Shaping Mesoporous Films Using Dewetting on X-ray Pre-patterned

Mar 4, 2011 - Division of Materials Science and Engineering, Commonwealth Scientific and Industrial Research Organisation (CSIRO),. Private Bag 33 ...
0 downloads 0 Views 4MB Size
ARTICLE pubs.acs.org/Langmuir

Shaping Mesoporous Films Using Dewetting on X-ray Pre-patterned Hydrophilic/Hydrophobic Layers and Pinning Effects at the Pattern Edge Stefano Costacurta,† Paolo Falcaro,‡ Luca Malfatti,§ Daniela Marongiu,§ Benedetta Marmiroli,|| Fernando Cacho-Nerin,|| Heinz Amenitsch,|| Nigel Kirkby,^ and Plinio Innocenzi*,§ †

Associazione CIVEN, Via delle Industrie 5, 30175 Venezia, Italy Division of Materials Science and Engineering, Commonwealth Scientific and Industrial Research Organisation (CSIRO), Private Bag 33, Clayton South MDC, Victoria 3169, Australia § Laboratorio di Scienza dei Materiali e Nanotecnologie, Universita di Sassari, D.A.P., CR-INSTM, Palazzo del Pou Salit, Piazza Duomo 6, 07041 Alghero, Sassari, Italy Institute of Biophysics and Nanosystems Research, Austrian Academy of Sciences, Schmiedlstrasse 6, 8042 Graz, Austria ^ Australian Synchrotron, 800 Blackburn Road, Clayton, Victoria 3168, Australia

)



ABSTRACT: Ordered mesoporous silica micrometer-sized structures have been fabricated via selective dewetting of the coating sol on a hydrophilic/hydrophobic fluorinated silica substrate, which had been pre-patterned using deep X-ray lithography with a synchrotron radiation source. We have observed that deposition of mesoporous films on the prepatterned areas can be used as a design tool for obtaining regions of specific geometry and dimensions. The evaporation of the solution in constrained conditions because of pinning at the pattern edges gives layers with thicker edges. This edge effect appears dependent upon the dimension of the pre-patterned hydrophilic/hydrophobic layer; in smaller patterns, the evaporation is too fast and thickening of the edges is not observed. We have used infrared imaging, optical profilometry, and atomic force microscopy to characterize the patterned layers and the edge effect, produced by pinning at the border of the microstructures.

’ INTRODUCTION The current strategies for the self-assembly of mesoporous ordered films allow for the control of porosity features, such as size, distribution, shape, and interconnectivity. In particular, mesoporous films (pore size between 2 and 50 nm)1 are typically synthesized via evaporation-induced self-assembly (EISA),2 in which amphiphilic surfactants or block co-polymers are the structure-directing agents forming micellar aggregates upon self-assembly.3 Although mesoporous ordered films having different chemical composition, crystallinity, and porosity have already demonstrated specific properties for practical applications, more effort is still required to integrate them in functional devices. The key technological hurdle is the combined ability to tailor coating properties and to fabricate the film in a precise shape, e.g., in circuits, dots, channels, etc. This material shaping process can be attained by a top-down process, which directs the macroscopic shape of a mesoporous film through either photonic/electron beam lithography or soft lithographic techniques (see ref 4 and references therein). For example, ultraviolet (UV) lithography has been used in combination with mesoporous sol-gel films, in which a photoacid generator induces condensation of the inorganic network upon UV illumination r 2011 American Chemical Society

through a mask, thereby generating chemical contrast between the exposed and unexposed portions of the film; the latter can be etched using a suitable solvent.5 Patterning of mesostructured and hierarchical films without the need of adding radiationsensitizing species to the film composition has been recently achieved using X-ray synchrotron radiation.6-8 The patterning is realized by the combined effects of organic removal (block copolymer) induced by radiation damage and inorganic condensation promoted by the high-energy and high-flux incident radiation through the formation of radical species. These lithographic techniques require at least two steps: exposure to radiation with a suitable power density spectrum and dose and chemical etching of the exposed material (developing). In these processes, the patterned material behaves as a negative resist; i.e., the exposed part remains on the substrate after developing, whereas the unexposed part is selectively removed. This can be a disadvantage if the patterned mesoporous film contains organic functional moieties, which could be damaged Received: September 27, 2010 Revised: January 20, 2011 Published: March 04, 2011 3898

dx.doi.org/10.1021/la103863d | Langmuir 2011, 27, 3898–3905

Langmuir

ARTICLE

Figure 1. (a) Surface functionalization by a FAS-13, (b) TEOS-FAS-13 co-condensed film, and (c) silica nanoparticle monolayer modified by FAS-13 and the water contact angle measured in the different surfaces before (middle) and after (bottom) X-ray exposure (50 J cm-2).

by exposure to the ionizing radiation.7 Furthermore, it has been shown that hard X-rays interact with not only mesostructured films of silica composition but also other oxides, such as hafnia.9 In this work, we have developed a new route for the fabrication of mesoporous circuits and devices, which is based on the prepatterning of a hydrophilic/hydrophobic thin film and the selective deposition of mesoporous layers on the hydrophilic regions of the surface. Hydrophilic/hydrophobic surfaces are obtained by patterning perfluoroalkyl-functionalized10-12 surfaces using X-rays, and these surfaces having spatially defined wettability can be exploited in microfluidic devices.13 Using an integrated approach, we have modulated not only the dimension of the mesoporous structures, which is in the micrometer range, but also the shape. In particular, we have used the pinning effect at the contact edge to obtain concave-convex shapes; this capability of modulating the shape is dependent upon the dimension of the patterns. The integration of different techniques, such as surface hydrophobization, deep X-ray lithography (DXRL), and EISA, has shown to allow for the design of porous ordered thin films with controlled dimensions of the pore sizes and 2D/3D geometry.

’ EXPERIMENTAL SECTION Tetraethoxysilane (TEOS), methyltriethoxysilane (MTES), 1H,1H,2H, 2H-perfluorooctyl-trichlorosilane (FAS-13), 3-aminopropyltriethoxysilane (APTES), fluorescein isothiocyanate (FITC), hydrochloric acid aqueous 1 M solution (HCl), and Pluronic F127 were purchased from Sigma-Aldrich. Ethanol (EtOH) and toluene were purchased from Carlo Erba Reagents. All chemicals were used without further purification. Boron-doped silicon wafers, one-side polished, with 10-20 Ω cm resistivity (Si-Mat, Germany) were employed as the substrates. The silicon substrate was modified using three different routes and dip-coated in a FAS-13 solution, in a solution in which FAS-13 was cohydrolyzed with TEOS, and in a solution containing silica nanoparticles and further modification with FAS-13.

Precursor Solution for Silicon Surface Hydrophobization. FAS-13 (15 cm3) was dissolved in ethanol (15 cm3), and this solution was used to deposit a thin layer on silicon substrates via dip-coating at 10 mm min-1 as the withdrawal rate, 10% relative humidity (RH), and 25 C. Substrates were dipped 5 times and, at the end of the last coating, were thermally treated in air for 1.5 h at 150 C. Precursor Solution for the Silica Fluorinated Thin Layer. A solution was prepared mixing EtOH (30 cm3), TEOS (4.2 cm3), FAS-13 (0.12 cm3), H2O (18 MΩ) (3.56 cm3), and 1 M HCl (0.4 cm3). The solution was stirred for 2 h to allow co-hydrolysis of FAS-13 with TEOS and deposited by dip coating (20% RH at 25 C) on silicon wafers; the withdrawal speed was set to 10 mm min-1. The samples were thermally treated in air for 1 h at 80 C and had a thickness of 15 nm [measured using an Alpha-Stepper IQ stylus profilometer (KLA Tencor)]. Silica Nanoparticle Layer. Silica nanoparticles were prepared by mixing ammonia (1.5 cm3, 30% in weight), EtOH (25 cm3), and TEOS (0.5 cm3); the solution was left to react at 25 C under stirring for 12 h. A silicon wafer was dipped 5 times in such solution at a withdrawal speed of 300 mm min-1, 10% RH, and 25 C in the deposition chamber. After the deposition of the silica nanoparticle layer, the surface was functionalized using the same procedure described for bare silicon. The hydrophobic functionalization on nanoparticle films and bare silicon has been similarly performed. Mesoporous Layer Precursor Solution. We have synthesized two different precursor solutions to obtain mesoporous silica and hybrid organic-inorganic mesoporous silica films. The solution for the deposition of silica mesoporous films was prepared in two steps: at first, a sol containing the silica source was prepared by mixing TEOS (4.3 cm3), EtOH (3 cm3), and 0.768 M HCl (0.355 cm3). This sol was stirred for 1 h at 25 C before mixing with another solution prepared dissolving Pluronic F127 (1.3 g) in EtOH (15 cm3), toluene (3 cm3), and 0.057 M HCl (1.5 cm3) solution. The solution for the deposition of hybrid silica mesoporous film was prepared following the same protocol previously described for silica films, but instead of only TEOS, a mixture of TEOS and methyltriethoxysilane (MTES) has been used (TEOS/MTES = 1:0.33). 3899

dx.doi.org/10.1021/la103863d |Langmuir 2011, 27, 3898–3905

Langmuir

ARTICLE

DXRL. The films were patterned using the DXRL beamline at Elettra Synchrotron Light Laboratory (Trieste, Italy) working at 2 GeV. A dose matrix (exposure to increasing doses) from 3.4 to 50 J cm-2 was performed without any mask; samples were also exposed through an X-ray mask containing a single channel with reservoir (width of 800 μm). The mask had a 20 μm thick gold absorber and a transparent membrane made of 4 μm of silicon nitride and 25 μm of SU8. Deposition of Mesoporous Layers on Patterned Fluorinated Substrates. The fluorinated layers were patterned through direct exposure to X-rays to create hydrophilic/hydrophobic patterned surfaces. In a second step, the selective dewetting effect was used and the patterned fluorurated film was dip-coated (20% RH at 25 C) in the mesoporous silica precursor solution using a withdrawal rate of 150 mm min-1. Once the mesoporous layers were deposited, calcination was used to remove the surfactant;16-20 the deposited layers were baked at 150 C for 1 h and calcined at 400 C for another hour to obtain mesoporous structures.

Surface Functionalization of Mesoporous Silica Patterns. The mesoporous patterned structures were immersed at 60 C for 12 h in a solution prepared by mixing 10 cm3 of APTES and 50 cm3 of toluene. A second functionalization was obtained by immersing the amino-functionalized mesoporous patterned layers into a solution prepared mixing FITC (15 mg) in EtOH (45 cm3). After 30 min, the samples were removed from the dye solution and thoroughly rinsed with methanol. Materials Characterization. The water contact angle was measured using a DataPhysics Contact Angle System (OCA 20); the measurement was performed at 22 C and 50% RH by dispensing a 5 μL droplet on the film surface. Fourier transform infrared (FTIR) spectroscopy and FTIR imaging were performed using a Bruker Hyperion 3000 microscope attached to a Bruker Vertex 70 interferometer working in the mid-IR range with a conventional Globar source and a KBr beamsplitter. The microscope was equipped with a liquidnitrogen-cooled mercury-cadmium-telluride (MCT) detector and a motorized sample stage that allowed for mapping of the samples using a rectangular aperture of 15  30 μm2. Each spectrum was obtained in transmission mode, averaging 700 interferograms at a resolution of 8 cm-1. The ordered porous mesostructured layers were investigated by grazing incidence small-angle X-ray scattering (GISAXS) at the smallangle X-ray scattering (SAXS) beamline of the Australian Synchrotron (Melbourne, Australia), averaging 10 acquisitions with 5 s of exposure time each. The instrumental glancing angle was set maintaining an incident X-ray beam (λ = 1.54 Å) smaller than 3. A two-dimensional pixel detector (Pilatus 1 M Dectris, Ltd., containing 3 modules in a row with a total of 1120  967 pixels) was used to record the GISAXS patterns. Dye-functionalized mesoporous silica structures were left in a closed vessel until they were measured with a laser scanner (GenePix 4000B, Axon Instruments); the measurements were performed with the maximum resolution (5 μm), under excitation with 532 and 633 nm wavelengths (laser power set to 33% and electronic gain set to 50).

’ RESULTS AND DISCUSSION Surface Functionalization with FAS-13 and Effect of X-rays. We have modified the substrate surface using three

different methods to modulate the wettability of the functionalized film on silicon; we have deposited via dip-coating (1) a fluorinated layer (FAS-13 solution), (2) a silica-fluorinated layer (TEOS-FAS-13 co-condensed sol), and (3) a monolayer of silica nanospheres followed by functionalization with FAS-13 (Figure 1). These three approaches can be changed according to the desired coating properties: the water contact angle is the highest in the silica nanoparticle-coated surface because of the roughness induced by the presence of the nanospheres, whereas

Figure 2. Water contact angle of a TEOS-FAS-13 co-condensed film exposed to deep X-rays with no mask, measured as a function of the X-ray dose (continuous line is a guide for the eyes).

in the second example, a co-condensed TEOS-FAS-13 layer is mechanically more robust. The different functionalizations allow for modulation of the water contact angle, which has been measured to be 70 (FAS-13 layer), 95 (FAS-13-TEOS layer), and 135 (silica nanoparticle-FAS-13 layer). These values dramatically decrease after X-ray exposure to a dose of 50 J cm-2; the contact angles fall to 20 (FAS-13 layer), 13 (TEOS-FAS-13 layer), and 3 (silica nanoparticle-FAS-13 layer). The exposure of the surface to X-rays, regardless of the functionalization method, produces, in any case, a drastic change of its wettability properties. This is a very important point that has to be stressed; in fact, examples of realization of mesostructures via dewetting on UV patterned self-assembled monolayers deposited by chemical vapor deposition (CVD) are already reported in the literature.14 However, the proposed sol-gel processing offers the possibility to deposit the coating in 1/300 of the time employed to prepare self-assembled monolayers by CVD. In addition the X-ray exposure reduces to 1/100 the time required for patterning with respect to UV irradiation. Even if hard X-rays are not easily available combining sol-gel processing with DXRL is a very fast method for mesoporous micropattern preparation. Accordingly, with the trend to move toward faster lithographic technologies,7 this integrated sol-gel/DXRL/EISA method appears as a competitive process for the fabrication of mesoporous-material-based devices. An important experimental point is the determination of the suitable X-ray dose to pattern the fluorinated silica films; this has been obtained performing a dose matrix. We have decided to use the TEOS-FAS-13 modified surface as the model sample, because it appeared more efficient with respect to the fluorinated layer and more robust and simple to obtain in comparison to the silica nanoparticle fluorinated layer. Fluorinated co-condensed TEOS-FAS-13 samples have been exposed to increasing X-ray doses, up to 50 J cm-2 with no mask; immediately after the exposure, the water contact angle has been measured as a function of the X-ray dose (Figure 2). The contact angle decreased from 95 to 13 with increasing doses because of decomposition and removal of the fluorinated alkyl moieties in the film caused by the incident radiation. In particular, it has been found that doses higher than 17 J cm-2 induce a dramatic change of the surface properties from hydrophobic to hydrophilic by completely removing the fluorinated silica layer. DXRL and Selective Dewetting for Mesoporous Micropattern Fabrication. Exposure of the fluorinated layers to X-rays 3900

dx.doi.org/10.1021/la103863d |Langmuir 2011, 27, 3898–3905

Langmuir

ARTICLE

Figure 3. Schematic representation of the procedure for the synthesis of a mesoporous structure via X-ray lithography and selective dewetting. A fluorinated film is (a) deposited and (b) exposed to X-rays through a mask, (c) creating a hydrophilic/hydrophobic patterned surface, (d) on which a mesoporous film is deposited via selective dewetting.

Figure 4. (a) Optical image and (b) AFM topography of a 8 μm wide silica mesoporous channel deposited by dewetting on a TEOS-FAS-13 prepatterned film.

completely removes the organic species, as we have seen; this effect can be used as a lithographic tool. We have developed a three-step lithographic technique based on the creation of a hydrophilic/hydrophobic patterned surface and the selective deposition on the hydrophilic regions of mesoporous films. The process is illustrated in Figure 3: in the first step, a silicon substrate is functionalized with a hydrophobic fluorinated layer, which reduces the surface wettability (Figure 3a). In the second step, the fluorinated species are locally removed by exposing the sample to X-ray synchrotron radiation through a mask; this allows for the creation of a hydrophilic/hydrophobic pattern (Figure 3b). In the third step, the patterned substrate is dipcoated into the mesoporous precursor sol and a film is selectively deposited on the hydrophilic regions. The driving force for the formation of the patterned film is dewetting: the sol selectively dewets from the unexposed hydrophobic regions during

withdrawal of the substrate, and a film is preferentially deposited on the exposed hydrophilic regions (panels c and d of Figure 3). We have, therefore, applied this procedure to fabricate structures in the micrometer scale, even if, in principle, the technique could be easily extended to a submicrometer range; an example of such a type of structure is shown in Figure 4. A 8 μm wide and 11 mm long silica mesoporous channel has been formed on the fluorinated silica film using a dose of 20 J cm -2 ; such a value has been selected because, according to the dose matrix, it guaranteed a small water contact angle (20) of the final film (see Figure 2). The patterned film has then been dip-coated into the mesoporous silica precursor sol within a few minutes since X-ray exposure. During withdrawal of the substrate from the precursor sol, the liquid wetted the whole film up to a height of 0.5 cm from the solution container, where 3901

dx.doi.org/10.1021/la103863d |Langmuir 2011, 27, 3898–3905

Langmuir

ARTICLE

Figure 5. GISAXS pattern showing the ordered mesophase of the hybrid organic inorganic (a) TEOS-MTES and (b) TEOS mesoporous films. Before measurements, the films were fired at 400 C for 30 min. The pattern on panel a was indexed as tetragonal with an I4/mmm space group, while the pattern on panel b was indexed as orthorhombic with an Fmmm space group.

Figure 6. Stylus profilometry of a mesoporous silica channel deposited by selective dewetting on a TEOS-FAS-13 pre-patterned film. The plot has been obtained by measuring the width across the channel. The enlargement with the correct (unexaggerated) proportions is shown for visualizing the inclination of the channel walls with respect to the substrate.

dewetting on the unexposed regions of the film caused the sol to flow toward the exposed areas (i.e., the channel). The optical images at two different enlargements of Figure 4a show that a well-defined channel, which appears in blue with respect to the brown fluorinated unexposed layer, is formed. The atomic force microscopy (AFM) image (Figure 4b) shows that the mesoporous silica channel has a height of 310 nm with a well-defined and sharp border. It is interesting to observe that this process can be used with not only pure mesoporous silica but also hybrid organicinorganic sols, such as TEOS-MTES mixtures, and it results in an ordered mesoporous film at the end of device processing. Panels a and b of Figure 5 show the GISAXS patterns obtained from TEOS-MTES and pure TEOS mesoporous-shaped films, respectively, after firing at 400 C for 30 min. The first pattern has been indexed as a tetragonal body centered ordered porous structure (I4/mmm space group), while the second pattern has been indexed as a rhombohedral ordered porous structure (Fmmm space group). These structures correspond to those

previously observed in hybrid mesoporous silica films obtained using the same precursor sol.15,16 Edge Phenomena during Selective Dewetting Deposition of Mesoporous Films. We have extended the application of the technique to produce more complex patterns of larger dimension with the potential target of microfluidic applications: a 800 μm wide channel with central reservoir (4000  2000 μm) and overall length of 11 000 μm has been patterned on the fluorinated silica film using the optimized dose of 50 J cm-2. On this pattern, a mesoporous film has been deposited; the thickness has been measured by a stylus profilometer across the channel width (Figure 6) and has been found to vary between 1 μm at the edges and 200 nm at the center of the channel. The profile of the edge appears sharp; this means that DXRL enables writing very precise hydrophilic/hydrophobic areas. The optical profile of the channel clearly indicates the thickening of the film border, which is an interesting effect because of chemical-physical phenomena correlated to the evaporation of the precursor sol. The thickened walls result to have a small inclination with respect to the substrate (see Figure 6). Interestingly, this thickening effect has not been found in smaller channels and seems dependent upon the pattern dimensions. Therefore, we have studied more in detail this effect using FTIR imaging to characterize the channel (Figure 7) and the central reservoir (Figure 8) of the pattern. The FTIR images in Figure 7 are shown in a false color scale that indicates the intensity of the integrated band that has been used for imaging. Panels a (3D graphical representation with integrated absorbance in false color scale in the z axis) and b (2D graphical representation with integrated absorbance in false color scale) of Figure 7 show the FTIR image obtained by integrating the spectroscopic signal associated with SiO2 (SiO-Si TO3 mode at 1170 cm-1) in a selected region of the patterned channel.17 Outside the channel, no FTIR signals related to SiO2 have been detected because of the small film thickness, which indicates the dewetting of the mesostructured films from the hydrophobic side of the substrate. The mesoporous film at the end of the process is deposited only on the region exposed to X-rays. In the FTIR image (Figure 7), the border regions of the channel show a more intense Si-O-Si signal compared to the center of the structure, which can be explained by a greater thickness of the film at the edges, as shown by profilometer measurements. We have also obtained FTIR images 3902

dx.doi.org/10.1021/la103863d |Langmuir 2011, 27, 3898–3905

Langmuir

ARTICLE

Figure 7. Three- and two-dimensional representation of the FTIR imaging results of a mesoporous silica channel deposited by selective dewetting on a TEOS-FAS-13 pre-patterned film (panels a and b, respectively). The false color scale indicates the intensity of the integrated band (Si-O-Si TO3 mode at 1170 cm-1) that has been used for making the image.

Figure 8. Optical (left side) and FTIR (right side) images of a mesoporous silica circuit that has been obtained via selective dewetting on a TEOS-FAS13 pre-patterned film. The FTIR image in false color scale is obtained by integrating the Si-O-Si stretching band peaking at 1170 cm-1.

from the central reservoir of the pattern; the flexibility of the lithographic process allows, in fact, obtaining structures that are not limited to regular arrays or linear geometries, as shown in Figure 8. The optical image (left side) shows different interference colors that already give an indication of some difference in thickness; the FTIR image on the right side (2D graphical representation with integrated absorbance in false color scale) has been obtained by integrating the silica band (Si-O-Si TO3 mode at 1170 cm-1) similarly to the image in Figure 7. The border of the patterned area appears again very sharp, but also in this case, the edge of the film is thicker than the central region, with the formation of a kind of concave patterned area in the reservoir of the circuit. The FTIR images, therefore, indicate that the deposition of films from liquid phases on patterned areas gives rise to a specific edge effect that produces thicker walls at the borders. The infrared images also show that the thicker edges are formed by silica. We think that this effect is due to evaporation in conditions of pinning, which is quite common during evaporation of droplets and is responsible for the so-called “coffee stain”. The edge effect is particularly evident in a confined environment for evaporation on a surface such as the one described here and is a common phenomenon caused by capillary flows to the edge of a liquid layer (the precursor sol) during solvent evaporation.18 More in general, this effect is observed whenever a geometrical constraint pins the contact line of an evaporating liquid layer to a defined position. In our system, the pinning effect is due to the chemical surface modification induced by the X-ray pattering, which produces sharp hydrophilic domains onto the hydrophobic

Figure 9. Illustrated representation of the concave and convex shapes that can be obtained: patterns of larger dimensions give concave shapes, while convex structures are obtained using contact pinning effects at the film edge in the case of patterns smaller than around 10 μm.

fluorinated coating. It has been demonstrated that, if pinning is accompanied by a non-uniform evaporation of the solvent, the mass conservation requires a flow toward the contact edge, producing the thickening effect.19 We have to underline that this edge effect is typically observed in sol-gel dip-coated films,20 thicker regions are observed at the side and bottom margin of a substrate, and their width is proportional to withdrawal speed. An important conclusion is that, even if we have not performed a systematic analysis of the phenomenon, the edge effect is clearly dependent upon the dimension of the 3903

dx.doi.org/10.1021/la103863d |Langmuir 2011, 27, 3898–3905

Langmuir

ARTICLE

Figure 10. Scheme of the procedure to functionalize a mesoporous silica film with fluorescein isothiocyanate to the amino-functionalized silica network (left side). Laser scanner image of the APTES-mesoporous silica channel functionalized with fluorescein isothiocyanate (right side).

patterned region, likely because a smaller pattern wets a smaller amount of solution, which evaporates fast enough to avoid a capillary flow to the edge. On the other hand, this effect can also be used for modulating the patterned areas, depending upon the geometry of the area available for the evaporation of the precursor sol layers of concave-convex geometry, and welldefined dimension can be produced just modulating the edge effect (Figure 9). Mesoporous Patterned Film Functionalization. Using selective dewetting on X-ray pattern layers, we have, therefore, been able to produce mesoporous ordered microstructures of different dimension and shape. If the dimension of these structures is larger than several micrometers, a pinning effect during film deposition allows for the production of a concave shape, while a convex geometry is observed for patterns smaller than around 10 μm. A further step ahead is the functionalization of the patterned convex structure; the mesoporous silica channels have then been functionalized using APTES to add NH2 moieties on the mesopore walls. The functionalized films have been subsequently immersed in a FITC solution to covalently link the fluorescein molecule to the amino-functionalized silica network, as schematically reported on the left side in Figure 10. The laser scanner image on the right side in Figure 10 indicates that functionalization occurred only in the amine-modified mesoporous channel; the mesoporous film modified with amines and doped by FITC forms the straight channel in light green color, and the unexposed side of the film is in dark green color. The fluorescence image did not decrease in intensity even after 5 rinsing cycles with methanol, demonstrating a stable bonding between the guest dye and the mesoporous matrix.

’ CONCLUSIONS In this work, we have shown that, through an integrated material processing is possible to design mesoporous ordered microstructures in terms of dimension and shape. DXRL has been used for obtaining hydrophobic/hydrophilic patterns on fluorinated layers. The technique is highly versatile and can be applied on substrates of different roughness to obtain patterns of different geometries. The combination of X-ray lithography and selective dewetting allows us to obtain ordered mesoporous micropatterned structures. Concave- or convex-shaped layers can be obtained using the pinning effect at the edge of the patterns; in patterns of small dimensions, convex channels can be obtained, while concave structures form in the case of large patterns. After shaping, the films can be easily functionalized, as shown by the successful surface modification by amine groups. ’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT The CSIRO OCE Science Leader Scheme is acknowledged for supporting this work. Stefano Costacurta acknowledges Guido Costacurta for help in data analysis. Part of this research was undertaken on the SAXS beamlines at the Australian Synchrotron, Victoria, Australia. We acknowledge travel funding provided by the International Synchrotron Access Program (ISAP) managed by the Australian Synchrotron. The ISAP is an initiative of the Australian Government being conducted as part of the 3904

dx.doi.org/10.1021/la103863d |Langmuir 2011, 27, 3898–3905

Langmuir

ARTICLE

National Collaborative Research Infrastructure Strategy. R. Trombone is acknowledged for inspiring discussion.

’ REFERENCES (1) Roquerol, J.; Avnir, D.; Fairbridge, C. W.; Everett, D. H.; Haynes, J. H.; Pernicone, N.; Ramsay, J. D. F.; Sing, K. S. W.; Unger, K. K. Pure Appl. Chem. 1994, 66, 1739–1758. (2) Brinker, C. J.; Lu, Y.; Sellinger, A.; Fan, H. Adv. Mater. 1999, 11, 579–585. (3) Soler-Illia, G.; Sanchez, C.; Lebeau, B.; Patarin, J. Chem. Rev. 2002, 102, 4093–4138. (4) Innocenzi, P.; Kidchob, T.; Falcaro, P.; Takahashi, M. Chem. Mater. 2008, 20 (3), 607–614. (5) Lu, Y.; Yang, Y.; Lu, M.; Huang, J.; Fan, H.; Haddad, R.; Lopez, G.; Burns, A. R.; Sasaki, D. Y.; Shelnutt, J.; Brinker, C. J. Nature 2001, 410, 913–917. (6) Falcaro, P.; Costacurta, S.; Malfatti, L.; Takahashi, M.; Kidchob, T.; Casula, M. F.; Piccinini, M.; Marcelli, A.; Marmiroli, B.; Amenitsch, H.; Schiavuta, P.; Innocenzi, P. Adv. Mater. 2008, 20, 1864–1869. (7) Falcaro, P.; Malfatti, L.; Vaccari, L.; Amenitsch, H.; Marmiroli, B.; Grenci, G.; Innocenzi, P. Adv. Mater. 2009, 21, 4932–4936. (8) Costacurta, S.; Malfatti, L.; Patelli, A.; Falcaro, P.; Amenitsch, H.; Marmiroli, B.; Grenci, G.; Piccinini, M.; Innocenzi, P. Plasma Processes Polym. 2010, 7, 459–465. (9) Malfatti, L.; Kidchob, T.; Costacurta, S.; Falcaro, P.; Schiavuta, P.; Amenitsch, H.; Innocenzi, P. Chem. Mater. 2006, 18 (19), 4553–4560. (10) Fustin, C.-A.; Glasser, G.; Spiess, H. W.; Jonas, U. Langmuir 2004, 20 (21), 9114–9123. (11) Tadanaga, K.; Morinaga, J.; Matsuda, A.; Minami, T. Chem. Mater. 2000, 12 (3), 590–592. (12) Tadanaga, K.; Fujii, T.; Matsuda, A.; Minami, T.; Tatsumisago, M. J. Sol-Gel Sci. Technol. 2004, 31, 299–302. (13) Falcaro, P.; Innocenzi, P. J. Sol-Gel Sci. Technol. 201010.1007/ s10971-009-2127-7. (14) Hozumi, A.; Kojima, S.; Nagano, S.; Seki, T.; Shirahata, N.; Kameyama, T. Langmuir 2007, 23, 3265. (15) Falcaro, P.; Costacurta, S.; Mattei, G.; Amenitsch, H.; Marcelli, A.; Guidi, M. C.; Piccinini, M.; Nucara, A.; Malfatti, L.; Kidchob, T.; Innocenzi, P. J. Am. Chem. Soc. 2005, 127, 3838. (16) Falcaro, P.; Grosso, D.; Amenistch, H.; Innocenzi, P. J. Phys. Chem. B 2004, 108, 10942. (17) Brinker, C. J.; Hurd, A. J.; Schunk, P. R.; Frye, G. C.; Ashley, C. S. J. Non-Cryst. Solids 1992, 147-148, 424–436. (18) Innocenzi, P.; Malfatti, L.; Piccinini, M.; Grosso, D.; Marcelli, A. Anal. Chem. 2008, 81, 551–556. (19) Deegan, R. D. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 2000, 61, 475. (20) Arfsten, N. J.; Eberle, A.; Otto, J.; Reich, A. J. Sol-Gel Sci. Technol. 1997, 8, 1099–1104.

3905

dx.doi.org/10.1021/la103863d |Langmuir 2011, 27, 3898–3905