Computational modelling of the flow of viscous fluids in carbon nanotubes

Carbon nanotubes will have extensive application in all areas of nano-technology, and in particular in the field of nano-fluidics, wherein they can be used for molecular separation, nano-scale filtering and as nano-pipes for conveying fluids. In the field of nano-medicine, nanotubes can be functionalized with various types of receptors to act as bio-sensors for the detection and elimination of cancer cells, or be used as bypasses and even neural connections. Modelling fluid flow inside nanotubes is a very challenging problem, since there is a complex interplay between the motion of the fluid and the stability of the walls. A critical issue in the design of nano-fluidic devices is the induced vibration of the walls, due to the fluid flow, which can promote structural instability. It has been established that the resonant frequencies depend on the flow velocity. We have studied, for the first time, the flow of viscous fluids through multi-walled carbon nanotubes, using the Euler-Bernoulli classical beam theory to model the nanotube as a continuum structure. Our aim has been to compute the effect of the fluid flow on the structural stability of the nanotubes, without having to consider the details of the fluid-walls interaction. The variations of the resonant frequencies with the flow velocity are obtained for both unembedded nanotubes, and when they are embedded in an elastic medium. It is found that a nanotube conveying a viscous fluid is more stable against vibration-induced buckling than a nanotube conveying a non-viscous fluid, and that the aspect ratio plays the same role in both cases.