2.1 Preparation of B₄ C dispersion

The purity of B₄ C (Mudanjiang Qianjin Boron Carbide Co., Ltd.) was greater than 99.99%. Commercially available B₄ C powders (3.6 µm, 2.45 g·cm^-3) were pretreated using ball milling at 300 rpm for 12 h. The mass ratio of the corundum ball and material was 2:1. The mass ratio of different diameters of corundum balls was ø20:ø10:ø6=1:3:6. All the other chemicals and analytical reagents were used as received. The B₄ C dispersion consisted of B₄ C, distilled water, ammonia (analytical reagent (AR), Aladdin) and dispersant (Two nekal, BX, Aladdin) with a mass ratio of 1:1:0.5:0.01. Ammonia was used to adjust the pH value to enhance the stabilization of the B₄ C dispersion. The dispersant BX was used to increase the wettability of the B₄ C particles and to prevent the re-agglomeration and settlement of B₄ C particles.

2.2 Preparation of pre-vulcanization NRL

The pre-vulcanization composition of the NRL is provided in Table 1. All the chemicals and analytical reagents were used as received. The compounding agent included vulcanizater (S, AR, Aladdin), active agent (ZnO, AR, Aladdin), accelerator (zinc diethyl dithiocarbamate, AR, Aladdin), antioxidant (D, N-Phenyl-2-naphthylamine, AR, Aladdin), and dispersant BX.^[25] The mass ratio of the compounding agents and distilled water was 1:1.5. The compounding agent dispersion was prepared via milling at 300 rpm for 12 h. The pre-vulcanized latex consisted of NRL (60 wt.% DRC HA, TVRTEX, Thailand, 60 mPa·s), compounding agent dispersion, KOH (AR, Aladdin), casein (AR, Aladdin), and Peregal-O (leveling Agent O, polyoxyethylene alkyl ether). The compounding agent dispersion was slowly added to NRL by stirring. Subsequently, the solution of KOH, casein, and Peregal-O was mixed into NRL. The viscosity of NRL was controlled by adjusting the compounding agent. After curing at 60 °C for 1 h and thereafter cooling to room temperature (25 °C), the pre-vulcanized NRL was ready to use.

2.3 Preparation of B₄ C/NRL flexible films

The B₄ C dispersion was added and dispersed into the pre-vulcanized NRL by stirring. After 48 h, filtration and defoaming were performed to obtain the desired latex (5–55 wt.% B₄ C/NRL). The proper viscosity of the B₄ C/NRL mixture used for dip-molding was 60–100 mPa·s.

The B₄ C/NRL flexible films were prepared using dip-molding. Ceramic plates or molds were dipped in NRL for several seconds and thereafter allowed to set. After drying, the molds were vulcanized by boiling water and dried in an oven. Subsequently, the B₄ C/NRL flexible films were ready for thermal neutron shielding.

The thickness of the film can be controlled by dipping multiple times. After a single dip, the thickness of the film ranged from 0.1 mm to 0.35 mm. The molds were dipped multiple times to achieve the desired thicker films.

2.4 Properties characterization

The viscosity of the latex was measured using a digital viscometer (NDJ-5S, Shanghai Ping Xuan Scientific Instrument Co., Ltd.) at room temperature. The median diameter (D₅₀ ) of B₄ C was detected using a laser particle size analyzer (Mastersizer 2000) at a scanning speed of 1000 times/second with deionized water as the dispersing agent. The dispersion state of B₄ C in NRL was analyzed using a scanning electron microscope (SEM, JSM6390 LV, Japan JEOL) and an energy dispersive spectrometer (EDS). Fourier transform infrared (FT-IR) spectra were recorded using Nicolet FT-IR Nexus, in the range of 4000–500 cm^-1. The test mode was set to total reflection. Wide-angle X-ray diffraction (WAXD) was performed on a D/Max-RB target X ray diffraction (Japan Rigaku) under the following conditions: Cu Kα radiation (λ = 0.15406 nm) at the voltage of 40 kV. The scanning rate was 15 °/min in the angle range of 10° to 80°. Mechanical properties of the films including the tensile strength and elongation at break were tested using a universal testing machine (CMT6103, Meitesi industry system (China) Co. Ltd.). The harnesses of the films were measured using a shore durometer (LX-A, China Jun Ping Machinery Factory).

2.5 Thermal neutron shielding property measurement

The thermal neutron shielding property of the films was evaluated using a 49-2 swimming-pool-type reactor at the Reactor Engineering Research and Design Institute of the Chinese Atomic Energy Academy. The power of the 49-2 reactor was 3500 MW, with light water used as the coolant and moderator. Fig. 1 illustrates a schematic diagram of the thermal neutron shielding assessment.

Fig. 1Full-screen View

Schematic diagram of the thermal neutron shielding assessment

During the measurement, the sample was placed 1600 mm from the thermal column. Three Dy-Al alloy foil detectors were placed on both sides of each sample. The positions of the six foil detectors on both sides of the sample did not overlap. A cadmium foil was used for shielding against lateral neutrons of the thermal column. The diameter of Dy-Al alloy foil was 4 mm with a thickness of 0.3 mm. The performances of Dy-Al alloy foil detectors were as follows: The correction coefficient between the foil detectors was less than 1%. The Dy detector was only sensitive to thermal neutrons.

Cadmium was used to absorb thermal neutrons and a small collimator aperture was used to reduce the counting rate. A 3-cm lead brick was used to screen the γ-rays to reduce their impact. The activation cross-section of the neutrons above the thermal neutrons was very small, and could be ignored relative to the thermal neutron activation cross-section. The most probable energy of the thermal neutron was 0.0253 eV. The activity of the foil detector after activation by the neutrons was proportional to the neutron flux rate on the detector. The normalization activity differences between the front and back Dy detectors reflected the thermal neutron shielding performance of the sample. The radioactive decay equation is as follows:^[16-17]

where A₀ is the initial normalization activity, A is the normalization activity at time t, and T_½ is half-time.

The thermal neutron shielding performance of the sample was calculated using the following equation:

where A_front and A_back represent the normalization activity of the front and back Dy detectors at time t, respectively.

After A_front and A_back are obtained, the macroscopic cross-section of the actual material was calculated using Lambert–Beer law (3):^[7,26]

where d is the thickness of the material and σ is the macroscopic cross-section of the material. The term σ was obtained by measuring A_front, A_back, and d.

3.1 Dispersion state of B₄ C in NRL

Particles in commercial B₄ C powders agglomerate owing to environmental humidity. The B₄ C used was pre-treated via ball milling to avoid particle agglomeration and precipitation. The median diameter (D₅₀ ) of B₄ C varied with the milling time, which was detected using a laser particle size analyzer. As shown in Fig. 2, D₅₀ of the B₄ C particles reduced from 3.6 µm to 1 µm after milling for 12 h. The granule shape of the B₄ C particles did not change after milling. No apparent precipitation was observed after 10 days, indicating that the stability of the pre-vulcanized latex containing B₄ C is suitable for practical applications.

Fig. 2Full-screen View

Median diameter (D₅₀ ) of B₄ C particles with milling time. Inset images: Grain size distribution and SEM images of B₄ C particles (a) before milling and (b) after milling for 12 h.

The SEM images of the flexible film with different contents of B₄ C are shown in Fig. 3. The bright spots in the figure are the B₄ C particles. The small pits are the trails left behind when B₄ C particles peeled off the interface. B₄ C particles are well distributed in NRL with no pores or cracks.

Fig. 3Full-screen View

SEM images of B₄ C/NRL flexible films with (a) 5 wt.%, (b) 10 wt.%, (c) 15 wt.%, (d) 20 wt.%, (e) 25 wt.%, (f) 35 wt.%, (g) 45 wt.%, and (h) 55 wt.% B₄ C

Fig. 4 demonstrates the EDS images of the B₄ C/NRL flexible films. The small bright spots in the figure represent the B₄ C particles, indicating that they are well distributed in the NRL. The number of small bright spots increased with the B₄ C concentration. The main component of NRL is a kind of polyisoprene, and its molecular formula is (C₅ H₈ )_n. The main components of the B₄ C/NRL flexible films are B₄ C and NR. Accordingly, the main elements of the B₄ C/NRL flexible films are B, C, and H. B and C can be detected using EDS; however, H cannot be detected using EDS. Excluding the weight of H, the weights of B and C in B₄ C/NRL obtained using EDS are presented in Table 2. The mass fraction of B in B₄ C is 0.7826. The mass fraction of B₄ C in B₄ C/NRL is calculated by dividing 0.7826 by the weight of B. In Table 2, the designed contents of B₄ C are consistent with the calculated values.

Fig. 4Full-screen View

EDS images of B₄ C/NRL flexible films with (a) 5 wt.%, (b) 10 wt.%, (c) 15 wt.%, (d) 20 wt.%, (e) 25 wt.%, (f) 35 wt.%, (g) 45 wt.%, and (h) 55 wt.% B₄ C

The SEM images of the outside and inside surfaces of the B₄ C/NRL flexible films are shown in Fig. 5. The inside surface is defined as the film face adjacent to the mold during dipping. The outside surface is defined as the film face away from the mold during dipping. Both the inside and outside surfaces were smooth, and no apparent particles peeled from the matrix. This indicates that the surface of the film is in a good condition and conducive to practical applications.

Fig. 5Full-screen View

SEM images of the outside surface (a, b) and inside surface (c, d) of a B₄ C/NRL flexible film containing 25 wt.% B₄ C

3.2 WAXD of the B₄ C/NRL films

The WAXD patterns of the B₄ C/NRL flexible films are presented in Fig. 6. The peak observed at 20° revealed the amorphous feature of NR (Fig. 6a). The strongest peak of B₄ C was observed at 37.6° (Fig. 6a). The peak amplitude at 20° became weakened with an increase of B₄ C (Fig. 6b), and thus, the relative peak amplitude (i.e., the ratio of peak amplitudes at 37.6°/20°) improved (Fig. 6b). The WAXD feature of NRL became weaker with an increase in the concentration of B₄ C, whereas the WAXD feature of B₄ C became stronger. When the content of B₄ C was in the range of 25–55 wt.%, the WAXD pattern of the B₄ C/NRL flexible films was weak.

Fig. 6Full-screen View

(Color online) WAXD pattern of (a) B₄ C and NRL, and (b) B₄ C/NRL flexible films

3.3 FT-IR of B₄ C/NRL films

The FT-IR spectra of the B₄ C/NRL flexible films are shown in Fig. 7. The addition of B₄ C did not change the peak position of the IR characteristic peak of NRL, indicating that B₄ C did not change the molecular structure of NRL. The wave numbers of the B₄ C/NRL composite are summarized in Table 3. Our results are consistent with the previous reports of IR spectra of NR illustrated as follows. The wave number of nearly 3500 cm⁻¹ was attributed to the N–H stretching from a small amount of protein in the NR. The protein content decreased after curing in boiling water and subsequent washing. Therefore, the corresponding peak was not apparent in the B₄ C/NRL films. The wave numbers of 2966 cm⁻¹ and 2913 cm⁻¹ were attributed to the C–H stretching,^[27] whereas the wave numbers of 2846 cm⁻¹, 1665 cm⁻¹, and 1438 cm⁻¹ corresponded to the C–H stretching, C=C stretching, and CH₂ deformation, respectively.^[28] The wave numbers of 1535 cm⁻¹ and 1076 cm⁻¹ were attributed to the O–H bending and C–C stretching, respectively.^[29] The wave number of 1380 cm⁻¹ was attributed to the CH₃ bending.^[30] The wave numbers of 1665 cm⁻¹ and 836 cm⁻¹ were attributed to the C=C stretching and C=C bending of the isoprene unit, respectively.^[31]

Fig. 7Full-screen View

(Color online) FT-IR spectra of the B₄ C/NRL flexible films

3.4 Mechanical property of B₄ C/NRL films

B₄ C/NRL flexible films with 5 wt.%, 10 wt.%, 15 wt.%, 20 wt.%, 25 wt.%, 35 wt.%, 45 wt.%, and 55 wt.% B₄ C were vulcanized by boiled water at 100 °C for 40 min, 60 min, or 80 min,. The mechanical properties are illustrated in Fig. 8. The minimum tensile strength of all the samples was greater than 12 MPa (Fig. 8a), which is higher than that of the reported rubber for thermal neutron shielding.^[19-23] The tensile strength of the 10% B₄ C/NRL film was the highest under different vulcanizing temperatures. The tensile strength was the highest when films with different contents of B₄ C were vulcanized for 60 min. The highest tensile strength among all the samples was 29 MPa. When considering the resulting strength and energy consumption, curing for 60 min was considered the optimal timeframe. As the B₄ C concentration increased, the tensile strength of the B₄ C/NRL films initially increased and thereafter decreased. The tensile strengths of the 5 wt.%, 10 wt.%, and 15 wt.% B₄ C/NRL films were greater than that of the NRL film. The tensile strengths of the 20 wt.%, 25 wt.%, 35 wt.%, 45 wt.%, and 55 wt.% B₄ C/NRL films were less than that of the NRL film. No other fillers were added to NRL as reinforcing agents. The added B₄ C played the role of a reinforcing agent in the NRL. The addition of 10 wt.% B₄ C achieved the best reinforcing effect.

Fig. 8Full-screen View

(Color online) Tensile strength (a) and elongation at break (b) of the B₄ C/NRL flexible films

The elongation at break of the B₄ C/NRL films did not vary significantly in the range of 600%–810% (Fig. 8b), which is higher than that of the reported rubber for thermal neutron shielding,^[19-23] indicating that the B₄ C/NRL films are suitable for the high elasticity requirement in some cases, such as gloves.

Fig. 9 shows the hardness of the B₄ C/NRL films. The shore hardness of the B₄ C/NRL films increased with the increase in concentration of B₄ C; however, the B₄ C/NRL films were still flexible for practical applications in comparison with common latex gloves. The hardness is lower than that of the reported rubber for thermal neutron shielding^[24].

Fig. 9Full-screen View

(Color online) Shore hardness of the B₄ C/NRL flexible films

3.5 Thermal neutron shielding property of B₄ C/NRL films

The density of the B₄ C/NRL film was calculated using the following equation:

where m, v, ρ, and x represent the mass, volume, density, and mass fraction, respectively, and subscripts 1 and 2 represent B₄ C and NRL, respectively. Thus, m₁ /m₂ = x₁ /x₂ = ρ₁ v₁ /ρ₂ v₂, and x₁ +x₂ =1. The density of NRL measured using the Archimedes principle was 1 g·cm^-3. The density of the B₄ C/NRL flexible film is shown in Fig. 10. The experimental values are consistent with the calculated values.

Fig. 10Full-screen View

(Color online) Macroscopic cross-section of the B₄ C/NRL flexible films for attenuating thermal neutrons. Inset image: Density of the B₄ C/NRL flexible films.

Analogous with the attenuation for γ-rays, the mass attenuation coefficient of the B₄ C/NRL composite was calculated using the following equation: ^[7,32]

where σ is the line attenuation coefficient or macroscopic cross-section of the composite; σ_m is the mass attenuation coefficient of the B₄ C/NRL composite; the terms σ_m,1 and σ_m,2 represent the mass attenuation coefficients of B₄ C and NR, respectively. The mass attenuation coefficient of B₄ C for attenuating thermal neutrons can be obtained from Monte Carlo simulations or the literature database. The macroscopic cross-section of B₄ C for thermal neutrons (0.0253 eV) fitted by an experiment was 58.537 cm^-1.^[33] As NR and high-density polyethylene (HDPE) are both mainly composed of carbon and hydrogen, the macroscopic cross-section of NR for attenuating thermal neutrons (0.0253 eV) was approximately equal to that of HDPE, which was determined by an experiment as 0.77 cm^-1.^[34]

The macroscopic cross-section of the B₄ C/NRL composite was calculated using the following equation:

The macroscopic cross-section of the B₄ C/NRL flexible films for attenuating thermal neutrons is shown in Fig. 10. The experimental values of σ w $$ ere obtained from Eq. (3). The experimental values and the values calculated using Eq. (6) are consistent with each other. However, our results are different from those in the literature.^[35] This was mainly because our calculations of the macroscopic cross-section were based on experiments with thermal neutrons of energy 0.0253 eV. The most probable energy of thermal neutrons in the literature was larger than 0.0253 eV. ^[35]

After the macroscopic cross-section of the B₄ C/NRL flexible films was obtained, the attenuation efficiency of the B₄ C/NRL flexible films for thermal neutrons was calculated for different thicknesses and percentages of B₄ C using Eq. (3). Fig. 11 illustrates the attenuation efficiency of the B₄ C/NRL flexible films for attenuating thermal neutrons (0.0253 eV). This result is favorable to engineering design. If the attenuation efficiency is set, suitable thickness and B₄ C content can be designed to satisfy the attenuation efficiency. For a B₄ C/NRL flexible film, the B₄ C content can be calculated by measuring its density and thickness, which is easy to measure; subsequently, the efficiency for attenuating thermal neutrons (0.0253 eV) can be obtained.

Fig. 11Full-screen View

(Color online) Shielding efficiency of the B₄ C/NRL flexible films for attenuating thermal neutrons