Morphology engineering of type-II heterojunction nanoarrays for improved sonophotocatalytic capability

Highlights

• •ZnO/ZnS nanostructures with type-II heterojunction are successfully synthesized.
• •Heterojunction nanotube exhibits better sonophotocatalysis than nanorod.
• •Effect of morphology on piezoelectric is experimentally and simulatively studied.
• •Morphology engineering can affect sonophotocatalysis through turning piezoelectric.

Abstract

Sonophotocatalysis is one of the most significant outcomes of the exploration of the interaction between piezoelectric field and charge carriers, which exhibits potential applications in dye degradation, water splitting, and sterilization. Although several heterojunction catalysts have been applied to improve the sonophotocatalytic capability, the importance of the morphology on the sonophotocatalytic capability has not been emphasized. In this study, brush-like ZnO nanorod arrays are synthesized on a stainless-steel mesh and subsequently vulcanized into ZnO/ZnS core–shell nanorod arrays to investigate the sonophotocatalytic capability of the heterojunction. The sonophotocatalytic capability increases from 25.1% to 45.4% through vulcanization. Afterward, the ZnO/ZnS nanorods are etched to ZnO/ZnS nanotubes without affecting the crystallography and distribution of the ZnS nanoparticle shell, further improving the capability to 63.3%. The improvement can be ascribed to the coupling effect of the enhanced piezoelectric field and the reduced migration distance, which suppresses the recombination of photoexcited electron–hole pairs while transforming the morphology from nanorod to nanotube, as proven by the electron spin resonance test and numerical simulations. This study explores a novel approach of morphology engineering for enhancing the sonophotocatalytic capability of heterojunction nanoarrays.

1. Introduction

The interaction between the polarization field and charge carriers in ferroelectric, pyroelectric, and piezoelectric nanomaterials has received considerable critical attention across many areas, including photocatalysis, field-effect transistors, sensors, and self-powered systems [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. As one of the most significant outcomes, ultrasound sonication-promoted photocatalysis (named sonophotocatalysis) has been developed using a sonication-induced piezoelectric field by suppressing the recombination of photoexcited electron–hole pairs [11], [12], [13], [14]. For the emerging strategy of non-centrosymmetric sonophotocatalysts, the macroscopic spontaneous polarization field derived from the directional accumulation of polar units can facilitate the rapid separation of charge carriers and their migration from the bulk to the surface in opposite directions, inducing an increment in the catalytic capability than those of traditional photocatalysts [15], [16], [17], [18]. Nevertheless, the spontaneous polarization field can be shielded by the migration of charge carriers, resulting in the loss of the driving force for carrier separation. To reduce the consumption of mechanical energy and improve the utilization of photo energy, a strong piezoelectric polarization field is required to afford a high carrier concentration.

In recent years, heterojunction construction has been attracting research interest in the area of sonophotocatalysis since the heterojunction interface can significantly improve the separation efficiency of charge carriers [19], [20], [21]. To date, type-II and Z-scheme heterojunctions composed of two homomorphic semiconductors have attracted the most attention in heterojunction sonophotocatalysis research. Most of the previous studies on heterojunction sonophotocatalysis emphasized the importance of the electronic band structure while ignoring the effect of the morphology [22], [23], [24]. In addition, it has been reported that morphology significantly affects piezoelectricity and photocatalysis. Analytical and numerical studies by Meguid et al. showed that GaN nanotubes can generate a significantly higher piezopotential than nanowires [25]. Suib et al. studied the effect of morphology on the photocatalysis of ZnO nanostructures synthesized by different hydrothermal routes [26]. Fan et al. synthesized butterfly-wing-like TiO₂, which exhibited a double photocatalytic property unlike plate-like TiO₂ [27]. Thus far, the effect of morphology on the sonophotocatalytic capability of heterojunctions has been minimally investigated. Moreover, an effective and inexpensive fabrication route for transforming the morphology without affecting the crystalline characteristic of the heterojunctions remains to be developed.

In this study, ZnO/ZnS core–shell nanorod arrays have been designed as the base material for type-II heterojunction sonophotocatalysis through the facile vulcanization of ZnO nanorods. ZnO is a good sonophotocatalyst candidate owing to its strong piezoelectricity, controllable morphology, and high optical absorption efficiency [28], [29], [30], [31]. Furthermore, ZnO can easily form a type-II heterojunction with ZnS because of their similar band structures and tunable lattice mismatch [32], [33], [34]. Therefore, ZnO nanostructures provide an ideal platform to investigate the effect of morphology on the sonophotocatalytic capability of heterojunction nanostructures. To investigate the effect of morphology on heterojunction sonophotocatalysis, the ZnO nanorod in the ZnO/ZnS heterojunction is corroded into a nanotube architecture through alkali etching, while the crystalline nature and type-II construction remain unchanged. The photocatalytic and sonophotocatalytic capabilities of the synthesized samples are evaluated by degrading methylene blue (MB) solution as an indicator. Compared with the sonophotocatalytic capability of the ZnO/ZnS nanorod, that of the ZnO/ZnS nanotube increases by ∼1.5 fold. This is because the nanotube provides a stronger piezoelectric field than the nanorod under the same sonication condition to separate the photoexcited electron–hole pairs. Furthermore, detailed experiments and numerical simulations support this observation. This study offers a new perspective on morphology engineering for improving the sonophotocatalytic capability of heterojunction nanoarrays.

2. Materials and method

2.1. Synthesis of ZnO nanorod arrays on a stainless-steel mesh

A precursor solution was formed by mixing 0.5 g of Zn(NO₃)₂·2H₂O, 2.5 mL of concentrated NH₃·2H₂O (28%), and 47.5 mL of deionized water in a beaker. After immersing a piece of pre-cleaned stainless-steel mesh (network density of 30 lines cm⁻¹, area of 2 cm × 2 cm), the beaker was sealed and maintained at 80 °C for 12 h to grow ZnO nanorod arrays.

2.2. Synthesis of ZnO/ZnS core–shell nanorods and nanotubes

ZnO/ZnS core–shell nanorods were synthesized by immersing the as-prepared ZnO nanorod arrays sample into 0.2 M Na₂S aqueous solution under mild stirring for 4 h at 60 °C. Subsequently, the ZnO/ZnS core–shell nanorods were immersed in diluted ammonia solution (10%) for 20 min at 50 °C to synthesize ZnO/ZnS core–shell nanotubes. For simplicity, the ZnO nanorods, ZnO/ZnS core–shell nanorods, and ZnO/ZnS core–shell nanotubes were denoted as S1, S2, and S3, respectively. Before characterization and measurement, all the samples were washed using flowing water and dried at 60 °C for 24 h.

2.3. Characterization

The elemental components of the samples were analyzed by energy-dispersive X-ray spectroscopy (EDS) and X-ray photoelectron spectroscopy (XPS). The morphologies, microstructures, and crystalline phases were examined by scanning electron microscopy (SEM), transmission electron microscopy (TEM) equipped with selected area electron diffraction (SAED), and X-ray powder diffraction (XRD). The optical properties were investigated by UV–Vis diffuse reflection spectroscopy (DRS) and photoluminescence (PL).

2.4. Sonophotocatalytic capability evaluation

The sonocatalytic, photocatalytic, and sonophotocatalytic capabilities were tested in a homemade sonophotocatalytic reaction system. The UV light and ultrasound (US) sources were a 500 W Hg-lamp and ultrasonic cleaner with tunable frequency and power, respectively. Before the catalysis, a piece of the sample was immersed in MB solution (50 mL, 10 mg L^–1) and mildly stirred for 2 h in dark for full adsorption. After catalyzing at 10 min intervals, 0.5 mL of the MB solution was collected and measured using a spectrophotometer to determine the concentration represented by the peak intensity at 664 nm. To measure the acceleration effect, the electron spin resonance (ESR) signals of the hydroxyl radical ($\bullet \mathrm{O}\mathrm{H}$) and superoxide radical (${\bullet \mathrm{O}}_2^{-}$) trapped by dimethylpyridine nitrogen oxide (DMPO) were tested.

2.5. Numerical simulations

The finite element method (FEM) was utilized to simulate the deformation and strain-induced piezoelectric field of the nanorods and nanotubes. To simplify the calculation, the hexagonal ZnO nanorod was represented by a cylinder, and the ZnS shell was ignored. The size of the nanorod was determined to be ϕ 200 nm × 3 μm. The length and outer diameter of the nanotube were the same as those of the nanorod, and the inner diameter was 160 nm.

3. Results and discussion

3.1. Design of ZnO/ZnS core–shell nanostructures

The design and synthesis procedures of the sonophotocatalysts are described in Scheme 1. In the present design, a piece of stainless-steel mesh was chosen as the substrate because the doping elements present in it can provide many nucleation sites for growing ZnO and the network architecture is beneficial for absorbing multidirectional incident light. Thus, brush-like ZnO nanorod arrays can be grown along the (0001) orientation on stainless-steel networks through a hydrothermal route at 80 °C, as expressed by Eq. (1). Through the in situ displacement of the O atom by the S atom on the surface, ZnO nanorods are sheathed by a ZnS nanoparticle layer to form ZnO/ZnS core–shell nanorods, as described by Eq. (2) [35]. To transform the morphology, ZnO/ZnS core–shell nanorods are immersed in a highly alkaline solution since ZnO can be dissolved in the presence of excess ${\mathrm{O}\mathrm{H}}^{-}$ ions at 50 °C, as expressed in Eq. (3) [36]. The exposed (0 0 1) plane is a polar face, and it is less stable than the nonpolar faces, leading to preferential corrosion. Therefore, the ZnO/ZnS nanorod can be etched into a nanotube. To prevent the corrosion of ZnS while etching ZnO, the reaction temperature, reaction time, and pH should be well optimized.

$$Z{n}^{2+}+2O{H}^{-}\stackrel{80{}^{°}C}{\to }Zn{O}_{\mathit{solid}}+H_{2}O$$ (1)

$${\mathit{ZnO}}_{\mathit{solid}}+S^{2-}+H_{2}O\to {\mathit{ZnS}}_{\mathit{solid}}+{2OH}^{-}$$ (2)

$$\mathit{Zn}{O}_{\mathit{solid}}+2O{H}^{-}\stackrel{50{}^{°}C}{\to }Zn{O}_2^{2-}+H_{2}O$$ (3)

Scheme 1.

Fabrication process of S1, S2, and S3.

3.2. Characterization of ZnO and ZnO/ZnS core/shell nanostructures

Figure 1a shows the EDS spectra of the three samples. Evidently, the Zn and O elements exist in all the samples, while the S element only exists in S2 and S3. XRD patterns were recorded to preliminarily estimate the crystal category of the three samples, as shown in Fig. 1b. For S1, all the predominant peaks can be well assigned to the pure phase of wurtzite ZnO (JCPDS No. 36-1451), and the strong (0 0 2) peak indicates the c-axis growth direction. For S2 and S3, the new diffraction peaks corresponding to ZnS (JCPDS No. 05-0566) suggest the coexistence of ZnO and ZnS. By rough calculation without refinement, the component proportion of ZnS:ZnO in S3 (11:1) is determined to be considerably higher than that in S2 (1:2), implying that most ZnO crystals have been subducted during the etching procedure.

Fig. 1.

Elemental components and crystal phases of the as-prepared samples. (a) EDS spectra. (b) XRD patterns.

Figure 2 shows the morphologies and elemental mapping profiles of the three samples. Fig. 2a shows the SEM image of the as-grown ZnO nanostructures on the stainless-steel mesh, exhibiting a brush-like array consisting of vertically aligned nanorods. Fig. 2b shows a magnified SEM image of several ZnO nanorods and indicates that the average diameter of the ZnO nanorod is ∼200 nm. After forming the ZnO/ZnS core–shell nanorods, the surface becomes rough (Fig. 2c). As the core–shell nanorod is subsequently etched, the ZnO/ZnS core–shell nanotube is formed with a wall thickness of ∼20 nm (Fig. 2d). The elemental mapping profiles of the ZnO/ZnS nanorod and nanotube are shown in Fig. 2e and f, respectively. Fig. 2e shows that ZnS is uniformly distributed on the surface of the ZnO nanorod, and Fig. 2f shows the hollow structure of the ZnO/ZnS nanotube. These results suggest that the ZnO/ZnS nanorods and nanotubes were successfully fabricated.

Fig. 2.

Characterization of the morphology and elemental distribution. (a) Low-magnification SEM image of S1. (b–d) High-magnification SEM images of (b) S1, (c) S2, and (d) S3. (e, f) Elemental mapping profiles of Zn, S, and O in (e) S2 and (f) S3.

The sizes and microstructures were further studied by TEM images, as shown in Fig. 3. Fig. 3a shows a TEM image of one ZnO nanorod, indicating that the length of the ZnO nanorod is ∼3 μm. According to the high-resolution TEM image (Fig. 3b), the interlayer spacing of 0.36 nm is in agreement with the (0 0 2) lattice plane, indicating the good crystallinity of the hexagonal wurtzite ZnO nanorod along the (0001) growth direction. For the SAED pattern (Fig. 3c), the dot spacings of 3.89, 6.32, and 7.26 nm⁻¹ in the periodic dot-matrix can be assigned to the (0 0 2), (1 1 0), and (1 1 2) planes of ZnO, respectively, also indicating the monocrystalline ZnO. Fig. 3d shows a TEM image of the ZnO/ZnS nanorod, indicating the coaxial core–shell structure. The high-resolution TEM of the ZnO/ZnS core–shell nanorod (Fig. 3e) shows that the interlayer spacing of 0.31 nm corresponds to the (1 1 1) plane of ZnS with an average diameter of ∼ 15 nm, and the interface between ZnO and ZnS sharply transforms without an intermediate layer. As shown in the SAED pattern (Fig. 3f), the dot matrix is raised from the single-crystalline ZnO core, and the concentric rings with radii of 3.21, 5.14, and 6.18 nm⁻¹, corresponding to the (1 1 1), (2 2 0), and (3 1 1) planes of the ZnS shell, respectively. The TEM image of S3 shows the nanotube structure (Fig. 3g). Fig. 3h displays a similar image to that of Fig. 3e, indicating the constant crystalline structures. As shown in Fig. 3i, the dot matrix disappears because of the loss of the ZnO crystal.

Fig. 3.

Microstructures of the samples. Low-resolution TEM images, high-resolution TEM images, and SAED patterns of S1 (a–c), S2 (d–f), and S3 (g–i).

XPS technique was employed to further investigate the chemical state of the present elements on the surface of the three samples, and the results are shown in Fig. 4. Fig. 4a shows the survey spectra containing O1s, Zn 2p, and S 2P, which are further analyzed by the fine spectra shown in Fig. 4b–d, respectively. For Zn 2p (Fig. 4b), the peaks with binding energies at approximately 1044.98 and 1021.88 eV are attributed to Zn 2p_1/2 and Zn 2p_3/2, respectively. Compared with those of S1, the Zn 2p peaks of S2 and S3 slightly shift toward low-energy regions, triggered by the formation of the Zn–S bond. As shown in Fig. 4c, the S 2p spectra of S2 and S3 are almost the same, where the peaks located at 162.78 and 161.68 eV are assigned to S 2p_1/2 and S 2p_3/2, respectively. Fig. 4d shows the O 1 s spectra of the three samples, where the peaks located at 531.58 and 530.08 eV are assigned to oxygen vacancy (O_V) and lattice oxygen (O_L), respectively. As observed, the content and proportion of the lattice oxygen are efficiently diminished by sulfuration. All the above results confirm the successful synthesis of ZnO/ZnS core–shell heterojunctions and transformation of their morphology from nanorod to nanotube without destroying their crystalline nature.

Fig. 4.

XPS spectra of the three samples. (a) Survey. (b) Zn 2p. (c) S 2p. (d) O 1 s.

3.3. Optical property

DRS and PL spectroscopy were employed to investigate the optical properties of the samples, and the results are shown in Fig. 5. Fig. 5a displays the DRS spectra to illustrate the optical absorption property. As observed, the absorption edges of the three samples are all in the UV range with similar absorption capabilities. The absorption peaks of S2 and S3 tend to locate around low wavelengths and are affected by the addition of ZnS since the bandgap of ZnS (∼3.6 eV) is larger than that of ZnO (∼3.2 eV). The luminescence properties of the samples were examined by PL spectra, and the results are shown in Fig. 5b. The minor peaks at approximately 380 nm and major peaks at approximately 500 nm are roughly attributed to intrinsic luminescence and defect luminescence, respectively [37]. The peak intensities of the defect luminescence are significantly higher than those of intrinsic luminescence, suggesting that the recombination of the electron–hole pairs occurring at the defect sites dominates the luminescence property of the as-synthesized samples. Furthermore, the defect luminescence peak intensities of S2 and S3 are considerably lower than those of S1, implying the suppressed recombination of the electron–hole pairs induced by the formation of the ZnO/ZnS type-II heterojunction, improving the photocatalytic quantum efficiency.

Fig. 5.

Optical properties of the three samples. (a) DRS spectra. (b) PL spectra.

3.4. Sonophotocatalysis evaluation

The sonophotocatalytic capabilities of the three samples were examined by degrading the MB solution under 120 W ultrasound sonication and 500 W UV irradiation, and the results are shown in Fig. 6. The original time-dependent photoabsorption spectra of the MB solution during the sonophotocatalytic procedure catalyzed by S1, S2, and S3 are shown in Fig. 6a–c, respectively. It is evident that the sonophotocatalysis is improved by sulfuration. To better characterize the sonophotocatalytic capability, the degradation rate curves are plotted, as shown in Fig. 6d. The degradation rate of S2 (45.4%) is significantly higher than that of S1 (25.1%), indicating that the interfacial polarization field induced by the ZnO/ZnS heterojunction enhances the sonophotocatalytic capability. As the nanorods are transformed into nanotubes, S3 exhibits a higher degradation rate (63.3%) than S2, suggesting that the sonophotocatalytic capability is significantly increased by only transforming the catalyst morphology. Moreover, S3 is comparable to the majority of recently reported nanomaterials for sonophotocatalytic degradation of MB solution, as shown in Table 1 [21], [38], [39], [40].

Fig. 6.

Sonophotocatalytic capabilities of the three samples. (a–c) Absorption curves of MB sonophotocatalyzed by (a) S1, (b) S2, and (c) S3. (d) Sonophotocatalytic degradation rates of the three samples. (e, f) ESR signals of (e) $\bullet \mathrm{O}\mathrm{H}$ and (f) ${\bullet \mathrm{O}}_2^{-}$ generated by the three samples under the same condition.

Table 1.

Capabilities of various nanomaterials for sonophotocatalytic degradation of MB solution.

Material	Dosage	MB volume, concentration	Time (min)	Conditions	Efficiency (%)	Refs.
CuS/ZnO Nanowires	100 mg	50 mL, 5 mg L–1	20	Vis: 500 W, US: 200 W	∼100	21
BaTiO3/TiO2 nanowire	50 mg	100 mL, 30 mg L–1	105	UV: 300 W, US: 200 W	99.8	38
NaNbO3 nanorods	100 mg	250 mL, 10−5 M	180	UV: 103 W m−2, US: 100 W	∼83	39
Fe3+-doped ZnO	25 mg	50 mL, 0.063 mM	60	UV: 160 W, US: 200 W	74.13	40
ZnO nanorod	1 piece	50 mL, 10 mg L−1	50	UV: 500 W US: 160 W	25.1	This work
ZnO/ZnS nanorod	1 piece	50 mL, 10 mg L−1	50	UV: 500 W US: 160 W	45.4	This work
ZnO/ZnS nanotube	1 piece	50 mL, 10 mg L−1	50	UV: 500 W US: 160 W	63.3	This work

Recently, the ESR technique has proven powerful for the quantitative study of catalytic mechanisms and capabilities by measuring the amount of produced reactive oxygen species (ROSs) [41]. The ${\bullet \mathrm{O}}_2^{-}$ radical generated from the free electron on the conduction band (E_CB) and the $\bullet \mathrm{O}\mathrm{H}$ radical generated from the hole on the valence band (E_VB) are the two most accredited ROSs, and they work as primary reaction sources for organic molecule degradation and disease treatment [42], [43], [44]. Further comparison has been carried out to investigate the capability of the three samples by measuring the ESR signals of the $\bullet \mathrm{O}\mathrm{H}$ and ${\bullet \mathrm{O}}_2^{-}$ radicals generated by sonophotocatalysis, as displayed in Fig. 6e and f, respectively. It is concluded that both the characteristic signal intensities of the $\bullet \mathrm{O}\mathrm{H}$ and ${\bullet \mathrm{O}}_2^{-}$ radicals for S3 significantly increase in contrast to those for S1 and S2, supporting the improved sonophotocatalytic capability of S3.

To investigate the influencing factors of the boosted sonophotocatalysis, the degradation rates catalyzed by S2 and S3 were examined under different conditions, and the results are shown in Fig. 7a and b, respectively. As shown in Fig. 7a, the degradation rates catalyzed by S2 under 120 W sonication and 500 W UV irradiation are ∼ 3-fold and ∼ 1.5-fold higher than those by sonocatalysis (17.5%) and photocatalysis (35.7%), respectively. Furthermore, the degradation rate increases as the sonication power increases, suggesting that the degradation rate significantly depends on the ultrasound-induced piezoelectric field. Fig. 7b shows that S3 exhibits a similar phenomenon and higher degradation rates compared with S2 under the same conditions. Fig. 7c displays the degradation rates catalyzed by S3 under 500 W UV irradiation and ultrasound sonication with different modes. The lowest capability is exhibited in the degassing mode. The dissolved oxygen molecule is eliminated in this mode such that few ${\bullet \mathrm{O}}_2^{-}$ radicals are generated. Interestingly, the degradation rate under 24 kHz ultrasound is higher than that under 40 kHz. According to our previous work, it can be ratiocinated that the 24 kHz ultrasound may be suitable for optimizing the pressure induced by the burst of vacuum bubble to bend the ZnO/ZnS nanotubes [45]. Stability plays an important role in sonophotocatalytic performance. The stabilities of the three samples in five degradation cycles are shown in Fig. 7d–f, respectively. The degradation rate of S1 rapidly drops from 25.1% to 18.8%, while S2 and S3 maintain 87.3% and 85.2% of the initial degradation rate, respectively. The good reusabilities of S2 and S3 can be ascribed to the heterojunction-enhanced photo-corrosion resistance of ZnO [46], [47], [48].

Fig. 7.

(a, b) Degradation rates of sonocatalysis, photocatalysis, and sonophotocatalysis with different ultrasound powers, catalyzed by (a) S2 and (b) S3. (c) Degradation rates of S3 with different ultrasound modes. (d–f) Recycling degradation rates of (d) S1, (e) S2, and (f) S3 for five cycles.

3.5. Numerical simulation

To explore the difference in the strain-induced piezoelectric field between the nanorod and nanotube, the deformation degrees and piezoelectric potential distributions of the nanorod and nanotube were numerically calculated by FEM, as shown in Fig. 8. To simplify the calculation, the ZnS shell was ignored as it does not benefit qualitative understanding. A static pressure of 80 MPa, inspired from the ultrasound sonication-induced pressure, was applied on the tip region along the horizontal direction to bend the nanostructure (Fig. 8a and b). Under the uniform applied force, the maximum displacement of the nanotube (275 nm) is distinctly higher than that of the nanorod (168 nm), as shown in Fig. 8c. Therefore, the generated piezoelectric field of the nanotube (0.98 V) is also higher than that of the nanorod (0.21 V), leading to a significant improvement in the separation efficiency of the electron–hole pairs (Fig. 8d).

Fig. 8.

Numerical analysis of the ultrasound-induced strain and the piezoelectric field of the ZnO nanorod and nanotube by FEM. (a, b) The 200-nN-force-induced deformations and piezoelectric potential of the (a) ZnO nanorod and (b) ZnO nanotube. (c) Strain-induced displacements along the axial direction on the stretched side of the ZnO nanostructures. (d) Piezoelectric potentials along the axial direction on the stretched side of the ZnO nanostructures.

3.6. Proposed reaction mechanism

The above results demonstrate that the sonophotocatalytic capability can be significantly improved by morphology engineering. The mechanism of the improved sonophotocatalytic capability of the ZnO/ZnS nanotube arrays is proposed and explained in detail, as shown in Fig. 9. In the ZnO/ZnS heterojunction, the E_CB and E_VB of ZnS are more negative and less positive than those of ZnO, respectively (Fig. 9a) [49]. Upon exposure to UV irradiation, the photoexcited electrons are accumulated at the E_CB of ZnO to generate ${\bullet \mathrm{O}}_2^{-}$ radicals and the photoexcited holes are accumulated at the E_VB of ZnS to generate $\bullet \mathrm{O}\mathrm{H}$ radicals, forming a type-II heterojunction. The spatial isolation provided by the heterojunction is suitable for suppressing the recombination of the electron–hole pairs and improving the inner quantum efficiency of catalysis. However, the interface electric field induced by the difference in the work functions of the ZnO/ZnS heterojunction plays a negative role by hindering the migration of the charge carriers across the interface. Thus, the improvement induced by heterojunction is limited. When the ZnO/ZnS nanorods are deformed by ultrasound sonication without UV irradiation, many polarization charges are generated at the surface by the piezoelectric effect, resulting in a piezoelectric field depending on the bending degree (Fig. 9b). Since the piezoelectric field is sufficiently strong, the interface electric field is significantly screened. When the heterojunction nanorods are simultaneously subjected to UV irradiation and ultrasound sonication, the piezoelectric field separates the photoexcited charge carriers inside the nanostructures and accelerates the diffusion of the photoexcited charge carriers through the interface (Fig. 9c). Therefore, the sonophotocatalytic capability is higher than the photocatalytic capability, which is attributed to the synergistic effect of photocatalysis, heterojunction, and the piezoelectric field. In contrast to nanorods, nanotubes have two functions to further improve their catalytic capability (Fig. 9d). On the one hand, the nanotube structure can reduce the migration distance from the interior to the surface, leading to more charge carriers participating in the ROS generation. On the other hand, the nanotube can generate a higher piezoelectric field to screen the interface electric field than that generated by the nanorod under the same force, thereby contributing to the inhibition of the recombination of charge carriers.

Fig. 9.

Band structure and charge transfer of the proposed mechanisms. (a–c) Diagrams of (a) photocatalysis, (b) piezoelectricity, and (c) sonophotocatalysis of the ZnO/ZnS nanorod heterojunction. (d) Diagram of the sonophotocatalysis of the ZnO/ZnS nanotube heterojunction.

4. Conclusion

In summary, the importance of morphology on sonophotocatalysis was demonstrated by comparing the catalytic performance of vulcanized ZnO nanorods and nanotubes. The sonophotocatalytic capability of the ZnO/ZnS nanorods is higher than that of ZnO nanorods, which is attributed to inhibited recombination of photoexcited charge carriers driven by the synergistic effect of photocatalysis, heterojunction, and piezoelectric field. The spatial isolation of electron-hole pairs can be achieved by heterojunction-induced diffusion through the interface, and it can be further enhanced by the screening of the interface electric field by the piezoelectric field. Under a periodic mechanical strain provided by ultrasound sonication, the nanotube construction can generate a superior piezoelectric field than that of the nanorod, as confirmed by FEM simulations. This work provides promising insights for the development of sonophotocatalysts.

CRediT authorship contribution statement

Lixia Guo: Investigation, Writing – original draft. Yaodong Chen: Investigation, Writing – original draft. Zeqian Ren: Methodology, Software. Xiu Li: Methodology, Software. Qiwei Zhang: Methodology, Software. Jizhou Wu: Investigation, Resources, Visualization. Yuqing Li: Investigation, Resources, Visualization. Wenliang Liu: Investigation, Resources, Visualization. Peng Li: Investigation, Resources, Visualization. Yongming Fu: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing. Jie Ma: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.