This page contains automatically translated content.
Detecting Nanoparticles Using X-Ray Films Composed of Two Images
The X-ray laser can generate two consecutive flashes of different colors that are bright enough to image a particle twice within femtoseconds. However, one major challenge remained: No detector is fast enough to record the two snapshots separately—they overlap in a single image, similar to a double exposure. Scientists have now successfully employed two complementary methods to separate two diffraction patterns captured by the same X-ray detector. The international team’s findings were published in two separate articles in *Nature Communications*.
While both methods exploit different colors—or photon energies—one relies on the detector’s ability to distinguish energy levels for each individual pixel, while the other uses mathematical reconstructions. The inventors refer to the latter method as “dichography.” One of the lead researchers compares the technique to an extremely fast high-speed camera: “To my knowledge, these are the fastest films ever recorded at the nanoscale, if by ‘film’ we mean several individual images of the same object,” says Alessandro Colombo of the Department of Physics at ETH Zurich.
“This opens up the possibility of observing rapid changes in nanoparticles—similar to studies that have been possible with small molecules for some time now,” says Yevheniy Ovcharenko, a scientist at the “Small Quantum Systems” (SQS) experimental station at European XFEL and lead author of the study. “These nanoparticles are less uniform than individual molecules, which is why it is necessary to observe processes using the same sample.” Experiments on single molecules that track chemical processes typically rely on a series of observations of different samples taken at varying time intervals following a triggering event. However, this approach does not yield precise results for particles that may differ in size, shape, orientation, or internal structure.
Using two X-ray images of the same particle taken at different time intervals, researchers now have the ability to directly track a single event—such as an expanding nanoplasm, a disintegrating cluster, or a light-induced structural rearrangement. “We are at the dawn of a new kind of research,” adds Marcel Mudrich of the University of Kassel, who proposed the experiment together with his SQS colleague. “We now have free-electron lasers that can generate pairs of X-ray pulses with different ‘colors,’ we have detectors that can distinguish images based on the color of the scattered X-ray light, and we have the analytical tools to reconstruct the shape of individual particles from the recorded X-ray snapshots.”
To develop the color-separation and densitography methods and demonstrate their feasibility, the scientists used the SQS experimental station at European XFEL. Two sections of the facility’s undulators—X-ray light sources with a total length of 120 meters—were tuned to different X-ray photon energies (approximately 1.0 and 1.2 kiloelectronvolts) or “colors”—thereby generating two X-ray pulses. The second X-ray pulse was emitted a few tens to a few hundreds of femtoseconds after the first. The researchers then focused both pulses onto free-floating helium nanodroplets and recorded the scattered X-ray light with a pnCCD detector. This device measures the charge generated by incoming photons, allowing their energy to be estimated at each pixel. By analyzing the charge accumulated in each pixel and in small groups of pixels, the team was able to assign many of the detected photons to either the first or second X-ray pulse.
“It’s more complicated than it sounds,” says Michael Meyer, senior researcher and Section Head at the SQS experimental station and one of the authors of the publications, “because we have a very large number of photons. A single pixel could have collected one photon from the first pulse and several photons from the second pulse. In addition, a single photon can also illuminate multiple pixels.” Linos Hecht, a doctoral student at ETH Zurich and first author of the publications, emphasizes that “the method is very flexible because it does not rely on specific properties of the sample. The method works particularly well in the less-illuminated areas of the detector at large scattering angles, where the finest structural information can be found.”
This pixel-level color separation provided experimental proof that the principle works. The second step was more ambitious: Instead of sorting the detected photons by color, the researchers wanted to reconstruct two actual images of the sample from a mixed diffraction pattern. This is the idea behind dichography, which was developed in a companion study. The method uses iterative phase recovery algorithms to search for two distinct structures whose diffraction intensities add up to the measured signal. Using European XFEL data, the team reconstructed two separate images of a helium nanodroplet containing xenon particles from individual overlapping patterns, with a time interval of 50 and 750 femtoseconds. The movies, composed of two individual images, showed that the embedded xenon nanoclusters appeared essentially unchanged: Although intense light can cause particles to explode, this process had barely begun within the observed time period.
The experiments and accompanying simulations also showed that challenges remain: dichography works only if the data is bright enough and the contributions of the two colors are reasonably balanced. “I am certain that further development at facilities such as the European XFEL will lead to these limitations becoming less and less significant, and our method will find an ever-wider range of applications,” says Colombo. “We are embarking on a new scientific adventure and finally making the dream of filming ultrafast structural changes in individual nanoparticles a reality.”
Original publications: