High-Throughput DFT-Based Discovery of Next Generation Two-Dimensional (2D) Superconductors

Abstract

High-throughput density functional theory (DFT) calculations allow for a systematic search for conventional superconductors. With the recent interest in two-dimensional (2D) superconductors, we used a high-throughput workflow to screen over 1000 2D materials in the JARVIS-DFT database and performed electron−phonon coupling calculations, using the McMillan−Allen− Dynes formula to calculate the superconducting transition temperature (T c ) for 165 of them. Of these 165 materials, we identify 34 dynamically stable structures with transition temperatures above 5 K, including materials such as W 2 N 3 , NbO 2 , ZrBrO, TiClO, NaSn 2 S 4 , Mg 2 B 4 C 2 , and the previously unreported Mg 2 B 4 N 2 (T c = 21.8 K). Finally, we performed experiments to determine the T c of selected layered superconductors (2H-NbSe 2 , 2H-NbS 2 , ZrSiS, FeSe) and discuss the measured results within the context of our DFT results. We aim that the outcome of this workflow can guide future computational and experimental studies of new and emerging 2D superconductors by providing a roadmap of high-throughput DFT data.

Inverse Design of Next-Generation Superconductors Using Data-Driven Deep Generative Models

“…We show the relationship between EPC parameters (λ, ω log , T c ) for the 61 materials we performed DFT calculations for in Figure c and d. Figure c depicts an inverse relationship between λ and ω log , and in Figure d, we observe a somewhat positive relationship between λ and T c . These are typical behaviors of BCS superconductors and were observed in our work on BCS bulk and 2D superconductors. , From the colormap of Figure c and d, it is clear that a balance of high λ and ω log is a necessary condition for a material to have a high T c . It is important to note that our focus on ambient condition stoichiometric BCS superconductors significantly limits the value of T c that can be achieved.…”

“…After superconductivity was discovered in 1911 by Onnes, the efforts to identify novel superconducting materials with high transition temperatures ( T c ) has been an intense area of research in materials science and condensed matter physics. , There have been systematic computational efforts to identify Bardeen–Cooper–Schrieffer (BCS) conventional superconductors , with high T c prior to costly experimental investigation, − where density functional theory-perturbation theory (DFT-PT) calculations have been performed to obtain the electron–phonon coupling (EPC) parameters. In addition, various machine learning approaches have been utilized to accelerate the search for high- T c superconductors. ,− However, these typical funnel-like screening-based approaches are not sufficient for inverse materials design, where, instead of engineering from structure to property, the goal is to engineer from a target property to the crystal structure.…”

“…JARVIS (Joint Automated Repository for Various Integrated Simulations, ) is a collection of databases and tools to automate materials design using classical force-field, density functional theory, machine learning calculations, and experiments. JARVIS-DFT is a density functional theory based database of over 75,000 materials with several material properties such as formation energy, band gap with different levels of theory, solar-cell efficiency, topological spin–orbit coupling spillage, − elastic tensors, dielectric tensors, piezoelectric tensors, infrared and Raman spectra, electric field gradients, exfoliation energies, two-dimensional (2D) magnets, and bulk and 2D superconductors, all with stringent DFT-convergence setup . Our ALIGNN and CDVAE models were trained on the 1058 DFT calculations of superconducting properties presented in ref .…”

“…The superconducting transition temperature, T c , can then be approximated using the McMillan–Allen–Dynes equation as follows: T normalc = ω log 1.2 .25em exp [ − 1.04 false( 1 + λ false) λ − μ * false( 1 + 0.62 λ false) ] where ω log = exp [ ∫ d ω α 2 F false( ω false) ω ln ⁡ ω ∫ d ω α 2 F false( ω false) ω ] In eq , the parameter μ* is the effective Coulomb potential parameter, which we take as 0.1. It is important to note that the robustness of this workflow was heavily benchmarked against experimental data and higher levels of theory in refs and , which indicates that the training data for these deep learning models is of high quality, given the level of theory used to produce the data.…”

High-Throughput First-Principles Prediction of Interfacial Adhesion Energies in Metal-on-Metal Contacts

“…Moreover in recent times, alongside the fundamental combination of theory and experiments on specific problems, high-throughput density functional theory (DFT) approaches are used in materials design and discovery. − Increasingly complex DFT-based high-throughput workflows have been developed, screening molecular adsorption energies and sites on intermetallic surfaces in view of electrochemical catalysis applications, looking for novel 2D superconductors, predicting lattice parameters and formation energies of high-entropy alloys, and identifying promising metal–organic frameworks for heterogeneous catalysis. , However, in this quickly developing framework, a systematic study of mechanical and tribological properties of solid–solid heterointerfaces has not been addressed yet. Most probably this is due to the inherent difficulties that this kind of system poses and to the fact that the community of references (the tribology, metallurgy, and mechanical manufacturing communities) most of the time relies on classical macroscopic engineering models.…”