FEL Activities at SINAP and FEL Stabilization

Considerations

Jianhui Chen (SINAP, CAS)On behalf of the FEL Team

October 19, 2012

7th International Conference on Mechanical Engineering Design of Synchrotron Radiation Equipment and Instrumentation (MEDSI 2012), October 15-19, 2012, Shanghai

To the organizers of MEDSI for invitation to this fantastic workshop

To the whole FEL team at SINAP: There are many NOT in the photos, my apology in advance.

Acknowledgements

Outline

Introduction to SINAPFEL activities at SINAP (limited to FEL projects and exps.)

SXFELSHXFELDCLSSDUV-FEL

FEL stabilization considerations (limited to orbit related issues)

RequirementsCountermeasures

Summary and conclusions

SINAP: Transforming……

Shanghai Synchrotron Radiation FacilityA world leading intermediate-energy 3rd generation light source

626

1238

747575

1032

516

91.9%83.4%

69.1%

0100200300400500600700800900

1000110012001300

2009 2010 2011

Num

ber o

f Pro

posa

ls

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

Approved/S

ubmitted

SubmittedApprovedApproved/Submitted

5958

12813.5

6899

26784048

1601

44.9%

31.6%

23.2%

0100020003000400050006000700080009000

10000110001200013000

2009 2010 2011

Num

ber o

f Shi

fts

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

50.0%

Allocated/R

equested

RequestedAllocatedAllocated/Requested

Gap

λu

undulatorelectron beam

photo cathode

accelerator

bunch compressoraccelerator

laser-pulse

What’s Next?

Compared to 3rd generation storage ring based synchrotron radiation facilities, the gain factorsare:

• peak brilliance: 109 at the FEL line

• average brilliance: 104 at FEL line (EUR)

• coherence: 109 at FEL line• pulse duration: 10-3 at FEL line

The XFEL will not replace the storage ring facilities,it opens the door to new science

Properties of XFEL radiation

XFELspontaneous

radiation

Peak brilliance

With an FEL one gets in a pulse of ~25 fsduration as many X-ray photons as with a

modern storage ring in 1 sec

Soft X-ray FEL

SDUV-FEL Test Facility

Hard X-ray FEL

2011 2012 2013 2015 2016 2017 2018 2019 202020142010

FEL strategy at SINAP

EEHG Cas. HGHG Users

Desi. Operation

Design R&Ds Operation

100

10

0.1

nm

Cons. Comm.

Comm.

Beamline

Construction

Shanghai photon science center

Compact XFELSXFEL

SXFEL in the SSRF campus

Approved Feb. 2011

SXFEL Main Parameters

Parameters HGHG Upgrade Unit

Output Wavelength 9 3 nmBunch charge 0.5~1 0.5~1 nCEnergy 0.84 1.2~1.3 GeVEnergy spread 0.1~0.15% 0.15%Energy spread (sliced) 0.02% 0.03%Normalized emittance 2.0~2.5 2.0~2.5 mm.mradPulse length (FWHM) 1. 1 psPeak current ~0.5 0.5 kARep. rate 1~10 1~10 Hz

Parameters Value Unit

Output Wavelength 0.1 nmBunch charge 250 pC

Energy 6.4 GeVNormalized emittance 0.4 mm.mradEnergy spread (sliced) 0.01%

Pulse length (Full) 100 fsPeak current 3 kA

Rep. rate 60 HzFEL parameter 3.41*10-4

Peak power 10 GWPeak brightness 2*1033

3D gain length 2.156 mSaturation length 50 m

Shanghai Hard X-ray FEL

λm=5cmNm=10

λm=2.5cm

100~500A1~2mm.mradσz =400fs

50~150nm15~85uJ1ps/130fs (FWHM)

EUV FEL for DICP, CASEUV FEL for DICP, CAS

FEL projects in Asia

PAL XFEL

DCLS

SXFEL

SACLA

SDUV-FEL Experiment Hall

Shanghai Deep UV FEL (SDUVShanghai Deep UV FEL (SDUV--FEL)FEL)

• 2009.04-08: Linac commissioning • 2009.09-12: SASE experiment• 2010.01-03: Seeded FEL Installations• 2010.05: Seeded FEL experiments start • 2010.05.17: HGHG signal• 2010.05.22: First Echo signal (‘double-peak’) • 2010.07-08: Install. for high harmonics EEHG• 2010.12: HGHG saturation• 2011. 04: EEHG amplification• 2011.07-08: Cascaded HGHG experiments begin• 2012.04: First cascaded HGHG signal

Milestones of FEL experiments

Beam energy 100-150MeV

Beam energy spread (projected)

<0.03%

Normalized emittance 4~5mm-mrad

Bunch Length (rms) 1~3ps

Bunch charge 100~300pC

Main Parameters of SDUV-FEL Linac

Linac Commissioning Results

100~150MeV4~5mm.mrad……

SASE Results (2009)

350 370 390 nm

SASE Spectrum

Period：25mmTotal length： 9 m

Gain length：~0.9mSaturation length：~20 m

~135MeV~370 nm

L.H. Yu, PRA 44, 5178 (1991)

HGHG principle

HGHG saturation, Dec. 2010

Gain curveFEL profile

ExperimentsS2Esimulations

G. Stupakov, PRL 102, 074801 (2009)

EEHG rationale

First lasing of EEHG FEL, April 2011

FEL profile

Gain curves

SpectrumExperiments

Simulation

A famous drawing from the 1970’s showing the relative importance of experiment vs theory

2-Stage cascade HGHG

Cascaded HGHG

1st Stage 2nd Stage

Final AmplifierSeed

Final Amplifier

nλλ =1 mn×

=λλ 2 2λλ =f

Seed-Laser λ

The world first 2The world first 2--stage cascaded HGHG FELstage cascaded HGHG FEL

150MeV

1200nm600nm 300nm

fresh bunch technique

CH01CV01

B15-18

Q19CH19CV19

CH04CV04

Q16Q13Q14

Q15

DS1-1B3

Q18 Q17DS1-2B4-B7Q30 UDR

Q20CH20CV20

Q21CH21CV21

Q22CH22CV22

Q23CH23CV23

Q24CH24CV24

PMUsRad

B8CH02CV02Q1-1Q1-2Q1-3

ADC-2 ADC-1

MC3-C

MC3-A

SeedInj-1SeedInj-2MC3-B

MC3-D

MC5-B MC5-A

MC5-CMC5-D

BPM4BPM5

UN-POP3/4 UN-POP9/10

UNPRF0ADC-POP2 ADC-POP1

UN-POP1/2 UN-POP5/6 UN-POP7/8 UN-POP11/12

CCD MotorUNPRF1UNPRF2UNPRF3UNPRF4

Coll.1Coll.2

UNPRF5

UNPRF6 UNPRF7CBPM

MOD2

Delay LineB19-B21

DS2B22-B24 B9

The two-stage cascaded HGHG FEL

Stage-1

Stage-2

First results from a twoFirst results from a two--stage cascaded HGHG FELstage cascaded HGHG FEL

Paper in preparation

Stability can maximizethe performance of a light source !

Courtesy of Hitoshi Tanaka

Stability can maximize the performance of a light source !

Courtesy of Hitoshi Tanaka

As Shintake urged in FEL’04, a linac-based light source is essentially unstable without an autonomous stabilization mechanism

Strategy and efforts on the beam-stabilization is more crucial for linac-based sources rather than for ring-based ones.

Spinning-Top:

VS

Archery: Linac-based

Ring-based

Courtesy of Hitoshi Tanaka

1.3775 1.378 1.3785 1.379 1.3795 1.38 1.3805 1.381 1.3815

x 104

0

1000

2000

3000

4000

5000

6000

7000

8000

1.3775 1.378 1.3785 1.379 1.3795 1.38 1.3805 1.381 1.3815

x 104

0

1000

2000

3000

4000

5000

6000

7000

8000

Relative Electron Energy (%)-

Seeded FEL Power Jitter vs. Seeded FEL Power Jitter vs. ee−− Beam Energy JitterBeam Energy Jitter

13.790GeV13.790GeV= 8.33 keV= 8.33 keV

0.05% = 8 eV0.05% = 8 eV

A. Lutman

FEL

Seed

ed P

ower

(arb

.)

>50% FEL power jitter

even at fixed energy

(20% best)…Expect ~10%

>50% FEL power jitter

even at fixed energy

(20% best)…Expect ~10%

6% of shots not seeded

due to e−

energy jitter (0.04% rms)

6% of shots not seeded

due to e−

energy jitter (0.04% rms)

(220)

• Transverse stability• Low energy (injector): transverse emittance

dilution(depends on the RF-frequency: X-band more sensitive than L-band)

• Beam break-up at high average current operation• High energy (undulator): gain reduction and

wavelength jitter

• Longitudinal stability (bunch / RF phase)• Energy fluctuations• Current fluctuations FEL gain fluctuations• Temporal jitter with synchronized sources

Electron-Beam Stability

typically a (few) micron orbit accuracy

Noise sources: Transverse

Steering coil current regulation

Quadrupole magnet transverse vibration

Quadrupole current regulation with misalignments

Wakefield kicks with charge jitter

CSR kicks with charge or bunch length jitter

Drive laser pointing jitter

Longitudinal Loops Stabilize:Longitudinal Loops Stabilize:DL1 energyDL1 energyBC1 energyBC1 energyBC1 bunch lengthBC1 bunch lengthBC2 energyBC2 energyBC2 bunch lengthBC2 bunch lengthFinal energyFinal energy

Transverse Loops Stabilize:Transverse Loops Stabilize:Laser spot on cathodeLaser spot on cathodeGun launch angleGun launch angleInjector trajectoryInjector trajectoryXX--band cavity positionband cavity positionLinac & LTU trajectory (5)Linac & LTU trajectory (5)Undulator trajectoryUndulator trajectory

Laser & Electron-Based Feedback Systems

L0L0

gungun

L3L3L2L2XXDL1 BC1 DL2

L1L1

σσzz11

δδ11ϕϕ11 VV11

σσzz22

δδ22ϕϕ22 VV22

δδ33

VV33

δδ00VV00

BC2

BPMsBPMsCER detectorsCER detectors

Steering LoopSteering LoopLaserLaser

D. Fairley, J. Wu, LCLS@SLACD. Fairley, J. Wu, LCLS@SLAC

• Electron beam orbit must be controlled over a few gain-lengths.

• Within this distance: gain reduction

• Over larger distances the overlap between optical field and the electron beam may disappear. In such a case, the micro-bunching will re-initiate the FEL process over a few undulator periods only.

Transverse Orbit / Undulator

gcc

gg

Lx

xLL

/,/

)1/( 2'

λθθθ ==

−=

T. Tanaka et al., NIM A 528 (2004) 172

Reduction of radiation efficiencySmearing of microbunching

LCLS Req. Specifications

• <2um over one gain length (~5m)

Summary and conclusions

FEL activities at SINAPSXFEL: approved in Feb. 2011, design studies underwayDCLS: approved and funded, design studies underway, construction will begin late 2012SDUV-FEL: nice test bed, however stability improvement highly desired

FEL stabilization considerationsA lot of engineering challenges ahead

Your inputs are greatly appreciated.

谢谢