Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
48th ICFA Advanced Beam Dynamics Workshop on Future Light SourcesMarch 1-5, 2010, SLAC, Menlo Park, California
XFELO X-ray Cavity
Feasibility Studies
Yuri Shvyd’ko
C1
C2 C3
C4
M1
M2
e-
undulator x-rays
Content
• Technical challenges• Feasibility studies:
⇒ reflectivity of diamond crystals⇒ heat load problem⇒ nanoradian angular stabilization⇒ radiation damage
• Conclusions and Outlook
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
XFEL Oscillator
x-rays
e-
Beam
dump
Linac
undulator
Permeablemirror
R1, T1
Mirror
R2 L 100 m
XFELO requires:
• ultra-low-emittance (εn . 10−7 m rad) electron beams,• low-loss x-ray crystal cavity (losses ' 15%) R1, R2 > 95%, T1 ' 4%
K.-J. Kim, Yu. Shvyd’ko, S. Reicher, PRL 100 (2008) 244802.
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Reflectivity of Si in backscattering
0 5 10 15
0.75
0.8
0.85
0.9
0.95
1.0
Si@120 K
0 5 10 15
0.75
0.8
0.85
0.9
0.95
1.0
Si@300 K
E [keV]
Refl
ecti
vit
y
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Theory predicts highest reflectivity from diamond
0 5 10 15
0.75
0.8
0.85
0.9
0.95
1.0
C@300 K
E [keV]
Refl
ecti
vit
y
Very high reflectivity (in theory)due to:
• High Debye Temperature, andthus high Debye-Waller factor
• Low Z, low photo absorption
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
undulator
L
S
H
H
2
2
A
B
M2
M1
e-
x-rays
RA × RB × RM1 × RM2 ' 0.9 TA ' 0.04
Diamond cavity for the X-FEL Oscillator
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Diamond crystal and mirror reflectivity @ 12 keV
0.0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
Tra
nsm
issi
vit
y
Ab
sorp
tio
n
-60 -40 -20 0 20 40 60
E - E0
[meV]
0.0
0.2
0.4
0.6
0.8
1.0
Refl
ecti
vit
y
E0=12.0404 keV
0.0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
Tra
nsm
issi
vit
y
Ab
sorp
tio
n
-60 -40 -20 0 20 40 60
E - E0
[meV]
0.0
0.2
0.4
0.6
0.8
1.0
Refl
ecti
vit
y
E0=12.0404 keV
Mirror calculations:www-cxro.lbl.gov
Narrow band mirrors:∆E ≈ 10 meV; ∆E ≈ ~/τp
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
undulator
L
S
H
H
2
2
A
B
M2
M1
e-
x-rays
E = EH cos Θ ⇒ Two-crystal scheme is not tunable.
Because, it is necessary to keep small φ . 2 mrad
and therefore small Θ . 2 mrad, for high reflectivity of the mirrors.
Two-Crystal Cavity is not Tunable
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
A four-crystal (A,B,C, and D) x-ray optical cavity allows photon energy E tuningin a broad range by changing the incidence angle Θ.
A
B C
D
M1
M2
e-
undulator
L
L’
S
G
G
F
H
H
H
2
2
x-rays
R.M.J. Cotterill, Appl. Phys. Lett., 12 (1968) 403
K.-J. Kim, and Yu. Shvyd’ko, Phys. Rev. STAB (2009)
E = EH cos Θ
Tunable Cavity
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
A
B C
D
M1
M2
e-
undulator
H
H
H
2
2
x-rays
X-ray Optics:
• Quality of diamond crystals:• is the theoretical reflectivity achievable?
• Heat load problem (reflection region variations . 1 meV).• Angular stability: δθ . 10 nrad (rms)• Spatial stability: δL . 3 µm (rms) → δL/L . 3 × 10−8
• Radiation damage
XFELO Technical Challenges
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Quality of Diamond crystals
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
3rd International Conference on Diamonds for Modern Light Sources, Awaji Yumebutai, 20th-23th May 2008ESRF15
Phase platePhase plate
White beam topograph in transmission
7.1 mm
110-oriented plate
Dislocation free areas of 4x4mm2
and more!!!
1-1-1-reflection
Härtwig, 2008 (ESRF)
Required diamond crystals:• high quality (dislocation free, etc.)• thickness: 20 − 2000 µm• small suffice: ' 1 mm2
Still open question:is the theoretical reflectivity achievable?
Quality of Diamond crystals
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
undulator
x-rays
bandwidth100 eV
C(111) cooledmonochromator
bandwidth1.7 eV
2 Si(220)
300 K
2 Si(220)
300 K
Si(15 11 9) channel-cut
123 K
T. Toellner
bandwidthE 1 meV
high-resolutionmonochromator
E=23.7 keV
(111)
(995)2
= 2 10-4
APD
diamond
L 10 mT. Toellner
Shvyd’ko, Stoupin, Cunsolo, Said, Huang, Nature Phys. 6, 196 (2010)
Experiment, 30-ID @ APS
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
10-3
2
5
10-2
2
5
10-1
2
5
1
Refl
ecti
vit
y,R
-15 -10 -5 0 5 10 15
E - E0, [meV]
C(995) theoryE
0.2
0.4
0.6
0.8
R
-10 0 10
E - E0
E =2.9 meV
R =91 %
C(995), EH
= 23.765 keV
Lcrystal = 415µm
Rtheory = 91%
Spectral Width and Reflectivity: Theory
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
10-3
2
5
10-2
2
5
10-1
2
5
1
Refl
ecti
vit
y,R
-15 -10 -5 0 5 10 15
E - E0, [meV]
experimentC(995) theory
E
0.2
0.4
0.6
0.8
R
-10 0 10
E - E0
E =2.9 meV
R =87 %R =91 %
C(995), EH
= 23.765 keV
δE = 1.5 meV
δE = hc/(2Lcrystal)
dcrystal = 415 ± 5µm
Rtheory = 91%
Rexperiment = 87%
Beam footprint' 1 mm2
Shvyd’ko, Stoupin, et al, Nature Phys. 6, 196 (2010)
Spectral Width and Reflectivity: Experiment
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
0.92
0.94
0.96
0.98
1.0
Peak
Refl
ectiv
ity
5 10 15 20 25
2
5
102
2
5
103
2
Pene
trat
ion
Dep
th[
m]
5 10 15 20 25
Bragg Energy EH
[keV]
1
10
102
Ene
rgy
Wid
th[m
eV]
5 10 15 20 25
R, L, and ∆E are interconnected. ∆E ∝ 1/L
R L ∆E
Smallness of ∆E is a hallmark of high quality crystal
Reflectivity vs. Penetration & Energy Width
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
0 1 2 3 4 5x [mm]
0
1
2
3
4
5
6
7
8
y[m
m]
(a)
0 1 2 3 4 5x [mm]
0
1
2
3
4
5
6
7
8
1
1.3
2
4
20
Spectral width, E/ Emin
(b)
0 1 2 3 4 5x [mm]
0
1
2
3
4
5
6
7
8
0
0.2
0.4
0.6
0.8
1
Reflectivity, R/Rmax
(c)
Shvyd’ko, Stoupin, et al, Nature Phys. 6, 196 (2010)
Spectral Width and Reflectivity Map
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
0 1 2 3 4 5x [mm]
0
1
2
3
4
5
6
7
8
y[m
m]
(a)
0 1 2 3 4 5x [mm]
0
1
2
3
4
5
6
7
8
1
1.3
2
4
20
Spectral width, E/ Emin
(b)
0 1 2 3 4 5x [mm]
0
1
2
3
4
5
6
7
8
0
0.2
0.4
0.6
0.8
1
Reflectivity, R/Rmax
(c)
Shvyd’ko, Stoupin, et al, Nature Phys. 6, 196 (2010)
Synthetic diamond crystals are available with theoretically high reflectivity andsufficiently large in size for XFELO cavity applications.
Spectral Width and Reflectivity Map
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Heat Load Problem
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Temperature gradient δT ⇒ energy spread δE/E = βδT .Requirement: δE . 1 meV, when the next pulse arrives.
Incident power ' 50 µJ/pulse.Absorbed power: ' 1 µJ/pulse (2%).Footprint: ' 100 × 100 µm2
Is it a problem?
Heat Load Problem
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
300
301
0
5
10
15
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Time [ s]
T(0)=300 K
T =0.5 K
= 1 10-6
K-1
Dia
mo
nd
Tem
per
aure
T[K
]
Ph
oto
nE
ner
gy
Sp
read
E[m
eV]
100 m
200 m
beamfootprint
H. Sinn simulations
Temperature gradient δT ⇒ energy spread δE/E = βδT .Requirement: δE . 1 meV, when the next pulse arrives.
• Big temperature jump δTafter the x-ray pulse arrival.
• T=300K: Big temperaturespread by the arrival of
Heat Load Problem
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
300
301
0
5
10
15
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Time [ s]
T(0)=300 K
T =0.5 K
= 1 10-6
K-1
100
105
110
0
2
4
6
8
10
12
14
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Time [ s]
T(0)=100 K
T =14 K
= 1 10-7
K-1
Dia
mo
nd
Tem
per
aure
T[K
]
Ph
oto
nE
ner
gy
Sp
read
E[m
eV]
100 m
200 m
beamfootprint
H. Sinn simulations
Temperature gradient δT ⇒ energy spread δE/E = βδT .Requirement: δE . 1 meV, when the next pulse arrives.
• Big temperature jump δTafter the x-ray pulse arrival.
• T=300K: Big temperaturespread by the arrival of
the next x-ray pulse.
• T=100K: Negligible temperaturespread by the arrival of
the next x-ray pulse.
Heat Load Problem
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
300
301
0
5
10
15
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Time [ s]
T(0)=300 K
T =0.5 K
= 1 10-6
K-1
100
105
110
0
2
4
6
8
10
12
14
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Time [ s]
T(0)=100 K
T =14 K
= 1 10-7
K-1
Dia
mo
nd
Tem
per
aure
T[K
]
Ph
oto
nE
ner
gy
Sp
read
E[m
eV]
100 m
200 m
beamfootprint
H. Sinn simulations
Temperature gradient δT ⇒ energy spread δE/E = βδT .Requirement: δE . 1 meV, when the next pulse arrives.
Solution: Maintain diamond at T < 100 K!
• Big temperature jump δTafter the x-ray pulse arrival.
• T=300K: Big temperaturespread by the arrival of
the next x-ray pulse.
• T=100K: Negligible temperaturespread by the arrival of
the next x-ray pulse.
• Reasons:1. High temperature diffusivity D2. Low temperature expansion β
Heat Load Problem
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Diamond Thermal Expansion
10-9
10-8
10-7
10-6
Thermalexpansion,[K-1]
0 50 100 150 200 250 300
T, [K]
Experiment 1 (March 09)Experiment 2 (May 09)
4.0 10-14T3
Polynomial fit
Stoupin, Shvyd’ko, PRL 104, 085901 (2010)
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
100101102103104105106107108109110111112
0
5
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Time [ s]
T(0)=100 K
T =14 K
= 3.9 10-8
K-1
50
55
60
65
70
75
80
85
90
95
0
2
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Time [ s]
T(0)=50 K
T =40 K
= 3.9 10-9
K-1
Dia
mo
nd
Tem
per
aure
T[K
]
Ph
oto
nE
ner
gy
Sp
read
E[m
eV]
100 m
200 m
beamfootprint
H. Sinn simulations
Temperature gradient δT ⇒ energy spread δE/E = βδT .Requirement: δE . 1 meV, when the next pulse arrives.
Solution: Maintain diamond at T < 100 K!
Corrected energy scaleXXy
Heat Load Problem
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Angular & Spatial Stability
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Required angular stability: δθ . 10 nrad (rms)Required spatial stability: δL . 3 µm (rms) ⇒ δL/L ' 3 × 10−8 (L = 100 m)
Angular & Spatial Stability
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Intensity,R(V)
Voltage, VVmaxV0
Required angular stability: δθ . 10 nrad (rms)Required spatial stability: δL . 3 µm (rms) ⇒ δL/L ' 3 × 10−8 (L = 100 m)
Solution: Null-detection harware feedback. (LIGO prototype)
X-ray intensity: linear response tosmall angular oscillations is propor-tional to angular deviation from themaximum of the rocking curve.
Angular & Spatial Stability
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Intensity,R(V)
Voltage, VVmaxV0
monitoring
Detector
Lock-in amplifier Integrator
piezo driver
piezo actuator
x-ray beam
IN OUT REF OUT IN REF IN OUT OUT
CONROL IN
R(V(t))
V v(t)
V + v(t)
V0+ V +v(t)
Required angular stability: δθ . 10 nrad (rms)Required spatial stability: δL . 3 µm (rms) ⇒ δL/L ' 3 × 10−8 (L = 100 m)
Solution: Null-detection harware feedback. (LIGO prototype)
X-ray intensity: linear response tosmall angular oscillations is propor-tional to angular deviation from themaximum of the rocking curve.
Feedback: correction signal is ex-tracted using lock-in amplification.
Angular & Spatial Stability
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
0
10000
20000
Intensity[a.u.]
-1000 -500 0 500 1000
3 [nrad]
-0.2 -0.1 0.0 0.1 0.2
Voltage [V]
FWHM
3=500 nrad=75mV
Region of stability=50 nrad
1
HHLMHRM
IC2
IC1IC3
APD
3
0Si6
Si5 Si4
Si3 Si2
Si1D2
D1
T. Toellner, D. Shu
HERIX Monochromator Stability Region
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
a b
0
1 104
2 104
3 104
4 104
5 104
6 104
7 104
8 104
Intensity[a.u.]
-1.0
-0.5
0.0
0.5
1.0
0 10 20 30 40 50 60
Time [minutes]
APDIC1
IC2IC3
HHLMHRM
0
1 104
2 104
3 104
4 104
5 104
6 104
7 104
8 104
-1.0
-0.5
0.0
0.5
1.0
Correctio
nsignal[V]
0 1 2 3 4 5 6 7 8 9 10
Time [hours]
APDIC1
IC2IC3
HHLMHRM
1
HHLMHRM
IC2
IC1IC3
APD
3
0Si6
Si5 Si4
Si3 Si2
Si1D2
D1
T. Toellner, D. Shu
HERIX Monochromator Stabilization
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
a b
0
1 104
2 104
3 104
4 104
5 104
6 104
7 104
8 104
Intensity[a.u.]
-1.0
-0.5
0.0
0.5
1.0
0 10 20 30 40 50 60
Time [minutes]
APDIC1
IC2IC3
HHLMHRM
0
1 104
2 104
3 104
4 104
5 104
6 104
7 104
8 104
-1.0
-0.5
0.0
0.5
1.0
Correctio
nsignal[V]
0 1 2 3 4 5 6 7 8 9 10
Time [hours]
APDIC1
IC2IC3
HHLMHRM
1
HHLMHRM
IC2
IC1IC3
APD
3
0Si6
Si5 Si4
Si3 Si2
Si1D2
D1
T. Toellner, D. Shu
' 15 nrad (rms) stabilization was demonstratedStoupin, Lenkszus, Laird, Goetze, K-J Kim, Shvyd’ko, RSI (submitted)
HERIX Monochromator Stabilization
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Radiation damage in diamond
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
XFELO generates:
50µJ/pulse @ 12 keV with ' 1 MHz rep. rateFootprint: A = 1.6 × 10−4 cm2 (rms)Flux ' 2 × 1020 ph/s/cm2
Time to ionize carbon atom with 100% probability: T ' 250 s Robin Santra
Can this produce irreversible changes in the perfect crystal lattice structure?
Radiation damage in diamond
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
XFELO generates:
50µJ/pulse @ 12 keV with ' 1 MHz rep. rateFootprint: A = 1.6 × 10−4 cm2 (rms)Flux ' 2 × 1020 ph/s/cm2
Time to ionize carbon atom with 100% probability: T ' 250 s Robin Santra
before after 1 yearof operations
APS undulators generate:
Flux ' 5 × 1017 ph/s/cm2
Time to ionize carbon atom with 100% probability: T ′ ' 105 s ' 1 day
Graphitization of the surface layer
of the diamond crystal is ob-served after several daysof operations. Though,no significant degradation in theperformance of the high-heat-load monochromator is observedafter a year of operations.
��9
Radiation damage in diamond
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
⇒ Operate diamond at cryogenic temperatures.⇒ Use isotopically pure 12C crystals.
Laser damage threshold is more than order of magnitude higher for isotopicallypure diamond.Anthony et al, PRB, 42 1104 (1990)
⇒ Apply electric fields to keep the electrons in diamond (K-J. Kim)⇒ ....
How to mitigate radiation damage in diamond?
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
A
B C
D
M1
M2
e-
undulator
H
H
H
2
2
x-rays
X-ray Optics:
• Quality of diamond crystals:• is the theoretical reflectivity achievable?
• Heat load problem: reflection region variations . 1 meV.• Angular stability: δθ . 10 nrad (rms)• Spatial stability: δL . 3 µm (rms) → δL/L . 3 × 10−8
• Radiation damage
XFELO Technical Challenges
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
Acknowledgments
Yu. Shvyd’ko FLS2010, SLAC, March 1, 2010
XFELO collaboration:
Kwang-Je Kim (APS)
Stanislav Stoupin (APS)Ryan Lindberg (APS)Sven Reiche (SLS)Bill Fawley, (LBNL)Frank Lenkszus (APS)Deming Shu (APS)Harald Sinn (European XFEL)
Thanks to:
Robert Winarski (APS)Tim Graber (APS)Robert Laird (APS)Stan Whitcomb (LIGO)Al Macrander (APS)Tim Roberts (APS)Kurt Goetze (APS)Emil Trakhtenberg (APS)Ayman Said (APS)Alessandro Cunsolo (APS)Tom Toellner (APS)Lahsen Assoufid (APS)