First, physical properties of nanotubes:

Nanotubes are either single-walled (SWNT for short) or multi-walled (MWNT).
Young's modulus for SWNTs ~= 1 TPa
                for MWNTs = 1.28 TPa

==> these values are for applying sideways loads to the end of a supported beam.  The values for compression and extension are different, I think.


The wall thickness of a SWNT is 0.335nm.

SWNTs have a diameter of from 1-~7nm
MWNTs have an outer diameter from I think <4nm to 33+nm.
  The inner diameter is between 1-6nm, with the average ~2nm for a MWNT of 10nm in diameter

The lengths of SWNTs are up to at least 10nm, possibly much longer
The lengths of MWNTs are up to at least several microns.

Carbon nanotubes behave like classical beams if they have only small deflections, i.e. for a tube loaded at one end and clamped at the other,
deltamax = F * (L^3 / 3*E*I)
where F = applied force, L = length, E = Young's modulus, I = moment of inertia = pi*(Router^4 - Rinner^4)/4

After a certain point, they develop ripples along their inner edge, at which point the effective spring constant drops a lot.  I suspect if nanotubes were used as spines, they would not be bent enough to cause this rippling.  
One paper had this buckling occuring with a deflection of ~160nm with a tube length of 813nm and diameter of 32.9nm [1].

Also, carbon nanotubes can severely bend (like 180-degree bends) relatively easily (past the rippling point I believe), but can spring back to their original position elastically most of the time.

 

References (not yet complete): [1] : Wong et al. Science; 26 Sept. 1997; vol.277, no.5334, p.1971-5

-- AlanAsbeck - 05 Nov 2004

Continuing with a NanoSpines? page