English

• X-rays are extensively utilized in security inspections, scientific research, and medical imaging due to their high penetration ability, short wavelength and high energy [1–6]. In principle, direct X-ray detection works by absorbing X-ray photons and converting them into electrical signals [7]. Based on the fundamental concept of X-ray energy conversion, the optimal detector provides the following key characteristics: a large atomic number (Z) with superior absorption of radiation; large carrier mobility, long charge lifetimes and low defect density to obtain effective carrier collection [8]. Currently, the conventional detection materials mainly include amorphous selenium (α-Se) [9], PbI₂ [10], CdTe [11], and Si [12], but these materials usually show inferior performance. For instance, the α-Se is restricted by its inadequate mobility-lifetime (μτ) product (approximately 10⁻⁷ cm²/V), resulting in poor sensitivity and detection limits [13]. In this situation, it is vitally desirable to explore new members of lead-free perovskites with low detection limits for high-sensitive X-ray detection.

  In recent years, organic-inorganic hybrid perovskites have been proved to be ideal photodetection materials owing to their high carrier mobility, prolonged carrier lifetime, and low-cost single-crystal growth [14–18]. Among them, lead-based halide perovskites contain a high atomic number and thus exhibit outstanding X-ray absorption coefficients [19–23]. For example, it has been reported that sensitive detectors based on MAPbI₃ (MA⁺ = CH₃NH₃⁺) single crystal (SC) and MAPbBr₃ SC attain a sensitivity of 2.1 × 10⁴ µC Gy_air⁻¹ cm⁻² [2,24], which is substantially greater than that of conventional α-Se X-ray detector [25]. However, with instability and hazardous Pb element, the potential of the above-mentioned compounds is significantly constrained [26]. In this situation, it is urgently desirable to explore lead-free hybrid perovskites as green X-ray detection materials. Bismuth, one of the nontoxic elements, has a large atomic number that will result in strong X-ray absorption [27], making it ideal for capturing charge carriers produced by X-rays [28]. The recently discovered one-dimensional (1D) (H₂MDAP)BiI₅ SC obtained a moderate sensitivity of 1 µC Gy_air⁻¹ cm⁻² at 5 V/mm [29], which provides the groundwork for improving the performance and demonstrates the potential for X-ray inspection. The direct X-ray detection was achieved in Bi-based perovskite (DMEDA)BiI₅ SC and exhibited a promising sensitivity of 72.5 µC Gy_air⁻¹ cm⁻² [30]. However, it needs to operate at a large field strength (494 V/mm), and its sensitivity and detection capability are insufficient [31]. Thus, it is critical to develop X-ray detectors with great sensitivity to detect faint X-ray signals at low operating voltage.

  In this research, we synthesized a non-lead compound (R)-(H₂MPz)BiI₅ (R-MPz = (R)-(-)-2-methylpiperazine, R-MBI) and demonstrated its suitability as an X-ray detector. The single crystal of R-MBI possessed a large bulk resistivity of 9.62 × 10¹⁰ Ω cm, and a high μτ product of 1.88 × 10⁻⁴ cm²/V. Consequently, planar-structured crystal devices based on R-MBI SC are capable of excellent X-ray detection, producing an exceptional sensitivity of 263.58 µC Gy_air⁻¹ cm⁻² at 50 V/mm and detection limit less than 4.35 µGy_air/s. Our work demonstrates the R-MBI is a viable candidate for direct X-ray detection and inspires research into the next generation of eco-friendly detectors.

  A centimeter-sized single crystal of 5 × 10 × 1.5 mm³ (Fig. 1a) was grown via the low-temperature solution method. The powder X-ray diffraction (XRD) pattern of the R-MBI matches the simulated results quite well, implying the purity of crystals (Fig. S1 in Supporting information). Meanwhile, R-MBI exhibits no weight loss until 595 K and no significant change in the XRD pattern after 30 days, indicating favorable thermostability and long-term air stability (relative humidity, 70%) (Figs. S2 and S3 in Supporting information). Structural analysis indicates that R-MBI crystallizes in the orthorhombic (P2₁2₁2₁) space group (Table S1 in Supporting information), which forms a 1D zigzag chain along the a-axis with the BiI₆ octahedrons interconnected by corner-sharing units (Fig. 1b). In addition, the BiI₆ octahedra have a slight deformation, as shown by the non-uniformity of the Bi-I bond length and I-Bi-I angle. The Bi-I bond distance is in the range of 2.946 Å to 3.215 Å, while the I-Bi-I angle specifically falls between 83.80° and 176.58° (Fig. S4 and Table S2 in Supporting information). It is notable that organic component (R)-(-)−2-methylpiperazine cations are bonded to the 1D infinite zigzag chain via N–H···I hydrogen bonds (Fig. S5 in Supporting information).

  Figure 1

  

  Figure 1.  (a) The simulated morphology based on the R-MBI SC (left) and solution-grown bulk crystal of R-MBI (right). (b) The crystal structure of R-MBI.

  DownLoad: Full-Size Img PowerPoint

  The ultraviolet-visible diffuse reflectance spectra of R-MBI in the 450–900 nm region were carried out to evaluate the light-absorbing properties (Fig. 2a). The R-MBI exhibits an absorption cutoff wavelength of approximately 665 nm and the band gap is inferred with a value of 1.92 eV according to the Tauc curve (inset of Fig. 2a). Moreover, this band gap is relatively lower than that of several bismuth-based perovskites, including (DFPIP)₄AgBiI₈ (~2.03 eV) and (NH₄)₃Bi₂I₉ (~2.04 eV) [32,33]. As illustrated in Fig. 2b, the calculated band gap of R-MBI is 1.98 eV with the valence band maximum at point Y and the bottom of the conduction band at point U, indicating its indirect feature. Besides, the partial density of states spectra prove the R-MBI energy band gap is determined by the BiI₆ octahedra of the inorganic 1D chain (Fig. S6 in Supporting information): The conduction band minimum is predominantly assigned to the Bi-6p orbital, while the valence band maximum is principally attributable to the I-5p orbital.

  Figure 2

  

  Figure 2.  (a) The optical absorption spectrum of R-MBI with the insert of corresponding Tauc plot. (b) The calculated energy band structure of the R-MBI.

  DownLoad: Full-Size Img PowerPoint

  The resistivity of R-MBI was calculated up to 9.62 × 10¹⁰ Ω cm (Fig. 3a), which is substantially higher than other lead-based perovskites (10⁷–10⁸ Ω cm) [2,19]. This high resistivity is advantageous in terms of lowering the dark current, resulting in a higher signal-to-noise ratio (SNR) [34]. Another significant parameter to evaluate the performance of semiconductors is trap density (n_trap). Thus, we measured the dark current-voltage (Ⅰ-Ⅴ) curve to investigate the n_trap of R-MBI by the space-charge-limited current method. The ohmic zone is denoted by the first smooth line in Fig. 3b, where the current signal grows linearly with voltage. As seen by the second steep line, the trap-filled area occurs when injected carriers fill the trap states. After that, we calculated the applied voltage at the intersection as the trap-filled limit voltage (V_TFL), which came out to be 13.3 V. The n_trap can also be determined from the following formula [35]:

  Figure 3

  

  Figure 3.  (a) Bulk resistivity of R-MBI. (b) The Ⅰ-Ⅴ curve of R-MBI using the SCLC analysis. (c) Linear attenuation coefficient of R-MBI, CdTe, α-Se, Si, and C versus different X-ray photons energy. (d) Bias-dependent photoconductivity of R-MBI SC detector.

  DownLoad: Full-Size Img PowerPoint

  (1)

  where ε is the relative dielectric constant, ε₀ is the vacuum dielectric constant, e is the elemental charge, and L is the thickness. The relative dielectric constant ε ~11.8 of R-MBI was measured. After fitting calculation, the corresponding n_trap was determined to be 1.22 × 10¹⁰ cm⁻³, which is considerably better than that of typical semiconductors Si (n_trap = 10¹³–10¹⁴ cm^–3) and CdTe (n_trap = 10¹¹–10¹³ cm^–3) [36,37]. Such low trap density would increase charge transport ability, thus facilitating the acquisition of a high μτ product.

  As previously stated, R-MBI possesses a low trap density, high resistivity, and the heavy element Bi, all of which are essential for an excellent X-ray detector. We used the NIST X-COM application to estimate the X-ray absorption coefficients of R-MBI and conventional semiconductors over a wide range of photon energies (0.01–100 MeV) to examine the detection capacity of R-MBI [38]. In the entire energy range of interest, the absorption coefficient of R-MBI is comparable to that of α-Se and CdTe, and significantly superior to that of conventional silicon and carbon materials (Fig. 3c). Due to the high attenuation capacities ensuring adequate X-ray photon absorption, it is easier to collect charge carriers. As an X-ray detector, the μτ product is a crucial indicator to gain insight into the performance of the device, and a higher μτ product can collect more carriers at low electric fields [39]. The photocurrent-voltage curve of the R-MBI SC device is depicted in Fig. 3d, along with the accompanying fitting lines. The μτ product is calculated by the modified Hecht equation [40]:

  (2)

  where I₀ is the saturated photocurrent, d is the thickness, and V is the applied bias. Especially, R-MBI displays a relatively high μτ product of 1.88 × 10⁻⁴ cm²/V, which is equivalent to the values for lead-based perovskites of (BDA)PbI₄ (4.43 × 10⁻⁴ cm²/V) and Cs₃Bi₂I₉ (7.97 × 10⁻⁴ cm²/V) [41,42].

  As a result, we constructed an electrode with dimensions of 1 × 2 × 1.2 mm³ to test the X-ray detection performance of R-MBI (Fig. 4a). By adjusting the X-ray tube current from 5 µA to 200 µA, the on/off photoelectric response was measured at different electric field strengths (Fig. 4b and Fig. S7 in Supporting information). The correlation between the response photocurrent and dose rates under various electric field strengths was depicted in Fig. 4c. The current density showed a nearly linear connection with the dosage rate of X-ray radiation, and the sensitivity could be calculated by measuring the slope of the linear relationship. Fig. 4d plots the computed X-ray detection sensitivity versus the electric field. The sensitivity of the R-MBI SC device is about 263.58 µC Gy_air⁻¹ cm⁻² at 50 V mm⁻¹ resulting from the linear fit. This characteristic outperforms previously published bismuth-based 1D perovskite (DMEDA)BiI₅ detectors (72.5 µC Gy_air⁻¹ cm⁻² at 494 V/mm) and the double perovskite (I-BA)₄AgBiI₈ X-ray devices (5.38 µC Gy_air⁻¹ cm⁻² at 4.5 V/mm) [30,43]. Additionally, we examined the noise current by computing the standard deviation of the response current. In various field strength ranges, we calculated that the SNR was more than 3 when the dosage rate was 4.35 µGy_air/s (Fig. S8 in Supporting information) [34]. As a consequence, the lowest detectable dose rate of the R-MBI SC device is 4.35 µGy_air/s, which is lower than the minimum dose rate needed for normal medical diagnosis (5.5 µGy_air/s) [44,45]. Moreover, the photoelectric response to the X-ray was investigated by cycling the incident X-ray on and off at regular intervals (Fig. S9 in Supporting information), indicating that the output photocurrent was quite steady without obvious degradation.

  Figure 4

  

  Figure 4.  (a) Device structure with R-MBI SC as the X-ray detector. (b) The photocurrent responses under different dose rates. (c) The relationship between photocurrent density and dose rate of R-MBI SC device at different field strengths. (d) Sensitivity under different electric fields of R-MBI SC detector.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, we designed a lead-free bismuth iodide perovskite, (R)-(H₂MPz)BiI₅ (R-MBI), and centimeter-sized crystals were successfully acquired. Moreover, the features of low trap density, high attenuation efficiency and μτ product make the R-MBI single-crystal device exhibits high-sensitive X-ray detection with a superior sensitivity of 263.58 µC Gy_air⁻¹ cm⁻² at 50 V/mm. Besides, the R-MBI single-crystal detector displays a detection limit as low as 4.35 µGy_air/s, meeting the dose rate required for medical diagnosis. This research advances the progress on high-performance bismuth-based perovskite X-ray devices and broadens the scope of environmentally friendly detectors for X-ray inspection.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (Nos. 22175177, 21971238, 22193042, 21833010, 22125110, 22122507, 21921001, and U21A2069), the Key Research Program of Frontier Sciences of the Chinese Academy of Sciences (No. ZDBS-LY-SLH024). The National Postdoctoral Program for Innovative Talents (No. BX2021315), and the National Key Research and Development Program of China (No. 2019YFA0210402).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.108708.

  
  
• 1. [1]

     Y. He, C.C. Stoumpos, I. Hadar, et al., J. Am. Chem. Soc. 143 (2021) 2068–2077. doi: 10.1021/jacs.0c12254

  2. [2]

     H. Wei, Y. Fang, P. Mulligan, et al., Nat. Photonics 10 (2016) 333–339. doi: 10.1038/nphoton.2016.41

  3. [3]

     J. Liu, B. Shabbir, C. Wang, et al., Adv. Mater. 31 (2019) 1901644. doi: 10.1002/adma.201901644

  4. [4]

     A. Sakdinawat, D. Attwood, Nat. Photonics 4 (2010) 840–848. doi: 10.1038/nphoton.2010.267

  5. [5]

     Z. Liu, J.A. Peters, H. Li, et al., J. Electron. Mater. 44 (2015) 222–226. doi: 10.1007/s11664-014-3372-2

  6. [6]

     Y. Li, Y. Lei, H. Wang, et al., Nano-Micro Lett. 15 (2023) 128. doi: 10.1007/s40820-023-01118-1

  7. [7]

     Z. Xu, X. Liu, Y. Li, et al., Angew. Chem. Int. Ed. 58 (2019) 15757–15761. doi: 10.1002/anie.201909815

  8. [8]

     W. Yuan, G. Niu, Y. Xian, et al., Adv. Funct. Mater. 29 (2019) 1900234. doi: 10.1002/adfm.201900234

  9. [9]

     S. Kasap, J.B. Frey, G. Belev, et al., Sensors 11 (2011) 5112–5157. doi: 10.3390/s110505112

  10. [10]

      Z. Gou, W. Liu, S. Huanglong, et al., IEEE Electron Device Lett. 40 (2019) 578–581. doi: 10.1109/led.2019.2897757

  11. [11]

      P. Duvauchelle, G. Peix, D. Babot, Nucl. Instrum. Methods Phys. Res. Sect. B 155 (1999) 221–228. doi: 10.1016/S0168-583X(99)00450-4

  12. [12]

      M. Jeong, W.J. Jo, H.S. Kim, et al., Nucl. Instrum. Methods Phys. Res. Sect. A 784 (2015) 119–123. doi: 10.1016/j.nima.2014.11.013

  13. [13]

      G. Belev, S.O. Kasap, J. Non-Cryst. Solids 345–346 (2004) 484–488.

  14. [14]

      D. Yang, R. Yang, K. Wang, et al., Nat. Commun. 9 (2018) 3239. doi: 10.1038/s41467-018-05760-x

  15. [15]

      J.C. Blancon, J. Even, C.C. Stoumpos, et al., Nat. Nanotechnol. 15 (2020) 969–985. doi: 10.1038/s41565-020-00811-1

  16. [16]

      H. Wei, D. DeSantis, W. Wei, et al., Nat. Mater. 16 (2017) 826–833. doi: 10.1038/nmat4927

  17. [17]

      X. He, Y. Deng, D. Ouyang, et al., Chem. Rev. 123 (2023) 1207–1261. doi: 10.1021/acs.chemrev.2c00404

  18. [18]

      L. Li, H. Chen, Z. Fang, et al., Adv. Mater. 32 (2020) 1907257.

  19. [19]

      D. Shi, V. Adinolfi, R. Comin, et al., Science 347 (2015) 519–522. doi: 10.1126/science.aaa2725

  20. [20]

      Y. He, L. Matei, H.J. Jung, et al., Nat. Commun. 9 (2018) 1609. doi: 10.1038/s41467-018-04073-3

  21. [21]

      S. Shrestha, R. Fischer, G.J. Matt, et al., Nat. Photonics 11 (2017) 436–440. doi: 10.1038/nphoton.2017.94

  22. [22]

      C. Shan, F. Meng, J. Yu, et al., J. Mater. Chem. C 9 (2021) 7632–7642. doi: 10.1039/d1tc01323h

  23. [23]

      W. Xu, M. Niu, X. Yang, et al., Chin. Chem. Lett. 32 (2021) 489–492. doi: 10.1016/j.cclet.2020.05.017

  24. [24]

      S. Yakunin, M. Sytnyk, D. Kriegner, et al., Nat. Photonics 9 (2015) 444–449. doi: 10.1038/nphoton.2015.82

  25. [25]

      W. Wei, Y. Zhang, Q. Xu, et al., Nat. Photonics 11 (2017) 315–321. doi: 10.1038/nphoton.2017.43

  26. [26]

      A. Babayigit, A. Ethirajan, M. Muller, et al., Nat. Mater. 15 (2016) 247–251. doi: 10.1038/nmat4572

  27. [27]

      R. Mohan, Nat. Chem. 2 (2010) 336. doi: 10.1038/nchem.609

  28. [28]

      R.E. Brandt, R.C. Kurchin, R.L.Z. Hoye, et al., J. Phys. Chem. Lett. 6 (2015) 4297–4302. doi: 10.1021/acs.jpclett.5b02022

  29. [29]

      K. Tao, Y. Li, C. Ji, et al., Chem. Mater. 31 (2019) 5927–5932. doi: 10.1021/acs.chemmater.9b02263

  30. [30]

      L. Yao, G. Niu, L. Yin, et al., J. Mater. Chem. C 8 (2020) 1239–1243. doi: 10.1039/c9tc06313g

  31. [31]

      C. Ji, S. Wang, Y. Wang, et al., Adv. Funct. Mater. 30 (2020) 1905529.

  32. [32]

      J.A. Steele, W. Pan, C. Martin, et al., Adv. Mater. 30 (2018) 1804450. doi: 10.1002/adma.201804450

  33. [33]

      C. Wang, H. Li, M. Li, et al., Adv. Funct. Mater. 31 (2021) 2009457. doi: 10.1002/adfm.202009457

  34. [34]

      W. Pan, H. Wu, J. Luo, et al., Nat. Photonics 11 (2017) 726–732. doi: 10.1038/s41566-017-0012-4

  35. [35]

      Q. Cui, X. Song, Y. Liu, et al., Matter 4 (2021) 2490–2507. doi: 10.1016/j.matt.2021.05.018

  36. [36]

      J.R. Ayres, J. Appl. Phys. 74 (1993) 1787–1792. doi: 10.1063/1.354782

  37. [37]

      A. Balcioglu, R.K. Ahrenkiel, F. Hasoon, J. Appl. Phys. 88 (2000) 7175–7178. doi: 10.1063/1.1326465

  38. [38]

      M.J. Berger, J.H. Hubbell, S.M. Seltzer, et al., XCOM: photon Cross Sections Database: NIST Standard Reference Database 8 (NIST), https://www.nist.gov/pml/xcom-photon-cross-sections-database, accessed.

  39. [39]

      Y. Liu, H. Ye, Y. Zhang, et al., Matter 1 (2019) 465–480. doi: 10.1016/j.matt.2019.04.002

  40. [40]

      J. Androulakis, S.C. Peter, H. Li, et al., Adv. Mater. 23 (2011) 4163–4167. doi: 10.1002/adma.201102450

  41. [41]

      Y. Shen, Y. Liu, H. Ye, et al., Angew. Chem. Int. Ed. 59 (2020) 14896–14902. doi: 10.1002/anie.202004160

  42. [42]

      Y. Zhang, Y. Liu, Z. Xu, et al., Nat. Commun. 11 (2020) 2304.

  43. [43]

      Z. Xu, H. Wu, D. Li, et al., J. Mater. Chem. C 9 (2021) 13157–13161. doi: 10.1039/d1tc03412j

  44. [44]

      D.R. Douglas, M. Bopaiah, Health Phys. 79 (2000) S20–S21. doi: 10.1097/00004032-200008001-00007

  45. [45]

      I. Clairand, J.M. Bordy, E. Carinou, et al., Radiat. Meas. 46 (2011) 1252–1257.

• 

  Figure 1  (a) The simulated morphology based on the R-MBI SC (left) and solution-grown bulk crystal of R-MBI (right). (b) The crystal structure of R-MBI.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) The optical absorption spectrum of R-MBI with the insert of corresponding Tauc plot. (b) The calculated energy band structure of the R-MBI.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Bulk resistivity of R-MBI. (b) The Ⅰ-Ⅴ curve of R-MBI using the SCLC analysis. (c) Linear attenuation coefficient of R-MBI, CdTe, α-Se, Si, and C versus different X-ray photons energy. (d) Bias-dependent photoconductivity of R-MBI SC detector.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Device structure with R-MBI SC as the X-ray detector. (b) The photocurrent responses under different dose rates. (c) The relationship between photocurrent density and dose rate of R-MBI SC device at different field strengths. (d) Sensitivity under different electric fields of R-MBI SC detector.

  下载: 全尺寸图片 幻灯片
•