pubs.acs.org/Langmuir © 2011 American Chemical Society
Island Organization of TiO2 Hierarchical Nanostructures Induced by Surface Wetting and Drying M. Fusi,† F. Di Fonzo,‡ C. S. Casari,†,‡ E. Maccallini,§ T. Caruso,§ R. G. Agostino,§ C. E. Bottani,†,‡ and A. Li Bassi*,†,‡ † NEMAS (Center for NanoEngineered MAterials and Surfaces) and Dipartimento di Energia, Politecnico di Milano, via Ponzio 34/3, 20133 Milano, Italy, ‡Center for Nano Science and Technology of IIT@PoliMI, Via Pascoli 70/3, 20133 Milano, Italy, and §CNISM-Dipartimento di Fisica, Universit a della Calabria, Ponte Bucci, Cubo 33c, I-87036 Arcavacata di Rende (CS), Italy
Received October 1, 2010. Revised Manuscript Received December 23, 2010 We report on the reorganization and bundling of titanium oxide nanostructured layers, induced by wetting with different solvents and subsequent drying. TiO2 layers are deposited by pulsed laser deposition and are characterized by vertically oriented, columnar-like structures resulting from assembling of nanosized particles; capillary forces acting during evaporation induce bundling of these structures and lead to a micrometer-size patterning with statistically uniform islands separated by channels. The resulting surface is characterized by a hierarchical, multiscale morphology over the nanometer-micrometer length range. The structural features of the pattern, i.e., characteristic length, island size, and channel width, are shown to depend on properties of the liquid (i.e., surface tension) and thickness and density of the TiO2 layers. The studied phenomenon permits the controlled production of multiscale hierarchically patterned surfaces of nanostructured TiO2 with large porosity and large surface area, characterized by superhydrophilic wetting behavior without need for UV irradiation.
Introduction High aspect ratio nanostructures (e.g., nanotubes, nanorods, nanowires) are nowadays studied for an incredibly wide range of applications as a consequence of their peculiar properties and of the constant demand for miniaturization in the fields of electronic, electromechanics, and fluidic devices as well as of the need for high surface area systems in the fields of biotechnology, energy, and chemistry (see e.g. refs 1-4). The interaction of these systems with liquids has recently attracted a great attention since surface and capillary forces acting during wetting and drying may have severe reorganization effects. Such effects are considered a major drawback for several applications,5-7 while in other cases they have been exploited to confer unique material properties.8-11 A growing number of papers have been devoted to these reorganization phenomena, *Corresponding author. E-mail:
[email protected]. (1) Bauer, L. A.; Birenbaum, N. S.; Meyer, G. J. J. Mater. Chem. 2004, 14, 517– 526. (2) Hsu, H.-C.; Wu, W.-W.; Hsu, H.-F.; Chen, L.-J. Nano Lett. 2007, 7, 885–889. (3) Chen, L. Q.; Chan-Park, M. B.; Zhang, Q.; Chen, P.; Li, C. M.; Li, S. Small 2009, 5, 1043–1050. (4) Haberkorn, N.; Gutmann, J. S.; Theato, P. ACS Nano 2009, 3, 1415–1422. (5) Lau, K. K. S.; Bico, J.; Teo, K. B. K.; Chhowalla, M.; Amaratunga, G. A. J.; Milne, W. I.; McKinley, G. H.; Gleason, K. K. Nano Lett. 2003, 3, 1701–1705. (6) Zhu, K.; Vinzant, T. B.; Neale, N. R.; Frank, A. J. Nano Lett. 2007, 7, 3739– 3746. (7) Stoykovich, M. P.; Cao, H. B.; Yoshimoto, K.; Ocola, L. E.; Nealey, P. F. Adv. Mater. 2003, 15, 1180–1184. (8) Correa-Duarte, M. A.; Wagner, N.; Rojas-Chapana, J.; Morsczeck, C.; Thie, M.; Giersig, M. Nano Lett. 2004, 4, 2233–2236. (9) Futaba, D. N.; Hata, K.; Yamada, T.; Hiraoka, T.; Hayamizu, Y.; Kakudate, Y.; Tanaike, O.; Hatori, H.; Yumura, M.; Iijima, S. Nature Mater. 2006, 5, 987–994. (10) Schierhorn, M.; Lee, S. J.; Boettcher, S. W.; Stucky, G. D.; Moskovits, M. Adv. Mater. 2006, 18, 2829–2832. (11) Pokroy, B.; Kang, S. H.; Mahadevan, L.; Aizenberg, J. Science 2009, 323, 237–240. (12) Chakrapani, N.; Wei, B. Q.; Carrillo, A.; Ajayan, P. M.; Kane, R. S. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 4009–4012.
Langmuir 2011, 27(5), 1935–1941
in particular involving 1D structures like nanorods, nanowires, and nanotubes.6,11-15 Nevertheless, the investigation of such phenomena on more complex systems such as elongated microstructures composed by a hierarchical self-assembly of nanoparticles is still lacking. Recently, we have developed, by means of pulsed laser deposition (PLD), titanium oxide (TiO2) hierarchical nanostructures composed by small particles (size ≈10 nm), which form columnar treelike assemblies grown on a substrate.16-18 Such systems can be considered a peculiar class of high aspect ratio nanostructured materials, with unique properties comparable for some aspects to those of anisotropic columnar structures like nanorods and nanowires and for others to those of aerogels or nanoparticle assemblies with a low density (down to 0.3 g/cm3) and a large surface area, of the order of 100 m2/g. In this paper we study the reorganization effects caused by the interaction of these hierarchical nanostructures with different liquids. Over a critical thickness (usually around 1 μm) the asdeposited uniform layer of nanostructures undergoes profound deformations during the drying phase: channels open and columnar structures aggregate in domains (islands) with irregular shape and similar size. We explored the dependence of deformation features (i.e., channels and island size, characteristic length of the pattern) on structure height (i.e., layer thickness), compactness, and liquid properties (i.e., surface tension). The observed behavior is for (13) Li, Q. W.; DePaula, R.; Zhang, X. F.; Zheng, L. X.; Arendt, P. N.; Mueller, F. M.; Zhu, Y. T.; Tu, Y. Nanotechnology 2006, 17, 4533–4536. (14) Zhao, Y.-P.; Fan, J.-G. Appl. Phys. Lett. 2006, 88, 103123. (15) Chandra, D.; Yang, S. Langmuir 2009, 25, 10430–10434. (16) Di Fonzo, F.; Casari, C. S.; Russo, V.; Brunella, M. F.; Li Bassi, A.; Bottani, C. E. Nanotechnology 2009, 20, 015604(1-7). (17) Torta, F.; Fusi, M.; Casari, C. S.; Bottani, C. E.; Bachi, A. J. Proteome Res. 2009, 8, 1932–1942. (18) Sauvage, F.; Di Fonzo, F.; Li Bassi, A.; Casari, C. S.; Russo, V.; Divitini, G.; Ducati, C.; Bottani, C. E.; Comte, P.; Graetzel, M. Nano Lett. 2010, 10, 2562– 2567.
Published on Web 01/19/2011
DOI: 10.1021/la103955q
1935
Article
Fusi et al.
some features comparable to that of nanorods, nanowires, and soil dewetting, while for other aspects it presents unique features. The resulting surfaces exhibit a morphology which is intrinsically characterized by features over different length scales in the nanometer-micrometer range. The presented study helps identifying design and handling solutions for the development of layers for targeted applications, in particular those for which a multiscale structuring is of interest. Moreover, pulsed laser deposited TiO2 layers with hierarchical structuring are shown to be superhydrophilic without need for UV irradiation.
Experimental Section Nanostructured titanium oxide was grown by PLD on Si(100) at room temperature exploiting UV laser pulses (duration ≈7 ns) from a quadrupled Nd:YAG laser (λ = 266 nm, 10 Hz repetition rate) focused on a TiO2 target (purity 99.99%) with an energy density on the target (fluence) equal to ∼3 J/cm2, a target-tosubstrate distance of 50 mm, and a background O2 pressure of 20-40 Pa. Film thickness was varied from 200 nm to 10 μm by tuning the number of laser pulses; deposition rate was about 2-4 nm/s depending on deposition pressure. Postdeposition thermal treatments at 500 °C were performed for 1 h in a muffle furnace under an ambient atmosphere to induce growth of crystalline phase (mainly anatase). Films have been immersed in Petri dishes containing different liquids and let dry in controlled conditions under a chemical hood. The employed solvents are distilled water (Milli-Q), ethanol, hexane, and acetonitrile. All chemicals have been purchased from Sigma-Aldrich. Wetting tests have been performed immediately after deposition to minimize surface modifications due to exposure to ambient atmosphere. After complete evaporation of the liquid, samples have been checked with an optical microscope, and scanning electron microscopy (SEM) images were acquired with a Zeiss Supra 40 field emission SEM. SEM images have been processed with an analysis software developed in-house. In particular, supposing a statistically isotropic distribution of the reorganization features (the islands), the circular average of the 2D power spectral density (CA-PSD) of SEM images was evaluated; typical CA-PSD presents a peak at a certain frequency (fp), indicating the presence of a characteristic length (L= fp-1) in the pattern, which is related to the typical island size and separation. Moreover, since islands have a statistically uniform size over the whole surface, SEM micrographs were converted into black and white images, by simply setting a gray threshold, in order to identify connected domains and compute the distribution of area (A) and diameter (D = A0.5) of the islands. Analysis of characteristic length L, equivalent diameter D, and island eccentricity was always performed by averaging data from at least three images for each condition (thickness and liquid). A FEI QUANTA FEG 400 F7 environmental SEM (eSEM) has been employed to acquire images of the TiO2 films during exposure at increasing water vapor partial pressure, from 1 mbar up to 8 mbar, at a nearly constant substrate temperature of 3 °C. After occurrence of complete wetting of the film the subsequent drying was also observed by acquiring images while lowering the chamber pressure. Contact angles were measured with a Deltaphysics optical contact angle (OCA20) instrument.
Results and Discussion Nanostructured TiO2 Layer Deposition and Characterization. As previously demonstrated,16,19 oxide films of variable morphology at the nanometer-micrometer scale can be obtained by PLD exploiting plume confinement effects and lateral scattering (19) Bailini, A.; Di Fonzo, F.; Fusi, M.; Casari, C. S.; Li Bassi, A.; Russo, V.; Baserga, A.; Bottani, C. E. Appl. Surf. Sci. 2007, 253, 8130–8135.
1936 DOI: 10.1021/la103955q
Figure 1. SEM images of PLD TiO2 layers before (A, B) and after (C, D) wetting-dewetting with water (background deposition pressure 40 Pa O2).
of expanding species by ablation in a background gas. In the deposition conditions used it is possible to grow nanoporous TiO2 films composed by peculiar high aspect ratio structures, similar to trees and resulting from anisotropic assembling of nanosized particles; the as-deposited TiO2 nanostructures are amorphous/ disordered, but upon annealing they become mainly crystalline anatase with a typical grain size around 10 nm.16 Rutile content increases with deposition pressure being about 10% for 20 Pa and 15% for 40 Pa. Increasing the deposition pressure from 20 to 40 Pa results in a decrease of density (from about 0.5 g/cm3 to 0.3 g/ cm3) and in an increased aspect ratio of the columnar structures. Self-Organized Pattern Formation Induced by Liquid Wetting and Drying. Nanostructured TiO2 layers (deposition pressure 40 Pa) with a thickness above a critical value of ∼1 μm undergo irreversible deformation upon immersion in different liquids and subsequent drying: when the liquid is completely evaporated, a pattern of micrometer-sized randomly shaped islands, separated by deep channels, is formed (as shown for water in Figure 1). The resulting surface is then stable with respect to further wetting-dewetting cycles. Similar cellular networks have been reported in the literature for a wide range of systems: ceramic glazes,20 clay soils,21 slurries of nanoparticles suspended in water,22 nanowires and nanotubes (in this case the reorganization is sometimes called nanocarpet effect).6,14 In most cases these patterns are formed in materials that contract during cooling or drying. In the case of drying-induced phenomena, such as the one here described, the pattern formation is generally caused by capillary forces that arise during liquid evaporation from the material and induce a tensile stress. In brittle materials or in systems with poor adhesion to the substrate, when stress exceeds a certain value, it is released by cracks formation. Cracks then propagate until meeting other cracks or until stress is reduced below a critical threshold. Instead, for well-adhered elastic materials, such as nanorods or nanowires, cellular networks may appear even in the absence of fractures formation. In this case capillary forces cause lateral deformation of the material and aggregation of nanowire bundles; if, after complete evaporation, there is a balance between adhesion and elastic energy, the (20) Bohn, S.; Platkiewicz, J.; Andreotti, B.; Adda-Bedia, M.; Couder, Y. Phys. Rev. E 2005, 71, 046215(1-7). (21) Colina, H.; Roux, S. Eur. Phys. J. E 2000, 1, 189–194. (22) Shorlin, K. A.; de Bruyn, J. R.; Graham, M.; Morris, S. W. Phys. Rev. E 2000, 61, 6950–6957.
Langmuir 2011, 27(5), 1935–1941
Fusi et al.
Article
10 μm) at increasing water pressure: (A) 1.00, (B) 7.40, (C) 7.50, and (D) 7.60 mbar.
Figure 3. eSEM images of a TiO2 PLD film (40 Pa O2, thickness 10 μm) as (A) water floods the surface at 7.70 mbar, and as water pressure is reduced after complete wetting at pressure >7.80 mbar: (B) 7.40, (C) 7.30, and (D) 7.20 mbar.
nanostructures are in a stable bended configuration (see also discussion below). In order to better understand the dynamics of pattern formation in PLD TiO2 columnar layers, we followed in real time, by means of an optical microscope, the morphological evolution of the system during drying (see also movie “la103955q_si_001. mpg” in the Supporting Information, for the case of water). Cracks appear when the material starts to emerge from the liquid layer (at the evaporation front) and seem to nucleate at surface pre-existing inhomogeneities; this step is very fast and takes place on a 1 s time scale. While evaporation proceeds, cracks/channels propagation/broadening and shrinkage of the material occur; at the same time secondary cracks/channels are formed. The whole process occurs in a few seconds. Further information about the process dynamics can be gained by inspection of SEM images. Figure 1C,D clearly indicates that some TiO2 columnar structures are broken at their base and that columns at the edges of the aggregates are slightly bended toward the center of the aggregate, while some material is left inside the cracks, on the substrate (Figure 1C). The spaces (cracks or channels) between islands are as deep as the layer thickness, and the bare substrate is exposed in these regions. Following a similar approach, environmental SEM images have been acquired to observe in situ the evolution of the surface morphology during wetting by water and subsequent drying. The experiment was conducted by introducing an increasing water vapor pressure in the measurement chamber until liquid condensation occurs on the film surface. In Figure 2 a first set of images are shown at selected pressures in the range 1-7.60 mbar, while the substrate was maintained at a constant temperature (3 °C). Small cracks (inhomogeneities) are already present in the pristine surface, i.e., at the lowest investigated water pressure (1 mbar, Figure 2A). At a pressure of about 7.40 mbar, i.e., upon approaching the saturated vapor pressure for water (which is about 7.60 mbar at the measurement temperature, even though the exact local temperature is unknown), a minor restructuring occurs by starting from surface inhomogeneities (Figure 2B,C) and connected channels begin to appear in the film. The channel width shows an increasing trend with water vapor pressure (Figure 2B-D) but remains still quite small (nearly
1 μm) even at the highest investigated pressure (Figure 2D), and no complete formation of separated aggregates (islands) is observed (as instead seen in the drying phase, see below). This reorganization is irreversible in the sense that a subsequent decrease of the pressure at each exposure step does not further modify the surface. When the saturated vapor pressure (about 7.60 mbar) is overcome, complete wetting of the layer occurs (Figure 3A). In Figure 3A the image is blurred, an indication that flooding of the surface by water has occurred; no droplet formation is observed even at a larger scale. After this, a subsequent reduction of the pressure leads to formation of the already described pattern with well-separated islands (Figure 3B-D), which is thus related to evaporation after flooding of the porous nanostructure. The mechanism suggested by comparing Figure 3A,B with Figure 3C,D, is that islands are formed by material shrinkage and columns displacement during drying of the liquid. The island profiles are determined by the largest cracks already present in the film surface; the cracks/channels broaden and propagate during liquid evaporation while the columns are packed forming separated aggregates. The minor evolution of the pre-existing inhomogeneities before complete wetting (Figure 2) can be also related to capillary forces acting after a local, nonuniform water condensation on the material nanostructured surface. These capillary forces may act at the base of nanostructured columns, or where liquid/vapor menisci are formed between the columns (either on the top surface or on the lateral sides), as due to a possible temperature nonuniformity, but are not strong enough to lead to a complete surface reorganization. According to these findings, the mechanism leading to the final pattern may be related to capillary forces associated with the liquid/vapor menisci at the free ends of the columnar structures (or at their base in case of liquid condensation from high vapor pressure on the cold substrate), which produce a torque acting on the nanostructured columns. Because of this torque, the material may fracture at his weakest point, i.e., in proximity of the column base. Columns, while maintaining an upright position, are then displaced by capillary forces and by convective flows of the evaporating liquid, causing material shrinkage and formation of aggregates.
Figure 2. eSEM images of a TiO2 PLD film (40 Pa O2, thickness
Langmuir 2011, 27(5), 1935–1941
DOI: 10.1021/la103955q
1937
Article
Fusi et al.
Figure 4. Circularly averaged 2D power spectral density (PSD, left) and island size D distribution (right, with a Gaussian fit) for a TiO2 layer (40 Pa O2, thickness 10 μm) after wetting/dewetting with water.
Figure 6. Mean aggregate eccentricity as a function of thickness for films deposited at 40 Pa O2 (dots, water; squares, ethanol).
Figure 5. (A) Scaling of characteristic length L with thickness for PLD films deposited at 40 Pa O2 (liquids: water and ethanol). (B, C) Scaling of characteristic length L (dots) and average aggregate diameter D (squares) with thickness (B, water; C, ethanol).
This mechanism presents features in common with bundle formation in high aspect ratio structures, since also in this case we have columnar structures which are bended during evaporation, 1938 DOI: 10.1021/la103955q
and with clay soils desiccation, since also our material fractures and shrinks. For these related phenomena the final pattern features present a dependence of on several parameters (i.e., structures dimension, adhesion to the substrate, properties of the wetting liquid, etc.). Analysis of Island Pattern Features and Discussion. We now show in more detail how the features of the obtained patterns depend on the TiO2 layer (i.e., height of the nanostructures) and the solvent properties. We define a few synthetic parameters to characterize the patterns, which show aggregates with a statistically uniform distribution and size over the whole surface: the characteristic length L (defined as the inverse of the peak frequency in the circularly averaged 2D power spectral density of SEM images; see Experimental Section) and the average island size D (the square root of the area A of connected domains; see examples of 2D power spectral density and island size distribution in the case of wetting with water in Figure 4). Results in Figure 5 show a dependence of L on layer thickness (T). L is larger for thicker layers; a power law (L = c þ βTR) accurately describes experimental data, and we obtained similar R values with two different liquids (R ∼ 1.7 ( 0.1: ethanol; R ∼ 1.9 ( 0.4: water), indicating that the scaling with thickness does not depend on the liquid employed. Values of the exponent R equal to 1.3 ( 0.3 (c = 0) and to 1.2 ( 0.1 (c = 0) have been reported respectively for scaling of the cracks width in experiments with TiO2 nanotubes6 and for scaling of L in experiments with Si nanowires,14 in the case of bending without fracturing. Besides experiments, some simple models have been proposed to describe the bundling phenomenon. For example, Zhu et al.6 presented a model for the maximum lateral deflection (δ) of nanotubes where the scaling law δ ∼ T3 is obtained starting from the expression of Langmuir 2011, 27(5), 1935–1941
Fusi et al.
Article
Figure 7. SEM images of patterns induced on films with increasing thickness (first row: ethanol 2, 4, 6, 10 μm; second row: water 2, 4, 6, and
10 μm).
the capillary force between two adjacent cylindrical structures with given elastic properties and dimensions;23 Zhao et al.,14 by energetically balancing mechanical bending and capillary interaction, estimated that the average diameter (D) of the bundles should follow the relation D ∼ T3/2; Journet et al.24 balanced the elastic energy of a bended nanowire and the capillary energy and estimated that the average bundle diameter scales according to D ∼ T2. For clay soils, instead, the scaling law is linear if fracture formation occurs due to a critical stress criterion while it follows a nonlinear law if it follows the Griffith criterion.22 It is evident that finding a model which precisely describes such phenomena is extremely difficult since it would require an accurate description of the type of structures, of the material properties, and of the forces involved at very small scales. Moreover, there is no general agreement on the parameters that are best suited to describe the surface patterns (i.e., L, D, or crack width). Anyhow, the R exponent that we estimated (referred to the scaling of L) differs from values reported in the literature for similar experiments pointing out that, despite some analogies, the peculiar features of our nanostructures lead to a unique restructuring mechanism. Noteworthy we observe that the mean aggregates diameter D and the characteristic length L follow a different scaling behavior with the thickness T (Figure 5). A deeper analysis reveals that this behavior can be ascribed to differences in the shape of the aggregates, and thus a direct comparison of the scaling of D and L over the whole range of thicknesses here considered is not appropriate. In order to describe the island shape, we made use of the mean eccentricity (similar results are obtained using the mean perimeter/ diameter ratio). Eccentricity is evaluated as the eccentricity of an ellipse having the same second moment of each connected domain. These two parameters have higher values when the shape of the aggregates is more irregular and less compact. In Figure 6 we plot the eccentricity values as a function of the layer thickness: eccentricity decreases with increasing thickness and, in both ethanol and water, the decrease is not progressive but marks a significant drop between 4 and 6 μm thickness. We can thus figure out two different regimes: below a certain thickness islands have an elongated/ irregular shape and are not completely separated; this is probably the result of primary cracks starting at surface inhomogeneities. Above this thickness islands have more compact and separated profiles, as it can be qualitatively seen by direct inspection of SEM images (Figure 7). This is probably related to the fact that that the (23) Kralchevsky, P. A.; Paunov, V. N.; Denkov, N. D.; Ivanov, I. B.; Nagayama, K. J. Colloid Interface Sci. 1993, 155, 420–437. (24) Journet, C.; Moulinet, S.; Ybert, C.; Purcell, S. T.; Bocquet, L. Europhys. Lett. 2005, 71, 104–109.
Langmuir 2011, 27(5), 1935–1941
Figure 8. Trend of characteristic length L as a function of surface tension for the different liquids investigated: hex = hexane, EtOH = ethanol, ACN = acetonitrile, H2O = water (film thickness 10 μm).
total force or torque acting on the columnar structures increases for increasing structure height, as predicted by the above-cited models, thus inducing more pronounced reorganization effects. We then investigated the role of liquid properties in determining features size after evaporation. First of all, we observe that hexane does not induce appreciable modifications for layers with a thickness lower than about 5 μm, and no complete formation of well-separated bundles is observed even for the maximum film thickness (10 μm, not shown). Figure 8 shows values of L for layers of equal thickness (10 μm) after wetting with liquids of different surface tension. To the best of our knowledge, only a limited number of studies involving different liquids and comparing their effects have been performed, mainly on high aspect ratio nanostructures.6,11 These studies generally claim that, together with the structures height and elastic properties (i.e., Young’s modulus), the solvent parameters influencing the final pattern are the surface tension of the liquid (γ) and the equilibrium contact angle of the liquid on the surface of the material (θ). Such dependences arise by deriving the capillary force from the equations that describe the shape of the capillary meniscus between adjacent columnar structures.23 According to these studies, the final pattern features are related to the liquid properties according to ∼γa cosb(θ), where 0.5 e a e 1 and 1 e b e 2 depending on the different approximations used to describe the system.6,14 We observe a reasonable agreement for the trend of L as a function of γ for the different liquids investigated (Figure 8); the trend for D is similar. Actually, acetonitrile represents an anomaly DOI: 10.1021/la103955q
1939
Article
Fusi et al. Table 1. Electric Dipole Moment, Surface Tension, and Vapor Pressure for the Different Liquids Employeda acetonitrile (C2H3N)
water (H2O)
electric dipole moment (D) 3.92 1.87 29.3 72.8 surface tension at 20 °C (10-3 N/m) vapor pressure at 20 °C (mbar) 96 23 a Data are from Sigma-Aldrich product data sheets or from tabulated literature data; see e.g. ref 25.
with respect to the L trend. Inclusion of the cos2(θ) or cos(θ) dependence does not significantly modify the trends. In fact, contact angle values before island formation are below 10° for all liquids. Since the macroscopic contact angle is severely affected by the surface morphology, we also evaluated the wettability of an extremely flat TiO2 surface (root-mean-square roughness 1 μm) are wetted for the first time with a water droplet, they are already strongly hydrophilic, with contact angles below 10°. As already discussed, following water evaporation, surface restructuring and patterning occur; the resulting patterned surfaces are stable toward further wetting-dewetting cycles and are extremely hydrophilic, with a contact angle for water close to zero. In particular, in the case 1940 DOI: 10.1021/la103955q
ethanol (C2H6O)
hexane (C6H14)
1.69 22.7 59
0.08 18.4 173
Figure 9. SEM images of patterns induced by ethanol and water on films with different porosity (deposition pressure: first column 40 Pa, second column 20 Pa O2; film thickness = 4 μm).
of a 10 μm patterned thick film (40 Pa), the contact angle is about 7° at initial time (
