Mechanism of Sorption of Phenols from Aqueous Solutions onto Surfactant-Modified Montmorillonite

Journal of Colloid and Interface Science 254, 234–241 (2002) doi:10.1006/jcis.2002.8629

Mechanism of Sorption of Phenols from Aqueous Solutions onto Surfactant-Modified Montmorillonite Ruey-Shin Juang,∗,1 Su-Hsia Lin,† and Kung-Hsuen Tsao∗ ∗ Department of Chemical Engineering, Yuan Ze University, Chung-Li 320, Taiwan; and †Department of Chemical Engineering, Nanya Institute of Technology, Chung-Li 320, Taiwan Received April 16, 2002; accepted July 22, 2002

Equilibrium and kinetic studies on the sorption of phenol, mnitrophenol (m-NP), and o-cresol from water onto montmorillonite modified with cetyltrimethylammounium bromide (CTAB) were conducted. Experiments were carried out as a function of solution pH, sorbate concentration, and temperature (25–55◦ C). It was shown that the sorption capacity decreased in the order phenol > ocresol > m-NP. The Langmuir, dual-mode sorption, and Redlich– Peterson models were tested to fit the sorption isotherms of singlesolute systems, whereas the Langmuir competitive model was used to describe bisolute sorption equilibria. Thermodynamic parameters (H◦ and S◦ ) and the mean free energy (E) for the sorption of phenols were determined from the temperature dependence of the distribution constant and the Dubinin–Radushkevick equation, respectively. A simplified kinetic model was proposed to confirm the sorption mechanism. C 2002 Elsevier Science (USA) Key Words: montmorillonite; CTAB; sorption mechanism; isotherms; kinetics; phenol; m-nitrophenol; o-cresol.

INTRODUCTION

Various physicochemical and biological methods have been applied to remove organic contaminants from wastewater. Among these techniques, the use of organically modified clays for sorption removal has been paid increasing attention because they are cheaper than other sorbents such as activated carbons. Clays are widely applied in many fields of technology and science, e.g., the removal of liquid impurities and the purification of gases (1–3). The wide usefulness of clays is a result of their high specific surface area, their high chemical and mechanical stability, and a variety of surface and structural properties. The chemical nature and pore structure of clays generally determine their sorption ability. For gas-phase sorption, the pore structure (that is, the nature and/or volume of the pores) is a predominant factor. In the case of liquid-phase sorption, the chemical properties of surface groups actually affect the extent of sorption. The Na+ -type of montmorillonite, which is a 2 : 1 layered silicate, swells as it contacts water. The inner layer is composed of

1 To whom correspondence should be addressed. Fax: +886-3-4559373. E-mail: [email protected].

0021-9797/02 $35.00

 C 2002 Elsevier Science (USA)

All rights reserved.

234

an octahedral sheet of general form M2–3 (OH)6 (M is typically Al), which is situated between two SiO4 tetrahedral sheets (4). Substitution of Al3+ for Si4+ in the tetrahedral layer and Mg2+ or Zn2+ for Al3+ in the octahedral layer results in a net negative charge on the clay surfaces. The charge imbalance is offset by the exchangeable cations such as H+ , Na+ , or Ca2+ on the layer surfaces. In aqueous solutions, water molecules are intercalated into the interlamellar space of montmorillonite, leading to an expansion of the minerals. Clays, which are inherently hydrophilic due to hydration of metal ions, become hydrophobic by ion-exchanging long-chain quaternary amine cations for metal ions on the clays (4). The sorption of organic contaminants from water using organoclays has been widely studied (5–14). Boyd et al. (10) modified smectite with cetyltrimethylammonium bromide (simply as CTAB–smectite) to sorb trichloroethylene and benzene and indicated that the sorption is affected by partition action. Lawrence et al. (11) studied the uptake of phenol and chlorinated phenols using tetramethylammonium (TMA) and tetramethylphosphonium (TMP) montmorillonite. They found that the selectivity is size- and shape-dependent and is not strongly affected by water solubility. Also, TMP smectite is a better sorbent than TMA, which does not measurably sorb any of the phenols. Lee et al. (12) also studied the sorption of aromatic compounds from water onto the TMA smectite and stated that the sorption behavior is different from that modified with CTAB. The isotherms of benzene, toluene, and o-xylene onto TMA smectite are nonlinear, indicating that the “sorption” occurs by sorption, rather than by partition like the sorption onto the CTAB smectite. Although the organoclays have found growing use in actual field applications, some mechanisms such as the role of partition in single- and bisolute systems have not been fully clarified. In this work, phenol, m-nitrophenol (m-NP), and o-cresol were selected to reveal the effect of the interaction between phenols and CTAB and the water solubility of phenols on sorption capacity. The Langmuir model (LM), the dual mode sorption model (DSM), and Redlich–Peterson model (RPM) were used to analyze the single-solute sorption isotherms. Prediction and fitting of bisolute sorption isotherms by the Langmuir competitive model (LCM) and the competitive dual-mode sorption model (CDSM) were discussed. The thermodynamic

235

PHENOL SORPTION ONTO MONTMORILLONITE

TABLE 1 The Physicochemical Properties of Phenols (13, 15)

and kinetic parameters for sorption of phenols were finally evaluated. MATERIALS AND METHODS

Materials Montmorillonite K30 (Fluka Co.) had an idealized formula of Al2 O3 · 4SiO2 · nH2 O. The particle size was 0.03 mm and its cation exchange capacity was 33.2 meq/100 g. The BET surface area and average pore size were measured to be 330 m2 /g and 3.3 nm, respectively, from N2 sorption isotherms with a sorptiometer (Quantachrome NOVA 2000). Phenols (phenol, m-NP, o-cresol) and other inorganic chemicals were supplied by Merck Co. as analytical-grade reagents. Table 1 lists the physicochemical properties of phenols (13, 15). The initial concentrations of phenols in the aqueous solution ranged from 0.2 to 50 mol/m3 , and the solution pH was adjusted by adding a small amount of 0.1 mol/dm3 HCl or NaOH if necessary. Preparation of the CTAB-Modified Clays The modification of the clay was carried out as follows. A known amount of CTAB (Aldrich Co.) ranging from 1 to 20 mol/m3 was dissolved in 1 dm3 of deionized water (Milli-Q, Millipore), to which 20 g of the raw (unmodified) clays and acetone (5 cm3 ) were added. Acetone was used to stabilize the suspensions. Preliminary experiments had shown that the amount of CTAB sorbed was kept unchanged after the sample was shaken for more than 1 h. The solution mixture was centrifuged and the solid was washed 5 times with deionized water to remove superficial CTAB attached on the surface. The amount of CTAB sorbed was measured by a CHN Elemental Analyzer (Carlo Erba, EA1108). Zeta potentials of the clay suspensions were measured using a Malvern Zetasizer 3000HS system. After modification with 20 mol/m3 CTAB, the BET surface area and average pore size of the clays changed to about 70 m2 /g and 5.4 nm.

Solute

MW

Phenol 94.11 m-NP 139.11 o-Cresol 108.14

Molecular pK a (25◦ C) area (A)

Water solubility (g/dm3 )

Octanol–water partition coefficient (25◦ C)

9.92 7.23 10.21

8.2 (15◦ C) 1.1 (20◦ C) 2.4 (20◦ C)

30.2 81.2 95.5

240 299 278

sorbed at equilibrium, qe , was obtained by qe = (C0 − Ce )V /W,

[1]

where C0 and Ce are the initial and equilibrium liquid-phase concentrations of phenols, respectively, V is the volume of the solution, and W is the amount of dry sorbent used. Kinetic experiments were performed in a Pyrex glass vessel of 80 mm I.D. and 120 mm H, fitted with four glass baffles, 8 mm W. An aqueous solution (0.5 dm3 ) was poured and stirred using a Cole–Parmer Servodyne agitator with six blades, flatbladed impeller (12 mm H, 30 mm W). The stirring speed was 500 rpm since above that the agitation has little effect on sorption. An amount of clays (10 g) was added and the timing was started. The vessel was immersed in a water bath fixed at 25 ◦ C (Haake K-F3). At preset time intervals, aqueous sample (2 cm3 ) was taken through a 0.45 µm membrane filter, and the concentrations of phenols were analyzed. The amount of sorption at time t, qt , was calculated similarly to the method for obtaining the amount of sorption at equilibrium (Eq. [1]). RESULTS AND DISCUSSION

Properties of the CTAB-Modified Montmorillonite Figure 1 illustrates the isotherm for CTAB sorption onto the clay, indicating that the sorption is saturated when the CTAB

Procedures of Sorption Experiments For equilibrium sorption of phenols, 0.5 g of air-dried CTAB– clay was used in a batch vessel at a fixed temperature. The solution mixture (25 cm3 ) was allowed to shake for 12 h and was then centrifuged at 2500 rpm for 20 min. After filtration with glass fibers, the aqueous samples were withdrawn and were adjusted to pH 4.0 with an acetate buffer. The concentrations of phenols in the samples before and after sorption were analyzed using GC (Varian CP-3800) equipped with a flame ionization detector. A fused silica capillary column (J&W Scientific, DB-5, 30 m × 0.53 mm) was used. The amount of phenols sorbed onto the clay was calculated by difference. The pH was measured with a Horiba F-23 pH meter. Each experiment was performed twice at least under identical conditions. The reproducibility of the measurements was mostly within 4%. The amount of phenols

FIG. 1.

Isotherm for the sorption of CTAB onto the montmorillonite clay.

236

FIG. 2. 2 g/dm3 ).

JUANG, LIN, AND TSAO

Zeta potentials of the raw and CTAB–clay suspensions (clay dose,

concentration in the aqueous solution exceeds about 15 mol/m3 . However, the sorption decreases when the CTAB concentration is larger than 30 mol/m3 . This is likely due to the more serious steric effect of the CTAB molecules at higher concentrations, making penetration of the molecules into the internal pores of the clay more difficult. Figure 2 shows the zeta potentials of the raw (unmodified) and CTAB-modified clay suspensions at different pH values (clay dose, 2 g/dm3 ). The surface charge of the modified clay is positive in the whole pH range when the CTAB concentration is above 15 mol/m3 . The small variance in sorption ability with the equilibrium pH (Fig. 3) supports this point. This confirms that above this concentration the surface of the clays is almost completely occupied by surfactant (Fig. 1). Thus, 20 mol/m3 CTAB was used in this work to modify the clays for further equilibrium and kinetic studies.

As shown in Fig. 3, the sorption is kept unchanged at pH < 5 but gradually increases with a further increase in pH. The sorption is slightly enhanced at a higher pH likely due to partial ionization of the solutes (effect of electrostatic force). In studying the sorption of benzene, toluene, and o-xylene onto the TMAclays, Cadena (16) also indicated that pH does not affect the sorption of nonionizing organic matters. However, Kim et al. (14) have recently compared the sorption of 2-chlorophenol, 3cyclophenol, and 4-nitrophenol onto the CTAB-montmorillonite at pH 7.0 and 11.5 and concluded that the sorption is highly pH-dependent. This is possibly a result of partial coverage of the surfactant on the clays, as in the case of 10 mol/m3 CTAB-montmorillonite in Fig. 1. Because of the weak pH effect on the sorption, the initial solution pH value was not adjusted in this work for the sorption experiments. In this case, for example, the solution pH values before and after sorption are 6.50/3.70, 7.31/3.88, and 6.93/3.68 for the sorption of phenol, m-NP, and o-cresol, respectively, under typical conditions (clay dose, 20 g/dm3 ; phenol, 1.06 mol/m3 ; m-NP, 0.72 mol/m3 ; o-cresol, 0.93 mol/m3 ). Single-Solute Sorption Model The two-parameter Langmuir model (LM) and the threeparameter Redlich–Peterson model (RPM) (9) were used to fit the single-solute isotherms, which both obey the thermodynamic boundary condition of Henry’s law over an infinitely dilute concentration range. The LM and RPM in single-solute systems are expressed respectively by q L K L Ce 1 + K L Ce

[2]

qe =

aCe , 1 + bCeα

[3]

where Ce is the equilibrium liquid-phase concentration and qe is the amount of sorption per unit mass of clay. In Eq. [2], q L and K L denote monolayer sorption capacity and a constant related to equilibrium constant, respectively. In RPM, a, b, and α are empirical constants. The parameters in LM and RPM are determined using the BSOLVE optimization procedure based on Marquardt’s leastsquares algorithm (Table 2). As a measure of the degree of fitness, the correlation coefficient R 2 for the sorption systems is computed by R2 =

FIG. 3. Effect of equilibrium pH on sorption removal of phenols onto the CTAB–clay in the single-solute systems.

qe =



2 qe,exp −



(qe,exp − qe,cal )2

 

2 qe,exp , [4]

where the subscripts exp and cal are the experimental and calculated data, respectively. It is found that the fit of the three-parameter RPM is not better than that of LM. Another three-parameter dual-mode sorption model (DSM) was tested (13), which assumes that some of solutes dissolve in the medium (partition) and the rest sorb onto

237

PHENOL SORPTION ONTO MONTMORILLONITE

TABLE 2 Model Parameters for the Sorption of Single Solute onto the CTAB–Clay at 25◦ C Langmuir model (LM) Solute Phenol m-NP o-Cresol

KL

(m3 /mol) 0.044 0.175 0.119

Redlich–Peterson model (RPM)

q L (mol/kg)

R2

1.01 0.76 0.90

0.9958 0.9852 0.9918

a

b

(m3 /mol)

α (−)

R2

0.034 0.468 0.128

5.7 × 10−3 2.6 × 10−3 2.5 × 10−3

1.45 0.54 0.82

0.9933 0.9861 0.9902

the sorption sites. The linear and Langmuir type of isotherms are assigned to describe the contribution by partition and sorption, respectively. In this way, DSM in the single-solute systems has the form qe = K H C e +

qDS K DS Ce . 1 + K DS Ce

Dual-mode sorption model (DSM)

(m3 /kg)

[5]

The parameters K H , qDS , andK DS in DSM are also evaluated using the above regression technique. As shown in Table 2, DSM exhibits a slightly better fit than LM. Especially, the fit of m-NP by the DSM improves to the largest extent. This means that the partition/sorption ratio of mNP is the largest, compared to the other solutes, ascribable to its having the lowest water solubility (Table 1). Owing to the fact that phenol has the least steric effect and smallest molecular size, it has the largest qDS value. In addition, the decreasing order of K H value is m-NP > o-cresol > phenol, which is just reversed from the order of water solubility. Figure 4 compares the measured and DSM-fitted isotherms. The sorption capacity (in terms of q L obtained in LM) decreases in the order phenol > o-cresol > m-NP. These results agree with those obtained previously (10–14). For example, Sung et al. (13) found that the sorption capacity (in g/kg) on the CTAB– montmorillonite has the order m-NP ∼ p-NP > o-NP > phenol (in the present work the order becomes m-NP > o-cresol >

KH

(m3 /kg)

0 0.026 0.017

qDS (mol/kg)

R2

0.029 0.763 0.298

1.59 0.22 0.41

0.9972 0.9922 0.9941

phenol if the capacity is in g/kg, although this is misleading). They explained the difference in sorption capacity mainly by the van der Waals interaction between a solute and long hydrocarbon groups of the CTAB, originating from the difference in water solubility of the solutes. The carbonyl oxygen on the carbon acts as the electron donor and the aromatic ring of the solute acts as acceptor (17). Nitro-substituted phenols thus enhanced the donor–acceptor interaction due to the electron-withdrawing nature of the nitro group. Bisolute Sorption Model The Langmuir competitive model (LCM) and competitive dual-mode sorption model (CDSM) are used to “predict” bisolute sorption behavior (13). LCM is an extended form of LM which allows predictions of the sorption amount of solute i, qe,i , in the mixtures, qe,i =

q L ,i K L ,i Ce,i ,  1 + Nj=1 K L , j Ce, j

[6]

where Ce,i is the equilibrium concentration of solute i in a mixture containing N solutes, and K L ,i and q L ,i are the parameters determined by fitting LM to the single-solute isotherms. This is a concept of model prediction. Koros (18) first extented DSM to describe sorption of a binary gas mixture in glassy polymers. The DSM has been extented to multisolute liquid-phase sorption systems (19). In a medium the partition of a solute is not affected by the presence of other solutes, not showing competitive sorption behavior until the sorbed quantity approaches the dissolving capacity of the medium. However, the “sorption” term should be modified to consider the competition among the sorbing solutes for the limited sorption sites, i.e., LCM. Therefore, the so-called CDSM has the form qe,i = K H,i Ce,i +

FIG. 4. Sorption isotherms measured and calculated by the dual-mode sorption model (DSM) in the single-solute systems.

K DS (m3 /mol)

qDS,i K DS,i Ce,i ,  1 + Nj=1 K DS, j Ce, j

[7]

where the parameters K H,i , qDS,i and K DS,i are obtained by fitting the DSM to the single-solute sorption data, based on the concept of model prediction. Bisolute sorption was performed using the three binary systems (a) phenol/m-NP, (b) phenol/o-cresol, and (c) o-cresol/ m-NP. The measured data and model predictions by LCM and

238

JUANG, LIN, AND TSAO

CDSM are compared in Fig. 5. Although the fit of the singlesolute sorption data by the DSM is better, as indicated above, the prediction of the bisolute sorption data under the concentration ranges studied by the LCM (R 2 > 0.934) is better than that by the CDSM. The CDSM tends to be over- or underestimated. That is, the contribution of partition is depressed or enhanced in the bisolute systems, compared to the single-solute systems. From the viewpoint of competitive effect, Fig. 6 compares the isotherms in the single- and bisolute systems. In this figure, the curves are calculated based on data fitting by the LM (singlesolute systems) or LCM (bisolute systems). This is a concept of “model fitting.” Table 3 lists the fitted parameters. The sorption capacity decreases when the second solute is present. The extent of reduction in sorption capacity decreases in the order phenol > m-NP > o-cresol. The reduction of o-cresol is least, likely as a result of its having the highest interaction strength on the hydrophobic surface of CTAB–clay (in terms of the largest octanol–water partition coefficient shown in Table 1).

FIG. 6. Comparison of the sorption isotherms of phenols measured and calculated in the single and bisolute systems.

Thermodynamic Parameters The amounts of sorption of the single phenols by CTAB– clay were also measured in the temperature range 25–55◦ C. The equilibrium distribution constant K d is calculated by K d = qe /Ce .

[8]

The following relationships have been used to evaluate the

TABLE 3 Model Parameters for Sorption of Binary Solutes onto the CTAB–Clay at 25◦ C K L (m3 /mol)

R2

q L (mol/kg)

Solute 1/Solute 2 Solute 1 Solute 2 Solute 1 Solute 2 Solute 1 Solute 2 FIG. 5. Sorption isotherms measured and calculated by the Langmuir competitive model (LCM) and competitive dual-mode sorption model (CDSM) in the bisolute systems.

Phenol/m-NP Phenol/o-cresol o-Cresol/m-NP

0.056 0.085 0.131

0.125 0.151 0.153

0.98 1.05 1.15

0.71 1.10 0.98

0.9688 0.9822 0.9712

0.9971 0.9921 0.9802

239

PHENOL SORPTION ONTO MONTMORILLONITE

FIG. 7. Temperature dependence of the sorption distribution constants in the single-solute systems.

FIG. 8. Plot of the Dubinin–Radushkevick equation for the sorption of phenols in the single-solute systems.

thermodynamic parameters G ◦ , H ◦ , and S ◦ (4, 20):

qm is the D–R monolayer capacity, and K is a constant related to sorption energy (mol2 /kJ2 ). It should be noted that in Eq. [11] qe is in mol/g and Ce in mol/dm3 . The parameters qm and K can be obtained from the intercept and slope of the plot as shown in Fig. 8. The mean free energy of sorption, E, is calculated by (20, 22)

G ◦ = −RT ln K d   S ◦ −H ◦ 1 ln K d = + . R T R

[9] [10]

At 25◦ C, the change in free energy (G ◦ ) is calculated to be −9.7 kJ/mol for phenol, −12.8 kJ/mol for m-NP, and −11.7 kJ/mol for o-cresol. The spontaneous nature (G ◦ < 0) indicates that the surfactant molecules have more affinity toward phenols (21). As shown in Eq. [10], a plot of ln K d vs. 1/T (Fig. 7) gives H ◦ and S ◦ (R 2 > 0.986). The values of H ◦ and S ◦ (25◦ C) are listed in Table 4. A negative H ◦ value represents strong bonding between the sorbent and solute. Krishna et al. (20) also found the same tendencies in the sorption of Cr(VI) anionic species onto the CTAB-montmorillonite. The Dubinin–Radushkevick (D–R) equation is more general than LM because it does not assume a homogeneous surface or a constant sorption potential (22). The linear D–R equation is expressed by (20, 22)

E = (−2K )−1/2 .

[12]

The D–R parameters and mean free energy are given in Table 5. The magnitude of E is useful for estimating the type of sorption reaction. The value of E is obtained to be 6.2, 7.7, and 7.2 kJ/mol for phenol, m-NP, and o-cresol, respectively. Although they are lower than the energy range of the ion-exchange reaction, i.e., 8–16 kJ/mol, this still indicates that the sorption of phenols onto the modified clay mainly proceeds by ion exchange (20). Sorption Kinetics

where ε is the Polanyi potential, which equals RT ln(1 + 1/Ce ),

Figure 9 shows the typical time changes of solid-phase solute concentration. Several simplified kinetic models including the pseudo-first-order equation, pseudo-second-order equation, and intraparticle diffusion model were tested here to check the validity (23, 24). Only the pseudo-second-order equation is

TABLE 4 Thermodynamic Parameters for Sorption of Single Solutes onto the CTAB–Clay

TABLE 5 The Dubinin–Radushkevick Parameters for the Sorption of Single Solutes onto the CTAB–Clay

ln qe = ln qm − K ε 2 ,

[11]

Solute

H (kJ/mol)

S (J/(mol K))

G (kJ/mol)

Solute

K (mol/kJ)2

qm (mol/kg)

E (kJ/mol)

R2

Phenol m-NP o-Cresol

−6.7 −5.9 −3.6

9.7 23.7 26.7

−9.7 −12.8 −11.7

Phenol m-NP o-Cresol

−1.3 × 10−2 −8.4 × 10−3 −9.6 × 10−3

0.17 0.13 0.16

6.2 7.7 7.2

0.9964 0.9992 0.9821

240

JUANG, LIN, AND TSAO

TABLE 6 Kinetic Parameters for the Sorption of Single Solutes onto the CTAB–Clay at 25◦ C

FIG. 9. Time changes of the sorption of phenols onto the CTAB–clay in the single-solute systems.

applicable which is expressed by (23, 24) dqt = k2 (qe − qt )2 . dt

k2 (kg/mol · min)

qe (mol/kg)

h (mol/kg · min)

R2

Phenol m-NP o-Cresol

158 185 231

0.028 0.031 0.040

0.121 0.178 0.373

0.9962 0.9974 0.9979

In fact, Kaur et al. (25) described the sorption of Cu2+ and Pb2+ onto the bottom ash by the pseudo-second-order equation. Sayrafi et al. (26) analyzed the kinetics of Cd2+ sorption from polluted water by this model. Because this equation is basically based on the sorption capacity, the description of sorption phenomenon suggests that the chemical reaction is rate controlling (24). It is indicated that these chemisorption systems involve vacancy forces through sharing or exchange of electrons between the sorbent and solutes. As shown in Tables 2 and 6, the decreasing order of sorption capacity is phenol > o-cresol > m-NP, but that of sorption rate is m-NP > o-cresol > phenol.

[13] CONCLUSIONS

According to the conditions t = 0, qt = 0 and t = t, qt = qt , the integrated form of Eq. [13] becomes t 1 1 = + t. 2 qt k 2 qe qe

Solute

[14]

where k2 (kg/mol · min) is the rate constant and k2 qe2 (=h) is the initial rate (mol/kg · min). The parameters k2 and qe are obtained from the linear plot of t/qt vs. t, as shown in Fig. 10 (R 2 > 0.9962).

The mechanism of sorption of phenol, m-nitrophenol (m-NP), and o-cresol from water onto montmorillonite modified with CTAB in single- and bisolute systems was studied. Zeta potential data could determine the amount of CTAB required for complete coverage of the clay. The sorption capacity decreased in the order phenol > o-cresol > m-NP. Of the three models, the dual-mode sorption model was best fitted to the singlesolute isotherms, especially for m-NP. This indicated the largest partition/sorption ratio of m-NP due to its lowest water solubility. The Langmuir competitive model yielded better prediction of the bi-solute sorption data (R 2 > 0.934) than the competitive dual-mode sorption model. The mean free energy of sorption (6.2–7.7 kJ/mol) obtained from the Dubinin-Radushkevick equation probably indicated ion-exchange reaction nature for the sorption of phenols onto the CTAB-clay. The good description of sorption processes by the pseudo-second-order kinetic model (R 2 > 0.9962) also supported chemisorption being rate controlling. APPENDIX: NOMENCLATURE

Ce Ct C0 k2

FIG. 10. Test of the pseudo-second-order equation for the sorption of phenols onto the CTAB–clay.

qe qt R2

equilibrium solute concentration in the aqueous phase (mol/m3 ) solute concentration in the aqueous phase at time t (mol/m3) initial solute concentration in the aqueous phase (mol/m3 ) pseudo-second-order rate constant defined in Eq. [13] (kg/mol · min) amount of sorption at equilibrium (mol/kg) amount of sorption at time t (mol/kg) correlation coefficient defined in Eq. [4]

PHENOL SORPTION ONTO MONTMORILLONITE

t V W

time (min) volume of the solution (m3 ) amount of dry sorbent used (kg) REFERENCES

1. Carrado, K. A., Appl. Clay Sci. 17, 1 (2000). 2. Mortland, M. M., Shaobai, S., and Boyd, S. A., Clays Clay Miner. 34, 581 (1986). 3. Zhang, Z. Z., Spark, D. L., and Scrlvner, N. C., Environ. Sci. Technol. 27, 1625 (1993). 4. Zhu, L., Ren, X., and Yu, S., Environ. Sci. Technol. 32, 3374 (1998). 5. Smith, J. A., Jaffe, P. R., and Chiou, C. T., Environ. Sci. Technol. 24, 1167 (1990). 6. Zhu, L., Shen, B., and Shen, X., Environ. Sci. Technol. 34, 468 (2000). 7. Zhu, L., Li, Y., and Zhang, J., Environ. Sci. Technol. 31, 1407 (1997). 8. Smith, J. A., and Galan, A., Environ. Sci. Technol. 29, 685 (1995). 9. Boyd, S. A., Shaobai, S., Lee, J. F., and Morland, M. M., Clays Clay Miner. 36, 125 (1988). 10. Boyd, S. A., Morland, M. M., and Chiou, C. T., Soil Sci. Soc. Am. J. 52, 652 (1988). 11. Lawrence, M. A. M., Kukkadapu, R. K., and Boyd, S. A., Appl. Clay Sci. 13, 13 (1998).

241

12. Lee, H. J., Song, D. I., and Jeon, Y. W., Sep. Sci. Technol. 32, 1975 (1997). 13. Sung, C., Song, D. I., and Jeon, Y. W., Sep. Sci. Technol. 33, 1981 (1998). 14. Kim, D. G., Song, D. I., and Jeon, Y. W., Sep. Sci. Technol. 36, 3159 (2001). 15. Lide, D. R., “CRC Handbook of Chemistry and Physics,” 80th ed., pp. 16-42–16-46. CRC Press, Boca Raton, FL, 1999. 16. Cadena, F., J. Environ. Eng. ASCE 115, 756 (1989). 17. Mattson, J. S., and Mark, H. B., “Activated Carbon-Surface Chemistry and Sorption from Solutions.” Marcel Dekker, New York, 1971. 18. Koros, W. J., J. Polym. Sci. Polym. Phys. Ed. (18), 981 (1980). 19. Huh, J. K., Song, D. I., and Jeon, Y. W., Sep. Sci. Technol. 34, 571 (1999). 20. Krishna, B. S., Murty, D. S. R., and Prakash, B. S. J., J. Colloid Interface Sci. 229, 230 (2000). 21. Naseem, R., and Tahir, S. S., Water Res. 35, 3982 (2001). 22. Gemeay, A. H., El-Sherbiny, A. S., and Zaki, A. B., J. Colloid Interface Sci. 245, 116 (2002). 23. Juang, R. S., Wu, F. C., and Tseng, R. L., J. Colloid Interface Sci. 227, 437 (2000). 24. McKay, G., and Ho, Y. S., Water Res. 33, 578 (1999). 25. Kaur, A., Mailik, A. K., Verma, N., and Rao, A. L. J., Ind. J. Environ. Pollut. 11, 4335 (1991). 26. Sayrafi, O., Salim, R., and Sayrafi, S. A., J. Environ. Sci. Health A 31, 2503 (1996).