[论文解读] Accelerating computational materials discovery with artificial intelligence and cloud high-performance computing: from large-scale screening to experimental validation

Chi Chen, Dan Thien Nguyen+9 more|arXiv (Cornell University)|Jan 8, 2024

Machine Learning in Materials Science被引用 13

一句话总结

本文将 AI 模型与云端 HPC 相结合，筛选超过 32 million 个材料候选，预测约 half a million 潜在稳定材料，并对新的 Li/Na 固态电解质进行实验证明。

ABSTRACT

High-throughput computational materials discovery has promised significant acceleration of the design and discovery of new materials for many years. Despite a surge in interest and activity, the constraints imposed by large-scale computational resources present a significant bottleneck. Furthermore, examples of large-scale computational discovery carried through experimental validation remain scarce, especially for materials with product applicability. Here we demonstrate how this vision became reality by first combining state-of-the-art artificial intelligence (AI) models and traditional physics-based models on cloud high-performance computing (HPC) resources to quickly navigate through more than 32 million candidates and predict around half a million potentially stable materials. By focusing on solid-state electrolytes for battery applications, our discovery pipeline further identified 18 promising candidates with new compositions and rediscovered a decade's worth of collective knowledge in the field as a byproduct. By employing around one thousand virtual machines (VMs) in the cloud, this process took less than 80 hours. We then synthesized and experimentally characterized the structures and conductivities of our top candidates, the Na$_x$Li$_{3-x}$YCl$_6$ ($0 < x < 3$) series, demonstrating the potential of these compounds to serve as solid electrolytes. Additional candidate materials that are currently under experimental investigation could offer more examples of the computational discovery of new phases of Li- and Na-conducting solid electrolytes. We believe that this unprecedented approach of synergistically integrating AI models and cloud HPC not only accelerates materials discovery but also showcases the potency of AI-guided experimentation in unlocking transformative scientific breakthroughs with real-world applications.

研究动机与目标

• Demonstrate a scalable AI+cloud HPC workflow for large-scale materials discovery.
• Identify stable materials predicting properties suitable for solid-state electrolytes.
• Experimentally validate top computationally discovered electrolyte candidates.
• Provide insights into ESW and Li/Na diffusion trends to guide future design.

提出的方法

• 构建通过离子替代在 ICSD 派生原型中的广泛初始候选集，覆盖 54 元素，得到 32,598,079 个候选。
• 使用 ML 力场来放松结构并评估热力学稳定性，缩小到 589,609 个稳定材料。
• 应用 AI-属性过滤器（带隙、ESW、氧化还原势），再结合 ML 力场的 DFT/MD 评估锂扩散性和稳定性，产生 147 个有前景的候选。
• 进一步过滤以适应实际电池集成（元素丰度、密度、体模/剪切模量），选出 23 个最终候选。
• 合成并表征顶级候选物（Li3YCl6 系列与 Na2LiYCl6 系列）以验证结构和离子导电性。
• 描述云端 HPC 基础设施（Azure Quantum Elements）用于可扩展、容器化的 ML/DFT 工作流和数据管理。

实验结果

研究问题

• RQ1 AI+云 HPC 工作流是否能够高效筛选超大化学空间以识别稳定、可合成的材料？
• RQ2 在广义 ESW 和 Li 导电性标准下，哪些材料族群有望成为有前景的固态电解质？
• RQ3 计算预测的候选材料是否呈现实验上可测的离子导电性，适用于固态电池？
• RQ4 在卤化物基电解质中，哪些结构特征与组成趋势与高 Li/Na 离子移动性相关？

主要发现

• 筛选 3260 万个候选，识别出 589,609 个预测稳定的材料。
• 在稳定性过滤池中，147 个候选物通过了所有 ESW、扩散性和实用性属性标准。
• 四种稀土卤化物组合（Li3YCl6、Li5YCl8、Li7Y2Cl13、Na2LiYCl6）在实验中得到验证，具有有意义的 Li/Na 导电性和合适的带隙。
• Na2LiYCl6 展示出显著的 Li-Na 双导电性并降低的活化能（0.46–0.66 eV），相较 Na3YCl6（0.82 eV）。
• ESW 和扩散趋势与先前研究一致，强调氯化物、氟化物、溴化物作为有利的卤化物电解质。