Transcript 2.Cook

A Very Brief and Selective Historical
Overview of ICF Capsule Fabrication
Robert Cook
Consultant for
General Atomics
Presented at
18th High Average Power Laser Program Workshop
Los Alamos, New Mexico
April 9, 2008
VG-1 4/9/08
What does the history of capsule fabrication tell us?
• ICF history is full of examples of seemingly insurmountable problems that
were solved, sometimes by a totally new approach, sometimes by
incremental changes.
• In the ICF capsule area, fabrication and characterization capability have
been closely coupled to design, certainly since the early 90’s.
• A firm theoretical understanding of a process generally leads to solutions,
and it puts realistic limits on what we can expect from a technology.
• Incremental improvements of existing technologies are important since
they cause us to look closely at a process. The understanding that comes
from this examination often is what leads to breakthroughs.
• The story is constantly changing - both the designs and the fabrication
methods.
The current HAPL foam shell development has come
a long way - due to advances in fabrication AND
changes in design - but still has a ways to go.
VG-2 4/9/08
Quick timeline
1972
1980
glass
shells
drop tower
plastic
shells
sorting!!!
Nova
0.5 mm
capsules
1990
2000
decomposable
mandrels
2 mm shells
for NIF
MICROENCAPSULATION
2008
CH, Be, and
polyimide
capsules
droplet generator
size control
HAPL size
R/F foam
shells
foam shells
interfacial
techniques
plasma polymerization
VG-3 4/9/08
X
In the beginning…..
• J. Nuckolls, L. Wood, A. Thiessen, and G. Zimmerman, “Laser Compression
of Matter to Super-High Densities: Thermonuclear (CTR) Applications,”
Nature 239, 139 (1972).
- Only a very rudimentary “Design” initially involving
glass shells
- Shells available commercially - a huge sorting
problem
- Fortunately characterization capability was
limited…,.
- Development of dedicated ICF glass shell towers….
- limited size (not a problem for early ICF
from J. H. Campbell, J. Z. Grens, and J.
F. Poco, “Preparation and Properties
of Hollow Glass Microspheres for Use in
Laser Fusion Experiments,” LLNL Report,
1983.
experiments)
- Size problem largely solved many years later with a
totally new glass shell production method….
- M. L. Hoppe, “Large Glass Shells from GDP Shells,”
Fusion Technol. 38, 42 (2000).
VG-4 4/9/08
In the early 80’s the designers found that lower Z - plastic
shells - were better
• The science for thin wall plastic shell fabrication via spray drying existed.
- Drop tower methods were perfected for ICF capsule mandrels.
- Still required lots of sorting….
- Size limitation maybe up to 1 mm, most work done at 0.5 mm (Nova Scale).
• With the development of surface characterization (SphereMapper - mid 90’s)
we found that shells were in fact quite smooth (good thing!) - and the game
changed…
300 0
R(q) equatorial traces
200 0
100 0
100 0
Capsule
nm
0
-100 0
Vacuum
chuck
AFM
AFM
-200 0
-300 0
0
90
18 0
27 0
36 0
Ak2 = power (nm 2 )
AFM
tip
10 0
10
1
0. 1
0.0 1
0.00 1
Air bearing
rotary stage
VG-5 4/9/08
R(q) = Akexp(-ikq)
1
10
10 0
100 0
k = mode number
Nova indirect drive capsules required thick (~50 µm) ablators.
70 mTorr
H 2 + C 4H 8
• Plasma polymerization techniques developed from
the late 70’s to current times allowed for the
deposition of a thick smooth ablator.
S. A. Letts, D. W. Myers, and L. A. Witt, “Ultrasmooth Plasma
Polymerized Coatings for Laser Fusion Targets,” J. Vac. Sci.
Technol. 19, 739 (1981).
G. W. Collins, S. A. Letts, E. M. Fearon, R. L McEachern, and T.
P. Bernat, “Surface Roughness Scaling of Plasma Polymer
Films,” Phys. Rev. Lett. 73, 708 (1994).
K. C. Chen, Y. T. Lee, H. Huang, J. B. Gibson, A. Nikroo, M. A.
Johnson, and E. Mapoles, “Reduction of Isolated Defects on
Ge Doped CH Capsules to Below Ignition Specifications,”
Fusion Sci. Technol. 51, 593 (2007).
• Our understanding of the deposition - and the
conditions that result in very smooth coatings on
well defined substrates - is quite advanced.
VG-6 4/9/08
H
.
-
H
e
.
Glass
tube
•
3
eCH
e
.
H
eH
C
3H
•
5
40 MHz
RF
.
H
.
Microshell
targets
Bouncer
The jump to NIF shells offered quite a few challenges.
• NIF capsules were 4 times larger than Nova capsules - drop tower methods for
mandrels simply didn’t work. Several approaches were explored.
• Microencapsulation was found to be the best answer.
- An old industrial technique, but developed by the Japanese for ICF.
a) µ-encapsulation
of PaMS solution
b) aqueous bath cure loss of solvent
c) shell hardening
aqueous core
removal in
vacuum
PaMS Shell
Microencapsulation is the basis for all capsule
targets in both the ICF and IFE programs
VG-7 4/9/08
How was the first surface specification arrived at for
NIF capsules?
10
• In 1999 I analyzed all of the power
spectra data that had been
acquired on Nova (0.5 mm)
capsules used in experiments.
10
10
10
• The initial design surface finish
requirement for NIF shells (2.0
mm) was based on best Nova
shells ever shot.
nm
VG-8 4/9/08
Nova Capsules
average power spectra for
capsules grouped by rms
over modes 10-1000
4
3
<20-30 nm rms>
8 capsules
2
2
10
10
10
• It was a serious question whether
this could be achieved on the
larger NIF shells, particularly at
lower modes.
5
10
<15-20 nm rms>
11 capsules
1
<10-15 nm rms>
18 capsules
0
<5-10 nm rms>
14 capsules
-1
NIFDesign
Design
NIF
(rms=9.1nm)
nm)
(rms=9.1
-2
2
4
6 8
2
10
4
mode
6
8
100
2
4
6
8
1000
The early 2-mm microencapsulated shells had problems
throughout the power spectrum.
• Mode 2 asymmetry > 5 µm (OOR ~ rn/
n = 2 or 3) Solved by increasing the
interfacial tension  with PAA (Takagi).
5
10
4
10
Mode 2
asymmetry
3
2
power (nm )
10
• Mid mode problems were thought to be
due to convection cells in the rapidly
curing shell wall. Solved by significantly
slowing down the curing rate (McQuillan).
Mid mode
bump
2
10
1
10
Before
After
0
10
-1
10
NIF
design
-2
10
• High frequency roughness was due to surface
debris from the curing bath. Solved by an
improved washing technique.
-3
10
High frequency
roughness
10
100
1000
mode number
The first two of these were NOT incremental changes but were
motivated by a theoretical understanding of relevant parameters.
VG-9 4/9/08
Early concern about vacuoles and wall thickness variations in
microencapsulated PaMS shells led to the development of
the “Decomposable Mandrel” technique.
PaMS
mandrel
300 °C
Decompose and
remove PaMS
mandrel
Plasma polymer
coater
Plasma
polymer
Completed plastic
shell
• This is the reason that the NC for NIF capsules is negligible.
• This was a key development and later was essential in the development of
both polyimide and Be shells because the plasma polymer shell is
thermally stable.
VG-10 4/9/08
The current NIF Ge-doped CH shells have very stringent specs.
• ~2.2 mm diameter
• 165 µm thick ablator with 4 different
Ge-doped layers
• Outer surface specs are very tight:
5
10
4
10
Plastic Shell
Surface Specifications
2
power (nm )
3
10
2
10
full
thickness
shells
1
10
0
10
modes 13-1000
rms = 13 nm
-1
10
mandrels
rms = 7 nm
-2
10
10
100
mode
VG-11 4/9/08
1000
Fill tube 10 µm OD
Laser drilled fill hole 5 µm
The NIF capsules meet the surface specs...
5
10
…and all other specs as well!
4
10
3
2
power (nm )
10
surface specification for
Ge-doped NIF capsules
2
10
power spectra
for 9 shells from
one batch
1
10
0
10
-1
10
representative
mandrels
-2
10
-3
10
2
3
4
5
6 7 8 9
10
2
3
4
5
6 7 8 9
100
2
3
4
5
6 7 8 9
1000
mode number
Meeting this goal is the result of 35 years of work, a number of key new
technologies as well as often slow incremental improvements to existing
technologies, and a good interaction between fabricators and designers.
VG-12 4/9/08
The history of foam shells is shorter.
• Suggested first by Sachs and Darling (1987) to symmetrize the fuel - at that
time they were thinking about liquid DT!
• First foam shells fabricated by Takagi, et al. in Japan in early 90’s. He used a
methacrylate chemistry, resulting in ~1-2 µm cells and thus opaque shells.
- The basic technique for the foam shell formation was microencapsulation
- An interfacial technique was developed to apply a solid outer layer to the shells
• Foam shell fabrication has been extended to a number of different chemistries.
• At LLNL there had been a great deal of work on resorcinol/formaldehyde(R/F)
low density foams. In the mid 90’s LLNL produced the first R/F foam shells.
• GA now fabricates ~1 mm R/F foam shells for Cryo experiments at Rochester.
• Work in 2006 indicated that the formation of an outer, gas tight coating by
interfacial techniques on HAPL shells was problematic.
- Subsequent solvent exchange steps created huge osmotic pressures
Current effort is to coat HAPL size dry R/F foam shells with GDP
VG-13 4/9/08
What are the important differences in the fabrication of
NIF capsules and HAPL foam shells?
• The R/F foam shell microencapsulation system is inverted relative to the NIF
mandrel system (and the DVB foam shell system).
- The bath additives controlling the interfacial tension are poorer (currently).
PaMS/oil
solution
outer aq bath
outer oil bath
inner aq
droplet
inner oil
droplet
PaMS mandrels
R/F foam shells
R/F aq
solution
• HAPL shells are twice as large, low mode symmetry will be worse.
• The microencapsulated wall is much thicker for HAPL foam shells, making
control of NC more difficult.
• The GDP coating on NIF shells is on a very smooth substrate - the foam
surface is much rougher.
VG-14 4/9/08
What are the biggest near-term challenges for meeting the
HAPL foam shell specifications?
• Out-of-Round, low mode symmetry.
- Controlled during the microencapsulation process by the interfacial tension, ,
between the aqueous R/F droplet and the oil bath.
- Current results are consistent with OOR ~ r2/.
- The primary handle is  - current experiments are focusing on this.
• Non-concentricity problems.
- May be helped when OOR is reduced.
- New dielectrophoresis (DEP) technology is intended to address the problem.
• High mode surface finish.
- Experience is that a GDP coating mimics the substrate, and since the substrate is
foam, not a smooth solid shell, a rough coating surface will result.
- However recent experiments at higher coating pressure suggest something new
is happening to the foam surface - our ability to find out how it works is critical.
• Gas tightness and solid layer thickness.
- Optimization of the GDP coating process should address these specifications.
VG-15 4/9/08
Summary
The history of ICF is filled with many examples of
“insurmountable” problems that have been solved by a
combination of new fabrication methods and steady
improvements in existing technologies.
HAPL foam shell development has built effectively on the
existing technology, and good progress has been made
due to advances in fabrication AND changes in design.
The remaining challenges will yield to new ideas
and increased understanding of our technology.
VG-16 4/9/08