Continuously Improved Photocatalytic Performance of Zn2SnO4/SnO2/Cu2O Composites by Structural Modulation and Band Alignment Modification

Improving the photocatalytic performance of multi-component photocatalysts through structural modulation and band alignment engineering has attracted great interest in the context of solar energy utilization and conversion. In our work, Zn2SnO4/SnO2 hierarchical architectures comprising nanorod building block assemblies were first achieved via a facile solvothermal synthesis route with lysine and ethylenediamine (EDA) as directing agents, and then chemically etched in NaOH solution to enlarge the surface area and augment active sites. The etched Zn2SnO4/SnO2 hierarchical architectures were further decorated by Cu2O nanoparticles though an in situ chemical deposition method based on band alignment engineering. In comparison with unetched Zn2SnO4/SnO2, the specific surface area of Zn2SnO4/SnO2/Cu2O hierarchical architectures became larger, and the responsive region and absorbance intensity became wider and higher in the whole visible-light range. Zn2SnO4/SnO2/Cu2O hybrid photocatalysts presented enormously improved visible-light photocatalytic behaviour for Rhodamine B (RhB) decomposition. The enhancement of photocatalytic behaviour was dominantly attributed to the synergy effect of the larger specific surface area, higher light absorption capacity, and more effective photo-induced charge carrier separation and migration. A proposed mechanism for the enormously promoted photocatalytic behaviour is brought forth on the basis of the energy-band structure combined with experimental results.


Introduction
As one of the widely investigated ternary oxide materials in the application of photodegrading pollutants [1][2][3][4][5][6], Zn 2 SnO 4 (ZSO) has the prominent advantages of relatively high electron mobility and long-term chemical stability. However, limited to its wide bandgap and unfavorable recombination rate, the photocatalytic efficiency of pure ZSO is not satisfied. Thus, researchers have taken some approaches, for instance, ion doping [2], morphological modulation [3][4][5][6], and coupling composites [7][8][9][10][11][12][13][14][15][16], to improve the photocatalytic behavior of ZSO based photocatalyst. Among those strategies mentioned above, relevant reports on doping in ZSO are very limited, probably because the concentration of dopants is not easily controlled and the impurity level can act not only as a platform for electron transition, but also serve as a recombination center. Further, we all know that it is very difficult for a single component photocatalyst to overcome the intrinsic drawbacks of a low quantum rate and high recombination rate only by morphological modulation. In this regard, constructing composites is one of the most facile and efficient solutions, because the construction of coupling ZSO with other matchable photocatalysts is beneficial for prolonging the lifespan of photo-generated charge carriers and accelerating charge carrier separation [14][15][16][17].
As previously reported [7][8][9][13][14][15], SnO 2 is available to construct ZSO/SnO 2 composites displaying the improvement of photodegradation rate. For instance, Chen et al. [7] took a one-pot hydrothermal synthesis method to fabricate SnO x /Zn 2 SnO 4 composites, which exhibited a higher and more stable photocatalytic performance toward the degradation of MO and gaseous C 6 H 6 under UV light irradiation. Junploy et al. [8] reported an approach involving co-precipitation combined with calcination for the synthesis of Zn 2 SnO 4 /SnO 2 , and the results showed that the as-obtained Zn 2 SnO 4 /SnO 2 displayed improved photocatalytic efficiency toward the degradation of MB under UV light illumination, as compared to Zn 2 SnO 4 and SnO 2 . Most previously reported studies were focused on the improvement of UV-light photocatalytic behavior over Zn 2 SnO 4 /SnO 2 photocatalysts, while the relevant studies on visible light photocatalytic activity were relatively scarce, probably because both Zn 2 SnO 4 and SnO 2 belong to the group of wide band-gap photocatalysts. Generally, the photocatalytic performance in visible-light region over Zn 2 SnO 4 /SnO 2 coupling composites is conceivably triggered by defect states originating from oxygen vacancy or interstitial ions. Han et al. [13] fabricated double-shell Zn 2 SnO 4 /SnO 2 microboxes with superior photocatalytic performance under sun-light illumination. Zhang et al. [14] designed a reduced method to fabricate surface-disordered Zn 2 SnO 4 /SnO 2 hybrid nanocomposites, which exhibited improved sun-light driven degradation performance of organic pollutant owing to the introduction of oxygen vacancy and amorphous shells. However, the promotion of light absorption capacity is limited only by the introduction of defects. It is desirable to indraught narrowed band photocatalyst for further boosting of the light adsorption of the Zn 2 SnO 4 /SnO 2 hybrid. Ho et al. [15] prepared a novel visible-light-response photocatalyst Zn 2 SnO 4 -SnO 2 /graphene (rGO) with improved photo-reactivity, and the results showed that rGO performed as an electron mediator to favor the high photocatalytic efficiency toward acetone and NO oxidation. Despite the previous achievements mentioned above, it is still a challenge to achieve a novel, multi-component Zn 2 SnO 4 -SnO 2 -based photocatalyst by simultaneously extending the light absorption range and increasing the spatial separation of photogenerated charge carriers.
Cu 2 O is regarded as a promising candidate for visible-light photocatalyst owing to its suitable band-gap (2.1-2.4 eV), amazing optical property, low cost, non-toxicity, and environmental acceptability [18][19][20][21]. However, pure Cu 2 O has a short charge diffusion length (20-100 nm), resulting in a high recombination rate of photoinduced electron-hole pairs and quite low quantum efficiency [22]. For this purpose, as an efficient co-catalyst, Cu 2 O could be added into some photocatalyst systems, e.g., ZnO [21,23], TiO 2 [24,25], CdS [26], ZnWO 4 [27], WO 3 [19,28], BiVO 4 [29], BiOBr [30], in order to broaden light absorption scope, facilitate charge carriers transfer and separation, and then boost up the photocatalytic performance. Inspired by previously reported achievements, we developed a novel "directing growth-etching treatment-coupling composites" route to prepare Zn 2 SnO 4 /SnO 2 /Cu 2 O hybrid photocatalyst based on structural modulation and band alignment modification in this work. After the overall modification mentioned above, the as-obtained Zn 2 SnO 4 /SnO 2 /Cu 2 O hybrids were simultaneously endowed with a higher specific surface area, wider spectral absorption range, and more efficient spatial separation of photogenerated charge carriers, giving rise to an enormous improvement of photodegradation efficiency for the removal of RhB. To our certain knowledge, this is the first report on the fabrication and insight into the mechanism of enhanced photocatalytic behavior of Zn 2 SnO 4 /SnO 2 /Cu 2 O hybrids. After that, 0.2 mmol of lysine was added to the above mixture, which was continuously kept stirring for another 30 min. The obtained suspension was transferred into a 100 mL Teflon-lined autoclave, and maintained at 190 • C for 20 h. After being naturally cooled to room temperature (the cooling time is about 7 h), the white precipitate was centrifuged and washed with deionized water and anhydrous alcohol repeatedly. The resulting precipitate was dried at 70 • C in an electric drying oven overnight, and then grinded into powder. The as-obtained product was named P-ZSO/SnO 2 . For comparison, Zn 2 SnO 4 /SnO 2 microspheres (M-ZSO) were prepared via a facile route, and the detailed procedure was presented in the supporting information. First, 0.5 g of P-ZSO/SnO 2 was added to a 60 mL sodium hydroxide (5 M) solution with constant stirring for 20 min. Then, the suspension mentioned above was transferred into a 100 mL Teflon-lined autoclave and kept at 60 • C for 1 h. The subsequent washing and drying process was similar to that of P-Zn 2 SnO 4 /SnO 2 . The as-obtained product was denoted as ET-ZSO/SnO 2 .

Synthesis of Zn 2 SnO 4 /SnO 2 /Cu 2 O Hierarchical Architectures
Typically, 0.02 mmol EDTA-2Na, 0.02 mmol Cu(NO 3 ) 2 ·5H 2 O and a specified amount of ET-ZSO/SnO 2 were put into 50 mL deionized water to form a suspension, and sonicated for 30 min. 0.05 mmol NaOH and 0.1 mmol AA were respectively dissolved into 20 mL deionized water under constant stirring. The NaOH-solution was added dropwise into the suspension, and then followed by the sluggish drip of the AA-solution with vigorous stirring for another 40 min. The above mixture was stirred for 240 min at room temperature. The resulting product was treated by centrifuging and washing with deionized (D.I.) water and anhydrous alcohol to eliminate the impurity, followed by drying at 120 • C in a vacuum oven overnight. In this manner, coupling composites of 0.5, 1, 1.5, and 2% Cu 2 O loaded onto ET-ZSO/SnO 2 were prepared and denoted as ZSC0.5, ZSC1, ZSC1.5 and ZSC2, respectively.

Characterizations
The crystal structure analysis of P-ZSO/SnO 2 , ET-ZSO/SnO 2 , and ZSC hybrids were performed by using a D8 Advance X-ray diffractometer (Bruker, Billerica, MA, USA) with a copper radiation (λ = 0.15406 nm). An ESCALAB 250Xi spectrometer (Thermo Fisher Scientific, New York, NY, USA) equipped with a monochromatic Al-Kα radiation was employed to examine the surface component of P-ZSO/SnO 2 and ZSC hybrids. Morphological and micro-structural investigations of P-ZSO/SnO 2 and ZSC hybrids were achieved on a Hitachi-S-4800 field emission scanning electron microscope (Hitachi, Toyko, Japan) as well as JEM-2100F transmission electron microscope (JEOL Ltd., Tokyo, Japan). The Brunauer-Emmett-Teller (BET) specific surface area of P-ZSO/SnO 2 and ZSC hybrids was determined on the basis of multi-point estimation using a NOVA-2000e sorption analyzer. UV-DRS of P-ZSO/SnO 2 , ET-ZSO/SnO 2 and ZSC hybrids were collected on a TU 1901 UV-vis spectrophotometer (Puxi, Beijing, China), using BaSO 4 as a reference. Photoluminescence (PL) spectra excitated with a wavelength of 320 nm were achieved on a LS55 fluorescence spectrophotometer (PE, Waltham, MA, USA).

Photocatalytic Experiments
Commonly used RhB dye in the textile industry was taken as a model pollutant in our work. The procedure of RhB photodegradation tests were performed in a homemade apparatus, which was reported in our previous studies [10][11][12]31,32]. Briefly, 50 mg of photocatalyst was coated on the bottom surface of the culture dish (9 cm) to form a thin film, followed by dispersing into 40 mL RhB aqueous solution (1.0 × 10 −5 M). Two daylight lamps (60 W, λ ≥ 420 nm) were designated as the visible light source for triggering the photodegration reaction of RhB. After a specified interval of the simulant sun-light illumination, the suspension was sampled and the concentration of the reacted RhB aqueous solution was measured on a TU 1901 UV-vis spectrophotometer with the wavelength ranging from 300 nm to 700 nm.

Photocurrent Measurements
The photo-electrochemical measurements of P-ZSO/SnO 2 , Cu 2 O, and ZSC1.5 samples were carried out on a three-electrode CHI660E electrochemical workstation (Chenhua Instruments Co., Shanghai, China), which contained the counter electrode (Pt wire), reference electrode (saturated Ag/AgCl), and working electrode (photocatalysts). The details for the photo-electrochemical measurements were presented in our previously reported study [31,32].

Results and Discussion
Considering that 3D hierarchical architectures composed of one dimensional (1D) nanobuilding blocks, with desired size, shape and composition, play substantial roles in determining the photo-activity of photocatalysts, the morphological structure and composition of the as-obtained products were characterized by field emission scanning electron microscope (FESEM) in combination with EDS analysis, and TEM. From the FESEM image with low-magnification ( Figure 1a), P-ZSO/SnO 2 was made up of massive flower-like architectures. High-magnification FESEM micrograph ( Figure 1b) revealed clearly morphological structure that 3D flower-like architectures of P-ZSO/SnO 2 were comprised of a large number of closely interconnected 1D nanorod building blocks with small size. Comparing Figure 1b with Figure 1c, we can easily find that the ZSC1.5 sample almost inherited the hierarchical morphology of the corresponding P-ZSO/SnO 2 product, signifying that the etching process in combination with chemical deposition procedure used in this experiment only exerted a limited effect on the morphological structure. Such a special morphological structure will be beneficial for achieving high photo-activity [33][34][35]. The result of energy dispersive spectrometer (EDS) mapping ( Figure 1e) perfectly authenticated the coexistence and homogeneous distribution of Zn, Sn, O, and Cu elements throughout the ZSC1.5 sample. Additionally, as is seen from Supplementary Figure S1, the M-ZSO sample was composed of microsphere-shaped particles with the size ranging from 1 to 2 µm.
A deep insight into detailed structural features of the ZSC1.5 sample was provided by TEM observation, as shown in Figure 2a,b. Obviously, the obtained ZSC1.5 sample exhibited 3D flower-like architectures with a mean size of 100-200 nm, built up from plenty of nanorods with an average diameter of approximately 10-20 nm. A close-view of a single unit was revealed from Figure 2b that oriented nanorods (referring to ZSO and SnO 2 ) were grown together with the common part in the middle and congregated with each other to shape into flower-like architectures. Noticeably, fine discrete Cu 2 O nanoparticles were tightly anchored on the surface of ZSO and SnO 2 nanorods to establish an intimate contact along the interface between ZSO, SnO 2 and Cu 2 O, which would effectively accelerate the interfacial charge migration and separation.       Clearly, only two crystalline phases of ZSO and SnO 2 can be perfectly identified in P-ZSO/SnO 2 , well indexed to JCPDS card no. 74-2184 and 41-1445, respectively. No other diffraction peaks resulting from other residues or impurity phases, e.g., ZnO, were found, indicating the successful synthesis of the as-obtained product. Notably, the ET-ZSO/SnO 2 sample also had the same crystalline phases with P-ZSO/SnO 2 even through the etching process, although the corresponding diffraction peaks became broader and the diffraction peak intensity of ET-ZSO/SnO 2 grew weaker, compared with those of P-ZSO/SnO 2 . Other than ZSO and SnO 2 , no crystal phase related with Cu 2 O was detected in the ZSC1.5 sample, which might be attributed to the low concentration of Cu 2 O in comparison with that of ZSO and SnO 2 . On the other hand, pure Cu 2 O has sharp diffraction peaks, which are well indexed to JCPDS card no. 05-0667. Based on the analysis of XRD results, the possible etching mechanism for ZSO and SnO 2 could be proposed as following. Being subject to a high concentrated solution of NaOH, Zn 2+ and Sn 4+ could respectively coordinate with a specified amount of OHto form soluble complex radicals of Zn(OH) 4 2− and Sn(OH) 6 2− under hydrothermal condition. As a result, the components of ZSO and SnO 2 were continuously converted into soluble complex radicals and dissolved into the solution with hydrothermal time prolonging. In fact, the mass of P-ZSO/SnO 2 decreased to a certain degree after etching treatment. Thus, such an explanation can be completely supported by the result that there is little difference in crystal phase and morphological structure before and after the etching treatment.
Nanomaterials 2019, 9, x FOR PEER REVIEW 6 of 15 resulting from other residues or impurity phases, e.g., ZnO, were found, indicating the successful synthesis of the as-obtained product. Notably, the ET-ZSO/SnO2 sample also had the same crystalline phases with P-ZSO/SnO2 even through the etching process, although the corresponding diffraction peaks became broader and the diffraction peak intensity of ET-ZSO/SnO2 grew weaker, compared with those of P-ZSO/SnO2. Other than ZSO and SnO2, no crystal phase related with Cu2O was detected in the ZSC1.5 sample, which might be attributed to the low concentration of Cu2O in comparison with that of ZSO and SnO2. On the other hand, pure Cu2O has sharp diffraction peaks, which are well indexed to JCPDS card no. 05-0667. Based on the analysis of XRD results, the possible etching mechanism for ZSO and SnO2 could be proposed as following. Being subject to a high concentrated solution of NaOH, Zn 2+ and Sn 4+ could respectively coordinate with a specified amount of OHto form soluble complex radicals of Zn(OH)4 2− and Sn(OH)6 2− under hydrothermal condition. As a result, the components of ZSO and SnO2 were continuously converted into soluble complex radicals and dissolved into the solution with hydrothermal time prolonging. In fact, the mass of P-ZSO/SnO2 decreased to a certain degree after etching treatment. Thus, such an explanation can be completely supported by the result that there is little difference in crystal phase and morphological structure before and after the etching treatment.  (Figure 4b) can be divided into three peaks with binding energies at 530.5, 531.5 and 532.6 eV, which can be respectively denoted as lattice oxygen, vacancy oxygen, and chemisorbed oxygen species [17,36,37]. The vacancy oxygen species play a crucial effect on the visible-light photo-activity, as they are capable of inducing the formation of  (Figure 4b) can be divided into three peaks with binding energies at 530.5, 531.5 and 532.6 eV, which can be respectively denoted as lattice oxygen, vacancy oxygen, and chemisorbed oxygen species [17,36,37]. The vacancy oxygen species play a crucial effect on the visible-light photo-activity, as they are capable of inducing the formation of shallow energy levels. In terms of Zn 2p spectrum (Figure 4c), two fitting peaks with bind energies at~1044.6 and 1022.4 eV could be ascribed to Zn 2p1 /2 and Zn 2p 3/2 , respectively [15,17]. The prominent peaks centred at~495.2 and 486.8 eV in Figure 4d can be assigned to the Sn 3d 1/2 and Sn 3d 3/2 of Sn 4+ valence state. As for the chemical status of Cu element, the high-resolution spectrum of Cu 2p displayed two characteristic peaks with binding energies at~952.4 and 932.7 eV, which verified the existence of Cu 2 O in the ZSC1.5 sample. Moreover, by calculation, the concentration of Cu 2 O is 1.4 wt%, being approximate to the theoretical value. adsorption-desorption measurements. It is unambiguous from Figure 5a that three samples all displayed type IV isotherm accompanying with typical H3 hysteresis loop in the light of Brunauer-Deming-Deming-Teller (BDDT) classification, signifying the appearance of slit-like mesopores because of subunits stacking [16]. Consulting to the Brunauer-Emmett-Teller (BET) method, the specific surface area (SBET) of P-ZSO/SnO2, ET-ZSO/SnO2 and ZSC1.5 samples were approximately calculated to be 22.4 m 2 /g, 49.2 m 2 /g and 43.5 m 2 /g, respectively. By contrast, the SBET value of ET-ZSO/SnO2 was much larger than that of P-ZSO/SnO2, signifying that the absorption capacity of ET-ZSO/SnO2 would be substantially enhanced due to the introduction of etching treatment. Although a slight decrease in the SBET value occurred after Cu2O decoration, it is still rationally inferred that the ZSC1.5 sample will be provided with more active sites and exceptional absorption capacity as a novel photocatalyst, as compared with P-ZSO/SnO2. From Figure 5b, it is easily found out that all three samples contained mesopores ranging from 2nm to 50 nm. In comparison with P-ZSO/SnO2, some more mesopores with small sizes appeared for ET-ZSO/SnO2 and ZSC1.5, which was mainly attributed to the etching treatment and Cu2O modification. Such special characteristics of surface area and porosity would promisingly lead to high photo-activity for a multi-component photocatalyst.  It is common knowledge that photocatalytic redox reactions occur on the interface between photocatalysts and pollutants. As one of the most significant factors affecting the photo-activity of photocatalysts, absorption capacity is highly dependent on the surface specific area. Generally speaking, a photocatalyst with a larger surface area is likely to adsorb more pollutant molecules, being indicative of its excellent absorption capacity. As a consequence, the specific surface area, together with the porosity of the as-obtained products, was evaluated through nitrogen adsorption-desorption measurements. It is unambiguous from Figure 5a that three samples all displayed type IV isotherm accompanying with typical H3 hysteresis loop in the light of Brunauer-Deming-Deming-Teller (BDDT) classification, signifying the appearance of slit-like mesopores because of subunits stacking [16]. Consulting to the Brunauer-Emmett-Teller (BET) method, the specific surface area (S BET ) of P-ZSO/SnO 2 , ET-ZSO/SnO 2 and ZSC1.5 samples were approximately calculated to be 22.4 m 2 /g, 49.2 m 2 /g and 43.5 m 2 /g, respectively. By contrast, the S BET value of ET-ZSO/SnO 2 was much larger than that of P-ZSO/SnO 2 , signifying that the absorption capacity of ET-ZSO/SnO 2 would be substantially enhanced due to the introduction of etching treatment. Although a slight decrease in the S BET value occurred after Cu 2 O decoration, it is still rationally inferred that the ZSC1.5 sample will be provided with more active sites and exceptional absorption capacity as a novel photocatalyst, as compared with P-ZSO/SnO 2 . From Figure 5b, it is easily found out that all three samples contained mesopores ranging from 2nm to 50 nm. In comparison with P-ZSO/SnO 2 , some more mesopores with small sizes appeared for ET-ZSO/SnO 2 and ZSC1.5, which was mainly attributed to the etching treatment and Cu 2 O modification. Such special characteristics of surface area and porosity would promisingly lead to high photo-activity for a multi-component photocatalyst. In general, different kinds of photocatalysts exhibit different absorption features, either in the light absorption intensity or light absorption range. For this purpose, the diffuse reflection spectra (DRS) of pristine P-ZSO/SnO2, ET-ZSO/SnO2, and ZSC samples with different weight contents were taken and the results are given in Figure 6a. Either P-ZSO/SnO2 or ET-ZSO/SnO2 presented a reasonably poor visible-light absorption property due to the wide band-gap of ZSO and SnO2. From another point of view, pure Cu2O displayed a wide light response range and outstanding light harvesting in the whole visible light region. After ET-ZSO/SnO2 are composited with Cu2O, all ZSC samples expectedly displayed evidently enhanced optical absorption in the whole visible light region, imputing to the interaction between fine Cu2O nanoparticles and ZSO and SnO2 nanorod assemblies. reasonably poor visible-light absorption property due to the wide band-gap of ZSO and SnO2. From another point of view, pure Cu2O displayed a wide light response range and outstanding light harvesting in the whole visible light region. After ET-ZSO/SnO2 are composited with Cu2O, all ZSC samples expectedly displayed evidently enhanced optical absorption in the whole visible light region, imputing to the interaction between fine Cu2O nanoparticles and ZSO and SnO2 nanorod assemblies. As is well known, photoluminescence (PL) emission spectrum is confirmed to be feasible and effective for revealing the migration and recombination of photo-induced hole-electron pairs [6,38]. Specifically, a lower PL intensity signifies the lower recombination possibility of photogenerated hole-electron pairs. For this purpose, PL measurements for pristine P-ZSO/SnO2, ET-ZSO/SnO2, and ZSC hybrids were performed and the results are presented in Figure 6b. For P-ZSO/SnO2, the strongest intensity of PL emission peak was clearly indicative of the occurrence of the highest recombination possibility of photo-induced hole-electron pairs. With the addition of Cu2O, the PL As is well known, photoluminescence (PL) emission spectrum is confirmed to be feasible and effective for revealing the migration and recombination of photo-induced hole-electron pairs [6,38]. Specifically, a lower PL intensity signifies the lower recombination possibility of photogenerated hole-electron pairs. For this purpose, PL measurements for pristine P-ZSO/SnO 2 , ET-ZSO/SnO 2 , and ZSC hybrids were performed and the results are presented in Figure 6b. For P-ZSO/SnO 2 , the strongest intensity of PL emission peak was clearly indicative of the occurrence of the highest recombination possibility of photo-induced hole-electron pairs. With the addition of Cu 2 O, the PL emission peak intensity of ZSC hybrids declined sharply at first, then reached the lowest value, and tended to rise thereafter. ZSC1.5 exhibited the lowest PL emission intensity, revealing that it had the most efficient suppression of charge carrier recombination in this hybrid system. Moreover, it is worth noting that the emission peak with weaker intensity (centred at about~480 nm) is probably ascribed to oxygen vacancies [39,40]. Based on the analysis above, wider photoresponse range and more efficient spatial separation of photogenerated charge carriers were simultaneously achieved for ZSC hybrids through the co-modification with etching treatment and Cu 2 O nanoparticle decoration.
In our work, the photocatalytic behaviour against the as-prepared photocatalysts was assessed with RhB as a target pollutant owing to its universal use and good long-term stability. For comparison, the photocatalytic degradation curve over M-ZSO is presented in Figure S2. Undoubtedly, the M-ZSO sample showed a very poor photocatalytic activity. Thus, it is particularly desired to carry out an overall investigation in order to explore the effect of the structural modulation and decoration on the photo-activity. As portrayed in Figure 7a, a certain amount of RhB molecules was adsorbed on the surface of photocatalysts during the absorption-desorption equilibrium process prior to illumination. By contrast, the adsorbed amount of RhB molecules for ET-ZSO/SnO 2 and ZSC samples with different weight contents was larger than that for P-ZSO/SnO 2 , suggesting that the absorption capacity was indeed improved by the introduction of etching treatment, which was well consistent with the result of BET analysis. Under visible-light illumination, the absorbance values decreased continuously for all test samples. Among them, P-ZSO/SnO 2 displayed the worst photodegradation efficiency, mainly owing to its characteristics of energy band and morphological structure. For ET-ZSO/SnO 2 and ZSC samples, an evident improvement in photodegradation rate can be found, demonstrating that structural modulation combined with modification actually exerted a continuously positive effect on the photocatalytic performance. Specifically, the photodegradation efficiency of ZSC hybrids increased first with the Cu 2 O loading amount and reduced thereafter. The ZSC 1.5 sample exhibited the highest photodegradation rate, approaching 98% over a period of 50 min of visible-light illumination. Perhaps, excess Cu 2 O nanoparticles were likely to aggregate with each other and shadow the surface of ZSO and SnO 2 from the incident visible-light photons, resulting in poor light exposure of ZSO and SnO 2 for photo-induced charge carrier production. In order to quantitatively examine the reaction dynamics, we presumed that the photocatalytic behaviour for RhB aqueous solution followed the first-order kinetic model plotted by the equation of −ln(C/C 0 ) = kt, in which k, C 0 and C, respectively, stand for reaction rate constant, initial concentration, and concentration at time t of RhB concentration. From Figure 7b, it is very clear that ZSC1.5 had the highest kinetic constant of approximately 0.07 min -1 , which was around 7 and 6 times higher than that of P-ZSO/SnO 2 and ET-ZSO/SnO 2 . Considering that the usability and stability are also vital factors other than photocatalytic efficiency in the potentially practical application, ZSC1.5 was determined as the representative to recycle RhB degradation experiments for five runs, in which the photocatalyst was recovered by centrifugation, washing and drying at 65 • C for 10 h. The stability result for ZSC1.5 is given in Figure 7c. Only a slight decline was detectable in the photodegradation rate from the first to the fifth run, perhaps resulting from inevitable weight loss in the process of collecting and washing. The as-obtained results are indicative of high stability of ZSC hybrid photocatalyst.
According to previously reported studies [41,42], the photocurrent response of photocatalysts can accurately shed light on photo-induced electrons and holes transfer and spatial separation. As a consequence, the photocurrent measurements of P-ZSO/SnO 2 , Cu 2 O and ZSC1.5 were carried out to further witness the contribution of Cu 2 O to the charge carrier spatial separation efficiency of ZSC hybrids. As shown in Figure 8a, ZSC1.5 presented the largest photocurrent among all the tested samples, suggesting that it owned the longest charge carrier lifespan and highest spatial separation efficiency. Furthermore, the capacity of interfacial charge separation was revealed by electrochemical impedance spectra (EIS) Nyquist plots of P-ZSO/SnO 2 , Cu 2 O and ZSC1.5 electrodes under visible-light illumination, as depicted in Figure 8b. Evidently, ZSC1.5 had a much smaller arc radius than P-ZSO/SnO 2 and Cu 2 O, which was attributed to its smaller charge transfer resistance and faster interfacial transfer rate of photoinduced charge carriers [43][44][45].     Trapping experiments were devised to reveal what active species would dominate the photodegradation reaction process in presence of ZSC1.5 photocatalyst. According to previous reports [46,47], four scavengers of benzoquinone (BZQ 2 mmol L −1 ), sodium oxalate (SO, 4 mmol L −1 ), tert-butyl-alcohol (t-BuOH, 2 mmol L −1 ), and FeSO4-EDTA (FS-EDTA, 0.2 mmol L −1 ) were individually put into the system to capture the active species of •O2 − , h + , •OH and H2O2. The Trapping experiments were devised to reveal what active species would dominate the photodegradation reaction process in presence of ZSC1.5 photocatalyst. According to previous reports [46,47], four scavengers of benzoquinone (BZQ 2 mmol L −1 ), sodium oxalate (SO, 4 mmol L −1 ), tert-butyl-alcohol (t-BuOH, 2 mmol L −1 ), and FeSO 4 -EDTA (FS-EDTA, 0.2 mmol L −1 ) were individually put into the system to capture the active species of •O 2 − , h + , •OH and H 2 O 2 . The variation of photodegradation rate upon exposure to different scavengers is presented in Figure 9. Overall, the suppression effect on the photodegradation rate with the introduction of four scavengers followed the order of t-BuOH > BZQ > SO > FS-EDTA. Specifically, the photodegradation rate of RhB over ZSC 1.5 was only suppressed to some extent upon exposure to BZQ or t-BuOH, suggesting that the contribution of •O 2 − and •OH was subordinate in the decomposition process. By contrast, the addition of SO and FS-EDTA led to an obviously significant decline in the photodegradation rate, implying that the active species of h + and H 2 O 2 could make a predominant contribution to photocatalytic oxidation reaction of RhB. variation of photodegradation rate upon exposure to different scavengers is presented in Figure 9.
Overall, the suppression effect on the photodegradation rate with the introduction of four scavengers followed the order of t-BuOH > BZQ > SO > FS-EDTA. Specifically, the photodegradation rate of RhB over ZSC 1.5 was only suppressed to some extent upon exposure to BZQ or t-BuOH, suggesting that the contribution of •O2 − and •OH was subordinate in the decomposition process. By contrast, the addition of SO and FS-EDTA led to an obviously significant decline in the photodegradation rate, implying that the active species of h + and H2O2 could make a predominant contribution to photocatalytic oxidation reaction of RhB. According to previous studies [15,21], the conduction band (CB) potentials for pure Cu2O, Zn2SnO4, and SnO2 are determined to be approximately −1.04 eV, −0.22 eV and 0.05 eV, respectively; while the valence band (VB) potentials for pure Cu2O, Zn2SnO4, and SnO2 are approximately 0.87 eV, 3.18 V and 3.25 eV, respectively. A proposed mechanism for the enormously promoted photocatalytic behavior is brought forth on the basis of energy-band structure combining with experimental results, as portrayed in Figure 10. In principle, pure Zn2SnO4 and SnO2 belong to the group of wide bandgap semi-conducted photocatalysts, and are incapable of being excited to generate electron-hole pairs under visible-light irradiation. As is revealed by aforementioned results of XPS and PL measurements, some defects, such as oxygen vacancy, existed in ZSC hybrid photocatalyst. As a result, shallow energy levels could be simultaneously produced in Zn2SnO4 and SnO2 photocatalysts [14,15,48]. Upon exposure to visible light illumination, the electrons from show energy levels of Zn2SnO4 and SnO2 are able to be excited and transfer to the CB positions of Zn2SnO4 According to previous studies [15,21], the conduction band (CB) potentials for pure Cu 2 O, Zn 2 SnO 4 , and SnO 2 are determined to be approximately −1.04 eV, −0.22 eV and 0.05 eV, respectively; while the valence band (VB) potentials for pure Cu 2 O, Zn 2 SnO 4 , and SnO 2 are approximately 0.87 eV, 3.18 V and 3.25 eV, respectively. A proposed mechanism for the enormously promoted photocatalytic behavior is brought forth on the basis of energy-band structure combining with experimental results, as portrayed in Figure 10. In principle, pure Zn 2 SnO 4 and SnO 2 belong to the group of wide bandgap semi-conducted photocatalysts, and are incapable of being excited to generate electron-hole pairs under visible-light irradiation. As is revealed by aforementioned results of XPS and PL measurements, some defects, such as oxygen vacancy, existed in ZSC hybrid photocatalyst. As a result, shallow energy levels could be simultaneously produced in Zn 2 SnO 4 and SnO 2 photocatalysts [14,15,48]. Upon exposure to visible light illumination, the electrons from show energy levels of Zn 2 SnO 4 and SnO 2 are able to be excited and transfer to the CB positions of Zn 2 SnO 4 and SnO 2 , respectively. As for narrow-bandgap Cu 2 O, the electrons from the VB position are easily excited, and then accumulate at the CB position through migration. Based on the energy band structure analysis above, a typical type II heterostructure is successfully established in the ternary component photocatalyst of Cu 2 O, Zn 2 SnO 4 and SnO 2 . Assisted by the built-in electric field, the photoinduced electrons on the CB of Cu 2 O could flow to that of Zn 2 SnO 4 , eventually arrive, and accumulate at the CB position of SnO 2 . Simultaneously, the photoinduced holes from VB of SnO 2 are available to accumulate at the VB position of Cu 2 O against the direction of the electron flow. Therefore, the efficient separation of photogenerated electron-hole pairs is successfully achieved in the ZSC hybrid system composed of Cu 2 O, Zn 2 SnO 4 , and SnO 2 components with a well-matched band structure. It is worthwhile to note that the CB potential of SnO 2 (0.22 eV) is negative enough to enable electrons to react with O 2 and to generate the species of H 2 O 2 (O 2 /H 2 O 2 , 0.695 eV) [49][50][51], significantly playing a positive role in the photocatalytic oxidation process of RhB molecules. Although the VB potential of Cu 2 O (0.87 eV) is not oxidative enough to convert the radicals of OHinto the active species of OH (OH − /•OH, 1.99 eV) [52][53][54][55], the accumulated holes on the VB of Cu 2 O could straightforwardly oxidize organic pollutant molecular to CO 2 and H 2 O. That is why the active species of h + and H 2 O 2 exerted a predominant contribution to the oxidization process of organic pollutant molecular.

Conclusions
In summary, Zn2SnO4/SnO2/Cu2O hybrid photocatalysts were successfully achieved via a novel "directing growth-etching treatment-coupling composites" route based on structural modulation and band alignment modification. The co-modification of the etching treatment and Cu2O decoration endowed ZSC hybrid photocatalysts with a larger specific surface area, better light absorption ability, and more efficient spatial separation of photogenerated charge carriers. Thus, the resultant ZSC hybrid photocatalyst exhibited a continuously improved photocatalytic performance towards the photodegradation of RhB, as compared with the P-ZSO/SnO2 sample. Especially, the k value of the ZSC1.5 photocatalyst reaches up to 0.07 min −1 , which was, respectively, about seven times higher than the corresponding counterparts for the P-ZSO/SnO2 sample. Such a simply and rationally designed route can provide an idea for the synthesis of highly efficient visible-light responsive photocatalysts applied in the territory of solar energy utilization and conversion.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Figure S1: SEM image Photocatalytic degradation curve of Zn2SnO4/SnO2 microspheres

Conclusions
In summary, Zn 2 SnO 4 /SnO 2 /Cu 2 O hybrid photocatalysts were successfully achieved via a novel "directing growth-etching treatment-coupling composites" route based on structural modulation and band alignment modification. The co-modification of the etching treatment and Cu 2 O decoration endowed ZSC hybrid photocatalysts with a larger specific surface area, better light absorption ability, and more efficient spatial separation of photogenerated charge carriers. Thus, the resultant ZSC hybrid photocatalyst exhibited a continuously improved photocatalytic performance towards the photodegradation of RhB, as compared with the P-ZSO/SnO 2 sample. Especially, the k value of the ZSC1.5 photocatalyst reaches up to 0.07 min −1 , which was, respectively, about seven times higher than the corresponding counterparts for the P-ZSO/SnO 2 sample. Such a simply and rationally designed route can provide an idea for the synthesis of highly efficient visible-light responsive photocatalysts applied in the territory of solar energy utilization and conversion.
Author Contributions: T.J. conceived and designed the experiments; J.A., F.F. and K.W. carried out the synthetic experiment and photo-catalytic performance of the as-prepared samples; D.Y. and J.L. measured the as-prepared samples; W.W. provided precise instruction and analyzed the data. T.J. wrote the paper.
Funding: The work was sponsored by Natural Science Foundation of Henan Province (182300410252, 182102410074, 172102410047), and Education Department of Henan Province for the fund support (18A 430020), and the National Natural Science Foundation of China (U1304520, 51702148), and Luoyang Institute of science and Technology (2018YZ06).

Conflicts of Interest:
The authors declare no conflict of interest.