Pulse-by-pulse multi-beam-line operation for x-ray free-electron lasers

The parallel operation of plural undulator beam lines is an important means of improving the efficiency and usability of x-ray free-electron laser facilities. After the installation of a second undulator beam line (BL2) at SPring-8 Angstrom compact free-electron laser (SACLA), pulse-by-pulse switching between two beam lines was tested using kicker and dc twin-septum magnets. To maintain a compact size, all undulator beam lines at SACLA are designed to be placed within the same undulator hall located downstream of the accelerator. In order to ensure broad tunability of the laser wavelength, the electron bunches are accelerated to different beam energies optimized for the wavelengths of each beam line. In the demonstration, the 30 Hz electron beam was alternately deflected to two beam lines and simultaneous lasing was achieved with 15 Hz at each beam line. Since the electron beam was deflected twice by 3° in a dogleg to BL2, the coherent synchrotron radiation (CSR) effects became non-negligible. Currently in a wavelength range of 4–10 keV, a laser pulse energy of $100–150\mu \mathrm{J}$ can be obtained with a reduced peak current of around 1 kA by alleviating the CSR effects. This paper reports the results and operational issues related to the multi-beam-line operation of SACLA.

Figure 1

Schematic of the SACLA facility.

Figure 2

Electron beam optics and magnet configuration for the beam transport from the end of the accelerator to the BL2 undulators. Solid lines indicate the horizontal beta function (red), vertical beta function (blue), and horizontal dispersion function (black), respectively. The inset on top shows the magnet configuration.

Figure 3

Configuration of the beam line switchyard composed of kicker and dc twin-septum magnets located at the end of the linear accelerator.

Figure 4

Block diagram of the kicker magnet power supply. FET units are water cooled.

Figure 5

An example of a switching pattern. Solid colored circles show the arrival timing of 60 Hz electron bunches. Half of the bunches, whose energy is 7 GeV, are deflected to BL2 (red circles). A quarter of them, having 6 GeV energy, are deflected to XSBT (green circles), and the rest of the bunches travel straight through the kicker magnet to BL3 with 8 GeV (blue circles).

Figure 6

Stability of the kicker magnetic fields measured by a gated NMR detector. The kicker magnet was operated at 60 Hz with a peak magnetic field of 0.61 T. The magnetic fields at the electron bunch arrival timing were measured and its relative variation was plotted with a sampling interval of one minute.

Figure 7

Longitudinal current profiles of the electron bunch measured by a C-band rf deflector [15]. The profile shown in red is the electron bunch with a high peak current used for the normal operation of BL3. The profile shown in blue is the electron bunch with a low peak current obtained after parameter optimization for BL2.

Figure 8

Averaged laser spectra measured by scanning a monochromator in the multi-beam-line operation, (a) BL2 and (b) BL3. The electron beam energy was 7.8 GeV and the undulator K-values were 2.85 and 2.1 for BL2 and BL3, respectively. The peak current is 1.2 kA for both beam lines.

Figure 9

Stability of the photon pulse intensity measured by an intensity monitor [21] in the multi-beam-line operation, (a) BL2 and (b) BL3. Red points represent single-shot results and blue lines show averaged values over 15 shots. The pulse repetition at each beam line was 15 Hz and the full horizontal scale corresponds to 10 minutes. The electron beam energy was 7.8 GeV and the undulator K-values were 2.85 and 2.1 for BL2 and BL3, respectively. The peak current is 1.2 kA for both beam lines.

Figure 10

Stability of the central photon energy measured by a wavelength monitor [22] in the multi-beam-line operation, (a) BL2 and (b) BL3. Red points represent single-shot results and blue lines show averaged values over 15 shots. The pulse repetition at each beam line was 15 Hz and the full horizontal scale corresponds to 10 min. The electron beam energy was 7.8 GeV and the undulator K-values were 2.85 and 2.1 for BL2 and BL3, respectively. The peak current is 1.2 kA for both beam lines.

Figure 11

Averaged laser spectra measured by scanning a monochromator in the multi-beam-line operation combined with the multienergy operation of the accelerator, (a) BL2 and (b) BL3. The electron beam energy of BL2 was 6.3 GeV and that of BL3 was 7.8 GeV. The undulator K-values were 2.85 and 2.1 for BL2 and BL3, respectively. The peak current is 1.2 kA for both beam lines.

Figure 12

Stability of the photon pulse intensity measured by an intensity monitor in the multi-beam-line operation combined with the multienergy operation of the accelerator, (a) BL2 and (b) BL3. Red points represent single-shot results and blue lines show averaged values over 15 shots. The pulse repetition at each beam line was 15 Hz and the full horizontal scale corresponds to 10 minutes. The electron beam energy of BL2 was 6.3 GeV and that of BL3 was 7.8 GeV. The undulator K-values were 2.85 and 2.1 for BL2 and BL3, respectively. The peak current is 1.2 kA for both beam lines.