Deep learning reveals key predictors of thermal conductivity in covalent organic frameworks

2024-09-10

arXiv.org

AI Summary

Accurate prediction of thermal conductivity in covalent organic frameworks (COFs) remains challenging due to limitations of conventional structural descriptors. Method: This study overcomes these limitations by developing an attention-based deep learning model capable of cross-structural generalization. Contribution/Results: We identify pendant molecular branches as the dominant structural determinant of COF thermal conductivity—first reported herein—and uncover a novel physical mechanism: vibrational mismatch induced by pendant groups suppresses phonon transport, validated via molecular dynamics simulations and vibrational density-of-states analysis. The model achieves 92% prediction accuracy on out-of-distribution COF structures. Feature importance analysis and MD-based verification jointly confirm the pivotal role of pendant functionalities. This work establishes an interpretable, high-accuracy paradigm for the rational design of COF-based thermal management materials.

Technology Category

• Machine Learning: Structured Learning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The thermal conductivity of covalent organic frameworks (COFs), an emerging class of nanoporous polymeric materials, is crucial for many applications, yet the link between their structure and thermal properties remains poorly understood. Analysis of a dataset containing over 2,400 COFs reveals that conventional features such as density, pore size, void fraction, and surface area do not reliably predict thermal conductivity. To address this, an attention-based machine learning model was trained, accurately predicting thermal conductivities even for structures outside the training set. The attention mechanism was then utilized to investigate the model's success. The analysis identified dangling molecular branches as a key predictor of thermal conductivity, a discovery supported by feature importance assessments conducted on regression models. These findings indicate that COFs with dangling functional groups exhibit lower thermal transfer capabilities. Molecular dynamics simulations support this observation, revealing significant mismatches in the vibrational density of states due to the presence of dangling branches.

Problem

• Covalent Organic Frameworks
• Thermal Conductivity
• Structure-Property Relationship

Innovation

• Deep Learning Attention Mechanism
• Covalent Organic Frameworks (COFs)
• Thermal Conductivity