Periodic Reporting for period 1 - HIGH-Q (Breaking resolution limits in ultrafast X-ray diffractive imaging)
Reporting period: 2022-04-01 to 2024-09-30
Our ability to observe processes and study function at the nanoscale is limited due to the compromise between temporal and spatial resolutions inherent to the majority of our far-field imaging techniques. The problem can be intuitively understood using the example of a simple pinhole camera. An almost infinitely sharp image of an object can be obtained using a small entrance pinhole which comes with the drawback of a long exposure time. A widening of the pinhole reduces the exposure time, but the resulting image is less crisp. The trade-off between temporal and spatial resolutions in far-field imaging limits our perspective on many non-equilibrium processes at the nanoscale such as chemical and catalytic reactions, ultrafast phase transitions, non-linear light matter interactions and biological processes. This applies particularly to dynamics in non-crystallin samples occurring at room temperature. One additional challenge is the damage inflicted upon fragile nanoscale samples during the required exposure time.
Theoretically, X-rays are ideal to mirror processes with high spatial and temporal resolutions. Xray photons have Å-short wavelengths, don´t repel each other like electrons, can penetrate through extended volumes, are insensitive to temperature and can produce images with high material contrast. One major limitation is that the X-rays are highly damaging as virtually all matter is being ionized by X-rays through photo-absorption. Intense and spatially coherent femtosecond-short X-ray pulses from free-electron laser (FEL) sources enable outrunning this photo-induced damage through 'diffraction-before destruction' coherent diffractive imaging (CDI) of individual nano-specimens within a single exposure. The PI and colleagues have
demonstrated that irreversible processes on the femtosecond time scale can be followed using FEL CDI with few nm resolution and 100 femtosecond time resolution at the single particle level. Other CDI FEL studies have found surprising morphologies in in aerosols, metastable shapes of metal nanoparticles, exotic states of water just to name a few. As such FEL CDI has theoretically wide-ranging applications ranging across chemistry, material sciences to biology, but so far is not widely utilized due to limited spatial resolution.
Project overview and objectives: The high spatial resolution information is lost in the dim photon signal at higher angle scattering signal. Most X-rays interact little with matter and if X-rays are scattered, the majority of the elastic scattering goes into the forward direction and only a fraction of photons changes their direction significantly. Despite significant efforts, it has been unclear how to improve the resolution of a single diffraction image much beyond 5-10 nanometers. Spatial resolution improvement of one order of magnitude requires four orders of increase in the product of X-ray photon flux Iph (photon number per area) multiplied by coherent scattering cross section σscat. Simply increasing Iph is extremely expensive and maybe even useless due to bleaching effects during the exposure. This proposal focuses on overcoming the current limitations by exploiting a previously little explored set of optimization parameters which can increase σscat by orders of magnitude through non-linear effects such as transient resonances and resonant stimulated emission. The results will have far-reaching consequences across disciplines such as photo-chemistry, catalysis and study of biological processes at room temperature.
Theoretically, X-rays are ideal to mirror processes with high spatial and temporal resolutions. Xray photons have Å-short wavelengths, don´t repel each other like electrons, can penetrate through extended volumes, are insensitive to temperature and can produce images with high material contrast. One major limitation is that the X-rays are highly damaging as virtually all matter is being ionized by X-rays through photo-absorption. Intense and spatially coherent femtosecond-short X-ray pulses from free-electron laser (FEL) sources enable outrunning this photo-induced damage through 'diffraction-before destruction' coherent diffractive imaging (CDI) of individual nano-specimens within a single exposure. The PI and colleagues have
demonstrated that irreversible processes on the femtosecond time scale can be followed using FEL CDI with few nm resolution and 100 femtosecond time resolution at the single particle level. Other CDI FEL studies have found surprising morphologies in in aerosols, metastable shapes of metal nanoparticles, exotic states of water just to name a few. As such FEL CDI has theoretically wide-ranging applications ranging across chemistry, material sciences to biology, but so far is not widely utilized due to limited spatial resolution.
Project overview and objectives: The high spatial resolution information is lost in the dim photon signal at higher angle scattering signal. Most X-rays interact little with matter and if X-rays are scattered, the majority of the elastic scattering goes into the forward direction and only a fraction of photons changes their direction significantly. Despite significant efforts, it has been unclear how to improve the resolution of a single diffraction image much beyond 5-10 nanometers. Spatial resolution improvement of one order of magnitude requires four orders of increase in the product of X-ray photon flux Iph (photon number per area) multiplied by coherent scattering cross section σscat. Simply increasing Iph is extremely expensive and maybe even useless due to bleaching effects during the exposure. This proposal focuses on overcoming the current limitations by exploiting a previously little explored set of optimization parameters which can increase σscat by orders of magnitude through non-linear effects such as transient resonances and resonant stimulated emission. The results will have far-reaching consequences across disciplines such as photo-chemistry, catalysis and study of biological processes at room temperature.
The overall goal of the HIGH-Q project is to identify strategies for improving the spatial resolution in ultrafast X-ray coherent imaging. The idea is to use very short FEL pulses tuned to excite non-linear effects, which will increase the brightness of single exposure X-ray diffraction patterns of nanoparticles. The brightness is a crucial parameter because currently the spatial resolution in ultrafast X-ray imaging is shot-noise limited in high scattering angles, which carry the information about the finest structures. We have successfully passed two critical steps, which are the corner stones of the entire project. Before the start of the project, we had already demonstrated that transient resonances can make soft X-ray coherent diffraction images brighter and lead to higher spatial resolution in images of Xe nanoparticles. These previous results are currently under review in Nature Communications. I have presented the Xe data and simulation at various international conferences. While the response from my scientific community was mostly positive, a few members were sceptical about the generality of the effect. In particular, it was pointed out that previous studies found that transient resonances in light elements, which are important for organic chemistry and biology, are rather detrimental to the imaging process. Also, the persistence of transient resonances in metals in the hard X-ray regime, which are crucial for high spatial resolution imaging of many samples relevant to chemistry and material sciences, was questioned.
During the course of the HIGH-Q ERC project, we were able to address and refute both concerns. We have demonstrated that (part I a) transient resonances can enhance coherent X-ray diffraction signal from metals in the hard X-ray regime, and that they can enhance the brightness of images of light elements (part I b). Overall, we were able to identify some unexpected difficulties as well as surprising new findings, which will define the next steps of the project as described below. These results have been demonstrated by our group at several international conference and will be used to write future experiment proposals.
The part II of the project focuses on the integration of super-resolution approaches into X-ray coherent diffraction imaging. Our first tests suggest that under certain conditions super-resolution algorithms can provide a better contrast and also a higher spatial resolution. However, in the extreme shot noise limited case, super-resolution approaches can suffer from artefacts, which need to be addressed in the next steps of the project.
During the course of the HIGH-Q ERC project, we were able to address and refute both concerns. We have demonstrated that (part I a) transient resonances can enhance coherent X-ray diffraction signal from metals in the hard X-ray regime, and that they can enhance the brightness of images of light elements (part I b). Overall, we were able to identify some unexpected difficulties as well as surprising new findings, which will define the next steps of the project as described below. These results have been demonstrated by our group at several international conference and will be used to write future experiment proposals.
The part II of the project focuses on the integration of super-resolution approaches into X-ray coherent diffraction imaging. Our first tests suggest that under certain conditions super-resolution algorithms can provide a better contrast and also a higher spatial resolution. However, in the extreme shot noise limited case, super-resolution approaches can suffer from artefacts, which need to be addressed in the next steps of the project.
The dominating view is that photo-ionisation has only detrimental effects on X-ray imaging. Even if the X-ray pulses are very short, this perspective is correct in most cases as photo-ionisation removes bound electrons from their orbitals and these electrons no longer contribute to the scattering cross section of their parent ions. This effect is called bleaching and has been extensively studied theoretically and experimentally. Overall, bleaching decreases the scattering efficiency of the parent ions and promotes the transparency of the specimen.
Within this simple model, our results are counter intuitive. Our data repeatedly suggests that ions with less bound electrons can scatter more efficiently than their neutral counter parts under certain conditions. The work conducted within HIGH-Q strongly indicates that a set of X-ray FEL parameters can excite very short lived resonances in most materials, which improve the brightness of the images of nanospecimens consisting of such materials.
Within this simple model, our results are counter intuitive. Our data repeatedly suggests that ions with less bound electrons can scatter more efficiently than their neutral counter parts under certain conditions. The work conducted within HIGH-Q strongly indicates that a set of X-ray FEL parameters can excite very short lived resonances in most materials, which improve the brightness of the images of nanospecimens consisting of such materials.