Graphene oxide–metal oxide nanocomposites: fabrication, characterization and removal of cationic rhodamine B dye

The fabrication and characterization of graphene oxide (GO) nanosheets and their reaction with Fe3O4 and ZrO2 metal oxides to form two nanocomposites, namely graphene oxide–iron oxide (GO–Fe3O4) and graphene oxide–iron oxide–zirconium oxide (GO–Fe3O4@ZrO2), have been examined. The fabricated nanocomposites were examined using different techniques, e.g.transmission electron microscopy, X-ray diffraction, zeta potential measurement and Fourier transform infrared spectroscopy. Compared to GO, the newly fabricated GO–Fe3O4 and GO–Fe3O4@ZrO2 nanocomposites have the advantage of smaller band gaps, which result in increased adsorption capacity and photocatalytic effects. The results also showed the great effect of the examined GO–metal oxide nanocomposites on the decomposition of cationic rhodamine B dye, as indicated by steady-state absorption and fluorescence, time correlated single photon counting and nanosecond laser photolysis techniques. The antibacterial activity of the fabricated GO and GO–metal oxides has been studied against Gram-positive and Gram-negative bacteria.


Introduction
Water pollution has been a vital environmental issue for the last few decades. 1,2 Industrial organic dyes and heavy metals are considered to be the most important sources of water pollution. 1,2 For this purpose, membrane technologies based on nanomaterials have been extensively examined for water puri-cation and desalination over the last few decades. Among the utilized nanomaterials in water treatment, graphene oxide (GO), with its fascinating 2D carbon framework with a honeycomblike structure, has attracted much attention in the last decade for its unique specic surface area, high charge carrier mobility and electron conductivity. [3][4][5] Recently, there has been great interest in fabricating and utilizing novel graphene oxide-metal oxide nanocomposites for environmental remediation by the degradation and elimination of toxic organic contaminants and heavy metals, and for antibacterial applications. [6][7][8] Compared with graphene oxide, graphene oxide-metal oxide nanocomposites show a unique structural morphology and photochemical properties which render them good candidates for water treatment projects. [9][10][11] Among the utilized metal oxides, zero valent iron (ZVI) has been widely used for separating water from harmful heavy metals 12,13 and organic species. [14][15][16][17][18][19][20][21][22] The fabricated graphene oxide-iron oxide nanocomposites showed high efficiency in the removal of tiny concentrations of chromium ions from water and industrial waste water. [23][24][25][26][27][28] In addition, the zirconium oxide (ZrO 2 ) nanoparticles showed unique electrochemical properties when combined with graphene oxide. [29][30][31][32][33][34][35][36] Compared with the widely used TiO 2 , zirconium oxide (ZrO 2 ) is less expensive and insoluble in water. According to the preparation method, ZrO 2 exhibited a band gap ranging from 3 to 5 eV. Such a wide band gap renders ZrO 2 a promising photocatalyst for the production of hydrogen in water decomposition. 37 Taking these unique properties into consideration, we report herein the fabrication and characterization of GO, graphene oxide-iron oxide (GO-Fe 3 O 4 ) and graphene oxide-iron oxidezirconium oxide (GO-Fe 3 O 4 @ZrO 2 ). This combination of graphene oxide with Fe 3 O 4 -ZrO 2 metal oxide and its application in the degradation of organic species is rare in the literature. Photocatalytic studies of the examined nanocomposites on the degradation of cationic rhodamine B dye (RhB) have been examined in detail using TEM, XRD, FTIR, steady-state absorption and uorescence and nanosecond laser ash photolysis techniques.

Characterization techniques
UV-vis absorption and uorescence measurements were taken using a Shimadzu UV-2450 spectrophotometer and a Shimadzu RF-5301PC spectrouorometer, respectively. Picosecond timeresolved uorescence lifetimes were recorded on a Fluo300 (PicoQuant, Germany). Lifetimes were evaluated using FluoFit soware, which was attached to the equipment. Nanosecond transient absorption studies were recorded using a nanosecond laser ash photolysis technique (LP980, Edinburgh Instruments, UK). The instrument was connected with a tunable laser source (NT342B-10, Ekspla). Fourier transform infrared (FT-IR) spectra were obtained using a JASCO spectrometer 4100, using a KBr pellet technique. The X-ray diffraction (XRD) measurements were conducted using a Shimadzu 6000 model with Cu Ka (l ¼ 1.5418Å) as the incident radiation. Transmission electron microscopy (TEM) images were taken using a JEOL 2010 microscope operating under a maximum acceleration voltage of 200 kV. Zeta potential results were obtained using a Brookhaven zeta potential/particle size analyzer.

Photocatalytic activity
The photocatalytic activities of GO, GO-Fe 3 O 4 @ZrO 2 and GO-Fe 3 O 4 nanocomposites were evaluated for the adsorption of dyes, such as rhodamine B, without light and their efficiency for the degradation of rhodamine B (RhB) dye under visible light irradiation (simulator of sunlight; 150 W Xenon lamp, l > 420 nm). 1 Â 10 À4 M RhB dye and 2 mg nanocomposite were dispersed in 10 ml H 2 O. Measurements were performed every 5 min aer exposure to visible light. This experiment was repeated using UV light at 256 nm. OHc radicals were generated more during the reaction, which can result in the rapid degradation of RhB dye molecules. 38 The photo degradation of RhB by graphene oxide-metal oxide nanocomposites was analyzed using steady-state absorbance and uorescence, time-resolved uorescence and nanosecond laser photolysis techniques.

Synthesis of nanocomposites based on graphene oxide (GO)
2.4.1 Synthesis of the GO nanostructure. Water dispersions and solid graphite oxide were prepared from natural graphite powder using a modied Hummers and Offeman's method. 39,40 In a typical reaction, 8 g graphite akes (Sigma Aldrich), 8 g NH 4 NO 3 and 368 ml 98% (w/w) H 2 SO 4 were mixed under stirring in an ice bath for 1 h. Then, 40 g KMnO 4 was slowly added to the mixture in the ice bath until the solution became green. The beaker was placed in a water bath at 35 C and the solution was stirred for about 1 h to form a thick paste. 640 ml highpurity water was then added to the formed paste and stirred at 90 C for 1 h. The formed solution turned brown. With the slow addition of 48 ml H 2 O 2 (30%), the color changed from dark brown to yellow. The solid was ltered, washed with 10% HCl aqueous solution (3.2 L) to remove metal ions and washed with water several times. The resulting graphene oxide was dried at 45 C for 24 h. The crystalline structure of the GO powder was identied using an XRD technique. Renement was carried out from a starting model based on information given in the Inorganic Crystal Structure Database (ICSD   (1)). 46

Characterization of GO, GO-Fe 3 O 4 and GO-Fe 3 O 4 @ZrO 2 nanocomposites
where K is a constant representing the shape factor ($0.9), l is the wavelength of the X-ray source (1.5405 A), b is the full width at half maximum of the diffraction peak and q is the angular position of the peak. The average crystallite sizes of GO-Fe 3 O 4 and GO-Fe 3 O 4 @ZrO 2 were determined to be 8 and 10 nm, respectively. Fig. 2 shows the TEM images of the fabricated GO, GO-Fe 3 O 4 and GO-Fe 3 O 4 @ZrO 2 nanocomposites with different magnications. From the images, GO appeared as nano-sheets, while GO-Fe 3 O 4 and GO-Fe 3 O 4 @ZrO 2 appeared as nano-spherical shapes. The samples were analyzed using EDX with uniform particle morphology (Fig. S2 †). The average size of the observed metal oxides on the surface of graphene oxide was $8 to 10 nm, which is in good agreement with that observed using XRD.
The absorption spectra of fabricated GO-Fe 3 O 4 and GO-Fe 3 O 4 @ZrO 2 were recorded in water, as shown in Fig S3. † The absorption spectra exhibited an absorption peak with a maximum at $228 nm, which was attributed to the p / p* transitions of the aromatic C]C bonds. 47,48 The absorption bands at 390 and 360 nm correspond to GO-Fe 3 O 4 @ZrO 2 and GO-Fe 3 O 4 , respectively. The band gap values of the nanocomposites were determined using eqn (2): 49,50 where a is the absorption coefficient, y is the frequency of light, h is Planck's constant, hy is the photon energy, A is a proportionality constant, E g is the band gap and n ¼ 1/2 for the direct transitions. 51 From the plot of (ahy) 2 versus hy (the insets of Fig. S3 †), the band gaps for GO, GO-Fe 3 O 4 @ZrO 2 and GO-Fe 3 O 4 were found to be 4.00, 3.20 and 3.66 eV, respectively. A zeta potential technique has been used to predict the long term stability of the nanoparticles in solution and to understand the state of the nanoparticle surface. As shown in Fig. 3, the spectra show negatively charged particles for GO (À33), GO-Fe 3 O 4 @ZrO 2 (À41) and GO-Fe 3 O 4 (À52). These negative values are related to the stability of the colloidal dispersions in water.    The observed rate constants for the photocatalytic degradation of RhB with GO nanocomposites were determined using eqn (3): where C o (mg l À1 ) is the initial dye concentration and k obs depends on the initial dye concentration (C o ). 55 The efficiency of the photocatalytic degradation process was determined using eqn (4): where A(RhB) o and A(RhB) t are the absorbance changes of RhB at 554 nm with time, in the dark and under light irradiation, respectively. 57,58 Fig. 7 59 The proposed mechanism for the photodegradation of cationic RhB dye using GO-metal oxide nanocomposites is shown in Scheme 1, [60][61][62][63] where the surface of GO has the ability to receive electrons from the high conduction band (CB) of ZrO 2 , which reacts with oxygen to produce superoxide anion radicals (O 2 c À ) and OHc radicals, leading to the rapid oxidation of the organic molecules.   GO showed a low efficiency that may be explained by the adsorption process of the dye over the GO surface.

Adsorption process of RhB on the surface of GO nanocomposites
Dye molecules were entrapped on the surface of GO and GOnanocomposites in aqueous solution. From Fig. 9 and S7, †  The uorescence measurements showed the same trend as observed for the absorption studies. As seen in Fig. 10 and S8, † the uorescence maximum band of RhB at 579 nm was signicantly decreased in the presence of GO, but not in the presence of GO-Fe 3 O 4 or GO-Fe 3 O 4 @ZrO 2 . The adsorption process could be explained by the dye binding with GO through hydrogen bonding, and electrostatic interactions. 64,65

Laser studies of the photodegradation process of RhB by GO composites
Fluorescence lifetime measurements showed the same trend as observed for the uorescence measurements ( Fig. 11 and S9 †). Upon exciting RhB with 470 nm laser light, the uorescence decay-time prole of the singlet-excited state of RhB ( 1 RhB*) decayed with a monoexponential decay, from which the uorescence lifetime of 1 RhB* was determined to be 1.7 ns. With increasing amounts of GO, the substantial quenching of the uorescence lifetime was considerable and the decay could be tted satisfactorily to a biexponential decay. The fast decaying component had a lifetime of 120 ps (58%), while the slow decaying component had a lifetime of 1.9 ns (42%). The lifetime   Nanosecond transient absorption spectroscopy was used to obtain further insight into the excited state interactions of RhB with GO, GO-Fe 3 O 4 @ZrO 2 and GO-Fe 3 O 4 , to corroborate the observed interaction by both steady-state and time-resolved uorescence techniques. To achieve this, RhB dye was probed with excitation at l ¼ 550 nm in an oxygen-free water solution. The nanosecond transient absorption spectrum of RhB in water was dominated by pronounced bleaching between 540 and 600 nm, which was due to the depletion of the singlet ground state ( Fig. 12 and S10 †). In the case of RhB-GO, it was observed that the singlet state of RhB recovered quickly with increasing   This journal is © The Royal Society of Chemistry 2018 amounts of GO, conrming the quenching of the singlet state of RhB by the GO. In the case of RhB with GO-Fe 3 O 4 @ZrO 2 , the intensity of the ground state bleaching remained almost unchanged with increasing amounts of GO-Fe 3 O 4 @ZrO 2 , suggesting that there was no interaction between RhB and Fe 3 O 4 @ZrO 2 .

Antibacterial activity of GO and nanocomposites
Antibacterial activity was tested against Gram-positive and Gram-negative bacteria using BHI agar plates and the agar diffusion method. The GO, GO-Fe 3 O 4 @ZrO 2 and GO-Fe 3 O 4 samples were evaluated. The resulting antibacterial effect could be rationalized by the diffusion of GO, Fe 3 O 4 @ZrO 2 and GO-Fe 3 O 4 over the agar surface, preventing bacterial growth in the specic area occupied by the nanocomposite. As seen from Fig. 13, we observed only a small zone of inhibition around GO, indicating limited bacterial toxicity against E. coli. 69 In contrast, the GO-Fe 3 O 4 sample showed a signicant inhibitory effect against E. coli. The presence of clear zones on the BHI agar surface proves that the GO-Fe 3 O 4 composite was able to inhibit the growth of E. coli, whereas no antibacterial activity was detected for raw GO and GO-Fe 3 O 4 @ZrO 2 against E. coli. The GO, Fe 3 O 4 @ZrO 2 and GO-Fe 3 O 4 samples showed no antibacterial activity against Steph. The experiment was conducted to characterize bacterial killing with concentration (0.5 mg ml À1 ) and the cellular viability was measured aer 24 h exposure time.

Conclusion
Novel nanocomposites of graphene oxide with iron oxide (GO-Fe 3 O 4 ) and iron oxide-zirconium oxide (GO-Fe 3 O 4 @ZrO 2 ) were fabricated and characterized using XRD, TGA, FTIR, and TEM techniques. From the optical absorption measurements, the energy band gap values were found to be 4.00, 3.66, and 3.20 eV for GO, GO-Fe 3 O 4 and GO-Fe 3 O 4 @ZrO 2 , respectively. All of the steady-state absorption and uorescence, time-resolved uorescence and nanosecond transient absorption spectroscopy results conrmed that RhB is efficiently adsorbed over the surface of graphene oxide ($93%). Different features were observed in the presence of metal oxides (Fe 3

Conflicts of interest
The authors declare no conict of interest.