Transcript HHG with ERL FEL

Compact ERL-FEL/Pulse Stacker Cavity
configurations:
new high rep rate, high intensity driver sources
for High Field Applications ?
Mufit Tecimer
THz-FEL Group, University of Hawai’i at Manoa
KEK, Tsukuba, Japan
April 20, 2012
The rationale of the presented study is an old idea
regarding electron beam based radiation sources:
 To tap on the (high) power deposited in the electron
beam by elaborating on schemes with high extraction
efficiency,
 Use of the generated radiation in Applications relevant
to the current research/ technological development.
High Field Applications
I.) Upfrequency conversion in the x-ray region
• phase-matched High order Harmonic Generation
(HHG) attosecond science
• x-ray Parametric Amplification (XPA)
II.) Laser driven plasma-based electron accelerators
• Laser Wake Field Accelerator (LWFA)
III.) Inverse Compton Scattering (ICS)
.....
Generation of coherent X-Ray pulses by HHG
Three-Step Model (Corkum 1993)
h  cutoff  I l  (single atom HHG)
2
Popmintchev et al.,
OSA/ CLEO 2011
Generation of coherent X-Ray pulses by HHG
The idea of using Mid-IR (ERL) FELs as drivers for HHG thought of or
considered by Kapteyn /Murnane (JILA), Foehlisch (Bessy) and others …
requirements imposed on drive lasers (Popmintchev et al.) :
Phase-matched HHG in keV region photons needs:
 preferably few cycle (CEP stabilized) to ~10 cycle drive laser pulses in NIR/MIR ,
 intensities in the range of 1-5x1014 W/cm2 ,
 noble gas filled hollow waveguide apertures: ~100mm-200mm, (He) gas pressure:
tens of atm)
OPCPA’s
•NIR sub-10 fs with 70 mJ energy at 100kHz.
• NIR sub-10 fs multi-kHz, multi-mJ
•Mid-IR (~3mm) sub-100 fs with a few micro-Joule energy at 100kHz
•3.9 mm sub-100 fs with ~9 mJ at 20Hz
HHG - Predictions & Measurements
Popmintchev et al.,
PNAS 106, 10516 (2009)
h  cutoff  I l  (single atom HHG)
2
Curves normalized to phase-matched HHG @ λ0=0.8µm
@ = 6µm, 10 MHz rep. rate (He)
estimated Photon flux : ~1013-14 ph/sec (1.0%BW)
@ = 3.9µm, 1 kHz rep. rate (35 - 40 atm. He)
Photon flux : ~108 ph/sec (1.0%BW)
(based on experiments)
Phase matched HHG @3.9mm, 6cycle, 20 Hz
Popmintchev et al.,
OSA/ CLEO 2011
M. Tecimer, FHI-Berlin (FEL Seminar), Sep. 29, 2011
HHG - Predictions & Measurements
He driven by 20 μm mid-IR lasers
may generate bright 25 keV beams.
[Ref.: Kapteyn/Murnane, Quantum
Physics and Nonlinear Optics at
High Energy Densities]
to be published by Kapteyn/Murnane
Group (JILA) in Science
XPA Experiments
J. Seres et al., Nature Phys. 6, 455 (2010).
B. Aurand et al., NIM A 653, 130
(2011)
Amplified
spontaneous
emission
Amplifier
with a seed
A claimed maximum gain of about
8000 at 50eV photon energy is
demonstrated.
(figure modified from H.Kapteyn, Quantum Physics
and Nonlinear Optics at High Energy Densities)
"Modified" Cascaded/Staged LWFA using FEL driver pulses
are repeatedly accelerated by the laser
 electrons
.
wakefields in a manner similar to the conventional
accelerators ...
J. S. Liu et al., PRL 107, 035001 (2011)
n~1017-1018 cm-3
~ 3 - 6 mm (?)
Joule level driver laser pulses
@ ~1 mm
multiple stages
FEL pulse
GeV
beam
e-
Tens of TWatts
few optical cycles
FEL pulse
FEL pulse
synchronized FEL pulses
(Figure modified from 'High Power Laser Technology',Wim Leemans, LBNL)
Reference:
C.B. Schroeder, E. Esarey, C.G.R. Geddes, C.
Benedetti, and W.P. Leemans,
Phys. Rev. ST Accel. Beams 13, 101301 (2010).
System parameters used in the Simulations
Beam parameters
Beam Energy
Bunch charge
_z rms bunch length
norm.Trans. Emittance
_e rms energy spread
FEL (~3-6mm)
100 (200)
80 (200)
0.1
5
0.5%
Units
MeV
pC
ps
mm.mrad
Wiggler period
Wiggler Krms
Periods
60
MeV
Bunch charge
110 (135)
pC
_z rms
150
fs
Peak current
~300
A
_e rms
0.1%
_e rms
0.5%
nor. trans. Emit.
mm
Units
115
(correlated)
planar
FEL (1.6mm)
Beam Energy
(uncorrelated)
Wiggler parameters
Type
Beam parameters
JLab IR FEL
rep. rate
8
mrad
~75
MHz
1.7-2.6
25 (23)
Trim Quads reading
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Apr. 12, 2011
Coherent OTR interferometer autocorrelation
scans for bunch length measurements
[S. Zhang et al., FEL 09 Conf. Proceedings]
Outline of the project:
 short term: carrying out the HHG experiments on an
existing FEL facility that meets the requirements set on the
mid-IR drive laser, verifying the theory throughout the mid-IR
(particularly at around 6 mm-7mm) (JLab, FHI-FEL, …?)
 long term: mid-IR ERL-FELs should be able to perform
better than atomic lasers in terms of :
 tunability (throughout the nir/mid IR and beyond)
- high rep rate (MHz) in generating mJ(s) of ultrafast pulses
with high average power
Ongoing simulation work is mainly focused on the latter :
(system requirements imposed on a compact ERL)
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Apr. 12, 2011
Suggested (3-6mm) MIR FEL & Pulse Stacker Cavities
I.)
II.)
compressor
stretcher
mode matching
telescope
 high-Q enhancement cavity
(EC) smoothes out power and
timing jitter of the injected pulses
inherent to FEL interaction.
 allows ~fs level
synchronization of the cavity
dumped mid-IR pulse with the
mode-locked switch laser.
Mode-locked
NIR Laser
PLE
dielectric mirror
NIR/MIR FELO
 Depending on the recombination time of the fast switch, sequence of
micropulses with several ns separation can be ejected from the EC !
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Nov.04, 2010 & Apr. 12, 2011
Enhancement Cavity @ JLab
Folded cavity
Input
Coupler
vacuum
vessel
FEL
Q ~ 40 (Finesse ~ 300 )
enhancement :~90
Q~ 50
enhancement :~130-140
estimated enhancement
@ JLab ~ 100
High
Reflector
Brewster W.
Opt. Switch
mount
T. Smith @ Stanford IR-FEL achieved enhancement of ~70 - 80
using an external pls stacker cavity (1996)
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Apr. 12, 2011
Ultrashort (few cycles) Pulse Generation in (IR-THz) FELs
• Chirped pulse generation in a FEL oscillator using a
chirped electron beam and pulse compression
• Mode-locking techniques in FELs
-Active mode-locking
- Passive mode-locking
• Generation of short electron pulses
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Nov. 04, 2010
Ultrashort Pulse Generation by passive modelocking
Passive modelocking in conventional (atomic) laser :
- Kerr Lens modelocking
- Semiconductor Saturable Absorber Mirrors (SESAM)
- Does FEL have a self (passive) modelocking mechanism ?
(for instance intensity dependent absorber)
Synchrotron Osc. Freq. :  s 
Er
FEL oscillator with perfectly
synchronized cavity (single spike,
high gain superradiant FEL oscillator)
Nonlinear reflectivity data for a
representative SESAM sample
(figure added to the original)
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Nov. 04, 2010
Ultrashort Pulse Generation in (Mid-IR) FELs
• Chirped pulse generation in a FEL oscillator using a
chirped electron beam and pulse compression (JLab)
• Mode-locking techniques in FELs
-Active mode-locking (multiple OK sections used in a
cavity)
 - Passive mode-locking (JAERI, lasing at ~22 mm)
(single spike, high gain superradiant FEL osc.)
 Generation of short electron pulses (JLab)
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Apr. 12, 2011
High Gain (superradiant) FEL Oscillator operating at cavity synchronization
Further studies:
- cascaded oscillator schemes
(problem: large momentum spread for the beam
transport/energy recovery)
- use of (assistant) SESAM mirrors
- checking the results with other well established
codes
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Nov. 04, 2010
coupled FEL oscillators
FEL oscillators with perfectly synchronized cavity
I.)

u1

u1

u2
II.)
• U(1) = U(2)

u2
(better U(1) > U(2) )
• Mirror coupling ratios of are optimized
relatively
large
Outcoupling
Cascaded system of coupled oscillators
I.)
II.)
• U(1) = U(2) > U(3) = U(4) …
• U(1) > U(2) > U(3) > U(4) …
• Mirror coupling ratios of are optimized
Amplifier stage follows the coupled FEL oscillators
Time domain multi-mode appraoch using SVEA
 
E   x , z , t   Re
q
 z  v g q   u
1
n
t
u qn  z , t  e
{  (



) e n  x, y 
}
n
 qn  z , t  
 q   0  qc  / L r

i k zn (  q ) z   q t
 i k zn (  q ) z   q t
 u  qn  z , t  e
1

e
 i ( k zn ( q ) z   q t )
2 S  qn

 
*
J  x , z , t   e n  dxdy
S
   c / L r
 z  v gn   u n  z , t   
1
1
t
e
 i ( k zn ( ) z   t )
0
0
2Sn

 
*
J  x , z , t   e n  dxdy
un 
 u qn
e

iq   z / v gqn  t
q
S
Space-frequency representation of the electromagnetic fields and current sources
 
 1
 E ( x  , z , t ) 
 
  Re 
 J ( x  , z , t ) 
 2 
~ 
E  x , z,  
(
u n z,  e
~ 

 E ( x  , z ,  )   i  t
  J~ ( x , z ,  )  e d  

0 



ik zn (  ) z
 u n z,  e
 ik zn (  ) z
) e n x, y 
n
 z u n z,   
1
2S n
e
 ik zn ( ) z
~ 
 J x , z,   e  n ( x, y )
*
dxdy
S
• Exact first order ordinary differential equations of the axial dimension without the need of
introducing any approximations.
• Inverse Fourier Transform is necessary to construct the fields used to determine particle’s motion.

Contrasting approaches used for FEL simulation
First Stage (master oscillator)
1D SVAE (complex field amplitude of a carrier wave)
250
a.)
b.)
1 .0
1 .0
c.)
200
100
50
n o rm . S p e c tra l In te n s ity
n o rm . P o w e r
P u lse E n e rg y [ m J]
0 .8
150
0 .6
0 .4
0 .2
0
0 .0
0
100
200
300
400
500
600
700
800
0 .8
0 .6
0 .4
0 .2
0 .0
0
10
20
P a ss
30
40
5
50
6
7
8
9
10
11
w a v e le n g th [ m m ]
R a d ia tio n cycle s
250
1 .0
d.)
1 .0
e.)
f.)
n o rm . S p e ctra l In te n sity
200
n o rm . P o w e r
P u ls e E n e rg y [m J ]
0 .8
150
100
50
0 .6
0 .4
0 .2
0
100
200
300
Pass
400
500
600
0 .6
0 .4
0 .2
0 .0
0 .0
0
0 .8
10
20
30
40
50
60
R a d ia tio n cycle s
5
6
7
8
9
10
11
w a v e le n g th [ m m ]
3D non-averaged, multifrequency (multimode) code
M. Tecimer, PRST-AB 15, 020703 (2012)
Simulated temporal/spectral characteristics of mid-IR pulses
250
1 .0
1 .0
I.)
n o rm . S p e ctra l In te n sity
200
n o rm . P o w e r
P u ls e E n e rg y [m J ]
0 .8
150
100
0 .6
0 .4
0 .2
50
0 .8
0 .6
0 .4
0 .2
0 .0
0 .0
0
0
100
200
300
400
500
10
20
Pass
50
5
60
n o rm . P o w e r
2 .0
1 .5
0 .6
0 .4
1 .0
0 .2
400
20
Pass
III.)
11
0 .2
0 .0
10
500
10
0 .4
0 .0
300
9
0 .6
0 .0
200
8
0 .8
0 .5
100
7
1 .0
0 .8
2 .5
0
6
w a v e le n g th [ m m ]
1 .0
3 .0
P u ls e E n e rg y [m J ]
40
n o rm . S p e ctra l In te n sity
II.)
30
R a d ia tio n cycle s
30
40
R a d ia tio n cycle s
50
6
60
7
8
9
10
w a v e le n g th [ m m ]
3 .0
1 .0
1 .0
0 .8
0 .8
1 .5
1 .0
n o rm . S p e ctra l In te n sity
2 .0
n o rm . P o w e r
P u ls e E n e rg y [m J ]
2 .5
0 .6
0 .4
0 .2
0 .6
0 .4
0 .2
0 .5
0 .0
0 .0
0 .0
0
100
200
Pass
300
400
500
10
20
30
40
R a d ia tio n cycle s
50
60
6
7
8
w a v e le n g th [ m m ]
9
10
Partial bilateral Coupling of FEL Oscillators
Optical Pulse locking
1 .0
1 .0
0 .8
0 .8
n o rm . P u ls e E n e rg y
n o rm . P u ls e E n e rg y
Beam & Optical Pulse locking
0 .6
0 .4
feedback ~5x10-4
0 .2
0 .6
0 .4
feedback ~5x10-4
0 .2
0 .0
0 .0
0
100
200
300
400
0
100
Pass
200
300
400
500
Pass
~ 5x10-4 ~5 – 10% of optimum output pulse energy
~10-7 feed back ~65-70% of optimum output,
• feed back reduced to less than 10-8 to reach nearly
the optimum output,
•
limit cycle oscillations reduce strongly
Master Oscillator: beam longitudinal phase space
210
a.)

200
190
180
170
2 .5 0 x1 0
-5
5 .0 0 x1 0
-5
7 .5 0 x1 0
-5
1 .0 0 x1 0
-4
1 .2 5 x1 0
1 .5 0 x1 0
-4
1 .7 5 x1 0
-4
2 .0 0 x1 0
z
Undulator exit
210
-4
b.)
200

190
180
170
160
150
2 .5 0 x1 0
-5
5 .0 0 x1 0
-5
7 .5 0 x1 0
-5
1 .0 0 x1 0
-4
1 .2 5 x1 0
z
-4
1 .5 0 x1 0
-4
1 .7 5 x1 0
-4
2 .0 0 x1 0
-4
-4
Slave FEL Oscillator: beam longitudinal phase space
Undulator entrance
210
a.)
200

190
180
170
160
150
2 .5 0 x1 0
220
210
-5
5 .0 0 x1 0
-5
7 .5 0 x1 0
-5
1 .0 0 x1 0
-4
1 .2 5 x1 0
-4
1 .5 0 x1 0
-4
1 .7 5 x1 0
-4
2 .0 0 x1 0
-4
z
Undulator exit
C
b.)
200

190
180
170
160
150
2 .5 0 x1 0
-5
5 .0 0 x1 0
-5
7 .5 0 x1 0
-5
1 .0 0 x1 0
-4
1 .2 5 x1 0
z
-4
1 .5 0 x1 0
-4
1 .7 5 x1 0
-4
2 .0 0 x1 0
-4
Slave FEL Oscillator: beam longitudinal phase space
Undulator entrance
210
200

190
180
170
160
150
2 .5 0 x1 0
-5
5 .0 0 x1 0
-5
7 .5 0 x1 0
-5
1 .0 0 x1 0
Undulator exit
-4
1 .2 5 x1 0
-4
1 .5 0 x1 0
-4
1 .7 5 x1 0
-4
2 .0 0 x1 0
-4
z
220
210
200

190
?
180
170
160
150
5 .0 x1 0
-5
1 .0 x1 0
-4
1 .5 x1 0
-4
2 .0 x1 0
z
-4
2 .5 x1 0
-4
3 .0 x1 0
-4
FEL Osc. sensitivity to temporal jitter
t/t = dL/L + df/f
e- bunch
t : timing jitter
L : cavity length
dL: cavity length detuning
f : bunch rep. frequency (perfectly synchronized to L)
t : cavity roundtrip time ( 2L/c)
 Bunch time arrival variation effectively has the same effect
as cavity length detuning.
 effect of the timing jitter on the FEL performance
In slippage dominated short pulse FEL oscillators cavity
detuning is necessary to optimize the temporal overlap
between optical and e- pulses (Lethargy effect).Timing jitter
induces fluctuations on the operational cavity detuning.
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Apr. 12, 2011
FEL Osc. sensitivity to temporal jitter
~ 6 mm
P o w e r [W a tts ]
6 x1 0
5 x1 0
4 x1 0
3 x1 0
2 x1 0
1 x1 0
10
7 x1 0
10
jitter 2.5 fs rms
10
9 x1 0
6 x1 0
10
5 x1 0
10
4 x1 0
10
3 x1 0
10
2 x1 0
10
1 x1 0
10
8 x1 0
7 x1 0
10
P o w e r [W a tts]
7 x1 0
10
10
10
6 x1 0
5 x1 0
4 x1 0
3 x1 0
2 x1 0
10
1 x1 0
0
0
0
100
200
300
400
500
600
0
100
6 x1 0
P o w e r [W a tts]
5 x1 0
4 x1 0
3 x1 0
2 x1 0
7 x1 0
10
6 x1 0
10
5 x1 0
10
4 x1 0
w/o initial jitter
10
10
10
10
10
10
10
300
400
500
600
0
100
200
300
400
500
600
tim e [fs]
10
10
jitter 2.5 fs rms
10
10
10
3 x1 0
10
10
2 x1 0
1 x1 0
200
tim e [fs]
10
10
0
tim e [fs ]
7 x1 0
10
10
10
1 x1 0
10
0
0
-1 x1 0
10
0
100
200
300
tim e [fs]
400
500
600
0
100
200
300
400
500
600
tim e [fs ]
M. Tecimer, Bessy-Berlin (Machine Group Seminar), Apr. 12, 2011
High Reflectivity Dielectric Mirrors for the mid-IR & THz regions
The presented coupled oscillator scheme
should be applicable to longer mid-IR (THz)
wavelengths by using the low loss, high
reflectivity dielectric mirrors developed for THzFEL applications.
M. Tecimer, K. Holldack and L. Elias, PRST-AB 13, 030703 (2010)
1 .0
3 rd
5 th
0 .9 9 9 7
R e fle cta n ce
0 .8
4 xS i
0 .8
0 .6
0 .6
0 .4
0 .4
0 .2
0 .2
0 .0
a .)
100
0 .0
200
300
400
500
w a ve le n g th [m icro n s]
0 .9 9 7 6
1 .0
600
5xQ
b .)
100
200
300
400
w a v e le n g th [m ic ro n s ]
500
Summary
100-200 MeV range superconducting ERL driven mid-IR FELs hold
great promise in filling a unique niche for generating multi-mJ level
(possibly much higher), ultrashort ( <10 cycles) pulses tunable within
the entire mid-IR region (and beyond) with at least many tens of MHz
repetition rates.
Because of their ability in providing high peak intensities with excellent
temporal and transversal coherence characteristics at unprecedented
high repetition rates across the entire NIR/MIR spectral range, they
have the potential to become attractive tools in various strong field
applications alone or in combination with high finesse enhancement
cavities.
References:
HHG:
• T. Popmintchev et al., Nature Photon. 4, 822 (2010).
• M.-C. Chen et al., Phys. Rev. Lett. 105, 173901 (2010).
• G. Andriukaitis,T. Balciunas, S. Alisauskas, A. Pugzlys,
A. Baltuska, T. Popmintchev, M. C. Chen, M. M. Murnane,
and H. C. Kapteyn, Opt. Lett. 36, 2755 (2011).
• Henry Kapteyn and Margaret Murnane, Quantum Physics and Nonlinear Optics
at High Energy Densities - Applications in Plasma Imaging
• R. J. Jones, K. D. Moll, M. J. Thorpe, and J. Ye, Phys.Rev. Lett. 94, 193201
(2005).
XPA:
• J. Seres et al., Nature Phys. 6, 455 (2010).
• L. Gallman, Nature Phys. 6, 406 (2010).
LWFA:
• J. S. Liu et al., PRL 107, 035001 (2011).
• Wim Leemans, LBNL ,White Paper of the ICFA-ICUIL Joint Task Force – High
Power Laser Technology for Accelerators.
and references in
M. Tecimer, PRST-AB 15, 020703 (2012)
THANK YOU FOR YOUR ATTENTION