X-Ray Free Electron LasersJ.S. WurteleUCB and LBNL

N

S

S

NS

SN

N

e-

w

Light

Davidson SymposiumPPPL

June 12, 2007

X-RayFELs

Two Livingston Plots

Particle accelerators Light Sources

Panofsky EU FEL

X-Ray FELs

Goals:• High average flux• High peak power • Temporal coherence• Spatial coherence• Attosecond pulses• Synchronization • Flexibility

• Implications (current technology): Large machines, GeV Energies

• Critical Physics• Optical manipulation of phase

space• High brightness beam generation

and preservation• Wiggler technology

Evolution of synchrotron radiation sources

X-ray sources expand

LCLS [SLAC] FEL

JAPAN [SPRING 8]

€

Current ~3.5kAEnergy ~13.6GeV Repetition rate

~120Hz

Peak X-Ray Power

~8GW

EU XFEL [DESY]

Vision for a future LBNL light source

ALSFEL array at the Bevatron site

Injector

Linac in tunnel

Vision for a future light source facility at LBNLVision for a future light source facility at LBNL A HIGH REP-RATE, SEEDED, VUV — SOFT X-RAY FEL ARRAY

Low-emittance, high rep-rate electron gun

Array of configurable FELsIndependent control of wavelength, pulse duration, polarizationConfigured with an optical manipulation technique; seeded, attosecond, ESASE

Laser systems, timing &

synchronization

Beam manipulation

and conditioning

Beam distribution and individual beamline tuning

~2 GeV CW superconducting linac

FEL BASICS

N

S

S

NS

SN

N

e -

w

Light

Spread in this term is harmful!

Limits

What drives X-ray FELs towards large energy electron beams?

1. Coherent emission--bunching at X-ray wavelengths

2. Limits on our ability to create and propagate high brightness electron beams

3. Limits on our ability to build short wavelength wigglers

z

undulator

zz

Dephasing from transverse motion

• allows relaxed emittance requirement in FEL--but we do not know how to produce required conditioning in a practical system (yet)

ΔE/E

with conditioning

€

+κJ⊥

€

dψdz

=

€

2kwγ − γ rγ r

€

J⊥

€

J⊥

€

J⊥

We are limited by our inability to make high quality beams

SASE FEL: amplification of fluctuations

Single pass synchrotron radiation spectrum (Catravas, et al, @BNL/ATF,)

SASE spectrum and temporal shape has spikes--poor longitudinal coherence

Seeded FEL ENHANCED CAPABILITIES FOR CONTROL OF X-RAY PULSE

Electron beam is 1.5 GeV, energy spread 100 keV, 250 A current, 0.25 micron emittance; laser seed is 100 kW at 32 nm; undulator period 1 cm

SASE

25 fs seed

500 fs seed

0.4

0.3

0.2

0.1

0.0

Photons/meV (X10

9)

12421241124012391238

Photon Energy (eV)

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

Photons/meV (X10

9)

12421241124012391238

Photon Energy (eV)

30

25

20

15

10

5

0

Photons/meV (X10

9 )

12421241124012391238

Photon Energy (eV)

Spectrum

0.6

0.5

0.4

0.3

0.2

0.1

0.0

Power (GW)

-700 -600 -500 -400 -300 -200 -100

Time (fs)

0.6

0.5

0.4

0.3

0.2

0.1

0.0

Power (GW)

-700 -600 -500 -400 -300 -200 -100

Time (fs)

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

Power (GW)

-700 -600 -500 -400 -300 -200 -100

Time (fs)

Pulse profile

Seeded FEL close to

transform limit No monochromator

Phase space manipulationManipulate beam phase space can have many

advantages: Enhanced gain Seeding radiation pulse for harmonic cascadesAttosecond pulsesSynchronizationRelax beam quality constraints (conditioning)Lower energy for given wavelength

Many of these ideas are realized by laser interactions with the electron beam prior to the radiation generation. Some examples…

slice ~ 1 fs

e-beam ~ 100 fs

Lasers manipulate longitudinal phase space

during interaction in wiggler

This cartoon is realized by manipulation of beam phase space with short pulse lasers. The idea is to condition and select specific slices of electrons to radiate differently (in direction, frequency, intensity, etc.). For synchrotron sources this has already been accomplished: Zholents & Zoloterev (1996); Schoenlein, et al, 2000; Khan, Part. Acc. Conf. 2005. For FEL see Zholents et al (2003-2007)

Harmonic cascade seed

Laser pulse ~ 5 fs

)t(E

Dispersive section introduces bunching

High Gain Harmonic Generation (HGHG) – seed with a laser pulse and radiate at a harmonic

L.-H. Yu et al, Science 289 932-934 (2000)L.-H. Yu et al, Phys. Rev. Let. Vol 91, No. 7, (2003)

dispersive chicane

phase

Input Outpute- beam phase space:

Laser modulates e-beam energy

Ener

gy

€

−π

€

πBunched beam radiates strongly

at harmonic in a downstream undulator

€

nπ

€

−nπ

ModulatorShort wiggler

laser pulse

e- bunch

€

0λ Radiator

Longer wiggler€

0λ n

Extends energy reach, lower power

Seed laser pulse

Tbunch >> TMO

PMO >> Pshot

FEL modulatorLW < LSATStrong

bunching

3rd - 5th harmonic radiator

3 - 5th harmonic FEL modulator /

low gain amplifierLW < LSAT

3rd - 5th harmonic radiator

Cascaded harmonic generation scheme

Delay bunch in micro-orbit-bump (~50 m)

Low electron pulse

Unperturbed electronsE ~ E (0)

seed laser pulse

tail head

radiator radiatormodulatormodulatordisrupted

region

HHG laser seed--an alternative to harmonic cascades

Example with seed at 30 nm, radiating in the water windowFirst stage amplifies low-power seed with “optical klystron”

More initial bunching than could be practically achieved with a single modulatorOutput at 3.8 nm (8th harmonic)

300 MW output at 3.8 nm (8th harmonic) from

a 25 fs FWHM seed

1 GeV beam500 A

1.2 micron emittance75 keV energy spread

Gullans et al. (2007)

Modulator30 nm, L=1.8 m

Modulator30 nm, L=1.8 m

Radiator3.8 nm, L=12 m

100 kW=30 nm

Courtesy H. Kapteyn

Or, X-ray laser seed

Gun Beam manipulation

FELs

linac

FEL performance is governed by beam brightness:

Brightness = # electrons/6D-phase space volume

This number will NOT get larger---determined bygun physics and can grow through various instabilities

RCD circa ‘89

Gullans et al, 2007

19

Conventional Linac

Ez : 10 - 200 MV/ mLint : km's

Δ

~ K l y s t r o n M i c r o w a v e

P o w e r S o u r c e

W a v e - g u i d e

s t r u c t u r e

Δ~

• E • •

h ′ω

W = e Ez Lint

0 1 02 0

3 04 0

5 06 0

la s e r p u ls e

Electron beam surfing on plasma electric field

( B. Shadwick, UCB/ CBP)

Laser dr iven plasma based linacEz : 1 0 - 1 0 0 GV/ m

Lint : laser dif fract ion lengt h

Plasma-based Electron Linac

1014

1015

1016

1017

1018

1019

1020

1021

1022

1023

1024

1025

1026

Average Brightness [Ph/(s 0.1% BW mm

2

mrad2)]

10-5 10-4 10-3 10-2 10-1 100 101 102 103

X-Ray Pulse Duration [ps]

2nd Generation Storage Rings

3rd Generation Storage Rings

Cornell ERL (6 keV)

European XFEL (12 keV)

Seeded FEL, 750 fs (1.2 keV)

LCLS (8 keV)

BESSY FEL (1 keV)

FLASH (200 eV)

Seeded FEL, 50 fs (1.2 keV)

Seeded FEL, 100 as (1.2 keV)

Performance comparison PARAMETER SPACE COMPLEMENTS OTHER FACILITIES

Many people within LBNL contribute to new light sourceMany people within LBNL contribute to new light source

Walter BarryDan BatesKen BaptisteAli BelkacemJohn ByrdChris CelataChris Coleman-SmithJohn CorlettStefano DeSantisLarry DoolittleRoger FalconeBill FawleyGraham FlemingMiguel FurmanTom GallantMike GreavesSteve GourlayMichael GullensGang HuangZahid Hussein

Preston JordanJerry KekosJanos KirzJim KrupnickSlawomir KwiatkowskiSteve LeoneDerun LiSteve LidiaSteve MarksBill McCurdyPat OddoneHoward PadmoreEmanuele PedersoliGregg PennDave PlateIlya PogorelovJi QiangAlex RattiIna ReichelDavid Robin

Kem RobinsonGlenna RogersRob RyneFernando SannibaleBob SchoenleinAndy SesslerKiran SonnadJohn StaplesChristoph SteierJean-Luc VayMarco VenturiniWill WaldronWeishi WanRussell WellsRussell WilcoxJonathan WurteleSasha ZholentsMike ZismanMax Zolotorev

Extras

800 nm

spectral broadening and

pulse compression

e-beam

harmonic-cascade FEL

one period wiggler tuned for FEL

interaction at 800 nm

2 nm light from FEL

2 nm modulator chicane-buncher

1 nm radiator

dump

endstation

1 nm coherent radiation e-beam

endstation

time delay chicane

Potential for attosecond x-ray production

e-beam

Zholentz and Fawley PRL 2004

X-rays from plasma sources

Already demonstrated—beams make x-rays More elaborate ideas based on ion channels [Whittum;

E157SLAC]

Rousse et al, PRL 04

Many groups worldwide are working on thisPlasma yield naturally short pulses, but hard to reachFEL intensities with spontaneous emission [N vs N^2]

The Advanced Photoinjector Experiment – APEX*)

*) J. Staples, F. Sannibale, S. Virostek, CBP Tech Note-366, October 2006

BeamDump

Coaxial Gun Cavity

Current Monitors

Solenoid &Trim CoilPackages

LaserPort

Beam Position Monitor

Retractable Cerenkov Monitor

Pepper Pot &Faraday Cup

RetractableCerenkov Monitor

& Faraday Cup

Cathode Mounted on Coaxial Center

Conductor

Frequency 65 MHzField 12-25 MV/mRF power at 20MV/m 70 kWPeak wall power density 8 W/cm2Vacuum 10-11 Torr200 MHz is also under consideration

High repetition rate RF photocathode gun