Updated Shielding Analysis
for Final Focus Magnets
by
Jeff Latkowski
ARIES Meeting
July 1-2, 2002
Work performed under the auspices of the U. S. Department of Energy by Lawrence Livermore National Laboratory under Contract W-7405-Eng-48.
Outline
• Overview of the problem:
– General requirements
– Why is this an issue now?
• Review of work from last several years
• Update for new point design
• Outstanding issues/future work
JFL—7/02 ARIES
General requirements for the final
focusing magnet shielding design
• Shielding design consistent with target beam requirements
(e.g., half-angle of array <24º)
• Magnets don’t quench on a per-shot basis (limits energy
deposition to ~100 mJ/cc/shot)
• Recirculating power for magnet cooling doesn’t cause
unacceptably large economic hit
• Magnets have reasonable radiation damage lifetime
• Avoid generation of above Class C waste
JFL—7/02 ARIES
What limits the radiation lifetime
of the final focusing magnets?
• Total dose to the insulators is limited to 100 MGy (ref. 1);
tends to be dominated (90-95%) by contribution from
gamma-rays
Inorganic insulators that may be usable with
high-temperature superconductors (HTS) offer
significantly (~103) greater limits (ref. 2)
• “Conservative limit” for the fast neutron fluence (En  0.1
MeV) in Nb3Sn superconductor is 1019 n/cm2 (ref. 1):
– Room-temperature anneal required after 3  1018 n/cm2
– 70% recovery assumed
HTS appear to have fast neutron
References:
fluence limits that are at least as
(1) Sawan and Walstrom
good as that for Nb3Sn (ref. 2)
(2) Bromberg, ARIES meeting (Jan. 2002)
JFL—7/02 ARIES
Why is this an issue when Osiris,
HIBALL, etc. had no problem?
• Previous studies had only 12-20 beams:
– Real estate for shielding was relatively cheap
– These designs devoted 30-40 cm for shielding inside the bore of
each magnet
• We now know that the cost of the accelerator pushes the
design towards many beams (100+):
–
–
–
–
Need roughly same size half-angle for the array
Available space for shielding falls as 1/Nbeam
Effectiveness of shielding goes as exp(-t/l)
Increase from 12  100 beams per side results in room for only ~5
cm of shielding per beam
– exp[-(40-5)/l] ~ 0.03 for l=10 cm  superconductor and
insulation exposed to doses & fluences that are higher by >30
JFL—7/02 ARIES
Protection of the final focusing magnets has
been an active area of research for ~4 years
• HIF Symposium results (3/99) showed that magnet lifetimes were only
~ 1 full-power-year—clearly this is not acceptable.
• IFSA-99 (9/99) results extended lifetimes to ~3 years, but showed that
recirculating power for cooling might be as high as 10% and
superconductor activation was a problem.
• 14th TOFE (10/00) results included investigation of shielding features
and effects such as: “cross-talk,” shielding thickness & composition,
beam clearances, focusing length & frontal shielding, “egg-crates,”
angle-of-attack to the target, tapered shielding, and 3-D effects. Made
significant progress in extending the lifetime to >30 FPY for a selfconsistent design provided by Wayne Meier’s IBEAM code.
• Recently, the thick-liquid HIF community has starting putting together
a new, self-consistent point design. Several shielding analyses have
been completed.
JFL—7/02 ARIES
We observed a strong peaking of the fast
neutron fluence at the center of the array
 Additional calculations showed
that cross-talk is due to
scattering between neighboring
penetrations (7 effect) rather
than scattering between
neighboring magnets (only a
30% effect)
Annual fast neutron fluence
vs. Position in the magnet array
Fast neutron fluence (n/cm2-y)
 Suggested that “cross-talk”
between neighboring
penetrations was very important
2.50E+19
2.00E+19
1.50E+19
1.00E+19
5.00E+18
0.00E+00
1
2
3
4
3
2
5
6
JFL—7/02 ARIES
1
4
5
6
Cut-away view of the HYLIFE-II
chamber shows inner Flibe pocket, first
wall and blanket, and array of magnets
for 288-beam base case (Mar. 1999).
Lifetime ~1.5 years
Lifetime ~40 years
72-beam final focus design for HYLIFE-II including
composite shielding design, slab jet arrays for protective flibe
pocket, and 3-D aspects of chamber/blanket (Oct. 2000).
Schematic of a beamline and definition of
terms used in magnet shielding calculations
1stwall,
blanket,
vac. vess.
Flibe
cross
jets
Beam profile
target
JFL—7/02 ARIES
Beam-to-flibe
clearance
Beam-to-structure
clearance
CL
The beam-to-structure clearance
is an important parameter
• Beam-to-structure clearance reduced from 5 mm to 1 mm:
• Beam-to-Flibe clearance held constant at 5 mm
• Total dose fell by 2.3; fast fluence fell by 2.2
• Since previous results showed ~2.5 reductions with zero
for both clearances, suggests that beam-to-Flibe clearance
isn’t very important
• Can we generate structures and/or liquids carefully enough
to shield within 1 mm of the beamline? Probably with
vortices
Due to the high-precision that is
possible, vortices are an attractive
option for beamline shielding
JFL—7/02 ARIES
Chamber Beam Map
design will further
28 Foot / 28 Main
stress the shielding requirements
(Same Both Sides)
Accelerator
Beampoint
Map
The new
(112 Beams Total)
Chamber Beam Map
28 Foot / 28 Main
(Same Both Sides)
Beam
Map
• Analyses Accelerator
completed
in late-2000
(112 Beams Total)
used 6  6 array; achieved 40Foot of
year lifetime with R.
half-angle
Pulse (28)
29º
L. Foot
Pulse(28)
R. Foot
Pulse (28)
• Even though all beam positions
are not occupied, new design
based upon 9  9 array; half-angle
limited to 24º
L. Main
Pulse
R.(28)
Main
ain
e (28)
R. Main
Pulse (28)
Pulse
(28)
Array
Angles:
min = 8.5°,
Main
Pulse
Main
Pulse
Foot Pulse
FootUnoccupied
Pulse
Unoccupied
max = 24.7°
Array Angles: min = 8.5°, max = 24.7°
JFL—7/02 ARIES
shielding
structure for
last magnet
set
target
injection
pathway
vortex tubes
for precision
beamline
shielding
cylindrical crossing flibe jets
for gross beamline shielding
target location
at shot time
1 cm of structure,
insulation, and H2O
cooling
1 cm of structure,
insulation, and LN2
cooling (inside and
outside coil)
size of beam
at front of
last magnet
alternating layers (1 cm
each/5 cm total) of W
and a proprietary
material (good for
neutrons)
3 cm thick coil
region (40 vol%
each of Cu and
Nb3Sn + 20 vol%
LHe)
2.3 cm of SS304 for
banding/shielding
beam profiles
target-facing end
of last magnet set
structure and liquid
injection/extraction
for vortex tube; liquid
stand-off from beam
is only 1 mm
target-facing end
of vortex tube
Three cases have been analyzed
for the new point design
• First case: minimal shielding between vortices:
– Insulator lifetime: 1.4 years
– Superconductor lifetime: 0.6 years
• Second case: took advantage of space between vortices (25
v% each SS, flinabe, Ti-hydride shielding, and void); used
alternating layers of high-/low-Z material for inner bore
shielding (end point from Oct. 2000):
– Insulator lifetime: 24 years
– Superconductor lifetime: 9 years
• Third case: tweaked inner bore shielding to emphasize
neutrons rather than gamma-rays:
– Insulator lifetime: 13 years
– Superconductor lifetime: 16 years
JFL—7/02 ARIES
Additional work is needed/planned
• Optimization to further increase magnet lifetime
• Add beam neutralization hardware, liquid jet nozzles
• Need to look at recirculating power for magnet cooling
(results already generated but have not been synthesized)
• Waste disposal needs drastic improvement: WDRcoil = 29
(94Nb)
• Evaluate/improve shielding design for 2nd-to-last & 3rd-tolast magnets; others? bending magnet?
• Consider high-temperature superconductors?
• Look at life-cycle waste volume vs. shielding design (long
life magnets may actually increase overall waste volume)
JFL—7/02 ARIES