In this work, titanium dioxide (TiO2) was modified with reduced graphene oxide (rGO), and then coated on filter paper to prepare the rGT/FP photoelectrode for the photoelectrocatalytic (PEC) decolorization of methylene blue (MB). The physicochemical properties of the rGT/FP photoelectrode were characterized using scanning electron microscopy (SEM), X-ray diffraction (XRD) analysis and UV-Vis diffuse reflectance spectroscopy (DRS). The decolorization results demonstrated that the photocatalytic (PC) and electrocatalytic (EC) efficiency of the photoelectrode could be significantly improved by the modification of rGO. The improvement of PC and EC efficiency might attribute to the existence of rGO, which could extend the light-harvesting efficiency, promote the photocurrent response value and suppress the charge recombination. Furthermore, the PEC decolorization of MB using the rGT/FP photoelectrode presented higher efficiency than the sum of PC and EC decolorization, indicating the synergistic effect between the photo and electrical energy.
The photocatalytic (PC) decolorization of organic pollutants using semiconductors has attracted numerous attentions in recent years, because of their use of synchronous light and activated molecules driven in chemical reactions (Hoffmann et al. 1995; Chen et al. 2010; Wang et al. 2018; Takle et al. 2019). Among these semiconductors, TiO2 has been recognized as the most potential PC material, due to its nontoxicity, good biocompatibility, low-cost and excellent PC degradation capability (Chen et al. 2018; Sánchez-Rodríguez et al. 2018; Sieland et al. 2018). Lots of reports had extensively investigated the PC performance of TiO2 with various nanostructures for decolorizing organic dye effluent (Rajkumar et al. 2015; Ren et al. 2017; Sathishkumar et al. 2017; Subramonian et al. 2017). However, the high recombination rate of photogenerated charge carriers of TiO2 hinders its practical implementation. There are two main strategies to improve the catalytic efficiency of TiO2, including modifying or doping TiO2 with other sensitive materials, and using the method of photoelectrocatalysis, which is the combination of photocatalysis and electrocatalysis.
Numerous efforts have been made to vary the chemical composition and structure by doping TiO2 with ions (Koketsu et al. 2017; Smirnova et al. 2017; Yan et al. 2017), coupling with noble metals (Ghasemi et al. 2013; Zhang et al. 2013) and forming hybrids with metal sulfides (Mao et al. 2013; Lindblad et al. 2014; Litke et al. 2016; Zhang et al. 2018), etc. Graphene, a typical two-dimensional honeycomb carbon material made from the compact accumulation of single layer carbon atoms, has become a new research hotspot since its discovery in 2004, due to its excellent electrical, thermal, mechanical and optical properties (Bolotin et al. 2008; Du et al. 2008; Liang et al. 2010; Zhang et al. 2010). Graphene has been widely used in the applications of photocatalysis (Williams et al. 2008; Xiang et al. 2011), electronic devices (Li et al. 2010; Wei et al. 2013), micro-nano sensors (Tao et al. 2016) and other fields. Recently, many reports have demonstrated the superior performance of nano-semiconductors coupled with reduced graphene oxide (rGO) for enhancing the PC degradation efficiency. For example, Garrafa Galvez et al. reported that the rGO/TiO2 compound possessed the enhanced PC degradation efficiency by a maximum of seven times than that of pristine TiO2 under UV irradiation (Garrafa-Galvez et al. 2019). Kim et al. (2017) found that doped TiO2 nanotube arrays with rGO showed much higher PC degradation efficiency than that of TiO2 nanotube arrays.
There are many studies that have investigated the PC degradation performance of rGO/TiO2, with the compound in the various structures of nanoparticle, nanorod or nanotube. Among the different structures, nanoparticles have obtained the most attention, because of the high adsorption capacity and their large specific surface area. However, there are two obvious obstacles that still restrict its application, including difficult recovery from dyed or other polluted waste solutions and confining the form of photocatalysis (Kim & Anderson 1994; Liao & Carter 2013; Zhai et al. 2013).
Photoelectrocatalysis has been proven as an efficient technique for pollutant degradation by applying an external bias to the photoelectrode. The photoexcited electrons could be accelerated and separated with holes by the external bias, leading both the electrons and holes to react with reactants (Dai et al. 2011; Qin et al. 2012).
In this work, the TiO2 based photoelectrode was prepared and used for photoelectrocatalytic (PEC) decolorization of methylene blue (MB). The reduced graphene oxide was used to modify the TiO2, to improve the degradation efficiency. Moreover, commercially available filter paper was used as the substrate material for the photoelectrode preparation.
TiO2 (P25, 80% anatase and 20% rutile) was purchased from Evonik Degussa Co., Ltd. Quantitative filter paper (GB/T1914-2007, Sinopharm Chemical Reagent Co., Ltd) were cut into pieces of 2 cm × 2 cm to obtain the electrode substrate. Ethyl cellulose ([C6H7O2(OC2H5)3]n), and methylene blue (MB) were purchased from Aladdin. Graphite, sulfuric acid, hydrochloric acid (HCl), sodium nitrate (NaNO3), potassium permanganate (KMnO4), and hydrogen peroxide (H2O2) were purchased from Sinopharm Chemical Reagent Co., Ltd. All reagents were of analytical grade and used without further purification.
Preparation of rGT/FP photoelectrocatalyst
The composite of rGO and TiO2 nanoparticles was denoted as rGT and prepared using the hydrothermal method (Guo et al. 2019). Details of preparing rGT are provided in the Supplementary Material. The weight ratio of rGO to TiO2 in the rGT composite was controlled as 5%.
The photoelectrode was obtained by coating rGT onto the filter paper. In detail, 3.0 g ethylcellulose and 0.8 g absolute ethyl alcohol were mixed and stirred to form a clear and viscous fluid; 0.1 g rGT was then added into the fluid to make the rGT mixture. Thereafter, filter paper was coated with the rGT mixture using a doctor-blade method, followed by heat treatment at 60 °C for 10 min to obtain the photoelectrode of rGT/FP. The photoelectrode of TiO2/filter paper (TiO2/FP) was also prepared with the same method.
The morphology of samples were observed by scanning electron microscope (SEM, SU1510, HITACHT Ltd, Japan). The X-ray diffraction (XRD) patterns of the samples were recorded via an X-ray diffractometer (D/Max 2200PC, Rigaku Corporation) in the range of 10°–80°, using a nickel filtrated CuKα radiation operated at 28 kV and 20 mA. The element distribution of rGT/FP photoelectrode was analyzed through an energy-dispersive x-ray spectroscopy (EDS, Hitachi S4800). The UV-Vis diffuse reflectance spectroscopy (DRS) were obtained from dry-pressed disk samples using the UV-Vis spectrophotometer (U-4100, Hitachi Ltd, Japan).
Photoelectrochemical measurements were performed in a conventional three-electrode, single-compartment quartz cell on an Ivium Electrochemical Workstation. A 250 W high-pressure mercury lamp was used to provide the light irradiation. A saturated calomel electrode (SCE) was used as the reference electrode and a Pt plate was used as the counter electrode. 0.1 M Na2SO4 was used as the electrolyte. The TiO2/FP and rGT/FP photoelectrodes were used as the working electrodes, respectively.
In the PEC experiments, the rGT/FP was used as the photoelectrode with the dimension of 2.3 cm × 2 cm. The clamping length of the electrode holder was 0.3 cm, and the effective area of the photoelectrode immersed in the electrolyte was 2 cm × 2 cm. A same-sized platinum foil acted as the cathode. The distance between the two electrodes was 5.0 cm. The light irradiation was supplied by a 250 W high-pressure mercury lamp. The irradiation distance between the lamp and the electrode was set as 10 cm. The electrolyte of 0.05 mol/L KCl was added to increase the conductivity of the MB solution. Prior to irradiation, the two electrodes were vertically aligned in the reactor. Then, 60 mL of the MB solution was added to the reactor and stirred in the dark for 30 min to establish an adsorption-desorption equilibrium. The applied potential in the PEC experiments was controlled by a DC power supply. The decolorization of the MB solution was determined at a wavelength of 664 nm by a UV-Vis spectrophotometer (Shanghai Meipuda Instrument Co., Ltd).
RESULTS AND DISCUSSION
The morphology of pristine TiO2, rGT, TiO2/FP and rGT/FP was observed by SEM. As shown in Figure 1(a), pristine TiO2 has a rough surface and loose structure, while a relatively smooth surface and compact structure are presented in rGT (Figure 1(b)), due to the distribution of powdered TiO2 on the flake rGO. Figure 1(c) and 1(d) show the surface morphology of TiO2/FP and rGT/FP, respectively. It could be observed that TiO2/FP possesses a smooth surface with few convex particles, while the rGT/FP exhibits an uneven surface. EDS mapping was used to investigate the distribution of O, C, and Ti elements on rGT/FP, and shown in Figure 2. It can be found that O and C elements are uniformly distributed on the filter paper. The results of EDS demonstrated that the TiO2 nanoparticles have been homogenously coated onto the filter paper by the doctor blade method.
XRD and UV-Vis analysis
Figure 3(a) shows the XRD patterns that correspond to the samples of pristine TiO2, rGT, TiO2/FP and rGT/FP. The diffraction peaks at 2θ = 25.4°, 37.8°, 48.1°, 54.0°, 55.1°, 62.7°, 68.8°, 70.3° and 75.1° are indexed to the crystalline planes of anatase TiO2 (101), (004), (200), (105), (211), (204), (116), (220) and (215), respectively, which correspond to the standard pattern of the anatase crystalline phase of TiO2 (JPCDS, No. 21-1272) (Almeida & Zanoni 2014). Furthermore, the diffraction peaks at 2θ values of 27.4°, 36.1°, and 41.3° correspond to the rutile TiO2 (110), (101) and (111) planes (JPCDS, No. 21-1276), respectively. The XRD pattern of rGT sample is quite similar to that of pure TiO2, while there is not any typical diffraction peak corresponding to rGO, due to the low content of carbon species in the composites (Liu et al. 2012; Zhang et al. 2009). Therefore, the characteristic peaks of rGO could not be detected on the XRD pattern. The diffraction peaks at 2θ values of 15.13°, 16.90°, and 22.92°, corresponding to the cellulose (JPCDS, No. 50-2241), are strong enough to cover the diffraction peaks of rGT and TiO2.
Optical characterizations of various samples were then investigated using the UV-Vis diffuse reflectance measurements. As shown in Figure 3(b), pristine TiO2 samples exhibit very low absorption in the visible region, because of the wide band gap of TiO2. On the other hand, rGT and rGT/FP have obvious enhanced absorption in the visible light region. Furthermore, a red shift could be detected on the spectra of rGT and rGT/FP. These results indicated that the modification with rGO could enhance the PC degradation activity of TiO2 (Min et al. 2017).
Figure 4(a) shows the transient photocurrent response of TiO2/FP and rGT/FP photoelectrodes, under UV-light irradiation without applied bias. The rGT/FP owns a higher photocurrent than that of TiO2/FP. The photocurrent response of rGT/FP photocatalyst electrodes is greatly enhanced by 3.5 fold, compared to the pristine TiO2. This result indicates the improved charge separation and transportation efficiency of the rGT/FP photoelectrode, because the rGO enhanced the light absorption and accelerated the transfer of photogenerated electrons. Figure 4(b) shows the photocurrent of TiO2/FP and rGT/FP in 0.1 M Na2SO4 electrolyte under a bias from −1.0 V to 1.0 V (vs Ag/AgCl), using the UV light as the illuminant. The saturated photocurrent of rGT/FP is higher than that of pure TiO2/FP, indicating that there are more photoexcited electrons on the rGT/FP electrode's flow to the cathode (Wang et al. 2012; Zhang et al. 2012; Zhou et al. 2014).
In order to investigate the catalytic performances of the TiO2/FP and rGT/FP photoelectrodes, the decolorization of MB in the supporting electrolyte was conducted in terms of the PC, electrocatalytic (EC), and PEC treatments.
In this work, two basic conditions of applied voltages and initial dye concentrations were investigated to understand their effects on the decolorization efficiency of the EC process.
The EC decolorization performances using the photoelectrodes of TiO2/FP and rGT/FP are firstly assessed at various external biases in darkness. It is of note that the EC decolorization rates using rGT/FP are much higher than that using TiO2/FP. Furthermore, as shown in Figure 5(a), the EC decolorization rates on the rGT/FP photoelectrode, after 60 min treatment, are 15.36%, 43.88%, 74.23% and 82.88% at applied voltages of 4, 8, 12 and 16 V, respectively. This result indicated that the EC decolorization efficiency can be significantly impacted by the applied voltage, because the applied voltage can reduce the recombination rate of e− and h+ pairs. However, the increment declined sharply as the applied voltage exceeded 12 V. This phenomenon is caused by the recombination reaction between the e− derived from external voltage and h+ formed on the TiO2, which resulted in the consumption of active species and the decrease of the MB decolorization rate (Kim et al. 2017). Therefore, the applied voltage of 12 V was chosen for further experiments.
MB solution with concentrations of 5, 10, 15 and 20 mg/L was used to investigate the influence of dye concentration on decolorization efficiency. As shown in Figure 5(b), the decolorization rate decreases with increasing dye concentration. The decolorization processes can be divided into the two main steps, including dye diffusion from the solution to the surface of the electrode, and dye degradation by the electrochemical reaction. At low dye concentrations, the decolorization rate was mainly controlled by the diffusion process, because the electrochemical reaction was faster than the diffusion rate and could rapidly degrade the dye molecules. As the concentration increased, more dye molecules could diffuse to the surface of the electrode and could not be degraded in time, resulting in a lower degradation rate (Rajkumar et al. 2007).
The decolorization capabilities of the TiO2/FP and rGT/FP were also evaluated with the PC decolorization process. As shown in Figure 6(a), the TiO2/FP achieved PC decolorization of 15.34% within 60 min under UV irradiation, while rGT/FP presented PC decolorization of 29.26%. This result indicated that the incorporated rGO increases absorption of visible light, and also acts like a charge separator to reduce the recombination rate of ‘electron-hole’ pairs (Garrafa-Galvez et al. 2019), which is consistent with the UV-Vis diffuse reflectance spectra in Figure 3(b). The enhancement of the decolorization rate can also be observed in the EC and PEC decolorization processes, using the TiO2/FP and rGT/FP photoelectrodes, respectively, as shown in Figure 6(b) and 6(c). The decolorization efficiency using the rGT/FP photoelectrode is much higher than that of the TiO2/FP photoelectrode. There are two main reasons for the enhancement of the PEC decolorization rate: (1) the electrical conductivity of rGO can accelerate the overall electron speed of the photoelectrode; and (2) the narrow band gap of rGO benefits the transfer of photogenerated electrons from the valence band to the conduction band (Zhai et al. 2013). Therefore, the recombination rate of ‘electron-hole’ pairs is decreased, resulted in the improvement of the decolorization efficiency (Wang et al. 2012; Min et al. 2017).
Most importantly, PEC treatment using the rGT/FP photoelectrode achieved the highest decolorization rate (almost 100% decolorization of MB within 50 min). Furthermore, as shown in Figure 6(d) and 6(e), the summary decolorization efficiency of MB by EC + PC treatment is lower than that using PEC treatment, which indicated the synergetic effect in the PEC process. The corresponding parameters, k and R2 (regression coefficients) of MB decolorization by the rGT/FP photoelectrode under the PC, EC and PEC processes are listed in Table S1 (Supplementary Material). The rate constants of the decolorization of MB over the PC, EC and the PEC activity are determined to be 0.0051, 0.0215 and 0.1152 min−1, respectively. Apparently, the kinetic constant of the PEC process is about 22-fold and 5-fold that of the PC process and EC process, respectively, which further demonstrates the enhanced decolorization efficiency from using PEC process.
The UV-Vis absorbance spectra of MB solution during PEC decolorization were measured and are shown in Figure S1 (Supplementary Material). The original MB solution before PEC decolorization exhibits two defined UV-Vis absorption peaks at 292 nm and 664 nm, respectively. The peak observed at 292 nm was ascribed to the aromatic rings in the MB molecules, and the peak at 664 nm was attributed to the dimethylamino structure of MB. During the degradation process, the absorption peaks decrease with increasing PEC time. For a PEC time of 60 min, these two peaks had completely disappeared, and no new peaks appeared in the visible region, which indicated that complete decolorization of MB was achieved.
Recyclability is another essential factor to evaluate the photoelectrode. The stability of the rGT/FP photoelectrode was investigated by recycling the PEC decolorization processes. As shown in Figure 7, no obvious decrease was observed in the PEC activity after the rGT/FP photoelectrode had been used for four cycles. The electrochemical performance of rGT/FP was also tested to measure its stability; the results are shown in Figure S2 (Supplementary Material). The excellent recyclability is mainly due to the film-making materials used for preparing the rGT/FP photoelectrode. As a typical cellulose ether, ethyl cellulose has a similar structure to cellulose, which can cohere with filter paper by hydrogen bonds. In consequence, ethyl cellulose and ethyl alcohol film provide the mechanical support for the filter paper and also improve its durability (Philip et al. 2008). Table S2 in the Supplementary Material presents the optical images of the rGT/FP photoelectrode in four cycles. There was no obvious changes that could be observed, implying the excellent stability of the rGT/FP photoelectrode.
In conclusion, the hybrid rGT/FP photoelectrode based on filter paper was prepared and used for PEC decolorization of MB. Compared with the TiO2/FP photoelectrode, the rGT/FP photoelectrode shows enhanced PEC decolorization efficiency. The UV-Vis reflectance spectra reveal that the incorporation of TiO2 with rGO can extend the absorption range to the visible region. The photocurrent and linear sweep voltage results indicate that the enhanced PEC decolorization efficiency using the rGT/FP photoelectrode is attributed to the existence of rGO sheets. Furthermore, the PEC process exhibits a synergistic effect, in which the decolorization efficiency of MB by PEC treatment is higher than the summary of EC and PC treatment. This work provides a novel strategy for developing cost-effective and efficient PEC materials for decolorization of organic pollutants.
This work was financially supported by the National Natural Science Foundation of China (No. 51903107, 51803076), the Postdoctoral Science Foundation of Jiangsu Province (1701012B), the China Postdoctoral Science Foundation (2017M611696, 2018M63223), the Natural Science Foundation of Jiangsu Province (No. BK20190619, BK20180629), and the Open Project Program of Fujian Key Laboratory of Novel Functional Textile Fibers and Materials (Minjiang University), China (No. FKLTFM1809).
The Supplementary Material for this paper is available online at https://dx.doi.org/10.2166/wst.2019.425.