Predicting the Temperature Dependence of Surfactant CMCs Using Graph Neural Networks

Area of Science:

• Physical Chemistry
• Computational Chemistry
• Materials Science

Background:

• The critical micelle concentration (CMC) is crucial for surfactant applications.
• Existing quantitative structure-property relationship (QSPR) and graph neural network (GNN) models predict CMC at room temperature but neglect temperature dependence.
• Temperature-dependent CMC is vital for real-world industrial applications.

Purpose of the Study:

• To develop a GNN model for predicting the temperature-dependent CMC of various surfactant classes.
• To assess the model's predictive performance across different temperature conditions and surfactant types.
• To evaluate the model's generalizability to novel and complex surfactant structures, including sustainable sugar-based surfactants.

Main Methods:

• Collected approximately 1400 data points for ionic, nonionic, and zwitterionic surfactants across multiple temperatures from public sources.
• Developed and trained a graph neural network (GNN) model to predict temperature-dependent CMC.
• Validated the model's predictive quality in two scenarios: with and without prior CMC data for specific surfactants at different temperatures.
• Tested the model's performance on sugar-based surfactants with complex molecular structures.

Main Results:

• The GNN model achieved high predictive performance with R² ≥ 0.95 on test data in both tested scenarios.
• The model demonstrated strong predictive capabilities for generalizing to unseen surfactants.
• Model performance was observed to vary across different surfactant classes.
• The model successfully predicted the CMC for complex sugar-based surfactants.

Conclusions:

• The developed GNN model effectively predicts temperature-dependent CMC for diverse surfactant classes.
• The model shows excellent generalizability and accuracy, even for surfactants not included in the training set.
• This approach is valuable for predicting the behavior of sustainable surfactants in the personal and home care industries.