Fabrication of Capsules with Angle Dependent Gold shims for Hohlraum Drive Symmetry Correction

1
Fabrication of Capsules with Angle Dependent Gold
shims for Hohlraum Drive Symmetry Correction
15th International Symposium on Heavy Ion
Fusion PPPL Princeton, New Jersey June 7-11, 2004
Work performed under General Atomics Internal RD
program
2
Innovation in target design and fabrication can
be used to compensate for drive asymmetry

• In the HIF hybrid target, most of the energy is
  deposited behind a shine shield
• Radiation flows around the shield and results in
  a bright source around the shield
• This causes a large P4 asymmetry
• The P4 asymmetry can be corrected using a shim to
  remove excess radiation
• Initial examination of this concept was done on
  a target shot in SNLs double ended Z-pinch

Z-axis
Au shim
Ge-CH
equator XY
Shims can be used to fix asymmetries in all
indirect drive targets (heavy ion, laser,
z-pinch)
3
Summary of angle dependent gold shim fabrication
process

• Ge-CH mandrels were made using well developed
  ICF capsule fabrication techniques
• Gold shim coating was deposited by magnetron
  sputtering
• Using masks patterned coatings were produced
• Coatings on flats were used as initial
  guidelines
• Gold thicknesses on shells were determined by
  x-ray transmission (contact radiography)
• Profile similar to desired profile was produced
• Adjustments to process is necessary to obtain
  desired profile

4
The decomposable mandrel technique was used to
fabricate Ge-CH shells

• Process allows fabrication of shells with
• Desired diameter ( 5 mm)
• Wall thickness ( 20-30 µm)
• Sphericity (gt99.9)
• Wall thickness uniformity (lt0.2 µm)
• Dopant content (germanium- 2 at.)

High aspect ratio- fragile
5
The desired gold shim profile was a combination
of P2 and P4 Legendre polynomials

• The relative contribution of P2 and P4 needed to
  be an adjustable parameter
• Shim coating process had to be flexible enough
  to accommodate such changes
• Use of multiple masks
• Adjustment of mask-shell distance

XZ Projection vs. polar angle
Designed to zero out P2, P4 double ended Z-pinch
asymmetry
6
Gold coating was deposited using physical vapor
deposition via magnetron sputtering

• Sputtering provides an important processing knob
• Background argon pressure determines mean free
  path of gold atoms
• Affects scattering and deposition pattern of
  gold atoms
• Masks were used to pattern gold coating

Sputter system
7
Combination of coatings through two different
masks could be used to obtain different profiles

• Aluminum plate with precision milled slots was
  used as mask
• Varying slot dimension varied coating profile
• Single and double slot masks could be combined
  to fine tune pattern

Combination mask
Coatings on flat substrates
8
Gold coating pattern could also be changed by
varying the mask-substrate distance
Movable mask

• For a given mask the separation between mask and
  substrate could be changed
• Coatings spread with separation of mask from
  substrate
• This allowed obtaining various profiles by
  adjusting mask-substrate distance

1mm
2mm
Coatings on flats
9
The desired profile could be produced on flats
very closely

• Coatings using
• 2mm mask-substrate distance
• 0.8 single slot 1.0 double slot
• produced desired pattern on flats

• Coatings on shells were likely to be deficient
  near the poles
• This, however, was used as a starting point for
  producing the shim on shells
• Coating rate on shells needed to be determined

10
Characterization of coatings on shells provided a
difficult challenge

• Interferometry or profilometry were not possible
• X-ray transmission detected by contact
  radiography was readily available and used
• Convolution of x-ray source, coating, x-ray film
  signal was calibrated on flats
• X-ray signals was linear for the range of
  interest
• Uncoated shell signal was subtracted
• Shell signal was corrected for cord length in ZX
  projection
• Data near poles suffers from saturation due to
  large cord length

11
Coatings on flat substrates were used to
calibrate x-ray transmission signal

• Coatings through single and double slots on
  flats were measured by interferometry
• Coatings were duplicated on thin CH film ( zero
  substrate signal)
• X-ray signals were recorded using various tube
  voltage and exposure times
• The settings that led to highest x-ray vs.
  interferometry linearity were used for measuring
  shells
• It also allowed thickness determination on shells

Single mask coating
8 bit digitized line out
12
X-ray signal was linear in most of the range of
interest
ZX profile cord length of desired profile

• X-ray signal measures projection of shells
  thickness on ZX plane
• Cord length for desired profile is 0.5-1.2 µm
• X-ray signal is linear vs. thickness for this
  range
• Proper exposure and source voltage settings are
  required!
• Region of linearity could be changed by changing
  settings

13
Shells were precisely positioned under the masks

• Shells were mounted on 200 µm tungsten stalks
• Stalk was attached to motor shaft with wobble of
  lt 50 µm
• Interferometer was used to determine position of
  top of shell to lt 1 µm
• Measuring microscope was used for lateral
  position determination (1 µm)
• XYZ micrometer stage was used to move mask over
  shells to within 2 µm

Tungsten stalk
Shell mounting for coating
14
X-ray signal from shells coated through single
mask was used to determine coating thickness
Uncoated shell
ZX thickness projection
Au coated shell

• Single slot pattern on shell was very similar to
  that on flats
• X-ray exposure settings were changed for these
  thinner coatings

15
X-ray signals from double slot mask indicated
lower rate near equator
Au coated shell

• Coatings using double slot mask were thicker
  near equator and thinner at the poles than
  expected from flat data
• Shell curvature and larger distance between
  poles and mask are responsible
• Data analyzed after shells were delivered for
  shots
• We will make adjustments for next series

ZX thickness projection
16
A profile similar to the desired profile was
produced on target shells

• Final profile was not exactly as desired on
  shells delivered for shots
• Coating near equator was thicker than desired-
  would it reverse P2 asymmetry?
• Coating near poles was thinner
• Adjustment of the double slot mask is needed
• Shells were assembled and shot at SNL

P2 symmetry was reversed in shot on SNL Z-pinch
as expected
17
We hope to simulate process using Monte Carlo
simulation to obtain better predictive capability

• Direct Monte Carlo simulation program DS2V of G.
  Bird
• Initially determine gold atom distribution near
  masks
• Use that distribution to simulate coating under
  masks as function of distance on flat and
  spherical substrates

18
We have produced angle dependent gold shim targets

• This innovative target can compensate for driver
  asymmetry
• Targets were produced using ICF target
  fabrication infrastructure at General Atomics
• Gold coating thickness could be patterned by
  coating shells through
• Various masks
• Adjusting mask-substrate distance
• The desired pattern was nearly produced
• However, adjustments are needed to faithfully
  produce desired profile
• Other characterization techniques also need to
  be examined in the future