A novel ZnFe2O4/TiO2/flake graphite composite material was synthesized and used as particle electrodes in a photoelectrocatalytic (PEC) system to investigate the degradation of rhodamine B as a model dye pollutant in water. Results showed that a PEC process with the new composite evidently led to enhanced degradation of rhodamine B due to a synergistic effect of photocatalysis and electrocatalysis. Operating variables including electrolyte concentration, applied cell voltage, air flow, composite dosage, solution pH, and dye concentration were also found to play important roles in rhodamine B removal. A 99.0% removal efficiency was observed within 30 min of treatment under optimum conditions of 0.01 mol/L Na2SO4, applied cell voltage of 15 V, air flow of 20 mL/min, composite dosage of 10 g/L, solution pH of 2, and rhodamine B concentration of 20 mg/L, with a pseudo-first-order rate constant of 0.278 min−1. These findings could provide new insights into the development of efficient PEC technologies on degradation of residual dyes in water.
In recent years, dyes have been frequently used in textile industries and the untreated dyes are often discharged into natural streams, causing a growing environmental concern about contamination of drinking water (Mourão et al. 2010). Most dyes are potentially carcinogenic and resist biological methods to decompose (Liu et al. 2012) due to their large and complex molecular structures, posing a great challenge for color removal and complete mineralization. Advanced oxidation technologies (AOTs) that can generate highly powerful free radicals such as hydroxyl radicals (•OH) have been found useful in the degradation of most dyes. Among them, photocatalytic oxidation represents a promising AOT in the degradation of dyes due to its low cost, formation of less harmful secondary byproducts, operability under standard atmospheric condition, and utilization of renewable solar energy (Lee et al. 2013).
The AOT based on the heterogeneous semiconductor has reactions occurring at the surface of photocatalysts. However, the recombination of photoinduced electron-hole pairs at the surface is an inevitable issue. In addition, powdered photocatalysts are difficult to separate and recycle from the reaction system. Therefore, effective approaches for restraining electron-hole pair recombination and recycling photocatalysts are highly desirable. To solve the first limitation, a photoelectrocatalytic (PEC) process that uses an external electric circuit has been proposed by combining the photocatalytic and electrocatalytic technologies. A PEC reactor has an applied external anodic bias, which drives the photoinduced electrons and holes in different directions to effectively restrict the photoinduced electron-hole pair recombination (Zhai et al. 2013). The second limitation on separation and recovery of the suspended photocatalysts can be solved by immobilization of semiconductors onto the electrodes using a two-dimensional or a three-dimensional electrode system (Liu et al. 2012; Ayoubi-Feiz et al. 2014). An et al. (2002) investigated the PEC degradation of methylene blue with a three-dimensional electrode system and found 95% decolorization efficiency and 87% COD reduction were achievable. Since the titanium dioxide (TiO2) photocatalysts they used were in a powdered form, their further research (An et al. 2004) on degradation of quinoline used TiO2 aggregated sand as particle electrodes for easy separation and recovery. However, the packed bed sand in a compact reactor may affect the light transmittance, and the weight percentage of the semiconductor in the sand aggregate is low due to limited active sites in the sand structure, which potentially inhibit the PEC degradation efficiency. A better support medium for efficient PEC degradation of dyes is needed.
To date, little information has been available on using flake graphite as a support medium for heterogeneous semiconductor aggregation. As a carbonaceous material, flake graphite has more active sites for aggregation of semiconductors compared to the conventional sand media. The adsorption of dye pollutants to the surface also allows for easy oxidation by the produced •OH. In addition, the crystallinity feature of flake graphite leads to efficient photocatalytic activity (Pan et al. 2010), and its good conductivity contributes to easy polarization by external electric field for the formation of numerous charged microelectrodes. Therefore, it would be meaningful to investigate the potential of using flake graphite as particle electrodes in a three-dimensional PEC system. Specifically, it is speculated that the incorporation of zinc ferrite (ZnFe2O4) into the TiO2 structure might benefit the photocatalytic performance due to a relatively narrow band-gap (ca. 1.9 eV), because the band overlap could promote the excitation of the valence band electrons. Besides, the use of microelectrodes shortens the distance between the reactants and the electrodes, increases the surface area of the electrode and mass transfer, and thus might benefit the electrocatalytic performance. Thus, a synergistic effect of a visible-light-responsive photocatalyst, a new support medium, and a three-dimensional PEC system design is worth investigating.
In this work, we demonstrate a novel ZnFe2O4/TiO2/flake graphite composite material as particle electrodes in a three-dimensional electrode system for the PEC degradation of rhodamine B as a model dye pollutant. Rhodamine B, a widely used fluorescent dye, has been linked with potential toxicity, mutagenicity and carcinogenic effects, and is an important representative xanthene dye with good stability and high water solubility. The objectives of this research are two-fold: (1) to prepare a ZnFe2O4/TiO2/flake graphite composite material by immobilizing ZnFe2O4 and TiO2 on flake graphite using a sol-gel method; (2) to investigate the PEC performance with the composite for rhodamine B removal under various operating conditions. The research could provide new insights into the development of efficient PEC technologies on degradation of residual dyes in water.
MATERIALS AND METHODS
All reagents and chemicals used were of analytical grade without further purification. Tetrabutyl titanate (Tianjin Kemiou Chemical Reagent Co. Ltd, China), zinc acetate (Shanghai Shanpu Chemical Co. Ltd, China) and ferric nitrate (Xilong Chemical Co. Ltd, China) were used as the sources for Ti, Zn and Fe respectively. Rhodamine B (Tianjin Guangfu Fine Chemical Research Institute, China) was used as the model dye pollutant for the degradation study.
Flake graphite with a diameter range between 0.38 and 0.83 mm was pre-washed with distilled water to remove any impurities and dried at 110 °C. The composite was prepared by a sol-gel process as follows: Solution A was made by dissolving 10 mL tetrabutyl titanate in 20 mL anhydrous ethanol, followed by adding 6 mL glacial acetic acid and stirring for 1 h. Solution B was prepared by mixing zinc acetate and ferric nitrate based on a Zn:Fe molar ratio of 1:2 with 20 mL anhydrous ethanol, followed by adding 3 mL distilled water and stirring for 1 h. Solution B was added dropwise to solution A at 1 drop per second while stirring to form a yellow gel. Ten grams of flake graphite were added to the yellow gel, stirred for 1 h, and sealed at room temperature for 24 h. Filtration was then applied to remove extra gel that did not attach to the flake graphite, and the residue was dried at 110 °C and calcined for 2 h in a muffle furnace at 450 °C. The formed composite was stored in a desiccator after ambient air cooling.
The surface morphology and structure of the ZnFe2O4/TiO2/flake grahite composite was characterized by N2 physisorption using the Brunauer-Emmett-Teller (BET) isotherm, scanning electron microscopy (SEM, Hitachi Model S-4800, Hitachi High-Technologies Corporation, Japan), X-ray diffraction (XRD, M21X, MAC Science Ltd, Japan) using Cu Kα radiation, and X-ray photoelectron spectroscopy (XPS, K-Alpha 1063, Thermo Fisher Scientific Inc., UK).
The PEC system (Figure 1) was composed of a quartz reactor with a capacity of 250 mL, a magnetic stirrer, a potentiostat, diffused air, a light source at the side of the reactor, two Pt plates (10 cm × 4.5 cm, effective area 4.0 cm × 4.5 cm) as anode and cathode, and the prepared ZnFe2O4/TiO2/flake graphite composite as particle electrodes. The reactor contained 100 mL rhodamine B solution as target dye pollutant and Na2SO4 as electrolyte. The distance between the light source and the reactor was 25 cm. The two Pt plates were divided by securing to the opposing side walls of the reactor.
Degradation studies and analysis
RESULTS AND DISCUSSION
Characterization of the composite
The specific surface area of the ZnFe2O4/TiO2/flake graphite composite was determined by a N2 physisorption isotherm method. The composite has a BET specific surface area of 2.7470 m2/g, which is higher than that of bare flake graphite, 1.7605 m2/g. The surface morphology of the composite and flake graphite measured by SEM is shown in Figure 2. The bare flake graphite is illustrated in Figure 2(a). Prior to any modification, the surface appeared smooth, without any obvious impurities. After aggregation with metal oxides and calcinations, the surface became rough and dispersed with particles (Figure 2(b)). This was in close agreement with the observation of increased specific surface area. The particles on the surface were produced by the aggregation of ZnFe2O4 and TiO2, and neither possess any definite shape. The increased specific surface area could benefit the heterogeneous photocatalysis by facilitating surface adsorption of reactants and thus promoting the interfacial charge transfer (Lee et al. 2013).
In order to examine the crystal phases of the prepared composite, qualitative analysis of the XRD pattern was performed. Figure 2(c) compares the XRD patterns for the composite and bare flake graphite. The peak of anatase TiO2 phase was observed at 2θ = 25.281° while the peak of ZnFe2O4 was not found in the XRD pattern of the composite. The non-existence of the aggregate was attributed to low content or high dispersion of the species within the composite (Xu et al. 2011). Meanwhile, the intensity of the XRD peak is also dictated by the degree of crystallization, which is influenced by the calcination temperature, and Lee et al. (2013) found that increasing the calcination temperature from 450 to 750 °C would lead to a higher peak formation. It is noted that the XRD pattern of the composite sample with optimum PEC performance was presented. Increasing the weight percentages of ZnFe2O4 visualized the peak, but was found to have lower PEC performance.
The chemical state of each element in the ZnFe2O4/TiO2/flake graphite composite was studied by XPS analysis. The XPS survey spectrum of the composite is illustrated in Figure 3(a) and the core level spectra of characteristic elements are presented in Figure 3(b)–3(e). The survey spectrum confirms that besides the C element in flake graphite, Fe, Ti, Zn, and O were also present in the composite material. The results were in accordance with the explanation of XRD results on the disappearance of ZnFe2O4 peak due to low content and high dispersion of the aggregate. High resolution XPS spectrum of Ti 2p was observed in Figure 3(b). The peaks in the Ti 2p spectra located at the binding energy of 464.4 eV and 458.6 eV were indicative of Ti 2p1/2 and Ti 2p3/2 of Ti4+, respectively (Joung et al. 2006). As shown in Figure 3(c), the binding energy of 710.6 eV and 724.4 eV were ascribed to the Fe 2p1/2 and Fe 2p3/2 of the oxidation state Fe3+, respectively. The peaks in the Zn 2p spectra, as shown in Figure 4(d), were indicative of Zn 2p1/2 and Zn 2p3/2 of Zn2+ (NuLi et al. 2004), located at the binding energy of 1,045.02 eV and 1,021.79 eV, respectively. The peak at 529.93 eV (Figure 3(e)) was due to O 1 s bonded to Ti and Fe, whereas the peak at higher binding energy of 531.82 eV was ascribed to the presence of the OH group (Palanisamy et al. 2013), indicating surface contamination by hydroxides from the atmosphere.
On the basis of the SEM images, XRD patterns, and XPS spectra, it is reasonable to suggest that the heterojunctions of the composite were present in the as-synthesized sample.
The influence of different processes including direct photocatalysis, electrocatalysis, and PEC degradation of rhodamine B were studied. Compared to the operating conditions of the PEC process, direct photocatalysis and electrocatalysis did not include cell voltage and light irradiation, respectively. The reactions for all three processes were carried out for 90 min and the progress of each reaction was monitored by UV absorbance measurement at a time interval of 10 min to examine the degradation performance.
The results indicate that rhodamine B could be removed from the solution by all three processes, and the profiles are presented in Figure 4. However, only 22.1% of rhodamine B was removed by direct photocatalysis within 90 min, compared to 63.2% removal by the electrocatalysis when a bias potential was applied. These two removal efficiencies were both apparently lower than that of the PEC process (96.2%). The increased removal efficiency for the PEC process implied an obvious synergistic effect between the electrocatalysis and photocatalysis, suggesting that the applied bias potential inhibited the recombination of photoinduced electron-hole pairs effectively and prolonged the lift of the photoinduced carriers (Hou et al. 2010). The external electric field could also result in direct (direct electron transfer on the anode) or indirect (electro-generated oxidizing species, i.e. •OH, from water electrolysis) electrochemical oxidation of organics. Besides, the carbonaceous material in flake graphite might serve as a catalyst to decompose H2O2 to •OH due to the presence of polyaromatic moieties and functional group (Can et al. 2014), leading to higher concentration of •OH. Moreover, the use of the composite as particle electrodes resulted in expanded specific surface areas contributing to a greater collecting capacity of photoinduced electrons than those with holes (An et al. 2004), and could form numerous charged microelectrodes shortening the distance between the reactants and the electrodes (Kong et al. 2006). Thus, the surface area of the electrodes and the mass transfer rates were increased, resulting in higher degradation efficiency. The above discussion explains the phenomenon of the enhanced rhodamine B removal by the PEC process. The life of the composite was also investigated by repeated uses of the composite, and the removal efficiency of PEC degradation decreased by 10% after five cycles of operation.
Adding electrolyte to a reaction solution can enhance the conductivity and increase the electron mass transfer. Na2SO4 is used as a supporting electrolyte, because it can offer excess levels of sulfate anions, which can enhance the charge transfer. It may also be oxidized to peroxydisulfate anions or an even stronger oxidant sulfate radical anions (Zhang et al. 2012). The influence of Na2SO4 on the PEC degradation of rhodamine B with the novel composite has been investigated in 0.005, 0.010, 0.020, and 0.030 mol/L concentrations. As shown in Figure 5(a), the rhodamine B removal efficiency increased as the Na2SO4 concentration increased from 0.005 to 0.010 mol/L. This was because the electricity current increased in the system and more excited electrons under light irradiation were transferred from the anode to the cathode, which resulted in lower recombination of the electron-hole pairs. There was no significant change in the removal efficiency when the Na2SO4 concentration increased further. This was attributed to the more parasitic reactions, i.e. the deposition of H2O2, as a result of higher current level. Thus, it would be more economical to use 0.010 mol/L Na2SO4 on an industrial scale.
Applied cell voltage
Applying a potential bias to electrodes using a potentiostat can minimize the recombination of photoinduced electron-hole pairs (Wang et al. 2012; Zhai et al. 2013). The cell voltage is also the driving force in the electrocatalytic reaction and plays an important factor in wastewater treatment involving the electrochemical method. The influence of an applied cell voltage on the PEC degradation of rhodamine B with the novel composite is shown in Figure 5(b). The removal efficiency increased with an increasing potential bias, up to 15 V. This phenomenon could be attributed to an increase in the electrostatic interaction and affinity between the surface of the electrode and ions due to increased adsorption at high potentials (Ayoubi-Feiz et al. 2014). The capability for direct or indirect electrochemical oxidation of the dye pollutant was enhanced. Furthermore, by increasing the electric field, the separation of photoinduced electron-hole pairs inside the semiconductors was enhanced, which might result in the enhancement of •OH generation (Jiang et al. 2004).
It should also be noted that when the applied cell voltage increased higher than 15 V, the removal efficiency did not further increase. This was because higher potential not only led to more parasitic reactions such as the decomposition of H2O2 and significant simultaneous H2 generation, but also caused higher energy consumption (Liu et al. 2012). The corrosion of electrodes and higher solution temperature might also lead to the leaching of adsorbed pollutants.
The PEC degradation of rhodamine B with the novel composite was compared under no aeration and aeration with air flow of 10, 20 and 30 ml/min. The observed rhodamine B removal percentages are shown in Figure 5(c). With aeration, the removal efficiency of rhodamine B increased. This was because aeration agitated the solution and offered better gas-liquid mass transfer efficiency. On the other hand, the molecular oxygen played an important role in the photocatalytic reaction by supplying essential oxygen to capture photoinduced electrons, reducing the recombination of photoinduced electron-hole pairs (An et al. 2004).
It is noted that the improvement in removal efficiency was not observed when the air flow further increased from 20 to 30 mL/min. This implied that necessary aeration was beneficial but a large air flow was not favorable. The phenomenon could be attributed to the saturation of dissolved oxygen and the reduced adsorption of the pollutant onto the composite surface caused by a large air flow.
The amount of catalyst in the PEC process is significant as it may strongly influence the degradation effect. The PEC experiments were carried out by varying the composite dosage from 6 to 12 g/L. As shown in Figure 5(d), the removal efficiency of rhodamine B increased when the amount of composite increased from 6 to 10 g/L, attributable to a pronounced increase in the active sites available on the composite surface. An optimum composite dosage also offered a reasonable light absorption to produce sufficient amounts of active •OH on the surface of the composite. A further increase in the dosage beyond 10 g/L showed a decrease in degradation performance. This was ascribed to the decrease of total active surface area by flake graphite overlapping, deactivation of activated molecules by collision with ground state molecules (Zhou et al. 2005), and agglomeration (Wu et al. 2009).
To study the effect of pH on the PEC degradation of rhodamine B with the novel composite, experiments were carried out in a pH range of 2–12. As shown in Figure 5(e), removal efficiency for rhodamine B increased as pH value decreased. It is known that the pH level may affect discolorization and mineralization of dyes differently (Zhang et al. 2012). In this study, the optimum pH value for the PEC degradation of rhodamine B was found to be 2. Since the point of zero charge (pHpzc) for TiO2 is around 6 (Selcuk & Bekbolet 2008), the catalyst surface is negatively charged when the pH is higher than the pHpzc or positively charged when the pH is lower. It is speculated that the solution pH affects the sorption of dyes on the catalyst surface, which depends on the dye functional group. Rhodamine B has a pKa of 3.0 and was negatively charged with only one ionized carboxylic group (Kasnavia et al. 1999) while the catalyst was positively charged at pH 2, the adsorption of rhodamine B onto the catalyst surface became much easier. Beltrán et al. (1996) also reported a disadvantageous effect existed with an increased solution pH due to a stronger scavenging effect until the pH value was higher than the pKa of hydrogen peroxide (11.8). This explained why a low pH was preferred for rhodamine B removal with the system.
To explore the maximum PEC degradation capacity with the novel composite, the effect of dye concentration on degradation performance was investigated at rhodamine B concentrations of 10, 20, 30, and 40 mg/L as a function of irradiation time at optimum conditions (Na2SO4 = 0.01 mole/L, applied cell voltage = 15 V, air flow = 20 mL/min, composite dosage = 10 g/L, and solution pH = 2). As shown in Figure 5(f), higher removal efficiencies were observed by starting with higher rhodamine B concentrations up to 20 mg/L. Increasing concentrations higher than 20 mg/L led to lower removal efficiencies. This was because the number of active sites on the composite adsorbent was not enough for all of the dye molecules in a high dye concentration (Wang et al. 2013). Furthermore, high concentration dyes absorbed part of the light, or the low transparency reduced the light intensity reaching the composite surface, leading to a reduction of efficiency due to decline in the amount of •OH generation (Cardoso et al. 2010).
It was found that a 99.0% removal efficiency was achievable within 30 min of PEC treatment under optimum conditions of 0.01 mol/L Na2SO4, applied cell voltage of 15 V, air flow of 20 mL/min, composite dosage of 10 g/L, solution pH of 2, and rhodamine B concentration of 20 mg/L. Table 1 presents the pseudo-first-order rate constants and regression coefficients for PEC degradation with the novel composite at rhodamine B concentrations of 10, 20, 30, and 40 mg/L while all other conditions were optimum. The results indicated that the degradation followed pseudo-first-order kinetics with high regression coefficients. The rate constant increased as the dye concentration increased from 10 to 20 mg/L and then decreased. The degradation rate constant at 10 mg/L was relatively low, because the process was dictated by the absorption rate of dye to the active sites of the composite, and a lower concentration led to a less favorable concentration gradient for diffusion of the dye onto the composite. On the other hand, the number of •OH formed as a result of the PEC process under certain light and voltage scenario was fixed or even decreased as the dye concentration increased further, and the reaction between •OH and dye became rate-limiting. At the optimum conditions, the pseudo-first-order rate constant was 0.278 min−1.
|Rhodamine B concentration||10 mg/L||20 mg/L||30 mg/L||40 mg/L|
|Pseudo-first-order rate constant k (min−1)||0.139||0.278||0.126||0.113|
|Regression coefficient R2||0.960||0.942||0.979||0.988|
|Rhodamine B concentration||10 mg/L||20 mg/L||30 mg/L||40 mg/L|
|Pseudo-first-order rate constant k (min−1)||0.139||0.278||0.126||0.113|
|Regression coefficient R2||0.960||0.942||0.979||0.988|
Rhodamine B – one of the most commonly used dyes in the textile and food industry, is highly carcinogenic, a neurotoxin and chronic as a pollutant in drinking water. In this study, dye concentrations up to 40 mg/L (0.084 mM) were explored to simulate wastewater contaminated with a high level of Rhodamine B. Thus, it must be treated carefully before being discharged into water streams. Although many researchers have confirmed that two-dimensional systems are able to get the treatment job done, with a three-dimensional electrode system, the electrochemical process shows a better process as a result of the formation of numerous micro-electrodes in the electrical field. Moreover, the heterogeneous photocatalysis appears to be a promising route combined with the three-dimensional system for the treatment of the dye as demonstrated by the high degradation efficiency in this study, and it overcomes previous major drawbacks such as inefficient light harvesting, design of photo reactor, and the recovery and reuse of titanium dioxide. The study presents a promising result which should provide motivation to build a larger scale PEC system and eventually to make commercial PEC systems a reality.
The ZnFe2O4/TiO2/flake graphite composite has been synthesized and characterized by SEM, XRD and XPS techniques. The results revealed that the composite is a good catalyst as particle electrodes for the PEC degradation of rhodamine B due to a synergistic effect of photocatalysis and electrocatalysis. The PEC degradation performance was dependent on electrolyte concentration, applied cell voltage, air flow, composite dosage, solution pH, and dye concentration. A 99.0% rhodamine B removal efficiency was observed within 30 min of PEC treatment under optimum conditions of 0.01 mol/L Na2SO4, applied cell voltage of 15 V, air flow of 20 mL/min, composite dosage of 10 g/L, solution pH of 2, and rhodamine B concentration of 20 mg/L. The ZnFe2O4/TiO2/flake graphite composite as particle electrodes is expected to have promising applications for the degradation of residual dyes in water.
This work was supported by National Science Foundation of China (Project No. 51178173 and 51202065) and the Start-up fund provided by College of Natural Sciences and Mathematics at Indiana University of Pennsylvania. The authors acknowledge Drs Shenglian Luo and Chengbin Liu for coordination of core facilities for this study.