Facile Synthesis of Graphene Oxide/Titanate Nanotube Composites and Their Application for Cobalt(II) Removal

School of Materials and Chemical Engineering, Anhui Jianzhu University, Hefei, Anhui 230601, China Key Laboratory of Functional Molecule Design and Interface Process, Anhui Jianzhu University, Hefei, Anhui 230022, China School of Environment and Energy Engineering, Anhui Jianzhu University, Hefei, Anhui 230601, China Research Institute of Wood Industry, Chinese Academy of Forestry, Beijing 100091, China


Introduction
During the last decade, rational disposal and management of the radioactive species have drawn great attention because a large amount of radioactive species has been discharged through unexpected nuclear power plant incidents [1][2][3]. Among various radioactive elements, 60 Co is widely used in radiotherapy, industrial radiography, sterilization, density measurements, and electroplating [4][5][6]. 60 Co also can be found in nuclear wastewaters as a result of the neutron bombardment of stable cobalt isotopes in the metal structures of nuclear power plants [7,8]. Upon exposure, skin burns, liver/kidney/bone cancers, and acute radiation sickness leading to death (EPA) can occur [9]. Hence, its stringent reduction to permissible levels in water should be performed.
Various technologies have been developed to purify the Co(II) ions contaminated with wastewater, such as precipitation [10], membrane separation [11], electrolysis [12], and adsorption [13]. It is well-accepted that adsorption is the wide-used Co(II) ions removal method due to the advantages of low cost, high flexibility, and easy operation [14]. However, most traditional adsorbents, such as active carbon [15], diatomite [16], and clay [17], often suffer from the low adsorption ability towards Co(II) ions. us, developing novel Co(II) adsorbents with high performance is still a big challenge.
Recently, titanate nanomaterials are widely studied as the promising adsorbent for metal ions removal [18]. Due to the great number of the exchangeable Na + /H + within the stable layers, titanate nanomaterials exhibit large adsorption performance to the heavy metal ions [19]. However, due to the small size, the further application of those titanate nanoadsorbents is still limited by the weak mechanical strength, strong tendency to agglomerate, and difficulty in separation. Graphene oxide (GO) is another well-studied adsorbent [20,21]. Owning to its large surface areas and oxygenated functional groups, GO and its derivatives are of great interest in fabricating hybrid adsorption. Many GObased nanoadsorbents with superior adsorption performances have been synthesized for water purification, such as TiO 2 /GO, Ag/GO, and Fe 3 O 4 /GO [6,22,23]. Although titanate and GO nanomaterials have been extensively studied for the adsorption of heavy metal ions from aqueous solutions, the reports mentioned that the titanate and GO hybrid nanoabsorbents for radioactive elements removal are still limited. In this work, the titanate nanotube/graphene oxide nanocomposites have been synthesized through a facile hydrothermal method, and their adsorption performance towards Co(II) ions also has been studied. In the composites, the titanate nanotubes are closely attached to the graphene sheet structure and overlap each other. is hierarchical morphology endows the GTNT composites with not only the high specific surface area of 236.9 m 2 /g but also abundant porous structure, both of which are benefits for Co(II) adsorption. e adsorption capacity of the GTNT composites reaches 211.1 mg/g. It is believed that the superior adsorption capacity of the GTNT composites makes it a potential candidate for radioactive wastewater treatment in future practical application.

Experimental
2.1. Synthesis. All reagents were commercially available from Sinopharm Chemical Reagent Co., Ltd. (China) with analytical grade and were used without further purification.
GO was synthesized through the improved Hummers method. Typically, 1 g of graphite flake, 1 g of NaNO 3 , and 45 mL of 98% sulfuric acid were added into a 500 mL flask, and 6 g of KMnO 4 was slowly added under the ice bath. e mixture was stirred for 2 hours. en, the flask was heated to 60°C and stirred for 2 hours. Next, 200 mL of deionized water was added into the reactant and stirred at 90°C for 5 hours. Finally, 10 mL of 30% hydrogen peroxide was added into the flask. e obtained solution was filtered and washed with 5% hydrochloric acid solution and deionized water for 5 times. After ultrasonic treatment for 2 hours, GO can be obtained.
Next, 40 mg graphene oxide, 2 g P25 powder, and 32 g NaOH were added into 80 mL deionized water, stirred 30 min, and then transferred to an autoclave. Put the autoclave into the oven, set the temperature to 200°C, and keep for 12 hours. e product was washed with dilute acid and ethanol and then dried in a vacuum oven at 60°C for 12 hours. In this way, the titanate nanotube/GO composites can be obtained.

Characterization.
e morphologies and structures of the samples were examined by field emission scanning electron microscopy (FE-SEM; FEI Sirion-200), high-resolution transmission electron microscopy (HRTEM; JEM-2010), and X-ray diffraction (XRD, Philips X'pert PRO) with Cu Ka radiation. e specific surface area and gas adsorption isotherms of the products were tested on a Coulter Omnisorp 100CX Brunauer-Emmett-Teller (BET). e concentration of the Co(II) ions was measured using the atomic fluorescence spectrometer (TAS-990).

Adsorption Experiments.
e sorption of Co(II) ions on the titanate nanotube/GO composites was studied using the batch technique. e adsorption experiments were carried out in 15 mL polypropylene flasks containing 10 mg of adsorbent and 10 mL of Co(II) solution with various concentration, and then, the flasks were shaken at 150 rpm in a shaker at 25°C for 24 h. e pH value was adjusted using negligible volumes of HCl and/or NaOH.
After the adsorption experiments, the adsorbent was separated from the solution by centrifugation, and the residual Co(II) concentration in solution was measured by ICP. e residual adsorbents were centrifuged, washed with water, and dried in the oven at 60°C.
In the kinetics study, the initial Co(II) concentration was 20 mg/L, the adsorbents dose was 1.0 mg/L, and the total suspension volume was 100 mL. e samples were placed on a shaker for stirring. At predetermined time intervals, stirring were interrupted, while 6 mL of supernatant solutions was pipetted and centrifuged for the determination of the remaining Co(II) concentrations. In this case, the centrifugal speed was 5000 rpm. e amount of Co(II) adsorbed on the adsorbents was calculated from the initial concentration (C 0 ) and the final concentration (C e ). e adsorption percentage (%) was calculated as e zeta potential test was carried out at Malvern Zetasizer Nano ZS90. Typically, GTNT composites were first ultrasonic dispersed in water. en, almost 3 mL of the suspension was injected into the test cell. After test, the ζ potential of the GTNT composites can be obtained. Figure 1 is a SEM photograph of the GO/titanate nanotubes composites (GTNT) obtained by the hydrothermal method. As shown in Figures 1(a) and 1(b), titanate nanotubes and graphene are intertwined to form a loose composite structure. Figure 1(c) shows the edge of a GTNT cluster, where the flake structure and curly edge of GO can be seen clearly. Figure 1(d) shows the enlarged view of the white box in Figure 1(c), in which the modification of titanate nanotubes on GO nanosheet can be clearly seen. Moreover, the titanate nanotubes not only modify on the surface of GO nanosheet but also exist between the layers of graphene. For further characterization, the structure of the GTNT composites is also investigates by TEM, and the results are shown in Figure 2. Figure 2(a) shows the low-magnification TEM image of the GO/titanate nanotubes composites. It can be seen that a huge number of titanate nanotubes are closely attached to the graphene sheet structure and overlap each other. From the magnified TEM images (Figures 2(b) and 2(c)), the nanotube structure of titanate can be confirmed. At the edge of the individual GO nanosheet, the uncovered graphene and the titanate nanotubes protruding from the edge are clearly visible. is result is well consistent with that obtained by SEM. Figure 2(d) shows the high-resolution TEM image of the GTNT composite. From which, it can be seen that the titanate nanotubes with a multiwall structure and a length of several hundreds of nanometers can be characterized, indicating that those nanotubes are formed by the bending of (100) plane titanate nanosheets. e smooth and ultrathin GO nanosheet also can be observed. Figure 3 shows the XRD pattern of GTNT composites. ose diffraction peaks located at 24.1°, 28.1°, 48.1°, and 60.6°c an be indexed to (220), (310), (331), and (521) planes of the Na 0.5 TiO 2 O 4 (JCPDS 80-1282). To further investigate the thermal behavior of GTNT composites, TGA/DTA measurement was conducted and is shown in Figure 4. e TGA curve shows weight loss of about 17.3% from 50°C to 600°C, and the corresponding DTA curve displays two endothermic peaks at about 80°C and 500°C. e DTA curve does not show any exothermic peak, which indicates that the product is pure crystalline phase without any other amorphous phase. e weight percentage loss at 150°C is about 10.3%, which is probably due to the removal of water molecules bound to the surfaces of the sample. e second peak at about 500°C is weight loss of 5.1%, which is probably due to the removal of water molecules confined in the interlayer spaces of the GTNT composites during the increase in temperature. e remaining weight loss of 2.2% up to at 600°C is probably due to the oxidation of carbon.

Characterization of the GTNT Composites.
In order to better understand the formation process of the GTNT composites, the influence of reaction time on the morphology of the sample was studied. As shown in Advances in Condensed Matter Physics the composites of titanate nanotubes and GO nanosheets are completely formed. In addition, the effect of NaOH concentration on the morphology of the samples also has been studied. Figure 6(a) shows the morphology of the sample synthesized without NaOH. After 24 h reaction, P25 has been completely modified to the surface of graphene. However, the granular structure of the nanoparticle does not change. When the concentration of NaOH increased to 0.1 M, the obtained samples remain the composite of GO nanosheets and nanoparticles ( Figure 6(b)). Continue to increase the NaOH concentration to 1 M; some products with nanoplate-like structure can be observed (Figure 6(c)). When the concentration of NaOH reaches 4 M, on the surface of GO nanosheets, the structures of modified materials totally change to nanotubes (Figure 6(d)). As the reaction continued, the strong alkali condition would destroy the octahedral crystal structure of adsorbed TiO 2   nanoparticles. e free [TiO 6 ] octahedron would bond together by OH bonds formed between titanium ions, which result in its growth along the (100) direction, and form a two-dimensional crystal layered structure on the GO surface [24]. In order to saturate the dangling bond and reduce the surface energy, those nanolayers would roll up and finally formed the titanate nanotubes/GO composites. Advances in Condensed Matter Physics N 2 adsorption/desorption isotherms are also employed to examine the microstructure of the GTNT composites. Figures 7(a) and 7(b) show the nitrogen adsorptiondesorption isotherms and pore size distribution curves of the GTNT composites. e corresponding results of P25 powders are also present as the comparison. According to IUPAC classification, the GTNT composites exhibited a type IV isotherm of a type H3 hysteresis loop, typical of mesoporous materials [25]. e pore size distribution of the GTNT composites shows a bimodal size distribution, with the smaller sizes peaking at 1.5 nm and the larger sizes at 30 nm, which is consist with the SEM and TEM results. e smaller mesopores can be related to the hollow structure of the titanate nanotubes, while larger mesopores would correspond to the spacing between of GO nanosheets and the gap between nanotubes. P25 also has two pore structures of 100 nm and 2 nm, which may be caused by the agglomeration of P25. Although the GTNT composites and P25 have similar size distribution, the BET specific surface area differs a lot. e BET specific surface area of the GTNT composites and P25 is 236.9 and 46.0 m 2 /g, respectively. Due to the advantages of abundant porous structure and high surface area, the GTNT composites would expect with excellent adsorption properties.

Adsorption Properties of the GTNT Composites.
Kinetics experiment was first employed to determine the removal rate of Co(II) ions from water at initial concentration of 20 mg/L. e adsorption kinetics of Co(II) on the GTNT composites is shown in Figure 8(a). As shown, in the first 10 min, the adsorption of Co(II) on the GTNT composites happens rapidly, where almost 95% of the Co(II) ions can be adsorbed. en, adsorption capacity gradually increased till it reached the equilibrium, where more than 98% of the contained Co(II) ions can be adsorbed. To further quantify the changes of Co(II) ions adsorption with time on the GTNT composites, the pseudo-first-order model (equation 1 [26]) and pseudo-second-order model (equation 2 [27]) are applied to simulate the adsorption kinetics process.
where q e and q t are the amounts of adsorbed fluoride (mg/g) at equilibrium and at any time t (min), respectively. k 1 (min −1 ) and k 2 (g/(mg·min)) are the equilibrium rate constants for pseudo-first-order and pseudo-second-order sorptions, respectively. Table 1 provides the fitting results of the experimental data fitted into pseudo-first-order and pseudo-secondorder models, respectively. Figure 8(b) shows the pseudosecond-order fitting of the experimental data. It can be seen that the regression coefficients (R 2 ) of the pseudo-  Advances in Condensed Matter Physics first-order model and pseudo-second-order model are 0.999 and 0.971, respectively. e higher R 2 value indicates that the adsorption kinetic fitting of the GTNT composites favors the pseudo-second-order Co(II) removal process. e calculated equilibrium adsorption capacity (q e , cal) is 19.85 when the initial Co(II) concentrations is 20 mg/L, which is in agreement with experimental equilibrium adsorption capacities (q e , exp). e adsorption isotherm of Co(II) adsorption on the GTNT composites is also investigated, and the results are shown in Figure 9. Two commonly used isotherm models, namely, Langmuir and Freundlich were used in the present study for the data analysis [28,29]. Table 2 provides isotherm parameters computed from the Langmuir and Freundlich models.
e higher R 2 value suggests that the Langmuir model presents better fitting for the adsorption data than the Freundlich model, implying that the adsorption of Co(II) on the GTNT composites is monolayer adsorption. e adsorption capacity of Co(II) on the GTNT composites is 211.1 mg/g when the equilibrium concentration is 88.9 mg/L. e comparison of the Co(II) adsorption capacity of different adsorbents is given in Table 3. It can be seen that, compared with other adsorbents, the as-prepared GTNT composites exhibit the superior adsorption capacity. e good performance in Co(II) adsorption can be attributed to the benefit of the relative high surface area and hierarchical porous structure. e porous structure and relative high surface area can provide more active sites to trap Co(II) ions. e hierarchical structure not only provides sufficient diffusion channels for Co(II) ions but also avoids the

Equations
First-order kinetics Second-order kinetics ln(q e − q t ) � ln q e − k 1 t t/q t � 1/k 2 q 2 e + t/q e C o (mg/L) q e,exp (mg/g) k 1 (1/min) q e,cal (mg/g) R 2 k 2 (g/(mg·min)) q e,cal (mg/g) e pH value of the wastewater is an important parameter in adsorption treatment technology, which would largely influence the adsorption processes. us, the effect of pH on Co(II) adsorption by the GTNT composites is studied in the pH ranging of 1.0-7.0, and results are shown in Figure 10. At pH � 7, the equilibrium uptake of Co(II) ions is 243.2 mg/g. e adsorption capacity of Co(II) on the GTNT composites quickly decreases with the decrease of pH value. e ζ potential of the GTNT composites at different pH is shown in Figure 11. With the decrease of pH, the zeta potential of the GTNT composites continuously increases. e lower the pH value, the higher the zeta potential of the GTNT composites. In the pH range of 4.0-7.0, the GTNT composites are negative charged, while, when the pH value is lower than 4.0, the GTNT composites are positively charged.    Co(II) ion is a cation; thus, the higher negative charged on the surface of the GTNT composites, the greater its electrostatic attraction to Co(II) ions. us, the GTNT composites always exhibit better Co(II) adsorption performance under higher pH values. e recycling experiment of GTNT composites was also investigated. e GTNT composites were recovered after the batch test and washed with dilute hydrochloric acid solution and then calcinated at 500°C for 2 h to get rid of adsorbed Co(II) ions. Figure 12 shows the Co(II) ions removal efficiency of the GTNT composites after 5 adsorption-desorption cycles. It can be found that, with the increase of reuse times, the performance of the GTNT composites decreases slightly; almost 87% Co(II) ions can be removal by GTNT composites after 5 adsorptiondesorption cycles. is decrease might be due to the blocking of some of the adsorption sites by Co(II) ions.
In common condition, many kinds of polluted metal ions often coexist in the wastewater; thus, the interference of coexisting ions on Co(II) adsorption performance of the GTNT composites are studied. As shown in Figure 13, it can be found that Na + , K + , and Mg 2+ ions have shown a negligible effect on the Co(II) adsorption of the GTNT composites. However, when Cu 2+ , Cd 2+ , and Pb 2+ ions exist in the system, the interference can be clearly observed on the Co(II) adsorption of the GTNT composites. e greater the ionic strength, the more significant the effect, which may be attributed to the ions' competition for active sites.

Conclusion
In summary, a novel GO/titanate nanotubes (GTNT) composite was synthesized through a facile alkaline hydrothermal method. e specific surface area of the GTNT composites is 236.9 m 2 /g, and the large surface area is beneficial to removal Co(II) from the radioactive wastewater. e batch adsorption experiments demonstrated that the GTNT composites have high adsorption capacity and rapid kinetics for Co(II) adsorption (10 min for equilibrium). e adsorption capacity of the GTNT composites is 211.1 mg/g. e adsorption kinetics of the GTNT composites fits well to the pseudo-second-order model, while the adsorption isotherm of which fits well to the Langmuir model. e adsorption performance of Co(II) ions on the GTNT composites has a great relationship with the pH value, and in the pH range of 1-7, the Co(II) adsorption capacity of the GTNT composites greatly increases with the increase of the pH value. In addition, the effect of coexisting anions on fluoride removal is also investigated. Na + , K + , and Mg 2+ ions have shown a negligible effect on the Co(II) adsorption efficiency of the GTNT composites. However, the existences of Cu 2+ , Cd 2+ , and Pb 2+ ions would clearly have an effect on the Co(II) adsorption of the GTNT composites. e adsorption mechanism is also discussed.
us, the GTNT composites can be considered as the potential functional material for removing the radioactive metals contained wastewater.

Data Availability
No data were used to support this study.  Advances in Condensed Matter Physics 9