Development of a deep learning-based group contribution framework for targeted design of ionic liquids
Sadah Mohammed, Fadwa Eljack, Monzure-Khoda Kazi, Mert Atilhan
Abstract
• A data-driven modeling framework for the design of targeted ILs. • Incorporating group contribution method in DL model for IL properties prediction • Developing DNN-GC and ANN-GC for ILs' viscosity and CO 2 solubility prediction. • Correlating IL viscosity and CO 2 solubility using merged DNN-GC and ANN-GC models. • Utilizing correlation to identify optimal IL structure for maximal CO 2 absorption. In this article, we present a novel deep learning-based group contribution framework for the targeted design of ionic liquids (ILs). This computational framework can expedite and improve the process of finding desirable molecular structures of IL via accurate property predictions in a data-driven manner. Our proposed framework consists of two essential steps: establishing a correlation between IL viscosity and CO 2 solubility by merging two deep learning models (DNN-GC and ANN-GC) and utilizing this correlation to identify the optimal IL structure with maximal CO 2 absorption capacity. Our model achieves high accuracy with R 2 values of 95%, 94.2%, and 96.4% for DNN-GC, ANN-GC, and DNN-ANN-GC, respectively. Correlation results align with the experimental data, affirming the applicability of our framework. Finally, the algorithm is employed in a CO 2 capture case study to generate and select the best-performing novel ILs, which exhibit behavior consistent with established ILs in the literature.