Reversible Tuning of the Wettability of Carbon Nanotube Arrays: The

Jun 14, 2011 - UV/ozone treatment is a well-known dry-oxidation process to functionalize the surface of nanotubes with oxygenated functional groups at...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/Langmuir

Reversible Tuning of the Wettability of Carbon Nanotube Arrays: The Effect of Ultraviolet/Ozone and Vacuum Pyrolysis Treatments Adrianus I. Aria and Morteza Gharib* Graduate Aeronautical Laboratories, California Institute of Technology, Pasadena, California 91125, United States

bS Supporting Information ABSTRACT: Among diverse types of synthetic materials, arrays of vertically aligned carbon nanotubes have attracted the most attention, mainly because of their exceptional mechanical, electrical, optical, and thermal properties. However, their wetting properties are yet to be understood. In this present study, oxygenated surface functional groups have been identified as a vital factor in controlling the wetting properties of carbon nanotube arrays. The results presented herein indeed show that a combination of ultraviolet/ozone and vacuum pyrolysis treatments can be used to vary the surface concentration of these functional groups such that the carbon nanotube array can be repeatedly switched between hydrophilic and hydrophobic.

’ INTRODUCTION In recent years, production of synthetic materials with tunable wetting properties has been reported.1 It has been fully understood that the wetting properties of a material is dictated by its surface chemistry2,3 and topographic structure.4 7 The ability to tune the wetting properties of a material is considered critical because of its relevance to numerous applications, including selfcleaning non-stick surfaces for preventing biofouling,8 hydrodynamic skin-friction drag reduction,9 and high power density supercapacitors for storing charge energy.10 Among these materials, vertically aligned multi-walled carbon nanotube (MWNT) arrays have gained enormous attentions, because of their simplicity fabrication process and inherent twolength scale topographic structure. Many reported studies suggest that the MWNT arrays can be easily made hydrophilic by functionalizing their surfaces with oxygenated functional groups. Such functional groups allow for a hydrogen bond between the surface of nanotubes and water molecules to form. However, complicated processes are always involved in producing superhydrophobic MWNT arrays. To make these arrays superhydrophobic, they have to be coated with non-wetting chemicals, such as polytetrafluoroethylene (PTFE), ZnO, and fluoroalkylsilane,11 13 or be pacified by plasma treatments, such as CF4, CH4, and NF3.14,15 Note that a superhydrophobic surface is defined as a surface that is extremely difficult to get wet and that has a static contact angle (CA) higher than 150°, along with a CA hysteresis less than 10°.4,16,17 The findings presented herein show that the wettability of MWNT arrays can be tuned straightforwardly via a combination of ultraviolet (UV)/ozone and vacuum pyrolysis treatments. UV/ozone treatment is a well-known dry-oxidation process to functionalize the surface of nanotubes with oxygenated functional r 2011 American Chemical Society

groups at standard room temperature and pressure.18 Basically, UV radiation attacks the end caps and outer sidewall of the nanotubes, allowing ozone to oxidize their surface defect sites.19,20 Vacuum pyrolysis treatment was used to reverse the effect of oxidation by removing these groups from the surface of nanotubes while keeping the microscopic structure of the array intact. Typically, a vacuum pyrolysis treatment performed at a mild vacuum and a moderate temperature is sufficient to deoxidize the array. Using these treatments, the amount of oxygenated functional groups can be controlled carefully such that the MWNT array can be repeatedly switched between hydrophilic and hydrophobic (Figure 1a). In this study, seven different sets of MWNT arrays with various degrees of oxidation were examined. The first two sets of samples were the as-grown MWNT arrays (AG) and the AG arrays that had been subjected to vacuum pyrolysis treatment (SH1). The next three sets of samples were the SH1 arrays that had been exposed to UV/ozone treatment for 1 min (UO1), 5 min (UO2), and 15 min (UO3). The next two sets of samples were the UO3 arrays that had been subjected to vacuum pyrolysis treatment for the second time (SH2) and the SH2 arrays that had been subjected to another 15 min of UV/ozone treatment (UO4). All vacuum pyrolysis treatments were conducted at a mild vacuum of 2.5 Torr and a moderate temperature of 250 °C for at least 3 h.

’ RESULTS AND DISCUSSION Here, the degree of wettability of each set of samples was assessed by measuring the average static CA for water (Figure 1b). Received: May 16, 2011 Revised: June 10, 2011 Published: June 14, 2011 9005

dx.doi.org/10.1021/la201841m | Langmuir 2011, 27, 9005–9011

Langmuir

ARTICLE

Figure 1. (a) UV/ozone and vacuum pyrolysis treatments are used to vary the wetting properties of MWNT arrays. Oxygen adsorption occurs during UV/ozone treatment, and oxygen desorption take places during vacuum pyrolysis treatment. (b) Static CA of the MWNT arrays with different wetting properties. (c) Low-magnification and (d) high-magnification scanning electron microscopy (SEM) images of MWNT arrays used in this study showing the overall vertical alignment and high packing density of the nanotubes.

As expected, the AG arrays were found to be mildly hydrophobic, with CA ≈ 143°. An extremely high CA exhibited by the SH1 arrays, with CA ≈ 171°, showed the effectiveness of vacuum pyrolysis treatment in rendering the MWNT arrays superhydrophobic. The effect of oxidation started to appear once the MWNT arrays were exposed to UV/ozone treatment. The UO1 arrays were found to be mildly hydrophobic, with CA ≈ 132°. At a prolonged duration of UV/ozone treatment, the MWNT arrays became increasingly hydrophilic such that their average static CA plummeted to CA ≈ 75° and CA ≈ 27° for UO2 and UO3 arrays, respectively. The effectiveness in rendering the MWNT arrays superhydrophobic was shown once again by the extremely high static CA of the SH2 arrays, with CA ≈ 171°. The effect of oxidation reappeared after the MWNT arrays were subjected to another round of UV/ozone treatment. Indeed, the average static CA for UO4 arrays was found to be CA ≈ 45°. MWNT arrays used in this study were grown by the standard chemical vapor deposition (CVD) technique on silicon substrates. The average length of all of the arrays was chosen to be about 14 ( 4 μm (Figure 1c), which was about the minimum length that can be made using the CVD technique while preserving the overall vertical alignment and high packing density of the arrays (Figure 1d). One of the reasons this length was chosen is for the difficulties in producing a superhydrophobic surface out of short MWNT arrays as reported in the previous studies.12,21,22 Another reason is for the simplicity of the experiment because a longer exposure time of UV/ozone treatment is needed to oxidize a taller MWNT array. For instance, a millimeter long MWNT array needs almost 4 h of UV/ozone treatment to reach the same static CA as the UO3 samples.

Fourier transform infrared (FTIR) spectrometry analysis was performed to evaluate the effect of UV/ozone and vacuum pyrolysis treatments on the surface chemistry of the MWNT arrays (Figure 2a). Three absorbance peaks at 810 1320, 1340 1600, and 1650 1740 cm 1 can be observed in almost all samples. The origin of the peak at 718 cm 1 is unknown; hence, it is unassigned at this time. The peaks at 970, 1028, 1154, and 1201 cm 1 correspond to C O stretching modes,23 and the broad shoulder band at 810 1320 cm 1 suggests the existence of C O C bonds from the ester functional group.19,20,24,25 The peaks at 1378, 1462, 1541, and 1574 cm 1 indicate the presence of CdC stretching vibration modes of the MWNT walls.19,20,23,24 The narrow band at a peak of 1703 cm 1 corresponds to CdO stretching modes of either quinone or carboxyl groups.19,20,23,25 Interestingly, all peaks that correspond to C O and CdO stretching modes are found to be very weak in the SH1 and SH2 samples, suggesting that the oxygen desorption process occurs during the vacuum pyrolysis treatment. Although the whole oxygen adsorption and desorption cycles via UV/ozone and vacuum pyrolysis treatments seem to be completely reversible, it is worth noting that a very small amount of carbon atoms is actually etched from the MWNT in the form of CO and CO2 molecules during the desorption process. Hence, the atomic-scale structure of the MWNT array is slightly altered each time UV/ozone and vacuum pyrolysis treatments are performed. Nevertheless, the MWNT array can still be repeatedly switched between hydrophilic and hydrophobic for many times as long as the overall microscopic structure of the array is still intact. Raman spectroscopy and transmission electron microscopy (TEM) analyses of the UV/ozone- and vacuumpyrolysis-treated MWNT arrays show that the graphitic structures 9006

dx.doi.org/10.1021/la201841m |Langmuir 2011, 27, 9005–9011

Langmuir

ARTICLE

Figure 2. (a) Typical FTIR spectra of MWNT arrays with various degrees of oxidation, showing absorbance peaks at 810 1320, 1340 1600, and 1650 1740 cm 1. (b) Integrated absorbance for C O (810 1320 cm 1) and CdO (1650 1740 cm 1) obtained from panel a as a function of the static CA of the MWNT arrays. (c) Oxygen/carbon atomic ratio (O/C ratio) of MWNT arrays obtained from EDS as a function of static CA of the MWNT arrays. The shaded region in panels b and c represents the superhydrophobic regime. AG, SH1, UO1, UO2, UO3, SH2, and UO4 arrays are represented in panels b and c by 4, /, O, ], 3, 0, and b, respectively. High-resolution XPS spectra of the C C 1s peak of (d) UO2, (e) UO3, and (f) SH2 arrays. Deconvolution of these spectra shows the presence of hydroxyl (C OH), carbonyl (CdO), and carboxyl ( COOH) groups.

of the arrays are still practically intact after several cycles (SI 1 in the Supporting Information). An intensive study to determine the maximum number of oxygen adsorption and desorption cycles that a MWNT array can handle via UV/ozone and vacuum pyrolysis treatments is yet to be performed. To directly correlate the effect of the surface concentration of oxygenated functional groups to the wetting properties of the MWNT arrays, the integrated absorbance of C O and CdO stretching modes is plotted against the static CA of the arrays. Interestingly, this plot shows a very strong correlation between the integrated absorbance and the static CA of the arrays, where the integrated absorbance decreases monotonically with the

increase of the static CA (Figure 2b). As mentioned before, the integrated absorbance of C O and CdO stretching modes decreases from 8.34 and 0.34 A cm 1, respectively, for the highly hydrophilic UO3 arrays with CA ≈ 27° to as low as 1.17 and 0.01 A cm 1, respectively, for the superhydrophobic SH2 arrays with CA ≈ 171°. Obviously, MWNT arrays with a higher surface concentration of oxygenated functional groups are more hydrophilic, and ones with a very low concentration exhibit superhydrophobic behavior. A quick and simple oxidation-time-independent relation can be observed by plotting the oxygen/carbon atomic ratio (O/C ratio) of the MWNT arrays to their static CA. In this study, the 9007

dx.doi.org/10.1021/la201841m |Langmuir 2011, 27, 9005–9011

Langmuir

ARTICLE

Figure 3. (a) Time-lapse images of the water droplet impacting the surface of MWNT arrays with different wetting properties. Scale bars indicate 5 mm. (b) Dispersion of MWNT with various wetting properties in DI water. The nanotubes from AG and UO2 arrays precipitated in several hours and days, respectively, after the dispersion were performed. The dispersion of nanotubes from UO3 arrays was found stable even after more than 2 months afterward. The wettability of the MWNT increases from left to right. (c) Bode impedance plot of MWNT arrays with various wetting properties in 1 M NaCl aqueous electrolyte. Only four spectra are shown for clarity. AG, UO2, UO3, and SH2 arrays are indicated by 4, ], 3, and 0, respectively. An arrow indicates the increase of static CA of the arrays.

O/C ratio of MWNT arrays was obtained from the energydispersive X-ray spectroscopy (EDS). Because EDS can be performed relatively easily, the O/C ratio can be used to quickly assess the degree of oxidation of the MWNT arrays by comparing the amount of oxygen atoms to the amount of carbon atoms present at the surface of the array. Therefore, a higher O/C ratio corresponds to a higher degree of oxidation and vice versa. The O/C ratio decreases as the static CA of the array increases, where the O/C ratio of hydrophilic UO3 arrays was measured to be as high as 15.49% and that of the superhydrophobic SH2 arrays was measured to be as low as 6.77%. Notice that the O/C ratio of SH2 arrays is not zero, which suggests that a small amount of oxygen cannot be easily removed by vacuum pyrolysis treatment. X-ray photoelectron spectroscopy (XPS) analysis was conducted to further verify this finding. Deconvolution of the highresolution XPS spectra at the binding energy of 283 293 eV shows four distinct peaks, with one primary peak associated with the presence of sp2 C C 1s bonds (∼284.9 eV) and three secondary peaks associated with the presence of hydroxyl C OH (∼285.4 eV), carbonyl CdO (∼287.4 eV), and carboxyl COOH (∼289.7 eV) functional groups.26,27 The XPS spectra of the AG arrays indicate that the peaks associated with C OH and CdO groups are considerably large, while the peak associated with the COOH group is found to be very weak. Because

the MWNT arrays undergo a prolonged UV/ozone treatment, all peaks associated with C OH, CdO, and COOH groups become more pronounced. Notice that, after short exposure of the UV/ozone treatment, the surface concentration of CdO groups increases at a much faster rate than that of C OH and COOH groups (Figure 2d). However, at a longer exposure time, the surface concentration of CdO groups decreases slightly, while that of C OH and COOH groups continues to increase (Figure 2e). There is a possibility that some CdO groups were further oxidized into COOH groups during the UV/ozone treatment. As expected, the amount of C OH, CdO, and COOH groups decreases significantly after the vacuum pyrolysis treatment (Figure 2f). Although the existence of these peaks suggests that the vacuum pyrolysis treatment does not completely deoxidize the array, the surface concentration of these groups is found to be extremely low. This finding does agree with the FTIR result, which shows that the MWNT arrays can be rendered superhydrophobic, with CA ≈ 171°, even though there are still some traces of C OH groups on the arrays. One may expect that the pristine MWNT arrays that are free from these oxygenated functional groups will exhibit a perfect superhydrophobic behavior with CA = 180°. To demonstrate that the UV/ozone- and vacuum-pyrolysistreated MWNT arrays were indeed hydrophilic and hydrophobic, 9008

dx.doi.org/10.1021/la201841m |Langmuir 2011, 27, 9005–9011

Langmuir respectively, the dynamic effect of a free-falling water droplet on the surface of these MWNT arrays was observed (Figure 3a). As a free-falling water droplet hit the surface of the AG arrays, it deformed heavily and then bounced off of the surface. However, because the AG arrays were mildly hydrophobic, there was always a small portion of the droplet pinned to the surface, preventing the droplet from bouncing off completely from the surface. This pinning phenomenon was more pronounced when the water droplet hit the surface of the more hydrophilic UO2, UO3, and UO4 arrays. For example, on the UO3 arrays, the droplet was pinned to the surface and came to rest shortly after the impact. A totally different behavior was observed for SH1 and SH2 arrays, where the free-falling water droplet bounced completely off of the surface. In contrast to the previous study,28 the droplet pinning did not occur on the superhydrophobic SH1 and SH2 arrays even at a considerably high impact velocity of 2.22 ms 1. The dispersions of the UV/ozone- and vacuum-pyrolysistreated MWNT in deionized (DI) water were observed to further verify their wetting properties. A small portion of each sample was scrapped from its growth substrates and ultrasonically dispersed in DI water (Figure 3b). As expected, the floating nanotubes on the surface of the water column indicated that the superhydrophobic nanotubes from SH1 and SH2 arrays could not be dispersed in DI water. Similarly, the mildly hydrophobic nanotubes from AG and UO1 arrays could not be dispersed easily in DI water, although a completely different behavior was observed. Instead of floating, the nanotubes from AG and UO1 arrays were completely submerged in the water column. The dispersibility of nanotubes in DI water was unquestionably increased after the nanotubes had been exposed to UV/ozone treatment. Some degree of dispersion of hydrophilic nanotubes from UO2 and UO4 arrays could be observed clearly. In fact, the nanotubes from UO3 arrays could be dispersed quite easily in DI water. In agreement with the previous study,27,29 hydrophilic nanotubes from UO3 arrays do not form sedimentation even after 2 months of settling time. The electrochemical characterization was conducted to provide additional insight on the effect of UV/ozone and vacuum pyrolysis treatments to the physiochemical properties of MWNT arrays. It is well-known that the presence of carboxyl groups may increase the capacitance of the array by a factor of 3, while the hydrophobic nanotubes exhibit a very low electrochemical capacitance in aqueous solution.30 In agreement with that study, the impedance of MWNT arrays was found to be highly dependent upon their wettability. In other words, the impedance of the MWNT arrays in aqueous electrolyte can be varied by manipulating the surface concentration of oxygenated functional groups via UV/ozone and vacuum pyrolysis treatments. The Bode impedance plot of the electrochemical impedance spectroscopy (EIS) data in 1 M NaCl aqueous electrolyte shows that the superhydrophobic MWNT arrays yield much higher impedance than the hydrophilic counterpart. The difference in impedance can be observed clearly at considerably low frequency of f < 1 kHz, where the effect of double-layer capacitance becomes increasingly dominant. In fact, the frequency at which the transition from pure resistive behavior to a capacitiveresistive behavior starts to occur decreases with the increase of wettability of the array (Figure 3c). The transition for the SH2 arrays occurs at a higher frequency of f ≈ 1.6 kHz, while the transition for the UO3 arrays occurs at a much lower frequency of f ≈ 42.8 Hz. The absolute impedance of the superhydrophobic SH2 arrays is measured to be about 3 orders of magnitude higher

ARTICLE

than that of the hydrophilic UO3 arrays at a very low frequency of f = 12.6 mHz, where |z| ≈ 163 kΩ for the SH2 arrays and |z| ≈ 650 Ω for the UO3 arrays. The extremely large discrepancy of absolute impedance at lower frequency suggests that the doublelayer capacitance of superhydrophobic SH2 arrays is very much smaller than that of hydrophilic UO3 arrays. These findings are expected because of the presence of a thin film of air on the interface between the surface of the superhydrophobic MWNT array and the aqueous electrolyte, which inhibits electron transfer from the arrays and blocks protons in the electrolyte to approach the surface of the array.30 In addition, this thin air film significantly reduces the effective contact area between the surface of the nanotubes and the electrolyte molecules, which results in a much smaller effective area of the Helmholtz layer and a dramatic decrease of double-layer capacitance. Thus, the impedance of the superhydrophobic MWNT arrays is measured more than 2 orders of magnitude larger than that of the hydrophilic MWNT arrays. A completely opposite behavior may be observed if a polar-aprotic-based or nonpolarbased electrolyte is used. In such a case, higher impedance is expected to be observed from a hydrophilic MWNT array and vice versa.

’ CONCLUSION In conclusion, the findings reported herein show that the wetting properties of MWNT arrays can be tuned easily and precisely by controlling the surface concentration of oxygenated functional groups, e.g., hydroxyl, carbonyl, and carboxyl groups, via UV/ozone and vacuum pyrolysis treatments. MWNT arrays with a very low amount of oxygenated functional groups exhibit a superhydrophobic behavior such that they cannot be dispersed in DI water and their impedance in aqueous electrolytes is extremely high. These arrays have an extreme water repellency capability such that a water droplet will bounce off of their surface upon impact. In contrast, MWNT arrays with a very high surface concentration of oxygenated functional groups exhibit an extreme hydrophilic behavior such that they can be dispersed easily in DI water and their impedance in aqueous electrolytes is tremendously low. Because the microscopic structures and packing density of the arrays are maintained during the UV/ ozone and vacuum pyrolysis treatments, all MWNT arrays can be repeatedly switched between hydrophilic and hydrophobic. These findings show a very high potential in many industrial applications, including skin friction drag reduction in fluid flow and charge energy storage. ’ METHODS MWNT Growth. The vertically aligned MWNT arrays used in this present study were grown using thermal chemical vapor deposition on silicon wafer substrates. These wafers were coated with a 1 nm iron catalyst layer using an electron beam evaporator (Temescal BJD 1800) and diced into 1  1 cm samples. The growth itself was performed in a 1 in. diameter quartz tube furnace (Lindberg/BlueM Single-Zone Tube Furnace) under the 490 standard cubic centimeters per minute (sccm) ethylene gas (Matheson 99.999%) and 210 sccm hydrogen gas (Airgas 99.999%) at a temperature of 750 °C and a pressure of 600 Torr. The flow rate and pressure of the gases were maintained by an electronic mass flow controller (MKS πMFC) and a pressure controller (MKS πPC). The overall growth quality, including the length of the array, was characterized under a scanning electron microscope (ZEISS LEO 1550VP). 9009

dx.doi.org/10.1021/la201841m |Langmuir 2011, 27, 9005–9011

Langmuir UV/Ozone and Vacuum Pyrolysis Treatments. All UV/ozone treatments were perforemed by exposing the MWNT arrays to 185 nm of UV radiation (Bioforce Nanosciences UV/Ozone Procleaner Plus) in air at room temperature and pressure. Here, all vacuum pyrolysis treatments were conducted in a vacuum oven (VWR Signature Vacuum Oven) at a mild vacuum of 2.5 Torr and a moderate temperature of 250 °C for at least 3 h. CA Measurement. All static CA measurements were conducted with a CA goniometer at room pressure and temperature. A 5 μL water droplet was dropped on the top surface of each sample using a 5 μL syringe (Hamilton 7105KH) equipped with a 31 gauge flat-tipped needle (Hamilton KF731). Once a water droplet had come to rest on the samples, images of the water droplet were then taken 10 s afterward to ensure that the equilibrium condition was achieved and to avoid the excessive evaporation of the droplet. The CAs were then measured by processing the captured images with LBADSA software.31 FTIR, EDS, and XPS. For the FTIR spectroscopy study, a small portion of the MWNT array was scraped from its growth substrate, dispersed in deuterated dichloromethane (Sigma-Aldrich, 99.9 atom %), drop-cast onto a KBr window, and then dried overnight under low vacuum (10 Torr), to remove the solvent leftover. The FTIR spectra were obtained by an infrared spectrometer (Nicolet 6700) using a blank KBr window as a baseline correction. The O/C ratio of each sample was measured using EDS (Oxford INCA Energy 300) during the SEM characterization at an acceleration voltage of 5 kV. The XPS spectra were obtained by an X-ray photoelectron spectrometer (Surface Science M-Probe XPS). Curve fitting and deconvolution of the XPS spectra were performed using a Gaussian Lorentzian peak curve fit with Shirley baseline correction. MWNT Dispersion. For MWNT dispersion study, a small portion of the MWNT array was scrapped from its growth substrates and ultrasonically dispersed in DI water for 1 h. The dispersions were then left untouched for more than 2 months to assess their stability. EIS. EIS was conducted in three-electrode configuration using a potentiostat (Biologic SP-200) at a frequency of 10 mHz 1 MHz. The MWNT arrays were set as the working electrode; a platinum wire was used as the counter electrode; and a saturated calomel electrode was used as the reference electrode.

’ ASSOCIATED CONTENT

bS

Supporting Information. Raman spectroscopy spectra and TEM images as well as Nyquist impedance plot of the MWNT arrays. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by The Charyk Foundation, The Fletcher Jones Foundation under grant number 9900600 and The Office of Naval Research under grant number N00014-11-10031. A. I. Aria was supported by a Fulbright Scholarship. The authors acknowledge the Kavli Nanoscience Institute at the California Institute of Technology for use of the nanofabrication instruments, the Molecular Materials Research Center of the Beckman Institute at the California Institute of Technology for use of the FTIR, XPS, and contact angle goniometer, and the Analytical Facility Division of Geological and Planetary

ARTICLE

Sciences of the California Institute of Technology for use of SEM/EDS. The authors also gratefully acknowledge Prof. George Rossman for giving access to the Raman spectrometer and Dr. Masoud Beizai for valuable discussion in interpreting the FTIR spectra.

’ REFERENCES (1) Tao, Z.; Tianhong, C. Tunable wetting properties of patterned silicon microchannels with varied surface free energy based on layer-bylayer nano self-assembly. J. Micromech. Microeng. 2011, 21 (4), No. 045015. (2) Emsley, J. Very strong hydrogen-bonding. Chem. Soc. Rev. 1980, 9 (1), 91–124. (3) Wenzel, R. N. Resistance of solid surfaces to wetting by water. Ind. Eng. Chem. 1936, 28 (8), 988–994. (4) Bhushan, B.; Jung, Y.; Koch, K. Micro-, nano- and hierarchical structures for superhydrophobicity, self-cleaning and low adhesion. Philos. Trans. R. Soc., A 2009, 367 (1894), 1631–1672. (5) Gao, L.; McCarthy, T. A commercially available perfectly hydrophobic material (θA/θR = 180°/180°). Langmuir 2007, 23 (18), 9125–9127. (6) Gao, L.; McCarthy, T. A perfectly hydrophobic surface (θA/θR = 180°/180°). J. Am. Chem. Soc. 2006, 128 (28), 9052–9053. (7) Krupenkin, T.; Taylor, J.; Schneider, T.; Yang, S. From rolling ball to complete wetting: The dynamic tuning of liquids on nanostructured surfaces. Langmuir 2004, 20 (10), 3824–3827. (8) Scardino, A. J.; Zhang, H.; Cookson, D. J.; Lamb, R. N.; Nys, R. d. The role of nano-roughness in antifouling. Biofouling 2009, 25 (8), 757–767. (9) Rothstein, J. Slip on superhydrophobic surfaces. Annu. Rev. Fluid Mech. 2010, 42 (1), 89–109. (10) Lee, S. W.; Yabuuchi, N.; Gallant, B. M.; Chen, S.; Kim, B.-S.; Hammond, P. T.; Shao-Horn, Y. High-power lithium batteries from functionalized carbon-nanotube electrodes. Nat. Nanotechnol. 2010, 5 (7), 531–537. (11) Huang, L.; Lau, S.; Yang, H.; Leong, E.; Yu, S.; Prawer, S. Stable superhydrophobic surface via carbon nanotubes coated with a ZnO thin film. J. Phys. Chem. B 2005, 109 (16), 7746–7748. (12) Lau, K.; Bico, J.; Teo, K.; Chhowalla, M.; Amaratunga, G.; Milne, W. Superhydrophobic carbon nanotube forests. Nano Lett. 2003, 3 (12), 1701–1705. (13) Feng, L.; Li, S.; Li, Y.; Li, H.; Zhang, L.; Zhai, J.; Song, Y.; Liu, B.; Jiang, L.; Zhu, D. Super-hydrophobic surfaces: From natural to artificial. Adv. Mater. 2002, 14 (24), 1857–1860. (14) Hong, Y.; Uhm, H. Superhydrophobicity of a material made from multiwalled carbon nanotubes. Appl. Phys. Lett. 2006, 88 (24), No. 244101. (15) Cho, S.; Hong, Y.; Uhm, H. Hydrophobic coating of carbon nanotubes by CH4 glow plasma at low pressure, and their resulting wettability. J. Mater. Chem. 2007, 17 (3), 232–237. (16) Wang, S.; Jiang, L. Definition of superhydrophobic states. Adv. Mater. 2007, 19 (21), 3423–3424. (17) Men, X.; Zhang, Z.; Yang, J.; Wang, K.; Jiang, W. Superhydrophobic/superhydrophilic surfaces from a carbon nanotube based composite coating. Appl. Phys. A: Mater. Sci. Process. 2010, 98 (2), 275–280. (18) Banerjee, S.; Wong, S. Rational sidewall functionalization and purification of single-walled carbon nanotubes by solution-phase ozonolysis. J. Phys. Chem. B 2002, 106 (47), 12144–12151. (19) Mawhinney, D.; Naumenko, V.; Kuznetsova, A.; Yates, J.; Liu, J.; Smalley, R. Infrared spectral evidence for the etching of carbon nanotubes: Ozone oxidation at 298 K. J. Am. Chem. Soc. 2000, 122 (10), 2383–2384. (20) Sham, M.; Kim, J. Surface functionalities of multi-wall carbon nanotubes after UV/ozone and TETA treatments. Carbon 2006, 44 (4), 768–777. 9010

dx.doi.org/10.1021/la201841m |Langmuir 2011, 27, 9005–9011

Langmuir

ARTICLE

(21) Liu, H.; Zhai, J.; Jiang, L. Wetting and anti-wetting on aligned carbon nanotube films. Soft Matter 2006, 2 (10), 811–821. (22) Liu, H.; Li, S.; Zhai, J.; Li, H.; Zheng, Q.; Jiang, L.; Zhu, D. Selfassembly of large-scale micropatterns on aligned carbon nanotube films. Angew. Chem., Int. Ed. 2004, 43 (9), 1146–1149. (23) Kuznetsova, A.; Mawhinney, D.; Naumenko, V.; Yates, J.; Liu, J.; Smalley, R. Enhancement of adsorption inside of single-walled nanotubes: Opening the entry ports. Chem. Phys. Lett. 2000, 321 (3 4), 292–296. (24) Socrates, G. Infrared and Raman Characteristic Group Frequencies: Tables and Charts, 3rd ed.; Wiley: Chichester, U.K., 2001. (25) Kim, U.; Liu, X.; Furtado, C.; Chen, G.; Saito, R.; Jiang, J. Infrared-active vibrational modes of single-walled carbon nanotubes. Phys. Rev. Lett. 2005, 95 (15), No. 157402. (26) Peng, Y.; Liu, H. Effects of oxidation by hydrogen peroxide on the structures of multiwalled carbon nanotubes. Ind. Eng. Chem. Res. 2006, 45 (19), 6483–6488. (27) Naseh, M. V.; Khodadadi, A. A.; Mortazavi, Y.; Pourfayaz, F.; Alizadeh, O.; Maghrebi, M. Fast and clean functionalization of carbon nanotubes by dielectric barrier discharge plasma in air compared to acid treatment. Carbon 2010, 48 (5), 1369–1379. (28) Jung, Y.; Bhushan, B. Dynamic effects of bouncing water droplets on superhydrophobic surfaces. Langmuir 2008, 24 (12), 6262–6269. (29) Chen, C.; Ogino, A.; Wang, X.; Nagatsu, M. Plasma treatment of multiwall carbon nanotubes for dispersion improvement in water. Appl. Phys. Lett. 2010, 96 (13), No. 131504. (30) Kim, Y.; Ito, Y.; Tadai, K.; Mitani, T.; Kim, U.; Kim, H. Drastic change of electric double layer capacitance by surface functionalization of carbon nanotubes. Appl. Phys. Lett. 2005, 87 (23), No. 234106. (31) Stalder, A.; Kulik, G.; Sage, D.; Barbieri, L.; Hoffmann, P. A snake-based approach to accurate determination of both contact points and contact angles. Colloids Surf., A 2006, 286 (1 3), 92–103.

9011

dx.doi.org/10.1021/la201841m |Langmuir 2011, 27, 9005–9011