Materials informatics approach using domain modelling for exploring structure-property relationships of polymers

Area of Science:

• Materials Science
• Polymer Chemistry
• Data Science

Background:

• Understanding the relationship between domain structure and physical properties is crucial for polymer material development.
• ¹H-static solid-state NMR (ssNMR) provides insights into polymer domains (mobile, intermediate, rigid) but spectral overlap complicates analysis.
• Accurate domain structure estimation is challenging due to spectral complexity.

Purpose of the Study:

• To develop a materials informatics approach for analyzing polymer domain structures and their relationship with physical properties.
• To overcome the limitations of spectral overlap in ¹H-static ssNMR for domain analysis.
• To integrate domain modeling with meta-information analysis for comprehensive structure-property insights.

Main Methods:

• Applied a materials informatics approach combining domain modeling and meta-information analysis.
• Processed ¹H-static ssNMR data from 120 polymer materials using short-time Fourier transform to extract frequency, intensity, and T₂ relaxation times for different domains.
• Integrated estimated domain proportions with meta-information (elements, functional groups, thermophysical properties) and analyzed using self-organization maps and market basket analysis.

Main Results:

• Characterized four distinct polymer domains based on T₂ relaxation times: Mobile (0.96 ms), Intermediate (Mobile) (0.55 ms), Intermediate (Rigid) (0.32 ms), and Rigid (0.11 ms).
• Successfully integrated domain proportion data with material meta-information.
• Demonstrated the utility of self-organization maps and market basket analysis for exploring structure-property relationships.

Conclusions:

• The developed materials informatics approach effectively addresses the challenges of analyzing complex polymer domain structures from ssNMR data.
• This method facilitates a deeper understanding of structure-property relationships in polymer materials with multiple domains.
• The approach offers a powerful tool for polymer material design and optimization.