Portable radiation detectors are important for measuring surface contamination in applications related to radiation protection, nuclear facilities, and environmental monitoring. In this work, a newly portable phoswich detector for simultaneous alpha and beta particles contamination monitoring was developed using a silicon photomultiplier (SiPM) sensor. The detector was designed from large area of (ZnS:Ag)/ (EJ-212) phoswich scintillator coupled with a very small area of SiPM sensor through a truncated cone light guide. The detector was integrated with a portable, lightweight, low-cost, and Wi-Fi-enabled ArduSiPM signal acquisition system. Detector's performance with radiation demonstrates a low background level (0.07 ± 0.02 cps) with a consistent and stable detection response over 4 hours. Additionally, the detector showed an absolute detection efficiency of 28.85 ± 0.41% at a distance of 5 mm for alpha particles from the 241Am source and 19.78 ± 0.83% for beta particles from the 99Tc source. The Minimum Detectable Activity (MDA) values were found to be a 0.39 Bq/cm2 and 0.68 Bq/cm2 for alpha particles and beta particles, respectively. The minimum particle transmission thresholds through the detector entrance structure were estimated to be about 2.1 MeV and 20 keV for alpha and beta particles, respectively. Because of the varying scintillation decay durations and deposited energies in the phoswich scintillator, the detector can differentiate between alpha and beta particles based on variations in their pulse-height spectral distributions, which makes it appropriate for use in mixed radiation fields.
The objective of this work is to improve the method to check the temperature calibration of automated thermoluminescence (TL) / optically stimulated luminescence (OSL) readers based on the luminescence lifetime of Al2O3:C. The improvements were achieved by: (a) performing the POSL measurements continuously during heating, and (b) using blue stimulated OSL from deep traps, instead of the green stimulated OSL from the main dosimetric traps. The first approach, which is currently possible only using the lexsygsmart readers, allows for more efficient data acquisition compared to the method in which the POSL is taken at the end of the heating stage. Another advantage is that the lifetime data is obtained with better temperature resolution. The second approach allowed measurements of the lifetime up to 200°C, a temperature at which the signal from the main dosimetric trap is too weak to allow meaningful results. An additional benefit of the proposed improved method is that several measurements can be taken repeatedly using the same sample without the need to prepare the samples again. The data from two lexsygsmart readers indicate a discrepancy in temperature calibration that requires correction by an offset and a scaling factor.
A method for reconstructing linear energy transfer (LET) fluence spectra of a complex cosmic radiation field using LiF-based fluorescent nuclear track detectors (FNTD) was developed. It is based on the relationship between track intensity and LET and includes corrections for the particle’s angle of incidence. The method was successfully applied to the analysis of the detectors exposed to radiation at lunar orbit in the frame of the MARE project (part of the NASA Artemis I mission).The obtained spectra were compared with the data from etched plastic nuclear track detectors (PNTD), which were located in the same places as FNTDs. The spectra measured with both detector systems agreed, with the exception of LETH2O >100 keV/μm, where FNTD overestimated the PNTD. This, in turn, caused overestimation of the dose (by up to 30%) and quality factor (by about 15%). When calculations were restricted to LETH2O<100 keV/μm, the agreement was within a few percent for dose and within 15% for quality factor.
Thallium-doped cesium iodide (CsI(Tl)) crystals are widely used in nuclear radiation detection and high-energy physics experiments, but their scintillation light yield is highly temperature-dependent, which impairs detector stability and accuracy in variable-temperature environments. To address this issue, we propose a high-accuracy method for determining the intrinsic relative light yield of CsI(Tl) crystals. Geant4 simulations confirm that the photon collection efficiency of small-sized CsI(Tl) crystals (1cm3) can be treated as a constant, eliminating wavelength interference and simplifying the measurement system. We decompose the photodiode (PD) quantum efficiency into wavelength-dependent (Q1(λ)) and temperature-dependent (Q2(T)) components, and systematically characterize their temperature behaviors as well as the wavelength-averaged quantum efficiency (Q(T)¯). Using this method, we measure the intrinsic relative light yield of CsI(Tl) crystals over the temperature range of −40 °C to 40 °C. The results show a non-monotonic temperature dependence: the relative light yield increases quadratically by 16% from −40 °C to 24 °C (peaking near 24 °C), and then decreases slightly by 0.8% as the temperature rises to 40 °C. Compared with conventional methods using photomultiplier tubes or silicon photomultipliers, our approach avoids temperature coupling interference from readout devices, achieving higher measurement accuracy with simplified procedures. This study provides accurate key parameters and a reliable theoretical basis for the design, calibration, and optimization of CsI(Tl)-based energy-spectrum detectors applied in complex temperature-varying scenarios such as space exploration and deep-sea surveys.
3D localization of radioactive material leakage sources is inherently challenging. In this study, we propose a Bayesian Fusion Neural Network (BFNN) for precise 3D localization by integrating information at multiple localization granularities. Specifically, region-level localization obtained from a robust multi-task learning architecture is fused with direct fine-grained 3D localization through Gaussian spatial priors, thereby refining the localization results and improving both region-level and coordinate-level accuracy. A 2D Gaussian distribution is introduced to model the spatial probability density of leakage sources within container regions. Through a Bayesian fusion (BF) mechanism, data-driven predictive confidence is effectively combined with spatial prior knowledge, enabling refined predictions and improved robustness. The proposed framework is evaluated under various distribution shifts between the training and testing data to emulate practical deployment scenarios and assess its generalization capability. Experimental results demonstrate that, even under complex noise conditions and mismatched data distributions, the proposed method consistently achieves more accurate 3D localization and outperforms conventional baseline models, providing a robust and intelligent solution for nuclear leakage localization.
This study presents a systematic investigation of neutron source-term modeling for a GE PETtrace-8 cyclotron, focusing on proton irradiation of an [18O]H2O target to produce 18F. Monte Carlo simulations were performed with PHITS 3.33 and MCNP6.3.0 to update previously reported neutron source terms and quantify code-dependent discrepancies associated with different nuclear data libraries and reaction models, particularly in the low-energy regime (<20 MeV). Neutron energy and angular distributions were computed using equivalent tally methodologies in both codes, considering ENDF/B-VIII.0, JENDL-5, and TENDL-2021 libraries, as well as CEM03.03 in MCNP and INCL4.6+GEM in PHITS. While the calculated neutron spectra exhibit similar overall shapes, significant differences are observed in angular distributions and absolute fluence values, even when the same evaluated data are used. These discrepancies are mainly attributed to differences in intranuclear cascade and pre-equilibrium models, which strongly affect neutron multiplicities and emission anisotropies in low-energy proton-induced reactions. Among all configurations, TENDL-2021-based simulations show the best agreement with the manufacturer's reference data. The results also indicate that MCNP systematically predicts harder spectra and enhanced backward emission relative to PHITS, underscoring the sensitivity of source-term characterization to the underlying physics models. Overall, the study demonstrates that inter-code discrepancies are governed not only by the choice of nuclear data libraries but also by model implementations, highlighting the need for careful benchmarking and validation. The TENDL-based source term obtained here is proposed as a consistent baseline for preliminary neutron-field characterization and shielding design in GE PETtrace-8 cyclotron facilities operating up to 16.5 MeV.
Unfolding the neutron energy spectrum from proton-recoil data requires a detector response matrix, which has conventionally been obtained by Monte Carlo particle transport — a step that is computationally costly and injects statistical noise into the inversion. We remove that step. Retaining the B-spline expansion of the neutron spectrum introduced for organic-scintillator unfolding, we instead build the response matrix of a proton recoil telescope (PRT) from a recently derived closed-form analytical response, which expresses the equal-radius, on-axis response in terms of elliptic integrals and includes proton energy loss in the converter. The result is a fully deterministic pipeline: the response matrix follows directly from closed-form expressions, with no Monte Carlo sampling at any stage, free of statistical fluctuations and reproducible to machine precision. The inversion is regularized by a second-derivative Tikhonov penalty with automatic L-curve selection of the regularization parameter, non-negativity is enforced exactly by bounded-variable least squares, and Poisson counting statistics are accounted for through a weighted (whitened) least-squares formulation. A first independent, out-of-model validation is provided: proton-recoil spectra generated by Geant4 transport at two neutron energies — 2.45 MeV (D–D), for five radiator thicknesses between 5 and 100 μm, and 14.1 MeV (D–T) — are unfolded with the analytical matrix and recover the neutron lines to within ≤4% of the true energies, without forward and inverse operators matched by construction; a joint two-line unfolding resolves both lines simultaneously. Matched-operator synthetic benchmarks, which characterize the inversion rather than the physics of the kernel, localize well-separated peaks to better than 0.15 MeV and yield bootstrap confidence intervals; peaks overlapping within their widths are not resolved. The complete pipeline runs in seconds — orders of magnitude faster than a direct Monte Carlo construction of an equivalent response matrix — making it well-suited to iterative detector design and rapid-turnaround neutron diagnostics.
Accurate and rapid measurement of beta surface contamination is essential during the decommissioning of nuclear facilities. However, conventional Geiger–Müller (GM) survey meters have limited detection areas and cannot measure contamination inside piping, resulting in significant labor and measurement inefficiency. In this study, we developed several plastic scintillation fiber (PSF)–based detectors designed for wide-area surface monitoring, in-pipe contamination measurement, and drainage-channel monitoring. The PSF enables radiation detection along its entire length, allowing large-area scanning and insertion into narrow spaces. For surface measurements, a 10-m PSF arranged in a zigzag configuration was mounted under a cart, enabling efficient mapping of beta contamination over asphalt floor surfaces in Fukushima (demonstration up to 5 m × 5 m). For piping applications, thin and flexible PSF probes were developed, enabling direct insertion into pipes with an inner diameter of 8 mm and achieving detection limits below the regulatory threshold of 4 Bq/cm2 within 1 min. Additionally, a waterproof PSF system was developed for monitoring radioactive contamination in drainage channels and successfully detected temporal fluctuations, including rainfall-induced increases in activity. Overall, the developed PSF systems provide practical, versatile, and highly efficient alternatives to conventional survey meters for contamination monitoring in nuclear decommissioning environments.
Image-guided radiation therapy (IGRT) techniques using computed tomography (CT) and cone-beam CT (CBCT) expose patients to additional radiation. While several dosimetric approaches have been used in earlier research, there is a need for a comprehensive, multi-modality evaluation of absolute organ doses that fully accounts for tissue inhomogeneities using thermoluminescent dosimeters (TLDs). This study aims to compare absolute absorbed doses to 18 critical organs using TLDs within an anthropomorphic phantom across five distinct IGRT techniques on a single platform. To evaluate the radiation doses to key organs in the head-and-neck, thorax, and pelvis, TLDs were placed inside matching anatomical areas of the phantom. For each region, CT, kV-MV port, and CBCT scans (using 200° and 360°, both filtered and unfiltered) were conducted with only that region's TLDs inserted. The absorbed doses were then measured for each modality and body site. Each measurement was repeated three times per technique and region, with mean values reported as the per-fraction dose. Cumulative imaging doses were estimated by multiplying these averages by standard numbers of therapeutic fractions. Dose contributions varied with imaging technique and settings, with standard deviations of per-fraction doses ranging from ±0.002 to ±0.049 cGy. Mean total regional absorbed doses (Dregion) for complete treatment courses across all monitored organs-at-risk were: 0.66 cGy (CT, single fraction), 1.06 cGy (kV port), 1.34 cGy (unfiltered CBCT), 46.07 cGy (filtered CBCT), and 94.53 cGy (MV port). Doses delivered to critical organs vary widely between IGRT protocols. CBCT, particularly with larger arc angles and higher mAs, delivers the largest dose. Clinical decisions should weigh the need for precise imaging against patient safety by selecting parameters consistent with the ALARA (As Low As Reasonably Achievable) principle.
Although magnetic resonance guided radiotherapy is considered among the most promising techniques for image-guided radiotherapy, the inherent complexity of magnetic field interactions with charged particles underscores the need for independent computational tools to validate dose distribution calculations. This study characterizes and validates the one-dimensional dosimetric accuracy of a self-developed voxelized geometry manager integrated with the PENELOPE Monte Carlo code -named Voxgeom-by comparing estimated on-axis depth-dose distributions against experimental measurements and Monaco® treatment planning system calculations for the Elekta Unity® MR-linac. Experimental measurements were performed using a PTW 60019 microDiamond and a PTW 31021 ionization chamber in a PTW BeamScan MR water phantom. The comparative analysis revealed promising agreement between Monte Carlo estimations and experimental measurements in a homogeneous medium, with mean relative deviations below 3% and χ2 test values below 1. Detector-specific analysis showed a better agreement for the microDiamond detector, while the ionization chamber exhibited non negligible discrepancies, primarily within the build-up region. An additional assessment was carried out using a solid water phantom with a lung-like gap in order to emulate inhomogeneities of the human body. In this case, mean relative deviations below 5% were observed when compared to Monaco® calculations, and preliminary reference dose calculations report differences up to 6%. The results presented in this study highlights the detector-dependent response variations in magnetic field environments and establishes the feasibility of Monte Carlo method as an independent dosimetric verification tool in MR-linac units, supporting enhanced quality assurance protocols in clinical practice.
Silicon photomultipliers (SiPMs) operated in high-energy physics facilities for a long time need to be in situ tested using specified light sources. A novel light source for this purpose is suggested. The source is based on scintillations due to natural radioactivity of the 176Lu isotope in Ce-doped lutetium yttrium oxyorthosilicate (LYSO:Ce) single crystals, whereas the light is delivered to the SiPM under testing via a bundle of optical fibers. Light emission and delivery are studied, and the tunability of light intensity by modifying the bundle configuration is considered.
Due to the extremely short penetration depth of tritium beta rays, their direct detection remains challenging. This study demonstrates direct detection of low-energy beta particles from tritium decay using a novel outer-layer scintillation (OLS) fiber, consisting of a surface scintillation layer, a cladding layer, and a wavelength-shifting core. In addition, enhanced photon collection efficiency using a fiber-bundle configuration is demonstrated.Tritium beta-ray detection was verified using a graphite plate containing tritium. By inserting a thin plastic film to selectively attenuate beta particles, pulse-height spectra of the OLS fiber were compared with and without the film. A significant increase in the count rate in the low pulse-height region was observed without the film, clearly indicating the detection of low-energy beta particles.Although a fraction of scintillation photons generated in the outer layer does not reach or get absorbed by the wavelength-shifting core in a single fiber, such escaped photons can be partially recovered by the adjacent fibers within the fiber-bundle configuration. This so-called fiber-bundle effect was examined using external scintillation photons emitted from a piece of OLS fiber irradiated with alpha particles. An increase in the detected signal was observed when the piece of OLS fiber on the alpha particle source was placed near the fiber bundle, despite the absence of direct optical coupling, indicating that scintillation photons generated in the piece of fiber were captured by adjacent fibers within the fiber bundle.These results demonstrate that OLS fiber bundles provide an effective approach to real-time, in situ monitoring of tritium in aqueous environments, with improved detection efficiency through enhanced photon collection.
This study theoretically evaluates the detection time window for assessing internal dose following the inhalation of radionuclides released in a postulated accident scenario at the European Spallation Source (ESS). The assessment focuses on gamma-emitting radionuclides and investigates the feasibility of high-resolution whole-body counting (WBC) and gamma spectrometric analysis of 24-h urinary excretion samples. In a study, an adult individual is assumed to inhale radionuclides released at a single point in time near the ESS. No atmospheric dispersion, dilution, or radioactive decay during transport is considered, and all radionuclides are inhaled simultaneously and distributed homogeneously in the body. In the modelling framework, uncertainty in the physical and chemical properties of potential releases is addressed by exploring multiple parameter combinations that describe different particle characteristics. Semi-synthetic spectra of radionuclides, selected by their radiotoxicity, were generated by coupling time-dependent whole-body retention and diurnal urinary excretion data from the ICRP Viewer with spectra generated in Nucleonica for representative detector configurations. Minimum detectable activities (MDA) were calculated for hypothetical intake scenarios (1–1000 mSv) and post-inhalation times (1–100 days). These were used to produce time-dependent minimum detectable doses (MDD) to assess possible detection windows. Results show that biokinetic variability is the dominant factor influencing detectability, causing differences of 2–3 orders of magnitude. High-resolution WBC consistently outperforms 24-h urinary excretion measurements in terms of lower MDD values and longer detection windows. These findings highlight the importance of biokinetic characterization for dose assessment and the value of maintaining WBC capability for emergency preparedness.
Accurate energy calibration of multi-pixel energy-resolving photon-counting detectors (ERPCDs) remains a critical challenge in X-ray detection and imaging, as calibration errors directly degrade the accuracy of photon energy discrimination. In practical clinical environments, electronic noise, temperature sensitivity, and the unpredictable effects of low-dose radiation on electronic components significantly distort the robustness of the measured energy spectral response, further exacerbating this challenge. In this study, we propose a pixel-wise energy calibration method for multi-pixel ERPCDs based on the spectral response characteristics of an X-ray tube. Precise energy calibration is achieved by establishing a quantitative mapping between simulated spectra corrected for detector response distortions and experimentally measured spectra that inherently include these effects. The proposed approach quantitatively analyzes the influence of detector response distortion on the measured spectra and constructs a distortion correction model under polychromatic X-ray irradiation. A global optimization strategy based on particle swarm optimization is employed to jointly fit the simulated and measured spectra, ultimately establishing a unique and consistent mapping between the detector digital-to-analog converter (DAC) settings and photon energy for each pixel. Numerical simulations and experimental results show excellent agreement, validating the accuracy and robustness of the proposed calibration method. This work provides a reliable and practical solution for routine energy calibration of multi-pixel ERPCDs and is expected to significantly advance high-precision X-ray spectral imaging for clinical diagnosis and industrial nondestructive testing.
