American Institute of Aeronautics and Astronautics

1

Mitigating the Effects of the Space Radiation Environment: A Novel Approach of Using Graded-Z Materials

William Atwell1

Boeing Research & Technology (BR&T), Houston, TX 77058

Kristina Rojdev2 NASA Johnson Space Center, Houston, TX 77058

Sukesh Aghara3 and Sirikul Sriprisan4 University of Massachusetts Lowell, Lowell, MA 01854

In this paper we present a novel space radiation shielding approach using various material lay-ups, called “Graded-Z” shielding, which could optimize cost, weight, and safety while mitigating the radiation exposures from the trapped radiation and solar proton environments, as well as the galactic cosmic radiation (GCR) environment, to humans and electronics. In addition, a validation and verification (V&V) was performed using two different high energy particle transport/dose codes (MCNPX & HZETRN). Inherently, we know that materials having high-hydrogen content are very good space radiation shielding materials. Graded-Z material lay-ups are very good trapped electron mitigators for medium earth orbit (MEO) and geostationary earth orbit (GEO). In addition, secondary particles, namely neutrons, are produced as the primary particles penetrate a spacecraft, which can have deleterious effects to both humans and electronics. The use of “dopants,” such as beryllium, boron, and lithium, impregnated in other shielding materials provides a means of absorbing the secondary neutrons. Several examples of optimized Graded-Z shielding lay-ups that include the use of composite materials are presented and discussed in detail. This parametric shielding study is an extension of some earlier pioneering work we (William Atwell and Kristina Rojdev) performed in 20041 and 20092.

Nomenclature GCR = galactic cosmic radiation GEO = geostationary earth orbit HZETRN = High Z and Energy Transport code MCNPX = Monte Carlo Nuclear Particle transport computer code MEO = medium earth orbit V & V = validation & verification

I. Introduction

RADED-Z shielding is a laminate of several materials with different Z values (atomic numbers) designed to protect against ionizing radiation. Compared to single-material shielding, the same mass of Graded-Z shielding has been shown to reduce electron penetration over 60%.3 Graded-Z shielding is commonly used in satellite-

based particle detectors and offers several benefits: protection from radiation damage, reduction of background noise for detectors, and lower mass compared to single-material shielding. Designs vary, but typically involve a gradient

1 Technical Fellow, Boeing Research & Technology, 13100 Space Center Blvd./HB 2-30, AIAA Associate Fellow. 2 Engineer, Systems Architecture and Integration Office, 2101 NASA Parkway, MC: EA-361, AIAA Student Member. 3 Associate Professor, Chemical Engineering (Nuclear), 1 University Ave., Perry Hall – 318. 4 Research Associate, Chemical Engineering (Nuclear), 1 University Ave., Perry Hall – 318.

G

https://ntrs.nasa.gov/search.jsp?R=20140002469 2020-06-11T16:52:03+00:00Z

American Institute of Aeronautics and Astronautics

2

from high-Z (usually tantalum [Ta]) through successively lower-Z elements such as tin, steel, and copper, usually ending with aluminum. Sometimes even lighter materials such as polypropylene and boron carbide are used.4,5 In a typical Graded-Z shield, the high-Z layer effectively scatters protons and electrons. It also absorbs gamma rays, which produces x-ray fluorescence. Each subsequent layer absorbs the X-ray fluorescence of the previous material, eventually reducing the energy to a suitable level. Each decrease in energy produces bremsstrahlung and Auger electrons, which are below the detector's energy threshold. Some designs also include an outer layer of aluminum, which may simply be the outer structure of a spacecraft or satellite. Fan, et al.6 reports “Shielding for space microelectronics needs to provide an acceptable dose rate with minimum shield mass. The analysis presented here shows that the best approach is, in general, to use a graded-Z shield, with a high-Z layer sandwiched between two low-Z materials. A Graded-Z shield is shown to reduce the electron dose rate by more than sixty percent over a single-material shield of the same areal density. For protons, the optimal shield would consist of a single, low-Z material layer. However, it is shown that a Graded-Z shield is nearly as effective as a single-material shield, as long as a low-Z layer is located adjacent to the microelectronics. A specific shield design depends upon the details of the radiation environment, system model, design margins/levels, compatibility of shield materials, etc.” In this paper we investigate several novel Graded-Z materials that are compared with baseline materials: aluminum (Al), high density polyethylene (HDPE), and water (H2O). In addition, we compare the results using two different computational computer codes, NASA HZETRN7 and MCNPX8 to validate and verify our results.

II. Study Assumptions

A. Graded-Z configurations The shielding configurations considered in this paper were two-layer shields with typical shielding thicknesses

and very large thickness configurations. The two layer shielding configurations were Al-HDPE, Ta-HDPE, and W-HDPE. The thicknesses of Al, Ta, and W were 150 mils, which correspond to 1.0287, 6.3237, and 7.3343 g/cm2, respectively, and the HDPE thicknesses were 5 and 10 g/cm2. The very large thickness configurations consisted of single layer H2O and two-layer H2O-SWX. SWX, specifically SWX-210, is a boron-doped hydrocarbon having a density of 1.19 g/cm3. The water thicknesses ranged from 100-300 g/cm2, and the SWX thicknesses were 10 and 20 g/cm2.

B. Environments For deep-space exploratory missions, the radiation sources are due to the naturally-occurring galactic cosmic

radiation (GCR) and from solar particle events (SPEs). The radiation exposure analyses discussed later is based on the GCR environment during solar minimum (epoch 1977), solar maximum (epoch 1982), and the classic series of solar proton events that occurred 19-24 October 1989. The October 1989 SPEs were four events that were combined using the Band fit.9,10 For all three environment cases, the energy spectrum was restricted to 2 GeV due to limitations in the MCNPX software, as shown in the differential energy spectrum below (Figure 1).

American Institute of Aeronautics and Astronautics

3

Figure 1. The Band fit differential energy spectrum for the 19-24 October 1989 SPEs.

C. High Energy Particle Transport/Dose Codes Two high energy particle transport/dose codes were used in the computations, HZETRN7 and MCNPX8, and the

results were compared as a check on the verification and validation of the absorbed doses. The HZETRN code was develop at NASA Langley Research Center in 1995. The high-charge-and energy (HZE) transport computer program HZETRN was developed to address the problems of free-space radiation transport and shielding. The HZETRN program is intended specifically for the design engineer who is interested in obtaining fast and accurate dosimetric information for the design and construction of space modules and devices. The program is based on a one-dimensional space-marching formulation of the Boltzmann transport equation with a straight-ahead approximation. The effect of the long-range Coulomb force and electron interaction is treated as a continuous slowing-down process. Atomic (electronic) stopping power coefficients with energies above a few MeV are calculated by using Bethe`s theory including Bragg`s rule, Ziegler`s shell corrections, and effective charge. Nuclear absorption cross sections are obtained from fits to quantum calculations and total cross sections are obtained with a Ramsauer formalism. Nuclear fragmentation cross sections are calculated with a semi-empirical abrasion-ablation fragmentation model. The relation of the final computer code to the Boltzmann equation is discussed in the context of simplifying assumptions.

The MCNPX code was developed at Los Alamos National Laboratory (LANL) and is described at their website11 as “MCNP is a general-purpose Monte Carlo N-Particle code that can be used for neutron, photon, electron, or coupled neutron/photon/electron transport. Specific areas of application include, but are not limited to, radiation protection and dosimetry, radiation shielding, radiography, medical physics, nuclear criticality safety, detector design and analysis, nuclear oil well logging, accelerator target design, fission and fusion reactor design, decontamination and decommissioning. The code treats an arbitrary three-dimensional configuration of materials in geometric cells bounded by first- and second-degree surfaces and fourth-degree elliptical tori. Point-wise cross-section data typically are used, although group-wise data also are available. For neutrons, all reactions given in a particular cross-section evaluation (such as ENDF/B-VI) are accounted for. Thermal neutrons are described by both the free gas and S(alpha,beta) models. For photons, the code accounts for incoherent and coherent scattering, the possibility of fluorescent emission after photoelectric absorption, absorption in pair production with local emission of annihilation radiation, and bremsstrahlung. A continuous-slowing-down model is used for electron transport that includes positrons, x-rays, and bremsstrahlung but does not include external or self-induced fields. Important standard features that make MCNP very versatile and easy to use include a powerful general source, criticality source, and surface source; both geometry and output tally plotters; a rich collection of variance reduction techniques; a flexible tally structure; and an extensive collection of cross-section data. MCNP contains numerous flexible tallies: surface current & flux, volume flux (track length), point or ring detectors, particle heating, fission heating, pulse height tally for energy or charge deposition, mesh tallies, and radiography tallies.”

1.E+02

1.E+03

1.E+04

1.E+05

1.E+06

1.E+07

1.E+08

1.E+09

1.E+10

0 500 1000 1500 2000

Diff

. Flu

ence

, pr

oton

s/cm

2-M

eV

Proton Energy, MeV

A. Two-lThere

aluminum high densithicknesses

The reswhile the G

Figur

The reenvironmeHowever, environmethe single l

layer Configurwere three ar(Al), Tantalumity polyethylens of Al or HDPsults for the AGraded-Z Al-H

re 2. Al-HDPE

sults of the Tents, the Ta-Hthe Ta-HDPE

ent, the reverselayer of HDPE

Americ

rations rrangements inm (Ta), or Tunne (HDPE) atPE.

Al-HDPE systemHDPE and singl

E two-layer sy

Ta-HDPE systeDPE system wcase with 10

e is true. The E is superior for

can Institute of

III.

nvestigated forgsten (W) andt 5 or 10 g/c

m (Figure 2) le layer HDPE

ystem compare

em (Figure 3)with 5 g/cm2

g/cm2 of HDPTa-HDPE systr dose mitigatio

f Aeronautics a

4

Results

r the two-layed corresponded cm2 thickness.

show that the E systems are co

ed with Alum

) show similarof HDPE is

PE is comparatem is now comon.

and Astronauti

er configuratiod to 150 mils o

These layer

single layer aomparable in d

minum and HD

r results to theslightly superi

able to the sinmparable with

ics

ons. The outf thickness. Trs were comp

aluminum casedose.

DPE at the sam

e Al-HDPE caior to the sing

ngle layer of Hh the single lay

ter layer was The inner layers

ared with the

e has the worst

me thicknesses

ase. For thegle layer of H

HDPE. For thyer of aluminu

either s were

e same

t dose,

s.

e GCR HDPE. he SPE um and

Figur

The reenvironmeg/cm2 of Hmore comp

re 3. Ta-HDPE

sults of the Wents, the W-HDHDPE is more cparable to the s

Americ

E two-layer sy

W-HDPE systeDPE system wcomparable to single layer of

can Institute of

ystem compar

em (Figure 4)with 5 g/cm2 of

the single layealuminum, and

f Aeronautics a

5

red with Alum

) show similarf HDPE is sliger of HDPE. Fd the single lay

and Astronauti

minum and HD

r results to thghtly superior, For the SPE enyer of HDPE is

ics

DPE at the sam

he Ta-HDPE cbut the W-HD

nvironment, thes the best radiat

me thicknesses

case. For theDPE system we W-HDPE systion mitigator.

s.

e GCR with 10 stem is

Figur

The twshielding pthicknesses

B. Very LFor the

water withand the twsame overcomparabl

re 4. W-HDPE

wo layer configupurposes. Ths change these

Large Thickne very large thh a borated pol

wo thicknesses rall thicknessee to the single

Americ

E two-layer sy

urations discushe next sectioresults.

ness Configurahickness configlyethylene knoof SWX consi

es was also colayer of H2O a

can Institute of

ystem compare

ssed in this secton will consid

ations gurations, thereown as SWX. idered were 10onsidered. That the same thic

f Aeronautics a

6

ed with Alum

tion are similarder very large

e were six desThe three thic

0 and 20 g/cm2

he results (Figcknesses.

and Astronauti

minum and HD

ar to historical se thicknesses

signs analyzed cknesses of wa2. For compargure 5) show

ics

DPE at the sam

spacecraft thicof water and

that consistedater were 100, rison, a single

that the H2O

me thicknesses

knesses for radwhether the

d of combinati200, and 300 layer of water

O-SWX system

s.

diation larger

ons of g/cm2

r at the ms are

Figure 5.

C. VerifiFor the

componentCarlo nucllayer confi

The radfrom left toMeV to 2 Gin the radMCNPX aexplicitly: The photon

The MCenergy rancombinatioto generatetissue phanover a cell

All calperformed statistical uto see beca

FigureSPE) for thsubstantialGCR (solalower comcombinatio

H2O-SWX ve

ication and Vae verification t of the GCR lear particle traigurations and diation source o right along z GeV. The radiaial direction a

and HZETRN neutrons, prot

n-electron tranCNPX code us

nges, and intranons of flux talle results. The ntom surfaces,volume, result

lculations presat a total of

uncertainty. Thause of its triviae 6 shows the phe five differenlly lower as coar minimum an

mpared to the Gon with water p

Americ

ery large thick

alidation and validationare transporte

ansport compuone case of the(GCR or SPEaxis toward a

al surface of thand to accompresults. The M

tons, photons, sport is not preses the mix-manuclear cascadly (F4), energyF4 tallies werand at variou

ting in dose (Gsented in this 2 million hist

he correspondinal values.

plot of neutron nt shielding casompared to thend maximum) cGraded-Z shielproduce a high

can Institute of

kness system c

n, high energed through shiuter code, MCe very large thiE) spectra is m

simple cylindehe cylinder is mplish a one diMCNPX simulelectrons and lesent in the HZatch algorithm e (INC) physic

y depositions pe used to evals phantom dep

Gy/source particstudy are per

tories. The calng error bars w

flux (secondarses. It is evidene Graded-Z shicases, respectivd cases. Addit

her thermal neu

f Aeronautics a

7

compared with

y proton parteld material a

CNPX. The MCickness configu

modeled as an er (of 15 cm ra

modeled as a reimensional simlations are perlight particles

ZETRN versionthat utilizes n

cs models to her track lengthuate energy sp

pths. The F6 tacle/s). rformed with lculated results

were plotted in

ry neutrons pront that the neutield cases. Figvely. Again notionally, we noutron flux as c

and Astronauti

h single layer

ticles producedand tissue equiCNPX simulaturation, for eacuniformly dist

adius). The inpeflecting boundmulation. This rformed with tincluding deut

n used in this snuclear data tabhandle interactih tally (F6), anpectra at a tranally is used to

MCNPX verss were set to the curves, bu

oduced, resultintron fluxes (10gure 7 and Figotice the neutrootice (more proompared to the

ics

of H2O at the

d by SPE eveivalent geomettions were perch environmentributed planar

put spectra sourdary to simulat

allows for cothe following teron, triton, h

study. bles to transporions at high en

nd cylindrical mnsition point bcalculate abso

sion 2.7.0. Ameet a minim

ut in most cases

ng from the pr-150 MeV) in gure 8 show ton fluxes for thonounced) thae water case. T

e same thickne

ents and the ptry using the M

rformed for thent case. r source proparce ranges frome an infinite momparisons beparticles trans

helium-3, and a

rt particles at cnergy regions.1

mesh tally werbetween shieldorbed dose, ave

All simulationsmum of 10% os, they were no

rimary protons the water shiel

the flux plots fhe water shield

at the SWX layThis may be du

esses.

proton Monte e two-

agating m 0.01

material etween ported alphas.

certain 12 The re used ing, at eraged

s were or less ot easy

of the lds are for the d to be yers in ue to a

higher primthese resulcompared tadditional

mary proton intlts are viewed to just the wateshielding.

Americ

teraction with in context wit

er case. This s

Figure

Figure

can Institute of

the shield mateth the dose ressuggests that th

6. October 19

7. 1977 GCR

f Aeronautics a

8

erial, resulting sults, it shows he more damag

989 SPE Band

(solar minimu

and Astronauti

g in a higher secthat the does

ging protons ar

d fit neutron fl

um) neutron f

ics

condary neutrois lower for th

re removed by

lux.

flux

on production. he SWX casesthe SWX, pro

When s when viding

Table

trend to thvalues. This one of thto the neutbetween thMCNPX adifferencesare higher necessary solar minim

1 shows the she HZETRN reese results are he major reasotron and light he results. Wheand HZETRN s in the transpofor the 1982 Gto understand mum results.

Americ

Figure 8

ummary of Mesults. Overallconsistent wit

ons for these diion transport n

en the ratios ofis much smallort calculation GCR solar maxwhy the differ

can Institute of

8. 1982 GCR (

MCNPX dose re, the MCNPXth previously pifferences. Alsnow available f the relative thler (5-10%). Th

algorithms anximum results rence is much

f Aeronautics a

9

(solar maximu

esults for the X dose values apublished workso the 2005 vein the 2010 rehicknesses of this is stated as

nd in the cross-(Table 3), appgreater than w

and Astronauti

um) neutron f

SPE. Clearly tare 15% to 30k.12, 13 The absersion of HZETelease.14 Thisthe shields are s the D ratio. -sections used proximately 50

what we observ

ics

flux.

the dose value0% higher thansence of EM trTRN does not ialso contribut compared, theThis differencby the two co

0% higher. Furve for the SPE

es show a conn the HZETRNransport in HZEinclude the upgtes to the diffee difference be

ce may be due odes. The differther investiga

E and the 1977

nsistent N dose ETRN grades

erences etween to the

erences ation is 7 GCR

*here th**the D

he HZETRN dD ratio is the ra

Americ

Table 1.

dose is higher thatio of dose val

Table 2.

can Institute of

Dose values f

han MCNPX lues from 10g/

Dose values fo

f Aeronautics a

10

for October 19

cm2 shield to t

for 1977 GCR

and Astronauti

989 SPE Band

the 5 g/cm2

(solar minimu

ics

d fit.

um).

This stresults shoenvironmecompared w

The vevalues betwneeded to e

FurtherunderstandresponsibleconfiguratiZ configurstructures. advantages

1Bartholet, BEnergy Spac 2AtwellComputatio16 July 200

3Fan, WNo. 6, 1996

4Smith, 5Pia, M

3649. 6Fan, W

IEEE Trans7Wilson

and TripathNASA Techn

8MCNP

tudy investigatowed some slient. Howeverwith a single la

erification and ween the codeselucidate why rmore, since td whether thee and how thions is importaration, but de In addition,

s over two-laye

B., Atwell, W., ce Radiation Env, W., Boeder, nal and Experim9.

W.C., et al., "Shie, pp. 2790–2796D.M., et al., “Th. G., et al., "PIX

W. C., Drumm, Csactions in Nuclen, J. W., Badavi,hi, R.K., “HZETnical Paper 349

PX, http://en.wik

Americ

Table 3.

ted the use of ight dose redu, the Graded-Zayer of HDPE.

validation shs were due to kthere are largerthis study has e incident radhis will affectant because if wcrease the ovefuture work w

er or single-lay

Clowdsley, M., vironment,” SPAP., Wilkins, R

mental Results,”

elding Considera6. he RHESSI Spec

XE Simulation w

C. R., Roeske, Sear Science, Vol, F. F., CucinottTRN: Descriptio5, 1995.

kipedia.org/wiki/

can Institute of

Dose values fo

IV.f Graded-Z mauctions in usinZ configuratio In the very la

howed similar known discrepr differences foonly focused

diation (protont the dose eq

we can acquireerall mass of will also inve

yer systems.

R

Reddell, B., andACE-2004, San D

R., and Rojdev,SAE Internation

ations for Satelli

ctrometer,” Solawith Geant4," IEE

S. B., and Scriv. 43, No. 6, 1996ta, F. A., Shinn,on of a Free-Sp

/MCNPX

f Aeronautics a

11

or 1982 GCR

Conclusionsaterials for radng Tantalum oon was actuallarge thickness ctrends betwee

pancies in methor the GCR sold on the dose,ns) or the proquivalent. Fu comparable rathe system, tstigate three-l

References

d Wilson, J., “ODiego, CA, Sept K., “Analyses

nal Conference o

te Microelectron

ar Physics, Vol. 2EE Transactions

vner, G. J., “Shi6. J. L., Badhwar,ace Ion and Nu

and Astronauti

(solar maxim

s diation mitigatior Tungsten wly detrimental configurations

en HZETRN ahods of calculalar minimum e, additional woduction of surthermore, anadiation mitigathere is still vayer systems

Optimized Spacet. 28-30, 2004. s of Several Son Environment

nics", IEEE Tran

210, No. 1-2, 20s on Nuclear Sci

elding Consider

, G. D., Silberbeucleon Transpor

ics

mum).

ion of proton with 5 g/cm2 o

with an SPEs, the results weand MCNPX. ation. Further environment cawork needs to secondary radn analysis of ation characterivalue in pursuto evaluate w

ecraft Material C

Space Radiationtal Systems (ICE

nsactions on Nu

002, pp. 33–60. ience, Vol. 56, N

rations for Satel

erg, R., Tsao, Crt and Shielding

environmentsof HDPE in aE environment ere comparable The differeninvestigation w

ase. be implemen

diation (neutrothe mass of

istics with a Guing these Grawhether they p

Composition in a

n-Mitigating MaES), Savannah, G

uclear Science, V

No. 6, 2009, pp.

llite Microelectr

C.H., Townsendg Computer Pro

. The a GCR

when e.

nces in will be

nted to ons) is f these

Graded-aded-Z provide

a High-

aterials: GA, 12-

Vol. 43,

. 3614–

ronics,”

, L.W., ogram,”

American Institute of Aeronautics and Astronautics

12

9Atwell, W., Tylka, A., Dietrich, W., and Badavi, F. F., “Radiation Exposure Estimates For Extremely Large Solar Proton Events,” 37th Committee on Space Research (COSPAR) Scientific Assembly, Montreal, Canada, 13-20 July 2008.

10Allan J. Tylka, William F. Dietrich and William Atwell, “Band Function Representations of Solar Proton Spectra in Ground-Level Events,” 38th Scientific Assembly of the Committee on Space Research (COSPAR , Bremen, Germany, 18-25 July 2010.

11http://mcnp.lanl.gov/ 12Aghara, S. K., Blattnig, S. R., Norbury, J. W., and Singleterry, R. C., “Monte Carlo Analysis of Pion Contribution to Absorbed Dose from Galactic Cosmic Rays,” Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms, Vol. 267, No. 7, 2009, pp. 1115-1124. 13Slaba, T. C., Blattnig, S. R., Reddell, B., Bahadori, A., Norman, R. B., and Badavi, F. F., “Pion and Electromagnetic Contribution to Dose: Comparisons of HZETRN to Monte Carlo Results and ISS Data,” Advances in Space Research, Vol. 52, No. 1, 2013, pp. 62-78. 14Slaba, T. C., Blattnig, S. R., Aghara, S. K., Townsend, L. W., Handler, T., Gabriel, T. A., et al., “Coupled Neutron Transport for HZETRN,” Radiation Measurements, Vol. 45, No. 2, 2010, pp. 173-182.