Heterojunction configuration-specific photocatalytic degradation of methyl orange and methylene blue dyes using ZnO-based nanocomposites

Graphical abstract

Highlights

• •ZnO-based composite nanostructures exhibit a specific photocatalytic degradation for MO and MB dyes.
• •The predominant active species for MO and MB degradation are photogenerated holes and superoxide radicals.
• •Such a specific photocatalytic degradation was demonstrated to be related to the heterojunction configuration.
• •Heterojunction band alignments govern the generation of active species for photocatalytic degradation of organic pollutants.

Abstract

Introduction

Heterostructured photocatalysts have shown an enormous potential in photocatalytic degradation of organic pollutants in wastewater. However, the efficacy of such heterojunction on the photocatalytic degradation behaviors has not yet been fully revealed.

Objectives

This work aims to demonstrate a specific photocatalytic degradation behavior of ZnO-based heterostructured nanocomposites toward methyl orange (MO) and methylene blue (MB) dyes based on a systematically comparative investigation for their physical and chemical properties.

Methods

A series of low-cost and efficient ZnO-based heterostructured nanocomposite photocatalysts including ZnO/CuO, ZnO/TiO₂ and ZnO/SnO₂ with 3 and 10 mol% of CuO/TiO₂/SnO₂ were synthesized by a simple strategy to combine the modified polymer-network gel and traditional sol–gel methods. The physical and chemical properties were analyzed using X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), transmission electron microscope (TEM), X-ray photoelectron spectra (XPS), ultraviolet–visible (UV–Vis) absorption spectra, photoluminescence (PL), surface photovoltage (SPV), electrochemical impedance spectroscopy (EIS) and zeta potential.

Results

Owing to the fast interfacial charge transfer at the heterojunction, all the three ZnO-based nanocomposite catalysts exhibited higher efficient separation of photogenerated electrons and holes, delivering an enhanced photocatalytic activity for the degradation of organic dyes compared with pure ZnO. Three photocatalysts of ZnO/3 %-CuO, ZnO/3 %-TiO₂ and ZnO/10 %-SnO₂ (marking as ZC3, ZT3 and ZS10, respectively) were capable of achieving the complete degradation of 4 mg/L concentration of MB dye within 50 min, and the first two could degrade MO within 80 min. However, the degradation rate of MO by ZS10 became significantly slower. For MO and MB degradation, the active species of photogenerated holes (${\text{h}}_{h\nu }^{+}$) and superoxide radicals ($·{\text{O}}_2^{-}$) play the predominant roles, respectively, followed by hydroxyl radicals ($·\text{OH}$). The differences in heterojunction configuration and dominant active species result in a specific photocatalytic degradation behavior of ZnO-based composite nanostructures.

Conclusion

The generation of the active species are influenced by the heterojunction configurations, of which the essence is that the different band alignments can results in the differences of interfacial charge transfer behaviors, and thus selective generation of the active species such as ${\text{h}}_{h\nu }^{+}$, $·{\text{O}}_2^{-}$ and $·\text{OH}$. Importantly, this work offers a fundamental understanding for specific photocatalytic degradation of the different heterojunction nanostructures towards the different organic dyes.

Introduction

Since the early 21st century, energy crisis and environmental pollution have always been global concerns for mankind [[1], [2], [3]]. The latter, in particular, brings pollution loads (of air, water and soil) that have exceeded the ecological carrying capacity, which is gradually destroying our ecosystem. The burden of disease and death attributable to this increasingly aggravates, igniting a burning desire for environmental protection and purification [[4], [5], [6]].

Water pollution is an urgent and knotty problem to be solved in our future, which is vital for the survival of human beings, animals and plants [[7], [8], [9]]. Whereas the issue is escalated with the rampant appearance of emerging contaminants (detergents, pesticides, pharmaceuticals, drugs, additives, pathogens, etc.) in wastewater discharges in the context of rising industrialization, urbanization, agriculture production and medical care [10]. This can be no longer be handled by the traditional wastewater treatment approaches (precipitation, coagulation, adsorption, ultrafiltration, ion exchange, etc.) [[11], [12], [13], [14]] with features of low-efficiency, cumbersome operation, incomplete removal and secondary pollution.

Heterogeneous photocatalysis, as an advanced oxidation technology (AOT), offers a green and promising strategy for the simple, high-efficient and complete removal of the emerging contaminants from wastewater by coupling solar energy with semiconducting photocatalysts [[15], [16], [17], [18]]. Nevertheless, such AOT is largely limited by the low light response capacity, high electron-hole recombination rate and poor stability of photocatalysts. To overcome the above obstacles, some effective strategies, including element doping [[19], [20], [21]], plasmonic metal decoration [14,[22], [23], [24]] and heterostructure construction [[25], [26], [27], [28]], have been developed to optimize the performances of photocatalysts for the photodegradation of organic pollutants. Among them, consciously constructing heterostructures by the various materials is a fascinating strategy to achieve highly efficient photocatalytic degradation systems.

For the heterostructured photocatalysts, ZnO-based composite nanostructures have been widely studied because of the suitable band-gap energy, versatile functions, easy synthesis, controlled morphology and environmental friendliness of the material [[29], [30], [31]]. These published works are primarily focused on the optimization of degradation efficiency of organic pollutants, and very few works pay attention to the difference of degradation behaviors of different organic pollutants. In fact, there are many reasons for the differentiated photocatalytic degradation behaviors of ZnO-based nanocomposites with different heterojunction configuration, which include morphology, composition, crystal structure, and so on. However, a very important influencing factor is overlooked and rarely researched, which is heterojunction configuration. Recently, some low-cost ZnO-based heterostructured nanophotocatalysts (such as ZnO/CuO [[32], [33], [34], [35]], ZnO/TiO₂ [[36], [37], [38], [39]] and ZnO/SnO₂ [40,41]) have shown the potential for large-scale applications. Their superior photocatalytic performances for the degradation of organic pollutants are attributed to the enhanced visible light absorption of the components with a narrow band-gap, efficient separation of photogenerated electron-hole pairs caused by the fast interfacial charge transfer and suppression of photocorrosion of ZnO. The differentiated heterojunction configurations can result in the specific photocatalytic degradation toward the different organic pollutants. However, the fundamental principle behind this has yet to be understood, of which the interpretation remains a great challenge.

Herein, pure ZnO and ZnO-based heterostructured nanocomposites (ZnO/CuO, ZnO/TiO₂ and ZnO/SnO₂) were synthesized, and their sunlight-driven photocatalytic degradation characteristic of MO and MB dyes in wastewater was investigated comparatively. The characterization for phase and composition, morphology and microstructure, surface chemical state, optical characteristics, charge separation and transfer behaviors were carried out to correlate between heterojunction configuration of ZnO-based heterostructured nanocomposites and photocatalytic degradation characteristic. In addition, the corresponding photocatalytic degradation mechanism was comprehensively discussed.

Experimental strategies

Reagents and materials

All chemical reagents were used as received without further purification, and deionized water with a resistivity of 18.25 MΩ·cm was made by a water purification system for all the experiments.

The reagents including zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O, purity: 99.0 %), L-(+)-tartaric acid (C₄H₆O₆, purity: 99.5 %), glucose anhydrous (C₆H₁₂O₆, purity: 99.0 %), acrylamide (C₃H₅NO, purity: 99.0 %), N,N'-methylene diacrylamide (C₇H₁₀N₂O₂, purity: 99 %), diethanolamine (C₄H₁₁NO₂, purity: 99.0 %), ethanol (C₂H₆O, purity: 99.7 %) and glacial acetic acid (CH₃COOH, purity: 99.5 %) as well as the materials including copper oxide (CuO, purity: 99.0 %), titanium oxide (TiO₂, purity: 99.0 %) and tin oxide (SnO₂, purity: 99.0 %) nanopowders are of analytic grade and purchased from Chengdu Chron Chemicals Co., Ltd. (Chengdu, China).

The reagents including MO and MB are of indicator grade, and purchased from Tianjin Jinbei Fine Chemical Co., Ltd. (Tianjin, China) and Chengdu Chron Chemicals Co., Ltd. (Chengdu, China), respectively.

The reagents including ethylenediamine tetraacetic acid disodium salt (EDTA-2Na, purity: 99.0 %), Isopropyl alcohol (IPA, purity: 99.7 %) and p-benzoquinone (BQ, purity: 97.0 %) are of analytic grade, and respectively purchased from Tianjin Fuchen Chemical Reagent Co., Ltd. (Tianjin, China), Tianjin Zhiyuan Chemical Reagent Co., Ltd. (Tianjin, China) and Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China).

Catalyst synthesis

Synthesis of ZnO nanocatalysts

Pure ZnO nanocatalysts were synthesized through a modified polymer-network gel method [42]. Typically, a certain amount of Zn(NO₃)₂·6H₂O (∼0.02 mol) and C₄H₆O₆ (∼0.03 mol) were successively dissolved in deionized water (∼75 mL) under magnetic stirring in a water bath at room temperature. After fully dissolved and chelated, ∼12 g of C₆H₁₂O₆, ∼10.77 g of C₃H₅NO and ∼4.97 g of C₇H₁₀N₂O₂ were added in turn to form a homogeneous and precursor solution, which was followed by transformed into a white solid gel as the temperature slowly heats up to 90 °C. The above solid gel was mashed and dried at 120 °C for 24 h to form a brown xerogel. Finally, the xerogel was calcined at 650 °C for 200 min, and then ground to obtain the light grey ZnO nanocatalysts, labelled as ZnO.

Synthesis of ZnO/CuO, ZnO/TiO₂, ZnO/SnO₂ nanocomposite catalysts

ZnO-based nanocomposite catalysts were synthesized by a simple sol–gel route. Firstly, the as-synthesized powdery ZnO nanocatalysts, commercially available CuO/TiO₂/SnO₂ nanopowders and 2.5 ml of C₄H₁₁NO₂ were successively dispersed to 60 ml of C₂H₆O by magnetic stirring for 2 h in a water bath at 60 °C. After standing for 48 h away from light, the resultant composite gels were dried at 90 °C, calcined at 450 °C for 3 h and ground to obtain ZnO/CuO, ZnO/TiO₂, ZnO/SnO₂ nanocomposite catalysts, respectively. In the nanocomposite catalysts, the molar ratios of CuO/TiO₂/SnO₂ were 3 mol% and 10 mol%, respectively. The corresponding samples were labeled as ZC3, ZC10, ZT3, ZT10, ZS3 and ZS10.

Material characterization

The phase composition and crystallinity of as-prepared catalysts were characterized by a D8 advance powder X-ray diffractometer (Bruker, Germany) using Cu Kα1 radiation (λ = 0.15418 nm) in the angular range of 2θ from 20° to 80°. Their morphologies and microstructures were examined on an Apreo 2C field emission scanning electron microscopy (FESEM, Thermo Fisher Scientific, USA) with an accelerating voltage of 10 kV and a Talos F200S G2 transmission electron microscope (TEM, FEI, USA) with an accelerating voltage of 200 kV. The X-ray photoelectron spectra (XPS) were recorded by an Escalab Xi+ photoelectron spectrometer (Thermo Fisher Scientific, USA) employing Al Kα excitation source (hυ = 1486.6 eV). All the binding energies were calibrated with respect to the C 1s peak at 284.80 eV, and an XPS PEAK 41 program was utilized to conduct the XPS spectra decomposition with Gaussian functions after subtraction of a Tougaard-type background. A Lambda 750 ultraviolet/visible/near-infrared spectrophotometer (PerkinElmer, USA) equipped with an integrating sphere was applied to obtain the ultraviolet–visible (UV–Vis) diffuse reflectance spectra. The photoluminescence (PL), surface photovoltage (SPV) and electrochemical impedance spectroscopy (EIS) spectra of the catalysts were measured by a LS55 fluorescence spectrophotometer (PerkinElmer, USA), PL-SPV/IPCE 1000 steady-state surface photovoltage spectrometer (PerfectLight, China) and CHI1760 electrochemical analysis workstation (Shanghai Chenhua Instrument Co., Ltd., China), respectively. The zeta potential of the catalysts were obtained by measuring their surface charges at various pH values via a Litesizer 500 Zeta potential and particle size analyzer (Anton paar, Austria).

Photocatalytic degradation of MO and MB

The photocatalytic activities of ZnO-based nanocomposite catalysts were evaluated by the degradation of MO and MB under simulated sunlight irradiation at room temperature (25 °C). Typically, 96 ml of deionized water and 4 ml of pre-prepared MO/MB dye aqueous solutions (100 mg/L) were mixed into the aqueous solutions with concentration of 4 mg/L (pH: ∼6.5). After that, 0.05 g of catalysts were added into the above dye aqueous solutions, accompanied by ultrasonic oscillation for 5 min and standing in the dark for 30 min to achieve adsorption–desorption equilibrium. During photocatalytic degradation, a 300 W high pressure xenon lamp (PLS-SXE300C, PerfectLight, China, light wavelength range of 300–2500 nm) was applied to simulate sunlight irradiation, which was installed on the top of the suspensions at 25 cm distance. The intensity of the light vertically irradiating to the upper layer of the suspensions was about 180 mW/cm². At regular intervals, ∼5 ml of suspension was dislodged and centrifuged at 6000 r·min⁻¹ for 5 min to remove the catalysts. The residual concentration of the dyes in the resultant supernatant was analyzed using a V-1100D visible spectrophotometer (Shanghai Mapada Instruments Co., Ltd., China) and Lambda 750 ultraviolet/visible/near-infrared spectrophotometer (PerkinElmer, USA). The degradation ratio and apparent rate constant (k) were determinated by the following formulas:

$$\text{Degradation}\text{ratio}\left(\%\right)=\frac{C_0-C_{\text{t}}}{C_0}\times 100\%=\frac{A_0-A_{\text{t}}}{A_0}\times 100\%$$ (1)

$$k=\mathrm{\Delta }\ln \frac{C_0}{C_{\text{t}}}/\mathrm{\Delta }t=\mathrm{\Delta }\ln \frac{A_0}{A_{\text{t}}}/\mathrm{\Delta }t$$ (2)

where A₀ and A_t are the absorbances of dye aqueous solutions without and with exposure to light irradiation measured via the visible spectrophotometer at specific wavelengths of light (463 nm and 664 nm for MO and MB), C₀ and C_t represent the residual concentration of the dyes in the aqueous solutions.

In addition, the pH value of MB solution was adjusted by adding a certain amount of CH₃COOH and ammonium hydroxide (NH₃·H₂O, purity: 25.0 %) to analyze the influence of dye solution pH on its photocatalytic degradation. EDTA-2Na, IPA, and BQ respectively as scavengers for photogenerated holes (${\text{h}}_{h\nu }^{+}$), hydroxyl radicals ($·\text{OH}$) and superoxide radicals ($·{\text{O}}_2^{-}$) were added into the dye solution to analyze the photocatalytic degradation mechanism.

Results and discussion

Phase and composition analysis

Fig. 1a depicts the XRD pattern of as-synthesized ZnO nanocatalysts, in which the characteristic diffraction peaks located at 31.84°, 34.48°, 36.34°, 47.61°, 56.65°, 62.93°, 66.43°, 68.02°, 69.15°, 72.60°, and 77.07° can be assigned to the (1 0 0), (0 0 2), (1 0 1), (1 0 2), (1 1 0), (1 0 3), (2 0 0), (1 1 2), (2 0 1), (0 0 4) and (2 0 2) planes of hexagonal wurtzite phase of ZnO (JCPDS file No. 36-1451), respectively. The strong and sharp XRD peaks suggest its good crystallinity. The XRD peaks of as-synthesized ZnO-based heterostructured nanocomposite catalysts are compared with those of pure ZnO, as shown in Fig. 1b and Fig. S1. As incorporating the other nanostructured metal oxides, the XRD peaks corresponding to the (1 1 −1), (1 1 1) planes of monoclinic CuO (JCPDS file No. 48-1548, marked with ‘♥’) for ZC3 and ZC10 catalysts, the (1 0 1), (0 0 4), (1 0 5), (2 1 1) planes of tetragonal anatase TiO₂ (JCPDS file No. 21-1272, marked with ‘⁎’) for ZT3 and ZT10 catalysts, and the (1 1 0), (1 0 1), (2 0 0), (1 1 1), (2 1 1), (2 2 0), (0 0 2), (3 1 0) and (1 1 2) planes of tetragonal rutile SnO₂ (JCPDS file No. 41-1445, marked with ‘#’) for ZS3 and ZS10 appear, of which the peak intensity enhances with the increase in the content.

Fig. 1.

XRD patterns of (a) pure ZnO and (b) ZnO-based heterostructured nanocomposite catalysts.

The average crystalline sizes and lattice parameters of ZnO nanocrystals in all the catalysts were calculated according to the following Scherrer's formula (3), Bragg's formula (4) and lattice constant calculation formula (5).

$$D=k\lambda /\beta \cos \theta$$ (3)

$$2d\sin \theta =\lambda$$ (4)

$$1/d^2=4(h^2+hk+k^2)/3a^2+(l^2/c^2)$$ (5)

where the value of k is taken as 0.89, λ represents the diffraction wavelength of X-rays which is about 0.15405981 nm, β and θ are the full width at half maximum and diffraction angle of the diffraction peak. d is the interplanar spacing. h, k, l are Miller indices, a and c are lattice constants.

As shown in Table S1, the ZnO nanocrystals in the nanocomposite catalysts exhibit the slightly larger average crystalline sizes and lattice parameters compared with pure ZnO nanocatalysts. This originates from the re-crystallization of ZnO nanocrystals induced by secondary calcining during the synthesis of ZnO-based nanocomposite catalysts, and the incorporation of CuO, TiO₂ and SnO₂ does not significantly affect the crystallinity of ZnO nanocrystals.

Morphology and microstructure analysis

The morphologies of all the as-synthesized catalysts are observed by FESEM. The SEM images in Fig. 2a–d and Fig. S2 illustrate that the pure ZnO nanocatalyst, ZC3, ZT3 and ZS3 nanocomposite catalysts consist of spheroidal nanoparticles with wide particle size ranges (insets of Fig. 2a–d), meanwhile there are some nanoplates with a thickness of ∼50 nm in ZC3. These nanoparticles are aggregated due to their large surface area. The degree of particle agglomeration decreases with the increase of CuO, TiO₂ and SnO₂ contents for ZC10, ZT10 and ZS10 nanocomposite catalysts, and furthermore, a large number of nanorods are formed in ZT10. The average particle sizes of ZnO, ZC3, ZT3 and ZS10 are determinated to be 62.26, 67.48, 67.17, 53.90 nm by a Statistical Sampling Method based on a Nano Measurer 1.2.5 software, respectively. The smaller the particle size and the less the particle agglomeration, the larger the active surface area for contact between catalysts and surroundings.

Fig. 2.

FE-SEM images of (a) pure ZnO and (b–d) ZnO-based heterostructured nanocomposite catalysts (b: ZC3, c: ZT3, d: ZS10), the corresponding particle size distribution are presented in the insets. HRTEM images of (e) ZC3, (f) ZT3 and (g) ZS10 heterostructured nanocomposite catalysts.

To analyse the microstructures of the nanocomposite catalysts, the ZC3, ZT3 and ZS10 were observed via TEM, and the corresponding high-resolution TEM (HRTEM) images are illustrated in Fig. 2e–g. These composite nanoparticles present a well-defined heterjunction structure. The lattice fringes of 0.231, 0.148, 0.334, 0.282, 0.261 and 0.248 nm shown in Fig. 2e–g are indexed to the interplanar spacing of the (2 0 0) plane of monoclinic CuO, (0 0 2) plane of tetragonal anatase TiO₂, (1 1 0) plane of tetragonal rutile SnO₂ as well as (1 0 0), (0 0 2) and (1 0 1) planes of hexagonal wurtzite ZnO, respectively.

Surface chemical state analysis

Fig. S3a shows the XPS survey spectra of pure ZnO and ZnO-based heterostructured nanocomposite catalysts, from which the strong signals related to Zn and O elements are detected for all the catalysts. In addition to this, for the ZC3&ZC10, ZT3&ZT10 and ZS3&ZS10, the Cu, Ti and Sn elements have also been respectively detected by XPS. The high resolution Zn 2p XPS spectra in Fig. 3a and Fig. S3b present the same intervals of 23.1 eV between the Zn 2p_3/2 and Zn 2p_1/2 peaks, demonstrating that Zn elements exist in all the catalysts in the form of Zn²⁺ lattice ions [42,43].

Fig. 3.

High-resolution XPS spectra of pure ZnO and ZnO-based heterostructured nanocomposite catalysts: (a) Zn 2p, (b) O 1s, (c) Cu 2p, (d) Ti 2p and (e) Sn 3d.

As a matter of fact, it is generally believed that the crystallization of nanostructured metal oxides during the chemical synthesis processes can induce the deficiency of oxygen atoms to form oxygen vacancy defects. Especially on their surfaces, it is easy to produce surface oxygen vacancy defects, which contribute to the visible-light absorption, charge carrier separation and surface adsorption of target molecules of catalysts for the enhanced photocatalytic activity [[43], [44], [45]]. As described in Fig. 3b and Fig. S3c, all the O 1s XPS spectra have an asymmetric profile with a broad shoulder peak at the higher binding energy, indicating the coexistence of abundant oxygen-containing species on the surfaces of the as-synthesized nanocatalysts, such as –OH, H₂O and O₂ [46,47]. Based on Gauss multi-peak Fitting, the O 1s XPS spectra can be decomposed into three peaks at 529.8 ± 0.5 eV, 531.7 ± 0.1 eV and 533.1 ± 0.3 eV, which correspond to lattice oxygen ions (O_L) in the catalyst matrix [[48], [49], [50]], and loosely bound oxygen atoms from the dissociatively adsorbed –OH (O_H) and physically adsorbed H₂O&O₂ (O_C) on the surface oxygen vacancy defect sites [43,51], respectively. Table S2 gives the calculated atomic percentages for three oxygen species according to the respective areas of the decomposed O 1s peaks, in which the total atomic percentages of O_H and O_C reach 41.86, 36.24, 33.13 and 34.21 at.% for pure ZnO, ZC3, ZT3, and ZS10, respectively. It indicates that compounding other metal oxide semiconductors improves the crystalline qualities of the ZnO-based nanocomposite catalysts, their surface oxygen vacancy defects decrease. This is ascribed to the secondary crystallization of the individual component in the nanocomposite catalysts during the sol–gel synthesis processes, which is consistent with the XRD results. Fig. 3c–e give the high-resolution Cu 2p, Ti 2p and Sn 3d XPS spectra of the nanocomposite catalysts, respectively. In Cu 2p XPS spectra (Fig. 3c), the two dominant peaks located at ∼932.8 eV and ∼952.6 eV correspond to Cu 2p_3/2 and Cu 2p_1/2, which are accompanied by the weaker satellite peaks located at ∼943.3 eV and ∼961.8 eV, respectively. The appearance of these peaks suggests Cu²⁺ lattice ions of ZC3 and ZC10 [52,53], which confirms the existence of CuO in the ZnO/CuO nanocomposite catalysts. In Ti 2p XPS spectra (Fig. 3d), the peaks assigned to Ti 2p_3/2 and Ti 2p_1/2 are separately centred at ∼458.6 eV and ∼464.2 eV. The binding energy differences are calculated to be ∼5.6 eV, representing a typical characteristic for the Ti⁴⁺–O bonds in ZnO/TiO₂ nanocomposite catalysts (ZT3 and ZT10) [54,55]. As for Sn 3d XPS spectra (Fig. 3e), the observed three peaks at ∼486.2 eV, ∼494.7 eV and ∼497.8 eV can be separately assigned to Sn 3d_5/2, Sn 3d_3/2 and Zn LMM, and the about 8.5 eV spin–orbit splitting of Sn 3d is consistent with the standard values of SnO₂, indicative of Sn⁴⁺ lattice ions in ZnO/SnO₂ nanocomposite catalysts (ZS3 and ZS10) [[56], [57], [58]].

Optical characteristic analysis

Fig. 4a shows the UV–Vis absorption spectra evolved from the UV–Vis diffuse reflectance data of the as-synthesized catalysts, in which the ZC3 has a much stronger light absorption in the visible region compared with pure ZnO owning to the introduction of p-type CuO with a low band-gap energy. Whereas introducing the TiO₂ and SnO₂ with high band-gap energies do not improve light absorption of the composite catalysts. According to the Kubelka-Munk equation, the band-gap energies of the ZnO, ZC3, ZT3 and ZS10 are estimated as 3.14, 3.12, 3.15 and 3.15 eV by drawing the plots of (F(R)hv)^1/2 versus photon energy (hv) (Fig. 4b), indicative of negligible influences of compounding other metal oxide semiconductors on the band-gap energy of ZnO. Correspondingly, the UV–Vis absorption spectra of pure oxides (CuO, TiO₂ and SnO₂) are shown in Fig. S4, from which the estimated band-gap energies are 1.58, 3.25 and 3.62 eV for CuO, TiO₂ and SnO₂, respectively (Fig. 4c).

Fig. 4.

(a) UV–Vis absorption spectra and (b) band-gap energies calculated by the Kubelka-Munk equation of pure ZnO and ZnO-based heterostructured nanocomposite catalysts. (c) Calculated band-gap energies, (d) XPS valence band spectra and (e) proposed band-gap diagrams of pure ZnO, CuO, TiO₂ and SnO₂.

In addition, the band alignments of the ZC3, ZT3 and ZS10 heterostructured composites are determined by the estimated band-gap energies (Fig. 4c) and XPS valence band spectra (Fig. 4d) of pure ZnO, CuO, TiO₂ and SnO₂, and the corresponding energy band diagrams are drawn in Fig. 4e. Their band-gap energies and valence-band maximums (VBMs) respectively reach up to 3.14, 1.58, 3.25, 3.62 eV and 2.71, 0.41, 2.31, 3.23 eV, and correspondingly the conduction-band minimums (CBMs) of −0.43, −1.17, −0.94, −0.39 eV. From these data, it can be inferred that a direct Z-scheme heterojunction might be established between ZnO and CuO for ZC3 because of the small difference in CBMs of ZnO and VBMs of CuO. However, the ZT3 and ZS10 demonstrate a typical type-Ⅱ charge transfer behavior between ZnO and TiO₂ or SnO₂.

Charge separation and transfer behavior analysis

In general, the fast interfacial charge transfer induced by the heterostructures formed among different semiconducting materials can effectively accelerate the separation of photogenerated electrons and holes, achieving the enhanced photocatalytic performances [[59], [60], [61]]. The PL signals of semiconducting materials come from their radiative recombination of photoinduced electrons and holes, and its emission profile can describe the separation and transfer behaviors of photogenerated charge carriers. Figs. 5a and S5(b) illustrate the room temperature PL spectra of the as-synthesized catalysts, from which all the catalyst samples present the near-band-edge emission at around 395 nm and relatively strong blue-green emission peaks at about 465 nm. These two main emission bands are ascribed to the free exciton recombination of ZnO [62], and the radiative transition from the shallow donor levels to the deep acceptor levels associated with surface oxygen vacancies of ZnO [51,62], respectively. For the whole PL emission band, the emission intensity reduces with the formation of binary heterojunction between ZnO and other metal oxide semiconductors. The enhanced separation efficiency of photogenerated electron-hole pairs caused by the fast charge transfer at the heterojunction interfaces is responsible for the reduced PL emission. In particular, the ZC3 exhibits much lower PL emission than those of other heterostructured nanocomposite catalysts (ZT3 and ZS10).

Fig. 5.

(a) PL spectra, (b) SPV spectra and (c) EIS of pure ZnO and ZnO-based heterostructured nanocomposite catalysts. SPV phase spectrum of (d) ZC3, (e) ZT3 and (f) ZS10 heterostructured nanocomposite catalysts.

The SPV spectra and EIS illustrated in Fig. 5b and c, Fig. S5c and d further confirm the PL results. As shown, the SPV signal intensity ranks in the order of ZC3 > ZT3 > ZS10 > ZnO, while the diameter of semicircular Nyquist plots of EIS features follows the order of ZC3 < ZS10 < ZnO. Accompanied by the construction of binary heterojunction between ZnO and other metal oxides, the electrons and holes are instantly transferred to both sides of the heterojunction interfaces, and thus the charge transfer resistance reduces i.e. the semicircular arc of Nyquist plot decreases. Such an interfacial charge transfer can effectively facilitate the photogenerated electron-hole separation. The electrons and holes separated at both sides of the heterojunction interfaces delivery the higher SPV signal. As for the ZC3, the lowest PL signal, the strongest SPV signal and the minimum semicircular arc of Nyquist plot indicate that its interfacial charge transfer behavior is different from ZT3 and ZS10 and the highest separation efficiency of the photogenerated electrons and holes.

Fig. 5d–f separately describe SPV phase spectrum of ZC3, ZT3 and ZS10, the phase value is between 0° and 180° for ZC3, which implies the photoinduced electrons preferentially migrate toward the surface irradiated by light [63,64]. In contrast, the phase values are between −90° and 0° for ZT3 and ZS10, meaning the photoinduced holes preferentially migrate toward the surface irradiated by light.

Photocatalytic degradation characteristic analysis

Photo-degradation of organic dyes under simulated sunlight irradiation

Fig. S6 depicts the simulated sunlight-driven photocatalytic degradation of MO and MB dyes at room temperature (25 °C) in the presence of as-synthesized catalysts. The degradation curves at the 20 min (−20–0 min) before exposed light irradiation indicates that all the as-synthesized catalysts adsorb the dye molecules weakly, and thus photocatalytic oxidation play a decisive role decomposition of MO and MB dyes [65,66]. In further observation, all catalysts exhibit a stronger adsorption capacity for MB molecules than for MO molecules. This is primarily because the adsorption of dye molecules is governed by electrostatic interactions and hydrogen bonding. At initial conditions (dye aqueous solution pH: ∼6.5), the electrostatic adsorption of the anionic dye MO is stronger than that of the cationic dye MB. Consequently, hydrogen bonding is dominant in the adsorption process. In other words, more hydrogen bonds are formed between MB molecules and the surfaces of catalysts, inducing a stronger adsorption. It enhances hydrogen bonding adsorption due to the presence of a greater number of hydrogen bond acceptors (aromatic C − H) and donors (N and O atoms) in MB molecules [[67], [68], [69]].

From Fig. S6a–c, the other heterostructured nanocomposite catalysts excepting ZS3 deliver the enhanced photocatalytic activities than that of pure ZnO nanocatalysts, which can implement efficient degradation of MB dye within 50 min. For the photo-degradation of MO dye, the ZnO/CuO and ZnO/TiO₂ nanocomposite catalysts show a higher degradation efficiency compared with pure ZnO nanocatalysts (Fig. S6d and e). However, the SnO₂ compounding ZnO deteriorates the photocatalytic activity of ZnO/SnO₂ nanocomposite catalysts for MO photo-degradation (Fig. S6f).

The above results indicate that the enhanced separation of photogenerated electrons and holes induced by binary heterojunction between ZnO and other metal oxides increases the photocatalytic degradation efficiency of ZnO-based heterostructured nanocomposite catalysts towards to organic dyes. The MB degradation is more efficient than the MO degradation under simulated sunlight irradiation. Interestingly, the ZnO/TiO₂ and ZnO/SnO₂ nanocomposite catalysts exhibit a diametrically opposed photocatalytic degradation characteristic towards MO and MB dyes. In order to clearly compare the differences in photo-degradation of organic dyes using the various nanocomposite catalysts, the data associated with photocatalytic degradation of pure ZnO, ZC3, ZT3 and ZS10 nanocatalysts towards MO and MB dyes are re-displayed in Fig. 6a and b, respectively. The insets of Fig. 6a and b show that the photocatalytic degradation of MO and MB dyes for all the catalysts obey the pseudo-first-order kinetic model reaction. The fitted apparent rate constants for MO degradation are 0.0327, 0.0482, 0.0516 and 0.0161 min⁻¹ for pure ZnO, ZC3, ZT3 and ZS10 nanocatalysts, while 0.0281, 0.0789, 0.0684 and 0.0786 min⁻¹ for MB degradation in sequence. Obviously, the reduced photo-degradation efficiency of MO by compounding SnO₂ contrasts starkly with the boosted photo-degradation efficiency of MB.

Fig. 6.

Degradation rate curves of (a) MO and (b) MB in dye wastewater (pH = 6.5) using pure ZnO and ZnO-based heterostructured nanocomposite catalysts under simulated sunlight irradiation at room temperature, insets show the pseudo-first-order kinetic curves. (c–e) Photodegradation rate curves of MB in dye wastewater with various pH values using ZnO-based heterostructured nanocomposite catalysts, insets give the variation in the UV–Vis absorption spectra of MB wastewater as the degradation time. Cycling runs for MB photodegradation (pH = 6.5) using (f) ZC3, (g) ZT3 and (h) ZS10 heterostructured nanocomposite catalysts.

Dye solution pH, catalyst dosage and initial dye concentration influences and cycling stability

In the external factors affecting the efficiency of photocatalytic degradation system, the dye solution pH, catalyst dosage and initial dye concentration levels play vital roles. Fig. 6c–e display the photocatalytic degradation of MB dye aqueous solutions at various pH levels under simulated sunlight irradiation at room temperature. No matter what kind of nanocomposite catalysts, the photo-degradation efficiency increases with the increase in MB dye solution pH level, which is largely because the H₂O molecules in the basic solution are more easily ionized into hydroxide anions (OH⁻), promoting the generation of the reactive oxygen species (ROSs) such as hydroxyl radicals ($·\text{OH}$) [70,71].

Figs. S8 and S9 demonstrate the effects of catalyst dosage and initial dye concentration levels on the degradation of MB and MO dyes under simulated sunlight irradiation at room temperature, respectively. For catalyst dosage level, all the catalysts deliver a trend of increasing first and then decreasing for the degradation of MB and MO dyes with increasing it (10–100 mg per 100 ml of dye solution). The upgraded degradation efficiency by increasing catalyst dosage to the optimal level is ascribed to the increased exposed surface areas of catalysts for incident photons and dye molecules. But continuing to increase the dosage can hinder the absorption of incident light and adsorption of dye molecules on its surfaces due to the stacking of the catalysts, reducing the efficiency of photocatalytic degradation. For initial dye concentration level, increasing it can lead to more dye molecules participating in the photocatalytic reactions, slowing down the degradation rate of MB and MO dyes.

Fig. 6f–h show the photocatalytic degradation of mb dye for five consecutive cycles by using ZC3, ZT3 and ZS10 nanocatalysts, respectively. All the nanocomposite catalysts achieve good reproducibility and stability, which keep a high and lossless photocatalytic activity after five successive runs. Furthermore, the as-synthesized composite photocatalysts exhibit the superior photocatalytic degradation performances compare with some existing counterparts, as shown in Table S3.

Sacrificial agent influence and ZPC values

For elucidation of the rationales behind the differentiated photocatalytic degradation behaviors of ZnO/CuO, ZnO/TiO₂ and ZnO/SnO₂ nanocomposite catalysts towards MO and MB dyes. The active radical trapping tests, and zeta potential measurements at various pH values for ZC3, ZT3 and ZS10 were carried out, and the corresponding results are recorded in Fig. 7. As can be seen from Fig. 7a–c that adding EDTA-2Na and BQ maximally reduce the photo-degradation percentage of MO and MB dyes, respectively, indicative of the predominant roles of ${\text{h}}_{h\nu }^{+}$ and $·{\text{O}}_2^{-}$ in the photocatalytic oxidation reaction processes for the MO and MB degradation. Furthermore, the degradation are also contained to some extent with addition of other free radical scavengers (IPA and BQ for MO degradation; EDTA-2Na and IPA for MB degradation). It indicates that the other active species ($·\text{OH}$ and $·{\text{O}}_2^{-}$ for MO degradation; ${\text{h}}_{h\nu }^{+}$ and $·\text{OH}$ for MB degradation) can also effectively oxidize and degrade the two organic dyes. In comparison, the ZC3 exhibits a quite different photo-degradation behavior from ZT3 and ZS10 in presence of various free radical scavengers, where the reduced photocatalytic activity of ZC3 is more obvious when adding EDTA-2Na and BQ. This can be attributed to the more positive valence band potential and more negative conduction band potential of ZC3, further confirming the formation of direct Z-scheme heterojunction of ZnO/CuO nanocomposite catalysts. Fig. 7d–f present separately the plots of zeta potentials of ZC3, ZT3 and ZS10 versus pH values, from which their zero point charge (ZPC) values (the pH values at zero zeta potential) are almost the same. They are approximately 7.86, 7.93 and 7.90, respectively.

Fig. 7.

Photodegradation percentage of MO (degradation time: 80 min) and MB (degradation time: 50 min) using (a) ZC3, (b) ZT3 and (c) ZS10 heterostructured nanocomposite catalysts in presence of various free radical scavengers (the related data from the Fig. S7). Zeta potential of (d) ZC3, (e) ZT3 and (f) ZS10 heterostructured nanocomposite catalysts at various pH values.

Photocatalytic degradation mechanism

Based on the above discussion, a deeper insight into possible mechanism for specific photocatalytic degradation of the different binary heterojunction nanostructures towards MO and MB dyes is provided. As shown in Fig. 8, the charge transfer behaviors in photocatalytic processes are governed by the heterojunction configuration of nanocomposite catalysts, determining the photocatalytic degradation efficiency. When the sunlight irradiates on the surfaces of nanocomposite catalysts, the electrons (e^-) in the filled VB are excited by the photons with the energies greater than the band-gap energies of individual components and transfer to the empty CB, resulting in the generation of energetic photogenerated electrons (${\text{e}}_{h\nu }^{-}$) and holes (${\text{h}}_{h\nu }^{+}$).

Fig. 8.

Schematic diagram of photocatalytic degradation mechanism of ZnO/CuO, ZnO/TiO₂ and ZnO/SnO₂ heterostructured nanocomposite catalysts for MO and MB in dye wastewater under sunlight irradiation.

For ZnO/CuO nanocomposite catalyst, its direct Z-scheme heterojunction induces the recombination of ${\text{e}}_{h\nu }^{-}$ in the CB of ZnO and ${\text{h}}_{h\nu }^{+}$ in the VB of CuO, and thus the ${\text{e}}_{h\nu }^{-}$ and ${\text{h}}_{h\nu }^{+}$ participating in the redox reactions are accumulated in the CB of CuO and VB of ZnO, respectively. For ZnO/TiO₂ and ZnO/SnO₂ nanocomposite catalysts, the conventional type-Ⅱ charge transfer occurs at the interfaces, where the ${\text{e}}_{h\nu }^{-}$ transfer from TiO₂ to ZnO for ZnO/TiO₂, and from ZnO to SnO₂ for ZnO/SnO₂, respectively, the ${\text{h}}_{h\nu }^{+}$ transfer in the opposite paths. Compared with the interfacial charge transfer of the type-Ⅱ heterojunction catalysts (ZnO/TiO₂ and ZnO/SnO₂), the direct Z-scheme charge transfer at the interfaces between ZnO and CuO endows the ZnO/CuO nanocomposite catalysts with higher separation efficiency of photogenerated electrons and holes. Meanwhile, the introduction of CuO with a narrow band-gap increases the visible light absorption. Consequently, the superior photocatalytic properties for degradation of MO and MB dyes are achieved for ZnO/CuO nanocomposite catalysts.

According to the References [72,73], the redox potentials of O₂/$·{\text{O}}_2^{-}$ and OH⁻/$·\text{OH}$ are −0.33 eV and + 1.99 eV, respectively, which are between the effective CB and VB potentials of ZC3 (−1.17 eV ∼ +2.71 eV), ZT3 (−0.43 eV ∼ +2.31 eV) and ZS10 (−0.39 eV ∼ +2.71 eV). Therefore, the ROSs of $·{\text{O}}_2^{-}$ and $·\text{OH}$ can be produced by the reactions of ${\text{e}}_{h\nu }^{-}$ and ${\text{h}}_{h\nu }^{+}$ with O₂ molecules and OH⁻, respectively. The ${\text{h}}_{h\nu }^{+}$, $·{\text{O}}_2^{-}$ and $·\text{OH}$ can degrade organic dye molecules into green products such as CO₂, H₂O and low-molecular-mass organic groups (LOGs [74,75]). In the present photocatalytic systems, ${\text{h}}_{h\nu }^{+}$ and $·{\text{O}}_2^{-}$ play the predominant roles for the degradation of MO and MB dyes, respectively. Meanwhile, the $·\text{OH}$ is also one of the most important ROSs for MO and MB degradation.

For all the nanocomposite catalysts, the diametrically opposed photocatalytic degradation characteristic of ZT3 and ZS10 for MO and MB dyes can be attributed to the differences of interfacial charge transfer behaviors and dye degradation mentioned above. ZT3 has the VB potential closest to the redox potentials of OH⁻/$·\text{OH}$, it is easiest to generate the $·\text{OH}$. However, the CB potential of ZS10 is closest to the redox potentials of O₂/$·{\text{O}}_2^{-}$, the $·{\text{O}}_2^{-}$ is most easily generated. As a consequence, the degradation of MO dye is more efficient in the presence of ZT3, and the MB dye is more efficiently degraded by ZS10.

Conclusions

A series of ZnO-based heterostructured nanocomposite photocatalysts including ZnO/CuO, ZnO/TiO₂ and ZnO/SnO₂ with the different molar ratios (3 mol% and 10 mol% of CuO/TiO₂/SnO₂) were synthesized by a combination of modified polymer-network gel and traditional sol–gel methods. Pure ZnO as a reference along with these ZnO-based nanocomposites were comparatively investigated on the photocatalytic degradation of MO and MB dyes in wastewater under simulated sunlight irradiation. Combining with an appropriate amount of other metal oxide semiconductors contributes to enhance the photocatalytic activity of ZnO for the degradation of organic pollutants, however, which is not the case for the MO degradation using ZnO/SnO₂ nanocomposite catalysts. The incorporation of 10 % SnO₂ increases the efficiency for the photocatalytic degradation of MB by ZnO from 0.0281 min⁻¹ to 0.0786 min⁻¹, while for the photocatalytic degradation of MO, it decreased from 0.0327 min⁻¹ to 0.0161 min⁻¹. The enhanced photocatalytic activity is largely ascribe to the separation of photogenerated electrons and holes facilitated by the fast charge transfer at interfaces between the different components. Such an effect is magnified by direct Z-scheme heterojunction of ZnO/CuO nanocomposite catalysts, achieving the superior photocatalytic properties for MO and MB degradation for ZC3. For ZnO/TiO₂ and ZnO/SnO₂ nanocomposite catalysts with a type-Ⅱ heterojunction, they exhibit the diametrically opposed photocatalytic degradation characteristic for MO and MB dyes (the more efficient MO degradation and less efficient MB degradation for ZnO/TiO₂, the opposite is true for ZnO/SnO₂). The reasons for the specific photocatalytic degradation are that the varied heterojunction configurations defines the different band alignments for governing the selective formation of active species (${\text{h}}_{h\nu }^{+}$, $·{\text{O}}_2^{-}$ and $·\text{OH}$) in the oxidative decomposition of dye molecules.

Compliance with ethics requirements

This article does not contain any studies with human or animal subjects.

CRediT authorship contribution statement

Mengjiao Wu: Methodology, Investigation, Visualization, Writing – original draft. Chengpu Lv: Validation, Investigation, Writing – review & editing. Yuling Xiong: Investigation, Visualization, Methodology. Wenglong Li: Investigation, Visualization, Methodology. Yuangui Lin: Investigation, Visualization, Methodology. Jing Li: Investigation, Visualization, Methodology. Fei Yu: Visualization, Methodology, Writing – review & editing. Huan Yuan: Visualization, Methodology, Writing – review & editing. Biao You: Visualization, Methodology, Writing – review & editing. Qiuping Zhang: Conceptualization, Data curation, Methodology, Writing – review & editing, Supervision. Ming Xu: Conceptualization, Methodology, Writing – review & editing, Supervision.

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.

Appendix A. Supplementary data

The following are the Supplementary data to this article: