ZeoNet: 3D convolutional neural networks for predicting adsorption in nanoporous zeolites
Yachan Liu, Gustavo Pérez, Zezhou Cheng, Aaron Sun, Samuel C. Hoover, Wei Fan, Subhransu Maji, Peng Bai
Abstract
ZeoNet, based on 3D convolutional neural networks and a volumetric distance-grid representation, delivers an exceptional performance in predicting Henry's constants for adsorption of long-chain hydrocarbon molecules in all-silica zeolites.
Topics & Concepts
NanoporousAdsorptionConvolutional neural networkHydrocarbonRepresentation (politics)GridZeoliteComputer scienceMoleculeMaterials scienceArtificial neural networkChemistryArtificial intelligenceNanotechnologyMathematicsOrganic chemistryCatalysisGeometryPoliticsLawPolitical scienceZeolite Catalysis and SynthesisMesoporous Materials and CatalysisCatalytic Processes in Materials Science