Innovative neutron shielding materials composed of natural rubber-styrene butadiene rubber blends, boron oxide and iron(III) oxide

Optimized flexible and lightweight neutron shielding materials were designed using the Monte Carlo N-Particle (MCNP) code. Thicknesses of 10 mm and 100 mm were tested for neutron shielding performances. Simulation results indicated that the 10 mm shielding material of natural rubber (NR) and styrene butadiene rubber (SBR) blend (1:1) with 60 part per hundred rubber (phr) boron oxide (B2O3) and 100 mm shielding material with four alternating layers of NR with 100 phr iron (III) oxide (Fe2O3) and of NR and SBR blend (1:1) with 10 phr B2O3 were most suitable for thermal neutron shielding and all-energy neutron shielding, respectively. Experimental results verified the shielding efficiency of these optimal designs and ease of fabrication.


Introduction
Neutron radiation is widely utilized in medical, agricultural and industrial applications. As neutrons have no electrical charge, they can pass through an electron cloud to interact directly with the atomic nucleus and transfer most of their energy to the nucleus with a mass close to that of a neutron. As the human body has a high water content, there is a high density of hydrogen atoms present. The mass of the hydrogen nucleus is close to that of the neutron. Therefore, exposure to free neutrons can be hazardous, since the interaction of neutrons with molecules in the body can cause disruption to molecules and atoms as well as cause reactions which give rise to other forms of radiation such as protons with high Linear Energy Transfer (LET) [1], [2]. Neutron shielding requires materials with an atomic mass close to that of the neutron. For example, hydrogen-rich materials are often used to shield against neutrons, since ordinary hydrogen both scatters and slows down neutrons, after which neutrons may be absorbed by an isotope exhibiting high affinity for adsorbing slow neutrons such as boron.
There have been several studies of polymeric composition containing materials with high neutron cross section to determine their thermal neutron shielding properties. Materials made of polyethylene or epoxy resin mixed with boron compound are widely used as neutron shielding materials, but they usually exhibit sizable weight and volume, which results in low-flexibility performances [3], [4]. 1 To whom any correspondence should be addressed. Flexible polymers such as natural rubber (NR) have been used as a matrix for neutron shielding materials, but they only shield against thermal neutrons [5], [6]. There is considerable natural rubber cultivation in many areas of Thailand. This rubber is a hydrocarbon polymer containing a large number of hydrogen atoms and can be easily fabricated into flexible materials. In addition, borate minerals such as boracite, or boron compounds, are found in the northeast of Thailand. Thus, not only is the production of neutron shielding materials comprising natural rubber/ SBR blends and a borate material useful for neutron-related work, but it also utilizes domestic resources.
The most effective neutron shielding material can be obtained by appropriately mixing high hydrogen-content materials, heavy elements and thermal neutron absorbers. High hydrogen-content materials can undergo elastic scattering with fast and intermediate-energy neutrons. Heavy elements can undergo inelastic scattering with fast neutrons and can attenuate the secondary gamma ray emission. Finally, neutron absorbers function to greatly reduce the number of thermal neutrons.
In this experimental work, rubbers, which are low-cost hydrocarbon polymers that can easily be fabricated into many different products and shapes, were selected as the substrate of the composite together with a very low-cost and often disregarded neutron absorbing material, B2O3. Fe2O3 (commonly known as rust), a very low-cost and usually disregarded material, was also utilized as the heavy element in the composite.
The method of calculation for neutron shielding can be found in references [7], [8]. This research incorporated inelastic scattering and shielding of secondary gamma ray emissions into the simulation. The compositions of neutron shielding materials were optimized by Monte Carlo simulations using the MCNP5 code to examine the neutron flux encompassing the entire range of interaction energy and accounting for secondary gamma ray emissions.
Neutron attenuation experiments were carried out to verify neutron and secondary gamma ray shielding performances of the designed neutron shielding materials using a simulated neutron source (energy 1x10 -8 -100 MeV) and an Am-Be neutron source.

Material selection and MCNP simulation
The main raw materials were selected from simple elements or compound materials, taking into consideration cost, radiation shielding performances and physical and thermal properties. The main compositions of the selected neutron shielding material to shield against the entire range of neutron energy and secondary gamma ray emission are listed in Table 1. They consisted of natural rubber (NR), styrene butadiene rubber (SBR-1502 from BST elastomer Co., Ltd. Thailand) and neutron absorber/scattering agent [boron oxide (B2O3), iron(III) oxide (Fe2O3) and lead dioxide (PbO2)]. 40 phr of each agent was used for each sample because preliminary fabrication indicated that at 40 phr, rubber still exhibited desirable properties. The lead or iron content is desirable for inelastic neutron scattering to effectively bring the neutron energy down to lower-energy regions. In addition, lead and iron are used for attenuation of gamma radiation. Neutrons in the lower-energy regions can be moderated by hydrogen and other light elements by elastic scattering. In the thermal energy region, boron (as B2O3) is used for neutron absorption. Because the interaction between a thermal neutron and boron-10 produces a secondary gamma ray from the following interaction: Thermal neutron + B-10  He-4 + Li-7* (478 keV gamma ray), without the gamma ray attenuating material (Pb or Fe), this neutron shielding material would itself become a significant source of gamma radiation.
For each sample in table 1, the input card of the MCNP5 transport code was appropriately prepared for testing and comparison of shielding properties of different materials. A simplified model was employed: a single radiation source, an isotropic emission of simulated neutron source with energy 1x10 -8 -100 MeV, was located at the center of a sphere filled with the shielding material as shown in . The spherical model of radius 10 mm or 100 mm was chosen for neutron shielding in the thermal region or in the entire-energy range, respectively, to ensure that all neutron particles interacted with the shielding material at least one time before leaving the system. For each shielding material, the dose equivalent rate per source strength was estimated on the outer surface of the sphere. and then compression molded by an electrically-heated hydraulic press at 160±2 ºC and 4 MPa for a suitable duration of each sample. The homogeneity of each shielding material sample was verified by neutron radiography at the Thai Research Reactor TRR-1/M1, which is a TRIGA-type reactor. A neutron imaging plate (ND 2040) attached to the shielding material by an aluminium tape was exposed to the neutron beam for 4 minutes. The neutron flux and the Cd ratio at the position of the shielding material were approximately 9 x 10 5 n cm -2 s -1 and 100, respectively. The neutron imaging plate was processed by an image reader (BAS 2500).

Experimental work on neutron attenuation
The neutron attenuation experiment was carried out at the Radiation Measurement Laboratory at the Nuclear Engineering Department, Faculty of Engineering, Chulalongkorn University to assess neutron shielding performances of the selected shielding materials shown in table 2. Shielding materials were slab shaped with thicknesses of 10 mm and 100 mm and dimensions of 150 mm x 150 mm. These slabs were stacked together to a height of 20 cm. The Am-Be neutron source whose strength was 3.08 x 10 3 n cm -2 s -1 was placed at the bottom of the polyethylene (PE) source holder with a distance of 40 mm from the neutron shielding material, as illustrated in figure 2. The neutron measurement was performed using a NE-905 glass scintillation neutron detector connected to a counting system.

MCNP Simulation
Results from the MCNP transport code revealed that the 10-mm thick sample #2 (NR+SBR+B2O3) and 100-mm thick sample #10 (4 alternating layers of NR+Fe2O3/NR+SBR+B2O3) exhibited excellent neutron and secondary gamma ray shielding performances, as seen in figures 3 and 4. For sample #2, because the thickness of 10 mm was smaller than the mean free path of fast neutrons in rubber, only thermal neutrons could be absorbed by this material, which also had the highest boron content. For the 100-mm thick sample #10, fast neutrons could undergo inelastic scattering with iron and elastic scattering with hydrogen present in the rubber to become thermal neutrons and finally be absorbed by boron. Thus, sample #2 and sample #10 are suitable for thermal neutron shielding and all-energy neutron shielding, respectively.  Figures 5(a) and 5(b) illustrate the neutron shielding efficiencies at different energy regions. For sample #2, the best shielding efficiency occurred in the thermal energy range (1x10 -8 -1x10 -2 MeV) because boron exhibits a high thermal neutron absorption cross-section. In the fast neutron regime (1x10 -1 -1x10 2 MeV), for 100-mm thick shielding designs, although not clearly visible in Figure 5 sample #10 outperformed sample #2 because the total neutron dose rate equivalent was lower in the entire fast neutron energy range. Moreover, for sample #10, it also performed better than the NR/SBR rubber compound (sample #1) and NR/SBR/Fe2O3 (sample #4) in the thermal energy range due to the presence of the boron compound.
To further determine the optimized amount of B2O3 and Fe2O3 additions, a simulation of a 10 mm thick sample #2 with varying amounts of B2O3 added from 0 to 70 phr was performed. Results are shown in figure 6(a). As more B2O3 was added, the total dose rate equivalent became reduced and appeared to saturate after about 60 phr. Thus, the most appropriate B2O3 addition for sample #2 was chosen to be 60 phr. Moreover, a simulation of a 100 mm thick sample #10 with varying amounts of B2O3 added from 0 to 70 phr and Fe2O3 added from 0 to 100 phr was performed. The reason why the amount of Fe2O3 added was limited to 100 phr was because during a trial fabrication test, the rubber became very rigid at 100 phr and more, making it much less flexible. Results in figure 6(b) revealed that the most appropriate amounts were 100 phr of Fe2O3 and 10 phr of B2O3.

Fabrication of rubber compound
The optimized designs according to the MCNP simulation were fabricated by trial and error. The successfully-fabricated shielding materials with slab geometry are shown in figures 7(a) and 8(a). Homogeneity was one of several factors determining the suitability of the fabricated shielding materials. In reality, for the rubber compounds of samples #2 and #10, if they were left exposed to moisture in room condition, B2O3 would quickly react with water vapour in the atmosphere, turning into boric acid crystals, causing a problem with homogeneity. Thus, the samples must be completely encased in a thin layer of conventional rubber compound. Therefore, figure 7(a) represents the rubber compound encased in the green conventional rubber. In figure 8(a), because the rubber with iron oxide didn't require any encasement, the red slab represented the actual rubber compound. The corresponding neutron radiographs displayed in figures 7(b) and 8(b) exhibiting a very even color tone indicated that the neutron absorbing/scattering materials (B2O3 or Fe2O3) can be homogenously dispersed in the rubber compounds of NR and SBR.

Neutron transmission test
From the neutron transmission test, the 10 mm rubber compound with 60 phr of B2O3 can reduce the neutron dose from the Am-Be source by 53.91%. The 100 mm rubber compound composed of 4 alternating layers with 100 phr Fe2O3 and 20 phr B2O3 can reduce the neutron dose by 72.56%.

Conclusion
Optimized flexible and lightweight neutron shielding materials were designed using the MCNP code. The neutron shielding materials with thicknesses of 10 mm and 100 mm were examined for neutron shielding performances. Simulation results indicate that the 10 mm shielding material of NR and SBR blend (1:1) with 60 phr B2O3, and the 100 mm shielding material having 4 alternating layers of NR with 100 phr Fe2O3 and of NR and SBR blend (1:1) with 10 phr B2O3 were most suitable for thermal neutron shielding and all-energy neutron shielding, respectively. Experimental results verified the shielding efficiency of these optimal designs and the ease of fabrication. The designed shielding materials are highly suitable for applications in nuclear science and technology.