Adsorptive removal of arsenic from real sample of

International Journal of Environmental Science and Technology https://doi.org/10.1007/s13762-018-2140-x

ORIGINAL PAPER

Adsorptive removal of arsenic from real sample of polluted water using magnetic GO/ZnFe2O4 nanocomposite and ZnFe2O4 nanospinel S. A. Hosseini1 · A. R. Abbasian2 · O. Gholipoor1 · S. Ranjan3 · N. Dasgupta3 Received: 14 September 2018 / Revised: 17 November 2018 / Accepted: 26 November 2018 © Islamic Azad University (IAU) 2018

Abstract Developing affordable and efficient materials for the removal of arsenic from drinking water is crucial for human and environmental safety. In the present study, the adsorptive performance of magnetic GO/ZnFe2O4 nanocomposite and ZnFe2O4 nanospinel for arsenic removal from aqueous water was analyzed. The adsorbents were characterized using Fourier-transform infrared spectroscopy, X-ray powder diffraction, transmission electron microscopy, selected area electron diffraction and vibrating sample magnetometer. The conditions were optimized by response surface methodology (RSM) by considering the main factors as adsorption time, arsenic concentration, dose of adsorbent and pH. The optimum condition for the removal of arsenic was observed at pH 9.76, 30 min of contact time, 13.4 mg L−1 of initial arsenic concentration and 0.048 g of adsorbent dosage. The predicted arsenic removal percent under optimized conditions was noted as 98%; on the other hand, the experimental values at optimized conditions were observed as 96%. The Pareto analysis predicted that pH of the polluted water is the major factor in adsorptive arsenic removal and the relative importance of the process factors was found in the following order: pH > arsenic concentration > contact time > adsorbent dosage. Thus, introduced compositions form a promising material for the decontamination of polluted water or using in environmental remediation programs. Keywords Arsenic removal · Magnetic nanocomposite · Graphene oxide · RSM · Adsorption

Introduction Heavy metals present in the aquatic system and industrial effluent have turned out to be a serious threat globally, especially in the developing countries due to their toxicity, pervasiveness and coexistence with different metal species (Tang et al. 2016). Arsenic can be absorbed and accumulated in the human body through food, air or water and result in various clinical symptoms like hyperpigmentation, the risk of skin,

Editorial responsibility: M. Abbaspour. * A. R. Abbasian [email protected] 1

Department of Applied Chemistry, Urmia University, Urmia, Iran

2

Department of Materials Engineering, Faculty of Engineering, University of Sistan and Baluchestan, Zahedan, Iran

3

Faculty of Engineering and the Built Environment, University of Johannesburg, Johannesburg, South Africa

cancer, cardiovascular disease and neuropathy, memory loss and death (Chowdhury et al. 2016; Mandal 2017). Arsenic is a metalloid, a known toxic element and is considered to be the 14th most abundant element in seawater, 20th in the terrestrial crust and 12th in the human body (Mandal et al. 2013). The presence of arsenic in the environment is the result of both anthropogenic and natural sources (Mandal et al. 2013). The natural sources of arsenic include weathering, soil erosion, arsenic minerals dissolution and volcanic emissions, whereas mining, the burning of fossil fuels, smelting of metal sores, sewage sludge, ceramic manufacturing industries are some of the anthropogenic activities which lead to increased arsenic contamination in groundwater (Lata and Samadder 2016; Mandal et al. 2013). Considering the toxic effects of arsenic, World Health Organization (WHO) has announced the highest acceptable arsenic concentration in drinking water as 10 μg L−1 (Yazdani et al. 2016). As(V) is the dominant arsenic species which occurs mainly in H 2AsO 4− form in the aerobic water surface while As(III) exists in the anaerobic form in underground water mostly as H3AsO3 (Chammui et al. 2014; Çiftçi and Henden 2015). Studies

13

Vol.:(0123456789)

International Journal of Environmental Science and Technology

suggest that compounds containing an inorganic form of As(III) are observed to have approximately 60–80 times higher toxic impact on the human body as compared to As(V). As(III) removal from wastewater or drinking water is more difficult than As(V) (Çiftçi and Henden 2015). Recently, various technologies have been developed for the removal of arsenic from waste- or drinking water. The main processes and techniques include coagulation, coprecipitation, ion exchange, adsorption, membrane filtration, oxidation and reverse osmosis. They are considered as promising technologies because of their low running cost, low regeneration of waste by-product, the best potential for overall treatment, easy operation, high efficiency and lower environmental impacts (Arabnezhad et al. 2017; Çiftçi and Henden 2015; Pan et al. 2014; Yazdani et al. 2016). Recently, the use of nano-sized spinel ferrite has been reported for the wastewater treatment due to their unique characteristics such as cost efficiency, thermal and physical stability, magnetic property for easy recovery and so on (Wei et al. 2014). Though Z nFe2O4 has been observed as an excellent adsorbent for different metals and dyes, the adsorbent efficiency of ZnFe2O4 for As(III) has not yet been investigated. Graphene is a unique two-dimensional nanomaterial. Due to the remarkable structure, graphene has been emerging fascinating material with application potential in water purification and the photocatalyst. Graphene oxide (GO), the oxidized derivative of graphene, consists of a single layer of graphene bound to oxygen-rich groups in the form of carboxyl, hydroxyl or epoxy groups which has been explored in many fields. Graphene oxide (GO) can also be synthesized by several methods including Staudenmaier, Hofmann, Hummers and Tour (Dhifaf et al. 2016; Kapitanova et al. 2012; Zhu et al. 2010). In this paper, the modified Hummers method was used for synthesizing the graphene oxide. The need for new technologies to improve the removal of As(III) is emerging. Magnetic engineered Z nFe2O4 by graphene oxide has been affirmed as an efficient adsorbent for different metals as well as dyes. In this work, the As(III) removal from wastewater using magnetic separation with engineered Z nFe2O4 nanoparticle

was investigated. The target of the present study was to develop the magnetic zinc ferrite as a novel, cheap and magnetic nanoadsorbent for the removal of As(III) from polluted water. Besides, its performance is compared with GO/ZnFe2O4 composite. The ZnFe2O4 and GO/ZnFe2O4 adsorbents are environmentally friendly and can be separated by an external magnet after the process finished. The performance of the adsorbents is also investigated for the treatment of a real groundwater sample. This study is first of its translational investigation for effectual As(III) removal from actual groundwater specimen and has the potential to have societal as well as industrial impact in the form of product or process development. To save the time and cost of the study, the experiments were designed by response surface methodology (RSM). The pH, contact time, adsorbent dosage and arsenic concentration were considered as effective factors for experimental design. A mathematical model was developed, the optimum conditions were predicted and the adsorbents were compared under the optimum conditions.

Materials and methods Materials All the solutions were prepared using analytical reagent grade chemicals unless otherwise specified and doubly distilled water. As2O3 was procured from Sigma (USA) while KIO3, NaOH, HCl, C H3COOH, C2H3O2NH4, ZnSO4∙7H2O, FeSO4∙7H2O, NH3 and rhodamine B were purchased from the Merck Company (Germany). The chemical structure and characteristics of rhodamine B are presented in Table 1.

Determination of arsenic dosages A stock solution of arsenic (100 ppm) was prepared by dissolving a definite amount of A s2O3 in distilled water, and working standard solutions of As (III), 4–50 ppm, were prepared by proper dilution of the stock solution. The solution of rhodamine B (0.05 M) was prepared and stored in an amber bottle for future use.

Table 1 Properties of rhodamine B Chemical structure

13

Molecular formula

Molecular weight (g mol−1)

Chemistry class

λmax (nm)

C28H31ClN2O3

479.02

Cationic

555


separated by centrifuging and washing with DI water several times till the neutral pH of the solution was achieved and further overnight drying at 60 °C. A solution for obtaining a suspension with a concentration of 1 mg L−1 was prepared and inside a bath for 20 min ultrasonicated and finally dried in an electric oven.

Volumes of 2 mL of 1% potassium iodate and 0.5 mL of 0.45 M HCl were added to As (III) solutions, and the reaction mixture was shaken for 2 min. Then, this was followed by the addition of 2 mL of acetate buffer (pH 4.5). After that, 0.5 mL of rhodamine B solution was added to the flasks. The solution was kept for 15 min and made up to the mark with deionized water (Pillai et al. 2000). The content of As(III) in the polluted water was determined by absorption spectroscopy using a double-beam PG instrument spectrophotometer. The absorbance of As–rhodamine B complex was measured at 554 nm against the reagent blank, prepared in same way as above. A real groundwater from Western Azerbaijan region in Iran contained 56 mg L−1 Cr(VI), 43 mg L−1 As(III) and 18 mg L−1 Pb(II) and used as a real aqueous solution.

Graphene oxide and ZnFe2O4 salts were dissolved in 20 mL of ethanol with a weight ratio of 1:1 and set for ultrasonication for 40 min (40 kHz, 100 W). The collected precipitate was sieved, washed using DI water and ultimately dried in a hot air oven at 110 °C for 2 h.

Synthesis of adsorbents

Characterization of adsorbents

Synthesis of ZnFe2O4

The samples obtained were further characterized for their structure, shape and size. The following characterization techniques were done.

ZnFe2O4 was prepared by the oxidative precipitation method. Briefly, ammonia solution (25 mL) was mixed with 80 mL deionized water in desired proportions until pH reached 9.2. The reaction was conducted under vigorous magnetic stirring in a water bath at 90–95 °C. Further, FeSO4∙7H2O and ZnSO4∙7H2O raw materials with a 2:1 molar ratio were mixed in deionized water and added dropwise using a burette. The air was continually blown into the solution during the reaction. The reaction was continued for 18 h, and the pH and temperature of the solution were controlled to keep at the above-mentioned values. The precipitate was further filtered, washed with deionized water, then dried in a hot air oven for 3 h at about 100 °C and finally calcined at 600 °C for 3 h in a muffle furnace. Synthesis of graphene oxide (GO) Modified Hummers method was used for synthesizing the graphene oxide (Hosseini and Babaei 2017). Briefly, 0.05% (w/v) solution of graphite powder in sulfuric acid was prepared and kept at room temperature for 1 h and was stirred vigorously for another 2 h in an ice bath. Further, 0.5 g of NaNO3 and 3 g of KMnO4 were supplemented gradually, and the resulted suspension was magnetically stirred for 2 h in an ice bath. After removing from ice bath, the solution was stirred 2 h at the room temperature and subsequently 50 mL of deionized water was added. After 15 min, 140 mL of deionized water was added. Next, the sample was stirred further for 2 h at 90 °C and cooled to the room temperature followed by dropwise addition of 3 mL H2O2 until the solution color changes from dark brown to yellow. The suspension was

Synthesis of GO/ZnFe2O4 nanocomposite

XRD confirmatory characterization The XRD patterns were obtained from Philips PW1800 apparatus using Cukα radiation at 40 kV and 30 mA. FTIR analysis The FTIR spectra of the samples were recorded and studied in transmittance at the room temperature by Bruker spectrometer (TENSOR 27 model) in diffuse reflectance mode in the range of 400–4000 cm−1. The wave number revealed about the chemical bonding and the structure of nanocomposite. Shape and size characterization The physical morphology of the samples was determined using the transmission electron microscopy (TEM, Philips EM-208S). Characterization of magnetic property Magnetic properties of ZnFe2O4 were determined by using a vibrating sample magnetometer (VSM, Daghigh Kavir, Iran) under magnetic fields up to 10 kOe.

13


Determination of pHPZC The pHPZC (point of zero charges) value is the pH point at which the net adsorbent surface charge is zero. The pH drift route was used to distinguish the pHPZC values of ZnFe2O4 and ZnFe2O4/GO. The final pH (pHf) was plotted against the initial pH (pHi) values, and the point of intersection was considered as the pHPZC (Rahman et al. 2014).

spectrophotometer (PG Instrument 80 plus) at λmax = 554 nm. The response (As(III) percent removal) was stated as percent of As(III) removal calculated by Eq. 2:

Removal% =

A0 − A × 100, A0

where A and A0 are the absorbencies of arsenic in the adsorbent presence and absence, respectively.

RSM experimental design and optimization

Results and discussion

RSM, a fractional factorial design, could be employed as a collection of mathematical and statistical techniques for designing the process experiments, building the models, examining the simultaneous impacts of various factors and hence searching the optimized values for appropriate and desired responses (Hosseini et al. 2017). The most critical factors for arsenic removal considered in this study include: initial concentration of arsenic (X1), dosage of adsorbent (X2), time of contact (X3) and pH (X4) to examining their impact on remediation of arsenic. Total of 31 experiment runs were designed and further optimized based on the RSM approach with central composite design method. Every experimental run was analyzed twice to avoid any experimental and handling errors, and the mean of response was further considered. The arsenic removal percent for every experimental run was presented with ± 1% precision. The second-order polynomial response model was considered to correlate the independent and dependent variables using Eq. 1:

Adsorbent characterization

Y = b0 +

n ∑

bi Xi +

i=1

n ∑ i=1

bii Xi2 +

n−1 n ∑ ∑

bij Xi Yi ,

(1)

i=1 j=i+1

where Y is the arsenic removal, Xi and Xj are coded factors which influence the final response (Y) and B0 is the constant. The ith linear coefficient, ith quadratic coefficient and the ijth interaction effect coefficient are as Bi, Bii and Bij, respectively. Analysis of variance (ANOVA) was used for interpreting the results using Minitab 17 software.

Adsorption studies Standard As(III) stock solution (100 mg mL−1) was prepared in 100 mL distilled water. Experiments were performed in accordance with matrix of RSM design as mentioned in Table 2. The pH, contact time, concentration of arsenic and dosage of adsorbent were in ranges 4–8, 40–60 min, 20–40 mg L−1 and 0.4–0.8 g, respectively. The adsorbents were separated with the help of a magnet. The arsenic concentration adsorbed was determined using ultraviolet–visible

13

(2)

The FTIR spectra of GO, ZnFe2O4 and ZnFe2O4/GO are presented in Fig. 1a. In the spectrum of GO, the peak at 1050.50 is allocated to the C–O stretching vibration. The peak at 1711.55 cm−1 is assigned to C=O. The peaks at 1580.28 and 3396.18 cm−1 resemble the C=C and hydroxyl group stretching vibrations of carboxylic acid, respectively. FTIR of ZnFe2O4 particles was also noted for better illustration and examination of surface groups. Absorption peaks at 593 and 442 cm−1 validate the cubic spinel structure formation of ZnFe2O4. The 548.36 cm−1 peak demonstrates metal–oxygen stretching bonds in a tetrahedral situation (mainly Zn). The intense peak in 456.57 cm−1 is attributed to the metal–oxygen stretching band in the octahedral site (mainly Fe) (Hosseini et al. 2018). The spectrum of FTIR for the Z nFe2O4/ GO demonstrates all of the peaks appeared at the Z nFe2O4 and GO spectra; just the peaks intensity has been relatively low. As depicted in Fig. 1b, the XRD pattern consists of high-intensity peaks, which further endorses the polycrystalline nature of the synthesized material. The planes corresponding to the diffraction peaks are (220) (311), (400), (422), (511) and (440) which afford a clear indication for the spinel structure formation of the ferrite which also resembles well the standard pattern (JCPDS 22-1012) file for Z nFe2O4. This observation resembles significantly the earlier findings and reports (Chammui et al. 2014; Yazdani et al. 2016). It is also illustrated that the ferrite concludes some residual a-Fe2O3 phase. Also, through the equation of Debye Scherer (Eq. 3) (Hosseini et al. 2017), the mean crystallite size of the ZnFe2O4 was estimated to be 32 nm:

D = 0.89𝜆∕𝛽cos𝜃 ,

(3)

where λ is X-ray wavelength and β is X-ray reflection effective line width. To understand the microstructural characteristic of synthesized powders, the transmission electron microscopic (TEM) images were taken. Figure 2a shows the image of synthesized ZnFe2O4 nanoparticles. It can be clearly seen the size of the particles is in the nanorange, entirely. Also,

International Journal of Environmental Science and Technology Table 2 Matrix of RSM design and the response values

Run

pH

Time (min)

Concentration (mg L−1)

Adsorbent dos- Experimental age (g) response %

Predicted response %

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

8 6 4 8 4 6 6 4 4 2 4 4 8 6 6 6 6 6 8 6 8 6 6 4 8 6 4 8 8 10 6

60 50 40 60 60 50 70 60 60 50 40 60 60 50 50 50 50 50 40 50 40 50 50 40 40 30 40 40 60 50 50

20 30 20 40 20 50 30 20 40 30 40 40 20 10 30 30 30 30 40 30 40 30 30 20 20 30 40 20 40 30 30

0.4 0.6 0.4 0.8 0.8 0.6 0.6 0.4 0.4 0.6 0.4 0.8 0.8 0.6 0.6 0.2 0.6 0.6 0.4 0.6 0.8 0.6 1.0 0.8 0.8 0.6 0.8 0.4 0.4 0.6 0.6

92.12 70.23 72.62 68.69 90.28 71.41 66.44 78.13 69.37 85.24 63.85 89.81 72.46 83.94 70.23 88.76 70.23 70.23 65.54 70.23 63.18 70.23 98.55 84.77 66.95 55.42 93.30 86.60 71.05 96.11 70.23

some weak aggregation occurred due to the high surface area of nanoparticles. Selected area electron diffraction (SAED) of ZnFe2O4 nanoparticles presented in Fig. 2b exhibits the diffused spotty pattern due to the well-crystalline nature of ZnFe2O4 powders (Abbasian et al. 2015; Mostaan et al. 2017). Figure 2c represents the morphological features of GO. Furthermore, SAED patterns of GO (Fig. 2d) showed a diffused diffraction ring pattern. This contributed to a disordered structure and illustrates the disruption of the conjugated structure of graphite due to chemical oxidation (Flyunt et al. 2014). Figure 2e represents a bright-field TEM image of a GO/ZnFe2O4 nanocomposite. Figure 2f gives the SAED pattern of a GO/ZnFe2O4 nanocomposite indicating bright arc-like diffraction spots with a circular configuration. This pattern is not a typical pattern of a single crystal, suggesting that ZnFe2O4 nanoparticles were formed by assembling with a slight misalignment among the GO.

93.20 72.61 73.00 72.00 92.00 71.00 65.49 73.00 71.00 88.00 64.00 96.00 72.00 86.00 69.00 89.00 67.00 72.00 63.00 72.00 63.00 69.00 99.97 83.00 64.00 58.02 92.00 88.00 72.00 68.00 70.00

The use of magnetic adsorbents facilitates the adsorbent removal after the process finished. Hence, determining the magnetization of adsorbents is of importance. Magnetic characterization of Z nFe2O4 and Z nFe2O4/GO was investigated with a vibrating sample magnetometer and attributed in Fig. 3. The magnetization values at 10 kOe magnetic field for ZnFe2O4 and ZnFe2O4/GO were found to be 9 and 7 emu g−1, respectively. However, even in this strong field, the sample is not saturated. The magnetization value for ZnFe2O4 was less than the value of some other magnetic spinels, but this value is still acceptable for magnetic separation applications. This is the first reported study where Z nFe 2O 4/GO nanocomposite is used to study arsenic removal from polluted water. Various studies have reported nanocomposites with ZnFe2O4 for other applications. For example, Zhou et al. (Zhou et al. 2017) reported that ZnFe2O4/BiOI

13


Fig. 1 FTIR spectra of ZnFe2O4, GO, GO/ZnFe2O4 (a) and the XRD patterns of Z nFe2O4 and ZnFe2O4/GO (b)

Fig. 2 TEM images of a ZnFe2O4, c GO, e ZnFe2O4/GO and SAED patterns of b ZnFe2O4, d GO and f ZnFe2O4/GO

13


Fig. 3 Magnetization curve of ZnFe2O4 and ZnFe2O4/GO nanoparticles

nanocomposite was able to degrade the methylene blue, rhodamine B and methylene orange in visible-light irradiation rather than pure BiOI and ZnFe2O4. Similarly, Zamiri et al. 2017 discussed the magnetic and optical properties of ZnO/ZnFe2O4 nanocomposite synthesized by a low-cost and straightforward chemical precipitation method.

RSM optimization RSM results could be employed to optimize and to study the interactions and the process statistical analysis of the process. Subsequently, the quadratic response surface model (Eq. 4) could significantly better fit the experimental results among various evaluated models:

Y = 70.231 − 3.133X1 + 2.448X2 + 2.755X3 − 4.033X4 + 1.861X12

(4)

+ 5.858X22 − 2.324X32 + 1.736X42 + 4.325X1 X2 − 3.075X1 X4 − 7.950X2 X4

X1 and X4 with a negative sign have an antagonistic impact on the response, signifies arsenic concentration and pH increase further decreases the arsenic removal percentage. Among particular terms, the X2 (adsorbent dosage) through the smallest coefficient has the minimum effects on the arsenic removal (response). Among binary terms, the interactions of X1X4 and X2X4 have a negative impact on the arsenic response. Table 3 shows the results of ANOVA for the designed model which suggest adequacy of the equation to describe the association among the arsenic removal (response) and the significant variables. The Fisher’s F test was used for verifying the model statistical significance and approved the relevance of the model with F value of 52.54. The model significance was evaluated by the correlation coefficient (R2), which is a correlation measure among the predicted response and experimental data. Figure 4 shows a graph of the predicted response plotted by the model against the experimental responses. The determination coefficient (R2) between the predicted and the experimental values was 0.9682, which shows the model validity. This further means that experimental data are in significantly better consistence with predicted data which further implies that 96.82% of the variations in removal efficiency of arsenic are enlightened by the independent vari2 ables. The predicted model determination coefficient (Rpred ) was 0.9040, indicating that 90.40% of the arsenic removal variation is attributed to the four independent model factors. 2 The adjusted R2 (Radj ) = 0.9497 was also statistically significant and commands the model correlation applicability. Table 4 predicts the coefficient of regression and corresponding t value, and on the contrary, the significance for the coefficient of regression in the model variables was examined by the p value. Also, the Student’s t test was applied for calculating the regression coefficient importance for the model terms, analyzing the possibilities of the true parameter is zero. The t value and p value for each model term are given in Table 4. Usually, the smaller the p value (p time of reaction > adsorbent dosage. Among independent factors, the arsenic concentration and pH are the most significant factors on the response. The Pareto analysis anticipates that the adsorbent dosage does not significantly effect on the response between rather to independent factors. On the other hand, the following relative importance order of model terms resulted:

13

Fig. 5 Pareto chart analysis

X2*X4 (35.66%) > X22 (19.36%) > X1*X2 (10.55%) > X4 (9.17%) > X1 (5.53%) > X1*X4 (5.33%) > X3(4.28%) > X2 (3.38%) > X23 (3.04%) > X21 (1.95%) > X24(1.70%). Figure 6a, b demonstrates 3D surface plot and 2D contour plot for the combined impact of dosage of adsorbent and concentration of arsenic on the adsorption capacity of arsenic, respectively. It is resulted from the figure that X1X2 term has a positive effect on the model response (adsorption capacity of arsenic) as evidenced from the positive terms in the model. The maximum impact on the response is observed at mid-levels of arsenic concentration and high level of adsorbent dosage, meaning that the increasing adsorbent dosage leads to increase the adsorption capacity. The statistical model results are similar with the earlier research and literature and reveal the RSM ability at experimental and prediction results. Figure 6c, d illustrates a 3D surface plot and 2D contour for the collective interaction of arsenic concentration and pH on the adsorption capacity of arsenic, respectively. The figure indicates that best interaction of As(III) concentration–pH occurs at mid-levels of arsenic concentration and low pH. Figure 6e, f shows 3D surface plot and 2D contour plot for the combined interaction of dosage of adsorbent and pH on the arsenic adsorption capacity, respectively. It results from the figure that the most impact on As(III) removal happens at mid-levels of pH and maximum adsorbent dosage. According to Eq. 4, the combined interaction of adsorbent dosage and contact time (X2X4) has a negative effect on arsenic removal. Detailed review of the literature indicates that the low pH is suitable for As(V) adsorption (Dutta et al. 2004; Wei et al. 2016). In the case of As(III), the maximum adsorption occurs at high pH (Burton et al. 2009; Qiao et al. 2012). Likewise, the optimized process condition of arsenic removal by Z nFe2O4 was expected by the RSM. The adsorption process optimized conditions were expected to be at pH,


Fig. 6 Response surface plot and contour plot of the decolorization efficiency as the function of arsenic concentration and adsorbent dosage (a, b), arsenic concentration and pH (c, d) and adsorbent dosage and pH (e, f)

time, arsenic concentration and adsorbent dosage of 9.76, 30 min, 13.4 mg L−1 and 0.048 g, respectively. The predicted response under these conditions was 98%, whereas the experimental test of predicted condition led to 96% removal of arsenic. In the same condition, the removal percent of arsenic by GO/ZnFe2O4 and GO was 94% and 91%, respectively. The capacity of Z nFe2O4 for As(III) adsorption is higher than that of GO. The GO/ZnFe2O4 composite is comprised of the same ratio of GO and ZnFe2O4 (weight ratio 1:1). Thus, it is logic that the adsorption capacity of the composite is between the capacity of GO and ZnFe2O4. To justify the results, firstly, we considered the distribution of arsenic species at different pH as well as the p Hpzc (the pH for the point of zero charge) of ZnFe2O4. Different distributions of As(III) exist in the solution phase as a pH function. Altering the solution pH alters the As(III) speciation through the following equations:

H3 AsO3 ↔ H+ + H2 AsO−3 pKa = 9.23

(6)

H2 AsO−3 ↔ H+ + H2 AsO2− pKa = 12.10 3

(7)

H2 AsO2− ↔ H+ + AsO3− pKa = 13.41 3 3

(8)

The pHpzc of ZnFe2O4 was determined to be 6.5. According to the RSM prediction, the optimum pH for maximum As(III) removal is 9.76. In this pH, the adsorbent possesses negative charge and the As(III) species are found as H2AsO3− in the solution. According to pHpzc and electrostatic factors, the ZnFe2O4 could not adsorb H2AsO3− species (Dutta et al. 2004; Wei et al. 2016). Therefore, the

p HPZC is not a suitable factor to investigate As(III). Based on DFT simulation reported in the literature (Wei et al. 2016), the surface of the adsorbent with positive charge would attract O atom of arsenic species and surface with negative charge would attract H atom of arsenic species. On the other hand, based on the adsorption energies by DFT calculations (Wei et al. 2016), the adsorption orders of As(III) are AsO33− > OH− > HAsO32− > H2AsO3− > H2O > H3AsO3. Under acidic and neutral conditions, H 3AsO3 is the dominant species in aqueous solution, whereas the adsorption ability of H 3AsO3 is weaker than that of water. Therefore, the adsorption of As(III) is low at low pH range. The adsorption of As (III) increased in alkaline conditions, because H2AsO3− should appear on the surface, and then the maximum adsorption of As(III) on Z nFe2O4 appears at pH 9–10, which is agreement with RSM model prediction and experimental confirmation. Finally, the adsorptive performance of Z nFe2O4 for real groundwater (obtained from Western Azerbaijan, Iran) containing 56 mg L−1 Cr(VI), 43 mg L−1 As(III) and 18 mg L−1 Pb(II) resulted in 90, 86 and 75% removal, respectively. It can be noted that this is first of its kind study for As(III) efficient removal from real samples using novel engineered nanocomposite. Recently, some researches are dedicated to the removal of arsenic from water. However, the optimized conditions have been reported in very few. Hussein and Abu-Zahra (2016) reported that by incorporating 15–20 nm iron oxide nanoparticles inside porous polyurethane foam, they were able to remove 40% arsenic using single-stage batch analysis. Shokri et al. (2016) synthesized polysulfone/

13


organoclay adsorptive nanocomposite membrane for the removal of arsenic from contaminated water. It had high adsorption capacity for As(V) solutions and was regenerable for multiple cycles. The use of ZnFe2O4 has shown an encouraging substituted technology for the treatment to remove arsenic from contaminated water. It establishes the higher sorption capability of nanoparticles for the elimination of arsenic. This possible greater arsenic removal has favored the implications of these nanoparticles to contaminated water systems with arsenic levels lesser than 1500 μg L−1. Nanosorbents are observed to be more impactful in arsenite and arsenate removal from wastewater or arsenic-contaminated water. It has been observed that even high arsenic pollutants of contaminated water can be significantly remediated using various nanosorbents with a comparatively smaller sorbent amount and shorter time. The nanosorbents should have the adequate capacity for adsorption with a lower dose of the adsorbent for an initial concentration of till 10 mg L−1 for the selection of nanosorbents as the adsorbent for remediation of arsenic from polluted water. In the present study, magnetic NPs are noted to have moderately improved the capacity of adsorption. The results give a strong emphasis on maximum real-life situations encountered in the regions of arsenic contamination throughout the globe.

Conclusion Magnetic Z nFe 2O 4 and Z nFe 2O 4/GO nanocomposites were efficaciously engineered by the oxidative precipitation approach, and their performance analysis shows an enhanced uptake of arsenic under optimized conditions to save time and cost of the study. The predicted percent of removal of arsenic under optimum conditions was found to be 98%, whereas the experimental results of optimized condition tend to be 96% of arsenic degradation. The Pareto analysis further expected that the relative importance order of the four independent factors is as follows: pH > arsenic concentration > reaction time > adsorbent dosage. Under the identified optimum conditions, GO/ZnFe2O4 nanocomposite exhibited better adsorption than GO. The study showed that ZnFe2O4-engineered nanocomposite with GO could be the effective adsorbent for industrial arsenic removal from polluted waters. Further implementation could also be achieved by integrating and incorporating the magnetic hetero-structures into contaminated real water sample purification systems. Worldwide, research is ongoing on this objective, and

13

most technologists and researchers around the world are achieving the utilization of the superparamagnetic behavior and mono-dispersed characteristics of these magnetic sorbents to develop large-scale treatment units for contaminated water with higher and efficient arsenic elimination potential from several water reservoirs to provide safer water with the common mass of people. Acknowledgements Authors would like to acknowledge Iranian Nanotechnology Initiative Council and Urmia University for financial support.

References Abbasian AR, Rahimipour MR, Hamnabard Z (2015) Hydrothermal synthesis of lithium meta titanate nanocrystallites. Proc Mater Sci 11:336–341. https://doi.org/10.1016/j.mspro.2015.11.110 Arabnezhad M, Shafieeafarani M, Jafari A (2017) Co-precipitation synthesis of ZnO–TiO2 nanostructure composites for arsenic photodegradation from industrial wastewater. Int J Environ Sci Technol. https://doi.org/10.1007/s13762-017-1585-7 Burton ED, Bush RT, Johnston SG, Watling KM, Hocking RK, Sullivan LA, Parker GK (2009) Sorption of arsenic(V) and arsenic(III) to schwertmannite. Environ Sci Technol 43:9202–9207. https:// doi.org/10.1021/es902461x Chammui Y, Sooksamiti P, Naksata W, Thiansem S, Arqueropanyo O-A (2014) Removal of arsenic from aqueous solution by adsorption on Leonardite. Chem Eng J 240:202–210. https://doi. org/10.1016/j.cej.2013.11.083 Chowdhury S, Mazumder MAJ, Al-Attas O, Husain T (2016) Heavy metals in drinking water: occurrences, implications, and future needs in developing countries. Sci Total Environ 569–570:476– 488. https://doi.org/10.1016/j.scitotenv.2016.06.166 Çiftçi TD, Henden E (2015) Nickel/nickel boride nanoparticles coated resin: a novel adsorbent for arsenic(III) and arsenic(V) removal. Powder Technol 269:470–480. https://doi.org/10.1016/j.powte c.2014.09.041 Dhifaf AJ, Neus L, Kostas K (2016) Synthesis of few-layered, highpurity graphene oxide sheets from different graphite sources for biology. 2D Mater 3:014006 Dutta PK, Ray AK, Sharma VK, Millero FJ (2004) Adsorption of arsenate and arsenite on titanium dioxide suspensions. J Colloid Interface Sci 278:270–275. https://doi.org/10.1016/j.jcis.2004.06.015 Flyunt R et al (2014) Mechanistic aspects of the radiation-chemical reduction of graphene oxide to graphene-like materials. Int J Radiat Biol 90:486–494. https://doi.org/10.3109/09553 002.2014.907934 Hosseini SA, Babaei S (2017) Graphene oxide/zinc oxide (GO/ZnO) nanocomposite as a superior photocatalyst for degradation of methylene blue (MB)-process modeling by response surface methodology (RSM). J Braz Chem Soc 28:299–307 Hosseini SA, Davodian M, Abbasian AR (2017) Remediation of phenol and phenolic derivatives by catalytic wet peroxide oxidation over Co–Ni layered double nano hydroxides. J Taiwan Inst Chem Eng 75:97–104. https://doi.org/10.1016/j.jtice.2017.03.001

International Journal of Environmental Science and Technology Hosseini SA, Majidi V, Abbasian AR (2018) Photocatalytic desulfurization of dibenzothiophene by NiCo2O4 nanospinel obtained by an oxidative precipitation process modeling and optimization. J Sulfur Chem 39:119–129. https://doi.org/10.1080/17415 993.2017.1369981 Hussein FB, Abu-Zahra NH (2016) Synthesis, characterization and performance of polyurethane foam nanocomposite for arsenic removal from drinking water. J Water Process Eng 13:1–5. https ://doi.org/10.1016/j.jwpe.2016.07.005 Kapitanova OO, Panin GN, Baranov AN, Kang TW (2012) Synthesis and properties of graphene oxide/graphene nanostructures. J Korean Phys Soc 60:1789–1793. https: //doi.org/10.3938/ jkps.60.1789 Lata S, Samadder SR (2016) Removal of arsenic from water using nano adsorbents and challenges: a review. J Environ Manag 166:387– 406. https://doi.org/10.1016/j.jenvman.2015.10.039 Mandal P (2017) An insight of environmental contamination of arsenic on animal health. Emerg Contam 3:17–22. https://doi. org/10.1016/j.emcon.2017.01.004 Mandal S, Sahu MK, Patel RK (2013) Adsorption studies of arsenic(III) removal from water by zirconium polyacrylamide hybrid material (ZrPACM-43). Water Resour Ind 4:51–67. https ://doi.org/10.1016/j.wri.2013.09.003 Mostaan H, Mehrizi MZ, Rafiei M, Beygi R, Abbasian AR (2017) Contribution of mechanical activation and annealing in the formation of nanopowders of Al(Cu)/TiC–Al2O3 hybrid nanocomposite. Ceram Int 43:2680–2685. https://doi.org/10.1016/j.ceram int.2016.11.082 Pan B, Li Z, Zhang Y, Xu J, Chen L, Dong H, Zhang W (2014) Acid and organic resistant nano-hydrated zirconium oxide

(HZO)/polystyrene hybrid adsorbent for arsenic removal from water. Chem Eng J 248:290–296. https: //doi.org/10.1016/j. cej.2014.02.093 Pillai A, Sunita G, Gupta VK (2000) A new system for the spectrophotometric determination of arsenic in environmental and biological samples. Anal Chim Acta 408:111–115. https://doi.org/10.1016/ S0003-2670(99)00832-6 Qiao J, Jiang Z, Sun B, Sun Y, Wang Q, Guan X (2012) Arsenate and arsenite removal by F eCl3: effects of pH, As/Fe ratio, initial As concentration and co-existing solutes. Sep Purif Technol 92:106– 114. https://doi.org/10.1016/j.seppur.2012.03.023 Rahman MM, Adil M, Yusof AM, Kamaruzzaman YB, Ansary RH (2014) Removal of heavy metal ions with acid activated carbons derived from oil palm and coconut shells. Materials 7:3634–3650 Shokri E, Yegani R, Pourabbas B, Kazemian N (2016) Preparation and characterization of polysulfone/organoclay adsorptive nanocomposite membrane for arsenic removal from contaminated water. Appl Clay Sci 132–133:611–620. https: //doi.org/10.1016/j. clay.2016.08.011 Tang X et al (2016) Chemical coagulation process for the removal of heavy metals from water: a review. Desalin Water Treat 57:1733– 1748. https://doi.org/10.1080/19443994.2014.977959 Wei J, Zhang X, Liu Q, Li Z, Liu L, Wang J (2014) Magnetic separation of uranium by C oFe2O4 hollow spheres. Chem Eng J 241:228– 234. https://doi.org/10.1016/j.cej.2013.12.035 Wei Z et al (2016) The effect of pH on the adsorption of arsenic(III) and arsenic(V) at the T iO2 anatase [101] surface. J Colloid Interface Sci 462:252–259. https://doi.org/10.1016/j.jcis.2015.10.018

13


Yazdani M, Tuutijärvi T, Bhatnagar A, Vahala R (2016) Adsorptive removal of arsenic(V) from aqueous phase by feldspars: kinetics, mechanism, and thermodynamic aspects of adsorption. J Mol Liq 214:149–156. https://doi.org/10.1016/j.molliq.2015.12.002 Zamiri R et al (2017) Optical and magnetic properties of ZnO/ZnFe2O4 nanocomposite. Mater Chem Phys 192:330–338. https://doi. org/10.1016/j.matchemphys.2017.01.066

13

Zhou Y et al (2017) Fabrication of novel ZnFe2O4/BiOI nanocomposites and its efficient photocatalytic activity under visible-light irradiation. J Alloys Compd 696:353–361. https://doi.org/10.1016/j. jallcom.2016.11.323 Zhu Y, Murali S, Cai W, Li X, Suk JW, Potts JR, Ruoff RS (2010) Graphene and graphene oxide: synthesis, properties, and applications. Adv Mater 22:3906–3924. https://doi.org/10.1002/adma.201001068