Answers are grounded in this paper and cite their sources.
CO2 oxidative dehydrogenation of ethane (CO2-ODHE) couples ethane upgrading with CO2 utilization, but reliable catalyst screening remains hindered by cross-literature heterogeneity and publication-specific data leakage. Here, 511 records from 59 publications were curated into 510 canonical entries, including 504 valid C2H4 yields. Six machine learning algorithms were evaluated using fold-contained preprocessing, DOI-grouped validation, chronological validation, and 50 repeated DOI-stratified holdouts. Gradient Boosting and XGBoost showed the most consistent cross-literature performance, with mean repeated-holdout R2 values of 0.539 ± 0.115 and 0.544 ± 0.105, respectively. However, chronological validation yielded R2 values of approximately –0.03 for both models, indicating weak generalization to more recent literature. Convex stacking reduced variability but did not improve mean predictive accuracy. Subgroup analysis revealed weaker transfer to nonCr catalysts. SHAP identified reaction temperature, BET surface area, and CO2 partial pressure as the dominant predictors. Applicability-domain analysis was further used to prioritize candidates for prospective validation, providing a more realistic framework for machine-learning-assisted CO2-ODHE catalyst screening.
No takes yet. Share an insight, caveat, or question.
Shi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: