Synthesis of an Ag 3 PO 4 /Nb 2 O 5 Photocatalyst for the Degradation of Dye

: In this work, the photocatalytic performance of Ag 3 PO 4 , Nb 2 O 5 and Ag 3 PO 4 /Nb 2 O 5 hybrid photocatalysts to degrade methyl orange dye, MO, in an aqueous solution under visible light irradiation was evaluated. The Ag 3 PO 4 and Ag 3 PO 4 /Nb 2 O 5 photocatalysts, with various Ag to Nb molar ratios, were prepared using a facile precipitation method. The photocatalysts were characterized by X-ray diffraction, UV–Visible, X-ray Photoelectron, and Photoluminescence spectroscopies. Upon the addition of Ag 3 PO 4 , the band gap energy of Nb 2 O 5 decreased from 3.0 eV to 2.7 eV, indicating the possible use of the Ag 3 PO 4 /Nb 2 O 5 hybrid photocatalysts under visible light irradiation. All of the prepared Ag 3 PO 4 /Nb 2 O 5 catalysts exhibited higher photocatalytic performance than Ag 3 PO 4 in degrading methyl orange dye under 23-watt visible light irradiation. The Ag 3 PO 4 /Nb 2 O 5 catalyst, with a mole ratio of 2:1, exhibited the fastest MO degradation rate of 7.3 × 10 − 2 min − 1 , which is twice faster than that of Ag 3 PO 4 . The catalyst also shows better stability, as it is reusable for up to six experimental cycles while maintaining its photocatalytic activity above 60%. the gap Tauc The band gap energy of Nb 2 O 5 (3.3 eV) and Ag 3 PO 4 (2.3 eV) is similar to the values reported in the literature [3,12]. the of


Introduction
Dyes, widely used in the textile, paint, ink, and paper industries, are the most significant chemical contaminants that cause water pollution [1]. Up to 200,000 tons of dyes are lost to effluents in textile industries every year during dyeing and finishing operations. The release of colored water in the effluents causes hazards and environmental problems. Since its discovery by Fujishima and Honda in 1972, heterogeneous photocatalysis has been widely explored as a water decontamination method [1]. Through heterogeneous photocatalysis, the complete mineralization of parents and their intermediate compounds is possible at ambient operating temperatures and pressures, and at low operational costs [2]. Nowadays, visible-light-driven photocatalysis has gained research interest, with Ag 3 PO 4 being reported as a promising photocatalyst for the photodegradation of the aqueous dye solution. Its ability to utilize visible light irradiation is due to its narrow band gap energy. Unfortunately, self-photo corrosion is a significant problem, which means that the stability of Ag 3 PO 4 remains an issue [3]. One of the initiatives to enhance the photocatalytic activity and the stability of the Ag 3 PO 4 photocatalyst is by coupling it with other semiconductors, such as TiO 2 [4][5][6], ZnO [7], and CeO 2 [8]. This coupling method decreased the recombination probability of the electron-hole pairs, and enhanced the photocatalytic efficiency. Niobium (V) oxide, Nb 2 O 5 , which has a band gap energy (3.4 eV), which is similar to TiO 2 , is one of the promising photocatalysts, as it exhibits excellent photocatalytic performance in the photodegradation of dyes [9]. Although Souza et al. (2016) reported similar catalytic activity between Nb 2 O 5 and TiO 2 -P25 in the degradation of textile wastewater [10], Prado and his co-workers (2008) showed that Nb 2 O 5 is a more stable photocatalyst as it successfully maintained its 85% catalytic activity in indigo carmine degradation after ten reaction cycles [11].
This work aimed to produce a stable visible-light-driven photocatalyst by coupling Ag 3 PO 4 with Nb 2 O 5 . There are very few works on Ag 3 PO 4 /Nb 2 O 5 photocatalysts available in the literature [3]. To the best of our knowledge, using a fluorescence lamp (23 W) as a source of light irradiation to degrade a dye over Ag 3 PO 4 /Nb 2 O 5 photocatalysts has not been reported. Previously, the degradation of rhodamine B over Ag 3 PO 4 /Nb 2 O 5 under simulated sunlight irradiation (Xe, 600 W) was reported [3]. In this work, Ag 3 PO 4 /Nb 2 O 5 photocatalysts were synthesized via the deposition-precipitation method, and their photocatalytic activity was evaluated by the photodegradation of methyl orange (MO) under fluorescent light (23 W) irradiation. The experiments were performed under various parameters in order to determine the best-suited conditions for the degradation of MO. The effect of scavengers and a reusability test were conducted in order to determine the active species responsible for the degradation, and to evaluate the stability of the catalysts, respectively   The optical response of the catalysts in the visible light region was determined using the UV-Vis DRS spectrum. The data collected was transformed using the Kubelka-Munk function, and the band gap energy was extrapolated from the intercept of the x-axis of the Tauc plot ( Figure 2). The band gap energy of Nb 2 O 5 (3.3 eV) and Ag 3 PO 4 (2.3 eV) is similar to the values reported in the literature [3,12]. Upon the addition of Ag 3 PO 4 , the band gap energy of Nb 2 O 5 slightly decreased to 3.2 eV.

Characteristics of the Photocatalysts
lysts 2021, 11, x FOR PEER REVIEW 3 o The optical response of the catalysts in the visible light region was determined us the UV-Vis DRS spectrum. The data collected was transformed using the Kubelka-Mu function, and the band gap energy was extrapolated from the intercept of the x-axis of Tauc plot ( Figure 2). The band gap energy of Nb2O5 (3.3 eV) and Ag3PO4 (2.3 eV) is sim to the values reported in the literature [3,12]. Upon the addition of Ag3PO4, the band g energy of Nb2O5 slightly decreased to 3.2 eV.
The optical properties and band structure of the samples were further evalua using steady-state photoluminescence analysis. The results are presented in Figure 3. the Nb2O5, two broad peaks ranged from (i) the UV region: 350 nm to 400 nm, and (ii) visible region: 400 nm to 550 nm. From Figure 3b, the first deconvoluted peak (blue) be ascribed to the near band edge emission of the bulk Nb2O5 samples, whereas the seco peak (green) can be referred to as the defect-assisted band alignment transition [13]. T Ag3PO4 sample also showed a broad emission peak from 400 nm to 600 nm, which can deconvoluted into four peaks via Gaussian function analysis. The peaks were located the wavelength 425 nm (violet), 480 nm (blue), 525 (green), and 575 nm (green). This res is consistent with the literature, stating that the carrier recombination arose from shallow defect between the [PO4] and highly distorted tetrahedral [AgO4] clusters [1 Upon the incorporation of Ag3PO4 into the Nb2O5 at a 1:1 molar ratio, it was found t the PL emission at the visible region from 400 nm to 600 nm (shallow defect) w quenched. This phenomenon signifies that the carrier recombination becomes m dominant in the heterojunction structure than the direct band edge at UV region of Nb (blue region in Figure 3b). In order to further evalutate the peak quenching, a normali peak with respect to the direct band edge of Nb2O5 (382 nm) was conducted for all of Ag3PO4/Nb2O5 and Nb2O5 samples. The result is presented in Figure 3d. It can be observ that the further increases of the Ag content in Nb2O5 have proved to suppress the intensity further in the shallow defcet states, as compared to the Ag3PO4 sample. None the defect states were observed in the Nb2O5 semiconductor material. This observat may be atttributed to the agglomeration of the Ag3PO4 cluster, which contributed to bulk trapping states. It is also worth mentioning that the increased Ag concentration alter the charged polarization between the distorted Ag cluster, and could eventua promote intrinsic and extrinsic defect trapping states in the electronic structure of Nb2O5 semiconductor materials. The plausible charge transfer mechanism is illustarted The optical properties and band structure of the samples were further evaluated using steady-state photoluminescence analysis. The results are presented in Figure 3. For the Nb 2 O 5 , two broad peaks ranged from (i) the UV region: 350 nm to 400 nm, and (ii) the visible region: 400 nm to 550 nm. From Figure 3b, the first deconvoluted peak (blue) can be ascribed to the near band edge emission of the bulk Nb 2 O 5 samples, whereas the second peak (green) can be referred to as the defect-assisted band alignment transition [13]. The Ag 3 PO 4 sample also showed a broad emission peak from 400 nm to 600 nm, which can be deconvoluted into four peaks via Gaussian function analysis. The peaks were located at the wavelength 425 nm (violet), 480 nm (blue), 525 (green), and 575 nm (green). This result is consistent with the literature, stating that the carrier recombination arose from the shallow defect between the [PO 4 ] and highly distorted tetrahedral [AgO 4 ] clusters [14]. Upon the incorporation of Ag 3 PO 4 into the Nb 2 O 5 at a 1:1 molar ratio, it was found that the PL emission at the visible region from 400 nm to 600 nm (shallow defect) was quenched. This phenomenon signifies that the carrier recombination becomes more dominant in the heterojunction structure than the direct band edge at UV region of Nb 2 O 5 (blue region in Figure 3b). In order to further evalutate the peak quenching, a normalised peak with respect to the direct band edge of Nb 2 O 5 (382 nm) was conducted for all of the Ag 3 PO 4 /Nb 2 O 5 and Nb 2 O 5 samples. The result is presented in Figure 3d. It can be observed that the further increases of the Ag content in Nb 2 O 5 have proved to suppress the PL intensity further in the shallow defcet states, as compared to the Ag 3 PO 4 sample . None of the defect states were observed in the Nb 2 O 5 semiconductor material. This observation may be atttributed to the agglomeration of the Ag 3 PO 4 cluster, which contributed to the bulk trapping states. It is also worth mentioning that the increased Ag concentration can alter the charged polarization between the distorted Ag cluster, and could eventually promote intrinsic and extrinsic defect trapping states in the electronic structure of the Nb 2 O 5 semiconductor materials. The plausible charge transfer mechanism is illustarted in Figure 3e. For the Nb 2 O 5 sample, a charge trasfer only occured between the valance band and the conduction band of the semiconductor material. For the Ag 3 PO 4 /Nb 2 O 5 sample, a charge carrier was photoexcited from the valance band to the conduction band of Nb 2 O 5 under the simulated sunlight. The charge carrier was later transferred to the conduction band of Ag 3 PO 4 before it further underwent the reduction process with the analyte during the photodegradation process. Therefore, it is believed that the Ag 3 PO 4 /Nb 2 O 5 semiconductor materials served as better photocatalysts than Nb 2 O 5, owing to its two charge transfer pathways, which lower the direct carrier recombination in the Nb 2 O 5 semiconductor.  (Figure 4c) are assigned to Ag 3d 5/2 and Ag 3d 3/2 , indicating the existence of Ag + in the composite [15]. The peak of P 2p, observed at 132.6 eV (Figure 4d), is due to the P 5+ of the phosphate. The deconvolution of the O 1s spectrum (Figure 4e) resulted in two peaks at 531.0 eV and 532.8 eV. The former corresponds to a crystal lattice oxygen of Ag 3 PO 4 [16,17] and Nb 2 O 5 [18,19]; the latter could be derived from the hydroxyl group on the surface of the catalyst. The high resolution of Nb 3d (Figure 4f) peaked for Nb 3d 5/2 at 207.3 eV, and Nb 3d 3/2 at 210.0 eV corresponds to Nb 5+ , in good agreement with the binding energy of Nb 2 O 5 [20].   The BET surface area and particle size determination of the Ag 3 PO 4, Nb 2 O 5, and Ag 3 PO 4 /Nb 2 O 5 composite photocatalysts were analyzed by N 2 adsorption/desorption and Zetasizer; the data are summarized in Table 2. Based on the results, as the Ag 3 PO 4 loading increased in the composites, a larger aggregate particle size was clearly observed compared to pristine Ag 3 PO 4 and Nb 2 O 5 .    The incorporation of Ag 3 PO 4 into Nb 2 O 5 resulted in the reduction of the surface area, which is attributed to the method of preparation. When contacted with the Nb 2 O 5 , the Ag + ions would be adsorbed onto the Nb 2 O 5 surface, possibly near the mouth of the pore or in the pore of Nb 2 O 5 . When the phosphate solution was added into the solution mixture, Ag + would react to precipitate Ag 3 PO 4 on the Nb 2 O 5 . The formation of Ag 3 PO 4 in the pore or on the mouth of the pore of Nb 2 O 5 would block the pore, hence reducing the accessibility of the N 2 gas to the adsorption sites. Consequently, the prepared catalysts exhibited a lower surface area compared to pristine Nb 2 O 5 and Ag 3 PO 4 .

Photocatalytic Activity
The photocatalytic activity of the prepared photocatalysts was evaluated by the photodegradation of methyl orange, and the results are as shown in Figure 6. The degradation of MO by Nb 2 O 5 was insignificant because it is inactive under visible light irradiation due to its wide band gap energy. The Ag 3 PO 4 and Ag 3 PO 4 /Nb 2 O 5 photocatalysts, however, exhibited a similar percentage of MO degradation, ranging from 91 to 96%. The experiment data was then fitted to the pseudo-first-order kinetic in order to determine the rate constant for the reaction. The MO degradation rate was faster when using Ag 3 PO 4 /Nb 2 O 5 catalysts. The reaction rate initially increased with increasing Ag 3 PO 4 loading up to a ratio of 2:1 before it decreased at a higher loading (3:1), which could be due to the agglomeration of Several factors influence photocatalytic efficiency, including the mass of the photocatalyst and the concentration of the pollutant. Figure 7 shows the photodegradation activity using a various amount of Ag:Nb = 2:1 photocatalyst and the initial concentration of MO. The MO degradation rate increases with increasing catalyst mass, up to 0.5 g, before it decreases at a higher catalyst mass (Figure 7a). The use of an excessive amount of catalyst loading reduces the degradation activity by increasing the Several factors influence photocatalytic efficiency, including the mass of the photocatalyst and the concentration of the pollutant. Figure 7 shows the photodegradation activity using a various amount of Ag:Nb = 2:1 photocatalyst and the initial concentration of MO. The MO degradation rate increases with increasing catalyst mass, up to 0.5 g, before it decreases at a higher catalyst mass (Figure 7a). The use of an excessive amount of catalyst loading reduces the degradation activity by increasing the light scattering effect and the opacity of the solution [21]. It also leads to the aggregation of the catalyst and reduces the interfacial area between the reaction solution and the photocatalyst. The fastest rate of degradation (Figure 7b) was observed when degrading 5 ppm of MO. The MO degradation rate, however, decreased with the increasing MO initial concentration. This decrement may occur due to the light screening effect, which reduces the photon's path length and the number of photons that arrive on the surface of the catalyst. Thus, it reduces the generation of electron-hole pairs, and consequently the photocatalytic activity of the photocatalyst [22]. In order to determine the species responsible for the photodegradation, 10 mL of a scavenger solution-such as ethylene diamine tetraacetic acid (EDTA), tert-butanol, or benzoquinone (BQ)-was added to the MO solution. These solutions scavenge the h + , •OH, and •O 2 radicals, respectively. In the absence of the scavengers, the photocatalytic degradation of MO was 96% (Figure 8). Upon the addition of the EDTA and BQ solutions, the photodegradation decreased significantly to 1% and 32%, respectively. However, there was no significant effect on the photodegradation efficiency observed with the ad- In order to determine the species responsible for the photodegradation, 10 mL of a scavenger solution-such as ethylene diamine tetraacetic acid (EDTA), tert-butanol, or benzoquinone (BQ)-was added to the MO solution. These solutions scavenge the h + , •OH, and •O − 2 radicals, respectively. In the absence of the scavengers, the photocatalytic degradation of MO was 96% (Figure 8). Upon the addition of the EDTA and BQ solutions, the photodegradation decreased significantly to 1% and 32%, respectively. However, there was no significant effect on the photodegradation efficiency observed with the addition of tert-butanol. These results demonstrate that the h + and •O − 2 species are responsible for the photodegradation of MO.

Reusability of the Ag3PO4/Nb2O5 Photocatalyst
The catalyst's lifetime is an essential parameter of the photocatalytic degradation process, so it is crucial to evaluate the catalyst's reusability for practical applications. As shown in Figure 9, the decrease in the photocatalytic efficiency of the Ag:Nb = 2:1 photocatalyst was gradual, but remained above 60% up to the fifth cycle of the photodegradation experiment. This gradual loss might be due to the self-photo corrosion of Ag3PO4. The enhanced photocatalytic activity and stability of the Ag:Nb = 2:1 photocatalyst may be due to the insoluble Nb2O5 layer, which can effectively protect the dissolution of Ag3PO4 to Ag + ions, thus improving the structural stability of the photocatalyst.

The Synthesis of the Ag3PO4/Nb2O5 Photocatalyst
Ag3PO4/Nb2O5 photocatalysts with different Ag3PO4 to Nb2O5 mole ratios were prepared via a facile deposition-precipitation method. In a typical experiment, a fixed amount of Nb2O5 (Merck, Darmstadt, Germany) was dispersed in 150 mL deionized water

Reusability of the Ag 3 PO 4 /Nb 2 O 5 Photocatalyst
The catalyst's lifetime is an essential parameter of the photocatalytic degradation process, so it is crucial to evaluate the catalyst's reusability for practical applications. As shown in Figure 9, the decrease in the photocatalytic efficiency of the Ag:Nb = 2:1 photocatalyst was gradual, but remained above 60% up to the fifth cycle of the photodegradation experiment. This gradual loss might be due to the self-photo corrosion of Ag 3 PO 4 . The enhanced photocatalytic activity and stability of the Ag:Nb = 2:1 photocatalyst may be due to the insoluble Nb 2 O 5 layer, which can effectively protect the dissolution of Ag 3 PO 4 to Ag + ions, thus improving the structural stability of the photocatalyst.

Reusability of the Ag3PO4/Nb2O5 Photocatalyst
The catalyst's lifetime is an essential parameter of the photocatalytic degradatio process, so it is crucial to evaluate the catalyst's reusability for practical applications. A shown in Figure 9, the decrease in the photocatalytic efficiency of the Ag:Nb = 2: photocatalyst was gradual, but remained above 60% up to the fifth cycle of th photodegradation experiment. This gradual loss might be due to the self-photo corrosio of Ag3PO4. The enhanced photocatalytic activity and stability of the Ag:Nb = 2: photocatalyst may be due to the insoluble Nb2O5 layer, which can effectively protect th dissolution of Ag3PO4 to Ag + ions, thus improving the structural stability of th photocatalyst.

The Synthesis of the Ag3PO4/Nb2O5 Photocatalyst
Ag3PO4/Nb2O5 photocatalysts with different Ag3PO4 to Nb2O5 mole ratios were pre pared via a facile deposition-precipitation method. In a typical experiment, a fixe

The Synthesis of the Ag 3 PO 4 /Nb 2 O 5 Photocatalyst
Ag 3 PO 4 /Nb 2 O 5 photocatalysts with different Ag 3 PO 4 to Nb 2 O 5 mole ratios were prepared via a facile deposition-precipitation method. In a typical experiment, a fixed amount of Nb 2 O 5 (Merck, Darmstadt, Germany) was dispersed in 150 mL deionized water and sonicated for 30 min. Various amounts of AgNO 3 salt (Merck, Germany) were added into the solution mixture and magnetically stirred for 30 min. In total, 150 mL of disodium hydrogen phosphate and Na 2 HPO 4 (Fisher Scientific, Shah Alam, Malaysia) solution were added dropwise into the solution mixture with continuous stirring. The mole ratio of AgNO 3 to Na 2 HPO 4 was kept constant at 3:1 in order to produce the Ag 3 PO 4 catalyst. The yellow precipitate formed was collected by filtration, washed several times with distilled water and ethanol, and dried in an oven at 80 • C overnight. The final product was ground and characterized before being used in the photodegradation studies. A similar procedure, without the addition of Nb 2 O 5 , was employed to produce a pure Ag 3 PO 4 photocatalyst.

Characterization of the Catalyst
The crystalline phase of the catalysts was analyzed using an XRD 6000 (Shimadzu, Kyoto, Japan) with Cu-Kα radiation, operated at λ = 1.54 nm, 30 kV, and 40 mA. The morphology and elemental composition of the catalysts were determined using a FESEM Nova Nanosem 230 (Fei, Eindhoven, The Netherlands) integrated with an energy dispersive X-ray detector. A fluorescence spectrophotometer (Perkin Elmer LS55, Waltham, MA, USA) with a wavelength excitation of 290 nm was used to record the photoluminescence spectra of the samples. The UV-vis diffuse reflectance spectra were recorded by a Shimadzu UV 3101 spectrophotometer at a wavelength between 220 and 800 nm. The chemical state of the elements was analysed using Auger Electron Spectroscopy with an X-Ray Photoelectron Spectrometer (Axis Ultra DLD, Kratos, Manchester, UK). The BET surface area was obtained by the application of BET modelling, conducting in TriStar II Plus (Micromeritics, Norcross, GA, USA). The particle size measurement was determined by Zetasizer, (Malvern Instruments, Worcestershire, UK).

Photoactivity of the Ag 3 PO 4 /Nb 2 O 5 Photocatalyst
The photodegradation of methyl orange dye (MO) was performed in a glass photoreactor, as shown in Figure 10. A fixed amount of photocatalyst was added into 0.5 L of 10 ppm MO solution and stirred in the dark for 30 min. Air was bubbled into the solution throughout the experiment in order to ensure a constant supply of oxygen. During irradiation with a compact fluorescent lamp (23 W, Phillips, 1600 Lumens), an aliquot of the sample was withdrawn from the bulk solution at predetermined time intervals and filtered to remove catalyst particles. The samples were analyzed using UV-vis Perkin-Elmer Lambda 30 spectrophotometer at λ max of 464.4 nm in order to determine the MO dye's residual concentration. The percentage of degradation and the amount of MO degraded are calculated using Equations (1) and (2), respectively: where C 0 is the initial MO concentration, C t is the MO concentration at a time 't', and V is the volume of the MO solution used in the reaction.

Conclusions
The Ag3PO4/Nb2O5 photocatalysts, synthesized via the deposition-precipitatio method, exhibited irregular and spherical shapes with a surface area of 0.89-1.70 m 2 / The catalysts also showed a faster MO degradation rate than Ag3PO4. Ag3PO4/Nb2O5, an a molar ratio of 2:1 showed the highest photocatalytic degradation of 10 ppm MO (96% with a catalyst loading of 0.5 g. The incorporation of Nb2O5 in Ag3PO4 enhanced the cat lysts' photocatalytic performance and stability, and thus can be used effectively und visible light irradiation for up to five cycles. h + and •O 2 are the main active species for th degradation. Funding: This research was funded by Universiti Putra Malaysia, grant number GPB-9629300.

Data Availability Statement:
The data presented in this study are available on request from th corresponding author.

Reusability Test
The reusability of the catalyst was evaluated by performing several cycles of MO photodegradation in the the best-suited conditions. Once the first experiment ended, the treated MO solution was removed by decantation after the catalyst had settled at the bottom of the photoreactor. A fresh MO solution was then added to the reactor in order to start the next cycle.

Conclusions
The Ag 3 PO 4 /Nb 2 O 5 photocatalysts, synthesized via the deposition-precipitation method, exhibited irregular and spherical shapes with a surface area of 0.89-1.70 m 2 /g. The catalysts also showed a faster MO degradation rate than Ag 3 PO 4 . Ag 3 PO 4 /Nb 2 O 5 , and a molar ratio of 2:1 showed the highest photocatalytic degradation of 10 ppm MO (96%) with a catalyst loading of 0.5 g. The incorporation of Nb 2 O 5 in Ag 3 PO 4 enhanced the catalysts' photocatalytic performance and stability, and thus can be used effectively under visible light irradiation for up to five cycles. h + and •O − 2 are the main active species for the degradation.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.