Advanced Chamber Concept with Magnetic Intervention: - Ion Dump with Phase Change (including Cu, Pb-17Li, Flibe) - SiCf/SiC Pb-17Li or Flibe Blanket

1
Advanced Chamber Concept with Magnetic
Intervention- Ion Dump with Phase Change
(including Cu, Pb-17Li, Flibe) - SiCf/SiC
Pb-17Li or Flibe Blanket

• A. René Raffray
• UCSD
• With contributions from
• M. Sawan, G. Sviatoslavsky, I. Sviatoslavsky
• UW
• HAPL Meeting
• PPPL, Princeton, NJ
• December 12-13, 2006

2
Outline

• Possible options for large chamber in case W
  armor does not work include (follow-up from last
  meeting)
• Allowing melt layer momentary liquid wall (Look
  at possibility of Cu in addition to W and example
  be case presented before)
• Moving solid wall (L. Snead)
• Engineered W armor samples provided by PPI to UNC
  ORNL to be tested provision of samples for
  additional testing elsewhere to follow.
• Separate ion dump chamber for magnetic
  intervention case
• Possibility of solid armor with phase change (W,
  Be, Cu)
• Possibility of wetted walls (Pb, Pb-17Li, flibe)
• SiCf/SiC blanket for magnetic intervention case
• Pb-17Li and flibe (see also posters)
• In the process of evolving chamber core of MI
  case with a number of other contributors

3
Momentary Liquid Walls (allowing solid to melt
and resolidify)

• Allowing W armor itself to melt is an option
  but concerns about stability of melt layer and
  integrity of high temperature solid W under melt
  layer
• Other possibility is to use a lower MP material
  in a W structure
• - e.g. gt90Cu in lt10 W structure
• - How to fabricate it?
• - Structure size to provide good melt layer
  retention through capillarity (microstructure
  size to be optimized for melt layer retention
  and integrity)

4
Histories of Temperature and Phase Change
Thickness for a Cu Armor as a Function of the
Chamber Sizes for the 350 MJ Target
1-mm Cu on 3.5 mm FS at 580 C No chamber
gas Can the W mesh be maintained at a
reasonable temperature acceptable
lifetime? (1250C for 10.75 m
chamber) Stability of 3-10 ?m melt layer of
Cu Minimal evaporation, 0.0001 nm on average
per shot for 10.75 m chamber, 1 ?g per shot
5
Magnetic Intervention Utilizing a Cusp Field to
Create a Magnetic Bottle Preventing the Ions from
Reaching the Wall and Guiding them to Specific
Locations at the Equator and Ends

• Utilization of a cusp field for such magnetic
  diversion has been experimentally demonstrated
  previously
• - 1980 paper by R.E. Pechacek et al.,
• Following the micro-explosion, the ions would
  compress the field against the chamber wall, the
  latter conserving the flux. Because of this flux
  conservation, the energetic ions would never get
  to the wall.
• One possibility would be to dissipate the
  magnetic energy resistively in the FW/blanket,
  which reduces the energy available to recompress
  the plasma and reduces the load on the external
  dumps
• - about 70 of ion energy dissipated in blanket
• - about 30 of ion energy in dump region

6
Seems Advantageous to Position Dump Plate In
Separate Smaller Chamber
Dry wall main chamber to satisfy target and
laser requirements. Separate phase-change dry
wall or wetted wall chamber to accommodate ions
and provide long life. Have to make sure no
unacceptable contamination of main chamber
7
Scoping Analysis of an Example Ion Dump Ring
Chamber
Some flexibility in setting chamber major and
minor radii so as not to interfere with laser
beams e.g., with Rmajor/Rminor 8/2.7 or 9/2.4
m, and assuming 35 of wetted wall area sees ion
flux with a peaking factor of 1 - Ion dump area
300 m2 - From 0 to 0.5 ?s, q 4.51x1010
W/m2 - From 0.5 to 1.5 ?s, q 6.53x1010 W/m2

• Dry Wall Armor with Phase Change
• - Example results for W and Be previously
  presented.
• - New case analyzed with Cu, possibly within
  high porosity W microstructure (80-90) for
  integrity and Cu melt layer retention
• Wetted Wall
• - Example results for Pb previously presented
• - New cases with Pb-17Li and flibe analyzed

8
Temperature and Phase Change Thickness Histories
for W, Be, Cu, Pb, Pb-17Li and Flibe for Example
Case
350 MJ target (ion energy 87.8 MJ) Ion dump
area 300 m2 From 0 to 0.5 ?s, q 4.51x1010
W/m2 (7.7 of ion energy) From 0.5 to 1.5 ?s,
q 6.53x1010 W/m2 (22.3 of ion energy)
9
Maximum Temperature and Phase Change Thicknesses
for W, Be, Cu, Pb, Pb-17Li and Flibe as a
Function of Ion Dump Area
350 MJ target (ion energy 87.8
MJ) Evaporation loss per shot relatively modest
for W but could be a concern for Cu or Be (1
nm/shot 0.43 mm/day) Stability of melt layer
is a concern (10?m for Cu or Be 1 ?m for
W) For wetted wall in particular, the
evaporated material (10 ?m for Pb-17Li, Pb or
flibe) must recondense within a shot and not
contaminate main chamber
10
Wetted-Wall Concept Could Consist of a Porous
Mesh Through Which Liquid (Pb-17Li or flibe)
Oozes to Form a Protective Film
Need to make sure that protective film is
reformed prior to each shot - radial flow
through porous mesh - circumferential flow of
recondensed liquid - no concern about any
droplets falling in chamber
11
Film Condensation in Ion Dump Chamber for
Pb-17Li and Flibe
Scoping calculations previously done for Pb as
example. Now extended to Pb-17Li and Flibe as
they are used as breeder/coolant in the
blanket. Ion energy from 350 MJ target
87.8 MJ - 7.7 of ion energy to dump over
0-0.5 ?s - 22.3 of ion energy over 0.5-1.5
?s Evaporated thickness and vapor
temperature rise from ion energy deposition
in ion dump chamber. Assume ion deposition area
300 m2 - e.g. 35 of chamber with Rmajor 9 m
and Rminor 2.4 m
jevap
Pg Tg
Tf
jcond
jnet net condensation flux (kg/m2-s) M
molecular weight (kg/kmol) R Universal gas
constant (J/kmol-K) G correction factor for
vapor velocity towards film sc, se
condensation and evaporation coefficients Pg, Tg
vapor pressure (Pa) and temperature (K) Pf, Tf
saturation pressure (Pa) and temperature (K) of
film
12
Scoping Analysis of Pb-17Li Condensation in
Example Ring Chamber
Characteristic condensation time very fast, lt
0.024 s It takes lt 0.24 s for vapor density to
reach saturation for final vapor temperature gt
773 K (assuming linear temporal decrease of
vapor temperature from initial to final value).
13
Scoping Analysis of Flibe Condensation in Example
Ring Chamber
Characteristic condensation time very fast, lt
0.02 s It takes lt 0.202 s is for vapor density
to reach saturation for final vapor temperature gt
773 K (assuming linear temporal decrease of vapor
temperature from initial to final value).
14
Blanket Study for Magnetic Intervention Chamber

• More detailed study of blanket using SiCf/SiC
  Pb-17Li or Flibe
• - Layout and thermal-hydraulics
• - Neutronics
• - Fabrication
• - Assembly and maintenance

(presented by M. Sawan)
(presented by G. Sviatoslavsky)
15
Conical Chamber Well Suited to Cusp Coil Geometry
and Utilizing SiCf/SiC for Resistive Dissipation
Armored ion dumps - designed for easier
replacement than blanket - shown inside
the chamber but could also be in separate ring
chamber SiCf/SiC blanket with liquid
breeder - TBR 1.3 for Pb-17Li
Water-cooled steel shield / vacuum vessel (0.5m
thick) is lifetime component and protects the
coil. Design accommodates laser ports.
Example Chamber Layout
Maintenance performed from the top by removing
the upper shield and the blanket modules from the
different region without having to move the coils.
16
Self-Cooled Pb-17Li SiCf/SiC Blanket Optimized
for High Cycle Efficiency
Simple annular submodule design builds on
ARIES-AT concept Pb-17Li flows in two-pass
first pass through the annular channel to
cool the structure and a slow second pass
through the large inner channel where the
Pb-17Li is superheated This allows for
decoupling of the outlet Pb-17Li temperature
from the maximum SiCf/SiC temperature limit
17
Submodule Configuration for Upper Mid-Blanket
Region
Submodule cross-section changes because of
conical geometry Pb-17Li enters through annular
channel at equator (C-C), turns at top (A-A),
flows through inner channel and exits at
A-A. 5 submodules joined (e.g. by brazing) to
form a modular unit for assembly and
maintenance Tight fit assembly so that all
submodules are pressure-balanced by adjacent
modules to avoid large stresses associated with
long radial span (particularly at A-A)
radial/toroidal dimensions A-A 1.06/0.196 m
B-B 0.88/0.33 m C-C 0.7/0.47 m
18
Submodules Shaped at Module End for Tight Fit
Assembly (Resistive Dissipation and
Pressure-Balancing of All Submodules)
Concerns exist about the possible domino effect
on all submodules in case of a catastrophic
failure of a submodule. Possible solutions
include isolating a limited number of modules by
including structurally independent wedges and/or
using pressure-sensitive valve system to drain
and decompress the coolant in such an accident
case.
Module A
Module C
Module B
19
2-D Stress Analysis of First Wall Performed with
ANSYS
At B-B, maximum heat loads q0.11 MW/m2
qSiC 31 MW/m3 Pb-17Li pressure is 1 MPa
(accounting for hydrostatic pressure 0.74 MPa
for 9 m elevation, ?Pblkt 0.2 MPa and some
margin). ?tot increases sharply as the wall
thickness is increased, indicating the dominating
effect of the increasing ?thermal over the
decreasing ?pressure. For the present scoping
design analysis, it seems reasonable to choose
?FW 5 mm the corresponding ?tot 100 MPa for
plane stress and 230 MPa for plane strain ,
compared to an assumed limit of 190 MPa for
SiCf/SiC. If more margin is needed in the
future, a slightly thinner wall of larger chamber
could be used.
Chamber dimension 6m
20
Possible Submodule Fabrication Method(rectangular
submodules shown for illustration)
Issue Complex concentric walls prevent assembly
of inner and outer channels
Solution Expendable core form fabrication
1. inner channel form
2. Lay-up infiltrate inner channel
3. Two-piece form fitted over inner channel
4. Lay-up infiltrate outer channel
6. Braze end caps
5. Consume both forms via chemical or thermal
process
7. Braze 5 submodules together to form module
21
Self-Cooled Pb-17Li SiCf/SiC Blanket Coupled
to a Brayton Cycle though a Pb-17Li/He HX
3 Compressor stages (with 2 intercoolers) 1
turbine stage DP/P0.05 1.5 lt rplt 3.5

• - DTHX 30C
• - hcomp 0.89
• - hturb 0.93
• - Effect.recup 0.95

22
Thermal-Hydraulic Optimization Procedure
Set blanket design parameters. - SiCf/SiC
?FW0.5 cm ?annulus0.5 cm - only the blanket
length is adjusted based on the chamber
size Simple MHD assumption based on assumed 1 T
field and flow laminarization with conduction
only (probably conservative).
For given chamber size and fusion power,
calculate combination of inlet and outlet Pb-17Li
temperatures that would maximize the cycle
efficiency for given SiCf/SiC temperature limit
and/or Pb-17Li/SiC interface temperature
limit. - SiCf/SiC Tmaxlt1000C - Pb-17Li/SiC
Tmaxlt950C - Assume conservatively
k15 W/m-K for SiCf/SiC
23
Brayton Cycle Efficiency as a Function of
Cone-Shaped Chamber Size and Corresponding Outlet
and Inlet Pb-17Li Temperatures
Pb-17Li/SiC Tmax lt 950C is more constraining
than SiCf/SiC Tmax lt1000C Both ?P and Ppump
show minima at a chamber dimension of 6 m
corresponding to the largest ?T between Pb-Li
inlet and outlet temperatures (and lowest flow
rate). For a 6 m chamber, Pb-17Li Tout1125C
?Brayton 0.59 Such a high-temperature also
allows for the possibility of H2
production Question about whether such a high
Pb-17Li Tout can be handled in out of reactor
annular piping and in heat exchanger.
24
Effect of Varying the Pb-17Li/SiC Interface
Temperature Limit
It is not clear what the allowable SiC/Pb-17Li
Tmax really is as it depends on a number of
conditions. Earlier experimental results at
ISPRA indicated no compatibility problems at
800C, whereas more recent results indicate a
higher limit. Decreasing the SiC/Pb-17Li Tmax
from 950C to 800C results in a marked reduction
in cycle efficiency, from 59 to 50.
Interestingly, the pressure drop and pumping
power minima correspond to an interface limit of
950C, and both increase significantly as the
interface temperature limit is decreased and an
increased in flow rate is required.
25
Adapting the Blanket for Flibe Requires a Be
region for Tritium Breeding
Flibe electrical resistivity well suited for
resistive dissipation of magnetic energy 1-1.5
cm Be region sufficient for TBR1.1 A Be plate
can be included in the previous submodule
design Be also used for chemistry control of
flibe The flibe flows in two-pass a first pass
through the annular channel to cool the
structure and a slow second pass through the
large inner channel where the flibe is superheated
26
Effect of Varying the Flibe/SiC Interface
Temperature Limit
Flibe Tout lt Pb-17Li Tout mostly because of
its poorer heat transfer properties For 6 m
conical chamber and 1000C limit Flibe
Tin/Tout 673/1000C Be Tmax 840 C ?P
0.16 MPa Ppump 0.27 MW ?Brayton 0.57
High temperature also allows for possibility
of H2 production Lower density of flibe
results in lower primary stresses in
module (design pressure 0.4-0.5 MPa compared
to 1 MPa for Pb-17Li)
6 m chamber SiCf/SiC Tmax lt 1000C
27
Summary

• Scoping study of self-cooled Pb-17Li or flibe
  SiCf/SiC blanket concept for use in the
  magnetic-intervention cone-shaped chamber
  geometry performed
• Separate dump chamber with melted solid wall
  (W, Cu, Be) or wetted wall (Pb-17Li, flibe, Pb)
  assessed for magnetic intervention case
• - Much relaxed atmosphere requirements for
  separate dump chamber
• - Encouraging results as condensation is very
  fast
• - Need to ensure no unwanted contaminants in
  main chamber
• Future work
• - Complete overall chamber layout of MI core
• - More detailed design of separate dump chamber

28
Summary of Blanket Study for MI Case

• A scoping design analysis has been performed of
  a self-cooled Pb-17Li SiCf/SiC blanket concept
  for use in the magnetic-intervention cone-shaped
  chamber geometry
• - Simple geometry with ease of draining and
  accommodation of 40 rectangular laser
  ports with vertical aspect ratio
• - Good performance, with the possibility of a
  cycle efficiency gt50 depending on chamber size
  and SiCf/SiC properties and temperature limits
• - Submodule side walls are pressure-balanced
  only the first wall and back wall are
  designed to accommodate the loads
• - Must be noted that SiCf/SiC is an advanced
  material requiring substantially more RD than
  more conventional structural material (e.g. FS)
• - Submodule design can be adapted to flibe as
  breeder by adding a layer of Be to ensure a TBR
  of 1.1 and provide for chemistry control
• - The high coolant temperatures result in high
  cycle efficiency and could also be used for H2
  production
• - However, issues of what outside coolant tube
  and HX material(s) to use at these
  temperatures need to be further investigated