Controlled Ohmic and nonlinear electrical transport in inkjet-printed single-wall carbon nanotube films

We present the fabrication and characterization of logic elements (transistors and interconnects) built using our recently developed inkjet-printer-controlled deposition of single-wall carbon nanotube network films. The method requires no preselection of “metallic” or “semiconducting” nanotubes. By selecting the number of prints on a specified region, it is possible to have low-density, nonlinear, gate-voltage controllable transistors or high-density, linear, high-current-throughput metallic interconnects without any gate-voltage response. Intermediate steps drive the films between the nonlinear and linear regimes with precise controllability. The transport mechanism in these films as a function of bias, gate voltage, and temperature dependence have been investigated and analyzed using junction properties of metal-semiconductors in the context of networks of carbon nanotubes.

Figure 1

(a) Layout of the Si chip used in the CNT thin-film FET measurements. The optical micrograph shows a set of Pt electrodes (source-drain) upon which the nanotube films were printed. (b) Atomic force microscopy images of the electrode area. The higher magnification image shows a network of nanotube bundles printed between the electrodes (on the gate oxide).

Figure 2

Optical and scanning electron microscope images of the samples printed on alumina substrates. (a) Optical microscopy image of a low-density CNT line (three prints) connecting the gap between two gold electrodes. The dotted lines are to guide the eye along the edges of the printed nanotube pattern, which is practically invisible (Ref. 37). (b) Scanning electron microscope (SEM) picture shows a partially connected low-density network on the substrate. The high contrast features show local charge accumulation in the substrate around nonpercolated nanotubes. (c) Low magnification SEM image of a dense CNT line (20 prints) in the vicinity of a grounded gold electrode. Charging of the alumina substrate takes place where no nanotube network or evaporated metal film is deposited. (d) SEM image of a dense CNT network on alumina substrate and (e) close-up image of CNTs bridging two alumina crystallites. Note that because of the good percolation, the network is grounded well through the contact electrodes and local charging of the substrate by the electron beam is negligible as compared to that of the sample shown in (b).

Figure 3

Current-voltage characteristics of inkjet deposited SWCNT films of various coverage (print repetitions) at the temperature range of $20–80°\mathrm{C}$ measured (a) in ambient air and (b) in nitrogen atmosphere. (c) Relative change of film resistances $\Delta r∕r_0$ of dense (20 prints) and sparse networks (3 prints) upon heating (h1) and subsequent cooling (c1) and reheating (h2) in nitrogen atmosphere up to $150°\mathrm{C}$.

Figure 4

$I\text{−}{V}_{\mathrm{SD}}$ and $I\text{−}{V}_{\mathrm{G}}$ sweeps on (a) diluted and (b) thick nanotube networks.