Transparent, conductive carbon nanotube films

Science

Volumn 305, Issue 5688, 2004, Pages 1273-1276

  Wu, Zhuangchun ^a   Chen, Zhihong ^a,c   Du, Xu ^a   Logan, Jonathan M ^a   Sippel, Jennifer ^a   Nikolou, Maria ^a   Kamaras, Katalin ^b   Reynolds, John R ^a   Tanner, David B ^a   Hebard, Arthur F ^a   Rinzler, Andrew G ^a  

^a University of Florida ^*  (United States)

^b INSTITUTE OF EXPERIMENTAL MEDICINE   (Hungary)

^c IBM T J WATSON RESEARCH CENTER   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

CARBON NANOTUBES; ELECTRIC CONDUCTIVITY; ELECTRIC FIELDS; INDIUM COMPOUNDS; OXIDES; SUBSTRATES; TIN COMPOUNDS;

CONDUCTIVE CARBON NANOTUBE FILMS; ELECTRICAL COUPLING; PHOTONIC DEVICES; SHEET RESISTANCE; SPECTRAL BANDS;

ULTRATHIN FILMS;

INDIUM; INDIUM TIN OXIDE; NANOTUBE; TIN DERIVATIVE; UNCLASSIFIED DRUG;

NANOTECHNOLOGY;

ARTICLE; DEVICE; ELECTRIC CONDUCTIVITY; ELECTRIC FIELD; FILM; INFRARED RADIATION; OPTICS; PHOTON; PRIORITY JOURNAL; SEMICONDUCTOR;

EID: 4344647530     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.1101243     Document Type: Article

References (27)

1. D. Ginley et al., Mater. Res. Soc. Symp. Proc. 668, H 2.7.1 (2001).
2. R. G. Gordon, MRS Bull. 25, 52 (2000).
3. K. Lee et al., Nano Lett. 4, 911 (2004).
4. See supporting text on Science Online.
5. Others have also used vacuum filtration as the initial step to fabricate free-standing SWNT films (6). However, their transfer process involved cutting a rectangular hole in adhesive tape, laying this tape onto the nanotube layer on the membrane, and peeling up the tape to obtain the free-standing film within the hole in the tape. This gives comparatively little control over the film thickness and is not amenable to film deposition over large areas compatible with microelectronic processing.
6. F. Hennrich et al., Phys. Chem. Chem. Phys. 4, 2273 (2002).
7. M. E. Itkis et al., Nano Lett. 2, 155 (2002).
8. F. Hennrich, R. Wellmann, S. Malik, S. Lebedkin, M. M. Kappes, Phys. Chem. Chem. Phys. 5, 178 (2003).
9. S. Kazaoui, N. Minami, R. Jacquemin, H. Kataura, Y. Achiba, Phys. Rev. B 60, 13339 (1999).
10. C. F. Windisch Jr., K. F. Ferris, G. J. Exarhos, S. K. Sharma, Thin Solid Films 420-421, 89 (2002).
11. S. J. Tans, R. M. Verschueren, C. Dekker, Nature 393, 49 (1998).
12. Work over recent years has shown that this simple picture for NFET operation is by no means complete and ignores Schottky barriers that develop at the nanotube-metal contact junctions, which are also modulated by the gate fields (13). Nonetheless, carrier concentration modulation contributes to the effect and under some conditions dominates (14).
13. S. Heinze et al., Phys. Rev. Lett. 89, 106801 (2002).
14. A. Javey, J. Guo, Q. Wang, M. Lundstrom, H. Dai, Nature 424, 654 (2003).
15. M. Kruger, M. R. Buitelaar, T. Nussbaumer, C. Schonenberger, L. Forro, Appl. Phys. Lett. 78, 1291 (2001).
16. S. Rosenblatt et al., Nano Lett. 2, 869 (2002).
17. x dielectric layer.
18. D. Mann, A. Javey, J. Kong, Q. Wang, H. Dai, Nature 424, 654 (2003).
19. Modulation of the three principal SWNT absorption bands in a three-terminal electrochemical cell has been reported (20, 21). There, the effect was ascribed to nanotube Fermi level shifts associated with intercalated species generated by electrochemical redox reactions in the electrolyte. The SWNT film (produced by airbrushing) used in those experiments evidently did not have sufficient intrinsic conductivity to be used without a thin, transparent platinum electrode (20) or ITO electrode (21) onto which the nanotubes were sprayed. The ionic liquid in our experiments acts merely as a near-lying gate electrode undergoing no complicating redox reactions. It also yields substantially cleaner SWNT spectra over broader voltage and spectral ranges. The spectra we show are raw data.
20. S. Kazaoui, N. Minami, N. Matsuda, H. Kataura, Y. Achiba, Appl. Phys. Lett. 78, 3433 (2001).
21. L. Kavan et al., J. Phys. Chem. B 105, 10764 (2001).
22. Although the S1 band overlaps the 1.55-μm wavelength important in optical communications, the ionic liquid-gated device (or any electrolyte-gated device) is far too slow to be useful for modulation at the rates needed in communication devices. Because of the high viscosity of the ionic liquid, the time scale for electrostatic gate equilibration is on the order of minutes.
23. Our all-solid-state device has similar spectral behavior with "gate" voltage; however, the magnitude of the modulation is much smaller (0.2% over ±6 V at the peak of S1) because electrostatic screening allows only the layer of nanotubes nearest the ITO counterelectrode to participate. Because gating is all electronic (as opposed to ionic), however, the response time is much faster than the ionic liquid device.
24. With the spectrophotometer in a mode that records the transmittance at fixed wavelength as a function of time, we sat on the peak of the S1 absorption band while driving the O-NFET with a square wave potential. No systematic changes in the amplitude of the modulation were observed over multiple measurements totaling several hundred cycles. Only small changes (a few percent) in the average transmittance, both up and down, were seen-entirely consistent with spectrometer drift over the long time scale of these measurements.
25. Four-probe measurements were performed with a Linear Research 700 resistance bridge using 2-mV excitation at 16 Hz.
26. P. Chandrasekhar et al., Synth. Metals 135-136, 23 (2003).
27. Supported by the U.S. Army through Center for Materials in Sensors and Actuators grant DAAD19-001-0002 (J.R.R., D.B.T., A.F.H., A.G.R.), NSF grants DMR 0101856 (A.F.H.) and ECS-0210S74 (A.G.R.), and NSF-MTA-OTKA international grants 021, N31622 (D.B.T., K.K.), NSF-INT-9902050 (D.B.T., K.K.), and OTKA 034198 (K.K.).