Molecular Identification with Atomic Force Microscopy and Conditional Generative Adversarial Networks

TL;DR

This paper introduces a novel machine learning model that uses AFM images at various distances to accurately identify molecular structures and chemical compositions, enhancing resolution and analysis capabilities.

Contribution

The paper presents a conditional GAN model that converts AFM image stacks into detailed molecular structures, enabling complete chemical and structural identification from microscopy data.

Findings

High accuracy in molecular identification demonstrated on simulated data.

Effective application to experimental AFM images shows practical potential.

Model captures detailed chemical and structural information from AFM images.

Abstract

Frequency modulation (FM) Atomic Force Microscopy (AFM) with metal tips functionalized with a CO molecule at the tip apex has provided access to the internal structure of molecules with totally unprecedented resolution. We propose a model to extract the chemical information from those AFM images in order to achieve a complete identification of the imaged molecule. Our Conditional Generative Adversarial Network (CGAN) converts a stack of AFM images at various tip-sample distances into a ball-and-stick depiction, where balls of different color and size represent the chemical species and sticks represent the bonds, providing complete information on the structure and chemical composition. The CGAN has been trained and tested with the QUAM-AFM data set, that contains simulated AFM images for a collection of 686,000 molecules that include all the chemical species relevant in organic chemistry. Tests with a large set of theoretical images and few experimental examples demonstrate the accuracy and potential of our approach for molecular identification.