Exploration of Lithium-Ion Conductors Based on Local Coordination Environments Using Crystallographic Site...

Area of Science:

• Materials Science
• Electrochemistry
• Computational Chemistry

Background:

• Developing high-performance solid-state electrolytes for lithium-ion batteries is crucial for next-generation energy storage.
• Many potential lithium-containing compounds for solid-state electrolytes remain undiscovered, hindering battery development.

Purpose of the Study:

• To accelerate the discovery of novel lithium-ion conductors using a semisupervised learning approach.
• To identify promising solid-state electrolyte candidates by focusing on local coordination environments.

Main Methods:

• Utilized semisupervised learning with four structure-representation descriptors for local coordination.
• Applied agglomerative clustering to a dataset of 3,835 lithium-containing structures.
• Evaluated shortlisted compounds using molecular dynamics simulations and experimental conductivity measurements.

Main Results:

• Identified 147 potential lithium-ion conductor candidates from cluster screening.
• Li3LaP2S8 was initially identified but showed low conductivity.
• Optimized lithium content in Li3LaP2S8 to yield Li3.1La0.9Sr0.1P2S8 with a conductivity of 2.1 × 10^-6 S cm^-1 at 298 K.

Conclusions:

• Demonstrated the efficacy of semisupervised learning in discovering advanced materials for solid-state electrolytes.
• Reported Li3LaP2S8 as a novel solid-state electrolyte candidate, with an optimized composition showing improved conductivity.
• Provided a valuable methodology for accelerating the development of next-generation solid-state battery technologies.