ARTICLE pubs.acs.org/Langmuir
Three-Dimensional Hierarchical Structures for Fog Harvesting H. G. Andrews, E. A. Eccles, W. C. E. Schofield, and J. P. S. Badyal* Department of Chemistry, Science Laboratories, Durham University, Durham DH1 3LE, U.K. ABSTRACT:
Conventional fog-harvesting mechanisms are effectively pseudo-2D surface phenomena in terms of water droplet-plant interactions. In the case of the Cotula fallax plant, a unique hierarchical 3D arrangement formed by its leaves and the fine hairs covering them has been found to underpin the collection and retention of water droplets on the foliage for extended periods of time. The mechanisms of water capture and release as a function of the surface 3D structure and chemistry have been identified. Of particular note is that water is retained throughout the entirety of the plant and held within the foliage itself (rather than in localized regions). Individual plant hairs form matlike structures capable of supporting water droplets; these hairs wrap around water droplets in a 3D fashion to secure them via a fine nanoscale groove structure that prevents them from easily falling to the ground.
1. INTRODUCTION Approximately one billion people live without access to clean water sources in rural areas of African, Asian, and Latin American countries.1 Therefore, the issue of water shortage and scarcity is one of major global concern. In contrast, indigenous plants found in such arid and semiarid locations readily cope with insufficient access to fresh water and lack of precipitation.2 Fog episodes occur frequently in many of these regions and help to augment water supplies for native botanic species through dew and fog collection as well as water vapor absorption.2 Inspired by nature, there are now extensive efforts being made to utilize fog interception to provide clean water to human settlements in these regions.3-6 Over the last 25 years, several projects have been introduced in rural communities situated in arid climates that use fog interception as the basis for their fresh water supply.3 Schemes of this nature have been successfully implemented in countries including South Africa,4 Namibia,5 and Chile.6 These are based upon two main types of passive fog-collecting devices: the standard fog (SFC) collector7 and the Atmospheric Sciences Research Center (ASRC) collector.8 They rely upon fogwater droplet deposition onto nylon mesh7 and Teflon fibers,8 respectively. These materials are known to resist complete wetting by water, instead allowing condensation to occur in the form of droplets on their surfaces. After initial deposition, the droplets are able to grow in size and merge with others until eventually they become too large to be supported by the surface and fall into a collection container below. Biomimetic replication of natural fog harvesting systems is becoming a topic of interest to the scientific community with the aim being to help maximize the fog-collection efficiency. Of r 2011 American Chemical Society
particular note has been the Stenocara sp. beetle that is able to harvest water on its back through the presence of hydrophilic spots on a hydrophobic background,9 a design principle that has been successfully replicated.9,10 Documented fog-harvesting mechanisms for plant families are also reported as being effectively two-dimensional in terms of water droplet-plant interactions where water retention occurs in localized regions.11-15 Fogcollection devices based on such pseudo-2D structures ultimately suffer from an inefficient interaction with the 3D fog medium because of their dependency upon the direction of fog-bearing winds. This has prompted the evaluation of naturally occurring 3D species (such as plants) for efficient fog-collection behavior. Indeed, there are a number of documented plant leaves onto which water is able to condense.11-13 However, their surface wettabilities tend toward either hydrophilic, leading to the formation of thin water films covering the exterior of the plant,16 or hydrophobic, where beads of water usually remain until they evaporate.14 The extreme in the latter case is where water droplets form on superhydrophobic leaves (the lotus leaf effect) and eventually roll off toward the ground in order to be absorbed by plant roots.17 In all of these cases, although the inherent plant structure itself is 3D, the water droplet capture mechanism remains effectively pseudo-2D (i.e., capture occurs on one side of the leaf surface).
Received: January 1, 2011 Revised: February 10, 2011 Published: March 07, 2011 3798
dx.doi.org/10.1021/la2000014 | Langmuir 2011, 27, 3798–3802
Langmuir
ARTICLE
Extensive screening studies have identified the Cotula fallax,; it is a tufted, cushion-forming alpine plant that is believed to be native to South Africa.18 For many years, it has been mistakenly identified as both Cotula lineariloba and Cotula hispida and is still often available under these names. This hardy plant thrives in temperate climates and can be grown throughout the world,18 thus making it a suitable candidate for widespread cultivation (as well as a template for man-made biomimetic fog-collection structures).
2. EXPERIMENTAL SECTION Preparation of Cotula fallax Samples. Cotula fallax plant cuttings were collected from outdoor beds, and any visible water droplets were allowed to evaporate in the laboratory prior to trimming down to approximately equal sizes. Scanning Electron Microscopy (SEM). Individual leaf specimens were prepared for scanning electron microscopy analysis by fixing overnight using 2% gluteraldehyde in phosphate buffer solution (Sigma, pH 7.4). The leaves were then rinsed twice in phosphate buffer solution before undergoing dehydration by progressing through a graded series of ethanol solutions. The drying process was completed using a criticalpoint dryer (Samdri 780). Each sample was then mounted onto a carbon disk supported on an aluminum stub and finally coated with a 15 nm layer of gold (Polaron SEM coating unit). Plant surface structure images were taken on a scanning electron microscope (Cambridge Stereoscan 240). Environmental Scanning Electron Microscopy (ESEM). Environmental scanning electron microscopy (FEI XL30 ESEM-FEG) operating in wet mode was used to observe water vapor condensation onto individual leaf samples in real time. The leaves were mounted onto a carbon disk and attached to a Peltier cooling stage. Vapor condensation was initiated by controlling the chamber pressure and the sample temperature.19 Micrographs were taken using a gaseous secondary electron detector (GSED) in conjunction with a 25 kV accelerating voltage and a working distance varying between 9 and 11 mm. Microcondensation Experiments. Water-collection measurements were performed by exposing plant samples to a fine mist of ultrahigh-purity water (ISO 3696 grade 1) generated by a nebulizer operating with an air flow of 11 L min-1. Individual specimens were mounted on top of a mass balance, and the amount of water collected was measured at 15 min intervals over a period of 2 h. Photographs of the plant were taken throughout this duration. Dynamic Contact Angle (DCA) Analysis. Dynamic contact angle analysis (Cahn DCA-322) of single hair fibers taken from the Cotula fallax leaves was performed by mounting them onto tape with sufficient overhang to allow positioning perpendicular to the surface of the test liquid (ultra-high-purity water (ISO 3696 grade 1)). The wettability of the individual hair samples was determined by allowing the hair to advance 0.5 mm beyond the point of contact with the test liquid surface using a stage speed set at 40 μm s-1. The hair fiber did not realign during approach toward the liquid surface. For wicking experiments, the software was programmed to detect the point of fiber impact with the water and then maintain that position for 5 min while recording the mass of the sample at approximately 0.2 s intervals. Air Plasma Treatment of Cotula fallax. Air plasma treatments of Cotula fallax specimens were carried out in a cylindrical glass reactor (5 cm diameter, 460 cm3 volume) surrounded by a copper coil (4 mm diameter, 9 turns, located 10 cm away from the gas inlet) enclosed within a Faraday cage.20 The vessel was evacuated by a 30 L min-1 rotary pump connected via a liquid-nitrogen cold trap (base pressure of less than 2 10-3 mbar and leak rate better than 6 10-9 mol s-1). All joints were grease-free. The output impedance of a 13.56 MHz radio frequency (rf) power generator was matched to the partially ionized gas load through
Figure 1. Images of Cotula fallax: (a) dry plant, (b) water collected by a dry plant, (c) water collected by a plant later in the day (scale bar = 20 mm), and (d) schematic representation of water collection by Cotula fallax over the course of a day, showing an interior matlike structure formed by overlapping hairs. an L-C matching unit. Individual plant samples were placed into the chamber and pumped down. Air was then fed into the system via a fine control needle valve at a pressure of 0.2 mbar, and the electrical discharge ignited at 40 W for a total exposure period of 30 s.
3. RESULTS Cotula fallax plants are densely sericeous with a biternate leaf arrangement (Figure 1a). Their water-harvesting behavior is easily visible to the naked eye throughout the day (Figure 1b, c). Small beads of moisture collect on the tips of individual leaves and increase in size. When these droplets become too large in size to balance on the leaf apex, they detach into the main body of the plant to coalesce with other fallen droplets. The newly formed, larger droplets are retained within the Cotula fallax plant structure via a combination of the leaves and hairs (Figure 1d). Eventually they become too large to be supported within the framework of the plant and fall to the underlying soil. Smaller droplets may also become detached because of the destabilization effects in the presence of wind. A key feature is the role played by the fine hairs that help to maximize the amount of water capture. Scanning electron microscopy (SEM) micrographs of the Cotula fallax leaves reveal their complex hierarchical structure (Figure 2a-f). Individual leaves measure approximately 4 mm in length and 1 mm in width. Each leaf is completely covered with hairs with an approximate diameter of 5 μm and two distinct lengths (long hairs of 2-5 mm length and short hairs measuring less than 500 μm, Figure 2a). This bimodal length distribution does not appear to change between seasons. Both types possess a fine nanoscale structure of well-defined ridges that run along the length of individual hairs, stretching from close to the tip to just short of the hair root (Figure 2b-f). These longitudinal grooves are present around the entire circumference of each hair. The shorter hairs lie adpressed to the leaf surface whereas the longer hairs grow freely outward, extending in all directions. Because of the proximity of individual leaves on the plant itself, the longer hairs from adjacent leaves are able to overlap and intertwine to form a matlike structure that effectively provides a netting to support water droplets (akin to the Cassie-Baxter21 effect). The behavior of droplets condensed on individual leaf surfaces was explored in real time using wet-mode environmental scanning electron microscopy (ESEM, Figure 2g-j). Microscale droplets are seen to condense initially over the entirety of the leaf surface and display a hydrophobic wetting interaction with 3799
dx.doi.org/10.1021/la2000014 |Langmuir 2011, 27, 3798–3802
Langmuir
ARTICLE
Figure 3. SEM micrographs of Cotula fallax leaf hairs (a) untreated and (b) following air plasma treatment.
Figure 2. SEM micrographs of Cotula fallax: (a) a single leaf covered in hairs, (b) a single leaf tip covered in hairs, (c) individual hairs, (d) a single hair tip at high resolution, (e) midsection of a single hair at high resolution, and (f) a hair root at the leaf surface. Wet-mode ESEM micrographs of condensed water droplets on Cotula fallax: (g) a single leaf tip covered in hairs with droplets, (h) midsection of a single leaf covered in hairs with droplets, (i) growth of a large droplet condensed in between hairs, and (j) the same droplet 2 min later.
individual hairs. Neighboring droplets eventually merge with each other to form larger droplets. At the leaf edges, these droplets soon become too large to be supported by a single hair and hence become detached, (Figure 2g). However, in the midsection of the leaf where there is a denser covering of hairs, droplets are able to grow in between the hairs that subsequently wrap around these larger droplets, effectively capturing them and preventing them from falling off (Figure 2i,j). It is of interest to note that as the droplets grow in size the capillary action occurring along the grooves of the individual hairs underpins the eventual wrapping of droplets in order to help accommodate the buildup of internal stresses within the plant structure due to the continued expansion of the water droplets (as exemplified by the single hair dynamic contact angle (DCA) measurements described later). To further understand this remarkable ability of Cotula fallax to collect water, its macroscopic behavior was followed in the vicinity of an artificially generated fine water mist. A comparison was made between two similarly sized specimens of which one had been exposed to a very short air plasma treatment (so as to perturb the plant surface chemistry slightly by increasing the surface energy) without causing any physical deterioration to the leaf or hair structure (as checked by SEM analysis, Figure 3). For
the untreated specimen, small water droplets located on the tips of the hairs are visible to the naked eye after 15 min and they continue to grow steadily (Figure 4a). In contrast, although small water droplets begin to form on the tips of the plasma-treated plant hairs initially (15 min), their size does not increase significantly with mist exposure time (Figure 4b). Furthermore, in the latter case the leaves close to the base of the sample become completely wet, which is indicative of water absorption and was confirmed by undertaking quantitative mass measurements (Figure 4c). There is very little difference among the amounts of water collected by each plant sample during the first 15 min of mist exposure. However, the air-plasma-treated Cotula fallax sample continues to collect water at a higher rate throughout the remainder of the experiment because of absorption. This illustrates how the surface chemistry also plays a key role in water collection. In the case of the plasma-treated specimen, although the mass of water collected is greater than that for the fresh plant, there is no way of retrieving this water back from the plant because it has become absorbed. In practice, it is advantageous for the plant not to become covered with a continuous water layer because it can lead to infection.16 Dynamic contact angle (DCA) analysis was performed on individual hair fibres in order to gain additional insight into their wetting behavior (Figure 5). It was found that a single untreated hair is subjected to a repulsive initial force, implying that its surface is hydrophobic, whereas a hydrophilic interaction is observed during the removal of the hair from bulk water (indicative of large hysteresis that is consistent with the built-in stress behavior observed in Figure 2i,j). The DCA analysis performed on individual plasma-treated hairs correlates well with the observations made in the aforementioned water mist collection experiments (Figure 4). The treated hairs exhibit an instantaneous hydrophilic interaction upon contact with the water surface, thereby providing confirmation that air plasma exposure imparts greater hydrophilicity (higher surface energy).
4. DISCUSSION It is well known that a number of plants16 possess pubescent leaves that display a composite Cassie-Baxter21 superhydrophobic surface through the trapping of tiny pockets of air at the liquid-solid interface.17,22 This minimization of contact between the droplet and the plant leaf lowers the adhesion to allow the droplet to roll off when the leaf is slightly inclined. Such behavior is often observed during rainfall, where the droplet diameters range from 0.5 to 10 mm,23 and they subsequently bounce toward the ground. However, for smaller droplet sizes that are comparable to the size of surface roughness features (for instance, during fog episodes where droplet size ranges from 1 to 40 μm9) air pockets are unable to form to provide a superhydrophobic state, and thus water is able to settle and spread across the plant surface.24 Such behavior has been reported for Alchemilla vulgaris (lady’s mantle)17 and Nelumbo nucifera (lotus 3800
dx.doi.org/10.1021/la2000014 |Langmuir 2011, 27, 3798–3802
Langmuir
ARTICLE
Figure 5. Dynamic contact angle profile of a single Cotula fallax hair being dipped in water: (a) untreated and (b) air-plasma-treated.
Figure 4. Water collection on Cotula fallax plant specimens over time: (a) fresh plant sample, (b) air-plasma-treated plant sample (scale bar = 10 mm), and (c) mass of water collected by each type of plant specimen. (Controls exposed to just vacuum confirmed that the observed difference is not just due to dehydration. A thermocouple placed into the electrical discharge indicated a negligible heating effect.)
leaf),24 where only larger droplets are able to lift away from the leaf surface as a consequence of composite Cassie-Baxter surface formation.25 Cotula fallax plants exhibit similar behavior toward rain droplets. However, they differ when exposed to fog because droplets are prevented from rolling off and falling to the ground by the presence of long entangled hair matlike structures that are capable of supporting the weight of large droplets and thus capturing them within the foliage itself (Figure 2). This ability of Cotula fallax plants to collect water is remarkable given that the individual hair and leaf components are significantly hydrophobic (as shown by ESEM and DCA analysis, Figures 2 and 5). The hairs themselves can also be considered to be in a Wenzel26 wetting state with accompanying contact line pinning27 attributable to the fine-scale roughness features detected by SEM. DCA analysis shows that the movement of the hair across the airwater interface (equating to the buildup of internal stress within the plant due to droplet growth) eventually leads to the liquid entering the ridgelike surface features running along the length of the hair. This corresponds to the wrapping of individual hairs around water droplets via capillary action along the grooves to relieve internal stress (Figure 2i,j). The parallel grooves prevent the wrapping of water around the hair fibers. Effectively, a transition between Cassie-Baxter and Wenzel states is occurring, which is already well documented for droplets deposited from a height or pushed into a surface and heavy droplets.28 Although it has been reported that plants with rosette structures consisting of long, narrow leaves are ideally suited to fog collection,12,13 it appears that Cotula fallax is able to maximize deposition by virtue of its unique yet compact hierarchical 3D structure combined with its inherently much greater surface area to volume ratio. Cotula fallax differs from other plants described in the literature because it does not withhold water in physical pockets13,15
or just directly channels water to the ground toward roots.11 It is worthwhile to note that there are a range of activities attributed to plant trichomes including the reflection of sunlight, absorption of water, and protection against predators and fungi.29 Therefore, it would be unreasonable to suggest that the fine hair matting structure of the Cotula fallax plant is solely present to collect water. Successful replication of this hierarchical 3D structure would represent a promising new approach for fog harvesting in geographical regions where there is a limited supply of fresh water (as well as commercial applications including filtration and smart textiles). The efficiency of such man-made 3D structures could be further fine tuned by plasmachemical surface functionalization to introduce smart behavior (e.g., antibacterial30 to help keep the collected water fresh) in combination with the adjustment of the longitudinal groove sizes. In addition, the widespread planting of Cotula fallax or similar species should be contemplated as an ecofriendly, environmentally sustainable approach to help tackle water shortages in arid climates and to counter-balance deforestation in other parts of the world.
5. CONCLUSIONS The fog-harvesting mechanism of the Cotula fallax plant can be attributed to the hierarchical 3D arrangement formed by its leaves and the fine hairs covering them, which aid the collection and retention of water droplets on the foliage for extended periods of time. Water is collected throughout the entirety of the plant and held within the foliage itself (rather than in localized regions as reported previously for other plant families). Individual plant hairs form matlike structures capable of supporting water droplets; these hairs are able to wrap around water droplets in a 3D fashion to secure them and prevent them from easily falling to the ground. This represents a previously unreported water-collection mechanism, and successful replication of the Cotula fallax plant structure would provide a promising new approach to fog harvesting in arid and semiarid regions of the world. ’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT We thank T. Davey of the Electron Microscopy Research Services at Newcastle University for the SEM imaging and P. 3801
dx.doi.org/10.1021/la2000014 |Langmuir 2011, 27, 3798–3802
Langmuir
ARTICLE
Carrick of the Chemical and Materials Analysis Services at Newcastle University for the ESEM imaging.
’ REFERENCES (1) Progress on Sanitation and Drinking Water: 2010 Update; World Health Organization: Geneva, 2010. (2) Agam, N.; Berliner, P. R. J. Arid Environ. 2006, 65, 572–590. (3) Schemenauer, R. S.; Cereceda, P. Ambio 1991, 20, 303–308. (4) Olivier, J.; de Rautenbach, C. J. Atmos. Res. 2002, 64, 227–238. (5) Shanyengana, E. S.; Henschel, J. R.; Seely, M. K.; Sanderson, R. D. Atmos. Res. 2002, 64, 251–259. (6) Schemenauer, R. S.; Fuenzalida, H.; Cereceda, P. Bull. Am. Meteorol. Soc. 1988, 69, 138–147. (7) Schemenauer, R. S.; Cereceda, P. J. Appl. Meteorol. 1994, 33, 1313–1322. (8) Falconer, R. E.; Falconer, P. D. J. Geophys. Res. 1980, 85, 7465–7470. (9) Parker, A. R.; Lawrence, C. R. Nature 2001, 414, 33–34. (10) Garrod, R. P.; Harris, L. G.; Schofield, W. C. E.; McGettrick, J.; Ward, L. J.; Teare, D. O. H.; Badyal, J. P. S. Langmuir 2007, 23, 689–693. (11) Dawson, T. E. Oecologia 1998, 117, 476–485. (12) Martorell, C.; Ezcurra, E. J. Veg. Sci. 2002, 13, 651–662. (13) Martorell, C.; Ezcurra, E. Oecologia 2007, 151, 561–573. (14) Shirtcliffe, N. J.; McHale, G.; Newton, M. I. Langmuir 2009, 25, 14121–14128. (15) Benzing, D. H.; Henderson, K.; Kessel, B.; Sulak, J. Am. J. Bot. 1976, 63, 1009–1014. (16) Koch, K.; Barthlott, W. Philos. Trans. R. Soc., A 2009, 367, 1487–1509. (17) Otten, A.; Herminghaus, S. Langmuir 2004, 20, 2405–2408. (18) Hind, N. Curtis’s Bot. Mag. 2006, 23, 307–313. (19) Stokes, D. J. Adv. Eng. Mater. 2001, 3, 126–130. (20) Hynes, A. M.; Shenton, M. J.; Badyal, J. P. S. Macromolecules 1996, 29, 18–21. (21) Cassie, A. B. D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 546–551. (22) Mock, U.; F€orster, R.; Menz, W.; R€uhe, J. J. Phys.: Condens. Matter 2005, 17, S639–S648. (23) Williams, C. R.; Kruger, A.; Gage, K. S.; Tokay, A.; Cifelli, R.; Krajewski, W. F.; Kummerow, C. Geophys. Res. Lett. 2000, 27, 1763–1766. (24) Zheng, Y.; Han, D.; Zhai, J.; Jiang, L. Appl. Phys. Lett. 2008, 92, 084106. (25) Wier, K. A.; McCarthy, T. J. Langmuir 2006, 22, 2433–2436. (26) Wenzel, R. N. Ind. Chem. Eng. 1936, 28, 988–994. (27) Zhao, Y.; Lu, Q.; Li, M.; Li, X. Langmuir 2007, 23, 6212–6217. (28) Patankar, N. A. Langmuir 2004, 20, 7097–7102. (29) Wagner, G. J.; Wang, E.; Shepherd, R. W. Ann. Bot. 2004, 93, 3–11. (30) Schofield, W. C. E.; Badyal, J. P. S. ACS Appl. Mater. Interfaces 2009, 1, 2763–2767.
3802
dx.doi.org/10.1021/la2000014 |Langmuir 2011, 27, 3798–3802
