Synergetic defect and local structure engineering to boost photocatalytic activity of ZnIn_2−xCu_xS₄ nanosheets for H₂O₂ production

Abstract

Photocatalytic generation of H₂O₂ has attracted considerable attention because of its environmental benignity and economic merit. However, for the commercialization of photocatalytic H₂O₂ production, it is necessary to improve the activity and selectivity of noble-metal-free photocatalysts for the reduction of O₂ to H₂O₂. In this study, we developed a synergetic defect and local structure engineering approach to enhance the photocatalytic performance of transition-metal sulfides toward H₂O₂ production via simultaneous Cu substitution and exfoliation of ZnIn₂S₄. Combined Cu substitution and exfoliation allowed the introduction of considerable S vacancies and regulated the local structural distortion and electronic configuration. The Cu-substituted ZnIn_2−xCu_xS₄ nanosheets exhibited significantly enhanced photocatalytic activity for hydrogen peroxide production compared to pristine ZnIn₂S₄ and Ni-substituted ZnIn_2−xNi_xS₄ nanosheets. The high efficacy of Cu substitution–exfoliation in optimizing the photocatalytic activity was ascribed to the increase in S vacancies, enhancement of tetragonal distortion around the Cu substituent, and regulation of the electronic structure, which enhanced O₂ adsorption, increased visible-light absorptivity, prevented charge recombination, and improved the charge transfer and H₂O₂ production kinetics. This defect and local structure engineering strategy provides an effective means of developing highly efficient metal chalcogenide photocatalysts.

---

Introduction

Photocatalytic H₂O₂ generation via the partial reduction of atmospheric O₂ gas has sparked intense research efforts owing to its significant advantages in energy consumption and production cost.^1,2 The photocatalyst-assisted production of H₂O₂ under ambient conditions can circumvent the deepening crisis of global climate change and depletion of fossil fuels.^3,4 The anthraquinone process has been commercialized as the most effective production method for hydrogen peroxide.^5,6 However, despite the high efficiency of this process, the necessity of H₂ gas as a reactant increases the production cost and the complexity of the production procedure.^7,8 To circumvent the drawbacks of the anthraquinone process, it is necessary to develop economically feasible noble-metal-free photocatalysts that are highly active for the reduction of O₂ gas to H₂O₂.⁹ In recent years, semiconducting metal chalcogenides such as ZnIn₂S₄ have attracted attention as promising photocatalysts for H₂O₂ production because of their low price and high activity.^10,11 To further enhance the photocatalytic performance of ZnIn₂S₄, diverse approaches have been exploited, such as morphology control, chemical doping, and cocatalyst deposition.^12–14 Considering the weak van der Waals-type interlayer interaction between ZnIn₂S₄ layers, liquid exfoliation of this material can provide an efficient route for producing monolayered ZnIn₂S₄ nanosheets in which the surface area-to-volume ratio and the number of surface reaction sites are maximized.^15,16 The reduction in lattice energy caused by exfoliation may facilitate the introduction of crystal defects, enhancing the attachment of O₂ molecules and surface reactivity.¹⁷ Alternatively, aliovalent substitution of In³⁺ ions with divalent metal ions such as Cu²⁺, Ni²⁺, and Mn²⁺ would offer a useful method for the controlled introduction of anion vacancies into the ZnIn₂S₄ lattice. Furthermore, the substitution-induced regulation of the local structure and electronic configuration of ZnIn₂S₄ can enhance the photocatalytic performance of ZnIn₂S₄via the optimization of light absorptivity, charge recombination, and charge-transport behaviors. In this regard, the Jahn–Teller-active Cu²⁺ ion is considered as a relevant substituent that induces local structural distortion in the ZnIn₂S₄ lattice and thus regulates the interaction with the catalysis reactant/product. The benefit of substitution-assisted defect/local structure engineering can be maximized by the combination with an exfoliation approach. While only a part of introduced anion vacancies and locally distorted sites can be surface-exposed in the bulk ZnIn₂S₄ material, atomically thin thickness of exfoliated ZnIn₂S₄ nanosheets allows for the complete exposure of anion vacancies and distorted metal sites to the surface. Considering that the anion vacancies and locally distorted metal ions can function as active sites for the adhesion and reduction of O₂ molecules, the combination of defect engineering and exfoliation strategies can offer a powerful means to optimize the photocatalyst performance of ZnIn₂S₄ for the production of H₂O₂.¹⁸ Additionally, the exfoliation of ZnIn_2−xCu_xS₄ into monolayer nanosheets is expected to improve charge-/mass-transport kinetics, which is also beneficial in enhancing its photocatalytic activity. Furthermore, in situ spectroscopic investigation of the obtained ZnIn_2−xCu_xS₄ nanosheets during H₂O₂ generation would provide a crucial fundamental understanding of the impact of the aliovalent substitution–exfoliation process on the photocatalyst efficiency.

Although several studies on the exfoliation of ZnIn₂S₄ nanosheets and the cation substitution of bulk ZnIn₂S₄ have been reported,^10,19 we are not aware of any prior study on the synergistically combined defect/local structure engineering and exfoliation approach to optimize the photocatalytic activity of ZnIn₂S₄ material toward partial O₂ reduction.

In the present study, simultaneous defect and local structure engineering of atomically thin ZnIn₂S₄ nanosheets was achieved by combining aliovalent substitution and exfoliation processes to develop efficient photocatalysts for H₂O₂ production. The effects of synergistic Cu substitution and exfoliation on the defect structure, local atomic arrangement, and electronic structure of the Znin₂S₄ nanosheets were examined using combined spectroscopic and diffraction analyses. Cu-substituted ZnIn_2−xCu_xS₄ nanosheets as well as Ni-substituted homologs were employed as photocatalysts for H₂O₂ generation to investigate the advantages of synergistic substitution and exfoliation approaches on the photocatalytic activity of ZnIn₂S₄ materials. The evolutions of the local structure and chemical bonding characteristics during H₂O₂ generation were investigated using in situ X-ray absorption spectroscopy.

Experimental

Preparation

The photocatalyst materials of ZnIn_2−xCu_xS₄ nanosheets were synthesized using a hydrothermal method. First, 0.0015 mol of Zn(CH₃COO)₂·2H₂O, 0.003−x mol of InCl₃, and x mol of Cu(CH₃COO)₂·H₂O were dissolved in 250 mL of distilled water and stirred for 30 min. Then, 0.008 mol of thioacetamide was added to the mixture, followed by an additional 30 min of stirring. After complete mixing, the solution was heated to 95 °C for 5 h under reflux in the dark. The resulting precipitate was collected via centrifugation and washed several times with distilled water. As depicted in Fig. 1a, the liquid exfoliation of ZnIn_2−xCu_xS₄ into monolayer nanosheets was conducted by dispersing bulk ZnIn_2−xCu_xS₄ powder in 200 mL of distilled water, followed by ultrasonication using an Elmasonic_P_70H device, wherein an effective power of 220 W was employed for 0.5 h at room temperature. The resulting colloidal suspension was then centrifuged at 6000 rpm for 5 min to obtain the nanosheets. This experimental process offered a highly reproducible synthesis method for the exfoliated ZnIn_2−xCu_xS₄ nanosheets. In this study, the Cu ratio x was set as 0.1, 0.2, and 0.3, and the corresponding materials were denoted as Cu0.1-ZIS, Cu0.2-ZIS, and Cu0.3-ZIS, respectively. As a reference material, ZnIn₂S₄ was prepared without a Cu source, and this material was denoted as ZIS. The exfoliation yield of the present nanosheets was found to decrease with increasing Cu content. The suspension concentrations of exfoliated ZIS, Cu0.1-ZIS, Cu0.2-ZIS, and Cu0.3-ZIS nanosheets were estimated to be approximately 2.0, 1.6, 1.0, and 0.45 mg mL⁻¹, respectively. Transmission electron microscopy (TEM) analysis revealed that the monolayer percentage of the exfoliated nanosheets was about 50%. For comparison, Ni-substituted ZnIn_2−xNi_xS₄ was also synthesized using NiCl₂ as precursor.

Fig. 1 (a) Schematic for the synergetic Cu substitution–exfoliation route to produce the ZnIn_2−xS₄ nanosheets. (b) Powder XRD patterns of ZnIn_2−xCu_xS₄ materials. (c) AFM, (d) STEM, (e) photoimages, and (f) FE-SEM/EDS-elemental mapping data of ZnIn_2−xCu_xS₄ nanosheets.

Characterization

To investigate the crystal shape of ZnIn_2−xCu_xS₄ nanosheets, atomic force microscopy (AFM) images were obtained using a Park Systems (NX-10) instrument. The effects of Cu substitution and exfoliation on the crystal structures of the ZnIn_2−xCu_xS₄ materials were characterized using powder X-ray diffraction (XRD) with Ni-filtered Cu Kα radiation (Rigaku MiniFlex, λ = 1.54184 Å, 298 K). Field emission-scanning electron microscopy (FE-SEM) and scanning transmission electron microscopy (STEM) were performed using JEOL JSM-7001F and JEOL JEM-ARM200F electron microscopes, respectively. The spatial distribution of the components was determined through energy-dispersive X-ray spectroscopy (EDS)-elemental mapping analysis. X-ray photoelectron spectroscopy (XPS) analysis was performed using a Thermo VG instrument (UK) with Al Kα radiation. The measured XPS spectra were energy-referenced to the adventitious C 1s peak (284.8 eV). Before and after the photocatalytic reactions, XPS data was measured using an AXIS Supra+ (Kratos Analytical) instrument. Diffuse-reflectance ultraviolet–visible (UV-Vis) spectra were obtained using a Jasco V-760 spectrometer. Photoluminescence (PL) spectra were acquired using a PerkinElmer FL8500 fluorescence spectrometer. The size distribution of present nanosheets was recorded using an ELS-Z1000 (Otsuka Electronics). The surface areas of the nanosheets were examined using N₂ adsorption–desorption isotherms measured at 77 K using a BelSorp Mini X analyzer. X-ray absorption near-edge structure/extended X-ray absorption fine structure (XANES/EXAFS) analyses were performed at the Cu K-edge, Zn K-edge, and In K-edge using beamline 10C of the Pohang Accelerator Laboratory (PAL, Pohang, Republic of Korea). Energy calibration of the XANES spectra was performed via simultaneous measurement of the Cu K-edge, Zn K-edge, and In K-edge of Cu, Zn, and In metal foils, respectively. Micro-Raman spectroscopy was conducted using a Horiba Jobin Yvon LabRam Aramis apparatus, with a 514.5 nm laser source. Electrochemical impedance spectroscopy (EIS) measurements were performed at an open-circuit potential in the frequency range of 10⁻²–10⁵ Hz with an IVIUM analyzer to determine the charge-transport properties of the ZnIn_2−xCu_xS₄ nanosheets using a three-electrode system in a 0.5 M Na₂SO₄ solution. A saturated calomel electrode (SCE) and a Pt mesh (dimensions of 1 × 1 cm²) were utilized as the reference and counter electrodes, respectively.

Photocatalytic activity measurement

Photocatalytic H₂O₂ production was monitored using a Pyrex reactor and a 300 W Xe arc lamp (Oriel) with a UV cutoff filter (λ > 420 nm). For the photocatalytic activity tests, 50 mg of ZnIn₂S₄ and ZnIn_2−xCu_xS₄ nanosheets was prepared in a volume of 90 mL using deionized water, and 10 mL of methanol was added as a sacrificial agent. O₂ gas (99.9999%) was purged for 30 min under dark conditions, followed by 1 h of irradiation. To detect the H₂O₂ product, 1 mL of the reacted solution was extracted every 15 min and filtered with a 0.2 μm pore syringe filter. The photogenerated H₂O₂ was detected using the Fe²⁺/Fe³⁺ detection method.²⁰ 1 mL of the reaction mixture was mixed with 0.9 mL of 0.5 M H₂SO₄ and 0.1 mL of 0.2 M FeSO₄. The resultant mixture was examined via UV-Vis spectroscopy to quantify the generated H₂O₂. To rule out interference from residual ions, we measured calibration curve using reference solutions. The photocurrents generated were measured using a 300 W Xe arc lamp and a UV cutoff filter (λ > 420 nm) in a 0.2 M Na₂SO₄ aqueous solution. Photocurrent measurements were performed three times, confirming the reproducibility of photocurrent data. Photocatalyst coating was conducted on 1 × 2 cm² fluorine doped tin oxide (FTO) glass to fabricate an electrode for photocurrent measurements. Mott–Schottky plots were recorded in the frequency range of 5000–100 Hz and the potential range of −1.0 to 1.0 V (vs. SCE).

Results and discussion

The substitution of Cu into the ZnIn_2−xCu_xS₄ lattice was confirmed using powder XRD analysis. As shown in Fig. 1b, regardless of Cu substitution, all ZnIn_2−xCu_xS₄ materials exhibit typical Bragg reflections of the ZnIn₂S₄ lattice without impurity-related peaks. The Cu substitution leads to a slight shift of the XRD peak toward the high-angle side, indicating a reduction in the lattice parameters. This is attributed to the replacement of smaller-sized cations and/or the formation of S vacancies upon the aliovalent substitution of Cu²⁺ (0.73 Å) for In³⁺ (0.8 Å). As presented in Fig. 1c and S1 (SI), exfoliation of the monolayered ZnIn_2−xCu_xS₄ nanosheets was confirmed by AFM data of Cu0.2-ZIS exhibiting a highly anisotropic two-dimensional (2D) nanosheet morphology with an atomic-scale thickness of ∼1.1 nm and an average lateral dimension of ∼300 nm. The STEM image in Fig. 1d reveals highly anisotropic 2D nanosheet crystallites with a very faint contrast, providing further evidence for the exfoliation of ZnIn_2−xCu_xS₄ nanosheets. The observation of significant color changes upon Cu substitution strongly suggests the Cu-substitution-driven modification of the electronic structure of ZnIn₂S₄, as displayed in Fig. 1e. In Fig. 1f, the FE-SEM results demonstrate house-of-cards-type stacking of the exfoliated 2D ZnIn_2−xCu_xS₄ nanosheets. The successful incorporation of Cu ions into the ZnIn_2−xCu_xS₄ nanosheets was supported by EDS elemental mapping analysis for Cu0.2-ZIS, which revealed the homogenous distribution of Zn, In, Cu, and S elements (Fig. 1f).

The distinct impact of Cu substitution on the electronic structure and chemical bonding character of ZnIn_2−xCu_xS₄ was supported by combined spectroscopic analyses. As shown in Fig. S2 (SI), the Cu0.2-ZIS nanosheet shows typical Cu 2p XPS features of Cu²⁺ species, indicating the divalent oxidation state of the substituted Cu ions. Both Zn 2p and In 3d XPS analyses (Fig. 2a and b) revealed that Cu substitution shifts the XPS features toward the low-energy side, reflecting the formation of anion vacancies caused by Cu substitution. In the S 2p XPS spectra shown in Fig. 2c, Cu substitution induces the blue shift of the high-energy shoulder peak, confirming the formation of S defects upon Cu substitution.²¹ All the observed Cu-substitution-induced XPS spectral changes are attributed to the aliovalent replacement of In³⁺ ions with Cu²⁺ ions, resulting in the formation of anion vacancies. In micro-Raman analysis (Fig. 2d), both the unsubstituted ZIS and Cu-substituted Cu0.2-ZIS materials display characteristic phonon lines of the ZnIn₂S₄ phase, emphasizing the retention of the ZnIn₂S₄ lattice after Cu substitution. The incorporation of Cu ions into the ZnIn₂S₄ lattice reduces the phonon energy, which is ascribed to lattice softening caused by the introduction of S vacancies.

Fig. 2 (a) Zn 2p XPS, (b) In 3d XPS, (c) S 2p XPS, and (d) micro-Raman data of ZnIn_2−xCu_xS₄ nanosheets.

The local structure and electronic configuration of the Cu substituent ions in the ZnIn_2−xCu_xS₄ nanosheets were characterized using X-ray absorption spectroscopy. In the Cu K-edge XANES spectra in Fig. 3a, Cu0.2-ZIS exhibits two main-edge peaks A and B assigned to the following dipole-allowed transitions: 1s → 4p_π with shakedown process and 1s → 4p_σ without shakedown process,²² respectively. The energies of these two features are nearly identical for Cu0.2-ZIS and the reference CuO, indicating the divalent oxidation state of the Cu substituent ions, as determined via XPS analysis. The very high intensity of peak A indicates significant tetragonal distortion around the Cu ions due to their Jahn–Teller-active electronic configuration (e.g., 3d⁹). The distortion is attributed to the elongation of the out-of-plane Cu–S bond in the tetragonally distorted CuS₆ octahedra, which facilitates the shakedown process owing to the weakening of the repulsion between the transferred electrons and axial ligands.²³ In the Zn K-edge and In K-edge XANES spectra in Fig. 3b and c, both ZIS and Cu0.2-ZIS show typical spectral features and edge energies of Zn²⁺ and In³⁺ ions, respectively, confirming the retention of ZnIn_2−xCu_xS₄ lattice upon Cu substitution.^10,24

Fig. 3 (a) Cu K-edge XANES, (b) Zn K-edge XANES, (c) In K-edge XANES, (d) Cu K-edge FT-EXAFS, (e) Zn K-edge FT-EXAFS, (f) In K-edge FT-EXAFS, and (g) contour plot of the WT-EXAFS data of ZnIn_2−xCu_xS₄ nanosheets. (h) Scheme for the impact of Cu substitution on the crystal structure of ZnIn₂S₄ lattice.

The local atomic arrangements of the component metal ions in the ZnIn_2−xCu_xS₄ material were investigated via EXAFS analysis, Fig. S3 (SI). In Fig. 3d, ZnIn_2−xCu_xS₄ exhibits a Fourier transform (FT) peak at ∼1.8 Å corresponding to Cu–S bonding pairs, whose distance agrees well with the average Cu–S bond distances in tetragonally distorted CuS₆ octahedra,²⁵ substantiating the stabilization of the substituted Cu ions in the octahedral sites of the ZnIn₂S₄ lattice. In the Zn K-edge FT-EXAFS data in Fig. 3e, both ZIS and Cu0.2-ZIS commonly exhibit an intense FT peak at ∼1.8 Å related to Zn–S bonding pairs. In contrast to the absence of a significant change in the peak position, both the exfoliation and Cu substitution processes led to a slight but distinct depression of this FT peak, indicating an increase in the crystal disorder around the Zn ions. Similarly, the exfoliation- and Cu-substitution-driven depressions of the FT peak at ∼2.0 Å related to the In–S coordination shell are discernible in the In K-edge FT-EXAFS spectra in Fig. 3f, confirming the increase in structural disorder upon the exfoliation and Cu substitution. However, in contrast to the Zn K-edge FT-EXAFS data, the In K-edge FT-EXAFS data display a shift of the In–S related FT peak toward the longer-bond distance side upon the Cu substitution, revealing the elongation of In–S bonds. The depression of the first FT peak corresponding to the metal–sulfur bond was further substantiated by the Zn K-edge and In K-edge wavelet transform (WT)-EXAFS contour data (Fig. 3g), in which the Zn–S- and In–S-related maxima at k = ∼4–8 Å⁻¹ and R = ∼2.0 Å were less intense for Cu0.2-ZIS than ZIS. The distinct impacts of exfoliation and Cu substitution on the local structure of ZnIn₂S₄ lattice were cross-confirmed by nonlinear curve fitting analyses for Zn K-edge and In K-edge EXAFS data, see Fig. 3e and f, S4 and Table S1 (SI). The coordination numbers of Zn–S and In–S bonding pairs become gradually reduced upon the exfoliation and Cu substitution processes, underscoring the enhancement of local structural disorder. In addition, the Cu substitution results in the significant elongation of the In–S bond, which is greater than the change of the Zn–S bond distance. As shown in Fig. 3h, the introduction of shorter and stronger Cu–S bonds into the ZnIn_2−xCu_xS₄ lattice weakens the adjacent In–S bonds owing to the bond-competition effect. Thus, the resulting elongation of In–S bonds provides compelling support for the substitution of Cu ions in octahedral In sites.

To evaluate the efficiency of the combined aliovalent substitution–exfoliation approach in improving the photocatalytic activity of zinc indium sulfide, the obtained ZnIn_2−xCu_xS₄ nanosheets were employed as visible-light-active photocatalysts for H₂O₂ generation. In Fig. 4a, the Cu-substituted ZnIn_2−xCu_xS₄ nanosheets show significantly higher activity for the photocatalytic production of H₂O₂via the partial reduction of O₂ molecules than the unsubstituted ZnIn₂S₄ nanosheets and pristine ZnIn_2−xCu_xS₄ materials. This underscores the benefit of the synergistic Cu-substitution–exfoliation strategy. Among the present materials, the Cu0.2-ZIS delivers the best photocatalytic performance with a high production rate of 1475 μM g⁻¹ h⁻¹, indicating an optimal substitution rate of x = 0.2. In addition, a control experiment was conducted in an Ar atmosphere to verify the origin of H₂O₂ product. In this O₂-free condition, no H₂O₂ is formed, confirming that the produced H₂O₂ is a product of the reduction of atmospheric O₂ molecule, as shown in Fig. S5 (SI). The recyclability of Cu0.2-ZIS was evidenced by conducting the repeated cycling tests. As shown in Fig. S6 (SI), Cu0.2-ZIS demonstrates an excellent retention of its initial photocatalytic activity without any significant degradation for four successive runs, highlighting its good reusability. The XRD and XPS analyses reveal that the Cu0.2-ZIS experiences no marked changes in these data after the photocatalytic reaction, see Fig. S7 (SI). This result indicates the negligible effect of photocatalytic activity test on the structural integrity and electronic structure of the catalyst. Moreover, the elemental analysis clearly demonstrates a good retention of Cu content before and after the reaction, stressing its good chemical stability with the absence of Cu leaching. The advantage of Cu substitution in improving the photocatalytic activity was further substantiated by photocurrent generation experiments. As shown in Fig. 4b, the Cu-substituted Cu0.2-ZIS nanosheet generates photocurrents approximately four times higher than those of the unsubstituted ZIS nanosheets, confirming the benefits of the combined Cu-substitution–exfoliation approach.

Fig. 4 (a) Photocatalytic activity data toward H₂O₂ generation, (b) photocurrent generation, (c) diffuse reflectance UV-Vis, (d) PL, (e) EIS, and (f) scheme for band structure of ZnIn_2−xCu_xS₄ nanosheets. The error bars (mean ± standard deviation) were obtained based on three independent photocatalytic experiments.

To clarify the origin of the beneficial impact of Cu substitution on the photocatalytic activity of ZnIn₂S₄ nanosheet, the evolution of the optical properties upon Cu substitution was characterized using diffuse-reflectance UV-Vis and PL spectroscopy. In Fig. 4c, the Cu-substituted ZnIn_2−xCu_xS₄ nanosheets exhibit far higher visible-light absorptivity with a smaller bandgap energy than the unsubstituted ZIS nanosheet, indicating Cu-substitution-driven narrowing of the bandgap energies. The PL results in Fig. 4d revealed that Cu substitution results in a marked depression of the PL signal, indicating that it effectively prevents charge recombination and the resulting extension of the electron–hole lifetime. The observed changes in the optical properties significantly contribute to the improvement in the photocatalytic activity upon Cu substitution. In addition, EIS data were measured to determine the effect of Cu substitution on the charge-transport properties of the ZnIn₂S₄ nanosheet.^26–28 As shown in Fig. 4e, all ZnIn_2−xCu_xS₄ materials show a semicircle related to the charge-transfer resistance (R_ct) in the mid-frequency region. The Cu substitution significantly reduces the radius of the semicircle, underscoring the improvement in the charge-transport properties following Cu substitution.²⁹ Because the migration of photoexcited electrons into surface active sites is crucial for optimizing the photocatalyst performance for H₂O₂ production, the observed Cu-substitution-driven improvement in the charge-transport properties is also responsible for the enhanced photocatalytic activity of the Cu-substituted ZnIn_2−xCu_xS₄ nanosheets. The profound impact of Cu substitution on the band structure of the ZnIn_2−xCu_xS₄ nanosheets was verified using Mott–Schottky plots and UV-Vis analyses, as shown in Fig. S8 (SI). Cu substitution lowers the position of the conduction band (CB) in ZnIn_2−xCu_xS₄, Fig. 4f and S9 (SI). Consequently, the CB position in ZnIn_2−xCu_xS₄ approaches the redox potential of O₂/H₂O₂, promoting interfacial electron injection into the adsorbed O₂ molecules.

To further investigate the mechanism underlying the benefits of Cu substitution, in situ EXAFS measurements were performed on Cu-substituted Cu0.2-ZIS and unsubstituted ZIS upon illumination with visible light. As shown in Fig. 5a, under the visible-light irradiation, the in situ Cu K-edge FT-EXAFS data for Cu0.2-ZIS indicate notable depression of the FT peak at ∼1.8 Å corresponding to the Cu–S bonding pair, suggesting an increase in local structural disorder. Because the adsorption of O₂ molecules and interfacial interaction with the adsorbed O₂ molecules cause the diversification of the Cu–S bond distance and chemical environment, the observed peak attenuation is convincing evidence of enhanced adsorption and interaction with the O₂ reactant. Similarly, the in situ Zn K-edge FT-EXAFS data in Fig. 5b display a marked depression in the FT peak assigned to the Zn–S bonding pairs for the Cu-substituted Cu0.2-ZIS nanosheets. In comparison, the photoinduced spectral change is weaker for the unsubstituted ZIS nanosheets than for the Cu0.2-ZIS (Fig. 5b). This finding provides spectroscopic evidence for the Cu-assisted enhancement of the interfacial interaction with the reactant O₂ molecules. Similarly, as shown in the in situ In K-edge FT-EXAFS spectra in Fig. 5c, a more prominent depression in the FT peak assigned to the In–S bonding pair under visible-light illumination is discernible for the Cu-substituted Cu0.2-ZIS nanosheets compared to the Cu-free ZIS nanosheets. All the in situ EXAFS results provide compelling evidence for the benefit of Cu substitution in enhancing the interaction of ZnIn_2−xCu_xS₄ with the adsorbed O₂ molecules, which can originate from the introduction of anion vacancies upon Cu substitution. As illustrated in Fig. 5d, the created anion vacancies can function as efficient adhesion sites for O₂ molecules, leading to the promotion of O* adsorption and Cu-substitution-induced enhancement of the photocatalytic activity of the ZnIn₂S₄ nanosheets. This conclusion is further supported by the recently reported density functional theory calculations showing the significant impact of S vacancy creation on the coordination structure and O₂ adsorption property of ZnIn₂S₄.¹⁰ Furthermore, the vacancy-induced introduction of mid-gap defect states provides multiple migration pathways for photogenerated electrons, facilitating multi-path charge separation. The formation of vacancies in sulfide materials also creates non-uniform charge layers at the interfaces of 2D nanosheet, which generates an interlayer electric field that promotes electron separation and thus enhances the photocatalytic activity.^30,31 Notably, the peak depression is stronger for the Cu K-edge FT-EXAFS data than for the Zn K-edge and In K-edge data, indicating the pivotal role of the Cu substituent in the photocatalytic reaction with O₂ molecules. Considering the Jahn–Teller-active 3d⁹ electronic configuration of Cu²⁺ ions, an elongation of the axial Cu–S bond facilitates the interaction between Cu and O, which contributes to the advantage of Cu substitution.

Fig. 5 (a) In situ Cu K-edge FT-EXAFS, (b) in situ Zn K-edge FT-EXAFS, and (c) in situ In K-edge FT-EXAFS data of ZnIn_2−xCu_xS₄ nanosheets. (d) Scheme for the impact of synergetic defect and local structure regulation on the photocatalytic activity of ZnIn_2−xCu_xS₄ toward H₂O₂ production.

Moreover, the porous structure of present materials was confirmed by N₂ adsorption–desorption isotherm measurements. The Cu-substituted Cu0.2-ZIS nanosheet exhibits a larger Brunauer–Emmett–Teller (BET) surface area (65.7 m² g⁻¹) compared to the ZIS nanosheet (48.9 m² g⁻¹) and bulk Cu0.2-ZIS (2.9 m² g⁻¹), indicating that both the liquid exfoliation and the Cu substitution processes commonly increase the surface area. The remarked expansion of surface area upon the combined Cu substitution and exfoliation process additionally contributes to the enhanced photocatalytic performance of Cu0.2-ZIS via the provision of more reaction sites (Fig. S10, SI).

To confirm the importance of the local structural modification of Cu ions, a Jahn–Teller-inactive divalent Ni²⁺ ion was employed as substituent ions for ZnIn_2−xM_xS₄ with an optimal substitution rate of x = 0.2. The obtained material was named Ni0.2-ZIS, respectively. The successful incorporation of divalent Ni²⁺ into the ZnIn₂S₄ lattice and following exfoliation were confirmed by powder XRD and STEM analyses (Fig. S11, SI). According to the Ni K-edge XANES analysis, Ni0.2-ZIS displays only a weak small pre-edge peak corresponding to dipole-forbidden 1s → 3d transition. Since this transition can be promoted by the local structural distortion from centrosymmetric NiS₆ octahedra,^32,33 the observed low intensity of this pre-edge peak provides convincing evidence for the stabilization of Ni²⁺ (3d⁸) ions in regular octahedral symmetry (Fig. S11c, SI). The obtained Ni0.2-ZIS was tested as photocatalysts for visible-light-driven H₂O₂ production. As shown in Fig. S12 (SI), both Ni0.2-ZIS nanosheets deliver notably lower photocatalytic activity toward H₂O₂ generation with respect to Cu0.2-ZIS. This result provides compelling support for the importance of local structural modifications induced by Jahn–Teller-active Cu²⁺ ions.³⁴

Conclusions

The combined aliovalent Cu-substitution–exfoliation approach provides a useful defect and local structure engineering method for optimizing the photocatalytic performance of ZnIn₂S₄ materials. Exfoliation of Cu-substituted ZnIn_2−xCu_xS₄ materials allows the synthesis of atomically thin photocatalytically active nanosheets with optimized surface structures, electronic configurations, and charge-transport properties. The aliovalent substitution of In³⁺ with Cu²⁺ leads to the introduction of S vacancies and local structural modifications, which increased the surface reactivity and the regulation of the electronic structure. The resulting Cu-substituted ZnIn_2−xCu_xS₄ nanosheets exhibit excellent photocatalytic activity for visible-light-driven H₂O₂ generation. The in situ EXAFS analysis under visible-light illumination clearly demonstrates that Cu substitution promoted O₂ adsorption on the anion vacancy sites and Jahn–Teller-active Cu sites. The outstanding photocatalytic activity of the Cu-substituted ZnIn_2−xCu_xS₄ nanosheets is attributed to the increase in visible-light absorptivity, suppression of electron–hole recombination, promotion of O₂ adsorption, and improved charge-transport behaviors. Considering the vast library of semiconducting layered metal chalcogenides that allow exfoliation into nanosheets,^35–38 the combined defect and local structure engineering approach is promising for developing highly efficient photocatalyst materials.

Author contributions

Conceptualization: SJH, XJ; methodology: HJ, DWL, SJH, XJ; investigation: HJ, DWL, XJ; formal analysis: HJ, DWL, RC, XJ; visualization: HJ, DWL, XJ; funding acquisition: RC, SJH, XJ; supervision: SJH, XJ; writing—original draft: SJH, XJ; writing—review & editing: RC, SJH, XJ.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Supplementary information: additional experimental results (12 figures and 1 table). See DOI: https://doi.org/10.1039/d5ta06161j.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (RS-2023-00208355, 2021M3H4A1A03049662) and the Global Science Research Center Program (RS-2024-00411134). This work was also supported by the 2025 Research Fund of the University of Seoul for Xiaoyan Jin. The experiments at PAL were supported in part by MOST and POSTECH.

References

1. Y. Zhang, C. Pan, G. Bian, J. Xu, Y. Dong, Y. Zhang, Y. Lou, W. Liu and Y. Zhu, Nat. Energy, 2023, 349, 361.
2. X. Zhang, H. Su, P. Cui, Y. Cao, Z. Teng, Q. Zhang, Y. Wang, Y. Feng, R. Feng, J. Hou, X. Zhou, P. Ma, H. Hu, K. Wang, C. Wang, L. Gan, Y. Zhao, Q. Liu, T. Zhang and K. Zheng, Nat. Commun., 2023, 14, 7115.
3. C.-H. Bao, L. Li, X.-F. Wang, S.-S. Xia, X. Wang, C.-C. Jin and Z. Chen, Nano Lett., 2025, 25, 4115.
4. M. Tan, Z. Yao, L. Ma, F. Cao, Y. Peng, J. Chen and Y. Bai, Coord. Chem. Rev., 2025, 544, 216993.
5. J. M. Campos-Martin, G. Blanco-Brieva and J. L. G. Fierro, Angew. Chem., Int. Ed., 2006, 45, 6962.
6. H. Huang, Q. Zhang, R. Shi, C. Su, Y. Wang, J. Zhao and T. Zhang, Appl. Catal., B, 2022, 317, 121731.
7. F. M. Yap, G. Z. S. Ling, B. J. Su, J. Y. Loh and W.-J. Ong, Nano Res. Energy, 2024, 3, e9120091.
8. X. Zhang, S. Cheng, C. Chen, X. Wen, J. Miao, B. Zhou, M. Long and L. Zhang, Nat. Commun., 2024, 15, 2649.
9. K. Dong, Y. Lei, H. Zhao, J. Liang, P. Ding, Q. Liu, Z. Xu, S. Lu, Q. Li and X. Sun, J. Mater. Chem. A, 2020, 8, 23123.
10. H. Peng, H. Yang, J. Han, X. Liu, D. Su, T. Yang, S. Liu, C.-W. Pao, Z. Hu, Q. Zhang, Y. Xu, H. Geng and X. Huang, J. Am. Chem. Soc., 2023, 145, 27757.
11. Y. Liu, Y. Wu, Y. Liu, Y. Wang, X. Sun, P. Chen and S.-F. Yin, ACS Catal., 2024, 14, 7726.
12. X. Xin, Y. Li, Y. Zhang, Y. Wang, X. Chi, Y. Wei, C. Diao, J. Su, R. Wang, P. Guo, J. Yu, J. Zhang, A. J. Sobrido, M.-M. Titirici and X. Li, Nat. Commun., 2024, 15, 337.
13. M. Yang, X.-Q. Zhan, D.-L. Ou, L. Wang, L.-L. Zhao, H.-L. Yang, Z.-Y. Liao, W.-Y. Yang, G.-Z. Ma and H.-L. Hou, Rare Metals, 2024, 44, 1024.
14. X. Ruan, S. Zhao, M. Xu, D. Jiao, J. Leng, G. Fang, D. Meng, Z. Jiang, S. Jin, X. Cui and S. K. Ravi, Adv. Energy Mater., 2024, 14, 2401744.
15. J. Lee, T. Kim, D. H. Sun, X. Jin and S.-J. Hwang, EnergyChem, 2025, 7, 100139.
16. Y. H. Park, Y. Sun, D. W. Lee, I. Y. Kim, X. Jin and S.-J. Hwang, Nano Res. Energy, 2025, 4, e9120163.
17. X. Jin, T. Lee, A. Soon and S.-J. Hwang, Adv. Funct. Mater., 2024, 34, 2316446.
18. C. Fu, M. Zhao, X. Chen, G. Sun, C. Wang and Q. Song, Appl. Catal., B, 2023, 332, 122752.
19. L. Huang, B. Han, X. Huang, S. Liang, Z. Deng, W. Chen, M. Peng and H. Deng, J. Alloy. Compd., 2019, 798, 553.
20. H. Piao, G. Choi, X. Jin, S.-J. Hwang, Y. J. Song, S.-P. Cho and J.-H. Choy, Nano-Micro Lett., 2022, 14, 55.
21. X. Wang, J. Bai, Y. Wang, X. Lu, Z. Zou, J. Huang and C. Xu, Green Energy Environ., 2022, 7, 755.
22. J.-H. Choy, W. Lee and S.-J. Hwang, Phys. C: Supercond. Appl., 1999, 322, 93.
23. J. Garcia, M. Benfatto, C. R. Natoli, A. Bianconi, A. Fontaine and H. Tolentino, Chem. Phys., 1989, 132, 295.
24. E. Zhang, Q. Zhu, J. Huang, J. Liu, G. Tan, C. Sun, T. Li, S. Liu, Y. Li, H. Wang, X. Wan, Z. Wen, F. Fan, J. Zhang and K. Ariga, Appl. Catal., B, 2021, 293, 120213.
25. Y. Mathey, H. Mercier, A. Michalowicz and A. Leblanc, J. Phys. Chem. Solids, 1985, 46, 1025.
26. J.-W. Ji, L.-J. Zhang, W.-J. Yi, Y. Wang, J.-H. Chen, X.-Z. Yue and S.-S. Yi, Chem. Eng. J., 2024, 500, 157277.
27. L. Zhao, J. Wang, R. Yan, S. Wang, Z. Liao, H. Yang, D. Yang, H. Hou and W. Yang, Small, 2025, 21, 2406109.
28. L. Zhao, H. Hou, S. Wang, L. Wang, Y. Yang, C. R. Bowen, J. Wang, Z. Liao, D. Yang, R. Yan and W. Yang, Adv. Funct. Mater., 2025, 35, 2416346.
29. N. H. Kwon, S.-J. Kim, T.-H. Gu, J. M. Lee, M. H. Kim, D. Paik, X. Jin, H. Kim and S.-J. Hwang, Adv. Sci., 2024, 11, 2408948.
30. Y. Jiang, H. Sun, J. Guo, Y. Liang, P. Qin, Y. Yang, L. Luo, L. Leng, X. Gong and Z. Wu, Small, 2024, 20, 2310396.
31. X. Bai, X. Wang, T. Jia, L. Guo, D. Hao, Z. Zhang, L. Wu, X. Zhang, L. Wu, X. Zhang, H. Yang, Y. Gong, J. Li and H. Li, Appl. Catal., B, 2022, 310, 121302.
32. M. Numan, G. Das, M. S. Khan, G. Manna, A. Banerjee, S. Giri, G. Aquilanti and S. Majumdar, Phys. Rev. B, 2022, 106, 214437.
33. S.-J. Hwang and J.-H. Choy, J. Phys. Chem. B, 2003, 107, 5791.
34. B. Munisha, L. Patra, J. Nanda and S. Mondal, RSC Adv., 2025, 15, 3110.
35. S. Manzeli, D. Ovchinnikov, D. Pasquier, O. V. Yazyev and A. Kis, Nat. Rev. Mater., 2017, 2, 83.
36. R. Wu, H. Zhang, H. Ma, B. Zhao, W. Li, Y. Chen, J. Liu, J. Liang, Q. Qin, W. Qi, L. Chen, J. Li, B. Li and X. Duan, Chem. Rev., 2024, 124, 10112.
37. H. Hou, C. R. Bowen, D. Yang and W. Yang, Chem, 2024, 10, 800.
38. N. H. Kwon, X. Jin, S.-J. Kim, H. Kim and S.-J. Hwang, Adv. Sci., 2022, 9, 2103042.

---

Footnote

† These authors contributed equally to this work.

---