I finally obtained a copy of the Nature article published October 24, 2002, volume 419, page 801. The researchers at Tsingua University in Beijing, China were able to draw continuous nanotube yarns (CNT) from a nanotube array on a silicon substrate in a manner similar to drawing thread from a silk cocoon. The yarns were 30 cm long and 200 micrometers wide.

A light bulb filament was created from a CNT and a DC potential of 70 volts was applied in a vacuum for 3 hours. The result was an increase in the tensile strength from 1 milli-Newton to 6.4 milli-Newtons. Using the best information given, assuming the yarn width can be interpreted as a diameter, then this tensile strength can be converted to approximately 200,000 Pascals. Previous publications had estimated the tensile strength of microscopic nanotubes to be 200 billion pascals, so the large scale CNT yarns produced by the chinese researchers seem to be a million times weaker than the constituent nanotubes.

It was speculated that the increase in tensile strength that occurs after 3 hours of resistive heating was due to "some welding effect may be occurring at the weak connection points of the CNTs during light emission, because these points have a higher resistivity and, as a result, a higher temperature when a current is applied." The reason for the strength shortfall of yarns compared to constituent nanotubes is implied by this as well to the trained reader: The yarns are composed of parellel filaments of varying lengths, held together by simple entangling, rather than chemical bonds or welding. As such a weak application of force can tear the CNT yarn asunder, , ,