Selective oxidation at low temperatures without alkali of biomassis a promising and sustainable avenue to manufacture glycolic acid (GA), a biodegradable functional material to protect the environment. However, producing glycolic acid with high selectivity and yield using the traditional research and development approach is time-consuming and labor-intensive. To this context, a hybrid deep learning framework driven by data and reaction mechanisms for predicting sustainable glycolic acid production was proposed, considering the lack of related reaction mechanisms in the machine learning algorithms. Results showed that the fully connected residual network exhibited superior performance (average R2=0.98) for the multi-task prediction of conversion rate, GA, and by-product yields, therefore employed for the following super parameters optimization by the genetic algorithm. The L further identifies that using the optimized operating parameters, the fossil energy demand and greenhouse emissions have decreased by 2.96% and 3.00%, respectively.

Rational regulation of local environment of Ti-sites in TS-1 harbours tremendous industrial and scientific significance to epoxidation reactions. Herein, we report a facile and environment-friendly strategy to boost catalytic selectivity by covering the Ti-sites on external surface of TS-1 (ITS-1) without sacrificing activity of internal Ti sites. By quantitative analysis of d3-acetonitrile and quinoline-DRIFTS, 1H MAS NMR and XPS, it is found that percentage of external Ti-sites decreased from 11% to 6% after depositing surface silica islands. This successfully inhibits ring-opening reaction of 1,2-epoxyhexane on catalyst surface, as demonstrated by slower kinetic decomposition rate of 1,2-epoxyhexane. Compared with conventional TS-1 catalyst, selectivity of 1,2-epoxyhexane over ITS-1 catalyst significantly increased from 83.5% to 98.5% while maintaining high 1-hexene conversion. Furthermore, overmuch surface silica coverage only leads to extremely low conversion (2%) due to inhibition of mass transfer. This work paves the way for rational construction of Ti-containing catalysts for 1-hexene epoxidation

A strategy to boost hydrogen generation from ammonia borane hydrolysis was proposed by engineering Ru local structures via the employment of Cl-free Ru precursors and acid-leaching of the bimetallic Ru-Co catalysts. The effects of Cl and Co were separately investigated by comparing the Cl-free/Cl-containing monometallic Ru catalysts and Co-leaching/Co-containing bimetallic Ru-Co catalysts, respectively. Decoupled from the Co electronic effects (electron transfer from Co to Ru), the Cl geometric effects (site blockage and steric hindrance) were identified as the main factors, which could be further promoted by introducing more low-coordinated Ru sites. The increase of Co content gives rise to more edge-like Ru active sites after acid leaching, with a lower H2O activation barrier for hydrogen production. Consequently, al-Ru1Co with the most edge-like Ru active sites demonstrated a 4.1-fold increase in catalytic activity. The insights revealed here might guide the design and preparation of metal catalysts with the maximum active sites.