Inverse design of solid-state materials with desired properties represents a formidable challenge in materials science. Although recent generative models have demonstrated potential, their adoption has been hindered by limitations such as inefficiency, architectural constraints and restricted open-source availability. The representation of crystal structures using the SLICES (Simplified Line-Input Crystal-Encoding System) notation as a string of characters enables the use of state-of-the-art natural language processing models, such as Transformers, for crystal design. Drawing inspiration from the success of GPT models in generating coherent text, we trained a generative Transformer on the next-token prediction task to generate solid-state materials with targeted properties. We demonstrate MatterGPT's capability to generate de novo crystal structures with targeted single properties, including both lattice-insensitive (formation energy) and lattice-sensitive (band gap) properties. Furthermore, we extend MatterGPT to simultaneously target multiple properties, addressing the complex challenge of multi-objective inverse design of crystals. Our approach showcases high validity, uniqueness, and novelty in generated structures, as well as the ability to generate materials with properties beyond the training data distribution. This work represents a significant step forward in computational materials discovery, offering a powerful and open tool for designing materials with tailored properties for various applications in energy, electronics, and beyond.
Microstructure Generation via Generative Adversarial Network for Heterogeneous, Topologically Complex 3D Materials
U.S. DOE National Energy Technology Laboratory(美国能源部国家能源技术实验室);Carnegie Mellon University(卡内基梅隆大学);Lawrence Livermore National Laboratory(劳伦斯利弗莫尔国家实验室);Leidos Research Support Team(莱多斯研究支持团队)
Order and randomness in dopant distributions: exploring the thermodynamics of solid solutions from atomically resolved imaging
作为掺杂剂浓度的函数的结构-性质关系的探索通常基于固溶体的平均场论。然而,这些理论在半导体中很好地工作,在具有强相关性的材料中往往会失败,无论是在电子行为还是化学分离方面。在这些情况下,原子排列的细节通常不会被探索和分析。材料的生成物理和化学的知识可以解决这个问题,因为可以生成作为原子级结构的随机表示的缺陷配置库,或者可以导出介观热力学模型的参数。为了获得这些信息,以改进预测,我们使用原子分辨的显微图像,可视化系统内复杂的结构相关性,并将其转化为结构形成的统计力学模型的数据。鉴于材料加工历史的微观方面存在显著的不确定性,以及可用图像的数量有限,我们将模型优化技术与统计假设检验的原则相结合。我们证明了从一系列的原子分辨扫描透射电子显微镜图像的Mo $_x$ Re $_{1-x}$ S $_2$在不同比例的Mo/Re化学计量的数据上的方法,为此,我们提出了一个有效的相互作用模型,然后用于生成原子配置,并在一系列的浓度和形成温度进行可测试的预测。
Boron-assisted synthesis of compositionally complex amorphous oxides via short-range-order-constrained generative design
Key Laboratory of Material Simulation Methods and Software of Ministry of Education, College of Physics, Jilin University(吉林大学物理学院教育部材料模拟方法与软件重点实验室);Aiiso Yufeng Li Family Department of Chemical and Nano Engineering, University of California, San Diego(加州大学圣地亚哥分校爱思唯尔俞峰李家族化学与纳米工程系);College of Chemistry and Molecular Engineering, Peking University(北京大学化学与分子工程学院);Institute of Molecular Engineering Plus, College of Chemistry, Fuzhou University(福州大学化学学院分子工程+研究院);Institute of Modern Physics, Fudan University(复旦大学现代物质结构研究所);Program in Materials Science and Engineering, University of California, San Diego(加州大学圣地亚哥分校材料科学与工程项目);Center for Renewable Energy and Storage Technologies, Physical Science and Engineering Division, King Abdullah University of Science and Technology(阿卜杜拉国王科技大学物理科学与工程学院可再生能源与储能技术中心);Department of Computer Science and Engineering, University of California, San Diego(加州大学圣地亚哥分校计算机科学与工程系)
A Differentiable DFT-Based Framework for Inverse Materials Design
Institute for Solid State Physics, University of Tokyo(东京大学固体物理研究所);Department of Precision Engineering, Graduate School of Engineering, The University of Osaka(大阪大学大学院工学系研究科精密工学科);Materials DX Research Center, National Institute of Advanced Industrial Science and Technology(产业技术综合研究所材料DX研究中心);Calmarion Inc.(Calmarion公司)
Combining physical models with dynamically acquired experimental information for the optimization of multicomponent NASICON fast ionic conductors in a self-driving laboratory
University of California Berkeley(加州大学伯克利分校);Lawrence Berkeley National Laboratory(劳伦斯伯克利国家实验室);Argonne National Laboratory(阿贡国家实验室);Bakar Institute of Digital Materials for the Planet, UC Berkeley(加州大学伯克利分校巴克拉星球数字材料研究所);Vanderbilt University(范德堡大学)
Space Group Informed Transformer for Crystalline Materials Generation
Beijing National Laboratory for Condensed Matter Physics(北京凝聚态物理国家实验室);School of Physical Sciences, University of Chinese Academy of Sciences, Beijing 100190, China(中国科学院大学物理科学学院,北京100190,中国);Key Laboratory of Material Simulation Methods and Software of Ministry of Education, College of Physics, Jilin University, Changchun 130012, China(教育部材料模拟方法与软件重点实验室,吉林大学物理学院,长春130012,中国);State Key Laboratory of Superhard Materials, College of Physics, Jilin University, Changchun 130012, P. R. China(超硬材料国家重点实验室,吉林大学物理学院,长春130012,中华人民共和国);Songshan Lake Materials Laboratory, Dongguan, Guangdong 523808, China(松山湖材料实验室,东莞,广东523808,中国)