Instrument

Overview

Conceived as an open port complementing the other two baseline instruments at SASE 3, the Soft X-ray Port (SXP) instrument is primarily designed for time- and spin-resolved X-ray photo-electron spectroscopy. However, investigations of complex chemical and bio-inorganic molecular systems with fluorescence spectroscopy as well as research on highly charged ions are also envisioned. SXP is located behind the SASE 3 soft X-ray undulator providing photons with horizontal polarization in an energy range between 260 eV and 3000 eV exceeding 1012 photons per pulse with up to 27000 pulses/s. Two high-quality elliptical mirrors in Kirkpatrick-Baez (KB) configuration can focus the femtosecond XFEL pulses to a spot size of approximately 1 μm in diameter, resulting in an intensity of more than 1018 W/cm2 in the interaction region. Together with various synchronized femtosecond optical laser (OL) systems, whose wavelength ranges can be extended into the infrared as well as extreme ultraviolet region, it will pave the way for ultrafast pump-probe investigations at the SXP instrument combining intense and tunable soft X-rays with versatile optical laser capabilities.

Picture
Figure1: Top view of the SXP instrument model including tentative laser beam paths.

A top view of the SXP instrument model without an experimental end-station but including the permanent beamline components and an exemplified outline of the laser beam paths for day one operation is presented in Fig 1. The FEL beam enters the SXP instrument on the rightern side of this figure and passes first through the alignment laser system (ALAS) entering the KB optics tank thereafter. Before the soft X-ray beam is focused into the interaction region, it is lead into the photon arrival time monitor (PAM) and the laser in-coupling unit (LIN). While the PAM is used to measure the relative arrival times between the ultrafast XFEL and the OL pulses, the LIN allows for coupling the OL beam into the path of the XFEL in a co-linear manner, such that both beams can be properly focused onto the sample under investigation.

References

  • SXP Technical Design Report - PART A

    Manuel Izquierdo, January 2022

    Download
  • The SXP instrument at the European XFEL

    Patrik Grychtol et al., December 2022

    Download
  • The Laser Infrastructure at the SXP instrument of the European XFEL

    Patrik Grychtol et al., December 2022

    Download

The alignment laser system - ALAS 

The ALAS is an optical laser and imaging system whose purpose is to (pre-)align the beamline components farther downstream. It is supposed to simplify the alignment procedure of the SXP instrument, thereby also saving valuable user beam time, and reducing the risk of beam damage as well as wear on all components that can potentially be exposed to the focused XFEL radiation. A mechanical model of the ALAS placed right in front of the KB optics tank is displayed in Fig 2. In this figure, the XFEL beam enters the ALAS chamber from the bottom through a CF40 flange and it passes through two combined 4-way-crosses, each of which hosting a manipulator.

The first manipulator is equipped with a drop-in mirror to inject an optical laser beam along the beam path of the XFEL, while the second manipulator holds a set of fluorescent screens to be used for monitoring the positions of both laser beams. In combination with additional pop-in screens located farther downstream in the PAM or the LIN sections, it is possible to align the SXP instrument without having to resort to valuable XFEL operation.

Picture
Figure 2: Top view of the ALAS mechanical model.

Mounted in the black box perpendicular to both manipulators, the corresponding laser diode module including a set of alignment optics and a CCD camera assembly can be found. To maximize the utility of the ALAS, it should not only propagate coaxially with the XFEL beam, but also mimic its profile and divergence. That is how it is also possible to adjust the KB mirrors to focus on the sample using an optical beam only. To this end, a telescope has been placed after the laser diode allowing for the expansion of the optical beam to the size of the X-ray beam.

Optical pump-probe laser systems - LAS

Ultrafast optical pulses at the SXP instrument are mainly provided by the central pump-probe laser developed by the European XFEL laser group. This laser operates in a 20 Hz burst mode with an intra-burst pulse structure matching that of the European XFEL, as illustrated in the inset of Fig 1. The 20 Hz operation allows for sending the beam simultaneously to the two baseline experiments at SASE 3. Thus, both instruments (SQS and SCS) are able to work at 10 Hz with the same pattern as the X-ray pulses. While it is foreseen to offer laser pulses in a wide range of wavelengths around the visible spectrum via various frequency conversion schemes, two beams will be delivered to the SXP experiment hutch in the first place: one centered at 800 nm with pulse durations in the femtosecond range (mainly 15 fs and 50 fs) and the other centered at 1030 nm with pulse durations in the picosecond regime (mainly 0.9 ps and 400 ps). Both beams are passed through the dedicated SCS instrument laser hutch entering the SXP experiment hutch by means of an elaborate delivery system, as seen on the left side of Fig 1. The SASE 3 central pump-probe laser has primarily four set points with the intra-burst pulse pattern matching the one of the XFEL and they are summarized in the following table:

 

set points
intra burst repetition rate [MHz]
pulse energy @ 800nm
pulse energy @ 1030nm
1
4.5
0.05
1
2
1
0.2
4
3
0.2
1
20
4
0.1
2
40

Both laser beams allow for optical excitations at discrete wavelengths in the ultraviolet region by frequency up-conversion through second-, third- and fourth-harmonic generation reaching as high as 257 nm. In case of the 800 nm beam, however, fourth-harmonic generation is not feasible at these high pulse energies, because absorption of the generated light by the conversion crystal leads to its rapid degradation. To fill the spectral gaps until about 250 nm, an optical parametric amplifier (OPA) system can be used, which at the same time allows for tuning the wavelength into the mid-infrared region until approximately 15 μm to enable resonant excitations of collective low-energy modes, such as phonons, magnons, spin density waves, etc.

Furthermore, the SXP instrument is equipped with an additional Yb fiber laser system centered at 1030 nm, delivering 250 fs short and 200 μJ strong laser pulses at a repetition rate of about 300 kHz. An internal acoustic-optical modulator allows for mimicking the characteristic 10 Hz bunch pattern and its pulse train can be fully synchronized to the European XFEL. As indicated at the bottom of Fig. 1, it is planned to compress the pulse duration below 40 fs using a Herriott-type multi-pass cell and extend the wavelength range into the extreme ultraviolet (XUV) region employing high harmonic generation (HHG) techniques.

The soft X-ray transport system

The SXP instrument shares the SASE 3 undulator and the first part of the X-ray transport system with the baseline instruments SCS and SQS. The X-ray transport system, depicted in Fig. 3, can be divided into four parts: (i) The offset mirrors, (ii) the monochromator, (iii) the distribution mirrors, and (iv) the focusing system. The offset mirrors and the monochromator parts are common to all the instruments. The offset mirror system is a combination of a flat mirror, M1, and an adaptive optics, M2, deflecting the beam horizontally. They are arranged in a chicane configuration to suppress the spontaneous radiation produced by the source. The incident angle of both mirrors can be adjusted between 6 mrad and 20 mrad in order to suppress the higher order harmonics. The M2 bending system is used to optimize the mirror aperture with respect to the source divergence and to produce an intermediate horizontal focus. The optics have extremely tight tolerances in slope errors to work in the diffraction limit: 20 nrad to 50 nrad (RMS) slope profile errors and 2 nm peak-to-valley (PV). All X-ray optical elements in the system are coated with a B4C layer, chosen to minimize damage effects at high pulse energies. Up to 17 mJ have already been demonstrated.

 

Picture
Figure 3: Sketch of the SASE3 X-ray transport system. The SXP instrument shares the first 340 m with SCS and SQS. The distribution mirror M6, located at 344.5 m from the source, deflects the FEL beam into the SXP experiment hutch.

The monochromator section is made of two pre-mirrors followed by two gratings, G1 and G2 and a plane mirror M4. The fix radius pre-mirrors are installed at fixed incidence angles: 20 mrad and 9 mrad for the low energy (LE) and high energy (HE) pre-mirrors, respectively. The LE mirror covers the energy range from 0.25 keV to 1.5 keV and the HE mirror that between 1.5 keV and 3 keV. M4 is flat and allows to use the pink beam. The grating G1 has 50 lines/mm and G2 150 lines/mm providing a resolving power of ∼3000 and ∼10000, respectively. The specific X-ray optics of SXP are the M6 distribution mirror and the KB focusing system. M6, positioned at 344.5 m from the source, steers the beam to the SXP branch. It is flat and has a nominal length of 0.95 m. The mirror quality is similar to that of the offset mirrors and it is also coated with B4C.

Picture
Figure 4: Focusing schemes of the bendable KB system. Three optical configurations are envisioned: F1 and F2 and a walk-through focus F3. The blue drawings represent the horizontal focusing plane, while the orange drawing refers to the vertical one.

The KB focusing system will be installed in the SXP experiment hutch. It has been conceived with two bendable mirrors that allow to set different interaction points after the last permanent beamline component (LIN). The X-ray beam will impinge on the horizontal focusing mirror first and on the vertical focusing mirror thereafter at a grazing angle of 9 mrad. This will focus the intermediate horizontal focused beam realized with M2 and the intermediate vertical focused beam produced by the monochromator system. Two interaction points F1 and F2, separated by 1.5 m, have been defined. They can be reached using the configurations BTL#1 and BTL#2 shown in Fig. 4. An additional work through focus point F3 has been defined to allow for changing the spot size between approximately 1-2 μm at F1, and 100 μm (FWHM). The first KB implementation will be realized by means of fixed radii mirrors, which are supposed to provide a spot size of about 30 μm.

The photon arrival time monitor - PAM

To perform pump-probe experiments with the highest possible time resolution combining an OL with a XFEL, it is not only important to generate ultrafast laser pulses, but also to properly synchronize these light sources and to compensate any residual jitter between them. By utilizing a dedicated balanced optical cross-correlation technique based on the actively stabilized distribution of an optical reference signal, a relative in-loop timing jitter on the sub-10 fs level has been achieved at European XFEL. A feasible approach to mitigate any residual jitter is to measure the relative arrival time of the pump and probe pulses on a single shot basis and sort the data accordingly in the subsequent analysis process, potentially reaching sub-femtosecond accuracy. For this purpose, the SXP instrument is equipped with a PAM placed downstream of the KB optics tank. As indicated in Fig. 5, it receives a weak portion of the OL pulses that are split off the main beam pumping the sample under investigation. After passing over a delay stage, these pulses are stretched in time propagating through a thick piece of glass and they are focused onto a thin semi-transparent membrane, which at the same time is illuminated by the soft X-ray beam coming from the KB optics chamber. The XFEL pulses change the refractive index and thus transmission of the membrane material, whose transient response is measured by the chirped optical pulses such that their relative delays to the soft X-ray pulses are mapped onto the optical laser spectrum. This established technique is known as spectral encoding. The PAM was designed and set up by the X-ray photon diagnostics group of the European XFEL. It has already been successfully employed to monitor the relative arrival times between XFEL and synchronized OL pulses resulting in a timing jitter of about 58 fs (FWHM). Moreover, it was demonstrated that applying corresponding time-of-arrival corrections in pump- probe experiments can significantly improve their temporal resolution.

 

Picture
Figure 5: Close-up view of the PAM (left) and LIN (right) showing the tentative 800 nm laser beam path. Some mechanical components in this figure have been omitted for the sake of illustration.

The laser incoupling unit - LIN

The LIN chamber is the most downstream beamline component of the SXP instrument placed right in front of any experimental end-station. As indicated in Fig. 5, it serves for collinearly coupling the OL beam, which is focused onto the sample under investigation by a curved mirror system outside the chamber, into the XFEL beam path. In order to admit a large bandwidth of laser wavelengths without having to break the vacuum, this chamber can host several in-vacuum mirrors mounted on a two story rotatable carousel, the properties of which can be chosen according to the requirements of a particular experiment. The mirrors can have a diameter of either 2” or 3” and they are mounted at an angle of 15° with respect to the XFEL beam, which can pass collinearly with the OL via holes milled through the centre of each mirror. Vacuum compatible piezo motor driven mirror mounts can be used to align the OL beam with nm precision. Downstream of the in-coupling mirrors, a scintillator on a manipulator can be moved into the beam paths to image the OL as well as the XFEL beams for spatial overlap, at the same time allowing for the alignment of the entire beamline when combined with the ALAS. Moreover, the manipulator is equipped with an ultrafast photodiode that can be used to narrow down the temporal overlap of both beams to less than 100 ps. A conservative assessment utilizing ray-tracing simulations reveal that by using a pair of curved mirrors placed onto the optical breadboard below the LIN chamber, a focus with a diameter of about 100 μm at the sample interaction region can be obtained. This translates to an intensity of more than 1015 W/cm2 per mJ pulse energy on target using the 800 nm beam with a pulse duration of 15 fs.

 

The TR-XPES experimental endstation

The first endstation that is going to be installed at the SXP instrumet is the time-resolved X-ray photoelectron spectroscopy (TR-XPES) experiment. As depicted in Fig. 6, it is composed of a main chamber equipped with the photoelectron spectrometer and a sample manipulator with a cryostat cooling down to 20 K, a sample preparation chamber, a LEED chamber and a load-lock for fast sample exchange. The main feature of the system is the time-of-flight (ToF) momentum microscope photoelectron spectrometer. It represents a novel approach to angle-resolved photoelectron spectroscopy (ARPES) allowing for measuring two dimensional maps over large momentum areas (kx, ky), thus providing the highest degree of parallelization possible in ARPES experiments. k –microscopy uses a basic optical concept: in imaging systems, the reciprocal image represents the distribution of the transversal momentum components. Since k is conserved in the photoemission process, the reciprocal image formed in the back-focal plane of a cathode lens maps the transversal momentum distribution of the electrons inside the crystal. Valence and core electrons can be directly accessed providing images of either the electronic structure in energy-momentum space or the chemical and atomic structure in real space.

Picture
Figure 6: Model of the momentum microscope adapted from Review of Scientific Instruments 91, 013109 (2020).

The soft X-ray range of the SXP instrument has required the development of new electron optics. The resultant lens accepts the full half-space above the surface with initial kinetic energies of up to several keVs. This value represents an increase of the maximum k field radius by a factor of 3 as compared to standard low-energy designs. The larger acceptance angle results in the enhancement of the total intensity by one order of magnitude. The first k image in the back-focal plane is formed by a special objective lens designed for minimum spherical aberration. The electron beam is transformed to the desired drift energy in the imaging ToF column. Energy-dependent k momentum maps are taken simultaneously in an energy interval defined by the drift energy. For high resolution, band-mapping intervals of up to 10 eV are imaged in best focus. Much larger intervals can be recorded with a reduced resolution determined by the chromatic aberration. To achieve a good energy resolution and minimize the “crosstalk” between the longitudinal and transversal momentum components, time markers with 10 ns spacing are defined as planar isochrones.

Besides the lens system, the other key components of the spectrometer are the 900 mm long drift tube and the detector. The applied voltage selects the energy interval of the electrons reaching the detector. The detector is an 80 mm diameter delay-line (DLD) with a spatial resolution of 80 μm and a temporal resolution of 150 ps. This allows for resolving about 1 Megapixel image points, which is higher than the resolution of the electron optics. The time resolution allows a maximum count rate of 8 · 106 counts per second. The DLD records all counting events in the energy window selected by the voltage of the drift tube. The energy resolution of the ToF microscope is defined by the length and potential of the drift tube and the temporal resolution of the detector. The SXP k -microscope can achieve an energy resolution ∆E between 70 meV and 9 meV, when working with drift energies between 60 eV and 10 eV, exceeding the energy resolution provided by the SASE 3 monochromator. Given the 150 ps time resolution of the DLD, the time gap of 222 ns (4.5 MHz mode) allows for resolving 1,500 time slices, thus enabling high-resolution work. However, the effect of temporal aliasing may restrict the usable working range. Spin detection capabilities will be added in a second stage by inserting a spin-filter crystal (SFC) as already shown in Fig. 6. A lens will focus a parallel electron beam on the spin-filter. The reflected spin filtered electrons will enter into another ToF column, a drift tube and a DLD detector similar to those of the spin-integrated detector. Azimuthal rotation of the SFC will allow for switching from the transversal to the in-plane spin component thus enabling vectorial spin detection.