An official website of the United States government
The rapid development of universal machine learning potentials (uMLPs) has enabled efficient, accurate predictions of diverse material properties across broad chemical spaces. While their capability for modeling phonon properties is emerging, systematic benchmarking across chemically diverse systems remains limited. We evaluate six recent uMLPs—EquiformerV2, MatterSim, MACE, and CHGNet—on 2429 crystalline materials from the Open Quantum Materials Database. Models were used to compute atomic forces in displaced supercells, derive interatomic force constants (IFCs), and predict phonon properties including lattice thermal conductivity (LTC), compared with density functional theory and experimental data. The EquiformerV2 pretrained model trained on the OMat24 dataset exhibits strong performance in predicting atomic forces and third‐order IFCs, while its fine‐tuned counterpart consistently outperforms other models in predicting second‐order IFCs, LTC, and other phonon properties. Although MACE and CHGNet demonstrated comparable force prediction accuracy to EquiformerV2, notable discrepancies in IFC fitting led to poor LTC predictions. Conversely, MatterSim, despite lower force accuracy, achieved intermediate IFC predictions, suggesting error cancellation and complex relationships between force accuracy and phonon predictions. This benchmark guides the evaluation and selection of uMLPs for high‐throughput screening of materials with targeted thermal transport properties.
more »
« less
Machine Learning a Universal Harmonic Interatomic Potential for Predicting Phonons in Crystalline Solids
Phonons, as quantized vibrational modes in crystalline materials, play a crucial role in determining a wide range of physical properties, such as thermal and electrical conductivity, making their study a cornerstone in materials science. In this study, we present a simple yet effective strategy for deep learning harmonic phonons in crystalline solids by leveraging existing phonon databases and state-of-the-art machine learning techniques. The key of our method lies in transforming existing phonon datasets, primarily represented in interatomic force constants, into a force-displacement representation suitable for training machine learning universal interatomic potentials. By applying our approach to one of the largest phonon databases publicly available, we demonstrate that the resultant machine learning universal harmonic interatomic potential not only accurately predicts full harmonic phonon spectra but also calculates key thermodynamic properties with remarkable precision. Furthermore, the restriction to a harmonic potential energy surface in our model provides a way of assessing uncertainty in machine learning predictions of vibrational properties, essential for guiding further improvements and applications in materials science.
more »
« less
- Award ID(s):
- 2317008
- PAR ID:
- 10534580
- Editor(s):
- Grein, Christoph
- Publisher / Repository:
- American Institute of Physics (AIP)
- Date Published:
- Journal Name:
- Applied physics letters
- Volume:
- 124
- Issue:
- 10
- ISSN:
- 0003-6951
- Page Range / eLocation ID:
- 102202-1-6
- Subject(s) / Keyword(s):
- Vibrational properties, Potential energy surfaces, Electrical conductivity, Phonons, Thermodynamic properties, Deep learning, Machine learning, Crystalline solids, Interatomic force constants, Interatomic potentials
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The discovery of advanced thermal materials with exceptional phonon properties drives technological advancements, impacting innovations from electronics to superconductors. Understanding the intricate relationship between composition, structure, and phonon thermal transport properties is crucial for speeding up such discovery. Exploring innovative materials involves navigating vast design spaces and considering chemical and structural factors on multiple scales and modalities. Artificial intelligence (AI) is transforming science and engineering and poised to transform discovery and innovation. This era offers a unique opportunity to establish a new paradigm for the discovery of advanced materials by leveraging databases, simulations, and accumulated knowledge, venturing into experimental frontiers, and incorporating cutting-edge AI technologies. In this perspective, first, the general approach of density functional theory (DFT) coupled with phonon Boltzmann transport equation (BTE) for predicting comprehensive phonon properties will be reviewed. Then, to circumvent the extremely computationally demanding DFT + BTE approach, some early studies and progress of deploying AI/machine learning (ML) models to phonon thermal transport in the context of structure–phonon property relationship prediction will be presented, and their limitations will also be discussed. Finally, a summary of current challenges and an outlook of future trends will be given. Further development of incorporating AI/ML algorithms for phonon thermal transport could range from phonon database construction to universal machine learning potential training, to inverse design of materials with target phonon properties and to extend ML models beyond traditional phonons.more » « less
-
Phonons play a crucial role in many properties of solid-state systems, and it is expected that topological phonons may lead to rich and unconventional physics. On the basis of the existing phonon materials databases, we have compiled a catalog of topological phonon bands for more than 10,000 three-dimensional crystalline materials. Using topological quantum chemistry, we calculated the band representations, compatibility relations, and band topologies of each isolated set of phonon bands for the materials in the phonon databases. Additionally, we calculated the real-space invariants for all the topologically trivial bands and classified them as atomic or obstructed atomic bands. We have selected more than 1000 “ideal” nontrivial phonon materials to motivate future experiments. The datasets were used to build the Topological Phonon Database.more » « less
-
Abstract Polar dielectrics are key materials of interest for infrared (IR) nanophotonic applications due to their ability to host phonon‐polaritons that allow for low‐loss, subdiffractional control of light. The properties of phonon‐polaritons are limited by the characteristics of optical phonons, which are nominally fixed for most “bulk” materials. Superlattices composed of alternating atomically thin materials offer control over crystal anisotropy through changes in composition, optical phonon confinement, and the emergence of new modes. In particular, the modified optical phonons in superlattices offer the potential for so‐called crystalline hybrids whose IR properties cannot be described as a simple mixture of the bulk constituents. To date, however, studies have primarily focused on identifying the presence of new or modified optical phonon modes rather than assessing their impact on the IR response. This study focuses on assessing the impact of confined optical phonon modes on the hybrid IR dielectric function in superlattices of GaSb and AlSb. Using a combination of first principles theory, Raman, FTIR, and spectroscopic ellipsometry, the hybrid dielectric function is found to track the confinement of optical phonons, leading to optical phonon spectral shifts of up to 20 cm−1. These results provide an alternative pathway toward designer IR optical materials.more » « less
-
Abstract Phonons are quantum elastic excitations of crystalline solids. Classically, they correspond to the collective vibrations of atoms in ordered periodic structures. They determine the thermodynamic properties of solids and their stability in the case of structural transformations. Here we review for the first time the existing examples of the phonon analysis of adsorption-induced transformations occurring in microporous crystalline materials. We discuss the role of phonons in determining the mechanism of the deformations. We point out that phonon-based methodology may be used as a predictive tool in characterization of flexible microporous structures; therefore, relevant numerical tools must be developed.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- The DOI is not currently available.
