Physical Properties of Aqueous Solutions of a Thermo-Responsive

Department of Physics, Institute for Energy Technology, Post Office Box 40, ... Norway, Department of Chemistry, University of Oslo, Post Office Box 1...
0 downloads 0 Views 232KB Size
Download PDF
8010

Langmuir 2005, 21, 8010-8018

Physical Properties of Aqueous Solutions of a Thermo-Responsive Neutral Copolymer and an Anionic Surfactant: Turbidity and Small-Angle Neutron Scattering Studies Ce´line Galant,† Anna-Lena Kjøniksen,‡ Kenneth D. Knudsen,† Geir Helgesen,† Reidar Lund,§ Antti Laukkanen,| Heikki Tenhu,| and Bo Nystro¨m*,‡ Department of Physics, Institute for Energy Technology, Post Office Box 40, N-2027 Kjeller, Norway, Department of Chemistry, University of Oslo, Post Office Box 1033, Blindern, N-0315 Oslo, Norway, IFF-Forschungszentrum Ju¨ lich, D-52425 Ju¨ lich, Germany, and Laboratory of Polymer Chemistry, University of Helsinki, PB 55, FIN-00014 Helsinki, Finland Received January 5, 2005. In Final Form: June 23, 2005 Aqueous mixtures of the anionic sodium dodecyl sulfate (SDS) surfactant and thermo-responsive poly(N-vinylcaprolactam) chains grafted with ω-methoxy poly(ethylene oxide) undecyl R-methacrylate (PVCLg-C11EO42) have been characterized using turbidimetry and small-angle neutron scattering (SANS). Turbidity measurements show that the addition of SDS to a dilute aqueous copolymer solution (1.0 wt %) induces an increase of the cloud point (CP) value and a decrease of the turbidity at high temperatures. In parallel, SANS results show a decrease of both the average distance between chains and the global size of the objects in solution at high temperatures as the SDS concentration is increased. Combination of these findings reveals that the presence of SDS in the PVCL-g-C11EO42 solutions (1.0 wt %) promotes the formation of smaller aggregates and, consequently, leads to a more homogeneous distribution of the chains in solution upon heating of the mixtures. Moreover, the SANS data results show that the internal structure of the formed aggregates becomes more swollen as the SDS concentration increases. On the other hand, the addition of moderate amounts of SDS (up to 4 mm) to a semidilute copolymer solution (5.0 wt %) gives rise to a more pronounced aggregation as the temperature rises; turbidity and SANS studies reveal in this case a decrease of the CP value and an increase of the scattered intensity at low q. The overall picture that emerges from this study is that the degree of aggregation can be accurately tuned by varying parameters such as the temperature, level of surfactant addition, and polymer concentration.

* To whom correspondence should be addressed. Telephone: +4722855522. Fax: +47-22855441. E-mail: [email protected]. † Institute for Energy Technology. ‡ University of Oslo. § IFF-Forschungszentrum Ju ¨ lich. | Laboratory of Polymer Chemistry.

the presence of an ionic surfactant6,7 reveal that, with an increasing surfactant concentration, the ratio between the number of hydrophobic polymer groups and the surfactant molecules inside the mixed clusters decreases until pure surfactant micelles and macromolecules saturated with the surfactant are present in the solution. Surfactants are believed to bind to the hydrophobic segments of the polymer and progressively decrease the amphiphilic character of the polymer, i.e., increase its solubility in water. On the other hand, extensive studies on amphiphilic block copolymers capable of forming stable micelles in aqueous environment, such as PS-b-PEO,8-10 disclose that the addition of the surfactant (anionic or cationic) does not necessarily destroy the micellar structure of the polymer. In some cases, depending on the composition of the surfactant and the polymer, the surfactant may absorb into the corona of the micelles and actually enhance the stability of the micellar structures. This behavior typically takes place in the lower concentration regime of the surfactant, and at higher concentrations, partial destabilization of the mixed micelles may occur. In addition,

(1) Goddard, E. D. In Interactions of Surfactants with Polymers and Proteins; Goddard, E. D., Ananthapadmananabhan, K. P., Eds.; CRC Press: Boca Raton, FL, 1993; p 123. (2) Brackman, J. C.; Engberts, J. B. F. N. Chem. Soc. Rev. 1993, 85, 1. (3) Lindman, B.; Carlsson, A.; Gerdes, S.; Karlstro¨m, G.; Piculell, L.; Thalberg, K.; Zhang, K. In Food Colloids and Polymers: Stability and Mechanical Properties; Walstra, P., Dickinson, E., Eds.; The Royal Society of Chemistry: London, U.K., 1993; p 113-125. (4) Macromolecular Complexes in Chemistry and Biology; Dubin, P., Bock, J., Davies, R. M., Schulz, D. N., Thies, C., Eds.; Springler-Verlag: Berlin, Germany, 1994. (5) Polymer-Surfactant Systems; Kwack, J. C. T., Ed.; Marcel Dekker: New York, 1998; Vol. 77.

(6) Almgren, M.; van Stam, J.; Lindblad, C.; Li, P.; Stilbs, P.; Bahadur, P. J. Phys. Chem. 1991, 95, 5677. (7) Hecht, E.; Mortensen, K.; Gradzielski, M.; Hoffmann, H. J. Phys. Chem. 1995, 99, 4866. (8) Bronstein, L. M.; Chernyshov, D. M.; Timofeeva, G. I.; Dubrovina, L. V.; Valetsky, P. M.; Obolonkova, E. S.; Khokhlov, A. R. Langmuir 2000, 16, 3626. (9) Bronstein, L. M.; Chernyshov, D. M.; Timofeeva, G. I.; Dubrovina, L. V.; Valetsky, P. M.; Khokhlov, A. R. J. Colloid Interface Sci. 2000, 230, 140. (10) Bronstein, L. M.; Chernyshov, D. M.; Vorontsov, E.; Timofeeva, G. I.; Dubrovina, L. V.; Valetsky, P. M.; Kazakov, S.; Khokhlov, A. R. J. Phys. Chem. 2001, 105, 9077.

Introduction An increasing interest is generated today toward systems involving nonionic water-soluble amphiphilic polymers and ionic surfactants, both from fundamental and technological points of view.1-5 Such systems play an important part in diverse industrial fields such as paints, pharmaceuticals, and cosmetics. Their behavior is governed by a subtle interplay between hydrophilic, hydrophobic, and electrostatic interactions. The interactions between amphiphilic polymer chains and ionic surfactants give rise to the formation of micelle-like clusters involving substituents from one or several polymer chains. Studies of dilute solutions of a nonionic amphiphilic polymer in

10.1021/la050036a CCC: $30.25 © 2005 American Chemical Society Published on Web 07/21/2005

Turbidity and Small-Angle Neutron Scattering Studies

Figure 1. Chemical structure of the grafted PVCL-g-C11EO42 copolymer.

studies of semidilute solutions of an amphiphilic polymer in the presence of moderate levels of an ionic surfactant11,12 report on interchain aggregation promoted by the surfactant binding to the polymer. In this case, the mixed micelles either act as junctions or strengthen already existing connections between segments of different polymer chains. At this stage, the surfactant-promoted interchain interactions induce an enhancement of the viscosity. The complexes of neutral thermo-responsive polymers and ionic surfactants have been shown to be particularly suitable for the design of novel systems with interesting structural properties, because physical crosslinking effects can be influenced both by the surfactants and by temperature.13 Such polymers, with a lower critical solution temperature, become more hydrophobic with an increase in temperature and, consequently, offer better nuclei for surfactant self-assembly at higher temperatures. Recently,14 we reported on the characterization in water (without any additives) of a neutral thermo-responsive copolymer composed of N-vinylcaprolactam and ω-methoxy poly(ethylene oxide) undecyl R-methacrylate (PVCLg-C11EO42), whose structure is depicted in Figure 1. The interest in this copolymer is connected with the fact that it is thermo-sensitive, water-soluble, and biocompatible.15 Therefore, it can find applications in biochemistry, medicine, and pharmaceuticals, for example, in the development of drug-delivery systems. By combination of several techniques, including turbidimetry, rheology, small-angle neutron scattering (SANS), and dynamic light scattering (DLS), we reported on shrinking of PVCL-g-C11EO42 coils and intermolecular aggregation in water upon an increase in temperature above the cloud point (CP) value. These findings were explained by the enhancement of the hydrophobic interactions of the nonpolar PVCL groups with a rise in temperature. In the present study, we investigate the effect of adding the anionic surfactant SDS to PVCL-g-C11EO42 solutions by using turbidity and SANS techniques. SANS is a powerful technique for revealing global and local structures of aggregates in solution because it provides a direct access to the different length scales of the systems. At a scale smaller than the aggregates size, a specific analysis of the SANS data collected at q > 1/Rg (Rg is the radius of gyration of the aggregates) provides information about the conformation of the local structures in the system. At larger scales (i.e., at smaller q values), SANS allows us to determine the effective interactions between aggregates and to estimate their average size. Moreover, in the case of heterogeneous aggregates, the contrast matching (11) Piculell, L.; Thuresson, K.; Ericsson, O. Faraday Discuss. 1995, 101, 307. (12) Kjøniksen, A.-L.; Nystro¨m, B.; Lindman, B. Macromolecules 1998, 31, 1852. (13) Makhaeva, E. E.; Tenhu, H.; Khokhlov, A. R. Macromolecules 1998, 31, 6112. (14) Kjøniksen, A.-L.; Laukkanen, A.; Galant, C.; Knudsen, K. D.; Tenhu, H.; Nystro¨m, B. Macromolecules 2005, 38, 948. (15) Vihola, H.; Laukkanen, A.; Valtola, L.; Tenhu, H.; Hirvonen, J. Biomaterials 2005, 26, 3055.

Langmuir, Vol. 21, No. 17, 2005 8011

method16 allows us to obtain the specific structures of the individual system components; the scattering from one of the components can be selectively suppressed by matching the scattering length density of that component to that of the solvent. Turbidity measurements make it possible to keep track of major changes in the thermodynamic conditions of the system and to determine CP as well as the tendency for macroscopic phase separation under various conditions. In the present study, we demonstrate by means of turbidimetry and SANS structural alterations in PVCL-g-C11EO42 solutions induced by surfactant addition and temperature change. Experimental Procedures Materials. The grafted copolymer PVCL-g-C11EO42 has been synthesized according to a procedure described elsewhere.14 The molecular weight of the sample used in this study has been measured by light scattering14 and is 200 000 g mol-1. The alkylpoly(ethylene oxide) content has been determined by 1H NMR and is about 15 wt %, i.e., 1.2 mol %.14 The SDS surfactant was supplied by Fluka Chemie AG, Buchs, Switzerland. It was used without further purification. A deuterated sodium dodecyl sulfate (d-SDS), purchased from Cambridge Isotope Laboratories, Andover, MA, has also been utilized in the SANS measurements. Turbidity Measurements. The turbidity of the samples was measured with a NK60-CPA Cloud Point Analyzer from Phase Technology, Richmond, British Columbia, Canada. This instrument uses a scanning diffusive light-scattering technique to determine the turbidity changes of the sample. A light beam (AlGaAs light source with a wavelength equal to 654 nm) is focused onto the sample, while directly above it, an optical system continuously monitors the scattered intensity (S) from the sample. A platinum resistance thermometer probes the temperature of the sample, and a thermoelectric device (Peltier elements) is used to cool and warm the sample with rates set to 0.2 °C/min (no effect of the heating rate on the signal was observed at low heating rates). To prevent evaporation of the solvent, the sample surface was covered with a layer of highly transparent silicon oil. More details about the functioning of the apparatus are given in the previous paper.14 The turbidity (τ) of the sample is then obtained from the relationship

τ (cm-1) ) 9.0 × 10-9S3.751

(1)

where S is the signal (S) from the Cloud Point Analyzer. Henceforth, all data from the Cloud Point Analyzer will be presented in terms of turbidity. SANS Measurements. The SANS experiments on PVCL-gC11EO42 in D2O with hydrogenated SDS were carried out with the KWS-1 instrument at the research reactor FRJ-2 at the Forschungszentrum Ju¨lich GmbH, Germany. The measurements were performed at three instrumental setups with sampledetector distances equal to 2, 8, and 20 m. The neutron wavelength was λ ) 7.0 Å with a spread of ∆λ/λ = 0.2. The resulting scattering vector, q, covered the range 2.3 × 10-3 e q e 0.15 Å-1 [q ) (4π/ λ)sin(θ/2), where θ is the scattering angle]. The SANS measurements with PVCL-g-C11EO42 and deuterated surfactant (d-SDS) were performed on the SANS installation at the IFE reactor at Kjeller, Norway. The wavelength was set by the aid of a selector (Dornier), using a high full width at half maximum for the transmitted beam (∆λ/λ ) 0.2), and maximized flux on the sample. When the wavelength is varied between 5.1 and 10.2 Å and the sample-detector distance is varied from 1.0 to 3.4 m, a q range from 8 × 10-3 to 0.25 Å-1 was covered. The raw data were corrected two-dimensionally for the empty cell scattering and background. The background scattering, primarily because of electronic noise and γ radiation, was measured using a Boron carbide sample to block the beam. Corrections for the spatial detector efficiency and sensitivity were made with a Plexiglas that also served as a secondary standard for absolute calibration. The measured intensities were normal(16) Cotton, J.-P. J. Phys. IV 1999, 9, 21.

8012

Langmuir, Vol. 21, No. 17, 2005

Figure 2. Temperature dependencies of the turbidity of PVCLg-C11EO42 (1.0 wt %)/SDS mixtures in D2O at different levels of SDS addition. ized to the monitor count rate of the incident intensity to take into account temporal variations of the output power provided by the reactor. The absolute value of the scattering cross-section was determined by calibration against vanadium (Ju¨lich) that has dΣ/dΩ ) 1.655 ( 0.005 cm-1 and direct beam measurements (IFE). After a radial average of the two-dimensional data, the measured scattering from the solvent was appropriately subtracted to obtain the coherent macroscopic scattering cross section of the system, dΣ/dΩ (q). The solutions were measured in standard 2 mm Hellma quartz cells using D2O as a solvent to minimize incoherent scattering and to maximize the scattering contrast. Transmission values were determined in situ by a neutron monitor at q ) 0. Values were in the range between 70 and 90% depending on the concentration and conditions. A mixture of 91% of D2O and 9% of H2O, of scattering length densities b equal to 6.39 × 1010 and -0.56 × 1010 cm-2, respectively, was also used as a solvent to match out the surfactant (bd-SDS ) 5.76 × 1010 cm-2). Under these conditions, only the scattering from PVCL-g-C11EO42 was observed (bPVCL-g-C11EO42 ) 0.80 × 1010 cm-2) as the surfactant concentration was varied.

Results and Discussion Turbidimetry Results. Figure 2 presents turbidity versus temperature curves for 1.0 wt % PVCL-g-C11EO42/ D2O solutions at different SDS concentrations. The characteristic feature of most of these curves is an abrupt increase in turbidity as the temperature is raised. This behavior signalizes aggregation upon heating the solutions and is interpreted as follows. When the temperature is increased above the CP value, the hydrophobicity of the PVCL segments is strongly enhanced, and this induces a “sticky” character of the copolymer molecules. Then, collision between the sticky chains in solution gives rise to the formation of large aggregates responsible for the augmented turbidity. In this study, the temperature at which the first deviation of the turbidity values from the baseline occurs is taken as the CP of the corresponding solution. In Figure 2, it is seen clearly that the turbidity transition observed for the 1.0 wt % copolymer mixtures is shifted toward higher temperatures as the SDS concentration increases. These results disclose significant interaction between the polymer and SDS even at the lowest level of surfactant addition studied in this work (i.e., 0.5 mm). The weak decrease that is seen for the turbidity of the 0.5 mm sample as the temperature rises below CP signals temperature-induced contraction of the polymer-surfactant complexes. This finding indicates that a possible

Galant et al.

Figure 3. Influence of the SDS concentration on the CP temperature of PVCL-g-C11EO42/SDS mixtures in D2O for two different copolymer concentrations. These values of CP are estimated with an accuracy of (0.1 °C.

critical aggregation concentration (cac) is even below the lowest SDS concentration considered here. In contrast, the critical micellar concentration (cmc), i.e., the concentration necessary for the SDS molecules to form selfassociated structures alone in water, is significantly higher than this, namely, 8.1 mM. In view of the chemical structure of the copolymer, it is even possible that the binding of the surfactant to the polymer chains is a noncooperative process that commences as soon as surfactant molecules are added to the system. Figure 3 demonstrates the rise of the CP temperature with an increasing surfactant concentration. Thus, the addition of increasing amounts of SDS to the solutions results in an enhancement of the copolymer solubility,17 because of better thermodynamic conditions and/or electrostatic repulsions. Moreover, Figure 2 shows that the turbidity values reached at high temperatures decrease when the surfactant concentration is increased. This result shows that an increase of the level of added SDS to the copolymer solutions leads to a reduction of the size of the formed aggregates, most likely caused by solubilization of polymer hydrophobic domains in SDS-dominated mixed micelles. For the highest SDS concentration used here (i.e., 16 mm), the measured turbidity has a low and nearly constant value at all temperatures. The fact that the turbidity value is so low is probably because, at this high surfactant concentration, the aggregates (to the extent that they are formed) are always much smaller than the light wavelength (654 nm), thus producing very little scattering. A quite different behavior is observed upon addition of SDS to a more concentrated (semidilute) copolymer solution. As shown in Figure 4 that displays turbidity versus temperature curves for 5.0 wt % PVCL-g-C11EO42/ SDS mixtures, no decrease in the turbidity values can be observed at high temperatures when the surfactant concentration rises. Besides, as shown in Figure 3, the addition of small amounts of surfactant to the semidilute system (up to 4 mm) initially leads to a reduction of the CP value, in contrast to the dilute copolymer solution (1.0 wt %), where there is a nearly linear increase in CP. This is probably because at low levels of SDS addition, in the semidilute system, the surfactant molecules act as crosslinkers between the polymer chains, giving rise to association complexes. At higher surfactant concentrations, (17) Lindman, B.; Carlsson, A.; Karlstro¨m, G.; Malmsten, M. Adv. Colloid Interface Sci. 1990, 32, 183.

Turbidity and Small-Angle Neutron Scattering Studies

Langmuir, Vol. 21, No. 17, 2005 8013

Figure 4. Temperature dependencies of the turbidity of PVCLg-C11EO42 (5.0 wt %)/SDS mixtures in D2O at different levels of SDS addition.

some of the hydrophobic microdomains are solubilized and the value of CP increases. It can also be noticed in Figure 3 that the CP values measured for the 5.0 wt % copolymer mixtures are always lower than those of the 1.0 wt %. The depression of the CP value with an increase of the copolymer concentration has previously been interpreted in terms of poorer solvent conditions and stronger association.14 It has been argued that an increase of the copolymer concentration at a constant temperature induces at the same time a closer contact between the molecules and a reduction of the solvent quality, both favoring the interchain interactions. Consequently, the probability of a chain adhering to a neighboring molecule (or to an already existing aggregate) is larger at higher polymer concentrations, leading to a lower value of the CP temperature. In our previous work,14 it was revealed by rheological measurements that the PVCL-g-C11EO42 chains alone in solution (without the surfactant) are nonentangled, even in the semidilute concentration regime (i.e., above 2.5 wt %). The initial decrease of the CP value observed in Figure 3 for the 5.0 wt % copolymer solution shows that aggregation in solution can be promoted in the presence of low levels of surfactant addition (up to 4mm). Such a behavior has already been reported in the literature for semidilute solutions of other neutral amphiphilic polymers in the presence of moderate levels of ionic surfactants.18 In such mixtures, a pronounced enhancement of viscosity can be observed. In these systems, the surfactant molecules form clusters together with hydrophobic polymer segments. The resulting mixed micelles either act as crosslinks or strengthen already existing connections between segments of different polymer chains, giving rise to the formation of larger aggregates or transient networks. However, a gradual disruption of the network is observed at higher surfactant concentrations, because the hydrophobic microdomains are progressively solubilized into ionic surfactant-dominated mixed micelles. Such a scenario can explain the turbidity behavior observed for the 5.0 wt % PVCL-g-C11EO42 mixtures. Thus, the formation of large aggregates or networks at moderate concentrations of SDS is responsible for the observed initial decrease in the CP value, whereas a gradual disruption of the formed assemblies at higher SDS concentrations leads to a subsequent increase in CP. (18) Lund, R.; Lauten, R. A.; Nystro¨m, B.; Lindman, B. Langmuir 2001, 17, 8001.

Figure 5. Influence of temperature on SANS from 1.0 wt % PVCL-g-C11EO42/D2O solutions at the levels of SDS addition indicated. Every fourth data point is shown.

A peculiar peak is observed in the CP curve (Figure 4) just below 30 °C in the presence of 2 mm and particularly 4 mm SDS. At these low surfactant concentrations, the behavior of the system is complex. As the temperature rises, the polymer becomes more hydrophobic and more surfactant adheres to it. However, at the same time, the increased hydrophobicity gives higher sticking probability between different polymer segments. There will thus be a competition between increased surfactant binding to the polymer, promoting a dissolution of aggregates and an enhanced hydrophobicity that favors aggregation. Besides, because SDS is charged, there will be a buildup of Coulomb forces as more surfactant is attached to the polymer. At a certain point, the corresponding electrostatic repulsion could induce a partial disruption of the clusters, leading to a reduction in the overall turbidity. A subsequent increase in temperature would then be necessary to obtain sufficient hydrophobicity gain for a continued increase in aggregate size. The anomalous peak may be a result of this intricate interplay. Repeated measurements (both on these and newly prepared samples) were conducted at this condition to check the effect, and the feature with the peak was found to be completely reproducible. Studies in progress19 have shown that this phenomenon is only observed in the semidilute polymer concentration regime at moderate SDS concentrations. SANS Results. The q dependencies of the scattered intensity from a 1.0 wt % PVCL-g-C11EO42 solution with and without SDS are depicted in Figure 5 at various temperatures. The results in the absence of SDS (Figure 5a) show an upturn of the scattered intensities at low q values that are strongly enhanced as the temperature increases and exceeds the CP value (CP ) 31.5 °C for the 1.0 wt % solution). The observed Porodlike behavior [I(q) ∼ q-4] is characteristic of formation of large-scale aggregates in the solution, which is ascribed to hydrophobic interchain associations as the hydrophobic character of the PVCL segments is reinforced at temperatures above CP. In the presence of SDS (parts b and c of Figure 5), the profiles of the scattered intensity curves are different. In (19) Kjøniksen, A.-L. et al. Preprint.

8014

Langmuir, Vol. 21, No. 17, 2005

Galant et al. Table 1. Characteristic Properties of the PVCL-g-C11EO42/SDS System from SANS Experiments 1 wt % PVCL-g-C11EO42 temperature (°C) 10.9 21.5 31.5 40.5

Figure 6. Effects of the SDS concentration on SANS from PVCL-g-C11EO42 (1.0 wt %) in D2O solutions at the temperatures indicated. Every fourth data point is shown.

this case, the spectra exhibit a peak in the intermediate q range, especially at the lowest temperatures. The observation of such a peak is a recurrent finding of SANS studies from polyelectrolyte solutions.20-22 From a theoretical viewpoint, the peak in the scattering pattern from polyelectrolyte solutions is interpreted as an ordered structure in the medium. This peak disappears progressively when salt is added, and consequently, it is interpreted as resulting from long-range electrostatic interactions that impose a preferential distance between the charged particles in the medium. According to parts b and c of Figure 5, the scattering profiles of the PVCL-gC11EO42/SDS system are dominated by electrostatic contributions (at least at low temperature), showing the existence of strong interactions between the nonionic copolymer and the anionic surfactant. The gradual disappearance of the interaction peak at higher temperatures is a result of enhanced hydrophobic interactions that reduce the effect of the electrostatic repulsive interactions. The Guinier-like plateau at elevated temperatures observed at low q values in the presence of SDS contrasts to the strong upturn of the scattered intensity observed in the absence of SDS. The difference in the profiles of the scattered intensity shows that the growth of large aggregates is strongly inhibited in the presence of SDS. From the observation of correlation peaks in Figure 5, it can be assumed that the adsorption of SDS onto the copolymer generates electrostatic interactions that impose a preferential distance d between the chains (or associated chains). This average distance d can be estimated from the relation d ) 2π/qmax, where qmax is the position of the maximum of the peak. Figure 6 displays SANS curves collected for mixtures with a constant polymer concentration (1.0 wt %) and different amounts of SDS for temperatures below and above CP. A close inspection of Figure 6 reveals that the position of the correlation peak is shifted toward higher q values as the SDS concentration is (20) Milas, M.; Rinaudo, M.; Duplessix, R.; Borsali, R.; Lindner, P. Macromolecules 1995, 28, 3119. (21) Ermi, B. D.; Amis, E. J. Macromolecules 1997, 30, 6937. (22) Borsali, R.; Nguyen, H.; Pecora, R. Macromolecules 1998, 31, 1548.

5 wt % PVCL-g-C11EO42

2 mm SDS

4 mm SDS

2 mm SDS

4 mm SDS

qmax (Å-1)

qmax (Å-1)

qmax (Å-1)

qmax (Å-1)

d (Å)

d (Å)

0.0150 418 0.0226 278 0.0150 418 0.0226 278 0.0150 418 0.0226 278

d (Å)

d (Å)

0.0203 310 0.0226 278 0.0226 278

increased. The values of qmax and the average interchain distance d for different polymer and surfactant concentrations are collected in Table 1 for some temperatures. We have not been able to detect any influence of temperature on qmax or the values estimated for the interchain distances, but at higher temperatures, the maximum of the interaction peak is more or less hidden by the overall increase of the scattered intensities at low q, characteristic of enhanced interchain aggregation. It is evident from Table 1 that the value of d decreases when the SDS concentration increases. This finding indicates a more homogeneous distribution of the chains in the presence of higher SDS concentrations, because of Coulomb forces between the chains and/or better dissolution of the polymer hydrophobic microdomains. Note that for the 5.0 wt % copolymer solutions, a smaller value of d was found in the presence of 2 mm SDS (d ) 310 Å), while in the presence of 4 mm SDS, a distance equivalent to the one estimated for the 1.0 wt % solutions (i.e., d ) 278 Å) was obtained. Thus, it seems that, in the presence of a sufficient amount of added surfactant (and in the range of 1.0-5.0 wt % copolymer solutions), the interchain distance d does not depend much on the copolymer concentration. In the range of low q values, it is obvious from Figure 6 that the influence of SDS addition is weak at temperatures well below CP (Figure 6a), whereas at higher temperatures, the effect of SDS addition is significant. At 50 °C (Figure 6c), i.e., above CP, large association structures characterized by high scattered intensity values are observed in the presence of 2 mm SDS, whereas smaller aggregates are observed at higher surfactant concentration. The general picture that emerges is that the presence of SDS in the dilute polymer solutions leads to amended thermodynamic conditions and repulsive electrostatic forces that reduce the tendency of the system to form hydrophobic associations. A different behavior appears for the semidilute copolymer solutions (5.0 wt %) in the presence of SDS. As shown in Figure 7, the scattered intensities measured at low q for these solutions below and close to the CP temperatures (parts a and b of Figure 7) exhibit similar features and are virtually not affected by the addition of SDS. However, a close inspection of the data at the highest temperature (well above CP, Figure 7c) reveals that the scattered intensity in the low q range displays an even stronger upturn in the presence of 4 mm of SDS than without the surfactant, showing an increase of the interchain aggregation in the presence of SDS. This effect is compatible with the depression of CP observed for the same sample, as discussed earlier (see Figure 3). Thus, the increase in SANS intensity observed in Figure 7 at high temperature shows that addition of a moderate amount of SDS enhances the number of effective crosslinks (mixed micelles shared by several polymer chains) and, consequently, gives rise to a transient network or strengthens an already existing network. In this context,

Turbidity and Small-Angle Neutron Scattering Studies

Figure 7. Influence of SDS addition on SANS from PVCLg-C11EO42 (5.0 wt %)/SDS mixtures in D2O at the temperatures indicated. Every fourth data point is shown.

Langmuir, Vol. 21, No. 17, 2005 8015

temperature, but the effect becomes less pronounced as the surfactant concentration increases. As discussed previously,14 an increase in the polymer concentration at a low constant temperature leads to poorer thermodynamic conditions of the system, resulting in shrinkage of the polymer coils. Dynamic light scattering results14 on this polymer at a low temperature without the surfactant also demonstrated a decrease in the hydrodynamic radius as the polymer concentration was increased. Thus, the behavior observed in Figure 8a is due to deteriorated thermodynamic conditions of the system. In the presence of SDS, a different picture emerges (parts b and c of Figure 8) with an interaction peak at intermediate q values and different features at low q. An increase of the copolymer concentration at a given temperature (T g 31.5 °C) promotes a strong upturn of the normalized data at low q values, and this effect is more pronounced as the temperature rises, i.e., when the “sticky” character of the PVCL segments is strengthened. This trend supports the surmise that, in the presence of a moderate level of added surfactant, interchain association can be induced by elevated temperature and high polymer concentration. In the analysis of SANS data in the low q region, our attention will now be focused on scattered intensity data with a plateaulike profile. These data have been analyzed with the aid of Guinier’s equation, yielding an overall apparent radius of gyration Rg of the aggregates 2 2 dΣ(0) q Rg dΣ(q) ) ln ln dΩ dΩ 3

Figure 8. Normalized SANS data from PVCL-g-C11EO42/D2O solutions with various levels of SDS addition, collected at different temperatures, for the two different polymer concentrations indicated. Every fourth data point is shown.

we should bear in mind that the polymer/surfactant ratio is much higher for the 5.0 wt % sample than for the 1.0 wt % sample. In Figure 8, the scattering profiles from 1.0 wt % solutions at different temperatures and levels of SDS addition are compared with the corresponding ones at the higher polymer concentration (5.0 wt %). To highlight the influence of the copolymer concentration, the SANS data presented in this figure have been normalized to the volume fraction of the solute and are called “normalized data”. In the absence of SDS (Figure 8a), the normalized data measured in the low q range are lower for the most concentrated solution, especially at low temperature. This trend is also visible in the presence of SDS at the lowest

(2)

Equation 2 is valid in the range where qRg e 1. Strictly speaking, the application of Guinier’s expression should be limited to noninteracting entities. Our point of view, however, is that the apparent radii of gyration estimated for the 1.0 wt % copolymer solutions at different conditions should correspond to dilute solutions of more or less charged aggregates and that this method should provide us with an average size of these entities. In this framework, quantitative analyses using Guinier’s method have also been reported for SANS from polyelectrolyte solutions.22-24 Typical illustrations of SANS data analyzed by means of Guinier’s method are shown in the inset of Figure 9. In these cases, the data could be well-described by eq 2. The values of the Rg estimated from this data treatment are presented in Figure 9 for 1.0 wt % solutions of PVCLg-C11EO42 as a function of the surfactant concentration at two different temperatures. The trend at the temperatures indicated that Rg falls off with an increasing SDS concentration. The apparent Rg values are reduced by approximately a factor of 2 when the SDS concentration is increased from 2 to 4 mm. This can be attributed to a gradual disruption of interchain association complexes as the level of added surfactant rises. Using Guinier’s equation, the scattered intensity extrapolated to q ) 0, dΣ/dΩ(0), has also been determined for the PVCL-g-C11EO42 (1.0 wt %)/SDS systems at high temperatures. For dilute and uniform particles, dΣ/dΩ(0) should vary pro(23) Matsuoka, H.; Schwahn, D.; Ise, N. Macromolecules 1991, 24, 4227. (24) Takahashi, Y.; Matsumoto, N.; Iio, S.; Kondo, H.; Noda, I.; Imai, M.; Matsushita, Y. Langmuir 1999, 15, 4120.

8016

Langmuir, Vol. 21, No. 17, 2005

Galant et al.

Figure 9. Influence of SDS addition and temperature on the Rg values resulting from the fits of eq 2 to the SANS data from PVCL-g-C11EO42 (1.0 wt %)/SDS mixtures in D2O. The solid lines in the inset plot represent the fits of eq 2 to the experimental data.

Figure 10. (a-c) Influence of the d-SDS concentration on the scattering from PVCL-g-C11EO42 (1.0 wt %) at the temperatures indicated. Every fourth data point is shown. (d) Solid lines represent fits of the qmax versus cd-SDS to power laws qmax ∼ cd-SDSβ.

portionally to the apparent weight average molecular weight Mw of the aggregates

5 of the apparent molecular weight of the scattering objects can be assumed. To obtain more detailed information about the polymer structure of the complexes formed in the presence of SDS, the scattering from PVCL-g-C11EO42 in the mixtures has been determined by using contrast-matching conditions; d-SDS has been added to the hydrogenated copolymer in a D2O/H2O mixture (91:9%). Under these conditions, the scattering length densities of the surfactant and the solvent are almost identical and, hence, the scattering from d-SDS is strongly suppressed. Figure 10 shows the evolution of the scattering from PVCL-g-C11EO42 in 1.0 wt % solutions when the d-SDS concentration is increased at different temperatures. It can be seen that the scattering profiles from PVCL-g-C11EO42 exhibit a correlation peak at all temperatures, which confirms that ionic surfactant molecules are bound to the copolymer chains (this gives rise to repulsive correlations between chains or associated chains). As expected for dilute PVCL-g-C11EO42 solutions involving increasing amounts of surfactant, the intensity of the correlation peak is reduced and its position is shifted toward higher q values as the surfactant concentration is increased (formation of smaller aggregates and more homogeneous distribution of the chains in the medium). Actually, the d-SDS concentration dependency of the scattering vector associated with the maximum of the correlation peak, qmax, follows a power law (qmax ∼ cd-SDSβ), as shown in Figure 10d. A decreased value of the power law exponent, i.e., lower concentration dependence, is found at higher temperatures. At elevated temperatures, the hydrophobic character of the polymer chains is strengthened and it is possible that the change of the electrostatic interactions induced by adding d-SDS is relatively less pronounced, and therefore, a smaller shift of the maximum of the interaction peak occurs with an increasing surfactant concentration. Generally, information concerning the local structure of the chains in solution can be obtained by analyzing the asymptotic behavior of the scattered intensities at large q values. In the range where qRg > 1, the variation of the scattered intensities as a function of q may be represented by a power law, dΣ/dΩ ∼ q-R. The exponent R is usually identified with the fractal dimension df for homogeneous systems that are characterized by self-similarity within some spatial range (the mass distribution in such structures varies with a power df of the length R).

dΣ(0) cMw ) 2 (bp - bD2O)2 dΩ F NA

(3)

In eq 3, c is the polymer concentration (mass/volume), F is the mass density of the polymer, NA is Avogadro’s constant, and bp and bD2O are the scattering length densities of the polymer particle and the solvent, respectively. For a single PVCL-g-C11EO42 chain, the use of eq 3 leads to a theoretical value dΣ/dΩ(0) = 10 cm-1. At 21.5 °C and in the absence of SDS, a value of dΣ/dΩ(0) equal to 13 ( 1 cm-1 has been determined for the 1.0 wt % PVCLg-C11EO42 solution, indicating that under these conditions of temperature and concentration, the probability of forming multichain aggregates is low. For aggregates formed at 50 °C in the presence of 2 and 4 mm of SDS, we have estimated values of dΣ/dΩ(0) equal to 1300 and 50 cm-1, respectively, assuming that the contribution of the surfactant to the scattered intensity is weak at very low q. Thus, the apparent average molecular weight of the aggregates decreases dramatically (by a factor of more than 20) when the surfactant concentration is increased from 2 to 4 mm, which is correlated to the decrease of their apparent Rg. At high temperatures, the decrease of the apparent size of the aggregates with an increase in the level of SDS is therefore due to reduced association between polymer chains. When the surfactant concentration is increased at a fixed polymer concentration, the development of mixed micelle-like clusters with a decreasing [PVCL]/[SDS] ratio can be anticipated, up to the formation of clusters involving only few PVCL segments belonging to the same copolymer chain in the limiting case. It should be noted that, in the investigated range of SDS concentrations, the formation of multichain aggregates is still detected upon heating the solutions. Indeed, as shown in Figure 5c, an increase in the scattered intensity values can still be observed as the temperature rises in the presence of 4 mm of SDS. Moreover, when dΣ/dΩ(0) determined for a single PVCLg-C11EO42 chain (i.e., 10 cm-1) is compared to, for example, dΣ/dΩ(0) estimated for PVCL-g-C11EO42 mixed with 4mm of SDS at 50 °C (i.e., 50 cm-1), an increase by a factor of

Turbidity and Small-Angle Neutron Scattering Studies

Figure 11. Effects of the surfactant concentration and temperature on the power law exponent R, characterizing the high q behavior of the scattered intensity, for dilute and semidilute solutions of PVCL-g-C11EO42.

Langmuir, Vol. 21, No. 17, 2005 8017

the polymer/surfactant ratio and a larger contribution from the polymer to the scattered intensity. To investigate the internal structure of the polymer alone, without the contribution to the scattered intensity from surfactant micelles and/or surfactant molecules adsorbed onto the polymer chains, polymer solutions with contrast-matched d-SDS are considered in Figure 11c. The general trend at each temperature is that the value of R drops off as the surfactant concentration increases, which indicates swelling of the structures. This supports that the increase of R with the SDS concentration observed in parts a and b of Figure 11 is primarily due to the contribution of the surfactant to the scattering. In Figure 11c, at higher levels of surfactant addition, the effect of temperature on R has practically vanished, because all of the studied temperatures are far below the CP of the system at these high d-SDS concentrations, and hence, this temperature change has little effect on the morphology. At low surfactant addition, the impact of temperature on R is strong because at this stage the hydrophobicity of the polymer segments is significantly affected by temperature; elevated temperatures give rise to contracted structures. Conclusions

The R values resulting from the analysis of the experimental data in the high q range (0.03 < q < 0.15 Å-1) are presented in Figure 11 for the systems indicated. The same general effect of temperature can be observed for both polymer concentrations in the absence and presence of the surfactant (parts a and b of Figure 11), namely, that the value of R increases with rising temperatures. The theoretical predictions25 of df for a selfavoiding chain (linear swollen polymer) and for a Gaussian chain (polymer in a θ solvent) are 1.7 and 2, respectively. Values up to 3 can be reached for more condensed structures, such as globules. Thus, the variation of R observed with the temperature for the dilute polymer solutions (Figure 11a) indicates a conformational transition of the chains from random to more compacted coils with increasing temperature. However, it appears from Figure 11a that the apparent compactness of the local structures depends on the level of surfactant addition. At intermediate temperatures, higher values of R are observed when the SDS concentration increases, indicating the formation of compact microdomains. This behavior may at first sight be surprising, because the presence of the surfactant in the dilute polymer solutions promotes electrostatic repulsions and amended thermodynamic conditions. However, the increase of R with the surfactant concentration at intermediate temperatures may be ascribed to the formation of dense SDS clusters adsorbed onto the polymer and whose contribution to the scattering is stronger as the SDS concentration is increased. At high temperatures, the scattering from the polymer dominates and the difference disappears. In the semidilute polymer solution (Figure 11b), similar trends with a temperature-induced contraction of the structural entities are observed, but the values of R are generally lower both with and without SDS addition. We note that the effect of SDS addition is also smaller, and to some extent, this may be related to the higher value of

This study reveals that aqueous mixtures of the PVCLg-C11EO42 copolymer and SDS can be easily manipulated in a reproducible and reversible way by changing temperature and surfactant concentration in the process of forming aggregates. When an aqueous PVCL-g-C11EO42 solution is heated above its CP temperature, large-scale aggregates are created, characterized by a sharp increase of both the turbidity of the solution and the scattered intensity. These large-scale aggregates possess an increasingly more compact structure as the temperature rises, as seen from a quantitative analysis of the SANS data. When the anionic surfactant SDS is added to a dilute PVCL-g-C11EO42 solution (1.0 wt %), smaller aggregates are produced, with a more homogeneous distribution in the medium. The addition of increasing amounts of SDS to the copolymer solution results in an increase of the CP value. Furthermore, the value of the fractal dimension, estimated from the SANS data at high q, shows that the smaller aggregates formed in the presence of SDS have a more swollen internal structure as the SDS concentration rises. These effects, i.e., the possibility of controlling the CP, the size, and the internal structure of association complexes by using an anionic surfactant, can be advantageous to various potential applications, in particular those where it is important to prevent the aggregation of the particles upon an increase in temperature. On the other hand, for semidilute PVCL-g-C11EO42 solutions (5.0 wt %), the addition of moderate amounts of SDS (up to 4 mm) leads to a decrease in CP because of an effective cross-linking of the copolymer chains as the temperature rises. At higher levels of SDS addition, the value of CP steadily increases as the SDS concentration rises. This work has clearly demonstrated that the degree of aggregation can be tuned by varying parameters such as the level of surfactant addition, polymer concentration, and temperature. The resulting transient assemblies may find a range of applications.

(25) De Gennes, P.-G. Scaling Concepts in Polymer Physics; Cornell University Press: Ithaca, NY, 1979.

Acknowledgment. B. N. and K. D. K. gratefully acknowledge financial support provided by the Norwegian

8018

Langmuir, Vol. 21, No. 17, 2005

Research Council through a NANOMAT Project (158550/ 431). K. D. K. and C. G. thank the Marie Curie Industry Host Project (contract number G5TR-CT-2002-00089) for support. Support from IFF-Forschungszentrum Ju¨lich is gratefully acknowledged. A. L. K. and C. G. thank the

Galant et al.

“Ju¨lich Neutron for Europe” program for a traveling grant. The technical assistance by Dr. W. Pyckhout-Hintzen and Dr. D. Schwahn is gratefully acknowledged. LA050036A