Low slice emittance preservation during bunch compression S. Bettoni M. Aiba, B. Beutner, M. Pedrozzi, E. Prat, S. Reiche, T. Schietinger
Outline. Introduction. Experimental studies a. Measurement procedure b. Characterization of the core emittance increase c. Systematic checks 3. Simulations a. Simulations scenarios b. Characterization of the core emittance blow-up 4. Discussion 5. Conclusions
Introduction: in other machines < Compression factor 5 Compression factor ~ 4.5 Compression factor = 5.5 Charge profile (a.u.) Slice emittance Slice emittance increase Slice emittance increase No slice emittance increase 3
SwissFEL: FEL in Switzerland Gun S band 00 MV/m S band ( x4 m) 4/6 MV/m Laser Heater Booster Booster E = 50 MeV S band (4x4 m) 5 MV/m X band ( x 0.75 m) 5MV/m BC Deflector 30 MeV Linac BC Linac C band (36 x m) 6.5 MV/m C band (6 x m) 7.5 MV/m, 0 º Switch Yard Energy tuning.5-3.4 GeV C band (8 x m) max 8.5 MV/m Athos Undulators Athos Linac Deflector Linac 3 Deflector Energy tuning C band (5 x m). GeV 3.0 GeV max 8.5 MV/m.- 5.8 GeV Aramis Undulators 0.7-7 nm, 00 Hz; 360 μj > nm: transform limited (0.8) - 7 Å 5 0 fs; 00 Hz; 50 μj Soft X-ray Hard X-ray Wavelength - 70 Å Pictures courtesy of I. Widmer June 05 Pulse duration 3 0 fs Maximum e- beam energy 5.8 GeV e- beam charge 0 00 pc Repetition rate Slice emittance (expected performances) Slice energy spread Saturation length 00 Hz 00 nm (0 pc) 300 nm (00 pc) 50-350 kev < 50 m Construction started in 03 Commissioning planned in Feb 06 Aramis user operation planned in Dec 07 Athos user operation planned in 0 4
SwissFEL Injector Test Facility (SITF) Missions Benchmark the simulation expectations and prove the feasibility of SwissFEL Develop and test components/systems and optimization procedures in SwissFEL dipole quadrupole BPM+screen screen Acceleration (S-band) Harmonic cavity (X-band) Deflecting cavity (S-band) FODO cells S-Band RF gun Compression Diagnostic section Commissioning phases Max e- beam charge 00 pc Laser longitudinal shape Flat-top Gaussian Laser longitudinal length 9.9 ps FWHM 3.7 ps RMS Laser transverse shape Cut Gaussian Laser transverse RMS 0.8 mm Max e- beam energy 66 MeV Repetition rate 00 Hz Phase : Electron source and diagnostics (00) Phase : Phase + (some) S-band accelerations (00-0) Phase 3: Machine in full configuration: all RF structures operational and bunch compressor installed (0-03) Phase 4: Undulator installed for several weeks (04) Phase 4+: PSI gun installed (Oct 04) 5
Slice emittance measurement ~4 m 35 m <bx< 40 m, by<0 m at the PM Phase advance in y ~70 º TDC Measurement quads Profile monitor Phase advance in x from 0 to ~80º y time (0 slices/rms bunch length) x.5 380.4 M = (β 0 α 0 α + γ 0 β) 360 340 Core emittance Mismatch.3. 30. 300-0 -5 0 5 0-0 -5 0 5 0 6
Measurements: compression factor The bunch compression factor, CF, is given by: CF = hr 56 h: relative longitudinal chirp R 56 : derivative of the path length on energy 000 800 600 Uncompressed CF = CF = 4 CF = 5 CF = 6 CF = 7 CF = 00 0-5 -0-5 0 5 0 5 7
Measurements: compression factor The bunch compression factor, CF, is given by: CF = hr 56 h: relative longitudinal chirp R 56 : derivative of the path length on energy 000 800 600 00 Uncompressed CF = CF = 4 CF = 5 CF = 6 CF = 7 CF = Core COM / UNC.5.5 0-5 -0-5 0 5 0 5 0 5 0 5 Compression factor Core emittance increases versus CF 8
Measurements: other parameters.5 Scan over E (CF < 0) CF = hr 56 Scan over R 56 (CF < 0).5 Core COM / UNC.5 Core COM / UNC.5 40 60 80 00 0 40 Beam energy (MeV).5 3 3.5 4 4.5 5 BC angle (deg) Core emittance increase does not depend on the beam energy Core emittance increase is smaller for smaller BC bending angles 9
Measurements: optics dependence At constant CF (<0) we varied the optics along the bunch compressor Mismatch 500 300 00 00-5 -0-5 0 5 0 5 3.5.5-5 -0-5 0 5 0 5 Uncompressed +.3 +0.48 +0.40 +0.37 +0.3 +0.8 b x (m) 0 8 6 4 0 8 6 4-0.50-0.5-0.0 0 0 4 6 8 0 z (m) +0.0 +0.5 +0.40 +0.50 High sensitivity of the core emittance increase on the optics along BC Smaller emittance blow-up for a ~ a 0 +0.3, corresponding to the optics with the small b x at the two last dipoles a 0 0
Systematic checks Before proceeding we answered to these questions: Are the measurements resolution limited? The slice energy spread increases with the compression factor: are spurious dispersion and chromaticity along the single slice increasing the core emittance? Tilt in the (s,x) plane: ~0% emittance blow-up compressed beam uncompressed beam 450 350 Uncompressed V = MV V = MV V = 3 MV V = 4 MV V = 5 MV 500 300 00 Uncompressed Decompressed Compressed voltage phase de-compressed beam 300 00 50-0 -5 0 5 0 Slice emittance does not depend on the TDC voltage (V>3 MV) 0-0 -5 0 5 0 Uncompressed and decompressed bunch have similar slice emittances
Simulations scenarios Astra Elegant Astra Elegant CSRTrack Elegant 3D space charge forces No CSR D space charge forces D CSR 3D space charge forces 3D CSR
D versus 3D Astra Elegant (m) 3.5 x 0-7 3.5 Astra Elegant Option Uncompressed-H CF = 5, V CF = 5, H CSRTrack (m) 4 x 0-7 3.5 3.5 Elegant Option (3D) Uncompressed CF = 5 CF = 5.5.5 -.5 - -0.5 0 0.5.5 s (m) x 0-4 -3 - - 0 3 s (m) x 0-4 Not observed core emittance increase for the option Observed core emittance increase versus CF using 3D version of CSRTrack In the following vertical emittance equivalent to horizontal uncompressed and CF = 5 3
Dependence on BC angle Simulated at constant CF the dependency of the slice emittance versus the BC bending angle (m) 6 x 0-7 5 4 3 SIMULATIONS Vertical BC angle = 3 deg BC angle = 4 deg BC angle = 5 deg MEASUREMENTS -.5 - -0.5 0 0.5.5 s (m) x 0-4 Core emittance increase is smaller for a smaller BC bending angle 4
Dependence on beam energy Simulated at constant CF the dependency of the slice emittance versus the beam energy (m).6 x 0-7.45..95.7 SIMULATIONS Vertical E = 0 MeV E = 00 MeV E = 70 MeV E = 48 MeV 600 550 500 450 350 MEASUREMENTS Uncompressed E = 00 MeV E = 75 MeV E = 50 MeV E = 5 MeV.45. - -0.5 0 0.5 s (m) x 0-4 300 50 00 50-0 -8-6 -4-0 4 6 8 0 slice index Core emittance increase does not dramatically depend on the beam energy 5
Dependence on optics along BC Varied the optics along BC at constant compression factor x 0-7.5 SIMULATIONS Vertical -0.5 +0.0 +0.5 500 MEASUREMENTS +0.40 (m) +0.50 300.5 00 - -0.75-0.5-0.5 0 0.5 0.5 0.75 s (m) x 0-4 Qualitative agreement 00-5 -0-5 0 5 0 5 6
Dependence on incoming emittance We measured less emittance blow-up starting from a larger slice emittance uncompressed bunch We simulated bunches at the entrance of the bunch compressor with different initial core emittances 000 800 MEASUREMENTS.4 SIMULATIONS 000 600 00 Uncompressed CF =.3 750 0-3 - - 0 3 000 800 600 Uncompressed CF = / 0.. 500 50 00 0-3 - - 0 3 00 600 800 0 (nm) The larger the initial slice emittance, the smaller the relative core emittance blow-up is 0 7
Discussion The simulations using the D model of CSRTrack or Elegant do not show any core emittance increase with the bunch compression Using the 3D model of CSRTrack a core emittance increase is observed: The dependency of the core emittance increase on the different beam setup qualitatively reproduces the behavior observed at SITF, and in particular: Variable Compression factor Beam energy Chicane bending angle Optics (b x at the end of BC) Emittance growth Linear Independent Smaller for smaller bending angle Smaller for smaller b x at the last dipoles Measurements and simulations indicate CSR and 3D effects as the responsible for the observed core emittance increase 8
Minimum measured core emittance By doing a full optimization we have achieved the following emittances for the 00 pc bunch charge: Uncompressed Compressed (CF = 8) Slice emittance (nm) 99±5* 99±7* 500 300 00 Mismatch 3.5.5 Uncompressed Measurement Measurement Measurement 3 00-5 -0-5 0 5 0 5-5 -0-5 0 5 0 5 These 3 values fulfill the SwissFEL requirements * Only statistical error included 9
Conclusions Measured core emittance growth during compression at SITF, and it is: Proportional versus the bunch compression factor (constant R 56 ) Smaller for smaller chicane bending angle Independent on the beam energy Strongly dependent on the optics along the bunch compressor Strongly dependent on the optics upstream the bunch compressor Using 3D model including CSR reproduced qualitatively all the measured dependencies Excluded some possible sources for the core emittance blow-up: Resolution limit of the measurement Slice energy spread via chromaticity or spurious dispersion Horizontal tilt along the bunch CSR and 3D effects considered to be the main responsible for the observed phenomenon Measured ~00 nm slice emittance along a small fraction (flatness of the mismatch) and less than 300 nm for the majority of the bunch length 0
Acknowledgment We would like to thank all the groups involved in the SwissFEL Test Facility and you for the attention