Although the thermal properties of millimeter-sized carbon nanotube mats and packed carbon nanofibers have been readily measured, measurements for a single nanotube are extremely difficult. Here, we report a novel method that can reliably measure the thermal conductivity of a single carbon nanotube using a suspended sample-attached T-type nanosensor. Our experimental results show that the thermal conductivity of a carbon nanotube at room temperature increases as its diameter decreases, and exceeds 2000 W/mK for a diameter of 9.8 nm. The temperature dependence of the thermal conductivity for a carbon nanotube with a diameter of 16.1 nm appears to have an asymptote near 320 K. The present method is, in principle, applicable to any kind of a single nanofiber, nanowire, and even single-walled carbon nanotube.

In the applications of single-walled carbon nanotubes (SWNTs), it is extremely important to separate semiconducting and metallic SWNTs. Although several methods have been reported for the separation, only low yields have been achieved at great expense. We show a separation method involving a dispersion-centrifugation process in a tetrahydrofuran solution of amine, which makes metallic SWNTs highly concentrated to 87% in a simple way.

In recent years, several researchers have reported the occurrence of reversible resistance switching effects in simple metal nanogap junctions. A large negative resistance is observed in the I-V characteristics of such a junction when high-bias voltages are applied. This phenomenon is characteristic behaviour on the nanometre scale; it only occurs for gap widths slightly under 13 nm. Furthermore, such a junction exhibits a non-volatile resistance hysteresis when the bias voltage is reduced very rapidly from a high level to around 0 V, and when the bias voltage is reduced slowly. This non-volatile resistance change occurs as a result of changes in the gap width between the metal electrodes, brought about by the applied bias voltage.

This letter reports on the measurements of the in-plane thermal conductivity and the electrical conductivity of a microfabricated, suspended, nanosized platinum thin film with the width of 260nm, the thickness of 28nm, and the length of 5.3μm. The experimental results show that the electrical conductivity, the resistance-temperature coefficient and the in-plane thermal conductivity of the nanofilm are greatly lower than the corresponding bulk values from 77to330K. The comparison results indicate that the relation between the thermal conductivity and the electrical conductivity of this nanofilm might not follow the Wiedemann–Franz law that describes the relation between the thermal conductivity and the electrical conductivity of a bulk metallic material.

Raw single-walled carbon nanotubes produced by arc discharge were first treated with K 2 S 2 O 8 in dilute H 2 -SO 4 solution to generate oxygenated functional groups such as carboxyl, hydroxyl, and carbonyl on varying carbon components. Further functionalization with octadecylamine was carried out via a condensation reaction between an amine group and a carboxyl group with the assistance of dicyclohexylcarbodiimide. The attachment of long alkyl chains leads the sample to be soluble in tetrahydrofuran and other organic solvents. The thus obtained stable dispersion was subjected to dispersion-centrifugation recycles, and then the metal catalysts and varying carbon impurities were separated with carbon nanotubes. The advantage of this procedure lies in high-yield and nondestructive recovery of the target material. Optical absorption spectroscopic and thermogravimetric analyses show that about 60% of the nanotubes in the starting material is extracted with a purity more than 90 wt %.