Multitask Scanning Probe Microscopy
2026-08-10 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a smart method for scanning probe microscopy that helps choose where and what type of measurement to take next on a material sample. Their system uses a special machine learning approach called multitask Gaussian processes to understand connections between different types of measurements and locations. This method helps avoid taking measurements everywhere, saving time and reducing damage to the sample or tip. They tested their system on a large atomic force microscope with an AlScN wafer and showed it can decide when to use quick scans versus slower, more detailed ones.
Scanning probe microscopyAtomic force microscopeMultitask Gaussian processActive learningComposition-spread waferTapping modeDual AC Resonance Tracking (DART)Multimodal mappingAutonomous measurementElectromechanical properties
Authors
Aditya Raghavan, Yu Liu, Ian Mercer, JP Maria, Sergei Kalinin
Abstract
Scanning probe microscopy provides nanoscale access to structural, electrical, electromechanical, magnetic, and mechanical properties of materials. Its increasing use for wafer-scale characterization and combinatorial materials exploration creates a need to distribute measurements efficiently across large spatial domains. This is particularly important when available modalities differ in acquisition time and potential for tip and sample damage, making exhaustive multimodal mapping over spatial grids impractical. Here, we demonstrate multitask scanning probe microscopy, a live, closed-loop workflow in which a multitask Gaussian process learns spatial and cross-modal relationships and autonomously selects both the next measurement location and the next experimental protocol. The approach is implemented on an automated large-sample atomic force microscope and demonstrated on a composition-spread AlScN wafer using tapping-mode and Dual AC Resonance Tracking (DART) measurements. Paired initial measurements establish the relation between the tasks, after which noncoincident measurements are used to update both response landscapes. The resulting workflow extends active learning in scanning probe microscopy from spatial sampling to autonomous allocation of measurement modalities and provides a basis for combining rapid, weakly perturbative imaging with slower contact, electrical, electromechanical, magnetic, or spectroscopic measurements.