CHEMICAL ENGINEERING TRANSACTIONS VOL. 78, 2020 A publication of The Italian Association of Chemical Engineering Online at www.cetjournal.it Guest Editors: Jeng Shiun Lim, Nor Alafiza Yunus, Jiří Jaromír Klemeš Copyright © 2020, AIDIC Servizi S.r.l. ISBN 978-88-95608-76-1; ISSN 2283-9216 Fabrication of Magnetic Iron Oxide/Graphene Oxide Nanocomposites for Removal of Lead Ions from Water Lu Thi Mong Thya, Pham Mai Cuongb, Tran Hoang Tua, Hoang Minh Namb, Nguyen Huu Hieua, Mai Thanh Phongb,* aVNU-HCM Key Laboratory of Chemical Engineering and Petroleum Processing (CEPP Lab) bFaculty of Chemical Engineering Ho Chi Minh City University of Technology - Vietnam National University (HCMUT - VNU) 268 Ly Thuong Kiet Street, Ward 14, District 10, Ho Chi Minh City, Vietnam mtphong@hcmut.edu.vn In this study, magnetic iron oxide/graphene oxide (Fe3O4/GO) nanocomposites were prepared by in situ method. It was found that the suitable ratio of Fe3O4/GO for lead ions (Pb2+) removal was 2:1 (FGO2). The adsorption conditions of Pb2+ onto FGO2 were studied using response surface methodology (RSM) with Box- Behnken design. In RSM model, the interactive effects of critical variables including contact time, initial concentration, and pH on the adsorption capacity were investigated. The adsorption process followed the Langmuir isotherm model. Based on these results, FGO2 could be used as an efficient adsorbent for removal of Pb2+ from water. 1. Introduction The increase of heavy metal ions concentration into the water which has caused hazardous impact on human health and ecosystems. This issue is especially serious in the case of lead pollution because of the high toxicity and the bioaccumulation ability in both human bodies and in other living organisms (Kumar et al., 2014). Even at very low concentrations, lead (II) (Pb2+) ions can significantly damage the human brain, kidneys, and circulatory system, leading to serious diseases such as anemia, mental disorders, and cancers (Zhao et al., 2011). Thus, the removal of Pb2+ ions from water has become an urgent environmental problem. Adsorption turns out to be one of the most effective solutions due to its simple treatment process, low cost, and highly efficiency (Thy et al., 2019). In recent years, graphene-based materials have attracted interest of researchers. Graphene oxide (GO), as a derivatives of graphene, has oxygen-containing groups on the surface such as hydroxyl, carboxyl, carbonyl, and epoxide groups, which could be used as anchoring sites for metal ion complexation and showed good prospect in water treatment (Marina et al., 2019). However, GO can be well dispersed which causes trouble in obtaining a complete recovery (Yu et al., 2019). To overcome these disadvantages, the magnetic graphene oxide-based nanocomposites have been researched. The magnetic iron oxide/graphene oxide (Fe3O4/GO) has the unique advantages of high adsorption capacity, chemical stability, reusability, and easy magnetic separation to make it a potential adsorbent for removal of heavy metal ions from water (Wang et al., 2012). The purpose of this study was to investigate the adsorption ability of Fe3O4/GO nanocomposites for Pb2+ ions. Further, response surface methodology (RSM) with Box-Behnken design (BBD) was used to study the simultaneous effects of factors on the adsorption capacity. 2. Experimental 2.1 Synthesis of Fe3O4/GO nanocomposites GO was prepared from graphite powder by improved Hummers’ method (Marcano et al., 2010). Fe3O4 nanoparticles (Fe3O4 NPs) were prepared by in-situ method. Fe3O4/GO nanocomposites were prepared by in- situ method (Thy et al., 2019). Briefly, 10 mL of FeCl3.6H2O and FeCl2.4H2O solution was added slowly into 50 DOI: 10.3303/CET2078047 Paper Received: 15/04/2019; Revised: 03/08/2019; Accepted: 15/11/2019 Please cite this article as: Thy L.T.M., Cuong P.M., Tu T.H., Nam H.M., Hieu N.H., Phong M.T., 2020, Fabrication of Magnetic Iron Oxide/Graphene Oxide Nanocomposites for Removal of Lead Ions from Water, Chemical Engineering Transactions, 78, 277-282 DOI:10.3303/CET2078047 277 mailto:mtphong@hcmut.edu.vn https://www.sciencedirect.com/topics/earth-and-planetary-sciences/ecosystem-health https://www.sciencedirect.com/topics/earth-and-planetary-sciences/ecosystem-health mL of GO suspension (6 mg/mL). The mixture was stirred and heated to 80oC at pH 10 for 2 h. The black precipitation was separated using a magnet, followed by washing with ethanol. Finally, the nanocomposites were obtained after drying at 50 °C for 24 h. The nanocomposites with different Fe3O4:GO mass ratios of 2:1, 3:1, and 4:1 were marked as FGO2, FGO3, and FGO4. The characterization of nanocomposites was investigated by X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), transmission electron microscopy (TEM), Brunauer-Emmett-Teller (BET) surface area, and vibrating sample magnetometer (VSM). 2.2 Adsorption experiments Batch experiments were carried by adding 20 mg of Fe3O4/GO adsorbent into 20 mL of Pb2+ solution at room temperature. After adsorption, Fe3O4/GO was separated from aqueous solution by an external magnet. The residual Pb2+ concentration in solution was determined by inductively coupled plasma mass spectrometry (ICP-MS 7500, Agilent, USA). RSM with Box–Behnken design was used to study the significance of independent variables and thereby determining optimal conditions for adsorption process. The regression analysis was performed to estimate the response function as a second order polynomial shown in Eq(1): Y = β0 + ∑ βixi 3 i=1 + ∑ βiixi 23 i=1 + ∑ ∑ βijxixj 3 j=1 3 i=1 (1) where Y is the predicted response; the parameter βo, βi, βii, and βij are the regression coefficients for intercept, linear effect, double interaction, and quadratic effect; xi and xj are the independent variables (Chen et al., 2011). Contact time (A), initial Pb2+ concentration (B), and pH (C) were taken as three input variables and the adsorption capacity (Y) of Fe3O4/GO towards Pb2+ was taken as the response. The range and levels of the independent variables are shown in Table 1. The statistical parameters were assessed from the analysis of variances (ANOVA) using Design Expert v.11. Table 1: Independent variables matrix and their encoded levels No. Independent variables Code Levels -1 0 +1 1 Contact time (min) A 30 60 90 2 Initial Pb2+ concentration (ppm) B 200 300 400 3 pH C 5.1 5.9 6.7 3. Results and discussion 3.1 Characterization of Fe3O4/GO nanocomposites The XRD patterns of graphite, GO, Fe3O4, and Fe3O4/GO nanocomposites are presented in Figure 1a and 1b. Figure 1: XRD patterns of (a) graphite and GO; (b) Fe3O4 NPs, and Fe3O4/GO nanocomposites, (c) FTIR spectra of GO, Fe3O4 NPs, and Fe3O4/GO nanocomposites In case of GO, the diffraction peak at 2 = 12.5o (0.8 nm) showed the formation of oxygen-containing groups on the surface of graphene sheets. In Fe3O4 NPs and Fe3O4/GO nanocomposites, several peaks at 2θ = 30.04, 35.53, 43.27, 57.31, and 62.62o were assigned to the (220), (311), (400), (511), and (440) planes (Omidinia et al., 2013). These peaks well fitted with the crystalline characteristics of Fe3O4 (JCPDS No.65-3107) (Chen et al., 2011). The disappearance of GO peak indicates that the Fe3O4 NPs were inserted between GO sheets, increasing the interlayer distance between GO sheets. Figure 1c shows that FTIR spectrum of GO contained these peaks of oxygen-containing groups at 3,391, 1,734.5, 1,628, 1,351, and 1,098 cm-1, corresponding to the stretching vibrations of hydroxyl (O-H), carbonyl (c) (a) (b) 278 https://serc.carleton.edu/research_education/geochemsheets/techniques/XRD.html https://en.wikipedia.org/wiki/Fourier_transform_infrared_spectroscopy https://en.wikipedia.org/wiki/Fourier_transform_infrared_spectroscopy https://en.wikipedia.org/wiki/Transmission_electron_microscopy (C=O), carboxylic (COOH), epoxide (-O-), and alkoxy (C-O-C) (Shahriary and Athawale 2014, 2011). FTIR spectra of Fe3O4 and Fe3O4/GO nanocomposites showed strong fluctuations at 583 cm-1 corresponding toFe- O bonds in tetrahedral and octahedral sites. This result was consistent with XRD results, showed the formation of Fe3O4 in the structure of nanocomposites. The decrease in the intensity of oscillations at 1725 and 1599 cm-1 in Fe3O4/GO are showed that the decoration of Fe2+, Fe3+ ions associated with high electronegativity positions on the GO surface (OH, C=O, and COOH groups), in accordance with the synthetic mechanism of the material (Kumar et al., 2014). From the above results, it can be seen that Fe3O4 NPs were successfully linked on GO surface. As shown in Figure 2a, the surface of GO film was wrinkled due to the crumpling and scrolling of sheets (Metin et al., 2014). For Fe3O4/GO nanocomposites, the Fe3O4 NPs with size ranging from approximately 10-15 nm were anchored on the surface of GO sheets. The decrease in the agglomeration of Fe3O4 NPs showed the efficient interactions between Fe3O4 NPs and GO during the synthesis procedure (Trinh et al., 2018). The FGO2 had a particle size smaller, and distribution more uniform than FGO3 and FGO4. When increasing Fe3O4:GO ratio, the agglomeration regions and the size of Fe3O4 NPs were enhanced, resulting in reducing BET specific surface area as well as adsorption sites of the materials. Figure 2: TEM images of (a)GO, (b) Fe3O4 NPs, (c) FGO2,(d) FGO3, (e) FGO4; (f) VSM data of FGO2, FGO3, and FGO4 The specific surface area of FGO nanocomposites were determined to be 190.70 - 200.40 m2/g, which are higher than of graphene-based materials as shown in Table 2. These results could be explained by Fe3O4 NPs were formed and anchored on the surface of GO, decreasing the aggregation of Fe3O4 and stacking of GO sheets (Yang et al., 2009). However, the ratio Fe3O4:GO enhances, increasing the agglomeration regions of Fe3O4 NPs, reducing the surface area and adsorption sites (Li et al., 2013). Table 2: BET specific surface areas of Fe3O4/GO nanocomposites and graphene-based materials Materials BET surface areas (m2/g) References GO 130.20 In this study Fe3O4 94.40 In this study FGO2 200.40 In this study FGO3 199.80 In this study FGO4 190.70 In this study FeOOH/GO 202.60 (Kuang et al., 2017) CoFe2O4/Graphene 126.36 (Santhosh et al., 2015) NiFe2O4/Graphene 57.11 (Santhosh et al., 2015) The saturation magnetization (Ms) values of FGO2, FGO3, and FGO4 were 21.79, 18.30, and 18.54 emu/g, which were high enough to be separated by a magnet (> 16 emu/g) as shown in Figure 2f (Bai et al., 2015). These values of Fe3O4/GO nanocomposites were lower than that of the bulk Fe3O4 (92 emu/g) due to the Fe3O4 NPs were decorated on GO surface, decreasing in the size of nanoparticles. Ms values of Fe3O4/GO nanocomposites decreased from FGO2 to FGO4 due to the accumulation of Fe3O4 NPs, leading to an increase in particle size, reducing the magnetic value of the material (Kuang et al., 2017). (a) (b) (c) (d) (e) (f) 279 3.2 Adsorption ability of Fe3O4/GO nanocomposites The Pb2+ adsorption capacities of FGO2, FGO3, and FGO4 were determined to be 62.35, 41.66, and 39.41 mg/g. The FGO2 has higher adsorption capacity than other ratios, which was consistent with the results of analyzing characteristic. The results of BET, TEM, and VSM showed that the FGO3 and FGO4 had the agglomeration regions of Fe3O4 NPs on the Fe3O4/GO surface, reducing the surface area and adsorption sites of materials. The increase of Fe3O:GO ratio was also reduced the oxygen-containing groups on Fe3O4/GO surface, the adsorption capacities of the materials were decreased. Therefore, FGO2 was selected to study the adsorption capacity for Pb2+. 3.3 Effects of adsorption variables on the adsorption capacity of FGO2 The statistical parameters from quadratic model were analysed using ANOVA to determine the suitability and reliability of the model as shown in Table 3. The F-value of model was calculated to be 48.51 indicating the model is significant. The insignificant values for of lack of fit (0.0868) and a low probability value of p-values (0.0002) were determined a very high significance for the regression model showing that the model consider to be statistically significant. The effect variables and their interactions on the responses were studied by using the comparison and evaluation with the p-values. The p-values of Co and pH are smaller than 0.0001, thus these variables are highly significant factors. The p-values of AC, BC, and A2 from ANOVA analysis are higher than 0.10, these coefficients are not statistically significant in model. The other variables with p-values are lower than 0.05, are significant parameters on model term. Therefore, the final equation is presented in Eq(2). q = 173,082+1,404t+0,796Co-99,892pH-0,00159tCo-0,000833Co 2 +10,645pH 2 (2) Table 3: ANOVA analysis of the Pb2+ adsorption of FGO2 Source Sum of squares Degree of freedom Mean square F-value p - value Model 4503.28 9 500.36 48.55 0.0002 significant A 249.20 1 249.20 24.18 0.0044 B 2024.71 1 2024.71 196.45 < 0.0001 C 1647.95 1 1647.95 159.89 < 0.0001 AB 91.49 1 91.49 8.88 0.0308 AC 20.52 1 20.52 1.99 0.2173 BC 1.28 1 1.28 0.1239 0.7392 A² 6.89 1 6.89 0.6687 0.4507 B² 255.97 1 255.97 24.84 0.0042 C² 171.30 1 171.30 16.62 0.0096 Residual 51.53 5 10.31 Lack of Fit 48.51 3 16.17 10.69 0.0868 not significant Pure Error 3.03 2 1.51 Correlation Total 4,554.81 14 The graph of actual versus and data reveal the well-fitting with high correlation coefficients (R2 of 0.9987 and adjusted R2 of 0.968) suggesting a high adequacy of the models. The deviation between experimental and predicted values was low, which was inferred from the CV value (2.66 %). The difference of R2 and the adjusted R2 is lower than 0.2 and the value of the adequate precision is much larger than 4 (24.476), the selected model is sufficiently accurate to predict the lead ion adsorption capacity of FGO2. The individual effect of variables on the adsorption capacity for FGO2 was studied as shown in Figure 3a. The curve of B has largest slope, indicating the initial Pb2+ concentration had a greater effect on the adsorption capacity than time and initial concentration. This result is consistent with the higher F-value of initial Pb2+ concentration with respect to other process variables. Three-dimensional (3D) response surface plot has been used to study the combined effects of all the process variables on adsorption capacity of FGO2. The adsorption capacity for the combined effect of variables at the RSM derived optimum conditions (contact time of 50 min, initial Pb2+ concentration of 380 ppm, and pH of 6.7) was obtained at 152.57 mg/g, which was closed to the predicted values (150.69 mg/g), indicating the suitability and accuracy of the suggested models. 280 Figure 3: (a) Interaction effects of variables; 3D response surface plots for interactive effects of (b) contact time - initial Pb2+ concentration; (c) contact time - pH; (d) initial Pb2+ concentration – pHAdsorption isotherm The adsorption data were analyzed by the Langmuir, Freundlich, and Temkin isotherm models and the results were presented in Table 4. The adsorption process was well-fitted to the Langmuir isotherm model with R2 of 0.989 and qmax of 169.49 mg/g. The Langmuir isotherm is based on monolayer adsorption on the active sites of the adsorbent with constant enthalpy of adsorption for all sites. Table 4: The parameters of Langmuir, Freundlich, and Temkin models of FGO2 for Pb2+ adsorption Langmuir Freundlich Temkin kL qmax (mg/g) R2 n kF R2 kT bT (kJ/mol) R2 0.027 169.49 0.989 2.94 22.65 0.934 0.43 0.08 0.953 The maximum adsorption capacities of FGO2 and other materials were presented in Table 5. FGO2 had better adsorption capacity in comparison to some other materials. The adsorption mechanism was explained by the electrostatic interaction of oxygen-containing groups on FGO2 surface with Pb2+, and the formation of cation - π interactions between hexagonal structure of FGO2 with Pb2+ ions. Table 5: Maximum Pb2+ adsorption capacities of FGO2 and other materials Materials qmax (mg/g) References Fe3O4/GO 169.49 In this study SiO2/Graphene 113.60 (Hao et al., 2012) Desiccated Coconut Waste 55.86 (Abdul et al., 2019) Kaolinite 10.40 (Tabrizi and Zamani, 2016) 4. Conclusions Fe3O4/GO nanocomposites were successfully fabricated by in-situ method. FTIR, XRD, and TEM results showed the Fe3O4 NPs with size ranging from 10-15 nm were deposited on GO surface. The quadratic equations were developed to model the adsorption of Pb2+ based on the nanocomposite in the high experimental compatibility. The adsorption capacity of Pb2+ was determined at 150.69 mg/g for remaining time in 50 min with initial Pb2+ concentration of 380 ppm, and pH of 6.7. The obtained results proved a great potential to apply the Fe3O4/GO nanocomposites with the well-defined adsorption characteristics for heavy metals treatment in solution. (a) (b) (c) (d) 281 Acknowledgments This research is funded by Ho Chi Minh City University of Technology, VNU-HCM, under grant number BK- SDH-2019-1880319. References Abdul R., Rabat N. E., Johari K., Saman N., Mat H., 2019, Removal of lead (II) ions from aqueous solution using desiccated coconut waste as low-cost adsorbent, Chemical Engineering Transactions, 72, 169-174. Bai X., Feng R., Hua Z., Zhou L., Shi H., 2015, Adsorption of 17β-estradiol (E2) and Pb (II) on Fe3O4/graphene oxide (Fe3O4/GO) nanocomposites. Environmental Engineering Science, 32, 370-378. Chen W., Li S., Chen C., Yan L., 2011, Self‐assembly and embedding of nanoparticles by in situ reduced graphene for preparation of a 3D graphene/nanoparticle aerogel. Advanced materials, 23, 5679-5683. Hao L., Song H., Zhang L., Wan X., Tang Y., Lv Y., 2012, SiO2/graphene composite for highly selective adsorption of Pb (II) ion. Journal of Colloid and Interface Science, 369, 381-387. Kuang L., Liu Y., Fu D., Zhao Y., 2017, FeOOH-graphene oxide nanocomposites for fluoride removal from water: Acetate mediated nano FeOOH growth and adsorption mechanism. Journal of Colloid and Interface Science, 490, 259-269. Kumar P. R., Kollu P., Santhosh C., Rao K. E. V., Kim D. K., Grace A. N., 2014, Enhanced properties of porous CoFe2O4–reduced graphene oxide composites with alginate binders for Li-ion battery applications. New Journal of Chemistry, 38, 3654-3661. Li Y., Du Q., Liu T., Sun J., Wang Y., Wu S., Xia L., 2013, Methylene blue adsorption on graphene oxide/calcium alginate composites. Carbohydrate Polymers, 95, 501-507. Marcano D.C., Kosynkin D.V., Berlin J. M., Sinitskii A., Sun Z., Slesarev A., Tour J.M., 2010, Improved synthesis of graphene oxide. ACS Nano, 4, 4806-4814. Marina G. B., Carmen P., Fernando G., Ángel Y., and Carmen B., 2019, CO2 Capture by Amino-functionalized Graphene Oxide. Chemical Engineering Transactions, 75, 637-642. Metin Ö., Aydoğan Ş., Meral K., 2014, A new route for the synthesis of graphene oxide–Fe3O4 (GO–Fe3O4) nanocomposites and their Schottky diode applications. Journal of Alloys and Compounds, 585, 681-688. Omidinia E., Shadjou N., Hasanzadeh M., 2013, (Fe3O4)-graphene oxide as a novel magnetic nanomaterial for non-enzymatic determination of phenylalanine. Materials Science and Engineering: C, 33, 4624-4632. Santhosh C., Kollu P., Felix S., Velmurugan V., Jeong S.K., Grace A.N., 2015, CoFe2O4 and NiFe2O4@ graphene adsorbents for heavy metal ions–kinetic and thermodynamic analysis. RSC Advances, 5, 28965- 28972. Shahriary L., Athawale A.A., 2014, Graphene oxide synthesized by using modified hummers approach, International Journal Renewable Energy and Environmental Engineering, 2, 58-63. Tabrizi, N.S., Zamani S., 2016, Removal of Pb (II) from aqueous solutions by graphene oxide aerogels. Water Science and Technology, 74, 256-265. Thy L.T.M., Thuong N.H., Tu T.H., Nam H.M., Hieu N.H., Phong M.T., 2019, Synthesis of magnetic iron oxide/graphene oxide nanocomposites for removal of cadmium ions from water. Advances in Natural Sciences: Nanoscience and Nanotechnology, 10, 025006. Trinh L. T., Quynh L. A. B., Hieu N. H., 2018, Synthesis of zinc oxide/graphene oxide nanocomposite material for antibacterial application. International Journal of Nanotechnology, 15, 108-117. Wang J. C., Chen P., Chen L., Wang K., Deng H., Chen F., Fu Q., 2012, Preparation and properties of poly (vinylidene fluoride) nanocomposites blended with graphene oxide coated silica hybrids. Express Polymer Letters, 6, 299-307. Yang X., Zhang X., Ma Y., Huang Y., Wang Y., Chen Y., 2009, Superparamagnetic graphene oxide– Fe3O4 nanoparticles hybrid for controlled targeted drug carriers. Journal of materials chemistry, 19, 2710-2714. Yu Y., Zhang G., Ye L., 2019, Preparation and adsorption mechanism of polyvinyl alcohol/graphene oxide‐ sodium alginate nanocomposite hydrogel with high Pb (II) adsorption capacity. Journal of Applied Polymer Science, 136, 47318. Zhao G., Ren X., Gao X., Tan X., Li J., Chen C., Wang X., 2011, Removal of Pb (II) ions from aqueous solutions on few-layered graphene oxide nanosheets. Dalton Transactions, 40, 10945-10952. 282