Data-Driven Deciphering Structure-Mössbauer Spectroscopy Relationships in Iron-Based Compounds.

Area of Science:

• Materials Science
• Computational Chemistry
• Spectroscopy

Background:

• Accurate material structure identification is vital for structure-property relationships.
• Metastable phases and their transformations complicate material analysis.
• Lack of reference data hinders spectroscopic identification of complex phases.

Purpose of the Study:

• To develop a unified strategy for identifying metastable phases in iron-based intermetallic compounds.
• To overcome limitations in experimental analysis due to complex phase coexistence and transformations.
• To create a data-driven framework complementing experimental techniques.

Main Methods:

• Integrating structure prediction and thermodynamic stability calculations.
• Employing machine learning models to link local structural features with Mössbauer parameters.
• Utilizing simulated Mössbauer spectra from a theoretical database for spectral interpretation.
• Focusing on iron-based intermetallic compounds, specifically iron sulfides.

Main Results:

• Successfully identified previously unresolved candidate iron sulfide metastable phases.
• Established quantitative structure-Mössbauer parameter relationships using machine learning.
• Created a theoretical spectral database to aid experimental spectrum interpretation.
• Demonstrated the framework's ability to analyze dynamically evolving systems.

Conclusions:

• The proposed data-driven framework significantly enhances the accuracy and efficiency of metastable phase identification.
• This integrated approach provides a robust complement to experimental methods in materials science.
• Enables more precise analysis of complex materials under reaction conditions.