Very high energy gain Very high energy gain at the Neptune IFEL experimentat the Neptune IFEL experiment

P. MusumeciAdvanced Accelerator Concepts

Stony Brook, NYJune 25th 2004

OutlineOutline

• Brief IFEL introduction• Inverse-Free-Electron-Laser accelerators around

the world• Neptune IFEL project

– Overview of Neptune Laboratory setup• Electron beam• Kurchatov Undulator• CO2 laser

• Experimental results • Conclusion

CollaboratorsP. Musumeci, S. Boucher, A. Doyuran, R.J. England, C.Pellegrini,

J.B. Rosenzweig, G.Travish, R. YoderUCLA Department of Physics and Astronomy

S. Tochitsky, C. Clayton, C. Joshi, J. Ralph, C. Sung

UCLA Department of Electrical Engineering

S.Tolmachev, A. Varfolomeev Jr., A. Varfolomeev, T. Yarovoi

RRCKI Kurchatov Institute

IFEL InteractionIFEL InteractionUndulator magnetic field to couple high power radiation with relativistic electrons

Br

Erx

yz

kr

wmckeBK =

kmceEKl 2

0=

Undulator magnet

Laser and electron beam

22 1

2 2w Kλγλ

≅ ⋅ + ⋅

Significant energy exchange between the particles and the wave happens when the resonance condition is satisfied.

Relative strength

IFEL characteristics: IFEL characteristics: a solid and reliable Advanced a solid and reliable Advanced

AcceleratorAccelerator• Laser accelerator: high gradients • Vacuum accelerator: good output beam quality• Microbunching: control and manipulation of beams at the optical

scale• Efficient mechanism to transfer energy from laser to electrons• State of the art requirements on laser and magnet technology• Synchrotron losses at high energy (can be controlled by

appropriate tapering of undulator)• Gradient is energy dependent

IFEL ExperimentsIFEL Experiments• IFELA: Wernick & Marshall 1992 (PRA, 46, 3566)

– First proof-of-principle IFEL experiment– 5 MW at λ = 1.6 mm, gradient 0.7 MV/m, gain 0.2 MeV

• BNL-IFEL: Van Steenbergen, Gallardo et al. 1996 (PRL 77, 2690)– Microbunching observed 1998 (PRL, 80 4418)– 1-2 GW at λ = 10.6 µm, gradient 2.5 MV/m, gain 1 MeV

• MIFELA: Yoder, Marshall, Hirshfield 2001 (PRL, 86, 1765)– All electrons accelerated, phase dependency of the acceleration– 6 MW at λ = 10 cm, gradient 0.43 MV/m, gain 0.35 MeV

• STELLA: Kimura et al. 2001 (PRL, 86, 4041)– First staging of two IFEL modules. – 0.1-0.5 GW at λ = 10.6 µm, gain up to 2 MeV

• STELLA 2 : Kimura et al. 2003 (PRL, 92, 054801)– Monoenergetic laser acceleration (80 % of electrons accelerated, energy

spread less than 0.5 % FWHM)– ~30 GW,at λ = 10.6 µm, gain up to 17 % of initial beam energy

MotivationMotivation• Proof-of-principle experiments successful• Upgrade to significant gradient and energy gain

– Technical challenges: • very high power radiation• strong undulator tapering

– Physics problems:• include diffraction effects in the theory• beyond validity of period-averaged classical

FEL equation• The Neptune Laboratory at UCLA has a high-power

laser and a high-brightness electron beam

Experimental LayoutExperimental Layout

E- beam

Laser beam

Vacuum translation stages insert in the middle of the undulator a probe for spatio-temporal alignment

High energy spectrometer

IFEL vacuum box

Kurchatov strongly tapered undulator

Final focus large aperture quadrupole magnets

To streak camera diagnostics

Neptune IFEL Design ParametersNeptune IFEL Design ParametersLaser Power 400 GW

Laser wavelength 10.6 µm

Laser beam size (w0) 340 µm

Rayleigh range 3.5 cm

Energy 14.5 MeV

Energy spread (rms) 0.5 %

Charge 300 pC

Pulse length (rms) 4 ps Rms transverse

Emittance 10 µ

Rms beam size at the focus 150 µm

ee--beam deliverybeam delivery

• New quadrupole magnets with larger (2.625”) aperture to avoid CO2 clipping– Up to 7 T/m– 9.1 cm effective length

The fringe field has a correction of ~ 10% on the radius of curvature

22 24 26 28 30 32 340

1x107

2x107

3x107

4x107

5x107

6x107

7x107

# of

cou

nts

(a.u

.)

Pressure in Cherenkov cell (psi)

14.5 MeV 13.9 MeV 13.2 MeV

• Cherenkov cell– Vary CO2 pressure to find

threshold for Cherenkovemission

• Calibration of 45 º two dipoles spectrometer– Alignment errors– Fringe field issue

KurchatovKurchatov IFEL UndulatorIFEL Undulator

• Unique “double tapered” 50 cm long undulator. – Final resonant energy 250 % bigger than

initial • Hall Probe measurements. • Pulse Wire tuning.

Initial Final

Period 1.5 cm 5.0 cmField

Amplitude 0.12 T 0.6 T

Peak K parameter 0.2 2.8

gap 12 mm 12 mm

0 100 200 300 400 500-8

-6

-4

-2

0

2

4

6

8

B (k

Gau

ss)

z (mm)

Hall probe scan

Diffraction Dominated InteractionDiffraction Dominated Interaction

2

2

( ) sin( )2( )1

l

w

r

kKK JJ Kz z z

z

γ ψγ

∂= ⋅

∂ −+

1/2 222

2

2

1( )11 ( ) cos( )

2 21

wwl

rlr

kk kz z zKK zKK JJ K z

ψ

ψ

γ

∂= + − −

∂ − ++ + + ⋅ ⋅ −

0 100 200 300 400 500

-0.003

-0.002

-0.001

0.000

0.001

0.002

Beam centroid trajectory CO2 beam size ( w0 )

X (m

m)

Z (mm)0.0 0.1 0.2 0.3 0.4 0.5

20

40

60

80

100

120

Ele

ctro

n en

ergy

(γ)

Distance along the undulator ( m )

DiagnosticsDiagnostics• Charge

– ICT in the folding box and Faraday Cup at the end of beamline.

• Energy spectrometer – Browne and Buechner geometry

to have the energy spectrum over as wide a range of energies as possible.

– 1.5” pole gap for CO2 dump– 11º edge angle for controlling

vertical size of the beam• Spatial and temporal

synchronization– Mid-of undulator graphite-coated

phosphorous screen– Ge crystal on the beam path for

temporal synchronization using e-beam controlled transmission of CO2

Experimental 2d field map of Browne-Buechner poles

200 400 600 800

0

2

4

6

8

10

12

x0 459.4±4.9σt 91.0±11.2a.

u.

Time delay (ps)

Measured cross-correlation curve fit reconstructed CO2 laser pulse

Pulse propagation in a COPulse propagation in a CO22 amplifieramplifier

0

50

100

150

200

250

0 5 10 15 20

Pulse delay in the MARS amplifier

Tim

e de

lay,

ps

Amplification - g x L, m -1

nactive medium = nres + nnonres

Classical Vgr = c/(n+νdn/dν)

nnonres

= 1+ 2π Nmα m +m

∑ ∆nvibr − ∆ne

m = CO2 , N2

nres

(ν ) = n0 +c(ν − ν 0 )g(ν )

2πν∆ν

Vgr = c

n0 +cg0

2π∆ν

t delay = Lg0

2π∆ν

tmin

tmax

Time delay of 100 ps < tdelay < 220 ps in the active medium of MARS high-power amplifier has to be compensated.

Streak camera: Streak camera: ““Live from the bunkerLive from the bunker””• Shot-to-shot measurements of the laser pulse length and the timing

between two pulses necessary because of the complex dynamics of the final amplification of the CO2 pulse.

• Optical Kerr Effect to get CO2 streaks• E-beam reference pick off the photocathode drive laser

532 nm pulse

CO2 pulse

0 500 1000 1500 200005

101520

250 ps1170 psa.

u.

time (ps)

Streak lineout

Measurement of timingMeasurement of timingbetween CO2 and e beambetween CO2 and e beam

900 950 1000 1050 1100 1150 1200 1250 1300 135016

17

18

19

20

21

22

23

24IF

EL

Out

put e

nerg

y (M

eV)

Relative timing between Photo-Injector Driver and CO2 laser (ps)

0

20

40

IFEL output energy vs timing Typical laser pulse profile

Optimization of IFEL outputOptimization of IFEL output

13.0 13.5 14.0 14.5 15.0

18

20

22

24

26

28

30

IFEL output vs. Injection energy

Injection energy (MeV)

Fina

l ene

rgy

(MeV

)

0

1

2

3

4

5

Fraction of particles trapped (%)

100 150 200 250 300 350 400 450

20

22

24

26

28

30

Laser Power (GW)

Fina

l ene

rgy

(MeV

)0

1

2

3

4

5

6IFEL output vs. Laser Power Fraction of captured particles (%

)

Output energy vs. focus positionOutput energy vs. focus position

-6 -5 -4 -3 -2 -1 022

24

26

28

30

32

34

36

Intensity at undulator entrance P= 400 GW, zr= 1.8 cm Intensity Threshold for trapping

Focus position (cm)

Fina

l ene

rgy

(MeV

)

20

30

40

50

60

70

80

I (GW

/cm2)

SideSide--effects of effects of very high power laser beamsvery high power laser beams

• Increase in laser intensity could be accomplished by increase inRayleigh range, or increase of power in the pulse…

Single crystal NaCL windowsSingle crystal NaCL lens

Copper mirrors

Single shot spectrum (laser on)Single shot spectrum (laser on)

10 15 20 25 30 350

400

800

1200

1600

2000

a.u.

Electron energy (MeV)

Single shot spectrumSingle shot spectrum(laser polarization 90(laser polarization 90ºº off)off)

10 15 20 25 30 350

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000a.

u.

E lectron energy (MeV)

Single shot spectrum (laser on)Single shot spectrum (laser on)

10 15 20 25 30 350

400

800

1200

1600

2000

a.u.

Electron energy (MeV)

24 26 28 30 32 340

50

100

150

200

a.u.

Energy (MeV)

Where are the energy peaks coming from?Where are the energy peaks coming from?2

2,

12

2

w

res n

K

n

λγ

λ

⋅ + =

⋅

Unfortunately we were not able to follow the red curve because of missing laser intensity, but if you slip out of the first resonance, the undulator is tapered enough that electrons can start to exchange energy with 10 µm photons through second harmonic coupling !!!

0 0.1 0.2 0.3 0.4 0.5

50

100

Fundamental2nd harmonic3rd harmonic

Distance along the undulator (m)

Res

onan

t ene

rgy

Neptune IFEL undulator resonant energy

Resonant conditions for energy transfer between particles and e.m. wave: Higher Harmonic IFEL

Higher Harmonic IFEL theoryHigher Harmonic IFEL theory

( )( ) sin (1 )2

nl w

n

JJ K KkK k z n kz tzγ ω ϕ

γ∂

= ⋅ ⋅ + + − +∂ ∑

( ) ( ) ( )( )2 2 2( ) ( ) ( ) ( )n m m n m nm

JJ K J K J K J Kξ σ σ∞

+ + +=−∞

= ⋅ +∑2

2( )

4 12

KKK

ξ =

+

( )w

KKk w

σγ

=where

0 1 2 3 4 50

0.10.20.30.40.50.60.70.80.9

1

Fundamental2nd harmonic3rd harmonic

Coupling coefficients for harmonics

K undulator

JJ fa

ctor

The even harmonics are coupled through the diffraction angle

In the limit σ(Κ)→0 we found the known result for odd harmonics

Higher Harmonic IFEL gives a lot of Higher Harmonic IFEL gives a lot of flexibility in flexibility in undulatorundulator design !!!design !!!

3d simulation3d simulation• Energy gain is in the first section of undulator. (20 MeV in 25 cm !! )• Higher Harmonic IFEL in the second section

Summary & ConclusionSummary & Conclusion

• IFEL Advanced Accelerator at the Neptune Laboratory– > 20 MeV energy gain ( + 150 % ) !!– trapped up to 10 pC in accelerating buckets !– accelerating gradient ~70 MeV/m !

• First experimental study of Strong Tapering & Diffraction Effects in IFELs

• Observation of Harmonic IFEL interaction in second section of undulator