Ultrafast electron diffraction has taken a major step toward becoming faster, smaller and more precisely synchronized. In a new proof-of-principle experiment, researchers generated electron beams for diffraction using a laser wakefield accelerator, or LWFA, and transported them through a miniaturized beamline built with permanent magnets. The system produced MeV-energy electron bunches with a measured temporal resolution of approximately 49.8 femtoseconds, while preserving enough charge to record diffraction patterns from single-crystalline gold. The result demonstrates that an all-optical electron source can be integrated into a practical ultrafast electron diffraction platform, potentially opening a path toward observing structural changes in matter on timescales of only a few femtoseconds.
Ultrafast electron diffraction is widely used to watch atoms move after a material absorbs energy from a laser pulse. In a typical experiment, a short optical pulse initiates a change in the sample, while a synchronized electron bunch acts as a probe. As the electrons pass through the material, they scatter from its atomic lattice and form a diffraction pattern that reveals how the structure evolves. Shorter electron pulses provide sharper snapshots, but producing and transporting such pulses is technically difficult. Conventional electron sources can require large radio-frequency accelerators and complex synchronization systems. Laser wakefield accelerators offer a different approach: an intense laser pulse drives a plasma wave capable of accelerating electrons over a distance far shorter than conventional machines.
The LWFA process can generate electron bunches with intrinsically short durations because the particles are injected and accelerated inside a rapidly evolving plasma structure. However, these beams also present challenges that complicate their use in diffraction. The electrons typically possess a significant energy spread, meaning that particles in the same bunch travel at slightly different velocities and follow different paths through the beamline. Even at relativistic energies, this difference can cause the bunch to stretch in time as it propagates. Shot-to-shot fluctuations in the electron energy can also change the arrival time at the sample, degrading the synchronization between the electron probe and the laser pulse that initiates the structural dynamics.
To address these problems, the researchers developed a compact transport system based on a double-bend achromat. This arrangement uses two magnetic bends together with focusing elements to control the electron trajectory and energy-dependent timing effects. In the experiment, the LWFA beam first underwent temporal stretching during transport because of its energy spread. The optimized double-bend achromat then compressed the bunch and was configured to operate in an isochronous regime. In an isochronous beamline, electrons with different energies are arranged to arrive at nearly the same time, suppressing the timing changes that would otherwise be caused by fluctuations in the beam energy. This design allowed the system to exploit one of LWFA’s most important advantages: the natural synchronization between the driving laser and the electrons produced by it.
The beamline also incorporated energy filtering to select a narrower portion of the electron spectrum. After filtering, the electron energy spread was reduced to approximately 3 percent full-width at half-maximum. Energy selection usually comes at a cost because removing unwanted electrons reduces the total charge available for an experiment. In this case, however, the system retained about 11.9 femtocoulombs per bunch, a sufficient quantity for recording diffraction signals. Maintaining both a relatively narrow energy distribution and useful charge is essential for UED. A broad energy spread can blur diffraction features and complicate the interpretation of lattice spacing, while too little charge can make single-shot measurements impractical.
The researchers measured the final electron pulse using a laser-driven terahertz deflector positioned near the sample. This device acts as an ultrafast streak camera for electrons. A time-varying terahertz electromagnetic field deflects electrons according to when they pass through it, converting their arrival time into a measurable spatial displacement on a detector. The measurement indicated an electron bunch duration of approximately 49.6 femtoseconds root mean square. The arrival-time jitter was approximately 4.7 femtoseconds root mean square, showing that the isochronous transport system effectively suppressed timing fluctuations linked to shot-to-shot energy changes. Combining the bunch duration and timing stability produced a temporal resolution of about 49.8 femtoseconds.
The experiment did more than characterize the electron beam; it also tested whether the compact LWFA source could perform real diffraction measurements. The electron pulses were directed onto single-crystalline gold samples, producing clear diffraction patterns in both single-shot and multi-shot modes. The patterns contained the expected information about the repeating atomic arrangement in the crystal. By analyzing the diffraction features, the researchers extracted the gold lattice constant, which agreed well with the known value. This agreement confirms that the transported LWFA beam retained the quality required for quantitative structural analysis, rather than merely producing a detectable scattering signal.
The study also points toward a substantially faster version of the instrument. Comprehensive start-to-end simulations, which model the electron beam from its generation in the plasma through magnetic transport and delivery to the sample, indicate that the bunch duration could potentially be reduced to around 10 femtoseconds root mean square. The simulated improvement would be accompanied by a lower energy spread of approximately 1.6 percent. Such performance would move LWFA-based UED closer to the sub-10-femtosecond regime, where researchers could investigate some of the earliest stages of atomic and electronic motion. These simulations represent a future capability rather than the directly measured result, but they show that the demonstrated beamline has room for further optimization.
The significance of the work extends beyond the size of the accelerator. A compact, laser-driven MeV electron source could make advanced diffraction experiments more accessible to laboratories that cannot accommodate large accelerator facilities. MeV electrons are particularly useful because they can penetrate samples more effectively than lower-energy electrons and can reduce some limitations associated with space-charge effects and sample thickness. Combining MeV energy with femtosecond duration, stable arrival timing and a miniaturized permanent-magnet beamline could create a flexible platform for examining phase transitions, chemical reactions, magnetic phenomena and light-induced changes in solids. The all-optical architecture may also simplify synchronization, since the same laser system can drive the accelerator and initiate the sample dynamics.
For now, the results establish a proof of principle rather than a complete replacement for established UED sources. LWFA beams can still exhibit fluctuations in charge, energy and pointing, and transporting them while preserving ultrashort duration requires careful control of plasma acceleration and magnetic optics. Nevertheless, the successful observation of gold diffraction demonstrates that these challenges can be managed well enough to perform meaningful structural measurements. With improvements in energy stability, charge control, beam transport and temporal compression, the technology could evolve into a powerful tool for filming matter in motion. The combination of a laser-driven accelerator, an isochronous double-bend achromat and terahertz timing diagnostics now provides a credible route toward ultrafast electron diffraction with temporal resolution approaching—and potentially passing—the ten-femtosecond frontier.
Subject of Research: Laser wakefield accelerator-based MeV ultrafast electron diffraction and femtosecond electron-beam transport.
Article Title: Ultrafast electron diffraction with MeV electron source from a laser wakefield accelerator
Article References: Fang, Y., Li, F., Hua, J. et al. “Ultrafast electron diffraction with MeV electron source from a laser wakefield accelerator.” Nature Photonics (2026). https://doi.org/10.1038/s41566-026-01980-6
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s41566-026-01980-6
Keywords: Ultrafast electron diffraction, laser wakefield accelerator, MeV electron beams, femtosecond science, electron-beam compression, isochronous beamline, double-bend achromat, terahertz deflector, structural dynamics, laser-driven electron source.
Tags: all-optical electron sourceatomic lattice diffractioncompact electron beamlinefemtosecond electron pulsesfemtosecond temporal resolutionhigh-charge electron buncheslaser wakefield acceleratorMeV electron sourcesingle-crystalline gold diffractionstructural change observationultrafast electron diffractionultrafast material dynamics
