Remediation of Contaminated Water With Crystal Violet Dye by Using Magnetite Nanoparticles: Synthesis, Characterization and Adsorption Mechanism Studies

Recently, Adsorbents with magnetic properties such as MNPs have attracted interest in many environmental engineering related applications due to a signifi cant eff ect in accelerating separation and improving the effi ciency of water treatment. With reported sizes ranging from 1 to 100 nm, high surface-to-volume ratio, and high loading capacity, MNPs were successfully used as adsorptive materials for pollutants [7-9]. Th erefore, these nanoparticles received signifi cant attention as a low-cost processing and operational easiness adsorbents for potential applications in the environmental treatment.


Introduction
Dyes are known as pollutants that not only aff ect aesthetic advantages of environment, but also reduce light penetration and photosynthesis, and some of them are considered toxic and even carcinogenic to human health [1]. Adsorption process provides an attractive alternative for the treatment of colored wastewaters due to its simplicity, selectivity, and effi ciency [2][3][4][5][6].
Recently, Adsorbents with magnetic properties such as MNPs have attracted interest in many environmental engineering related applications due to a signifi cant eff ect in accelerating separation and improving the effi ciency of water treatment. With reported sizes ranging from 1 to 100 nm, high surface-to-volume ratio, and high loading capacity, MNPs were successfully used as adsorptive materials for pollutants [7][8][9]. Th erefore, these nanoparticles received signifi cant attention as a low-cost processing and operational easiness adsorbents for potential applications in the environmental treatment.
CV dye is highly toxic to mammalian cells and could cause skin and digestive tract irritation. Also, It may leads to respiratory and kidney failure in extreme conditions. Th us it is necessary to remove this dye from waste water before its fi nal disposal.
Th e study scope is to investigate the adsorption behavior of MNPs as a surfactant for removal of crystal violet (CV). Based on this, the eff ects of variables infl uencing the removal of dyes were evaluated.
Th e adsorption kinetics and isotherms for CV dye onto MNPs were studied.

Experimental Materials
Crystal Violet (C 25 H 30 N 3 Cl) as a cationic azo dye, ferric chloride (FeCl 3 .6H 2 O), ferrous sulfate (FeSO 4 .7H 2 O), ammonium hydroxide (NH 4 OH), diluted hydrochloric acid (HCl) and sodium hydroxide (NaOH) were used for this study. A stock solution of CV (1000 mg l -1 ) was prepared in distilled water. Th is solution was diluted with distilled water to prepare stock solutions with the concentration of 10, 25 and 50 mg l -1 of CV.

Preparation of magnetite nanoparticles (MNPs)
Stock solutions of ferrous ions (Fe +2 ; 1M) and ferric ions (Fe +3 ; 2M) were prepared by dissolving appropriate weight in 200 ml of distilled water, the stock solutions were mixed and heated at 90 °C , ammonium hydroxide solution (33%) was introduced by syringe dropwise until pH adjust around 10-12. Th e appearance of black colour was indicated the formation of magnetite particles. Th e black mixture was heated at 80 °C for 60 minutes, fi ltered and washed with deionized water repeatedly until pH become neutral and dried at room temperature [10].

Preparation of crystal violet dye solution
A stock solution of CV (1000 mg/l) was prepared by dissolving 1 gm in 1 liter distilled water. Th en, it diluted by distilled water to prepare diff erent concentrations of dye according to dilution law.    Th e dye was made up in stock solution of concentration 1000 mg l -1 (1 g CV in 1000 ml of distilled water) and was subsequently diluted to the required concentrations from 10 to 100 mg l -1 . Th e eff ect of some parameters such as pH (3)(4)(5)(6)(7)(8)(9) and adsorbent dosage (0.02-0.2 g) was made by known amount of MNPs and 25 ml of CV solution. Th e pH of dye solution was adjusted using with 0.1 M HCl or 0.1 M NaOH solutions. Th e sorption studies were performed at diff erent temperatures (25 o C-95 o C). Th e mixture was shaken using water bath at 240 rpm (25 °C for 30 min). Aft er each adsorption process, the residual CV solution was separated the absorbance of the fi ltrate was measured at λ max 576 nm. In all these diff erent parameters, the amount of dye taken by adsorbent is calculated as: q e = (C i -C e ) v/m (15) Where: (q e ; the amount of dye taken by adsorbent (mg/g)), (C i and C e ; the concentrations of dye at initial and equilibrium (mg/l), respectively), (V; the volume of solution (l)) and (m; the mass of adsorbent (g)). Also the removal effi ciency is calculated as: %Removal = (C i -C f )/Ci * 100 (16) Where: (C i and C f : the initial and fi nal dye concentrations (mg/l), respectively).

Characterization
FT-IR spectra of the samples were carried out using (FT/IR-4100) spectrophotometer (Th ermoFisher Nicolet IS10, USA) in KBr pellets at room temperature. Micrographs of the samples were taken using SEM (JSM-6510, JEOL, Ltd.). Th e images of TEM were taken by a JEM-2100 operated at an accelerating voltage of 200 kV. UVvisible spectroscopic analysis was carried out on Oasis Scientifi c (PG Instruments T80).

Morphology Characterization
Th e morphology and particle size of the prepared MNPs was investigated by TEM. As shown in Figure 1, MNPs exhibited spherical morphologies with a uniform particle size (about 11 nm).
Th e morphologies of MNPs (a: before adsorption) and MNPs-CV (b: aft er adsorption) are investigated by SEM and shown in Figure 2. It can be seen from this Figure 2 (2a) that MNPs shows a spherical shape with aggregated porous surface, but MNPs-CV (2b) has a relatively loose and fibrous surface, which indicated the penetration of dye molecules into the MNPs powder.
Th e FT-IR spectra of Fe 3 O 4 before and aft er adsorption were shown in Figure 3. Th e presence of strong broad absorption band at around 447-634 cm -1 shows the formation of magnetic nanoparticles. Th e absorption band at 447 cm -1 attributed to tetrahedral and octahedral sites and peak at 3385 cm -1 due to the O-H stretching adsorbed on the surface of the Fe 3 O 4 nanoparticles [11].
In the case of CV-Fe 3 O 4 nanoparticles (curve b in Figure 3); the adsorption of CV is established by the appearance of peaks at 2918 and 2843 cm -1 considered to be the stretching vibrations of -CH 3 . Th e appearance of peaks at 1538 and 1630 cm -1 are assigned to the C-C   Th e appearance of peak at 1025 cm -1 (C-N stretch) confi rms adsorption of CV (dimethylamino). Th e observed shift of bands to higher and /or to lower wavelength is taken as a strong evidence for the adsorption process ( Figure 3).

Adsorption Parameters
Eff ect of pH: When the pH of the dye solution increased, the amount of dye adsorbed also increases ( Figure 4). Th e adsorption of cationic dye is promoted due to electrostatic forces of attraction between the negatively charged adsorbent and the positively charged dyes molecules [12,13]. It is noticeabe that high removal effi ciency even at neutral pH (74.75%, 89.23% and 92.23%) for concentrations of 50, 25 and 10, respectively.
Proposed mechanism for adsorption: Based on the result obtained from the eff ect of pH; at high pH, an oxide surface will probably exist and this will increase the negative charge, while with deceasing pH the hydroxide form will instead of oxide form (positive polarization).    Th is confi rms that the adsorption process is promoted as electrostatic attraction mechanization at higher pH.

Eff ect of sorbent dose
From Figure 6, the removal effi ciency of investigated dye increases rapidly with increasing amount of MNPs. At the same time, further increasing the adsorbent dose leads to a slightly decrease of the removal effi ciency. Th e removal rate (%) decreases with increase in concentration and takes prolonged period to reach equilibrium because of the verity that with increase in dye concentration, there will be increase in contest amongest the dye molecules and the adsorption process will increasingly slowing down [14].

Eff ect of dye concentration
Th e eff ect of the concentration of the dye adsorption process was investigated (10-100 mg l -1 at 25 o C and pH 9). Th e concentration of the dye at max was obtained using a standard calibration curve. Th e adsorption is very rapid in the initial stages of the adsorption until concentration of 50 (higher than 90%) and then decreased for higher concentrations (Figure 7). Th is can be clarifi ed by a large number of active centers at the starting of adsorption and saturation of these centers on the surface of the adsorbent with attaining equilibrium. Th e requisite time for reaching the equilibrium increases with increasing the concentration due to the fact that adsorption involves fi lm diff usion and internal diff usion [15].

Eff ect of temperature
Temperature is a point for the adsorption whether it is an endothermic or exothermic process. Th e solubility and chemical potential of dye is aff ected by increasing the temperature. Figure 8 represents the sorption of CV onto MNPs at various temperatures (25 o C to 95 o C) with a fi xed initial dye concentrations (10, 25 and 50 mg l -1 ). With increasing the temperature the rate of adsorption decreases that indicates the adsorption process is exothermic. Th e decrease in rate of adsorption with increasing temperature, this may be due to when the temperature increased the solubility of dye will be increased.

Adsorption isotherms
Th e sorption profi le of CV was specifi ed over a wide range of equilibrium concentrations (10-100 ppm) as shown in Figure 9.

Adsorption isotherms
Langmuir isotherm: Langmuir equation used is given below: Where: (C e ; equilibrium concentration of dye), (q e ; the amount of dye adsorbed at equilibrium), (q m ; Langmuir monolayer adsorption capacity (mg/g)), and (b; Langmuir constant (l/mg)). Th e equation is linearized by diff erent ways to give the next equation: Th e value of adsorption capacity of MNPs was compared with the adsorption capacities of various adsorbents (Table 1). According to that comparison, MNPs have a good adsorption capacity for CV

Freundlich isotherm
It is oft en used to depict non-specifi c adsorption that comprises of heterogeneous surfaces.
Th e equation is linearized by diff erent ways to give the next equation: ln qe = ln k f + (1/n) ln C e (4) Where: (k f ; Freundlich coeffi cients (m/g)) and (n; related to adsorption intensity) were obtained from the slope and the intercep of the linearized Freundlich plots.
Th e plot of ln C e versus ln q e of CV in Figure 11 is straight line over the entire concentrations of dye in the aqueous phase. Th e values of k f and 1/n computed from the intercept and slope of the plot were equal to 1.191 and 0.84 for CV. Th e amount of 1/n < 1 indicates that the sorption strength is slightly miniature at lower equilibrium concentration and the isotherm does not predict any saturation of the solid surface of the adsorbent by the adsorbate [16,17].

Temkin isotherm
It assumes that the heat of adsorption of all the molecules in the layer decreases linearly with coverage due to adsorbate species adsorbent interactions, and adsorption is described by a regular distribution of binding energies, up to some maximum binding energy [18]. q e = q m ln (t + C e ) (5) Th e equation is linearized by diff erent ways to give the next equation: q e = q m ln t + q m ln C e (6) Where: (q e ; the concentration of CV removed (mg/g)), (t; the equilibrium binding constant (l/g)) and (q m ; constant related to heat of adsorption (J/mol).
Th e linear plot of q e versus ln C e for both the adsorption system gave good fi t for the Temkin isotherm as shown in Figure 12. Th e computed values of q m and t from the slope and intercept were equal 3.669 and 1.103 for CV dye, respectively.

Dubinin-Radushkevich isotherm (D-R)
It is used to evaluate the characteristic porosity of the adsorbent and the apparent energy of adsorption. It is illustated by the equation:     q e = q m e (-DƐ2) (7) Th e equation is linearized by diff erent ways to give the next equation: Ln q e = ln q m -DƐ2 (8) Ɛ = RT ln (1+1/C e ) (9) Where: (q e ; amount of CV adsorbed on MNPs), (q m ; highest adsorption capacity mg/g), (D; constant related to the energy of adsortion (mol 2 KJ 2 ), (C e ; equilibrium concentration of CV in mg/l) and (ɛ; RT ln (1+1/C e ), which is called as Polanyi Potential).
A plot of ln q e versus ɛ 2 as shown in Figure 13 Table 2.
Th e results obtained from the linear and the nonlinear regressions with Standard Error of the Estimate function were compared. It was found that the regression methods revealed that CV adsorption were better fi tted to the Freundlich isotherm (nonlinear regression) in terms of the R 2 and error values, because of the higher R 2 values and lower error values than those of other models.

Adsorption Kinetics
Th e dynamics of the adsorption of CV onto MNPs were inspected using the Lagergren's pseudo-fi rst order (Eq. 10) and pseudo-second order (Eq. 11) equations: Log (q e -q t ) = Log q e -k 1 t / 2.303 (10) t / q = 1/ K 2 q e 2 -[1 / q e ] t (11) Where: (q e and q t ; the amount of dye adsorbed at equilibrium and at any time t, respectively), (k 1 ; the fi rst-order adsorption rate constant (1/min)), (k 2 ; the pseudo-second order rate constant (g/mg min)). Th e K 1 and K 2 have been calculated from the intercept of the corresponding of log (q e -q t ) versus t as shown in Figure 14 and t/q versus t shown in Figure 15, and are tabulated in Table 3.
Th e correlation coeffi cient value (R 2 ) for the pseudo-second order rate equation was higher than the pseudo-fi rst order rate equation, so through these results, it was assured that the adsorption system pursued the pseudo-second order rate equation.

ΔG = Δ H -TΔS
Th e values of the thermodynamic parameters reported in Table 4 were obtained from the linear plot of ln K versus 1/T ( Figure 16). Th e positive value of ΔH indicates the endothermic nature of adsorption. Th e negative amount of ΔG indicates the spontaneous nature of the adsorption of CV onto MNPs. Th e positive amount of ΔS shows the increased disorder and randomness at the solid solution interface during the adsorption dye on the adsorbent [19][20][21] . Th e magnitude of ΔG, ΔH and ΔS vales indicate that the adsorption of CV onto MNPs is nonspontaneous at high temperature and spontaneous at low temperature. It can be said that the adsorption of CV MNPs is physical adsorption and the forces can be Van deer Waals forces or electrostatic attraction between CV and MNPs surface.

Conclusion
Th e present study highlights the potential application of iron oxide nanoparticles to remove the crystal violet dye. Th e adsorption of CV was found to increase with increase in time, decrease temperature, decrease dye concentration and increase pH up to equilibrium amont. It was found the adsorption process of CV on MNPs was exothermic and spontaneous. A pseudo-second order equation well explained the kinetic data and revealed the physisorption.