Effects of local Joule heating on the reduction of contact resistance between carbon nanotubes and metal electrodes

Abstract

We report here a practical application of known local Joule heating processes to reduce the contact resistance between carbon nanotubes and metallic electrical contacts. The results presented in this study were obtained from a series of systematic Joule heating experiments on 289 single-walled carbon nanotubes (SWCNTs) and 107 multiwalled carbon nanotubes (MWCNTs). Our experimental results demonstrate that the Joule heating process decreases the contact resistances of SWCNTs and MWCNTs to 70.4% and 77.9% of their initial resistances, respectively. The I-V characteristics of metallic nanotubes become more linear and eventually become independent of the gate voltages (Vgs). For semiconducting nanotubes, the contact resistance has a similar decreasing tendency but the dependency of source-drain current (Ids) on Vgs does not change with the Joule heating process. This suggests that the reduction of the contact resistance and the decrease of the transport potential barrier are largely attributed to the thermal-energy-induced desorption of adsorbates such as water and oxygen molecules from the nanotube surface and the interface region, as well as thermal-energy-enhanced bonding between the nanotubes and electrode surfaces. In comparison to several other methods including rapid thermal annealing, e-beam lithography patterning of the top metal layer, and focused ion beam induced metal deposition of the top layer, the Joule heating process not only effectively reduces the contact resistance but also simultaneously measures the resistance and monitors the change in the transport potential barrier at the interface region.

Field emission from a selected multiwall carbon nanotube

“…Once a first and quite stable CNT/metal contact has been obtained, a high current value (10 -4 A) has been applied across the entire tungsten-CNT-metal system, thus resulting, via Joule heating effect, in the cutting of the CNT in correspondence of a defective point 18,19 . At the same time, the current produced an anneal of the tungsten-nanotube interface, due to heating stimulated contaminants desorption 18,19 , yielding the formation of stable, though quite resistive, metal-nanotube contact. Furthermore, the open and sharp CNT tip morphology, which is a direct consequence of the current induced cutting procedure, is expected to enhance the emitting properties of the CNT, making the resulting system particularly suitable for field emission investigation 5 .…”

Local-gated single-walled carbon nanotube field effect transistors assembled by AC dielectrophoresis

“…After the trapping process is complete, the chip is rinsed with IPA and blow dried with nitrogen gas to remove any unwanted nanotubes or impurities in the suspension. Other groups have dispersed CNT in an aqueous sodium dodecyl sulfate (SDS) [14][15][16][17][18][19][20]23], isopropyl alcohol (IPA) [21], dimethylformamide (DMF) [24], and DCE [25] for DEP assembly. Although an aqueous SDS solution has been used most frequently and has been shown to disperse CNT well, it is not compatible with our device fabrication process.…”

“…In DEP, CNTs are assembled from solution using a non-uniform AC electric field. However, all DEP assembled CNT-FETs reported in the literature used only a global back gate [14][15][16][17][18][19][21][22]24]. Global back gated devices give poor device performance due to inefficient gate coupling and contact-controlled operation.…”

“…Recently, AC dielectrophoresis (DEP) has been utilized for large scale assembly of individual single-walled carbon nanotube (SWNT) or bundles at selected positions of the circuit [14][15][16][17][18][19][20][21][22][23][24][25]. In DEP, CNTs are assembled from solution using a non-uniform AC electric field.…”

In situ TEM study on the improvement of contact resistance between a carbon nanotube and metal electrodes by local melting

“…Many techniques for decreasing the contact resistance have been studied: electron beam irradiation to the contact region, [9] annealing the CNT device, [10] and Joule heating. [11] Only a few experiments, however, have been performed to observe the structure of the contact directly, [12,13] because it is difficult to observe the contact structure at an atomic scale with the simultaneous measurements of current and voltage by means of conventional methods. Because the contact resistance is governed by the contact structure, [14,15] it is highly required to acquire the structural information of the contact region to improve the electric properties of the contact.…”