Fixed-bed column studies of pentachlorophenol removal by use of alginate-encapsulated pillared clay microbeads

Journal of Colloid and Interface Science 379 (2012) 101–106

Contents lists available at SciVerse ScienceDirect

Journal of Colloid and Interface Science www.elsevier.com/locate/jcis

Fixed-bed column studies of pentachlorophenol removal by use of alginate-encapsulated pillared clay microbeads Mouloud Lezehari a,b, Michel Baudu a, Omar Bouras b, Jean-Philippe Basly a,⇑ a b

Université de Limoges, EA 4330 Groupement de Recherche Eau Sol Environnement, Faculté des Sciences et Techniques, 123 Avenue Albert Thomas, 87060 Limoges Cedex, France Université Saad Dahlab, Faculté des Sciences de l’Ingénieur, Département de Chimie Industrielle, BP 270, 09000 Blida, Algeria

a r t i c l e

i n f o

Article history: Received 28 February 2012 Accepted 17 April 2012 Available online 28 April 2012 Keywords: Pillared clays/alginate microbeads Pentachlorophenol Sorption Fixed bed Breakthrough curves Thomas model

a b s t r a c t Columns were packed with two alginate/pillared clays microbeads (aluminium-pillared clay and surfactant-modified aluminium-pillared clay). Pentachlorophenol sorption performance was assessed under variable operating conditions: different bed heights, influent pentachlorophenol concentrations and flow rates. These conditions greatly influenced the breakthrough time/volume, the saturation time/volume and the uptake capacity. Higher values of experimental uptake capacities were obtained for the encapsulated surfactant-modified aluminium-pillared clay compared with the encapsulated aluminium-pillared clay, and the values were compared with those obtained with other low-cost sorbents. The experimental breakthrough curves were modelled using Bed Depth Service Time (BDST), Wolborska and Thomas models. Linear relationship was obtained for the BDST model, indicating the suitability of this model; bed capacity increased sharply with the introduction of CTAB in the inorgano-pillared clay. Wolborska model was applied only to the initial part of the curves. Thomas model was no doubt the most suitable description of the adsorption mechanisms for the entire breakthrough curves. Experimental and Thomas modelpredicted equilibrium uptake capacities were in accordance. Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction Pentachlorophenol (PCP), a long been used broad spectrum biocide, is a weak acid present in the environment as both neutral and charged (anionic) forms, and its biodegradation in water is slow and incomplete. PCP was found to be persistent in either the environment or the organisms which results in its widespread presence throughout the food chain leading to harmful effects to public health [1,2]. Among the various treatment options investigated to remove PCP from wastewaters (coagulation and flocculation, membrane filtration, biological treatments, adsorption, advanced oxidation processes), adsorption has been regarded as a promising technology [3–6]. Extensive investigations are being carried out actually to identify alternative sorbents, and pillared clays have been proposed as a new class of microporous materials due to easy availability and catalytic properties in different reactions. These solids are obtained by introducing large polyoxycations into their interlayer regions, and the separation between the layers can be kept stable and depends on the polyoxycation used. The polymeric compounds most frequently used as pillaring agents are species of ⇑ Corresponding author. Address: Université de Limoges, EA 4330 Groupement de Recherche Eau Sol Environnement, Faculté de Sciences et Techniques, 123 Avenue Albert Thomas, 87025 Limoges Cedex, France. Fax: +33 555 45 75 28. E-mail address: [email protected] (J.-P. Basly). 0021-9797/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2012.04.054

Al, Ti, Cr and Fe [7,8]. Moreover, the organic–inorganic modification of clay minerals offers the potential to be adsorbent in the removal of both organic and inorganic pollutants from wastes, due to their structural diversity [9]. In this way, numerous studies have been directed towards the use of mixed-pillared clay for the removal of metal ions, organic pollutants and dyes from waters and wastewaters [7–10]. Processes in water treatment could require granular materials of controlled size and the encapsulation of pillared clays into alginate, a polysaccharide copolymers of bd-mannuronic acid and a-l-guluronic acid extracted from brown seaweeds, could widen the field of applications. The sorption properties of composite beads including alginate have been demonstrated in earlier publications for various organic and inorganic pollutants: extractants [11,12], magnetosorbents [13–16] grape stalks [17,18], solid waste [19] activated carbon [20] and clays [21–24]. Continuous flow conditions are considered useful in large-scale industrial wastewater treatment because of their simplicity, ease of operation, handling and regeneration capacity. In this framework, the aim of the present work is to examine the feasibility of using two alginate/pillared microbeads, for example, encapsulated aluminium-pillared clay (Al-Mont-EnPILC) and encapsulated surfactant-modified aluminium-pillared clay (CTAB-Al-Mont-EnPILC) for the removal of PCP from aqueous solutions in continuous mode. The influence of different operational parameters affecting the

102

M. Lezehari et al. / Journal of Colloid and Interface Science 379 (2012) 101–106

process in a column such as bed height, flow rate and initial pollutant concentration on breakthrough curves was examined. The breakthrough curves behaviour was also modelled using BDST, Wolborska and Thomas models. 2. Material and methods 2.1. Pillared clays encapsulation Aluminium-pillared clay (Al-Mont-PILC) and surfactant-modified pillared clay (CTAB-Al-Mont-PILC) were synthesized from a west Algerian bentonite powder (Maghnia deposit – size <70 lm) supplied by ENOF (Entreprise Nationale des Substances Utiles et des Produits Non Ferreux, Algeria) according to previous reported methods [8,25]. Surface properties of the pillared clays are summarized in Table 1. The beads were prepared according to the ionic gelation method [11,16,17,21,24]. Four grams of aluminium-pillared clay (Al-Mont-PILC) or surfactant-modified pillared clay (CTA-Al-Mont-PILC) was dispersed in 100 mL of Milli-Q water. An alginate (supplied by Fluka) suspension (1%, w/w; V = 100 mL) was added, and the mixture was stirred for 2 h. Once the mixture was homogeneous, it was forced through a micropipette tip by a peristaltic pump, and the resulting droplets were collected in a stirred reservoir containing 200 mL of 0.1 M CaCl2. After 10 h, the beads (Al-Mont-EnPILC and CTAB-Al-Mont-EnPILC) were filtered, washed several times with Milli-Q water and used immediately after preparation. Physicochemical properties of the microbeads are given in Table 1.

diameter. The aqueous PCP solution (pH = 5.3 without adjustment during the experiments) was pumped through the column at different bed height, flow rates and PCP initial concentrations. Samples were collected from the outlet of the column, and the PCP concentrations were determined spectrophotometrically at 319 nm. The effect of bed depth Z on the breakthrough curves was investigated using bed heights of 0.15, 0.20 and 0.25 m. The feed solution had a PCP concentration of 50 lM in all cases and was allowed to flow through the beds with an effluent rate of 1.3 mL min1 (0.69 m h1). The effect of effluent flow rate U0 (1.3, 2.5 and 5 mL min1 – 0.69, 1.32 and 2.65 m h1) was investigated using columns with a fixed bed depth of 0.15 m and PCP concentration of 50 lM. The effect of the PCP concentration (10 lM, 25 and 50 lM) was investigated using columns with a fixed bed depth of 0.15 m and an effluent flow rate of 1.3 mL min1 (0.69 m h1). Parameters in part 2.2 were optimized by ORIGIN 7.5 software. 2.3. Modelling of the breakthrough curves Three breakthrough curve models were selected in this study. 2.3.1. Bed Depth Service Time Bed Depth Service Time (BDST) is one of the most widely used and simple model [26]. It assumes that the rate of adsorption is governed by the surface reaction between the adsorbate and the unused capacity of the adsorbent. The BDST model (Eq. (1)) shows a linear relationship between the bed height and the time and can be helpful to scale up the process for other flow rates without further experimental runs.



2.2. Dynamic sorption

t ¼ N0 =C 0 U 0 Z  1=kBDST C 0 ln Pentachlorophenol (Fluka) stock solution (C0 = 50 lM, 13.3 mg L1) was prepared by dissolving the appropriate amount of PCP in high purity de-ionized water (Milli-Q system: resistivity 18.2 MO cm and TOC < 10 lg L1) at room temperature, shaken for several days and then filtered through Sartorius cellulose nitrate membrane (0.45 lm) before use. The solutions (stock and diluted solutions) were kept at 4 °C in the dark to prevent any possible photodegradation. All column experiments were conducted at room temperature in glass column of 30 cm height and 1.2 cm

Table 1 Physicochemical properties of the pillared clays and alginate/pillared clays microbeads [8, 24, 25].

Clays f (mV) ± 5 mV pH 3 pH 6.5 pH 9.7 BET specific surface area (m2 g1) Basal spacing d(001) (Å) Pore (nm) Cation exchange capacity (meq/100 g)

Microbeads Experimental weight ratio (clay/ alginate) Microbeads diameter (mm) EWC (%) k2 (gdry beads lmol1 h1)

Al-Mont-PILC

CTAB-Al-Mont-PILC

20 6 14 230 20 <21 14

42 29 n.da 12 22 fb n.da

Al-MontEnPILC

CTAB-Al-MontEnPILC

2.8/1

1.7/1

2.6 93.5 0.0301

2.6 93.7 0.0225

f: zeta potential; d(001): basal spacing; experimental ratio: amount of clay (mg) included in alginate bead vs. dry weight of alginate (mg); EWC: experimental water content; k2: pseudo-second-order kinetic constant. a n.d: Not determined. b f: Filling pore.

C0 1 C

 ð1Þ

where t is the time (h), C0 and C are the influent and the effluent concentrations (lmol L1), Z is the bed height (m), U0 is the linear flow rate (m h1), kBDST is the adsorption rate constant that describes the mass transfer from the liquid phase to the solid phase (L lmol1 h1) and N0 is the bed capacity (lmol L1). 2.3.2. Thomas model The Thomas model is a general sorption model; it had been previously used for dynamic sorption studies with alginate [27,28] and chitosan sorbents [29,30]. This model [31] (Eq. (2))] assumes plug-flow behaviour in the bed and uses the Langmuir isotherm for equilibrium and second-order reversible reaction kinetics. It assumes a constant separation factor but is applicable to both favourable and unfavourable sorption conditions.

C 1   ¼ C 0 1 þ exp kTHO qo m  kTHO C 0 t U0

ð2Þ

where kTHO is the Thomas rate constant [mL/(h lmol)], q0 is the equilibrium uptake capacity (lmol/g), m the mass of sorbent packed in the column (g) and U0 is the volumetric flow rate (m h1); C and C0 are the concentrations (lmol L1) of PCP in the effluent and in the influent at any time (h), respectively. 2.3.3. Wolborska model Wolborska model (Eq. (3)) [32] is used for the description of adsorption dynamics using mass transfer equations for diffusion mechanisms in the range of the low concentration breakthrough

  C b C 0 t ba Z ¼ exp a  C0 N0 U0

ð3Þ

with ba kinetic coefficient of the external mass transfer (h1), and N0 (lmol L1) the exchange capacity.

M. Lezehari et al. / Journal of Colloid and Interface Science 379 (2012) 101–106

All the above-mentioned models in part 2.3 were fitted to the experimental breakthrough curves using nonlinear regression method and STATISTICA 6.0 software; their suitability was assessed on the basis of R2, sum of squares errors (SSE – Eq. (4)) and standard deviation of residues (syx – Eq. (5)).

SSE ¼

n X ðC exp;i  C model;i Þ2

ð4Þ

i¼1

syx ¼

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi SSE=n  p

ð5Þ

Cexp,i and Cmodel,i are the experimental and calculated PCP concentrations (lM) at the column outlet, n is the number of points and p the degree of freedoms. 3. Results and discussion In a previous work [24], two alginate/pillared microbeads were proven effective in the elimination of pentachlorophenol in batch systems. The adsorption isotherms were best fitted by Langmuir model, while the kinetic was well described by a pseudo-secondorder model.

103

3.1. Impact of bed height, flow rate and concentration on the dynamic sorption Breakthrough curves present an S-shaped profile characteristic of small adsorbates with simple molecular structure (Fig. 1). The overall performance of a flow-through biosorption column is related to: (a) the column breakthrough time tb (the time at which the PCP concentration in the effluent reached 5% of the influent concentration), (b) the exhaustion time te (the time at which the PCP concentration in the effluent reached 95% of the influent concentration), (c) the total quantity of PCP biosorbed in the column or uptake capacity q0 (lM/gdry bead) calculated from the area above the breakthrough curve multiplied by the flow rate. Results are presented in Table 2. Higher values of q0 were obtained for the encapsulated surfactant-modified aluminium-pillared clay compared with the encapsulated aluminium-pillared clay (an increase of 10% was observed). This result can be compared with our earlier studies involving batch reactors [24] (12% PCP sorption increase with the introduction of CTAB in the inorgano-pillared clay; qLangmuir = 580 lmol/g for Al-Mont-EnPILC and qLangmuir = 651 lmol/g for CTAB-Al-Mont-EnPILC).

Fig. 1. Influence of operational parameters (A) bed height (B) PCPconcentration (C) flow rate on the pentachlorophenol breakthrough curves. Modelled straight lines were obtained with the Thomas model. Bed heights: 0.15, 0.20 and 0.25 m; flow rate: 1.3 mL/min; [PCP] = 50 lM, bed height: 15 cm; flow rates: 1.3 mL/min; [PCP] = 10, 25 and 50 lM, bed height: 15 cm; flow rate: 1.3, 2.5 and 5.0 mL/min; [PCP] = 50 lM.

104

M. Lezehari et al. / Journal of Colloid and Interface Science 379 (2012) 101–106

Table 2 Column data and parameters with different bed height, flow rates and initial pentachlorophenol concentrations. Breakthrough Z (m) U0 (m.h-1) C0 (lM)

curves conditions 0.15 0.20 0.69 0.69 50 50

0.25 0.69 50

0.15 0.69 25

0.15 0.69 10

0.15 1.32 50

0.15 2.65 50

Al-Mont-EnPILC 6.7 Tb (h) Vb (L) 0.53 Te (h) 21.8 Ve (L) 1.70 q0 (lM/g) 112.0

10 0.78 27.7 2.16 113.3

14.8 1.15 33.1 2.58 114.6

11.2 0.87 28.5 2.22 77.2

17.7 1.38 34.1 2.66 40.9

2.8 0.42 11.5 1.72 107.0

1.8 0.54 5.5 1.65 99.8

CTAB-Al-Mont-EnPILC Tb (h) 9.3 Vb (L) 0.72 Te (h) 25.4 Ve (L) 1.98 q0 (lM/g) 143.7

15.7 1.22 31.0 2.42 148.0

23.1 1.80 38.0 2.96 151.3

16.4 1.28 34.6 2.70 107.7

18.8 1.46 46.6 3.63 62.9

4.6 0.69 13.1 1.97 142.9

2.0 0.60 6.4 1.92 137.7

Z: bed height; U0: flow; C0: PCP initial concentration; Tb: breakthrough time; Vb: breakthrough volume = Tb  flow; Te: exhaustion time; Ve: exhaustion volume = Te  flow; q0: PCP (lM) removed by gdry bead.

The bed depth (Fig. 1A) has a great effect on the column performance [33,34]. Both the breakthrough and exhaustion times increased for the inorgano- and organo–inorgano-pillared clays studied with increasing bed height from 15 to 20 and 25 cm, respectively, as more binding sites became available for sorption. Introduction of CTAB delayed the breakthrough time by 40–50%. Nevertheless, the sorption capacities q0 were not significantly modified by an increase in the bed height. This may be due to no limitation of pollutant availability in the column. As expected, a change in inlet PCP concentration (Fig. 1B) affected the sorption characteristics of the packed sorbent in the column. The earlier appearance of the breakthrough point with increasing PCP concentration is due to the arrival of a high mass of solute per unit area on the surface of the adsorbent and a fast saturation of the binding sites. The shape of the breakthrough curves were not modified by an increase in inlet concentration. Flow rate (Fig. 1C) is an important characteristic affecting the performance of a biosorbent in the continuous mode. The steepness (dC/dt) of breakthrough curves increased with increasing flow rate. An increase in uptake capacity q0 (Table 2) with increasing 2

uptaake cappacity (m mg/g)

40 1

6

20 3

10

0

this study

this study

[36]

5

[37]

[38]

3.2. Mathematical modelling of the breakthrough curves 3.2.1. BDST model Linear relationship was obtained for the BDST model (Table 3) with R2  0.99, indicating the suitability of this model. Bed capacity increased sharply with the introduction of CTAB in the inorgano-pillared clay, while in contrast, kBDST decreased. More than one rate-limiting step in the adsorption process could be involved since BDST line does not pass through the origin [41,42]. The magnitude of N0 (745 mg/L and 1270 mg/L for Al-Mont-EnPILC and CTAB-Al-Mont-EnPILC, respectively) was higher than those observed for pine bark [38] (32 mg/L) but lower as compared to almond shell [36] (6322 mg/L). 3.2.2. Wolborska model The Wolborska model was fitted, in a first time, to the full part of the breakthrough curves. However, it showed poor fitness (R2 < 0.92) for all the breakthrough curves (data not shown). There, Wolborska model was applied only to the initial part of the curves (0.05 < C/C0 < 0.30). Nevertheless, parameters of this model (Table 4) suggest important points regarding column parameters. The saturation concentration N0 increased with increasing PCP concentration; in contrast, bed height and flow rate did not affect greatly the saturation concentration. Conversely, kinetic constants b decreased with PCP concentration and increased with flow rate. The migration velocity, v, defined by Eq. (6) [43]:

m ¼ U 0 C 0 =C 0 þ N0

ð6Þ

remains constant with bed height but increases with concentration and flow rate. Increasing flow rate from 1.3 to 5.0 mL min1 increased the kinetic constant values since increased turbulence reduces the film boundary layer surrounding the sorbent particle. The maximum amount of PCP determined by this model did not show a larger deviation from the values calculated using the BDST model. 3.2.3. Thomas model In the batch equilibrium study [24], it was found that the Langmuir model and the pseudo-second-order kinetic model fit the removal of PCP by encapsulated pillared clays. Thomas model was no doubt the most suitable description of the adsorption mechanisms for the entire breakthrough curves (Table 5). Fits between experimental and calculated Ceffluent/Cinfluent vs. time using Thomas model are reported in Fig. 1. Experimental and Thomas model-predicted equilibrium uptake capacities were in accordance for both AlMont-EnPILC and CTAB-Al-Mont-EnPILC (q0 in Tables 2 and 4). The rate constant KTHO that characterized the rate of solute transfer

30

4

residence time (low flow rate), which indicates that the sorption process is controlled by intraparticle mass transfer [35], was not evidenced in this study. Uptake capacities were compared with other low-cost sorbents (Fig. 2) with good performance for the two pillared clays studied; however, the values for the encapsulated pillared clays were 25– 30% those obtained with Granular Activated Carbon [40] (150 mg g1).

[39]

1 : Al-Mont-EnPILC 2 : CTAB-Al-Mont-EnPILC

Table 3 BDST Parameters.

3: Almond Shell 4 : Coal fly ash 5 : Pine Bark

N0 (lmol L1) kBDST (L lmol1 h1) R2

6: n Fig. 2. Uptake information.)

capacities.

(See

above-mentioned

references

for

further

Al-Mont-EnPILC

CTAB-Al-Mont-EnPILC

2795 ± 300 (10.3 ± 3.2)  103 0.99

4760 ± 535 (5.1 ± 1.3)  103 0.99

Bed height = 0.15, 0.20, 0.25 m; flow rate = 1.3 mL/min; C0 = 50 lmol/L.

105

M. Lezehari et al. / Journal of Colloid and Interface Science 379 (2012) 101–106 Table 4 Breakthrough curves: Wolborska model. Breakthrough curves conditions Z (m) 0.15 0.69 U0 (m h1) C0 (lM) 50

0.20 0.69 50

0.25 0.69 50

0.15 0.69 25

0.15 0.69 10

0.15 1.32 50

0.15 2.65 50

Al-Mont-EnPILC N0 (lM L1) ba (h1) m (mm h1) R2 SS syx

3370 ± 38 23.3 ± 0.4 10.1 0.988 6.7 0.63

3401 ± 46 21.3 ± 0.5 10.0 0.981 4.1 0.47

3480 ± 35 20.2 ± 0.4 9.7 0.990 12.2 0.67

2342 ± 53 30.9 ± 1.0 7.3 0.977 0.66 0.15

1161 ± 16 48.1 ± 1.7 5.9 0.950 0.44 0.13

3016 ± 41 44.3 ± 07 21.5 0.989 21.2 1.00

3056 ± 49 74.8 ± 1.3 42.7 0.972 28.0 1.13

CTAB-Al-Mont -EnPILC N0 (lM L1) ba (h1) m (mm h1) R2 SS syx

3735 ± 29 33.5 ± 0.5 9.1 0.992 4.6 0.48

4558 ± 114 24.7 ± 1.3 7.5 0.963 8.2 0.57

4131 ± 31 32.8 ± 1.2 8.3 0.959 19.9 1.14

3152 ± 59 36.8 ± 1.2 5.4 0.976 2.0 0.29

1736 ± 23 61.5 ± 3.0 4.0 0.978 0.18 0.08

3907 ± 43 55.4 ± 1.0 16.7 0.992 2.8 0.36

3979 ± 58 96.5 ± 2.2 32.9 0.987 5.0 0.46

Z: bed height; U0: flow; C0: PCP initial concentration; N0: exchange capacity; ba: external mass transfer coefficient.

Table 5 Breakthrough curves: Thomas model. Breakthrough curves conditions Z (m) U0 (m h1) C0 (lM)

0.15 0.69 50

0.20 0.69 50

0.25 0.69 50

0.15 0.69 25

0.15 0.69 10

0.15 1.32 50

0.15 2.65 50

Al-Mont-EnPILC kTHO  103 (L lmol1 h1) q0 (lmol g1) R2 SS syx

6.68 ± 0.12 115.0 ± 0.5 0.994 179 1.38

6.48 ± 0.09 115.4 ± 0.3 0.997 68.5 0.77

6.44 ± 0.08 116.2 ± 0.2 0.997 36.8 0.52

13.36 ± 0.18 78.8 ± 0.2 0.997 17.2 0.38

34.56 ± 0.50 41.5 ± 0.1 0.997 1.16 0.09

12.58 ± 0.24 109.9 ± 0.5 0.994 204 1.17

24.90 ± 0.51 102.6 ± 0.6 0.992 236 1.59

CTAB-Al-Mont -EnPILC kTHO  103 (L lmol1 h1) q0 (lmol g1) R2 SS syx

6.52 ± 0.12 146.6 ± 0.6 0.993 145 1.18

8.07 ± 0.09 150.7 ± 0.2 0.998 30.8 0.49

7.48 ± 0.11 150.2 ± 0.2 0.996 41.0 0.52

12.69 ± 0.19 109.3 ± 0.2 0.995 17.2 0.35

31.67 ± 0.39 63.5 ± 0.1 0.997 1.37 0.10

12.42 ± 0.23 146.1 ± 0.5 0.993 134 1.14

24.77 ± 0.47 141.2 ± 0.6 0.993 232 1.50

Z: bed height; U0: flow; C0: PCP initial concentration; kTHO: Thomas rate constant coefficient; q0: Thomas uptake capacity.

from the liquid to the solid phase increased linearly with flow rate and same order of values were obtained for both Al-Mont-EnPILC and CTAB-Al-Mont-EnPILC. The highest value indicates that the column adsorption is kinetically favourable. An increase in flow rate did not increase the correlation between the experimental and calculated data. 4. Conclusion The removal of PCP in a packed bed system using PILC/alginate microbeads is an effective and feasible method. Breakthrough times and uptake capacities largely depend on bed height, influent PCP concentrations and flow rate. Higher values of removal were obtained for the encapsulated surfactant-modified aluminium-pillared clay compared with the encapsulated aluminium-pillared clay. Uptake capacities were compared with other low-cost sorbents with good performance for the two pillared clays studied; however, these values were lower than those obtained with Granular Activated Carbon. Linear relationship was obtained for the BDST model; bed capacity increased sharply with the introduction of CTAB in the inorgano-pillared clay, while in contrast, kBDST decreased. Thomas model successfully predicted breakthrough curves obtained under varying experimental conditions. Experimental and Thomas model-predicted equilibrium uptake capaci-

ties were in accordance for both Al-Mont-EnPILC and CTAB-AlMont-EnPILC.

References [1] D.G. Crosby, Pure Appl. Chem. 53 (1981) 1051–1080. [2] J.A. Field, R. Sierra-Alvarez, Rev. Environ. Sci. Technol. 7 (2008) 211–241. [3] W. Jianlong, Q. Yi, N. Horan, E. Stentiford, Bioresour. Technol. 75 (2000) 157– 161. [4] Z. Shaokui, Y. Zhifeng, H.J. Do, H.P. Yun, Water Res. 38 (2004) 2315–2322. [5] R. Leyva-Ramos, P.E. Diaz-Flores, J. Leyva-Ramos, A. Femat-Flores, Carbon 45 (2007) 2280–2289. [6] S. Deng, R. Ma, Q. Yu, J. Huang, G. Yu, J. Hazard. Mater. 165 (2009) 408–414. [7] O. Bouras, T. Chami, M. Houari, H. Khalaf, J.C. Bollinger, M. Baudu, Environ. Technol. 23 (2002) 405–411. [8] O. Bouras, J.C. Bollinger, M. Baudu, H. Khalaf, Appl. Clay Sci. 37 (2007) 240–250. [9] J. Smith, S. Bartlett-Hunt, S. Burns, J. Hazard. Mater. 96 (2003) 91–97. [10] R. Yu, S. Wang, D. Wang, J. Ke, X. Xing, N. Kumada, N. Kinomura, Catal. Today 139 (2008) 135–139. [11] T. Vincent, A. Parodi, E. Guibal, Sep. Purif. Technol. 62 (2008) 470–479. [12] E. Guibal, A.F. Pinol, M. Ruiz, T. Vincent, C. Jouannin, A.M. Sastre, Sep. Sci. Technol. 45 (2010) 1935–1949. [13] S.F. Lim, J.P. Chen, Appl. Surf. Sci. 253 (2007) 5772–5775. [14] S.F. Lim, Y.M. Zheng, S.W. Zou, J.P. Chen, Environ. Sci. Technol. 42 (2008) 2551– 2556. [15] V. Rocher, A. Bee, J.M. Siaugue, V. Cabuil, J. Hazard. Mater. 178 (2010) 434–439. [16] A. Bee, D. Talbot, S. Abramson, V. Dupuis, J. Colloid Interface Sci. 362 (2011) 486–492. [17] C. Escudero, N. Fiol, I. Villaescusa, Environ. Chem. Lett. 4 (2006) 239–242.

106

M. Lezehari et al. / Journal of Colloid and Interface Science 379 (2012) 101–106

[18] N. Fiol, C. Escudero, J. Poch, I. Villaescuesa, React. Funct. Polym. 66 (2006) 795– 807. [19] C. Escudero, N. Fiol, I. Villaescusa, J.C. Bollinger, J. Hazard. Mater. 164 (2009) 533–541. [20] Y.L. Lai, G. Annadurai, F.C. Wang, J.F. Lee, J. Chem. Technol. Biotechnol. 83 (2008) 788–798. [21] A. Ely, M. Baudu, J.P. Basly, M.O.S.O Kankou, J. Hazard. Mater. 171 (2009) 405– 409. [22] H.A. Shawky, J. Appl. Polym. Sci. 119 (2011) 2371–2378. [23] A. Ely, M. Baudu, M.O.S.O Kankou, J.P. Basly, Chem. Eng. J. 178 (2011) 168–174. [24] M. Lezehari, J.P. Basly, M. Baudu, O. Bouras, Colloid Surface A 366 (2010) 88– 94. [25] H. Khalaf, O. Bouras, V. Perrichon, Microporous Mater. 8 (1997) 141–150. [26] R.A. Hutchins, Am. J. Chem. Eng. 80 (1973) 133. [27] L. Huidong, L. Ting, L. Zhao, D. Le, Bioresour. Technol. 99 (2008) 2234–2241. [28] D. Charumathi, N. Das, Desalination 285 (2012) 22–30. [29] C.M. Futalan, C.C. Kan, M.L. Dalida, C. Pascua, M.W. Wan, Carbohyd. Polym. 83 (2011) 697–704. [30] D. Chauhan, N. Sankararamakrishnan, J. Hazard. Mater. 185 (2011) 55–62. [31] H.C. Thomas, J. Am. Chem. Soc. 66 (1944) 1664–1666.

[32] A. Wolborska, Water Res. 23 (1989) 85–91. [33] K. Vijayaraghavan, Y.S. Yun, Chem. Eng. J. 145 (2008) 44–49. [34] M.E. Fernandez, G.V. Nunell, P.R. Bonelli, A.L. Cukierman, Bioresour. Technol. 101 (2010) 9500–9507. [35] Z. Aksu, S.S. Cagatay, F. Gönen, J. Hazard. Mater. 143 (2007) 362–371. [36] B.N. Estevinho, E. Ribeiro, A. Alves, L. Santos, Chem. Eng. J. 136 (2008) 188–194. [37] B.N. Estevinho, I. Martins, N. Ratola, A. Alves, L. Santos, J. Hazard. Mater. 143 (2007) 535–540. [38] I. Bras, L. Lemos, A. Alves, M.F.R. Almeida, in: 4th European Bioremediation Conference, September 3–6, 2008, Chiana (Greece). [39] O. Rubilar, G.R. Tortella, R. Cuevas, M. Cea, S. Rodriguez-Couto, M.C. Diez, Water Air Soil Poll. (published online 26.12.11). [40] W.W. Eckenfelder, Jr., W. Wesley, Industrial Water Pollution Control, third ed., McGraw-Hill, International edition, Singapore, 2000 cited by B.N. Estevinho et al. in reference 36. [41] D.C. Sharma, C.F. Forster, Bioresour. Technol. 52 (1995) 261–267. [42] Y.S. Al-Degs, M.A. M Khraisheh, S.J. Allen, M.N. Ahmad, J. Hazard. Mater. 165 (2009) 944–949. [43] O. Hamdaoui, J. Hazard. Mater. 161 (2009) 737–746.