Skip to main content

Carbon Nanotube Devices at the Quantum Limit

Exploitable results

The work of WP1 has concentrated in understanding phenomena that affect RF transport dynamics and charge sensitivity in Nano-FET type of devices made of semi-conducting carbon nanotubes or semi-metallic graphene sheets. We have studied, both theoretically and experimentally, at room and cryogenic temperatures, the GHz operation of single wall nanotube FETs (NT-FETs) and single layer graphene FETs (GR-FETs). Measured transport parameters are i) the drain-source conductance ii) the transconductance iii) the gate capacitance, iv) the transit frequency and v) the shot noise. In most cases we have provided the first GHz determination of these quantities and in some cases we have been able to compare quantitatively their absolute or relative values with theoretical predictions for the ultimate quantum limit, meeting thereby one of the main scopes of the CARDEQ project. The work of WP2 has concentrated in understanding phenomena that affect charge sensitivity in rf-SET type of devices. We have investigated gate modulation both theoretically and experimentally, and obtained a quite good overall understanding of the phenomena. Charge detection sensitivity using SWNT rf-SETs has been brought to a new level: Our charge sensitivity is seven times better than reached earlier in nanotubes, while our gain-bandwidth product is by more than one order of magnitude better than for typical radio-frequency single-electron transistors. The work of WP3 has concentrated on contacting single walled and multiwalled carbon nanotubes, including some work on single layers of graphene. We have employed different metals for making the contacts as well as various cleaning/annealing procedures for improving their conductance. Ti and Pd have been found to be the most reliable metals for making the contacts. Gold, with a thin (2-3 nm) Ti or Cr sticking layer is also quite good, and thanks to its stability, it has been used for etched, suspended samples. WP 4 has established the development of CNT-based Josephson junctions as planned in the proposal. Concerning the gate control of the Josephson current in single- and multi-walled CNTs, substantial progress has been made in our experimental work, and several different regimes of nanotube Josephson junctions have been investigated. In FP regime, critical currents up to 4.8 nA have been observed. In this regime, the nanotube together with superconducting leads can be considered as a resonant level quantum dot, and thus the two-barrier Breit-Wigner model is applicable to model its behavior. The Kondo regime is very rich in phenomena and a lot of additional work is needed before it becomes fully understood. The CARDEQ work has just scratched the surface so far. In multiwalled tubes, we have also reached good contact transparency, but scattering in the tube affects strongly the gate modulation pattern. Instead of a Fabry-Perot interference, the patterns are reminiscent of universal conductance fluctuations. The scattering also leads to diffusive proximity junctions in multiwalled tubes, contrary to the resonant junction of short single walled tubes. In WP5 mechanical vibrations of suspended carbon nanotubes (and graphene) have been investigated. The main goal has been to develop nanotube resonators as inertial mass sensors. Specifically, the mass responsivity and the mass resolution were measured, two important parameters for the evaluation of the performance for mass sensing. The values for these quantities were found to be excellent; they surpass those reported previously for resonators made of nanotube or any other material. These results show that carbon nanotube is the material of choice for fabricating ultrasensitive mechanical resonators. The reason is that the mass of a nanotube is ultralow so that even a tiny amount of atoms deposited onto the nanotube makes up a significant fraction of the resonator's total mass. This result offers many new perspectives for mass spectrometry. The main achievements reached in WP5 are (i) enhancement of nanotube resonator's Q-value up to 5100, (ii) an increase of the resonant frequency up to GHz range and mapping of the shape of the modes both in CNTs and graphene, (iii) a mass detection using a mechanical CNT resonator with a resolution of 1.7 zg at He temperature, and (iv) observation of coupling mechanical vibrations with charge transport.