AI‑Driven System Lets Atomic Force Microscopes Zero In on Key Nanoscale Features

Scientists at the Department of Energy's Oak Ridge National Laboratory have introduced a machine‑learning framework that enables atomic force microscopes to automatically pinpoint the most informative regions of a sample, cutting down the time required for high‑resolution imaging.

Atomic force microscopy (AFM) is a cornerstone technique for visualizing surfaces at the nanometer scale, but the process traditionally relies on researchers manually selecting scan locations. That approach can miss subtle but crucial features and often involves repeated trial‑and‑error to capture the right area.

The new AI system tackles that limitation by continuously analysing the data streamed from the probe and using predictive algorithms to decide where the microscope should focus next. In practice, the software evaluates early‑stage measurements, ranks potential scan zones by their informational value, and directs the instrument to the most promising spots without human intervention.

Initial trials at ORNL showed that the autonomous routine could locate critical nanoscale structures more rapidly than standard, operator‑driven scans. By prioritising high‑impact regions, the framework reduced overall imaging time while preserving, and in some cases enhancing, the quality of the resulting maps.

Accelerated, data‑rich AFM measurements have far‑reaching implications for fields ranging from semiconductor manufacturing to energy‑storage research. Faster identification of defects, grain boundaries, or phase transitions can speed up material‑design cycles, aligning with the Department of Energy’s broader goals of advancing clean‑energy technologies.

Looking ahead, the team plans to expand the platform to work with other scanning probe instruments and to refine the learning models with larger datasets. If successful, the technology could become a standard component of nanoscale laboratories, offering a blend of speed and precision that could reshape how scientists explore the atomic landscape.