Article pubs.acs.org/Langmuir
Controlled Orientation and Alignment in Films of Single-Walled Carbon Nanotubes Using Inkjet Printing Simon T. Beyer* and Konrad Walus Department of Electrical and Computer Engineering, The University of British Columbia, British Columbia, Canada S Supporting Information *
ABSTRACT: An inkjet printing procedure for depositing films of carbon nanotubes (CNTs) that exhibit a very high degree of long-range mutual alignment as well as a controlled orientation with respect to the printed geometry is presented. CNT self-assembly was induced by the intrinsic lyotropic liquid crystallinity of CNT suspensions. Sufficient concentrations are reached by matching the inkjet deposition rate to the numerically modeled local evaporation rate of the printed feature and enable the CNT suspension to be printed using standard inkjet printing. Surface alignment was verified using scanning electron microscopy (SEM) and polarized light microscopy. In addition, the bulk morphology was investigated and found to be composed of stacked planar layers that did not necessarily have the same long-range orientation found on the surface. The bulk morphology was characterized by removing layers through an elastomeric peeling process and by observing cross sections of the films using SEM. CNT concentration and length were spanned experimentally, and it was found that very short and very long CNTs as well as low concentration suspensions did not yield long-range alignment.
■
INTRODUCTION Since their discovery two decades ago, carbon nanotubes (CNTs) have been the subject of intense research interest. Their extraordinary intrinsic properties are well documented, and their potential applications are vast, ranging from electronics to biotechnology. To date, exquisite control over the placement and orientation of CNTs in printed films has remained a challenge. Such films have significant potential for applications in large-area electronics and sensors, and their realization using printing can facilitate the low-cost fabrication of such devices. There is evidence suggesting that films of aligned nanotubes, all oriented in the same direction, exhibit superior characteristics to films composed of randomly oriented CNTs.1−3 Significant results have already been reported toward this objective, particularly with suspension-based deposition methods.3−5 However, some of these methods require that the CNT suspension sit undisturbed for long periods of time,4,6 while others have not demonstrated macroscopic pattern control.7−9 Here we show that films of self-assembled, aligned, and controllably oriented CNTs may be implemented using a single, inexpensive and high throughput process: inkjet printing. Inkjet printing is an increasingly popular drop-ondemand deposition technology that is particularly well suited for patterning nanoparticles from suspension. The method described herein utilizes the lyotropic liquid crystallinity of highly concentrated CNT suspensions and the natural forces present in solvent evaporation. Previous reports have demonstrated that upon reaching a critical concentration, φc, © 2012 American Chemical Society
CNTs suspended in solvent experience a phase transition from an isotropic suspension to a liquid crystal (LC).4,6,8,10−12 The concentration required for this transition is largely a function of the CNT aspect ratio and has been shown by others to occur in the range of 1−4 wt %.12 Such a phase transition is predicted by LC theory of rigid rods and was explained as early as 1949 by Onsager.13 Furthermore, theory predicts an intermediate phase at a bulk concentration φi < φc, at which a thin LC film wets the surface of the still isotropic bulk suspension. This surface crystallization is found under the condition that the suspended particles are stable on the surface.14 Accumulation of particles to a liquid−vapor interface is a well-known phenomenon in colloidal and polymer chemistry, often referred to as “crusting”.15 Macromolecules and small particles are more thermodynamically stable at the interface and tend to accumulate there, increasing the local concentration. Similar supporting experimental work on lyotropic LC systems showed that ordered phases were more thermodynamically stable at the liquid−vapor surface.16 These earlier results suggest that the film surface may have a morphology that is different than that of the bulk. To the best of our knowledge, the subsurface morphology has not yet been investigated in aligned CNT films. Many of the prior studies investigating liquid crystallinity in CNTs rely solely on surface characterization techniques such as polarized Raman spectroscopy, scanning electron microsReceived: February 22, 2012 Revised: May 1, 2012 Published: May 9, 2012 8753
dx.doi.org/10.1021/la300770b | Langmuir 2012, 28, 8753−8759
Langmuir
■
copy (SEM), atomic force microscopy, and reflective polarized light microscopy. As such, these measurements do not provide direct insight into the subsurface properties. The results presented here show that the quality of alignment and long-range orientation of the CNTs is exceptional across the surface and is not necessarily reflected deeper into the bulk. This was determined by comparing SEM micrographs and reflective polarized light microscope (PLM) images of the film before and after the removal of a thin surface layer, peeled off using a polydimethylsiloxane (PDMS) stamp. Additionally, cross sections of the films were investigated using SEM and showed a stacked layer structure suggestive of smectic ordering. To realize such films, we have developed a simple way of controllably increasing the CNT concentration during the inkjet printing process. While it is possible to simply create a highly concentrated suspension of CNTs using surfactants prior to printing,11 the viscosity of such a suspension is well beyond the capability of inkjet printing technology. However, it may be possible to directly deposit the viscous suspension using other printing techniques. It was also found that the movement of the receding liquid edge in both a semicylindrical line and spherical cap was sufficient to establish long-range orientation of the CNTs in the film. Since this edge is defined by the geometry of the printed pattern, we show that it is possible to control the orientation of the aligned films. Such a flexible degree of pattern control has never before been demonstrated. A number of attempts at achieving long-range alignment using solvent evaporation have already been reported,7−9,17 including an example utilizing inkjet printing.18 Most reported experiments of this system rely on the same alignment mechanism: the internal hydrodynamic flow in an evaporating droplet. When a droplet of liquid evaporates, it does so in one of two basic modes: (1) with a pinned perimeter and decreasing contact angle or (2) with a receding perimeter and constant contact angle.19 These will be referred to as pinned mode and receding mode, respectively. In colloidal systems, evaporation shows hysteresis between advancing and receding contact angles, initially proceeding in the pinned mode followed by a transition to the receding mode once the receding contact angle is reached. The pinned mode is most notably characterized by the existence of an internal hydrodynamic flow directed outward from the center of the droplet. This flow acts to replenish liquid lost at the outermost edges due to evaporation and carries with it any suspended particles often resulting in the well-known coffee ring deposition pattern.20−23 If the droplet consists of a dilute suspension of CNTs, many of the CNTs will be shuttled toward the droplet edge, increasing the local concentration there. If it is increased up to or beyond φc, the CNTs near the edge may transition to a LC phase and will also align parallel to the edge.9 The parallel orientation may be explained most simply by the development of a flow-induced torque on the CNT as one of its ends becomes pinned by the contact line. The first CNTs that reach the edge may align in this manner and act to seed the LC director as the concentration increases over time. A shortcoming of this approach is that the CNT concentration at the onset of the receding mode is well below φc; thus, CNTs deposited throughout the inner region of the film possess a random orientation.8,18 To obtain complete film alignment, the bulk CNT concentration needs to be increased beyond φc.
Article
EXPERIMENTAL SECTION
Once a pattern such as a semicylindrical line has been printed but has not yet reached the receding mode of evaporation, the CNT concentration may be increased by the continued addition of CNT ink at the same rate as the evaporation rate of the liquid pattern. If the deposition and evaporation rates are well matched, the size and shape of the pattern will remain stationary while the concentration of CNTs will continually increase. However, it was found that the ink cannot simply be deposited at one location as this would result in local slippage of the contact line and bulging of the printed pattern. In addition, areas furthest away from the deposition point will begin to recede. Deposition must therefore be performed across the surface of the printed pattern. The evaporation rate is spatially varying and, in the case of a semicylindrical line, is enhanced toward the ends of the line. In order to improve printing yield, it was necessary to model the evaporation profile of the printed feature and tailor the deposition profile accordingly. The evaporation model used in this study was implemented and evaluated using a finite-element analysis method within the software COMSOL Multiphysics. The model was developed following the work of Hu and Larson21 and is the solution to the time invariant diffusion equation, ∇2C(r) = 0. This equation is used under two assumptions. The first is that the process is diffusion limited, meaning that the evaporation rate of molecules from the liquid is greater than their diffusion away from the surface. The consequence of this is that a saturated vapor layer surrounds the liquid feature, and the resulting equilibrium allows the diffusion equation to be used. This assumption has been shown to be a reasonable one under typical room conditions.21 The second assumption is that the geometry of the liquid feature is invariant in time, in which case the diffusion equation becomes time invariant. Because liquid is continually being added to the feature for the purpose of maintaining its geometry, this is also found to be a reasonable assumption. The boundary conditions for the model are determined by measuring the local temperature and humidity in the printing apparatus. The surface of the liquid feature is fully saturated and the air at the boundaries has a relative humidity that is measured using a temperature/humidity sensor (Sensirion). The size of the box encapsulating the liquid pattern was chosen to be 20 times the pattern length on each side. The evaporation flux into the substrate is zero as set by the boundary condition ∇C = 0. The local J (r,t), can be found using evaporation flux at the liquid−air interface, ⇀ Fick’s first law, and the total evaporation rate over a segment of the liquid surface, Ji, can be found through integration using the relation Ji =
∫S (⇀J ·⇀n ) dS i
n is a normal vector to the liquid surface and Si=1,2,3,...,10 is a where ⇀ section of the liquid surface. The entire liquid surface is divided into 10 segments, and the local evaporation rate is evaluated for each segment. The boundary conditions are depicted in Figure 1a. The experimental parameters were initialized to reflect the current environmental conditions, and the model was evaluated prior to each experiment. The evaporation flux profile and corresponding deposition rate for a typical experiment may be seen in Figure 1b. Inkjet printing was performed on a custom-built printer containing a single 60 μm orifice diameter, piezoelectric nozzle (Microfab). The peak-to-peak voltage used to actuate the nozzle was 35 V using a bipolar pulse. We used poly(ethylene terephthalate) (PET) as the substrate material, which was treated with oxygen plasma for 10 s immediately prior to use to reduce substrate hydrophobicity. Line patterns were composed of individually printed droplets, ∼50 μm in diameter upon ejection, spaced 75 μm apart. The overall pattern consisted of 1000 drops, forming a 0.75 mm by 7.5 mm rectangle. Specifically, the line pattern was created by rasterizing 10 individual lines in sufficient proximity that they merged forming a single liquid feature as shown schematically in Figure 2a. This pattern was printed five times to accumulate a suitable amount of liquid and ensure total connectivity of the liquid pattern. At this point deposition proceeded at a rate matching the modeled evaporation for another 15−60 layers, 8754
dx.doi.org/10.1021/la300770b | Langmuir 2012, 28, 8753−8759
Langmuir
Article
also at 0.5 wt %. The suspension was then sonicated using a high power probe ultrasonicator (Heat Systems Inc.) and centrifuged to remove any remaining agglomerates. The effect of length distribution on alignment was investigated by varying ultrasonication power and centrifugation time and speed.24 Short CNTs were isolated in solution by first ultrasonicating for 20 h at 120 W followed by ultracentrifugation at 170000g for 3 h, where g is gravitational force. The top 50% of the supernatant was extracted. Next we obtained a suspension of longer CNTs by ultrasonicating for 20 h at 20 W followed by 2 h of centrifugation at 20000g, taking the top 80% of the supernatant. Lastly, an effort was made to minimize scission by sonicating in a low power bath sonicator (Branson) for 20 h followed by centrifugation at 20000g for 2 h, taking the top 80% of the supernatant. These three CNT samples will be referred to as short, medium, and long, respectively, for the remainder of this paper. The average lengths of CNTs in each of these three regimes, as taken from an analysis of SEM images of 0.005 wt % suspensions, are 131, 337, and 787 nm with standard deviations of 60, 190, and 360 nm, respectively. The printed films were studied using a field emission SEM (Hitachi S4700) and with a PLM in reflective mode (Nikon Eclipse E600). Dried films were rinsed in an ethanol bath for ∼10 min prior to imaging to remove any surfactant. Samples on PET being imaged with the SEM were sputter-coated with 3 nm of Cr and surrounded with conductive carbon paste to reduce the effects of charging. For the peeling procedure, a PDMS (Sylgard) stamp was used. The PDMS was mixed at a 10:1 base to hardener ratio, degassed, and cast into a Petri dish where it was then cured at 65 °C for 3 h.
Figure 1. (a) Schematic showing the boundary conditions of the COMSOL model. CS indicates the saturated vapor concentration on the surface of the liquid, CA is the ambient vapor concentration, and L is the length of the liquid line. (b) The spatially varying deposition rate along the length of the line. The inset shows the evaporation profile as evaluated by the model, and the bar graph indicates the deposition rate associated with each segment of the inset.
■
RESULTS AND DISCUSSION As shown in the SEM micrographs in Figure 3a−f, the CNTs on the surface of the inkjet printed line are highly aligned and possess an orientation that is parallel to the geometrical edge of the line. Figure 3g shows a PLM image of the same printed line oriented parallel to the microscope analyzer, and Figure 3h shows the line rotated by 45°. The entire region becomes bright at 45°, indicating that the CNTs in that area are oriented parallel to the edge of the line, in agreement with the SEM results. It can be noted that the PET substrate is birefringent, though this has no influence on the results as the CNT films are highly opaque and the PLM uses reflected light. Additionally, it can be seen that the edge of the line is bright in both Figure 3g and Figure 3h. This suggests that some of the CNTs along the periphery have a 45° orientation with respect to the edge of the line. Indeed, this behavior can be seen in Figure 3a. The evaporation behavior of the line initially proceeded in the pinned mode followed by depinning of the ends and lastly by inward recession of all edges, as depicted in Figure 4. The orientation of the CNTs was found to be parallel to the geometrical edge of the line whether the edge was pinned or receding. This behavior is observed in other geometries as well, for example in the drop-cast circular droplet deposited on copper, shown in Figure 5. Figure 6 shows a PLM image of a 2 μL drop-cast circular droplet on untreated PET. This film displays a striking birefringent pattern consisting of an outer ring and a vibrant inner circle, each showing the characteristic dark cross of circular alignment. SEM images of both regions may be found in the Supporting Information and show CNT alignment on the surface in both regions. This unique inner region was not apparent in the printed lines. Both copper and untreated PET are more hydrophobic than plasma treated PET and allow a greater volume-to-footprint ratio, enabling a considerable concentration enhancement as the droplet evaporates. We did not find it possible to inkjet print stable, high aspect ratio patterns on copper or untreated PET. Long
Figure 2. Rasterized nozzle path taken when inkjet printing: (a) the first 5 layers and (b) up to 60 additional layers. The nozzle velocity is constant; therefore, droplet spacing is varied to modify the local deposition rate. Each circle represents a single droplet. corresponding to an overall concentration enhancement of 4−13 times. The frequency of deposition was spatially nonuniform in order to match the local evaporation rate, shown schematically in Figure 2b. Substrates were cooled to 20 °C using a Peltier element to slow the evaporation rate and prevent the regions of the pattern furthest from the nozzle from receding. The amount of cooling required will depend on the geometry of the pattern and the velocity of the nozzle positioning system. Semiconducting (>90%, Southwest Nanotechnologies Inc.) singlewalled carbon nanotubes purchased in powder form were used for all experiments. The CNTs were suspended in distilled water at a concentration of 0.5 wt % using the surfactant sodium cholate (SC), 8755
dx.doi.org/10.1021/la300770b | Langmuir 2012, 28, 8753−8759
Langmuir
Article
Figure 5. SEM images showing the morphology of a circular CNT film. The orientation of the CNTs follows the edge in a circular pattern and persists across the entire surface.
Figure 6. PLM image of a circular CNT film showing the characteristic dark cross of circular orientation in both the outer and inner regions. Regions perpendicular and parallel to the analyzer are dark, whereas regions at 45° are bright.
Figure 3. Surface morphology of an inkjet printed line: (a−f) shows that the CNT orientation across the width of the line is consistently parallel to the edge. The white arrows indicate the local orientation of the CNTs and the black arrows indicate the longitudinal orientation of the line. (g) and (h) show PLM results of the same line oriented parallel to the microscope analyzer (g) and rotated by 45° (h). The cross indicates the orientation of the polarizer and analyzer.
printed lines spontaneously segregate into smaller, lower surface energy segments on these substrates. Evaporation behavior on copper and untreated PET proceeded qualitatively in the same manner as on plasma treated PET. We have recognized and investigated three key variables contributing to long-range alignment: (1) bundling of CNTs, (2) the concentration of CNTs suspended in solution, and (3) the length of the CNTs. First, it was necessary to remove all CNT bundles, agglomerates, and impurities; otherwise alignment was poor or nonexistent. It is thus beneficial to start with as pure a source of CNTs as possible, though measures may be taken to increase purity.25 SEM images of impure and agglomerated CNTs can be found in the Supporting Information. As predicted by LC theory, the mesogenicity of a particle is a strong function of its aspect ratio.13 Zhang et al.26 showed that very short multiwalled CNTs did not form a LC phase while longer ones did and that the quality of the LC was improved when the short CNTs were selectively removed. However, work done by Puech et al.11 suggested that the opposite was true for single-walled CNTs. The difference prompted us to investigate this variable. We found that short CNTs did not undergo a long-range LC phase transition regardless of their concentration. The
Figure 4. Schematic representing the experimentally observed evaporation behavior of an inkjet printed line and the resulting CNT orientation.
8756
dx.doi.org/10.1021/la300770b | Langmuir 2012, 28, 8753−8759
Langmuir
Article
were matched for 20 layers. That is, the pattern was repeated 20 consecutive times using the printing scheme described previously. A significant solute ring formed around the perimeter of the line as seen in the topological profile acquired with an optical profilometer (Wyko, Veeko) and shown in Figure 8. SEM results show clear alignment in the central
morphology of these CNTs may be seen in Figure 7a. Medium length CNTs showed a dramatic improvement in long-range
Figure 8. Topology of a 20-layer printed line segment (inset) and its profile.
region and suggest that the concentration was sufficiently high to undergo a LC phase transition. Doubling the volume deposited (40 layers) did not yield any qualitative difference in the film morphology, and the ratio of CNTs between the coffee ring and center remained relatively constant. In order to investigate the subsurface film morphology, a PDMS stamp was used to peel an ∼80 nm layer off the top of the CNT films, leaving the majority of the film on the substrate. This was done by applying the PDMS to the CNT film and rapidly removing it. PDMS has been previously shown to have an adhesive strength that is proportional to the rate at which it is removed; the faster it is removed, the greater the adhesive force. 29 Scotch tape (3M), wafer dicing tape (Epak Electronics), and thermal release tape (Nitto Denko) also proved capable of removing a thin layer from the CNT film. After removing the aligned surface layer of CNTs we found that the bulk morphology was a combination of aligned and isotropic regions as shown in Figure 9a. This image suggests that some CNTs are only partially lifted at the peeled interface. The ability to peel off thin sheets of CNTs from the bulk film is suggestive of a layered structure, not an entirely random network. An SEM micrograph of the cross section of an aligned film, shown in Figure 9b, shows that the bulk film appears to be composed of stacked layers. In this case no peeling was performed. Most articles on CNT liquid crystallinity conclude that the LC structure is nematic; however, the layered structure observed here is indicative of smectic order and requires further study. Furthermore, while alignment is evident in subsurface layers, the orientation is not necessarily preserved and may not correspond to the geometrical edge of the film, as can be seen in Figure 9c. While the mechanism is not entirely understood, there are a number of potential influences that may be contributing to surface alignment. Liquid crystal theory of rigid rods suggests that lyotropic liquid crystals at an interface will undergo a LC phase transition prior to the bulk as the bulk concentration is increased.14 As explained by Matsuyama and Kato,14 and based on work done by Flory,30 the difference between surface and bulk behavior is a result of a difference in free energy at the two
Figure 7. Typical morphology of (a) CNTs shortened through high power ultrasonication and (b) of long bath sonicated CNTs.
alignment as shown in Figures 3, 5, and 6. Despite the theoretical prediction that mesogenicity should continue to increase with particle aspect ratio, it was found that the long CNTs used in this study did not exhibit a long-range LC phase transition, as shown in Figure 7b. There are a number of possible explanations for this: the longer, more massive CNTs may have a LC phase transition time constant that is much greater than the evaporation time; low power bath sonication may be insufficient to debundle the CNTs; or it may be a result of the flexibility of long CNTs. To investigate the LC phase transition time, we cooled a 5 μL droplet of long CNTs to marginally above the dew point of the local atmosphere (∼16 °C), allowing it to evaporate over ∼8 h. The resulting film did not possess any improvement in alignment, suggesting that if the transition time is the cause, it is too long to be apparent in our experiment. Bath sonication has been shown to be an effective method of suspending and dispersing CNTs,27 suggesting that it is some other property of long CNTs that is limiting the formation of the LC phase. As reported by others, it may be a result of entanglements caused by their increased flexibility.6,11 Our experiments suggest that an intermediate length of CNT is ideal for forming films with long-range alignment. Being a lyotropic LC, the phase transition of the CNT suspension is inherently concentration dependent.28 The exact concentration at which the transition occurs depends on the mesogenicity of the particles, and this property of CNTs is still not fully understood. As expected, films made from dilute CNT suspensions only exhibited alignment along the periphery (Supporting Information). For the inkjet printed lines of medium length CNTs, the evaporation and deposition rates 8757
dx.doi.org/10.1021/la300770b | Langmuir 2012, 28, 8753−8759
Langmuir
Article
initial conditions were the same, we dispersed a sample of each length regime into 15 wt % SC solutions at a 5:1 dilution and mixed them using bath sonication for 10 min followed by a post-evaporation rinse in ethanol. The results showed that short CNTs had a C/Na ratio of 29, and the medium length CNTs had a ratio of 48, suggesting that medium length CNTs have less bound surfactant and may have a greater affinity to the liquid−vapor surface as a result. This surface effect has been scarcely studied with CNTs but holds important implications for CNT films manufactured using their innate liquid crystallinity, namely, that the film properties are likely to be a complex mixture of the properties of isotropic and aligned films of varying orientation.
■
CONCLUSION Utilizing the intrinsic liquid crystal behavior of highly concentrated CNT suspensions, we have realized films of highly aligned CNTs with controlled long-range orientation using solely inkjet printing. We have developed a means of arbitrarily increasing the concentration of the deposited material by matching the deposition and evaporation rates during inkjet printing and have used this method to promote the liquid crystalline behavior of CNTs in suspension, which was found to produce films with an orientation that is governed by the printed pattern. The bulk morphology of these films was explored and found to exhibit stacked planar layers, suggestive of smectic ordering. Planar layers were selectively removed using both adhesive tape and a PDMS stamp. It was further found that the alignment and orientation were most striking on the surface layer with more complex behavior through the depth of the film.
■
ASSOCIATED CONTENT
S Supporting Information *
Optical profilometry images of CNT distribution along printed lines; PLM and SEM images of printed circles with different dimensions and concentrations; SEM images of contaminated and agglomerated CNTs; SEM images of a film made using a dilute CNT suspension. This material is available free of charge via the Internet at http://pubs.acs.org.
Figure 9. Subsurface morphology of an aligned CNT film: (a) shows the film after a single layer has been removed using a PDMS stamp, (b) shows a cross section of the film, and (c) shows a topographical view of a large fissure, revealing several subsurface layers.
■
locations. Additionally, it has been shown that nanoscale particles are very thermally stable at liquid−air interfaces.16 However, the theory assumes that each particle is the same shape and size, which is not the case in our system. Though we have used sonication and centrifugation to narrow the length distributions of the suspensions, we have not taken efforts to unequivocally isolate one single length. Short CNTs with low mesogenicity will inevitably be present in the medium length suspension. Differences in surfactant loading and wrapping between the medium, long, and short tubes could possibly lead to an accumulation of high mesogenicity, medium length tubes at the surface leading to the enhanced surface alignment. Previous research utilizing atomistic simulations of surfactant wrapping on CNTs have studied the surfactants SC31 and sodium dodecyl sulfate32,33 and suggest the possibility of nonuniform surfactant wrapping at different loading values. Differences in surfactant wrapping and loading of long and short CNTs may be caused by preferential surfactant binding at the ends of the tubes. Using energy dispersive spectroscopy, we investigated the ratio of carbon and sodium atoms in the short and medium length regimes. To account for any surfactant gradients resulting from the centrifugation process and ensure
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors acknowledge the financial support of the Natural Science and Engineering Research Council (NSERC) of Canada Strategic Projects Grant, the Canadian Foundation for Innovation (CFI) Leaders Opportunity Fund, the University of British Columbia, and equipment provided by CMC Microsystem. SEM imaging was performed at the UBC BioImaging Facility. The authors also thank John Madden, Boris Stoeber, Edmond Cretu, Vahid Bazargan, and Dansik Yoo for helpful discussions as well as Chris Sherwood for assistance with the ultracentrifugation.
■
REFERENCES
(1) Pimparkar, N.; Kocabas, C.; Kang, S. J.; Rogers, J.; Alam, M. A. Limits of Performance Gain of Aligned Cnt over Randomized
8758
dx.doi.org/10.1021/la300770b | Langmuir 2012, 28, 8753−8759
Langmuir
Article
Network: Theoretical Predictions and Experimental Validation. IEEE Electron Device Lett. 2007, 28, 593−595. (2) Tsukruk, V. V.; Ko, H. Liquid-Crystalline Processing of Highly Oriented Carbon Nanotube Arrays for Thin-Film Transistors. Nano Lett. 2006, 6, 1443−1448. (3) Avouris, P.; Engel, M.; Small, J. P.; Steiner, M.; Freitag, M.; Green, A. A.; Hersam, M. C. Thin Film Nanotube Transistors Based on Self-Assembled, Aligned, Semiconducting Carbon Nanotube Arrays. ACS Nano 2008, 2, 2445−2452. (4) Chen, W.; Lu, L. H. Large-Scale Aligned Carbon Nanotubes from Their Purified, Highly Concentrated Suspension. ACS Nano 2010, 4, 1042−1048. (5) Lagerwall, J. P. F.; Scalia, G. Carbon Nanotubes in Liquid Crystals. J. Mater. Chem. 2008, 18, 2890−2898. (6) Anglaret, E.; Zamora-Ledezma, C.; Blanc, C.; Maugey, M.; Zakri, C.; Poulin, P. Anisotropic Thin Films of Single-Wall Carbon Nanotubes from Aligned Lyotropic Nematic Suspensions. Nano Lett. 2008, 8, 4103−4107. (7) Huang, L. M.; Cui, X. D.; Dukovic, G.; O’Brien, S. P. SelfOrganizing High-Density Single-Walled Carbon Nanotube Arrays from Surfactant Suspensions. Nanotechnology 2004, 15, 1450−1454. (8) Windle, A. H.; Zhang, S. J.; Li, Q. W.; Kinloch, I. A. Ordering in a Droplet of an Aqueous Suspension of Single-Wall Carbon Nanotubes on a Solid Substrate. Langmuir 2010, 26, 2107−2112. (9) Li, Q. W.; Zhu, Y. T.; Kinloch, I. A.; Windle, A. H. SelfOrganization of Carbon Nanotubes in Evaporating Droplets. J. Phys. Chem. B 2006, 110, 13926−13930. (10) Simmons, T. J.; Bravo-Sanchez, M.; Vidal, M. A. Liquid Crystal Behavior of Single Wall Carbon Nanotubes. Carbon 2010, 48, 3531− 3542. (11) Poulin, P.; Puech, N.; Blanc, C.; Grelet, E.; Zamora-Ledezma, C.; Maugey, M.; Zakri, C.; Anglaret, E. Highly Ordered Carbon Nanotube Nematic Liquid Crystals. J. Phys. Chem. C 2011, 115, 3272− 3278. (12) Kumar, S.; Zhang, S. J. Carbon Nanotubes as Liquid Crystals. Small 2008, 4, 1270−1283. (13) Onsager, L. The Effects of Shape on the Interaction of Colloidal Particles. Ann. N. Y. Acad. Sci. 1949, 51, 627−659. (14) Matsuyama, A.; Kato, T. Theory of Surface Ordering on Solutions of Rigid-Rod-Like Molecules. Macromolecules 1995, 28, 131−135. (15) Style, R. W.; Peppin, S. S. L. Crust Formation in Drying Colloidal Suspensions. Proc. R. Soc. London, Ser. A 2011, 467, 174. (16) Braun, C.; Lang, P.; Findenegg, G. H. Surface-Induced Shift of the Hexagonal-to-Isotropic Phase-Transition in a Lyotropic System Studied by X-Ray Reflectivity. Langmuir 1995, 11, 764−766. (17) Strano, M. S.; Sharma, R.; Lee, C. Y.; Choi, J. H.; Chen, K. Nanometer Positioning, Parallel Alignment, and Placement of Single Anisotropic Nanoparticles Using Hydrodynamic Forces in Cylindrical Droplets. Nano Lett. 2007, 7, 2693−2700. (18) Blayo, A.; Denneulin, A.; Bras, J.; Carcone, F.; Neuman, C. Impact of Ink Formulation on Carbon Nanotube Network Organization within Inkjet Printed Conductive Films. Carbon 2011, 49, 2603−2614. (19) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Contact Line Deposits in an Evaporating Drop. Phys. Rev. E 2000, 62, 756−765. (20) Barash, L. Y.; Bigioni, T. P.; Vinokur, V. M.; Shchur, L. N. Evaporation and Fluid Dynamics of a Sessile Drop of Capillary Size. Phys. Rev. E 2009, 79, 046301. (21) Hu, H.; Larson, R. G. Evaporation of a Sessile Droplet on a Substrate. J. Phys. Chem. B 2002, 106, 1334−1344. (22) Petsi, A. J.; Burganos, V. N. Potential Flow inside an Evaporating Cylindrical Line. Phys. Rev. E 2005, 72, 047301. (23) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Capillary Flow as the Cause of Ring Stains from Dried Liquid Drops. Nature 1997, 389, 827−829. (24) Hennrich, F.; Krupke, R.; Arnold, K.; Stutz, J. A. R.; Lebedkin, S.; Koch, T.; Schimmel, T.; Kappes, M. M. The Mechanism of
Cavitation-Induced Scission of Single-Walled Carbon Nanotubes. J. Phys. Chem. B 2007, 111, 1932−1937. (25) Haddon, R. C.; Sippel, J.; Rinzler, A. G.; Papadimitrakopoulos, F. Purification and Separation of Carbon Nanotubes. MRS Bull. 2004, 29, 252−259. (26) Windle, A. H.; Zhang, S. J.; Kinloch, I. A. Mesogenicity Drives Fractionation in Lyotropic Aqueous Suspensions of Multiwall Carbon Nanotubes. Nano Lett. 2006, 6, 568−572. (27) Islam, M.; Rojas, E.; Bergey, D.; Johnson, A.; Yodh, A. High Weight Fraction Surfactant Solubilization of Single-Wall Carbon Nanotubes in Water. Nano Lett. 2003, 3, 269−273. (28) Collings, P. J.; Patel, J. S. Handbook of Liquid Crystal Research; Oxford University Press: New York, 1997; p xv, 600 pp. (29) Rogers, J. A.; Meitl, M. A.; Zhu, Z. T.; Kumar, V.; Lee, K. J.; Feng, X.; Huang, Y. Y.; Adesida, I.; Nuzzo, R. G. Transfer Printing by Kinetic Control of Adhesion to an Elastomeric Stamp. Nat. Mater. 2006, 5, 33−38. (30) Flory, P. J. Statistical Thermodynamics of Semi-Flexible Chain Molecules. Proc. R. Soc. London, Ser. A 1956, 234, 60−73. (31) Blankschtein, D.; Lin, S. C. Role of the Bile Salt Surfactant Sodium Cholate in Enhancing the Aqueous Dispersion Stability of Single-Walled Carbon Nanotubes: A Molecular Dynamics Simulation Study. J. Phys. Chem. B 2010, 114, 15616−15625. (32) Zerbetto, F.; Calvaresi, M.; Dallavalle, M. Wrapping Nanotubles with Micelles, Hemimicelles, and Cylindrical Micelles. Small 2009, 5, 2191−2198. (33) Striolo, A.; Tummala, N. R. Sds Surfactants on Carbon Nanotubes: Aggregate Morphology. ACS Nano 2009, 3, 595−602.
8759
dx.doi.org/10.1021/la300770b | Langmuir 2012, 28, 8753−8759
