Abstract

In this study, an ordered mesoporous silica modified with lanthanum oxide was synthesized using diatomite as silica source and applied for adsorption of phosphate from aqueous solution. By taking cost-effectiveness for practical application into consideration, the adsorbent with a theoretical La/SiO2 molar ratio of 0.2 (La0.2M41) possessed a promising performance. In the batch adsorption tests, the adsorbents with La2O3 loading possessed markedly enhanced adsorption capacities. Phosphate uptake by La0.2M41 was pH-dependent with the highest sorption capacities observed over a pH range of 3.0–6.0. Coexistent anions displayed an adverse effect on phosphate adsorption following the order of CO32−  > F  > NO3 > Cl > SO42−. In the kinetic study, phosphate adsorption onto La0.2M41 followed the pseudo-second-order equation better than the pseudo-first-order, suggesting chemisorption. The Langmuir isothermal model well described the adsorption isotherm data, showing a maximum adsorption capacity for phosphate of up to 263.16 mg/g at 298 K. In a real treated wastewater effluent with phosphate concentration of 2.5 mg P/L, La0.2M41 efficiently reduced the phosphate concentration to 28 µg P/L.

INTRODUCTION

The excessive presence of phosphate in water bodies, which often triggers eutrophication and algal bloom, lowers water quality and affects biodiversity in aquatic ecosystems (Gao et al. 2007). In this regard, several methods, including biological, physical, chemical and a combination of these methods, have been investigated to remove phosphate from aqueous solution. Enhanced biological treatment can achieve higher phosphate removal rate and it is low cost; however, the high microbial sensitivity to shock loadings and pH changes of wastewater makes the implementation of this process not feasible for wastewater treatment (Delgadillo-Mirquez et al. 2016). There are also many shortcomings in chemical methods. Phosphate can precipitate with calcium giving a Ca3(PO4)2 compound, but the competition for calcium ions between phosphate and organic ligands affects its removal efficiency (Lemlikchi et al. 2015). Large scale application of the H2O2/UV oxidation system for malathion degradation is often technically limited by the high values of pH and temperature (Chenna et al. 2016). Physical processes have proved to be challenging due to the high cost and limited removal performance (Kartashevsky et al. 2015). A combination of adsorption and precipitation led to rapid adsorption: the removal percentage of phosphate could reach a maximum in 5–10 min (Bouamra et al. 2018). The adsorption technique is regarded as one of the promising routes because of its simple operation, high efficiency and fast adsorption rate even at low P concentrations (Wu et al. 2017).

Recently, mesoporous materials have received considerable attention owing to their highly ordered structure, large specific surface areas and favorable surface chemical property for functionalization or modification (Huang et al. 2013). However, the preparation of mesoporous materials using toxic organic silica sources as precursor is expensive and unsafe. Natural materials are abundant and cheap, and can offer an environmental process and reduce the cost of synthesis. In recent years, efforts have been made to develop nontoxic and economical substitute precursors; some natural material such as kaolinite, vermiculite and coal fly ash can be used as silica source for the synthesis of mesoporous silica (Sanhueza et al. 2006; Hadjar et al. 2008; Sun et al. 2015; Yacob et al. 2016). As compared to pure mesoporous silica with little affinity for phosphate, metal doping or deposition is an effective way to enhance the phosphate adsorption capacities of adsorbents. In particular, lanthanum-modified adsorbents could provide a great quantity of coordination sites and exhibit good sensitivity affinity toward phosphate even at trace levels. The incorporation of lanthanum onto supports for phosphate adsorption have exhibited a series of advantages such as high removal rate, superior adsorption capacity, environmentally benign and wide operating pH range (Zhang et al. 2016).

Diatomite is a kind of natural siliceous sedimentary rock (Wang et al. 2009), which is relatively cheap, widely available and eco-friendly, and has been used in various industrial applications. The raw diatomite can be used as a suitable carrier or precursor material to improve the adsorption capacity. In this study, taking diatomite as the raw materials, the hexagonally ordered mesoporous silica was prepared by the hydrothermal method, and then functionalized with La2O3 by the ethanol evaporation method. The prepared materials were characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM) and N2 adsorption–desorption analysis. The fundamental adsorption behavior was investigated for phosphate removal from aqueous solutions, including adsorption isotherm, the adsorption kinetics, and the effect of initial pH and competing anions.

EXPERIMENTAL

Materials and chemicals

Diatomite was obtained from Jilin province, China. Its main chemical composition (wt%) is: SiO2, 90.66%; Fe2O3, 2.96%; Al2O3, 4.28%; K2O, 0.67%; and P2O5, 0.43%. The lanthanum nitrate was obtained from Sinopharm Chemical Reagent Co. Ltd (China). The phosphate stock solution (500 mg PO43−/L) and coexisting ion solutions were generated by dissolving KH2PO4 and their corresponding sodium salts into deionized water. All chemicals were of analytical grade and used without further purification.

Synthesis of MCM-41

The ordered mesoporous silica (MCM-41) material was prepared using diatomite as silica source and hexadecyltrimethylammonium bromide (CTAB) as the template. In a typical synthesis, diatomite (1.00 g) was dissolved in 8 mL of 1 M NaOH, and then stirred for 0.5 h at room temperature, followed by heating at 150 °C for 4 h. Then the aqueous solution of CTAB (2 g of CTAB dissolved in 15 mL of deionized water) was added into the above solution under vigorous stirring. The pH of the mixture was adjusted to 10. After stirring for 1 h at room temperature, the obtained homogeneous gel was transferred to an autoclave and hydrothermally treated under autogenous pressure at 105 °C for 24 h. It was then filtered, washed with deionized water and dried at 80 °C, followed by calcination at 550 °C for 6 h to remove the template.

Preparation of La2O3-modified MCM-41

The lanthanum nitrate was selected as the lanthanum precursor and incorporated into MCM-41 using the ethanol evaporation method (Huang et al. 2015). Typically, 0.5 g of MCM-41 was added into 100 mL of ethanol solution containing a desired amount of lanthanum precursor. The mixture was stirred at 60 °C for 24 h, and then the temperature was increased to 85 °C to evaporate the ethanol completely. The product was calcined at 550 °C for 5 h. The samples were presented as LaxM41, where x indicates the La/SiO2 molar ratio.

Characterization

The crystal structure of the sorbents was analyzed using an X-ray diffraction spectrometer (D/max-2400, Rigaku Corporation, Japan) with Cu Kα radiation (40 mA, 40 kV) over the 2θ range of 1–10°. TEM (JEM-2100, JOEL, Japan) was used to characterize the sorbents at an accelerating voltage of 20 kV. The N2 adsorption–desorption isotherms were measured at 77 K on a nitrogen adsorption apparatus (Micromeritics ASAP 2020).

Adsorption experiments

All batch adsorption experiments were conducted at 25 °C with La0.2M41 dosage of 0.6 g/L. Except for the initial pH study, all tests were performed under natural pH. The experiment was carried out to obtain optimal La/SiO2 molar ratio; 100 mL solutions were equilibrated with La/SiO2 molar ratio varying from 0 to 0.5. To investigate the effect of coexisting anions on phosphate adsorption, a certain amount of NaCl, Na2SO4, NaNO3, NaF and Na2CO3 was added to adjust to different ionic concentration. The effect of the initial pH on phosphate removal was analyzed at initial pH levels ranging from 2.0 to 10.0; 0.1 M HCl and 0.1 M NaOH solutions were used for the pH adjustments. The tests were performed for 3 h at the initial concentration of 50 mg PO43−/L. For kinetic studies, a certain concentration (30 mg PO43−/L, 50 mg PO43−/L, and 100 mg PO43−/L) was chosen as the initial concentration of phosphate solution; samples were drawn periodically for phosphate concentration analysis. Sorption isotherm experiments at temperature of 25 °C were performed to find out the maximum sorption capacity. The sorbents were dispersed into 100 mL of phosphate solution with initial concentration ranging from 30 to 300 mg PO43−/L. After adsorption reaction, the sample was filtered by using a 0.45 μm membrane filter and the phosphate concentrations in filtrates were analyzed by the ammonium molybdate spectrophotometric method. The adsorption capacity of adsorbent qe(mg/g) and the percentage of removal () were calculated by the following Equations (1) and (2): 
formula
(1)
 
formula
(2)
where C0 and Ce(mg/L) are the initial and equilibrium liquid-phase phosphate concentrations, respectively, V(L) is the volume of the solution and m(g) is the mass of the dry sorbent used.

RESULTS AND DISCUSSION

The phosphate adsorption capacity and removal percentage for MCM-41 and LaxM41 are presented in Figure 1. It clearly shows that all of the LaxM41 adsorbents possess higher phosphate removal capacities, as compared to that of MCM-41. The phosphate adsorption is mainly attributed to the incorporation of La2O3 onto mesoporous silica materials. It is noted that the successive increase of phosphate adsorption capacity was dramatic as the La/SiO2 molar ratio increased from 0 to 0.2, followed by a limited increase for La0.3M41. This is likely due to the higher La2O3 loadings in the resulting MCM-41. However, with further increasing La/SiO2 molar ratio to 0.5, the uptake amount of phosphate demonstrated a slight decrease. This indicated excessive La2O3 loading might attack and partially destruct ordered porous frameworks of silica resulting in reduced access to adsorption sites inside ruined pores and in turn lower adsorption. Considering the efficiency of La usage, we selected the adsorbent La0.2M41 for further study on its phosphate adsorption performance.

Figure 1

The phosphate adsorption capacity and removal percentage for MCM-41 and LaxM41.

Figure 1

The phosphate adsorption capacity and removal percentage for MCM-41 and LaxM41.

Characterization of adsorbent

X-ray diffraction

The XRD patterns of MCM-41, La0.2M41 and La0.5M41 are shown in Figure 2. For MCM-41, four typical diffraction peaks were observed: a prominent peak of (100) planes as well as three weaker peaks of (110), (200) and (210), indicating the typical hexagonal long-range arrays structure of mesoporous material. However, compared with patterns of MCM-41, the intensity of the (100) peak became weaker and the (110) (200) (210) peaks disappeared for La0.2M41, and all typical diffraction peaks disappeared for La0.5M41. This indicated that there is no significant impact on hexagonal structure when a moderate amount of La2O3 is incorporated into MCM-41. However, a large amount of La2O3 incorporated into the framework of doped MCM-41 would lead to the collapse of the hexagonal long-range arrays structure and even a disordered structure. A similar finding has also been reported that distortion of the flexible silica framework is caused by the stress arising from a greater amount of metal oxide impregnated into the mesopores (Huang et al. 2014).

Figure 2

X-ray diffraction patterns of MCM-41, La0.2M41 and La0.5M41.

Figure 2

X-ray diffraction patterns of MCM-41, La0.2M41 and La0.5M41.

Transmission electron microscopy

The TEM image of MCM-41 and La0.2M41 are shown in Figure 3. The perfectly parallel well-organized mesopore channels with streak structural features were observed in MCM-41 (Figure 3(a)). Figure 3(b) shows mesopores with long channel structure are preserved after La2O3 is introduced into MCM-41. No separate dark spots outside of MCM-41 particles can be observed, implying the lanthanum oxides are homogeneously dispersed in the pore channels. However, hexagonal arrays of mesopores are decreased with La2O3 in the mesoporous structure. This is in good agreement with the results obtained from the XRD analyses.

Figure 3

TEM images of MCM-41 (a) and La0.2M41 (b).

Figure 3

TEM images of MCM-41 (a) and La0.2M41 (b).

N2 adsorption–desorption isotherms

The N2 adsorption–desorption isotherms of MCM-41 and La0.2M41 are shown in Figure 4. Both samples exhibit typical type IV isotherms with hysteresis loops, as expected for mesoporous materials. The quantity adsorbed of the adsorption isotherms increased sharply in the range between 0.3 and 0.45 of relative pressure (P/P0), representing capillary condensation of nitrogen within the uniform mesopore. In Table 1, the Brunauer–Emmett–Teller (BET) surface area, pore diameter and total pore volume of MCM-41 are 815.42 m2/g, 4.32 nm and 0.82 cm3/g, respectively, whilst those of La0.2M41 are 323.67 m2/g, 4.25 nm and 0.31 cm3/g, respectively. Obviously, structure characteristics were changed after being incorporated with lanthanum, which reduces the BET surface and pore volumes and narrows the pore sizes.

Table 1

Structure properties of MCM-41 and La0.2M41

Sample BET surface area (m2/g) Pore diameter (nm) Total pore volume (cm3/g) 
MCM-41 815.42 4.32 0.82 
La0.2M41 323.67 4.25 0.31 
Sample BET surface area (m2/g) Pore diameter (nm) Total pore volume (cm3/g) 
MCM-41 815.42 4.32 0.82 
La0.2M41 323.67 4.25 0.31 
Figure 4

N2 adsorption–desorption isotherm of MCM-41 (a) and La0.2M41 (b).

Figure 4

N2 adsorption–desorption isotherm of MCM-41 (a) and La0.2M41 (b).

Phosphate adsorption

Adsorption kinetics

Adsorption kinetics helps to evaluate the adsorption rate and understand the adsorption mechanism. Figure 5(a) illustrates the phosphate adsorption on La0.2M41 at different contact times in the solution with initial phosphate concentrations of 30, 50 and 100 mg/L. As seen, the adsorption process could be divided into two stages, a rapid stage at the very beginning followed by a gradually slower stage until the adsorption equilibrium was achieved with the increase of time. The extremely fast phosphate uptake in the beginning time attributed to a higher driving force, provided by higher phosphate concentration, to overcome mass transfer resistance of ions from the solution to active sites. As time proceeded, the concentration gradient became reduced and the diffusion rate became slow owing to the accumulation of phosphate adsorbed on the surfaces sites, resulting in the reduction in adsorption rate at the later stages.

Figure 5

(a) Adsorption of phosphate on La0.2M41 as a function of contact time at various initial phosphate concentrations. (b) The simulated pseudo-first-order kinetics. (c) The simulated pseudo-second-order kinetics.

Figure 5

(a) Adsorption of phosphate on La0.2M41 as a function of contact time at various initial phosphate concentrations. (b) The simulated pseudo-first-order kinetics. (c) The simulated pseudo-second-order kinetics.

To further examine the phosphate adsorption of La0.2M41. The pseudo-first-order and pseudo-second-order models were used to investigate the adsorption kinetics of phosphate. The equations used are as follows (Equations (3) and (4)).

The pseudo-first-order model: 
formula
(3)
The pseudo-second-order model: 
formula
(4)
where qt(mg/g) and qe(mg/g) are the amounts of phosphate adsorbed at time t(h) and at equilibrium, respectively; k1 and k2 are the pseudo-first-order and pseudo-second-order rate constant, respectively.

The kinetic data recorded in the phosphate solution with different initial concentrations are fitted into both of the pseudo-first-order and pseudo-second-order models (Figure 5(b) and 5(c)); and the corresponding parameters and correlation coefficients are listed in Table 2. The results are found to better match with the pseudo-second-order model (R2 > 0.99) rather than the pseudo-first-order model (R2 > 0.94), suggesting that chemisorption occurred between phosphate and La0.2M41 and the process was controlled by surface chemical reaction.

Table 2

Parameters for kinetic models of phosphate adsorption

C0 (mg/L) qe(exp) (mg/g) First-order kinetics
 
Second-order kinetics
 
k1 (min−1qe(cal) (mg/g) R2 k2 (g/(mg·min)) qe(cal) (mg/g) R2 
30 49.85 0.0203 55.09 0.9408 6.201 × 10−4 54.05 0.9972 
50 82.92 0.0093 38.42 0.9736 5.032 × 10−4 86.21 0.9998 
100 152.14 0.0051 97.19 0.9888 9.565 × 10−5 161.29 0.9988 
C0 (mg/L) qe(exp) (mg/g) First-order kinetics
 
Second-order kinetics
 
k1 (min−1qe(cal) (mg/g) R2 k2 (g/(mg·min)) qe(cal) (mg/g) R2 
30 49.85 0.0203 55.09 0.9408 6.201 × 10−4 54.05 0.9972 
50 82.92 0.0093 38.42 0.9736 5.032 × 10−4 86.21 0.9998 
100 152.14 0.0051 97.19 0.9888 9.565 × 10−5 161.29 0.9988 

Adsorption isotherms

The equilibrium adsorption isotherm is usually employed to illustrate the interaction between adsorbates and adsorbents and deduce the capacity of the adsorbent when designing adsorption systems. To better understand the characteristics of the adsorption process, the Langmuir and the Freundlich models were used to investigate the adsorption isotherms of phosphate. The equations used are as follows (Equations (5) and (6)): 
formula
(5)
 
formula
(6)
where Ce(mg/L) is the equilibrium concentration of phosphate, qe(mg/g) is the amount adsorbed under equilibrium, qm(mg/g) is the theoretical maximum adsorption capacity of the adsorbent for phosphate, and KL(L/mg) is a Langmuir binding constant related to the energy of adsorption; KF and n are the Freundlich empirical constants. The empirical constant n represents the adsorption intensity and the type of isotherm to be favorable (0.1 < 1/n <0.5) or unfavorable (1/n > 2).

The Langmuir and Freundlich isotherm fitting plots of La0.2M41 at various initial phosphate concentrations are presented in Figure 6. The corresponding isotherm constants at 25 °C are summarized in Table 3. As observed, the fitting results of Langmuir model were more satisfactory due to high correlation coefficient values (R2 > 0.99), indicating that the adsorption process occurs on the surface of La0.2M41 via homogeneous monolayer adsorption. In addition, the n value was larger than 1, indicative of high adsorption intensity (Benhamou et al. 2009).

Table 3

Langmuir and Freundlich parameters for phosphate adsorption

Temperature (°C) Langmuir model
 
Freundlich model
 
qm (mg/g) KL (L/mg) R2 KF (mg/g) R2 
25 °C 263.16 85.47 0.9984 93.40 4.34 0.9367 
Temperature (°C) Langmuir model
 
Freundlich model
 
qm (mg/g) KL (L/mg) R2 KF (mg/g) R2 
25 °C 263.16 85.47 0.9984 93.40 4.34 0.9367 
Figure 6

The Langmuir adsorption isotherm (a) and Freundlich adsorption isotherm (b) fitting plots of La0.2M41 samples.

Figure 6

The Langmuir adsorption isotherm (a) and Freundlich adsorption isotherm (b) fitting plots of La0.2M41 samples.

Effect of initial pH

The initial pH is a key parameter during the adsorption process. The effect of pH on phosphate capacity of La0.2M41 is presented in Figure 7(a). Clearly, the adsorption capacity increased drastically when the initial pH changed from 2.0 to 3.0, and then reached a plateau with pH from 3.0 to 6.0 with a high phosphate adsorption capacity of around 81.5 mg/g. With a further increase of pH from 6.0 to 10.0, the sorption started to reduce at a relatively fast speed, and declined to 67.87 mg/g as the pH approached 10.0. This is because, at different solution pH, phosphate can exist in the form of H3PO4, H2PO4, HPO42− and PO43− (Zhang et al. 2010). At pH ≤ 3.0, the predominant species of phosphate is H3PO4 which is weakly attached to the sites of the sorbent. With increase of solution pH, the dominant phosphate species is monovalent H2PO4. The high adsorption capacities observed between pH 3.0 and 6.0 indicate that La0.2M41 provides a great affinity to the single charged phosphate species (H2PO4). H2PO4 could be captured onto La0.2M41 by electrostatic attraction and ligand exchange. In addition, the decreased phosphate uptake in the pH range of 6.0 and 10.0 was due to electrostatic repulsion between the deprotonated surface and highly charged phosphate. At higher pH, the high concentration of hydroxide did not favor the protonation of lanthanum oxide and will strongly compete with phosphate on active sites (Qiu et al. 2017).

Figure 7

(a) Effect of initial pH on phosphate adsorption of La0.2M41. (b) Effect of coexisting anions on the phosphate adsorption capacity. (c) Effect of contact time on phosphate removal in the treated wastewater effluent by using La0.2M41.

Figure 7

(a) Effect of initial pH on phosphate adsorption of La0.2M41. (b) Effect of coexisting anions on the phosphate adsorption capacity. (c) Effect of contact time on phosphate removal in the treated wastewater effluent by using La0.2M41.

Effect of coexisting anions

Various types of inorganic anion are present in natural water, which may interfere in the removal of phosphate through competing for adsorption sites on the adsorbent surface. The influence of competitive anions Cl, SO42−, NO3, F and CO32− on the phosphate adsorption capacity of La0.2M41 is presented in Figure 7(b). The influence of anion on phosphate adsorption followed the order: CO32−  > F  > NO3 >  Cl > SO42− . The removal efficiency of phosphate was 91.74% without coexisting ions. When the concentration of Cl, SO42−, NO3, F and CO32− was 500 mg/L, the removal ratio of phosphate decreased to 83.50%, 84.12%, 81.00%, 41.68% and 15.33%, respectively. Compared with the other four anions, CO32− might have higher affinity to the adsorbent to compete with phosphate for adsorption sites. This is attributed to the lower Ksp (solubility product constant) of La2(CO3)3 compared with that of LaPO4, which favors the displacement of adsorbed PO43− on La0.2M41 by CO32− ions and the subsequent transformation of formed LaPO4 to La2(CO3)3, thus deteriorating the phosphate adsorption capacity (Huang et al. 2015). In addition, the initial pH of the solution with 500 mg/L CO32− is above 10.0. The presence of a large amount of OH may compete with PO43− for adsorption active sites, thus hindering the ligand-exchange mechanism and lowering phosphate adsorption.

Phosphate removal from treated wastewater effluent

In order to prevent eutrophication, different countries and regions have applied strict P effluent limits, and there is a need for wastewater facilities to meet a very low phosphate limit, e.g. the discharge level of 50 μg P/L or less. Based on enhanced phosphate removal performance of the La0.2M41, the material's applicability for treatment of real wastewater effluent was also tested. The sewage contained a number of ionic ingredients including Cl, NO3 and SO42− as shown in Table 4. It had a pH of 7.26, a chemical oxygen demand of 45.72 mg/L, and a phosphate concentration of around 2.2 mg/L, respectively. As shown in Figure 7(c), 97.65% of the final adsorption capacity of La0.2M41 is reached in the first 10 min at a dosage of 0.6 g/L, suggesting that the phosphate removal rate is dramatically fast. The whole adsorption process reaches the equilibrium in 40 min and the concentration of phosphate was reduced from 2.2 mg P/L to 28 μg P/L, which satisfies the stringent phosphate discharge limit of 50 μg P/L. The La0.2M41 shows great potential as a highly efficient and fast-acting adsorbent for phosphate removal.

Table 4

Some soluble components (mg/L) in the real wastewater effluent

Cl NO3 SO42− Na+ K+ Ca2+ Mg2+ 
123.82 63.21 151.86 82.37 29.85 102.37 128.32 
Cl NO3 SO42− Na+ K+ Ca2+ Mg2+ 
123.82 63.21 151.86 82.37 29.85 102.37 128.32 

CONCLUSIONS

La2O3-modified mesoporous silica was prepared and the adsorption of phosphate was investigated in batch systems. The phosphate adsorption equilibrium data were fitted better with the use of the Langmuir model than the Freundlich model, and the maximum adsorption capacity of phosphate reached 263.16 mg/g. In addition, the adsorption kinetic data were better described by the pseudo-second-order model, suggesting that mainly chemisorption contributes to phosphate adsorption. The solution pH values had a major impact on phosphate adsorption with optimal removal observed around pH 3.0–6.0. Coexisting ions decreased the phosphate adsorption capacity. Moreover, the use of La0.2M41 could lower the phosphate concentration of 28 μg P/L in a real wastewater; after treatment, residual phosphate concentration is well below stringent discharge standards of P.

According to previous research, the main problem is that the cost of LaxM41 preparation is high (Zhang et al. 2010). In this study, preparation of mesoporous sorbents using natural diatomite as precursor greatly reduces costs and is more environmentally friendly. Moreover, superior phosphate adsorption capacities were achieved in a shorter adsorption time.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the financial support of the National Natural Science Foundation (21667017 and 51741805).

REFERENCES

REFERENCES
Benhamou
A.
,
Baudu
M.
,
Derriche
Z.
&
Basly
J. P.
2009
Aqueous heavy metals removal on amine-functionalized Si-MCM-41 and Si-MCM-48
.
J. Hazard. Mater.
171
,
1001
1008
.
Bouamra
F.
,
Drouiche
N.
,
Abdi
N.
,
Grib
H.
,
Mameri
N.
&
Lounici
H.
2018
Removal of phosphate from wastewater by adsorption on marble waste: effect of process parameters and kinetic modeling
.
Int. J. Environ. Res.
12
,
1
15
.
Delgadillo-Mirquez
L.
,
Lopes
F.
,
Taidi
B.
&
Pareau
D.
2016
Nitrogen and phosphate removal from wastewater with a mixed microalgae and bacteria culture
.
Biotechnol. Rep.
11
,
18
26
.
Gao
G.
,
Qin
B.
,
Sommaruga
R.
&
Psenner
R.
2007
The bacterioplankton of Lake Taihu, China: abundance, biomass, and production
.
Hydrobiologia
581
,
177
188
.
Hadjar
H.
,
Hamdi
B.
,
Jaber
M.
,
Brendie
J.
,
Kessaissia
Z.
,
Balard
H.
&
Donnet
J. B.
2008
Elaboration and characterisation of new mesoporous materials from diatomite and charcoal
.
Micropor. Mesopor. Mater.
107
,
219
226
.
Huang
W.
,
Zhu
Y.
,
Tang
J.
,
Yu
X.
,
Wang
X.
,
Li
D.
&
Zhang
Y.
2014
Lanthanum-doped ordered mesoporous hollow silica spheres as novel adsorbents for efficient phosphate removal
.
J. Mater. Chem. A
2
,
8839
8848
.
Huang
W.
,
Yu
X.
,
Tang
J.
,
Zhu
Y.
,
Zhang
Y.
&
Li
D.
2015
Enhanced adsorption of phosphate by flower-like mesoporous silica spheres loaded with lanthanum
.
Micropor. Mesopor. Mater.
217
,
225
232
.
Lemlikchi
W.
,
Drouiche
N.
,
Baaziz
B.
&
Mecherri
M. O.
2015
Formation of mixed complexes of type phosphate-Ca-dye
.
Sep. Sci. Technol.
50
,
2676
2679
.
Sanhueza
V.
,
López-Escobar
L.
,
Kelm
U.
&
Cid
R.
2006
Synthesis of a mesoporous material from two natural sources
.
J. Chem. Technol. Biotechnol.
81
,
614
617
.
Wang
Y.
,
Shang
Y.
,
Zhu
J.
,
Wu
J.
,
Ji
S.
&
Meng
C.
2009
Synthesis of magadiite using a natural diatomite material
.
J. Chem. Technol. Biot.
84
,
1894
1898
.
Zhang
J.
,
Shen
Z.
,
Shan
W.
,
Chen
Z.
,
Mei
Z.
,
Lei
Y.
&
Wang
W.
2010
Adsorption behavior of phosphate on Lanthanum(III) doped mesoporous silicates material
.
J. Environ. Sci. China
22
,
507
511
.