Electro-Optic Longitudinal Bunch Profile Measurements at FLASH: Experiment, Simulation, and Validation Bernd Steffen, DESY FEL 2007 Novosibirsk, August 29th 2007
Electro-Optic Bunch Length Detection fs laser scanning delay P EO λ/4 λ/2 P photo detector - Coulomb field of electron bunch induces birefringence in EO crystal. - birefringence is sampled by Ti:Sa laser pulse.
The Electro-Optic Effect: Coulomb Field induced Birefringence Y=[0,0,1] EO crystal Phase retardation (in the small signal limit): n 1 n 2 E Coulomb E Laser (1,1,0)-plane X=[-1,1,0]
Effect of the half wave plate θ=0 θ=2 I det,v /I laser 0.03 0.025 0.02 0.015 0.01 signal θ=2 0.005 bg θ=2 signal θ=0-0.2-0.1 0 0.1 0.2 Γ in rad
Response function of the GaP crystal 0.25 0.2 d=65 μm 175 μm 300 μm G(f) [pv/m] 0.15 0.1 0.05 0 0 5 10 15 20 frequency [THz] EO coeff., transmission, velocity matching Signal distortion esp. for thick crystals!
The simulation program Effective Coulomb pulse calculated from electron bunch profile and EO response function Phase retardation Γ from effective THz pulse Complex electric field of the modulated stretched laser pulse calculated according to: Temporal and spectral intensity in both polarisations can be calculated.
Simulated EOTD signals of Gaussian THz pulses in 65 µm GaP 1.5 Γ sim E THz 1 σ = 10 fs σ = 30 fs σ = 90 fs Γ 0.5 σ =?? σ = 48 fs σ = 94 fs 0 0.2 0 0.2 0.4 0.6 time in ps 0.2 0 0.2 0.4 0.6 time in ps 0.2 0 0.2 0.4 0.6 time in ps All pulse length: σ of a fitted Gaussian
FLASH Free-Electron Laser in Hamburg RF gun BC BC 450-700 MeV ACC1 ACC2 ACC3 ACC4 ACC5 CTR beamline OTR screen for TDS Bypass Undulator 5-200 µj 13-47 nm TDS (LOLA) EOS BAM Collimator TEO Dump Longitudinal Diagnostics Electron bunches: 30 fs duration 700 MeV electron energy 0.5-1 nc charge 1 ka peak current
Bunch length measurements using the transverse deflecting structure (TDS) vertical RF field E y (t) fast horizontal kicker ee σ z 2 3.66 m head tail ~σ z vertical streak and horizontal kick 200 Resolution: approx. 20 fs at a time window of 2 ps x [mm] ρ/ρ max 0-2 -4 1 0.5 0 0 0.5 1 1.5 2 t [ps] 0 0.5 1 1.5 2 t [ps] image charge density σ spike 30 fs Q spike 0.12 nc (23 %) 150 100 50 0
fs laser σ 0 EO Spectral Detection optical stretcher σ c P EO A grating CCD Linear relationship between wavelength and long. position in laser pulse ( linear chirp ) Bunch profile is transferred to spectral profile of the laser pulse - Problem: Frequency mixing with Coulomb field creates new frequency components: Distortions at large chirp α 1/ σ0 σc σmin σ0 σ c 2.6 200 fs (for Gaussian pulses!)
Spectrally resolved detection: Comparison of measured to simulated Signals 5 consecutive bunches, corrected for different arrival times Simulation: EOSD signal of a bunch measured with TDS EOSD signal I/I laser 0.02 0.01 σ = 210 fs simulation measurements 1st reflection GaP 175 µm θ=2 σ 0 =6 fs σ c =1.5ps 0 0 2 4 time [ps] Excellent agreement with simulation in shape and amplitude, but much wider than electron bunch due to response function and frequency mixing
EOSD: Distortions due to frequency mixing for thin crystal and large chirp arb. units 2.5 2 1.5 1 0.5 0 TDS trace E Coulomb, eff EO simulated -bg GaP 175 μm thick Laser σ 0 =6 fs chirped to: σ c =1.5 ps θ=2-1 -0.5 0 0.5 1 1.5 2 2.5 time [ps]
EO Temporal Detection σ 0 fs laser fixed delay σ c optical stretcher P EO A BBO CCD EO pulse BBO crystal - Single shot cross-correlation with fs pulse in a frequency doubling crystal (BBO) - approx. 100 µj pulse energy necessary for 10 ps time window gate pulse time
EO Temporal Detection EOTD signal 6 4 2 σ = 80 fs θ = 0 2 EO signal μ E Coul 0 4 6 8 10 12 14 16 time [ps] EOTD signal 10 5 σ = 95 fs θ = 1 EO signal μ E Coul 0 4 6 8 10 12 14 16 time [ps]
Comparison of EOTD vs. TDS measurements EOTD and TDS signal TDS raw signal EOTD raw signal background -6-4 -2 0 2 4 time [ps] TDS corrected signal GaP, d=65µm EOTD corrected signal -0.5 0 0.5 1 time [ps] - 10th bunch in bunch train: electro-optic detection - 11th bunch: TDS
Comparison of EOTD vs. TDS measurements 20 fs TDS signal 26 fs THz pulse EOTD signal 55 fs measurement 43 fs simulation 43 fs EO signal simulation -0.2 0 0.2 0.4 Time [ps] -0.2 0 0.2 0.4 Time [ps] - Good agreement between measurement and simulation - close to the resolution limit of GaP Signal due to wake fields?
TDS and EOTD measurement of overcompressed bunches TDS (a) EOTD and TDS signal EOTD In good agreement with the electron bunch shape -0.5 0 0.5 1 Time [ps] Signal due to wake fields?
Conclusions Benchmarked EO detection against TDS Simulations based on published material data are consistent in shape and amplitude with measured signals for GaP EOTD signals measured with of 55 fs (rms) length (linear in field and without deconvolution!) are close to the resolution limit of GaP
Thanks to G. Berden, A.F.G. van der Meer (FELIX) S. Jamison (ASTeC, Daresbury Laboratory) P.J. Phillips, W.A. Gillespie (University of Dundee) A.M. MacLeod (School of Computing and Advanced Technologies, University of Abertay, Dundee) V. Arsov, E.-A. Knabbe, H. Schlarb, B. Schmidt, P. Schmüser (DESY)