Imaging surfaces at the space–time limit: New perspectives of time-resolved scanning tunneling microscopy for ultrafast surface science

Many fundamental processes in nature occur on ultrashort time scales within picoseconds to attoseconds, and on intrinsic length scales from nanometers to picometers. The structure of crystalline solids is dictated by long range order and the periodic arrangement of atoms, but the elementary excitations that define its interaction with the environment may vary locally at the atomic scale. Multiple domains and phases can coexist on length scales down to a few nanometer, and impurities and defects can influence the collective many-body response of solids at the single-atom level. Ultrafast pump–probe techniques provide valuable information about fundamental many-body interactions in solids and at surfaces, but spatially average over macroscopic spot sizes such that the influence of local order or disorder at angstrom scales is not directly accessible. Therefore, real-space observation of ultrafast dynamics with atomic spatial resolution is highly desirable, and motivates the development of time-resolved ultrafast scanning tunneling microscopy (USTM) since the early 1990’s. Tremendous progress has been made in this field in the past decade, and a number of breakthrough achievements have significantly advanced our possibilities to add ultrafast time resolution to the angstrom spatial resolution of STM. This article reviews new technical approaches and developments in the field of USTM. A particular focus will be the classification of light-matter interaction in tunnel junctions, based on the criteria for adiabatic tunneling from Keldysh's theory of strong-field ionization and a tunneling time as defined by Büttiker and Landauer, and on Tucker's definition of quantum detection in a tunnel junction mixer. Moreover, various mechanisms to generate an ultrafast tunneling current in USTM are discussed and are to some extent related to those from other techniques such as optical spectroscopy or photoemission spectroscopy. The resulting new possibilities for imaging the ultrafast dynamics of electronic and vibrational excitations at surfaces with USTM will be highlighted. Finally, the article outlines possible future directions of USTM for studying ultrafast processes and light-induced phenomena at surfaces and in quantum materials.