High Energy Density Science and Free-electron Lasers P. A udeb ert, H. B aldis, J. B enage, M . Bergh, C . C aleman, R. C aub le, P. C elliers, M .H. C hen, H.K. C hung, G. C ollins, M. Fajardo, R. Falcone, R. Fedosejevs, E. Foerster, J. Gauthier, S. Glenzer, E. Glover, G. Gregori, J. Hajdu, P. Heimann, S. L. Johnson, L. Juha, F. Y. Khattak, J. Krzywinski, R. W. Lee, A. Lindenbergh, J. Meyer-ter-Vehn, S. Moon, T. Möller, W .L. Morgan, M. Murillo, A. Nelson, A. Ng, Y. Ralchenko, R. Redmer, D. Riley, F. Rogers, S. J. Rose, F. Rosmej, W . Rozmus, R. Schuch, H. A. Scott, T. Schenkel, D. Schneider, J. R. Seely, R. Sob ierajski, K. Sokolowski-Tinten, T. Stoelker, S. Toleikis, T. Tschentscher, H. W ab nitz, J. S. W ark., K. W idmann, P. Zeitoun… LULI, UC Davis, LANL, Uppsala, LLNL, IST-GoLP, UC Berkeley, Jena, CELIA, LBNL, RAL, Stanford, PSI/SLS, Czech Academy, QU Belfast, Polish Academy, SLAC, MPI, TU Berlin , Kinema, NIST, Stockholm, Rostock, AW E, Marseille, Alberta, W arsaw, Essen, GSI, DESY, Oxford, LIXAM…
48
Embed
High Energy Density Sciencephotonscience.slac.stanford.edu/.../talks/RLee_HED_Physics.pdf · High Energy Density Science and ... •Frontiers in High Energy Density Physics: ... need
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
High Energy Density Science
and
Free-electron LasersP. Audebert, H. Baldis, J. Benage, M. Bergh, C. Caleman, R. Cauble, P. Celliers, M.H. Chen, H.K.Chung, G. Collins, M. Fajardo, R. Falcone, R. Fedosejevs, E. Foerster, J. Gauthier, S. Glenzer, E.Glover, G. Gregori, J. Hajdu, P. Heimann, S. L. Johnson, L. Juha, F. Y. Khattak, J. Krzywinski, R. W. Lee,A. Lindenbergh, J. Meyer-ter-Vehn, S. Moon, T. Möller, W .L. Morgan, M. Murillo, A. Nelson, A. Ng, Y.Ralchenko, R. Redmer, D. Riley, F. Rogers, S. J. Rose, F. Rosmej, W . Rozmus, R. Schuch, H. A. Scott,T. Schenkel, D. Schneider, J. R. Seely, R. Sobierajski, K. Sokolowski-Tinten, T. Stoelker, S. Toleikis, T.Tschentscher, H. W abnitz, J. S. W ark., K. W idmann, P. Zeitoun…
LULI, UC Davis, LANL, Uppsala, LLNL, IST-GoLP, UC Berkeley, Jena, CELIA, LBNL, RAL,Stanford, PSI/SLS, Czech Academy, QU Belfast, Polish Academy, SLAC, MPI, TU Berlin , Kinema,NIST, Stockholm, Rostock, AW E, Marseille, Alberta, W arsaw, Essen, GSI, DESY, Oxford, LIXAM…
• 1995: 1st comprehensive report on HEDS (energy density > 105J/cm3)
• Science on High Energy Lasers (Lee, Petrasso, & Falcone)see http://www.llnl.gov/science_on_lasers/
• 2003: Two NAS reports highlighted HEDS:
• Connecting Quarks with the Cosmos:Eleven Science Questions for the New Century
• Frontiers in High Energy Density Physics:The X-Games of Contemporary Science
• 2004: National Taskforce on HEDS formed to setpriorities and develop coordinated interagency plan• Specifically addressed “fostering…HED physics in US”
• Frontiers for Discovery for High Energy Density Physics(Davidson Report, July 2004).
Interest in HEDS is growing within thescientific community at large
High Energy Density matter isinteresting because it occurs widely
• Hot Dense Matter(HDM) occurs in:• Supernova, stellar interiors,
• Directly and indirectly driveninertial fusion experiments
• Warm Dense Matter(WDM) occurs in:• Cores of large planets
• Systems that start solid andend as a plasma
• X-ray driven inertial fusionexperiments
HED
WDM
Hydrogen phase diagram
Estimates of HED time scalesElastic e--e- collision frequency:
Elastic ion-e- collision frequency:
Inelastic ion-e- collision rate:
Photopumping Rate:
Spontaneous Decay Rate:
Ion plasma frequency:
Hydrodynamic time scale:
!
"ee# 6 $10
%6ne&eeT3 / 2
|| T(eV ),ne(cm
%3),&
ee' ln(
ee~ 2
!
"ie#1$10
%9niZ2&
ieT3 / 2
|| ni(cm
%3)
!
Rexcitation
" 3 #10$8ne
e$(E
UL/T )
EUL
T || E
UL= excitation energy (eV)
!
Rionization " 3 #10$6ne%
E1(Ip /T)
Ip T || % =#valence e$; Ip = ionization potential (eV)
!
Rphoto " 774 fLUF
# laserElaser
2
!
fLU = absorption oscillator strength; F = flux (W/cm2
)
" = fractional bandwidth; E = laser energy (eV)
!
Avalue
" 4.7 #108Z2 || atomic # = Z; H - like 2 $1
!
" pi #141 ni /Z
!
thydro "1.2 #10
$10
T || time surface moves 1 µm
WDM and HDM rates indicate short pulsesrequired to access the physical processes
2x10125x1013νion plasma
5x10105x108Avalue
2x10112x1011Rphotopump
(f~0.2;Δ~0.003, F=1014,EL~5000)
5x10115x1014Rionization (EIP~T)
1x10103x1013Rexcitation (Eul ~T)
9x1093x1012νei
1x10132x1016νee
HDMZ~ 10, ne~1022 cm-3, T~500 eV
WDMZ~ 1, ne~1023 cm-3, T~10 eV
• To remove hydrodynamic effects one requires probe/pump at less than 1 ps
Hot DenseMatter
For HDM the short pulse intense x-raysource creates a unique initial state• Population kinetics is complex for realistic cases
• The model construct requires vast amounts of atomic data• Atomic data: Energy levels, oscillator strengths, autoionization rates• Collisional cross-sections for excitation (BB) and ionization (BF)
processes
• Due to the vast number of states and the effects of the plasmaenvironment, additional model assumptions are required
Creating Warm Dense Matter with FLASHis initial step for eventual XFEL research
• Isochoric heating• 40 fs 60 Å VUV-FEL heats a Al foil 500 Å uniformly
• ⇒(1012 x 200 eV) / Volume = 3/2 (1.7x1023) x 10 eV
• Volume = Area x 500 Å ⇒ Area = 50 µm spot
• For 1 eV plasma a 140 µm spot is needed
• Isentropic expansion• A optical FDI probe measures the isentropic expansion
!
E
V= 3
2neTe
+ niIP
i
i
"
Heating
500 Å Al
VUV-FEL
Simulations of FLASH VUV-FEL confirmsimple estimates for creating WDM:• 500 Å Al irradiated by split FLASH 40 fs beam
• Temperature and density at 200 fs after the FLASH pulse140 µm spot 30 µm spot
FLASH experiment are straightforward:Transmission vs Intensity• Disparity between the various approximations represents
the state of uncertainty
• Varying intensity to >1016 accesses important regime• 5x1016 represents ~40 µJ - within current FLASH operation
I (W/cm2)
Abso
rbed
frac
tion
IBCold κν<.025 eV; EOS >.025 eVCold κν< 10 eV; EOS > 10 eV
13.5 nm
50 nm Si3N4 foil
diode
Thin uniformly heated sampleis essential
To study LCLS interaction with matterneed non-Maxwellian electrons kinetics
Electron thermalization due to elastic collisions with e- and ions Collisional excitation/de-excitation and ionization/recombination Sources such as collisional, photo and Auger electrons Sinks such as 3-body, radiative recombination and e- capture
Elastic losses to phonon (deformation potential) scattering Ionization potential depression using quasi-bound states Treatment of extremely fast particles
• Study a VUV-FEL case:• 200 eV; 200 fs pulse; ΔE/E~0.003; 1012 photons; 40µm spot
!
" ne (#)
" t=" ne (#)
" t
$
% &
'
( ) Elastic
+" ne (#)
" t
$
% &
'
( ) Inelastic& Superelastic
+" ne (#)
" t
$
% &
'
( ) Sources
*" ne (#)
" t
$
% &
'
( ) Sinks
+" ne (#)
" t
$
% &
'
( ) Electron*Electron
Al
FEL-solid interaction creates uniquephotoelectron generated plasmas• Case study for λ ~ 200 eV (FLASH)• Primary innershell photoelectrons produced at 105 eV• e- thermalize due to inelastic electron-ion collisions• Average e- energy sharply decreases then rises
0.0001
0.001
0.01
0.1
1
10
1 10 100
5 attoseconds
0.0001
0.001
0.01
0.1
1
10
1 10 100
24 as
0.0001
0.001
0.01
0.1
1
10
1 10 100
120 as
0.0001
0.001
0.01
0.1
1
10
1 10 100
1 fs
0.0001
0.001
0.01
0.1
1
10
1 10 100
3 fs
Electron energy (eV)
f e (#
/cm
-3/e
V)
0.0001
0.001
0.01
0.1
1
10
1 10 100
10 fs • At 5 attoseconds: Te ~65 eV Ne ~1016 cm-3
Ni ~6x1022 cm-3
• e--e- elastic νee : Coulomb ~1.4x109 s-1
• e--ion inelastic νei :
excitation ~5x1016 s-1
ionization ~2x1016 s-1
High PressureStates
Two areas of interest for studies ofdynamics of materials under high pressure• For studies of material strength one requires both high
pressure and high strain rates.
• HEDS capability will generate high pressures for > 100 ns• In situ studies of dislocation dynamics can be performed at LCLS• Phenomenology and MD simulation predict dislocation densities
orders of magnitude larger than measured post-shock• Creation and destruction of dislocation is dynamic => need short
duration high intensity x-ray pulse as an in situ probe
• For phase transformations the LCLS HEDS capability willprovide information on sub-ps timescales
• Phase transformations can occur on times scales <100 ps• MD simulations indicate, e.g., Fe goes through a ~1ps phase
transformation
High pressure studies illustrate a uniquefeature of the intense short pulse x-rays
• Hydrodynamic times are usually considered slow (>> 1ps)
• In cases where phase changes occur two aspects ofdiffraction require sub-ps pulses
• First, when one wants to look at a sample the undergoes bulksolidification the smearing of the signal due to locally rapidmodification will compromise the data (Ta study by Steitz)
• Second, there are currently indication that some, i.e.,diffusionless orMartensitic, transitions may undergo phase changes very rapidly(Fe study by Kadau)
Real-time measurements of solidificationprocess are possible with the LCLS
Nucleation of solid SolidificationLiquid state
• Test done to find converged result as function of sample size• Use 1.6x106 atoms (50 nm cube) at 5000 K compressed isotropically• Total physical time was ~ 1 ns• Used 1 CPU millenium on BlueGene/Light
• Use output of Ta solidification MD to run diffraction LCLS “experiment”
MD MD MD
Iron is important due to our geophysicsand developments of modern technology• Phase diagram shows Fe is BCC at ambient conditions and under a
shock goes to HCP
S. K. Saxena & L. S. Dubrovinsky, American Mineralogist 85, 372 (2000).J. C. Boettger & D. C. Wallace, Physical Review B 55, 2840 (1997).C. S. Yoo et al., Physical Review Letters 70, 3931 (1993).
A.M. Dziewonski and D. L. AndersonPhysics of the Earth and Planetary Interiors 25, 297 (1981).
LCLS enables real-time, in situ study ofdeformation at high pressure and strain rate
• MD simulation of FCC copper
• X-ray diffraction image using LCLS probe ofthe (002) shows in situ stacking fault data
0
0
Diffusescattering
from stackingfault
Peak diffractionmoves from 0,0due to relaxationof lattice under
pressure
Periodic features ⇒ averagedistance between faults
However, the MD simulations indicatesthat the transition takes ~ 1 ps
Grey = static BCC Blue = compressed BCC Red = HCP
• 8x106 atoms, total run time 10 ps(K. Kadau LANL)
• The transition shows that theshock along [100] axis goes fromBCC (α) to HCP(ε)
• However, if shock along [111]70% goes to to FCC(γ)
• This may indicate diffusionlesstransitions at very short timesmay take place
• In any event only a sub-psintense x-ray source will be ableto disentangle these results.
XFELas a
probe
Current x-ray phase-contrast imaging at ~ 5µm resolution uses laser-plasma sources
dzzr
! ""#
$%%&
'() 1),(
0*
*
0
)(
!
! "r
0
)(
I
rI !
Current techniques are limited by spatial coherence & flux of laser-plasmax-ray source [D. G. Hicks 2006]
IterativePhase
Retrieval
Tomo-graphic
inversion
Projected density Density profileImage intensity
Laserdrive
Laser-plasmasource
Quasi-sphericalshock
5.2 keVHe-α x rays
+200
-200
0
+2000-200
µm
µm
shockfront
interface
CH Al
LCLS will enable coherent diffractive x-raymicroscopy at the nanoscale
Dynamic processes on the nanoscale: shock front size (viscosity), phasetransition kinetics, nucleation & growth, grain structure deformation
CoherentXFEL beam
Shock front @ 2 µmresolution
Shock front @ 50 nmresolution
Phase-contrastimaging (near-field)
Coherent-diffractiveimaging (far-field)
Coherentx-ray beam
Zoneplate
Order-sorting
aperture
Shockedsample
Laserdrive
Coherent Microscopy
Single-pulse imaging:> 107 photons/pulse
Phaseretrieval
g
Satisfy imageconstraints
G=|G|eiϕℑ{g}
Satisfy objectconstraints
G’=|F|eiϕg’ ℑ-1{G’}
Object space Image space
X-ray ‘Thomson Scattering’ will provide aunique probe for HED matter
• Scattering from free electronsprovides a measure of the Te, ne,f(v), and plasma damping
⇒ structure alone not sufficient for plasma-like matter
• Due to absorption, refraction andreflection neither visible norlaboratory x-ray lasers can probehigh density ⇒ little to no high density data
• FEL scattering signals will be wellabove noise for all HED matter
• Direct multiphoton ionization: KxLyMz+2hν XFEL K0LyMz +2e
XFEL
spectrometer
t > 1 ps XFELpumps Mg
plasma
• Study the K0LyMz K1Ly-1Mz+hνemitted
Hollow ion studies in the HED regime will yielddata on kinetic processes and diagnostics
• LCLS will create unique states of matter and provide first hollow ions• Simulations: 5x1010 photons, 30 µm spot into a ne=1021 cm-2 plasma
• Recombination kinetics take > 10 ps• Time-integrated spectrum shows
dominance of hollow ion emission
• At 1.85 keV maximize He-like state
• 3.10 keV ionize to bare nucleus in < 50 fs
In Warm Dense Matter regime the hollow ionsprovide time-resolved diagnostic information
• XFEL forms unique states and can provide in situ diagnostics with100 fs resolution
• 5x1010 1.85 keV photons in 30 µm spot into a ne=1023 cm-2 plasma• Strong coupling parameter, Γii = Potential/Kinetic Energy ~ 10
• Spectra vary measurably with Te • At high ne emission lasts ~100 fs
Saturating the continuum using the FELmay provide a ~100 fs absorption source• He-like B plasma at 30 eV, 5x1022 cm-3, 1 mm in length• FEL tuned to H-like 1 -2 transition
Opacity and Emissivity Continuum rises rapidly and last for ~100 fs
Summary of HEDS using sub-ps intensex-ray sources• For both the hot and warm dense matter regimes the
possibilities opened up by the FELs are important
• For WDM the FELs provide• Fast uniform heating source to create WDM• Diagnostic potential: Thomson Scattering, Kα temperature
measurement, fast absorption sources, phase contrast imaging,diffraction for high pressure states
• For HDM the FELs provide:• Fast deposition may create hot, high pressure matter (not shown)• Plasma spectroscopic probes of kinetic and radiative processes• Diagnostic potential: Thomson scattering