Available and Future Laser Pulse Shaping Technology Reality

Available and Future Laser Pulse Shaping Technology (Reality and Future Directions for Spatio-Temporal Laser Pulse Shaping) T. QUAST, HELMHOLTZ-ZENTRUM BERLIN Bild von 3 d Ellipsoid courtesy of T. Rao courtesy of L. Hein Future Light Sources Workshop 2012, JLab A. Neumann, SRF 2011 Chicago, FRIOA 07 1

Laser Pulse Shaping – why do we need it ? Space Charge. . best solution: Space charge forces are linear for 3 d ellipsoid: Space charge force of a Gaussian distribution „laser world is gaussian“ 2 nd best solution: Transversal flattop + Longitudinal flattop „beercan“ T. Quast, Future Light Sources Workshop 2012, JLab 2

Introduction into Laser Shaping THREE „STAGES“ OF LASER PULSE SHAPING Transversal + Longitudinal „easy“ ≠ Spatio-Temporal (3 d) „very ambitious“ „advanced“ Difficulty level transversal I(r) 3 Quality gain ++ Stability Good – if reasonably done Good - Longitudinal I(z) 7 + (with feedback control) spatio-temporal I{r(z)} T. Quast, Future Light Sources Workshop 2012, JLab 10 ? Poor – relying on nonlinear effects 3

Spatial Shaping TRANSVERSAL SHAPING – EVERY ONE DOES IT. . Gauss to flattop – the simple way: principle pro con 1. Transport laser to vincinity of gun 2. Overilluminated iris cuts out only the inner „flat“ part 3. Iris is imaged onto the cathode 1. High position Stability (iris located near cathode) 2. Robust and simple setup 3. Spot size and laser power can be varied easily (but not independent) 4. Any pattern can be imaged onto cathode 1. what a waste of laser power ! 2. Problems from laser transmission beamline (spot size / shape stability) are transfomed into intensity fluctuations Has to be done carefully T. Quast, Future Light Sources Workshop 2012, JLab 4

Transversal Flattop – Pi Shaper ASHERIC OPTICS – PI SHAPER - aspheric refractive (reflective) elements - High transmission (90%) - very sensitive to input laser parameters - tilt, decenter, size (few mrad, 10 ths of µm) - TEM 00 mode required (difficult with UV) Use both: „mild“ shaper and pinhole w=1. 1 w 0 w=w 0 T=70% w=0. 9 w 0 from G. Klemz, I. Will, Proc. FEL 06 alternative: deformable mirror and genetic algorithm T. Quast, Future Light Sources Workshop 2012, JLab 5

Longitudinal shaping – different methods 1. Direct space to time (with grating and mask) Works well but transmission 10 E-2 2. DAZZLER - (Acousto Optic Programmable Dispersive Filter) - only up to 100 k. Hz Rep. rate (this rules out many of the existing bunch patterns) - shaping up to a few ps (restricion from possible crystal length) T. Quast, Future Light Sources Workshop 2012, JLab 6

Longitudinal shaping II 3. Spatial light modulator (SLM) Practically only for fs pulses PULSE STACKING … (LARGE VARIETY) 4. Pulse stacking with polarizer (pol. beam splitters) Courtesy of S. Schreiber, DESY • difficult geometric alignment • intensity variations due to imperfect polarizers T. Quast, Future Light Sources Workshop 2012, JLab 7

Longitudinal shaping III Birefringent crystals – reduced complexity with Linear setup from H. Tomizawa, Rad. Phys. Chem 80, 10 (2010) T. Quast, Future Light Sources Workshop 2012, JLab 8

Longitudinal shaping IV 3 STAGE STACKER W. BIREFRINGENT CRYSTALS 3 x YVO 4 d=(24, 12, 6 mm) T= 62%; 532 nm from: A. K. Sharma et al. PRSTAB 12, 033501 (2009) T. Quast, Future Light Sources Workshop 2012, JLab 9

Longitudinal shaping V HIGH PRECISION PULSE SHAPER (MBI) Taken from: Will, Klemz, Optics Express 16 (2008) , 4922 -14935 Theory for N = 10 crystals: 1024 components aranged in 11 groups T. Quast, Future Light Sources Workshop 2012, JLab 10

shaper for high resolution 13 CRYSTAL PULSE SHAPER FOR HIGH RESOLUTION motorized rotation stage Shaped ouput pulses Gaussian input pulses Laser pulse shape measured: edge 10 -90 ~ 2. 2 ps FWHM = 25 ps temperature controlled birefringent crystal OSS signal (UV) edge 10 -90 ~ 2 ps birefringent shaper, 13 crystals Will, Klemz, Optics Express 16 (2008) , 4922 -14935 T. Quast, Future Light Sources Workshop 2012, JLab 11

Pulse shaping DIFFERENT POSSIBLE PULSE SHAPES Gaussian: Flat-top: FWHM ~ 2 ps FWHM ~7 ps FWHM ~ 11 ps FWHM ~ 17 ps FWHM ~ 24 ps FWHM ~ 19 ps Feedback with optical sampling system (OSS): Simulated pulse-stacker - dynamic range of streak camera not sufficient - scanning of 100 subsequent pulses (~0. 2 ps res. ) without feedback - shaping is done after oscillator in IR FWHM - sampling for feedback signal in the UV ~ 24 ps courtesy of I. Will, MBI T. Quast, Future Light Sources Workshop 2012, JLab 12

Self evolving SELF EVOLVING BEAM - SPACE CHARGE FORCE • start with a parabolic (or half sphere) Laser intensity profile • automatic evolution into a uniform ellipsoidal (3 D) beam pro: - Easy - no longitudinal laser shaping - only a short (100 fs) pulse (clipped gaussian) needed con: - cannot put high charge in it - short pulse may damage cathode - only fast response photocathode material => metal - requires high accelerating gradient O. J. Luiten et al. , Phys. Rev. Lett. 93, 094802 (2004) T. Quast, Future Light Sources Workshop 2012, JLab 13

3 d pulse shaping USE CHROMATIC ABERRATION OF A DISPERSIVE LENS • Refractive index n is a function of frequency (dispersion) • Focal length of focussing lens changes with frequency • Parabolic frequency change by giving the pulse a cubic phase spatial shaper Zn. Se lens achromatic lens camera DAZZLER Courtesy of Yuelin Li T. Quast, Future Light Sources Workshop 2012, JLab 14

3 d pulse shaping A FIRST PROOF OF PRINCIPLE EXPERIMENT In principle it is working -quality suffers from AOPDF limitations - No pinholes in transport ! (changing size) - no dispersion in transport ! - only IR so far (conversion ? ? ) From Y. Li et al. PRSTAB 12, 020702 (2009) T. Quast, Future Light Sources Workshop 2012, JLab 15

3 d pulse shaping (alternative) STACKING A 3 D-ELLIPSOID… con • Alignment ? • Coherence and diffraction ? • Slice number limited pro • No dependance on nonlinear effects Not been demonstrated yet Z. He et al. Proc of PAC 2011, TUP 200 T. Quast, Future Light Sources Workshop 2012, JLab 16

…back to Reality… SURVEY ON LONGITUDINAL SHAPING JLab Information provided by Simulations Conclusion Reality Experim ent Observation Shukui Zhnag, Pavel Evtuschenko Gaussian 3 s Better for longitudinal emittance Gaussian yes Regular operation, optimal longitudinal emittance for FEL Gaussian with new laser, pulses shorter than old yes Reduced gun voltage, new injector settings => reduced longitudinal brightness Pulses stretched in 3 birefingent crystals yes Stretcher almost never stable; changes in pulse shape made optimization hard; diff. injector setup necessary; after optimization same long. brightness as before Laser modified for longer Gaussian bunches Superposition of 2 n G. tails 3 s. min. ripple center Better for transverse emittance Real longitudinal profiles streak camera pictures approx. by computer code Necessary for agreement between simulations and measurement T. Quast, Future Light Sources Workshop 2012, JLab 17

…reality II… SURVEY (CONT. ) Information provided by Alice Yuri Saveliev, Boris Militsyn SPARC Massimo Ferrario HZB Thorsten Kamps, Bettina Kuske Simulations Conclusion Reality Alive studies Paper in Boris’ email only gun beam line 7 ps rms Gaussian 2 -16 -2 ps flat top Experim ent Observation Pulse stacker (4 x 7 ps) 28 ps FWHM Near future 7 ps Gauss pulse stacker incl. recirculato r Potential benefits wiped out by nonflatness Flat top pulses yes Benefits in transverse emittance for charges above 300 p. C At beginning, transverse nonlinearities reduced & frozen Blow out regime profitable for long. phase space for compression Laser comb Gaining importance THz sources FEL plasma acc. Longitudinal: current distribution similar, Energy spread larger in FT Transverse: low charge slices over-focussed T. Quast, Future Light Sources Workshop 2012, JLab 18

pulse shaping SUMMARY • transversal pulse shaping has to be done carefully • with advanced pulse shaping the beam transport becomes an issue (dont mess it up. . ) • advanced and controlled pulse stacking setup for flat top laser pulses work nicely • first steps into 3 d ellipsoidal shaping are done – further exploration needed REMARKS § § § Overall performance of a gun largely depends on careful technical implementation of the Ph. Cath Laser Stable laser parameters improve the overall performance Put more emphasis on laser diagnostic and feedback More complicated shaping schemes, utilizing nonlinear effects causes unwanted coupling of parameters o Less degrees of freedom o Potential source of instability Is it worth it ? T. Quast, Future Light Sources Workshop 2012, JLab 19

Z-polarization gun - laser induced shottky effect z-polarization (interesting concept) workfunction is lowered of the intense laser field -Requires very moderate laser parameters: 2. 6µJ, 100 fs, ~800 nm -Focussing down to 20µm results in 21 p. C „If the Schottky-effect-induced Z-field is large enough, we expect that electrons will make oscillations with the Zfield frequency on the outermost surface of the metal cathode and will be extracted with the external electric field of the RF cavity“ H. Tomizawa, Proc FEL 2010 T. Quast, Future Light Sources Workshop 2012, JLab 20