I don't think so. My impression (without reading the technical details) are that the crystal structure of the Al and O atoms in some of the planes seems to be positioning the carbon atoms in just the proper locations to begin the formation of the tube structure. Given the pictures it appears that the nanotubes grow in a linear fashion along the surface of the crystal planes. The advantage appears to be that due to the structure of the crystal planes one gets arrays of nanotubes in linear alignment.

Whether you could use an AFM and a synthesis tip to dope the nanotubes with various atoms/molecules on the surface of the sapphire I don't know. But keep in mind that until we have *extremely* large arrays of parallel AFMs, presumably under independent control, this kind of approach is likely to be much slower than liquid or gas phase solutions using chemistry, particularly if they can get accuracy and/or complexity using self-assembly based approches. What this appears to demonstrate is substrate assisted (pseudo-catalytic?) self-assembly of nanotubes.

So if you attach Freitas' diamond synthesis tip to one of those faces, could you grow a nanotube on top of it?