X-ray free electron lasers Alexander Zholents LBNL An
X-ray free electron lasers Alexander Zholents LBNL An introduction to the afternoon session John Arthur, SLAC Applications of the intense coherent x-ray pulses from LCLS John Corlett, LBNL Proposals and concepts for future FELs Andrew Sessler, LBNL Transverse-longitudinal correlations: FEL performance and emittance exchange A. Zholents, PQE, January, 2006 X-ray FEL essentials Electron beam production system: Q=1nC, n=1 mm-mrad Electron beam delivery system: E=20 GeV, Ipeak=5kA Electron beam utilization for emission of x-rays: l=1, =100fs, E=1mJ courtesy T. Limberg Layout of XFEL at DESY (Germany) m More info about future x-ray FEL facilities in the talk of John Corlett and use of these facilities in the talk of John Arthur in the following session A. Zholents, PQE, January, 2006 Emission of x-rays
u N S N S N S By peak magnetic field N S N S N S N S N
S vz S N S N S N S N S N S N S 2
1. Undulator 2. Electron beam N S N 2 vz 1 v =1 2 c c 2 eBy u v K = ; K= c 2 2 mc 3. Laser (not always needed) undulator parameter A. Zholents, PQE, January, 2006 FELs power is in bunching
N S S N N S S N N S S N N S S N
N S S E ( )single electron = E0 ( )e N N S S N N S S N * N S S
N i ( 0 )t N E ( )N electrons =E0 ( )e i ( 0 )t j j =1 Radiation power N WN electrons =Wsingle electron e j =1 2 i ( 0 )t j 1 (k z )2 =Wsingle electron N + N (N 1)e 2 Electrons should stay bunched within
*) Motz 1953; Phillips 1960, Madey 1971 z 1 / k = x / 2 A. Zholents, PQE, January, 2006 RF buncher E/E E/E E/E chirp z V = V0sin() RF RFAccelerating Accelerating Voltage Voltage z z z = R56E/E Path PathLength-Energy Length-Energy courtesy P.Emma Dependent
DependentBeamline Beamline A. Zholents, PQE, January, 2006 Laser buncher Energy modulation of electrons in the undulator by the laser light B = B sin(k z) 0 e - u S N Light S N N S N S u
Electron trajectory through undulator Magnetic field in the undulator Undulator period V E B u =22 /(1+ K 2 / 2) L E V 0 k k B Laser wavelength E V FEL resonance condition While propagating one undulator period, the electron is delayed with respect to the light on one optical wavelength A. Zholents, PQE, January, 2006 Laser buncher (2)
Field energy in the far field region of undulator radiation in the presence of the laser field: 2 A~ E ( ,r)+ E ( ,r) dSd L R Laser field A=A + A + 2 A A / cos( ) L R L R L R R L Laser pulse energy Spontaneous emission R 1 ; L 1 energy Nu N L Laser pulse width Number of undulator periods Number of optical cycles E () =2 A A / cos( ) L R L R E
o Electron spontaneous emission z Energy modulation is the electron phase relative to the laser wave at the undulator entrance A. Zholents, PQE, January, 2006 Laser buncher (3) Spontaneous emission energy AR 4 hL , =1/137 ( for K >> 1 ) photon energy 2 E 8 hL AL Laser pulse energy Numerical example: K=3
h L = 45eV L = R A L = 5 nJ E 150 keV (30-th harmonic of Ti:Sapphire laser produced using High-order Harmonic Generation) (in 50 fs FWHM pulse) to be compared with uncorrelated electron beam energy spread of ~ 100 keV A. Zholents, PQE, January, 2006 Laser buncher (4) Optical klystron* 30 nm, 5 nJ, 50 fs = 100 kW XUV light e- modulator radiator Nu=100 Nu=100 fragment of e-beam: modulation *) Skrinsky, Vinokurov,1977
30 nm, 30 MW XUV light bunching chicane e- fragment of e-beam: bunching A. Zholents, PQE, January, 2006 Harmonic cascade FEL bunching chicane Laser light time delay chicane - modulator radiator 240 nm 48 nm e-
* e light bunching chicane light modulator radiator 48 nm 12 nm e- evolution of e-beam phase space energ y New undulator resonant at L/n, and bunched beam radiates at n-th harmonic
phas e r ~100-fs seed laser pulse tail z Fresh bunch technique Position of FEL pulse in full electron beam pulse radiator modulator radiator head Unperturbed electrons A. Zholents, A. Zholents,
San-Diego, PQE, January, April, 2004 2006 At each modulator, radiation interacts with fresh e- 4.0 GW At each harmonic upshift (modulator to 4.0 radiator), macro- GW particle phase multiplied by n Bunching effects of dispersive 1.2 section visible in GW change from Z=6 m in 48-nm modulator to GINGER simulations Z=0.4 m LBNL
W. Fawley, 240-nm modulato r 48-nm radiato r Energy (MeV) Power vs. z and1.0 GW - scatterplots 250 8 249 2 - 251 0 Z=0 m (radians)) Z=1.8 m +
Z=0 m 249 0 -5 250 4 (radians)) +5 Z=3.6 m Z=4.4 m Z=2.4 m 48-nm 48-nm modulat modulato or r Z=0 m 12-nm radiato r 249 6 - 250
4 249 6 -4 (radians)) Z=3 m + Z=0.4 m +4 (radians)) Z=3.4 m Z=6 m Z=5.4 m A. Zholents, PQE, January, 2006 Noise evolution from imperfect seed* Input laser seed initialized with broadband (a) phase noise (b) amplitude noise
Fields resolved in simulation on 240 nm/c temporal resolution or better RMS phase noise d(t)/dt after removal of average component d(t)/dt (A.U.) GINGER simulation of 4stage cascade configuration (240 nm 1 nm); W. Fawley P Psignal 2 signal n Pnoise out Pnoise in (a) (b) Results: Noise reaches minimum at 48-nm stage (slippage aveg.) In later stages noise increases due to harmonic
multiplication of low frequency components Noise can be aproblem at 1 EXIT 240 nm *) Saldin et al., 2002 48 nm 12 nm 4 nm 1 nm A. Zholents, PQE, January, 2006 Self-Amplified Spontaneous Emission FEL* courtesy S. Reiche PradPbeam Similar to optical lasers, SASE x-ray FEL starts from spontaneous emission butavoids use of mirrors Density modulation (shot noise at start
or microbunching latter) drives energy modulation and vice-versa Instability reaches saturation after all electrons are microbunched (or rate of de-bunching equals rate of bunching) *) Kondratenko, Saldin 1980; Bonifacio, Pellegrini, Narducci 1984 u gain length 4r courtesy Z. Huang A. Zholents, PQE, January, 2006 The FEL parameter Small diffraction, radiation field interacts locally with the electron beam, i.e. optical guiding* (some similarity with fiber optics) for K>>1 peak current I x u r I A 4 b 2 1 3 IA = 17 kA beta-function
e-beam emittance No guiding, strong diffraction light emittance; x x-ray wavelength I r~ I A I Key parameters: ; IA x ; E 4 b 1 2 beam energy spread causes de-bunching *) Moore 1984; Scharlemann, Sessler, Wurtele 1985 A. Zholents, PQE, January, 2006
Transverse coherence x When b 4 spontaneous undulator radiation consists of many spatial modes, i.e. incoherent sum of individual electron emissions X b e-beam x light beam 4 (diffraction limited) x Radiation field at different locations along the undulator But FEL gain is localized within the electrons and higher-order modes have stronger diffraction : gain guided selection of fundamental mode results in fully transverse coherence even at x b 4 Zholents, PQE, January, 2006 courtesy S.A.Reiche
Temporal coherence SASE output exhibits chaotic light properties Cooperation length (slice): 15 x c 2 r E(t), a.u. E(t ) = e i( t k z j ) j -15 -60. -40. -20. tc/ length Bunch 00. 20. 40.
60. Number of longitudinal modes: M (bunch length)/slice Fluctuation in the x-ray pulse energy ~ 1/M Slice properties, i.e. slice peak current, emittance and energy spread define performance A. Zholents, PQE, January, 2006 Temporal coherence (2) M decreases as coherence builds up during the exponential gain reaching minimum at saturation (~200 at LCLS) courtesy W. Fawley A. Zholents, PQE, January, 2006 Production of bright electron beams: generation I unites in a single expression key e-beam peak brightness ~ 2 parameters for x-ray FELs b E Peak brightness of different photocathode e-guns (2002) ~100 A, n=1 mm-mrad DESY-Zeuthen new generation of e-guns courtesy P. Piot
rapid acceleration near to the cathode to avoid space charge dilution A. Zholents, PQE, January, 2006 Production of bright electron beams: preservation Physics phenomena affecting the e-beam while acceleration and compression q Non-linear effects in bunch compression: rf waveform, T566 q Longitudinal and transverse wakefields in accelerator q Space charge effects (mainly longitudinal) q Coherent synchrotron radiation (CSR) and emittance excitation q Resistive wall wakefields in undulators Technical issues q Jitter in the rf phase and amplitude in accelerating structures q Intensity and timing jitters in photocathode gun laser q Misalignment of rf structures and magnetic elements q Power supply ripples A. Zholents, PQE, January, 2006 Coherent Synchrotron Radiation (CSR) Powerful radiation generates energy spread in bends Energy spread breaks achromatic system Causes bend-plane emittance growth (short bunch worse) coherent radiation z for z L0 R
e bend-plane emittance growth l x overtaking length: L0 ~(24zR2)1/3 s E/E = 0 E/E < 0 x = R16(s)E/E CSR wake is strong at very small scales (~1 m) A. Zholents, PQE, January, 2006 Longitudinal space charge, CSR and microbunching instability Initial density modulation induces energy modulation through longitudinal space charge forces, converted to more density modulation by a compressor Current Gain=10 10% 1%
t Space charge Energy compression saturation due to overmodulation stops the growth growth of slice energy spread (and emittance) courtesy Z. Huang t A. Zholents, PQE, January, 2006 Microbunching instability (2) Entire machine with its accelerating sections, drifts and chicanes acts as an amplifier for initial density perturbation and can be characterized by a spectral gain function (in an analogy to the FELs) * 1400 FERMI FEL project Instability increases rms energy spread by a factor of 5-10 50
(m) 150 200 Spectral dependence of the gain of the microbunching instability *) Z. Huang et. al, Phys. Rev. ST Acc. and Beams, v.5, 074401(2002) A. Zholents, PQE, January, 2006 Laser heater as an instrument for a suppression of microbunching instability1,2 Laser heater (laser-e-beam interaction induce energy spread) provides Landau damping effect through controlled increase of the energy spread at the beginning of acceleration Laser heater 2.5 keV 1200 Laser heater 5 keV 800 Laser heater 7.5 keV 300 Laser heater 10 keV 150 1) E.L. Saldin, E.A. Schneidmiller, and M.V. Yurkov, DESY Report No. TESLA-FEL-2003-03, 2003. A. Zholents, PQE, January, 2006 2) Z. Huang, et. al, Phys. Rev. ST Acc. and Beams, V.7, 074401, (2004).
Wakefields Alignment errors and orbit distortions are responsible for transverse wakefields produced by e-beam, and transverse wakefields twist e-beam into a banana shape courtesy P. Emma EA z AL Other wakefields: Longitudinal wakes, Resistive wall wakes, Surface roughness wakes also do not affect slices and produce similar global variations that nevertheless can be dangerous for FEL performance Slice emittance is not affected Centroid shift and variation can be important A. Zholents, PQE, January, 2006 Pushing over the limits further improvements can be
obtained by using: 1) electron beam conditioning* 2) enhanced SASE * Talk of J. Wurtele, Wednesday evening A. Zholents, PQE, January, 2006 Electron beam conditioning* provides correlation of electron transverse amplitudes with electron energies to prevent de-bunching of electrons (more in Sesslers talk in the following session) y undulator z y z y E/E without conditioning with conditioning allows relaxed emittance requirement in FEL *) Sessler, Whittum, Yu in 1992
A. Zholents, PQE, January, 2006 Laser-assisted electron beam conditioning* e-beam 1 ~ sin ( k L z1 ) 2 ~ sin ( k L z 2 ) 1 z 2 =z1 + ( x2 + y2 )ds 2 ~ Jx + Jy d =1 2 ~ k L ( J x + J y ) cos( k L z1 ) Proposed scheme gains factor of 105 in efficiency by utilizing laser and wiggler for electron energy modulation instead of RF cavities: k L / k RF ~ 105 !!! Caution: approximately one half of electrons have wrong sign of correlation *) Zholents 2005 A. Zholents, PQE, January, 2006 Laser-assisted electron beam conditioning (2) Example: LCLS-like FEL with 2 times of LCLS emittance GENESIS simulations Beam parameters: energy = 14 GeV
peak current = 3.4 kA, energy spread = 1.2 MeV, emittance = 2.4 mm-mrad, beta-function = 20 m. 1 - no conditioning, 2 - ideal conditioning (all electrons), 3 - proposed conditioner. A. Zholents, PQE, January, 2006 ESASE-Enhanced Self-Amplified Spontaneous Emission* 30-100 fs pulse L~0.8 to 2.2m Modulation Acceleration E ~ 4.5 GeV E ~ 14 GeV Peak current, I/I0 One optical cycle Laser peak power ~ 10 GW (easy) Short wiggler, ~ 10 periods *) Zholents 2004 Bunching 20-25 kA Laser pulse width
z /L Electron beam after bunching A. Zholents, PQE, January, 2006 ESASE (2) Example for LCLS with =12 m ESASE Start-to-End simulations 3-m FODO lattice period drifts+quads occupy ~ 0.5m not compatible with current LCLS lattice design standard LCLS courtesy W. Fawley ESASE cases saturate by 50 m with 50-100 times power contrast over unmodulated part of the electron bunch - the opportunity for an absolute synchronization of a probe x-ray pulse to a pump laser pulse A. Zholents, PQE, January, 2006 Attosecond x-rays using ESASE* Combined field of two lasers *) Zholents, Penn 2005
Energy modulation of electrons produced in interaction with two lasers A. Zholents, PQE, January, 2006 Attosecond x-rays using ESASE (2) two lasers one laser Peak current after chicane A. Zholents, PQE, January, 2006 Attosecond x-rays using ESASE (3) background from 100 fs e-bunch main peak main peak entire bunch attosecond pulse side side peaks peaks 35 GW x-ray pulse energy growth over the length of the undulator ~350as A. Zholents, PQE, January, 2006
Summary X-ray FELs are as good as the electron beam is, i.e.: peak current slice emittance slice energy Production of aspread high-brightness electron beams and preservation of the electron beam quality is affected by: space charge coherent synchrotron radiation microbunching instability various wake fields Laser-assisted manipulation of electrons in the phase space is a promising concept for future FELs : electron beam conditioning enhanced self-amplified spontaneous emission A. Zholents, PQE, January, 2006 Outlook: future FEL-based multi-user x-ray facility Laser(s) FEL farm High repetition rate linac Future facility will be as much laser beam based as electron beam based and will have a multi -FEL x-ray production farm. This farm will be fed by a highrepetition rate linac (up to MHz) equipped with a high brightness source of electrons.
Optical lasers will be used for a production and shaping the electron bunches and for seeding the x-ray radiation. An advent of high-average power lasers will boost highrepetition rate FELs. A. Zholents, PQE, January, 2006 Outlook: future FEL-based multi-user x-ray facility Use of FELs will expand beyond FELs based on the SASE method (in a construction phase at present) towards FELs producing laser-like nearly Fourier transform limited x-ray beams at various wavelengths with controlled pulse duration, bandwidth, and polarization. A. Zholents, PQE, January, 2006 Gratefully acknowledged: K. Bane, J. Corlett, M. Cornaccia, P. Craevich, S. DiMitri, D. Dowel, P. Emma, W. Fawley, W. Graves, J. Hasting, Z. Huang, K.-J. Kim, S. Leone, S. Lidia, G. Penn, R. Schoenlien, A. Sessler, J. Staples, G. Stupakov, J. Wu, J. Wurtele, M. Zolotorev Thank you ! A. Zholents, PQE, January, 2006 courtesy B. Faatz A. Zholents, PQE, January, 2006 A. Zholents, PQE, January, 2006
Enzymes function at a specific temperature, pH, and concentration (amount of enzyme) Ex: pepsin enzyme in the stomach works at a pH of 2. Enzymes, Cont. Enzyme names usually end in -ase. Ex: Lipase (breaks down fats) Catalase (break down...
Mayer Riff and Ann Borden of Marin Country Day School, "Indexed Tuition and Financial Aid" (Presentation at CAIS Trustee/School Head Conference, Jan. 2012) I recommend removing the quote marks in the top and last line of questions. It looks like...
The Data Deluge " EIGHTEEN months ago, Li & Fung, a firm that manages supply chains for retailers, saw 100 gigabytes of information flow through its network each day.Now the amount has increased tenfold. During 2009, American drone aircraft flying...
He continued to replicate this experiment on the previously healthy plants which had become infected due to sap and was able to show that the infection rate did not decrease. In 1935, Wendell Stanley was able to crystallize the infectious...
Salt isnt going to vaporize and go into drop. Precipitating agent concentration First commercially Available crystallization Screening kit. Hampton Crystal Screen 1 Phase problem was first solved in 1960. Kendrew & Perutz soaked heavy atoms into a hemoglobin crystal, just...
Hook. Bridge. Thesis Statement . Body Paragraphs. Option One. Body 1- Explains how the two topics are the same. Body 2- Explains how the two topics are different. Option Two. Body 1- Explain how the two topics are same/different.