Transcript 21 SNEAD on optics 23Oct0855lancesne

Performance of Dielectric Mirrors for
Inertial Fusion Application
Lance Snead, Keith Leonard, and Jay Jellison
Oak Ridge National Laboratory
Mohamed Sawan
University of Wisconsin, Madison
Tom Lehecka
Penn State University
High Average Power Laser Program
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Workshop
University of Wisconsin, Madison
October 22-23, 2008
Background : Dielectric Mirrors
• Mirrors are composed of alternating layers of a high and low refractive
index films deposited on a substrate. The path difference between the
thinner high index films and the thicker low index films results in
constructive interference of the reflected light.
• Mirrors is tailored to achieve high reflectivity in a specific wavelength band.
• Ultra-high reflectivity (>99%), as compared to metal mirrors in the UV range:
– Aluminum (80-90%)
– Molybdenum (50-60%)
The reflectivity (R) of a lossless multilayer stack of N
successive quarter wave layers of alternating high (nHi)
and low (nLi) refractive index.
– Tungsten (40-50%)
– Silver and Gold (<40%)
High refractive index layer
Low refractive index layer
1 b 
R 

1 b 
2
n 
b   Ni 0 Li 
n Hi 
Substrate
Geometrical Model Used in 3-D Neutronics Analysis
Bio-Shield
Shield
Focusing (M2)
Blanket
Turning (M3)
GIMM (M1)
Beam Duct
Fast Neutron Flux Distribution in Final Optics of HAPL
.0003 dpa
lifetime
1.0 dpa
2 year
SiC GIMM
M2
.02 dpa
lifetime
Flux (n/cm2s)
M3
Background: Neutron Irradiation of Dielectric Mirrors
Differing opinions as to the use of dielectric mirrors in nuclear environments.
E.H. Farnum et al. (1995)
•
HfO2/SiO2, ZrO2/SiO2, and TiO2/SiO2 mirrors on SiO2 substrates.
•
Neutron fluence: 1019 n/cm2, 270-300ºC.
•
Excessive damage in HfO2/SiO2 and ZrO2/SiO2 mirrors, including flaking and crazing of films.
Orlovskiy (2005)
•
TiO2/SiO2, ZrO2/SiO2 mirrors on KS-4V silica glass.
•
Neutron fluence: up to 1019 n/cm2, 50 ºC.
•
Dielectric mirrors showed no significant damage under irradiation, mirrors were severely
damaged upon annealing (crazing.)
Observations and opportunity ?
•
Fewer and thinner bi-layers may improve resistance to radiation and thermal effects.
•
Poor performance from SiO2 substrates may be limiting performance; suggested use of more
damage resistant substrates (eg: sapphire.)
•
Damage resistance is sensitive to quality/purity of materials. Explore very high purity.
HAPL Irradiation : Test Samples
•
Test samples consisted of 3 dielectric mirror types along with single-layer
films to evaluate film / substrate interactions.
•
Higher damage tolerant sapphire substrates used instead of SiO2.
•
Films deposited by electron beam with ion-assist; Spectrum Thin Films Inc.
•
GE-124 fused silica bars included in test matrix.
Sample
Quantity
Film Thickness / Description
Sapphire substrates only
18
6 mm diameter x 2 mm thickness
Al2O3 single-layer on sapphire
18
Film thickness (36 nm)
SiO2 single-layer on sapphire
18
Film thickness (40 nm)
HfO2 single-layer on sapphire
18
Film thickness (27 nm)
Al2O3 / SiO2 mirror on sapphire
18
26 Bi-layers, 2036 nm total thickness
Al2O3 / HfO2 mirror on sapphire
18
14 Bi-layers, 924 nm total thickness
HfO2 / SiO2 mirror on sapphire
18
11 Bi-layers, 768 nm total thickness
Pre-Irradiation Mirror Thickness
Neutron
Irradiation
mirror
Fused
silica
SiC TM
•3 samples of each mirror, monolayer and substrate irradiated to 0.001, 0.01 and 0.1 dpa
•One order higher than Farnum and Vukolov
•Factor of five higher than HAPL M2 mirror
•Irradiation temperature 175-200°C
Post Irradiation Examination and Testing
• Visual inspection.
Signs of delamination, cracking or flaking.
• Measurement of relative specular reflectance.
Perkin Elmer, Lamda 900 photospectrometer, equipped with 6º
relative specular reflectance accessories.
Measurements were made on the dielectric mirrors relative to an
aluminum mirror standard.
• Thermal annealing treatment.
300 and 400ºC, 1.5 hrs with 3ºC/min heating/cooling rate
Vacuum <1x10-6 torr.
Density of bars by density gradient column.
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Visual Inspection of Neutron Irradiated Samples
• Changes in color are observed
with increasing neutron
exposure.

Highest dose samples nearly
opaque to visible light.
controls

Some annealing of color
centers observed following
thermal treatments.
• No visible signs of cracking or
delamination.
• Slight speckling appearing on
some annealed samples: no
correlation between
temperature, dose or material
type.
• Unirradiated controls are all
clear to visible light.
0.001 dpa
0.01 dpa
0.1 dpa
Visual Inspection of Neutron Irradiated Samples
• Compared to Vukolov study,
current material is quite stable
upon post-irradiation thermal
annealing.
controls
0.001 dpa
0.01 dpa
Vukolov
2005
Substrate KS-4V Fused Silica
TiO2/SiO2 layers
0.1 dpa
Fused Silica Bar Samples
3
Percent Change (%)
2
Refractive Index
Primak (1958)
Vitreous Silica
1
Amorphous SiO2,
through gamma or
neutron irradiation
rapidly densifies.
0
-1
Dimensional Change
(this work)
GE-124 Fused Silica
-2
-3
0
2 1023
4 1023
6 1023
8 1023
2
1 1024 1.2 1024
Fast Neutron Fluence (n/m ; E>0.1 MeV)
Annealing will recover
both dimension and
refractive index.
Stress Induced in Dielectric Mirrors
Volumetric Change (%)
3
2
polycrystalline alumina
Pells (1994)
1
0
-1
SiO2
HfO2
GE-124
(this study, 175-200°C)
-2
SiO2
Al2O3
HfO2
Al2O3
sapphire
-3
0
2 10
24
4 10
24
6 10
24
2
8 10
24
1 10
25
Fast Neutron Fluence (n/m ; E>0.1 MeV)
• Sapphire was chosen as a substrate for it relatively stable performance under
irradiation: relatively small swelling, and shallow temperature dependence
• Assumption: by closely matching irradiation-induced dimensional change of substrate
and layers, induced stress will be minimized, increasing lifetime.
• However, we don’t know the irradiation performance of these microcrystalline or
amorphous materials.
Optical Property Changes: HfO2 / SiO2 mirrors
248
•Gradual shift in or peak reflectivity range to lower wavelengths with dose.
•Limited effect at 0.01 dpa.
•Maximum reflectivity measurement may have a systematic error due to use of
limiting aperture (due to small sample) on the normal spot size of the spectrometer.
Optical Property Changes
HfO2 / SiO2 mirrors
Neutron Irradiation Effects
• Slight shift in working range, little
or no reduction in reflectivity.
Annealing Effects
• Shifting of peak reflectivity range
to lower wavelengths occurring
with increasing annealing
temperature.
• Shifting observed in both irradiated
and control materials.
• Spectra of 0.1 dpa irradiated mirror
annealed at 400ºC suggests
QuickTime™
and a
considerableTIFF
damage
to mirror.
(Uncompressed) decompressor
are needed to see this picture.
Optical Property Changes: Al2O3 / SiO2 mirrors
248
• Doses up to 0.01 dpa resulted in a peak shift to slightly higher wavelengths.
• Reflectance spectra exhibits limited change with dose.
Optical Property Changes
Al2O3 / SiO2 mirrors
Irradiation Effects
• Relatively small amount of change
measured of all mirror types.
Annealing Effects
• Annealing resulted in limited
shifting of the peak reflectivity as
compared to other mirror types.
• Differences between 300 and
400ºC annealing diminishes with
increasing dose.
• A significant loss in reflectivity
with 400ºC annealing is possible
(further evaluation underway.)
Optical Property Changes: Al2O3 / HfO2 mirrors
248
• Peak reflectivity of as-received mirrors were off-specifications.
• No significant change in reflectance spectra to 0.01 dpa.
• Lower wavelengths shift observed following irradiation to 0.1 dpa.
Optical Property Changes: Al2O3 / HfO2 mirrors
Neutron Irradiation Effects
• No significant change in
reflectance curves to 0.01
dpa.
Annealing Effects
• Mirror type appears more
stable to thermal anneal than
that of other mirror types
examined.
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Summary
• Samples exposed up to 0.1 dpa with and without thermal annealing at 300 and
400ºC show no signs of delamination or cracking.
• Mirrors show no significant degradation in reflectance up to doses of 0.1 dpa,
with shifts in the peak reflectance curve of up to 10 nm towards lower
wavelengths occurring at higher doses.
• Of the materials studied, the HfO2/SiO2 mirrors show the most sensitivity to
radiation dose and thermal effects despite having the fewest number of film
layers.
• Reflectance spectra of Al2O3 / HfO2 mirrors appear least sensitive to radiation
and combined radiation + annealing.
–Stability and matched behavior of constituent materials, low number of film layers?
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Implication for HAPL
• The initial poor performance and resulting dismissal of dielectric mirrors as
unstable in reactor environments appears unfortunate and misguided.
• The combination of a more stable substrate (sapphire,) combined with higher
quality materials, and the selection of more behavior matched materials under
irradiation appear to have led to more stable materials.
Farnum(95) : glass substrate, glass/ceramic layers, failed by 0.01 dpa
Vukolov (2005) : fused silica substrate glass/ceramic layers, failed by 0.01 dpa
This Work : sapphire substrate, glass/ceramic and ceramic/ceramic, ok to 0.1 dpa
• Results of this work, while preliminary, are encouraging for use of dielectric in HAPL
- HAPL first mirror (assumed dielectric), lifetime dose is 0.03 dpa, appears ok.
- HAPL final mirror (assumed grazing incidence metal mirror), has substantially
flux. Its 2-year dose is approximately 1 dpa. Suggests possibility of dielectric.
Future Work
Optical Testing
•
•
•
Spectrophotometer – further evaluate absolute reflectance of films
through transmission technique (may not be possible on high dose
samples)
Ellipsometry – evaluate film thickness changes in single-layer deposited
samples.
Carry out LIDT of non-irradiated mirrors, if results are acceptable, carry
out irradiated testing.
Structural Characterization
•
•
•
X-ray diffraction analysis.
Cross-sectional transmission electron microscopy.
Raman microprobe – evaluate interfacial strains at interface.
Higher Dose Irradiation
•
•
Complete irradiation to 1 dpa (2 year assumed limit for GIMM) and 3 dpa.
Include representative bulk materials for bulk property measurement.
Silica densified 0.77%
20.02±0.01mm
. The fast flux at the focusing dielectric mi rror M2 is 2.28e18 that gives 7.2e17 n/cm2 per
FPY and 2.8e19 n/cm2 at end-of-lif e. Your results are good since they confirm they are
lif etim e components. In the past we used to assume 1e19 limi t. At the GIMM location we
have 4.4e20 n/cm2 per FPY. If limi t for dielectric is raised above 1e20 maybe we can
think of using dielectric in place of GIMM. For GIMM we assumed 1e21 limi t giving
about 2 FPY lif e
Coating
Working range,
nm
Materials
Layers
PT#
590 – 740
TiO2/SiO2
13
PZ#
590 – 670
ZrO2/SiO2
15
LT#
550 – 650
TiO2/SiO2
17
LZ#
640 – 740
ZrO2/SiO2
23
Label
Substrate:
Silica glass KS-4V, 25 x 2 mm
Manufacturer
“Luch”,
Podolsk,
Russia
“LOGF”,
Lytkarino,
Russia
Neutron Irradiation
Container
Samples
Fluence,
Flux, n/cm2s
n/cm2
Duration,
hours
1
PZ5, PT5, LT5, LZ7
1017
4
7 x 1012
2
PT6, PT7, LT8, LT9
1019
80
3 x 1013
The samples were inserted in two
hermetic aluminium containers which
In result the reflectivity bands of
LT8 and LT9 shifted towards short
wavelengths, the spectra of other
samples remained unchanged.
then were filled with He gas. Irradiation
performed
in
the
water-pool
nuclear reactor IR-8. The fast neutron
fluences
were
measured
by
accompanying iron films of isotope
Fe54 enriched to 99,92%.
Specular reflectance, %
was
100
before
after
80
60
40
20
0
400
500
600
700
800
wavelength, nm
900
1000
Thermal Tests
Heating regimes in Vacuum and Atmosphere
300
Temperature, °C
250
200
150
100
50
Vacuum
Atmosphere
0
0
60
120
180
240
300
Time, min
Specular reflectance, %
100
20°C
64°C
Coatings of all Lytkarino samples were
damaged during heating. Podolsk
samples kept their coatings and optical
properties
80
60
Reflectivity bands of all mirrors
shifted towards short wavelengths in
heated state ~2 nm / 10C. Working
range of Lytkarino samples remained
shifted at 2 nm after getting colder
40
20
0
450
550
650
wavelength, nm
750
850
Expected IFE Mirror Irradiation Environment
Expected doses:
• Total neutron flux to mirror:
~ 2.2x1013 n/cm2s (first mirror) to 1x1011n/cm2s (final mirror)
• Total neutron fluence in IFE in one year, assuming 80 % plant availability:
About 7.2 x 1017n/cm2 per FPY (first mirror), ~0.02 dpa for 30 year lifetime
About 4.4 x 1020 n/cm2 per FPY (final mirror), ~ 1 dpa in 2 years
Effects on mirrors:
• Differences in radiation and thermally induced swelling or contraction of the film
layers.
• Changes in surface roughness.
• Radiation / thermally induced structural changes within a given layer.
• Radiation / thermally induced mixing or formation of interlayer compounds.
• Reduction in peak reflectivity and shift towards lower wavelengths.