Journal of Colloid and Interface Science 317 (2008) 1117 www.elsevier.

com/locate/jcis

Adsorption of sodium dodecylbenzenesulfonate on activated carbons: Effects of solution chemistry and presence of bacteria
M.I. Bautista-Toledo, J.D. Mndez-Daz, M. Snchez-Polo, J. Rivera-Utrilla , M.A. Ferro-Garca
Departamento de Qumica Inorgnica, Facultad de Ciencias, Universidad de Granada, 18071 Granada, Spain Received 4 May 2007; accepted 16 September 2007 Available online 20 September 2007

Abstract The objective of the present investigation was to determine the effectiveness of activated carbon in removing sodium dodecylbenzenesulfonate (SDBS) and to analyze the chemical and textural characteristics of the activated carbons that are involved in the adsorption process. Studies were also performed on the inuence of operational variables (pH, ionic strength, and presence of microorganisms) and on the kinetics and interactions involved in the adsorption of this pollutant on activated carbon. The kinetics study of SDBS adsorption revealed no problems in its diffusion on any of the activated carbons studied, and WeiszPrater coefcient (C WP ) values were considerably lower than unity for all activated carbons studied. SDBS adsorption isotherms on these activated carbons showed that: (i) adsorption capacity of activated carbons was very high (260 470 mg/g) and increased with larger surface area; and (ii) dispersive interactions between SDBS and carbon surface were largely responsible for the adsorption of this pollutant. SDBS adsorption was not signicantly affected by the solution pH, indicating that electrostatic adsorbent adsorbate interactions do not play an important role in this process. The presence of electrolytes (NaCl) in the medium favors SDBS adsorption, accelerating the process and increasing adsorption capacity. Under the working conditions used, SDBS is not degraded by bacteria; however, the presence of bacteria during the process accelerates and increases SDBS adsorption on the activated carbon. Microorganism adsorption on the activated carbon surface increases its hydrophobicity, explaining the results observed. 2007 Elsevier Inc. All rights reserved.
Keywords: SDBS; Activated carbon; Adsorption; Ionic strength; Microorganisms

1. Introduction The progressive shortage of water suitable for drinking and the increasing water consumption by industry and agriculture pose a challenge to the technology currently used in water treatment systems. One solution is the rational and efcient administration of available resources, and wastewater reutilization is an essential part of sustainable water use. However, water treatment plants are not prepared for treating new emerging pollutants and have shown a low effectiveness in their removal. Furthermore, the new European Water Framework Directive has instituted a progressive control of these new pollutants. These compounds, including pharmaceutical products, pesticides, and surfactants, are very persistent and highly soluble pollutants.
* Corresponding author. Fax: +34 958248526.

E-mail address: jrivera@ugr.es (J. Rivera-Utrilla). 0021-9797/$ see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2007.09.039

Hence, they can easily penetrate all stages of the water cycle, presenting a potential risk to drinking water supplies. It is difcult and relatively expensive to remove these pollutants at water treatment plants. Therefore the most reasonable approach may be to address the problem at its origin by removing these compounds from wastewater and industrial efuents before they are emptied into the ecosystem [13]. Over the past few decades, indiscriminate surfactant use by the chemical industry (paper, textile) and domestically has led to the detection of these compounds in various drinking water sources [4,5]. Surfactants found in the environment include sodium dodecylbenzenesulfonate (SDBS), currently one of the most widely used anionic surfactants in detergents. Concentrations of SDBS up to 50 g/g have been measured in sediments on the coasts of Spain [6]. Although various authors have shown that SDBS is biodegradable [7], the amount of this product emptied into waters is too large to be completely elimi-

12

M.I. Bautista-Toledo et al. / Journal of Colloid and Interface Science 317 (2008) 1117

nated. Raw sewage concentrations (13 mg/L) were efciently reduced at sewage treatment plants, but leaving a sufciently high concentration in sludge to present a potential environmental problem [8]. Moreover, recent studies [9] have shown that SDBS has a low reactivity to ozone; therefore high doses of ozone are required to entirely remove this pollutant. Adsorption on activated carbon, one of the oldest water treatment technologies [10], has been recognized by the U.S. Environmental Protection Agency as one of the best methods available to remove organic and inorganic compounds from water intended for human consumption. The value of activated carbons in this eld resides in the chemical and textural properties of their surfaces [1115]. However, this technology is not very widely used to treat wastewater and industrial efuents, mainly because of their high organic matter content. Importantly, some experimental parameters of activated carbon adsorption remain unknown, and the mechanisms involved in each system have yet to be elucidated. This knowledge is essential for optimizing the use of activated carbon in water treatments. With this background, the objective of the present study was to investigate the behavior of activated carbons with different chemical and textural characteristics in the adsorption of the anionic surfactant sodium dodecylbenzenesulfonate (SDBS). The study also evaluated the combined participation of microorganisms and activated carbon in the removal of SDBS from water and the inuence on the adsorption process of the chemical nature of the solution (pH and ionic strength). 2. Experimental 2.1. Activated carbons This study used the commercial carbons Sorbo-Norit (S), Merck (M), and Ceca AC40 (C) and an activated carbon prepared in our laboratory from almond shells (A) as described elsewhere [16]. The particle size used was between 0.5 and 0.8 mm. Activated carbon samples were characterized texturally (N2 adsorption at 77 K and mercury porosimetry) and chemically (elemental analysis, determination of mineral matter components, acidbase evaluation and pH at the point of zero charge). The techniques and methods used for this characterization were previously published [17]. Results of the textural and chemical characterization of the carbons are listed in Tables 13. 2.2. Adsorption process SDBS adsorption kinetics on activated carbons were obtained by adding 0.1 g of activated carbon to Erlenmeyer asks containing 100 mL of SDBS solution at a concentration of 250 mg/L; the concentration of these solutions was determined as a function of time. SDBS adsorption isotherms were obtained by adding 0.1 g of activated carbon to Erlenmeyer asks containing 100 mL of solution with increasing SDBS concentrations (100550 mg/L). The asks were kept in a thermostatic bath at 298 K with agitation for 7 days (time needed to reach equilibrium), and the SDBS concentration in each ask was

then determined. Adsorption kinetics and isotherms were obtained at a pH of 67. The inuence of the pH on SDBS adsorption on these activated carbons was examined by adding 0.1 g of activated carbon into Erlenmeyer asks containing 100 mL of SDBS solution (250 mg/L) at different pH values (pH 213). The working pH was obtained by adding the appropriate volume of HCl (0.1 N) and NaOH (0.1 N) to the SDBS solution. The effect of electrolyte presence on SDBS adsorption on the activated carbons was studied by obtaining adsorption isotherms (see procedure above) in presence of 0.02 M NaCl. In order to determine the efciency of the combined action of activated carbon and microorganisms (bacteria) to remove SDBS, adsorption isotherms were obtained on the four carbons in presence of bacteria. A mixture of microorganisms obtained from secondary efuents at a wastewater treatment plant was used; 5 mL of this water was added to an Erlenmeyer ask containing 25 mL of Tryptic Soy Broth solution buffered at pH 7. The ask was then sterilized in autoclave and agitated at a temperature of 303 K until turbidity was observed in the solution. Following the procedure described above to determine the SDBS adsorption isotherms, 1 mL of this bacteria suspension was added to each ask, which had rst been centrifuged and repeatedly washed with distilled and sterile water. SDBS was determined by spectrophotometry at a wavelength of 223 nm, using a Genesys 5 spectrophotometer. Elemental analysis of the activated carbons was carried out in a model 1108 CHNS elemental analyzer from Fisons Instruments. Toxicity of SDBS solutions was measured with the LUMIStox system, consisting of an instrument that measure bioluminescence plus an incubation unit, in accordance with guideline UNE-EN ISO 11348-2. The measurement is based on inhibition of the luminosity intensity of marine bacteria Vibrio sheri, NRRL-B-11177, after a 15-min exposure with the toxic sample. SDBS toxicity is expressed as the percentage of bacteria inhibition, as a function of SDBS concentration. 3. Results and discussion 3.1. SDBS removal by means of activated carbon. Inuence of chemical and textural properties of the carbon 3.1.1. Adsorption kinetics Fig. 1 depicts SDBS adsorption kinetics on the activated carbons. Regardless of the carbon used, virtually all of the SDBS present was removed after 175 h of contact. SDBS was removed from the medium most rapidly by carbon M and most slowly by carbon S. In order to quantify the rate of SDBS adsorption on these carbons and identify the chemical and textural properties involved, adsorption rate constants (k a ) were determined for each activated carbon using adsorption kinetics data, which showed a perfect t to a rst-order kinetic model [18]. Very similar rate constants (k a ) were obtained for the four adsorption processes (Table 4). Thus, the highest k a value (for carbon M) was only

M.I. Bautista-Toledo et al. / Journal of Colloid and Interface Science 317 (2008) 1117

13

Table 1 Textural characteristics of activated carbons Sample C M S A

b c d e f g

S N2 a (m2 g1 ) 1201 1301 1225 1600

S ext b (m2 g1 ) 21.3 41.9 46.9 65.9

V 2c (cm3 g1 ) 0.046 0.101 0.044 0.225

V 3d (cm3 g1 ) 0.409 0.284 0.481 0.309

E0e (kJ mol1 ) 16.2 15.1 16.4 16.9

L0 f (nm) 2.24 2.94 2.14 1.96

V H2 O g (cm3 g1 ) 0.835 0.457 0.983 0.512

a S N2 Apparent surface area determined by applying BET equation to N2 adsorption isotherm.

S ext External surface area determined by mercury porosimetry. V 2 Volume of pores with diameter of 506.6 nm, determined by mercury porosimetry. V 3 Volume of pores with diameter above 50 nm, determined by mercury porosimetry. E 0 Energy of N2 adsorption from the DubininRadushkevich equation. L0 Average micropore width obtained from N2 adsorption isotherm. V H2 O Pore volume accessible to water, determined by picnometric density.

Table 2 Elemental analysis (wt%) and pH of the point of zero charge (pHPZC ) of activated carbon samples Sample C M S A C 89.0 90.5 89.7 98.5 H 0.4 0.5 0.3 0.4 N 0.6 0.7 0.2 0.7 S 0 1 0 0 O by difference 10.1 7.3 9.8 0.4 Ash 8.3 5.3 6.1 0.1 pHPZC 6.0 7.7 12.1 11.1

Table 3 Major metals present in the mineral matter of activated carbons (wt%) Sample C M S Fe 1.07 1.28 2.1 Ca 9.8 0.72 16.0 Mg 1.4 0.1 4.4 Na 0.69 0.4 0.5

Table 4 WeiszPrater coefcients (C WP ), kinetic constants of SDBS adsorption (k a ) obtained using a rst-order kinetic model, and results of applying both Langmuir and Freundlich equations to the adsorption isotherms Sample C WP k a (h1 ) Langmuir r2 C M S A
a b c d

Freundlich Xma (mg g1 ) 322.4 348.1 265.4 468.8 BXmb (L g1 ) 46.5 44.6 33.8 33.2 r2 0.9442 0.9659 0.9876 0.9811 1/nc 0.25 0.26 0.22 0.34 kd (mg g1 ) (L mg1 ) 97.68 96.16 88.74 86.22

0.016 0.020 0.011 0.022

0.0256 0.0274 0.0225 0.0232

0.9987 0.9964 0.9961 0.9904

Xm = adsorption capacity calculated by Langmuir equation. BXm = relative afnity calculated by Langmuir equation. k = specic adsorption constant calculated by Freundlich equation. 1/n = empirical parameter related to the intensity of adsorption.

22% higher than the lowest value (for carbon S). WeiszPrater coefcients [19] (C WP ) were calculated in order to determine whether k a values were inuenced by SDBS diffusion phenomena on the activated carbon surface CWP =
2 (rM )Rp

D e CM

(1)

where (r M ) is the SDBS adsorption rate (mol/L s), R p is the particle radius (0.050.08 cm), D e is the effective diffusivity (cm2 /s), and C M is the SDBS concentration (mol/L). The C WP values obtained for each carbon (Table 4) were all much lower than unity (C WP 1), indicating that the kinetic

constant values were not inuenced by the intraparticular diffusion of SDBS to access the interior of activated carbon surface pores. According to these ndings, the textural properties of activated carbons appear to have no major effect on the kinetics of SDBS adsorption. No clear relationship was observed between the k a values (Table 4) and textural properties of the activated carbons (Table 1). However, when the k a values were related to the chemical characteristics of the carbons, it was found that those with the highest k a values (carbons M and C) had the closest pHPZC values to the working pH (pH 67). These results may indicate that SDBS adsorption kinetics is potentiated when the carbon

14

M.I. Bautista-Toledo et al. / Journal of Colloid and Interface Science 317 (2008) 1117

Fig. 1. SDBS adsorption kinetics on activated carbons. pH 7, [SDBS]0 = 250 mg/L, T = 298 K, [activated carbon] = 1 g/L. (E) S, (P) M, (!) C, (1) A.

Fig. 2. SDBS adsorption isotherms on activated carbons. pH 7, T = 298 K, [activated carbon] = 1 g/L. (E) S, (P) M, (!) C, (1) A.

surface area has a zero net surface charge (pHsolution pHPZC ). = Hence, the main adsorbateadsorbent interactions involved in SDBS adsorption on activated carbon surface are nonelectrostatic. This nding is discussed in greater depth in the next section. 3.1.2. Adsorption isotherms SDBS adsorption isotherms on these activated carbons are depicted in Fig. 2. These isotherms were modeled using both Freundlich and Langmuir equations; the results obtained are given in Table 4. Langmuir equation tted much better the experimental data than Freundlich equation. The values of the parameters k f and 1/n from Freundlich equation were between 8698 ((mg/g1 ) (L/mg)1 ) and 0.22034, respectively. The values of 1/n, less than 1, indicate a favorable SDBS adsorption, regardless of activated carbon sample considered. The SDBS adsorption capacity (Xm) and relative afnity (BXm) of each carbon were determined by applying the Langmuir equation to the adsorption isotherms (see Table 4). A high adsorption capacity was found for all activated carbons, regardless of their chemical and textural characteristics, with values ranging from 265.4 (carbon S) to 468.8 mg g1 (carbon A), in-

dicating that this treatment is highly effective to remove SDBS from the medium. The carbon adsorption capacity increased in the order S < C < M < A (Table 4). When these results were related to the chemical and textural characteristics of the activated carbons (Tables 1 and 2), it was observed that the carbon with the highest adsorption capacity (carbon A) had a very high surface area (SN2 ), external surface area (S ext ), and pHPZC values and, especially, low oxygen and ash content values. In their extensive review of the literature on the adsorption of organic compounds on carbons, Radovic et al. [10] concluded that the adsorption mechanisms involved remain controversial. Their review showed that both specic and dispersive adsorbentadsorbate interactions participate in the adsorption of aromatic compounds. According to Mattson et al. [20], the adsorption of aromatic compounds on activated carbons takes place by the formation of a donoracceptor complex, with surface carbonyl groups of the activated carbon acting as electron donors and the aromatic rings of organic compounds acting as acceptors. However, Coughlin and Ezra [21] suggested that the adsorption mechanism is based on dispersive interactions between electrons of the aromatic ring of the organic compound and electrons of graphene layers of the activated carbon. In addition, Leon & Leon et al. [22] reported that basic carbons (pHPZC > 7) with a low oxygen percentage, e.g., carbon A, are characterized by a high content of electron-rich sites in graphene layers and a low concentration of surface electronattracting oxygen groups, thereby enhancing aromatic compound adsorption in accordance with the mechanism proposed by Coughlin. Results obtained (Table 4) indicate that, in general, SDBS adsorption on activated carbon is favored by a lower percentage of oxygen in the carbon, i.e., a greater hydrophobicity of the carbon. This nding conrms previously published results [23] and indicates that SDBS adsorption on activated carbons mainly takes place by nonelectrostatic interactions. This can arise, given their hydrophobic nature, between aliphatic chains of the surfactant and the carbon surface and/or by dispersive forces between electrons of the SDBS aromatic ring and activated carbon basal plane. 3.2. Inuence of solution pH on SDBS adsorption on activated carbons Fig. 3 shows the percentage of SDBS adsorption as a function of the solution pH. It can be observed that the solution pH had no major effect on the adsorption of SDBS on any of these activated carbons. These results indicate that electrostatic adsorbentadsorbate interactions do not play an important role in SDBS adsorption on activated carbon surfaces, since the progressive ionization of surface oxygenated groups with higher working pH values did not reduce SDBS adsorption on activated carbon, which suggests that repulsive electrostatic interactions between carbon surface (negatively charged at pH values above its pHPZC ) and SDBS (negatively charged at pH above 3 [pKa = 3]) were not established.

M.I. Bautista-Toledo et al. / Journal of Colloid and Interface Science 317 (2008) 1117

15

Fig. 3. Inuence of pH in SDBS adsorption processes. [SDBS]0 = 250 mg/L, T = 298 K, [activated carbon] = 1 g/L. (E) S, (P) M, (!) C, (1) A.

Fig. 4. SDBS adsorption isotherms in presence of ionic strength. pH 7, [activated carbon] = 1 g/L, T = 298 K, [NaCl] = 0.02 M. (E) S, (P) M, (!) C, (1) A.

These results corroborate the ndings discussed in the previous section, supporting the proposition that nonelectrostatic interactions are responsible for SDBS adsorption processes on these carbons. 3.3. Inuence of ionic strength on SDBS adsorption process on activated carbons The ionic strength of the solution considerably affects the adsorption of a given pollutant. Thus, Radovic et al. [10] reported that the presence of electrolytes in solution can modify the strength of adsorbateadsorbent electrostatic interactions. Both attractive and repulsive interactions can be increased or reduced by varying the ionic strength of the solution, due to a screening effect of the carbon surface charge produced by the added salt. Thus, an increase in ionic strength decreases the adsorption capacity (Xm) when electrostatic interactions between carbon surface and adsorbate are attractive but increases the adsorption capacity when they are repulsive. Fig. 4 and Table 5 show the results of SDBS adsorption on the activated carbons in presence of NaCl (0.02 M). The presence of NaCl in the solution produced an increase in the adsorption capacity (Xm) of the activated carbons, ranging from 21% (carbons C and M) to 44% (carbon A), but a decrease in their relative adsorbentadsorbate afnity (BXm). Results presented in the above sections demonstrated that electrostatic interactions between carbon surface and SDBS play no major role in the adsorption process. Hence, an increase in the ionic strength of the solution should not signicantly affect the adsorption process. However, the adsorption capacity markedly increased when NaCl was added to the system (Table 5). There are two main explanations for this behavior: (i) Decrease in SDBS solubility. It is well known that the addition of electrolytes tends to decrease the solubility of many substances in water [24] as the case of SDBS [25,26]. Thus, addition of NaCl to the system decreases solvatation of the hydrophilic part of SDBS, thereby favoring its adsorption on activated carbon.

(ii) Screening between SDBS molecules. Electrolyte addition produces a higher ion concentration near the hydrophilic part of SDBS [24], which is negatively charged, therefore the presence of Na+ produces a screening effect that reduces electrostatic repulsions among hydrophilic parts of SDBS molecules negatively charged, allowing them greater packing during the adsorption process. 3.4. Inuence of the presence of microorganisms on SDBS adsorption on activated carbons (SDBS bioadsorption) Presence of microorganisms in wastewater can have considerable effects on the effectiveness of activated carbon treatments. Fig. 5 depicts the SDBS adsorption isotherms on activated carbons in presence of microorganisms. Results obtained by applying the Langmuir equation to these isotherms are shown in Table 5. Presence of microorganisms during the SDBS adsorption process markedly increased the adsorption capacity of the activated carbons but decreased the relative adsorbateadsorbent afnity values. The presence of bacteria produced the highest Xm increase in carbon S (113%) and the lowest in carbon A (14%). Research into the effects of microorganism adsorption on the chemical and textural properties of activated carbons [27,28] have shown that microorganism adsorption (a) decreases the surface area, due to pore blockage and (b) reduces the pHPZC value, increasing the negative surface charge density of activated carbon. Furthermore, since the external walls of bacteria are formed by phospholipids [29], their adsorption on activated carbon increases the surface hydrophobicity. In previous sections, it was demonstrated that SDBS adsorption is mainly controlled by the surface chemistry of the activated carbon. Therefore, textural modications caused by microorganism adsorption should have no major effect on the SDBS adsorption process. Nevertheless, an increase in carbon surface hydrophobicity caused by microorganism adsorption might explain the increased SDBS adsorption capacity of the carbons. It should, however, be noted that the increase in carbon adsorption capacity varies considerably among carbons,

16

M.I. Bautista-Toledo et al. / Journal of Colloid and Interface Science 317 (2008) 1117

Table 5 Results obtained by applying the Langmuir equation to SDBS adsorption isotherms in presence of NaCl and microorganisms Sample C M S A NaCl Xm (mg g1 ) 389.9 420.1 330.7 675.1 BXm (L g1 ) 27.0 24.7 18.7 36.4 Microorganisms Xm (mg g1 ) 472.6 444.4 565.6 534.0 BXm (L g1 ) 17.9 50.2 12.5 29.9

Fig. 5. SDBS adsorption isotherms in presence of microorganisms. pH 7, T = 298 K, [activated carbon] = 1 g/L. (E) S, (P) M, (!) C, (1) A.

Fig. 6. SDBS toxicity as a function of its concentration. pH 7, T = 298 K.

as mentioned above, because microorganism adsorption on the carbon surface depends on its chemical and textural characteristics. Thus, a recent publication [27] showed that microorganism adsorption is favored on activated carbons with basic surface chemical properties (pHPZC > 7), since attractive electrostatic interactions are enhanced between the carbon surface (positively charged at working pH) and bacterial cell walls (negatively charged). Microorganism adsorption is also potentiated by presence of mineral matter in the carbon and by its macroporosity [27]. Accordingly, the large (113%) increase in the adsorption capacity of carbon S in presence of bacteria may be explained by the high capacity of this carbon to adsorb bacteria, which can be attributed to its high pHPZC , large macropore volume (V3 ), and high ash content. The SDBS adsorption capacity of carbon C, which has a large macropore volume and high ash content, also showed a major (46%) increase when in presence of bacteria. 3.5. SDBS toxicity SDBS toxicity was measured in order to determine maximum concentration that can be considered innocuous. Samples were prepared with increasing SDBS concentrations and their toxicity was determined as described in Section 2. Fig. 6 shows the percentage of bacteria inhibition as a function of SDBS concentration. It can be observed that bacteria were not inhibited by concentrations below 5 mg/L but were harmed at higher concentrations, and a linear increase in toxicity was found. Furthermore, experimental results showed a lineal increase of toxicity as a function of SDBS concentration. Therefore, activated

carbon adsorption is an adequate technology to remove SDBS from waters, because the concentrations of the surfactant after this treatment are not toxic. 4. Conclusions SDBS adsorption on activated carbon is not inuenced by diffusion phenomena, since C WP values were considerably below unity for all activated carbons studied. No clear relationship was observed between adsorption kinetic constants and textural properties of the carbons. Activated carbons have a high SDBS adsorption capacity, regardless of their chemical and textural properties. In general, Xm values increase with (i) larger activated carbon surface area and (ii) lower oxygen concentrations on the carbon surface. The main interactions involved in SDBS adsorption on activated carbon are (i) dispersive forces between aromatic ring electrons of the compound and basal layer electrons of the activated carbon and (ii) hydrophobic interactions between the aliphatic chain of SDBS molecules and the carbon surface. Solution pH has no major effect on SDBS adsorption on activated carbon, indicating that adsorbateadsorbent electrostatic interactions do not play an important role in this process. However, an increase in the ionic strength of the solution increases its adsorption capacity. The addition of electrolytes to the system reduces solvatation of SDBS hydrophilic part, thereby favoring its adsorption on the activated carbon. It also increases the concentration of ions around the hydrophilic part of SDBS; this produces a screening effect that reduces adsorbate adsorbate electrostatic repulsions between hydrophilic parts of SDBS molecules negatively charged, favoring the adsorption process.

M.I. Bautista-Toledo et al. / Journal of Colloid and Interface Science 317 (2008) 1117

17

The SDBS adsorption capacity of activated carbons is higher in presence of bacteria, because adsorption of the bacteria increases the hydrophobicity of the carbon surface. The greatest increase in adsorption capacity is observed in carbons with the most appropriate characteristics for adsorbing bacteria from the medium. Acknowledgments The authors are grateful for the nancial support provided by MEC-DGI, FEDER (Project: CTQ2004-07783-C02-01) and Junta de Andaluca (Project: RNM547). References
[1] S.K. Srivastava, V.K. Gupta, A. Mohan, D. Mohan, Ind. J. Chem. A34 (5) (1995) 342350. [2] S.W. Musselman, S. Chander, J. Colloid Interface Sci. 256 (1) (2002) 91 99. [3] S.W. Musselman, S. Chander, Colloids Surf. A Physicochem. Eng. Aspects 206 (13) (2002) 497513. [4] C.F. Tabor, L.B. Barber, Environ. Sci. Technol. 30 (1996) 161171. [5] E. Barbieri, I. Rocha de Oliverira, P.C.S. Serralheiro, J. Exp. Marine Biol. Ecol. 277 (2002) 109127. [6] E. Gonzlez Mazo, J.M. Quiroga, D. Sales, A. Gomez-Parra, Toxicol. Environ. Chem. 59 (14) (1997) 7787. [7] R.A. Kirmele, R.D. Swisher, Water Res. 11 (1977) 3137. [8] W. Giger, A.C. Alder, P.H. Brunner, A. Marcomimi, H. Siegrist, Tens. Surf. Det. 26 (2) (1989) 95100. [9] F.J. Beltrn, J.F. Garca-Araya, P.M. lvarez, Ind. Eng. Chem. Res. 39 (2000) 22142220. [10] L.R. Radovic, C. Moreno-Castilla, J. Rivera-Utrilla, in: L.R. Radovic (Ed.), Chemistry and Physic of Carbon, vol. 27, Dekker, New York, 2000, pp. 227403.

[11] C. Moreno-Castilla, J. Rivera-Utrilla, M.V. Lpez-Ramon, F. CarrascoMarn, Carbon 33 (1995) 845851. [12] M. Franz, H.A. Arafat, N.G. Pinto, Carbon 38 (2000) 18071819. [13] M.A. Montes-Morn, J.A. Menndez, E. Fuente, D. Surez, J. Phys. Chem. B 102 (1998) 55955601. [14] R.G. Peel, A. Benedek, Environ. Sci. Technol. 14 (1980) 6671. [15] H. Tamar, T. Yoshida, M. Sasaki, H. Yasuda, Carbon 37 (1999) 983 989. [16] C. Moreno-Castilla, M.A. Ferro-Garca, J.P. Joly, M.I. Bautista-Toledo, F. Carrasco-Marn, J. Rivera-Utrilla, Langmuir 11 (1995) 43864392. [17] J. Rivera-Utrilla, M. Snchez-Polo, Langmuir 20 (2004) 92179222. [18] O. Levenspiel, Ingeniera de las Reacciones Qumicas, eighth ed., Limusa, Mxico, 2004. [19] R.C. Reid, J.M. Prousnitz, B.E. Poiling, Properties of Gases and Liquids, fourth ed., McGrawHill, New York, 1987. [20] J.A. Mattson, H.B. Mark, M.D. Malbin, W.J. Weber Jr., J.C. Crittenden, J. Colloid Interface Sci. 31 (1969) 116125. [21] R.W. Coughlin, F.S. Ezra, Environ. Sci. Technol. 2 (1968) 291298. [22] C.A. Leon & Leon, J.M. Solar, V. Calemma, L.R. Radovic, Carbon 30 (1992) 797811. [23] P. Pendleton, S.H. Wu, A. Badalyan, J. Colloid Interface Sci. 246 (2002) 235240. [24] L.E. Scrinen, in: K. Mitta (Ed.), Micellization, Solubilization and Microemulsion, Plenum Press, New York, 1977, p. 877. [25] C.C. West, J.H. Harwell, Environ. Sci. Technol. 26 (1992) 23242330. [26] J.E. Flaming, R.C. Knox, D.A. Sabatini, T.C. Kibbey, Vadose Zone J. 2 (2003) 168176. [27] J. Rivera-Utrilla, M.I. Bautista-Toledo, M.A. Ferro-Garca, C. MorenoCastilla, J. Chem. Technol. Biotechnol. 76 (2001) 12091215. [28] J. Rivera-Utrilla, I. Bautista-Toledo, M.A. Ferro-Garca, C. Moreno-Castilla, Carbon 41 (2003) 323330. [29] H. Mikaido, H. Vaara, in: F.C. Neidhardt (Ed.), Cellular and Molecular Biology, vol. 1, American Society for Microbiology, Washington, DC, 1987, chap. 3, pp. 322.