English

• All-solid-state lasers employing second-harmonic generation (SHG) are crucial for expanding output wavelengths and achieving high-energy laser output, playing a vital role in laser frequency conversion and optical communication [1-9]. Consequently, high-performance nonlinear optical (NLO) crystals operating in the short-wave ultraviolet (UV) region (200–300 nm) have garnered significant research interest [10-13]. However, to date, only a few UV NLO crystals like KH₂PO₄ (KDP), β-BaB₂O₄ (BBO), LiB₃O₅ (LBO) have been commercialized, and these materials face a series of challenges in practical application [14-17]. Although KDP exhibits good UV transmission, its SHG response is modest (d₃₆ ≈ 0.39 pm/V) [18]. As for BBO, while pursuing strong SHG response, the birefringence of the material is often too large, especially when the π-conjugated group is the NLO active unit. The BBO crystal, based on π-conjugated [B₃O₆] group is a typical example. Its large birefringence (Δn≈0.12@546 nm) leads to a large walk-off angle, which limits its conversion efficiency under high power conditions [19-21]. Consequently, the design and synthesis of novel short-wave UV NLO materials that simultaneously exhibit strong SHG response (>1.0×KDP), broad ultraviolet transparency (λ_cutoff<300 nm), moderate birefringence (Δn = 0.05–0.10), and high stability has emerged as a critical challenge.

  Recently, organic-inorganic hybrid NLO crystals containing cyclic planar π-conjugated groups have received widespread attention from researchers [22-25]. Cyclic planar π-conjugated groups (e.g., C₅H₆NO⁺, C₃N₃O₃^-) typically possess large HOMO-LUMO gaps, polarization anisotropy (δ), and hyperpolarization (χ²), rendering them classic UV NLO-active groups [26,27], with numerous high-performance NLO crystals have been reported, such as (C₅H₆ON)(H₂PO₄) (3.0×KDP, Δn_cal = 0.25@1064 nm) [28], (C₇H₄NO₄)(IO₃) (3.6×KDP, Δn = 0.35@546 nm) [29] and 2(C₃H₇N₆)·2Cl·H₂O (4.3×KDP, Δn = 0.28@546 nm) [30]. Nevertheless, despite their excellent SHG performance, the large birefringence stemming from strong π-π interactions and the substantial polarization anisotropy of organic π-conjugated rings can result in excessive walk-off angles, thereby limiting their effective application. Consequently, non-π-conjugated organic groups have gained considerable interest, leading to a series of notable NLO crystals, including [Me₃NCH₂Cl]CdCl₃ (0.7×KDP) [31], H₁₁C₄N₂CdI₃ (6.0×KDP) [32] and (PIP)·CdI₃(HDABCO) (2.1×KDP) [33]. Among these, 1,4-diazabicyclo[2.2.2]octane (DABCO) has aroused our interest due to its rigid bridge ring skeleton and easy growth of single crystals, and has successfully obtained a new (H₂DABCO)CdI₄ in the preliminary work [33]. However, the inherent small polarization anisotropy (δ = 2.05) and hyperpolarization (χ² = 9.27, Fig. S1a in Supporting information) caused by non π-conjugated (H₂DABCO)²⁺ cation leads to a small birefringence (0.02@546 nm), which result in non-phase matching under 1064 nm radiation. Phase-matching (PM) requires a zero wavevector difference (Δk) between fundamental and second-harmonic waves. Typically, crystal birefringence (Δn) determined by optical anisotropy can be used to compensate for Δk and achieve PM [34-37]. Therefore, we have focused on a similar structures: 1-Azabicyclo[2.2.2]octan-3-one (3-QUO, C₇H₁₁NO). Compared with (H₂DABCO)²⁺, (3-QUO)⁺ has a moderate HOMO-LUMO gap (6.09 eV, Fig. S1b in Supporting information), and the introduction of π-conjugated C = O groups significantly enhances polarization anisotropy (δ = 13.12) and hyperpolarization (χ² = 69.28) through electron delocalization effect, which can optimize the birefringence for phase matching under 1064 nm radiation (Fig. 1a). Moreover, the n→π* transition and localized π→π* leaps facilitate the generation of fluorescence, providing favorable conditions for the design of multifunctional optical crystals.

  Figure 1

  

  Figure 1.  The comparison of the calculated hyperpolarizability, polarizability anisotropy and the HOMO-LUMO gap of (a) the representative NLO active (non-)π-conjugated organic groups, (b) views of the [(3-QUO)₂MX₄] unit, (c) overall architecture of (3-QUO)₂MX₄.

  DownLoad: Full-Size Img PowerPoint

  Following this rationale, we initially designed and synthesized a new organic-inorganic metal halide, (3-QUO)₂CdI₄. It is speculated that it may have strong SHG response and be able to achieve phase matching under 1064 nm radiation, but the opposite is true (non-phase matching, 0.9×KDP). Considering the strong dispersion of heavy atoms Cd and I, which may cause a red shift in the phase matching wavelength, we replaced Cd²⁺ with Zn²⁺ and successfully obtained a new isomorphic (3-QUO)₂ZnI₄, but still exhibited non-phase matching. Nonetheless, we noticed a blue shift in its theoretical Type-I phase matching wavelength (from 544 nm to 510 nm). Previous studies have demonstrated that manipulating normal dispersion presents a viable alternative for achieving PM [34,38]. Since the refractive index demonstrates stronger frequency-dependent enhancement near the bandgap (E_g) than in off-resonance regions, normal dispersion can be reduced by enlarging E_g in NLO materials. This bandgap-widening strategy enables effective dispersion modulation to attain phase matching at shorter wavelengths. Therefore, it is reasonable to infer that halogen substitution can increase the bandgap and induce a blueshift in the PM wavelength. And we propose a halogen-substitution-mediated strategy to achieve phase matching in the ultraviolet band, resulting in a dual increase in bandgap and SHG response.

  We further substituted I with Br based on the two iodine compounds synthesized previously, and successfully obtained two new (3-QUO)₂ZnBr₄ and (3-QUO)₂CdBr₄ which can achieve short-wave ultraviolet phase matching. Remarkably, compared to their iodide counterparts, both bromides exhibit shorter UV absorption edges and wider bandgaps ((3-QUO)₂ZnBr₄: 220 nm, 5.10 eV; (3-QUO)₂CdBr₄: 245 nm, 4.55 eV) as well as stronger phase-matchable SHG responses (1.5×KDP and 1.8×KDP, respectively). Additionally, they exhibit yellowish-white fluorescence emissions under the ultraviolet light irradiation. The fluorescence excitation peaks for both (3-QUO)₂ZnBr₄ and (3-QUO)₂CdBr₄ are located in the blue light region, suggesting their potential for developing single-component, blue-light-excitable white-light-emitting materials.

  (3-QUO)₂MX₄ (M = Zn, Cd; X = Br, I) has a similar structure, and all four compounds crystallize in the polar and non-centrosymmetric (NCS) space group Pmn2₁ (Tables S1-S5 in Supporting information). Their asymmetric units all consists of one C₇H₁₂NO⁺ (3-QUO)⁺ cation, half of a M atom and two X atoms. In (3-QUO)⁺ cations, except for the C atom on the carbonyl group which is sp² hybridized, all other C and N atoms are sp³ hybridized, and the C–C/C–N bond length and C = O bond length are consistent with the previously reported compounds (Table S3). Each M²⁺ cation is connected to four halogen atoms, forming distorted ZnBr₄, ZnI₄, CdBr₄, CdI₄ tetrahedra, and each [MX₄]^2- anion is connected to two (3-QUO)⁺ cations through hydrogen bonds N—H^…X to balance charges, forming a [(3-QUO)₂MX₄] unit (Fig. 1b). Neighboring [(3-QUO)₂MX₄] units are isolated from each other and arranged in a quasi-two-dimensional (quasi-2D) [(3-QUO)MX₄]_∞ layer (Fig. S2 in Supporting information), parallel to the bc plane. These pseudo-layers stack anti-parallel along the a-axis to form the entire 3D network (Fig. 1c).

  Powder X-ray diffraction analysis (Fig. S3 in Supporting information) and the refined PXRD data (Fig. S4 in Supporting information) confirmed the purity of the synthesized samples. Field-emission scanning electron microscopy (FESEM) analyses detected the presence of C, N, O, Cd/Zn and Br/I elements in (3-QUO)₂MX₄, the ratio of each element is consistent with structural calculations (Fig. S5 in Supporting information). The infrared spectra (Fig. S6 and Table S6) for title compounds show the C = O stretching vibration peaks around 1740 cm^-1, as well as the absorption peaks between 2900 cm^-1 and 3000 cm^-1 corresponding to N—H and C–H stretching vibrations. The peaks between 800 cm^-1 and 1500 cm^-1 correspond to the stretching and bending vibrations of C–C/C–N bonds on the quinine ring skeleton, with the strong peak near 1364 cm^-1 corresponding to the C–N stretching vibration and the quinine ring skeleton vibration corresponding to the peak near 445 cm^-1. TG-DTA curves indicate that all four title compounds are stable up to at least 260℃, with good thermal stability (Fig. S7 in Supporting information).

  Analysis of the ultraviolet-visible (UV–vis) absorption spectra reveals a systematic variation in optical properties across the series. The absorption edge decreases in wavelength in the order: (3-QUO)₂CdI₄ (294 nm) > (3-QUO)₂ZnI₄ (269 nm) > (3-QUO)₂CdBr₄ (245 nm) > (3-QUO)₂ZnBr₄ (220 nm) (Fig. 2a). Conversely, the optical bandgap energy derived via Kubelka-Munk transformation increases in the order: (3-QUO)₂CdI₄ (3.75 eV) < (3-QUO)₂ZnI₄ (3.94 eV) < (3-QUO)₂CdBr₄ (4.55 eV) < (3-QUO)₂ZnBr₄ (5.10 eV) (Fig. 2b). This tunability, spanning the UV region from 294 nm to 220 nm (3.75–5.10 eV). The band gaps of (3-QUO)₂CdBr₄ and (3-QUO)₂ZnBr₄ have significant advantages over some of OIMHs containing cyclic π-conjugated organic cations, such as C₆H₁₁N₂PbCl₃ (3.87 eV) [39], (C₃N₆H₇)(C₃N₆H₆)HgCl₃ (4.40 eV) [22], (C₆H₁₀N₂)ZnX₄·H₂O (X = Cl/Br, 3.94 and 3.84 eV) [40], (C₁₃N₃H₁₄)₂MBr₄ (M = Zn/Cd, 3.96 and 3.98 eV) [41], (C₁₀H₁₁N₃)PbX₄ (X = Cl/Br, 3.35 and 2.95 eV) [42]. The larger the bandgap, the shorter the UV transmission band, which is beneficial for the application of high-energy UV NLO crystals, highlights the potential values of (3-QUO)₂CdBr₄ and (3-QUO)₂ZnBr₄.

  Figure 2

  

  Figure 2.  (a) UV–vis diffuse reflectance spectra and (b) bandgaps for title compounds. (c) Phase-matching curves and (d) oscilloscope traces of the SHG signals (150–210 µm) with 1064 nm laser radiation. KDP was used as references for the SHG measurements. (e) The photoluminescence excitation/emission spectra of (3-QUO)₂MBr₄ (M = Zn, Cd). (f) The CIE coordinates of (3-QUO)₂MBr₄ (M = Zn, Cd).

  DownLoad: Full-Size Img PowerPoint

  The SHG responses were evaluated by the Kurtz-Perry method under a laser irradiation at 1064 nm with KDP sieved as standard reference. The relationship between the SHG intensity and the particle size curve indicates the type-Ⅰ phase-matchable behavior of (3-QUO)₂ZnBr₄ and (3-QUO)₂CdBr₄, while the non-phase-matchable behavior of (3-QUO)₂ZnI₄ and (3-QUO)₂CdI₄ (Fig. 2c). For (3-QUO)₂ZnI₄ and (3-QUO)₂CdI₄, they exhibit moderate but non-phase-matchable SHG efficiency of 0.6 times and 0.9 times that of KDP, respectively (Fig. 2d). For (3-QUO)₂ZnBr₄ and (3-QUO)₂CdBr₄, they achieved relatively large SHG responses at the size of 150–210 µm, which were 1.5 times and 1.8 times that of KDP, respectively (Fig. 2d). Their SHG efficiencies are superior to some of NLO crystals containing organic planar π-conjugated moieties or MX₄ tetrahedral groups, such as (2-AMP)₂BiBr₇·H₂O (0.3×KDP) [43], (C₄H₁₀NO)PbBr₃ (0.8×KDP) [44], (C₆H₅N₂)₂ZnCl₄ (1.2×KDP) [45], (C₆H₁₀N₂)ZnCl₄·H₂O (1.3×KDP) [40], (C₁₃N₃H₁₄)₂ZnBr₄ (1.12×KDP)[41] and (C₁₃N₃H₁₄)₂CdBr₄ (0.98×KDP) [41].

  Under UV light irradiation, all four title compounds exhibit yellowish-white fluorescence emission (Fig. S8 in Supporting information). The photoluminescence (PL) and photoluminescence excitation (PLE) spectra of (3-QUO)₂CdBr₄ and (3-QUO)₂ZnBr₄ were shown in Fig. 2e. Both (3-QUO)₂CdBr₄ and (3-QUO)₂ZnBr₄ exhibit broad emission peaks ranging from 400 nm to 700 nm, with dual PLE peaks at 440 and 330 nm. The Commission Internationale de l’Eclairage (CIE) color coordinates of (3-QUO)₂CdBr₄ and (3-QUO)₂ZnBr₄ are (0.4075, 0.5327) and (0.3703,0.4216), respectively (Fig. 2f). It is worth noting that the optimal excitation peaks for both (3-QUO)₂CdBr₄ and (3-QUO)₂ZnBr₄ are located in the blue light region. Compared with UV-light-excitable materials, blue-light-excitable materials have lower energy consumption and less damage to the encapsulant, which is significant to solid-state lighting application [46,47]. Therefore, this type of OIMH holds potential for the development of a series of single-component white-light-emitting materials excited by blue light. Meanwhile, since all four compounds can maintain good thermal stabilities (Fig. S7), which would be beneficial for their further applications.

  The experimental birefringence (Δn_exp) of four title compounds was measured, with the single crystals in original and complete extinction states shown in Fig. 3a. As a result, the Δn_exp was determined to be 0.06, 0.07, 0.05 and 0.05 at 546 nm for (3-QUO)₂ZnBr₄, (3-QUO)₂ZnI₄, (3-QUO)₂CdBr₄, and (3-QUO)₂CdI₄, respectively (see Supporting information). Furthermore, to understand the phase-matching (PM) capabilities, the theoretical birefringence can be obtained via the calculation of the refractive index, with the values of 0.04 for (3-QUO)₂ZnBr₄ and 0.05 for others at 546 nm, respectively (Figs. 3b-e). These theoretical values show good agreement with the experimental results. The shortest calculated wavelengths for type-Ⅰ phase-matched second-harmonic generation have also been shown in Figs. 3b-e, which are approximately 405, 510, 442 and 544 nm for (3-QUO)₂ZnBr₄, (3-QUO)₂ZnI₄, (3-QUO)₂CdBr₄, and (3-QUO)₂CdI₄, respectively. Since the SHG wavelength corresponding to the 1064 nm fundamental laser used in experiments is 532 nm, these calculations predict that all four title compounds should be phase-matchable. However, according to the experimental results, (3-QUO)₂ZnI₄ and (3-QUO)₂CdI₄ exhibit non-phase-matchable behavior, which might be due to the fact that their theoretical phase-matching wavelengths have approached 532 nm. Conventional studies on isostructural OIMHs NLO crystals suggest that halogen substitution from Cl to Br to I typically yields a sequential enhancement in SHG response accompanied by bandgap narrowing, thereby limiting their operational spectral ranges [42,48,49]. Notably, in the isomorphic OIMHs systems (X = Br/I), the situation where bromides are phase-matched while iodides are not is extremely rare, as exemplified by (MDA)Pb₂X₆ (X = Br, I) [50]. While birefringence (Δn) conventionally compensates for Δk in crystals, its inherent optical anisotropy limits direct tunability. In this study, since the title compounds have almost the same birefringence, the bandgap widening caused by halogen substitution plays a major role in the phase matching process. This phenomenon fully validates the halogen substitution-mediated strategy, which through bromine replacement of iodine that substantially enlarges bandgaps while blue-shifting phase-matching wavelengths. This behavior originates from broader bandgaps producing flatter normal dispersion profiles, which shift the intersection point of n(ω) and n(2ω) toward shorter wavelengths (blue-shifted).

  Figure 3

  

  Figure 3.  (a) Original and complete extinction crystals for title compounds. Calculated birefringence and refractive index dispersion curves for fundamental and second-harmonic light of (b) (3-QUO)₂ZnBr₄, (c) (3-QUO)₂ZnI₄, (d) (3-QUO)₂CdBr₄ and (e) (3-QUO)₂CdI₄. The type-Ⅰ phase matching wavelengths in different planes were evaluated based on the calculated refractive index, in which we consider the type-Ⅰ phase matching condition of n(ω) = n(2ω).

  DownLoad: Full-Size Img PowerPoint

  DFT calculations were employed to elucidate the origin of optical properties. It has indicated that both (3-QUO)₂ZnBr₄ and (3-QUO)₂CdBr₄ are indirect bandgaps, with bandgaps of 3.18 and 3.32 eV, respectively; whereas (3-QUO)₂ZnI₄ and (3-QUO)₂CdI₄ are direct band gaps, with bandgaps of 2.68 and 2.81 eV, respectively (Fig. S9a in Supporting information). As anticipated, these values underestimate the experimental bandgaps, a common characteristic of the GGA-PBE functional. To correct for this underestimation and better align the calculated electronic structure with experimental observations, scissor operators of 1.92 eV for (3-QUO)₂ZnBr₄, 1.26 eV for (3-QUO)₂ZnI₄, 1.23 eV for (3-QUO)₂CdBr₄ and 0.94 eV for (3-QUO)₂CdI₄ were applied, respectively. Naturally, larger experimental bandgaps necessitate larger scissor operators. Our calculations reveal a clear correlation: as the scissor operator magnitude increases, the shortest phase-matching wavelength systematically blueshifts (Fig. S10 in Supporting information). This provides a powerful theoretical insight: for an isostructural series of compounds, increasing the bandgap via simple elemental substitution (e.g., metal or halogen tuning) directly enables a shorter phase-matching wavelength without altering the fundamental crystal packing (Fig. S11 in Supporting information). Consequently, these results theoretically validate the central innovation of our work—substituting bromine for iodine effectively yields organic-inorganic metal halides with both a wider bandgap and a more favorable phase-matching range in the short-wavelength UV region.

  Additionally, the total and partial density of states (TDOS and PDOS) was conducted to clarify orbital contributions near the band edges (Fig. S9b in Supporting information). These analyses reveals distinct halogen-dependent characteristics in the band edge compositions. For bromide systems, the valence band maxima (VBM) primarily arise from Br-4p orbitals hybridized with transition metal d-states (Zn/Cd-3d) and C-2p orbitals. Conversely, the conduction band minima (CBM) in both bromides exhibit dominant contributions from C-2p and O-2p orbitals. In their iodide analogues, I-5p orbitals become the principal component of VBM alongside Zn/Cd-3d and C-2p orbitals, while the CBM maintains similar C/O-2p character as observed in the bromide counterparts. This electronic structure evolution highlights the critical role of halogen p-states in modulating VBM energetics without altering CBM orbital composition. Further confirmed that for all four compounds, both the inorganic [MX₄]^2- tetrahedrons and the organic (3-QUO)⁺ cations synergistically contribute to the optical properties. Theoretical calculations also indicate that in (3-QUO)₂CdBr₄, the highest occupied molecular orbital (HOMO) is primarily composed of p-orbitals from the Br atoms of the inorganic framework, while the lowest unoccupied molecular orbital (LUMO) is localized on the non-conjugated framework of the organic ligand (Fig. S12 in Supporting information). Thus, molecular orbital analysis attributes the emission to halide-to-ligand charge transfer (XLCT) [51]. The fluorescence characteristics and the HOMO-LUMO orbitals of the other three compounds are similar, and their luminescence mechanisms might be attributed to XLCT.

  Due to the strong SHG responses, wide bandgaps, and phase matching of the two Br compounds, we further analyzed their SHG origins. Typically, in NLO crystals, the greater the net dipole moment, the stronger the SHG response [52,53]. Thus, we have calculated the dipole moments of the inorganic units and the net dipole moments of the unit cells using a simple bond-valence approach (Table S7 in Supporting information). The calculated dipole moment of the [CdBr₄]^2- tetrahedron in (3-QUO)₂CdBr₄ (4.71 D) is significantly larger than that of the [ZnBr₄]^2- tetrahedron in (3-QUO)₂ZnBr₄ (3.20 D). More importantly, due to the approximate crystal structure, the three-dimensional stacked [MBr₄] tetrahedron (M=Cd and Zn) mainly synergistically stack along the c-axis direction, leading to a significant difference in the net dipole moments per unit cell: 11.31 D for (3-QUO)₂CdBr₄ compared to 7.40 D for (3-QUO)₂ZnBr₄, which is consistent with the experimental SHG responses. For most NLO materials, the dipole moment per unit cell normally positively correlating with the SHG intensity (e.g., the net dipole moments per unit cell: 4-HPYSOF (68.36 D) and 4-APSF (11.74 D), corresponding to SHG responses of 12 and 2×KDP, respectively [24]; S-[Cd₂(SIAP)₂(bpy)₂] (9.62 D) and S-[Zn₂(SIAP)₂(bpy)₂] (7.28 D), corresponding to SHG responses of 2.1 and 1.5×KDP, respectively [54]).

  In summary, we present a halogen-substitution strategy to unlock short-wavelength UV phase matching in OIMHs by replacing iodine with bromine in the (3-QUO)₂MX₄ system. The resultant bromide compounds, (3-QUO)₂ZnBr₄ and (3-QUO)₂CdBr₄, exhibit significantly widened bandgaps (5.10 and 4.55 eV), blue-shifted UV absorption edges (220 nm and 245 nm), and enhanced phase-matchable SHG responses (1.5 and 1.8×KDP), outperforming their iodide counterparts. These materials also demonstrate moderate birefringence (Δn ≈ 0.05–0.06) and unique blue-light-excitable yellowish-white fluorescence, highlighting their dual functionality for nonlinear optics and solid-state lighting. This work establishes a rational approach to simultaneously optimize bandgap, SHG efficiency, and phase-matching capabilities, offering a promising pathway for designing advanced UV nonlinear optical crystals with balanced performance for next-generation photonic technologies.

  CRediT authorship contribution statement

  Ming-Chang Wang: Writing – original draft, Formal analysis, Data curation. Zhi Lin: Writing – original draft. Jia-Jia Li: Visualization. Jia-Min Lian: Visualization, Data curation. Yun-Xia Hu: Visualization, Data curation. Yan Chen: Writing – review & editing, Supervision. Ke-Zhao Du: Writing – review & editing, Supervision, Conceptualization. Jin Chen: Writing – review & editing, Conceptualization.

  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

  Our work has been supported by the National Natural Science Foundation of China (Nos. 22205037 and 22373014) and the Natural Science Foundation of Fujian Province (No. 2023J01498).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111639.

  
  
• 1. [1]

     X. Dong, L. Huang, G. Zou, Acc. Chem. Res. 58 (2025) 150–162. doi: 10.1021/acs.accounts.4c00704

  2. [2]

     P.S. Halasyamani, Nat. Photon. 17 (2023) 639–640. doi: 10.1038/s41566-023-01255-4

  3. [3]

     A.D. Poletayev, M.C. Hoffmann, J.A. Dawson, et al., Nature 625 (2024) 691–696. doi: 10.1038/s41586-023-06827-6

  4. [4]

     Y. Yang, J.W. Henke, A.S. Raja, et al., Science 383 (2024) 168–173. doi: 10.1126/science.adk2489

  5. [5]

     L. Gao, J. Chen, X. Shi, et al., Sci. Adv. 10 (2024) eadr2389. doi: 10.1126/sciadv.adr2389

  6. [6]

     Y. Zhou, L.T. Jiang, X.M. Jiang, et al., Chin. Chem. Lett. 36 (2025) 109740. doi: 10.1016/j.cclet.2024.109740

  7. [7]

     C. Hu, M. Wu, J. Han, et al., Chem. Commun. 60 (2024) 2653–2656. doi: 10.1039/d3cc05784d

  8. [8]

     J.H. Wu, B. Zhang, T.K. Jiang, et al., Chin. J. Struct. Chem. 42 (2023) 100016.

  9. [9]

     L. Gao, Y. Han, C. Lin, et al., Chin. Chem. Lett. 35 (2024) 109529. doi: 10.1016/j.cclet.2024.109529

  10. [10]

      H. Qiu, F. Li, C. Jin, et al., Angew. Chem. Int. Ed. 63 (2024) e202316194. doi: 10.1002/anie.202316194

  11. [11]

      X. Zhou, X. Mao, P. Zhang, et al., Inorg. Chem. Front. 11 (2024) 3221–3228. doi: 10.1039/d4qi00791c

  12. [12]

      H. Zhao, J. Song, S. Li, et al., Chin. Chem. Lett. 37 (2026) 111011. doi: 10.1016/j.cclet.2025.111011

  13. [13]

      W. Wang, D. Mei, S. Wen, et al., Chin. Chem. Lett. 33 (2022) 2301–2315. doi: 10.1016/j.cclet.2021.11.089

  14. [14]

      G. Ghosh, Opt. Commun. 163 (1999) 95–102. doi: 10.1016/S0030-4018(99)00091-7

  15. [15]

      H.T. Luo, T. Tkaczyk, E.L. Dereniak, et al., Opt. Lett. 31 (2006) 616–618. doi: 10.1364/OL.31.000616

  16. [16]

      Y. Tian, W. Zeng, X. Dong, et al., Angew. Chem. Int. Ed. 63 (2024) e202409093. doi: 10.1002/anie.202409093

  17. [17]

      L.Q. Yang, X.M. Jiang, Y. Chen, et al., Adv. Opt. Mater. 12 (2024) 2301897. doi: 10.1002/adom.202301897

  18. [18]

      L. Ye, Z.L. Fang, M.Y. Guo, et al., Chin. Chem. Lett. 37 (2025) 2026. doi: 10.1016/j.cclet.2025.110826

  19. [19]

      H. Qiu, F. Li, Z. Li, et al., J. Am. Chem. Soc. 145 (2023) 24401–24407. doi: 10.1021/jacs.3c09573

  20. [20]

      S. Wu, G. Wang, J. Xie, et al., J. Cryst. Growth 245 (2002) 84–86. doi: 10.3901/JME.2002.11.084

  21. [21]

      J. Wang, H. Wu, Z. Hu, et al., J. Mater. Chem. C 11 (2023) 1320–1328. doi: 10.1039/d2tc04914g

  22. [22]

      Z. Bai, J. Lee, H. Kim, et al., Small 19 (2023) 2301756. doi: 10.1002/smll.202301756

  23. [23]

      P. Zhang, X. Mao, X. Dong, et al., Chin. Chem. Lett. 35 (2023) 109235.

  24. [24]

      L. Qi, X. Jiang, K. Duanmu, et al., J. Am. Chem. Soc. 146 (2024) 9975–9983. doi: 10.1021/jacs.4c00666

  25. [25]

      Y. Deng, F. Li, Z. Zhou, et al., Chin. Chem. Lett. 35 (2024) 109085. doi: 10.1016/j.cclet.2023.109085

  26. [26]

      C. Shen, J. Liu, D. Sun, et al., Small 21 (2025) 2501471. doi: 10.1002/smll.202501471

  27. [27]

      C. Yang, J. Gou, Y. Zhu, et al., Angew. Chem. Int. Ed. 64 (2025) e202420810. doi: 10.1002/anie.202420810

  28. [28]

      J. Lu, X. Liu, M. Zhao, et al., J. Am. Chem. Soc. 143 (2021) 3647–3654. doi: 10.1021/jacs.1c00959

  29. [29]

      M.-B. Xu, Q.Q. Chen, B.X. Li, et al., Chin. Chem. Lett. 36 (2024) 110513. doi: 10.1016/j.cclet.2024.110513

  30. [30]

      L. Liu, C.L. Hu, Z. Bai, et al., Chem. Commun. 56 (2020) 14657–14660. doi: 10.1039/d0cc06558g

  31. [31]

      C. Shen, J. Liu, K. Wu, et al., CrystEngComm. 25 (2023) 2264–2270. doi: 10.1039/d3ce00145h

  32. [32]

      H.Y. Wu, C.L. Hu, M.B. Xu, et al., Chem. Sci. 14 (2023) 9533–9542. doi: 10.1039/d3sc03052k

  33. [33]

      M.C. Wang, M.F. Zhuo, Z. Lin, et al., J. Mater. Chem. C 13 (2025) 11651–11660. doi: 10.1039/d5tc01362c

  34. [34]

      B.W. Liu, H.Y. Zeng, X.M. Jiang, et al., CCS Chem. 3 (2020) 964–973. doi: 10.1002/pi.6050

  35. [35]

      Z. Yang, A. Tudi, B.H. Lei, et al., Sci. China Mater. 63 (2020) 1480–1488. doi: 10.1007/s40843-020-1279-6

  36. [36]

      X. Jiang, S. Luo, L. Kang, et al., ACS Photon. 2 (2015) 1183–1191. doi: 10.1021/acsphotonics.5b00248

  37. [37]

      M. Mutailipu, J. Han, Z. Li, et al., Nat. Photon. 17 (2023) 694–701. doi: 10.1038/s41566-023-01228-7

  38. [38]

      M.Y. Li, X. Liu, R.X. Wang, et al., Angew. Chem. Int. Ed. 64 (2025) e202502143. doi: 10.1002/anie.202502143

  39. [39]

      Y. Kang, C. Yang, J. Zhang, et al., Chin. Chem. Lett. 37 (2026) 111385. doi: 10.1016/j.cclet.2025.111385

  40. [40]

      C. Shen, C. Zhang, Q. Xing, et al., Chem. Eng. J. 509 (2025) 161389. doi: 10.1016/j.cej.2025.161389

  41. [41]

      J. Wu, Y.F. Fu, W. Liu, et al., Inorg. Chem. Front. 11 (2024) 7090–7097. doi: 10.1039/d4qi01729c

  42. [42]

      L.L. Zhang, Q. Ding, P. Wang, et al., Inorg. Chem. Front. 11 (2024) 3618–3625. doi: 10.1039/d4qi00832d

  43. [43]

      D. Sun, D. Wang, H. Chen, et al., Inorg. Chem. 61 (2022) 15247–15255. doi: 10.1021/acs.inorgchem.2c02433

  44. [44]

      C. Shen, D. Sun, Y. Dang, et al., Inorg. Chem. 61 (2022) 16936–16943. doi: 10.1021/acs.inorgchem.2c03042

  45. [45]

      D.X. Yang, Y.L. Lv, J.D. Guo, et al., Inorg. Chem. 64 (2025) 3643–3648. doi: 10.1021/acs.inorgchem.5c00076

  46. [46]

      Y.P. Lin, S. Hu, J. Xu, et al., Chem. Eng. J. 468 (2023) 143818. doi: 10.1016/j.cej.2023.143818

  47. [47]

      Y. Zhong, Y.E. Huang, T. Deng, et al., Inorg. Chem. 60 (2021) 17357–17363. doi: 10.1021/acs.inorgchem.1c02840

  48. [48]

      Y. Zhu, J. Gou, C. Yang, et al., Angew. Chem. Int. Ed. 64 (2025) e202509290. doi: 10.1002/anie.202509290

  49. [49]

      D. Chen, S. Hao, L. Fan, et al., Chem. Mater. 33 (2021) 8106–8111. doi: 10.1021/acs.chemmater.1c02896

  50. [50]

      M.A. Cassingham, Y.G. Goh, E.T. McClure, et al., ACS Appl. Mater. Interfaces 15 (2023) 18006–18011. doi: 10.1021/acsami.2c20648

  51. [51]

      Y. Shen, B. Wang, P. Wang, et al., Inorg. Chem. 63 (2024) 12073–12080. doi: 10.1021/acs.inorgchem.4c01047

  52. [52]

      H. Yu, M.L. Nisbet, K.R. Poeppelmeier, J. Am. Chem. Soc. 140 (2018) 8868–8876. doi: 10.1021/jacs.8b04762

  53. [53]

      Y. Liu, Y.P. Gong, S. Geng, et al., Angew. Chem. Int. Ed. 61 (2022) e202208875. doi: 10.1002/anie.202208875

  54. [54]

      J. Lee, J.B. Cho, Y. Li, et al., Small 20 (2024) 2309323. doi: 10.1002/smll.202309323

• 

  Figure 1  The comparison of the calculated hyperpolarizability, polarizability anisotropy and the HOMO-LUMO gap of (a) the representative NLO active (non-)π-conjugated organic groups, (b) views of the [(3-QUO)₂MX₄] unit, (c) overall architecture of (3-QUO)₂MX₄.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) UV–vis diffuse reflectance spectra and (b) bandgaps for title compounds. (c) Phase-matching curves and (d) oscilloscope traces of the SHG signals (150–210 µm) with 1064 nm laser radiation. KDP was used as references for the SHG measurements. (e) The photoluminescence excitation/emission spectra of (3-QUO)₂MBr₄ (M = Zn, Cd). (f) The CIE coordinates of (3-QUO)₂MBr₄ (M = Zn, Cd).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Original and complete extinction crystals for title compounds. Calculated birefringence and refractive index dispersion curves for fundamental and second-harmonic light of (b) (3-QUO)₂ZnBr₄, (c) (3-QUO)₂ZnI₄, (d) (3-QUO)₂CdBr₄ and (e) (3-QUO)₂CdI₄. The type-Ⅰ phase matching wavelengths in different planes were evaluated based on the calculated refractive index, in which we consider the type-Ⅰ phase matching condition of n(ω) = n(2ω).

  下载: 全尺寸图片 幻灯片
•