Langmuir 2007, 23, 3723-3731
3723
Self-Aggregation of Cationic Surfactants onto Oxidized Cellulose Fibers and Coadsorption of Organic Compounds S. Alila,† F. Aloulou,† D. Beneventi,‡ and S. Boufi*,† LMSE, Faculte´ des Sciences de Sfax, BP 802-3018 SfaxsTunisie, and LGP2-UMR5518, Ecole Franc¸ aise de Papeterie et des Industries Graphiques (INPG), BP 65, F-38402 St. Martin d’He´ res, France ReceiVed October 25, 2006. In Final Form: December 29, 2006 In this work, the adsorption of cationic surfactant and organic solutes on oxidized cellulose fibers bearing different amounts of carboxylic moieties was investigated. The increase in the amount of -COOH groups on cellulose fibers by TEMPO oxidation induced a general rise in surfactant adsorption. For all tested conditions, that is, cellulose oxidation level and surfactant alkyl chain length (C12 and C16), adsorption isotherms displayed a typical three-region shape with inversion of the substrate ζ-potential which was interpreted as reflecting surfactant adsorption and aggregation (admicelles and hemimicelles) on cellulose fibers. The addition of organic solutes in surfactant/cellulose systems induced a decrease in surfactant cac on the cellulose surface thus favoring surfactant aggregation and the formation of mixed surfactant/solute assemblies. Adsorption isotherms of organic solutes on cellulose in surfactant/cellulose/ solute systems showed that solute adsorption is strictly correlated to (i) the surfactant concentration, solute adsorption increases up to the surfactant cmc, where solute partitioning between the cellulose surface and free micelles causes a drop in adsorption, and to (ii) solute solubility and functional groups. The specific shape of solutes adsorption isotherms at a fixed surfactant concentration was interpreted using a Frumkin adsorption isotherm, thus suggesting that solute uptake on cellulose fibers is a coadsorption and not a partitioning process. Results presented in this study were compared with those obtained in a previous work investigating solute adsorption in anionic surfactant/cationized cellulose systems to better understand the role of surfactant/solute interactions in the coadsorption process.
Introduction Since the beginning of the last century, surface modification of finely dispersed mineral particles by the selective adsorption of organic compounds has been extensively used in the mining industry to promote the separation of ore minerals from the gangue.1 Minerals surface modification relied on a semiempirical basis for over half a century, and a mechanistic model describing surfactant adsorption and self-assembly at solid/liquid interfaces has been proposed in more recent times in the pioneering works of Gaudin and Fuerstenau.2-4 Ever since, surfactant aggregation at solid/liquid interfaces has been the subject of intense research activities aimed at better understanding the role of surfactant, substrate, and electrolyte nature on the morphology of adsorbed aggregates. 5-7 Models generally used to describe ionic surfactants adsorption and aggregation on metal oxides with opposite charge8-10 foresee the presence of three to five distinct regions in the adsorption isotherm,3,11-14 as summarized in Figure 1. In the first region, * Corresponding author. E-mail:
[email protected]. Fax: 00 21 674 27 44 37. † Faculte ´ des Sciences de Sfax. ‡ Ecole Franc ¸ aise de Papeterie et des Industries Graphiques. (1) Fuerstenau, D. W. Centenary of Flotation Symposium; Brisbane, Australia, June 6-9, 2005; p 13. (2) Somasundaran, P.; Fuerstenau, D. W. J. Phys. Chem. 1966, 70, 90. (3) Somasundaran, P.; Fuerstenau, D. W. Trans. AIME 1972, 252, 275. (4) Gaudin, A. M.; Deker, T. G. J. Colloid Interface Sci. 1967, 24, 151. (5) Golub, T. P.; Koopal, L. K. Langmuir 1997, 13, 673. (6) Bo¨hmer, M. R.; Koopal, L. K. Langmuir 1992, 8, 2649. (7) Scamehorn, J. F.; Schechter, R. S.; Wade, W. H. J. Colloid Interface Sci. 1982, 85, 494. (8) Łajtar, L.; Narkiewicz-Michałek, J.; Rudzinski, W. Langmuir 1994, 10, 3764. (9) Li, B.; Ruckenstein, E. Langmuir, 1996, 12, 5052. (10) Drach, M.; Łajtar, L.; Narkiewicz-Michałek, J.; Rudzinski, W.; Zaja¸ c, J. Colloids Surf., A 1998, 145, 243. (11) Fuerstenau, D. W. J. Colloid Interface Sci. 2002, 256, 79.
Figure 1. Five-region adsorption isotherm of ionic surfactants on oppositely charged surfaces. The full line represents the total amount of adsorbed surfactant, dotted and thin lines represent the partitioning of surfactant molecules in the three possible adsorbed states, namely monomer, hemimicelle, and admicelle. (a) Three-region adsorption isotherm typical for a continuous transition from monolayer to bilayer aggregation.
surfactants adsorb sparsely as monomers via an ion exchange mechanism. In the second region, lateral hydrophobic interactions promote surfactant assembly, monolayered structures (hemimicelles) are formed, and a sharp increase in adsorption due to aggregates growth is observed. In the third region, monolayer (12) Koopal, L. K.; Lee, M. E.; Bo¨hmer, M. R. J. Colloid Interface Sci. 1995, 170, 85. (13) Harwell, J. H.; Hoskin, J.; Schechter, R. S.; Wade, W. H. Langmuir 1985, 1, 251. (14) Adler, J. J.; Singh, P. K.; Patist, A.; Rabinovich, Y. I.; Shah, D. O.; Moudgil, B. M. Langmuir 2000, 16, 7255.
10.1021/la063118n CCC: $37.00 © 2007 American Chemical Society Published on Web 03/03/2007
3724 Langmuir, Vol. 23, No. 7, 2007
saturation induces a decrease in surfactant adsorption. A plateau is attained in the adsorption isotherm, and the surface charge of the oxide surface is progressively neutralized. In the fourth region, dispersive forces among dissolved monomers and surface monolayered aggregates promote surfactant adsorption and the formation of bilayered structures (admicelles). The charge at the oxide surface is reversed and electrostatic repulsive forces progressively hinder surfactant adsorption until a plateau is reached in the adsorption isotherm, viz. region five. Further increase in surfactant concentration in the bulk solution leads to the formation of free micelles. Although five-region adsorption isotherms clearly show different surfactant aggregation states, most of experimental isotherms display a three-region shape (Figure 1a), thus indicating the dominant contribution of dispersive forces in promoting self-association and a continuous transition from monolayer to bilayer aggregation.15 The understanding of surfactant self-assembly at metal oxides interfaces and the ability of bilayered aggregates to act, like free micelles, as reservoirs for organic compounds opened the way to new applications in soil remediation16 and water treatment,16,17 where organic pollutants are removed by their adsolubilization/ coadsorption in surfactant aggregates formed at the surface of mineral particles (silica, clays, ...). In thin film coating18,19 polymeric films are formed at the surface of metal oxides by admicellar polymerization, and in drug carrier targeting20 the elution rate of chloroquin drug adsolubilized on zeolites is governed by the ionic strength of the surrounding aqueous medium. Recent studies21,22 showed that, even if negative electric charges at the surface of natural cellulose fibers and metal oxides have different origins, namely, ionization of discrete functional groups (-COOH) for cellulose and homogeneous adsorption of hydroxyl anions for metal oxides, surfactant adsorption models used to interpret adsorption isotherms of cationic surfactants onto metal oxides surfaces can be used for cellulose/surfactant systems. Indeed, despite their relatively low surface charge and specific surface area in the dry state (∼2 to 5 m2 g-1), cellulose fibers are able to adsorb amounts of cationic surfactants comparable with those adsorbed by finely dispersed mineral oxides23-25 giving three- to five-region adsorption isotherms. This high surfactant adsorption has been correlated with the ability of cellulose fibers to swell in water generating a microporous structure with specific surface area comparable to that of mineral particles, that is, 150200 m2 g-1.26 Furthermore, the selective oxidation of cellulose primary hydroxyl groups into carboxylic groups has been used to increase the fiber surface charge and to improve surfactant adsorption and self-aggregation onto cellulose.25 Similarities between the behavior of minerals/surfactant and cellulose/ (15) Singh, P. K.; Adler, J. J.; Rabinovich, Y. I.; Moudgil, B. M. Langmuir 2001, 17, 468. (16) Christian, S. D.; Scamehorn, J. F. Solubilization in Surfactant Aggregates; Surfactant Science Series 38; Marcel Dekker: New York, 1995. (17) Wang, Y.; Banziger, J.; Dubin, P. L.; Filippelli, G.; Nuraje, N. EnViron. Sci. Technol. 2001, 35, 2608. (18) Sakhalkar, S. S.; Hirt, D. E. Langmuir 1995, 11, 3369. (19) Pongprayoon, T.; Yanumet, N.; O’Rear, E. A. J. Colloid Interface Sci. 2002, 249, 227. (20) Hayakawa, K.; Pouri, Y.; Maeda, T.; Stake, I.; Sato, M. Colloid Polym. Sci. 2000, 278, 553. (21) Aloulou, F.; Boufi, S.; Chakchouk, M. Colloid Polym. Sci. 2004, 282, 699. (22) Aloulou, F.; Boufi, S.; Belgacem, N.; Gandini, A. Colloid Polym. Sci. 2004, 283, 344. (23) Goloub, T. P.; Koopal, L. K.; Bijsterbosch, B. H.; Sidorova, M. P. Langmuir 1996, 12, 3188. (24) Dickson, J.; O’Haver, J. Langmuir 2002, 18, 9171. (25) Alila, S.; Boufi, S.; Belgacem, M. N.; Beneventi, D. Langmuir 2005, 21, 8106. (26) Herrington, T. H.; Petzold, J. C. Cellulose 1995, 2, 83.
Alila et al.
surfactant systems have been recently used to elaborate cellulose/ polymer composites by generating a thin polystyrene layer onto cellulose fibers by admicellar polymerization27 and to coadsorb organic pollutants on cellulose fibers in the presence of cationic surfactants bearing different alkyl chain lengths.22 The high coadsorption capacity of cellulose/surfactant systems, close to that obtained with minerals/surfactant systems,21,24,28,29 put forward cellulose fibers as a low-cost and lightweight organic substrate for surfactant-based remediation technologies. The ability of natural cellulose to coadsorb organic compounds and the direct relationship existing between the net amount of adsorbed surfactant and the coadsorption capacity22 motivated this study which was aimed at better understanding the influence of cellulose oxidation on surfactant adsorption and the ensuing coadsorption of organic compounds. Materials and Methods Materials. A bleached soda pulp from the Tunisian annual plant esparto (alfa tenassissima) was used in this study. Cellulose fibers were highly porous and had a specific surface area in a dry state of 3 m2 g-1, as measured by mercury porosimetry. Since cellulose fibers swell in water, their specific surface area when dispersed in water can increase up to 300 m2 g-1 for the adsorption of molecules with molecular weight below 2000.30 Thereafter, the specific surface area of the dry state was not used to interpret adsorption isotherms. The mean fiber length and width measured by optical microscopy were 0.75 mm and 14.2 µm, respectively, thus corresponding to an aspect ratio of 52. The carboxylic content of the original fibers, determined using conductometric and polylectrolyte titration, was 45 µmol g-1. Chemicals used to oxidize cellulose fibers, namely, TEMPO (2,2,6,6-teramethyl-1-piperidinyoxy radical), sodium bromide NaBr, and a 12% sodium hypochlorite solution, were commercial products of laboratory grade used without further purification. Two analytical-grade cationic surfactants (Aldrich), namely hexadecyltrimethyl ammonium bromide (C16), and dodecyltrimethyl ammonium bromide (C12), were dissolved in distilled water to obtain high-concentration mother solutions. Their critical micelle concentration (cmc) in distilled water (25 °C) determined by conductimetry was 1.05 × 10-3 and 1.3 × 10-2 mol L-1 for C16 and C12, respectively. All organic solutes used in this study, namely, chlorobenzene, dichlorobenzene, trichlorobenzene, nitrobenzene, quinoline, and 2-naphthol, were of analytical grade commercial products. Oxidation Procedure of Cellulose Fibers. Cellulose fibers (10 g) were dispersed in distilled water (500 mL). TEMPO (25 mg) and NaBr (250 mg) were added in the suspension and the pH was adjusted to 10 by the addition of a 0.5 mol L-1 NaOH solution. A 12% sodium hypochlorite solution was then added to the suspension during 30 min, which was thereafter stirred during 2 h.25 The pH of the suspension was continuously adjusted to 10 by addition of the NaOH solution. The oxidation was stopped by adding 100 mL of ethanol, and oxidized fibers were filtered and washed two times with a 1 mol L-1 NaOH solution to remove the traces of residual lignin. The recovered fibers were washed with distilled water until the pH of the filtrate was close to 7. The content of carboxylic moieties in the oxidized fibers was evaluated by using two techniques: (i) conductometric titration of 1% fiber suspension with 10-3 mol L-1 HCl31 and (ii) colloid titration technique, using cationic 3-6 ionene (Polybrene, Aldrich), Mn ) 1500, as adsorbing polyelectrolyte, potassium polyvinyl sulfate as the anionic titrant, and orthotoluidine blue as indicator.32 (27) Boufi, S.; Gandini, A. Cellulose 2001, 8, 303. (28) Asvapathnagul, P.; Malakul, P.; O’Haver, J. J. Colloid Interface Sci. 2005, 292, 305. (29) Esumi, K. J. Colloid Interface Sci. 2001, 241, 1. (30) Wagberg, L.; O ¨ dberg, L.; Lindstro¨m, T. Colloids Interface 1987, 27, 163. (31) Katz, S.; Beatson, R. P.; Scallan, A. M. SVen. Papperdistidn. 1984, 87 (6R), 48.
Self-Aggregation of Cationic Surfactants
Langmuir, Vol. 23, No. 7, 2007 3725 Table 1. The cmc of C16 and C12 in the Presence of Organic Solutes (Csolute ) 3 × 10-3 mol L-1) as Determined by Conductimetry (Figure 2) surfactant C16
C12
Figure 2. Influence of C16 concentration on electric conductivity in the absence and in the presence of different organic solutes (Csolute) 3 × 10-3 mol L-1). T ) 40 °C.
solute no solute
T cmc solute solubility43 (°C) (mmol L-1) 25 °C (mmol L-1)
trichlorobenzene quinoline 2-naphthol
25 40 40 40 40
no solute 2-naphthol
25 25
1.05 1.65 1.40 1.15 1.15 13 10
0.22 46 5 5
a 12 h shaking to reach the coadsorption equilibrium, fibers were isolated by centrifugation and ethanol extracted. The coadsorbed amount of the organic solute was then determined by UV spectrophotometry on the extracted ethanol fraction. The surfactant concentration was determined potentiometrically by using a cationic ion-selective electrode.25,33 The cmc of C16 and C12 in the presence of a constant concentration of organic solutes, that is, 3 × 10-3 mol L-1, was determined by conductimetry and used to evaluate the contribution of tested organic solutes in promoting the formation of surfactant/solute mixed micelles in the bulk solution. The cmc measurements were performed at 40 °C to promote quinoline dissolution. ζ-Potential. In order to confirm the contribution of organic solutes to promote surfactant self-assembly in both the bulk solution and on cellulose surface, a commercial ζ-potential analyzer (Malvern 2000) was used to measure the electrophoretic mobility of cellulose fines (Cel-450) in the presence of C16 and of a fixed concentration of 2-naphthol, that is, 6.9 × 10-5 mol L-1. Measurements were conducted on the suspension-fine fraction obtained after filtration of the original suspension through a 45 µm screen.34 To avoid the fluctuation of ζ-potential due to ionic strength changes generated by the increase in the surfactant concentration, the aliquot sample was diluted with a concentrated solution of KCl to buffer the ionic strength to 10-2 mol L-1. Four consecutive measurements were taken for each sample at room temperature and averaged. ζ-Potential was calculated from electrophoretic mobility data by using the Smoluchowski equation.
Results and Discussion Figure 3. Adsorption isotherm of C16 (hexadecyltrimethylammonium bromide) on oxidized cellulose Cel-150, in the absence and in the presence of different solutes. Conductivity measurements were carried out in a thermostatic bath, (0.1 °C, using a platinum electrode. Adsorption Isotherms of Cationic Surfactants and Coadsorption of Organic Solute. Adsorption isotherms of cationic surfactants were carried out at 25 °C using the depletion method. An appropriate volume of a high-concentration mother solution, which was heated at 30 °C to ensure C16 solubilization, was added to a 1% cellulose fibers suspension to obtain surfactant concentrations ranging from 10-5 to 10-2 mol L-1. The pH of the suspension was not affected by the addition of cationic surfactant, and pH was close to 8-8.5 for all cellulose samples, independent of the oxidation level. The suspension was then stirred during 24 h at room temperature to reach adsorption equilibrium. Suspensions were then centrifuged at 2500 rpm during 15 min, and the amount of free surfactant in the supernatant was determined by potentiometric titration using a surfactant selective electrode for the dosage of C16 and C12. The titration’s protocol was described elsewhere.33 Coadsorption measurements were performed in two ways. In the first method the organic solute and the surfactant were added at the desired concentration simultaneously, and in the second method the organic solute was added after the adsorption of the surfactant. After (32) Horn, D. Prog. Colloid Interface Sci. 1978, 65, 251. (33) Ishibashi, N.; Masadome, T.; Imato, T. Anal. Sci. 1986, 2, 487.
Adsorption Isotherms of Cationic Surfactants in the Presence of Organic Solutes. For both C16 and C12 solutions, the presence of organic solutes led to the formation of mixed micelles and to a general shift of the cmc toward lower surfactant concentrations. Figure 2 shows that the larger decrease in the cmc of C16 was given by 2-naphthol and quinoline, whereas trichlorobenzene gave a negligible cmc shift. As summarized in Table 1, a similar trend was observed with the C12/2-naphthol system. In agreement with the behavior of 2-naphthol/cetylpyridinium systems,35 the observed trends were associated with the preferential interaction of 2-naphthol with the C16 ammonium head group. This phenomenon has been confirmed by viscosity measurements which revealed a significant increase in the viscosity of surfactant solution after the addition of 2-naphthol. Indeed, the viscosity at shear rate of 100 s-1 of a 5 × 10-3 mol L-1 C16 solution rises from 2 mPa‚s to 21 mPa‚s in the presence of 3 × 10-3 mol L-1 of 2-naphthol. The adsorption isotherm of C16 on oxidized fibers bearing the lowest amount of carboxylic groups, that is, 150 µmol g-1, shown in Figure 3 is characterized by a three-region shape with the onset of adsorption at 3 × 10-5 mol L-1 and a plateau value attained in the vicinity of the cmc of C16, namely 7.5 × 10-4 (34) Aloulou, F.; Boufi, S.; Beneventi, D. J. Colloid Interface Sci. 2004, 280, 350. (35) Monticone, V.; Mannebach, M. H.; Treiner, C. Langmuir 1994, 10, 2395.
3726 Langmuir, Vol. 23, No. 7, 2007
Figure 4. Adsorption isotherm of C16 on oxidized cellulose Cel450, in the absence and in the presence of different solutes.
L-1.
mol Referring to the shape of the adsorption isotherm and to our previous study,25 C16 adsorption and self-aggregation on cellulose substrates proceed via a progressive formation of more densely packed aggregates, hemimicelles and admicelles, and stop when free micelles start to form in the bulk solution. These hypothesis were confirmed by ζ-potential measurement and by the dosage of the released surfactant counterion. Results of this study25 revealed that in some cases adsorption isotherms were characterized with a five-region shape. In region I, the amount of adsorbed surfactant was low and did not exceed 30-40 µmol g-1. When the free surfactant concentration reached a critical level, a significant increase in adsorption (onset of region II) was noted. The adsorption was slowed as the adsorption attained a certain level (region III), and then when the surfactant concentration was further increased, it increased again (region IV) until it reached a constant level (region V). With the increase in the ionic strength, a progressive transition from a five- to a threeregion adsorption was observed. The steep increase in the slope of adsorption isotherms was associated to a change in the conformation of adsorbed molecules from a flat to a perpendicular orientation, where alkyl chains form highly packed aggregates, minimizing their contact area with water molecules and therefore increasing the entropy of the system. Surfactant molecules are adsorbed with head groups facing toward the surface with the formation of monolayered aggregates until charge neutralization of the surface. In the presence of C16 the adsorption is further promoted beyond this point with a generation of a bilayer domains, with surfactant headgroups facing into the solution. When compared to the system surfactant/cellulose fibers, the presence of organic solutes in the C16/cellulose-150 system induced a general shift toward lower concentrations of regions II and III, thus showing that organic solutes promote the aggregation of the surfactant molecules on the cellulose surface, probably through the formation of mixed surfactant/solute aggregates. The same phenomena account for the shift of cmc in bulk solution in the presence of an added solute. The contribution of organic solutes in promoting surfactant aggregation on cellulose is further emphasized when oxidized cellulose with a carboxylic group content of 450 µmol g-1 is used as substrate. Figure 4 shows that adsorption isotherms of C16 on cellulose-450 have a three-region shape with a clearly detectable transition between region I and region II, corresponding to the critical aggregation concentration (cac) and to the onset of the
Alila et al.
Figure 5. Adsorption isotherm of C16 on oxidized cellulose Cel600, in the absence and in the presence of different solutes. Table 2. Parametersa of the Frumkin Adsorption Isotherm Used To Interpret the Adsorption of C16 on Cellulose-450 and C12 on Cellulose-600 surfactant
soluteb
cacc (105 mol L-1)
C16
no solute tcb nb quinoline 2-naphthol no solute 2-naphthol
9.1 4.1 3.3 2.4 2.5 26 12
C12
n
a
∆Gads ln K (kJ mol-1)
3.35 9.0 2.8 3.3 9.6 3.36 9.9 3.2 10.0 3.24 10.1 2.8 2.83 8.3 2.82 9.1
-32.3 -33.7 -34.6 -34.9 -35.1 -30.6 -32.6
a See text for parameter definitions. b tcb ) trichlorobenzene; nb ) nitrobenzene. c cac ) critical aggregation concentration.
surfactant aggregation whose value is determined from the intercept of the two regions, that is, region I and II. As shown in a previous work,25 the increase in the amount of carboxylic groups promoted C16 adsorption via electrostatic interactions with a corresponding increase in the maximum amount of adsorbed C16 which was proportional to the amount of -COOH moieties carried on oxidized fibers. As observed for C16/solute/ cellulose-150 systems (Figure 3), the presence of organic solutes induced a general shift in the adsorption isotherms of C16/ cellulose-450 systems (Figure 4) with a decrease of the cac from 9 × 10-5 mol L-1 to about 6.3 × 10-5, 4.1 × 10-5, 3.3 × 10-5, and 2.5 × 10-5 mol L-1 in the presence of trichlorobenzene, nitrobenzene, 2-naphthol, and quinoline, respectively. Figure 5 shows the adsorption isotherm of C16 and C12 on cellulose-600 in the absence, and in the presence of 2-naphthol. In the presence of C12, the low amount of surfactant adsorbed on cellulose fibers and the shift of the adsorption isotherm toward high concentrations when comparing C12 to C16 was associated to the decrease of dispersive interactions between surfactant alkyl chains and therefore of their tendency to self-assembly on cellulose fibers and in the bulk solution. When adding 2-naphthol to C16/ and C12/cellulose-600 systems surfactant self-assembly on cellulose was boosted, and the adsorption isotherm of C12/2naphthol/cellulose-600 was superposed to that of C16/cellulose600. The contribution of 2-naphthol in promoting surfactant adsorption was similar to an increase of four carbon atoms in the surfactant alkyl chain. The cac of C16 and C12 on cellulose450 and cellulose-600 is given in Table 2. To further support the promoting effect of 2-naphthol on the surfactants adsorption on oxidized cellulose substrate, the
Self-Aggregation of Cationic Surfactants
Langmuir, Vol. 23, No. 7, 2007 3727
surfactant alkyl chains39 and, as observed for aliphatic alcohols/ ionic surfactant systems,40 with the onset of interactions between the solute polar group and the surfactant ammonium head group, that is, for 2-naphthol and quinoline. The comparison of the cmc*/cmc and cac*/cac ratios (where / indicates the critical concentration in the presence of solute), that is, 0.7 and 0.3 for 2-naphthol and quinoline and 0.8, 0.5 for trichlorobenzene, shows that organic solutes favor surfactant aggregation at the surface of oxidized cellulose more than in the bulk solution, thus emphasizing the role of substrate/solute and substrate/surfactant hydrophobic interactions in promoting hemimicelle and admicelle formation. Adsorption isotherms of C16 and C12 on oxidized cellulose were interpreted by using the Frumkin equation which is often used to model adsorption involving lateral interaction among adsorbate molecules41,42
θ e-2aθ ) Kcsol (1 - θ)n Figure 6. ζ-potential of cellulose Cel-450/C16 solutions in the absence and in the presence of 2-naphthol, 6.9 × 10-5 mol/L.
variation of the ζ-potential of cellulose-450 was measured as a function of the surfactant bulk concentration (Figure 6). The ζ-potential follows the same trend of the adsorption isotherm of C16 on cellulose-450, and the abrupt increase in ζ-potential was associated with the onset of surfactant self-assembly and the progressive neutralization of cellulose carboxylic moieties by the ammonium bromide surfactant head group. The reversal of the ζ-potential and the progressive formation of a plateau region are in agreement with the hypothesis of the generation of aggregates and the saturation of the cellulosic substrate. In the presence of 2-naphthol, the ζ-potential is shifted to lower surfactant concentrations and displays a quick rise at a concentration close to that corresponding to the steep raise in the adsorption isotherm, with a reverse in ζ-potential and a plateau close to 35 mV. Both adsorption and ζ-potential isotherms support the hypothesis that tested organic solutes, particularly 2-naphthol, promote surfactant self-aggregation on oxidized cellulose. As previously mentioned for free micelle formation, the different contribution of tested solutes in promoting surfactant aggregation on cellulose fibers, Figures 3-5, does not seem to be associated with the solute hydrophobic character. Indeed, the geometry and the presence of specific functional groups in the solute structure seem to be at the origin of this behavior. Two main kinds of interactions can be supposed to control surfactant/ solute aggregation: (1) a direct interaction between the polar moieties on the solute molecule and the surfactant ammonium group. This kind of interaction has been supposed to govern surfactant/solute aggregation in the presence of dyes,36 drugs,37 and 2-naphthol;35,38(2) a hydrophobic or van der Waals-type interaction between the solute hydrocarbon chains/rings and the surfactant alkyl chain. Tested solutes displayed different contributions in promoting C16 and C12 aggregation, the main shift in the cac and in the cmc was given by 2-naphthol and quinoline, followed by nitrobenzene and trichlorobenzene, respectively. This behavior agrees with the ability of aromatic rings to boost hydrophobic interactions among surfactant molecules by intercalating between (36) Esumi, K.; Sugimara, A.; Yamada, T.; Meguro, K. Colloid Surf., A 1992, 62, 249. (37) Janssen, J.; Treiner, C.; Vaution, C.; Puisieux, F. Int. J. Pharm. 1994, 103, 19. (38) Klumpp, E.; Heitman, H.; Schwuger, M. J. Colloid Surf., A 1993, 73, 93.
(1)
where θ ) qe/qmax is the surface coverage ratio, qe and qmax are the equilibrium and the maximum adsorption capacity, csol is the surfactant concentration in the bulk solution at equilibrium, a, n, and K are coefficients correlated to the lateral interaction between surfactant alkyl chains, the surfactant molecular area, and the free energy of adsorption, respectively. By using the linearized Frumkin equation,
ln
θ ) ln K + 2aθ csol(1 - θ)n
(2)
a and K can be obtained from the slope and the intercept of the plot ln(θ)/(csol(1 - θ)n) versus θ and the free energy of adsorption can be calculated as
∆Gads ) -RT ln K
(3)
Linearized adsorption isotherms of C16 and C12 in the presence and in the absence of organic solutes, Figure 7, were interpolated by using eq 2 and a n coefficient typical for the C16 surfactant,41 namely n ) 2.8. The slope of interpolating straight lines, viz. 2a, was not affected by the presence of different solutes, thus showing that the intensity of lateral interactions between surfactant alkyl chains was mainly affected by the alkyl chain length. The lateral interaction coefficient, a, was 3.3 for C16 and 2.8 for C12, respectively. Figure 7a,b shows a relevant increase in the intercept of linearized adsorption isotherms when aromatic solutes are added to surfactant/cellulose systems. This trend was interpreted as reflecting a general decrease in the free energy of adsorption when surfactant adsorption and aggregation at the surface of cellulose proceeds in the presence of solutes. Free adsorption energies calculated by using eq 3 are given in Table 2. This result further supports the hypothesis that the presence of organic solute promotes the adsorption and aggregation of C16 and C12 surfactants on oxidized cellulose fibers. Coadsorption of Organic Solutes. The second part of our study concerned the effect of the surfactant adsorption on the (39) Lee, B. H.; Christian, S. D.; Tucker, E. E.; Scamehorn, J. F. J. Phys. Chem. 1991, 95, 360. (40) Baglioni, P.; Kevan, L. J. Phys. Chem. 1987, 91, 1516. (41) Atia, A. A.; Farag, F. M.; Youssef, F. Colloids Surf., A 2006, 278, 74. (42) Stubenrauch, C.; Fainerman, V. B.; Aksenenko, E. V.; Miller, R. J. Phys. Chem. B 2005, 109, 1505. (43) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. In EnVironmental Organic Chemistry; John Wiley & Sons: New York, 1993.
3728 Langmuir, Vol. 23, No. 7, 2007
Alila et al.
Figure 9. Coadsorption isotherm of organic solutes on Cel-450 in the presence of C16. Table 3. Maximum Amount of Organic Solute Coadsorbed on Cellulose Fibers in the Presence of Increasing Surfactant Concentrations and a Fixed Solute Concentration C0 ) 1000 µmol g-1 (T ) 25 °C)
Figure 7. Frumkin adsorption isotherm of C16 on cellulose fibers in the absence and in the presence of organic solutes. (a) Oxidized cellulose 450, (b) oxidized cellulose 600.
Figure 8. Coadsorption isotherm of organic solutes on Cel-150 in the presence of C16.
coadsorption behavior of the cellulose fibers. The relationship between the equilibrium adsorbed surfactant concentration in the bulk solution and the corresponding amount of coadsorbed solute is shown in Figures 8 and 9. The comparison of these two isotherms clearly illustrates the progressive solute coadsorption or adsolubilization accompanying the corresponding increase of surfactant adsorption, up to a sharp critical value, after which the extent of solute uptake decreases dramatically. Although solute coadsorption was anticipated by a decrease in the cac of surfactant/solute/cellulose systems, the analysis of Figures 8-10 shows the following:
substrate
surfactant
2-naphthol (µmol g-1)
nba (µmol g-1)
Cel-0 Cel-150
C16 C16 C12 C16 C16
180 530 680 890 930
125 470 345 570 680
Cel-450 Cel-600 a
tcba quinoline (µmol g-1) (µmol g-1) 147 540 630 850 880
550
nb ) nitrobenzene; tcb ) trichlorobenzene.
(1) In the absence of surfactant, the solute adsorption on cellulose fibers does not exceed 80-140 µmol g-1. (2) Solute coadsorption increases steadily with surfactant concentration until a maximum is reached. This maximum is reached for a surfactant concentration close to the cmc, when free micelles start forming in the bulk solution and solute molecules partition between admicelles and free micelles. (3) The maximum amount of solute adsorbed on cellulose does not seem to be correlated with its solubility in water. Furthermore, with all tested solutes the increase in the cellulose carboxylic group content gave a general increase in the maximum amount of coadsorbed solute. This trend, observed for both C16 and C12 (Table 3), was associated with the increase of surfactant adsorption on cellulose via electrostatic interactions when increasing the surface charge of the substrate25 (Figures 3 and 4) and with the presence of a larger number of hydrophobic sites available for the solute coadsorption. Adsorption isotherms of solutes on cellulose fibers pretreated in a surfactant solution to obtain a constant surface coverage ratio, θ ) 0.75, display a typical Langmuir-type profile (Figure 10) with a steady increase of adsorption at low-solute concentration followed by the progressive saturation of the substrate. As mentioned in a previous study,34 the retention of solutes on cellulose fibers was attributed to a coadsorption process and not to the solute partitioning between surfactant aggregates and the aqueous medium (adsolubilization). Presumably, the adsorption is driven by van der Waals interactions between the solute molecules and the adsorbed surfactant molecules. Figure 10 shows that the maximum coadsorption capacity of cellulose fibers strongly depends on the amount of carboxylic groups on cellulose fibers and on the nature of the solute. Indeed,
Self-Aggregation of Cationic Surfactants
Langmuir, Vol. 23, No. 7, 2007 3729 Table 4. Maximum Amount of Organic Solute Adsorbed by Cellulose Fibers in the Presence of C16 and C12 in the Presence of a Constant Surfactant Coverage Ratio (θ ) 0.75)
Figure 10. Adsorption isotherms of organic solutes on cellulose fibers in the presence of C16 (coverage ratio θ ≈ 75%). (a) Not oxidized cellulose Cel-0, (b) oxidized cellulose Cel-450, oxidized cellulose Cel-600.
for a series of homologous solutes, namely chloro-, dichloro-, and trichlorobenzene, the decrease in water solubility due to the increase in the number of chloro moieties can be directly correlated with the maximum amount of adsorbed solute (Table 4). However, when considering nitrobenzene, whose solubility is 4 times higher than that of chlorobenzene, the adsorption of nitrobenzene remains favored thus indicating that the nitro group participates to the coadsorption process. The role of polar groups in promoting the solute coadsorption is further emphasized by 2-naphthol which, among tested solutes, gave the highest coadsorption on cellulose fibers even displaying a solubility 10 times higher than that of dichlorobenzene (Table 4). In a previous work aimed at better understanding coadsorption mechanisms in anionic surfactant/cationized cellulose/organic solutes systems34 chlorobenzene homologues and nitrobenzene
displayed an opposite behavior. Indeed, the increase in the number of chloro moieties or even the presence of a nitro group gave a decrease in the solute coadsorption capacity, whereas, in line with results presented in this study, 2-naphthol gave the highest coasdorption. The different coadsorption behavior observed in the presence of surfactants bearing a sulfate and a trimethylammonium bromide head group was ascribed to the specific interaction between surfactants and solute functional groups: polar chloro and nitro groups are supposed to affect solute interaction with surfactant aggregates via attractive and repulsive dipole/charge interactions with the surfactant ammonium and sulfate moieties, respectively. 2-Naphthol displayed the highest coadsorption in the presence of cationic surfactants. This behavior was associated with the asymmetric structure generated by the presence of the polar hydroxyl and the hydrophobic-planar naphthalene moiety. 2-Naphthol is therefore supposed to coadsorb on admicelles with the naphthalene moiety intercalated among surfactant alkyl chains and the hydroxyl group oriented toward water, whereas, chlorobenzene homologues are supposed to coadsorb with a flat configuration thus maximizing the surface area occupied per molecule which can be roughly estimated as the double of a 2-naphthol molecule adsorbed perpendicularly to the surface. Proposed solute coadsorption mechanisms are summarized in Figure 11. Results obtained with C16 and C12 show that, although promoting surfactant aggregation by improving lateral interactions among surfactant molecules (Table 2), the length of the surfactant alkyl chain seems to play a secondary role in the coadsorption of organic solutes. At both the saturation of cellulose surface (Table 3) and at a fixed surfactant coverage ratio (Table 4) the maximum amount of coadsorbed solute is slightly affected by
3730 Langmuir, Vol. 23, No. 7, 2007
Alila et al.
Figure 11. Schematic representation of solute coadsorption on cellulose in the presence of surfactants. (a) Coadsorption of 2-naphthol on anionic/cationized cellulose in the presence of hemimicelles generated by a surfactant with opposite charge. Naphthalene moieties assume an orientation parallel to surfactant alkyl chain with a dense packing density. (b) Coadsorption of 2-naphthol in the presence of admicelles, naphthalene moieties are intercalated between surfactant alkyl chain with hydroxyl groups pointing toward water. This configuration is assumed to be favored by both the flat shape of naphthalene and interactions between hydroxyls and the ionic surfactant head groups (i.e., trimetylammonium bromide and sulfate). (c) Coadsorption of chlorobenzene homologues on cationic surfactant hemimicelles formed on anionic cellulose. The aromatic ring lies flat on the admicelle hydrophobic core thus maximizing its surface area, low packing density. (d) Coadsorption of chlorobenzene homologues in admicelles. Solute can form bilayers and/or assume a configuration with the aromatic ring intercalated between surfactant alkyl chains and the polar chloro group pointing toward the surfactant cationic head groups. This orientation is favored by charge/dipole interactions between trimethylammonium and chloro moieties. (e) Coadsorption of chlorobenzene homologues on anionic surfactant hemimicelles formed on cationized cellulose. The aromatic ring lies flat on the admicelle hydrophobic core thus maximizing its surface area, low packing density. (f) Coadsorption of chlorobenzene homologues in admicelles. Solute can form bilayers parallel to the substrate surface but it is not supposed to align parallel to the surfactant alkyl chain since charge/dipole interactions between sulfate and chloro moieties do not favor this configuration.
the surfactant alkyl chain length, and it is not possible to identify a general trend. Experimental results show that the increase in the surfactant alkyl chain length promotes the formation of surfactant aggregates on the cellulose surface (i.e., by shifting the cac toward lowsurfactant concentrations) with the aggregate hydrocarbon core acting as hydrophobic adsorption site for organic solutes. However, at both saturation of the cellulose surface and at a fixed surfactant coverage ratio, C16 and C12 aggregates displayed almost the same solute coadsorption capacity. In other words, the alkyl chain length affects interactions between surfactant molecules during the aggregation process but not between aggregates and solutes close to the maximum coadsorption capacity. To better understand the mechanism of organic solute coadsorption, laser induced luminescence using benzophenone or 2-naphthol as a probe will be used to investigate the structure of surfactant/solute aggregates. This work is under progress.
Conclusions The present investigation has shown the following: (i) Surfactant aggregation on cellulose surface is strongly affected by the amount of carboxylic group (i.e., oxidation degree) on the fiber surface. The increase of the -COOH surface
concentration is associated with a corresponding increase in the maximum amount of adsorbed surfactant. As expected from micellization in solution, the decrease in the surfactant alkyl chain length induced a decrease in the free energy of aggregation at the water/cellulose interface and an ensuing increase in the cac of surfactant/cellulose systems. (ii) The addition of organic solutes to surfactant/cellulose systems induced a general increase in the free energy of adsorption corresponding to a decrease in the cac. The modest ∆Gads lead to suppose that this effect is mainly due to dispersive interactions between solute molecules and surfactant alky chains; however, the different behavior observed for chlorobenzene homologues and 2-naphthol shows that chloro and hydroxyl group interactions with the surfactant ionic head group cannot be neglected when considering solute/surfactant interactions. (iii) The formation of surfactant aggregates on the surface of cellulose fibers boosts solute uptake by cellulose. The shape of solute adsorption isotherms for a fixed surfactant surface coverage ratio led to the supposition that solute uptake proceeds via a coadsorption process. (iv) The comparison of results presented in this work with a previous work focusing on solute coadsorption on cationized cellulose in the presence of anionic surfactant shows that
Self-Aggregation of Cationic Surfactants
solute coadsorption is a complex process governed by the substrate surface charge (i.e., amount of carboxylic groups on cellulose), solute solubility, polarity/charge of solute/surfactant functional groups, and geometry of the solute molecule. A simple model was proposed to illustrate coadsorption mechanisms; however, further investigation is needed to explain this complex process.
Langmuir, Vol. 23, No. 7, 2007 3731
Acknowledgment. This work was financially supported by the Tunisian “Ministe`re de la Recherche Scientifique, technologique et de developpement des compe´tences” and by the International Foundation for Science through Grant W/3358-2. LA063118N
