English

• Non-centrosymmetric materials, lacking structural spatial inversion symmetry, exhibit various properties including ferroelectricity [1,2], piezoelectricity [3,4], pyroelectricity [3,5,6], and frequency-doubling effect [7]. Particularly, asymmetric single-phase compounds can also generate stable photocurrents and above-bandgap photovoltages under continuous uniform illumination [8,9], known as the bulk anomalous photovoltaic (BAPV) effect, which is anticipated to achieve a high-performance self-powered photoelectric response [10–12]. Currently, the exploration of BAPV has largely focused on a few insulating inorganic oxides like LiNbO₃ [13,14], BaTiO₃ [15–17], and BiFeO₃ [18,19]. However, the majority of these devices suffer from poor electronic properties which limit their application in the field of photovoltaics. Fortunately, the proposition of the pyro-phototronic effect aims to enhance photoelectric performance by naturally connecting the photo-induced pyroelectric effect with semiconductor properties [20–23]. The charges generated by photo-induced pyroelectric polarization can efficiently modify carrier separation and transportation. Upon light illumination, the transient temperature rises in these materials can generate current signals in the polarization direction to enhance the light response. Therefore, the exploration of novel non-centrosymmetric materials with BAPV effect accompanied by pyro-phototronic effect is desirable.

  Organic−inorganic halide perovskites have become mainstream semiconducting materials with wide-ranging applications, including photovoltaics [24], light-emitting devices [25], and photodetectors [26,27]. Furthermore, the flexibility of the perovskite structures enables the incorporation of chiral ligands, giving rise to a novel concept known as chiral hybrid perovskites (CHPs) [28,29]. As a consequence, by combining the advantages of the perovskite structure with inherent chiral features, CHPs are widely used in chiral optoelectronic devices, such as spintronic devices, nonlinear optics devices and circularly polarized light detectors [30,31]. Given the unique properties, introducing chiral organic cations disrupts spatial inversion symmetry, opening avenues for high-performance photovoltaic applications. For instance, previously reported chiral perovskites, such as (R/S-1-phenylpropylamine)₂BiI₅ (0.63 V) [32], (R/S-1-phenylpropylamine)EAPbCl₄ (0.5 V) [33], and (R-β-methylphenethylammonium)₄AgBiI₈ (0.36 V) [34], have fully exhibited their fascinating bulk photovoltaic effect. On the other hand, the introduction of chiral cations renders CHPs more likely to crystallize in five chiral-polar point groups, namely C₁, C₂, C₃, C₄, and C₆ [28,35,36]. These chiral-polar perovskites with inherent polarization will readily generate the pyro-phototronic effect [37,38].

  Herein, we synthesized a pair of zero-dimensional (0D) bismuth-based CHPs (S-α-MBA)₄Bi₂I₁₀ and (R-α-MBA)₄Bi₂I₁₀ (1S and 1R), by assembling chiral methylbenzylamine (S/R-α-MBA) cations. These compounds crystallized in the chiral-polar point group (P2₁), induced by the organic chiral ligands. Remarkably, the intrinsic asymmetric structure of these compounds gives rise to a notable BAPV effect, characterized by a steady-state photovoltage (V_oc) reaching an impressive 15 V. The synergism of BAPV and pyro-phototronic effects can further modulate the optoelectronic processes, enabling highly enhanced self-powered detection based on CHPs. These merits enable 1S and 1R to exhibit a broadband response range that surpasses the inherent bandgap limitation, spanning from 405 nm to 785 nm. The photodetector can realize simultaneous multi-signal sensing and effectively utilize the electric energy generated by photo-thermal coupling, opening up possibilities for the design and development of prominent self-powered materials.

  Single crystals of 1S and 1R with a size up to 10 × 5 × 5 mm³ have been successfully grown using the temperature-lowering crystallization method. Scheme 1 provides schematic illustrations for the crystal growth process. Powder X-ray diffraction (XRD) analysis confirmed the high purity by closely matching the simulation results (Fig. S1 in Supporting information). Furthermore, the thermogravimetric analysis revealed no mass loss from the crystals up to 495 K, disclosing its high thermostability (Fig. S2 in Supporting information). The crystallization of 1S and 1R is largely dependent on the addition of chiral components. These crystals adopted a chiral-polar space group of P2₁ (detailed crystallographic data are provided in Table S1 in Supporting information). Their chiral-polar characteristic was confirmed through the second harmonic generation signals (Fig. S3 in Supporting information). Given the similarity between the crystallographic structural parameters of the two enantiomers, the crystal structure of 1R will be described in detail.

  Scheme 1

  

  Scheme 1.  Schematic illustration of the growth processes of 1S and 1R crystals.

  DownLoad: Full-Size Img PowerPoint

  Fig. 1a illustrates the structural arrangement, comprising four R-MBA⁺ cations and one [Bi₂I₁₀]⁴⁻ dimer as the fundamental structural unit. Each [Bi₂I₁₀]⁴⁻ anionic dimer is formed by BiI₆ octahedra sharing their edges through two iodine atoms. The chiral organic molecules are firmly anchored to iodine atoms of dimers via N-H···I hydrogen bonds. Moreover, the Bi-I bond lengths and I-Bi-I bond angles are 2.92−3.33 Å and 84.56°−177.30°, indicating the tilts and distortion of the inorganic skeleton (Fig. 1b, Tables S2 and S3 in Supporting information). The electric polarization (P_s) of 1R primarily arises from the arrangement of the electric dipole moments of the organic cation [28]. The dipole moment of the S/R-α-MBA ion is estimated to be 7.75 Debye by DFT (Fig. S4 in Supporting information). Then, the calculated P_s along the b-axis was found to be 0.4 µC/cm² using the point charge model (Tables S4 and S5 in Supporting information). Thus, this chiral polarity feature empowers 1S and 1R with considerable potential to display the BAVE effect for self-driven detection.

  Figure 1

  

  Figure 1.  (a) Structures of 1S and 1R crystals. (b) The inorganic frameworks and bond lengths of 1S and 1R.

  DownLoad: Full-Size Img PowerPoint

  The optical properties of 1S and 1R were investigated first. The ultraviolet-visible absorption spectra reveal they present a sharp absorption edge at around 610 nm (Fig. 2a and Fig. S5 in Supporting information). By fitting the Tauc curve, the optical bandgap (E_g) was calculated to be 2.0 eV [39]. This value is comparable to that of other 0D binuclear bismuth halide perovskites, such as (R/S-1-phenylpropylamine)₂BiI₅ (2.17 eV) and (R-1-(4-fluoro)-phenylethylammonium)₄Bi₂I₁₀ (2.09 eV) [32,40]. Such a relatively narrow bandgap holds the potential for achieving broadband optoelectronic detection. To verify the intrinsic chirality of 1S and 1R crystals, circular dichroism (CD) signals were recorded for the thin films (Fig. 2b). The prepared films also show a preferred orientation of the (100) plane (Fig. S6 in Supporting information). Notably, the CD spectra exhibit that the CD signals of 1S and 1R are mirror images of each other and correspond to the optical intrinsic absorption of perovskites. These results provide compelling evidence of the intrinsic chirality feature of the crystals, further confirming the efficient transfer of chirality from organic molecules to inorganic parts.

  Figure 2

  

  Figure 2.  (a) The absorption spectrum and optical bandgap from the Tauc plot of 1R. (b) The CD spectra of 1S and 1R films.

  DownLoad: Full-Size Img PowerPoint

  Subsequently, we fabricated photodetectors using both 1S and 1R chiral crystals to investigate the photovoltaic effect. Under 405 nm laser illumination, the current-voltage (I–V) curves clearly showed the BAPV effect when the electrode direction was aligned parallel to the polar b-axis, as depicted in Fig. 3a. In this configuration, the V_oc impressively reached a maximum of 15 V, approximately 7.5 times larger than its optical bandgap. Particularly, the V_OC of 1R increases with incident light intensity (Fig. S7 in Supporting information). Such a huge V_oc should be the first demonstration in the CHPs, much higher than that of double perovskite (4-bromobenzylammonium)₂CsAgBiBr₇ (3.2 V) and hybrid perovskite ferroelectrics (n-hexylammonium)₂(ethylammonium)₂Pb₃Br₁₀ (7.4 V) [10,11]. Crucially, the BAPV effect is intricately connected to the P_s of the crystals. The V_oc and short-circuit photocurrent (I_sc) are observed for the opposite P_s direction: I_sc < 0 and V_oc > 0 for 1R (b-axis//[010]) and I_sc < 0, V_oc > 0 for 1S (b-axis//[0 1 ¯ 0]). These opposing electric polarizations between enantiomers result from the crystallographic constraints imposed by the chiral polar system. To verify polarity dependency, the crystal was rotated 180° around the b-axis. This resulted in a reversal of the V_oc and I_sc signs (Fig. S8 in Supporting information): I_sc > 0, V_oc < 0 for 1R (b-axis//[010]) and I_sc > 0, V_oc < 0 for 1S (b-axis//[010]). Because of the chirality remained unchanged, the sign change of I_sc may be explained by the change of P_s direction.

  Figure 3

  

  Figure 3.  (a) I-V curves at 405 nm laser for devices 1S and 1R. (b) I-V curves for 1S and 1R devices when the electrode direction is vertical to the b-axis. (c) The V_OC of 1R versus the time when the light is turned on. (d) Variation of V_OC with the switching “on/off” cycles.

  DownLoad: Full-Size Img PowerPoint

  To clarify the correlation between P_s and the BAPV effect, we measured photocurrents perpendicular to the b-axis. As expected, there are no I_sc and V₀ signals (Fig. 3b). This result strongly suggests that the BAPV effect is intricately linked to polarity along the b-axis. In other words, the photovoltaic effect, as manifested solely along the b-axis, reaffirms its origin in the chiral-polar character of these crystals. Fig. 3c and Fig. S9 (Supporting information) illustrate the photovoltaic-time (V-t) curve under continuous irradiation, exhibiting above-bandgap V_oc = 15 V in 1S and 1R. Additionally, the V-t curve measured at zero bias during the light switching on/off cycles verified the good stability of the BAPV effect (Fig. 3d). The remarkable BAPV effect, independent of bandgap, highlights 1S and 1R as promising candidates for prominent pyro-phototronic detection.

  In pyro-phototronic devices, the simultaneous operation of pyroelectric and photovoltaic effects synergistically promotes the transport of photogenerated carriers and reduces the recombination rate of electron-hole pairs [22,41,42]. The principle is illustrated in Fig. 4a. When exposed to light, charge carriers are generated within the chiral-polar structure, while the increased temperature of the 1R crystal induces pyroelectric polarized potential. The transient current and four-stage behavior reveal the pyro-phototronic effect (Fig. 4b) [43]. In stage ‘Ⅰ’, the pyroelectric material undergoes a transient current peak characterized by a sudden increase in temperature (dT/dt > 0). This current comprises both the output current from the pyroelectric and photovoltaic effects, represented as I_pyro+photo. Under continuous illumination, the temperature remains constant, leading to a stable plateau in the output current without the effect of pyroelectric, represented as I_photo (dT/dt = 0, stage ‘Ⅱ’). However, when the light is turned off, a reverse pyroelectric peak I_pyro’ is generated (dT/dt < 0, Stage Ⅲ). In stage ‘Ⅳ’, the pyroelectric current gradually disappears as the temperature returns to room temperature (dT/dt = 0).

  Figure 4

  

  Figure 4.  (a) Working mechanism of pyro-phototronic effect. (b) Four stages of pyroelectric effect. (c) I–t curves of 1R detector under different laser illumination (245 mW/cm²). (d) The response time during one cycle. (e) Repeated switching cycles of the pyro-phototronic effect.

  DownLoad: Full-Size Img PowerPoint

  Fig. 4c depicts the I–t curves of 1R within the spectral range from 405 nm to 785 nm, indicating the capability of a broadband response, which breaks the limit of optical bandgap. The curves reveal two distinct stages: initial sharp transient current peaks (I_pyro+photo), followed by a steady plateau (I_photo) without pyroelectric effect as the temperature remains constant. The current peak, a result of the combined BAPV and pyroelectric effects, shows a clear wavelength dependence and a high enhancement. Moreover, at zero bias, the peak pyroelectric current monotonically increases with increasing incident power (Fig. S10 in Supporting information). Since the temperature gradient increases with stronger light, it leads to higher densities of electrons and holes. These findings show that 1R not only generates stable photovoltaic I_photo directly but also amplifies the response via the pyro-phototronic effect. The rising and falling components of the response time of 1R inferred from single switching cycle are approximately 0.5 s and 3.4 s, respectively (Fig. 4d). Furthermore, the photostability of 1R devices under 520 nm irradiation is depicted in Fig. 4e, demonstrating that the transient photoinduced pyroelectric current remains consistently stable without obvious degradation. These findings highlight the synergistic interaction between the BAPV and pyroelectric effects, offering a promising path toward advanced broadband self-powered photoelectric detection.

  In summary, a pair of CHPs (S-α-MBA)₄Bi₂I₁₀ and (R-α-MBA)₄Bi₂I₁₀ have been synthesized, which crystallize in the chiral-polar space group of P2₁. This chiral-polar structure induces a superior bulk anomalous photovoltaic effect (BAPV), which exhibits an above-bandgap photovoltage extraordinary 15 V. As far as we know, this should be the first discovery of the BAPV effect in lead-free chiral hybrid perovskites. In addition, such BAPV effect gives rise to a distinctive pyro-phototronic effect, which further enables a highly enhanced self-powered detection with a broad response from 405 nm to 785 nm. Such photo-thermal coupling is capable of simultaneous multi-signal sensing and effectively enhances photoelectric response. These unique features establish a potential platform for optoelectronic applications, particularly in the field of photovoltaics.

  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.

  CRediT authorship contribution statement

  Xin Dong: Writing – original draft, Data curation. Zhijin Xu: Formal analysis. Jing Liang: Formal analysis. Tianqi Chen: Investigation. Junhua Luo: Writing – review & editing. Lina Li: Writing – review & editing.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (NSFC, Nos. 22322506, 22175177, 22193042, 22125110, 22122507, 21921001 and U21A2069), the NSF of Fujian Province (No. 2023J06052), the Key Research Program of Frontier Sciences of the Chinese Academy of Sciences (No. ZDBS-LY-SLH024).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111268.

  
  
• 1. [1]

     J.F. Scott, Science 315 (2007) 954–959. doi: 10.1126/science.1129564

  2. [2]

     Y. Chen, C. Gao, T. Yang, et al., Chin. J. Struct. Chem. 41 (2022) 2204001–2204011.

  3. [3]

     X. Yuan, J. Shi, Y. Kang, et al., Adv. Mater. 36 (2024) 2308726. doi: 10.1002/adma.202308726

  4. [4]

     H.S. Wu, B.T. Murti, J. Singh, P.K. Yang, M.L. Tsai, Adv. Sci. 9 (2022) 2104703. doi: 10.1002/advs.202104703

  5. [5]

     R. Mondal, M.A.M. Hasan, J.M. Baik, Y. Yang, Mater. Today 66 (2023) 273–301. doi: 10.1016/j.mattod.2023.03.023

  6. [6]

     L. Tang, W. Weng, H. Chen, et al., Adv. Funct. Mater. 33 (2023) 2214858. doi: 10.1002/adfm.202214858

  7. [7]

     J. Chen, C.L. Hu, F. Kong, J.G. Mao, Acc. Chem. Res. 54 (2021) 2775–2783. doi: 10.1021/acs.accounts.1c00188

  8. [8]

     C. Paillard, X. Bai, I.C. Infante, et al., Adv. Mater. 28 (2016) 5153–5168. doi: 10.1002/adma.201505215

  9. [9]

     H.Y. Ye, Y. Zhang, D.W. Fu, R.G. Xiong, Angew. Chem. Int. Ed. 53 (2014) 11242–11247. doi: 10.1002/anie.201406810

  10. [10]

      Y. Liu, W. Guo, L. Hua, et al., J. Am. Chem. Soc. 145 (2023) 16193–16199. doi: 10.1021/jacs.3c05020

  11. [11]

      L. Hua, L. Tang, Y. Liu, et al., Small. 19 (2023) 2207393. doi: 10.1002/smll.202207393

  12. [12]

      R. Chakraborty, P.K. Rajput, G.M. Anilkumar, et al., J. Am. Chem. Soc. 145 (2023) 1378–1388. doi: 10.1021/jacs.2c12034

  13. [13]

      A.M. Glass, D. von der Linde, T.J. Negran, Appl. Phys. Lett. 25 (1974) 233–235. doi: 10.1063/1.1655453

  14. [14]

      J. Carnicero, O. Caballero-Calero, M. Carrascosa, J.M. Cabrera, Appl. Phys. B 79 (2004) 351–358. doi: 10.1007/s00340-004-1543-1

  15. [15]

      J.E. Spanier, V.M. Fridkin, A.M. Rappe, et al., Nat. Photonics 10 (2016) 611–616. doi: 10.1038/nphoton.2016.143

  16. [16]

      W.T.H. Koch, R. Munser, W. Ruppel, P. Würfel, Ferroelectrics. 13 (1976) 305–307. doi: 10.1080/00150197608236596

  17. [17]

      S.H. Wemple, M. Didomenico, I. Camlibel, J. Phys. Chem. Solids 29 (1968) 1797–1803. doi: 10.1016/0022-3697(68)90164-9

  18. [18]

      S.Y. Yang, J. Seidel, S.J. Byrnes, et al., Nat. Nanotechnol. 5 (2010) 143–147. doi: 10.1038/nnano.2009.451

  19. [19]

      A. Bhatnagar, A.R. Chaudhuri, Y.H. Kim, D. Hesse, M. Alexe, Nat. Commun. 4 (2013) 2835. doi: 10.1038/ncomms3835

  20. [20]

      Y. Wang, L. Zhu, Y. Feng, Z. Wang, Z.L. Wang, Adv. Funct. Mater. 29 (2019) 1807111. doi: 10.1002/adfm.201807111

  21. [21]

      Y. Zhang, J. Chen, L. Zhu, Z.L. Wang, Nano Lett. 21 (2021) 8808–8816. doi: 10.1021/acs.nanolett.1c03171

  22. [22]

      Y. Dai, X. Wang, W. Peng, et al., Adv. Mater. 30 (2018) 1705893. doi: 10.1002/adma.201705893

  23. [23]

      Z. Wang, R. Yu, X. Wang, W. Wu, Z.L. Wang, Adv. Mater. 28 (2016) 6880–6886. doi: 10.1002/adma.201600884

  24. [24]

      I.H. Park, K.C. Kwon, Z. Zhu, et al., J. Am. Chem. Soc. 142 (2020) 18592–18598. doi: 10.1021/jacs.0c08189

  25. [25]

      L. Li, W. Wu, D. Li, et al., CCS Chem. 4 (2022) 2491–2497. doi: 10.31635/ccschem.021.202101037

  26. [26]

      Z. Xu, X. Dong, L. Wang, et al., J. Am. Chem. Soc. 145 (2023) 1524–1529. doi: 10.1021/jacs.2c12300

  27. [27]

      B. Shi, P. Wang, J. Feng, et al., Nano-Micro Lett. 15 (2023) 3. doi: 10.1007/s40820-022-00961-y

  28. [28]

      P.J. Huang, K. Taniguchi, H. Miyasaka, J. Am. Chem. Soc. 141 (2019) 14520–14523. doi: 10.1021/jacs.9b06815

  29. [29]

      Y. Dang, X. Liu, B. Cao, X. Tao, Matter. 4 (2021) 794–820. doi: 10.1016/j.matt.2020.12.018

  30. [30]

      Y. Dong, Y. Zhang, X. Li, et al., Small. 15 (2019) 1902237. doi: 10.1002/smll.201902237

  31. [31]

      J. Ma, H. Wang, D. Li, Adv. Mater. 33 (2021) 2008785. doi: 10.1002/adma.202008785

  32. [32]

      S. You, Z.K. Zhu, S. Dai, et al., Adv. Funct. Mater. 33 (2023) 2303523. doi: 10.1002/adfm.202303523

  33. [33]

      T. Zhu, H. Wu, C. Ji, et al., Adv. Opt. Mater. 10 (2022) 2200146. doi: 10.1002/adom.202200146

  34. [34]

      J. Wu, S. You, P. Yu, et al., ACS Energy Lett. 8 (2023) 2809–2816. doi: 10.1021/acsenergylett.3c00629

  35. [35]

      Y.L. Zeng, X.Q. Huang, C.R. Huang, et al., Angew. Chem. Int. Ed. 60 (2021) 10730–10735. doi: 10.1002/anie.202102195

  36. [36]

      T. Zhu, J. Bie, C. Ji, et al., Nat. Commun. 13 (2022) 7702. doi: 10.1038/s41467-022-35441-9

  37. [37]

      N. Ma, Y. Yang, Nano Energy 40 (2017) 352–359. doi: 10.1016/j.nanoen.2017.08.043

  38. [38]

      W. Peng, X. Wang, R. Yu, et al., Adv. Mater. 29 (2017) 1606698. doi: 10.1002/adma.201606698

  39. [39]

      J. Tauc, Mater. Res. Bull. 3 (1968) 37–46. doi: 10.1016/0025-5408(68)90023-8

  40. [40]

      N. Dehnhardt, M. Axt, J. Zimmermann, et al., Chem. Mater. 32 (2020) 4801–4807. doi: 10.1021/acs.chemmater.0c01605

  41. [41]

      Z. Wang, R. Yu, C. Pan, et al., Nat. Commun. 6 (2015) 8401. doi: 10.1038/ncomms9401

  42. [42]

      S. Sahare, P. Ghoderao, M.K. Sharma, et al., Nano Energy 107 (2023) 108172. doi: 10.1016/j.nanoen.2023.108172

  43. [43]

      S. Tian, B. Li, Y. Dai, Z.L. Wang, Mater. Today 68 (2023) 254–274. doi: 10.1016/j.mattod.2023.07.001

• 

  Scheme 1  Schematic illustration of the growth processes of 1S and 1R crystals.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  (a) Structures of 1S and 1R crystals. (b) The inorganic frameworks and bond lengths of 1S and 1R.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) The absorption spectrum and optical bandgap from the Tauc plot of 1R. (b) The CD spectra of 1S and 1R films.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) I-V curves at 405 nm laser for devices 1S and 1R. (b) I-V curves for 1S and 1R devices when the electrode direction is vertical to the b-axis. (c) The V_OC of 1R versus the time when the light is turned on. (d) Variation of V_OC with the switching “on/off” cycles.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Working mechanism of pyro-phototronic effect. (b) Four stages of pyroelectric effect. (c) I–t curves of 1R detector under different laser illumination (245 mW/cm²). (d) The response time during one cycle. (e) Repeated switching cycles of the pyro-phototronic effect.

  下载: 全尺寸图片 幻灯片
•