Direct Visualization of a Self-Organized Multilayer Film of Low Tg

May 4, 2007 - ... University of Leeds, Leeds LS2 9JT, U.K., and Department of Chemistry, The University of Sheffield, Brook Hill, Sheffield S3 7HF, U...
0 downloads 0 Views 890KB Size
5536

2007, 111, 5536-5541 Published on Web 05/04/2007

Direct Visualization of a Self-Organized Multilayer Film of Low Tg Diblock Copolymer Micelles Emelyn G. Smith,† Grant B. Webber,‡ Kenichi Sakai,£,# Simon Biggs,£ Steven P. Armes,§ and Erica J. Wanless*,† School of EnVironmental and Life Sciences, UniVersity of Newcastle, Callaghan, NSW 2308, Australia, Particulate Fluids Processing Centre, Department of Chemical and Biomolecular Engineering, The UniVersity of Melbourne, ParkVille, Victoria 3010, Australia, School of Process, EnVironmental and Materials Engineering, UniVersity of Leeds, Leeds LS2 9JT, U.K., and Department of Chemistry, The UniVersity of Sheffield, Brook Hill, Sheffield S3 7HF, U.K. ReceiVed: March 21, 2007; In Final Form: April 20, 2007

The first in situ proof of the presence of true nanoscale micelle structure within alternating layers of a selfassembled cationic and anionic diblock copolymer micelle-micelle multilayer film is presented using atomic force microscopy. Apparently similar layer morphologies are distinguished by the interaction force curves normal to each layer. The three-dimensional order in these low Tg diblock copolymer films, together with their mechanical resilience and strongly hydrated nature, suggests many new technological applications.

Use of the layer-by-layer (LbL) technique in the development of novel functional surface coatings has attracted increasing interest in recent years.1,2 The LbL methodology was initially established for multilayers of polyelectrolytes on planar surfaces,1 although more recent studies have focused on coating colloidal substrates such as sols, latexes, and emulsion droplets. It has been demonstrated that one or more layers of linear polyelectrolyte can be replaced with either globular proteins,3 particles,4 or diblock copolymer micelles.5,6 This has greatly broadened the range of possible applications of LbL films; films incorporating one or more layers of colloidal material have been suggested as biosensors, microreactors, and as antireflective coatings.7 Early investigations into the incorporation of colloids into LbL films used linear polyelectrolytes to bind (oppositely charged) particles.4 Recently, Ma et al. proved the existence of diblock copolymer micelles of poly(acrylic acid-b-styrene) (PAA-b-PS: anionic PAA corona) as a second layer on top of an initial layer of linear poly(diallyldimethylammonium chloride) using in situ atomic force microscopy.5 Building upon this promising result, LbL films composed solely of diblock copolymer micelles have been recently reported.7,8 For example, Qi et al. constructed alternating multilayers composed of micelles of quaternized poly(4-vinylpyridine)-b-poly(tert-butyl acrylate), (PQ4VP-b-PtBA: cationic PQ4VP corona) and poly(acrylic acid)-b-poly(4-vinylpyridine), (PAA-b-P4VP: anionic PAA corona),8 while Cho et al. reported the formation of micellar multilayers based on P4VP-b-PS and PAA-b-PS (with cationic P4VP and anionic PAA coronas, respectively).7 In these * To whom correspondence should be addressed. E-mail: erica.wanless@ newcastle.edu.au. † University of Newcastle. ‡ The University of Melbourne. £ University of Leeds. § The University of Sheffield. # New address: Faculty of Science and Technology, Department of Pure and Applied Chemistry, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan

10.1021/jp072231z CCC: $37.00

cases, the retention of the relatively high Tg micellar cores during multilayer buildup was inferred from UV-vis spectroscopy, fluorescence spectroscopy, or by ex situ analysis with scanning electron microscopy (SEM) or tapping mode atomic force microscopy (TM-AFM). To the best of our knowledge, the present study represents the first direct in situ visualization of a LbL multilayer film that solely comprises diblock copolymer micelles at each stage of layer development. Moreover, both types of micelle cores are stimulus-responsive, rather than permanently hydrophobic as in many of the previous studies. The self-assembly of block copolymers into micelles and other aggregates has been the subject of much endeavor due to their versatility of function and ease of design; recent synthetic advances have significantly broadened the range of monomers which may be incorporated into block copolymers.9,10 A wide range of aggregate morphologies has been reported for block copolymers (spherical micelles, crew-cut micelles, vesicles, etc.), while aggregates of water-soluble block copolymers have been shown to be effective carriers for hydrophobic agents such as pigments and drugs, as well as quantum dots.8 The reversible self-assembly of diblock copolymers in response to stimuli such as temperature, ionic strength, or pH has been shown,11 and in principle, transferring such behavior to the solid-liquid interface offers the potential for a novel range of tailored uptake/release coatings and thus devices. It is well-known that the rate of desorption for polymers adsorbed at the solid-liquid interface, even in the absence of bulk polymer concentration, is significantly slower than that for small molecule surfactants.12 Furthermore, it has been demonstrated that diblock copolymer micelles adsorbed onto both mica and silica retain their core-shell structure when the bulk copolymer concentration is reduced below the solution critical micelle concentration.13 In principle, this may eliminate problems with premature release of entrained molecules. It has been suggested14 that the use of a multilayer of block copolymer micelles may enhance the stability of the adsorbed layer as well © 2007 American Chemical Society

Letters

Figure 1. Structure of (a) 10qPDMA93-b-PDEA24 diblock copolymer (quaternization of the tertiary amines is statistical but restricted to the much more reactive PDMA block) and (b) PMAA50-b-PDEA59. Subscripts denote the mean degrees of polymerization of the respective blocks.

as offer a potential route to optimizing the rate of release of the entrapment active, either via diffusion through the film or by limiting the loading to specific layers. We recently reported the first example of a LbL film of stimulus-responsive diblock copolymer micelles on a particulate substrate.14 Here, the entrainment of a hydrophobic dye within the micelle cores of each layer was monitored, and the data suggested the retention of micelle character for each new layer. However, at that time, it was not possible to directly visualize the developing LbL film to confirm the micellar morphology of each layer. This is the focus of the current work. Herein, we report on the direct in situ AFM visualization of a multilayer film of alternating layers of oppositely charged, stimulus-responsive diblock copolymer micelles self-assembled onto a planar silica substrate from aqueous solution at pH 9. At this pH, the cationic corona micelles used were 10% statistically quaternized poly(2-(dimethylamino)ethyl methacrylate)-b-poly(2-(diethylamino)ethyl methacrylate) (10qPDMAb-PDEA) and the anionic corona micelles used were poly(methacrylic acid)-b-poly(2-(diethylamino)ethyl methacrylate), (PMAA-b-PDEA).15,16 The structure of both diblock copolymers is shown in Figure 1. The 10qPDMA-b-PDEA diblock copolymer was synthesized using group-transfer polymerization (GTP), as described elsewhere.10 The molecular weight of this diblock was determined to be 19 100 g mol-1 by gel permeation chromatography (THF eluent, poly(methyl methacrylate) standards), and the PDMA content was found to be 79 mol % by 1H NMR spectroscopy. The mean degrees of polymerization for the PDMA and PDEA blocks were thus calculated to be 93 and 24, respectively. Following the procedure of Bu¨tu¨n et al.,17 the tertiary amine residues of the PDMA block were selectively quaternized using a substoichiometric amount of methyl iodide based on the DMA residues. The degree of quaternization was determined by NMR

J. Phys. Chem. B, Vol. 111, No. 20, 2007 5537 to be 10%, and this diblock is therefore referred to as 10qPDMA-b-PDEA. The PMAA-b-PDEA diblock copolymer was also synthesized by GTP, as described elsewhere.9 The mean degrees of polymerization for the PDEA and PMAA blocks were determined to be 59 and 50, respectively, by 1H NMR spectroscopy. Aqueous solutions (Milli-Q grade water, 200 ppm) of both copolymers were adjusted to pH 9 immediately prior to injection into the AFM fluid cell using small quantities of concentrated HNO3 and KOH solutions as appropriate; the 10qPDMA-b-PDEA required molecular dissolution in dilute HNO3 (pH ∼ 3) prior to adjustment to pH 9, while the PMAA-b-PDEA was directly soluble in alkaline solution. All solutions were 10 mM in background electrolyte (KNO3, used as received). Both types of micelles have been characterized in bulk aqueous solution at the deposition pH of 9. For the cationic corona micelles of 10qPDMA-b-PDEA, the hydrodynamic diameter was 13 nm, and the zeta potential was +18 mV,13 whereas the anionic corona micelles of PMAA-b-PDEA had a hydrodynamic diameter of 33 nm and a zeta potential of -40 mV by dynamic light scattering and microelectrophoresis, respectively. These two types of diblock copolymer micelles are highly hydrated (water content ∼80%)13 and, in view of their low Tg micelle cores, are expected to undergo rapid exchange with individual copolymer chains (unimers) in the surrounding solution.18-20 This adds complexity to the construction of micelle multilayers as it may result in interlayer mixing. Moreover, the final multilayers will be rather soft in character when compared to multilayers constructed with, for example, high Tg polystyrene cores. However, the unimer-micelle exchange kinetics are expected to be retarded for the adsorbed copolymer micelles (as found for cationic surfactant molecules in surface micelles on silica),21 which should facilitate the retention of micelle structures within each successive layer. While these factors make multilayer construction potentially more difficult, it also suggests their enhanced utility in applications such as drug delivery. To date, the focus of many controlled-release studies has been on relatively hydrophobic drugs that require encapsulation for efficient transport through the aqueous environment of the host.22 A logical release trigger in this case is to induce a change in capsule hydrophobicity in order to deliver the active molecule to the site of interest. However, controlled delivery and dosage can be just as important for more hydrophilic drugs. In this case, controlled release rates based on the slow diffusion of drug molecules through micelle multilayers may be the logical route. The lightly quaternized 10qPDMA-b-PDEA copolymer was selected as the first layer in view of the structural integrity and robustness of the adsorbed micelle layer formed by this diblock copolymer on silica (this is confirmed in Figure 2a).13 We have previously shown that, while the nonquaternized copolymer also forms a close-packed micellar film, it is not particularly robust to rinsing with electrolyte at pH 9.23 In contrast, the adsorbed amount (1.9 mg m-2, degree of hydration 83%) recorded for the 10qPDMA-b-PDEA copolymer remained unchanged after rinsing with electrolyte at pH 9, and a layer morphology of fairly close-packed micelles (streaming potential: +29 mV) was observed by AFM.13 A surface aggregation number, Nagg(surface), of 81 was derived from the in situ AFM images. This is significantly greater than the solution aggregation number for this copolymer.13 A further reason for selecting this copolymer was that its cationic coronal charge was expected to increase its electrostatic interaction with the second (anionic) micelle layer. Micelles of the PMAA-b-PDEA copolymer were chosen

5538 J. Phys. Chem. B, Vol. 111, No. 20, 2007

Letters

Figure 2. In situ AFM deflection images recorded for micelle multilayers adsorbed onto silica in 10 mM KNO3 at pH 9; 1 × 1 µm2 images of (a) 10qPDMA-b-PDEA Layer 1 (interaggregate spacing 32 ( 5 nm), (b) PMAA-b-PDEA Layer 2 (44 ( 5 nm), (c) 10qPDMA-b-PDEA Layer 3 (44 ( 5 nm), (d) PMAA-b-PDEA Layer 4 (55 ( 10 nm), (e) 10qPDMA-b-PDEA Layer 5 (55 ( 8 nm), and (f) 2 × 2 µm2 image of Layer 5. Note (f) is at lower magnification to show the roughness of the multilayer.

as the second layer due to their significant anionic charge density at pH 9. This choice also ensured that the nature of the hydrophobic micelle cores would be identical for each micellar layer; the PDEA block is pH responsive and has a pKa of around 7.0-7.5. A NanoScope III AFM (Veeco, CA) was used for in situ imaging to assess the morphology of the alternating electrostatically assembled diblock copolymer micelle-micelle multilayer.14 A cantilever with integral silicon nitride tip (Veeco, CA) was cleaned by UV irradiation (∼9 mW cm-2 at 254 nm for 30 min) prior to use. The nominal spring constant of the cantilever was 0.21 N m-1. Silicon wafers (Silicon Valley Microelectronics, CA, with a thermal oxide layer of 115 nm) were used as the substrate and subjected to a three-step treatment before use. This involved UV irradiation for 30 min followed by ultrasonication in ethanol for 20 min. The silica was then stored in ethanol, rinsed immediately prior to use with Milli-Q water, soaked in 10 wt % aqueous NaOH for 10 min, and rinsed in copious amounts of Milli-Q water to give a hydroxylated silica surface. Once assembled, the AFM cell was filled with Milli-Q water, adjusted to pH 9, and left for 30 min to allow the silica substrate to equilibrate. The copolymer solutions were then passed through a 0.2 µm syringe-mounted filter as they were injected into the cell. Images were recorded in deflection mode using a scanning speed of 7.63 Hz. Unless otherwise specified, all images have been recorded using the soft-contact imaging technique,24 in which the adsorbed layer is imaged using the minimum repulsive force required, and were recorded after supernatant replacement with 10 mM KNO3 at pH 9 following equilibration for at least 30 min in the relevant copolymer solution. A series of soft-contact deflection images are presented in Figure 2; images are presented for the first five layers of the

multilayer. For Layer 1, the 10qPDMA-b-PDEA copolymer shows a micelle structure as expected. Similar to the previously reported images of this copolymer adsorbed on silica, the micelles are reasonably close-packed with good lateral uniformity; the intermicellar spacing measured from a number of images was found to be 32 ( 5 nm.23 Interestingly, for the adsorption of the anionic corona PMAA-b-PDEA micelles (Layer 2, Figure 2b), the micelle structure observed in solution is also present in the adsorbed layer morphology, again with a fairly close-packed arrangement (intermicellar spacing of 44 ( 5 nm) with a clear increase in layer roughness when compared to the image of Layer 1. Unlike the case in which this copolymer is adsorbed directly onto silica, where it forms a laterally homogeneous brush-like morphology in order to minimize unfavorable electrostatic interactions between the PMAA block and the negatively charged substrate (AFM image not shown), here, the underlying cationic corona surface-adsorbed 10qPDMA-b-PDEA micelles ensure strong electrostatic interactions with the anionic coronal chains of the PMAA-b-PDEA micelles. The substantial intermicellar spacing suggests a combination of micelle adsorption and rearrangement during the formation of Layer 2. Other unknown factors are whether slow exchange occurs between chains within the multilayer and unimers in bulk solution or whether there is any intermixing between the micelles in adjacent layers (it is expected that there will be some degree of interpenetration between the layers, as observed for multilayers of linear polyelectrolytes,1 but we are unable to quantify this in the current study). Regardless, the morphology of Layer 2 is quite stable to both imaging and rinsing with electrolyte, although soft-contact imaging conditions were utilized to minimize any physical damage. Multilayer formation was continued further with three alternating layers of copolymer micelles sequentially added to

Letters

J. Phys. Chem. B, Vol. 111, No. 20, 2007 5539

Figure 3. (a) In situ AFM deflection 2 × 2 µm2 image of Layer 4, PMAA-b-PDEA. The central hole was created by scanning this region using a greater imaging force. (b) Cross section of the hole.

Figure 4. In situ AFM deflection images obtained for Layer 6, PMAAb-PDEA. Image (a) is 2 × 2 µm2, and (b) is 5 × 5 µm2.

the existing two layers, where odd-numbered layers consist of the cationic 10qPDMA-b-PDEA micelles and even-numbered layers consist of the anionic PMAA-b-PDEA micelles, as shown in Figure 2 up to Layer 5. Each layer clearly shows a micellar morphology that remains intact after rinsing with electrolyte, and comparing layers of the same copolymer (i.e., 1, 3, and 5 and 2 and 4), it is evident that there is an increase in the lateral size of the aggregates. This is attributed to enhanced flattening during adsorption and any subsequent rearrangement, as a consequence of the increasing underlying surface roughness. The duration of the experiment was approximately 8 h, and the solution pH was kept constant over this period. Subsequent layers have slightly higher surface roughness; this is more evident on a larger scale, as shown in Figure 2f. Further evidence of the nature of each micelle layer is given in Figure 3a, which is an AFM image of Layer 4 (PMAA-bPDEA micelles). This image was captured after scanning for several minutes at a higher force, over an area of approximately 1 × 1 µm2. From a cross-sectional analysis of the corresponding height image (Figure 3b), the hole depth is 12 ( 1 nm. The image clearly indicates the underlying micelle layer within the scan hole. This depth is significantly less than the bulk solution diameter of the PMAA-b-PDEA micelles, implying that a substantial flattening of the micelles occurs during their adsorption, similar to that reported previously on flat, rigid substrates.13 It is intriguing that the micelle layer within the hole is evidently resistant to the greater imaging force that was employed to create it. From this image, we infer that interlayer mixing is not sufficient to completely disturb the micelle structures within a layer (at least within the time scale of these observations). The well-behaved LbL micelle adsorption behavior was eventually disrupted. With the addition of Layer 6 (PMAA-b-

PDEA, Figure 4), large, ill-defined features were observed, which are attributed to complexation between the two different polyelectrolytes present. In addition, these images of Layer 6 are much less stable than those of the first five layers; a significant amount of the adsorbed material became mobile during scanning, particularly within the larger scan ranges (Figure 4b). The large features that are observed are presumably ill-defined electrostatic copolymer complexes of reduced net charge that are only weakly bound to the underlying film. We have previously shown that adsorption of PDMA-b-PDEA micelles to a low-charge-density substrate of relatively high surface roughness (silica) resulted in a poorly defined adsorbed layer that was more susceptible to desorption during rinsing compared to adsorption onto a flat, high-charge-density surface (mica).13 Additionally, when preparing micelle multilayers of closely related diblock copolymer micelles adsorbed on silica sols, the cationic surface charge after each addition of PDMA-b-PDEA micelles was reduced as the overall layer number increased.14,25 Conversely, the magnitude of the anionic surface charge remained essentially constant. It is possible that Layer 5 (cationic 10qPDMA-b-PDEA) is relatively diffuse, resulting in sufficiently low charge density, and that the adsorbing anionic PMAA-b-PDEA micelles can strongly penetrate Layer 5, hence disrupting the micelles and thereby producing the observed ill-defined complexes. The section analysis of the complementary height image to Figure 4b (and related images) indicates mean heights of about 11 nm for these features, which is comparable to the thickness of Layer 4 determined in Figure 3. This suggests that the degree of polyelectrolyte interpenetration between each of the first five micelle layers is not substantial.

5540 J. Phys. Chem. B, Vol. 111, No. 20, 2007

Figure 5. Force-distance data for each micelle layer monitored in the rinse solution (10 mM KNO3 at pH 9) during the construction of micelle-micelle multilayers. Data were recorded during (a) the approach and (b) the retraction of the tip to/from the surface. The inset of (a) shows the data replotted on a log-normal scale. Data were recorded after the formation of (blue s) Layer 1, (blue 4) Layer 2, (green s) Layer 3, (green 0) Layer 4, (red s) Layer 5, and (red O) Layer 6.

Interaction force curves were also recorded during the construction of these micelle multilayers, and these data are reported in Figure 5. The approach curves normal to each layer are rather similar (see Figure 5a) and indicate that electrosteric repulsion occurs within an average apparent separation of 17 nm. It is noteworthy that, as with all AFM experiments where soft material is bound to one or both substrates, precise determination of the zero separation distance is not possible and tip-sample separation comparisons are, at best, qualitative. The observed similarities were unexpected given the alternating charge on the micelle multilayers.14,25 Given the multilayer nature of the film, one might expect instabilities in the forcedistance data as the cantilever tip pushes through successive layers of adsorbed micelles. The absence of such instabilities suggests that the tip no longer acts as a sharp point, presumably due to copolymer adsorption onto the cantilever. Replotting of the approach data on a log-normal scale, shown in the inset of Figure 5a, yields an approximate decay length of 6 nm, which

Letters is substantially greater than that expected for a purely electrostatic interaction (approximately 3 nm for a 1:1 electrolyte at 10 mM). We conclude that the steric nature of the interaction dominates and is a consequence of copolymer adsorption on both the silica substrate and the cantilever tip, yielding a repulsive interaction; adsorption to these two surfaces during the construction of each micelle layer results in both surfaces being either cationic (odd layers) or anionic (even layers), and consequently facilitates the soft-contact imaging technique. In contrast, the retraction curves recorded for each consecutive micelle layer clearly demonstrate that the two different types of micellar layers may be readily distinguished; see Figure 5b. Here, the cationic layers of 10qPDMA-b-PDEA display weak adhesion, while the anionic layers exhibit strong adhesion. This difference was also manifest while imaging each successive layer, where it was found that the anionic layers were more easily imaged than the cationic layers. These observed differences in the adhesive interaction may be attributed to a combination of two effects. First, the cationic 10qPDMA-b-PDEA and anionic PMAA-b-PDEA micelles have differing coronal charge densities. When 10qPDMA-b-PDEA micelles comprise the outer layer, the interaction during measurement of normal forces is between two layers of relatively low cationic charge; thus, there is only weak adhesion between these two layers upon retraction. Conversely, the zwitterionic nature of the PMAA-b-PDEA copolymer means that while these micelles have a high overall anionic charge (due to the coronal PMAA blocks), some residual cationic charge remains within the low Tg PDEA cores (the pKa of PDEA is ∼ 7.3).10 Thus, it is feasible that adsorption of this zwitterionic copolymer to both the multilayer and the tip leads to attractive electrostatic interactions and, hence, results in a significant adhesive force upon retraction. Such interactions would not be evident in the approach data due to long-range steric shielding but would become manifest during the intimate contact required to enable interlayer polymer penetration. The relatively high proportion of the PDEA block within the PMAA-b-PDEA copolymer should enhance such behavior. It is also possible, however, to attribute the observed differences in adhesion to variations in the adsorbed layer morphologies for each respective layer. From previous studies,13 and indeed from the images shown in Figure 2, the 10qPDMAb-PDEA micelles clearly adsorb in a somewhat flattened conformation both on bare silica and also onto a PMAA-bPDEA micelle layer. This leads to minimal extension of polymer chains as “loops and tails” normal to the interface; thus, little or no entanglement between interacting chains on the surface and tip would be expected. On the other hand, it is likely that the more highly charged zwitterionic PMAA-b-PDEA micelles adsorb in a more extended conformation; thus, the propensity for entanglement between chains on the surface and those adsorbed on the AFM tip is greater for these layers. This will allow greater interpenetration of other copolymer chains into the layer, enhancing the effect of the zwitterionic PMAA-bPDEA copolymer on the adhesion and resulting in the formation of the ill-defined electrostatic complexes that are observed during the construction of Layer 6 (Figure 4). It is not known if the observed collapse of the ordered selfassembly at Layer 6 is due solely to these electrostatic forces (and also the hydrophobic interactions between the PDEA core blocks in adjacent layers) or whether the compressive forces due to increasing adsorbed copolymer mass and the relatively high fraction of entrained water also play a significant role. Our observations suggest that low Tg micelle-micelle multilayers

Letters of more than five layers are more technically challenging than those constructed from micelles comprising high Tg cores. While the role of the AFM tip should not be ignored, its contribution to the layer collapse is expected to be minimal given that the multilayer was compromised within the first few seconds of scanning Layer 6. In summary, we report the first in situ AFM images of up to six layers of a LbL film comprising solely diblock copolymer micelles constructed via electrostatic self-assembly in aqueous solution. The presence of micelles is unambiguously confirmed within each layer for the first time. Moreover, interaction force curves recorded normal to each layer indicate substantial adhesive forces between adjacent layers. The electrostatic nature of this interaction provides direct evidence for the mode of selfassembly. Unlike in previous reports of micelle-micelle multilayers, the diblock copolymers used in this study have subambient Tg values; thus, the micelles are soft and may well undergo some exchange with unimers in bulk solution during the construction of the multilayer film. The high water content and stimulus-responsive nature of such multilayer films augur well for potential controlled uptake/release applications. We plan to quantify these behaviors in a future publication using alternative experimental techniques. Acknowledgment. The Australian Research Council is thanked for Research Grant DP0343783. The EPSRC is thanked for the linked Research Grants GR/S60419 and GR/S60402 awarded to S.P.A. and S.B., respectively. Dr. M. Vamvakaki is thanked for the synthesis of the zwitterionic PMAA-b-PDEA diblock copolymer and Dr. V. Bu¨tu¨n for the synthesis of the PDMA-b-PDEA diblock copolymer. We also thank Dr. S. Liu for assistance with the quaternization reaction. S.P.A. is a recipient of a five year Royal Society-Wolfson Research Merit Award. References and Notes (1) Decher, G. Science 1997, 277, 1232-1237. (2) Donath, E.; Sukhorukov, G. B.; Caruso, F.; Davis, S. A.; Mo¨hwald, H. Angew. Chem., Int. Ed. 1998, 37, 2202-2205.

J. Phys. Chem. B, Vol. 111, No. 20, 2007 5541 (3) Caruso, F.; Mo¨hwald, H. J. Am. Chem. Soc. 1999, 121, 60396046. (4) Caruso, F.; Caruso, R. A.; Mo¨hwald, H. Science 1998, 282, 11111114. (5) Ma, N.; Zhang, H.; Song, B.; Wang, Z.; Zhang, X. Chem. Mater. 2005, 17, 5065-5069. (6) Ma, N.; Wang, Y.; Wang, Z.; Zhang, X. Langmuir 2006, 22, 39063909. (7) Cho, J.; Hong, J.; Char, K.; Caruso, F. J. Am. Chem. Soc. 2006, 128, 9935-9942. (8) Qi, B.; Tong, X.; Zhao, Y. Macromolecules 2006, 39, 5714-5719. (9) Lowe, A. B.; Billingham, N. C.; Armes, S. P. Macromolecules 1998, 31, 5991-5998. (10) Bu¨tu¨n, V.; Armes, S. P.; Billingham, N. C. Polymer 2001, 42, 5993-6008. (11) Forster, S.; Abetz, V.; Muller, A. H. E. Polyelectrolytes with Defined Molecular Architecture II; Wiley: New York, 2004; Vol. 166, p 173. (12) Fleer, G. J.; Cohen Stuart, M. A.; Scheutjens, J. M. H. M.; Cosgrove, T.; Vincent, B. Polymers at Interfaces, 1st ed.; Chapman and Hall: London, 1993. (13) Sakai, K.; Smith, E. G.; Webber, G. B.; Schatz, C.; Wanless, E. J.; Bu¨tu¨n, V.; Armes, S. P.; Biggs, S. Langmuir 2006, 22, 5328-5333. (14) Biggs, S.; Sakai, K.; Addison, T.; Schmid, A.; Armes, S. P.; Vamvakaki, M.; Bu¨tu¨n, V.; Webber, G. AdV. Mater. 2007, 19, 247-250. (15) Dai, S.; Ravi, P.; Tam, K. C.; Mao, B. W.; Gan, L. H. Langmuir 2003, 19, 5175-5177. (16) Mao, B. W.; Gan, L. H.; Gan, Y. Y.; Tam, K. C.; Tan, O. K. Polymer 2005, 46, 10045-10055. (17) Bu¨tu¨n, V.; Armes, S. P.; Billingham, N. C. Macromolecules 2001, 34, 1148-1159. (18) Tian, M.; Qin, A.; Ramireddy, C.; Tuzar, Z.; Munk, P. Polym. Mater. Sci. Eng. 1993, 69, 172-173. (19) Creutz, S.; van Stam, J.; Antoun, S.; De Schryver, F. C.; Je´roˆme, R. Macromolecules 1997, 30, 4078-4083. (20) Sakai, K.; Smith, E. G.; Webber, G. B.; Schatz, C.; Wanless, E. J.; Bu¨tu¨n, V.; Armes, S. P.; Biggs, S. J. Phys. Chem. B 2006, 110, 1474414753. (21) Clark, S. C.; Ducker, W. A. J. Phys. Chem. B 2003, 107, 90119021. (22) Malmsten, M. Soft Matter 2006, 2, 760-769. (23) Sakai, K.; Smith, E. G.; Webber, G. B.; Baker, M.; Wanless, E. J.; Bu¨tu¨n, V.; Armes, S. P.; Biggs, S. Langmuir 2006, 22, 8435-8442. (24) Manne, S.; Cleveland, J. P.; Gaub, H. E.; Stucky, G. D.; Hansma, P. K. Langmuir 1994, 10, 4409-4413. (25) Webber, G. B.; Sakai, K.; Wanless, E. J.; Armes, S. P.; Vamvakaki, M.; Bu¨tu¨n, V.; Biggs, S. World Congress of Chemical Engineering, 7th; Glasgow, United Kingdom, July 10-14, 2005; 2005; pp 86505/1-86505/ 10.