Atomic-scale structural inversion of interfacial water from atomic force microscopy.

Area of Science:

• Surface Science and Interfacial Phenomena
• Atomic Force Microscopy (AFM) and Nanotechnology
• Computational Chemistry and Materials Science

Background:

• The atomic structure of interfacial water is critical for diverse scientific fields, including electrochemistry, catalysis, and biological engineering.
• Atomic Force Microscopy (AFM) offers high spatial resolution but struggles with direct atomic water structure determination due to weak hydrogen contrast and complex image interpretation.
• Existing methods face challenges in linking AFM images to precise atomic configurations, especially for hydrogen-rich interfacial systems.

Purpose of the Study:

• To develop a novel framework for accurately determining the atomic-scale structure of interfacial water from multi-height AFM images.
• To establish a direct, interpretable link between AFM contrast and the underlying atomic positions and hydrogen orientations.
• To provide a robust initialization strategy for AFM inverse problems, reducing search space for complex interfacial systems.

Main Methods:

• Developed a closed-loop, physics-informed structural inversion framework integrating conditional generative adversarial learning.
• Incorporated an explicit structural descriptor encoding atomic positions and hydrogen orientations for direct AFM contrast linkage.
• Trained the framework on simulated AFM data and applied automated/structure-aware postprocessing for experimental data.

Main Results:

• Achieved high accuracy in localizing atomic positions and determining hydrogen orientation from simulated AFM data.
• Generated physically plausible atomic structures from experimental AFM images that reproduce observed contrast after relaxation.
• Demonstrated robustness against experimental noise and limited height sampling, yielding reliable structural initializations.

Conclusions:

• The developed framework successfully bridges the gap between AFM imaging and atomic-scale water structure determination.
• This approach offers a significant advancement for analyzing interfacial water and other hydrogen-rich systems using AFM.
• Provides a generalizable strategy for tackling complex inverse problems in nanoscience and materials characterization.