Preparation and characterization of lanthanum(III) loaded granular ceramic for phosphorus adsorption from aqueous solution

Preparation and characterization of lanthanum(III) loaded granular ceramic for phosphorus adsorption from aqueous solution

Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789 Contents lists available at SciVerse ScienceDirect Journal of the Taiwan Ins...

1MB Sizes 0 Downloads 8 Views

Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789

Contents lists available at SciVerse ScienceDirect

Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice

Preparation and characterization of lanthanum(III) loaded granular ceramic for phosphorus adsorption from aqueous solution Nan Chen a,b, Chuanping Feng a,b,*, Zhenya Zhang c, Ruopeng Liu b, Ya Gao b, Miao Li d, Norio Sugiura c a

Key Laboratory of Groundwater Cycle and Environment Evolution (China University of Geosciences (Beijing)), Ministry of Education, Beijing 100083, China School of Water Resources and Environment, China University of Geosciences (Beijing), Beijing 100083, China c Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba 305-8572, Japan d School of Environment, Tsinghua University, Beijing 100084, China b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 5 January 2012 Received in revised form 20 March 2012 Accepted 8 April 2012 Available online 12 May 2012

A La(III)-loaded granular ceramic adsorbent, consisting mainly of different forms of metal oxide minerals, was developed for phosphorus removal from aqueous solution. Batch experiments were performed to investigate the effects of various experimental parameters, such as contact time (3–48 h), initial phosphorus concentration (5–25 mg/L), pH (2.0–12.0), temperature (20, 30 and 40 8C) and anions on phosphorus adsorption. The adsorption process was well described by a pseudo-second-order kinetic model and equilibrium was achieved at 30 h. The adsorption data closely fitted the Langmuir isotherm model at temperatures ranging from 20 8C to 40 8C. Thermodynamic study indicated a spontaneous, favorable and exothermic adsorption on the La(III)-loaded granular ceramic adsorbent. The optimum pH for phosphorus removal ranged from 7.0 to 9.0. Phosphorus adsorption was impeded by the presence of F, followed by Cl, SO42 and NO3 and the adsorption process appeared to be controlled by a chemical precipitation process. The mechanism may involve ion complexation during subsequent adsorption of phosphorus on lanthanum hydroxides. ß 2012 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Keywords: La(III)-loaded Granular ceramic Kinetic Adsorption Phosphorus adsorption Thermodynamic

1. Introduction During recent years, considerable amounts of soluble phosphorus in waste streams from households, agricultures and industries have often been responsible for eutrophication problems [1–3]. Various methods, including biological processes [4], ion-exchange [1], tertiary filtration [5], chemical precipitation, electro-coagulation [6], crystallization [7], and adsorption [8,9], have been applied to remove phosphorus from water and wastewater. Adsorption is an attractive method for removal of phosphorus because of its high removal efficiency, relatively simplicity, economy in application and straightforward operation. Several adsorbents have been tested for the removal of phosphorus from aqueous solution, namely aluminum oxide hydroxide [10], iron oxides [11,12], zeolite [13], Fe–Mn binary oxide [14], fly ash [15], a polymeric ligand exchanger [16], zirconium hydroxide [17] and basic oxygen furnace slag [18]. However, most of these adsorbents may be not suitable for practical application owing to their financial and

* Corresponding author at: School of Water Resources and Environment, China University of Geosciences (Beijing), No. 29 Xueyuan Road, Haidian District, Beijing 100083, China. Tel.: +86 10 82322281; fax: +86 10 82321081. E-mail address: [email protected] (C. Feng).

technological constraints, such as high cost, low removal efficiency and narrow pH range of operation [10,19]. Lanthanum has a known affinity toward phosphorus. In the 1970s, Melnyk et al. [20] and Recht et al. [21] reported that phosphate precipitation by lanthanum was more effective over a wider pH (4.5–8.5) range than with either Fe(III) or aluminum salts. Lanthanum compounds have been used in water treatment processes, because La is less expensive than other rare earth elements (REEs) and the point of zero charge of lanthanum oxide is higher than that of other well-known adsorbents [22]. Encai et al. [23] indicated that phosphate removal would be near 100% with 0.3 g/L La-meso-SiO2 with a Si/La molar ratio 10 after 3 h. Li et al. [24] found that the adsorptive capacity was 1.32 mg/g in lanthanum-doped vesuvianite (La/vesuvianite mass ratio 0.14) at pH between 6.0 and 9.0 after 40 h. Haghseresht et al. [25] reported that the monovalent phosphate anion, H2PO4, had the greatest affinity for a lanthanum-modified bentonite surface. However, most of the above adsorbents for P adsorption are fine powders or fibers, which are not suitable for filtration in a column adsorption reactor as they will block during sewage treatment in practical applications. Thus, in the present study, we introduced lanthanum as a component for preparation of granular ceramic materials (with 3–5 mm particle diameter) as an adsorbent for phosphorus removal from aqueous solution. Batch studies were conducted to gain insights into the phosphorus uptake behavior of the

1876-1070/$ – see front matter ß 2012 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jtice.2012.04.003

784

N. Chen et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789

2. Materials and methods

method at 700 nm (Hitachi-DR/5000, USA) [27]. A surface morphology analysis was carried out using scanning electron microscopy (SEM) (SSX-550, Shimadzu, Japan). The specific surface areas of samples were determined by the Brunauer–Emmett– Teller (BET) method with N2 adsorption (Coulter SA3100, Japan). The quantitative element and mapping analysis was performed using an electron probe microanalyzer (EPMA1600, Shimadzu, Japan). The residual metal elements were analyzed by an inductively coupled plasma optical emission spectrometer (ICPOES) (iCAP6300, Thermol, USA).

2.1. Preparation of La-loaded granular ceramic materials

3. Results and discussion

Loess was supplied by the Sunrise store (Beijing, China) (particle size less than 140 mesh), montmorillonite (grain size less than 100 mesh), soluble starch and lanthanum nitrate hexahydrate (La(NO3)36H2O) were supplied by the Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). Briefly, the La(III)loaded samples were made from the prepared powders (montmorillonite, loess, soluble starch and La(NO3)36H2O, at 40%:30%:20%:10% (in weight)). Montmorillonite and loess, mainly consisting of silicon and aluminum oxide minerals, were used as both aggregates and binders of the ceramic; soluble starch was used as a porosifier to create additional pores after the green granules were fired at high temperature; La(NO3)36H2O was introduced as an additive to load La, which was expected to form a hydrous lanthanide oxide complex that would enhance the adsorption ability toward phosphorus. The prepared powders were mixed with ultrapure water and kneaded into spherical shapes (gray and spherical, with diameter of 3–5 mm) by hand. The granular ceramic samples were then dried at 105 8C for 24 h in an oven, resulting essentially in a starch paste (the soluble starch granules swelled and consolidated), hence increasing porosity after sintering [26]. Finally, the dried ceramic samples were sintered at 600 8C for 1 h in a muffle furnace and cooled in a desiccator to room temperature for further studies.

3.1. Effect of initial phosphorus concentration and adsorbent dose

La-loaded granular ceramic adsorbent by studying its physicochemical properties, adsorption kinetics and equilibrium. A mechanism for phosphorus adsorption is proposed, based on data from adsorption isotherms and scanning electron microscopy (SEM) analyses. In addition, the effect of pH and coexisting anions (F, Cl, NO3, and SO42) on the phosphorus adsorption process was explored and the key behavior and morphology characteristics of the adsorbent are described.

2.2. Batch adsorption experiments All reagents used in this study were of analytical grade. Stock P solution (100 mg/L) was prepared initially by dissolving 0.7741 g of NaH2PO4 in a 2 L volumetric flask using ultrapure water. All solutions for phosphorus removal experiments and analysis were prepared by an appropriate dilution from the stock solution. Batch experiments to study the removal of phosphorus from solution were carried out by reacting 100 mL of P solution in 250 mL conical flasks with 1.0 g of the adsorbent. The flasks were immersed in a water bath at predetermined temperatures (20, 30, 40 8C). The aqueous samples were taken at known intervals and then filtered through a 0.45 mm cellulose acetate filter. The effect of initial phosphorus concentration and adsorbent dose was studied at pH 7.0  0.2 and 20 8C. The effect of contact time (3–48 h) was examined at 20 8C and 40 8C with an initial P concentration of 10 mg/L. Adsorption isotherm studies were conducted by varying the initial P concentration (5–25 mg/L) at several temperatures (20, 30 and 40 8C). The effect of pH was investigated by adjusting the pH from 2.0 to 12.0 using 0.1 M NaOH and HCl solutions, with an initial P concentration of 10 mg/L. Finally, the effect of other competitive anions (Cl, NO3, F, and SO42) was studied with an initial P concentration of 10 mg/L.

Fig. 1 shows the effect of initial P concentration and adsorbent dose on P removal by the La(III)-loaded granular ceramic adsorbent. The phosphorus adsorption percentages at equilibrium increased from 19.01 to 98.2%, 14.82 to 96.02% and 10.11 to 89.96% as the adsorbent dose increased from 2.5 to 20 g/L for all the initial P concentrations (5, 15 and 25 mg/L), respectively. The increase in P removal with the increase in adsorbent dose is the consequence of the greater amount of surface area and available binding sites on the surface of the adsorbent. 3.2. Effect of contact time The P adsorption on La(III)-loaded granular ceramic adsorbent was carried out with 2.0 g of adsorbent and 200 mL of initial P solution (10 mg/L) at 20 and 40 8C. The results in Fig. 2 show that adsorbed P increased with an increase in contact time. The adsorption process can be divided into two steps: a rapid initial adsorption (3–9 h) and a much slower subsequent adsorption process. The second, slower adsorption process was likely due to the decrease in adsorption sites on the surface of the adsorbents. About 70% of the P was removed in 9 h at 40 8C, while only 50% was removed at 20 8C. Both of the P concentrations remained relatively constant with contact time after 30 h at 20 and 40 8C. Therefore, subsequent adsorption experiments under other conditions were a conducted with the contact time of 30 h. Fig. 2 also reveals that the adsorption rate increased with temperature. Haghseresht et al. [25] observed a similar trend when they used lanthanum-modified bentonite for P removal. Other investigators also reported that the P adsorption rate and capacity increased with temperature (e.g., ZnCl2 activated coir pith carbon [28] and dolomite [29]). 3.3. Adsorption kinetic study Using experimental kinetic data, both pseudo-first-order and pseudo-second-order mechanisms have been used to elucidate the

2.3. Analytical methods Solution pH was monitored with a standard pH meter (ORION 8157BNUMD, USA). Phosphorus analysis was carried out with a UV–vis spectrophotometer using the molybdenum antimony

Fig. 1. Effect of initial P concentration and adsorbent dose on P adsorption by La(III)-loaded granular ceramic adsorbent (initial solution pH = 7.0  0.2, temperature = 20  1 8C).

N. Chen et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789

Fig. 2. Effect of contact time on P adsorption by La(III)-loaded granular ceramic adsorbent (initial P concentration = 10 mg/L, adsorbent dosage = 10 g/L, initial solution pH = 7.0  0.2).

Fig. 3. Effect of solution pH for P adsorption by La(III)-loaded granular ceramic adsorbent (initial P concentration = 10 mg/L, adsorbent dosage = 10 g/L, temperature = 20  1 8C).

mechanisms of adsorption and potential rate controlling steps [30,31]. logðqe  qt Þ ¼

log qe  k1 t 2:303

According to the calculated results in Table 1, the removal of phosphorous follows a pore diffusion process, because the values of D (9.26  1013 cm2/s) are in the range of 1012–1013 cm2/s for both temperatures of 20 and 40 8C.

(1)

t 1 t ¼ þ qt k2 qe 2 qe

3.4. Effect of solution pH

(2)

Because the adsorption of anions is usually influenced by the solution pH, it is important to assess the adsorption behavior of phosphorus in solution at different pH values. The adsorption capacity first increased with increasing pH from 2.01 to 6.90 and reached a plateau over the pH range from 6.90 to 9.11, then significantly decreased from 9.11 to 12.01 (Fig. 3). The adsorption process being favored in neutral or slightly alkaline solutions may be attributed to the equilibrium pH value (about 10.50 on average) (Fig. 3) of La(III)-loaded granular ceramic adsorbent, which is approximately the zero point charge under alkaline conditions. Moreover, phosphate acid undergoes dissociation at different pHs, yielding different species (Eq. (4)) [35]:

where qt and qe are the amounts of adsorbed phosphorus (mg/g) at time t and at equilibrium time, respectively; and k1, k2 are the firstorder and second-order rate constants for P adsorption. The results are shown in Table 1. The correlation coefficient values for the pseudo-second order adsorption model (R2 = 0.9942 and 0.9973) are slightly higher than those obtained from the pseudofirst model (R2 = 0.9898 and 0.9919). The calculated equilibrium adsorption capacities qe (0.7300 and 0.8900 mg/g) are consistent with the experimental values (0.9871 and 1.0476 mg/g). Therefore, it can be concluded that pseudo-second order adsorption model was superior for describing the adsorption process kinetics of the La(III)loaded granular ceramic adsorbent. Similar results were reported by Rodrigues [32] and Zhang et al. [33]. Assuming spherical geometry for the adsorbent, the time for half adsorption can be correlated to the pore diffusion coefficient [34]: t 1=2 ¼

0:03r 0 2 D

785

pK a1

pK a2

pK a3

H3 PO4 !H2 PO4 2 þ Hþ !HPO4  þ 2Hþ !PO4 3 þ 3Hþ

(4)

where pKa1 = 2.15, pKa2 = 7.20 and pKa3 = 12.33, respectively. When the pH  2.15, the predominant species of phosphorus is H3PO4, which is weakly attached to the sites of the adsorbent. The adsorption capacity increased when the concentration of the H2PO4 and HPO42 species was increased (Fig. 3), indicating that the lanthanum ions had a greater affinity for the monovalent dihydrogen and phosphate [25]. When the pH is in the range 9.11– 12.01, lanthanum exists mainly in the insoluble form, La(OH)3, which can only sparingly form La(PO4)3 (note: the solubility of La(OH)3 is less than La(PO4)3).

(3)

where t1/2 is the time for half adsorption (s), r0 is the radius of the adsorbent particle (cm), and D is the diffusion coefficient (cm2/s). Values of D have been calculated for various phosphorous concentrations and temperatures. The radius of the adsorbent particle was selected to be about 0.3 cm in the experiment.

Table 1 Kinetic constants and pore diffusion coefficient for P adsorption on La(III)-loaded granular ceramic adsorbent at initial solution pH 7.0  0.2, initial P concentration 10 mg/L, adsorbent dosage 10 g/L. Kinetic constants for La(III)-loaded granular ceramic adsorbent Temperature (8C)

20 40

Experiment qe,exp (mg/g)

0.7300 0.8900

Pseudo-first-order

Pseudo-second-order

k1 (1/h)

qe,cal (mg/g)

R2

k2 [g/(mg h)]

qe,cal (mg/g)

R2

0.1699 0.1568

1.0316 1.2712

0.9898 0.9919

0.1184 0.2106

0.9871 1.0476

0.9942 0.9973

Pore diffusion coefficient for La(III)-loaded granular ceramic adsorbent Phosphorous conc. (mg/L)

Temperature (8C)

Equilibrium time (min)

t1/2 (s)

D  1013 (cm2/s)

10

20/40

1800

54,000

9.2593

qe,exp and qe,cal represent the measured value and predicted by kinetic model value of solid phase phosphorus concentrations at equilibrium.

786

N. Chen et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789

3.5. Adsorption isotherm study The data obtained from the experiments was used to analyze the adsorption isotherms to estimate the adsorption maxima, constants and density. The data was fitted to both the Langmuir and the Freundlich isotherm models described in Eqs. (5) and (6), respectively [36,37]. 1 1 1 ¼ þ qe qm BC e qm log qe ¼ log K f þ

(5) 1 log C e n

Table 2 Langmuir and Freundlich isotherm parameters for the adsorption on La(III)-loaded granular ceramic adsorbent at initial P concentration 5–25 mg/L, adsorbent dosage 10 g/L and temperature 20–40 8C. Temperature (8C) Langmuir isotherm

Freundlich isotherm

qm (mg/g) 1/B (mg/L) R2 20 30 40

0.8515 0.8722 0.9015

0.3734 0.2929 0.1667

Kf (mg/g)(L/mg)n n

0.9896 0.5527 0.9919 0.5871 0.9900 0.6483

R2

5.6402 0.9645 5.9988 0.9494 6.7751 0.9341

(6)

where qe (mg/g) is the amount of solute adsorbed per unit weight of material; Ce (mg/L) is the equilibrium concentration of phosphorus; qm (mg/g) and B (L/mg) are the Langmuir constants corresponding to a saturated monolayer adsorption capacity and the binding energy of the adsorption system, respectively; Kf and n are empirical constants in the Freundlich isotherms, indicating the adsorption capacity and adsorption intensity, respectively. Langmuir and Freundlich isotherms are plotted in Fig. 4, and the parameters can be calculated from a linear plot of 1/qe vs. 1/Ce and log qe vs. log Ce. These isotherm parameters are shown in Table 2. The applicability of the isotherm equations was compared by evaluating the correlation coefficient values (R2 ranges between 0.9341 and 0.9919). Both indicate a good fit of the data by both the Langmuir and Freundlich isotherm models but the Langmuir isotherm yields the better fit. In previous studies, many researchers reported that phosphorus adsorption on a range of adsorbents fits the Langmuir equation. Li et al. [24] reported that phosphate adsorption on lanthanum doped vesuvianite was fitted better by the Langmuir equation than by the Freundlich equation. Moreover, the increase in the maximum adsorption capacity with increasing solution temperature indicates that the adsorption of phosphorus by the La(III)-loaded granular ceramic adsorbent most likely occurs through chemical rather than physical interactions. Similar results were reported for phosphorus adsorption on lanthanum-modified bentonite [25], bauxsol [38], Mg/Mn-layered

Fig. 4. Isotherm modeling of P adsorption by La(III)-loaded granular ceramic adsorbent: (a) Langmuir isotherm; (b) Freundlich isotherm (adsorbent dosage = 10 g/L, initial P concentration = 5–25 mg/L, temperature = 20–40 8C).

double hydroxides [39] and mesoporous ZrO2 [40] for phosphorus adsorption. The inverse exponent n in the Freundlich isotherm is related to the strength and intensity of phosphorus adsorption. The calculated 1/n value lies in the range between 0.15 and 0.18, denoting favorable adsorption of P onto the La(III)-loaded granular ceramic adsorbent. Greater values of n indicate stronger bonds. Table 2 shows that n increases with increasing solution temperature, suggesting that a high temperature would strengthen the adsorption bond between P and the surface sites on the adsorbent [41]. 3.6. Effect of coexisting anions In practice, phosphorus-contaminated water contains several other anions that may influence the adsorption process. This study assessed P adsorption behavior in the presence of 100 mg/L salt solutions of chloride, nitrate, fluorine and sulfate ions, individually, at an initial P concentration of 10 mg/L. The effects of these coexisting anions on P removal are shown in Fig. 5. The P removal efficiency decreased slightly from 89 to 75.5 and 72.4%, respectively, in the presence of nitrate and sulfate anions, while it decreased significantly from 89 to 67.3 and 22.8%, respectively, in the presence of chloride and fluorine anions. It is expected that some anions would enhance the coulombic repulsion forces and some would compete with phosphate for the active sites. Generally, multivalent anions are adsorbed more readily than monovalent anions [42]. Phosphorus removal in the presence of anions increased in the order F < Cl < SO42 < NO3, which correlates with the Z/r (charge/radius) values of F (1/ 1.15) > SO42 (2/2.40) > Cl (1/1.81) > NO3 (1/2.81). In addition, Cl had a stronger effect on the P adsorption than SO42, which may be explained on the basis of ion exchange forces. The process of P adsorption on the adsorbent occurs between phosphorus ions and the La hydroxide ions located on the adsorbent surface, and Cl

Fig. 5. Effect of interfering anions for P adsorption by La(III)-loaded granular ceramic adsorbent (initial P concentration = 10 mg/L, salt concentration = 100 mg/ L, adsorbent dosage = 10 g/L, temperature = 20  1 8C).

N. Chen et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789

787

Table 3 Thermodynamic parameters for P adsorption on La(III)-loaded granular ceramic adsorbent at different temperatures. Temperature (8C)

DG (kJ/mol)

DS (J/mol K)

DH (kJ/mol)

20 30 40

2.2055 2.9266 3.6478

72.1181

18.9252

may have a better affinity than SO42 for displacing the PO43, HPO42 and H2PO4 adsorbed on the surface sites of adsorbent. 3.7. Thermodynamic study Fig. 6. Pore size distribution of La(III)-loaded granular ceramic adsorbent.

To evaluate the thermodynamic feasibility and to confirm the nature of the adsorption process, three fundamental thermodynamic parameters, changes in standard free energy (DG), standard enthalpy (DH) and standard entropy (DS), were calculated from the following equations: Kc ¼

C Ae Ce

(7)

DG ¼ RT ln K c

(8)

DG ¼ DH  T DS

(9)

where CAe and Ce represent the concentrations (mg/L) of phosphorus on the adsorbent and in solution, respectively; R is the universal gas constant (8.314 J/mol K), and T is the temperature in Kelvin (K). Values of the energy parameters DG, DH and DS are given in Table 3. The positive value of DS (72.12 J/mol K) indicates an increase in randomness during the ongoing process and hence a good affinity of P with the adsorbent. The positive value of DH (18.93 kJ/mol) for P adsorption suggests the adsorption interaction process was endothermic in nature. Moreover, it can be concluded that P adsorption occurred via chemisorption, which is consistent with the results of the kinetic study. Negative values of DG reveal that the adsorption of P on the La(III)-loaded granular ceramic adsorbent was favorable and spontaneous under standard conditions. The decrease in DG with the increase in temperature shows an increase in feasibility of adsorption at higher temperatures [43]. Similar results have been reported for P adsorption on hydrous niobium oxide [32] and ZnCl2 activated coir pith carbon [28]. 3.8. Characterization of adsorbent and possible adsorption mechanism The EMPA results of adsorbent component analysis are shown in Table 4. The weight percent of the La element in pristine adsorbent was 7.94%, indicating the presence of La in the form of lanthanum oxide compounds on the surface of the prepared granular ceramic. In aqueous solution, most of lanthanum oxide was converted to lanthanum hydroxide and complex compounds. P and La elements appeared on the surface of the adsorbed adsorbent at weight percentages of 0.81 and 5.72%, respectively, which indicates that P was bound with La-hydroxide/complex

minerals (Table 4). The content of Fe, Mn, Ca and Al ions showed no significant changes between pristine and adsorbed adsorbent in Table 4, which suggests that the amount of P on the surface of adsorbent possibly depended only on the La content of the particle surface. The BET specific surface area and total pore volume of the La(III)-loaded granular ceramic were 26.83 m2/g and 0.056 cm3/g, respectively. A pore size distribution plot, using the desorption branch, demonstrated that the majority of the pores had a diameter of approximately 15.2 nm (Fig. 6). The BET specific surface area and pore structure of samples may assist in maximizing the exposure of the adsorption sites. The Vickershardness (HV) of La(III)-loaded ceramic particles was 98.7 kg/mm2, which indicates the stability and firmness of the adsorbents. The transformation of minerals during the adsorption was also indicated by the SEM images. Fig. 7(a) and (b) compares the surface features of the adsorbents before and after P adsorption at a magnification of 1000. The lanthanum oxides on the granular ceramic surfaces initially formed a rough surface structure (Fig. 7(a)). However, the surface changed to smooth, polyhedron, and stretched cubic structures (Fig. 7(b)) with extensive reaction process exposure (for 30 h) by P adsorption. La mapping in Fig. 7(c) also reveals that La seems to be very well dispersed throughout the entire surface of the adsorbent. Comparison with P mapping (Fig. 7(d)) suggests that La is closely associated with P, which implies that the amount of P adsorbed on the surface of the adsorbent depended on the La content of the particle surface. Based on the above, the mechanism of P adsorption by La(III)loaded granular ceramics may involve electrostatic attraction and a surface precipitation process between the P species in solution and the La hydroxides on the solid adsorbents. The possible transformation was assumed to be: LaOH2þ =LaðOHÞ2 þ þ H2 PO4  ! LaðH2 PO4 Þ2 þ þ OH

(10)

LaOH2þ =LaðOHÞ2 þ þ HPO4 2 ! LaðHPO4 Þþ þ OH

(11)

Similar changes in mineralogical and structural characteristics were reported by [22]. A chemical precipitation process was the major factor controlling the P adsorption characteristics, which is consistent with the kinetic and thermodynamic study discussed above.

Table 4 Component analysis of loess, montmorillonite, pristine La(III)-loaded granular ceramics and adsorbed La(III)-loaded granular ceramics. Composition (wt%)

O

Si

Al

La

Ca

Mg

Fe

P

Loess Montmorillonite La(III)-GC (pristine) La(III)-GC (adsorbed)

58.79 55.96 45.78 48.58

27.44 31.17 32.89 33.22

5.15 9.58 8.39 7.05

– – 7.94 5.72

5.86 0.76 2.46 2.36

1.55 2.53 1.56 1.22

1.21 – 0.98 1.04

– – – 0.81

The effect on LOI (600 8C) has been neglected.

788

N. Chen et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789

Fig. 7. SEM images of (a) pristine La(III)-loaded granular ceramic adsorbent; (b) adsorbed La(III)-loaded granular ceramic adsorbent. Mapping of elements by EPMA: (c) La in pristine adsorbent; (d) P in adsorbed adsorbent.

3.9. Desorption and residual ion test High stability and reusability are important features for an advanced adsorbent. Such an adsorbent not only possesses efficient adsorption stability but also shows good reusability, which will significantly reduce the cost for the adsorbent per adsorption cycle. Desorption studies were carried out using 10 g/L P-adsorbed La(III)-loaded granular ceramic adsorbent at various pH values by adding 0.1 M HCl or NaOH. It was found that the amount of the desorbed P increased with decreasing pH. When the pH was 2.0, 84% of the phosphorus could be released into the solution. However, a maximum of only 39% phosphorus could be desorbed when the pH exceeded 6.0. The results of these desorption studies may indicate that P adsorption on the La(III)-loaded ceramic adsorbent is not completely reversible. Electrostatic attraction, ion exchange and chemical precipitation processes are operative mechanisms for the P adsorption on the La(III)-loaded ceramic adsorbent. The phosphorus adsorbed by electrostatic attraction and ion exchange could be desorbed, while the phosphorus adsorbed by chemical precipitation could not be desorbed. The concentrations of Al, Fe, Cu and Mn ions in treated aqueous solution were 13.77, 0.48, 1.62 and 0.14 mg/L, respectively. These residual ion concentrations did not exceed the World Health

Organization limits [44]. Moreover, Zn and La ions were not detected in treated aqueous solutions. In general, treating phosphate aqueous solutions with the La(III)-loaded ceramic adsorbent had little adverse effect on water quality. 4. Conclusions La(III)-loaded granular ceramic is a cost-effective adsorbent for phosphorus removal from aqueous solutions. The phosphorus adsorption capacity increased with increasing contact time. The adsorption rate was rapid initially, followed by a slower adsorption stage, which fitted well to a pseudo-second-order kinetic model. The adsorption process followed both the Langmuir and Freundlich isotherm models but the Langmuir isotherm was superior. The calculated thermodynamic parameters showed that phosphorus removal process was spontaneous, favorable and endothermic, indicating that the adsorption capacity would increase with temperature. Phosphorus removal was independent of the initial pH between 2.0 and 12.0. The presence of NO3 and SO42 had only a slight effect on phosphorus removal, while F and Cl had significant effects. Chemical precipitation may be the major factor controlling phosphorus adsorption behavior. Lanthanum or other metal ions had little adverse effects on the treated water quality. The results presented here support the potential of the

N. Chen et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 783–789

new La(III)-loaded granular ceramic as a material for the treatment of phosphorus in aqueous solutions. Acknowledgments The authors thank the Fundamental Research Funds for the Central Universities, the Development of Catch-Up Type Ceramics Material with Higher Pollutant Removal Efficiency and Its Application to Water Renovation of JST (Japan) for the financial support of this work. References [1] Zhao D, Sengupta AK. Ultimate removal of phosphate from wastewater using a new class of polymetric ion exchangers. Water Res 1998;32:1613–25. [2] De-Bashana LE, Bashan Y. Recent advances in removing phosphorus from wastewater and its future use as fertilizer. Water Res 2004;38:4222–46. [3] Mulkerrins D, Dobson ADW, Colleran E. Parameters affecting biological phosphate removal from wastewaters. Environ Int 2004;30:249–59. [4] Gieseke A, Arnz P, Amann R, Schramm A. Simultaneous P and N removal in a sequencing batch biofilm reactor: insights from reactor and microscale investigations. Water Res 2002;36(2):501–9. [5] Morse GK, Brett SW, Guy JA, Lester JN. Review: phosphorus removal and recovery technologies. Sci Total Environ 1998;212:69–81. [6] Feng CP, Sugiura N, Shimada S, Maekawa T. Development of a high performance electrochemical wastewater treatment system. J Hazard Mater 2003;103(1–2):65–78. [7] Eggers E, Dirkzwager AH, Honing HVD. Full-scale experiences with phosphate crytallisation in a crystalactor. Water Sci Technol 1991;24:333–4. [8] Kawasaki N, Ogata F, Tominaga H. Selective adsorption behavior of phosphate onto aluminum hydroxide gel. J Hazard Mater 2010;181:574–9. [9] Ania CO, Cabal B, Parra JB, Arenillas A, Arias B, Pis JJ. Naphthalene adsorption on activated carbons using solvents of different polarity. Adsorption 2008;14:343– 55. [10] Tanada S, Kabayama M, Kawasaki N, Sakiyama T, Nakamura T, Araki M, et al. Removal of phosphate by aluminum oxide hydroxide. J Colloid Interface Sci 2003;257:135–40. [11] Zeng L, Lia X, Liu J. Adsorptive removal of phosphate from aqueous solutions using iron oxide tailings. Water Res 2004;38:1318–26. [12] Boujelben N, Bouzid J, Elouear Z, Feki M, Jamoussi F, Montiel A. Phosphorus removal from aqueous solution using iron coated natural and engineered sorbents. J Hazard Mater 2007;308:47–55. [13] Wu D, Zhang B, Li C, Zhang Z, Kong H. Simultaneous removal of ammonium and phosphate by zeolite synthesized from fly ash as influenced by salt treatment. J Colloid Interface Sci 2006;304:300–6. [14] Zhang GS, Liu HJ, Liu RP, Qu JH. Removal of phosphate from water by a Fe–Mn binary oxide adsorbent. J Colloid Interface Sci 2009;335:168–74. [15] Ugurlu A, Salman B. Phosphorus removal by fly ash. Environ Int 1998;24(8):911– 8. [16] Liu RX, Guo JL, Tang HX. Adsorption of fluoride, phosphate, and arsenate ions on a new type of ion exchange fiber. J Colloid Interface Sci 2002;248:268–74. [17] Chitrakar R, Tezuka S, Sonoda A, Sakane K, Ooi K, Hirotsu T. Selective adsorption of phosphate from seawater and wastewater by amorphous zirconium hydroxide. J Colloid Interface Sci 2006;297:426–33. [18] Xue YJ, Hou HB, Zhu SJ. Characteristics and mechanisms of phosphate adsorption onto basic oxygen furnace slag. J Hazard Mater 2009;162:973–80. [19] Tian SL, Jiang PX, Ning P, Su YH. Enhanced adsorption removal of phosphate from water by mixed lanthanum/aluminum pillared montmorillonite. Chem Eng J 2009;151:141–8.

789

[20] Melnyk PB, Norman JD, Wasserlauf M. Lanthanum precipitation: alternative method for removing phosphates from waste water. In: Haschke JM, Eick HA, editors. Proc. 11th Rare Earth Res. Conf.. Springfield: NTIS; 1974. p. 4–13. [21] Recht HL, Ghassemi M, Kleber EV. Precipitation of phosphates from water and waste water using lanthanum salts. In: Proceedings of the 5th international water pollution research. Pergamon; 1970. p. 1–17. [22] Shin EW, Karthikeyan KG, Tshabalala MA. Orthophosphate sorption onto lanthanum-treated lignocellulosic sorbents. Environ Sci Technol 2005;39:6273–9. [23] Encai O, Zhou JJ, Mao SC, Wang JQ, Xia F, Min L. Highly efficient removal of phosphate by lanthanum-doped mesoporous SiO2. Colloids Surf A Physicochem Eng Aspects 2007;308:47–53. [24] Li H, Ru JY, Yin W, Liu XH, Wang JQ, Zhang WD. Removal of phosphate from polluted water by lanthanum doped vesuvianite. J Hazard Mater 2009; 168:326–30. [25] Haghseresht F, Wang SB, Do DD. A novel lanthanum-modified bentonite, Phoslock, for phosphate removal from wastewaters. Appl Clay Sci 2009;46:369–75. [26] Chen RZ, Xue Q, Zhang ZY, Sugiura N, Yang YN, Li M, et al. Development of longlife-cycle tablet ceramic adsorbent for geosmin removal from water solution. Appl Surf Sci 2011;257:2091–6. [27] 4th ed., Water and wastewater monitoring analysis method, vol. 4, 4th ed. China Environmental Science Press; 2002. p. 246–8. [28] Namasivayam C, Sangeetha D. Equilibrium and kinetic studies of adsorption of phosphate onto ZnCl2 activated coir pith carbon. J Colloid Interface Sci 2004;280:359–65. [29] Karaca S, Gu¨rses A, Ejder M, Ac¸ıkyıldız M. Kinetic modeling of liquid-phase adsorption of phosphate on dolomite. J Colloid Interface Sci 2004;277:257–63. [30] Lagergren S, Svenska K. About the theory of so called adsorption of soluble substances. K Sven Vetenskapsad Handl 1898;24(4):1–39. [31] Ho YS, Mckay G. Pseudo-second order model for sorption process. Process Biochem 1999;34:451–65. [32] Rodrigues LA, Pinto da Silva MLC. Thermodynamic and kinetic investigations of phosphate adsorption onto hydrous niobium oxide prepared by homogeneous solution method. Desalination 2010;263:29–35. [33] Zhang JD, Shen ZM, Shan WP, Chen ZY, Mei ZJ, Lei YM, et al. Adsorption behavior of phosphate on Lanthanum(III) doped mesoporous silicates material. J Environ Sci 2010;22(4):507–11. [34] Michelson LD, Gideon PG, Pace EG, Kutal LH. Office of Water Research and Technology, U.S. Department Industry, Bull No. 74; 1975. [35] Zhang JD, Shen ZM, Shan WP, Wang WH. Adsorption behavior of phosphate on lanthanum(III)-coordinated diamino-functionalized 3D hybrid mesoporous silicates material. J Hazard Mater 2011;186:76–83. [36] Langmuir I. The constitution and fundamental properties of solids and liquids. Part 1. Solids. J Am Chem Soc 1916;38:2221–95. [37] Freundlich HMF. Uber die adsorption in losungen. Z Phys Chem 1906;57A:385– 470. [38] Akhurst DJ, Jones GB, Clark M, McConchie D. Phosphate removal from aqueous solutions using neutralized bauxite refinery residues (BauxsolTM). Environ Chem 2006;3:65–74. [39] Chitrakar R, Tezuka S, Sonoda A, Sakane K, Ooi K, Hirotsu T. Adsorption of phosphate from seawater on calcined MgMn-layered double hydroxides. J Colloid Interface Sci 2005;290:45–51. [40] Liu H, Sun X, Yin C, Hu C. Removal of phosphate by mesoporous ZrO2. J Hazard Mater 2008;151:616–22. [41] Guo HM, Li Y, Zhao K, Ren Y, Wei C. Removal of arsenite from water by synthetic siderite: behaviors and mechanisms. J Hazard Mater 2011;186:1847–54. [42] Onyango MS, Kojima Y, Aoyi O, Bernardo EC, Matsuda H. Adsorption equilibrium modeling and solution chemistry dependence of fluoride removal from water by trivalent-cation-exchanged zeolite F-9. J Colloid Interface Sci 2004;279(2):341–50. [43] Stach H, Mugele J, Ja¨nchen J, Weiler E. Influence of cycle temperatures on the thermochemical heat storage densities in the systems water/microporous and water/mesoporous adsorbents. Adsorption 2005;11:393–404. [44] World Health Organization. Guidelines for drinking-water quality, 3rd ed. Geneva; 2004. p. 1.