Shanghai University

About: Shanghai University is a education organization based out in Shanghai, Shanghai, China. It is known for research contribution in the topics: Microstructure & Catalysis. The organization has 59583 authors who have published 56840 publications receiving 753549 citations. The organization is also known as: Shànghǎi Dàxué.

Ion sieving in graphene oxide membranes via cationic control of interlayer spacing

Abstract: Graphene oxide membranes-partially oxidized, stacked sheets of graphene-can provide ultrathin, high-flux and energy-efficient membranes for precise ionic and molecular sieving in aqueous solution. These materials have shown potential in a variety of applications, including water desalination and purification, gas and ion separation, biosensors, proton conductors, lithium-based batteries and super-capacitors. Unlike the pores of carbon nanotube membranes, which have fixed sizes, the pores of graphene oxide membranes-that is, the interlayer spacing between graphene oxide sheets (a sheet is a single flake inside the membrane)-are of variable size. Furthermore, it is difficult to reduce the interlayer spacing sufficiently to exclude small ions and to maintain this spacing against the tendency of graphene oxide membranes to swell when immersed in aqueous solution. These challenges hinder the potential ion filtration applications of graphene oxide membranes. Here we demonstrate cationic control of the interlayer spacing of graphene oxide membranes with angstrom precision using K+, Na+, Ca2+, Li+ or Mg2+ ions. Moreover, membrane spacings controlled by one type of cation can efficiently and selectively exclude other cations that have larger hydrated volumes. First-principles calculations and ultraviolet absorption spectroscopy reveal that the location of the most stable cation adsorption is where oxide groups and aromatic rings coexist. Previous density functional theory computations show that other cations (Fe2+, Co2+, Cu2+, Cd2+, Cr2+ and Pb2+) should have a much stronger cation-π interaction with the graphene sheet than Na+ has, suggesting that other ions could be used to produce a wider range of interlayer spacings.

Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks

Abstract: In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active toward a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria), it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecule...