principles calculations, we investigate the thermodynamic stability of HECs over their entire composition space, and show that more than 50% of the investigated compositions are stable with respect to phase segregation to the competing binary ingredients at the experimental synthesis temperature. We further studied the enthalpic effect of the individual elements and showed that Sn and Se lower the enthalpy of mixing, while S is detrimental to the phase stability of HECs. Our work demonstrates the potential of entropy stabilization in the discovery of novel multicomponent semiconductor alloys.   

Defect-Informed Descriptor for Predicting New High-Entropy Materials

Single-phase high-entropy materials (HEM) consisting of at least five elements in nearly equiatomic ratios constitute an enormously large but least explored compositional space. Over recent years, these materials have gained significant interest due to their potential applications and remarkable functional properties. Despite considerable progress has been made in understanding the salient features of these materials, predictive composition-structure-property relationships are rare, and the prediction of new HEM remains a grand challenge. In this work, we propose a new physical descriptor that can enable high-throughput screening of single-phase HEM. Based on the energy distribution spectrum of various defect configurations, this descriptor measures the relative propensity of forming random chemical disorders in a single phase. Applying this descriptor to disordered refectory five-metal carbides, all experimentally synthesized ones in single-phase have been successfully identified and separated from specific elemental combinations that are formed as multiple phases. This method can also be extended to the search for new higher-order (six or more metal elements) high-entropy systems without increasing computational costs.   

Predicting magnetic anisotropy energies using site-specific spin-orbit coupling energies and machine learning: Application to iron-cobalt nitrides

We perform high-throughput first-principles calculations to predict the magnetic anisotropy energies of a variety of iron-cobalt nitrides. We illustrate the efficacy of a spatial decomposition technique that divides the total magnetic anisotropy energy into spin-orbit coupling energy contributions from individual sites. The spatial decomposition scheme that we utilized works for a wide range of magnetic anisotropy energies. We also construct a machine-learning model by combining the site-specific spin-orbit coupling energies with structural details on each atomic site. We adopt the same approach to predicting site-specific magnetic moments. We show the capability of our machine-learning model to accelerate computational screening of candidate materials with high magnetization and large magnetic anisotropy energy.   

Discovery of rare-earth-free magnetic ternary compounds using machine learning assisted adaptive genetic algorithms

Finding new materials with desired properties is a challenging task owing to the vast number of possible compositions and crystal structures. In order to address this problem,  we outline a feedback loop scheme consisting of machine learning assisted high-throughput first-principles calculations and adaptive genetic algorithm. Our scheme enables efficient and accurate predictions of materials properties through a wide range of compositional and structural space, allowing the fast discovery of materials with desired properties. We illustrate the procedure to a ternary Fe-Co-B system, where we discovered hundreds of new metastable Fe-Co-B structures across the ternary phase space.   Many of  many of these new structures possess promising magnetic properties that can be used as rare-earth-free magnets.   

Line-graph-lattice crystal structures of stoichiometric materials

Many quantum-material phenomena may be intimately related to the presence of flat electronic bands. In quantum simulation, such bands have been realized through line-graph lattices. Based on that work, we conduct a high-throughput screening for line-graph lattices among the crystalline structures of the Materials Flatband Database and report on new candidates for line-graph materials and lattice models [1]. We find materials with line-graph-lattice structures beyond the well-known kagome and pyrochlore lattices. We identify the most frequently represented line-graph lattices and those with gapped flat bands. With the identification of real stoichiometric materials and theoretical lattice geometries, the results of this work may inform future studies of flat-band many-body physics in both condensed matter experiment and theory.

[1] C.S. Chiu, A.N. Carroll, N. Regnault, and A.A. Houck. arXiv: 2109.13940 (2021).   

High-throughput Discovery of a Promising Thermoelectric Heteroanionic Materials Family

Heteroanionic materials, or mixed-anion materials, have interesting properties due to the presence of multiple anions, but are much fewer experimentally reported comparing to single-anion compounds. Here, we demonstrate how high-throughput DFT calculations can accelerate the discovery of heteroanionic compounds created from homoanionic materials. CsCuLa₂Se₄ and CsCu₃Gd₂Se₅, two typical compounds in ACuLnQ (1124 and 1325, A=Cs, Rb; Ln=rare earth; Q=S, Se, Te) materials family, were discovered years ago. Using the Open Quantum Materials Database (OQMD) platform, our high-throughput study finds many unreported yet stable compounds for both 1124 and 1325 ACuLnQ materials. The thermal transport property calculations demonstrate that this material family is a good candidate as a thermoelectric material due to ultra-low thermal conductivity. Further, we find three different Wyckoff positions of anion sites in both 1124 and 1325 structures and create hypothetical mixed-anion materials by replacing all anions on one Wyckoff position with some other anion. High-throughput calculations predict 61 new stable and over 200 metastable mixed-anion compounds, largely expanding the phase space of ACuLnQQ’ materials family.   

Density Functional Theory for Determining Adsorptive Strength of Highly Porous Crystalline Materials

Perfluoroalkyl substances (PFAS) are a large family of chemicals which, due to their strong fluorinated carbon backbone, are difficult to remove through conventional means. They also cause adverse health effects, including immune system issues and cancer. This leads to a need for novel remediation techniques, including new materials. One potential candidate are metal-organic frameworks (MOFs), a large class of highly porous coordination polymers, owing to their high adsorption capacity, large variety of structures, and ability for functionalization. While MOFs show promise, research thus far has been limited to well-known MOF structures. It can be costly to synthesize new MOFs, and with higher risk of failure. In this work we used Density Functional Theory (DFT) to study the adsorption of PFAS molecules in MOFs and other porous materials to gain physical insights. In particular, we studied the MOF74 structure, as well as various covalent-organic frameworks (COFs). A high throughput approach is also being used to generate large amounts of data describing PFAS interactions with these porous materials. This lays the groundwork for future calculations to identify new porous materials for enhanced PFAS remediation.   

High throughput computational design of novel nitride perovskites

Perovskites constitute an exceptionally tunable materials family with diverse applications in electronics, optoelectronics, energy, and quantum technologies. Out of the thousands of known perovskites, the majority of compounds are oxides, halides, and chalcogenides. In contrast, only two nitride perovskites are currently known. In this work we employed an ab initio high-throughput workflow to screen for new semiconducting nitride perovskites. We identified two potential new ABN₃ semiconductors with band gaps approaching the Shockley-Queisser optimum values for maximizing photovoltaic energy conversion efficiency. We report a detailed analysis of thermodynamic stability, dynamical stability, and optical properties, and we show that these compounds are highly efficient light absorbers. The present findings reveal a potentially new class of nitride semiconductors with promise for electronics, optoelectronics, and light harvesting.   

Screening Fluoride Perovskites for use in Piezoelectrics using High-throughput Calculations.

The search for good piezoelectric materials without toxic elements such as lead is an active area of research. The high-performance piezoelectric lead zirconate titanate relies on the presence of a morphotropic phase boundary (MPB) in the composition-temperature phase diagram where the macroscopic polarization vector quasi continuously changes direction and/or magnitude, often coupled to a distinct softening of the corresponding elastic constants. A method to design materials that facilitates an MPB is to alloy two systems with different structural configuration in their respective ground states. The combination of the systems may then exhibit a desired MPB at some composition ratio between the two endpoint systems.

This talk presents a computational large-scale investigation of candidate perovskite alloys for potential use in high-performance piezoelectrics by first screening for viable endpoint systems, and then investigating promising alloys.