Trans. Nonferrous Met. Soc. China 28(2018) 1465-1474

Recovery of magnetite from waste ferrous sulfate using polyethylene glycol (PEG) as a dispersant

Ying-lin PENG¹, Wang YU², Ya-jie ZHENG²

1. School of Materials and Chemical Engineering, Hunan City University, Yiyang 413000, China;

2. School of Metallurgy and Environment, Central South University, Changsha 410083, China

Received 14 March 2017; accepted 25 July 2017

Abstract: The effect of PEG dispersant on the magnetic separation of magnetite (Fe₃O₄) synthesized from ferrous sulfate solution via co-precipitation method with calcium hydroxide as the precipitant was investigated. The results indicated that a PEG dispersant could significantly affect Fe₃O₄ recovery. Adding PEG during the preparation of Fe₃O₄ was unfavorable for Fe₃O₄ recovery. When the PEG-6000 concentration was increased from 0 to 8 g/L, the iron grade and median particle size of the Fe₃O₄ product decreased from 65.58% and 2.35 μm to 57.79% and 1.35 μm, respectively. However, adding PEG during the wet milling of the mixed product promoted the subsequent recovery of Fe₃O₄. When the amount of PEG-200 increased from 0% to 4% of the powder mass, the grade of iron in the Fe₃O₄ product increased from 65.58% to 68.32%. While the relative molecular mass of PEG at an amount of 4% of the powder mass increased from 200 to 20000, the grade of iron was reduced from 68.32% to 66.70%.

Key words: ferrous sulfate; Fe₃O₄; PEG; co-precipitation; magnetic separation; wet milling

1 Introduction

Ferrous sulfate (FeSO₄·7H₂O) is a by-product of titanium dioxide manufacture via the sulfate method. In this method, 3-4 t ferrous sulfate is generated to obtain 1 t titanium dioxide. In China, about 98% of the titanium dioxide is manufactured by the sulfate method, thereby yielding more than 7×10⁶ t ferrous sulfate waste each year, and the annual growth rate in waste emissions exceeds 10% as titanium dioxide production expands [1,2]. Unfortunately, this waste is less marketable and difficult to be utilized due to its high impurity contents (i.e., Mg, Mn, Ti, Zn and Al) [3], which causes not only severe environmental problems but also the waste of iron resource.

Using ferrous sulfate to produce iron oxides is an effective method for ferrous sulfate re-utilization because of the increasing demand for iron oxides driven by the increasing consumption in industry. Fe₃O₄, with fascinating properties, including good biocompatibility, high surface area, low toxicity, super-paramagnetism, and high saturation field [4], has attracted much attention from both academic and industrial interests in many application fields, such as magnetic storage media, solar energy transformation, catalysis, biological and  chemical sensing, ferrofluids and environmental remediation [5-9]. In environmental engineering, however, the utilization of Fe₃O₄ particles prepared from reagent-grade chemicals could not be cost-effective in the full-scale process, due to the characteristics of the water or wastewater, such as high flow rate, and the presence of various pollutants and associated high consumption and loss of particles [10]. In this work, we proposed a novel and low cost method to synthesize Fe₃O₄ particles from waste ferrous sulfate.

Numerous methods have been developed for preparing Fe₃O₄ particles, such as co-precipitation [11], microemulsions [12], sol-gel syntheses [13], and hydrothermal or solvothermal process [14]. However, the most common method for producing synthetic Fe₃O₄ particles is the co-precipitation of Fe²⁺/Fe³⁺ ions (molar ratio 1:2) by sodium hydroxide or ammonia solution [15]. In our approach, Fe₃O₄ was recovered from ferrous sulfate solution by co-precipitation and magnetic separation with calcium hydroxide as the precipitant [16,17]. We used lime instead of other alkali reagents due to its availability and low cost. Although caustic soda (NaOH) can be used for the pH adjustment, its price is 5 times higher than that of lime [18].

In the magnetic separation process, the Fe₃O₄ particles tend to heterocoagulate, so a good dispersion is required to improve their separation efficiency. The application of a polymeric dispersant in the magnetic separation of Fe₃O₄, however, has not drawn considerable attention. Since PEG is a water soluble polymer together with the other properties, such as bio- compatibility, low immunogenicity, and ease of use [19], we investigated the effect of a PEG dispersant on the magnetic separation of Fe₃O₄ synthesized from waste ferrous sulfate. The aim of this work is to investigate the effect of adding PEG during the preparation of Fe₃O₄ or during the wet milling of the mixed product on magnetite recovery.

2 Experimental

2.1 Materials

Reagent-grade NaOH, CaO, and PEG (Guangdong Xilong Chemical Co., Ltd., China) were used directly without further purification. The dried ferrous sulfate was obtained from Guangdong Hui Yun Titanium Industry Corporation Limited, China. The dried ferrous sulfate sample was further dried in a vacuum atmosphere at 80 °C for 12 h, and then its chemical composition was analyzed by X-ray fluorescence (XRF), as given in Table 1. The sample was composed of O (39.24%), Fe (36.08%), S (22.11%) and other impurities.

Table 1 Chemical composition of dried ferrous sulfate (mass fraction, %)

The ferrous sulfate solution was purified with the addition of NaOH solution to precipitate impurities [16]. The solution pH was adjusted to 3.5-4.0 by 5 mol/L NaOH solution, followed by continuous stirring for 30 min at room temperature. After filtration, the solution was diluted with distilled water. The diluted solution was used as an iron source for the Fe₃O₄ preparation and its composition is given in Table 2.

Table 2 Chemical composition of purified ferrous sulphate solution (g/L)

2.2 Experimental procedures

Fe₃O₄ was prepared from ferrous sulfate solution by co-precipitation method with calcium hydroxide as the precipitant. 4 L of purified ferrous sulfate solution (Table 2) was added to a 5 L, four-necked flask fitted with a reflux condenser and electric heater. Lime milk was obtained by adding distilled water to calcium oxide (CaO) at a liquid-to-solid mass ratio of 3:1 under vigorous stirring. Then, the lime milk with a 1.4:1 molar ratio of CaO/Fe²⁺ was added dropwise to the solution under sufficient stirring. The reactor was heated to 80 °C, and air was bubbled into the solution at 0.6 m³/h. When the mole ratio of Fe³⁺/Fe²⁺ in the reaction solution reached 1.9:1-2.1:1, the reaction was stopped. The obtained mixture of Fe₃O₄ and gypsum (CaSO₄·2H₂O) was then filtrated, dried, and ground for use in subsequent experiment. The wet-milling pretreatment was conducted at room temperature in a ball mill using a slurry with a mixed product-to-water mass ratio of 1:40 and a rotation speed of 130 r/min. The time of milling was 20 min, and the treated mixed product was directly subjected to magnetic separation. Magnetic separation was performed using a low-intensity magnetic separator (CRIMM DC CXGφ50, Changsha Research Institute of Mining and Metallurgy Co., Ltd., China), with a working electrical current of 1.75 A. The magnetic fractions were washed, filtered, dried, weighed and subjected to various analyses. Fe₃O₄ can be formed from ferrous salts according to the following reactions [20]:

CaO+H₂O→Ca(OH)₂                          (1)

FeSO₄+Ca(OH)₂+2H₂O→Fe(OH)₂+CaSO₄·2H₂O     (2)

4Fe(OH)₂+O₂+2H₂O→4Fe(OH)₃                  (3)

2Fe(OH)₃+Fe(OH)₂→Fe₃O₄+4H₂O                (4)

In this work, the effect of PEG dispersant on the magnetic separation of Fe₃O₄ was investigated as  follows: 1) During the preparation of Fe₃O₄, PEG-6000 with a concentration ranging from 0 to 8 g/L was added to the reaction solution; 2) During the wet milling of the mixed product, PEG-200, in amounts ranging from 0% to 16% of the powder mass, was added to the slurry;  PEG, with a relative molecular mass ranging from 200 to 20000, was also studied at an amount under 4% of the powder mass.

The procedure used to recover Fe₃O₄ from ferrous sulfate is shown in Fig. 1.

2.3 Analysis

The samples were analyzed to examine the ferrous and ferric iron contents according to GB/T 1863-2008 (Chinese National Standard). The contents of trace elements in sample were determined by inductively coupled plasma optical emission spectrometry (ICP-OES, iCAP 7000 SERIES, Thermo Scientific) after the total dissolution of samples. Phase analysis of the products was conducted by a Rigaku D/max-TTR III X-ray diffractometer (XRD) with Cu K_α radiation (λ=  1.54056 , voltage of 40 kV, and current of 250 mA) and at the scanning rate of 10 (°)/min from 10° to 70°. To identify the functional groups present in the product, Fourier transform infrared (FT-IR) spectrometer (WQF-510A, Beijing Beifen-Ruili Analytical Instrument (Group) Co., Ltd., China) was used to record the spectrum from 400 to 4000 cm^-1. The raw material sample was measured by X-ray fluorescence (XRF) with a Bruker S4 Pioneer system equipped with two X-ray detectors. Surface morphology of the samples was observed by using a FEI Quanta 200 scanning electron microscope (SEM), coupled with an energy dispersive X-ray spectroscope (EDS). Laser particle size analyzer (LS-pop(6), Zhuhai OMEC Instrument Co., Ltd., China) was used to analyze the size distribution of samples. Magnetic properties of the samples were measured with a vibrating sample magnetometer (VSM, Quantum Design) at room temperature.

After the experiment, the recovery rate of iron in the Fe₃O₄ product (R) was calculated based on the mass of iron in the raw material (the purified ferrous sulfate solution) and defined as follows:

R=m₁/m×100%                               (5)

where m₁ is the mass of iron in the Fe₃O₄ product and m is the mass of iron in the raw material.

The grade of iron in the Fe₃O₄ product (G) was calculated based on the mass of the Fe₃O₄ product and defined as follows:

G=m₁/m₂×100%                              (6)

where m₂ is the mass of the Fe₃O₄ product.

3 Results and discussion

3.1 Characterization of Fe₃O₄ product without adding PEG

In this work, a mixed product was first prepared from ferrous sulfate solution by co-precipitation with calcium hydroxide as the precipitant. Figure 2(a) shows the XRD pattern of the mixed product. XRD analysis indicated that Fe₃O₄ (JCPDS, No. 19-0629) and CaSO₄·2H₂O (JCPDS, No. 33-0311) were the major components. The SEM image of the mixed product is shown in Fig. 2(b). From Fig. 2(b), two distinct phases (zones A and B) were observed: the phase in zone A was irregularly shaped plates, while fine phase in zone B were apt to aggregate. The subsequent EDS analysis proved that the two phases were CaSO₄·2H₂O (zone A) and Fe₃O₄ (zone B), which is in agreement with the above XRD results.

Fig. 1 Flow sheet of Fe₃O₄ recovery from ferrous sulfate

Fig. 2 XRD pattern (a) and SEM image (b) of mixed product without adding PEG

After the co-precipitation stage, Fe₃O₄ was obtained from the mixed product with wet milling prior to the magnetic separation. The grade of iron in the mixed product without adding PEG was determined by titration to be 20.55%, while that in the Fe₃O₄ product was 65.58%. The XRD pattern and SEM image of the Fe₃O₄ product are shown in Figs. 3(a) and (b), respectively. Figures 3(a) and (b) show that a highly crystallized Fe₃O₄ product with a nearly spherical shape was obtained. However, a peak attributable to calcite (CaCO₃) emerged in the product, as shown in Fig. 3(a). The reaction to form CaCO₃ can be expressed by the following reaction equation:

Ca(OH)₂+CO₂→CaCO₃+H₂O                   (7)

In subsequent experiments, the PEG dispersant was added before the magnetic separation to decrease the content of calcite in the Fe₃O₄ product.

Fig. 3 XRD pattern (a) and SEM image (b) of Fe₃O₄ product without adding PEG

3.2 Adding PEG during preparation of Fe₃O₄

3.2.1 Influence of concentration of PEG on magnetic separation of Fe₃O₄

To investigate the influence of the PEG dispersant on the magnetic separation of Fe₃O₄, a series of experiments were performed with the PEG-6000 concentrations ranging from 0 to 8 g/L. PEG-6000 was added to the reaction solution before the co-precipitation stage, and the effects of PEG-6000 concentration on the grade and recovery rate of iron in the Fe₃O₄ product are shown in Fig. 4. It is revealed that both the grade and recovery rate of iron decreased rapidly with the increase of PEG-6000 concentration. When the PEG-6000 concentration fluctuated from 0 to 8 g/L, the grade and recovery rate of iron decreased from 65.58% to 57.79% and from 80.35% to 27.20%, respectively. These results indicate that the addition of the PEG dispersant during the preparation of Fe₃O₄ was not conducive to Fe₃O₄ recovery.

Fig. 4 Effects of PEG-6000 concentration on grade and recovery rate of iron in Fe₃O₄ product

It has been proposed that Fe₃O₄ is formed by air-oxidation of Fe(OH)₂ via dissolution and recrystallization [21]. PEG, as a crystalline water-soluble polymer, can greatly increase the viscosity of the solution system, which restrains the hydrated iron and ferrous ions migrating from the solution to the surface of the Fe₃O₄ crystal nucleus. Hence, the size of the Fe₃O₄ particle is not too large because of the growth-restraint of the Fe₃O₄ microcrystals [4,22]. This inhibition is proportional to the concentration of PEG in the reaction system. The particle size distributions of the Fe₃O₄ products are presented in Fig. 5. As shown, the size of the Fe₃O₄ particles decreased gradually with an increase in the PEG-6000 concentration, and the median particle sizes of the Fe₃O₄ particles obtained at 0, 2, 4 and 8 g/L PEG-6000 were 2.35, 1.80, 1.66 and 1.35 μm, respectively.

Fig. 5 Size distributions of Fe₃O₄ products obtained at different PEG-6000 concentrations

In this work, Fe₃O₄ was obtained from the mixed product by magnetic separation. However, magnetic, gravity and hydrodynamic drag forces are the major forces that govern the overall behavior of mineral particles in a magnetic separator [23]. While magnetic forces attract magnetic particles, gravity and drag forces work against magnetic forces. Particle size of minerals has a pronounced influence on the magnitude of these forces. It has been reported that the hydrodynamic drag forces are dominant for fine particles, the gravitational forces are dominant for the coarse particles, and the magnetic forces are dominant for particles in the intermediate size range [24]. Thus, adding PEG dispersant during the preparation of Fe₃O₄ would be unfavorable for Fe₃O₄ recovery because of the size reduction of the Fe₃O₄ particles.

3.2.2 XRD patterns and FT-IR spectra

Figure 6(a) gives the XRD patterns of the Fe₃O₄ products obtained at different PEG-6000 concentrations. All the XRD patterns were in good agreement with the standard pattern for Fe₃O₄. These peaks were very intensive, indicating a high crystallinity. However, a peak attributable to CaCO₃ emerged in all the XRD patterns.

Fig. 6 XRD patterns (a) and FT-IR spectra (b) of Fe₃O₄ products obtained at different PEG-6000 concentrations

More intense diffraction peaks indicate more extensive crystallization of the particles [25]. Figure 6(a) shows that the peak intensity of Fe₃O₄ decreased with the increase of PEG-6000 concentration, indicating that the obtained product contained less Fe₃O₄ phase with a lower degree of crystallization. These results are consistent with the results shown in Fig. 4.

The FT-IR spectra of the Fe₃O₄ products obtained at different PEG-6000 concentrations are shown in Fig. 6(b). All the products showed characteristic absorption bands at wave numbers around 579, 876, and 1429 cm^-1. The IR bands at 577-579 cm^-1 could be indexed as the Fe—O bond of Fe₃O₄ [26,27], whereas the IR bands at 876 cm^-1 and 1429-1436 cm^-1 could be assigned to the bending vibrations of the  group in the calcite structure [28,29]. These results are in agreement with the above XRD analysis.

3.3 Adding PEG during wet milling of mixed product

3.3.1 Influence of amount of PEG on magnetic separation of Fe₃O₄

Generally, in many industries handling fine powders, small quantities of dispersants can remarkably improve the grinding efficiency, so they have been widely used in ultrafine wet grinding processes [30]. In this work, Fe₃O₄ was recovered from the mixed product by wet milling prior to magnetic separation, and the PEG dispersant was added to the mixed product slurry at the beginning of the milling process. To study the effect of the amount of PEG on the magnetic separation of Fe₃O₄, a series of experiments were performed with the amount of PEG-200 ranging from 0% to 16% of the powder mass. Figure 7 shows the effects of the PEG-200 amount on the grade and recovery rate of iron in the Fe₃O₄ product. It can be seen that both the grade and recovery rate increased significantly with the amount of PEG-200 ranging from 0% to 4% and achieved their maximum (68.32% and 83.14%) at 4%. A slight downward trend was observed with a further increase of the PEG-200 amount. It is clearly demonstrated by the obtained results that adding the PEG dispersant during the wet milling of the mixed product was favorable for Fe₃O₄ recovery and the optimal amount of PEG-200 was 4% of the powder mass.

Fig. 7 Effects of PEG-200 amount on grade and recovery rate of iron in Fe₃O₄ product

In the magnetic separation process, fine particles are attracted to each another by van der Waals attractive forces to form aggregates or flocs; hence, a good dispersion is required to improve the separation efficiency [31]. However, by the addition of a small amount of a water-soluble surface-active polymer, it is possible to change the nature of the interparticle forces to be entirely repulsive. The fine particles will therefore not flocculate, but remain as discrete entities [32,33]. For small PEG dosages, the newly-created fracture surface generated in the mixed product grinding process could not adsorb enough PEG to form a stable dispersion system. However, when an excessive amount of PEG was used, it led to the polymer bridging among the particles [34], which had an adverse effect on the stability of the mixed product slurry. Therefore, the optimal PEG-200 dosage for further experiments was chosen to be 4% of the powder mass. A potential mechanism of the PEG-200 dispersant action in milling prior to the magnetic separation process is schematized in Fig. 8.

The particle size distributions of the Fe₃O₄ products obtained at different PEG-200 amounts are shown in  Fig. 9. The size distributions of the Fe₃O₄ particles were self-similar, and the median particle sizes of Fe₃O₄ particles obtained at 0%, 2%, 4%, and 16% PEG-200 were 2.35, 2.35, 2.48 and 2.42 μm, respectively. It can be inferred that the addition of the PEG dispersant during the milling process could not change the particle size of the Fe₃O₄ product. This is because the shape of the powders is determined by the grinding mechanisms operating in the mill, but the grinding mechanisms cannot be altered by the presence of dispersants [35].

The XRD patterns of the Fe₃O₄ products obtained at different PEG-200 amounts are given in Fig. 10. All the XRD patterns had obvious diffraction peaks and matched well with the reflections of the standard Fe₃O₄. CaCO₃ was only observed when the PEG-200 amount was 0%. This result indicates that it was not easy to obtain a pure-phase product without adding the PEG dispersant, which is consistent with the results shown in Fig. 7. Figure 11 gives the FT-IR spectra of the Fe₃O₄ products obtained at different PEG-200 amounts. All the products exhibited an absorption peak at around 577 cm^-1, corresponding to the Fe—O bond in Fe₃O₄. Notably, the product obtained at 0% PEG-200 exhibited absorption peaks at 876 cm^-1 and 1436 cm^-1, which were attributed to the  group in CaCO₃. These results are in agreement with the above XRD results.

The chemical composition of the Fe₃O₄ products obtained at different PEG-200 amounts was determined by titration and ICP-OES, and the results are presented in Table 3. It is shown that the product obtained by adding PEG contained higher concentrations of Fe, Ti, and Mn compared with the product obtained without adding PEG, whereas the product obtained by adding PEG contained less Ca, S and Mg, and the content of Ca was notably reduced from 2.01% to 0.81%. The results well accord with the previous results shown in Figs. 7 and 10. Therefore, it can be concluded that the effective magnetic separation of Fe₃O₄ can be realized by adding PEG dispersant during the wet milling of the mixed product, through a substantial reduction of the content of calcite in the Fe₃O₄ product.

Fig. 8 Schematic diagram of changes of potential in milling prior to magnetic separation process using PEG-200 as dispersant

Fig. 9 Size distributions of Fe₃O₄ products obtained at different PEG-200 amounts

Fig. 10 XRD patterns of Fe₃O₄ products obtained at different PEG-200 amounts

Fig. 11 FT-IR spectra of Fe₃O₄ products obtained at different PEG-200 amounts

3.3.2 Influence of relative molecular mass of PEG on magnetic separation of Fe₃O₄

To illustrate the influence of the relative molecular mass of the PEG dispersant on the magnetic separation of Fe₃O₄, a series of experiments were performed with the relative molecular mass of PEG ranging from 200 to 20000. Figure 12 shows the grade and recovery rate of iron in the Fe₃O₄ product obtained at different PEG relative molecular masses. It is shown that by increasing the relative molecular mass from 200 to 20000, the grade of iron was gradually reduced from 68.32% to 66.70% while the recovery rate of iron increased from 83.14% to 85.01%. These results indicate that the increase in the PEG relative molecular mass led to a reduction in the efficiency of the magnetic separation. REN et al [36] and YANG et al [37] reported that polymers with larger relative molecular masses can form a better conjunction with Fe₃O₄ particles, leading to well-enwrapped large particles, which results in the remarkable decrease of the magnetic property. Hence, the increase of the PEG relative molecular mass resulted in a poor magnetic separation probably because of a reduction in the magnetic property of the Fe₃O₄ particles.

Table 3 Chemical compositions of Fe₃O₄ products obtained at different PEG-200 amounts (mass fraction, %)

Fig. 12 Effects of PEG relative molecular mass on grade and recovery rate of iron in Fe₃O₄ product

3.4 Magnetic properties

The magnetic properties of the Fe₃O₄ products obtained without and with the PEG dispersant were measured at room temperature, and the magnetic hysteresis (M-H) loops are presented in Fig. 13. It is shown that the Fe₃O₄ product obtained without adding PEG exhibited the highest magnetization saturation (M_s) of 79.963×10^-3 A·m²/g, a coercivity value (H_c) of 5878.44 A/m, and a remanence value (M_r) of 8.8972×10^-3 A·m²/g (Fig. 13(a)). While adding PEG during the preparation of Fe₃O₄, the M_s and M_r for the Fe₃O₄ product decreased to 58.796×10^-3 and 7.7917×10^-3 A·m²/g, respectively, but the H_c increased to 6527.60 A/m (Fig. 13(b)). These observations are consistent with the results in Figs. 5 and 6. It is well known that the decrease in the particle size can lead to an increasing surface-to-volume ratio, which in turn increases the surface spin disorder and consequently results in a reduction in the saturation magnetization [38]. The reduction in the overall magnetic phase concentration or poor crystallization may also affect the magnetic behavior of the as-synthesized powder by reducing the saturation magnetization and slightly increasing the coercive force [39,40]. However, when adding PEG during the wet milling of the mixed  product, M_s for the Fe₃O₄ product increased to 83.656×10^-3 A·m²/g, which is not far from the theoretical value of bare Fe₃O₄ (~90×10^-3 A·m²/g) [41], and H_c decreased to 5729.28 A/m (Fig. 13(c)). This result indicates that the obtained Fe₃O₄ product possessed stronger magnetic properties after adding the PEG dispersant during the wet milling of the mixed product.

Fig. 13 Magnetization versus magnetic field curves of Fe₃O₄ products obtained without adding PEG (a), with adding PEG-6000 (4 g/L) during preparation of Fe₃O₄ (b) and with adding PEG-200 (4% of powder mass) during wet milling of mixed product (c)

4 Conclusions

1) The effect of PEG dispersant on the magnetic separation of Fe₃O₄ synthesized from a ferrous sulfate solution via co-precipitation with calcium hydroxide as the precipitant was investigated. The results indicated that the PEG dispersant had a significant influence on Fe₃O₄ recovery. When PEG was added during the preparation of Fe₃O₄ with PEG-6000 concentration increasing from 0 to 8 g/L, the grade of iron in the Fe₃O₄ product decreased from 65.58% to 57.79%. This result could be explained by the size reduction of the Fe₃O₄ particles, which was unfavorable for the magnetic separation of Fe₃O₄.

2) When PEG was added during the wet milling of the mixed product and the PEG-200 amount was increased from 0% to 4% of the powder mass, the grade of iron in the Fe₃O₄ product increased from 65.58% to 68.32%. The addition of PEG in the milling process made it possible to change the nature of the interparticle forces to be entirely repulsive, which caused an increase in the efficiency of the subsequent magnetic separation. While the relative molecular mass of PEG at an amount of 4% of the powder mass increased from 200 to 20000, the grade of iron was reduced from 68.32% to 66.70%. Polymers with larger molecular mass can give rise to improved conjunction to Fe₃O₄ particles and lead to a more remarkable decrease of the magnetic property, which results in a poor magnetic separation. In addition, the obtained Fe₃O₄ product possessed stronger magnetic properties after adding the PEG dispersant during the wet milling of the mixed product.

References

[1] HUANG Peng-hui, JIANG Bing, ZHANG Zhi-ye, WANG Xin-long, CHEN Xiao-dong, YANG Xiu-shan, YANG Lin. Recycling sulfur and iron resources in the waste ferrous sulfate [J]. Journal of Thermal Analysis and Calorimetry, 2015, 119(3): 2229-2237.

[2] HUANG Peng-hui, DENG Shao-gang, ZHANG Zhi-ye, WANG Xin-long, CHEN Xiao-dong, YANG Xiu-shan, YANG Lin. A sustainable process to utilize ferrous sulfate waste from titanium oxide industry by reductive decomposition reaction with pyrite [J]. Thermochimica Acta, 2015, 620: 18-27.

[3]  R, VACA F. Physicochemical characterization of raw materials and co-products from the titanium dioxide industry [J]. Journal of Hazardous Materials, 2009, 166(2-3): 1429-1440.

[4] SUN Min-min, ZHU Ai-mei, ZHANG Qiu-gen, LIU Qing-lin. A facile strategy to synthesize monodisperse superparamagnetic OA-modified Fe₃O₄ nanoparticles with PEG assistant [J]. Journal of Magnetism and Magnetic Materials, 2014, 369: 49-54.

[5] KARAPINAR N. Magnetic separation of ferrihydrite from wastewater by magnetic seeding and high-gradient magnetic separation [J]. International Journal of Mineral Processing, 2003, 71(1-4): 45-54.

[6] LI Yi-ran, WANG Jun, ZHAO Ying, LUAN Zhao-kun. Research on magnetic seeding flocculation for arsenic removal by superconducting magnetic separation [J]. Separation and Purification Technology, 2010, 73(2): 264-270.

[7] LAKSHMANAN R, RAJARAO G K. Effective water content reduction in sewage wastewater sludge using magnetic nanoparticles [J]. Bioresource Technology, 2014, 153: 333-339.

[8] XIA Hong, WANG juan, TIAN Ye, CHEN Qi-Dai, DU Xiao-bo, ZHANG Yong-lai, HE Yan, SUN Hong-bo. Ferrofluids for fabrication of remotely controllable micro-nanomachines by two-photon polymerization [J]. Advanced Materials, 2010, 22: 3204-3207.

[9] WANG juan, XIA Hong, XU Bin-bin, NIU Li-gang, WU Dong, CHEN Qi-dai, SUN Hong-bo. Remote manipulation of micronanomachines containing magnetic nanoparticles [J]. Optics Letters, 2009, 34(5): 581-583.

[10] WEI Xin-chao, VIADERO R C Jr. Synthesis of magnetite nanoparticles with ferric iron recovered from acid mine drainage: Implications for environmental engineering [J]. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2007, 294: 280-286.

[11] NABIYOUNI G, JULAEE M, GHANBARI D, ALIABADI P C, SAFAIE N. Room temperature synthesis and magnetic property studies of Fe₃O₄ nanoparticles prepared by a simple precipitation method [J]. Journal of Industrial and Engineering Chemistry, 2015, 21: 599-603.

[12] ZHANG Dong-en, TONG Zhi-wei, LI Shan-zhong, ZHANG Xiao-bo, YING Ai-ling. Fabrication and characterization of hollow Fe₃O₄ nanospheres in a microemulsion [J]. Materials Letters, 2008, 62(24): 4053-4055.

[13] ALBORNOZ C, JACOBO S E. Preparation of a biocompatible magnetic film from an aqueous ferrofluid [J]. Journal of Magnetism and Magnetic Materials, 2006, 305(1): 12-15.

[14] HAW C Y, MOHAMED F, CHIA C H, RADIMAN S, ZAKARIA S, HUANG N M, LIM H N. Hydrothermal synthesis of magnetite nanoparticles as MRI contrast agents [J]. Ceramics International, 2010, 36(4): 1417-1422.

[15] VALENZUELA R, FUENTES M C, PARRA C, BAEZA J, DURAN N, SHARMA S K, KNOBEL M, FREER J. Influence of stirring velocity on the synthesis of magnetite nanoparticles (Fe₃O₄) by the co-precipitation method [J]. Journal of Alloys and Compounds, 2009, 488(1): 227-231.

[16] YU Wang, PENG Ying-lin, ZHENG Ya-jie. Effect of purification pretreatment on the recovery of magnetite from waste ferrous sulfate [J]. International Journal of Minerals, Metallurgy, and Materials, 2016, 23(8): 891-897.

[17] YU Wang, PENG Ying-lin, ZHENG Ya-jie. Recovery of iron from waste ferrous sulphate by co-precipitation and magnetic separation [J]. Transactions of Nonferrous Metals Society of China, 2017, 27(1): 211-219.

[18] BISWAL A, DASH B, TRIPATHY B C, SUBBAIAH T, SHIN S M, SANJAY K, MISHRA B K. Influence of alternative alkali reagents on Fe removal during recovery of Mn as electrolytic manganese dioxide (EMD) from Mn sludge [J]. Hydrometallurgy, 2013, 140: 151-162.

[19] LV Chao, XIA Hong, SHI Qing, WANG Gong, WANG Ying-shuai, CHEN Qi-dai, ZHANG Yong-lai, LIU Lian-qing, SUN Hong-bo. Sensitively humidity-driven actuator based on photopolymerizable PEG-DA films [J]. Advanced Materials Interfaces, 2017, 4(9): 1-8.

[20] YU Wang, PENG Ying-lin, ZHENG Ya-jie. Recovery of magnetite from FeSO₄·7H₂O waste slag by co-precipitation method with calcium hydroxide as precipitant [J]. Journal of Central South University, 2017, 24(1): 62-70.

[21] ISHIKAWA T, NAKAZAKI H, YASUKAWA A, KANDORI K, SETO M. Influences of Co²⁺, Cu²⁺ and Cr³⁺ ions on the formation of magnetite [J]. Corrosion Science, 1999, 41(8): 1665-1680.

[22] SHIRINOVA H, PALMA L D, SARASINI F, J, RAMAZANOV M A, HAJIYEVA F, SANNINO D, POLICHETTI M, GALLUZZI A. Synthesis and characterization of magnetic nanocomposites for environmental remediation [J]. Chemical Engineering Transactions, 2016, 47: 103-108.

[23] AMBASHTA R D, M. Water purification using magnetic assistance: A review [J]. Journal of Hazardous Materials, 2010, 180(1-3): 38-49.

[24] AROL A I, AYDOGAN A. Recovery enhancement of magnetite fines in magnetic separation [J]. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2004, 232: 151-154.

[25] SHEN La-zhen, QIAO Yong-sheng, GUO Yong, TAN Jun-ru. Preparation and formation mechanism of nano-iron oxide black pigment from blast furnace flue dust [J]. Ceramics International, 2013, 39(1): 737-744.

[26] ZHANG Wen-xue, LU Bin, TANG Hui-hui, ZHAO Jing-xiang, CAI Qing-hai. Reclamation of acid pickling waste: A facile route for preparation of single-phase Fe₃O₄ nanoparticle [J]. Journal of Magnetism and Magnetic Materials, 2015, 381: 401-404.

[27] CHEN Ru-fen, SONG Shan-shan, WEI Yu. Study on the formation and phase transformation of 1D nanostructured Fe₃O₄ particles under an external magnetic field [J]. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2012, 395: 137-144.

[28] GUNASEKARAN S, ANBALAGAN G. Spectroscopic study of phase transitions in natural calcite mineral [J]. Spectrochimica Acta: Part A, 2008, 69(4): 1246-1251.

[29] LI Tian-tian, SUI Feng-feng, LI Fu-chun, CAI Yuan-feng, JIN Zhang-dong. Effects of dry grinding on the structure and granularity of calcite and its polymorphic transformation into aragonite [J]. Powder Technology, 2014, 254: 338-343.

[30]  B, UCURUM M. An experimental study on the ultra-fine grinding of gypsum ore in a dry ball mill [J]. Powder Technology, 2016, 291: 186-192.

[31] BHAGAT R P, PATHAK P N. The effect of polymeric dispersant on magnetic separation of tungsten ore slimes [J]. International Journal of Mineral Processing, 1996, 47(3-4): 213-217.

[32] GREENWOOD R, ROWSON N, KINGMAN S, BROWN G. A new method for determining the optimum dispersant concentration in aqueous grinding [J]. Powder Technology, 2002, 123(2-3): 199-207.

[33] ZHANG Xue, HU Hui-ren. Synthesis and application of a polyacrylate dispersant on the preparation of ultrafine ground calcium carbonate in a laboratory stirred media mill [J]. Powder Technology, 2014, 266: 218-227.

[34] LEE H H, YAMAOKA S, MURAYAMA N, SHIBATA J. Dispersion of Fe₃O₄ suspensions using sodium dodecylbenzene sulphonate as dispersant [J]. Materials Letters, 2007, 61(18): 3974-3977.

[35] KAPUR P C, HEALY T W, SCALES P J, BOGER D V, WILSON D. Role of dispersants in kinetics and energetics of stirred ball mill grinding [J]. International Journal of Mineral Processing, 1996, 47(1-2): 141-152.

[36] REN Jie, HONG Hai-yan, REN Tian-bin, TENG Xin-rong. Preparation and characterization of magnetic PLA-PEG composite particles [J]. Materials Letters, 2005, 59(21): 2655-2658.

[37] YANG Jing-hai, ZOU Ping, YANG Li-li, CAO Jian, SUN Yun-fei, HAN Dong-lai, YANG Shuo, WANG Zhe, CHEN Gang, WANG Bing-ji, KONG Xiang-wang. A comprehensive study on the synthesis and paramagnetic properties of PEG-coated Fe₃O₄ nanoparticles [J]. Applied Surface Science, 2014, 303: 425-432.

[38] SHEN La-zhen, QIAO Yong-sheng, GUO Yong, MENG Shuang-ming, YANG Guo-chen, WU Mei-xia, ZHAO Jian-guo. Facile co-precipitation synthesis of shape-controlled magnetite nanoparticles [J]. Ceramics International, 2014, 40(1, Part B): 1519-1524.

[39] FAIYAS A P A , VINOD E M, JOSEPH J, GANESAN R, PANDEY R K. Dependence of pH and surfactant effect in the synthesis of magnetite (Fe₃O₄) nanoparticles and its properties [J]. Journal of Magnetism and Magnetic Materials, 2010, 322(4): 400-404.

[40] KUMAR R, SAKTHIVEL R, BEHURA R, MISHRA B K, DAS D. Synthesis of magnetite nanoparticles from mineral waste [J]. Journal of Alloys and Compounds, 2015, 645: 398-404.

[41] LIN Chia-chang, HO Jui-min, WU Min-shan. Continuous preparation of Fe₃O₄ nanoparticles using a rotating packed bed: Dependence of size and magnetic property on temperature [J]. Powder Technology, 2015, 274: 441-445.

以PEG为分散剂从废绿矾中回收Fe₃O₄

彭映林¹，余 旺²，郑雅杰²

1. 湖南城市学院 材料与化学工程学院，益阳 413000；

2. 中南大学 冶金与环境学院，长沙 410083

摘  要：以CaOH为沉淀剂通过共沉淀从绿矾溶液中回收Fe₃O₄，研究分散剂PEG对Fe₃O₄磁分离的影响。结果表明，分散剂PEG对Fe₃O₄的回收有显著影响。在Fe₃O₄制备过程中添加PEG不利于Fe₃O₄的回收。当PEG-6000浓度从0增加到8 g/L时，Fe₃O₄产品的铁品位从65.58%降至57.79%，中值粒径从2.35 μm 降至1.35 μm。然而，在混合产物的湿磨过程中添加PEG可促进随后的Fe₃O₄回收。当PEG-200用量从粉末质量的0%增加到4%时，Fe₃O₄产物中铁的品位从65.58% 升高到68.32%。当PEG-200用量为4%时，改变PEG相对分子质量从200增加到20000，Fe₃O₄产物中铁品位从68.32%降至66.70%。

关键词：绿矾；Fe₃O₄；PEG；共沉淀；磁分离；湿磨

(Edited by Bing YANG)

Foundation item: Project (2016JJ6017) supported by Hunan Provincial Natural Science Foundation of, China; Project (15C0245) supported by the  Research Foundation of Education Bureau of Hunan Province, China; Project (2013A090100013) supported by the Special Foundation  of Major Science & Technology Program of Guangdong Province, China

Corresponding author: Ya-jie ZHENG; Tel: +86-731-88836285; E-mail: zyj@csu.edu.cn

DOI: 10.1016/S1003-6326(18)64785-5