Adsorption and kinetic studies of cationic and anionic dyes on pyrophyllite from aqueous solutions

Adsorption and kinetic studies of cationic and anionic dyes on pyrophyllite from aqueous solutions

Journal of Colloid and Interface Science 286 (2005) 53–60 www.elsevier.com/locate/jcis Adsorption and kinetic studies of cationic and anionic dyes on...

257KB Sizes 0 Downloads 14 Views

Journal of Colloid and Interface Science 286 (2005) 53–60 www.elsevier.com/locate/jcis

Adsorption and kinetic studies of cationic and anionic dyes on pyrophyllite from aqueous solutions a,∗ Aslıhan Gücek a , Sava¸s Sener ¸ , Sedat Bilgen b , M. Ali Mazmancı a a Department of Environmental Engineering, University of Mersin, 33343 Mersin, Turkey b Department of Civil Engineering, University of Mersin, 33343 Mersin, Turkey

Received 23 September 2004; accepted 11 January 2005 Available online 19 February 2005

Abstract The adsorption of cationic Methylene Blue (MB) and anionic Procion Crimson H-EXL (PC) dyes from aqueous medium on pyrophyllite was studied. Changes in the electrokinetics of pyrophyllite as a function of pH were investigated in the absence and presence of multivalent cations. The results show that pyrophyllite in water exhibits a negative surface charge within the range pH 2–12. Pyrophyllite is found to be a novel adsorbent for versatile removal of cationic and anionic dyes. The negative hydrophilic surface sites of pyrophyllite are responsible for the adsorption of cationic MB molecules. The adsorption of anionic PC dye is possible after a charge reversal by the addition of trivalent cation of Al. Nearly 2 min of contact time are found to be sufficient for the adsorption of both dyes to reach equilibrium. The experimental data follow a Langmuir isotherm with adsorption capacities of 70.42 and 71.43 mg dye per gram of pyrophyllite for MB and PC, respectively. For the adsorption of both MB and PC dyes, the pseudo-second-order chemical reaction kinetics provides the best correlation of the experimental data.  2005 Elsevier Inc. All rights reserved. Keywords: Pyrophyllite; Zeta potential; Dye; Adsorption isotherm; Adsorption kinetics

1. Introduction Many industries use dyes and pigments to color their products. The remaining dye molecules are common water pollutants in trace quantities in the wastewater. Their presence in water, even at very low concentrations, is highly visible and undesirable and may significantly affect photosynthetic activity in aquatic life due to reduced light penetration. The most widely used methods for removing dyes from wastewater systems include physicochemical, chemical, and biological methods, such as flocculation, coagulation, precipitation, adsorption, membrane filtration, electrochemical techniques, ozonation, and fungal decolorization [1]. However, wastewaters containing various dyes, due to their com* Corresponding author. Fax: +903-61-0032.

E-mail address: [email protected] (S. Sener). ¸ 0021-9797/$ – see front matter  2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2005.01.012

plex aromatic structure, are very difficult to treat using conventional wastewater treatment methods, since the dyes are generally stable under the influence of light and oxidizing agents, and reactive dyes are especially resistant to aerobic digestion [2]. Of the numerous techniques mentioned, adsorption in particular is an effective process for the removal of dyes from waste effluents. Currently, the most common procedure involves the use of activated carbons [3–5] as adsorbents because of their higher adsorption capacities. However, because of them relatively high cost, there have been attempts to utilize low-cost and naturally occurring adsorbents. There are very different studies on the use of low cost materials for removing dyes, such as various agricultural wastes [6–9], coal [10], lignite [11], sawdust [12], fuller’s earth [13], chitosan [14], fly ash [15], recycled alum sludge [16], slag [17], calcined alunite [18], alumina and kaolinite [19], perlite [20], sepiolite [21], montmoril-

54

A. Gücek et al. / Journal of Colloid and Interface Science 286 (2005) 53–60

Fig. 2. The molecular structure of methylene blue.

used to model the isotherm data for their applicability. Batch kinetic experiments were performed to provide appropriate equilibrium times. The experimental data were evaluated by applying the pseudo-first- and second-order and the intraparticle diffusion adsorption kinetic models. Fig. 1. Schematic diagram of the structure of pyrophyllite as viewed along the x-axis [55].

lonite [22], saponite [23], zeolite [3], and raw and modified diatomite [24,25]. Several studies have been performed on the use of pyrophyllite as an adsorbent for removing some contaminants such as heavy metals [26–28], boron ions [29], and cyanide [30]. However, no studies on the adsorption of dyes on pyrophyllite are available in the literature. Pyrophyllite is a nonswelling hydrous aluminum silicate with the chemical formula Al2 Si4 O10 (OH)2 . It belongs to the family of silicate minerals that are composed of three infinite layers formed by the sharing of oxygen ions at three corners of the silica tetrahedra. A layer of octahedrally coordinated Al–OH ions holds the two layers of tetrahedrally coordinated Si–O ions together as a three-layer sheet [31]. Since this three-layer unit is electrically balanced as neutral on the basal plane due to the very small ionic substitution, the crystal is held together by van der Waals forces, which are comparatively weak with respect to the primary covalent or ionic forces that hold the atoms in the unit layers together [32]. Consequently, easy cleavage takes place along the plane of the layers. During the grinding to produce pyrophyllite powder, some surfaces result from the easy cleavage between the layers, whereas lateral surfaces (edges) are formed by the fracture of the ionic and covalent bonds. Therefore, the basal surfaces where the interlayer space is devoid of the hydrated counterions [33] are hydrophobic, whereas edges resulting from the breaking of the sheets are hydrophilic in aqueous solutions, due to the occurrence of OH groups (essentially silanol) [34]. A schematic diagram of the structure of pyrophyllite is given in Fig. 1. In this study, pyrophyllite was used as an adsorbent for removal of both cationic and anionic dyes. The effect of the electrokinetic properties of pyrophyllite on adsorption was assessed in terms of the charge and magnitude of the zeta potential as a function of pH and multivalent cations. The number of dye molecules adsorbed at equilibrium were measured. The Langmuir and Freundlich isotherm models were

2. Materials and methods 2.1. Materials Lump-sized samples of pyrophyllite ore taken from Malatya, Turkey were subjected to crushing and grinding in a ball mill, followed by sedimentation to produce pyrophyllite fines (<2 µm). The sample was analyzed for its chemical composition and found to contain 63.47% SiO2 , 29.28% Al2 O3 , 0.16% Fe2 O3 , 0.18% TiO2 , 0.14% MgO, 0.30% CaO, 0.59% K2 O, 0.54% Na2 O, and 5.11% loss on ignition. The chemical analysis indicated that the concentrated pyrophyllite samples to be used as an adsorbent were composed of 91.25% of pyrophyllite. The specific surface area of the sample was determined as 7.03 m2 /g by a Micromeritics surface-area/pore-volume analyzer (Model 2100D) BET apparatus using nitrogen gas as the adsorbate. Detailed information on BET measurements is given by Davies and Kent [35]. The inherent pH of the suspension of pyrophyllite in distilled water was measured as 7.15. A thiazin group cationic, Methylene Blue (MB) (Merck grade C.I. 52015), and an azo group anionic, Procion Crimson H-EXL (PC) (a commercially available dye produced by Dystar Textilfarben GmbH & Co.), were used as model dyes without purification. Both dyes are solid phases. The molecular structure diagram of MB is given in Fig. 2 that of PC could not be provided for commercial reasons. The concentration of the dye in the solution after equilibrium adsorption was determined with a Shimadzu Brand UV-160 UV–visible spectrophotometer by measuring absorbance at λmax of 664 and 546 nm for MB and PC, respectively. All the chemicals used were of analytical grade of Merck. 2.2. Methods Electrokinetic studies were performed to determine the relationship between the zeta potential of the surface and the adsorption capacity. A Rank Brothers Mark II electrophoresis apparatus was used for electrokinetic measurements. The effect of pH and different inorganic electrolytes on the sign

A. Gücek et al. / Journal of Colloid and Interface Science 286 (2005) 53–60

and values of zeta potential of pyrophyllite were examined. In the apparatus, platinum electrodes were used and all measurements were conducted at 80 V and 298 K. The stock suspension was prepared with distilled water at a solid/liquid ratio of 1 g/L. To investigate the effect of various cations on the zeta potential of pyrophyllite, equimolar 4 × 10−4 M of KCl, CaCl2 and AlCl3 were used as supporting electrolytes. The suspension was ultrasonicated in a Bandelin Sonorex BK-106 ultrasonic bath (35 kHz) for 5 min to break up aggregates in the suspension prior to measurement. The resultant suspension (100 ml) was conditioned for 10 min, during which pH was adjusted to a predetermined pH by addition of a NaOH or HCl (0.1 N) solution under continuous stirring. A WTW 340i Model pH meter was used for measurement of the pH of solutions. The zeta potential values were calculated from electrophoretic mobility using the Smoluchowski equation [36]. Adsorption tests were done as a single-stage batch test using a Velp Model DLH mechanical stirrer. A suspension containing 1 g of adsorbent sample was mixed by stirring the mixture at 120 rpm with a 1-L aqueous solution of dye at a known initial concentration in a flask that was immersed in thermostatted water in a bath keeping a constant temperature of 298 K. Aliquot of the solution were withdrawn at a predetermined time intervals and were centrifuged at 10,000 rpm for 10 min to remove any adsorbent particles. The residual dye concentration in the filtrate was subsequently determined using a spectrophotometer at the wavelength corresponding to the maximum absorbance. The adsorption tests were continued until the equilibrium concentration was reached. The experiments were carried out by varying the concentration of dye solution from 20 to 150 mg/L. In the adsorption tests for the anionic dye, 4 × 10−4 M AlCl3 was first added into the adsorbent suspension to make the pyrophyllite surface charge positive prior to addition of dye solution. The data obtained from the adsorption tests were then used to calculate the adsorption capacity, qe (mg/g), of the adsorbent by a mass–balance relationship, which represents the amount of adsorbed dye per the amount of dry adsorbent, (C0 − Ce )V , (1) W where C0 and Ce are the initial and equilibrium concentrations of dye in solution (mg/L), respectively, V is the volume of the solution (L), and W is the weight of the dry adsorbent used (g). Finally, the adsorption capacity, qe , was plotted against equilibrium concentration, Ce . qe =

3. Results and discussion 3.1. Zeta potential of pyrophyllite The electrokinetic properties of pyrophyllite were necessary to gain a comprehensive understanding of adsorption

55

Fig. 3. Zeta potential of pyrophyllite as a function of pH (supporting electrolytes are equimolar 4 × 10−4 M of KCl, CaCl2 , and AlCl3 ).

mechanism of dye molecules at the solid/solution interface. The zeta potential of pyrophyllite as a function of suspension pH and in the presence of KCl, CaCl2 , and AlCl3 with equimolar ionic strength of 4 × 10−4 is presented in Fig. 3. As seen in Fig. 3, pyrophyllite exhibited a net negative surface charge and no charge reversal occurred over a wide pH range, 2–12, in the absence of an electrolyte. The negative zeta potential is a reflection of the charge from the edges. Hydrophobic surfaces of pyrophyllite are accessible to nonpolar organic species, acting as neutral adsorption sites. However, the fracture of the ionic and covalent bonds on the hydrophilic edges and some substitutions in the tetrahedral or octahedral layers, such as Al3+ instead of Si4+ or Mg2+ , Fe2+ and Ti2+ instead of Al3+ form negatively charged sites that are available for the adsorption of polar dye molecules from aqueous solutions [37–39]. Therefore, the negative hydrophilic surface sites of pyrophyllite particles were investigated by adsorption of cationic MB molecules. As the pH increases toward more alkaline values, the absolute value of the zeta potential increases toward more negative values, ultimately reaching a plateau at around −73 mV at pH 8–10 and −81 mV at pH 12. The results are in good agreement with those reported in the literature [31,40]. The undercoordinated cations and anions in the octahedra and tetrahedra sheets, by the fracture of chemical bonds, are Lewis acid and base sites that are unstable in the presence of water. Water molecules chemisorbed to Lewis acid–base sites stabilize edge faces both crystallochemically and electrostatically; the Coulomb potential, measured either within the layer or parallel to the layer, has a distinct negative trend near the edge face that can be traced to chemisorbed water molecules. The increase in zeta potential with increasing pH may be explained by the deprotonation of the water molecule coordinating Al at the edge face converting it into a hydroxyl ion [41]. In the presence of KCl and CaCl2 as supporting electrolytes, the monovalent and the divalent cations resulted in a decrease in the absolute value of the negative zeta potentials

56

A. Gücek et al. / Journal of Colloid and Interface Science 286 (2005) 53–60

of pyrophyllite, but no charge reversal was observed within the studied pH range (Fig. 3). This may be due to the precipitation of K2+ and Ca2+ as the counter ions on the negatively charged pyrophyllite surface or due to the ion exchange of the adsorbed hydronium cation with Ca2+ [41]. However, the addition of 4 × 10−4 M AlCl3 significantly reversed the charge of zeta potential to positive. As the pH increases, the zeta potential increases, reaches a maximum value of +75.14 mV at pH 8, and decreases toward more alkaline conditions. The reversal in the zeta potential sign can be interpreted via the strong specific adsorption of trivalent cation (Al3+ ), a lattice ion, on the negatively charged surface. The results obtained from the electrokinetic measurements were used to determine the electrostatic interaction between dye molecules and the surface of the pyrophyllite. The charge reversal made it possible to use pyrophyllite in the adsorption of anionic PC dye molecules.

3.3. Equilibrium adsorption isotherms The equilibrium adsorption of dyes was studied as a function of concentration. The amount of dye adsorbed (qe ), plotted against the equilibrium concentration (Ce ) for MB and PC, is given in Fig. 6. Initially, the adsorption isotherms of dye molecules show a steeply rising part, suggesting a strong affinity of the dye molecules for the polar surface sites on pyrophyllite. Then the amount of adsorption reaches a limiting value of around 70 and 71 mg/g for MB and PC, respectively. The equilibrium adsorption of dyes increases with the increase of initial dye concentration, showing the adsorption process to be dependent on the initial concentration. As also seen in Fig. 6, adsorption of both dyes onto pyrophyllite forms a typical Langmuir-type isotherm, which indicates that dye molecules outcompete water molecules for the sites available on the surface of pyrophyllite.

3.2. Adsorption studies To determine the equilibration concentration and time, the adsorption of the cationic MB and anionic PC dyes onto pyrophyllite was studied as a function of contact time. Figs. 4 and 5 show the effect of initial concentration of dyes on the adsorption capacity of pyrophyllite with a solid/ liquid ratio of 1 g/L at pH 8. At all initial dye concentrations studied, the adsorption takes place very fast initially: typically 90–95% of the ultimate adsorption occurs within the first 2 min of contact and gradually tails off thereafter. The independence of time required to achieve definite fraction of equilibrium adsorption on initial concentration may suggest that the adsorption process is second-order [42,43] which will be discussed later. Fig. 5. Effect of initial concentration on the adsorption of PC onto pyrophyllite. Conditions: 1 g/L adsorbent concentration, 4 × 10−4 M AlCl3 supporting electrolyte, 298 K temperature, and pH 8.

Fig. 4. Effect of initial concentration on the adsorption of MB onto pyrophyllite. Conditions: 1 g/L adsorbent concentration, 298 K temperature, and pH 8.

Fig. 6. Equilibrium isotherms for MB and PC adsorption onto pyrophyllite.

A. Gücek et al. / Journal of Colloid and Interface Science 286 (2005) 53–60

57

3.4. Adsorption isotherms Adsorption isotherms describe how adsorbates interact with adsorbents and are critical in optimizing the use of adsorbents. Therefore, the correlation of equilibrium data by either theoretical or empirical equations is essential to the practical design and operation of adsorption systems. To optimize the design of a sorption system to remove dyes from effluents, it is important to establish the most appropriate correlation for the equilibrium curves [14]. The experimental data of equilibrium isotherms for both dyes were modeled using isotherms by Freundlich [44] and Langmuir [45]. The applicability of the Freundlich adsorption isotherm is analyzed by plotting log(qe ) versus log(Ce ), but data are not found to be in good agreement, with correlation coefficients of 0.729 and 0.730 for MB and PC dyes, respectively. The Langmuir isotherm has been widely used to describe single-solute systems. It is based on the assumption that intermolecular forces decrease rapidly with distance and consequently it predicts monolayer coverage of the adsorbate on the outer surface of the adsorbent. The isotherm equation further assumes that adsorption takes place at specific homogeneous sites within the adsorbent and there is no significant interaction among adsorbed species. Theoretically, the adsorbent has a finite capacity for the adsorbate. Once a dye molecule occupies a site, no further adsorption can take place at that site. The rate of sorption to the surface should be proportional to a driving force which times an area. The driving force is the concentration in the solution, and the area is the amount of bare surface [52]. The saturated or monolayer capacity can be represented by the expression qe =

Q0 bCe , 1 + bCe

(2)

where qe is solid-phase adsorbate concentration at equilibrium (mg/g), Ce is aqueous-phase adsorbate concentration at equilibrium (mg/L), Q0 (mg/g) is the maximum amount of adsorbate per unit weight of adsorbent to form a complete monolayer on the surface, and b is the Langmuir isotherm constant (L/mg), related to the affinity of the adsorption sites. A plot of Ce /qe versus Ce gives a straight line of slope 1/Q0 and intercept 1/Q0 b, where Q0 gives the theoretical monolayer saturation capacity. Therefore, a linear expression for the Langmuir equation is Ce 1 1 = 0 + 0 Ce . qe Q b Q

(3)

The adsorption data were analyzed using the linear form (Eq. (3)) of the Langmuir isotherm. The plots of specific sorption, Ce /qe , against the equilibrium concentration, Ce , for MB and PC are shown in Fig. 7. The isotherm constants, b, and equilibrium monolayer capacities, Q0 , are presented in Table 1.

Fig. 7. Langmuir isotherms of MB and PC dyes adsorbed on pyrophyllite. Table 1 Langmuir adsorption isotherm constants for MB and PC adsorption onto pyrophyllite Dye MB PC

Langmuir isotherm Q0 (mg/g)

b (L/mg)

R2

70.42 71.43

2.06 1.94

0.999 0.998

The Langmuir isotherms for MB and PC dyes (Fig. 7) are nearly parallel to one another and yield very close values for the coefficients b (2.06 and 1.94 for MB and PC, respectively). The Langmuir monolayer capacity Q0 of MB and PC dyes were 70.42 and 71.43 mg/g, respectively. The Langmuir model is applicable when there is a strong specific interaction between the surface and the adsorbate so that a single adsorbed layer forms. The large b values relative to the energy of adsorption (more than unity) for MB and PC dyes imply strong bonding, most probably chemisorption, of the dye molecules onto the Lewis base and acid sites of the pyrophyllite, respectively. The isotherms of the MB and PC dyes were found to be linear over the whole concentration range and the correlation coefficients, R 2 , were extremely high (0.999 and 0.998 for MB and PC, respectively), suggesting that the adsorption of both dyes onto pyrophyllite closely follow a Langmuir isotherm. The high fit to the Langmuir model for both dyes suggests that the adsorption is limited with monolayer coverage and the surface is relatively homogeneous in terms of functional groups. However, it is well known that pyrophyllite, very similarly to talc, actually exhibits a structural heterogeneity that is caused by both the surface topology and the chemical composition [46]. The hydrophilic sites (lateral surface or edge) influence the waterwetting behavior of the surface or the adsorption of polar organic molecules, whereas the hydrophobic sites (basal surface) are responsible for the adsorption of nonpolar molecules [38]. It was considered that the polar MB molecules chemically adsorb on the edge of the crystal, forming inner-sphered binding with the tetrahedrally and octahedrally coordinated Lewis base sites that are previously hydrated with hydroxyls of the water molecules. Similarly, the adsorption of PC molecules is considered to be

58

A. Gücek et al. / Journal of Colloid and Interface Science 286 (2005) 53–60

Table 2 Comparison of the pseudo-first- and second-order and intraparticle diffusion adsorption constants at different initial concentrations of MB and PC dyes Initial concentration, C 0 (mg/L)

q e,exp (mg/g)

Pseudo-second-order ks (g/(mg min))

MB

20 40 60 80 100 150

19.69 39.67 56.82 66.93 69.42 69.99

PC

20 40 60 80 100 150

19.73 39.69 57.00 69.02 69.28 71.01

Dye

q e,cal (mg/g)

R2

Pseudofirst-order R2

Intraparticle diffusion R2

3.9 × 10−2 3.2 × 10−2 3.8 × 10−2 6.2 × 10−3 6.5 × 10−3 7.4 × 10−3

19.65 39.37 56.50 66.67 68.97 70.42

0.998 0.998 0.998 1.000 0.998 0.999

0.486 0.401 0.488 0.517 0.551 0.526

0.458 0.416 0.384 0.493 0.512 0.561

8.8 × 10−2 8.2 × 10−2 7.5 × 10−2 6.8 × 10−2 6.2 × 10−2 7.4 × 10−2

19.46 39.22 56.60 68.03 68.49 70.92

1.000 0.999 0.998 0.998 0.998 0.997

0.288 0.245 0.311 0.293 0.425 0.272

0.526 0.485 0.392 0.444 0.524 0.548

chemically bonded on the Lewis acid sites, which are coordinated by Al3+ added into the system for charge reversal. The adsorption of both dye molecules on the hydrophobic basal surface may be negligible. The hydrophobic character of the basal surface of pyrophyllite is consistent with experimental values of the high contact angle, which are measured as 79.2◦ [47] and as 65◦ [31]. In our case, therefore, even if the surface of pyrophyllite is actually structural heterogeneous, it is suggested that the surface is energetically homogeneous with respect to adsorption energy, since the only active adsorption sites for the polar dye molecules are hydrophilic edge surfaces. The effect of isotherm shape can be used to predict whether a sorption system is favorable or unfavorable in batch processes [48]. According to Hall et al. [49], the essential features of the Langmuir isotherm can be expressed in terms of a dimensionless constant separation factor or equilibrium parameter KR , which is defined by the relationship

3.5. Kinetics of adsorption

a pseudo-first-order [6,50], a pseudo-second-order [51,52], and an intraparticle diffusion model [53]. The correlation coefficients, R 2 , (given in Table 2) of the pseudo-first-order model resulted as 0.551 and 0.425 and underestimated the amount of the dye adsorbed at equilibrium for MB and PC, respectively. The calculated qe values obtained from the first-order kinetic model do not give reasonable values, which are too low compared with experimental qe values. This finding suggests that the sorption of MB and PC by pyrophyllite is not diffusion-controlled and the process does not follow the pseudo-first-order adsorption rate expression of Lagergren. The nature of the rate-limiting step in a batch system can also be assessed from the properties of the solute and adsorbent. Weber and Morris [53] stated that if intraparticle diffusion is the rate-controlling factor, uptake of the adsorbate varies with the square root of time. Thus, rates of adsorption are usually measured by determining the adsorption capacity of the adsorbent as a function of the square root of time [54]. Best-fit straight lines (plots are not shown) that do not pass through the origin indicating that there is an initial boundary layer resistance. The correlation coefficients, R 2 , (given in Table 2) for the intraparticle diffusion at different initial concentrations (20–150 mg/g) did not exceed the values of 0.561 and 0.548 for MB and PC dyes, respectively. The results also indicate that adsorption of MB and PC dyes onto pyrophyllite is not diffusion controlled. On the other hand, the rate of pseudo-second-order reaction is dependent on the amount of solute adsorbed on the surface of adsorbent and the amount adsorbed at equilibrium. The pseudo-second-order model can be represented in the form

It is important to be able to predict the rate at which contamination is removed from aqueous solutions in order to design an adsorption treatment plant. In order to investigate the mechanism of adsorption and potential rate controlling steps such as mass transfer and chemical reaction, the kinetic adsorption data given in Figs. 4 and 5 were modeled using

dqt (5) = ks (qe − qt )2 , dt where ks is the rate constant of the pseudo-second-order model (g/(mg min)) and qe and qt are the amount of the dye adsorbed at equilibrium and at any time t, in mg/g, respectively. After integration of Eq. (5) for boundary conditions

KR =

1 , 1 + bC0

(4)

where KR is a dimensionless separation factor, C0 is initial concentration (mg/L), and b is the Langmuir constant (L/mg). The parameter KR indicates the shape of the isotherm accordingly: The value of KR indicates the type of the isotherm to be either unfavorable (KR > 1), linear (KR = 1), favorable (0 < KR < 1), or irreversible (KR = 0). The KR values calculated indicate that adsorption of MB and PC dyes on pyrophyllite is favorable (0 < KR < 1) for all initial dye ion concentrations.

A. Gücek et al. / Journal of Colloid and Interface Science 286 (2005) 53–60

59

The straight lines for all initial concentrations (Figs. 8 and 9) with extremely high correlation coefficients (<0.998) for the pseudo-second-order kinetic model compared to those for the pseudo-first-order and the intraparticle diffusion kinetic models for the adsorption of MB and PC dye ions onto pyrophyllite strongly suggest that all the adsorption systems are a pseudo-second-order model, based on the assumption [42] that the rate-limiting step may be chemical sorption or chemisorption involving valency forces through sharing or exchange of electrons between the hydrophilic edge sites of pyrophyllite and polar dye ions. The calculated qe values also agree very well with the experimental data in the case of pseudo-second-order kinetics. The adsorption process involved a chemical reaction demonstrates that the pseudo-second-order rate constant, ks , is a function of the surface area of the adsorbent. Fig. 8. Plot of amount of adsorption versus time for MB dye onto pyrophyllite at various initial concentrations.

4. Conclusion The study presented revealed that pyrophyllite can be used as low-cost adsorbents for removing both cationic and anionic dyes. The equilibrium adsorption data of the dyes can be best modeled using a Langmuir approach with adsorption capacities of 70.42 and 71.43 mg/g for MB and PC, respectively. For the adsorption of both MB and PC dyes, chemical reaction seems significant in the rate-controlling step and the pseudo-second-order chemical reaction kinetics provide the best correlation for the experimental data. Acknowledgments

Fig. 9. Plot of amount of adsorption versus time for PC dye onto pyrophyllite at various initial concentrations.

t = 0 to t = t and qt = 0 to qt = qt , the following form can be obtained: 1 1 t = + t. (6) qt ks qe2 qe Plotting t/qt against t, a line can be obtained and a value for qe can be calculated. The values of t/qt shown in Figs. 8 and 9 were calculated from the kinetic data of Figs. 4 and 5 and plotted against time for adsorption of MB and PC on pyrophyllite, respectively, at different initial concentrations (20–150 mg/g). The results show that an increase in the initial dye concentrations produces a reduction in the percentage removal of dyes from the water and the adsorption processes for both dyes are highly dependent on the initial concentration. The slopes and intercepts of the plots of t/qt versus t were used to calculate the ks and qe . The calculated ks and qe values and correlation coefficients for MB and PC dyes are given in Table 2.

The authors thank the Scientific and Technical Research Council of Turkey, TÜB˙ITAK, for financially supporting this research through Grant ˙IÇTAG-Ç048 (102I034). The authors also acknowledge Dr. C. Kantar for his valuable critical comments. Appendix A. Nomenclature b C0 Ce KR ks Q0 qe qt qt,cal qt,exp R t V W

Langmuir constant (L/mg) initial dye concentration in solution (mg/L) equilibrium concentration (mg/L) dimensionless separation factor rate constant of pseudo-second-order model (gadsorbent /(mgdye min)) maximum adsorption capacity (mg/g) adsorption capacity in equilibrium (mg/g) amount of adsorption at time t (mg/g) amount of adsorption value calculated by the model experimental amount of adsorption value linear correlation coefficient time (min) volume of solution (L) mass of adsorbent (g)

60

A. Gücek et al. / Journal of Colloid and Interface Science 286 (2005) 53–60

References [1] A. Dabrowski, Adv. Colloid Interface Sci. 93 (1–3) (2001) 135–224. [2] J. Orthman, H.Y. Zhu, G.Q. Lu, Sep. Purif. Technol. 31 (1) (2003) 53–59. [3] V. Meshko, L. Markovska, M. Mincheva, A.E. Rodrigues, Water Res. 35 (14) (2001) 3357–3366. [4] C. Pelekani, V.L. Snoeyink, Carbon 38 (10) (2000) 1423–1436. [5] G.M. Walker, L.R. Weatherley, Sep. Sci. Technol. 35 (9) (2000) 1329– 1341. [6] Y.S. Ho, G. McKay, Water Res. 33 (2) (1999) 578–584. [7] R.-S. Juang, F.-C. Wu, R.-L. Tseng, J. Colloid Interface Sci. 227 (2) (2000) 437–444. [8] T. Robinson, B. Chandran, P. Nigam, Environ. Int. 28 (2002) 29–33. [9] T. Robinson, B. Chandran, P. Nigam, Water Res. 36 (2002) 2824– 2830. [10] A.K. Mittal, C. Venkobachar, J. Environ. Eng. A SCE (1993) 366–368. [11] S.J. Allen, G. McKay, K.Y.H. Khader, J. Chem. Technol. Biotechnol. 45 (4) (1989) 291–302. [12] V.K. Garg, R. Gupta, A.B. Yadav, R. Kumar, Bioresour. Technol. 89 (2) (2003) 121–124. [13] G. Atun, G. Hisarli, W.S. Sheldrick, M. Muhler, J. Colloid Interface Sci. 261 (1) (2003) 32–39. [14] Y.C. Wong, Y.S. Szeto, W.H. Cheung, G. McKay, Process Biochem. 39 (6) (2004) 695–704. [15] D. Mohan, K.P. Singh, G. Singh, K. Kumar, Ind. Eng. Chem. Res. 41 (15) (2002) 3688–3695. [16] M. Basibuyuk, C.F. Forster, Process Biochem. 38 (9) (2003) 1311– 1316. [17] V.K. Gupta, I.A. Suhas, D. Mohan, J. Colloid Interface Sci. 265 (2) (2003) 257–264. [18] M. Özacar, ˙I.A. Sengil, ¸ J. Hazard. Mater. 98 (1–3) (2003) 211–224. [19] R.G. Harris, J.D. Wells, B.B. Johnson, Colloids Surf. A Physicochem. Eng. Aspects 80 (1–2) (2001) 131–140. [20] M. Doˇgan, M. Alkan, Chemosphere 50 (4) (2003) 517–528. [21] G. Rytwo, D. Tropp, C. Serban, Appl. Clay Sci. 20 (6) (2002) 273– 282. [22] A.H. Gemeay, A.S. El-Sherbiny, A.B. Zaki, J. Colloid Interface Sci. 245 (1) (2002) 116–125. [23] N. Miyamoto, R. Kawai, K. Kuroda, M. Ogawa, Appl. Clay Sci. 16 (3– 4) (2000) 161–170. [24] E. Erdem, G. Çölgeçen, R. Donat, Colloid Interface Sci., in press. [25] M.A.M. Khraisheh, Y.S. Al-Degs, W.A.M. Mcminn, Chem. Eng. J. 99 (2004) 177–184. [26] R.G. Ford, D.L. Sparks, Environ. Sci. Technol. 34 (12) (2000) 2479– 2483. [27] M. Prasad, S. Saxena, S.S. Amritphale, N. Chandra, Ind. Eng. Chem. Res. 39 (8) (2000) 3034–3037. [28] A.M. Scheidegger, G.M. Lamble, D.L. Sparks, Environ. Sci. Technol. 30 (2) (1995) 548–554.

[29] R. Keren, D.L. Sparks, Soil Sci. Soc. Am. J. 58 (1994) 1095–1100. [30] S. Saxena, M. Prasad, S.S. Amritphale, N. Chandra, Sep. Purif. Technol. 24 (1–2) (2001) 263–270. [31] Y. Hu, X. Liu, Z. Xu, Miner. Eng. 16 (3) (2003) 219–227. [32] R. Keren, D.L. Sparks, Soil Sci. Soc. Am. J. 59 (2) (1995) 430–435. [33] C.H. Bridgeman, A.D. Buckingham, N.T. Skipper, M.C. Payne, Mol. Phys. 89 (3) (1996) 879–888. [34] B.J. Teppen, C.H. Yu, D.M. Miller, L. Schafer, J. Comput. Chem. 19 (2) (1998) 144–153. [35] J.A. Davis, D.B. Kent, in: M.F. Hochella, A.F. White (Eds.), Mineral– Water Interface Geochemistry, Rev. Mineral., vol. 23, Mineral Soc. of America, Washington, 1990, pp. 177–259. [36] A.L. Smith, in: Parfitt (Ed.), Dispersion of Powders in Liquids: With Special Reference to Pigments, second ed., Applied Science, London, 1973, pp. 86–131. [37] M. Erdemo˘glu, S. Erdemo˘glu, F. Sayılkan, M. Akarsu, S. ¸ Sener, ¸ H. Sayılkan, Appl. Clay Sci. 27 (2004) 41–52. [38] C.I. Sainz-Diaz, V. Timon, V. Botella, E. Artacho, A. Hernandez-Laguna, Am. Mineral. 87 (7) (2002) 958–965. [39] H. Sayılkan, S. Erdemo˘glu, S. ¸ Sener, ¸ F. Sayılkan, M. Akarsu, M. Erdemo˘glu, J. Colloid Interface Sci. 275 (2) (2004) 530–538. [40] S.M. Zhao, D.Z. Wang, Y.H. Hu, B.D. Liu, J. Xu, Miner. Eng. 16 (12) (2003) 1391–1395. [41] W.F. Bleam, G.J. Welhouse, M.A. Janowiak, Clays Clay Miner. 41 (3) (1993) 305–316. [42] Y.S. Ho, G. McKay, Process Biochem. 34 (1999) 451–565. [43] A.K. Jain, V.K. Gupta, A. Bhatnagar, I.A. Suhas, J. Hazard. Mater. 101 (1) (2003) 31–42. [44] H.M.F. Freundlich, J. Phys. Chem. 57 (1906) 385–470. [45] I. Langmuir, J. Am. Chem. Soc. 40 (1918) 1361. [46] C. Charnay, S. Lagerge, S. Partyka, J. Colloid Interface Sci. 233 (2001) 250–258. [47] R.F. Giese, P.M. Costanzo, C.J. Vanoss, Phys. Chem. Miner. 17 (7) (1991) 611–616. [48] V.J.P. Poots, G. McKay, J.J. Healy, J. Water Pollut. Control Fed. 50 (1978) 926–935. [49] K.R. Hall, L.C. Eagleton, A. Acrivos, T. Vermeulen, Ind. Eng. Chem. Fundam. 5 (1966) 212–223. [50] S. Lagergren, Kungl. Sv. Vetensk. Handl. 24 (1898) 1–39. [51] Y.S. Ho, G. McKay, Process Saf. Environ. Prot. B 76 (1998) 332– 340. [52] Y.S. Ho, J.C.Y. Ng, G. McKay, Sep. Sci. Technol. 36 (2) (2001) 241–261. [53] W.J. Weber, J.C. Morris, J. Sanit. Eng. Div. Am. Soc. Civ. Eng. 89 (1963) 31–60. [54] Y.S. Ho, G. McKay, Process Biochem. 38 (2003) 1047–1061. [55] P.A. Ciullo, C.S. Thompson, in: D.D. Carr (Ed.), Industrial Minerals and Rocks, Society for Mining, Metallurgy, and Exploration, Littleton, CO, 1994, pp. 815–826.