MOFGalaxyNet: a social network analysis for predicting guest accessibility in metal-organic frameworks utilizing...

Area of Science:

• Materials Science
• Chemistry
• Computational Science

Background:

• Metal-organic frameworks (MOFs) are versatile porous materials with applications in adsorption, separation, and catalysis.
• Predicting MOF properties like guest accessibility is crucial for targeted applications but challenging with bottom-up synthesis.
• Current methods often rely on post-synthesis characterization or complex theoretical models, hindering rapid development.

Purpose of the Study:

• To develop a predictive tool for MOF guest accessibility using only information on metal ions and organic linkers.
• To overcome limitations in bottom-up crystal engineering for high-performance MOF design.
• To facilitate the industrial-scale adoption of MOFs by simplifying performance prediction.

Main Methods:

• Utilized a social network approach, MOFGalaxyNet, to cluster diverse MOFs.
• Employed a Graphical Convolutional Network (GCN) for predicting guest accessibility based on network data.
• Integrated linker and metal ion information into the predictive model.

Main Results:

• Successfully predicted MOF guest accessibility, a key performance indicator.
• Demonstrated the robustness of the MOFGalaxyNet and GCN approach for MOF screening.
• Provided a reliable method for evaluating MOFs for host-guest interactions.

Conclusions:

• The developed tool offers a significant advancement in predicting MOF guest accessibility from basic components.
• This approach simplifies the screening process for MOFs, accelerating materials discovery.
• The findings support the rational design and efficient application of MOFs in various fields.