Small Crystals, Big Insights: How Electron Diffraction is Transforming Materials, Life Science, and Chemistry Research

X-ray diffraction relies on the elastic scattering of X-rays from the electron density surrounding atoms within a sample. Electron diffraction, however, involves the elastic scattering of electrons deflected by the Coulomb potential within the specimen. This means electrons are sensitive to both the electron shells and the nucleus, positioning their interaction sensitivity between that of X-rays and neutrons. Crucially, the interaction cross-section for electrons is significantly larger (up to five orders of magnitude) than for X-rays, allowing useful diffraction signals to be obtained from sub-micron crystals.

Due to the strong interaction of electrons with matter, electron diffraction can obtain a useful signal from crystals less than a micron in size, which are too small for conventional single-crystal X-ray diffraction. For powder X-ray diffraction, while it works with small crystals, it only provides a one-dimensional spectrum, making structural determination from complex mixtures or multi-phase samples very difficult. Electron diffraction acts as a single-crystal technique on individual grains within a powder sample, allowing for the separate measurement and characterization of different phases present in a mixture. This "powder diffraction grain by grain" approach makes multi-phasing and analyzing impure or messy samples much more tractable.

Inelastic scattering processes, which deposit energy into a sample and cause damage, are detrimental to diffraction experiments. For X-rays, the ratio of inelastic to elastic collisions is quite unfavorable, leading to significant sample damage per useful elastic scattering event. While electrons also cause damage, the ratio is up to three orders of magnitude better than for X-rays. This higher tolerance for damage in electron diffraction makes it compatible with obtaining good diffraction data from very small crystals where a higher radiation dose per unit volume is necessary to extract a sufficient signal.

Modern electron diffraction is referred to as 3D ED or micro ED largely because it developed somewhat in parallel within different scientific communities (crystallography and electron microscopy). The key paradigm shift, which occurred around 20 years ago, was the realization that collecting electron diffraction data by continuously rotating a crystal in the beam, similar to X-ray crystallography, and staying off specific zone axes, yielded data that was much more "kinematical." This means the intensities of the diffraction spots were more closely proportional to the square of the structure factors, allowing standard crystallographic methods to be applied for structure solution, which was previously thought impossible due to strong dynamical diffraction effects at zone axes.

Several practical challenges exist for electron diffraction. Nano-sized crystals are often invisible to light microscopy, requiring screening and searching for suitable crystals using electrons within the diffractometer itself. Electrons are immediately absorbed by air, necessitating a vacuum environment and sample transfer systems, as well as protection for vacuum-sensitive samples. High radiation doses per unit volume are inherent, requiring sensitive detectors and careful dose management in the control system. Finally, because samples are often loaded as powders with thousands of grains, automation and big data processing are crucial for efficiently analyzing multicomponent or impure samples.

Data collection methods for electron diffraction have evolved significantly. Initially, data was collected in a stepwise tomography approach. Later, beam precession was added to improve data quality by better integrating over diffraction spots. The latest advancement is continuous rotation schemes, where the crystal rotates continuously and data is collected in a shutterless manner, similar to modern X-ray diffractometers. The current trend focuses heavily on automation, allowing instruments to automatically find and collect data from numerous crystal grains, which is particularly beneficial for analyzing mixed or multi-component samples.

Unlike X-ray diffraction where anomalous dispersion provides information about absolute structure, electron diffraction leverages dynamical diffraction effects. These effects cause the intensities of Friedel pairs (reflections from opposite sides of the reciprocal lattice) to be unequal, even without anomalous dispersion. By explicitly modeling these intensities using a full dynamical theory during refinement, it is possible to distinguish between enantiomers and determine the absolute structure. Comparing the refinement metrics (like R values) for different possible absolute structures allows identification of the correct one, often with high confidence quantified by a z-score.

Electron diffraction is expanding beyond solving structures of pure, well-ordered crystals. It is increasingly being applied to understand defects and disorder in materials by analyzing the total scattering signal. There is also growing interest in using electron diffraction to study fully or partially amorphous samples by calculating pair distribution functions (PDFs). The ability to perform these analyses on individual grains without rotational averaging, leading to 3D delta PDFs that quantify short-range order along crystallographic directions, is a promising development in understanding disordered materials like battery components. The ability to conduct experiments in different environments (cryo, gas, liquid) also opens up new avenues for studying materials under operando or in-situ conditions.