An ultrahot neutron beam shaping device based on a three-layer fluoride structure

Abstract

The application relates to the technical field of neutron beam shaping, in particular to an epithermal neutron beam shaping device based on a three-layer fluoride structure. The technical scheme comprises a moderator, the moderator is sequentially composed of a first bismuth fluoride layer, a main magnesium fluoride layer and a second bismuth fluoride layer along a neutron incident direction, the thickness of the first bismuth fluoride layer is 1.8-2.2 cm, the thickness of the main magnesium fluoride layer is 18.5-20.5 cm, and the thickness of the second bismuth fluoride layer is 4.0-5.2 cm. The three-layer structure of only two materials is used, the thickness of each layer is kept in the range of easy processing size, the front and rear bismuth fluoride layers are cooperated, the epithermal neutron beam output meeting the BNCT treatment standard is realized, and the structure simplification, manufacturing convenience and beam quality are considered.

Description

Technical Field

[0001] This invention relates to the field of neutron beam shaping technology, and more particularly to an ultrathermal neutron beam shaping device based on a three-layer fluoride structure. Background Technology

[0002] Boron neutron capture therapy (BNCT) is a binary targeted radiotherapy technique. Its basic principle is to preferentially enrich boron-containing drugs in tumor cells, and then irradiate the lesion area with a neutron beam. 10 B(n,α) 7 Li nuclear reactions release alpha particles with high energy transfer line density (LET) and 7 Li heavy ions selectively kill tumor cells while causing less damage to surrounding normal tissues.

[0003] BNCT therapy places strict requirements on the quality of the neutron beam. The required neutron beam energy range should be within the hyperthermic neutron energy region, as hyperthermic neutrons have a suitable penetration depth, allowing them to be thermalized upon reaching deep tumor tissue and trigger a boron neutron capture reaction. However, the initial neutron beam produced by reactor or accelerator neutron sources consists of mixed-energy neutrons with energies ranging from 10... -3 eV to 10 7 eV, in which both fast neutron (>10 keV) and thermal neutron (<0.5 eV) components are detrimental to treatment—fast neutrons cause a non-selective dose burden on normal tissues, while thermal neutrons are largely absorbed in superficial tissues and cannot reach deep tumors. Therefore, the core function of BSA is to convert mixed-spectrum neutrons into a therapeutically suitable ultrathermal neutron beam, while suppressing both fast and thermal neutron components and controlling the accompanying gamma-ray dose.

[0004] There are two main technical approaches to the design of moderators in existing BSAs. One approach uses a combination of multiple layers of different materials, such as a complex structure with up to six layers including lead fluoride / bismuth fluoride / magnesium fluoride / beryllium oxide / heavy water. While this design can achieve good neutron spectrum modulation, it has the following prominent problems: (1) the large number of material types leads to high procurement, processing and assembly costs; (2) some layers are extremely thin, <1 cm, which contributes very little to neutron performance but significantly increases manufacturing difficulty and tolerance control requirements; (3) some materials, such as heavy water D2O, require sealed containers, introducing additional structural complexity and neutron loss. The other approach uses a single moderator material, such as pure MgF2, which has a simple structure but limited spectrum modulation capability, making it difficult to achieve an ideal balance between superthermal neutron flux and beam quality. Therefore, this application proposes a superthermal neutron beam shaping device based on a three-layer fluoride structure. Summary of the Invention

[0005] The purpose of this invention is to address the technical problems in the prior art, such as too many types of materials, complex structures, high manufacturing costs, or insufficient performance, by proposing a superthermal neutron beam shaping device based on a three-layer fluoride structure.

[0006] The technical solution of the present invention is as follows: a superthermal neutron beam shaping device based on a three-layer fluoride structure, comprising a moderator, wherein the moderator is composed of a first bismuth fluoride layer, a magnesium fluoride main moderator layer and a second bismuth fluoride layer in sequence along the neutron incident direction.

[0007] Optionally, the thickness of the first bismuth fluoride layer is 1.8–2.2 cm, the thickness of the magnesium fluoride main moderating layer is 18.5–20.5 cm, and the thickness of the second bismuth fluoride layer is 4.0–5.2 cm.

[0008] Optionally, it also includes a rearward reflection ring disposed behind the neutron source target. The rearward reflection ring is made of magnesium fluoride, with an axial length of 25-35 cm, an inner diameter equal to the target radius, and an outer diameter equal to the outer radius of the reflector.

[0009] Optionally, it may also include a radial reflector disposed around the moderating body, the radial reflector being made of magnesium fluoride.

[0010] Optionally, it also includes a thermal neutron absorbing layer disposed on the emitting end face, the thermal neutron absorbing layer being made of cadmium and having a thickness of 0.005 to 0.02 cm.

[0011] Optionally, it also includes an extended collimating ring disposed at the outlet end of the moderator. The extended collimating ring is made of magnesium fluoride, and its inner hole has a conical structure. The inlet radius is equal to the radius of the moderator, and the outlet radius is equal to the radius of the measuring hole. The thickness of the extended collimating ring is 1 to 20 cm.

[0012] Optionally, the radius of the moderator is 12-18 cm, the outer radius of the reflector is 35-45 cm, the radius of the target is 8-12 cm, and the radius of the exit hole is 5-9 cm.

[0013] Optionally, the moderator may be composed of only two fluorides: bismuth fluoride and magnesium fluoride.

[0014] Optionally, the density of the first bismuth fluoride layer and the second bismuth fluoride layer is 8.0–8.5 g / cm³. 3 The density of the magnesium fluoride primary moderating layer is 2.8–3.2 g / cm³. 3 .

[0015] Compared with the prior art, this application includes at least one of the following beneficial technical effects:

[0016] This invention uses only two moderating materials, BiF3 and MgF2, to form a three-layer symmetrical structure. Reducing the number of materials involved in the procurement process decreases the variety of materials and the number of suppliers, thus lowering the complexity of procurement management. In the processing stage, only processing procedures need to be established for these two materials, reducing process preparation time and equipment adjustment frequency. In the assembly stage, fewer component types and simpler interface relationships simplify the assembly process. These factors combined effectively control the overall manufacturing cost.

[0017] In this invention, each layer of the three-layer structure has a thickness of 2.0 cm or more, a thickness range suitable for conventional machining. Thicker components exhibit better structural rigidity during clamping, positioning, and machining, making them less prone to deformation. Dimensional tolerances for each layer are easily guaranteed during machining, eliminating the need for special processing techniques for thin sheet materials. In the assembly stage, the handling, stacking, and positioning of thicker components are simple, and the fit accuracy between layers easily meets design requirements. These features allow the manufacturing process of this invention to operate independently of high-precision thin-sheet machining equipment, reducing processing difficulty and manufacturing costs.

[0018] In this invention, bismuth fluoride layers are located at both ends of the moderator, forming a three-layer configuration that encloses the magnesium fluoride master moderator. The first bismuth fluoride layer at the incident end utilizes the inelastic scattering cross-section provided by the high atomic number of bismuth to initially moderate the incident high-energy fast neutrons, reducing the average energy of neutrons entering the master moderator. The second bismuth fluoride layer at the exit end performs tail-shaping on the neutron energy spectrum after it has been moderated by the master moderator, maximizing the neutron flux in the hyperthermal region through the energy spectrum modulation effect of bismuth. Each of the three layers performs its specific function: the front layer shares the moderation load of the master moderator, while the rear layer supplements the master moderator's shortcomings in fine-tuning the energy spectrum.

[0019] This invention, through the synergistic effect of the aforementioned three-layer fluoride structure, achieves a hyperthermal neutron beam at the outlet that meets the clinical requirements for BNCT treatment. Monte Carlo simulations have verified that all quality indicators of the outlet neutron beam, including hyperthermal neutron flux, fast neutron dose ratio, gamma dose ratio, thermal neutron contamination level, and beam forward bias, meet the constraints stipulated for BNCT treatment, ensuring the usability and safety of the device in a clinical setting.

[0020] In summary, this invention, with its simple three-layer structure using only two materials, achieves superthermal neutron beam output that meets BNCT treatment standards by maintaining the thickness of each layer within an easily machinable range and through the division of labor and synergy between the front and rear bismuth fluoride layers, while keeping the thickness of each layer within an easily machinable range. It achieves a balance between structural simplification, manufacturing convenience, and beam quality. Attached Figure Description

[0021] Figure 1 A schematic diagram of the longitudinal cross-sectional geometry of the BSA device of the present invention is shown.

[0022] Figure label:

[0023] 1. First bismuth fluoride layer; 2. Magnesium fluoride main moderator layer; 3. Second bismuth fluoride layer; 4. Backward reflector ring; 5. Radial reflector; 6. Thermal neutron absorber layer; 7. Extended collimation ring; 8. Lithium target; 9. Accelerator tube. Detailed Implementation

[0024] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0025] Example 1

[0026] I. Neutron Source Parameters. This embodiment of the invention uses a Li(p,n) reaction neutron source as the neutron source. The proton energy is 2.5 MeV, and the proton beam current is 20 mA. The lithium target 8 adopts a layered structure, consisting of a copper backplate, a cooling water layer, and a copper frontplate. The proton beam bombards the lithium target 8 through the accelerating tube 9 to produce mixed-energy spectral neutrons, with neutron energies ranging from 10... -3 eV to 10 7 eV.

[0027] II. BSA Geometry and Parameters. The overall geometric parameters of the BSA device in this embodiment are shown in Table 1.

[0028] Table 1 BSA Geometric Parameters

[0029]

[0030] The moderator is composed of a first bismuth fluoride layer 1, a magnesium fluoride main moderator layer 2, and a second bismuth fluoride layer 3 along the neutron incident direction (positive z-axis). The specific parameters of each layer are shown in Table 2.

[0031] Table 2. Moderator sequence parameters

[0032]

[0033]

[0034] The three-layer structure described above constitutes the core moderator of this invention. The entire moderator is composed of only two fluorides, bismuth fluoride and magnesium fluoride, and does not contain any other moderator materials, thereby minimizing the types of materials and manufacturing costs while ensuring performance.

[0035] III. Detailed structural and functional descriptions of each component.

[0036] 1. Detailed description of the three-layer structure of the moderating body.

[0037] (1) The first bismuth fluoride layer 1 is located at the incident end of the moderator, that is, the position closest to the neutron source lithium target 8. The thickness of this layer is 1.8 to 2.2 cm, and in this embodiment it is 2.0 cm, with a density of 8.30 g / cm³. 3 .

[0038] Bismuth fluoride was chosen as the moderator material in this layer based on the following physical mechanism: Bismuth has a high atomic number (Z=83) and a large inelastic scattering cross section for high-energy fast neutrons (energy >10 keV). Fast neutrons interact with the bismuth nucleus through inelastic scattering, converting some of their kinetic energy into the internal energy of the nucleus, thus rapidly reducing the neutron energy. This process plays a crucial role in subsequent moderation—if the energy of fast neutrons cannot be effectively reduced at the inlet, the subsequent magnesium fluoride main moderator layer 2 will bear an excessive moderation burden, potentially requiring an increase in its thickness, thereby reducing the hyperthermal neutron emission flux.

[0039] The thickness of this layer is set at 1.8–2.2 cm, based on the following design considerations: If the thickness is less than 1.8 cm, the initial slowing down of fast neutrons will be insufficient, allowing a large number of high-energy fast neutrons to enter the main slowing layer. This would result in a high fast neutron content at the exit of the main slowing layer, potentially exceeding the fast neutron dose ratio required for BNCT treatment (≤7.00 × 10⁻⁶). -13 Gy·cm 2 If the thickness exceeds 2.2 cm, the absorption of neutrons by bismuth (radiative trapping cross section) begins to increase significantly, leading to a decrease in the superthermal neutron flux and a reduction in the superthermal neutron conversion efficiency. Therefore, a thickness range of 1.8–2.2 cm represents the optimal balance between sufficiently slowing down fast neutrons and minimizing neutron absorption losses.

[0040] Furthermore, bismuth fluoride is chemically stable and does not react with air or moisture at room temperature and pressure, requiring no special sealed packaging, making it convenient for long-term storage and use. Bismuth fluoride materials can be prepared into dense bulk materials using a hot-pressing sintering process, and then cut into thin layers of the required thickness using CNC machining, with processing accuracy controllable within ±0.05 mm.

[0041] Monte Carlo simulations verified that with a layer thickness of 2.0 cm, the average energy of fast neutrons >10 keV at the incident end decreased to approximately 3.2 MeV after passing through this layer, reducing the total number of fast neutrons by about 62%. This effectively alleviated the moderation burden on the magnesium fluoride main moderator layer 2, creating favorable conditions for subsequent hyperthermal neutron conversion, ultimately achieving 1.67 × 10⁻⁶ keV. 9n / cm 2 The superthermal neutron flux is 6.75 × 10⁻⁶ / s and 6.75 × 10⁻⁶. -13 Gy·cm 2 Both have a fast neutron dose ratio that is superior to the beam quality requirements of BNCT therapy.

[0042] (2) The magnesium fluoride main moderation layer 2 is located after the first bismuth fluoride layer 1 and is the core moderation region in the three-layer structure. The thickness of this layer is 18.5–20.5 cm, preferably 19.6 cm, and the density is 3.00 g / cm³. 3 .

[0043] Magnesium fluoride was selected as the main moderator material in this layer based on the following physical mechanisms: (1) Fluorine has a high elastic scattering cross section, and the energy loss efficiency is high when neutrons undergo elastic scattering with fluorine nuclei. The average logarithmic energy reduction (ξ) per collision is large, which is conducive to the rapid decay of neutron energy; (2) Magnesium is a low atomic number material (Z=12), and the elastic scattering of neutrons with light nuclei has good energy transfer efficiency. A large amount of energy can be lost in a single collision; (3) MgF2 has a moderate macroscopic absorption cross section, and the capture and absorption loss of neutrons is low, ensuring a high neutron utilization rate; (4) The density of MgF2 material is moderate at 3.00 g / cm³. 3 This ensures sufficient scattering nucleus density per unit volume without causing excessive neutron absorption due to excessive density.

[0044] The thickness of this layer, set at 18.5–20.5 cm, was determined after extensive Monte Carlo parameter scanning optimization. The design principle is as follows: After neutrons are initially moderated by the first bismuth fluoride layer 1, their energy drops to approximately 3 MeV. At this point, they enter the magnesium fluoride main moderation layer 2 for further moderation. Along the moderation path of 18.5–20.5 cm, neutrons undergo multiple elastic and inelastic scattering processes with the magnesium fluoride, gradually reducing their energy to the ultrathermal region, i.e., 0.5 eV–10 keV. If the thickness of the magnesium fluoride main moderator layer 2 is less than 18.5 cm, the moderation path is insufficient, and a large number of neutrons with energies above 10 keV remain at the exit, making it difficult to meet the fast neutron dose ratio constraints for BNCT treatment. If the main moderator layer thickness is greater than 20.5 cm, although fast neutron filtration is more thorough, the probability of hyperthermal neutrons being further moderated into thermal neutrons (<0.5 eV) within the moderator increases, leading to a decrease in hyperthermal neutron flux. Simultaneously, the angular distribution of hyperthermal neutron emissions becomes diffuse, and the forward J / Φ index may fall below the requirement of 0.700. Therefore, a thickness range of 18.5–20.5 cm represents the optimal interval between sufficient fast neutron filtration and maximizing hyperthermal neutron flux.

[0045] Monte Carlo simulations verified that, with a layer thickness of 19.6 cm, after passing through the magnesium fluoride main moderator layer 2, the proportion of neutrons in the ultrathermal region (0.5 eV–10 keV) reaches over 68% of the total emitted neutrons, the proportion of thermal neutrons (<0.5 eV) is effectively controlled below 3.2%, the thermal / ultrathermal neutron ratio is 0.032, and the proportion of fast neutrons is reduced to a negligible level. The magnesium fluoride main moderator layer 2 contributes over 80% to the entire moderation process and is the most crucial component in the three-layer structure.

[0046] Magnesium fluoride exhibits excellent machinability. With a Mohs hardness of approximately 5.5, it can be precision milled and ground using diamond tools, achieving a surface roughness of Ra≤0.8μm. The 19.6 cm thickness of this layer falls within the range of standard machining dimensions, eliminating the machining accuracy and assembly tolerance issues associated with extremely thin layers (<1 cm), thus significantly reducing manufacturing costs.

[0047] (3) The second bismuth fluoride layer 3 is located at the moderator exit end, i.e., the position closest to the treatment outlet. The thickness of this layer is 4.0 to 5.2 cm, preferably 4.6 cm.

[0048] This layer also uses bismuth fluoride, but its function differs from the first layer. The core function of this layer is to finely shape the tail of the neutron energy spectrum after passing through the magnesium fluoride main moderator layer 2. The specific mechanism is as follows: After passing through the magnesium fluoride main moderator layer 2, most neutrons are in the ultrathermal energy region, but the energy spectrum distribution still exhibits some inhomogeneity—some neutrons are close to the upper limit of 10 keV, while others are close to the lower limit of 0.5 eV. The second bismuth fluoride layer 3 utilizes the inelastic scattering cross-section characteristics of bismuth in the energy spectrum range at the boundary between the ultrathermal and fast energy regions to perform final filtering of the high-energy portion, i.e., the residual fast neutrons near >10 keV. Simultaneously, through appropriate elastic scattering, it further moderates some neutrons with energies slightly higher than 10 keV to the ultrathermal energy region, thereby making the exit energy spectrum more concentrated and the peak more prominent, ultimately maximizing the ultrathermal neutron flux.

[0049] The thickness of this layer is set at 4.0–5.2 cm, preferably 4.6 cm, based on the following design considerations: If the thickness is less than 4.0 cm, the energy spectrum shaping effect is insufficient, a small amount of fast neutrons remain at the exit, and the fast neutron dose ratio may approach the constraint limit, resulting in insufficient safety margin; if the thickness is greater than 5.2 cm, the absorption of neutrons in the ultrathermal region by bismuth is enhanced, leading to a significant decrease in ultrathermal neutron flux and reducing the overall conversion efficiency. Therefore, the thickness range of 4.0–5.2 cm represents the optimal balance between adequately filtering residual fast neutrons and maintaining ultrathermal neutron flux.

[0050] Monte Carlo simulations confirmed that with a layer thickness of 4.6 cm, the ultrathermal neutron flux at the exit reaches 1.67 × 10⁻⁶. 9 n / cm 2 / s, which is about 18% higher than the scheme without the second bismuth fluoride layer 3. At the same time, the fast neutron dose is about 1.2 × 10⁻⁶ ppm higher than that without the second layer. -12 Gy·cm 2 Reduced to 6.75×10 -13 Gy·cm 2 The energy spectrum reduction reached 44%, which fully demonstrates the energy spectrum shaping effect of the second bismuth fluoride layer 3.

[0051] 2. Detailed description of the reflector system.

[0052] (1) The rearward reflection ring 4 is located behind the neutron source lithium target 8 (in the region from z=-30 to z=0). The material is magnesium fluoride, the inner diameter is equal to the target radius (10 cm), the outer diameter is equal to the outer radius of the radial reflector 5 (40 cm), and the axial length is 25-35 cm, preferably 30 cm.

[0053] Neutrons produced by neutron source target 8 are emitted into 4π space, with a portion escaping behind the lithium target 8 (away from the beam direction). If left unused, these neutrons would be completely lost, resulting in a significant decrease in neutron source utilization. The function of the backscattering ring 4 is precisely to reflect these escaping neutrons back into the beam direction via elastic scattering, allowing them to re-enter the moderator and participate in the moderation process, thereby improving the overall neutron utilization.

[0054] The length of the backreflector ring 4 is designed to be 25-35 cm because a longer ring increases the probability of back-escaping neutrons being reflected back into the beam. However, due to limitations in the overall size of the device and material costs, the length cannot be increased indefinitely. When the length reaches 30 cm, more than 85% of the back-escaping neutrons are reflected back into the moderator, and the marginal benefit of further increasing the length decreases significantly. Simulation results show that with the addition of the backreflector ring 4, the ultrathermal neutron flux increases by approximately 23% compared to when no reflection ring is added (from approximately 1.36 × 10⁻⁶). 9 n / cm² / s increased to 1.67×10 9 n / cm² / s).

[0055] The back reflection ring 4 uses the same MgF2 material as the main moderator layer. This facilitates unified procurement and processing. In addition, MgF2 itself has excellent neutron reflection performance, with a high elastic scattering cross section and a low absorption cross section, making it an ideal reflective material.

[0056] (2) The radial reflector 5 is located around the moderating body (in the region from r=15 cm to r=40 cm), and the material is magnesium fluoride (MgF2).

[0057] During the passage of neutrons through the moderator, some escape from the side (radial direction). The function of the radial reflector 5 is to reflect these radially escaped neutrons back into the moderator, allowing them to continue participating in the moderation process. The radial reflector 5 covers the entire length of the moderator, effectively recovering radially leaked neutrons and further improving the neutron source utilization.

[0058] The outer radius of the radial reflector 5 is designed to be 40 cm, an optimized result considering both reflection effect and space volume. Simulation results show that with the radial reflector 5 added, the ultrathermal neutron flux increases by approximately 15% compared to without the reflector, from approximately 1.45 × 10⁻⁶. 9 n / cm 2 / s increased to 1.67×10 9 n / cm 2 / s.

[0059] Similar to the rearward reflector ring 4, the radial reflector 5 also uses MgF2 material, which achieves uniformity of material types and simplifies the procurement and processing procedures.

[0060] 3. An extended collimating ring 7 is disposed at the outlet end of the moderating body. The material is magnesium fluoride, and the thickness is 1–20 cm, which is 5.8 cm in this embodiment. Its inner hole has a conical structure, with the inlet radius equal to the moderating body radius and the outlet radius equal to the measuring hole radius.

[0061] The functions of the extended collimation ring 7 mainly include two aspects:

[0062] Neutron beam collimation – The conical inner hole structure makes the emitted neutron beam more concentrated in space, improves the beam forward directionality J / Φ, ensures that the neutron beam has good directionality when it reaches the patient's body surface, and reduces the irradiation of normal tissues caused by lateral scattering;

[0063] Further reflection and moderation – As an extension structure of MgF2 material, the extended collimating ring 7 also has the function of further moderation and reflection of emitted neutrons. In particular, for neutrons that still have some energy near the exit, they can undergo additional scattering with the ring wall in the conical channel, which further optimizes the energy spectrum.

[0064] The thickness of the extended collimation ring 7 is designed to be 1–20 cm, and in this embodiment, it is 5.8 cm. This is based on the following considerations: if the thickness is too small (<1 cm), the collimation and moderation effects are insufficient, and the forward direction and energy spectrum quality are not ideal; if the thickness is too large (>20 cm), the hyperthermal neutrons are excessively moderated into thermal neutrons or absorbed by the material inside the extended collimation ring 7, resulting in a decrease in the outlet flux. After simulation optimization, 5.8 cm is the optimal thickness. At this thickness, the forward direction index J / Φ = 0.715, which meets the BNCT treatment requirement of ≥0.700, while maintaining the maximum hyperthermal neutron flux.

[0065] 4. A thermal neutron absorbing layer 6 is disposed on the emitting end face, and is made of cadmium (Cd) with a thickness of 0.005–0.02 cm, preferably 0.01 cm. Cadmium has an extremely high thermal neutron absorption cross-section. 113 The absorption cross section of Cd for 0.025 eV thermal neutrons is approximately 2 × 10⁻⁶. 4 Barns is an ideal thermal neutron filter material. This layer is located at the very end of the exit face, and its function is to absorb residual thermal neutrons (<0.5 eV) in the emitted neutron beam, preventing thermal neutrons from entering the patient's body and causing unnecessary dose burden on superficial normal tissues.

[0066] The thickness of the thermal neutron absorbing layer 6 is designed to be 0.005–0.02 cm, and in this embodiment, it is 0.01 cm. This is based on the following considerations: A 0.005 cm cadmium layer can absorb approximately 99% of thermal neutrons, but the processing precision requirements for foils thinner than 0.005 cm are high during manufacturing. A thickness of 0.01 cm ensures a thermal neutron absorption rate >99.9% and facilitates the procurement of standard cadmium foil for processing and assembly, making it the optimal choice. When the thickness is increased to 0.02 cm, the thermal neutron absorption rate further increases to nearly 100%, but approximately 1% of the superthermal neutrons are lost due to the small amount of absorption by cadmium in the superthermal energy region, slightly reducing the superthermal neutron flux. Therefore, a thickness range of 0.005–0.02 cm is the optimal range between ensuring sufficient thermal neutron absorption and protecting the superthermal neutron flux.

[0067] Simulation results show that when the thickness of this layer is 0.01 cm, the thermal / superthermal neutron ratio at the outlet decreases from 0.058 without the cadmium layer to 0.032, which is below the limit required for BNCT treatment (≤0.050). Meanwhile, the superthermal neutron flux is only reduced by about 0.8%, which is negligible.

[0068] In this embodiment, the moderator radius, i.e., the inner diameter, is 12–18 cm, preferably 15 cm, and the diameter is 30 cm. The design of the moderator radius directly affects the neutron moderation path length and the beam exit area. If the radius is too small (<12 cm), the lateral size of the moderator is insufficient, and some high-energy neutrons leak from the side without being sufficiently moderated, resulting in a decrease in the hyperthermal neutron conversion efficiency. If the radius is too large (>18 cm), although neutrons in the central region of the moderator can be sufficiently moderated, neutrons in the edge region are over-modulated and absorbed in large quantities, while the overall size of the device and material costs increase significantly. The preferred moderator radius of 15 cm is the optimal value between the moderation effect and material utilization rate.

[0069] The outer radius of the radial reflector 5 is 35–45 cm, preferably 40 cm. The thickness of the radial reflector 5, i.e., the difference between the outer radius of the radial reflector 5 and the radius of the moderator, is 20–30 cm, preferably 25 cm. This thickness ensures that most of the radially leaking neutrons are reflected back to the moderator, and the marginal benefit of further increasing the thickness of the radial reflector 5 is very limited.

[0070] The radius of the lithium target 8 is 8–12 cm, preferably 10 cm. The design of the radius of the lithium target 8 is related to the beam spot size and neutron yield distribution of the neutron source. When the radius of the lithium target 8 is 10 cm, it can cover the beam spot region of the proton beam, with a beam cross-section diameter of about 8 cm, and make the generated neutrons emitted towards the moderator in an approximately point source manner, which is beneficial to the symmetry of the subsequent moderation process.

[0071] The exit aperture radius is 5–9 cm, preferably 7 cm. The exit aperture radius determines the size of the treatment field. A 7 cm exit aperture radius can cover the field size required for most BNCT clinical treatments, while avoiding an excessively large exit aperture that would lead to a decrease in beam quality, i.e., forward direction and spectral purity.

[0072] The following is a detailed explanation of the key points of assembly and manufacturing.

[0073] 1. Material Preparation. Both bismuth fluoride and magnesium fluoride materials are prepared from high-purity (≥99.5%) powder raw materials using a hot-pressing sintering process to form dense bulk materials. Specific process parameters are as follows: hot-pressing temperature: BiF3 controlled at 800℃ and MgF2 controlled at 1100℃; pressure: 30–50 MPa; holding time: 2–4 hours; sintering atmosphere: inert gas protection. The relative density of the sintered bulk material reaches over 99% of the theoretical density, ensuring the stability of neutronics properties.

[0074] 2. Machining. Each moderating layer is precision-machined using CNC machine tools. The first bismuth fluoride layer 1 is 2.0 cm thick, and surface grinding is used to ensure the parallelism of the two surfaces is ≤0.02 mm. The magnesium fluoride main moderating layer 2 is 19.6 cm thick, and external cylindrical grinding is used to ensure the cylindricity is ≤0.05 mm. The second bismuth fluoride layer 3 is 4.6 cm thick, and the machining requirements are the same as the first layer.

[0075] The rearward reflector ring 4 is an annular structure, with its inner and outer circles machined using CNC turning. The inner diameter is 20 cm, the outer diameter is 80 cm, and the axial length is 30 cm. The radial reflector 5 is a cylindrical structure with an inner diameter of 30 cm and an outer diameter of 80 cm, machined using the same process. The inner hole of the extension collimating ring 7 is a tapered structure, requiring shaping using a five-axis CNC machining center to ensure cone angle accuracy and surface finish.

[0076] 3. Assembly Sequence. The assembly sequence of this device is as follows: First, the radial reflector 5 is fixed inside the outer casing of the device; then, the three modestors, namely the first bismuth fluoride layer 1 → the magnesium fluoride main moderator layer 2 → the second bismuth fluoride layer 3, are stacked sequentially inside the radial reflector 5 along the neutron incident direction, with precision positioning pins used between each layer to ensure coaxiality ≤0.5 mm; next, the rearward reflection ring 4 is installed at a predetermined position behind the lithium target 8; then, the extended collimating ring 7 is installed at the exit end of the moderator; finally, the cadmium thermal neutron absorbing layer 6 is installed on the exit end face and fixed by adhesive or compression. No adhesive is needed between the layers; the relative positions are maintained by mechanical compression and positioning pins, facilitating disassembly and maintenance.

[0077] 4. Tolerance Control. To ensure the repeatability and consistency of neutronics performance, the thickness tolerance of each layer is controlled within ±0.1 mm, and the radius / diameter tolerance is controlled within ±0.5 mm. These tolerance requirements fall within the range of conventional machining accuracy for thicknesses greater than 2.0 cm and radii greater than 15 cm, and do not increase special manufacturing costs.

[0078] Neutron transport simulations were performed on the BSA device of this embodiment using the OpenMC Monte Carlo program in conjunction with the ENDF / B-VII.1 nuclear database. The number of simulated particles was 20 million, and the statistical error was controlled within 1%.

[0079] The performance indicators are shown in Table 3.

[0080]

[0081] Table 3 shows the performance indicators, verified with 20 million particles and a 20 mA proton beam. The superthermal neutron conversion efficiency, i.e., the ratio of the superthermal neutron flux at the exit to the total yield of the neutron source, is 1.17%, indicating that the three-layer fluoride structure of this invention has high neutron utilization efficiency.

[0082] The above results fully demonstrate that the present invention, with its simple design using only two materials and a three-layer structure, achieves technical effects that surpass the beam quality requirements of BNCT treatment in all key performance indicators. While simplifying the structure and reducing manufacturing costs, its performance is comparable to or even superior to that of more complex multi-layer design schemes in the prior art.

[0083] Example 2

[0084] The difference between this embodiment and Embodiment 1 is that the thickness of the first bismuth fluoride layer 1 is 1.8 cm. All other structures, materials, and process parameters are the same as in Embodiment 1.

[0085] In this embodiment, due to the slightly thinner thickness of the first bismuth fluoride layer 1, the initial slowing effect of fast neutrons is slightly weaker than in Example 1. Monte Carlo simulations verified that the fast neutron dose ratio at the exit was 6.92 × 10⁻⁶. -13 Gy·cm2 It still satisfies the BNCT constraint condition ≤7.00×10 -13 The superthermal neutron flux is 1.72 × 10⁻⁶. 9 n / cm 2 The flux is slightly higher than in Example 1, as the bismuth layer is thinned, resulting in reduced neutron absorption loss and a slight increase in flux. This example achieves a higher hyperthermal neutron flux while maintaining a fast neutron dose ratio close to the constraint limit, making it suitable for therapeutic scenarios with higher flux requirements and relatively relaxed fast neutron dose ratio requirements.

[0086] Example 3

[0087] The difference between this embodiment and Embodiment 1 is that the thickness of the first bismuth fluoride layer 1 is 2.2 cm. All other structures, materials, and process parameters are the same as in Embodiment 1.

[0088] In this embodiment, due to the slightly thicker thickness of the first bismuth fluoride layer 1, the initial slowing down of fast neutrons is more complete. Monte Carlo simulations verified that the fast neutron dose ratio at the exit is 6.58 × 10⁻⁶. -13 Gy·cm 2 This is superior to Example 1, offering a greater safety margin. The superthermal neutron flux is 1.62 × 10⁻⁶. 9 n / cm 2 / s, slightly lower than Example 1, with a slight increase in neutron absorption loss due to the thicker bismuth layer. This example demonstrates superior fast neutron suppression and is suitable for therapeutic scenarios with more stringent requirements for fast neutron dose ratios.

[0089] Example 4

[0090] The difference between this embodiment and Embodiment 1 is that the thickness of the magnesium fluoride main moderator layer 2 is 18.5 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the slightly thinner thickness of the magnesium fluoride main moderator layer 2, the neutron moderation path is slightly insufficient. Monte Carlo simulations verified that the fast neutron dose ratio at the exit is 6.88 × 10⁻⁶. -13 Gy·cm 2 It is still within the constraint limit of ≤7.00×10 -13 The superthermal neutron flux is 1.70 × 10⁻⁶. 9 n / cm 2 / s, slightly higher than in Example 1. This example is suitable for applications requiring high ultrathermal neutron flux.

[0091] Example 5

[0092] The difference between this embodiment and Embodiment 1 is that the thickness of the magnesium fluoride primary moderator layer 2 is 20.5 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the slightly thicker magnesium fluoride primary moderator layer 2, neutron moderation is more complete, and fast neutron filtration is more thorough. Monte Carlo simulation verification shows that the fast neutron dose ratio at the exit is 6.52 × 10⁻⁶. -13 Gy·cm 2 This is superior to Example 1, with an ultrathermal neutron flux of 1.61 × 10⁻⁶. 9 n / cm 2 / s, slightly lower than in Example 1. The thermal neutron content increased slightly, and the thermal / hyperthermal neutron ratio was 0.036, still satisfying the constraint condition ≤0.050. This example demonstrates excellent performance in fast neutron suppression.

[0093] Example 6

[0094] The difference between this embodiment and Embodiment 1 is that the thickness of the second bismuth fluoride layer 3 is 4.0 cm. All other structures, materials, and process parameters are the same as in Embodiment 1.

[0095] In this embodiment, due to the slightly thinner thickness of the second bismuth fluoride layer 3, the tail-shaping effect of the energy spectrum is slightly weaker. Monte Carlo simulations verified that the ultrathermal neutron flux at the exit is 1.72 × 10⁻⁶. 9 The fast neutron dose ratio (n / cm² / s) is slightly higher than in Example 1, but the fast neutron dose ratio is 6.92 × 10⁻¹³ Gy·cm², close to the constraint limit. This example needs to consider the margin of the fast neutron dose ratio while maintaining the flux advantage.

[0096] Example 7

[0097] The difference between this embodiment and Embodiment 1 is that the thickness of the second bismuth fluoride layer is 5.2 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the slightly thicker second bismuth fluoride layer, the energy spectrum tail shaping is more complete. Monte Carlo simulations verified that the fast neutron dose ratio at the exit is 6.58 × 10⁻⁶. -13 Gy·cm 2 This is superior to Example 1, with an ultrathermal neutron flux of 1.62 × 10⁻⁶. 9 n / cm 2 / s, slightly lower than in Example 1. This example offers a greater safety margin in fast neutron suppression.

[0098] Example 8

[0099] The difference between this embodiment and Embodiment 1 is that the axial length of the back reflection ring 4 is 25 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the shorter length of the back reflection ring 4, some of the back-escaping neutrons are not completely reflected back to the beam direction. Monte Carlo simulations verified that the ultrathermal neutron flux is 1.58 × 10⁻⁶ cm⁻¹. 9 n / cm 2 / s, slightly lower than in Example 1, but still meets the BNCT treatment requirement of ≥5.0×10 8 This embodiment is applicable to application scenarios where the axial space of the device is limited.

[0100] Example 9

[0101] The difference between this embodiment and Embodiment 1 is that the axial length of the back reflection ring 4 is 35 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the increased length of the back reflection ring 4, the probability of back-escaping neutrons being reflected back into the beam direction is further increased. Monte Carlo simulations verified that the ultrathermal neutron flux is 1.70 × 10⁻⁶ cm. 9 n / cm 2 / s, slightly higher than in Example 1. This example is suitable for applications with higher requirements for ultrathermal neutron flux, but the axial length of the device is increased accordingly.

[0102] Example 10

[0103] The difference between this embodiment and Embodiment 1 is that the thickness of the cadmium thermal neutron absorbing layer 6 is 0.005 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the thinner cadmium layer, the thermal neutron absorption rate is approximately 99%. Monte Carlo simulations verified that the thermal / hyperthermal neutron ratio is 0.035, still satisfying the BNCT constraint condition ≤0.050. The hyperthermal neutron flux is 1.68 × 10⁻⁶. 9 n / cm 2 / s, slightly higher than in Example 1. Due to the thinning of the cadmium layer, the absorption loss of superthermal neutrons is slightly reduced. This example is suitable for scenarios with high requirements for superthermal neutron flux, but it is still necessary to pay attention to whether the thermal neutron composition is within the constraint limit.

[0104] Example 11

[0105] The difference between this embodiment and Embodiment 1 is that the thickness of the cadmium thermal neutron absorbing layer 6 is 0.02 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the increased thickness of the cadmium layer, the thermal neutron absorption rate is close to 100%. Monte Carlo simulations verified that the thermal / hyperthermal neutron ratio is 0.030, which is superior to that of Embodiment 1. The hyperthermal neutron flux is 1.65 × 10⁻⁶. 9 n / cm 2 / s, slightly lower than in Example 1, with a slight increase in absorption loss for superthermal neutrons due to the thicker cadmium layer. This example offers a greater safety margin in thermal neutron suppression.

[0106] Example 12

[0107] The difference between this embodiment and Embodiment 1 is that the thickness of the extended collimating ring 7 is 1 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the thinner thickness of the extended collimating ring 7, the collimation effect is relatively limited. Monte Carlo simulations verified that the forward tendency J / Φ = 0.705, still satisfying the BNCT constraint condition ≥ 0.700. The ultrathermal neutron flux is 1.70 × 10⁻⁶. 9 n / cm 2 / s, slightly higher than in Example 1. This example is suitable for application scenarios where the space at the device's outlet end is limited.

[0108] Example 13

[0109] The difference between this embodiment and Embodiment 1 is that the thickness of the extended collimation ring 7 is 20 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, the increased thickness of the extended collimation ring 7 results in a more complete collimation effect. Monte Carlo simulations verified that the forward tendency J / Φ = 0.725, which is superior to Embodiment 1. The ultrathermal neutron flux is 1.63 × 10⁻⁶. 9 n / cm 2 / s, slightly lower than in Example 1, because the superthermal neutrons are partially slowed down or absorbed within the collimation ring. This embodiment offers a greater safety margin in terms of forward directionality and is suitable for therapeutic scenarios with high beam directionality requirements.

[0110] Example 14

[0111] The difference between this embodiment and Embodiment 1 is that the moderator radius is 12 cm, the outer radius of the radial reflector 5 is adjusted to 35 cm, the reflector thickness remains at 23 cm, the lithium target 8 radius is adjusted to 8 cm, and the outlet aperture radius is adjusted to 5 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. In this embodiment, due to the reduced moderator radius and shortened lateral moderation path, some high-energy neutrons leak from the side. Monte Carlo simulations verified that the hyperthermal neutron flux is 1.52 × 10⁻⁶. 9 n / cm 2 / s, fast neutron dose ratio is 6.85×10 -13 Gy·cm 2 The thermal / superthermal neutron ratio is 0.034, and the forward neutron ratio J / Φ = 0.718. All indicators still meet the BNCT constraint conditions. This embodiment is suitable for application scenarios with strict limitations on the overall size of the device.

[0112] Example 15

[0113] The difference between this embodiment and Embodiment 1 is that the radius of the moderator is 18 cm, the outer radius of the radial reflector 5 is adjusted to 45 cm, the reflector thickness is maintained at 27 cm, the radius of the lithium target 8 is adjusted to 12 cm, and the radius of the outlet hole is adjusted to 9 cm. All other structures, materials, and process parameters are the same as in Embodiment 1.

[0114] In this embodiment, due to the increased radius of the moderator, the transverse moderation of neutrons is more complete. Monte Carlo simulations verified that the ultrathermal neutron flux is 1.74 × 10⁻⁶. 9 n / cm 2 / s, slightly higher than in Example 1, with a fast neutron dose ratio of 6.62 × 10⁻⁶. -13 Gy·cm 2 This embodiment is superior to Example 1, with a thermal / superthermal neutron ratio of 0.031 and a forward flux J / Φ = 0.712. All indicators meet the BNCT constraint conditions. This embodiment is suitable for applications with higher requirements for superthermal neutron flux and beam quality, but the overall size of the device and material costs are correspondingly increased.

[0115] Example 16

[0116] The difference between this embodiment and Embodiment 1 is that the moderator material consists only of two fluorides: bismuth fluoride and magnesium fluoride. The thickness of each layer is the median of its respective range: the first bismuth fluoride layer 1 is 2.0 cm, the magnesium fluoride main moderator layer 2 is 19.6 cm, and the second bismuth fluoride layer 3 is 4.6 cm. All other structures, materials, and process parameters are the same as in Embodiment 1. This embodiment aims to verify the core technical solution of the invention—achieving beam quality that meets the requirements of BNCT treatment using only two materials and a three-layer structure. Monte Carlo simulations verified that all performance indicators are consistent with those of Embodiment 1, with a superthermal neutron flux of 1.67 × 10⁻⁶. 9 n / cm 2 / s, fast neutron dose ratio is 6.75×10 -13 Gy·cm 2 The gamma dose ratio was 1.48 × 10⁻⁶. -13 Gy·cm 2 The thermal / superthermal neutron ratio is 0.032, and the forward J / Φ = 0.715, fully meeting the beam quality requirements for BNCT treatment. This embodiment fully demonstrates that the technical solution of the present invention can achieve excellent beam quality without introducing other moderation materials (such as lead fluoride, beryllium oxide, heavy water, etc.), reducing the types of materials to a minimum and significantly lowering manufacturing costs and assembly complexity.

[0117] This invention addresses the technical problems of existing boron neutron capture therapy (BNCT) hyperthermal neutron beam shaping (BSA) devices, such as the wide variety of materials, complex structures, high manufacturing costs, or insufficient performance. It provides a hyperthermal neutron beam shaping device based on a three-layer fluoride structure. The device consists of a first bismuth fluoride layer 1, a magnesium fluoride main moderator layer 2, and a second bismuth fluoride layer 3, arranged sequentially along the neutron incident direction, forming a BiF3 / MgF2 / BiF3 three-layer symmetrical structure. The moderator material is composed only of bismuth fluoride and magnesium fluoride. Through the synergistic effect of the three layers, functional division is achieved: the first bismuth fluoride layer 1 utilizes the high atomic number of bismuth to enhance the inelastic scattering cross section, initially moderating high-energy fast neutrons; the magnesium fluoride main moderator layer 2 utilizes the high elastic scattering cross section of fluorine and the low atomic number of magnesium to moderate fast neutrons to the hyperthermal energy region, while effectively filtering residual fast neutrons; the second bismuth fluoride layer 3 shapes the tail of the neutron energy spectrum, maximizing the hyperthermal neutron flux at the exit. Each layer is at least 2.0 cm thick, falling within a range easily machinable, thus avoiding the machining accuracy and assembly tolerance issues associated with extremely thin layers. Monte Carlo simulations verified that the 2.0 cm BiF3 / 19.6 cm MgF2 / 4.6 cm BiF3 scheme achieved a hyperthermal neutron flux of 1.67 × 10⁻⁶ under a 20 mA proton beam. 9 n / cm 2 / s, fast neutron dose ratio is 6.75×10 -13 Gy·cm 2 The gamma dose ratio was 1.48 × 10⁻⁶. -13 Gy·cm 2 The thermal / hyperthermal neutron ratio is 0.032, and the forward J / Φ is 0.715, all of which meet the beam quality requirements for BNCT treatment. This invention achieves beam quality that meets clinical treatment requirements using only two moderator materials, while simplifying the processing and assembly procedures and reducing manufacturing costs. It provides a simple, reliable, and easily engineering-implemented hyperthermal neutron beam shaping device solution for BNCT.

[0118] The above specific embodiments are merely several optional embodiments of the present invention. Based on the technical solutions of the present invention and the relevant teachings of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A superthermal neutron beam shaping device based on a three-layer fluoride structure, comprising a moderator, characterized in that, The moderator is composed of a first bismuth fluoride layer, a magnesium fluoride main moderator layer, and a second bismuth fluoride layer in sequence along the neutron incident direction.

2. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, The thickness of the first bismuth fluoride layer is 1.8–2.2 cm, the thickness of the magnesium fluoride main moderating layer is 18.5–20.5 cm, and the thickness of the second bismuth fluoride layer is 4.0–5.2 cm.

3. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, It also includes a rearward reflection ring disposed behind the neutron source target. The rearward reflection ring is made of magnesium fluoride, with an axial length of 25-35 cm, an inner diameter equal to the target radius, and an outer diameter equal to the outer radius of the reflector.

4. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, It also includes a radial reflector disposed around the moderating body, the radial reflector being made of magnesium fluoride.

5. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, It also includes a thermal neutron absorbing layer disposed on the emission end face, the thermal neutron absorbing layer being made of cadmium and having a thickness of 0.005 to 0.02 cm.

6. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, It also includes an extended collimating ring disposed at the outlet end of the moderator. The extended collimating ring is made of magnesium fluoride and has a conical inner hole. The inlet radius is equal to the radius of the moderator, and the outlet radius is equal to the radius of the measuring hole. The thickness of the extended collimating ring is 1 to 20 cm.

7. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, The radius of the moderator is 12-18 cm, the outer radius of the reflector is 35-45 cm, the radius of the target is 8-12 cm, and the radius of the exit hole is 5-9 cm.

8. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, The moderator is composed of only two fluorides: bismuth fluoride and magnesium fluoride.

9. The superthermal neutron beam shaping device based on a three-layer fluoride structure according to claim 1, characterized in that, The density of the first bismuth fluoride layer and the second bismuth fluoride layer is 8.0–8.5 g / cm³. 3 The density of the magnesium fluoride primary moderating layer is 2.8–3.2 g / cm³. 3 .

Images