Some Ideas of Informational Deep Neural Networks Structural Organization

Smolin, Vladimir

doi:10.1007/978-3-319-40663-3_66

Cited by 1 publication

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Combined with computer simulations, machine learning (ML) has recently emerged as an extraordinary tool in many research areas, − especially in predicting the variation of physicochemical properties of materials with multiple variables , and in efficient material mining. − Although various ML methods have been proposed, the artificial neural network (ANN) is one of the most widely employed ML strategies. − In this work, an ANN is employed to explore the interfacial elastocapillarity of microgel capsules on substrates, aiming to identify the adsorption regime over wider ranges of multiple variables. Finally, the structure–property relationship between the external conditions and the particular features of structures is established and discussed.…”

Section: Introductionmentioning

confidence: 99%

Interfacial Engineering of Thermoresponsive Microgel Capsules: Polymeric Wetting vs Colloidal Adhesion

et al. 2019

View full text Add to dashboard Cite

Control of the microstructure of microgels adsorbed on solid surfaces plays an essential role in various fields, including lithography, optical sensing, and biocatalysis. Here, we adopt an experimentally validated molecular dynamics simulation approach to investigate the structural properties and deswelling dynamics of thermoresponsive microgel capsules on solid substrates. Specifically, by examining the interfacial elastocapillarity of the adsorbed microgel capsules, we find that the poorly cross-linked microgel capsules (i.e., cross-link density ψ ≤ 0.0217) display a crossover adsorption regime between polymeric wetting and colloidal adhesion, whereas the highly cross-linked microgel capsules present only colloidal adhesion adsorption. As the system temperature increases, the microgel capsules progressively transform from the fully swollen state to the fully collapsed state on the solid substrates, whereas the capsule architecture remains. The adsorption regime of the microgel capsule is mutually determined by the elastic deformation and surface attraction strength. In addition, the elastic deformation is attributed to the internal structure of the microgel capsule, which varies with the system temperature and cross-link density. Aiming to identify the adsorption regime over wider ranges of the control variables, a machine learning study on an artificial neural network is further carried out and a three-dimensional phase diagram of the adsorption regime and multiple control variables is constructed, unraveling the comprehensive relationship among the adsorption regime and the operational parameters.

show abstract