Realization of the transition from an indirect band gap to a direct band gap through halogen regulation in the Cs–Ag–X system

Bingliang Cheng ab, Wenjuan Ma c, Cong Hu b, Yuhan Lv b, Zongxiao Li b, Jianchao Zhao b, Tao Bo ab and Wenwen Lin *ab
aQianwan Institute of CNITECH, Ningbo 315336, China. E-mail: linwenwen@nimte.ac.cn
bZhejiang Key Laboratory of Data-Driven High-Safety Energy Materials and Applications, Ningbo Key Laboratory of Special Energy Materials and Chemistry, Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, Ningbo 315201, China
cCollege of Physical Science and Technology, Department of Physics, Xiamen University, Xiamen 361005, China

Received 7th January 2026 , Accepted 24th February 2026

First published on 25th February 2026


Abstract

Scintillation crystals need to have a direct band gap. Here, two new optical crystals, CsAgICl and Cs2AgBr2I, were obtained through halogen regulation, achieving the transition from an indirect band gap (CsAgCl2) to a direct band gap. This work enriches the structural chemistry of the Cs–Ag–X system and is conducive to the exploration of new scintillation crystals.


Nuclear radiation detection crystals are the core materials of detection devices and have significant applications in medical diagnosis, security screening, high-energy physics, and homeland safety.1–4 The metal halide perovskite CsPbBr3 has attracted much attention due to its excellent photon-to-electron conversion efficiency, high defect tolerance factor, and high carrier mobility-lifetime product.5–7 However, the CsPbBr3 crystal has encountered some inherent problems, such as the toxicity of Pb and ionic migration caused by its three-dimensional (3D) structure, which restricts its environmental friendliness and long-term performance.8–13

In recent years, considering the limitations imposed by the aforementioned various challenges, other lead-free alternative materials, such as Cs3Cu2I5, Cs2AgBiBr6, Cs3Bi2I9, etc., have also been studied and explored.14–20 Their superior stability and environmentally friendly characteristics have attracted widespread attention and achieved rapid development, demonstrating excellent performance in all-inorganic nuclear radiation detection. For these lead-free systems, research on the silver halide Cs–Ag–X system is relatively scarce.21,22 Among them, CsAg2I3 has been reported to have an excellent UV photodetector response speed, due to the [Ag2X6] octahedral units in its structure.23

In this work, based on the halogen regulation strategy, two new Ag-based halides with [AgX4] units were obtained, namely CsAgICl and Cs2AgBr2I. Compared with the known CsAgCl2, the main framework of CsAgICl is a layered structure composed of [AgCl2I2] units, while the anionic framework of Cs2AgBr2I is a 1D [AgBr2I] chain structure composed of [AgBr2I2] units. Meanwhile, the introduction of Br and I changed the coordination environment of the Ag atom from 5-coordination in CsAgCl2 to 4-coordination in CsAgICl and Cs2AgBr2I. Unfortunately, the [AgCl2I2] unit and the [AgBr2I2] unit failed to form the [Ag2X6] octahedral unit with enhanced luminescence properties through edge-sharing connection modes. The results of first-principles calculations indicate that the change of halogen regulates the band gap and realizes the transition from an indirect band gap (CsAgCl2) to a direct band gap (CsAgICl and Cs2AgBr2I). These results will open a new avenue for exploring nuclear radiation detection materials in the Cs–Ag–X system with direct band gaps.

Single crystals of CsAgICl and Cs2AgBr2I were obtained by a high-temperature solid-state method in a vacuum system, and their polycrystalline samples (Fig. S1) can be prepared by solid-state and antisolvent methods (experimental details are presented in the SI). Compared with the solid-state method, the antisolvent method for synthesizing polycrystalline samples is more time-efficient. The purity of the phase was determined by powder X-ray diffraction (XRD) (Fig. 1d and f). Meanwhile, to compare the effect of halogen regulation on the band gap, polycrystalline samples of CsAgCl2 were synthesized. Compared with the previously reported synthesis of CsAgCl2 nanocrystals using the anti-solvent method,21 the crystals synthesized in this study through the supersaturation method have significantly larger sizes, resulting in micron-sized single crystals (Fig. 1a and b). Crystallographic data of CsAgICl and Cs2AgBr2I are presented in Table S1. Their bond lengths and bond angles are comparable to those of known metal halides (Tables S3 and S4). The rationality of the two structures was confirmed by energy dispersive spectroscopy (Fig. 1c and e) and bond valence calculation24 (Table S2).


image file: d6dt00040a-f1.tif
Fig. 1 Microscopic morphology of CsAgCl2 obtained by (a) the antisolvent method and (b) the supersaturation method. (c) Energy dispersive spectroscopy analysis and (d) experimental powder XRD patterns of CsAgICl. (e) Energy dispersive spectroscopy analysis and (f) experimental powder XRD patterns of Cs2AgBr2I.

Two novel Ag-based halides, CsAgICl and Cs2AgBr2I, were obtained by introducing Br/I atoms into the CsAgCl2 structure. CsAgICl crystallizes in the centrosymmetric space group Cmcm, which is the same as that of CsAgCl2. Its cell parameters are a = 4.8341(4) Å, b = 15.6709(13) Å, c = 7.6430(5) Å, and Z = 4. In the asymmetric unit, there is one crystallographically independent Cs atom, one Ag atom, one Cl atom, and one I atom. The central atom Ag is coordinated to two Cl atoms and two I atoms to form the [AgCl2I2] fundamental building block (FBB) (Fig. 2b) with dAg–Cl = 2.646 Å and dAg–I = 2.764 Å. The introduction of larger-sized I atoms alters the coordination of the Ag atoms in the [AgCl5] unit of CsAgCl2 (Fig. 2a). The Cs atom is coordinated to four Cl atoms and four I atoms to form a [CsCl4I4] polyhedron. The bond lengths of dCs–Cl and dCs–I in the [CsCl4I4] polyhedron are 3.482 Å and 3.884–4.124 Å, respectively. The formed [AgCl2I2] units polymerize to form 2D [AgICl] layers through corner-sharing. These layers extend in the bc plane and are connected by [CsCl4I4] polyhedra. Finally, the overall structure of CsAgICl is composed of 2D [AgICl] layers and Cs atoms (Fig. 2b). The difference from the layer structure of CsAgCl2 (Fig. 2a) is that the layers of CsAgICl are not flat. Specifically, when viewed along the a-axis, the angle of Cl–Ag–Cl in CsAgICl is 92.447°, while in CsAgCl2, it is 162.050°.


image file: d6dt00040a-f2.tif
Fig. 2 Crystal structures: (a) CsAgCl2, (b) CsAgICl, and (c) Cs2AgBr2I.

Cs2AgBr2I crystallizes in the centrosymmetric space group Pnma, which is different from that of CsAgCl2. Its cell parameters are a = 10.0800(16) Å, b = 4.8981(8) Å, c = 19.635(3) Å, and Z = 4. In the asymmetric unit, there are two crystallographically independent Cs atoms, one Ag atom, two Br atoms, and one I atom. The central atom Ag is coordinated to two Br atoms and two I atoms to form the [AgBr2I2] FBB (Fig. 2c), with dAg–Br = 2.717–2.724 Å and dAg–I = 2.834 Å, which is different from the [AgCl5] FBB of CsAgCl2. The Cs (1) atom is coordinated to four Br atoms and three I atoms to form a [CsBr4I3] polyhedron, and the bond lengths of dCs–Br and dCs–I in the [CsBr4I3] polyhedron are 3.561–3.629 Å and 3.859–3.889 Å, respectively. The Cs (2) atom is coordinated to six Br atoms and one I atom to form a [CsBr6I] polyhedron, and the bond lengths of dCs–Br and dCs–I in the [CsBr6I] polyhedron are 3.612–3.658 Å and 3.861 Å, respectively. The formed [AgBr2I2] units polymerize to form 1D [AgBr2I] chains through I atoms, with the Br atoms being suspended. These chains extend along the b-axis and are connected by [CsBr4I3] and [CsBr6I] polyhedra. Finally, the overall structure of Cs2AgBr2I is composed of 1D [AgBr2I] chains and Cs atoms (Fig. 2c). The introduction of Br and I atoms has transformed the layer structure of CsAgCl2 into the chain structure of Cs2AgBr2I.

The band gap is a key parameter of optical crystals,25–27 and the UV-vis spectra of CsAgCl2 (Fig. S2), CsAgICl (Fig. 3a) and Cs2AgBr2I (Fig. 3d) were tested by the Tauc plot method.28 The results show that the experimental band gap of CsAgCl2 is 4.17 eV, which is consistent with previous reports,21 and the band gaps of CsAgICl and Cs2AgBr2I are 2.82 eV and 3.53 eV, respectively. Furthermore, the previous reports indicate that CsAgCl2 has an indirect band gap with a calculated value of 3.82 eV (Fig. S3),21 while first-principles calculations in this work indicate that CsAgICl and Cs2AgBr2I have direct band gaps with values of 2.30 eV and 2.73 eV, respectively (Fig. 3b and e). The introduction of Br atoms and I atoms with weaker electronegativity can effectively reduce the band gap, which is consistent with the result that halogens regulate the band gap of optical crystals. The contribution of the orbitals to the band gap was analyzed through partial density of states (PDOS) and integrated projected density of states (IPDOS, ranging from −2.0 to 0.0 eV). For CsAgICl, near the Fermi level, the contributions of the Ag-4d, Cl-3p, and I-5p orbitals to the top of valence bands are 43%, 27%, and 26%, respectively, while the Ag-5s orbitals contribute the most to the bottom of conduction bands, which is 50% (Fig. 3c). The calculated results of Cs2AgBr2I are similar. The Ag-4d, Br-4p, and I-5p orbitals contribute the most to the top of valence bands, accounting for 26%, 49% and 20%, respectively, while the Ag-5s orbitals contribute 44% to the bottom of conduction bands (Fig. 3f). By comparing the PDOS of these three compounds, it can be seen that the optical band gap is mainly determined by the [AgXn] units. The introduction of different halogen atoms alters the coordination environment (from the [AgCl5] unit in CsAgCl2 to the [AgCl2I2] unit in CsAgICl and the [AgBr2I2] unit in Cs2AgBr2I) and the distribution of electronic orbitals of the Ag atom, causing the wave vector positions of the conduction band bottom and the valence band top to tend to overlap, thus transforming from an indirect band gap to a direct band gap. Therefore, the introduction of Br/I atoms in this work transformed the indirect band gap into a direct band gap, which is a prerequisite for scintillation crystals.


image file: d6dt00040a-f3.tif
Fig. 3 (a) UV-vis optical absorption spectrum, (b) calculated band gap, and (c) calculated density of states of CsAgICl. (d) UV-vis optical absorption spectrum, (e) calculated band gap, and (f) calculated density of states of Cs2AgBr2I.

In conclusion, two new metal halide compounds, CsAgICl and Cs2AgBr2I, were obtained through halogen regulation, and their polycrystalline samples were synthesized through solid-state and antisolvent methods. Meanwhile, micron-sized single crystals of CsAgCl2 were grown through the supersaturation method. Furthermore, the optical band gaps and electron structures were characterized and analyzed through experiments and calculations. The results show that the introduction of different halogens can effectively regulate the band gap, converting the indirect band gap to the direct band gap, which will boost the discovery of new metal halide scintillation materials.

Conflicts of interest

The authors declare no competing financial interest.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: details of synthesis and characterization of compounds (PDF). See DOI: https://doi.org/10.1039/d6dt00040a.

CCDC 2519961 and 2519962 contain the supplementary crystallographic data for this paper.29a,b

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (52272171) and the Ningbo Yongjiang Talent Introduction Project (2021A-106-G).

References

  1. Y. H. He, M. Petryk, Z. F. Liu, D. G. Chica, I. Hadar, C. Leak, W. J. Ke, I. Spanopoulos, W. W. Lin, D. Y. Chung, B. W. Wessels, Z. He and M. G. Kanatzidis, Nat. Photonics, 2021, 15, 36–42 CrossRef CAS.
  2. Y. H. He, L. Matei, H. J. Jung, K. M. McCall, M. Chen, C. C. Stoumpos, Z. F. Liu, J. A. Peters, D. Y. Chung, B. W. Wessels, M. R. Wasielewski, V. P. Dravid, A. Burger and M. G. Kanatzidis, Nat. Commun., 2018, 9, 1609 CrossRef PubMed.
  3. J. P. Deng, J. L. Li, Z. Yang and M. Q. Wang, J. Mater. Chem. C, 2019, 7, 12415–12440 RSC.
  4. Q. Xu, J. Wang, W. Y. Shao, X. Ouyang, X. Wang, X. L. Zhang, Y. Guo and X. P. Ouyang, Nanoscale, 2020, 12, 9727–9732 RSC.
  5. W. W. Lin, J. G. He, K. M. McCall, C. C. Stoumpos, Z. F. Liu, I. Hadar, S. Das, H. H. Wang, B. X. Wang, D. Y. Chung, B. W. Wessels and M. G. Kanatzidis, Adv. Funct. Mater., 2021, 31, 2006635 CrossRef CAS.
  6. Y. Wu, X. M. Li and H. B. Zeng, ACS Energy Lett., 2019, 4, 673–681 CrossRef CAS.
  7. X. L. Xu, S. Y. Wang, Y. Chen, W. W. Liu, X. P. Wang, H. T. Jiang, S. Y. Ma and P. D. Yun, ACS Appl. Mater. Interfaces, 2022, 14, 39524–39534 CrossRef CAS PubMed.
  8. C. Liu, H. Chen, P. Lin, H. H. Hu, Q. Y. Meng, L. B. Xu, P. Wang, X. P. Wu and C. Cui, J. Mater. Sci.: Mater. Electron., 2022, 33, 24895–24905 CrossRef CAS.
  9. Y. Y. Zhou and Y. X. Zhao, Energy Environ. Sci., 2019, 12, 1495–1511 RSC.
  10. J. S. Manser, M. I. Saidaminov, J. A. Christians, O. M. Bakr and P. V. Kamat, Acc. Chem. Res., 2016, 49, 330–338 CrossRef CAS PubMed.
  11. H. W. Zhou, L. Fan, G. H. He, C. Yuan, Y. Y. Wang, S. Z. Shi, N. Sui, B. L. Chen, Y. T. Zhang, Q. X. Yao, J. S. Zhao, X. X. Zhang and J. Yin, RSC Adv., 2018, 8, 29089–29095 RSC.
  12. M. Shen, Y. L. Zhang, B. L. Cheng, W. J. Ma, X. L. Huang, L. Zhang, Z. F. Chai and W. W. Lin, CrystEngComm, 2024, 26, 3162–3166 RSC.
  13. X. L. Huang, B. L. Cheng, W. J. Ma, S. M. Qi, Y. L. Zhang, M. Shen, T. Bo, L. Zhang and W. W. Lin, Inorg. Chem., 2024, 63, 11924–11929 CrossRef CAS PubMed.
  14. Q. Yao, J. M. Li, X. S. Li, Y. S. Ma, H. H. Song, Z. Y. Li, Z. G. Wang and X. T. Tao, Adv. Mater., 2023, 35, 2304938 CrossRef CAS PubMed.
  15. L. X. Yin, H. D. Wu, W. C. Pan, B. Yang, P. H. Li, J. J. Luo, G. D. Niu and J. Tang, Adv. Opt. Mater., 2019, 7, 1900491 CrossRef CAS.
  16. M. M. Yang, A. F. Li, X. Hao, L. L. Wu, W. B. Tian, D. Y. Yang and J. Q. Zhang, Adv. Opt. Mater., 2023, 11, 2203066 CrossRef CAS.
  17. M. A. Hadi, M. N. Islam and J. Podder, RSC Adv., 2022, 12, 15461–15469 RSC.
  18. M. N. Islam, M. A. Hadi and J. Podder, AIP Adv., 2019, 9, 125321 CrossRef.
  19. Q. Q. Jia, Z. X. Zhang, H. F. Ni, Q. Lin, G. Teri, P. G. Liu, J. Q. Luo, P. Z. Huang, Z. J. Wang, C. F. Wang, Z. Q. Liu, Y. Zhang and D. W. Fu, Angew. Chem., Int. Ed., 2025, 64, e202505163 CrossRef CAS PubMed.
  20. Q. Q. Jia, G. Teri, Q. F. Zhou, J. Q. Luo, H. F. Ni, P. Z. Huang, P. G. Liu, L. Pan, C. F. Wang, Z. Q. Liu, Z. X. Zhang, Y. Zhang and D. W. Fu, Angew. Chem., Int. Ed., 2025, 64, e202514669 CrossRef CAS PubMed.
  21. D. F. Wu, J. E. Zhou, W. Kang, K. An, J. Y. Yang, M. Zhou, P. He, Q. Huang and X. S. Tang, J. Phys. Chem. Lett., 2021, 12, 5110–5114 CrossRef CAS PubMed.
  22. D. W. Yang, F. Wang, S. Li, Z. F. Shi and S. F. Li, J. Phys. Chem. C, 2024, 128, 2223–2230 CrossRef CAS.
  23. M. M. Yao, Q. Zhang, D. Wang, R. L. Chen, Y. C. Yin, J. Xia, H. Tang, W. P. Xu and S. H. Yu, Adv. Funct. Mater., 2022, 32, 2202894 CrossRef CAS.
  24. I. D. Brown, Phys. Chem. Miner., 1987, 15, 30–34 CrossRef CAS.
  25. B. L. Cheng, W. J. Ma, A. Tudi, F. F. Zhang, Z. L. Chen, X. L. Hou, Z. H. Yang and S. L. Pan, Chem. Mater., 2023, 35, 5671–5679 CrossRef CAS.
  26. B. L. Cheng, Z. J. Li, Y. Chu, A. Tudi, M. Mutailipu, F. F. Zhang, Z. H. Yang and S. L. Pan, Natl. Sci. Rev., 2022, 9, nwac110 CrossRef CAS PubMed.
  27. B. L. Cheng, W. J. Ma, A. Tudi, F. F. Zhang, X. F. Long, S. L. Pan and Y. Yang, Mater. Today Chem., 2025, 45, 102623 CrossRef CAS.
  28. J. Tauc, R. Grigorovici and A. Vancu, Phys. Status Solidi, 1966, 15, 627–637 CrossRef CAS.
  29. (a) CCDC 2519961: Experimental Crystal Structure Determination, 2026,  DOI:10.25505/fiz.icsd.cc2ql72n; (b) CCDC 2519962: Experimental Crystal Structure Determination, 2026,  DOI:10.25505/fiz.icsd.cc2ql73p.

Footnote

These authors contributed equally to this work.

This journal is © The Royal Society of Chemistry 2026
Click here to see how this site uses Cookies. View our privacy policy here.