In this work, pyridinium-functionalized silica nanoparticles adsorbent (PC/SiO2/Fe3O4) was synthesized for phosphate removal from aqueous solutions. The removal efficiency of phosphate on the PC/SiO2/Fe3O4 was carried out and investigated under various conditions such as pH, contact temperature and initial concentration. The results showed that the adsorption equilibrium could be reached within 10 min, which fitted a Langmuir isotherm model, with maximum adsorption capacity of 94.16 mg/g, and the kinetic data were fitted well by pseudo-second-order and intra-particle diffusion models. Phosphate loaded on the adsorbents could be easily desorbed with 0.2 mol/L of NaOH, and the adsorbents showed good reusability. The adsorption capacity was still around 50 mg/g after 10 times of reuse. All the results demonstrated that this pyridinium-functionalized mesoporous material could be used for the phosphate removal from aqueous solution and it was easy to collect due to its magnetic properties.

INTRODUCTION

Phosphorus is one of the main elements to sustain life in the earth system, and is also an essential element in the metabolic process (Ramasahayam et al. 2012). However, phosphorus containing compounds in water could lead to serious problems such as eutrophication, water degradation and potentially harmful impact on humans or animals (Duranceau et al. 2014). China requires that the maximum level of phosphorus should not exceed 0.5 mg/L, which affects the survival of fish and other aquatic organisms (Chen et al. 2015). Therefore, controlling phosphorus emission is a key factor for preventing eutrophication phenomena in water bodies (Ghaneian et al. 2014; Ge et al. 2015).

To date different technologies have been developed to remove phosphorus from wastewater including chemical precipitation (Zhang et al. 2014), denitrification (Wang et al. 2014), biological removal (Sibag & Kim 2012), and adsorption (Zhong et al. 2014). Among these various methods of wastewater treatment, adsorption is one of the most environmentally friendly, simple and economic methods for phosphorus removal. A variety of adsorbents for phosphorus removal from water such as fly ash (Chen et al. 2007), ion exchange resin (Yoon et al. 2014), silicates (Moharami & Jalali 2013), and bone charcoal (Ghaneian et al. 2014) have been used in water treatment. Many of these adsorbents such as Ti and Al oxides can uptake phosphorus through innersphere complexation (Moharami & Jalali 2014). However, it is difficult to recycle these adsorbents after treatment, especially when the adsorbents are dispersed into water as fine particles. The difficulty in separation substantially limits the practical application of these powered adsorbents.

Recently, magnetic nanoparticles (MNPs) have attracted more attention in water treatment, due to their convenient separation from water with a simple magnetic process and low cost (Tang & Lo 2013). Several studies have reported that heavy metals, phenolic compounds, drugs, and others are adsorbed onto MNPs (Lin et al. 2010). We also notice that exchange quaternary ammonium cations can adsorb anions from water. For example, Qiu et al. (2006) prepared an anion phase with N-methylimidazolium immobilized on silica to separate common inorganic anions. Qiu et al. (2010) reported poly(ionicliquid)-grafted silica hybrid materials could be exchanged with tetrafluoroborate, hexafluorophosphate, and trifluoromethanesulfonate through a simple aqueous anion-exchange reaction. Xin et al. (2012) prepared cetyl pyridine bromide-modified bentonites (CPB-Bent) to remove benzoic acid from water. Among these materials, pyridinium modified materials, which have high adsorption, chemical inertness and low production cost, have attracted some researchers’ attentions. Therefore, we designed and synthesized pyridinium modified MNPs, and they were applied to remove anions from water (Ma et al. 2015). In this study, MNPs modified with pyridinium are expected to be a promising adsorbent to remove phosphorus from water, and can be easily separated from water after treatment. Consecutive cyclic experiments were conducted to evaluate the reusability of the adsorbent.

MATERIALS AND METHODS

Materials

Ferric chloride hexahydrate (FeCl3·6H2O), tetraethyl orthosilicate (TEOS), iron sulfate heptahydrate (FeSO4·7H2O), ammonia solution (NH3·H2O 25%), KH2PO4 stock solution (1,000 mg/L), N,N-dimethyl formamide (DMF), and 1,3-dichloropropane (C3H6Cl2) were purchased from New Sanli Chemical Reagents Company (Xi'an, Shaanxi, China) in analytical grade and used as received.

Adsorbent preparation and characterization

Pyridinium-functionalized magnetic mesoporous silica (PC/SiO2/Fe3O4) was prepared according to our method (Ma et al. 2015). Briefly, Fe3O4 nanoparticles were prepared via a coprecipitation method according to the literature. The core structured SiO2/Fe3O4 microsphere was synthesized by a modified Stöber method. Then, PC/SiO2/Fe3O4 was synthesized using 1-(3-chloropropyl) pyridin-1-ium chloride and SiO2/Fe3O4. Subsequently, the adsorbent was characterized by transmission electron microscopy (TEM) on a JEOL-2010 microscope (JEOL, Japan), Brunauer-Emmett-Teller (BET) on a V-Sorb 2800P specific surface area and porosity analyzer (Gold APP, China), magnetic measurements on a MPMS-XL SQUID magnetometer (Quantum Design, USA), Fourier transform infrared (FT-IR) spectra on a BRUKER TENSOR 27 spectrometer (Bruker, Germany) and X-ray photoelectron spectroscopy (XPS) on a ESCALAB 250 X-ray photoelectron spectrometer (Thermo Fisher Scientific, USA).

Adsorption experiments

The adsorption of phosphorus by PC/SiO2/Fe3O4 was conducted in batch experiments and the effects of adsorbent dosage, the solution pH, initial phosphate concentration, temperature, and contact time were investigated. Batch sorption experiments were performed at 20 °C under neutral pH conditions in a constant temperature shaker platform at 120 rpm for 2 h. An amount of 20 mg of adsorbent was added into a capped plastic centrifuge tube containing 20 mL of 100 mg/L phosphate solutions. 0.1 mol/L HCl or NaOH solutions were used to adjust the pH value. After the adsorption equilibrium, the suspension was separated by an external magnet. For each flask, 1 mL of the supernatant was sampled for analysis. The residual total phosphate concentration in the solutions was measured by ammonium molybdate spectrophotometry method (Yoon et al. 2014) at a wavelength of 880 nm using a UV-Vis NE300 spectrophotometer (Thermo Fisher Scientific, USA). All the treatments were performed in triplicate.

RESULTS AND DISCUSSION

Adsorbent characterization

The results of adsorbent characterization refer to our research literature (Ma et al. 2015). The BET surface areas for SiO2/Fe3O4 and PC/SiO2/Fe3O4 are 52.7 and 51.5 m2/g, pore volumes are 0.052 and 0.046 cm3/g and pore diameters are 2.6 and 2.2 nm, respectively. Figure 1(a) represents the TEM microstructure of Fe3O4 (a), (b), and PC/SiO2/Fe3O4 (c), (d). Figure 1(a) and 1(b) show that Fe3O4 was dispersed with irregular shape and average diameter of about 20 nm. Figure 1(c) and 1(d) illustrate that the dark Fe3O4 nanoparticles were embedded in the light grey SiO2 with average thickness of about 4 nm. The result demonstrated that Fe3O4 was coated with SiO2 on the surface of material successfully, but micrograph could not confirm that pyridinium was linked to the surface of SiO2. Therefore, XPS and FT-IR were applied to investigate the groups and chemical elements on the surface of the sorbent.
Figure 1

(a) TEM images of Fe3O4 top and PC/SiO2/Fe3O4 bottom; (b) magnetic hysteresis loops of Fe3O4, SiO2/Fe3O4 and PC/SiO2/Fe3O4; (c) FT-IR spectra of SiO2/Fe3O4 and PC/SiO2/Fe3O4; (d) XPS wide-scan spectrum of PC/SiO2/Fe3O4.

Figure 1

(a) TEM images of Fe3O4 top and PC/SiO2/Fe3O4 bottom; (b) magnetic hysteresis loops of Fe3O4, SiO2/Fe3O4 and PC/SiO2/Fe3O4; (c) FT-IR spectra of SiO2/Fe3O4 and PC/SiO2/Fe3O4; (d) XPS wide-scan spectrum of PC/SiO2/Fe3O4.

The magnetic hysteresis loops of Fe3O4, SiO2/Fe3O4 and PC/SiO2/Fe3O4 are shown in Figure 1(b). The saturation magnetization values were measured to be 85.6, 58.7 and 50.9 emu/g for Fe3O4, SiO2/Fe3O4 and PC/SiO2/Fe3O4, respectively. The saturation magnetization values decrease is attributed to the decrease of the magnetite fraction after silica coating and pyridinium linking (Ren et al. 2013). The adsorbent could be separated from solution with a magnet easily and re-dispersed quickly when the magnet was taken away due to its superparamagnetic property and large saturation magnetization (Zhang et al. 2013).

The spectra FT-IR are shown in Figure 1(c). The typical adsorption peak for Fe3O4 at 585 cm−1 was due to the Fe–O stretching vibration peak (Idris 2015). A broad strong band at 1,090 cm−1 contributes to the Si–O–Si stretching vibration peak, while bands at 800 cm−1 and 465 cm−1 due to the Si–O bending and stretching mode were obvious (Idris 2015), which confirmed the successful coating of silica layers on Fe3O4. The broad adsorption peak at around 3,425 cm−1 resulted from O–H stretching vibration, and 1,633 cm−1 and 958 cm−1 were characteristic bending bands of O–H. Adsorption peaks around 1,644 cm−1, 1,571 cm−1 belonged to C=N and C–N bonds respectively, and 1,329 cm−1 was attributed to the –CH2 bending of the connecting nitrogen ion (Ren et al. 2013), demonstrating pyridinium was successfully grafted to the surface of SiO2/Fe3O4 particles.

The wide-scan XPS spectra for PC/SiO2/Fe3O4 are shown in Figure 1(d). The elements of Si2s (154.8 eV), Si2p (103.5 eV), C1s (285.9 eV), Cl2p (196.9 eV), N1 s (400.8 eV), and O1s (532.7 eV) were detected on the surface of PC/SiO2/Fe3O4 (Konno & Yamamoto 1987; Hernández-Morales et al. 2012). The results agreed with the TEM micrograph of the sorbent. The results indicated that PC/SiO2/Fe3O4 was prepared successfully.

Determination of adsorption parameters

The effect of pH range of 2–10 at 20 °C on phosphate removal is shown in Figure 2(a). The results indicated that the solution pH was a significant parameter controlling the process of adsorption. It can be observed that the adsorption of phosphate increased rapidly within the pH range from 2.0 to 4.0, and maintained a high adsorption at pH in the range of 4.0 to 8.0. However, the adsorption decreased obviously at pH above 8.0. The reason for this phenomenon may be that the main form of phosphorus is H3PO4 as the solution pH value was lower than 3.0, which is difficult to adsorb by the adsorbent. The existing forms of phosphorus are H2PO4 and HPO42− within the pH range from 3.0 to 8.0, which are easily absorbed through ion exchange by the adsorbent. Contrarily, more OH exists in the solution and competes with HPO42− and PO43− at higher pH (above pH 8.0).
Figure 2

Effect of (a) pH, (b) temperature and (c) adsorbent dosage on adsorption of phosphate onto PC/SiO2/Fe3O4.

Figure 2

Effect of (a) pH, (b) temperature and (c) adsorbent dosage on adsorption of phosphate onto PC/SiO2/Fe3O4.

Figure 2(b) expresses the effect of temperature in the range of 20–40 °C for the adsorption of phosphate on the PC/SiO2/Fe3O4 under neutral pH for 60 min. The equilibrium sorption capacity slightly decreased with the solution temperature increased from 20 to 40 °C, indicating that adsorption process had an exothermic nature (Keränen et al. 2015).

The effect of adsorbent dosage in the range of 10–60 mg on the adsorption of phosphate was carried out at 20 °C under neutral pH for 60 min, as shown in Figure 2(c). The equilibrium adsorption capacity gradually decreased, with the increment of adsorbent dosage. While the removal rate was rising suggesting that phosphate adsorption onto adsorbent depended on availability of positively charged adsorption sites. This is consistent with the expectation that higher adsorbent dosages will result in lower adsorption efficiencies per gram, as the adsorbed phosphate is distributed among more available binding sites. Hence, the optimal adsorbent dosage for the investigation of maximum adsorption capacity is very important. According to the experimental results, the adsorption isotherm experiments were performed with 20 mg of phosphate.

Adsorption isotherms

Solutions of different initial concentrations 100, 150, 200, 250, 300, 350, 400, 500, 600 mg/L KH2PO4 were used to evaluate the effect of concentration on the phosphate removal with 20 mg of adsorbent in 20 mL of solution and shaken in a thermostat at 120 rpm for 60 min under 20 °C. Figure 3 shows the adsorption isotherm of phosphate by PC/SiO2/Fe3O4. The data were fitted by Langmuir isotherm and Freundlich isotherm, the results are shown in Table 1. From Figure 3 and Table 1, it is evident that the maximum adsorption capacity of PC/SiO2/Fe3O4 was 94.16 mg/g, and the Langmuir isotherm model showed more significant correlation (R2 = 0.999) than the Freundlich isotherm model (R2 = 0.912). Therefore, the adsorption process could be described as a monolayer adsorption due to the homogenous and negligible interaction between adsorbed molecules (Ge et al. 2015).
Table 1

Parameters for Langmuir and Freundlich equations

AdsorbentLangmuir equation
Freundlich equation
Qmax (mg/g)b (L/mmol)R2K (mg(1−1/n)L1/n/g)nR2
PC/SiO2/Fe3O4 94.16 5.62 × 10−2 0.999 43.14 7.931 0.912 
AdsorbentLangmuir equation
Freundlich equation
Qmax (mg/g)b (L/mmol)R2K (mg(1−1/n)L1/n/g)nR2
PC/SiO2/Fe3O4 94.16 5.62 × 10−2 0.999 43.14 7.931 0.912 

Qmax: maximum adsorption capacity; b: energy of adsorption; K: proportionality constant; n: dimensionless exponent; R2: correlation coefficient.

Figure 3

The equilibrium sorption isotherms of the phosphate.

Figure 3

The equilibrium sorption isotherms of the phosphate.

Adsorption kinetics

To illustrate the adsorption kinetics and the rate determining step of phosphate by PC/SiO2/Fe3O4, the pseudo-first-order model, pseudo-second-order model and intra-particle diffusion model were employed (El-Khaiary 2007). The description of the pseudo-first-order equation is following:
formula
1
where k1 (min−1) is the rate constant of sorption equilibrium in the first order reaction, qe (mg/g) is the amount of phosphate at the equilibrium state, qt (mg/g) is the amount of phosphate on the surface of the sorbent at any time.
The description of the pseudo-second-order equation is as follows:
formula
2
where k2 (g/(mg min)) is the rate constant of sorption equilibrium in the second order reaction.
The intra-particle equation is as follows:
formula
3
where kp (g/(mg min0.5)) is the intra-particle rate constant.
To distinguish which of the three models was more appropriate, a comparison between adsorption capacity calculated (qe, cal) and experimental adsorption capacity from the kinetic adsorption test was carried out. From Table 2, although high correlation coefficients (R2) were obtained, the correlation coefficients (R2) of pseudo-second-order (Figure 4(c)) were higher than those of pseudo-first-order (Figure 4(b)). The equilibrium adsorption capacity calculated (qe, cal) by the pseudo-second-order equation is closer to the experimental ones from the pseudo-first-order model, which proved that the phosphate adsorption on adsorbents is probably controlled by chemisorptions (Moattari et al. 2015).
Table 2

Parameters of adsorption kinetics

  Pseudo-first-order equation
Pseudo-second-order equation
Intra-particle diffusion equation
kp1kp2kp3
C (mg/L)qeexp(mg/g)qe1 (mg/g)k1 (min−1)R2qe2 (mg/g)k2 (g/(mg min))R2(mg/(g min0.5))
50 45.0 45.3 0.244 0.960 45.79 0.0155 0.999 13.79 3.383 0.076 
100 64.1 64.28 0.264 0.991 64.85 0.0131 0.999 20.34 5.372 0.008 
200 80.4 80.21 0.299 0.990 78.43 0.0348 0.999 24.99 1.940 0.032 
  Pseudo-first-order equation
Pseudo-second-order equation
Intra-particle diffusion equation
kp1kp2kp3
C (mg/L)qeexp(mg/g)qe1 (mg/g)k1 (min−1)R2qe2 (mg/g)k2 (g/(mg min))R2(mg/(g min0.5))
50 45.0 45.3 0.244 0.960 45.79 0.0155 0.999 13.79 3.383 0.076 
100 64.1 64.28 0.264 0.991 64.85 0.0131 0.999 20.34 5.372 0.008 
200 80.4 80.21 0.299 0.990 78.43 0.0348 0.999 24.99 1.940 0.032 

qeexp: experimental value; qe1 and qe2: calculated values; k1, k2, kp1, kp2 and kp3: rate constants.

Figure 4

(a) Adsorption kinetics of phosphate, (b) pseudo-first-order kinetic model, (c) pseudo-second-order kinetic model and (d) intra-particle diffusion model onto PC/SiO2/Fe3O4.

Figure 4

(a) Adsorption kinetics of phosphate, (b) pseudo-first-order kinetic model, (c) pseudo-second-order kinetic model and (d) intra-particle diffusion model onto PC/SiO2/Fe3O4.

Due to the mesoporous structure of adsorbent, diffusion was also tested. Figure 4(d) shows the intra-particle diffusion plot of phosphate on PC/SiO2/Fe3O4. As could be seen, the adsorption process appeared as three linear portions which were not linear during the whole time. The diffusion rate constants in every step follow the order of kp,1 > kp,2 > kp,3, which are described in Table 2. It can be demonstrated that the first portion with high slope showed the instant diffusion stage, in which large numbers of phosphate were adsorbed rapidly by adsorbent. After almost every active bonding site was filled, the phosphate entered into the pores of the adsorbent, then the phosphate was absorbed by the interior surface of pores, which appeared in the second stage. In the third stage, the intra-particle diffusion rate constants were almost zero, demonstrating equilibrium was reached finally. The regression did not pass through the origin, indicating that the intra-particle diffusion was not the only rate controlling step and that some other kinetic processes were also involved.

Results of phosphate adsorption onto different materials from other researchers are shown in Table 3. It can be seen that the maximum adsorption capacity of PC/SiO2/Fe3O4 for phosphate is matched and even higher than that of other adsorbents, which maybe contribute to more pyridinium cations coated on the surface per unit mass MNP. Moreover, due to the absorbing process being mainly controlled by an ion exchange process (Ma et al. 2015), the rate of adsorption is relatively fast. In addition, the adsorbent with magnetic properties could be separated easily from solution with a magnet. So, this material has advantages of high adsorption capacity, rapid adsorption, and easy separation in removing phosphate in aqueous solution.

Table 3

Comparison of adsorption capacities of various adsorbents for phosphate

Adsorbent nameqm (mg/g)Equilibrium time (min)References
PC/SiO2/Fe3O4 94.2 10 This work 
Iron/reduced iron oxide nanoparticles 43.7 – Ramasahayam et al. (2012)  
Bone charcoal (BC) 30.2 120 Ghaneian et al. (2014)  
TiO2 28.3 180 Moharami & Jalali (2014)  
Fe3O4 24.4 180 Moharami & Jalali (2014)  
Al2O3 21.5 180 Moharami & Jalali 2014)  
Wheat straw anion exchanger (WS–AE) 45.7 150 Xu et al. (2010)  
Nano-bimetal ferrites 41.4 120 Tu & You (2014)  
Fe3O4@Zn–Al–LDH 36.9 60 Yan et al. (2015)  
Fe3O4@Mg–Al–LDH 31.7 40 Yan et al. (2015)  
Fe3O4@mZrO2 119.8 200 Sarkar et al. (2010)  
Magnetic iron oxide nanoparticles 83.2 >100 Yoon et al. (2014)  
Fly ashes 130.4 >100 Chen et al. (2007)  
Adsorbent nameqm (mg/g)Equilibrium time (min)References
PC/SiO2/Fe3O4 94.2 10 This work 
Iron/reduced iron oxide nanoparticles 43.7 – Ramasahayam et al. (2012)  
Bone charcoal (BC) 30.2 120 Ghaneian et al. (2014)  
TiO2 28.3 180 Moharami & Jalali (2014)  
Fe3O4 24.4 180 Moharami & Jalali (2014)  
Al2O3 21.5 180 Moharami & Jalali 2014)  
Wheat straw anion exchanger (WS–AE) 45.7 150 Xu et al. (2010)  
Nano-bimetal ferrites 41.4 120 Tu & You (2014)  
Fe3O4@Zn–Al–LDH 36.9 60 Yan et al. (2015)  
Fe3O4@Mg–Al–LDH 31.7 40 Yan et al. (2015)  
Fe3O4@mZrO2 119.8 200 Sarkar et al. (2010)  
Magnetic iron oxide nanoparticles 83.2 >100 Yoon et al. (2014)  
Fly ashes 130.4 >100 Chen et al. (2007)  

qm: maximum adsorption capacity.

Desorption and reusability

To achieve the reusability of adsorbent, the phosphate loaded on the adsorbents should be desorbed carefully without damaging the structure of the adsorbents. Different consumption of NaOH (0.2 mol/L) was applied to desorb phosphate (KH2PO4 64.1 mg/g) after the adsorbent (20 mg) was separated from solution with a magnet and the supernatant was decanted. The results (Figure 5(a)) showed that over 90% of adsorbed phosphate could be desorbed with NaOH (12 mL), therefore, we chose 12 mL NaOH (0.2 mol/L) for desorption in the next experiments.
Figure 5

(a) Desorption of phosphate with NaOH and (b) reusability of the PC/SiO2/Fe3O4 for phosphorus adsorption.

Figure 5

(a) Desorption of phosphate with NaOH and (b) reusability of the PC/SiO2/Fe3O4 for phosphorus adsorption.

Phosphate adsorbed on the adsorbents was desorbed with desorption solution, and the adsorbents were collected and rinsed with deionized water thoroughly for reuse. The reusability of PC/SiO2/Fe3O4 is shown in Figure 5(b). The adsorption capacity was 64.1 mg/g in the first use of adsorbents, and it decreased to around 54.8 and 50.9 mg/g in the second and third use. The decrease of the adsorption capacity in the first two reuses might result from some of the phosphate entering into the pores of the adsorbent, which was not an irreversible adsorption. With the reduction of this kind of irreversible adsorption process, the adsorption capacity remained stable in the following reuse. The adsorption capacity was still around 50 mg/g after 10 times of recycle, indicating excellent reusability of the adsorbents.

CONCLUSION

In this study, a magnetic adsorbent PC/SiO2/Fe3O4 was successfully synthesized. Containing a magnetic core, PC/SiO2/Fe3O4 could be conveniently separated from water in short time by using a simple magnetic process. The PC/SiO2/Fe3O4 exhibited excellent capacity for phosphate removal within the pH range from 3.0 to 8.0, and its maximum adsorption calculated by Langmuir model was 94.16 mg/g. The phosphate adsorption reached the equilibrium state within 10 min, and the kinetic data were well fitted by pseudo-second-order model, and intra-particle pore diffusion process which confirmed the phosphate adsorption was limited by the chemisorption of phosphate on the porous structure of adsorbent. The PC/SiO2/Fe3O4 with sorbed phosphate can be effectively regenerated in NaOH solution, and the regenerated PC/SiO2/Fe3O4 can be used at least 10 times in the sorption–desorption cycles without any significant loss of the adsorption capacities. Therefore, we consider that PC/SiO2/Fe3O4 adsorbent can be reused as a promising, effective and magnetic separation adsorbent for phosphorus removal from aqueous solution.

ACKNOWLEDGEMENTS

This work was financially supported by National Natural Science Foundation of China under Grant No. 41101288; and Northwest A&F University PhD Degree Scholar Research Projects under Grant No. 2013BSJJ120.

REFERENCES

Duranceau
S. J.
Biscardi
P. G.
Barnhill
D. K.
2014
Screening the toxicity of phosphorous–removal adsorbents using a bioluminescence inhibition test
.
Environ. Toxicol.
31
,
489
495
.
Ghaneian
M. T.
Ghanizadeh
G.
Alizadeh
M. T. H.
Ehrampoush
M. H.
Said
F. M.
2014
Equilibrium and kinetics of phosphorous adsorption onto bone charcoal from aqueous solution
.
Environ. Technol.
35
,
882
890
.
Hernández-Morales
V.
Nava
R.
Acosta-Silva
Y.
Macías-Sánchez
S.
Pérez-Bueno
J.
Pawelec
B.
2012
Adsorption of lead (II) on SBA-15 mesoporous molecular sieve functionalized with –NH2 groups
.
Microporous Mesoporous Mater.
160
,
133
142
.
Keränen
A.
Leiviskä
T.
Hormi
O.
Tanskanen
J.
2015
Removal of nitrate by modified pine sawdust: effects of temperature and co-existing anions
.
J. Environ. Manage.
147
,
46
54
.
Lin
J. H.
Wu
Z. H.
Tseng
W. L.
2010
Extraction of environmental pollutants using magnetic nanomaterials
.
Anal. Methods
2
,
1874
1879
.
Ma
F.
Du
H. T.
Wang
Q.
Li
R. H.
Zhang
Z. Q.
2015
Preparation of Pyridinium-functionalized magnetic adsorbent and its application for nitrate removal from aqueous solution
.
Water Air Soil Pollut.
226
,
1
11
.
Moattari
R. M.
Rahimi
S.
Rajabi
L.
Derakhshanb
A. A.
Keyhani
M.
2015
Statistical investigation of lead removal with various functionalized carboxylate ferroxane nanoparticles
.
J. Hazard. Mater.
283
,
276
291
.
Moharami
S.
Jalali
M.
2014
Effect of TiO2, Al2O3, and Fe3O4 nanoparticles on phosphorus removal from aqueous solution
.
Environ. Prog. Sustain. Energy
33
,
1209
1219
.
Ramasahayam
S. K.
Gunawan
G.
Finlay
C.
Viswanathan
T.
2012
Renewable resource-based magnetic nanocomposites for removal and recovery of phosphorous from contaminated waters
.
Water Air Soil Pollut.
223
,
4853
4863
.
Xin
X.
Yang
J.
Feng
R.
Zhao
J.
Chen
G.
Wei
Q.
Du
B.
2012
Preparation, characterization and adsorption performance of cetyl pyridine bromide modified bentonites
.
J. Inorg. Organomet. P
22
,
42
47
.
Yoon
S. Y.
Lee
C. G.
Park
J. A.
Kim
J. H.
Kim
S. B.
Lee
S. H.
Choi
J. W.
2014
Kinetic, equilibrium and thermodynamic studies for phosphate adsorption to magnetic iron oxide nanoparticles
.
Chem. Eng. J.
236
,
341
347
.
Zhang
J. M.
Zhai
S. R.
Li
S.
Xiao
Z. Y.
Song
Y.
An
Q. D.
Tian
G.
2013
Pb(II) removal of Fe3O4@SiO2-NH2 core-shell nanomaterials prepared via a controllable sol-gel process
.
Chem. Eng. J.
215
,
461
471
.