Decoding Interfacial Water Orientation to Predict Surface Charge Density on a Model Sheet Using a Deep Learning Algorithm

The anomalous behavior of water at the aqueous interface of a surface has been thoroughly investigated over the past decades. However, indirect surface characterization based on the properties of interfacial water has not been well-studied at the molecular level despite its practical significance in...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physical chemistry. C 2020-01, Vol.124 (4), p.2574-2582
Main Authors: Oh, Myong In, Oh, Chang In, Gupta, Mayuri, Weaver, Donald F
Format: Article
Language:English
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The anomalous behavior of water at the aqueous interface of a surface has been thoroughly investigated over the past decades. However, indirect surface characterization based on the properties of interfacial water has not been well-studied at the molecular level despite its practical significance in science and medicine. Here, we present the theoretical development of an inference of surface charge density from orientational polarization of interfacial water using an integrated approach of information theory, molecular dynamics, and deep learning. Our findings demonstrate that the “message” about surface charge is encoded in interfacial water orientation, but the orientation angle itself is a weak predictor. The shape of the orientational distribution curve is instead a better predictor and decodable to surface charge with high accuracy using an artificial neural network. This study stimulates the design of a tunable protocol to predict surface properties using information stored in interfacial water particularly when direct surface characterization is impractical.
ISSN:1932-7447
1932-7455
DOI:10.1021/acs.jpcc.9b11442