• High Power Laser Science and Engineering
  • Vol. 3, Issue 3, 03000001 (2015)
Siegfried Schreiber and Bart Faatz
Author Affiliations
  • Deutsches Elektronen-Synchrotron, Notkestrasse 85, 22603 Hamburg, Germany
  • show less
    DOI: 10.1017/hpl.2015.16 Cite this Article Set citation alerts
    Siegfried Schreiber, Bart Faatz. The free-electron laser FLASH[J]. High Power Laser Science and Engineering, 2015, 3(3): 03000001 Copy Citation Text show less
    Schematic layout of FLASH (not to scale); the electron beam direction is from left to right. The total length of the facility, including the experimental halls, is 315 m.
    Fig. 1. Schematic layout of FLASH (not to scale); the electron beam direction is from left to right. The total length of the facility, including the experimental halls, is 315 m.
    Schematic drawing of the FLASH RF-gun. The beam is emitted from the photocathode and exits to the right (path indicated by the red line). The laser beam illuminating the cathode enters along the electron beam path from the right (blue line). The RF is fed in via a coaxial waveguide coupler. Drawing courtesy: Elmar Vogel, DESY.
    Fig. 2. Schematic drawing of the FLASH RF-gun. The beam is emitted from the photocathode and exits to the right (path indicated by the red line). The laser beam illuminating the cathode enters along the electron beam path from the right (blue line). The RF is fed in via a coaxial waveguide coupler. Drawing courtesy: Elmar Vogel, DESY.
    QE evolution of a cathode in continuous operation for 436 days at FLASH.
    Fig. 3. QE evolution of a cathode in continuous operation for 436 days at FLASH.
    Picture of a TESLA-type 9-cell superconducting niobium cavity. The length is 1 m. Courtesy: DESY.
    Fig. 4. Picture of a TESLA-type 9-cell superconducting niobium cavity. The length is 1 m. Courtesy: DESY.
    Sketch of a TESLA-type superconducting accelerating module as installed at FLASH. The outer cryostat is not shown. Each cavity has its own RF-power coupler. Courtesy: DESY.
    Fig. 5. Sketch of a TESLA-type superconducting accelerating module as installed at FLASH. The outer cryostat is not shown. Each cavity has its own RF-power coupler. Courtesy: DESY.
    Installation of the cryo-module containing four 3.9 GHz superconducting cavities (red) into the FLASH injector in 2009. The first accelerating module with eight 1.3 GHz cavities has already been installed (yellow). Courtesy: Kai Jensch, DESY.
    Fig. 6. Installation of the cryo-module containing four 3.9 GHz superconducting cavities (red) into the FLASH injector in 2009. The first accelerating module with eight 1.3 GHz cavities has already been installed (yellow). Courtesy: Kai Jensch, DESY.
    View of a FLASH tunnel section with accelerating modules. Courtesy: Heiner Müller-Elsner and DESY.
    Fig. 7. View of a FLASH tunnel section with accelerating modules. Courtesy: Heiner Müller-Elsner and DESY.
    Photon energy along a photon pulse train of 430 pulses measured with a GMD at FLASH. The detector is able to resolve single FEL pulses (blue line). Also shown are the average over many shots (green) and maximum energies recorded (yellow). In this example, the wavelength is 18.2 nm, the pulse spacing . With a single-pulse energy of and 4300 pulses per second, the average SASE power is 350 mW.
    Fig. 8. Photon energy along a photon pulse train of 430 pulses measured with a GMD at FLASH. The detector is able to resolve single FEL pulses (blue line). Also shown are the average over many shots (green) and maximum energies recorded (yellow). In this example, the wavelength is 18.2 nm, the pulse spacing . With a single-pulse energy of and 4300 pulses per second, the average SASE power is 350 mW.
    Measured single-shot spectra at FLASH. The bold line shows an averaged spectrum over 300 shots. The spectra are obtained in saturation. The circles indicate a simulation of the averaged spectrum with the 3D code FAST[64]. Adapted from Ref. [8]. Adapted by permission from Macmillan: Nature Photonics, [8], Copyright (2007).
    Fig. 9. Measured single-shot spectra at FLASH. The bold line shows an averaged spectrum over 300 shots. The spectra are obtained in saturation. The circles indicate a simulation of the averaged spectrum with the 3D code FAST[64]. Adapted from Ref. [8]. Adapted by permission from Macmillan: Nature Photonics, [8], Copyright (2007).
    Time-resolved double ionization of He (dots). The solid line is a Gaussian fit to the autocorrelation data with a width of 39 fs (FWHM). Assuming a Gaussian FEL pulse shape, this gives a duration of (FWHM). The dashed line represents a three-pulse structure with temporal separations of the side peaks by 12 and 40 fs, with an added chirp of 50 fs. Reprinted with permission from Ref. [67]. Copyright (2009) by the American Physical Society.
    Fig. 10. Time-resolved double ionization of He (dots). The solid line is a Gaussian fit to the autocorrelation data with a width of 39 fs (FWHM). Assuming a Gaussian FEL pulse shape, this gives a duration of (FWHM). The dashed line represents a three-pulse structure with temporal separations of the side peaks by 12 and 40 fs, with an added chirp of 50 fs. Reprinted with permission from Ref. [67]. Copyright (2009) by the American Physical Society.
    Aerial view of the FLASH Facility at DESY, Hamburg. The accelerator is from top right to the lower left, with the two experimental halls; Kai Siegbahn hall (left) and Albert Einstein hall (right) in the lower left corner. The curved hall (left) and the construction site (bottom) belong to the synchrotron radiation facility PETRA III. Courtesy: DESY, July 2014.
    Fig. 11. Aerial view of the FLASH Facility at DESY, Hamburg. The accelerator is from top right to the lower left, with the two experimental halls; Kai Siegbahn hall (left) and Albert Einstein hall (right) in the lower left corner. The curved hall (left) and the construction site (bottom) belong to the synchrotron radiation facility PETRA III. Courtesy: DESY, July 2014.
    Basic scheme for splitting the bunch trains. The train is split into two parts, one to be sent to the FLASH1 beamline, the other to FLASH2. The gap between the sub-trains is large enough to ramp up the kicker system.
    Fig. 12. Basic scheme for splitting the bunch trains. The train is split into two parts, one to be sent to the FLASH1 beamline, the other to FLASH2. The gap between the sub-trains is large enough to ramp up the kicker system.
    An example of the steps in amplitude (top) and phase (bottom) within a RF pulse in one of the modules, needed to optimize compression for different charges at FLASH1 and FLASH2. The part from 0 to is for the sub-train to be sent to FLASH1, the part from 500 to is for FLASH2. The position where the step occurs is adjustable according to the length of each sub-train. The green curves show the setpoint for the step and the blue curves show the achieved step.
    Fig. 13. An example of the steps in amplitude (top) and phase (bottom) within a RF pulse in one of the modules, needed to optimize compression for different charges at FLASH1 and FLASH2. The part from 0 to is for the sub-train to be sent to FLASH1, the part from 500 to is for FLASH2. The position where the step occurs is adjustable according to the length of each sub-train. The green curves show the setpoint for the step and the blue curves show the achieved step.
    Example of simultaneous SASE at FLASH1 with single-bunch operation (left) and FLASH2 with 10 bunches (right) measured with GMDs. The top plots show the calibrated ion signal, the bottom row the single-shot electron signals resolving each pulse in the pulse train. The blue color indicates the last value. In addition, average (green) and peak values (yellow) are displayed as well. Note that the FLASH2 GMD is not yet calibrated.
    Fig. 14. Example of simultaneous SASE at FLASH1 with single-bunch operation (left) and FLASH2 with 10 bunches (right) measured with GMDs. The top plots show the calibrated ion signal, the bottom row the single-shot electron signals resolving each pulse in the pulse train. The blue color indicates the last value. In addition, average (green) and peak values (yellow) are displayed as well. Note that the FLASH2 GMD is not yet calibrated.
    Electron beam
    Energy rangeGeV0.35–1.25
    Peak currentkA2.5
    Bunch chargenC0.06–1
    Emittance (rms), norm.1.4
    Energy spread (rms)keV200
    RF pulse length800
    Number of bunches/train1–800
    Repetition rateHz10
    Bunch separation1–25
    Undulator
    TypePlanar, fixed gap
    Periodmm27.3
    Gapmm12
    Peak magnetic fieldT0.48
    K1.23
    Segment lengthm4.5
    Number of segments6
    Average -functionm10
    FEL radiation delivered to experiments
    (user runs 2014/2015)
    Wavelength (fundamental)nm52–4.2
    Average single-pulse energy10–500
    Pulse duration (FWHM)fs50–200
    Bandwidth (FWHM)%0.3–2.0
    Peak powerGW1–3
    Photons per pulse
    Photon pulses per second10–5000
    Peak spectral brilliance*
    Av. spectral brilliance*
    Table 1. Basic FLASH electron and photon beam parameters.
    Electron beam
    Energy rangeGeV0.5–1.25
    Peak currentkA2.5
    Bunch chargenC0.02–1
    Emittance (rms), norm.1.4
    Energy spread (rms)keV500
    Number of bunches/train1–800
    Repetition rateHz10
    Bunch separation1–25
    Undulator
    TypePlanar, variable gap
    Periodmm31.4
    K0.7–2.8
    Segment lengthm2.5
    Number of segments12
    Average -functionm6
    Expected SASE properties
    Wavelength (fundamental)nm60–4
    Average single-pulse energy10–500
    Pulse duration (FWHM)fs50–200
    Bandwidth (FWHM)%0.7–2.0
    Peak powerGW1–3
    Photons per pulse
    Photon pulses per second10–7500
    Peak spectral brilliance*
    Av. spectral brilliance*
    Table 2. Expected parameters for FLASH2.