Model-Based Control of the Scanning Tunneling Microscope: Enabling New Modes of Imaging, Spectroscopy, and Lithography

The invention of the STM dates back to the work of Binnig and Rohrer in the early 1980s [27], [S1], [S2], whose seminal contribution was rewarded by the 1986 Nobel Prize in Physics "for the design of the scanning tunneling microscope." Forty years later, the STM remains the best existing tool for studying electronic, chemical, and physical properties of conducting and semiconducting surfaces with atomic precision. It has opened entirely new fields of research, enabling scientists to gain invaluable insight into the properties and structure of matter at the atomic scale. Recent breakthroughs in STM-based automated hydrogen depassivation lithography (HDL) on silicon have resulted in the STM being considered a viable tool for the fabrication of error-free silicon-based quantum electronic devices. Despite the STM's unique ability to interrogate and manipulate matter with atomic precision, it remains a challenging tool to use. It turns out that many issues can be traced back to the STM's feedback control system, which has remained essentially unchanged since its invention about 40 years ago. This article explains the role of the feedback control system of the STM and reviews some of the recent progress made possible in imaging, spectroscopy, and lithography by making appropriate changes to the STM's feedback control loop. We believe that the full potential of the STM is yet to be realized, and the key to new innovations will be the application of advanced model-based control and estimation techniques to this system.