L

Ja

b

h

•

•

•

•

a

ARRAA

KALFW

1

ag1Cso

h0

Journal of Hazardous Materials 278 (2014) 343–349

Contents lists available at ScienceDirect

Journal of Hazardous Materials

j o ur nal ho me pa ge: www.elsev ier .com/ locate / jhazmat

a3+-modified activated alumina for fluoride removal from water

iemin Chenga,∗, Xiaoguang Mengb, Chuanyong Jingb, Jumin Haob

College of Population Resources and Environment, Shandong Normal University, Jinan 250014 ChinaCenter for Environmental Systems, Stevens Institute of Technology, Hoboken, NJ 07030, USA

i g h l i g h t s

A La3+-modified activated aluminaadsorbent was prepared for effectiveremoval F−.SEM/EDS and EXAFS analyses deter-mined the formation of La(OH)3

coating on the AA.The La-AA had much high adsorptionrate and capacity than the AA.The La-AA was promising adsorbentfor effective removal of F− fromwater.

g r a p h i c a l a b s t r a c t

r t i c l e i n f o

rticle history:eceived 2 January 2014eceived in revised form 4 June 2014ccepted 6 June 2014vailable online 13 June 2014

eywords:

a b s t r a c t

A La3+-modified activated alumina (La-AA) adsorbent was prepared for effective removal of fluoride fromwater. The surface properties of adsorbent were characterized with zeta potential analysis, SEM-EDS andEXAFS. Batch and column experiments were conducted to evaluate improvement of F− removal by theLa-AA. SEM/EDS and EXAFS analyses determined the formation of La(OH)3 coating on the AA and strongbonding interactions between La3+ and the Al atoms. The points of zero charge (pHPZC) of AA and La-AAwere at pH 8.94 and 9.57, respectively. Batch experimental results indicated that the La-AA had much

−

ctivated aluminumanthanumluorideater

higher adsorption rate and capacity than the AA. The F adsorption processes on La-AA and AA followedthe pseudo-second-order kinetics and the Langmuir isotherm. Column filtration results shows that theLa-AA and AA treated 270 and 170 bed volumes of the F−-spiked tap water, respectively, before F−

breakthrough occurred. The results demonstrated that the La-AA was a promising adsorbent for effectiveremoval of F− from water.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

Fluoride contamination of groundwater by natural as well asnthropogenic sources is a major problem worldwide. The WHOuideline value for fluoride in drinking water is 1.5 mg/L [1]. Above.5 mg/L mottling of teeth may occur to an objectionable degree.

oncentrations between 3 and 6 mg/L may cause skeletal fluoro-is. Continued consumption of water with fluoride levels in excessf 10 mg/L can result in crippling fluorosis [2]. However, fluorine

∗ Corresponding author. Tel.: +86 53186179100.E-mail address: jmcheng2002@hotmail.com (J. Cheng).

ttp://dx.doi.org/10.1016/j.jhazmat.2014.06.008304-3894/© 2014 Elsevier B.V. All rights reserved.

content in groundwater is above the WHO standards in manycountries and regions. Excessive fluoride concentrations have beenreported in ground waters of more than 20 developed and devel-oping countries including India where 19 states are facing acutefluorosis problems [3]. In some parts of India, the fluoride levelsare as high as 30 mg/L have been reported [4]. Fluoride concentra-tions in the groundwater of some villages in China were also greaterthan 8 mg/L [5].

Adsorptive filtration is one of commonly used techniques for

fluoride removal from water [6]. The effectiveness of adsorptiontechniques is greatly dependent on the physicochemical propertiesof the adsorptive materials. A large number of adsorption materi-als have been tested for the removal of fluoride from water [7–11].

3 rdous

Arflbflaarbrrcf

mamztmudllatpdnsEcr

taaic

2

2

t042

2

T1uTLw

l(taA

44 J. Cheng et al. / Journal of Haza

ctivated aluminum has been reported to be effective in fluorideemoval [12–14]. Maliyekka et al. [15] improved the efficiency ofuoride removal with magnesia-amended AA granules. The adsor-ent, which was a mixture of rare earth oxides, was found to adsorbuoride rapidly and effectively. Raichur and Jyoti [16] show thepplicability of naturally occurring rare earth oxides as selectivedsorbent for fluoride from solutions. Kedar [17] found that fluo-ide removal with the phosphorylated orange waste or AA loadedy cerium(IV) was superior to AA. Modification of the organic mate-ials with Fe3+, La3+, Al3+, Sn4+ and Ce4+ could also enhance fluorideemoval from water [17,18]. However, hardly any study has beenarried out on the modification of AA by La3+ for fluoride removalrom water.

Modification of adsorbents with multivalent cations may dra-atically change the surface properties of the adsorption materials

nd their affinity for fluoride [19]. Several surface analysis instru-ents are used to characterize the adsorbent surfaces. The

eta-potential analysis of sorbent particles can be used to inves-igate the variation of the surface charge caused by surface

odification [20]. The pHPZC of several activated aluminas prod-cts have been reported [21]. Scanning electron microscopy-energyispersive x-ray spectrometry (SEM-EDS) is a non-destructive ana-

ytical method for elemental analysis, with a potential detectionimit of 0.1–0.5 wt. % for most elements [22]. Extended x-raybsorption fine structure (EXAFS) have been developed as a quan-itative, short-range structural probe in the 1970s following theioneering work of Sayers, Stern, and Lytle [23]. EXAFS can provideirect structure information, including bond length, coordinationumber and chemical identity of near neighbors. Researchers haveuccessfully studied lanthanum oxides structure using La LIII-edgeXAFS analysis [24,25]. However, limited EXAFS studies have beenarried for La coated AA system and its application in fluorideemoval.

In the present study, a novel La-AA was prepared for effec-ive removal fluoride from water. The surface properties of La-AAnd AA were systematically characterized with the zeta-potentialnalysis, SEM-EDS, and EXAFS. A series of adsorption kinetic andsotherm tests were conducted to investigate the adsorption pro-esses and capacity of the La-AA for F−.

. Materials and methods

.1. Preparation of La-AA

The AA used in this study was supplied by the Alcoa Indus-rial Chemicals Port Allen Works. The particle size of the AA was.38 mm. The AA was modified by mixing 2.0 g the adsorbent with0 mL lanthanum nitrate solution (0.075 M of La(NO3)3·6H2O) for

days. The solid was washed with deionized water and air tried.

.2. Surface characterization of the adsorbents

The AA and La-AA samples were dispersed into 10 mM NaCl.he pH of the suspension was adjusted to a range between 2 and2 using HCl and NaOH solutions for analysis the zeta potentialsing a Nano Zeta Sizer (Malvern Instrument, UK) at 25 ± 0.2 ◦C.he surface morphology and element composition of the AA anda-AA samples were examined using LEO 982 field-emission SEMith an Oxford energy dispersive x-ray analyzer.

The La LIII-edge spectra the La-AA and La(OH)3 samples were col-ected at beamline X23B of the National Synchrotron Light Source

NSLS) in Brookhaven National Laboratory at Upton, NY. The spec-ra were taken under standard NSLS operation conditions (2.8 GeVnd 150–280 mA) with a double-crystal Si(1 1 1) monochromator.

collimating mirror in front of the monochromator with a cutoff

Materials 278 (2014) 343–349

energy of approximately 10.5 keV provides harmonic rejection, andtherefore detuning was not required. The spectra were collected influorescence mode using a Stern-Heald fluorescence ion chamberdetector positioned at a 90◦ angle to the incident beam at room tem-perature. Three scans were collected from each sample, inspectedfor overall quality and averaged to improve the signal/noise ratio.

EXAFS data analysis was performed using the ATHENA andAETEMIS program in the IFEFFIT computer package [26,27]. Theanalysis procedure was similar to our previous studies [28,29]. Theraw data measured in intensities were converted to �(E), and aver-aged spectra were used in the analysis. The EXAFS signal �(k) wasextracted from the measured data using the AUTOBK algorithm [30]where k is the photoelectron wave number. The primary quantityfor EXAFS was the �(k), the oscillations as a function of photoelec-tron wave number. �(k) was weighted by k2 to account for thedampening of oscillations with increasing k. The different frequen-cies in the oscillations in �(k) correspond to different near neighborcoordination shells which can be described and modeled accordingto the EXAFS equation

x(k) =∑

j

Njfj(k)e−2k2�2

j

kR2j

sin[2kRj + ıj(k)

](1)

where f(k) and �(k) represent the photoelectron backscatteringamplitude and phase shift, respectively, N is the number of neigh-boring atoms, R is the distance to the neighboring atom, and the�2 is the Debye-Waller factor representing the disorder in theneighbor distance. The k2 weighted EXAFS in k-space (Å−1) wasFourier transformed (FT) to produce the radial structure func-tion (RSF) in R-space (Å). The experimental spectra were fittedwith single-scattering theoretical phase-shift and amplitude func-tions calculated with the ab initio computer code FEFF6 [31] usingatomic clusters generated from the crystal structure of �-La2O3and SrLa(AlO4). The many-body amplitude reduction factor (S0

2)was established as 0.78 by isolating and fitting the first-shell La-O of La(OH)3 spectrum. The spectrum was fit by first isolatingand fitting the first-shell La-O to estimate �E0, the difference inthreshold energy between theory and experiment. Then, �E0 wasfixed to the best fit value from first-shell fitting and kept thesame for all interatomic shells in a given spectrum. In addition,the �E0 value was allowed to float by no more than ±10 eV. Theparameters such as interatomic distance (R), coordination num-ber (CN), and Debye-Waller factor (�2) were first established withreasonable guesses and were fitted in R-space. The error in theoverall fits was determined using R-factor, the goodness-of-fitparameter: R − factor =

∑(xdata − xfit)2/

∑(xdata)2. Good fits occur

for R-factor < 0.05.

2.3. Batch adsorption experiments

F− adsorption kinetics experiments were conducted by mix-ing suspensions containing 1.25 g/L of AA or La-AA, 10 mg/L F−,and 0.04 mol/L NaCl as electrolyte. Aliquots of suspension sampleswere taken at desired mixing time and immediately centrifuged at10000 rpm to separate the solution from the solid. The residual F−

concentration in solution samples was measured using the using afluoride selective electrode (Thermo Orion 9609 BN) according tothe standard method [32].

The adsorption isotherms were obtained by mixing suspensionscontaining 2 g/L of AA or La-AA, 0 to 20 mg/L of F− for 2 h. The sus-pension also contained 1:1 volume of total ionic total ionic strength

adjustment buffer (TISAB) solution to control the suspension pH at7.0 ± 0.2 and ionic strength at 0.04 M. A liter of the TISAB solutioncontained 57 ml glacial acetic acid, 58 g NaCl, 4.0 g 1,2-cyclohexlyenediuminetracctic acid (CDTA), and 125 ml 6 mol/L NaOH. After

J. Cheng et al. / Journal of Hazardous

FN

mt

2

5awaewc

3

3

3

Taffwtutva

3

LsiaafFBtgtAHo

i

ig. 1. Zeta potential of the AA and La-AA at different pH values, t = 25 ◦C andaCl = 10 mM.

ixing, the suspensions samples were centrifuged for analysis ofhe equilibrium F− concentration.

.4. Column filtration experiment

Columns with a 10 mm ID and 100 mm length were packed with mL of adsorbents. A 10 mg/L of F− solution was prepared usingged tap water and the pH was adjusted to 7.0 ± 0.2. Filtration testsere carried out by pumping the solution through the columns

t a constant flow rate of 1.0 mL/min, which corresponded to anmpty bed contact time (EBCT) of 5 min. Effluent water samplesere collected at different time intervals for the analyses of F−

oncentrations.

. Results and discussion

.1. Surface characteristics of AA or La-AA

.1.1. Zeta potentialThe zeta potentials of both the AA and La-AA are shown in Fig. 1.

he pHPZC of AA and La-AA were 8.94 and 9.57, respectively. Choind Chen [21] reported that the pHPZC of AA adsorbents rangedrom 6.2 to 8.9. Ku and Chiou [12] reported a pHPZC value of 8.0or a commercial AA. The pHPZC of the AA employed in this workas similar to the values reported in the literature. The pHPZC of

he La-AA shifted to higher pH by 1.37 unit. Zeta potential is widelysed for quantification of the magnitude of the electrical charge inhe double layer [20]. The La-AA particles had higher zeta potentialalues than the AA particles at the same pH, and should have higherdsorption affinity for F− anions than AA.

.1.2. Surface composition of the adsorbentsFig. 2 shows the SEM images of AA, lanthanum hydroxide (La),

a-AA, and La-AA with fluoride adsorbed on it (La-AA-F). The EDSpectra for the materials were also included in the Fig. 3. The chem-cal compositions of the material surfaces determined using EDSnalysis were summarized in Table S1. As expected, only Al and O,nd La and O were detected on AA (Fig. 3a) and La (Fig. 3b) sur-aces, respectively. The La-AA sample exhibited a small La peak inig. 3c. La content on the solid surface was about 3% (Table S1).ased on the amount of La used in the preparation of the La-AA andhe amount of La in the suspension which was separated from theranular La-AA, the bulk content of La in the sample was calculatedo be X%. The results indicated the La was coated on the AA surface.

small La peak was also detected on the La-AA-F sample surface.

owever, no F was detectable on the sample because the contentf F was very low.

The Na and Cl peaks were observed in EDS spectra (Fig. 3d) andts composition was 0.2% and 0.3%, respectively. The trace amount

Materials 278 (2014) 343–349 345

of Na and Cl was caused the presence of 0.04 M NaCl used in theadsorption suspension.

3.1.3. Specific interactions of La with AAEXAFS spectroscopy was employed to determine the La local

coordination environment in the La(OH)3 and La-AA samples. Thek2 weighted La LIII-edge EXAFS spectra, the corresponding radialstructure functions (RSF) as magnitude and real part of the Fouriertransformation (FT) vs. radial distance are shown in Fig. 4. Theoptimal parameters listed in Table S2 were obtained by fitting thetheoretical curves to the experimental spectra. The FT of EXAFSspectra can isolate the contributions of different coordinationshells, in which the peak positions correspond to the interatomicdistances. However, these peak positions are uncorrected for phaseshift so that they are shifted from the true distance by 0.3–0.5 A inthe figure.

Based on the fit of the theoretical to the experimental spec-tra, the first and strongest peak in the FT curve was contributedby 7.6 oxygen atoms at an average distance of 2.58 A for La(OH)3sample. One oxygen and 3.3 La atoms were at distance of 3.56 and4.22 A, respectively. The distances and coordination numbers (CN)of La-O and La-La are in good agreement with previously publisheddata [24,25]. When La formed a coating on AA, the first La-O shellremained similar as in the La(OH)3. Aluminum atoms were detectedat 3.19 A with CN of 3.7. This structure change from La(OH)3 demon-strated that La was specifically adsorbed on the surface of AA. TheLa LII-edge at 410 eV beyond the LIII-edge increased the complexityof the data analysis because truncation led to a limited data range ink-space. Therefore, the FT peaks higher than 4.5 A are unresolved.

3.2. Batch adsorption of F− on La-AA and AA

3.2.1. Adsorption kineticsFig. 5 shows the adsorption kinetic data of F− on the AA and

La-AA at pH 7.0 and the best-fit model curves. The amount of F−

adsorbed by AA increased slowly over a period of 6 h. Tang et al.[33] reported that fluoride adsorption on AA reached equilibriumin 10 h. On the other hand, F− adsorption by La-AA occurred rapidlyand reached equilibrium within about 1 h.

The pseudo-first-order kinetic equation (2) [34] and pseudo-second-order kinetic Eq. (3) [35] were applied to fit the adsorptionkinetic data in Fig. 5.

ln(q1e − qt) = ln q1e − k1 × t (2)

t

qt= 1

k2q22e

+ t

q2e(3)

where q1e and q2e are the adsorption capacity at equilibrium (mg/g),q1t and q2t are the amount of F− adsorbed at time t, and k1 and k2are the rate constants.

The best-fit parameters for the pseudo-first-order model andpseudo-second-order model are listed in Table S3. The R2 valuesof the pseudo-second-order model were higher than those of thepseudo-first-order model, which indicated that the F− adsorptionon the La-AA and AA followed the pseudo-second-order kinetics.The much higher rate constant k2 for La-AA (i.e. 0.0422) than forAA (i.e. 0.0023) also indicated the much faster adsorption kineticsof F− on La-AA than on AA.

In most cases, the pseudo-first-order equation does not fit wellfor the whole range of contact time and is generally applicable overthe initial 20–30 min of the sorption process [36]. F− is adsorbed onmetal oxides and hydroxides through bonding interactions with

the surface metal sites [37]. The rapid adsorption of F− on La-AAwithin the first 0.5 h (Fig. 5) could be attributed mainly to the spe-cific adsorption of the F− by the La(OH)3 coating on the AA surfacebecause it was readily accessible to F−. The slower adsorption after

346 J. Cheng et al. / Journal of Hazardous Materials 278 (2014) 343–349

Fig. 2. SEM images for activated alumina AA (a), Lanthanum oxide (b), La coated AA (c) and F-laden La coated AA (d).

Fig. 3. EDS spectra for activated alumina AA (a), Lanthanum oxide (b), La coated AA (c) and F-laden La coated AA (d). The element and its composition in percentage (inparenthesis) are labeled in EDS spectra. Break was used in plotting y-axis to emphasis the fine peaks with less intensity.

J. Cheng et al. / Journal of Hazardous Materials 278 (2014) 343–349 347

6543210

R(Å)

FT M

agni

tude

(a)

(b)

6543210

R(Å)

(a)

(b)

C B

10987654321k (Å-1)

k2 χ(k

)(a)

(b)

A

Fig. 4. Normalized k2-weighted observed (dotted line) and model calculated (solid line) La LIII-edge EXAFS spectra (A), the corresponding Fourier transformed magnitude( e peak positions are uncorrected for phase shift.

tA

0(wssLabpLi

3

AF

0

1

2

3

4

5

6

7

8

6.565.554.543.532.521.510.50Time (h)

F ad

sorb

ed (m

g/g)

AALa-AA

Fig. 5. Adsorption kinetics of fluoride on AA or La-AA at pH = 7.0, 1.25 g/L adsorbents,

B) and real parts of Fourier transform (C) for (a) La(OH)3 and (b) La-AA samples. Th

he initial 0.5 h was ascribed to diffusion of F− into the pores of theA to be absorbed by the internal surface sites.

The adsorption rate k2 of F− on AA increased from 0.002 to.019 mg/g min with decreasing particle size from 0.38 to 0.15 mmFig. S1). The higher rate of F− adsorption by smaller AA particlesas due to greater specific surface area, especially and external

urface area, and greater accessibility to pores and the internalurface area [38,39]. However, the adsorption rate of F− on thea-AA was not affected much by the particle size. All three La-AAbsorbents had higher adsorption rate than the three AA adsor-ents. The results suggested that the La(OH)3 coating on the AAlayed an important role in the adsorption and the content of thea(OH)3 in the La-AA adsorbents of different particle size was sim-lar.

.2.2. Adsorption isothermsFig. 6 illustrated the adsorption isotherms of F− on the La-AA and

A at pH 7.0. La treatment AA significantly increased the amount of− adsorbed. The best-fit parameters are listed in Table S4. The R2

10 mg/L F− , and T = 25 ◦C.

348 J. Cheng et al. / Journal of Hazardous

Fig. 6. Adsorption isotherms of fluoride on La-AA and AA at pH = 7.0 and T = 25 ◦C.

FAfl

vtAch

pLtFt[

3

vctearLt

4

Lbzs

[

[

[

[

[

[

[

[

[

[

[

ig. 7. Column filtration results for F− removal from spiked tap water by AA and La-A. Adsorbent volume = 5 mL, influent F− concentration = 10 mg/L and pH = 7.0 ± 0.2,ow rate = 1 mL/min.

alues suggested that the Langmuir isotherm described the adsorp-ion data better than the Freundlich isotherm. The qmax value forA was 2.74 mg/g, which was similar to the maximum adsorptionapacity of 2.69 mg/g by Roberto et al. [13]. The La-AA had muchigher qmax (6.70 mg/g) than the AA.

The removal of fluoride was strongly dependent on the solutionH. The maximum adsorption pH was 5.84 and 4.99 for the AA anda-AA, respectively (Fig. S2). The La-AA had better F− removal thanhe AA in a pH range between 3 and 9. The optimum pH range for− removal by the adsorbents was from 4 to 8. The effect of pH onhe adsorption of F− was similar as that reported in the literature33].

.3. F− removal by La-AA and AA columns

Fig. 7 shows the effluent F− concentration as a function of bedolumes (BV = volume of filtered water/volume of adsorbent in theolumn) [40]. The La-AA column filtered more F−-spiked tap waterhan the AA columns before F− breakthrough occurred. When theffluent F− concentration reached the WHO MCL of 1.5 mg/L, 170nd 270 BV of the water was treated by the AA and La-AA columns,espectively. The column filtration results demonstrated that thea modification significantly improved the F− removal capacity ofhe adsorbent.

. Conclusion

The SEM/EDS and EXAFS analyses determined the formation of

a hydroxide coating on the AA and strong bonding interactionsetween La3+ and the Al atoms. The La-AA particles had highereta potential and adsorption capacity than the AA particles at theame pH. The adsorption behavior of La-AA and AA was described

[

[

Materials 278 (2014) 343–349

well with the pseudo-second-order kinetics and the adsorptionisotherms were described with the Langmuir equation. The max-imum adsorption capacities of AA and La-AA determined by theLangmuir equation were 2.74 and 6.70 mg/g at pH = 7.0, respec-tively. The optimum pH range for F− removal by the adsorbentswas from 4 to 8. The results of the column tests indicated thatthe La modification greatly improved the F− removal capacity ofthe adsorbent. The La-AA was a more effective adsorbent than thecommonly used AA for F− removal from water.

Acknowledgements

We acknowledge the staff on beamline X23B at the NationalSynchrotron Light Source (NSLS) for their assistance with EXAFSdata collection.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.jhazmat.2014.06.008.

References

[1] World Health Organization (WHO), Guidelines for Drinking-Water Quality. 3rdEd., WHO, Geneva, Switzerland, 2004.

[2] C.A. Yeung, A systematic review of the efficacy and safety of fluoridation, Evid.Based Dent. 9 (2008) 39–43.

[3] R.C. Meenakshi Maheshwari, Fluoride in drinking water and its removal, J.Hazard. Mater. 137 (2006) 456–463.

[4] B.K. Handa, Geochemistry and genesis of fluoride-containing ground water inIndian Groundwater, Ground Water 13 (1975) 275–281.

[5] Anonymous Drinking water atlas of China. Beijing, China Cartographic Publish-ing House. (1990)91–92.

[6] X. Fan, D.J. Parker, M.D. Smith, Adsorption kinetics of fluoride on low costmaterials, Water Res. 37 (2003) 4929–4937.

[7] W. Nigussie, F. Zewge, B.S. Chandravanshi, Removal of excess fluoride fromwater using waste residue from alum manufacturing process, J. Hazard. Mater.147 (2007) 954–963.

[8] S. Venkata Mohan, S.V. Ramanaiah, B. Rajkumar, P.N. Sarma, Removal of fluoridefrom aqueous phase by biosorption onto algal biosorbent Spirogyra sp.-IO2:sorption mechanism elucidation, J. Hazard. Mater. 22 (141) (2007) 465–474.

[9] M. Karthikeyan, K.K. Satheeshkumar, K.P. Elango, Removal of fluoride ionsfrom aqueous solution by conducting polypyrrole, J. Hazard. Mater. 167 (2009)300–305.

10] B. Yupin, G. Yunan, G. Haijun, F. Jinxian, Y. Yong, Zhang pingping chitosanadsorption process for treatment of fluoride pollution in groundwater, WaterPurif. Technol. 31 (2012) 30–33.

11] W. Gong, J. Qu, R. Liu, H. Lan, Effect of aluminum fluoride complexation onfluoride removal by coagulation, Colloid. Surface. A 395 (2012) 88–93.

12] Y. Ku, H.W. Chiou, The adsorption of fluoride ion from aqueous solution byactivated alumina, Water Air Soil Poll. 133 (2002) 349–360.

13] L.-L. Roberto, N.A. Medellin-Castillo, J.-A. Araceli, M.-Z. Jovita, L.E. Landin-Rodriguez, J.M. Martinez-Rosales2, A.-P. Antonio, Fluoride removal from watersolution by adsorption on activated alumina prepared from pseudo-boehmite,J. Environ. Eng. Manage. 18 (2008) 301–309.

14] M. Aneeza Rafique, A. Awan, A. Wasti, I. Ishtiaq Qazi, A. Muhammad, Removalof fluoride from drinking water using modified immobilized activated alumina,J. Chem. (2013) 7, http://dx.doi.org/10.1155/2013/386476, Article ID 386476.

15] S.M. Maliyekka, S. Shukla, P. Ligy, N.M. Nambi, Enhanced fluoride removal fromdrinking water by magnesia-amended activated alumina granules, Chem. Engg.J. 140 (2008) 183–192.

16] A.M. Raichur, M. Jyoti, Adsorption of fluoride onto mixed rare earth oxides, Sep.Purif. Technol. 24 (2001) 121.

17] G.N. Kedar, Effective removal of fluoride onto metal ions loaded orange waste,J. Nepal Chem. Soc. 27 (2011) 61–66.

18] Y. Huo, W. Ding, X. Huang, J. Xu, J. Zhao Jingnian1, Fluoride removal by lan-thanum alginate bead: adsorbent characterization and adsorption mechanism,Chin. J. Chem. Eng. 19 (2011) 365–370.

19] H. Guo, W. Li, Z. Chang, H. Wang, Y. Zhou, Mechanism study of fluoride adsorp-tion by hydrous metal oxides, Spectrosc. Spect. Anal. 31 (2011) 2210–2214.

20] P. Peng, A. Keller Arturo, Natural and engineered nano and colloidal transport:role of zeta potential in prediction of particle deposition, Langmuir 25 (2009)6856–6862.

21] W.W. Choi, K.Y. Chen, The removal of fluoride from waters by adsorption, J. Am.Water Works Ass. 71 (1979) 562–570.

22] S.M. Haley, A.D. Tappin, P.R. Bond, M.F. Fitzsimons, A comparison of SEM-EDS with ICP-AES for the quantitative elemental determination of estuarineparticles, Environ. Chem. Lett. 4 (2006) 235–238.

rdous

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

J. Cheng et al. / Journal of Haza

23] D.E. Sayers, E.A. Stern, F.W. Lytle, New technique for investigating non-crystalline structures – Fourier analysis of extended x-ray –absorption finestructure, Phys. Rev. Lett. 27 (1971) 1204–1207.

24] P. Malet, M.J. Capitan, M.A. Centeno, J.A. Odriozola, I. Carrizosa, Exafs data-analysis for lanthanide aesquioxides, J. Chem. Soc. Faraday T. 90 (1994)2783–2790.

25] F. Ali, A.V. Chadwick, M.E. Smith, EXAFS analysis of the structural evolution ofgel-formed La2O3, J. Mater. Chem. 7 (1997) 285–291.

26] B. Ravel, M. Newville, ATHENA, ARTEMIS HEPHAESTUS: data analysis for x-ray absorption spectroscopy using IFEFFIT, J. Synchrotron. Radiat. 12 (2005)537–541.

27] M. Newville, IFEFFIT: interactive XAFS analysis and FEFF fitting, J. Synchrotron.Radiat. 8 (2001) 322–324.

28] C. Jing, J.S. Liu, M. Patel, X. Meng, Arsenic leachability in water treatment adsor-bents, Environ. Sci. Technol. 39 (2005) 5481–5487.

29] C. Jing, X. Meng, S. Liu, S. Baidas, R. Patraju, C. Christodoulatos, G.P. Korfiatis,Surface complexation of organic arsenic on nanocrystalline titanium oxide, J.Colloid Interface Sci. 290 (2005) 14–21.

30] M. Newville, P. Livins, Y. Yacoby, J.J. Rehr, E.A. Stern, Near-edge x-ray-absorptionfine-structure of Pb - a comparison of theory and experiment, Phys. Rev. B. 47(1993) 14126–14131.

31] J.M. Deleon, J.J. Rehr, S.I. Zabinsky, R.C. Albers, Abinitio curved-wave x-ray-absorption fine-structure, Phys. Rev. B. 44 (1991) 4146–4156.

[

Materials 278 (2014) 343–349 349

32] APHA. Standard methods. 19th Edition. American Public Health Association,Washington, DC. (1995).

33] Y. Tang, X. Guan, T. Su, N. Gao, J. Wang, Fluoride adsorption onto activatedalumina: modeling the effects of pH and some competing ions, Colloid. Surface.A 337 (2009) 33–38.

34] S. Lagergren, About the theory of so-called adsorption of soluble substances,Kungliga Svenska Vetenskapsakademiens Handlingar 24 (1898) 1–39.

35] Y.S. HO, G. McKAY, A comparison of chemisorption kinetic models applied topollutant removal on various sorbents, Trans IChemE. B76 (1998) 332–340.

36] Y.S. Ho, Review of second-order models for adsorption systems, J. Hazard.Mater. 136 (2006) 681–689.

37] W. Xiaomei, Y. Zhang, D. Xiaomin, Z. Bei, M. Yang, Fluoride adsorption onan Fe–Al–Ce trimetal hydrous oxide: characterization of adsorption sites andadsorbed fluorine complex species, Chem. Engg. J. 223 (2013) 364–370.

38] S. Yean, L. Cong, C.T. Yavuz, J.T. Mayo, W.W. Yu, A.T. Kan, V.L. Colvin, M.B. Tom-son, Effect of magnetite particle size on adsorption and desorption of arseniteand arsenate, J. Mater. Res. 20 (2005) 3255–3264.

39] R. Krishna, A.V.V.S. Hema Swamy, The effect of particle size and adsorption

kinetics for the removal of Cr(VI) from aq. solutions, Eur. Chem. Bull. 1 (2012)258–262.

40] J. Hao, M. Han, C. Wang, X. Meng, Enhanced removal of arsenite from waterby a mesoporous hybrid material–thiol-functionalized silica coated activatedalumina, Micropor. Mesopor. Mater. 124 (2009) 1–7.