2026-10-13T05:14:58Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33972916 2026-09-23T11:38:22Z category_1 category_8 category_873 category_734 category_931 portal_63 item_type_3 month_year_09_2026

A Transferable Four-Tier Decision Architecture for Gas Solubility Prediction in Ionic Liquids Jiaping Zhou (17361696) Yang Lei (136316) Hong Huang (119275) Yuqiu Chen (6414422) Biophysics Microbiology Chemical Sciences not elsewhere classified Biological Sciences not elsewhere classified Information Systems not elsewhere classified step ablation protocol span distinct regions shap analysis reveals physicochemical interaction spectrum gas solubility prediction gas chemistries represented tier decision architecture electrostatic binary switch 2 </ sub electrostatic interactions volume effects thermodynamic baseline state classifier specific artifacts solvation thermodynamics optimal thermodynamic mediated microtuning ionic liquids input representation groupkfold cross evaluation beyond encodes knowledge data set autonomously learned We develop a transferable, physics-inspired feature engineering framework that encodes knowledge of solvation thermodynamics, electrostatic interactions, free-volume effects, and acid–base chemistry directly into the input representation, ensuring that the model learns causal physical relationships rather than data set-specific artifacts. Through a four-step ablation protocol under GroupKFold cross-validation, where generalization is evaluated on entirely unseen ionic liquid (cation + anion)gas compositional triplets, we identify a Pareto-optimal Thermodynamic + Electrostatic model. Critically, SHAP analysis reveals that the model has autonomously learned a four-tier decision architecture that mirrors established physicochemical hierarchies: a thermodynamic baseline, structure-mediated microtuning, an electrostatic binary switch, and a phase-state classifier. The transferability of this architecture was further evaluated through zero-shot generalization to three entirely unseen gases (C₂H₄, SO₂, and N₂), which span distinct regions of the physicochemical interaction spectrum and extend the evaluation beyond the gas chemistries represented during training. 2026-09-23T00:00:00Z Dataset Dataset 10.1021/acs.jpcb.6c05017.s001 https://figshare.com/articles/dataset/A_Transferable_Four-Tier_Decision_Architecture_for_Gas_Solubility_Prediction_in_Ionic_Liquids/33972916 CC BY-NC 4.0