XFEL serial femtosecond crystallography (SFX) has become a useful technique for structure determination that bypasses the necessity for large single crystals. In an SFX experiment, thousands of single microcrystals in random orientations are delivered to a pulsed X-ray beam at room temperature to collect a complete dataset. SFX has been successful at solving important macromolecular structures at high resolution and has recently been used successfully to solve structures of network solids and small molecule microcrystals. Cryo-EM has achieved atomic resolution 3D imaging of single macromolecules, but exclusively under cryogenic conditions. High intensity, ultrafast XFEL pulses can be used for single particle imaging (SPI) at room temperature while outrunning radiation damage. SPI has produced 3D reconstructions of single cells, cell organelles, and viruses at up to 10 nanometer resolution, but atomic resolution imaging of single molecules or clusters has yet to be achieved. Weak scattering produces data with low signal to noise (SNR), but this can be ameliorated with specialized sample delivery. Optimizing experimental data collection rates is also challenging. Thousands and thousands of randomly oriented particles must be imaged to reconstruct a 3D structure at atomic resolution. Currently, the aerosol injectors used in SPI experiments have low sample delivery efficiency and data collection rates remain low.
SFX and SPI both stand to benefit greatly from Megahertz high-repetition rate XFEL sources, where complete datasets could be obtained in minutes and adequate multiplicity can be obtained to improve 3D reconstruction resolution. We cannot make full use of these facilities or these methods if we cannot collect data efficiently. Here, we present the current state of SFX and SPI at XFELs and present methods for improving data collection efficiency to prepare for next generation XFEL capabilities and experiments.
