Imperfect Dissolution in Nonionic Block Copolymer and Surfactant

Feb 2, 2009 - Imperfect Dissolution in Nonionic Block Copolymer and Surfactant. Mixtures. Karin Shimoni and Dganit Danino*. Department of Biotechnolog...
0 downloads 0 Views 2MB Size
2736

Langmuir 2009, 25, 2736-2742

Imperfect Dissolution in Nonionic Block Copolymer and Surfactant Mixtures Karin Shimoni and Dganit Danino* Department of Biotechnology and Food Engineering, Technion-Israel Institute of Technology, Haifa, Israel 32000 ReceiVed NoVember 11, 2008. ReVised Manuscript ReceiVed December 27, 2008 Self-assembled copolymer micelles have been widely explored for numerous applications including cosmetic formulations and detergency, drug delivery, and agriculture. In many of these technologies at least trace amounts of surfactants and detergents are present, yet little is known regarding their effect on the copolymer micelle structure. In this paper we examine the influence of a nonionic micelle-forming surfactant, Triton X-100, on spherical, nonionic polymeric micelles composed of poly(butadiene)-co-poly(polyethylene oxide). Using cryo-TEM we find that relatively small surfactant concentrations (less than 1:1 molar ratio) are sufficient to disrupt the copolymer assemblies, and to yield, via dimerization, mixed polymer-surfactant micelles with characteristic diameters. Saturation of the polymeric micelles is reached with approximately 3 mM surfactant (1:8 mol ratio). Upon saturation, and in high surfactant excess, coexistence of two homogeneous micellar populations is found: saturated polymer-surfactant micelles, and much smaller micelles of pure surfactant. The lack of complete demicellization of the polymeric micelles is explained by packing constraints of the polymer hydrophobic chains by the added surfactant. This behavior is found to be characteristic of polymeric molecules with hydrophobic-to-hydrophilic molecular weight ratio close to, or exceeding, 0.75. We further found that structural transitions in polymer-surfactant mixtures are fast, and the systems reach equilibrium at time scales characteristic to the small molecule, in contrast with the slow equilibration in polymer-polymer mixtures.

Introduction Amphiphilic block copolymer micelles and vesicles are promising candidates for many environmental, agricultural, pharmaceutical, cosmetic, detergent, and drug delivery applications.1-4 In drug delivery, for example, polymeric carriers show enhanced stability and longer circulation time in vivo when compared to their small-molecule counterparts.5-7 The coexistence of amphiphilic block copolymers with small amphiphiles in many of these formulations8 advanced intense research on the interaction between the two classes of molecules and their mixing behavior (see for example reviews by Sastry and Hoffmann9 and Dan and Safran,10 and references therein). Indeed, it was shown that diblock copolymer vesicles are highly sensitive to dissolution by surfactants and small molecule amphiphiles,9,11-15 and * Corresponding author. Phone: 972-4-829-2143(office), 3168 (laboratory); fax: +972-4-829-3399; e-mail: [email protected]. (1) Torchilin, V. P. J. Controlled Release 2001, 73, 137–172. (2) Nishiyama, N.; Kataoka, K. Polym. Drugs Clin. Stage 2003, 519, 155–177. (3) Peng, C. L.; Shieh, M. J.; Tsai, M. H.; Chang, C. C.; Lai, P. S. Biomaterials 2008, 29, 3599–3608. (4) Lee, E. S.; Oh, K. T.; Kim, D.; Youn, Y. S.; Bae, Y. H. J. Controlled Release 2007, 123, 19–26. (5) Geng, Y.; Discher, D. E. Polymer 2006, 47, 2519–2525. (6) Cai, S. S.; Vijayan, K.; Cheng, D.; Lima, E. M.; Discher, D. E. Pharm. Res. 2007, 24, 2099–2109. (7) Kwon, G. S.; Kataoka, K. AdV. Drug DeliVery ReV. 1995, 16, 295–309. (8) Soo, P. L.; Eisenberg, A. J. Polym. Sci. Polym. Phys. 2004, 42, 923–938. (9) Sastry, N. V.; Hoffmann, H. Colloid Surface A 2004, 250, 247–261. (10) Dan, N.; Safran, S. A. AdV. Colloid Interface Sci. 2006, 123-126, 323– 331. (11) Hecht, E.; Mortensen, K.; Gradzielski, M.; Hoffmann, H. J. Phys. Chem. 1995, 99, 4866–4874. (12) Santore, M. M.; Discher, D. E.; Won, Y. Y.; Bates, F. S.; Hammer, D. A. Langmuir 2002, 18, 7299–7308. (13) Uddin, M. H.; Morales, D.; Kunieda, H. J. Colloid Interface Sci. 2005, 285, 373–381. (14) Dan, N.; Shimoni, K.; Pata, V.; Danino, D. Langmuir 2006, 22, 9860– 9865. (15) Pata, V.; Ahmed, F.; Discher, D. E.; Dan, N. Langmuir 2004, 20, 3888– 3893.

undergo a transition to smaller structures in the presence of low surfactant concentrations. While the effect of low molecular weight amphiphiles on polymeric vesicles has been documented, less is known regarding their effect on polymeric micelles. Early neutron scattering16 and NMR17 studies suggested uniform mixing between nonionic copolymers and the surfactant sodium dodecyl sulfate (SDS). Copolymer micelle destruction by SDS and dissolution until complete solubilization was also reported by Hecht and Hoffmann18 and Desai et al.19 All these studies indicated mixed aggregates containing a single or two macromolecules within the surfactant cluster coexisting with pure surfactant micelles, in excess surfactant. More recently, Bahadur and co-workers20 reported on a decrease followed by an increase in the size of the polymer-surfactant mixed aggregates with increasing surfactant concentration, for Pluronic and ionic low molecular weight amphiphile mixtures. They, too, found coexistence of large mixed assemblies with smaller pure surfactant micelles. Kelarakis and co-workers showed that, depending on the block polymer type, a given charged surfactant may induce growth of large surfactantspolymer mixed aggregates or lead to the formation of smaller assemblies compared to those formed by the pure polymeric micelles.21 Formation of supramolecular clusters was also reported for polystyrene-polyethylene oxide block copolymer with added surfactant.22 (16) Cabane, B.; Duplessix, R. J. Phys. (Paris) 1982, 43, 1529–1542. (17) Almgren, M.; Vanstam, J.; Lindblad, C.; Li, P. Y.; Stilbs, P.; Bahadur, P. J. Phys. Chem. 1991, 95, 5677–5684. (18) Hecht, E.; Hoffmann, H. Langmuir 1994, 10, 86–91. (19) Desai, P. R.; Jain, N. J.; Sharma, R. K.; Bahadur, P. Colloid Surf., A 2001, 178, 57–69. (20) Bharatiya, B.; Ghosh, G.; Bahadur, P.; Mata, J. J. Dispers. Sci. Technol. 2008, 29, 696–701. (21) Kelarakis, A.; Chaibundit, C.; Krysmann, M. J.; Havredaki, V.; Viras, K.; Hamley, I. W. J. Colloid Interface Sci. 2009, 330, 67–72.

10.1021/la8037439 CCC: $40.75  2009 American Chemical Society Published on Web 02/02/2009

Dissolution of Spherical Block Copolymer Micelles

Only a few reports examined the interaction in uncharged copolymer-surfactant systems. Zheng and Davis23 investigated by cryogenic transmission electron microscopy (cryo-TEM) the dissolution of spherical micelles composed of diblock copolymers by a nonionic amphiphile (C12E5) known to form cylindrical micelles under the experimental conditions used. The copolymersurfactant mixtures resembled common mixed surfactants systems and displayed good mutual miscibility, even when only ∼5% (molar) copolymer was present. Other studies, with wormlike micelles of a nonionic triblock copolymer indicated transition into smaller spherical ones when mixed with C12E5. In this system, however, complete disappearance of copolymer micelles was not observed even at high surfactant loadings.23 Similarly, Nordskog et al.24 found that the size of diblock copolymer wormlike micelles decreased with SDS concentration until, at a certain limiting concentration of surfactant, only spherical mixed micelles with a characteristic radius were observed. Increasing the SDS concentration did not further affect the micelle size. What determines the type of structures formed in mixtures of diblock copolymers and (nonionic) amphiphiles? In all cases, the effective MW decreases as the small amphiphile concentration increases. However, the amphiphile also affects the ‘packing parameter’ of the mixture, namely, the preferred geometry of the aggregate.25 As discussed by Discher and Eisenberg,26 and shown by Jain and Bates,27 diblock copolymers with relatively large hydrophobic tail groups form bilayers. As the ratio of tail to head MW decreases, the phases formed are as follows: coexistence between bilayers and cylindrical micelles, cylindrical micelles, coexistence between cylindrical and spherical micelles, and in systems where the head groups are relatively large, spherical micelles. Thus, mixing copolymers with a particular packing parameter and preferred geometry, with an amphiphile with a different packing parameter and preferred geometry, could yield either coexistence of the copolymer and amphiphilic structures, if there is no mixing, or mixed structures with a different geometry and size. If mixing does occur, the effective tail-to-head ratio may be taken to be proportional to the amphiphile content. The goal of this paper is to examine the effect of small molecule amphiphiles on the structure and packing of spherical amphiphilic polymer micelles. Because indirect methods cannot conclusively determine the geometry of the structures,9 we use cryo-TEM to define the aggregates’ morphology and directly resolve fine, local, and aggregate-specific assembly details. Cryo-TEM has been successfully applied to numerous polymer systems of single molecules and mixtures.14,23,27-31 In addition to showing the morphology, this methodology reveals intermediate states and the coexistence of multiple structures. In some block-copolymer micelles cryo-TEM successfully resolved the micelle core and corona23,29 and disclosed gentle morphological features such as the formation of junctions,27 swollen micellar endcaps,29 and SAB (sort-armed branched) micelles.14 (22) 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. B 2001, 105, 9077–9082. (23) Zheng, Y.; Davis, H. T. Langmuir 2000, 16, 6453–6459. (24) Nordskog, A.; Futterer, T.; von Berlepsch, H.; Bottcher, C.; Heinemann, A.; Schlaad, H.; Hellweg, T. Phys. Chem. Chem. Phys. 2004, 6, 3123–3129. (25) Israelachvili, J. N.; Mitchell, D. J.; Ninham, B. W. J. Chem. Soc., Faraday Trans. 2 1976, 72, 1525–1568. (26) Discher, D. E.; Eisenberg, A. Science 2002, 297, 967–973. (27) Jain, S.; Bates, F. S. Science 2003, 300, 460–464. (28) Jain, S.; Bates, F. S. Macromolecules 2004, 37, 1511–1523. (29) Zheng, Y.; Won, Y. Y.; Bates, F. S.; Davis, H. T.; Scriven, L. E.; Talmon, Y. J. Phys. Chem. B 1999, 103, 10331–10334. (30) Li, Z. B.; Kesselman, E.; Talmon, Y.; Hillmyer, M. A.; Lodge, T. P. Science 2004, 306, 98–101. (31) Won, Y. Y. Korean J. Chem. Eng. 2004, 21, 296–302.

Langmuir, Vol. 25, No. 5, 2009 2737 Table 1. Properties of the PB-PEO Diblock-Copolymer Used in the Solubilization Studya

sample name

formula

Mtotal

P3017-BdEO (P30) Bd102-EO114 10500

WEO MPEO (weight poly(g/mol) fraction) dispersity 5000

0.48

1.05

a

Bd, butadiene; EO, ethylene oxide; MPEO, molecular weight of the PEO block; WEO, weight fraction of EO group.

The copolymer-amphiphile system we studied was chosen to satisfy several criteria. First, the copolymer phase diagram had to be well understood, so that copolymer-surfactant structures could be directly compared to the pure copolymer system. The (pure) surfactant had to be known to form a single type of aggregate, independent of concentration, temperature, or ionic strength, so that the appearance of new structures could be directly related to mixing. This requirement eliminates ionic surfactants such as SDS, as well as the nonionic CiEj type surfactants whose aggregate type varies with both temperature and surfactant concentration. Our chosen polymer, poly(butadiene)-co-poly(ethylene oxide), PB-PEO, has been examined as a function of the copolymer composition and MW by Jain and Bates,27 and polymers similar to the one used in this study were shown to form spherical micelles. The nonionic surfactant chosen is Triton X-100, widely used in dissolution studies and known to form only spherical micelles under all system conditions. Because both the block copolymer and surfactant have EO head groups, good compatibility between the two was expected.

Materials and Methods Materials. A nonionic diblock-copolymer P3017-BdEO (denoted by P30) was purchased from Polymer Source Inc. (Canada). The block copolymer properties are presented in Table 1. Phosphatebuffered saline (PBS) was purchased from Biological Industries and filtered through a 0.22 µm Poretics polycarbonate membrane (Osmonics Inc.). Triton X-100 (4-(1,1,3,3-tetramethylbutyl)phenylpolyethylene glycol, t-Oct-C6H4-(OCH2CH2)xOH, x ) 9-10) was purchased from BioLab and used as received. Micelle Preparation and Solubilization. Micelles were prepared by film rehydration, following the Avanti Polar Lipid procedure.32 Briefly, a given quantity of amphiphilic block copolymer in a glass tube was dissolved in chloroform and dried to the form of a thin film by manually rotating the tube while adding a gentle stream of nitrogen gas. The dry polymer was then kept under vacuum overnight to ensure complete evaporation of the organic solvent. To form the micelles, a known amount of PBS was added to the dry polymer film, and the polymer-solvent mixture was incubated in an oven at 60 °C for 48 h. This procedure ensures the entire polymer film is solubilized in the PBS, allowing accurate determination of the polymer concentration. To study the solubilization of the block copolymer-based aggregates, P30 solutions were mixed with known quantities of a micellar Triton X-100 solution, to yield the required polymer-to-surfactant ratio and the required final concentration. Spectrophotometery. The transformation of the block copolymer nanostructures into smaller ones following the addition of Triton X-100 was monitored with a Milton Roy Spectronic 301 spectrophotometer. The measurements were carried out at a wavelength of 380 nm, as it provided the largest range between turbidity of pure polymer and surfactant solutions. These turbidity changes enabled location of the surfactant concentration range where the dissolution of the copolymer-based assemblies took place. Dynamic Light Scattering (DLS). DLS measurements of 4 mg/ mL block copolymer micellar solutions and Triton X-100 concentrations of up to 10 mM were performed with a Brookhaven laser light scattering system (BIC, BI-200SM Research Goniometer (32) http://www.avantilipids.com/PreparationOfLiposomes.html (accessed Nov 10, 2008).

2738 Langmuir, Vol. 25, No. 5, 2009 System). Experiments were conducted with a diode-pumped solidstate 300 mW laser at a wavelength of 532 nm, at a constant temperature of 25 °C. The scattered light was detected at an angle of 90° by a photomultiplier tube. The hydrodynamic radii, RH, of polymer micelles and the mixed aggregates were determined from the Stokes-Einstein equation. The intensity autocorrelation function was measured with a BI-9000AT digital signal processor and analyzed using the CONTIN method. Cryogenic-Transmission Electron Microscopy (cryo-TEM). Specimens were prepared in a controlled environment vitrification system (CEVS)33 at a controlled temperature of 25 °C and at saturation. We used two preparation procedures: In the ‘conventional’ procedure (procedure I), the diblock-copolymer and surfactant solutions were mixed in a vial, and a drop of the mixture was then applied to a Formvar-coated TEM grid held by tweezers inside the CEVS. In the second procedure (procedure II), on-the-grid-processing (OTGP) was used:34 a drop of the block copolymer solution was placed on the TEM grid, and then a drop of the surfactant solution was added; thus, mixing occurred on the grid, resulting in contact times between the polymer and surfactant of only a few seconds.14,35,36 In both procedures the drop was blotted, and the sample was plunged into liquid ethane to form a vitrified specimen and transferred to liquid nitrogen for storage. Specimens were examined in a Philips CM120 TEM at 120 kV. The temperature was kept below -175 °C. Images were recorded digitally, under low-dose conditions, on a cooled Gatan MultiScan 791 CCD camera using DigitalMicrograph 3.1 software.37 Image Processing. The core and the corona diameters of the pure block copolymer micelles were measured using the Nikon NIS image processing software. All the spherical micelles in the field of view were analyzed, taken from several images of various preparations. These measurements allowed analysis of the micelle size distribution and calculation of their average aggregation number. We also measured the diameter of copolymer-surfactant mixed micelles at intermediate and complete solubilization, namely, after treatment with the highest surfactant concentrations. This analysis was further used to estimate the number of polymer molecules in the smallest, saturated mixed micelles.

Results and Discussion The phase diagram constructed by Jain and Bates27,28 predicts the formation of spherical micelles by P30 (Bd102-EO114), in coexistence with some cylindrical ones. The formation of spherical micelles by P30 also meets the general criteria set by Discher and Eisenberg,26 according to which spheres are created when the molecular weight fraction of the hydrophilic block to the total molecular weight of the copolymer (fhydrophilic) > 0.45. As shown in Figure 1, we find that in aqueous solution the pure copolymer indeed forms primarily spherical micelles and very few elongated micelles. As expected for micelles of this block copolymer family,14,29 structural details of the micelles, including their core and corona domains (see white and black arrows in Figure 1A) and swollen micellar end-caps (white arrowheads), are clearly resolved. Overall, we find the formation of these micelle populations to be insensitive to the copolymer concentration in solution in the range of concentration studied (2-10 mg/mL), in agreement with previous work on PB-PEO block copolymers.38 (33) Bellare, J. R.; Davis, H. T.; Scriven, L. E.; Talmon, Y. J. Electron Microsc. Tech. 1988, 10, 87–111. (34) Danino, D.; Talmon, Y. In Physical Chemistry of Biological Interfaces; Baskin, W.; Norde, E., Ed.; Marcel Dekker: New York, 2000; Ch. 24, pp 799821. (35) Danino, D.; Moon, K. H.; Hinshaw, J. E. J. Struct. Biol. 2004, 147, 259–267. (36) Konikoff, F. M.; Danino, D.; Weihs, D.; Rubin, M.; Talmon, Y. Hepatology 2000, 31, 261–268. (37) Danino, D.; Bernheim-Groswasser, A.; Talmon, Y. Colloid Surf., A 2001, 183, 113–122. (38) Won, Y. Y.; Davis, H. T.; Bates, F. S.; Agamalian, M.; Wignall, G. D. J. Phys. Chem. B 2000, 104, 9054–9054.

Shimoni and Danino

Figure 1. Representative cryo-TEM images of pure P30 (fEO ) 0.48) in PBS, at 10 mg/mL (A, C) and 4 mg/mL (B). In both concentrations mainly uniform spherical micelles form, ∼75 nm in diameter, in coexistence with a small population of short threadlike micelles. The hydrophilic PEO corona and hydrophobic PB core are evident, marked in A by black and white arrows, respectively. White arrowheads show the swollen micelle end-caps. These details are also highlighted in the higher magnification section shown in C. Bars equal 50 nm in all panels.

Several studies applied direct imaging cryo-TEM to gain quantitative measures on self-assembled nanostructures. Examples include an early work where the evolution of size distribution of emulsion droplets by Ostwald ripening39 was calculated and studies by Zasadzinski and co-workers40,41 where cryo-TEM techniques were used to measure the size distribution of vesicles and estimate membrane elastic and thermodynamic properties. Here we used cryo-TEM to quantify the pure polymer micellar dimensions and polydispersity, as well as to estimate their aggregation number. As displayed in Figure 2, we find that over 80% of the polymeric micelles are characterized with a mean core diameter, DC, of 40.5 ( 2.5 nm. This value is in excellent agreement with a core diameter of ∼39 nm calculated by Jain et al.28 for a PB block with a number-average molecular weight of 5850. The value we found also agrees well with the study of Poppe et al.42 where micelles with DC of 35 nm were reported for a poly(ethyleneco-propylene)-PEO (PEP-PEO) diblock copolymer with core MW of 5970 and corona of 5130 g/mol, similar to the chain lengths and hydrophilic-hydrophobic ratio of our copolymer. To estimate the aggregation number, we multiply the core volume by the density of polybutadiene, which is on the order of 0.93 g/cm3.43 The resulting core mass is then divided by the butadiene chain MW, yielding an aggregation number of ∼3500. Our cryo-TEM micrographs also allow measurement of the micelles’ mean diameter, DM, finding that almost 80% of the spherical micelles have a mean diameter of 75.5 ( 3.5 nm, thus an overall mean corona dimension of ∼35.5 nm (a radius of ∼18 nm). The ratio between the micelles core and corona radii fits (39) De Smet, Y.; Danino, D.; Deriemaeker, L.; Talmon, Y.; Finsy, R. Langmuir 2000, 16, 961–967. (40) Jung, H. T.; Lee, S. Y.; Kaler, E. W.; Coldren, B.; Zasadzinski, J. A. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15318–15322. (41) van Zanten, R.; Zasadzinski, J. A. Curr. Opin. Colloid Interface Sci. 2005, 10, 261–268. (42) Poppe, A.; Willner, L.; Allgaier, J.; Stellbrink, J.; Richter, D. Macromolecules 1997, 30, 7462–7471. (43) Brandrup, J.; Immergut, E. H.; Grulke, E. A. Polymer Handbook; Wiley: New York, 1999.

Dissolution of Spherical Block Copolymer Micelles

Figure 2. Statistical analysis performed on 250 spherical micelles shows a relatively uniform population of micelles, with ∼80% of the micelles having a mean core diameter of 40.5 ( 2.5 nm and a total micelle diameter of 75.5 ( 3.5 nm.

Figure 3. Apparent absorbance of 4 mg/mL diblock copolymer P30 solutions containing increasing Triton X-100 concentrations. Arrows point to surfactant concentrations further examined by cryo-TEM and DLS.

very well with the chemical composition of P30 (Table 1). The corona block might be partially solvated by water; hence, it is not possible to use this measurement to calculate the aggregation number. Yet, roughly, following the approach described above, we find a good correlation between the chemical composition of our block copolymer and the measured volumes of the core and corona. The low polydispersity in the micelle diameter may be attributed to the distribution in the block copolymer molecular weight (see Table 1). To examine the effect of Triton X-100 on the structure of the diblock copolymer micelles, we first measure the apparent absorbance (turbidity) of the system as a function of the surfactant content. As shown in Figure 3, we find that the absorbance decreases steeply as a function of Triton X-100 concentration, indicating that the solution becomes transparent, until reaching a plateau at high surfactant concentrations (∼4 mM). The value of the absorbance in the plateau region (namely, in excess surfactant) is somewhat higher than that expected for a suspension of surfactant micelles of similar concentration, indicating the presence of some larger objects whose size and concentration does not vary with increasing surfactant content. This type of turbidity profile seems to suggest a continuous process and is

Langmuir, Vol. 25, No. 5, 2009 2739

quite different from the multistage turbidity profile found in mixtures of Triton X-100 with vesicle-forming copolymers. The turbidity profile shown in Figure 3 for the micellar copolymer/surfactant mixtures can be explained by two, very different processes. In the first, the polymeric micelle size decreases continuously with Triton X-100 concentration (namely, second order-like transition), until reaching a limit value. The second possibility is that mixing with surfactant leads to coexistence between the large, pure polymer micelles and small polymer/surfactant mixed micelles (first-order-like process). As the surfactant concentration increases, the number of pure polymer micelles decreases, and that of mixed ones increases, until reaching a limit where all pure polymer micelles are dissolved. In both scenarios, at high surfactant concentrations, excess surfactant forms pure Triton X-100 micelles, whose effect on the measured turbidity is small. The turbidity data reveal that the dissolution of the P30 spherical copolymer micelles by Triton X-100 is initiated at extremely low surfactant concentrations. The composition at which the plateau is reached (∼4 mM Triton X-100 to 4 mg/mL block copolymer, i.e., ∼0.38 mM polymer) may be taken to be an indicator of the composition of the ‘equilibrium’ mixed polymer/ surfactant aggregates. Thus, the surfactant-saturated polymer aggregate is roughly composed of 10 Triton X-100 molecules per one polymer chain. Turbidity, however, cannot distinguish between the two scenarios, first- or second-order transition. Moreover, the turbidity measurements cannot be used to determine the mixed micelles’ shape and dimensions or to evaluate the time scales over which structural transitions occur. We therefore apply cryo-TEM. To distinguish between the two possible scenarios (first- or second-order transition), we examine the size of the mixed polymer/surfactant micelles in three surfactant-to-polymer molar ratios: 1:1 (0.4 mM surfactant), 1:8 (3 mM surfactant), and 1:26 (10 mM surfactant). As indicated in Figure 3, these values correspond to the initial polymeric micelle dissolution, mid to late stages, and excess surfactant. To evaluate the rate of structural transition in the mixed polymer/surfactant system, we use on-the-grid-processing (OTGP),35,36 which allows fixation of the structures a few seconds after mixing the polymer solution with the surfactant solution. As shown in Figure 4, we find structural changes, within a few seconds of mixing, and formation of mixed micelles that are smaller in size. The sensitivity and rapidity of OTGP cryomicroscopy also enables us to capture the mechanism by which the larger, pure copolymer micelles transition into the smaller, surfactant-polymer mixed ones. We find elongation of the polymer micelles, creation of pinnate-shaped aggregates, and then ‘dimerization’ into two smaller micelles. It should be noted that overlap of pure copolymer micelles is suppressed (see Figure 1) by the strong steric repulsion between the polymeric coronas. Interestingly, as seen in Figure 4, most pair micelles have equal diameters, and only a few pairs with unequal diameter are seen. The mean core diameter of the newly formed micelles is ∼31 nm, which suggests an aggregation number of ∼1700 molecules per micelle (Table 2). This is roughly half the number of molecules per micelle in the untreated micelles, in support of dimerization. We also made measurements on several unequal pairs. The deviation in the total aggregation numbers of coupled micelles compared with Nagg of the untreated micelles was ∼10%, again agreeing with their formation via dimerization. Thus, overall, the data suggest a stepwise breaking of the micelles and the formation of micelles of defined sizes. Note also that, unlike in the pure polymer micelles, the corona region is not seen in

2740 Langmuir, Vol. 25, No. 5, 2009

Shimoni and Danino

Figure 4. Cryo-TEM images of 4 mg/mL P30 micelles mixed on the grid with 0.4 mM Triton X-100 (final concentration), after 7-12 s of contact time between polymer and surfactant. (a) Upon addition of surfactant the spherical micelles become distorted and adopt a peanutlike shape (white arrows). However, they remain intact, and their hydrophilic corona remains visible. (b) Next, dimerization of the block copolymer micelles takes place, and pairs of round, smaller micelles form (black arrows), whose corona is invisible because of looser packing of the ethylene oxide chains caused upon mixing with the surfactant molecules. Most pair micelles have identical diameters, although a few unequal micelle pairs also form. The distribution in the micelle diameter is, however, very small. The newly formed micelles often overlap and appear darker at the overlapping regions. Two pairs of partially overlapping newly formed micelles are enclosed within the white dashed rectangle drawn in b.

Figure 5. Representative cryo-TEM images of mixtures containing Bd102-EO114 (P30) and Triton X-100. (a) 3 mM Triton X-100, 5 min after mixing; (b) 10 mM Triton X-100, 5 min after mixing; (c) 10 mM Triton X-100, 9 days after mixing. In all images mixed micelles of fairly uniform diameter (∼22 nm, few are marked with white arrowheads) coexist with small spherical surfactant micelles (seen as small black dots in the images). Images are displayed at identical magnifications. Bar ) 50 nm.

Table 2. Cryo-TEM Measurements of the Mean Diameters of the Copolymer P30 and the Mixed Micelles, and Aggregation Numbers, as a Function of Triton X-100 Concentration Triton X-100 concentration (mM) 0 0.4 3 10

stage of the process pure block copolymer micelles onset of mixed copolymersurfactant micelle formation saturation of mixed micelles high surfactant excess

mean micelle core diameter (nm)

aggregation number

40 31

3500 1700

21-22 21-22

600 600

the mixed micelles. This is due to the lower density of the polymer EO chains in the corona resulting from the dilution by the surfactant. At the limit of excess surfactant (above 3 mM, or 1:8 molar polymer to surfactant ratio), coexistence between two uniform micellar populations is evident, as shown in Figure 5a. One population is that of mixed micelles with core size of 21-22 nm (with unresolved corona), approximately half the core diameter of the pure polymer micelles. These are taken to be the surfactantsaturated polymeric micelles. The others are much smaller pure Triton X-100 micelles, which are ∼4 nm in diameter. The number of copolymer molecules in the saturated mixed micelles was found to be on the order of 600 (about 15% of the aggregation

Figure 6. Schematic representation of the block copolymer molecules packing (a) alone, and (b) in the saturated mixed micelle. It is suggested that the Triton X-100 hydrophobic tail penetrates the PB core, and the head groups of the two classes of molecules, both which are made of PEO chains, form the mixed-micelle corona.

number of the pure copolymer micelle), suggesting that each block copolymer micelle breaks into about six smaller micelles. Interestingly, as shown in Figure 5b and 5c, even at a high excess of surfactant, no further decrease in the size of the micelles was

Dissolution of Spherical Block Copolymer Micelles

Langmuir, Vol. 25, No. 5, 2009 2741

Figure 7. (A, B) Cryo-TEM image of pure Bd46-EO30 vesicles. (C) Representative cryo-TEM image of a mixture containing Bd46-EO30 and Triton X-100 showing a bimodal distribution of micelles: mixed polymer-surfactant micelles are marked with black arrowheads and smaller Triton X-100 micelles with white arrowheads.

noted. Together, our data lead to two main conclusions: (1) packing constraints force the arrangement of the block copolymer molecules into micelles of discrete sizes, and, (2) because of the increase in the surface area-per volume ratio when the micelles decrease in size, the hydrophilic chains fail to effectively coat the hydrophobic domain. This prevents complete solubilization of the block copolymer micelles and sets a minimum for the mixed spherical micelle size. Specifically for P30, we find the minimal core diameter in excess surfactant to be ∼22 nm, with ∼600 assembled polymer molecules in a mixed micelle. Schematic representations of the pure block copolymer micelle and the mixed saturated micelle are shown in Figure 6. We thought the restricted solubilization is either due to the large hydrophobic chain (Bd102) of the macromolecule or due to the hydrophobic-to-hydrophilic molecular weight ratio which for P30 is ∼1 (see Table 1). Previous cryo-TEM studies by Zheng and Davis23 described two pathways for solubilization of di- and triblock copolymer micelles by the nonionic surfactant C12E5. For the Bd45-EO124 diblock polymer (small hydrophobic chain, hydrophobic-to-hydrophilic ratio 1.5, supports our conclusion that imperfect dissolution is linked with the hydrophobic-to-hydrophilic molecular weight ratio, rather than with the hydrophobic block size. The size of the mixed micelles we find (14 nm ( 1 nm) also agrees well with the 13 nm reported by Nordskog.24 Another clear finding is the coexistence of the large mixed polymer-surfactant micelles with small surfactant micelles in excess surfactant. The latter assemblies are small (∼4 nm in diameter) and only become directly visible at high resolution imaging and when optimal sample and imaging conditions are being applied. We also studied the effect of mixing time on the observed structures, by examining samples prepared shortly after mixing, as well as after long periods of time when equilibrium should have been reached. As shown in Figure 5, we find that the mixed structures seen immediately after mixing the surfactant solution with the copolymer suspension remain stable over days and even weeks. The only difference in their appearance was that when prepared for cryo-TEM using conventional mixing in a tube (rather than OTGP), the micelles were distributed more randomly on the grid. Their morphology, size, and appearance, however, were identical to those observed using OTGP. To estimate micelle size and further confirm that the assemblies formed shortly after mixing do not change with time, DLS measurements were performed 10 min and 2 months after mixing the copolymer and surfactant solutions. In support of the cryoTEM data, no significant changes in particle size were detected as a function of time. Additionally, we find DLS measurements to be complementary to turbidity and cryo-TEM, showing a decrease in the particle size, as more surfactant is present in solution, up to 4 mM. Specifically, a mean diameter of 79 nm in the absence of surfactant, and 38 and 28 nm in presence of 0.4 and 3 mM Triton X-100, respectively, were measured by DLS. Past studies indicated that DLS is shifted toward larger particles compared to direct cryo-TEM observations,44,45 as also found here. The slightly larger hydrodynamic diameter of the pure copolymer micelles compared with measurements from the cryo-TEM data may also result from DLS measuring the (44) Coldren, B.; van Zanten, R.; Mackel, M. J.; Zasadzinski, J. A.; Jung, H. T. Langmuir 2003, 19, 5632–5639. (45) Hans, M.; Shimoni, K.; Danino, D.; Siegel, S. J.; Lowman, A. Biomacromolecules 2005, 6, 2708–2717.

Shimoni and Danino

entire micellar population, while our cryo-TEM analysis excluded the threadlike micelles. In the case of mixed micelles, a real comparison between the two methods cannot be made, since the entire micelle diameter is likely measured by DLS, while only the core is visible (and thus measured) by cryo-TEM.

Conclusions In this paper we examine the dissolution of diblock copolymer spherical micelles by a nonionic surfactant, Triton X-100, which is known to form spherical micelles under the conditions of the experiment. Unlike polymer-polymer mixtures where mixing did not take place over extremely long periods of time,28,31 we find structural transitions in the copolymer micelles within less than 10 s of mixing with surfactant. Moreover, the initial structures were the same as those found in the solutions days after mixing, which indicates rapid equilibration of the system. The time scales of mixing are similar to those found for dissolution of vesicleforming diblock copolymers with surfactants,14 and they show that the dynamics in the mixed surfactant-polymer system is controlled by the small molecule. Cryo-TEM revealed that upon mixing with Triton X-100 the micelle core diameter decreases from ∼40 nm in the pure polymer case to ∼22 nm at the limit value in excess surfactant (1:26 ratio). Our cryo-TEM data also enable determination of the mechanism of copolymer micelle dissolution, identifying that the copolymer micelles elongate and then ‘dimerize’, i.e., divide into two smaller micelles. At the limit of solubilization, coexistence of large mixed polymer-surfactant micelles and small surfactant micelles is found. Our new results and the literature data we analyze show that limited solubilization of the polymeric assemblies occurs (1) with di- and triblock copolymers, (2) in the presence of charged and uncharged surfactants, and (3) with polymeric molecules having small and large hydrophobic domains. Additionally we show that imperfect dissolution which was reported for threadlike micelle-forming block copolymers23,24 can occur with vesicleforming block copolymer (Figure 7) and even with spherical micelle-forming block copolymers (Figure 5). Thus, we interpret the lack of complete dissolution of the polymeric assemblies by packing constraints and the inability of the head-group chains to efficiently shield the hydrophobic domain from the solvent, when the hydrophobic domain size relative to the size of the hydrophilic part of the macromolecule is greater than ∼0.75. This seems to be a sufficient criterion to determine complete or restricted solubilization. Acknowledgment. We thank Nily Dan for many key discussions and careful reading of the manuscript. The work was supported in part by the Israel Science Foundation and the Russell Berrie Nanotechnology Institute (RBNI) at the Technion. The assistance of Mrs. Sigal Goren in the image processing is appreciated. LA8037439