Machine Learning Reviews Composition Dependent Thermal Stability in Halide Perovskites

2025-04-05

AI Summary

Halide perovskite thermal stability is highly composition-dependent, yet its degradation mechanisms remain complex and poorly predictable. Method: We employ high-throughput in situ environmental photoluminescence (PL) characterization coupled with machine learning—specifically, a dual-mode XGBoost framework integrating composition-agnostic and composition-aware modeling. The methodology incorporates t-SNE/UMAP dimensionality reduction, feature importance analysis, and correlation heatmap visualization. Contribution/Results: We discover, for the first time, a significant negative correlation between Cs content and thermal stability—a counterintuitive trend. Our standardized, transferable ML pipeline achieves 85% PL stability prediction accuracy in the composition-agnostic mode and 75% in the composition-aware mode; Cs-related features dominate with 99% importance. This work establishes a generalizable analytical paradigm applicable to arbitrary perovskite systems and substantially accelerates the high-throughput screening of photovoltaic-grade stable perovskites.

Technology Category

• Machine Learning: Hardware-aware ML
• Humans and AI: Human-in-the-loop Machine Learning
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Halide perovskites exhibit unpredictable properties in response to environmental stressors, due to several composition-dependent degradation mechanisms. In this work, we apply data visualization and machine learning (ML) techniques to reveal unexpected correlations between composition, temperature, and material properties while using high throughput, in situ environmental photoluminescence (PL) experiments. Correlation heatmaps show the strong influence of Cs content on film degradation, and dimensionality reduction visualization methods uncover clear composition-based data clusters. An extreme gradient boosting algorithm (XGBoost) effectively forecasts PL features for ten perovskite films with both composition-agnostic (>85% accuracy) and composition-dependent (>75% accuracy) model approaches, while elucidating the relative feature importance of composition (up to 99%). This model validates a previously unseen anti-correlation between Cs content and material thermal stability. Our ML-based framework can be expanded to any perovskite family, significantly reducing the analysis time currently employed to identify stable options for photovoltaics.

Problem

• Reveal correlations between composition and thermal stability
• Predict material properties using machine learning models
• Reduce analysis time for stable photovoltaic materials

Innovation

• Machine learning analyzes perovskite thermal stability
• XGBoost predicts photoluminescence with high accuracy
• Data visualization reveals composition degradation correlations