New Amphiphilic Diblock Copolymers: Surfactant Properties and

4044

Langmuir 2006, 22, 4044-4053

New Amphiphilic Diblock Copolymers: Surfactant Properties and Solubilization in Their Micelles Se´bastien Garnier† and Andre´ Laschewsky*,†,‡ UniVersita¨t Potsdam, P.O Box 601553, D-14415 Potsdam-Golm, Germany, and Fraunhofer Institute for Applied Polymer Research FhG-IAP, Geiselbergstr. 69, D-14476 Potsdam-Golm, Germany ReceiVed January 6, 2006. In Final Form: February 19, 2006 Several series of amphiphilic diblock copolymers are investigated as macrosurfactants in comparison to reference low-molar-mass and polymeric surfactants. The various copolymers share poly(butyl acrylate) as a common hydrophobic block but are distinguished by six different hydrophilic blocks (one anionic, one cationic, and four nonionic hydrophilic blocks) with various compositions. Dynamic light scattering experiments indicate the presence of micelles over the whole concentration range from 10-4 to 10 g‚L-1. Accordingly, the critical micellization concentrations are very low. Still, the surface tension of aqueous solutions of block copolymers decreases slowly but continuously with increasing concentration, without exhibiting a plateau. The longer the hydrophobic block, the shorter the hydrophilic block, and the less hydrophilic the monomer of the hydrophilic block is, the lower the surface tension is. However, the effects are small, and the copolymers reduce the surface tension much less than standard low-molar-mass surfactants. Also, the copolymers foam much less and even act as anti-foaming agents in classical foaming systems composed of standard surfactants. The copolymers stabilize O/W emulsions made of methyl palmitate as equally well as standard surfactants but are less efficient for O/W emulsions made of tributyrine. However, the copolymer micelles exhibit a high solubilization power for hydrophobic dyes, probably at their core-corona interface, in dependence on the initial geometry of the micelles and the composition of the block copolymers. Whereas micelles of copolymers with strongly hydrophilic blocks are stable upon solubilization, solubilization-induced micellar growth is observed for copolymers with moderately hydrophilic blocks.

1. Introduction Amphiphilic block copolymers are a particular class of surfactants. They are composed of at least one hydrophobic block and one hydrophilic block1-4 and are usually called polymeric surfactants or “macrosurfactants”. In comparison to classical surfactants, amphiphilic block copolymers often exhibit a reduced mobility and slower diffusion rates.5 As a direct consequence, the equilibrium between polymeric micelles can take several days.4,6 Moreover, macrosurfactants have much lower critical micelle concentrations (CMC) than their low-molecular-mass counterparts.7-10 Typically, the CMCs of macrosurfactants have been reported in the concentration range from 10-9 to 10-4 mol‚L-1,5,11-13 whereas common surfactants such as sodium dodecyl sulfate (SDS) or cetyltrimethylammonium bromide (CTAB) exhibit CMCs on the order of 10-3 to 1 mol‚L-1.14 The * To whom correspondence should be addressed. Phone: +493315681327. Fax: +493315683000. E-mail: [email protected]. † Universita ¨ t Potsdam. ‡ Fraunhofer Institute for Applied Polymer Research FhG-IAP. (1) Laschewsky, A. Tenside Surf. Det. 2003, 40, 246-249. (2) Alexandridis, P. Curr. Opin. Colloid Interface Sci. 1997, 2, 478-489. (3) Cameron, N. S.; Corbierre, M. K.; Eisenberg, A. Can. J. Chem. 1999, 77, 1311-1326. (4) Riess, G. Prog. Polym. Sci. 2003, 28, 1107-1170. (5) Creutz, S.; van Stam, J.; De Schryver, F. C.; Je´roˆme, R. Macromolecules 1998, 31, 681-689. (6) Won, Y.-Y.; Davis, H. T.; Bates, F. S. Macromolecules 2003, 36, 953955. (7) Storsberg, J.; Laschewsky, A. SO ¨ FW J. 2004, 130, 14-18. (8) Fo¨rster, S. Ber. Bunsen-Ges. Phys. Chem. 1997, 101, 1671-1678. (9) Astafieva, I.; Zhong, X. F.; Eisenberg, A. Macromolecules 1993, 26, 73397352. (10) Vieira, J. B.; Thomas, R. K.; Li, Z. X.; Penfold, J. Langmuir 2005, 21, 4441-4451. (11) Jada, A.; Siffert, B.; Riess, G. Colloids Surf. 1993, A 75, 203-209. (12) Antoun, S.; Gohy, J.-F.; Je´roˆme, R. Polymer 2001, 42, 3641-3648. (13) Chang, Y.; Bender, J. D.; Phelps, M. V. B.; Allcock, H. R. Biomacromolecules 2002, 3, 1364-1369. (14) Riess, G.; Labbe, C. Macromol. Rapid Commun. 2004, 25, 401-435.

CMC might be even absent for macrosurfactants.15 The extremely low CMCs are advantageous for many applications, since only traces of polymer are required to form micelles. High dilution effects, problematic in the case of classical surfactants, do not alter polymeric micelles. Furthermore, the surface activity of new polymeric surfactants may strongly differ from the classical surface-active behavior of surfactants. Because of their much lower diffusion coefficients and their much more complex conformations at the interface air/water in comparison to their low-molar-mass counterparts, amphiphilic block copolymers might behave differently and offer new property profiles at interfaces. In fact, macrosurfactants have been reported to exhibit specific advantages in numerous applications, compared to their low-molar-mass counterparts. For instance, they were reported to be efficient dispersers, e.g., for the stabilization of latexes.14,16-20 They have been investigated as rheology modifiers,4 efficiency boosters in microemulsions,21-24 or flocculants.7 Furthermore, numerous studies have reported enhanced stability of emulsions via (electro)-steric stabilization,25-29 including multiple emul(15) Rager, T.; Meyer, W. H.; Wegner, G. Macromolecules 1997, 30, 49114919. (16) Riess, G. Colloids Surf. A 1999, 153, 99-110. (17) Tauer, K.; Zimmermann, A.; Schlaad, H. Macromol. Chem. Phys. 2002, 203, 319-327. (18) Lieske, A.; Jaeger, W. Macromol. Chem. Phys. 1998, 199, 255-260. (19) Kukula, H.; Schlaad, H.; Tauer, K. Macromolecules 2002, 35, 25382544. (20) Tan, B.; Grijpma, D. W.; Nabuurs, T.; Feijen, J. Polymer 2005, 46, 13471357. (21) Jakobs, B.; Sottmann, T.; Strey, R. Langmuir 1999, 15, 6707-6711. (22) Endo, H.; Allgaier, J.; Mihailescu, M.; Monkenbusch, M.; Gompper, G.; Richter, D.; Jakobs, B.; Sottmann, T.; Strey, R. Appl. Phys. A 2002, 74, 392-395. (23) Byelov, D.; Frielinghaus, H.; Allgaier, J.; Gompper, G.; Richter, D. Physica B 2004, 350, 931-933. (24) Frielinghaus, H.; Byelov, D.; Allgaier, J.; Gompper, G.; Richter, D. Physica B 2004, 350, 186-192. (25) Nikova, A. T.; Gordon, V. D.; Cristobal, G.; Ruela Talingting, M.; Bell, D. C.; Evans, C.; Joanicot, M.; Zasadzinski, J. A.; Weitz, D. A. Macromolecules 2004, 37, 2215-2218.

10.1021/la0600595 CCC: $33.50 © 2006 American Chemical Society Published on Web 03/29/2006

New Amphiphilic Diblock Copolymers

sions, in the presence of amphiphilic block copolymers. Micelles of macrosurfactants are efficient solubilizing agents for hydrophobic substances,30-33 if their structure is properly designed for a given solubilizate. This, in combination with the extremely low CMCs, makes amphiphilic diblock copolymers attractive materials for the design of controlled drug delivery systems, for example.34-38 Importantly, the use of amphiphilic block copolymers has the added benefit of choosing from a much larger pool of useful hydrophilic and hydrophobic segments in the surfactant design,1 when classical hydrophobic alkyl chains and hydrophilic ethylene oxide chains or ionic headgroups are not adequate or optimal for a targeted application. However, the vast majority of studies on macrosurfactants has been limited to block copolymers bearing poly(ethylene oxide) as the hydrophilic block so far and mostly using poly(propylene oxide) as the hydrophobic block.2,39-42 The next best investigated system is poly(styrene)block-poly(acrylic acid) whose hydrophobic block is glassy at ambient temperature at which most studies have been performed.3,4,33,43-48 Therefore, comparative studies with a larger variety of the chemical structures are desirable to improve the understanding of the surfactant behavior of block copolymer amphiphiles. We have recently described the synthesis of several series of amphiphilic diblock copolymers by reversible addition fragmentation chain transfer (RAFT) polymerization and reported on their aggregation properties in water and selective organic solvents.49-51 Here, we focus on the surface-active properties of these macrosurfactants at the air/water and oil/water interfaces, on one hand, and on the solubilization capacity of their micelles in water, on the other hand. The systems studied share as common element the nature of the hydrophobic block, namely poly(butyl acrylate), poly(M1), whose glass transition temperature is about -47 °C,50 i.e., well below the experimental conditions. The macrosurfactants are distinguished by the nature of their (26) Gref, R.; Babak, V.; Bouillot, P.; Lukina, I.; Bodorev, M.; Dellacherie, E. Colloids Surf. A 1998, 143, 413-420. (27) Hoerner, P.; Riess, G.; Rittig, F.; Fleischer, G. Macromol. Chem. Phys. 1998, 199, 343-352. (28) Ivanova, R.; Balinov, B.; Sedev, R.; Exerowa, D. Colloids Surf. A 1999, 149, 23-28. (29) Gosa, K.-L.; Uricanu, V. Colloids Surf. A 2002, 197, 257-269. (30) Stepanek, M.; Krijtova, K.; Prochazka, K.; Teng, Y.; Webber, S. E.; Munk, P. Acta Polym. 1998, 49, 96-102. (31) Kim, J.-H.; Emoto, K.; Iijima, M.; Nagasaki, Y.; Aoyagi, T.; Okano, T.; Sakurai, Y.; Kataoka, K. Polym. AdV. Technol. 1999, 10, 647-654. (32) Chen, X. L.; Jenekhe, S. A. Langmuir 1999, 15, 8007-8017. (33) Choucair, A.; Eisenberg, A. J. Am. Chem. Soc. 2003, 125, 11993-12000. (34) Kataoka, K.; Harada, A.; Nagasaki, Y. AdV. Drug DeliVery ReV. 2001, 47, 113-131. (35) Kabanov, A. V.; Batrakova, E. V.; Alakhov, V. Y. J. Controlled Release 2002, 82, 189-212. (36) Kakizawa, Y.; Kataoka, K. AdV. Drug DeliVery ReV. 2002, 54, 203-222. (37) Bertin, P. A.; Watson, K. J.; Nguyen, S. T. Macromolecules 2004, 37, 8364-8372. (38) Dong, C. M.; Chaikof, E. L. Colloid Polym. Sci. 2005, 283, 1366. (39) Chu, B. Langmuir 1995, 11, 414-421. (40) Booth, C.; Attwood, D. Macromol. Rapid Commun. 2000, 21, 501-527. (41) Mortenson, K. Polym. AdV. Technol. 2001, 12, 2-22 (42) Kozlov, M. Y.; Melik-Nubarov, N. S.; Batrakova, E. V.; Kabanov, A. V. Macromolecules 2000, 33, 3305-3313. (43) Burguie`re, C.; Chassanieux, C.; Charleux, B. Polymer 2003, 44, 509518. (44) Laruelle, G.; Francois, J.; Billon, L. Macromol. Rapid Commun. 2004, 25, 1839-1844. (45) Zhao, J.; Allen, C.; Eisenberg, A. Macromolecules 1997, 30, 7143-7150. (46) Choucair, A.; Lavigueur, C.; Eisenberg, A. Langmuir 2004, 20, 38943900. (47) Voulgaris, D.; Tsitsilianis, C. Macromol. Chem. Phys. 2001, 202, 32843292. (48) Chan, Y.; Mi, Y. Polymer 2004, 45, 3473-3480. (49) Mertoglu, M.; Garnier, S.; Laschewsky, A.; Skrabania, K.; Storsberg, J. Polymer 2005, 46, 7726-7740. (50) Garnier, S.; Laschewsky, A. Macromolecules 2005, 38, 7580-7592. (51) Garnier, S.; Laschewsky, A. Colloid Polym. Sci., accepted for publication.

Langmuir, Vol. 22, No. 9, 2006 4045

hydrophilic blocks poly(Mx), providing different hydrophilicities (Figure 1A). These are, from the least to the most hydrophilic ones: poly(N-acryloyl pyrrolidine) [poly(M2)], poly(dimethyl acrylamide) [poly(M3)], poly(2-(acryloyloxylethyl) methyl sulfoxide) [poly(M4)], comblike poly(oligo(ethyleneglycol) methyl ether acrylate) [poly(M5)], anionic poly(2-acrylamido-2-methylpropanesulfonic acid) [poly(M6)], and cationic poly(3-acrylamidopropyltrimethylammonium chloride) [poly(M7)] (Figure 1A). The ratio f between the length of the hydrophilic and the hydrophobic blocks varies from 0.4 to 4 (see Table 2). The unusually broad compositional range of the block copolymers under investigation, combined with the broad chemical variation of the hydrophilic blocks, provide a gradual variation of the hydrophilicity, and a direct comparison of nonionic, permanently anionic, and permanently cationic systems. This enables an instructive comparison of the surfactant behavior of amphiphilic block copolymers with standard surfactants (see Figure 1B). Thus, the properties of the diblock copolymers at interfaces and the ability of their micelles to solubilize a hydrophobic azo-dye (see Figure 1D) are investigated, compared to those of reference low-molar-mass or polymeric surfactants (see Figure 1B) and correlated to macromolecular parameters such as the polarity of the hydrophilic block and the relative and absolute molar masses of the blocks. This study makes a valuable addition to the understanding of block copolymer surfactants, as most reports have focused so far on the influences of the nature and/or of the absolute molar mass of the hydrophobic block on the aggregation behavior and surface-active properties,2,5,52-55 while the influence of the hydrophilic segments is still hardly explored. 2. Experimental Section 2.1. Materials. The synthesis of the amphiphilic diblock copolymers by the RAFT polymerization technique is described elsewhere.50 After polymerization, the block copolymers in solution in a cosolvent were purified via dialysis against water to separate the block copolymers from inevitable and undesirable traces of hydrophobic and/or hydrophobic homopolymers, which are inherent to the mechanism of the RAFT technique. The absence of any lowmolar-mass homopolymers or species in the macro-surfactants was demonstrated by narrow monomodal molar mass distributions.50 The synthesis of azo-dye [4-(4-butyl-phenylazo)-phenyl]-diethylamine (S2) used for solubilization experiments was described elsewhere.56 Poly(ethylene glycol) hexadecyl ether (Brij56) (cryst. grade, HLB 12.0), tributyrin (98%) and methyl palmitate (97%) were purchased from Aldrich. Sodium dodecyl sulfate (SDS) (cryst. grade) and cetyltrimethylammoniumbromide (CTAB) (cryst. grade) were used as received from Serva (Heidelberg, Germany). Poly(allyl alcohol 1,2-butoxylate)-block-poly(ethoxylate) (HLB 9.9) (P1) was purchased from Aldrich. Its number average molar mass Mn was determined by membrane osmometry in THF (from Acros, HPLC grade) to be 7500 ( 500 g‚mol-1. The HLB concept of Davies,57 combined with Mn, allowed the estimation of the composition of block copolymer P1, namely (allyl alcohol (1,2butoxylate)37-block-(ethoxylate)100). Dialysis tubes “Zellu Trans” (nominal molar mass cut off 3500) from Roth were used to prepare aqueous micellar solutions. Water for all experiments was purified by a Millipore Qplus water purification system (resistance 18 MΩcm). (52) van Stam, J.; Creutz, S.; De Schryver, F. C.; Je´roˆme, R. Macromolecules 2000, 33, 6388-6395. (53) Deng, Y.; Young, R. N.; Ryan, A. J.; Fairclough, J. P. A.; Norman, A. I.; Tack, R. D. Polymer 2002, 43, 7155-7160. (54) Gohy, J.-F.; Lohmeijer, B. G. G.; De´camps, B.; Leroy, E.; Boileau, S.; van den Broek, J. A.; Schubert, D.; Haase, W.; Schubert, U. S. Polym. Int. 2003, 52, 1611-1618. (55) Hruby, M.; Konak, C.; Ulbrich, K. J. Appl. Polym. Sci. 2005, 95, 201211. (56) Anton, P.; Laschewsky, A. Colloid Polym. Sci. 1994, 272, 1118-1128. (57) Davies, J. T. Proc. Int. Congr. Surf. ActiVity 1959, 1, 426.

4046 Langmuir, Vol. 22, No. 9, 2006

Garnier and Laschewsky

Figure 1. Chemicals used in this study: Amphiphilic diblock copolymers (A), reference surfactants (B), oils for emulsification experiments (C), and hydrophobic substance for solubilization experiments (D). 2.2. Methods. Micellar solutions of block copolymers were prepared as described in detail elsewhere.50 When the hydrophobic block was shorter than the hydrophilic one, polymers were directly dissolved in purified water, with a concentration of 1 g‚L-1. When the hydrophobic block was longer than the hydrophilic one, the polymers were dissolved in a cosolvent (typically dioxane) for 24 h and dialyzed against purified water for 3 days (nominal molar mass cutoff: 3500). Micellar solutions of reference surfactants were prepared by direct dissolving the surfactant in Millipore water. Dynamic light scattering (DLS) of micellar solutions was performed with a high performance particle sizer (HPPS, from Malvern Instruments) equipped with an He-Ne laser (633 nm). The measurements were made at the scattering angle θ ) 173° (“backscattering detection”) at 25 °C. The autocorrelation functions were analyzed with the CONTIN method. The apparent hydrodynamic diameters DH of micelles or aggregates were calculated according to the Stokes-Einstein equation, DH ) kT/3πηDapp, where Dapp is the apparent diffusion coefficient and η is the viscosity of the solution. In the case of copolymers with ionic hydrophilic blocks, the derived nominal values should be considered only as estimates as no salt was added to suppress charge effects. Prior to measurement, the polymer solutions were filtered using a Sartorius Ministar-plus 0.45 µm disposable filter and were placed in a polystyrene cuvette. Surface tension measurements were performed with a Wilhelmy plate (platinum) K12 tensiometer from Kru¨ss (Hamburg, Germany) at room temperature (20 °C). Only directly water-soluble polymers were studied over the concentration range from 1 × 10-4 to 10

g‚L-1. Aqueous polymer micellar stock solutions with concentrations of 10, 1, and 0.1 g‚L-1 were prepared by directly dissolving the block copolymers in MilliQ water and vigorous stirring for 3 days. For each block copolymer, about 5 further solutions were subsequently prepared by dilution of the stock solutions with MilliQ water and by stirring for 3 days. Block copolymers which had to be dispersed in water by the dialysis method50 were investigated at the single concentration of 1 g‚L-1. The reference surface tension of pure water γMilliQ-water ) 73.6 mN‚m-1 ((1.0 mN‚m-1) for the geometry of the plate used was measured before each measurement series. All micellar solutions were placed in crystallizing dishes of identical size (40 mL, 50 mm diameter, 30 mm height from Roth, Germany) and were allowed to equilibrate at least 7 days between two subsequent measurements. Constant values of the surface tension γ after about one month of quiet standing were considered to correspond to the equilibrium. The crystallizing dishes were sealed, to avoid the evaporation of water in the equilibration periods. After each measurement, the plate was carefully cleaned with acetone and MilliQ water. The absence of residual polymer or of dust traces on the plate and the conservation of the form of the plate were regularly verified by measuring the surface tension of MilliQ water as a reference. Error intervals of about (1 mN‚m-1 were found. Each polymer or surfactant solution was systematically characterized by DLS in parallel over the concentration range studied. For foaming experiments, aqueous solutions of diblock copolymers and low-molar-mass surfactants with a concentration of 1 g‚L-1 and with a constant volume were prepared as described elsewhere50 and

New Amphiphilic Diblock Copolymers mechanically shaken for 1 min in identical narrow glass tubes. The height of the resulting foam was plotted vs time. Mixtures of surfactants were studied, too, by mixing (1/1 v/v) two different surfactant solutions (1 g‚L-1) for 3 days before shaking. Ultrasonication for the preparation of emulsions was performed with a US50 apparatus (IKA Labortechnik, Germany). The emulsions were prepared as follows. 4.0 g of water, 1.0 g of oil, and 0.015 g of block copolymer or reference surfactant were poured together into similar glass tubes. The mixtures were sonicated with 1 s impulses, for 1 min. Emulsions were obtained as white one-phase mixtures with an initial maximal height of 18 mm in the glass tubes. The height of the emulsion phase as well as the size of the droplets formed were determined as a function of time. DLS for the characterization of emulsions used a Mastersizer X from Malvern Instruments (U.K.), equipped with a 2 mW He-Ne laser (λ ) 633 nm). The emulsion phase sample, whose volume was smaller than 1 mL, was placed in a small volume sample preparation unit MSX1 (Malvern), which diluted the emulsion with water to an overall volume of 100 mL, stirred the mixture, and pumped the resulting diluted emulsion to the measuring cell of the DLS apparatus. The diameter d of the droplets was subsequently determined. UV-visible spectra were recorded with a Cary-1 UV-vis spectrophotometer (Varian) equipped with temperature controller (Julabo F-10). Quartz cuvettes (Suprasil, Heraeus, Germany) with an optical path length of l ) 1 cm were used. For quantitative measurements, the Lambert-Beer law was applied: A ) lC, where A is the absorbance, is the extinction coefficient of the sample [L‚mol-1‚cm-1], and C is the concentration of the chromophore [mol‚L-1]. The absorbance maximum of the lowest energy π-π transition of azo-dye S2 was found at λ ) 417 nm in water, at λ ) 412 nm in butyl acetate, at λ ) 409 nm in 2-propanol, and at λ ) 405 nm in hexane. The extinction coefficient in 2-propanol was determined as ) 1.7 × 104 L‚cm-1‚mol-1. Each solubilization experiment was subsequently performed 3 times in parallel (with the same micellar solution), according to the following procedure. A total of 1 mg of solid S2 per mL of solution was added to aqueous micellar solutions of amphiphilic diblock or reference surfactants with concentrations of 1 or 4 g‚L-1, respectively (to ensure the presence of micelles). As one week was necessary to achieve the maximum values, the suspensions were shaken for 3 weeks at 20 °C to ensure equilibration. After allowing most of the undissolved dye to settle down during 4 days, the more or less colored solutions were filtered using a Sartorius Ministar-plus 0.45 µm disposable filter in order to remove the excess of unsolubilized dyes and characterized by UV-vis spectroscopy. The solubilization capacity Ω, defined as the concentration of solubilized hydrophobic substance normalized to the concentration 1 g‚L-1 of surfactant, was calculated from the peak absorbance A, using the extinction coefficient of the dye determined in 2-propanol. The Ω values differed between solubilization experiments performed in parallel for a given micellar solution and a given hydrophobic substance by (20%. The error interval for the values of the absorbance maximum λmax of the dyes was found to be (1 nm. The micellar solutions were systematically characterized by DLS before and after solubilization.

3. Results and Discussion 3.1. Surface Activity and Critical Micelle Concentration (CMC). The surface tension of aqueous solutions of amphiphilic diblock copolymers and reference (macro)-surfactants was measured, using a Wilhelmy plate, classical geometry for the characterization of aqueous solutions of amphiphilic block copolymers.58,59 Prior to measurements, the absence of any lowmolar-mass molecules such as oligomers in the polymer samples, which could strongly modify the surface tension of the aqueous (58) Tan, B.; Grijpma, D. W.; Nabuurs, T.; Feijen, J. Polymer 2005, 46, 13471357. (59) Crothers, M.; Attwood, D.; Collett, J. H.; Yang, Z.; Booth, C.; Taboada, P.; Mosquera, V.; Ricardo, N. M. P. S.; Martini, L. G. A. Langmuir 2002, 18, 8685-8691.

Langmuir, Vol. 22, No. 9, 2006 4047

Figure 2. Surface tension γ vs concentration of aqueous solutions of reference surfactants SDS and P1 (A), block copolymers poly(M1)b-poly(M7) (B), block copolymers poly(M1)-b-poly(M6) (C), and block copolymers with various hydrophilic blocks (D). (×) SDS, (+) P1, (2) (M1)81-b-(M7)105, (4) (M1)81-b-(M7)55, (1) (M1)81b-(M6)136, (3) (M1)95-b-(M6)58, and (9) (M1)81-b-(M5)95. The arrows indicate the CMC of SDS and of P1 according to DLS studies.

polymer solutions, was verified by size exclusion chromatography.50 For practical reasons, only the directly water soluble block copolymers were studied over the concentration range from 10-4 to 10 g‚L-1, namely poly(M1)-b-poly(M5), poly(M1)b-poly(M6), and poly(M1)-b-poly(M7) (see Figure 2).50 Systematically, DLS measurements were performed in parallel, to obtain simultaneous information on the aggregation behavior of the amphiphiles as a function of the concentration. The surface tensions of solutions of (M1)37-b-(M3)70, (M1)37-b-(M4)59, (M1)37-b-(M4)108 and (M1)95-b-(M4)190, which are not directly soluble in water due to the moderate hydrophilicity of their hydrophilic block,50 were measured at one single concentration of 1 g‚L-1 (see Table 1). It has been shown elsewhere that the polymeric micellar systems studied are dynamic but exhibit very slow diffusion and exchange coefficients.50 As a consequence, a reliable comparison of the surface activity of the block copolymer systems with that of reference surfactants was only possible after about one month of equilibration time for the block copolymers, i.e., after the surface tension became constant with time. SDS was used as a low-molar-mass reference surfactant (see Figure 1B). As depicted in Figure 2A, the aqueous solution of SDS exhibited a continuous decrease of γ with the concentration, till the concentration above γ remained constant. This is a typical surface-active behavior for a surfactant. The break of the curve of γ plotted against concentration can be considered to occur at the CMC of SDS. The analysis of the data indicated a CMC of about 2.2 g‚L-1, in excellent agreement with the literature.58 The curve of γ vs concentration exhibits a minimum in the vicinity of the CMC, which is attributed to traces of dodecanol, inherent to the synthesis and/or hydrolysis of SDS. The characterization of the solutions by DLS indicated the presence of micelles (DH ) 4 nm) at concentrations above 2.2 g‚L-1 but no aggregates below this critical concentration, thus corroborating the CMC. For a better panel of reference surfactants, the surface activity of a polymeric reference surfactant was studied, too. Commercial block copolymer (allyl alcohol (1,2-butoxylate)37-block(ethoxylate)100) was chosen (see Figure 1B), whose block lengths are comparable to those of the block copolymers

4048 Langmuir, Vol. 22, No. 9, 2006


Table 1. Surface Tension γ of Aqueous Solutions of Amphiphilic Block Copolymers at the Concentration of 1 g‚L-1 block copolymer

γ (mN‚m-1)

(M1)37-b-(M3)70 (M1)95-b-(M4)190 (M1)37-b-(M4)106 (M1)37-b-(M4)59 (M1)81-b-(M5)95 (M1)81-b-(M6)136 (M1)95-b-(M6)58 (M1)81-b-(M7)55 (M1)81-b-(M7)105

52.9 ( 1 65.1 ( 1 59.3 ( 1 56.0 ( 1 51.7 ( 1 64.7 ( 1 56.5 ( 1 50.9 ( 1 59.1 ( 1

investigated in this work. Figure 2A gives the plot of the surface tension γ of aqueous solutions of P1 vs concentration. The behavior of the polymeric reference surfactant clearly differs from that of SDS, with γ decreasing already at much lower concentrations. It is noteworthy that the shape of this part of the γ vs log(concentration) curve is convex, rather than concave as expected for a standard surfactant. At concentrations above 0.2 g‚L-1, γ becomes nearly constant with ca. 32 mN‚m-1. The observed behavior suggests the presence of a CMC, although the discontinuity in the γ vs log(concentration) curve is less evident than in the case of SDS. The hypothetical CMC of 0.2 g‚L-1 for P1 is in the typical concentration range for block copolymers with relatively low molar mass.4 DLS measurements indicated the presence of micelles (DH ) 15 nm) at concentrations above 0.2 g‚L-1, whereas no aggregates were observed below this critical concentration, confirming the presence of a CMC for P1 at about 0.2 g‚L-1. The differences in the surface tension plots of the two reference surfactants SDS and P1 (Figure 2A) are obvious. They might be derived from the inherent polydispersity of the polymer surfactant or from low molar mass or oligomeric impurities. Alternatively, the differences could be due to the low diffusion and exchange coefficients of the polymers. In any case, the references exemplify that the determination of the CMC of macrosurfactants via the classical plot of γ vs concentration may be arduous. Table 1 and Figure 2B-D illustrate the behavior of the investigated amphiphilic block copolymers. Their limiting surface tensions of about 50 mN‚m-1 at concentrations as high as 10 g‚L-1 are substantially higher than those of the two reference surfactants SDS and P1 or than those reported for classical poly(ethylene oxide)-block-poly(propylene oxide) systems in the 33-43 mN‚m-1 range.60 It is noteworthy that γ continuously decreases from 73 to about 50 mN‚m-1 for concentrations from 1 × 10-4 to 10 g‚L-1. This differs from the situation observed for polymeric reference surfactant P1 and from many reports mentioning the presence of a pseudoplateau58,59,61 or at least of a discontinuity62-64 of γ vs log(concentration) plots of aqueous solutions of macrosurfactants. In our case, no indication for a CMC can be detected via surface tension measurements. However, DLS experiments show the presence of micelles of constant DH over the whole concentration range studied,51 demonstrating that the new amphiphilic diblock copolymers exhibit CMCs below 1 × 10-4 g‚L-1, if they have a measurable CMC at all. The differences between reference macrosurfactant P1 and our block copolymers are striking, but the reasons are not obvious. (60) Alexandritis, P.; Athanassiou, V.; Fukuda, S.; Hatton, T. Langmuir 1994, 10, 2604-2612. (61) Hussain, H.; Busse, K.; Kressler, J. Macromol. Chem. Phys. 2003, 204, 936-946. (62) Fischer, A.; Brembilla, A.; Lochon, P. Polymer 2001, 42, 1441-1448. (63) de Paz Ba´n˜ez, M. V.; Robinson, K. L.; Armes, S. P. Macromolecules 2000, 33, 451-456. (64) Ravi, P.; Sin, S. L.; Gan, L. H.; Gan, Y. Y.; Tam, K. C.; Xia, X. L.; Hu, X. Polymer 2005, 46, 137-146.

Both hydrophobic blocks poly(butylenoxide) and poly(butylacrylate) are well above their glass transitions at 20 °C. A possible explanation for the much lower CMCs could be the higher hydrophobicity of the hydrophobic block poly(M1). As at least samples (M1)37-b-(M3)70, (M1)37-b-(M4)59, (M1)37-b-(M4)106 have a very similar size to P1, the different behavior should a priori not be due to differences in the molar masses. However, it cannot be excluded that the observed behavior of P1 is dominated by surface-active oligomers, which had been carefully removed from the poly(M1)-b-poly(Mx) block copolymers by extensive dialysis. As DLS indicates the presence of micelles throughout the concentration range investigated, the small but continuous decrease of γ with increasing concentration asks for an explanation. Despite the painstaking cleaning procedures, the presence of surface-active trace impurities in the polymer solutions cannot be rigorously excluded but is highly improbable. It has been shown that the equilibria in an aqueous solution of macrosurfactants differ from those of low-molar-mass surfactants with increasing concentration. Whereas the concentration of unimers, and thus γ, is constant once the micelles are formed for classical surfactant systems, the concentration of polymeric unimers was reported to continue to increase after micellization.65 This could explain the absence of any plateau of γ vs concentration for all of the diblock copolymers studied. Alternatively, one might consider that polymeric micelles might be slightly surface active, causing the continuous decrease of γ with increasing concentration. In fact, a slow but continuous decrease of the surface tension with concentration of amphiphilic block copolymers in the presence of polymer micelles has been reported occasionally but was attributed to a particular behavior of strongly charged hydrophilic blocks.66 In our case, copolymers with nonionic hydrophilic blocks show such a behavior, too. This slow but continuous decrease of the surface tension with concentration in the presence of micelles is an interesting phenomenon,66 which needs theoretical explanation. The observed surface-activity of the block copolymers was subsequently correlated with the nature of their hydrophilic blocks. First, the influence of the absolute length of the hydrophilic block was considered, as illustrated in Figure 2B for macrosurfactants poly(M1)-b-poly(M7). Block copolymer (M1)81-b(M7)105 is slightly less surface active than (M1)81-b-(M7)55. This can be explained by the fact that the block copolymer has a higher affinity for water with increasing molar mass of the hydrophilic block, the length of the hydrophobic block being constant, reducing the density of the monolayer at the air/water interface and thus increasing γ. This goes along with previous studies, which report the decrease of the surface-activity of Brijtype surfactants with increasing length of the poly(ethylene oxide) fragments67 or of amphiphilic block copolymers with increasing the length of the hydrophilic block.62 The comparison of systems (M1)95-b-(M6)58 (f ) 0.6) and (M1)81-b-(M6)136 (f ) 1.7) (Figure 2C) demonstrates that the surface activity of amphiphilic block copolymers is controlled by the relative length of the hydrophilic block, too. The more hydrophilic the macrosurfactant is, the less surface-active it is (see Table 1, too). The comparison of the γ values at 1 g‚L-1 for block copolymers (M1)37-b-(M4)59 (f ) 1.6, γ ) 56.0 mN‚m-1), (M1)37-b-(M4)106 (f ) 2.9, γ ) 59.3 mN‚m-1), and (M1)95-b-(M4)190 (f ) 2.0, γ ) 65.1 mN‚m-1) (65) Morel, A.; Cottet, H.; In, M.; Deroo, S.; Destarac, M. Macromolecules 2005, 38, 6620-6628. (66) Kaewsaiha, P.; Matsumoto, K.; Matsuoka, H. Langmuir 2005, 21, 99389945. (67) Schambil, F.; Schwuger, M. J. Surfactants in Consumer Products; Falbe, J., Ed.; Springer: Berlin, 1986; p 142.


(see Table 1) allows us to conclude that the absolute, and not relative, length of the hydrophilic segments is the main factor governing the surface activity of diblock copolymers poly(M1)b-poly(M4). The shorter the hydrophilic block, the higher the surface activity of the macrosurfactant, whatever the length of the hydrophobic block. This differs from the behavior of aqueous solutions of poly(ethylene oxide)-block-poly(propylene oxide) copolymers, whose surface tension was reported to be controlled only by the absolute length of the hydrophobic block poly(propylene oxide).68 Other block copolymer structures should be compared to verify this tendency. Finally, the influence of the nature of the hydrophilic block was analyzed (Figure 2D). The nonionic block copolymer poly(M1)-b-poly(M5) is clearly more surface-active than ionic block copolymers poly(M1)-b-poly(M6) and poly(M1)-b-poly(M7). Once again, this can be correlated to the lower hydrophilicity of the nonionic block compared to the ionic ones. In line with these observations are reports on increasing surface activity of poly(M1)-b-poly(acrylic acid) with decreasing degree of neutralization,70 as well as reports on increasing surface activity with decreasing protonation and quaternization of amphiphilic diblock copolymers containing a polymeric amine as hydrophilic block.71,72 This comparative study highlights that the surface activity of macrosurfactants can be tailored by the hydrophilicity, i.e., the length and the nature, of the hydrophilic segments. 3.2. Foaming Properties of Block Copolymers. As discussed above, the amphiphilic diblock copolymers are less surfaceactive than the surfactants SDS or P1 for concentrations above their respective CMCs. This should confer polymeric surfactants different foaming properties from those of low-molar-mass surfactants. Therefore, the ability of the amphiphilic diblock copolymers to form and stabilize foams was studied, in comparison to that of reference low-molar-mass surfactants SDS, CTAB, and Brij56, and polymeric P1 (see Figure 1B). As depicted in Figure 3A, plotting the foam height vs time after shaking, the ionic surfactants SDS and CTAB are the best foaming agents and foam stabilizers among the various amphiphiles studied, with anionic surfactant SDS exhibiting the same foaming behavior as CTAB. This is in agreement with previous works reporting the high foaming properties of ionic surfactants.69 In comparison, the nonionic surfactant Brij56 is a poor foaming agent, whereas the nonionic polymeric surfactant P1 exhibits intermediate foaming ability. The amphiphilic diblock copolymers of all poly(M1)-b-poly(Mx) series clearly foam less than any of the reference surfactants after shaking, with higher foam collapse rates. Surprisingly, ionic block copolymers poly(M1)-b-poly(M6) and poly(M1)-b-poly(M7) do not form any foam. Among the nonionic block copolymers of the poly(M1)-b-poly(Mx) series, the foaming behavior is virtually independent of the nature of the hydrophilic block or on the block length. Thus, whatever their composition, the synthesized macrosurfactants are very poor foaming agents. This could be explained by their relatively low surface activity at the air/water interface, as discussed above, and by their low diffusion coefficients in comparison to the experimental shaking time of 1 min. The foaming of mixtures of the synthesized block copolymers with the low molar mass reference surfactants (1/1 v/v) was (68) Wanka, G.; Hoffmann, H.; Ulbricht, W. Macromolecules 1994, 27, 41454159. (69) Tadros, T. F. Applied Surfactants; Wiley-VCH Verlag: Weinheim, Germany, 2005. (70) Hartenstein, M. PhD Thesis, Bayreuth, Germany, 2002. (71) Minoda, M.; Sawamoto, M.; Higashimura, T. Macromolecules 1990, 23, 1897-1901. (72) Baines, F. L.; Billingham, N. C.; Armes, S. P. Macromolecules 1996, 29, 3416-3420.

Langmuir, Vol. 22, No. 9, 2006 4049

Figure 3. Foam height vs time for aqueous solutions of surfactants. (A) pure amphiphilic diblock copolymers and reference surfactants. (B) Mixtures of amphiphilic diblock copolymer (M1)95-b-(M2)157 and reference surfactants; the curves of the pure reference surfactants are shown for comparison. Note the different time scales for (A) and (B).

studied, too. Figure 3B illustrates the behavior observed for all of the block copolymers and the surfactants tested, by the example of (M1)95-b-(M2)157. The presence of block copolymer (M1)95b-(M2)157 in the reference surfactant solutions of CTAB and Brij56 markedly lowers the initial foam formation and foam stabilization abilities of the low-molar-mass surfactants. The same behavior was found for the ionic block copolymers, too. Thus, the poly(M1)-b-poly(Mx) macrosurfactants act as antifoaming agents in classical foaming systems. This might be explained by the adsorption of the polymers onto the surfactant interfacial film, modifying its mechanical properties. The solubilization of the low-molar-mass surfactants in the polymeric micelles could be an alternative explanation. The formation of foams or antifoam properties is still far from being well understood. Further experiments are needed to obtain more detailed information about the antifoaming properties exhibited by these diblock copolymers. 3.3. Stabilization of Emulsions by Block Copolymers. Despite the generally lower ability of amphiphilic block copolymers to decrease the interfacial tension between two immiscible phases in comparison to low-molar-mass surfactants, they have been shown to stabilize emulsions via steric stabilization.27,29,73-75 Thus, emulsification by the series of poly(M1)-b-poly(Mx) macrosurfactants was investigated, in comparison to that by the low-molar-mass reference surfactants. To foresee which type of emulsion would be appropriate to study, the HLB values of the macrosurfactants were estimated (Table 2). The empirical HLB concept predicts the emulsion type to be expected,76 but for the poly(M1)-b-poly(Mx) block copolymers, (73) Imbert, P.; Sadtler, V. M.; Dellacherie, E. Colloids Surf. A 2002, 211, 157-164. (74) Faers, M. A.; Luckham, P. F. Colloids Surf. A 1994, 86, 317-327. (75) Yang, Z.; Sharma, R. Langmuir 2001, 17, 6254-6261. (76) Griffin, W. C. J. Cosm. Chem. 1954, 1, 311-328.

4050 Langmuir, Vol. 22, No. 9, 2006


this method suffers from the lack of data for the HLB numbers of functional groups present in the synthesized macrosurfactants. A second method, applicable to block copolymers, allows the estimation of an approximate HLB values from the composition of the macrosurfactants, according to the equation

HLB )

WH 20 WH + WL

with WH and WL being the weight fraction of the hydrophilic and lipophilic segments, respectively.58 According to this method, HLB values of the macrosurfactants vary from 6.7 for (M1)133b-(M4)53 (f ) 0.4) to 15.0 for (M1)37-b-(M3)145 (f ) 3.9). Although this method does not take into account the nature of the hydrophilic blocks poly(Mx) and thus their different polarities, one may expect that most polymers stabilize O/W emulsions according to the estimated HLB values. Typically, surfactants with HLB values in the range 3-6 stabilize W/O emulsions, whereas those with HLB values in the range 8-18 rather stabilize O/W emulsions.69 In the case of O/W emulsions stabilized by amphiphilic block copolymers, it was reported that the hydrophobic block is not adsorbed at the interface but rather extended in the oil droplet.77 Thus, the oil phase should be compatible with the hydrophobic block poly(M1) to ensure a good stabilization of the emulsion. Accordingly, two oils of medium polarities but of different densities were chosen, namely methyl palmitate (d25 ) 0.852 g‚mL-1) and tributyrine (d25 ) 1.032 g‚mL-1) (see Figure 1C). Methyl palmitate is frequently used,78 for instance in personal care applications, whereas tributyrine has been shown to be an effective antitumor agent.79 Directly after the preparation of the emulsion, no difference could be observed macroscopically between the use of the block copolymers and of the low molar mass reference surfactants SDS, Brij 56, and CTAB as emulsifiers. With both oils, a single white emulsion phase was obtained for all of the systems, confirming the surface activity of the amphiphiles used at the interface between the two liquids. The plot of the height of the emulsion phase vs time, giving the rate of creaming or sedimentation, is depicted in Figure 4 for selected amphiphiles. For the systems composed of methyl palmitate as oil (see Figure 4A), the creaming occurred in a similar way and with comparable rates with the amphiphilic block copolymers as with the lowmolar-mass surfactants, over the period of 15 days. The emulsion height of all of the systems decreased rapidly during the first 2 days after emulsification, from 18 to about 4 mm, till a plateau was reached, corresponding to the equilibrium emulsion volume (lower phase). The stabilization efficiency of the reference surfactants did not differ from each other, i.e., being relatively poor, since the end emulsion volume represented only about 20% of the overall volume. The ability of the block copolymers to stabilize the emulsions was similar to that of SDS, CTAB, and Brij56. No differentiation between the block copolymers as a function of their composition could be seen, and steric or electrosteric stabilization by the polymers does not produce a particular advantage in this system. In the case of tributyrine as oil (see Figure 4B), CTAB seems to be particularly efficient for the stabilization of O/W emulsions, yielding a 94% (in volume) upper emulsion phase after 15 days. Generally, sedimentation occurred much more slowly with the low-molar-mass surfactants as emulsifiers than with the macro(77) Perrin, P.; Lafuma, F. J. Colloid Interface Sci. 1998, 197, 317-326. (78) Nuisin, R.; Ma, G.-H.; Omi, S.; Kiatkamjornwong, S. J. Appl. Polym. Sci. 2000, 77, 1013-1028. (79) Su, J.; Ho, P. C. J. Pharm. Sci. 2004, 93, 1755-1765.

Figure 4. Height of the emulsion phase vs time with methyl palmitate (A) or tributyrine (B) as oil, with low-molar-mass surfactants CTAB (*) or Brij56 (g), or block copolymers (M1)95-b-(M6)58 (3) or (M1)81b-(M7)55 (4) as emulsifiers (0.3% w/w), or without surfactant (9).

Figure 5. Droplet diameter d of O/W emulsions of methyl palmitate stabilized by block copolymers (M1)37-b-(M4)106, (M1)81-b-(M5)95 and (M1)95-b-(M6)58, and reference surfactants Brij56 and CTAB (dotted lines) vs time.

surfactants, indicating that the steric effects of the block copolymers are not sufficient at the concentration studied. As a general tendency, it seems that block copolymers with a higher relative length of the hydrophilic block are somewhat more efficient to stabilize the emulsions but in any case still less efficient than the low-molar-mass reference surfactants. Interestingly, the electrostatic effect does not seem to be a factor governing the emulsion stabilization capacity among the block copolymers of the various poly(M1)-b-poly(Mx) series. Even in the case of the emulsions made with methylpalmitate, a somewhat lower stabilization efficiency of the block copolymers in comparison to their low-molar-mass counterparts was revealed by the determination of the oil droplet diameter d as a function of time, as exemplified in Figure 5. Directly after emulsification (t ) 0), the droplets stabilized by the low molar mass surfactant are clearly smaller than those stabilized by the block copolymers. The droplet diameter increases with time, indicating Ostwald ripening or coalescence for both systems, till reaching a plateau after about 10 days. The end emulsion state is composed of droplets with a size of about 20 µm for the low-molar-mass and of and about 30 µm for the polymeric surfactants as emulsifiers. The fast increase of the droplet size when using the macro-


Langmuir, Vol. 22, No. 9, 2006 4051

Table 2. Solubilization of Dye S2 by Amphiphilic Diblock Copolymers at 20 °C, Followed by DLS and UV/vis Spectroscopy surfactants diblock copolymer (M1)95-b-(M2)157 (M1)37-b-(M3)70 (M1)86-b-(M3)125 (M1)86-b-(M3)138 (M1)133-b-(M3)146 (M1)95-b-(M4)190 (M1)133-b-(M4)53 (M1)133-b-(M4)106 (M1)81-b-(M5)95 (M1)95-b-(M5)42 (M1)81-b-(M6)136 (M1)95-b-(M6)58 (M1)81-b-(M7)55 (M1)81-b-(M7)105 Brij56 SDS CTAB P1 water

fa 1.7 1.9 1.5 1.6 1.1 2.0 0.4 0.8 1.2 0.4 1.7 0.6 0.7 1.3

2.7

DLS measurements calculated HLB 12.3 11.9 10.6 11.1 9.2 14.3 6.7 10.0 16.0 12.2 14.6 9.9 10.5 13.5 12.0

PDIb 1.26 1.33 1.17 1.21 1.20 1.31

1.34

DsH c

DHc [nm] before solubilization

[nm] after solubilization

51 (93%)/303 (6%) 26 65 (95%)/450 (5%) 54 83 (16%)/458 (83%) 81 93 143 52 31 (87%)/245 (12%) 43 54 261 268

80 (23%)/360 (76%) 112 (47%)/305 (53%) 121 187 222 130 255 117 76 120 (24)/515 (75) 40 64 170 249

3 3 16

224 241 15

UV studies λmaxd [nm]

Ωe [mg‚L-1]

416 414 414 415 417 416 416 416 416 414 414 417 417 416 410 409 417 417

< 0.1 21.5 125.0 46.8 74.6 24.6 112.4 136.1 32.0 81.0 38.6 78.7 54.9 50.4 47.5 8.3 50.5 32.6 2.4

a f is the ratio (length of the hydrophilic block/length of the hydrophobic block). b Apparent polydispersity index according to SEC in NMP [data taken from ref 50]. c Hydrodynamic diameter of micelles and of second population of aggregates before (DH) and after (DsH) solubilization, respectively, determined by DLS. 51 (93%) means 93 vol % of aggregates with DH of 51 nm. No bracket means 100 vol %. d λmax of the absorbance peak of the dye in micellar solutions. e Ω is the solubilization capacity of the surfactant ) concentration of solubilized dye normalized to 1 g‚L-1 of surfactant. Amount of solubilized dye calculated from the absorbance maximum using the extinction coefficient determined in 2-propanol (cf. Table 1).

surfactants as emulsifier indicates that the steric stabilization effect is not high enough to prevent from Ostwald ripening or coalescence. These findings show that the amphiphilic diblock copolymers investigated are surface active at the interface between the liquid phases but are less prone to adsorb onto the interface than lowmolar-mass surfactants. Their much lower ability to decrease interfacial tensions in comparison to their low-molar-mass counterparts is not compensated by a steric stabilization effect at low concentrations. These findings markedly contrast from previous reports describing the efficient stabilization of emulsions by polymeric surfactants.73-75 A possible explanation could be that a perfect match in terms of solubility and polarity between the hydrophobic block and the oil is necessary, to obtain a good emulsifier performance. Also, one can speculate that the steric stabilization effect might be most efficient in emulsion systems composed of mixtures of amphiphilic block copolymers and lowmolar-mass surfactants, as previously reported.80 3.4. Solubilization by Aqueous Micellar Solutions of Diblock Copolymers. The solubilization capacity Ω of the block copolymers in water was investigated and compared to that of reference low-molar-mass and polymeric surfactants, investigating the uncharged azo-dye S2 as solubilizate (see Figure 1D). The results are summarized in Table 2. Looking at solvatochromic effects, the absorbance maxima of the solubilized dye typically inform on the polarity of the solubilizate microenvironment in micellar solution and thus about the solubilization site, i.e., in the core, at the interface, or in the corona of the micelles.56 This was done by comparing the absorbance maxima of the hydrophobic dye in the micellar solutions with those of the dye in solvents of different polarities, namely water, butyl acetate, 2-propanol, and hexane (cf. Experimental Section). Note that dye S2 is readily soluble in these organic solvents but is only sparingly soluble in water (see concentration of solubilized dye in pure water in Table 2, last line). S2 shows a moderate

solvatochromism, λmax increasing with the polarity of the solvent. The solvatochromic data listed in Table 2 show that the λmax values of the aqueous micellar solutions are found between those in water and in 2-propanol or butyl acetate, for both the block copolymers and the reference surfactants. This suggests that the dye is solubilized at polar sites of the micelles. Furthermore, the λmax values in the aqueous micellar solutions vary from a polymer to another. The nature of the hydrophobic block being constant, this excludes the inner hydrophobic micellar core as the primary solubilization site. The λmax values of the solubilized dye are, within the experimental error, nearly the same for block copolymers with a given hydrophilic block. This can be explained by a more or less close proximity of the hydrophobic dye to the hydrophilic corona chains. However, the solubilization capacity seems to be primarily influenced by the characteristics of the hydrophobic block. Therefore, the observations support a localization of the solubilizate at the interface between the micellar core and the micellar corona. This justifies the use of the extinction coefficients of the dye determined in 2-propanol to calculate the solubilization capacity of the micelles. As often found in solubilization studies,56,81 the results of the solubilization capacity of the (macro-) surfactants are complex and difficult to correlate with the composition of the block copolymers, keeping in mind that the thermodynamic equilibrium might not be reached. Nevertheless, the large number of samples of macrosurfactants shows some general trends. The synthesized amphiphilic diblock copolymers exhibit a high ability to solubilize dye S2, as indicated by their high solubilization capacities Ω in comparison to micellar solutions of the low molar-mass reference surfactants (see Table 2). For example, S2 is solubilized by block copolymer (M1)133-b-(M4)106 about 60 times more than in pure water and 16 times more than by SDS. For a meaningful correlation between the solubilization data and the composition of the macrosurfactants, many factors must be taken into account. First, the initial state of the micellar solutions, i.e., the micellar size distribution before solubilization, is a crucial factor governing

(80) Pons, R.; Taylor, P.; Tadros, T. F. Colloid Polym. Sci. 1997, 275, 769776.

(81) Zhang, Y.; Zhang, Q.; Zha, L.; Yang, W.; Wang, C.; Jiang, X.; Fu, S. Colloid Polym. Sci. 2004, 282, 1323-1328.

4052 Langmuir, Vol. 22, No. 9, 2006


Figure 7. Size changes of aggregates of amphiphilic block copolymer in water upon solubilization of hydrophobic dye S2 (from the left to the right): - ) (M1)86-b-(M3)138 before solubilization, - - - ) (M1)133-b-(M4)53 before solubilization, ‚‚-‚‚ ) (M1)86-b-(M3)138 after solubilization, ‚‚‚ ) (M1)133-b-(M4)53 after solubilization.

Figure 6. Solubilization capacity of polymer solutions (1 g‚L-1) for dye S2, as a function of the factor f (ratio length of the hydrophilic block/length of the hydrophobic block): (a) (O) series poly(M1)b-poly(M3), (3) series poly(M1)-b-poly(M4), (4) ) series poly(M1)-b-poly(M5), (×) ) series poly(M1)-b-poly(M6), (+) ) series poly(M1)-b-poly(M7); (b) as in part a but normalized on the weight fraction of the hydrophobic block poly(M1) in the copolymers.

the apparent solubilization capacity of the macrosurfactants. For instance, aqueous solution of (M1)86-b-(M3)125 (Ω ) 125.0 mg‚L-1) initially exhibited micelles of DH of 65 nm (95 vol %) and additional aggregates of about 460 nm (5 vol %) according to DLS experiments. In contrast, the very similar (M1)86-b-(M3)136 forms monomodal micelles (DH ) 54 nm) in water and exhibited a much lower Ω value of 46.8 mg‚L-1. Accordingly, large aggregates seem to improve the Ω values. For this reason, comparative solubilization studies with surfactants showing multimodal aggregate size distributions in water should be avoided. Second, for micellar solutions with monomodal aggregate size distributions, the initial micellar size and shape might play a crucial role for the solubilization capacity of the macrosurfactants, rendering the comparisons with the block copolymer composition arduous. As reported elsewhere,51 the geometry of the micelles formed by the amphiphilic diblock copolymers in water is mainly governed by the nature of the hydrophilic block. This could explain the very different solubilization capacities obtained between the different diblock copolymer classes. For example, block copolymers poly(M1)b-poly(M4), forming large elongated ellipsoids or large spherical nonmicellar aggregates in water,51 solubilized much better than the other polymeric and low-molar-mass systems. Thus, comparisons between block copolymers having the same hydrophilic block nature and the same micellar shape with initial monomodal size distributions should be preferred. As a general tendency, the solubilization capacity increases with increasing relative and absolute length of the hydrophobic block poly(M1) for a given hydrophilic block. This tendency is even visible when comparing the copolymers with different hydrophilic blocks, as illustrated for the relative length of the poly(M1) block in Figure 6. The comparison of panels a and b in Figure 6 demonstrates that this effect cannot be explained by the inherently higher content of hydrophobic block for a given copolymer concentration but must be a structural effect. The

finding agrees with previous reports about solubilization by other amphiphilic block copolymers.81 For instance, for a macrosurfactant series with a given hydrophilic block, (M1)133-b-(M4)106 (f ) 0.8, Ω ) 136.1 mg‚L-1) solubilizes better than (M1)95b-(M4)190 (f ) 2, Ω ) 24.6 mg‚L-1) or (M1)95-b-(M6)58 (f ) 0.6, Ω ) 78.7 mg‚L-1) solubilizes better than (M1)81-b-(M6)136 (f ) 1.7, Ω ) 38.6 mg‚L-1). Also, (M1)81-b-(M7)55 (f ) 0.7, Ω ) 54.9 mg‚L-1) solubilizes better than (M1)81-b-(M7)105 (f ) 1.3, Ω ) 50.4 mg‚L-1). Therefore, the solubilization capacity of the amphiphilic diblock copolymers can be fairly correlated to the relative size of the hydrophobic domains. The sensitivity of the micellar systems to the solubilization effects was studied by comparing the DLS data before and after solubilization (see Table 2). The behavior of the amphiphilic diblock copolymers is apparently correlated to the nature of the hydrophilic block as well as to the initial state of the micellar solutions. For block copolymers initially forming large aggregates in water such as (M1)133-b-(M4)53, the solubilization of dye S2 was accompanied by a marked growth of the aggregates (e.g., before solubilization DH ) 93 nm, after solubilization DsH ) 255 nm, see Figure 7). This might correspond to a solubilizationinduced morphological transition. Micellar systems with initial multimodal size distribution, i.e., containing small micelles and large aggregates, undergo a dramatic micellar growth, too, and evolved in the direction of the large aggregates. This is exemplified by block copolymer (M1)95-b-(M5)42 (DH ) 31 nm (87%) and 245 nm (12%), DsH ) 120 nm (24%) and 515 nm (75%)). For these systems, the incorporation of hydrophobic dyes destabilized the micelles in favor of the large aggregates. Monomodal micelles of poly(M1)-b-poly(M3) were destabilized, too, as illustrated by block copolymer (M1)86-b-(M3)138 forming only large aggregates in water after solubilization (DH ) 54 nm, DsH ) 187 nm, see Figure 7). Considering the high DsH values for block copolymers poly(M1)-b-poly(M2) to poly(M1)-b-poly(M4), one can conclude that large metastable aggregates are formed during the solubilization process for these polymers. This can be explained by the moderate hydrophilicity of their hydrophilic blocks which is not high enough to stabilize the small micellar systems when additional hydrophobic material, i.e., the dye, is incorporated in the core chains of poly(M1). Thus, for these systems whose thermodynamic balances are modified by the incorporation of hydrophobic dyes, the term “solubilization in micelles” should be employed with care. In contrast, monomodal micelles of poly(M1)-b-poly(M5) and poly(M1)-b-poly(M6) were stable upon solubilization. This is attributed to their highly hydrophilic solvating blocks. In the most cases, a limited solubilizationinduced growth of the micelles was observed, presumably due to an increase of the size of the micellar core.82 Noteworthy, the


large aggregates of poly(M1)-b-poly(M7) rather shrank upon solubilization, probably because of a change in the interactions stabilizing their initial morphology. Possibly, these aggregates are clusters of micelles which disintegrate in the course of the solubilization process. Note that micellar morphology transitions occurred for reference low-molar-mass surfactants, too. Solubilization of dye S2 in micelles of CTAB or SDS caused a dramatic increase of the aggregate size from 3 to about 220-240 nm. Our observations go along with recent reports describing morphological changes of micelles of low-molar-mass surfactants induced by the solubilization of hydrophobic substances.83 Interestingly, the λmax values of S2 solubilized in CTAB or SDS micelles were much lower (about 410 nm) than for the macrosurfactant samples, suggesting a solubilization site more oriented toward the micellar core. This can be attributed to the hydrophobic alkyl chains of the low-molar-mass surfactants, which are much less polar than the ester functional groups of the hydrophobic block of the macrosurfactants. This suggests that the micelles were destabilized by strong hydrophobic interactions between the dye containing a C4-alkyl chain and the alkyl chain of the surfactants.

4. Conclusions Amphiphilic diblock copolymers composed of poly(butyl acrylate) as constant hydrophobic block and of various hydrophilic blocks form micelles at very low concentrations and are only weakly surface-active and weakly foaming. The more hydrophilic the hydrophilic block is, either due to its increased size or due to an increased hydrophilicity of the repeat units, the lower the surface activity is. These macrosurfactants rather act as anti(82) Chevalier, Y.; Zemb, T. Rep. Prog. Phys. 1990, 53, 279-371. (83) Mishael, Y. G.; Dubin, P. L. Langmuir 2005, 21, 9803-9808.

Langmuir, Vol. 22, No. 9, 2006 4053

foaming agents in classical low-molar-mass foaming surfactant systems. Their interfacial behavior strongly differs from that of reference surfactants, as the surface tension continuous decreases with increasing concentration even when micelles are present. This unusual phenomenon suggests either that the unimer concentration continues to increase after micellization or that the polymeric micelles are surface-active. The block copolymers studied are less able to stabilize standard emulsions with oils of medium polarity compared to low-molar-mass surfactants. In contrast, the polymeric micelles more efficiently solubilize hydrophobic substances of medium polarity in water. Solubilization leads to a dramatic aggregate growth of the copolymers with moderately hydrophilic blocks and probably to morphological transitions, whereas macrosurfactants composed of ionic blocks exhibit excellent stability, i.e., no solubilization-induced precipitation or morphological transitions. Owing to the low toxicities of the poly(ethyleneoxide) macromonomer and the sulfoxide-based hydrophilic block, to the high capacity of the macrosurfactants to solubilize hydrophobic substances in aqueous medium, and to the exceptional stability of their micelles against dilution, such polymeric amphiphiles offer a high potential for drug delivery systems or encapsulation of active substances for personal care for instance. Acknowledgment. We thank N. Zuber (Ecole Nationale Supe´rieure de Chimie de Lille, France) for help with surface tension measurements and emulsification studies. We acknowledge the helpful comments of the reviewers for the discussion the solubilization behavior. Financial support by Fonds der Chemischen Industrie and DFG (Project La611/4-1) is gratefully acknowledged. LA0600595

New Amphiphilic Diblock Copolymers: Surfactant Properties and

Recommend Documents