Photocatalytic activity of attapulgite–TiO2–Ag3PO4 ternary nanocomposite for degradation of Rhodamine B under simulated solar irradiation

An excellent ternary composite photocatalyst consisting of silver orthophosphate (Ag3PO4), attapulgite (ATP), and TiO2 was synthesized, in which heterojunction was formed between dissimilar semiconductors to promote the separation of photo-generated charges. The ATP/TiO2/Ag3PO4 composite was characterized by SEM, XRD, and UV-vis diffuse reflectance spectroscopy. The co-deposition of Ag3PO4 and TiO2 nanoparticles onto the surface of ATP forms a lath-particle structure. Compared with composite photocatalysts consisting of two phases, ATP/TiO2/Ag3PO4 ternary composite exhibits greatly improved photocatalytic activity for degradation of rhodamine B under simulated solar irradiation. Such ternary composite not only improves the stability of Ag3PO4, but also lowers the cost by reducing application amount of Ag3PO4, which provides guidance for the design of Ag3PO4- and Ag-based composites for photocatalytic applications.


Background
Organic pollutant degradation has been a critical process towards resolving environmental pollution. Fujishima et al. reported in 1972 that TiO 2 has the capability of utilizing solar energy for water splitting and hydrogen production [1]. Since then, semiconductor-based photocatalytic technology has become a promising, and yet effective approach to resolve environment pollution. Over the past decades, a number of semiconductors, such as TiO 2 , Ag 3 PO 4 , BiVO 4 , WO 3 , and g-C 3 N 4 , have been extensively investigated for photocatalytic application [2]. Among them, TiO 2 has received extensive attention due to its good chemical stability, nonphotocorrosion, low cost, and nontoxicity. Because of its wide band gap (3.2 eV) and lacking visible light absorption, however, TiO 2 exhibits low photocatalytic efficiency.
The application of TiO 2 -based photocatalysts was thus hampered severely.
The photocatalysts, such as Ag 3 PO 4 [3], Bi2MoO 6 [4], WO 3 [5], and g-C 3 N 4 [6], can exhibit high-efficiency under visible light irradiation, and thus have drawn extensive research efforts. For example, Ye et al. reported that silver orthophosphate (Ag 3 PO 4 ) exhibited much stronger photooxidative capabilities and higher efficiency for photocatalytic degradation [3] than most other known photocatalysts such as WO 3 [5] and BiVO 4 [7]. However, the photocatalytic stability of Ag 3 PO 4 could be deteriorated by the photoreduction of Ag + into metallic Ag. The low photostability and high cost of Ag 3 PO 4 are concerning issues that will limit its photocatalytic applications. In this context, Ag 3 PO 4 -based composite photocatalysts have been investigated with the goal of improving its photostability and photocatalysis, such as TiO 2 /Ag 3 PO 4 [8], Ag 3 PO 4 /graphene [9], and Ag 3 PO 4 /Ag/WO 3-x [10]. Attapulgite (ATP) is a kind of rod-shaped fiber hydrated magnesium aluminum silicate non-metallic mineral, which has remarkable physical and chemical properties, such as exchangeable cations, water absorption, adsorption discoloration, and large specific surface area [11]. ATP is thus considered to be an ideal catalyst carrier with rod morphology, and its high surface area is benefit for absorbing catalyst and pollutant. Although Ag 3 PO 4 -and TiO 2 -based and attapulgite/Ag 3 PO 4 binary composite photocatalysts have been reported, attapulgite-based ternary composite materials have rarely been investigated.
In this work, the ATP/TiO 2 /Ag 3 PO 4 ternary composites were synthesized by a facile two-step method for improving the photostability and photocatalysis of Ag 3 PO 4 and suppressing the consumption of noble metal Ag. The crystalline structure and microstructure of novel ternary composites were characterized by XRD and SEM, respectively, while their photocatalytic activities and stability were measured by degradation of organic dye rhodamine B (RhB) under simulated solar irradiation. This ternary composite exhibits higher photocatalytic efficiency than pure silver phosphate and excellent photocatalytic stability.

Synthesis of samples
The ATP/TiO 2 /Ag 3 PO 4 ternary composite was synthesized by a facile two-step method. Well-dispersed ATP nanorods and TiO 2 nanoparticles with mass ratio of 5:2 were first added into deionized water and stirred for 4 h. Through physical and surface electronic absorption, the TiO 2 nanoparticles were attached to the surface of ATP nanorods. After centrifugal separation, the precipitate was washed with deionized water and then dried at 60°C for 6 h to obtain ATP/TiO 2 composites. By a simple precipitation method, Ag 3 PO 4 nanoparticles were deposited on the surface of ATP/TiO 2 and ATP/TiO 2 /Ag 3 PO 4 ternary composites were then prepared. [12] In a typical preparation process, 20 ml silver nitrate solution (0.1 mol/L) was dissolved in ATP/TiO 2 aqueous suspension with 0.7 g ATP/TiO 2 composites and 50 ml deionized water by ultrasonic stirring for 30 min. 20 ml Na 2 HPO 4 aqueous solution (0.1 mol/L) was then added slowly into the above solution with ultrasonic stirring in dark condition for another 40 min. Then, the light yellowish-brown precipitate was centrifuged, washed several times with ethanol absolute, and dried at 60°C for 12 h, to obtain ATP/TiO 2 / Ag 3 PO 4 ternary composites. The powder samples of Ag 3 PO 4 , Ag 3 PO 4 /ATP, Ag 3 PO 4 /TiO 2 , and ATP/TiO 2 were also synthesized using the similar method.
Characterization X-ray diffraction was collected using XRD Rigaku D/ max-RB) for phase analysis of the powders under 40 kV and 30 mA. The microstructures were evaluated by scanning electron microscopy (SEM, INSPECTF FEI, Netherlands). Ultraviolet-visible (UV-vis) diffuse reflection spectroscopy of the photocatalyst was investigated using U-3010 Hitach UV-vis spectrophotometer using BaSO 4 as reference.

Photocatalytic experiment
Photocatalytic degradation of RhB was tested under simulated solar irradiation. 50 mg ATP/TiO 2 /Ag 3 PO 4 was added to 100 ml RhB solution with a concentration of 5 mg/L and stirred in dark for 40 min to ensure adsorption-desorption equilibrium. The light source was a 300 W Xe lamp (Microsolar300, PerfectLight, Beijing, China) at about 150 mW/cm 2 (as tested by a radiometer FZ-A, Photoelectric Instrument Factory of Beijing Normal University, China). After opening the lamp, 4 ml solution was taken out at known time intervals and separated through centrifugation (10,000 rpm, 10 min). The supernatants were analyzed by recording variations of absorption peak (554 nm) in the UV-vis spectra using UV/vis spectrophotometer (T6, PERSEE, Beijing, China).
The degradation degree of RhB dye was determined according to the following equation: where c 0 and c are the initial concentration and concentration after photocatalysis of the solution, respectively; and A 0 and A are the absorbance values of the solution before and after photocatalytic reaction, respectively.

Results and discussion
Characterization of the ATP-Ag 3 PO 4 -TiO 2 composites The XRD patterns of ATP, TiO 2 , Ag 3 PO 4 , and nanocomposites are shown in Fig. 1. The diffraction peaks in Fig. 1a can be indexed as ATP phase with monoclinic structure (JCPDS # 21-0958), which implies that the ATP had been specially purified and no impurity phases exist. Figure 1b displays typical diffraction peaks of anatase TiO 2 without any impurities, while Fig. 1c shows the diffraction peaks corresponding to pure Ag 3 PO 4 phase, in good agreement with JCPDS # 06-0505. There are not any impurity phases or structure destabilization for all the nanocomposite samples of ATP/TiO 2 (Fig. 1d), Ag 3 PO 4 / TiO 2 (Fig. 1e), ATP/Ag 3 PO 4 (Fig. 1f), and ATP/TiO 2 / Ag 3 PO 4 (Fig. 1g). In XRD patterns of ATP/TiO 2 /Ag 3 PO 4 (Fig. 1g), main characteristic peaks associated with both Ag 3 PO 4 and TiO 2 can be detected, while the diffraction peaks from ATP phase are much weaker. The phenomenon implies that the ATP nanorods are cladded by TiO 2 and Ag 3 PO 4 nanoparticles.
The morphological and microstructure of the composite photocatalysts are shown in Fig. 2. ATP nanorods exhibited an average length less than 1 μm and a diameter less than 100 nm (Fig. 2a). Due to surface physical and chemical adsorption, TiO 2 nanoparticles with diameter of about 40 nm attached to the surface of ATP nanorods and formed ATP/TiO 2 composites, as shown in Fig. 2b. In Fig. 2c, the ATP nanorods were fully covered by Ag 3 PO 4 and TiO 2 particles in ATP/TiO 2 / Ag 3 PO 4 ternary composite, while Ag 3 PO 4 appeared on the surface of ATP/TiO 2 composites in the form of uniform spheroidal particles with diameter of about 50 nm.

Absorption spectra
The UV-vis absorption spectra of Ag 3 PO 4 , ATP, TiO 2 , and ATP/TiO 2 /Ag 3 PO 4 are shown in Fig. 3a. Similar with the reported results, [3] Ag 3 PO 4 exhibits good absorption from the UV to the visible light region with a wavelength up to about 500 nm. On the other hand, TiO 2 exhibits an excellent UV absorption without obvious absorption in visible light region. ATP shows a lower UV absorption and little absorption in visible light region. As expected, ATP/TiO 2 /Ag 3 PO 4 ternary nanocomposite exhibits a strong UV absorption benefiting from TiO 2 and ATP and the enhanced visible-light absorption imposed by Ag 3 PO 4 . The optical band gap (E g ) can be estimated from the optical absorption edge according to the Eq. (1). [13,14] αhv where α is the spectral absorption coefficient, "hv" is the photon energy, A is a constant, and m is equal to 0.5 or 2 for direct and indirect transitions, respectively. TiO 2 [15] is generally regarded as an indirect bandgap semiconductor, and its indirect E g is determined by the interception of a straight line fitted through the low-energy side of the curve (αhυ) 1/2 versus hυ as shown in Fig. 3b, with an estimated value of about 3.20 eV. Ag 3 PO 4 was reported as an indirect bandgap semiconductor, and its direct gap at the Gamma point and the indirect gap are very close in terms of the calculated results. [16] Its direct gap of about 2.45 eV was regarded as the bandgap of Ag 3 PO 4 in most reports. Here, the indirect E g and direct E g are determined by the interception of the straight line fitted through the low-energy side of the curve (αhυ) 1/m (m = 2 and 0.5) versus hυ, respectively. The results of Ag 3 PO 4 reveal an indirect bandgap of 2.33 eV (Fig. 3b) and a direct bandgap of 2.49 eV (Fig. 3c). The direct E g of 2.49 eV is more matched with its absorption band edge than the indirect bandgap of 2.33 eV. Thus, the E g of Ag 3 PO 4 is determined as 2.49 eV. Similarly, ATP shows an indirect bandgap of 3.37 eV (Fig. 3b) and a direct bandgap of 3.75 eV (Fig. 3c), and the E g of ATP is determined as 3.75 eV. The above bandgap values of TiO 2 , Ag 3 PO 4 and ATP are quite close to the reported results. [17] In the ATP/TiO 2 /Ag 3 PO 4 ternary

Photocatalytic activities
The photocatalytic activity of the resulting samples was evaluated by the degradation of RhB under Xe light irradiation, Fig. 4. After immersing photocatalysts, RhB solutions were stirred for 40 min in dark condition to establish adsorption-desorption equilibrium with the goal of eliminating the interference of adsorption. Figure 4a shows the evolution of absorption spectra during the only a little decrease in the absorption peak intensity is observed for RhB, which indicates a weak dye adsorption of the nanocomposite. After irradiation for 20 min, the characteristic absorption peak of RhB nearly disappeared, implying almost complete degradation of the dye in the solution. Under similar Xe light irradiation condition, the photocatalytic degradation of RhB with different photocatalysts is compared in Fig. 4b. The photocatalysts of single-phase TiO 2 and ATP showed lower degradation rate than 50% under 60 min irradiation, while Ag 3 PO 4 displayed much stronger and faster photocatalytic degradation, in good agreement with previous reports on photocatalysis of TiO 2 and Ag 3 PO 4 [18]. Ag 3 PO 4 was reported as a strong photocatalyst, but its stability of photocatalytic activity is low and its cost is high. The ternary nanocomposites revealed a fast degradation rate of around 81.1% only after 3 min irradiation and almost complete degradation after 20 min irradiation, which are obviously higher than that of single-phase Ag 3 PO 4 and other binary composite photocatalysts including ATP/ Ag 3 PO 4 and TiO 2 /Ag 3 PO 4 as seen in Fig. 4b. ATP has little photocatalytic activity, but it has been reported with good ability of adsorption, [19] which facilitates dye molecules adhering to its surface, and results in a higher degradation rate of RhB by the ATP/TiO 2 /Ag 3 PO 4 ternary nanocomposite photocatalysts. Interestingly, the ATP/ TiO 2 /Ag 3 PO 4 photocatalysts showed stronger photocatalytic degradation efficiency than TiO 2 /Ag 3 PO 4 or Ag 3 PO 4 with the same weight. As a result, the application amount of high-cost Ag 3 PO 4 is reduced. The stability of the photocatalysts for photodegradation of RhB under Xe light irradiation was evaluated by repeated photocatalytic experiments. Similar test was also performed on Ag 3 PO 4 for comparison. After each run of photocatalytic degradation, the photocatalysts were separated, washed, dried, and then recycled for the next run. The initial concentration of RhB and the dosage of photocatalyst were kept consistent during each run of photocatalytic degradation. The results are shown in Fig. 5. After every run, the activity of Ag 3 PO 4 significantly decreased as expected [20]. In the photocatalytic process, the active sites were covered by Ag appearing on the surface of Ag 3 PO 4 particles. The photocatalytic activity of the ATP/TiO 2 /Ag 3 PO 4 ternary nanocomposite remained unchanged even after five cycling runs of photodegradation of RhB. This result indicates that the photocatalysis is very stable in ATP/TiO 2 /Ag 3 PO 4 ternary nanocomposites.

Possible mechanism in photocatalytic process
In photocatalytic degradation processes, the common reactive oxygen species include •OH radicals, O 2 •radicals and holes (h + ). [2] The trapping experiments were carried out to monitor the reactive oxygen species involved in photocatalytic process of ATP/TiO 2 /Ag 3 PO 4 composites over RhB. Three chemicals of tert-butanol (TBA), benzoquinone (BQ), and disodium ethylenediaminetetraacetate (Na 2 -EDTA) were used as scavengers of •OH radicals, O 2 •radicals and holes, respectively. [9] The experimental results under Xe light irradiation are shown in Fig. 6. The introduction of 1 mM TBA (•OH radical scavenger) has no obvious influence on the photocatalytic activity of the composite photocatalyst (Fig. 6b). This result indicated that OH· radicals are not the main active oxygen species in the photocatalytic process. The addition of 1 mM BQ (O 2 •radical scavenger) reduces the photocatalytic degradation degree of RhB to 42% in 60 min (Fig. 6c), which indicates that O 2 •radicals make an important but only segmental contribution to photocatalytic performance. After adding the hole scavenger Na 2 -EDTA (1 mM) into the photocatalytic system, the photocatalytic degradation activity of ATP/TiO 2 /Ag 3 PO 4 nanocomposites is almost completely suppressed (Fig. 6d), and the degradation degree of RhB decreases to less than 5% after 60 min. This result implies that holes play a key role in photocatalytic •radicals are the main reactive radicals in the ATP/TiO 2 /Ag 3 PO 4 photocatalytic process degrading RhB under Xe light irradiation.
Based on the discussion mentioned above, a possible photocatalytic mechanism was proposed to explain the photocatalytic degradation of RhB by ATP/TiO 2 /Ag 3 PO 4 ternary composite photocatalysts, as shown in Fig. 7. The potentials for conduction band (CB) and valence band (VB) of TiO 2 are − 0.5 eV vs. NHE, and + 2.70 eV vs. NHE, respectively [21,22]. These values are more negative than that of both Ag 3 PO 4 (CB + 0.45 eV vs. NHE, VB + 2.97 eV vs. NHE) [3,16] and ATP (CB − 0.25 eV vs. NHE, VB + 3.50 eV vs. NHE). Therefore, the photo-generated electrons in the CB of TiO 2 can easily transfer to that of Ag 3 PO 4 , while the photo-induced holes in the VB of Ag 3 PO 4 will migrate to that of TiO 2 , which promotes the effective separation of photogenerated electron-hole pairs and decreases the recombination probability of electrons and holes. As a result, the ATP/TiO 2 /Ag 3 PO 4 composite photocatalyst can exhibit higher photocatalytic activities than single phase Ag 3 PO 4 . Meanwhile, the holes in VB of TiO 2 , which has strong oxidation characteristics, not only could significantly accelerate the photocatalytic reaction rates of RhB degradation, but also could oxidize H 2 O to generate O 2 . The reduction potential of O 2 •is − 0.28 eV, while the potentials of CB for TiO 2 and Ag 3 PO 4 are − 0.3 and + 0.45 eV, respectively. Therefore, the resulting O 2 at the surface of photocatalysts then could capture photogenerated electrons to produce O 2 •radicals, and the Ag + ions in Ag 3 PO 4 could be protected from photoreduction into metallic Ag (Ag + + e − → Ag) since the electrons were consumed in the reaction with O 2 . In consequence, the composite photocatalyst with TiO 2 and Ag 3 PO 4 shows much higher stability than single-phase Ag 3 PO 4 photocatalyst.

Conclusions
In conclusion, we synthesized ATP/TiO 2 /Ag 3 PO 4 ternary composite through a simple method: TiO 2 nanoparticles were absorbed on the surface of ATP to form a binary structure, and then Ag 3 PO 4 nanoparticles were deposited on ATP/TiO 2 composite through electrostatic interaction. The heterogeneous junction formed in the ternary composite improves the photocatalytic efficiency and stability. In comparison with pure Ag 3 PO 4 phase, this kind of composite photocatalyst not only reduces the consumption of the precious metal silver to a larger extent, but also improves the efficiency of photocatalysts. Our results will provide guidance to design Ag-based composites for photocatalytic application.