Photocatalytic Degradation of Quinoline Yellow over Ag3PO4

In this study, the ability of Ag3PO4 to achieve the photocatalytic degradation of quinoline yellow (QY) a hazardous and recalcitrant dye, under UVA and visible light was investigated. The photocatalyst Ag3PO4 was synthesized through a precipitation method, and characterized by X-ray diffraction (XRD), scanning electron microscope (SEM), BET Brunauer–Emmett-Teller (BET) analysis, UV-Differential Reflectance Spectroscopy (DRS) and Fourier transform infrared spectroscopy (FTIR). Ag3PO4 could successfully induce the photocatalytic degradation of QY under UVA and visible light. Optimal parameters were 0.5 g·L−1 of the catalyst, 20 ppm of QY and pH~7. Ag3PO4 was 1.6-times more efficient than TiO2 Degussa P25 under UVA light in degrading QY. Total organic carbon (TOC) analyses confirmed the almost complete QY mineralization. At least eight intermediate degradation products were identified by liquid chromatography coupled to high resolution mass spectrometry. The stability of Ag3PO4 was satisfactory as less than 5% Ag metal appeared in XRD analyses after 3 reuse cycles. These results show that under optimized conditions Ag3PO4 can efficiently achieve quinolone yellow mineralization.


Introduction
Advanced Oxidation Processes (AOPs), especially heterogeneous photocatalysis, is an environmentally friendly technique [1] that is used for pollutants elimination in water and air [2], bacterial inactivation [3], reduction of heavy metals to less harmful species and water splitting [4]. Photocatalysis initiates oxidation reactions through the generation of holes (h + ) and electrons (e − ) pairs. The resulting radicals act as oxidants of organic pollutants and convert them into CO 2 , H 2 O, and inorganic salts [5]. For this reason, extensive investigation was directed toward semiconducting materials that have positive points like the environmental safety and facile preparation. On the other hand, popular photocatalysts such as ZnO [6], TiO 2 [7,8], and SnO 2 are attractive because of their low cost, environmental friendliness [4], thermal stability and bio safe property [9,10]. However, because they possess a wide gap (Eg > 3 eV), their efficacy and sensitivity only occur under UV-light making solar-light application not enough efficient [11,12]. In this respect, active research is directed to the elaboration of catalysts active under visible light for the wastewater treatment [13][14][15].

Characterization of Ag 3 PO 4
The XRD pattern of Ag 3 PO 4 had narrow peaks, characteristic of a high crystallization (Figure 1a). The simulated and the measured diffractograms showed a good fit of 1.6 with R wp and R exp (%) less than 10% (Figure 1b). The microstructural parameters obtained after the Rietveld refinement were used to determine the dislocation density after elaboration. The coherent domains of diffraction (Dv) had an average value of 178 nm and the corresponding microstrain (ε) a percentage of 2.5 × 10 −3 . It is well known that the presence of crystallographic defects, such as dislocations, has a strong influence on the properties of materials. The dislocation density (ρ) calculated from Equations (1)-(3) was ρ = 2.6 × 10 12 in this work. This value is acceptable and confirm the crystal quality. During the elaboration, the dislocation assisted slip activities occurred in the studied material. In addition, the processes used Catalysts 2020, 10, 1461 3 of 15 in this study had no effect on the lattice parameters (a = 6.0316 nm in this work), variation of which is only sensitive to the effects of electronic linkages of the chemical species. ρ = (ρ D ρ ε ) 1/2 (1) where ρ D and ρ ε are calculated using Equations (2) and (3): b the Burger vector, and K is a constant equal to 10 for a Gaussian distribution of the deformations.
Catalysts 2020, 10, x FOR PEER REVIEW 3 of 16 processes used in this study had no effect on the lattice parameters (a = 6.0316 nm in this work), variation of which is only sensitive to the effects of electronic linkages of the chemical species. ρ = (ρDρε) 1/2 (1) where ρD and ρε are calculated using Equations (2) and (3): b the Burger vector, and K is a constant equal to 10 for a Gaussian distribution of the deformations. The SEM micrographs of Ag3PO4 are illustrated in Figure 2. It can be observed that the morphology of the powder was uniform and agglomerated in spherical grains with regular size lying between 1 and 1.5 µm. The particles surface was relatively smooth. The N2 adsorption-desorption isotherm of Ag3PO4 is shown in Figure 3. The isotherm belonged to the type IV according to IUPAC classification, which indicates a well-developed mesoporous nature. In addition, the desorption isotherm showed a H3-hysteresis, characteristic of a capillary condensation observed in mesoporous structures [36]. The detailed textual parameters of Ag3PO4 are gathered in Table 1. The specific surface area and total pore volume were equal to 6 m 2 ·g −1 and 0.0192 cm 3 ·g −1 , respectively. The SEM micrographs of Ag 3 PO 4 are illustrated in Figure 2. It can be observed that the morphology of the powder was uniform and agglomerated in spherical grains with regular size lying between 1 and 1.5 µm. The particles surface was relatively smooth. The N 2 adsorption-desorption isotherm of Ag 3 PO 4 is shown in Figure 3. The isotherm belonged to the type IV according to IUPAC classification, which indicates a well-developed mesoporous nature. In addition, the desorption isotherm showed a H3-hysteresis, characteristic of a capillary condensation observed in mesoporous structures [36]. The detailed textual parameters of Ag 3 PO 4 are gathered in Table 1. The specific surface area and total pore volume were equal to 6 m 2 ·g −1 and 0.0192 cm 3 ·g −1 , respectively.  The formation of Ag3PO4 was further supported by the FTIR spectroscopy, using the routine KBr technique in the 400-4500 cm −1 region. The peaks centered at 545 and 1082 cm −1 (Figure 4) were assigned to molecular vibrations of P-O of the phosphate (PO4 3− ) unit [18], and the bands at 3401 and 1641 cm −1 were attributed to the stretching vibration of O-H groups of adsorbed water. The H2O molecules and OH groups adsorbed on the surface of Ag3PO4 are oxidized into free hydroxyl radicals that achieve the photodegradation of organic pollutants [37]. The intense peak at 1383 cm −1 corresponded to the nitrate impurity, AgNO3 having been used for the preparation of Ag3PO4 [38].
The knowledge of the gap band (Eg) is of great utility in the solar energy conversion; it is evaluated from DRS by using the known relationship: αλ (cm −1 ) is the optical adsorption coefficient and hν (eV) the energy of the incident photon. The nature of the optical allowed transition is deduced from the exponent k (=2, direct) or (0.5, indirect). The gap Eg is deduced from the intersection of the plot (αhν) 2 with the abscissa-axis. The direct     The formation of Ag 3 PO 4 was further supported by the FTIR spectroscopy, using the routine KBr technique in the 400-4500 cm −1 region. The peaks centered at 545 and 1082 cm −1 (Figure 4) were assigned to molecular vibrations of P-O of the phosphate (PO 4 3− ) unit [18], and the bands at 3401 and 1641 cm −1 were attributed to the stretching vibration of O-H groups of adsorbed water. The H 2 O molecules and OH groups adsorbed on the surface of Ag 3 PO 4 are oxidized into free hydroxyl radicals that achieve the photodegradation of organic pollutants [37]. The intense peak at 1383 cm −1 corresponded to the nitrate impurity, AgNO 3 having been used for the preparation of Ag 3 PO 4 [38].   The knowledge of the gap band (E g ) is of great utility in the solar energy conversion; it is evaluated from DRS by using the known relationship: is the optical adsorption coefficient and hν (eV) the energy of the incident photon. The nature of the optical allowed transition is deduced from the exponent k (=2, direct) or (0.5, indirect). The gap E g is deduced from the intersection of the plot (αhν) 2 with the abscissa-axis. The direct transition at 2.52 eV ( Figure 5) was in conformity with the yellow color of Ag 3 PO 4 that is a promising photocatalyst under visible light [20].

Photocatalytic Degradation of QY
Neither the irradiation of QY under UVA for 4 h in absence of Ag 3 PO 4 nor the stirring of QY solution pH 7 in presence of Ag 3 PO 4 in the dark for 24 h led to a significant concentration loss. These results demonstrated that direct photolysis and adsorption on the photocatalyst were negligible in these conditions. The very poor adsorption of QY on Ag 3 PO 4 in neutral medium can be explained by electrostatic repulsion between the negatively charged surface of the photocatalyst and anionic QY as observed for methyl orange and orange G [39]. In contrast, when the suspension containing Ag 3 PO 4 and QY was exposed to UVA or visible light, an almost complete degradation was achieved within 40 min and 70 min under visible light and UVA, respectively ( Figure 6). Experiments were then undertaken to optimize parameters.

Parameters Optimization
In this set of experiments, Ag3PO4 suspensions were irradiated with visible light. The degradation profile of QY was monitored until 40 min. The QY decrease rate, r, obeyed an apparent pseudo-first order model (r = k × C); therefore, the degradation rate constants, k, were obtained from the slope of the plots ln(C/C0) vs. irradiation time.
The photocatalytic degradation of QY (20 ppm) by Ag3PO4 (0.5 g·L −1 ) was studied by varying pH in the range of 4-11. As shown in Figure 7, the degradation rate of QY was pH-dependent. The highest rate constant (0.094 min −1 ) was observed at pH 7. The rate constant drastically decreased below pH 7 by 65-90% while by 40-50% above ( Figure 8). This highest efficiency at neutral pH, close to the natural environment, is a very positive point in terms of application.

Parameters Optimization
In this set of experiments, Ag 3 PO 4 suspensions were irradiated with visible light. The degradation profile of QY was monitored until 40 min. The QY decrease rate, r, obeyed an apparent pseudo-first order model (r = k × C); therefore, the degradation rate constants, k, were obtained from the slope of the plots ln(C/C 0 ) vs. irradiation time.
The photocatalytic degradation of QY (20 ppm) by Ag 3 PO 4 (0.5 g·L −1 ) was studied by varying pH in the range of 4-11. As shown in Figure 7, the degradation rate of QY was pH-dependent. The highest rate constant (0.094 min −1 ) was observed at pH 7. The rate constant drastically decreased below pH 7 by 65-90% while by 40-50% above ( Figure 8). This highest efficiency at neutral pH, close to the natural environment, is a very positive point in terms of application.    The reaction was also studied at QY concentrations in the range of 5-40 ppm, at a fixed dose of Ag 3 PO 4 (0.5 g·L −1 ) and at pH 7. Between 5 and 20 ppm, the rate constant laid between 0.065 and 0.072 min −1 , while at 30 and 40 ppm the value dropped to 0.02 and 0.011 min −1 , respectively. At 40 ppm, the absorbance of the QY solution was equal to 3.2 at 434 nm, QY had therefore a very strong filter effect and prevents Ag 3 PO 4 to absorb light.
The Ag 3 PO 4 photocatalyst dose was varied in the range of 0.25-1 g·L −1 to avoid the photocatalyst excess and ensure a total absorption of efficient incident photons. Figure 9 shows that the degradation rate increased with the catalyst dose and reached a maximum for 0.5 g·L −1 . The rate constant k was equal to 8.6 min −1 . An increase in the catalyst dose up to 1.5 g·L −1 led to a reduced photoactivity due to a lower availability of photoactive sites on the Ag 3 PO 4 surface and a weak light utilization [40].

Compared Photocatalytic Activity Ag 3 PO 4 /TiO 2 Degussa P25
The photocatalytic efficiency Ag 3 PO 4 and TiO 2 Degussa P25 ( Table 2) were compared at the dose of 0.5 g·L −1 . Under visible light, Ag 3 PO 4 was active while TiO 2 non active and, under UVA, Ag 3 PO 4 was 1.6-times more efficient that TiO 2 . This demonstrates the interest of Ag 3 PO 4 as a photocatalyst.

Product Studies
HPLC and UHPLC-ESI-HRMS analyses were conducted for products identification. QY is sold by Sigma-Aldrich as a mixture of mono and disulfonic acids C 18 H 12 NO 5 SNa-C 18 H 11 NO 8 S 2 Na 2 . Yet, we only detected two peaks at m/z = 353.0352 in ES − . They correspond to two isomeric forms of the monosulfonic acid (Figure 10a,b). Three samples were analyzed after 44, 70 and 175 min of irradiation under UVA. In the first two samples, the QY conversion extent was of 50 and 80%, and in the third one QY was fully degraded (Figure 10c). A typical HPLC chromatogram is shown Figure 10c and Table 3 gives the accurate and experimental m/z, the chemical formula in the limit of 5 ppm and the possible structure of the 8 detected photoproducts (TPs) by UHPLC-ESI-HRMS. In general, TPs were detected by negative electro spray ionization.    by Sigma-Aldrich as a mixture of mono and disulfonic acids C18H12NO5SNa-C18H11NO8S2Na2. Yet, we only detected two peaks at m/z = 353.0352 in ES − . They correspond to two isomeric forms of the monosulfonic acid (Figure 10a,b). Three samples were analyzed after 44, 70 and 175 min of irradiation under UVA. In the first two samples, the QY conversion extent was of 50 and 80%, and in the third one QY was fully degraded (Figure 10c). A typical HPLC chromatogram is shown Figure 10c and Table 3 gives the accurate and experimental m/z, the chemical formula in the limit of 5 ppm and the possible structure of the 8 detected photoproducts (TPs) by UHPLC-ESI-HRMS. In general, TPs were detected by negative electro spray ionization. by Sigma-Aldrich as a mixture of mono and disulfonic acids C18H12NO5SNa-C18H11NO8S2Na2. Yet, we only detected two peaks at m/z = 353.0352 in ES − . They correspond to two isomeric forms of the monosulfonic acid (Figure 10a,b). Three samples were analyzed after 44, 70 and 175 min of irradiation under UVA. In the first two samples, the QY conversion extent was of 50 and 80%, and in the third one QY was fully degraded (Figure 10c). A typical HPLC chromatogram is shown Figure 10c and Table 3 gives the accurate and experimental m/z, the chemical formula in the limit of 5 ppm and the possible structure of the 8 detected photoproducts (TPs) by UHPLC-ESI-HRMS. In general, TPs were detected by negative electro spray ionization. by Sigma-Aldrich as a mixture of mono and disulfonic acids C18H12NO5SNa-C18H11NO8S2Na2. Yet, we only detected two peaks at m/z = 353.0352 in ES − . They correspond to two isomeric forms of the monosulfonic acid (Figure 10a,b). Three samples were analyzed after 44, 70 and 175 min of irradiation under UVA. In the first two samples, the QY conversion extent was of 50 and 80%, and in the third one QY was fully degraded (Figure 10c). A typical HPLC chromatogram is shown Figure 10c and Table 3 gives the accurate and experimental m/z, the chemical formula in the limit of 5 ppm and the possible structure of the 8 detected photoproducts (TPs) by UHPLC-ESI-HRMS. In general, TPs were detected by negative electro spray ionization. only detected two peaks at m/z = 353.0352 in ES − . They correspond to two isomeric forms of the monosulfonic acid (Figure 10a,b). Three samples were analyzed after 44, 70 and 175 min of irradiation under UVA. In the first two samples, the QY conversion extent was of 50 and 80%, and in the third one QY was fully degraded (Figure 10c). A typical HPLC chromatogram is shown Figure 10c and Table 3 gives the accurate and experimental m/z, the chemical formula in the limit of 5 ppm and the possible structure of the 8 detected photoproducts (TPs) by UHPLC-ESI-HRMS. In general, TPs were detected by negative electro spray ionization. only detected two peaks at m/z = 353.0352 in ES − . They correspond to two isomeric forms of the monosulfonic acid (Figure 10a,b). Three samples were analyzed after 44, 70 and 175 min of irradiation under UVA. In the first two samples, the QY conversion extent was of 50 and 80%, and in the third one QY was fully degraded (Figure 10c). A typical HPLC chromatogram is shown Figure 10c and Table 3 gives the accurate and experimental m/z, the chemical formula in the limit of 5 ppm and the possible structure of the 8 detected photoproducts (TPs) by UHPLC-ESI-HRMS. In general, TPs were detected by negative electro spray ionization. only detected two peaks at m/z = 353.0352 in ES − . They correspond to two isomeric forms of the monosulfonic acid (Figure 10a,b). Three samples were analyzed after 44, 70 and 175 min of irradiation under UVA. In the first two samples, the QY conversion extent was of 50 and 80%, and in the third one QY was fully degraded (Figure 10c). A typical HPLC chromatogram is shown Figure 10c and Table 3 gives the accurate and experimental m/z, the chemical formula in the limit of 5 ppm and the possible structure of the 8 detected photoproducts (TPs) by UHPLC-ESI-HRMS. In general, TPs were detected by negative electro spray ionization.  TP1 showed m/z = 208.0074 corresponding to C9H7O3NS for the neutral molecule. This TP was likely quinoline sulfonic acid formed by C-C cleavage and its detection allowed us to locate the sulfonic group of QY on the quinoline ring. With m/z = 251.9977, TP2 had the molecular formula of C10H7O5N and corresponded to TP1 + CO2. It was, therefore, likely a carboxylated quinoline sulfonic acid formed by cleavage of the 5-carbon ring. TP3 with m/z = 149.0229 corresponded to a molecular formula of C8H6O3. On the other hand, TP4 that showed m/z = 165.018 corresponded to C8H6O4 and most probably to phthalic acid. TP3 might be therefore the corresponding aldehyde. TP5 was identified to benzoic acid (m/z = 121.0286 and C7H6O2). TP6 and TP7 with m/z = 89.0231 and 87.0074 were two aliphatic acids with 3 atoms of carbon produced by oxidation and opening of the rings.   TP1 showed m/z = 208.0074 corresponding to C9H7O3NS for the neutral molecule. This TP was likely quinoline sulfonic acid formed by C-C cleavage and its detection allowed us to locate the sulfonic group of QY on the quinoline ring. With m/z = 251.9977, TP2 had the molecular formula of C10H7O5N and corresponded to TP1 + CO2. It was, therefore, likely a carboxylated quinoline sulfonic acid formed by cleavage of the 5-carbon ring. TP3 with m/z = 149.0229 corresponded to a molecular formula of C8H6O3. On the other hand, TP4 that showed m/z = 165.018 corresponded to C8H6O4 and most probably to phthalic acid. TP3 might be therefore the corresponding aldehyde. TP5 was identified to benzoic acid (m/z = 121.0286 and C7H6O2). TP6 and TP7 with m/z = 89.0231 and 87.0074 were two aliphatic acids with 3 atoms of carbon produced by oxidation and opening of the rings.

TP8
61.9873 61.9868 −4.6 NO 3 − TP1 showed m/z = 208.0074 corresponding to C 9 H 7 O 3 NS for the neutral molecule. This TP was likely quinoline sulfonic acid formed by C-C cleavage and its detection allowed us to locate the sulfonic group of QY on the quinoline ring. With m/z = 251.9977, TP2 had the molecular formula of C 10 H 7 O 5 N and corresponded to TP1 + CO 2 . It was, therefore, likely a carboxylated quinoline sulfonic acid formed by cleavage of the 5-carbon ring. TP3 with m/z = 149.0229 corresponded to a molecular formula of C 8 H 6 O 3 . On the other hand, TP4 that showed m/z = 165.018 corresponded to C 8 H 6 O 4 and most probably to phthalic acid. TP3 might be therefore the corresponding aldehyde. TP5 was identified to benzoic acid (m/z = 121.0286 and C 7 H 6 O 2 ). TP6 and TP7 with m/z = 89.0231 and 87.0074 were two aliphatic acids with 3 atoms of carbon produced by oxidation and opening of the rings. Nitrate ions (m/z = 62.0049) were also detected. TP1-TP5 were detected after 44 and 70 min of irradiation while TP6, TP7 and nitrate ions in the samples irradiated for 175 min. The degradation pathways are summarized in Figure 11. Catalysts 2020, 10, x FOR PEER REVIEW 10 of 16 Figure 11. Proposed degradation mechanism for QY in Ag3PO4 process.

Mineralization
TOC evolution during the photocatalytic reaction was monitored to explore the percentage of QY mineralization under the optimal conditions (QY 20 ppm, Ag3PO4 dose =0.5 g·L −1 and pH 7) ( Figure 12). After 70 min of irradiation under UVA corresponding to the full degradation of QY, only 5% of OC were transformed into CO2. Total OC removal was reached after 48 h of irradiation time. This long time to achieve complete mineralization indicates that intermediates were difficult to oxidize.

Catalyst Stability and Reusability
To assess the photostability and reusability of Ag3PO4 under visible irradiation for the photodegradation of QY, three photodegradation experiments were successively performed. After each cycle, the catalyst was centrifuged, filtered, thoroughly rinsed with water and ethanol, and finally dried at 70 °C. After three recycles, 87, 78 and 64% of QY were photodegraded (Figure 13a) within 40 min. These results revealed some deactivation effects. This reduction can be explained by the reduction of Ag + to metallic state Ag in the presence of photogenerated electrons. To quantify the percentage of photoreduced Ag + , both XRD and AAS analyses were carried out (Table 4). XRD results after the three cycles are shown in Figure 13b. Small peaks of Ag appeared (<5%) [20]. On the other hand, results of the literature showed that, when dispersed in water and irradiated under visible light Figure 11. Proposed degradation mechanism for QY in Ag 3 PO 4 process.

Mineralization
TOC evolution during the photocatalytic reaction was monitored to explore the percentage of QY mineralization under the optimal conditions (QY 20 ppm, Ag 3 PO 4 dose =0.5 g·L −1 and pH 7) ( Figure 12). After 70 min of irradiation under UVA corresponding to the full degradation of QY, only 5% of OC were transformed into CO 2 . Total OC removal was reached after 48 h of irradiation time. This long time to achieve complete mineralization indicates that intermediates were difficult to oxidize.
Catalysts 2020, 10, x FOR PEER REVIEW 10 of 16 Figure 11. Proposed degradation mechanism for QY in Ag3PO4 process.

Mineralization
TOC evolution during the photocatalytic reaction was monitored to explore the percentage of QY mineralization under the optimal conditions (QY 20 ppm, Ag3PO4 dose =0.5 g·L −1 and pH 7) ( Figure 12). After 70 min of irradiation under UVA corresponding to the full degradation of QY, only 5% of OC were transformed into CO2. Total OC removal was reached after 48 h of irradiation time. This long time to achieve complete mineralization indicates that intermediates were difficult to oxidize.

Catalyst Stability and Reusability
To assess the photostability and reusability of Ag3PO4 under visible irradiation for the photodegradation of QY, three photodegradation experiments were successively performed. After each cycle, the catalyst was centrifuged, filtered, thoroughly rinsed with water and ethanol, and finally dried at 70 °C. After three recycles, 87, 78 and 64% of QY were photodegraded (Figure 13a) within 40 min. These results revealed some deactivation effects. This reduction can be explained by the reduction of Ag + to metallic state Ag in the presence of photogenerated electrons. To quantify the percentage of photoreduced Ag + , both XRD and AAS analyses were carried out (Table 4). XRD results after the three cycles are shown in Figure 13b. Small peaks of Ag appeared (<5%) [20]. On the other hand, results of the literature showed that, when dispersed in water and irradiated under visible light

Catalyst Stability and Reusability
To assess the photostability and reusability of Ag 3 PO 4 under visible irradiation for the photodegradation of QY, three photodegradation experiments were successively performed. After each cycle, the catalyst was centrifuged, filtered, thoroughly rinsed with water and ethanol, and finally dried at 70 • C. After three recycles, 87, 78 and 64% of QY were photodegraded (Figure 13a) within 40 min. These results revealed some deactivation effects. This reduction can be explained by the reduction of Ag + to metallic state Ag in the presence of photogenerated electrons. To quantify the percentage of photoreduced Ag + , both XRD and AAS analyses were carried out (Table 4). XRD results after the three cycles are shown in Figure 13b. Small peaks of Ag appeared (<5%) [20]. On the other hand, results of the literature showed that, when dispersed in water and irradiated under visible light for a long time, Ag 3 PO 4 remained stable (dissolution < 1 ppm) [41]. Atomic absorption spectroscopy analyses confirmed this result (Table 4).

Synthesis of Ag3PO4
Ag3PO4 was prepared by precipitation as previously described [20]. Briefly, 15 mL of a Na2HPO4 solution (3 mM) was mixed dropwise with 30 mL of AgNO3 solution (3 mM) under magnetic stirring during 3 h at room temperature. The precipitate with a yellow color characteristic of Ag3PO4, was

Synthesis of Ag 3 PO 4
Ag 3 PO 4 was prepared by precipitation as previously described [20]. Briefly, 15 mL of a Na 2 HPO 4 solution (3 mM) was mixed dropwise with 30 mL of AgNO 3 solution (3 mM) under magnetic stirring during 3 h at room temperature. The precipitate with a yellow color characteristic of Ag 3 PO 4 , was collected and washed with deionized water and ethanol. The yellow powder was recovered after drying at 70 • C under static vacuum overnight.

Characterization of Ag 3 PO 4
The synthesized phase was confirmed by X-ray diffraction using Co Kα line (λ = 1.790300 Å) at a 2θ scan rate of 2 per min. The micrograph of the sample was examined by SEM analysis using a JEOL microscope (Model JSM 6360-LV). The BET surface area and pore volume were calculated from the N 2 adsorption/desorption isotherm at 77 K using an ASAP 2010 Micrometrics equipment. The external surface area, micropore area and micropore volume were computed by the t-plot technique. The total pore volume was determined from the liquid N 2 volume at high relative pressure (0.99). The subtraction of the micropore volume from the total volume gave the volume of mesopore. The DRS was measured using a spectrophotometer (Specord 200 Plus). The concentration of Ag before and after recycling in hydrolyzed samples was determined using Atomic Absorption Spectrophotometry (AAS) (Perkin-Elmer A700). A calibration curve was created by analyzing Ag (1, 2, 3, 4 µg·mL −1 ) standard solutions, and the instrument was adjusted to 328.1 nm.

Photocatalytic Experiments
Irradiations were conducted in two devices; one was equipped with fluorescent tubes (Philips HPK 15 W) emitting in the near UV region (maximum at 365 nm) while the other with LED (Delleled 5 W) emitting within the wavelength range 400-800 nm. Pyrex reactors were connected to a thermostatically controlled water bath. For irradiation under visible 50 mL of suspensions containing Ag 3 PO 4 and QY were added to the reactor of 100 mL of capacity while for UVA irradiation 15 mL of suspensions were placed in the reactor of 30 mL capacity. Solution pH was adjusted using NaOH or HClO 4 . In both systems, the suspensions were kept at 25 • C under constant agitation (600 rpm). Aliquots of 0.5 mL were regularly withdrawn, vigorously centrifuged and filtered through a 0.45 µm filter (Waters, Millipore) before analyses. The degradation yield (%) was evaluated from the relationship: C 0 being the initial QY concentration and C t the concentration at time (t).

Analytical Analyses
A UV-vis spectrophotometer (Model Specord 200 plus) was used to monitor the degradation of QY (λ max = 434 nm). QY and its photoproducts were separated using a Shimadzu 8040 HPLC system, equipped with a SPD-M30A DAD detector, and a Nucleodur 100-3 C 18 end-capped column (125 mm × 4.0 mm, 3 µm particle size). The mobile phase was a mixture of (A) 30 mmol·L −1 sodium acetate and (B) acetonitrile in the proportions of 80% A and 20% B, at a constant flow rate of 0.75 mL·min −1 . The detection wavelength was set at 290 nm and the temperature oven at 30 • C.
UHPLC-ESI-HRMS was performed on an Orbitrap Q-Exactive (Thermo Scientific, Waltham, MA, USA) mass spectrometer coupled to an Ultimate 3000 RSLC (Thermo Scientific, Waltham, MA, USA) apparatus. The mass spectrometer with heated electrospray ionization was operated in both positive and negative ESI modes with a spray voltage of 3 kV for both modes. The column was a Waters analytical column (2.1 × 100 mm, 1.7 µm particle size) and the column oven temperature was regulated at 30 • C. The aqueous solvent (A) was a mixture of formic acid (0.1%) and acetonitrile as organic phase (B). The separation was achieved using a gradient program consisting of 0-7.5 min 5%, 7.5-8.5 min 99% of the mobile phase B. The injection volume was taken at 10 µL while the flow rate was fixed at 0.450 mL·min −1 .
A Shimadzu 5050 TOC analyzer was used to monitor the mineralization of QY. The percentage of mineralization was calculated using the following equation: where OC 0 is the organic carbon present in the solution at time 0 and OC t at time t. OC is obtained by subtracting the inorganic carbon from total carbon.

Conclusions
Ag 3 PO 4 was synthesized by precipitation at ambient temperature. It was shown to induce the photocatalytic oxidation of Quinoline Yellow under visible irradiation with an efficiency much higher than that of TiO 2 Degussa P25 under UVA. The optimal parameters such as initial pH, QY concentration and catalyst dose were determined. Photoproducts of QY were identified based on the UHPLC-ESI-HRMS analysis and were different than those identified using TiO 2 . The complete mineralization of QY was be achieved after 48 h of irradiation against 70 min for QY elimination. Some XRD peaks of Ag metal (~5%) were noticeable after three reuses. It can be concluded that Ag 3 PO 4 needs efficient hetero-junction to avoid its photocorrosion. This is the subject of our future investigations.