Introduction

Currently, photocatalysis applications for addressing environmental issues such as environment pollution and energy crises have attracted more and more attention and gradually become a research hotspot1,2,3,4. In most case, non-toxic, chemically stable, controllability of redox power through materials engineering, and the capable of retrieval and extended use without substantial loss of activity are often recognized as the rubric of semiconductor photocatalysis5,6,7. However, conventional homogeneous photocatalysis have inherent drawbacks such as the easy recombination of photo-induced electron-hole (e−/h+) pairs and the absorption of light only at ultraviolet region (λ < 400 nm)8. Developing heterogeneous photocatalysis has been an effective strategy to enlarge the range of wavelengths of light absorption and promote the separation of the charge-carrier9,10. Therefore, the heterogeneous photocatalysis often shows more appealing than its homogeneous counterpart.

Among the multi-heterogeneous systems that have been developed, TiO2-based photocatalysts have triggered considerable interest due to their unique photocatalytic activity and good chemical stability11. Heterostructures of TiO2 and other oxides, such as ZnO12, SnO213,14,15, and Fe2O316 to form the semiconductor coupling is believed to overcome the facile recombination of e−/h+ pairs. In the suitably assembled semiconductors, the efficient charge transfer can be occurred, ultimately leading to the spatially separation of the charge-carrier. In particular, the SnO2/TiO2 system with high photocatalytic activity has attracted extensive interest. On one hand, SnO2 possessed a high electron mobility (~100–200 cm2V−1s−1)17, which gives rise to a faster transport of photoinduced electrons. On the other hand, the conduction band (CB) edge of SnO2 is more positive than that of TiO218, which may lead to the transformation of photoexcited electrons from the CB of TiO2 to that of SnO2, and the opposite transformation direction for holes in the heterojunction between TiO2 and SnO219. By this way, the recombination of charge carriers can be greatly suppressed, thereby resulting in an enhanced photocatalytic performance.

Additionally, in regard to the weak visible light response, decorating TiO2 with noble metal nanoparticles(NPs) to construct the noble metal/TiO2 composite was suggested to be an effective method to extend the photoresponse of TiO2 to visible light region due to the localized surface plasmon resonance (SPR) for metallic nanoparticles20. In these heterogeneous systems, the noble metal NPs can be excited by visible light in the ways that the oscillating electric field of the light interacts with the conduction electrons21. As a result, a strong oscillation of these electrons appears when the incident photon frequency is comparable to the collective oscillation of the conduction electrons22. Still now, there are several methods have been reported for incorporation of noble metal NPs into TiO2, such as adsorption of preformed noble metal colloids23 and photo-deposition24. But unfortunately, all of these strategies are suffered from the weak interaction between TiO2 and noble metal NPs, and also the problem of inhomogeneity distribution of noble metal NPs25. As is well known, the strong linkage of noble metal NPs to TiO2 may enhance the electron transfer between noble metal NPs and TiO2 and can also prevent the metal NPs to leach from TiO2 surface26. Moreover, the surface homogeneity distribution of small noble metal NPs might increase the density of metal/TiO2 interface. All of these enhance the photocatalytic activity of the composites comprised with the pure TiO2. In these regards, it is necessary to develop a facile strategy to prepare the composite photocatalyst that coupled with noble metal NPs, SnO2 and TiO2.

Recently, our group27 reported a convenient Sn2+ reduction method to prepare a ternary heterostructure TiO2/SnOx-Au photocatalyst and the photocatalyst exhibited an enhanced visible photocatalytic performance as compared with TiO2/SnO2 and TiO2/Au binary composites. In this catalytic system, the Au nanoparticles were connected with SnOx surface directly, which may weaken the interaction with noble metal NPs and TiO2 matrix inevitably. In this regard, it may be more meaningful to construct a ternary heterostructure in which noble metal NPs were coupled with both TiO2 matrix and tin oxides accelerant.

In the present work, we have constructed a ternary heterostructure TiO2/Ag/SnO2 photocatalyst with Ag nanoparticles by SnO2, by which we want to maximize the potential of Ag NPs for extending the visible-light absorption and the SnO2 species for inhibiting the recombination rate of photo-generated h+/e− pairs. The preparation route was shown in Fig. 1. The Ag NPs in the ternary composites exhibited a quite uniform distribution with the particles size could be facilely tuned from 2.04 nm to 3.94 nm. The ternary TiO2/Ag/SnO2 composites offered an enhanced catalytic activity for degradation of methylene blue under the visible light irradiation as compared with the single TiO2 and the binary hybrid materials (TiO2/Ag and TiO2/SnO2).

Figure 1
figure 1

Methodology used for the preparation of the TiO2/Ag/SnO2 photocatalysts.

Full size image

Experiment

Synthesis of TiO2/Ag/SnO2 composites

In a typical experiment, 1.0 g of P25 TiO2 was dispersed in deionized water (100 mL) by ultrasonic processing. Then a mixture aqueous solution containing SnCl2 (0.5 g) and hydrochloric acid (3.0 mL) were added into the above solution, which was allowed for stirring at room temperature for 12 h. The precipitate was collected by centrifugation, followed by washing with water and redispersed into 70 mL water. Subsequently, 0.5 mL of AgNO3 solution (50 mM) was added. After reaction for 30 min, 2 mL of 0.15 M sodium formate solution was added. The mixture was allowed to stir for another 4 h, and then the product was collected by centrifugation, washing with water and drying in a vacuum oven at 75 °C. The obtained samples were labeled as TiO2/Ag/SnO2(x wt%), where the x denoted the nominal content of Ag NPs in the ternary composites.

Characterization

Transmission electron microscopy (TEM) experiments were conducted on a JEM-1230 microscope operated at 100 kV. The samples for the TEM measurements were suspended in ethanol and supported onto a Cu grid. The powder X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Advance Diffractometer (Germany) with Cu Kα radiation (λ = 1.5406 Å). X-ray photoelectron spectra (XPS) measurements were carried out in a Thermo ESCALAB 250 instruments (USA) using non-monochromatic Al Kα1486.6 radiation. The nitrogen adsorption and desorption isotherms were measured at −196 °C on an ASAP 2020 (Micromertics USA).The specific surface area was determined from the linear part of the BET equation (P/P0 = 0.05–0.25). The pore size distribution was derived from the desorption branch of the N2 isotherm using the Barrett-Joyner-Halenda (BJH) method. UV-vis spectra were recorded on a Shimadzu UV 3600 spectrometer.

Evaluation of photocatalytic performance

40 mg TiO2/Ag/SnO2 composites were added into 100 mL of 3.12 × 10−5 mg/L methylene blue (MB) solution. A 500 W xenon lamp, the main wavelength of which lies in the 365–720 nm range, was used as the visible light source. Before irradiation, the solution was stirred for 30 min in dark in order to achieve absorption-desorption equilibrium. Then, the aforementioned mixture solution was irradiated in a photochemical chamber under continuous stirring with reflux water to keep the temperature constant. At certain time intervals, 2 mL solution was drawn out and centrifuged to obtain clear liquid. The quantitative determination of MB was performed by measuring the intensity of its absorption peak with a UV-vis spectrophotometer.

Results and Discussion

The TiO2/Ag/SnO2 ternary composites were prepared followed by the strategy containing a facile one step reduction approach by using stannous chloride as both SnO2 precursor and reducing agent. The commercial Degussa P25 TiO2 composed with 80% anatase and 20% rutile was used as the support due to its good application prospect28. In the first step, the TiO2 particles were activated with Sn2+ by the inorganic grafting between Sn2+ and surface hydroxyl groups on TiO2 particles29. Secondly, AgNO3 solution was added into the TiO2/Sn2+ species. It is well known that the isoelectric point of Degussa P25 TiO2 is about 6.230. Therefore, the surface of TiO2 is possessed of negative charge in the neutral environment. Except the part of the TiO2 surface being neutralized by Sn2+, the residual position with negative charge can be served as the adsorption site for self-assembly of Ag+. Since the standard reduction potential of the Sn4+/Sn2+ (0.151 V vs. redox pair the standard hydrogen electrode, SHE) is lower than that of Ag+/Ag(0.80 V vs. SHE), the deposited Ag+ can be easily reduced to Ag NPs at ambient temperature by the surrounding Sn2+ species, with the Sn2+ species being oxidized to SnO2. As a result, Ag NPs were incorporated onto TiO2 surface and isolated by the SnO2 species.

Figure 2(a) showed the TEM image of TiO2/Ag/SnO2 photocatalyst with the nominal Ag content was 1 wt%. It was obviously that the Ag NPs with the average particle size of about 2.04 nm were well-distributed on the TiO2 surface (with the size range of Ag NPs is 1.1 nm~2.9 nm). We attributed it to the evenly distribution of Sn2+ species on TiO2 that gives rise to the uniform adsorption sites for Ag+. As a result, the Ag NPs were homogeneously distributed in the ternary composites. Interestingly, the particles size of the Ag NPs on the support could be tuned by facilely changing the dosage of AgNO3 (Fig. 2(b,c)).When the nominal dosage of AgNO3 was 2 wt%, the average size of the formed Ag NPs increased to ~2.66 nm (with the size range of Ag NPs is 1.5 nm~3.6 nm). Further increase the nominal concentration of Ag to 5 wt%, the Ag NPs possessed a larger particles size (~3.94 nm, with the size range of Ag NPs is 2.0 nm~6.4 nm), suggesting that the particle size of the incorporated Ag NPs on TiO2 surface could be easily controlled. It should be mentioned that the SPR effect is strongly related to the content and particle size of noble metal NPs31, therefore it seems that the convenient adjustment of Ag NPs is crucial for optimizing the photocatalysis. But beyond that, it should be mentioned that Ag NPs in all the heterostructure composites exhibited uniform distribution even with the growth of Ag NPs. Figure 2(d) displayed the typical HRTEM images of the TiO2/Ag/SnO2(2 wt%) photocatlaysts. The image indicated that the ternary composites were composed of Ag NPs located on TiO2 matrix and immobilized by the surrounded SnO2.

Figure 2
figure 2

TEM images of as-prepared TiO2/Ag/SnO2 photocatlaysts with different Ag contents: (a) 1 wt%, (b) 2 wt% and (c) 5 wt%. (d) HRTEM images of TiO2/Ag/SnO2(2 wt%) photocatlaysts.

Full size image

Figure 3 displayed the energy dispersive X-ray (EDX) spectroscopy of TiO2/SnO2/Ag photocatlaysts. The result showed that the content of Ti, O and Sn in all samples is almost the same, suggesting that the incorporation of Ag NPs in the photocatalysts have little influence on the content of TiO2 and Sn species. Additionally, the Ag content in TiO2/SnO2/Ag samples increased with the increased dosage of AgNO3, demonstrating that the concentration of Ag NPs can be easily tuned in our experiment. Figure 3 also displayed the HAADF-STEM of the TiO2/Ag/SnO2. Though the images appear blurry, it might be concluded that Ag nanoparticles were tightly covered by Sn species from the red frames in the mapping images of Ag(Ag-L) and Sn(Sn-L), which is bright in Ag images but dark in Sn images.

Figure 3
figure 3

TOP: EDX analysis of (a) TiO2/Ag/SnO2(1 wt%) and (b) TiO2/Ag/SnO2(5 wt%). (c) A representative EDS-Mapping of TiO2/Ag/SnO2(5 wt%). Down: HAADF-STEM image of the as-obtained TiO2/Ag/SnO2(2 wt%).

Full size image

XRD patterns of the TiO2/Ag/SnO2 samples were shown in Fig. 4. All the samples exhibited the mixed crystalline phase containing anatase and rutile, which is on account of the P25 TiO2 that used as the support in our experiment containing anatase and rutile phases32. The result also suggested that the loading of SnO2 or Ag had minor influence on the crystalline phase of original TiO2. No obvious Ag and Sn species peaks could be seen in the composites that incorporated 1 wt% and 2 wt% Ag, which might be attributed to the relatively low crystallinity of Sn species and the quite small Ag NPs, respectively33. As increasing the size of the supported Ag NPs to 3.94 nm(TiO2/Ag/SnO2(5 wt%)), the diffraction peaks that could be assigned to the Ag NPs with face-centered cubic structures (fcc) emerged.

Figure 4
figure 4

XRD patterns for (a)TiO2/SnO2, (b) TiO2/Ag/SnO2(1 wt%), (c) TiO2/Ag/SnO2(2 wt%) and (d) TiO2/Ag/SnO2(5 wt%).

Full size image

The chemical and bonding environments of the ternary composites were ascertained by XPS measurements. As shown in Fig. 5, the fully scanned spectra revealed that the presence of Ti, O, Sn and Ag in all the ternary TiO2/Ag/SnO2 composites. The high-resolution spectrum of the Ag 3d region in the TiO2/Ag/SnO2(5 wt%) displayed two peaks corresponding to metallic Ag at binding energies of 367.8 eV and 373.8 eV in Ag 3d3/2 and Ag 3d5/2 levels, with the splitting of the 3d doublet is 6.0 eV, revealing the complete reduction of Ag+ in the experiment34. It was obvious that the peaks of Ag 3d shifted to the lower position as compared to these of bulk Ag (368.3 eV for Ag 3d3/2 and 374.3 eV for Ag 3d5/2)35, indicating the increase in electrons density of Ag species and also revealing the strong interaction between Ag and the semiconductors36. Moreover, the binding energies of Sn 3d5/2 and Sn 3d3/2 are 486.8 and 495.1 eV, respectively, indicating that tin, in all samples, is in the +4 oxidation state37,38. In the experiment, Sn2+ was completely oxidized by the excess Ag+, and the formed Sn4+ species could not be easily reduced by sodium formate39,40. As a consequence, it is reasonably to calculate that the Sn species in TiO2/Ag/SnO2 ternary heterostructures are dominated by the SnO2. The Ti 2p spectrum in TiO2/Ag/SnO2(1 wt%) can be ascribed to Ti 2p3/2 and Ti 2p1/2 that centered at binding energies of 464.3 and 458.6 eV correspondingly. The splitting of the binding energies was ~5.7 eV, which indicated that the typical Ti4+ in the composite sample41,42. The Ti 2p spectrum do not vary with the increase the content of Ag NPs, suggesting that the incorporation of higher quantity of Ag have few influence on the chemical environments of TiO2.We attributed it to the lager amount of TiO2 species in the composites that could offer abundant electrons without the alteration of the chemical state of TiO2 particles.

Figure 5
figure 5

XPS patterns of (a) TiO2/Ag/SnO2(1 wt%), (b) TiO2/Ag/SnO2(2 wt%) and (c) TiO2/Ag/SnO2(5 wt%).

Full size image

The nitrogen adsorption-desorption isotherm and corresponding size distribution of the as-prepared TiO2/Ag/SnO2 samples were shown in Fig. 6. It was obviously that all the TiO2/Ag/SnO2 samples exhibited the typical porous characteristics, as evidenced by the significant type IIb curves in the N2 absorption-desorption isotherm of TiO2/Ag/SnO2 sample43. This pore should be ascribed to the space among the stacking of the grain in the TiO2 matrix or the SnO2 shells. Additionally, the feature of the isotherm was not be changed as the alteration of Ag content, indicating that Ag NPs were mainly deposited on the surface of TiO2 instead of filling into the pores of TiO2 matrix. The same result could also be verified from the pore size distribution of the sample in Fig. 6(ii), which showed the similar pore size distribution of the two samples, with the average pore size being ~40 nm.

Figure 6
figure 6

N2 adsorption-desorption isotherm and corresponding pore size distribution curve (inset) of (a) TiO2/Ag/SnO2(1 wt%), and (b) TiO2/Ag/SnO2(5 wt%).

Full size image

The optical properties of the as prepared samples were characterized by UV-vis DRS, and the results were shown in Fig. 7. Compared with the unmodified TiO2 nanoparticles, the absorption measurements of the TiO2/Ag/SnO2 sample exhibited enhanced photoabsorption in the range of 400–650 nm, which can be attributed to the light-harvesting enhancements by the surface plasmon resonance of Ag NPs27. Moreover, the absorption edge of the TiO2/Ag/SnO2 composite extended an unambiguous red-shift compared to the TiO2 and TiO2/SnO2, which reflected that the electronic structure and optical properties of the TiO2 and TiO2/SnO2 have been modified by the incorporated Ag species. The band-gap energy of the samples can also be determined from the plot of (ahν)1/2 versus hν (Fig. 7(ii))44. The optical band gap energy of the TiO2/Ag/SnO2(1 wt%) composite was measured to be 2.7 eV, which was relatively lower than that of TiO2(3.2 eV) and TiO2/SnO2(3.0 eV). These result also confirmed that the incorporation of Ag and SnO2 caused a red shift of the UV-Vis absorption spectrum and narrowed the optical band gap energy of TiO2. Moreover, the TiO2/Ag/SnO2(2 wt%) composite exhibited a higher absorption and an even narrower band gap energy (2.3 eV) than the TiO2/Ag/SnO2(1 wt%) composite, possibly suggesting that the enhanced electron transfer existed among the components of TiO2/Ag/SnO2(2 wt%). Benefiting from such broad light absorption width and high absorption intensity in the visible region, the ternary composite photocatalyst was expected to improve the solar-energy utilization efficiency and perhaps possessed an enhanced visible-light-driven photocatalytic performance.

Figure 7
figure 7

(i) UV-vis diffuse reflectance spectra and (ii) the corresponding Tauc plot of (a) original TiO2, (b) TiO2/SnO2, (c) TiO2/Ag/SnO2(1 wt%) and (d) TiO2/Ag/SnO2(2 wt%).

Full size image

Visble-light photocatalytic activity

The photocatalytic performances of the as-prepared TiO2/Ag/SnO2 composites were evaluated by monitoring its characteristic absorption band at 650 nm to measure the degradation rate of MB under visible light irradiation. Before irradiation, the reaction mixtures were stirred in dark for 30 min to ensure that the adsorption-desorption equilibrium of MB was established. As shown in Fig. 8(i), the ternary TiO2/Ag/SnO2 heterostructure exhibited a slightly increase adsorption efficiencies for MB (~10%) as compared with the commercial TiO2 and binary heterostructure(TiO2/Ag, and TiO2/Sn2+), demonstrating that the incorporation of Ag and Sn species can increase the adsorption of MB.

Figure 8
figure 8

Photocatalytic activities (i) and kinetics (ii) of the different catalysts. (a) No catalyst, (b) TiO2, (c) TiO2/SnO2, (d) TiO2/Ag, (e) TiO2/Ag/SnO2(1 wt%), (f) TiO2/Ag/SnO2(2 wt%), and (g) TiO2/Ag/SnO2(5 wt%) for degradation of MB under visible light irradiation.

Full size image

The photodegradation of MB does not occur without the presence of photocatalysts, as evidenced by little change in the absorption peak after visible light irradiated for 2 h. After addition of a trace amount of photocatalyst into the solution, the adsorption peak significant decreased, suggesting the degradation of the organic pollutant proceeded. Figure 8(i) showed plots of Ct/C0 in the degradation of MB, where C0 is initial concentration of MB and Ct is the concentrations of the MB at time t, respectively. It is undisputed that the degradation rate of MB followed the order of TiO2/Ag/SnO2 > TiO2/Ag > TiO2/SnO2 > TiO2 after the same irradiation time. The result indicated that the degradation efficiency of MB could be improved in the presence of Ag or SnO2 modified TiO2 photocatalytic systems as compared to pure TiO2, particularly remarkably enhanced with the Ag/SnO2 co-decorated TiO2 photocatalyst. Furthermore, the photocatalytic degradation of the organic pollutant could be regarded as a pseudo-first-order kinetics reaction to evaluate the degradation rate. The linear relationships between ln(Ct/C0) and reaction time using the samples were shown in Fig. 8(ii), and the plots of all the samples were well matched the first-order reaction kinetics. The calculated rate constant k for TiO2, TiO2/SnO2,TiO2/Ag and TiO2/Ag/SnO2(2 wt%) samples were 0.002 min−1, 0.004 min−1, 0.008 min−1 and 0.019 min−1, respectively. It is apparent that TiO2/Ag/SnO2(2 wt%) exhibited the best degradation efficiency among the above four samples, giving a 9.5 times higher rate constant of MB degradation than the commercialize TiO2.

As discussed before, the excellent photocatalytic performance for the TiO2/Ag/SnO2(2 wt%) sample, on one hand, should be largely attributed to the SPR effect of Ag NPs induced broadband optical absorption enhancement. The porous characteristic of the ternary composites may also promote the connection between the embedded Ag NPs and the external environment, leading to a strong SPR effect of Ag NPs. On the other hand, it could be ascribable to the SnO2 species may serve as an electron tank to accept the photogenerated electrons and facilitate charge carriers separation45. Interesting, as compared the rate constant with that of TiO2/SnOx/Au reported previously(0.014 min−1)27, it could be found that the k values of the samples(0.019 min−1 for TiO2/Ag/SnO2(2 wt%)) even exhibited an increased photocatalytic activity. In the previous report, it was suggested that the electron trapping capability of Au was excellent than that of Ag upon the higher electron affinity of Au NPs, which inevitably give rise to a higher photocatalytic activity for the Au modified photocatalysts46. In this work, the abnormal higher activity of TiO2/Ag/SnO2 can be attributed to the intimate connection among the Ag, SnO2 and TiO2. In our experiment, the Ag NPs were initially absorbed on TiO2 surface, and reduced by the surrounding Sn2+ species, rustling in Ag NPs directly located on the TiO2, and anchored by the SnO2 species. This structure may allow the maximum improvement level of interaction between each component of the photocatalyst, resulting in an enhanced photocatalytic performance. While for the TiO2/SnOx/Au, the Au NPs were located on the SnOx surface, which may weaken the interaction between noble metal NPs and TiO2 matrix. Indeed, comparing the rate constant with that of Au-based and Ag-based photocatalysts reported previously listed in Table 1, it can be found that the activity of TiO2/Ag/SnO2(2 wt%) in our experiment is much higher than that of most reports for the photodegradation of MB.

Table 1 Comparison of rate constant for the photodegration of MB using catalysts containing Ag nanoparticles.
Full size table

Besides, one can see that the TiO2/Ag/SnO2 heterostructure catalysts with Ag content 2% revealed the highest photocatalytic activity. When the Ag content is relatively low(<2%), the photodegration of MB enhanced gradually with increase of the Ag content, which may be attributed to the increase of the electron transfer interface both of Ag-TiO2 and Ag-SnO2. However, when the Ag content exceeded 2%, an opposite phenomenon was observed by further increase the content of Ag. The lower photocatalytic performance should be ascribed to the re-combined electrons and holes upon the excess content of Ag NPs.

Because the practical application of photocatalyst requires its renewable, we carried out four cycling tests to degrade MB using the as-prepared TiO2/Ag/SnO2(2 wt%) photocatalyst to investigate its stability, as shown in Fig. 9a. It is observed that the TiO2/Ag/SnO2(2 wt%) sample only exhibited a slight decline for photocatalytic decomposition of MB shows after four cycles(9.1%). Almost no changes can be found form the XRD pattern of catalyst after reaction (Fig. 9b), demonstrating the excellent stability of the catalyst.

Figure 9
figure 9

(a) Conversion of 4-NP in 4 successive cycles of reduction using TiO2/Ag/SnO2(2 wt%) as photocatalyst, (b) XRD pattern of the TiO2/Ag/SnO2(2 wt%) catalyst after reaction, (c) Effects of a series of scavengers on the degradation efficiency of MB by TiO2/Ag/SnO2(2 wt%) photocatalyst.

Full size image

Based on the above experimental results and discussion, a possible mechanism for degradation of MB by visible-light-driven TiO2/Ag/SnO2 photocatalyst was proposed and presented in Fig. 10. Under visible light irradiation, only the Ag SPR can be excited, resulting in the emerging of electrons in the CB while the holes (h+) remain in the VB of Ag NPs47. These photo-induced electrons can get sufficient energy to surmount the Schottky barrier on Ag/TiO2 and Ag/SnO2 and move into the CB of semiconductor (TiO2 or SnO2) through their tightly-coupled interfaces, where the electrons can be consequently trapped by the adsorbed molecular oxygen and produce superoxide radicals(▪O2−)48. At the same time, the h+ on the VB of Ag in turn accept electrons from water or the dye molecules adsorbed on the surface of catalysts, resulting hydroxyl radicals(▪OH) generating in the surface of Ag NPs49. These reactive oxygen species and h+ are potent oxidizing agents for the degradation of methylene blue molecules. In order to distinguish the roles of the active species, we have taken the trapping experiment with scavenger investigation. Three reagents, t-BuOH, p-benzoquinone and EDTA, were used as the scavengers of •OH, •O2− and h+, respectively. Figure 9c showed the degradation rates of MB by TiO2/Ag/SnO2(2 wt%) in the conditions of adding scavengers. When t-BuOH (5 mM) was added into reaction solution, the photocatalytic efficiency was slightly reduced to 85.3%. However, the photocatalytic efficiency exhibited a significant decrease with the addition of p-benzoquinone (5 mM) or EDTA (5 mM), and the degradation rates of MB were reduced to 42.9% and 21.6%, respectively. Obviously, the results suggested that •O2− and h+ are the main active species in the current photocatalytic system. The unique structure of TiO2/Ag/SnO2 with Ag NPs tightly immobilized on TiO2 by SnO2 species may promote the interactions between Ag and the semiconductor (TiO2 and SnO2), which can accelerate the separation of photo-induced holes and electrons. As a result, the holes and electrons can be entirely involved in the photocatalytic reactions, and a strong photocatalytic activity is expected.

Figure 10
figure 10

Proposed photocatalytic mechanism for degradation of MB by TiO2/Ag/SnO2 nanocomposites under visible light irradiation.

Full size image

Conclusions

In summary, the ternary TiO2/Ag/SnO2 heterostructure composites were successfully prepared by a facile one-step reduction approach using SnCl2 as both SnO2 precursor and reducing agent. The Ag+ was first adsorbed on TiO2 surface and then reduced by the surrounding Sn2+ species, resulting in the formation of the TiO2/Ag/SnO2 composite. The obtained TiO2/Ag/SnO2 heterostructure exhibited enhanced photocatalytic activity toward MB degradation under visible light irradiation as compared to individual TiO2 or the binary composite (TiO2/Ag or TiO2/SnO2). The significantly improved photocatalytic activity should be attributed to the SPR effect of Ag NPs and also the fast charge separation by the formation of tightly connected interface of TiO2/Ag and SnO2/Ag. Moreover, the TiO2/Ag/SnO2 composite exhibited good visible-light photocatalytic stability and reusability.