AQVolt26: High-Temperature r$^2$SCAN Halide Dataset for Universal ML Potentials and Solid-State Batteries

2026-04-02

AI Summary

This work addresses the limited accuracy of general-purpose machine learning potentials in predicting energies and forces for halide solid-state electrolytes under high-temperature and highly distorted conditions, which undermines their reliability in ion transport simulations. To overcome this, the authors construct the AQVolt26 dataset comprising 322,656 r²SCAN single-point calculations derived from high-temperature molecular dynamics sampling, augmented with relaxed structures from the Materials Project. They introduce a targeted co-training strategy leveraging high-temperature data and systematically reveal, for the first time, the performance blind spots of generic models in high-temperature distorted regimes. Their results demonstrate that domain-specific sampling is essential for dynamic screening of halide electrolytes. The trained model substantially improves force prediction accuracy at elevated temperatures, whereas incorporating only near-equilibrium data, while enhancing local performance, compromises robustness under extreme strain.

Abstract

The demand for safe, high-energy-density batteries has spotlighted halide solid-state electrolytes, which offer the potential for enhanced ionic mobility, electrochemical stability, and interfacial deformability. Accelerating their discovery requires extensive molecular dynamics, which has been increasingly enabled by universal machine learning interatomic potentials trained on foundational datasets. However, the dynamic softness of halides poses a stringent test of whether general-purpose models can reliably replace first-principles calculations under the highly distorted, elevated-temperature regimes necessary to probe ion transport. Here, we present AQVolt26, a dataset of 322,656 r$^2$SCAN single-point calculations for lithium halides, generated via high-temperature configurational sampling across $\sim$5K structures. We demonstrate that foundational datasets provide a strong baseline for stable halide chemistries and transfer local forces well, however absolute energy predictions degrade in distorted higher-temperature regimes. Co-training with AQVolt26 resolves this blind spot. Furthermore, incorporating Materials Project relaxation data improves near-equilibrium performance but degrades extreme-strain robustness without enhancing high-temperature force accuracy. These results demonstrate that domain-specific configurational sampling is essential for the reliable dynamic screening of halide electrolytes. Furthermore, our findings suggest that while foundational models provide a robust base, they are most effective for dynamically soft solid-state chemistries when augmented with targeted, high-temperature data. Finally, we show that near-equilibrium relaxation data serves as a task-specific complement rather than a universally beneficial addition.

Problem

• halide solid-state electrolytes
• machine learning potentials
• high-temperature dynamics
• ionic transport
• configurational sampling

Innovation

• machine learning potentials
• halide solid-state electrolytes
• high-temperature sampling
• r2SCAN dataset
• ionic transport