Production of actinium-225 from a (n,p) reaction: Feasibility and pre-design studies
, , , , und | 08. Juni 2021
Article Category: Original Paper
Online veröffentlicht: 08. Juni 2021
Seitenbereich: 61 - 67
Eingereicht: 18. Aug. 2020
Akzeptiert: 06. Apr. 2021
DOI: https://doi.org/10.2478/nuka-2021-0008
Schlüsselwörter
© 2021 Fouad A. Abolaban et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Recent studies have shown that mortality among people due to prostate cancer is above 358 000 worldwide, and is ranked as the second most occurrence of cancer after lung cancer [1, 2, 3]. Metastatic castration-resistant prostate cancer (mCRPC) is an advanced prostate cancer [4, 5]. The prostate-specific membrane antigen (PSMA) is widely known as a glycoprotein cell surface that can be used as a metastatic prostate cancer indicator. Besides, the PSMA can be also used as a malignant neoplasms indicator of neovasculature [6]. This advanced stage of prostate cancer can be considered a non-curable disease depending on the PSMA spreading stage. Recent studies and clinical reports show that molecular radiotherapy using radionuclides such as Lu-177, Bi-213, and Ac-225 [7, 8] can increase the survival rate of patients [7, 9, 10, 11, 12]. This involves labelling the PSMA with alpha or beta emitter sources where the alpha has a high linear energy transfer (LET) and relative biological effect (RBE) compare to beta. Actinium-225 is the preferred radionuclide as it is an alpha emitter and has both high LET and RBE almost 100 times more than beta emitters [6].
The generated amount of these radiopharmaceuticals is a major challenge as the demand is high while the supply is not sufficient [12]. Several studies were reported by Morgenstern
This paper aims to study the feasibility by first generating p in a thin layer of one material from an (n,p) reaction, as shown in Fig. 1, and then Ac-225 atoms in a thin layer of a second material from a (p,2n) reaction. This study will evaluate the activation yield by utilizing different materials and the amount of Ac-225 gained from each material.
Fig. 1
Ac-225 generation process.
Several research studies have been published explaining the different methods of generating Ac-225 [10, 14, 18]. The accelerator method that uses Ra-226 has been investigated theoretically and experimentally and shows promising results [12, 15, 19]. The target of RaCl2 can be irradiated by proton, neutrons or gamma rays [19]. In this work, the plan is to evaluate the generation of Ac-225 in a two-stage process using Ra-226 (p,2n) reaction: neutrons incident on material no. 1 shown previously in Fig. 1 to generate high energy protons, and the generated protons will then be incident on material no. 2, which is the RaCl2 in this study. The evaluation is carried out using the GATE Monte-Carlo simulation toolkit.
GATE incorporates the well-validated physics of GEANT4 with a more user-friendly simulation environment [20]. GATE has shown great potential in a wide range of medical applications, including radiotherapy [21, 22, 23], diagnostic radiology [24, 25], and dosimetry [26, 27, 28].
The simulation can be split into three stages. The first two stages focus on finding the optimum parameters for producing Ac-225 using a RaCl2 target (see Table 1), while the third employs these parameters into studying the feasibility and viability of producing the radionuclide using the said stages. The quark–gluon string model with a high neutron precision package QGSP_BIC_HP was used. This reference physics package provides a good description of secondary particles produced by protons and neutrons in addition to a very reliable transport modelling for neutrons below 20 MeV [29].
Stages to find the optimum parameters for Ac-225 generation
| Simulation | |||
|---|---|---|---|
| Stage 1: Bombarding RaCl2 target with high energy protons | |||
| Steps | 1 | 2 | 3 |
| Vary proton energy | Vary RaCl2 target thickness | Find optimum conditions to produce Ac-225 | |
| Stage 2: Bombarding several materials with neutrons to generate high energy protons found in the optimum condition for stage 1 | |||
| Steps | 1 | 2 | 3 |
| Vary neutron energy | Vary material type and thickness | Find optimum conditions to generate high energy protons | |
| Stage 3: Combining results from stages 1 & 2 to simulate two hollow cylinders: the inner cylinder is a proton-producing target, and the outer cylinder is RaCl2 to produce Ac-225 | |||
| Steps | 1 | 2 | |
| Proton production (n,p) | Ac-225 production (p,2n) | ||
The first simulation stage involves finding the proton energy threshold for producing Ac-225 by bombarding a RaCl2 target with high-energy protons. This is achieved by embedding a monoenergetic proton point source at the centre of a target composed of a radium chloride (RaCl2) cylinder. Both the proton energy and the target thicknesses are varied to find the optimum conditions for producing Ac-225. Proton energies considered in this set of experiments are 2, 5, 7.5, 10, 15, and 20 MeV. Thicknesses considered are 0.04, 0.06, 0.08, 0.1, 0.25, and 1 mm. The number of particles used in this stage is 108.
In the second simulation stage, the objective is to find the neutron energy and target material required to produce protons that meet the minimum requirements found in the previous step for producing Ac-225. Therefore, a neutron monoenergetic point source is placed at the centre of a cylindrical proton-producing target. The neutron energy, cylinder material and its thickness are varied. For each run, the energy spectrum of the produced protons is recorded. The feasibility of each neutron energy, material type and thickness depends on the quantity and energy of the produced protons. Several materials were investigated, including nickel (Ni), iron (Fe), potassium (K), stainless-steel (SS), magnesium (Mg) and manganese (Mn). The number of particles used in this stage was 108.
The outcomes of the first and second stages form the basis of the third stage. The setup of the third stage consists of two hollow co-axial cylinders. The inner cylinder acts as the proton-producing target. The proton-producing target composition and thickness are dependent on the outcomes of the second stage. The outer cylinder represents the target and is composed of RaCl2 with the optimum thickness found in the first stage. At the centre of the proton-producing target, a point neutron source is placed. All available neutron sources with energies meeting the criteria are used against this setup to find the probability of producing Ac-225 for each neutron source.
The number of neutrons used should be high enough to produce enough protons to produce Ac-225. The fraction of protons produced by neutrons and Ac-225 produced by protons is expected to be relatively low. Therefore, a very high number of neutrons is required to produce Ac-225. Using a very high number of particles would extend the simulation time significantly. Alternatively, to achieve low statistical uncertainty in calculating Ac-225 production probability while keeping the simulation time reasonable, the simulation is split into two parts.
The splitting is carried out by adding a very thin virtual layer (1 μm) between the two cylinders. The purpose of this layer is to record the properties of all protons produced in the proton-producing target and leaving its volume towards the target.
When the proton-producing target is bombarded with 108 neutrons, some protons will be produced. Several of these protons will leave the proton-producing target and transverse the thin layer where their momentum and spatial properties are stored in a phase space file. Subsequently, the simulation is relaunched using the same setup but with the neutron source being replaced by a proton source. This proton source emits from the phase-space layer with the source properties derived from the phase-space file. The number of protons emitted from the phase-space file is set to 5 × 106 and by that, the number of protons is amplified by a factor ranging approximately from 104 to 105.
The cross section for Ac-225 production from RaCl2 material depends mainly on the proton energy interacting with RaCl2. Figure 2 shows the dependence of producing Ac-225 on the proton energy. Even though six proton energies were simulated in the first stage, starting at 2.5 MeV, the figure only shows three curves because the probability of producing Ac-225 from protons with energies below or equal to 10 MeV is between 0.0006% and 0.0008%. The optimum RaCl2 thickness to produce Ac-225 for 15 MeV and 20 MeV protons was determined to be 5 mm. To calculate the activity of the generated Ac-225 using the curves from Fig. 2, a 1 Ci proton source incident on a 5 mm thick RaCl2 target will result in 32 μCi and 68 μCi for the 10 MeV and the 15 MeV proton energies, respectively.
Fig. 2
Probability of producing Ac-225 from bombarding RaCl2 with high energy protons inside the cylinder target as a function of proton energy. The target density is 4.9 g/cm3. An isotropic point proton source was placed inside a RaCl2 cylinder with a 3 cm height and varying diameter (thickness) in an air volume.
The generation of high-energy protons is crucial in producing actinium as shown in the previous figure. To produce high-energy protons inside the Ac-225 generator, a neutron beam with energies from 10 MeV up to 20 MeV was incident on nickel (Ni), iron (Fe), potassium (K), stainless-steel (SS), magnesium (Mg) and manganese (Mn). The following discussion is for nickel, manganese and iron as they produced the best results.
Figure 3 shows the proton average energy generated at different nickel thicknesses from a neutron beam with 10, 12.5, 15, 17.5, and 20 MeV. It should be noted that the maximum proton energy generated cannot exceed the maximum neutron energy from the (n,p) reaction. The figure demonstrates that the higher the neutron energy, the higher the average proton energy generated at all suggested nickel thickness. The maximum thickness in the
Fig. 3
Average proton energy generated from neutron beams interacting with nickel at several thicknesses.
Fig. 4
Average proton energy generated from neutron beams interacting with manganese at several thicknesses.
The average produced protons energy ranges from 0 to approximately the maximum incident neutron energy, i.e. a 10 MeV neutron can produce 10 MeV protons, because of the conservation of energy. The incident neutron energy and the average produced proton energy are directly proportional. The maximum average proton energy was around 5.6 MeV at 0.25 mm from the 20 MeV neutrons.
Figure 4 shows the average proton energy generated at different manganese thicknesses from a neutron beam of 10, 12.5, 15, 17.5, and 20 MeV. Figure 4 confirms what was demonstrated in the nickel case, where the higher the neutron energy, the higher the average proton energy generated is at all thicknesses. The maximum thickness was kept to 1 mm to reduce the probability of proton absorption in the manganese.
The maximum average proton energy generated by 20 MeV neutrons is approximately 7.7 MeV at ~0.05 mm, which is higher than the nickel case.
Figure 5 shows the average proton energy generated at different iron thicknesses from a neutron beam of 10, 12.5, 15, 17.5, and 20 MeV. Figure 5 follows the same trend observed in the nickel and manganese cases, where the higher the incident neutron energy is, the higher the average proton energy is generated at all thicknesses. The maximum average proton energy generated by 20 MeV neutrons is around 6 MeV at ~0.05 mm, which is slightly higher than the nickel case.
Fig. 5
Average proton energy generated from neutron beams interacting with iron at several thicknesses.
Figure 6 shows the probability of producing protons at the same nickel thicknesses used in Fig. 3 as a function of neutron energy. The probability is maximum at 0.1 mm thickness for the 17.5 MeV neutron energy and slowly decreases as the nickel thickness increases for all other neutron energies.
Fig. 6
Probability of proton production from neutrons incident on nickel at different thicknesses as a function of neutron energy. The target density is 8.9 g/cm3. An isotropic point neutron source was placed inside a RaCl2 cylinder with a 3 cm height and varying diameter (thickness) in an air volume.
Figure 7 shows the probability of producing protons at different manganese thicknesses as a function of neutron energy. The probability is maximum at 0.25 mm thickness for the 20 MeV neutron energy and decreases as the neutron energy decreases. There is a tradeoff between the maximum average proton energy that can be produced by nickel and manganese and the probability of producing the protons for the two metals.
Fig. 7
Probability of proton production from neutrons incident on manganese at different thicknesses as a function of neutron energy. The target density is 7.43 g/cm3. An isotropic point neutron source was placed inside a RaCl2 cylinder with a 3 cm height and varying diameter (thickness) in an air volume.
Figure 8 shows the probability of producing protons at different iron thicknesses as a function of neutron energy. The probability is maximum at 0.25 mm thickness for the 20 MeV neutron energy and decreases as the neutron energy decreases. The iron and manganese have similar proton production probabilities, both are equal but less than the nickel case.
Fig. 8
Probability of proton production from neutrons incident on iron at different thicknesses as a function of neutron energy. The target density is 7.87 g/cm3. An isotropic point neutron source was placed inside an RaCl2 cylinder with a 3 cm height and varying diameter (thickness) in an air volume.
The results for potassium, stainless-steel, and magnesium are not discussed here because either the probability of proton production is very low so that it is not feasible to bombard the RaCl2 with sufficient protons to generate Ac-225, or the generated proton energy is not enough to produce Ac-225. Combining the results for nickel, manganese and iron, it can be concluded that the neutron source must possess energy >10 MeV to produce Ac-225 in addition to a high probability of producing protons. Thus, nickel was chosen as the target for the neutron source inside the generator. The steps needed to generate Ac-225 are shown in Fig. 9. The first two steps of the process are made from three concentric cylinders. The innermost is a Cf-252 source, the middle cylinder is made from nickel and the outermost cylinder is made from RaCl2. A chemical separation stage is needed to be applied for each batch produced by the generator that contains Ac-225 to get rid of the unwanted fragments generated by the irradiation process in the third step before preparing a patient dose in the fourth and final step. Additional shielding should be placed around the generator to protect the users from neutrons and gamma rays.
Fig. 9
Proposed Ac-225 generator.
Since Cf-252 emits a spectrum of energies, additional simulations were run to find the optimum nickel thickness between 0.01 mm and 0.06 mm. The 0.01 mm thickness produced 3.3 MeV average proton energy and the 0.05 mm thickness produced 2.7 MeV average energy with about 90% more protons than the 0.01 mm thickness. Thus, the 0.05 mm thickness of nickel was chosen to be paired with Cf-252. On the other hand, all RaCl2 thicknesses produced a similar amount of Ac-225, and so a 5 mm thickness was used. Even though it was stated at the beginning of the results section that the proton energy needed to produce Ac-225 should be >10 MeV, this does not contradict the 2.7–3.3 MeV average proton energy generated from bombarding nickel with Cf-252 because the maximum generated proton energy from the simulation was between 10 MeV and 19 MeV for the 0.05 mm nickel thickness depending on the neutron energy.
The simulation results demonstrated that in order to produce a single Ac-225 atom: (1) a Cf-252 source with a minimum activity of 7.15E12 Bq should be used, (2) the neutrons from the Cf-252 should be incident on a nickel target with a thickness of 0.05 mm, (3) the nickel target will produce a 2.7 MeV average energy protons, (4) the protons will be incident on an RaCl2 5 mm thick target. A further set of simulations were carried out replacing the Cf-252 by a typical neutron generator with a 14 MeV energy. The geometry for the generator simulation differs from the Cf-252 one. The generator will be incident on a 5 mm height and 5 mm diameter cylinder needle with a 0.05 mm nickel thickness. The nickel needle material in the needle will produce high-energy protons that are incident on 5 mm RaCl2 volume inside the cylinder shape needle as this was experimentally tested [30]. The Ac-225 atoms are produced at the back of the RaCl2 volume. The minimum activity needed to produce a single Ac-225 atom/s with the neutron generator is 2.21E9 Bq.
The Cf-252 and the neutron generator produced isotopes/fragments in addition to the Ac-225. Some of the isotopes are radioactive and some are not, so the chemical separation stage is of great importance to produce a pure Ac-225 solution that does not contain other radioactive materials to ensure the dose delivered to the patient is optimized.
In terms of real-life applicability, the process of producing Ac-225 is feasible. Because of the high activity Cf-252 source, there will be a need to shield against the high energy neutrons and the high energy gamma emitted by the source. A two-metre diameter barrel shield made of high-density polyethylene (HDPE) with a small lead cylinder around the Cf-252 source will be used to shield against the neutrons and gamma radiations. A solution containing Ac-225 will be generated from the apparatus containing the neutron source; nickel and RaCl2 cylinders will generate batches that will require chemical separation. Once Ac-225 is separated, the radiation dose will be prepared to use for patients. Further studies are needed to determine the amount of Ra in the RaCl2 solution/target. An automated system will be required to optimize the user radiation exposure to unnecessary radiation dose [31].
Two of the most important fragments/isotopes generated in the solution containing Ac-225 are Ac-224 (
It has been demonstrated by employing GATE, an open-source Monte Carlo simulation toolkit, that the production of Ac-225 from a neutron generator is feasible. This work suggests that a design consisting of three concentric cylinders, the innermost a Cf-252 neutron source, the middle nickel cylinder acting as a proton-producing target and the outer cylinder a RaCl2 target may provide a feasible design outline for an Ac-225 generator. The optimum dimensions for the proton-producing target are 0.05 mm nickel and the optimum RaCl2 thickness to produce Ac-225 was determined to be 5 mm. Such a design concept may help address the issue of a lack of radiopharmaceutical production to address worldwide demand, particularly in low and middle-income countries.