XtalOpt version 14: Variable-composition crystal structure search for functional materials through Pareto optimization

Version 14 of XtalOpt, an evolutionary multi-objective global optimization algorithm for crystal structure prediction, is now available for download from its official website https://xtalopt.github.io, and the Computer Physics Communications Library. The new version of the code is designed to perform a ground state search for crystal structures with variable compositions by integrating a suite of ab initio methods alongside classical and machine-learning potentials for structural relaxation. The multi-objective search framework has been enhanced through the introduction of Pareto optimization, enabling efficient discovery of functional materials. Herein, we describe the newly implemented methodologies, provide detailed instructions for their use, and present an overview of additional improvements included in the latest version of the code. NEW VERSION PROGRAM SUMMARY Program Title: XtalOpt CPC Library link to program files: https://doi.org/10.17632/jt5pvnnm39.5 Developer's repository link: https://github.com/xtalopt/XtalOpt Code Ocean capsule: (to be added by Technical Editor) Licensing provisions: 3-clause/BSD. Programming language: C++. Journal reference of previous version: Computer Physics Communications 304 (2024) 109306. Does the new version supersede the previous version?: Yes. Reasons for the new version: Implementation of the variable-composition evolutionary search feature and Pareto optimization within the XtalOpt program package. Summary of revisions: Implemented evolutionary global optimization of structures with variable compositions, the Pareto algorithm for multi-objective optimization, and the multi-cut crossover operation. Various improvements have been made to the user interface, and bugs have been fixed. Nature of problem: For a given set of chemical constituents the XtalOpt algorithm can search for (meta)stable crystal structures with fixed or varying compositions and optionally with specific functionalities – a grand challenge in computational materials science, chemistry and physics. Solution method: During the search process, the convex hull of the chemical system is calculated and updated. Instead of enthalpy, the “distance above the convex hull” is used as the target value for global optimization. The genetic operations are revised to enable the evolution of parent structures with different compositions, and to possibly produce new compositions. To further enhance the code's capability of performing a multi-objective search, the Pareto optimization scheme is implemented. This allows the user to choose from the previously implemented generalized fitness function, and the Pareto optimization scheme in searches for novel functional materials. For fast and efficient exploration of phase diagrams, easy-to-use interfaces for machine learning interatomic potentials are added to the code's package.