Study on Concrete Activation Reduction in a PET Cyclotron Vault

Article information

J. Radiat. Prot. Res. 2020;45(3):130-141
Publication date (electronic) : 2020 September 30
doi : https://doi.org/10.14407/jrpr.2020.45.3.130
1Division of Advanced Nuclear Engineering, Pohang University of Science and Technology (POSTECH), Pohang, Korea
2Pohang Accelerator Laboratory, Pohang University of Science and Technology (POSTECH), Pohang, Korea
Corresponding author: Hee-Seock Lee, Department of Radiation Protection, Pohang Accelerator Laboratory, Pohang University of Science and Technology, 80 Jigok-ro 127beon-gil, Nam-gu, Pohang 37673, Korea, E-mail: lee@postech.ac.kr, https://orcid.org/0000-0002-7285-481X
Received 2020 June 07; Revised 2020 August 07; Accepted 2020 September 05.

Abstract

Background

Concrete activation in cyclotron vaults is a major concern associated with their decommissioning because a considerable amount of activated concrete is generated by second-ary neutrons during the operation of cyclotrons. Reducing the amount of activated concrete is important because of the high cost associated with radioactive waste management. This study aims to investigate the capability of the neutron absorbing materials to reduce concrete activa-tion.

Materials and Methods

The Particle and Heavy Ion Transport code System (PHITS) code was used to simulate a cyclotron target and room. The dimensions of the room were 457 cm (length), 470 cm (width), and 320 cm (height). Gd2O3, B4C, polyethylene (PE), and borated (5 wt% natB) PE with thicknesses of 5, 10, and 15 cm and their different combinations were selected as neutron absorbing materials. They were placed on the concrete walls to determine their effects on thermal neutrons. Thin B4C and Gd2O3 were placed between the concrete wall and additional PE shield separately to decrease the required thickness of the additional shield, and the thermal neutron flux at certain depths inside the concrete was calculated for each condition. Subsequently, the optimum combination was determined with respect to radioactive waste reduction, price, and availability, and the total reduced radioactive concrete waste was estimated.

Results and Discussion

In the specific conditions considered in this study, the front wall with respect to the proton beam contained radioactive waste with a depth of up to 64 cm without any additional shield. A single layer of additional shield was inefficient because a thick shield was required. Two-layer combinations comprising 0.1- or 0.4-cm-thick B4C or Gd2O3 behind 10 cm-thick PE were studied to verify whether the appropriate thickness of the additional shield could be maintained. The number of transmitted thermal neutrons reduced to 30% in case of 0.1 cm-thick Gd2O3+10 cm-thick PE or 0.1 cm-thick B4C+10 cm-thick PE. Thus, the thickness of the radioactive waste in the front wall was reduced from 64 to 48 cm.

Conclusion

Based on price and availability, the combination of the 10 cm-thick PE+0.1 cm-thick B4C was reasonable and could effectively reduce the number of thermal neutrons. The amount of radioactive concrete waste was reduced by factor of two when considering whole concrete walls of the PET cyclotron vault.

Introduction

Medical proton cyclotrons operating below 20 MeV are used to produce radioisotopes for medical diagnostic imaging. Typical short-lived radioisotopes used in positron emission tomography (PET) are 11C, 13N, 15O, and 18F among which 18F is in particular interest. More than 40 cyclotrons are installed in Korea including 20 non-self-shielded cyclotrons. These cyclotron facilities are expected to be dismantled before the designed lifespan due to upgrade, changes of purpose and place of use, etc. Activated materials inside the cyclotron vault and in concrete walls result in radiological hazards during the maintenance and decommissioning stage, respectively. Additionally, a huge amount of radioactive waste will be generated, after a long period of cyclotron operation, which is one of the main issues during the decommissioning stage because of the high cost of radioactive waste management. The primary radionuclides that cause the concrete activation are 152Eu (T1/2=13.52 y) and 60Co (T1/2=5.27 y). Although 151Eu and 59Co are trace element in the concrete constitutes, they have significant thermal neutron absorption cross sections (Table 1). After a long operation, the specific activities will exceed the order of clearance level [13]. The clearance level of the radionuclide in this work is based on the international standards [4]. A material is disregarded as the radioactive waste based on the following clearance criterion:

Major Radionuclides in Concrete and Corresponding Characteristics

(1) iCiCLi1

where Ci is the specific activity of the isotope i (Bq·g−1) and CLi is the clearance level for the isotope i (Bq·g−1).

The goal of this work is to reduce the total volume of radioactive concrete waste by a factor of two by considering the capability, availability, and the cost of the shielding materials. To achieve this goal, the thermal neutrons, which are responsible for the induced radioactivity in the concrete, were reduced by using the absorbing materials and the reduced amount of activated concrete was estimated. Activation reduction methods are important to decrease the activation level of the concrete in the depth for non-self-shielded cyclotrons that are already installed and are in operation.

Several researchers have proposed different ways to reduce the amount of radioactive waste such as using low activation concrete [5, 6], or reusing the low activated materials [7]. One effective method is to use neutron absorbing materials with high neutron capture cross sections. In addition to the conventional shielding materials as polyethylene (PE) and borated polyethylene (BPE), other novel materials such as boron carbide, and gadolinium are gaining interest [1, 8]. In this work, the effects of conventional and novel materials are considered and discussed. These materials can be used around the target or on the walls [9] in order to absorb thermal neutrons and consequently to reduce the activated concrete. A non-self-shielded PETtrace cyclotron (GE Healthcare, Chicago, IL, USA) has been selected to investigate the neutron reduction method by Monte Carlo code, Particle and Heavy Ion Transport code System (PHITS) version 3.02 [10]. The PETtrace is one of the popular medical cyclotrons which are used to generate 18F by impinging 16.5-MeV protons on a water target with enriched 18O via 18O(p,n)18F reaction. In our previous work, the neutron production yields from the H218O target have been investigated [11]. The secondary neutrons that are produced along with 18F activate the surrounding materials and the bulk concrete, especially, in case of non-self-shielded PET cyclotrons [2, 1214].

Materials and Methods

In order to consider the activity reduction in the concrete, a geometry of an H218O target and a cyclotron vault was simulated in PHITS-3.02 code [10] as shown in Fig. 1. In the PHITS code, by default, proton-induced reactions at energies up to 3 GeV are simulated by the intra-nuclear cascade of Liège model version 4.6 (INCL-4.6) [15]. The INCL-4.6 model is followed by the generalized evaporation model (GEM) [16]. In this simulation, JENDL-4.0 [17] cross sections library was applied for neutron interactions. INCL-4.6 was used for proton-induced reactions on 18O.

Fig. 1

PETtrace room geometry simulated using the PHITS code (A) top view, (B) vertical view, (C) 16.5-MeV proton beam striking the H218O target. Neutron spectra were scored inside the room and in the concrete at positions 1, 2, 3, and 4, respectively. The size of the room is 457 cm in length, 470 cm in width and 320 cm in height. PHITS, Particle and Heavy Ion Transport code System.

The dimension of the cyclotron vault was 457 cm in length, 470 cm in width and 320 cm in height. The neutron spectrum in the forward direction was scored before reaching the concrete (position 1 in Fig. 1). Position 1 was selected in this study as it is in the forward direction with respect to the proton beam and the maximum neutron flux is expected. The spherical detector radius is 10 cm and distance to the concrete wall is 40 cm. Thermal neutrons in the concrete at different depths were scored as well (positions 2, 3, and 4 in Fig. 1). Other tallies were also used to score the neutrons at different directions as shown in Fig. 1. However, the results from positions 1, 2, 3, and 4 are shown in this work. In this simulation, the concrete wall was divided into layers with the dimensions of 30×30 cm2 and the thickness of 3 cm. Neutrons were scored at each layer using the [T-Track] tally in the PHITS code. The scored neutron fluence was folded with the cross sections taken from JENDL-4.0 to consider the impact of the thermal, epithermal and fast neutrons on the radionuclides production. The thickness of the H218O target was 1.04 cm with a diameter of 1.4 cm. The target is bombarded with 16.5-MeV protons. The radioactivity is due to trace element in the concrete such as Eu and Co. The magnitude of activation will highly depend on the concrete composition. In this calculations, concrete compositions are taken from PAL-XFEL-2017 (Pohang Accelerator Laboratory X-ray Free Electron Laser) [18] with the density of 2.337 g·cm−3.

The operation scheme was selected based on the survey that was conducted about the cyclotrons in Korea [19] so that the maximum operation frequency and daily operation was selected to be 5 days per week and 4 hours a day. Additionally, it is expected that many of the cyclotrons would stop the operation before 10 years because of upgrade, changes of purpose, and place of use [20]. Therefore, the operation scheme in this work is quite conservative.

Boron, gadolinium, and hydrogen have high neutron absorption cross sections (Fig. 2) so that they can reduce thermal neutrons significantly and in case of hydrogen, it is relatively an effective neutron moderator. The whole walls of the cyclotron vault were then covered by neutron absorbing materials so as the effects of each material could be investigated separately. In the calculations, Gd2O3 (7.407 g·cm−3), B4C (2.52 g·cm−3), BPE (1.01 g·cm−3), and PE (0.94 g·cm−3), with thicknesses of 5, 10, and 15 cm were considered for each material. In addition, a combination of PE+Gd2O3, and PE+B4C were studied as well. Optimum material was determined by considering the thermal neutron absorption capability, the material cost and availability. Finally, the concrete thickness to be the radioactive waste were determined.

Fig. 2

Neutron absorption cross sections for natGd, natB, and natH taken from JENDL-4.0 [17]. The vertical dashed line indicates the thermal neutron energy of 0.4 eV.

Results and Discussion

1. Thermal Neutron Depth Profile

The neutron spectrum in the forward direction was scored before reaching the concrete (position 1 in Fig. 1) and is shown in Fig. 3. The spectrum contains 20% thermal neutrons and 80% epithermal and fast neutrons that can be down-scattered to thermal neutrons. The neutron spectra without any absorbing material are shown in Fig. 4A at several concrete depths. It is seen that fast neutrons are higher at the very beginning of the concrete depth and decrease as the concrete thickness increases. On the other hand, the number of thermal neutrons increases in the higher depths of concrete. Suppressing these thermal neutrons is extremely important to reduce the radioactivity in the concrete. The impact of the epithermal and fast neutrons on the 152Eu and 60Co production were also considered. Even though considering the wall position where those flux are maximum, the contribution of the epithermal and fast neutrons to the production of 152Eu was less than 10%, while they contribute to the 60Co production by 15% at the concrete surface and in the deeper positions it was less than 10%. Therefore, the thermal neutrons were the major concern in the 152Eu and 60Co production.

Fig. 3

Neutron spectrum in the forward direction respect to the proton beam indicated as position 1 in Fig. 1 at 40 cm before the concrete. The spectrum contains 20% thermal neutrons and 80% epithermal and fast neutrons. The vertical dashed line indicates the thermal neutron energy of 0.4 eV.

Fig. 4

(A) Neutron fluence at different concrete depths without neutron absorbing materials, (B) integrated neutron fluence (En<0.4 eV) as a function of concrete thickness in the front wall indicated as position 2 in Fig. 1. The proton beam current is 60 μA. Horizontal and vertical dashed lines correspond to the thermal neutron flux of 105 n·cm−2·s−1 and 64 cm concrete to be radioactive waste, respectively.

Fig. 4B shows the integrated thermal neutron fluence for energies less than 0.4 eV as a function of concrete depth, and illustrates that thermal neutron fluence reaches a maximum around 10 cm. The reason of this increase is that neutrons coming into the concrete interact with the hydrogens in a concrete and are thermalized, and then reach the maximum around the depth of 10 cm. Thermalized neutrons are captured by the elements of concrete and the number of thermal neutrons decreases after 10 cm depth. The level of concrete activity mainly depends on the amount of Eu and Co, and the thermal neutron flux. We measured the average values of Eu and Co as 1 and 15 ppm, respectively [18]. By considering the operation scheme of 4 hours per day, 5 days per week over 520 weeks (10 years), and the thermal neutron flux of 105 n·cm−2·s−1, the specific activities of 152Eu and 60Co were matched the clearance level. In other words, where the thermal neutron flux exceeded 105 n·cm−2·s−1, the concrete was considered as the radioactive waste in this work. The horizontal and vertical lines in Fig. 4B correspond to the thermal neutron flux of 105 n·cm−2·s−1 and the concrete depth which it is radioactive waste, respectively. It is seen that around 64 cm of the concrete becomes radioactive waste in the front wall.

In the simulations, the walls were covered by the absorbing materials with different thicknesses. Table 2 shows the thickness of each material and the corresponding transmitted thermal neutrons ratios obtained as Φwith shieldno shield at depth of 64 cm in the concrete wall. The thermal neutron fluence and the Φwith shieldno shield versus the depth in concrete for single materials are shown in Fig. 5 and for different combinations are shown in Fig. 6.

Neutron Absorbing Materials and Thicknesses Used in the Calculations and Corresponding Transmitted Thermal Neutrons at Depth of 64 cm in Concrete

Fig. 5

Thermal neutron fluences for different neutron absorbing materials and corresponding ratios, (Φwith shieldno shield) as function of concrete thickness in the front wall indicated as position 2 in Fig. 1. (A, B) for Gd2O3, (C, D) for B4C, (E, F) for PE, and (G, H) for BPE. Horizontal dashed lines correspond to thermal neutron flux of 105 n·cm−2·s−1. PE, polyethylene; BPE, borated polyethylene.

Fig. 6

Thermal neutron fluences for different neutron absorbing materials and corresponding ratios, (Φwith shieldno shield) as function of concrete thickness in the front wall indicated as position 2 in Fig. 1. (A, B) for PE+B4C and (C, D) for PE+Gd2O3. The layers of B4C and Gd2O3 are placed between the wall and PE. Horizontal dashed lines correspond to thermal neutron flux of 105 n·cm−2·s−1. PE, polyethylene.

Fig. 5A and 5B indicate the thermal neutron fluence against depth in concrete and corresponding ratios to case with no shield, respectively. The radioactive concrete waste in the front wall decreases to 59, 58, and 49 cm in depth by using 5, 10, and 15 cm-thick Gd2O3 shield and transmitted thermal neutrons are 65%, 40%, and 25%, respectively. It is seen from Fig. 5C that B4C can reduce the concrete thickness as the radioactive waste to 55, 43, and 30 cm. According to Fig. 5D, the thermal neutrons decrease dramatically so that for 5, 10, and 15 cm-thick B4C, the transmitted thermal neutrons are 45%, 20%, and 10%, respectively. Although boron carbide is quite effective in thermal neutron abortion, it is one of the costly materials, in particular, for covering whole wall of cyclotron walls with thick layers.

In case of PE, thermal neutrons increase at the surface of the concrete due to thermalization process in PE. At inner positions, however, thermal neutrons are attenuated by concrete (Fig. 5E, 5F). Placing 5, 10, and 15 cm of PE shield, the concrete thickness as the waste is 56, 48, and 43 cm, and the transmitted thermal neutrons are 58%, 31%, and 19%, respectively. On the other hand, for BPE, neutrons are elastically scattered by the hydrogen atoms in PE and lose energy and are absorbed by boron. Thus, the thermal neutron fluence is lower than that of only PE at shallow positions in the concrete. The activated thickness of the concrete after using 5, 10, and 15 cm-thick BPE shield is 56, 48, and 43 cm and the transmitted neutrons are 59%, 35%, and 20%, respectively (Fig. 5G, 5H).

10 cm-thick PE+10 cm-thick B4C can reduce dramatically the radioactive concrete waste to 20 cm in depth and the transmitted thermal neutrons to 7% (Fig. 6A, 6B). As indicated in Fig. 6C and 6D, the activated concrete thickness is reduced to 40, 45, and 38 cm and the transmitted thermal neutrons reach 20% by using 10 cm-thick PE+10 cm-thick Gd2O3, and 23% for 10 cm-thick PE+5 cm-thick Gd2O3, and 5 cm-thick PE+10 cm-thick Gd2O3.

It should be mentioned that in economical view, using a large amount of these absorbing materials are not cost effective. Therefore, thin layer shields of B4C and Gd2O3 were placed between 10 cm-thick PE and the concrete wall one by one and the results were compared together. As shown in Fig. 6A and 6B, using 0.1 or 0.4 cm-thick B4C layer, the thermal neutron fluence was reduced absolutely in the concrete depth up to 24 cm. Because the fast neutrons are thermalized in the PE shield and then are absorbed by boron before the concrete wall. On the other hand, the results for 0.1 cm and 0.4 cm-thick B4C with 10 cm-thick PE are almost similar so as the transmitted thermal neutrons was reduced to 30%. The thickness of concrete wall to be considered radioactive was also reduced to 48 cm in the front wall. According to Fig. 6C and 6D, the layered shields of 10 cm-thick PE+0.4 cm-thick Gd2O3 and 10 cm-thick PE+0.1 cm-thick Gd2O3 reduce the radioactive concrete thickness to 48 cm and the transmitted thermal neutrons were 30%.

Fig. 6 illustrates the results for the front wall where the effect of 10 cm-thick PE and 10 cm-thick PE+0.1 cm-thick B4C on the waste reduction seems similar. However, it is obvious in Fig. 7 to figure out the effect of B4C on the radioactive waste reduction compared with only PE layer. The depth profile of thermal neutrons at backward and left side walls, which are indicated as position 3 and position 4 in Fig. 1, are also shown in Fig. 7. It can be seen that in backward (Fig. 7A) and left side (Fig. 7B) walls, no radioactive waste is generated by using 10 cm-thick PE+0.1 cm-thick B4C, while the concrete waste thickness is around 20 cm for 10 cm-thick PE in both positions.

Fig. 7

Thermal neutron fluences for PE and PE+B4C combinations as function of concrete thickness. (A) In the backward wall indicated as position 3 in Fig. 1 and (B) in the left side wall indicated as position 4 in Fig. 1. Horizontal dashed lines correspond to thermal neutron flux of 105 n·cm−2·s−1. PE, polyethylene.

By placing 0.1 cm-thick B4C+10 cm-thick PE or 0.1 cm-thick Gd2O3+10 cm-thick PE on the concrete wall, the depth of concrete wall treated as the radioactive waste was reduced similarly, from 64 cm of no shield case to around 48 cm. Although the price of 10 cm-thick PE+0.1 cm-thick Gd2O3 is around 20% lower than that of 10 cm-thick PE+0.1 cm-thick B4C. The Gd2O3 is in the form of powder and needs to be fabricated in a correct way to be used after the PE layer. On the other hand, B4C is commercially available in different forms and it is easy to be shaped and used after the PE layer. Therefore, 10 cm-thick PE+0.1 cm-thick B4C is considered to be the best combination as additional shielding materials.

2. Estimation of Radioactive Concrete Reduction

The thermal neutron versus concrete thickness shown in Fig. 6 are only for the forward direction with respect to the proton beam (front wall). In order to estimate the total reduction of radioactive concrete waste, the thermal neutron distribution inside the cyclotron room as well as all concrete walls are shown in Fig. 8 and discussed. The figures were obtained for the cases without any additional shielding material (Fig. 8A, 8B), with 10 cm-thick PE (Fig. 8C, 8D), and with 10 cm-thick PE+0.1 cm-thick B4C shield (Fig. 8E, 8F). Fig. 8 illustrates the thermal neutron distributions (En<0.4 eV) calculated by the PHITS code in the vertical (left column) and top views (right column). The concrete depths at which the thermal neutron exceeds 105 n·cm−2·s−1 are indicated for each case. Without neutron absorbing materials, more than 60 cm of the concrete was considered to be radioactive waste in the front wall regarding the proton beam as well as in the floor and more than 40 cm in all other walls. By using 10 cm-thick PE, the radioactive waste thickness in the front wall and floor is 48 cm and that of ceiling is 30 cm (Fig. 8C). The radioactive waste thickness is 20 cm for the left side and backward walls and 32 cm for the right side wall respecting to the proton beam (Fig. 8D). On the other hand, 10 cm-thick PE+0.1 cm-thick B4C can reduce the concrete waste thickness to 48 cm in the front wall as well as in the floor and 24 cm in the ceiling (Fig. 8E). Thickness of 28 cm of the concrete is radioactive waste in the right side wall regarding to the proton beam. No activated concrete was generated in the backward and left side walls (Fig. 8F).

Fig. 8

Distributions of thermal neutron flux (En<0.4 eV) inside the cyclotron room and the concrete walls in vertical (left column) and top views (right column) (A, B) without neutron absorbing materials, (C, D) with 10 cm-thick PE, and (E, F) with 10 cm-thick PE+0.1 cm-thick B4C. The thin layer of B4C is placed between the wall and PE. The numbers indicate the depths at which thermal neutron flux exceeds 105 n·cm−2·s−1 and concrete is radioactive waste. PE, polyethylene.

At forward direction, the effect of 10 cm-thick PE and 10 cm-thick PE+0.1 cm-thick B4C seems to be similar. For the side walls, ceiling, floor, and back wall, however, 0.1 cm-thick B4C can absorb the scattered neutrons dramatically as can be seen from Figs. 7 and 8. The reason is that the neutrons are scattered and become thermal and can be captured by B4C. As a result, the total radioactive waste volume decreases. In the forward direction, most neutrons are epithermal and fast neutrons that cannot be captured by B4C so that the difference between the 10 cm-thick PE and 10 cm-thick PE+0.1 cm-thick B4C is small. It is found that the amount of radioactive concrete waste can be reduced by a factor of 2 using the 10 cm-thick PE+0.1 cm-thick B4C shield. It can be seen from Fig. 8 that the front wall respecting to the proton beam and the floor are the hot spots where the amount of radioactive waste is quite high. Thus, 10 cm-thick PE+0.1 cm-thick B4C can be used for these areas, and for other side walls, thinner shielding materials is enough to reduce thermal neutron fluence inside the walls and results in lowering the price of the shielding materials. In a real cyclotron vault, however, neutrons can be scattered by the cyclotron body remarkably and reach to the thermal energy which can be captured by boron in the B4C. Therefore, 10 cm-thick PE+0.1 cm-thick B4C is expected to be more effective to reduce the radioactive waste in concrete than the simple geometry considered in this work. It is emphasized that the operation scheme in this work was conservative so that in the case of the operations corresponding to around 20% of our operation scheme and using 10 cm-thick PE+0.1 cm-thick B4C, one should not consider the concrete as the radioactive waste. By considering the installation cost of the shielding material, it seems to be very beneficial to reduce the radioactive waste storage expenses.

It should be pointed out that the B4C is available in rubber form that can be shaped easily and can be attached to the walls. The amount of B4C in the rubber material is 80% and 20% is the glue to form the rubber. Therefore, the required thickness of the B4C rubber is 0.12 cm to make equivalent to 0.1 cm-thick B4C.

3. Production of Other Radionuclides

Together with the 152Eu and 60Co radionuclides, the 54Mn, 22Na, and 3H radionuclides can be produced through the 54Fe(n,p)54Mn, 23Na(n,2n)22Na, 2H(n,γ)3H, and 6Li(n,α)3H reactions. 54Mn and 22Na are generated by fast neutrons so as their production were approximately in the order of 152Eu and 60Co at the concrete surface. The 54Mn and 22Na production decreased exponentially with the concrete depth and became one order lower than 152Eu and 60Co at 30 cm. The production of 3H in the concrete was about one order higher than other radionuclides due to the high reaction cross section of the 6Li(n,α)3H reaction. These results are in agreement with the results reported in the literature [2, 13]. However, the clearance level of 3H is 100 Bq·g−1 which is three orders higher than that of 152Eu and 60Co. It was also confirmed that 3H is produced mainly by the thermal neutrons and the transmitted thermal neutrons decrease to 30% by using the 10 cm-thick PE+0.1 cm-thick B4C shielding material and the production of the 3H could reduce correspondingly. The number of fast neutrons at different concrete depths without and with the 10 cm-thick PE+0.1 cm-thick B4C shielding are shown in Fig. 9A and corresponding ratio is indicated in Fig. 9B. It is seen that the transmitted fast neutrons also decrease to 40% by using the shielding material. It means that the 3H, 54Mn, and 22Na production could also decrease considerably and they could not be a major issue for the radioactive waste management of concrete.

Fig. 9

(A) Fast neutron fluence (En>1 MeV) against the concrete depth without and with the 10 cm-thick PE+0.1 cm-thick B4C shielding, (B) transmitted fast neutrons, (Φwith shieldno shield) as a function of concrete thickness in the front wall indicated as position 2 in Fig. 1. PE, polyethylene.

Structural concrete is often reinforced with steel bars that become activated via 54Fe(n,γ)55Fe, 54Fe(n,p)54Mn, and 59Co(n,γ)60Co reactions. The reinforcing bars contribution to the activation can be significant [21]. However, the number of fast and thermal neutrons significantly decrease by using 10 cm-thick PE+0.1 cm-thick B4C as the shielding material and the activation of the reinforcing bars are also decreased proportionally. It is noticed that the activated bars are localized and could be separated from the concrete during the dismantling. They are not as voluminous as the concrete to become a major concern for the radioactive waste storage. For the case of 54Mn, its activation level remains constant after several years of operation [21, 22]. By considering the typical cooling time before the decommissioning the activity level of 54Mn could reduce as well.

Conclusion

The use of non-self-shielded cyclotrons for the PET isotopes production results in serious concrete activation after a long operation. In this study, intensive Monte Carlo calculations were performed to mitigate the concrete activation by using additional neutron absorbing materials. A non-self-shielded PETtrace cyclotron vault and an H218O target was simulated by the PHITS code to study the effects of neutron absorbers on thermal neutrons reduction. The conservative cyclotron operation scheme of 4 hours per day, 5 days per week over 520 weeks (10 years) and 60 μA proton beams was selected in this work. The amounts of Eu and Co were 1 and 15 ppm, respectively. The thermal neutron flux of 105 n·cm−2·s−1 was obtained at above specific condition, which the concrete was considered as the radioactive waste. Results showed that Gd2O3, B4C, PE, and BPE would be effective to decrease the activation level of concrete. However, a combination of these materials is more effective than using single shield material to decrease the price of the shielding. It was concluded that placing 0.1 cm-thick B4C between 10 cm-thick PE and the concrete wall could reduce the total radioactive concrete waste by a factor of 2 in the whole cyclotron room. However, this combination is expected to reduce more thermal neutrons in a real cyclotron vault. Additionally, the amount of radioactive waste decreases dramatically at cyclotron with lower operating parameters. By taking the price of 0.1 cm-thick B4C and 10 cm-thick PE into account, it is still cost-effective to use it as a shielding material to achieve a significant reduction of radioactive concrete waste volumes.

Acknowledgements

This work was supported by the Nuclear Safety Research Program through the Korea Foundation of Nuclear Safety (KOFONS), granted financial resource from the Nuclear Safety and Security Commission (NSSC) of Republic of Korea (No. 1303026 and 1603005).

Notes

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Author Contribution

Conceptualization: Bakhtiari M. Data curation: Bakhtiari M. Funding acquisition: Lee HS. Methodology: Bakhtiari M, Mokhtari Oranj L, Jung NS, Lee A. Project administration: Lee HS. Writing - original draft: Bakhtiari M, Jung NS. Writing - review & editing: Bakhtiari M, Mokhtari Oranj L, Lee HS. Software: Bakhtiari M. Supervision: Lee HS.

References

1. Kumagai M, Sodeyama K, Sakamoto Y, Toyoda A, Matsumura H, Ebara T, et al. Activation reduction method for a concrete wall in a cyclotron vault. J Radiat Prot Res 2017;42:141–145.
2. Wang Q, Masumoto K, Bessho K, Matsumura H, Miura T, Shibata T. Evaluation of the radioactivity in concrete from accelerator facilities. J Radioanal Nucl Chem 2007;273:55–58.
3. Dodd AC, Shackelton RJ, Carr DA, Ismail A. Activation of air and concrete in medical isotope production facilities. AIP Conf Proc 2017;1845:020006.
4. International Atomic Energy Agency. Application of the concepts of exclusion, exemption and clearance: safety guide (RS-G-1.7) Vienna, Austria: International Atomic Energy Agency; 2004.
5. Sato S, Morioka A, Kinno M, Ochiai K, Hori J, Nishitani T. Experimental study on induced radioactivity in boron-doped low activation concrete for DT fusion reactors. J Nucl Sci Technol 2004;41(sup4):66–69.
6. Sato S, Maegawa T, Yoshimatsu K, Sato K, Nonaka A, Takakura K, et al. Development of a low activation concrete shielding wall by multi-layered structure for a fusion reactor. J Nucl Mater 2011;417:1131–1134.
7. Ehrlicher U, Pauli H. Waste reduction by re-use of low activated material. In : Proceedings of the International Conference on Radioactive Waste Management and Environmental Remediation; 2009 Oct 11–15; Liverpool, UK. p. 423–430.
8. Wang P, Tang X, Chen F, Chen D. The design, fabrication and safety evaluation of a novel spent fuel storage basket material. Nucl Eng Des 2015;284:91–96.
9. Major B. PET cyclotron design for decommissioning and waste inventory reduction. In : Proceedings of the International Conference on Radioactive Waste Management and Environmental Remediation; 2009 Oct 11–15; Liverpool, UK. p. 19–24.
10. Sato T, Iwamoto Y, Hashimoto S, Ogawa T, Furuta T, Abe SI, et al. Features of particle and heavy ion transport code system (PHITS) version 3.02. J Nucl Sci Technol 2018;55:684–690.
11. Bakhtiari M, Oranj LM, Jung NS, Lee A, Lee HS. Estimation of neutron production yields from H218O as the 18F-production target bombarded by 18-MeV protons. Radiat Phys Chem 2020;177:109120.
12. Fujibuchi T, Horitsugi G, Yamaguchi I, Eto A, Iwamoto Y, Obara S, et al. Comparison of neutron fluxes in an 18-MeV unshielded cyclotron room and a 16.5-MeV self-shielded cyclotron room. Radiol Phys Technol 2012;5:156–165.
13. Masumoto K, Miura T, Bessho K, Matsumura H, Toyoda A, Wang Q, et al. Evaluation of radioactivity in concrete samples obtained from various accelerator facilities. In : Proceedings of the 2nd Asian and Oceanic Congress for Radiation Protection (AOCRP); 2006 Oct 9–13; Beijing, China.
14. Masumoto K, Toyoda A, Eda K, Izumi Y, Shibata T. Evaluation of radioactivity induced in the accelerator building and its application to decontamination work. J Radioanal Nucl Chem 2003;255:465–469.
15. Boudard A, Cugnon J, David JC, Leray S, Mancusi D. New potentialities of the Liège intranuclear cascade model for reactions induced by nucleons and light charged particles. Phys Rev C 2013;87:014606.
16. Furihata S. Statistical analysis of light fragment production from medium energy proton-induced reactions. Nucl Instrum Methods Phys Res B 2000;171:251–258.
17. Shibata K, Iwamoto O, Nakagawa T, Iwamoto N, Ichihara A, Kunieda S, et al. JENDL-4.0: a new library for nuclear science and engineering. J Nucl Sci Technol 2011;48:1–30.
18. Lee A, Jung NS, Lee HS. Composition analysis of ordinary concrete to estimate residual isotopes in the decommissioning of particle accelerator. In : Proceedings of the 14th Workshop on Shielding Aspects of Accelerators, Targets and Irradiation Facilities (SATIF-14); 2018 Oct 30–Nov 2; Gyeongju Korea.
19. Gwon DY, Kim Y, Jeong KY, Jung NS, Lee HS. Regulation improvement of cyclotrons in the Republic of Korea through analysis of radioactivation and questionnaire on operation status. Indian J Public Health Res Dev 2018;9:675–680.
20. Jung NS, Lee A, Oranj LM, Lee HS, Gwon DY, Kim Y, et al. Activation assessment of self-shielded and non-self-shielded PET cyclotrons. In : Proceedings of the 14th Workshop on Shielding Aspects of Accelerators, Targets and Irradiation Facilities (SATIF-14); 2018 Oct 30–Nov 2; Gyeongju, Korea.
21. Vichi S, Infantino A, Zagni F, Cicoria G, Braccini S, Mostacci D, et al. Activation studies for the decommissioning of PET cyclotron bunkers by means of Monte Carlo simulations. Radiat Phys Chem 2020;174:108966.
22. Martinez-Serrano JJ, Diez de los Rios A. Prediction of neutron induced radioactivity in the concrete walls of a PET cyclotron vault room with MCNPX. Med Phys 2010;37:6015–6021.
23. Mughabghab SF. Neutron cross sections: neutron resonance parameters and thermal cross sections (Part B, Z=61–100) Orlando, FL: Academic Press; 1984.

Article information Continued

Fig. 1

PETtrace room geometry simulated using the PHITS code (A) top view, (B) vertical view, (C) 16.5-MeV proton beam striking the H218O target. Neutron spectra were scored inside the room and in the concrete at positions 1, 2, 3, and 4, respectively. The size of the room is 457 cm in length, 470 cm in width and 320 cm in height. PHITS, Particle and Heavy Ion Transport code System.

Fig. 2

Neutron absorption cross sections for natGd, natB, and natH taken from JENDL-4.0 [17]. The vertical dashed line indicates the thermal neutron energy of 0.4 eV.

Fig. 3

Neutron spectrum in the forward direction respect to the proton beam indicated as position 1 in Fig. 1 at 40 cm before the concrete. The spectrum contains 20% thermal neutrons and 80% epithermal and fast neutrons. The vertical dashed line indicates the thermal neutron energy of 0.4 eV.

Fig. 4

(A) Neutron fluence at different concrete depths without neutron absorbing materials, (B) integrated neutron fluence (En<0.4 eV) as a function of concrete thickness in the front wall indicated as position 2 in Fig. 1. The proton beam current is 60 μA. Horizontal and vertical dashed lines correspond to the thermal neutron flux of 105 n·cm−2·s−1 and 64 cm concrete to be radioactive waste, respectively.

Fig. 5

Thermal neutron fluences for different neutron absorbing materials and corresponding ratios, (Φwith shieldno shield) as function of concrete thickness in the front wall indicated as position 2 in Fig. 1. (A, B) for Gd2O3, (C, D) for B4C, (E, F) for PE, and (G, H) for BPE. Horizontal dashed lines correspond to thermal neutron flux of 105 n·cm−2·s−1. PE, polyethylene; BPE, borated polyethylene.

Fig. 6

Thermal neutron fluences for different neutron absorbing materials and corresponding ratios, (Φwith shieldno shield) as function of concrete thickness in the front wall indicated as position 2 in Fig. 1. (A, B) for PE+B4C and (C, D) for PE+Gd2O3. The layers of B4C and Gd2O3 are placed between the wall and PE. Horizontal dashed lines correspond to thermal neutron flux of 105 n·cm−2·s−1. PE, polyethylene.

Fig. 7

Thermal neutron fluences for PE and PE+B4C combinations as function of concrete thickness. (A) In the backward wall indicated as position 3 in Fig. 1 and (B) in the left side wall indicated as position 4 in Fig. 1. Horizontal dashed lines correspond to thermal neutron flux of 105 n·cm−2·s−1. PE, polyethylene.

Fig. 8

Distributions of thermal neutron flux (En<0.4 eV) inside the cyclotron room and the concrete walls in vertical (left column) and top views (right column) (A, B) without neutron absorbing materials, (C, D) with 10 cm-thick PE, and (E, F) with 10 cm-thick PE+0.1 cm-thick B4C. The thin layer of B4C is placed between the wall and PE. The numbers indicate the depths at which thermal neutron flux exceeds 105 n·cm−2·s−1 and concrete is radioactive waste. PE, polyethylene.

Fig. 9

(A) Fast neutron fluence (En>1 MeV) against the concrete depth without and with the 10 cm-thick PE+0.1 cm-thick B4C shielding, (B) transmitted fast neutrons, (Φwith shieldno shield) as a function of concrete thickness in the front wall indicated as position 2 in Fig. 1. PE, polyethylene.

Table 1

Major Radionuclides in Concrete and Corresponding Characteristics

Radionuclide Reaction Half-life Thermal neutron absorption cross section (barn) [17, 23] Clearance level (Bq·g−1)
152Eu 151Eu(n,γ)152Eu 13.52 yr 5,900 0.1
134Cs 133Cs(n,γ)134Cs 2.06 yr 29 0.1
60Co 59Co(n,γ)60Co 5.27 yr 37 0.1
46Sc 45Sc(n,γ)46Sc 83.79 day 26.5 0.1

Table 2

Neutron Absorbing Materials and Thicknesses Used in the Calculations and Corresponding Transmitted Thermal Neutrons at Depth of 64 cm in Concrete

Material Thickness (cm) Transmitted thermal neutrons (%)
Gd2O3 5 65
10 40
15 25

B4C 5 45
10 20
15 10

PE 5 58
10 31
15 19

BPE 5 59
10 35
15 20

PE+Gd2O3 10+10 20
10+5 23
10+0.4 30
10+0.1 30
5+10 23

PE+B4C 10+10 7
10+0.4 30
10+0.1 30