Nanotubes are being increasingly investigated for use in biomolecule separation devices, as encapsulation media for biomolecule storage and delivery, and as channels for selective and rapid gas flow [1-3]. Many recent theoretical and computational studies have focused on non-diffusive water transport through short (< 10 nm) carbon nanotubes [4-6], as well as gas diffusion in long carbon nanotubes [7]. Hydrophilic metal oxide nanotubes of ~ 10-100 nm length are being synthesized in our group and are being investigated as realistic candidates for the above applications. These nanotubes are attractive candidates for use as nanoscale artificial water/ion channels, nanofluidic devices, and ultra-thin (sub-100-nm) nanotube membranes as a result of their well-defined solid-state structure, hydrophilic interior, tunable length and diameter [8,9], and functionalizable interior for obtaining transport and adsorption selectivity. Furthermore, they are synthesized inexpensively through aqueous-phase chemistry at mild temperatures below 100°C.

Here we present the first computational studies of diffusive water transport in nanotubes of precisely tunable diameters ranging from a diameter of 2.2nm to 3.3 nm. The axial self-diffusivities of water molecules through the nanotubes are calculated at different loadings of water molecules. Transport diffusivities were calculated via the “Darken” approximation using a thermodynamic correction factor obtained from computed adsorption isotherms [10]. The transport diffusivities of water in the nanotubes are comparable to that in liquid water, so that very high diffusive fluxes should be possible. We quantify these predictions with comparisons of the predicted water fluxes through the present nanotube materials with those measured and predicted in carbon nanotubes and zeolites. Furthermore, the transport fluxes depend substantially on the nanotube diameter, which in our system is a well- tunable parameter from an experimental viewpoint. We discuss the extension of these studies to the onset of ballistic transport in short (10 nm) nanotube membranes that can be constructed from realistic nanotube materials; and the possibility of using nanotube functionalization to obtain selective behavior mimicking that of biological ion-selective channels.

References: [1] Miller, S.A.; Young, V.Y.; Martin, C.R. J. Am. Chem. Soc. 2001, 123, 12335. [2] Gao, H. J.; Kong, Y.; Cui, D. X.; Ozkan, C. S. Nano Lett. 2003, 3, 471. [3] Skoulidas, A.I.; Ackerman, D.M.; Johnson, J.K.; Sholl, D.S. Phys. Rev. Lett. 2002, 89, 185901-1. [4] Dujardin, E.; Ebbeson, T.W.; Hiura, H.; Tanigaki, K. Science 265, 1850, 1994. [5] Koga, K.; Gao, G. T.; Tanaka, H.; Zeng, X.C. Nature 412, 802, 2001. [6] Kalra, A.; Garde, S.; Hummer, G. Proc. Natl. Acad. Sci. U.S.A. 2003,100, 10175. [7] Jakobtorweihen, S.; Verbeek, M.G.; Lowe, C.P.; Keil, F.J.; Smit, B. Phys. Rev. Lett. 2005, 95, 044501-1. [8] Mukherjee, S.; Bartlow, V.M.; Nair, S. Chem. Mater. 2005, 17, 4900. [9] Konduri, S.; Mukherjee, S.; Nair, S. Phys. Rev. B (Submitted) 2006. [10] Skoulidas, I.; Sholl, D.S. J. Phys. Chem. 2001, 105, 3151.