NANO LETTERS
Electric Field Induced, Superhydrophobic to Superhydrophilic Switching in Multiwalled Carbon Nanotube Papers
2008 Vol. 8, No. 9 2693-2696
Bhalchandra Kakade, Rutvik Mehta, Apurva Durge, Sneha Kulkarni, and Vijayamohanan Pillai* Physical and Materials Chemistry DiVision, National Chemical Laboratory, Pune 411 008, India Received April 9, 2008; Revised Manuscript Received June 13, 2008
ABSTRACT Superhydrophobic multiwalled carbon nanotube bucky paper, fabricated after ozonolysis, shows fascinating electrowetting behavior, which could be remarkably tuned by changing key solution variables like the ionic strength, the nature of the electrolyte, and the pH of the water droplet. More significantly, the droplet behavior can be reversibly switched between superhydrophobic, Cassie-Baxter state to hydrophilic, Wenzel state by the application of an electric field, especially below a threshold value.
Solid surfaces revealing water contact angles greater than 150° and lower than 30° are described by many as superhydrophobic and superhydrophilic respectively, although other diagnostic criteria such as hysteresis and low rolling angle are also important to describe the extent of this.1-4 Many leaves (e.g., Lotus) have hierarchical surfaces that prevent dirt and even bacteria from sticking. This “Lotus leaf effect” is related to the superhydrophobicity and has attracted tremendous attention from both academia and industry due to its wide applications involving water-repellent and stainless coatings, self-cleaning and antifouling surface designs, and laboratory-on-a-chip devices.5 In many of these applications, it would be of immense importance to externally control the interaction of liquids with a superhydrophobic surface, including the modification of contact angle, droplet mobility, and degree of penetration of the liquid. Consequently, several researchers have attempted to achieve a dynamic control over the wettability of superhydrophobic surfaces, and carbon nanotube (CNT) surfaces provide a convenient means for this because of easy manipulation of both geometrical and chemical parameters.4,6-8 More interesting would be the behavior of bucky papers, since they are usually porous and hydrophobic though their wetting characteristics can be modulated between the extremes by changing functionalization strategies, so that one could prepare papers which are wetted selectively.9 Also, forest of aligned CNTs shows superhydrophobic behavior due to * Corresponding author. E-mail:
[email protected]. Fax: (22)91-2025902636. Phone: +91-20-25902588. 10.1021/nl801012r CCC: $40.75 Published on Web 07/30/2008
2008 American Chemical Society
a higher degree of surface roughness as a result of microdomains between nanotubes grown on a substrate.10 This way of modulation of the wetting of carbon nanotubes, indeed, has many profound implications in nanofluidic transport, irrespective of the nature of the application. Wetting properties of such superhydrophobic surfaces could also be effectively controlled using external forces like electric and magnetic field. For example, Froumkine and coworkers have shown the wetting of polarized metal surface by electrolyte,11 similar to the one used by Lippman.12 Similarly, nanostructured parylene films deposited on the arrays of aligned multiwalled CNTs (MWCNTs) have shown effective electrowetting without any saturation limit up to 80 V.13 However, aligned growth of nanotubes on any solid substrate is practically a tedious process as this needs a stringent control over various parameters such as the selection of catalysts, precursor gas/liquid, and the carrier gas (and its flow rate) used during the synthesis. In order to achieve flexibility in such superhydrophobic nanotube films, fabrication of bucky paper could be an alternative with concomitant control over their wetting properties. Surprisingly, there are no reports on the electrowetting behavior of a continuous film of MWCNTs, despite its significance in many aforementioned applications. Hence, a correlation of electric field on the wetting behavior of nanotube paper as a function of ionic strength, nature of electrolyte, and pH is of immense importance to understand the fundamental chemistry and physics of nanotube surfaces.
Figure 1. (a) Electrowetting of bucky paper using deionized water, indicating oxygen evolution after 15 V and SEM images of bucky paper (b) showing both regions of before (white circled) and after electrowetting (dotted region); (c) magnified portion of electrowetted area indicated as black circle in (b), revealing the surface corrugation effects after electrowetting.
In the present work, we report electrowetting of bucky papers of MWCNTs using various aqueous solutions of KCl, NaCl, Na2SO4, and LiClO4 (0.05, 0.1. 0.5, and 1 M). Hydrophobic bucky papers of MWCNTs have been prepared after ozonolysis, since it is difficult to form a paper of pristine nanotube sample. We further demonstrate an interesting transition from superhydrophobic, Cassie-Baxter to hydrophilic, Wenzel form. An initial contact angle of 156° with slight pinning properties (roll off angle of 16°) has been made superhydrophilic with an external electric field. Effect of critical parameters like concentration and pH of the drop has been systematically studied to reveal their relative importance. In order to prepare superhydrophobic bucky paper, ozonolysis was carried out on indigenously synthesized CNTs.9 In brief, 120 mg of MWCNTs was mixed with 50 mL of dichloromethane and was sonicated for 30 min to get a fine dispersion. This was subsequently subjected to ozonolysis for about 60 min. Ozonolysis is an addition reaction where, the double bond between carbon atoms could be broken to generate many oxygenated functional groups through ozonide type of intermediates. It is expected to selectively generate more functional groups like carboxylic groups, ketone, esters, or alcohol on the side walls. Although there are abundant reports on the filter deposition of nanotubes, the pretreatment of ozonolysis would cause the purification of nanotubes along with the generation of hydrophobic functional groups like esters but not carboxylate or hydroxyl groups by quenching with dimethyl sulfide.14 The ozonolyzed dispersion of MWCNTs was then filtered through a PTFE membrane (0.2 µm pore diameter) under vacuum, and the bucky paper on the surface of membrane was peeled off and dried at 60 °C. Figure 1a shows a droplet profile of deionized water on the bucky paper surface as a function of voltage, which 2694
reveals the spreading of liquid in a systematic fashion. The observation of an initial contact angle of 156° reveals the superhydrophobic nature of the bucky paper, perhaps due to the presence of several hydrophobic functional groups produced during ozonolysis.15 A continuous decrease in the contact angle with respect to direct current (dc) voltage shows a remarkable tuning of the surface properties of nanotube paper, especially with respect to its change in its surface oxygen content (Supporting Information S2). Appearance of bubbles at 15 V is indicative of electrolysis of water with subsequent reduction in the interfacial tension, generating more hydrophilic groups. The liquid remains suspended at the top of the bucky paper and forms a stable “immobile ball” because of a slight pinning action with an angle of 156°. Interestingly, the application of electric field induces a transition from the superhydrophobic to the hydrophilic state where the immobile droplet undergoes a distortion which could be seen even with the naked eye. However, some of the liquid penetrates through the bucky paper substrate as a result of this transition. This could be attributed to the fact that the electrical energy induces both external and internal wetting (capillarity) due to the filling of the lumen of MWCNTs. An approximate estimation of the capillary rise, assuming constant inner diameter (6 nm) and an initial contact angle between capillary surface and water to be zero, gives 0.49 mm, and this water could be seen in the low resolution TEM images (shown in Supporting Information S3). This effect of electrocapillarity is strictly based on electrostatic control of the solid-liquid interfacial tension.11 Nevertheless, at a critical value of the applied potential, wetting commences to move water up along the CNT side walls, concomitantly filling the nanotube cores to show a capillary action. This is in excellent agreement with the recent observation of Barber Nano Lett., Vol. 8, No. 9, 2008
Figure 2. (a) Water droplet profile with respect to the ionic strength of electrolyte using various concentrations of KCl, indicating that a very small amount of electrolyte could reduce the surface energy of the nanotubes; (b) reversible wetting and dewetting of bucky paper using 0.1 M KCl, showing the tuning between the contact angles of 100° (at cathodic potential) and 110° (at anodic potential); (c) electrowetting as a function of type of electrolyte and (d) as a function of pH, showing better stability at alkaline pH.
et al., where an enhanced internal wetting for the open ended individual MWCNTs is demonstrated using Wilhelmy force balance.16 This also explains why water exhibits a significantly larger interaction with the nanotube (as a function of diameter) causing unique behavior as compared with organic liquids.17 Figure 1b clearly demonstrates the comparative surface topography (in SEM images) of bucky paper before (white circled) and after electrowetting (black circled) experiments revealing the formation of surface roughness. More specifically Figure 1c shows a magnified SEM image of only the electrowetted (black circled portion from Figure 1b) region confirming the change in surface roughness developed after electrowetting. Interestingly, the compositional variations, clearly seen in energy dispersive X-ray analysis (ED) spectrum, also confirm these findings (Supporting Information S2). For example, the oxygen peak has been severely affected by electrowetting, revealing the field-induced functionalization capable of generating more oxygenated groups on the side walls. This could be an efficient route toward functionalization of nanotubes to overcome the surface inertness as substantiated by a decrease in the C/O peak intensity ratio, facilitating enhanced hydrophilic nature of the bucky paper. The dangling bonds present on the side walls of nanotubes with hydrogen atoms also show probable variations in surface tension presumably due to the hydrogen bonding when the surface comes in contact with a water droplet. Nano Lett., Vol. 8, No. 9, 2008
Figure 2a shows the variation of electrowetting results with respect to the ionic strength where the contact angle decreases by increasing the concentration of the electrolyte. It means that a very small amount of electrolyte could reduce the surface energy of the nanotubes to an appreciable extent. More interestingly, the wetting could be reversed by changing the polarity as shown in Figure 2b (100° becomes 110° on reversal) although adsorption effects are not taken into account. Indeed, the droplet behavior can be reversibly switched between the superhydrophobic, Cassie-Baxter state and the hydrophilic, Wenzel state by the application of an electric field. However, the hysteresis of pristine sample shows an appreciable extent of variation, since the reversal, clearly exhibits a surface manipulation (contact angle of 110° as compared with initial 155°) due to the pinning action during both advancing and receding modes. Interestingly, this reversal of switching is possible only when the voltage applied is lower than a threshold voltage. An approximate relation between the threshold voltage and the wetting as a function of ionic strength is shown in Supporting Information S4, where a gradual decrease in the threshold could be observed with an increase in KCl concentration. The application of electric field as a function of the type of cation and anion also exhibits a remarkable change in the wetting behavior of bucky paper. For example, Figure 2c shows a systematic variation of the surface properties as a result of ionic size, especially the size of the anion. A small amount of bigger sized SO42- ions requires a lower voltage 2695
for the liquid to spread over nanotube surface. Hence, the surface hydrophilicity could be enhanced by increasing the concentration of larger anions, where a possible trend, in order to achieve a better hydrophility, would be SO42- > ClO4- > Cl-, despite a reversal in switching below a threshold voltage. A similar possible trend (with decreasing threshold voltage) for various electrolytes using 0.05 M concentration is KCl > NaCl > LiClO4 > Na2SO4 (Supporting Information S4). Our electrowetting results also demonstrate better droplet stability at alkaline pH () 12) compared with that of deionized water especially above higher voltages (>50 V), whereas around 50 V, both acidic and neutral droplets show a steep variation in contact angle (Figure 2d). However, the variation of contact angle with electric field below 50 V is obviously less for the case of deionized water droplet due to less field penetration, unless electrolysis commences at this threshold value. In other words, electrowetting at acidic and alkaline pH shows pronounced effects of electrolysis and adsorption thus exhibiting sudden drop in the contact angle at initial voltages. It could also be explained by the fact that, during electrowetting, the easy hydrolysis of the ester groups of nanotubes take place in acidic pH rather than at alkaline pH. However, other factors including preferential solvation, surface diffusion, specific adsorption, and so forth also could contribute to such variations in surface properties with time as a function of pH. If we assume a negligible role for these factors, the change in the wetting of the surface with pH could be entirely dominated by the surface tension of solid-liquid interface (γSL) and not by that of liquid-vapor (γLV) and solid-vapor (γSV) interfaces, since γSL is determined by the energy of interaction of the liquid with a collection of groups on the exterior surface of the solid.18,19 Hence, the observed molecular level changes with variation in pH can be considered as a direct consequence of the ionization of surface functional groups.18,19 Indeed, capillary effects along with water spreading through the porous bucky paper (formed due to the wide diameter distribution of MWCNTs) are expected to show a transition from Cassie state to the Wenzel state. Thus the contact angle between a MWCNT buckypaper and a salt-containing droplet can be tuned and indeed reversed from hydrophobic to hydrophilic by the application of an electric field. In conclusion, the superhydrophobic MWCNT bucky paper, fabricated after ozonolysis, shows fascinating wetting behavior as a result of electric field, which could be remarkably tuned by changing key solution variables like ionic strength, nature of electrolyte, and pH of the droplet.
2696
More significantly, the droplet behavior can be reversibly switched between superhydrophobic, Cassie-Baxter state to hydrophilic, Wenzel state by the application of electric field, especially below a threshold. Our present findings may have important benefits for a wide range of applications of CNT surfaces such as design of biosensors, smart composites, variable lenses, and display screen. Acknowledgment. Authors thank CSIR for funding this work through a NMITLI programme. B.K. and S.K. thank UGC and CSIR for financial support. We also thank Dr. Sivaram, Director NCL, for valuable discussions.
Note Added in Proof: One recent paper by A. Ahuja et al. has appeared on “electric tuning of superlyophobic surfaces” in Langmuir 2008, 24, 9–14 which confirms our findings of manipulation of wetting by electric field using other liquids. Supporting Information Available: Schematic, TEM, EDS, and contact angle measurements at different threshold voltages. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Martines, E.; Seunarine, K.; Morgan, H.; Gadegaard, N.; Wilkinson, C. D. W.; Riehle, M. O. Nano Lett. 2005, 5, 2097. (2) Gao, L.; McCarthy, T. J. Langmuir 2006, 22, 6234. (3) Feng, X.; Jiang, L. AdV. Mater. 2006, 18, 3063. (4) Liu, H.; Zhai, J.; Jiang, L. Soft Matter 2006, 2, 811. (5) Feng, L.; Li, S. H.; Li, Y. S.; Li, H. J.; Zhang, L. J.; Zhai, J.; Song, Y. L.; Liu, B. Q.; Jiang, L.; Zhu, D. B. AdV. Mater. 2002, 14, 1857. (6) Ichimura, K.; Oh, S.-K.; Nakagawa, M. Science 2000, 288, 1624. (7) Lahann, J.; Mitragotri, S.; Tran, T.-N.; Kaido, H.; Sundaram, J.; Choi, I. S.; Hoffer, S.; Somorjai, G.; Langer, R. Science 2003, 299, 371. (8) Isaakson, J.; Tengstedt, C.; Fahlman, M.; Robinson, N.; Berggren, M. AdV. Mater. 2004, 16, 316. (9) Kakade, B. A.; Pillai, V. K. J. Phys. Chem. C 2008, 112, 3183. (10) Zhu, L.; Xu, J.; Xiu, Y.; Sun, Y.; Hess, D. W.; Wong, C. J. Phys. Chem. B 2006, 110, 15945. (11) Froumkine, A. Actualites Scientifiques 1976, 373, 5. (12) Lippman, G. Ann. Chim. Phys. 1875, 5, 495. (13) Wang, Z.; Ou, Y.; Lu, T.-M.; Koratkar, N. J. Phys. Chem. B 2007, 111, 4296. (14) Banerjee, S.; Hemraj-Benny, T.; Balasubramanian, M.; Fischer, D. A.; Misewich, J. A.; Wong, S. S. Chem. Phys. Chem. 2004, 5, 1416. (15) Dorrer, C.; Ruh¨e, J. Langmuir 2006, 22, 7652. (16) Barber, A. H.; Cohen, S. R.; Wagner, H. D. Phys. ReV. B 2005, 71, 115443/1. (17) Ghosh, S.; Sood, A. K.; Kumar, N. Science 2003, 299, 1042. (18) Holmes-Farley, S. R.; Reamey, R. H.; McCarthy, T. J.; Deutch, J.; Whitesides, G. M. Langmuir 1985, 1, 725. (19) Holmes-Farley, S. R.; Bain, C. D.; Whitesides, G. M. Langmuir 1988, 4, 921.
NL801012R
Nano Lett., Vol. 8, No. 9, 2008
