Predicting hydration layers on surfaces using deep learning.

Yashasvi S Ranawat,Adam S Foster,Ygor M Jaques

doi:10.1039/d1na00253h

Yashasvi S Ranawat, Adam S Foster + Show 1 more

Open Access

https://doi.org/10.1039/d1na00253h

Copy DOI

Journal: Nanoscale advances	Publication Date: Jan 1, 2021
Citations: 7	License type: CC BY 3.0

Affiliation: Aalto University, Kanazawa University

Abstract

Characterisation of the nanoscale interface formed between minerals and water is essential to the understanding of natural processes, such as biomineralization, and to develop new technologies where function is dominated by the mineral–water interface. Atomic force microscopy offers the potential to characterize solid–liquid interfaces in high-resolution, with several experimental and theoretical studies offering molecular scale resolution by linking measurements directly to water density on the surface. However, the theoretical techniques used to interpret such results are computationally intensive and development of the approach has been limited by interpretation challenges. In this work, we develop a deep learning architecture to learn the solid–liquid interface of polymorphs of calcium carbonate, allowing for the rapid predictions of density profiles with reasonable accuracy.

Highlights

Characterisation of the nanoscale interface formed between minerals and water is essential to the understanding of natural processes, such as biomineralization, and to develop new technologies where function is dominated by the mineral–water interface
We develop a deep learning architecture to learn the solid–liquid interface of polymorphs of calcium carbonate, allowing for the rapid predictions of density profiles with reasonable accuracy
The demonstration of a direct relationship between experimental Atomic force microscopy (AFM) force data and water densities opened an easier route to interpret the images through simulations,[19,20] which were further advanced by inclusion of the in uence of the tip's hydration structure in the forces[21] and an analysis of the role of tip radius.[17]

Summary

Introduction

Characterisation of the nanoscale interface formed between minerals and water is essential to the understanding of natural processes, such as biomineralization, and to develop new technologies where function is dominated by the mineral–water interface. These technical challenges encumber nanoscale characterisation of solid–liquid interfaces through AFM and is in stark contrast to the breakthroughs in molecular characterization offered by low-temperature functional-tip AFM in ultra-high vacuum.[22,23] In an effort to provide a rapid and reliable interpretation tool, in this work we develop a machine learning approach that can predict water densities, and AFM images, directly from the atomic structure.

Objectives

Results

Conclusion