Enhancing hard X-ray beamline performance at SwissFEL through spontaneous radiation measurements

Abstract

The hard X-ray beamline (Aramis) of the Swiss free-electron laser (SwissFEL) has been in user operation since the end of 2017 and its performance has been continuously monitored and enhanced over the past eight years. From the beginning, spontaneous radiation has been utilized to monitor the behavior of its 13 individual undulator modules: the pointing direction of the electron beam in each module, their alignment relative to the electron beam, and the calibration of their magnetic field strength (K calibration). This article introduces the methods employed at the Aramis beamline to optimize performance using spontaneous radiation, traces the evolution of these improvements, and highlights the recently achieved record performance.

Keywords: X-rays; free-electron laser; synchrotron radiation; undulator.

Figure 1

Sequence of photon beam instruments in the Aramis beamline and the Bernina endstation. From left to right: 13 undulator modules; variable-aperture slits (APU44 and APU92) for photon orbit definition; double-crystal monochromator (DCM) for wavelength selection; double-stage multi-channel plate (MCP) with a phosphorus screen (PS) and camera for profile monitoring; and silicon photodiode for direct intensity measurement, via a data acquisition system (DAQ) (used only during the commissioning period). All devices can be individually removed from the beamline to avoid obstructing the photon path.

Figure 2

MCP intensity versus undulator gap scan at 10.8 keV photon energy (left). Red numbers indicate the gap positions corresponding to the MCP images shown on the right (averaged over 25 shots). The image sequence illustrates the evolution of the transverse profile during the gap scan around a K value of 1.4. The reference frame is arbitrarily set around an initial guess of the gap.

Figure 3

Schematic of the source point triangulation concept (horizontal plane): the undulator module source is imaged twice on the MCP screen, first with slit APU44 (A) closed and later with slit APU92 (B) closed. Similar considerations can be applied to the vertical plane.

Figure 4

Source point analysis of the undulator modules along the Aramis beamline applying the described method. The horizontal and vertical positions are shown with an associated uncertainty of ±10 µm, and their linear fits are superimposed as solid red lines. The z positions of the undulator modules are indicated by gray rectangular boxes, the locations of the apertures and the MCP are marked by vertical lines.

Figure 5

Gap scans at fixed photon energy set by the monochromator performed (a) at different heights and (b) at different pitches. The zero height and zero pitch correspond to the undulator module center, which was determined prior to this measurement. Negative values are omitted for clarity, as they are identical to their positive counterparts due to the field’s symmetry about the center.

Figure 6

(a) Height and (b) pitch scans at fixed photon energy set by the monochromator and a fixed gap, which was selected to lie on the blue-edge side of the spectrum.

Figure 7

(a) Gap scans at a fixed photon energy set by the monochromator, performed for different currents in the Earth field coil. (b) Intensity as a function of current in the Earth field coil at a fixed gap corresponding to the blue edge.

Figure 8

Example of K calibration scans at a specific monochromator photon energy of 10.06 keV (i.e. a single K value of 1.5) for all undulator modules of the Aramis line recorded in November 2024.

Figure 9

Final result of the K calibration procedure carried out in November 2024. The plot shows the K values (plus signs) as a function of the undulator gap (center of the error function fit) for each undulator module. For the interpolation a fourth-order polynomial fit is used. Solid lines mark the latest calibration fits, dashed lines show the previous calibration fits for comparison.

Figure 10

Differences between the newly measured calibration fits (undulator gap to K) and the previous ones for the 13 individual modules over the measured K range. The systematic shift towards smaller gaps may be attributed to the electron beam energy calibration.

Figure 11

Performance evolution of SwissFEL over the past years. The pulse energies for photon energies between 11.8 and 12.4 keV are shown, where a straight electron trajectory and precise alignment and calibration of undulator modules are of particular importance.

Figure 12

Top: spatial distribution of undulator radiation (indicating FEL alignment and K calibration) from each of the 13 undulator modules as seen by the MCP downstream: (a) after the e-BBA, (b) after the subsequent p-BBA, and (c) after application of the K calibration. Bottom: changes in electron trajectory between e-BBA and p-BBA.