ICF-related research at Strathclyde Paul McKenna University of Strathclyde
Dec 21, 2015
ICF-related research at Strathclyde
Paul McKenna
University of Strathclyde
EPSRC grant: EP/E048668/1 Key physics for ICF diagnosed by ion emission
1. Fast electron generation and transport in dense plasma
2. Shock propagation physics
3. Laser-ion source development (ion fast ignition)
• (Nuclear diagnostics of laser-plasma)
Fast electron generation and transport:
Diagnostic Energy range
Issues
K emission 10s keV Wavelength shift with temperature
CTR / OTR emission
MeV Limited to thin targets due to electron bunch dephasing
“Escaped” electron spectrometry
MeV Target charges to MV potentials
Notoriously difficult to measure fast electrons in solid targets. Each diagnostic has limitations due to assumptions, model dependences etc.
Examples:
Our approach: Ion emission as a diagnostic
Laser pulse
Target RCF stack Sample RCF
Beam sampling for analysis
Protons
Structured sheath
• Maximum proton energy electron density (MeV energies)• Intensity distribution electron transport filamentation• Proton divergence with energy electron sheath profile • Proton spectrum electron temperature (model)
• Thick solid density targets can be investigated (>mm)
0 100 20015
20
25
30
35
40
Scale length, LO
(m)
Ma
xim
um
pro
ton
en
erg
y (
Me
V)
0 100 2000
1
2
3
4
5
6
7
Scale length, LO
(m)
En
erg
y c
on
ve
rsio
n e
ffic
ien
cy (
%)
Iabl
t
0 100 20015
20
25
30
35
40
Scale length, LO
(m)
Ma
xim
um
pro
ton
en
erg
y (
Me
V)
0 100 2000
1
2
3
4
5
6
7
Scale length, LO
(m)
En
erg
y c
on
ve
rsio
n e
ffic
ien
cy (
%)
0 100 20015
20
25
30
35
40
Scale length, LO
(m)
Ma
xim
um
pro
ton
en
erg
y (
Me
V)
0 100 2000
1
2
3
4
5
6
7
Scale length, LO
(m)
En
erg
y c
on
ve
rsio
n e
ffic
ien
cy (
%)
Iabl
t
0 100 20015
20
25
30
35
40
Scale length, LO
(m)M
axim
um
pro
ton
en
erg
y (
Me
V)
0 100 2000
1
2
3
4
5
6
7
Scale length, LO
(m)
En
erg
y c
on
ve
rsio
n e
ffic
ien
cy (
%)
0 100 20015
20
25
30
35
40
Scale length, LO
(m)
Ma
xim
um
pro
ton
en
erg
y (
Me
V)
0 100 2000
1
2
3
4
5
6
7
Scale length, LO
(m)
En
erg
y c
on
ve
rsio
n e
ffic
ien
cy (
%)
1011
1012
1013
0
5
10
15
20
25
30
35
40
Intensity (W/cm2)
Shock f
ront po
sition (u
m)
Cu 0.5 ns
Cu 1.5 ns
Cu 3.5 nsAu 0.5 ns
Au 1.5 ns
Au 3.5 ns
-60 -40 -20 0 2010
-4
10-3
10-2
10-1
100
101
102
X-position (microns)
Densi
ty (g/c
m3)
0.5 ns
1.5 ns3.5 ns
1011
1012
1013
0
5
10
15
20
25
30
35
40
Intensity (W/cm2)
Shock f
ront po
sition (u
m)
Cu 0.5 ns
Cu 1.5 ns
Cu 3.5 nsAu 0.5 ns
Au 1.5 ns
Au 3.5 ns
-60 -40 -20 0 2010
-4
10-3
10-2
10-1
100
101
102
X-position (microns)
Densi
ty (g/c
m3)
0.5 ns
1.5 ns3.5 ns
Proton measurements show that controlled preplasma expansion leads to enhanced energy coupling to fast electrons
1: Laser propagation and energy absorptionP. McKenna et al, LPB 26 591-596 (2008) D.C. Carroll et al, CRP 10 188-196 (2009)
OSIRIS Simulations
-200 -100 0
1018
1020
1022
1024
X (m)
Ele
ctro
n d
en
sity
(cm
-3)
-10 0 10 2010
23
1024
1025
X (m)
n e (cm
-3)
Pollux 0.5 nsPollux 3.5 nsExpt. 0.5 nsExpt. 3.6 ns
• Preplasma expansion enhances electron energy spectrum
• Self focusing and beam break-up observed
• changes to the electron injection angle
0 20 4010
3
104
105
Electron energy (MeV)
Num
ber
of e
lect
ron
(arb
. uni
ts) Sharp gradient
Pollux 0.5 nsPollux 3.5 ns
Sharp density gradient
0.5ns Pollux density profile
3.5ns Pollux density profile
Simulations by Roger Evans
Evidence of collimation of fast electrons in solid targets by self-generated B-field observed using proton emission
2. Collimation of fast electron transport
Yuan…McKenna., submitted (2009)
0 300 600 900 1200 15000
10
20
30
40
Target thickness (m)
Max
imum
pro
ton
ener
gy (
Me
V)
ballistic model (27o) RCFTP-Spec
0 300 600 900 12000
100
200
300
400
500
Target thickness (m)
Sh
ea
th d
iam
ete
r (
m)
ballistic model (27o)Inferred from expt.
Simulations with 2-D hybrid LEDA code
Electron refluxing within thin targets perturbs B-field structure
Ne no B field Ne with B field
Simulations by Alex Robinson (RAL)
3: Effects of target material on beam filamentation
CH SiO2 glassBk7 glass
Li Al Au
Laser pulse
Target RCF stack
Sample RCF
Protons
Target Effective Z Resistivity[Ω.m]
Al 13 10-8
C3H6 5.4 1013
Li 3 10-7
SiO2 11.6 1014
ZEPHROS hybrid-PIC simulationsSimulations by Alex Robinson (RAL);
Li curve calculation by Mike Desjarlais (Sandia)
4. Shock propagation physics• Exception sphericity of implosion required for ICF
• Non-uniformities in illumination or target roughness amplified by Richtmeyer-Meshkov and Rayleigh-Taylor instabilities
• Uniform drive pressure can result in non-uniform shock propagation depending on grain alignment in the material
• e.g. Be is naturally polycrystalline with different shock velocities along different crystal axes – grain size is ~10 m
D Swift et al.,
Shock uniformity measurements using proton emission
Our approach – use proton emission imaging to measure perturbations of the initial shock breakout
CPA illumination timed to coincide with shock breakout thus imprinting the rear surface geometry on the ion emission.
Proof-of-principle tests in January 2010
Sub-micron structure on target surface
Reproduced in proton beam
M. Roth et al., PR-STAB 5, 061301 (2002)
Lindau et al., PRL 95, 175002 (2005)
Proton emission is sensitive to shock breakout
Laser pulse energy (J)0 100 200 300 400
Con
vers
ion
effic
ienc
y (%
) to
pro
tons
w
ith e
nerg
y gr
eate
r th
an 4
MeV
0
1
2
3
4
5
6
7
8
10 micron25 micron
Robson et al, Nat Phys 2007
EL
10 µm Al, ~5 µm
25 µm Al, ~5 µmControlling the front surface density
gradient gives a factor of 2 increase in conversion efficiency
2 µm Al, ~80 µm, ~1019 Wcm-2
Thin targets and defocused laser spot gives even higher conversion efficiency
• DT fuel at 300g/cc• 35 m ignition spot
Curved proton rich target
5: Laser-ion source development
• Proton energy scaling with ps pulse
• Spectral shaping with dual CPA pulses
• Techniques to enhance conversion efficiency
1. Proton emission applied as a diagnostic of fast electron generation and transport
Examples:
• Electron generation as a function of plasma scale length
• Collimating effect of self-generated magnetic fields
• Electron transport filamentation
• Electron transport in compressed targets (HiPER, LULI)
2. Shock propagation physics
• Ion diagnostic technique to be trialled in January 2010
3. Laser-ion source development (ion fast ignition)
• Spectral control and enhancement of conversion efficiency
4. Nuclear diagnostics of laser-plasmas
Summary of ICF-related physics at Strathclyde
Collaboration:
P. McKenna et alSUPA, Department of Physics, University of Strathclyde
D. Neely, A.P.L. Robinson et al STFC, Rutherford Appleton Laboratory
R G Evans Imperial College London
M. Borghesi, M. Zepf et al School of Mathematics and Physics, Queen’s University Belfast.
J. Fuchs et al LULI Ecole Polytechnique, France
M. P. DesjarlaisSandia National Laboratories, New Mexico
6: Nuclear activation
1 – Development of laser-plasma nuclear diagnostics
• choice of activation reactions with well-known cross sections;
• spectral, spatial and yield measurements of n, , ions;
• significant development work required for in-situ measurements in noisy plasma environment, using radiation hardened detectors;
2 – Innovative nuclear diagnostics
Examples may include:
• fusion reaction history measurements using gamma detectors (NIF) (D + T + 5He);
• charged particle detection to measure yield of neutronless reactions (e.g. D + 3He p (15 MeV) + 4He);
• Higher nuclear yields expected; observation of lower cross section and higher threshold energy reactions;
Effect of angle change with energy
Magnetic field generation during electron propagation is described by combining Ohm’s law with Faraday’s law:
fjE Ohm’s law
= resistivity
= fast electron current density
fj
Generates a magnetic field that pushes electrons towards regions of higher current density
Generates a magnetic field that pushes electrons towards regions of higher resistivity
ff jjB
EB
t
t
Robinson and Sherlock, Phys. Plasmas, 14, 083105 (2007)
Fast electrons
B field
Homogeneous plasma
laser
Resistive generation of toroidal B-field. B-field pinches the fast electron beam
Magnetic collimation
Results: Maximum proton energy
Ballistic transport and P. Mora PRL 2003 plasma expansion model.
sheatheppNppiBp netTkE /1~~,2/,)]1[ln(2 22
The scaling with target thickness is significantly different than expected from ballistic electron transport
Yuan et al., Vulcan TAP Fuchs et al., LULI (Nat. Phys. 2006)
Carroll et al, Phys. Rev. E 76, 065401 (2007)
Bea
m d
iver
gen
ce (
deg
rees
)
Proton energy (MeV)
Assume sheath profile
),()(),( max txHtFEtxE
0w
Ionizes hydrogen proton
Ion front profile
Sheath profile
Beam divergence
Good fit?
0w
NoYes
Measured divergence
0w
modify
Initial sheath diameter
Sheath expansion is modelled
10um
Divergence
Source size
Model is benchmarked using grooved target results
Scaling of the collimation effect
Bell-Kingham theory: In the limit of substantial heating, collimation parameter:2
,2/15/2
sec5/22/1
511,10/3
511,5/15/25/25/3
23 )2(ln13.0 radpmffTW tRTTPZn
Electron temperature from LEDA simulations
0 50 100 150 200 250
186
188
190
192
194
196
p / mec
log
(f 00(p
)) (
at o
bs.
ce
ll)
0 50 100 150 200 250
186
188
190
192
194
196
p / mec
log
(f 00(p
)) (
at o
bs.
ce
ll)
LEDAfit at 13MeVfit at 7MeVfit at 8MeV
Temperature variation
Black line is electron spectrum at rear surface of a 400 micron Al target
Red is fit using the input electron distribution and temperature (9.2 MeV)Same temperature!
Artificially increasing scattering:
Increase electron-ion scattering rate LogΛ = 2 → 10 → 100
Marginal effect on beam smoothness
Field duration variation
P Mora PRL 2003 plasma expansion model
0 500 1000 15000
10
20
30
40
50
60
Target thickness (um)
Ma
xim
um
pro
ton
en
erg
y (M
eV
)
p
2p
4p
-600 -400 -200 0 200 400 6000
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5x 10
11
Position (um)
Ele
ctric
al f
ield
(V
/m)
E-field with space at peak field time
The temporal evolution is assumed to be a combination of Gaussian increase and exponential decrease.
This trend is supported by LEDA simulation, as well as the previous reports.
McKenna et al PRL 98, 145001 (2007)Kar et al PRL 100, 105004 (2008)
Electrical field transverse distribution is assumed to be parabolic function
Carroll et al PRE 76, 065401 (2007)Brambrink et al PRL 96, 154801 (2006)
),()(),( txHtFEtxE p
0)/exp(
0)2/exp()(
0
22
ttt
tttF
)(4/1),( 2 tpxtxH 2/)()( 2
0 vtwtp
-1 -0.5 0 0.5 1 1.5 2 2.50
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-2 -1 0 1 2 30
200
400
600
800
1000
1200
1400
1600
Time (ps)S
heat
h fu
ll w
idth
(um
)
-800 -600 -400 -200 0 200 400 600 8000
1
2
3
4
5
6
7x 1011
Position (um)
She
ath
stre
ngth
(V
/m)
W0
Sheath size and source size with time
An example fit of beam divergenceExample sheath field profiles at different time
Transverse expansion velocity with time
From Patrizio Antici’s PhD thesis E Brambrink et al PRL 96, 154801 (2006)
Both suggest an exponential decrease of expansion velocity with time
Comparing simulation and experiment results
Simulation densities used in plasma expansion model
Sheath size as a function of target thickness
Reduced growth in sheath size for thick targets
Lateral expansion of the fast electrons is limited
Self-induced fields become more important in thicker targets
title
I = 5 x 1020 W/cm2
Lancaster et al., PRL 98, 125002 (2007) Green et al., PRL 100, 015003 (2008)
Target thicknesses ~100 µmDiagnostics:
•K emission•XUV emission•Shadowgraphy
title
Fast electrondensity
Magnetic Fields