Dosimetric characterization of organic-inorganic hybrid perovskite scintillators for low-energy diagnostic X-ray detection

Perovskite materials have recently emerged as promising candidates for radiation detection due to their strong X-ray absorption, efficient charge transport, and tunable optical properties. However, their performance in scintillation-based dosimetry remains insufficiently explored. This study systematically evaluates the radiation attenuation, radioluminescence, and dosimetric characteristics of two-dimensional organic–inorganic perovskite scintillators, B A 2 P b B r 4 and P E A 2 P b B r 4, in comparison with conventional plastic (E J 212) and inorganic (B G O, L Y S O) scintillators under low-energy X-ray irradiation. The perovskite samples exhibited mass attenuation coefficients significantly higher than those of the plastic scintillator and comparable to dense inorganic crystals, reflecting their high- Z composition. Under X-ray excitation, both materials exhibited narrow radioluminescence emission bands centred near 420 n m. The measured light yields were approximately 9.8 × 1 0 3 and 4.8 × 1 0 3, p h o t o n s M e V − 1 for B A 2 P b B r 4 and P E A 2 P b B r 4, respectively, indicating efficient conversion of absorbed X-ray energy into visible emission. Dosimetric measurements showed excellent dose linearity (R 2 > 0.999), high sensitivity despite small thicknesses, and stable, repeatable responses with relative standard deviations below 1.3 %. The perovskites also maintained consistent dose-rate behaviour across 0.11– 1.12 m G y s − 1. Energy-dependent measurements revealed enhanced responses at lower photon energies, consistent with photoelectric absorption. Notably, differences between B A 2 P b B r 4 and P E A 2 P b B r 4 highlight the ability to tailor energy response through modifications of the organic spacer or inorganic framework, enabling compositionally engineered scintillators optimized for specific dosimetric tasks. Overall, these results show that organic–inorganic perovskite scintillators bridge the performance gap between plastic and inorganic materials, offering high sensitivity, stability, and compositional tunability, positioning them as versatile candidates for next-generation diagnostic X-ray detection and real-time dosimetry.

Perovskite materials have recently attracted significant attention as next-generation candidates for radiation detection owing to their exceptional optical and electronic properties (Wei and Huang, 2019; Yakunin et al., 2015). Their versatility enables operation in both direct and indirect detection modes, supported by high photon absorption coefficients, efficient charge-carrier transport, intrinsic defect tolerance, and cost-effective synthesis (Jeng et al., 2013; Wang et al., 2019; Saiji et al., 2024). Consequently, perovskites have emerged as promising alternatives to conventional detector materials such as amorphous selenium (α -Se), surpassing them in several key detection metrics (Wei et al., 2016). Metal halide perovskites have demonstrated remarkable potential for direct X-ray detection and dosimetry, exhibiting high sensitivity and efficient charge transport (Yakunin et al., 2015; Saiji et al., 2024). For example, methyl-ammonium lead iodide (M A P b I 3) has achieved outstanding X-ray detection sensitivity, reinforcing the suitability of perovskites for medical imaging and radiation monitoring applications (Wei and Huang, 2019). This high sensitivity arises from the large X-ray attenuation coefficients provided by the high-Z elements Pb and I, which allow for efficient stopping of medical-energy X-rays in sub-millimeter to millimeter-thick layers (Wei et al., 2016; Datta et al., 2020; Xu et al., 2021), combined with excellent charge transport properties, including high mobility–lifetime (μ τ) products, long carrier diffusion lengths, and lifetimes that approach or rival those of CdZnTe and HgI 2 (Wei et al., 2016; Deumel et al., 2021; Datta et al., 2020; Musiienko et al., 2021). Despite these promising advances, direct-detection perovskite systems still face intrinsic challenges related to charge collection efficiency, ion migration, and structural instability under prolonged irradiation (Yang et al., 2024). These limitations restrict their long-term applicability and motivate the exploration of indirect detection strategies, such as scintillation-based approaches, which decouple the radiative process from charged transport and thereby enhance operational stability.

In scintillation detection, the absorbed X-ray energy is converted into visible light through electron–hole recombination. The efficiency of this process depends on both the material’s absorption coefficient and its ability to transfer energy radiatively without non-radiative losses. Compared with traditional scintillators, perovskites offer several advantages, including tunable emission spectra, high exciton binding energy, and potential for low-temperature, solution-based fabrication. The structure of two-dimensional (2D) perovskites enhances exciton confinement and promotes efficient radiative recombination, resulting in intense emission bands (Birowosuto et al., 2016) that are attractive for dosimetry and imaging applications.

Two-dimensional perovskite scintillators, such as (B A) 2 P b B r 4 (Sheikh and Nag, 2019; Li et al., 2023) and (P E A) 2 P b B r 4 (Xie et al., 2020), have recently emerged as promising materials for radiation scintillation. Their hybrid organic–inorganic architectures incorporate heavy atoms such as lead and bromine, which give rise to large photon absorption cross sections; in the diagnostic X-ray energy range, where the photoelectric effect dominates, the mass attenuation and mass energy-absorption coefficients scale strongly with atomic number (approximately proportional to Z 4), so Pb–Br perovskites provide substantially higher attenuation than low- Z polymer-based scintillators. Consequently, these perovskites exhibit improved X-ray detection efficiency compared with plastic scintillators. They also display high light output, efficient radiative recombination, and fast nanosecond-scale decay components, producing intense and spectrally stable radioluminescence under X- and γ -ray excitation (Xie et al., 2020; Sheikh and Nag, 2019; Li et al., 2023). These characteristics make two-dimensional perovskites promising candidates for radiation dosimetry, combining higher detection efficiency and light yield than plastic scintillators used as tissue-equivalent detectors with significantly faster scintillation than many conventional inorganic scintillators. Despite these advantages, their dosimetric performance has not been comprehensively evaluated. Key parameters—including dose linearity, reproducibility, dose-rate dependence, and energy response—remain insufficiently characterised, as most existing studies have primarily focused on optical properties, imaging performance, or direct-detection mechanisms rather than systematic dosimetry.

Beyond conventional diagnostic and therapeutic radiation beams, emerging high dose-rate radiotherapy techniques such as ultra-high dose rate (UHDR) “FLASH” electron therapy introduce additional challenges for accurate dosimetry. These challenges include strict requirements for dose-rate linearity and the need for detectors capable of fast scintillation responses without saturation. Recent studies using point scintillator detectors in UHDR electron beams have demonstrated that scintillation-based dosimetry can be successfully applied under FLASH conditions, highlighting the promise of scintillators for reliable dose measurements in this demanding regime (Vanreusel et al., 2022). In this context, the development and detailed characterization on the dose rate dependency of novel scintillator materials are particularly important to ensure accurate and dependable dosimetry across both conventional and ultra-high dose rate radiotherapy treatments.

To address the current lack of comprehensive dosimetric data for organic–inorganic hybrid perovskite scintillators, this study systematically evaluates the performance of (B A) 2 P b B r 4 and (P E A) 2 P b B r 4 under diagnostic X-ray irradiation. Their fundamental scintillation characteristics, including mass attenuation behaviour, radioluminescence emission, temporal stability, and dose–response performance, are examined in direct comparison with conventional plastic (EJ212) and inorganic (BGO, LYSO) scintillators. In addition, key dosimetric parameters such as reproducibility, dose-rate dependence, and energy dependence are assessed to determine the stability, accuracy, and overall reliability of these materials under clinically relevant low-energy X-ray (20– 80 k V p) conditions.

Scintillator samples

As reference scintillators, lutetium–yttrium oxyorthosilicate (LYSO) and bismuth germanate (BGO) were selected to represent conventional inorganic scintillators widely implemented in clinical imaging systems. These materials provide relatively high light yields—approximately 29,000 and 8,200 p h o t o n s M e V − 1 for LYSO and BGO, respectively. In contrast, the plastic scintillator EJ212 was employed as a representative tissue-equivalent material, exhibiting a lower light yield corresponding to approximately 11,000 p h o t o n s M e V − 1 (Eljen Technology, 2023). The two classes of scintillators exhibit markedly different radiation attenuation behaviours, primarily governed by their effective atomic numbers and mass densities.

In total, five scintillators were investigated: the hybrid organic–inorganic perovskites B A 2 P b B r 4 (10 × 10 × 0.3, m m 3) and P E A 2 P b B r 4 (approximately 7 × 5 × 0.5, m m 3), the inorganic crystals LYSO and BGO (each 10 × 10 × 2, m m 3), and the plastic scintillator EJ212 (10 × 10 × 1, m m 3). The corresponding optical and physical parameters of all samples are summarized in Table 1.

Synthesis and preparation of (B A) 2 P b B r 4 and (P E A) 2 P b B r 4 single crystals

(B A) 2 P b B r 4 single crystals were synthesized using a slow-cooling growth approach. In a fume hood, 79 μ L of n -butylamine (BA) and 0.0379 g of lead (II) acetate trihydrate (P b (A c) 2 ⋅ 3 H 2 O) were dissolved in 2.80 m L of hydrobromic acid (HBr) under cooling in an ice bath to moderate the initial reaction kinetics. The precursor solution was heated to 100 ° C until fully transparent, ensuring complete complexation, and then allowed to cool naturally to room temperature to initiate controlled nucleation. The uncovered vial was subsequently placed inside a sealed beaker containing at least 2 g of solid KOH to regulate the local vapour environment and promote gradual supersaturation. The solution was left undisturbed at room temperature, yielding millimeter-scale (B A) 2 P b B r 4 single crystals (approximately 1 × 1 c m 2 lateral size) after approximately 5 weeks, with well-defined facets indicative of high crystalline quality. For comparison, (P E A) 2 P b B r 4 single crystals were prepared using a controlled solvent-evaporation method at room temperature (Di et al., 2020). Stoichiometric amounts of phenethylammonium bromide (PEABr) and lead(II) bromide (P b B r 2) were dissolved in anhydrous N,N -dimethylformamide (DMF) under gentle stirring to obtain a clear precursor solution. After filtration to remove residual impurities, the solution was left undisturbed in a partially covered vial, allowing slow solvent evaporation and progressive supersaturation. Over approximately 10–15 days, millimeter-sized P E A 2 P b B r 4 crystals formed with well-defined facets (Zhang et al., 2019). The slow growth conditions effectively suppressed rapid nucleation, resulting in high crystalline quality consistent with previous reports (Ciavatti et al., 2024).

X-ray irradiation and dosimetric properties measurements

To prevent direct irradiation of the photomultiplier tube (PMT) by the primary X-ray beam, each scintillator was positioned within a custom-designed light guide prior to optical coupling. The assembled detector was then positioned 10 cm from the X-ray source, as shown in Figure 1. The light guide was 3D-printed from an ultra-clear polymer substrate and coated with three layers of reflective paint (EJ-510, Eljen Technology) to enhance optical collection efficiency. Each scintillator was optically coupled to a PMT (model 9256B) via a high-grade optical fibre, with the entire assembly enclosed in an aluminium housing to minimize the influence of scattered radiation and ambient light.

Due to the optical coupling configuration, approximately 90 % of the scintillation light was attenuated during transmission, requiring additional signal amplification. The PMT output current was amplified using a transimpedance amplifier (model DHPCA-100, FEMTO) and subsequently recorded with a digital oscilloscope. A gain value of 1 0 3 was used for the inorganic scintillators (BGO and LYSO), whereas a higher gain of 1 0 4 was applied to the perovskite and plastic scintillators to produce comparable signal amplitudes. Before each measurement, dark current and background noise were recorded and subtracted from the acquired data to ensure accurate quantification of the scintillation signals.

To evaluate the radiation response, all scintillators were irradiated using a Metrix SB-120-350-0.5 X-ray generator operated at tube voltages of 50–80 kVp and tube currents of 10– 100 μ A. The source-to-detector distance (SDD) was fixed at 10 cm. Dose rates were adjusted between 0.11 and 1.12 m G y s − 1 and calibrated using a reference parallel plate ionization chamber RC-1800. The stability of the X-ray output was continuously verified using the built-in kV monitor, with voltage fluctuations maintained within ± 0.2 k V throughout all measurements.

The dosimetric characterization of all scintillators was performed to evaluate their dose linearity, reproducibility, dose-rate dependence, and energy dependence under diagnostic X-ray irradiation. For the dose–linearity evaluation, each scintillator was irradiated with total doses ranging from 10 to 100 m G y at a fixed dose rate of 0.1 m G y s − 1. The photocurrent signal from each sample was recorded in real time and integrated to determine the total charge, which was analysed as a function of dose to assess linearity.

The reproducibility of the radiation response was assessed by performing ten repeated irradiations for each scintillator at a fixed dose of 30 m G y. The stability of the output signal was quantified using the relative standard deviation (RSD), defined as where σ and x ̄ denote the standard deviation and mean of the measured signal intensities, respectively.

Dose-rate and energy-dependence measurements were carried out under identical geometric configurations using standardized RQR2–RQR6 beam qualities with thin aluminium attenuation layers in accordance with IEC 61267 (Pernicka and McLean, 2007). For the dose–rate dependence study, each scintillator’s response was measured across dose rates between 0.11 and 1.12 m G y s − 1 and normalized to its response at 0.56 m G y s − 1, representing the mid-range reference condition. The energy dependence was examined across RQR2–RQR6 beam qualities corresponding to mean photon energies from 22.0 to 37.8 k e V, with each response normalized to the RQR6 beam. All measurements were corrected for photon flux variations using readings from a calibrated transmission ionization chamber to ensure accurate dose normalization and cross-comparison across different beam qualities.

Radioluminescence (RL) measurement and relative light yield

Radioluminescence (RL) spectra were acquired using a high-sensitivity UV–visible spectrometer (Maya 2000/Maya 2000-Pro, Ocean Insight). Each scintillator was continuously excited by 80 k V p X-ray radiation at 0.1 m A, and the emitted light was collected for 100 s. The recorded spectra were corrected for instrumental response and normalized to the scintillator surface area to enable direct comparison of emission intensities. All measurements were performed at room temperature with consistent geometric alignment for every sample to ensure reproducibility.

The relative light yields of B A 2 P b B r 4, and P E A 2 P b B r 4 were determined using a relative comparison method with LYSO (29,000 p h o t o n s M e V − 1) as the reference scintillator. The light yield of each sample was calculated according to the following relation: where L is the light yield, I is the integrated emission intensity, μ en is the mass energy-absorption coefficient, and A is the scintillator surface area.

Data processing and analysis

The data were acquired by measuring the voltage output from the oscilloscope and subsequently converted to photocurrent per unit area using the amplifier gain and the sample surface area. The resulting current signals were processed and analysed using OriginPro 2023. The total charge signal (Q) was obtained by integrating the current–time curve for each irradiation cycle. Dose–response linearity was evaluated by performing linear regression analysis between charge and the absorbed dose, yielding the coefficient of determination (R 2). A linear relationship with R 2 > 0.990 was considered indicative of excellent proportionality between dose and output signal.

Fundamental scintillation characteristics

This section presents the fundamental scintillation characteristics of the perovskite materials, including photon attenuation properties, radioluminescence emission, temporal stability, and dose–response behaviour, in comparison with conventional scintillators.

Figure 2a presents the mass attenuation coefficients of water, the plastic scintillator, BGO, B A 2 P b B r 4, and P E A 2 P b B r 4 across the photon energy range of 0.01 to 100 M e V. Both perovskite and inorganic scintillators exhibit significantly higher attenuation than the plastic scintillator, particularly in the low-energy region where photoelectric absorption dominates. The attenuation curves of B A 2 P b B r 4 and P E A 2 P b B r 4 closely follow that of BGO, reflecting their strong attenuation arising from the presence of high-atomic-number elements such as Pb and Br within the crystal lattice. Within the perovskite family, P E A 2 P b B r 4 showed a slightly lower mass attenuation coefficient, approximately 15 % lower across the 10– 50 k e V range compared with B A 2 P b B r 4, consistent with compositional variations and the influence of the organic spacer on the effective atomic number.

Figure 2b shows the radioluminescence (RL) emission spectra of the perovskite and reference scintillators under 80 k V p X-ray excitation (0.1 m A) at room temperature. Both B A 2 P b B r 4 and P E A 2 P b B r 4 exhibit narrow emission peaks centred at approximately 420– 430 n m, with intensities comparable to BGO. In contrast, BGO displays a broad emission band extending from roughly 400 to 800 n m, while LYSO:Ce shows its strongest emission within the 400– 600 n m range. The relative light yields, calculated using LYSO as the reference scintillator, reveal that B A 2 P b B r 4 produces an estimated ∼ 9,800 p h o t o n s M e V − 1, whereas P E A 2 P b B r 4 yields ∼ 4,800 p h o t o n s M e V − 1. Notably, B A 2 P b B r 4 surpasses the calculated light output of BGO (∼ 8,200 p h o t o n s M e V − 1), demonstrating efficient radiative recombination and strong X-ray-to-photon conversion efficiency.

Figure 2c illustrates the time-dependent photocurrent response of the scintillators under X-ray exposure. All samples exhibited stable and repeatable signal profiles throughout the measurement period, confirming excellent temporal stability. The integrated charge increased linearly with accumulated dose, and the corresponding dose–response curves (Figure 2d) yielded coefficients of determination (R 2) between 0.9994 and 0.9998, verifying highly linear behaviour across the tested dose range. Radiation sensitivity, defined as the slope of the dose–response curve and normalized to the BGO reference scintillator, showed distinct material-dependent differences. The normalized relative sensitivities represent the effective response of the complete scintillation detection system under identical experimental conditions, incorporating scintillation efficiency, X-ray absorption, PMT spectral response, optical coupling efficiency, and signal acquisition characteristics. LYSO exhibited the highest relative sensitivity (2.65 a. u.), followed by B A 2 P b B r 4 (0.48 a. u.) and P E A 2 P b B r 4 (0.20 a. u.). Despite their limited thicknesses of 0.3 m m and 0.5 m m, respectively, both perovskite samples outperformed EJ212, which showed a responsivity of only 0.06 a. u. Under the present experimental configuration, the corresponding absolute sensitivity of the BGO reference scintillator was 22.99 μ C c m − 2 m G y − 1.

It should be noted that, in addition to the intrinsic light yield of each scintillator, the measured radiation sensitivity is influenced by several experimental factors, including the spectral matching between the scintillation emission and the photomultiplier tube (PMT) quantum efficiency, as well as the X-ray detection efficiency of the material. In the present configuration, the detection efficiency at an average photon energy of 40 k e V was approximately 100 % for the inorganic and perovskite scintillators, whereas the plastic scintillator exhibited a significantly lower efficiency of below 5 %.

The dose reproducibility of the scintillators was evaluated at a fixed dose rate of 0.1 m G y s − 1, with each sample exposed to a cumulative dose of 30 m G y (Figure 3). To assess variability across repeated exposures, ten consecutive measurements were performed for each scintillator under identical conditions. Reproducibility was quantified using the relative standard deviation (RSD) calculated from the measured responses across all trials.

All scintillators demonstrated good reproducibility, with RSD values below 2 %. The commercial plastic scintillator EJ212 exhibited the largest response variation, attributable to its comparatively low light yield and the consequent reduction in signal-to-noise ratio. In contrast, the hybrid perovskite scintillators demonstrated superior stability, with relative standard deviations (RSDs) of 1.21 % for B A 2 P b B r 4 and 0.56 % for P E A 2 P b B r 4, indicating highly reproducible responses under repeated irradiations.

Figure 4 presents the dose rate response of all scintillators, normalized to their respective response at 0.56 m G y s − 1. Measurements were performed across a dose rate range of 0.11 to 1.12 m G y s − 1 at a fixed dose of 50 m G y, with the current density recorded in 50 s intervals over a total acquisition time of 1000 s.

Across this range, all scintillators exhibited generally stable responses, though the degree of variability differed among materials. The inorganic scintillator LYSO showed the most pronounced dose-rate dependence, with normalized values varying from 89.47 % to 105.54 % (± 10.53 %). In contrast, BGO demonstrated exceptional stability, with values constrained within 98.31 % to 101.03 % (± 1.69 %). The plastic scintillator EJ212 displayed intermediate behaviour, with responses between 98.18 % and 107.74 % (± 7.74 %). The perovskite scintillators showed consistently favourable dose-rate stability throughout the measurement range. B A 2 P b B r 4 exhibited a variation of ± 4.47 %, with normalized responses from 99.56 % to 104.47 %, while P E A 2 P b B r 4 varied by ± 5.50 %, with values between 98.55 % and 105.50 %. These results indicate that the perovskite materials maintain stable light output across varying dose rates, outperforming the plastic scintillator and approaching the level of stability observed in conventional inorganic scintillators.

Figure 5 presents the energy-dependent responses of the scintillators across standardized diagnostic beam qualities (RQR2 to RQR6), corresponding to mean photon energies from approximately 22.0 k e V to 37.8 k e V (Pernicka and McLean, 2007). Each scintillator’s response was normalized to its value at the RQR6 energy level, and the incident photon flux was corrected using an RC-1800 ionization chamber for dose normalization. According to the manufacturer data, the ionization chamber exhibits minimal energy dependence in the 20 to 80 k e V photon energy range (Radcal Corporation, 2025).

Overall, the relative response of the inorganic scintillators decreased with increasing photon energy, whereas the plastic scintillator showed the opposite trend, exhibiting a gradual increase across the same range. The decreasing behaviour observed in the inorganic materials reflects the characteristic dependence of photoelectric absorption, which dominates in this low-energy X-ray region. Because the photoelectric cross-section scales strongly with atomic number (approximately proportional to Z 3 to Z 4) and decreases rapidly with increasing photon energy (proportional to E − 3), low-energy photons are absorbed efficiently in high- Z materials such as BGO, LYSO, and the lead-halide perovskites. As photon energy increases, the probability of photoelectric absorption diminishes sharply, leading to reduced light yield per unit dose.

In contrast, the modest increase in the response of the plastic scintillator EJ212 originates from its low effective atomic number, resulting in weak photoelectric absorption and a nearly constant mass energy-absorption coefficient across this energy range. Consequently, as photon energy increases, a greater fraction of incident photons penetrates the material and deposits energy more uniformly, producing a slightly enhanced signal. As a result, EJ212 exhibited the lowest relative response across all tested energies (Hubbell and Seltzer, 2004).

The perovskite and inorganic scintillators showed markedly higher and more energy-dependent responses than the plastic scintillator. B A 2 P b B r 4 and P E A 2 P b B r 4 exhibited strong energy dependence, with relative response ranges of 1.18–2.76 and 1.15–2.39, respectively, compared with the narrower ranges of LYSO (1.17–1.62) and BGO (1.04–1.50). Within the perovskites, B A 2 P b B r 4 showed a stronger energy dependence than P E A 2 P b B r 4, consistent with its higher mass attenuation coefficient. These trends reflect the dominant influence of high- Z components in enhancing photoelectric absorption and sensitivity to lower-energy photons.

The markedly higher attenuation exhibited by the perovskite scintillators compared with the plastic scintillator is consistent with the presence of heavy elements within their metal–halide frameworks. Although these materials contain organic spacer cations, the X-ray interaction is primarily governed by the inorganic Pb and Br atoms in the lattice, resulting in attenuation characteristics that more closely resemble inorganic crystals than organic polymers. The difference in attenuation between B A 2 P b B r 4 and P E A 2 P b B r 4 can be attributed to the differing effective atomic contributions of the BA and PEA spacer cations, which influence the overall effective atomic number of the crystal. This indicates that variations in the organic spacer affect both the composition and the radiation–matter interaction behaviour. These results highlight how relatively small structural modifications in hybrid perovskites can modulate attenuation performance. They also demonstrate the possibility of tailoring the mass energy-absorption coefficient of the crystal for specific applications by selecting appropriate organic cation molecules. This tunability underscores the versatility and flexibility of hybrid perovskites as an adaptable scintillator family for radiation detection, and is particularly advantageous for radiation dosimetry, where achieving tissue-equivalent response is often required. Collectively, these findings demonstrate that two-dimensional hybrid perovskites effectively bridge the performance gap between organic and inorganic scintillators by combining strong photon absorption from their heavy-element sublattice with compositional tunability arising from their hybrid architecture.

The radioluminescence behaviour further supports the strong scintillation capability of these materials. The comparable RL intensities of the perovskite samples to BGO indicate highly efficient radiative recombination and effective conversion of absorbed X-ray energy into visible photons. Their sharp emission features near 420 to 430 n m suggest well-defined excitonic recombination pathways that differ from the broader, multi-component emission typically observed in traditional scintillators. The calculated light-yield values align well with earlier studies, including the work by Li et al. (2021), who reported a radioluminescence yield of approximately 7,000 p h o t o n s M e V − 1 for B A 2 P b B r 4 along with an ultrafast decay time of about 2.5 n s. Combined with the present measurements, these results reaffirm the high X-ray conversion efficiency of perovskite scintillators. Their strong light output, narrow emission features, and rapid decay dynamics underscore their potential as efficient alternatives to conventional inorganic scintillators for low-energy X-ray imaging and dosimetry.

The dose-response behaviour of the scintillators further demonstrates their suitability for quantitative dosimetry. All materials exhibited a strong linear relationship between dose and photocurrent, confirming predictable and proportional energy conversion under the tested X-ray conditions. The relatively high sensitivities of B A 2 P b B r 4 and P E A 2 P b B r 4, despite their modest thicknesses, emphasize the role of intrinsic optical efficiency and attenuation properties in determining overall responsivity. Their superior performance compared with EJ212 reflects the higher stopping power and greater light-yield capability of the perovskite structures. These results reinforce the promise of two-dimensional perovskites for diagnostic X-ray dosimetry, where linearity, sensitivity, and stability are essential. The excellent dose linearity observed for all scintillators (R 2 > 0.999) further supports their applicability in accurate and quantitative radiation measurement.

Although the present measurements demonstrate highly linear and measurable scintillation responses over the investigated dose range of 10– 100 m G y, a rigorous determination of the system-level limit of detection (LoD) was not performed in this study. Accurate LoD evaluation requires dedicated low-dose measurements below the present dose range, together with independent characterization of noise contributions originating from the scintillator, PMT, optical coupling, amplifier, and acquisition electronics. Therefore, the lowest experimentally verified dose in this work was 10 m G y, while detailed LoD determination under optimized low-noise conditions will be addressed in future studies.

Regarding reproducibility, the perovskite scintillators exhibited stable and consistent responses across repeated irradiations. Their low relative standard deviation values indicate reliable and reproducible scintillation behaviour under repeated X-ray exposure. Compared with the plastic scintillator, which showed higher variability, the perovskites demonstrated improved measurement stability. Combined with their strong light yield and efficient attenuation, this reproducibility further supports their viability as high-performance alternatives to conventional scintillator materials for applications requiring consistent and accurate dose monitoring.

The observed stability suggests that the scintillation response of the investigated perovskites remains well controlled under the present irradiation conditions. In scintillation-based detection, the emitted signal primarily originates from radiative excitonic recombination within the inorganic framework, and efficient recombination pathways are essential for maintaining stable light output and minimizing signal fluctuations. Furthermore, the experimentally observed low relative standard deviations (< 1.3 %), stable temporal signal profiles, and consistent dose-rate responses collectively suggest limited defect-induced fluctuations in the scintillation process under the investigated operating conditions. Nevertheless, further characterization of defect-assisted noise processes, afterglow behaviour, and long-term scintillation stability would provide deeper insight into the scintillation dynamics of these materials and remains an important direction for future investigation.

The perovskite scintillators also demonstrated favourable dose-rate dependence, maintaining a stable response across the full range of tested flux conditions. This behaviour may be associated with the fast excitonic recombination dynamics commonly reported in two-dimensional perovskite scintillators, which can help reduce signal saturation or intensity loss at higher excitation levels. In contrast, LYSO showed noticeably larger dose-rate variations, suggesting the presence of charge trapping or non-linear recombination processes under increased flux (Wei et al., 2021). These results indicate that the hybrid organic–inorganic structure of the perovskites may moderate recombination kinetics, enabling smoother and more consistent performance across varying dose rates.

The energy-dependent measurements revealed clear distinctions in how the scintillators respond across diagnostic X-ray beam qualities. As photon energy increased, the inorganic scintillators (BGO and LYSO) and the lead-halide perovskites (B A 2 P b B r 4 and P E A 2 P b B r 4) exhibited decreasing relative responses, consistent with the rapid decline in photoelectric absorption at higher energies. In contrast, the plastic scintillator EJ212 showed a slight increase due to its low effective atomic number and nearly energy-independent mass energy-absorption coefficient. The perovskites displayed the strongest spectral dependence, reflecting the influence of their high- Z constituents (Pb and Br), which provide substantial X-ray absorption similar to dense inorganic crystals. This behaviour is further reinforced by their efficient radiative exciton recombination: excitons generated within the confined two-dimensional inorganic recombine rapidly with minimal non-radiative loss, yielding sharp emission features and ultrafast decay components, As previously reported (Horimoto et al., 2020). When considered together with the photon attenuation characteristics, these findings highlight an important opportunity: the energy response of hybrid perovskite scintillators can be intentionally tailored through compositional design. Both the organic spacer molecules and the inorganic Pb–Br lattice contribute to the overall effective atomic number and mass energy-absorption coefficient of the material. Thus, modifying either component provides a direct and practical route for tuning the spectral sensitivity of the scintillator. This is supported by the present results, where B A 2 P b B r 4 exhibited a stronger energy dependence than P E A 2 P b B r 4, consistent with its higher mass energy-absorption coefficient. Such compositional tunability suggests that perovskites can be engineered to meet the performance requirements of specific diagnostic imaging or dosimetric applications.

Moreover, when these results are considered alongside the behaviour of the plastic scintillator, which exhibits the lowest and most tissue-equivalent response, an important design opportunity emerges for future detector development. A composite or hybrid detector design that integrates the over-response of high- Z perovskite materials with the under-response of plastic scintillators could effectively combine the strengths of both systems. In such a configuration, the high light yield and strong absorption efficiency of the perovskites would be balanced by the tissue-equivalent characteristics of the plastic scintillator. By offsetting the over-response of the perovskite with the under-response of the organic material, the detector could be tuned to achieve a functional “sweet point” where enhanced sensitivity and tissue equivalence are simultaneously realized. This concept suggests a promising pathway for engineering next-generation detectors that provide both high signal output and clinically accurate dose representation. Taken together, these insights highlight the broader versatility of hybrid halide perovskites, not only as high-performance stand-alone scintillators but also as adaptable, tunable components within composite systems optimized for precision X-ray detection and dosimetry.

Beyond the short-term dosimetric stability demonstrated in the present measurements, long-term operational durability under prolonged radiation exposure remains an important consideration for practical detector deployment. Extended evaluation under cumulative radiation exposure will therefore be valuable for assessing potential changes in scintillation efficiency, defect accumulation, and structural stability under realistic operating conditions. Encouragingly, recent studies on layered hybrid perovskites have reported favourable radiation tolerance and defect stability under ionizing radiation exposure (Lédée et al., 2022; Zhong et al., 2024), supporting their continued development for radiation detection applications. Future work will investigate long-term scintillation stability and cumulative dose effects in greater detail.

Taken together, organic–inorganic hybrid perovskites emerge from this work as a compelling and highly tunable class of scintillators for X-ray detection and dosimetry. Their exceptional attenuation, concentrated radioluminescence, and stable short-term dosimetric response combine to deliver performance that, in several key aspects, surpasses that of established inorganic materials. Although their energy dependence requires consideration in broadband applications, it simultaneously opens opportunities for engineered spectral responses tailored to specific diagnostic needs. With their inherent compositional flexibility offering routes to integrate high light yield with tissue-equivalent behaviour, hybrid perovskites stand poised to reshape the design of next-generation X-ray detectors and dosimetric tools.

This study evaluated the X-ray scintillation and dosimetric performance of two-dimensional hybrid perovskites, B A 2 P b B r 4 and P E A 2 P b B r 4, in comparison with plastic (EJ212) and inorganic (BGO, LYSO) scintillators. The perovskite materials demonstrated a compelling combination of properties, including strong X-ray attenuation, intense radioluminescence near 420 n m, excellent dose linearity (R 2 > 0.999), high sensitivity, and highly stable, reproducible responses (RSD < 1.3 %). These characteristics confirm their suitability for accurate and quantitative X-ray detection.

Their energy-dependent behaviour, governed by high- Z photoelectric absorption, produced enhanced sensitivity at lower photon energies. Although this spectral dependence may require calibration for broadband dosimetry, it also highlights a key advantage of hybrid perovskites: their energy response can be tuned through compositional modification. Differences observed between B A 2 P b B r 4 and P E A 2 P b B r 4 demonstrate that adjusting the organic spacer or inorganic lattice allows controlled manipulation of attenuation and spectral sensitivity, enabling scintillators engineered for specific diagnostic or dosimetric needs.

The results also point toward opportunities for composite detector architectures, where a tailored perovskite is combined with a tissue-equivalent plastic scintillator. Such an approach could balance the high light yield and strong absorption characteristics of perovskites with the energy-equivalent response of organic materials, achieving detectors that offer both enhanced sensitivity and clinically accurate dose representation.

Collectively, these findings position hybrid perovskites as highly promising and versatile scintillator materials with excellent radiation dosimetry performance. Moreover, their compositional tunability, strong scintillation efficiency, and stable dosimetric behaviour establish them as promising candidates for next-generation X-ray imaging and radiation measurement technologies.