Remarkable Viscoelasticity in Mixtures of ... - ACS Publications

Sep 9, 2014 - Received: July 28, 2014. Revised: August 28 ... the presence of C12EO3;20−22 the cosurfactant acts by reducing ... were recorded on a ...
0 downloads 0 Views 5MB Size
Download PDF
Article pubs.acs.org/Langmuir

Remarkable Viscoelasticity in Mixtures of Cyclodextrins and Nonionic Surfactants Á ngela García-Pérez,† Marcelo A. da Silva,† Jonny Eriksson,‡ Gustavo González-Gaitano,§ Margarita Valero,∥ and Cécile A. Dreiss*,† †

Institute of Pharmaceutical Science, King’s College London, Franklin-Wilkins Building, 150 Stamford Street, London SE1 9NH, UK Department of ChemistryBMC, Uppsala University, SE-751 23 Uppsala, Sweden § Departamento de Química y Edafología, Universidad de Navarra, 31080 Pamplona, Spain ∥ Departamento de Química Física, Universidad de Salamanca, 37008 Salamanca, Spain ‡

S Supporting Information *

ABSTRACT: We report the effect of native cyclodextrins (α, β, and γ) and selected derivatives in modulating the self-assembly of the nonionic surfactant polyoxyethylene cholesteryl ether (ChEO10) and its mixtures with triethylene glycol monododecyl ether (C12EO3), which form wormlike micelles. Cyclodextrins (CDs) generally induce micellar breakup through a host−guest interaction with surfactants; instead, we show that a constructive effect, leading to gel formation, is obtained with specific CDs and that the widely invoked host−guest interaction may not be the only key to the association. When added to wormlike micelles of ChEO10 and C12EO3, native β-CD, 2-hydroxyethyl-β-CD (HEBCD), and a sulfated sodium salt of β-CD (SULFBCD) induce a substantial increase of the viscoelasticity, while methylated CDs rupture the micelles, leading to a loss of the viscosity, and the other CDs studied (native α- and γ- and hydroxypropylated CDs) show a weak interaction. Most remarkably, the addition of HEBCD or SULFBCD to pure ChEO10 solutions (which are low-viscosity, Newtonian fluids of small, ellipsoidal micelles) induces the formation of transparent gels. The combination of small-angle neutron scattering, dynamic light scattering, and cryo-TEM reveals that both CDs drive the elongation of ChEO10 aggregates into an entangled network of wormlike micelles. 1H NMR and fluorescence spectroscopy demonstrate the formation of inclusion complexes between ChEO10 and methylated CDs, consistent with the demicellization observed. Instead, HEBCD forms a weak complex with ChEO10, while no complex is detected with SULFBCD. This shows that inclusion complex formation is not the determinant event leading to micellar growth. HEBCD:ChEO10 complex, which coexists with the aggregated surfactant, could act as a cosurfactant with a different headgroup area. For SULFBCD, intermolecular interactions via the external surface of the CD may be more relevant.

■

of starch contain six, seven, or eight glucose units, for α-, β-, and γ-cyclodextrins, respectively, resulting in different sizes of the apolar cavity. Additionally, industrially produced cyclodextrins bear differently substituted groups and induce different properties, in particular a better solubility of β-CD (markedly less soluble than α and γ species). Because of their ability to form soluble complexes, their low toxicity, and their immunogenicity, cyclodextrins are used in numerous industrial applications, such as in the food, pharmaceutical, and cosmetic industries.2−4 In this paper, we explore the ability of cyclodextrins to enhance, induce, or disrupt soft gels, based on simple surfactant systems. While supramolecular gels based on inclusion complexes with cyclodextrins have been reported, they are

INTRODUCTION A strong focus of research in soft matter over the past couple of decades has been dedicated to the design of new and responsive materials by exploiting noncovalent interactions. Host−guest chemistry, based on the favorable localization of a guest molecule within the cavity of a host, is a particularly versatile tool for creating soft materials, such as hydrogels, which have a plethora of applications as matrices for storage and delivery or artificial scaffolds for tissue repair.1 Cyclodextrins, with their toroidal shape and relatively hydrophobic cavity, are a particularly useful host molecule for this purpose, being available in large quantities at low-cost and largely biocompatible.2 Cyclodextrins are cyclic oligosaccharides comprising glucose units connected through α-(1,4)-glucosidic bonds. Their specific architecture imparts them the ability to form water-soluble inclusion complexes by encapsulating a variety of molecules inside their cavity through noncovalent interactions. The three main cyclodextrins formed by enzymatic degradation © 2014 American Chemical Society

Received: July 28, 2014 Revised: August 28, 2014 Published: September 9, 2014 11552

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

dextrins (HPACD, HPBCD, and HPGCD); randomly methylated βcyclodextrin (RAMEB); heptakis(2,6-di-O-methyl)-β-cyclodextrin (DIMEB, FW = 1331.36 g·mol−1); heptakis(2,3,6-tri-O-methyl)-βcyclodextrin (TRIMEB); 2-hydroxyethyl-β-cyclodextrin (HEBCD); and a sulfated sodium salt of β-cyclodextrin (SULFBCD) (structures of most compounds are given in our previous publication15 and are also available as Supporting Information, Scheme SI 1). For the methylated β-CDs, the degree of substitution (number of methyl groups per molecule of β-cyclodextrin) increases in the following order RAMEB < DIMEB < TRIMEB.24 For the remaining CDs, the degree of substitution is as follows: HPBCD, 0.8 molar substitution; HPGCD, 0.6 molar substitution; HEBCD, 0.7 molar substitution; SULFBCD, 12−15 substituents per molecule, as provided by the supplier. Pyrene was obtained from Sigma-Aldrich UK. For the preparation of the samples for SANS measurements, D2O was used instead of water (Euriso-top, 99.85% purity). For all other experiments, ultrapure water (18.2 MΩ·cm, Millipore-filtered) was used. Methods. Rheological Measurements. Small-amplitude oscillatory and steady-shear experiments were performed on a dynamic straincontrolled rheometer ARES (TA Instruments) using cone-and-plate geometry (50 mm, 0.02 rad), with a temperature-controlling Peltier unit and a solvent trap. Experiments were run at 20.0 °C. All solutions were left to rest at least 2 weeks at room temperature after preparation. They were then vortex-mixed and equilibrated at 20.0 °C (in a water bath) for 24 h before conducting the rheological measurements. Measurements were carried out in duplicates or triplicates, with very good reproducibility. Oscillatory frequency sweep tests covering the range from 100 to 0.1 rad·s−1 were performed in the linear viscoelastic regime (5% strain), as determined by dynamic strain sweep measurements. Steady-shear rate flow curves were acquired covering the range from 0.01 to 500 s−1. Fluorescence Measurements. Aqueous solutions were prepared containing 6.1 μM pyrene, an increasing concentration of ChEO10 (from 6 × 10−4 to 6 × 10−2 wt %), and either no cyclodextrin or a constant molar ratio of HPBCD or SULFBCD (with surfactant/CD = 0.62 for HEBCD and 1.24 for SULFBCD, corresponding to a ratio where a “gel-phase” was observed). The fluorescence emission spectra were recorded on a Cary Eclipse fluorescence spectrophotometer (Varian, Oxford, UK) with λexc = 340 nm at 373, 384, and 393 nm. The critical micelle concentration value was determined from the break of the curve showing the intensity vs ChEO10 concentration (from three repeated measurements). Nuclear Magnetic Resonance (NMR). Proton NMR spectra were recorded either on a Bruker Advance 400 MHz instrument (systems with DIMEB and SULFBCD) or on a Bruker Advance 700 MHz instrument (HEBCD system) by collecting 128 scans. Samples were prepared in D2O, using the residual HDO signal as the reference. Small-Angle Neutron Scattering (SANS). Small-angle neutron scattering experiments were performed on the instrument KWS-2 at the Jülich Centre for Neutron Science (JCNS), Munich, Germany. An incident wavelength of 4.97 Å was used with detector distances of 1.7 and 7.6 m to cover a q-range from 0.007 to 0.3 Å−1. The temperature was controlled with a Peltier system with an accuracy of 0.1 °C. All samples were measured in quartz cells (Hellma) with a path length of 2 mm, using D2O as the solvent (Aldrich, >99.9% in D; Armar Chemicals, >99.8% in D). Raw data were corrected for electronic background and empty cell and normalized by water using the software Qti-KWS v1.0. Data were fitted to an ellipsoidal cylinder model using the SasView 2.2.1 software package (NSF DANSE project).25 Further details on the modeling procedure can be found elsewhere.26 Dynamic Light Scattering. Size distributions were obtained by photon correlation spectrometry with a Malvern Zetasizer Nano ZS, with a laser wavelength of 633 nm. The samples were filtered prior to the measurements with 0.22 μm syringe filters into a semimicro glass cell. The temperature of the sample was controlled with 0.1 °C accuracy at 20.0 °C by a built-in Peltier unit. Data analysis of the autocorrelation function was carried out with the Zetasizer software in the high-resolution mode to distinguish potentially overlapping size distributions.

either based on the threading of cyclodextrins on polymeric chains or the inclusion of small guest molecules covalently attached to macromolecules.1,5 Cyclodextrin/surfactant gels instead have scarcely been reported.6,7 Generally also, the interactions of cyclodextrins with surfactants have mostly focused on the low concentration range [below the critical micelle concentration (cmc)], rather than on micellar aggregates.8−10 Cyclodextrins may act with surfactants as structural “modulators”, in the vast majority of cases disrupting the organized assemblies by “snatching” one (or more) of the components by forming an inclusion complex.11,12 In a few instances, however, CDs have been reported to act as “constructive” modulators, promoting aggregate growth in surfactant mixtures,11,13 either by the selective removal of the less bound surfactant molecules from the aggregates or through CD:guest complex coaggregating through noninclusion complex formation 14 or other mechanisms. The effect of CD:surfactant complexation on wormlike micelles, however, has been scarcely investigated,15−17 and the use of CDs as a handle to tune the rheological response of micellar solutions has not been reported, except in our earlier report.15 Wormlike micelles (WLM) are elongated, and flexible aggregates of surfactants, which, above a threshold concentration, entangle into transient viscoelastic networks, showing, in some systems, remarkable “gel-like” properties.18,19 The expected outcome of cyclodextrins−WLM interaction is a disruption of the micellar aggregates by complexation, leading to a collapse of the networks and thus a drop in rheological properties, hence offering a handle for an effective “sol/gel” transition. However, in a recent study on wormlike micelles based on two nonionic surfactants, polyoxyethylene cholesteryl ether (ChEO10, a surfactant based on cholesterol linked to 10 oxyethylene units) and monocaprylin (MCL), we recently demonstrated that, depending on the type of substitution of the β-cyclodextrin, they could act either as destructive or constructive modulators.15 In this paper, we extend our study to a different surfactant system (replacing the cosurfactant MCL with triethylene glycol monododecyl ether, C12EO3) and a wider range of cyclodextrins, including native α-, β- and γCDs and their derivatives, to assess the impact of cavity size and substitution. ChEO10 is known to form wormlike micelles in the presence of C12EO3;20−22 the cosurfactant acts by reducing the effective headgroup area per molecule, increasing the value of the packing parameter23 and leading to uniaxial micellar growth. In addition, we also study the interaction of cyclodextrins with the single surfactant system (ChEO10) and show for the first time the formation of transparent gels in mixtures of cyclodextrins to a liquid-like micellar solution. The rheological behavior of the solutions is characterized by flow curves and small-amplitude oscillatory shear frequency sweeps. Fluorescence and NMR spectroscopy are used to gain insight into the interactions at the origin of the drastic macroscopic changes observed. Finally, small-angle neutron scattering, dynamic light scattering, and cryo-TEM are used to reveal the nanoscale morphology underlying the surprising gellike behavior.

■

EXPERIMENTAL SECTION

Materials. ChEO10 was purchased from Ikeda Corp. (Yokohama, Japan) and C12EO3 was obtained from TCI Europe N. V. (Zwijndrecht, Belgium). All cyclodextrins were obtained from Sigma-Aldrich UK: native α-, β-, and γ-cyclodextrins; their hydroxypropyl derivatives 2-hydroxypropyl-α-, -β-, and -γ-cyclo11553

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

Figure 1. (A) Oscillatory frequency sweep measurements showing the storage (G′) and loss (G″) moduli as a function of angular frequency (ω). (B) Flow curves for 10 wt % ChEO10 (aq) solutions with 0.0, 1.0, and 2.5 wt % C12EO3. (C and D) Effect of adding methylated β-cyclodextrins (9 wt %) on the rheological response of ChEO10/C12EO3 solutions. Cryogenic Transmission Electron Microscopy (cryo-TEM). CryoTEM measurements were performed with a Zeiss TEM Libra 120 instrument (Carl Zeiss NTS). The instrument was operated at 80 kV in zero-loss bright-field mode. Digital images were recorded under low-dose conditions with a BioVision Pro-SM Slow Scan CCD camera system (Proscan elektronische Systeme GmbH) and iTEM software (Olympus Soft Imaging Solutions GmbH). An underfocus of 1−3 μm was used in order to enhance contrast. The preparation procedure has been described in detail elsewhere.27 Specimens were prepared in a climate chamber with temperature and humidity control (temperature of 25 °C and relative humidity of approximately 98−99%). Thin films were prepared by placing a small drop of the sample on a copper-gridsupported perforated polymer film, covered with thin carbon layers on both sides. After the drop was blotted with filter paper, thin sample films (10−500 nm) spanned the holes in the polymer film. Immediately after blotting, samples were vitrified in liquid ethane just above its freezing point. Samples were kept below −165 °C and protected against atmospheric conditions during both transfer to the TEM and examination. Several images were taken of each sample studied, and representative examples are presented here.

wormlike micelles, but this unusual behavior has been reported in some wormlike micelle systems.19,28 Steady-state measurements show a typical Newtonian behavior in the absence of or at low concentrations of C12EO3, which becomes shearthinning beyond ∼1 wt %, marking the onset of WLM entanglement (Figure 1B). In the remainder of the paper, wormlike micelle composition is kept constant at a 10 wt % ChEO10/2.5 wt % C12EO3 ratio. The effect of adding varying concentrations of a range of cyclodextrinsboth native and modifiedto viscoelastic ChEO10/C12EO3 solutions was studied next. Due to the low solubility of β-CD in water (ca. 1.8 wt %), wormlike micelle solutions were diluted down to ChEO10 0.63 wt %/C12EO3 0.18 wt % in mixtures with native β-CD to keep a comparable surfactant/CD ratio to the mixtures with the other CDs studied. Figure 1C,D shows that the addition of the randomly methylated and di- and trimethylated β-CDs produces a remarkable loss of the solid-like viscoelastic behavior, as visible from the drop in viscosity and storage and loss moduli. This effect is dependent on the extent of CD methylation: the largest effect is produced by the thrice substituted CD (TRIMEB), where G″ is higher than G′ over the whole range of frequencies and the viscosity drops the most dramatically, followed by the twice (DIMEB) and then randomly methylated CD (RAMEB). This effect concurs with the expected scenario of a CD:surfactant host−guest interaction in which the hydrophobic tail of the surfactant forms a complex with the cavity of the cyclodextrin, thus competing with micelle formation.11,15 In contrast, Figure 2A,B shows that the addition of native βCD induces a remarkable increase of the solid-like response of the WLM. Starting from a ∼14 times diluted WLM solution

■

RESULTS Rheology of Cyclodextrins and Wormlike Micelles: Breakup or Reinforcement. Mixtures of 10 wt % ChEO10 with C12EO3 in water form highly viscoelastic wormlike micellar solutions over a range of concentrations.21,22 Frequency sweeps reveal a change from a predominantly liquid-like response below C12EO3 ∼ 1 wt % to a viscoelastic, solid-like behavior. This is typical of a transition from short micelles to a network of entangled wormlike micelles, as has been reported for this system.22 G′ is largely independent of frequency over the range measured [Figures 1A and SI 1, Supporting Information (SI)] and as such departs from the typical Maxwell response of 11554

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

departing from the expected host−guest interaction, leading to surfactants being extracted from the micelles. The most notable finding is that the addition of either hydroxyethyl-β-CD (HEBCD) or β-CD sulfated sodium salt enhances the elasticity of the systems (Figure 3). This reinforcing effect of HEBCD had also been reported in ChEO10/MCL wormlike micellar systems,15 while SULFBCD has not been studied before.

Figure 2. Effect of native cyclodextrins on ChEO10/C12EO3 wormlike micelles. (A) Oscillatory frequency sweeps of a ∼14-times dilution of ChEO10/C12EO3 (0.63 wt %/0.18 wt %) solutions with and without native β-CD. (B) Flow curves of the samples shown in part A. (C) Appearance of mixtures of ChEO10/C12EO3 wormlike micelles (10/2.5 wt %) with native γ-CD. (D) Appearance of sample C after lifting the upper plate of the rheometer, showing its stringy appearance.

Figure 3. Rheology of ChEO10 10% (wt %)/C12EO3 2.5% (wt %) WLM with and without 9 wt % HEBCD or 6 wt % SULFBCD.

Making Gels in Mixtures of Cyclodextrins and One Surfactant. Macroscopic Behavior. Following the study on the ChEO10/C12EO3 surfactant system, we next turn to the effect of adding cyclodextrins to single ChEO10 micellar solutions. ChEO10 forms small, slightly elongated micelles in aqueous solution;21,22 at 10 wt %, the solutions display no viscoelasticity (Figure 1). On the basis of the observation that both HEBCD and SULFBCD enhance the viscoelasticity of the ChEO10/C12EO3 wormlike micelle solutions (Figure 3) and hypothesizing that this behavior could arise from a specific interaction of ChEO10 with the cyclodextrins, we next explore the effect of adding these two modified CDs to the pure ChEO10 solutions. Figure 4 shows the appearance of formulations containing 25 wt % HEBCD and increasing HEBCD/ChEO10 molar ratios (1:0.43−1:0.8). Samples with the highest amount of ChEO10 (1:0.8 molar ratio) can clearly withstand their own weight after 48 h when tipped upside down, thus showing a surprising gel-like behavior. Similar observations were made for mixtures of SULFBCD with ChEO10. On the basis of these results, phase diagrams of ChEO10 with either HEBCD or SULFBCD were constructed (2 weeks after preparation of the mixtures) and are shown in Figure 5. Gel-like samples, defined as samples not undergoing flow when turned upside down, form in a well-defined area of the phase diagram (corresponding to HEBCD concentrations between 20 and 30 wt % and ChEO10/HEBCD molar ratios between ca. 0.4 and 0.8, i.e., between just over 1 and 2 cyclodextrins per surfactant molecule; for SULFBCD, 20 wt % CD and between ca. 1.1 and 1.5 ChEO10/HEBCD ratios, i.e., 2−3 cyclodextrins for 3 surfactant molecules). To our knowledge, this is the first report of transparent gels formed in a mixture of a single surfactant with cyclodextrins. Frequency sweep rheology measurements and flow curves corroborate the visual observations (Figure 6). Twenty-five weight percent HEBCD solutions alone display a very weak

with a very weak rheological response (G′ ∼ G″ ∼ 0.01 Pa and Newtonian behavior, with a viscosity comparable to that of pure water), the addition of native β-CD induces the formation of a whitish, solid-like viscoelastic material, with a ∼103-fold increase in G′ and shear-thinning behavior. The observed onset of a solid-like behavior is in disagreement with the general observation of cyclodextrins breaking up micelles. Interestingly, the addition of β-CD to ChEO10 or C12EO3 alone leads to phase separation, with a clear phase above a white precipitate; this is typical of a complex formation between β-CD and a surfactant, which was confirmed by NMR (data not shown). However, in the presence of the cosurfactant C12EO3, the viscoelasticity increase suggests that neither of the two surfactants are removed from the micellar aggregates. The addition of native α- and γ-CD to the ChEO10/C12EO3 WLM induces the formation of white samples with a viscoelasticity similar to that of the original WLM, however displaying surprising stringy characteristics (Figure 2C,D). These characteristics could arise from the coexistence of CDs or surfactant:CD aggregates embedded in the WLM network. Strain sweeps show a reduction of the linear viscoelastic range (LVE) with the addition of α- or γ-CD, with an inverse dependence of the LVE range on CD concentration (Figure SI 2, SI), suggesting that WLM are more restricted and unable to accommodate the strain imposed, leading to an earlier rupture than in the original systems. Given the inhomogeneous appearance of the samples, the formation of large aggregates (bundles) of WLM may be at the origin of the observed behavior. The addition of hydroxypropyl-α-, -β-, and -γ-CDs has a limited impact on the viscoelasticity of the samples, leaving the overall rheology pattern unchanged, hence markedly differing from the methylated cyclodextrins (Figure SI 3, SI) and 11555

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

Figure 4. Pictures showing the time evolution of mixtures of HEBCD (25 wt %) with varying molar ratios of ChEO10 after tilting the vials upside down.

Figure 5. Phase diagrams showing the appearance of ChEO10/HEBCD (left) and ChEO10/SULFBCD (right) aqueous systems. Shaded symbols correspond to hazy or turbid samples.

Figure 6. Oscillatory frequency sweeps and flow curves of HEBCD 25 wt %/ChEO10 (A, B) and SULFBCD 20% (wt %)/ChEO10 (C, D) aqueous mixtures at different molar ratios.

rheological response (with G′ ∼ G″ ∼ 0.1 Pa), with a viscosity hardly above that of water; similarly, 10 wt % ChEO10 solutions display a Newtonian behavior with a viscosity just above that of water.26 When combining both solutions, a remarkable solidlike response is obtained, with a 106-fold increase in zero-shear viscosity and G′ and G″ being hardly dependent on frequency over the range measured. Gradually increasing ChEO10 content

induces a steady increase in both storage and loss moduli, up to 0.8 mol of surfactant per cyclodextrins (ca. 12.6 wt % ChEO10), beyond which phase separation occurs. Similarly, mixtures of 20 wt % SULFBCD with ChEO10 induce a solid-like response, which is reinforced by increasing amounts of ChEO10 (up to ca. 1.5 mol ChEO10/CD) (Figure 6). 11556

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

Figure 7. Zoomed in view of the 1H NMR spectra of the aliphatic part of (A) ChEO10 1%; (B) ChEO10 1% + HEBCD 2.5%; and (C) 1:1 mixture of ChEO10/DIMEB, 3 mM. Zoomed in view of the CD region of (D) DIMEB, 3 mM; (E) HEBCD 5%; (F) 1:1 mixture of ChEO10/DIMEB, 3 mM.

presence of CD. These results, interestingly, point to major differences between SULFBCD and HEBCD: while HEBCD forms a complex with pyrene, SULFBCD does not interact with it, suggesting a lack of affinity for hydrophobic moieties (most likely due to the charged SO42− groups around the rim, blocking access to the cavity). This hints at potentially very different mechanisms involved in the gel formation. Cyclodextrin−Surfactant Interactions Probed by NMR. NMR spectra were recorded in order to detect the presence or otherwise of interactions between ChEO10 and the two CDs. Clear changes between the spectrum of ChEO10 alone (Figure 7A) and in the presence of HEBCD are detected (Figure 7B). Figure 7A shows the proton spectrum of 1 wt % surfactant (well above the cmc). A broad, unresolved signal corresponding to the cholesterol moiety is observed (0.6−2.4 ppm), along with a peak due to the EO groups (around 3.6 ppm, not shown). The broadening of resonances from a surfactant is typical of the micellized state, as the protons are in a dynamic magnetic environment occurring on the time scale of NMR in which monomers are in equilibrium with micelles. In the presence of 2.5 wt % HEBCD, the broad set of signals from the cholesterol group becomes partly resolved (Figure 7B), indicating a shift of the equilibrium toward the free form of the surfactant (in agreement with the cmc measurements). On the other hand, the signal from the protons of HEBCD (between 3.3 and 4.2 ppm) is barely affected. For comparison purposes, the spectra of an equimolar mixture of ChEO10/ DIMEB (3 mM) is shown (Figure 7C,F). DIMEB has a defined substitution at the 2 and 6 positions, leading to a better resolved spectrum (Figure 7D). In addition to a better resolution of the cholesterol set of signals (Figure 7C) more marked than with HEBCDthe inner H3 and H5 protons of DIMEB shift upfield (Figure 7F) as do, to a lesser extent, the outer protons H1, H2, and H4. This evidence points unambiguously to the formation of an inclusion complex, ChEO10:DIMEB, responsible for the strong disaggregating capacity of DIMEB (which concurs with the loss of viscoelasticity, Figure 1C,D). Unlike DIMEB, the substitution at the primary and secondary rims of HEBCD is random, thus

This gel behavior was not observed with any of the other substituted cyclodextrins at similar compositions. With the native α- and γ-cyclodextrins, white gels are obtained, in appearance quite similar to mixtures of these cyclodextrins with poly(ethylene oxide) or Pluronics.29 These gels are quite inhomogeneous and brittle and likely result from the threading of the cyclodextrins onto the surfactant’s short poly(ethylene oxide) chain, leading to phase separation by hydrogen-bond formation between CDs threaded on adjacent chains.29 Having established the existence of this intriguing gel phase in mixtures of ChEO10 with either HEBCD or SULFBCD, the next two sections report fluorescence and NMR spectroscopy results, aiming to identify molecular interactions responsible for the drastic macroscopic changes. Cmc Determination by Fluorescence Spectroscopy. The cmc was determined by fluorescence spectroscopy using pyrene as a probe in ChEO10 solutions, both in the absence and presence of HEBCD or SULFBCD. In all cases, a clear break point was obtained in the fluorescence curves, which was identified as the cmc at (5.0 ± 0.1) × 10−5 g·mL−1. The cmc increases slightly in the presence of either HEBCD or SULFBCD, to (9.6 ± 0.6) × 10−5 or (9.1 ± 1.0) × 10−5 g· mL−1, respectively, suggesting a competition between micelle formation and complexation of the surfactant with the CDs. The effect however is quite weak and certainly weaker than with “destructive” modulators such as DIMEB.15 In order to evaluate the possible competition between HEBCD or SULFBCD and ChEO10 for pyrene (which could affect cmc measurement), the binding constants of both CDs with pyrene were measured by studying the dependence of the intrinsic pyrene fluorescence with increasing amounts of CD, according to the procedure described in ref 30 (Figure SI 4, SI). No change in the pyrene fluorescence was observed with SULFBCD, reflecting the absence of any significant interaction. For HEBCD, however, a 1:1 complex formation was detected with K ∼ 343 ± 19 M−1. Thus, in the case of HEBCD/pyrene solutions and under the conditions used for cmc determination, ca. 90% of pyrene is free to interact with the surfactant; therefore, the cmc can be estimated unambiguously even in the 11557

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

model. Given the substantial changes in viscoelasticity, these results are surprising. However, the length scales probed by neutrons do not cover the typical persistence length reached by wormlike micelles; in fact, the scattering from WLM can be assimilated to a succession of rods, which correlate with the worms persistence length, rather than their contour (total) length. Therefore, these data show that the micellar crosssection is conserved but do not exclude that micelles may be of different total lengths, since the plateau region of the form factor is beyond the q-range measured. In addition, these samples are dilutions of the gel samples, and total concentration is likely to affect structure, in particular micellar length. Overall, under the conditions of concentration and solvent used, SANS show that the addition of either HEBCD or SULFBCD conserves the micellar cross-section and rodlike geometry of the micelles. Microstructure over Larger Length Scales. Dynamic light scattering measurements were performed next to expand the characterization to larger length scales. Figure 9 shows the scattering from ChEO10 solutions (1 wt %) and in mixtures with SULFBCD and HEBCD at varying ratios, corresponding to dilutions of the gels identified in Figure 5. In the absence of CD, a polydisperse size distribution is obtained, with micelles of ca. 20 nm diameter, in good agreement with the total rod length obtained from SANS measurements. With the addition of both CDs, the distribution becomes broader and shifts to larger sizes (ca. 10-fold), more marked in the case of HEBCD than SULFBCD at the compositions studied. In conclusion, while light scattering reveals a substantial increase in aggregate size by the addition of either CD to ChEO10 solutions, SANS, on the other hand, show that the cross-section of the aggregates does not vary. These results overall demonstrate that the gels are induced by the one-dimensional growth of ChEO10 into elongated objects; at high concentrations, their entanglement can explain the remarkable viscoelastic properties observed. Cryo-TEM Images. Cryo-TEM measurements give a direct visualization of the samples microstructure. Figure 10 shows aqueous mixtures of ChEO 10 with either HEBCD or SULFBCD. The presence of very long wormlike micelles is clearly visible in both samples; we note, however, that the sample with SULFBCD was not very homogeneous, and different types of structures were found. This thus confirms that the gel-like characteristics are attributable to the lengthening and entanglement of wormlike micelles induced by the addition of cyclodextrins. In comparison, no elongated structures are detectable in pure ChEO10 solutions, as reported previously using SANS21,26,32 and confirmed here by cryo-TEM (Figure SI 5, SI). Effect of External Parameters on the Gels: Temperature and Addition of Electrolytes. In order to test the responsiveness of the samples to temperature, the rheology of a HEBCD/ChEO10 gel (25/10 wt %) was measured as a function of temperature over the temperature range 10−45 °C, above which the sample becomes turbid (Figure SI 6,SI). While a small decrease in G′ and G″ is observed above 40 °C, followed by a slight drop before phase separation occurs, the viscoelastic properties appear to be quite insensitive to temperature, and the sample maintains its gel-like properties over this temperature range. The effect of adding salts was assessed by adding either CaCl2 or KCl; the cations Ca2+ and K+ were selected as being at extremes in the Hoffmeister series.33 Either 1.0 M CaCl2 or 1.0 M KCl was added to SULFBCD/ChEO10 (20/10 wt %) or

producing a broadening of the signals, which complicates the interpretation of the spectrum and consequently any conclusion on the exact mode of binding (Figure 7E). Nonetheless, the fact that, as with DIMEB, the cholesterol region becomes better resolved with the addition of HEBCD clearly suggests a similar mode of interaction, which, as a result, increases the amount of ChEO10 in the form of a complex. HEBCD would thus induce some “destructive” effect, which is somehow contradictory with the formation of a gel. Whatever the nature of the complex, however (inclusion or “shallow”), it must be quite weak, as suggested by the low binding constant of HEBCD with pyrene. In addition, no clear cross-peaks between the cholesterol moiety and those of the CD were observed in a 2D-ROESY spectrum, by using 1% and 5% concentrations, respectively (data not shown). Thus, within the range of concentrations studied, the complex is in minor proportion, but high enough to induce major changes in the macroscopic behavior, thus necessarily affecting the self-assembled structures. Instead, the addition of SULFBCD does not change the NMR spectrum of ChEO10, suggesting an absence of interaction. The highly charged nature of this CD makes any type of binding of the hydrophobic moiety of the surfactanteither inclusion type or shallowdifficult to envisage (indeed pyrene does not interact with SULFBCD). Hence, the mechanism of gelation induced by SULFBCD is likely to be different from the one obtained with HEBCD. Nanoscale Morphology of the Aggregates. Small-angle neutron scattering (SANS) measurements were performed both on pure ChEO10 micelles and mixtures with HEBCD or SULFBCD in D2O, to characterize the morphology of the aggregates. An important point to note is that phase diagrams were markedly different in D2O, thus pointing to the importance of hydrogen bonding in the self-assembly process (since D2O is a more structured liquid than H2O31). A selection of samples that formed clear, gel-like samples in D2O was measured and diluted twice (to avoid interference peaks in the pattern), corresponding to 5 wt % ChEO10 with 2.5 wt % HEBCD or 5 or 10 wt % SULFBCD (Figure 8). All scattering curveseither in the presence or absence of cyclodextrins overlap over the whole q range measured. The scattering of ChEO10 micelles originates from elongated objects, which can be fitted by a cylinder model with an elliptical cross-section (R1 = 28 Å, R2 = 55 Å, and L ∼ 200 Å), as reported previously.26 The samples containing CDs can be fitted with the same

Figure 8. SANS measurements performed at 5% (wt %) ChEO10 in the presence and absence of HEBCD or SULFBCD. 11558

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

Figure 9. Dynamic light scattering measurements of mixtures of ChEO10 (1 wt %) aqueous solutions, alone and in mixtures with varying ratios of HEBCD (A) and SULFBCD (B), corresponding to the gel phase. Ratios are given as molar ratios of CD:surfactant.

does not disrupt the micellar networks but induces a remarkable reinforcement of the viscoelastic properties, when added to diluted, liquid-like WLM (Figure 2). This is all the more surprising since cholesterol and β-CD are known to form a strong complex, β-CD extracting cholesterol from membranes.36 For this reason, a strong interaction between β-CDs and the cholesteryl group of ChEO10, leading to the removal of ChEO10 from the aggregatesand thus a shortening or breakup of the micelleswas expected, instead of the observed reinforcement. In fact, when native β-CD is added to ChEO10 alone, precipitation is observed, suggesting the formation of an inclusion complex followed by aggregation, promoted by intermolecular hydrogen bonding, a phenomenon similar to the one observed for solutions of native α-CD and PEO.37 The same behavior was also observed in mixtures of β-CD and C 12EO3: precipitation of the inclusion complex of βCD:C12EO3; therefore, β-CD interacts with both surfactants constituting the wormlike micelles. The destructive effect of methylated CDs had been observed in our previous work on ChEO10/MCL wormlike micelles and ascribed to the complexation of one or both of the surfactants with the CDs, inducing a transition from WLM to shorter micelles by depletion of one of the components.15 In stark contrast to this classic “destructive” behavior, all the other βCD derivatives and native α-, β-, and γ-CDs preserve the viscoelasticity of the worms, and remarkably for native β-CD, HEBCD, and SULFBCD, reinforce it; this is in disagreement with a host−guest interaction leading to the extraction of surfactants from the micelles. The similarity of the response obtained with hydroxypropylated α-, β-, and γ-cyclodextrins points again to the importance of the substituents’ nature rather than to the cavity size (Figure SI 3, SI). None of these modified CDs show a strong effect on the rheological behavior, suggesting a weak interaction with either surfactant. On the other hand, hydroxyethyl-β-CD shows a constructive interaction with the WLM, reinforcing the fact that small changes on the substituent and its position can markedly impact the nature of the CD−surfactant (or CD− micelle) interaction. The synergistic effect is observed only with three of the CDs (HEBCD, SULFBCD, native β-CD) and it seems to correlate with two factors: the hydrogen-bond capacity of the CDs (substituted or native) and its affinity for hydrophobic substrates. For instance, methylated CDs, which disrupt WLMs, are also less able to form intermolecular hydrogen bonding14 and have a higher affinity for certain hydrophobic

Figure 10. Cryo-TEM images of the wormlike micelles observed in (A) 5 wt % ChEO10/12.5 wt % HEBCD mixtures and (B) 5 wt % ChEO10/10 wt % SULFBCD mixtures. Scale bars are 100 nm.

HEBCD/ChEO10 (25/10 wt %). In all cases, the addition of salts induces turbidity, with a clear phase separation in the case of SULFBCD/ChEO10, while a total liquid-like behavior is obtained with HEBCD/ChEO10 samples (Figure SI 7, SI). Clearly, salts dramatically affect the interaction between the cyclodextrins and the surfactant responsible for the gel-like behavior.

■

DISCUSSION In this work, we have reported the effect of a range of cyclodextrins, both native and modified, on two micellar systems: the first, a mixture of two surfactants, ChEO10/C12EO3 (10/2.5 wt %), known to form wormlike micelles and exhibiting solid-like viscoelastic properties, and, the second, a micellar solution of the single surfactant ChEO10, displaying purely liquid-like behavior. Our rheological results show that the nature of the cyclodextrins (α, β, or γ) and the substituent type in modified CDs is highly selective in controlling the macroscopic behavior of the mixtures and thus, clearly, the underlying self-assembled structures. The largely prevailing outcome of mixing cyclodextrins with surfactants is a CD:surfactant complex formation, competing with micellar formation,7,34,35 which, in the case of WLM, would likely result in a loss of the solid-like rheological response. Surprisingly, this effect was only observed with the methylated cyclodextrins RAMEB, DIMEB, and TRIMEB (Figure 1C,D). The effect scales with the degree of methylation, pointing at the importance of the substituents and perhaps the hydrogen bond capacity (or indeed lack thereof, in the case of TRIMEB). Indeed, native β-CD itself 11559

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

species.24 Our cmc data for HEBCD and SULFBCD show them to have a lower affinity for the surfactants studied than the methylated β-CDs.15 This suggests that to obtain a reinforcing effect, inclusion complex formation may not be the main mechanism of interaction; instead, intermolecular interactions via the external surface of the CD could be more relevant. Intermolecular interactions via hydrogen bonding leading to CD aggregation are a known phenomenon for native CDs.14 In fact, methylation is commonly used to improve CD solubility in water by breaking intramolecular hydrogen bonding of the native β-CD.2 Thus, it is expected that CD:surfactant inclusion complexes are capable of forming supramolecular aggregates via intermolecular hydrogen bonding of the CDs external surface, as suggested by Loftsson and co-workers,14 in particular with nonionic surfactants, where no electrostatic repulsion of charged headgroups is present. The surprising reinforcing effect of HEBCD had also been observed with ChEO10/MCL wormlike micelles,15 thus pointing to a specific interaction with the main surfactant, ChEO10. In contrast, HEBCD fully breaks up wormlike micelles of the cationic potassium oleate/KCl, in line with the “disruptive” role of cyclodextrins.15 This effect could indeed be ascribed to electrostatic repulsions between the anionic surfactant headgroups overcoming the coaggregation of the complexes within the micelles; in that case, the surfactant:CD complex formed are likely to solubilize in the bulk solvent rather than the micellar aggregates. Expanding from mixtures of CDs with WLM in mixed surfactants, we then showed that in single-surfactant micelles of ChEO10, the addition of HEBCD or SULFBCD induced a surprising viscoelastic response, enhancing G′ 100-fold and the viscosity 100 000-fold. Structural studies on both systems reveal the formation of long wormlike micelles, responsible for the gel-like appearance. Hence, both CDs are able to induce micellar growth via their interaction with the surfactant. This finding brings further support to our hypothesis of HEBCD coadsorbing at the interface and participating with the aggregates.15 Changes in micellar shape are necessarily driven by changes in surfactant packing within the aggregates. If we consider Israelachvili’s packing parameter p, micellar shape is tied to the relation p = v/al, where v and l are the volume and the length of the hydrophobic moiety and a is the optimal headgroup surface area.23 For p < 1/3, spherical micelles are expected, while cylindrical micelles form for 1/3 < p < 1/2. Thus, in order to promote a sphere-to-rod transition, the addition of CDs needs to increase v or decrease a. An increase of v could arise from the complexation of the hydrophobic cholesterol moiety followed by its incorporation in the micellar aggregate. However, CDs are generally insoluble in organic solvents2,38 and thus unlikely to localize in the micellar core. Nevertheless, given the unusual structure of ChEO10, with its flat, rigid steroid group, the hydrophobic core region is unlikely to be as welldefined as in alkyl-tail surfactants. Instead, some staggering of the molecules is expected,39 as well as unusual micellar shapesas seen indeed for bile salts, for instance40which could possibly accommodate a CD inclusion complex. On the other hand, one could also envisage a micellar shape transition induced by a change in headgroup area arising, for instance, from the coaggregation of the cyclodextrin between the cholesteryl group and the EO, driven by hydrogen bonding, bringing the PEO headgroups closer together. This second scenario is more plausible, since both HEBCD and SULFBCD were seen to interact weakly with pyrene and thus are expected

to interact weakly with hydrophobic moieties. Loftsson and coworkers14 indeed observed that for mixtures of uncharged modified CDs and hydrophobic molecules, phase-solubility behavior could not be explained solely on the basis of the formation of inclusion complexes but that the CD complexes were self-assembling into supramolecular aggregates, which in turn could further solubilize the hydrophobic molecules via noninclusion association. In this scenario, the surfactant:CD complex would act as a “headgroup” in the aggregates; this complex may be an inclusion complex but could also be a complex formed via the external surface of the CD. While the nature of the self-assembly processes involved in gel formation are still not clear, hydrogen bonds must be key; this is also reinforced by the very different phase behavior obtained when changing the solvent from H2O to D2O or adding salts to the systems.

■

CONCLUSION In this work, we explored the impact of native and substituted cyclodextrins (CD) in modulating the self-assembly, and as a result the macroscospic behavior, of a single nonionic surfactant, polyoxyethylene cholesteryl ether (ChEO10), and its mixtures with C12EO3. The outcome was found to depend critically on the nature of the substituents along the CD rim: as a result, the viscoelastic response varied substantially. In the surfactants mixture, studied at a composition known to produce highly viscoelastic wormlike micelles, the addition of methylated β-CDs induced micellar disruption and, thus, a loss of the viscoelasticity, while hardly any effect was observed with HPACD, HPBCD, and HPGCD. Instead, native β-CD, HEBCD, and SULFBCD enhanced the solid-like character of the micellar solutions. Therefore, it is possible to access a large range of self-assembly structures and, consequently, macroscopic behavior, solely by changing the substituents on the CD. More surprisingly, HEBCD and SULFBD were capable of inducing gel formation when added to the single surfactant system (ChEO10). A combination of small-angle neutron scattering, dynamic light scattering, and cryo-TEM revealed a transition from short ellipsoidal micelles to long, flexible micelles, whose entanglement explains the strong viscoelastic solid-like behavior observed. This is, to our knowledge, the first report on transparent gel-like materials in mixtures of a single surfactant with cyclodextrins. While the disruptive effect can easily be ascribed to the formation of an inclusion complex with the surfactant, leading to its depletion and subsequent breakup of the micelles, the reinforcement of the worms (in mixed systems) and gel formation (in single-surfactant system) are very puzzling. Fluorescence spectroscopy using pyrene as a probe and NMR results strongly suggest that this effect cannot be explained solely by inclusion complex formation, since surfactant−CD interactions appear to be very weak. The enhancement seems to correlate with a hydrophilic CD exterior and the capacity to form hydrogen bonds (as in HEBCD), while strong surfactant:CD complex formation and limited hydrogen-bond capacity (as in the methylated CDs) lead to the solubilization of the complexes in the bulk, rather than coaggregation within the micelles. Hence, the reinforcement likely results from a fine balance between surfactant−CD and complex−micelle interactions, with the coaggregation of the CDs or surfactant:CD complexes affecting the critical packing parameter in favor of micellar elongation. 11560

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

■

Article

(11) Jiang, L. X.; Yan, Y.; Huang, J. B. Versatility of cyclodextrins in self-assembly systems of amphiphiles. Adv. Colloid Interface Sci. 2011, 169, 13−25. (12) Yan, Y.; Jiang, L. X.; Huang, J. B. Unveil the potential function of CD in surfactant systems. Phys. Chem. Chem. Phys. 2011, 13, 9074− 9082. (13) Jiang, L. X.; Peng, Y.; Yan, Y.; Deng, M. L.; Wang, Y. L.; Huang, J. B. “Annular ring” microtubes formed by SDS@2 beta-CD complexes in aqueous solution. Soft Matter 2010, 6, 1731−1736. (14) Messner, M.; Kurkov, S. V.; Jansook, P.; Loftsson, T. Selfassembled cyclodextrin aggregates and nanoparticle. Int. J. Pharm. 2010, 387, 199−208. (15) da Silva, M. A.; Weinzaepfel, E.; Afifi, H.; Eriksson, J.; Grillo, I.; Valero, M.; Dreiss, C. A. Tuning the viscoelasticity of nonionic wormlike micelles with beta-cyclodextrin derivatives: A highly discriminative process. Langmuir 2013, 29, 7697−7708. (16) Xu, H.-N.; Ma, S.-F.; Chen, W. Unique role of beta-cyclodextrin in modifying aggregation of Triton X-114 in aqueous solutions. Soft Matter 2012, 8, 3856−3863. (17) Wang, D.; Long, P. F.; Dong, R. H.; Hao, J. C. Self-assembly in the mixtures of surfactant and dye molecule controlled via temperature and beta-cyclodextrin recognition. Langmuir 2012, 28, 14155−14163. (18) Kumar, R.; Kalur, G. C.; Ziserman, L.; Danino, D.; Raghavan, S. R. Wormlike micelles of a C22-tailed zwitterionic betaine surfactant: From viscoelastic solutions to elastic gels. Langmuir 2007, 23, 12849− 12856. (19) Chu, Z. L.; Feng, Y. J. Amidosulfobetaine surfactant gels with shear banding transitions. Soft Matter 2010, 6, 6065−6067. (20) Acharya, D. P.; Kunieda, H. Formation of viscoelastic wormlike micellar solutions in mixed nonionic surfactant systems. J. Phys. Chem. B 2003, 107, 10168−10175. (21) Afifi, H.; da Silva, M. A.; Nouvel, C.; Six, J.-L.; Ligoure, C.; Dreiss, C. A. Associative networks of cholesterol-modified dextran with short and long micelles. Soft Matter 2011, 7, 4888−4899. (22) Afifi, H.; Karlsson, G.; Heenan, R. K.; Dreiss, C. A. Structural transitions in cholesterol-based wormlike micelles induced by encapsulating alkyl ester oils with varying architecture. J. Colloid Interface Sci. 2012, 378, 125−134. (23) Israelachvili, J.; Mitchell, D. J.; Ninham, B. W. Theory of selfassembly of hydrocarbon amphiphiles into micelles and bilayers. J. Chem. Soc., Faraday Trans. 2 1976, 72, 1525−1568. (24) Schönbeck, C.; Westh, P.; Madsen, J. C.; Larsen, K. L.; Städe, L. W.; Holm, R. Methylated β-cyclodextrins: Influence of degree and pattern of substitution on the thermodynamics of complexation with tauro- and glyco-conjugated bile salts. Langmuir 2011, 27, 5832−5841. (25) SasView for small angle scattering analysis. http://www.sasview. org/ (accessed July 21, 2014). (26) Afifi, H.; Karlsson, G.; Heenan, R. K.; Dreiss, C. A. Solubilization of oils or addition of monoglycerides drives the formation of wormlike micelles with an elliptical cross-section in cholesterol-based surfactants: A study by rheology, SANS, and cryoTEM. Langmuir 2011, 27, 7480−7492. (27) Almgren, M.; Edwards, K.; Karlsson, G. Cryo transmission electron microscopy of liposomes and related structures. Colloids Surf., A 2000, 174, 3−21. (28) Raghavan, S. R.; Kaler, E. W. Highly viscoelastic wormlike micellar solutions formed by cationic surfactants with long unsaturated tails. Langmuir 2001, 17, 300−306. (29) Dreiss, C. A.; Cosgrove, T.; Newby, F. N.; Sabadini, E. Formation of a supramolecular gel between alpha-cyclodextrin and free and adsorbed PEO on the surface of colloidal silica: Effect of temperature, solvent, and particle size. Langmuir 2004, 20, 9124− 9129. (30) Rodríguez, P.; Sánchez, M.; Isasi, J. R.; González-Gaitano, G. Fluorescence quenching investigation of the complexes of dibenzofuran with natural cyclodextrins. Appl. Spectrosc. 2002, 56, 1490−1497. (31) Soper, A. K.; Benmore, C. J. Quantum differences between heavy and light water. Phys. Rev. Lett. 2008, 101, 065502.

ASSOCIATED CONTENT

S Supporting Information *

Structures of the surfactants used in this study and repeating unit of the β-CD derivatives; oscillatory frequency sweeps and flow curves for ChEO10/C12EO3 WLM in the absence and presence of HPACD, HPBCD, and HPGCD, and α-CD and γCD; fluorescence data used for binding constant calculation; cryo-TEM image of ChEO10 solutions; oscillatory temperature sweep of HEBCD:ChEO 10 solutions; and pictures of SULFBCD/ChEO10 and HEBCD/ChEO10 salt mixtures. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge financial support provided by JCNS at the Heinz Maier-Leibnitz Zentrum (MLZ), Garching, Germany, and Aurel Rădulescu for his help with the SANS experiments. Jayne Lawrence (King’s College London) is acknowledged for access to the light scattering instrument. Gloria Tardajos (Universidad Complutense de Madrid, Spain) is acknowledged for the NMR measurements on the Advance 700 MHz and data interpretation. The rheometer used in this study was bought on EPSRC grant EP/F037902/1.

■

REFERENCES

(1) Appel, E. A.; del Barrio, J.; Loh, X. J.; Scherman, O. A. Supramolecular polymeric hydrogels. Chem. Soc. Rev. 2012, 41, 6195− 6214. (2) Szejtli, J. Introduction and general overview of cyclodextrin chemistry. Chem. Rev. 1998, 98, 1743−1754. (3) Szetjli, J. Cyclodextrins as food ingredients. Trends Food Sci. Technol. 2004, 15, 137−142. (4) Loftsson, T.; Brewster, M. Pharmaceutical applications of cyclodextrins: Basic science and product development. J. Pharm. Pharmacology 2010, 62, 1607−1621. (5) Van de Manakker, F.; Vermonden, T.; Van Nostrum, C. F.; Hennink, W. E. Cyclodextrin-based polymeric materials: Synthesis, properties, and pharmaceutical/biomedical applications. Biomacromolecules 2009, 10, 3157−3175. (6) Xing, P. Y.; Chu, X. X.; Li, S. Y.; Xin, F. F.; Ma, M. F.; Hao, A. Y. Switchable and orthogonal self-assemblies of anisotropic fibers. New J. Chem. 2013, 37, 3949−3955. (7) Jiang, L. X.; Yan, Y.; Huang, J. B. Zwitterionic surfactant/ cyclodextrin hydrogel: Microtubes and multiple responses. Soft Matter 2011, 7, 10417−10423. (8) de Lisi, R.; Milioto, S.; Muratore, N. Thermodynamic evidence of cyclodextrin−micelle interactions. J. Phys. Chem. B 2002, 106, 8944− 8953. (9) Alami, E.; Abrahmsén-Alami, S.; Eastoe, J.; Grillo, I.; Heenan, R. K. Interactions between a nonionic gemini surfactant and cyclodextrins investigated by small-angle neutron scattering. J. Colloid Interface Sci. 2002, 255, 403−409. (10) Jiang, L. X.; Deng, M. L.; Wang, Y. L.; Liang, D. H.; Yan, Y.; Huang, J. B. Special effect of beta-cyclodextrin on the aggregation behavior of mixed cationic/anionic surfactant systems. J. Phys. Chem. B 2009, 113, 7498−7504. 11561

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562

Langmuir

Article

(32) Moitzi, C.; Freiberger, N.; Glatter, O. Viscoelastic wormlike micellar solutions made from nonionic surfactants: Structural investigations by SANS and DLS. J. Phys. Chem. B 2005, 109, 16161−16168. (33) Collins, K. D.; Washabaugh, M. W. The Hofmeister effect and the behavior of water at interfaces. Q. Rev. Biophys. 1985, 18, 323−422. (34) Guo, R.; Zhu, X. J.; Guo, X. The effect of beta-cyclodextrin on the properties of cetyltrimethylammonium bromide micelles. Colloid Polym. Sci. 2003, 281, 876−881. (35) Du, X.; Chen, X.; Lu, W.; Hou, J. Spectroscopic study on binding behaviors of different structural nonionic surfactants to cyclodextrins. J. Colloid Interface Scie. 2004, 274, 645−651. (36) Frijlink, H. W.; Eissens, A. C.; Hefting, N. R.; Poelstra, K.; Lerk, C. F.; Meijer, D. K. F. The effect of parenterally administered cyclodextrins on cholesterol levels in the rat. Pharm. Res. 1991, 8, 9− 16. (37) Harada, A.; Kamachi, M. Complex formation between poly(ethylene glycol) and α-cyclodextrin. Macromolecules 1990, 23, 2823−2824. (38) Silva, O. F.; Silber, J. J.; de Rossi, R. H.; Correa, N. M.; Fernandez, M. A. On the possibility that cyclodextrins’ chiral cavities can be available on AOT n-heptane reverse micelles. A UV−visible and induced circular dichroism study. J. Phys. Chem. B 2007, 111, 10703− 10712. (39) Castanho, M. A. R. B.; Brown, W.; Prieto, M. J. E. Rod-like cholesterol micelles in aqueous-solution studied using polarized and depolarized dynamic light-scattering. Biophys. J. 1992, 63, 1455−1461. (40) Madenci, D.; Egelhaaf, S. U. Self-assembly in aqueous bile salt solutions. Curr. Opin. Colloid Interface Sci. 2010, 15, 109−115.

11562

dx.doi.org/10.1021/la503000z | Langmuir 2014, 30, 11552−11562