Synthesis of Superparamagnetic Fe₃O₄ Nano-Adsorbent Using an Energy-Saving and Pollution-Reducing Strategy for the Removal of Xylenol Orange Dye in Water

Abstract

The superparamagnetic Fe₃O₄ nanoparticles as an absorbent with a size distribution of 4.8–6.4 nm were synthesized using a simple one-pot hydrothermal strategy at 200 °C for 24 h, where iron citrate and distilled were the sum total of raw materials. The as-synthesized Fe₃O₄ powders showed rapid and efficient adsorption for xylenol orange with a saturated adsorption amount of 42.5 mg/g according to Langmuir linear fitting, and the adsorption reaction between xylenol orange adsorbate and Fe₃O₄ adsorbent was mostly completed within 10 min. The Fe₃O₄ nanoparticles not only had superparamagnetism with a saturation magnetization value of 54.9 emu/g at 15 kOe but also possessed strong magnetic response, making them easy to separate easily from aqoeous solution under the attraction of magnet. In this work, the Fe₃O₄ particles can be totally attracted toward the magnet within 15 s, leaving the suspension a clear solution.

1. Introduction

Energy development and environmental protection have always been the common concern of people because energy is the important material base for the life and development of human society, and the environment is the fundamental condition for human survival [1,2,3,4]. Water pollution caused by dyes is a serious disposal problem because most of the dyes in water are highly visible and undesirable. The dyes can severely interfere with the absorption and reflection of sunlight entering the water, which affects the replication and producing of dye-degrading bacteria; as a result, these dyes are not biodegradable in the water [5,6,7,8,9]. Therefore, removal of such colored dye contaminants from polluted aqueous media has become an impendent issue. To solve this problem, a number of available technologies for the removal of dyestuffs have been developed and implemented, such as physical [10,11], chemical [12,13], electrical [14,15], and biological [16,17] strategies. Among these technologies, the adsorption technology has been regarded as one of the most competitive methods for wastewater treatment in terms of its flexibility and simplicity of design, operational ease, and insensitivity to pollutants; moreover, no harmful substances are formed during the adsorption reaction [18,19,20].

Many absorbent materials have been developed to remove dyes from aqueous solution, such as graphene oxide nanocomposites [21], activated carbon [22], ultrafiltration membrane [23], natural fiber [24,25], and magnetic nanocellulose [26]. Among all available candidates, the nanostructured Fe₃O₄ particles are attractive because of their superparamagnetism, which facilitates the targeting control and recyclable separation just using an applied magnetic field. This superparamagnetism of Fe₃O₄ particles is beneficial to the simple and rapid separation of Fe₃O₄ after adsorption of pollutants, which can not only save the cost, but also shorten the operation cycle. In addition, the size effect of nanomaterials makes them easier to capture the pollutant due to enhanced surface activity [27,28]. So far, numerous methods have been reported to synthesize magnetic Fe₃O₄ nanoparticles. For example, Chen et al. [29] synthesized the superparamagnetic Fe₃O₄ nanofibers with hollow characteristic morphology based on a polymer-assisted thermochemical reduction process, and the diameter and wall thickness of these nanofibers were ranging from 100 to 200 nm and 15 to 25 nm, respectively. Yang et al. [30] synthesized the magnetite (Fe₃O₄) nanoparticles with different shapes and sizes by the thermal decomposition method. Liu et al. [31] also synthesized Fe₃O₄ nanoparticles with a diameter of ~500 nm by a hydrothermal method using FeCl₃·6H₂O, ethylene glycol and CH₃COONa as the main raw materials. Moreover, Eskandari et al. [32] synthesized the Fe₃O₄ nanoparticles (6 nm) based on a chemical co-precipitation method with an alternating magnetic field and ultrasonic-assisted. Although these methods could synthesize the nanostructured Fe₃O₄ particles with superparamagetism, but the synthesis process is still not the easiest, as well as the used raw materials. It is still challenging to further simplify the operation process to reduce costs and energy consumption.

Xylenol orange (XO) is a common synthetic dye, also usually used as potentiometric reagent and complexometric indicator for the determination of metal ions in the science laboratory. Therefore, the XO effluents from manufacturing industries and laboratories will further attract heavy metal ions, finally leading to various ailments to the living body. To date, only a few studies had reported on the removal of XO from the aqueous solution. Hyperbranched polyethyleneimine based gels [33] and polyvinyl alcohol/cellulose nanocrystals hydrogels [34] were developed as absorbent materials for the removal of XO from the aqueous solution. However, the process of collecting these used sorbents from water was cumbersome after the adsorption reaction. For that, in this work, a facile one-pot hydrothermal procedure was developed for the synthesis of superparamagnetic Fe₃O₄ nanoparticles, in which iron citrate and distilled were the sum total of raw materials. The main advantages of hydrothermal method are as follows. The cheap and readily available water served as the solvent during the hydrothermal method, which could synthesize the desired products under medium and low temperature conditions without subsequent high-heat treatment. The closed conditions of hydrothermal process could reduce the emission of toxic and harmful gases, effectively reducing environmental pollution, in line with the requirements of energy-saving and emission reduction. Moreover, the as-synthesized superparamagnetic Fe₃O₄ nanoparticles could serve as an adsorbent for the removal of XO, as a comparison, the adsorptions of other dyes, Basic Orange 2 (BO2) and Acid Orange 7 (AO7) were also investigated.

2. Experimental

2.1. Materials

Iron citrate (AR) was obtained from Shanghai Yien Chemical Technology Co., Ltd., Xylenol Orange (XO, 98%) dye was obtained from Shanghai Bide Medical Technology Co., Ltd., Basic Orange 2 (BO2, AR) dye was obtained from Shanghai Maclin Biochemical Technology Co., Ltd., Acid Orange 7 (AO7, >97.0%) was obtained from Tokyo Chemical Industry Co., Ltd., and ethanol (≥99.7%) was purchased from Chengdu Kelong Chemical Co., Ltd. The general characteristics of XO, BO2, and AO7 dyes, including Cas number, chemical number, and maximum absorption wavelength are (λ_max), are shown in Table 1. These reagents were used as received without further purification.

2.2. Synthesis of Fe₃O₄

Superparamagnetic Fe₃O₄ nanoparticles were synthesized by a sample one-step hydrothermal procedure based on our previous report [35]. Iron citrate (4.0 mmol) and distilled water (30 mL) was directly added into a 50 mL Teflon-lined stainless-steel autoclave, and maintained for 24 h at 200 °C. After cooling to room temperature, the resulting precipitate was separated with the help of a magnet, and washed with distilled water and ethanol. Finally, the Fe₃O₄ powders were obtained under vacuum at 60 °C for 24 h.

2.3. Characterization

The crystallographic phase of Fe₃O₄ sample was characterized by X-ray diffraction (XRD, DX-2700). The morphology and size of Fe₃O₄ particles were evaluated by transmission electron microscopy (TEM, JEM-2100F). The magnetic property of Fe₃O₄ powders was obtained by physical performance measurement system (PPMS 9).

2.4. Evaluation of Adsorption Capacity

The adsorption capacities of Fe₃O₄ powders were evaluated by the removal of XO, BO2, and AO7 dyes from simulated wastewater at room temperature without pH preadjustment. Briefly, 0.1 g Fe₃O₄ powder was dispersed into 100 mL dye solution with different concentrations, and the mixture was stirred at a constant speed of 200 rpm. Then, a small amount of suspension was withdrawn at regular intervals and separated by an applied magnetic field, and the absorbance of supernatant was measured at the maximum absorption wavelength of the dye using an ultraviolet-visible spectrophotometer (U-3900). The adsorption efficiencies (η_t, %) and adsorption amount (q, mg/g) for dyes were calculated using Equations (1) and (2), respectively:

$${\eta }_{\mathrm{t}}=\frac{C_0-C\mathrm{t}}{C_0}\times 100\%$$

(1)

$$q=\frac{(C_0-C\mathrm{e})V}{m}$$

(2)

where C₀ (mg/L) is the initial concentration of dyes, C_t (mg/L) is the concentration of dyes at time t (t = 0–60 min), m (g) is the mass of Fe₃O₄ powders, and V (L) is the volume of dyes aqueous solution.

Table 1. General characteristics of AO7 dye.

Generic Name	Abbreviation	Cas Number	Chemical Structure	λmax (nm)
Xylenol orange	XO	1611-35-4		435
Basic orange 2	BO2	532-82-1		452
Acid orange 7	AO7	633-96-5		484

Table 1. General characteristics of AO7 dye.

Generic Name	Abbreviation	Cas Number	Chemical Structure	λmax (nm)
Xylenol orange	XO	1611-35-4		435
Basic orange 2	BO2	532-82-1		452
Acid orange 7	AO7	633-96-5		484

3. Results and Discussion

Figure 1a shows the XRD pattern of the sample. All broad peaks had a good match with the standard Fe₃O₄ pattern (JCPDS No. 65-3107), suggesting that the as-synthesized Fe₃O₄ particles had a good crystallinity with small grain size. The mean grain size of Fe₃O₄ was about 6.6 nm, as calculated by the Scherrer formula. Figure 1b shows the high-resolution XPS spectra of Fe 2p and O 1s of the as-synthesized Fe₃O₄ particles. For the Fe2p core-level XPS spectra, two distinct peaks with binding energies of 724.1 and 710.5 eV appeared, which were assigned to the characteristic doublets of Fe 2p_1/2 and Fe 2p_3/2 from iron oxide. For the O 1s core-level XPS spectra in the inset of Figure 1b, the O 1s centered at binding energy of 529.8 eV belonged to O²⁻ species, and these data are consistent with the reported literature [36]. The above XRD and XPS results confirmed the formation of Fe₃O₄ phase in the hydrothermal system. TEM was employed to characterize the morphology and size of Fe₃O₄ particles. As observed in Figure 1c, the morphology of particles was an equiaxed shape. Moreover, these size values of Fe₃O₄ particles were demonstrated by a statistical analysis, the size distribution histograms were showed in Figure 1d, and most of the Fe₃O₄ particles were mainly concentrated in 4.8–6.4 nm.

The room-temperature magnetization hysteresis loop of the as-synthesized Fe₃O₄ was measured by a physical performance measurement system. Figure 2a shows the magnetic hysteresis curve of Fe₃O₄ powders. From Figure 2a, it could be found that the saturation magnetization value of Fe₃O₄ nanoparticles was 54.9 emu/g. Moreover, the Fe₃O₄ nanoparticles were essentially superparamagnetic with negligible hysteresis, as observed by the enlarged partial curve of the surrounding origin in Figure 2a. In practical application, it is critical for practical applications that the magnetic materials should exhibit prompt responsiveness to an applied magnetic field without retaining any magnetism once the applied magnetic field was removed [37]. The magnetic manipulation of such superparamagnetic Fe₃O₄ was performed in water upon the application of a NdFeB permanent magnet near the glass bottle. As observed in Figure 2c, the Fe₃O₄ particles can be totally attracted toward the magnet within 15 s, leaving the suspension a clear solution. Moreover, the congregated Fe₃O₄ particles can be easily and quickly re-dispersed again by shaking after the removal of magnet. Hence, this superparamagnetic Fe₃O₄ powders have the potential to be easily recovered after liquid phase adsorption reaction, which could greatly facilitate the practical running of an industrial pollutant cleanup.

Figure 3 shows that the time-dependent adsorption efficiencies of superparamagnetic Fe₃O₄ nanoparticles for XO, BO2, and AO7 dyes. It was observed that Fe₃O₄ nanoparticles had larger adsorption affinity for XO dye, whereas the adsorption of AO7 dye can be ignored. The adsorption efficiencies achieved within 60 min of reaction was 98.5, 15.9, and 5.5% for XO, BO2, and AO7 dyes, respectively. Furthermore, it could be clearly observed that the adsorption of XO dye was rapid at the early stages of the process. In fact, the adsorption reaction was mostly completed within 10 min, and the removal rate was up to 93.6%. No significant changes were observed from 20 to 60 min, indicating that the adsorption-desorption equilibrium between the dye adsorbate and Fe₃O₄ adsorbent was reached within the first 10 min of adsorption reactions. Compared with the adsorption reaction of XO dye, those of BO2 and AO7 dyes reached the adsorption-desorption equilibrium within 30 min, and the removal rates were only 23.0 and 6.2%.

The saturated adsorption amounts (q_m) of XO, BO2, and AO7 dyes were obtained according to the Langmuir isotherm model, and the Langmuir linear fittings based on the adsorption data of these dyes onto the superparamagnetic Fe₃O₄ nanoparticles were showed in Figure 4a–c, while the relevant parameters of Langmuir fittings calculated were listed in

Table 2. Relevant parameters of Langmuir fittings for XO, BO2, and AO7 dyes adsorbed onto the superparamagnetic Fe₃O₄ powders.

Langmuir Parameters	$\mathbf{Langmuir}\mathbf{Isotherm}\mathbf{Model}:\frac{\mathit{C}\mathit{e}}{{\mathit{q}}_{\mathbf{e}}}=\frac{1}{{\mathit{q}}_{\mathit{m}}}\mathit{C}\mathit{e}+\frac{1}{{\mathit{K}}_{\mathit{L}}{\mathit{q}}_{\mathit{m}}}$
	XO Dye	BO2 Dye	AO7 Dye
qm (mg/g)	42.5	1.3	3.8
R2	0.9991	0.9546	0.7904

. By comparing with the associated correlation coefficients (R²), it could be found that the Langmuir isotherm model was most suitable for modeling the adsorption of XO dye (R² = 0.9991) than that of BO2 (R² = 0.9546) and AO7 (R² = 0.7904) dyes, and the value of q_m is 42.5 mg/g for XO dye according to the Langmuir linear fitting.

Table 3. Recent literature on adsorbent development for the removal of XO dye.

Authors	Adsorbent Name	Synthetic Method	qm (mg/g)
Ishaq [38]	Coal ash	Heated at 750 °C	0.74
Bai [39]	Porous amino-cellulose membrane	TEMPO oxidation and ethylenediamine grafting	15
Pang [40]	Expansion graphite	Chemical oxidation intercalation of potassium permanganate and vitriol	18.15
Garrudo-Guirado [41]	Vitreous tuff mineral (VT)	Milled and sieved to 60 mesh	45.17
Wang [42]	Bi-porous chitosan monoliths	Unidirectional freeze-drying method under vacuum less than 20 Pa for 48 h	153.8
Zhu [33]	Hyperbranched polyethyleneimine (HPEI) based gel	Cross-linking reaction between HPEI and N,N′-methylene-bis-acrylamide	3312.06
Xu in this work	Fe3O4 nanoparticles	One-pot hydrothermal process at 200 °C for 24 h	42.5

shows the q_m values of XO dye adsorbed on other adsorbents from the recent literature [33,38,39,40,41,42]. Despite the adsorption capacity of superparamagnetic Fe₃O₄ nanoparticles synthesized in this work being moderate among these reported adsorbents by comparing the q_m of various adsorbents, the Fe₃O₄ adsorbent still had obvious advantages in energy consumption and cost, which are due to the advantages of synthetic process and the superparamagnetism of Fe₃O₄ nanoparticles synthesized in this work. The equipment used in the hydrothermal process was simple and economical, and the Fe₃O₄ phase could be obtained in a simple one-pot hydrothermal process. In addition, the as-synthesized Fe₃O₄ nanoparticles in this work had excellent magnetic sensitivity, which made them easy to separate from liquid medium under the attraction of a magnet.

The sorption kinetics of XO, BO2, and AO7 dyes onto Fe₃O₄ nanoparticles were tested using a pseudo-first-order kinetic model by plotting log(q_e-q_t) versus t (Figure 5a,c,e), as well as pseudo-second-order kinetic model by plotting t/q_t versus t (Figure 5b,d,f). As observed in Figure 5, the adsorption of XO, BO2, and AO7 dyes using a pseudo-second-order model (Figure 5b,d,f) exhibited a better linear fit than those using a pseudo-first-order model (Figure 5a,c,e). Moreover, the relevant kinetic parameters, such as the equilibrium adsorption amount (q_e1,cal, q_e2,cal), rate constant (k₁ and k₂), and correlation coefficient (R²), could be obtained by fitting with these two models, and the results were listed in

Table 4. Kinetic parameters for the adsorption of XO, BO2, and AO7 dyes onto Fe₃O₄ nanoparticles at room temperature.

Dye	qe,exp (mg/g)	Pseudo-First-Order Kinetic Model $\mathbf{log}({\mathit{q}}_{\mathbf{e}1,\mathbf{cal}}-{\mathit{q}}_{\mathit{t}})=-\frac{{\mathit{k}}_1}{2.303}\mathit{t}+\mathbf{log}{\mathit{q}}_{\mathbf{e}1,\mathbf{cal}}$			Pseudo-Second-Order Kinetic Model $\frac{\mathit{t}}{{\mathit{q}}_{\mathit{t}}}=\frac{1}{{\mathit{q}}_{\mathit{e}2,\mathbf{cal}}}\mathit{t}+\frac{1}{{\mathit{k}}_2{\mathit{q}}_{\mathit{e}2,\mathbf{cal}}^2}$
		qe1,cal (mg/g)	k1 (1/h)	R2	qe2,cal (mg/g)	k1 (g/mg·h)	R2
XO	9.847	1.7906	7.1318	0.7328	10.0452	6.4352	0.9998
BO2	2.444	6.2516	0.2983	0.6316	2.8740	2.8272	0.9859
AO7	0.614	8.9600	0.0383	0.5564	0.7223	18.9232	0.9959

. As observed in Table 4, all pseudo-second-order equations showed higher correlation coefficients (R² > 0.98) than their respective pseudo-first-order equations (R² < 0.74), and the adsorption amounts at equilibrium (q_e,cal) were much closer to the respective experimental one (q_e,exp). Therefore, the pseudo-second order model is more suitable to describe the adsorption kinetics of XO, BO2, and AO7 dyes onto Fe₃O₄ nanoparticles, indicating that the chemisorption is the rate controlling step during the attachment process.

4. Conclusions

In summary, Fe₃O₄ nanoparticles with a main size distribution of 4.8–6.4 nm have been successfully synthesized via a simple one-pot hydrothermal strategy at 200 °C for 24 h, in which iron citrate and distilled were the sum total of raw materials. The as-synthesized Fe₃O₄ nanoparticles exhibited apparent superparamagnetism, and the saturation magnetization value was 54.9 emu/g. Moreover, Fe₃O₄ nanoparticles possessed sensitive magnetic responsiveness, which can be totally attracted toward the magnet within 15 s from suspension, and quickly re-dispersed again by shaking after the removal of magnet. Such superparamagnetic Fe₃O₄ nanoparticles exhibited the effective adsorption affinity for XO dye, and the adsorption reactions were mostly completed within 10 min. The Langmuir isotherm model is most suitable for modeling the adsorption of XO dye (R² = 0.9991) compared to those of BO2 and AO7 dyes, and the value of the saturated adsorption amount is 42.5 mg/g for XO dye according to Langmuir linear fitting. This superparamagnetic Fe₃O₄ nanoparticles had the potential to be easily recovered after liquid phase adsorption reaction, which could greatly facilitate the practical running of an industrial pollutant cleanup.