- Email: [email protected]
Binding of ionic surfactants to purified humic acid
Journal of Colloid and Interface Science 275 (2004) 360–367 www.elsevier.com/locate/jcis
Binding of ionic surfactants to purified humic acid Luuk K. Koopal,∗ Tanya P. Goloub, and Thomas A. Davis Laboratory of Physical Chemistry and Colloid Science, Wageningen University, Dreijenplein 6, 6703 HB Wageningen, The Netherlands Received 17 October 2003; accepted 24 February 2004 Available online 9 April 2004
Abstract The binding of organic contaminants to dissolved humic acids reduces the free concentration of the contaminants in the environment and also may cause changes to the solution properties of humic acids. Surfactants are a special class of contaminants that are introduced into the environment either through wastewater or by site-specific contamination. The amphiphilic nature of both surfactants and humic acids can easily lead to their mutual attraction and consequently affect the solution behavior of the humics. Binding of an anionic surfactant (sodium dodecyl sulfate, SDS) and two cationic surfactants (dodecyl- and cetylpyridinium chloride, DPC and CPC) to purified Aldrich humic acid (PAHA) is studied at pH values of 5, 7, and 10 in solutions with a 0.025 M ionic strength (I ). Monomer concentrations of the surfactants are measured with a surfactant-selective electrode. At I = 0.025 M, no significant binding is observed between the anionic surfactant (SDS) and PAHA, whereas the two cationic surfactants (DPC, CPC) bind strongly to PAHA over the pH range investigated. The binding is due both to electrostatic and hydrophobic attraction. The initial affinity increases with increasing pH (i.e., negative charge of PAHA) and tail length of the surfactant. Binding reaches a pseudo-plateau value (2–5 mmol/g) when the charge associated with PAHA is neutralized by that of the bound surfactant molecules. The pseudo-plateau values for DPC and CPC are very similar and depend on the solution pH. The cationic surfactant–PAHA complexes precipitate when the charge neutralization point is reached. This occurs at approximately 10% of the critical micelle concentration or CMC. This type of phase separation commonly occurs during surfactant binding to oppositely charged polyelectrolytes. For CPC, the precipitation is complete, but in the case of DPC, a noticeable fraction of PAHA remains in solution. At very low CPC concentrations (less than 0.1% of the CMC), CPC binding to PAHA is cooperative. The investigated range of concentrations for DPC was too limited to reach a similar conclusion. The results of this study demonstrate that the fate of humic acids will be strongly affected by the presence of low cationic surfactant concentrations in aqueous environmental systems. 2004 Elsevier Inc. All rights reserved. Keywords: Humic acid; Anionic surfactant; Cationic surfactant; Surfactant binding; Surfactant electrode; Electrostatic interaction; Hydrophobic interaction; Humic acid precipitation; Surfactant micellization
1. Introduction Humic acids (HA) are a specific class of natural organic particles that are soluble in aqueous solutions in the pH range of 2–10. They are often considered to be polydisperse, structured polyelectrolytes of an amphiphilic nature [1,2]. The solubility discriminates HA from humins (insoluble) and fulvic acids (also soluble below pH 2). Due to their solubility, HAs can easily be transported in the aqueous phase through soil and other natural waters and it is well established that they play an important role in the distribution of contaminants in the environment [3]. Contaminant bind* Corresponding author. Fax: +31 317 483777.
E-mail address: [email protected] (L.K. Koopal). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.02.061
ing to HAs may also significantly impact the total and free contaminant concentrations present in surface and ground waters. The free contaminant concentrations are highly relevant to their bioavailability and toxicity [3]. Furthermore, the mobility of contaminants through ground waters and due to fresh water dynamics is an important factor to consider when evaluating the risks associated with contamination. Contaminant solubility and mobility in natural waters can either be reduced by binding to precipitated humic matter or increased by binding to dissolved humic matter. Precipitated HA will become part of the sediment matrix and thus the co-precipitated contaminant and its mobility in the aqueous phase are diminished. The extent of flocculation depends on the pH, the temperature, degree of hydrophobicity of the humic acid, and the nature of the contaminant. The
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
mobility of dissolved HAs can also be reduced by its binding to soil mineral particles [4–9]. In principle, this binding will also be influenced by the presence of contaminants in the environment. Many studies have focused on the binding of both inorganic (e.g., [10–14]) and organic (e.g., [15–21]) contaminants. However, surfactants, a special class of organic contaminants that may affect the fate of HA more strongly than other contaminants, have received limited attention. Surfactants can be introduced into the environment by wastewater discharge, point-discharge pollution, deliberate action, or natural secretion from aquatic plants. Wastewater treatment may remove some of the surfactants, yet detectable levels may persist [22–24]. Point-discharge pollution may result from foams used to extinguish hydrocarbon fuel fires [25]. On occasion, surfactants are deliberately introduced into the environment to remediate contaminants from soil (e.g., 2,4,6-trinitrotoluene [26]) or from groundwater (e.g., light hydrocarbons [27]). Natural plant-derived surfactants have been detected in river water at concentrations sufficiently high to produce persistent foams [28]. Previous studies performed on surfactant–HA interactions include those of Tombacz et al. [29,30] in which X-ray and interfacial tension measurements were performed on alkylammonium–humate complexes. A possible structure was discussed, but without reference to binding isotherms. Traina et al. [31] found that the association of linear alkylbenzene sulfonates with humic substances increased with increasing chain length. Otto et al. [32] indicated that humic substances enhance the aggregation of sodium dodecyl sulfate prior to micellization and form ion pairs with cetyltrimethylammonium bromide. Adou et al. [33] demonstrated that cationic surfactant–HA interactions could lead to phase separation (precipitation) of the dissolved humics. De Wuilloud et al. [34] used a cationic surfactant to preconcentrate humic and fulvic acids present at trace levels in samples of natural water. Preconcentration with Triton X100, a nonionic surfactant, was also achieved [35]. In general, it is well known that interactions between ionic surfactants and oppositely charged polyelectrolytes are quite strong [36,37]. Pikulell [38] discussed the phase behavior of polymer–surfactant systems and also the potential for phase separation. The phase separation observed by Adou et al. [33] is probably of a similar nature. The authors state that humic acid is removed from the aqueous phase by forming neutral hydrophobic complexes with cationic surfactants; however, no further evidence for the binding mechanism was presented. The objective of this study is to quantify the binding of surfactants to HA and to characterize their mutual interaction. This will permit us to assess in greater detail the relative importance of surfactant–HA interactions in determining HA solubility and, therefore, contaminant mobility in the environment. In order to consider both the electrostatic and hydrophobic components of the interaction, the binding of two cationic surfactants composed of aliphatic tails of dif-
361
fering lengths, as well as one anionic surfactant, to a purified humic acid is investigated at three solution pH values and at constant ionic strength.
2. Materials and methods 2.1. Water and chemicals Water used for the experiments was twice deionized and filtered through an activated carbon column and a micro filter (EASYpure UV), which resulted in a resistivity greater than 18.3 M cm. The inorganic chemicals used were of analytical grade quality (obtained from Merck). 2.2. Humic acid (PAHA) Humic acid in the sodium form was obtained from Aldrich (H1, 675-2) and was chosen because it is easily available. Furthermore, physicochemical studies of various humic acid samples have shown [39,40] that the ion-binding and hydrodynamic behavior of purified Aldrich humic acid is similar to that of other humic acids. The sample was purified by placing 10 g of humic acid into 1 L of NaOH (pH 10), stirred overnight, and then centrifuged to remove undissolved organic and inorganic matter. The humic acid fraction was precipitated by adjusting the solution pH to 2 with 1 M HCl, stirred for 24 h, and then centrifuged at 10,000 rpm for 30 min (Beckman JA-20). The precipitate was subsequently converted to the protonated form by rinsing several times with 0.01 M HCl. This was followed by a rinse with water, and then the sample was filtered with an ultrafiltration membrane (Millipore, regenerated cellulose, NMWL 10,000) using an AMICON cell. Finally, the sample was shaken in contact with an acid resin (DOWEX 50W-X8) for two days in order to remove trace metals. The final product was freeze-dried and is denoted as PAHA (purified Aldrich HA) in this work. 2.3. Surfactants The n-dodecylpyridinium chloride (DPC) sample was synthesized and purified according to the method of Colichman [41]. Purification was performed by recrystallizing the sample twice from acetone. The n-hexadecyl- or cetylpyridinium chloride (CPC; >98% purity) and sodium dodecyl sulfate (SDS; 99% purity) samples were supplied by BDH Chemicals and used as received. These surfactants display a linear aliphatic tail. The purity of the surfactants was investigated by measuring the surface tension of aqueous solutions of the samples. The absence of a minimum in a plot against log surfactant concentration indicated the absence of surface-active impurities.
362
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
2.4. Surfactant electrode and calibration: monomer surfactant concentration The equilibrium concentration of surfactant monomers in solution was determined with a commercial membrane electrode obtained from Metrohm A.G., Herisau, Switzerland (Type 6.0507.120) that is sensitive to ionic surfactants. The electrode potential (EMF in mV) was measured relative to an Ag/AgCl (3.3 mol/L KCl) reference electrode equipped with a ceramic plug. The use of surfactant electrodes has now become routine in the study of polymer–surfactant interactions [42–46] and was also deemed the most suitable for our experimental system. The surfactant electrode was calibrated at a given pH by titrating a concentrated surfactant solution (1 to 20 times the critical micelle concentration or CMC) into a reaction cell containing 25 ml of 0.025 M electrolyte solution at constant pH. Surfactant additions were made using a motorized piston burette (Schott T100) and an automated titration device (Schott TR250). A mixing time of 1 min was allowed after each aliquot addition (0.05–0.2 ml). The stirring was then stopped and the electrode potential was recorded. The closed reaction cell was kept at a constant temperature of 25 ± 0.5 ◦ C, and maintained in an argon atmosphere. The solution was mixed with a magnetic stirrer. A surfactant electrode that functions properly should yield a straight line (mV versus the logarithm of the concentration) for a blank experiment and a correlation coefficient of at least 0.999. Blanks were measured both before and after each humic acid titration and they should both give the same result within the experimental error (±3%). The data presented in this study adhere to these criteria. NaCl was used as the supporting electrolyte for experiments performed at pH 5 (ionic strength I = 0.025 M) and the pH was adjusted by adding the required amount of 0.1 M HCl. Experiments performed at pH 7 and 10 were buffered for both pH and ionic strength (I = 0.025 M) with phosphate (1.023 g Na2 HPO4 + 0.619 g NaH2 PO4 per liter) and carbonate (0.6831 g Na2 CO3 + 0.5041 g NaHCO3 per liter) buffers, respectively.
centration of PAHA in the solutions ranged between 0.2 and 0.5 g/L. The pH was controlled or buffered as indicated in the surfactant electrode section.
3. Results and discussion 3.1. Surfactant titrations in the absence and presence of PAHA Three typical surfactant–PAHA titration curves and corresponding blank titrations are shown in Fig. 1 for the SDS–,
2.5. Binding isotherms Binding isotherms were generated by employing the surfactant titration protocol in the presence of PAHA. After each aliquot addition, a maximum equilibration time of 10 min (equilibration criterion 5 mV/min) was allowed. The recorded EMF is a measure of the equilibrium monomer surfactant concentration in solution [42,44]. The quantity of free surfactant is calculated from the equilibrium monomer concentration and the known total solution volume. Binding to PAHA is determined by subtracting the free amount from the quantity initially added to the solution. Accordingly, the binding isotherm is obtained. Binding isotherms were determined at pH 5 and 7 for the SDS–PAHA system and at pH 5, 7, and 10 for the DPC– and CPC–PAHA systems. The con-
Fig. 1. Surfactant–PAHA titrations at pH 7 and I = 0.025 M. The cell potential or EMF (mV) is plotted as a function of the logarithm of the total (a) SDS, (b) DPC, and (c) CPC concentrations (mmol/L). Circles () depict the blank titrations (i.e., calibration lines) before electrode contact with PAHA; crosses (×) depict blank titrations after the surfactant–PAHA binding experiments. Triangles () depict surfactant titrations in the presence of 0.4 g/L PAHA.
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
DPC–, and CPC–PAHA systems at pH 7. The calibrations were performed at surfactant concentrations sufficiently low to allow accurate determination of the monomer concentration for all systems. Similar results are obtained at other solution pH values and PAHA concentrations. The basic property of the blank titration or calibration curves is their linear response as a function of the logarithm of the concentration. The slope obtained is typically within the range of 42–45 mV for all three surfactants. Slopes are lower than for a Nernstian response; according to the electrode manufacturer, this behavior is frequently observed in practice [47]. Upon the gradual increase of the monomer surfactant concentration, the CMC is reached. CMC values measured in the absence of PAHA and at an ionic strength of 0.025 M will be denoted as CMC◦ . At this concentration, the chemical potential of the surfactant monomers becomes practically constant due to micelle formation, and hence, the EMF response of the electrode remains constant. This corresponds to the small plateau at the end of the blank titration curves shown in Fig. 1. The CMC◦ values extracted from the blank curves were 4.0 ± 0.4, 10.1 ± 0.5, and 0.11 ± 0.03 mmol/L for SDS, DPC, and CPC, respectively. Within the quoted uncertainties, the values are independent of pH and are in agreement with the CMC values (I ≈ 0.025 M) reported by Van Os et al. [48] and Mukerjee and Mysels [49]. The fact that the values are independent of the solution pH implies that they are also counterion-independent. The latter observation is probably related to the relatively high ionic strength which provides a generic screening of the charges since, in the absence of added salts, the CMC of pyridinium surfactants is affected by the surfactant counterion [48]. Finally, these buffers are well suited for the binding studies because no significant differences in the determined CMC◦ values were detected with the different buffers. The SDS–PAHA system shown in Fig. 1a indicates that there is no discernable difference between the blank titration curve and the curve obtained from SDS in the presence of PAHA. The same results were obtained at pH 5 (not shown). No titration was performed at pH 10 because no attraction was expected to take place at higher pH values, due to the increased electrostatic repulsion. In contrast to the SDS– PAHA system, titrations of PAHA in the presence of the cationic surfactants DPC and CPC (Figs. 1b and 1c, pH 7) deviate strongly from the calibration lines. The results depicted in Figs. 1b and 1c are typical, and similar curves were obtained for the experiments performed at pH 5 and 10. The large deviations from the calibration lines indicate that the cationic surfactants bind strongly to PAHA at all measured concentrations. This observation is in accordance with literature results [31–34]. 3.2. SDS–PAHA binding A comparison between the SDS–PAHA titration curves and their corresponding blank titrations emphasizes two important features. The first is that binding of SDS to PAHA
363
Fig. 2. DPC–PAHA binding (mmol/g) as a function of the logarithm of the equilibrium DPC concentration. The isotherms are obtained at pH 10, I = 0.025 M, and the three PAHA concentrations indicated in the figure. A duplicate is presented for 0.4 g PAHA/L.
is not observed at I = 0.025 M. Apparently, the strong electrostatic repulsion between the negatively charged SDS and PAHA molecules outweighs any hydrophobic attraction. The present observation does not exclude the possibility that binding may occur at higher ionic strengths, at which the electrostatic repulsion will be suppressed. Other studies [31,32] have demonstrated some form of interaction between anionic surfactants and HA. The second feature is that the presence of PAHA does not influence the response of the surfactant electrode, thus indicating the absence of any interaction between PAHA and the electrode surface. 3.3. Cationic surfactant–PAHA binding and PAHA concentration Binding isotherms were calculated from the data presented in Figs. 1b and 1c, for the DPC– and CPC–PAHA systems, as well as from results obtained for the same systems at different pH values and PAHA concentrations. Fig. 2 depicts DPC–PAHA binding isotherms at three different PAHA concentrations (0.2, 0.4, and 0.5 g/L PAHA) and pH 10. It is evident from these data that the surfactant binding (mmol/g) is independent of the PAHA concentration (as it should be) and is highly reproducible. The overall reproducibility of the experiments can also be observed from the duplicates depicted in Fig. 3 in which the binding isotherms for both DPC– and CPC–PAHA are plotted at three different pH values and two PAHA concentrations (0.2 and 0.4 g/L). The reproducibility of all the experiments taken together lends credibility to the method employed and also indicates the absence of any interaction of PAHA with the surfactant selective electrode. 3.4. DPC– and CPC–PAHA binding isotherms The binding isotherms for the DPC– and CPC–PAHA systems at pH 5, 7, and 10 are plotted in Fig. 3 as a function of the logarithm of the reduced surfactant concentration (i.e., the ratio of the free surfactant concentration to the CMC◦ ). Plotting the results in this manner corrects the data
364
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
Fig. 3. Binding of DPC and CPC to PAHA (mmol surfactant/g PAHA) at I = 0.025 M and (a) pH 5, (b) pH 7, and (c) pH 10. The reduced monomer concentration, c/CMC◦ , normalizes the concentration differences between DPC and CPC. The arrows indicate the amount of negative charge (mmol/g) of PAHA in 0.02 M KNO3 at the given pH values. Duplicate experiments are indicated with different symbols: DPC () and CPC (×) refer to a PAHA concentration of 0.2 g/L; DPC () and CPC (+) to 0.4 g/L.
for hydrophobic effects (i.e., surfactant chain length) in the solution. The results, therefore, illustrate more clearly the similarities and/or differences in the interaction of the two surfactants with PAHA. In general, the binding isotherms for DPC– and CPC–PAHA are similar. Initially, the binding increases gradually with the reduced surfactant concentration. Subsequently, the binding levels off at a small pseudoplateau region. The slope of the initial part of the isotherm and the pseudo-plateau value both increase with increasing pH. Precipitation of the surfactant–PAHA complex, as was observed visually during the titrations, occurred when the reduced surfactant concentration value that corresponds to the pseudo-plateau was reached. The isotherms rise steeply at a reduced surfactant concentration close to the CMC◦ . This rise is especially prominent for the DPC–PAHA system and is due to the formation of surfactant micelles and not to surfactant binding to PAHA. For DPC–PAHA, micellization begins at a concentration that is somewhat lower than the
CMC◦ , whereas in the case of the CPC–PAHA system, the micellization concentration corresponds to the CMC◦ . Micellization may begin at concentrations that are less than the CMC◦ if the presence of PAHA causes a reduction in the Gibbs energy of micellization. The general features of the isotherms will be discussed in more detail below. The slope of the initial part of the binding curves in Fig. 3 increases with increasing pH and is a measure of the binding affinity. This behavior indicates that the increasing negative charge of PAHA (resulting from increasing pH) increases the affinity of PAHA for cationic surfactants. This fact, combined with the observation that SDS does not bind to PAHA at the investigated ionic strength, demonstrates that electrostatic attraction is an important feature in the determination of the binding affinity. A fairly steep initial slope is often observed in polyelectrolyte systems with oppositely charged surfactants and indicates a cooperative binding process [50–54]. The degree of cooperativity depends on the nature of the polymer and is lower for hydrophobic [55] and/or cross-linked [56] polymers than for linear hydrophilic polymers. In view of the observations of Avena et al. [40,57], who demonstrated that PAHA is structured [40] and that it readily adsorbs onto hydrophobic surfaces [57], it is likely that the moderate slopes observed for the binding of DPC and CPC to PAHA indicate behavior similar to that of cross-linked and/or partly hydrophobic polyelectrolytes. A close comparison of the DPC– and CPC–PAHA isotherms at a given pH reveals that the middle sections of the isotherms coincide. This is due to the fact that the isotherms are plotted versus the reduced concentration. It follows that the tendency of these surfactants to bind to PAHA that is already complexed with some surfactant parallels their tendency to form micelles. It appears that the Gibbs energy for transferring an alkyl group into the PAHA– surfactant complexes correlates with a similar transfer into micelles. This indicates that both processes are controlled by hydrophobic interactions of a similar nature. Correspondingly, micelles and PAHA–surfactant complexes display similar levels of hydrophobicity. It should be noted that the initial surfactant binding will enforce the hydrophobicity of the humic acid core and this will, in turn, enhance the surfactant–PAHA binding. Two conclusions emerge: (1) hydrophobic attraction is an important driving force for the binding of the surfactants to PAHA, and (2) the “core” of the PAHA–surfactant complex is largely apolar (hydrophobic). Another key observation, in addition to the fact that the DPC– and CPC–PAHA isotherms have the same pseudoplateau values for a given pH, is that the plateau values increase with increasing pH. The small difference between plateau values for DPC and CPC at pH 5 can be attributed to slightly poorer pH control (no buffer present) at this pH than at pH 7 and 10 (buffer present). The similarity in pseudoplateau values for both surfactants and their dependence on pH are strong indications that the plateau values are related to the charge density of the humic acid. For this reason,
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
the PAHA charge density in 0.02 M KNO3 , expressed as mmol negative charges per g PAHA, at each of the three pH values are indicated in Fig. 3 with an arrow. The values are 1.7, 3.3, and 4.3 mmol negative charge/g at pH 5, 7, and 10, respectively [58]. For each pH value, the amount of bound surfactant (mmol/g) equals the amount of negative charge of PAHA (mmol/g) at a surfactant concentration approximately one order of magnitude lower than the CMC◦ (0.1 CMC◦ ). It is at approximately this concentration that binding takes place within the pseudo-plateau region for all of the isotherms. Once the PAHA charge is compensated, the electrostatic driving force for binding is eliminated and superequivalent surfactant binding will lead to an electrostatic repulsion. This behavior explains the presence of the pseudo-plateau in the isotherms. Therefore, both the pH dependency of the initial slopes of the isotherms and the pseudo-plateau values near the point of charge compensation emphasize the importance of the electrostatic interaction. Flocculation or phase separation of the surfactant–PAHA complex is visually observed near the point of charge compensation. This phase separation is induced both by charge neutralization and by the hydrophobicity of the surfactant– PAHA complex and is sustained at higher surfactant concentrations. Equilibrium phase separation (i.e., a polymer-rich versus a polymer-poor phase) is commonly observed with the binding of ionic surfactants to oppositely charged polyelectrolytes (e.g., [52,54,55,59]). For the surfactant–PAHA system, a phase separation of a similar nature will most likely occur. Therefore, beyond the point of charge compensation, the system will be composed of a phase that is PAHA–surfactant concentrated (precipitated material) and a phase that is very dilute in PAHA–surfactant (dissolved material). The more hydrophobic the surfactant, the more hydrophobic the precipitated PAHA–surfactant complex, and therefore, the stronger the resulting phase separation. In the studied system, the surfactant binding in mmol/g is the same for the two cationic surfactants at the point of charge compensation, however, the CPC tail is 4 CH2 segments longer than the DPC tail. Therefore, the aliphaticity of the precipitated CPC–PAHA complex is considerably higher than for that of the DPC–PAHA complex. Consequently, for CPC, the remaining concentration of PAHA in the dilute surfactant–PAHA phase will be smaller than that in the presence of DPC. This corresponds with our experimental observation that micellization of DPC in the dilute surfactant– PAHA phase begins at concentrations less than the CMC◦ , whereas for CPC it begins at, or very close to, the CMC◦ . Apparently, in the case of CPC, the PAHA concentration in the dilute PAHA phase is too low to enhance CPC micellization. Due to the lower hydrophobicity of DPC, phase separation, in this case, leaves a small fraction of PAHA in solution. This small fraction of dissolved PAHA is sufficient to induce micellization of DPC at a concentration somewhat lower than the CMC◦ . As a consequence of its heterogeneity and the polydisperse character of PAHA, the phase separation also leaves
365
open the possibility that the PAHA remaining in solution is chemically distinct from that which precipitated with the surfactant. The PAHA remaining in the dilute PAHA–DPC phase will likely have a lower molecular mass as well as being more hydrophilic (e.g., having a higher negative charge density) than the starting material. Due to its polyelectrolyte nature, the HA fraction remaining in solution may screen the charge of the DPC micelles more effectively than simple inorganic ions do and this may possibly be the reason for the decrease in the observed CMC. 3.5. Behavior at very low surfactant concentrations Double logarithmic plots of the binding curves are well suited for gaining insight into the binding behavior of surfactants at very low concentrations. This type of plot is commonly employed in the characterization of ion binding to humic acids [14] and is useful for detecting heterogeneity effects at trace concentration levels. The plots are also used to detect positive cooperativity in the binding of surfactants to mineral surfaces that results from the formation of hemimicelles [60–62]. The concept behind this manner of plotting is that at extremely low concentrations “ideal” binding behavior may be expected whereby, at equilibrium, a direct proportionality exists between the concentration and the binding (the so-called “Henry region”). Ideal binding corresponds to a slope of unity in a log–log plot. This implies that (a) all binding sites are equivalent, (b) the fraction of binding sites that is occupied by surfactant molecules is negligible, (c) the bound surfactant molecules do not interact mutually, and (d) structural changes of the sorbent (as a consequence of the binding) do not change the binding affinity. Provided that measurements are made at sufficiently low concentrations, deviations from a slope of unity will indicate binding heterogeneity or cooperativity effects. Values of the initial slope smaller than unity indicate that the binding sites are heterogeneous [63]. A slope lower than unity is frequently observed in the binding of metal ions by humic acids at extremely low ion concentrations [14]. Vermeer et al. [13] have demonstrated that this is also the case for Cd–PAHA ion binding. Furthermore, potentiometric titrations of PAHA reveal that the affinity of protons for PAHA is strongly heterogeneous [64]. The interaction between the cationic surfactant head group and the acid groups of PAHA is, therefore, likely also heterogeneous. Moreover, the interaction between the aliphatic surfactant tails and PAHA will depend on the heterogeneity of the core of the humic acid molecules. The case in which an initial slope of unity is followed by a slope smaller than unity indicates negative cooperativity. This may be due to either or both coverage effects (a noticeable fraction of the sites are occupied) and a diminished attraction. The latter may easily occur for ion adsorption onto an oppositely charged surface. An initial slope of unity followed by a slope larger than unity, or an initial slope larger than unity, both indicate positive cooperativity. The greater the deviation from unity, the stronger the cooperativity ef-
366
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
ativity. Clearly, this cooperativity is strong enough to mask any possible heterogeneity of the PAHA–surfactant binding. Since the amount bound is still very low (especially at pH 5), the cooperativity is likely due in large part to structural rearrangements of the PAHA molecules that are induced by the binding of CPC molecules. Previous experiments on binding to solid surfaces [58] have demonstrated that PAHA molecules can undergo structural rearrangements. Such rearrangements, in combination with the presence of the bound surfactant molecules, can lead to an increase in the hydrophobic attraction between surfactant and PAHA. They may be induced by both the increase of the hydrophobicity and the decrease of electrostatic repulsion occurring within the PAHA molecules.
4. Conclusions
Fig. 4. Double logarithmic plots of the binding isotherms of (a) DPC and (b) CPC to PAHA at pH 5, 7, and 10. Duplicates are indicated by different symbols.
fect. For surfactants, positive cooperativity is often due to sorption induced self-assembly caused by hydrophobic attraction (e.g., [60–62]). Log–log plots for DPC– and CPC–PAHA are depicted in Figs. 4a and 4b, respectively. The line drawn from the origin has a slope of unity. At binding levels lower than the pseudo-plateau, the surfactant concentration required for a particular binding level increases with decreasing pH (i.e., when the charge of the PAHA molecules is reduced). This illustrates the electrostatic contribution to the affinity. For DPC–PAHA, the initial slopes are approximately unity and are steeper than the slopes at higher concentrations. Therefore, our results provide no indication of heterogeneity or positive cooperativity in the case of DPC–PAHA. However, additional experiments at even lower surfactant concentrations and/or other salt concentrations will be required to further investigate heterogeneity and/or self-assembly effects for DPC (see also the results for CPC). For the CPC–PAHA system the reduced concentration axis is extended to lower concentrations than for DPC and initial slopes are observed to be larger than unity. Slopes equal to, or lower than, unity should be present at even lower concentrations. Some scatter appears in the data for pH 5, however, the trend is clear. At pH 7, the three lowest concentrations for the two isotherms display a slope greater than unity. At pH 10, this occurs only for the two lowest concentrations. It becomes more difficult to observe cooperativity effects at high pH values due to the stronger electrostatic attraction. At pH 5 and 7, CPC–PAHA binding displays a slope larger than unity even at the lowest measured equilibrium concentrations and hence displays positive cooper-
At an ionic strength of 0.025 M no interaction is observable between PAHA and SDS. However, the cationic surfactants DPC and CPC bind strongly to PAHA as a result of both hydrophobic and electrostatic attraction. Surfactant binding reaches a pseudo-plateau value and causes charge neutralization of PAHA followed by a phase separation (precipitation) at surfactant concentrations of approximately 10% of the CMC◦ . For CPC, the precipitation of the humic acid appears to be so exhaustive that micellization in the dilute CPC–PAHA phase mimics the behavior of a system with no remaining PAHA. For DPC, with its shorter aliphatic tail, the precipitation is less extensive and micellization in the dilute DPC–PAHA phase occurs at a concentration somewhat lower than that which occurs in the PAHAfree system. In the case of CPC, trace quantities of surfactant are sufficient to induce cooperative hydrophobic binding effects. This cooperativity is most probably due to surfactant induced structural rearrangements of PAHA. In the case of DPC, no cooperative effects were observed, but the concentration range studied was too limited to categorically exclude such interactions. Experiments at even lower surfactant concentrations and different ionic strengths will be required to fully understand surfactant–humic acid interactions. Considering the general applicability of both electrostatic and hydrophobic interactions and their relevance to cationic surfactant–humic acid systems, it is to be expected that the trends observed for PAHA in this study will apply generally to other humic acids. The significance of this is that trace levels of long-chain surfactants present in the environment will increase the hydrophobicity of humic acids as well as influence their (and other contaminants) adsorption to them. Furthermore, cationic surfactant concentrations that are approximately an order of magnitude smaller than the CMC will likely lead to the precipitation of humic acids in aqueous environmental systems. Therefore, the mobility of humic acid–cationic surfactant complexes will be strongly reduced. Ultimately, these findings have important environmental implications for the transport and bioavailability of
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
surfactants and other contaminants capable of binding to humic substances.
Acknowledgments The “Wageningen Institute for Environment and Climate Research,” WIMEK, is kindly acknowledged for providing financial support for T.G. Financial support for T.A.D. was provided in the form of a Canadian postdoctoral fellowship from “Le fonds québécois de la recherche sur la nature et les technologies (NATEQ).”
References [1] T.F. Guetzloff, J.A. Rice, Sci. Total Environ. 152 (1994) 31. [2] S.C.B. Myneni, J.T. Brown, G.A. Martinez, I.W. Meyer, Science 286 (5443) (1999) 1335. [3] J. Buffle, Complexation Reactions in Aquatic Systems, Ellis Horwood, Chichester, 1988. [4] E. Tipping, R. Griffith, J. Hilton, Croat. Chem. Acta 56 (1983) 613. [5] E.M. Murphy, J.M. Zachara, S.C. Smith, J.L. Philips, Sci. Total Environ. 117/118 (1992) 413. [6] B. Gu, J. Schmitt, Z. Chen, L. Liang, J.F. McCarty, Environ. Sci. Technol. 28 (1994) 38. [7] J.P. Pinheiro, A.M. Mota, M.S. Goncalves, H.P. Van Leeuwen, Environ. Sci. Technol. 28 (1994) 2112. [8] A.W.P. Vermeer, W.H. Van Riemsdijk, L.K. Koopal, Langmuir 14 (1998) 2810. [9] A.W.P. Vermeer, L.K. Koopal, Langmuir 14 (1998) 4210. [10] J. Buffle, in: H. Sigel (Ed.), Circulation of Metals in the Environment, in: Metal Ions in Biological Systems, vol. 18, Dekker, New York, 1984, p. 165. [11] G. Sposito, CRC Crit. Rev. Environ. Control 16 (1986) 193. [12] R.M. Town, H.K.J. Powell, Anal. Chim. Acta 279 (1993) 221. [13] A.P.W. Vermeer, J.K. McCulloch, W.H. Van Riemsdijk, L.K. Koopal, Environ. Sci. Technol. 33 (1999) 3892. [14] D.G. Kinniburgh, W.H. Van Riemsdijk, L.K. Koopal, M. Borkovec, M.F. Benedetti, M.J. Avena, Colloids Surf. A 151 (1999) 147. [15] S.L. Stangroom, J.N. Lester, C.D. Collins, Environ. Technol. 21 (2000) 845. [16] J.-W. Ding, J.-C. Wu, Water Sci. Technol. 35 (1997) 139. [17] P. Schmitt, D. Freitag, I. Trapp, A.W. Garrison, M. Schiavon, A. Kettrup, Chemosphere 35 (1997) 55. [18] L.F. Schultz, T.M. Young, Environ. Toxicol. Chem. 18 (1999) 1710. [19] Y.-H. Yang, L.K. Koopal, Colloids Surf. A 151 (1999) 201. [20] Y. Laor, M. Rebhun, Environ. Sci. Technol. 36 (2002) 955. [21] C.L. Lee, L.J. Kuo, H.L. Wang, P.C. Hsieh, Water Res. 37 (2003) 4250. [22] M. Stalmans, E. Matthijs, N.T. De Oude, Water Sci. Technol. 43 (1991) 1. [23] J. Waters, T.C.J. Feijtel, Chemosphere 30 (1995) 1939. [24] H.R. Rogers, Sci. Tot. Environ. 185 (1996) 3. [25] C.A. Moody, J.A. Field, Environ. Sci. Technol. 34 (2000) 3864. [26] M.R. Taha, I.H. Soewarto, R.J. Acar, R.J. Gale, M.E. Zappi, Water Air Soil Pollut. 100 (1997) 33. [27] J.H. Harwell, D.A. Sabatini, R.C. Knox, Colloids Surf. A 151 (1999) 255. [28] C. Wegner, M. Hamburger, Environ. Sci. Technol. 36 (2002) 3250. [29] E. Tombacz, K. Varga, F. Szanto, Colloid Polym. Sci. 266 (1988) 734. [30] E. Tombacz, I. Regdon, in: N. Senesi, T.M. Miano (Eds.), Humic Substances in the Global Environment and Implications on Human Health, Elsevier Science, Amsterdam, 1994, p. 139.
367
[31] S.J. Traina, D.C. McAvoy, D.J. Versteeg, Environ. Sci. Technol. 30 (1996) 1300. [32] W.H. Otto, D.J. Britten, C.K. Larive, J. Colloid Interface Sci. 261 (2003) 508. [33] A.F.Y. Adou, V.S. Muhandiki, Y. Shimizu, S. Matsui, Water Sci. Technol. 43 (2001) 1. [34] J.C.A. de Wuilloud, R.G. Wuilloud, B.B.M. Sadi, J.A. Caruso, Analyst 128 (2003) 453. [35] R.L. Revia, G.A. Makharadze, Talanta 48 (1999) 409. [36] K. Shirahama, in: J.C.T. Kwak (Ed.), Polymer–Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 143. [37] P. Linse, L. Piculell, P. Hansson, in: J.C.T. Kwak (Ed.), Polymer– Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 193. [38] L. Pikulell, in: J.C.T. Kwak (Ed.), Polymer–Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 65. [39] C.J. Milne, D.G. Kinniburgh, W.H. Van Riemsdijk, E. Tipping, Environ. Sci. Technol. 37 (2003) 958. [40] M.J. Avena, A.P.W. Vermeer, L.K. Koopal, Colloids Surf. A 151 (1999) 213. [41] E.I. Colichman, J. Am. Chem. Soc. 72 (1950) 1834. [42] T. Shimizu, M. Seki, J.C.T. Kwak, Colloids Surf. 20 (1986) 289. [43] D.M. Bloor, W.M.Z. Wan-Yunus, W.A. Wan-Badhi, Y. Li, J.F. Hozwarth, Langmuir 11 (1995) 334. [44] D.M. Bloor, H.K.O. Mwakibete, E. Wyn-Jones, J. Colloid Interface Sci. 178 (1996) 334. [45] S.M. Ghoreishi, Y. Li, J.F. Holzwarth, E. Khoshdel, J. Warr, D.M. Bloor, E. Wyn-Jones, Langmuir 15 (1999) 1938. [46] S.M. Ghoreishi, Y. Li, D.M. Bloor, J. Warr, E. Wyn-Jones, Langmuir 15 (1999) 4380. [47] Metrohm Application Bulletin 233/3 e. [48] N.M. Van Os, J.R. Haak, L.A.M. Rupert, Physicochemical Properties of Selected Anionic, Cationic and Nonionic Surfactants, Elsevier Science, Amsterdam, 1993. [49] P. Mukerjee, K.J. Mysels, Critical Micelle Concentrations of Aqueous Surfactant Systems, National Bureau of Standards, Washington, DC, 1971, Report NSRDS-NBS 36. [50] A.P. Rodenhiser, J.C.T. Kwak, in: J.C.T. Kwak (Ed.), Polymer– Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 1. [51] T. Shimizu, J.C.T. Kwak, Colloids Surf. A 82 (1994) 163. [52] T. Shimizu, Colloids Surf. A 84 (1994) 239. [53] T. Shimizu, Colloids Surf. A 94 (1995) 115. [54] K. Kogej, J. Skerjanc, in: T. Radeva (Ed.), Physical Chemistry of Polyelectrolytes, in: Surfactant Science Series, vol. 99, Dekker, New York, 2001, p. 973. [55] E.D. Goddar, in: E.D. Goddar, K.P. Ananthapadmanabhan (Eds.), Interactions of Surfactants with Polymers and Proteins, CRC Press, Boca Raton, FL, 1992, p. 123. [56] J.P. Gong, Y. Osada, J. Phys. Chem. 99 (1995) 10971. [57] M.J. Avena, L.K. Koopal, Eviron. Sci. Technol. 33 (1999) 2739. [58] M.J. Avena, L.K. Koopal, W.H. Van Riemsdijk, J. Colloid Interface Sci. 217 (1999) 37. [59] E.D. Goddar, K.P. Ananthapadmanabham, in: J.C.T. Kwak (Ed.), Polymer–Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 21. [60] T.P. Goloub, L.K. Koopal, Langmuir 13 (1997) 673. [61] P. Somasundaran, D.W. Fuerstenau, J. Phys. Chem. 70 (1966) 90. [62] L.K. Koopal, T.P. Goloub, in: R. Sharma (Ed.), Surfactant Adsorption and Surface Solubilization, in: ACS Symposium Series, vol. 615, American Chemical Society, Washington, DC, 1995, p. 78. [63] W.H. Van Riemsdijk, L.K. Koopal, in: J. Buffle, H.P. van Leeuwen (Eds.), Environmental Particles, in: Environmental Analytical and Physical Chemistry Series, vol. 1, Lewis Publishers, Boca Raton, FL, 1992, p. 455. [64] A.P.W. Vermeer, L.K. Koopal, J. Colloid Interface Sci. 212 (1999) 176.
Binding of ionic surfactants to purified humic acid Luuk K. Koopal,∗ Tanya P. Goloub, and Thomas A. Davis Laboratory of Physical Chemistry and Colloid Science, Wageningen University, Dreijenplein 6, 6703 HB Wageningen, The Netherlands Received 17 October 2003; accepted 24 February 2004 Available online 9 April 2004
Abstract The binding of organic contaminants to dissolved humic acids reduces the free concentration of the contaminants in the environment and also may cause changes to the solution properties of humic acids. Surfactants are a special class of contaminants that are introduced into the environment either through wastewater or by site-specific contamination. The amphiphilic nature of both surfactants and humic acids can easily lead to their mutual attraction and consequently affect the solution behavior of the humics. Binding of an anionic surfactant (sodium dodecyl sulfate, SDS) and two cationic surfactants (dodecyl- and cetylpyridinium chloride, DPC and CPC) to purified Aldrich humic acid (PAHA) is studied at pH values of 5, 7, and 10 in solutions with a 0.025 M ionic strength (I ). Monomer concentrations of the surfactants are measured with a surfactant-selective electrode. At I = 0.025 M, no significant binding is observed between the anionic surfactant (SDS) and PAHA, whereas the two cationic surfactants (DPC, CPC) bind strongly to PAHA over the pH range investigated. The binding is due both to electrostatic and hydrophobic attraction. The initial affinity increases with increasing pH (i.e., negative charge of PAHA) and tail length of the surfactant. Binding reaches a pseudo-plateau value (2–5 mmol/g) when the charge associated with PAHA is neutralized by that of the bound surfactant molecules. The pseudo-plateau values for DPC and CPC are very similar and depend on the solution pH. The cationic surfactant–PAHA complexes precipitate when the charge neutralization point is reached. This occurs at approximately 10% of the critical micelle concentration or CMC. This type of phase separation commonly occurs during surfactant binding to oppositely charged polyelectrolytes. For CPC, the precipitation is complete, but in the case of DPC, a noticeable fraction of PAHA remains in solution. At very low CPC concentrations (less than 0.1% of the CMC), CPC binding to PAHA is cooperative. The investigated range of concentrations for DPC was too limited to reach a similar conclusion. The results of this study demonstrate that the fate of humic acids will be strongly affected by the presence of low cationic surfactant concentrations in aqueous environmental systems. 2004 Elsevier Inc. All rights reserved. Keywords: Humic acid; Anionic surfactant; Cationic surfactant; Surfactant binding; Surfactant electrode; Electrostatic interaction; Hydrophobic interaction; Humic acid precipitation; Surfactant micellization
1. Introduction Humic acids (HA) are a specific class of natural organic particles that are soluble in aqueous solutions in the pH range of 2–10. They are often considered to be polydisperse, structured polyelectrolytes of an amphiphilic nature [1,2]. The solubility discriminates HA from humins (insoluble) and fulvic acids (also soluble below pH 2). Due to their solubility, HAs can easily be transported in the aqueous phase through soil and other natural waters and it is well established that they play an important role in the distribution of contaminants in the environment [3]. Contaminant bind* Corresponding author. Fax: +31 317 483777.
E-mail address: [email protected] (L.K. Koopal). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.02.061
ing to HAs may also significantly impact the total and free contaminant concentrations present in surface and ground waters. The free contaminant concentrations are highly relevant to their bioavailability and toxicity [3]. Furthermore, the mobility of contaminants through ground waters and due to fresh water dynamics is an important factor to consider when evaluating the risks associated with contamination. Contaminant solubility and mobility in natural waters can either be reduced by binding to precipitated humic matter or increased by binding to dissolved humic matter. Precipitated HA will become part of the sediment matrix and thus the co-precipitated contaminant and its mobility in the aqueous phase are diminished. The extent of flocculation depends on the pH, the temperature, degree of hydrophobicity of the humic acid, and the nature of the contaminant. The
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
mobility of dissolved HAs can also be reduced by its binding to soil mineral particles [4–9]. In principle, this binding will also be influenced by the presence of contaminants in the environment. Many studies have focused on the binding of both inorganic (e.g., [10–14]) and organic (e.g., [15–21]) contaminants. However, surfactants, a special class of organic contaminants that may affect the fate of HA more strongly than other contaminants, have received limited attention. Surfactants can be introduced into the environment by wastewater discharge, point-discharge pollution, deliberate action, or natural secretion from aquatic plants. Wastewater treatment may remove some of the surfactants, yet detectable levels may persist [22–24]. Point-discharge pollution may result from foams used to extinguish hydrocarbon fuel fires [25]. On occasion, surfactants are deliberately introduced into the environment to remediate contaminants from soil (e.g., 2,4,6-trinitrotoluene [26]) or from groundwater (e.g., light hydrocarbons [27]). Natural plant-derived surfactants have been detected in river water at concentrations sufficiently high to produce persistent foams [28]. Previous studies performed on surfactant–HA interactions include those of Tombacz et al. [29,30] in which X-ray and interfacial tension measurements were performed on alkylammonium–humate complexes. A possible structure was discussed, but without reference to binding isotherms. Traina et al. [31] found that the association of linear alkylbenzene sulfonates with humic substances increased with increasing chain length. Otto et al. [32] indicated that humic substances enhance the aggregation of sodium dodecyl sulfate prior to micellization and form ion pairs with cetyltrimethylammonium bromide. Adou et al. [33] demonstrated that cationic surfactant–HA interactions could lead to phase separation (precipitation) of the dissolved humics. De Wuilloud et al. [34] used a cationic surfactant to preconcentrate humic and fulvic acids present at trace levels in samples of natural water. Preconcentration with Triton X100, a nonionic surfactant, was also achieved [35]. In general, it is well known that interactions between ionic surfactants and oppositely charged polyelectrolytes are quite strong [36,37]. Pikulell [38] discussed the phase behavior of polymer–surfactant systems and also the potential for phase separation. The phase separation observed by Adou et al. [33] is probably of a similar nature. The authors state that humic acid is removed from the aqueous phase by forming neutral hydrophobic complexes with cationic surfactants; however, no further evidence for the binding mechanism was presented. The objective of this study is to quantify the binding of surfactants to HA and to characterize their mutual interaction. This will permit us to assess in greater detail the relative importance of surfactant–HA interactions in determining HA solubility and, therefore, contaminant mobility in the environment. In order to consider both the electrostatic and hydrophobic components of the interaction, the binding of two cationic surfactants composed of aliphatic tails of dif-
361
fering lengths, as well as one anionic surfactant, to a purified humic acid is investigated at three solution pH values and at constant ionic strength.
2. Materials and methods 2.1. Water and chemicals Water used for the experiments was twice deionized and filtered through an activated carbon column and a micro filter (EASYpure UV), which resulted in a resistivity greater than 18.3 M cm. The inorganic chemicals used were of analytical grade quality (obtained from Merck). 2.2. Humic acid (PAHA) Humic acid in the sodium form was obtained from Aldrich (H1, 675-2) and was chosen because it is easily available. Furthermore, physicochemical studies of various humic acid samples have shown [39,40] that the ion-binding and hydrodynamic behavior of purified Aldrich humic acid is similar to that of other humic acids. The sample was purified by placing 10 g of humic acid into 1 L of NaOH (pH 10), stirred overnight, and then centrifuged to remove undissolved organic and inorganic matter. The humic acid fraction was precipitated by adjusting the solution pH to 2 with 1 M HCl, stirred for 24 h, and then centrifuged at 10,000 rpm for 30 min (Beckman JA-20). The precipitate was subsequently converted to the protonated form by rinsing several times with 0.01 M HCl. This was followed by a rinse with water, and then the sample was filtered with an ultrafiltration membrane (Millipore, regenerated cellulose, NMWL 10,000) using an AMICON cell. Finally, the sample was shaken in contact with an acid resin (DOWEX 50W-X8) for two days in order to remove trace metals. The final product was freeze-dried and is denoted as PAHA (purified Aldrich HA) in this work. 2.3. Surfactants The n-dodecylpyridinium chloride (DPC) sample was synthesized and purified according to the method of Colichman [41]. Purification was performed by recrystallizing the sample twice from acetone. The n-hexadecyl- or cetylpyridinium chloride (CPC; >98% purity) and sodium dodecyl sulfate (SDS; 99% purity) samples were supplied by BDH Chemicals and used as received. These surfactants display a linear aliphatic tail. The purity of the surfactants was investigated by measuring the surface tension of aqueous solutions of the samples. The absence of a minimum in a plot against log surfactant concentration indicated the absence of surface-active impurities.
362
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
2.4. Surfactant electrode and calibration: monomer surfactant concentration The equilibrium concentration of surfactant monomers in solution was determined with a commercial membrane electrode obtained from Metrohm A.G., Herisau, Switzerland (Type 6.0507.120) that is sensitive to ionic surfactants. The electrode potential (EMF in mV) was measured relative to an Ag/AgCl (3.3 mol/L KCl) reference electrode equipped with a ceramic plug. The use of surfactant electrodes has now become routine in the study of polymer–surfactant interactions [42–46] and was also deemed the most suitable for our experimental system. The surfactant electrode was calibrated at a given pH by titrating a concentrated surfactant solution (1 to 20 times the critical micelle concentration or CMC) into a reaction cell containing 25 ml of 0.025 M electrolyte solution at constant pH. Surfactant additions were made using a motorized piston burette (Schott T100) and an automated titration device (Schott TR250). A mixing time of 1 min was allowed after each aliquot addition (0.05–0.2 ml). The stirring was then stopped and the electrode potential was recorded. The closed reaction cell was kept at a constant temperature of 25 ± 0.5 ◦ C, and maintained in an argon atmosphere. The solution was mixed with a magnetic stirrer. A surfactant electrode that functions properly should yield a straight line (mV versus the logarithm of the concentration) for a blank experiment and a correlation coefficient of at least 0.999. Blanks were measured both before and after each humic acid titration and they should both give the same result within the experimental error (±3%). The data presented in this study adhere to these criteria. NaCl was used as the supporting electrolyte for experiments performed at pH 5 (ionic strength I = 0.025 M) and the pH was adjusted by adding the required amount of 0.1 M HCl. Experiments performed at pH 7 and 10 were buffered for both pH and ionic strength (I = 0.025 M) with phosphate (1.023 g Na2 HPO4 + 0.619 g NaH2 PO4 per liter) and carbonate (0.6831 g Na2 CO3 + 0.5041 g NaHCO3 per liter) buffers, respectively.
centration of PAHA in the solutions ranged between 0.2 and 0.5 g/L. The pH was controlled or buffered as indicated in the surfactant electrode section.
3. Results and discussion 3.1. Surfactant titrations in the absence and presence of PAHA Three typical surfactant–PAHA titration curves and corresponding blank titrations are shown in Fig. 1 for the SDS–,
2.5. Binding isotherms Binding isotherms were generated by employing the surfactant titration protocol in the presence of PAHA. After each aliquot addition, a maximum equilibration time of 10 min (equilibration criterion 5 mV/min) was allowed. The recorded EMF is a measure of the equilibrium monomer surfactant concentration in solution [42,44]. The quantity of free surfactant is calculated from the equilibrium monomer concentration and the known total solution volume. Binding to PAHA is determined by subtracting the free amount from the quantity initially added to the solution. Accordingly, the binding isotherm is obtained. Binding isotherms were determined at pH 5 and 7 for the SDS–PAHA system and at pH 5, 7, and 10 for the DPC– and CPC–PAHA systems. The con-
Fig. 1. Surfactant–PAHA titrations at pH 7 and I = 0.025 M. The cell potential or EMF (mV) is plotted as a function of the logarithm of the total (a) SDS, (b) DPC, and (c) CPC concentrations (mmol/L). Circles () depict the blank titrations (i.e., calibration lines) before electrode contact with PAHA; crosses (×) depict blank titrations after the surfactant–PAHA binding experiments. Triangles () depict surfactant titrations in the presence of 0.4 g/L PAHA.
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
DPC–, and CPC–PAHA systems at pH 7. The calibrations were performed at surfactant concentrations sufficiently low to allow accurate determination of the monomer concentration for all systems. Similar results are obtained at other solution pH values and PAHA concentrations. The basic property of the blank titration or calibration curves is their linear response as a function of the logarithm of the concentration. The slope obtained is typically within the range of 42–45 mV for all three surfactants. Slopes are lower than for a Nernstian response; according to the electrode manufacturer, this behavior is frequently observed in practice [47]. Upon the gradual increase of the monomer surfactant concentration, the CMC is reached. CMC values measured in the absence of PAHA and at an ionic strength of 0.025 M will be denoted as CMC◦ . At this concentration, the chemical potential of the surfactant monomers becomes practically constant due to micelle formation, and hence, the EMF response of the electrode remains constant. This corresponds to the small plateau at the end of the blank titration curves shown in Fig. 1. The CMC◦ values extracted from the blank curves were 4.0 ± 0.4, 10.1 ± 0.5, and 0.11 ± 0.03 mmol/L for SDS, DPC, and CPC, respectively. Within the quoted uncertainties, the values are independent of pH and are in agreement with the CMC values (I ≈ 0.025 M) reported by Van Os et al. [48] and Mukerjee and Mysels [49]. The fact that the values are independent of the solution pH implies that they are also counterion-independent. The latter observation is probably related to the relatively high ionic strength which provides a generic screening of the charges since, in the absence of added salts, the CMC of pyridinium surfactants is affected by the surfactant counterion [48]. Finally, these buffers are well suited for the binding studies because no significant differences in the determined CMC◦ values were detected with the different buffers. The SDS–PAHA system shown in Fig. 1a indicates that there is no discernable difference between the blank titration curve and the curve obtained from SDS in the presence of PAHA. The same results were obtained at pH 5 (not shown). No titration was performed at pH 10 because no attraction was expected to take place at higher pH values, due to the increased electrostatic repulsion. In contrast to the SDS– PAHA system, titrations of PAHA in the presence of the cationic surfactants DPC and CPC (Figs. 1b and 1c, pH 7) deviate strongly from the calibration lines. The results depicted in Figs. 1b and 1c are typical, and similar curves were obtained for the experiments performed at pH 5 and 10. The large deviations from the calibration lines indicate that the cationic surfactants bind strongly to PAHA at all measured concentrations. This observation is in accordance with literature results [31–34]. 3.2. SDS–PAHA binding A comparison between the SDS–PAHA titration curves and their corresponding blank titrations emphasizes two important features. The first is that binding of SDS to PAHA
363
Fig. 2. DPC–PAHA binding (mmol/g) as a function of the logarithm of the equilibrium DPC concentration. The isotherms are obtained at pH 10, I = 0.025 M, and the three PAHA concentrations indicated in the figure. A duplicate is presented for 0.4 g PAHA/L.
is not observed at I = 0.025 M. Apparently, the strong electrostatic repulsion between the negatively charged SDS and PAHA molecules outweighs any hydrophobic attraction. The present observation does not exclude the possibility that binding may occur at higher ionic strengths, at which the electrostatic repulsion will be suppressed. Other studies [31,32] have demonstrated some form of interaction between anionic surfactants and HA. The second feature is that the presence of PAHA does not influence the response of the surfactant electrode, thus indicating the absence of any interaction between PAHA and the electrode surface. 3.3. Cationic surfactant–PAHA binding and PAHA concentration Binding isotherms were calculated from the data presented in Figs. 1b and 1c, for the DPC– and CPC–PAHA systems, as well as from results obtained for the same systems at different pH values and PAHA concentrations. Fig. 2 depicts DPC–PAHA binding isotherms at three different PAHA concentrations (0.2, 0.4, and 0.5 g/L PAHA) and pH 10. It is evident from these data that the surfactant binding (mmol/g) is independent of the PAHA concentration (as it should be) and is highly reproducible. The overall reproducibility of the experiments can also be observed from the duplicates depicted in Fig. 3 in which the binding isotherms for both DPC– and CPC–PAHA are plotted at three different pH values and two PAHA concentrations (0.2 and 0.4 g/L). The reproducibility of all the experiments taken together lends credibility to the method employed and also indicates the absence of any interaction of PAHA with the surfactant selective electrode. 3.4. DPC– and CPC–PAHA binding isotherms The binding isotherms for the DPC– and CPC–PAHA systems at pH 5, 7, and 10 are plotted in Fig. 3 as a function of the logarithm of the reduced surfactant concentration (i.e., the ratio of the free surfactant concentration to the CMC◦ ). Plotting the results in this manner corrects the data
364
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
Fig. 3. Binding of DPC and CPC to PAHA (mmol surfactant/g PAHA) at I = 0.025 M and (a) pH 5, (b) pH 7, and (c) pH 10. The reduced monomer concentration, c/CMC◦ , normalizes the concentration differences between DPC and CPC. The arrows indicate the amount of negative charge (mmol/g) of PAHA in 0.02 M KNO3 at the given pH values. Duplicate experiments are indicated with different symbols: DPC () and CPC (×) refer to a PAHA concentration of 0.2 g/L; DPC () and CPC (+) to 0.4 g/L.
for hydrophobic effects (i.e., surfactant chain length) in the solution. The results, therefore, illustrate more clearly the similarities and/or differences in the interaction of the two surfactants with PAHA. In general, the binding isotherms for DPC– and CPC–PAHA are similar. Initially, the binding increases gradually with the reduced surfactant concentration. Subsequently, the binding levels off at a small pseudoplateau region. The slope of the initial part of the isotherm and the pseudo-plateau value both increase with increasing pH. Precipitation of the surfactant–PAHA complex, as was observed visually during the titrations, occurred when the reduced surfactant concentration value that corresponds to the pseudo-plateau was reached. The isotherms rise steeply at a reduced surfactant concentration close to the CMC◦ . This rise is especially prominent for the DPC–PAHA system and is due to the formation of surfactant micelles and not to surfactant binding to PAHA. For DPC–PAHA, micellization begins at a concentration that is somewhat lower than the
CMC◦ , whereas in the case of the CPC–PAHA system, the micellization concentration corresponds to the CMC◦ . Micellization may begin at concentrations that are less than the CMC◦ if the presence of PAHA causes a reduction in the Gibbs energy of micellization. The general features of the isotherms will be discussed in more detail below. The slope of the initial part of the binding curves in Fig. 3 increases with increasing pH and is a measure of the binding affinity. This behavior indicates that the increasing negative charge of PAHA (resulting from increasing pH) increases the affinity of PAHA for cationic surfactants. This fact, combined with the observation that SDS does not bind to PAHA at the investigated ionic strength, demonstrates that electrostatic attraction is an important feature in the determination of the binding affinity. A fairly steep initial slope is often observed in polyelectrolyte systems with oppositely charged surfactants and indicates a cooperative binding process [50–54]. The degree of cooperativity depends on the nature of the polymer and is lower for hydrophobic [55] and/or cross-linked [56] polymers than for linear hydrophilic polymers. In view of the observations of Avena et al. [40,57], who demonstrated that PAHA is structured [40] and that it readily adsorbs onto hydrophobic surfaces [57], it is likely that the moderate slopes observed for the binding of DPC and CPC to PAHA indicate behavior similar to that of cross-linked and/or partly hydrophobic polyelectrolytes. A close comparison of the DPC– and CPC–PAHA isotherms at a given pH reveals that the middle sections of the isotherms coincide. This is due to the fact that the isotherms are plotted versus the reduced concentration. It follows that the tendency of these surfactants to bind to PAHA that is already complexed with some surfactant parallels their tendency to form micelles. It appears that the Gibbs energy for transferring an alkyl group into the PAHA– surfactant complexes correlates with a similar transfer into micelles. This indicates that both processes are controlled by hydrophobic interactions of a similar nature. Correspondingly, micelles and PAHA–surfactant complexes display similar levels of hydrophobicity. It should be noted that the initial surfactant binding will enforce the hydrophobicity of the humic acid core and this will, in turn, enhance the surfactant–PAHA binding. Two conclusions emerge: (1) hydrophobic attraction is an important driving force for the binding of the surfactants to PAHA, and (2) the “core” of the PAHA–surfactant complex is largely apolar (hydrophobic). Another key observation, in addition to the fact that the DPC– and CPC–PAHA isotherms have the same pseudoplateau values for a given pH, is that the plateau values increase with increasing pH. The small difference between plateau values for DPC and CPC at pH 5 can be attributed to slightly poorer pH control (no buffer present) at this pH than at pH 7 and 10 (buffer present). The similarity in pseudoplateau values for both surfactants and their dependence on pH are strong indications that the plateau values are related to the charge density of the humic acid. For this reason,
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
the PAHA charge density in 0.02 M KNO3 , expressed as mmol negative charges per g PAHA, at each of the three pH values are indicated in Fig. 3 with an arrow. The values are 1.7, 3.3, and 4.3 mmol negative charge/g at pH 5, 7, and 10, respectively [58]. For each pH value, the amount of bound surfactant (mmol/g) equals the amount of negative charge of PAHA (mmol/g) at a surfactant concentration approximately one order of magnitude lower than the CMC◦ (0.1 CMC◦ ). It is at approximately this concentration that binding takes place within the pseudo-plateau region for all of the isotherms. Once the PAHA charge is compensated, the electrostatic driving force for binding is eliminated and superequivalent surfactant binding will lead to an electrostatic repulsion. This behavior explains the presence of the pseudo-plateau in the isotherms. Therefore, both the pH dependency of the initial slopes of the isotherms and the pseudo-plateau values near the point of charge compensation emphasize the importance of the electrostatic interaction. Flocculation or phase separation of the surfactant–PAHA complex is visually observed near the point of charge compensation. This phase separation is induced both by charge neutralization and by the hydrophobicity of the surfactant– PAHA complex and is sustained at higher surfactant concentrations. Equilibrium phase separation (i.e., a polymer-rich versus a polymer-poor phase) is commonly observed with the binding of ionic surfactants to oppositely charged polyelectrolytes (e.g., [52,54,55,59]). For the surfactant–PAHA system, a phase separation of a similar nature will most likely occur. Therefore, beyond the point of charge compensation, the system will be composed of a phase that is PAHA–surfactant concentrated (precipitated material) and a phase that is very dilute in PAHA–surfactant (dissolved material). The more hydrophobic the surfactant, the more hydrophobic the precipitated PAHA–surfactant complex, and therefore, the stronger the resulting phase separation. In the studied system, the surfactant binding in mmol/g is the same for the two cationic surfactants at the point of charge compensation, however, the CPC tail is 4 CH2 segments longer than the DPC tail. Therefore, the aliphaticity of the precipitated CPC–PAHA complex is considerably higher than for that of the DPC–PAHA complex. Consequently, for CPC, the remaining concentration of PAHA in the dilute surfactant–PAHA phase will be smaller than that in the presence of DPC. This corresponds with our experimental observation that micellization of DPC in the dilute surfactant– PAHA phase begins at concentrations less than the CMC◦ , whereas for CPC it begins at, or very close to, the CMC◦ . Apparently, in the case of CPC, the PAHA concentration in the dilute PAHA phase is too low to enhance CPC micellization. Due to the lower hydrophobicity of DPC, phase separation, in this case, leaves a small fraction of PAHA in solution. This small fraction of dissolved PAHA is sufficient to induce micellization of DPC at a concentration somewhat lower than the CMC◦ . As a consequence of its heterogeneity and the polydisperse character of PAHA, the phase separation also leaves
365
open the possibility that the PAHA remaining in solution is chemically distinct from that which precipitated with the surfactant. The PAHA remaining in the dilute PAHA–DPC phase will likely have a lower molecular mass as well as being more hydrophilic (e.g., having a higher negative charge density) than the starting material. Due to its polyelectrolyte nature, the HA fraction remaining in solution may screen the charge of the DPC micelles more effectively than simple inorganic ions do and this may possibly be the reason for the decrease in the observed CMC. 3.5. Behavior at very low surfactant concentrations Double logarithmic plots of the binding curves are well suited for gaining insight into the binding behavior of surfactants at very low concentrations. This type of plot is commonly employed in the characterization of ion binding to humic acids [14] and is useful for detecting heterogeneity effects at trace concentration levels. The plots are also used to detect positive cooperativity in the binding of surfactants to mineral surfaces that results from the formation of hemimicelles [60–62]. The concept behind this manner of plotting is that at extremely low concentrations “ideal” binding behavior may be expected whereby, at equilibrium, a direct proportionality exists between the concentration and the binding (the so-called “Henry region”). Ideal binding corresponds to a slope of unity in a log–log plot. This implies that (a) all binding sites are equivalent, (b) the fraction of binding sites that is occupied by surfactant molecules is negligible, (c) the bound surfactant molecules do not interact mutually, and (d) structural changes of the sorbent (as a consequence of the binding) do not change the binding affinity. Provided that measurements are made at sufficiently low concentrations, deviations from a slope of unity will indicate binding heterogeneity or cooperativity effects. Values of the initial slope smaller than unity indicate that the binding sites are heterogeneous [63]. A slope lower than unity is frequently observed in the binding of metal ions by humic acids at extremely low ion concentrations [14]. Vermeer et al. [13] have demonstrated that this is also the case for Cd–PAHA ion binding. Furthermore, potentiometric titrations of PAHA reveal that the affinity of protons for PAHA is strongly heterogeneous [64]. The interaction between the cationic surfactant head group and the acid groups of PAHA is, therefore, likely also heterogeneous. Moreover, the interaction between the aliphatic surfactant tails and PAHA will depend on the heterogeneity of the core of the humic acid molecules. The case in which an initial slope of unity is followed by a slope smaller than unity indicates negative cooperativity. This may be due to either or both coverage effects (a noticeable fraction of the sites are occupied) and a diminished attraction. The latter may easily occur for ion adsorption onto an oppositely charged surface. An initial slope of unity followed by a slope larger than unity, or an initial slope larger than unity, both indicate positive cooperativity. The greater the deviation from unity, the stronger the cooperativity ef-
366
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
ativity. Clearly, this cooperativity is strong enough to mask any possible heterogeneity of the PAHA–surfactant binding. Since the amount bound is still very low (especially at pH 5), the cooperativity is likely due in large part to structural rearrangements of the PAHA molecules that are induced by the binding of CPC molecules. Previous experiments on binding to solid surfaces [58] have demonstrated that PAHA molecules can undergo structural rearrangements. Such rearrangements, in combination with the presence of the bound surfactant molecules, can lead to an increase in the hydrophobic attraction between surfactant and PAHA. They may be induced by both the increase of the hydrophobicity and the decrease of electrostatic repulsion occurring within the PAHA molecules.
4. Conclusions
Fig. 4. Double logarithmic plots of the binding isotherms of (a) DPC and (b) CPC to PAHA at pH 5, 7, and 10. Duplicates are indicated by different symbols.
fect. For surfactants, positive cooperativity is often due to sorption induced self-assembly caused by hydrophobic attraction (e.g., [60–62]). Log–log plots for DPC– and CPC–PAHA are depicted in Figs. 4a and 4b, respectively. The line drawn from the origin has a slope of unity. At binding levels lower than the pseudo-plateau, the surfactant concentration required for a particular binding level increases with decreasing pH (i.e., when the charge of the PAHA molecules is reduced). This illustrates the electrostatic contribution to the affinity. For DPC–PAHA, the initial slopes are approximately unity and are steeper than the slopes at higher concentrations. Therefore, our results provide no indication of heterogeneity or positive cooperativity in the case of DPC–PAHA. However, additional experiments at even lower surfactant concentrations and/or other salt concentrations will be required to further investigate heterogeneity and/or self-assembly effects for DPC (see also the results for CPC). For the CPC–PAHA system the reduced concentration axis is extended to lower concentrations than for DPC and initial slopes are observed to be larger than unity. Slopes equal to, or lower than, unity should be present at even lower concentrations. Some scatter appears in the data for pH 5, however, the trend is clear. At pH 7, the three lowest concentrations for the two isotherms display a slope greater than unity. At pH 10, this occurs only for the two lowest concentrations. It becomes more difficult to observe cooperativity effects at high pH values due to the stronger electrostatic attraction. At pH 5 and 7, CPC–PAHA binding displays a slope larger than unity even at the lowest measured equilibrium concentrations and hence displays positive cooper-
At an ionic strength of 0.025 M no interaction is observable between PAHA and SDS. However, the cationic surfactants DPC and CPC bind strongly to PAHA as a result of both hydrophobic and electrostatic attraction. Surfactant binding reaches a pseudo-plateau value and causes charge neutralization of PAHA followed by a phase separation (precipitation) at surfactant concentrations of approximately 10% of the CMC◦ . For CPC, the precipitation of the humic acid appears to be so exhaustive that micellization in the dilute CPC–PAHA phase mimics the behavior of a system with no remaining PAHA. For DPC, with its shorter aliphatic tail, the precipitation is less extensive and micellization in the dilute DPC–PAHA phase occurs at a concentration somewhat lower than that which occurs in the PAHAfree system. In the case of CPC, trace quantities of surfactant are sufficient to induce cooperative hydrophobic binding effects. This cooperativity is most probably due to surfactant induced structural rearrangements of PAHA. In the case of DPC, no cooperative effects were observed, but the concentration range studied was too limited to categorically exclude such interactions. Experiments at even lower surfactant concentrations and different ionic strengths will be required to fully understand surfactant–humic acid interactions. Considering the general applicability of both electrostatic and hydrophobic interactions and their relevance to cationic surfactant–humic acid systems, it is to be expected that the trends observed for PAHA in this study will apply generally to other humic acids. The significance of this is that trace levels of long-chain surfactants present in the environment will increase the hydrophobicity of humic acids as well as influence their (and other contaminants) adsorption to them. Furthermore, cationic surfactant concentrations that are approximately an order of magnitude smaller than the CMC will likely lead to the precipitation of humic acids in aqueous environmental systems. Therefore, the mobility of humic acid–cationic surfactant complexes will be strongly reduced. Ultimately, these findings have important environmental implications for the transport and bioavailability of
L.K. Koopal et al. / Journal of Colloid and Interface Science 275 (2004) 360–367
surfactants and other contaminants capable of binding to humic substances.
Acknowledgments The “Wageningen Institute for Environment and Climate Research,” WIMEK, is kindly acknowledged for providing financial support for T.G. Financial support for T.A.D. was provided in the form of a Canadian postdoctoral fellowship from “Le fonds québécois de la recherche sur la nature et les technologies (NATEQ).”
References [1] T.F. Guetzloff, J.A. Rice, Sci. Total Environ. 152 (1994) 31. [2] S.C.B. Myneni, J.T. Brown, G.A. Martinez, I.W. Meyer, Science 286 (5443) (1999) 1335. [3] J. Buffle, Complexation Reactions in Aquatic Systems, Ellis Horwood, Chichester, 1988. [4] E. Tipping, R. Griffith, J. Hilton, Croat. Chem. Acta 56 (1983) 613. [5] E.M. Murphy, J.M. Zachara, S.C. Smith, J.L. Philips, Sci. Total Environ. 117/118 (1992) 413. [6] B. Gu, J. Schmitt, Z. Chen, L. Liang, J.F. McCarty, Environ. Sci. Technol. 28 (1994) 38. [7] J.P. Pinheiro, A.M. Mota, M.S. Goncalves, H.P. Van Leeuwen, Environ. Sci. Technol. 28 (1994) 2112. [8] A.W.P. Vermeer, W.H. Van Riemsdijk, L.K. Koopal, Langmuir 14 (1998) 2810. [9] A.W.P. Vermeer, L.K. Koopal, Langmuir 14 (1998) 4210. [10] J. Buffle, in: H. Sigel (Ed.), Circulation of Metals in the Environment, in: Metal Ions in Biological Systems, vol. 18, Dekker, New York, 1984, p. 165. [11] G. Sposito, CRC Crit. Rev. Environ. Control 16 (1986) 193. [12] R.M. Town, H.K.J. Powell, Anal. Chim. Acta 279 (1993) 221. [13] A.P.W. Vermeer, J.K. McCulloch, W.H. Van Riemsdijk, L.K. Koopal, Environ. Sci. Technol. 33 (1999) 3892. [14] D.G. Kinniburgh, W.H. Van Riemsdijk, L.K. Koopal, M. Borkovec, M.F. Benedetti, M.J. Avena, Colloids Surf. A 151 (1999) 147. [15] S.L. Stangroom, J.N. Lester, C.D. Collins, Environ. Technol. 21 (2000) 845. [16] J.-W. Ding, J.-C. Wu, Water Sci. Technol. 35 (1997) 139. [17] P. Schmitt, D. Freitag, I. Trapp, A.W. Garrison, M. Schiavon, A. Kettrup, Chemosphere 35 (1997) 55. [18] L.F. Schultz, T.M. Young, Environ. Toxicol. Chem. 18 (1999) 1710. [19] Y.-H. Yang, L.K. Koopal, Colloids Surf. A 151 (1999) 201. [20] Y. Laor, M. Rebhun, Environ. Sci. Technol. 36 (2002) 955. [21] C.L. Lee, L.J. Kuo, H.L. Wang, P.C. Hsieh, Water Res. 37 (2003) 4250. [22] M. Stalmans, E. Matthijs, N.T. De Oude, Water Sci. Technol. 43 (1991) 1. [23] J. Waters, T.C.J. Feijtel, Chemosphere 30 (1995) 1939. [24] H.R. Rogers, Sci. Tot. Environ. 185 (1996) 3. [25] C.A. Moody, J.A. Field, Environ. Sci. Technol. 34 (2000) 3864. [26] M.R. Taha, I.H. Soewarto, R.J. Acar, R.J. Gale, M.E. Zappi, Water Air Soil Pollut. 100 (1997) 33. [27] J.H. Harwell, D.A. Sabatini, R.C. Knox, Colloids Surf. A 151 (1999) 255. [28] C. Wegner, M. Hamburger, Environ. Sci. Technol. 36 (2002) 3250. [29] E. Tombacz, K. Varga, F. Szanto, Colloid Polym. Sci. 266 (1988) 734. [30] E. Tombacz, I. Regdon, in: N. Senesi, T.M. Miano (Eds.), Humic Substances in the Global Environment and Implications on Human Health, Elsevier Science, Amsterdam, 1994, p. 139.
367
[31] S.J. Traina, D.C. McAvoy, D.J. Versteeg, Environ. Sci. Technol. 30 (1996) 1300. [32] W.H. Otto, D.J. Britten, C.K. Larive, J. Colloid Interface Sci. 261 (2003) 508. [33] A.F.Y. Adou, V.S. Muhandiki, Y. Shimizu, S. Matsui, Water Sci. Technol. 43 (2001) 1. [34] J.C.A. de Wuilloud, R.G. Wuilloud, B.B.M. Sadi, J.A. Caruso, Analyst 128 (2003) 453. [35] R.L. Revia, G.A. Makharadze, Talanta 48 (1999) 409. [36] K. Shirahama, in: J.C.T. Kwak (Ed.), Polymer–Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 143. [37] P. Linse, L. Piculell, P. Hansson, in: J.C.T. Kwak (Ed.), Polymer– Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 193. [38] L. Pikulell, in: J.C.T. Kwak (Ed.), Polymer–Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 65. [39] C.J. Milne, D.G. Kinniburgh, W.H. Van Riemsdijk, E. Tipping, Environ. Sci. Technol. 37 (2003) 958. [40] M.J. Avena, A.P.W. Vermeer, L.K. Koopal, Colloids Surf. A 151 (1999) 213. [41] E.I. Colichman, J. Am. Chem. Soc. 72 (1950) 1834. [42] T. Shimizu, M. Seki, J.C.T. Kwak, Colloids Surf. 20 (1986) 289. [43] D.M. Bloor, W.M.Z. Wan-Yunus, W.A. Wan-Badhi, Y. Li, J.F. Hozwarth, Langmuir 11 (1995) 334. [44] D.M. Bloor, H.K.O. Mwakibete, E. Wyn-Jones, J. Colloid Interface Sci. 178 (1996) 334. [45] S.M. Ghoreishi, Y. Li, J.F. Holzwarth, E. Khoshdel, J. Warr, D.M. Bloor, E. Wyn-Jones, Langmuir 15 (1999) 1938. [46] S.M. Ghoreishi, Y. Li, D.M. Bloor, J. Warr, E. Wyn-Jones, Langmuir 15 (1999) 4380. [47] Metrohm Application Bulletin 233/3 e. [48] N.M. Van Os, J.R. Haak, L.A.M. Rupert, Physicochemical Properties of Selected Anionic, Cationic and Nonionic Surfactants, Elsevier Science, Amsterdam, 1993. [49] P. Mukerjee, K.J. Mysels, Critical Micelle Concentrations of Aqueous Surfactant Systems, National Bureau of Standards, Washington, DC, 1971, Report NSRDS-NBS 36. [50] A.P. Rodenhiser, J.C.T. Kwak, in: J.C.T. Kwak (Ed.), Polymer– Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 1. [51] T. Shimizu, J.C.T. Kwak, Colloids Surf. A 82 (1994) 163. [52] T. Shimizu, Colloids Surf. A 84 (1994) 239. [53] T. Shimizu, Colloids Surf. A 94 (1995) 115. [54] K. Kogej, J. Skerjanc, in: T. Radeva (Ed.), Physical Chemistry of Polyelectrolytes, in: Surfactant Science Series, vol. 99, Dekker, New York, 2001, p. 973. [55] E.D. Goddar, in: E.D. Goddar, K.P. Ananthapadmanabhan (Eds.), Interactions of Surfactants with Polymers and Proteins, CRC Press, Boca Raton, FL, 1992, p. 123. [56] J.P. Gong, Y. Osada, J. Phys. Chem. 99 (1995) 10971. [57] M.J. Avena, L.K. Koopal, Eviron. Sci. Technol. 33 (1999) 2739. [58] M.J. Avena, L.K. Koopal, W.H. Van Riemsdijk, J. Colloid Interface Sci. 217 (1999) 37. [59] E.D. Goddar, K.P. Ananthapadmanabham, in: J.C.T. Kwak (Ed.), Polymer–Surfactant Systems, in: Surfactant Science Series, vol. 77, Dekker, New York, 1998, p. 21. [60] T.P. Goloub, L.K. Koopal, Langmuir 13 (1997) 673. [61] P. Somasundaran, D.W. Fuerstenau, J. Phys. Chem. 70 (1966) 90. [62] L.K. Koopal, T.P. Goloub, in: R. Sharma (Ed.), Surfactant Adsorption and Surface Solubilization, in: ACS Symposium Series, vol. 615, American Chemical Society, Washington, DC, 1995, p. 78. [63] W.H. Van Riemsdijk, L.K. Koopal, in: J. Buffle, H.P. van Leeuwen (Eds.), Environmental Particles, in: Environmental Analytical and Physical Chemistry Series, vol. 1, Lewis Publishers, Boca Raton, FL, 1992, p. 455. [64] A.P.W. Vermeer, L.K. Koopal, J. Colloid Interface Sci. 212 (1999) 176.