Introduction

Recent industrial development has led to increase in the quantity of phosphate found in the industrial effluents (Nur et al. 2016). Discharged phosphorus causes eutrophication in fresh-water bodies, such as rivers and lakes, and the anoxic zones generated by these algal blooms result in the deaths of local fish species (Hong et al. 2015; Kim et al. 2015; Lee et al. 2015; Xie et al. 2013). Phosphorus is an essential element used in various day-to-day applications, such as semiconductors, fertilizers, and food (Tang et al. 2015; Yang et al. 2016). However, earth’s reserves of phosphorus are limited and will be depleted within 50–100 years, indicating the importance of not only removal of phosphorus but also its recovery (Cordell 2010; Cordell et al. 2009). Many researchers have recently started focusing on methods to recover phosphorus from the adsorbed phosphate wastewater. Table 1 shows the phosphate adsorption capacity of different adsorbents. As shown in Table 1, phosphate adsorption capacity of magnetite was performed 15.2 mg/g at pH 4.0 in our previous study (Choi et al. 2016).

Table 1 Adsorption capacity of phosphate adsorption by various adsorbents
Full size table

In this study, we have used magnetite for phosphate adsorption. The phosphate adsorption capacity of tantalum and zirconium was greater than that of magnetite. Neither of the materials, tantalum and zirconium, are economical, and both exhibit difficult desorption of phosphate (Thomas et al. 2016), so suitable phosphate recovery is difficult. Ce–Zr binary oxides and iron-doped activated carbon have smaller adsorption capacities than magnetite. Magnetite also has other benefits such as being comparatively inexpensive, and being highly convenient in terms of recycling as the recovery ratio of magnetite particles is greater than 99%. Magnetite has a strong magnetic field, resulting in easy recovery of particles using magnets and as a result. Therefore, magnetite is considered as a suitable material for phosphate recovery (Prasad and Satyanarayana 2012).

Phosphorus is contained in different effluents such as anaerobic digestion leachate, livestock-generated wastewater, and wastewater discharged from semiconductors (Huang et al. 2017; Kawai et al. 2016; Li et al. 2016). Recovery of phosphates from such sources of wastewater is essential for protecting the environment and reducing the depletion of phosphates.

In this study, we synthesized magnetite particles and performed their characterization. We also evaluated the effects three different types of coexisting components, i.e., anions, organic matter, and heavy metals, on phosphate adsorption on magnetite surface.

Materials and methods

Preparation of magnetite particles

Magnetite particles were prepared using a co-precipitation method. The first step in this preparation was to mix 250 mL of 0.5 M iron(II) sulfate heptahydrate (FeSO4•7H2O) solution with 1.0 M iron chloride hexahydrate (FeCl3•8H2O) solution. A 6 M sodium hydroxide (NaOH) solution was then added drop-wise into this mixture using an overhead stirrer (GT-SS20D, Green tech, Korea). The resulting precipitates were aged at 70 °C for 12 h, washed two times using deionized (DI) water and ethanol, and subsequently dried at 50 °C. Finally, the dried particles were crushed using a mortar and pestle, and a sieve was used to separate out particles with size less than 20 μm.

Characterization of magnetite particles

The particle-size distribution was measured using a particle-size analyzer (Dandong Bettersize Instruments Ltd., Bettersize2000, China). Specific surface area and pore distribution of magnetite was measured using a surface area analyzer (BET, BEL Japan Inc., Japan). The surface formation and components of magnetite were examined using a scanning electron microscopy (SEM, Carl Zeiss, Germany) and an energy dispersive X-ray spectroscopy (EDS, Carl Zeiss, Germany) operating at 15 kV.

Effect of coexisting components on phosphate adsorption

Solutions containing phosphate mixed with coexisting anions were prepared by mixing 200 mg/L of potassium dihydrogen phosphate with 1.0:0.5, 1.0:1.0, 1.0:1.5, and 1.0:2.0 M ratios of sodium nitrate, sodium bicarbonate, sodium sulfate, and sodium chloride, respectively, with DI water. The coexisting anions used in this study were nitrate, bicarbonate, sulfate, and chloride.

We examined the effects of coexisting organic matter on phosphate adsorption by analyzing the TOC concentration of the organic matter. A calibrated curve-fitting was done for each type of organic matter to determine the TOC concentration. Ethanol, glucose, citric acid (CA), oxalic acid (OA), and humic acid (HA) were all used as organic matter sources. Organic matter concentrations of 125, 250, 500, and 1000 mg/L were used on the basis of the TOC concentration findings, with 200 mg/L of phosphate prepared using potassium dihydrogen phosphate.

To determine the effects that coexisting heavy metals have on phosphate adsorption, we conducted experiments using different molar ratios of phosphate and heavy metals (P:heavy metal = 1.0:0.5, 1.0:1.0, 1.0:1.5, and 1.0:2.0) of Cu, Pb, and As(III) standard solution at pH 2–9. Furthermore, 1 g of magnetite particles and 30 mL of a solution containing phosphate mixed with coexisting anions were added drop-wise into a 50-mL conical tube that was continuously stirred using a multi-rotator (GTR-100, Green Tech Co., Korea) at room temperature.

Sampling and analysis

Each sample after the phosphate adsorption reaction was collected as a solution using a 0.45-μm syringe filter for removing magnetite particles. The phosphate concentration of each sample was then measured using the standard method 8048 with a high-range limit of 100 mg/L PO43− using an UV-spectrophotometer (Hach, DR3900, USA).

Results and discussion

Characterization

EDS is semi-quantitative analysis technique used for analysis of elements and their chemical composition. Figure 1 shows an EDS analysis of the magnetite used for examining magnetite surface components; carbon, oxygen, and iron were found as the components. Table 2 shows the magnetite composition obtained through the EDS analysis. Carbon impurity was found because the carbon tape used to set the sample for the EDS analysis. The atomic ratio of oxygen and iron was approximately 3:4, which is roughly the correct composition of magnetite.

Fig. 1
figure 1

EDS analysis of the magnetite particles

Full size image
Table 2 Composition of magnetite
Full size table

The particle-size distribution of magnetite is shown in Fig. 2. Overall, the range of the particle-size distribution was approximately 1–100 μm, and the average particle size was 24.77 μm. Figure 3 depicts BET analysis of magnetite particles. The specific surface area of magnetite was 85.6 m2/g and the pore size and pore volume were 14.7 nm and 0.32 cm3, respectively. The BET isotherm followed a type II isotherm, which means the magnetite particles exhibited a non-porous surface. SEM images of the magnetite are shown in Fig. 4. The particle size, as observed from the SEM images, was less than 100 μm, and the magnetite had an angular shape on the surface, and there are no existing pores on the magnetite surface.

Fig. 2
figure 2

Particle-size distribution of the magnetite

Full size image
Fig. 3
figure 3

BET analysis of magnetite

Full size image
Fig. 4
figure 4

SEM images of the magnetite. a Particle size of 100 μm. b Enlarged view to observed one particle size

Full size image

Effect of coexisting components on phosphate adsorption

Anions

The effect of coexisting anions on phosphate adsorption is shown in Fig. 5a–d. Four anions, NO3−, SO42−, HCO3−, and Cl−, were used to investigate the effects of the coexisting anions. Overall, we found that the coexisting anions had no effect on phosphate adsorption. Thus, the magnetite particles were considered to be the strongest when competing to bond with the magnetite surface as compared with other material like anions.

Fig. 5
figure 5

Effect of the coexisting anions on phosphate adsorption. a Nitrate. b Bicarbonate. c Chloride. d Sulfate

Full size image

Organic matter

Figure 6a, b shows the effect of general organic matter on phosphate adsorption for various TOC concentrations. Glucose and ethanol showed no effect on phosphate adsorption on the magnetite surface. However, the adsorbed quantity of phosphate on the magnetite surface was lower than that for the coexisting anions during the first 15 min of the experiment. Therefore, we can determine that glucose and ethanol only have an effect on phosphate adsorption only during lower adsorption kinetics.

Fig. 6
figure 6

Effect of general organic matter on phosphate adsorption. a Glucose. b Ethanol

Full size image

The effect of coexisting NOM on phosphate adsorption is shown in Fig. 7a–c. The phosphate adsorbed on the magnetite surface decreased with increasing TOC concentration of the NOMs, i.e., CA, OA, and HA. A comparison between phosphate adsorption using magnetite in a phosphate solution and that with magnetite in a phosphate solution mixed with 125 mg/L TOC of NOM indicates that the NOMs significantly decreased the amount of phosphate adsorbed. This is because a negative charge existed on the surface of the NOMs; therefore, both phosphate and NOMs competed for being adsorbed on to the magnetite surface (Guan et al. 2006; Liu et al. 2008b).

Fig. 7
figure 7

Effect of natural organic matter on phosphate adsorption. a Citric acid. b Oxalic acid. c Humic acid

Full size image

Heavy metals

Figure 8 show the effect of coexisting heavy metals, i.e., Pb and Cd, on phosphate adsorption. Under acidic conditions, the amount of phosphate adsorbed on to the magnetite surface in the presence of Pb and Cd did not change; however, under alkaline conditions, the amount of phosphate adsorbed decreased as with increasing concentration of Pb and Cd. This phenomenon was caused by the distribution of Pb and Cd at different pH conditions. The existing Pb and Cd cations were below the pH of 9–10 in water (Choi et al. 2013; Jin et al. 2015). However, for As(III), the amount of phosphate adsorbed decreased at all pH levels with increasing As(III) concentrations. This is because As(III) existed as an anion in water at almost all pH levels (Mähler et al. 2013).

Fig. 8
figure 8

Effect of heavy metals on phosphate adsorption. a Lead. b Cadmium. c Arsenic

Full size image

Conclusions

This study was conducted to examine the effects of coexisting components on phosphate adsorption when magnetite was used. The oxygen and iron on the surface of the magnetite were measured to have an atomic ratio of 3:4 and the particle-size range was 1–100 μm with an average particle size of approximately 24.77 μm. Coexisting anions were found to have a negligible effect on phosphate adsorption, and phosphate is fully adsorbed on to magnetite in this scenario. General organic matter also showed no effect on phosphate adsorption. In contrast, NOMs proved to be highly effective in decreasing the amount of phosphate adsorption. This phenomenon occurred because of the NOMs competing with phosphate for being adsorbed on to the magnetite surface as NOMs have a negative charge on the surface. The reaction of NOM adsorption on to the magnetite surface was mainly electrostatic. In the case of heavy metals, Pb and Cd showed no effect under acidic conditions but the amount of phosphate adsorbed was reduced in the presence of these metals near pH 10. This was because these metals are mostly present as cations at all pH levels in water. As(III), however, was effective in lowering the amount of phosphate adsorbed at all pH levels because As(III) exists in an anionic form at all pH levels in water. These anions competed with phosphate for being adsorbed on to the magnetite surface.