Letter pubs.acs.org/NanoLett
Photodetection and Photoswitch Based On Polarized Optical Response of Macroscopically Aligned Carbon Nanotubes Ling Zhang,† Yang Wu,† Lei Deng,‡ Yi Zhou,† Changhong Liu,† and Shoushan Fan*,† †
Tsinghua-Foxconn Nanotechnology Research Center and Department of Physics, Tsinghua University, Beijing 100084, People’s Republic of China ‡ Department of Electrical Engineering, Tsinghua University, Beijing 100084, People’s Republic of China S Supporting Information *
ABSTRACT: Light polarization is extensively applied in optical detection, industry processing and telecommunication. Although aligned carbon nanotube naturally suppresses the transmittance of light polarized parallel to its axial direction, there is little application regarding the photodetection of carbon nanotube based on this anisotropic interaction with linearly polarized light. Here, we report a photodetection device realized by aligned carbon nanotube. Because of the different absorption behavior of polarized light with respect to polarization angles, such device delivers an explicit response to specific light wavelength regardless of its intensity. Furthermore, combining both experimental and mathematical analysis, we found that the light absorption of different wavelength causes characteristic thermoelectric voltage generation, which makes aligned carbon nanotube promising in optical detection. This work can also be utilized directly in developing new types of photoswitch that features a broad spectrum application from nearultraviolet to intermediate infrared with easy integration into practical electric devices, for instance, a “wavelength lock”. KEYWORDS: carbon nanotube, polarimetry, photodetector, photoswitch, thermoelectric effect conventional polarizer. Spinning CNT yarns and film can work in ultraviolet range of tens of nanometers.14 In the terahertz region, aligned single-walled CNT film and stacks exhibits high polarization degree and extinction ratios.15,22 Moreover, the optical absorption spectrum of CNTs shows no absorption features from near-ultraviolet to intermediate infrared, strongly suggesting advantage in wide spectrum functional devices that can be exploited from the abundant physical properties of CNTs. On this basis, we report a method of photodetection based on aligned CNT film (ACNTF) in light of its polarization selectivity. The absorption of polarized light is a function of included angle between CNT alignment and polarization direction. To improve the detection efficiency, the absorption ratios of polarized light between parallel and perpendicular configurations are measured via thermoelectric effects, due to the conductive nature of CNTs.28 In addition, the data are interpreted in terms of mathematical modeling, yielding that the reciprocal of square root of absorption is an elliptical function of included angle in the polar coordinates. Accordingly, the fundamental principle of CNT photodetector and photoswitch are established. The simple preparation, broad-spectrum response, high robustness and good embedability suggested advantages to integrate with commercial
T
he unique optical property of nanomaterials is the fundamental of ground-breaking optical devices with ultrafast response, broadband spectrum detection, and facile embedability1,2 and thus triggered a rich number of interests in the fields of laser sources, biological systems, and photodetection.3−5 As one of the promising low-dimensional materials, carbon nanotube (CNT) possesses multifaceted optical properties such as ultraviolet absorption, infrared detection and polarization selectivity, which can be regarded as a consequence of its characteristic electronic structures.6−10 Importantly, the large aspect ratio and the 1D geometry of CNT result in an anisotropic response to electromagnetic radiation due to the constrained electron motion and electron− phonon coupling.11−16 In contrast to typical 2D or 0D nanomaterials such as graphene, transition metal dichalcogenides, and semiconductor quantum dots, whose optical properties are correlated to the energy band structure, chemical compositions, and physical thickness, CNT has exhibited a distinctive full spectrum response to polarized light. However, the random orientation in macroscopic CNTs such as buckypaper, CNT sponge, and CNT film has inevitably compromised the exceptional polarization property.17−19 To address the issue of misalignment, freestanding CNT films that consists of parallel CNT yarns directly pulled from spinnable superaligned CNT arrays is a rational approach that can lead to promising applications, for instance, polarizer, high tensile fibers, and bionic navigation systems.14,15,20−27 Especially, CNT polarizer shows ideal performance in a wider range than © XXXX American Chemical Society
Received: July 5, 2016 Revised: September 6, 2016
A
DOI: 10.1021/acs.nanolett.6b02778 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 1. (A, B) Photographs of ACNTF under different arraying directions on LCD. The red and white arrows indicate the polarization direction of the light from LCD and the alignment of CNTs, respectively. The side length of PTFE frame is 4 cm in both images. (C) SEM image of the superaligned CNT.
Figure 2. Schematic illustration of (A) ACNTF transmittance measurement and (B) TP and Seebeck coefficient measurement and testing.
devices. To demonstrate, a “wavelength lock” device, which can only be triggered by the light with specific wavelength is fabricated to verify the application intuitively. The CNT used in the experiment was prepared from spinnable superaligned CNT arrays grown with a low-pressure chemical vapor deposition method that has been described in previous publications.21,29 Please note that the CNT grown in this method is multiwalled with a diameter of 12 nm and around 10 graphite layers. The CNTs are spun along one direction from the superaligned CNT array and extended continuously. In order to achieve freestanding and macroscopically ACNTFs, the extracted CNTs are suspended onto a Teflon square frame with an effective sample area of 28 mm × 28 mm. The extraction was conducted for only once. A proof-of concept polarizer made of macroscopically ACNTFs is rapidly demonstrated by placing it on a smartphone screen (Figure 1). We observed a transparent ACNTF when the alignment direction of CNT is perpendicular to the polarization direction of light that was emitted from an LCD screen (Figure 1A). The direction of polarization is denoted by red arrow, whereas the alignment direction of CNT is depicted in white arrow in Figure 1A. A landscape photo is displayed as a background image. The transparency of the ACNTF was significantly decreased when the Teflon frame was rotated by 90° so that the CNT alignment direction is parallel to the light polarization direction (Figure 1B). As a result, we observed a darkened image because the transmission was suppressed this
time and we verified that parallel CNT yarns were naturally polarizers. The microstructure of the ACNTF is analyzed with scanning electron microscopy (SEM) and a typical SEM image is shown in Figure 1C. According to the SEM analysis, the bundled CNTs are arranged and stretched continuously in a uniform direction, suggesting a linear polarizer is effectively realized by aligned 1D conductors. The apparatuses and optical path for the measurement of transmittance spectra and thermoelectric potential (TP) were sketched in Figure 2. The transmission spectrum from 300 to 1100 nm was recorded in a PerkinElmer Lambda 950 UV/vis spectrometer. The laser is polarized with a linear polarizer before reaching the sample (Figure 2A). The light intensities before and after passing through sample are measured so as to calculate the transmittance. To unambiguously present the experimental data, we use “included angle” to denote the angle between polarization direction of light and the CNT alignment. The change of included angle is realized by rotating the linear polarizer. To investigate the interaction between the laser and the ACNTF, we choose monochromic laser sources at 973, 650, 450, and 405 nm, respectively. The alignment direction of CNT was always fixed during the measurement, whereas the included angle was varied by adjusting the polarization direction of the polarizer. The measurement of temperature increase arising from the optical absorption of ACNTF was illustrated in Figure 2B. The ACNTF sample was suspended between two copper blocks, B
DOI: 10.1021/acs.nanolett.6b02778 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 3. (A) Transmittance spectra of polarized lights as a function of wavelength at selected included angles between the polarization of light and the CNT alignment. (B) TP difference of CNT as a function of temperature difference exemplified by a 973 nm laser. The Seebeck coefficient, the slope of linear fitting, is 9.03 μV/K. (C) TP ratio of 90° and 0° polarizations as a function of incident light wavelength. Both low power mode and high power mode are applied. (D) Reciprocal of the square root of absorption as a function of included angle in the polar coordinates showing ellipses.
the included angle of 90°. This embodies the transmittance monotony and broadband characters of ACNTF, which differs ACNTF from common-used polarizers, such as polymer films. Because of the irregular preparation process and high reflection, a traditional polarizer shows multivalued absorption and comparatively narrow working spectrum (Figure S1). The low reflection of freestanding CNT keeps off the factors that reduce transmittance monotony such as film interference.30 The measurement of absorption of CNT by electric voltage via the thermoelectric effect of CNT yields a simple approach to detect the polarization. In Figure 3B, the TP, generated by the heat effect of laser, is plotted versus the temperature difference where an infrared laser (973 nm) was used to shine the ACNTF. The linear fitting yielded a Seebeck coefficient of 9.03 μV/K. In Figure 3C, we measured TPs when the included angles are at 90° and 0°, respectively, and calculated their ratio. The same measurements were also performed for 405 nm, 450 nm, 650 nm, and 973 nm lasers. Such TP ratio increases as the wavelength increases. Furthermore, at every specific wavelength, high and low power modes for each laser source are
which were insulated by a PMMA substrate. The copper also served as electrodes and heat sinks. Samples were painted to the copper by high-purity silver paste to reach a good electrical and thermal contact. A laser beam shined on one end of the sample, whereas the other side can be approximated to the room temperature due to the large heat capacity, relative to that of the ACNTF, of the copper substrate. In the meantime, the polarization angle was changed continuously by spinning the polarizer, and the actual effect of the included angle on both temperature and TPs of samples are also measured. The full transmittance spectra of the ACNTF at different included angles are plotted in Figure 3A. The spectra are recorded from 300 to 1100 nm, corresponding to the range from near-ultraviolet to intermediate infrared. Included angles are selected as 0°, 30°, 60°, and 90°. Generally, a slight increase in transmittance was observed as the wavelength increases. The lowest transmittance was in the parallel configuration where the included angle was 0°, consistent with the observation in Figure 1. As the included angle increases to 30° and 60°, the transmittance also increases, until to the maximum is reached at C
DOI: 10.1021/acs.nanolett.6b02778 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 4. (A) Schematic illustration of aligned CNT-based photoswitch. The arrow marks the clockwise spinning of sample stage when the photoswitch is operating. (B) Temperature difference between the cold end and the hot end of the ACNTF as a function of included angle when shined by different lasers..
aspect ratio increases as the light wavelength increases. For example, we observed a narrower ellipse corresponding to 973 nm laser but a much rounder ellipse corresponding to 405 nm laser. Because of the limitation of instrument, there is no original absorption data from included angle 325° to 30°. It is suggestive that the unique aspect ratio, or the “ellipse shape” is the fingerprint to recognize the light with certain wavelength. Because the macroscopic ACNTF shows the capability in recognizing linearly polarized light irrespective of input power, a new type of photoswitch could be developed based on the optical properties of such ACNTF. The CNT-based photoswitch is sketched in Figure 4A. The ACNTF is connected with a thermocouple to measure TP signals, which are then transmitted into a processor that controls the device. The device is placed on a rotating stage so that the function between TP and included angles could be recorded. In Figure 4B, the temperature differences corresponding to 450 and 973 nm lasers, respectively, are provided. By rotating the stage for 180°, the maximum and minimum values of temperature and TPs are measured. By using these TP values, the aspect ratio of aforementioned absorption ellipse is calculated. Then the data is analyzed in the processor. Only when the aspect ratios of the ellipse were consistent with that of the prestored ellipse of a specific wavelength that served as a key would the photoswitch be activated. Thus, this photoswitch designed based upon ACNTF is programmable toward a specific light wavelength. For a better demonstration, we also provided an actual operation video clip in which the photoswitch (Supporting Information), or the “wavelength lock”, can only be triggered by a 650 nm laser whose data were prestored. By changing the prestored data, the photoswitch can respond to different “keys” selected according to requirements. This application demonstrates the wide-spectrum and well-performed embedability of CNT photoswitch, especially compared with other lowdimensional nanomaterials, and reveals the property of selective wavelength response and the capability in functional devices.2,32−34 The CNT spinning and measuring electric signal instead light signal ensures high robustness of the device. In conclusion, we reported a new type of photodetecting device realized by freestanding ACNTF. Due to its 1D anisotropic property, we demonstrate that such ACNTF can be directly utilized in polarized light detection. To further
examined and the exact power of each laser is compiled in Supporting Information (Table S1). The small variation of the TP ratio, when the power mode was switched, may be attributed to the nonlinear optical absorption of CNT.31 However, it is important to point out that the trend is not significantly affected (Figure 3C). In fact, by measuring the TP ratio of included angles 90° over 0°, we can determine the incident light wavelength with little interference from the light intensity. The generation of temperature differences and TPs of ACNTF when interacting with polarized light can be attributed to the electron−photon interaction.31−33 Starting from this, we can derive the actual absorption with respect to the light polarization in the form of γλ(θ ) =
Iλ(θ ) = αλ 2cos 2 θ + βλ 2 sin 2 θ I
where θ is the included angle, I is the light intensity, and αλ and βλ are the CNT’s absorption coefficients to polarized light in parallel and perpendicular directions, respectively. It should be pointed out that the plot of γ−1/2 (θ) is ellipse in polar λ coordinates with an aspect ratio of αλ that defines the ellipse βλ
shape. To derive
αλ , βλ
we denote Iλ(θ) ∝ I2λ(θ), in which Uλ(θ) is
the TP generated by light with an intensity of Iλ(θ), if assuming sample a constant energy conversion efficiency. Therefore, the TP ratio can be used to derive that
αλ βλ
=
Uλ(0°) . Uλ(90°)
The detailed
derivation process can be found in the Supporting Information. We next analyze the results in different cases. When the wavelength of the incident laser is fixed, the ellipse will only grow “bigger” but keep the same aspect ratio when the laser intensity is increased. This is because that the aspect ratio is only the function of αλ , regardless of the intensity, as suggested βλ
previously. Therefore, switching from low power mode to high power mode will not significantly change the shape of ellipses, as shown in Figure 3C. On the other hand, when the wavelength changes, the aspect ratio αλ changes considerably βλ
due to the different absorption coefficients. In Figure 3D, we present this relation by plotting the ellipse for each wavelength. It is clear that according to the derived elliptical function, the D
DOI: 10.1021/acs.nanolett.6b02778 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
(12) Einarsson, E.; Edamura, T.; Maruyama, S.; Murakami, Y. Phys. Rev. Lett. 2005, 94, 087402. (13) Milkie, D. E.; Kane, C. L.; Yodh, A. G.; Kikkawa, J. M.; Islam, M. F. Phys. Rev. Lett. 2004, 93, 037404. (14) Jiang, K.; Li, Q.; Fan, S. Nature 2002, 419, 801−801. (15) Ren, L.; Pint, C. L.; Booshehri, L. G.; Rice, W. D.; Wang, X.; Hilton, D. J.; Takeya, K.; Kawayama, I.; Tonouchi, M.; Hauge, R. H.; Kono, J. Nano Lett. 2009, 9, 2610−2613. (16) Bachilo, S. M.; Strano, M. S.; Kittrell, C.; Hauge, R. H.; Smalley, R. E.; Weisman, R. B. Science 2002, 298, 2361−2366. (17) Li, Z.; Dharap, P.; Nagarajaiah, S.; Barrera, E. V.; Kim, J. D. Adv. Mater. 2004, 16, 640−643. (18) Gui, X.; Wei, J.; Wang, K.; Cao, A.; Zhu, H.; Jia, Y.; Shu, Q.; Wu, D. Adv. Mater. 2010, 22, 617−621. (19) Zhang, L.; Zhang, G.; Liu, C.; Fan, S. Nano Lett. 2012, 12, 4848−4852. (20) DeHeer, W. A.; Bacsa, W. S.; Châtelain, A.; Gerfin, T.; Humphrey-Baker, R.; Forro, L.; Ugarte, D. Science 1995, 268, 845− 847. (21) Zhang, X.; Jiang, K.; Feng, C.; Liu, P.; Zhang, L.; Kong, J.; Zhang, T.; Li, Q.; Fan, S. Adv. Mater. 2006, 18, 1505. (22) Ren, L.; Pint, C. L.; Arikawa, T.; Takeya, K.; Kawayama, I.; Tonouchi, M.; Hauge, R. H.; Kono, J. Nano Lett. 2012, 12, 787−790. (23) Zhang, D.; Ryu, K.; Liu, X.; Polikarpov, E.; Ly, J.; Tompson, M. E.; Zhou, C. Nano Lett. 2006, 6, 1880−1886. (24) Wu, Z.; Chen, Z.; Du, X.; Logan, J. M.; Sippel, J.; Nikolou, M.; Kamaras, K.; Reynolds, J. R.; Tanner, D. B.; Hebard, A. F.; Rinzler, A. G. Science 2004, 305, 1273−1276. (25) Kyoung, J.; Jang, E. Y.; Lima, M. D.; Park, H.; Robles, R. O.; Lepró, X.; Kim, Y. H.; Baughman, R. H.; Kim, D. Nano Lett. 2011, 11, 4227−4231. (26) Titova, L. V.; Pint, C. L.; Zhang, Q.; Hauge, R. H.; Kono, J.; Hegmann, F. A. Nano Lett. 2015, 15, 3267−3272. (27) He, X.; Fujimura, N.; Lloyd, J. M.; Erickson, K. J.; Talin, A. A.; Zhang, Q.; Gao, W.; Jiang, Q.; Kawano, Y.; Hauge, R. H.; Léonard, F.; Kono, J. Nano Lett. 2014, 14, 3953−3958. (28) He, X.; Wang, X.; Nanot, S.; Cong, K.; Jiang, Q.; Kane, A. A.; Goldsmith, J. E. M.; Hauge, R. H.; Léonard, F.; Kono, J. ACS Nano 2013, 7, 7271−7277. (29) Jiang, K.; Wang, J.; Li, Q.; Liu, L.; Liu, C.; Fan, S. Adv. Mater. 2011, 23, 1154−1161. (30) Zhang, M.; Fang, S.; Zakhidov, A. A.; Lee, S. B.; Aliev, A. E.; Williams, C. D.; Atkinson, K. R.; Baughman, R. H. Science 2005, 309, 1215−1219. (31) Mishra, S. R.; Rawat, H. S.; Mehendale, S. C.; Rustagi, K. C.; Sood, A. K.; Bandyopadhyay, R.; Govindaraj, A.; Rao, C. N. R. Chem. Phys. Lett. 2000, 317, 510−514. (32) West, J. L.; Halas, N. J. Annu. Rev. Biomed. Eng. 2003, 5, 285− 292. (33) Robinson, J. T.; Perkins, F. K.; Snow, E. S.; Wei, Z.; Sheehan, P. E. Nano Lett. 2008, 8, 3137−3140. (34) Irie, M.; Fukaminato, T.; Sasaki, T.; Tamai, N.; Kawai, T. Nature 2002, 420, 759−760.
improve the accuracy and the efficiency of photodetection, TP generation was employed to distinguish the wavelength of light. Both experiments and mathematical analysis show that the working principle of this photoswitch is directly derived from the anisotropic optical response of aligned CNT to linearly polarized light. Peculiarly, the reciprocal of the square root of absorption of ACNTF gives an elliptical function of included angle in the polar coordinates. As the wavelength increases, the absorption difference between parallel and perpendicular directions also increases. The optical device using ACNTF features broad-spectrum response to linearly polarized light and high robustness to application therefore can be easily integrated into practical commercial devices. In a word, this work could be applied in the field of polarized light detection, broadband spectrum detection, and photosensitive devices, such as “wavelength lock”, based upon the basic optical property of CNT.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.6b02778. Movie clip explaining working principle and actual scene of “wavelength lock” prepared based on ACNTF. (MPG) Information of SEM details, power values of both low power and high power modes, the details of PerkinElmer Lambda spectrometer to prevent potential interference from inconsistent intensity along different polarization directions, as well as rotation limitation of polarizer in the spectrometer. Structure and working details of “wavelength lock”, transmittance spectra of polarized lights for a polymer polarizer, model derivation and theoretical derivation of ellipse form function. (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Telephone: +86 10 62796011. Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Lal, S.; Link, S.; Halas, N. J. Nat. Photonics 2007, 1, 641−648. (2) Russew, M. M.; Hecht, S. Adv. Mater. 2010, 22, 3348−3360. (3) Hu, J.; Li, L.; Yang, W.; Manna, L.; Wang, L.; Alivisatos, A. P. Science 2001, 292, 2060−2063. (4) Huang, M. H.; Mao, S.; Feick, H.; Yan, H.; Wu, Y.; Kind, H.; Weber, E.; Russo, R.; Yang, P. Science 2001, 292, 1897−1899. (5) Alivisatos, P. Nat. Biotechnol. 2004, 22, 47−52. (6) Knupfer, M.; Golden, M. S.; Fink, J.; Rinzler, A.; Smalley, R. E.; Pichler, T. Phys. Rev. Lett. 1998, 80, 4729−4732. (7) Kataura, H.; Kumazawa, Y.; Maniwa, Y.; Umezu, I.; Suzuki, S.; Ohtsuka, Y.; Achiba, Y. Synth. Met. 1999, 103, 2555−2558. (8) Williamson, A. J.; Reboredo, F. A.; Galli, G.; Puzder, A. Phys. Rev. Lett. 2003, 91, 157405. (9) Bubke, K.; Gnewuch, H.; Hempstead, M.; Hammer, J.; Green, M. L. H. Appl. Phys. Lett. 1997, 71, 1906−1908. (10) Fraser, J. M.; Finnie, P.; Homma, Y.; Lefebvre, J. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 69, 075403. (11) Tang, Z. K.; Liu, H. J.; Wang, N.; Chan, C. T.; Saito, R.; Okada, S.; Li, G. D.; Chen, J. S.; Nagasawa, N.; Tsuda, S.; Li, Z. M. Phys. Rev. Lett. 2001, 87, 127401. E
DOI: 10.1021/acs.nanolett.6b02778 Nano Lett. XXXX, XXX, XXX−XXX
