A U.S. Department of Energy Office of Science Laboratory Operated by The University of Chicago Argonne National Laboratory Office of Science U.S. Department of Energy Beam Source Requirements for High Beam Power Accelerators John W. Lewellen Accelerator Systems Division Advanced Photon Source

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 2 Acknowledgements Charles Brau (Vanderbilt) Jym Clendennin (SLAC) Dave Dowell (SLAC/LCLS) Klaus Flöttmann (DESY) Fred Kiewit (Eindhoven) Yuelin Li (ANL) Kwang-Je Kim (ANL) John Noonan (ANL) John Petillo (SAIC) John Power (ANL) John Schmerge (SLAC) Charles Sinclair (Cornell) Frank Stephan (DESY-PITZ) Todd Smith (Stanford) Alan Todd (AES) and many others for interesting discussions on gun design & improvement Katherine Harkay (ANL) – ongoing discussions & support

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 3 Overview of the Talk Acknowledgements & Introduction -What this talk is not about -Accelerator Taxonomy -What drives the requirements? Source requirements -Storage-ring replacements -X-FELs -MW-class IR-FEL facilities Conclusions

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 4 What am I not talking about? Injector & linac technology specifics -DC vs. RF -NC vs SRF -recirculation Particular proposed future sources in any detail -provide broad brush strokes -generalization of performance figures & concepts “Operational” performance (i.e. uptime, MTBF, MTBM, etc.)  “User Perspective” (from machine builder’s standpoint)

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 5 Accelerator Taxonomy “Big Iron” accelerators -Linear colliders -Next-generation x-ray light sources “Desktop” accelerators -Electron microscopy -Electron beam lithography -Small laboratory experiments “Mini-Me” accelerators -Radioisotope generation -High-power free-electron lasers -Slow positron production -Pulse radiography Courtesy SLAC/LCLS Courtesy LLNL

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 6 What drives the requirements? Linac-based light sources -bifurcation between high-flux and high-brightness -minimum performance enhancement to be worthwhile -our ability to make magnetic structures “Mini-Me MW” FELs -wavelength -maximum desired electron beam power -size requirements

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 7 Sense of Scale: X-ray Facility Advanced Photon Source APS: 3 rd -generation storage-ring light source Circumference: ~ 1 km Beamlines: ~ 35 bending magnet ~ 40 undulator/ID Local user group offices Users per year: thousands Scheduled run hours per year: thousands

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 8 What improvements can be made? Increase the x-ray beam flux (photons / sec unit bandwidth) Increase the x-ray beam brightness (peak or average) -more “useful photons” per unit flux delivered Increase the temporal resolution (shorter x-ray pulses) -time-resolved phenomena -pump-probe experiments Coherent light production -“light bulb-to-laser” transition

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 9 Storage Rings vs. Linac-Based Light Sources Electrons circulate many times Beam power delivery =  losses Electron beam properties set by the ring (injector irrelevant) Electrons make one pass Beam power delivery = E beam ·I beam +  losses Electron beam properties set by the linac (injector is everything) Storage Ring Light Sources Linac-Based Light Sources

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 10 Storage-Ring Replacements & X-FELs “1 st -generation” Linac Based Light Sources Storage-Ring Replacement (SRR): Higher Peak Brightness Improved Temporal Resolution X-ray Free Electron Laser (X-FEL): Coherence Temporal Resolution

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 11 SRR – Performance Requirements >10 2 peak brightness improvement over existing SR sources < ps temporal resolution (i.e. bunch length) Source Norm. Transverse Emittance [  m] Bunch Charge [nC] Bunch Length [ps] Peak Current [A] Peak Brightness Enhancement Avg. Brightness Enhancement APS43 x 0.53 –  11*1* S-band Gun1 x †  4545 * New Gun (I)0.1 x  1000 †  * New Gun (II)0.1 x  1000 †  ‡ † With linac-based bunch compressor ‡ Assuming 10 mA average beam current * Assuming 100 mA average beam current

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 12 But here’s the kicker… I mean = Q b /  t where: I mean = average beam current  t = bunch-to-bunch interval Q b = single-bunch charge  t: as large as possible for timing exp. Q b : as small as possible for  n  define SRR in terms of brightness, not flux

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 13 The Pierce Parameter and X-FEL Performance The larger the Pierce parameter… the smaller the gain length the shorter the X-FEL undulator the less expensive the facility is Emittance too big? Pump up K and u … but there’s a price.  -2  u for fixed, so slow progress

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 14 X-FEL – Performance Requirements Maintain LCLS wavelength performance Minimize overall systems cost Fixed Parameter = x-ray wavelength = 1.5 Å Parameters we can adjust u = undulator period > 1¼ cm K = undulator strength ~ 1 Parameter to Minimize  = electron beam Lorentz factor With existing undulator technology, need about a 4 GeV beam to make 1.5 Å x-rays … assuming we can get the gain.

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 15 X-FEL Design Optimization… Assume: Existing undulator technology Existing linac technology “Classic” SASE design

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 16 Summary: SRR and X-FEL sources Single-bunch parameters for X-FEL and SRR sources are remarkably similar: ~ 0.1  m normalized emittance (or better) ~ 1 kA peak beam current (post-compressor) SRR will need 10 – 100 mA beam current to maintain reasonable average brightness X-FELs might want long inter-pulse spacing to change out samples

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 17 A Word on Linear Collider Guns… Why are these desired? -Damping rings are expensive as well as challenging -Makes most sense when LC combined with X-FEL Basic parameters not dissimilar from X-FEL guns Additional requirements: -flat-beam production (100:1 emittance ratios) -polarized electron beam production A positron gun would be nice…

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 18 IR-FEL – Driving Parameters “1-ish”  m operating wavelength ~ 100 MeV max. beam energy 1 MW optical power output Notional FEL extraction efficiency: ~ 1%  1-A average beam current  100 MW circulating e-beam power Large outcoupling (50-ish %)  reduce circulating optical power (save the mirrors)  high gain per pass  fairly high beam quality

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 19 High-power IR and UV FELs Injector 7 MeV 50 MeV MeV Beam Parameters I peak ~ 500 – 1000 A  n ~ 5 – 10  m  E/E~ 0.1 – 0.25% Electron beam power: 100 MW Optical beam power: MW RF power used: 7 MW (to dump)Average beam current: 1A Wallplug efficiency: ~ 10 – 20% : 2 MW (FEL)

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 20 Source Parameter Estimation For the moment, assume no “exotic” designs -no tapered undulator -no linac/wiggler combinations -etc. Assume some “reasonable” undulator parameters -  u ~ 3 cm -K ~ 2  ~ 10 m Assume some “bulk” beam properties -E beam ~ 100 MeV -  E/E ~ 0.25% (rms)

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 21 What do we get…?  x = mm Spot size in undulator is reasonable; no exotic focusing schemes needed r =  m Laser wavelength is in the desired range L g = m Gain length (1 e-folding length) is a reasonable match for a few-m undulator in a high-loss cavity.  nx = 10  m, I p = 1 kA Adjust to get reasonable L g

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 22 Summary – MW-class ERL-FEL Sources 1-A average beam current; Q b = 1 amp / f rf Transverse emittance: ~ 10  m (rms ~ 1.5 nC / bunch Longitudinal emittance: ~ 100 keV ps (rms)

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 23 General Source Requirements

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 24 A Word on Recirculation Losses… Megawatt-class IR FEL -1 A x 100 MeV = 100 MW -1 kW losses = P beam x = 0.001% -1 W loss = ≈ 90 electrons / bunch (from 1.4 nC) Storage-Ring Replacement X-ray Source -100 mA x 5 GeV = 500 MW -1 kW losses = % beam loss -1 W loss = 2·10 -9 ≈ 1¼ electrons / bunch (from 100 pC) Controlling halo no longer means winning an X-Box game. Source parameters must be consistent with zero-loss operation.

Pioneering Science and Technology Office of Science U.S. Department of Energy John Lewellen8 – 10 November 2004Workshop on High-Power Beams 25 Summary X-FEL wants: -order-of-magnitude improvement in emittance MW IR-FEL wants: -true CW operation -high beam currents (ca. 1 A) at modest energy -~ recirculation losses SRR wants: -true CW operation -high beam currents (ca. 10 – 100 mA) at 7-ish GeV -order-of-magnitude improvement in emittance -True zero-loss recirculation