Langmuir 2008, 24, 13591-13597
13591
Superhydrophobicity of Biological and Technical Surfaces under Moisture Condensation: Stability in Relation to Surface Structure Bernd Mockenhaupt,†,‡ Hans-Ju¨rgen Ensikat,*,† Manuel Spaeth,† and Wilhelm Barthlott† Nees Institute for BiodiVersity of Plants, UniVersity of Bonn, Meckenheimer Allee 170, D-53115 Bonn, Germany, and Federal Institute of Hydrology, Department of Animal Ecology, Am Mainzer Tor 1, D-56068 Koblenz, Germany ReceiVed July 22, 2008. ReVised Manuscript ReceiVed September 18, 2008 The stability of superhydrophobic properties of eight plants and four technical surfaces in respect to water condensation has been compared. Contact and sliding angles were measured after application of water drops of ambient temperature (20 °C) onto cooled surfaces. Water evaporating from the drops condensed, due to the temperature difference between the drops and the surface, on the cooled samples, forming “satellite droplets” in the vicinity of the drops. Surface cooling to 15, 10, and 5 °C showed a gradual decrease of superhydrophobicity. The decrease was dependent on the specific surface architecture of the sample. The least decrease was found on hierarchically structured surfaces with a combination of a coarse microstructure and submicrometer-sized structures, similar to that of the Lotus leaf. Control experiments with glycerol droplets, which show no evaporation, and thus no condensation, were carried out to verify that the effects with water were caused by condensation from the drop (secondary condensation). Furthermore, the superhydrophobic properties after condensation on cooled surfaces from a humid environment for 10 min were examined. After this period, the surfaces were covered with spherical water droplets, but most samples retained their superhydrophobicity. Again, the best stability of the water-repellent properties was found on hierarchically structured surfaces similar to that of the Lotus leaf.
1. Introduction During the last years, a variety of materials with superhydrophobic surfaces have been developed biomimicking plant surfaces such as the Lotus (Nelumbo nucifera) leaf.1 A serious problem for technical applications is the loss of superhydrophobicity caused by enduring contact with water or by condensation of moisture. The durability of superhydrophobicity of plant surfaces differs significantly, depending on their morphology and chemistry. Many plants lose their water-repellent properties temporarily after a longer contact with water, for example, during long lasting rainfalls. Some other plants, such as Lotus (Nelumbo nucifera) or some Eucalyptus species, retain their superhydrophobicity even when they are immersed under water for days.2 Publications about effects of water condensation on superhydrophobic surfaces are controversial. Several authors studied condensation on microstructured surfaces and found a strongly increased wettability or pinning of the growing droplets.3-5 Examinations of several nanostructured and hierarchically structured surfaces showed a constant water repellency under condensation conditions.6,7 Few data about condensation effects on superhydrophobic plant surfaces are published. Cheng and Rodak8 studied the behavior of Lotus leaves after condensation of hot water vapor * To whom correspondence should be addressed. E-mail: ensikat@ uni-bonn.de. † Nees Institute for Biodiversity of Plants, University of Bonn. ‡ Federal Institute of Hydrology.
(1) Barthlott, W.; Neinhuis, C. Planta 1997, 202, 1–8. (2) Martin, J. T.; Juniper, B. E. The cuticles of plants; Edward Arnold: London, 1970. (3) Narhe, R. D.; Beysens, D. A. Phys. ReV. Lett. 2004, 93, 1–4. (4) Wier, K. A.; McCarthy, T. J. Langmuir 2006, 22, 2433–2436. (5) Jung, Y. C.; Bhushan, B. J. Microsc. 2008, 229, 127–140. (6) Chen, Y. C.; He, B.; Lee, J.; Patankar, N. A. J. Colloid Interface Sci. 2005, 281, 458–464. (7) Lau, K. K.; 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. (8) Cheng, Y.-T.; Rodak, D. E. Appl. Phys. Lett. 2005, 86, 1–3.
(boiling water) and found a strongly increased wettability. In situ observations of water condensation on Lotus leaves9,10 in an environmental scanning electron microscope (ESEM) showed a partial wettability of the surface by the growing microdroplets under these special conditions (vacuum, electron beam irradiation). Wu et al.11 presented ESEM images of nonwetting condensing droplets on several plant surfaces covered with different epicuticular waxes. The present study resolves the correlation between the influence of condensation on the superhydrophobicity of various surfaces and different degrees of hierarchical roughness. It will help to answer whether condensation generally causes an increase of wettability or even a loss of superhydrophobicity, or whether certain structures are resistant and retain their extreme water repellency.
2. Materials and Methods Material. Eight plant species with superhydrophobic leaf surfaces and four technical surfaces were selected for this study. The plants were cultivated in the Botanical Gardens, Bonn (BG BONN): Alocasia macrorrhiza (L.) G. Don. (Giant Elephant Ear); Apocynum cannabinum L., (Indian Hemp); Argemone mexicana L., (Mexican Poppy); Brassica oleracea var. gongylodes L. (Kohlrabi); Colocasia esculenta (L.) Schott. (Taro); Nelumbo nucifera Gaertn. (Lotus); Tropaeolum majus L. (Nasturtium); Xanthosoma robustum Schott. (Elephant Ear). The upper side of the leaves was examined, except for Alocasia and Xanthosoma, where the lower side was used. The technical samples were two structured copper foils which were hydrophobized, and two superhydrophobization sprays applied on smooth glass surfaces. “Bolta-foil” (Bolta-Werke GmbH Gottmadingen, Germany) and “Circuit-foil” (Circuit Foil Luxembourg) (9) Zheng, Y.; Han, D.; Zhai, J.; Jiang, L. Appl. Phys. Lett. 2008, 92, 1–3. (10) Cheng, Y.-T.; Rodak, D. E.; Angelopoulus, A.; Gacek, T. Appl. Phys. Lett. 2005, 87, 1–3. (11) Wu, Y.; Saito, N.; Nae, F. A.; Inoue, Y.; Takai, O. Surf. Sci. 2006, 600, 3710–3714.
10.1021/la802351h CCC: $40.75 2008 American Chemical Society Published on Web 10/30/2008
13592 Langmuir, Vol. 24, No. 23, 2008
Mockenhaupt et al.
Figure 1. SEM micrographs of the tested plant surfaces, in the order of increasing stability of superhydrophobicity, equal to Figure 3. Only Alocasia (a) is not covered with wax crystals; the other surfaces are covered with wax crystals of different size and shape. Bar ) 10 µm. The insets show the wax crystals in detail. Bar ) 500 nm.
are galvanically roughened copper foils, which were hydrophobized by the epilamisator antispread F 2/50 FK60 (Dr. Tillwich GmbH, 72160 Horb, Germany). “Creavis-spray” (Creavis Innovations GmbH, Marl, Germany) and “Mincor-spray” (BASF Future Business GmbH, Ludwigshafen, Germany) formed “nanoparticulated” superhydrophobic coatings. They were applied on glass slides in compliance with the instruction of use. Methods. Sample Temperature Regulation. The samples were mounted on a Peltier element (Conrad Electronic, Hirschau, Germany), which was attached to the contact angle measurement system. The surface temperature was measured with a copperconstantan thermocouple made of 0.1 mm wires. At an ambient air temperature of 20 °C, the samples were cooled to a surface temperature of either 15, 10, or 5 °C. For control measurements, the samples were slightly warmed (up to 35 °C) or measured at 20 °C. Contact and Sliding Angle Measurement on Cooled Surfaces. A contact angle measurement system (OCA 30-2, Dataphysics Instruments GmbH, Filderstadt, Germany) equipped with a tilting base device was used. After stabilization of the desired surface temperature of the sample, a drop of 15 µL of water with ambient temperature (20 °C) was applied onto the surface. The contact angles were measured 20 s after application of the water drop, when the temperature of the drop had approximated the surface temperature, and the sliding angle measurement started after the same time period. An average of 10 readings was used for the contact and sliding angles. Every drop was placed on another area of the surface.
For the contact angle measurements, the water drop remained in contact with the application capillary in order to avoid the drop moving across the superhydrophobic surface. Control measurements were made with 100% glycerol. Condensation Experiment from a Humid Atmosphere. A closed chamber was filled up to one-quarter of its capacity with 25 °C warm water, which led to an atmosphere with 100% relative air humidity. Above the water surface, the samples were mounted on a cooled metal block of 5 °C. The condensation process from the humid atmosphere onto the sample surfaces was observed with a binocular. After 10 min, when the surfaces were densely covered with condensed water droplets, the samples were immersed in water and it was determined whether the samples still retained the characteristic air layer under water and whether the surfaces were wetted partially, completely, or appeared dry. Scanning Electron Microscopy (SEM). The plant material was prepared by “liquid substitution” with glycerol12 which avoids shrinkage artifacts and alteration of the epicuticular waxes. All SEM samples were sputter-coated with 20 nm of gold. SEM micrographs were made with a Stereoscan S 200 instrument (Cambridge, Cambridge, U.K.) or with a LEO 1450 scanning electron microscope (Carl Zeiss SMT, Oberkochen, Germany). (12) Ensikat, H. J.; Barthlott, W. J. Microsc. 1993, 172, 195–203.
Biological, Technical Surface Superhydrophobicity
Langmuir, Vol. 24, No. 23, 2008 13593
Figure 2. Two examples (Brassica oleracea and Colocasia esculenta) showing the decrease of hydrophobicity with increased surface cooling, which results in increased sliding angles (a,c) as well as decreased contact angles (b,d). Brassica (a,b) shows a much stronger alteration than Colocasia (c,d). Moderate warming had no significant effect.
Figure 3. Sliding angles and contact angles on different plant surfaces measured at 20 °C (striped) and 5 °C (black). The species are arranged in the order of increasing stability with regard to the sliding angles (compare SEM images, Figure 1). Sliding angles of 90° denote that drops did not roll off even at 90° tilt.
3. Results Morphology of the Plant Surfaces. All selected plant species are superhydrophobic with contact angles above 150° but vary in their surface morphologies referring to their micro- and nanostructure (Figure 1). The surface of Alocasia macrorrhiza shows papillose cells with cuticular foldings but without epicuticular wax crystals. Apocynum cannabinum has tabular cells covered with small wax platelets. The flat cells of Brassica oleracea are covered with long and fragile branched wax rodlets. Argemone mexicana and Tropaeolum majus have convex cells which are covered with wax tubules. The surface structure of Xanthosoma robustum is, similar to that of Colocasia esculenta, formed by papillose cells covered with wax platelets. Nelumbo nucifera also shows papillae, but they are covered with small wax tubules.
Effects of Temperature Differences between the Surface and Drop on the Contact and Sliding Angles on Superhydrophobic Plant Surfaces. With the exception of Nelumbo, all samples showed an increased wettability on cooled surfaces, while a moderate warming had no significant effect. The characteristic behavior is shown in Figure 2 for two examples (Brassica and Colocasia). The changes of sliding and contact angles became apparent even at slight cooling to ∼15 °C, and with increased cooling the changes became more distinctive. However, the changes indicate a gradual decrease of hydrophobicity, but not an abrupt change to a hydrophilic state. Comparison of Cooling Effects on Different Plant Surfaces. The decrease of hydrophobicity due to secondary condensation was different for all specimens (Figure 3). Only on Nelumbo nucifera (Lotus) leaves, no significant influence could be detected.
13594 Langmuir, Vol. 24, No. 23, 2008
The alterations are best displayed by the sliding angles. The contact angles show the same tendency, but the angle differences are smaller and the precision of contact angle measurements on rough surfaces may be lower. When comparing the change in contact and sliding angles of the different samples (Figure 3) with respect to their surface morphology (Figure 1), a clear correlation can be found. Figure 1 shows SEM images of the leaf surfaces in the same order as that in Figure 3, with increasing stability of hydrophobicity. The lowest stability was found on surfaces with no or with very small three-dimensional wax crystals (Alocasia, Apocynum, Figure 1a,b), although the cells are papillose or convex. A higher stability of hydrophobicity was observed on surfaces with a thicker layer of wax crystals, even on flat, tabular cells (Figure 1c-e). By far the highest stability is found on hierarchically structured surfaces formed by papillae, which are covered with wax crystals (Figure 1f-h). The morphology of wax crystals is not a primary factor, but it has a gradual influence on the stability. Nelumbo, Argemone, and Tropaeolum bear wax tubules, which are known to cause strong hydrophobicity; other very stable hydrophobic surfaces are covered with wax platelets (Colocasia, Xanthosoma). Certain waxes, such as rodlets (Brassica), are very fragile and can be damaged easily. Does Condensation of “Satellite Droplets” Cause the Decrease of Hydrophobicity? The sliding angles increased only when the sample surface was colder than the drop. Under isothermal conditions, when the surface and drop were cooled to the same temperature, the sliding angles did not change significantly. In a light microscope, it could be seen that, immediately after application of the water drop onto a cooled surface, satellite-droplets occur around the primary drop (Figure 4a). These satellite-droplets grow for a while until the temperature of the primary drop reaches the sample surface temperature. They then evaporate slowly, if the environmental air humidity is low enough. The observation of these “satellite droplets” in the vicinity of the applied water drop (“secondary condensation”) led to the assumption that condensation near the primary drop and in the air-filled spaces below the drop could be the main cause for the reduction in hydrophobicity on cooled surfaces (Figure 4b,c). In order to test this hypothesis, the effect of surface cooling on the contact and sliding angles of glycerol, which shows no evaporation and condensation activities due to its low vapor pressure, was tested. On all plant surfaces, cooling had no significant effect on the contact and sliding angles of glycerol. (Measurements were performed at low air humidity, because cooling in high air humidity may cause water condensation, which disturbs the measurements with glycerol.) Effects of Continuing Condensation from a Humid Atmosphere. The observation of condensation for at least 10 min on superhydrophobic plant surfaces at 5 °C showed small spherical droplets after ∼2 min. They grew continuously and merged with proximate drops, dashing across the surface. The growing droplets retained their spherical shape with a high contact angle (Figure 5). During immersion in water, the surfaces retained an air layer, and after lifting out of the water all superhydrophobic plant surfaces except Alocasia appeared dry. Only the leaf of Alocasia, which is not covered with wax crystals, exhibited a reduced air layer under water and a partially wetted surface, particularly at leaf veins. However, our tests with several other hydrophobic plant surfaces showed that condensation increases the wettability gradually.
Mockenhaupt et al.
Figure 4. (a) Top view of the appearance of condensed satellite-droplets (SD) in the vicinity of an applied water drop (temperature, 20 °C; diameter, 3 mm) on a cooled superhydrophobic surface immediately after drop application. (b,c) Images of a water drop (diameter, 3 mm) on a Brassica leaf. (b) On the cooled leaf (5 °C), 20 s after application of the water drop of 20 °C, the contact angle decreases and small condensed “satellite droplets” (SD) are visible. (c) On a leaf with ambient temperature (20 °C), no condensation occurs and the contact angle remains high.
Effects of Temperature Differences between the Surface and Drop on the Contact and Sliding Angles on Different Superhydrophobic Technical Surfaces. The four samples selected were superhydrophobic, with contact angles of 160-166°. The surface morphologies differ (Figure 6). “Bolta-foil” has a distinct hierarchical structure with coarse grains which are covered with particles of less than 100 nm size. “Circuit-foil” has mainly a coarse structure with grains of several micrometers in size. “Creavis-spray” and “Mincor-spray” form a nanostructured surface with a grain size of
