Beamline Instrumentationfor Precise Characterization

of X ray FELs

1

Makina YABASHIRIKEN SPring 8 Center (RSC), Beamline R&D Group

yabashi@spring8.or.jpAugust 27, 2013 @New York

FEL13

SPring 8 Angstrom Compact free electron LAser

Contents

1. Introduction– Importance of x ray diagnostics

2. Photon diagnostics (I): pulse energy & transversephase space

3. Photon diagnostics (II): Longitudinal phase space4. Summary

2

Why are photon diagnostics important forXFEL facilities ?

• Photon beam is a final product of XFEL source to userexperiments– e.g. studies on X ray nonlinear interactions

• Photon beam contains a number of key information formachine operation– could be more accurate and robust than electrons

• Photon diagnostics become strong linkage betweenmachine and users

3MachineMachine UsersUsersPhoton

diagnosticsPhoton

diagnostics

Objective

1. X ray beam profiles in 6 D phase space• Total photon number (pulse energy)• Spatial profile, centroid• Wavefront/Angular divergence• Spectrum• Temporal profile• Arrival timing to external laser• Polarization

2. Key• Shot to shot (non average)• Nondestructive (for users)• Wide dynamic range (spontaneous to fully saturated x rays)

4Introduce experiences at SACLA

5

SACLA @SPring 8

Compact XFELConstruction: FY2006~2010First lasing: June 2011User Operation:March 2012~

Experimental Building

6

Beamline

7

EH4: Open hutch

SACLA SPring 8Exp. Facility

Laser booth(CPA, OPA)Power amp: ~ 3 TW

BL3

5 beamlines (final)Start with BL3 (XFEL) & BL1 (SX)

EH1: R&D, beamconditioning optics

EH2: Pump & Probe (CPA/OPA)w unfocused beam

EH3: 1 um focusing(Imaging/Non linear)

BL1

OH: Common optics& diagnostics

Common X ray optics anddiagnostics have beencentralized in Optics HutchExperimental stations provideonly basic infrastructure (e.g.,optical laser, focusing system)Empty space for various

experimental instruments

Comparison of BL design concept:SACLA vs. LCLS

8

Brand new machine with unknownelectron beam propertiesMitigate risks by introducingphoton diagnosticsUtilize X ray diagnostics in opticshutch for both machine and users

Machine can operate withmatured e diagnosticsX ray optics & diagnostics arelocated in every experimentalhutch (only for users)

from LCLS web page

9

X ray beamline (optics and photon diagnostics) played a key role forenabling quick lasing in two months

Nat. Photon 6, 540 (2012)

Example: Commissioning of SACLA

10

Precise ID tuning using beamline optics/diagnostics

Overlap between e and p beam with an accuracy of ~urad isrequiredE beam trajectory easily bends at the edges of undulator modulesP beam goes through along the e beam in the undulatorP beam can be used to monitor the e beam trajectory in each

undulator moduleCf. central cone of the undulator radiation

1/g= 1/16000 = 60 urad for 8 GeV

11

Key components for SACLA commissioning

DCM Si(111)High stability;UHV compatible(Ohashi et al.)

single sensorMPCCD(Hatsui et al)

12

Spatial profile of monochromaticspontaneous radiation

~1 mm

Angular diameter: ~10 uradCentroid: 1 urad

Profile of monochromatic spontaneous rad

Photon energy

2 2

2 2 1 12 cos 2 2 2Ru u

U

2 2

2 2 1 12 cos 2 2 2Ru u

T. Takashi et alPRST 12

Pulse Energy 0.4 mJ @10 keVPhoton energy range 4.5 to 15 keV Stability Intensity I/I < 10%

Pointing 3 ~7% of beam sizeRepetition rate 20 Hz (Max. 60 Hz)

13

SACLA: Summary of Performance

Contents

1. Introduction2. Photon diagnostics (I): pulse energy &

transverse phase space3. Photon diagnostics (II): longitudinal phase space4. Summary

14

15

Pulse energy measurement

Pulse to pulse, wide dynamic range w/o saturation

Ion chamber using gas ionizationsemi transparent in wide wavelength rangeabsolute measurementneed care to nonlinearityneed cost space for vacuum system

Foil based backscattering monitorsemi transparent above a few keVneed care to damage to foil

Calorimeterabsolute measurementdestructiveaverage measurement

16

Pulse energy diagnostics:Thin foil backscattering monitor

Geometry

X rays

y

Semi transparentfilm

Photodiode

x

y

U

L

D

R

Quadrant photodiodesHamamatsu, S3590 09t300 um, 10 x10 mm2

xxx IKIIIIKx

RL

RL

yyy IKIIIIKy

DU

DU

ComptonScattering

Slope: KxL R U DI I I I IIntensity

Position

Collaboration with LCLS

Tono et al., RSI 82, 023108 (2011)

CVD ”nano” DiamondSmooth Debye Scherrer ringSpeckle freeSemi transparent (>97%@10 keV)Wide d range (10 nJ ~ 10 mJ)

Application (1): Gain curve measurement

17

Lg ~ 2.3 m

Lg ~ 4.6 m (2011 autumn)

Linear ScaleLog Scale

18

Application (2)Routine monitoring at control room

Pulse energy (uJ)

Vertical position (um)

Horizontal position (um)

Application (III)Monitoring of Beamline Transmittance

19

Mirror1

Mirror 2a(

Photon energy /keV

Pulse energy / J

Radiometer XGMD4.4 32.26 0.35 32.9 2.05.8 104.2 1.3 106.6 6.19.6 95.3 2.3 93.9 6.113.6 42.2 1.1 40.8 2.916.8 0.96 0.03 ---

20

“From Relative to Absolute”International Collaboration

20T. Tanaka et al., NIMA 659, 528 (2011).K. Tiedtke et al., JAP. 103, 94511 (2008).

Thin-foil monitor(SACLA)

Calorimeter(AIST)

RT-type:Worked!

Gas monitor detector(DESY/PTB)

Nov. 24, 2011

Max. Deviation = 3.3%Kato et al, Appl. Phys. Lett.101, 023503 (2012)

New portable calorimeter operatingat room temperature

RT-type (2013/4)Cryo-type (2011/11)

Saito san(AIST)

Tanaka, Saito et al, in preparationPush to commercially available device

Developed by AIST and RIKEN, tested on April 2013

Spatial Profile Diagnostics (I)

22

Ce:YAG (300 um t)high efficiency (~ 1 uJ/pls)destructiveCVD B doped diamond (30 um t)low efficiencysemi transparent (>97% for 10 keV)speckle free> Available as inline, routine monitors

Retractable screens Phosphor coupled MCP monitor

Very high efficiency to imagespontaneous radiationRoutine use for undulator

alignment in every two weeks

Spatial Profile & Pointing Stability

23

Temperature stabilization of injector section (2012 autumn)

July 25, 2012 Sep. 18, 2012

V40 um, H50 um

V10 um, H20 um

Spatial Profile Diagnostics (II)

24

Slit/knife edge scanning methodStability ? (intensity/position)

1 um Focusing at EH3

-0.1

0.1

0.3

0.5

0.7

0.9

1.1

0 10 20 30 40 500

0.02

0.04

0.06

0.08

0.1

0.12

0 10 20 30 40 50

II0 I / I0

Wire position /a.u.

Sig

nal /

a.u.

Wire position /a.u.

First derivative

Yumoto et al Nature Photon. 7, 43 (2013)

25

Measurement of nano scale spotwith two stage focusing system

0.0

0.2

0.4

0.6

0.8

1.0

-600 -300 0 300 600

Inte

nsity

(arb

. uni

t)

Position (nm)

Beam profileWire scan profileGaussian 45nm(FWHM)

0.0

0.2

0.4

0.6

0.8

1.0

-600 -300 0 300 600In

tens

ity (a

rb. u

nit)

Position (nm)

Beam profileWire scan profileGaussian 55nm(FWHM)

Vertical direction Horizontal direction

45nm(FWHM)

55nm(FWHM)

X ray energy: 10keVMimura et al, in preparation

~1020 W/cm2

Beamline Optics & Diagnostics

Photodiode & Screen monitor x7intensity/profile

Gas Monitor x1

Thin-foil monitor x2

Mirror 1

Mirror 2a Mirror 2b

Monochromator

Slit

In-line spectrometer x1

Tono et al., New J Phys. 15, 083035 (2013)

Contents

1. Introduction2. Photon diagnostics (I): Intensity & transverse phase space3. Photon diagnostics (II): Longitudinal phase space4. Summary

27

28

70

60

50

40

30

20

10

0

2 (d

eg)

20181614121086

photon energy (keV)

Dia(111) Dia(220) Dia(311)

111

220

311

Spectral range 6~20 keV

MPCCD

XFEL

Thin Diamond film

Central wavelength monitor

phot

on e

nerg

y

shot number

300 eV

Inubushi sanDiffraction ring from CVD diamond

HighWavelength StabilityRoutinely Achieved

29

10.0

9.9

9.8

9.7

9.6

9.5

phot

on e

nerg

y (k

eV)

200150100500

shot number

E

XFEL 1D focusPulse energy < 10 uJ

S. M. Durbin, et al., X ray pump optical probe cross correlation studyof GaAs Nature Photonics 6, 111

Conduction band

Egap < EphotonAbsorption

Valence band

Egap > EphotonTransmission

30

ArrivalTiming

Transient transmission of optical laser inducedwith X rays; Spatial decoding

Sato san (SACLA/U Tokyo)Ogawa san, Togashi san (SACLA)

10 keV X rays

Harmand et al, Nat. Photon 2013

Jitters and Resolution

Delay 0

Sato et al., in preparation Single shot image

Histogram of arrival timeTiming jitter: 130 fs (rms)

~ 10 fs resolution

Sorting

Shot

num

ber

900 um 2.1 ps

31

• Transmission grating• Efficiency can be tuned from 0.1% to 40%• Install to optics hutch as a permanent component

Collaboration with Ch. David (PSI)

Si grating 300 nm p/1 um t

75020070150150

64015015070

GG GGGG G

G G G G G G G G G G GG

L

P

L

L L

L

P

L L

PL

PPL PLG

L

LPL

“From destructive to nondestructive”Branching with beam splitter

32

T mon.

Exp.

Pulse Energy 0.4 mJ @10 keVPhoton energy range 4.5 to 15 keV Stability Intensity I/I < 10%

Pointing 3 ~7% of beam sizeRepetition rate 20 Hz (Max. 60 Hz)Pulse duration ??Peak Power ??

33

Summary of Performance

Temporal profile/Pulse duration

THz streaking methodGrguraš et al, Nat Photon. 2012

ref. T CAV

Frequency domainTTF 2 PRL (2002)Shintake et al, Nat Photon (2008)

35

Temporal vs. Frequency

FT-10 0 100

20

40

60

80

Time (fs)

Pow

er (G

W) =6 fs

9750 9760 9770 9780 97900

1

2

3[ 10 11 ]

Photon energy (eV)

Phot

on fl

ux(p

hoto

ns/p

ls/0

.1%

b.w

.)

Temporal domain Frequency domain

-30 -20 -10 0 10 20 300

20

40

60

Time (fs)

Pow

er (G

W)

=30 fs

9700 9710 9720 9730 97400

0.5

1[ 10 12 ]

Photon energy (eV)

Phot

on fl

ux(p

hoto

ns/p

ls/0

.1%

b.w

.)

9715 9716 9717 9718 9719 97200

0.5

1

Photon energy (eV)

~100 meV

Resolution

Recent developments of single shotspectrometers

LCLS: Collimated beam +bent, thin crystal analyzer

D. Zhu et al, APL 101, 034103 (2012)

SACLA: Diverged beam (elliptical mirror) + flat,thick crystal analyzer

M. Yabashi et al, PRL 97, 084802 (2006)Y. Inubushi et al, PRL 109 144801 (2012)

10 keVUndulator

MPCCD camera

Si crystal

Elliptical mirror

SwissFEL: Grating

P. Karvinen et al, Opt. Lett. 37 5073 (2012)36

37

PrincipleDivergent beam+ Flat Crystal Analyzer+ Position Sensitive Detector

Flat crystalanalyzer

B

B

min

22 22

cot

cot 3.2 cos 4cot

B

B x B es B hkl hkl

c

EE

s r F dL v

focussize extinction

Detectorpixel size

Higher order reflection >higher resolution

Resolution

-40 -20 0 20 400

0.5

1

-40 -20 0 20 40

Position ( m)

FWHM =16.9 meV

Inte

nsity

(a.u

.)

Energy (meV)

Measured at SPring 8M. Yabashi et al, PRL97, 084802 (2006)

140x103

120

100

80

60

40

20

Inte

nsity

(a.u

.)

10.1010.0510.009.959.90

Photon energy (keV)

120

100

80

60

40

20

0

inte

nsity

(a.u

.)

-2 -1 0 1 2Relative photon energy (eV)

100 m

10 keVUndulator

MPCCD camera

Si crystal

Elliptical mirror

Si (555)Resolution :

~14 meV

Si (111)Resolution :

~1 eV

Inubushi san (SACLA)

Development at SACLA

38

39

Single shot spectra: wide range (~100 eV)with Si111

Spike profiles change in every shot due to SASE schemeBut central energy is stable

40

Single shot spectra: high resolution (14 meV)with Si 555

Bunch compressionSpike width in freq. domain

WeakNarrow

StrongBroad

Fourier relationship between temporal and frequency

Experiment

Simulation“SIMPLEX”

41

E=600 meV (FWHM)E=290 meVE=110 meV

=31 fs =8.9 fs =4.5 fs (FWHM)

Y. Inubushi, et al, PRL. 109 144801 (2012).

Evaluation of pulse duration using XFEL simulation

E = 400 60 meV ~6.5 1.5 fs

300 uJ/pls (@ 10 keV) P > 40 GW

Recent operation condition (Jan 2013)

42

Pulse Energy 0.3 mJ @10 keVPhoton energy range 4.5 to 15 keV Stability Intensity I/I < 10%

Pointing 3 ~7% of beam sizeRepetition rate 10 Hz (Max. 60 Hz)Pulse duration < 10 fsPeak Power > 40 GW

43

Summary of Performance

Note: recent nonlinear X ray experiment suggests shorter pulseduration (FWHM: 2.5 2.8 fs, Tamasaku et al, PRL 2013)Continue to develop new methods

Normalization of SASE spikes by comparingtwo spectra from +1 and 1 order Katayama san

(SACLA)

+1

1

Application (II): Dispersive Absorption Spectroscopy

Collaborated with Profs. Toshinori Suzuki (Kyoto U), Kazuhiko Misawa (TUAT)Kiyotaka Asakura (Hokkaido U) 44

2.1462.146 2.1362.1362.1312.131

2.1262.1262.1212.121

2.1172.1172.1402.140

• Zn thickness 20 m• 200 shots (one K value)• Observable range 30 ~ 40 eV

(one K value)• resolution ~0.4 eV200 shots (20 s) for measuringXAS with good S/N level

Dispersive X ray absorption spectrum

Katayama et al, APL, in press

10 shots100 shots

Static artifact

rms deviation = 0.9%

K value

45

Summary• Photon diagnostics are practically working for reliable and stable operation

of XFEL machine• Advanced diagnostics promote to develop user experimental methods• Close collaboration among machine, beamline, and control interface

people is essential

• Absolute pulse energy: easily measured with RT calorimeter• Spatial profile: screen based and edge scan based methods are available,

single shot high resolution method is demanded• Wavefront: grating interferometer shows excellent performance• Spectrum: high resolution below 20 meV with crystal spectrometer, useful

for seeded operation• Arrival timing: below 10 fs is possible• Pulse duration: still need to develop, especially below 10 fs region

Spectral spikes, THz streaking, auto/cross correlation w XNL

• Still a lot of exciting challenges. Your valuable inputs, proposals, andcollaboration are highly appreciated 46

AcknowledgementAll SACLA/SPring 8 MembersEspecially,Kensuke Tono, Tadashi Togashi, Yuichi Inubushi, Tetsuo Katayama, KanadeOgawa, Takaki Hatsui, Yasumasa Joti, Togo Kudo, Takashi Kameshima, YoichiKirihara, Shun Ono, Hiroaki Kimura, Hiromitsu Tomizawa, Haruhiko Ohashi,Hirokatsu Yumoto, Takahisa Koyama, Shunji Goto, Kenji Tamasaku, MitsuruNagasono, Kazuaki Togawa, Takashi Tanaka, Toru Hara, Hitoshi Tanaka,Ryotaro Tanaka, Mitsuhiro Yamaga, Toru Ohata, Yukito Furukawa, TakashiSugimoto, Hideo Kitamura, & Tetsuya Ishikawa

Yoshiro Fujiwara & Engineering TeamOsaka University

Yasuhisa Sano, Satoshi Matsuyama, Taito Osaka,Univ. of Tokyo

Hidekazu Mimura, Takahiro Sato, Ichiro Inoue

47

Thank you for your attention

48

END