Introduction to 3D Electron Diffraction

Three-dimensional electron diffraction (3D ED) has emerged as a transformative technique for determining the atomic structures of nanocrystalline materials. Unlike traditional X-ray diffraction, which requires crystals larger than a few micrometers, 3D ED can extract high-quality structural data from crystals as small as a few tens of nanometers. The method involves scanning a focused electron beam over a rotating nanocrystal, collecting a series of two-dimensional diffraction patterns, and then computationally reconstructing the three-dimensional reciprocal space to solve the atomic arrangement. Recent progress in instrumentation, detector sensitivity, and data processing algorithms has dramatically expanded the applicability and reliability of 3D ED, making it a cornerstone technique in modern materials characterization.

The ability to determine structures from sub-micrometer crystals addresses a long-standing gap in crystallography. Many technologically important materials—such as metal-organic frameworks, zeolites, pharmaceutical polymorphs, and battery electrode materials—often grow only as small nanocrystals. Before the maturation of 3D ED, these systems were either intractable or required indirect methods. Today, the technique routinely provides atomic-resolution models for materials that were previously considered "uncrystallizable." This article reviews the latest technological advances driving 3D ED forward and explores key applications across materials science, chemistry, and biology.

Recent Technological Developments

Detector Technology

The most significant driver of progress in 3D ED has been the development of fast, low-noise, direct-electron detectors. Hybrid pixel array detectors (HPADs) such as the Timepix and Medipix families are now widely used for continuous rotation microED experiments. These detectors offer high dynamic range, single-electron sensitivity, and frame rates exceeding 1000 Hz. The ability to collect diffraction data with minimal dead time reduces electron dose on the sample, mitigating radiation damage—a critical advantage for organic and biological nanocrystals. Furthermore, improved detective quantum efficiency at higher electron energies enables researchers to work at accelerating voltages of 200–300 kV, which balances penetration depth with scattering contrast.

Complementary to HPADs, complementary metal-oxide-semiconductor (CMOS) detectors with fiber-optic coupling are also gaining traction. These detectors offer larger active areas and lower noise at lower frame rates, making them suitable for the static-beam step-rotation 3D ED methods. The combination of fast framing and low noise allows researchers to capture very weak reflections that were previously lost in detector background, directly improving the completeness and resolution of the final structure.

Data Acquisition and Processing Algorithms

Software developments have kept pace with hardware improvements. Algorithms for indexing diffraction patterns, merging multi-crystal data, and handling dynamical scattering effects have matured significantly. Modern routines such as those implemented in DIALS, PETRUS, and RED suite can automatically index hundreds of patterns per crystal and merge data from several crystals to produce a complete dataset. Multi-crystal merging is particularly important for small crystals that may not completely cover all reciprocal space in a single rotation. Advanced outlier rejection algorithms and scaling protocols now produce merged datasets with R-factors comparable to those from single-crystal X-ray diffraction.

Dynamical scattering—a major complication in electron diffraction—is increasingly addressed through multislice simulations and refinement algorithms. Software packages like PetroDynamics and eMAP allow refinement of structural parameters against dynamical diffraction data, often yielding accurate atomic positions even for medium-weight elements. These algorithms reduce the systematic errors that historically prevented ED from achieving the same accuracy as X-ray diffraction for precise bond length determinations.

Instrumentation and Automation

Automation of data collection is transforming 3D ED from a specialist technique into a routine tool. Commercial transmission electron microscopes equipped with automated sample stages, cryogenic holders for cryo-EM, and computer-controlled beam tilt now allow unattended operation overnight. Platforms such as the JEOL JEM-F200 and Thermo Fisher Scientific Glacios are commonly used for microED, with software scripts that handle crystal centering, rotation, and data saving. The ability to collect data from tens or even hundreds of crystals per day dramatically increases throughput, making 3D ED viable for high-throughput screening of polymorphs and for combinatorial materials discovery.

Additionally, the integration of electron energy loss spectroscopy (EELS) and energy-dispersive X-ray spectroscopy (EDS) into the same workflow allows simultaneous structural and chemical characterization. This multimodal approach is extremely powerful for understanding the relationship between composition and crystal structure in heterogeneous nanomaterials.

Applications in Materials Science and Chemistry

Nanocrystalline Inorganic Materials

3D ED has been instrumental in solving the structures of inorganic nanocrystals that are crucial for catalysis, energy storage, and electronics. For example, the structures of zeolites, which are microporous aluminosilicates used in petroleum refining and gas separation, have been solved from crystals smaller than 50 nm—far below the size needed for single-crystal X-ray diffraction. Recent studies have revealed new zeolite topologies and silicon/aluminum distributions that were inaccessible to powder diffraction. Similarly, metal-organic frameworks (MOFs), which often form as sub-micrometer particles, have been characterized using 3D ED to understand pore geometry and guest molecule positions, enabling rational design of next-generation sorbents and catalysts.

Another area of rapid progress is the structural characterization of battery electrode materials. Lithium metal oxides, such as the layered oxides used in lithium-ion batteries, undergo complex phase transformations during cycling. 3D ED has allowed researchers to follow these phase transitions in operando or at least ex situ, identifying intermediate structures that may be responsible for capacity fade. The ability to obtain atomic-resolution models from nanocrystals extracted from cycled electrodes provides direct insight into degradation mechanisms, guiding the development of more durable materials.

Pharmaceuticals and Small-Molecule Crystallography

The pharmaceutical industry has embraced 3D ED for polymorph screening and structural characterization of lead compounds. Most active pharmaceutical ingredients (APIs) can form multiple polymorphs, each with potentially different solubility, stability, and bioavailability. Traditional X-ray methods require large single crystals for structure determination, which may not exist for all polymorphs. 3D ED can provide structure solutions from the same nanocrystalline powders that appear in formulation development, eliminating the need for special crystallization protocols.

Moreover, the cryogenic environment used in many microED experiments (cryo-ED) dramatically reduces radiation damage for organic molecules. This has enabled structure determination of many pharmaceutical compounds that degrade rapidly under electron illumination at room temperature. Recent reports have shown that cryo-ED can provide nearatomic resolution (routinely better than 1.5 Å) for molecules as small as 100 Da, with high accuracy for bond angles and torsions. The technique is now considered complementary to X-ray and neutron diffraction for drug discovery workflows.

Biological Structures and Complex Nanomaterials

While single-particle cryo-electron microscopy (cryo-EM) excels at large macromolecular complexes, 3D ED (often called microcrystal electron diffraction—MicroED—in the cryo-EM community) is uniquely suited for smaller biological molecules and nanocrystalline arrays. Membrane proteins, which are notoriously difficult to crystallize in large volumes, can often form only tiny microcrystals. MicroED has been used to solve high-resolution structures of several membrane proteins, including ion channels and transporters, from crystals less than 500 nm in size. The method couples the cryogenic protection of the specimen with the single-crystal geometry, giving access to multiple structural states without the need for large expression and purification amounts.

Natural products and metabolites also benefit from MicroED. Many biologically active small molecules from plants and microorganisms are isolated in low yields and cannot be crystallized to adequate size for X-ray diffraction. MicroED has provided the absolute configuration of several complex natural products, such as the neurotoxin saxitoxin, with confidence levels equivalent to those from X-ray analysis.

Comparison with Traditional Techniques

To appreciate the role of 3D ED, it is helpful to place it alongside established methods. Single-crystal X-ray diffraction remains the gold standard for structural determination when sufficiently large crystals are available (typically >10 μm). It provides the highest resolution (<0.5 Å) and lowest uncertainty in atomic positions. However, for many modern materials, such large crystals simply do not exist. Powder X-ray diffraction can handle small crystals but suffers from peak overlap and loses information about symmetry and unit cell content, especially for complex organic molecules.

Electron microscopy techniques, including high-resolution TEM (HRTEM) and scanning transmission electron microscopy (STEM), can image individual atoms but are limited to very thin specimens and often involve significant beam damage. 3D ED bridges these gaps: it works with crystals in the 50–500 nm range, provides atomic-resolution structure solutions, and can be combined with cryogenic preservation to handle beamsensitive samples. The principal limitation is the need to handle dynamical scattering effects, especially for thick crystals and for light elements, where scattering is strong and multiple scattering is unavoidable. Nevertheless, the ongoing development of multislice refinement and precession electron diffraction continues to reduce these errors.

Limitations and Challenges

Despite its successes, 3D ED still faces several challenges. The dynamical scattering problem, while addressed by software, remains a source of systematic error for very thick crystals or very low symmetry space groups. Precessing the beam can reduce but not eliminate multiple scattering. Additionally, the size of the crystal—at the lower end, below about 20 nm—leads to broad diffraction spots and insufficient signal-to-noise for structure resolution. Radiation damage, even under cryogenic conditions, limits the experimental lifetime of many organic materials, requiring data collection to be completed rapidly. This imposes constraints on rotation speed and number or frames collected.

Crystallographers also note that data completeness is often lower than in X-ray diffraction due to the limited tilt range of the sample stage (typically ±70° or less). The missing wedge of data can be filled by merging multiple crystals, but that introduces potential errors from crystal variability. Furthermore, the technique currently requires specialized equipment (a TEM with proper electron optics and detectors) and expert operators, although automation is rapidly democratizing access. Combined, these challenges point to areas where further technological development is most needed.

Future Perspectives

The trajectory of 3D ED suggests several exciting directions. One is the integration of the technique with synchrotron-based X-ray methods, such as serial synchrotron crystallography. By combining the high throughput of X-ray sources with the nanocrystal capability of electron diffraction, a hybrid facility could screen polymorphs and then obtain high-resolution structures from the same batch of crystals. Another area is the development of in situ 3D ED, where the electron microscope is equipped with gas flow, heating, or electrochemical cells to observe structural changes under realistic conditions. Early experiments with gas phase carbon dioxide adsorption in MOFs already show the power of this approach.

On the software side, machine learning is beginning to influence data processing. Neural networks can be trained to predict dynamical scattering patterns and to deconvolute multiple crystal contributions, potentially allowing structure solution from extremely small crystals with high defect concentrations. Furthermore, improved algorithms for ab initio phase determination (direct methods and charge flipping) adapted for electron diffraction will reduce reliance on known structural fragments, opening up the technique to fully unknown materials.

Finally, the spread of user-friendly instrumentation—such as automated single-crystal electron diffractometers now offered by several vendors—will make 3D ED a routine tool in academic and industrial laboratories. The International Union of Crystallography has recognized this trend by establishing a Commission on Electron Crystallography and by devoting special issues of its journal, Acta Crystallographica Section C, to microED structures. With these developments, 3D electron diffraction is poised to become a primary method for nanocrystal structure determination in the next decade, complementing and extending the reach of traditional X-ray crystallography.

In conclusion, recent progress in detector technology, data processing, and automation has elevated 3D ED from a niche technique to a mainstream tool for structural chemistry and materials science. The ability to solve atomic-level structures from nanocrystals no longer than the wavelength of visible light opens up entire classes of materials that were previously inaccessible. As the methodology continues to mature, we can expect to see an accelerating pace of discovery in areas ranging from new catalysts to pharmaceutical formulations to biological signaling molecules. For researchers who work with small crystals, 3D electron diffraction is not just a promising development—it is a necessary evolution in the art and science of crystallography.