Water Treatment for Degradation of Carbamazepine

In situ growth of Ag-SnO₂ quantum dots on silver phosphate for photocatalytic degradation of carbamazepine: performance, mechanism and intermediates toxicity assessment

Abstract

The occurrence of carbamazepine (CBZ) in aquatic environments constitutes a potential risk to aquatic life including g bacteria, algae, invertebrates and fish, and exhibits growth inhibition of human embryonic cells. In this work, for the first time a novel Ag-SnO₂ quantum dots (QDs)/silver phosphate composite (AgSn/AgP) was synthesized and used as a photocatalyst for CBZ degradation. SEM, TEM, XRD, XPS, UV-vis DSR, PL, EIS and DRS tests have been used to study the structural and physicochemical characteristics of the AgSn/AgP. After 120 min visible light irradiation, the CBZ photodegradation efficiency over 10AgSn/AgP (with 10% wt% Ag-SnO₂ QDs content) reached 86.5%, which was much higher than that over pure Ag-SnO₂ QDs (21.2%), Ag₃PO₄ (28.5%), binary Ag/Ag₃PO₄ (25.8%) and SnO₂ QDs/Ag₃PO₄ (63.6%). The enhanced photocatalytic activity could be mainly ascribed to highly efficient charge separation through a synergistic effects of Ag-SnO₂ QDs, Ag₃PO₄, and in situ photo-reduced Ag nanoparticles. Radical trapping experiments and electron spin resonance measurement revealed that photogenerated holes (h⁺), superoxide radical (·O₂^–) and ·OH jointly participated in the decomposition of CBZ. A possible photocatalytic mechanism for the charge-transfer process was proposed to account for the enhanced photocatalytic performance of 10AgSn/AgP for CBZ degradation. In addition, eight degradation intermediates were identified by liquid chromatography-mass spectrometry (LC-MS). Toxicity evaluation by the ECOSAR program showed that the acute and chronic aquatic toxicity of most oxidation intermediates was much lower than that of CBZ. This study could provide useful information for the potential application of 10AgSn/AgP to treat water and wastewaters containing CBZ.

Keywords: Ag-SnO₂ quantum dots; Ag₃PO₄; visible light photocatalysis; Carbamazepine; Toxicity assessment

1. Introduction

The increasing global crisis of energy shortage and environ-mental issues are becoming serious threats to the sustainable development of human society. With outstanding merits including environmentally friendly and inexhaustible supply, sunlight has been served as the most ideal power to resolve the energy shortage and pollution removal [1].

In the past few years, much efforts have been made in developing novel visible light responsible photocatalysts. Silver phosphate (Ag₃PO₄) is regarded as a significant breakthrough in the field of visible light responsive photocatalysts and has attracted considerable attention since it was discovered by Yi et al. in 2010 [2]. Ag₃PO₄ possesses a suitable band gap of 2.45 eV, an absorption edge at ~530 nm and a quantum yield of more than 90% in visible light [3, 4]. Similar to other narrow-bandgap semiconductors, Ag₃PO₄ photocatalyst also suffers from a high recombination rate of photogenerated charge carriers and limited responsive range within the wide sunlight spectrum [5, 6]. Moreover, due to its slight solubility in solution (0.02 g/L) and its own characteristic of energy-band structure, the uncontrolled photo-corrosion in the photocatalysis process also restrict its wide application [7-9]. Recently, efficient strategies have been developed to improve the visible-light photocatalytic performance and stability of Ag₃PO₄ by coupling it with other semiconductors that possesses a different band structure, with successful cases such as Ag₃PO₄/AgCl [10], Ag₃PO₄/SrTiO₃ [11], Ag₃PO₄/MoS₂ [12], Ag₃PO₄/BiVO₄ [13], Ag₃PO₄/Co₃O₄ [14], Ag₃PO₄/MoO₃ [15], In₂O₃/Ag/Ag₃PO₄ [16].

As a typical n-type semiconductor, tin oxide (SnO₂) has wide range of potential application in the field of catalysts, transparent electrodes, gas sensors, and batteries [17, 18]. The structure, band gap, and chemical stability of SnO₂ are similar to those of titanium oxide (TiO₂), which is a widely used photocatalyst. Moreover, unlike the above mentioned semiconductors (SrTiO₃, MoS₂, BiVO₄, Co₃O₄, and In₂O₃), SnO₂ possesses exceptional properties of low cost, easy production, environmental friendliness, good chemical and biological inertness [19, 20]. Thus SnO₂ is potentially an ideal photocatalyst. Unfortunately, SnO₂ normally exhibits high photocatalytic activity only under ultraviolet light irradiation due to its wide band gap (∼3.6 eV), which significantly limits its practical applications in environmental remediation [21]. SnO₂ quantum dots (QDs) exhibits interesting properties of high specific surface area, short charge transport paths and strong quantum confinement to reach superior photocatalytic performance for dyes degradation [22]. However, the high recombination rate of the photogenerated electron-hole pairs in the pure SnO₂ QDs has still restricted its photocatalytic efficiency. Several studies has been made to improve the photocatalytic activity of SnO₂ QDs. Babu et al. reported that the photocatalytic performance of SnO₂ QDs for dyes was highly enhanced by doping Ag [23]. Sinha et al. synthesized Ag-SnO₂ nanocomposites for efficient removal of industrially emerging pollutants, the degradation efficiency of methylene blue, rose Bengal, methyl violet 6B, and 4-nitrophenol were found to be 99.1%, 99.6%, 99.5%, and 98.4%, respectively [24]. Doping small mass content of Ag can not only enhance the conductivity of SnO₂ QDs owing to the tailoring of the resistivity, the carrier concentrations and mobility, but also increase the visible light adsorption of SnO₂, which could be beneficial for photocatalysis.

However, to the best of our knowledge, no study has been performed on combining Ag-SnO₂ QDs with Ag₃PO₄ for photocatalyst synthesis, especially in application of the composite for photocatalytic degradation of carbamazepine (CBZ). Thus, in this work a novel Ag-SnO₂ QDs/Ag₃PO₄ (AgSn/AgP) photocatalysts with different weight ratios of Ag-SnO₂ QDs were prepared through an in situ precipitation method and were well characterized. Detailed investigations of the photocatalytic activity of AgSn/AgP toward CBZ was carried out under visible light irradiation. And a possible mechanism of the charge-transfer process was proposed to account for the enhanced photocatalytic performance of 10AgSn/AgP for CBZ degradation based on active species trapping experiments and ESR analysis. Moreover, for the first time, we have reported the acute and chronic toxicity of CBZ and its degradation intermediates in photocatalysis system to fish, daphnia and green algae using ECOSAR program.

2. Experimental section

All the chemicals in this study were of analytical grade and used as received without further purification. Deionized water was used throughout the experiments.

2.1. Synthesis of Ag-SnO₂ QDs

The Ag-SnO₂ QDs was synthesized through the following procedure. Briefly, 2.5 g SnCl₄·5H₂O and 0.125 g AgNO₃ was dissolved in 125 mL deionized water. The mixture was stirred continuously for 30 min, and then ammonia solution was dropwise added into the mixture until the pH was achieved 11.0. Subsequently, 2.0 mL of hydrazine was added to the above solution and kept stirring for another 90 min. Thereafter, the resulting homogeneous solution was transferred to a Teflon-lined stainless steel autoclave, sealed and heated at 120 ℃ for 18 h. After naturally cooling to room temperature, the solid products were washed with deionized water and ethanol several times, and dried.

2.2. Synthesis of AgSn/AgP composites

The AgSn/AgP composites were prepared via a simple in situ precipitation method as follows: a certain amount of the as-prepared Ag-SnO₂ QDs was distributed in 100 mL deionized water by ultrasonic treatment. 1 g AgNO₃ was added into the above solution and kept stirring for 1 h. Subsequently, 23 mL NaH₂PO₄·12H₂O (0.71 g) solution was added into the suspension dropwise under stirring. After that, the liquid mixture was centrifugalized and the centrifugate was washed repeatedly with deionized water. The AgSn/AgP composites with 1, 5, 10, 15, 20 wt % of Ag-SnO₂ QDs, which were expressed as 1AgSn/AgP, 5AgSn/AgP, 10AgSn/AgP, 15AgSn/AgP, 20AgSn/AgP, respectively, were obtained after drying at 60 ℃ under vacuum conditions.

2.3. Characterization

The morphology and microstructure of prepared samples were characterized by a scanning electron microscope (SEM, Hitachi s-4800, Japan) and Transmission electron microscopic (TEM, FEI TECNAI G2 F30, USA) with 300 kV acceleration voltage. X-ray diffraction (XRD) was carried out at room temperature using a D8 Advance X-ray diffractometer employing Ni-filtered Cu Kα radiation at a scan rate (2θ) of 0.02°/s. The chemical composition and state of the samples were conducted using X-ray photoelectron spectroscopy (XPS) using an ESCALAB 250 apparatus (Thermo Fisher Scientific) at 3×10^-10 mbar using Al Kα X-ray beam (1486.6 eV). UV-vis DRS were recorded on a spectrophotometer (UV-vis, DRS Shimadzu, UV-3600). PL spectra were obtained by using a Fluorescence spectrofluorometer (F-7000, Hitachi, Japan) at room temperature with an excitation wavelength of 350 nm, and all the samples were pressed into pellets in the sample holder. The photocurrent and EIS of the samples were obtained using an electrochemical workstation (CHI 660e, Shanghai, China) with a standard three-electrode system. A Pt wire was employed as a counter electrode and the Ag/AgCl electrode as a reference electrode. A 0.5 M Na₂SO₄ solution acted as the electrolyte and a 500 W Xe lamp served as a light source. The working electrodes were prepared as follows: 4 mg of samples were added into 0.9 mL water and ultrasonication for 15 min to form a homogeneous suspension. Then, 0.3 mL suspension was dropped directly onto the surface of 10 mm × 40 mm ITO glass. Thereafter, the ITO glass with samples were dried for 2 h in a vacuum oven.

2.4. Photocatalytic tests

The photocatalytic activities of as-prepared photocatalysts were examined toward CBZ in aqueous solution under visible light irradiation (λ > 420 nm, 300 W Xenon lamp). Typically, 20 mg as-prepared samples were added into a CBZ (50 mL, 0.5 mg/L) solution, and followed by stirring under dark condition for 30 min to achieve the saturated adsorption. After that, turn on the lamp and 0.75 mL aliquots were extracted from each sample at regular intervals and centrifuged to remove the catalysts. CBZ concentrations during the photocatalytic process were monitored by high performance liquid chromatography (HPLC, Agilent 1200, USA) outfitted with a XDB-C18 column (150 × 4.5 mm i.d., 5 μm particle, Waters, USA) and fitted with a UV detection at 286 nm. The mobile phase was a mixture of acetonitrile and deionized water (60:40) at a flow rate of 1.0 mL/min. The degradation rate constant (k) of CBZ over the photocatalysts were calculated using the equation:

k=1tlnCC0, where C and C₀ are the solution concentration at reaction time t and 0 respectively.

3. Results and discussion

3.1. Characterization

The microstructure of as-prepared samples was investigated SEM and TEM. As depicted in Fig. 1a and d, Ag-SnO₂ QDs particles are regular sphere with an average size of ca. 10 nm. While sphere-like Ag₃PO₄ particles with an average size of ca. 250-500 nm can be observed in Fig. 1b and e. SEM image of 10AgSn/AgP represented in Fig. 1c indicates that Ag-SnO₂ QDs particles are closely distributed on the surface of Ag₃PO₄. This observation can be further clarified in the TEM image (Fig. 1f), in which Ag-SnO₂ QDs and Ag₃PO₄ are well contacted, facilitating the transfer of charge at the interface. Moreover, the EDS patterns (Fig. 1g) demonstrates the as-prepared samples consist of elemental Ag, Sn, O, and P, and they are homogeneously distributed throughout the whole material, and no other elements are present. The results are powerful evidences for the successful synthesis of Ag-SnO₂/Ag₃PO₄.

Fig. 1. SEM images of (a) Ag-SnO₂ QDs, (b) Ag₃PO₄, and (c) 10AgSn/AgP, TEM images of (d) Ag-SnO₂ QDs, (e) Ag₃PO₄, and (f) 10AgSn/AgP, (g) HRTEM images of CN/AgSn-30 and corresponding EDX elemental mapping of Ag, Sn, P and O for 10AgSn/AgP.

Composition and crystallographic structure of Ag-SnO₂ QDs, Ag₃PO₄, and 10AgSn/AgP were determined by XRD patterns as illustrated in Fig. 2a, b and c. As shown in Fig. 2a, all of the diffraction peaks in the patterns including (110), (101), (211), (220), (002), (310), (112), (301), (202) and (321) can be indexed to the tetragonal phase of SnO₂ (JCPDS, No. 41-1445) [25]. A weak XRD diffraction peaks at about 37.9° can be indexed to the (111) facet of metallic Ag, and the d spacing value (d₁₁₀) of Ag is calculated to be 0.229 nm by using Bragg’s equation. The diffraction peaks of 2θ at 20.8°, 29.7°, 33.3°, 36.5°, 42.4°, 47.8°, 52.7°, 55.0°, 57.4°, 61.6° and 65.8° in patterns of Fig. 2b, indexed to (110), (200), (210), (211), (220), (310), (222), (320), (321), (400) and (330) planes of the cubic Ag₃PO₄ phase (JCPDS No. 74-1876) [26]. The sharp and intense XRD peaks can be attributed to the highly crystallized structure of Ag₃PO₄. As for 10AgSn/AgP (Fig. 2c), all the diffraction peaks related to the Ag₃PO₄ can be observed but without the detection of characteristic peaks of Ag-SnO₂, which is ascribed to the low intensity and low weight ratio of Ag-SnO₂ in the composite. No additional crystal phases could be found in all patterns and presented with sharp diffraction peaks, implying no impurities generated in the fabrication procedure.

Fig. 2. XRD patterns of (a) Ag-SnO₂ QDs, (b) Ag₃PO₄, and (c) 10AgSn/AgP, (d) XPS spectrum of 10AgSn/AgP before and after photocatalytic reaction.

All signals of Ag, Sn, O, and P are clearly observed in the full XPS spectrum of the fresh and used 10AgSn/AgP (Fig. 2d), which is consistent with the EDS results. To further prove the existence of Ag, Sn, O, and P and their chemical states, high-resolution of spectra of Ag 3d, Sn 3d, P 2p and O 1s are provided (Fig. S1). The two typical peaks of Ag 3d_5/2 and Ag 3d_3/2 located at 368.1 and 374.1 eV are related to Ag⁺ [27] (Fig. S1a). The peaks at 487.4 and 495.9 eV in Fig. S1b can be attributed to Sn 3d_5/2 and Sn 3d_3/2 of Sn 3d [28].

The UV-vis diffuse reflectance spectra of the as-prepared samples are depicted in Fig. 3. The pristine Ag-SnO₂ QDs shows a strong absorption peak between 250-400 nm, while the Ag₃PO₄ exhibits an absorption edge at about 550 nm, which agrees with previous reports [29, 30]. Apparently, the combination of Ag-SnO₂ and Ag₃PO₄ results in an enhanced photo-response over the whole visible light region. Therefore, the optical absorption property is conductive to the efficient utilization of solar energy and superior photocatalytic performance for the 10AgSn/AgP.

According to the Kubelka-Munk function, the band gap of pure Ag-SnO₂ QDs and Ag₃PO₄ can be calculated using the equation [31]:

αhv=Ahv-Egn/2, where A, α, h, ν and E_g are the proportional constant, absorption coefficient, Planck constant, light frequency and band gap, respectively. The n value is determined by the type of semiconductor (n = 1 for direct transition and n = 4 for indirect transition). From the plot of (αhν)² versus E_g in Fig. 3b, the transition band gaps estimated from the onset of the curve edges are about 2.42 and 2.88 eV for pure Ag₃PO₄ and Ag-SnO₂ QDs, which are in consistent with previous literature [8, 23]. In addition, the band edge positions of the valance band (VB) and conduction band (CB) can be calculated through the empirical formulas: E_VB = X – E_e + 0.5E_g, E_CB = E_VB – E_g, where X is the absolute electronegativity of the semiconductor, E_e is a constant of approximately 4.5 eV, representing the energy of free electrons on the hydrogen scale, E_g is the band gap energy of the semiconductor [32]. Thus, the E_VB of Ag₃PO₄ and Ag-SnO₂ QDs are calculated to be 2.89 and 2.68 eV respectively. Accordingly, the E_CB of these two samples are 0.47 and -0.20 eV respectively.

Fig. 3. (a) UV-vis diffuse reflectance spectra of the as-prepared samples, (b) band gap of pure Ag-SnO₂ QDs and Ag₃PO₄.

3.2. Photocatalytic performance evaluation

Fig. 4a and b displays the photocatalytic degradation of CBZ as a function of irradiation time over different photocatalysts. On the basis of the experiments without any photocatalyst, direct photocatalysis of CBZ under visible light could be neglected. For pure Ag-SnO₂ QDs and Ag₃PO₄, after 120 min visible light irradiation the photocatalytic degradation of CBZ were 21.2% and 28.5% respectively, while the binary Ag/Ag₃PO₄ and SnO₂ QDs/Ag₃PO₄ composites exhibited 25.8% and 63.6% of CBZ degradation respectively. Obviously, the introduction of Ag-SnO₂ QDs into Ag₃PO₄ greatly enhanced the photocatalytic performance of Ag₃PO₄. As shown in Fig. 4b, with increasing the Ag-SnO₂ QDs content, the photocatalytic activities of AgSn/AgP increased at first and then declined. After 120 min irradiation, the CBZ removal efficiencies over 1AgSn/AgP, 5AgSn/AgP, 10AgSn/AgP, 15AgSn/AgP, and 20AgSn/AgP were 70.2%, 73.5%, 86.5%, 75.5%, and 71.6%, respectively. The overloading of Ag-SnO₂ QDs in the composite can promote the utilization of photoelectrons, but can also suppress the light adsorption of semiconductor host. The results indicated that the AgSn/AgP composite exhibited higher photocatalytic activity than Ag-SnO₂ QDs and Ag₃PO₄ as well as binary Ag/Ag₃PO₄ and SnO₂ QDs/Ag₃PO₄ composites. In this study, Ag-SnO₂ QDs (10%) was demonstrated to be optimal ratio between these two contradictory factors. As shown in Fig. S2, the photocatalytic reaction rate could be described by pseudo-first-order kinetics. Fig. 4c shows the k values for the CBZ degradation over different samples. The 10AgSn/AgP displayed the highest rate constant (0.017 min^-1), which was about 8.5, 5.7, 5.7, and 1.9 times higher than that of Ag-SnO₂ QDs (0.002 min^-1), Ag₃PO₄ (0.003 min^-1), Ag/Ag₃PO₄ (0.003 min^-1), and SnO₂ QDs/Ag₃PO₄ (0.009 min^-1), respectively.

Fig. 4. (a) The photocatalytic degradation of CBZ based on different photocatalysts; (b) effect of Ag-SnO₂ mass loading on CBZ photocatalytic degradation; (c) pseudo-first-order kinetics rate constant of CBZ degradation over different photocatalysts; (d) cycling experiments of 10AgSn/AgP and Ag₃PO₄ for CBZ photocatalytic degradation; high resolution XPS spectrum of Ag 3d for (e) Ag₃PO₄ and (f) 10AgSn/AgP after 3 runs of photocatalytic reaction.

Apart from the degradation efficiency, the stability of catalyst is also an important factor in practical applications. The photo corrosion of Ag₃PO₄ has been the key problem for its practical application as Ag₃PO₄ is soluble in water and the free Ag⁺ ion can be easily reduced to Ag⁰ by photoelectrons [33]. Therefore, the stability of Ag₃PO₄ is one of the biggest challenges before putting it into applications. Recycling experiments for the CBZ photocatalytic degradation under visible light irradiation were carried out with the as-prepared Ag₃PO₄ and 10AgSn/AgP. As presented in Fig. 4d, both the photocatalytic stabilities of Ag₃PO₄ and 10AgSn/AgP gradually deteriorated because of an increment of Ag⁰ phase during photocatalysis. The degradation efficiency of CBZ by the 10AgSn/AgP decreased from 86.5% to 66.6% from the first run to the third run. However, as for Ag₃PO₄, the photocatalytic performance declined by more than a half from 29.5% to 11% after 3 cycles, indicating that a large number of Ag₃PO₄ could be greatly reduced to Ag⁰ by photogenerated electrons. With the increase of Ag⁰ content, there would be a shield layer on the Ag₃PO₄ surface formed by metallic Ag, blocking the light absorption of Ag₃PO₄, thereby inhibiting the transfer of carrier between Ag₃PO₄ and solution. Moreover, it could occupy some active sites of reaction molecules, resulting in the decline of catalytic activity [34].

High resolution XPS spectrum of Ag 3d for Ag₃PO₄ and 10AgSn/AgP after reaction were also examined. As shown in Fig. 4e, two peaks appearing at 368.1 and 374.1 eV could be attributed to the characteristic peaks of Ag 3d_5/2 and Ag 3d_3/2 respectively, suggesting the valence state of Ag was +1. However, the peaks located at 368.1 and 374.1 eV were related to Ag⁰. Comparing Fig. 4e and f, it’s obvious that the intensities of two peaks at 368.3 and 374.3 eV in Ag₃PO₄ were much higher than that of 10AgSn/AgP. This observation demonstrated that a large amount of Ag⁰ was formed in Ag₃PO₄ after 3 cycles due to the self-decomposition. In contrast, only a negligible amount of Ag⁰ was present in 10AgSn/AgP due to its excellent anti-photocorrosion performance.

3.3. Charge transport and separation efficiency

In general, the interfacial charge transport in heterostructured photocatalysts can promote the charge separation efficiency and enhance the photocatalytic performances. To further explicate the activity promotion of 10AgSn/AgP, the charge transport and separation have been investigated by electrochemical impedance spectroscopy (EIS), transient photocurrent, and photoluminescence (PL) characterizations. Fig. 5a displays the transient photocurrent responses of pristine Ag-SnO₂ QDs, Ag₃PO₄, and 10AgSn/AgP. The photocurrent intensity over Ag₃PO₄ suddenly increased rapidly when illuminated by visible light, but decreased sharply while the light was turn off. The photocurrent intensity over 10AgSn/AgP was much higher than that over pristine Ag₃PO₄, while that over Ag-SnO₂ QDs could be negligible. Since higher photocurrent implies more efficient charge migration and lower carrier recombination [35], the charge recombination over 10AgSn/AgP should be severely depressed, resulting in the activity enhancement. This deduction was further supported by the EIS and PL measurements. Fig. 5b illustrates the EIS Nyquist plots for Ag-SnO₂ QDs, Ag₃PO₄, and 10AgSn/AgP. Obviously, the arc radius of 10AgSn/AgP was the smallest of the three samples. Because a smaller arc radius generally represents lower resistance and higher charge transport rate [36], thus, the 10AgSn/AgP composites possessed a higher carrier transporting efficiency compared with bare Ag-SnO₂ QDs and Ag₃PO₄. PL analysis was conducted to examine the recombination rate of the photogenerated carriers of the samples. As shown in Fig. 5c, the PL emission intensities were in the order Ag-SnO₂ QDs > Ag₃PO₄ > 10AgSn/AgP. It’s well known that a high PL emission intensity means a big recombination rate of the photogenerated charge carriers. Therefore, as Ag-SnO₂ QDs was introduced into Ag₃PO₄ nanoparticles, the PL intensity weakened significantly, suggesting the higher separation efficiency of the electron-hole pairs in 10AgSn/AgP than pure Ag-SnO₂ QDs or Ag₃PO₄, thereby acquiring the enhanced photocatalytic performance.

Fig. 5. (a) Transient photocurrent responses, (b) EIS plots, (c) PL emission spectra of Ag-SnO₂ QDs, Ag₃PO₄, and 10AgSn/AgP.

3.4. Possible photocatalytic mechanism

It’s known that hydroxyl radical (·OH), superoxide radical (·O₂⁻), and hole (h⁺) are three main active species in the photocatalytic reaction process [14, 37]. To explore the photocatalytic mechanism of the 10AgSn/AgP and identify the dominant reactive species in more details, the trapping experiments were performed as shown in Fig. 6. In this study, benzoquinone (BQ), isopropanol (IPA), and triethanolamine (TEOA) were employed to act as ·O₂⁻, ·OH, and h⁺ scavengers, respectively [38]. The addition of IPA did not affect the CBZ photodegradation (Fig. 6a), suggesting that ·OH radicals were not the main reactive species in the photocatalytic process. On the contrary, the photocatalytic degradation was obviously suppressed in the presence of BQ and TEOA, and the rate constants declined from 0.017 min^-1 to 0.002 and 0.001 min^-1 respectively (Fig. S3). According to the result, it can be concluded that ·O₂⁻ and h⁺ acted as the dominant species for the photocatalytic degradation of CBZ, while ·OH played a minor role in the process. The trapping experiments of active species during the photocatalytic degradation of CBZ over other Ag₃PO₄-based photocatalysts also suggested the much more important role of ·O₂⁻ and h⁺ in the photocatalytic process (Fig. S4).

Fig. 6. (a) The trapping experiments of the active species during the photocatalytic degradation of CBZ over 10AgSn/AgP; ESR spectra of (b) DMPO-·O₂^– adducts in methanol solution and (c) DMPO-·OH adducts in aqueous solution recorded with 10AgSn/AgP in the dark and under visible light irradiation (λ> 420 nm); (d) Proposed photocatalytic degradation mechanism of CBZ over 10AgSn/AgP under visible light irradiation.

In order to further elucidate the photocatalytic mechanism, electron spin resonance (ESR) spin-trap signals of 10AgSn/AgP were determined with 5,5-dimethyl-1-pyrroline N-oxide (DMPO). As illustrated in Fig. 6b, no ESR signals were detected with the light off. However, six characteristic peaks of the DMPO-·O₂⁻ species appeared with the light on, implying that 10AgSn/AgP could facilitate to produce ·O₂⁻ under visible light irradiation. As for DMPO-·OH, no signals appeared under the dark and four characteristic peaks showed up under visible light irradiation (Fig. 6c). The ·OH radicals might be produced via the hydroxide oxidation by the photogenerated h⁺. However, the signals of ·OH were weaker than that of ·O₂⁻, indicating that ·O₂⁻ radicals had a greater impact on CBZ decomposition than ·OH radicals. Radicals and holes trapping experiments and ESR signals characterization suggested that ·O₂⁻, ·OH and h⁺ jointly participated the photodegradation process of CBZ, following the influence order:  h⁺ >·O₂⁻ > ·OH.

Based on all the results and analysis, a possible charge separation mechanism for photocatalytic degradation of CBZ over 10AgSn/AgP is tentatively proposed in Fig. 6d. At first, under visible light irradiation, Ag₃PO₄ as well as Ag-SnO₂ QDs could be easily excited to yield photogenerated electron-hole pairs, and the photogenerated electrons from the VB of Ag-SnO₂ QDs and Ag₃PO₄ migrated into their corresponding CB of Ag-SnO₂ QDs and Ag₃PO₄ respectively [13]. In the meantime, metallic Ag nanoparticles acted as the charge separation center to form the visible-light-driven 10AgSn/AgP Z-scheme system. That’s to say, the weak reductive photoelectrons from the CB of Ag₃PO₄ could easily transfer into Ag nanoparticles through the Schottky barrier because of the more positive Fermi energy of Ag than the CB level of Ag₃PO₄. Simultaneously, the holes in the VB of Ag-SnO₂ QDs could also shift to metallic Ag nanoparticles. The photogenerated electrons left on the CB of Ag-SnO₂ QDs can be trapped by O₂ in solution to produce ·O₂⁻ active species owing to the relative negative CB potential of Ag-SnO₂ QDs (-0.20 eV vs. SHE) to O₂/·O₂⁻ (-0.046 eV vs. SHE) [29, 39]. And the photogenerated holes left on the VB of Ag₃PO₄ can oxidize CBZ directly or to oxidize H₂O to form ·OH active species.

3.5. Proposed degradation pathway

LC-MS was carried out to monitor the process of CBZ degradation and eight intermediates were detected as listed in Table S1. The proposed CBZ degradation pathway over 10AgSn/AgP was illustrated in Fig. 7. The parent compound of CBZ in accordance with the formula of C₁₅H₁₂N₂O (M = 236) was recorded at m/z 236.1000 with the retention time of 9.2999 min. The identification of acridine-9-carboxyaldehyde (P1, m/z=251.9000), acridine (P2, m/z=180.1000), and P3 (m/z=194.1000) suggested the rearrangement of the seven membered ring of CBZ radical cation to produce P1 [40, 41]. The formation of P2 with the loss of amido and carbonyl, and its further reaction with ·O₂⁻ was responsible for the production of P3, P4, P5 and P6 [42]. In an alternative pathway, CBZ underwent a hydrolysis step to produce intermediate P7 (m/z=210.1000). Intermediate P8 (m/z=234.9000) probably resulted from CBZ through a ring closing reaction about phenolic hydroxyl hydrolysis of amide, which was reported previously [43].

Fig. 7. Proposed pathways for photocatalytic degradation of CBZ by 10AgSn/AgP under visible light irradiation.

3.6. Eco-toxicity assessment of CBZ and its degradation intermediate

ECOSAR program is commonly used to predict the toxicity of parent compound and its transformation intermediates [44, 45]. It is useful for the reduction and control the formation of toxic intermediates via CBZ degradation by identifying the toxicity of intermediates. As depicted in Fig. 8, according to the standard established by the Globally Harmonized System of Classification and Labelling of Chemicals (GHS) [46] (Table S2), CBZ can be classified to be very toxic due to its EC₅₀ value and chronic toxicity values (ChVs) to green algae. Therefore, the toxicity of oxidation intermediates during the catalytic degradation process should be considered. As can be seen in Fig. 8, the acute and chronic toxicity of all intermediates (except for P2, P7 and P8) to three typical aquatic species was less than that of the parent compound CBZ. Despite the slightly high toxicity of P2, P7 and P8, their concentrations were limited. This suggests that CBZ photodegradation over 10AgSn/AgP was very significant for the safety of aquatic environment.

Fig. 8. Predicted toxicity for fish, daphnia and green algae of CBZ and its transformation products by ECOSAR. (LC₅₀ represents Half Lethal Concentration and EC₅₀ represents Half Effective Concentration).

4. Conclusions

A novel Ag-SnO₂ quantum dots (QDs)/silver phosphate composite (AgSn/AgP) Z-scheme photocatalyst was prepared for the first time. The optimum AgSn/AgP hybrid material (10AgSn/AgP with 10 wt% of Ag-SnO₂ QDs loading) exhibited an improved photocatalytic performance under visible light irradiation for CBZ (0.5 mg/L) removal. The obtained degradation efficiency over 10AgSn/AgP reached 86.5% in 120 min, which was much higher than that over pure Ag-SnO₂ QDs (21.2%), Ag₃PO₄ (28.5%), binary Ag/Ag₃PO₄ (25.8%) and SnO₂ QDs/Ag₃PO₄ (63.6%). The enhanced photocatalytic activity of this 10AgSn/AgP composite could be mainly attributed to highly efficient charge separation through the synergistic role of Ag-SnO₂ QDs, Ag₃PO₄, and in situ photoreduced Ag nanoparticles. Radicals and holes trapping experiments and ESR signals characterization indicated that ·O₂⁻, ·OH and h⁺ jointly participated the photodegradation process of CBZ, following the order:  h⁺ >·O₂⁻ > ·OH. With the help of HPLC/MS analysis, eight main reaction intermediates/products were identified and a deductive degradation pathway of CBZ over 10AgSn/AgP was proposed. Eco-toxicity assessment of CBZ and its oxidation intermediate using ECOSAR program demonstrated that CBZ photodegradation over 10AgSn/AgP was very significant for the safety of aquatic environment.

References

[1] A. Sinhamahapatra, J.-P. Jeon, J.-S. Yu, A new approach to prepare highly active and stable black titania for visible light-assisted hydrogen production, Energy & Environmental Science, 8 (2015) 3539-3544.

[2] Z. Yi, J. Ye, N. Kikugawa, T. Kako, S. Ouyang, H. Stuart-Williams, H. Yang, J. Cao, W. Luo, Z. Li, Y. Liu, R.L. Withers, An orthophosphate semiconductor with photooxidation properties under visible-light irradiation, Nature Materials, 9 (2010) 559.

[3] H. Zhang, H. Huang, H. Ming, H. Li, L. Zhang, Y. Liu, Z. Kang, Carbon quantum dots/Ag₃PO₄ complex photocatalysts with enhanced photocatalytic activity and stability under visible light, Journal of Materials Chemistry, 22 (2012) 10501-10506.

[4] S. Huang, Y. Xu, T. Zhou, M. Xie, Y. Ma, Q. Liu, L. Jing, H. Xu, H. Li, Constructing magnetic catalysts with in-situ solid-liquid interfacial photo-Fenton-like reaction over Ag₃PO₄@ NiFe₂O₄ composites, Applied Catalysis B: Environmental, 225 (2018) 40-50.

[5] H. Wang, Y. Bai, J. Yang, X. Lang, J. Li, L. Guo, A facile way to rejuvenate Ag₃PO₄ as a recyclable highly efficient photocatalyst, Chemistry-A European Journal, 18 (2012) 5524-5529.

[6] Z. Liu, Y. Liu, P. Xu, Z. Ma, J. Wang, H. Yuan, Rational design of wide spectral-responsive heterostructures of Au nanorod coupled Ag₃PO₄ with enhanced photocatalytic performance, ACS applied materials & interfaces, 9 (2017) 20620-20629.

[7] D.J. Martin, G. Liu, S.J. Moniz, Y. Bi, A.M. Beale, J. Ye, J. Tang, Efficient visible driven photocatalyst, silver phosphate: performance, understanding and perspective, Chemical Society Reviews, 44 (2015) 7808-7828.

[8] S. Mandal, R. Ananthakrishnan, Sustainable design of hierarchically porous Ag₃PO₄ microspheres through a novel natural template and their superior photooxidative capacity, ACS Sustainable Chemistry & Engineering, 6 (2017) 1091-1104.

[9] X. Chen, Y. Dai, X. Wang, Methods and mechanism for improvement of photocatalytic activity and stability of Ag3PO4: a review, Journal of Alloys and Compounds, 649 (2015) 910-932.

[10] J. Guo, H. Shi, X. Huang, H. Shi, Z. An, AgCl/Ag₃PO₄: A Stable Ag-Based Nanocomposite Photocatalyst with Enhanced Photocatalytic Activity for the Degradation of Parabens, Journal of colloid and interface science, (2018).

[11] X. Guan, L. Guo, Cocatalytic effect of SrTiO₃ on Ag₃PO₄ toward enhanced photocatalytic water oxidation, Acs Catalysis, 4 (2014) 3020-3026.

[12] J. Wan, X. Du, E. Liu, Y. Hu, J. Fan, X. Hu, Z-scheme visible-light-driven Ag₃PO₄ nanoparticle@ MoS₂ quantum dot/few-layered MoS₂ nanosheet heterostructures with high efficiency and stability for photocatalytic selective oxidation, Journal of Catalysis, 345 (2017) 281-294.

[13] C. Li, P. Zhang, R. Lv, J. Lu, T. Wang, S. Wang, H. Wang, J. Gong, Selective deposition of Ag₃PO₄ on monoclinic BiVO₄ (040) for highly efficient photocatalysis, Small, 9 (2013) 3951-3956.

[14] C. Tang, E. Liu, J. Wan, X. Hu, J. Fan, Co₃O₄ nanoparticles decorated Ag₃PO₄ tetrapods as an efficient visible-light-driven heterojunction photocatalyst, Applied Catalysis B: Environmental, 181 (2016) 707-715.

[15] W. Teng, X. Tan, X. Li, Y. Tang, Novel Ag₃PO₄/MoO₃ pn heterojunction with enhanced photocatalytic activity and stability under visible light irradiation, Applied Surface Science, 409 (2017) 250-260.

[16] X. Chen, R. Li, X. Pan, X. Huang, Z. Yi, Fabrication of In₂O₃-Ag-Ag₃PO₄ composites with Z-scheme configuration for photocatalytic ethylene degradation under visible light irradiation, Chemical Engineering Journal, 320 (2017) 644-652.

[17] H. Wang, F. Fu, F. Zhang, H.-E. Wang, S.V. Kershaw, J. Xu, S.-G. Sun, A.L. Rogach, Hydrothermal synthesis of hierarchical SnO₂ microspheres for gas sensing and lithium-ion batteries applications: Fluoride-mediated formation of solid and hollow structures, Journal of Materials Chemistry, 22 (2012) 2140-2148.

[18] W. Tian, C. Zhang, T. Zhai, S.-L. Li, X. Wang, M. Liao, K. Tsukagoshi, D. Golberg, Y. Bando, Flexible SnO₂ hollow nanosphere film based high-performance ultraviolet photodetector, Chemical Communications, 49 (2013) 3739-3741.

[19] M.T. Uddin, Y. Nicolas, C.l. Olivier, T. Toupance, L. Servant, M.M. Müller, H.-J. Kleebe, J.r. Ziegler, W. Jaegermann, Nanostructured SnO₂–ZnO heterojunction photocatalysts showing enhanced photocatalytic activity for the degradation of organic dyes, Inorganic chemistry, 51 (2012) 7764-7773.

[20] L. Yang, J. Huang, L. Shi, L. Cao, H. Liu, Y. Liu, Y. Li, H. Song, Y. Jie, J. Ye, Sb doped SnO₂-decorated porous g-C₃N₄ nanosheet heterostructures with enhanced photocatalytic activities under visible light irradiation, Applied Catalysis B: Environmental, 221 (2018) 670-680.

[21] Y. Du, N. Zhang, C. Wang, Photo-catalytic degradation of trifluralin by SnO₂-doped Cu₂O crystals, Catalysis Communications, 11 (2010) 670-674.

[22] X. Liu, L. Pan, T. Chen, J. Li, K. Yu, Z. Sun, C. Sun, Visible light photocatalytic degradation of methylene blue by SnO₂ quantum dots prepared via microwave-assisted method, Catalysis Science & Technology, 3 (2013) 1805-1809.

[23] B. Babu, M. Cho, C. Byon, J. Shim, One pot synthesis of Ag-SnO₂ quantum dots for highly enhanced sunlight-driven photocatalytic activity, Journal of Alloys and Compounds, 731 (2018) 162-171.

[24] T. Sinha, M. Ahmaruzzaman, P.P. Adhikari, R. Bora, Green and environmentally sustainable fabrication of Ag-SnO₂ nanocomposite and its multifunctional efficacy as photocatalyst and antibacterial and antioxidant agent, ACS Sustainable Chemistry & Engineering, 5 (2017) 4645-4655.

[25] Y. Zou, Y. Wang, Microwave solvothermal synthesis of flower-like SnS₂ and SnO₂ nanostructures as high-rate anodes for lithium ion batteries, Chemical engineering journal, 229 (2013) 183-189.

[26] N. Güy, K. Atacan, E. Karaca, M. Özacar, Role of Ag₃PO₄ and Fe₃O₄ on the photocatalytic performance of magnetic Ag₃PO₄/ZnO/Fe₃O₄ nanocomposite under visible light irradiation, Solar Energy, 166 (2018) 308-316.

[27] T. Yan, J. Tian, W. Guan, Z. Qiao, W. Li, J. You, B. Huang, Ultra-low loading of Ag₃PO₄ on hierarchical In2S3 microspheres to improve the photocatalytic performance: The cocatalytic effect of Ag and Ag3PO4, Applied Catalysis B: Environmental, 202 (2017) 84-94.

[28] S. Su, H. Zhang, L. Zhao, M. He, C. Ling, Band alignment of n-SnO₂/p-GaN hetero-junction studied by x-ray photoelectron spectroscopy, Journal of Physics D: Applied Physics, 47 (2014) 215102.

[29] J. Guo, H. Shi, X. Huang, H. Shi, Z. An, AgCl/Ag₃PO₄: A stable Ag-Based nanocomposite photocatalyst with enhanced photocatalytic activity for the degradation of parabens, Journal of colloid and interface science, 515 (2018) 10-17.

[30] Y. Bu, Z. Chen, C. Sun, Highly efficient Z-scheme Ag₃PO₄/Ag/WO_3−x photocatalyst for its enhanced photocatalytic performance, Applied Catalysis B: Environmental, 179 (2015) 363-371.

[31] M. Butler, Photoelectrolysis and physical properties of the semiconducting electrode WO₂, Journal of Applied Physics, 48 (1977) 1914-1920.

[32] X. Xiao, C. Liu, R. Hu, X. Zuo, J. Nan, L. Li, L. Wang, Oxygen-rich bismuth oxyhalides: generalized one-pot synthesis, band structures and visible-light photocatalytic properties, Journal of Materials Chemistry, 22 (2012) 22840-22843.

[33] T. Cai, Y. Liu, L. Wang, S. Zhang, Y. Zeng, J. Yuan, J. Ma, W. Dong, C. Liu, S. Luo, Silver phosphate-based Z-Scheme photocatalytic system with superior sunlight photocatalytic activities and anti-photocorrosion performance, Applied Catalysis B: Environmental, 208 (2017) 1-13.

[34] L. Xu, W.-Q. Huang, L.-L. Wang, G.-F. Huang, P. Peng, Mechanism of superior visible-light photocatalytic activity and stability of hybrid Ag₃PO₄/graphene nanocomposite, The Journal of Physical Chemistry C, 118 (2014) 12972-12979.

[35] F. Chen, Q. Yang, J. Sun, F. Yao, S. Wang, Y. Wang, X. Wang, X. Li, C. Niu, D. Wang, Enhanced photocatalytic degradation of tetracycline by AgI/BiVO₄ heterojunction under visible-light irradiation: mineralization efficiency and mechanism, ACS applied materials & interfaces, 8 (2016) 32887-32900.

[36] J. Zhang, X. Yuan, L. Jiang, Z. Wu, X. Chen, H. Wang, H. Wang, G. Zeng, Highly efficient photocatalysis toward tetracycline of nitrogen doped carbon quantum dots sensitized bismuth tungstate based on interfacial charge transfer, Journal of colloid and interface science, 511 (2018) 296-306.

[37] D. Xia, T. An, G. Li, W. Wang, H. Zhao, P.K. Wong, Synergistic photocatalytic inactivation mechanisms of bacteria by graphene sheets grafted plasmonic AgAgX (X= Cl, Br, I) composite photocatalyst under visible light irradiation, Water research, 99 (2016) 149-161.

[38] T. Wang, Z. Jiang, T. An, G. Li, H. Zhao, P.K. Wong, Enhanced Visible-Light-Driven Photocatalytic Bacterial Inactivation by Ultrathin Carbon-Coated Magnetic Cobalt Ferrite Nanoparticles, Environmental Science & Technology, 52 (2018) 4774-4784.

[39] Y. Tang, Z. Jiang, G. Xing, A. Li, P.D. Kanhere, Y. Zhang, T.C. Sum, S. Li, X. Chen, Z. Dong, Efficient Ag@ AgCl cubic cage photocatalysts profit from ultrafast plasmon‐induced electron transfer processes, Advanced Functional Materials, 23 (2013) 2932-2940.

[40] D.C. McDowell, M.M. Huber, M. Wagner, U. von Gunten, T.A. Ternes, Ozonation of carbamazepine in drinking water: identification and kinetic study of major oxidation products, Environmental science & technology, 39 (2005) 8014-8022.

[41] Y.-H. Guan, J. Ma, X.-C. Li, J.-Y. Fang, L.-W. Chen, Influence of pH on the formation of sulfate and hydroxyl radicals in the UV/peroxymonosulfate system, Environmental Science & Technology, 45 (2011) 9308-9314.

[42] L. Wang, W. Lu, D. Ni, T. Xu, N. Li, Z. Zhu, H. Chen, W. Chen, Solar-initiated photocatalytic degradation of carbamazepine on excited-state hexadecachlorophthalocyanine in the presence of peroxymonosulfate, Chemical Engineering Journal, 330 (2017) 625-634.

[43] H. Chen, X. Wang, W. Bi, Y. Wu, W. Dong, Photodegradation of carbamazepine with BiOCl/Fe₃O₄ catalyst under simulated solar light irradiation, Journal of colloid and interface science, 502 (2017) 89-99.

[44] J.-F. Yang, M. He, T.-F. Wu, A.-P. Hao, S.-B. Zhang, Y.-D. Chen, S.-B. Zhou, L.-Y. Zhen, R. Wang, Z.-L. Yuan, Sulfadiazine oxidation by permanganate: Kinetics, mechanistic investigation and toxicity evaluation, Chemical Engineering Journal, 349 (2018) 56-65.

[45] I. Carpinteiro, R. Rodil, J.B. Quintana, R. Cela, Reaction of diazepam and related benzodiazepines with chlorine. Kinetics, transformation products and in-silico toxicological assessment, Water research, 120 (2017) 280-289.

[46] N. UNECE, Globally harmonized system of classification and labelling of chemicals (GHS), United Nations New York, 2011.