Thermal resistance of the nanoscale constrictions between carbon nanotubes and solid substrates

Abstract

We have determined the thermal resistance for transferring heat between individual single-walled carbon nanotube devices and solid substrates. Using sapphire and comparing our results to previous results obtained from SiO 2 , we find that the resistance is dominated by interfacial resistance rather than the spreading resistance of heat for diffusing into the substrate. Our results are in agreement to a recent model for the thermal resistance of nanoscale constrictions. Our results suggest that relatively short contact lengths ͑ϳ10-30 nm͒ to a typical solid should be sufficient to transfer heat efficiently into carbon nanotubes, underscoring the potential of carbon nanotubes for nanoscale thermal management.

Controlling the thermal contact resistance of a carbon nanotube heat spreader

“…2,[5][6][7][8][9][10][11][12] This incongruity in some cases can be attributed to variations in thermal contact resistance R c , 13 a parameter that has recently received much attention, with studies on the subject reporting values that vary widely. 13,[14][15][16][17][18][19][20][21][22] For example, Maune et al perform thermometry employing electrical breakdown of CNTs to report R c values of 0.6--3.0 K·m/W, 15 whereas Tsen et al…”

“…13,[15][16][17][18][19][20] However, many of these other studies use SEM imaging, 13,16,19,20 which is known to deposit significant amounts of carbonaceous material onto CNTs during imaging. 27 This material may reduce R c in the same manner that Pd does here.…”

“…Similarly, other studies use a CNT manipulation-and-placement routine that explicitly calls for embedding the CNT in an acrylic adhesive. 2,28 For other studies that do not use SEM imaging or other adhesive coatings, 15,17,18 the measurements are performed at elevated temperatures, where near-field thermal radiation 29 may be playing a significant role in the transport phenomena, reducing the observed R c value. The results we report here do not use SEM imaging or adhesives, are well-characterized by TEM to exclude the possible presence of gross amounts of contaminating materials, and are carried out at relatively low temperatures, where thermal radiation effects are expected to play a diminished role in the transport.…”

Trapping cold atoms using surface-grown carbon nanotubes

“…We find that this configuration demands simpler fabrication, and as noted earlier, it will enable more complex geometries and larger currents than those possible for suspended nanotubes [40]. For example, to the best of our knowledge, the maximum current thus far achieved for suspended CNTs of length L is ∼ (10/L) µA [40], compared to 20 µA (L ≥ 3 µm) and 45 µA (L ≤ 1.5 µm) for surface-grown CNTs [41,42]. To verify simplicity in fabrication, we have also conducted a fabrication feasibility study, the detailed results of which will be made available elsewhere [28].…”

“…We usually assume a CNT current of 20 µA; noting that currents > 40 µA have been achieved experimentally [42], we perform some calculations at 35 µA in order to realistically characterize the nanotube trap sensitivity to current.…”