9998
Langmuir 2005, 21, 9998-10003
Complexation of Hydrophobically Modified Polyelectrolytes with Surfactants: Anionic Poly(maleic acid/octyl vinyl ether)/Anionic Sodium Dodecyl Sulfate† Puspendu Deo, Namita Deo, and P. Somasundaran* NSF IUCR Center for Advanced Studies in Novel Surfactants, Langmuir Center for Colloids and Interfaces, Columbia University, New York 10027 Received February 28, 2005. In Final Form: July 1, 2005 Interactions of surfactants with hydrophobically modified polyelectrolytes in aqueous solutions are important in several applications such as detergents, cosmetics, foods, and paints. Fundamental questions arise on the mechanisms of complexation of the polyelectrolyte and surfactant that control their behavior. In this work, the complexation was studied by examining interactions in aqueous solutions of a hydrophobically modified polymer, poly(maleic acid/octyl vinyl ether) (PMAOVE), with sodium dodecyl sulfate (SDS) by monitoring viscosity, pyrene solubility, light scattering, and analytical ultracentrifugation. When the anionic surfactant SDS was added to aqueous solutions of the similarly charged polymer PMAOVE, the surfactant was incorporated into the hydrophobic nanodomains of PMAOVE even far below the cmc of the surfactant. On the basis of viscosity, pyrene solubility, and analytical ultracentrifugation data, it is proposed that PMAOVE undergoes structural unfolding and at higher SDS concentrations mixed micelles are formed.
Introduction Interactions of surfactants with polymers in aqueous solutions have been investigated for several years.1,2 Polymers such as hydrophobically modified polyelectrolytes (HMP) have attracted substantial interest recently because of their unique associative and rheological properties.3-7 These interactions are considered to be the result of complexes formed between the surfactant and the polymer due to electrostatic and hydrophobic forces. These complexes, besides having important practical applications for detergents, cosmetics, food, and paints, also raise some fundamental questions about the polymersurfactant interactions that control their behavior.1,2,8-11 Prior studies have been devoted mainly to systems with attractive interactions, either weak (e.g., between nonionic polymers and anionic surfactants) or strong (between oppositely charged polyelectrolytes and surfactants).1,2,12 Although the importance of attractive hydrophobic interactions was recognized a long time ago, the role of such interactions between HMP and surfactants has been †
Part of the Bob Rowell Festschrift special issue. * Corresponding author. E-mail:
[email protected].
(1) Interactions of Surfactants with Polymers and Proteins; Goddard, E. D., Ananthapadmanabhan, K. P., Eds.; CRC Press: Boca Raton, FL, 1993. (2) Kwak, J. C. T. Polymer-Surfactant Systems; Marcel Dekker: New York, 1998. (3) Qiu, Q.; Lou, A.; Somasundaran, P.; Pethica, B. A. Langmuir 2002, 18, 5921. (4) Hsu, J.-L.; Strauss, U. P. J. Phys. Chem. 1987, 91, 6238. (5) Barbieri, B. W.; Strauss, U. P. Macromolecules 1985, 18, 411. (6) Zdanowicz, V. S.; Strauss, U. P. Macromolecules 1993, 26, 4770. (7) Bokias, G.; Hourdet, D.; Iliopoulos, I. Macromolecules 2000, 33, 2929. (8) Sen, S.; Sukul, D.; Dutta, P.; Bhattacharyya, K. J. Phys. Chem. B 2002, 106, 3763. (9) Deo, P.; Jockusch, S.; Ottaviani, M. F.; Moscatelli, A.; Turro, N.; Somasundaran, P. Langmuir 2003, 19, 10747. (10) Nakagaki, M.; Handa, T. In ACS Symposium Series 253; Rosen, M. J., Ed.; American Chemical Society: Washington, DC, 1984; p 73. (11) Wesley, R. D.; Cosgrove T., Thompson L., Armes S. P.; Baines F. L. Langmuir 2002, 18, 5704. (12) Hayakawa, K.; Kwak, J. C. T. J. Phys. Chem. 1982, 86, 3866.
Chart 1. Structures of the Polymer {Poly(maleic acid/octyl vinyl ether)} (PMAOVE), Surfactant {Sodium Dodecyl Sulfate} (SDS), and Probe Molecule Pyrene
studied systematically only during the past decade.13-22 In this article, we consider the case when surfactant is added to solutions where HMP, poly(maleic acid/octyl vinyl ether) (PMAOVE) (Chart 1), has already formed nanodomains. The binding surfactant of the same charges sodium dodecyl sulfate (SDS) to HMP is very strongs starts at low concentration. To study these interactions, we used a multi-technique approach involving measurements of viscosity, light scattering, pyrene solubility, binding, and analytical ultracentrifugation. (13) Piculell, L.; Guillemet, F.; Thuresson, K.; Shubin, V.; Ericsson, O. Adv. Colloid Interface Sci. 1996, 63, 1. (14) Philippova, O. E.; Hourdet, D.; Audebert, R.; Khokhlov, A. R. Macromolecules 1996, 29, 2822. (15) Bromberg, L.; Temchenko, M.; Colby, R. H. Langmuir 2000, 16, 2609. (16) Colby, R. H.; Plucktaveesak, N.; Bromberg, L. Langmuir 2001, 17, 2937. (17) Maltesh, C.; Somasundaran, P. Colloids Surf. 1992, 69, 167. (18) Loyen, K.; Iliopoulos, I.; Audebert, R.; Olsson, U. Langmuir 1995, 11, 1053. (19) Hansson, P.; Lindman, B. Curr. Opin. Colloid Interface Sci. 1996, 1, 604. (20) Iliopoulos, I. Curr. Opin. Colloid Interface Sci. 1998, 3, 493. (21) Tan, H.; Tam, K. C.; Jenkins, R. D. Langmuir 2000, 16, 5600. (22) Iliopoulos, I.; Wang, T. K.; Audebert, R. Langmuir 1991, 7, 617.
10.1021/la050539g CCC: $30.25 © 2005 American Chemical Society Published on Web 08/06/2005
Complexation of Polyelectrolytes with Surfactants
Langmuir, Vol. 21, No. 22, 2005 9999
Experimental Section Materials. Polymers. The hydrophobically modified polymer PMAOVE, provided by International Specialty Products, Inc., was synthesized using free-radical polymerization of a 1:1 mole ratio of maleic anhydride and octyl vinyl ether in toluene with Vazo-69 (azo bis-valeryl nitrile) as the initiator. The products were purified twice by first dissolving in acetone (5 wt %), followed by precipitation with an excess of tert-butyl alcohol (40 times in volume). Residual solvent was removed in vacuum at 50 °C until the product had a constant mass. The anhydride moiety of the polymer was then hydrolyzed in triple-distilled water to make an approximately 5 wt % solution. The solution was stirred at 500 rpm at 70 °C for about 12 h and then freeze dried. As determined by gel permeation chromatography, the weightaverage molecular weight (Mw) was 160 000 Da with a polydispersity index of 1.23. Probe and Additives. The fluorescence probe pyrene (Aldrich), sodium dodecyl sulfate (SDS) (Aldrich, 99+% pure), NaCl (Fluka, 99.5% pure), and solutions of 0.1 N hydrochloric acid (Fisher Scientific) and 0.1 N NaOH (Fisher Scientific) were used as received. Triple-distilled water was used in all experiments. Methods. Viscosity Measurements. A calibrated capillary viscometer (Canon Instrument) was used for measuring the relative viscosity based on that of the triple-distilled water at 25 ( 0.05 °C. The viscometer was cleaned with chromic acid and triple-distilled water and thoroughly dried with acetone before the measurements. Before every measurement, the efflux time for triple-distilled water was checked for reproducibility. Pyrene Solubility. Excess pyrene crystals were added to solutions of interest contained in 20 mL glass vials. The samples were then wrapped with aluminum foil to avoid possible decomposition by light and subjected to wrist-action shaking at 250 rpm for a week. No further solubilization was detected at longer times. The samples were centrifuged at 4000 rpm for half an hour, and the supernatant was analyzed for pyrene concentration by UV absorbance at 335 nm using a Perkin-Elmer Lambda 25 UV-vis spectrophotometer. Dynamic Light Scattering. Dynamic light scattering was performed using a Brookhaven research-grade system with BI900AT correlator and BI-200SM goniometer with adjustable angles of detection from 15 to 155°. A water-cooled Lexel argon laser light source was used at a wavelength of 488 Å. The samples were temperature controlled to (1 °C, and decalin, a refractive index matching liquid, was used to reduce light bending at the glass interfaces. To minimize the dust effect encountered frequently in lightscattering measurements, the sample solutions were filtered through a 0.2 µm Nalgene membrane prior to use. The filtration process was found to produce no detectable effect on the polymer concentration, and because of the relatively large size, it is assumed not to change the conformation of the polymer or the polymer-surfactant complexes. The sample was thermostated for at least 30 min to equilibrate the system. All of the measurements were made in dynamic mode (at 90° detection) (i.e., the instrument measures the diffusion coefficient of the molecules and back calculates the effective diameter assuming a spherical shape). Binding Isotherm. An ultrafiltration technique23 was used to obtain the binding density of SDS to PMAOVE. Solutions at desired polymer/surfactant concentrations were filtered through a 50 000 molecular weight cutoff membrane (Millipore Co.) by subjecting them to centrifugal force at 4000 rpm for 2 min. The SDS concentration in the filtrate was analyzed using a twophase titration method24,25 with 1 mM cetyltrimethylammonium bromide (CTAB) as the cationic titrant, dimidium bromidedisulfine blue as the indicator, and chloroform as the organic phase. The binding density is expressed as the ratio of SDS molecules depleted from solution per PMAOVE monomer. (23) Dubin, Paul L.; Rigsbee, D. R.; Gan, L. M.; Fallon, M. A. Macromolecules 1988, 21, 2555. (24) Reid, V. W.; Longman, George F.; Heinerth, E. Tenside 1967, 4, 292. (25) Rosen, M. J.; Goldsmith, H. A. Systematic Analysis of SurfaceActive Agents, 2nd ed.; Chemical Analysis; Wiley-Interscience: New York, 1972; Vol. 12, p 624.
Figure 1. Relative viscosity of aqueous solutions in the absence (b) and presence (a) of 0.1% (wt/wt) PMAOVE as a function of SDS concentration. SD ) (2%.
Figure 2. Pyrene solubility of aqueous solutions in the absence (b) and presence (a) of 0.1% (wt/wt) PMAOVE as a function of SDS concentration. SD ) (2%. Ultracentrifugation Study. A Beckman Optima XL-1 analytical ultracentrifuge with scanning optics with interference systems was used to perform sedimentation velocity experiments. The interference optical system provides the total concentration by measuring the refractive index difference between the sample cell and the reference cell at each radial position as indicated by the vertical displacement of a set of evenly spaced horizontal fringe. The running condition was set at a motor speed 40 000 rpm and a temperature of 25 °C. The direct boundary fitting method was applied to get the average sedimentation coefficient from the solute boundary shift on the absorbance spectrum versus rotor radius. Also, time-derivative sedimentation analysis was used to get qualitative information about the heterogeneous nature of the solution while interacting with SDS.
Results and Discussion The interaction of PMAOVE with SDS was investigated first by monitoring the dependence of the viscosity on SDS concentration in the presence and absence of PMAOVE as shown in Figure 1. In the absence of PMAOVE, the viscosity increases slightly above the cmc (8 mM) because of the formation of micelles (Figure 1b). In the presence of 0.1% PMAOVE, the viscosity increased at an SDS concentration of ca. 2 mM (Figure 1a), indicating a change in the conformation of the polymer structure or aggregation. Upon further increase in the SDS concentration, the viscosity continued to increase, suggesting gradual polymer restructuring and the formation of mixed micelles of PMAOVE and SDS. These results are in accord with that expected from an increase in the effective volume occupied by PMAOVE-SDS mixed micelles compared to that occupied by the pure PMAOVE (coiled structure).
10000
Langmuir, Vol. 21, No. 22, 2005
Figure 3. Correlation of the binding density of SDS to PMAOVE with pyrene solubility in binary mixtures of PMAOVE-SDS.
Figure 4. Effect of PMAOVE-SDS complexation on polymer configuration studied by dynamic light scattering.
The interaction of PMAOVE with monomers of SDS involves hydrophobic forces between the polymer and the surfactant species, with the possible formation of hydrophobic clusters of SDS molecules along the polymer chain. Figure 2 shows the pyrene solubility in SDS solutions with and without PMAOVE. Pyrene is barely soluble in water, but when hydrophobic microdomains form, its solubility increases in proportion to the volume of the nanodomains. Hence, it can be concluded that PMAOVE
Deo et al.
forms nanodomains at all concentrations with 1000 ppm PMAOVE solublizing 10-5 M pyrene. This is demonstrated in SDS solutions where a noticeable increase in pyrene solubility is detected only above the cmc and a linear slope is obtained in that region. In the presence of 1000 ppm PMAOVE, the solubility of pyrene is increased at an SDS concentrations of 2 mM, indicating the restructuring of polymer and the formation of different types of nanoaggregates. Further addition of SDS to the solutions enhances pyrene solubilization, suggesting the formation of small nanoaggregates of polymer hydrophobic groups and SDS molecules. These results are in agreement with the observed increase in the pyrene solublization by a large number of nanoaggregates of polymer hydrophobic groups and SDS molecules on the polymer backbone compared to the PMAOVE nanodomains. The pyrene solubility in the presence of PMAOVE with added SDS shows two distinct regions: an initial region where there is little increase in pyrene solubility, followed by a region with a continuous linear increase in the pyrene solubility. This observation can be compared to the binding of SDS to PMAOVE, determined by the ultrafiltration method, with the solubilization of pyrene in the complex as shown in Figure 3.The saturation of SDS binding correlates with the onset of the linear increase in pyrene solubilization at around 12 mM SDS (as shown by a dotted line in Figure 3). The initial pyrene solubilization region is associated with the SDS molecule interacting with the nanodomains of PMAOVE, whereas the linear region results from the formation of mixed micelles on the polymer backbone. At the start of the linear region, the binding is approximately 0.4 bound SDS molecule per PMAOVE monomer, which is sufficient to reorient the polymer to form mixed nanoaggregates. At the saturation point of the PMAOVE/SDS complex, the binding is approximately 2 bound SDS molecules per PMAOVE monomer. Hydrodynamic measurements were performed using dynamic light scattering to determine the size of the PMAOVE-SDS complexes as function of SDS concentration, and the results obtained are given in Figure 4. With the initial increase in the SDS concentration, the computed diameter of the PMAOVE-SDS complexes remained
Figure 5. Typical sedimentation velocity experiment surfactant micelles. Scanning started after 5 min running with a 2.5 h scanning interval.
Complexation of Polyelectrolytes with Surfactants
Langmuir, Vol. 21, No. 22, 2005 10001
Figure 6. Distribution of apparent sedimentation coefficient: (a) 25 mM SDS; (b) 1000 ppm PMAOVE; and (c) 1000 ppm PMAOVE + 1.5 mM SDS.
Figure 7. Distribution of the apparent sedimentation coefficient: (a) 1000 ppm PMAOVE + 2.2 mM SDS and (b) 1000 ppm PMAOVE + 3.5 mM SDS.
constant up to 1.5 mM. A dramatic increase in the hydrodynamic diameter of the PMAOVE-SDS complex was observed above 2 mM, and further addition of SDS caused a gradual increase in the hydrodynamic diameter reaching a plateau at 12 mM SDS, suggesting restructuring of PMAOVE and possible formation of mixed micelles of hydrophobic groups of polymer and SDS molecules on the polymer backbone. With the addition of more surfactant, there was very little change in the diameter. The hydrodynamic measurement results are in agreement with the results obtained by means of viscosity and pyrene solubility. The above hypothesis was further tested using an analytical ultracentrifugation study and by correlating the ultracentrifuge results with the relative viscosity and
pyrene solublization data. Sedimentation velocity processes are caused by the centrifugal force field acting on a solute particle, leading to the movement of the solute toward the bottom of the centrifuge cell. The centrifugal force on the solute is partially counterbalanced by the buoyant force of the displaced solvent. The net sedimentation behavior of the solute particle in a centrifugal field is described by the Svedberg equation:26
S)
M(1 - uˆ F) V ) 2 Nf ωr
(1)
The above equation suggests that the rate of sedimentation v depends on several factors: the strength of the
10002
Langmuir, Vol. 21, No. 22, 2005
Deo et al.
Figure 8. Distribution of the apparent sedimentation coefficient: (a) 1000 ppm PMAOVE +5 mM SDS; (b) 1000 ppm PMAOVE + 6.5 mM SDS; (c) 1000 ppm PMAOVE + 10 mM SDS; (d) 1000 ppm PMAOVE + 15 mM SDS; and (e) 1000 ppm PMAOVE + 25 mM SDS.
centrifugal force field, ω2r (where r is the radial distance from the center of rotation); the molecular mass M; the molecular size and shape (related to the frictional coefficient coefficient f); the density of the solvent, F; and the partial specific volume of the solvent, uˆ . The combination of sedimentation and diffusion in the ultracentrifuge cell is described in terms of the flow, F
F ) Sω2rc - D
dc dr
(2)
where S is the sedimentation coefficient as described in eq 1, D is the diffusion coefficient, c is the solute concentration, and dc/dr is the solute concentration gradient in the form of a boundary or band when diffusional spreading begins to occur. The sedimentation velocity experiments done in this study refer to SDS in the absence of PMAOVE, PMAOVE in the absence of SDS, 1.5 mM SDS + PMAOVE, 2.2 mM SDS + PMAOVE, 3.5 mM SDS + PMAOVE, 5 mM SDS + PMAOVE, 6.4 mM SDS + PMAOVE, 10 mM SDS + PMAOVE, 15 mM SDS + PMAOVE, and 25 mM SDS + PMAOVE. Sedimentation velocity measurements were made at 30 000 rpm using the boundary procedure as illustrated in Figure 5. Figure 5 shows the typical absorbance distribution during the sedimentation velocity experiment of polymers, surfac(26) Hansen, J. C.; Lebowwitz, J.; Demeler, B. Biochemistry 1994, 15, 13155.
tants, and polymer/surfactant complexes. These scans were taken starting 5 min after initiating a run at 30 000 rpm and then every 2.5 h thereafter. The time-derivative concentration curve gives evidence of the nature of the PMAOVE-SDS complexes. The distributions of the apparent sedimentation coefficient of SDS, PMAOVE, and PMAOVE-SDS complexes are given in Figures 6-8 at seven different SDS concentrations. Sedimentation curves for SDS micelles, PMAOVE without SDS, and PMAOVE with an SDS concentration up to 1.5 mM (Figure 6a-c) show only a single peak, suggesting only a monomodal molecular distribution. PMAOVE with SDS concentrations at 2.2 and 3.5 mM (Figure 7a and b) exhibit two peaks, and this is attributed to the presence of two types of species: a partially unfolded PMAOVE and a completely unfolded PMAOVE. With increasing SDS concentration (Figure 8a-e), all PMAOVE will be unfolded and hence a monomodal distribution appears, but with a sedimentation coefficient much higher than that of PMAOVE alone. Because the sedimentation coefficient is proportional to the mass of the species, the mixed-micelle species formed are proposed to be larger in size than the nanodomains formed by PMAOVE and SDS micelles. Discussion On the basis of the viscosity, pyrene solubility, binding isotherm, and ultracentrifuge results, we propose a model illustrated in Figure 9 for PMAOVE-SDS interactions.
Complexation of Polyelectrolytes with Surfactants
Langmuir, Vol. 21, No. 22, 2005 10003
Figure 9. Effects of SDS on the structural reconformation of PMAOVE (0.1% wt/wt).
PMAOVE in aqueous solutions forms a coiled structure, with the nonpolar chains forming hydrophobic nanodomains.3 The fluorescence polarity parameter with pyrene as the probe reveals that the packing of the hydrophobic chains is tight and the polarity of the nanodomains is lower than that of SDS micelles.9 When small amounts of SDS were added ([SDS] < 2 mM) to PMAOVE solutions, only negligible changes in the properties (viscosity, pyrene solubility, and ultracentrifuge) were observed. Above approximately 2 mM SDS, significant changes in the above properties were observed. Because the viscosity (Figure 1a), and pyrene solubility (Figure 2a) increased and two peaks appeared on the sedimentation distribution curves (Figure 7a and b), it is concluded that SDS is incorporated into the hydrophobic microdomains of PMAOVE at an SDS concentration of 2 mM. The sedimentation coefficients observed for PMAOVE-SDS complexes (SDS > 2 mM) were high compared to those for PMAOVE and SDS micelles. This behavior is obviously caused by an increase in anisotropy accompanying the restructuring of the polymer. The SDS concentration of 2 mM was therefore termed the critical complexation concentration (formation of mixed micelles of SDS and the hydrophobic n-octyl chains of PMAOVE). As the SDS concentration is increased above 2 mM, more SDS is incorporated into the PMAOVE-SDS complex to form mixed micelles of hydrophobic groups of PMAOVE and SDS on the PMAOVE backbone (Figure 9). The SDS concentration of 12 mM corresponds to approximately 2 SDS molecules bound (from binding isotherm measurements) per n-octyl chain of the PMAOVE
(mixed micelles are mixtures of SDS monomers and hydrophobic side chains of PMAOVE). The SDS concentration of 12 mM is termed the saturation concentration. Conclusions Interactions of a hydrophobically modified anionic polymer (PMAOVE) with a similarly charged surfactant (SDS) were studied using viscosity, pyrene solubility, the binding isotherm, and analytical ultracentrifugation techniques. The n-octyl chains of PMAOVE by themselves form hydrophobic nanodomains in aqueous solutions. If a surfactant of the same charge (SDS) is added to PMAOVE solutions, then it gets incorporated into the existing hydrophobic polymer nanodomains. Two inflection points were observed on viscosity, pyrene solubility, binding isotherm, and analytical ultracentrifugation curves corresponding to the critical complexation concentration (formation of mixed micelles) and the saturation concentration (saturation of the polymer with SDS molecules). The above process involves a structural reconformation in the form of unfolding of PMAOVE. It is believed that the above findings on the interaction of PMAOVE with SDS should aid in pharmaceutical, cosmetic, and personal care applications of hydrophobically modified polymers. Acknowledgment. We acknowledge financial support from the Industrial/University Cooperative Research Center (IUCR) at Columbia University through National Science Foundation grant EEC-9804618 and the industrial sponsors of the IUCR Center. LA050539G
