Transcript ppt

Progress on GIMM Fabrication & Testing
M. S. Tillack, J. Pulsifer, K. Sequoia
High Average Power Laser Program
Project Meeting
University of Wisconsin – Madison
24–25 September 2003
Background (1):
GIMM design concept
85Þ
The reference mirror concept consists
of stiff, light-weight, radiation-resistant
substrates with a thin metallic coating
optimized for high reflectivity
(Al for UV, S-pol, shallow q)
~50 cm
Background (2):
Key Issues
• Shallow angle stability
• Damage resistance/lifetime
Goal = 5 J/cm2, 108 shots
• Fabrication & optical quality
• Contamination resistance
• Radiation resistance
When last we met...
• Defects on thin-film mirrors were plaguing us.
• Schafer Al coatings on superpolished SiC showed promise,
but pin-point defects and darkening were observed.
• Some of these surfaces operated over long periods of time
after surface changes occurred. Extended damage studies
were planned.
• Overcoating the Al to eliminate oxide effects was considered.
• Monolithic Al mirrors provided good resistance previously.
More testing of polished and diamond-turned Al, as well as
Al-coated Al and novel Al microstructures were considered.
What we’ve done...
• Continued to work with Schafer to improve coatings,
and MER to develop substrates (see posters).
• Resolved the issue of “darkening”:
– Built a new chamber with cryopump.
– While waiting for the new chamber, used He and Ne
backfill to eliminate pump oil decomposition.
• Extended the testing to shot counts up to 100,000.
• Tested more GA diamond-turned Al.
• Obtained and tested electroplated mirrors.
• Started to explore scale-up issues.
Summary of Schafer collaboration
• Source of pin-point defects
identified; defect-free substrates
yielded defect-free coatings.
• Reactive oxidation used to
overcoat Al in-situ.
• Stripping and recoating
successfully demonstrated.
• Scale-up pathway 31550 cm
identified.
mirror #41, s/n 10157-024
50 nm sputter+1 mm e-beam
500 shots at 5 J/cm2
A new vacuum chamber was built
mirror #38, s/n 10157-021
100 nm sputter+2.0 mm e-beam
5.0 J/cm2 for 1000 shots
• Cryopumped for higher purity
• Added flexibility in sample manipulation
• Improved diagnostic access
viewing port
beam diagnostics
dump
cube
1/2 waveplate
specimen
mount
cube
dump
In-situ monitoring helps us identify the
onset of damage
• Brightfield beam profiling
• Darkfield beam profiling
• Surface imaging dump
test specimen
translation
main beam
camera
probe laser
profiler
microscopy
in-situ imaging
darkfield
Testing continues...
• Thin films on superpolished substrates
– CVD SiC, 2-3Å roughness, 2-3 nm flatness over 3 cm
– magnetron sputtering up to 250 nm
– e-beam evaporation up to 2 mm
• Solid polycrystalline metal
– polished
– diamond-turned
• Electroplated and
turned Al
Thin films are delicate, and damage easily
and catastrophically
250 nm e-beam
23,000 shots @4 J/cm2
1.5 mm e-beam
86,000 shots @4 J/cm2
Nevertheless, we are continuing to explore methods to
improve the coating quality and survivability
Diamond-turned Al exhibits superior
damage resistance
•
•
•
•
Exposed for 50,000 shots in He at 3–4 J/cm2
No obvious damage
Minimal (if any) grain boundary separation
Polishing appears to introduce impurities
and pre-stress the grain boundaries, whereas
diamond-turning helps stabilize the surface
polished sample
for comparison
Electroplated Al solves problems with
coating thickness and weak grains
• 50-100 mm Al on Al-6061 substrate
• Grain size ~10 mm
• Survived 100,000 shots at 3-4 J/cm2
• No discernable change to the surface
• The performance, design
•
•
flexibility and scalability
make this our leading
concept
Still need to demonstrate
Al on SiC
Thick e-beam coatings are
another possible option
Damage was obtained finally at 11 J/cm2
•
•
•
Exposed to 78,500 shots at 11 J/cm2
Apparently melted at “micro-scratches” (which are smaller
than diamond lines), probably caused in shipping
Damage resistance should improve if these micro-scratches
can be eliminated
QuickTime™ and a TIFF (Uncompressed) decompressor are needed to see this picture.
QuickTime™ and a TIFF (Uncompressed) decompressor are needed to see this picture.
Optic scale-up: multiplexed beams enable
smaller, more tolerant final optics
drawing courtesy of J. Sethian, NRL
LONG PULSE AMPLIFIER
(~ 100's nsec)
Last Pulse
Demultiplexer
Array
(mirrors)
Multiplexer
Array
(beam
splitters)
Target
FRONT END
( 20 nsec)
Only three pulses shown for clarity
Final optic concept: many advantages to
mirror segmentation and multiplexing
amp 1
amp 2
1. Easier to fabricate
2. Easier to maintain
3. Less variation of laser and
1’ x 2’
neutrons over one optic
1-kJ mirror
4. Beam overlap reduces requirements on both mirror and laser
5. Can be tested on Electra & Mercury
For Reference:
NASA Technology Goals for JWST
James Webb Space Telescope (formerly known as NGST)
Deployment in 2011
7-m diam. lightweight optic
$825M project budget
Goal mirror cost of $300k/m2
Different candidates considered
(Be is prime candidate)
http://ngst.gsfc.nasa.gov
Based on a 1996 Optical Telescope Assembly study, the following
requirements were placed on JWST's optics:
The mirror should be sensitive to 1-5 microns (0.6-30 extended).
It should be diffraction limited to 2 microns.
It will have to operate at 30-60 K.
It should have an areal density of less than 15 kg/m2.
Future Plans
• Choose electroplated Al on R&H CVD SiC as our prime
candidate mirror coating and substrate (for now).
• Continue to develop alternate coatings and substrates.
• Fabricate and test a small batch of electroplated Al on SiC.
• After successful demonstration to 105 shots, place an order
large enough to satisfy all testing (x-ray, ion, neutrons, etc.)
• Fill out damage curves with long-term exposures.
• Scale up (fabricate) mirrors to 500 J (25 W absorbed).
• Install optics testing capability at Electra.
• Perform large-scale tests.
• Perform radiation damage tests (XAPPER, others?)
Acknowledgements and Links
Schafer Corp.
www.schafercorp.com
Rohm and Haas
www.cvdmaterials.com
Alumiplate
www.alumiplate.com
II-VI
www.ii-vi.com
Sigma Technologies
www.sigmalabs.com
MER corporation
www.mercorp.com
Surface Optics
www.surfaceoptics.com
Backup
X-ray dose to the final optic
• Attenuation calculation verified J. Latkowski’s earlier result:
we need a fair bit of gas to protect the optic
Cooling requirements
• Currently:
–
20 mW absorbed power
–
V=5 cc, r=3.2 g/cc, mass ~15 g, Cp~1 J/mol-K,
MW=10 g/mole, C=0.1 J/g-K
–
adiabatic dT/dt=Q/mCp = 0.02/1.5 = 1/75 K/s
• Prototype power plant optic
–
100 W absorbed power
–
r=15 kg/m2, L=0.2 m2, mass ~3 kg, Cp~1 J/mol-K,
MW=10 g/mole, C=0.1 J/g-K
–
adiabatic dT/dt=Q/mCp = 100/300 = 1/3 K/s
Defect-free surfaces are needed for
damage resistance in thin film coatings
Fabrication and handling protocols are under development:
1. Ensure the substrate has no defects
• micrographic and scattered light inspection
2. Clean the substrate adequately before coating
• established cleaning protocols
3. Provide an Al coating that is defect-free
• use clean sputter chambers
4. Ensure that the natural or applied overcoat is defect-free
• explore reactive oxidation, natual oxide, overcoating
5. Ship samples in a clean container
• custom containers?
6. Examine the samples before testing
7. Perform laser cleaning very carefully
• protocol developed, additional optics purchased
Logic Behind Coating Development
1. Al was chosen as the most promising reflector
2. Coatings are desired because pure Al is not an
attractive substrate (mechanical & radiation issues)
3. Thick coatings generally suffer from damage at
grain boundaries and intragrain slip
4. Thin (amorphous) coatings suffer from differential
stress at interface
5. Environmental overcoats are desirable (but
possibly not necessary)
6. Whatever coating we adopt must be scalable