The optimal way to deal with the organic pollutants in wastewater is an important issue for environmental protection. Semiconductor photocatalytic technology displays promising applications in the decomposition of organic compounds in polluted water ^[1-4]. The photocatalyst TiO₂ has been widely used in the degradation of various pollutants. However, the wide band gap energy (~3.2 eV) of TiO₂ leads to its low application efficiency of solar light because TiO₂ can only absorb the small ultraviolet light fraction of solar light (about 2%-3%) ^[5-8]. Therefore, many studies have focused on developing visible light response semiconductor photocatalysts with a narrow band gap ^[9-11].

Among the numerous narrow band gap photocatalysts, although pure Ag₂S displays poor photocatalytic performance in the degradation of pollutants, as a narrow band gap and stable direct semiconductor, Ag₂S exhibits excellent performance in accelerating the electron transfer ^[12]. The formation of a heterojunction can highlight this effect ^[13-16]. For example, the heterojunctions of Ag₂Mo₃O₁₀-Ag₂S ^[14], TiO₂-Ag₂S ^[18], ZnS-Ag₂S ^[19] and NiO-Ag₂S ^[20] display superior photocatalytic performance. As a novel and stable semiconductor material, Ag₂WO₄ has been widely used in catalysis, sensing, and antibacterial and photoluminescence (PL) applications ^[21-24]. However, as a photocatalyst, Ag₂WO₄ can only absorb UV light owing to its large band gap (~3.5 eV). Additionally, Ag₂WO₄ exhibits poor stability because photocorrosion readily takes place over Ag-based semiconductors with light illumination, if no sacrificial reagent is present ^[25].

Selecting a suitable synthetic route is the key to a successful photocatalyst because the textural properties and photocatalytic performance are determined by how they are prepared. A nanostructure/microstructure is highly desirable in photocatalyst synthesis. Recently, the application of high-intensity ultrasound or sonochemistry has provided a green and effective synthetic strategy for nanostructured photocatalyst that is often unavailable with the traditional methods. In the sonochemical synthesis, the acoustic cavitation effect is able to generate extraordinary physicochemical conditions, for example, high local temperatures (5000 ℃), high pressures (100 MPa), and rapid cooling and heating rates (＞10¹⁰ ℃/s). These unusual physicochemical conditions could greatly benefit the fabrication of photocatalysts with novel nanostructures ^[26-30].

In the present work, Ag₂S/Ag₂WO₄ composite microrods were fabricated by a facile sonochemical route and exhibited excellent photocatalytic activity and stability.

All chemicals were of analytical grade obtained from the Shanghai Chemical Company and were used without further purification. For the sonochemical fabrication of Ag₂S/Ag₂WO₄ (denoted as U-Ag₂S/Ag₂WO₄), 10 mmol AgNO₃ was dissolved in 20 mL of deionized (DI) water under stirring, to obtain solution A. Next, 2.5 mmol Na₂S was dissolved in 20 mL of DI water, and the resulting solution was added dropwise to solution A. After stirring for 3 h, 20 mL Na₂WO₄ solution (2.5 mmol) was added dropwise to the Ag₂S liquid suspension. Then, a VOSHIN 10-501D ultrasonic generator connected by a Φ10 luffing rod (selected ultrasonic power was 10%, each ultrasonication was 5 s, with a rest period of 2 s) was used to irradiate the mixture for 1 h. After the irradiation, the suspensions were further stirred for 3 h. The resulting powder was collected by filtration, and washed with DI water and ethanol for several times. Finally, U-Ag₂S/Ag₂WO₄ products were obtained after drying at 70 ℃ for 4 h. U-Ag₂S and U-Ag₂WO₄ were fabricated by the same method.

To evaluate the effects of ultrasonic irradiation, the corresponding Ag₂S and Ag₂WO₄ samples were fabricated by a common precipitation method, and Ag₂S/Ag₂WO₄ composites were fabricated by the successive precipitation method. For example, for the synthesis of the Ag₂S/Ag₂WO₄ composite, 10 mmol AgNO₃ was dissolved in 20 mL of DI water under stirring, to obtain solution A. Na₂S (2.5 mmol) was dissolved in 20 mL of DI water, and the resulting solution was added dropwise to solution A. After stirring for 3 h, 20 mL Na₂WO₄ solution (2.5 mmol Na₂WO₄) was added dropwise to the Ag₂S liquid suspension and was further stirred for 3 h. The resulting powder was collected by filtration and washed several times with DI water and ethanol. Finally, Ag₂S/Ag₂WO₄ products were obtained after drying at 70 ℃ for 4 h.

Powder X-ray diffraction (XRD) patterns were recorded at a scanning rate of 0.05°/s using a Bruker D8 Advance X-ray diffractometer at 40 kV and 40 mA with a Cu K_α radiation source (λ = 0.15418 nm). The Brunauer-Emmett-Teller (BET) specific surface area (S_BET) was measured on a Micromeritics ASAP2020 nitrogen adsorption apparatus (USA). All the samples were degassed at 120 ℃ for 1 h before nitrogen adsorption measurement. The S_BET was determined by a multipoint BET method using the adsorption data in the relative pressure (P/P₀) range of 0.05-0.3. Scanning electron microscopy (SEM) images were collected using an MLA650F scanning electron microscope and were used to investigate the sample morphologies. Transmission electron microscopy (TEM) and high- resolution TEM (HRTEM) images were collected using a TECNAI G2F20 microscope. Ultraviolet-visible diffuse reflectance spectra (UV-vis DRS) were recorded using a UV-vis spectrophotometer (UV-2550, Shimadzu). PL emission spectra were obtained with excitation at 325 nm using an F-4600 spectrometer. Surface compositions were determined by X-ray photoelectron spectroscopy (XPS) using an ESCALAB 250 XPS system with a monochromatic Al K_α source and a charge neutralizer. Photocurrent measurements were performed on a CHI 660E electrochemical work station (Chenhua Instrument, China) in a conventional three electrode configuration with a Pt foil as the counter electrode and a Ag/AgCl (saturated KCl) as the reference electrode. A 500-W Xe lamp served as a light source. 0.1 mol/L Na₂SO₄ aqueous solution was used as the electrolyte. The working electrodes were prepared as follows: 10 mg of the as-prepared photocatalyst and 0.5 mL Nafion dispersing reagent were added in 5 mL absolute ethanol and son icated for 30 min. The slurry was then spread on a 1.0-cm × 1.0-cm indium-tin oxide (ITO) glass substrate and dried in air. The photoresponse of the samples, as light on and light off, was measured at 0.0 V.

The photocatalytic activities of the as-prepared samples were determined by the degradation of methylene blue (MB) and methyl orange (MO) dyes in aqueous solution. The photocatalytic reaction was carried out in a photochemical reactor (XPA, XuJiang Electromechanical Plant, Nangjing). The light source was a 400-W metal halide lamp. In each test, 30 mg of the as-prepared samples were added into 60 mL of 20 mg/L MB or MO aqueous solution. Then, the suspension was stirred in the dark for 40 min to achieve the adsorption-desorption equilibrium prior to visible light irradiation. During the degradation process, the reaction temperature was maintained at 20 ℃ by the circulation of water. After fixed irradiation intervals, ~2-mL aliquots of suspension were sampled, and the solids were removed by centrifugation. The MB and MO concentration of the clear upper layer was determined by UV-vis spectrophotometry.

The phase composition and crystal properties of the samples were analyzed by XRD. Fig. 1(a) shows the XRD patterns of the as-prepared Ag₂S, Ag₂WO₄ and Ag₂S/Ag₂WO₄ samples. The Ag₂S sample corresponded to the monoclinic phase of Ag₂S (JCPDS 14-0072). The main characteristic peaks of Ag₂WO₄ at 16.7°, 30.2°, 31.5°, 32.9°, 45.2°, 54.5° and 58.1° can be indexed to the (011), (002), (231), (400), (060), (233) and (262) crystal planes of Ag₂WO₄ (JCPDS 70-1719) with orthorhombic phase. The observed main diffraction peaks of Ag₂S/Ag₂WO₄ are the diffraction peaks of Ag₂WO₄. Additionally, a series of weak diffraction peaks at 2θ = 20°-30° and 35°-45° for Ag₂S are observed. Fig. 1(b) shows the XRD patterns of the ultrasonically prepared U-Ag₂S, U-Ag₂WO₄ and U-Ag₂S/Ag₂WO₄ samples. Careful observation shows that ultrasonic irradiation exerts an obvious influence on the crystal structure. For U-Ag₂S, after ultrasonic irradiation, the diffraction peaks at 21.2°, 24.6°, 27.6°, 30.3°, 32.8°, 35.1°, 37.3°, 39.5°, 43.4°, 45.3° and 50.6° can be indexed to the (102), (112), (210), (202), (212), (004), (300), (222), (204), (214) and (400) crystal planes of Ag₈S₄(O₄) (JCPDS 032-1023) with tetragonal phase. The unique physicochemical conditions generated by the acoustic cavitation effect could cause the O to enter the Ag₂S lattice. On the whole, the U-Ag₂S/Ag₂WO₄ maintained the Ag₂WO₄ crystal phase structure. However, the intensity of the characteristic peaks at 30.3°, 32.9° and 45.2° was obviously increased with respect to Ag₂S/Ag₂ WO₄. This could arise from the ultrasonic irradiation resulting in an increase in crystallinity.

The average crystallite size of the as-prepared samples was estimated using the Scherrer equation. The strongest diffraction peaks, ~39.5° for U-Ag₂S and ~31.5° for the other samples, were used in this calculation. Table 1 shows that ultrasonic irradiation obviously decreased the average crystallite size of the samples. The surface area of the fabricated samples was determined by nitrogen physical adsorption and the obtained results are shown in Table 1. Obviously, ultrasonic irradiation has a distinct influence on the surface area. A slight increase in surface area after ultrasonic irradiation was observed, which could arise from the ultrasonic irradiation, improved the dispersion and decreased the crystallite sizes.

The morphologies of the prepared samples were characterized by SEM and the results were shown in Fig. 2. The inset photographs in the corresponding images display the color of the samples. Fig. 2(a) shows that pure Ag₂WO₄ exhibits microrod morphology with lengths of 0.2-1 μm and diameters of 20-30 nm. After ultrasonic irradiation, some cracks were observed in the U-Ag₂WO₄ microrods (Fig. 2(b)). Fig. 2(c) suggests that pure Ag₂S shows no special morphology. The Ag₂S sample was composed of small irregular particles. Compared with Ag₂S, U-Ag₂S (Fig. 2(d)) has a rounder and smoother surface, and the connections among particles become closer. Fig. 2(e) and (f) show the SEM images of Ag₂S/Ag₂WO₄ and U-Ag₂S /Ag₂WO₄, respectively. Over Ag₂S/Ag₂WO₄, many segregated nanoparticles (NPs) were attached on the surface of a microrod. As for U-Ag₂S/Ag₂WO₄, the attached NPs fused into fluffy particles. The inset photographs indicate that the colors of Ag₂WO₄, Ag₂S and Ag₂S/Ag₂WO₄ are white, black and light brown, respectively. Ultrasonic irradiation slightly changed the color of samples. The colors for Ag₂WO₄, Ag₂S and Ag₂S/ Ag₂WO₄ changed to light brown, deep brown and brown, respectively. Ultrasonic irradiation changed the texture and color, which will certainly result in variations in the surface area and light absorption.

Fig. 3 shows the TEM images for the U-Ag₂S/Ag₂WO₄ composite. Fig. 3(a) clearly shows that the Ag₂S NPs were attached on the smooth surface of the Ag₂WO₄ microrod. We selected one of the attached nanoparticles to analyze with high resolution HRTEM, and the result is shown in Fig. 3(b). Three clear lattice spacings of approximately 0.297, 0.255 and 0.227 nm correspond to the (202), (004) and (222) planes of Ag₂S, respectively, which confirms the high crystallinity of the attached Ag₂S NPs. Fig. 3(c) shows the HRTEM image of one of the Ag₂WO₄ microrods. In this image, two lattice spacings of approximately 0.297 and 0.284 nm are clearly observed, which are attributed to the (002) and (231) planes of Ag₂WO₄, respectively. Furthermore, an intimate interface between Ag₂S and Ag₂WO₄ was observed from the TEM images. The good interfacial contact in U-Ag₂S/Ag₂WO₄ composite could promote the transfer of charge carriers between the Ag₂WO₄ host photocatalytic material and the Ag₂S co-catalyst, thus promoting the photocatalytic performance.

The surface element composition and chemical state of U-Ag₂S/Ag₂WO₄ were evaluated by XPS. As shown in Fig. 4(a), the XPS survey spectrum proves the co-existence of Ag, S, W and O elements. In the high resolution XPS spectrum of Ag 3d (Fig. 4(b)), the two separate peaks at 373.3 and 367.3 eV are assigned to the Ag⁺ for Ag 3d_3/2 and Ag 3d_5/2 ^[17], respectively. Fig. 4(c) shows the high resolution spectrum of O 1s. The peaks at 530.6 and 529.7 eV correspond to the lattice oxygen in Ag₂WO₄ and absorbed oxygen ^[31], respectively. The peak at 531.8 eV is assigned to the chemisorbed oxygen in S-O ^[31]. The band energies of S 2p (Fig. 4(d)) at 168.2 and 167.1 eV arise from the S in Ag-S-O ^{[32, 33]}. This suggests that an Ag₂S/Ag₂WO₄ heterojunction could be formed between the Ag₂S NPs and Ag₂WO₄ microrods. The two peaks at 162.3 and 161.0 eV can be assigned to S 2p_1/2 and S 2p_3/2 ^[16], respectively. Fig. 4(e) shows the W 4f spectrum, and the two peaks at 36.6 and 34.4 eV are assigned to W 4f_7/2 and W 4f_5/2 for W⁶⁺ in Ag₂WO₄, respectively.

UV-vis DRS was used to evaluate the light absorption behavior of the samples in the UV to visible light range (λ = 250- 800 nm). Fig. 5(a) shows that the absorption edge of pure Ag₂WO₄ is at ~380 nm and its light absorption is focused in the ultraviolet region. Ag₂S displays obvious ultraviolet and visible light absorption and its absorption is mainly at ~450 nm. After the introduction of Ag₂S, the absorption edge of Ag₂S/Ag₂WO₄ samples shows a red-shift and an increase in the visible light absorption appears. Comparing the absorption spectrum of Ag₂S/Ag₂WO₄ with that of U-Ag₂S/Ag₂WO₄, we can observe that the ultrasonic treatment obviously shifts the light absorption edge to the visible light range and, additionally, greatly enhances its light absorption ability. The optical band gap energy of a semiconductor can be estimated from the Tauc plot (i.e., the curve of the converted (αhv)ⁿ versus hv from the UV-vis spectrum, in which α, h and v are the absorption coefficient, Planck constant and light frequency, respectively, and n = 2 for a direct band gap material and n = 1/2 for an indirect band gap material ^[34]). Here, the value of n was selected as 2 and Fig. 5(b) shows a good linear fit when the value n = 2 was applied. The E_g values of U-Ag₂WO₄ and U-Ag₂S were thus determined to be 3.29 and 2.19 eV, respectively, by measuring the x-axis intercept of the extrapolated line from the linear regime of the curve. The loading of Ag₂S resulted in a decrease in the E_g. The value of E_g for U-Ag₂S/Ag₂WO₄ was determined to be 1.88 eV.

PL analysis was performed to investigate the separation efficiency of the photogenerated electrons and holes. Fig. 6 displays the typical PL spectra of Ag₂WO₄, Ag₂S/Ag₂WO₄ and U-Ag₂S/Ag₂WO₄. The emission peak appearing at 468 nm could be attributed to the radiative recombination process of the photo-excited carriers in Ag₂WO₄. Fig. 6 shows that pure Ag₂WO₄ displays the strongest intensity PL peak among the three samples, indicating that the highest recombination probability of the charge carrier is in Ag₂WO₄. The positions of the emission band for Ag₂S/Ag₂WO₄ and U-Ag₂S/Ag₂WO₄ are similar to Ag₂WO₄, while the PL peak intensity of U-Ag₂S/Ag₂WO₄ is the weakest, indicating that both the coupling of Ag₂S and ultrasonic irradiation could promote the separation of charge carriers.

The charge separation efficiency was further investigated by the transient photocurrent response. The transient photocurrent responses were monitored by electrodes consisting of Ag₂S, Ag₂WO₄, Ag₂S/Ag₂WO₄ and U-Ag₂S/Ag₂WO₄ samples. Fig. 7 shows the comparison of the photocurrent-time (I-t) curves of the samples with several on-off cycles under intermittent irradiation from a Xe lamp. Fast photocurrent responses were observed in the samples when the light was turned on, and the photocurrent rapidly decayed when the light was turned off. This indicates that most of the photogenerated electrons are transported back to the contact across the sample to produce photocurrent under Xe lamp irradiation. Ag₂S/Ag₂WO₄ and U-Ag₂S/Ag₂WO₄ exhibited a much higher photocurrent intensity than that of Ag₂WO₄ and Ag₂S. It is proposed that the stronger photocurrent intensity reveals a higher separation efficiency of electrons and holes. This result further demonstrates that a noticeable improvement in the separation of photogenerated electrons and holes, realized by coupling of Ag₂S and ultrasonic irradiation, could also largely benefit the boosting of photocatalytic activity.

MB was used as the photodegradation target under metal halide lamp illumination. Fig. 8(a) and (b) shows that the addition of Ag₂S as a photocatalyst did not result in an obvious degradation reaction for MB, suggesting the poor photocatalytic activity of Ag₂S. Ag₂WO₄ displays an obvious photocatalytic activity for MB decomposition. The highest degradation rate of MB (99%) was observed over U-Ag₂S/Ag₂WO₄. Fig. 8(c) shows the degradation dynamics curves of MB over the as-prepared samples. The largest degradation rate constant over U-Ag₂S/Ag₂WO₄ was 0.149 min^-1, which is 4.7 times and 29.8 times higher than that of bare Ag₂WO₄ (0.032 min^-1) and Ag₂S (0.005 min^-1), respectively. Fig. 8(d) shows the UV-vis absorption spectra of MB with different reaction times in the presence of U-Ag₂S/Ag₂WO₄.

The photocatalytic activities of the as-prepared samples were further evaluated by the degradation of another dye. Fig. 9(a) and (b) display the concentration changes of MO over different samples. Both Ag₂S and Ag₂WO₄ did not exhibit good photocatalytic performance. After ultrasonic treatment, U-Ag₂S displayed a significant improvement in the photocatalytic activity. Once Ag₂S was loaded into Ag₂WO₄, Ag₂S/Ag₂WO₄ showed a large enhancement in the photocatalytic activity. Especially, U-Ag₂S/Ag₂WO₄ showed the highest photocatalytic activity, and the degradation rate of MO over it reached 75%.

To explore the degradation mechanism, radical capture experiments were carried out. Generally, hydroxyl radicals (^•OH), superoxide radicals (^•O₂^-) and photogenerated holes (h⁺) are the main radicals for the photocatalytic degradation of organic pollutants. Here, a series of radical capturing experiments were performed by using tert-butyl alcohol (TBA, a quencher of ^•OH) ^[35], benzoquinone (BQ, a quencher of ^•O₂^-) ^[36] and triethanolamine (TEOA, a quencher of h⁺) ^[37] in the U-Ag₂S/Ag₂WO₄ reaction system. Fig. 10 shows the effect of the addition of quenchers on the photocatalytic degradation of MO over U-Ag₂S/Ag₂WO₄. It was shown that the degradation rate of MO was maintained around 75% in the presence of TBA, which means that ^•OH has little effect on the photodegradation of MO. When the TEOA or BQ was added into the reaction solution, the photocatalytic degradation rate was substantially decreased. It was proposed that h⁺ and ^•O₂^- are the major reactive species in this U-Ag₂S/Ag₂WO₄ system.

We proposed a possible mechanism to explain the enhanced activity over the U-Ag₂S/Ag₂WO₄ composite. As shown in Fig. 11, upon light irradiation, electrons (e^-) in the valence band (VB) will be excited to the conduction band (CB) and holes (h⁺) will be left in the VB. Then, the photogenerated e^- in the CB will be trapped by adsorbed O₂ to produce ^•O₂^- radicals. Both ^•O₂^- and h⁺ are powerful oxidative species which can break down the chromophores of organic pollutants into small molecules, for example, CO₂ and H₂O. Moreover, the intimate interface of Ag₂S/Ag₂WO₄, with a matched band energy structure between Ag₂S and Ag₂WO₄ phases, could promote the transfer of charge carriers between Ag₂WO₄ and Ag₂S. Thus, a relatively large separation of photogenerated holes and electrons was obtained over U-Ag₂S/Ag₂WO₄, which was confirmed by the former PL and transient photocurrent tests. The radical capture experiments suggested that h⁺ and ^•O₂^- are the major reactive species in the U-Ag₂S/Ag₂WO₄ system. The relatively large separation of the photogenerated holes and electrons provide more active substances (e.g., h⁺ and e^-→^•O₂^-) to decompose the dye molecules.

The stability of the catalyst is an important factor to evaluate its potential for practical applications. Therefore, the performance of Ag₂WO₄, Ag₂S/Ag₂WO₄ and U-Ag₂S/Ag₂WO₄ was compared through recycling degradation reactions. Fig. 12 shows that after three cycles, the activity of pure Ag₂WO₄ decreased substantially. The degradation rate was reduced from 61.5% to 5.6%. Compared with Ag₂WO₄, the stability of the Ag₂S/Ag₂WO₄ was greatly improved. U-Ag₂S/Ag₂WO₄ yielded the highest stability. After three cycles, the degradation rate of MB over it was still maintained at ~80%.

Ag₂S/Ag₂WO₄ composite microrods were fabricated by a sonochemical route. Ultrasonic irradiation resulted in an obvious improvement of the textural properties, for example, the increase in crystallinity and surface area. Moreover, sonochemically fabricated Ag₂S/Ag₂WO₄ microrods exhibited strong visible light absorption and a transient photocurrent response. The produced intimate Ag₂S/Ag₂WO₄ interface between Ag₂S and Ag₂WO₄ phases was able to promote the separation of photogenerated h⁺ and e^-. A high photocatalytic activity and stability were obtained over U-Ag₂S/Ag₂WO₄ composite microrods. The degradation rate constant of U-Ag₂S/Ag₂WO₄ was 4.7 times and 29.8 times higher than that of bare Ag₂WO₄ and Ag₂S, respectively. This sonochemical route provides interesting information for the design of highly efficient visible light driven Ag-based semiconductor photocatalysts.