LEAP/E163: Laser Acceleration at the NLCTA Who we are PIs: Robert H. Siemann (50%), SLAC & Robert L. Byer, Stanford Staff PhysicistsGraduate Students Postdoctoral.

LEAP/E163: Laser Acceleration at the NLCTAWho we arePIs: Robert H. Siemann (50%), SLAC & Robert L. Byer, Stanford

Staff Physicists Graduate Students Postdoctoral RAEric R. Colby (100%), Spokesman Melissa Berry Rasmus Ischebeck (50%) Robert J. Noble (30%) Ben Cowan James E. Spencer (70%) Melissa Lincoln E163 Collaborators

Chris McGuiness Tomas PlettnerStaff Engineer Chris Sears Jamie Rosenzweig Dieter Walz (CEF, 10%) Sami Tantawi, Zhiyu Zhang (ATR)

What we doDevelop laser-driven dielectric accelerators into a useful accelerator technology by:• Developing and testing candidate dielectric laser accelerator structures• Developing facilities and diagnostic techniques necessary to address the unique technical challenges of laser acceleration

Motivation• Lasers can produce far higher energy densities than can microwave sources, hence larger electric fields• Dielectric materials can hold off field stresses of >1 GV/m for picosecond-class pulses• Lasers are a large-market technology with rapid R&D by industry (DPSS lasers: ↑0.22 B$/yr vs. ↓0.060B$/yr for microwave power tubes)• Short wavelength acceleration naturally leads to sub-femtosecond bunches • Technology to handle laser materials lithographically is rapidly evolving an all solid-state accelerator

Work supported by Department of Energy contracts DE-AC02-76SF00515 (SLAC) and DE-FG03-97ER41043-II (LEAP).

Proof-of-Principle Demonstration

We have shown that “direct” (no plasma) acceleration of electrons with light can be done with useful gradients and a very simple geometries

-50 -25 0 25 50-1000

-500

0

500

Position (mm)

Offset (microns)

Centroid Trajectory Inside Undulator

ElectronLaser

Figure 1: a) Above, laser & electron trajectories inside undulator for a gap of 5.4 mm. b) Left, gap scan data with simulation. The data shows clear peaks matching the simulation. Scan is composed of 164 separate runs with a fixed gap position for each run.

4 5 6 7 8 9 10 11 120

5

10

15

20

25

30

35

40

Undulator Gap (mm)

IFEL Modulation (keV; FWHM)

IFEL Gap Scan Data

Simulation x 0.67Data

4th

5th

6th

Inverse Transition Radiation Acceleration Harmonic Inverse FEL Acceleration

C. M. Sears, et al, Phys. Rev. Lett., 95, 194801 (2005).T. Plettner, et al, Phys. Rev. Lett., 95, 134801 (2005).

A single metal boundary illuminated by linearly polarized light at the transition radiation angle

Demonstrated: •Acceleration of appreciable charge (q~107 e-) by visible light•A peak longitudinal field of Ez>40 MV/m•“Large” interaction distance: ~1 mm or ~1200

The next step is to thoroughly explore the physics and technical limits of these and other more advanced structures.

A 3-period variable-gap undulator

Demonstrated: •Acceleration of appreciable charge (q~107 e-) by visible light•Interaction between electrons and higher-order undulator resonances (4th,5th, 6th)

This IFEL will be used to energy-modulate the beam as part of an optical prebuncher for staging experiments.

Inverse Transition Radiation Experiments = 800 nm100 m spotT ~ 2 psec ½ mJ/pulseE0 ~ 2.3 GV/mIo ~ 1.1 J/cm2

Laser pulsegaussian time

and spatial profile

-30 -20 -10 0 10 20 300

0.2

0.4

0.6

0.8

1

boundary angle = 45°

norm

aliz

ed e

nerg

y ga

in

laser crossing angle (degrees)

= 0.5 2 MeV 10 MeV 50 MeV

-40 -30 -20 -10 0 10 20 30 400

5

10

15

20

25

30

35

40

laser crossing angle (mrad)

energy gain (keV)

Umax ~ 37 keV

E163 (60 MeV)opt ~ 8.6 mradUmax ~ 37 keV

HEPL (30 MeV)opt ~ 16.8 mradUmax ~ 18.1 keV

-60 -40 -20 0 20 40 600

10

20

30

40

50

60

70

80

90

100

110

interaction distance (mm)

energy gain (keV)

phase reset ret

ret =

ret =

ret =

U

U

53 keV

75 keV

37 keV

= 800 nm100 m spotT ~ 2 psec ½ mJ/pulse

E0 ~ 2.3 GV/mIo ~ 1.1 J/cm2

Laser pulsegaussian time and

spatial profile

ret

= 8.3 mrad

1. U() Normal Boundary Reflective

2. U() Inclined Boundary Reflective

3. U() Inclined Boundary Transmissive

Is acceleration the result of F=qE (the fields couple directly to the accelerated electrons), or the result of F=kqq’/r2, (the fields induce surface currents on a boundary, which in turn accelerate the electrons)?

4. U Normal Boundary Absorbing ITR

Basic Physics Issue:

Guoy phase shift compensated

• Accelerating mode in planar photonic bandgap structure has been located and optimized

• Developed method of optical focusing for particle guiding over ~1m; examined longer-range beam dynamics

• Simulated several coupling techniques• Numerical Tolerance Studies: Non-

resonant nature of structure relaxes tolerances of critical dimensions (CDs) to ~λ/100 or larger

Structure contour shown for z = 0; field normalized to Eacc = 1

Vacuum defectbeam path is into the page silicon

Synchronous (=1) Accelerating Field

X (m)

Y (

m)

Planar Photonic Accelerator Structures

This “woodpile” structure is made by stacking gratings etched in silicon wafers, then etching away the substrate.

Goals:1. Design fibers with

band gaps to confine vphase = c modes

2. Calculate accelerating mode properties: ZC, vgroup, damage factor,…

Codes:1. RSOFT –

commercial photonic fiber code using Fourier transforms

2. CUDOS – Fourier-Bessel expansion from Univ of Sydney

kza

a/

c

v = c

lowest band gap

CUDOS: Poynting Vector and Accel. Field in silica PBG Fiber

Modeling PBG Band Gaps and Defect-Guided Modes RSOFT: Model of Blaze Photonics Fiber

Large band gap where expected at = 1.5

50 100 150 200

-1.5

-1

-0.5

0

0.5

1

1.5

Frequency (Hz)

Amplitude log. scale (arb. units)

Coil Scan of 3rd PMQ

dipole

Quadrupole

Sext Oct

Dec

Dodec

Developed techniques for designing (Radia), fabricating (EDM), and measuring fields (hall

scans, pulsed wire, and rotating coil).

Flip coil1.0x1.5 mm!

PM Focusing Triplet

PM Undulator

Hybrid Chicane

Flip-coil measurement of triplet

Laser Accelerator Injection OpticsMatching beam from a conventional rf accelerator into the dielectric structures is a challenge:x x y~100x100 m 2x2 m or less t~0.5 ps = (0.5o at s-band) (10o at =0.8 m) = 0.2 as [attoseconds!]Requiring:3 period undulator (IFEL) and hybrid chicane for microbunching>500 T/m gradient PM quad triplet for microfocus (*=1 mm)

Harmonic Analysis of PMQ Quad

0 1 2 3 4 5

x 10-6

0.5

1

1.5

2

2.5

3

3.5

4

x 106

Freespace Wavelength (m)

Power Radiated (arb. units)

Excitation by short pulse

mesh size 0.25 m

ResonantWavelength1.5m

2 4 6 8

x 10-4

10-5

10-4

Initial Spot Size Entering PMQT

Final Focused Spot Size (m)

PMQ Focusing

HorizontalVertical

*=1mm + Aberratio

ns dominate

Tracking simulation of electron beam spot sizes show ~50% transmission of E163 beam through 1 mm long x 5 m dia. hole.

Total radiated energy:0.16 nJ (~109 ) at 1.5 μm

Optical Injector Tests

Magic 2D simulation of single-particle wake in Bragg fiber

Initial PBG fiber tests will be made by witnessing the radiation spectrum generated in the fiber by an optically pre-bunced beam

Resonant Emission from Optical Structure

e- bunch

Focusing Bunching

=1320 nm

HeNe

Mode filter

OPA light from FEL4

Knife edge/alignment target: Razor blade with white tape on surface

Final focus lens on translation

stage

Pyro

Beam sampler: Fused silica

wedgeSample

Pyro detector OR Ophir head

Microscope slide mounted on translation stage, rotation stage, and vertically translating post holder

ND filter

wheel

Beam sampler

Si diode

CCD

Onset of damage

Silica and silicon show no change in near-IR transmission properties after a ~300 kGy Co60 dose

Telecom Band

Si Bandgap

Silicon Wafer Before (white) and After (black) 314

kGy of Co60

Opt

ical

Tra

nsm

issi

on

Silica Sample Before (white)

and After (black) 295 kGy of

Co60

Both silicon and silica show excellent resistance to laser and radiation damage in the near-IR.

The most efficient lasers are in this wavelength range

Semiconductor lithography is capable of CD tolerances of ~20 nm (/100) now, and is steadily improving; SEM metrology precision is already sub-nm

Excellent optical instruments (optical network analyzers, spectrometers) are available in this range

Damage Studies of Dielectric Materials

Near-IR Laser Damage Threshold Measurements

PUMP

PROBE

• Coupling of electron beam and laser into the same fiber– Explore coupling with sufficient free space

• Measurement of the transmission bandwidth

• Coupling of radially polarized light (TEM*01) into the fiber

– Creation of an accelerating mode

• Measurement of mode profiles– Far field intensity distribution– Near-field distribution at the exit of the fiber

• Michelson interferometer for – Thermal dependence of

phase velocity– Vibration sensitivity

beamsplitter

focusing optics

fiber

focusing optics

mirror

mirror

detector

source

Planned interferometer to measure phase velocity stability

Modeling PBG Band Gaps and Defect-Guided Modes

Core DIA51m

Successfully cleaved PBG fiber

Free-space to fiber coupling setupNear-field mode pattern

Prototype fiber acceleration experiment

Status June 2006

RF PhotoInjector

Ti:Sapphire LaserSystem

Next Linear Collider Test Accelerator

RF System

Cl. 10,000 Clean Room

NLCTA; T’Gun Removed

New Expt. Chambere-

Counting Room(b. 225)

Optical Microbuncher

60 MeV Experimental Hall

Gun Spectrometer

Beamline quads

ESB

• Completed since the last DOE Review (June 2005):– New NLCTA injector (rf gun) installed and commissioned– Extraction line magnets have been completed, and installation has begun– Safety systems (fire, laser, and radiation) for the Experimental Hall have been

installed and are nearing completion– Power & control installation for new beamline is well underway– Developed several ways to improve QE of copper cathodes

• Plans– Commission E163 extraction beamline late summer– Start first science with ITR, IFEL experiments early autumn– Commission optical microbuncher in late 2006/early 2007– Conduct first staging experiments (IFEL bunch, ITR accel) in 2007– Commence PBG microstructure tests

• Silica-fiber based structures• Silicon-based structures

This summer’s commissioning of the E163 beamline will mark the completion of a user facility for advanced accelerator R&D.

Interested users are welcome to submit proposals the the SLAC EPAC.

LEAP/E163 Accomplishments and Plans

SLAC FacultyRobert Siemann (25%)

Staff PhysicistMark Hogan (100%),Spokesperson

EngineerDieter Walz (CEF, 10%)

Non-ARDB SLAC Staff (<10% time)Franz-Josef Decker, Paul Emma, Rick Iverson and Patrick Krejcik

Plasma Wakefield Acceleration in the FFTB (E-164X & E-167)

PIs: Bob Siemann (SLAC), Chan Joshi (UCLA) and Tom Katsouleas (USC)

Postdoctoral RAsRasmus Ischebeck (50%)

StudentsChris BarnesMelissa BerryIan BlumenfeldNeil KirbyCaolionn O’Connell

University Collaborators (Faculty, Physicists and Engineers)UCLA: Chris Clayton, Ken Marsh and Warren MoriUSC: Patric Muggli

University StudentsUCLA: Chengkun Huang, Devon Johnson, Wei Lu and Miaomiao ZhouUSC: Suzhi Deng and Erdem Oz

U C L A

Laser Driven Plasma Accelerators:

• Accelerating Gradients > 100GeV/m (measured)• Narrow Energy Spread Bunches• Interaction Length limited to mm’s

Beam Driven Plasma Accelerators:

Large Gradients:• Accelerating Gradients > 30 GeV/m (measured!)• Interaction Length not limited

Unique SLAC Facilities:• FFTB• High Beam Energy• Short Bunch Length• High Peak Current• Power Density• e- & e+

Scientific Question:• Can one make & sustain high gradients in plasmas for lengths that give significant energy gain?

Laser Driven Plasma Accelerators:

• Accelerating Gradients > 100GeV/m (measured)• Narrow Energy Spread Bunches• Interaction Length limited to mm’s

Beam Driven Plasma Accelerators:

Large Gradients:• Accelerating Gradients > 30 GeV/m (measured!)• Interaction Length not limited

Unique SLAC Facilities:• FFTB• High Beam Energy• Short Bunch Length• High Peak Current• Power Density• e- & e+

Scientific Question:• Can one make & sustain high gradients in plasmas for lengths that give significant energy gain?

Plasma AcceleratorsShowing Great Promise!

U C L A

PWFA:Plasma Wakefield Acceleration

Ez: accelerating fieldN: # e-/bunchz: gaussian bunch lengthkp: plasma wave numbernp: plasma densitynb: beam density

Ez,linear∝Nσ z

2

kpσz ≅ 2 or np ∝1σ z

2For and

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

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

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

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

-

---- ------

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

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

- -- - -- - -

---------

------

electron beam

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

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

- --

--- --

EzEz

AcceleratingDecelerating

Short bunch!

€

kpσ r <<1

Linear PWFA Theory:

Looking at issues associated with applying the large focusing (MT/m) and

accelerating (GeV/m) gradients in plasmas to high energy physics and colliders Built on E-157 & E-162 which observed a wide range of phenomena with both

electron and positron drive beams: focusing, acceleration/de-acceleration, X-ray

emission, refraction, tests for hose instability…

A single bunch from the linac drives a large amplitude plasma wave which focus and accelerates particles For a single bunch the plasma works as an energy transformer and transfers energy from the head to the tail

U C L A

Located in the FFTB

FFTB

PWFA Experiments @ SLACShare Common Apparatus

e-

N=1.81010

z=20-12µmE=28.5 GeV

Optical TransitionRadiators

Li Plasma Ne < 4x1017 cm-3

L≈10-120 cm

Plasma light

X-RayDiagnostic,

e-/e+

Production

CherenkovRadiator Dump

∫Cdt

ImagingSpectrometer

xz

y

EnergySpectrum“X-ray”

25m

CoherentTransition

Radiation andInterferometer

FFTB

Wakefield Acceleration e-

Focusing e-

Phys. Rev. Lett. 88, 154801 (2002)

Beam-Plasma Experimental Results (6 Highlights)

X-ray Generation

Phys. Rev. Lett. 88, 135004 (2002)

0

50

100

150

200

250

300

-2 0 2 4 6 8 10 12

05160cedFIT.graph

ψ= *K L∝ne1/L

Plasma Entrance

=5µm

εN=1×1-5( )m rad

=1.16m

0

100

200

300

400

500

600

0 2 4 6 8 10 12 14

BetatronFitShortBetaXPSI.graph

Plasma OFF

Plasma ON

Envelope

Ψ

=1.4L m

=14µm

εN=18×1-5 -m rad

=6.1cm

=-.6

Phase Advance Ψ ne1/2L

Matching e-

Phase Advance Ψ ne1/2L

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

-8 -4 0 4 8

05190cec+m2.txt 8:26:53 PM 6/21/00impulse model

BPM data

( )mrad

φ( )mrad

1/sin

≈o BPM Data– Model

Electron Beam Refraction at the Gas–Plasma Boundary

Nature 411, 43 (3 May 2001)

Wakefield Acceleration e+

Phys. Rev. Lett. 90, 214801 (2003)Phys. Rev. Lett. 93, 014802 (2004)

Phys. Rev. Lett. 93, 014802 (2004)

0

0.4

0.8

1.2

1.6

-100 -50 0 50 100CombinedCTRInterferogramsSm 0

10

20

30

40

50

60

0 5 10 15 20 25 30 35 40

SigmazMylar12.7_3WandBS

Autocorrelationz( )µm

/ w o Filtering

w Filtering

z≈9 µm

≈60 fs

z ≈ 9 µm

z≈18 µm

GaussianBunch

or

First Measurement of SLAC Ultra-short Bunch Length!

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

Autocorrelation:

CTR Michelson Interferometer• Fabry-Perot resonance:=2d/nm, m=1,2,…, n=index of refraction• Modulation/dips in the interferogram• Smaller measured width:Autocorrelation < bunch !• Other issues under investigation:- Detector response (pyro vs. Golay)- Alternate materials:HDPE, TPX, Si, Diamond ($$$)

CTR Michelson Interferometer• Fabry-Perot resonance:=2d/nm, m=1,2,…, n=index of refraction• Modulation/dips in the interferogram• Smaller measured width:Autocorrelation < bunch !• Other issues under investigation:- Detector response (pyro vs. Golay)- Alternate materials:HDPE, TPX, Si, Diamond ($$$)• “All Silicon” CTR scanning interferometer.• Eliminates many of the material dependent features

• “All Silicon” CTR scanning interferometer.• Eliminates many of the material dependent features

U C L A

Plasma Source Starts withMetal Vapor in a Heat-Pipe Oven

€

E = 6GV /mN

2x1010

20μ

σ r

100μ

σ z

Peak Field For A Gaussian Bunch: Ionization Rate for Li:

See D. Bruhwiler et al, Physics of Plasmas 2003

Space charge fields are high enough to field (tunnel) ionize - no laser!- No timing or alignment issues- Plasma recombination not an issue

- However, can’t just turn it off! - Ablation of the head

0 0Pressure

zLi HeHe

Boundary Layers

OpticalWindowCoolingJacketCoolingJacket

HeaterWick

InsulationPumpHe

OpticalWindow

L

0 0 zLi HeHe

Boundary Layers

OpticalWindowCoolingJacketCoolingJacket

HeaterWick

InsulationPumpHe

OpticalWindow

L

Summer 2004: • Single electron bunch drives then samples all phases of the wake resulting in large energy spread• Future experiments will accelerate a second “witness” bunch• Electrons gained > 2.7GeV over maximum incoming energy in 10cm!• Confirmation of predicted dramatic increase in gradient with move to short bunches• First time any PWFA gained more than 1 GeV• Two orders of magnitude larger than previous beam driven results

Summer 2004: • Single electron bunch drives then samples all phases of the wake resulting in large energy spread• Future experiments will accelerate a second “witness” bunch• Electrons gained > 2.7GeV over maximum incoming energy in 10cm!• Confirmation of predicted dramatic increase in gradient with move to short bunches• First time any PWFA gained more than 1 GeV• Two orders of magnitude larger than previous beam driven results

Summer 2005:• Increased beamline apertures• Plasma length increased to 30cm• Energy gain >10GeV• Scales linearly with length

Summer 2004: • Single electron bunch drives then samples all phases of the wake resulting in large energy spread• Future experiments will accelerate a second “witness” bunch• Electrons gained > 2.7GeV over maximum incoming energy in 10cm!• Confirmation of predicted dramatic increase in gradient with move to short bunches• First time any PWFA gained more than 1 GeV• Two orders of magnitude larger than previous beam driven results

Summer 2005:• Increased beamline apertures• Plasma length increased to 30cm• Energy gain >10GeV• Scales linearly with length

…but moving forward will

require spectrometer

redesign to transport larger

energy spread

U C L A

April 2006:“The Last Hurrah!”

1. Constructed a meter long

plasma source

2. Raised linac energy to 42GeV

3. Installed spectrometer dipole

and temporary beam stopper

immediately after the plasma

4. Two screen energy diagnostic

1. Constructed a meter long

plasma source

2. Raised linac energy to 42GeV

3. Installed spectrometer dipole

and temporary beam stopper

immediately after the plasma

4. Two screen energy diagnostic

At the 2005 DOE Review we set an ambitious goal for the coming year:“Make the highest energy electrons ever at SLAC!”

Sorry, this image is part of a paper being prepared for a journal with strict embargo policies and cannot be put out on public ftp until it’s published.

U C L A

Effective Plasma LengthLimited By Head Erosion to ~90cm

A Simulation to Illustrate the Idea of Head Erosion(not current experimental parameters)

QuickTime™ and aYUV420 codec decompressor

are needed to see this picture.

Solution will likely involve either a low density pre-ionizationor integrated permanent magnet focusing

Solution will likely involve either a low density pre-ionizationor integrated permanent magnet focusing

Trapped Particles (Part 1):Electrons Are Trapped at He Boundaries and Accelerated Out of

the Plasma

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

Trapped Particles

Li Oven Heaters

Plasma LightSpectrograph

Dipole

Mask

Two Main Features• 4 times more charge• >104 more light!

Two Main Features• 4 times more charge• >104 more light!

Two energy populations (MeV & GeV)Two energy populations (MeV & GeV)

Note: Primary beam is also radiating!

Trapped Particles (Part 2):Visible Light Spectrum Indicates Time Structure of Trapped

Electrons

€

=2π

Bunch Spacing = cτ ≈ 70 μ,

plasma wavelength, λ p = 64 μ .

€

=2π

Bunch Spacing = cτ ≈ 70 μ,

plasma wavelength, λ p = 64 μ .

OSIRIS Simulations:• He electrons in several buckets• Spaced at plasma wavelength• Bunch length ~fs

OSIRIS Simulations:• He electrons in several buckets• Spaced at plasma wavelength• Bunch length ~fs

U C L A

Future ExperimentsNeed an FFTB Replacement

SABER (South Arc Beamline Experimental Region):

5.7GeV in 39cm

Evolution of a positron beam/wakefiled and final energy gain in a self-ionized plasma

€

Nb = 8.79 ×109,σ r =11μm, σ z =19.55μm, np =1.8 ×1017cm−3

Three Phases:1. Short e- early as 20072. Short e-/e+ 20083. Bypass line 2009

Three Phases:1. Short e- early as 20072. Short e-/e+ 20083. Bypass line 2009

Still interesting work to be done with electrons, but…Short Pulse e+ Are the Frontier

Still interesting work to be done with electrons, but…Short Pulse e+ Are the Frontier

U C L A

Over the past 5 yearsOver 20 Peer reviewed publications covering all aspects of beam plasma interactions: Focusing (e- & e+), Transport, Refraction, Radiation Production, Acceleration (e- & e+)

E-167 Accomplishments

Plasma Wakefield AcceleratorResearch Summary

Future Plans:Experiments @ SABER

Diagnostic Development: Measurement of SLAC

Ultra-short Electron Bunch

Understanding PhysicsOf Trapped Electrons in

Self-Ionized PWFA

Sorry, this image is part of a paper being prepared for a journal with strict embargo policies and cannot be put out on public ftp until it’s published.

A rich experimental program in advanced

accelerator research is ongoing at SLAC

Primarily looking at issues associated applying

lasers (E-163) and plasmas (E-167) to high energy

physics and colliders

Through strong collaborations with University

groups, SLAC has developed not only facilities for

doing unique physics, but also many of the techniques

and the apparatus necessary for conducting these

experiments

New facility in ESB/NLCTA about to turn on with E-

163

Need an FFTB replacement - SABER

“Build it and they will come…”

Summary