Introduction

As a “green” technology, semiconductor photocatalysis has been widely used for the treatment of water polluted by organic dye [Mahmoodi et al. 2011; Xiao et al. 2012]. TiO2 was one of the most widely researched photocatalyst, with the advantages of non-toxicity, high stability, low-cost. Unfortunately, TiO2 can only utilize the ultraviolet ray consisting of about 5 % of the solar spectrum due to the relatively wide band gap and the high recombination rate, which seriously confine the photocatalysis performance and practical application of TiO2 [Katsumata et al. 2013]. From the view of using solar light for environmental remediation, exploring more efficient visible-light-induced photocatalysts has attracted considerable attention.

In recent years, silver orthophosphate (Ag3PO4) is being thoroughly investigated by increasing researchers, because of excellent photocatalytic performance and efficient separation of photoexcited electrons and holes. Ag3PO4 exhibited extremely excellent oxidative capacities for the evolution of O2 from water, as well as the degradation of organic dyes under the solar irradiation [Yi et al. 2010]. It was reported by Bi et al., that the shape and facet effects of Ag3PO4 crystals on their photocatalytic properties, where the photocatalytic performance of single-crystalline Ag3PO4 rhombic dodecahedrons with only {110} facets exposed was much higher than that of cubes bounded entirely by {100} facets, ascribing to the higher surface energy of {110} facets [Bi et al. 2011]. In addition, other researchers successfully synthesized different shapes of Ag3PO4 single crystals, such as the cubes [Yang et al. 2014], tetrapods [Wang et al. 2013], myriapods [Teng et al. 2015; Wang et al. 2014c], branches [Li et al. 2014; Wang et al. 2014a], and spherical nanocrystals [Liu et al. 2014]. In spite of high efficiency of Ag3PO4 previously reported, the self-corrosion and bad stability always limit the application of Ag3PO4 photocatalyst. Therefore, it becomes very important for Ag3PO4 to successful regeneration.

As a type of wide pollutant, the pharmaceuticals and personal care products (PPCPs) are focused on increasing concern by scientists [Daughton and Ternes 1999; Ellis 2006; Ternes et al. 2004]. Among these, antibiotics are widely used for human and animal health, which are paid more comprehensive attention than other PPCPs [Golovko et al. 2014; Ortiz de Garcia et al. 2014].

In general, the Ag3PO4 photocatalysis degradation under visible-light irradiation of organic pollutants has mainly been applied to colored dyes, including methylene blue, rhodamine B, and methyl orange [Hua et al. 2015], while few reports have focused on antibiotics compounds [Wang et al. 2015; Wu et al. 2013].

As the second most usage of antibiotics, tetracycline (TC) has been used as broad-spectrum antibacterial agent for human and animal health, leading to a large amount of residual in aquatic and soil environment, which in turn threatened human health for appearance of drug-resistant bacteria and clinical side effects on liver and kidney [Martins et al. 2014]. Therefore, it is essential to exploit cost-efficient and feasible treatment technologies to remove these contaminations in the aquatic environment.

In this paper, we successfully prepared high-efficiency, micron-scale polyhedral Ag3PO4-PN photocatalyst with sharp corners and edges with the synergetic assistances of PVP and NH3·H2O. The catalysts were characterized by a series of methods. In this paper, tetracycline hydrochloride (TC), a kind of common broad-spectrum antibiotics, was selected as the target pollutant. Then, we investigated the photocatalysis performance and mineralization capacity of TC by as-synthesized Ag3PO4-PN. Moreover, the photocatalytic reaction mechanism, photocatalyst stability, and regeneration of obtain Ag3PO4-PN were also surveyed.

Experimental

Materials and methods

All the reagents used were of analytical purity and were used without further purification. Silver nitrate (AgNO3), disodium hydrogen phosphate dodecahydrate (Na2HPO4 · 12H2O) and ammonium oxalate (AO) were purchased from Tianjin Kermel Chem. Reagent Co., Ltd. Polyvinyl pyrrolidone (PVP, Mw 58000, K 29–32) was purchased from Shanghai Aladdin BioChem. Tech. Co., Ltd. Ammonia solution (NH3·H2O) was purchased from Luoyang Chem. Reagent Factory. Benzoquinone (BQ) and t-butyl alcohol (TBA) were Flukaimport reagents subpackaged by Shanghai Chemical Dispensing Factory and Shanghai Asus Fine Chemical Co., Ltd. Hydrogen peroxide (H2O2 30 %) was purchased by Tianjin Fengchuan Chemical Reagent Technologies Co., Ltd.

In a typical synthesis, 0.2 g PVP was added in 300 ml deionized water with continuous stirring for 10 min. 0.6 g AgNO3 was dissolved in the above solution and stirred for 20 min in darkness. Ammonia solution (1:1, v:v) was added until solution became clear, then continuously stirred for 10 min. Then 50 ml, 0.15 M Na2HPO4 solution was added drop-by-drop to above mixture solution, and stirred for 4 h. The yellow precipitate was centrifuged and washed with deionized water and absolute ethanol for several times and dried at 60 °C for 12 h and was designed as Ag3PO4-PN. And three similar procedures were used to prepare Ag3PO4 without the presence of PVP, NH3·H2O, PVP, and NH3·H2O, and were designed as Ag3PO4-N, Ag3PO4-P, and Ag3PO4-A, respectively.

Characterization

X-ray diffraction (XRD) analysis was performed on an X-ray diffractometer (X’Pert PRO MRD, PANalytical, the Netherlands) equipped with a Cu Kα X-ray source. The morphology of the samples was investigated by scanning electron microscopy (SEM, Quonxe-2000, Philips, the Netherlands). The surface elemental component and the chemical state of the sample were analyzed by X-ray photoelectron spectroscopy (XPS, PHI 5300, Perkin Elmer) with a monochromatized Al Kα X-ray source (hν = 1487 eV). UV–vis diffuse reflection spectra (UV–vis DRS) of the samples were obtained using a UV–vis spectrophotometer (Caly 5000, Agilent, USA). The band gap energy (Eg) of these samples was evaluated using the following equation [Guan et al. 2013; Liang et al. 2012]:

$$ \upalpha \left(\mathrm{h}\upnu \right)=A{\left(\mathrm{h}\upnu -{E}_{\mathit{\mathsf{g}}}\right)}^2 $$
(1)

where α, ν, Eg, and A are the absorption coefficient, light frequency, band gap energy, and a constant, respectively [Yi et al. 2010].

Measure of photocatalytic activity

The as-prepared photocatalyst activity was measured by the degradation of TC. A 100 mg photocatalyst was added to 50 mL of contamination solution as the initial concentration was 30 mg/L. Before illumination, the solution with a certain amount of photocatalyst was stirred for 30 min in darkness to reach the adsorption and desorption equilibrium. A 420-nm filter was placed between xenon lamp and reaction solution to filter ultraviolet. And the experimental device was illustrated in Fig. 1. At the process of the illumination of 350 W xenon lamp, a specified volume of solution was fetched at a given time interval and filtered immediately with 0.45 μm membrane filter. The concentration of TC solution were evaluated by UV–vis spectroscopy (UV-2450, Shimadzu, Japan), with 357.5 nm of the maximum absorption wavelength. The degradation efficiency was calculated by 1-c/c0, while the c was instantaneous concentration and c0 was solution concentration after adsorption process. The data of degradation experiment was the average value of three times parallel experiments. And the standard deviations of the average of independent runs are all between 1.2 and 2.9 %.

Fig. 1
figure 1

The illustration of photocatalysis experiment device

Full size image

The mineralization degree of TC and residual of long-lived organic intermediates in the solutions [Zhang et al. 2013] were measured by total organic carbon (TOC). The TOC concentration of filtrate was determined by using Total Organic Carbon Analyzer (TOC-VCPH/CPN, Shimadzu, Japan). The NO3 − and Cl− concentrations of TC solution were determined by ion chromatograph (IC), equipped with anion chromatography column (AS-14), anion guard column (AG-14), chemical suppressor (AMMS 30), and 8.0 mM Na2CO3/1.0 mM NaHCO3 as a leacheate with 0.8 mL min−1 of flow rate and dilute sulphuric acid as regeneration solution.

Detection of reactive oxygen species

The scavenging experiments of reactive oxygen species were similar to the photodegradation experiments. Different quantity of scavengers was added into the TC solution prior to addition of catalysts. These experiments were carried out according to the previous study [Lin et al. 2012]. The concentrations of scavengers were 1 × 10−3 mol/L (7.2 mg), 1.6 × 10−2 mol/L (0.1 g), and 1 × 10−3 mol/L (5.4 mg) for TBA, AO, and BQ, respectively. The concentration of solution was also measured by solution absorbance using a spectrophotometer (Shimadzu UV-2450).

Results and discussion

Characterization of photocatalysts

Crystal structure

The X-ray diffraction (XRD) patterns of Ag3PO4 samples (Fig. 2) show that all the diffraction peaks of the samples could be indexed to the body-centered cubic structure of Ag3PO4 (JCPDS card No. 06-0505). No peaks assignable to other phases are observed, indicating a high purity of Ag3PO4 samples. The sharp and strong diffraction peaks indicate that the Ag3PO4 catalysts were well crystallized. In addition, the peak intensity of Ag3PO4-N and Ag3PO4-PN are stronger than those of Ag3PO4-A and Ag3PO4-P, indicating the addition of NH3·H2O is in favor of formation of crystals.

Fig. 2
figure 2

XRD patterns of different Ag3PO4 powers: a Ag3PO4-A; b Ag3PO4-P; c Ag3PO4-N; d Ag3PO4-PN

Full size image

In order to investigate the valence states and surface chemical compositions of the Ag3PO4-PN microstructure, XPS analysis was employed and the results are shown in Fig. 3. Figure 3a presents an overview of the XPS spectrum of pure Ag3PO4-PN microstructures. Ag, P, O, and adventitious C coexist in the Ag3PO4-PN microstructures. Figure 3b–d display the high resolution XPS spectra of Ag, P, and O, respectively. In Fig. 3b, the peaks centered at 373.7 and 367.7 eV are ascribed to Ag 3d3/2 and 3d5/2, respectively, revealing the Ag exists mainly in the +1 oxidation state in the sample. As shown in Fig. 3c, the peak of P 2p is located at 132.5 eV, indicating that the oxidation state of P is +3 in the sample [Bertrand 1981]. In addition, the spectrum of O 1 s is located at 530.4 eV, which is assigned to the O2 − in Ag3PO4 [Katsumata et al. 2014].

Fig. 3
figure 3

a Overview XPS spectrum of Ag3PO4-PN microstructure, b–d are the XPS spectra of Ag3PO4-PN microstructure for the Ag 3d spectrum, P 2p spectrum, and O 1 s spectrum

Full size image

Crystal morphology

The morphology and size of the Ag3PO4 samples were surveyed by SEM. Figure 4a reveals the as-synthesized Ag3PO4-A is constituted of quasi-spherical particles with diameters around 600 nm. Figure 4b shows the quasi-tetrahedron morphology of Ag3PO4-P due to the addition of PVP as the structure inducer with the size in the range of 1–10 μm. Ag3PO4-N has a cube structure with the size of 2–10 μm (Fig. 4c), while Ag3PO4-PN exhibited a polyhedron structure with the size of 0.5–7 μm (Fig. 4d). The formation of sharp edges and corners on the surfaces of cube and polyhedron may be benefit from the introduction of NH3·H2O [Guo et al. 2015].

Fig. 4
figure 4

SEM of the obtained Ag3PO4 samples: a Ag3PO4-A; b Ag3PO4-P; c Ag3PO4-N; d Ag3PO4-PN

Full size image

Optical property

The optical property of Ag3PO4 samples was measured by UV–vis DRS, and the UV–vis spectra are exhibited in Fig. 5. The four samples are all Ag3PO4 prepared only with different additives, so the optical absorption edges of the samples have rarely significant difference and it is largely related to electronic structure of materials. It can be clearly seen that all the Ag3PO4 samples display photo-absorption in the UV and visible region. The absorption edges of Ag3PO4-A and Ag3PO4-PN are ca. 530 nm, while those of Ag3PO4-P and Ag3PO4-N are ca. 550 nm, which have slight shift to red. It is noted that the absorbance of Ag3PO4-P sample is obviously lower than those of Ag3PO4-A, Ag3PO4-N, and Ag3PO4-PN samples in the range of 330–500 nm, which may adverse to degradation performance of Ag3PO4-P sample.

Fig. 5
figure 5

UV–vis DRS of prepared Ag3PO4 samples

Full size image

Photocatalytic activity

Comparison of degradation activity of four Ag3PO4 catalysts

The photocatalytic activities of the four Ag3PO4 samples were tested by photodegradation of TC under visible-light irradiation and the results are shown in Fig. 6a. The Ag3PO4-PN exhibited the highest degradation efficiency, arriving 91 % after 120 min of visible-light irradiation. The degradation efficiencies of TC by Ag3PO4-A, Ag3PO4-P, and Ag3PO4-N are 78, 72, and 85 %, respectively. The degradation of TC by Ag3PO4 under visible-light irradiation followed pseudo-second-order kinetics and the reactive rate constants of four Ag3PO4 samples are 0.0006, 0.0009, 0.0013, 0.0023 L mg−1 min−1, respectively, which are shown in Fig. 6b. The higher degradation efficiencies of Ag3PO4-N and Ag3PO4-PN are attributed to generation of smooth surface, sharp edges and corners, which might contribute to the photocatalysis process at the surface owing to [100] plane with high surface energies [Bi et al. 2011; Guo et al. 2015]. The enhanced photocatalytic activity of Ag3PO4-PN benefits from the synergistic effect of PVP and NH3·H2O which not only inhibits the crystal aggregation but also facilitates the order array of crystals. Besides, the strong visible-light absorption capacity and appropriate particle sizes of Ag3PO4-PN are also conducive to the improvement of photocatalytic activity. The lowest degradation efficiency of Ag3PO4-P is by reason of low absorbance in visible region, which is consistent with results in Fig. 5.

Fig. 6
figure 6

Photocatalytic degradation efficiency (a) and degradation kinetic fitting (b) of different Ag3PO4 samples under solar irradiation with the 2.0 g L−1 of catalyst concentration and 30 mg L−1 of initial concentration of TC

Full size image

Effect of the reaction conditions

Since the enhanced photocatalytic activity, Ag3PO4-PN was selected to degrade contamination in the later experiments. To investigate the effects of various reaction conditions on the photocatalytic efficiency of tetracycline hydrochloride (TC), three reaction conditions were researched, including catalyst dosage, TC concentration, and pH of TC solution. The catalyst dosage was ranged from 0.25 to 3.0 g/L. The result shows that the degradation efficiency of TC increased with the increase of catalyst dosage, while it barely continuously rose until catalyst dosage increasing to 2 g/L (Fig. 7a). It has been reported that in a certain dosage range, radicals increases with the increase of catalyst dosage owing to the increase of the generated carriers [Bertelli and Selli 2004]. But the scatter reflection will result in a low utilization of irradiation light because of an over dosage [Wang et al. 2014b]. To take both into account, the dosage of 2 g/L is used in the later experiments. The effect of TC concentration on the photocatalytic efficiency of TC was investigated (Fig. 7b) with the range of 20–60 mg/L. The results exhibited that the degradation efficiency of TC by Ag3PO4-PN increased with the solution concentration up to 30 mg/L, and then decreased with the increase of TC concentration. The effect of pH of TC solution on the photocatalytic efficiency of TC was also investigated (Fig. 7c). The result indicates that the photocatalytic efficiencies of TC under weak acid and neutral conditions were higher than those under strong acid and alkaline conditions, which may because of instability of Ag3PO4 photocatalyst under strong acid and alkaline conditions. From above conclusion, the degradation efficiency of TC by Ag3PO4-PN was highest with 2 g/L of catalyst dosage, 30 mg/L of TC concentration, and 5 of pH of TC solution.

Fig. 7
figure 7

Effects of the Ag3PO4-PN dosage (a), TC concentration (b) and pH of TC solution (c) on the degradation efficiencies of TC under visible-light irradiation, with irradiation for 120 min

Full size image

Photocatalytic degradation mechanism

TOC and IC

Total organic carbon (TOC) was frequently applied as an important index to represent the concentration of organic matter in contamination solution [Sakthivel et al. 2003]. Herein, the mineralization of TC is evaluated by TOC content and quantitative determination of inorganic ions by Ag3PO4-PN as the photocatalyst. Figure 8a shows TOC removal increases with increase in irradiation time and arrived ca. 45 % through consecutive irradiation of visible light at 120 min. Moreover, due to the presence of nitrogen and chlorine elements, the detection of inorganic ions was also used to assess the mineralization degree of TC. The quantity measurements of NO3 − and Cl− were carried out by IC during the photodegradation process (Fig. 8b). The Cl− concentration of TC solution rapidly decreases during dark adsorption process, which may by reason of the generation of AgCl. With the increase of irradiation time, the Cl− concentration remains drop, owing to the catalytic oxidization. In addition, the NO3 − concentration showed a gradually increased tendency with the illumination irradiation, indicating the nitrogen of TC molecular was partly oxidized into NO3 −. Previous studies [Bandara et al. 1997] have demonstrated the conversion of intermediate nitrite partly into ammonia and nitrate. The formation of NH3 which escapes may account for the loss of nitrogen in the mass balance.

Fig. 8
figure 8

TOC removal (a) and Cl− and NO3 − concentration (b) of TC solution vs. irradiation time with TC concentration of 30 mg/L and Ag3PO4-PN dosage of 2 g/L

Full size image

Oxidative radical species

To investigate the photocatalytic mechanism of Ag3PO4-PN into more detail, the effects of active species scavengers on the degradation of TC were examined to shed light on the predominant reactive oxygen species in the photocatalytic process (Fig. 9). In this study, AO, BQ, and TBA were added into the reaction solution as h+, ·O2−, and ·OH scavenger, respectively [Lin et al. 2012]. Without the addition of the scavengers, the photocatalytic efficiency of TC is 91 % after 120 min of visible-light irradiation. As is clear from Fig. 9, the addition of TBA scarcely affected the degradation efficiency of TC by Ag3PO4-PN, which indicates that ·OH does not appear to be the main reactive oxygen species in the photocatalytic process of TC by Ag3PO4-PN.

Fig. 9
figure 9

Effect of various scavengers (AO, BQ and TBA) on the degradation of TC by Ag3PO4-PN, with TC concentration of 30 mg/L and catalyst dosage of 2 g/L

Full size image

The visible-light photocatalytic degradation of TC was significantly inhibited by BQ with 23 % of reduction of degradation efficiency, confirming that ·O2− indeed plays an important role in the degradation of TC by Ag3PO4-PN. Besides, the addition of AO likewise suppresses the photocatalytic degradation of TC with 19 % of decrease of degradation efficiency. It is, therefore, most likely that the photoinduced holes, with a high oxidation potential of +2.45 V vs. NHE, directly react with organic compounds [Yi et al. 2010]. Therefore, it may be concluded that h+ and ·O2− are major active species in the photodegradation of TC by Ag3PO4-PN.

Photocatalyst stability and regeneration

The photocatalyst stability was assessed by the cyclic photocatalytic reaction experiments of TC by Ag3PO4-PN. The cycle experiments were carried out at the same conditions as usual degradation tests and Ag3PO4-PN catalyst was washed twice by distilled water and dried before reused. As shown in Fig. 10, the photodegradation activity of Ag3PO4-PN was availably held after five times loop experiments, indicating Ag3PO4-PN has relatively high stability under visible-light irradiation. Five degradation efficiencies of repeated experiments were 89, 85, 82, 77, and 74 %, respectively. Nevertheless, it was found that Ag3PO4 materials always possessed photo-corrosion characteristics, and reductive elemental silver wrapped around the surface of Ag3PO4 catalyst may have a negative effect on the photo-absorption performance and photocatalytic activity of Ag3PO4, which prevented practical application of Ag3PO4 as a recyclable high-efficient photocatalyst [Yi et al. 2010].

Fig. 10
figure 10

Five times recycling runs of Ag3PO4-PN and three times degradation experiment of regenerative Ag3PO4-PN for the degradation of TC under visible-light irradiation. Each run of photocatalytic reaction lasted for 120 min (catalyst dosage = 1.0 g/L, TC concentration = 20 mg/L, pH = 4.3)

Full size image

A facile wet chemical-oxidation method was utilized to rejuvenate Ag3PO4-PN from weakly active Ag, which was previously reported by Wang and his coauthors. During the rejuvenation process, weakly active Ag can be oxidized by dropwise adding H2O2 into saturated PO4 3− solution. While the acid environment of NaH2PO4 and H3PO4 aqueous solution can dissolve the Ag3PO4, and the fast decomposition of H2O2 will occur under strong alkaline condition of Na3PO4 solution, weaker alkaline Na2HPO4 was chosen as the source for PO4 3− ions [Wang et al. 2012]. The photodegradation efficiency of TC by rejuvenated Ag3PO4-PN during three times recycling runs were 86, 84, and 80 %, respectively, indicating a successful rejuvenation of Ag3PO4, which greatly promotes the development of Ag3PO4 for practical applications.

Conclusion

A facile, environmental-friendly Ag3PO4-PN photocatalyst was successfully synthesized by a simple precipitation method at room temperature with the presence of NH3·H2O and PVP. The Ag3PO4-PN exhibits a superior photocatalytic activity compared to those of Ag3PO4-A, Ag3PO4-P, and Ag3PO4-N under visible-light irradiation. The enhancement of photocatalytic efficiency of Ag3PO4-PN is strongly dependent on the excellent photo-absorption capacity, sharp edges and corners, and synergistic effect of PVP and NH3·H2O. The effect researches of reaction conditions indicated that the optimum condition were 2.0 g/L of catalyst dosage, 30 mg/L of TC solution concentration, and 5 of pH value of TC solution. Radical detection experiments indicate the h+ and ·O2− are major active species in the degradation of TC by Ag3PO4-PN. TOC test and IC analysis indicate TC molecular was partially mineralized under the visible-light irradiation by Ag3PO4-PN. Moreover, photocatalyst stability and regeneration experiments exhibit a favorable stability and rejuvenation ability, suggesting a promising prospect of practical application of Ag3PO4 in the wastewater treatment.