High‐performance 110 kVp hard x‐ray detector based on all‐crystalline‐surface passivated perovskite single crystals

Abstract

Halide perovskite single crystals (SCs) have attracted much attention for their application in high‐performance x‐ray detectors owing to their desirable properties, including low defect density, high mobility–lifetime product (μτ), and long carrier diffusion length. However, suppressing the inherent defects in perovskites and overcoming the ion migration primarily caused by these defects remains a challenge. This study proposes a facile process for dipping Cs_0.05FA_0.9MA_0.05PbI₃ SCs synthesized by a solution‐based inverse temperature crystallization method into a 2‐phenylethylammonium iodide (PEAI) solution to reduce the number of defects, inhibit ion migration, and increase x‐ray sensitivity. Compared to conventional spin coating, this simple dipping process forms a two‐dimensional PEA₂PbI₄ layer on all SC surfaces without further treatment, effectively passivating all surfaces of the inherently defective SCs and minimizing ion migration. As a result, the PEAI‐treated perovskite SC‐based x‐ray detector achieves a record x‐ray sensitivity of 1.3 × 10⁵ μC Gy_air⁻¹ cm⁻² with a bias voltage of 30 V at realistic clinical dose rates of 1–5 mGy s⁻¹ (peak potential of 110 kVp), which is 6 times more sensitive than an untreated SC‐based detector and 3 orders of magnitude more sensitive than a commercial α‐Se‐based detector. Furthermore, the PEAI‐treated‐perovskite SC‐based x‐ray detector exhibits a low detection limit (73 nGy s⁻¹), improved x‐ray response, and clear x‐ray images by a scanning method, highlighting the effectiveness of the PEAI dipping approach for fabricating next‐generation x‐ray detectors.

INTRODUCTION

The demand for sensitive and highly stable x-ray detectors is steadily increasing owing to their diverse applications in medical diagnostics, homeland security, material characterization, space exploration, and nuclear power plant.^1–4 Several commercialized semiconductors, including silicon (Si),⁵ amorphous selenium (α-Se),⁶ and cadmium zinc telluride (CZT),⁷ had been extensively studied for their applicability in direct x-ray detection and imaging. However, commercially available Si and α-Se have relatively low attenuation coefficients due to low Z atoms, limiting their x-ray detection capabilities below 50 keV.⁸ Although single crystal (SC) CZTs exhibit excellent properties, the poor transport properties of holes, process complexity, and high processing cost due to high-temperature (800–1150°C) growth methods such as the Czochralski, Bridgman, and traveling heater methods still need to be addressed to ensure widespread application of these detectors.^9–11 Therefore, the development of an ideal semiconductor material that meets all the requirements for x-ray detection and imaging, such as high x-ray absorption, low operating voltage, non-toxicity, and low growth temperature, is still challenging.

Recent studies have actively explored organic–inorganic metal halide perovskite (OIMHP) SCs as promising candidates for high-performance next-generation x-ray detectors. OIMHP SCs possess several favorable characteristics, including high x-ray absorption, low-temperature solution processing, low manufacturing cost, easily tunable optical bandgap, low defect density, and excellent charge transporting properties such as high mobility-lifetime product (μτ) and long carrier diffusion length.^12–16 In particular, ultra-high sensitivities (≥10⁶ μC Gy_air⁻¹ cm⁻²), one of the most important properties for x-ray detection applications, have been achieved by engineering the SC growth for perovskite x-ray detectors through A-/X-site doping, ultrasound-assisted crystallization, and ligand-assisted growth.^16–18 However, all these studies employed lower dose rates than those typically used in actual medical diagnostics. Lower dose rates require longer exposure times to achieve the necessary overall dose for image contrast and sharpness in x-ray imaging. Unfortunately, longer exposure times can lead to blurred and degraded images due to patient movement, so a clinical dose rate on the order of mGy_air⁻¹, similar to the one used for standard chest x-rays (approximately 120 kVp, 160–320 mA, and 10–40 ms), is essential.^19,20 In addition, Basirico et al.²¹ reported that even when high sensitivity is achieved at low dose rates, it may not be possible to achieve high sensitivity at higher dose rates, especially at clinical dose rates. This suggests that the sensitivity measured at low dose rates may not be sufficient for application to clinical conditions. Accordingly, Zhang and Hua^22,23 investigated the x-ray response of Bridgman-grown CsPbBr_2.9Cl_0.1 and film-fed-grown CsPbBr₃ at higher dose rates and found the sensitivities to be 6.3 × 10⁴ and 4.6 × 10⁴ μC Gy_air⁻¹ cm⁻², respectively, with the former value being a record-high at high clinical dose rates. However, none of the two perovskite SCs are readily accessible, as they are grown by a melt-growth method that requires high temperatures. Therefore, to fabricate excellent perovskite SCs suitable for commercial x-ray detectors, it is essential not only to grow SCs with excellent properties using simple processes, but also to evaluate the grown SC-based detectors at clinical dose rates to achieve high sensitivity.

Surface defects, where various defects are concentrated, are a primary pathway for charge recombination. Therefore, effective surface defect passivation significantly improves the device performance of x-ray detectors by inducing compensation of dangling bonds and extending carrier lifetime. Enhancing device performance involves passivating interface defects and forming heterojunction structures to reduce trapping centers and improve charge collection efficiency in semiconductor devices such as solar cells and radiation detectors.^24–36 In particular, spin coated 2-phenylethylammonium iodide (PEAI) has been demonstrated to be a successful passivator that compensates for surface defects in perovskite SCs, suppresses ionic conductivity, and exhibits improved sensitivity despite low dose rates of x-rays.³⁴ However, an important issue that arises when passivating 3D SCs with PEAI solutions using spin coating techniques is that only one surface of the SC (i.e., the interface between perovskite and electrode) can be passivated, while other surfaces remain un-passivated and can act as carrier recombination pathways.

In this study, we have successfully grown triple-cation perovskite SCs of Cs_0.05FA_0.9MA_0.05PbI₃ (CFMPI) using the solution-based inverse temperature crystallization (ITC) method and passivated all SC surfaces by simple dipping the SCs into a PEAI solution. Unlike PEAI spin coating, dipping the entire SC in a PEAI solution is much more efficient in that all surfaces of the SC can be fully passivated. The immersion of SCs in the PEAI solution facilitated the formation of a 2D PEA₂PbI₄ layer on the SC surfaces without further treatment, which effectively reduced the number of inherent defects and suppressed ion migration. In addition, the treated CFMPI exhibited enhanced surface lifetime and linear photocurrent response to various x-ray doses. Utilizing this facile PEAI passivation approach, we achieved a record-high x-ray sensitivity of 1.3 × 10⁵ μC Gy_air⁻¹ cm⁻² at a clinical dose rate and a peak tube voltage of 110 kVp. Moreover, we successfully obtained distinguishable x-ray images of a custom-made phantom under bias voltages of 0 and 20 V, demonstrating the feasibility of a self-powered x-ray detectors. This effective passivation strategy, based on a simple method of dipping CFMPI SCs in a PEAI solution, has immense potential for synthesizing perovskite SC-based devices for next-generation x-ray detectors and imaging devices.

RESULTS AND DISCUSSION

CFMPI SCs were grown by incorporating both methylammonium (MA⁺) and cesium (Cs⁺) cations into the FAPbI₃ lattice. Compared to FA⁺ ions, the smaller MA⁺ and Cs⁺ ions are expected to release lattice stress,^16,37 effectively inhibiting ion migration. Detailed θ–2θ x-ray diffraction (XRD) was conducted to assess the single crystallinity of the CFMPI crystals and the phase changes following the dipping process as shown in Figure 1. In this study, we dipped for a variety of times (3 s, 5 s, 10 s, 20 s, 30 s, 1 m, and 2 m), but only showed cases that that represented the same result (pristine, 10 s, and 30 s). Figure 1A shows the SC XRD patterns of the CFMPI crystal dipped for 10 s (PEAI 10 s) and the crystal dipped for 30 s (PEAI 30 s), alongside the XRD patterns of the undipped CFMPI crystal (control) and PEAI powder for comparison. The inset in Figure 1A shows a photograph of the final CFMPI crystal. The control CFMPI crystals exhibited excellent single crystallinity with only a stable α phase and no change in the crystal structure after the passivation process. However, as the dipping time increased to 30 s, an extremely weak PbI₂ peak was observed at 12.93° (Figure 1B), which may be attributed to the partial dissolution of the perovskite SCs during dipping in the PEAI solution containing IPA. Yoo et al. reported that IPA effectively dissolves formamidinium iodide (FAI) because of its high polarity and ability to form hydrogen bonds.^38–41 They also observed a PbI₂-rich surface on IPA-treated perovskite films, which is consistent with our findings. However, in this study, the amount of the PbI₂ phase was likely minimal, given the extremely low peak intensity. Furthermore, Figure 1C represents enlarged pattern of Figure 1A, revealing a slight shift of the (110) peak toward a lower angle with respect to dipping process. This shift is attributed to the expansion of the lattice spacing caused by the bonding between PEAI and SC surfaces. This suggests the formation of an additional layer or chemical bonding occurring on the surface of the SC. However, up to a dipping time of 2 m at room temperature, no PEAI-related peaks were observed in the patterns obtained with conventional XRD, leading to the conclusion that the additional passivation layer is extremely thin and that its thickness did not increase proportionally with the dipping, but saturated.

[IMAGE OMITTED. SEE PDF]

Grazing incidence x-ray diffraction (GIXRD) was conducted to further examine the effect of dipping the CFMPI crystal surface into the PEAI/IPA solution. When the incident angle (ω) was set to 0.5°, the corresponding detection depth for FAPbI₃ was estimated to be ∼86 nm.⁴² The depth of x-ray penetration (detection) in GIXRD can be controlled by adjusting the incident angle of the x-rays and considering the absorption coefficient of the perovskite.⁴³ Figure 1D shows the GIXRD patterns for three different surfaces (control, PEAI 10 s, and PEAI 30 s) at a fixed incident angle (ω) of 0.5°. It should be noted that no single crystalline phase was detected in Figure 1D because ω was fixed at 0.5°. For the control sample (Figure 1D, bottom), only a weak but detectable peak was observed at 11.71°, indicating the presence of a polycrystalline yellow δ phase. Because the SC is primarily composed of the FAPbI₃ phase, it can be unstable even at low humidity and at surface depths of ~80 nm. Hence, the presence of this δ phase at surfaces is not surprising. When the SC sample was dipped for 10 s (Figure 1D, center), a new polycrystalline peak (shown as “*”) appeared at 5.37°, whose intensity continued to increase until 30 s. For PEAI 30 s, two types of XRD peaks were clearly identified. The new phase was identified to be a two-dimensional (2D) PEA₂PbI₄ phase. Interestingly, no distinct diffraction peak was observed at 4.7°, suggesting the absence of a PEAI layer. Periodic diffraction can be clearly observed at a PEAI dipping time of 30 s, which is a typical pattern for 2D PEA₂PbI₄. As discussed previously, IPA can partially dissolve the perovskite phase on the surface to form PbI₂. The PbI₂ phase exposed on the surface reacts with PEAI to form 2D PEA₂PbI₄ without further treatment. Importantly, we report the facile growth of a distinct 2D PEA₂PbI₄ overlayer on all surfaces of a perovskite SC without any further treatment. This overlayer serves as an all-surface passivating layer, which is expected to contribute to the enhancement of device performance and stability.

X-ray photoelectron spectroscopy (XPS) was further conducted to confirm the presence of the 2D PEA₂PbI₄ layer on the SC surface. Figure 2A shows the XPS spectra of various samples, including the PEAI-coated glass as a reference. The XPS spectrum of the untreated control SC contained CC, CN, and CO peaks at 284.6, 286.1, and 288.1 eV, respectively. Among these peaks, the CO peak can be attributed to the bonding of the perovskite to oxygen and moisture.³³ As shown in Figure 2A, the intensity of the CO peak in the PEAI-treated SCs was significantly reduced. In addition, the broad peak at approximately 293 eV can be ascribed to the π–π bonding of the phenyl group of the PEA⁺ cations.³⁴ Figure 2B also presents the XPS spectra of the I 3d core levels for the three SCs, indicating peaks of I 3d_3/2 and I 3d_5/2 at 618.76 and 630.29 eV, respectively. Notably, after PEAI passivation, a shift toward a higher binding energy was observed, resulting in an XPS peak shift of 0.3 eV. This observation suggests that alterations in the chemical environment can be attributed to the formation of a 2D PEA₂PbI₄ phase on the surface, which is consistent with the abovementioned XRD results. We hypothesize that the formation of such a phase facilitates the interaction between PEA⁺ and uncoordinated I (I⁻) originating from the [PbI₆]₄⁻ ions through coordinated covalent bonding.⁴⁴ Such phase formation allowed effective bonding of the PEA⁺ cations, compensating for possible iodine vacancies. Based on the XRD and XPS results, it can be concluded that a 2D PEA₂PbI₄ layer was successfully re-crystallized on the surface of the 3D perovskite SCs via a simple dipping process without any additional processing (Figure 2C).

[IMAGE OMITTED. SEE PDF]

To compare the effect of dipping time on the chemical composition of each CFMPI SC, the Pb 4f and I 3d core-level XPS spectra were used to estimate the Pb:I ratios of the three surfaces (Figure 2D−F). As shown in Figure 2D, the Pb:I ratio of the control perovskite SC was calculated to be 1:2.69, suggesting a significant iodine deficiency on the surface, which could result in considerable point defects. After dipping for 10 s (PEAI 10 s), the Pb:I ratio increased to 1:2.92, indicating that the iodine vacancies were likely filled by the formation of the 2D layer (Figure 2E). However, after dipping for 30 s (PEAI 30 s), the Pb:I ratio decreased again to 1:2.20. This observation was supported by the presence of a metallic Pb peak (Pb⁰) formed by the long immersion time of IPA,^38–41 as seen in the XP spectrum of PEAI 30 s (Figure 2F). The presence of metallic Pb can adversely affect device performance and stability⁴⁵; therefore, as shown in Figure 2C, dipping the SC in PEAI for 10 s is the optimal process, which can effectively passivate the surface of the SCs and form a moisture-resistant layer to increase its stability.

To analyze the effect of the PEAI solution dipping technique on carrier dynamics, photoluminescence (PL) spectroscopy was conducted. Figure 3A shows the steady-state PL spectra for the three samples. The PL spectrum of PEAI 10 s had a PL peak with the highest intensity, which was significantly stronger than that of the control SC. This enhancement in the PL intensity can be attributed to the passivation of surface trap states.⁴⁶ However, with increased dipping time to 30 s, a decrease in PL intensity was observed, which was attributed to the formation of metallic Pb, as previously discussed. The PL peak position remained unchanged in all the samples, indicating minimal change in the bandgaps. Furthermore, time-resolved photoluminescence (TRPL) analysis was performed to investigate the passivation effect induced by the PEAI solution (Figure 3B). TRPL study is employed to examine surface lifetimes under different conditions, aiming to comparatively assess the impact of PEAI passivation. The fitted parameters utilized in TRPL analysis are summarized in Table S1. The TRPL spectra of the PEAI-passivated SCs exhibit significantly longer PL lifetimes. Specifically, the PL lifetime of the control SC was calculated to be 61 ns, whereas the PL lifetimes of PEAI 10 s and PEAI 30 s increased to 208 and 130 ns, respectively. Fluorescence lifetime imaging microscopy (FLIM) images of the SCs with and without surface passivation are presented in Figure 3C−E showing the PL events and lifetimes over a large area. Typically, the fabrication procedure of radiation detectors includes mechanical processes such as lapping/slicing, which cause macro defects and deteriorate performance.^47,48 Such a defect and the related trap density in perovskites typically serve as nonradiative recombination centers for carriers, causing a notable decrease in the PL intensity and lifetime. PEAI 10 s exhibited the longest PL lifetime, highlighting its ability to passivate surface trap states. PL events and improved PL lifetime were observed along with the polishing scratches, meaning PEAI-based passivation effectively healed the mechanically caused defects. In the morphology of CFMPI SCs of Figure S1, it is similarly observed that the polishing scratches look filled with increasing immersion time, which supports previously mentioned defects-healing effect through passivation. However, the PEAI 30 s in Figure 3E exhibited a diminished lifetime, a reduction attributed to the formation of metallic Pb, as seen in steady state PL changes.

[IMAGE OMITTED. SEE PDF]

Figure 4A−C illustrates the results of the temperature-dependent ionic conductivity measurements of the control and PEAI-treated SCs to evaluate the effect of PEAI treatment on ion migration. The activation energies (E_as) of ion migration for control, PEAI 10 s, and PEAI 30 s SCs were extracted from the slope of the fitted line and found to be 0.38, 0.73, and 0.56 eV, respectively. The E_a values of the SCs increased significantly after PEAI passivation for 10 s (Figure 4A,B), indicating that ion migration was effectively inhibited. However, as the dipping time further increased, the E_a value decreased (as apparent for the PEAI 30 s sample), which is probably due to the degradation caused by the prolonged exposure to IPA (Figure 4C). Space charge limited current (SCLC) measurements were performed to evaluate the trap state densities (N_trap) of the SCs with and without PEAI passivation, as shown in Figure 4D−F. The calculated N_trap values for the control SCs, PEAI 10 s, and PEAI 30 s were 4.9 × 10¹⁰, 3.9 × 10⁹, and 1.8 × 10¹⁰ cm⁻³, respectively. Notably, the PEAI 10 s SCs exhibited the lowest trap density, which is in good agreement with the highest ion migration activation energy of this sample. Thus, the facile PEAI-dipping treatment is an efficient way to inhibit ion migration and passivate interface traps. It is worth noting that a longer dipping time may lead to the formation of metallic Pb due to the degradation of the perovskite SCs by IPA, which may adversely affect the ion activation energy and trap density, thus weakening the passivation effect.

[IMAGE OMITTED. SEE PDF]

We previously confirmed that the PEAI immersion treatment effectively mitigated the formation of dangling bonds on the surface of the CFMPI SCs because of the development of a 2D PEA₂PbI₄ interlayer and suppression of CO bonds. This caused the reduction of trap densities and extension of carrier lifetimes, enabling enhancement of x-ray response properties.^3,4 The improved SC material properties achieved through PEAI passivation can also enhance the performance of x-ray detectors. To determine the suitable dimensions for an x-ray detector, simulations studies were conducted using the physical properties of SCs; the results are shown in Figure 5A−C. Figure 5A shows the calculated mass attenuation coefficients (MACs) of the CFMPI SC in this study and the Cd_0.9Zn_0.1Te SC,⁸ which is widely used in room-temperature semiconductor detectors. Both SCs exhibited similar MACs within the diagnostic x-ray energy range of 10–110 keV, indicating that the CFMPI SC possessed sufficient x-ray absorption capability. Figure 5B shows the incident 110 kVp x-ray spectrum (black) used in this x-ray detection evaluation, and the corresponding absorbed x-ray spectrum (red) of a CFMPI SC with an arbitrary thickness of 2 mm. This figure indicates that most of the incident x-ray beam is absorbed in all x-ray energy ranges (10–110 keV), with 32 keV x-rays having the largest absorption counts, except for the characteristic x-rays, which are the most effective in terms of contrast and gray scale in the x-ray image. Therefore, the absorption rate of an incident 32 keV photon as a function of the thickness of the CFMPI SC is shown in Figure 5C. The figure demonstrates that more than 97% of the 32 keV x-rays are absorbed when the thickness of the CFMPI exceeds 1 mm. Consequently, based on this simulation calculation, the thickness of the CFMPI SC in the detector was set to 2 mm to ensure the maximum x-ray absorption.

[IMAGE OMITTED. SEE PDF]

We confirmed the impact of PEAI dipping passivation on the defect concentration of CFMPI SCs, causing a strong x-ray response of an x-ray detector with passivated crystals, as shown in Figure 5D–I. To evaluate the x-ray responses in this study, we used clinical dose rates ranging from 1 to 5 mGy s⁻¹ and peak tube voltage (110 kVp) commonly employed in practical applications.^19,20 Figure 5D shows the x-ray response of the CFMPI SCs at various dose rates. For the control detector (black), the x-ray photocurrent increased with the dose rate, however, the x-ray pulse shape and leakage current were unstable. In this case, the migration of ions through the defects and vacancies present on the perovskite surface causes the leakage current to fluctuate rapidly and degrades the x-ray response characteristics. However, for the detector based on the PEAI-treated CFMPI SC (red and blue), the leakage current was significantly stable during x-ray exposure at different dose rates. This stabilized x-ray response is attributed to the removal of stability-reducing C=O bonds and a reduction in the defect density on the surface. In particular, the PEAI 10 s-based detectors show high pulse response characteristics over the entire range. However, when the CFMPI SC was dipped in PEAI solution for 30 s (blue), the photocurrent density decreased, becoming lower than that of the control CFMPI (black). This reduction of photocurrent density was attributed to the formation of a thick 2D perovskite layer with a higher bandgap within the interface between the CFMPI bulk and electrode, which may inhibit charge carrier collection. The decrease in sensitivity is also likely due to the formation of metallic Pb inside the perovskite and further recombination of carriers.^49,50 In Figure 5E–G, the x-ray photocurrent density of the control, the PEAI 10 s- and PEAI 30 s-based detectors at different bias voltages and clinical dose rates are shown. While the control detector indicates the unstable response tendency under the different x-ray exposures, both the PEAI 10 s- and 30 s- detectors demonstrated a stable response to x-rays, with a gradual increase in the photocurrent as the bias voltage and dose rate increased. In Figure 5H, the signal-to-noise ratios (SNRs), particularly important for image quality, improved for both PEAI-treated detectors, but the SNR of the PEAI 30 s detector was lower than that of the PEAI 10 s detector. During the longer dipping period, the PEAI 30 s SC contained metallic lead (Pb) and a thicker 2D perovskite layer with a wider bandgap at the interface between the bulk and the electrode, which potentially induced recombination of charge carriers and hindered efficient collection of charge, respectively. As a result, the photocurrent density of the PEAI 30 s detector decreased and the SNR decreased compared to the PEAI 10 s detector.

Figure 5I shows the calculated sensitivities of all the detectors under various voltages at a clinical dose rate of 0.985 mGy s⁻¹. The control, PEAI 10 s-, and PEAI 30 s-based detectors exhibited sensitivities of 2.0 × 10⁴, 1.3 × 10⁵, and 7.4 × 10³ μC Gy_air⁻¹ cm⁻², respectively, when a 30 V external voltage was applied at an x-ray tube voltage peak of 110 kVp. The sensitivity of 1.3 × 10⁵ μC Gy_air⁻¹ cm⁻² achieved by the PEAI 10 s-based detector at clinical dose rates is the highest recorded value compared with previous studies (Figure S2).^22,51–60 The sensitivity in this study is approximately five orders of magnitude higher than that of commercially available α-Se based detectors (20 μC Gy_air⁻¹ cm⁻²).⁵¹ In addition, a thorough comparison with charge collection devices using perovskite crystals from literature studies is presented in Table S2, which includes details on the bulk material, passivation, device type, sensitivity, and presence of images obtained with the device.^{16,34,61–64} In addition, to demonstrate the effectiveness of all-surface passivation by dipping approach in a situation where the passivation effect at the perovskite-electrode interface is excluded, the devices with preformed electrodes was subjected to PEAI dipping passivation. As shown in Figure S3, all-surface passivation by dipping, regardless of the electrode type, increases the current response, proving that dipping scheme does indeed plays a significant role in the current response and showing that this dipping method is much more effective than passivation by spin coating.

The x-ray image acquisition by detector is critical to medical imaging and has been studied extensively.^58,65,66 To verify the x-ray imaging performance of the x-ray detectors based on PEAI 10 s-treated CFMPI SCs, an XY stage with a phantom was installed under the x-ray generator, as shown in Figure S4. Figure 6A describes the x-ray imaging acquisition of a customized KU phantom (background of Figure 6B and Figure S4G) consisting of a high atomic number part (Sn) and a low atomic number part (paraffin wax). In Figure 6B, each pattern displays the photocurrent measured at an applied voltage of 20 V, with an x-ray detector based on PEAI 10 s-treated CFMPI when the corresponding location on the KU phantom was scanned with a 110 kVp x-ray. The photocurrent decreases when x-rays scan the high atomic number part (the “KU” letter), and vice versa. Accordingly, the PEAI 10 s-treated CFMPI could successfully distinguish the photocurrent depending on the x-ray flux attenuated by the customized KU phantom. An excellent x-ray image was formed from the scanned photocurrent data, providing sufficient information for determining the original KU character, as shown in Figure 6C. We additionally utilized a low-dose x-ray generator to characterize the x-ray photoresponse, as shown in Figure 6D, to confirm its ability to operate at low dose rates and minimum detection limits (MDLs).⁶⁷ The measured current increased proportionally as the dose rate increased from 0.186 to 83 μGy s⁻¹. The calculated SNR and MDL are shown in Figure 6E. The extrapolated line of the SNRs reached 3 at an MDL of 72.11 nGy s⁻¹. Interestingly, the measured MDL was obtained at a biased voltage of 0 V, which implies that the detector can operate in a self-powered x-ray detection mode. Similarly, we have previously documented the realization of a self-powered perovskite detector utilizing the built-in potential arising from the continuous polarization effect.⁴⁸ Furthermore, we obtained zero bias x-ray images after optimizing various parameters such as the source-to-detector distance, dose rate, and scanning speed, as shown in Figure 6F, confirming that x-ray images can be obtained without an external bias. While the current x-ray images are preliminary for future applications of PEAI-treated CFMPI SCs in x-ray detectors incorporating commercial flat panel arrays, the strong x-ray response, and good imaging characteristics at clinical dose rates demonstrate the superiority of the PEAI passivation via simple immersion of perovskite SCs. In particular, the formation of a 2D PEA₂PbI₄ layer improved the defect passivation and ion migration resistance of the CFMPI SCs. The PEAI-treated CFMPI SC-based x-ray detector is expected to be utilized as a commercially attractive diagnostic medical detector because of its favorable characteristics, such as high sensitivity at clinical dose rates, good SNR/MDL, easy fabrication, self-powered operation, and distinguishable x-ray image formation without multiplexer circuit, which will ensure excellent image quality when data acquisition circuit is introduced.

[IMAGE OMITTED. SEE PDF]

CONCLUSION

This study introduces a novel defect passivation approach using a facile PEAI solution dipping method for triple cationic perovskite (CFMPI) SCs synthesized via a solution-based ITC technique. By applying this method, the x-ray response performance of the corresponding x-ray detectors was significantly improved. Interestingly, the facile dipping of the SCs in a PEAI solution resulted in the formation of a 2D PEA₂PbI₄ layer on all surfaces without further treatment, as confirmed by the XRD, GIXRD, and scanning electron microscopy results. Furthermore, the PL and SCLC results demonstrated that the defect density of the PEAI-passivated SCs was effectively reduced by more than one order of magnitude. The PEAI-treated SCs showed high activation energies of ion migration, indicating that PEAI passivation effectively inhibited the ion migration. To verify the applicability of the as-prepared SCs in practical systems, we evaluated the x-ray detection characteristics of the corresponding detectors at clinical dose rates and hard x-ray energies. The x-ray response and sensitivity evaluation results revealed that the SC-based detectors incorporating a 2D PEA₂PbI₄ passivation layer achieved a stable pulse response and high sensitivity. Notably, the obtained x-ray sensitivity of 1.3 × 10⁵ μC Gy_air⁻¹ cm⁻² is record high at clinical dose rates and over five orders of magnitude higher than that of commercially available α-Se detectors, allowing us to obtain good x-ray images without a multiplexer circuit. In conclusion, this study demonstrates the simplicity and effectiveness of the PEAI solution dipping method for passivating perovskite SCs, which is expected to play a significant role in the practical utilization of SCs in future x-ray detection applications.

EXPERIMENTAL SECTION

Chemicals and reagents

Methylammonium iodide (CH₃NH₃I, MAI, >99.99%), formamidinium iodide (CH(NH₂)₂I, FAI, >99.99%), and 2-phenylethylamine hydroiodide (C₆H₅(CH₂)₂NH₃I, >99%) were purchased from GreatCell Solar. Formic acid (CH₂O₂, >98% purity) was purchased from TCI. Gamma butyrolactone (C₄H₆O₂, GBL, 99%) was purchased from DAEJUNG Co., Ltd. CsI (99.999%), PbI₂ (99.999%), and 2-propanol ((CH₃)₂CHOH; IPA, 99.5%) were purchased from Sigma-Aldrich.

Synthesis of CFMPI SCs

The CFMPI SCs with the composition of Cs_0.05FA_0.9MA_0.05PbI₃ were synthesized using the ITC method.^16,68 Initially, a precursor solution was prepared by dissolving stoichiometric molar ratios of PbI₂, CsI, FAI, and MAI in a GBL solvent. The final concentration of the precursor solution was set to be 1 M. More specifically, CsI (0.1299 g; 0.5 mmol), FAI (1.5477 g; 9 mmol), MAI (0.0795 g; 0.5 mmol), and PbI₂ (4.6101 g; 10 mmol) were dissolved in GBL (10 mL). The dissolved precursor solution was then stirred at 60°C for 24 h. To ensure complete dispersion, formic acid (200 μL) was added dropwise to the solution and then stirred for 10 min. Subsequently, the solution was filtered into a beaker using a 0.2 μm pore size poly-(tetrafluoroethylene) (PTFE) filter. The filtered solution was heated to promote the growth of the CFMPI SCs. The temperature was gradually increased from 60 to 110°C with the rate of 5°C per hour, and then the temperature was maintained to maximize size of single crystal at the temperature with which the seed took place. The temperature was maintained at 60°C for 5 h and then gradually increased to 110°C for 5 h until the SC reached its maximum size. Finally, the SCs of suitable sizes were obtained through this process.

Surface treatment and detector fabrication

The grown CFMPI SCs were subjected to a lapping process using two different types of SiC paper. For PEAI dipping method, the PEAI powder was dissolved in IPA (concentration: 1 mg mL⁻¹). The SCs were dipped in the PEAI solution for varying durations (3 s–2 m) to passivate the defects on all surfaces. Finally, Au electrode deposition was performed using a thermal evaporation system, and Au wire and Ag paste were used to establish contact between the custom PCB board and each sample. The mentioned fabrication procedures are schematically shown in Figure S5.

Materials characterization

The crystal structures were determined via XRD (Rigaku Ultima) conducted using Cu Kα radiation at 40 kV and 40 mA. GIXRD (EMPYREAN, PANalytical B.V.) with Cu Kα radiation was performed at incident angle of 0.5° to characterize the crystal surface. Field emission scanning electron microscopy (FE-SEM; SU8010, Hitachi) was performed to examine the morphology of the synthesized SC surfaces. x-ray photoelectron spectroscopy (XPS) was conducted using a K-Alpha system (Thermo Fisher Scientific). Steady-state photoluminescence (SSPL) spectroscopy, time-resolved photoluminescence (TRPL) spectroscopy, and fluorescence lifetime imaging (FLIM) were performed using a measurement system (Korea Basic Science Institute) equipped with an inverted-type scanning confocal microscope (MicroTime 200, PicoQuant, Germany) and a 100× (oil-immersion) objective. The temperature-dependent ionic conductivity was measured at a low vacuum (~10⁻² torr) using Keithley 4200 SCS. The ionic conductivity was extracted via galvanostatic measurement with current bias (15 pA) and compliance voltage (10 V), excluding electronic conductivity. The thermally evaporated Au lateral electrodes (distance of 150 μm) on SCs were used for the measurements. The obtained current data under each condition were fitted to the Nernst–Einstein equation given below^69,70: 1 $\sigma \left(T\right)=\frac{{\sigma }_0}{T}\exp \left(\frac{-E_a}{k_{B}T}\right),$ where σ(T), σ₀, and k_B are the conductivity as a function of T, pre-exponential constant, and Boltzmann constant, respectively. The SCLC method was used to determine the trap density (n_t) of the SCs. The dark current–voltage curve and SCLC were measured using a Keithley 2636B source meter. For SCLC measurement, the voltage at the boundary between the ohmic and trap-filled regions is considered as the trap-filled limit voltage (V_TFL). The following equation associated with the SCLC method can be used to calculate the trap density: 2 $V_{\mathit{TFL}}=\frac{e{n}_t{d}^2}{2\epsilon {\epsilon }_0},$ where ɛ₀ represents the vacuum dielectric constant (8.854 × 10⁻¹² C² N⁻¹ m⁻²), ɛ is the relative dielectric constant, and d refers to the physical dimensions of the sample in the direction of the current flow.

X-ray response and sensitivity measurement

The x-ray responses were measured using a Keithley 2636B source meter. X-rays were produced using a GXR-S x-ray generator (DRGEM) at 110 kVp, 10–50 mA, and 2.5 s. The x-ray field size was 10 × 10 cm² and the distance between the x-ray source and detector was 100 cm. The pulse shape and response at different bias voltages were observed by setting the voltages to 5, 10, 20, and 30 V.

The sensitivity to x-ray response can be determined using Equation (3): 3 $S=\frac{I_{\mathrm{on}}-I_{\mathrm{off}}}{\text{Dose rate}},$ where I_on and I_off are the currents under irradiated and dark conditions, respectively. Therefore, I_on–I_off is the photocurrent generated upon irradiation. The dose rate was calculated from the measured values using an exposure dosimeter (Radical Corporation, model 20X6-60E). The initial measured value was 283.0 mR for 2.5 s exposure, which was converted to the exposure dose rate (mR s⁻¹) as follows: 4 $\text{Exposure dose rate}=\frac{283.0\mathrm{mR}}{2.5\mathrm{s}}=113.2\mathrm{mR}{\mathrm{s}}^{-1}$

The calculated value was multiplied by 0.0087 to convert the exposure dose rate to the absorbed dose rate.⁷¹ 5 $\begin{array}{l}\text{Absorbed dose rate}\\ =0.0087\frac{{\mathrm{Gy}}_{\mathrm{air}}}{\mathrm{R}}\times 113.2\mathrm{mR}{\mathrm{s}}^{-1}\\ =0.98484\mathrm{m}{\mathrm{Gy}}_{\mathrm{air}}{\mathrm{s}}^{-1}\end{array}$

The calculated (absorbed) dose rates were obtained using Equation (3) by utilizing the measured photocurrent (I_on–I_off) values. To identify the MDL, SNR values at low dose rates were taken by calculating the photocurrent tendency⁶⁷; the low dose rate x-rays were generated by the 150 kV Microfocus x-ray source L8121-03 (Hamamatsu photonics). Before experimenting with the x-ray response of the device, we performed a simulation to establish its accurate thickness. 6 $I=I_{\mathrm{o}}\exp \left(-\mathit{\mu \rho x}\right).$

Here, I and I_o are the numbers of the transmitted and initial photons, respectively. μ, ρ, and x are the critical parameters representing the mass attenuation coefficient, density, and thickness of materials, respectively. X-rays have a continuous spectrum; therefore, the attenuation of each photon depends on its energy. Assuming that the x-ray spectrum is divided into N energy bins: 7 $I_{\mathrm{o}}=\sum _{i=1}^N{I}_{\mathrm{o}i},I=\sum _{i=1}^N{I}_i$

Considering Equation (7) and the fact that μ is a function of energy (E_i), the number of transmitted photons (I_i) with material thickness x_j can be expressed as Equation (8). 8 $I_i=I_{\mathrm{o}i}\exp \left(-\mu \left(E_i\right)\rho x_j\right)$

From Equation (7), which represents the number of transmitted photons, the number of absorbed photons was calculated using Equation (9). 9 $\text{Absorbed photons}=I{\prime }_i=I_{\mathrm{o}i}\left[1-\exp \left(-\mu \left(E_i\right)\mathit{\rho x}\right)\right]$

The results in Figure 5B,C were simulated under the conditions of 0 < E_i < 110 keV (N = 220), as well as Figure 5B was simulated at x = 2 mm as an additional condition. The result in Figure 5C was simulated under the conditions of 0 ≤ x ≤ 2 mm and E = 32 keV. The values of μ and ρ were extracted from previous report⁷² and the input spectrum was obtained from the SRS-78 code.⁷³

X-ray imaging

X-ray imaging of SC x-ray defectors was carried out utilizing XY stage system (JMT, J Motion Tech) attached with customized KU phantom (Figure S4E–G). KU phantom was made up of Sn solder, paraffin wax, and plastic substrate. The current from SC was recorded with Keithley 2636B, when scanning KU phantom under 110 kVp x-rays exposure (GXR-S x-ray generator, DRGEM). The scanning positions of XY stage were matched with photocurrent obtained at corresponding points. The measured current data were reformed and assigned to each pixel of x-ray images by using Matlab. All the measurements were conducted at room temperature to exclude temperature-induced influence.

ACKNOWLEDGMENTS

This research was mainly supported by the Challengeable Future Defense Technology Research and Development Program through the Agency for Defense Development (ADD) funded by the Defense Acquisition Program Administration (DAPA) in 2022 (No. UI220006TD). This research was supported in part by the Challengeable Future Defense Technology Research and Development Program through the Agency for Defense Development (ADD) funded by the Defense Acquisition Program Administration (DAPA) in 2024 (No. 912765601). This work was supported in part by Human Resources Development Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Ministry of Trade, Industry and Energy, Republic of Korea (No. RS-2023-00237035).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

References

Sakdinawat A, Attwood D. Nanoscale x‐ray imaging. Nat Photonics. 2010;4(12):840‐848.

Kraushaar WL, Clark GW, Garmire GP, et al. High‐energy cosmic gamma‐ray observations from the OSO‐3 satellite. Astrophys J. 1972;177:341.

Wei H, Huang J. Halide lead perovskites for ionizing radiation detection. Nat Commun. 2019;10(1):1066.

He X, Deng Y, Ouyang D, et al. Recent development of halide perovskite materials and devices for ionizing radiation detection. Chem Rev. 2023;123(4):1207‐1261.

Granfors PR, Aufrichtig R. Performance of a 41×41‐cm amorphous silicon flat panel x‐ray detector for radiographic imaging applications. Med Phys. 2000;27(6):1324‐1331.

Sultana A, Wronski MM, Karim KS, Rowlands JA. Digital x‐ray imaging using avalanche a‐Se photoconductor. IEEE Sens J. 2009;10(2):347‐352.

Sordo SD, Abbene L, Caroli E, et al. Progress in the development of CdTe and CdZnTe semiconductor radiation detectors for astrophysical and medical applications. Sensors. 2009;9(5):3491‐3526.

X‐ray mass attenuation coefficients. https://physics.nist.gov/PhysRefData/XrayMassCoef/tab3.html

Derby JJ, Brown RA. On the dynamics of Czochralski crystal growth. J Cryst Growth. 1987;83(1):137‐151.

Park B, Kim Y, Seo J, et al. Bandgap engineering of Cd_{1− x}Zn_xTe_{1− y}Se_y (0< x< 0.27, 0< y< 0.026). Nucl Instrum Methods Phys Res A. 2022;1036: [eLocator: 166836].

Roy UN, Burger A, James RB. Growth of CdZnTe crystals by the traveling heater method. J Cryst Growth. 2013;379:57‐62.

Yakunin S, Sytnyk M, Kriegner D, et al. Detection of x‐ray photons by solution‐processed lead halide perovskites. Nat Photonics. 2015;9(7):444‐449.

Zhou Y, Chen J, Bakr OM, Mohammed OF. Metal halide perovskites for x‐ray imaging scintillators and detectors. ACS Energy Lett. 2021;6(2):739‐768.

Dong Q, Fang Y, Shao Y, et al. Electron‐hole diffusion lengths>175 μm in solution‐grown CH₃NH₃PbI₃ single crystals. Science. 2015;347(6225):967‐970.

Shi D, Adinolfi V, Comin R, et al. Low trap‐state density and long carrier diffusion in organolead trihalide perovskite single crystals. Science. 2015;347(6221):519‐522.

Liu Y, Zhang Y, Zhu X, et al. Triple‐cation and mixed‐halide perovskite single crystal for high‐performance x‐ray imaging. Adv Mater. 2021;33(8): [eLocator: 2006010].

Li WG, Wang XD, Huang YH, Kuang DB. Ultrasound‐assisted crystallization enables large‐area perovskite quasi‐monocrystalline film for high‐sensitive x‐ray detection and imaging. Adv Mater. 2023;35(31): [eLocator: 2210878].

Liu Y, Zheng Y, Fang Y, et al. Ligand assisted growth of perovskite single crystals with low defect density. Nat Commun. 2021;12(1):1686.

Chung M, Lee KY, Lee BY, et al. Diagnostic reference level of patient dose during a plain chest radiography examination. J Korean Radiol Soc. 2010;62(6):523‐528.

Bacher K, Smeets P, Bonnarens K, de Hauwere A, Verstraete K, Thierens H. Dose reduction in patients undergoing chest imaging: digital amorphous silicon flat‐panel detector radiography versus conventional film‐screen radiography and phosphor‐based computed radiography. AJR Am J Roentgenol. 2003;181(4):923‐929.

Basiricò L, Ciavatti A, Cramer T, Cosseddu P, Bonfiglio A, Fraboni B. Direct x‐ray photoconversion in flexible organic thin film devices operated below 1 V. Nat Commun. 2016;7(1):13063.

Zhang P, Hua Y, Xu Y, et al. Ultrasensitive and robust 120 keV hard x‐ray imaging detector based on mixed‐halide perovskite CsPbBr_3−nI_n single crystals. Adv Mater. 2022;34(12): [eLocator: 2106562].

Hua Y, Zhang G, Sun X, et al. Suppressed ion migration for high‐performance x‐ray detectors based on atmosphere‐controlled EFG‐grown CsPbBr₃ single crystals.

Abate A, Saliba M, Hollman DJ, et al. Supramolecular halogen bond passivation of organic–inorganic halide perovskite solar cells. Nano Lett. 2014;14(6):3247‐3254.

Zheng X, Chen B, Dai J, et al. Defect passivation in hybrid perovskite solar cells using quaternary ammonium halide anions and cations. Nat Energy. 2017;2(7):1‐9.

Zhao T, Chueh CC, Chen Q, Rajagopal A, Jen AKY. Defect passivation of organic–inorganic hybrid perovskites by diammonium iodide toward high‐performance photovoltaic devices. ACS Energy Lett. 2016;1(4):757‐763.

Abdi‐Jalebi M, Andaji‐Garmaroudi Z, Cacovich S, et al. Maximizing and stabilizing luminescence from halide perovskites with potassium passivation. Nature. 2018;555(7697):497‐501.

Wu Y, Feng J, Yang Z, Liu Y, Liu S(F). Halide perovskite: a promising candidate for next‐generation x‐ray detectors. Adv Sci. 2023;10(1): [eLocator: 2205536].

Park B, Kim Y, Seo J, Kim K. Effectiveness of parylene coating on CdZnTe surface after optimal passivation. Nucl Eng Technol. 2022;54(12):4693‐4697.

He Y, Pan W, Guo C, Zhang H, Wei H, Yang B. 3D/2D perovskite single crystals heterojunction for suppressed ions migration in hard x‐ray detection. Adv Funct Mater. 2021;31(49): [eLocator: 2104880].

Zhou Y, Zhao L, Ni Z, et al. Heterojunction structures for reduced noise in large‐area and sensitive perovskite x‐ray detectors. Sci Adv. 2021;7(36): [eLocator: eabg6716].

Yan J, Gao F, Tian Y, et al. Controllable perovskite single crystal heterojunction for stable self‐powered photo‐imaging and x‐ray detection. Adv Opt Mater. 2022;10(17): [eLocator: 2200449].

Jiang Q, Zhao Y, Zhang X, et al. Surface passivation of perovskite film for efficient solar cells. Nat Photonics. 2019;13(7):460‐466.

Han M, Xiao Y, Zhou C, et al. Suppression of ionic and electronic conductivity by multilayer heterojunctions passivation toward sensitive and stable perovskite x‐ray detectors. Adv Funct Mater. 2023;33(35): [eLocator: 2303376].

Xu Y, Li Y, Peng G, et al. Asymmetric metal halide film with suppressed leakage current for high sensitive x‐ray detection and imaging. IEEE Electron Device Lett. 2022;43(10):1709‐1712.

Li Y, Lei Y, Wang H, Jin Z. Two‐dimensional metal halides for x‐ray detection applications. Nano‐Micro Lett. 2023;15(1):128.

Basiricò L, Senanayak SP, Ciavatti A, Abdi‐Jalebi M, Fraboni B, Sirringhaus H. Detection of x‐rays by solution‐processed cesium‐containing mixed triple cation perovskite thin films. Adv Funct Mater. 2019;29(34): [eLocator: 1902346].

Yoo JJ, Wieghold S, Sponseller MC, et al. An interface stabilized perovskite solar cell with high stabilized efficiency and low voltage loss. Energ Environ Sci. 2019;12(7):2192‐2199.

Suo J, Yang B, Jeong J, et al. Interfacial engineering from material to solvent: a mechanistic understanding on stabilizing α‐formamidinium lead triiodide perovskite photovoltaics. Nano Energy. 2022;94: [eLocator: 106924].

Tan S, Huang T, Yavuz I, et al. Surface reconstruction of halide perovskites during post‐treatment. J Am Chem Soc. 2021;143(18):6781‐6786.

Wang D, Wright M, Elumalai NK, Uddin A. Stability of perovskite solar cells. Sol Energy Mater Sol Cells. 2016;147:255‐275.

Shan X, Wang S, Fu M, et al. Precisely preparing lead iodide passivation layer for enhancing the performance of triple cation perovskite solar cells using krypton fluoride excimer laser. J Power Sources. 2023;555: [eLocator: 232368].

Dimitrievska M, Fairbrother A, Gunder R, et al. Role of S and Se atoms on the microstructural properties of kesterite Cu₂ZnSn(S_xSe_{1− x})₄ thin film solar cells. Phys Chem Chem Phys. 2016;18(12):8692‐8700.

Naikaew A, Kumnorkaew P, Wattanathana W, et al. Investigation of double‐layered Pb‐Sn perovskite absorbers: formation, structure, band alignment, and stability. J Phys Chem C. 2022;126(3):1623‐1634.

Zhang H, Wang S, Hou Y, et al. Comparison of surface‐passivation ability of the BAI salt and its induced 2D perovskite for high‐performance inverted perovskite solar cells. RSC Adv. 2021;11(38):23249‐23258.

Chen L, Wang H, Zhang W, et al. Surface passivation of MAPbBr₃ perovskite single crystals to suppress ion migration and enhance photoelectronic performance. ACS Appl Mater Interfaces. 2022;14(8):10917‐10926.

Zheng Q, Dierre F, Crocco J, et al. Influence of surface preparation on CdZnTe nuclear radiation detectors. Appl Surf Sci. 2011;257(20):8742‐8746.

Park B, Ko J, Byun J, et al. Solution‐grown MAPbBr₃ single crystals for self‐powered detection of x‐rays with high energies above one megaelectron volt. Nanomaterials. 2023;13(15):2157.

Tumen‐Ulzii G, Qin C, Klotz D, et al. Detrimental effect of unreacted PbI₂ on the long‐term stability of perovskite solar cells. Adv Mater. 2020;32(16): [eLocator: 1905035].

Gujar TP, Unger T, Schönleber A, et al. The role of PbI₂ in CH₃NH₃PbI₃ perovskite stability, solar cell parameters and device degradation. Phys Chem Chem Phys. 2018;20(1):605‐614.

Kasap SO. X‐ray sensitivity of photoconductors: application to stabilized a‐Se. J Phys D. 2000;33(21):2853‐2865.

Zhu M, Du X, Niu G, et al. Template directed perovskite x‐ray detectors towards low ionic migration and low interpixel cross talking. Fund Res. 2022;2(1):108‐113.

Kim YC, Kim KH, Son DY, et al. Printable organometallic perovskite enables large‐area, low‐dose x‐ray imaging. Nature. 2017;550(7674):87‐91.

Shrestha S, Fischer R, Mat GJ, et al. High‐performance direct conversion x‐ray detectors based on sintered hybrid lead triiodide perovskite wafers. Nat Photonics. 2017;11(7):436‐440.

Ciavatti A, Sorrentino R, Basiricò L, et al. High‐sensitivity flexible x‐ray detectors based on printed perovskite inks. Adv Funct Mater. 2021;31(11): [eLocator: 2009072].

Pan W, Yang B, Niu G, et al. Hot‐pressed CsPbBr₃ quasi‐monocrystalline film for sensitive direct x‐ray detection. Adv Mater. 2019;31(44): [eLocator: 1904405].

Zhang Y, Liu Z, Xu H, et al. Nucleation‐controlled growth of superior lead‐free perovskite Cs₃Bi₂I₉ single‐crystals for high‐performance x‐ray detection. Nat Commun. 2020;11(1):2304.

Wu H, Chen X, Song Z, et al. Mechanochemical synthesis of high‐entropy perovskite toward highly sensitive and stable x‐ray flat panel detector. Adv Mater. 2023;35(29): [eLocator: 2301406].

He X, Xia M, Wu H, et al. Quasi‐2D perovskite thick film for x‐ray detection with low detection limit. Adv Funct Mater. 2022;32(7): [eLocator: 2109458].

Yang B, Pan W, Wu H, et al. Heteroepitaxial passivation of Cs₂AgBiBr₆ wafers with suppressed ionic migration for x‐ray imaging. Nat Commun. 2019;10(1):1989.

Wei H, Fang Y, Mulligan P, et al. Sensitive x‐ray detectors made of methylammonium lead tribromide perovskite single crystals. Nat Photonics. 2015;10(5):333‐339.

Kim T, Jeong S, Kim K‐H, Shim H, Kim D, Kim HJ. Engineered surface halide defects by two‐dimensional perovskite passivation for deformable intelligent photodetectors. ACS Appl Mater Interfaces. 2022;14(22):26004‐26013.

Mahapatra A, Parikh N, Kumari H, et al. Reducing ion migration in methylammonium lead tri‐bromide single crystal via lead sulfate passivation. J Appl Phys. 2020;127(18): [eLocator: 185501].

Wang K, Chen F, Yao Q, et al. Synergistic enhancement of the optoelectronic performance and stability of MA and Cs in FA_xMA_yCs_1−x−yPbI_zBr_3−z single crystals. J Mater Chem C. 2023;11(38):12959‐12967.

Pang J, Zhao S, Du X, Wu H, Niu G, Tang J. Vertical matrix perovskite x‐ray detector for effective multi‐energy discrimination. Light Sci Appl. 2022;11(1):105.

Song Z, Du X, He X, et al. Rheological engineering of perovskite suspension toward high‐resolution x‐ray flat‐panel detector. Nat Commun. 2023;14(1):6865.

Pan L, Shrestha S, Taylor N, Nie W, Cao LR. Determination of x‐ray detection limit and applications in perovskite x‐ray detectors. Nat Commun. 2021;12(1):5258.

Chen L, Tan YY, Chen ZX, et al. Toward long‐term stability: single‐crystal alloys of cesium‐containing mixed cation and mixed halide perovskite. J Am Chem Soc. 2019;141(4):1665‐1671.

Lee JW, Kim SG, Yang JM, Yang Y, Park NG. Verification and mitigation of ion migration in perovskite solar cells. APL Mater. 2019;7(4): [eLocator: 041111].

Zhao Y, Yavuz I, Wang M, et al. Suppressing ion migration in metal halide perovskite via interstitial doping with a trace amount of multivalent cations. Nat Mater. 2022;21(12):1396‐1402.

Khan FM, Gibbons JP. Khan's the Physics of Radiation Therapy. Lippincott Williams & Wilkins; 2014.

Ma L, Yang Z, Zhou X, et al. A polymer controlled nucleation route towards the generalized growth of organic‐inorganic perovskite single crystals. Nat Commun. 2021;12(1):2023.

Meyer P, Meyer E, Mertz L, et al Evaluation of the use of six diagnostic X‐ray spectra computer codes. Br J Radiol. 2004;77(915):224‐230.