In situ crystallography has become an essential method for monitoring structural evolution during materials synthesis, enabling researchers to observe atomic-scale changes in real time rather than after the fact. By capturing diffraction data while a sample undergoes chemical or physical transformations, this technique provides direct insight into reaction pathways, intermediate phases, and the mechanisms that govern material formation. The ability to track these dynamics is critical for designing advanced materials with tailored properties, from catalysts to battery electrodes to pharmaceuticals.

Fundamentals of Crystallography and Diffraction

To appreciate in situ crystallography, it is helpful to review the basic principles of diffraction. When a crystalline material is exposed to X-rays, neutrons, or electrons, the periodic arrangement of atoms causes the incident beam to scatter constructively at specific angles, producing a diffraction pattern. The positions and intensities of these diffraction peaks encode information about the unit cell dimensions, atomic positions, and long-range order of the material. Traditional crystallography analyzes this pattern from a static, fully formed crystal. In situ crystallography extends this approach by collecting a series of diffraction patterns over time while the sample is actively changing – for instance, during heating, gas exposure, electrochemical cycling, or solution-phase growth.

The Challenge of Real-Time Observation During Synthesis

Monitoring structural changes during synthesis poses several technical challenges. Synthesis conditions often involve elevated temperatures, varying pressures, reactive atmospheres, or liquid environments that are incompatible with standard crystallographic setups. Moreover, the transformations of interest – such as nucleation, phase transitions, or crystal growth – can occur on timescales from milliseconds to hours. Capturing these transient states requires a combination of intense radiation sources, fast detectors, and specialized sample environments that can maintain controlled conditions while allowing unobstructed diffraction measurements. Early efforts in this field were limited by radiation sources that were too weak to yield high-quality patterns in short exposures and by detectors that were too slow to follow rapid dynamics. Over the past two decades, advances in synchrotron light sources, detector technology, and environmental cells have largely overcome these obstacles, making in situ crystallography a routine tool in many laboratories.

Key Technological Advances Driving In Situ Crystallography

Synchrotron Radiation Sources

The advent of third-generation synchrotron facilities provided a dramatic increase in X-ray brightness – several orders of magnitude greater than conventional laboratory sources. This high flux allows researchers to collect diffraction patterns in seconds or milliseconds, capturing intermediate states that would be unobservable with weaker beams. Modern synchrotrons, such as the European Synchrotron Radiation Facility (ESRF) and the Advanced Photon Source (APS), also offer tunable wavelengths, enabling anomalous diffraction studies that can probe specific elements within a complex mixture. The development of fourth-generation sources, based on multi-bend achromat lattices, has further increased coherence and brightness, opening the door to coherent diffraction imaging and ptychography in situ. These facilities are now complemented by compact synchrotron-like sources based on inverse Compton scattering, which bring synchrotron-quality beams to smaller laboratories.

Ultra-Fast and Low-Noise Detectors

Detector technology has advanced in parallel with synchrotron sources. Hybrid photon-counting detectors (e.g., Pilatus, Eiger) and complementary metal-oxide-semiconductor (CMOS) detectors can now read out images at kilohertz rates with virtually no noise. These fast detectors enable time-resolved studies where diffraction patterns are acquired at frames per second, matching the speed of many chemical reactions. For example, the ability to collect a full diffraction pattern every millisecond has allowed researchers to track the formation of metastable phases during the crystallization of zeolites and metal-organic frameworks. In addition, the wide dynamic range of modern detectors allows the simultaneous measurement of weak and strong reflections, improving the accuracy of structural models derived from in situ data.

Advanced Sample Environments

To simulate realistic synthesis conditions, researchers have developed a variety of specialized sample chambers. For solid-state reactions, micro-furnaces that heat samples to over 1000°C with precise control are now compact enough to mount on a goniometer. For gas-solid reactions, capillary or flat-plate reactors allow controlled flow of reactive gases while maintaining good diffraction geometry. For solution-based syntheses – such as the nucleation of nanoparticles or the crystallization of proteins – microfluidic chips have been designed that mix reagents on-chip and deliver them to the X-ray beam within milliseconds. These chips combine optical transparency for simultaneous microscopy with thin, low-scattering windows for diffraction. Electrochemical cells for batteries enable operando measurements during charge and discharge, with the sample environment mimicking real device conditions. Each of these setups requires careful engineering to balance the conflicting demands of sample control, beam access, and signal-to-noise ratio.

Computational Methods for Real-Time Analysis

The data volumes produced by in situ experiments are enormous – a single experiment can generate thousands of diffraction patterns. Extracting structural information in real time requires robust algorithms for peak finding, indexing, and profile fitting. Parametric Rietveld refinement, which models the evolution of structural parameters as a function of time or temperature, has become a standard approach. More recently, methods based on machine learning have been introduced to automatically identify phase transitions and extract unit cell parameters from large datasets without human supervision. Convolutional neural networks can classify diffraction patterns into known phases, while autoencoders can detect anomalous patterns that may correspond to new or unexpected phases. These computational tools are essential for turning raw diffraction data into actionable insights during the course of an experiment, and they enable adaptive control of synthesis conditions – for example, stopping a reaction at the point when a target phase appears.

Applications Across Materials Science

Catalyst Development

In catalysis, the nature of the active site often changes under reaction conditions. In situ crystallography has been used to monitor the evolution of metal nanoparticles during oxidation and reduction cycles, revealing the formation of oxide shells, core-shell structures, and sintering processes. For example, researchers studying palladium catalysts for methane combustion observed the reversible formation of a surface oxide layer using in situ X-ray diffraction and pair distribution function analysis. Similar studies on zeolite catalysts have tracked the migration of cations and the formation of Bronsted acid sites during activation. These observations allow scientists to correlate structural dynamics with catalytic performance, guiding the rational design of more active and stable catalysts.

Nanomaterials and Nucleation

The nucleation and growth of nanoparticles occur on timescales that are difficult to capture with conventional methods. In situ synchrotron-based X-ray scattering, combined with small-angle X-ray scattering (SAXS) and wide-angle X-ray scattering (WAXS), has been used to follow the formation of quantum dots, metal nanoparticles, and oxide nanostructures from precursor solutions. A landmark study on the crystallization of iron oxide nanoparticles showed that the process involves the formation of a transient amorphous intermediate before the appearance of the crystalline phase. Such insights challenge classical nucleation theory and point to the importance of non-classical pathways involving prenucleation clusters or dense liquid phases. In situ crystallography is now being applied to understand the role of additives and capping agents in directing nanoparticle morphology.

Battery Materials

Operando studies of battery electrodes are among the most active areas of in situ crystallography. During charge and discharge, intercalation, conversion, and alloying reactions cause significant structural rearrangements that affect capacity and cycle life. For lithium-ion batteries, in situ XRD has been used to track the phase transitions in cathode materials such as LiFePO₄ and NMC (nickel-manganese-cobalt oxides), revealing the formation of solid-solution domains and the influence of particle size on phase boundary movement. For beyond-lithium chemistries like sodium-ion and magnesium batteries, in situ crystallography is essential for understanding the stability of host structures against repeated ion insertion. Recent developments include combining diffraction with electrochemical impedance spectroscopy and imaging to correlate structural changes at the particle level with overall device performance.

Pharmaceuticals and Polymorphism

Many active pharmaceutical ingredients (APIs) can exist in multiple crystalline forms, or polymorphs, each with different solubility, stability, and bioavailability. Controlling which polymorph forms during manufacturing is a critical quality attribute. In situ crystallography has been employed to monitor the solvent-mediated transformation between polymorphs, as well as the conversion of amorphous drug formulations to crystalline forms during storage. For example, in situ XRD was used to study the transformation of the metastable form II to the stable form I of a popular antifungal drug. By understanding the kinetics of these transformations under different humidity and temperature conditions, formulators can design processes that consistently yield the desired polymorph. The technique is also used to study the formation of cocrystals and salts, which are increasingly used to improve drug properties.

Emerging In Situ Crystallography Techniques

Time-Resolved Serial Crystallography

At X-ray free-electron lasers (XFELs), extremely short pulses (femtoseconds) can outrun radiation damage and capture atomic motions in real time. Serial crystallography, where a stream of microcrystals is delivered to the X-ray beam, enables pump-probe experiments that observe structural changes triggered by light or chemical stimuli. While primarily used for protein dynamics, this technique is now being adapted for materials science, for instance, to visualize the photoinduced phase transitions in organic semiconductors. The challenge lies in preparing a stable stream of microcrystals and synchronizing the pump pulse with the X-ray probe, but ongoing advances in sample delivery and data analysis are making these experiments more routine.

Microfluidic and Droplet-Based Platforms

Microfluidic systems offer precise control over mixing, temperature, and reaction time, making them ideal for in situ studies of crystallization kinetics. Droplet-based platforms encapsulate nanoliter volumes of solution, allowing each crystal to grow in isolation while the entire population is monitored by diffraction. This approach has been used to study the nucleation of calcite and the formation of calcium phosphate clusters relevant to biomineralization. Coupled with on-chip optical microscopy, these platforms provide complementary information about morphology and solution conditions. Future developments aim to integrate multiple analytical probes – such as Raman spectroscopy and UV-vis – into the same chip for a more complete picture of the synthesis process.

Combined In Situ Techniques

No single technique can capture all aspects of a material’s evolution during synthesis. Increasingly, researchers combine in situ crystallography with other characterization methods. For example, simultaneous XRD and PDF (pair distribution function) analysis provides both long-range order and local structural information. Combining XRD with X-ray absorption spectroscopy (XAS) gives insight into the electronic state of specific elements during a reaction. Raman spectroscopy and XRD can be performed on the same sample to link vibrational modes with crystallographic changes. These multi-modal approaches are particularly powerful for complex systems such as heterogeneous catalysts, where the active phase may be a minority component or present only transiently.

Integration with Machine Learning and Artificial Intelligence

The flood of data from in situ experiments has accelerated the adoption of machine learning for automated analysis. Deep learning models can now predict crystal structures from powder diffraction patterns with high accuracy, even for noisy or limited data. For in situ studies, recurrent neural networks (RNNs) and long short-term memory (LSTM) networks have been used to forecast the appearance of new phases based on earlier diffraction patterns, enabling real-time decision-making. Another promising direction is the use of reinforcement learning to autonomously optimize synthesis parameters – for instance, adjusting temperature ramps to maximize the yield of a desired phase. These AI-driven approaches promise to close the loop between characterization and synthesis, turning in situ crystallography from a passive monitoring tool into an active control system for materials discovery.

Future Directions and Remaining Challenges

Despite impressive progress, several challenges remain. Achieving spatial resolution at the nanoscale is still difficult: most in situ experiments probe the average behavior of a large ensemble of crystallites, masking heterogeneity. Techniques such as scanning nanobeam diffraction and Bragg coherent diffractive imaging are pushing toward sub-100 nm resolution, but they require extremely brilliant and coherent X-ray beams, which are only available at a few facilities worldwide. Temporal resolution also needs to improve for the fastest processes, such as chemical reactions on microsecond timescales. XFELs can achieve femtosecond resolution, but they are not suitable for long-duration experiments, and the data analysis for serial crystallography is demanding. Sample environments must become more versatile, allowing simultaneous control of temperature, pressure, gas, and liquid with minimal background scattering. Finally, handling the massive data volumes – a single experiment can generate hundreds of gigabytes – requires robust, scalable data management and analysis pipelines, ideally with real-time feedback.

Another important direction is the integration of in situ crystallography with multiscale modeling. By combining experimental observations with molecular dynamics simulations and density functional theory (DFT), researchers can build mechanistic models that explain not only what phases appear but why they appear. This synergy between experiment and computation is essential for moving from description to prediction in materials synthesis.

The growing availability of user-friendly in situ cells and automated beamlines at synchrotrons is democratizing the technique, allowing more research groups to incorporate in situ crystallography into their workflow. Commercial vendors now offer plug-and-play heating, cooling, and gas-flow stages that fit standard diffractometers, making in situ experiments feasible in home laboratories for certain applications.

Conclusion

In situ crystallography has matured into a powerful and versatile method for understanding structural changes during material synthesis. Driven by advances in synchrotron sources, fast detectors, environmental control, and machine learning, the technique now provides real-time atomic-scale insights that were unimaginable a generation ago. Its applications span catalysis, nanomaterials, battery research, pharmaceuticals, and beyond. As experimental capabilities continue to improve – toward higher spatial and temporal resolution, more seamless multi-modal integration, and AI-enabled autonomous experiments – in situ crystallography will play an increasingly central role in the rational design and engineering of new materials. The ability to watch materials form as they form is not only scientifically fascinating but also practically indispensable for accelerating the development of next-generation technologies.

For further reading on the current state of the technique, the following resources are recommended: