Capabilities of the FXE instrument

FXE instrument started user operation in September 2017 [A. Galler , et al., J. Synch. Rad., 26, 1432 (2019)]. Several components and standard setups have been commissioned and validated during recents years by performing measurements, including pump-probe experiments, or developing new methodologies to perform new research. On this page we present current experimental parameters of the instrument and its standard components with their performance specifications. Additional components in the commissioning phase and under development are given in Section 4. 

This page is being constantly updated and currently still in construction. To receive most recent information on specific topics please contact the FXE-support mailing list (fxe-support@xfel.eu) or directly the FXE Leading Scientist Chris Milne (christopher.milne@xfel.eu).

1. Main experimental parameters currently reachable at FXE

Typical experimental parameters currently available for experiments at the FXE instrument are listed below (state on April 2025), but for up-to-date details please contact the instrument.

2.  Standard setup for pump-probe spectroscopy and scattering in solution jets 

The setup is optimized for performing femtosecond optical pump/X-ray probe experiments mainly on solutions of molecular complexes or suspensions of nanoparticles. Quasi-collinear optical excitation is typically done using either the pump-probe laser harmonics (800/400/266 nm) or using a Topas OPA which can be tuned from 240 nm to 5 um. The pulse duration of the optical pulses are typically 50 fs, with the time-resolution of these measurements often limited by the liquid jet thickness, which leads to group-velocity mismatch between the laser and X-ray pulses. For a 100 um thick liquid jet the expected time resolution is 100 fs, thinner jets are possible but require testing to ensure compatibility with the sample and solvent. X-ray probing of photo-induced transient dynamics in solution can be done using any combination of X-ray scattering (2.2) or spectroscopy techniques (2.3). For more details also see [D. Khakhulin, et al., Appl. Sci., 10, 995 (2020)] and [F. Lima et al. J. Synch. Rad. 30, 1168 (2023)].

2.1 Sample environment

The main sample environment is designed as a sealed chamber with Kapton windows large enough to allow simultaneous collection of X-ray emission and wide-angle X-ray scattering [F. Lima et al. J. Synch. Rad. 30, 1168 (2023)]. The sample chamber can be filled with an inert gas (typ. He or N2) to ensure anaerobic or dry atmosphere. The chamber is equipped with SMA feedthroughs for installing additional 0D detectors for total X-ray fluorescence yield measurements and I_zero normalization. There are three view ports for sample microscopes. The main X-ray beam propagates in a He-purged flight tube and enters the chamber directly without a window. The optical laser path is quasi-collinear with the X-ray beam.

Solution samples are circulated using an HPLC pump based system and supplied through quartz nozzles with round opening of various diameters, typically 200, 100 um, or 50 um. With a 100 um round nozzle the typical linear speed of a water jet reaches ca. 60 m/s allowing to refresh the sample volume between individual pump-probe events at the repetition rate of up to 564 kHz. The sample chamber has also been used with GDVN nozzles [M. Aleksich et al. JACS 145, 17042 (2023)] and a Drop-on-Demand system [S. Perrett et al. Struct. Dyn. 11, 024310 (2024)], both currently provided by the Sample and Environment group (SEC).

Additionally to the main pump two sample circulation pumps are available for in-line static UV-VIS spectroscopy and solution/solvent refill, all pumps are controlled remotely. The nozzles can be manipulated by three XYZ translation stages, a YAG fluorescence screen and fast diode can be mounted next to the nozzle and brought into the beam without opening the chamber.

2.2. von Hamos X-ray emission spectrometer 

A wavelength dispersive von Hamos spectrometer is based on an array of 16 independently controlled cylindrical analyzer crystals with 0.5 m bending radius and allows measuring entire emission lineshapes in a single X-ray shot (or train) without scanning. Emission spectra from individual crystals are focused and overlapped on a Jungfrau 500k or 1M detector running at 10 Hz framerate, i.e. one image per X-ray train, or can be used with a Gotthard II detector if single-shot measurements are possible and required. For the Gotthard II to be used the sample should have sufficient concentration to allow single-shot X-ray emission to be measured, otherwise the JF detector is more suitable due to its lower noise level. Due to the flexibility and large number of analyzers simultaneous collection of multiple emission lines from the same or different elements can be performed, provided a compatible configuration of analyzer crystals exists and feasible. The setup geometry accommodates Bragg angles in the range 67-83 degrees. Details on how this approach has been scaled towards working at MHz repetition rates can be found in [M. Biednov et al. Nucl. Inst. and Meth. A 1055, 168540 (2023)]. An online tool is available to select analyzer crystals with our spectrometer geometry.

Below we provide an example of simultaneous full Ni K-line emission spectrum collected from a 10 mM aqueous solution of a Ni salt flowed in a 100 um round jet. Incoming X-ray photons at 9.3 keV with 150 pulses per train were used for this measurement.

2.3. Large Pixel Detector (LPD) for wide-angle X-ray scattering 

Wide-angle X-ray scattering on solutions systems is collected using LPD at a typical detector distance of about 20-30 cm. The LPD-1M is a modular detector consisting of 16 super-modules, 256x256 pixels each, with 500 um square pixels and 500 um thick silicon sensor. LPD offers 3 gain stages with respective gain reduction factors of ca. 8.7 and 107 for the medium and low gain stages respectively, thus enabling simultaneous dynamic range of ca. 10⁴ at 12 keV. X-ray scattering images can be recorded in the 510 total memory cells at the repetition rate of up to 4.5 MHz thus storing up to 510 single-shot X-ray images per pulse train. The unique feature of LPD is the parallel gain readout technology so that the final image is constructed from individual pixels each in the optimal gain stage by software-defined thresholds excluding possible analog gain switching uncertainties. That means that now when the gain transitions have been characterized and thresholds optimized the detector response is expected to be linear over the entire dynamic range covered by the three gain stages. The detector has been used to record scattering up to 22 keV, allowing it to measure out to 11 1/Å.

2.4. Sample excitation capabilities 

A range of optical wavelengths are available for excitation at FXE ranging from the UV to the IR, with some limited capability in the THz frequency range. Laser radiation from the main Pump-Probe Laser system of European XFEL with 800 nm (50 and 15 fs) and 1030 nm (850 fs) wavelengths are readily available together with their frequency harmonics (see table below). A wavelength-tunable OPA source, TOPAS,  has been installed and commissioned in the hutch and is available (see the measured tuning curve below). Standard operation of the laser is with 50 fs pulse duration and 282 kHz maximum repetition rate.

The optical excitation beams are delivered to the sample interaction region using a flexible optical setup able to accommodate various configurations in terms of excitation geometry and wavelengths (see images below). This includes collinear and non-collinear (15 degrees) laser  incoupling, ability to change excitation wavelength within one shift preserving the focusing conditions (currently not including TOPAS).

3. Standard setup for single-shot X-ray diffraction in transmission geometry

The setup consisted of a vacuum chamber (down to 10^-5 mbar) with Kapton window for forward scattering typically collected by the LPD (section 2.3).  The optical pump and the X-ray probe beams are focused on the sample at normal incidence in quasi-collinear geometry. The direct X-ray beam is off-centered on downstream Kapton window and on the detector behind to enable large Q-range, a motorized beam block is used to stop the direct beam after the sample and before the Kapton window. Thin film samples for single-shot measurements are commonly mounted on Si₃N₄ or single crystal Si membranes in a standard frame with typical pitch of 0.5-1 mm between the membranes. Fast XY translation stages allow to center membrane targets on the X-ray beam using predefined optimal positions currently at a rate of ca. 0.5 Hz. Once the target has reached the position a single pair of X-ray and optical laser pulses is delivered upon a trigger from the control system. The setup is capable of measuing from pulse-on-demand mode through to higher repetition rates depending on the measurement. A cryo cooler for sample cooling is also available. An example of the type of measurement this setup is capable of can be found in J. Antonowicz et al. Acta Materialia 276, 120043 (2024). A modification of this setup to allow for measurements on molecular crystals, with sample rotation is also available. Results from this solid target, molecular crystallography setup are published in D. Vinci et al. Nat. Comm. 16, 2043 (2025).

4. X-ray absorption spectroscopy by simultaneous scanning of the undulator gap and cryogenic Si (111) monochromator.

Femtosecond X-ray absorption scanning at the FXE instrument is currently fully operational. X-ray energy scanning using a 2-bounce monochromator geometry is standard and has been used routinely on liquid jet samples and solid state materials (scattering and spectroscopy). Scanning the monochromator can be used with total fluorescence yield detection, transmission X-ray detection and any of the FXE secondary spectrometers (von Hamos, Johann and Dumond/Laue). The undulator gap is synchronized to the photon energy changes allowing large energy ranges to be scanned (e.g. 500 eV) making it possible to perform extended energy range measurements (EXAFS). Due to power load limitations on the monochromator the number of pulses, repetition rate and X-ray focus onto the mono crystals can all be configured as necessary for stable beam delivery. Incoming X-ray intensities can be monitored using up to 4 separate I_zero measurements to cover different dynamic ranges and instrument configurations.

5. Setup for simultaneous grazing incidence XRD and/or X-ray spectroscopy on solid state systems

At FXE we have performed measurements using several grazing incidence diffraction setups using a Z-axis goniometer in vertical geometry (see photograph below) or a Kappa goniometer. Both geometries use a Jungfrau detector mounted on the robot arm for detection, which can be configured using diffraction geometry coordinates. Additionally the von Hamos spectrometer can be aligned to collect X-ray emission or a diode used for X-ray absorption from the surface of the sample. This setup is also compatible with a liquid nitrogen cryoblower, which allows temperature control down to around 100K in ambient conditions. A modified version of the small goniometer is also availabe in a He enclosure for low-background or low X-ray photon energy measurements (e.g. < 6 keV).

6. Johann-type X-ray emission spectrometer

FXE is equipped with a Johann-type X-ray emission spectrometer consisting of 5 independently controlled spherical analyzer crystals with 1m radius of curvature. The emission signals are collected by a detector on the robotic arm that can be either Jungfrau area detector or a 0D APD depending on the application. The spectrometer has been commissioned (See Co K-beta emission from a metal foil measured using a Jungfrau detector below) and is available for use. The spectrometer geometry relative to the sample is limited in its flexibility, but it has been used in both its standard 90 degree configuration and in a forward scattering configuration.