review of ultra high-gradient acceleration schemes ...epaper.kek.jp/e02/talks/mozgb004.pdf · m....

RA EPAC02 1

Review of Ultra HighReview of Ultra High--gradient gradient Acceleration Schemes, Results of Acceleration Schemes, Results of

ExperimentsExperimentsR. R. AssmannAssmann, CERN, CERN--SLSL

Reporting also on work done at SLAC, as member of and in collaboration with the Accelerator Research Department B (ARDB) under R. Siemann.

Thanks to my former co-workers and colleaguesM. Hogan (SLAC), C. Joshi (UCLA), T.Katsouleas (USC), W. Leemans (LBNL), J. Rosenzweig (UCLA), R. Siemann (SLAC)F. Amiranoff (Ecole Polytechnique)

for help and advice with this talk.Picture courtesy T. Katsouleas

RA EPAC02 2

1) Why ultra-high gradients?The Livingston plotLimits for new particle acceleratorsBasic requirements for particle physics The promise of ultra-high gradients

2) How to get ultra-high gradients?Overview on schemes being followedPrinciples (plasma and vacuum concepts)

3) Experimental results - HighlightsPlasma wakefield accelerationLaser wakefield accelerationLaser acceleration

4) Moving towards usage in a linear colliderConsiderationsThe plasma booster for the Higgs?

5) Conclusion

RA EPAC02 3

The Livingston plotThe Livingston plot

M. Tigner: �Does Accelerator-Based Particle Physics have a Future?�Physics Today, Jan 2001 Vol 54, Nb 1

The Livingston plot shows a saturation effect!

Practical limit for accelerators at the energy frontier:

Project cost increases as the energy must increase!Cost per GeV C.M. proton has decreased by factor 10 over last 40 years (not corrected for inflation)!

Not enough: Project cost increased by factor 200!

New technology needed�

RA EPAC02 4

Basic requirements Basic requirements beyondbeyond next next generation egeneration e++ee-- linear colliderslinear colliders

Beam energy c.m. > 5 TeV

Luminosity > 1035 cm-2 s-1

Beam power(consequence): ~ 100 MW

Example:

Energy = 2.5 TeVLuminosity = 1035 cm-2 s-1

IP spot size = 10 nmBunch population = 5 109

New technologies for high energy colliders must be compatible with reasonable beam power:

Good efficiencyLow emittance (beam size)High bunch charge

RA EPAC02 5

The promise of ultraThe promise of ultra--high gradientshigh gradients

Final energy = Active acceleration length × Accelerating gradient

1. Accelerators become longer.

2. Without �cheap� technology they become more expensive.

3. They do not fit on the sites of existing laboratories (more difficult

approval, additional infrastructure cost).

Exponential increase in energy without exponential increase in accelerating gradient:

Ultra-high gradients of up to 100 GV/m have been observed (gain of 3 orders of magnitude).

What is the status of this technology? Can we build ultra-short high-energy colliders?

RA EPAC02 6

Feasible gradients in metallic structures:

Collider Gradient

SLC 20 MeV/m SLAC 1988-1997

ILC 70 - 85 MeV/m SLAC, KEKTESLA 25 - 35 MeV/m DESYCLIC 150 MeV/m CERN

What are ultraWhat are ultra--high accelerating gradients?high accelerating gradients?

Ultra-high accelerating gradients: 1 - 100 GeV/m

�obtained with various new technologies!

See talk by W. Wuensch�

RA EPAC02 7





5) Conclusion

RA EPAC02 8

Advanced accelerator researchAdvanced accelerator research

Two types of advanced accelerators:

1. Wakefield accelerators

Dielectrics and microstructures (RF and two beam)

Plasma

2. Laser accelerators

Vacuum

Plasma

RA EPAC02 9

! Laser Wake Field Accelerator(LWFA)A single short-pulse of photons

! Self Modulated Laser Wake Field Accelerator(SMLWFA)Raman forward scattering Instability

! Plasma Beat Wave Accelerator(PBWA)Two-frequencies, i.e., a train of pulses

! Plasma Wake Field Accelerator(PWFA)A high energy electron bunch

Concepts For PlasmaConcepts For Plasma--Based AcceleratorsBased AcceleratorsPioneered by J.M. Dawson

evolves to

Courtesy T. Katsouleas

RA EPAC02 10

Basic principle and scaling rules 1Basic principle and scaling rules 1

I) Generate homogeneous plasma channel:GasLaserPlasma

II) Send dense electron beam towards plasma:

Beam density nb> Gas density n0

= ion = electron

Beam excited plasma. Also lasers can be used (laser wakefield acceleration).

RA EPAC02 11

III) Excite plasma wakefields:

Electrons are expelled

Ion channelr

z


Space charge force of beam ejects all plasma electrons promptly along radial trajectories

Pure ion channel is left: Ion-focused regime, underdense plasma

RA EPAC02 12

Equilibrium condition: n

r

n0

Drive beam

Ion channel

Quasineutralplasma

an(neutralization radius)

Ion charge neutralizes beam charge:

0nna b

rn ⋅=σ

Beam size

Beam and plasma densities determine mostcharacteristics of plasma wakefields!

SLC:nb/n0 = 10


RA EPAC02 13

driving force: Space charge of drive beam displaces plasma electrons.

restoring force: Plasma ions exert restoring force

Electron motion solved with ...

Space charge oscillations(Harmonic oscillator)

++++++++++++++ ++++++++++++++++

----- --- ----------------

--------------

--------- ----

--- -------------------- - --

---- - -- ---

------ -

- -- ---- - - - - - ------ - -

- - - - --- --

- -- - - - - -

---- - ----

------

electron beam

+ + + + + + + + + + ++ + + + + + + + + + + + + + ++ + + + + + + + + + + + + + ++ + + + + + + + + + + + + + +-

- --

--- --

EzEz


Longitudinal fields can accelerate and decelerate!

RA EPAC02 14


N re e

z

− ⋅→ ≈

σ1 (SLC: 0.2)

W n V mz ≈ ⋅ ⋅100 0 /

Accelerating field

Plasma density n0 1014 to 1015...

~ GV/m

Wavelength:

λ p mmcm

n≈ ⋅

−

11015 3

0

λp ~ mm

∝ N b zσ 2

(roughly)

RA EPAC02 15


W

rn e

T

m

n

cmr = ⋅ ⋅ = ⋅

−2 960

1002 0

14 3π π

Plasma ions move relatively little Constantfocusinggradient

β γϖ

= ⋅2c

p

Plasma �structures� are also super-strong �quadrupoles�!

(many thousand T/m)

... need to handle acceleration and focusing!

RA EPAC02 16

Basic observationsBasic observations

� High accelerating gradients (many GV/m)

� Short wavelengths (< mm)

Short bunches (< 100µm)

� Strong transverse focusing (> kT/m), small β (< cm)

Small matched beam sizeRigid tolerances

Plasma based acceleration:

Availability of short bunches for experiments is limited!

RA EPAC02 17





5) Conclusion

RA EPAC02 18

10 2

10 3

10 4

100 120 140 160 180 200

3.5 MeV4.1MeV6.5MeV5.0MeV7.3MeV8.6MeV5.2 MeV,

Cloud Chamber

}

Ele

ctro

n S

igna

l (A

U)

Pressure (mTorr)

Calculated trajectory

Acceleratedelectron track

Low energynoise

7.25 cm

3.4 cm

Bext

e-

Ultrahigh-gradient acceleration of injected electrons by laser-excited relativistic electron plasma waves

C.E. Clayton et al., Phys. Rev. Letters 70, 37 (1993)

Cloud chamber photograph Resonance curve

� Conclusive evidence of injected electron acceleration.

� First evidence of resonant acceleration.

UCLA

Courtesy C. Joshi, UCLA

RA EPAC02 19

100

101

102

103

104

105

106

1 10

30% model25% model

Elec

trons

/MeV

Electron energy (MeV)303

InjectionEnergy

TrappingEnergy

Trapped electron acceleration by a laser-driven relativistic plasma wave

M. Everett et al. Nature 368, 527 (1994)

� First observation of energy gain beyond trapping threshold

� Evidence for energy loss as well as energy gain by plasma wave

UCLA


RA EPAC02 20

Observation of energy gain:

100 MeV

over

0.6 mm

Observation of electron energies beyond the linear dephasing limit from a laser-excited relativistic plasma wave

D. Gordon et al., Phys. Rev. Letters 80, 2133 (1998)UCLA

=> at least 160 GeV/m acceleration gradient


Done at Rutherford Laboratories, UK

RA EPAC02 21

The NEPTUNE Facility for 2nd Generation Advanced Accelerator Experiments

C. Joshi, J. Rosenzweig, C. Pellegrini, K.A. Marsh,S. Tochitsky, and C.E. Clayton

� Acceleration of injected, high-current beam to 100 MeV.� Acceleration of a pre-bunched electron beam.� Beam-physics studies.

Goals:

Laser-oscillatorroom

Experimenthall

Injectorlinac

Plasmaaccelerator

Amplifier hall

Controlroom

UCLA


RA EPAC02 22

Table-top Laser-Driven Electron Acceleration(via self-modulated laser wakefield in plasma)

D. Umstadter,S.-Y. Chen,A. Maksimchuk,R. Wagner,

Courtesy T. Katsouleas, USC

RA EPAC02 23

Antonio C. Ting, Chris I. Moore, Rich Fischer, Dmitri Kaganovich, Ted Jones

NRL EXPERIMENTAL PROGRAM

Self-Modulated Laser Wakefield Accelerator:

Production of up to

100 MeV electrons

at >100 GV/m gradient

has been observed (over ~ mm). A. Ting, et al, Phys. Plasmas, 4, 1889 (1997)D. Kaganovich, et al, PRE, 59, R4769 (1999)

6 8 10 20 40 60 80100 200

103

104

105

106Shot 12 (10 kG)Shot 26 (10 kG)Shot 29 (5 kG)

Shot 33 (5 kG)Shot 39 (2.5 kG)Shot 40 (2.5 kG)

Rel

ativ

e #

of e

lect

rons

/MeV

/Ste

radi

anElectron energy (in MeV)

SM-LWFA electron energy spectrum

Courtesy T. Katsouleas, USC

Injection and channel guiding for the GeV range (longer).

RA EPAC02 24

Laboratoire pour l�Utilisation des Lasers Intenses (LULI)CNRS - CEA - Ecole Polytechnique - Université Paris 6

LPNHE - LSI �LPGP - IC - LOA - CPTH100 TW LULI laser facility

35 fs, 600 mJ, 2.1019W/cm2, 10 Hz

≈ 100 MeV in

E >100 GV/m

Few nC > 4 MeVDivergence : few mrad

106

107

108

109

1010

1011

0 10 20 30 40 50 60 70Num

ber o

f ele

ctro

ns [/

MeV

/sr]

Electron energy [MeV]

detection threshold

ne=1.510 20 cm-3 T=2.6 MeV

ne=5.10 19 cm-3 T=8.1 MeV

Self-modulated wakefield experiments - 1 single laser pulse focused in a gas jet

RA EPAC02 25

The potential energy gain is huge

The effective acceleration length is limited by diffraction of the laser beam

What about guiding the laser beam over 1 m ?

l0 = 1 µm , ne = 1017e-/cm3 , g = 100 : ∆W = 20 GeV over L = 1 m

Some laser guiding scheme is necessary to reach high energy gains

L=2ZR

E

Laser beam

Plasma wave

Laser guiding in dielectric capillary tubes

RA EPAC02 26

Production of electron beams with low energy spreads and pulse-to-pulse energy stability in a plasma-based accelerator requires:

� Injection of femtosecond electron bunches

� Injection with femtosecond timing

Laser triggered injection of background plasma electrons into a plasma wake by beating of two colliding pulses

[E. Esarey et al., PRL, 79, 2682 (1997)]

Colliding pulse injector

Reference:E.Esarey et al., PRL '97C.B. Schroeder et al., PRE '99

ψ = plasma wake phase = tzk pp ω−

a2a1

a0

-17.5 -15 -12.5 -10 -7.5 -5 -2.5 0-0-0.75

-0.5

-0-0.25

0

0.5

0.751

0.25

Nor

mal

ized

po

tent

ials

Courtesy W. Leemans LBNL

RA EPAC02 27

E-157 & E-162 & E-164 Collaborations

R. Assmann, F.-J. Decker, P. Emma, M. J. Hogan*, R.H. Iverson, C. O�Connell, P. Krejcik, P. Raimondi, R.H. Siemann, D. Walz

Stanford Linear Accelerator Center

B.E. Blue, C.E. Clayton, C. Huang, C. Joshi*, K.A. Marsh, W.B. Mori, S. Wang

University of California at Los Angeles

T. Katsouleas*, S. Lee, P. Muggli University of Southern California

� Extraordinarily high fields developed in beamplasma interactions

� Many questions related to the applicability ofplasmas to high energy accelerators andcolliders

Address these questions via experiments "

� E-157: First experiment to study Plasma WakefieldAcceleration (PWFA) of electrons over meter scaledistances

� Physics for positron beam drivers qualitativelydifferent (flow-in vs. blow-out) "E-162

� Opportunity for dramatically shorter bunches in theFFTB in 2003 with correspondingly highergradients (> GeV/m) " E-164

Plasma Wakefield Acceleration in Meter-long Plasmas

UCLA

Courtesy M. Hogan, T. Katsouleas

RA EPAC02 28

Located in the FFTB

e- or e+

N=2·1010

σz=0.6 mmE=30 GeV

IonizingLaser Pulse

(193 nm)Li Plasma

ne≈2·1014 cm-3

L≈1.4 m

CerenkovRadiator

Streak Camera(1ps resolution)

BendingMagnet

X-RayDiagnostic

Optical TransitionRadiators Dump

12 m

∫Cdt

Experimental Layout

FFTB

UCLA

RA EPAC02 29

0

100

200

300

400

500

600

0 2 4 6 8 10 12 14

BetatronFitShortBetaXPSI.graph

Plasma OFFPlasma ONEnvelope

σ x (µm

)

Ψ

L=1.4 mσ0=14 µm

εN=18×10-5 m-radβ0=6.1 cmα0=-0.6

� Smaller �matched� beam size at the plasma entrance reduces amplitude of the betatron oscillations measured at the OTR downstream of the plasma� Allows stable propagation through long plasmas (> 1 meter )

0

50

100

150

200

250

300

-2 0 2 4 6 8 10 12

05160cedFIT.graph

σX

DS

OTR

(µm

)

ψ=K*L∝ ne1/2L

σ0 Plasma Entrance=50 µm

εN=12×10-5 (m rad)β0=1.16m

E-157 E-162 Run 2

Beam Propagation Through A Long Plasma

C. E. Clayton et al., PRL 1/2002

Phase Advance Ψ ∝ ne1/2L

Plasma OFF

Phase Advance Ψ ∝ ne1/2L

s x (µ

m)

UCLA

E-157/E-162 collaboration

RA EPAC02 30

UCLA

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

-8 -4 0 4 8

θ (m

rad)

φ (mrad)

θ∝ 1/sinφ

θ≈φo BPM DATA

Impulse Model

rc=α(nb/ne)1/2rb

θφ

Head

Plasma, ne

AsymmetricChannel

Beam Steering

SymmetricChannel

Beam Focusing

e-++++

++++++++

++++++++

++++++++

++++++++

+++++++++ + ++++

- - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Core

Electron Beam Steering:Refraction At Plasma�Gas Boundary

P. Muggli et al., Nature 411, 2001

� Vary plasma � e- beam angle φusing UV pellicle

� Beam centroid displacement@ BPM6130, 3.8 m from theplasma center

E-157 collaboration

RA EPAC02 31

Refraction of an Electron Beam:Interplay Between Simulation & Experiment

Laser off Laser on

3-D OSIRIS

PIC Simulation

Experiment(Cherenkov

images)

1 to 1 modeling of meter-scale experiment in 3-D!

P. Muggli et al., Nature 411, 2001

UCLA

E-157 collaboration

RA EPAC02 32

Betatron Radiation of X-rays

l Peak brightness ~ 1019 photons/sec-mm2-mrad2-.1%bw!

Plasma focusing strength of 6000T/m acts as a strong undulator

UCLA

E-157 collaboration

RA EPAC02 33

Average energy gain (slice average): 156 ±40 MeV(≈1.5 ×108 e-/slice)

Average energy loss (slice average): 159 ± 40 MeV

PicosecondGaussian

Slice Analysisof Many Events

-200

-150

-100

-50

0

50

100

150

200

-6 -4 -2 0 2 4 6 8

SliceEnergyGain3curves.graph

ne=1.6×1014 (cm-3)

ne=2.0×1014 (cm-3)

ne=(2.3±0.1)×1014 (cm-3)

Rel

ativ

e En

ergy

(MeV

)

τ (ps)

+σz +2σz +3σz-2σz -σz

E-162: Use Imaging SpectrometerTo Measure Energy Loss & Gain

Preliminary�

UCLA

E-164: > 1 GeV/m (2003) E-162 collaboration

RA EPAC02 34

Other schemesOther schemes

Laser acceleration in vacuum (LEAP):

Two crossed laser scheme: �Only tantalizing of the interaction signature�.Power source: Laser.New proposal�

Inverse free electron laser (STELLA):

Power source: Laser.Successful staging. 90 MeV/m goal.Limit from synchrotron radiation at ~ 200 GeV. Bunches of femto-second length.

Di-electric wakefield accelerator:

Possible gradients: 100 MeV � 1GeVPower source: Drive beam.

RA EPAC02 35





5) Conclusion

RA EPAC02 36

Efficiency 1Efficiency 1

High beam power requires:

Efficient power generation

Progress in the field of lasers:

High peak power is possible (~ 100 TW )

Wall-plug-to-photon efficiency: 20 � 40 %

Better stability (pointing stability in sub µrad regime)

Efficiency of beam-driven PWFA: ~ 30% LEAP at the ORION Center

Robert L. Byer & Robert H. Siemann

RA EPAC02 37

C. D. Barnes et al, SLAC PROPOSAL E-163

Efficiency 2Efficiency 2

Accelerating Microstructures:Re-use the laser beams for increasing efficiency!

RA EPAC02 38

High quality beam transport 1High quality beam transport 1

Short accelerator, but what about emittance?R. Assmann, K. Yokoya, NIM 1997

Parameter PWFA LWFAAcceleration 1 GeV/m 30 GeV/mWavelength 2 mm 100 µmFocusing field 6,000 T/m 600,000 T/mModule length 6 m 1 mInjection energy 1 GeV 1 GeVFinal energy 1 TeV 1 TeVAcc. length 1 km 33 m

RA EPAC02 39

Matched beam size at injection energy:

0

2

4

6

8

10

12

1e-08 1e-07 1e-06 1e-05

σ 0 [

µm]

γε [m-rad]

SLC

NLC

PWFALWFA

NLC type emittance requires sub-micron spot size!


4 10-8 m rad

RA EPAC02 40


Tolerance for alignment beam - plasma wakefield:

� emittance 4 10-8 m rad with 200% emittance growth� Number of plasma cells: 167 or 33 (LWFA)

RMS alignment tolerance: < 300 nm (PWFA) 30 nm (LWFA)

High energy plasma-wakefield accelerator will have difficult alignment problems, but...

New ideas: Hollow plasma channels

No ions on axis, no focusing!Acceleration still works...

RA EPAC02 41

0

100

200

300

400

500

600

700

800

900

0 0.5 1 1.5 2 2.5

channel Radius R wp/c

e+

e-

Positron longitudinal wake amplitude is comparable to e- in a hollow plasma:

Hollow plasma channels with eHollow plasma channels with e++ and eand e--

T. Katsouleas, USC

RA EPAC02 42

The first step: A plasma afterburner?The first step: A plasma afterburner?

A 100 GeV-on-100 GeV e-e+ Collider at SLAC based on Plasma Afterburners

50 GeV e50 GeV e-- 50 GeV e50 GeV e++e-WFA e+WFA

IP

LENSESAfterburners

SLAC facilities

3 km

30 m

E-157 collaboration

RA EPAC02 43

Afterburner vs. E157 ComparisonAfterburner vs. E157 Comparison

� Bunch: � Number: 1 - σz= .65mm

� Plasma� Density: 0 - 5e14 cm-3

� Length: 1.4 m

� Betatron Oscillations:� 0-5

� Bunches:� Number: 2- σz= .065mm, .033 mm� Separation: .019 mm

� Plasma� Density: 2e16 cm-3

� Length: 7 m

� Betatron Oscillations:� 100

E-157 Afterburner

Can we control 7 m of plasma?

Studies are ongoing for this idea�

RA EPAC02 44





5) Conclusion

RA EPAC02 45

Advanced accelerator concepts have made impressive progress:

High gradient acceleration up to 160 GV/m demonstrated.

Beam-plasma interaction of 30 GeV SLAC beam in 1.4m long plasma.

Experimental results in excellent agreement with predictions.

New applications (fsec science, plasma wiggler, kicker, lens,�).

Many problems still to be addressed for building a high energy collider�� but no fundamental show stopper!?

Accelerator physics moves into the right direction (small emittance, short bunches).

Can a plasma afterburner on a �conventional� accelerator be the first step?

ConclusionConclusion

RA EPAC02 46

Plasma Accelerator Progress and the Livingston Curve

0

1

2

3

4

5

6

7

1930 1940 1950 1960 1970 1980 1990 2000 2010

LivingstonCurrent statusfuture

Log

of E

nerg

y in

MeV

Year

RAL/IC/UCLA/LULI NRL, UM

UCLA

UCLA Port J. LLNL/UCLA

?

SLAC/USC/UCLA/LBL

LEP

E-164/ORIONE-162

T. K

atso

ulea

s, U

SC

review of ultra high-gradient acceleration schemes ...epaper.kek.jp/e02/talks/mozgb004.pdf · m....

Documents