During the extensive course of development of silver salt semiconductor photocatalysts, silver phosphate (Ag3PO4) receives tremendous attention. Ag3PO4 can improve the electron transmission because it possesses the d-state wave function, which has the properties of high delocalization and isotropic distribution [1]. Ag3PO4 has strong oxidation capacity, which can be attributed to the suitable valence band position allowing to generate numerous holes. Further, its narrow band gap (2.5 eV) provides it with excellent visible-light absorption capability. Moreover, Ag3PO4 has the advantages of photosensitivity, easy preparation, and non-toxicity. Heretofore, Ag3PO4 is considered as an ideal material for photocatalytic activity in photocatalytic degradation. Recently, some Ag3PO4 adorned semiconductor systematic photocatalysts have been considered. Li's group [2] obtained a Ag2WO4/Ag3PO4 hetero-structure, which exhibited a highly efficient mineralization of bisphenol A under simulated solar light. Pang et al. [3] demonstrated that Ag3PO4 could completely degrade rhodamine B (RhB) after 36 h and the degradation rate of RhB by Ag3PO4 polypods (0.08099 h−1) was 46 times than that by N-doped TiO2 (0.00173 h−1). Further, Ag3PO4/AgI prepared by an in situ anion-exchange method showed efficient electron transfer rate and good stability [4]. However, Ag3PO4 particles easily agglomerate, inhibiting the electron transport during photocatalytic degradation. The strategy of introducing semiconductor materials as supports, with the properties of facile synthesis and low cost, can be considered to improve the activity of photocatalysts [5].
As one kind of support material, graphitic carbon nitride (g-C3N4), has rich resources, excellent optical properties, and recyclable characteristics. Besides, g-C3N4 nanosheets (CNs) provide numerous surface reactive sites because of their high specific surface area [6, 7]. Moreover, CNs have a shorter transfer/diffusion distance than g-C3N4, reducing the recombination probability of the photo-generated charge carriers. CN, as a supporter, is an ideal material to solve the uncontrolled aggregation problems [8-10]. To date, many studies have shown that the Z-scheme heterostructure can improve the performance of photocatalysts. Zhu et al. [11] synthesized ZnO/CeO2, a direct Z-scheme heterostructure, which displayed a high degradation rate of RhB. The Z-scheme g-C3N4/Bi4NbO8Cl heterostructure, which was prepared by You's group, solved the issues of high redox potentials and low charge separation efficiency [12]. Further, an efficient visible-light-driven photoelectrical-converted Z‑scheme CdTe-Bi2S3 heterostructure was synthesized by Liu et al. [13]. Considering the certainty that a Z‑scheme heterostructure reduces the carrier recombination rate, the Ag3PO4/g-C3N4 composite catalyst belongs to this type [14, 15]. Therefore, the strategy of introducing CN-dispersed Ag3PO4 and forming heterostructures has been considered as a favorable approach.
Noble-metal-modified semiconductor photocatalysts for wastewater treatment have attracted widespread attention. They serve as positively contributed, producing the localized surface plasmon resonance (LSPR) effect, for the process of visible-light-driven photocatalysis [16, 17]. When the photon energy is consistent with the resonance energy of its oscillation, the catalyst can produce the LSPR effect, causing excellent free electron mobility. The LSPR effect can be triggered through the combination of Ag and semiconductors. Hence, Ag-modified semiconductors are extensively researched [18, 19]. For example, Jeong et al. [20] acquired a Ag/BiVO4 composite whose SPR effect enhanced photocatalytic water splitting. Shen's group [21] synthesized a green Ag-TiO2 catalyst with excellent performance for removing dyes and oxidizing desulfurization. A novel three-dimensional Ag/AgBr/BiVO4 photocatalyst successfully prepared by Wei's group exhibited excellent photocatalytic performance [22]. Therefore, an Ag/Ag3PO4 complex can improve the absorption range of visible light and speed up the electronic transmission rate.
Poly N-isopropyl acrylamide (PNIPAM), as a thermo-responsive polymer, has reversible hydrophilic-to-hydro- phobic conversion around the lower critical solution temperature (LCST) (32 ℃) [23-25]. With temperature decrease, the hydrophilicity of PNIPAM is increased, showing changes from shrinking agglomeration to a stretching swelling status [26, 27]. Therefore, constructing a temperature-sensitive composite photocatalytic material can effectively utilize the temperature of the reaction system to achieve the regulation of the photocatalytic degradation process. PNIPAM also serves as a protective layer, preventing the loss of photosensitive materials.
Herein, a temperature-sensitive responsive Z-scheme PNIPAM@Ag/Ag3PO4/CN heterostructure was prepared by precipitation and emulsion polymerization methods to photodegrade tetracycline (TC) under visible light. Interestingly, with the PNIPAM@Ag/Ag3PO4/CN composite a responsive control of temperature was achieved. Moreover, the PNIPAM@Ag/ Ag3PO4/CN composite prevented the loss of Ag3PO4 and increased the adsorption capacity for the TC molecules. The photocatalytic activity of the Z-scheme PNIPAM@Ag/Ag3PO4/CN heterostructure was measured by the degradation of TC. Furthermore, a strong interfacial interaction between CN and Ag3PO4 promoted rapid migration of the electron–hole pairs. These results showed that PNIPAM@Ag/Ag3PO4/CN realized the reversible conversion property at different temperatures and remarkably improved the photocatalytic degradation performance of the photocatalyst.
Ag/Ag3PO4/CN photocatalyst was synthesized by the chemical precipitation method. Specifically, cyanuric acid and melamine in a molar ratio of 1:1 were mixed in 120 mL of ethanol. Then, the mixture was calcinated at a high temperature (550 ℃) with a heating rate of 2.3 ℃/min for 8 h. CN was obtained by a simple one-step method. Moreover, 0.4 g of yellow products were dispersed in 150 mL water and sonicated for 1 h to achieve uniform solution. Approximately 0.073 g of Na3PO4 was dispersed in 40 mL water and stirred for 30 min. After that, 2.9 mL of 0.1 mol/L AgNO3 solution was added dropwise under an ultrasonic procedure for 3 h. Additionally, the reactant was irradiated for 20 min by a 250-W UV lamp to produce a part of Ag. Lastly, the reactant was transferred to a 100-mL reactor and heated to 150 ℃ for 12 h. The final product was centrifugated for several times by water and ethanol and then vacuum dried at 60 ℃ for 12 h. The resulting product is called as Ag/Ag3PO4-10/CN. Furthermore, different mass ratios of the composite photocatalysts were obtained by adding different amounts of Na3PO4 and AgNO3, respectively. These are referred as Ag/Ag3PO4-10/CN, Ag/Ag3PO4-15/CN, Ag/Ag3PO4-20/CN, Ag/Ag3PO4-25/CN, Ag/Ag3PO4-30/CN, Ag/Ag3PO4-40/CN, and Ag/Ag3PO4-50/CN.
In a distinctive procedure, the PNIPAM@Ag/Ag3PO4-20/CN photocatalyst was synthesized by the emulsion polymerization method. Here, 0.5 g of the Ag/Ag3PO4-20/CN complex was dispersed into 60 mL of ethanol and 10 mL of 3-(trimethoxysilyl) propyl methacrylate (MPS) with magnetic stirring, using a water bath to maintain a constant temperature (50 ℃) under nitrogen (N2) atmosphere. When the reaction was completed, the product was centrifuged and dried, and is named as Ag/Ag3PO4-20/CN. After that, 0.2 g of N-isopropylacrylamide (NIPAM) and 0.03 g of N, N′-methyl-enebis (acrylamide) (MBA) were mixed into 50 mL of water by sonication. Then 0.2 g of the surface-modified Ag/Ag3PO4-20/CN was added. Ammonium persulphate ((NH4)2S₂O₈) was added as an initiator at 60 ℃ for 12 h in the presence of N2. Lastly, PNIPAM@Ag/Ag3PO4-20/CN was washed a few times with water and ethyl alcohol. Finally, the products were dried under vacuum at 60 ℃ for 12 h.
The crystalline properties were characterized by the powder X-ray diffraction (XRD model MAC Science, Japan) with Ni-filtrated Cu Kα radiation. The scan range of 2θ was 10°–80° at a scan rate of 5°/min. The elements on the surface of photocatalysis and surface electronic states were measured on X-ray photoelectron spectroscopy (XPS) by thermo ESCALAB 250X (America) electron spectrometer. The morphology of as-prepared samples and energy dispersive spectrometer (EDS) images were observed by scanning electronic microscopy (SEM) on an S-4800 field emission. The transmission electron microscopies (TEM) of as-prepared samples were obtained on JEM-2010 electron microscope with accelerating voltage of 200kV. Fourier transforms infrared (FT-IR) spectra of the samples were recorded with KBr pellets with a Bruker Vertex 70 spectrometer from 4000–500 cm−1. Raman spectra were measured on a confocal laser micro-Raman spectrometer (Thermo Fischer DXR, USA). Specific surface area was detected by the Brunauer-Emmett-Teller (BET) technique (TristarⅡ3020) from the N2 adsorption isotherm, the pore size distribution was determined by desorption isotherm using the Barrett-Joiner-Halenda (BJH) method. UV-Vis diffuse reflectance spectra (DRS) of the samples were obtained from an UV-2450 using BaSO4 as a reflectance standard. The diffuse reflectance spectra (DRS) of the catalysts were performed in the range of 250–850 nm using a UV-Vis spectrophotometer (UV-2450; Shimazu, Japan) equipped with an integrating sphere. Transient photocurrent response used Na2SO4 solution (0.5 mol/L) as a supporting electrolyte in the voltage of 0.5 V. Electrochemical impedance spectroscopy (EIS) analysis were measured by using a CHI 760E electrochemical workstation. The photoluminescence (PL) spectra were detected on a F4500 (Hitachi, Japan) photoluminescence detector. Thermogravimetric analysis (TGA) of the samples was conducted by a Diamond TG/DTA Instrument (STA 449C Jupiter, Netzsch, Germany). Angle meter was investigated using an optical contact angle-measuring device (KSV, CM200, Finland). The active species of ∙OH and ∙O2− radicals were verified with 5, 5-diamethyl-1-pyrroline N-oxide (DMPO) at room temperature by electron spin resonance (ESR) (Bruker A300, Germany) experiment.
In this work, TC was used to analyze the performance of the samples at different temperatures. Briefly, 50 mg of the photocatalyst was dissolved in 100 mL of TC solution with a concentration of 20 mg/L, and a constant temperature was maintained by passing it through condensed circulating water. The solution was reacted for 30 min first in the dark until it reached an adsorption/desorption balance and then under visible light by a 250-W Xenon arc lamp through a filter for 2 h. A sample was collected every 15 min and was immediately centrifuged and filtered through the filter. Finally, the filtrate was detected by UV-vis spectrophotometry, exhibiting a maximum absorption peak at 357 nm. The degradation rate could be calculated as follows:
where Ct means the remaining concentration after irradiation and C0 represents the initial concentration of the TC solution.
In order to further prove that PNIPAM has the ability of improving the adsorption properties of Ag/Ag3PO4-20/CN, we conducted adsorption experiments for the removal of TC, which are shown in Fig. 1. In detail, 50 mg of each sample was dissolved in 100 mL of the TC solution having a concentration of 20 mg/L, in a 250-mL flask. The adsorption experiment was carried out for 90 min, and the target solution was collected every 10 min. The concentration of the TC solution was tested by UV-Vis spectrophotometry. The removal efficiency and adsorption capacity were calculated by the following formulas (Eqs. (2) and (3)):
where Qe (mg/g) is called the adsorption amount reaching adsorption equilibrium, and Ce (mg/L) and V (mL) are the equilibrium concentration and solution volume, respectively. The initial concentration and the concentration of any time t of TC are C0 and Ct, respectively. In addition, W (mg) is the quality of the photocatalyst involved in the reaction [28]. We can find that the removal efficiency is high in the first 10 min and then gradually decreases. The removal efficiencies of Ag/Ag3PO4- 20/CN and PNIPAM@Ag/Ag3PO4-20/CN for TC are 25.27% and 32.65%, respectively. In addition, the adsorption ability of PNIPAM@Ag/Ag3PO4-20/CN (Qe = 13.059 mg/g) is improved compared to Ag/Ag3PO4-20/CN (10.1084 mg/g).
In order to further study the kinetic properties, the pseudo-first-order kinetic (FOK) model and pseudo-second-order kinetic (SOK) model were used to fit the adsorption data. Table 1 shows the corresponding parameters. The kinetic curves of Ag/Ag3PO4-20/CN and PNIPAM@Ag/Ag3PO4-20/CN quickly reach equilibrium within 30 min, as shown in Fig. 2A. This phenomenon is attributed to the fact that the adsorption on the outer surface of the sample can quickly reach saturation. The correlation coefficients of Ag/Ag3PO4-20/CN and PNIPAM@Ag/Ag3PO4-20/CN are presented in Table 1. PNIPAM@Ag/Ag3PO4-20/CN exhibits SOK during the adsorption process, for which the correlation coefficient (0.9986) is greater than that for FOK (0.9967). Furthermore, PNIPAM@Ag/Ag3PO4-20/CN shows a higher value of K2 than Ag/Ag3PO4-20/CN, indicating that PNIPAM@Ag/Ag3PO4-20/CN has excellent adsorption capacity, which agrees with the experimental results. The kinetic adsorption data are used to better research the kinetic properties of the adsorption process by the intra-particle diffusion model, as displayed in Fig. 2B. The related fitting parameters of the intra-particle diffusion model are listed in Table 2. According to the intra-particle diffusion model, the fitted line has three curves, which present a multi-linear relationship. This indicates that the PNIPAM@Ag/Ag3PO4-20/CN photocatalyst adsorption process has multiple rate control steps and its slope gradually decreases. The process is mainly divided into the following three steps. Firstly, the TC molecules rapidly diffuse from the original solution to the catalyst surface. Second, the TC molecules on the surface of the adsorbent diffuse into the channel. Finally, the adsorption tends to reach equilibrium [29].
The photocatalytic activities of the as-obtained samples were verified by photodegrading TC. Prior to the photocatalytic tests, the prepared samples underwent adsorption reactions for 30 min under dark conditions. Fig. 3A shows the activities of different mass ratio samples (Ag/Ag3PO4-10/CN, Ag/Ag3PO4- 15/CN, Ag/Ag3PO4-20/CN, Ag/Ag3PO4-25/CN, Ag/Ag3PO4-30/ CN, Ag/Ag3PO4-40/CN, and Ag/Ag3PO4-50/CN). We find that the different mass ratio Ag/Ag3PO4/CN photocatalysts have different photocatalytic activities: 64.36%, 77.27%, 84.82%, 79.93%, 76.63%, 66.27%, and 61.70%, respectively. When Ag3PO4 is above about 20%, it caused a reduced photocatalytic activity, which can be attributed to the additional Ag3PO4 granules being adhered to the surface of CN, inhibiting the contact with the contaminants in the photocatalyst system [30]. It is noteworthy that Ag/Ag3PO4-20/CN exhibits excellent catalytic performance. Moreover, PNIPAM@Ag/Ag3PO4-20/CN (88.96%) exhibits a much higher activity than Ag/Ag3PO4-20/ CN, as displayed in Fig. 3C. This demonstrates that PNIPAM accelerates the electron transport rate and increases the photocatalytic performance. The corresponding first-order linear dynamic curves between ln (C0/Ct) and the time are presented in Figs. 3B and 3D. The rate constants (k) for the as-prepared samples are obtained from the slope. Clearly, the PNIPAM@Ag/Ag3PO4-20/CN photocatalyst exhibits the highest k value (0.0207 min−1), which is 7.14 and 1.05 times higher than those of CN (0.0029 min−1) and Ag/Ag3PO4-20/CN (0.0198 min−1), respectively. Therefore, the photocatalytic property of PNIPAM@Ag/Ag3PO4-20/CN presents a comparatively more excellent performance for the photodegradation of TC.
In order to test the temperature-sensitive properties of the synthesized samples, we tested the degradation of TC at different temperatures. Figure 3E shows that the PNIPAM@Ag/Ag3PO4-20/CN photocatalyst achieves thermo-response below and above the LCST values. The photocatalytic degradation rates (88.96%, 82.42%, 76.54%, 69.97%, 65.87%, and 56.73%) are displayed in Fig. 3E. They indicate that the degradation efficiency is reduced with increasing ambient temperature. Therefore, intermolecular hydrogen bonds are formed between the amide groups and water molecules at a low temperature, so that a stable hydrated structure is formed around the hydrophobic group of the macromolecule. Further, the polymer molecular chain swells in water and stretches into a hydrophilic form, when the random coil state releases the Ag/Ag3PO4-20/CN photocatalyst. When T > LCST, the hydrogen bonds between the amide groups and water molecules are replaced by the intramolecular hydrogen bonds formed between the C=O and –N–H in the amide group. Thus, the hydration structure is destroyed, resulting in a rapid collapse of the polymer chain, which curls into a hydrophobic tight ball. Therefore, the catalyst is so packaged that it is difficult to be in contact with the contaminants, reducing the catalytic activity. Furthermore, the UV absorption peak intensity of TC at 357 nm is significantly reduced and has a certain positional shift, as presented in Fig. 3F. This demonstrates that the TC pollutant is degraded into other small molecular substances. Therefore, PNIPAM@Ag/Ag3PO4-20/CN achieves controllability to the surrounding environment and has excellent photocatalytic degradation ability [31].
In order to evaluate the mineralization ability of the as-prepared photocatalyst for contaminants during photocatalytic degradation, the total organic carbon (TOC) test results are presented in Fig. 4. It can be seen that only 37.21% and 47.63% TOC are removed using pure CN and Ag/Ag3PO4-20/CN, respectively. However, the removal rate of TOC reaches up to 69.24% within 120 min using PNIPAM@Ag/Ag3PO4-20/CN. This indicates that the PNIPAM@Ag/Ag3PO4-20/CN photocatalyst has a higher mineralization efficiency for TC photodegradation than CN and Ag/Ag3PO4-20/CN [32].
It is well comprehended that the stability of a material is an important basis for evaluating the quality and further application of the photocatalyst. The PNIPAM@Ag/Ag3PO4-20/CN and Ag/Ag3PO4-20/CN composites were tested by four recycling experiments, which are shown in Fig. 5A. The degradation efficiency of Ag/Ag3PO4-20/CN decreases slightly after four cycles of the degradation of TC. It is worth noting that the stability of the PNIPAM@Ag/Ag3PO4-20/CN composite is significantly enhanced comparing to that of Ag/Ag3PO4-20/CN. This indicates that the introduction of PNIPAM can effectively inhibit the photo-etching and deactivation of Ag3PO4. To the further research the stability of PNIPAM@Ag/Ag3PO4-20/CN, the X-ray diffraction (XRD) patterns of PNIPAM@Ag/Ag3PO4-20/CN for the degradation of TC before and after the four recycling experiments are displayed in Fig. 5B. The characteristic diffraction peak position and crystal structure of PNIPAM@Ag/Ag3PO4-20/CN show little change, indicating good stability when PNIPAM is added as a photoactive protective layer on the surface of Ag/Ag3PO4-20/CN.
In order to prove that the PNIPAM@Ag/Ag3PO4-20/CN material exhibits different hydrophilic and hydrophobic properties after phase transformation at different temperatures, the contact angle test is presented in Fig. 6. It shows photographs of the contact angles when the surface of PNIPAM@Ag/Ag3PO4-20/CN is in contact with a water droplet within 8 s at 25 and 45 ℃ in the same plane. At 25 ℃, the shape of the water droplet on the surface of PNIPAM@Ag/Ag3PO4-20/CN gradually becomes flat with increasing time, as displayed in Fig. 6A. In addition, the contact angle value of the PNIPAM@Ag/Ag3PO4-20/CN surface gradually decreases from 69.0° to 31.3°, which indicates that the composite presents a hydrophilic state in Fig. 6B. Furthermore, Ag/Ag3PO4-20/CN is released to participate in the degradation reaction, because PNIPAM exhibits a diastole state at 25 ℃. The shape of the water droplet on the surface of PNIPAM@Ag/Ag3PO4-20/CN hardly changes within 8 s with increasing time, at 45 ℃. PNIPAM tightly covers the catalyst so that PNIPAM@Ag/Ag3PO4-20/CN exhibits a contracted state [33, 34]. The above results prove that PNIPAM@Ag/Ag3PO4-20/CN achieves controllability of degradation at different temperatures.
In order to further observe the dispersion of the PNIPAM@Ag/Ag3PO4-20/CN system in the process of a reversible temperature change from 20 to 45 ℃ in water, we conducted a corresponding test as shown in Fig. 7. At 25 ℃, PNIPAM@Ag/Ag3PO4-20/CN is dispersed in water to form a homogeneous solution, showing good dispersibility in water. With temperature increase, as shown in Fig. 7(b–e), PNIPAM@Ag/Ag3PO4-20/CN begins to aggregate and finally agglomerates to precipitate at the bottom of the bottle. As shown in Fig. 7(g–j), PNIPAM@Ag/Ag3PO4-20/CN changes from its aggregated state to the original dispersion solution as the temperature is lowered. This indicates that the as-prepared catalyst has a temperature-sensitive function[35, 36].
The typical XRD diffraction patterns of the as-prepared composites are exhibited in Fig. 8. The notable peaks located at 20.9°, 29.7°, 33.4°, 36.6°, 47.6°, 52.7°, 54.9°, 57.3°, 61.7°, 70.1°, and 72.2° are indexed to the (110), (200), (210), (211), (310), (222), (320), (321), (400), (420), and (421) crystal planes of Ag3PO4 (PDF = 43-0542), respectively [37, 38]. Notably, four distinct peaks are detected, appearing at 38.1°, 44.3°, 64.2°, and 77.5° in the composite, which are indexed to the Ag (111), (200), (220), and (311) planes, respectively [39]. The diffractive peaks at 12.9° and 27.8° are attributed to the (100) and (002) planes of CN owing to the layer stacking factor [40]. This indicates that CN, Ag3PO4, PNIPAM@Ag/Ag3PO4-20/CN, and the different mass ratios of Ag/Ag3PO4/CN are successfully prepared. No diffraction peaks are observed in Ag/Ag3PO4-10/CN because of the low content of Ag3PO4. When PNIPAM is added, there are no significant changes in the shapes of the peaks of the Ag/Ag3PO4-20/CN composite, indicating that PNIPAM does not change the peak structure of the complex [41].
In order to further detect the surface element compositions and electronic states of the as-prepared photocatalysts, the results of the X-ray photoelectron spectroscopy (XPS) are exhibited in Fig. 9. The survey spectrum (Fig. 9A) verifies that the PNIPAM@Ag/Ag3PO4-20/CN samples contain C, N, Ag, P, and O elements, which is in accordance with the XRD results. As shown in Fig. 9B, the C 1s spectrum includes three peaks at approximately 283.7, 287.2, and 397.9 eV, which are attributed to the binding energies of C=C, C–N, and N=C–N, respectively [42, 43]. In Fig. 9C, the N 1s spectrum can be fitted into four peaks located at 397.9, 399.3, 400.3, and 403.6 eV, which are ascribed to the bonds of C=N–C, N–C3, and N–H and π excitation [44, 45]. In Figs. 9D and F, we note that the peaks of PNIPAM@Ag/Ag3PO4-20/CN and Ag/Ag3PO4-20/CN have a slight shift toward a higher binding energy than those of Ag3PO4. This indicates that there is an interaction between Ag3PO4 and CN [46, 47] and that the Ag element has been successfully prepared. Moreover, the binding energy peaks at 530.2 and 532.1 eV are in good accordance with the O 1s state, and they are assigned to the Ag–O and –OH group, respectively [48, 49]. The XPS results further verify that the PNIPAM@Ag/Ag3PO4-20/CN heterostructure is obtained.
Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were performed to determine the morphology, and the results are presented in Figs. 10 and 11. The SEM images of CN are displayed in Fig. 10A and 10B. It can be seen that uniform ultra-thin carbon nitride nanosheets are synthesized. Further, Ag3PO4 granules are clearly exhibited on the surface of CN. As shown in Figs. 10D and 10E, PNIPAM@Ag/Ag3PO4-20/CN presents a blurred surface compared to the Ag/Ag3PO4-20/CN surface, which shows that PNIPAM is successfully covering the surface of Ag/Ag3PO4-20/ CN. It is noticeable that the elements of C, N, Ag, P, and O are discovered in the energy-dispersive X-ray (EDX) analysis of the PNIPAM@Ag/Ag3PO4-20/CN photocatalyst. Fig. 11A exhibits that the CN morphology has a wrinkled lamellar structure and is almost transparent. Moreover, the black Ag3PO4 particles of the as-prepared Ag/Ag3PO4-20/CN display excellent dispersion in Fig. 11B. It is easy to find that there exists a layer of a substance, like a fog, on the surface of the CN and Ag/Ag3PO4 particles. This indicates that Ag/Ag3PO4-20/CN is successfully modified by PNIPAM. The distributions of C, N, Ag, P, and O in the mapping images of PNIPAM@Ag/Ag3PO4-20/CN further confirm that the as-prepared materials are synthesized.
The Fourier transform infrared (FTIR) spectra of CN, Ag3PO4, PNIPAM, Ag/Ag3PO4-20/CN, and PNIPAM@Ag/Ag3PO4-20/CN are displayed in Fig. 12. The typical peaks at 810, 1100–1750, and 3100–3400 cm−1 are attributed to the characteristic respiratory peak of the tri-s-triazine units, telescopic vibration of the C–N heterocycles, and N–H groups, respectively [50]. For the Ag3PO4 sample, the two peaks at 1045 and 843 cm−1 correspond to the P–O stretching oscillation in PO43− [48, 51]. The position at around 3150 cm−1, marked with a dotted line, is the N–H symmetric stretching peak of CN [52]. We can see that the PNIPAM@Ag/Ag3PO4-20/CN material has a new peak of PNIPAM at 2976 cm−1, indicating that Ag/Ag3PO4-20/CN is successfully modified by PNIPAM. To further research the structural compositions of the as-prepared materials, the Raman spectra are presented in Fig. 13. For Ag3PO4, the Raman peaks at 558 and 913 cm−1 correspond to the oscillation of PO43− [38]. The Raman spectrum of CN is dominated by the vibrational modes of C–N at the vibration absorption peaks of 1158, 1232, 1475, and 1649 cm−1 [53, 54]. We have found that the Ag/Ag3PO4-20/CN composite contains the peaks of C–N and PO43−, which indicate that Ag3PO4 and CN coexist in the composites. In the PNIPAM@Ag/Ag3PO4-20/CN photocatalyst, all the peaks are shifted to a low wavenumber, indicating that Ag/Ag3PO4-20/CN is modified by PNIPAM [55].
The specific surface area is an influential factor for enhancing photocatalytic activity, because a large specific surface area has more surface-active sites. The adsorption-desorption characteristics of N2 are presented in Fig. 14. The adsorption-desorption isotherms are type IV with a hysteresis loop[56, 57]. The Brunauer-Emmett-Teller (BET) surface area of CN is 91.0253 m2/g, as presented in Table 3. In comparison, the BET area of Ag/Ag3PO4-20/CN is lowered to 74.9156 m2/g, which is to the loading of Ag3PO4 particles on the CN surface. Moreover, PNIPAM@Ag/Ag3PO4-20/CN has a larger surface area (83.9404 m2/g) than the Ag/Ag3PO4-20/CN photocatalyst. This indicates that PNIPAM can effectively increase the specific surface area of Ag/Ag3PO4-20/CN. At the same time, the photocatalytic activity of PNIPAM@Ag/Ag3PO4-20/CN is higher than that of Ag/Ag3PO4-20/CN. The results show that one of the factors affecting the photocatalytic activity is the surface area of the photocatalyst.
The light absorption property and energy band characteristics of Ag3PO4, Ag/Ag3PO4-20/CN, PNIPAM@Ag/Ag3PO4-20/CN, and CN were analyzed by UV-vis diffuse reflection spectroscopy (DRS). Ag3PO4 and CN exhibit a distinct absorption edge at the positions of 521 and 473 nm, as presented in Fig. 15A [58]. In comparison to the spectra of Ag/Ag3PO4-20/CN and PNIPAM@Ag/Ag3PO4-20/CN, they exhibit a slight red shift, implying that the ability of the response to visible light is improved [59]. Ag3PO4 and CN can form a heterostructure, which can easily generate electrons and holes to improve the light absorption capacity. Moreover, the band gap energy (Eg) of Ag3PO4 and CN is evaluated using the following equation:
where α, ν, h, A, and Eg denote the absorption coefficient, the light frequency, Planck's constant, a constant, and the band gap energy, respectively[60]. Further, n for both Ag3PO4 and CN is 1 [61, 62]. The Eg values are computed by Eq. (4), and those of Ag3PO4 and CN are 2.38 and 2.62 eV, respectively. The Eg values are also given in Fig. 15B.
To investigate the photoelectrochemical performances of the as-obtained materials, the transient photocurrent response and electrochemical impedance spectroscopy (EIS) results are presented in Fig. 16. Rapid and stable photocurrent responses of CN, Ag3PO4, Ag/Ag3PO4-20/CN, and PNIPAM@Ag/ Ag3PO4-20/CN are detected under visible-light illumination. Importantly, the photocurrent responses of the Ag/ Ag3PO4-20/CN and PNIPAM@Ag/Ag3PO4-20/CN composites are much higher than those of the pure CN and Ag3PO4. This indicates that a heterostructure is formed between CN and Ag3PO4, which accelerates the electron transport and improves the separation efficiency of the electrons and holes [63, 64]. Meanwhile, PNIPAM@Ag/Ag3PO4-20/CN exhibits a higher photocurrent response than Ag/Ag3PO4-20/CN, revealing that PNIPAM efficaciously elevates the electron and hole separation and transfer. In addition, the EIS are presented in Fig. 16B. The arc radius of PNIPAM@Ag/Ag3PO4-20/CN is the smallest among those of CN, Ag3PO4, and PNIPAM@Ag/Ag3PO4-20/CN, indicating that PNIPAM@Ag/Ag3PO4-20/CN has a relatively higher surface reaction rate [65, 66]. Therefore, PNIPAM@Ag/Ag3PO4-20/CN possesses a high charge separation efficiency after the addition of PNIPAM.
To research the catalytic efficiencies of the photocatalysts, we studied the photoluminescence (PL) spectra and the free recombination of the photoinduced electrons and holes in each sample by PL spectroscopy [41]. The samples were excited at 360 nm. Fig. 17 shows the PL spectra of the CN, Ag/Ag3PO4-20/CN, and PNIPAM@Ag/Ag3PO4-20/CN samples, and the CN exhibits that the strongest emission peak is located at 467 nm. This is attributed to the high recombination rates of the carriers in CN. In addition, the PL intensity of PNIPAM@Ag/Ag3PO4-20/CN is remarkably lower than that of Ag/Ag3PO4-20/CN, illustrating that the recombination rate of PNIPAM@Ag/Ag3PO4-20/CN is smaller than that of Ag/Ag3PO4-20/CN. This indicates that the catalyst extends the life of the charge carrier and reduces the electron-hole pair binding rate after the introduction of PNIPAM [67]. Therefore, the PL results illustrate that the Z-scheme PNIPAM@Ag/Ag3PO4-20/CN heterostructure can remarkably improve the charge transfer and the photocatalytic performance in the process of photocatalytic degradation [68, 69].
To reveal the thermal stability of the as-prepared materials, thermogravimetric (TG) analysis was performed. The temperature detection range was from room temperature to 800 ℃ at a heating rate of 10 ℃/min. The TG thermograms for CN, Ag3PO4, PNIPAM, Ag/Ag3PO4-20/CN, and PNIPAM@Ag/ Ag3PO4-20/CN are clearly displayed in Fig. 18. It is clear that CN undergoes a weight loss in the range of normal temperature to 100 ℃. The loss of mass corresponds to the quantity of water, and it is caused by the adsorption of a small amount of water vapor from the outside of the material. Furthermore, CN becomes unstable when the heat temperature is above 530 ℃, whereas heating to 690 ℃ does not yield any residue of the material being observed [70]. In addition, Ag3PO4 has a weight loss after 500 ℃, probably caused by the presence of residual compounds in the sample [71]. For PNIPAM, there are three weightless stages. The weight loss phase before 100 ℃ may be owing to the sample removing water and organic solvents. When the temperature is in the range of 200 to 500 ℃, the molecular chain in PNIPAM will dehydrogenate as the temperature increases, during which stage, the dopant will be released. When the temperature continues to rise to 400 ℃, the molecular chain of PNIPAM will break and decompose [72]. This phenomenon will lead to serious weight loss in the thermogravimetric analysis curve of PNIPAM [73, 74]. Furthermore, PNIPAM@Ag/Ag3PO4-20/CN has more residue than Ag/Ag3PO4-20/CN at 694 ℃, indicating that the PNIPAM@Ag/Ag3PO4-20/CN composite has high thermal stability.
To ascertain the active species in the degradation process, holes, ∙O2−, and ∙OH active species were trapped by ethylenediamine tetraacetic acid disodium salt (EDTA-2Na), ascorbic acid, and isopropyl alcohol (IPA), respectively. When ascorbic acid was added, the degradation rate of TC significantly decreased (20.96%), compared to that without scavengers (88.96%) (Fig. 19). Therefore, the ∙O2− radicals play a major role in the degradation process. In contrast, the degradation rate (62.11%) of TC is slightly lower with the addition of IPA, indicating that the ∙OH radicals also participate in the degradation reaction. When EDTA-2Na is added, the degradation rate of TC is also significantly decreased (36.61%). Therefore, holes are also the main active species. ESR is further performed to explore the active species in a reaction. As can be seen from Fig. 19, no ESR signals are observed in the sample under dark conditions. For CN, Ag/Ag3PO4-20/CN, and PNIPAM@Ag/ Ag3PO4-20/CN, the characteristic peaks of ∙O2− radicals are noted in Fig. 20A. Furthermore, the PNIPAM@Ag/Ag3PO4-20/ CN sample displays the strongest intensities, compared to those of CN and Ag/Ag3PO4-20/CN. This reveals that more ∙O2− radicals are produced in PNIPAM@Ag/Ag3PO4-20/CN. Concurrently, Fig. 20B reveals that the ∙OH signals of the PNIPAM@Ag/Ag3PO4-20/CN complex also exhibit a stronger intensity than those of CN and Ag/Ag3PO4-20/CN, indicating that the PNIPAM@Ag/ Ag3PO4-20/CN complex has the ability to generate numerous ∙OH radicals. Consequently, ∙O2− and ∙OH play a significant role in the photocatalytic degradation process.
PNIPAM, as a thermo-sensitive polymer, has the property of hydrophilic–hydrophobic reversible conversion near the LCST temperature. At low temperatures, the hydrophilicity of PNIPAM is enhanced. Moreover, its morphology changes from a shrinking agglomeration to a stretch swelling. Therefore, the Ag/Ag3PO4-20/CN catalyst is released and the PNIPAM@Ag/ Ag3PO4-20/CN composite achieves temperature controllability. Moreover, the PNIPAM surface can improve the adsorption of the TC molecules. The band gaps of CN and Ag3PO4 are 2.62 eV and 2.38 eV, respectively, which are presented in Fig. 15B. In addition, the conduction band (CB) positions of CN and Ag3PO4 are detected by the Mott-Schottky technology in Fig. 21. Thus, the valence band (VB) position of CN and Ag3PO4 can be calculated by Eq. (5):
where Eg, EVB, and ECB are the band gap of the semiconductor, VB energy, and CB energy, respectively. The values of ECB are -1.11 and 0.31 eV for CN and Ag3PO4, respectively. Therefore, the EVB values of CN and Ag3PO4 are 1.51 and 2.69 eV, respectively. The calculated results are consistent with the result of the XPS VB spectra presented in Fig. 20. Based on the analyses of the above results, the possible mechanism of the photocatalytic TC degradation reaction over PNIPAM@Ag/ Ag3PO4-20/CN is proposed in Fig. 22. When PNIPAM@Ag/ Ag3PO4-20/CN is illuminated by visible light, CN and Ag3PO4 are excited and electrons are transferred from the CB to the VB. Moreover, the holes remain in the VB of CN and Ag3PO4. If a hole is transferred from the VB of Ag3PO4 to CN, it will not oxidize OH− to produce ∙OH. This is because the VB position of CN is 1.51 eV, which has a lower value than the standard oxidation–reduction potential of OH−/∙OH (1.99 eV) [75]. Moreover, the electrons in the CB of Ag3PO4 cannot reduce O2 to generate ∙O2− because the CB potential of Ag3PO4 (0.31 eV) is lower than the standard redox potential of O2/∙O2− (−0.33 eV) [76]. Therefore, the Z-scheme Ag/Ag3PO4-20/CN heterostructure must be prepared because ∙O2− and ∙OH are discovered by the ESR spin-trap technique and trapping experiments. The electrons in the CB of Ag3PO4 and holes in the VB of CN recombine in the inner heterostructure. Therefore, the separation efficiency of the electrons and holes is increased in the CB of CN and in the VB of Ag3PO4. Meanwhile, Ag nanoparticles (NPs) can be used as electron transport receptors, accelerating the electron transfer from the CB of Ag3PO4 to the VB of CN. Meanwhile, electrons receive more energy to transfer by the LSPR effect from the Ag NPs to CN across the barrier. Further, the O2 captures the electrons of the Ag NPs, producing ∙O2− radicals. Then, the electrons in the CB of CN are trapped, resulting in the O2 forming ∙O2− radicals to oxidize TC to CO2 and H2O. Simultaneously, the holes in the VB of Ag3PO4 reduce H2O into ∙OH radicals and act on TC. Moreover, the holes can directly reduce TC, forming CO2 and H2O.
In summary, an intelligent controllable Z-scheme PNIPAM@Ag/Ag3PO4/CN heterostructure is successfully fabricated by precipitation and emulsion polymerization methods, and it has an efficient and enhanced photocatalytic activity relative to those of Ag/Ag3PO4/CN, CN and Ag3PO4. The photocatalytic degradation rate at 25 ℃ (88.96%) is significantly higher than that at 45 ℃ (56.73%). This indicates that the PNIPAM@Ag/Ag3PO4/CN catalyst changes from a contracted state to a relaxed state and Ag/Ag3PO4/CN is released when T < LCST. Moreover, at 25 ℃, the contact angle value of the PNIPAM@Ag/Ag3PO4/CN surface decreases from 69.0° to 31.3°, which indicates that the composite is in a hydrophilic state. The shape of the water droplets on the surface of PNIPAM@Ag/Ag3PO4-20/CN hardly changes within 8 s with increasing time at 45 ℃. Remarkably, the PNIPAM@Ag/ Ag3PO4-20/CN photocatalyst exhibits the highest k value (0.0207 min−1), which is 7.14 and 1.05 times that of CN (0.0029 min−1) and Ag/Ag3PO4-20/CN (0.0198 min−1), respectively. The excellent photocatalytic activity can be attributed to the Ag NPs and Z-scheme heterostructure, which enhance the separation efficiency of the charge carriers. The current work achieves controllability to the surrounding environment and creates conditions for the uncontrollability of the environment during the degradation process.
2020,
Vol. 41 Issue (10): 1573-1588 DOI: 10.1016/S1872-2067(20)63554-9 
PDF
2020,
Vol. 41 Issue (10): 1573-1588 DOI: 10.1016/S1872-2067(20)63554-9 PDF