Three-dimensional electron diffraction (3D ED) has emerged as a powerful tool for solving the structures of small crystals down to nanometre-scale sizes. Despite advancements in automating data acquisition for 3D ED, the subsequent data processing and structure solution have largely relied on human intervention and have been mostly conducted offline. This reliance on expertise in electron crystallography and the lack of real-time feedback on data quality and structural information have limited the broader adoption of 3D ED. Here, we introduce Instamatic-solve, a fully automated, real-time structure solution pipeline for 3D ED deployed on a JEOL JEM 2100 transmission electron microscope. Instamatic-solve streamlines the entire process by automating the subsequent data processing and structure solution, providing real-time assessments of data quality and structural information. Moreover, the pipeline can handle offline 3D ED data acquired from various transmission electron microscope platforms. Using Instamatic-solve, we have successfully solved the crystal structures of diverse materials, including seven inorganic zeolites, two inorganic–organic hybrids and four organic molecules (including pharmaceuticals), all within 2 min. Instamatic-solve mimics the typical manual structure solution process, and its outcomes depend heavily on data quality. Our results indicate that a routine and reliable structure solution is achievable in most cases, provided that the data meet critical quality criteria, namely completeness ≥50% and resolution better than 1.0 Å. By enabling efficient, automated and real-time structure solution for crystalline materials, Instamatic-solve spans various scientific disciplines.

The structure of novel large pore borosilicate zeolite EMM-59 (|C₁₉H₄₂N₂|₈[B_5.2Si_218.8O₄₄₈]) with localized framework boron sites was determined by using three-dimensional electron diffraction (3D ED) and scanning transmission electron microscopy (STEM) imaging. EMM-59 was synthesized using 2,2-(cyclopentane-1,1-diyl)bis(N,N-diethyl-N-methylethan-1-aminium) as an organic structure-directing agent (OSDA). The framework has a three-dimensional intersecting channel system delimited by 12 × 10 × 10-ring openings and contains 28 T and 60 oxygen atoms in the asymmetric unit, making it the most complex monoclinic zeolite. The 3D ED data collected from as-made EMM-59 under cryogenic conditions revealed three symmetry-independent locations of the OSDAs, and STEM imaging showed that the OSDAs are flexible and adopt different molecular conformations in channels with identical structural environments. The framework boron atoms were exclusively found in T-sites of 4-rings, especially those shared by multiple 4-rings. The tetrahedral BO₄ with the highest boron content (38.6%) was transformed into a trigonal BO₃ after the OSDAs were removed upon calcination. Its location and boron content could also be identified by STEM imaging.

Rapid phase elucidation of polycrystalline materials is essential for developing new materials of chemical, pharmaceutical and industrial interest. Yet, the size and quantity of many crystalline phases are too small for routine X-ray diffraction analysis. This has become a workflow bottleneck in materials development, especially in high-throughput synthesis screening. Here we demonstrate the application of serial rotation electron diffraction (SerialRED) for high-throughput phase identification of complex polycrystalline zeolite products. The products were prepared from a combination of multiple framework T atoms ([Si,Ge,Al] or [Si,Ge,B]) and a simple organic structure-directing agent. We show that using SerialRED, five zeolite phases can be identified from a highly complex mixture. This includes phases with ultra-low contents undetectable using X-ray diffraction and phases with identical crystal morphology and similar unit cell parameters. By automatically and rapidly examining hundreds of crystals, SerialRED enables high-throughput phase analysis and allows the exploration of complex synthesis systems. It provides new opportunities for rapid development of polycrystalline materials.

Zeolites with large cavities that are accessible via wide pore windows are desirable but very rare. They have been dominantly used as catalysts in industry. Here we report a novel porous germanosilicate SCM-25, the zeolite structure containing ordered meso-cavities (29.9 × 7.6 × 6.0 ų) interconnected by 10- and 12-ring channels. SCM-25 was synthesized as nanosized crystals by using a simple organic structure-directing agent (OSDA). Three-dimensional (3D) electron diffraction shows that SCM-25 crystallizes in the orthorhombic space group Cmmm with a = 14.62 Å, b = 51.82 Å, c = 13.11 Å, which is one of the zeolites with the largest unit cell dimensions. We demonstrate that 3D electron diffraction is a powerful technique for determining the complex structure of SCM-25, including the disorders and distributions of framework atoms silicon and germanium. SCM-25 has a high surface area (510 m²/g) and high thermal stability (700 °C). Furthermore, we propose a potential postsynthetic strategy for the preparation of zeolites with ordered meso-cavities by applying the ADOR (assembly–disassembly–organization–reassembly) approach.

Zeolites have unique pore structures of molecular dimensions and tunable compositions, making them ideal for shape selective catalysis and separation. However, targeted synthesis of zeolites with new pore structures and compositions remains a key challenge. Here, we propose an approach based on a unique 3D-3D topotactic transformation, which takes advantage of weak bonding in zeolites. This is inspired by the structure transformation of PST-5, a new aluminophosphate molecular sieve, to PST-6 by calcination. The structure of nano-sized PST-5 crystals is determined by 3D electron diffraction. We find that the 3D-3D topotactic transformation involves two types of building units where penta- or hexa-coordinated Al is present. We apply this approach to several other zeolite systems and predict a series of new zeolite structures that would be synthetically feasible. This method provides a concept for the synthesis of targeted zeolites, especially those which may not be feasible by conventional methods.

A DigitalMicrograph script, InsteaDMatic, has been developed to facilitate rapid automated 3D electron diffraction/microcrystal electron diffraction data acquisition by continuous rotation of a crystal with a constant speed, denoted as continuous rotation electron diffraction. The script coordinates microscope functions, such as stage rotation, and camera functions relevant for data collection, and stores the experiment metadata. The script is compatible with any microscope that can be controlled by DigitalMicrograph and has been tested on both JEOL and Thermo Fisher Scientific microscopes. A proof of concept has been performed through employing InsteaDMatic for data collection and structure determination of a ZSM-5 zeolite. The influence of illumination settings and electron dose rate on the quality of diffraction data, unit-cell determination and structure solution has been investigated in order to optimize the data acquisition procedure.

Serial rotation electron diffraction (SerialRED) has been developed as a fully automated technique for three-dimensional electron diffraction data collection that can run autonomously without human intervention. It builds on the previously established serial electron diffraction technique, in which submicrometre-sized crystals are detected using image processing algorithms. Continuous rotation electron diffraction (cRED) data are collected on each crystal while dynamically tracking the movement of the crystal during rotation using defocused diffraction patterns and applying a set of deflector changes. A typical data collection screens up to 500 crystals per hour, and cRED data are collected from suitable crystals. A data processing pipeline is developed to process the SerialRED data sets. Hierarchical cluster analysis is implemented to group and identify the different phases present in the sample and to find the best matching data sets to be merged for subsequent structure analysis. This method has been successfully applied to a series of zeolites and a beam-sensitive metal-organic framework sample to study its capability for structure determination and refinement. Two multi-phase samples were tested to show that the individual crystal phases can be identified and their structures determined. The results show that refined structures obtained using automatically collected SerialRED data are indistinguishable from those collected manually using the cRED technique. At the same time, SerialRED has lower requirements of expertise in transmission electron microscopy and is less labor intensive, making it a promising high-throughput crystal screening and structure analysis tool.

Over the past decade, electron diffraction methods have aroused more and more interest for micro-crystal structure determination. Compared to traditional X-ray diffraction, electron diffraction breaks the size limitation of the crystals studied, but at the same time it also suffers from much stronger dynamical effects. While X-ray crystallography has been almost thoroughly developed, electron crystallography is still under active development. To be able to perform electron diffraction experiments, adequate skills for using a TEM are usually required, which makes ED experiments less accessible to average users than X-ray diffraction. Moreover, the relatively poor data statistics from ED data prevented electron crystallography from being widely accepted in the crystallography community.

The thesis focused on both application and method development of continuous rotation electron diffraction (cRED) technique. The cRED method was first applied to a beam sensitive metal-organic framework sample, Co-CAU-36, and the structure was determined and refined within one working day. More importantly, the guest molecules in the pores were also located using only electron diffraction data. To facilitate general users to perform cRED data collection for useful data, software was developed to automate the overall data collection procedure. Through combination of hierarchical cluster analysis tools, the automatically collected data showed comparable quality to those from recent publications, and thus were useful for structure determination and even phase identification. To deal with dynamical refinement for ED data, a frame orientation refinement algorithm was designed to calculate accurate frame orientations for rotation data. Accuracy for the method was validated and compared to an existing software, and the behavior of TEM goniometer was studied by applying the method to an experimental data set.

Single-crystal electron diffraction has shown to be powerful for structure determination of nano- and submicron-sized crystals that are too small to be studied by single-crystal X-ray diffraction. However, it has been very challenging to obtain high quality electron diffraction data from beam sensitive crystals such as metal-organic frameworks (MOFs). It is even more difficult to locate guest species in the pores of MOF crystals. Here, we present synthesis of a novel porous cobalt MOF with 1D channels, [Co-2(Ni-H4TPPP)]center dot 2DABCO center dot 6H(2)O, (denoted Co-CAU-36; DABCO=1,4-diazabicyclo[2.2.2]octane), and its structure determination using continuous rotation electron diffraction (cRED) data. By combining a fast hybrid electron detector with low sample temperature (96 K), high resolution (0.83-1.00 angstrom) cRED data could be obtained from eight Co-CAU-36 crystals. Independent structure determinations were conducted using each of the eight cRED datasets. We show that all atoms in the MOF framework could be located. More importantly, we demonstrate for the first time that organic molecules in the pores, which were previously difficult to find, could be located using the cRED data. A comparison of eight independent structure determinations using different datasets shows that structural models differ only on average by 0.03(2) angstrom for the framework atoms and 0.10(6) and 0.16(12) angstrom for DABCO and water molecules, respectively.

Single-crystal electron diffraction (SCED) is emerging as an effective technique to determine and refine the structures of unknown nano-sized crystals. In this work, the implementation of the continuous rotation electron diffraction (cRED) method for high-throughput data collection is described. This is achieved through dedicated software that controls the transmission electron microscope and the camera. Crystal tracking can be performed by defocusing every nth diffraction pattern while the crystal rotates, which addresses the problem of the crystal moving out of view of the selected area aperture during rotation. This has greatly increased the number of successful experiments with larger rotation ranges and turned cRED data collection into a high-throughput method. The experimental parameters are logged, and input files for data processing software are written automatically. This reduces the risk of human error, and makes data collection more reproducible and accessible for novice and irregular users. In addition, it is demonstrated how data from the recently developed serial electron diffraction technique can be used to supplement the cRED data collection by automatic screening for suitable crystals using a deep convolutional neural network that can identify promising crystals through the corresponding diffraction data. The screening routine and cRED data collection are demonstrated using a sample of the zeolite mordenite, and the quality of the cRED data is assessed on the basis of the refined crystal structure.