Data-driven deep generative design of stable spintronic materials

Siriwardane, Edirisuriya M. Dilanga; Zhao, Yong; Hu, Jianjun

doi:10.1039/d3ce00765k

Cited by 2 publications

(3 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, a compound’s stability is inversely proportional to its energy above the convex hull; the less stable it is, the higher its E hull . In many data-driven approaches to generating materials, E hull has become a standard measure to assess the thermodynamic stability of the generated materials. ,− ,− …”

Section: Resultsmentioning

confidence: 99%

“…Additionally, graphbased diffusion models have shown great promise in tasks like designing inorganic crystals, 59 generating 1D and 2D materials, 60,61 producing potential superconducting materials, 62 and curating stable spintronic materials. 63 In this work, GNNs were utilized in conjunction with highthroughput DFT calculations to methodically develop an accelerated data-driven approach for 2D magnet discovery. This was achieved through a four-step process.…”

Section: ■ Introductionmentioning

confidence: 99%

“…These models eliminate the need to wade through countless potential stoichiometries and symmetries, adopting a data-driven approach instead. Additionally, graph-based diffusion models have shown great promise in tasks like designing inorganic crystals, generating 1D and 2D materials, , producing potential superconducting materials, and curating stable spintronic materials …”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Accelerated Data-Driven Discovery and Screening of Two-Dimensional Magnets Using Graph Neural Networks

Elrashidy,

Della-Giustina,

Yan

2024

J. Phys. Chem. C

View full text Add to dashboard Cite

In this study, we employ graph neural networks (GNNs) to accelerate the discovery of novel 2D magnetic materials which have transformative potential in spintronic applications. Using data from the Materials Project database and the Computational 2D materials database, we train three GNN architectures on a dataset of 1190 magnetic monolayers with energy above the convex hull (E hull ) less than 0.3 eV/atom. Our Crystal Diffusion Variational Autoencoder generates 11,100 candidate crystals. Subsequent training on two Atomistic Line GNNs achieves 93% accuracy in predicting magnetic monolayers and a mean average error of 0.039 eV/atom for E hull predictions. After narrowing down candidates based on magnetic likelihood and predicted energy, constraining the atom count in the monolayers to five or fewer, and performing dimensionality checks, we identified 190 candidates. These are validated using density-functional theory to confirm their magnetic and energetic favorability, resulting in 167 magnetic monolayers with E hull < 0.3 eV/atom and a total magnetization of ≥0.5 μ B . Our methodology offers a way to accelerate the exploration and prediction of potential 2D magnetic materials, contributing to the ongoing computational and experimental efforts aimed at the discovery of new 2D magnets.

show abstract