JOURNAL OF COLLOID AND INTERFACE SCIENCE ARTICLE NO.
205, 97–105 (1998)
Adsorption of Polyoxyethylenic Nonionic and Anionic Surfactants from Aqueous Solution: Effects Induced by the Addition of NaCl and CaCl2 D. M. Nevskaia,1 A. Guerrero-Ruı´z, and J. de D. Lo´pez-Gonza´lez Departamento de Quı´mica Inorga´nica y Quı´mica Te´cnica, Facultad de Ciencias, Universidad Nacional de Educacio´n a Distancia (U.N.E.D.), c/Senda del Rey s/n, Madrid 28040, Spain Received October 17, 1997; accepted May 1, 1998
important in their adsorption processes by solids in water solutions, and that the presence of salinity in those solutions causes modifications of the adsorption capacity of surfactants in a solid–liquid interface. For instance, Denoyel et al. (1) found that the presence of NaCl shifts the “plateau” position of TX-100 adsorbed on quartz toward lower equilibrium concentrations, which means that there is a diminution of the critical micelle concentration (c.m.c.). At the same time, these authors observed a rise in adsorption at the “plateau.” They attributed this behavior to an increase in lateral interactions between polar chains, when salinity increases. However, Partyka et al. (2), in a study of DDTAB (C12H13N1(CH3)3Br2) adsorption isotherms on silica, in the presence of NaBr, observed that the adsorbed amount decreases when salt concentration is increased. This fact was explained by competitive adsorption between surfactant and Na1 ions on the silica surface. Davies et al. (3) proposed the same explanation for PLL (poly-Llysine) adsorption on mica when NaCl is added. On the other hand, the presence of inorganic salts affects the surfactant behavior. For instance, displacements of the c.m.c. of surfactants when a variety of inorganic salts are added to the water, in connection with “salting out” (micelle dehydration) and “salting in” effects, have been observed (4). The salting out effect is produced when the salt, competing with the surfactant for hydration water, reduces the amount of water available in the micelles for polar chain hydration. Thus, micelle formation will be produced at lower surfactant concentration. The salting in effect (micelle hydration) is produced because the salt ions “break” the water structure (intramolecular interactions), making the water molecules more accessible for hydration of the surfactant molecules (5). The salt addition also affects the solids in suspension and the specific adsorption of salt ions on a solid surface or flocculation processes may result from it. When the two phases came in contact, a potential difference is established and the majority of the most important properties of such a colloidal system can be determined, directly or indirectly, by surface charge or by potential differences between the particles. The electrical double layer plays a very
The adsorption processes of two different types of surfactant from aqueous solutions have been studied on several solids. The adsorbates used were a nonionic (TX-100) and a series of anionic (NP4S, NP10S, and NP25S) oxyethylenic surfactants. As adsorbent, five nonporous solids, including three quartz (QA, QB, and QC), a kaolin, and a dolomite, were chosen for this study, since these types of materials are frequently found in oil reservoirs. Great differences have been found in the adsorption plateaus, depending on the nature of the surfactant (anionic or nonionic). The influence of the presence of NaCl and CaCl2 in the solutions has been also studied. NaCl affects the adsorption of anionic surfactants on quartz and kaolin samples in a similar way. When this salt is added, the amount of anionic surfactants adsorbed on the solid surfaces increases. Some differences in the adsorption of the TX-100 surfactant, depending on the nature of the surface and the type of salt added, have been detected. Basically, three different adsorption behaviors have been found when NaCl is added. The amounts of TX-100 adsorbed decrease when NaCl is added to the solution on the QA sample; the amounts increase on the QB and kaolin samples; no alteration is observed on QC and dolomite samples. Changes in adsorption isotherms, depending on whether NaCl or CaCl2 is added, have also been observed. For the same five adsorbents, zeta potential measurements also have been carried out. When the nonionic TX-100 surfactant is adsorbed, a decrease in the negative values of the zeta potential takes place. However, for the adsorption of anionic surfactants, an increase in the negative values of the zeta potential is detected. The surface charge has been also determined by potentiometric titration (in presence and in absence of TX-100), and a decrease in surface charge when TX-100 is adsorbed on the sample surfaces has been detected. © 1998 Academic Press Key Words: adsorption; polyoxyethylenic surfactants; salt effect; surface charge.
It is well know that the nature and electric charge of polyoxyethylenic surfactants (nonionic and anionic) are very 1
To whom correspondence should be addressed. E-mail: [email protected]
0021-9797/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
´ PEZ-GONZA ´ LEZ NEVSKAIA, GUERRERO-RUI´Z, AND LO
important role in fields of colloidal chemistry such as micelle formation, stability, electroadsorption, flotation, polyelectrolyte properties, and electrokinetic studies (6 – 8). Thus far, in an oxide surface (like quartz) amphoteric sites can be charged, positively or negatively, depending on the pH values (9 –12). Some surface processes producing these charges are 1 AS 2 OH1 2 ↔ AS 2 OH 1 H
AS 2 OH ↔ SO2 1 H1,
where AS is a surface site. In the case of aluminosilicates such as kaolin, the formation of a double layer is controlled by the processes related to the SiO2 and Al2O3 skeletons, as though it were a mixed oxide. Thus, in addition to processes  and  the following ones can take place (13, 14): 2 AlOH1 2 1 OH 3 AlOH 1 H2O
AlOH 1 OH2 3 AlO2 1 H2O.
Moreover, it is generally thought that kaolin has two different charge regimes (15). There are negative face charges which are originated by the substitution of higher valency cations for others of lower valency (e.g., Fe21 for Al31 or Al31 for Si41) in the crystal structure of the lattices. These types of charge are practically invariant with pH or electrolyte concentration. The other types of charge occur on kaolin edges. It seems that these charges are positive in acid or neutral solutions (15). The edge charges are probably due to hydroxyl groups (16) which can also play an important role in adsorption processes. The aim of this work is to study the differences in the adsorption behavior of some anionic and nonionic surfactants on different solids (quartz, kaolin, and dolomite), and to point out how the presence of salinity (NaCl and CaCl2) can affect each adsorption system. The mentioned solids have been chosen because of their different surface properties and because these types of materials are frequently found in oil reservoirs. The influence of the surface nature has been also studied. We have also attempted to correlate the electrokinetic behavior of such surface samples with their adsorption behavior. EXPERIMENTAL
The solid samples used were a natural quartz (Sifraco C-600, 98.8% quartz) which is denoted as QA. This sample (QA) was treated with a HCl solution (pH 1.5) and subsequently washed with distilled water until the total elimination of chloride ions had occurred, as detected with an AgNO3 solution. Afterward, it was dried in a vacuum oven at 393 K for a few hours and denoted as QB. Sample QC was obtained by calcination of sample QB in an oven at 1273 K for 9 h. A kaolin from Sigma
and a natural dolomite from Granada (Spain) are also used as adsorbents (17). The surfactants used were commercial ones, a nonionic TX-100 (C8H17-Ph-(OCH2-CH2)n-OH from Rohm and Haas Co. (n 5 9.5) with a degree of purity .98%, and a series of anionic surfactants with a degree of purity .96%, and a general formula 1 C9H19-Ph-(OCH2-CH2)n-SO2 4 Na ,
where n 5 4 for NP4S (from Stepan Europe), n 5 10 for NP10S (from CFPI), and n 5 25 for NP25S (from Witco Chemicals). The specific surface areas of the solids were determined by nitrogen adsorption at 77 K, taking 0.162 nm2 for the crosssectional area of the nitrogen-adsorbed molecule, and using the B.E.T. method. An automatic Micromeritics ASAP 2000 volumetric system was used to obtain the corresponding gas adsorption isotherms. The values of the specific surface areas so obtained are 5.1 for QA, 5.5 for QB, 4.0 for QC, 19.3 for Kaolin, and ;0.3 m2 g21 for dolomite (17) with an accuracy $0.1 m2 g21. The surfactant adsorption isotherms from aqueous solutions were obtained by the immersion method (18). A few grams of each solid were inmersed in surfactant water solutions (20 cm3) of known concentrations. Equilibrium was reached after 24 h with ellipsoidal stirring at 298 K; the stirring rate was 100 r.p.m. After equilibrium the solids were separated by centrifugation, at the same temperature (298 K) and 3000 g. The amount adsorbed at 298 K for each equilibrium point was calculated from the changes in concentration of the solutions, determined by U.V. spectroscopy at 275 nm. The experimental error was less than 5%. In order to study the salt effect, 1, 10, and 20 g/liter of NaCl and 1 and 10 g/liter of CaCl2 were added to the surfactant solutions. Details about this experimental method are given elsewhere (17). Zeta potential measurements were carried out in a ZetaMeter 3.01 at 298 K. In our case, x z a @ 1 (x is the reciprocal of Debye length and a is the particle radius) so, to obtain the zeta potential (z), Smoluchowski’s equation was applied (19),
m 5 e reoz/h,
where m is the electrophoretic mobility, er is the dielectric constant of the surrounding fluid, eo is the permittivity of free space, and h is the fluid viscosity. The amount of sample used in each point was approximately 80 ppm. To obtain each point, ten measurements have been made. The pH was adjusted with HCl, NaOH, and Ca(OH)2. Solutions were stabilized 24 h before measurements. The accuracy was ;5%. The surface charge of the samples was obtained by potentiometric titration (716 DMS Titrino of Methrom) following the Van Raij and Peech method (20). Amounts from 0.8 to 3 g
ADSORPTION OF NONIONIC AND ANIONIC SURFACTANTS
FIG. 1. Adsorption isotherms of TX-100 at 298 K on the QA sample with different NaCl concentrations.
(Fig. 1). In the second case, for QB (Table 1) and for kaolin (Fig. 3), the behavior is similar; that is, when the amount of NaCl added is increased up to 20 g/liter the adsorbed amounts of TX-100 are greater (the differences between the two samples will be discussed later). Finally, in the third case, the adsorbed amount of TX-100 on QC and dolomite (Fig. 2 and Table 1) seems not to be practically affected by the increase in NaCl concentration. In the presence of CaCl2 in the water solutions (1 and 10 g/liter) the adsorption capacities of QB, QC, and kaolin present practically the same behavior as that in NaCl (Table 1). In the case of QA and the dolomite samples, the behavior is different in CaCl2 than in NaCl. As has been said above, the adsorbed amount of TX-100 on QA sample decreased when the concentration of NaCl increased; however, when 10 g/liter of CaCl2 is added, a slight increase in the amount adsorbed is observed. In the case of dolomite, when 1 g/liter of CaCl2 is added, the amount of adsorbed TX-100 decreases, and when the amount of CaCl2 is increased to 10 g/liter, the adsorbed amount of TX-100 also rises. The adsorption isotherms in aqueous solutions of the anionic surfactants NP4S, NP10S, and NP25S show that kaolin adsorbs about nine times more anionic surfactants than QB (Table 2), while QB adsorbs twice more TX-100 than kaolin (Table 1). On the other hand, all the anionic surfactants of the NPS series have shown similar behavior on QB and kaolin samples in the
of solid samples were put in contact with 50 cm3 of TX-100 solutions in a closed vessel. Blanks (without TX-100) were also prepared. Afterward, the pH of the samples and the blanks was fixed by addition of acid or base. Suspensions were stabilized 24 h before pH measurements. The amount of acid or base necessary (for samples and blanks) to obtain the pH fixed initially was determined by potentiometric titration. Under these conditions, the surface charge of a sample will be the amount of acid (base) spent minus the amount of acid (base) spent by its corresponding blank. The accuracy was about 2%. RESULTS
Adsorption Isotherms In Figs. 1 to 3, some adsorption isotherms of TX-100 at 298 K on QA, QC, and kaolin samples in presence of 1, 10, and 20 g/liter of NaCl, respectively, are shown as representative of our total experimental results. As it may be observed, all isotherms are S-shaped. The quantity adsorbed corresponding to the “plateau” of the isotherms in the presence of different amounts of NaCl and CaCl2, for all samples studied, is given in Table 1. Basically, three different behaviors have been found when NaCl is added to the TX-100 surfactant solution. First, in the case of the QA sample, a decrease of adsorbed amounts of TX-100 is observed when the amount of NaCl added increases
FIG. 2. Adsorption isotherms of TX-100 at 298 K on the QC sample with different NaCl concentrations.
´ PEZ-GONZA ´ LEZ NEVSKAIA, GUERRERO-RUI´Z, AND LO
sense that when the polar chain of the surfactant increases the adsorbed amount decreases from the first to the last term by a factor of about 2. Comparing the adsorbed amount of TX-100 with the corresponding adsorbed amounts of the anionic surfactant NP10S (which has the same oxyethylenic chain length), it can be seen that TX-100 is adsorbed on the QB sample about 32 times more than NP10S, while on kaolin the adsorbed amounts of TX-100 and NP10S are practically the same. The addition of NaCl (1g/liter of NaCl in our case) to the solution originates an increase in the adsorbed amounts of NP4S, NP10S, and NP25S (Table 2) on QB as well as on kaolin samples. This effect is smaller as the polar chain length increases, but higher for the QB than for the kaolin sample. Zeta Potential Measurements In order to ascertain the mode of interaction of NaCl and CaCl2 with the surfaces, zero point charges (z.p.c.) of the solids were determined at 298 K (Tables 3 and 4). When the NaCl salt concentration is increased, z.p.c. shifts toward higher pH values (Table 3). All the samples, except the QA quartz, show similar z.p.c. displacement when CaCl2 (Table 4) is added. For the QA sample, a z.p.c. shift is not detected. Figure 4 represents zeta potential, z, vs concentration of NaCl for the five solids. As may be observed, when NaCl concentration is increased, a change in the zeta potential values (from negative to positive) is originated in all cases. The same behavior is observed in the presence of CaCl2, except for with QA quartz, in which z remains constant whithin the interval of concentrations used. In Fig. 5 the adsorption isotherm and the zeta potential of the QB sample in the presence of 1 g/liter of CaCl2 are represented vs Ceq of TX-100. Similar patterns were obtained for QA, QC, kaolin, and dolomite samples (in NaCl and CaCl2), which are not reported here for the sake of brevity. When TX-100 is adsorbed on the QB surface, the zeta potential become less negative at increasing TX-100 concentrations (21, 22). Besides that, the evolution of the zeta potential curve runs practically parallel to the adsorption isotherm. On the other hand, when an anionic surfactant is adsorbed, the behavior is the opposite (Fig. 6), and the zeta potential negative values increase with the amount adsorbed.
FIG. 3. Adsorption isotherms of TX-100 at 298 K on the kaolin sample with different NaCl concentrations.
Surface Charge In Table 5 the surface charge of solids, measured in the presence and absence of TX-100, are reported. The pH at which these surface charges were determined is the pH of equilibrium when solids and water or surfactant solutions were put in contact (pH 8.4 for QA, pH 5.4 for QB, pH 7.0 for QC, pH 5.5 for kaolin, and pH 9.7 for dolomite) (23). Surface charge in the presence of TX-100 was obtained at the adsorption “plateau.” In general, when TX-100 is adsorbed on solid surfaces a decrease in the surface charge is observed.
TABLE 1 Adsorbed Amounts of TX-100 (mmol/m2) at the Isotherm’s “Plateau” TX-100 1 NaCl
TX-100 1 CaCl2
QA QB QC Kaolin Dolomite
0.72 1.60 2.71 0.66 0.42
0.52 1.52 2.31 0.80 0.35
0.19 1.75 2.48 0.81 0.45
0.21 2.11 2.81 0.88 0.48
0.73 1.77 2.47 0.81 0.27
0.85 1.78 2.63 0.86 0.68
ADSORPTION OF NONIONIC AND ANIONIC SURFACTANTS
TABLE 2 Adsorbed Amounts of “NPS” (mmol/m2) at the Isotherm’s “Plateau” NP4S 1 NaCl
NP10S 1 NaCl
NP25S 1 NaCl
As has been reported earlier, from the XPS examination (17), the sample QA contains 4% Al and 0.2% Na on its surface, and the O/Si ratio is very high (near four). These facts indicate that some kind of aluminum hydroxide is present on the silica surface. So, it could be postulated that there are two different types of OH group, one associated with the silica surface and the other with aluminum hydroxide. Furthermore, the amount of OH groups per nm2 is very high (41.9 OH nm22) (17) compared with the accepted value of approximately 5 Si-OH nm22 (25). This density of OH groups on the silica surface is so high that the OH groups can be present in a three-dimensional layer (26). These OH groups are mainly hydrogen-bound, as has been determined by IR (17). The Na1 (from added NaCl salt) ions are specifically adsorbed (discussed later) on the QA surface. It is probable that the Na1 ions interact with AlOOH groups so strongly that
TX-100 is not able to displace them. Thus, when the amount of NaCl added to surfactant solution increases (from 1 to 10 g/liter), more and more reactive surface sites are occupied by Na1, and less TX-100 is adsorbed. The degree of coverage of the QA sample by TX-100 falls from 31% (17) to about 9%. On the other hand, it can be observed (from Table 1) that there is no essential difference between the adsorbed amounts of TX-100 when 10 or 20 g/liter of NaCl is added. This seems to indicate that a saturation of surface adsorption sites is produced when 10 g/liter of NaCl is added. When 1 g/liter of CaCl2 is added (the Ca21 ions seem to be indifferent ions for the QA sample) variation of adsorbed TX-100 is not detected. However, when 10 g/liter of CaCl2 is added, the adsorbed amount increases (Table 1). This is probably due to the fact that the amount of 10 g/liter of CaCl2 comes to be enough to screen considerably the high surface charge of the QA sample (Table 5), allowing a greater amount of TX-100 to be adsorbed. After treatment with HCl, the surface of QB does not contain aluminum oxide (17), and the amount of OH groups per nm2 diminishes a little, down to 35.2 OH nm22, though not enough. Thus, in the QB is still present a gel layer on its surface, and by IR studies it has been determined that hydroxyl groups are mainly hydrogen bound (17). The addition of 1 or 10 g/liter of NaCl or CaCl2 salts does not affect the adsorption behavior (the differences observed are within the experimental error). However, when 20 g/liter of NaCl are added, the adsorbed amount of TX-100 increases (Table 1). The degree of coverage of the QB sample by TX-100 rises from 50% (17) to about 65%. This behavior is the opposite of that of the QA sample. Since on the QB surface there is no aluminum hydroxide, the
TABLE 3 Z.P.C. of Solids in the Presence of NaCl
TABLE 4 Z.P.C. of Solids in the Presence of CaCl2
In general, it is accepted that adsorption of polyoxyethylenic surfactants on the type of solid surfaces that we have studied is produced by hydrogen bonds between ether groups of oxyethylenic chains and the surface hydroxyl groups, specifically the isolated hydroxyl (17, 24). The results obtained from adsorption isotherm measurements (Fig. 1–3) clearly show that adsorption behavior in the presence of salts depends not only on the interaction between the salt and the surfactant but fundamentally on the nature of each specific solid surface. Effect of NaCl and CaCl2 Concentration on the TX-100 Adsorption
Z.P.C. in NaCl solutions
Z.P.C. in CaCl2 solutions
QA QB QC Kaolin Dolomite
2.17 2.05 2.29 3.04 12.25
2.32 2.16 2.46 3.19 12.38
2.44 2.35 2.61 3.46 12.53
2.53 2.47 3.35 3.55 12.66
QA QB QC Kaolin Dolomite
2.41 2.25 2.87 2.93 11.80
2.41 2.65 3.07 3.66 12.22
2.41 2.83 3.45 4.21 12.31
´ PEZ-GONZA ´ LEZ NEVSKAIA, GUERRERO-RUI´Z, AND LO
FIG. 4. Zeta potential vs NaCl concentration at 298 K.
Na1 ions interact weakly with the silica hydroxyl groups, allowing the surfactant to displace them from the QB reactive sites. The amount of 20 g/liter of NaCl is enough to screen at least part of the surface charge, and this increases the adsorbed amount of TX-100. The QC sample, due to its thermal treatment, has only isolated hydroxyl groups, and its amount of OH groups per nm2 is 3.3 (17). As has been said above, the oxyethylenic surfactants form preferentially hydrogen bonds with isolated hydroxyl groups. This explains the much higher adsorbed amount of TX-100 on the QC sample (Table 1). The addition of NaCl or CaCl2 does not have any effect on the adsorbed amount of TX-100 (the differences observed are within the experimental error). This is probably due to the fact that TX-100 (without salt) covers completely the QC surface (27) and the salt action is not strong enough to modify the situation. As was mentioned above, the kaolin has two distinct sources of charge (the basal planes and the edges). It is generally considered that the edges are more reactive sites than the basal plane surfaces (28). Furthermore, it is accepted that the edge charges are positive in acid or neutral solutions. In our case, a pH 5 for solid–surfactant equilibrium solutions has been obtained. This signifies that the edges of our sample should be positively charged. However the Na1 from the added NaCl or the Ca21 from the added CaCl2, which is specifically adsorbed on the kaolin surface (discussed later), produces an increase of TX-100 adsorption (Table 1). The degree of coverage of kaolin by TX-100 rises from 25% (17) to about 33%. This seems to indicate that surfactant is adsorbed on the basal plane and the
Na1 and Ca21 ions simply screen the surface charge facilitating the TX-100 adsorption. It can also be observed that the rise in the salt concentration seems not to produce a parallel increase in the adsorbed amount of surfactant; it is as if there is a maximum of screening that can be reached. In the case of dolomite, the influence of NaCl salt is not observed (the differences observed are within the experimental error, Table 1). It seems that the Na1 are displaced from the dolomite surface by TX-100. However, when CaCl2 is added, the adsorption behavior changes. When 1 g/liter of CaCl2 is added, a competition between Ca21 ions and TX-100 molecules by adsorption sites can be operating, diminishing the adsorbed amount of surfactant. On the other hand, when 10 g/liter of CaCl2 is added the opposite effect is observed, and an increase in the adsorbed amount is detected. We have not found, yet, any logical explanation for this effect. The differences observed in the adsorption behavior of QA, QB, kaolin, and dolomite samples with the concentration of the added salts suggest that the driving force for the adsorption of TX-100 on these surfaces is essentially of an electrostatic nature. In the case of TX-100 adsorption on the QC sample, where no salt effect is observed, the interaction between surfactant and surface should be attributed to hydrogen bonds.
FIG. 5. Adsorption isotherm of (TX-100 1 1g/liter CaCl2) at 298 K on the QB sample and the corresponding zeta potential.
ADSORPTION OF NONIONIC AND ANIONIC SURFACTANTS
anionic surfactants are adsorbed on the basal planes. And thus, the compensation of negative surface charge of the basal planes by Na1 addition increases the adsorbed amounts of anionic surfactants. Interactions between Aqueous Solutions of NaCl and CaCl2 and the Solid Surfaces In Figure 4, the zeta potentials, z, vs NaCl concentration are represented. A change of sign is observed in all cases when the NaCl concentration reaches values from 0.4 to 3.5 1021 mol/ liter. This means that a specific adsorption of the salt on the surfaces takes place (30). When the NaCl salt concentration increases, a displacement of z.p.c. toward higher pH values indicates that this kind of adsorption is of a “chemical” nature or, in other words, salt ions are chemisorbed on the solid surface. Moreover, the fact that the z.p.c. shifts to basic values of pH signifies that Na1 ions are chemisorbed (Table 3). Exactly the same behavior is observed for CaCl2 (Table 4), except for with the QA sample; for this sample, a shift of z.p.c. with salt concentration is not detected. This implies that, in this case, CaCl2 seems to be an indifferent electrolyte for QA quartz in our experimental conditions. Zeta Potential Measurements FIG. 6. Adsorption isotherm of (NP25S 1 1g/liter NaCl) at 298 K on kaolin, and the corresponding zeta potential.
Effect of NaCl on the Adsorption of Anionic Surfactants The adsorption data corresponding to the adsorption isotherms of the series of anionic surfactants on QB and kaolin are summarized in Table 2. One can observe that when oxyethylenic chain increases, the adsorbed amount of anionic surfactant decreases, on both samples (QB and kaolin), due to the larger size of the corresponding molecules (1). In general, the surface of these samples, at equilibrium pH, is negatively charged (Table 5). The anionic surfactants (NP4S, NP10S, and NP25S) also have a negative charge and, consequently, their adsorbed amounts per square meter of surface on quartz QB are very low in comparison with the corresponding TX-100. Addition of NaCl originates an increase in the adsorbed amounts of anionic surfactants on the quartz surface (Table 2). This could be due to a compensation of the negative surface charge by the Na1 ions. On the other hand, adsorption on kaolin is nine times greater (Table 2) than on QB. This behavior could be attributed to some positive charges on the kaolin edges (29), where anionic surfactants could be adsorbed. When Na1 ions are added to the solution, the adsorbed amounts of anionic surfactants rise. Since an increasing in the interlaminar distance of kaolin when Na1 ions were added was not detected by X-ray, it seems that openings of kaolin layers are not produced. Thus, the increase in adsorbed amounts of anionic surfactants when NaCl is added seems to indicate that an additional part of the
In Figure 5, where the adsorption isotherm and the corresponding zeta potential on QB vs Ceq of TX-100 in the presence of 1 g/liter of NaCl are represented; a decrease in the absolute values of the negative zeta potential can be observed when TX-100 is adsorbed on the QB surface. This can be attributed to a shift of the share plane toward bulk solution, as a consequence of formation of a surfactant layer on the surface (31) or a decrease of surface charge due to surfactant adsorption (32) since TX-100 molecules interact with the OH groups. The fact that the evolutions of the zeta potential and the adsorption isotherm curves are parallel, reaching a “plateau” in both cases, seems to be logical. If the amount adsorbed is constant at the “plateau,” the zeta potential values also should remain stable at this concentration interval. Thus, as can be seen from Fig. 5, an approximate estimation of c.m.c. can be also given from the zeta potential curve. When NP25S is adsorbed on the kaolin surface (or on QB
TABLE 5 Surface Charges of Solids in 1 g/liter NaCl Solution
swithout TX-100 (mcul/m2)
swith TX-100 (mcul/m2)
QA QB QC Kaolin Dolomite
2159.0 241.1 24.5 248.4 249.53
2136.8 219.8 23.0 220.9 24.50
´ PEZ-GONZA ´ LEZ NEVSKAIA, GUERRERO-RUI´Z, AND LO
quartz) an increase in zeta potential absolute negative values can be observed (Fig. 6 and Table 2). This effect seems to occur because the surfactant has a net negative charge, and it could seem unusual that a negatively charged surfactant could be adsorbed on a negative solid surface. It should be considered that, in the case of the QB sample (Table 2), the adsorbed amounts are very low if we compare them with the corresponding amounts of TX-100 adsorbed. The reason for this low adsorption could be that the surfactant is linked to OH surface groups by its ether groups. The negatively charged group diminishes the adsorbed amount by charge repulsion, but it is not able to totally avoid this adsorption. On the other hand, kaolin adsorbs greater amounts of anionic surfactants than QB samples (Table 2). This fact could be explained by the presence of positive charges on the edges of kaolin. Surface Charge The first detailed experimental evidence about oxide–surface solution interactions obtained by Hunter et al. (33) and Lyklema (34) indicates that excessively high values of surface charge with respect to electrokinetic potential measured are produced. From Table 5, it may be observed that the QA sample (159 mcul/m2) exhibits a surface charge value much higher than the QB (41 mcul/m2) and the QC (4.5 mcul/m2). Lyklema (34) suggested that such high values of surface negative charge were due to a gel layer formation on the solid surface. The basic idea was that H1 and OH2 ions could penetrate the surface layers of the oxide and react with the surface OH groups. Thus, only a part of the surface charge could be due to charged groups. At the same time, if contraions can penetrate in this porous layer, the net electric potential in the other side of the porous layer could be considerably reduced in magnitude. Thus, a “high charge density” could be compatible with low values of electrokinetic potential. By thermal analysis determination of surface OH groups (17) we found that on QA and QB, surfaces had a very high OH content compared with the accepted values (25). Thus, these two facts (high OH content and high charge density) seem to indicate that on QA and QB surfaces a gel layer is present. Finally, from Table 5 it can be also observed that when TX-100 is previously adsorbed on samples, the surface charge decreases by the consumption of some surface OH groups due to the formation of hydrogen bonds with TX-100 ether groups, as was expected from the generally accepted adsorption mechanisms. CONCLUSIONS
Due to the different surface charge regimes of quartz and kaolin, the adsorption behaviors of nonionic and anionic surfactants are also different. The QB sample adsorbs twice more TX-100 than kaolin per surface square meter while kaolin adsorbs about nine times more anionic surfactants than this quartz.
When NaCl salt is added to the solution of TX-100, three different adsorption behaviors have been found. Besides that, in some samples (QA and dolomite) there is a change in the adsorption behavior of TX-100 depending on the nature of the added salts (NaCl or CaCl2). However, in the case of QB, QC, and kaolin no significant influence of the two different salts was detected. These differences are associated with the interaction of salt cations with different surface hydroxyl groups (isolated or hydrogen-bonded) and with the presence of surface impurities rather than salt–TX100 interactions. The anionic surfactants (NP4S, NP10S, and NP25S) show the same behavior on quartz and on kaolin when NaCl is added. The rise in adsorbed amounts on the quartz surface when NaCl is added is attributed to the negative surface charge compensation by Na1 ions. On kaolin, the anionic surfactants, in the absence of salt, could be adsorbed on the edges (while the nonionic surfactant is adsorbed on surface basal planes). When Na1 ions are added, the adsorbed amounts of anionic surfactants also rise, indicating that an additional part of these surfactants is adsorbed on the basal planes. When a nonionic surfactant, as is the case of TX-100, is adsorbed on negatively charged surfaces the zeta potential values and the surface charge decrease. This may be explained by the fact that the OH groups that are responsible for surface charge are bound with ether groups of the surfactants. On the contrary, the adsorption of anionic surfactants (negatively charged) produces an increase in the negative values of the zeta potentials, as was expected. ACKNOWLEDGMENT The authors acknowledge the financial support of Project JOULE by the E.U. (Contract CT-91-0062).
REFERENCES 1. Denoyel, R., and Rouquerol, I., J. Colloid Interface Sci. 143(2), 555 (1991). 2. Partyka, S., Lindheimer, M., and Faucompre, B., Colloids Surf. A: Physicochem. Eng. Asp. 76, 267 (1993). 3. Davies, R. J., Dix, L. R., and Toprakcioglu, C., J. Colloid Interface Sci. 129(1), 145 (1989). 4. Nishikido, N., and Matuura, R., Bull. Chem. Soc. Japan 6, 793 (1977). 5. Schott, H., and Royce, A. E., J. Pharm. Sci. 76(6), 793 (1984). 6. Delsanti, M., Moussaid, A., and Munch, J. P., J. Colloid Interface Sci. 157, 285 (1993). 7. Makino, K., Ohshima, H., and Kondo, T., J. Colloid Interface Sci. 115, 65 (1987). 8. Brooks, D. E., and Seaman, G. V. F., J. Colloid Interface Sci. 43, (1973) 670. 9. Parks, G. A., and de Bruyn, P. L., J. Phys. Chem. 66, 87 (1962). 10. Levine, S., and Smith, A. L., Disc. Faraday Soc. 52, 290 (1971). 11. Healy, T. W., Yates, D. E., White, L. R., and Chan, D. J. Electroanal. Chem. 80, 57 (1977). 12. Healy, T. W., and White, L. R., Adv. Colloid Interface 9, 303 (1978). 13. Moreno, R., Moya, J. S., and Requena, J., Bol. Soc. Esp. Ceram. Vidr. 26(6), 355 (1987).
ADSORPTION OF NONIONIC AND ANIONIC SURFACTANTS 14. Herrington, T. M., Clarke, A. Q., and Watts, J. C., Colloids Surf. 68, 161 (1992). 15. van Olphen, H., “Introduction to Clay Colloid Chemistry.” Interscience, New York, 1963. 16. “Chemistry of Clays and Clay Minerals,” Mineralogical Society Monograph No 6, A. C. D. Newman, Ed., 1987. 17. Nevskaia, D. M., Rojas Cervantes, M. L., Guerrero Ruı´z, A., and Lo´pez Gonza´lez, J. D., J. Chem. Technol. Biotechnol. 63, 249 (1995). 18. Everett, D., Pure Appl. Chem. 58, 967 (1986). 19. Smoluchowski, M., Z. Phys. Chem. 93, 129 (1918). 20. Van Raij, B., and Peech, M., Soil. Sci. Soc. Amer. Proc. 36, 587 (1972). 21. Somasundaran, P., Snell, E. D., Fu, E., and Xu, Q., Colloids Surf. 63, 49 (1992). 22. Esumi, K., and Matsui, H., Colloids Surf. 30, 273 (1993). 23. Nevskaia, D. M., Ph.D. thesis, U.N.E.D., Spain, 1995. 24. Lijour, Y., Calves, J-Y., and Saumagne P., J. Chem. Soc. Faraday Trans. 1, 83, 3283 (1987).
25. Zhuravlev, L. T., Langmuir 3, 592 (1980). 26. Yates, D. E., and Healy, T. W., J. Colloid Interface Sci. 55, 9 (1976). 27. Nevskaia, D. M., Guerrero-Ruı´z, A., and Lo´pez-Gonza´lez, J. de D., J. Colloid Surf. Sci. 181, 571 (1996). 28. Bottero, J. Y., Bruant, M., and Cases, J. M., J. Colloid Interface Sci. 124, 515 (1988). 29. Peng, F. F., and Pingkuan, D., J. Colloid Interface Sci. 164, 229 (1994). 30. Schott, H., and Han, S. K., J. Pharm. Sci. 64, 658 (1975). 31. Hunter, R. J., in “Zeta Potential in Colloid Science. Principles and Applications” (R. H. Ottewill and R. L. Rowell, Eds.). Academic Press, London, 1988. 32. Sidorova, M., Golub, T., and Musabekov, K., Adv. Colloid Interface Sci. 43, 1 (1993). 33. Hunter, R. J., and Wright, H. J. L., J. Colloid Interface Sci. 37, 564 (1971). 34. Lyklema, J., Croatica Chem. Acta 43, 249 (1971).