Low charge, ultra-short beams for FEL and Advanced Accelerator applications

J.B. RosenzweigUCLA Dept. of Physics and Astronomy

ICFA Workshop on the Physics and Applications of High Brightness Electron BeamsMaui, November 19, 2009

Outline

• Proposal for ultra-short beams in SASE FEL• Ultra-high brightness electron beams at low Q• Single spike operation• Breaching attosecond, short wavelength frontier

• Coherent radiation from ultra-short beams• Scaling the PWFA to short wavelength• TV/m PWFA experiment at the LCLS

Brightness Begets Brightness: Electrons and Photons

• Revolution in light sources due to electron beam improvements

• Two orders of magnitude produce qualitatively new light source, the X-ray Free-electron Laser – Intense cold beam; instability

• To obtain needed e-beam brightness, must compress to <100 fs

• Ultra-bright, Å, coherent, sub-psec light source– Many orders of magnitude; LCLS is undeniably real…– Investigate atomic/molecular structure at length-time

scales of ion motion

Be 2I

n2 21014 A/m2

Ultra-short XFEL pulses: motivation• Investigations at atomic electron spatio-temporal scales

– Angstroms-nanometers (~Bohr radius)– Femtoseconds (electronic motion, Bohr period)

• Femtochemistry, etc. • 100 fs accessible using standard techniques• Many methods proposed for the fsec frontier

– Slotted spoiler; ESASE; two stage chirped pulse– Unsatisfactory (noise pedestal, low flux, etc.)– Still unproven

• Use “clean” ultra-short, low charge electron beam– Myriad of advantages in FEL and beam physics

• Mitigate collective effects dramatically– Robust in application: XFEL, coherent optical source, PWFA…

Ultra-high brightness beams: How?

• Brightness• Low Q in injector: shorter st, smaller e• Velocity bunching at low energy, recover I• Chicane bunching (1 or 2 stages)• The rules change much in our favor

– Low charge makes all manipulations easier– Higher brightness gives new possibilities, design

application with open imagination… • Illustrate 1st with original example: SPARX

B2I

n2

Q

tn2

Photoinjector scaling• Beam at lower energy is single

component relativistic plasma• Preserve optimized dynamics:

change Q, keeping plasma frequency (n, aspect ratio) same

• Dimensions scale– Shorter beam…

• Emittances:

• At low Q, x,th dominates (Ferrario WP, SPARC/LCLS)

i Q1/ 3

a1 0.11

a2 0.18

a3 0.23

n mm-mrad a1Q nC 2 / 3a2Q nC 4 / 3

a3Q nC 8 / 30.33Q nC 1/ 3

mm-mrad

J.B. Rosenzweig and E. Colby, Advanced Accelerator Concepts p. 724 (AIP Conf. Proc. 335, 1995).

Bunching for high currrent• Magnetic recently enhanced by velocity bunching

• Velocity bunching avoids coherent beam degrading effects in bends

• Also with low charges…

Longitudinal phase space schematic for velocity bunching

Recent velocity bunching results from SPARC (INFN-LNF, Frascati

Velocity bunching• Enhance current at low energy, avoid bending• Inject near zero crossing• Apply optimized focusing, manage e

evolution

Longitudinal phase space schematic for velocity bunching

Recent SPARC results (accepted in PRL)

VB example: SPARX (2008), 1 pC

- e growth manageable- 1 order of magnitude compression- “Thermalized longitudinal phase space;

limits subsequent compression

Beam transverse size, emittance Beam length during velocity bunching

Final longitudinal PS after velocity bunching

Chicane Compression• Run off-crest in linac, “chirp” longitudinal PS• Remove chirp with chicane compressor

– Limited by “thermal” energy spread, CSR

Long beams not easy to compress, large longitudinal e due to RF curvature

,rms kRF

2 z3

2

*

p,th

kRF cot 0 4.810 3 RF m Q pC 1/ 3

p0 MeV cot 0

Final compression for short beams limited by “thermal” energy spread

Original goal: single spike XFEL operation

• 1D dimensionless gain parameter

• 1D gain length• Cooperation length• Single spike operation

1D JJ Krms Krmskp

4ku

2 / 3

Lg,1D u

4 31D

Lc,1D r

4 31D

b,SS 2Lc ,1D r

2 31D

Numerical example: SPARX @ INFN-LNF

• Take 2 GeV operation, “standard SPARC undulator”, u=2.8 cm

• Peak current I=2 kA,• Estimate single spike threshold:

• Note: with ultra-small Q, is enhanced – Spike is a bit shorter…– FEL gain better

1D 1.810 3

b,SS 0.48 m (1.6 fsec)

Ultra-short pulses at SPARX (LNF)

• Scaling indicates use of ~1 pC beam for single spike• sz =4.7 mm after velocity bunching

• June 2008 version of SPARX lattice – compression no longer at end, at 1.2 GeV (Final 2.1 GeV)

• Very high final currents, – some CSR hor. emittance growth, for 1 pC– Longitudinal tails, higher peak brightness (2 orders of magnitude!)

nx 7.510 8 m - rad

Q=1 pC case

B21017 A/m2

FEL performance: 1 pC

• Single spike with some structure• > 1 GW peak power at saturation (30 m)• 480 attosecond rms pulse at 2 nm

s (mm)

t1.67

(l nm) z (m)

z

Higher Q SPARX case: 10 pC

• Put back beam power, X-ray photons• Velocity bunch to 10.3 mm rms

– Still space dominated charge scaling ~Q1/3

• 1 fs pulse• Notable emittance growth

n 7.6 10 7 m - rad

Longitudinal phase space, 10 pC Current profile; some pedestal visible

SPARX FEL at 10 pC

• 10 GW peak FEL power• Saturation at 20 m (v. high brightness)• Quasi-single spike

– Lower brightness=longer cooperation length– 4x1011 photons, good for many applications

s ( m)z (m)

Extension to LCLS case • 1 pC does not give quasi-single spike operation

– Beam too long, need to scale with /l r

• Use 0.25 pC (1.5M e-), obtain en=0.033 mm-mrad• Yet higher brightness; saturation expected in 60 m

E / E 410 4 Over 350A peak

LCLS Genesis results

Deep saturation achieved Minimum rms pulse: 150 attosec

Quasi-single spike, 2 GW

t 1.1

Alternative scenario: new undulator for very short loperation

• Use high brightness beam to push to short wavelength– LCLS example

• Shorter period undulator– Shorter still?

Ku 1

lu (cm) 1.5

b (m) 2-6

U (GeV) 4-14

Use Table-top XFEL undulator?• LMU MPQ-centered collaboration (BESSY, LBNL, UCLA, etc.)• UCLA collaboration on advanced hybrid cryo-undulator

(Pr-based, SmCo sheath, Fe pole), 9 mm period, >2 T• Need short lu high field undulator for X-rays @1 GeV –

critical for traditional linac sources too…• With ultra-high brightness beam, one may have very

compact, extended capability FELs

Simulated 9 mm cryounduator performance at 30K (Maxwell, Radia, Pandira)

Example: SPARX w/sub-fs pulse• Wavelength reduction (3 nm->6.5 Å)• Ultra-short saturation length (10 m)• LCLS photon reach at 2.1 GeV on 5th harmonic

Simulated performance on fundamentalMW peak powerat 1.5 Å, 5th harmonic

Example: LCLS w/sub-fs pulse• Use even shorter 0.25 pC beam, 150 as pulse

– Single spike w/standard LCLS undulator• Obtain ultra-compact “LCLS” at 4.3 GeV• Extend energy reach to 83 keV (0.15Å)

Gain evolution for 1.5 Å 4.3 GeV (0.25 pC)Gain evolution at 13.6 GeV, 0.15 Å

Progress at SLAC on low-Q, high brightness beams

• 20 pC case measured– Diagnostic limitation

• Excellent emittance after injector – No velocity bunching

• Chicane compression– As low as 2 fs rms, with some e growth

Slice Emittance at LCLS: Low Charge Case

20 pC, 135 MeV, 0.6-mm spot diameter, 400 µm rms bunch length (5 A)

OTR screen with transverse

deflector ON

0.14 µm

Emittance near calculated thermal emittance limit,

sz 0.6 mm

8 kA?

LiTrack(no CSR)

Measurements and Simulations for 20-pC Bunch at 14 GeV

Photo-diode signal on OTR screen after BC2 shows minimum compression at L2-linac phase of -34.5 deg.

Horizontal projected emittance measured at 10 GeV, after BC2, using 4 wire-scanners.

Y. Ding

LCLS FEL simulation at 1.5 Å based on measured injector beam and Elegant tracking, with CSR, at 20 pC.

1.5 Å,3.61011 photonsIpk = 4.8 kAge 0.4 µm

Ultra-short beam application: coherent, sub-cycle radiation

• Coherent transition radiation• Non-destructive: coherent edge radiation (CER)

Angular distribution of far-field radiation, by polarization: measured in color, QUINDI in contours

Total emitted CER spectrum (BNL ATF, UCLA compressor), measured compared to QUINDI.Coherent THz pulses…

Combined ER and SR

Coherent optical-IR sub-cycle pulse• CER and CTR cases simulated with QUINDI

– Coherent IR, sub-cycle pulse (SPARX 1 pC case)– Unique source at these wavelengths (~30 MW, peak)– Use in tandem w/X-rays in pump-probe– Would like to try this at LCLS… in context of other

applications (e.g. PWFA)

• CER and CTR cases simulated with QUINDI– Coherent IR, sub-cycle pulse (SPARX 1 pC case)– Unique source at these wavelengths (~30 MW, peak)– Use in tandem w/X-rays in pump-probe– Would like to try this at LCLS… in context of other

applications (e.g. PWFA)

Physics opportunity: focusing ultra-short beams

• 2 fs (600 nm) beam predicted to have Ip=8 kA

• Focus to sr<200 nm (low emittance enables…)• Surface fields

• 2 fs (600 nm) beam predicted to have Ip=8 kA

• Focus to sr<200 nm (low emittance enables…)• Surface fields

eEr remec2Ip /ec r

Final focusing system• Need few mm b-function• Use ultra-high field permanent magnet quads

– mitigate chromatic aberration limits– FF-DD-F triplet, adjustable through quad motion

• Developed 570 T/m (!) PMQ fields– Use need slightly stronger, no problem (Pr, >1kT/m)

Small aperture PMQ B’=570 T/m Halbach PMQ as

builtFinal beam sizes: ~130 nm

Ultra-short, high brightness beam:IR wavelength PWFA

• Ultra-high brightness, fs beams impact HEP strongly…• Use 20 pC LCLS beam in high n plasma• In “blowout” regime: total rarefaction of plasma e-s

– Beam denser than plasma– Very nonlinear plasma dynamics– Pure ion column focusing for e-s– Linac-style EM acceleration – General measure of nonlinearity:

˜ Q Nbkp

3

n0

4kpreNb

1, linear regime

1, nonlinear "blowout"

MAGIC simulation of blowout PWFA caseWakes in blown out region

Optimized excitation at LCLS

• Beam must be short and narrow compared to plasma skin depth

• In this case implies , blowout • With 2 fs LCLS beam we should choose • For 20 pC beam, we have• Linear “Cerenkov” scaling

• 1 TV/m fields (!)• Collaboration initiated

– UCLA-SLAC-USC

nb n0

r kp 1

z kp 1

˜ Q 1

n0 71019 cm-3

˜ Q 7OOPIC simulation of LCLS case

eEz,dec 4e2Nb

z2

eEz,dec e2Nb

n k 1

n k dk e2Nbkp

2

What are the experimental issues?• Length of plasma: ~1 mm

– ~1 GeV energy change, perturbative at beginning

• Plasma formation– ~3 atm gas jet– Beam-induced ionization (fundamental physics)

• Beam focusing– Need few hundred nm beam– Use mini-beta PMQ, easily installed, minimal disruption

• Ion collapse– Deal breaker for applications

• Beam diagnostics in entirely new regime– True for all high-brightness fs beam scenarios

Beam Field-Induced Ionization• Advantage of a HB, well focused drive bunch

– beam field is enormous – plasma created by ionization early in bunch

• Pre-2004 FFTB PWFA used laser pre-ionization• Later, compressed beams at gave ionization in Li

– Previous generation beams had much smaller fields– FFTB: 50 GV/m surface fields: our case: ~1 TV/m

• New regime accessible…

The Tunneling Regime (ADK theory)

Coulomb potential diminished by applied E-field

Electron may quantum mech-anically tunnel out of atom

Regime well understood Perturbation theory First developed for lasers Theory and simulation

(OOPIC) benchmarked to experiment

Tunneling Ionization

The barrier suppression regime (BSI)

In BSI electron can classically escape atom

Focused LCLS beams produce E-fields strong enough for BSI to occur

Previously only reached experimental by lasers

Not so well understood theoretically Non-perturbative! Empirical formulas

Plasma formation can be faster, than ADK theory

Barrier Suppression Ionization

Beam-field induced ionization in OOPIC• Need to focus beam to < 200 nm rms

– A beam size diagnostic• Radial E-field > TV/m; extreme physics…• Ionization studied in hydrogen gas (ADK model)

Complete ionization in LiHydrogen, ionization complete inside beamOOPIC study, 3.15 atm hydrogen ionized by beam

OOPIC predictions for 3.15 atm case

~7E19/cc plasma, >TV/m peak wakefield!

Calculation of BSI ionization• Extension to unipolar field pulse of the

approach of Bauer, et al. in laser context• BSI important above 40 GV/m, but tunneling

has already been accomplished… trust OOPIC

Fractional ionization due to BSI, 800 GV/m, 2 fs gaussian pulse

Plasma (Ion) Focusing

• The beam focuses as due to initial mismatch

Plasma Focusing and Mismatch

• Very small beam sizes even in 1 mm plasmas• Ion’s may collapse…

Ion Collapse• Positive ions “focused” by ultra-dense e-beam fields

• Nonlinear fields, emittance growth• Detect 10-100 keV ions (hydrogen)

– fusion?

Where and when• Option 1: LCLS beam time proposal

– Downstream of undulator– Competitive– Disruptive due to mini-b insert

• Option 2: LCLS bypass line– Proposal for test beam (HEP)– Possible addition of undulator (BES)– Under consideration now, could be fast track

• Option 3: FACET after upgrade– Need RF photoinjector– Optics not yet established– Long lead time…

Status/organization• Physical Review Letter submitted

– Frontier plasma/accelerator/atomic physics– Gift from FEL to HEP

• Collaboration– Based on E169– Proposal for LCLS beam time

• Initial explorations with SLAC stake-holders– Accelerator physicists, SLAC Director Drell– Bucksbaum (work on Rydberg atoms), Stohr

(magnetic switching)• Hope for first experiments in 2011