Accelerated Scheme to Predict Ring-Opening Polymerization Enthalpy: Simulation-Experimental Data Fusion and Multitask...

Area of Science:

• Polymer Chemistry
• Computational Chemistry
• Materials Science

Background:

• Ring-opening enthalpy (ΔHROP) is crucial for designing recyclable polymers via ring-opening polymerization (ROP).
• Current computational methods for calculating ΔHROP are computationally intensive, hindering efficient polymer design.
• Developing accurate and rapid methods to predict ΔHROP is essential for advancing sustainable polymer chemistry.

Purpose of the Study:

• To develop a generalizable machine learning (ML) model for predicting ring-opening enthalpy (ΔHROP).
• To enable faster and more efficient design of novel depolymerizable polymers.
• To bridge the gap between computational cost and the need for accurate thermodynamic data in polymer science.

Main Methods:

• Developed a machine learning model trained on experimental data and computed simulation results for ΔHROP.
• Utilized a combination of experimental measurements and first-principles simulation data for model training.
• Focused on creating a model that is both accurate and computationally inexpensive for predictions.

Main Results:

• The ML model achieves a prediction accuracy of approximately 8 kJ/mol for ΔHROP, nearing chemical accuracy.
• Predictions using the developed ML model are virtually instantaneous, drastically reducing computational time.
• The model demonstrates generalizability, allowing for systematic exploration of chemical diversity.

Conclusions:

• The developed ML model offers a computationally efficient and accurate approach to predict ΔHROP.
• This work facilitates the accelerated design and discovery of new recyclable and depolymerizable polymers.
• The ML model serves as a valuable tool for advancing sustainable polymer development and applications.