Search NASA⌕ Search
Home/Archive/DOE OSTI

DOE OSTI · 3739680

Predicting Band-Gap of Inorganic Materials Using Neuromorphic Graph Learning

Mulet, Ian [University of Tennessee (UT)]·Lim, Seung-Hwan [ORNL] (ORCID:0000000194616866)·Gautam, Ashish [ORNL]·Date, Prasanna [ORNL] (ORCID:000000021664069X)·Kulkarni, Shruti [ORNL] (ORCID:0000000168949851)·Cong, Guojing [ORNL]·Potok, Thomas [ORNL] (ORCID:0000000166873435)·Zhu, Kevin [George Mason University, Virginia]·Parsa, Maryam [George Mason University, Virginia]·Schuman, Catherine D [University of Tennessee, Knoxville (UTK)]
View original source ↗Read source document ↗Export metadata ↓

Abstract

Predicting properties of inorganic materials is a heavily researched topic, with several new prediction approaches emerging as competitors. One such competitor is graph neural networks, which leverage the structure of the graph to aid in the prediction process. In this work, we propose integration of neuromorphic computation into the graph neural network pipeline. We call this approach Neuromorphic Graph Learning (NGL). We utilize the NGL approach to leverage evolutionary algorithms and a novel Spike Pipeline for Raster Analysis (SPIRE) for the prediction of band gap in inorganic materials.

Keep this discovery

BibTeXRIS

Mulet, Ian [University of Tennessee (UT)], Lim, Seung-Hwan [ORNL] (ORCID:0000000194616866), Gautam, Ashish [ORNL], Date, Prasanna [ORNL] (ORCID:000000021664069X), Kulkarni, Shruti [ORNL] (ORCID:0000000168949851), Cong, Guojing [ORNL], Potok, Thomas [ORNL] (ORCID:0000000166873435), Zhu, Kevin [George Mason University, Virginia], Parsa, Maryam [George Mason University, Virginia], Schuman, Catherine D [University of Tennessee, Knoxville (UTK)]. 2026-06-01. Predicting Band-Gap of Inorganic Materials Using Neuromorphic Graph Learning. https://doi.org/10.1109/nice69539.2026.11567459

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related discoveries

DOE OSTI2026

Machine learning approach for vibronically renormalized electronic band structures

Here, we present a machine learning (ML) method for efficient computation of vibrational thermal expectation values of physical properties from first principles. Our approach is based on the nonperturbative frozen phonon formulation in which stochastic Monte Carlo algorithm is employed to sample configurations of nuclei in a supercell at finite temperatures based on a first-principles phonon model. A deep-learning neural network is trained to accurately predict physical properties associated with sampled phonon configurations, thus bypassing the time-consuming ab initio calculations. To incorporate the point-group symmetry of the electronic system into the ML model, group-theoretical methods are used to develop a symmetry-invariant descriptor for phonon configurations in the supercell. We apply our ML approach to compute the temperature dependent electronic energy gap of silicon based on density functional theory (DFT). We show that, with less than a hundred DFT calculations for training the neural network model, an order of magnitude larger number of sampling can be achieved for the computation of the vibrational thermal expectation values. Our work highlights the promising potential of ML techniques for finite temperature first-principles electronic structure methods.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2025

Dimensional Reduction Guides Electronic Structure Evolution in the A n Cu 4–n SnS 4 Semiconductor Series

The search for new functional materials with tunable properties remains a central challenge in chemistry, particularly for applications in energy and electronics. In this work, we present a framework for predictive crystal design in alkali metal chalcogenides that enables controlled dimensional reduction of a parent covalent motif, yielding a broad range of electronic structures, which systematically evolve from one parent to the other. We present 11 new members of the A n Cu 4–n SnS 4 family (A = alkali metal; n = 0–4), which reduce the three-dimensional (3D) covalent network of Cu 4 SnS 4 into various 3D, 2D, 1D, and 0D [Cu 4–n SnS 4 ] n− motifs through the substitution of Cu with alkali metals of various radii. The end members of the family set the range in achievable band gaps at 0.99 eV for fully covalent Cu 4 SnS 4 (n = 0) and 3.38 eV for K 4 SnS 4 (n = 4) with 0D [SnS 4 ] n− tetrahedra. As the dimensionality of [Cu 4–n SnS 4 ] n− systematically reduces within A n Cu 4–n SnS 4 (n = 1–3), a stepwise increase in band gap energy occurs through a gradual decrease in the energy of the valence band maximum and an increase in the conduction band minimum, with an increase in the effective masses of charge carriers. Furthermore, irrespective of the alkali metal, the thermal stability decreases with decreasing [Cu 4–n SnS 4 ] n− dimensionality within the quaternary members. Most importantly, we demonstrate that predictable crystal structure and property evolution for a given composition space is possible by deriving a general formula based on substituting the covalent metals of a parent structure with alkali metals.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Structure–Property Linkage in Alloys Using Graph Neural Network and Explainable Artificial Intelligence

Deep learning tools have recently shown significant potential for accelerating the prediction of microstructure–property linkage in materials. While deep neural networks like convolution neural networks (CNNs) can extract physics information from 3D microstructure images, they often require a large network architecture and substantial training time. In this research, we trained a graph neural network (GNN) using phase field generated microstructures of Ni-Al alloys to predict the evolution of mechanical properties. We found that a single GNN is capable of accurately predicting the strengthening of Ni-Al alloys with microstructures of varying sizes and dimensions, which cannot otherwise be done with a CNN. Additionally, GNN requires significantly less GPU utilization than CNN and offers more interpretable explanation of predictions using saliency analysis as features are manually defined in the graph. We also utilize explainable artificial intelligence tool Bayesian Inference to determine the coefficients in the power law equation that governs coarsening of precipitates. Overall, our work demonstrates the ability of the GNN to accurately and efficiently extract relevant information from material microstructures without having restrictions on microstructure size or dimension and offers an interpretable explanation.

Chemistry↗