This work fabricates an electrically controllable all-optical switch using a dye-doped liquid crystal (DDLC) cell on which is pasted a buckypaper. The buckypaper serves as a thin film resistant heater; it can maintain the temperature of the DDLC cell constant by environmental cooling. The threshold power of the DDLC cell, which is the input optical power at which the output optical power begins to fall, is controlled by applied voltages, and the switch-off power of the cell, which is the minimum input optical power at which the output optical power falls to zero, is also controlled by the applied voltages. Therefore, such a cell can be developed into optical limiting devices with adjustable threshold and switch-off powers for multi-purpose use.
© 2015 Optical Society of America
Dye-doped liquid crystals (DDLCs) have been used to develop all-optical switches [1–4]. The all-optical switches can reduce the power of light if it exceeds a threshold, and allow the light to pass through them if its power is below a threshold. Therefore, such switches are ideal for use in arc-welding helmets and windshields. One of their shortcomings, however, is that they have constant thresholds. If the power of the light is below the threshold but high enough to damage human eyes, then these switches may not protect users’ eyes against high-power light. Although this issue can be overcome using all-optical switches with relatively low thresholds, such switches may cause poor visibility for users. Therefore, researchers want to develop all-optical switches that can eliminate incident light of varying power.
Buckypaper is a thin sheet of an aggregate of carbon nanotubes. Buckypaper is highly efficient at generating and dissipating heat, and the heat generation and dissipation depend on its electrical resistivity and thermal conductivity, respectively [5,6]. Hence, buckypaper, when cooling to the environment, can act as a thin and flexible thermostat in temperature-sensitive devices .
This paper comprises five parts. In the first part, a DDLC cell on which is pasted a buckypaper is fabricated, and this cell is placed between two crossed polarizers to develop an all-optical switch. The second part studies the electrothermal characteristics of the buckypaper, including its efficiencies in generating and dissipating heat, thermal stability, thermal uniformity, and thermal responses. The third part investigates the effect of buckypaper-induced heating on the all-optical switching characteristics of the DDLC cell. A comparative study of DDLC cells with a buckypaper heater (proposed approach, tested sample) and an indium tin oxide heater (a common approach, reference sample) is performed. The final part presents two methods for increasing the transmittances of optical devices on which are pasted opaque buckypaper heaters.
Figure 1 presents the configuration and a photograph of a DDLC cell on which is pasted a buckypaper. The buckypaper is made from multi-walled carbon nanotubes by vacuum filtration, as described elsewhere [8,9]. The area, thickness, electrical resistivity and thermal conductivity of the buckypaper are approximately 17 mm × 15 mm, 25 μm, 1.30 × 10−2 Ω⋅cm and 65 W/m⋅K, respectively. The DDLC mixture is prepared by mixing azo dyes (DR1 from Sigma-Aldrich) with nematic LC (E7 from Merck). The mixing ratio of DR1: E7 in the DDLC mixture is 3:97 by weight, and the clearing point of the DDLC mixture is approximately 58 °C. An empty cell is fabricated using two mutually orthogonal-aligned glass substrates with an area of ~20 mm × 15 mm, which are separated by 4.3μm-thick spacers. The empty cell is filled with the DDLC mixture, and then sealed with epoxy gel. A square hole with an area of ~5 mm × 5 mm is punched in the center of the buckypaper, to enable incident light to pass through the DDLC cell. Finally, the punched buckypaper is pasted onto one of the glass substrates using epoxy glue.
Figure 2 presents the experimental setup for developing an all-optical switch. The cell is placed between two crossed polarizers. The transmission axis of the polarizer is set parallel to the x axis, as presented in Fig. 2, and that of the analyzer is set perpendicular to it. A green light that is emitted from a DPSS laser (λ = 532 nm) is used to pump the azo dye molecules and probe the transmission of the cell. After the linearly-polarized light has passed through a beam splitter with a splitting ratio of 50:50, it is divided into two beams whose diameters are approximately 1.5 mm. One of the beams is incident on the DDLC cell from the side without the buckypaper, and the other is incident on photodiode 2. Hence, the input power (Pin) of the light that is incident to the DDLC cell can be obtained from the optical power that is detected by photodiode 2. After the light passes through the DDLC cell and the analyzer, photodiode 1 detects the output power (Pout) of the transmitted light. The detected Pout is a function of Pin, and the Pout(Pin) curve is plotted.
3. Electrothermal effect of buckypaper
The buckypaper is used to heat the DDLC cell in this experiment. Therefore, the dependence of the cell temperature on applied voltage must firstly be known. A DC power supply (2220-30-1 from Keithley) is utilized to apply voltages to the buckypaper, and a thermal imager (Ti10 from Fluke) is used to measure the cell temperature. The cell temperature is measured at room temperature (T0), 25.5 ± 0.2 °C. Figure 3(a) presents the first three minutes of dynamic measurements of the cell temperature at various applied voltages, and Fig. 3(b) plots the dynamic measurements made over one hour. From Fig. 3(a), the application of voltages to the buckypaper increases the cell temperature by the electrothermal effect of the carbon nanotubes . Additionally, the buckypaper keeps the cell temperature constant owing to the thermal equilibrium that is established between the heating of the buckypaper and its cooling to the environment . The cell temperatures are 25.5 ± 0.2 °C, 33.1 ± 0.2 °C, 40.8 ± 0.2 °C, 48.0 ± 0.2 °C and 55.0 ± 0.2 °C at applied voltages 0 V, 1.43 V, 1.92 V, 2.23 V and 2.48 V, respectively. Since the cell temperature has a small tolerance of ± 0.2 °C at each of the applied voltages, the mean cell temperatures 25.5 °C, 33.1 °C, 40.8 °C, 48.0 °C and 55.0 °C are used in the presentation of the experimental results. The results of Fig. 3(b) reveal that the cell temperatures are stable in an hour at the applied voltages. Therefore, the buckypaper and its environment together function as a thin film thermostat. If the temperature of the environment changes after one hour, the cell temperature can be maintained by changing the applied voltage. The cell temperatures return to room temperature after the applied voltages are removed, as shown in Fig. 3(c).
The on-time (relaxation time) of the buckypaper is defined as that time taken for the rise (drop) of the cell temperature from room temperature (a steady-state temperature) to a steady-state temperature (room temperature). As is estimated from Fig. 3(a), the on-times are approximately 70 s, 100 s, 120 s and 135 s at 33.1 °C, 40.8 °C, 48.0 °C and 55.0 °C, respectively. As is estimated from Fig. 3(c), the relaxation times are approximately 50 s, 80 s, 100 s and 150 s at 33.1 °C, 40.8 °C, 48.0 °C and 55.0 °C, respectively. The on-times and relaxation times increase with the increase in the steady-state temperatures. The on-times can be reduced by using an optimized buckypaper. Efforts are being made at the authors’ laboratory to optimize the performance of the DDLC cell on which is pasted buckypaper, the results of which will be published in the near future.
Figure 3(d) shows the thermal images of the DDLC cell with the buckypaper at the applied voltages. These thermal images indicate that the temperature distributions in the DDLC cell with the buckypaper are uniform at all of the applied voltage. This result follows from the fact that the punched buckypaper heats the edge of the glass substrate, establishing a temperature gradient with a minimum at the center of the hole in the buckypaper in the early stage of heating. The temperature gradient accelerates the thermal conduction throughout the cell because heat flux is proportional to the change in temperature. Accordingly, the cell with the buckypaper exhibits high thermal uniformity. In the right-most thermal image in Fig. 3(d), the temperature is the highest above the center of the DDLC cell. This phenomenon has two causes. The first reason is that the epoxy glue layer between the DDLC cell and the buckypaper is irregular, and the second is that the copper tapas may be pasted roughly onto the buckypaper.
Figure 3(e) plots the dependence of the mean cell temperature () on the applied voltage (V), revealing a non-linear relationship between and V. When a voltage is applied to the buckypaper, a current of 0 mA, 111 mA, 163 mA, 223 mA or 253 mA passes through it at V = 0 V, 1.43 V, 1.92 V, 2.23 V or 2.48 V, respectively. Multiplying the current by the voltage yields the applied electrical power. Figure 3(f) plots the dependence of the mean cell temperature on the applied electrical power (Pe). In Fig. 3(f), the cell temperature is directly and linearly proportional to the applied electrical power. The exact relationship is obtained by best fitting using OriginPro software, and is given byFig. 3(f). Equation (1) easily derives the mean cell temperature from the applied electrical power. Therefore, the buckypaper is likely to be a commercially successful thermostat.
The buckypaper and its environment are used to hold the cell at a constant temperature that exceeds room temperature. Most thin-film thermostats require heat sinks, mechanical fans or water cooling systems to maintain the samples at constant temperature. Such thermostats increase the cost of fabrication and have high electrical power consumption. In our investigation, only buckypaper and its environment can maintain the cell temperature for one hour because the buckypaper has a large specific surface area . Therefore, the buckypaper thermostat is low-cost and consumes relatively little power.
4. Results and discussion
Figure 4 plots the Pout(Pin) curves of the DDLC cell at the various applied voltages. In all curves, Pout increases and then decreases as Pin increases. The increase in Pout is caused by the 90° twisted-nematic alignment and the constant light absorption coefficient of the DDLC cell. After the linearly-polarized light passes through the DDLC cell, the direction of polarization of the transmitted light is parallel to the transmission axis of the analyzer because the experimental conditions are consistent with the Mauguin rule. In addition, the light absorption coefficient of the DDLC cell is constant. Consequently, Pout increases with Pin, as presented in Fig. 4. The decline in Pout is attributable to the isothermal nematic→isotropic phase transition of the LC via the photoisomerization of the DR1 dyes. The rod-like trans-isomers of the DR1 dyes are typically stable in the dark, so the guest-host effect aligns them with the long axes of the LC molecules in the DDLC cell. When the intensity of the incident light is sufficiently high, the rod-like trans-isomers are transformed into bent cis-isomers. The bent cis-isomers disturb the orientation of the LC director in the 90° twisted-nematic DDLC cell, inducing the isothermal nematic→isotropic phase transition of the LC. After the linearly-polarized light passes through the DDLC cell, the direction of polarization of the transmitted light is still parallel to the x axis in Fig. 2 because the LC molecules are in an isotropic phase. The transmitted light does not pass through the analyzer because its direction of polarization is orthogonal to the transmission axis of the analyzer. Therefore, Pout decreases as Pin increases, as presented in Fig. 4. Since Pout increases and then decreases as Pin increases, the DDLC cell can act as an all-optical switch.
To describe the experimental results in Fig. 4, the threshold power (Pth) is defined as the input optical power at which the output optical power begins to fall, and the switch-off power (Poff) is the minimum input optical power at which the output optical power falls to zero. In Fig. 4, Pth and Poff decline as increases. This phenomenon has two causes. The first reason is that the buckypaper-induced heat reduces the required concentration of the cis-isomers, and the second is that buckypaper-induced heating reduces the birefringence of the LC. With respect to the first cause, the threshold is reached when the concentration of the bent cis-isomers is high enough to begin to destroy the mesophase. Switch-off occurs when the concentration of the bent cis-isomers is high enough to make the DDLC mixture isotropic. Preheating the DDLC cell reduces the difference between the cell temperature and the clearing point of the DDLC mixture. A small temperature difference makes the required concentration of the cis-isomers low, resulting in a low threshold power and a low switch-off power. Therefore, Pth and Poff decline as increases. The second cause is as follows. Preheating the DDLC cell reduces the LC birefringence. A low LC birefringence makes the order parameter low, resulting in a low threshold power and a low switch-off power. Therefore, Pth and Poff fall as increases. The second cause is a minor effect. In Fig. 4, all of the curves are superposed on each other before the threshold is reached. This result reveals that although the LC birefringence varies with cell temperature, the output power does not change until the mesophase begins to be destroyed at a particular concentration of the cis-isomers. Therefore, the first cause is the major effect.
The threshold power of the DDLC cell is controlled by the voltage that is applied to the buckypaper. This fact can be exploited in developing an electrically controllable optical limiting device for multi-purpose use. For example, consider a DDLC-based windscreen with adjustable threshold power. If the threshold power of the windscreen is adjusted to be less than the power of direct morning sunlight, then the windscreen can protect the driver’s eyes against it. However, such a windscreen will provide shade against the sunlight at midday, causing poor visibility for drivers, because the power of sunlight is commonly higher at midday than in the morning. To prevent shading by the windscreen, the threshold power of the windscreen must be adjusted to exceed the power of sunlight at midday. Accordingly, the windscreen will allow the sunlight to enter the car, providing good visibility at midday. In Fig. 4, the switch-off power of the DDLC cell is controlled by varying the voltage that is applied to the buckypaper. Therefore, the DDLC cell on which is pasted the buckypaper can eliminate incident light of varying power.
If a DDLC-based all-optical switch without buckypaper is operated at low ambient temperature, then the threshold power, which is inversely proportional to the ambient temperature, will be high. In our experiment, the application of the voltages to the buckypaper reduces the threshold power of the DDLC cell. Therefore, the buckypaper can assist the DDLC cell in switching off the light at low ambient temperature. 5CB LC with a low clearing temperature (approximately 36 °C) are commonly the selected host materials in DDLC cells that are used to fabricate all-optical switches with low threshold power and a fast response [1–3]. These optical switches lose functionality at high ambient temperature (> 36 °C) because the 5CB LCs are in the isotropic phase. The clearing point (approximately 58 °C) of the DDLC mixture in the DDLC cell that is used in the experiment herein is approximately 30 °C above mean room temperature (25.5 °C). Therefore, the dye-doped E7 LC cell can switch off light at high ambient temperature. Briefly, the use of the buckypaper and E7 LC in this work makes the DDLC-based all-optical switch functional over a wide range of ambient temperatures.
Figure 5 plots the dynamic output powers at Pin = Poff (25.5 °C), Poff (40.8 °C) and Poff (55.0 °C). To describe the experimental results in Fig. 5, the response time of optical switching is defined as that time taken for the output power to fall to of its maximum value. In Fig. 5, the response times are approximately 4.4 s, 2.5 s, 0.3 s at = 25.5 °C, 40.8 °C and 55.0 °C, respectively. The response time of the DDLC-based all-optical switch is too long for most applications that are considered herein. The response time of the switch depends on the azo dyes used, the doping concentration of the dyes, and the cell gap. The use of an optimized dye-doped LC cell may reduce the response time of the all-optical switch. Efforts are being made at the authors’ laboratory to optimize the dye-doped LC cell, the results of which will be published in the near future. In Fig. 5, the response time at = 55.0 °C is approximately 1/15th that at = 25.5 °C. This result may be exploited to develop high-speed all-optical switches. In , the DDLC-based all-optical switch that is operated at a wavelength of 750 nm exhibits a short response time of 25 ns in optical switching. If buckypaper is pasted on this switch, then the response time of the optical switching may approach several nanoseconds because a high cell temperature corresponds to a short response time for optical switching.
Indium tin oxide (ITO) electrodes are now extensively used in LC displays. ITO thin films can serve not only as electrodes but also as resistive heaters. ITO heaters have the advantages of low optical loss and high optical uniformity. However, one of the ITO components, indium, is a rare metal and more expensive than gold. A carbon-based material, buckypaper, may be able to replace ITO as a thin film heater. A comparative study of DDLC cells with the buckypaper heater (proposed approach, tested sample) and an ITO heater (a common approach, reference sample) will be carried out. The substrates in the reference sample are normal glass and commercial ITO glass, and the ITO thin film functions as a resistive heater. The tested sample and reference sample have the same configuration while the transparent ITO heater does not require any hole punching treatment, as presented in Fig. 6. The transparent ITO heater outperforms an opaque buckypaper heater in terms of both optical loss and uniformity. In this work, a square hole is punched in the center of the buckypaper, to enable incident light to pass through the DDLC cell. Therefore, the punched buckypaper heater has the same optical loss and uniformity as the ITO heater since there is no buckypaper in the path of the light.
Identical voltages are applied to the two heaters to evaluate their efficiency in heating the DDLC cells. Therefore, the voltages [Fig. 3(a)] that are applied to the buckypaper heater are also applied to the ITO heater. If a given voltage is applied to both heaters, then the one with the lower sheet resistance will yield more heat in a DDLC cell because the consumed electrical energies of the heaters are inversely proportional to their sheet resistances at a constant voltage. A piece of ITO glass (Aimcore Co.) with a sheet resistance of 5 Ω/sq is used in the comparative study. The sheet resistance of the buckypaper heater is also 5 Ω/sq because its electrical resistivity and film thickness are 1.30 × 10−2 Ω⋅cm and 25 μm, respectively. The buckypaper heater and ITO heater had the same sheet resistance, so these two heaters may have the same heating efficiency at a particular voltage.
Figure 7(a) plots the dynamic temperature measurements of the reference sample at various voltages for the first four minutes. From Fig. 7(a), the temperatures of the reference sample are 25.5 ± 0.2 °C, 31.3 ± 0.2 °C, 37.5 ± 0.2 °C, 44.1 ± 0.2 °C and 52.2 ± 0.2 °C at applied voltages 0 V, 1.43 V, 1.92 V, 2.23 V and 2.48 V, respectively. Although the two heaters have the same sheet resistance at room temperature, the mean temperature of the tested sample exceeds that of the reference sample when a particular voltage is applied to both heaters, as presented in Figs. 3(a) and 7(a). This result follows from the fact that the electrical resistivity of the buckypaper falls as the sample temperature increases . Semiconductors also exhibit this characteristic. This property causes the sheet resistance of the buckypaper to decrease as the sample temperature increases. Therefore, the buckypaper heater has greater heating efficiency than the ITO heater.
Estimated from Fig. 7(a), the on-times taken for the rise of the mean cell temperature from room temperature to 31.3 °C, 37.5 °C, 44.1 °C and 52.2 °C are approximately 55 s, 90 s, 130 s and 180 s, respectively. The relaxation times for the drop of the mean cell temperature from 31.3 °C, 37.5 °C, 44.1 °C and 52.2 °C to room temperature, estimated from Fig. 7(c), are approximately 40 s, 70 s, 130 s and 190 s, respectively. The on-times and relaxation times of the ITO heater are smaller than those [Figs. 3(a) and 3(c)] of the buckypaper heater at the low applied voltages of 1.43 V and 1.92 V. This result arises from the fact that low steady-state temperatures yield short on-times and short relaxation times. The on-times and relaxation times of the ITO heater exceed those [Figs. 3(a) and 3(c)] of the buckypaper heater at the high applied voltages of 2.23 V and 2.48 V. Since the buckypaper film has a highly porous three-dimensional (3D) network with a large specific surface area , the thermal conductivity (65 W/m⋅K) of the buckypaper heater exceeds that (10 W/m⋅K) of the ITO heater. Therefore, a thermal equilibrium between the heating of the buckypaper and its cooling to the environment can be rapidly established under an applied voltage, and the heat in the DDLC cell is rapidly dissipated to the environment by the 3D network when the voltage is removed. From the on-times and relaxation times, the buckypaper heater dissipates heat more efficiently than does the ITO heater, especially at high steady-state temperatures.
Figure 7(b) plots the dynamic temperature measurements of the reference sample in one hour at the applied voltages. From Figs. 3(b) and 7(b), the mean temperatures of the two samples are constant for one hour at each applied voltage. Therefore, the two samples exhibit the same thermal stability.
Figure 7(d) presents the thermal images of the reference sample under the applied voltages. The temperature distributions in the reference sample are uniform at all of the applied voltages. In Fig. 3(d), the temperature distributions in the tested sample are uniform at the applied voltages as well. The experimental results in Figs. 3(d) and 7(d) demonstrate that the two samples exhibit the same temperature uniformity.
Unlike ITO, which is sputtered, the buckypaper is pasted to the DDLC cell. The pasted buckypaper may not tolerate very high heat, and it may peel off during welding. The thermal reliability of the DDLC cell with the buckypaper is discussed as follows. Sparks and fume particles are emitted from metal joints during welding, and they carry a very large amount of heat energy. Since the sparks and fume particles have small dimensions between 0.5 and 2 mm, most of their heat is dissipated to the environment . Commercial welding helmets comprise front cover lenses, air gaps and photochromic lenses . As the sparks and fume particles impinge on the welding helmets, they are blocked by the front cover lenses. Air is a bad conductor of heat so the heat that is conducted or convected from the front cover lenses to the photochromic lenses does not damage the photochromic lenses. Consider a DDLC cell on which is pasted a buckypaper using silicone rubber, which is resistant to extreme temperatures from −55 °C to + 300 °C. If the DDLC cell with the buckypaper replaces the photochromic lens of a welding helmet, then the buckypaper will not be peeled off from the cell during welding because commercial welding helmets are designed to allow their front cover lenses to tolerate a maximum temperature of 135 °C . Therefore, the DDLC cell on which is pasted the buckypaper can be used in welding. If a DDLC cell on which is sputtered an ITO thin film is used in welding, the film will tolerate the huge heat from the sparks and fume particles because the melting point of ITO exceeds 1500 °C. Since the sputtered ITO thin film can endure more heat than the pasted buckypaper, the ITO has higher thermal reliability.
A comparative table is presented (Table 1), clarifying the advantages of the punched buckypaper heater over the ITO heater.
Most metal materials exceed the buckypaper in terms of generating and dissipating heat because they exhibit lower electrical resistivities and higher thermal conductivities than the buckypaper. However, oxidization and hydrogen embrittlement may occur in metal thin films during long-term use. The oxidization and hydrogen embrittlement will increase the electrical resistivities and decrease the thermal conductivities. Therefore, the buckypaper heater is used in this study.
6. Use of opaque buckypapers in transmissive optical devices
Opaque buckypaper heaters absorb and scatter visible light, so they cannot be used in transmissive optical devices. Two methods may be used to increase the transmittances of the transmissive optical devices on which are pasted buckypapers. One of the methods (edge-pasting method) is to paste buckypapers on the edges of the transmissive optical devices. A related experiment is performed to demonstrate the feasibility of the edge-pasting method. Two photochromic lenses that darken upon exposure to UV light are used in the experiment. On the edge of one of the lenses is pasted highly flexible buckypaper while the other has no buckypaper, as presented in Fig. 8(a). A DC voltage of 4.0 V is applied to the buckypaper to heat the photochromic lens. Two UV beams (λ = 375 nm) with equal intensity (5 mW/cm2) impinge separately on the two lenses. A digital video camera captures the two lenses under UV irradiation, as presented in Media 1. Figures 8(a) and 8(b), which are images taken from Media 1, present the two lenses following UV irradiation for 0 s and 15 s. After the UV beams irradiate the two lenses for 15 s, the irradiated region of the left lens is darker than that of the right lens, as presented in Fig. 8(b). It is attributable to the fact that the heat that the buckypaper generates accelerates the optical response of the left lens . Therefore, the buckypaper does not lose its heating function even if the buckypaper is pasted to the edge of the left lens. The other method for increasing the transmittances of the transmissive optical devices on which are pasted buckypapers is to burn holes into the buckypapers using laser light (hole-burning method). Related experimental results have been presented elsewhere . Therefore, the edge-pasting and hole-burning methods can overcome the optical absorption and scattering of opaque buckypapers.
This work fabricates the all-optical switch with adjustable threshold and switch-off powers using the DDLC cell on which is pasted the buckypaper. Preheating the DDLC cell reduces the difference between the cell temperature and the clearing point of the DDLC mixture. A small temperature difference makes the required concentration of the cis-isomers low, resulting in a low threshold power and a low switch-off power. Therefore, the threshold power and switch-off power decline as the mean cell temperature increases. The all-optical switch can be developed into an electrically controllable optical limiting device for multi-purpose use. The use of buckypaper and E7 LCs herein makes the DDLC-based all-optical switch functional over a wide range of ambient temperatures. The buckypaper can be pasted on other DDLC cells for use in high-speed optical switches. Buckypaper that resembles a sticker can be conveniently applied.
The authors would like to thank the National Science Council of the Republic of China, Taiwan, for financially supporting this research under Contract No. NSC 102-2112-M-029-001-MY3.
References and links
1. I. C. Khoo, J. H. Park, and J. Liou, “All-optical switching of continuous wave, microsecond lasers with a dye-doped nematic liquid crystal,” Appl. Phys. Lett. 90(15), 151107 (2007). [CrossRef]
2. I. C. Khoo, J. Liou, and M. V. Stinger, “Microseconds–nanoseconds all-optical switching of visible-near infrared (0.5 μm–1.55 μm) lasers with dye-doped nematic liquid crystals,” Mol. Cryst. Liq. Cryst. (Phila. Pa.) 527(1), 109–118 (2010). [CrossRef]
3. I. C. Khoo and A. Diaz, “Theory and experimental studies of all-optical transmission switching in a twist-alignment dye-doped nematic liquid crystal,” J. Opt. Soc. Am. B 25(11), 1931–1937 (2008). [CrossRef]
4. A. Shishido, O. Tsutsumi, A. Kanazawa, T. Shiono, T. Ikeda, and N. Tamai, “Rapid Optical Switching by Means of Photoinduced Change in Refractive Index of Azobenzene Liquid Crystals Detected by Reflection-Mode Analysis,” J. Am. Chem. Soc. 119(33), 7791–7796 (1997). [CrossRef]
5. Y. H. Yoon, J. W. Song, D. Kim, J. Kim, J. K. Park, S. K. Oh, and C. S. Han, “Transparent film heater using single-walled carbon nanotubes,” Adv. Mater. 19(23), 4284–4287 (2007). [CrossRef]
6. M. O. Memon, S. Haillot, and K. Lafdi, “Carbon nanofiber based buckypaper used as a thermal interface material,” Carbon 49(12), 3820–3828 (2011). [CrossRef]
7. H. K. Ma, P. C. Chien, M. C. Tsai, H. Y. Miao, J. H. Liu, and C. Y. Huang, “Application of buckypaper for preserving cholesteric liquid crystal cells within a certain temperature range,” Opt. Mater. Express 4(4), 719–724 (2014). [CrossRef]
8. Z. Wang, Z. Liang, B. Wang, C. Zhang, and L. Kramer, “Processing and property investigation of single-walled carbon nanotube (SWNT) buckypaper/epoxy resin matrix nanocomposites,” Comp. Pt. A: Appl. Sci. Manufact. 35(10), 1225–1232 (2004). [CrossRef]
9. Y. Chen, H. Y. Miao, M. Zhang, R. Liang, C. Zhang, and B. Wang, “Analysis of a laser post-process on a buckypaper field emitter for high and uniform electron emission,” Nanotechnology 20(32), 325302 (2009). [CrossRef] [PubMed]
10. Y. H. Yoon, J. W. Song, D. Kim, J. Kim, J. K. Park, S. K. Oh, and C. S. Han, “Transparent film heater using single-walled carbon nanotubes,” Adv. Mater. 19(23), 4284–4287 (2007). [CrossRef]
12. K. Yang, J. He, P. Puneet, Z. Su, M. J. Skove, J. Gaillard, T. M. Tritt, and A. M. Rao, “Tuning electrical and thermal connectivity in multiwalled carbon nanotube buckypaper,” J. Phys. Condens. Matter 22(33), 334215 (2010). [CrossRef] [PubMed]
13. I. Pires, L. Quintino, R. M. Miranda, and J. F. P. Gomes, “Fume emissions during gas metal arc welding,” Toxicol. Environ. Chem. 88(3), 385–394 (2006). [CrossRef]
14. Optrel Company, “Product List,” http://optrel.com/en.
15. B. V. Gemert, “The commercialization of plastic photochromic lenses: a tribute to John Crano,” Mol. Cryst. Liq. Cryst. (Phila. Pa.) 344(1), 57–62 (2000). [CrossRef]
16. Y. Chen, H. Y. Miao, R. J. Lin, M. Zhang, R. Liang, C. Zhang, and B. Wang, “Emitter spacing effects on field emission properties of laser-treated single-walled carbon nanotube buckypapers,” Nanotechnology 21(49), 495702 (2010). [CrossRef] [PubMed]