AVS2001 Session NT-WeM: Nanotubes: Nanoelectronics

Wednesday, October 31, 2001 8:20 AM in Room 133

Wednesday Morning

Start	Invited?	Item
8:20 AM	Invited	NT-WeM-1 When are Carbon Nanotubes Ballistic Conductors? W.A. de Heer (Georgia Institute of Technology) Frank, de Heer et al (Science 280,1744 (1998) found that freely suspended multiwalled carbon nanotubes (MWNTs) are 1-D conductors; the current flows on the outer layer, while large current densities are sustained. Quantized conductance was also found. These findings imply that carbon nanotubes could be ballistic conductors at room temperature. However there are difficulties that need clarification. One is that the conductance is only half what is expected from the theory of single wall nanotubes. Experiments performed on substrate supported and lithographically contacted nanotubes do not (yet) exhibit quantized conductance. Moreover electrical transport in carbon nanotubes is still quite confusing. Our experiments show that the conductance of clean, well-contacted nanotubes is quantized and independent of intercontact distance. Contamination, defects, non-ideal contacts and substrate interactions may explain the discrepancies between our experiments and others.
9:00 AM		NT-WeM-3 Electrical Characterization of Carbon Nanotube - Metal Contacts R. Vajtai, B.-Q. Wei, Y.V. Shusterman (Rensselaer Polytechnic Institute); K.A. Dunn, K. Dovidenko (The University at Albany-SUNY); L.J. Schowalter, P.M. Ajayan (Rensselaer Polytechnic Institute) Carbon nanotubes are excellent candidates to be used as interconnects or even active elements in nano-electronics. To harness the electrical properties of nanotubes in future nano-circuits, one needs to handle the most local neighboring effects such as contacting, insulation from the substrate and localized charges on the surface. Here we present results on spatially resolved electronic conductance of multiwalled nanotubes and nanotube networks on samples prepared by conventional and focused ion beam (FIB) lithography. Our results show that electrical potential changes along the nanotubes, drop at the contacts and spread into the perpendicular direction causing measuring difficulties for the nanoprobe, but also causing changes in the local electric field sensed by the nanotube. In this talk we will present the topographic and related spatially resolved electrostatic potential distribution in nanotube networks and we will describe the possible effect on the applications of nanotube-metal and nanotube-nanotube interconnects. Investigations made on similar configurations also showed long-term, high-current durability of the system, and low noise-factors resulting in most probably from the good coupling between the nanotubes and the metal electrodes.
9:20 AM		NT-WeM-4 Analysis of Carbon Nanotube Metal-Semiconductor Diode Device T. Yamada (NASA Ames Research Center) We study recently reported drain current (I_d)-drain voltage (V_d) characteristics of a carbon nanotube metal-semiconductor diode device with the gate voltage (V_g) applied to modulate the carrier density in the nanotube.¹ The diode was kink-shaped at the metal-semiconductor interface. It was shown that (1) larger negative V_g blocked I_d more effectively in the negative V_d region, resulting in the rectifying I_d-V_d characteristics, and that (2) positive V_g allowed I_d in the both V_d polarities, resulting in the non-rectifying characteristics. The negative V_d was the Schottky reverse direction, judging from the negligible I_d behavior for a wide region of -4 V < V_d < 0 V, with V_g = -4 V. Such negative V_g would attract positive charges from the metallic electrodes (charge reservoir) to the nanotube and lower the nanotube Fermi energy (E_F). With larger negative V_g, the experiment showed that the Schottky forward direction (V_d > 0) had a smaller turn-on voltage and the Schottky reverse direction (V_d < 0) was more resistant to the thermionic breakdown. Therefore, the majority carriers in the transport would be electrons since they can see a lower tunneling barrier (shallower built-in potential) in the forward direction when E_F is lowered, and a thicker tunneling barrier (Schottky barrier) in the reverse direction due to the reduction in the electron density when E_F is lowered. ¹Z. Yao, H. W. Ch. Posma, L. Balents, and C. Dekker, Nature 402, 273 (1999).
9:40 AM		NT-WeM-5 Electrical Transport through Vertically Aligned In-situ Contacted Multiwall Carbon Nanotubes F. Kreupl, A. Graham, E. Unger, M. Liebau, W. Hönlein (Infineon Technologies, Corporate Research, Germany) In order to exploit the high electrical performance of carbon nanotubes (CNTs), they have to be connected at both ends with highly conductive, i.e., metallic contacts. In addition to the quality of the grown CNTs, it is also very important that the contact resistances are as low as possible. CNTs grown on metallic surfaces give inherently a low contact resistance as the catalyst particles from which the nanotubes grow out of contact almost all layers of the multi-wall CNTs. We will report on electrical transport measurements on arrays of vertically aligned multi-wall CNTs contacted at the top and bottom. The bottom electrode with catalyst particles is covered completely with a thick silicon dioxide layer, in which holes of various sizes are etched down to the bottom electrode with reactive ion beam etching. The holes serve as templates in which CNTs are grown with chemical vapor deposition (CVD) directly from catalyst particles at the bottom electrode and contacted on top with a second metal electrode. Electrical measurements of the CNTs yielded extremely low resistances indicating electrical transport through almost all shells of the individual CNTs.
10:00 AM		NT-WeM-6 Electronic Rectification, Ballistic Switching, and Logic Gates with Carbon Nanotube `Y-Junctions' D. Srivastava (NASA Ames Research Center); A. Andriotis (Foundation for Research and Technology, Greece); M. Menon (University of Kentucky); L. Chernozetonski (Russian Academy of Sciences) I-V characteristics of single-wall carbon nanotube Y-junctions are calculated using an efficient embedding Green's Function formalism that allows for conductance across nanotube multiple junctions and realistic nanotube metal-lead contacts. The current vs voltage characteristics show the assymetry and rectification, in agreement with recent experimental results on multiwall nanotube Y-junctions. In symmetric Y-nanotube junctions the rectification has a weak dependence on the angle at the junction, and can support both ballistic rectification and/or ballistic switching operating modes. Although structural symmetry of the Y-junction is found to be necessary condition for rectification, it may not be sufficient for all cases. Transport calculations as a 3-terminal device show modulation of the current in the bias channel and a branch as a fucntion of applied voltage across the other branch. This behavior has been exploited to propose simple logic gates with Y junction carbon nanotubes.
10:20 AM		NT-WeM-7 Development of N-type Carbon Nanotube Transistors and Fabrication of the First Nanotube Logic Circuits V. Derycke, R. Martel, Ph. Avouris (IBM T.J. Watson Research Center) Single carbon nanotubes can be used as the active channel in field effect transistors (FETs).^1,2 Without any special treatment, the obtained FETs are always P-type: the current carriers are holes and the devices are off for positive gate bias. Here we show that this transistor behavior is due to a strong pinning of the Fermi level at the valence band of the semiconducting nanotubes and the presence of a large Schottky barrier for electron injection into the device. The fabrication of N-type FETs has recently been achieved by doping the device with an electron donor such as potassium.³ We introduce a novel approach of converting SWNT-based P-FETs into N-FETs without the use of dopands. The technique relies on the control of the electrostatic barriers and the modification of the contacts. By a specific treatment of the device we can position the Fermi level at any position within the band gap in a controlled manner and therefore modify the character of the device. For intermediate modification, the Fermi level lies in the middle of the band gap and the device shows both hole and electron conduction at negative and positive gate biases respectively. In completely converted devices, the Fermi level is pinned at the conduction band edge and the resulting FET is fully N-type. Using our ability to prepare both P- and N-type transistors we were able to fabricate the first nanotube-based integrated logic circuit, a "NOT" gate (voltage inverter). Inverters using K-doped tubes were also fabricated. We will compare the voltage inverters made by the two techniques. ¹ S. J. Tans, A. R. M. Verschueren, and C. Dekker, Nature (London) 393, 49 (1998). ² R. Martel, T. Schmidt, H. R. Shea, T. Hertel, and Ph. Avouris, Appl. Phys. Lett. 73, 2447 (1998). ³ M. Bockrath, J. Hone, A. Zettl, P.L. McEuen, A.G. Rinzler and Richard E. Smalley, Phys. Rev. B 61, R10606 (2000).
11:00 AM		NT-WeM-9 Contacting Carbon Nanotubes by Electrodeposition of Metal D.W. Austin, M.A. Guillorn (University of Tennessee); D.B. Geohegan (Oak Ridge National Laboratory); A.A. Puretzky (University of Tennessee); P.F. Britt, M.L. Simpson (Oak Ridge National Laboratory) We report our progress on the development of controllably contacting carbon nanotubes as part of our efforts to develop molecular-scale electronic devices. We are applying two approaches for electrically contacting single wall carbon nanotubes (SWNTs) that span metal electrodes. In the first approach, electrode metal is evaporated over the nanotubes, at the risk of damaging the tubes with the patterning electron beam. The second approach is to deposit the SWNTs onto prefabricated electrodes and then locate nanotubes that bridge two or more contacts. In the latter method, the nanotube/electrode resistance is typically on the order of 1 MΩ, due to the high contact resistance between the electrode and the sidewall of the nanotubes. We have had success in depositing SWNTs across prefabricated electrodes and making charge transport measurements. Our next step is to establish a contact method that will improve the conductivity across the nanotube/metal junction, and lead to the development of SWNT-based devices. We are conducting a set of experiments to determine if nanotubes can be contacted by electroplating metals such as gold, palladium, and platinum onto existing electrodes. In this approach, the electrodeposition of metal takes place long enough to allow the electrodes to contact the ends of the SWNTs. We will report the results of our electrodeposition experiments and describe our ongoing efforts towards the development of carbon nanotube-based molecular electronic devices.