Sage Journals: Discover world-class research

Abstract

Polyethylene and polyethylene-based composites increasingly draw researchers’ attention in space exploration applications because of their unique characteristics such as lightweight and radiation shielding. This study focuses on the space radiation shielding of polyethylene and polyethylene fiber reinforced polyethylene matrix composites for applications such as long duration human exploration beyond the low Earth orbit (LEO) to minimize crew and equipment exposure to the interplanetary radiation environment. Polyethylene-based composites with fiber reinforcements including ultra-high molecular weight polyethylene (UHMWPE) fibers are developed and their radiation shielding efficacy is evaluated. Specialized coating methods and incorporation of the coated substrate in the composite for shielding against secondary neutrons are presented. Results from the physics-based Particle and Heavy Ion Transport Code System (PHITS) are used to estimate the efficacy of the composite materials for shielding against Galactic Cosmic Rays (GCR). Simulation results for secondary neutrons in the energy range of thermal to 10 GeV is also discussed. Secondary neutron test results are obtained using a PuBe source in the energy range of 1 – 10 MeV. The radiation shielding results of the thermoplastic composite show its great potential for safe long duration human exploration beyond LEO.

Keywords

Thermoplastic composite radiation shielding vacuum plasma coating galactic cosmic rays secondary neutrons

Introduction

NASA and the commercial sector’s vision for space exploration includes long duration human travel beyond low earth orbit (LEO) and sustained human presence on other planetary surfaces. For this vision to be a reality, one of the major challenges that must be overcome would be to manage the radiation exposure to crew and equipment from the interplanetary radiation environment. The management of that risk will be guided by NASA’s planned adoption of a career exposure limit of 600 mSv.¹ In LEO, the radiation exposure to astronauts is kept below that limit by limiting their exposure time and by taking advantage of the shielding still offered by the geo-magnetic field.² However, exposure to free space cosmic radiation during a round trip to Mars or during extended stays of up to a few months on the surface of the Moon or Mars could result in significant exposures at or above the 600 mSv limit to crew members and potential effects to electronics. Calculations have shown that using an Al capsule or spacecraft with a 30 g/cm² areal density (>10 cm in thickness) will limit the total mission trip to Mars to 4 years.³ Escaping earth’s gravitational well with such a metallic structure can be cost and mass prohibitive. Most current crew capsules are built with an external skin that is only a few millimeters thick and uses an internal honeycomb structure for strength. To address the multifunctional material requirement of sufficient shielding and low-density structural material, polyethylene (PE) has been found to be one of the best suited materials [4^, 5]. To explain the suitability of a material such as PE mandates a brief discussion on the nature of the material, the deep space radiation environment, and the interaction of the radiation with the shield material.

Polyethylene, a thermoplastic polymer with a chemical formula of (C₂H₄)_n, can be categorized into several types based on their molecular weight, including high density polyethylene (HDPE) and ultra-high molecular weight polyethylene (UHMWPE).⁶ HDPE has molecular weight that ranges between 50,000 and 250,000 and it is commonly used as the matrix material for various applications. Ultra-high molecular weight polyethylene has a molecular weight between 3 million and 7.5 million, and it has been used in producing lightweight and high strength fibers for applications that require lightweight and high strength materials. The degree of crystallinity and the level of alignment can reach up to 85% and 95%, respectively. Both of them have the same chemical formula with a hydrogen mass content of approximately 14%.⁷ As will be discussed later it is the high hydrogen content of PE that forms the basis of superior radiation shielding.

The galactic radiation environment consists primarily of a continuous flux of galactic cosmic rays (GCRs)⁸ and transient but intense fluxes of solar energetic particles (SEPs).^8,9 The primary constituents of the GCR spectrum are 89-90 % protons, 8-9% alpha particles, and 1-2% heavy nuclei with energies ranging from 10 MeV/nucleon to 10s of TeV/nucleon.¹⁰ The SEPs on the other hand consist primarily of protons and alpha particles with energies ranging from a few MeV/nucleon to few 100s of MeV/nucleon.^9–11

On the Lunar or Martian surface crew members are protected from the continuous GCR flux over 2π radians by the planet. However, interaction of the GCR flux with the planetary soil or regolith produces secondary neutrons. The dose equivalent of secondary neutrons on the Lunar surface can vary depending on location. Monte Carlo simulations have shown that during a solar minimum the Lunar Highlands (Al and Ca rich) region have a smaller dose equivalent for secondary neutrons compared to the Mare (Fe and Ti rich) regions.¹² Ti and Fe have a larger cross-section for secondary neutron production compared to Al and Ca. This secondary neutron dose on the Lunar surface can be categorized based on their energy levels as thermal (E <0.4 eV), epithermal (0.4 eV <E< 1 MeV) and fast (E >1 MeV). Calculations based on the Geant4 code have shown that on average the total neutron dose is approximately 23% of the total dose on the Lunar surface and is dominated by the fast neutrons.¹² Other calculations predict that up to 50% of the dose equivalent on a Lunar base can come from neutrons.^13,14 It has also been demonstrated that during a worst-case solar particle event (SPE) an astronaut in a space suit on the Lunar surface can receive an acute dose of protons and neutrons significantly in excess of the 30 day limit.¹⁵ The Lunar Neutron Probe Measurements conducted as far back as Apollo 17 showed that the flux of thermal and epithermal neutrons increased significantly up to 1 m below the Lunar surface.¹⁶ More recently, the Chang’E Lunar Lander Neutron and Dosimetry (LND) experiment on the far side of the moon measured a surface dose rate of 23.5% for neutrons.¹⁷ Irrespective of the habitat location, it is necessary to develop habitat materials that will minimize the generation of secondaries and provide protection to humans from the GCR flux and the neutron flux that leaks to the planetary surface from the regolith.

A material-based countermeasure against the free space radiation environment is dictated by two specific interactions, namely energy loss and fragmentation. The energy loss of charged particles per unit length of material traversed (also known as the stopping power or the Linear Energy Transfer, LET) is directly proportional to the square of their atomic number (or, more precisely, their charge state) and inversely proportional to the square of their velocity. Fragmentation of the incident heavy ion projectile leads to the formation of smaller fragments generally moving at the same velocity and in the same direction as the incident particle. Light secondary ions such as protons, deuterons, tritons, He isotopes and neutrons can be emitted in all directions over an extensive energy range. The charged secondaries are less ionizing than the primary HZE ion due to their lower atomic number, but are also more penetrating when emitted at velocities close to the incoming HZE ion’s velocity. Breaking up the heavy ions in the GCR flux into smaller fragments with lower ionizing power creates a complex secondary radiation field that must be included in any design of passive radiation shielding. It is also important in this process to minimize the production of secondaries from target fragmentation that can otherwise add to the radiation risk.^18,19 Secondary neutrons from target fragmentation are one such constituent that can potentially increases radiation risk.

Considering that a potential radiation shield should maximize the likelihood of stopping the primary particle while limiting the production of lighter and more penetrating secondary particles, polyethylene (PE), as mentioned earlier, has been found to be one of the best suited materials for radiation shielding^4,5 since it has a very high density of hydrogen (H) atoms. As H has the greatest charge to mass ratio (Z/A = 1, whereas other elements have Z/A = 0.5 or less) it provides the highest stopping power per atom, while at the same time provides the smallest target and smallest cross section for nuclear interactions, which in turn minimizes the production of penetrating secondaries and target fragments. Moreover, the absence of elements heavier than carbon (C) reduces the number of target fragment neutrons produced in nuclear interactions when compared to heavier elements.

This paper discusses the development, testing and simulation of a multilayer PE composite material that will provide shielding against cosmic radiation and secondary neutrons generated from interactions with planetary regolith or crew vehicles and habitats. The choice of a multilayer architecture was driven by the findings from previous studies where dispersion of B containing particulate or fiber compounds did not adequately address the requirement for multifunctionality, namely shielding and structural capability and not simply a passive shield adding to the mass of a space vehicle or habitat. For example, in injection molded HDPE and BN composites the mass absorption of thermal neutrons increased by a factor of 2 when the BN weight percent was increased from 1 % to 30%. However, the maximum achieved tensile strength and elastic modulus were limited to 42 MPa and 0.7 GPa respectively.²⁰ The low tensile properties with addition of B fillers such as particles and fibers have been attributed to lack of interfacial bonding. PE with a hydrophobic surfaces and fillers such as BN and BC with polar surfaces are expected to have poor adhesion.^21,22 To circumvent this problem surface functionalization of the filler materials with compounds such as silane has been attempted. Such attempts have resulted in inconsistent structural results with minimal increase of mechanical properties.^21–23 However, the presence of the surface treated B compounds did demonstrate improvement in neutron attenuation with a 30% reduction in low energy neutron transmission.²³ More recently the potential of polyethylene/graphene nanoplatelet (GNP) composites, processed via fused filament fabrication (FFF) additive manufacturing, for long-term space missions has been explored.²⁴ The innovative intent here was to repurpose polyethylene waste—such as astronaut food packaging—into 3D-printed tools and components. Although incorporating 1 wt% GNP into medium-density polyethylene (MDPE) improved mechanical strength, the maximum tensile strength reported was 5.5 MPa. The effect on strength after simulated space environment exposure was not quantified. Using polyethylene as the base and FFF additive manufacturing techniques with other interesting additives such as multi-walled carbon nanotubes (MWCNT), and hybrid MWCNT/GNP nanofillers have been pursued for space applications.²⁵ The effect of proton irradiation on the stability of such composites in terms of electrical properties, wettability, thermal properties, and surface morphology was studied. Further studies are required to quantify the effect on mechanical properties and radiation shielding efficacy after such irradiation. Compared to different graphene-based fillers, radiation simulation has shown that graphene oxide (GO) additive was the least deleterious in terms of radiation shielding effectiveness due to the presence of hydrogen-rich functional groups immobilized on the graphene planes and edges.²⁶ Transport simulations have also shown that incorporating lithium hydride (LiH) into HDPE provides better shielding than even pure HDPE since LiH contains approximately 12.7% hydrogen by weight.²⁷ Unfortunately, LiH can ignite almost spontaneously even in low humidity air making it difficult to handle during processing. In comparison to composites with additives, this paper discusses a multilayer composite architecture approach balancing the shielding efficacy against cosmic and secondary neutron radiation with demands for structural integrity. Compared to other studies, our multilayer composite approach discussed in the following section has demonstrated high tensile strength in the range of 250 MPa and modulus of 10 GPa, surpassing the specific strength and modulus of commonly used aerospace aluminum alloys such as Al 2219 and Al 2195.²⁸ As opposed to using simulated space environments, detailed characterization of the multilayer composite was carried out before and after long duration exposure to the actual space environment of LEO.²⁸

This paper focuses on the radiation shielding efficacy of such a composite against the primary GCR radiation spectrum and secondary neutrons that can be generated from interactions of the GCR spectrum with planetary regolith or crew vehicles and habitats. Transport calculations were used for estimating the efficacy against GCRs and testing using a neutron source was used to quantifying the secondary neutron shielding efficacy.

Materials and Methods

Detailed description of the composite architecture using a combination of ultra-high molecular weight polyethylene (UHMWPE) fabric, B coated high strength carbon fabric, low density polyethylene (LDPE) and the associated fabrication process is presented in our recent publication.²⁴ For the convenience of the reader the relevant details are summarized here. The materials and their treatments used for fabrication of the composite panel are summarized in Table 1.

Table 1.

Details of materials used for composite fabrication.

Material	Manufacturer	Material information
UHMWPE fabric	Honeywell	Spectra® 1000, Plain Weave, ultimate tensile strength 3 GPa, tensile modulus 103 GPa, density 0.97 g/cm³.
Carbon fabric	Toray carbon fibers America inc	Toray - T300, 3K tow size, plain weave 0/90o, ultimate tensile strength 3.5 GPa, tensile modulus 103 GPa, density 1.76 g/cm³. Ar-H2 plasma was used to deposit the boron on the carbon fabric at a power of 100 kW and a chamber pressure of 50 kPa.
B Powder	SB Boron Corp	Elemental amorphous B – SB Boron 95, <10 μm particle size, >95% boron purity
LDPE sheet	Visqueen	Thickness of 0.11 mm.

The surface of the UHMWPE fabric was treated with an argon plasma at atmospheric conditions to modify the surface roughness and improve wettability between the fabric and the LDPE binder. The intent was to obtain superior interfacial bonding between the UHMWPE fabric and the B coated C fabric. The expected superior bonding was to prevent premature delamination. However, radiation shielding capabilities are dictated primarily by the overall composition of the composite and not by the extent of interfacial bonding. The Vacuum Plasma Spray, VPS, process was successfully used to deposit elemental B on the C fabric. In this process gases (usually Ar or N₂ as the primary gas and He, H₂ or N₂ as the secondary gas) are flowed between a tungsten cathode and a water-cooled tungsten lined, copper anode. An electric arc initiated between the two electrodes ionizes the gases creating high-pressure plasma. A measured quantity of B powder feedstock is introduced into the gas stream either just outside the torch or in the diverging exit region of the nozzle (anode). Plasma and B feedstock interactions caused melting or vaporization of B, where it was subsequently deposited on to a C fabric substrate to produce a B coating. Optical microscopy inspection indicated complete coverage of the C fabric with B.²⁴ At least three thickness measurements were performed on each C fabric and the corresponding 3 different batches of the B coated C fabric using a micrometer. The average thickness of the coated C fabric was 311 μm. The C fabric itself was 260 μm thick. The average coating thickness was calculated to be 51 μm, i.e., approximately 25 μm on either side of the C fabric, with a standard deviation of 6 μm. With this level of coating thickness and coverage uniformity the effect on neutron attenuation is discussed in Section - Neutron Shielding Test Results and Discussions.

The VPS parameters for obtaining a uniform coating and with minimal damage to the C fabric substrate were optimized to an Ar-H2 plasma at a power of 100 kW, a chamber pressure of 50 kPa, and a standoff distance of 20 cm between the plasma gun and the fabric.

A two-step pressure and temperature assisted compression molding process was used to fabricate the radiation shielding composite panel. First, a material stack was arranged with a sequence of LDPE/UHMWPE/LDPE/UHMWPE/LDPE/UHMWPE (Stack 1) and subsequently compression molded. Stack 1 was compressed at a maximum temperature of 145°C to prevent melting of the UHMWPE fabric and yet to ensure melting and infiltration of LDPE into the woven UHMWPE fabric. A second material stack consisting of boron coated carbon fabric (BC) was compression molded with two LDPE films on each side or LDPE/LDPE/BC/LDPE/LDPE (Stack 2). Stack 2 was consolidated under a load of 6 tons (on a mold area of 304.8 × 304.8 mm, resulting in a consolidation pressure of approximately 0.6 MPa) at a maximum temperature of 190°C to ensure infiltration of the B coated layer. Cooling was started after 30 minutes dwell time at maximum temperature. For the second compression molding step the overall layup sequence was:

Stack1/Stack 2/Stack1/Stack 2/Stack1/Stack 2/Stack1/Stack 2/Stack1.

A molding force of 10 tons (resulting in a pressure of approximately 1.1 MPa) was applied throughout this second molding step. The temperature was maintained at 145°C for 30 minutes before cooling. The panel had an overall thickness of 4.4 mm after molding.

A typical fabricated panel is shown in Figure 1(a) and the machined pieces used for radiation testing are shown in Figure 1(b). The different steps in the fabrication sequence are illustrated in Figure 2. To evaluate the effects of the LEO environment a 15.25 × 10.16 × 0.64 cm³ composite panel was attached to the outside of the International Space Station (ISS) on the Materials International Space Station Experiment (MISSE) platform. For this mission, MISSE-13, the composite panel faced the Zenith orientation (opposite side to earth) with maximum exposure to solar radiation. The panel stayed on orbit for approximately 10 months. This paper discusses the simulation results for GCR and secondary neutron shielding efficacy. Neutron test results prior to LEO exposure and after return to earth are also discussed. The performance of the composite panel against UV and AO exposure in LEO is a subject of our recent publication.²⁸

Figure 1.

(a) Compression molded polyethylene based composite panel fabricated for delivery to NASA for LEO exposure and (b) ground samples used in radiation shielding test.

Figure 2.

Fabrication steps used for radiation shielding composite panel exposed to LEO environment.

Radiation Transport Simulation

GCR Transport Simulation and Discussions

The efficacy of the composite materials for shielding GCR has been estimated through different approaches. PHITS transport code²⁹ (PHITS version 3.02) has been used to transport a GCR spectrum through varying thicknesses. The PHITS code uses physics models and cross-section databases to simulate interactions of all primary and secondary particles with atoms and nuclei in the composite. For these calculations, data libraries were used for neutron and proton interactions below 20 MeV, and the JQMD (JAERI Quantum Molecular Dynamics) quantum molecular dynamics model coupled with the GEM (General Evaporation Model) evaporation model were used to simulate all other interactions.

Following the methodology of Reference 30, a slab geometry composed of a thin water slab (0.3-mm thick) was sandwiched between two slabs of equal thicknesses to simulate an enclosed environment in space. A pencil beam of GCR ions was incident normally on the center of the first shielding slab. The BON2014 Badhwar-O’Neill GCR model³⁰ was used to generate the incident GCR spectrum, with GCR ions ranging from protons up to Fe, and energies up to 10 GeV/nucleon. A solar minimum GCR spectrum was used for all calculations. The shielding slabs were cylinders with a diameter of five m and were made of either aluminum or composite material. The elemental composition of the composite architecture was based on the number of UHMWPE fabrics, C fabrics, LDPE binder sheets, and total thickness of ¹⁰ B coating, and was determined to be C_6.88H_8.10B_0.86. Natural isotopic abundances for each element were used for the material definitions in the calculations.

The resulting dose in the thin water slab has been determined via the T-deposit tally in PHITS, which records the energy deposition in the volume from primary and secondary particles, including neutrons. Energy deposition was recorded in units of MeV, which after conversion to Joules and then divided by the mass of the water slab, yielded the dose in units of Gy per source GCR particle. The linear energy transfer (LET) of each particle depositing dose was used in the code to determine the dose equivalent using the current ICRP quality factors.

Simulations were run for composite shielding and aluminum shielding at 0, 2, 4, 6, 8, 10, 20, 30, 40, 50, 60 and 100 g/cm². Figure 3 shows the dose (Gy/GCR source ion) for both the composite and aluminum as a function of areal density. In all following figures thickness is represented as areal density as opposed to linear dimensions. This is to account for the varying densities of different materials and for a more accurate representation of the shielding effectiveness, when considering the different densities of materials and the way radiation interacts with them. Statistical uncertainties vary between 0.5 and 1%. Figure 4 shows the dose equivalent integrated over 1 year of solar minimum conditions as a function of aluminum and composite areal densities. Statistical uncertainties for the dose equivalents vary between 1 and 3%. At 20 g/cm², it has been shown that the decrease in dose equivalent reaches a plateau.²⁹ At this areal density, the composite material reduces the dose compared to the aluminum dose by 16% (from 1.72 × 10⁻¹⁵ to 1.45 × 10⁻¹⁵ Gy per source ion). At 20 g/cm², the composite shield reduces the dose equivalent by 44% relative to the dose equivalent behind aluminum shielding (from 0.362 Sv/y to 0.203 Sv/y). Even at a lower thickness of 10 g/cm², the composite is 30% more effective than Al from a dose equivalent standpoint. The composite material has greater mass stopping power than aluminum, and because the composite material produces fewer light secondary particles (protons, deuterons, tritons, ³He, ⁴He, and neutrons), the reduction in dose is also augmented with a reduction in average LET, leading to a greater reduction in dose equivalent.

Figure 3.

Calculated dose in a thin water slab between two shielding slabs of equal dimensions composed of aluminum or composite as a function of areal density (shielding thickness).

Figure 4.

Calculated dose equivalent in 1 year of solar minimum conditions as a function of aluminum and composite areal density (shielding thickness). Statistical uncertainties are not shown but are less than 2.5%.

In addition to comparisons between the composite material and aluminum, calculations were also conducted using polyethylene at the same areal densities used for the other two materials. Figure 5 shows the ratio of the composite dose to the polyethylene dose (blue) and the ratio of the composite dose equivalent to the polyethylene dose equivalent (orange). The general trend in the ratio of the doses suggests that the composite dose is 1 to 2% higher than the polyethylene dose. The ratio for dose equivalent indicates that the composite dose equivalent is 2 to 15% higher, with a general trend indicating that the ratio increases as the shielding thickness increases. The fact that the doses are nearly the same, but the dose equivalents are higher for the composite material, indicating that there is a larger fraction of high-LET particles contributing to the dose behind the composite than with polyethylene. Polyethylene has a larger fraction of hydrogen atoms than the composite, which in turn leads to fewer high-LET target fragments contributing to the dose. The relatively low numbers of heavy ions in the GCR spectrum and corresponding relatively low number of heavy ions in the simulation but larger contribution to dose equivalent leads to larger uncertainties in the comparison of dose equivalent when compared to dose.

Figure 5.

Calculated ratio of the composite dose to the polyethylene dose (blue) and the ratio of the composite dose-equivalent to the polyethylene dose-equivalent (orange) as a function of areal density (shielding thickness). Statistical uncertainties are shown.

Based on PHITS transport calculations, the composite material provides better shielding in a free-space GCR environment than does aluminum. When compared to polyethylene, the composite material displays similar shielding qualities, with nearly identical dose reduction. As discussed previously the design and fabrication of the composite was driven by optimum combination of radiation shielding and multifunctionality. The slightly higher dose and dose equivalent compared to HDPE is balanced significantly by higher mechanical strength and neutron attenuation.

Neutron Transport Simulation and Discussions

The contribution to the dose and dose-equivalent from neutrons can be significant. As highly penetrating particles, they present a challenge in reducing the dose in space. Because neutrons in space come solely from primary radiation interactions in shielding and other materials, effective shielding material both reduces the number of secondary neutrons produced by those interactions as well as absorb those neutrons after they are produced.

To study the dependence of the neutron energy spectrum on shielding material, PHITS simulations were run with an incident, isotropic GCR spectrum on a 6-m diameter spherical shell and a material thickness of 20 g/cm². Shielding in the range of 20-30 g/cm² provides optimal shielding for GCR,²⁹ and the choice of 20 g/cm² in these simulations represents a realistic shielding scenario. Neutrons were scored as they entered a 30-cm diameter water sphere placed in the center of the outer spherical shell. Figure 6 shows the neutron energy spectra for 20 g/cm² of aluminum, polyethylene, and composite in units of number of neutrons per incoming GCR ion. Statistical uncertainties in the calculations range between 1 and 5% except at the points of lower yields where the error bars are visible. Energies range from thermal to 10 GeV. Above 100 MeV, there is no difference between the three materials. There is a clear difference between aluminum and the other two materials below 100 MeV. Aluminum has a higher yield of neutrons between 100 keV and 100 MeV, the energy where neutron weighting factors are largest. Below 100 MeV, the effects of neutron moderation can be seen with both polyethylene and composite shielding, where higher energy neutrons have been shifted down to lower energies. Below 1 eV, the boron-loaded composite reduces the number of thermal and epithermal neutrons relative to the polyethylene shield.

Figure 6.

Calculated neutron energy spectrum from aluminum, composite and polyethylene. Spectra are given in units of number of neutrons per incoming GCR ion.

Table 2 shows the total number of neutrons per incoming GCR ion (column 2) and the total neutron effective dose in pSv per incoming ion for the three shielding materials. Statistical uncertainties on the neutron yields are less than 1%. Uncertainties on the dose equivalent per incoming ion are shown in the table. Both polyethylene and the composite reduce the overall neutron fluence and effective dose by a significant fraction relative to aluminum shielding.

Table 2.

Total neutron yield (number per incoming GCR ion) and total neutron effective dose (pSv per incoming ion) as a function of shielding material.

Material	Total neutrons per GCR ion	pSv per incoming ion
Aluminum	0.0118	2.86 ± 0.06
Polyethylene	0.0051	1.51 ± 0.03
Composite	0.0058	2.015 ± 0.040

Neutron Shielding Measurements

The neutron attenuation properties of the composite panel were measured at the University of Tennessee’s PuBe neutron irradiation facility, located in the Science and Engineering Research Facility on the Knoxville campus. The PuBe source emits neutrons isotropically, and as such the source was placed inside a “howitzer”, as shown schematically in Figure 7(a), to create a beam of neutrons that stream out of a bore hole in the device. The howitzer is a cylinder 76.2 cm in diameter and 1.2 m in length filled with a solid hydrocarbon material, except for a 5.08 cm diameter hole bore in the center to a depth of 61 cm. The PuBe source is placed in the bottom of the bore. Neutrons spanned the range of energies between thermal and approximately 10 MeV. The measurement campaign was broken into two halves – one for “high” energy neutron above approximately 2 MeV, and another one for energies below 2 MeV.

Figure 7.

(a) A schematic (not to scale) of the howitzer which holds the PuBe source and creates a stream of neutrons exiting the top of the device. The howitzer is filled with a solid hydrocarbon material shown in blue that moderates and attenuates neutrons emitted from the source, except along the bore hole. The red dot indicates the position of the PuBe source; (b) View from the top looking down into the 1.9 cm (0.75 inch) bore polyethylene blocks. The polyethylene block on top has a 5.08 cm (2 inches) diameter bore; (c) Composite shielding blocks placed inside the 5.08 cm bore polyethylene block, covering the 1.9 cm (0.75 inch) diameter hole at beam exit; (d) Experimental setup with the neutron detector (EJ 301 liquid scintillator) on top of the polyethylene blocks, which in turn are on top of the howitzer.

Since the composite samples were 2.54 × 2.54 cm², and as such, additional polyethylene blocks with a 1.9 cm diameter hole in the center were placed on top of the bore hole to create a neutron beam 1.9 cm diameter. Figure 7(b) shows a picture of a top-down view of the 1.9 cm bore polyethylene blocks with an additional polyethylene block with a 5.08 cm bore hole on top of the 1.9 cm bore blocks. During testing, various numbers of shielding samples were taped together to form a thick shielding block, and were then placed inside the top polyethylene block and covered the 1.9 cm (¾”) bore hole when in place, as shown in Figure 7(c). Figure 7(d) shows the position of the fast neutron detector when taking measurements.

The neutron detector used to measure the fast neutron flux was an EJ-301 liquid scintillator capable of discriminating neutron events from gamma-ray events in the detector. With this detector, neutrons above 2 MeV were measured and recorded for each shielding thickness, along with a run with no composite shielding in place. Measurements of neutrons below 2 MeV were performed with the use of Bonner spheres. The diameters of the polyethylene Bonner balls were 5.08 cm (2 inches), 7.62 cm (3 inches), 12.7 cm (5 inches), and 20.32 cm (8 inches), with an additional measurement made with no sphere. The detector that was used was a Ludlum BF₃ detector. A Bonner sphere measurement will yield a neutron spectrum covering thermal neutron energies up to a maximum of 2 MeV when the largest sphere used is 20.32 cm (8 inches) in diameter. A simple unfolding technique was used that incorporated the measurements made with the five configurations (no sphere, 5.08 cm, 7.62 cm, 12.7 cm, and 20.32 cm) that produces a five-point spectrum covering the energy range between thermal and 2 MeV. The five points roughly cover the following energy ranges:

• Point 1 (plotted at 0.1 eV) – thermal up to 2 eV

• Point 2 (plotted at 10 eV) – 2 eV up to 100 eV

• Point 3 (plotted at 200 eV) – 100 eV up to 50 keV

• Point 4 (plotted at 100 keV) – 50 keV up to 400 keV

• Point 5 (plotted at 1 MeV) – 400 keV up to 2 MeV

Measurements were made at areal densities of 5.2 g/cm², 10.4 g/cm², 15.5 g/cm² and 18.2 g/cm².

Neutron Shielding Test Results and Discussions

Data were taken with the liquid scintillators for 30 minutes using 0 (unshielded), 4.8, 9.6, 14.5 and 16.9 g/cm2 of composite shielding in place. Figure 8 shows the neutron energy spectrum above 2 MeV. Figure 9 shows the resulting neutron spectra from the Bonner sphere measurements, with shielding thicknesses of 0 (unshielded), 2.4, 4.8, 7.2, 9.6, and 14.5 g/cm2.

Figure 8.

Unfolded neutron energy spectra normalized to the number of incident neutrons per 30 minutes at the indicated neutron energies. The spectra are shown after passing through 4.8, 9.6, 14.5, and 16.9 g/cm² of composite shielding, as well as no shield. Statistical uncertainties vary between 1 and 4%. The error bars show the 15% systematic uncertainty based on efficiency uncertainties for the liquid scintillators used in these measurements and the uncertainty in the unfolding technique.

Figure 9.

Unfolded neutron energy spectra as measured with a Bonner Sphere spectrometer behind the indicated shielding thicknesses in units of g/cm². The uncertainties in energy are shown with the unshielded data and are the same values for the measurements conducted at the indicated thicknesses of composite shielding.

The high energy spectra in Figure 8 show the effects of moderation as the shielding thickness increases. The moderation at these energies is dominated by elastic scattering off the atomic constituents of the shielding (hydrogen, carbon, boron). The elastic scattering cross sections are energy-dependent, especially for carbon, which shows several strong resonances between 1 and 10 MeV. As a result, the amount of moderation and subsequent loss of neutron flux is energy dependent, which in turn results in the varying degrees of attenuation as a function of energy seen in Figure 8. The peaks around 3.5-4 MeV and 5-5.5 MeV that are typical of unshielded PuBe sources are present, but as shielding thickness increases, the peaks become less pronounced by the combination of loss of flux at those energies due to moderation, and some feeding into those energies by moderation of higher energy neutrons. Because the flux generally decreases as neutron energy increases, the loss of flux from moderation is generally greater than the gain in flux from the moderation of higher energy neutrons.

Below 1-MeV, an interesting feature is seen with the shielding in place (Figure 9). For example, a minimum in the flux between 100 eV and 1 keV is observed that is relatively constant, with equal loss of flux via moderation and gain of flux via moderation of higher energy neutrons. Moderation via the capture of neutrons by ¹⁰B via the n(¹⁰B,⁷Li)⁴He reaction becomes significant below 0.1 MeV, with a cross section larger than any other reaction cross sections from the other atomic constituents. As such, the advantage of a boron-loaded material can be seen at the lowest energies. As neutron energy decreases, the boron capture cross section increases, resulting in a greater reduction in neutron flux as neutron energy decreases.

Tables 3 and 4 list the fractional attenuation (equal to the unshielded flux minus the shielded flux, then divided by the unshielded flux) for the high-energy, liquid scintillator data and the Bonner sphere data, respectively. The statistical uncertainties in Table 3 are less than 1% and the systematic uncertainties are estimated to be 15%. The uncertainties shown in Table 4 are the statistical uncertainties.

Table 3.

Fraction of high-energy flux lost relative to the unshielded flux as a function of shielding thickness and neutron energy range. Data were taken with the liquid scintillators. Statistical uncertainties are less than one percent, systematic uncertainties in detection efficiency and dose are on the order of 15-20%.

Energy range (MeV)	4.8 g/cm²	9.6 g/cm²	14.5 g/cm²	16.9 g/cm²
2.5 - 3	0.56	0.73	0.82	0.83
3 - 3.5	0.49	0.73	0.78	0.78
3.5 - 4	0.45	0.61	0.67	0.69
4 - 4.5	0.35	0.63	0.72	0.72
4.5 - 5	0.35	0.58	0.64	0.63
5 - 5.5	0.37	0.44	0.57	0.6
5.5 - 6	0.34	0.54	0.61	0.6
6 - 6.5	0.21	0.47	0.52	0.51
6.5 - 7	0.41	0.48	0.55	0.6
7 - 7.5	0.3	0.46	0.6	0.63
7.5 - 8	0.31	0.49	0.59	0.58
8 - 8.5	0.41	0.5	0.59	0.59
8.5 - 9	0.33	0.43	0.54	0.57
9 - 9.5	0.22	0.3	0.42	0.42

Table 4.

Fraction of low-energy flux lost relative to the unshielded flux as a function of shielding thickness and neutron energy range. Data were taken with the Bonner sphere spectrometer.

Energy (MeV)	2.4 g/cm²	4.8 g/cm²	9.6 g/cm²	14.5 g/cm²
1.00 E-07	0.42 ± 0.11	0.51 ± 0.14	0.56 ± 0.16	0.69 ± 0.23
1.00 E-05	0.23 ± 0.09	0.28 ± 0.11	0.28 ± 0.11	0.56 ± 0.26
2.00 E-04	0.12 ± 0.06	0.04 ± 0.02	0.19 ± 0.09	0.42 ± 0.23
1.00 E-01	0.42 ± 0.12	0.45 ± 0.13	0.59 ± 0.19	0.64 ± 0.22
1.00 E + 00	0.29 ± 0.08	0.28 ± 0.12	0.51 ± 0.15	0.60 ± 0.19

As expected, the attenuation increases as shielding thickness increases, but as the thickness approaches 20 g/cm², the loss of neutron flux begins to level off. This trend is also seen in the rates of GCR dose attenuation with increasing shielding thickness in Reference 29 and the calculations reported here. The data also supports the features seen in Figure 6, where the addition of boron greatly aids in the reduction of thermal and epi-thermal neutrons.

At energies above 1 MeV (Table 3), interactions are dominated by elastic scattering, and as such, energy-dependent interaction lengths (the thickness of material where the transmitted neutron flux is equal to 1/e of the incident flux) can be used to approximate a functional relationship between transmitted neutron flux and shielding thickness. For example, for energies between 2.5 and 3.0 MeV, a fit to the first three thicknesses (before the plateau at 16.9 g/cm²) gives the transmitted neutron flux = 0.67e^−0.092x, where x is the thickness in g/cm².

Summary

A composite comprised of UHMWPE fabric, boron coated carbon fiber, and LDPE was fabricated to shield against GCRs and secondary neutrons. Using optimized vacuum plasma spray parameters, it was demonstrated that a uniform coating of ¹⁰B on the C fabric can be achieved. The proposed composite architecture was evaluated using the PHITS transport simulation code. The composite material was shown to have a greater mass stopping power than aluminum, and since it produces fewer secondary particles, the reduction in dose is also augmented with a reduction in average LET, leading to a greater reduction in dose equivalent. Assuming worst case solar minimum conditions it was determined that for the same areal density of 20 g/cm² the composite reduced the dose equivalent by 44% relative to the dose equivalent behind aluminum shielding.

Secondary neutron simulation results indicated a clear difference between aluminum and the composite below 100 MeV. Aluminum has a much higher yield of neutrons between 100 keV and 100 MeV compared to the composite. This is critical since the neutron weighting factors are highest in this energy range. Below 100 MeV, the effects of neutron moderation are evident with the composite shielding, where higher energy neutrons have been shifted down to lower energies. Below 1 eV, the boron-loaded composite reduces the number of thermal and epi-thermal neutrons even relative to the polyethylene shield. The relevance of such composites therefore for applications on the Lunar surface is evident.

Neutron testing performed in the energy range 1 – 10 MeV using PuBe source confirmed the simulation results. Moderation of high energy neutrons as the thickness of the composite increased is clearly evident from the liquid scintillator data. Below 1-MeV, the Bonner sphere data reveals an interesting feature in that there is a minimum in the flux between 100 eV and 1 keV that is relatively constant within the range of shield thicknesses tested here, with equal loss of flux via moderation and gain of flux via moderation of higher energy neutrons. The test data also confirmed the advantage of the boron-loaded composite as is evident at the lowest energies, where the boron capture cross sections are greatest and the reduction of flux is also the greatest. This study can provide a potential material solution to the safe long duration human exploration beyond LEO.

Footnotes

Acknowledgements

The authors are grateful to NASA SBIR Contract # 80NSSC18C0169 for funding this investigation. The authors gratefully acknowledge the technical mentorship of Dr Sheila A. Thibeault and Dr Keith Gordon of NASA Langley Space Center and Miria Finckenor of NASA Marshall Space Flight Center, during this investigation.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the NASA SBIR (# 80NSSC18C0169).

ORCID iDs

Yongzhe Yan

Haibin Ning

References

National Academies of Sciences E, Medicine . Space radiation and astronaut health: managing and communicating cancer risks. 2021.

NCRP . Guidance on radiation received in space activities: NCRP Report No. 98; 1988.

Dobynde

Shprits

Drozdov

, et al. Beating 1 sievert: optimal radiation shielding of astronauts on a mission to Mars. Space Weather 2021; 19(9): e2021SW002749.

Dahche

Kaul

Barghouty

. Radiation transport properties of polyethylene-fiber composites. 2003.

Guetersloh

Zeitlin

Heilbronn

, et al. Polyethylene as a radiation shielding standard in simulated cosmic-ray environments. Nucl Instrum Methods Phys Res Sect B Beam Interact Mater Atoms 2006; 252(2): 319–332.

Ning

. Thermoplastic composites: principles and applications. Walter de Gruyter GmbH & Co KG, 2021.

Wang

, et al. Upcycling polyethylene to high‐purity hydrogen under ambient conditions via mechanocatalysis. Angew Chem Int Ed Engl 2025; 64: e202417644.

Wilson

. Overview of radiation environments and human exposures. Health Phys 2000; 79(5): 470–494.

Shea

Smart

. A summary of major solar proton events. Sol Phys 1990; 127: 297–320.

10.

Parnell

Watts

John

, et al. Radiation effects and protection for moon and mars missions. Space 981998: 232–244.

11.

NCRP . Relative biological effectiveness of radiations of different quality: NCRP report No. 104; 1989.

12.

Hayatsu

Hareyama

Kobayashi

, et al. HZE particle and neutron dosages from cosmic rays on the lunar surface. J Physical Soc Japan 2009; 78: 149–152.

13.

Hayatsu

Kobayashi

Yamashita

, et al. Environmental radiation dose on the moon. Astroparticle, particle and space physics, detectors and medical physics applications. World Scientific. 2008; 792–796.

14.

Heilbronn

Townsend

Braley

, et al. (eds). Secondary neutron production from space radiation interactions: advances in model and experimental data base development. In: 2002 Microgravity Materials Science Conference, 2003.

15.

Wilson

Simonsen

Shinn

, et al. Astronaut exposures to ionizing radiation in a lightly-shielded spacesuit. 1999.

16.

Woolum

Burnett

Furst

, et al. Measurement of the lunar neutron density profile. Moon 1975; 12: 231–250.

17.

Zhang

Wimmer-Schweingruber

, et al. First measurements of the radiation dose on the lunar surface. Sci Adv 2020; 6(39): eaaz1334.

18.

Cucinotta

Townsend

Wilson

, et al. Light ion components of the galactic cosmic rays: nuclear interactions and transport theory. Adv Space Res 1996; 17(2): 77–86.

19.

Wilson

. Galactic and solar cosmic ray shielding in deep space: National aeronautics and space administration. Langley Research Center, 1997.

20.

Herrman

Baxter

Mishra

, et al. Mechanical characterization of polyethylene-based thermoplastic composite materials for radiation shielding. Compos Commun 2019; 13: 37–41.

21.

Lin

, et al. The advancement of neutron shielding materials for the storage of spent nuclear fuel. Science and Technology of Nuclear Installations 2021; 2021(1): 1–13.

22.

Harrison

Weaver

Bertelsen

, et al. Polyethylene/boron nitride composites for space radiation shielding. Applied Polymer. 2008; 109: 2529–2538.

23.

Shin

Lee

, et al. Polyethylene/boron-containing composites for radiation shielding. Thermochim Acta 2014; 585: 5–9.

24.

Laurenzi

Zaccardi

Toto

, et al. Fused filament fabrication of polyethylene/graphene composites for In-Space manufacturing. Materials 2024; 17: 1888.

25.

Zaccardi

Toto

Rastogi

, et al. Impact of proton irradiation on medium density Polyethylene/carbon nanocomposites for space shielding applications. Nanomaterials 2023; 13: 1288.

26.

Laurenzi

Zanet

Gde.

Santonicola

. Numerical investigation of radiation shielding properties of polyethylene-based nanocomposite materials in different space environments. Acta Astronaut. 2020; 170: 530–538.

27.

Unagar

Makwana

Barala

, et al. Investigation of HDPE composites with B4C and LiH fillers for space radiation shielding applications in LEO. Adv Space Res 2025; 76(6): 3717–3725.

28.

Sen

O’Dell

Yan

, et al. Space environmental effects on multifunctional radiation shielding materials. Earth Space Sci 2024; 11(11): e2024EA003681.

29.

Sato

Iwamoto

Hashimoto

, et al. Features of particle and heavy ion transport code system (PHITS) version 3.02. J Nucl Sci Technol 2018; 55(6): 684–690.

30.

Slaba

Bahadori

Reddell

, et al. Optimal shielding thickness for galactic cosmic ray environments. Life Sci Space Res 2017; 12: 1–15.

Study of radiation shielding efficacy of polyethylene and polyethylene-based composite against galactic cosmic rays and secondary neutrons

Abstract

Keywords

Introduction

Materials and Methods

Radiation Transport Simulation

GCR Transport Simulation and Discussions

Neutron Transport Simulation and Discussions

Neutron Shielding Measurements

Neutron Shielding Test Results and Discussions

Summary

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iDs

References