Calcination of reduced graphene oxide decorated TiO2 composites for recovery and reuse in photocatalytic applications

Calcination of reduced graphene oxide decorated TiO2 composites for recovery and reuse in photocatalytic applications

Ceramics International xx (xxxx) xxxx–xxxx Contents lists available at ScienceDirect Ceramics International journal homepage: www.elsevier.com/locat...

2MB Sizes 0 Downloads 21 Views

Ceramics International xx (xxxx) xxxx–xxxx

Contents lists available at ScienceDirect

Ceramics International journal homepage: www.elsevier.com/locate/ceramint

Calcination of reduced graphene oxide decorated TiO2 composites for recovery and reuse in photocatalytic applications ⁎

Yanyan Zhanga, Xinggang Houb, , Tingting Sunb, Xinlei Zhaob a b

School of Mathematical Sciences, Tianjin Normal University, Tianjin 300387, China Department of Physics, Tianjin Normal University, Tianjin 300387, China

A R T I C L E I N F O

A BS T RAC T

Keywords: RGO/GO–TiO2 Calcination Recovery of photocatalyst Visible and solar light irradiation Decomposition

Reduced graphene oxide (RGO)–TiO2 composites were synthesized hydrothermally. Then the samples were calcined at 450 °C for 30 min under air atmosphere to recovery and improve their photocatalytic efficiency. Those composites were characterized by XRD, TEM, SEM, TGA, BET, XPS, Raman, FTIR and DRS. Photocatalytic activities of RGO–TiO2 composites and calcinated samples were measured by decomposition of methyl orange (MO) under UV, visible and solar light irradiation. Results showed that some RGO was burn out, and others turned into GO after calcination. Both RGO–TiO2 composites and calcined samples exhibited red shift of the absorption edge. TiO2 particles were well dispersed on the surface of GO sheets, and calcinated samples with smaller TiO2 particles were observed. Graphene amount showed an apparent influence on photocatalytic activities of both RGO–TiO2 composites and calcined samples. At 5 wt% graphene concentration, both samples exhibited the best photocatalytic activities under UV light irradiation, while the concentration was 10 wt% under visible light irradiation. Moreover, calcined samples showed better photocatalytic activities than that of RGO–TiO2 composites. The enhanced photocatalytic activities were because of the oxidation of residual organics in the RGO–TiO2 composites, and better crystalized TiO2 particles with smaller diameter after calcination. Calcination is a valid method to recover the photocatalytic activities of RGO–TiO2 composites.

1. Introduction The availability of clean drinking water is an essential right for human existence and quality of life [1]. Therefore, cheap, feasible and sustainable methods to recycle and reuse water are becoming very important. Photo disinfection and degradation can improve the safety of the water by reducing the loading of pathogenic microorganisms and hard-degraded pollutant. One effective method to improve the activity of solar disinfection and degradation is the utilization of TiO2 [2–4]. However, practical applications of TiO2 are severely limited by rapid recombination of photo generated electron–holes, poor light absorption in visible region and declined photocatalytic efficiency after recycling [5]. To date, a variety of strategies have been adopted to enhance the photocatalytic performance of TiO2, including the design of textured photocatalysts with increased porosity and surface area, the use of ions doping, and the deposition of noble metal nanoparticles, metal oxide or metal hydroxide [6–13]. The combination of TiO2 with carbon materials, including graphite, carbon black, activated carbon, carbon fibers, carbon nanotubes, fullerenes and graphene, have also been



investigated [14–17]. Among these carbon materials, graphene, a twodimensional sp2-hybridized allotrope of carbon, has attracted a great deal of attention [18]. Based on the huge surface area providing good contact with TiO2 nanoparticles, considerable efforts have been devoted to synthesizing graphene oxide (GO) or reduced graphene oxide (RGO) decorated TiO2 composites [19–35]. It is suggested that RGO/ GO acts as an excellent electron scavenger, effectively decreasing the recombination rate of photogenerated carriers of TiO2. Furthermore, RGO/GO is considered an ideal sensitizer to enhance the visible light photocatalytic performance of TiO2. The hydrothermal method has been found to be an effective way to synthesize RGO–TiO2 composites, during which chemical bonds are formed between the TiO2 and RGO, and the TiO2 particles are distributed uniformly on the surface of the RGO [2,36,37]. The combination of TiO2 and graphene has been proved to have the superior photocatalytic activities under both visible and UV light illumination. For example, Wang et al. fabricated a hierarchical macro/mesoporous TiO2–graphene composites photocatalyst with a significant enhancement in photodegradation of acetone in air [38]. Zhou et al. obtained 3D urchin-like TiO2-reduced graphene micro/

Corresponding author. E-mail address: [email protected] (X. Hou).

http://dx.doi.org/10.1016/j.ceramint.2016.10.056 Received 20 September 2016; Received in revised form 7 October 2016; Accepted 8 October 2016 Available online xxxx 0272-8842/ © 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Please cite this article as: Zhang, Y., Ceramics International (2016), http://dx.doi.org/10.1016/j.ceramint.2016.10.056

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al.

nano structure composite photocatalysts for photodegradation of RhB [39]. Rong et al. prepared TiO2–graphene photocatalyst with significant enhancement of photocatalytic activity for degrading the MB under visible light irradiation [40]. Qu et al. developed graphene quantum dots decorated TiO2 nanotubes for degrading the MO under UV–vis light irradiation [41]. Rahimi et al. prepared TiO2–graphene nanocomposite sensitized with tetrakis porphyrin which were used as catalyst for disinfection of E. coli under visible light irradiation [42]. Luo et al. used fluorinated anatase TiO2/reduced graphene oxide nanocomposites for degradation of bisphenol A [43]. However, the aforementioned studies mainly focused on the enhanced photocatalytic behavior of TiO2–graphene composites via the modified morphology of TiO2 or the decoration of TiO2. There is still scarce work on the reuse of TiO2–graphene composites after repeated usage. It is well-known that the photocatalytic efficiency of RGO–TiO2 composites would decrease after using a certain times compared with the first-cycle results [27,44,45]. Therefore, for the goal of lasting photocatalytic ability in practice, the method to recovery the photocatalytic ability of RGO– TiO2 is of particularly important. Usually, calcination can improve the photocatalytic ability of TiO2 prepared by hydrothermal method due to the crystallization of TiO2 [2,3]. For RGO–TiO2 composites, however, there have been few reports of thermal treatment greater than 350 °C, probably because of concern about the oxidation of the RGO. However, it was reported that the carbon skeleton of RGO was completely oxidized with temperatures of 450–550 °C [22,27,46]. So, because the well anatase phase of TiO2 is formed at 450 °C, the thermal treatment at this temperature can be chosen to recover the photocatalytic ability of RGO–TiO2 after using a certain times. The aim of this work was to investigate the probability to recovery and reuse RGO–TiO2 by post calcination. To achieve these goals, RGO– TiO2 composites were synthesized by a hydrothermal method. And then the composites were calcined at 450 °C to investigate the influence of post thermal treatment on the stability of the chemical bonds between GO and TiO2. The properties of all those materials in photocatalytic degradation of methyl orange (MO) and disinfection of E. coli were tested (displayed in SI).

composite by weight. Details of the preparation of GO and TiO2 sol were given in SI. In the second stage of the synthesis, described here for RGO–TiO2–5 as an example, 0.1 g as-prepared GO was sonicated for 1 h in a mixture of 20 mL deionized water and 10 mL ethanol until it was well dispersed. After the sonication, the dispersion was added to 36.55 mL of the TiO2 sol and stirred for 2 h to obtain a homogeneous mixture. This mixture was transferred to a 100 mL Teflon-lined autoclave and maintained at 180 °C for 24 h to simultaneously convert the GO to RGO and to deposit the TiO2 particles on the carbon substrate. The resulting composite was then washed several times with deionized water and dried overnight at room temperature. TiO2 and RGO–TiO2–X composites were annealed in air at 450 °C for 30 min with heating rates of 1.5 °C min−1, and then allowed to cool naturally. The obtained composites were denoted by TiO2–c and GO– TiO2–X.

2. Experimental

2.3. Measurement of photocatalytic activities

2.1. Synthesis of RGO–TiO2 and GO–TiO2 composites

The photocatalytic activities of the prepared composites were investigated by the degradation of MO. The ultraviolet (UV) source was a 500 W high–pressure mercury lamp (with a 365 nm band pass filter), and a 300 W Xe lamp with a cut–off filter to eliminate any irradiation below 400 nm was used as visible light source. All solar photocatalytic experiments were carried out at Tianjin normal

2.2. Characterization X-ray diffraction (XRD) patterns were obtained with an X′Pert PRO MPD diffractometer (PANalytical, Holland). Transmission electron microscopy (TEM) was performed with an H600 microscope (Hitachi, Japan). The microscopic structure of the films was performed by scanning electron microscopy (SEM, Hitachi SU8010). Fourier– transform infrared (FTIR) spectra were recorded on an IR Affinity–1 spectrometer (Shimadzu, Japan). Raman spectra were recorded using a LavRAM Aramis Raman microscope (Horiba Jobin Yivon, France) with 532 nm excitation source at room temperature. X–ray photoelectron spectroscopy (XPS) measurements were obtained with an ESCA750 photoelectron spectrometer (Shimadzu, Japan). Peak deconvolution of the XPS spectra was performed using the Gaussian component after a Shirley background subtraction. Thermogravimetric analysis (TGA) was carried out on a DTG–60H instrument (Shimadzu, Japan) at a heating rate of 100 °C/min under air flow. UV–visible diffuse reflectance spectra (DRS) were recorded on a UV–3600 spectrophotometer (Shimadzu, Japan). The Brunauer–Emmett–Teller (BET) specific surface area SBET was measured at −196 °C with a BELSORP–max volumetric apparatus (Bel, Japan).

GO was synthesized by a modification of Hummers’ method [47]. RGO–TiO2 composites were prepared by hydrothermal method with different proportions of GO and TiO2. These composites were denoted by RGO–TiO2–X, where X=0.1, 1, 5 and 10 is the GO percentage of the

Fig. 1. XRD patterns of (1) TiO2 and RGO–TiO2–X: (a) TiO2, (b) RGO–TiO2–0.1, (c) RGO–TiO2–1, (d) RGO–TiO2–5 and (e) RGO–TiO2–10; (2) TiO2–c and GO–TiO2–X: (a) TiO2–c, (b) GO–TiO2–0.1, (c) GO–TiO2–1, (d) GO–TiO2–5 and (e) GO–TiO2–10.

2

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al.

Compared to TiO2 and RGO–TiO2–X, the intensity of diffraction peaks of TiO2–c and GO–TiO2–X became stronger, indicating better anatase crystallization. The crystallite sizes of the pure TiO2 particles were around 15 nm, calculated from the (101) peak (2θ=25.24°, Fig. 1(1) (a)) using the Scherrer formula. The average crystal size of TiO2 nanoparticles from Fig. 1(2)(a) was found to be around 30 nm, indicating increases in the anatase crystallite size after calcination. For GO–TiO2–X nanoparticles, there was a significant reduction in crystallite size with increasing X (i.e. increasing GO amount). Fig. 2 shows TEM images of GO, TiO2, GO–TiO2–0.1 and GO– TiO2–5. The characteristic crumpled morphology with several stacking layers of graphene sheets can be seen (Fig. 2(a)). The TiO2–c nanoparticles were in the size range 30–40 nm and agglomerated severely (Fig. 2(b)). It can be seen from Fig. 2(c) and (d) that in the GO–TiO2–X composites, the sheet-like structure of GO was covered by TiO2, as a consequence of the bonding between the TiO2 and GO [21]. It should be noted that although the composites were calcined in air at 450 °C for 30 min, the GO had not disappeared entirely. Together with the XPS, Raman and FTIR results, the TEM images indicate that the RGO had not been destroyed completely, and the sheet-like structure of the composites consists of GO. Furthermore, the particle dispersion of GO–TiO2–5 was better than that of TiO2–c, which was in agreement with the results of SBET measurements. The representative SEM images of TiO2, TiO2–c, RGO–TiO2–X and GO–TiO2–X are shown in Fig. 3 respectively. Compared Fig. 3(a) with Fig. 3(e), TiO2 particles with a diameter range of 15 nm were packed randomly, while TiO2–c particles aggregated to form larger particles with size about 30–40 nm after calcination. However, the changing of particles size was different when GO incorporated with TiO2. All the RGO–TiO2–X and GO–TiO2–X composites possessed particles smaller than 10 nm. Compared Fig. 3(b)–(c) with Fig. 3(f)–(g), it can be found that the particles of GO–TiO2–X was smaller than that of RGO–TiO2– X when X is the identical value, although GO–TiO2–X was annealed with the same temperature and time as TiO2–c. Furthermore, the particles of RGO–TiO2–X and GO–TiO2–X presented similar size below 5 nm as the GO amount was increased to 10 wt%. In the case of RGO–TiO2–X, TiO2 well distributed on the surface of the RGO

University, China, in June and November 2015. The experiments were performed in the daytime between 12:30 pm to 16:00 pm. Each composite (200 mg) was mixed with a series of aqueous solution of MO (200 mL) with a concentration of 3.5×10−5 mol/L and the resulting suspension was sonicated for 15 min. The aqueous system was stirred in the dark for 1 h to attain adsorption equilibrium between the MO and the samples before light irradiation. The series of suspension of each composite were irradiated by UV, visible and solar light for a regular interval respectively. After irradiation for each suspension, a series of 20 mL samples were collected and centrifuged to remove the suspended particles for analysis. The MO concentration was analyzed with a UV-3600 spectrophotometer. 3. Results and discussion 3.1. Materials characterization Fig. 1(1) shows the XRD patterns of TiO2 and RGO–TiO2–X composites. Six main characteristic peaks of anatase phase TiO2 clearly indicated the successful synthesis of TiO2 nanoparticles by the hydrothermal method. No diffraction peaks due to the layered structure of RGO can be seen, indicating that the main characteristic peak of RGO might be shielded by the main peak of anatase TiO2 [21–23]. The colour of TiO2 was brown, and the colour of RGO–TiO2–X turned into black gradually when the amount of GO was increased (Fig. S2). However, due to the relatively low diffraction intensity of GO and its partial reduction to RGO during hydrothermal reaction, two of the reflection peaks of GO were only observed at 32° and 43° in sample with the highest amount of GO as shown in Fig. 1(1)(e). The patterns of TiO2–c and GO–TiO2–X composites are shown in Fig. 1(2). It can be seen that those composites exhibited similar XRD peaks. There were no diffraction peaks of GO, indicating that the GO amount contained in composites decreased greatly after calcination. The colour of TiO2–c was light gray, while the colour of all GO–TiO2–X composites was turned into white (Fig. S2). It is worth noting that the intensities of the diffraction peaks for the composites showed a slight decrease as the GO amount increased for both hydrothermal and annealing samples.

Fig. 2. TEM images of (a) GO, (b) TiO2–c, (c) GO–TiO2–0.1, and (d) GO–TiO2–5.

3

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al.

Fig. 3. SEM images of (a) TiO2, (b) RGO–TiO2–1, (c) RGO–TiO2–5, (d) RGO–TiO2–10, (e) TiO2–c, (f) GO–TiO2–1, (g) GO–TiO2–5, (h) GO–TiO2–10.

4

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al.

anatase [27,32], suggesting the formation of TiO2 on the surface of GO sheets. There are two typical characteristic peaks at around 1330 and 1595 cm−1, corresponding to the D and G bands of graphite, respectively [48–50]. The D band is generally ascribed to one breathing mode of K–point photons of A1 g symmetry, related to the edges, disordered carbon and other defects in GO. The G band arises from the zone– center E2 g mode, corresponding to ordered sp2–bonded carbon atoms [26,28]. Thus, the ratio of the intensity of the D band to that of the G band, ID/IG, is a measure of the degree of disorder in the GO or RGO. It can be seen from Fig. 6 that, compared with the ID/IG ratio for GO, there was a decrease in ID/IG for both RGO–TiO2–5 and GO–TiO2–5, indicating the presence of GO in the composites. It has been reported that ID/IG is inversely proportional to the size of the in-plane sp2 domain. Therefore, the decrease in ID/IG for RGO–TiO2–5 indicated an increase in the average size of the in-plane sp2 domains and the partial reduction of GO after hydrothermal reaction [23,24,27,29,30]. The higher ID/IG of GO–TiO2–5 compared with RGO–TiO2–5 could be ascribed to oxidation of the carbon. Fig. 7(a), (b) and (c) show the FTIR spectra of TiO2, GO and GO– TiO2–5, respectively, in the range 400–4000 cm−1. For GO–TiO2–5, there was a broad absorption band around 500 cm−1. This is attributable to the vibrations of Ti–O–Ti and Ti–O–C (798 cm−1) [22,23,27]. The presence of Ti–O–C bonds indicated strong bonding between the TiO2 nanoparticles and GO. Furthermore, an absorption band at 1567 cm−1 could clearly be observed for GO–TiO2–5, attributable to skeletal vibration of the graphene [20,22]. All of these results confirmed the presence of GO in the GO–TiO2–5 composite. The weakness of the peak in the GO–TiO2–5 spectrum at 1384 cm−1, due to C–H vibration, indicates that the organic groups presented in RGO–TiO2–5 have been almost entirely eliminated from the calcined composite [22]. The existence of residual organics would have a detrimental effect on the photocatalytic performance of the composite, and thus the calcination was of benefit. UV–visible diffuse reflectance spectra of TiO2–c, RGO–TiO2–5, and GO–TiO2–X (X=1, 5 and10) are compared in Fig. 8(a). Compared with bare TiO2–c, it can be seen that there was a red shift of the absorption edge after the addition of GO to TiO2 (Fig. 8a). This shift was due to chemical bonding between the TiO2 nanoparticles and carbon, which caused a narrowing of the band gap of TiO2 [22,26,32,33]. Compared to RGO–TiO2–5, the greater narrowing of the band gap occurred with GO–TiO2–5, and broad background absorption of it in the visible region was also observed. These results suggest that interaction between TiO2 and GO in the calcined sample was stronger than in the sample synthesized by hydrothermal reaction alone. After calcination, the red shift increased. With the formation of GO, the broad background absorption of RGO–TiO2 composites in the visible region was extended [21,23,26], which implies that GO had bonded with the

nanosheets. The structure may favor to hinder the TiO2 from agglomeration and enable their smaller particles appearance on the RGO. It should be noted that nanosized crystallites tend to agglomerate during calcination, so calcination of RGO–TiO2–X is beneficial to the elimination of agglomerates. Furthermore, the presence of particles less than 5 nm indicates that the growth of TiO2 nanoparticles was restricted in combination with GO, which is attributable to the confining effect of the GO sheets. Fig. 4 shows the TG analysis curves of TiO2 and RGO–TiO2–X. Chun et al. observed rapid weight loss of GO in the range 300–700 °C due to initial dehydroxylation of the catalyst surface and combustion of carbon substrates [46]. However, all RGO–TiO2–X curves show a gentle slope similar to TiO2. Weight loss of 12.1%, 13.6%, 17.6%, 19.5% and 26.4% are corresponding to TiO2 and RGO–TiO2–0.1, 1, 5 and 10 respectively in the temperature range of 30–450 °C, which showed that 1.5%, 5.4%, 7.4% and 14.3% more organic materials were decomposed in comparison with TiO2. There might be two reasons to account for this result. One is that the GO amount of RGO–TiO2–X was higher than that added into TiO2 sol, the other is that oxidization of carbon skeleton of GO enhanced the elimination of organic materials in TiO2. All RGO–TiO2–X curves kept its slop until the temperature rised to 500 °C corresponding to entire oxidation of GO, and then a horizontal line appeared, which means the vigorous oxidation of RGO didn’t occur in the range of 300–1000 °C. A comparison of the TGA curves of TiO2 and RGO–TiO2–X confirms that embedded TiO2 particles improved thermal stabilization of GO, probably as a result of bonding between oxygen functional groups on the surface of GO and TiO2 nanoparticles. It should be pointed that RGO–TiO2–10 showed faster weight loss compared to RGO–TiO2–5, which is partly ascribed to greater loss of GO from RGO–TiO2–10 during the calcination compared with that from RGO–TiO2–5, because of more extensive combustion of the carbon constituents in RGO–TiO2–10. In order to study the effect of calcination on the RGO–TiO2–X composites, SBET was measured, and the results are shown in Table 1. The RGO–TiO2–X composites had larger SBET values than that of TiO2, and it is also clear that SBET increased with increasing GO amount, which may be attributed to increased SBET of the RGO sheets and TiO2 arising from the higher fraction of RGO in the composites [21,24,25]. So, the SBET values of RGO–TiO2–5 and RGO–TiO2–10 were not very different, which is because, above a certain amount of GO, further increases dis not greatly enhance the SBET of the nanoparticles. The decrease in the SBET of TiO2–c was due to a progressive increase in TiO2 particle size. However, the particle size of the GO–TiO2–X composites decreased after calcination. So, different from TiO2 particles, the main reason of the reduced SBET of GO–TiO2–X composites was attributable to combustion of part of the GO and release of oxygenated surface groups. The GO–TiO2–X composites could effectively sustain local strain during calcination and mesopores were prevented from collapsing for the remained GO sheets. Thus, the decrease in SBET was not so great for GO–TiO2–5 and GO–TiO2–10. The chemical states of the elements in the GO, RGO–TiO2–5 and GO–TiO2–5 composites were analyzed by XPS, as shown in Fig. 5. Fig. 5(a) shows the deconvolution of the C 1 s peak of GO. There are four peaks, centered at 284.8 eV (due to C–C, C=C and C–H bonds), 285.8 eV (C–O(H)), 287.3 eV (C=O) and 288.8 eV (O=C–OH) [22–24]. For RGO–TiO2–5 (Fig. 5b), the intensities of all C 1 s peaks due to carbon–oxygen bonds decreased dramatically, which indicates that most of the oxygen–containing functional groups have been partially removed, and it is worthy of attention that the C=O bonds were removed completely after reduction [25–27]. As can be seen in Fig. 5(c), after calcination, the peaks due to the oxygen–containing functional groups decreased slightly compared with the pre-calcination composite, which implies that RGO and residue organics were partly oxidized during calcination at 450 °C under air atmosphere. The Ti spectrum of GO–TiO2–5 (Fig. 5(d)) shows two peaks centered at 458.5 and 464.2 eV, which corresponded to the binding energy of Ti4+ in bare

100

TiO2-c GO-TiO2-0.1 GO-TiO2-1 GO-TiO2-5 GO-TiO2-10

Weight (%)

95 90 85 80 75 70

200

400

600

800

Temperature (°C) Fig. 4. Thermal gravimetric analysis curves.

5

1000

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al.

Sample

SBET/m2 g−1

Sample

SBET /m2 g−1

TiO2 RGO–TiO2–0.1 RGO–TiO2–1 RGO–TiO2–5 RGO–TiO2–10

56.8 82.8 86.8 138.6 139.2

TiO2–c GO–TiO2–0.1 GO–TiO2–1 GO–TiO2–5 GO–TiO2–10

40.1 61.7 74.7 121.9 125.4

Intensity (a.u.)

Table 1 The BET specific surface area of the composites.

TiO2. Fig. 8(a) shows that increased band-gap narrowing was proportional to the GO amount of the composite. The reason for this may be the formation of the chemical bonding between TiO2 and the specific sites of carbon (Ti-O-C bond) similar to that found for P25–GO composites [30,50,51]. This effect was proportional to the GO amount, and a weaker absorption in the visible region was observed for the composites with lower amount of GO. The evaluated band gaps for GO–TiO2–10, GO–TiO2–5, RGO–TiO2–5, GO–TiO2–1 and TiO2 were about 2.8, 2.9, 3.05, 3.05 and 3.15, respectively (Fig. 8b).

290

292

286.1 eV

280

282

Binding Energy (eV)

1800

284

288.3 eV

286

288

290

292

Binding Energy (eV) 458.5 eV

(d)

Intensity (a.u.)

284.8 eV

Intensity (a.u.)

(c)

1600

284.6 eV

Intensity (a.u.)

Intensity (a.u.)

(b)

288.8 eV

288

1400

286.1 eV

464.2 eV

288.8 eV

280

282

284

286

288

290

2000

The enhanced photocatalytic efficiency of RGO–TiO2–X was due to the hydrothermal reaction, in which chemical bonds were formed between TiO2 and GO. The resulting good connection between TiO2 particles and GO sheets resulted in a synergistic effect between TiO2 and GO, which improved the photocatalytic properties of the composites. Obviously, RGO–TiO2–5 exhibited the highest photocatalytic efficiency in 60 min irradiation under UV light, which means that overloading of GO would weaken the photocatalytic performance of TiO2. It was worth noting that the photodegradation rate of RGO–TiO2–10 (39.7%) was lower than that of TiO2 (41.9%). The phenomenon can be ascribed to the following reasons: (1) with increasing GO amount, light harvesting

285.8 eV

286

1200

ID/IG=1.46

Fig. 6. Raman spectra of (a) GO, (b) RGO–TiO2–5, and (c) GO–TiO2–5.

284.8 eV

284

×1

Raman Shift (cm-1)

287.3 eV

282

ID/IG=1.44

1000

The photocatalytic activities of TiO2, TiO2–c, RGO–TiO2–X and GO–TiO2–X were evaluated for degradation of MO under UV light irradiation, as shown in Fig. 9(a) and (b). It can be seen from Fig. 9(a) that, after 60 min magnetic stirring and irradiation under UV light, the photodegradation rates of TiO2, RGO–TiO2–0.1, RGO–TiO2–1, RGO– TiO2–5 and RGO–TiO2–10 composites reached to 41.9%, 51%, 59.8%, 68.8% and 39.7%, respectively. The efficiency of photodegradation of MO was proportional to the GO amount in the range of 0.1–5 wt% GO.

280

ID/IG=1.71

×1

(b)

(c)

3.2. Photocatalytic decomposition of MO

(a)

×2/5

(a)

292

450

Binding Energy (eV)

455

460

465

Binding Energy (eV)

Fig. 5. The C 1s XPS spectra of (a) GO, (b) RGO–TiO2–5, and (c) GO–TiO2–5; (d) The Ti 2p XPS spectrum of GO–TiO2–5.

6

470

Ceramics International xx (xxxx) xxxx–xxxx

Transmittance (%)

Y. Zhang et al.

of RGO–TiO2–X and GO–TiO2–X under visible light irradiation were lower than that under UV light irradiation. The photodegradation rates of TiO2, RGO–TiO2–0.1, RGO–TiO2–1, RGO–TiO2–5 and RGO–TiO2– 10 composites were 1.3%, 2.6%, 5.9%, 13.4% and 19.2% after 60 min irradiation, respectively. By comparison, the corresponding photodegradation rates of TiO2–c, GO–TiO2–0.1, GO–TiO2–1, GO–TiO2–5 and GO–TiO2–10 composites were 1.3%, 2.4%, 5.1%, 8% and 11.9%. TiO2 and TiO2–c showed photocatalytic activities because of the dye self-sensitized photocatalytic mechanism. Different from the degradation of MO under UV light irradiation, the photocatalytic activities of both RGO–TiO2–X and GO–TiO2–X increased with increasing GO amount. RGO–TiO2–10 and GO–TiO2–10 exhibited the highest photocatalytic activities, and the RGO–TiO2–X composites showed better photocatalytic activities than GO–TiO2–X when X was the same value. The enhanced photocatalytic activities of both RGO–TiO2–X and GO– TiO2–X composites under visible light irradiation can be explained as the heterogeneous electronic structure of GO which allow GO sheets to act as sensitizers (electron donors) of TiO2 [30]. So, the mechanism of degradation of MO under UV light irradiation was different from visible light irradiation. In the case of UV light irradiation, photogenerated electrons in TiO2 were transferred to GO, thus promoting charge separation. However, electrons was excited by visible light in localized sp2 states of RGO/GO with suitable energy, which would be injected to the conduction band of TiO2, where electrons can be readily scavenged by O2 molecules to produce reactive radicals that would attack pollutant molecules. There was no competition between TiO2 and RGO/GO when the composites were irradiated by visible light. Thus the photocatalytic activities of both RGO–TiO2–X and GO–TiO2–X increased with increasing GO amount. Although RGO–TiO2–10 showed faster weight loss compared to RGO–TiO2–5 in TGA, the photocatalytic degradation of MO suggested that the GO amount of GO–TiO2–10 was higher than that of GO–TiO2–5. The result of calcination of RGO–TiO2–X composites was closer combination between GO and TiO2 particles with smaller size and better crystallization. It is apparent that the photocatalytic degradation efficiencies of MO by RGO–TiO2–X were superior to that by GO–TiO2–X under visible light irradiation, which was due to the more effective charge transfer from RGO to TiO2 than GO to TiO2 [27]. To explore the photocatalytic activities in real environment, the composites with the best photocatalytic efficiency under UV light irradiation were chosen to degradate MO using solar illumination. Fig. 10 shows the corresponding photocatalytic activities and UV dose measured by an UV irradiance meter (UV-A, BNU, China) with a 365 nm sensor. It is clearly observed that the photocatalytic activity of GO–TiO2–5 was higher than that of RGO–TiO2–5 under the same irradiation. Under strong solar irradiation in the summer, the complete

(a) (b)

Ti-O-Ti/Ti-O-C

(c)

C-H skeletal vibration of graphene

4000

3000

2000

1000

Wavnumbere (cm ) -1

Fig. 7. FTIR spectra of (a) TiO2, (b) GO and (c) GO–TiO2–5.

competition between TiO2 and GO was intensified, which leads to the decreased of TiO2 absorbing UV light with high GO amount; (2) the excessive RGO could act as a charge–carrier recombination center and promoted the recombination of electron–hole pairs in the RGO [23,25,34]. From Fig. 9(b), it can be seen that, after calcination, the photocatalytic efficiencies of all the samples were improved further. After 60 min magnetic stirring and irradiation under UV light, the photodegradation rate of TiO2–c, GO–TiO2–0.1, GO–TiO2–1, GO– TiO2–5 and GO–TiO2–10 composites reached to 48.2%, 62.1%, 79.8%, 91.9% and 50.4%, respectively. Compared with RGO–TiO2–X composites, the influence of graphene amount on the photocatalytic efficiencies didn’t changed. GO–TiO2–5 composite showed the best photodegradation rate. Furthermore, compared to TiO2, calcination had more efficient enhancement on photodegradation rate of GO–TiO2–X. For TiO2, only 6.2% enhancement was found. However, the enhancements were 11.1%, 20%, 23.1% and 10.7% corresponding to RGO– TiO2–0.1, RGO–TiO2–1, RGO–TiO2–5 and RGO–TiO2–10, respectively. It is obviously that the photocatalytic efficiency of GO–TiO2–10 was greater than that of TiO2–c, which may be due to the more decreased GO amount. From the XRD, XPS, Raman, FTIR, SBET, SEM and TEM results, it can be deduced that the greater photocatalytic efficiency of GO–TiO2–X than RGO–TiO2–X under UV light irradiation was attributable to the smaller TiO2 particles, better crystallization, lower amounts of organic residues, similar BET surface area and retained GO. As seen from Fig. 9(c) and (d), it provided photocatalytic activity for oxidation of MO using TiO2, TiO2–c, RGO–TiO2–X and GO–TiO2–X under visible light irradiation. As expected, the degradation efficiencies

2.0

GO-TiO2-5

1.6

8

RGO-TiO2-5 1/2

GO-TiO2-1

1.2

(b)

GO-TiO2-10

(a)

TiO2-c

(αhν)

Absorbance (a.u.)

10

GO-TiO2-10

0.8 0.4

GO-TiO2-5 GO-TiO2-1

6

TiO2-c 4 2

0.0 300

400

500

600

700

800

0 2.0

2.5

3.0

3.5

4.0

4.5

5.0

hν (ev)

Wavelength (nm)

Fig. 8. (a) Diffuse reflection spectra of GO–TiO2–10, GO–TiO2–5, RGO–TiO2–5, GO–TiO2–1 and TiO2–c. (b) Corresponding plots of transformed Kubelka–Munk functions versus energy of incident light.

7

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al.

(a)

1.0

(b)

1.0 0.8

C/C0

C/C0

0.8 TiO

0.6

RGO-TiO -0.1

0.6

TiO -c GO-TiO -0.1

0.4

GO-TiO -1

RGO-TiO -1

0.4

RGO-TiO -5

GO-TiO -5

0.2

RGO-TiO -10

GO-TiO -10

0.0

0.2 0

20

40

0

60

20

1.0

(c)

(d)

TiO2-c

C/C0

C/C0

60

1.0

TiO2

0.9

40

Time (min)

Time (min)

RGO-TiO2-0.1

GO-TiO2-0.1

RGO-TiO2-1

GO-TiO2-1

0.9

RGO-TiO2-5

GO-TiO2-5

RGO-TiO2-10

GO-TiO2-10

0.8 0

30

60

0

20

Time (min)

40

60

Time (min)

Fig. 9. Photocatalytic degradation of MO. (a) and (b) under UV light irradiation; (c) and (d) under visible light irradiation.

2000

2000 1.0

(b)

1500

C/C0

0.8 0.6

1000

0.4 500 0.2 0.0 30

60

90

120

150

1500

0.8 0.6

1000

0.4 500 0.2 0.0

0 0

2

2

UV Intensity (μW/cm ) C/C0

(a)

UV Intensity (μW/cm )

1.0

180

0 0

30

60

Time (min)

90

120

150

180

Time (min)

Fig. 10. Photocatalytic degradation of MO under solar light irradiation at different time with (a) RGO–TiO2–5 and (b) GO–TiO2–5. Date: (▼, Δ) June 14, 2015; (▲,▽) June 20, 2015; (■, ○) December 15, 2015; (•, □) December 28, 2015.

GO

TiO2

RGO-TiO2

Hydrothermal reaction

GO-TiO2 Calcination

Scheme 1. Schematic illustration of the formation of hyperfine TiO2 particles by calcination of RGO.

photodegradation of MO for RGO–TiO2–5 was observed after 120 min (Fig. 10(a)), whereas in the presence of the same loading of GO–TiO2– 5, nearly 90 min was needed to achieve complete degradation of MO. The effectiveness of RGO–TiO2–5 and GO–TiO2–5 were clearly observed for the photocatalytic removal of MO under solar illumination in summer. However, the photocatalytic activities of RGO–TiO2–5 and GO–TiO2–5 decreased dramatically under solar irradiation in winter.

The maximum percentage of photodegradation of MO for GO–TiO2–5 was 50.3% after 180 min, while using RGO–TiO2–5, only 36.2% of MO was removed. The higher photocatalytic activity of GO–TiO2–5 can be attributed to more sensitive response to UV light. Though the photocatalytic abilities of TiO2 under visible light irradiation were modified by adding RGO or GO, the photocatalytic efficiencies of TiO2 in real environment still depended on the intensity of UV light in solar 8

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al.

Fig. 11. Recycling test over RGO–TiO2–5 for (a) photocatalytic disinfection of E. coli (visible light) and (b) degradation of MO (UV light).

88.1% of photocatalytic efficiency for decomposition of MO under UV light irradiation. The diminished photocatalytic activities of RGO– TiO2–X indicated the disadvantage of this kind of photocatalysis in real applications. So the method of calcination increases the service time of RGO–TiO2–X composites immensely. In this work, photocatalysts were involved in the form of powders, needing for an additional treatment to separate and recover the photocatalyst particles from the purified retentate. In practice, one method to solve this problem is immobilizing photocatalyst into a membrane support. Ceramic membranes possess excellent chemical, physical, and biological stabilities [52–54]. Most important of all, ceramic materials are resistant to high temperature, which means practicality of calcination. Then synergistic coupling of ceramic membrane filtration and RGO–TiO2–X photocatalytic abilities can overcome the disadvantages mentioned above.

irradiation.

3.3. Mechanisms and utilization From the experimental results, it can be found that the improvement in photocatalytic abilities of GO–TiO2–X was enhanced further after calcination. Hydrothermal treatment is one of the most widely used methods for preparing crystallized TiO2. Usually the as-prepared samples were calcined at high temperature to remove organic residual and improve the crystallinity of TiO2. The organic residual can be efficiently burned out with the calcination temperature of 450 °C. Generally, the procedure of calcination leaded to the growth of TiO2 nanoparticles, which means some negative effects on the photocatalytic activity such as larger crystalline size and lower surface area. The optimal photocatalytic ability of calcinated TiO2 was the results of competition between crystallization and size growth of nano particles. However, the smaller TiO2 particles was found after calcination according to the XRD, TEM and SEM, and residual GO was detected from the results of TEM, FTIR, Raman, XPS and TGA. On the basis of experimental data and discussion above, we postulate a mechanism about calcination of RGO–TiO2–X composites to explain the superior photocatalytic ability of GO–TiO2–X composites, as illustrated in Scheme 1. In the hydrothermal procedure, the GO with functional groups provided reactive and anchoring sites for nucleation and growth of TiO2 nanoparticles. Then fine amorphous particles were coated on GO sheets, which led to selective growth of TiO2. In the subsequent hydrothermal treatment, amorphous TiO2 crystallized and grew directly on graphene, showing strong interaction with the underlying GO sheets. So the growth of TiO2 nanoparticles was confined, suggested by smaller particle size and higher surface area of RGO–TiO2–X than pure TiO2. The high photocatalytic activity of RGO–TiO2–X could be attributed to strong coupling between TiO2 and RGO to facilitate interfacial charge transfer, prolong electron-hole recombination and expand the absorption spectrum. On the other hand, the carbon skeleton of RGO began to oxidize at about 400 °C. Due to the presence of TiO2 nano particles, the burning of RGO was restrained and parts of GO was retained. Meanwhile, organic residual contained in TiO2 nano particles were combusted. Because of the limitation of GO and combustion of organic residual, the size of TiO2 was diminished. The most active hydrothermal photocatalyst (RGO–TiO2–5) was also tested to evaluate its stability in successive photocatalytic disinfection and decomposition. After each photocatalytic experiment the catalyst was washed with deionized water and dried in the oven at 100 °C for 3 h before reused. As presented in Fig. 11, after being used for ten consecutive times, compared to the first cycle, the RGO–TiO2–5 composites remained approximately 87.4% of bacterialcidal ratio for disinfection of E. coli under visible light irradiation, and approximately

4. Conclusion We have demonstrated the successful synthesis of GO–TiO2 hybrid materials using a hydrothermal method together with calcination. TiO2 particles were uniformly dispersed on the surface of RGO sheet, which favored stronger interaction between TiO2 and RGO. GO in the GO– TiO2–X composites was not completely oxidized or removed by calcination. The calcined composites exhibited smaller particles size, large BET surface areas and broad background absorption in the visible region. These characteristics were taken as prerequisites for higher photocatalytic efficiencies. From this work, we can conclude that incorporation of the appropriate amount of GO into TiO2 composites, with careful control of the calcination process in air, will recover the photocatalytic abilities and enlong the service life of TiO2. Acknowledgement This work was supported by High Technology Research and Development Program of China (863 Program, No. 2015AA034702) and National Natural Science Foundation of China (51472180 and 51272176). Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ceramint.2016.10.056. References [1] D.A. Keane, K.G. McGuigan, P. Fernández Ibáñez, et al., , Solar photocatalysis for water disinfection: materials and reactor design, Catal. Sci. Technol. 4 (2014) 1211–1226. [2] O. Carp, C.L. Huisman, A. Reller, Photoinduced reactivity of titanium dioxide, Prog.

9

Ceramics International xx (xxxx) xxxx–xxxx

Y. Zhang et al. Solid State Chem. 32 (2004) 33–177. [3] A. Fujishima, X.T. Zhang, D.A. Tryk, TiO2 photocatalysis and related surface phenomena, Surf. Sci. Rep. 63 (2008) 515–582. [4] T. Froschl, U. Hormann, P. Kubiak, et al., High surface area crystalline titanium dioxide: potential and limits in electrochemical energy storage and catalysis, Chem. Soc. Rev. 41 (2012) 5313–5360. [5] S. Malato, P. Fernandez-Ibanez, M.I. Maldonado, et al., Decontamination and disinfection of water by solar photocatalysis: recent overview and trends, Catal. Today 147 (1) (2009) 1–59. [6] V. Etacheri, C.D. Valentin, J. Schneider, et al., Visible-light activation of TiO2 photocatalysts: advances in theory and experiments, J. Photochem. Photobio. C 25 (2015) 1–29. [7] G.S. Li, D.Q. Zhang, J.C. Yu, A new visible-light photocatalyst: CdS quantum dots embedded mesoporous TiO2, Environ. Sci. Technol. 43 (2009) 7079–7085. [8] J. Gong, P.W. Yang, J. Zhang, Tungsten and nitrogen Co-doped TiO2 electrode sensitized with Fe–chlorophyllin for visible light photoelectrocatalysi, Chem. Eng. J. 209 (2012) 94–101. [9] F. Dong, W. Zhao, Z. Wu, Characterization and photocatalytic activities of C, N and S co-doped TiO2 with 1D nanostructure prepared by the nano-confinement effect, Nanotechnology 19 (36) (2008) 1–10. [10] C. Chusaksri, J. Lomda, T. Saleepochn, et al., Photocatalytic degradation of 3,4dichlorophenylurea in aqueous gold nanoparticles-modified titanium dioxide suspension under simulated solar light, J. Hazard. Mater. 190 (2011) 930–937. [11] F. Sayilkan, M. Asitürk, P. Tatar, et al., Photocatalytic performance of Sn-doped TiO2 nanostructured mono and double layer thin films for Malachite Green dye degradation under UV and vis-lights, J. Hazard. Mater. 144 (2007) 140–146. [12] W. Wang, C.H. Lu, Y.R. Ni, et al., Fabrication of CNTs and GP/AuGP modified TiO2 photocatalyst with two-channel electron conduction path for significantly enhanced photocatalytic activity, Appl. Catal. B: Environ. 129 (2013) 606–613. [13] J.W. Kim, M.S. Damian, W.C. Wonyong, Simultaneous production of hydrogen with the degradation of organic pollutants using TiO2 photocatalyst modified with dual surface components, Energy Environ. Sci. 5 (2012) 7647–7656. [14] D. Mo, D. Ye, Surface study of composite photocatalyst based on plasma modified activated carbon fibers with TiO2, Surf. Coat. Technol. 203 (9) (2009) 1154–1160. [15] M. Inagaki, F. Kojin, B. Tryb, et al., Carbon-coated anatase: the role of the carbon layer for photocatalytic performance, Carbon 43 (2005) 1652–1659. [16] P.F. Du, L.X. Song, J. Xiong, et al., Dye-sensitized solar cells based on anatase TiO2/multi-walled carbon nanotubes composite nanofibers photoanode, Electrochim. Acta 87 (2013) 651–656. [17] Y.L. Pang, A. Zuhairi, Graphene/TiO2 nanocomposites: synthesis, characterization and application in hydrogen evolution from water photocatalytic splitting, Chem. Eng. J. 214 (2013) 129–138. [18] S. Morales-Torres, L.M. Pastrana-Martínez, J.L. Figueiredo, et al., Design of graphene-based TiO2 photocatalysts — a review, Environ. Sci. Pollut. Res. 19 (2012) 3676–3687. [19] E.W. Lee, J.Y. Hong, H.Y. Kang, et al., Synthesis of TiO2 nanorod-decorated graphene sheets and their highly efficient photocatalytic activities under visiblelight irradiation, J. Hazard Mater. 219–220 (2012) 13–18. [20] H. Zhang, X.J. Lv, Y.M. Li, et al., P25-graphene composite as a high performance photocatalyst, ACS Nano 4 (1) (2010) 380–386. [21] X.Y. Zhang, H.P. Li, X.L. Cui, et al., Graphene/TiO2 nanocomposites: synthesis, characterization and application in hydrogen evolution from water photocatalytic splitting, J. Mater. Chem. 20 (2010) 2801–2806. [22] Y.Y. Gao, X.P. Pu, D.F. Zhang, et al., Combustion synthesis of graphene oxide–TiO2 hybrid materials for photodegradation of methyl orange, Carbon 50 (2012) 4093–4101. [23] W.Q. Fan, Q.H. Lai, Q.H. Zhang, et al., Nanocomposites of TiO2 and reduced graphene oxide as efficient photocatalysts for hydrogen evolution, J. Phys. Chem. C 115 (21) (2011) 10694–10701. [24] Q.J. Xiang, J.G. Yu, M. Jaroniec, Enhanced photocatalytic H2-production activity of graphene-modified titania nanosheets, Nanoscale 3 (2011) 3670–3678. [25] D.T. Wang, X. Li, J.F. Chen, et al., Enhanced photoelectrocatalytic activity of reduced graphene oxide/TiO2 composite films for dye degradation, Chem. Eng. J. 198–199 (2012) 547–554. [26] Z. Zhang, W.S. Yang, X.X. Zou, et al., One-pot, solvothermal synthesis of TiO2– graphene composite nanosheets, J. Colloid Interface Sci. 386 (2012) 198–204. [27] A.A. Ismail, R.A. Geioushy, H. Bouzid, et al., TiO2 decoration of graphene layers for highly efficient photocatalyst: impact of calcination at different gas atmosphere on photocatalytic efficiency, Appl. Catal. B: Environ. 129 (2013) 62–70. [28] Y.J. Liu, M. Aizawa, W.Q. Peng, et al., Carbon nanosheet-titania nanocrystal composites from reassembling of exfoliated graphene oxide layers with colloidal titania nanoparticles, J. Solid State Chem. 197 (2013) 329–336. [29] B.T. Liu, Y.J. Huang, Y. Wen, et al., Highly dispersive {001} facets-exposed

[30]

[31]

[32]

[33]

[34]

[35] [36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47] [48] [49] [50] [51] [52] [53]

[54]

10

nanocrystalline TiO2 on high quality graphene as a high performance photocatalyst, J. Mater. Chem. 22 (2012) 7484–7491. L.M. Pastrana-Martínez, S. Morales-Torres, V. Likodimos, et al., Advanced nanostructured photocatalysts based on reduced graphene oxide–TiO2 composites for degradation of diphenhydramine pharmaceutical and methyl orange dye, Appl. Catal. B: Environ. 123–124 (2012) 241–256. N.R. Khalid, E. Ahmed, Z.G. Hong, et al., Nitrogen doped TiO2 nanoparticles decorated on graphene sheets for photocatalysis applications, Curr. Appl. Phys. 12 (2012) 1485–1492. N.R. Khalid, E. Ahmed, Z.L. Hong, et al., Synthesis and photocatalytic properties of visible light responsive La/TiO2-graphene composites, Appl. Surf. Sci. 263 (15) (2012) 254–259. P.F. Wang, Y.H. Ao, C. Wang, et al., Enhanced photoelectrocatalytic activity for dye degradation by graphene-titania composite film electrodes, J. Hazard. Mater. 223– 224 (2012) 79–83. Y.Y. Liang, H.L. Wang, H.S. Casalongue, et al., TiO2 nanocrystals grown on graphene as advanced photocatalytic hybrid materials, Nano Res 3 (10) (2010) 701–705. L.L. Tan, S.P. Chai, A.R. Mohamed, Synthesis and Applications of Graphene-Based TiO2 Photocatalysts, ChemSusChem. 5 (10) (2012) 1868–1882. M. Pal, J.S. Garcia, P. Santiago, et al., Size-controlled synthesis of spherical TiO2 nanoparticles: morphology, crystallization, and phase transition, J. Phys. Chem. C. 111 (2007) 96–102. G.S. Li, L.P. Li, J. Boerio-Goates, et al., High purity anatase TiO2 nanocrystals: near room-temperature synthesis, grain growth kinetics, and surface hydration chemistry, J. Am. Chem. Soc. 127 (2005) 8659–8666. W.G. Wang, J.G. Yu, Q.J. Xiang, et al., Enhanced photocatalytic activity of hierarchical macro/mesoporous TiO2–graphene composites for photodegradation of acetone in air, Appl. Catal. B: Environ. 119–120 (2012) 109–116. Y. Zhou, Y.W. Wu, Y.H. Li, et al., The synthesis of 3D urchin-like TiO2-reduced graphene micro/nano structure composite and its enhanced photocatalytic properties, Ceram. Int. 42 (2016) 12482–12489. X.S. Rong, F.X. Qiu, C. Zhang, et al., Preparation, characterization and photocatalytic application of TiO2–graphene photocatalyst under visible light irradiation, Ceram. Int. 41 (2015) 2502–2511. A.L. Qu, H.L. Xie, X.M. Xu, et al., High quantum yield graphene quantum dots decorated TiO2nanotubes for enhancing photocatalytic activity, Appl. Surf. Sci. 375 (2016) 230–241. R. Rahimi, S. Zargari, A. Yousefi, et al., Visible light photocatalytic disinfection of E. coli with TiO2–graphene nanocomposite sensitized with tetrakis (4-carboxyphenyl) porphyrin, Appl. Surf. Sci. 355 (2015) 1098–1106. L.J. Luo, Y. Yang, A. Zhang, et al., Hydrothermal synthesis of fluorinated anatase TiO2/reduced graphene oxide nanocomposites and their photocatalytic degradation of bisphenol A, Appl. Surf. Sci. 353 (2015) 469–479. P. Calza, C. Hadjicostas, V.A. Sakkas, et al., Photocatalytic transformation of the antipsychotic drug risperidone in aqueous media on reduced graphene oxide—TiO2 composites, Appl. Catal. B: Environ. 183 (2016) 96–106. P. Wang, J. Wang, X.F. Wang, et al., One-step synthesis of easy-recycling TiO2-rGO nanocomposite photocatalysts with enhanced photocatalytic activity, Appl. Catal. B: Environ. 132–133 (2013) 452–459. H.H. Chun, W.K. Jo, Adsorption and photocatalysis of 2-ethyl-1-hexanol over graphene oxide–TiO2 hybrids post-treated under various thermal conditions, Appl. Catal. B: Environ. 180 (2016) 740–750. D.C. Marcano, D.V. Kosynkin, J.M. Berlin, et al., Improved synthesis of graphene oxide, ACS Nano 4 (8) (2010) 4806–4814. K.N. Kudin, B. Ozbas, H.C. Schniepp, et al., Raman spectra of graphite oxide and functionalized graphene sheets, Nano Lett. 8 (1) (2008) 36–41. A.C. Ferrari, J. Robertson, Interpretation of Raman spectra of disordered and amorphous carbon, Phys. Rev. B 61 (2000) 14095–14107. D. Graf, F. Molitor, K. Ensslin, et al., Raman imaging of graphene, Nano Lett. 7 (2) (2007) 238–242. H. Zhang, X. Lv, Y. Li, et al., P25-graphene composite as a high performance photocatalyst, ACS Nano 4 (2009) 380–386. Y. Zhang, Z.R. Tang, X. Fu, et al., TiO2-graphene oxide nanocomposite as advanced photocatalytic materials, ACS Nano 4 (2010) 7303–7314. C.P. Athanasekou, S. Morales-Torres, V. Likodimos, et al., Prototype composite membranes of partially reduced graphene oxide/TiO2 for photocatalytic ultrafiltration water treatment under visible light, Appl. Catal. B: Environ. 158–159 (2014) 361–372. C.P. Athanasekou, N.G. Moustakas, S. Morales-Torres, et al., Ceramic photocatalytic membranes for water filtration under UV and visible light, Appl. Catal. B: Environ. 178 (2015) 12–19.