Electro-Optic Longitudinal Profile Diagnostics for CLIC Electro-Optic Longitudinal Profile Diagnostics for CLIC Current Status & Future Development S.P. Jamison, Accelerator Science and Technology Centre, STFC Daresbury Laboratory W.A. Gillespie, Carnegie Laboratory of Physics, University of Dundee A.M. MacLeod, University of Abertay Dundee G. Berden, FELIX, ‘Rijnhuizen’ FOM, Netherlands Steven Jamison, CTC, CERN September 21, 2010

Electro-Optic Techniques... spectral upconversion FELIX FELIX DESY RAL(CLF) MPQ Jena,... Temporal Decoding SLAC DESY LBNL,... Spatial Encoding FELIX DESY LBNL... Spectral Decoding several different techniques same underlying physics of electron-beam interaction differing optical implementation differing capability.... our CLIC proposal Temporal decoding for resolution down to ~60fs Spectral up-conversion to be further developed & demonstrated for shorter time scale system Choices available, based on time resolution & simplicity /robustness Steven Jamison, CTC, CERN September 21, 2010

Low time resolution... Attractive simplicity low time resolution measurements e.g. injector diagnostics Rely on t- relationship of input pulse for interpreting output optical spectrum Resolution limits come from fact that EO- generated optical field doesn't have same t- relationship Spectral Decoding Resolution related to laser parameters... BUT “ideal” laser systems still limited to > 1.0 ps resolution (for all reasonable observation windows) Laser requirement can be modest Suitable turn-key commercial systems widely available (fibre laser) Steven Jamison, CTC, CERN September 21, 2010

Laser requirement can be modest Suitable turnkey commercial systems widely available (fibre laser) Could be developed to directly link to the optical timing distribution low charge, low energy measurement on ALICE (Daresbury) 40pC, single shot, Spectral decoding - majority of expts with complicated Ti:S systems - can be done with freq. doubled Erbium fibre lasers (if “pulsed” clock distribution employed) low energy injector, >> 1ps diagnostics - Robust stand-alone Yb-fibre laser systems being developed by PSI & DESY Low time resolution... Spectral Decoding Steven Jamison, CTC, CERN September 21, 2010 Compact fiberised assembly at PSI (from Steffen et al. DIPAC09)

Explicit high time resolution approaches... Temporal Decoding Two distinct techniques demonstrated Profile retrieval not yet adequately demonstrated - complications of specific test Potential for moderately complex laser systems (fibre) Potential for all-optical integration with timing distribution system Demonstrated 120fs FWHM, 60fs rms resolution Benchmarked against RF deflecting cavity Requires high pulse energy laser system (>50 uJ per pulse) capable of explicit temporal profiles with features down to 120fs FWHM...will return to definition of time resolution Further improvements face significant issues laser: - generation & transport of <30fs pulses (physics & engineering) - ultrafast optical characterisation (physics) non-linear materials: bandwidth limitations (physics) Spatial Encoding Steven Jamison, CTC, CERN September 21, 2010

Electro-optic experiments on FLASH... Remote adjustment & alignment control required Accelerator location required (ultrafast laser transport) reliable remote diagnostic alignment achievable goal) true turn-key laser systems not available now. …CLIC time scale..? solution for high-power ultra-short laser transport..? Steven Jamison, CTC, CERN September 21, Vacuum transport of Ti:S to tunnel - Hands-on operation of amplified Ti:S CLIC context...

observing stabilising effect of slow feedback CDR feedback onCDR feedback off realtime control-room profiles high time resolution (+ wakefields ?) 65  m thick GaP Electro-optic experiments on FLASH Control-room diagnostic… BUT only with laser room preparation and support Steven Jamison, CTC, CERN September 21, 2010

Temporal decoding time resolution comparison with RF deflecting cavity Electro-optic Transverse Deflecting Cavity Phys Rev Lett 99, (2007) Phys. Rev. ST, 12, (2009) shot-shot variations not observable in (existing) EO... Steven Jamison, CTC, CERN September 21, 2010 Resolution limited by both material properties & laser pulse duration

Coulomb field of relativistic bunch probe laser Decoding of information from laser pulse non-linear crystal (electro-optic effect) Encoding choices for complexity limiting factor for “spectral decoding” Temporal resolution limits Temporal resolution limits are primarily in material properties Laser and “decoding” limits follow not far behind What are current limits? How to overcome them. Encoding process physics: non-linear frequency mixing between Coulomb field and laser phase-velocity matching of full Coulomb spectrum with laser (refractive index variations near phonon resonances ) Constraints: Decoding physics: ultrafast optical pulse characterisation Constraints:ultrafast laser generation & transport EOTD & EOSD Steven Jamison, CTC, CERN September 21, 2010

Coulomb spectrum shifted to optical region Coulomb pulse replicated in optical pulse envelopeoptical field Physics of EO encoding... time varying refractive index is a restricted approximation to physics New concepts & understanding of very high time res techniques come from generality of frequency mixing physics description) Jamison et al. Opt. Lett (2006)Jamison., Appl.Phys B. Laser & Opt (2008) Steven Jamison, CTC, CERN September 21, 2010

Encoding Time Resolution... material response R() ZnTe Velocity mismatch of Coulomb field and probe laser Frequency mixing efficiency,  (2) () GaP Phonon dips...missing data Origin of ringing artefacts Steven Jamison, CTC, CERN September 21, 2010

Time resolution from frequency response Encoding “time resolution” not a RMS measure... Better described as “temporal limitation” structure on longer time scale recorded faithfully short time scale structure not observed, PLUS ringing artefacts Steven Jamison, CTC, CERN September 21, 2010

50 fs FWHM bunches, separated by 110fs practical problems with 10um thickness ringing not solved by thinner crystals Laser pulse duration NOT included in calculations At sub-50fs resolution, short pulse preservation needs to be addressed 100 um thick GaP 50 um thick GaP 10 um thick GaP...Time resolution from frequency response Steven Jamison, CTC, CERN September 21, 2010

achieving even better resolution...? Detector Material: –Move to new material? ( phase matching,  (2) considerations ) –Could use GaSe, DAST, MBANP..... or poled organic polymers? –use multiple crystals, and reconstruction process –Introduce shorter pulse lasers (<30 fs) –Use (linear) spectral interferometry –Use FROG Measurement (initially attempted at FELIX, 2004) Encoding Decoding or Alternative techniques: spectral upconversion If drop requirement for explicit time information at high frequencies, other options also become available Steven Jamison, CTC, CERN September 21, 2010

polymer (MA9 in PMMA) alternative EO materials.. from Cao et al Opt Lett (2002) alternative materials exist with higher frequency response all materials have resonances & bandwidth limitations Leads to consideration of multi-crystal EO detectors from Ding et. al, J. non-linear Opt. materials (2003) GaSe varying crystal angles Explicit temporal reconstruction of combined crystal measurements…? removal of “ringing” from individual crystals ?? Steven Jamison, CTC, CERN September 21, 2010

Spectral upconversion diagnostic... accepting loss of phase information & explicit temporal information... gaining potential for determining information on even shorter structure... gaining measurement simplicity use long pulse, narrow band, probe laser same physics as “standard” EO.... different observational outcome. Two approaches.. Temporal decoding for ~60 fs resolution Higher resolution with alternative materials & reconstruction Explicit temporal information Implicit spectral information from spectral-upconversion Steven Jamison, CTC, CERN September 21, 2010 Full spectral information, [mm wavelength -> optical ] Robust capability [feedback..?] laser complexity reduced, reliability increased laser transport becomes trivial (fibre) problematic artefacts of spectral decoding become solution

What’s different...? Spectral decoding & Temporal decoding Spectral upconversion laser bandwidth bunch spectral extent >> laser bandwidth bunch spectral extent << issues of laser transport laser complexity/expense/reliability material effects (e.g. group velocity dispersion) low power, ‘simple’ lasers OK fibre transport now an option simple – linear – spectral detection without artefacts of spectral decoding Important: can measure non-propagating long wavelength components not accessible to radiative techniques (CSR/CTR/SP) Steven Jamison, CTC, CERN September 21, 2010

Spectral upconversion diagnostic experiments at FELIX Free-space optical transport Synchronised 5-50ps Ti:S and YAG lasers For these experiments… Steven Jamison, CTC, CERN September 21, 2010

Spectral upconversion diagnostic experiments at FELIX difference frequency mixing sum frequency mixing Appl. Phys. Lett (2010) FELIX temporal profile inferred FELIX spectrum

Spectral upconversion with FEL radiation... Explicit confirmation of frequency upconversion process with tunable, monochromatic, Far-IR and mid-IR optical side bands from =150m FEL radiation High fields effects !!! Appl. Phys. Lett (2010) Standard “EO” expectation x10 increase in field (…charge)

Proposed spectral upconversion system turn-key commercial fibre lasers fibre optic distribution to diagnostic location in accelerator fibre optic transport to optical spectrometer Compact crystal assembly in beam line multiple crystals azimuthally offset laser split and independently sent to each crystal commercially available link to telecoms technology (robustness) Required development: Short bunch testing of spectral upconversion Testing of alternative materials, including in accelerator environment Simultaneous testing of multiple-crystals Steven Jamison, CTC, CERN September 21, 2010

Pathway to meeting CLIC requirements Materials investigation Testing with laser generated THz (material properties) FEL experiments for non-linearity and mid-IR capability EO technique development and demonstration: Reducing laser requirements of EOTD FROG extension of EOTD…allows sub-laser duration measurements Multi-crystal reconstruction methods…theory -> expt test Spectral upconversion High field effects… Tests at short-bunch test facilities (PSI, FACET, …?) Steven Jamison, CTC, CERN September 21, 2010

Summary Low time resolution (>1ps structure) spectral decoding offers explicit temporal characterisation relatively robust laser systems available High time resolution (>60 fs rms structure) proven capability significant issues with laser complexity / robustness Very higher time resolution (<60 fs rms structure) Limited by EO material properties (& laser) Spectral up conversion + alternative (multiple) EO materials gives implicit ultra-short bunch structure Development & short-bunch demonstrations required (including alternative materials, GaSe,...) Steven Jamison, CTC, CERN September 21, 2010

Current capabilities & options Readiness for implementation Outstanding issues Development & Prototyping Pathways time resolution vs simplicity multi-bunch operation integration in optical timing distribution systems Steve Jamison, CTC, Sept 2010 Electro-Optic Longitudinal Profile Diagnostics for CLIC Steven Jamison, CTC, CERN September 21, 2010

Spatial Encoding Rely on t-x relationship between laser and Coulomb field In principle: expect same/similar capabilities as TD from Cavaleria et al. PRL 2005 SPPS (SLAC) measurements less widely demonstrated: from A. Azima et. al EPAC06 FLASH (DESY) measurements SLAC and DESY expts had significant additional complications of long transport in fibre... possible solution for simplifying laser requirements BUT.... cannot improve on temporal decoding not upgradable to FROG-like measurement without adding TD complexity (limited to laser pulse duration  issues of laser transport) + system under test (SPARC...)

 (2) (  thz,  opt )  opt +  thz convolve over all combinations of optical and Coulomb frequencies. includes field phase (chirp), general phase matching, optical GVD etc Electro-optic encoding is a consequence of sum- and difference-frequency mixing  thz  opt  opt -  thz  opt Frequency domain description of EO detection... EO crystal Coulomb field probe laser Refractive index formalism comes out as subset of solutions (restriction on laser parameters) for arbitrary probe and Coulomb pulses...

alternative ways forward... Current limitations are from material properties Phonon-resonance at 3-15 THz (material dependent) All materials will have some phonon resonance effects Can we use a set of crystals to cover larger range? requires (uncertain) reconstruction to find temporal profile (relative phase shifts, phase matching, efficiency between crystals) complication of system would multiply If reconstruction needed anyway, reconsider spectral techniques... BUT traditional spectral techniques have difficulties : long wavelength / DC component transport extreme (“100%”) spectral bandwidths for detection A solution : Electro-optic spectral upconversion

Steve Jamison, CTC, Sept 2010

Laser transport... in FLASH experiments EO Temporal decoding setup Vacuum laser transport from laser “hut” to tunnel. Some relay lenses in transport Remote adjustments on many mirrors More complicated setup Less access to setup in accelerator

Spectral upconversion diagnostic Aim to measure the bunch Fourier spectrum accepting loss of phase information & explicit temporal information... gaining potential for determining information on even shorter structure... gaining measurement simplicity use long pulse, narrow band, probe laser laser complexity reduced, reliability increased laser transport becomes trivial (fibre) problematic artefacts of spectral decoding become solution  -function NOTE: the long probe is still converted to optical replica same physics as “standard” EO different observational outcome new slide 20

In Summary... Proven capability for explicit temporal characterisation up to ~100 fs rms electron bunch structure high time resolution techniques have problems with reliability & necessary infrastructure but there exist avenues available for improving time resolution & robustness (depending on beam diagnostics requirements) For high time resolution, both alternative materials & spectral upconversion are under investigation

Selected References (Daresbury-Dundee Group): Free-electron laser pulse shape measurements with 100fs temporal resolution using a 10fs Ti:sapphire laser and differential optical gating X.Yan, A.M. MacLeod, W.A.Gillespie et al. Nucl. Instr. Meths. Phys.Res. Vol A429 (1999) 7- 9 Sub-picosecond electro-optic measurement of relativistic electron pulses X. Yan, A.M. MacLeod, W.A. Gillespie, G.M.H. Knippels, D. Oepts, A.F.G. van der Meer. Physical Review Letters 85 (2000) Single-shot electron bunch length measurements I. Wilke, A.M. MacLeod, W.A. Gillespie, G. Berden, G.M.H. Knippels, A.F.G. van der Meer Phys. Rev. Lett. 88 No 12 (2002) /1-4 Real-time, non-destructive, single-shot electron bunch-length measurements G. Berden, S.P. Jamison, A.M.MacLeod, W.A. Gillespie, B. Redlich and A.F.G. van der Meer Physical Review Letters 93 (2004) Temporally resolved electro-optic effect S.P.Jamison, A.M. Macleod, G. Berden, D.A. Jaroszynski and W.A. Gillespie Optics Letters 31, 11 (2006) Benchmarking of Electro-Optic monitors for Femtosecond electron bunches G. Berden, W.A.Gillespie, S.P. Jamison, B. Steffen, V. Arsov, A.M. MacLeod, A.F.G. van der Meer, P.J. Phillips, H. Schlarb, B. Schmitt, and P. Schmüser Phys. Rev. Lett (2007) Electro-optic time profile monitors for femtosecond electron bunches at the soft X-ray free-electron laser FLASH B. Steffen, V. Arsov, G. Berden, W.A. Gillespie, S.P. Jamison, A.M. MacLeod, A.F.G. van der Meer, P.J. Phillips, H. Schlarb, B. Schmitt, and P. Schmüser Physical Review Special Topics – Accelerators and Beams (2009)

Temporal decoding measure optical replica with t-x mapping in 2 nd Harmonic Generation Temporal profile of probe pulse  Spatial image of 2 nd harmonic limited by gate pulse duration (although FROG etc could improve) EO encoding efficiency, phase matching Rely on EO crystal producing a optical temporal replica of Coulomb field Practical limitations: complexity of laser systems transporting short pulse laser (gate pulse only)