Shielding Design of Distributed X-ray Source

TANG Huaping1,2, CHEN Zhiqiang1, LI Guoyu2, HE Wu2, WANG Biao2, LAI Sheng2
1. Department of Engineering Physics, Tsinghua University, Beijing 100084, China
2. Nuray Technology Company Limited, Changzhou 213200, China

Abstract

[Background]: The shielding structure is an important guarantee for the safe application of X-ray tubes and plays a key role in engineering implementation. Unlike traditional single-focus X-ray tubes, the radiation protection and shielding design of distributed multi-focus X-ray tubes requires targeted research. [Purpose]: This study aims to design and validate an effective shielding structure for distributed X-ray sources to ensure radiation safety while maintaining practical feasibility, providing a viable technical solution for shielding design and protection of distributed X-ray sources. [Methods]: This paper employs Monte Carlo simulation methods to model and investigate the radiation field and shielding structure of distributed X-ray tubes. Simulations were performed with 45 focal spots distributed linearly, operating at 160 kV anode voltage and 15 mA anode current. The study explored lead shielding thicknesses ranging from 1 to 6 mm. [Results]: Simulation results demonstrate that distributed X-ray sources exhibit line-source characteristics. By analyzing leakage dose rate levels across various shielding surfaces at different thicknesses, a lead shielding structure with 5–6 mm thickness on each surface was designed and assembled with an X-ray tube. The results indicate that adopting a uniform thickness shielding design on all surfaces, especially along the focal distribution direction, is feasible for the new distributed X-ray tube. The deviation between simulation and actual measurements was consistently below 25%, validating the reliability of the simulation approach. The points with higher leakage rates were located in the front lower middle and rear lower middle regions, with rates of 2.4 μGy/h and 2.92 μGy/h respectively, consistent with the line-source characteristics and reflection target structure of distributed X-ray tubes. The leakage rates at the top, bottom, left, and right surfaces were lower than 2 μGy/h and 1 μGy/h respectively. [Conclusions]: By adopting a shielding structure with uniform thickness on all sides of the X-ray tube, the leakage dose rate at each point meets the requirement of a leakage dose rate limit of less than 5 μGy/h.

Keywords: Distributed X-ray source; Shielding structure; Monte Carlo simulation; Line-source characteristic; Leakage dose rate

Introduction

In recent years, novel multi-focus distributed X-ray tubes have brought about significant transformations in radiation imaging technology, attracting widespread attention and research. Thanks to the high penetrability of X-rays and the differential attenuation of various materials, X-ray radiation imaging can obtain internal object information and is widely used in medical, security inspection, and non-destructive testing applications. A multi-focus distributed X-ray tube refers to an X-ray tube containing multiple focal spots distributed at different positions within a single vacuum envelope, where each focal spot can independently and controllably generate X-rays [1]. Radiation imaging systems based on distributed X-ray sources can perform perspective imaging of the same object from different angles without requiring source movement, enabling rapid three-dimensional imaging of the object, such as in stationary CT systems [2]. This avoids the issues associated with traditional high-speed slip-ring mechanical structures, offering better stability, higher detection speed, and improved efficiency. These advantages have prompted numerous institutions to pursue research in this area [3-6].

Shielding structures are critical for the safe application of X-ray tubes and play a key role in successful engineering implementation. Current shielding design typically relies on parameter guidance from ICRP reports and charts combined with empirical formulas for X-ray attenuation [7,8], or employs Monte Carlo modeling calculations as guidance [9,10]. However, existing radiation shielding research focuses on single-focus X-ray sources, typically using the focal spot as the center and performing calculations or simulations based on the different energy spectrum distributions and intensity distributions of X-rays in various angular directions [11]. The operating modes and shielding structures of novel multi-focus X-ray sources differ significantly from those of single-focus sources.

Unlike single-point X-ray sources, the shielding design and performance study of multi-focus distributed X-ray sources represent a completely new research topic due to the "non-central" characteristic of X-ray emission positions. Such sources exhibit non-central features approaching those of linear radiation sources [12,13], but their X-ray emission pattern differs from the uniform linear distribution of line sources, instead consisting of intermittent point distributions. Therefore, targeted research on shielding design methods is required. This paper employs Monte Carlo simulation methods to study the radiation field characteristics of a 45-focus distributed X-ray tube under sequential focal spot operation, revealing its line-source characteristics. A shielding design methodology using uniformly thick shielding layers along the focal distribution direction is proposed.

1 Distributed X-ray Tube

The distributed X-ray tube studied in this work features 45 focal spots arranged in a linear distribution on a long strip anode [14], housed within a rectangular frame structure, as shown in [FIGURE:1]. In addition to the rectangular main structure, the tube includes cylindrical high-voltage connection structures, oil cooling connection structures, and auxiliary components such as exhaust pipes and ion pumps.

1.1 Structural Modeling of Distributed X-ray Tube

A schematic diagram of the distributed X-ray tube structure is shown in [FIGURE:2]. The 45 anode focal spots are arranged linearly with equal spacing in a left-right symmetrical structure. The central focal spot position is defined as the origin O, establishing an X-Y-Z coordinate system with the X-ray emission direction designated as forward. To reduce Monte Carlo computational difficulty and improve efficiency, the calculation model was simplified by excluding auxiliary components such as exhaust pipes, ion pumps, and lead terminals.

The shielding structure positions are shown in [FIGURE:2], with distances from focal spot center O to the front, rear, top, bottom, and side surfaces being 94 mm, 81 mm, 254 mm, 178 mm, and 362 mm respectively.

1.2 Operating Parameters of Distributed X-ray Tube

The distributed X-ray tube operates with positive high voltage supplied by a high-voltage generator, multi-channel constant current control equipment (ECS) developed specifically for controlling different cathodes, real-time anode cooling by an oil cooler, and high vacuum maintenance inside the tube by an ion pump controller. The X-ray tube system and equipment setup are shown in [FIGURE:3], with operating parameters listed in [TABLE:1].

TABLE:1 Parameter of Multi-Beam X-ray tube

Item Technical data Maximum anode voltage / kV 160 Maximum anode current / mA 15 Continuous operation power / kW 2.4 Number of focal spots / pcs 45 Single point working pulse width / µs 50~2000

2 Monte Carlo Simulation Design

Based on the distributed X-ray tube structural model shown in [FIGURE:2], shielding structures were positioned at the red frame locations using lead as the shielding material, with dose sampling points set on the outer surfaces of the shielding structures. In this simulation, a multi-focus X-ray tube cathode structure model was established using electron bombardment of the anode target to generate X-rays. The electron source parameters were set according to [TABLE:1], with the anode target material set as tungsten. A 160 keV electron beam bombarded the 45 focal spot positions to generate X-rays with tungsten characteristic peaks and bremsstrahlung [15-18]. Since Monte Carlo programs cannot easily configure multiple electron sources at different positions to sequentially bombard the anode target, 45 separate programs were written, each with a single electron source. The 15 mA current was distributed across the 45 focal spots to achieve 2.4 kW operating power, with final data processing performed through superposition. The simulation yielded an X-ray radiation field with an average energy of approximately 110 keV and angular distribution consistent with line-source characteristics. Uniform thickness shielding was then applied to the six outer surfaces of the tube. When lead shielding thickness on each surface ranged from 1–6 mm, leakage dose rates were analyzed at positions near the center point (point O in [FIGURE:2]) on each surface. The resulting attenuation curves of leakage dose rate versus lead thickness are shown in [FIGURE:4].

The results show that for a 160 kV X-ray tube, each additional 1 mm of lead shielding thickness reduces the leakage dose rate by approximately one order of magnitude. Using the 5 μGy/h leakage dose rate limit as a reference line, shielding thicknesses of 4–6 mm on each surface satisfy radiation protection requirements.

Based on the Monte Carlo results in [FIGURE:4], using uniform lead thickness along the focal distribution direction, the same thickness was applied across each shielding surface area, simplifying shielding structure design, manufacturing, and assembly. After incorporating appropriate safety margins, the designed lead shielding thicknesses for each surface are shown in [TABLE:2].

TABLE:2 Thickness of lead shield at different locations

Location Lead shielding thickness / mm Bottom front 6 Bottom front 6 Bottom 6 Bottom right 6 Bottom right 6

Under the lead shielding thicknesses shown in [TABLE:2], Monte Carlo calculations indicate that leakage dose rates on all surfaces meet the 5 μGy/h requirement. The surface distribution characteristics of leakage dose rates for the front, rear, and bottom surfaces are shown in [FIGURE:5]. Each small square in the figure represents a detector, dividing the entire surface into small area segments to tally leakage X-rays passing through each segment. Due to relative errors inherent in Monte Carlo calculations, these fluctuations cause the distribution edges to appear jagged.

The figures reveal that maximum leakage dose rates on the front and rear panels occur in regions 3–5 cm below the focal spots. The bottom, front lower, rear lower, left lower, and right lower panels have 6 mm shielding thickness, greater than the 5 mm thickness used for the top, rear upper, left upper, and right upper panels, consistent with the reflection target structure where X-ray yield is higher in the forward and downward directions. In the direction along the anode focal distribution, uniform shielding thickness in each region achieves shielding effectiveness meeting safety objectives, demonstrating the line-source characteristics of the 45-focus distributed X-ray source. At both ends of the long anode, leakage dose rates are lower than in the central region, exhibiting the truncation effect at line-source ends.

3 Shielding Structure Design and Testing

Based on Monte Carlo simulation results, shielding structures were designed for the distributed X-ray tube according to the lead shielding thicknesses in [TABLE:2]. After manufacturing, the shielding structures were assembled with the distributed X-ray tube. Leakage rate testing was conducted using standard operating parameters from [TABLE:1], with actual operating conditions set to anode voltage 160 kV and anode current 15 mA (each focal spot pulse width 95 µs, pulse interval 5 µs, pulse current set to 15.8 mA, with 45 focal spots operating sequentially in multiple cycles) to verify the radiation protection effectiveness of the shielding structure.

3.1 Shielding Structure Design

Shielding structure design must consider manufacturing and assembly convenience and feasibility. The design and assembly process proceeded as follows: (1) A square hexahedral structure was built around the X-ray tube with structural frames at the edges; (2) 4 mm thick steel plates were designed at both ends of the tube for mounting the tube to the frame structure; (3) 5 mm thick lead plates were designed for the top, rear upper, rear lower, left upper, and right upper positions; (4) 6 mm thick lead plates were designed for the front upper, front lower, bottom, left lower, right lower, and rear lower positions; (5) Appropriate labyrinth structures were added externally at locations where connection cables and cooling oil pipes pass through. The final shielding structure design is shown in [FIGURE:6].

During actual manufacturing and assembly, seamless joining at the edges of the six lead plates proved difficult. Therefore, lead strips were attached to the inner sides of the structural frame edges to seal assembly gaps and mounting screws, ensuring radiation protection effectiveness.

3.2 Radiation Protection Effectiveness Testing

During shielding testing, the main beam port was sealed with a shielding fixture using 8 mm thick lead sheet overlapping the main beam window by approximately 40 mm. The X-ray source was activated, and an F451p ionizing radiation monitor was used to perform survey measurements at 5 cm from the shielding surface, with data recorded by zone. The leakage dose rate test results are shown in [FIGURE:7], where the right and left panels were essentially identical and not separately marked. The test data demonstrate that leakage dose rates at 5 cm from the shielding surface are all less than 5 μGy/h, achieving the predetermined radiation protection objective.

TABLE:3 Comparison of Simulated and Actual Measured Dose Distribution After Lead Shielding

Location Simulated dose rate / μGy·h-1 Measured dose rate / μGy·h Deviation between measurement and simulation Bottom front 2.4 1.8 -25% Bottom front 2.4 1.8 -25% Bottom 2.4 2.0 -16% Bottom right 2.4 1.9 -20% Bottom right 2.4 2.0 -15% Bottom right 2.4 2.3 -1%

As shown in [TABLE:3], the deviation between actual measurements and simulation results for the 45-focus distributed X-ray tube is within 25%, demonstrating that Monte Carlo simulation methods can effectively guide the development and design of radiation protection and shielding structures for X-ray tubes.

Differences between measured and calculated results likely arise because the simplified top high-voltage connection structure, oil cooling connection structure, exhaust pipes, ion pumps, and other auxiliary components in the simulation, as well as the shielding frame and labyrinth structures, actually block some radiation in practice, causing most measured values to be slightly lower with varying deviations across different locations. Both Monte Carlo simulation and actual test results show that leakage dose rates for the 45-focus distributed X-ray tube are below 5 μGy/h, meeting radiation shielding design requirements and complying with radiation protection dose rate limit standards.

This study employs Monte Carlo simulation methods to establish a computational model of a 45-focus distributed X-ray tube, investigates its X-ray field distribution characteristics as a line source, and proposes a uniform-thickness shielding design methodology for all surfaces, particularly along the focal distribution direction. By modeling shielding bodies with uniform thickness on each surface, the relationship between shielding thickness and leakage dose rate was simulated, revealing that lead shielding thicknesses of 4–6 mm in different regions can satisfy the leakage dose rate limit requirement of 5 μGy/h. Based on these results, a shielding structure with uniform thickness on each surface (5–6 mm varying by surface) was designed and manufactured. Assembly and measurement demonstrated that leakage dose rates at all locations are below the 5 μGy/h radiation protection dose rate limit standard, achieving the radiation protection design objective. This validates the reliability of uniform-thickness shielding design for distributed X-ray sources, particularly along the focal distribution direction, which simplifies shielding design, manufacturing, and assembly. Maximum leakage dose rates appeared in the middle of the front lower and rear lower regions at 2.4 μGy/h and 2.92 μGy/h respectively, consistent with the line-source characteristics and reflection target structure of multi-focus distributed X-ray tubes. Comparison between measured and simulated results shows deviations below 25% with good consistency, demonstrating that Monte Carlo simulation is feasible for designing shielding structures for novel distributed X-ray tubes. This research explores shielding technology for novel distributed X-ray tubes and proposes a uniform-thickness shielding design methodology for all surfaces, particularly along the focal distribution direction, providing technical and experiential references for radiation protection research on novel X-ray tubes.

Author Contributions

TANG Huaping conducted research on the line-source characteristics of the distributed X-ray source radiation field, proposed the uniform-thickness shielding design methodology along the focal distribution direction, guided Monte Carlo simulation and shielding structure design, and performed the main manuscript writing. CHEN Zhiqiang guided technical issues throughout the research process, including the physical model and result analysis for Monte Carlo simulation. LI Guoyu was responsible for developing the distributed X-ray tube shielding structure and participated in assembly and testing verification. HE Wu and WANG Biao performed X-ray source system construction, shielding structure assembly, testing, and data analysis. LAI Sheng performed Monte Carlo simulation modeling and analysis, data processing and analysis, and participated in manuscript writing.

References

1. TANG H P. History and Recent Advances of Distributed X-ray Sources [J]. Vacuum Electronics, 2023, 06: 1-9. DOI: 10.16540/j.cnki.cn11-2485.
2. JIN X, ZHANG S Y, LI L, et al. A Dual-energy Stationary CT System Based on Dual-plane Scanning Mode [J]. Computerized Tomography Theory and Applications, 2020, 29(1): 31-37. DOI:10.15953/j.1004-4140.2020.29.01.04.
3. C. Puett, C. Inscoe, Y. Z. Lee, et al. Phantom-based study exploring the effects of different scatter correction approaches on the reconstructed images generated by contrast-enhanced stationary digital breast tomosynthesis[J]. Journal of Medical Imaging, 2018, 5(1): 013502. DOI: 10.1117/1.JMI.5.1.013502.
4. P. R. Schwoebel1, J. M. Boone, J. Shao. Studies of a prototype linear stationary x-ray source for tomosynthesis imaging[J]. Physics in Medicine and Biology, 2014, 59: 2393–2413. DOI: 10.1088/0031-9155/59/10/2393.
5. Park, Junyoung, Jung, et al. Multi-beam x-ray sources with carbon nanotube emitter for tomosynthesis system[J]. Proceedings of SPIE, 2020, 11312: 113122E. DOI:10.1117/12.2550142.
6. C. J. Tan, C. X. Tang, W. H. Huang, et al. Beam simulation and experiment of multi-focal spot electron gun for distributed X-ray sources[J]. Nuclear Inst. and Methods in Physics Research, A, 2019, 920: 43–49. DOI:10.1016/j.nima.2018.11.039.
7. Zhang L, Chen N, Zeng D G. Different Methods of X-ray Room Shielding Optimization Analysis [J]. China Health Industry, 2015, 12(15): 79-82. DOI: CNKI:SUN:WSCY.0.2015-15-037.
8. Zheng L Z. Industrial Safety and Environmental Protection [J], Industrial Safety and Environmental Protection. 2020, 46(11): 89-92. DOI: 10.3969/j.issn.1001-425X.2020.11.022.
9. Liu R, Yu H W, Zhu Q, et al. Monte Carlo simulation analysis of X-ray density logging shields [J]. Nuclear Techniques, 2024, 47(8): 080202. DOI: 10.11889/j.0253-3219.2024.hjs.47.080202.
10. Zhu Z H, Luo W, Chai F C, et al. MC Simulation of Medical Diagnostic X-rays and Optimization Design of Composite Shielding Materials[J]. Journal of East China University of Technology (Natural Science), 2019, 42(2): 169-172. DOI: 10.3969/j.issn.1674-3504.2019.02.010.
11. H W YANG, T Xun, J M GAO, et al. X-ray shielding estimations for rep-rate high power microwave accelerators [J]. High Power Laser and Particle Beams, 2018, 30(7): 073005. DOI: 10.11884/HPLPB201830.170532.
12. WANG Y, LIU X, HUANG W X, et al. Line focal X-ray source imaging [J]. Acta Physica Sinica, 2016, 65(21), 219501: 1-7. DOI:10.7498/aps.65.219501.
13. NIE X C, LI J, ZHAO P H, et al. Application and Comparison of MCNP Variance Reduction Techniques for Deep Penetration Shielding Calculation [J]. Nuclear Electronics Detection Technology, 2016, 36(7): 729-733. DOI: 10.3969/j.issn.0258-0934.2016.07.016.
14. TANG H P, CHEN Z Q, HE W, et al. Development and Application of a 160kV Distributed X-ray Tube Based on Carbon Nanotube Cathodes [J]. Vacuum Electronics, 2024, 04: 26-32. DOI:10.16540/j.cnki.cn11-2485/tn.2024.04.06.
15. LI W W, WANG L Q. Simulation Optimization Design of Tungsten Anode Target in Reflection Type X-Ray Tubes [J]. Vacuum Electronics, 2023(4): 26-31. DOI: 10.16540/j.cnki.cn11-2485/tn.2023.04.05.
16. ZHAO Y. Simulation of X-Ray Fluorescence Spectrum Distribution in Different Materials Based on Monte Carlo Method [J]. Material Sciences, 2023, 13(12): 1095-1101. DOI: 10.12677/ms.2023.1312122.
17. XIANG A, LIANG L, CHEN R T. Monte Carlo simulation and analysis of X-ray tube's heel effect [J]. Nuclear Techniques, 2016, 39(06): 39-43. DOI: 10.11889.0253-3219.2016.his.39.060202.
18. LIU C F, LI D H, HAO G H, et al. Research on the dosimetric characterize of the new mobile low energy X ray based on Monte Carlo [J]. China Measurement & Test, 2024, 50(9): 174-180. DOI: 10.11857/j.issn.1674-5124.2022070134.