Polarization-assisted hydride transfer in metal-free benzophenone hyper-crosslinked polymers: Integrated experimental–DFT–ML mechanistic insight
Surfaces and Interfaces, cilt.100, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 100
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.surfin.2026.110905
- Dergi Adı: Surfaces and Interfaces
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, INSPEC
- Anahtar Kelimeler: DFT, Hydrogen evolution, Machine learning, Metal-free catalysis, NaBH4 methanolysis, Structure–activity relations
- Dokuz Eylül Üniversitesi Adresli: Evet
Özet
Benzophenone-derived hyper-crosslinked polymers (BzPHCPs) were prepared by Friedel–Crafts chemistry and subsequently investigated for the catalytic release of H2 generation via NaBH4 in methanolysis. Structural characterization (FTIR, XPS, BET, SEM, and TGA) confirmed the formation of stable π-conjugated networks with tunable porosity (319.7–679.5 m2 g-1) and donor–acceptor functionalities. Among the series, the asymmetric BzPHCP-4, bearing l -tryptophan-derived donor and 2,3-pyridinedicarboxylic-acid-derived acceptor units, exhibited the highest hydrogen generation rate (4.0 × 104 mL H2 min-1 gcat-1) together with a low activation energy of 35.8 kJ mol-1. Density functional theory (DFT) calculations elucidated how electrostatic potential gradients, charge delocalization, and frontier-orbital localization govern hydride activation. BzPHCP-4 displayed the narrowest HOMO–LUMO gap and dual-site reactivity arising from donor–acceptor asymmetry, consistent with its enhanced catalytic performance. Statistical and data-driven analyses, including adaptive neuro-fuzzy inference system (ANFIS), analysis of covariance (ANCOVA), and light gradient boosting machine (LightGBM) regression, identified reaction time and temperature as the dominant kinetic parameters. The LightGBM model achieved the highest predictive accuracy (R2 = 0.9981), whereas the ANFIS model also demonstrated excellent agreement with experimental data (R2 = 0.9971). Shapley additive explanations (SHAP) analysis further confirmed the key influence of thermal and temporal factors on hydrogen evolution. By integrating experimental, computational, and data-driven approaches, this study establishes a mechanistically informed framework linking molecular polarization, charge dynamics, and macroscopic catalytic performance. These findings highlight benzophenone-based hyper-crosslinked polymers as promising and sustainable candidates for metal-free hydrogen-evolution systems.