A new universal force-field for the Li₂S-P₂S₅ system.

Area of Science:

• Materials Science
• Computational Chemistry
• Electrochemistry

Background:

• Lithium thiophosphate electrolytes are key for all-solid-state batteries.
• Ab initio molecular dynamics (AIMD) simulations reveal ion conduction but are computationally expensive for complex materials.

Purpose of the Study:

• To develop a computationally efficient interatomic potential for classical molecular dynamics (CMD) simulations of lithium solid-state electrolytes.
• To enable the simulation of complex materials like those with grain boundaries and multiphase glass-ceramics.

Main Methods:

• Generated training datasets from various sulfide electrolytes (crystals and glass) using AIMD simulations.
• Optimized Class II and Stillinger-Weber potential parameters by minimizing differences in forces, stresses, and energies between CMD and AIMD.
• Validated the developed potential by comparing simulated Li+ dynamics and structures with AIMD results.

Main Results:

• The optimized interatomic potential accurately reproduces Li+ dynamics and structures in crystalline and glassy lithium sulfide electrolytes.
• CMD simulations with the new potential showed ionic conductivity differences between Li7P3S11 crystal and glass, matching AIMD predictions.
• The developed force-field enables CMD simulations of complex materials, including amorphous-crystalline interfaces and multiphase glass-ceramics.

Conclusions:

• The developed interatomic potential significantly reduces the computational cost for simulating lithium solid-state electrolytes.
• This advancement allows for the investigation of ion conduction mechanisms in more complex and realistic battery material systems.
• The new potential is a valuable tool for designing and optimizing materials for next-generation all-solid-state batteries.