Arsenic removal through supercritical carbon dioxide-assisted modified magnetic starch (starch–Fe3O4) nanoparticles

Original Paper
  • 274 Downloads

Abstract

Arsenic contamination in groundwater is an alarming issue in Terai region (the low-lying land at the foot of the Himalayas) of Nepal as groundwater serves as chief drinking water source; therefore, it is required to decrease arsenic below the maximum permissible limit (50 ppb). Supercritical carbon dioxide-assisted green chemical technology was used to synthesise starch-modified magnetic nanoparticles (starch-MNPs). Synthesised starch-MNPs were used to remove arsenic from water. Various techniques were used to characterise magnetic nanoparticles (MNPs) and starch modification on MNPs surface, viz. transmission electron microscopy, scanning electron microscopy, X-ray diffraction, Fourier transform infrared spectroscopy and vibrating sample magnetometer. These analyses revealed the crystalline structure of magnetite nanoparticle with size 16 nm and uniform coating of starch onto MNPs with magnetic property. Results showed the highest arsenite [As(III)] removal in a synthetic laboratory solution of 10 ppm with at pH 6 was 99% with 0.01 g starch-MNPs, while complete removal of arsenic from groundwater sample was attained within 15 min using starch-MNPs. Arsenite adsorption agreed well with pseudo-second-order kinetic model and Langmuir isotherm model. Adsorption of arsenite on starch-MNPs reached 124 mg/g of starch-MNPs at pH 6 starch-MNPs could be regenerated in NaOH solutions and could retain more than 50% arsenite removal capacity even in fifth regeneration/reuse cycle.

Keywords

Magnetic nanoparticles Arsenic Starch-modified magnetic nanoparticles Adsorption Supercritical carbon dioxide 

Introduction

There is cumulating evidence of underground water contamination with arsenic [1]. Underground water pollution is an alarming issue especially in the Terai region of Nepal as groundwater is the primary source of water for household purposes and drinking water. Also, this poses a threat to environment and ecosystem as this water is consumed by not only humans but also other living creatures, the consequences of which are not entirely predictable and understood. Therefore, it is imperative to reduce arsenic content to permissible limit (50 ppb) as this is a matter of public health and environmental concern. This arsenic contamination is the result of various natural phenomena such as release from arsenic-containing rocks. A study on 25,058 tube wells and dug wells tested in 20 districts in 2003 found that nearly one-third of the samples contained arsenic levels above World Health Organization (WHO) drinking water quality guideline value—0.01 mg/L (10 ppb). Of these, 8% samples contained arsenic above National Drinking Water Quality Guidelines of Nepal (50 ppb) [2].

Various traditional laboratory-based techniques are widely preferred for removing arsenic from water. Common techniques employed are reverse osmosis, chemical precipitation, ion exchanges, adsorption and conventional coagulation [3]. Among these treatment techniques, adsorption method is considered more sophisticated for arsenic removal from contaminated water because of its elimination effectiveness, low resource requirement, simple and stable operation and easy handling [4, 5]. For this purpose, iron-containing compounds have been extensively studied as they are very efficient in spontaneous adsorption of arsenic [6]. Recently, research using Fe3O4 nanoparticles has attracted wide attention in arsenic removal from water as small size of Fe3O4 nanoparticles aids them to have higher surface area and porosity in 3 dimensions, which accelerates surface adsorption capacity and form stable complexes with water species from the medium. Correspondingly, the characteristic superparamagnetic properties [7] prevent magnetic accumulation and make easy removal process from the sample, solely applying an external magnetic field.

Despite these advantages, MNPs (Fe3O4) tend to oxidise in air and easily aggregate due to its anisotropic dipolar attraction and so are difficult to recycle [7]. To overcome the drawback of MNPs, development of composites incorporating Fe3O4 nanoparticles is a suitable alternative. Adhering suitable functional groups (carboxylic acids, phosphoric acid, amines, biomolecules, polymer, ligands) to MNPs leads to upsurge adsorption capacity improving the overall activity [8]. Thus, it is considered as a commonly preferred way for modification of nanoparticles. Saiz et al. [9] synthesised and used silica-coated magnetite nanoparticles functionalised with N[3(trimethoxysilyl)propyl]ethylenediamine (TMPED) to remove arsenic from groundwater that resulted in maximum arsenic adsorption capacities of 14.7 mg/g As(III). Similarly, Feng et al. [10] also synthesised superparamagnetic acid-coated Fe3O4 nanoparticles with a high specific surface area through an environmentally friendly hydrothermal route for As(III) with maximum adsorption capacities of 46.06 mg/g. Another study by Vunain et al. [11] showed polymer nanocomposite of Fe3O4 for the removal of As(III) with maximum adsorption capacity of 2.83 mg/g at 26 ± 2 °C, pH 8.6 and adsorbent dose 1.7 g/L. Though these reported studies introduce noble magnetic nanoparticles for arsenic removal, these nanocomposites either have low adsorption capacity or are synthesised using the traditional toxic method. Thus, this study aims to provide a reliable adsorption method using MNPs (Fe3O4) and its modified form as an adsorbent for arsenic removal. The modification is done with starch, a biopolymer. Starch is considered as a potential candidate for coating due to its non-toxicity, biocompatibility, biodegradability and stability. Mosaferi et al. [12] stabilised nanoscale zero-valent iron with polymer starch and carboxymethyl cellulose (CMC) to remove As(III) from water and the examination concluded 30% better adsorption by starch-stabilised than CMC. Nevertheless, the use of various solvents foists some sort of toxicity even after purification. The primary disadvantage of the conventional techniques is the presence of residual solvent that degrades the quality and efficiency of the final product. Therefore, SC-CO2 is used for surface modification of magnetic nanoparticles as it is an environmentally benign process. SC-CO2-modified nanoparticles (NPs) have biomedical applications, mostly for drug delivery studies; however, only a handful is there in the environmental application. Use of the SC-CO2 reactor system in the processing of nanocomposites for environmental studies bears excellent prospects because it is an anti-solvent process, is inexpensive and has minimal contaminants or by-products [13]. It produces small uniform particles of narrow size distribution that are unachievable when using traditional techniques. Moreover, it follows the principle of recycling as it utilises one of the abundant greenhouse gases, i.e. carbon dioxide, in the synthesis process.

Considering these traits of SC-CO2, the study employs starch as biopolymer for coating onto MNPs (Fe3O4) surface and reports a noble method for preparation of starch-modified magnetic nanoparticles (starch-MNPs) in supercritical carbon dioxide reactor. The adsorption capacity of synthesised NPs for arsenite [As(III)] was investigated by varying pH, temperature, adsorbent dose and analyte concentration as a function of contact time. Equilibrium data from the batch experiments were utilised to reveal best fit adsorption isotherm and rate of kinetics. Besides, reusability study of synthesised NPs was estimated and was utilised in field samples to compare their efficacy in total arsenic removal.

Materials and methods

Reagents and solution

All chemicals used were of analytical grade. Ferrous chloride (Fisher Scientific Purity ≥ 99%), ammonia (Fisher Scientific Purity ≥ 99%) and starch (Purity ≥ 99%) were used without further purification for synthesis and experiments. Sulphuric acid, potassium permanganate, ammonium molybdate, hydrazine hydrate, sodium hydroxide were some of the other chemicals used in experiments. Borosil glasswares were used, and all solutions were prepared in distilled water.

A stock solution of arsenic(III) was prepared by dissolving 0.00132 kg arsenic trioxide (As2O3) in water containing 0.004 kg sodium hydroxide (NaOH) and diluted to 0.001 m3 [4]. Working arsenic solutions were freshly prepared for analysis by serial dilution of stock solution.

Synthesis of MNPs

MNPs (Fe3O4) comprising selected particle size were synthesised by co-precipitation method using ammonia as the precipitating agent. MNPs were synthesised following a similar procedure described by Bisht and Zaidi [14] where magnetic nanoparticles were formed by the regular addition of ammonium hydroxide (7.0 M, 5.0 × 10−5 m3) to the solution containing different concentrations of FeCl2·4H2O at 90 °C. The precipitate obtained was separated out using external magnet, which yielded MNPs with different crystallite size and magnetic properties [15].

Modification of MNPs

Surface modification of MNPs with starch was done in a stainless steel high-pressure compact reactor (4.5 × 107 kg/m3), model 5513 equipped with a temperature controller, manufactured by Parr Instrument Company, USA. Vortexing and ultrasonication were used for preparing a mixture of distilled water solution consisting starch and MNPs. 10% (w/w) loading of MNPs was taken for modification in an effort to get modified particles to have homogeneous surface morphology with no agglomerating mass and excellent magnetic property. Then the reaction mixture was fed into the reactor and was steadily heated with electrical heating tape wrapped around the cell at 363.15 ± 1 K and 8.27 × 106 Pa. Then the reaction mixture was condensed, and CO2 was expelled out from the cell at 298.15 ± 1 K, and thus, modified MNPs were obtained [16].

Experimental

Analytical methods

A Genesis 10, Thermo Scientific, USA, UV–Vis spectrophotometer was used to quantify arsenic. Also, a Solar 969 atomic adsorption spectrometer, Thermo Elemental, UK, equipped with flow injection hydride generator (HG-AAS) in the research laboratory of Environment and Public Health Organization (ENPHO), Kathmandu, was used for analysis of arsenic in groundwater samples.

High-pressure compact reactor (Parr Instrument Company, USA), mechanical overhead stirrer (Remi, India), magnetic stirrer (Optics Technology, India) and incubator shaker (Optics Technology, India) were used for the synthesis of nanoparticles.

The MNPs and starch-MNPs characterisation was carried out using a Shimadzu 8400 FT-IR spectrophotometer. Energy-dispersive X-ray spectroscopy system (EDX) and high-resolution transmission electron microscope (TEM) model—JEOL 1011—jointly provided the information about the morphological structure of MNPs. The magnetisation property of nanoparticles was determined by Princeton EG & G applied research model 155 with extreme current 30 A. X-ray diffraction (XRD) consisting Cu Kα radiation of wavelength 1.54056 Å with range 2θ from 20° to 80° was used for determining crystallinity of nanoparticles.

A BEL analytical balance (0.0001 g), pH meter (Hanna UK), conductivity meter (Hanna UK) and turbidity meter (Hanna UK) were also used.

Batch adsorption experiments

The adsorption experiments were performed according to the batch adsorption technique to evaluate the efficiency of synthesised NPs and its modified variant. The tests were carried out by shaking As(III) solution at room temperature for an equilibrium time. Then the adsorbent was separated by using external magnet and arsenic concentration in solution was determined. The pH of the solution was adjusted using 0.5 M sodium hydroxide (NaOH) and 0.5 M acetic acid (CH3COOH) solutions, and pH meter was employed to measure the pH. The experiments were conducted at a variable adsorbent dose (5–25 mg), pH (2–12), contact time (0–600 min), temperature (30–50 °C) and initial concentration (5–30 mg/L).

The amount of As(III) adsorbed was calculated from Eq. (1):
$$ q = \left( {C_{0} - C_{\text{e}} } \right) /mV $$
(1)
where q is As(III) adsorbed (mg/g); C0, initial concentration of As(III) (mg/L); Ce, concentration of As(III) in solution at equilibrium (mg/L); V, solution volume (L); m, NPs dosage (g).
The per cent adsorption (%A) of As(III) ions is given by Eq. (2)
$$ \% A = 100 \left({C_{0} - C_{\text{e}}} \right) /C_{0}. $$
(2)

Determination of arsenic in aqueous samples

The UV–visible spectrophotometer was used for recording absorption of arsenic. The arsenic concentration was determined by molybdenum blue method measuring the absorbance of the produced arsenomolybdate complex at a wavelength of 840 nm with the help of calibration curve plotted at concentration range 0–12 ppm. Each experiment was run in triplicate, and mean values were reported. For producing the arsenomolybdate complex, specified amount of diluted solutions of As(III) was pipetted out and transferred into volumetric flasks. To each flask, sulphuric acid (1.5 N), potassium permanganate (0.1 N), ammonium molybdate (0.5%) and hydrazine hydrate (0.5 M) solutions were added. Then the volume was made up to the mark of the volumetric flask by adding distilled water. In order to allow for maximum complex formation, the solutions were left for 20 min at room temperature, and their absorbance against blank solution was measured.

Adsorption data analysis

The metal cations As(III) forms a coordinate covalent bond with surface anions O2− from MNPs and OH from starch-MNPs. This dynamic equilibrium established between adsorbent and adsorbate results uptake of arsenic by nanoparticles [17]. In the present work, isotherm models, i.e. Langmuir, Freundlich and Temkin, were used. The linearised form of Langmuir, Freundlich and Temkin are given as follows.

Langmuir isotherm

Langmuir isotherm equation is [18],
$$ C_{\text{e}} /q_{\text{e}} = 1 / \left({q_{\rm m}\,b} \right) + C_{\text{e}} /q_{\text{m}} $$
(3)
where qe is the amount of As(III) adsorbed at equilibrium (mg/g), Ce is the equilibrium analyte concentration (mg/L), and the values of Langmuir constants b (binding energy constant) and qm (monolayer adsorption capacity in mg/g) represent the energy and capability of the adsorption process.

Freundlich isotherm

Freundlich isotherm in linear form [18] is given by
$$ \log \, q_{\rm e} = \log \,K_{\rm f} + 1 /n \left({\log \,C_{\rm e}} \right) $$
(4)
where Freundlich constants Kf and 1/n are the adsorption capacity and adsorption intensity, respectively.

Temkin isotherm

Temkin isotherm in linear form [19]
$$ q_{\rm e} = B \ln A + B \ln C_{\rm e} $$
(5)
where B = RT/b, b (J/mol) is the Temkin constant associated with heat of adsorption, A (L/g) is Temkin isotherm constant, R (8.314 J/mol K) is the gas constant and T is the temperature in Kelvin (K). For Temkin isotherm, the values of A and b were obtained by the plotting of Qe versus ln Ce.
Additionally, the Chi-square (χ2) statistical test is carried out to examine the suitability of best fitting isotherm curve. Chi-square test [20] is determined by the following equation:
$$ \chi^{2} = \sum \frac{{(q_{\text{calc}} - q_{\exp})^{2}}}{{q_{\exp}}} $$
(6)
where qcalc is adsorption capacity (qe) value obtained via calculation using an adsorption isotherm model and qexp is equilibrium qe value at time t after which there is no increase in adsorption capacity.

Kinetic study

The kinetic data were analysed to determine the rate of reaction.

Pseudo-first-order kinetics

The equation [21] in linear form is:
$$ {\rm log} \left(q_{\rm e} - q_{t} \right) = {\rm log} \left[q_{\rm e} - ((k_{1}) /(2.303)) t\right]. $$
(7)

Pseudo-second-order kinetics

The equation [21, 22] is:
$$ t /q_{t} = 1 /\left({k_{2} q_{\rm e}^{2}} \right) + \left({\left(1 \right) /\left({q_{\rm e}} \right)} \right) $$
(8)
where As(III) adsorbed (mg/g) at equilibrium and time t (min) are denoted by qe and qt, respectively, k1 is pseudo-first-order rate constant (min−1) and k2 is the rate constant of pseudo-second-order kinetics (mg/g min−1). The rate constant (k1 and k2) and equilibrium adsorption capacity (qe) were calculated from slope and intercept of the linear plot of the pseudo-first- and pseudo-second-order equation.

Field sample analysis

For field sample analysis, five samples were collected from five different locations from three districts of Lumbini zone, situated in Terai region of Nepal. The three districts (Nawalparasi, Rupandehi and Kapilbastu) located in outer Terai between elevation of 200–300 m north and 63 m south above sea level were considered for sample collection [23]. These areas reported arsenic concentration in groundwater above 50 ppb [2]. The samples were collected from hand pumps in Chisapani and Laguna village in Nawalparasi, Fachkaiya village in Kapilbastu and Mandangram village in Rupandehi during post-monsoon season. The samples were collected analysing the arsenic concentration in the collection area itself using arsenic testing kit. The analytical test strips quantitatively measure different levels of arsenic by relating the reaction area of the test strip to that of the colour scale. The measuring range for strip test is 0.005–0.01–0.025–0.05–0.1–0.25–0.5 mg/L arsenic.

The samples were acidified to prevent metal precipitation and then were carried to the laboratory. Treatment of these samples with NPs was performed using 0.01 g adsorbent for a selected contact time of 15 min. Since the presence of various ions in these samples interferes arsenomolybdate complex formation and emanates errors in the quantitative analysis of arsenic, UV spectrophotometer cannot be used in the determination. Therefore, these samples were analysed by AAS to attain the concentration of total arsenic (As(III) and As(V)) content.

Result and discussion

Characterisation

The morphology and microstructure of the synthesised MNPs sample were investigated by transmission electron microscopy (TEM). As shown in Fig. 1a, b, the sample consisted of small nanoparticles agglomerated together. TEM histogram (Fig. 1c) confirmed the average size of MNPs within 9–24 nm. Furthermore, characteristic peaks for Fe and O were observed in EDX spectra (Fig. 1d). The atomic ratio O/Fe of MNPs attained from spectra, i.e. 0.72, was similar to the theoretical value 0.75. This corresponds to our result from XRD.
Fig. 1

TEM image of MNPs at scale a 100 nm and b 200 nm. Here, high-magnification TEM image reveals agglomerated nanoparticles c a histogram showing particle size distribution of MNPs used in the As(III) removal experiments and d EDX spectra of MNPs

Figure 2a, b depicts SEM images at a scale of 1 and 10 µm where growth in the size of crystal grains was due to surface modification of MNPs by starch. Probably, coated nanoparticles offer electrostatic repulsion between each molecule, and hence, a decrease in agglomeration tendency was observed [24]. The coated magnetic nanoparticles had solid, dense structure with a spherical shape. This confirms stabilisation is required to avoid the formation of nanoparticle clusters.
Fig. 2

SEM image of starch-MNPs at scale a 1 µm and magnification 10 K × and b 10 µm and magnification 1 K ×

X-ray diffraction (XRD) patterns in Fig. 3 reveal the nanocrystal nature of MNPs and its modified form. It showed the XRD analysis of spinel structured MNPs with diffraction peaks located at 30.054, 35.528, 43.25, 57.217 and 62.739, corresponding to their indices (220), (311), (400), (333) and (440), respectively, appeared in Fig. 3a, concluding that MNPs are magnetite in nature [25]. The average crystallite size of MNPs was obtained 16 nm by calculation using Debye–Scherrer equation at the highest peak 2θ (hkl) = 35.528(311). Figure 3b shows that starch-MNPs retained the crystalline structure of MNPs with identical peaks at 30.12, 35.5, 43.067, 57.195 and 62.683 at hkl (220), (311), (400), (333) and (440), respectively. The width of peaks and the starch content from the XRD of starch-MNPs confirm the coating and size influence by starch.
Fig. 3

X-ray diffraction (XRD) patterns of a MNPs and b starch-MNPs. The crystal grain size calculated from Scherrer’s equation for the strongest diffraction peak ([3 1 1] at 35.528°) is 16 nm for MNPs, and similar diffraction peaks of starch-MNPs in b confirmed that the starch coating did not influence in crystallinity structures of modified magnetic nanoparticle

In Fourier transmission infrared (FT-IR) (Fig. 4) of all samples, feature bands in the range of 3500–3000 cm−1 are assigned to the O–H vibration [16]. For magnetic nanoparticle (Spectra A), the band at 570 cm−1 are typical for distinguishing the Fe–O that is seen as a sharp peak in arsenic-adsorbed nanoparticles [10, 14]. The IR spectrum of starch (Spectra B) shows C–H stretching vibration at 2931 cm−1 [26] and C–O stretching vibrations at 1240 cm−1. These bands are present in IR spectra (Spectra C and E) of starch-MNPs and starch-MNPs with adsorbed As(III). In Spectra B, C and E, there are two strong bands at 1004 and 1074 cm−1 which are presently assigned to C–O–C symmetric stretching mode. The bands near 2800 and 1500 cm−1 in Spectra C and E correspond to CH2 stretching modes. The band at 570 cm−1 in spectra of starch-MNPs is not distinct and is suppressed by starch. In arsenic-adsorbed nanoparticles (Spectra D and E), the 570 cm−1 peak is associated with the metal–oxygen absorption band (As–O bonds in the crystalline lattice of Fe3O4). After arsenic adsorption in MNPs and starch-MNPs (Spectra D and E), the broadband observed around 800 cm−1 is ascribed to adsorbed arsenic.
Fig. 4

FT-IR spectra of a MNPs, b starch, c starch-MNPs, d MNPs with adsorbed As(III) and e starch-MNPs with adsorbed As(III)

Figure 5 shows the vibrating sample magnetometer (VSM) loops of MNPs before and after functionalisation with starch. The MNPs (Fig. 5a) exhibit somewhat superparamagnetic characteristics and magnetisation behaviour with the VSM value of 0.05 emu/g, whereas after functionalisation with starch (Fig. 5b) it exhibits the VSM value of 0.006 emu/g. The particle size of nanoparticles attained from SEM, TEM and XRD data is in accordance with increased particle size leading MNPs to acquire lower coercivity (126 Oe) over starch-MNPs (133.49 Oe). Also, this has further contributed to a reduction in the saturation magnetisation (emu/g) of starch-MNPs (4.157) over MNPs (47.25) that is ascribed to the surface adsorption of starch. Because of surface functionalisation reaction, added starch is coated onto the surface of MNPs causing to drop the magnetisation value of MNPs. In 2010, Warner et al. [27] studied the VSM of magnetic nanoparticle before and after functionalisation with five different ligands. They reported that after functionalisation of NPs, the VSM value was reduced. In the present investigation, after the surface functionalisation with starch, the VSM value decreased. A similar observation was noted by Saikia et al. [26].
Fig. 5

Hysteresis loop of a MNPs and b starch-MNPs

Batch adsorption experiments

Effect of contact time

The effect of contact time was studied on the removal of arsenite. The time courses of the adsorption of As(III) by MNPs and starch-MNPs are shown in Fig. 6. It is seen that at initial hours, per cent removal of As(III) significantly increased with the time (Fig. 6). Likewise, the difference in percentage removal between MNPs and its modified variant showed that starch-MNPs are more efficient in removal than unmodified MNPs. This is because hydrophilic nature of magnetic nanoparticle surface forbids dispersal in water; however, starch being chemisorbed on nanoparticle surface makes particles hydrophobic and facilitates better nanoparticle dispersion increasing adsorption capacity. The result obtained showed starch-MNPs have 99% removal efficiency, while that of MNPs is 94% after 10 h. This result is better in comparison with the study of Savina et al. [28] where maximum removal for 2 ppm within ~ 5 h using 0.5 g of Fe3O4 adsorbent in 500 mL was obtained at ~ 80%. According to Chowdhury and Yanful [29], equilibrium for As(III) was achieved in almost 3 h at initial concentrations of 1 and 2 ppm with maghemite–magnetite nanoparticle and the removal efficiency of As(III) was 92%. Also, they noted that the percentage removal of As(III) was lower at the higher initial concentration, for a given amount of adsorbent. In the present study, higher initial arsenite concentration (10 ppm) might be the probable reason for removal percentage over 80% achieved only after 6 h.
Fig. 6

Effect of contact time on the As(III) adsorption efficiency by MNPs and starch-MNPs (initial concentration of As(III) solution: 10 mg/L; the amount of NPs: 0.01 g; room temperature). Error bars signify standard deviations from triplicate experiments

Effect of pH

Study of the pH effect on As(III) ion removal from aqueous medium is shown in Fig. 7. The adsorption efficiency was highly pH dependent. For both nanoparticles, As(III) percentage removal was highest at pH 6. Thus, this is the optimum pH for As(III) removal by MNPs or starch-modified adsorbents. Further, at a given pH, the starch-MNPs samples showed better removal efficiency than the corresponding MNPs as shown in Fig. 7b. The highest percentage removal of arsenic(III) at pH 6 for MNPs is 82 and for starch-MNPs is 92. At pH more than 6, the removal process is minimal due to the presence of As(III) in anionic form H2AsO31− and HAsO32− in basic pH range of 7.5–9. At pH ≥ 10, competition for active sites by an excessive amount of hydroxyl ions present in the water reduces the removal process deserting certain content of arsenic in the solution. At low pH, As(III) exists in non-ionic form (H3AsO3), so positive metals like As(III) have higher electrostatic repulsion with adsorbent surface, thus decreasing the adsorption of positive metals. Thus, pH 6 is optimum pH for As(III) removal.
Fig. 7

Effect of pH on As(III) removal by starch-MNPs and MNPs adsorbents, i.e. % of arsenic(III) removal versus pH. Error bars signify standard deviations from triplicate experiments

Effect of adsorbent dose

The effect of adsorbent dose on As(III) removal was studied at pH 6 and room temperature and is represented in Fig. 8. It was observed that increased dosage amplified the removal percentage of As(III). Higher content of adsorbent has greater surface area that provides adequate free sites for the metal ion adsorption and thus favours significant removal of arsenic.
Fig. 8

Effect of adsorbent dose, i.e. As(III) adsorption efficiency by MNPs and starch-MNPs (initial concentration of As(III): 10 mg/L; time: 6 h; solution pH: 6, room temperature). Error bars signify standard deviations from triplicate experiments

Effect of analyte concentration

Figure 9 depicts the variation in adsorption capacity (qe) with different analyte concentration. For MNPs and starch-MNPs, this quantity increases as the initial concentration of As(III) changes from 5 to 30 ppm. Thus, in all cases, the starch-MNPs adsorbents exhibit significantly greater removal percentage over corresponding MNPs adsorbents. Further, Figure 9 shows that percentage removal for MNPs and starch-MNPs decreased with increasing analyte concentration because abundant active adsorbent sites are present at a lower concentration. On the other hand, available active sites remain same as that in the previous case; thus, the removal efficiency increases at higher concentration decreasing As(III) adsorbed amount or the percentage removal.
Fig. 9

Effect of adsorbate concentration, i.e. As(III) adsorption efficiency by MNPs and starch-MNPs (amount of NPs: 0.01 g; time: 6 h; solution pH: 6, room temperature). Error bars signify standard deviations from triplicate experiments

Effect of temperature

The effect of temperature upon adsorption process was studied with temperature from 30 to 50 °C (Fig. 10). The experimental observation indicated that within the range of study, increase in temperature increased removal percentage because heat favours physiochemical adsorption and maximum removal was attained at 50 °C. Some literature reported a similar trend in studies of metal ion removal [30, 31].
Fig. 10

Effect of temperature on As(III) removal by starch-MNPs and MNPs adsorbents, i.e. % of As(III) removal versus temperature Error bars signify standard deviations from triplicate experiments

Regeneration and reusability study

For obtaining the regeneration percentage, used nanoparticles were eluted with 0.5 M aqueous NaOH and were subjected to adsorption of arsenite again [32]. The arsenite on the surface of 0.01 g of recovered adsorbent was washed out almost entirely with 10 mL of eluent. Successive adsorption cycles were performed for the same adsorbent to examine its appropriateness and firmness. The procedure was carried out five times, and the results are shown in Fig. 11. It can be seen that the nanoparticles retained over 50% of their initial As(III) removal capacity after being regenerated. Thus, starch-MNPs have higher regeneration percentage in comparison with MNPs. The As(III) removal ability of starch-MNPs and MNPs were 43 and 26%, respectively, after the final recovery. Moderate reduction in the removal capacity in comparison with the original values 95 and 81% indicates the declined number of adsorption sites and agglomeration of adsorbents during the renewal process. The results clearly demonstrate that magnetic nanoparticles can be used repeatedly in an adsorption cycle. In addition, the eluted arsenic can be transformed into useful products and thus avoids the secondary pollution. This is an essential technology for arsenite removal that is practically advantageous.
Fig. 11

Percentage of As(III) adsorption behaviour of MNPs and starch-MNPs up to five cycles. Error bars signify standard deviations from triplicate experiments

Adsorption isotherm study

For adsorption isotherm study, experiments were performed at optimum condition, and the relationship between the symmetry of As(III) adsorbed and the solution concentration was investigated using Langmuir, Freundlich and Temkin isotherms. The fitted constants for three isotherms with their regression coefficients are tabulated in Table 1.
Table 1

Parameters for Langmuir, Freundlich and Temkin isotherm model

Sorbent

Langmuir model

Freundlich model

Temkin model

B (L/mg)

r

qe (mg/g)

R 2

χ 2

Kf (mg/g)

1/n

χ 2

R 2

b (J/mol)

R 2

χ 2

MNPs

0.16

0.29

109

0.98

0.001

17.81

0.58

0.107

0.96

0.0049

0.98

0.510

Starch-MNPs

0.15

0.31

124

0.99

0.385

18.93

0.6

0.455

0.97

0.0054

0.99

0.392

Langmuir isotherm

Plot of Ce/qe versus qe representing Langmuir isotherm is presented in Fig. 12a. Besides, an essential Langmuir isotherm criterion for MNPs and starch-MNPs can be described by another dimensionless constant, also called equilibrium parameter r which is defined by the following equation:
$$ r = 1/ \left({1 + bC_{0}} \right) $$
(9)
where C0 is the preliminary concentration of As(III) (mg/L) and r is the Langmuir isotherm constant. The feasibility of isotherm can be evaluated from the values of r, where r < 1 signifies favourable adsorption process, while r > 1 represents unfavourable adsorption. The calculated value of r for the initial As(III) concentration of 10 mg/L was found to be 0.31 for starch-MNPs and 0.29 for MNPs. These values being less than 1 represent favourable adsorption.
Fig. 12

Linear plots of a Langmuir, b Freundlich and c Temkin isotherm for MNPs and starch-MNPs

Freundlich isotherm

The linear plot of log qe versus log Ce in Fig. 12b represents Freundlich isotherm. A parameter in Freundlich isotherm equation-adsorption intensity (1/n) implies adsorption strength between adsorbate and adsorbent. If (1/n) > 1, then adsorption is chemical process, and if (1/n) < 1, then adsorption is physical process. Value of 1/n greater than 1 acts as characteristic of cooperative adsorption where marginal adsorption increases with increased surface concentration and strong interaction between adsorbate and adsorbent occurs. Likewise, its value lower than 1 indicates decreased marginal adsorption with increased surface concentration that arises when interaction of adsorbate for adsorption sites is minimal. In the present investigation, the calculated value was found to be 0.60 for starch-MNPs and 0.58 for MNPs. This specifies that there is decrease in adsorbate–adsorbent interaction [33].

Temkin isotherm

In Fig. 12c, plot of Qe versus ln Ce demonstrates Temkin isotherm. The slope and intercept of the line in plot reveals the value of A and b.

Comparing the results of regression (R2) value and Chi-square (χ2) value for three isotherms, Langmuir isotherm has the highest regression value and lowest Chi-square (χ2) value in comparison with Freundlich and Temkin isotherm. Thus, adsorption of As(III) by iron oxide nanoparticles fits better with Langmuir isotherm, indicating that As(III) adsorption phenomenon deviates more towards Langmuir model representing chemisorption and physisorption. The adsorption capacity (qe) obtained from Langmuir isotherm for starch-MNPs and MNPs is 124 and 109 mg/g, respectively, signifying the better adsorption capacity of starch-MNPs over MNPs. Also, these results are higher in comparison with adsorption capacity of starch-stabilised Fe° nanoparticles, i.e. 12.2 mg/g reported by Mosaferi et al. [12]. In addition, Table 2 shows comparison of the adsorption capacity of MNPs (Fe3O4) nanoparticles or its modified variant in accordance with the Langmuir equation reported in other literature.
Table 2

Comparison of As(III) adsorption capacity from Langmuir isotherm model with MNPs (Fe3O4) and its modified variant as adsorption materials reported in the literature

Adsorbent

Concentration range (mg/L)

pH

Adsorbent dose (g/L)

Adsorption capacity of As(III) (mg/g)

References

Fe3O4 nanoparticles

0.3

7

2

As(III) = 8.19

[34]

Fe3O4 and MnO2 nanoparticle-modified graphene oxide

0.01–10

7

0.5

As(III) = 14.04

[3]

Ascorbic acid-coated Fe3O4 nanoparticles

0.1

7

0.06

As(III) = 46.06

[10]

Mesoporous silica-coated iron oxide nanoparticles

20

3.33

As(III) = 14.7

[9]

Magnetite nanoparticles

1–7

6

1

As(III) = 8.0

[35]

Magnetite nanoparticles

0.1

8

0.05

As(III) = 56.3

[36]

MNPs (Fe3O4)

10

6

1

As(III) = 109

This study

Starch-MNPs

10

6

1

As(III) = 124

This study

Benefits of supercritical carbon dioxide (SC-CO2) as polymerising agent over other conventional methods

Green properties of SC-CO2 have allowed it to be extensively used in biotechnological applications [37, 38], nanomaterial processing [39] and polymerisation [40]. It is low-toxic solvent having high diffusion coefficient that enhances the catalytic activities in enzymatic or chemical reactions by preventing side reactions and their low viscosity increases mass transfer [41]. Over the past decade, SC-CO2 has been employed widely in environmental application especially in extraction and removal of heavy metal ions as it is eco-friendly substitute to conventional techniques with significant limitations such as complicated and expensive method, use of toxic and flammable solvents and problems with product isolation [42, 43, 44]. There are numerous studies utilising SC-CO2 for extraction or removal of metal ions. However, there are no any studies employing supercritically processed polymer-modified nanoparticle for heavy metal removal. A study by Wang and Guan [45] utilised a noble method for removal of arsenite and arsenate from samples via supercritical carbon dioxide extraction and pre-ion pairing with reagent tetrabutylammonium bromide (Bu4NBr). The pre-ion pairing supercritical extraction with reagent removed only about 75.6% arsenic in spiked sample when dose of reagent 50 mg/g at 42 °C, 20 MPa, CO2 5 NL/min and ethanol 0.5 mL/min for 180 min were provided. The removal efficiency could be increased using higher dosage of reagent. Likewise, Hajeb et al. [46] studied four different oil extraction processes for reduction in toxic elements in fish oil. They found that major reduction of mercury (85–100%), cadmium (97–100%) and lead (100%) content was extracted from fish oil using supercritical carbon dioxide, while the fish oil extracted from conventional extraction methods contained these toxic metals in higher quantity than accepted limit. Besides these researches emphasising the benefits of SC-CO2, results of the present study also illustrate similar result. Higher adsorption capacity of starch-MNPs was attained in comparison with nanoadsorbents employed in other studies (Table 3).
Table 3

Different method for preparation and modification of Fe3O4 used for As(III) removal reported in the published literature

Nanoparticle

Nanoparticle synthesis

Polymer used

Solvent

Performance

References

Fe3O4 magnetic nanoparticles

Thermal decomposition iron(III) acetylacetonate

Thiol (DMSA)

Benzyl ether medium

97% removal of As(III) at pH 8

[47]

Fe3O4

Microwave-assisted hydrothermal synthesis technique

Binding capacity of 100 ppb As(III) = 32.2 μg/g

[48]

Nanophase Fe3O4

Precipitation method

Binding capacity for 300 ppb As(III) = 8.196 mg/g at 1 h and 5.680 mg/g at 24 h contact time

[36]

Magnetite nanoparticles

Aerosol-assisted chemical vapour deposition (AACVD) process

 

Dilution of Fe(II) chloride in methanol used as precursor solution

Removal efficiency of 87% for As(III)

[49]

Fe3O4 magnetic nanoparticles (MNPs)

Co-precipitation method

Starch

Supercritical carbon dioxide

Removal efficiency of As(III) by starch-MNPs = 124 mg/g

This study

Kinetic study

Figure 13a, b demonstrates the plots for pseudo-first- and pseudo-second-order kinetics study. The kinetic data were analysed to determine the rate of reaction. The related parameters for pseudo-first- and pseudo-second-order kinetics model are estimated and are shown in Table 4. It was found by comparing linear regression coefficient (R2) that the fits of data with the pseudo-first-order equation (Fig. 13a) were slightly weaker than that with the pseudo-second-order equation (Fig. 13b). Thus, adsorption data are well represented by pseudo-second-order kinetics and support the assumption that the rate-limiting step of arsenic adsorption on NPs may be chemisorption. Further, from the SEM image, it is seen that the presence of starch prevents agglomeration of Fe3O4 nanoparticles. This is because the starch chains wrapped around Fe3O4 nanoparticle provides colloidal stability due to multiple hydroxyl groups of starch chelating with an iron atom of Fe3O4. Arsenite is adsorbed by electrostatic attraction and complexation between the positively charged surface hydroxyl group and arsenite. FT-IR study shows changes in the band position and intensity for MNPs and starch-MNPs. This can be attributed to the changes resulting from inner-sphere complex formation (formation of Fe–O–As complex). Also obtained starch-coated nanoparticle is easily dispersed in water due to the existence of abundant hydrophilic group surrounding nanoparticle. Similar is reported by Feng et al. [10] as properties of ascorbic acid-coated Fe3O4 nanoparticles. Besides, the adsorbents can be separated from solution via external magnet. Thus, these physiochemical interactions such as hydrogen bonding and electrostatic attraction are the mechanisms for adsorption of arsenic onto starch-MNPs.
Fig. 13

a Pseudo-first-order kinetics and b pseudo-second-order kinetics for adsorption of As(III) by MNPs and starch-MNPs

Table 4

Comparison between pseudo-first-order and pseudo-second-order kinetic models for adsorption of As(III) by MNPs and starch-MNPs

 

Pseudo-first-order kinetics

Pseudo-second-order kinetics

K 1

R 2

K 2

qe (mg/g)

R 2

MNPs

0.009

0.75

0.0003

54

0.89

Starch-MNPs

0.017

0.84

0.0002

60

0.96

Field sample analysis

Total arsenic concentrations ranging from 0.05 to 0.1 mg/L were found in groundwater samples as detected by strip test. The samples with and without adsorption using adsorbents (MNPs and starch-MNPs) were subjected to AAS analyses, and the results are shown in Table 5. As can be seen, arsenic content in these groundwater samples ranged from 30 to 70 ppb. In water samples treated with starch-MNPs, arsenic content was not detected, whereas with MNPs, arsenic content was within WHO guideline (10 ppb). MNPs adsorption on groundwater samples showed the higher percentage of arsenic removal when the concentration is higher. Alike results in earlier publications also indicate the positive correlation between lower initial concentration of arsenic and declined adsorption of arsenic on magnetic nanomaterials [50]. This result supports the effectiveness of synthesised nanomaterials, i.e. starch-MNPs, over unmodified substrate MNPs. Thus, starch-MNPs are more efficient for arsenic removal from water in comparison with its unmodified equivalent MNPs.
Table 5

Arsenic content in water samples analysed in AAS

Sample no.

As content in untreated water (ppb)

As content in MNP-treated water (ppb)

% removal of As in MNP-treated water

As content in starch-MNP-treated water (ppb)

1

30

9

70

ND

2

50

10

80

ND

3

60

10

83.3

ND

4

70

10

85.7

ND

5

60

10

83.3

ND

ND not detected

Conclusion

MNPs were synthesised by co-precipitation method and modified using a biopolymer, i.e. starch, and the modification was carried out in SC-CO2 reactor optimising the necessary conditions of MNPs dosage. These nanoparticles were employed in the removal of As(III) from water, and their removal efficiencies concerning various parameters (time, temperature, adsorbent dose, analyte concentration) were compared. These nanoparticles were characterised by microscopic (SEM and TEM), spectral (XRD, FT-IR) and magnetic (VSM) analysis. Adsorption results showed that starch-MNPs could remove up to 99% As(III), while MNPs could remove only about 94% after 10-h incubation. Also, the starch-MNPs synthesised from SC-CO2 modification process increased the adsorption capacity (qe) relative to that of MNPs. After As(III) adsorption, nanomaterials could be regenerated with a simple alkalisation process, and 42.3% As(III) removal rates could be achieved with starch-MNPs, which is two times higher than with MNPs in the fifth adsorption cycle. The adsorption models (Langmuir, Freundlich and Temkin) were used to demonstrate the adsorption mechanism, and Langmuir isotherm clarified itself as best fit to the experimental data with adsorption capacities of 109 and 124 mg/g for MNPs and starch-MNPs, respectively. Kinetic adsorption studies of As(III) were conducted, and pseudo-second-order model described the adsorption kinetics. Groundwater samples were treated with MNPs and starch-MNPs and then analysed in AAS. And the results obtained showed complete removal of arsenic using starch-MNPs in 15 min, while in case of MNPs less than maximum concentration limit (10 ppb) was attained. Likewise, the initial arsenic concentrations are usually minimum in practical situations, and the process here shows the suitable capability to remove arsenic ions at low concentrations. Therefore, magnetite nanoparticles and starch-modified MNPs could be utilised as an efficient, convenient and cost-effective adsorbent for dealing with arsenic-contaminated water.

Notes

Acknowledgements

The authors acknowledge International Foundation for Science (IFS) co-financed by the Organisation for the Prohibition of Chemical Weapons (OPCW) for funding this research under the Grant No. 5580 and TWAS Individual Research grant Ref. No. 14-187 RG/CHE/AS_I, UNESCO FR: 324028568 for equipment support.

Compliance with ethical standards

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

References

  1. 1.
    Mazumder DNG, Ghosh A, Majumdar KK et al (2010) Arsenic contamination of groundwater and its health impact on population of district of Nadia, West Bengal, India. Indian J Community Med 35:331CrossRefGoogle Scholar
  2. 2.
    Thakur JK, Thakur RK, Ramanathan AL et al (2010) Arsenic contamination of groundwater in Nepal—an overview. Water 3:1–20CrossRefGoogle Scholar
  3. 3.
    Luo X, Wang C, Luo S et al (2012) Adsorption of As (III) and As (V) from water using magnetite Fe3O4-reduced graphite oxide–MnO2 nanocomposites. Chem Eng J 187:45–52CrossRefGoogle Scholar
  4. 4.
    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 4:51–67CrossRefGoogle Scholar
  5. 5.
    Yao S, Liu Z, Shi Z (2014) Arsenic removal from aqueous solution by adsorption onto iron oxide/activated carbon magnetic composite. J Environ Health Sci Eng 12:1CrossRefGoogle Scholar
  6. 6.
    Mohan D, Pittman CU (2007) Arsenic removal from water/wastewater using adsorbents—a critical review. J Hazard Mater 142:1–53CrossRefGoogle Scholar
  7. 7.
    Xu P, Zeng GM, Huang DL et al (2012) Use of iron oxide nanomaterials in wastewater treatment: a review. Sci Total Environ 424:1–10CrossRefGoogle Scholar
  8. 8.
    Dave PN, Chopda LV (2014) Application of iron oxide nanomaterials for the removal of heavy metals. J Nanotechnol 2014:1–14.  https://doi.org/10.1155/2014/398569 CrossRefGoogle Scholar
  9. 9.
    Saiz J, Bringas E, Ortiz I (2014) Functionalized magnetic nanoparticles as new adsorption materials for arsenic removal from polluted waters. J Chem Technol Biotechnol 89:909–918CrossRefGoogle Scholar
  10. 10.
    Feng L, Cao M, Ma X et al (2012) Superparamagnetic high-surface-area Fe3O4 nanoparticles as adsorbents for arsenic removal. J Hazard Mater 217:436–449Google Scholar
  11. 11.
    Vunain E, Mishra AK, Krause RW (2013) Fabrication, characterization and application of polymer nanocomposites for arsenic (III) removal from water. J Inorg Organomet Polym Mater 23:293–305CrossRefGoogle Scholar
  12. 12.
    Mosaferi M, Nemati S, Khataee A et al (2014) Removal of Arsenic (III, V) from aqueous solution by nanoscale zero-valent iron stabilized with starch and carboxymethyl cellulose. J Environ Health Sci Eng 12:74CrossRefGoogle Scholar
  13. 13.
    Bisht G, Neupane S, Makaju R (2016) Supercritical CO2 assisted synthesis of EDTA-Fe3O4 nanocomposite with high adsorption capacity for hexavalent chromium. J Nanomater 2016:1–10.  https://doi.org/10.1155/2016/2192647 CrossRefGoogle Scholar
  14. 14.
    Bisht G, Zaidi MGH (2015) Supercritical synthesis of poly (2-dimethylaminoethyl methacrylate)/ferrite nanocomposites for real-time monitoring of protein release. Drug Deliv Transl Res 5:268–274CrossRefGoogle Scholar
  15. 15.
    Palanikumar S, Meenarathi B, Kannammal L, Anbarasan R (2015) Synthesis, characterization and catalytic activity of furosemide-functionalized ferrite on the sedimentation behavior of starch. Appl Nanosci 5:83–91CrossRefGoogle Scholar
  16. 16.
    Dung TT, Danh TM, Hoa LTM et al (2009) Structural and magnetic properties of starch-coated magnetite nanoparticles. J Exp Nanosci 4:259–267CrossRefGoogle Scholar
  17. 17.
    De D, Mandal SM, Bhattacharya J et al (2009) Iron oxide nanoparticle-assisted arsenic removal from aqueous system. J Environ Sci Health Part A 44:155–162CrossRefGoogle Scholar
  18. 18.
    Nassar NN (2010) Rapid removal and recovery of Pb(II) from wastewater by magnetic nanoadsorbents. J Hazard Mater 184(1–3):538–546.  https://doi.org/10.1016/j.jhazmat.2010.08.069 CrossRefGoogle Scholar
  19. 19.
    Saikia A, Banerjee S, Veer V (2015) Adsorption isotherm, thermodynamic and kinetic study of arsenic (III) on iron oxide coated granular activated charcoal. Int Res J Environ Sci 4:64–77Google Scholar
  20. 20.
    Ncibi MC (2008) Applicability of some statistical tools to predict optimum adsorption isotherm after linear and non-linear regression analysis. J Hazard Mater 153:207–212CrossRefGoogle Scholar
  21. 21.
    Fang XB, Fang ZQ, Tsang PKE et al (2014) Selective adsorption of Cr(VI) from aqueous solution by EDA-Fe3O4 nanoparticles prepared from steel pickling waste liquor. Appl Surf Sci 314:655–662CrossRefGoogle Scholar
  22. 22.
    Akin I, Arslan G, Tor A et al (2012) Arsenic (V) removal from underground water by magnetic nanoparticles synthesized from waste red mud. J Hazard Mater 235:62–68CrossRefGoogle Scholar
  23. 23.
    Yadav IC, Devi NL, Mohan D et al (2013) Assessment of groundwater quality with special reference to arsenic in Nawalparasi district, Nepal using multivariate statistical techniques. Environ Earth Sci 72(1):259–273.  https://doi.org/10.1007/s12665-013-2952-4 CrossRefGoogle Scholar
  24. 24.
    Wu W, Wu Z, Yu T et al (2016) Recent progress on magnetic iron oxide nanoparticles: synthesis, surface functional strategies and biomedical applications. Sci Technol Adv Mater 16(2):023501.  https://doi.org/10.1088/1468-6996/16/2/023501 CrossRefGoogle Scholar
  25. 25.
    Khan A (2008) Preparation and characterization of magnetic nanoparticles embedded in microgels. Mater Lett 62:898–902CrossRefGoogle Scholar
  26. 26.
    Saikia C, Hussain A, Ramteke A et al (2014) Crosslinked thiolated starch coated Fe3O4 magnetic nanoparticles: effect of montmorillonite and crosslinking density on drug delivery properties. Starch-Stärke 66:760–771CrossRefGoogle Scholar
  27. 27.
    Warner CL, Addleman RS, Cinson AD et al (2010) High-performance, superparamagnetic, nanoparticle-based heavy metal sorbents for removal of contaminants from natural waters. Chemsuschem 3:749–757CrossRefGoogle Scholar
  28. 28.
    Savina IN, English CJ, Whitby RL et al (2011) High efficiency removal of dissolved As (III) using iron nanoparticle-embedded macroporous polymer composites. J Hazard Mater 192:1002–1008CrossRefGoogle Scholar
  29. 29.
    Chowdhury SR, Yanful EK (2010) Arsenic and chromium removal by mixed magnetite–maghemite nanoparticles and the effect of phosphate on removal. J Environ Manag 91:2238–2247CrossRefGoogle Scholar
  30. 30.
    Al-Saad KA, Amr MA, Hadi DT et al (2012) Iron oxide nanoparticles: applicability for heavy metal removal from contaminated water. Arab J Nucl Sci Appl 45:335–346Google Scholar
  31. 31.
    Ashour RM, Abdel-Magied AF, Abdel-khalek AA et al (2016) Preparation and characterization of magnetic iron oxide nanoparticles functionalized by l-cysteine: adsorption and desorption behavior for rare earth metal ions. J Environ Chem Eng 4:3114–3121CrossRefGoogle Scholar
  32. 32.
    Kango S, Kumar R (2016) Magnetite nanoparticles coated sand for arsenic removal from drinking water. Environ Earth Sci 75:1–12CrossRefGoogle Scholar
  33. 33.
    Delle Site A (2001) Factors affecting sorption of organic compounds in natural sorbent/water systems and sorption coefficients for selected pollutants. A review. J Phys Chem Ref Data 30:187–439CrossRefGoogle Scholar
  34. 34.
    Shipley HJ, Yean S, Kan AT, Tomson MB (2009) Adsorption of arsenic to magnetite nanoparticles: effect of particle concentration, pH, ionic strength, and temperature. Environ Toxicol Chem 28:509–515CrossRefGoogle Scholar
  35. 35.
    Song K, Kim W, Suh C-Y et al (2013) Magnetic iron oxide nanoparticles prepared by electrical wire explosion for arsenic removal. Powder Technol 246:572–574CrossRefGoogle Scholar
  36. 36.
    Luther S, Borgfeld N, Kim J, Parsons JG (2012) Removal of arsenic from aqueous solution: a study of the effects of pH and interfering ions using iron oxide nanomaterials. Microchem J 101:30–36CrossRefGoogle Scholar
  37. 37.
    Brown RA, Pollet P, McKoon E et al (2001) Asymmetric hydrogenation and catalyst recycling using ionic liquid and supercritical carbon dioxide. J Am Chem Soc 123:1254–1255CrossRefGoogle Scholar
  38. 38.
    Leitner W (2002) Supercritical carbon dioxide as a green reaction medium for catalysis. Acc Chem Res 35:746–756CrossRefGoogle Scholar
  39. 39.
    Ohde H, Hunt F, Wai CM (2001) Synthesis of silver and copper nanoparticles in a water-in-supercritical-carbon dioxide microemulsion. Chem Mater 13:4130–4135CrossRefGoogle Scholar
  40. 40.
    Kendall JL, Canelas DA, Young JL, DeSimone JM (1999) Polymerizations in supercritical carbon dioxide. Chem Rev 99:543–564CrossRefGoogle Scholar
  41. 41.
    Ikushima Y (1997) Supercritical fluids: an interesting medium for chemical and biochemical processes. Adv Colloid Interface Sci 71:259–280CrossRefGoogle Scholar
  42. 42.
    Albarelli JQ, Rabelo RB, Santos DT et al (2011) Effects of supercritical carbon dioxide on waste banana peels for heavy metal removal. J Supercrit Fluids 58:343–351CrossRefGoogle Scholar
  43. 43.
    Wang J, Marshall WD (1994) Recovery of metals from aqueous media by extraction with supercritical carbon dioxide. Anal Chem 66:1658–1663CrossRefGoogle Scholar
  44. 44.
    Lin F, Liu D, Maiti Das S et al (2014) Recent progress in heavy metal extraction by supercritical CO2 fluids. Ind Eng Chem Res 53:1866–1877CrossRefGoogle Scholar
  45. 45.
    Wang T, Guan Y (2003) Removal of arsenic in the form of anions from solid matrices by supercritical CO2 extraction with ion-pairing. Sep Purif Technol 31:225–230CrossRefGoogle Scholar
  46. 46.
    Hajeb P, Selamat J, Afsah-Hejri L et al (2015) Effect of supercritical fluid extraction on the reduction of toxic elements in fish oil compared with other extraction methods. J Food Prot 78:172–179CrossRefGoogle Scholar
  47. 47.
    Singh S, Barick KC, Bahadur D (2011) Surface engineered magnetic nanoparticles for removal of toxic metal ions and bacterial pathogens. J Hazard Mater 192:1539–1547CrossRefGoogle Scholar
  48. 48.
    Parsons JG, Lopez ML, Peralta-Videa JR, Gardea-Torresdey JL (2009) Determination of arsenic (III) and arsenic (V) binding to microwave assisted hydrothermal synthetically prepared Fe3O4, Mn3O4, and MnFe2O4 nanoadsorbents. Microchem J 91:100–106CrossRefGoogle Scholar
  49. 49.
    Monárrez-Cordero BE, Amézaga-Madrid P, Leyva-Porras CC et al (2016) Study of the adsorption of arsenic (III and V) by magnetite nanoparticles synthetized via AACVD. Mater Res 19:103–112CrossRefGoogle Scholar
  50. 50.
    Bavio MA, Lista AG (2013) Synthesis and characterization of hybrid-magnetic nanoparticles and their application for removal of arsenic from groundwater. Sci World J 2013:1–7.  https://doi.org/10.1155/2013/387458 CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Chemical Science and Engineering, School of EngineeringKathmandu UniversityDhulikhelNepal
  2. 2.Department of Environment Science and Engineering, School of ScienceKathmandu UniversityDhulikhelNepal

Personalised recommendations