Gradient Built‐In Electric Field Engineering in Hierarchical Co_0.85Se Quantum Dots‐Modified Cadmium Sulfide Nanorods for Boosted Photocatalytic H₂ Evolution

Abstract

Efficient charge separation and durable cocatalyst interfaces are critical for advancing visible‐light photocatalytic hydrogen production. Here, we develop an in situ selenization strategy that anchors defect‐rich Co_0.85Se quantum dots onto CdS nanorods while simultaneously forming a Se‐doped CdS surface layer, yielding a tightly coupled QDs–shell–core heterostructure. This construction establishes a spatial dual built‐in electric field arising from Se‐doping‐induced band modulation and the Co_0.85Se/CdS interfacial junction, which strongly promotes directional electron transfer. Experimental characterizations and DFT calculations jointly reveal that the defect‐rich Co_0.85Se QDs act as efficient electron sinks, accelerating H₂ evolution. Consequently, the optimized photocatalyst delivers a hydrogen evolution rate of 61.66 mmol g⁻¹ h⁻¹ with only 0.7 wt% Co_0.85Se loading, which is far exceeding pristine CdS and outperforming Pt/CdS and presents markedly enhanced photostability. This study offers a practical pathway for designing low‐loading, non‐noble‐metal cocatalyst systems for high‐performance solar hydrogen production.

This work develops an in situ selenization strategy to anchor defect‐rich Co_0.85Se quantum dots on CdS nanorods while simultaneously forming a Se‐doped surface layer, constructing a dual built‐in electric field‐driven heterostructure. This design achieves efficient visible‐light photocatalytic hydrogen evolution at 58.95 mmol g⁻¹ h⁻¹, surpassing Pt/CdS in performance while demonstrating significantly enhanced stability.

1. Introduction

The accelerating depletion of fossil fuel reserves and the resulting environmental pressures, including greenhouse gas accumulation, global warming, and air and water pollution, have underscored the urgent need for sustainable energy technologies [1, 2, 3, 4, 5, 6, 7, 8, 9]. As an abundant, clean, and renewable energy source, solar energy offers a compelling pathway for transitioning toward carbon‐neutral energy systems [10, 11, 12, 13, 14, 15, 16, 17]. Nevertheless, its inherent intermittency and uneven spatiotemporal distribution require efficient conversion and storage strategies capable of transforming intermittent solar irradiation into storable chemical fuels [18, 19, 20, 21, 22, 23]. Photocatalytic water splitting for hydrogen evolution represents one of the most promising approaches due to its simplicity, environmental compatibility, and potential for large‐scale deployment [24, 25, 26, 27, 28, 29].

Cadmium sulfide (CdS) has been widely studied as a classic visible‐light‐responsive photocatalyst owing to its narrow bandgap (~2.4 eV), strong light absorption capacity, and suitable band‐edge positions for hydrogen evolution reactions [30]. However, the practical application of pristine CdS is significantly hampered by several intrinsic limitations, such as the rapid recombination of photogenerated charge carriers, susceptibility to photocorrosion, and a scarcity of surface‐active sites, all of which collectively undermine its photocatalytic efficiency and long‐term stability.

To overcome these challenges in photocatalytic H₂ evolution, various modification strategies have been developed, including elemental doping, the construction of heterojunctions, and the introduction of cocatalysts [31, 32, 33, 34, 35, 36, 37]. Among the available approaches, cocatalyst decoration has attracted considerable interest due to its direct role in promoting surface redox reactions and suppressing charge recombination [27, 38]. Quantum dots (QDs)‐based catalysts represent a particularly promising avenue of research [39, 40, 41]. By leveraging their high specific surface area, abundant lattice defects, and pronounced quantum confinement effects, QD‐based cocatalysts achieve an optimal balance between catalytic activity, structural stability, and synthetic feasibility when compared to bulk or single‐atom counterparts.

In this study, Co_0.85Se QDs were anchored onto the surface of CdS nanorods (NRs) via an in situ selenization process, during which simultaneous Se doping of the CdS surface was achieved. The resulting hierarchical structure consists of three distinct components from the exterior to the interior: Co_0.85Se QDs, a Se‐doped CdS layer, and the pristine CdS core. Unlike single doping or cocatalyst deposition, the present in situ selenization simultaneously generates a Se‐modified CdS surface and anchors defect‐rich Co_0.85Se QDs through Cd–Se–Co interfacial linkages. The resulting QD/shell/core arrangement spatially couples the CdS/Se‐modified CdS and Se‐modified CdS/Co_0.85Se potential gradients, providing directional charge extraction and surface H₂‐evolution sites. This synergistic configuration not only facilitates efficient electron transfer but also improves the surface reaction pathway. As a result, the Co_0.85Se QDs/CdS NRs composite exhibits a remarkable H₂ evolution rate of 61.66 mmol g⁻¹ h⁻¹, which is 6.97 times higher than that of pristine CdS NRs and surpasses the performance of the benchmark Pt/CdS system.

2. Experimental Section

2.1. Materials

All reagents used in this study were of analytical grade and employed without further purification.

2.2. Synthesis of CdS NRs

CdS NRs were synthesized through a solvothermal route. Typically, 4.05 mmol of Cd(NO₃)₂·4H₂O and 12.15 mmol of thiourea were dissolved in 20 mL of ethylenediamine under continuous stirring for 30 min to form a clear precursor solution. The solution was then transferred to a 50 mL Teflon‐lined autoclave and heated at 160 °C for 24 h. After cooling naturally to room temperature, the yellow precipitate was collected by centrifugation, washed repeatedly with deionized water and ethanol, and dried under vacuum at 60 °C to obtain CdS NRs.

2.3. Synthesis of Co_0.85Se

Co_0.85Se was prepared through a hydrothermal method and used as a reference material. A mixed solvent consisting of diethylenetriamine and deionized water (volume ratio 2:1) was first prepared to yield 40 mL of solution. Subsequently, 1 mmol of cobalt acetate tetrahydrate and 1 mmol of sodium selenite were added, followed by ultrasonication for 1.5 h to ensure uniform dispersion. The resulting mixture was transferred to a 50 mL Teflon‐lined autoclave and maintained at 180 °C for 16 h. The obtained black precipitate was recovered by centrifugation, washed several times with water and ethanol, and dried at 60 °C under vacuum.

2.4. Synthesis of Co_0.85Se/CdS Composites

Co_0.85Se/CdS nanocomposites with varying Co_0.85Se loadings were fabricated via an in situ solvothermal deposition process, as shown in Figure 1. First, 200 mg of CdS NRs was dispersed in 40 mL of diethylenetriamine/deionized water mixture (2:1 v/v), followed by the addition of predetermined amounts of cobalt acetate and sodium selenite precursors. The suspension was ultrasonicated for 1.5 h to promote homogeneous mixing. This mixture was then sealed in a 50 mL autoclave and heated at 180 °C for 16 h. After natural cooling, the solids were collected by centrifugation, washed thoroughly with deionized water and ethanol, and vacuum‐dried at 60 °C. Samples were denoted as Co_0.85Se/CdS–X, where X corresponds to the nominal precursor dosage.

FIGURE 1.

Synthetic route of Co_0.85Se/CdS nanocomposite photocatalyst.

2.5. Preparation of Mechanically Mixed Co_0.85Se/CdS

A physically mixed composite (denoted as Co_0.85Se/CdS–grinding) with the same nominal composition as the optimized in situ sample was prepared by manually grinding CdS NRs and presynthesized Co_0.85Se powders in an agate mortar.

2.6. Characterization

X‐ray diffraction (XRD) was conducted on a Bruker D8 diffractometer. Sample morphology and microstructure were examined by field emission scanning electron microscope (FE‐SEM) and transmission electron microscopy (TEM). X‐ray photoelectron spectroscopy (XPS, Al Kα source) was used to analyze surface composition and chemical states. UV–vis diffuse reflectance spectra were collected on a UV‐2600 spectrophotometer using BaSO₄ as the reference. Photoluminescence (PL) spectra were recorded using a Quanta Master fluorometer. The specific surface area was obtained from N₂ adsorption–desorption isotherms at 77 K using the Brunauer–Emmett–Teller (BET) method.

2.7. Photoelectrochemical Measurements

Electrochemical impedance spectroscopy (EIS) and transient photocurrent measurements were performed on a standard workstation using Pt and Ag/AgCl as counter and reference electrodes, respectively. The working electrodes were prepared by coating catalyst ink onto FTO glass (1.0 × 1.5 cm). A 0.2 M Na₂SO₄ solution served as the electrolyte. Mott–Schottky (M–S) analysis was conducted under the same conditions, except that a glassy carbon electrode was used as the working electrode.

2.8. Photocatalytic H₂ Evolution

Photocatalytic H₂ evolution was carried out under illumination from a 300 W Xe lamp (AM 1.5, ~100 mW cm⁻²) using lactic acid (LA) as the sacrificial agent. A suspension containing 10 mg of catalyst and 100 mL of LA solution (LA: water = 1:9) was ultrasonicated for 30 min. The system was degassed for 20 min before irradiation and maintained at 5 °C during the reaction to avoid photothermal interference.

2.9. Theoretical Calculations

DFT calculations were performed using CASTEP with the Perdew‐Burke‐Ernzerhof (PBE) functional and a 380 eV plane‐wave cutoff [42]. Electronic self‐consistency was achieved at an energy tolerance of 2 × 10⁻⁵ eV, and atomic structures were optimized until the residual forces were below 0.02 eV Å⁻¹.

3. Results and Discussion

The crystal structures of CdS, Co_0.85Se, and Co_0.85Se/CdS composites were examined by XRD, as shown in Figure 2a. The diffraction peaks of pristine CdS at 24.8°, 26.5°, 28.2°, 36.6°, 43.7°, 47.8°, 51.8°, 58.3°, 66.8°, 70.9°, and 75.5° correspond to the (100), (002), (101), (102), (110), (103), (201), (202), (203), (211), and (105) planes of hexagonal wurtzite CdS (JCPDS#41‐1049), confirming its high crystallinity. For pure Co0.85Se, characteristic peaks at 33.3°, 44.7°, 50.5°, 60.3°, 61.8°, 69.9°, and 71.3° are indexed to the (101), (102), (110), (103), (112), (202), and (004) planes of hexagonal Co_0.85Se [43]. In the Co_0.85Se/CdS composites, only the diffraction peaks of CdS were clearly observed, and no distinct peaks for Co_0.85Se appeared, which can be ascribed to the ultralow loading content, poor crystallinity, or high dispersion of Co_0.85Se QDs on the CdS surface. The XRD pattern of Se‐doped CdS is shown in Figure S1. The positions of the diffraction peaks do not show obvious changes compared to pure CdS, which proves that the Se doping has not disrupted the crystal structure.

FIGURE 2.

(a) The XRD patterns, (b–e) TEM and HRTEM images of Co_0.85Se/CdS‐2, (f) HAADF image, and (g–k) element distribution.

TEM and high‐resolution TEM (HRTEM) were employed to further investigate the microstructure of the Co_0.85Se/CdS heterostructures (Figure 2b–e). The TEM image revealed that the composite retained a uniform one‐dimensional NR morphology with an average diameter of approximately 20 nm and a length of around 150 nm. The introduction of Co_0.85Se QDs did not alter the overall structure of CdS. In the HRTEM images, lattice fringes with a spacing of 0.33 nm were indexed to the (002) plane of CdS, while spacings of 0.20 and 0.27 nm corresponded to the (101) and (102) planes of Co_0.85Se, respectively [44, 45]. These observations confirmed that Co_0.85Se QDs, with an average diameter below 5 nm, were uniformly anchored on the CdS NRs surface, forming intimate heterointerfaces. Moreover, the Co_0.85Se QDs exhibited partially disordered lattice regions, indicating the presence of abundant surface defects that could serve as active sites for the H₂ evolution reaction. Elemental mapping (Figure 2f–k) demonstrated the homogeneous distribution of Cd, S, Co, and Se elements, further verifying the successful formation of well‐integrated Co_0.85Se/CdS composites.

The optical properties of the prepared samples were evaluated to understand their light‐harvesting behavior. As shown in Figure 3a, CdS presents a typical absorption edge around 550 nm, while Co_0.85Se exhibits broad absorption extending deep into the near‐infrared region. Incorporation of Co_0.85Se significantly enhances visible light absorption and induces a redshift of the absorption edge, and the effect becomes more pronounced with increasing Co_0.85Se content. Tauc plot analyses (Figure 3b) indicate a gradual narrowing of the optical bandgap from 2.22 eV for pristine CdS to 1.93 eV for the composite with the highest Co_0.85Se loading. This bandgap modulation suggests that Co_0.85Se contributes to electronic structure rearrangement, thereby facilitating the generation of additional photocarriers under visible light.

FIGURE 3.

(a) UV–vis spectra and (b) bandgap width of CdS, Co_0.85Se and CdS/Co_0.85Se‐2. (c) N₂ adsorption–desorption isotherms and (d) core size distribution curves of CdS, Co_0.85Se and CdS/Co_0.85Se‐2.

Nitrogen adsorption–desorption tests (Figure 3c,d) show that all samples exhibit type‐III isotherms, characteristic of slit‐like mesopores formed by stacked nanostructures. Pristine CdS has a relatively high surface area (38.57 m² g⁻¹), whereas Co_0.85Se possesses a low value of 1.70 m² g⁻¹. The composite shows a moderate surface area (25.88 m² g⁻¹), which is lower than CdS due to partial pore obstruction caused by QDs deposition.

The surface chemical composition and electronic interaction between CdS Se‐doped CdS and Co_0.85Se were analyzed using XPS, as shown in Figure 4. In the Cd 3d spectra (Figure 4a), the two peaks at around 402 and 408 eV were assigned to Cd 3d_5/2 and Cd 3d_3/2, respectively [46]. In the S 2p spectrum (Figure 4b), peaks at around 158 and 159 eV correspond to S 2p_3/2 and S 2p_{1/2, respectively} [47, 48, 49, 50]. Compared with pristine CdS, both Cd and S peaks in the Se‐doped CdS shifted toward lower binding energies, suggesting electron transfer from Se‐doped CdS to CdS. Such a shift indicates strong interfacial coupling and charge redistribution at the heterojunction interface. In Figure 4c, the Se 3d spectrum, the peaks in Co_0.85Se/CdS‐2 moved to the positive direction compared to the pure Co_0.85Se. In the Co 2p spectrum (Figure 4d), the peaks at 778.33 and 781.31 eV (Co 2p_3/2) and at 793.37 and 797.24 eV (Co 2p_1/2), along with satellite peaks at 786.16 and 802.53 eV, correspond to Co—Se bonds in Co_0.85Se [51, 52, 53, 54, 55]. Upon forming the composite, the Co 2p and Se 3dpeaks slightly shifted toward higher binding energies, consistent with electron migration from Co_0.85Se to CdS [56]. These results collectively confirm the formation of strong interfacial electronic coupling between CdS and Co_0.85Se, which promotes efficient charge transfer and separation. The XPS results of Se 3d in Se‐doped CdS were also characterized and are shown in Figure S2, and the presence of Cd 3d, S 2p, and Se 3d spectrums can prove the successful doping Se.

FIGURE 4.

XPS spectra of (a) Cd 3d, (b) S 2p, (c) Se 3d, and (d) Co 2p.

XPS analysis confirms the formation of a Se‐doped surface layer on CdS NRs during the Co0.85Se‐supported synthesis process, which can be attributed to the use of an excess selenium precursor that facilitates reaction with the CdS surface. To understand how this Se doping influences the electronic structure and interfacial charge transfer behavior, we conducted systematic theoretical calculations on the relevant crystal planes of pristine CdS and Co_0.85Se. As shown in Figures S3 and 5a–c, the band structure and projected density of states (PDOS) analyses reveal that pristine CdS has a bandgap of approximately 1.35 eV. Surface Se doping narrows this value to 0.96 eV, suggesting enhanced electronic conductivity. In contrast, the density of states (DOS) of Co_0.85Se crosses the Fermi level, indicating metallic character and efficient electron transport capability. Work function measurements (Figure 5d–f) further clarify the driving force for interfacial electron transfer. The calculated work functions for CdS (100), Se‐doped CdS (100), and Co_0.85Se (101) surfaces are 5.82, 5.37, and 3.93 eV, respectively. This progressive decrease in work function from the inner CdS to the outer Co0.85Se establishes a sequential energy gradient. Under equilibrium conditions, electrons transfer spontaneously from Co_0.85Se (lower work function) toward Se‐doped CdS and further to pristine CdS (higher work function), forming a stepwise built‐in electric field that promotes charge separation. In the plane‐averaged charge density difference (Figure 5g), the interfacial charge‐density difference and planar‐averaged electrostatic potential of the optimized Co0.85Se/Se‐modified‐CdS model show electron redistribution across the outer junction, while the CdS/Se‐modified‐CdS work‐function difference defines the inner contribution [48]. These calculations, together with the control‐dependent photoelectrochemical response and XPS shifts, support a serial potential gradient. Moreover, the higher electron density of pristine CdS relative to Se‐doped CdS aligns well with the work‐function‐guided electron transfer direction.

FIGURE 5.

(a–c) The PDOS CdS, Se doped CdS, Co_0.85Se, (d–f) the calculated work function results of CdS (1 0 0), Se doped CdS, Co_0.85Se (1 0 1), and (g) the planar‐average charge density of Co_0.85Se QDs on Se doped CdS.

HRTEM confirms that the deposited Co_0.85Se exists in the form of QDs with a particle size below 5 nm. This nanoscale morphology suggests the presence of abundant surface defects, which are expected to influence the photocatalytic H₂ evolution reaction. To clarify how these defect site modulate the H₂ evolution reaction activity, we performed theoretical calculations by selecting Co atoms on the Co_0.85Se (101) facet as hydrogen adsorption sites. A defect model was constructed by removing one Co and one adjacent Se atom near the target Co site, and the Gibbs free energy of hydrogen adsorption (ΔG_H*) was evaluated before and after introducing the defect.

The photocatalytic H₂ evolution reaction performance of Co_0.85Se/CdS‐X composites with varying Co_0.85Se loading amounts was systematically evaluated, and the corresponding results are presented in Figure 6a,b. For the pristine CdS sample, a H₂ evolution reaction rate of 8.46 mmol g⁻¹ h⁻¹ was observed. Notably, after hybridization of Co_0.85Se with CdS, the photocatalytic H₂ evolution reaction activity of the resulting composites was significantly enhanced. Moreover, the H₂ evolution reaction rate showed a strong dependence on the Co_0.85Se loading amount. Among all the synthesized composites, the optimized Co_0.85Se/CdS‐2 sample achieved the highest H₂ evolution rate of 61.66 mmol g⁻¹ h⁻¹. This value is nearly 7 times higher than that of pristine CdS and also markedly superior to the H₂ evolution performance of the Pt/CdS composite (42.70 mmol g⁻¹ h⁻¹). The photocatalytic H₂ evolution rate of Se‐doped CdS is about 0.61 mmol g⁻¹ h⁻¹ (Figure S4), which is much lower than that of pristine CdS. The apparent quantum efficiency (AQE) of Co_0.85Se/CdS‐2 was also calculated, as shown in Figure S8, and the values at 420, 450, and 500 nm were 10.6%, 11.9%, and 6.3%, respectively. This can prove that the surface Co0.85Se are the real active centers. Given the significantly enhanced photocatalytic performance, it is evident that interfacial charge dynamics rather than surface area dominate catalytic activity. We have added a H₂ evolution performance comparison table from the literatures, as shown in Table S2, which compares the photocatalytic H₂ performance using the selenide of cobalt acts as cocatalysts, while CdS or other sulfides serve as the photocatalyst. From the results, it can be seen that the activity of the catalyst we prepared belongs to the leading group.

FIGURE 6.

(a) Time‐dependent profiles of photocatalytic H₂ evolution and (b) corresponding H₂‐evolution rates for various samples. (c) Photocatalytic H₂ generation with various sacrificial agents for Co_0.85Se/CdS‐2. (d) UV–vis absorption spectrum Co_0.85Se/CdS‐2 and the wavelength dependent H₂ production activity, (e) cycling test for H₂ evolution over CdS/Co_0.85Se‐2 and CdS under visible light irradiation, and (f) the attenuation amplitude of activity after achieving cycle stability test.

The optimal loading was identified experimentally from the Co_0.85Se/CdS‐X series. Inductively‐coupled plasma mass spectrometry (ICP‐MS) showed that the best‐performing Co_0.85Se/CdS‐2 sample contained 0.7 wt% Co_0.85Se. Below this loading, the number of electron‐extraction interfaces and surface H₂ evolution sites is insufficient; above it, increased light shielding, surface coverage, and possible recombination centers offset the benefit of additional Co_0.85Se. To elucidate the influence of in situ interface formation on catalytic performance, a mechanically mixed control sample (Co_0.85Se/CdS grinding) with an identical composition was fabricated for comparative studies. This physically mixed control sample exhibited a notably lower H₂ evolution reaction rate of 16.86 mmol g⁻¹ h⁻¹. This result explicitly underscores the indispensable role of the in situ constructed heterointerface in enhancing H₂ generation activity. The outstanding catalytic activity of the CdS/Co_0.85Se‐2 sample can prove the advance of dual built‐in electric field.

The influence of sacrificial reagents on the H₂ evolution rate was also examined (Figure 6c), and the results indicate that the activity achieved using LA as a sacrificial agent is superior to that of vitamin C (Vc), triethanolamine (TEOA), methanol (MeOH), and Na₂S/Na₂SO₃. The wavelength‐dependent photocatalytic experiments (Figure 6d) confirmed that the photocatalytic activity correlated well with the light absorption profile of the composite, further demonstrating the beneficial role of Co_0.85Se in broadening light absorption and promoting visible‐light utilization. The long‐term stability of photocatalysts is a crucial criterion for their practical application. As CdS is prone to photocorrosion, cycling tests were conducted to assess the durability of Co_0.85Se/CdS‐2 (Figures 6e,f and S9). After going through five photocatalytic hydrogen production test cycles, the activity of Co_0.85Se/CdS‐2 was 84.76% of the initial activity, while CdS only had 29.48% of the initial activity after the cycle tests. Postreaction XRD and TEM results have been added for Co_0.85Se/CdS‐2 recovered after five cycles, as shown in Figures S6 and S7. Compared with the fresh sample, the XRD and morphology results after reaction have no obvious change, supporting the phase stability. This indicates that the modified sample has superior corrosion resistance and stability.

The efficient separation and migration of photogenerated charge carriers play a critical role in determining the overall photocatalytic performance. To gain insight into the carrier dynamics within the Co_0.85Se/CdS system, a series of photoelectrochemical and spectroscopic characterizations were conducted. Transient photocurrent response measurements were first carried out to evaluate the photoinduced charge separation efficiency. As illustrated in Figure 7a, all Co_0.85Se‐modified samples exhibited significantly enhanced photocurrent densities compared to those of pristine CdS. Notably, the Co_0.85Se/CdS‐2 composite demonstrated the highest and most stable photocurrent response, indicating superior electron–hole separation and transfer capability [57, 58, 59]. This improvement can be attributed to the role of Co_0.85Se QDs as effective electron sinks, which rapidly extract photogenerated electrons from CdS and thereby suppress charge recombination. EIS analysis provided further evidence for facilitated charge transfer. The Nyquist plots in Figure 7b reveal that pure Co_0.85Se shows the smallest Nyquist arc, consistent with its metallic transport characteristic. Among the CdS‐based samples, Co_0.85Se/CdS‐2 exhibits the smallest arc, indicating that in situ anchoring lowers the interfacial charge transfer resistance relative to pristine CdS and the other CdS‐based controls.

FIGURE 7.

(a) Photocurrent curve, (b) impedance curves, (c) PL curve, (d) electron paramagnetic resonance (EPR) results, ESR spectra of (e) DMPO‐•OH, and (f) DMPO‐•O₂ ⁻.

Co_0.85Se/CdS‐2 possessed the smallest arc radius among all samples, suggesting the lowest charge transfer resistance at the semiconductor–electrolyte interface [60]. This reduced resistance is consistent with the formation of highly conductive pathways through the intimate interfacial contact between Co_0.85Se QDs and CdS NRs.

In Figure 7c, the steady‐state PL spectra showed a pronounced quenching of the fluorescence signal after Co_0.85Se modification, corresponding to suppressed recombination of photogenerated charge carriers and thereby improved separation efficiency in the CdS photocatalyst. Electron paramagnetic resonance (EPR) measurements were performed to evaluate and compare the defect concentrations of the samples. The results reveal a significantly stronger EPR signal intensity for Co_0.85Se/CdS‐2 compared to Co_0.85Se/CdS grinding (Figure 7d), suggesting a substantially higher defect concentration in the former [14].

The catalytic activity toward the hydrogen evolution reaction was further assessed by linear sweep voltammetry (LSV). As presented in Figure S5, the Co_0.85Se/CdS‐2 composite required a lower overpotential and delivered a higher cathodic current density compared to bare CdS, highlighting the favorable hydrogen evolution reaction kinetics enabled by the Co_0.85Se modification.

To corroborate these findings from a radical generation perspective, electron spin resonance (ESR) spectroscopy was performed. As depicted in Figure 7e,f, the Co_0.85Se/CdS‐2 composite generated significantly stronger signals of •O₂ ⁻ and •OH radical species under light irradiation compared to those of pristine CdS. This enhanced radical production directly evidences the improved availability of photogenerated electrons and holes for surface redox reactions, further supporting the conclusion that the Co_0.85Se/CdS heterojunction effectively promotes charge separation and utilization.

To further clarify the band structures of the samples, M–S measurements were performed, with the results shown in Figure 8a–c. All samples displayed positive slopes that confirms they belong to n‐type semiconductors [61, 62, 63]. Subsequent analysis allowed for the determination of flat‐band potentials relative to the normal hydrogen electrode (NHE): CdS exhibited a flat‐band potential of −0.47 V, Co_0.85Se of 0.22 V, and the Co_0.85Se/CdS‐2 composite of −0.36 V. The positive shift in flat‐band potential observed after composite formation provides direct evidence for the migration of photogenerated electrons from CdS to Co_0.85Se. This result not only validates the successful establishment of an effective electron‐transfer interface between the two components but also supports the proposed mechanism for efficient charge carrier separation. To visualize this process more clearly, a schematic energy band alignment model is presented in Figure 8d,e, which explicitly illustrates the charge transfer pathway and further reinforces the rationality of the catalytic mechanism. A tightly coupled QDs/shell/core heterostructure, characterized by defect‐rich Co_{0 .85}Se QDs anchored on a Se‐doped CdS surface, was yielded by an in situ selenization strategy. In this construct, a spatial dual built‐in electric field was formed due to Se doping‐induced band modulation and the Co_{0 .85}Se/CdS interfacial junction. The two built‐in field contributions are spatially distinct but electronically coupled. The inner contribution arises across the CdS/Se‐modified CdS region, whereas the outer contribution is localized at the Se‐modified CdS/Co_0.85Se interface. Their serial arrangement produces a continuous QD/shell/core potential gradient. Following Fermi‐level equilibration, the associated band bending exerts a concerted driving force for photogenerated electron transport toward the surface Co_0.85Se sites while suppressing back transfer and bulk recombination.

FIGURE 8.

Mott–Schottky plots for (a) CdS, (b) Co_0.85Se, (c) Co_0.85Se/CdS‐2, (d) schematic diagram of interface electron transfer caused by work function and formation of built‐in electric field, and (e) schematic diagram of photocatalytic H₂ production.

4. Conclusion

In summary, we have developed a highly efficient Co_0.85Se quantum‐dot‐modified CdS NR photocatalyst through an in situ selenization approach that couples QDs anchoring with simultaneous Se surface doping. Structural and spectroscopic analyses demonstrate that the tightly bonded Cd–Se–Co interface and the gradient electronic structure arising from Se doping form a dual built‐in electric field that significantly promotes directional electron transfer from CdS to Co_0.85Se. As a result, the optimized composite achieves a hydrogen evolution rate of 61.66 mmol g⁻¹ h⁻¹ with only 0.7 wt% Co_0.85Se, surpassing not only pristine CdS but also the Pt/CdS reference catalyst. The system also exhibits markedly improved photostability, attributed to suppressed photocorrosion enabled by accelerated charge extraction. This study establishes a clear mechanistic connection between interfacial electronic coupling, defect‐regulated surface energetics, and macroscopic photocatalytic performance. The insights gained here provide a rational design paradigm for constructing low‐loading, non‐noble‐metal cocatalyst systems capable of highly efficient and durable solar‐driven hydrogen production.

Funding

This study was supported by the National Natural Science Foundation of China (22572148, 22502145, 22005123) and the Natural Science Foundation of Shandong Province, China (ZR2023MB049).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Data Availability Statement

The authors will supply the relevant data in response to reasonable requests.