Pushing the Size Limits in the Replication of Nanopores in Anodized

Article pubs.acs.org/Langmuir

Pushing the Size Limits in the Replication of Nanopores in Anodized Aluminum Oxide via the Layer-by-Layer Deposition of Polyelectrolytes Mohammad Raoufi, Davide Tranchida,† and Holger Schönherr* Department of Chemistry and Biology, Science & Technology, Physical Chemistry I, University of Siegen, Adolf-Reichwein-Str. 2, 57076 Siegen, Germany S Supporting Information *

ABSTRACT: We report on the successful replication of the smallest pores in anodized aluminum oxide (AAO) via the layer-by-layer (LBL) deposition of polyelectrolytes to date to yield free-standing, open nanotubes with inner and outer diameters (±2σ) down to 37 ± 4 and 52 ± 19 nm, respectively. This work is based on the fabrication of defined arrays of highly regular nanopores by anodic oxidation of aluminum. Pores with pore diameters between 53 ± 9 and 356 ± 14 nm and interpore distances between 110 ± 3 and 500 ± 17 nm were obtained using an optimized twostep anodization procedure. 3-(Ethoxydimethylsilyl)propylamine-coated pores were replicated by alternating LBL deposition of poly(styrenesulfonate) and poly(allylamine). The detrimental adsorption of polyelectrolyte on the top surface of the template that typically results in partial pore blocking was eliminated by controlling the surface energy of the top surface via deposition of an ultrathin gold layer. The thickness of the deposited LBL multilayer assembly at the pore orifice agreed to within the experimental error with the thicknesses measured by variable angle spectroscopic ellipsometry and atomic force microscopy (AFM) for layers assembled on flat substrates. The selective dissolution of the alumina template afforded free-standing, open polymer nanotubes that were stable without any cross-linking procedure. The nanotubes thus obtained possessed mean outer diameters as small as 52 nm, limited by the size of the AAO template.

■

INTRODUCTION For the fabrication of defined nanoscale polymer structures and objects two alternative routes have been successfully pursued in recent years.1 Both self-assembly (or self-organization) approaches2 and strategies utilizing the replication of sacrificial templates3 were shown to result in nanoscale structures with unprecedented control of dimensions, chemical composition, and functionality.4 In particular, template replication is a viable pathway to obtain highly defined tunable micro- and nanostructures. Instructive examples include the filling of micro- and nanometer scale pores with polymers5 or polymerizable precursors, as exploited in micromolding in capillaries (MIMIC)6 and UV nanoimprint lithography,7 as well as the application of colloidal lithography for functional structures that can be applied in plasmonics,8 among others. As reviewed recently by Caruso and co-workers,9 layer-bylayer (LBL) assembly of polyelectrolytes10 around sacrificial microparticles followed by dissolution of the template particle led to the development of responsive capsules for encapsulation of active compounds or for application in cosmetics, etc.11 Among the stimuli reported are optical,12 biochemical,13 and electrochemical stimuli.14 The same methodology has been investigated as a means to fabricate polymeric nanotubes by replicating porous membrane structures, e.g., in track etched membranes15 or in anodized aluminum oxide (AAO) templates (Figure 1),16 which can be obtained electrochemically.17,18 © 2012 American Chemical Society

The application of AAO substrates in label-free sensing was previously reported.19 These reports bear an immediate consequence for template replication via multilayer formation inside the pores, as discussed here. For example, Janshoff and co-workers showed that protein may diffuse without hindrance on time scales of thousands of seconds into sub-100 nm wide pores in AAO up to depths of several micrometers.20 In these experiments the adsorption of molecules was limited to a (sub)monolayer. By contrast, for LBL functionalization of porous templates, unlike for the replication of particles, the deposited polyelectrolytes may result in an increasingly difficult mass transport (diffusion), since the pore diameter decreases with increasing number of layers. In another very recent paper, Roy et al. have analyzed this process using porometry measurements and identified two regimes of deposition.21 In early reports it was speculated that the pore orifice may be easily blocked by adsorbed polymer. To prevent clogging of submicrometer wide pores, a pressure-filter-template technique (pore diameter 330 nm)16 and much more recently optimized solution conditions (∼100 nm pore size)22 have been proposed. The smallest pores that have been successfully replicated to date by the LBL approach measured 70 nm in Received: April 26, 2012 Revised: May 30, 2012 Published: June 20, 2012 10091

dx.doi.org/10.1021/la3017062 | Langmuir 2012, 28, 10091−10096

Langmuir

Article

Figure 1. Schematic of the replication of nanoporous template by LBL deposition of polyelectrolyte multilayer and subsequent dissolution of the substrate to yield isolated nanotubes.

Table 1. Etching Conditions AAOa regime

DI (nm)

DP (nm)

first step

voltage (V)

1 2

110 345

35−70 150−250

MA HA

40 170

3

500

185−400

MA

195

a

electrolyte (concentration) oxalic acid (0.3 M) oxalic acid (0.3 M) + phosphoric acid (0.1 M) phosphoric acid (0.1 M)

second step

voltage (V)

MA MA

40 136

MA

195

electrolyte (concentration) oxalic acid (0.3 M) oxalic acid (0.05 M) + phosphoric acid (0.1 M) phosphoric acid (0.1 M)

DI: interpore distance; DP: pore diameter.

diameter.22 This was rendered possible according to the authors by using appropriate ionic strength and particularly divalent cations to decrease the size of the dissolved polyelectrolyte chains, which is in line with the data reported by Rubner.23 However, even though the entropic forces22 may become important in some situations, a comparison of pore diameter, polymer coil size (rms end-to-end distance on the order of 12 nm for the polymers utilized in ref 15 and here), and a typically observed (dry) thickness increment, which is on the order of 2.5 nm/bilayer, shows that this is not a fully satisfactory explanation for pores with diameters ≫50 nm, even if swelling of the layer is taken into account.24 On the other hand, pore clogging is clearly observed for pores with a mean internal diameter of 53 nm for as few as two bilayers (corresponding to a multilayer thickness of ∼5 nm; see Figure 3). Hence there is a clear need to improve the process of depositing multilayers via the LBL approach in order to be able to fabricate much smaller nanotubes. In this paper we report on the identification of polymer adsorption at or very close to the pore opening as the prime contributor for pore clogging and the implementation of a new strategy to suppress the adsorption of polyelectrolyte on the top surface of the template and the pore orifice. Thereby, clogging of the pore is effectively circumvented. By exploiting selective surface passivation, pores with outer diameters of 55 nm, limited by the size of the AAO template, have been successfully replicated, thus paving the way to stable, open polymeric nanotubes after using appropriate template removal. Such truly nanoscale polymeric tubes may find application in the areas of tissue engineering, mechanical sensing, or advanced biomolecule separation, where the reduced radius of curvature and the high surface area to volume ratio may be exploited.

■

acid (85%, Chemische Fabrik Budenheim), oxalic acid (C.N. 6153-566, Merck), perchloric acid (60−62%, J.T. Baker), sulfuric acid (95%, Chemsolute), Milli-Q water (from a Millipore Direct-Q 8 system with resistivity of 18.0 MΩ/cm, Millipore, Schwalbach, Germany), KOH (C.N. 1310-58-3, Roth), NaOH (Sigma-Aldrich), NaCl (C.N. 764714-5, Baker), chromium oxide (C.N. 27081-50G-F, Sigma-Aldrich), dialysis membrane (7Spectra/Por dialysis membrane, Spectrum Laboratories; volume/length: 2.5 mL/cm; diameter: 18 mm; molecular weight cutoff: 50 kDa), HCl (37%, VWR), H2O2 (30%, Roth), and silicon (100) wafers (P/Boron type, manufactured by OKMETIC, Finland) were purchased from the suppliers listed. AAO Fabrication. For the anodization a home-built setup was used, comprising a cooling system and power supply (EA-PSI 8160-04 T, Electro-Automatic GmbH, Viersen, Germany), in which the temperature and the applied voltage are controlled (Figure 2). A multichannel electrochemical cell was designed to keep both sides of the aluminum sample at a preset constant temperature. A lowtemperature cooling system (LAUDA PK20, −40 to +150 °C) was used to control the temperature in an outer thermal reservoir. A high purity Al plate, mounted on a copper plate, served as the anode. The polished side of the Al sample was pointing toward the electrolyte, and the other side was in contact with the cold liquid of the thermal reservoir. A rubber O-ring (Wilhelm Jung GmbH) was used on the polished side of the sample to seal the compartments. The cathode made of an Al sheet (2.5 cm × 9.0 cm, Bikar, Bad Berleburg, Germany) was positioned at a distance of 20 mm from the anode. A rotary motor ensured convection of the electrolyte during the anodization. AAO templates with different interpore distances were fabricated by the two-step anodization method.17 The anodization was carried out with two different methods, namely mild and hard anodization.17,18 Before starting the anodization, the high-purity aluminum plate was degreased in acetone and washed with Milli-Q water. Subsequently, the aluminum was electropolished at room temperature in a 1:3 (vol:vol) mixture of perchloric acid and ethanol at a constant current density of 100 mA cm−2 for 3 min to reduce the surface roughness. The mild anodization (MA) was carried out with constant voltage (details see Table 1). When MA was used for both the first and the second step, the corresponding anodization times were 15 and 2 h, respectively. The first step of the anodization in regime number 3 (MA, 195 V, Table 1) was started with a low potential (100 V), which was increased to the desired potential (195 V) for 30 min, after which it was kept constant for the rest of the process. HA was not carried out with constant voltage. Instead, the

EXPERIMENTAL SECTION

Materials. Aluminum (99.9999%, 0.5 mm thick plate, Chempure), poly(allylamine hydrochloride) (C.N. 71550-12-4, Mw = 120 000− 200 000 g/mol, Alfa Aesar), poly(styrenesulfonate) (C.N. 25704-18-1, Mw = 70 000 g/mol, Sigma-Aldrich), 3-(ethoxydimethylsilyl)propylamine (97%, C.N. 18306-79-1, Sigma-Aldrich), phosphoric 10092

dx.doi.org/10.1021/la3017062 | Langmuir 2012, 28, 10091−10096

Langmuir

■

voltage was changed during the anodization to create a protective layer that prevents the sample from electric breakdown (burning) at high voltage. The anodization was started at a constant voltage of 40 V for 15 min; subsequently, the voltage was increased to the desired voltage with a particular rate and was kept constant until the end. In HA the motor pumped the electrolyte toward the sample during the anodization. The temperature for low voltage (40 V) and high voltages (≥105 V) was kept constant at 15 and 0 °C, respectively. The porous oxide layer that was obtained in the first step was chemically removed by a treatment in an aqueous mixture of chromic acid (1.8 wt %) and phosphoric acid (6 wt %) for 5 h at 48 °C. The remaining Al surface exhibited surface corrugations that stem from the bottom parts of the originally formed pores. The corresponding anodization voltages and acid concentrations are summarized in Table 1. The pores were widened, as required, by chemical etching in 0.5 M aqueous phosphoric acid at 30 °C. Preparation of Multilayered Polymeric Nanotubes and Thin Films. 2 cm × 1 cm pieces of a silicon wafer were cleaned by treatment with piranha solution (70/30 mixture by volume of concentrated sulfuric acid and hydrogen peroxide for 2 min, followed by rinsing with copious amounts of Milli-Q water. (Caution: piranha solution should be handled with extreme caution! It has been reported to detonate unexpectedly.) The AAO was sonicated in Milli-Q water three times for 10 min. The AAO templates (or silicon samples) and one drop of 3-ethoxydimethylsilylpropylamine were kept in desiccators at a reduced pressure of 16 mbar in room temperature for 24 h to obtain a monolayer of the silane exposing amino groups.25 The AAO template was then coated under an angle of 45° with a thin Au film by sputtering ∼7 nm gold (Cressington 108auto, Dülmen, Germany) to afford a passivation primer layer that prevents the polyelectrolyte from adsorbing at the surface of template. For LBL deposition, the cleaned substrates were immersed consecutively into solutions of poly(styrenesulfonate) (PSS, concentration 0.5 g/L in 0.15 M NaCl in Milli-Q water), Milli-Q water, poly(allyl amine) (PAH, concentration 0.5 g/L in 0.15 M NaCl in Milli-Q water), and again Milli-Q water. Dipping times in polyelectrolyte solutions were 30 min; the washing step in Milli-Q water lasted for 10 min. The adsorption and rinsing steps were repeated as required. After the deposition of the desired number of polyelectrolyte layers, the samples were dried in a stream of nitrogen. The polymeric nanotubes were released from the AAO template by dissolving the aluminum oxide in KOH (5 wt %, room temperature). To isolate the nanotubes from the KOH solution, the pH of the solution was adjusted to 7 by adding HCl. Then dialysis was performed for 3 days against Milli-Q water; the water was exchanged each 3 h on the first day and each 12 h on the second and third day. Ellipsometry. Film thicknesses of the PSS/PAH multilayer films on silicon were determined using an α-SE Ellipsometer (J.A. Woollam Co., Lincoln, NE). Using the EEASE software, the layered systems was modeled as a three-layer system, comprising silicon as substrate, silicon oxide (2 nm), and the polyelectrolyte layer (refractive index 1.570 at 600 nm). The measurements were done at three different angles of incidence and reflection (65°, 70°, and 75°). Finally, the data were cross-checked by AFM thickness measurements. Scanning Electron Microscopy (SEM) Measurements. The SEM data were acquired on a Zeiss Ultra 55cv field emission scanning electron microscope (FESEM) with 30 kV maximum operating voltage (Zeiss, Oberkochen, Germany). All measurements were performed with an operation voltage of 10 kV with the Inlens secondary electron detector. For the analysis of the FESEM micrographs, SPIP software (scanning probe image processor, Version 5.0.7) was used. Atomic Force Microscopy (AFM) Measurements. A MultiMode IIIa AFM (Bruker/Veeco/Digital Instruments, Santa Barbara, CA) and an Asylum Research MFP-3D Bio (Asylum Research, Santa Barbara, CA) were used for AFM measurements in tapping-mode, employing silicon cantilever-tip assemblies (Veeco type MLCT, nominal resonance frequency of 15 Hz and spring constant of 0.03 N/m).

Article

RESULTS AND DISCUSSION The AAO templates used in this study were obtained by the well-established two-step anodization procedure (Figure 2).

Figure 2. Schematic diagram (A) and photograph (B) of the reactor used for the anodization (for details see Methods). The cathode was fixed in front of the anode. The reactor was placed in a thermal reservoir filled with water and ethanol, which was connected with an external cooling system. The temperature of the electrolyte in the reactor was adjusted by controlling the temperature of the cooling solution in the thermal reservoir during the anodization process.

Both mild and hard anodization protocols (MA and HA) as well as widening by a phosphoric acid etch were applied and afforded uniform AAO substrates (for SEM data see Supporting Information, Figure S-1). The experimental conditions used are summarized in Table 1. The surfaces of these templates were modified by gas phase silanization to render them amine functionalized. Subsequently, we sputtered a 7 nm thin gold layer at an angle of 45° onto some of the templates. This layer was found in independent trial experiments sufficient effective to suppress polyelectrolyte adsorption (see also Figure 5). In the first set of experiment, we investigated how the gold layer affects the literature known deposition of multilayers of PSS and PAH by the LBL technique. We chose on purpose deposition conditions that match those reported by Caruso and co-workers for the deposition in pores with diameters of 400 nm.15 Thus, unlike in the work reported by Genzer and coworkers,22 no increased ionic strengths or divalent ions were utilized. In Figure 3, the top view SEM micrographs acquired on an original AAO template with a pore diameter of 53 ± 9 nm, on an uncoated sample that was modified with two PSS/PAH bilayers, and on a gold-passivated sample that was modified with five PSS/PAH bilayers are compared. The original regular pore structure discernible in Figure 3A is almost completely blocked after deposition of two bilayers (Figure 3B). The morphology observed is consistent with a layer of polyelectrolyte deposited on top of the AAO as well as at/near the pore orifice. By contrast, the gold passivated sample exposes open pores even after the deposition of five bilayers (Figure 3C). The deposited polyelectrolyte can be differentiated from the gold-coated top due to its different secondary electron emission efficiency. The open pore diameter (in the dry state) was found to be 37 ± 4 nm. These data demonstrate that the prime factor that is responsible for pore clogging is the uncontrolled 10093

dx.doi.org/10.1021/la3017062 | Langmuir 2012, 28, 10091−10096

Langmuir

Article

The sample shown in Figure 3C exhibited a more nonuniform layer, which may be tentatively attributed in parts to the onset of the transition from regime 1 to regime 2.21 However, a slight inhomogeneity of the gold coating in conjunction with a nonideal pore cross section may also contribute to the reduced apparent uniformity of the LBL multilayer. A detailed analysis of the dry thickness of the multilayer as observed by SEM at the pore mouth provided an estimate for the wall thickness. These data were compared to dry thicknesses determined by ellipsometry and AFM for LBL adsorption on flat silicon and alumina substrates as well as goldcoated slides (Figure 5). While the deposited polymer layer

Figure 3. SEM: (A) original pores with 53 ± 9 nm pore diameter; (B) AAO after deposition of 2 bilayers; (C) AAO (top surface was passivated with gold) after deposition of 5 bilayers. The gold may be deposited to a very limited extent inside the templte pore close to the pore orifice.

adsorption of polyelectrolyte on the top surface or near the pore mouth of the AAO. Not surprisingly, we observed a very similar passivation function of the gold coating on the AAO templates with different pore diameters. For pores with original inner diameters >100 nm, the multilayers were very uniform in thickness at the pore opening (Figure 4); thus, the multilayer was clearly deposited under regime 1 conditions, as defined by Jonas and co-workers.21

Figure 5. (A) Dry ellipsometric thickness of multilayers as a function of the number of bilayers (PSS/PAH)n deposited on flat silicon and alumina substrates both primed with the aminosilane as well as on the gold substrates. The solid lines correspond to linear least-squares fits of the data shown. (B) Wall thicknesses of nanotubes determined from SEM images versus number of bilayers deposited for different initial pore diameters. The solid lines show a comparison of the linear fit of the thickness data of bilayers inside the AAO template with the fit of the thickness data of bilayers deposited on flat silicon substrate (for latter data see panel A).

thickness on gold leveled off at a thickness of 3 nm, the silicon and flat aluminum oxide slides showed a multilayer buildup with to within the experimental error the same thicknesses, independent of the method used for analysis. Independent SEM data on fracture surfaces of multilayercoated AAO templates as well as AFM data acquired on isolated nanotubes obtained by dissolving the template (see below) support the thickness data for the multilayers inside the pores. Compared to multilayers on flat silicon and alumina, the multilayers inside the pore showed to within the experimental

Figure 4. Sections of SEM images: (A) original AAO, pore diameter 250 ± 11 nm; (B) AAO (the top surface was passivated with gold) after deposition of 5 bilayers. The entire SEM micrographs are shown in the Supporting Information. 10094

dx.doi.org/10.1021/la3017062 | Langmuir 2012, 28, 10091−10096

Langmuir

Article

Figure 6. SEM micrographs of nanotubes of (PSS/PAH)5 multilayer templated in pores with original inner diameter of 53 nm: (A, B) free-standing nanotubes on AAO template (∼7 μm long); (C, D) isolated nanotube after complete removal of the AAO matrix.

37 ± 4 and 52 ± 19 nm, respectively, were obtained, limited only by the size of the AAO template.

error the same thickness and, more importantly, a near identical thickness increment (oxidized aluminum, slope: 2.63 ± 0.03 nm/bilayer; AAO template, slope: 2.70 ± 0.17 nm/bilayer). This can be rationalized by the fact that (i) polymer molecules can indeed diffuse into and out of the pores to at least 8 μm depths20,25 and (ii) that the thickness of the deposited layer was in all cases (much) smaller than the original pore radius. Hence, regime 1 conditions are applicable. For the smallest pores the measured polyelectrolyte layer thickness exhibited a broader distribution. Since a less uniform layer morphology was observed when 5 bilayers were deposited in the smallest pores (53 nm original diameter), this system may be as already mentioned likely close to the regime 1−regime 2 transition (vide supra).21 The multilayers deposited were found to very robust and stable. Removal of the AAO template by phosphoric acid, KOH or NaOH proved successful and stable free-standing nanotubes were obtained for all pore diameters investigated (see Figure 6 for data on the 53 nm diameter pores). Unlike in reports by Genzer and co-workers22 and Caruso et al.,15 our tubes composed of 5 bilayers (corresponding to ∼11 nm wall thickness) were found to be stable in the absence of any thermal cross-linking. The SEM data also showed that the tubes were open at the orifice. By completely removing the substrate, neutralizing the solution, and subsequent purification by dialysis, the nanotubes with lengths between 2 and 8 μm could be isolated and studied by AFM, among others (see Supporting Information).

■

ASSOCIATED CONTENT

S Supporting Information *

Additional characterization data of the porous templates and the nanotubes (SEM, AFM) as well as distributions of pore sizes and tube dimensions. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]; Tel (+49) 271/ 740-2806; Fax (+49) 271/740-2805. Present Address †

Borealis Polyolefine GmbH, InnoTech Operational Support, Advanced Polymer Characterization, Sankt Peter Strasse 25, 4021 Linz, Austria.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge Prof. Dr. Xin Jiang and Ms Petra auf dem Brinke (Department of Mechanical Engineering, University of Siegen, Germany) for access and help with the FESEM measurements as well as Dipl. Ing. Gregor Schulte (Physical Chemistry I), Mr. Dieter Gaumann, and Mr. Bernd Meyer (both mechanical workshop of the Department of Chemistry and Biology, University of Siegen, Germany) for their excellent advice and assistance in fabricating the anodization cell. This work was generously supported by the Siegener Graduate School (Development of Integral Heterosensor Architectures for the n-Dimensional (Bio) Chemical Analysis), the University of Siegen, the Alexander von Humboldt Foundation (postdoc stipend to D.T.), and the Deutsche Forschungsgemeinschaft (DFG grant INST 221/87-1 FUGG).

■

CONCLUSIONS The extension of the versatile approach to replicate nanoporous anodized aluminum oxide (AAO) by the layer-by-layer (LBL) deposition of polyelectrolytes to the nanoscale was achieved by minimizing the detrimental adsorption of polyelectrolyte on the top surface of amino-silane primed AAO by applying a passivating gold layer. The smallest multilayered poly(styrenesulfonate)/poly(allylamine) nanotubes isolated thus far were shown to be open at their orifice. Hence, free-standing open nanotubes with inner and outer diameters (±2σ) down to 10095

dx.doi.org/10.1021/la3017062 | Langmuir 2012, 28, 10091−10096

Langmuir

■

Article

Membrane Templates with Sub-100-nm Pore Diameters. Small 2010, 6, 2683−2689. (23) Yoo, D.; Shiratori, S. S.; Rubner, M. F. Controlling Bilayer Composition and Surface Wettability of Sequentially Adsorbed Multilayers of Weak Polyelectrolytes. Macromolecules 1998, 31, 4309−4318. (24) Wong, J. E.; Rehfeldt, F.; Hänni, P.; Tanaka, M.; Klitzing, R. V. Swelling Behavior of Polyelectrolyte Multilayers in Saturated Water Vapor. Macromolecules 2004, 37, 7285−7289. (25) Ruckenstein, E.; Li, Z. F. Surface Modification and Functionalization through the Self-Assembled Monolayer and Graft Polymerization. Adv. Colloid Interface Sci. 2005, 113, 43−63.

REFERENCES

(1) Liu, T. B.; Burger, C.; Chu, B. Nanofabrication in Polymer Matrices. Prog. Polym. Sci. 2003, 28, 5−26. (2) Decher, G.; Hong, J. D.; Schmitt, J. Buildup of Ultrathin Multilayer Films by a Self-Assembly Process: III. Consecutively Alternating Adsorption of Anionic and Cationic Polyelectrolytes on Charged Surfaces. Thin Solid Films 1992, 210−211, 831−835. (3) Martin, C. R. Nanomaterials: A Membrane-Based Synthetic Approach. Science 1994, 266, 1961−1966. (4) Shchepelina, O.; Kozlovskaya, V.; Singamaneni, S.; Kharlampieva, E.; Tsukruk, V. V. Replication of Anisotropic Dispersed Particulates and Complex Continuous Templates. J. Mater. Chem. 2010, 20, 6587− 6603. (5) Steinhart, M.; Wendorff, J. H.; Greiner, A.; Wehrspohn, R. B.; Nielsch, K.; Schilling, J.; Choi, J.; Gösele, U. Polymer Nanotubes by Wetting of Ordered Porous Templates. Science 2002, 296, 1997. (6) Kim, E.; Xia, Y.; Whitesides, G. M. Micromolding in Capillaries: Applications in Materials Science. J. Am. Chem. Soc. 1996, 118, 5722− 5731. (7) Viheriälä, J.; Tommila, J.; Leinonen, T.; Dumitrescu, M.; Toikkanen, L.; Niemi, T.; Pessa, M. Applications of UV-Nanoimprint Soft Stamps in Fabrication of Single-Frequency Diode Lasers. Microelectron. Eng. 2009, 86, 321−324. (8) Fredriksson, H.; Alaverdyan, Y.; Dmitriev, A.; Langhammer, C.; Sutherland, D. S.; Zäch, M.; Kasemo, B. Hole−Mask Colloidal Lithography. Adv. Mater. 2007, 19, 4297−4302. (9) Wang, Y. J.; Hosta-Rigau, L.; Lomas, H.; Caruso, F. Nanostructured Polymer Assemblies Formed at Interfaces: Applications from Immobilization and Encapsulation to Stimuli-Responsive Release. Phys. Chem. Chem. Phys. 2011, 13, 4782−4801. (10) Decher, G. Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites. Science 1997, 277, 1232−1237. (11) Skirtach, A. G.; Yashchenok, A. M.; Möhwald, H. Encapsulation, Release and Applications of LbL Polyelectrolyte Multilayer Capsules. Chem. Commun. 2011, 47, 12736−12746. (12) Radt, B.; Smith, T. A.; Caruso, F. Optically Addressable Nanostructured Capsules. Adv. Mater. 2004, 16, 2184−2189. (13) Chandrawati, R.; Hosta-Rigau, L.; Vanderstraaten, D.; Lokuliyana, S. A.; Städler, B.; Albericio, F.; Caruso, F. Engineering Advanced Capsosomes: Maximizing the Number of Subcompartments, Cargo Retention, and Temperature-Triggered Reaction. ACS Nano 2010, 4, 1351−1361. (14) Ma, Y. J.; Dong, W. F.; Hempenius, M. A.; Möhwald, H.; Vancso, G. J. Redox-Controlled Molecular Permeability of CompositeWall Microcapsules. Adv. Mater . 2006, 5, 724−729. (15) Liang, Z. J.; Susha, A. S.; Yu, A. M.; Caruso, F. Nanotubes Prepared by Layer-by-Layer Coating of Porous Membrane Templates. Adv. Mater. 2003, 15, 1849−1853. (16) Ai, S. F.; Lu, G.; He, Q.; Li, J. B. Highly flexible polyelectrolyte nanotubes. J. Am. Chem. Soc. 2003, 125, 11140−11141. (17) Masuda, H.; Fukuda, K. Ordered Metal Nanohole Arrays Made by a Two-Step Replication of Honeycomb Structures of Anodic Alumina. Science 1995, 268, 1466−1468. (18) Lee, W.; Ji, R.; Gösele, U.; Nielsch, K. Fast Fabrication of LongRange Ordered Porous Alumina Membranes by Hard Anodization. Nat. Mater. 2006, 5, 741−747. (19) Alvarez, S. D.; Li, C. P.; Chiang, C. E.; Schuller, I. K.; Sailor, M. J. A Label-Free Porous Alumina Interferometric Immunosensor. ACS Nano 2009, 3, 3301−3307. (20) Lazzara, T. D.; Behn, D.; Kliesch, T. T.; Janshoff, A.; Steinem, C. Phospholipids as an Alternative to Direct Covalent Coupling: Surface Functionalization of Nanoporous Alumina for Protein Recognition and Purification. J. Colloid Interface Sci. 2012, 366, 57−63. (21) Roy, C. J.; Dupont-Gillain, C.; Demoustier-Champagne, S.; Jonas, A. M.; Landoulsi, J. Growth Mechanism of Confined Polyelectrolyte Multilayers in Nanoporous Templates. Langmuir 2010, 26, 3350−3355. (22) Cho, Y.; Lee, W.; Jhon, Y. K.; Genzer, J.; Char, K. Polymer Nanotubules Obtained by Layer-by-Layer Deposition Within AAO10096

dx.doi.org/10.1021/la3017062 | Langmuir 2012, 28, 10091−10096