Prediction of phase transition in layered oxide cathode materials

Highly stable P2-type layered oxides are a major driving force in advancing Na-ion batteries. The identification of compositions that do not undergo phase transitions under electrochemical operating conditions is crucial for the strategic design of cathode active materials. To achieve this, we developed a machine learning (ML) model, trained on experimentally observed phase transitions, to predict the structural evolution of P2-type layered oxides during cycling. To facilitate future materials design, we corroborated the ML-identified trends using a combination of electrostatic and density functional theory (DFT) calculations. The combination of these three methods enabled the elucidation of microscopic driving forces governing the phase transition of Na-based layered oxides under operating conditions. We discovered that the concentrations of monovalent and divalent elements, changes in the transition metal (TM) ionic potential, and the Na content at the initial and final states are the key factors driving desodiation-induced phase transitions. Taking into account its moderate accuracy (∼69 %), the model is not intended as a standalone predictive framework, but rather as a tool to identify key trends and to provide an initial assessment of promising compositions for subsequent experimental and theoretical studies.

Authors

Institutions

Publication Details

Journal
Journal of Power Sources
Published
2026-11-18
DOI
https://doi.org/10.1016/j.jpowsour.2026.241896
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Prediction of phase transition in layered oxide cathode materials

Payam Kaghazchi, Jyh‐Chiang Jiang, Liang‐Ting Wu, Dijana Milosavljević
2 links
Journal of Power Sources
Machine Learning in Materials Science
article

Prediction of phase transition in layered oxide cathode materials

Payam Kaghazchi, Jyh‐Chiang Jiang, Liang‐Ting Wu, Dijana Milosavljević
article en

Abstract

Highly stable P2-type layered oxides are a major driving force in advancing Na-ion batteries. The identification of compositions that do not undergo phase transitions under electrochemical operating conditions is crucial for the strategic design of cathode active materials. To achieve this, we developed a machine learning (ML) model, trained on experimentally observed phase transitions, to predict the structural evolution of P2-type layered oxides during cycling. To facilitate future materials design, we corroborated the ML-identified trends using a combination of electrostatic and density functional theory (DFT) calculations. The combination of these three methods enabled the elucidation of microscopic driving forces governing the phase transition of Na-based layered oxides under operating conditions. We discovered that the concentrations of monovalent and divalent elements, changes in the transition metal (TM) ionic potential, and the Na content at the initial and final states are the key factors driving desodiation-induced phase transitions. Taking into account its moderate accuracy (∼69 %), the model is not intended as a standalone predictive framework, but rather as a tool to identify key trends and to provide an initial assessment of promising compositions for subsequent experimental and theoretical studies.

Journal of Power SourcesVol. 700
National Taiwan University of Science and Technology (TW), Forschungszentrum Jülich (DE), University of Twente (NL)
Bundesministerium für Bildung und Forschung, Forschungszentrum Jülich, Ministry of Education, India, National Science and Technology Council
Affordable and clean energy
Machine Learning in Materials Science
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon