1 Introduction

Dye is one of the major sources of environment pollutions. In cosmetics, food, leather, paper, and textile industries various dyes are frequently applied as colorants to process products and factories often discharge large amounts of contaminants containing dyes into the aquatic ecosystem. The presence of these dyes in ambient environment threats the environmental sustainability due to massive discharge of wastewater. Dyes’ wastewater will cause allergic, carcinogenic, mutagenic and toxic hazards to biological organisms, which is even harmful to the survival and development of plants, animals, and humans [1, 2]. Hence, it is urgent to remove dye from industries’ effluents. The dye uptakes in wastewaters have always been a troublesome problem for the environment protection. The treatment for dyes’ wastewaters is difficult due to complex and varied water bodies that contain large quantity of organic and inorganic dyes [3]. Various techniques for removing dyes from the wastewater have been developed, the most common techniques of dye treatment include coagulation, oxidation, adsorption, reverse osmosis, and so forth. In addition to the employ of organic solvents and cumbersome operation, these techniques can also cause damage to equipment, resulting in costly repair costs. Nevertheless, the adsorption as a treatment technology of industrial wastewater has displayed effective and promising result, because of its advantages of high efficiency, simple control process, cost-effectiveness, and environmental-friendliness [4, 5].

Various adsorbents, such as peanut shell [6], chitosan [7], and mesoporous activated carbon [8], have been intensively investigated for the removal of dyes. They involve defects of the high treatment costs, sluggish kinetics, low adsorption capacity, and so on, thus limiting further applications. In order to enhance the removal efficiency, various nanoadsorbents are studied and applied to adsorb dye with the development of nanomaterials, and they represent a high adsorption capacity [9, 10]. Compared with nonmagnetic adsorbents, magnetic nanomaterials that are easy to separate from aqueous solutions, have been gaining attention as a recyclable absorbent. Most of the previous studies indicate that the magnetic adsorbents have many advantages such as excellent mechanical properties, large specific surface area, facile separation, and high recycling performance. The numerous active sites of magnetic nanomaterials make the theoretically extraordinary application potential in adsorption. It is worthy that the magnetic nanoabsorbents are considered as a highly efficient adsorbent for organic and inorganic substances [11].

As adsorbent to remove dyes, magnetic MgFe2O4 nanoparticle is one of the significant nanomaterials due to its excellent adsorption performance. Many approaches have been developed to fabricate MgFe2O4 nanoparticles, such as the autoclave route [12], the hydrothermal process [13, 14], the sol–gel method [15], the glycine–nitrate method [16], etc. The preparation procedures of the above approaches have been limited by the long pretreatment processes, use of toxic solvents, expensive devices. However, the alcohol combustion technique is introduced as a convenient, facile, and novel method, which provides the advantages of low cost, short preparation period, homogeneous products, and free of dispersants [17,18,19,20].

In this project, we have successfully prepared the magnetic MgFe2O4 nanoparticles via the facile alcohol combustion process, and the adsorption characteristics and the electrochemical performance of reactive red (RR) onto them was investigated.

2 Experimental details

2.1 Preparation and characteristics of magnetic MgFe2O4 nanoparticles

The magnetic MgFe2O4 nanoparticles were prepared via the alcohol-assisted combustion process. Typically, analytical grade magnesium nitrate and iron nitrate were used, and the molar ratio of Mg and Fe was 1:2. a mixture of 2.56 g of Mg(NO3)2·6H2O and 8.08 g of Fe(NO3)3·9H2O were dispersed in 20 mL of absolute ethyl alcohol, the mixture was maintained at room temperature with constant stirring for 1 h to form the homogeneous solution, which subsequently was put into a crucible and ignited. Ignition combustion process replaced the process of forming xerogel in traditional sol–gel method, which significantly reduced the preparation time [21]. After the flame was extinguished, the gel was obtained in the crucible. Crucible with obtained gel was placed in the calciner with the heat rate of 3 °C/min; it was kept at 600 °C for 2 h to form MgFe2O4 nanoparticles that were applied as adsorbent for RR removal. X-ray diffraction (XRD) pattern of the as-synthesized nanoparticles was collected with Holland Philips X’Pert X-ray diffractometer (Cu Ka radiation, λ = 1.5406 Å, 30 mA and 40 kV) for 2θ over 20°–80°. The morphology and structure studies were performed by SEM and TEM techniques. Magnetic saturation experiment was carried out using a vibrating sample magnetometry (VSM) with a maximum magnetic field of 10 kOe. The Brunauer–Emmett–Teller (BET) surface area via nitrogen adsorption was determined with the instrument of NOVA 2000e.

2.2 Adsorption performance of RR onto MgFe2O4 nanoparticles

Adsorption experiments were performed in a batch of 50 mL flasks and the flasks were shaken for the required contact time. In kinetics experiment, an average of 0.05 g of magnetic MgFe2O4 nanoparticles was added to RR aqueous solution (20 mL) with the initial concentration of 100–400 mg/L. The suspensions were agitated. Then, the flasks were taken out at predetermined time intervals, the adsorbents were separated under the magnetic field and the residual RR amounts in the solution was determined by measuring the absorbance. To study the adsorption isotherms, the experiments were carried out by mixing 0.05 g of magnetic MgFe2O4 nanoparticles and 20 mL RR with various concentrations in flasks, these flasks were maintained at room temperature and agitated. The supernatant was collected after reaching equilibrium. The absorbance of RR at 530 nm was investigated by UV spectrophotometer (UV-2250).

The adsorption capacity of the MgFe2O4 nanoparticles for RR could be calculated by the following [22]:

$$q = \frac{{\left( {C_0 - C_t} \right) \cdot V}}{m},$$
(1)

where C0 (mg/L) and Ct (mg/L) referred the initial concentration of RR solution and the concentration for any time, respectively; m represented the weight of MgFe2O4 nanoparticles (g), V was the volume of RR solution.

2.3 Electrochemical measurement

The electrodes were constructed using a three-electrode system. The platinum wire and Ag/AgCl2 were used as counter and reference electrode, respectively. During preparation process of working electrodes, magnetic MgFe2O4 nanoparticles and post adsorption nanoparticles were mixed in aqueous solution at concentration of 1 mg/mL. The 2 μL solution was coated in the center of the magnetic glassy carbon electrode (MGCE), and the samples were placed in a vacuum oven to dry at 30 °C. The bare MGCE, as-prepared magnetic MgFe2O4 nanoparticles and post adsorption nanoparticles were applied as work electrode to measure the electrochemical property. The experiments of cyclic voltammetry (CV, potential range from −0.2 to 0.6 V, scan rate of 0.1 V/s) and electrochemical impedance spectroscopy (EIS, frequency range of 1–104 Hz, amplitude of 5 mV) were performed using an electrochemical workstation (CHI660E).

3 Results and discussion

3.1 Characterization of MgFe2O4 nanoparticles

The characteristics of magnetic MgFe2O4 nanoparticles calcined at 600 °C for 2 h with absolute alcohol of 20 mL were evinced in Fig. 1. Figure 1a showed the SEM micrograph of magnetic MgFe2O4 nanoparticles’ microstructure, and it was clear that the average diameter of nanoparticles was around 120 nm and the size distribution was comparatively uniform, however there was a certain degree of aggregation between the particles. Figure 1b illustrated the TEM image of MgFe2O4 nanoparticles, obviously, the MgFe2O4 nanoparticles had a polycrystalline structure, and the grain size was around 120 nm, which was in accordance with the SEM morphology. The XRD pattern of MgFe2O4 nanoparticles was displayed in Fig. 1c, when the calcination temperature was at 600 °C for 2 h, the diffraction peaks of MgFe2O4 nanoparticles matched well with (220), (311), (222), (400), (422), (511), (440), (620), (533), and (444) planes, the result indicated that crystal structure-obtained products were single-phase spinel MgFe2O4 (JCPDS No. 88-1936). The average crystallite size (D) of MgFe2O4 nanoparticles could be calculated from the value of full-width (FWHM) at half-maximum of (311) plane by Scherrer’s formula [18]. The average crystallite size of MgFe2O4 was determined to be 32 nm, which was smaller than particle sizes observed in TEM and SEM images. The reason was that MgFe2O4 nanoparticles were formed by the aggregation of multiple crystals. The hysteresis loop of the MgFe2O4 nanoparticles was exhibited in Fig. 1d, it could be observed that the saturation magnetization (Ms) of magnetic MgFe2O4 nanoparticles was 176.2 Am2/kg with a typical soft magnetization behavior.

Fig. 1
figure 1

SEM (a), TEM (b), XRD (c), and VSM (d) of magnetic MgFe2O4 nanoparticles calcined at 600 °C for 2 h with absolute alcohol of 20 mL

Full size image

The BET measurement for the magnetic MgFe2O4 nanoparticles was shown in Fig. 2. It could be seen that the curve belonged to the IV nitrogen adsorption isotherm whose adsorption curve was not coincidence with desorption curve, thus forming a hysteresis loop. This nitrogen adsorption–desorption isotherm exhibited the adsorption and desorption behavior of porous materials. The specific surface area was calculated as 33.4 m2/g, and the estimated pore size distribution was shown in the inset of Fig. 2, it could be observed that the pore size was concentrated at 4 nm, which led a larger specific surface area. Both large surface area and highly porous structure provided more adsorption sites for the material, which were beneficial for dye removal.

Fig. 2
figure 2

Nitrogen sorption isotherm of magnetic MgFe2O4 nanoparticles calcined at 600 °C for 2 h with absolute alcohol of 20 mL

Full size image

3.2 Adsorption of RR onto magnetic MgFe2O4 nanoparticles

3.2.1 Adsorption kinetics

In order to better design the treatment process from sewage, it is significant to evaluate the removal efficiency of pollutants in aqueous solution. The experimental data obtained from RR kinetic experiments were fitted using pseudo-first-order model [7], pseudo-second-order model [23], and the intraparticle diffusion model [24, 25], and the adsorption mechanism of RR adsorption onto magnetic MgFe2O4 nanoparticles was explicated. The three equations of kinetic model were represented as follows:

$$q_t = q_e\left( {1 - e^{ - k_1t}} \right),$$
(2)
$$q_t = \frac{{q_e^2k_2t}}{{1 + q_ek_2t}},$$
(3)
$$q_t = c_i + k_it^{1/2},$$
(4)

where k1, k2, and ki represented the rate constants of the pseudo-first-order model, the pseudo-second-order kinetic model, and the intraparticle diffusion models, qe (mg/g) was the adsorbed RR amounts at equilibrium state, qt (mg/g) was associated to adsorption capacity at any time, ci was boundary layer thickness.

Origin software was applied to fit all the models, and the accurate theoretical model was selected to describe the adsorption behavior of RR onto magnetic MgFe2O4 nanoparticles based on the correlation coefficient (R2) values as indicator, which better represented the experimental data. The experimental data of RR solution at different concentrations were respectively fitted by the pseudo-first-order, the pseudo-second-order, and the intraparticle diffusion models. The kinetic parameters of the relevant fitting models were summarized in Table 1. The fitting effects of the models were evaluated by correlation coefficients (R2). It could be observed that R2 value of pseudo-second-order kinetic exceeded 0.98 at diverse RR concentrations (100, 200, 300, 400 mg/L). As shown in Fig. 3, The high value of R2 suggested that the kinetics of RR onto magnetic MgFe2O4 nanoparticles followed the pseudo-second-order kinetic model and hence not a diffusion-controlled phenomenon.

Table 1 Fitted kinetics parameters for adsorption of RR from aqueous solutions onto magnetic MgFe2O4 nanoparticles
Full size table
Fig. 3
figure 3

Pseudo-second-order kinetic model for adsorption of RR onto magnetic MgFe2O4 nanoparticles at various initial concentrations

Full size image

Figure 4 showed the experimental data at predetermined contact time and fitted curve of pseudo-second-order kinetics model under different concentrations of RR. As displayed in Fig. 4, the adsorption process would be affected by initial RR concentration. The adsorption capacity of RR onto magnetic MgFe2O4 nanoparticles drastically increased during the initial stage of adsorption and then increased at a slow rate until equilibrium was reached after 60 min, which might be due to more vacancies on the surface of the adsorbent at the initial stage. When the initial concentration of RR was 400 mg/L, the adsorption capacity could reach 96.9 mg/g. Figure 5 showed the linear fits of the pseudo-second-order kinetics model, and the relative errors for the experimental qe and qe calculated from the Eq. (3) were <0.4%, which implied that the experimental data followed the regularity of pseudo-second-order kinetic model and there was excellent linear relation between them. The pseudo-second-order adsorption kinetics model included all processes of adsorption namely, extra-granular diffusion, outer surface adsorption, intraparticle diffusion, and internal surface adsorption. MgFe2O4 nanoparticles had small particle sizes, large specific surface areas, and a certain surface energy, and the adsorption between the adsorbent, and the adsorbate was caused by an intermolecular force. The adsorption process of dye first occurred on the surface of MgFe2O4 nanoparticles by van der Waals force and hydrogen bonding, then partial dye molecules by anionic groups were electrostatically attracted with positively charged MgFe2O4 nanoparticles. In addition, ion exchange or other chemical forces may be present during the adsorption process. Therefore, the adsorption process had both physical adsorption and chemical adsorption.

Fig. 4
figure 4

Fits of the pseudo-first-order kinetics model, pseudo-second-order kinetics model and intraparticle diffusion kinetics model at initial RR concentrations of 100 mg/L (a), 200 mg/L (b), 300 mg/L (c), 400 mg/L (d)

Full size image
Fig. 5
figure 5

Linear fits of the pseudo-second-order kinetics model at initial RR concentrations of 100–400 mg/L

Full size image

3.2.2 Adsorption isotherms

Adsorption isotherm was vital to describe how adsorbent surface interacted with molecules of adsorbate. Therefore, the experimental data of RR adsorption equilibrium on MgFe2O4 nanoparticles was interpreted using theoretical equation. The models of Langmuir, Freundlich, and Temkin were applied for the analysis of equilibrium data.

The theoretical model of Langmuir isotherm assumed that the adsorbent surface was homogeneously covered by a monolayer of the adsorbate. The adsorption energy was uniformly dispersed on the adsorbent surface consisted of active sites and was constant, surface heterogeneity of the adsorbent was not taken into account in this model. The equation of Langmuir model was represented as [26]

$$q_e = \frac{{q_{\max }K_LC_e}}{{1 + K_LC_{\mathrm{e}}}},$$
(5)

where qe (mg/g) was the equilibrium adsorption capacity of RR onto magnetic MgFe2O4 nanoparticles, qmax (mg/g) gave an idea of the maximum adsorption capacity in theory, KL (L/mg) represented Langmuir constant related to the adsorption rate of the adsorbent.

The Freundlich adsorption isotherm presumed that the adsorption process of multilayer took place on heterogeneous surface of adsorbent. The multilayer adsorption of active sites with heterogeneous energetic distribution and interaction between adsorbed molecules should be further considered. This empirical equation was mostly applied to describe the adsorption of inorganic and organic components in solution, expressed by the following equation [27]:

$$q_e = K_{\mathrm{F}}C_e^{1/n},$$
(6)

where KF (mg·L1/n·mg−1/n·g−1) was isotherm constant that reflected the adsorption capacity. The 1/n value ranging from 0 to 1 represented dimensionless factor related to adsorption intensity.

The Temkin isotherm model took into account the effects of the interaction between adsorbent and adsorbate. As the coverage area of the adsorbate increased, the heat of the adsorbent encapsulated inside decreased constantly. and the binding energy was uniformly distributed during the adsorption process, reaching a certain maximum. The Temkin adsorption isotherm expression was expressed as follows [28]:

$$q_e = B\,{\mathrm{ln}}\left( {A_TC_e} \right),$$
(7)

where B (=RT/bT) was the Temkin constant related to adsorption heat, AT (L/g) was equilibrium binding constant corresponding to maximum binding energy, T (K) represented the absolute solution temperature, and R (8.314J·mol−1·K−1) was the universal gas constant.

Figure 6 exhibited the experimental plot and fitted curves of Langmuir, Freundlich, and Temkin models. The relevant parameters (KL, qmax, KF, 1/n, AT, and B) obtained from the three isotherm models were calculated and summarized in Table 2. By comparing the values of the correlation coefficients (R2) between the isotherm models, the Langmuir isotherm model could give the highest R2 value that exceeded 0.99. The Langmuir yielded the best fit (R2 = 0.9912). While, the R2 values of Freundlich and Temkin isotherm (R2 = 0.8634, 0.9374) were relatively lower, which indicated that the adsorption of RR onto magnetic MgFe2O4 nanoparticles could be best described by Langmuir model. The mechanism of Langmuir isotherm suggested that the adsorption process of the RR occurred at a specific site on the surface of the MgFe2O4 nanoparticles. When the homogeneous sites were occupied, the adsorption couldn’t produce in the same site.

Fig. 6
figure 6

The Langmuir, Freundlich, and Temkin isotherms for adsorption of RR onto magnetic MgFe2O4 nanoparticles at room temperature

Full size image
Table 2 Evaluated model parameters of adsorption isotherms for RR onto magnetic MgFe2O4 nanoparticles at room temperature
Full size table

3.3 Effect of pH on adsorption

The pH in solution was a crucial factor affecting the removal efficiency of adsorbents. Adsorption experiments were performed in the range of solutions pH from 5 to 9. Figure 7 showed the effects of pH on adsorption for MgFe2O4 nanoparticles calcined at 600 °C for 2 h with absolute alcohol of 20 mL, it could be seen that the adsorption capacity firstly improved until 7 of pH value and then dropped sharply with further increase of pH value. The phenomenon was caused by electrostatic interaction of MgFe2O4 nanoparticles and ionized dye. When the pH of solution was less than RR isoelectric point that was about 7, the positive charge of RR increased as the pH decreased. The electrostatic repulsion was enhanced between RR and positively charged MgFe2O4, which would cause a reduction in the removal efficiency of RR. When pH was greater than 7 in solution, hydroxide ions were generated and dye had negative charge. Meanwhile, the activated surface sites of MgFe2O4 nanoparticles would produce electrostatic attraction to hydroxide ions. The emergence of competitive adsorption limited the removal of RR between hydrogen ions and the dye. The above results indicated that the pH of solution was a vital driving force for whole adsorption phase, and the variation of pH disturbed the situation to which the ionized dye molecules transported electrostatic charges.

Fig. 7
figure 7

Effect of pH on the adsorption of RR onto magnetic MgFe2O4 nanoparticles

Full size image

The performance comparison in terms of adsorption capacities of the RR dye on various adsorbents was presented in Table 3. It could be seen that the adsorbent reported in this study had a relatively high adsorption capacity in comparison with all of the other adsorbents. Magnetic MgFe2O4 nanoparticles not only had higher adsorption ability for RR dye, but also had facile magnetic separation ability. Magnetization and availability of more active adsorption sites indicated that MgFe2O4 nanoparticles could be considered a convenient and effective adsorbent in wastewater treatment.

Table 3 Comparison of the adsorption capacity of RR dye on various adsorbents
Full size table

3.4 Magnetic response test

Figure 8 showed photograph of magnetic MgFe2O4 nanoparticles adsorption behavior and magnetic separation process. Figure 8a displayed RR solution with the concentration of 200 mg/L, while Fig. 8b showed that magnetic MgFe2O4 nanoparticles had good dispersion in RR solution, and Fig. 8c provided the magnetic separation image of MgFe2O4 nanoparticles from RR solution, which indicated strong adsorption effect of MgFe2O4 nanoparticles to RR, and the nanoparticles adsorbed dye could be effectively separated under the external magnetic field.

Fig. 8
figure 8

Photograph of magnetic MgFe2O4 nanoparticles adsorption behavior and magnetic separation process: 200 mg/L RR solution (a), RR solution with MgFe2O4 nanoparticles (b), magnetic separation of magnetic MgFe2O4 nanoparticles after adsorption of RR for 12 h (c)

Full size image

3.5 Electrochemical properties

In order to evaluate electrochemical properties of magnetic MgFe2O4 nanoparticles before and after RR adsorption. Electrochemical performance changes of bare MGCE, MGCE/MgFe2O4 and MGCE/post adsorption nanoparticles were further investigated. The CV analysis of the three samples were displayed in Fig. 9a, it could be observed that the peak currents of the bare MGCE were larger than the MGCE/post adsorption nanoparticles. While for the response on the MGCE/MgFe2O4 nanoparticles, the corresponding peak was widest and the peak currents were minimal. The curve of EIS consisted of arc and straight line, the radius of the arc represented the resistance degree to the electron transfer from the electrode. The linear part at region of low frequency was related to the limited rate step of diffusion [29]. As exhibited in Fig. 9b, the maximum radius of MGCE/MgFe2O4 nanoparticles could be seen, that clearly indicated that MGCE/MgFe2O4 nanoparticles had maximum resistance. Which was consistent with the CV consequence in Fig. 9a. These results confirmed that the RR successfully occupied the surface sites of the magnetic MgFe2O4 nanoparticles.

Fig. 9
figure 9

CV analysis (a) and EIS (b) of bare MGCE, MGCE/MgFe2O4 and MGCE/post adsorption nanoparticles

Full size image

4 Conclusions

(1) Magnetic MgFe2O4 nanoparticles were prepared successfully via the facile alcohol combustion process, and the magnetic MgFe2O4 nanoparticles calcined at 600 °C for 2 h with 20 mL of absolute alcohol were characterized with average particle size of around 120 nm and the specific magnetization of 176.2 Am2/kg.

(2) The kinetic parameters of pseudo-first-order, the pseudo-second-order, and the intraparticle diffusion models for various initial concentrations from 100 to 400 mg/L were calculated. The adsorption process of RR from aqueous solutions onto the magnetic MgFe2O4 nanoparticles followed pseudo-second-order model.

(3) Analysis of equilibrium data indicated that the Langmuir model could describe the adsorption mechanism of RR onto the magnetic MgFe2O4 nanoparticles, which was dominated by monolayer adsorption.