AI Imaging of Molecules in Complex Mixtures

Illustration enlarge

AI "looking" at AFM images of organic molecules and making decisions for finding the best imaging parameters.

Scientific Achievement

CFN staff and collaborators implemented AI methods to automatically collect high-resolution atomic force microscopy (HR-AFM) images of complex petroleum-based mixtures.

Significance and Impact

  • Broad potential uses imaging complex molecular mixtures, such as petroleum, combustion or pyrolysis products, and organic compounds found in meteorites and oceans.
  • HR-AFM is the only tool for imaging individual molecules but is complex and exceedingly time-consuming.

Research Details

High-resolution atomic force microscopy (HR-AFM) is a unique tool used to study molecules and complex molecular mixtures because it can image individual molecules and reveal their chemical structure. However, the job is very labor intense and time consuming plus requires expert knowledge and years of experience. A typical single molecule takes 20 to 30 minutes to image once the necessary setup is complete, a process that may itself take days. To get reasonable statistics on molecular mixtures, weeks to months of 24-7 microscope operations are required. This is workable for a handful of molecules, but after many hours even the best operator gets tired and may make mistakes, which can result in lost time and may even require re-starting the experiment. Thus, automatization is key to more effective machine operation and achieving high throughput. 

Publication Reference

Arias S., Zhang Y., Zahl P., Hollen S. “Autonomous Molecular Structure Imaging with High-Resolution Atomic Force Microscopy for Molecular Mixture Discovery” J. Phys. Chem. A 127, 29 (2023).


Acknowledgment of Support

This research used the LT-STM/HR-AFM facility of the Center for Functional Nanomaterials (CFN), which is a U.S. Department of Energy Office of Science User Facility, at Brookhaven National Laboratory under Contract No. DE-SC0012704. This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Workforce Development for Teachers and Scientists, and Office of Science Graduate Student Research (SCGSR) program. The SCGSR program is administered by the Oak Ridge Institute for Science and Education (ORISE) for the DOE. ORISE is managed by ORAU under Contract No. DE-SC0014664. All opinions expressed in this paper are the authors’ opinions and do not necessarily reflect the policies and views of DOE, ORAU, or ORISE. 

2024-21821  |  INT/EXT  |  Newsroom