Isotropy in electrical properties is a desirable trait for electronic materials, allowing current to experience the same resistance regardless of the pathway it takes. In order to obtain isotropic electrical properties, one- and two-layers of multiwalled carbon nanotubes were deposited from aqueous dispersion onto two filter papers of small (5 μm) and large (25 μm) pore sizes using dropcasting. Two different drying arrangements were used wherein the composite material was placed on a surface which was hotter than the conditions around it in either a 65°C heater or at ambient conditions. The time of exposure or the temperature of the surface was varied. It was found that the resistance at the surface and through the thickness could be made equal within experimental error at two deposited layers when placed on a hot surface at 95°C with the surroundings at room temperature for 10 minutes or when placed on a hot surface at 95°C with the surroundings at 65°C for 10–15 minutes.
Paper is ubiquitous in everyday life with applications spanning from the ordinary, such as the vehicle for drafting a research article, to the complex, such as transformer boards.1 In the electronics industry, paper’s usage is often limited to the role of the dielectric or separator material. Because of the low cost, flexibility, abundance, and wide variability in the structure of paper, there has been increasing interest in using the paper’s architecture as part of the active electronic material through the addition of a conducting filler material.2 Still, in many of these cases, the focus of the research and application are on the properties of the surface coating, relegating the paper to be a sample support.3 It is possible to use the paper backbone as part of the active electronic material by utilizing the structure of the paper such that during the fabrication step, the filler material deposits both on the surface and within the thickness of the paper.
Previous research has been done on paper-based composite materials made via dropcasting dispersions of multiwalled carbon nanotubes (MWNT) onto paper substrates and drying via vacuum filtration. While exceptional electrical properties can be obtained (sheet resistances as low as 100 Ω/sq.), when either the number of layers (1–20) or dispersion concentration (0.1 to 5 mg/mL) is varied, the properties measured on the surface (in-plane) and through the thickness (thru-plane) of the composite material do not match when vacuum filtration is used.4 By decreasing the degree of MWNTs pulled through the paper by using a unidirectional heating method, isotropy in the electrical properties of the composites may be able to be achieved with only a few number of deposited layers. In this study, MWNTs are deposited onto and into two paper backbones which vary only by pore size (5 μm vs. 25 μm) from aqueous dispersion and dried using unidirectional heating methods. By using a hot zone below the composite, the MWNT dispersion dries in the direction of the hot zone and is pulled into the paper.5,6
MWNTs were acquired from cheaptubes.com with the following specifications: 0.5–2.0 μm in length, 8–15 nm in diameter, >95 wt% purity with <1.5 wt% ash. The MWNTs were added to DI water at a concentration of 1 mg/mL and sonicated for 10 minutes to begin the dispersion process. Over the next 10 minutes, sodium dodecylbenzenesulfonate (SDBS) surfactant was slowly added to the mixture while it is undergoing ultrasonication until 10 mg/mL concentration is reached. The mixture is then ultrasonicated for another 60 minutes and finally magnetically mixed overnight to ensure that the MWNTs are completely dispersed.
Once a stable dispersion is achieved, the composites are formed by dropcasting 150 μL per layer of the MWNT dispersion onto qualitative filter papers obtained from vwr.com: (1) 413 paper with a 5 μm pore size; (2) 415 paper with a 25 μm pore size. The composites are dried using one of the two methods shown in Figure 1 which are modified drying techniques from Refs. 5 and 6. Figure 1(a) shows the composite being placed in a 65°C heater on top of a ceramic heating board which can be heated to temperatures higher than the heater. Figure 1(b) shows the composite being placed on a temperature-controlled hot plate with room temperature surroundings.
Figure 1. Schematics of the different drying methods used in this study. The sample shows the MWNT dispersion (black) deposited on top of a paper substrate (white). (a) The composite is placed in a heater at 65°C on top of a ceramic heating board which can be varied in temperature. (b) The composite is placed on top of a temperature-controlled hot plate in ambient surroundings.
Both time of exposure or the temperature of the heating board/hot plate can be varied, allowing for a maximum of four drying methodologies. If time of exposure is varied, the ceramic heating board or the temperature controlled-hot plate are put at 95°C and time for drying is varied from 5 minutes to 60 minutes (called 65/95 for the ceramic heating board samples and RT/95 for the hot plate). For the temperature variation studies, the ceramic heating board is varied in ten degree increments from 65°C to 125°C (denoted as 65/T) and the hot plate is varied in ten degree increments from room temperature to 80°C (denoted as RT/T). The composite is dried for 5 minutes. In the case of two-layer composites, the first layer is dried before the second one is deposited.
The electrical properties were measured using impedance spectroscopy using a Solartron 1296 connected to the Solartron 1260. To measure the in-plane properties, a four point probe stage with spacing of 1.5875 mm was used. The high current and high voltage cables were shorted together and the low current and low voltage cables were shorted together such that only two probes were used for the measurement. Frequency was varied from 21 MHz to 1 Hz using an AC voltage of 0.1 V. To measure the thru-plane properties, the same impedance equipment was used, but the sample was placed in a parallel plate electrode arrangement. Frequencies were scanned from 10 MHz to 0.01 Hz using an AC voltage of 4 V. Values of resistance reported are those corresponding to the lowest frequency. Each film was measured at least three times. A Celestron PentaView 5 MP LCD Digital Microscope was used to take 10x magnification images of the macroscale surfaces of the paper and the different composites. Figure 2 shows images of the paper surfaces (a,b), a 1 layer composite formed from the 65/95 technique (c,d), and a 2 layer composite formed from the 65/95 technique (e,f) for the 413 (left) and 415 (right) papers, respectively
Figure 2. Optical micrographs for the plain papers and 1 and 2 layer samples dried using the 65/95 drying technique. (a) blank 413 paper; (b) blank 415 paper; (c) 1 layer deposited and dried on the 413 paper; (d) 1 layer deposited and dried on the 415 paper; (e) 2 layers deposited and dried on the 413 paper; (f) 2 layers deposited and dried on the 415 paper.
In a previous study, the presence of a hot zone had been shown to influence the pulling of MWNTs into the thickness of the paper backbone. By measuring the MWNT-MWNT junction resistance of the film, which is formed on the surface of the paper backbone, it was shown that the hot zone leads to an improvement of the junction resistance as compared to samples made without the hot zone.6 However, the thru-plane properties were not reported; instead, a combination of scanning electron microscopy images and impedance spectroscopy showed the differences in the created surfaces and the electrical properties of the junctions.6 The research reported here was done to resolve whether temperature gradient between the surroundings and the hot board or the time of exposure had a greater impact on the measured properties. Both the in-plane and thru-plane properties are reported.
1-layer of MWNTs was deposited onto both the 413 and 415 papers and dried using the two different methods described in figure 1 through varying either the time of exposure or the temperature difference between the hot zone and the surrounding environment. The measured properties of the different composites made from 1-layer of MWNTs are shown in figure 3. Figures 3a and 3b show the properties of the heater-dried composites where 3a shows the results for the 65/95 study and 3b shows the results for the 65/T study. Figures 3c and 3d show the results for the hot plate-dried composites where 3c shows the results for the RT/95 study and 3d shows the results for the RT/T study. For all graphs, the in-plane resistance is plotted with closed data points and the thru-plane resistance is plotted with open data points. The square data points correspond to the results of the MWNT-413 paper composites while the circle data points correspond to the results of the MWNT-415 paper composites. The x-axis of time describes the time of drying for the composite. The x-axis of temperature gradient describes the difference between the temperature of the hot zone and the temperature of the surroundings.
Figure 3. The in-plane (solid data points) and thru-plane (open data points) properties measured for the 413 (square data points) and the 415 (circle data points) of the 1-layer composites formed via the (a) 65/95, (b) 65/T, (c) RT/95, and (d) RT/T drying techniques.
From the 1-layer results, it can be seen that the thru-plane measurements are consistently lower in resistance (by 2 to 5 orders of magnitude) than the in-plane for all of the fabricated composites suggesting that the pore structure of the paper leads to a greater degree of interconnection through the paper than it aids in the formation of a surface film. Additionally, the variation in time or the temperature gradient did not seem to impact the in-plane properties; however, increasing the time of exposure did cause a gradual decrease in the measured resistance of the thru-plane properties for both the experimental setup indicating an increased amount of MWNTs penetrating through the paper as time increases. In all cases, the error bars are imperceptible due to the large range on the y-axis as well as the small relative error of each sample. The small errors are indicative of a sample of uniform deposition around the measurement area.
Because the in-plane and thru-plane properties never converged for the 1-layer composites regardless of the drying methodology, a layer study was conducted to see where the electrical properties of the two directions overlap. In this study, the 65/95 drying technique was used to deposit multiple layers of MWNTs (from 1–20) onto the 413 paper. Each layer was dried for 5 minutes before the next layer was added. Figure 4 shows the percolation curves plotting resistance as a function of the number of MWNT layers for the in-plane (closed triangles) and the thru-plane (open triangles). There is an intersection between the in-plane and the thru-plane, as denoted by the circle in figure 4, between 2 and 3 deposited layers, at 106–107 Ω. After that point, given the same number of layers, the thru-plane is always more resistive than the in-plane. The thru-plane also decreases in resistance at a much slower rate than the in-plane suggesting that as layers are added, fewer MWNTs are being pulled into the paper and are instead depositing at the surface. The increase in the resistance of the thru-plane after 12 layers is likely due to increased surfactant retention within the composite as more layers are added.
Figure 4. Percolation curves showing resistance vs. layers for the in-plane and thru-plane properties of MWNT networks deposited on the 413 paper using the 65/95 drying method at 5 minutes per layer.
The intersection point for the in-plane and thru-plane resistance on the percolation graph occurs between 2 and 3 deposited MWNT layers on the 413 paper using the 65/95 drying technique (at 5 minutes drying time per layer). There was also an improvement in the thru-plane properties when time was varied for the 1-layer experiments. These two results lead to 2-layers of MWNTs to be deposited onto the 413 and 415 papers using both drying setups with time variation in order to find the conditions at which isotropy in the electrical properties is achieved. Figure 5 shows the results of the RT/95 (5a) and the 65/95 (5b) time variation studies as well as optical micrographs of the surfaces of 2-layered composites fabricated using the RT/95 methodology and dried at 5 minutes per layer on the 413 (5c) and the 415 (5d) papers and dried at 60 minutes per layer on the 413 (5e) and the 415 (5f) papers.
Figure 5. Results and images from the 2-layer deposition studies. (a) Resistance vs. time per layer for the in-plane (open) and the thru-plane (closed) for composites formed on the 413 (square) and 415(circle) papers using the RT/95 experimental design. (b) Resistance vs. time per layer for the in-plane (open) and the thru-plane (closed) for composites formed on the 413 (square) and 415(circle) papers using the 65/95 experimental design. (c) Optical micrograph of 2-layer film for the RT/95 dried at 5 minutes per layer on the 413 paper. (d) Optical micrograph of 2-layer film for the RT/95 dried at 5 minutes per layer on the 415 paper. (e) Optical micrograph of 2-layer film for the RT/95 dried at 60 minutes per layer on the 413 paper. (f) Optical micrograph of 2-layer film for the RT/95 dried at 60 minutes per layer on the 415 paper.
For the RT/95 study, there is an intersection at ~8×105 Ω between the in-plane and the thru-plane electrical properties at 10 minutes of drying per deposited MWNT layer on the 413 paper. Prior to this point, the in-plane and the thru-plane resistances are an order of magnitude different. After this point, the degree of separation between the in-plane and thru-plane resistance grows and are eventually separated by 3 orders of magnitude for the long drying times per layer. As drying time increases, the in-plane becomes more and more resistive, matching the corresponding 1-layer properties. Also of note is that as drying time increases, the thru-plane begins to increase in resistance as well. It seems that at high drying times, the composite is breaking down both on the surface and through the thickness of the 413 paper, resulting in either some kind of charring of the composite or massive loss of the MWNTs during drying. The difference in surface on the 413 paper can be observed in figures 5c and 5e which show the in-plane structure for films dried at 5 minutes per layer and at 60 minutes per layer, respectively. The dark regions correspond to regions of MWNTs on the surface of the paper and are observed much more prevalently in figure 5 c than in 5e.
For the 2-layered composites made on the 415 paper, intersection between the in-plane and thru-plane does not happen, but the properties do remain relatively stable with increasing time. This can be seen visually in figures 5d and 5f which show the surfaces of the 415 paper for 2-layered composites dried at 5 minutes and 60 minutes per layer, respectively. The structures presented in these figures are very similar and therefore, the electrical properties presented for the in-plane in figure 5b are also similar for each composite. While the intersection point for the 65/95 study is slightly higher in time (between 10 and 15 minutes per layer) and resistance (~7×106) than the RT/95 study, the trends experienced by the samples are the same. The higher resistance value for the intersection point of the 65/95 sample than the RT/95 sample suggests that the increased temperature gradient does play a role in isotropic property formation for 2-layer composites at low time scales. Processing can be done at room temperature conditions.
Achieving isotropy in electrical properties allows for the resistance that electrical current experiences to be path independent. It is possible to achieve isotropy between the in-plane and the thru-plane electrical properties of MWNT-paper composite materials by employing a unidirectional drying method in which the deposited MWNTs are pulled into the paper due to a temperature difference between a hot region below the composite and its surroundings. In this study, it was found that isotropy could be reached for composites made up of 2-layers of MWNTs deposited onto the 413 paper using either a temperature-controlled hot plate in ambient surroundings or by using a ceramic heating board in a heated furnace.
For the RT/95 experiment, isotropy was reached at 10 minutes drying time per layer while for the 65/95 experiment, isotropy occurred somewhere between 10 and 15 minutes drying time per layer. The MWNT/415 paper composites did not reach isotropy for any of the experimental conditions as 2 layers of MWNTs were not sufficient to lower the in-plane resistance as the pore network of the 415 paper results in more MWNTs being pulled into the paper structure than depositing on the surface.
The authors would like to acknowledge NSF DMR-1207323 for project funding and support.
1. D.J.T. Hill, T.T. Le, M. Darveniza, and T. Saha, “A study of degradation of cellulosic insulation materials in a power transformer, part 1. Molecular weight study of cellulose insulation paper,” Polymer Degradation and Stability, 48, (1995), 79–87.
2. D. Tobjork and R. Osterbacka, “Paper Electronics,” Advanced Materials, 23, (2011), 1935–1961.
3. L. Hu, J.W. Choi, Y. Yang, S. Jeong, F. La Mantia, L.-F. Cui, and Y. Cui, “Highly conductive paper for energy storage devices,” PNAS, 106, (2009), 21490–21494.
4. R.L. Muhlbauer and R.A. Gerhardt, “Determining In-plane and Thru-plane Percolation Thresholds for Carbon Nanotube Thin Films Deposited on Paper Substrates Using Impedance Spectroscopy,” Mater. Res. Soc. Symp. Proc., 1549, (2013), DOI: 10.1557/opl.2013.1032.
5. L. Huang, K. Chen, C. Peng, and R.A. Gerhardt, “Highly conductive paper fabricated with multiwalled carbon nanotubes and poly(3,4-ethylenedioxylthiophene)-poly(styrenesulfonate) by unidirectional drying. J. Mater. Sci., 46, (2011), 6648–6655.
6. R.L. Muhlbauer, S.M. Joshi, and R.A. Gerhardt, “The effect of substrate pore size on the network interconnectivity and electrical properties of dropcasted multiwalled carbon nanotube thin films,” J. Mater. Res., 28, (2013), 1617–1624.