SSC15-IV-3
Passive Space Radiation Shielding: Mass and Volume Optimization of Tungsten-Doped
PolyPhenolic and Polyethylene Resins
Benjamin Klamm
NASA Ames Research Center
Mail Stop 202-2; 650-604-6347
[email protected]
ABSTRACT
The use of commercial, off-the-shelf (COTS) components have become an increasingly attractive option to develop
small satellites that satisfy high-quality and reliability, as well as low-cost and development time demands.
However, the use of COTS may introduce mission design constraints associated with the mass and volume of
appropriate radiation shielding required for various mission lifetime and/or reliability objectives. The purpose of this
study is to provide insights towards optimization that minimizes the mass and/or volume of a graded-Z or
composite-Z shield. In this study, three model space radiation environments will be attenuated via mass shields
using numerical analysis, with a discussion provided on the manufacturability and integration of such shields within
the small sat design envelope. The use of graded-Z shielding for multi-particle attenuation is not a new concept.
However as complex shield systems are yet emerging, few concepts have the development maturity at a Technology
Readiness Level of three (TRL-3). Previous design approaches employed an optimization strategy that equally
weighed the structural ability of the given system equally against its shielding performance. The current study
removes the specific structural constraint, and examines the mass shielding efficiency associated with constant
thickness Tungsten-doped PolyPhenolic and Polyethylene resin shields.
allowing ever smaller satellites operating with enhanced
mission lifetimes.
INTRODUCTION
The low cost, high performance and availability of
COTS electronics are attractive to low budget small
satellite projects, as they allow direct competition with
larger space platforms. However, their use forces the
assumption of a risk posture and/or predicted mission
lifetime that are directly related to intensity of the
mission radiation environment. This is due to the broad
reaching unknowns associated with the radiation
susceptibility of COTS components, as the
supermajority are not designed towards or tested in
simulated space radiation environments. In order to
effectively reduce mission risk without participating in
often expensive and time consuming rad-testing, these
components should be insulated from the space
environment sufficiently so as to meet even the most
basic dose requirement. The current state-of-the-art for
insulating said components from the space radiation
environment involves using mass shielding with a
judicious material selection process and design.
In order to maintain low total system costs for both a
COTS component and its required additional shielding,
the direct material cost and manufacturing method are
additional influences within the associated shield
system engineering process.
Shield ease of
manufacturing is especially important as in-house
methods can be developed and applied to significantly
reduce process costs and development timeframes.
These include, but are not limited to: Fused Filament
Fabrication (FFF), Selective Laser Sintering (SLS),
Metal Injection Molding (MIM), etc.
SPACE RADIATION SHIELDING
The Spacecraft Environment
The natural space radiation environment within our
heliosphere consists of two general populations:
particles emitted from the sun including photons,
protons, electrons and the fully ionized nuclei Helium
through Uranium and particles from cosmic source.
These include the same particle variety only with a
higher specific-energy spectral range and peak bulk
energy, but a lower flux on the order of one percent of
the total count. Unsteady phenomena within the Sun
spill over into the heliosphere, known as solar particle
events (SPEs) and consist of two types: solar particle
Due to the inherent restrictions on small satellites, the
desire to drive the said mass shield efficiency higher is
presently an active research area within the aerospace
industry and academia. Shield efficiency could be
defined as a ratio between the shield’s ability to
attenuate particles divided by its mass and/or volume.
As the efficiency of shields increase, so too does the
performance density of the COTS/shield system,
Klamm
1
29th Annual AIAA/USU
Conference on Small Satellites
 1 2me c 2  2 2Tmax

  2  ...
 ln
2
2
I


dE
Z 1     C (  )


 Kz 2


zL
(

)

...
1

dx
A 2  2
Z
 2

 z L2 (  )



events and coronal mass ejections. For planets with
active magnetospheres, both solar and cosmic particles
can become trapped within specific belt-like regions
along the magnetic plane of the planet, in relation to the
current energy of the particle and local magnetic field
strength. The Earth’s trapped radiation regions consist
of high flux, low energy particles due to the relative
weakness of Earth’s magnetic field. Higher energy
particles are able to either penetrate to the atmosphere
proper or escape back into interplanetary space.
where,
K  4N A re2 me c 2
Space contains a very low energy plasma of protons
and electrons from the plasmasphere out to
interplanetary space, which is easily shielded by thin
material. However, it can damage some surface
materials over time and produce differential charging
effects in dielectric materials.2,3
Tmax 
(2)
2mec 2  2 2
(3)
 2me   me 
1 

m p   m p 

2
The Z/A term (atomic number to atomic mass ratio) of
the shielding material is generally proportional to its
stopping power, where materials with high
concentrations of low Z elements are preferential. This
basically translates into selecting materials with the
highest hydrogen content ratio, as seen in Table 1.
Radiation Effects in Electronics
Electronic components undergo damage and
degradation due to three general mechanisms: total
ionizing dose (TID), non-ionizing dose (often referred
to as displacement damage dose, DDD), where both
accumulate over time, and through single event effects
(SEEs) which are the ionizing and non-ionizing dose
associated with single particle strikes in critical areas.
The type of damage experienced is related to the
incident particle type and relative energy. COTS
components are generally more at risk to these damage
modes than components designed specifically for space
operations and are widely assumed to begin failing
around 5 krad in TID.1 As such, they require additional
mass and volume from the spacecraft operational
budget to attenuate said effects in the form of mass
shielding.
This mitigation strategy is especially
effective if the component is also preferentially selected
for inherent dose and single event effect robustness, and
used in conjunction with a watchdog system for soft
errors and latch-up resets, and current limiting devices
to help prevent device burnout.1
The production of secondary particles due to proton
bombardment can generally be neglected for small
spacecraft since the maximum energy of ionized
electrons, also proportional to Z/A, is very low (< 10
KeV) and the minimal amount of neutrons produced in
thin, non-fissionable materials. Neutron production has
been empirically shown to be directly proportional to
increasing atomic mass, where materials with heavy
nuclei produce more neutrons per incident proton
proportional to its incident energy (>120 MeV protons).
For ionic bombardment, fractionation of the incident
particle will also generate neutrons, alpha particles and
lesser lighter nuclei. Therefore high-Z materials will
generate more neutrons per unit thickness, but
preferential selection of the high-Z material will allow
for the re-absorption and scattering of any spallatial
neutrons. The strongest known neutron moderator is
Gadolinium, as shown in Figure 2. Tungsten is an
average neutron moderator for high-Z elements, and is
used herein primarily due to the cost and availability of
its associated microparticle powders.
Particle-Matter Interactions
The governing phenomena for particle–matter
interactions are related to the relativistic momentum of
the charged particle.
For moderately relativistic
charged hadronic particles (i.e.- protons, ions, etc.), this
is dominated by the inelastic coulombic collision of
particles with conducting and valence band electrons
via electron excitation and ionization. The corrected
Bethe formula describes this one dimensional energy
loss rate as found below:4
Klamm
(1)
Incident electron shielding is characterized by three
interactions: elastic nuclear scattering (small deflection
angle), inelastic nuclear scattering (large deflection
angle) leading to gamma-ray photon emission (a.k.a.
Bremsstrahlung), and inelastic momentum transfer
collisions with atomic electrons leading to atomic
electron excitation or ionization. For low energies
below the critical energy Ec, incident electron-atomic
electron collisions dominate, roughly proportional to
Z/A.
Above the critical energy radiative losses
dominate, proportional to Z2/A.
Elastic nuclear
2
29th Annual AIAA/USU
Conference on Small Satellites
scattering exists in small amounts at all energies, also
proportional to Z2/A. Because the critical energy
decreases with increasing Z, the combination of elastic
and inelastic nuclear scattering dominates at lower
electron energies in high-Z materials, where the Z2/A
term is orders of magnitude larger in high-Z materials
allowing them to be more effective at stopping
electrons.
610
Ec 
MeV
Z  1.24
Table 1:
Select shielding materials sorted by
density stopping power
Element or Density Avg. Avg.
Molecule [g/cm3] A
Z
(4)
Conversely, this radiative stopping produces more
Bremsstrahlung, where high-Z materials are more
capable at stopping these gamma-ray photons. Above
5-10 MeV per photon, pair production dominates for Z
> 20 materials. The produced positron quickly
annihilates with a nearby electron, generating two
photons each of at least 511 keV which are further
scattered, where the generated electron also attenuates
accordingly depending on initial energy. This
attenuation process is again generally proportional to
Z2/A. Compton scattering dominates at mid-range
energies for high-Z materials, where atomic electrons
are ionized by an incident photon which is scattered,
and is proportional to Z/A. Finally, the photoelectric
effect dominates for high-Z materials at lower energies,
where a photon is fully absorbed leading to electron
ionization, and is proportional to Z4/A.6 A diagram of
this is shown in Figure 1. Due to the combination of
these processes, high-Z materials tend to be better
photon attenuators, but this is slightly dependent on the
specific energy of the incident photon. The mass
attenuation length is shown in Figure 3.
Z/A Z^2/A
Density Density*
*(Z/A) (Z^2/A)
tungsten
19.25 183.80 74.00 0.40 29.79
7.75
573.52
tantalum
16.69 180.94 73.00 0.40 29.45
6.73
491.55
lead
11.34 207.20 82.00 0.40 32.45
4.49
368.00
gadolinium
7.90 157.25 64.00 0.41 26.05
3.22
205.78
aluminum
oxide
3.95
20.39 10.00 0.49
4.91
1.95
19.38
diamond
3.50
12.00 6.00
0.50
3.00
1.75
10.50
aluminum
2.70
26.98 13.00 0.48
6.26
1.30
16.91
boron carbide
2.52
11.05 5.20
0.47
2.45
1.18
6.17
sucrose
1.59
7.60
4.04
0.74
2.27
1.18
3.60
phenolic
novolac
1.36
5.68
3.14
0.80
1.86
1.08
2.53
graphite
2.15
12.00 6.00
0.50
3.00
1.08
6.45
glycerol
1.26
6.57
3.57
0.79
2.07
0.99
2.61
Aramid fiber
1.44
8.50
4.43
0.68
2.39
0.98
3.45
lithium oxide
2.01
10.00 4.67
0.45
2.19
0.91
4.41
PEEK
1.32
8.47
4.41
0.68
2.38
0.89
3.14
polycarbonate
1.21
4.06
4.06
0.71
2.24
0.86
2.71
water
1.00
6.00
3.33
0.83
2.00
0.83
2.00
HDPE
0.97
4.67
2.67
0.83
1.67
0.81
1.62
liquid
hydrogen
0.71
1.00
1.00
1.00
1.00
0.71
0.71
lithium nitride
1.27
8.75
4.00
0.45
1.84
0.57
2.34
Shield Material and Configuration Optimization
There have been several previous studies, and some
ongoing, examining the use of mixed-Z (or bi-layer)
and graded-Z (three to seven layers of varying Z
gradients) systems. It has been shown that a bi-layer
outside facing high-Z then low-Z shield proved to be
more effective in attenuating both protons and electrons
than aluminum, where graded-Z systems improved on
this attenuation efficiency by reducing mass 40 to
60%.6,10,11 These systems can be realized as either
independent shielding, or more commonly as an
integrated component to the structure of a given
spacecraft or habitat. Incorporating shielding and
structure is most beneficial when a majority of the
spacecraft requires shielding, as in human space
transport and extraterrestrial habitat construction. This
increases the total mass efficiency of the system and
potentially decreases the overall manufacturing
complexity.
When only smaller sections of the
spacecraft require shielding, also called spot shielding,
it is more mass-efficient to separate the structure and
Figure 1: Photon attenuation cross section as a
function of energy5
Klamm
3
29th Annual AIAA/USU
Conference on Small Satellites
Figure 2: Neutron Absorption and Scattering for Hydrogen through Uranium13
Figure 3: Photon absorption length for select elements4
Klamm
4
29th Annual AIAA/USU
Conference on Small Satellites
shield systems, where the shield’s structural constraints
are mostly removed. For spot shielding, minimizing the
mass and/or density remain as primary concerns but are
driven solely by shielding ability per unit
mass/thickness, where space qualification, material
cost/availability and ease of manufacturing exist as
more threshold-type constraints. This study examines
the use of both composite-Z and graded-Z spot shield
systems, where a low-Z resin matrix is doped with
high-Z metal microparticles in varying mass fractions.
This differs from some earlier studies which focused on
foil or deposition of pure material layers.6,8,10,11
Microparticle doping into a resin matrix decreases the
manufacturing complexity of the system by allowing
advanced manufacturing techniques such as fused
filament fabrication and injection molding: both allow
the low and high Z materials to be processed
concurrently, and both are highly advantageous to
incorporating shields into conformal-shaped volumes of
the interior of a small satellite. Also, structural
delamination, thermal warping, and other layered
composite concerns become a non-issue with
microparticle doped resin construction.
Of the possible high-Z materials the common selection
is between tungsten, tantalum and gadolinium due to
their availability, but each with differing good high-Z
characteristics. Gadolinium is most beneficial when
there is a large amount of secondary neutrons due to its
very large neutron absorption cross section. However,
for small spacecraft this is not commonly a problem, so
materials with better electron and photon attenuation
are used, such as tungsten and tantalum. As seen in
Table 1, tungsten is a slightly better electron attenuator
than tantalum and has essentially the same high energy
photon shielding characteristics due to their nearly
identical atomic numbers, with the added benefit of
being 20% denser. It should be noted that although
tungsten is a decent neutron absorber, it is the worst of
the three mentioned. It was selected as the example
high-Z microparticle over tantalum primarily due to its
higher availability and lower cost.
In order to effectively compare shielding materials, it is
often helpful to define an absolute quantity through
which to gauge shielding ability rather than relative
shielding as done in previous studies. Two such
quantities, the mass shielding coefficient and the
volume shielding coefficient, are defined below:
Generally, low-Z materials that require the fewest
atomic collisions to stop high energy particles (least
mass) also require the most volume to serve as effective
shields, due to their low density/high hydrogen content.
Previous studies have examined the effectiveness of
low-Z materials such as: liquid hydrogen,
polypropylene,
polyethylethyl-ketone,
aluminum,
polyethylene, carbon fiber, aramid fiber, silica fiber,
boron fiber, etc.6,7,8,10,11 This study examines the effects
of driving the low-Z material density higher by
increasing the mass fraction of the high-Z dopant; one
such material is examined in detail in reference to
standard shield sources. Phenolic Novolac resin was
compared to both the de facto practical hadron
attenuator High Density Polyethylene and the universal
standard spacecraft shield/structural source, Aluminum.
Phenolic Novolac was selected from a shielding
standpoint due to both its large stopping power (Z/A)
relative to HDPE and low secondary particle production
potential (Z2/A), while also being inexpensive, easy to
manufacture and process, widely available, highly
temperature stable and flame resistant, with very low
outgassing after a TVAC bakeout treatment. The only
issue with Phenolic resin, as for most polymers, is they
tend to serve as good di-electrics. In order to mitigate
charging within the spacecraft near sensitive
electronics, a neutralizing conducting layer must be
inserted below the spot shield. This plays well into the
shielding configuration in general, because said
electronics will already likely require a thermal
pathway regardless of covering material.
Klamm
m 
1
mD
[(rad/day)-1(g/cm2)-1]
(5)
v 
1
tD
[(rad/day)-1(cm)-1]
(6)
Where m and t represent mass (in g/cm2) and thickness
(cm) respectively, and D is the TID in silicon (rad/day)
for a given environment. This allows for a direct
comparison of shielding efficiency both within and
between environments, where larger coefficients
represent more mass or volume efficient systems.
In summary, this study is examining the mass and
volume shielding effectiveness in varying radiation
environments between:
1.
2.
3.
Graded-Z versus Composite-Z layering
Phenolic versus HDPE low-Z resins
Percentages of tungsten microparticle doping
Radiation Simulation Results
The TID variation deposited in a standard Silicon wafer
behind a constant thickness shield, with thicknesses
shown in Table 2, due to three model environments was
examined. These are: a standard “mid-altitude” GPS
orbit with results found in Figure 4 & Figure 7, a high
inclination sun-synchronous LEO orbit as in Figure 5 &
Figure 8, and an interplanetary heliocentric orbit at 1
5
29th Annual AIAA/USU
Conference on Small Satellites
AU as in Figure 6 & Figure 9. Details for each
environment, along with model parameters, are found
in Appendix A. Note that the dose values are not
representative of worse case conditions at each location,
since the parameters for each environment were
selected to show average conditions for the most
common general case. SPENVIS was used to define
these environments, as well as implement MULASSIS
to determine the attenuation characteristics of both the
graded-Z and composite-Z shield systems.12 The
standard SHIELDOSE-2Q results of aluminum were
included, also from SPENVIS. MULASSIS was ran
using the default cuts-in-range at 1E7 particles per run
to minimize error. Even with a large stochastic
simulation, both geometric effects and the default
particle generation cuts-in-range contributed to a
general 11% error. The 3 mm shield thickness was
selected because it was assumed to be the maximum
practical thickness available to a spot shield in small
satellites. Note that there is 1 mm of aluminum
included in all simulations due to the exterior structure
of the spacecraft itself.
Notes on Shield Manufacturing
Of the manufacturing options available for small scale
production of PolyPhenolic and HDPE test coupons,
three were examined in this study: fused filament
fabrication (FFF), injection molding (IM) and reactioncompression molding (RCM). These all afford in house
production of customizable shielding systems that
involve both relatively inexpensive capital equipment
and low manufacturing complexity. Red iron oxide
325-mesh was used as the microparticle simulant due to
its high visibility for evaluating doping uniformity and
very low cost.
Phenolic Novolac is a two part resin where PhenolFormaldehyde monomers are not fully polymerized due
to an excess of Phenol.
A cross-linker,
Hexamethylenetetramine (Hexa), is then added to
provide free formaldehyde, doing so as it decomposes
around 90 C to fully polymerize the mix into a
hardened thermoset. Since Phenolic Novolac starts
partially polymerized, it is a brittle solid thermoplastic
at room temperature. In order to effectively mix with
hexa, a powder, both must be ground together in a slow
process to not prematurely cure the mix within the
grinder itself. For the test coupon a simple handcranked burr mill coffee grinder was used to
successively shrink and blend the Phenolic with Hexa.
It is important to not grind-blend the hexa into the
phenolic until a sufficiently fine phenolic powder has
been achieved to prevent pre-curing. This is due to the
fact that once mixed even at room temperature, the two
powders will begin to cure at a very slow rate and the
grinding occurs at a moderately higher temperature the
coarser the beginning powder. After grind-blending
both components to fine powder, any dopants can then
be added and mixed. At this point the red iron oxide
was added and blended sufficiently, and the three part
powder was poured into a circular mold and vibrated
until level. It was baked at 150 C for approximately 15
minutes under atmospheric pressure. This partially
cures the coupon, which then must be die pressed to
remove voids and create better conformity to the
desired mold. This process could be considered a
combination of reaction molding and compression
molding, or RCM. Figure 10 shows the test coupon
after manufacture as described.
At each environment Phenolic and HDPE a compositeZ configuration was compared to graded-Z shields of
the same materials, with increasing Tungsten doping
mass fractions, while also maintaining a constant 3 mm
spot shield thickness. The graded-Z systems can then
be directly compared to the composite-Z by way of
average tungsten doping, where the total amount of
tungsten used is held constant and only its location
varied. Six shielding configuration were run for each
environment, as shown in Appendix B, with varying
levels of the tungsten doping starting with the pure
resin case. Mass fractions of 5%, 15% and 35%
tungsten were selected for the composite-Z system and
12% and 35% average tungsten for the graded-Z. Note
in Table 2 the thickness variation between the 12% and
35% graded-Z layers, due to the required doping
density exceeding the resin’s capability within the 1
mm layer, therefor a thicker layer was used.
Table 2: Simulation arrangement and thickness per
slab. %W represents the percent mass of Tungsten
microparticles
Composite-Z
Graded-Z 12% W avg. Graded-Z 35% W avg.
Al chassis 1 mm
Al chassis
1 mm
Al chassis
1 mm
Resin+%W 1 mm
Resin
1 mm
Resin
0.6 mm
Resin+%W 1 mm Resin+35% W 1 mm Resin+58% W 1.8 mm
Resin+%W 1 mm
Si sensor
Klamm
2 mm
Resin
1 mm
Resin
0.6 mm
Si sensor
2 mm
Si sensor
2 mm
6
29th Annual AIAA/USU
Conference on Small Satellites
Figure 4: Mass efficiency as a function of mass for a
standard GPS orbit
Figure 7: Volume efficiency as a function of mass for
a standard GPS orbit
Figure 5: Mass efficiency as a function of mass for a
standard Sun-sync orbit
Figure 8: Volume efficiency as a function of mass for
a standard Sun-sync orbit
Figure 6: Mass efficiency as a function of mass for a
standard interplanetary orbit
Klamm
Figure 9: Volume efficiency as a function of mass for
a standard interplanetary orbit
7
29th Annual AIAA/USU
Conference on Small Satellites
Figure 11: HDPE pellets in the hopper of a filament
extruder (left) and after repeated runs through the
extruder (right). Note the increasing doping density
(darker red) obtained with each pass through the
extruder, from right to left.
Figure 10: 4" diameter Fe2O3 doped Phenolic test
wafer
Generating the HDPE test article was a slightly more
arduous process, due to trouble with the FFF process,
whereby IM was utilized instead. Initially, IM-grade
HDPE was obtained in pellet form and fed at 140 C
nozzle temperature through a filament extruder made
specifically for generating polymer filament from
thermoplastics. This temperature generates a consistent
diameter filament, with before and after images show in
Figure 11. After producing a consistent filament of
approximately 20% iron oxide by weight, it was fed
into a 3D printer to produce a square test coupon. This
process proved to be very difficult, as the HDPE
consistently clogged the nozzle through a range of
nozzle temperatures from 160 to 230 C. This might
have been due to the printer design, where the filament
feed gear was located far from the nozzle. This low
pressure feed system did not sufficiently force the high
viscosity HDPE through the nozzle at a high enough
rate, where the polymer then melted within the isolation
chamber generating a clogged system. A printer
utilizing a direct feed system might alleviate this
problem. Without such a printer, a different direction
was taken and an injection molding system was
developed. This IM system served to both mix the
microparticle dopant into the HDPE resin with multiple
initial pass-throughs, and then inject the mixture into a
mold. This HDPE test coupon was unavailable for
photographic documentation at the time of publication,
but will be included in the conference presentation
proceedings.
As expected the shielding mass coefficient for both
resins increases in GPS orbit as the tungsten mass
fraction is increased, due to the increasing amount of
the more effective high-Z material in this electron-rich
environment. The same coefficient decreases for the
other two mostly hadronic environments, where
tungsten adds more mass than effective shielding
ability. Including high-Z materials into a shield system
in electron rich environments decreases total shield
mass, which was previously known. Importantly, note
that the shielding volume efficiency increased for all
simulated environments. Also, the composite system of
each material performed better than its graded system in
all cases, while the phenolic resin is preferential in
electron-rich environments and HDPE in hadron-rich.
One successful and one unsuccessful attempt at
generating a space qualification test article was
completed using lean, in-house processes as a first step
in supporting rapid generation and integration into
small satellite systems. The very low cost, readily
available procurement and ease of in-house
manufacture make each doped resin system both highly
attractive and highly effective for low budget missions.
For future work, a Mid-Z dopant such as iron or
molybdenum could be included into a composite
system for improved electron and Bremsstrahlung
shielding as compared against traditional graded Z
systems with both mid and high-Z layers in varying
configurations.
CONCLUSIONS
The mass and volume shielding efficiency associated
with constant thickness Tungsten-doped PolyPhenolic
and Polyethylene resins in graded-Z and composite-Z
shield configurations were evaluated against aluminum
in three model space radiation environments and found
to be absolutely superior for thin shields.
Klamm
Acknowledgments
I would like to thank Sasha Weston, Dayne Kemp,
Hemil Modi, Chad Frost, Tony Ricco, Andres Henke,
Alex Mazhari and Bryan Hackett for their contributions
to both my research and sanity.
8
29th Annual AIAA/USU
Conference on Small Satellites
13.
References
1.
Sinclair, D. and J. Dyer, Radiation Effects and
COTS Parts in SmallSats, Proceedings of 27th
annual AIAA/USU Conference on Small
Satellites; 2013 Aug 10–15; Logan, UT, USA;
2013.
2.
Holmes-Siedle, A. and L. Adams, Handbook of
Radiation Effects, p. 16, Oxford University Press,
Oxford, 1993.
3.
Frederickson, R., Upsets Related to Spacecraft
Charging, IEEE Trans. on Nucl. Science, Vol. 43,
No. 2, pp. 426-441, April 1996.
4.
Groom, D.E. and S.R. Klein, The European
Physical Journal C, Vol 15. Issue 1-4, pp. 163173, 2000.
5.
Krane, K., Introductory Nuclear Physics, John
Wiley and Sons, Inc., 1988.
6.
Koerselman,
J.R.,
A
Multidisciplinary
Optimization of Composite Space Enclosures,
Master of Science Thesis, Delft Univ. of Tech.,
2012.
7.
Likar, J.J. et al, Novel Radiation Design
Approach for CubeSat Based Missions,
Proceedings of 24th annual AIAA/USU
Conference on Small Satellites; 2010 Aug 9–12;
Logan, UT, USA; 2010.
8.
Gutierrez, N. et al, Passive Hybrid Thermoplastic
Graded-Z Radiation Shielding for a Europa
CubeSat Mission, University of Washington
Undergraduate Research Symposium 2015, Univ.
of Washington, Seattle, WA, USA, 2015.
9.
Clemens, M.A., Energy Deposition Mechanisms
for Proton and Neutron-Induced Single Event
Upsets in Modern Electronic Devices, Doctoral
Dissertation, Vanderbilt University, Nashville,
TN, USA, 2012.
10.
Wrobel, J. et al, Versatile Structural Radiation
Shielding and Thermal Insulation Through
Additive Manufacturing, Proceedings of 27th
annual AIAA/USU Conference on Small
Satellites; 2013 Aug 10–15; Logan, UT, USA;
2013.
11.
Fan, W.C., Drumm, C.R., Roeske, S.B., Scrivner,
G.J., Shielding considerations for satellite
microelectronics, IEEE Trans. Nucl. Sci. 43 (6),
2790–2796, 1996.
12.
Kruglanski, M. et al, Space Environment
Information System (SPENVIS), 38th COSPAR
Scientific Assembly 38, 4176, 2010.
Klamm
9
Dove, M.T., An Introduction to the Use of
Neutron Scattering Methods in Mineral Sciences,
European Journal of Mineralogy, 12, 203-224,
2002.
29th Annual AIAA/USU
Conference on Small Satellites
APPENDIX A
Table A.1: Radiation environment parameters
GPS
20,200 km Altitude
Trapped
Protons
Trapped
Electrons
Sun Sync
55 deg inclination
Solar Protons
Solar particle
Trapped proton Trapped electron model: ESPmodel: AP9
model: AE9 PSYCHIC total
fluence
Model run
mode: mean
Model run mode: Ion range: H mean
U
800 km Altitude
Interplanetary
98.7 deg inclination
GCR H-U
Trapped
Protons
Trapped
Solar
GCR H-U
Electrons Protons
GCR model:
ISO 15390
Same as
GPS
Same as Same as
GPS
GPS
Same as
GPS
Ion range: H U
External
magnetic field
model: OlsonPfitzer quiet
model
Prediction
period: 1.00 yr
Prediction
period: 1.00 yr (
Prediction
1.00 yr in solar
Prediction
period: 1.00 yr max., 0.00 yr in period: 1.00 yr
solar min, over
1 solar cycles)
Solar Protons
GCR H-U
Solar particle
model: ESP- GCR model: ISO
PSYCHIC total
15390
fluence
Ion range: H - U Ion range: H - U
Internal
50.00%
Internal magnetic
magnetic field
probability of Solar activity
field model:
model: IGRF
fluences not data: May 1996
IGRF (2012)
(2012)
being exceeded
Magnetic
Magnetic
shielding:
shielding:
External
Stormer
Stormer
magnetic field upgrade/quiet upgrade/quiet
model: Olson- magnetosphere/ magnetosphere/
Pfitzer quiet
CREME96
CREME96
model
magn. mom./all magn. mom./all
arrival
arrival
directions
directions
1 AU
95.00%
Solar
probability of
activitydata: May
fluences not
1996
being exceeded
No magnetic
shielding
No magnetic
shielding
Prediction
period: 1.00 yr (
1.00 yr in solar
Prediction
max., 0.00 yr in period: 1.00 yr
solar min, over 1
solar cycles)
Table A.2: MULASSIS parameters
Klamm
Particle Type
Trapped Protons, Solar Protons and GCR Ions
Trapped Electrons, HE photons
Geant4 physics list
QBBC
emstandard_opt3
Particle count
1E+7
1E+7
Resin
Chemical Formula
Phenolic
Pure: C16H26O2, Doped: C1024H1664O128W#
HDPE
Pure: C2H4, Doped: C1000H2000W#
10
Doping Relation
5% = W5, 15% = W15, 35% = W46,
58% = W119
5% = W4, 15% = W13, 35% = W41,
58% = W107
29th Annual AIAA/USU
Conference on Small Satellites
APPENDIX B
Table B.1
Environment ->
0% W
GPS
Sun Sync
Interplanetary
Resin ->
areal density
[g/cm2]
Phenolic
HDPE
Phenolic
HDPE
Phenolic
HDPE
4.1E-01
2.9E-01
4.1E-01
2.9E-01
4.1E-01
2.9E-01
dose [rad/day in Si]
1.6E+04
2.8E+04
6.0E+01
7.9E+01
5.9E+02
7.3E+02
5.7E-02
4.6E-02
1.5E+01
1.6E+01
1.5E+00
1.7E+00
7.7E-02
4.4E-02
2.0E+01
1.5E+01
2.0E+00
1.7E+00
areal density
[g/cm2]
7.0E-01
5.6E-01
7.0E-01
5.6E-01
7.0E-01
5.6E-01
dose [rad/day in Si]
9.2E+00
1.7E+01
1.0E-01
1.2E-01
1.1E+00
1.3E+00
1.6E-01
1.0E-01
1.4E+01
1.5E+01
1.3E+00
1.4E+00
3.6E-01
1.9E-01
3.3E+01
2.9E+01
3.0E+00
2.6E+00
1.2E+00
1.1E+00
1.2E+00
1.1E+00
1.2E+00
1.1E+00
1.3E+00
2.1E+00
7.5E-02
7.7E-02
7.0E-01
7.3E-01
shield mass
coefficient
shield volume
coefficient
Comp. Z
5% W
shield mass
coefficient
shield volume
coefficient
areal density
[g/cm2]
dose [rad/day in Si]
15% W
35% W
shield mass
coefficient
shield volume
coefficient
areal density
[g/cm2]
6.6E-01
4.4E-01
1.1E+01
1.2E+01
1.2E+00
1.3E+00
2.6E+00
1.6E+00
4.5E+01
4.3E+01
4.8E+00
4.5E+00
2.3E+00
2.2E+00
2.3E+00
2.2E+00
2.3E+00
2.2E+00
dose [rad/day in Si]
1.8E-01
2.8E-01
5.3E-02
5.3E-02
4.1E-01
4.0E-01
2.5E+00
1.6E+00
8.3E+00
8.6E+00
1.1E+00
1.1E+00
1.9E+01
1.2E+01
6.3E+01
6.3E+01
8.1E+00
8.4E+00
areal density
[g/cm2]
1.0E+00
9.3E-01
1.0E+00
9.3E-01
1.0E+00
9.3E-01
dose [rad/day in Si]
2.4E+00
3.2E+00
8.6E-02
8.8E-02
8.7E-01
9.3E-01
shield mass
coefficient
shield volume
coefficient
Graded-Z
12% W
average
35% W
average
Klamm
shield mass
coefficient
shield volume
coefficient
areal density
[g/cm2]
4.1E-01
3.4E-01
1.1E+01
1.2E+01
1.1E+00
1.2E+00
1.4E+00
1.1E+00
3.9E+01
3.8E+01
3.8E+00
3.6E+00
2.3E+00
2.2E+00
2.3E+00
2.2E+00
2.3E+00
2.2E+00
dose [rad/day in Si]
2.0E-01
3.3E-01
5.9E-02
6.0E-02
4.6E-01
4.7E-01
2.2E+00
1.4E+00
7.5E+00
7.5E+00
9.5E-01
9.5E-01
1.6E+01
1.0E+01
5.7E+01
5.6E+01
7.2E+00
7.0E+00
shield mass
coefficient
shield volume
coefficient
11
29th Annual AIAA/USU
Conference on Small Satellites