Transcript ppt

XAPPER Progress & Plans
XAPPER
Presented by: Jeff Latkowski
XAPPER Team: Ryan Abbott, Steve Payne, Susana Reyes, Joel Speth
April 9, 2003
Work performed under the auspices of the U. S. Department of Energy by Lawrence Livermore National Laboratory under Contract W-7405-Eng-48.
The XAPPER experiment will be used to
study damage from rep-rated x-ray exposure
 Source built by PLEX LLC;
delivered 10/02; operational
11/02; system testing and
characterization now complete
 Uses rf-initiated star-pinch to
generate plasma
 Operates with Xe (113 eV), Ar
(250-300 eV), N (430 eV)
JFL—4/03 HAPL
XAPPER Mission
 XAPPER is to perform rep-rated, x-ray exposures to look
for “sub-threshold” effects such as roughening and
thermomechanical fatigue.
 XAPPER provides large doses of soft (100-400 eV) x-rays;
Dose is a reasonable figure of merit, not fluence.
 XAPPER cannot match exact x-ray spectrum, but it can
replicate a selected figure of merit. For example, the peak
surface temperature, dose, stress, etc. that would occur in a
real IFE system can be matched on XAPPER.
 XAPPER will be used in the study of x-ray damage to
optics and chamber wall materials.
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
December – Crack in the ceramic within the plasma
head. Loss of vacuum, water leak into plasma
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
December – Crack in the ceramic within the plasma
head. Loss of vacuum, water leak into plasma
Resolution – Replaced ceramic with higher thermal
conductivity material; Added epoxy layer as vacuum
barrier; Operation to >100,000 pulses (everything from
1-10 Hz) without problem.
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
January-April – Fluence on sample 40x lower than spec
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
January-April – Fluence on sample 40x lower than spec
Note: We don’t actually need the full 40x
full currently envisioned experiments –
we would be quite happy with a 5x
improvement.
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
January-April – Fluence on sample 40x lower than spec
Resolution – Direct source measurements to ensure that
problem is with optic, rather than source. Confirmed that
source output is adequate.
JFL—4/03 HAPL
Source measurements indicate that
output is within 1.4x of specification
 Sampled source through foil comb at 23º
(rough center of condensing optic, when used)
 Source output is ~0.24 J/sr (0.33 expected)
 Indicates that majority of problem is with
optic
60
X-ray power (mW)
50
40
30
20
10
0
0
2
4
6
Rep rate (Hz)
JFL—4/03 HAPL
8
10
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
March-April – Continued problems with second optic.
Reticle
Phosphorescent
material
Zr filter
(passes 7-17 nm)
Incoming x-rays
CCD imaging of inner spot
JFL—4/03 HAPL
1.3 cm
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
March-April – Continued problems with second optic.
Resolution – Discussions with internal optics experts.
Testing and analysis of current optics. Decision to remove
optic from PLEX contract. Internal team to sub-contract
mandrels but coat optics internally.
JFL—4/03 HAPL
Spot size measurements made with HeNe
 Tabletop visible (HeNe) spot size measurements suggest
error must be a wavelength-dependent effect
 Vendors and EUV experts agree that likely explanation is
mid-frequency spatial roughness
~3 mm spot
JFL—4/03 HAPL
We are removing the condensing optics from the
PLEX contract; A combination of LLNL expertise
and external vendors will be used
 LLNL’s Materials Science & Technology Division (MSTD)
routinely makes collimating optics that far surpass our
figuring & roughness specifications
 Current plan:
– Outside vendor for mandrels (3):
• Roughness specification <1.5 nm RMS
• Slope error specification <1 arc-minute
– MSTD to coat optics: C, Pd, Cu, Ni
– Should get 4 good optics per mandrel
– Total cost: $6-10K/optic
 For now, we will switch to study of Al mirrors (we have
more than enough fluence for this)
JFL—4/03 HAPL
MSTD has previously produced collimating optics that far
exceed our specifications for roughness and slope error
X-ray Radial Intensity Distribution
100
simulation
80
gain
60
40
20
0
Multiple optics produced from a single mandrel
experiment
0
0.5
1
radial position (cm)
1.5
When measured figure errors (from mandrel) are accounted for,
calculations agree well with measured intensities  suggests that coating
process is not significantly degrading optical quality
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
February – Question raised if damage could be due to ions.
JFL—4/03 HAPL
XAPPER activities since December 2002
Sample plane
Star-pinch
plasma
Ellipsoidal
condenser
February – Question raised if damage could be due to ions.
Resolution – Conducted simple experiment to verify
damage due to x-rays.
JFL—4/03 HAPL
We have confirmed that damage is being
caused by focused x-rays, not stray ions
 Sample is ½” diameter Al mirror from Newport (Al on SiO2):
– Exposed to 3000 pulses at 8 Hz; tpulse ~40 ns
– Translated focusing optic (perpendicular to axis of symmetry) by ~0.9
mm between 1st and 2nd exposure sequences
 Observed movement of
damage spot, indicating
that damage is caused by
x-rays, which are focused
by the condensing optic
 Ions, if present, would not
be focused to new spot
JFL—4/03 HAPL
2nd sequence:
~0.13 J/cm2
1st sequence:
~0.19 J/cm2
Gantry was installed & spectrometer has been
mounted/aligned—testing is underway
EUV Spectrometer is mounted vertically to
intercept x-rays directed upon the pinch axis
JFL—4/03 HAPL
ABLATOR has been used to predict the
time/temperature history of an Al GIMM
370
Secondary
x-ray pulse
Temperature (K)
360
350
340
Prompt
x-ray pulse
330
320
Laser 30 ns
pulse
310
300
0.0E+00
2.0E+02
4.0E+02
6.0E+02
8.0E+02
1.0E+03
1.2E+03
1.4E+03
Time (ns)
Assumes 99% reflectivity GIMM @ 85° and 30 m, 10 mTorr Xe, 1 ns prompt,
and 1 ms secondary x-ray pulselengths. Surface zone is 10 nm thick.
nd x-ray pulse.
Full
46
MJ
assumed
for
2
JFL—4/03 HAPL
Increasing the gas pressure to 50 mTorr
helps attenuate the x-rays
Prompt
x-ray pulse
325
Temperature (K)
320
Secondary
x-ray pulse
315
310
305
Laser 30 ns
pulse
300
295
0.0E+00
2.0E+02
4.0E+02
6.0E+02
8.0E+02
1.0E+03
1.2E+03
1.4E+03
Time (ns)
Assumes 99% reflectivity GIMM @ 85° and 30 m, 50 mTorr Xe, 1 ns prompt,
and 1 ms secondary x-ray pulselengths. Surface zone is 10 nm thick.
Full 46 MJ assumed for 2nd x-ray pulse.
JFL—4/03 HAPL
We have completed a multi-material version
of ABLATOR; Testing is underway
 New version needed to analyze exposure of Newport mirrors
(and components such as tungsten armor and dielectric mirrors):
– Treatment as single, thin layer (100 nm Al) way too conservative
– Treatment as thicker Al layer non-conservative due to high conductivity
 Calculation agrees with experimental observations:
– Removal of Al at only 0.18 J/cm2
– Can actually see plasma burn through Al layer
2300
Aluminum
Fused silica
Temp (K)
2100
t=0.4 ps
X-ray pulse off
t=5 ns
1900
t=10 ns
1700
t=15 ns
t=20 ns
1500
t=30 ns
1300
t=40 ns
1100
t=100 ns
t=200 ns
900
t=300 ns
700
t=400 ns
500
300
0
JFL—4/03 HAPL
5
10
15
20
Zone #
25
30
35
40
Summary: Source characterization is completed
(for now); Ready to start hitting Al samples
 Resolve optic issues: outside contractor for mandrels and
LLNL-produced coatings
 Spectral characterization and tuning (EUV spectrometer)
 Enhance diagnostic capabilities:
– Fast (<1 ns resolution) photodiode
– Procure/install fast optical thermometer (from UCSD)
 Add ion heating to ABLATOR
 Sample testing and evaluation:
– Campaign for Al to begin (actually need to reduce fluence for optics
experiments); return to tungsten once new optics are available
– Explain effect of energy, number of pulses, fluence, etc.
JFL—4/03 HAPL
Back-up slides
PLEX LLC produces a source
that meets our needs
 Uses a Z-pinch to produce x-rays:
– 1 GHz radiofrequency pulse pre-ionizes
low-pressure gas fill
– Pinch initiated by ~100 kA from
thyratrons
– Operation single shot mode up to 10 Hz
Sample
plane
 Operation with Xe (11 nm, 113 eV):
– 70% of output at 113 eV (tunable)
– 3 mm diameter spot
Significant
– Fluence of ≥7 J/cm2
margin for laserIFE simulations
 Several million pulses
before minor maintenance
JFL—4/03 HAPL
Ellipsoidal
condenser
Star-pinch
plasma
The ellipsoidal condenser
is not performing to specification
 Specification calls for <3 mm spot size,
which provides >7 J/cm2
 Experiments using a phosphorescent disk
indicate a large (~1.5 cm) spot
 Expected energy appears to be there; will be
confirmed with calorimeter experiments
Reticle
Phosphorescent material
Zr filter (passes 7-17 nm)
Incoming x-rays
CCD image of
inner spot
JFL—4/03 HAPL
OptiCAD spot
calculation
(with MFSR)
X-ray fluences in IFE and
ICF systems will be significant
 Direct-drive dry-walls:
 Indirect-drive liquid walls:
– Thick-liquid jets: ~1 kJ/cm2
– Wetted wall/vortices:
30-80 J/cm2
8
Total = 115 MJ
X-ray output (J/keV)
– Chamber: ~1 J/cm2
– Final optics: ~100 mJ/cm2
10
10
6
10
4
Total = 6.1 MJ
10
2
NRL target
HIF target
0
 NIF ignition targets:
– Diagnostic @ 1 m: ~40 J/cm2
– First wall @ 5 m: ~3 J/cm2
– Final optic @ 6.8 m: ~2 J/cm2
JFL—4/03 HAPL
10 -3
10
10
-2
10
-1
0
10
10
Energy (keV)
Target output calculations
(1-D LASNEX) courtesy of
John Perkins, LLNL
1
10
2
Photodiode signal = 4.1 V (~0.18 J/cm2);
3000 pulses @ 8 Hz
JFL—4/03 HAPL
The x-ray exposure significantly
reduced the mirror reflectivity
1.0
0.9
0.8
0.7
Reflectivity (%)
 Reflectivity
measurement
averaged over a
5-mm-diameter area
centered over
obvious damage site
11-27-02, Newport AL-2 (broadband aluminum) mirror
before and after 3000 pulses
before
after
0.6
0.5
0.4
0.3
0.2
0.1
0.0
200
400
600
Wavelength (nm)
NOTE: This mirror looks very different from
what an IFE final optic would look like.
JFL—4/03 HAPL
800
1000
X-ray damage: need for rep-rated exposures
Chamber Wall Temperature (deg C)
 Design
can provide systems
3000
Surface
that avoid
significant
single1 micron
2600
5 microns
shot damage
2200
10 microns
1800
100 microns
 Single-shot
results are not
1600
adequate;
1200 miss:
154 MJ Target
Tungsten wall @ 6.5 m radius
– Thermal
fatigue
600
No gas in chamber
– Surface
200 roughening (RHEPP
2
4
6
8
10
results,0UW analyses)
(msec)
– Difficult to assesstime
very
small
ablation levels
 Analyses need to consider
multi-shot effects; rep-rated
exposures are needed
JFL—4/03 HAPL
Result from UCSD
3000
Surface temperature
2500
2000
1500
400 MJ Target
Graphite wall @ 8.25 m radius
25 mTorr Xe in chamber
1000
500
0
2
4
6
time (msec)
8
Single-shot results
are not sufficient
10
Single-shot results, (Cont’d.)
 Single-shot, laser-induced
damage threshold is ~140 J/cm2
Data courtesy of Mark Tillack,
University of California at San Diego
532 nm light
fluence quoted is normal to beam
 Multiple-shot operation is only
safe at a small fraction (~40%?)
of the single-shot threshold
 Gradual optical degradation
explained (ref: Ghoniem) as
roughening caused by migration
of dislocation line defects
 While length scales will differ
(eV vs. keV), laser/x-ray
physics should be quite similar
JFL—4/03 HAPL
Rep-rated x-ray damage
studies are needed
Significant damage was found throughout the
unshielded region using white-light interferometry
 ~250 nm removed over visible
damage site
 Peak-to-valley removal >500 nm
 Considerable pitting throughout
unshielded region
(concentrated in
obvious damage area)
 Semi-regular
“roughening” observed
– seems consistent
with RHEPP results
JFL—4/03 HAPL