A novel p–n junction Ag₃PO₄/BiPO₄-based stabilized Pickering emulsion for highly efficient photocatalysis

Abstract

This study demonstrates a new kind of photocatalytic system via utilization of the superior specific properties inherent in Pickering emulsion. We designed the new photocatalytic system using a novel p–n heterojunction Ag₃PO₄/BiPO₄ (AB) as a photocatalytic active component, multiwalled carbon nanotubes (MWCNTs) and graphene (GR) as a hydrophobic conducting nanostructure to form the stabilized Pickering emulsion. The photocatalytic activity of the as-prepared stabilized Pickering emulsion-based system (Pe-bp system) was studied by monitoring the change in Acid Blue 92 dye (AB92) concentration under both visible and UV light irradiation. The results revealed that the Pe-bp system exhibits a noticeable improvement in both efficiency and rate of AB92 photodegradation in comparison with the traditional solution-dispersed photocatalytic system. The observed results are discussed in terms of (1) the self-assembled nanohybrid at the water/oil (w/o) interface provides a large surface area, (2) the use of MWCNTs and GR promotes the generation of amphiphilic nanostructures self-assembled at the w/o interface, reducing the charge recombination by shuttling and capturing photogenerated electrons and (3) the great separation of the products from the reactants during the photocatalytic reactions facilitates these processes. A possible mechanism explaining the origin of enhanced performance of formed nanohybrids in the Pe-bp system is also proposed. In addition to the high efficiency, the rapid and simple procedures used for demulsifying and re-emulsifying crucially make the Pe-bp system technically simple and practically applicable for environmental remediation.

---

1. Introduction

Over the past decades, the “Green-life” concept has inspired enthusiasm to exploit novel and high-efficiency photocatalysts for environmental purification. Photocatalysis has recently been an intensively pursued topic since it has been considered as a vanguard solution focused on the viewpoint of environmental accountability and energy conversion.^1,2 The application of heterogeneous semiconductor photocatalysts in water cleaning and environmental remediation has recently attracted extensive attention.^3–6 There are generally two distinct areas in the development of a photocatalytic system. First, the focus is on designing and synthesizing nanoscale porous photocatalysts with high activity under visible light irradiation.^7–9 Second is to establish a novel photocatalytic system which creates large active surface area and facilitates efficient mass transfer on the photocatalyst surface.^10–13

To date, extensive efforts have been undertaken in both areas for the synthesis of advanced photocatalytic materials to effectively utilize visible light in order to design very effective photocatalytic systems. As is well known, photocatalytic systems have been broadly employed by either suspending a photocatalyst into a solution to generate a solution-dispersed photocatalytic system or immobilizing it onto a substrate to generate a surface-immobilized photocatalytic system. The former system provides a large surface area for the photocatalyst, but the separation of the photocatalyst from the solution remains a problem. In the latter system, however, photocatalyst separation from the bulk solution is easily achieved; the restriction of mass transport of substrate to the photocatalyst surface has not yet been solved.¹³ In searching for new approaches to improve photocatalytic degradation of hazardous contamination, a Pickering emulsion-based photocatalytic system (Pe-bp system) has stood out as a novel means which has shown its usefulness in catalytic processes.^13–15 An emulsion stabilized by solid particles rather than an organic surfactant is called a Pickering emulsion. In Pickering emulsions, the nanometer- to micrometer-sized particles are strongly adsorbed on the water/oil (w/o) interfaces because of the decrease of the total free energy.^16–20

From this point of view, a Pe-bp system has outstanding advantages in comparison to traditional photocatalytic systems. Firstly, the much dispersed photoactive solid emulsifiers self-assembled at the w/o interface greatly enhance the particle stability against aggregation, ensuring a large active surface area for photocatalysis. Secondly, photoactive solid emulsifiers can be established via the combination of photocatalyst emulsifiers and conducting nanostructure, in which the conducting nanostructure can be employed to capture and shuttle photogenerated electrons, which will reduce the recombination process of e⁻/h⁺ pairs. Thirdly, because of the different solubility in the w/o phases, the spatial separation of the reaction products from the reactants enhances the photocatalytic efficiency. Finally, Pickering emulsions can be easily demulsified through centrifuging or sonicating and re-emulsified via shaking the mixture. These advantages effectively make the Pe-bp systems technically simple and thus very applicable for various kinds of uses. However, the applicability of this photocatalysis system has not yet been proven in a wider context of photocatalysis.^13,21

Herein, the Pe-bp system was selected as a novel photocatalysis system candidate that was prepared from a novel p–n junction Ag₃PO₄/BiPO₄ (AB) photoactive solid emulsifier through sensitizing BiPO₄ nano-cocoons with Ag₃PO₄ particles, and graphene (GR) and multiwalled carbon nanotubes (MWCNTs) as a conducting nanostructure. We demonstrate a technique for degrading Acid Blue 92 (AB92) dye, which is compared to previous reported methods in an effort to elucidate the role of this combination method in photocatalytic removal. The photocatalytic activity of Pe-bp systems was measured through degrading AB92 under both visible and UV light illumination. Moreover, the roles of reactive species and the activity enhancement mechanism were also investigated in detail. This was verified by the effects of scavengers. This work mainly shows novel photocatalytic systems with a high efficiency and less technical demanding based on Pickering emulsion science and technology. The enhanced performance and the underlying mechanism of the Pe-bp system reported in this work have been scarcely investigated previously.

2. Experimental

2.1. Chemicals and apparatus

AgNO₃, Bi(NO₃)₃·5H₂O, Na₂HPO₄ and concentrated NH₃·H₂O utilized for the synthesis of Ag₃PO₄/BiPO₄ nanocomposite were purchased from Merck. For the GR synthesis, graphite powder was purchased from Fluka. Concentrated sulphuric acid, K₂S₂O₈, P₂O₅, H₂O₂ and KMnO₄ were purchased from Merck. Acid Blue 92 dye (MW = 695.58 g mol⁻¹, CI = 13 390), provided by Iran Color Research Center, was selected as a probe molecule to test the photocatalytic activity. Raw MWCNTs (purity >95%, surface area 198 m² g⁻¹) were obtained from Aldrich.

X-ray diffraction (XRD) patterns of the samples were recorded using a Philips X'pert instrument operating with Cu Kα (λ = 0.15406 nm) radiation as an X-ray source at 40 kV/40 mA. Morphological analysis was carried out via an XL30 field-emission scanning electron microscope. The hydrophilicity was measured with an optical contact angle measuring device (OCA20110524; Data Physics Instruments, Germany) by using a droplet of water as an indicator. The electrochemical measurements were carried out with an Autolab potentiostat/galvanostat (EG&G model 263A, USA).

2.2. Catalyst preparation

2.2.1. Synthesis of GR. Few-layered graphene oxide (GO) was synthesized by using natural graphite powder according to the modified Hummers' method.²² Firstly, 3.00 g of graphite powder was added into a mixture of 15 mL of H₂SO₄, 2.50 g of K₂S₂O₈ and 2.50 g of P₂O₅. After that, the suspension was made by diluting the prepared solution with DI water, filtering using 0.2 μm Nylon Millipore film, washing several times and finally drying under ambient conditions. In the second step, the oxidized graphite was added to 460 mL of H₂SO₄ in an ice bath and then stirred. Afterward, 6 g of KMnO₄ was added slowly with controlling the temperature below 10 °C, and then stirred for 2 h at 35 °C. After 2 h, 2.8 L of DI water and 50 mL of 30% H₂O₂ were added to the prepared mixture. The resulting mixture was centrifuged and washed with 20% HCl solution followed by 5 L of DI water to eliminate the acid. Finally, the product was subjected to dialysis for a week to eliminate all impurities and dried under vacuum at ambient temperature. In order to prepare GR, the synthesized GO underwent reduction by the hydrothermal method at 120 °C for 4 h.

2.2.2. Synthesis of Ag₃PO₄/BiPO₄ photoactive solid emulsifier. The Ag₃PO₄/BiPO₄ (AB) photoactive emulsifier was synthesized via a simple co-precipitation hydrothermal method in a two-step process according to a reported method.⁹ In the first step, 0.003 mol of Bi(NO₃)₃·5H₂O was dissolved in a highly dispersed solution consisting of 1 mL of concentrated HNO₃ and 8 mL of DI water. Then, AgNO₃ was added to the above solution and also sonicated for 15 min. After that, an aqueous solution of Na₂HPO₄·2H₂O was added to the above-mentioned mixture while it was stirring vigorously. The pH of the mixed solution was adjusted to 7 using NH₃·H₂O. In the second step, the resulting mixture was rapidly transferred into a Teflon-lined stainless steel autoclave under autogenous pressure and maintained at 160 °C for 6 h.

2.2.3. Emulsion preparation. In order to obtain a stable Pickering emulsion, AB and GR components were mixed by homogenizing a dispersion of GR in isooctane (0.4 mL, 1.67 g L⁻¹) with an aqueous dispersion of AB (4 mL, 7.85 g L⁻¹) in a sonication bath for 20 min. After sonicating, the water-to-oil volume fraction was adjusted to 4 : 10 by adding an appropriate volume of isooctane into the mixture while the total amount of the mixture was 14 mL. A stable emulsion was finally prepared via shaking the mixture for 2 min. An emulsion was considered as a stable one when the emulsified state was retained for more than 1 day, whereas it was recognized as unstable if it demulsified within 24 h. The prepared emulsions could be broken simply by either centrifugation for 5 min at 7000 rpm or sonication for 2 min and re-emulsified via shaking the as-broken emulsion for 2 min. The conditions for preparing the stable Pickering emulsion in the presence of MWCNTs were the same as for GR.

2.2.4. Photocatalytic degradation of AB92 dye. Acid Blue 92 (AB92) was chosen as a target molecule to investigate the photocatalytic activity of the prepared photocatalysts. Photocatalytic reaction was performed in a Pyrex reactor. A Xe arc lamp was used as the visible light source (λ ≥ 420 nm) and a 125 W mercury lamp was used as the source of UV light. In a typical experiment, 0.2 mL of dye solution with a concentration of 1.34 × 10⁻³ mol L⁻¹ was applied into the emulsion system. Prior to irradiation, the emulsion system was kept in a darkroom for 30 min to establish the adsorption/desorption equilibrium of AB92 dye on the surface of the nanohybrids to exclude the effect of dye adsorption in the photodegradation process, and then the photocatalytic reaction was started. At regular time intervals (5 min), samples of the emulsion system consisting of water and oil phases were taken out and centrifuged for 10 min at a rate of 10 000 rpm. Then, the water phase containing AB92 dye was diluted five times. The concentration of AB92 dye in water phase was analyzed using a UV-visible spectrophotometer at 572 nm recording variations of the absorption band maximum with DI water as a reference sample. For comparison, the decomposition of AB92 dye in a solution-dispersed system was conducted with AB nanocomposite. In this case, AB as well as dye concentration were kept almost the same in both systems.

In addition, to exclude the kinds of reactive species directly taking part in the AB92 photocatalytic degradation, an AB/GR emulsion system was first formed using the procedures mentioned above, with the addition of different scavengers (fluoride ion, iodide ion, persulfate and t-BuOH) into a water phase as a diagnostic tool. A control experiment was conducted as follows: 0.2 mL of AB92 solution (67 μM) was added into 3 mL of an aqueous solution of different scavengers (0.15 mM). The prepared emulsion system was illuminated for 25 min to study the effect of different scavengers on the AB92 photocatalytic degradation.

3. Results and discussion

3.1. Catalyst characterization

3.1.1. UV-visible absorption spectra of GO and GR. The UV-visible absorption spectra of GO and GR are shown in Fig. 1. The broad absorption peak around 230 nm in the UV-visible absorption spectrum of GO (Fig. 1a) is related to the π–π* transitions of unoxidized aromatic C=C bonds, and the small shoulder peak around 290–300 nm corresponds to the n–π* transition of the C=O bonds.²² After hydrothermal reduction of GO, the absorption peak around 230 nm red-shifts to higher wavelength and the shoulder disappears (Fig. 1b). This is accompanied by the change of GO solution colour from yellowish brown to black which also confirms the reduction of GO to GR by hydrothermal reduction treatment.²³

Fig. 1 UV-visible absorption spectra of (a) GO and (b) GR.

3.1.2. XRD analysis. Crystallinity of the fresh and used AB composite was analyzed by using XRD. The XRD pattern of as-prepared sample is illustrated in Fig. 2. The prepared samples were well crystallized, in which the diffraction peaks could be assigned to the known hexagonal BiPO₄ (PDF File no. 15-0766) and the cubic Ag₃PO₄ (PDF File no. 06-0505).⁹ It is interesting to note that only the diffraction peaks relating to Ag₃PO₄ and BiPO₄ can be seen for the fresh AB composite. Therefore, this shows that no metallic silver (Ag⁰) was present in the used AB composite in the Pe-bp system, suggesting that only a tiny amount of Ag⁰ had been generated.

Fig. 2 XRD patterns of fresh and used AB composites.

3.1.3. FE-SEM analysis. The surface morphology of catalysts was characterized from FE-SEM images (Fig. 3 and 4). It can be seen that the AB nanocomposite was made up of irregular nano-cocoons, while very small spherical-like particles adhered on the surface of the nano-cocoons (Fig. 4).²⁴ Based on the literature, it is noteworthy that the solubility product constants of Ag₃PO₄ (8.89 × 10⁻¹⁷) and BiPO₄ (1.3 × 10⁻²³) are so different.²⁵ Therefore, Ag₃PO₄ particles would be generated after the formation of BiPO₄ nano-cocoons. Additionally, for the preparation of AB sample, the AgNO₃ concentration was much lower than that of pure Ag₃PO₄ particle synthesis. Therefore, only low content of Ag₃PO₄ nanoparticles could be formed and inserted on the surface of BiPO₄ nano-cocoons. The results mentioned above indicate that a close heterojunction between Ag₃PO₄ and BiPO₄ cocoons could be efficiently formed, which would facilitate the effective separation of charge carriers.

Fig. 3 The SEM images of (a) MWCNTs and (b) GR sheets.

Fig. 4 SEM images of Ag₃PO₄/BiPO₄ composites.

3.1.4. EIS characteristics of GR and MWCNTs. One of the most important roles of GR and MWCNTs in the Pe-bp system is electron transport. GR and MWCNTs, as one kind of carbon nanostructure, have been proved to possess a good capability of capturing and shuttling electrons through the π–π network. As a result, the hybridization of nanocomposite photocatalysts (e.g., AB) with them would make a highly impressive approach to the establishment of the Pe-bp system. Therefore, the fast transport of electrons could be obtained and an efficient charge separation subsequently accomplished in the Pe-bp system. The electron transfer in GR and MWCNTs through the π–π network can be shown by using electrochemical impedance spectroscopy (EIS). For the sake of in-depth investigation of the conductivity of GR and MWCNTs in the Pe-bp system, Nyquist plots of these samples have been obtained. In the electrochemical plot, the arc is attributed to the charge transfer limiting process and can be assigned to the double-layer capacitance (C_dl) in parallel with the charge transfer resistance (R_ct) at the contact interface between the electrode and electrolyte solution.^26–28 As shown in Fig. 5, it is observed that the arc in the plot of GR nanosheets becomes much shorter demonstrating a decline in both solid state interface layer resistance and charge transfer resistance on the surface. Compared to the MWCNTs, the arc diameter of the GR semicircle is smaller, implying a better conductivity than that of MWCNTs.

Fig. 5 EIS changes of GR and MWCNT electrodes. The EIS measurements were carried out in the presence of a 2.5 mM K₃[Fe(CN)₆]–K₄[Fe(CN)₆] (1 : 1) mixture as a redox probe in KCl aqueous solution (0.1 M). The inset is the magnification of EIS changes of GR and MWCNT electrodes.

3.2. Emulsion characterization

3.2.1. Contact angle measurement. Surface wettability has been proven to be a key factor affecting the stability of Pickering emulsions formed by solid particles.^29–32 This factor can be described with the contact angle (θ) measured between the solid particle and the w/o interface. In general, efficiency arises at an intermediate contact angle about 90° which is optimal for stabilizing emulsions. In this manner, the emulsion stability decreases via any deviation of θ from 90°.^29,30 This is consistent with our experimental results. As shown in the schematic diagram (Fig. 6a), contact angle of the synthesized nanohybrid at the w/o interface was analysed as follows. Firstly, a cast film of AB/GR nanohybrid was soaked in isooctane. After that, one water droplet was slowly placed on the prepared film with an isooctane environment around the droplet, and then the contact angle was measured.

Fig. 6 (a) Schematic diagram of a cast film prepared for contact angle measurement and (b) the contact angle of AB/GR nanohybrid with content of ∼4 wt% GR.

We found that the particle contact angle or wettability of both AB/GR and AB/MWCNT nanohybrids can be controlled by changing the concentration of GR and MWCNTs. The stability of emulsions can be controlled as the GR and MWCNT concentration was varied, which indicates that the efficiency of emulsification improved as the emulsifier concentration changed. This provides a factor that can be controlled during emulsion formation. By increasing GR and MWCNTs concentration (Fig. 7), the stability of emulsions increased and the contact angle became closer to 90°. As shown in Fig. 7, the water contact angles of the AB/GR hybrid with higher GR mass ratios are smaller than those of the AB/MWCNT with the same mass ratios of MWCNTs. The process of emulsification along with the GR sheet absorption process at the interface of the two immiscible liquids (water and isooctane) leads to the reduction in the free energy of the system. This means that the unique structure of GR sheets enables them to be trapped at the interface and to be wrapped around the oil droplets.

Fig. 7 Contact angle (θ) as a function of the contents of GR or MWCNTs in the prepared nanohybrids.

Fig. 8 shows photographs and optical micrographs of a typical system obtained with AB/GR nanohybrids. As shown in Fig. 8a, the stable w/o emulsion is realized in the lower part of the vial with the upper part being pure isooctane phase. These behaviours are essentially the same as those observed for the typical system obtained with AB/MWCNT nanohybrids. Fig. 8b and c show photographs of a typical system obtained with AB/GR nanohybrid before and after addition of AB92 solution, respectively.

Fig. 8 (a) Photograph of Pickering emulsion stabilized by AB/GR nanohybrid (volume ratio of w/o was 4 : 10, containing ∼3 wt% GR). (b and c) Photographs of this emulsion before and after the addition of AB92 dye solution.

3.3. Photocatalytic experiments

3.3.1. Photocatalytic activity of the prepared Pe-bp system. We emphasize the significance of establishing a novel Pe-bp system, a new category of photocatalytic systems distinct from the traditional ones, which have excellent universality for degrading various pollutants. Therefore, the development of these Pe-bp systems has become one of the desired directions and most important topics in the photocatalytic field. In order to design a Pe-bp system, photoactive solid nanoparticles have to be prepared to concurrently possess both emulsifying ability and photocatalytic activity. Nevertheless, almost all of these nanoparticles suffer from the drawback of a poor emulsifying ability although they have been demonstrated to have a high photocatalytic activity. This drawback greatly inhibits their wider practical application in Pe-bp systems. In order to overcome this problem, the introduction of another way for promoting photoactive solid nanoparticles with an amphiphilic property has to be proposed.

In fact, the emulsifying ability of photoactive nanoparticles is closely related to the surface wettability.^29,30 Therefore, water-in-oil (w/o) and oil-in-water (o/w) emulsions can be individually designed by logically combining hydrophobic/hydrophilic solid nanoparticles with hydrophilic/hydrophobic components.¹⁷ By using this proposed strategy, one may form photocatalyst-based nanoparticles with both emulsifying ability and photocatalytic activity. It is a feasible and effective combination strategy for making a stable Pickering emulsion.

Herein, a Pe-bp system is demonstrated with AB as a photocatalyst-based photoactive solid nanoparticle. Ag₃PO₄ is a p-type semiconductor; meanwhile, BiPO₄ is determined as an n-type one.^33–36 As an active visible light sensitive photocatalyst, Ag₃PO₄ displayed high performances for contaminant removal.³³ So, this yellow p-type semiconductor is a potential sensitizing agent to sensitize white BiPO₄ nano-cocoons. Thus, an efficient p–n junction between Ag₃PO₄ and BiPO₄ will be formed, which will decrease the recombination rate of charge carriers.⁹ Pure BiPO₄ only absorbs UV light while Ag₃PO₄ has good visible light absorption ability. Obviously, the AB nanocomposites demonstrate the combined absorption property of both Ag₃PO₄ and BiPO₄ semiconductors.^33–36

Unfortunately, the AB nanocomposite is intrinsically hydrophilic and thus unable to generate a Pickering emulsion. To circumvent this problem, we have to use hydrophobic components as a scaffold for the semiconductor nanoparticles, and more importantly, these hydrophobic components such as GR and MWCNTs supply an emulsifying ability. Moreover, these components have been demonstrated to possess a good ability to capture and shuttle electrons through the π–π networks. Consequently, the hybridization of AB nanocomposites with GR or MWCNTs would provide highly impressive approaches for the establishment of the Pe-bp system.

Degradation efficiency for dye removal was calculated from eqn (1):

where X is the degradation efficiency for dye removal, and C₀ and C are dye concentration initially and after degradation at time (t), respectively.¹⁰

As discussed, although stable emulsions were prepared by using high concentration of MWCNTs or GR, their light harvesting also needed to be considered. Hence, we studied the effect of the hydrophobic component content on the photocatalytic activity in the Pe-bp system. As typically shown in Fig. 9, the degradation efficiency of AB92 dye in the Pe-bp systems reaches a maximum value in the content of 3 wt% both GR and MWCNTs. As the concentration of hydrophobic component increases, the degradation efficiency decreases (red curve goes up). This result demonstrates that the light harvesting by the larger content of GR and MWCNTs causes a smaller number of photons absorbed by the photoactive AB nanocomposite and finally can prevent the light absorption by photoactive components. So, to develop a very efficient photocatalytic system, it remains very crucial to decrease the proportion of GR and MWCNTs in the as-prepared nanohybrid in the Pe-bp system. Nevertheless, it was found that this process has a restriction: the water droplets were no longer stabilized by the nanohybrid and the emulsions collapse. Thus, by simultaneously evaluating the light harvesting activity of the hydrophobic component and the stability of the Pickering emulsions, we adjusted the concentration of GR and MWCNTs in the as-prepared nanohybrid to be 3 wt% in our following experiments.

Fig. 9 Photodegradation of AB92 dye and the contact angle (θ) as a function of the contents of hydrophobic components in the prepared (a) AB/MWCNT and (b) AB/GR Pe-bp system under UV light irradiation.

The degradation efficiencies for both systems under UV and visible light irradiations are represented in Fig. 10a and b. To highlight the enhanced photocatalysis in the prepared Pe-bp system, we compared the dye degradation in this system with that of a traditional solution-dispersed system under the same conditions. As depicted in Fig. 10, decomposition of AB92 dye under both UV and visible light irradiations in the prepared Pe-bp system was much quicker than that in the solution-dispersed system. Strikingly, the results are essentially explained by the fact that O₂ is a nonpolar molecule and is thus much more soluble in isooctane phase than in water. At the w/o interface, H₂O reacts with the photogenerated holes of the AB composite, resulting in the formation of O₂ under light irradiation. Due to the higher solubility of O₂ molecules in the oil phase, O₂ molecules tend to migrate out of the water phase and dissolve in the isooctane phase. We infer from the above discussion that the removal of products from the reaction system can essentially accelerate the photocatalytic reactions. The observed significantly enhanced decolorization rate of AB92 in the Pe-bp system might be considered to result from the O₂ migration from the water to isooctane phase at the w/o interface. Thereby, O₂ migration makes the reaction equilibrium shift toward the side of products. Hence, the spatial separation of the reaction products from the reactants, which is a unique property of the Pe-bp system, could essentially accelerate the photocatalytic reactions for which the products show different solubility in the water phase and the isooctane phase.

Fig. 10 The degradation efficiency of AB92 dye in both Pe-bp and solution-dispersed systems under (a) UV and (b) visible light irradiation.

Moreover, the results indicated that the dye degradation in the AB/GR Pe-bp system is higher than in the AB/MWCNT system. Carbon nanotubes (CNTs) and GR, which are cylindrical and planar forms of sp²-hybridized carbon, respectively, have been demonstrated to enhance catalytic efficiency in the Pe-bp system due to their large specific surface areas, outstanding electronic mobility and molecular stability. Holistically, electron accepting and transporting properties of GR and MWCNTs in the Pe-bp system could contribute to the effective suppression of charge carrier recombination and thereby a higher photocatalysis rate would be obtained.^{16,31,37–42}

MWCNTs did not behave as well in the Pe-bp system as GR that has regular geometrical contours and a planar layered structure of sp²-hybridized carbon. GR, unlike CNTs, has edges that can interact well chemically. This may be because GR is composed of partially broken sp²-carbon networks and the substituent groups on the aromatic rings in AB92 dye can interact with the edges of the GR resulting in stronger interactions between GR and AB92 dye. On the other hand, degrading AB92 in the AB/GR Pe-bp system indicates that GR has stronger π–π interactions with AB92 dye and thus it enhances the photocatalytic efficiency in the Pe-bp system.

Additionally, since photocatalytic reactions are driven by the energetic photogenerated electrons, these reactions presumably stand to benefit most markedly from the enhanced electrical mobility of GR and MWCNTs. On the basis of extensive characterization of the electrical properties, as discussed in the EIS section, it is concluded that the improved electrical mobility of GR facilitates photocatalytic reactions in the Pe-bp system. With its unique electronic properties, GR is proposed to functionalize and tailor the Pe-bp system consisting of AB photoactive nanoparticles for improved reactivity. It appears that GR has superior electronic coupling to AB photoactive nanoparticles in the Pe-bp system. In particular, the greater electrical mobility of GR compared to MWCNTs implies a longer electronic mean free path. This enables energetic electrons to diffuse farther from the AB/GR interface in the Pe-bp system, thus both decreasing the likelihood of their recombination with holes on AB and enhancing the likelihood of interaction with adsorbed dye.⁴³ In this manner, the lifetime of holes on AB can be prolonged for the AB/GR nanohybrid, which is consistent with the observed enhancement in photodegradation of AB92 dye. As a whole, these results are in agreement with the concept that the photocatalytic degradation reaction is caused by the photogenerated e⁻/h⁺ pairs which are greatly separated.

Moreover, studies on the wettability of the AB/GR Pe-bp system have also shown that the water contact angle, in the range of 84–130°, is much closer to 90°, which demonstrates that GR is partially hydrophobic and tends to stabilize w/o emulsions because the particle surface resides more in water than oil. Additionally, the close proximity of the emulsion droplets did not cause them to coalesce, which shows that the presence of GR prominently hinders coalescence and phase separation and thus efficiently stabilizes the emulsion. The reason for this behaviour of GR sheets is that the basal planes of the carbon networks and AB component that are on those phases endow the Pe-bp with both hydrophilic and hydrophobic properties, which make them act like a functional surfactant. The photocatalytic experiments are in good agreement with both experimental impedance and contact angle data, which validate the proposed results of this work.

This further proves that both π–π interactions and superior electrical mobility have a significant influence on the Pe-bp system and explains why conductive components are beneficial for the formation of stable Pickering emulsions. Typical absorption spectra of the AB92 solution illuminated by UV light at different times in the AB/GR Pe-bp and solution-dispersed systems are shown in Fig. 11.

Fig. 11 Typical absorption spectra of AB92 solution illuminated by UV light at different times in the (a) AB/GR Pe-bp and (b) solution-dispersed systems (the initial concentration of AB92 dye and the volumes of water phase used in two systems were kept almost the same).

For reference, the evaluation of catalytic activity for the fabricated systems under both UV and visible light irradiation in the presence of P25 has been done as a control. The results reveal that AB composite exhibited markedly improved efficiency for AB92 photodegradation in comparison with P25 in Pe-bp and solution-dispersed systems under both visible and UV light illumination (see the ESI†).

3.3.2. Kinetic analysis. The photocatalytic degradation kinetics was studied. The linear simulation of AB92 photodegradation in both systems under UV and visible light irradiations can be realized with a pseudo-first-order model, called the Langmuir–Hinshelwood (L–H) kinetics model:¹⁰

In this equation r is degradation rate (mg L⁻¹ min⁻¹), C is dye concentration after various intervals of time (mg L⁻¹), t is irradiation time (min), k is reaction rate constant (min⁻¹) and K is the adsorption coefficient of dye (L mg⁻¹).

At low initial concentration of the AB92 molecules, the L–H kinetics model is well established for heterogeneous photocatalysis systems.²⁴ Hence, the above equation is changed to eqn (3). According to eqn (3), k_app was obtained for dye degradation from plotting the graph of ln(C₀/C_t) versus t (Fig. 12):

Fig. 12 Linear-log plot under (a) UV light irradiation, (b) visible light irradiation and (c) the degradation rate constants of AB92 photodegradation as a function of UV and visible light irradiation.

To make a more useful comparison, the pseudo-first-order rate law was applied to both systems for investigating the AB92 degradation kinetics. The obtained results are presented in Fig. 12 and they indicate a good correlation with pseudo-first-order reaction kinetics. As seen in Fig. 12, this comparison demonstrates that the degradation rate of AB92 dye under both UV and visible light irradiation in the Pe-bp system was significantly higher than of the traditional solution-dispersed system. The largest improvement in the rate of degradation was achieved by the AB/GR nanohybrid under UV light illumination. Under visible light irradiation, the AB/GR nanohybrid also displayed the fastest rate of AB92 dye degradation.

3.4. Discussion of the effects of reactive species on the photodegradation of AB92

In order to determine the influence of reactive species directly taking part in the AB92 photocatalytic process, quantities of different scavengers (fluoride ion, iodide ion, persulfate and t-BuOH) were applied to the AB92 degradation process using AB/GR nanohybrid in the Pe-bp system. The mechanism of photocatalysis has been reported earlier, summarized in eqn (4)–(12).⁴⁴ The deduction will be manifested by the addition of different scavengers, which are further discussed in this study.

Photocatalyst + hν → photocatalyst (e_CB⁻ + h_VB⁺) (4)

h_VB⁺ + OH_ads⁻ → ˙OH_ads (5)

h_VB⁺ + H₂O → ˙OH_Free + H⁺ (6)

e_CB⁻ + O₂ → O₂˙⁻ (7)

e_CB⁻ + O₂˙⁻ + 2H⁺ → H₂O₂ (8)

H₂O₂ + O₂˙⁻ → ˙OH_ads + O₂ + OH⁻ (9)

H₂O₂ + e_CB⁻ → ˙OH_ads + OH⁻ (10)

OH⁻ + h_VB⁺ → ˙OH_ads (11)

AB92 + ˙OH and/or O₂˙⁻ → degradation products (12)

AB92 + h_VB⁺ and/or e_CB⁻ → degradation products (13)

3.4.1. Effect of fluoride ion. As seen in Fig. 13, AB92 decolorization was strongly inhibited in the presence of fluoride ion as a diagnostic tool. Since the redox potential of the F˙/F⁻ pair is around 3.6 V, F⁻ can be effectively adsorbed on the catalyst surface at the w/o interface in the Pe-bp system. Moreover, the adsorbed fluoride ion is very stable against oxidation even by h_VB⁺.^44–46 So, the addition of excess F⁻ ion to the AB92 solution in the Pe-bp system could significantly prevent dye from being adsorbed onto the catalyst surface at the w/o interface. Therefore, AB92 degradation was significantly prevented by fluoride ion addition. Further convincing evidence to explain the trend described was obtained by a comparison of the blue and red curves (Fig. 13). Comparison between these curves indicates that AB92 decolorization mainly occurred by the surface charge process. It was deduced that AB92 is preferentially adsorbed onto the catalyst surface at the w/o interface before reacting with the active species. These results provide further support for the degradation mechanism of AB92 dye. It is concluded that the photocatalytic activity allows decomposition of organic species such as azo dyes which are adsorbed onto the catalyst surface at the w/o interface or dissolved in the surrounding liquid upon irradiation in the Pe-bp system. The photoactive particles located at the w/o interface photocatalytically decompose dyes dissolved in the water phase in the Pe-bp system.

Fig. 13 Effects of scavengers on the photodegradation of AB92 dye in the AB/GR Pe-bp system.

3.4.2. Effect of iodide ion. AB92 photodegradation was strongly inhibited by the addition of excess KI, as typically shown in Fig. 13. Iodide ion is a scavenger and reacts with both h_VB⁺ and ˙OH_ads at the w/o interface in the Pe-bp system, reducing the number of active species available for AB92 decolorization. The redox potential of I˙/I⁻ couple is 1.3 V, so the iodide ion can be an outstanding scavenger for capturing both h_VB⁺ and ˙OH_ads according to eqn (14)–(16) summarized as follows:^44,45

I⁻ ↔ I˙ + e⁻ (14)

I⁻ + I˙ → I₂˙⁻ (15)

I₂˙⁻ ↔ I₂ + e⁻ (16)

As seen in eqn (17)–(19), I⁻ reaction with ˙OH_ads is also possible because the rate constant of reaction with ˙OH_ads is 1.2 × 10¹⁰ M⁻¹ s⁻¹.⁴⁴

˙OH_ads + I⁻ → OH⁻ + I˙ (17)

˙OH_ads + I₂ → HOI + I˙ (18)

I⁻ + I˙ → I₂˙⁻ (19)

Conclusively, the amount of active species is reduced in the Pe-bp system by the addition of excess KI. When KI was used to capture both h_VB⁺ and ˙OH_ads, AB92 decolorization was significantly inhibited. Accordingly, if h_VB⁺ and/or ˙OH_ads are effective for AB92 decolorization, the decolorization rate should be mostly decreased by the addition of KI scavenger, which is consistent with the result shown in Fig. 13. It should be noticed that the determination wavelength for AB92 dye (572 nm) was not affected by the KI addition, because the absorption peak of KI is at a wavelength of 220 nm and the I₂ aqueous solution has two important absorption peaks at 287 and 353 nm.^44,45

3.4.3. Effect of tert-butyl alcohol (t-BuOH). To evaluate the effect of ˙OH radicals in the Pe-bp system, t-BuOH was added to the AB/GR system as a typical example. t-BuOH is a well-known ˙OH scavenger, which can react quickly with all the hydroxyl radicals produced in the AB/GR Pe-bp system with a rate constant of 6.0 × 10⁸ M⁻¹ s⁻¹ according to the following eqn (20):^44,47,48

t-BuOH + ˙OH → t-BuOH(–H) + H₂O (20)

The results show that adding t-BuOH as an ˙OH scavenger did not change the decolorization rate of AB92 textile dye: the results were much the same in the presence and absence of t-BuOH. This implies that the ˙OH radicals do not play a dominant role in AB92 degradation.

3.4.4. Effect of persulfate. The application of inorganic compounds as scavengers such as S₂O₈²⁻ has been carried out to capture the photogenerated electrons according to eqn (21) and (22). At the w/o interface, the sulfate radical anion (SO₄⁻˙) may react with H₂O producing ˙OH, as seen in eqn (23).⁴⁴ As displayed in Fig. 13, K₂S₂O₈ addition showed a negligible effect on AB92 decolorization, suggesting that O₂˙⁻ and ˙OH produced in the Pe-bp system (eqn (7) and (23)) are not the principal active species in the photodegradation of AB92 dye.

S₂O₈²⁻ + e_CB⁻ → SO₄⁻˙ + SO₄²⁻ (21)

SO₄⁻˙ + e_CB⁻ → SO₄²⁻ (22)

SO₄⁻˙ + H₂O → H⁺ + ˙OH + SO₄²⁻ (23)

It should be noted that the suppressive effects for AB92 degradation are conspicuously different for the various trapping agents, signifying O₂˙⁻ and ˙OH radicals and photogenerated holes play different roles in the degradation reaction; thereby, they result in different degradation approaches of AB92 dye. Taken together, combined with the results obtained by the additions of NaF, KI, persulfate and t-BuOH as scavengers, it can be concluded that h_VB⁺ is the predominant contributor to the AB92 degradation while the others display a weak effect.

3.5. Verification of the proposed mechanism

It is interesting to explore the plausible mechanism for the high photocatalytic performance of the nanohybrid in the prepared Pe-bp system. From the above discussion, reactive species are produced in the successive steps and further degraded AB92 azo dye. In addition, the other species would take part in the AB92 degradation process partially.

Combining the above effects, a possible mechanism for AB92 degradation by the AB heterojunction photocatalyst under visible light irradiation is proposed as illustrated in Fig. 14. As can be seen in Fig. 14a, the schematic diagram of the activity improvement of AB composite clearly demonstrates that the Ag₃PO₄ quantum dot sensitizer provides the electron–hole pairs under visible light irradiation. As displayed in Fig. 14b, some photogenerated electrons in the Ag₃PO₄ can be transferred directly to the BiPO₄ nano-cocoons or to C atoms on the GR sheets, after which reaction with AB92 dye to decolorize it is possible. Studies have shown that the CB electrons of Ag₃PO₄ can be injected into the GR sheets in the prepared Pe-bp system because of the high mobility of electrons on the GR sheets. The BiPO₄ in this system can also accept electrons and act as active sites for dye degradation. Additionally, the photogenerated electrons in the CB of Ag₃PO₄ can be transferred to BiPO₄ nano-cocoons through the GR sheets (which act as a conductive electron transport “highway”) and then react with the adsorbed AB92 dye in the w/o interface.

Fig. 14 Schematic illustration of (a) activity enhancement of AB composite and (b) a highly efficient Pe-bp system formed by self-assembling AB/GR nanohybrid at the w/o interface.

Due to the notable property of the Pe-bp system, the aforementioned ways in which photogenerated electrons are transferred (i) suppress the recombination of charge carriers, (ii) enlarge the reaction space and active adsorption sites, (iii) improve the interfacial charge transfer, and (iv) intensify the spatial separation of the reaction products from the reactants and consequently enhance the photocatalytic activity for AB92 photodegradation.

4. Conclusions

In conclusion, we have systematically demonstrated a complete utilization of the superior specific properties inherent in Pickering emulsions which would expedite generating a novel kind of photocatalytic system with a notably improved efficiency distinct from the traditional systems. The highly increased photocatalytic performance of the as-prepared Pe-bp systems has been clarified in terms of superior characteristics of the Pickering emulsions involving (i) enlarged reaction space, (ii) intensified continuous separation of the products from reactants at the w/o interface and (iii) effectively hindered charge recombination by the hydrophobic components. As a proof-of-concept, this research could offer not only new insights into novel photocatalytic systems with a notably high efficiency and practical applicability, but also greatly extend the potential applicability based on Pickering emulsion science and technology for environmental applications.

Acknowledgements

We gratefully thank the University Language Center, especially Mrs Bazrafshan, director of Language Center, for her cooperation in reviewing and editing this paper.

References

1. J. H. Huang, Y. J. Cui and X. C. Wang, Environ. Sci. Technol., 2010, 44, 3500–3504.
2. Y. Liu, G. Su, B. Zhang, G. Jiang and B. Yan, Analyst, 2011, 136, 872–877.
3. W. Zhou, H. Liu, J. Wang, D. Liu, G. Du and J. Cui, ACS Appl. Mater. Interfaces, 2010, 2, 2385–2392.
4. Y. F. Li and Z. P. Liu, Particle Size, J. Am. Chem. Soc., 2011, 133, 15743–15752.
5. A. Kubacka, M. Fernández-García and G. Colón, Chem. Rev., 2012, 112, 1555–1614.
6. Q. Zhu, W. Wang, L. Lin, G. Gao, H. Guo, H. Du and A. Xu, J. Phys. Chem. C, 2013, 117, 5894–5900.
7. X. Wang, S. Li, H. Yu, J. Yu and S. Liu, Chem.–Eur. J., 2011, 17, 7777–7780.
8. M. R. Elahifard, S. Rahimnejad, S. Haghighi and M. R. Gholami, J. Am. Chem. Soc., 2007, 129, 9552–9553.
9. H. Lin, H. Ye, B. Xu, J. Cao and S. Chen, Catal. Commun., 2013, 37, 55–59.
10. N. Mohaghegh, M. Tasviri, E. Rahimi and M. R. Gholami, Mater. Sci. Semicond. Process., 2014, 21, 167–179.
11. W. Y. Teoh, J. A. Scott and R. Amal, J. Phys. Chem. Lett., 2012, 3, 629–639.
12. C. McCullagh, N. Skillen, M. Adams and P. K. J. Robertson, J. Chem. Technol. Biotechnol., 2011, 86, 1002–1017.
13. W. Zhai, G. Li, P. Yu, L. Yang and L. Mao, J. Phys. Chem. C, 2013, 117, 15183–15191.
14. T. Nakato, H. Ueda, S. Hashimoto, R. Terao, M. Kameyama and E. Mouri, ACS Appl. Mater. Interfaces, 2012, 4, 4338–4347.
15. W. Wu, S. Gao, W. Tu, J. Chen and P. Zhang, Particuology, 2010, 8, 453–457.
16. X. Song, Y. Yang, J. Liu and H. Zhao, Langmuir, 2011, 27, 1186–1191.
17. R. Aveyard, B. P. Binks and J. H. Clint, Adv. Colloid Interface Sci., 2003, 100, 503–546.
18. A. Boker, J. He, T. Emrick and T. P. Russell, Soft Matter, 2007, 3, 1231–1248.
19. B. P. Binks and T. S. Horozov, Angew. Chem., Int. Ed., 2005, 44, 3722–3725.
20. Y. Lin, H. Skaff, T. Emrick, A. D. Dinsmore and T. P. Russell, Science, 2003, 299, 226–229.
21. S. Crossley, J. Faria, M. Shen and D. E. Resasco, Science, 2010, 327, 68–72.
22. S. Ghasemi, A. Esfandiar, S. Rahman Setayesh, A. Habibi-Yangjeh, A. Irajizad and M. R. Gholami, Appl. Catal., A, 2013, 462–463, 82–90.
23. Y. Zhou, Q. Bao, L. A. L. Tang, Y. Zhong and K. P. Loh, Chem. Mater., 2009, 21, 2950–2956.
24. F. Xue, H. B. Li, Y. C. Zhu, S. L. Xiong, X. W. Zhang, T. T. Wang, X. Liang and Y. T. Qian, J. Solid State Chem., 2009, 182, 1396–1400.
25. J. A. Dean, Lange's Handbook of Chemistry, R. R. Donnelley & Sons Company, New York, 1968.
26. D. Wang, D. Choi, J. Li, Z. Yang, Z. Nie, R. Kou, D. Hu, C. Wang, L. V. Saraf, J. Zhang, I. A. Aksay and J. Liu, ACS Nano, 2009, 3, 907–914.
27. H. L. Guo, X. F. Wang, Q. Y. Qian, F. B. Wang and X. H. Xia, ACS Nano, 2009, 3, 2653–2659.
28. X. Bai, R. Zong, C. Li, D. Liu, Y. Liu and Y. Zhu, Appl. Catal., B, 2014, 147, 82–91.
29. B. P. Binks and S. O. Lumsdon, Langmuir, 2000, 16, 8622–8631.
30. B. P. Binks, Curr. Opin. Colloid Interface Sci., 2002, 7, 21–41.
31. Y. He, F. Wu, X. Sun, R. Li, Y. Guo, C. Li, L. Zhang, F. Xing, W. Wang and J. Gao, ACS Appl. Mater. Interfaces, 2013, 5, 4843–4855.
32. M. Tang, X. Wang, F. Wu, Y. Liu, S. Zhang, X. Pang, X. Li and H. Qiu, Carbon, 2014, 71, 238–248.
33. Z. G. Yi, J. H. Ye, N. Kikugawa, T. Kako, S. X. Ouyang, H. Stuart-Williams, H. Yang, J. Y. Cao, W. J. Luo, Z. S. Li, Y. Liu and R. L. Withers, Nat. Mater., 2010, 9, 559–564.
34. H. Xu, Y. G. Xu, H. M. Li, J. X. Xia, J. Xiong, S. Yin, C. J. Huang and H. L. Wan, Dalton Trans., 2012, 41, 3387–3394.
35. C. Pan and Y. Zhu, Environ. Sci. Technol., 2010, 44, 5570–5574.
36. C. Pan and Y. Zhu, J. Mater. Chem., 2011, 21, 4235–4241.
37. H. Wang and E. K. Hobbie, Langmuir, 2003, 19, 3091–3093.
38. R. K. Wang, R. D. Reeves and K. J. Ziegler, J. Am. Chem. Soc., 2007, 129, 15124–15125.
39. M. Shen and D. E. Resasco, Langmuir, 2009, 25, 10843–10851.
40. J. Kim, L. J. Cote, F. Kim, W. Yuan, K. R. Shull and J. Huang, J. Am. Chem. Soc., 2010, 132, 8180–8186.
41. Q. Xiang, J. Yu and M. Jaroniec, Chem. Soc. Rev., 2012, 41, 782–796.
42. H. Zhang, X. Lv, Y. Li, Y. Wang and J. Li, ACS Nano, 2010, 4, 380–386.
43. Y. T. Liang, B. K. Vijayan, K. A. Gray and M. C. Hersam, Nano Lett., 2011, 11, 2865–2870.
44. S. Song, L. J. Xu, Z. Q. He, J. M. Chen, X. Z. Xiao and B. Yan, Environ. Sci. Technol., 2007, 41, 5846–5853.
45. Z. He, Y. Shi, C. Gao, L. Wen, J. Chen and S. Song, J. Phys. Chem. C, 2014, 118, 389–398.
46. S. Y. Yang, L. P. Lou, K. Wang and Y. X. Chen, Appl. Catal., A, 2006, 301, 152–157.
47. S. Song, F. Y. Hong, Z. Q. He, Q. L. Cai and J. M. Chen, J. Colloid Interface Sci., 2012, 378, 159–166.
48. L. H. Chia, X. M. Tang and L. K. Weavers, Environ. Sci. Technol., 2004, 38, 6875–6880.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra14294b

---