Demonstra*on of Two- color XFEL Opera*on and Autocorrela*on Measurement at SACLA

Transcription

1 Demonstra*on of Two- color XFEL Opera*on and Autocorrela*on Measurement at SACLA Toru Hara, Yuichi Inubushi, Tetsuya Ishikawa, Takahiro Sato, Hitoshi Tanaka, Takashi Tanaka, Kazuaki Togawa, Makina Yabashi RIKEN SPring- 8 Center Tetsuo Katayama, Tadashi Togashi, Kensuke Tono JASRI/SPring- 8

2 Contents Laser pulse length measurement with an autocorrelajon method. Two- color XFEL operajon. Bunch by bunch mulj- energy operajon of a linear accelerator. Future upgrade plans.

3 Electron beam Installa*on of a chicane in SACLA BL3 Undulators ID01~08 (first section) Chicane (= one ID length) max ~0.15 Undulators ID09~19 (second section) 0.3 m 1.6 m 5.1 m Self- seed experiments. Laser pulse length measurements using an autocorrelajon technique. Time delay for two- color operajon. Maximum Jme delay 45 fs for 8 GeV with sub- fs resolujon.

4 Orbit correc*on of chicane Orbit deviajon at maximum delay with respect to 0 delay. Beam orbit deviajons are within±2 μm. Feed forward orbit correcjon as a funcjon of the chicane current using a pair of steering magnets.

5 Gain curve at 10 kev µ Measured gain length indicates a laser pulse length around 2-3 fs.

6 Laser pulse length measurement autocorrela*on technique Gaussian pulse assumed.

7 Laser pulse length measurement spectral spike width Single shot spectrum measurement using Si (660) with a resolution ~70 mev. Normal operajon Long bunch with smaller compression factor.

8 Laser pulse length measurement spectral spike width The laser pulse length of SACLA is less than 10 fs (FWHM). Some current spikes exist?

9 Two- color XFEL opera*on in SACLA First two- color operajon in hard x- rays. Simple scheme: 19 undulators of of BL3 are divided into 2 secjons with different K- values. Two- color FEL is already open to the user experiments. Two wavelengths are tunable with maximum separajon of ~30 %. Time delay between two pulse can be adjusted between 0~40 fs with a sub- femtosecond resolujon.

10 Averaged spectrum of 2- color operajon K=1.92 (9 kev) for the first half of undulators (ID01-08), K=2.1 (8 kev) for the last half (ID10-19). Total pulse energy is about 130 μj. No delay dependent output change.

11 Averaged spectrum of 2- color operajon To equilibrate two outputs, the lasing of the first color was stopped well before saturajon.

12 Two- color XFEL opera*on in SACLA CorrelaJon between the first- and second- color intensijes.

13 Two- color XFEL opera*on in SACLA The maximum separajon of two wavelengths is about 30 %. Peak power of the laser pulse reaches several GW.

14 Two- color XFEL opera*on in SACLA SpaJal separajon of two- color pulses. Electron beam is deflected by 10 µrad at chicane. Undulators and Q- mags are realigned along the deflected orbit (2-3 mins).

15 Mul*- beamline opera*on of SACLA

16 How to adjust the beam energy? Laser wavelength Electron beam energy λ = λ! u 1+ K 2 # 2γ 2 " 2 $ & % Undulator gap Impossible µ

17 Beam energy control from bunch to bunch Bunch to bunch distribujon to mulj- beamline using a fast switching magnet (kicker + DC septum). Electron beam energy needs to be changed from bunch to bunch. Fast parameter change from bunch to bunch leads to instable operajon of RF. Since all RF systems operate under steady condijon, the developed method does not degrade the accelerator stability.

18 Demonstra*on of the mul*- energy accelera*on Since a fast switching magnet has not yet installed, all bunches are sent to BL3. The beam energy is measured at upstream chicane of BL3 undulators. Bunch repejjon 10 Hz. 8 C- band RF units (16 accelerator structures) are operated at 5 Hz.

19 Demonstra*on of the mul*- color lasing from bunch to bunch 7.3 GeV 7.8 GeV Bunch repejjon C- band 4 RF units 10 Hz 5 Hz 7.3 GeV 7.55 GeV 7.8 GeV Bunch repejjon C- band 2 RF units C- band 2 RF units 10 Hz 5 Hz 1 Hz

20 Mul*- energy opera*on from bunch to bunch Beam envelope and orbit can be readjusted aper the switching magnet. Expected energy range, for example 4~8.5 GeV. Beam envelop can be matched to the undulator FODO.

21 Future upgrade plans of SACLA Self- seed experiments to generate Fourier transform limited pulses (autumn 2013). Beam injecjon to SPring- 8 (SACLA to booster synchrotron, autumn 2013) InstallaJon of the second beamline BL2 (summer 2014) Move SCSS test facility to SACLA undulator hall (from 2013 to 2014) Development of a fast switching magnet.

22 Near future plans of SACLA 0.5 MeV 35 MeV 400 MeV 1.4 GeV Aperture Beam dump Gun CB-correction 476MHz Energy slit Chopper 238MHz Injector L-APS(2m 2) LB-correction BC1 (R56~-41mm) S-band (3m 8) BC2 (R56~-37mm) C-band (1.8m 24) Z=0 m Z~13 m Z~50 m Z~120 m Q-mag for dispersion correction BC3 (R56~-7.5mm) Modified YAG screen to avoid COTR Switching magnet 8 GeV 3 BL3 SCSS 0.36 BL2 ID01-08 BL1 Undulators (5m 15) λu=15 mm Chicane ID09-19 Undulators (4.5m 2) λu=15 mm Beam dump To users SR from SPring-8 Deflector cavity (1.8m 2) Main C-band (1.8m 104) Undulators (5m 19) λu=18 mm Beam dump Combination of XFEL and SR XSBT To SPring-8 Z~390 m Z~615 m Experimental hall