Colloquium: To the nanoworld and beyond, in real time: what can in situ electron microscopy do for us?

Aberration corrected scanning transmission electron microscopy (STEM) combined with electron energy-loss spectroscopy (EELS) and in situ capabilities have brought sub-Ångstrom electron beams into visualization and quantification with high accuracy of atomistic phenomena at work. Energy can be transferred into exciting many different degrees of freedom so dynamics, kinetics and other dimensions can be explored at the flick of a switch. This talk will review several examples regarding how high resolution in situ STEM techniques (including imaging, spectroscopy or diffraction) can be used to map elusive physical phenomena in functional materials. Examples will include in the first place the mapping of exotic electronic phenomena in nanomaterials for spintronics. Cu nanowires with very low Bi doping levels are very interesting systems due to the giant spin Hall effect reported in bulk, associated with the presence of the high spin orbit coupling Bi species. In situ electric current injection through the wires reveals the presence of an orbital Hall effect (OHE) which can be spatially mapped using energy-loss magnetic chiral dichroism (EMCD) spectroscopic techniques. The change of sign in the dichroic signal measured on the nanowire surface upon inversion of electric current exhibits promising fingerprints of OHE derived orbital accumulation within length scales of 2-3 nm. Another example can be found in studies of the magnetic properties of complex oxide ferromagnetic La0.7Sr0.3MnO3 / insulating SrTiO3 interfaces. In these systems, an externally applied electric field may provide a knob to tune the amount of O vacancies at the interface, thus changing the local magnetic properties which can be mapped in situ. On a different front, in situ 4D-STEM under bias provides the means for atomic resolution characterization of polarization switching phenomena in ferroelectric heterostructures, which may also be affected by local defects. However, noise hampers such measurements, so interpretation relies on a multidisciplinary approach involving advanced characterization, modelling and analysis of the large experimental datasets. Attempts to establish on-the-fly, real time quantification tools that can help with the decision making process on-the-fly will be discussed.