I meant slightly different mechanism of junction - not to stick the nanotube onto the Si tip edge (as appear in the paper for pyramid-shaped tips) but to stick part of the tube into the bulk of Si tip.
This should promote better static friction within the junction (better stability) in order to liquid imaging and so on (I do not want to use "glue").
The reason I am asking is that I have a plenty of AFM tips, after tribology-related experiments, with radius of curvature of about 50-200 nm. They do not have pyramid shape.
These tips are sticked to levers with relatively low k [N/m] (