Adsorption of reactive red 2BF onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles fabricated via the ethanol solution of nitrate combustion process

Magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles were fabricated via the ethanol solution of nitrate combustion process with absolute ethanol as fuel and solvent [13]. Typically, 6.89 g Fe(NO₃)₃ · 9H₂O, 1.26 g Zn(NO₃)₂ · 6H₂O, 0.75 g Ni(NO₃)₂ · 6H₂O, and 0.49 g Co(NO₃)₂ · 6H₂O were placed into 30 ml absolute ethanol. The ethanol solution of the nitrates was stirred until the homogeneous solution was formed. Then, the crucible with mixed solution was ignited. As the crucible naturally cooled, the crucible in which gel had been formed was put into the program-controlled furnace following 2 h calcination at 500 °C to form nanoparticles.

XRD was applied to characterize the nanoparticles' phase identification by means of Cu–Kα radiation, the magnetic property was measured using a VSM, the SEM image was obtained to analyze morphology of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles, and their microstructure was investigated by the transmission electron microscopy.

20 ml RR-2BF aqueous solution in range of 100–400 mg · l⁻¹ and 0.05 g Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles fabricated at 500 °C for 2 h with 30 ml absolute ethanol in glass bottles were shaken with the speed of 100 rpm at ambient temperature. After that, the glass bottles were pulled out at designed intervals, and the magnetic field was applied to separate the Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles. The residual RR-2BF concentrations in solutions were measured using a UV–vis spectrophotometer. If the flasks were taken out for enough time, the adsorption of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles for RR-2BF would achieved equilibrium, and then the equilibrium adsorption capacities were obtained.

The adsorbance values of RR-2BF adsorbed onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles were determined by the following equation (1) [22].

$$\begin{eqnarray}&&q=\displaystyle \frac{V({C}_{0}-C)}{m}\end{eqnarray} \tag{ 1 }$$

wherein q represented the adsorbance in mg g⁻¹, V represented the volume of RR-2BF solution in L, C and C₀ (mg · l⁻¹) were the RR-2BF concentrations at the predetermined and initial time, and m referred to the weight of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles in g.

The characteristics for magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles fabricated at 500 °C for 2 h with 30 ml absolute ethanol were shown in figure 1. The SEM morphology and TEM image for Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles were revealed in figures 1(a) and (b), and the particle size of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles was about 30 nm. From the XRD pattern of as-fabricated nanoparticles, the observed peaks at 62.85°, 57.24°, 53.70°, 43.26°, 35.60°, and 30.22° could be assigned to the crystal plane of (440), (511), (422), (400), (311), and (220), respectively. It is universally acknowledged that the sharp and intense peaks suggested that the products were well-crystallized. At the same time, it seemed that the dopings with Co²⁺ and Zn²⁺ ions had no obvious effects on the cubic spinel crystal construction of NiFe₂O₄ (JCPDS No. 03-0875). For Co²⁺ and Zn²⁺ doped samples, no corresponding peaks for cobalt and zinc compounds appeared, which indicated that Co²⁺ and Zn²⁺ ions were effectively inlayed into crystal lattices of NiFe₂O₄ and formed a stable Ni_0.3Co_0.2Zn_0.5Fe₂O₄ crystal phase. Figure 1(d) showed the hysteresis loop of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles. Obviously, the magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles with specific magnetization (Ms) of 203.2 emu · g⁻¹ showed the typical soft magnetization behavior.

Zoom In Zoom Out Reset image size

Before exploring the adsorption mechanism, we first selected to focus on the appropriate dosage of adsorbent. The adsorption capacity of the adsorbent with different dosages in 20 ml RR-2BF solution (200 mg l⁻¹) was examined, as displayed in figure 2. Obviously, when the amount of adsorbent was greater than 0.05 g, the adsorbance showed a downward trend. This was for the reason that the magnetic properties of the nanoparticles were very strong. When the volume of dye solution remained constant, the more magnetic nanoparticles, the worse their dispersion and the more likely they were to aggregate with each other. Moreover, the surface of the agglomerated nanoparticles could not fully contact with the dye molecules, so the adsorption capacity decreased. As a result, in the adsorption process, 0.05 g adsorbent was selected for the test.

Zoom In Zoom Out Reset image size

The solution pH value was a key condition that could affect the adsorption process. In order to explore the adsorption mechanism, 0.05 g magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles were weighed to adsorb 20 ml dyes of 200 mg/l with pH of 2, 4, 6, 8 and 10. The results were shown in figure 3. As the increase of pH, the decreased adsorption capacity could be observed. The reason may be that RR-2BF was an anionic dye, and the material was positively charged. When the nanoparticles were dispersed in the reactive red dye, there was electrostatic attraction between the material and the RR-2BF molecules, which resulted in the adsorption of outer surface of the nanoparticles for the RR-2BF molecules. As the pH value was changed from acidity to alkalinity, the amount of OH⁻ in the solution increased gradually, and nanoparticles had adsorption effect on both OH⁻ and RR-2BF molecules. However, the competitive adsorption of the two made the competitive ability of RR-2BF molecules weakened gradually with the increase of pH. As the active sites on the nanoparticles were occupied by OH⁻, the adsorbance of RR-2BF molecules onto the nanoparticles decreased gradually when the solution pH value increased.

Zoom In Zoom Out Reset image size

It is significant to investigate adsorption rate of adsorbate onto the adsorbent for the application of the nanoparticles in wastewater treatment. The kinetics models of the pseudo-first-order, pseudo-second-order and intraparticle diffusion were often employed to measure the process of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles adsorbing RR-2BF.

These three kinetics models were listed as equations (2)–(4), respectively [23, 24].

$$\begin{eqnarray}&&{q}_{t}={q}_{e}\left(1-{e}^{-{k}_{1}t}\right)\end{eqnarray} \tag{ 2 }$$

$$\begin{eqnarray}&&{q}_{t}=\displaystyle \frac{{q}_{e}^{2}{k}_{2}t}{1+{q}_{e}{k}_{2}t}\end{eqnarray} \tag{ 3 }$$

$$\begin{eqnarray}&&{q}_{t}={x}_{i}+{k}_{i}{t}^{\tfrac{1}{2}}\end{eqnarray} \tag{ 4 }$$

Wherein q_e and q_t (mg · g⁻¹) referred to quantities of RR-2BF adsorbed onto the magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles at equilibrium and an arbitrary time, correspondingly; k₁ (min⁻¹), k₂ (g · mg⁻¹ · min⁻¹) and k_i (mg · g⁻¹ · min⁻¹) referred to the adsorption rate constants for three models, accordingly; x_i referred to the boundary layer thickness.

Figure 4 displayed the adsorbances (q_t) at the arbitrary times when the initial concentrations of RR-2BF were 100–400 mg · l⁻¹. Notably, the adsorbances of RR-2BF on Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles rapidly ascended in the initial adsorption phase, and continuously ascended with the stretch of time until a state of equilibrium was achieved. This process was mainly affected by the amount of dye molecules, for Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles surface might be exposed to outer active surface sites of the nanomaterial occupied by the dye. When the nanoparticles initially contacted RR-2BF molecule, the active sites were sufficient, so the equilibrium capacity for RR-2BF adsorption increased rapidly. When the contact time was gradually extended, the dye concentration was relatively reduced, and the adsorption site on the surface of the material was also occupied, so the adsorption capacity increased slowly. Therefore, the effect of time on the adsorption process of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles was mainly reflected in the early stage. When the adsorption site on the material under the current dye concentration was in a condition of equilibrium, the saturation state at equilibrium occurred, and the adsorption state would hardly change with time.

Zoom In Zoom Out Reset image size

The fittings of three kinetics models for RR-2BF onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles with dye concentrations ranging from 100 to 400 mg · l⁻¹ were shown in figure 5. The experimental data were calculated, and the corresponding kinetic parameters were listed in table 1. The calculation results suggested that the pseudo-second-order model had the largest variance (R²) values; while, the models of intraparticle diffusion and pseudo-first-order had lower R² values, which illustrated poorer fitted results. So, the pseudo-second-order kinetics model could accurately demonstrate the adsorption behavior of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles for RR-2BF. On the other hand, the adsorption kinetics in the whole process of adsorption was mainly controlled by chemical action rather than by the material transfer step. And the line relations of t/q_t with t were shown in figure 6 with the large variances closed to 1.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The interrelation of the nanoparticles and RR-2BF molecules was significant, and the significance of adsorption isotherm was self-evident. So, the adsorption equilibrium data of RR-2BF onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles were fitted by the two-parameter models. The isotherms parameters could be determined by Langmuir, Freundlich and Temkin in this research.

Langmuir model presumed that the adsorbate was uniformly single-deck coverage on the adsorbent surface, however, the surface heterogeneity of adsorbent was often overlooked, and the adsorption only took place at the uncovered sites of adsorbent, with the uncovered site being defined as the active site. Langmuir model assumed that the active sites on the adsorbent surface had an average energy; therefore, the adsorption energy for Langmuir isotherm was considered constant, and its expression was shown in equation (5). The surface heterogeneity effect of adsorbent was raised by empirical equation of Freundlich model, which assumed that the binding strength was inversely proportional to the binding site after the binding site was first occupied. Therefore, Freundlich model, which could be expressed in formula (6), ulteriorly considered multilayer adsorption was accompanied by the interactions of adsorbed molecules and active sites with heterogeneous energetic distribution. Temkin model considered the interactions of the adsorbent and the adsorbate; it assumed that the distribution of the adsorption binding energy was uniform, the heat from adsorption decreased linearly with the coverage of adsorbent, and the binding energy finally reached maximum. And Temkin adsorption isotherms was expressed by the equation (7) [25, 26].

$$\begin{eqnarray}&&{q}_{e}=\displaystyle \frac{{q}_{\max }{K}_{L}{C}_{e}}{1+{K}_{L}{C}_{e}}\end{eqnarray} \tag{ 5 }$$

$$\begin{eqnarray}&&{q}_{e}={K}_{F}{C}_{e}^{\tfrac{1}{n}}\end{eqnarray} \tag{ 6 }$$

$$\begin{eqnarray}&&{q}_{e}=B{\rm{In}}\left({A}_{T}{C}_{e}\right)\end{eqnarray} \tag{ 7 }$$

Wherein q_e (mg · g⁻¹) referred to the weights of RR-2BF adsorbed onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles; The concentration of RR-2BF at equilibrium was expressed in terms of C_e (mg · l⁻¹); K_L (l · mg⁻¹), B (=R_T/b_T), A_T (L · g⁻¹) , and K_F (mg^1−(1/n) · l^1/n · g⁻¹) referred to the constants for three models; 1/n (0–1) referred to the dimensionless factor, which reflected the surface heterogeneity or adsorption intensity.

When the adsorption rate of RR-2BF onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles was equal to desorption, the adsorption equilibrium data were obtained to fit three isotherms models, and the results were revealed in figure 7. With the increase of the initial concentration, the adsorption capacity of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles for RR-2BF showed a trend of first increasing and then decreasing. This phenomenon was also related to the limited adsorption sites on the surface of nanoparticles. Table 2 summarized the calculated parameters for three isotherms models. Temkin model yielded the largest variance (R² = 0.9953) compared the variance values for three models with each other. However, the variances for Langmuir and Freundlich models (R² = 0.9831, 0.9698) showed lower values in comparison with Temkin isotherm. Thus, they were not fit to demonstrate the equilibrium process of RR-2BF adsorbed onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles. Based on the assumption from theoretical model of Temkin, the adsorption of RR-2BF onto Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles was of hybrid absorbing mechanism caused by multilayer adsorption and monolayer adsorption.

Zoom In Zoom Out Reset image size

Figure 8 revealed the regeneration of Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles for adsorption of RR-2BF in aqueous solution. Due to the collapse of nanoparticle pores caused by cyclic calcination, the active sites were subsequently reduced. Therefore, the figure elucidated that the amounts of RR-2BF adsorbed on adsorbent decreased with the increase of the number of cycles. The magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles having adsorbed RR-2BF for 8 times had a capacity of 53.7 mg · g⁻¹, which exceeded 75% of the initial equilibrium capacity, indicating that the magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles had superior regeneration ability and could be widely applied in dye adsorption in the future.

Zoom In Zoom Out Reset image size

The adsorption capacity of different adsorbents to dyes was investigated, and the results were displayed in table 3. The equilibrium capacity of magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles was at a medium level, and they could be reused for many times with higher adsorption amount. The magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles, as an excellent adsorbent, could rapidly achieve a state of equilibrium in a short time, and the adsorbent could be recycled in a way of calcination and pH adjustment to reduce secondary pollution. Therefore, magnetic Ni_0.3Co_0.2Zn_0.5Fe₂O₄ nanoparticles was a promising dye adsorbent.