[Paper Review] 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 Science13 citations

TL;DR

The paper combines AI models with cloud HPC to screen over 32 million material candidates, predict about half a million potentially stable materials, and experimentally validate new Li/Na solid electrolytes.

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.

Motivation & Objective

• 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.

Proposed method

• Construct a broad initial candidate set by ionic substitution across 54 elements into ICSD-derived prototypes, yielding 32,598,079 candidates.
• Use ML potentials to relax structures and assess thermodynamic stability, narrowing to 589,609 stable materials.
• Apply AI-property filters (band gap, ESW, redox potentials) followed by DFT/MD with ML potentials to evaluate Li diffusivity and stability, producing 147 promising candidates.
• Further filter for practical battery integration (elemental abundance, density, bulk/shear moduli) to select 23 final candidates.
• Synthesize and characterize top candidates (Li3YCl6 family and Na2LiYCl6 series) to validate structure and ionic conductivity.
• Describe cloud-HPC infrastructure (Azure Quantum Elements) for scalable, containerized ML/DFT workflows and data management.

Experimental results

Research questions

• RQ1 Can an AI+cloud HPC workflow efficiently screen an ultra-large chemical space to identify stable, synthesizable materials?
• RQ2 What material families emerge as promising solid-state electrolytes under a broad ESW and Li conductivity criteria?
• RQ3 Do computationally predicted candidates exhibit experimentally measurable ionic conductivities suitable for solid-state batteries?
• RQ4 What structural features and composition trends correlate with high Li/Na ionic mobility in halide-based electrolytes?

Key findings

• Screened 32.6 million candidates to identify 589,609 predicted to be stable.
• From stability-filtered pool, 147 candidates passed all ESW, diffusion, and practical-property criteria.
• Four rare-earth halide compositions (Li3YCl6, Li5YCl8, Li7Y2Cl13, Na2LiYCl6) validated experimentally with meaningful Li/Na conductivity and suitable band gaps.
• Na2LiYCl6 demonstrated notable Li-Na dual conductivity and reduced activation energy (0.46–0.66 eV) compared to Na3YCl6 (0.82 eV).
• ESW and diffusivity trends align with prior studies, highlighting chlorides, fluorides, bromides as favorable halide electrolytes.