Automated Structure Discovery for Scanning Tunneling Microscopy

Scanning tunnelling microscopy (STM) with a functionalized tip apex reveals the geometric and electronic structure of a sample within the same experiment. However, the complex nature of the signal makes images difficult to interpret and has so far limited most research to planar samples with a known chemical composition. Here, we present automated structure discovery for STM (ASD-STM), a machine learning tool for predicting the atomic structure directly from an STM image, by building upon successful methods for structure discovery in non-contact atomic force microscopy (nc-AFM). We apply the method on various organic molecules and achieve good accuracy on structure predictions and chemical identification on a qualitative level, while highlighting future development requirements to ASD-STM. This method is directly applicable to experimental STM images of organic molecules, making structure discovery available for a wider SPM audience outside of nc-AFM. This work also opens doors for more advanced machine learning methods to be developed for STM discovery.