Defect Engineering of ZnIn₂S₄ Photocatalysts for Enhanced Hydrogen Evolution Reaction

Abstract

ZnIn₂S₄, a visible-light-responsive layered sulfide photocatalyst with a suitable bandgap (~2.4 eV), exhibits considerable potential for the photocatalytic hydrogen evolution reaction (PHER) due to its low toxicity, excellent stability, and appropriate band alignment. Nevertheless, its practical deployment is limited by inherent issues such as rapid charge carrier recombination, scarce surface-active sites, and slow oxidation kinetics. Defect engineering strategies—including sulfur, zinc, and indium vacancies, as well as heteroatom doping—have been developed to mitigate these shortcomings. This review not only summarizes recent advances in these strategies but also elucidates the fundamental physicochemical mechanisms behind the enhanced photocatalytic performance. A systematic quantitative evaluation is presented, highlighting improvements in critical performance metrics such as hydrogen evolution rate, light absorption range, apparent quantum yield (AQY), and charge separation efficiency. Furthermore, the review offers a critical perspective on the current state of defect-engineered ZnIn₂S₄ systems. Promising future research pathways are outlined, with emphasis on atomic-precision synthesis and operando characterization techniques. Finally, we discuss persistent challenges in the field, including reproducibility in synthesis, long-term operational stability, and scalability toward industrial hydrogen production.

1. Introduction

Mitigating the intensifying climate crisis necessitates a global transition towards carbon neutrality, demanding transformative technologies for clean energy production and storage [1,2,3,4,5]. Hydrogen, envisioned by the International Energy Agency (IEA) to constitute approximately 13% of the world’s final energy consumption by 2050, emerges as a pivotal clean energy carrier. However, the current hydrogen landscape is dominated by production methods reliant on fossil fuel reforming, which incur significant carbon dioxide emissions, which starkly contradicts the imperative for decarbonization. Consequently, the development of sustainable and carbon-neutral hydrogen evolution reaction HER technologies is critically urgent [6,7,8,9]. Among the promising solutions, photocatalytic hydrogen evolution reaction (PHER) through water splitting stands out [10,11,12,13]. This approach harnesses abundant solar energy to directly dissociate water molecules, generating storable hydrogen without the associated carbon emissions [14,15,16,17]. By mimicking natural photosynthesis, photocatalysis offers a potentially scalable pathway to convert solar energy directly into chemical fuels, representing a cornerstone strategy for a sustainable hydrogen economy [18,19,20,21].

Among visible-light photocatalysts, hexagonal ZnIn₂S₄ (ZIS, E_g ≈ 2.4 eV) exhibits compelling advantages [22,23,24,25]. Its hexagonal phase features advantageous [S–Zn–In–S] layers that facilitate charge transport and modification, combined with optimal band positions, good chemical stability, and an eco-friendly composition [26,27,28,29]. However, three key limitations impede its practical application as shown in Figure 1 [30,31,32,33]:

(i)

Severe charge recombination: Due to inter-layer potential barriers (>0.5 eV) and deep-trap states (0.8–1.4 eV below the conduction band minimum) induced by sulfur vacancies (V_S), which function as Auger-mediated recombination centers. These constraints confine carrier transport primarily to out-of-plane pathways, culminating in >80% bulk recombination within 5 ns rapid carrier recombination due to inter-layer barriers and inherent defects [15,34,35].

(ii)

Insufficient HER kinetics: Arising from low active-site density (<5% S-atom utilization). Density functional theory (DFT) calculations reveal that this originates from suboptimal orbital hybridization: the d-band center of surface In atoms resides at −2.8 eV below E_F, weakening S 3p–H 1s coupling. Consequently, the Gibbs free energy of hydrogen adsorption (ΔG_H) deviates substantially from thermoneutrality (ΔG_H ≈ 0.35 eV), placing ZIS on the weak-adsorption branch of the HER [36,37,38].

(iii)

Photocorrosion instability: Mediated by V_S sites, which initiate an autocatalytic degradation cycle. Under illumination, V_S-derived trap states (0.8 eV below CBM) accumulate photogenerated holes, oxidizing adjacent S atoms and releasing H₂S, thereby inducing irreversible structural collapse [39,40,41,42].

Recent surface engineering strategies, such as defect engineering, heterogeneous structure construction, surface functionalization, morphology and facet control, have been effectively addressing these issues. For instance, surface vacancies (such as sulfur, zinc, or indium vacancies) induce trap-dominated recombination by optimizing conduction band alignment or reconstructing coordination environments to suppress sulfur oxidation thermodynamics [43,44]; metal or non-metal decoration optimizes hydrogen adsorption (ΔG_H = −0.03 eV) and suppresses interface recombination, and single-atom metal decoration will amplify active-site density [45]; heterojunction construction achieves spatial charge separation through built-in fields [46,47,48] and also the surface protective layer to suppress sulfur oxidation [49,50]. It should be noted that the defect engineering provides a precise solution to overcome the three issues listed in Figure 1, through atomic-scale interface manipulation (sulfur, zinc, or indium vacancies; heteroatoms doping), enhancing charge separation efficiency, optimizing active-site density, and suppressing photocorrosion via protective coatings or defect passivation [51,52].

Herein, this comprehensive review critically discusses defect engineering strategies specifically tailored for ZIS-based photocatalysts to advance the photocatalytic hydrogen evolution reaction (PHER). We systematically examine three major modification approaches, including sulfur, zinc, or indium vacancies, as well as heteroatom doping. Beyond summarizing these strategies, the review elucidates the underlying physicochemical mechanisms responsible for enhanced performance. A rigorous quantitative assessment benchmarks the improvements in key metrics, such as the PHER rate, light absorption, apparent quantum yield (AQY), and carrier separation efficiency. Importantly, the review adopts a critical perspective toward these surface-engineered ZIS photocatalytic systems. Furthermore, it highlights promising future research directions with an emphasis on atomic-precision synthesis and operando diagnostics. Finally, we address ongoing challenges related to synthesis reproducibility, long-term operational stability, and process integration for large-scale HER.

2. Classification and Mechanisms of Defect Engineering Strategies

2.1. Sulfur Vacancies (V_S)

Sulfur vacancy (V_S) engineering represents an effective approach to enhancing optical, electrical, and surface properties, thereby improving PHER performances [53]. As a typical anion defect in ZIS, V_S induce lattice distortion by breaking the local coordination balance of S atoms, thus reconstructing the electronic structure of the material [54]. Introducing V_S into ZIS is a common defect engineering strategy aimed at improving its photocatalytic performance (such as promoting the separation of photoinduced charge carriers, enhancing adsorption capacity, etc.). V_S can serve as active sites or alter the material’s electronic structure. V_S serve as efficient electron traps by introducing mid-gap defect states in ZIS, which significantly enhance light absorption and prolong photogenerated carrier lifetimes through suppressed recombination. Also, the V_S site could act as active sites for the absorption of H₂O molecules and accelerate the HER. As shown in Figure 2, the formation of V_S could be obtained by creating a sulfur-deficient chemical environment or applying energy (heat, plasma, radiation) to facilitate the escape of sulfur atoms, which mainly rely on the following methods: high-temperature calcination, hydrothermal/solvothermal reduction, plasma treatment technology, and other high-energy application methods. Although the introduction of V_S in ZIS could enhance the PHER by creating defect energy levels, the control and modulation of V_S are very important to PHER performances. Moderate concentrations of V_S enhance charge separation by trapping photogenerated electrons/holes, thereby suppressing recombination and boosting the HER. However, excessive V_S formation may establish recombination centers that degrade photocatalytic efficiency. Therefore, there is a critical need for precise V_S modulation.

2.1.1. High-Temperature Calcination Method

The calcination atmospheres (such as air, N₂, Ar, H₂/Ar, and so on) can create a sulfur-deficient chemical environment to achieve selective desulfurization at 200–800 °C, which facilitates the escape of sulfur atoms for the formation of V_S [55,56,57]. And the gradient control for the density of V_S can be realized by adjusting the calcination temperature and treated time. This method offers the advantages of a clear underlying principle; relatively precise control over vacancy concentration via temperature, holding duration, and gas concentration parameters; and scalability for large-scale treatment. As shown in Figure 3a–h, Jing et al. determined to conduct calcination experiments at temperatures below 500 °C by comparing the crystal phase conditions at 350 °C and 500 °C for different durations for the modulation of V_S [56]. ZIS engineered with V_S with the highest HER performances were fabricated through controlled calcination at 350 °C for 4h. The optimized light absorption characteristics and extending charge carrier lifetimes were obtained. Consequently, the ZIS-350-4h photocatalyst achieved an apparent quantum efficiency (AQE) of 0.21% at 420 nm and a solar-to-hydrogen (STH) conversion efficiency of 0.035%—representing the highest reported performance for pristine ZIS to date. As shown in Figure 3i–p, they also prepared rhombohedral ZIS nanosheets enriched with V_S under a N₂ atmosphere at 400 °C and 800 °C, ZIS enriched with V_S obtained at 800 °C with Pt/Cr cocatalysts presents a HER rate of 3.40 μmol·h⁻¹ [55]. An appropriate amount of V_S can improve the overall water splitting performance; the STH conversion efficiency of sulfur vacancy-enriched rhombohedral ZIS reaches 0.021%, which is the highest reported efficiency for single-phase ZIS. As shown in Figure 3k, the shift in S 2p₃/₂ (~161.4 eV) and S 2p₁/₂ (~162.6 eV) peaks in S 2p XPS spectra corresponded to the presence of low-valence sulfur or defect-associated sulfur states. Additionally, a decreased S/(Zn + In) atomic ratio serves as further evidence of V_S. Electron paramagnetic resonance (EPR) directly captures the characteristic signals of unpaired electrons localized at V_S, and its intensity is linearly related to the vacancy density (see Figure 3d,l). As shown in Figure 3m, ZIS-800 shows a red-shift in the absorption edge in the UV-vis spectra, which is generally associated with various defects, further confirming the presence of rich V_S and are consistent with the EPR results. Moreover, ZIS-800 has a lower PL intensity, indicating a higher separation efficiency of photogenerated electron–hole pairs. The lifetime of ZIS-800 (1.902 ns) is higher than that of ZIS (1.506 ns) in TRPL spectra (Figure 3n), effectively avoiding carrier recombination, and V_S as active sites can significantly improve the HER rate. Similarly, Huang et al. controlled ZIS that engineered V_S via calcination under a N₂ atmosphere for two hours at a much lower temperature (300 °C) for piezo-catalytic HER; a high HER rate of approximately 3164.67 µmo1∙g⁻¹∙h⁻¹ under ultrasound was obtained with the optimal defect level [58].

2.1.2. Hydrothermal/Solvothermal Reduction Method

During the hydrothermal/solvothermal reaction, sulfur vacancy formation is facilitated by controlling reaction conditions (temperature, time, and pH) [59,60]. This reduces the stability of the sulfur source or inhibits its complete reaction, allowing V_S to form naturally during crystal growth. Alternatively, different sulfur sources are used (e.g., thioacetamide (TAA) decomposes more readily than Na₂S and thiourea) or complexing agents are added to modulate reaction kinetics during the hydrothermal/solvothermal reaction to obtain V_S. The approach provides relative process simplicity while enabling simultaneous control over both material morphology and sulfur vacancy concentration. Specifically, the vacancy concentration can be precisely tuned by modifying key parameters such as the reductant amount, reaction temperature, and time. This is proved by Xu et al., who selected an appropriate sulfur source (TAA) and optimized the hydro/solvothermal conditions to generate a certain level of V_S [61]. As shown in Figure 4, they investigated various sulfur vacancy levels in pristine ZIS (denoted as ZIS-X; and P, M, and R represent poor, moderate, and rich contents of V_S, respectively), proved in the EPR spectra. ZIS-M with a moderate level of sulfur vacancies demonstrated the highest PHER performance of around 25 mmol·g⁻¹. ZIS-M, with the calculated theoretical work functions of 4.14 eV, possessed the most favorable hydrogen adsorption-free energy, resulting in superior HER performances. Interestingly, Ren et al. fabricated V_S-ZIS with efficient PHER performances (5437 μmol⁻¹·h⁻¹∙g⁻¹) via a one-step solvothermal process using ethylene glycol (EG); the addition of EG plays a key role in producing defects in the V_S during the solvothermal reaction [62].

In addition, a moderate amount of mild reducing agent, such as hydrazine hydrate (N₂H₄·H₂O), sodium borohydride (NaBH₄) or sodium hypophosphite (NaH₂PO₂) can be added; these reductants reduce part of the sulfur ions (S²⁻), causing them to escape as hydrogen sulfide (H₂S) and leaving V_S during the reaction process. It should be noted here that these agents require extreme care as they are highly reducing, toxic, and flammable at elevated temperatures; the experiments should be operated strictly in a fume hood and with precise controlled quantities. ZIS engineered by V_S was prepared via a N₂H₄·H₂O-assisted hydrothermal method. Typically, 100 mg as-synthesized ZIS was dispersed into 20 mL of deionized water for 1 h; then, 5 mL of N₂H₄·H₂O was added into the mixing solution and stirred for another 30 min. After that, the mixture was transferred to a 50 mL stainless steel autoclave and maintained at 240 °C oven for 5 h. Finally, the precipitate was separated by centrifugation and washed with deionized water several times, then dried at 60 °C for 10 h. The obtained light-yellow powder was labeled as ZIS engineered by V_S [63]. Reducing agents (e.g., NaBH₄) drive in situ formation of V_S during crystal growth. Li et al. utilized this method to anchor Fe-Ni₂P cocatalysts onto ZIS nanosheets rich with V_S [64]. The sites of V_S acted as nucleation centers for atomic-level interfacial bonding, facilitating electron transfer and yielding a HER rate of 15.6 mmol·g⁻¹·h⁻¹.

2.1.3. Plasma Treatment Technology

The generation of V_S in ZIS via plasma treatment operates through kinetic energy transfer and reactive species etching. Plasma-generated ions (e.g., Ar⁺, H⁺) or electrons (10–100 eV kinetic energy) collide with surface S atoms. Energy transfer exceeds the S-Zn/In bond dissociation energy (~2–3 eV), ejecting S atoms via momentum transfer, leading to the formation of V_S. Liu et al. reported ZIS nanosheets with V_S (ZIS-V_S) synthesized through Ar plasma treatment, and its defect concentration was modulated by varying the treatment times, which were confirmed by EPR spectra [65]. As shown in Figure 5a–g, Ar plasma-treated ZIS with optimized sulfur vacancies (radio frequency discharge: 70 W, 13.56 MHz; duration: 5 min) achieves a visible-light PHER rate of 261.9 μmol·h⁻¹, representing a 2.06-fold enhancement over pristine ZIS. The AQE reaches 51.2% at 420 nm of monochromatic irradiation. H₂O-adsorption FTIR spectroscopy confirms the enhanced adsorption of water molecules at vacancy sites. Furthermore, density functional theory (DFT) calculations elucidate the altered HER kinetics in alkaline media between V_S-ZIS and pristine ZIS. These results demonstrate that Ar plasma treatment effectively engineers surface sulfur vacancies on ZIS nanosheets to boost the PHER efficiency. Jiang et al. carried out a comparative study on V_S-ZIS obtained through two facile approaches, plasma etching and annealing, for enhancing the photocatalytic performance [66]. The defect formation energies for atoms at various sites within a single unit cell of ZIS were calculated and are presented in Figure 5i. From the density of states (DOS) of both ZIS and V_S-ZIS in Figure 5j,k, it is evident that the valence band maximum (VBM) of the ZIS microsphere is primarily composed of S 2p and In 2s orbitals, while the conduction band minimum (CBM) is dominated by S 2p orbitals. In contrast to pristine ZIS, a deep defect level emerges within the bandgap of V_S-ZIS, contributed mainly by S 2p, S 2s, and In 2s orbitals. This localized state lies above the Fermi level and represents an occupied electron state. It is widely recognized that such states can enhance electron–hole separation by trapping and exciting electrons; however, they may also act as recombination centers due to relatively low hole mobility, which could potentially reduce photocatalytic activity. Therefore, Ar plasma-etched V_S-ZIS achieved an HER of 706 μmol·g⁻¹·h⁻¹, which was 5 and 1.2 times higher than that of pure ZIS and annealed ZIS, respectively (see Figure 5l,m), due to optimized surface-localized V_S, whereas bulk vacancies from annealing acted as recombination centers.

In H₂/Ar plasma, reactive H radicals reduce surface S²⁻ resulting in sulfur etching and leaving V_S. For instance, ZIS nanosheet arrays were modified by H₂-Ar plasma to introduce V_S; EPR spectra unambiguously confirmed the formation of V_S. By varying plasma power (20–100 W), the modulation of the concentration of V_S was controlled. This defect engineering significantly enhanced photoelectrochemical performance: plasma treatment at 60 W yielded a photocurrent density of 0.3 mA cm⁻² at 0.3 V of V_S. RHE—approximately double that of untreated samples—with a corresponding HER rate of 2.74 μmol∙cm⁻²∙h⁻¹ versus 0.96 μmol∙cm⁻²∙h⁻¹ for pristine ZIS material. These performance enhancements validate EPR as a reliable quantitative indicator of vacancy states [67].

2.1.4. Other Methods

Other high-energy methods (such as ball milling and laser) were also used to generate V_S for ZIS. The V_S were formed predominantly through mechanochemical reactions and lattice defect engineering during high-energy ball milling (HEBM) [68]. Intense mechanical collisions and friction transfer substantial energy to the ZIS material, facilitating sulfur diffusion and migration from the lattice to create V_S. Concurrently, HEBM induces structural modifications including reduced crystallite size and increased lattice distortion, which collectively regulate the distribution density of V_S. The thermal hotspots formed during mechanical energy conversion also accelerate surface property modifications, creating additional catalytic active sites beneficial for photocatalytic reactions. Beyond V_S, HEBM introduces co-existing defects such as Zn/In vacancies and interstitial atoms. Daria Roda et al. proposed a new method for preparing ZIS thin films using pulsed laser deposition (PLD) [69]. Future research should focus on improving the durability of ZIS through methods such as protective coatings and doping strategies, so as to mitigate degradation and maintain a high photoelectrochemical (PEC) performance.

The advancement of the engineering of V_S in photocatalysts faces three critical challenges: (1) regarding stability limitations, high-concentration sites of V_S are prone to oxidation by reactive oxygen species; (2) precision control demands, as existing methods lack an atomic-layer spatial resolution for the distribution of V_S, requiring hybrid techniques such as ALD-EBID for site-specific vacancy creation; and (3) cross-scale mechanism gaps. The intrinsic charge transport properties and the dynamic evolution of V_S under reaction conditions fail to form a coherent mechanism chain spanning atomic, mesoscopic, and macroscopic scales. As the core regulatory unit of ZIS surface engineering, V_S break through the bottleneck of the PHER through the trinity mechanism of “electron capture-bandgap narrowing-adsorption optimization”. In the future, it is necessary to integrate atomic manufacturing, in situ diagnosis and machine learning to achieve a precise editing of the concentration/position of V_S and promote the practical application of solar HER [58].

2.2. Zinc or Indium Vacancies

In the PHER, zinc (Zn) or indium (In) vacancies in the surface engineering of ZIS significantly enhance catalytic performances by regulating the electronic structure, optimizing the active sites, and improving the carrier separation efficiency [70,71]. Xu et al. reported that ZIS nanosheets with controlled Zn vacancies (V_Zn) had an efficient PHER rate of 212.0 μmol under 4h light irradiation, which is 1.9 times that of pristine ZIS [72]. As shown in Figure 6a, V_Zn-ZIS was synthesized by calcination at 200 °C for 20 min in air. The composition of the CBM for both ZIS and V_Zn-ZIS, derived from the DOS calculations in Figure 6b, reveals a significantly enhanced DOS near the Fermi level. This suggests that V_Zn can introduce new defect states below the CBM and promote the accumulation of photogenerated electrons in the conduction band of ZIS. Furthermore, the partial charge density distributions of ZIS and V_Zn-ZIS visually demonstrate that the charge density near the CBM of pristine ZIS is primarily localized around the In atoms and adjacent S atoms, consistent with the DOS results. Notably, a pronounced increase in charge density is observed on S atoms near the V_Zn sites, confirming that Zn vacancies facilitate the enrichment of photoexcited electrons on these sulfur atoms and promote efficient charge transfer. In ZIS, the S atoms surrounding Zn exhibit strong electron localization, leading to a high adsorption but poor desorption capability for H*. In contrast, V_Zn-ZIS shows a more uniform redistribution of electron density due to interactions among neighboring S atoms, resulting in a balanced H* adsorption and desorption behavior. The terminal S site adjacent to the V_Zn exhibits an excellent hydrogen adsorption-free energy (ΔG_H*) of 0.19 eV, indicating optimal adsorption/desorption properties that are highly beneficial for the PHER. As present in Figure 6e, the defect states of charge density enhanced by V_Zn near the ZIS’s CB promote the migration of photogenerated carriers around Zn vacancies. In addition, the V_Zn can redistribute the charge density of the surrounding S atoms, resulting in high-efficiency surface HER sites. Also, the V_Zn-ZIS could be obtained by the treatments of 5% HF solution as shown in Figure 6f–i; the introduction of V_Zn does not affect the lattice structure of ZIS but causes progressively weaker visible-light absorption and an increase in the bandgap values, which is beneficial for the separation of photogenerated carriers [73]. Among these samples, V_Zn-ZIS₂₄ exhibited the highest HER rate of 4.18 mmol∙g⁻¹∙h⁻¹, which was 1.42 times that of pristine ZIS. The photocurrent data of V_Zn-ZIS proved the enhanced carrier separation and charge transfer that resulted from the introduction of V_Zn. Additionally, V_Zn-ZIS exhibits a lower Φ and a higher E_f; such energy level alignment facilitates the transfer of photogenerated electrons. Furthermore, Wu et al. prepared ZIS rich with V_Zn via a solvothermal method and fabricated the Pt/V_Zn-ZIS 3D nanoflower photocatalyst (Figure 6j) [74]. During the HER process, the quantum efficiency of 1.0% Pt/ V_Zn-ZIS reached 13.12% at 350 nm and 6.59% at 400 nm (Figure 6k). Platinum atoms occupied the zinc vacancies in ZIS to form Pt-S electron channels, and the ΔG_H* significantly decreased from −0.611 eV to −0.321 eV, which effectively reduced the energy barrier of the HER (Figure 6l) and effectively promoted the reduction in protons by photogenerated electrons. Chong et al., based on the two-dimensional layered structural characteristics of zinc indium sulfide (ZIS) and the specific coordination effect between ethylene glycol and Zn²⁺, incorporated ethylene glycol and zinc vacancies into the intra-layer and inter-layer of ZIS [75]. V_Zn, in synergy with the doped ions, form a strong built-in electric field, promoting the directional migration of photogenerated electron–hole pairs. In addition, V_Zn can serve as an H⁺ adsorption site, reducing the energy barrier of the HER and enhancing the absorption range of visible light.

Compared with Zn²⁺ ions, In³⁺ ions with more positive charges are easier to combine with small molecule chelators. Normally, it is easier for Zn²⁺ ions to form tetrahedral-possible zinc complexes with poor stabilities, while it is easier for In³⁺ ions to form more stable octahedral chelates of indium complex compounds. The addition of small molecule chelators could produce an In vacancy (V_In) in ZIS. Therefore, He et al. prepared a rich V_In- and poor V_In-ZIS via the hydrothermal process by adjusting the temperature from 393 to 473 K every 20 K using excess cysteine (C₃H₇NO₂S) as the chelating agent [76]. The EPR signal intensities at the g-value of 2.005 indicated that the V_In-rich ZIS possessed a higher concentration of V_In than the V_In-poor ZIS (See Figure 7a–c).

Studies on dual vacancies in ZIS for the PHER (HER) have focused on their regulatory role in catalytic performances. As shown in Figure 7d–f, Zhao et al. successfully introduced dual Zn and In vacancies into ZIS flower like photocatalysts via cetyltrimethylammonium chloride (CTAC)-assisted hydrothermal method, where CTAC was used to induce structural defects [77]. A series of ZIS catalysts with different bimetallic defect concentrations were fabricated in the presence of different amounts of CTAC, which denoted them as ZIS-x. The increased intensity of the EPR signal in Figure 7f at a g-factor of 2.003 indicated increased vacancy concentrations in the order of ZIS < ZIS-1 < ZIS-5 < ZIS-10. As shown in Figure 7g, the Zn and In vacancies in ZIS-5 delocalize electrons and lead to charge accumulation around the defect sites from the charge density structures of ZIS and ZIS-5. As shown in Figure 7h–k, Zhong et al. investigated the effect of the concentrations of indium (In) and sulfur (S) vacancies on the photocatalytic activity of non-centrosymmetric zinc indium sulfide (ZIS) nanosheets in the HER, and a positive correlation was observed between the concentration of dual In and S vacancies and the PHER rate on ZIS nanosheets [78]. The synergistic effect of the dual vacancies increases the voltage gradient within the nanosheets and improves the charge separation efficiency after light absorption, thereby significantly enhancing the HER. Overall, these studies collectively demonstrate that different types of dual vacancies (Zn/In and In/S) in ZIS regulate the PHER performances through distinct yet effective mechanisms, and both can ultimately achieve the goal of promoting the HER.

2.3. Doping Engineering (Metal/Non-Metal Doping)

Doping engineering in ZIS, as a core strategy to tune material properties, achieves electronic structure optimization primarily through the strategic implantation of metal/non-metal elements, which effectively creates intermediate bandgap states to regulate electron transfer and energy band alignment [31,79,80]. Transition metals (such as Mn, Co, Ni, Zn, and Fe) can introduce intermediate energy levels that match the conduction bands/valence bands of ZIS, thereby adjusting the electron transition paths [81,82,83,84,85]; noble metals (such as Au, Ag, and Pt; often as nanoparticles formed on the defect sites) can not only form intermediate bandgap states but also synergistically improve carrier utilization efficiency through the surface plasmon resonance effect [86,87]. In addition, other common metals (like Cu and Na) can regulate the electronic transmission characteristics of ZIS by modifying the crystal lattice environment and inducing charge compensation effects, which also contributes to the optimization of the material’s overall electronic structure alongside the intermediate bandgap states [88]. Non-metal doping (e.g., N, S, and P) mainly modifies the band edge positions of ZIS, and when combined with the constructed intermediate bandgap states, it optimizes the range of light absorption and the kinetics of electron transmission [89,90]. Additionally, bimetallic co-doping—including combinations of transition metals (e.g., Co-Ni), noble metals (e.g., Au-Ag and Au-Pt), or transition metals with precious metals—leverages the electronic interaction between different metal ions to achieve precise regulation of intermediate bandgap states, further promoting the separation efficiency of photogenerated carriers [91,92]. In essence, all these doping modes revolve around optimizing the electronic structure of ZIS, laying a foundation for improving its photocatalytic performance in fields.

Noble metals are often used to enhance the performance of photocatalysts. He et al. prepared different structured Pt deposited ZIS photocatalysts, achieving an efficient HER. Under photothermal catalytic conditions (80 °C), the HER amount of 1% Pt/ZIS-300 reached 1218 μmol/(g catalyst) after 4 h of reaction [86]. However, their large-scale application is limited due to scarcity and high costs. Under this constraint, the design of transition metal-doped photocatalysts has become increasingly urgent. As shown in Figure 8a,b, Shen et al. investigated ZIS doped with different transition metals (Cr, Mn, Fe, and Co), and Mn-doped ZIS showed great improvement in the PHER, which may be attributed to the bandgap narrowing caused by Mn doping [93]. However, for Cr-, Fe-, and Co-doped ZIS, the inhibition of their photocatalytic activity should be ascribed to the dopant-related impurity energy levels; these levels can either cause the localization of charge carriers or act as non-radiative recombination centers for photoexcited electrons and holes. Summarily, He et al. achieved the goal of improving the PHER by Mn²⁺ doping and applying an external magnetic field [85]. Compared with ZIS, the H₂ production rate of Mn_0.15-ZIS reached as high as 32.75 mmol·g⁻¹·h⁻¹, which was 13.87 times that of ZIS. This demonstrates great potential in regulating spin-polarized electrons and provides an effective strategy for enhancing photocatalytic performances. As shown in Figure 8c–e, Xue et al. reported cobalt-doped ZIS hierarchical nanotubes (Co-ZIS HNTs) with bifunctions for the HER and benzaldehyde production [82]. Without any cocatalysts, the HER rate of Co-ZIS HNTs reached 2823.7 μmol·h⁻¹·g⁻¹, which was 1.93 times that of undoped ZIS microspheres (ZIS MSs), which may be according to the enhanced separation and transfer efficiency of photoinduced carriers in Co-ZIS HNTs.

Metal doping can regulate charge separation and construct gradient channels for hydrogen migration, thereby enhancing the performance of the PHER. As shown in Figure 8f–h, copper doping generates adaptive V_S and establishes gradient hydrogen migration channels, reducing hydrogen diffusion resistance by 62% and collectively enhancing charge dynamics and HER efficiency [94]. The intensity of EPR signals at a 2.003 g-value in xCu–ZIS samples were related to the self-adapting vs. induced with atomic Cu introduction, which can tune charge separation and construct a gradient channel for H migration. Therefore, all samples exhibit photocatalytic H₂ production under irradiation, with undoped ZIS showing a baseline rate of 0.6640 mmol·g⁻¹·h⁻¹. The H₂ evolution rate increases with Cu dopant concentration, peaking at 9.8647 mmol·g⁻¹·h⁻¹ for 5%Cu-ZIS. Zeng et al. synthesized Sn-doped ZIS nanosheets via a one-step hydrothermal method to realize microstructural optimization, electronic property regulation, and active site modification of two-dimensional (2D) hexagonal-phase ZIS. The HER rate of Sn-ZIS under visible light reaches 62.18 μmol/h [96]. Also, for the modification of ZIS with noble metals and the introduction of defects present another effective regulatory approach, He et al. demonstrated that Ag-modified ZIS ultrathin nanosheets rich with V_S could promote photocatalytic performances, which enriched the functional diversity of ZIS-based photocatalysts through structural and surface engineering [54]. Bimetallic co-doping to form photocatalysts with efficient charge separations and transfer performances is also crucial for a high-efficiency PHER. As shown in Figure 8i–l, Zhu et al. constructed a ZIS catalyst co-doped with Sn²⁺ and Sn⁴⁺ (Sn²⁺/Sn⁴⁺-ZIS) via a one-step solvothermal method [95]. Surface photovoltage (SPV) spectra show that all samples exhibit a strong response within the wavelength between 300 nm and 500 nm, and the Sn²⁺/Sn⁴⁺-ZIS sample has the strongest SPV intensity, indicating that the carrier separation efficiency is much higher than those of other catalysts; thus, the Sn²⁺/Sn⁴⁺-ZIS catalyst exhibited excellent HER catalytic activity (2.48 mmol·g⁻¹·h⁻¹). This strategy enhanced the metallic conductivity of ZIS and reduced the recombination of photogenerated electrons and holes. The co-doping of Sn²⁺ and Sn⁴⁺ ions increased the charge density and narrowed the bandgap for light harvesting. Bimetallic co-doping opens up a promising avenue for the development of high-efficiency photocatalysts for a solar-driven PHER. An et al. constructed a three-dimensional Ag_XAu_1−X alloy/ZIS system (Ag_XAu_1−X). In this system, Ag_XAu_1−X alloy nanoparticles (NPs) are deposited on ZIS microspheres self-assembled from ZIS nanosheets, forming tight contact with ZIS. The HER capacity of Ag_0.6Au_0.4/ZIS was 7.1 times that of pure ZIS, and the apparent quantum yield (AQY) of Ag_0.6Au_0.4/ZIS was approximately 8.06% [91]. Considering the collective excitation of plasmonic nanoparticle assemblies, An et al. constructed a Au@Pt/ZIS photocatalyst through the assembly of core–shell structured Au@Pt nanoparticles on three-dimensional (3D) ZIS microspheres. This photocatalyst exhibited extraordinary catalytic performance in the water splitting for the HER under visible light (≥420 nm) [92]. Under visible light, the HER amount and HER rate of Au₁₆@Pt/ZIS reached 41747 μmol·g⁻¹ and 4174.7 μmol·g⁻¹·h⁻¹, respectively, which were approximately 10 times those of ZIS. Despite having achieved significant progress, doping engineering still faces some persistent challenges and requires further regulation.

Besides metal doping, non-metal doping can induce dual p-n charge characteristics within a single ZIS crystal structure, triggering charge transfer behavior, and promoting the PHER. As shown in Figure 9a, Chong et al. have demonstrated that the p-type characteristics induced by N doping lead to favorable charge redistribution in the ZIS framework [89]. N-doped ZIS (N-ZIS) showed a blue shift in the UV-vis spectra, accompanied by the dopant-induced color change (Figure 9b). The optimal 10N-ZIS remarkably boosts HER performance with the rate of 187.17 μmol∙g⁻¹∙h⁻¹, about 1575.71 μmol∙g⁻¹ that of H₂ after 6 h of visible-light irradiation (Figure 9c,d). To further investigate the changes in electronic properties and charge behavior induced by N doping, we computed the DOS and partial DOS (PDOS). As shown in Figure 9e, the valence band (VB) of pristine ZIS is composed of hybridized S 3p and Zn 3d orbitals, while the conduction band (CB) consists of S 3p and In 5s orbitals. The incorporation of N dopants into the ZIS lattice introduces a deep acceptor level above the valence band maximum. As a result, the Fermi level of N_(4)(4’)ZIS shifts downward toward the VB, indicating a p-type conduction behavior. Moreover, N doping significantly enhances the electron density near the valence band maximum (VBM), suggesting a larger pool of ground-state electrons that can be photoexcited into the conduction band, thereby facilitating participation in the PHER. As shown in Figure 9f–i, Feng et al. reported that for phosphorus (P)-doped ZIS for the PHER, P-ZIS showed a good light absorption ability in the visible-light region, and the absorption edge of P-ZIS-1.0 was a little broader than that of the pristine ZIS, corresponding to its smaller bandgap values [97]. Thus, the P-ZIS exhibited stronger catalytic activity than the pristine ZIS, the HER rate of the optimized P-ZIS was 1566.6 μmol·g⁻¹·h⁻¹, which was 3.8 times that of the ZIS (411.1 μmol·g⁻¹·h⁻¹). As present in Figure 9j–m, Goswami et al. spectroscopically investigated the effects of O and N doped ZIS for the PHER [98]. OZIS and NZIS (ZIS doped with oxygen and nitrogen, respectively) exhibited reduced bandgap values. Furthermore, N and O doping resulted in an upward shift of the conduction band minimum (CBM) by approximately 0.06 eV and 0.21 eV for NZIS and OZIS, respectively, compared with pristine ZIS. The broad PL spectrum observed in the red region for OZIS suggests the existence of closely spaced defect states within the bandgap. Following O doping, the PL intensity undergoes an approximately 2-fold increase, which is likely attributed to the increased number of charge carriers in the system. In the doped ZIS system, the defect-state-mediated HER enabled the O-doped ZIS to result in a three-fold increase in the PHER rate. Also, Zhang et al. synthesized N, O co-doped ZnIn₂S₄ (NO-ZIS) and the N and O co-doping enhanced light absorption, increased specific surface area, and improved charge separation and transfer compared with unmodified ZIS [99]. As a result, the as-prepared NO-ZIS exhibited superior photocatalytic performance in both H₂ evolution and Cr(VI) photoreduction relative to bare ZIS. Specifically, the H₂ evolution rate of NO-ZIS reached 2254.0 μmol·g⁻¹·h⁻¹, which is 2.8 times that of pristine ZIS. In addition, the metal and non-metal co-doped (such as Mo/N, Ni/N, Co/P) ZIS system were also reported with enhanced PHER performances [100,101]. These findings collectively demonstrate the versatility of doping engineering in optimizing ZIS-based photocatalysts.

2.4. Performance Enhancements and Mechanistic Insights

The defect engineering strategies of ZIS effectively enhance photogenerated carrier separation and surface reaction kinetics through atomic-level modification. Based on modification mechanisms and experimental data, we construct a quantitative comparison in

Table 1. Comparison of HER performance of ZIS under different modification strategies.

Catalysts	Light Source	Scavenger	HER Rate (μmol·g−1·h−1)	Ref.
VS-ZIS (with Pt/Cr cocatalysts)	300W Xenon lamp	10 vol% acetone	3.40	[55]
VS-ZIS/Mo2N	300W Xenon lamp	15 vol% TEOA	10,280	[102]
VS-ZIS/Cu	300W Xenon lamp (λ > 420 nm)	0.2 M ascorbic acid	9864.7	[103]
VS-ZIS/Ni-P@C	300W Xenon lamp (λ > 400 nm)	14 vol% TEOA	11,064	[104]
VS-ZIS/CdS	300W Xenon lamp (λ > 420 nm)	10 vol% lactic aqueous	16,630	[105]
VS-ZIS/TpPa-1	300W Xenon lamp (λ > 420 nm)	0.05 M L-ascorbic acid	2745	[106]
VS-ZIS nanosheets	300W Xenon lamp (λ > 420 nm)	10 vol% TEOA	5437	[62]
VS-ZIS (plasma-etched ZIS)	300W Xenon lamp	Na2S/Na2SO3 (0.35 M/0.25 M)	261.9	[65]
VS-ZIS-P (plasma-etched ZIS)	300W Xenon lamp	Na2S/Na2SO3 (0.35 M/0.25 M)	706	[66]
VZn-ZIS nanoflowers	300W Xenon lamp	10 vol% TEOA	21,430	[73]
VZn-ZIS	300W Xenon lamp	10 vol% TEOA	212	[72]
Pt/VZn-ZIS	300W Xenon lamp	4-methylbenzyl alcohol	6410	[74]
Sn-ZIS	300W Xenon lamp (λ > 400 nm)	20 vol% TEOA	62.18	[96]
Pt-ZIS	simulated solar light (λ > 420 nm)	10 vol% TEOA	17,500	[107]
Pt-ZIS	300W Xenon lamp	15 vol% TEOA	577	[86]
Mn-ZIS	300W Xenon lamp	1% benzyl alcohol	32,750	[85]
Ni-ZIS	300W Xenon lamp (λ > 420 nm)	14 vol% TEOA	8910	[108]
Cu-ZIS	300W Xenon lamp	0.2 M ascorbic acid	9864.7	[94]
Pd-ZIS	300W Xenon lamp	0.25 M Na2SO4	2310	[34]
Na-ZIS/CoSe2	300W Xenon lamp	20 vol% TEOA	4525	[88]
Au@Pt/ZIS	Xenon lamp (λ > 420 nm)	Na2S/Na2SO3 (0.35 M/0.25 M)	41,747	[92]
Ag0.6Au0.4/ZIS	300W Xenon lamp	Na2S/Na2SO3 (0.35 M/0.25 M)	54,007	[91]
P-ZIS	300W Xenon lamp (420 nm < λ < 760 nm)	Na2S/Na2SO3 (0.25 M/0.35 M)	1566.6	[97]
N-ZIS/(tungsten-based polyoxometalate)	300W Xenon lamp (λ>320nm)	15 vol% TEOA	17,345.53	[109]
N, O co-doped ZIS	300W Xenon lamp	Na2S/Na2SO3 (0.35 M/0.25 M)	2254	[99]

as follows.

3. Conclusions and Outlooks

Although defect engineering strategies have significantly improved the PHER performance of ZIS as summarized in this review, several critical scientific and technological challenges remain.

(1) The real-time tracking of dynamic evolution at the atomic-scale—particularly the migration of vacancy defects during catalysis—remains unresolved. This limitation obstructs a deeper understanding and precise manipulation of the structure–performance relationship.

(2) For scalable applications, the high cost of noble-metal cocatalysts and the insufficient efficiency of non-precious alternatives present two major obstacles to practical implementation.

(3) The design of bifunctional active sites that can concurrently promote hydrogen evolution along with sacrificial agent oxidation, operate in tandem with the oxygen evolution reaction (OER) for overall water splitting, or facilitate synergistic organic pollutant degradation remains a significant challenge in system integration [110,111].

In the future, bridging the gap between laboratory research and industrial deployment will require integrated efforts across computational modeling, atomic-precision synthesis, and reactor engineering. With guidance from dynamic operando studies enabling precise defect construction, ZIS-based photocatalysts are poised to contribute substantially to the development of scalable solar-driven hydrogen production.