Towards near-atomic resolution imaging of microsecond nanoscale dynamics

Time-resolved electron microscopy has made significant progress in recent years, with some groups now working on instruments that offer attosecond temporal resolution. While much of the research in the field revolves around the improvement of temporal resolution, atomic-resolution imaging of nanoscale dynamics has remained elusive.This thesis describes the development of two methods for time-resolved electron microscopy which afford near-atomic spatial resolution. The first consists of a modification of a commercial transmission electron microscope to generate bright and intense microsecond electron pulses, which are then used to image fast and irreversible nanoscale dynamics with atomic resolution (Chapters 2 and 3). The second is a novel approach to time-resolved cryo-electron microscopy which boosts the temporal resolution to the microsecond timescale (Chapters 4 and 5).Chapter 2 describes the irradiation of a Schottky emitter with microsecond laser pulses. The temperature of the filament rises to extreme values for brief periods of time, causing a significant increase in emission current. Even though the temperatures reached by the filament tip during laser irradiation are well beyond the maximum value recommended by the manufacturer, we show that the brief and localized heating provided by the focused laser pulse provides a way to extract large currents without damaging the filament.An electrostatic deflector, placed below the accelerator, chops the laser-boosted electron beam into microsecond pulses, as described in Chapter 3. We show that a 5 µs pulse generated with this method is brighter than the continuous electron beam and can be used to capture an atomic-resolution image of a gold nanoparticle in a single shot. Two possible applications of these pulses are then discussed. Drift-corrected imaging, especially in the presence of large amounts of drift, is significantly improved when bright electron pulses are used instead of the continuous beam. In addition, these pulses can be employed to capture irreversible dynamics occurring on the microsecond timescale with atomic spatial resolution. Chapter 4 provides details of a novel method for microsecond time-resolved cryo-electron microscopy. The high temporal resolution is achieved by irradiating a cryo specimen with a laser beam, causing it to locally melt. The embedded biomolecules can undergo dynamics in liquid until the laser is switched off, at which point the sample revitrifies within a few microseconds trapping particles in their intermediate configurations. The chapter shows that it is possible to obtain a near-atomic resolution reconstruction from revitrified sample areas, and the result looks identical to a map obtained from conventional sample areas. In addition, the projection angles are more uniformly distributed after revitrification.Chapter 5 shows that it is not necessary to modify a transmission electron microscope to perform melting and revitrification experiments on a cryo sample. The chapter introduces a simplified setup, requiring an optical microscope, that allows performing such experiments and verify their outcome on the fly. We present the advantages and disadvantages of this new setup, in the hope it will encourage the adoption of our method by other research groups and boost the development of microsecond time-resolved cryo-electron microscopy.