Bioinspired notched volvox-like nested Z-scheme heterostructure improves solar-energy utilization for high visible-light-driven hydrogen production†

Yi Chang‡ ^a, Bowen Pang‡ ^a, Weiyi Cheng ^b, Penghui Song ^b, Ruijuan Qi ^c, Xiaobing Wang ^b, Zhengyu Bai *^a, Yuming Guo ^b, Nana Ma *^b and Xiaoming Ma *^a
aKey Laboratory of Green Chemical Media and Reactions, Ministry of Education, School of Chemistry and Chemical Engineering, Henan Normal University, Xinxiang, Henan 453007, China. E-mail: mxm@htu.edu.cn
^bCollaborative Innovation Center of Henan Province for Green Manufacturing of Fine Chemicals, Henan Normal University, Xinxiang, Henan 453007, China
^cKey Laboratory of Polar Materials and Devices, Ministry of Education, Department of Electronic Engineering, East China Normal University, Shanghai 200241, China

First published on 24th October 2023

---

---

Introduction

In recent decades, many urgent social and environmental issues, such as the global energy crisis, greenhouse effect, air pollution, and other problems caused by the large-scale use of fossil fuels, have forced the public to seek green and sustainable new energy. Hydrogen energy, one of the cleanest and most high-efficiency new energy sources, has been widely applied in many areas, such as spaceflight, traffic, and fuel cells.¹ Due to the disadvantages of the traditional production of hydrogen, such as the high energy consumption and undesirable yield, developing facile and efficient approaches to generate hydrogen fuel from renewable energy sources has become urgently needed.^2,3 Photocatalytic hydrogen production utilizing green and sustainable solar energy is considered a promising alternative.⁴ The structure and composition of photocatalysts have important impacts on their solar energy utilization, photocatalytic efficiency, and durability. Therefore, the design and construction of efficient photocatalysts with elaborate structures and suitable compositions have been a challenging concern for researchers and are attracting increased attention.⁵

Nature has provided good inspirations to design and synthesize photocatalysts with sophisticated structures for capturing and storing solar energy.⁶ Elaborate spatial natural structures with sufficient active sites and a strong adsorption capability to solar energy endow photosynthetic biosystems with efficient and sustainable solar-to-chemical conversion capabilities.^7,8 The structural complexity, referring to highly sophisticated geometric morphologies, shell architectures, and interior architectures, can not only offer a large specific surface area and abundant active sites but also improve the light-scattering effect and the charge-transport, which would improve the utilization efficiency of solar energy.^9,10 Volvox, a multicellular algae plant, possesses a multi-core nested hollow structure with many small spines on the cell surface, which are helpful for enhancing the surface area and light-scattering ability (Fig. S1†). In addition, the shell and cores with different compositions could facilely adjust the band gaps to expand the light-absorption region of photocatalysts and promote the separation and migration of charge carriers.^11–15 So, heterojunctions with a sophisticated spatial structure could show distinct structural and compositional advantages for solar-energy-conversion reactions.^16–18

Herein, inspired by the nested structure of volvox with multiple sub-cells, a notched volvox-like double-core ZnO/CdS nested heterostructure, which could effectively allow capturing and utilizing solar energy, was constructed according to a facile and universal symbiosis hydrothermal process (Scheme 1a). It is worth mentioning that the unique nested structure could maximize the capture and adsorption of light in photocatalysis. Different from the reported multiple cores,¹⁹ the double cores here had exactly the same morphologies as the notched maternal shell, which looked like a twin of the gravida. Hereof, the double-core nested heterostructure was vividly named the “mother–twin” structure. Furthermore, the number of cores could be controlled accurately by tailoring the concentration of metal salts. CdS nanosheets could be directly grown on the surface of the mother–twin hollow ZnO through a chemical deposition method (see Experimental details in the ESI†). By adjusting the CdS content, a series of mother–twin ZnO/CdS nested heterostructures with different ZnO/CdS ratios were obtained, and were denoted as mother–twin HZC-n: mother–twin HZC-1, mother–twin HZC-2, mother–twin HZC-3, and mother–twin HZC-4 (“n” represents the different mole numbers of CdCl₂ added: 0.02, 0.05, 0.10, and 0.20 mmol, respectively). The mother–twin HZC with a complex spatial structure and tight interfacial contact could enhance the light-scattering effect and improve the charge-transport property. Thereinto, the mother–twin HZC-3 exhibited the highest photocatalytic hydrogen evolution rate, up to 18.70 mmol g⁻¹ h⁻¹, under visible-light driven condition, and good recyclability during the photocatalytic reaction. The excellent photocatalytic performance could be ascribed to the synergetic effect of the complex mother–twin nested hollow structure and heterostructure feature of the ZnO/CdS nanomaterials.

Results and discussion

Bioinspired construction of the volvox-like nested heterostructure

The notched volvox-like ZnO possessing a peculiar “mother–twin” nested hollow structure was constructed by a facile and universal symbiosis hydrothermal process (Scheme 1b). The symbiosis synthetic approach meant that the zinc ionic was adsorbed into carbonaceous microspheres (CMSs) during the CMSs’ formation process via a hydrothermal method. So, more Zn²⁺ could disperse into the CMSs with the increase in the hydrothermal time, which favored the construction of the complex interior structure. In a typical synthetic process, zinc sulfate and glucose solution were mixed first under mild conditions. Subsequently, intact CMSs loaded with Zn²⁺ were produced through a hydrothermal process. The obtained microspheres appeared to be uniform and had an average diameter of about 4.65 μm (Fig. S3†). Then, the CMSs-Zn²⁺ was calcined in an air atmosphere and the CMSs templates were removed to obtain the complex hollow volvox-like ZnO. Of note, the core numbers could be regulated by adjusting the hydrothermal time of Zn²⁺ and glucose, which determined the amount of Zn²⁺ adsorbed into the CMSs (Fig. S4†). Also, the hollow, single-core and dual-core structure of the volvox-like ZnO could be accurately controlled with the symbiosis time increasing from 1 to 40 h, indicating that the enhancement of Zn²⁺ adsorption was beneficial for the formation of the complex interior structure (Fig. 1 and S5†). It is remarkable that there was a moderate opening on the structure of the volvox-like ZnO, which was very beneficial to enlarging the specific surface area. Besides, the size of the volvox-like ZnO slightly increased alongside the complexity enhancement of the interior structure (Fig. S6†). Following the same preparation method, CaCO₃ and La₂O₃ hollow spheres with the same “mother–twin” nested structure were also successfully obtained, indicating that the symbiotic hydrothermal method is a simple and universal process for preparing different materials with “mother–twin” nested hollow structures (Fig. S7†).

The structure and morphology of the notched mother–twin hollow ZnO were determined by FESEM, TEM, HRTEM, and STEM, respectively. The FESEM images showed that the obtained ZnO possessed the uniform and intact volvox-like nested hollow structure (Fig. 1a). The twin cores could be observed clearly inside the ZnO hollow structure (Fig. 1b). Furthermore, the surface of the ZnO shell was rough and consisted of ZnO nano-aggregates fabricated by the assembly of ZnO nanoparticles, just like the small spines on the volvox surface, which could enlarge the specific surface area of the materials to improve the photocatalytic performance (Fig. 1c). The TEM image also confirmed that the uniform ZnO with the unique internal complex structure was constructed successfully, which looked like the volvox with a double-core hollow nested structure (Fig. 1d). The magnified images not only clearly showed the mother–twin shell–core structure but also revealed the porous properties of the shell–core by the obvious contrast between the dark edge and pale center, resulting from the assembly of the ZnO nano-aggregates (Fig. 1e and f). The HRTEM image showed the lattice fringes with a d-spacing of 0.247 nm attributed to the (101) lattice plane of ZnO, while the clear dotted-circle pattern in the selected area electron diffraction (SAED) revealed the polycrystalline structures of the ZnO (Fig. 1g).²⁰ Meanwhile, the result further showed regions of light contrast between the individual ZnO nanoparticles, indicating that the presence of mesopores within the shells, resulting from the assembly of ZnO nanoparticles. Therefore, it could be speculated that hierarchical pores existed in the mother–twin structure of ZnO. To be specific, the hierarchical porous features could supply more channels and mean less diffusion blockage, which would lead to a more efficient light-scattering ability and transport of mass/charge when compared with the single pores.²¹ The X-ray diffraction (XRD) analysis confirmed that the samples were hexagonal ZnO with good crystallinity (JCPDS 36-1451), while Fourier-transform infrared (FT-IR) spectroscopy of the sample also confirmed the formation of ZnO (Fig. S8†). The STEM image clearly showed the volvox-like nested hollow structure of the mother–twin ZnO, while the corresponding elemental mapping images further proved that Zn and O were uniformly dispersed in the sample (Fig. 1h–k). The chemical states of Zn and O were confirmed by X-ray photoelectron spectroscopy (XPS) (Fig. S9†). Zn 2p_3/2 and Zn 2p_1/2 peaks located at 1021.1 and 1044.2 eV were observed and assigned to the typical Zn²⁺ characteristics (Fig. S9a†). The O 1s peaks located at 530.0 and 531.1 eV were assigned to the metal–oxygen bonds of the typical Zn–O and the surface hydroxyl oxygen, respectively (Fig. S9b†).²²

To further determine the hierarchical pores in the nested hollow structure, ZnO with hollow, single-core, and dual-core (mother–twin) structures were characterized by nitrogen adsorption–desorption isotherms, respectively (Fig. S10†). The three samples exhibited typical type IV with H3 type hysteresis loops in the relatively high pressure region (P/P₀ = 0.8), indicating that the samples had complex pore structures.²³ It could be observed that the mother–twin hollow ZnO possessed the largest specific surface area (45.82 m² g⁻¹), and the special mesopores (3.84 nm and 12.80 nm) verified the existence of a hierarchical mesoporous structure, which is beneficial to increasing the accessible reaction sites for photocatalysis (Table S1†). The combination of the complex hollow structure and the special hierarchical pores endowed the volvox-like ZnO with both a large specific surface area and good permeability.

In order to discuss the evolution processes of the mother–twin hollow structure, a series of samples obtained at different calcination times (from 0.5 to 3.0 h) were analyzed. From the FESEM and TEM images (Scheme 1c and d), it was found that the volvox-like appearance of the ZnO/C composites had gradually taken shape from solid CMSs, then turned into core–shell hollow spheres, and finally to mother–twin hollow particles with the increase in calcination time from 0.5 to 3.0 h. During the calcination process, the ZnO shell was prepared by the decomposition of ZnSO₄ adsorbed into the outer layer of the CMSs along with the combustion of the outer layer of the CMSs. Meanwhile, the inner part of CMSs with Zn²⁺ adsorbed as the core still remained. With the conduction of heat continuously from the outer layer shell to the inner core after calcination for 1.5 h, the core showed a splitting trend when enough Zn²⁺ was loaded into the inner CMSs (as the dotted-line shows in Scheme 1d). Accompanied by continuous heating, the single core completely split into two smaller twin mortar-shaped ZnO cores along with the complete combustion of the CMSs. The XRD results of the products at different calcination stages clearly showed that the characteristic peaks of ZnO gradually became more obvious with the increase in calcination time (Fig. S11a†). The stretching vibration peak of Zn–O at 425 cm⁻¹ of the final product also indicated the formation of ZnO (Fig. S11b†).²⁴

Due to the strong oxidation ability of the photogenerated holes from ZnO and the high reduction power of the photoexcited electrons of the metal sulfides, the ZnO/metal sulfides heterostructures appear to be promising photocatalysts for hydrogen evolution.^25–27 Herein, mother–twin hollow ZnO/CdS (mother–twin HZC) with volvox-like nested structures were facilely constructed by the direct growth of CdS nanosheets onto the mother–twin HZ through a chemical deposition method (Fig. 2). In order to discuss the influence of different CdS contents on the structure and photocatalytic performances of the mother–twin HZC, a series of volvox-like mother–twin HZC compounds with different ZnO/CdS ratios were obtained by adjusting the CdS content (Table S2†). Comparing the XRD patterns of the mother–twin ZnO, there was an increasing CdS peak (101) at around 28.1° in the mother–twin HZC with the increase in the CdS mass ratio, indicating the formation of CdS (JCPDS no. 41-1049) (Fig. 2a and S12†). Furthermore, the CdS nanosheets on the surface of ZnO also became more and more obvious with the increase in the CdS content (Fig. S13†).

The series of as-prepared heterostructures still displayed the volvox-like mother–twin nested structure with an average size of about 2.10 μm (Fig. 2b and c), indicating that the successful deposition of CdS nanosheets hardly affected the morphology and unique structure of the host ZnO matrix. The magnified FESEM images showed that the CdS nanosheets consisted of small nanoparticles, indicating the hierarchical structure of the nanosheets (Fig. 2d). The heterostructure of the mother–twin HZC was further investigated by HRTEM. A first observation was that there were two distinguishable domains in the HRTEM image (Fig. 2e). Otherwise, it could be found that the CdS nanosheets were composed of uniform CdS nanoparticles (about 10 nm) with good crystallinity (Fig. 2f and S14†). Simultaneously, the clear crystal lattice fringes and the distinct interface region at the boundary of the ZnO and CdS nanoparticles revealed the successful formation of the heterostructure between the two phases (Fig. 2f).^28–30 The marked d-spacing of 0.316 nm in the outside corresponded to the (101) lattice fringes of CdS. The interplanar distance of 0.247 nm labeled in the inside region matched well with the (101) lattice fringes of ZnO. Furthermore, there were intimate interfaces observed between the ZnO and CdS nanoparticles, which may enhance the charge-carrier separation efficiently (Fig. S15†). Besides, the STEM images showed that the newly formed CdS nanosheets were deposited on not only the surface of the ZnO shell but also the surface of the double cores (Fig. 2g–k). The complex hollow ZnO/CdS heterostructures inside and outside could help to reduce the charge-transport distance and enhance the light-scattering ability, which would further improve the photocatalytic performance.^31–33

The surface compositions and chemical states of the mother–twin HZC were investigated by XPS. The high-resolution XPS spectra of the mother–twin HZC, mother–twin HZ, and pure CdS without light irradiation are shown in Fig. S16.† In comparison with the sample of the mother–twin HZ, the characteristic peaks of Zn (Zn 2p_3/2 and Zn 2p_1/2 states) at 1020.8 and 1043.9 eV, and O (the lattice oxygen species and the chemisorbed oxygen species) at 529.8 and 531.1 eV, respectively, in the mother–twin HZC showed obvious shifts to the lower energy direction (Fig. S16a and b†). In contrast, the characteristic peaks of Cd (3d_5/2 and 3d_3/2 states) at 404.9 and 411.7 eV, and S (S 2p_3/2 and S 2p_1/2 states) at 161.1 and 162.2 eV, respectively, in the mother–twin HZC exhibited shifts to the higher energy direction compared to those of the pure CdS (Fig. S16c and d†). These indicated the strong interaction and possible chemical bonding between ZnO nanoparticles and CdS nanosheets,^34–36 and the electrons will migrate from the decoration of CdS to ZnO in the mother–twin HZC. The XPS analysis of different samples before and after light irradiation is discussed in the following photocatalytic mechanism section.

The hierarchically porous features of the mother–twin HZ were still retained after the growth of CdS nanosheets on the volvox-like nested structure, and were characterized by nitrogen adsorption–desorption isotherms (Fig. S17†). Macropores in the ZnO/CdS heterostructures could be observed, which could be ascribed to the aggregation of CdS nanosheets. In addition, the specific surface areas of the heterostructures were improved distinctly with the increase in the CdS nanosheets content, which may provide more active sites for the reactions to enhance the photocatalytic performances (Table S3†).³⁷

Photocatalytic performance analysis

The photocatalytic activities for hydrogen production over the as-prepared samples were tested using Na₂S and Na₂SO₃ as sacrificial agents and without the addition of any co-catalysts under light illumination. To exclude the interference of photocatalysis in the dark, experiments were carried out in the dark under the same conditions in the beginning. No hydrogen could be detected in the test groups without light irradiation or a photocatalyst. Then, the influence of the mass of the photocatalysts on the photocatalytic performances was studied. As shown in Fig. S18,† the hydrogen production using different amounts of mother–twin HZC-3 and mother–twin HZC with different ZnO/CdS ratios were measured, respectively. It could be observed that the photocatalytic hydrogen evolution under the set conditions increased gradually from 15 to 25 mg, while the photocatalytic hydrogen evolution decreased from 25 to 35 mg, indicating that the series of mother–twin HZC samples had the optimal hydrogen production efficiency when the mass was about 25 mg.³⁸ This could be ascribed to the fact that an excessive amount of photocatalyst may hinder the light utilization, which is not helpful to improving the photocatalytic performance.^39,40

To reveal the advantages of the unique volvox-like mother–twin nested structure and the synergistic heterojunction of ZnO/CdS in photocatalytic hydrogen evolution, the photocatalytic performances of mother–twin hollow ZnO (HZ), commercial ZnO, and CdS were also determined for comparison, respectively. As shown in Fig. 3a, the hydrogen yield of HZC was much higher than that of CdS, HZ, and commercial ZnO. The samples with a single component exhibited fairly low photocatalytic performance due to the rapid recombination of photogenerated electrons and holes. Accompanied by the introduction of the ZnO/CdS heterojunction, the hydrogen evolution of HZC rose sharply. This should be ascribed to the formation of a heterojunction, which adjusted the band gaps for a better visible-light-absorption capacity and promoted the charge migration for higher photocatalytic efficiency. In addition, by comparison with the hydrogen yield of commercial ZnO (0.12 mmol g⁻¹ h⁻¹), HZ with the volvox-like mother–twin structure obviously exhibited a higher hydrogen evolution activity (0.88 mmol g⁻¹ h⁻¹). Thanks to the multiple reflections of the visible light inside the interior cavities of the fascinating notched mother–twin hollow structure, an appreciable enhancement of the visible-light absorption can be achieved (Fig. S19† and the inset of Fig. 3a).⁴¹

Furthermore, the H₂ production rates of the HZC heterostructures exhibited an obvious difference when there was an increasing amount of CdS. For the series of HZC samples, the H₂ production rate of HZC-3 increased and climbed sharply, finally giving the highest yield (18.70 mmol g⁻¹ h⁻¹) (Fig. 3b). The enhancement of the photocatalytic performance varied obviously along with the loading amount of CdS on the surface of ZnO. It is worth noting that the H₂ production of the mother–twin HZC-4 with the highest amount of CdS dramatically decreased to 6.79 mmol g⁻¹ h⁻¹. It could be speculated that an excess of CdS nanosheets may block the interior pores of the mother–twin hollow ZnO, which would obstruct the direct interaction between ZnO and the sacrificial agents. In addition, the excess CdS nanosheets here just took the role of electron–hole recombination sites, which reduced the photocatalytic performances of the sample.⁴²

After consulting the literature about the hydrogen evolution rate of the ZnO/CdS heterostructure, we found that the hydrogen evolution rate of the mother–twin HZC-3 was the highest in those heterostructures without co-catalysts (Table S4†). To prove that hydrogen production was driven by light, the apparent quantum efficiencies (AQE) of the mother–twin HZC-3 were tested at monochromatic wavelengths of 365, 380, 400, and 420 nm, respectively, as shown in Fig. 3c and Table S5.† The mother–twin HZC-3 achieved an AQE value of 33.98% at 365 nm, and the AQE decreased with increasing the wavelength, indicating that the photocatalytic efficiency of the monochromatic wavelengths had an obvious wavelength dependence, which was consistent with the light-absorption properties. For a comprehensive evaluation to the AQE values of the as-prepared photocatalysts, parallel experiments in the dark were also carried out. The results showed that there was no hydrogen production without a light source, which ruled out the contribution of possible mechanisms operating in the dark in the hydrogen evolution process.⁴³

In order to determine the durability of the photocatalysts, a cyclic H₂ production experiment was carried out on the mother–twin HZC-3, which showed the optimal photocatalytic performance among the series tested. There was an attractive result that the photocatalytic performance of the mother–twin HZC-3 in the first cycle was higher than in the other cycles, as shown in Fig. 3d. The reason for this may lie in the formation of ZnS in the first cycle of H₂ production, which was caused by the slight photocorrosion at the heterojunction surface during the photocatalytic reaction (Fig. S20a†).⁴² As the cyclic experiments went on, the mother–twin HZC-3 exhibited a good durability without an obvious decrease during the 20 h cycling test. Also, the morphology of the mother–twin HZC-3 did not show obvious changes after the photocatalytic test, demonstrating the desired structure stability of the mother–twin HZC-3 (Fig. S20b and c†). Similarly, the morphologies of the other proportions of ZnO/CdS with the mother–twin structure also did not show obvious changes after hydrogen production, accompanied by the original disappearance of ZnO and the appearance of ZnS (Fig. S21†). These results suggest the reasonable stability and effective photocatalytic performance of HZC in hydrogen production.

Photocatalytic mechanism for the enhanced performance

The optical properties of photocatalysts, which are greatly affected by their structure and internal electron-transfer behavior, are important indicators for evaluation of the photocatalytic activities.²⁵ The light-harvesting abilities of all the as-prepared photocatalysts were thus investigated by their UV-visible light-absorption spectra (Fig. 4a and S22†). It was found that commercial ZnO only exhibited strong absorption in the UV region because of its broad band gap, while CdS with a relatively narrow band gap showed extremely strong absorption in a broader region, from the ultraviolet region to the red light region (∼600 nm). Compared to the pure ZnO and CdS phase, the absorption band edges of the series of HZC samples with different contents of CdS deposited exhibited an obvious red-shift toward the long wave direction. Two spatially distinguishable absorption band edges of 400 and 500 nm, corresponding to the bandgaps of ZnO and CdS, were observed in the CdS/ZnO heterostructures. Also, the degrees of this absorption shift were relative to the amount of CdS deposited on the surface of ZnO, indicating the successful construction of an intimate contact between ZnO and CdS.⁴⁴ The Tauc plots for optical-band-gap determination of all the samples are shown in Fig. 4b and S23.† According to the equation (Rhν)ⁿ = A(hν − E_g),⁴⁵ the band gaps (E_g) of CdS, commercial ZnO, and the HZ, HZC-1, HZC-2, HZC-3, and HZC-4 heterostructures were calculated as 2.28, 3.26, 3.24, 3.11, 3.10, 3.03, and 3.07 eV, respectively. It could be found that HZC-3 had a lower band gap energy than ZnO and the other HZC samples, which could be ascribed to the just-right ratio of CdS deposited on the surface of ZnO.

The separation and transfer efficiency of photogenerated charge carriers play a key role in the photocatalytic hydrogen production. The separation and transfer behavior of photoexcited carriers can be investigated by photoluminescence (PL) spectroscopy and their photocurrent response. The PL spectra of ZnO and the CdS/ZnO heterostructures at 380 nm excitation are shown in Fig. 4c. Evidently, the commercial ZnO showed the strongest emission at the wavelength of around 563 nm, which could be ascribed to the quick recombination of photogenerated electrons and holes. The HZ series showed a slight decrease in the PL spectra compared to the commercial ZnO, proving that the construction of a complex structure indeed helped to improve the separation of photogenerated electrons and holes. After the introduction of CdS, the HZC heterostructures exhibited an apparent decrease in PL intensity, indicating the significant inhibition effect on the recombination of photogenerated electrons and holes due to the formation of the CdS/ZnO heterojunction. However, the increased PL intensity of HZC-4 compared with HZC-3 implied that increasing the concentration of CdS in the heterostructures would not increase the inhibition effect constantly, and only a suitable proportion between CdS and ZnO could result in the optimal efficient separation of photogenerated electrons and holes. When the content of CdS exceeding the optimal ratio to ZnO, the process that holes generated in ZnO pass through CdS to react with sacrificial agents was interrupted. Besides, when the holes pass through CdS, a recombination of photogenerated electrons and holes is likely to happen. Therefore, excessive CdS here acted as an additional electron–hole recombination center, which eventually inhibited the photocatalytic activity of the samples.⁴²

To investigate the enhanced separation and migration ability of photogenerated electron–holes in the hollow nested heterostructures, transient photocurrent responses and electrochemical impedance spectroscopy (EIS) tests of HZC were carried out. The transient photocurrent–time curves of the as-prepared samples were measured by several on–off cycles of intermittent illumination under visible irradiation (Fig. 4d). Due to the fastest recombination rate of photogenerated charge carriers in all the test samples, HZ showed the lowest photocurrent response.⁴⁶ It could be obviously seen that the photocurrent responses of the series of HZC were higher than that of pure ZnO, which could be attributed to the relatively narrower band gaps derived from CdS. Furthermore, compared with the other three HZC samples, HZC-3 exhibited the strongest photocurrent intensity, which could be attributed to the effective separation and migration of photogenerated electrons and holes from the package of moderate CdS nanosheets.

In addition, the resistance of electrode/electrolyte interfaces can be shown by the radius of each arc, where a smaller radius reflects a lower charge-transfer resistance.⁴⁷ As Fig. 4e and S24† show, the arc radius of HZC was smaller than that of pure ZnO and CdS, indicating that the charge-transfer resistance of the heterostructures was lower than that of the single components. Due to the existence of the heterojunction, the photogenerated carriers appeared to separate and transfer more effectively, which is essential to improving the photocatalytic activity. However, it did not show a linear relationship between the changes in the charge-transfer resistance and the amount of CdS deposited. Also, HZC-3 showed the highest charge-transfer rate presented in these heterojunctions. So, to achieve the optimal charge-transfer property, it is necessary to optimize the ratio of the components during the construction of heterostructures for photocatalysis.

The possible transfer mechanism of charge carriers in the mother–twin nested heterostructure was explored by the detection of hydroxyl radicals (˙OH) carried out over CdS, HZ, and HZC-3. Terephthalic acid (TA) is generally utilized as a probe molecule, which can react with the formed ˙OH to generate the fluorescent agent 2-hydroxyterephthalic acid (TAOH), and the fluorescence intensity is directly proportional to the amount of ˙OH produced.⁴⁸ It is known that the ˙OH generated during the photocatalytic process may be derived from two different ways in the ZnO/CdS heterostructure.⁴⁹ One comes from the photogenerated holes of ZnO, which can oxidize the hydroxyl ions (OH⁻) or H₂O to get the ˙OH. The other way refers to the photo-induced electrons generated from either ZnO or CdS, which reduce the dissolved oxygen to the form of ˙O²⁻. Then ˙O²⁻ will further reduce the H⁺ to ˙OH. To reduce confusion during the investigation of the photocatalytic mechanism based on HZC system, strict deoxygenation operation was carried out before the measurement to exclude the interference of the ˙OH generated from oxygen. Then, the ˙OH detected could be simply regarded as the oxidation product of (OH⁻) or H₂O. It can be found in Fig. 4f that the PL signals of HZ were obvious and increased continuously with the extension of the irradiation time, which was consistent with the mechanism of the ˙OH generation according to the oxidation route. In strong contrast, there was almost no fluorescence intensity in the presence of CdS. However, we could see a gradual increase in the PL intensity after the construction of the ZnO/CdS heterostructure in HZC-3 with the increase in the irradiation time. This means that the photo-induced holes of ZnO in this heterostructure still stayed in its valence band and they did not transfer to the valence band of CdS. Therefore, the heterojunction should not be the one with the characteristic charge transfer of a type-II. Actually, a direct Z-scheme heterojunction should have been naturally formed between ZnO and CdS in HZC.

To further explore the charge migration in the constructed heterojunction, theory calculations were applied to investigate the electron property and electrostatic potentials of CdS and ZnO.^38,49 The work functions of CdS and ZnO were calculated relative to the vacuum energy level. From Fig. 5a and b, the work functions of CdS(101) and ZnO(101) were calculated to be 6.31 and 6.93 eV, respectively, indicating that the electrons in CdS will flow to the ZnO from the surface of the CdS to set a balance of the Fermi level between them. Further charge-difference calculations were performed based on the formed heterojunction ZnO(101)/CdS(101) (Fig. 5c). The results showed that there were newly formed Cd–O bonds with an average bond length of 2.25 Å at the interface of the heterojunction, and the charge-difference calculations also suggested that the charge transfer was from CdS to ZnO, which would form a built-in electric field with the electric field direction from CdS to ZnO at the interface of the heterojunction. Additionally, the density of states in Fig. 5d proved that the VB and CB of CdS were more negative than those of ZnO, respectively, which were in accord with the band structure requirements of the Z-scheme heterojunction.

The charge-transfer mode was further confirmed by the XPS measurement before and after light irradiation. As shown in Fig. 5e and f, the binding energies of Zn 2p in the mother–twin HZC after light irradiation showed a 0.2 eV shift to the higher energy direction in comparison with the binding energies before it was treated with light irradiation. By contrast, the binding energies of Cd 3d in the mother–twin HZC exhibited a shift of about 0.2 eV to the lower energy direction compared with the sample tested in the dark. This proved that the electrons transferred from ZnO to CdS under illumination, which conformed with a direct Z-scheme mechanism. That is, when the mother–twin HZC heterostructure was excited by visible-light irradiation, electrons in the VB of ZnO and CdS were excited to their respective CB, leaving holes in their VB. Due to the internal electric field between ZnO and CdS, the electrons in the ZnO CB recombined with the holes in the CdS VB through the interface, leaving electrons in the CdS CB and holes in the ZnO VB, which could efficiently enhance the separation and migration of photogenerated electrons and holes.

Based on the experimental results of photocatalytic H₂ production, a possible photocatalytic mechanism of the mother–twin ZnO/CdS is proposed and illustrated in Scheme 2a. It could be observed that the VB position of ZnO (+2.96 V vs. NHE) (Fig. S25a†) was more positive than the E(˙OH/OH⁻) (+2.38 V vs. NHE) and E(˙OH/H₂O) (+2.27 V vs. NHE),⁵⁰ indicating that the holes in the VB of ZnO can react with adsorbed water molecules (or surface hydroxyl groups) to generate ˙OH radicals. In contrast, the VB position of CdS (+1.78 V vs. NHE) was more negative (Fig. S25b†), so the holes in VB cannot react with these groups to generate ˙OH radicals. Combined with the results of the ˙OH-capture experiments, it was confirmed that the holes originated from the VB of ZnO. The CB positions of ZnO and CdS could be obtained from the VB and band gap, which were −0.30 eV and −0.50 eV, respectively. Under light irradiation, the electrons are excited to the CB and the holes are left in the VB of ZnO and CdS, respectively. Subjected to the built-in electric field, the electrons transfer from the CB of ZnO to the VB of CdS, resulting in electrons being retained in the CB of CdS and holes staying in the VB of ZnO. Then the holes in the VB of ZnO will be consumed by the sacrificial agent, while the accumulated electrons in the CB of CdS are combined with H⁺ to produce hydrogen. The conduction band potential of the polar semiconducting CdS was more negative than the redox potential of standard H⁺/H₂ (0 V vs. NHE), which is the most important requirement for evolution of hydrogen.

As for the hydrogen evolution process, the generated H* would combine to produce H₂, and the free energy of hydrogen adsorption (ΔG_H*) on the surface of photocatalyst is often used as a good indicator of the activity of hydrogen evolution. Generally, the ΔG_H* value of the hydrogen atom to this site should be close to 0 eV, which indicates that the catalyst has better active sites for H₂ evolution.⁵¹ Herein, the ΔG_H* values of different active sites in HZC-3, CdS, and ZnO were calculated as shown in Scheme 2b. According to the calculated ΔG_H* values, the optimal hydrogen adsorption energies were 0.05, −0.84, and −2.41 eV for HZC-3, CdS, and ZnO, respectively. Compared to CdS and ZnO, the H binding free energy (ΔG_H*) of the S site on the HZC-3 surface (0.05 eV) was very close to the ideal optimization value, showing a relatively high catalytic activity and the advantages of running the H evolution reaction. Therefore, HZC-3 is an efficient catalyst for hydrogen evolution due to it having the most suitable hydrogen adsorption energy. The results indicate that the surface of HZC-3 has a higher electron density for adsorbing H⁺ due to the efficient separation and transfer of photoexcited charges from CdS to ZnO in HZC-3 through the interface between ZnO and CdS.

Conclusions

In summary, bioinspired by the multi-core nested hollow structure of volvox, a high-efficiency hierarchical ZnO/CdS heterostructure for hydrogen production was rationally designed by combining the merits of mother–twin hollow nested structures and a hybrid system. The shell and core structure could be regulated according to adjusting the precursor amount. The mother–twin hollow ZnO/CdS heterostructures were successfully constructed by the direct chemical deposition of CdS nanosheets on the surface of mother–twin ZnO. The CdS molar ratio in the ZnO/CdS composite was optimized to achieve the maximum hydrogen production rate of 18.70 mmol g⁻¹ h⁻¹ under visible-light irradiation with the Na₂S–Na₂SO₃ sacrificial system. Further functional investigation showed that this rationally designed system also exhibited excellent stability in photocatalytic H₂ generation. The hydroxyl radical detection experiments and the work function calculations verified the direct Z-scheme charge-transfer mechanism between ZnO and CdS under light illumination. The developed synthetic strategy may provide new inspirations for the design and construction of high-performance photocatalysts with advanced structures for solar-to-chemical energy conversion.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (21877027 and 21601052); Science and Technology Innovation Talents in Universities of Henan Province (23HASTIT002); Natural Science Foundation of Henan Province (212300410009); Science and Technology Innovation Teams in Universities of Henan Province (19IRTSTHN023); Henan Center for Outstanding Overseas Scientists (GZS2022017).

Notes and references

1. I. Staffell, D. Scamman, A. V. Abad, P. Balcombe, P. E. Dodds, P. Ekins, N. Shah and K. R. Ward, Energy Environ. Sci., 2019, 12, 463–491 RSC .
2. C. Higman and S. Tam, Chem. Rev., 2014, 114, 1673–1708 CrossRef CAS PubMed .
3. D. Kim, K. K. Sakimoto, D. Hong and P. Yang, Angew. Chem., Int. Ed., 2015, 54, 3259–3266 CrossRef CAS PubMed .
4. H. Sheng, J. Wang, J. Huang, Z. Li, G. Ren, L. Zhang, L. Yu, M. Zhao, X. Li, G. Li, N. Wang, C. Shen and G. Lu, Nat. Commun., 2023, 14, 1528 CrossRef CAS .
5. Y. Goto, T. Hisatomi, Q. Wang, T. Higashi, K. Ishikiriyama, T. Maeda, Y. Sakata, S. Okunaka, H. Tokudome, M. Katayama, S. Akiyama, H. Nishiyama, Y. Inoue, T. Takewaki, T. Setoyama, T. Minegishi, T. Takata, T. Yamada and K. Domen, Joule, 2018, 2, 509–520 CrossRef CAS .
6. C. B. Stefano, H. Zhang, J. M. Kim, Y. Shen and P. D. Yang, Nat. Catal., 2020, 3, 245–255 CrossRef .
7. J. Aizenberg, J. C. Weaver, M. S. Thanawala, V. C. Sundar, D. E. Morse and P. Fratzl, Science, 2005, 309, 275–278 CrossRef CAS .
8. R. Croce and H. van Amerongen, Nat. Chem. Biol., 2014, 10, 492–501 CrossRef CAS .
9. X. Lai, J. E. Halpert and D. Wang, Energy Environ. Sci., 2012, 5, 5604–5618 RSC .
10. P. Zhang, B. Y. Guan, L. Yu and X. W. Lou, Chem, 2018, 4, 162–173 CAS .
11. J. Qi, X. Lai, J. Wang, H. Tang, H. Ren, Y. Yang, Q. Jin, L. Zhang, R. Yu, G. Ma, Z. Su, H. Zhao and D. Wang, Chem. Soc. Rev., 2015, 44, 6749–6773 RSC .
12. J. Zhao, J. Wang, R. Bi, M. Yang, J. Wan, H. Jiang, L. Gu and D. Wang, Angew. Chem., Int. Ed., 2021, 60, 25719–25722 CrossRef CAS .
13. X. Xia, Y. Wang, A. Ruditskiy and Y. Xia, Adv. Mater., 2013, 25, 6313–6333 CrossRef CAS PubMed .
14. J. Liu, N. P. Wickramaratne, S. Z. Qiao and M. Jaroniec, Nat. Mater., 2015, 14, 763–774 CrossRef CAS .
15. H. Zhang, Z. Wang, J. Zhang and K. Dai, Chin. J. Catal., 2023, 49, 42–67 CrossRef CAS .
16. P. Zhang and X. W. Lou, Adv. Mater., 2019, 31, 1900281 CrossRef PubMed .
17. Y. Wei, J. Wan, J. Wang, X. Zhang, R. Yu, N. Yang and D. Wang, Small, 2021, 17, 2005345 CrossRef CAS .
18. L. Liu, Z. Wang, J. Zhang, O. Ruzimuradov, K. Dai and J. Low, Adv. Mater., 2023, 35, e2300643 CrossRef .
19. L. Yu, H. Hu, H. B. Wu and X. W. Lou, Adv. Mater., 2017, 29, 1604563 CrossRef .
20. S. Bai, J. Guo, X. Shu, X. Xiang, R. Luo, D. Li, A. Chen and C. C. Liu, Sens. Actuators, B, 2017, 245, 359–368 CrossRef CAS .
21. Y. Zhao and L. Jiang, Adv. Mater., 2009, 21, 3621–3638 CrossRef CAS .
22. T. Chen, S. Yu, X. Fang, H. Huang, L. Li, X. Wang and H. Wang, Appl. Surf. Sci., 2016, 389, 303–310 CrossRef CAS .
23. Z. Dong, Y. Wu, N. Thirugnanam and G. Li, Appl. Surf. Sci., 2018, 430, 293–300 CrossRef CAS .
24. J. Wang, G. Wang, J. Jiang, Z. Wan, Y. Su and H. Tang, J. Colloid Interface Sci., 2020, 564, 322–332 CrossRef CAS .
25. D. Ma, J.-W. Shi, Y. Zou, Z. Fan, X. Ji and C. Niu, ACS Appl. Mater. Interfaces, 2017, 9, 25377–25386 CrossRef CAS PubMed .
26. S. Khanchandani, S. Kundu, A. Patra and A. K. Ganguli, J. Phys. Chem. C, 2012, 116, 23653–23662 CrossRef CAS .
27. S. Mukhopadhyay, I. Mondal, U. Pal and P. S. Devi, Phys. Chem. Chem. Phys., 2015, 17, 20407–20415 RSC .
28. X. Huang, M.-G. Willinger, H. Fan, Z.-l. Xie, L. Wang, A. Klein-Hoffmann, F. Girgsdies, C.-S. Lee and X.-M. Meng, Nanoscale, 2014, 6, 8787–8795 RSC .
29. Y. Pina-Perez, O. Aguilar-Martinez, P. Acevedo-Pena, C. E. Santolalla-Vargas, S. Oros-Ruiz, F. Galindo-Hernandez, R. Gomez and F. Tzompantzi, Appl. Catal., B, 2018, 230, 125–134 CrossRef CAS .
30. K. Yu, H.-B. Huang, J.-T. Wang, G.-F. Liu, Z. Zhong, Y.-F. Li, H.-L. Cao, J. Lue and R. Cao, J. Mater. Chem. A, 2021, 9, 7759–7766 RSC .
31. Y. Yang, Q. Meng, X. Jiang, S. Meng, X. Zheng, S. Zhang, X. Fu and S. Chen, ACS Appl. Energy Mater., 2020, 3, 7736–7745 CrossRef CAS .
32. X. Zhao, J. Feng, J. Liu, J. Lu, W. Shi, G. Yang, G. Wang, P. Feng and P. Cheng, Adv. Sci., 2018, 5, 1700590 CrossRef .
33. J. Yan, X. Fang, L. Zhang, Y. Bando, U. K. Gautam, B. Dierre, T. Sekiguchi and D. Golberg, Nano Lett., 2008, 8, 2794–2799 CrossRef CAS .
34. X. Wang and X. Yao, Carbon, 2014, 77, 667–674 CrossRef CAS .
35. X. Wang, G. Liu, G. Q. Lu and H.-M. Cheng, Int. J. Hydrogen Energy, 2010, 35, 8199–8205 CrossRef CAS .
36. Z. Zhao, Z. Wang, J. Zhang, C. Shao, K. Dai, K. Fan and C. Liang, Adv. Funct. Mater., 2023, 33, 2214470 CrossRef CAS .
37. D. Xu, B. Cheng, S. Cao and J. Yu, Appl. Catal., B, 2015, 164, 380–388 CrossRef CAS .
38. S. Cao and L. Piao, Angew. Chem., Int. Ed., 2020, 59, 18312–18320 CrossRef CAS PubMed .
39. Z. Yu, X. Yue, J. Fan and Q. Xiang, ACS Catal., 2022, 12, 6345–6358 CrossRef CAS .
40. H. Kisch, Angew. Chem., Int. Ed., 2010, 49, 9588–9589 CrossRef CAS PubMed .
41. M. Sayed, F. Xu, P. Kuang, J. Low, S. Wang, L. Zhang and J. Yu, Nat. Commun., 2021, 12, 4936 CrossRef CAS PubMed .
42. S. Wang, B. Zhu, M. Liu, L. Zhang, J. Yu and M. Zhou, Appl. Catal., B, 2019, 243, 19–26 CrossRef CAS .
43. M. Melchionna and P. Fornasiero, ACS Catal., 2020, 10, 5493–5501 CrossRef CAS .
44. X. Wang, L. Yin, G. Liu, L. Wang, R. Saito, G. Q. Lu and H.-M. Cheng, Energy Environ. Sci., 2011, 4, 3976–3979 RSC .
45. R. Ye, H. Fang, Y.-Z. Zheng, N. Li, Y. Wang and X. Tao, ACS Appl. Mater. Interfaces, 2016, 8, 13879–13889 CrossRef CAS PubMed .
46. B. Wang, S. Guo, X. Xin, Y. Zhang, Y. Wang, C. Li, Y. Song, D. Deng, X. Li, A. J. Sobrido and M.-M. Titirici, ACS Appl. Mater. Interfaces, 2020, 10, 2001575 CAS .
47. H.-B. Huang, K. Yu, J.-T. Wang, J.-R. Zhou, H.-F. Li, J. Lu and R. Cao, Inorg. Chem. Front., 2019, 6, 2035–2042 RSC .
48. J.-W. Shi, D. Ma, Y. Zou, Z. Fan, J. Shi, L. Cheng, X. Ji and C. Niu, J. Power Sources, 2018, 379, 249–260 CrossRef CAS .
49. G. Sun, B. Xiao, H. Zheng, J.-W. Shi, S. Mao, C. He, Z. Li and Y. Cheng, J. Mater. Chem. A, 2021, 9, 9735–9744 RSC .
50. D. Ma, J.-W. Shi, Y. Zou, Z. Fan, X. Ji, C. Niu and L. Wang, Nano Energy, 2017, 39, 183–191 CrossRef CAS .
51. X. Yan, M. Xia, H. Liu, B. Zhang, C. Chang, L. Wang and G. Yang, Nat. Commun., 2023, 14, 1741 CrossRef CAS PubMed .

---

Footnotes

† Electronic supplementary information (ESI) available: Details on chemicals, instruments, synthesis and characterizations of different products, photocatalytic measurement, photoelectrochemical measurement, photocatalytic stability test, apparent quantum efficiency test and computational contents. See DOI: https://doi.org/10.1039/d3gc03152g

‡ These authors contributed equally.

This journal is © The Royal Society of Chemistry 2024