Polymorphism Crystal Structure Prediction with Adaptive Space Group Diversity Control

2025-06-12

AI Summary

To address the weak global search capability and premature convergence in inorganic polymorph structure prediction, this paper proposes ParetoCSP2, a multi-objective genetic algorithm. Its core innovation is the first-of-its-kind neural network interatomic potential (NNIP)-guided space-group-adaptive diversity control mechanism, integrated with space-group–aware encoding, optimized population initialization, and iterative relaxation strategies—collectively enhancing configurational space exploration efficiency and convergence stability. On a benchmark set of dual-polymorphic systems with identical atomic compositions, ParetoCSP2 achieves near-perfect accuracy (~100%) in space-group and structural similarity identification. Compared to state-of-the-art baseline algorithms, it improves energy-ranking accuracy by 44.8%–87.04% and polymorph identification rate by 2.46×–8.62×. This work establishes a new, efficient, and robust paradigm for inorganic polymorph prediction.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Mixed Discrete/Continuous Search
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Crystalline materials can form different structural arrangements (i.e. polymorphs) with the same chemical composition, exhibiting distinct physical properties depending on how they were synthesized or the conditions under which they operate. For example, carbon can exist as graphite (soft, conductive) or diamond (hard, insulating). Computational methods that can predict these polymorphs are vital in materials science, which help understand stability relationships, guide synthesis efforts, and discover new materials with desired properties without extensive trial-and-error experimentation. However, effective crystal structure prediction (CSP) algorithms for inorganic polymorph structures remain limited. We propose ParetoCSP2, a multi-objective genetic algorithm for polymorphism CSP that incorporates an adaptive space group diversity control technique, preventing over-representation of any single space group in the population guided by a neural network interatomic potential. Using an improved population initialization method and performing iterative structure relaxation, ParetoCSP2 not only alleviates premature convergence but also achieves improved convergence speed. Our results show that ParetoCSP2 achieves excellent performance in polymorphism prediction, including a nearly perfect space group and structural similarity accuracy for formulas with two polymorphs but with the same number of unit cell atoms. Evaluated on a benchmark dataset, it outperforms baseline algorithms by factors of 2.46-8.62 for these accuracies and improves by 44.8%-87.04% across key performance metrics for regular CSP. Our source code is freely available at https://github.com/usccolumbia/ParetoCSP2.

Problem

• Predict polymorph crystal structures efficiently
• Control space group diversity adaptively
• Improve convergence speed and accuracy

Innovation

• Adaptive space group diversity control technique
• Neural network interatomic potential guidance
• Improved population initialization method