Abstract

Titanium dioxide-graphene (TiO2-G) composite was used for the photodegradation of alkylphenols in wastewater samples. The TiO2-G composites were prepared via sonochemical and calcination methods. The synthesized composite was characterized by X-ray diffraction (XRD), infrared spectroscopy (IR), scanning electron microscopy (SEM), transmission electron microscopy (TEM), energy dispersive X-ray analysis (EDX), and fluorescence spectroscopy. The photocatalytic efficiency was evaluated by studying the degradation profiles of alkylphenols using gas chromatography-flame ionization detector (GC-FID). It was found that the synthesized TiO2-G composites exhibit enhanced photocatalytic efficiencies as compared to pristine TiO2. The presence of graphene not only provides a large surface area support for the TiO2 photocatalyst, but also stabilizes charge separation by trapping electrons transferred from TiO2, thereby hindering charge transfer and enhancing its photocatalytic efficiency.

1. Introduction

Alkylphenol ethoxylates (APEs) are a class of nonionic surfactants that have been used extensively as detergents, emulsifiers, and dispersing agents [1]. It has been reported that during the biodegradation of APEs in conventional wastewater treatment plants (WWTPs), alkylphenols (APs) like octylphenol (OP) and nonylphenol (NP) are produced as persistent metabolites which possess higher toxicity and estrogenic activity and have greater tendency towards bioaccumulation [2]. With widespread use of these APs and poor efficiency of biodegradation, it is difficult to remove completely from wastewater treatment plants. NP is listed as a priority pollutant in the Water Framework Directive (WFD) [3], and studies have shown that trace levels of OP and NP can exert estrogenic effects on aquatic organisms and mammals [4]. The concentrations of APs detected in wastewater effluents were in the range of 0.01–15 gL−1 [1] while the proposed environmental quality standard (EQS) of the WFD for nonylphenol and octylphenol in surface waters is 0.3 gL−1 and 0.1 gL−1, respectively [5].

Advanced oxidation process (AOP) involving UV/TiO2 is a promising remediation technique for these phenolic endocrine-disrupting compounds since there is usually UV disinfection step in wastewater treatment plants and it is cheap. Titania is the most common semiconductor photocatalyst used due to its strong oxidizing power, nontoxicity, and long-term photostability [6]. In addition, studies have shown that complete mineralization of alkylphenols to CO2 and H2O can be achieved via the use of TiO2 [7]. However, it suffers from the major drawback of low photocatalytic efficiency due to the high rate of electron/hole pair recombination [8]. Thus the development of new materials for modifying TiO2 is needed to increase the photocatalytic activity for its practical applications in WWTPs.

When the TiO2 photocatalyst is irradiated with energy equal to or greater than the bandgap energy (3.2 eV), the electrons move to the conduction band (CB) to generate positive holes in the valence band (VB) [10]. The positive holes can react with adsorbed H2O to form hydroxyl radicals while the electrons react with O2 to form superoxide radicals. Being highly reactive, the OH and O2 radicals can oxidize the pollutants in solution or react with adsorbed pollutants. Otherwise, these electron and hole pairs can recombine. The high rate of electron and hole pair recombination thus restricts the efficiency of TiO2 photocatalysis. Many efforts are focused on enhancing the photoactivity of TiO2, and some of these include increasing the adsorption abilities of the photocatalyst surface by adding a coadsorbent [7] and modification of TiO2 with metal or semiconductor to improve the separation between free carriers [11].

Graphene is a flat monolayer of carbon atoms tightly packed in a two-dimensional honeycomb lattice, which has received tremendous attention over the years due to its extraordinary electronic, thermal, and mechanical properties [12]. It is a 0 ev bandgap semiconductor [13] with high mobility of charge carriers [14], high specific area (2600 m2g−1) [15], and a high adsorption capacity. Given the properties of graphene, the presence of graphene can enhance the photocatalytic efficiency of TiO2 [1619]. Since cost is an important consideration for practical applications in WWTPs, one added advantage of using TiO2-graphene composites as photocatalyst is that bulk production of graphene sheets can be produced from the inexpensive and abundant graphite at low costs [9].

In this study, TiO2-graphene composite was synthesized and characterized. The photocatalytic efficiencies for the degradation of 4-n-heptylphenol, 4-n-octylphenol, and 4-nonylphenol of the synthesized composites were compared with TiO2 using GC-FID. The degradation profiles of these APs were studied because they are the reported byproducts of the widely used commercial formulations of APEs.

2. Experimental

2.1. Chemicals and Reagents

The acid-functionalized graphite oxide was obtained from Associate Professor Loh Kian Ping’s research group at the National University of Singapore and was synthesized via Hummer’s method [20]. Tetrabutyl orthotitanate () was purchased from Fluka (Buchs, Switzerland), and hydrazine (35 weight % solution in water) was obtained from Sigma-Aldrich (Milwaukee, WI, USA). Technical ethanol was obtained from Riverbank Chemicals Pte Ltd (Singapore). The commercial photocatalyst, TiO2 P25 (Degussa AG, Germany) with primary particle diameter of 21 nm and specific surface area of  m2g−1, was used for comparison purposes. The following chemical standards () were obtained from Wako Chemicals (Tokyo, Japan): 4-n-heptylphenol, 4-n-octylphenol, and 4-nonylphenol (mixture of NP isomers). The derivatization agent bis(trimethylsilyl)trifluoroacetamide (BSTFA) () was obtained from Merck (Darmstadt, Germany). Ultrapure water was prepared on a Milli-Q water system (Milford, MA, USA). For the HPLC-grade organic solvents, acetonitrile was obtained from Lab-Scan Analytical Sciences (Bangkok, Thailand), and hexane was obtained from Fischer Scientific (Loughborough, UK). Stock standard solutions of the individual standards were prepared in acetonitrile at 1000 mgL−1 and stored at 4˚C. Aqueous solutions (200 mL) containing 5 mg/L mixtures of the analytes were prepared using ultrapure water.

2.2. Materials

Photodegradation was initiated via the use of a 15 W germicidal mercury lamp with a working wavelength of 254 nm from Ster-L-Ray, Atlantic Ultraviolet Corp. (Hauppauge, NY, USA). A jacketed Pyrex glass column (310 mm height and 70 mm inner diameter) was bought from UFO Labglass (S) Pte Ltd and was used as the photoreactor.

2.3. Synthesis of TiO2-Graphene Photocatalyst

The graphene sheets were combined with TiO2 nanoparticles through sonochemical and calcination methods [21]. Firstly, graphite oxide was sonicated in technical ethanol (20 mL) for 15 minutes to disperse them well and form graphene oxide solution (GO). Tetrabutyl orthotitanate (TBOT) was used as the titanium precursor, and it was added to the GO solution. Next, the mixture was sonicated for 30 minutes to improve the interaction between these chemicals. Milli-Q water (80 mL) was added, and the sol sample formed by hydrolysis was treated with ultrasonic irradiation in an ultrasonic cleaning bath (Branson 1210-USA). The sol sample was then aged at 25˚C for 20 hours to further hydrolyze TBOT and form monodispersed TiO2 nanoparticles. The samples were dried in a 100˚C oven. In order to reduce the GO to graphene (G), hydrazine solution (5 mL) was added to the dried sample and it was stirred for 24 hours. The reduced sample, TiO2-graphene (TiO2-G), was then washed with ultrapure water and dried in a dessicator. Finally, in order to convert the amorphous TiO2 to crystalline TiO2 which possesses higher photocatalytic activity, the mixtures were calcined under nitrogen atmosphere at 400˚C for 1 hour. TiO2-G composites with different mass ratios of titanium and carbon were synthesized (1 : 0.1 and 1 : 0.2). The preparation process of TiO2 was similar to that of TiO2-G, except that the addition of GO and reduction with hydrazine steps were skipped.

2.4. Characterization Techniques

Powder X-ray diffraction (XRD) was used for phase identification and to estimate the crystallite size of the anatase nanoparticles. The XRD spectra were obtained at room temperature with a Siemens X-ray diffractometer (Model: B5005) using Cu K radiation ( Ǻ), and data were collected from 2 = 20˚ to 80˚ at a step size of 0.020˚/s. FTIR spectra of TiO2, GO, and TiO2-graphene were obtained via a Varian 3100 FT-IR spectrophotometer using KBr pellets. Transmission electron microscopy (TEM) images and selected area electron diffraction (SAED) of the synthesized composite were performed via a JEOL JEM 3010 electron microscope to observe the surface morphology of the TiO2-G composite. EDX analysis was recorded on a JEOL JED-2300 scanning electron microscope to identify the elemental composition. The charge recombinations of TiO2-G and TiO2 were compared by obtaining the fluorescence emission spectra. Fluorescence spectroscopy analysis was carried out at room temperature via a Perkin Elmer Luminescence spectrometer (LS 55) equipped with a powder holder accessory, and the excitation wavelength was set at 300 nm.

2.5. Study of Photodegradation Profiles

The photocatalytic efficiencies of the synthesized composites were evaluated by studying the degradation profiles of alkylphenols in aqueous media under Ultraviolet C ( nm) irradiation. A batch reactor, which was housed in a black box, was used for the photocatalytic experiments, and the schematic diagram is shown in Figure 1. The solution was continuously stirred in order to obtain a homogeneous solution. Cooling water was circulated in the outer jacket in order to keep the reaction temperature constant. In each photodegradation test, 20 mg of photocatalyst was dispersed in 200 mL solution containing 5 ppm mixture, and the amount of suspended photocatalyst was kept at 0.1 g/L. To allow the adsorption-desorption equilibrium of alkylphenols on the photocatalyst, the suspension was stirred for 5 minutes under dark conditions. Prior to turning on the UV lamp, the concentration of the solution was determined, which was considered as the initial concentration ().

2.6. Analytical Procedure

In order to monitor the variation of analyte concentrations with time and obtain the degradation profiles, sample aliquots were withdrawn from the photoreactor at 45-minute intervals (over a period of 3 hours) and filtered through a 0.2 m syringe filter to remove the photocatalyst particles before analyses. Liquid-liquid extraction with previously optimized conditions [1] was used for study the degradation profile; briefly, a 5 mL of sample was adjusted to pH 2 using 1 M HCl in a 10 mL centrifuge tube and extracted twice with 200 L of dichloromethane. To remove trace amounts of water, anhydrous sodium sulphate was added to the organic layer and 100 L of bis(trimethylsilyl)trifluoroacetamide (BSTFA) was added. The mixture was kept in a 60˚C water bath for 20 min to complete the derivatization. From this, 2 L was injected to a GC-FID for analysis.

2.7. GC-FID Analysis

GC analyses were performed via an Agilent 7890A GC-FID system with a 7683B autosampler. The GC was equipped with a DB-5 fused silica capillary column, 30 m × 0.25 mm i.d. × 0.25 m film thickness (J&W Scientific, Folsom, CA, USA). Helium was used as the carrier gas at a flow rate of 1 mLmin−1. 2 L of the derivatized sample was injected into GC-FID under the splitless mode. The injection temperature was set at 300˚C and the temperature programming was as follows: 60˚C (held for 2 min), 20˚C/min to 180˚C, and 5˚Cmin−1 to 220˚C (held for 8 minutes).

3. Results and Discussion

3.1. XRD Characterization

Although there are three crystalline forms of TiO2 (anatase, brookite, and rutile), the anatase form of titania is reported to give the best combination of photoactivity and photostability [22]. The XRD spectra of the synthesized TiO2 and TiO2-graphene are shown in Figure 2. The peaks at 2 = 25.56˚, 38.34˚, 48.26˚, 54.48˚, 63.08˚ are attributed to the anatase phase [23]. The absence of the peaks at 2 = 27.5˚ and 30.8˚ indicates that the rutile and brookite phases are not present. In addition, it is observed that the peak width broadened slightly with the introduction of graphene. It can be seen from Table 1 that the average crystallite size of TiO2-G is smaller than that of pristine TiO2. It is thus inferred that due to the strong interactions between TiO2 and graphene during the hydrolysis of the sol samples, the TiO2 crystalline particles were unable to grow larger [18].

3.2. FTIR Spectroscopy

FTIR spectroscopy was used to characterize the interaction between graphene and TiO2 nanoparticles. The FTIR spectrum of graphene oxide (Figure 3(a)) shows bands which are attributed to the epoxide ( at 1250 cm−1), carboxylic acid ( at 1720 cm−1), and hydroxyl groups ( at 1365 cm−1) [24]. Conversely, it can be seen from Figure 3(c) that most of these oxygen functionalities had been removed in the synthesized TiO2-graphene composite. A new band at 1580 cm−1 which may be attributed to the skeletal vibrations of the graphene sheets [25] and a strong band at 500–700 cm−1, which is due to the Ti–O–Ti vibration [26] confirms the presence of both graphene and TiO2 in the synthesized composite.

The IR spectrum of TiO2 shows a broad band at 3400 cm−1 and the band at 1625 cm−1 originates from the surface-adsorbed water, and this indicates the presence of −OH groups on the surface of titania [27]. Comparatively, the band at 2400–3400 cm−1 of TiO2-graphene is broader, thus indicating the presence of hydrogen bonding in the synthesized composite. Since hydrazine reduction introduces nitrogen groups onto graphene, the OH groups present on the surface of titania could form hydrogen bonds with the NH groups present on the graphene sheets to form the composite. In addition, a new band at 1210 cm−1 is observed in the spectrum obtained for TiO2-G, and this can be assigned to Ti–O–C vibrations [28]. Since hydrazine does not cause a full reduction of graphene oxide [20], the Ti–O–C bond may be formed when OH groups present in TiO2 react with residual OH groups present on the graphene sheets in removing water.

3.3. Characterization of TiO2-G by Transmission Electron Microscopy

From the TEM images, it can be seen that the graphene sheets are coated with TiO2 nanoparticles. Figure 4(c) illustrates the homogeneous dispersion of TiO2 nanoparticles on the graphene sheet, and it can be speculated that under ultrasonic irradiation, the graphene sheets interact with TiO2 through strong chemisorption and physisorption [29]. In addition, the graphene sheets appear transparent and are folded over one edge [24]. Thus, this indicates that single graphene sheets have been isolated.

3.4. Energy Dispersive X-Ray Elemental Microanalysis

The EDX spectrum shows the presence of Ti, C, and O atoms in the synthesized TiO2-graphene composite (Figure 5). It can be seen from Table 2 that the atom percentage of O is slightly more than twice that of Ti. This indicated that some oxygen functionalities remained on the graphene sheets after hydrazine reduction, which is in agreement with the IR spectrum obtained.

3.5. Fluorescence Emission Spectroscopy

Since photocatalytic activity is a function of lifetime and trapping of electron and hole pairs, fluorescence emission spectra are a useful characterization technique to investigate the efficiency of charge carrier trapping and to understand the fate of e−/h+ pairs in semiconductor particles like TiO2 [21]. Recombination of electron and hole pairs can emit energy in the form of fluorescence [30], and as seen from Figure 6, the fluorescence intensity of TiO2-G is lower than that of TiO2. This indicates that the recombination of electron and hole pairs is suppressed in the presence of graphene, which possibly indicates an increase in photocatalytic efficiency. In addition, the fluorescence emission spectra displayed 3 main peaks at 380 nm, 420 nm, and 484 nm for both TiO2 and TiO2-G, which are attributed to the self-trapped excitons localized on TiO6 octahedra and oxygen vacancies [31]. Thus, this indicates that the presence of graphene does not alter the mechanism of TiO2 photocatalysis.

3.6. Method Validation

Due to the polarity of APs, derivatization of the analytes is required prior to GC-FID analyses [32]. Silylation using bis(trimethylsilyl)trifluoroacetamide (BSTFA) is a rapid and commonly used derivatization technique in which the active hydrogen of the hydroxyl groups is converted into trimethylsilyl (TMS) group [33]. Thermally stable and highly volatile derivatives are obtained, and this results in improved gas chromatographic parameters such as accuracy, reproducibility, sensitivity, and resolution [34]. Liquid-liquid extraction was employed to monitor the degradation rates since it is a simple, fast, and reliable technique for monitoring the disappearance of the parent compound [35]. In order to access the practicality and suitability of this proposed LLE method, the optimized extraction conditions were used to determine the extraction method’s repeatability, linearity, limits of detection (LOD), and limits of quantification (LOQ). The results are summarized in Table 3. Repeatability was evaluated by triplicate analysis at the various analyte concentrations within the linear range of the extraction method. Satisfactory repeatability of relative standard deviations (RSDs) 9 to 12% was obtained. The linearity of this extraction method was evaluated at five different concentrations, ranging from 5 to 1000 g/L. The limit of detection of the analytes was determined (between 0.4 and 0.8 g/L) based on S/N ratio is 3, while the limit of quantification (LOQ) was calculated (between 1.3 and 2.6 g/L) based on the definition of .

3.6.1. Comparison of Photocatalytic Efficiency of the Synthesized TiO2-G, TiO2, and Commercial Photocatalyst P25 on Alkylphenols in Wastewater Samples

Prior photocatalysis, wastewater samples were collected from local drainages and extracted using optimized LLE method. No target compounds were detected in the wastewater samples. Therefore, experiments were designed with spiked (100 g/L of alkylphenols) wastewater samples. The degradation profiles of alkylphenols are presented in Figure 7. The concentrations of photocatalyst used for each experiment is 0.1 g/L. A low dosage of photocatalyst loading was chosen for the comparison of photocatalytic efficiency since it would be more cost effective for practical applications. In order to confirm that the source of decrease in analyte response was due to photodegradation rather than sample loss from extraction, a control experiment was carried out in which the sample solution was continuously stirred over a period of 3 hours and the variation in concentration was monitored. It can be seen that the percentage of the original analyte left remained unchanged. Thus photodegradation and not sample loss was the main source of declining peak areas which was observed in the subsequent experiments.

In addition, to monitor the effects due to direct photolysis only, the sample solution was irradiated with UV light. It was observed that the concentrations decreased slightly with UV irradiation, with more than 90% of the original solution still remaining at the end of 3 hours.

When the photocatalysts were added, the peak areas of the analytes declined significantly, thus indicating the efficacy of advanced oxidation processes for the removal of these phenolic endocrine disrupting compounds. Figures 8(a) and 8(b) show the chromatogram after the dark adsorption period () and the chromatogram after the addition of TiO2-G photocatalyst with UV irradiation for 3 hours, respectively. The percentage of individual analytes that remained after 3 hours of UV irradiation with the different photocatalysts is shown in Table 4. It was observed that TiO2-graphene composite with Ti : C ratio of 1 : 0.1 exhibited enhanced photocatalytic efficiencies for the photodegradation of the analytes over that of pristine TiO2, which was synthesized in the same way. As a comparison, the photocatalytic efficiencies were compared to that of the commercial TiO2 photocatalyst, which has a phase composition of 80% anatase and 20% rutile. The synthesized TiO2-graphene (1 : 0.1) also displayed higher photocatalytic efficiency as compared to that of the commercial TiO2 photocatalyst P25 for the degradation of 4-nonylphenol and 4-n-octylphenol. On the other hand, the TiO2-graphene photocatalyst with a higher carbon loading (Ti : C ratio of 1 : 0.2) had the highest photocatalytic efficiency for the photodegradation of 4-n-heptylphenol. The degradation rate of 4-nonylphenol was the lowest since technical nonylphenol is a mixture of isomers which contain both straight- and branched-chain NPs [36].

3.6.2. Performance of TiO2-G Composite on Photocatalytic Efficiency

From the fluorescence emission spectra obtained, it can be seen that the presence of graphene can suppress the recombination of electron and hole pairs of TiO2. Under UV irradiation, electrons are excited from the valence band (VB) to the conduction band (CB) of the anatase, thus creating a hole in the valence band. In the absence of graphene, most of these charges recombine and only a small fraction participate in photocatalytic reactions (1%), thus resulting in low photoefficiency. Given the semiconducting and unique electronic properties of graphene, electrons from the anatase particles can be transferred to the graphene sheets, thus allowing charge separation, stabilization, and hindered recombination [37]. Consequently, the holes on the anatase particles are longer-lived and thus the TiO2-G composite has higher photocatalytic efficiency.

In addition, the large specific surface area of graphene sheets can also account for the enhanced photocatalytic efficiency of TiO2-G. Graphene sheets can adsorb the alkylphenol molecules in aqueous solutions and concentrate them on the surface of titania [38]. It was noted that the APs at time = 0 min are largely decreased from the prepared initial concentration due to strong adsorption on the photocatalyst surface [32]. Given the highly hydrophobic nature of the long-chained alkylphenols, they tend to adsorb on the highly hydrophobic graphene sheets. As a result, the dispersion of TiO2 on graphene sheets and adsorption of the alkylphenol molecules created many active sites for photocatalytic degradation and thus enhanced the photocatalytic efficiency.

The photocatalytic efficiency of the TiO2-G composite would increase with increase of the graphene percentage at optimum experimental conditions. The TiO2-G composite with Ti : C mass ratio of 1 : 0.1 displayed highest photocatalytic efficiency for the degradation of 4-nonylphenol and 4-n-octylphenol. One reason for the reduced photocatalytic efficiency with increase of graphene content is that higher composition of graphene sheets can shield the UV light from absorption by TiO2 nanoparticles [18]. Thus, with reduced UV adsorption by the TiO2 particles, the composite would have reduced photocatalytic efficiency. Further studies are required to investigate this mechanism. From the experimental results obtained, higher photocatalytic efficiencies can be achieved with a graphene composition which is only 10% that of TiO2 present for 4-n-nonylphenol and 4-octylphenol whereas a higher graphene loading (20% of TiO2 present) in the synthesized composite was favorable for photodegradation of 4-heptylphenol. Thus, the presence of a small percentage of graphene can have synergetic effects on the photocatalytic efficiency of TiO2.

4. Conclusion

In this study, TiO2-graphene composite has been successfully synthesized and characterized. Compared to that of pristine TiO2, TiO2-Graphene exhibited enhanced photocatalytic efficiency of alkylphenols. By varying the composition of graphene in the photocatalyst, the photocatalytic efficiency was also varied. The photocatalytic efficiency of the photocatalyst exhibited better photocatalytic efficiency than the commercial TiO2 photocatalyst, P25 with optimal loading of graphene. This might be due to the following reasons: (i) large surface area of graphene allowed it to act as a coadsorbent for the alkylphenol molecules and it resulted in more active sites for photocatalytic reactions to occur and (ii) graphene could inhibit the recombination of electron and hole pairs of TiO2, thus enhancing charge transfer and allowing the oxidation of adsorbed molecules.

Acknowledgments

The author gratefully acknowledges the support of this research by the National University of Singapore and also would like to acknowledge the support provided by the King Abdul Aziz City for Science and Technology through the Science and Technology Unit at the King Fahd University of Petroleum and Minerals for funding this work through Project no. 10-WAT1396-04 as part of the National Science Technology and Innovation Plan.