Vol.:(0123456789) MRS Energy & Sustainability (2025) 12:233–253 doi:10.1557/s43581-025-00142-5 MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    233 © The Author(s), 2025 Progress and opportunities in bismuth‑based materials for X‑ray detection REVIEW Joydip Ghosh and Robert L. Z. Hoye , Inorganic Chemistry Laboratory, University of Oxford, Oxford OX1 3QR, UK �Priyanka Priyadarshini and Judith L. MacManus‑Driscoll, Department of Materials Science and Metallurgy, University of Cambridge, Cambridge, UK �Zubaida T. Younus and Quanxi Jia, Department of Materials Design and Innovation, University at Buffalo - The State University of New York, Buffalo, NY 14260, USA �Wanyi Nie, Department of Physics, University at Buffalo - The State University of New York, Buffalo, NY 14260, USA �Address all correspondence to Robert L. Z. Hoye at robert.hoye@chem.ox.ac.uk (Received: 14 April 2025; accepted: 11 July 2025; published online: 25 August 2025) ABSTRACT Lead-free bismuth-based perovskites and derivatives are promising eco-friendly materials for sensitive X-ray detection, crucial for medi- cal imaging and security inspection applications. This review highlights their key properties, recent developments, strategies to enhance performance, as well as commercialization challenges. Over the past decade, lead halide perovskites have gained significant interest for ionizing radiation detection, owing to their exceptional perfor- mance, and cost-effective fabrication in a wide range of form factors, from thick films to large single crystals. However, the toxicity of lead, limited environmental and thermal stability of these materials, as well as dark current drift due to ionic conductivity, have prompted the development of alternative materials that can address these challenges. Bismuth-based compounds (including perovskite derivatives and nonperovskite materials) have similarly high atomic numbers, leading to strong X-ray attenuation, but have lower toxicity, tend to be more environmentally stable, and can have lower ionic conductivity, especially in low-dimensional materials. These materials are also advantageous over commercial direct X-ray detectors by being able to detect lower dose rates of X-rays than amorphous selenium by at least two orders of magnitude, are potentially more cost-effective to mass produce than cadmium zinc telluride, and can operate at room temperature (unlike high-purity Ge). Given the strong interest in this area, we here discuss recent advances in the development of bismuth-based perovskite derivatives (with 3D, 2D and 0D structural dimensionality), and other bismuth-based perovskite-inspired materials for direct X-ray detection. We discuss the critical properties of these materials that underpin the strong performances achieved, particularly the ability to detect low-dose rates of X-rays. We cover key strategies for enhancing the performance of these materials, as well as the challenges that need to be overcome to commercialize these emerging technologies. Keywords  Bi · perovskites · electronic material · devices · defects · electron–phonon interactions DISCUSSION • Material upscaling is an ongoing effort for Bi-based materials to enable large-area radiation imaging. • Approaches to control charge-carrier transport and localization are not well understood in the Bi-based material systems, and it will be important to develop materials design principles to improve charge-carrier transport. • The chemistry and physics of defects in Bi-based semiconductors are not well understood, and addressing defect-mediated non-radiative recombination would greatly improve performance. http://orcid.org/0000-0002-7675-0065 234         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal Introduction X-rays are high-energy photons in the high-frequency part of the electromagnetic spectrum, and are capable of penetrat- ing through solids and soft tissue. As a result, X-ray detection is widely used for non-destructive measurements across a variety of disciplines, including medical imaging, security screening, and in-field inspection.1,2 The popular approaches for X-ray detection are based on scintillators and solid-state devices for indirect and direct conversion of the X-ray photon signal to an electrical signal, respectively. A scintillator is comprised luminescent materi- als that can downconvert high-energy X-ray photons to lower- energy visible photons. It is typically coupled to a visible light detector, such as an avalanche photodiode or a photon multi- plication tube. This is also called an indirect detector because of the multiple conversion processes involved (X-rays to vis- ible photons, visible photons to an electrical signal). Direct detectors, on the other hand, utilize semiconductors to con- vert X-ray photons to electrical signals. Semiconductors com- monly used commercially for direct conversion include silicon, germanium, GaAs, Cd-Zn-Te and amorphous selenium (α-Se).3 Directly converting X-ray photons to an electrical signal can lead to higher conversion efficiencies, especially in the low- dose range, as well as higher spatial resolution compared to scintillators.4 The types of detectors employed depend on the specific end uses. For instance, for experiments requiring fast time reso- lution, scintillators are typically chosen for their rapid radio- luminescence decay on the sub-nanosecond timescale.5 The response time of the direct conversion detector is determined by the charge extraction time, which is typically longer than 10 s of nanoseconds. In the case of medical imaging, where the main consideration is minimizing the patient’s X-ray exposure dose, it is more important to have a highly sensitive detector with a strong signal-to-noise ratio (SNR) in a low flux regime. In this scenario, a direct detection mechanism with a high detection quantum efficiency is preferred. In imaging applications, spatial resolution is another important figure- of-merit to consider when choosing detector materials. Here, direct detectors are advantageous, in that the pixels used can be made as small as 10 s of microns, enabling high spatial reso- lution to be achieved. For radiology, a resolution of 5.7 lp mm–1 is required, while mammography demands a higher resolution, with a minimum requirement of 10 lp mm–1.6,7 However, there are currently few direct detector materials in the market right now. One limitation is the high cost of building digital panels using classical semiconductors, such as cadmium zinc telluride (CZT). Recently, remarkable developments have been made in low-cost lead halide perovskites (LHPs) for direct radiation detection thanks to the ability to achieve high mobility-life- time products using simple solution-based synthesis meth- ods, along with high X-ray attenuation coefficients, large bulk resistivities, and high radiation hardness.7–11 Solution-grown LHP single crystals (SCs) have demonstrated high sensitivities, and the capability of detecting remarkably low-dose rates of X-rays.11 Organic–inorganic hybrid perovskite direct X-ray detectors have achieved a record X-ray sensitivity of ~ 2.2 × 108 µC Gyair −1 cm−2, along with a low detection limit of 0.62 nGyair s−1.12–14 By contrast, industry-standard amorphous selenium direct detectors have sensitivities of 20 μC Gyair −1  cm−2 and the lowest detectable dose rate of 5500 nGyair s−1.15 However, the commercial application of LHP-based devices is mainly restricted by poor ambient stability. LHP devices degrade in the presence of oxygen, moisture, high temperatures, and extreme light.16 Another shortcoming of LHP radiation detectors is the toxicity of lead, which is bioaccumulative, and can readily be released from LHPs owing to their water solubility. Toxic sol- vents (e.g., N,N-dimethylformamide or chlorobenzene) are often used in the synthesis of LHPs, and the ambient stability of these materials is limited, such that their commercial use depends on effective encapsulation. The exceptional perfor- mance of LHPs, as well as their limitations, have prompted a broad search for more sustainable alternatives, and Bi-based materials have emerged as a prominent class of materials. In this review, we discuss the status, challenges and future opportunities of using Bi-based perovskite-inspired materials for direct X-ray detection. We begin with the key requirements of X-ray detectors, the properties of Bi-based materials that allow them to fulfill these properties, and the progress in applying these materials in devices. We conclude with a discussion of the market opportunities for these materials, and the challenges in commercializing these devices. Key properties of X‑ray detector materials Achieving high-resolution X-ray imaging with minimal radia- tion exposure to the subject (e.g., medical patient) requires an in-depth understanding of the interplay between materials prop- erties and device performance. For instance, the X-ray attenua- tion coefficient, ionization energy, and mobility-lifetime product are closely related to the chemistry, microstructure, and defect structure of the materials.17 These also affect important device properties, such as dark current, charge collection efficiency, sensitivity, detection limit, and response time.18,19 In the follow- ing, we will discuss some key properties and figures of merit for radiation detector materials. Radiation attenuation The attenuation of radiation in materials typically occurs through Rayleigh scattering, photoelectric effects, Compton scattering, and pair production. X-ray attenuation is quantified by the Beer-Lambert law: where I is the attenuated radiation intensity at a given depth of x from the surface exposed to X-rays, I0 is the incident radia- tion intensity at x = 0, and μMAC and ρMD are the mass attenuation coefficient and the mass density of the material, respectively. (1)I = I0 exp {−µMACρMDx} MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    235 The value of μMAC is proportional to Z4, where Z is the average atomic number of the material. Therefore, effective radiation detection requires materials with high atomic numbers (Z > 40) and mass density to ensure that sufficient X-ray attenuation can be achieved without requiring excessively large thicknesses that are challenging for charge-carrier extraction. For instance, it has been reported that the effective Z (Zeff) value is 73.6 for BiOI. Such a material could lead to a stopping power nearly double that of CZT (Zeff = 48–52) at energies > 100 keV, and is also sub- stantially higher than amorphous Se (Zeff = 34).2,20 Charge collection efficiency The charge collection efficiency (CCE) is the ratio of the total charge collected to the charge generated within a material when the device is exposed to radiation. CCE reflects how effectively a detector converts incoming radiation into a measurable electri- cal signal. As depicted by Eq. (2), the mobility-lifetime product ( µτ ) of charge carriers plays an important role in determining the CCE at a given generation rate of electron–hole carriers, where µ and τ are the carrier mobility and lifetime, respectively. A large µτ product will ensure that generated charge carriers can be collected before they are lost due to non-recombination. Sensitivity Sensitivity (S) is defined by the charge accumulated per unit area when the device is exposed to radiation. This performance metric ref lects a material’s efficiency in converting irradi- ated photons into electrical signals. Mathematically, S can be expressed as where D is the irradiation dose rate, A is the active area of the detector, and Iradiation and Idark are the currents from the device with and without irradiation, respectively. To enhance S for a given device design, it is critical to reduce Idark, as well as to increase CCE to increase the photocurrent signal. Reductions in Idark could be achieved by increasing the bandgap of the mate- rial (thus lowering the concentration of thermally generated charge-carriers), lowering the doping level (e.g., by reducing the concentration of donor/acceptor defects), or by lowering the electronic dimensionality of the material. For example, many efforts have been made to synthesize single crystals (SCs) or large-grained polycrystalline materials in order to improve µτ products (by reducing grain boundary scattering of charge- carriers, as well as reducing non-radiative recombination), and lower dark currents due to point defects.21 It is worth noting that increases in the measured sensitivity can also be achieved from photoconductive gain, which could be detrimental if it also increases the noise current, and increases the detection limit.22 (2)CCE = µτV L2 [ 1− exp ( − L2 µτV )] (3)S = Iradiation − Idark DA Signal‑to‑noise ratio The SNR is a measure of the device’s ability to produce a detectable signal. Using current as the detected signal, the SNR can be expressed as23,24 where ISC and INC are the signal current and noise current, respectively, where ISC is taken as the photocurrent (Iradiation— Idark). The INC can, however, come from different sources. For example, the Johnson noise current and the current resulting from the limited shunt resistance of the device can all contribute to INC. Limit of detection In addition to sensitivity, another key metric for X-ray detec- tors is the lowest detectable dose rate (LoDD), where having a low LoDD is critical for minimizing harm to the patient dur- ing medical imaging.24,25 The LoDD for radiation detectors is defined by IUPAC as the radiation dose rate at which the SNR of the detector is 3. The SNR is typically calculated as the ratio of the average photocurrent signal to the standard deviation of the photocurrent, while the LoDD is determined by measuring the SNR across various dose rates. Key properties of bismuth‑based materials for X‑ray detection Bismuth-based materials have emerged as a promising class of compounds for X-ray detection because of their low toxicity, strong attenuation of X-rays, and, in many cases, high environ- mental and thermal stability.26 Bismuth is the heaviest (Z = 83) element that is not radioactive, and the high effective atomic number of Bi-based compounds is critical for their high mass attenuation coefficients for ionizing radiation (Sect. “Radiation attenuation”). Some Bi-based materials have also exhibited high mobility-lifetime products (> 10–4 cm2  V–1), along with band- gaps that can be tuned to the optimal range for X-ray detectors (1.4–2.5 eV).7,27 Prominent classes of Bi-based materials are hal- ide elpasolites (A2MIMIIX6, e.g., Cs2AgBiBr6), vacancy-ordered triple perovskites (A3Bi2X9; (NH4)3Bi2I9, MA3Bi2I9, Cs3Bi2I9), and other binary or mixed-anion bismuth-based compounds (BiOI, BiI3, Bi2O3). Bismuth‑based double perovskites Lead-free elpasolites (also known chemically as double perovskites), with the general formula A2MIMIIIX6 (A = mono- valent cation, MI = monovalent cation, MIII = Bi3+, X = halide anion), have gained attention as alternatives to LHPs that maintain the perovskite crystal structure, but without toxic (4)SNR = ISC INC 236         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal elements that are restricted for use in electronics. Two Pb2+ cations are substituted by a combination of one monovalent cation (MI) and one trivalent cation (MIII), thus maintain- ing the same overall charge as two Pb2+ cations would have (presented in Figs.  1, 2, 3). Among bismuth-based double perovskites, the Bi3+-Ag+ system is one of the most studied set of materials,11 particularly Cs2AgBiBr6. This material has a cubic structure (space group Fm3m ) with corner sharing metal-halide octahedra, and lattice parameter a = 11.27 Å at room temperature. There are several appealing features that are conducive toward high-performance in X-ray detectors. Firstly, Cs2AgBiBr6 has a Zeff value of 53.1, which exceeds those of MAPbBr3 (Zeff = 45.1) and α-Se (Zeff = 34), thus allow- ing higher X-ray attenuation coefficients (Fig. 4a). Secondly, single crystals of this material have high resistivity (10⁹–1011 Ω cm), which exceeds the resistivity of methylammonium lead halide single crystals (10⁷–10⁸ Ω cm). This results in reduced dark and noise current in devices. Furthermore, Cs2AgBiBr6 has lower field-driven ion migration, which is important for maintaining stable performance with the application of an electric field.28,29 Other bismuth-based elpasolites that have been investigated include two-dimen- sional (2D) (BA)2CsAgBiBr7 (BA = n-butylammonium),30 (DFPIP)4A gBiI8 (DFPIP = 4,4-dif luoropiperidinium),31 (CPA)4A gBiBr8 (CPA = chloropropylammonium),32 and (PA)4AgBiBr8 (PA = propylammonium).33 These materials are advantageous because of improved stability compared to their inorganic counterparts due to the hydrophobic nature of the long-chain organic A-site cations. Meanwhile, there are several properties that limit the per- formance of Cs2AgBiBr6. In particular, electron–phonon cou- pling plays an important role. The coupling between charge- carriers and longitudinal optical (LO) phonons (known as Fröhlich coupling) reduces mobilities, while pronounced charge-carrier coupling with acoustic phonons results in small polaron formation. By forming small polarons, charge-carrier mobilities are substantially reduced, limiting CCEs.48–50 This carrier localization process arises due to the high acoustic deformation potentials in Cs2AgBiBr6, and is intrinsic to the material itself. An important question is whether the chemis- try of these materials could be changed to lower this acoustic deformation potential.49 Various strategies, including dop- ing/alloying, external energy treatments such as thermal annealing, laser/photonic (UV/X-ray/IR exposure) irradia- tion, plasma and pressure treatments, the use of  heterojunc- tion structures, and bond length compression (using chemi- cal/mechanical pressure) have been attempted to reduce the strength of electron–phonon coupling to increase detector performance.28,29,51 For example, Steele and Roeffaers found that annealing Cs₂AgBiBr₆ at 160°C for 1 h led to a reduction of over 10% in the strength of charge-carrier coupling with longitudinal optical phonons (ϒₗₒ), decreasing from 226 to 201 meV.52 Similarly, applying pressure (up to 31 GPa) at room temperature in Cs₂AgBiCl₆ resulted in a blue shift of the broad PL emission along with a red shift of the absorption edge. This behavior was attributed to the decreased lattice relaxation energy caused by lattice compression. This significantly reduced Fröhlich coupling, as indicated by the Huang-Rhys parameter, S, which is estimated from Stokes shift energy and the LO phonon mode energy (EStokes = 2S�ωLO) . By increas- ing the pressure from atmospheric to 4.5 GPa, the S value reduced from 18.1 to 8.5. This lattice compression effectively suppressed ionic activity under high pressure, all while pre- serving the highly symmetric cubic structure.53 There has been less work on understanding how the coupling to acoustic phon- ons could be reduced, but a recent investigation into CuSbSe2 suggests that having regular free volume in the structure (by having a layered material), enables reduced acoustic deforma- tion potentials, enabling band-like transport.54 Bismuth‑based perovskite derivatives Vacancy-ordered triple perovskites (VTPs) have also been widely explored, and have the general formula A3Bi2X9, where A refers to the monovalent cation and X is a halogen. These are described as VTPs because the structure can be thought of as derived from a perovskite, where there is a vacancy in the cation site for every three formula units. VTPs typically adopt either a 0D (isolated dimers of Bi2X9 3− surrounded by the A-site cation) or 2D structure (which has corner sharing BiX6 3− octahedra). As with halide elpasolites, VTPs have been synthesized as SCs, 2D flakes ((NH4)3Bi2I9, Rb3Bi2I9), nanocrystals (NCs) (Cs3Bi2Cl9), and quantum dots (MA3Bi2I9), and these have been developed into X-ray detectors.55–57 2D-VTPs are obtained by removing every third layer of Bi3+ along the (111) direction in the perovskite structure to maintain charge balance (presented in Fig. 1). This layered structure leads to anisotropic electronic properties and X-ray detector perfor- mance. Similarly, like 2D VTPs, 0D-A3Bi2X9 VTPs tend to have high effective masses due to the low electronic dimensionality. While this reduces the μT product, it can be beneficial for X-ray detectors by enabling lower dark currents and noise. These 2D/0D VTPs have resistivities ~ 1012 Ω cm, which is 2–3 orders of magnitude larger than 3D LHPs, and are comparable to the resistivities of commercial inorganic semiconductors (e.g., CZT: 109–1011 Ω cm, CdTe: 109 Ω cm), and close to α-Se (1014–1015 Ω cm).58 Furthermore, Bi-based VTPs have high attenuation coeffi- cients for ionizing radiation.7,59 For example, the calculated X-ray attenuation coefficient of (NH4)3Bi2I9 is comparable to some well-known semiconductors Cs2AgBiBr6, MAPbBr3, CdTe, α-Se and Si.38 It is estimated that (NH4)3Bi2I9 with 0.99 mm thickness can attenuate 99% of 50 keV X-ray pho- tons (i.e., 99% attenuation efficiency), while MAPbBr3 needs 2.28 mm to reach the same attenuation efficiency. Similarly, 2D layered Rb3Bi2I9 more strongly attenuates ionizing radia- tion than CsPbBr3 and Si.37 Other VTP variants, includ- ing 0D Cs3Bi2I9,36 and MA2Bi2I9 40 also exhibit strong X-ray attenuation. Cs3Bi2I9 SCs with 0.5  mm thickness is able to attenuate 94.7% of the incident X-rays, compared to MAPbI3 (87.8%), Cs2AgBiBr6 (65.9%), MAPbBr3 (65.9%) and MAPbCl3 (54.2%).55,60,61 MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    237 The low electronic dimensionality of these VTPs results in spatial confinement of charge carriers, enhancing the Cou- lombic attraction between them, such that the exciton bind- ing energy is high (~ 100 meV), often surpassing the thermal energy (~ 25 meV) at room temperature. This reduces the μτ product, as does ion migration. To address these limitations, various strategies, such as surface passivation, blending with higher-dimensional perovskites, and integrating appropriate charge transport layers, should be explored to improve charge- carrier extraction and enhance detectivity.62 For example, Zhang et al. employed an epitaxial growth strategy to develop 2D/3D heterocrystals, (BA)₂CsAgBiBr₇/Cs₂AgBiBr₆, for X-ray detection.63 The introduction of 2D VTPs induced steric hin- drance and increased the activation energy barrier (0.19 eV in the dark) for ion migration. Other bismuth‑based materials Apart from structural perovskite derivatives, other classes of Bi-based compounds are appealing because of the similarities in the composition of orbitals at band-edges to LHPs, which is conducive towards achieving defect tolerance.64,65 Prominent examples include bismuth oxyiodide (BiOI), bismuth sulfoiodide (BiSI), bismuth iodide (BiI3), bismuth chalcogenides (Bi2X3; X = O, S, Se, and Te), and AgBi2I7.35,66 BiI3 and BiOI have a lay- ered structure belonging to the space group R 3 and P4/nmm (a = b = 3.99 Å and c = 9.21 Å) at room temperature.42 BiI3 adopts highly polar covalent Bi-I bonds in the layer and weak van der Waals bonding between layers.42 BiOI has the stochiometric I-Bi-O-Bi-I layers stacked along the c-axis (presented in Fig. 1). Similarly, bismuth chalcogenides have the same rhombohedral structure (R 3 m) with a quintuple 2D layer. Figure 1.   Schematic representation of different key properties of bismuth-based materials for X-ray detection. 238         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal In 1999, Dmitriyev et al. 67 achieved good resistivity along the c-axis of BiI3 SCs (ρ ~ 108–109 Ω cm), along with decent μτ products of electrons (>10-5 cm2 V-1) and holes (~ 10-7 cm2 V-1). A breakthrough came in 2002, when Matsumoto et al. first reported α-particle detection using an 82 μm-thick BiI3 detector with a clear 5.48 MeV peak and an energy resolution of 2.2 MeV FWHM.68,69 Since then, there has been much ongoing research into BiI3 semiconductors for X-ray detection.42,68,70 Similarly, a high effective atomic number (Zeff =73.2) coupled with a high mass density of 7.97 g cm-3 and high linear attenua- tion coefficient (102 cm-1 at 50 keV) for BiOI makes it a promi- nent contender for X-ray detection. Only 2% of the incident X-rays generated from a source with 40 kV voltage were trans- mitted through a 0.4 mm thick stack of BiOI single crystals, while 78% were transmitted through Si of the same thickness.20 The photo-excited charge carriers in BiOI couple to intralayer breathing phonon modes, forming large polarons. Unusually for Bi-halide semiconductors, carrier localization is avoided in this material, and free carriers occur at room temperature due to a low exciton binding energy.20 At the same time, electron-phonon coupling results in an unavoidable non-radiative loss channel and low carrier mobility at room temperature, which limits the PL lifetime to 2 ns at room temperature, thus limiting diffusion lengths. However, high mobility-lifetime products of 10-3 cm2 V-1 (out-of-plane) and 10-2 cm2 V-1 (in-plane) are still achieved. For example, applying a bias of only 1.8 V across BiOI in the out-of-plane direction (where the mobility-lifetime product is 1.1 × 10-3 cm2 V-1) results in a drift length of 1 mm, exceeding the drift distance required (0.18 mm). We attribute this to the application of an electric field decoupling charge-carriers from the renormalization of the lattice, such that the non-radiative loss channel arising from electron-phonon coupling is avoided, and the drift lifetime then exceeds the diffusion lifetime. An overall summary of the key properties of these bismuth- based materials discussed here for X-ray detection is shown in Table 1. Overall comparison Having discussed the details of Bi-based compounds explored thus far for X-ray detection, we can draw an overall comparison with established commercial materials and lead halide perovs- kites (Figs. 2 and 3). As shown in Fig. 2, Bi-based compounds exhibit equivalent or higher attenuation coefficients than Si, α-Se, CdTe, and MAPbI3 perovskite. This is due to the high effec- tive atomic numbers of these Bi-based compounds (Table 1), meaning that thinner active layers are required, such that the distance for charge-carrier transport can be shorter. Beyond a shorter charge-carrier transport distance, achieving high charge collection efficiency also requires a high μτ product. As shown in Fig. 3b, structurally 2D or 3D Bi-based materials exhibit μτ products (10-4–10-2 cm2 V-1) that are higher than commercial Si, α-Se (~10-7 cm2 V-1), and comparable to CZT (~10-3 cm2 V-1), and III-V semiconductors, which are expensive to manufacture.71 Furthermore, it is important to have high resistivity to effec- tively suppress noise and dark current, which, together with a high μτ product and attenuation coefficient, increase the signal-to- noise ratio and sensitivity, along with reducing the detection limit of the device. As shown in Fig. 3a, Bi-based compounds exhibit higher resistivity than Si, but comparable resistivity to commercial materials. Notably, the bulk resistivity of 0D Cs3Bi2Br9 VTP SC is higher (~1012 Ω.cm), exceeding that of 3D MAPbI3 (107–1010 Ω.cm) and other commercial materials including Si (104 Ω cm), GaAs (108 Ω cm), CZT (1010 Ω cm) and CdTe (109 Ω cm). Beyond these highly promising materials properties, many of the Bi-based compounds explored thus far have demonstrated Figure 2.   Comparison of the linear attenuation coefficient versus X-ray energy of different bismuth-based materials with conventional lead-halide perovskites and semiconductors used commercially in X-ray detectors. (b) Attenuation efficiency versus thickness of bismuth-based materials compared with conventional semiconductors. Calculations were made based on 50 keV X-ray photon irradiation. Data for Fig (a) are obtained from the NIST database,34 while part (b) is calculated using the data shown in part (a) and using Eq. (1). MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    239 high radiation hardness. For instance, Cs2AgBiBr6 SC has shown remarkable stability under X-ray irradiation with no significant change in dark current even after exposure to doses up to 9.2 Gyair, which is equivalent to 92 000 times the dose required for a chest X-ray.28 Even in case of AgBi2I7, high radiation hard- ness has been observed for continuous X-ray irradiation dose of 58 Gyair (43 keV mean energy), which equals 580 000 times the dose required for a single chest radiograph. After this large dose, there was only a small change in dark current, sensitivity and SNR of the detector.35,72 In contrast, CsPbBr3 was reported to degrade after exposure to more than 2 Gyair of radiation, suf- fering a loss of spectral resolution, and requiring post-anneal- ing to recover.73 Similarly, conventional semiconductors, such as Si and CZT exhibit even lower radiation hardness, showing performance degradation after exposure to just a few grays due to defect formation, resulting in increased noise and reduced detection efficiency under high dose rates.74 Compared to these conventional materials, bismuth-based semiconductors exhibit superior radiation hardness, withstanding cumulative doses up to the kGy range without electrical or structural degradation. The origin of the high radiation hardness of Bi-based materials is not yet thoroughly investigated, however, it is likely attributed to the high resistivity that effectively supresses leakage current, and possibly also defect tolerance and self healing in these mate- rials, which allows high resistivities and large μτ products to be maintained.72,74 X‑ray detectors with bismuth‑based materials Bismuth‑based double perovskites Cs2AgBiBr6 was first reported for X-ray detection by Pan et al., 28 who grew SCs by slow cooling from a heated precursor solution. After washing and annealing the surfaces of these SCs to remove impurities, the SCs exhibited a high resistivity of 1.6 × 1011 Ω cm, Figure 3.   Comparison of (a) resistivity, and (b) mobility-lifetime product of some of the best performing Bi-based materials, compared with conventional state-of-art semiconductors used for X-ray detection. The μτ product of Bi-based materials showed slight lower value than MAPbI3, but exceeds most semiconductors used commercially in X-ray detectors. Data collected from the references in Table 1. SC refers to single-crystal material. 240         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal with a trap density of 1.74 × 10⁹ cm⁻3 (Fig. 4b), as determined from space charge-limited current density (SCLC) measurements.75 The corresponding charge-carrier mobility was estimated to be 11.81 cm2 V−1 s−1 using Mott–Gurney law.76 The sensitivity of the detector with the structure of Au/ Cs2AgBiBr6 SC/Au was meas- ured to be 105 µC Gyair −1 cm−2 under an electric field of 25 V mm−1 (Fig. 4c). However, the performance of their device was poorer than LHPs SC X-ray detectors, and further controlled materials growth and device optimisation will be needed.77,78 Steele and co-workers investigated the X-ray detector per- formance of Cs2AgBiBr6 double perovskite SCs at both room temperature and low temperature.44 The Au/Cs2AgBiBr6 SC/ Au detector (Fig. 4d) exhibited a marked increase in X-ray sen- sitivity when cooled to liquid nitrogen temperatures, rising from 316 µC Gyair⁻1 cm⁻2 (room temperature) to 988 µC Gyair⁻1 cm⁻2 (liquid nitrogen temperature) under an applied electric field of 50 V mm−1 (Fig. 4e). A linear fit to the sensitivity measure- ments at different temperatures suggests a coefficient of –3.3 µC Gyair⁻1 cm⁻2 K⁻1. This rise in sensitivity with decreasing tem- perature was attributed to 1) an increase in the mobility-lifetime product, and 2) an increase in resistivity (from 5.5 × 1011 Ω cm at room temperature to 3.6 × 1012 Ω cm at liquid nitrogen tem- perature). The charge-carrier mobility increase was attributed to reduced electron–phonon scattering, which was also partially responsible for the increase in lifetime from 700 ns (room temper- ature) to > 1500 ns (liquid nitrogen temperature). Furthermore, reductions in non-radiative recombination at lower temperatures were also considered to contribute to improved lifetimes. The rise in resistivity with a reduction in temperature was attributed simply to a reduction in thermally generated charge-carriers, in addition to reduced non-radiative recombination.44 Typically, structural distortion in Cs2AgBiBr6 SCs can occur due to the disordered arrangement of Ag⁺ and Bi3⁺, which arises due to their similar ionic radii, and can negatively impact photoelectric performance. In the fully ordered structure of Cs2AgBiBr6, each [AgX6]5− octahedron is surrounded by six Table 1.   Key properties of representative bismuth-based materials for X-ray detection. a The X-ray attenuation efficiency is the fraction of the incident X-ray intensity attenuated within a specified thickness of material Materials Zeff Mass Density (g cm–3) Linear Attenuation efficiencya Bandgap (eV) Reference Cs2AgBiBr6 60.0 4.65 99% (1.18 mm, 50 keV) 2.10–2.27 28, 29 BA2CsAgBiBr7 – – 99% (20 nm, 70 keV) 2.38 30 (DMEDA)BiI5 – 3.83 93.2% (690 μm, 50 keV) 1.86 27 (4,4-DPP)4AgBiI8 – 2.85 99% (1 mm, 40 keV) 2.03 31 AgBi2I7 – – ~ 97% (100 keV, 0.5 mm) 1.73 35 Cs3Bi2I9 – 5.02 94.7% (0.5 mm, 40 keV) 1.94–2.0 36 Rb3Bi2I9 61.6 4.67 90% (30 keV, 0.4 mm) 1.89 37 (NH4)3Bi2I9 30.9 4.30 99% (50 keV, 0.99 mm) 2.05 38 (H2MDAP)BiI5 – 4.36 99.8% (40 keV, 0.4 mm) 1.83 39 MA3Bi2I9 – 3.8–4.1 90% (40 keV, 0.3 mm) 1.98 40 FA3Bi2I9 – – 99.8% (40 keV, 0.9 mm) 2.08 41 BiI3 83.5 5.78 99.8% (50 keV, 1 mm) 1.7–2.2 70 Bi2O3 79.3 8.9 ~ 63% (50 μm, 70 kV, 124 GW) 2.83 43 BiOI 73.6 7.97 90%, (30 keV, 134 μm) 1.93 20 MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    241 [BiX6]3− octahedra. However, in the case of partial or com- plete disorder, one or more of these six [BiX6]3− octahedra may be replaced by [AgX6]5−octahedra. Yuan et  al. incorporated phenethylammonium bromide (PEABr) into the Cs2AgBiBr6 per- ovskite precursor to synthesize PEA-Cs2AgBiBr6 SCs for X-ray detection.45 It was shown that PEABr can effectively suppress the order–disorder phase transition in Cs2AgBiBr6 SCs, thereby enhancing the X-ray sensitivity of the device. They quantitatively assessed the degree of ordering in Cs2AgBiBr6 by evaluating the diffraction intensity ratio between the (111) and (022) X-ray diffraction peaks (I111/I022). The ordering parameter (H) was determined by comparing the ratio of the observed superlattice reflection (111) to the base lattice reflection (022) in the SC with the calculated intensity ratio for a perfectly ordered structure.79 (5)H 2 = ( I111 I022 ) observed ( I111 I022 ) calculated Figure 4.   (a) Comparison of the attenuation coefficients of Cs2AgBiBr6 double perovskite with MAPbBr3, CdTe, and Si as a function of photon energy. (b) Current–voltage characteristics of Cs2AgBiBr6 single crystal, featuring linear and quadratic fittings based on the space charge-limited current (SCLC) model. (c) The sensitivity of the SC detector under different bias voltages. Reproduced from, Ref. 28 with permission from Springer Nature. (d) Top: photograph of Au/Cs2AgBiBr6/Au X-ray detector, bottom: Temporal X-ray current of the detector at different measurement temperatures and applied electric field of 50 V mm−1. (e) X-ray sensitivities of the device at different measurement temperatures. Reproduced from, Ref. 44 John Wiley & Sons. © 2018 WILEY–VCH Verlag GmbH &Co. KGaA, Weinheim. (f) I–V curves of pristine Cs2AgBiBr6 and PEA-Cs2AgBiBr6 SC devices. The inset illustrates the schematic of the device’s working mechanism. (g) Comparison of X-ray sensitivities of pristine Cs2AgBiBr6 and PEA-Cs2AgBiBr6 SC detector under different applied biases. Reproduced from Ref. 45 © 2019 WILEY–VCH Verlag GmbH & Co. KGaA, Weinheim. (h) Comparison of the X-ray sensitivity of Cs2AgBiBr6 SCs synthesized by the natural and controlled cooling method under different applied electric fields. Reproduced from Ref.46© 2019 WILEY–VCH Verlag GmbH & Co. KGaA, Weinheim. (i) Schematic illustration of the Cs2AgBiBr6 double perovskite crystal structure (left) and site-specific substitutions (Cs-site, cyan, and Bi-site magenta) (right). (j) Sensitivity for the undoped (middle, labeled “D”) and doped (Cs-site cyan and Bi-site magenta shade) detectors. Reproduced from Ref.47 © 2024, American Chemical Society. 242         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal As the PEA concentration increases, the I111/I022 ratio becomes larger, indicating an enhanced ordering of Bi3⁺ and Ag⁺. This suggests that the PEABr precursor effectively pro- motes cation ordering. The µτ value of PEA-Cs2AgBiBr6 SCs (1.94 × 10–3 cm2 V−1 s−1) exceeded that of pristine Cs2AgBiBr6 SCs (9.14 × 10−4 cm2  V−1  s−1) (Fig. 4f). The sensitivity of the detector was also improved up to 288.8 µCGyair⁻1 cm⁻2 at an electric field of 22.7 V mm−1 (Fig. 4g). Chemical treatment applied after growth can further improve crystal quality and enhance detector performance. BiBr3 is a cru- cial precursor in the synthesis of Cs2AgBiBr6 SCs, and its residue on the double perovskite SC surface contributes to the formation of surface conduction channels.77 Zhang and colleagues inves- tigated various post-growth treatment processes and found that rinsing Cs2AgBiBr6 SCs with isopropanol and applying thermal annealing effectively mitigated field-driven ion migration and surface conduction channels, as well as reducing the detector’s noise current.77 The X-ray sensitivity of the detector was esti- mated to be 316.8 µC Gyair −1 cm−2 at an electric field of 6 V mm−1. Yin et al. reported a controlled cooling process for the synthesis of Cs2AgBiBr6 SC for X-ray detection.46 In comparison to the conventional natural cooling method, the controlled cooling process resulted in bismuth-based SC with smooth surfaces, higher resistivity, and improved reproducibility. Thus, the detec- tor exhibits an X-ray sensitivity of 1974 µC Gyair −1 cm−2 under an applied electric field of 50 V mm−1 which is higher than the sensitivity value of SC grown by the natural cooling method (Fig. 4h). The SC detector also demonstrates a very low LoDD of 45.7 nGyair s−1 at 50 V mm−1 applied electric field, surpass- ing previous reports for Cs2AgBiBr6 SC X-ray detectors. Donato et al. demonstrated that growing the perovskite in a slightly Bi- deficient and Eu-enriched environment significantly boosts X-ray sensitivity from 17 to 120 μC Gyair –1 cm–2. Furthermore, substituting Cs sites with imidazolium enhances the sensitivity even more, reaching over 180 μC Gyair –1 cm–2 due to higher X-ray attenuation.47 Figure 4i presents a schematic illustration of the Cs2AgBiBr6 crystal structure on the left, alongside site-specific substitutions on the right. Cs-site substitutions are ammonium (A), guanidinium (G), triazolium (T), and imidazolium (Im). Figure 4j shows the trend of the X-ray sensitivity of the pristine (D) and doped samples. The overall increase in sensitivity can be attributed to the fact that lanthanide cations have a K-edge energy value of approximately 50 keV, helping to give to higher X-ray attenuation. Thus, many engineering strategies have been demonstrated to successfully improve the performance of Cs2AgBiBr6. However, the LoDD and sensitivities reported are currently still at least an order of magnitude inferior to LHPs, and the inherent self-trapping present in Cs2AgBiBr6 could be a key limiting factor, as discussed earlier. It is therefore important to explore alternative Bi-based materials that could overcome these limitations. Bismuth‑based 2D perovskites 2D Cs3Bi2Br9 VTPs can be an excellent candidate for sensi- tive X-ray detection due to high X-ray attenuation, and high bulk resistivity with a decent mobility-lifetime product (Table 1). The layered structure of Cs3Bi2Br9 VTP results in anisotropic electronic properties. The limited carrier transport in the out- of-plane direction of the SC helps to reduce noise and lower dark current drift due to ion migration in vertically structured devices, ultimately contributing to a lower LoDD. However, the vertical device can exhibit lower X-ray sensitivity compared to the planar device due to its reduced mobility. Saqr et  al. reported solution-grown Cs3Bi2Br9 SCs for direct X-ray detec- tion.80 Ag/Cs3Bi2Br9 SC/Ag vertical devices exhibited a resis- tivity of 1.79 × 1011 Ω cm and μτ product of 5.12 × 10−4 cm2 V−1. Compared to the low-temperature solution-growth technique, the high-temperature melt growth method offers significant advantages in producing high-quality, large-area SCs.81,82 This approach enables the formation of crystals with superior structural integrity and substantially reduced defect density. Xiang and co-workers reported high-quality 2D Cs3Bi2Br9 VTP SCs grown from a melt via the Bridgman method shown in Fig. 5a.83,84 The bismuth-based perovskite SC exhibited a high resistivity of 1.41 × 1012 Ω cm and mobility-lifetime product of 8.32 × 10–4 cm2 V−1. The Au/Cs3Bi2Br9 SC/Au device demon- strates impressive sensitivity, achieving 1705 µC Gyair −1 cm−2 under an applied electric field of 1000 V mm−1 (Fig. 5b), along with an exceptionally low detection limit of 0.58 nGyair s−1 for detecting 120 keV hard X-rays (Fig. 5c). The Cs3Bi2Br9 detector also shows remarkable operational stability, featuring a mini- mal dark current drift of 2.8 × 10–10 nA cm−1 s−1 V−1 and long- term stability in air under a high electric field of 1000 V mm−1, attributed to the high activation energy barrier to ion migration, which arises as a result of its 2D structure. Thus, the device per- formance of bismuth-based 2D perovskite SCs is on par with that of conventional 3D LHP SCs.85 Zhi synthesized 2D Cs3Bi2Br9 nanoflakes using inversion temperature crystallization (ITC) for high-performance X-ray detection.86 Without the addition of AgBr in the precursor solu- tion, the nanoflakes show a rectangular morphology with CsBiO3 impurities. After adding AgBr to the precursor, the Br− vacan- cies in the lattice of Cs3Bi2Br9 were passivated. Br atoms at the vertices of [BiBr6]⁻ octahedra in Cs3Bi2Br9 can easily migrate, creating Br− vacancies. These vacancies trap electrons and desta- bilize the structure, making it prone to oxidation in HBr solu- tion, leading to the formation of CsBiO3 and Cs3Bi2Br9 hybrids. Adding AgBr to the precursor solution reduces Br⁻ vacancies by filling trap states and inhibiting Br migration, thus enhanc- ing structural stability. The synthesized highly crystalline 2D Cs3Bi2Br9 nanoflakes exhibit a direct bandgap, with a mobility- lifetime product reaching 9.8 × 10⁻4 cm2 V⁻1. Notably, devices constructed from these 2D nanoflakes demonstrate a high sen- sitivity of 1.9 × 105 μC Gyair −1 cm−2, attributed to photoconduc- tive gain. Figure 5d shows the comparison of the dark current and X-ray photocurrent of 2D Cs3Bi2Br9 nanoflake detectors under X-ray irradiation with a dose rate of 220 µGyair s−1, and the optical micrograph of the device is inset. The sensitivity of the detector increases up to 1.9 × 106 µC Gyair −1 cm−2 as the elec- tric field increases from 0 to 3.3 V µm−1 as shown in Fig. 5e. The gain of the detector is approximately 8.9 × 108, confirming that MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    243 the photoconductive gain mechanism contributes to the excep- tional sensitivity of devices made from 2D Cs3Bi2Br9 nanoflakes (Fig. 5e). While photoconductive gain can enhance sensitivity, it also increases the noise current, which in turn restricts the LoDD. Li et al. demonstrated Cs3Bi2Br9 thick films grown by CVD.89 The vertical device structure, with the configuration of Au/ Cs3Bi2Br9 films /SnO2/ITO, demonstrates good X-ray sensi- tivity of 593 μC Gyair −1 cm−2, a low detection limit of 187.7 nGyair s−1, and outstanding stability, maintaining performance after 20 days in ambient conditions and during continuous operation for over 2 h. X-ray detection without an applied bias is highly desirable for developing energy-efficient, portable detectors, with potential applications in biomedical imaging, radiation dose monitoring, and security scanning for remote or inaccessible locations. Wu and co-workers demonstrated a chirality-induced polar pho- tovoltaic effect in a chiral-polar 2D bismuth-based perovskite (R-MPA)4AgBiI8 (R-MPA = R-β-methylphenethylammonium) SCs for self-powered X-ray detection.90 The strong spontaneous electric polarization in bismuth-based SC results in a notable polar photovoltage of 0.36 V, which facilitates the separation and transport of X-ray-generated charge-carriers, enabling self- powered detection. As a result, X-ray detectors constructed from high-quality SCs of (R-MPA)4AgBiI8 demonstrate a high sensitiv- ity of 46.3 μC Gyair –1 cm–2 and a low LoDD of 85 nGyair s –1 at zero bias. The performance of this self-powered X-ray detector was comparable to that of other reported lead-based 2D chiral-polar perovskite SCs.91–93 This sensitivity can be further enhanced to 949.6 μC Gyair –1 cm–2 when applying a bias of 50 V across the electrodes. Solution-grown bismuth-based layered 2D hybrid double per- ovskite SCs have also been reported for X-ray detection.30,31,94 Wu et al. reported bismuth-based (4-AP)2AgBiBr8 (4-AP = 4-ami- dinopyridine) Dion-Jacobson (DJ) 2D perovskite SC for sensi- tive X-ray detection (Fig. 5f).87 In this DJ structure, the AgBr₆ and BiBr₆ octahedra are alternately arranged and corner shar- ing, creating a 2D inorganic monolayer. Figure 5g shows the variation of the photocurrent density of the Ag/(4-AP)2AgBiBr8 SC/Ag detector with X-ray irradiation dose rates under differ- ent applied bias voltages. The slope of the linear fit represents the sensitivity of the detector. The SC detector exhibits a high sensitivity value of 1117.3 μC Gyair –1  cm–2 under applied bias of 80 V. Huanyu and co-workers presented the comparison of X-ray detection performance of three different layered hybrid silver bismuth bromine SCs: (BDA)2AgBiBr8 (BDA = 1,4-diam- inobutane), (BA)4AgBiBr8 (BA = n-butylamine), PA4AgBiBr8 (PA = n-propylamine).88 Figure  5h presents the comparison of different device parameters of three different SCs in the in- plane (parallel) and out-of-plane (perpendicular) directions. The (BDA)2AgBiBr8 SC demonstrated a high μτ value, bulk resistivity, and an excellent on/off ratio, making it a promising candidate for X-ray detection. In optimized in-plane devices, the detectors based on (BDA)2AgBiBr8 achieved a sensitiv- ity of 2638 µC Gyair⁻1 cm⁻2 and an exceptionally low detection limit of 7.4 nGyair s⁻1 (Fig. 5i). 2D bismuth-based (F-PEA)3BiI6 [(F-PEA) = 4-fluorophenethylammonium] pressed wafer with an area of 1.33 cm2 was reported with sensitivity of 52.6 µC Gyair⁻1 cm⁻2 and LoDD of 30 nGyair s −1.95 It is important to note that large-area devices are highly desirable for the fabrication of multi-pixel X-ray imagers. Thus, low-cost solution-grown bis- muth-based layered perovskites could be promising candidates for X-ray detection. Bismuth‑based 0D perovskites Zhang et  al. developed a nucleation-controlled solution method to synthesize large-size high-quality Cs3Bi2I9 perovs- kite zero-dimensional (0D) SCs as shown in Fig.  6a.36 After filtration, the CsI and BiI3 precursor solution was placed in a temperature-controlled oven. The temperature was raised to 80°C, such that sub-millimeter Cs3Bi2I9 SCs were precipitated out. To eliminate nucleation seeds, the system was maintained at this temperature for 24 h. Once the solution reached saturation, the excess material recrystallized. The supernatant was carefully transferred to a new container to grow large single crystals. The structure of these VTPs was discussed earlier in Sect. “Bismuth- based perovskite derivatives”. The SC exhibited a high resistiv- ity of 2.79 × 1010 Ω cm with a μτ value of 7.97 × 10–4 cm2  V−1, and devices achieved a sensitivity of 1652.3 μCGyair −1  cm−2 at 50 V mm−1 applied electric field (Fig. 6c, with an LoDD of 130 nGyair s−1. Liu et  al. reported inch-sized 0D MA3Bi2I9 (MA = CH3NH3) SCs grown by solution processing method for sensitive X-ray detection and imaging.96 The bismuth-based SC detector demonstrated a high resistivity of 3.74 × 1010 Ω cm and a substantial μτ product of 2.87 × 10−3 cm2  V−1. The SC X-ray detector exhibited a very high sensitivity of 1947 μC Gyair −1 cm−2 under an electric field of 60 V mm−1 (Fig. 6d), a low detection limit of 83 nGyair s −1, and a short response time of 23.3 ms. Addi- tionally, the bismuth-based SC is utilized further for demon- stration of X-ray imaging as shown in Fig. 6e–f. Zheng reported bismuth-based MA3Bi2I9 SCs with a high sensitivity of 10620 µC Gyair −1 cm−2 and ultra-low LoDD of 0.62 nGyair s −1.14 The copla- nar detector made from MA3Bi2I9 SC demonstrates impressive performance, including a low dark noise current and exceptional X-ray response.40 Charge-carrier transport within the MA3Bi2I9 SC differs along the [010] and [001] directions due to scatter- ing effects. In a coplanar device, carriers transfer along the [Bi2I9]3− intralayer, while in a vertical device, they must travel through the interlayer along the [001] direction. During the growth of the SC, potential barriers such as ion vacancies, traps, and disorder states are typically introduced into the interlayer between the [Bi2I9]3− monolayers. These act as scattering cent- ers, which reduce carrier mobility. As a result, carrier mobility is expected to be higher in a coplanar structure than in a vertical one. It features a high sensitivity of 872 μC Gyair⁻1 cm⁻2, a rapid response time of 266 μs, and a low detection limit of 31 nGyair s⁻1.40 Additionally, the detector offers a spatial resolution of 4.22 lp mm⁻1 and long-term stability, with a small area single pixel device. Li and co-workers reported FA3Bi2I9 (FA = CH(NH2)2) SCs grown by the nucleation-controlled secondary solution constant temperature evaporation (SSCE) method.41 The 244         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal temporal X-ray response of Au/FA3Bi2I9/Au detector to X-rays with dose rates from 0.5 to 13.1 μGys−1 under a bias of 180 V is shown in Fig. 6g. The SC crystal detector exhibited a sensitivity of 598.1 μC Gyair −1 cm−2 and an LoDD of 0.2 μGyair s −1, which are improved over commercial α-Se detectors. It is important to note that large-area films and wafers are highly desired for the production of flat-panel X-ray detectors, which are essential for X-ray imaging applications. Polycrystal- line bismuth-based 0D perovskite thick films and pellets have also been reported for X-ray detection.61,98 Xu and co-workers reported a high-quality large-area amorphous MTP3Bi2X9 (meth- yltriphenylphosphonium = MTP and X = Cl, Br, or I) wafers grown by the melt-quenching method.97 Figure  6h depicts the photographs and ultraviolet–visible (UV–vis) transmission spectra of the MTP3Bi2Cl9, MTP3Bi2I9, and MTP3Bi2I9 wafers. MTP3Bi2I9 amorphous wafer exhibited high X-ray sensitivity of 7601 μC Gyair −1 cm−2 under 200 V of applied bias (Fig. 6i). However, further research is needed to fabricate and optimize multipixel, large-area X-ray imagers using this bismuth-based material for potential commercial applications. Figure 5.   (a) Photograph of as-grown 2D Cs3Bi2Br9 SC grown by the Bridgman method. (b) X-ray sensitivities of Cs3Bi2Br9 and Cs2AgBiBr6 SC detectors tested under 120 keV hard X-rays. (c) The SNRs of Cs3Bi2Br9 SC and Cs2AgBiBr6 SC under varying X-ray dose rates. The dashed line indicates an SNR of 3 to measure LoDD. Reproduced from Ref. 83 American Chemical Society © 2022. (d) The dark current and photocurrent of a 2D Cs3Bi2Br9 nanoflake device under X-ray irradiation of 220 µGy s⁻1. The inset shows the optical microscopy image of the device. (e) The sensitivity and gain of the detector under different applied electric fields. Reproduced from Ref. 86 John Wiley and Sons © 2023 Wiley‐VCH GmbH. (f) Crystal structure of (4-AP)2AgBiBr8 2D perovskite. (g) The variation of photocurrent density of the Ag/(4-AP)2AgBiBr8 SC/Ag under varying applied voltages and irradiation dose rates. Reproduced from Ref. 87 John Wiley and Sons © 2024 Wiley‐VCH GmbH. (h) Comparison of different detector parameters of (PA)4AgBiBr8, (BA)4AgBiBr8, and (BDA)2AgBiBr8 SC. (i) The X-ray dose rate dependence of SNR of the detectors under an external bias voltage of 200 V. The LoDD is determined from the fitting line corresponding to an SNR of 3. Reproduced from Ref. 88 © John Wiley and Sons 2023 Wiley‐VCH GmbH. MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    245 Other bismuth‑based materials for X‑ray detection In addition to bismuth-based perovskite materials, high-per- formance X-ray detectors have also been reported using Bi-based compounds without the perovskite structure, including AgBi₂I₇ SCs, BiI₃, Bi₂O₃, and layered BiOI SCs. Tie et  al. reported AgBi2I7 SCs (Fig. 7a) grown by a vertical Bridgman technique.35 The μτ values of the SC detector were 3.4 × 10–3 and 1.2 × 10–3 cm2 V–1 for electrons and holes, respectively. The AgBi2I7 SC exhibited very high mobility values of 492.1–859.3 cm2 V–1 s–1 and 296.2–702.5 cm2 V–1 s–1 for electrons and holes, respectively. The X-ray sensitivity of the Au/AgBi2I7 SC/Au SC detector was obtained to be 282.5 μC Gyair –1 cm–2 (Fig. 7b) while the detec- tor demonstrated a LoDD of 72 nGyair s–1. Sun and co-workers demonstrated free-standing BiI3 SC flakes grown by the physical vapor transport method as shown in Fig. 7c.99 The SC detector exhibited very high sensitivity of 1.22–1.36 × 104 µCGyair −1 cm−2 along [001] direction (Fig. 7d). These enhancements in the measured sensitivity can also result from photoconductive gain. Au/BiI3 SC/Au X-ray detector was reported with a high signal- to-noise ratio of 896.4 and a sensitivity up to 0.526 × 104 μC Gyair −1 cm−2 along the c-axis direction under an electric field of 0.02 V μm−1 and X-ray dose rate of 489.78 μGy h.−1.100 Fan and co-workers recently reported bismuth vanadate (BiVO4) sintered pellets for 110 kVp hard X-ray detection.101 The comparison of the X-ray attenuation coefficient with the photon energy of BiVO4 with other detector materials is pre- sented in Fig.  7e. BiVO4 offers excellent X-ray attenuation, particularly for photons with energies greater than 90 keV. At a thickness of 1 mm, BiVO₄ achieves an impressive attenuation efficiency of 97.1%, significantly outperforming other materi- als, which remain below 80%. Ultra-stable BiVO₄ metal oxide X-ray detectors exhibit a high sensitivity of 3164 μC Gyair⁻1 cm⁻2 and a low detection limit of 20.76 nGyair s⁻1 under 110 kVp hard Figure 6.   (a) Photographs of as-grown Cs3Bi2I9 0D SC. (b) Crystal structure of a 2 × 2 × 1 supercell of Cs3Bi2I9. (c) X-ray sensitivity of the detector under different applied electric fields. Reproduced from Ref. 36 Springer Nature Copyright © 2020. (d) X-ray sensitivity of the MA3Bi2I9 SC detector under different applied electric fields. (e) Photograph and (f) corresponding X-ray images obtained from MA3Bi2I9 SC detector. Reproduced from Ref. 96 © 2020 Elsevier Inc. (g) Temporal response of Au/FA3Bi2I9/Au detector to X-rays with different dose rates under a bias of 180 V. Reproduced from Ref. 41 Copyright © 2021, American Chemical Society (h) Photographs and UV–vis transmittance spectra of MTP3Bi2Cl9, MTP3Bi2I9 and MTP3Bi2I9 wafers. (i) X-ray sensitivity of Au/ MTP3Bi2I9 amorphous wafer/Au detector under different applied bias. Reproduced from Ref. 97 © 2024 Wiley‐VCH GmbH. 246         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal X-rays, setting a new benchmark for X-ray detectors based on polycrystalline Bi-halides and metal oxides. BiVO4 pellet X-ray detector was reported with a large resistivity of 1.3 × 1012 Ω cm, negligible current drift of 6.18 × 10−8 nA cm−1 s−1 V−1, a high µτ value of 1.75 × 10−4 cm2  V−1, an X-ray sensitivity of 241.3 µC Gyair −1 cm−2 and a detection limit of 62 nGyair s−1 under 40 kVp X-ray illumination.103 Ceramic BiVO4 wafers exhibit lower charge-carrier mobility and mobility-lifetime products than bismuth-based perovskite single crystals, mainly due to their polycrystalline nature. This limitation in electronic properties can impact the overall efficiency and performance of detectors utilizing BiVO₄ ceramics. Additionally, the fabrication process for BiVO₄ ceramic wafers involves multiple complex steps. One of the critical stages is energy-intensive high-temperature sin- tering, conducted within a temperature range of 650 to 800°C. Furthermore, Praveenkumar et al. synthesized phase-pure Bi5O7I NCs, and X-ray detectors based on this achieved a sensi- tivity of 1.92 ×10-2 μC Gyair -1cm-2.104,105 Due to their low cost and compatibility with solution processing, organic semiconductors based on conjugated polymers and small molecules hold prom- ise for use in X-ray detectors, particularly f lexible, wearable detectors. However, their performance is limited by inherently poor X-ray attenuation. To address this limitation, inorganic nanomaterials with high atomic numbers can be incorporated to enhance the sensitivity of organic semiconductor devices for ionizing radiation detection. Jayawardena et al. showed a direct X-ray detector on Bi2O3 nanoparticles dispersed in poly(3-hex- ylthiophene-2,5-diyl) (P3HT) and [6,6]-phenyl C71 butyric acid methyl ester (PC70BM) with sensitivity 160 μC mGyair −1 cm−3 exhibits almost 100% attenuation.43 Thirimanne et al. incor- porated bismuth oxide (Bi2O3) nanoparticles with high atomic number into an organic bulk heterojunction for X-ray detec- tion.43,102 Figure 7g depicts the variation of X-ray sensitivity of ITO/PEDOT:PSS/P3HT:PC70BM:Bi2O3/Al device with Bi2O3 nanoparticle loading. These hybrid detectors demonstrated X-ray sensitivities of 1712 µC mGyair −1cm−3 for soft X-rays and ~30 and 58 µC mGyair −1 cm−3 under 6 and 15 MV hard X-rays. Recently, Jagt et al. demonstrated layered bismuth oxyiodide (BiOI) SCs for X-ray detection with low LoDD. As discussed in Section “Key properties of bismuth-based materials for X-ray detection”, BiOI is unusual among Bi-halide materials by having band-like transport, enabling high mobility-lifetime products of 1.1× 10−3 cm2 V-1 (out-of-plane) and 6 × 10−2 cm2 V−1 (in-plane).20 These SC detectors exhibited ultra-low LoDD of 1.1 nGyair s−1 (Fig. 7h) and a high sensitivity of 1.1 × 103 μC Gyair −1 cm−2 under 5 V applied bias in the out-of-plane direction (Fig. 7i). Despite exhibiting band-like charge-carrier transport, BiOI still suffers from non-radiative losses due to strong electron-phonon cou- pling. Its layered crystal structure limits single-crystal thickness results in anisotropic growth, such that SCs are only a few hun- dred microns thick (despite having lateral dimensions > 5 mm), which is insufficient for high stopping power with high-energy X-rays. Additionally, the low yield, high temperature processing, and challenges with large crystal growth limit the practical appli- cations of these SCs. As a result, polycrystalline BiOI wafers or thick films offer a more viable alternative. They allow for greater thickness and area coverage, improving X-ray attenuation, and are compatible with scalable fabrication, and should be explored in the future. The surface of SCs usually contains a large number of dan- gling bonds, under-coordinated atoms, surface dislocations, and chemical impurities.106 There are notable differences in the physical properties between the surface and bulk of SCs. Therefore, surface passivation and heterojunction forma- tion can be effective strategies to improve detection perfor- mance. Recently, solution-grown thick BiI/BiI3/BiI (BixIy) van der Waals heterostructures were reported with a sensitiv- ity up to 4.3 × 104 μC Gyair −1 cm−2 and a detection limit as low as 34 nGyair s−1.70 Therefore, bismuth iodide and various oxide materials show significant promise for the development of highly sensitive X-ray detection and imaging technologies. A summary of the X-ray sensitivity, LoDD, and other device parameters for various bismuth-based materials is presented in Table 2. Conclusions and outlook In conclusion, Bi-based compounds exhibit several appeal- ing properties for X-ray detection that have sparked a resur- gence of interest in this area. The composition of heavy ele- ments, and the high mass density of these materials lead to strong attenuation of ionizing radiation. Combined with the high mobility-lifetime products (reaching > 10–2 cm2  V−1 in some cases) and low dark current densities, high sensitivities (> 104 μC Gyair −1 cm−2) and low LoDD < 10 nGyair s−1 have been achieved in Bi-based materials used in direct X-ray detectors. The high versatility of these materials is such that the struc- tural dimensionality can be tuned from 3D (e.g., Cs2AgBiBr6) to 0D (e.g., (MA)3Bi2I9). Owing to higher effective masses, low-dimensional materials, especially VTPs, benefit from reduced dark currents that are conducive toward achieving lower LoDDs. Bi-based compounds, in addition to having low toxicity, also exhibit high ambient stability in many cases (e.g., BiOI, and most VTPs and double perovskites), with no phase impurities forming after weeks of storage in air. Out of this wide variety of materials, we especially highlight VTPs, which have achieved both the lowest detection limits and highest sensitivities reported in Table 2. Both MA3Bi2I9 and Cs3Bi2Br6 exhibit LoDDs < 1 nGyair s−1, along with sensitivities exceeding 103 μC Gyair −1 cm−2. Sensitivities as high as > 105 μC Gyair −1 cm−2 were also reported from 2D Cs3Bi2Br6, and this was likely enhanced through photoconductive gain. Other notable materials include passivated BixIy and BiOI, which also have low LoDD and high sensitivities. BiOI is particu- larly promising because of its absence of exciton formation at room temperature (unlike VTPs) or carrier localization (unlike Cs2AgBiBr6 elpasolites). We therefore believe that these mate- rials are especially worth emphasis in future efforts at develop- ing X-ray detectors from Bi-based materials. There has been thriving research in the area of Bi-based com- pounds for X-ray detection. Taking these materials forward, it is important to go beyond simple demonstrations of Bi-based MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    247 materials in single test devices at the lab scale. For practical use in medical imaging, large-area multi-pixel flat-panel detectors are required. Here, it is important to increase the size of the devices from the mm-level (as achievable with single crystals) to the cm-level, such that these devices can be tiled to form flat-panel imagers. Promising routes forward include the fab- rication of polycrystalline wafers (by pressing together single crystals or powders), deposition of thick films (> 200 μm), or the formation of flexible nanocomposite arrays. These materials have the advantage of being more cost-effective to synthesize than large single crystals, but it will be essential to mitigate any increases in dark current or ion migration via grain bounda- ries or structural defects. This includes performing in-depth characterization to understand the role of grain boundaries on non-radiative recombination and charge-carrier scattering, for example through cathodoluminescence mapping, or fabricating the materials into thin film transistors and measuring the tem- perature-dependence of the field-effect mobility.113 Strategies to address deleterious effects at grain boundaries, surfaces and interfaces include developing passivation approaches and post- deposition heat treatments to reduce the density of structural defects or strain in the materials, such that mobility-lifetime products approach the values of their single-crystal counter- parts. Heteroepitaxial passivation has arisen as a particularly promising route. For example, a heteroepitaxial layer of BiOBr was grown onto polycrystalline Cs2AgBiBr6 to suppress non- radiative recombination and ion migration, such that these materials performed comparably to single crystals in X-ray Figure 7.   (a) Photograph of the AgBi2I7 crystals grown by the vertical Bridgman technique. (b) Current density (Js) of the AgBi2I7 detector as a function of irradiation dose rate. The slope of the linear fit corresponds to the sensitivity of the device. Reproduced from Ref. 35 Copyright © 2020, American Chemical Society. (c) Schematic of the fabrication process of BiI3 SC via the physical vapor transport method. (d) Sensitivity of multiple BiI3 detectors. Reproduced from Ref. 99 Copyright © 2023 Wiley–VCH GmbH. (e) Comparison of the attenuation coefficient of BiVO4 as a function of photon energy with MAPbI3, CZT, and α-Se. (f) X-ray sensitivity of the BiVO4 detector as a function of applied bias. Reproduced from Ref. 101 Copyright © 2018 Springer Nature. (g) Variation of X-ray sensitivity of ITO/PEDOT:PSS/P3HT:PC70BM:Bi2O3/Al device with Bi2O3 nanoparticle loading. Reproduced from Ref. 102 Copyright © 2018 Springer Nature. (h) SNR of Au/BiOI/Au SC detector as a function of X-ray dose rate, and the dashed red line shows the corresponding dose rate (1.1 nGyair s −1) at which the SNR value is 3. (i) X-ray sensitivity of the perpendicular BiOI SC detector as a function of dose rates. Reproduced from Ref. 20 Copyright © 2023 Springer Nature. 248         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal Table 2.   Summary of the performance of X-ray detectors made from bismuth-based materials. Device configuration μτ product (cm2 V−1) Resistivity (Ω cm) Electric field (V mm−1) LoDD (nGyair s −1) Sensitivity (µC Gyair −1 cm−2) Reference Au/Cs2AgBiBr6 SC/Au 6.3 × 10−3 1.6 × 1011 25 59.7 105 28 Au/Cs2AgBiBr6 SC/Au – 3.6 × 1012 50 – 316 at 300 K 988 at 77 K 44 Au/PEA-Cs2AgBiBr6 SC/Au 1.94 × 10−3 – 22.7 – 288.8 45 Au/Cs2AgBiBr6 SC/Au – – 6 – 316 77 Au/Cs2AgBiBr6 SC/Au 5.95 × 10−3 3.31 × 1010 50 45.7 1974 46 Au/Imidazolium sub- stituted Cs2AgBiBr6 SC/Au – – Bias = 10 V – 180 47 Au/Cs2AgBiCl6 SC/Au 5.36 × 10–4 3.1 × 1010 40  < 241 325.8 107 Au/ BiOBr passivated Cs2AgBiBr6 wafer/ Au 5.51 × 10−3 1.6 × 1010 500 95.3 250 108 Au/Cs2AgBiBr6 film/ Au – – – 145.2 1.8 × 104 109 Au/Cs2AgBiBr6:PVA film/Au flexible device – 2.0 × 1011 4000 – 40 110 Au/ Cs3Bi2Br9 SC /Au 8.32 × 10−4 1.41 × 1012 1000 0.58 1705 83 Au/ Cs3Bi2Br9 SC /Au – 6.8 × 1011 100 – 230.4 84 Ag/ Cs3Bi2Br9 SC /Ag 5.12 × 10−4 1.79 × 1011 – – – 80 Au/Cs3Bi2I6Br3 SC/Au – 2.3 × 1010 30 – 55.62 81 Au/2D Cs3Bi2Br9 nanoflakes/Au 9.8 × 10−4 1.6 × 107 7142 – 1.9 × 105 86 ITO/SnO2/Cs3Bi2Br9 films /Ag – –  ~ 625 187.7 593 89 Au/(BA)2CsAgBiBr7 SC/Au 1.21 × 10−3 – 5 – 4.2 30 MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    249 Table 2.   (continued) Device configuration μτ product (cm2 V−1) Resistivity (Ω cm) Electric field (V mm−1) LoDD (nGyair s −1) Sensitivity (µC Gyair −1 cm−2) Reference Au/(4,4-DPP)4AgBiI8 SC/Au – – Bias = 50 V 3130 188 31 Ag/(I-BA)4AgBiI8 SC/Ag 2.28 × 10−3 3.04 × 1010 4.5 – 5.38 94 Ag/(R-MPA)4AgBiI8 SC/Ag 2.2 × 10−5 1.54 × 1010 0 85 46.3 90 Ag/(4-AP)2AgBiBr8 SC/Ag 4.8 × 10–4 1.7 × 1011 Bias = 80 V 279 1117.3 87 Carbon/ (BDA)2AgBiBr8/ Carbon 3.12 × 10–2 4.25 × 109 Bias = 200 V 7.4 2638 88 Ag/(H2MDAP)BiI5 SC/Ag – 2.1 × 1010 5 – 1.0 39 Au/(F-PEA)3BiI6 wafer /C60/ BCP/Cr 8.3 × 10–5 2 × 1011 100 30 52.6 95 Au/Cs3Bi2I9 SC/Au 7.97 × 10–4 2.79 × 1010 50 130 1652.3 36 Au/ MA3Bi2I9 SC/Au 2.87 × 10−3 3.74 × 1010 60 83 1947 96 Au/MA3Bi2I9 SC/Au 2.8 × 10−3 5.27 × 1011  ~ 120 0.62 10,620 14 Au/MA3Bi2I9 SC/Au – 4.7 × 1010 2860 31 872 40 ITO/MA3Bi2I9 film/Au 3.89 × 10–5 5 × 1011 136 140 35 98 Au/ MA3Bi2I9 pellet/ Au 4.6 × 10−5 2.28 × 1011 210 9.3 563 61 Au/FA3Bi2I9 SC/Au 2.4 × 10−5 7.8 × 1010 555 200 598 41 Au/Rb3Bi2I9 SC/Au 2.51 × 10−3 2.3 × 109 300 8.32 159.7 37 Au/ MTP3Bi2I9 wafer/ Au 1.88 × 10−4 – 400 19.69 7601 97 Ag/ (NH4)3Bi2I9 SC/Ag 4.0 × 10−3 – 6.5 55 803 38 250         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal detectors.108 Beyond this, there are a wide range of successful passivation strategies that have been implemented with lead halide perovskites, including using ligands with functional groups that coordinate to surface defects,114 doping with alkali halides,115 or through physisorption of O2/H2O species on the surface.116 Such strategies could act as inspiration for the devel- opment of approaches to passivate surfaces and interfaces in Bi-based materials. Furthermore, more efforts are needed to integrate these materials with ASICs to develop imagers. A simple approach is to separately grow the wafer or single crystals, and electrically integrate these with the ASIC via bonding techniques. Current approaches include wire bonding, flip-chip bonding, and iso- tropic conductive film bonding. A critical challenge is that the spatial resolution will depend on the size of the pixels used. For example, 83.2 μm sized pixels on a CMOS with CsPbBr3 integrated onto it had a spatial resolution of 5.0 lp mm−1.117 This is close to the requirement for radiography, but below the requirement for mammography (10 lp mm−1). Another challenge is that binding the X-ray attenuating medium and the ASIC can potentially damage the soft Bi-based materials, since this often involves the application of pressure or heating. Such a challenge could be overcome by directly depositing the X-ray attenuating medium onto the ASIC, for example, through thick film deposition from the vapor phase or from solution. However, in this latter approach, the processing temperature of each layer needs to be kept typically below 125°C to avoid damaging the ASIC. Beyond these rigid device applications, there are also future opportunities for f lexible and wearable devices. Here, it is particularly advantageous to have the Bi-based materials in nanocrystal form and integrated into a f lexible polymer matrix. When using X-ray detectors in portable and wearable applications, it is a significant advantage if these devices are self-powered, obviating the need for a bulky energy storage device with a limited lifetime. Self-powered operation has been reported in Bi-based materials, for example, through the for- mation of ferroelectric domains when using chiral molecules Table 2.   (continued) Device configuration μτ product (cm2 V−1) Resistivity (Ω cm) Electric field (V mm−1) LoDD (nGyair s −1) Sensitivity (µC Gyair −1 cm−2) Reference Ag/ (HIS)BiI5 SC/ Ag 2.81 × 10−4 2.31 × 1011 2.5 36.4 103 111 Au/AgBi2I7 SC/Ag 1.2 × 10−3 1.3 × 108 0.38 7261 282.5 35 Au/BiI3 SC/Au – 6.4 × 1011 – – 1.36 × 104 99 Au/BiI3 SC/Au – 3.43 × 1011 20 – 5.26 × 103 100 Ag/ PMMA polysty- rene-BiI3/Ag – – 1 5000 189 µC Gy−1 cm−3 112 Ag/BiVO4 pellet /Ag 1.15 × 10−4 3.61 × 1011 1100 20.76 3164 101 Au/ BiVO4 pellet/ Au 1.75 × 10–4 1.3 × 1012 62.3 62 241.3 103 Al/BCP/ P3HT:PC70BM:Bi2O3 nps /PEDOT:PSS/ ITO – –  ~ 500 – 1712 µC mGy−1 cm−3 102 Au/ P3HT:PCBM:Bi2O3 pellet/Au 1.7 × 10–6 – 1200 – 160 μC mGy−1 cm−3 43 Au/ BiOI SC/Au (1.1 ± 1.4) × 10−3 1.1 × 1012 2780 1.1 1100 20 Cu/BixIy/Cu 3.0 × 10−3 4.1 × 109 24 34 4.3 × 104 70 MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    251 within the structure.90 Other means of achieving self-powered operation could also be possible, for example, by having a large built-in field engineered via the device interfaces, or by making use of ion migration to form a built-in field in these devices. Finally, beyond these practical and device-related chal- lenges, it is also important to address key fundamental chal- lenges with these materials. One of the most important bar- rier is carrier localization, which has been widely found among Bi-based perovskite-inspired materials, and which severely restricts mobility-lifetime products. Although important pro- gress has been made in identifying the chemical and structural factors that allow carrier localization to be overcome, these proposed design principles need to be tested and applied to develop compounds that could achieve delocalized free carri- ers. But, as found in the case of BiOI, even if band-like trans- port is achieved, non-radiative losses can still occur as a result of electron–phonon coupling. Furthermore, defects could also induce extrinsic self-trapping, and these effects need to be understood more in these materials. Author contributions  R. L. Z. H. conceived of this review, and decided on the struc- ture with J. G. W. N. wrote the introduction, Z. A and Q. J. wrote the section on properties required for X-ray detectors. P. P. and J. L. M.-D. wrote the section on the materials properties of Bi- based compounds, while J. G. wrote the sections on the perfor- mance of Bi-based X-ray detectors (Sect. “X-ray detectors with bismuth-based materials” and “Other bismuth-based materials for X-ray detection”). R. L. Z. H. and J. G. wrote the conclusions and outlook. Funding  The authors thank the Engineering and Physical Sciences Research Council (EPSRC) and National Science Foundation (NSF) for support through an ECCS-EPSRC collaborative grant (EPSRC no. EP/Y032942/1; NSF no. ECCS-2313755). The authors also thank 5N Plus for support. R. L. Z. H. thanks the Science & Technology Facility Council (STFC) and Royal Acad- emy of Engineering (RAEng) for support through the Senior Research Fellowship scheme (no. RCSRF/2324-18-68). Declarations  Conflicts of interest  The authors declare no conflicts of interest. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Crea- tive Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Crea- tive Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​ licen​ses/​by/4.​0/. REFERENCES 1. L. Yi, B. Hou, H. Zhao, X. Liu, Nature 618, 281–286 (2023). https://​doi.​ org/​10.​1038/​s41586-​023-​05978-w 2. T. Yang, F. Li, R. Zheng, Mater. Adv. 2(21), 6744–6767 (2021). https://​doi.​ org/​10.​1039/​D1MA0​0569C 3. A. Owens, J. Synchrotron Radiat. 13, 143–150 (2006). https://​doi.​org/​10.​ 1107/​S0909​04950​50333​39 4. M.J. Yaffe, J.A. Rowlands, Phys. Med. Biol. (1997). https://​doi.​org/​10.​ 1088/​0031-​9155/​42/1/​001 5. P.A. Rodnyi, Radiat. Meas. 33, 605–614 (2001). https://​doi.​org/​10.​1016/​ S1350-​4487(01)​00068-3 6. S. Procz, G. Roque, C. Avila, J. Racedo, R. Rueda, I. Santos et al., IEEE Trans. Med. Imag. 39, 3766–3778 (2020). https://​doi.​org/​10.​1109/​TMI.​ 2020.​30046​48 7. K.R. Dudipala, T.-H. Le, W. Nie, R.L.Z. Hoye, Adv. Mater. 36, 2304523 (2024). https://​doi.​org/​10.​1002/​adma.​20230​4523 8. J. Ghosh, J. O’Neill, M.G. Masteghin, I. Braddock, C. Crean, R. Dorey et al., ACS Appl. Nano Mater. 6, 14980–14990 (2023). https://​doi.​org/​10.​1021/​ acsanm.​3c025​31 9. G.J. Yukta, M.A. Afroz, S. Alghamdi, P.J. Sellin, S. Satapathi, ACS Photon. 9, 3529–3539 (2022). https://​doi.​org/​10.​1021/​acsph​otoni​cs.​2c007​76 10. S. Choudhary, J. Ghosh, S. Pathak, S.K. Saini, N.K. Tailor, P. Sellin et al., Small 21, 2409962 (2025). https://​doi.​org/​10.​1002/​smll.​20240​9962 11. X. He, Y. Deng, D. Ouyang, N. Zhang, J. Wang, A.A. Murthy et al., Chem. Rev. 123, 1207–1261 (2023). https://​doi.​org/​10.​1021/​acs.​chemr​ev.​2c004​ 04 12. A. Glushkova, P. Andričević, R. Smajda, B. Náfrádi, M. Kollár, V. Djokić et al., ACS Nano 15, 4077–4084 (2021). https://​doi.​org/​10.​1021/​acsna​no.​ 0c079​93 13. Y. Song, L. Li, W. Bi, M. Hao, Y. Kang, A. Wang et al., Research (2020). https://​doi.​org/​10.​34133/​2020/​59582​43 14. X. Zheng, W. Zhao, P. Wang, H. Tan, M.I. Saidaminov, S. Tie et al., J. Energy Chem. 49, 299–306 (2020). https://​doi.​org/​10.​1016/j.​jechem.​2020.​02.​ 049 15. S.O. Kasap, J. Phys. D Appl. Phys. 33, 2853 (2000). https://​doi.​org/​10.​ 1088/​0022-​3727/​33/​21/​326 16. C.C. Boyd, R. Cheacharoen, T. Leijtens, M.D. McGehee, Chem. Rev. 119, 3418–3451 (2019). https://​doi.​org/​10.​1021/​acs.​chemr​ev.​8b003​36 17. A. Rogalski, Appl. Phys. Lett. (2024). https://​doi.​org/​10.​1063/5.​02280​01 18. M.H. Miah, M.U. Khandaker, M. Aminul Islam, M. Nur-E-Alam, H. Osman, M.H. Ullah, RSC Adv. 14, 6656–6698 (2024). https://​doi.​org/​10.​1039/​ D4RA0​0433G 19. S.H. Bennett, J. Ghosh, E. Gros-Daillon, F. Lédée, J. Mayén Guillén, J.-M. Verilhac et al., Front. Detect. Sci. Technol. (2023). https://​doi.​org/​10.​ 3389/​fdest.​2023.​12498​92 20. R.A. Jagt, I. Bravić, L. Eyre, K. Gałkowski, J. Borowiec, K.R. Dudi- pala et al., Nat. Commun. 14, 2452 (2023). https://​doi.​org/​10.​1038/​ s41467-​023-​38008-4 21. Z. Li, F. Zhou, H. Yao, Z. Ci, Z. Yang, Z. Jin, Mater. Today 48, 155–175 (2021). https://​doi.​org/​10.​1016/j.​mattod.​2021.​01.​028 22. Y. Liu, Y. Zhang, X. Zhu, J. Feng, I. Spanopoulos, W. Ke et al., Adv. Mater. 33, 2006010 (2021). https://​doi.​org/​10.​1002/​adma.​20200​6010 http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1038/s41586-023-05978-w https://doi.org/10.1038/s41586-023-05978-w https://doi.org/10.1039/D1MA00569C https://doi.org/10.1039/D1MA00569C https://doi.org/10.1107/S0909049505033339 https://doi.org/10.1107/S0909049505033339 https://doi.org/10.1088/0031-9155/42/1/001 https://doi.org/10.1088/0031-9155/42/1/001 https://doi.org/10.1016/S1350-4487(01)00068-3 https://doi.org/10.1016/S1350-4487(01)00068-3 https://doi.org/10.1109/TMI.2020.3004648 https://doi.org/10.1109/TMI.2020.3004648 https://doi.org/10.1002/adma.202304523 https://doi.org/10.1021/acsanm.3c02531 https://doi.org/10.1021/acsanm.3c02531 https://doi.org/10.1021/acsphotonics.2c00776 https://doi.org/10.1002/smll.202409962 https://doi.org/10.1021/acs.chemrev.2c00404 https://doi.org/10.1021/acs.chemrev.2c00404 https://doi.org/10.1021/acsnano.0c07993 https://doi.org/10.1021/acsnano.0c07993 https://doi.org/10.34133/2020/5958243 https://doi.org/10.1016/j.jechem.2020.02.049 https://doi.org/10.1016/j.jechem.2020.02.049 https://doi.org/10.1088/0022-3727/33/21/326 https://doi.org/10.1088/0022-3727/33/21/326 https://doi.org/10.1021/acs.chemrev.8b00336 https://doi.org/10.1063/5.0228001 https://doi.org/10.1039/D4RA00433G https://doi.org/10.1039/D4RA00433G https://doi.org/10.3389/fdest.2023.1249892 https://doi.org/10.3389/fdest.2023.1249892 https://doi.org/10.1038/s41467-023-38008-4 https://doi.org/10.1038/s41467-023-38008-4 https://doi.org/10.1016/j.mattod.2021.01.028 https://doi.org/10.1002/adma.202006010 252         MRS ENERGY & SUSTAINABILITY  //  VOLUME  12  //  www.mrs.org/energy-sustainability-journal 23. J. Ghosh, S. Parveen, P.J. Sellin, P.K. Giri, Adv. Mater. Technol. 8, 2300400 (2023). https://​doi.​org/​10.​1002/​admt.​20230​0400 24. L. Pan, S. Shrestha, N. Taylor, W. Nie, L.R. Cao, Nat. Commun. 12, 5258 (2021). https://​doi.​org/​10.​1038/​s41467-​021-​25648-7 25. J. Ghosh, N.K. Tailor, K. Ghosh, J. Jayarathne, S. Choudhary, C. Crean et al., Adv. Opt. Mater 13, 2402497 (2025). https://​doi.​org/​10.​1002/​ adom.​20240​2497 26. R.L.Z. Hoye, J. Hidalgo, R.A. Jagt, J.-P. Correa-Baena, T. Fix, J.L. Mac- Manus-Driscoll, Adv. Energy Mater. 12, 2100499 (2022). https://​doi.​org/​ 10.​1002/​aenm.​20210​0499 27. L. Yao, G. Niu, L. Yin, X. Du, Y. Lin, X. Den et al., J. Mater. Chem. C 8, 1239–1243 (2020). https://​doi.​org/​10.​1039/​C9TC0​6313G 28. W. Pan, H. Wu, J. Luo, Z. Deng, C. Ge, C. Chen et al., Nat. Photon. 11, 726–732 (2017). https://​doi.​org/​10.​1038/​s41566-​017-​0012-4 29. H. Lei, D. Hardy, F. Gao, Adv. Funct. Mater. 31, 2105898 (2021). https://​ doi.​org/​10.​1002/​adfm.​20210​5898 30. Z. Xu, X. Liu, Y. Li, X. Liu, T. Yang, C. Ji et al., Angew. Chem. Int. Ed. 58, 15757–15761 (2019). https://​doi.​org/​10.​1002/​anie.​20190​9815 31. C.-F. Wang, H. Li, M.-G. Li, Y. Cui, X. Song, Q.-W. Wang et al., Adv. Funct. Mater. 31, 2009457 (2021). https://​doi.​org/​10.​1002/​adfm.​20200​9457 32. W. Guo, X. Liu, S. Han, Y. Liu, Z. Xu, M. Hong et al., Angew. Chem. Int. Ed. 59, 13879–13884 (2020) 33. P. Liu, Y. Xiao, Z. Yang, S. Yu, X. Meng, Opt. Mater. 133, 112972 (2022). https://​doi.​org/​10.​1016/j.​optmat.​2022.​112972 34. Berger M. http://​www.​nist. gov/pml/data/xcom/index cfm, (2010). https://​doi.​org/​10.​18434/​T48G6X. 35. S. Tie, W. Zhao, W. Huang, D. Xin, M. Zhang, Z. Yang et al., J. Phys. Chem. Lett. 11, 7939–7945 (2020). https://​doi.​org/​10.​1021/​acs.​jpcle​ tt.​0c023​43 36. Y. Zhang, Y. Liu, Z. Xu, H. Ye, Z. Yang, J. You et al., Nat. Commun. 11, 2304 (2020). https://​doi.​org/​10.​1038/​s41467-​020-​16034-w 37. M. Xia, J.-H. Yuan, G. Niu, X. Du, L. Yin, W. Pan et al., Adv. Funct. Mater. 30, 1910648 (2020). https://​doi.​org/​10.​1002/​adfm.​20191​0648 38. R. Zhuang, X. Wang, W. Ma, Y. Wu, X. Chen, L. Tang et al., Nat. Photon. 13, 602–608 (2019). https://​doi.​org/​10.​1038/​s41566-​019-​0466-7 39. K. Tao, Y. Li, C. Ji, X. Liu, Z. Wu, S. Han et al., Chem. Mater. 31, 5927–5932 (2019). https://​doi.​org/​10.​1021/​acs.​chemm​ater.​9b022​63 40. Y. Liu, Y. Zhang, Z. Yang, J. Cui, H. Wu, X. Ren et al., Adv. Opt. Mater. 8, 2000814 (2020). https://​doi.​org/​10.​1002/​adom.​20200​0814 41. W. Li, D. Xin, S. Tie, J. Ren, S. Dong, L. Lei et al., J. Phys. Chem. Lett. 12, 1778–1785 (2021). https://​doi.​org/​10.​1021/​acs.​jpcle​tt.​1c000​90 42. P.M. Johns, J.E. Baciak, J.C. Nino, Appl. Phys. Lett. (2016). https://​doi.​ org/​10.​1063/1.​49622​93 43. K.D.G.I. Jayawardena, H.M. Thirimanne, S.F. Tedde, J.E. Huerdler, A.J. Parnell, R.M.I. Bandara et al., ACS Nano 13, 6973–6981 (2019). https://​ doi.​org/​10.​1021/​acsna​no.​9b019​16 44. J.A. Steele, W. Pan, C. Martin, M. Keshavarz, E. Debroye, H. Yuan et al., Adv. Mater. 30, 1804450 (2018). https://​doi.​org/​10.​1002/​adma.​20180​4450 45. W. Yuan, G. Niu, Y. Xian, H. Wu, H. Wang, H. Yin et al., Adv. Funct. Mater. 29, 1900234 (2019). https://​doi.​org/​10.​1002/​adfm.​20190​0234 46. L. Yin, H. Wu, W. Pan, B. Yang, P. Li, J. Luo et al., Adv. Opt. Mater. 7, 1900491 (2019). https://​doi.​org/​10.​1002/​adom.​20190​0491 47. D. Valli, H. Zhang, M. Betušiak, G. Romolini, A. Meulemans, D. Escudero et al., Adv. Opt. Mater. 2, 2075–2084 (2024). https://​doi.​org/​10.​1021/​ acsaom.​4c002​65 48. S.R. Rondiya, R.A. Jagt, J.L. MacManus-Driscoll, A. Walsh, R.L.Z. Hoye, Appl. Phys. Lett. (2021). https://​doi.​org/​10.​1063/5.​00717​63 49. B. Wu, W. Ning, Q. Xu, M. Manjappa, M. Feng, S. Ye et al., Sci. Adv. 7, eabd3160 (2021). https://​doi.​org/​10.​1126/​sciadv.​abd31​60 50. A.D. Wright, L.R.V. Buizza, K.J. Savill, G. Longo, H.J. Snaith, M.B. John- ston et al., J. Phys. Chem. Lett. 12, 3352–3360 (2021). https://​doi.​org/​10.​ 1021/​acs.​jpcle​tt.​1c006​53 51. A.H. Slavney, T. Hu, A.M. Lindenberg, H.I. Karunadasa, J. Am. Chem. Soc. 138, 2138–2141 (2016). https://​doi.​org/​10.​1021/​jacs.​5b132​94 52. J.A. Steele, P. Puech, M. Keshavarz, R. Yang, S. Banerjee, E. Debroye et al., ACS Nano 12, 8081–8090 (2018). https://​doi.​org/​10.​1021/​acsna​no.​8b029​ 36 53. L. Zhang, Y. Fang, L. Sui, J. Yan, K. Wang, K. Yuan et al., ACS Energy Lett. 4, 2975–2982 (2019). https://​doi.​org/​10.​1021/​acsen​ergyl​ett.​9b021​55 54. Y. Fu, H. Lohan, M. Righetto, Y.-T. Huang, S.R. Kavanagh, C.-W. Cho et  al., Nat. Commun. 16, 65 (2025). https://​doi.​org/​10.​1038/​ s41467-​024-​55254-2 55. F. Cao, X. Xu, D. Yu, H. Zeng, Nanophotonics (2020). https://​doi.​org/​10.​ 1515/​nanoph-​2020-​0632 56. F. Zhou, Z. Li, W. Lan, Q. Wang, L. Ding, Z. Jin, Small Methods 4, 2000506 (2020). https://​doi.​org/​10.​1002/​smtd.​20200​0506 57. N.K. Tailor, J. Ghosh, N. Singh, P.K. Nayak, J. Hall, D. Ghosh et al., ACS Appl. Mater. Interfaces (2025). https://​doi.​org/​10.​1021/​acsami.​5c022​22 58. Y. Shen, C. Ran, X. Dong, Z. Wu, W. Huang, Small 20, 2308242 (2024). https://​doi.​org/​10.​1002/​smll.​20230​8242 59. J. Ghosh, P.J. Sellin, P.K. Giri, Nanotechnology (2022). https://​doi.​org/​ 10.​1088/​1361-​6528/​ac6884 60. W. Liang, Z. Shi, Y. Li, J. Ma, S. Yin, X. Chen et al., ACS Appl. Mater. Inter- faces 12, 37363–37374 (2020). https://​doi.​org/​10.​1021/​acsami.​0c103​23 61. S. Tie, W. Zhao, D. Xin, M. Zhang, J. Long, Q. Chen et al., Adv. Mater. 32, 2001981 (2020). https://​doi.​org/​10.​1002/​adma.​20200​1981 62. W. Ning, F. Gao, Adv. Mater. 31, 1900326 (2019). https://​doi.​org/​10.​1002/​ adma.​20190​0326 63. X. Zhang, T. Zhu, C. Ji, Y. Yao, J. Luo, J. Am. Chem. Soc. 143, 20802–20810 (2021). https://​doi.​org/​10.​1021/​jacs.​1c089​59 64. I. Mosquera-Lois, Y.-T. Huang, H. Lohan, J. Ye, A. Walsh, R.L.Z. Hoye, Nat. Rev. Chem. 9, 287–304 (2025). https://​doi.​org/​10.​1038/​ s41570-​025-​00702-w 65. R.E. Brandt, V. Stevanović, D.S. Ginley, T. Buonassisi, MRS Commun. 5, 265–275 (2015). https://​doi.​org/​10.​1557/​mrc.​2015.​26 66. H. Zhu, I. Turkevych, H. Lohan, P. Liu, R.W. Martin, F.C.P. Massabuau et al., Int. Mater. Rev. 69, 19–62 (2024). https://​doi.​org/​10.​1177/​09506​ 60823​12130​65 67. N.D. Yuriy, R.B. Paul, J.C. Leonard, B.K. Mikhail, S.S. Kanai, Proc. SPIE 3768, 521–529 (1999). https://​doi.​org/​10.​1117/​12.​366625 68. R. Chaudhari, C. RaviKant, Sens. Actuators, A 346, 113863 (2022). https://​ doi.​org/​10.​1016/j.​sna.​2022.​113863 69. M. Matsumoto, K. Hitomi, T. Shoji, Y. Hiratate, IEEE Trans. Nucl. Sci. 49, 2517–2520 (2002). https://​doi.​org/​10.​1109/​TNS.​2002.​803883 70. R. Zhuang, S. Cai, Z. Mei, H. Liang, N. Zhao, H. Mu et al., Nat. Commun. 14, 1621 (2023). https://​doi.​org/​10.​1038/​s41467-​023-​37297-z 71. N. Ali, K. Shehzad, S. Attique, A. Ali, F. Akram, A. Younis et al., Small 20, 2310946 (2024). https://​doi.​org/​10.​1002/​smll.​20231​0946 72. K.R. Dudipala, T.-H. Le, W. Nie, R.L.Z. Hoye, Adv. Mater. 36, 2304523 (2024). https://​doi.​org/​10.​1002/​adma.​20230​4523 73. W. Ma, L. Liu, H. Qin, R. Gao, B. He, S. Gou et  al., Sensors 23, 2017 (2023). https://​doi.​org/​10.​3390/​s2304​2017 74. X. Xu, Y. Wu, Y. Zhang, X. Li, F. Wang, X. Jiang et al., Energy Environ Mater 7, e12487 (2024). https://​doi.​org/​10.​1002/​eem2.​12487 75. D. Shi, V. Adinolfi, R. Comin, M. Yuan, E. Alarousu, A. Buin et al., Science 347, 519–522 (2015). https://​doi.​org/​10.​1126/​scien​ce.​aaa27​25 76. Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao et al., Science 347, 967–970 (2015). https://​doi.​org/​10.​1126/​scien​ce.​aaa57​60 77. H. Zhang, Z. Gao, R. Liang, X. Zheng, X. Geng, Y. Zhao et al., IEEE Trans. Electron Devices 66, 2224–2229 (2019). https://​doi.​org/​10.​1109/​TED.​ 2019.​29035​37 78. D. Cao, G. Yang, Mater. Today Commun. 33, 104242 (2022). https://​doi.​ org/​10.​1016/j.​mtcomm.​2022.​104242 79. N. Setter, L.E. Cross, J. Mater. Sci. 15, 2478–2482 (1980). https://​doi.​org/​ 10.​1007/​BF005​50750 80. S. Alshogeathri, D. Cao, D. Kim, G. Yang, Front. Phys. (2023). https://​doi.​ org/​10.​3389/​fphy.​2023.​11293​01 81. W. Chen, H. Sun, Y. Jin, H. Yang, Y. He, X. Zhu, J. Mater. Sci. Mater. Elec- tron. 34, 496 (2023). https://​doi.​org/​10.​1007/​s10854-​023-​09897-4 82. Y. He, M. Petryk, Z. Liu, D.G. Chica, I. Hadar, C. Leak et al., Nat. Photon. 15, 36–42 (2021). https://​doi.​org/​10.​1038/​s41566-​020-​00727-1 83. X. Li, P. Zhang, Y. Hua, F. Cui, X. Sun, L. Liu et al., ACS Appl. Mater. Inter- faces 14, 9340–9351 (2022). https://​doi.​org/​10.​1021/​acsami.​1c240​86 https://doi.org/10.1002/admt.202300400 https://doi.org/10.1038/s41467-021-25648-7 https://doi.org/10.1002/adom.202402497 https://doi.org/10.1002/adom.202402497 https://doi.org/10.1002/aenm.202100499 https://doi.org/10.1002/aenm.202100499 https://doi.org/10.1039/C9TC06313G https://doi.org/10.1038/s41566-017-0012-4 https://doi.org/10.1002/adfm.202105898 https://doi.org/10.1002/adfm.202105898 https://doi.org/10.1002/anie.201909815 https://doi.org/10.1002/adfm.202009457 https://doi.org/10.1016/j.optmat.2022.112972 http://www.nist https://doi.org/10.18434/T48G6X https://doi.org/10.1021/acs.jpclett.0c02343 https://doi.org/10.1021/acs.jpclett.0c02343 https://doi.org/10.1038/s41467-020-16034-w https://doi.org/10.1002/adfm.201910648 https://doi.org/10.1038/s41566-019-0466-7 https://doi.org/10.1021/acs.chemmater.9b02263 https://doi.org/10.1002/adom.202000814 https://doi.org/10.1021/acs.jpclett.1c00090 https://doi.org/10.1063/1.4962293 https://doi.org/10.1063/1.4962293 https://doi.org/10.1021/acsnano.9b01916 https://doi.org/10.1021/acsnano.9b01916 https://doi.org/10.1002/adma.201804450 https://doi.org/10.1002/adfm.201900234 https://doi.org/10.1002/adom.201900491 https://doi.org/10.1021/acsaom.4c00265 https://doi.org/10.1021/acsaom.4c00265 https://doi.org/10.1063/5.0071763 https://doi.org/10.1126/sciadv.abd3160 https://doi.org/10.1021/acs.jpclett.1c00653 https://doi.org/10.1021/acs.jpclett.1c00653 https://doi.org/10.1021/jacs.5b13294 https://doi.org/10.1021/acsnano.8b02936 https://doi.org/10.1021/acsnano.8b02936 https://doi.org/10.1021/acsenergylett.9b02155 https://doi.org/10.1038/s41467-024-55254-2 https://doi.org/10.1038/s41467-024-55254-2 https://doi.org/10.1515/nanoph-2020-0632 https://doi.org/10.1515/nanoph-2020-0632 https://doi.org/10.1002/smtd.202000506 https://doi.org/10.1021/acsami.5c02222 https://doi.org/10.1002/smll.202308242 https://doi.org/10.1088/1361-6528/ac6884 https://doi.org/10.1088/1361-6528/ac6884 https://doi.org/10.1021/acsami.0c10323 https://doi.org/10.1002/adma.202001981 https://doi.org/10.1002/adma.201900326 https://doi.org/10.1002/adma.201900326 https://doi.org/10.1021/jacs.1c08959 https://doi.org/10.1038/s41570-025-00702-w https://doi.org/10.1038/s41570-025-00702-w https://doi.org/10.1557/mrc.2015.26 https://doi.org/10.1177/09506608231213065 https://doi.org/10.1177/09506608231213065 https://doi.org/10.1117/12.366625 https://doi.org/10.1016/j.sna.2022.113863 https://doi.org/10.1016/j.sna.2022.113863 https://doi.org/10.1109/TNS.2002.803883 https://doi.org/10.1038/s41467-023-37297-z https://doi.org/10.1002/smll.202310946 https://doi.org/10.1002/adma.202304523 https://doi.org/10.3390/s23042017 https://doi.org/10.1002/eem2.12487 https://doi.org/10.1126/science.aaa2725 https://doi.org/10.1126/science.aaa5760 https://doi.org/10.1109/TED.2019.2903537 https://doi.org/10.1109/TED.2019.2903537 https://doi.org/10.1016/j.mtcomm.2022.104242 https://doi.org/10.1016/j.mtcomm.2022.104242 https://doi.org/10.1007/BF00550750 https://doi.org/10.1007/BF00550750 https://doi.org/10.3389/fphy.2023.1129301 https://doi.org/10.3389/fphy.2023.1129301 https://doi.org/10.1007/s10854-023-09897-4 https://doi.org/10.1038/s41566-020-00727-1 https://doi.org/10.1021/acsami.1c24086 MRS ENERGY & SUSTAINABILITY  //  VOLUME 12  //  www.mrs.org/energy-sustainability-journal                    253 84. X. Li, X. Du, P. Zhang, Y. Hua, L. Liu, G. Niu et al., Sci. China Mater. 64, 1427–1436 (2021). https://​doi.​org/​10.​1007/​s40843-​020-​1553-8 85. H. Wei, J. Huang, Nat. Commun. 10, 1066 (2019). https://​doi.​org/​10.​ 1038/​s41467-​019-​08981-w 86. Z. Zheng, H. Li, L. Hai, R. Ma, R. Liu, C. Zhai et al., Adv. Funct. Mater. 34, 2307093 (2024). https://​doi.​org/​10.​1002/​adfm.​20230​7093 87. G. Chen, H. Dai, Z.-K. Zhu, J. Wu, P. Yu, Y. Zeng et al., Small 20, 2312281 (2024). https://​doi.​org/​10.​1002/​smll.​20231​2281 88. H. Chen, Z. Li, S. Wang, G. Peng, W. Lan, H. Wang et al., Adv. Mater. 36, 2308872 (2024). https://​doi.​org/​10.​1002/​adma.​20230​8872 89. J. Li, L. Li, L. Wang, L. Tao, D. Yang, Y. Fang, J. Mater. Sci. Mater. Electron. 35, 488 (2024). https://​doi.​org/​10.​1007/​s10854-​024-​12279-z 90. J. Wu, S. You, P. Yu, Q. Guan, Z.-K. Zhu, Z. Li et al., ACS Energy Lett. 8, 2809–2816 (2023). https://​doi.​org/​10.​1021/​acsen​ergyl​ett.​3c006​29 91. A. Wang, C. Zhang, Q. Guan, H. Ye, R. Li, H. Li et al., Small 21, 2407843 (2025). https://​doi.​org/​10.​1002/​smll.​20240​7843 92. G. Chen, Z.-K. Zhu, J. Wu, P. Yu, Y. Zeng, H. Dai et al., ACS Appl. Mater. Interfaces 16, 67970–67978 (2024). https://​doi.​org/​10.​1021/​acsami.​ 4c149​63 93. Y. Lei, Y. Li, G. Peng, Y. Xu, H. Wang, H. Fan et al., IEEE Electron Device Lett. 45, 1441–1444 (2024). https://​doi.​org/​10.​1109/​LED.​2024.​34174​37 94. Z. Xu, H. Wu, D. Li, W. Wu, L. Li, J. Luo, J. Mater. Chem. C 9, 13157–13161 (2021). https://​doi.​org/​10.​1039/​D1TC0​3412J 95. M. Li, H. Li, W. Li, B. Li, T. Lu, X. Feng et al., Adv. Mater. 34, 2108020 (2022). https://​doi.​org/​10.​1002/​adma.​20210​8020 96. Y. Liu, Z. Xu, Z. Yang, Y. Zhang, J. Cui, Y. He et al., Matter 3, 180–196 (2020). https://​doi.​org/​10.​1016/j.​matt.​2020.​04.​017 97. Y. Xu, Z. Li, C. Shi, Y. Li, Y. Lei, G. Peng et al., Adv. Mater. 36, 2406128 (2024). https://​doi.​org/​10.​1002/​adma.​20240​6128 98. D. Xin, S. Dong, M. Zhang, S. Tie, J. Ren, L. Lei et al., J. Phys. Chem. Lett. 13, 371–377 (2022). https://​doi.​org/​10.​1021/​acs.​jpcle​tt.​1c039​22 99. H. Sun, X. Zhu, P. Wangyang, X. Gao, S. Zhu, B. Zhao, J. Mater. Sci. Mater. Electron. 29, 20003–20009 (2018). https://​doi.​org/​10.​1007/​ s10854-​018-​0130-x 100. Y. Liu, H. Sun, D. Yang, P. Wangyang, X. Gao, Z. Gou et al., Mater. Res. Express 6, 055902 (2019). https://​doi.​org/​10.​1088/​2053-​1591/​aaff87 101. Z. Fan, L. Lei, S. Tie, S. Dong, R. Yuan, B. Zhou et al., Small 20, 2401213 (2024). https://​doi.​org/​10.​1002/​smll.​20240​1213 102. H.M. Thirimanne, K.D.G.I. Jayawardena, A.J. Parnell, R.M.I. Bandara, A. Karalasingam, S. Pani et al., Nat. Commun. 9, 2926 (2018). https://​doi.​ org/​10.​1038/​s41467-​018-​05301-6 103. S. Jia, Y. Xiao, N. Bu, N. Li, D. Li, Z. Yang et al., Adv. Funct. Mater. 33, 2213563 (2023). https://​doi.​org/​10.​1002/​adfm.​20221​3563 104. L. Mao, Y. Li, H. Chen, L. Yu, J. Zhang, Nanomaterials 11, 1832 (2021). https://​doi.​org/​10.​3390/​nano1​10718​32 105. P. Praveenkumar, G.D. Venkatasubbu, P. Thangadurai, T. Prakash, Mat. Sci. Semicond. Process 104, 104686 (2019). https://​doi.​org/​10.​1016/j.​mssp.​ 2019.​104686 106. L. Chen, H. Wang, W. Zhang, F. Li, Z. Wang, X. Wang et al., ACS Appl. Mater. Interfaces 14, 10917–10926 (2022). https://​doi.​org/​10.​1021/​ acsami.​1c219​48 107. N.K. Tailor, J. Ghosh, M.A. Afroz, S. Bennett, M. Chatterjee, P. Sellin et al., ACS Appl. Electron. Mater. 4, 4530–4539 (2022). https://​doi.​org/​10.​1021/​ acsae​lm.​2c007​52 108. B. Yang, W. Pan, H. Wu, G. Niu, J.-H. Yuan, K.-H. Xue et al., Nat. Commun. 10, 1989 (2019). https://​doi.​org/​10.​1038/​s41467-​019-​09968-3 109. H. Zhang, G. Dun, Q. Feng, R. Zhao, R. Liang, Z. Gao et al., IEEE Trans. Electron Devices 67, 3191–3198 (2020). https://​doi.​org/​10.​1109/​TED.​ 2020.​29987​63 110. H. Li, X. Shan, J.N. Neu, T. Geske, M. Davis, P. Mao et al., J. Mater. Chem. C 6, 11961–11967 (2018). https://​doi.​org/​10.​1039/​C8TC0​1564C 111. Z. Zhao, Q. Fan, Y. Liu, H. Rong, H. Ni, L. Wei et al., ACS Appl. Mater. Interfaces 16, 38283–38289 (2024). https://​doi.​org/​10.​1021/​acsami.​ 4c086​48 112. R. Chaudhari, C.R. Kant, A. Garg, MRS Commun. 12, 358–364 (2022). https://​doi.​org/​10.​1557/​s43579-​022-​00185-6 113. Z. Li, S.P. Senanayak, L. Dai, G. Kusch, R. Shivanna, Y. Zhang et al., Adv. Funct. Mater. 31, 2104981 (2021).  https://​doi.​org/​10.​1002/​adfm.​20210​ 4981 114. N. Fiuza-Maneiro, J. Ye, S.K. Sharma, S. Chakraborty, S. Gómez-Graña, R.L.Z. Hoye et al., ACS Energy Lett. 10, 1623–1632 (2025). https://​doi.​ org/​10.​1021/​acsen​ergyl​ett.​5c001​85 115. H. Zhang, L. Pfeifer, S.M. Zakeeruddin, J. Chu, M. Grätzel, Nat. Rev. Chem. 7, 632–652 (2023). https://​doi.​org/​10.​1038/​s41570-​023-​00510-0 116. Y. Yang, M. Yang, D.T. Moore, Y. Yan, E.M. Miller, K. Zhu et al., Nat. Energy 2, 16207 (2017). https://​doi.​org/​10.​1038/​nener​gy.​2016.​207 117. Y. Liu, C. Gao, D. Li, X. Zhang, J. Zhu, M. Wu et al., Nat. Commun. 15, 1588 (2024). https://​doi.​org/​10.​1038/​s41467-​024-​45871-2 Publisher’s Note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. https://doi.org/10.1007/s40843-020-1553-8 https://doi.org/10.1038/s41467-019-08981-w https://doi.org/10.1038/s41467-019-08981-w https://doi.org/10.1002/adfm.202307093 https://doi.org/10.1002/smll.202312281 https://doi.org/10.1002/adma.202308872 https://doi.org/10.1007/s10854-024-12279-z https://doi.org/10.1021/acsenergylett.3c00629 https://doi.org/10.1002/smll.202407843 https://doi.org/10.1021/acsami.4c14963 https://doi.org/10.1021/acsami.4c14963 https://doi.org/10.1109/LED.2024.3417437 https://doi.org/10.1039/D1TC03412J https://doi.org/10.1002/adma.202108020 https://doi.org/10.1016/j.matt.2020.04.017 https://doi.org/10.1002/adma.202406128 https://doi.org/10.1021/acs.jpclett.1c03922 https://doi.org/10.1007/s10854-018-0130-x https://doi.org/10.1007/s10854-018-0130-x https://doi.org/10.1088/2053-1591/aaff87 https://doi.org/10.1002/smll.202401213 https://doi.org/10.1038/s41467-018-05301-6 https://doi.org/10.1038/s41467-018-05301-6 https://doi.org/10.1002/adfm.202213563 https://doi.org/10.3390/nano11071832 https://doi.org/10.1016/j.mssp.2019.104686 https://doi.org/10.1016/j.mssp.2019.104686 https://doi.org/10.1021/acsami.1c21948 https://doi.org/10.1021/acsami.1c21948 https://doi.org/10.1021/acsaelm.2c00752 https://doi.org/10.1021/acsaelm.2c00752 https://doi.org/10.1038/s41467-019-09968-3 https://doi.org/10.1109/TED.2020.2998763 https://doi.org/10.1109/TED.2020.2998763 https://doi.org/10.1039/C8TC01564C https://doi.org/10.1021/acsami.4c08648 https://doi.org/10.1021/acsami.4c08648 https://doi.org/10.1557/s43579-022-00185-6 https://doi.org/10.1002/adfm.202104981 https://doi.org/10.1002/adfm.202104981 https://doi.org/10.1021/acsenergylett.5c00185 https://doi.org/10.1021/acsenergylett.5c00185 https://doi.org/10.1038/s41570-023-00510-0 https://doi.org/10.1038/nenergy.2016.207 https://doi.org/10.1038/s41467-024-45871-2 Abstract Anchor 2 Discussion Introduction Key properties of X-ray detector materials Radiation attenuation Charge collection efficiency Sensitivity Signal-to-noise ratio Limit of detection Key properties of bismuth-based materials for X-ray detection Bismuth-based double perovskites Bismuth-based perovskite derivatives Other bismuth-based materials Overall comparison X-ray detectors with bismuth-based materials Bismuth-based double perovskites Bismuth-based 2D perovskites Bismuth-based 0D perovskites Other bismuth-based materials for X-ray detection Conclusions and outlook References Publisher’s Note