The evaluation of absorption, distribution, metabolism, exclusion, and toxicity (ADMET) properties plays a key role in a variety of domains including industrial chemicals, agrochemicals, cosmetics, environmental science, food chemistry, and particularly drug development. Since molecules are often intrinsically described as molecular graphs, graph neural networks have recently been studied to improve the prediction of ADMET properties. Among many graph neural networks published in recent years, Graph Isomorphism Network (GIN) is a relatively recent and very promising one. In this paper, we propose an enhanced GIN, called MolGIN, via exploiting the bond features and differences influence of the atom neighbors to end-to-end predict ADMET properties. Based on GIN, MolGIN concatenates the bond feature together with node feature in the feature aggregator and applies a gate unit to adjust the atomic neighborhood weights to map the differences in the interaction strength between the central atom and its neighbors, such that more meaningful structural patterns of molecules can be explored toward better molecular modeling. Extensive experiments were conducted on seven public datasets to evaluate MolGIN against four baseline models with benchmark metrics. Experimental results of MolGIN were also compared with state-of-the-art results published in the last three years on each dataset. Experimental results in terms of RMSE and AUC show that MolGIN significantly boosts the prediction performance of GIN and markedly outperforms the baseline models, and achieves comparable or superior performance to state-of-the-art results.

In this work, we established a novel microstructured chemical system to intensify the synthesis process of poly­(vinyl butyral) (PVB) based on the mixing enhancement principle of poly­(vinyl alcohol) (PVA)–n-butanal aqueous solution with hydrochloric acid. Because of the high mixing performance, a higher temperature (30–60 °C) with much higher reaction rate was applied. Through a study of the parameters of flow rate, temperature, and feeding ratios of n-butanal and H+ to the hydroxyl groups in PVA, the apparent butyral group content in PVB reached more than 60% with a residence time of 1 min in a microreactor, and reached 75% with an additional hour of aging time, much shorter than the conventional 8–10 h in low temperature batch reactors. The size of the primary PVB particle was smaller than 10 μm, which might reduce the time and water consumption in the post washing process. This work provides a good example for operating reactions between macromolecules and small molecules in a microreactor system.

In this work, we established a cascade cooling and two-step feeding technology for the energy-saving and fast synthesis of polyvinyl butyral (PVB) with a low molecular weight in a microreactor system. Because of the strong mixing capacity of the microreactor and the relatively higher initial reaction temperature of 60 °C compared to existing methods, the condensation reaction between the raw materials of poly(vinyl alcohol) (PVA) and n-butanal was highly intensified and the reaction rate was faster, along with less energy consumption for cooling. By reducing the aging temperature to 40−55 °C, which is far below the glass transition temperature of PVB, we could also lower energy consumption and improve the morphology of PVB. Benefiting from the new technology, the acetalization degree (AD) of PVB reached more than 78%. By applying the two-step n-butanal feeding method in the new technology, the utilization ratio of n-butanal was dramatically increased to more than 99.7%, which was the best result among current published reports. Moreover, a PVB product with a low molecular weight has been successfully prepared with the new technology. This work demonstrates a promising continuous microreaction process for greener and energy-saving synthesis of PVB.