Progress and Challenges of Reverse Design of AI-enabled Functional Materials
DOI:
https://doi.org/10.54097/s7zdxg59Keywords:
Functional Materials, Inverse Design, Artificial Intelligence, Generative Models.Abstract
Against the backdrop of global energy structure transformation and high-end manufacturing industry upgrading, the development of new functional materials has become the key to scientific and technological competition. This paper systematically reviews how artificial intelligence (AI) drives the profound change of the functional material research and development paradigm. Through in-depth analysis of the atomic component evolution of AI in energy catalysis, the geometric topology construction in the field of optoelectronic and electromagnetic, and the multi-scale optimization in structural mechanical materials, this paper shows the great potential of deep coupling between generative models and physical laws. At the same time, this paper also faces technical bottlenecks, such as the lack of experimental data, the poor synthesis of design structure, and the lack of transparency of the model. The research conclusion points out that with the intervention of cutting-edge technologies such as physical information assistance, automatic laboratory and large language models, the material research and development will move towards the automatic closed-loop direction of deep integration of data, experiment and theory. This review not only reveals the theoretical value of AI in complex performance optimization but also has important academic and practical significance in breaking the bottleneck of the material development cycle and accelerating the large-scale customization of high-performance materials.
Downloads
References
[1] Sanchez-Lengeling B, Aspuru-Guzik A. Inverse molecular design using machine learning: Generative models for matter engineering. Science, 2018, 361(6400): 360-365.
[2] Kim C, Chandrasekaran A, Huan T D, et al. Polymer genome: A data-powered polymer informatics platform for property predictions. The Journal of Physical Chemistry C, 2018, 122(29): 17575-17585.
[3] Gomez-Bombarelli R, Wei J N, Duvenaud D, et al. Automatic chemical design using a data-driven continuous representation of molecules. ACS Central Science, 2018, 4(2): 268-276.
[4] National Academies of Sciences, Engineering, and Medicine. The Materials Genome Initiative: 10 years on. Washington, D.C.: National Academies Press, 2023.
[5] Zheng J, Wang S, Deng S, et al. Accelerating the screening of modified MA₂Z₄ catalysts for hydrogen evolution reaction by deep learning-based local geometric analysis. Energy & Environmental Materials, 2024, 7(6): e12743.
[6] Chen X, et al. Accelerating computational materials discovery with artificial intelligence and cloud high-performance computing: From large-scale screening to experimental validation. arXiv preprint, 2024.
[7] Alhashmi A, Kanoun M B, Goumri-Said S. Machine learning for halide perovskite materials ABX₃ (B = Pb, X = I, Br, Cl): Assessment of structural properties and band gap engineering for solar energy. Materials, 2023, 16(7): 2657.
[8] So S, Rho J. Designing nanophotonic structures using conditional deep convolutional generative adversarial networks. Nanophotonics, 2019, 8(7): 1255-1261.
[9] Qi X, Sun Z, Li Y, et al. Machine learning-guided design of single-phase high-entropy alloys with superior mechanical properties. npj Computational Materials, 2023, 9: 191.
[10] Ma Z, Liu H, Liu Y, et al. Design of new energy-absorbing lattice cell configuration by dynamic topology optimization. Metals, 2024, 14(12): 1348.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Highlights in Science, Engineering and Technology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
