Posted on 07/05/2004 12:45:33 PM PDT by LibWhacker
The structure of water inside carbon nanotubes has been debated for several years. Now some experimental light has been shed on the issue.
Proposed structure of water confined within a single-wall carbon nanotube. The interior 'chain' water molecules have been coloured yellow to distinguish them from the exterior 'shell' water molecules in red. Reprinted with permission from ref. 1. Copyright (2004) American Physical Society.
Water held inside carbon nanotubes is very different from normal water, researchers in the USA have found.
They say that it adopts a structure quite unlike that seen in the bulk liquid or in ice. The 'nanotube water' shows 'soft', liquid-like behaviour even at temperatures as low as 8 K. And it displays no abrupt melting transition between a solid and a liquid as it is warmed up1.
This state of water, confined in a nanoscale hydrophobic channel, may sound unearthly, but it could nevertheless be relevant to real-world situations. Water is confined in this manner within pore-like cavities in some membrane proteins, such as aquaporin (which regulates water transport through cell walls), gramicidin and bacteriorhodopsin.
Water in nanotubes has been studied previously by molecular dynamics computer simulations2–5, but there remains no consensus on its molecular structure and behaviour in this environment — partly because water is a notoriously difficult molecule to simulate.
Alexander Kolesnikov of Argonne National Laboratory in Illinois and co-workers have now addressed the problem experimentally, by using neutron scattering to probe the structure and dynamics of the confined state. They use neutron diffraction to discern how the molecules are arranged, and inelastic neutron scattering — a form of spectroscopy, where energy is exchanged between the neutron probe beam and the sample — provides information about the molecular motions. Neutron scattering is ideally suited to this problem because of the strong scattering power of protons, allowing the water molecules to show up clearly.
The researchers made measurements with single-walled carbon nanotubes 1.4 nm wide. They found that the nanotubes seemed to be fully filled, with no excess water on the outsides (which would complicate the neutron-scattering results), at a water/nanotube ratio of 11.3 per cent.
The experimental results are hard to interpret in isolation, and so to assist this, Kolesnikov and colleagues conducted molecular-dynamics simulations of the system. In these simulations, water in the channels adopted a 'shell–chain' structure: a cylindrical shell of 'frozen' molecules hydrogen-bonded in a square grid, and a one-dimensional chain of hydrogen-bonded molecules running down the centre.
The 'square ice' has a quite different structure to that of normal ice, in which water molecules are hydrogen-bonded into hexagonal rings. Nonetheless, it retains fourfold coordination of molecules, as in hexagonal ice. In the central water chain, the coordination is much lower — one would expect it to be twofold, but in fact the average coordination number is around 1.86, because hydrogen bonds are continually being broken and re-forged.
This shell–chain structure matched both the neutron diffraction and inelastic scattering measurements, whereas the square-ice shell alone, a lone water chain, or a tube filling of hexagonal ice did not.
The most striking feature of the work is that the nanotube water is very 'soft', even at low temperatures. This is clear both from the low energies of the hindered rotational movements (librations) of the hydrogen-bonded molecules, relative to bulk hexagonal ice, and from the very large mean-square displacements of the hydrogen atoms, which can be calculated from the elastic (that is, the Bragg peak) scattering intensity of the neutron beam.
Kolesnikov find that at temperatures down to around 50 K the hydrogen atoms in the central water chain seem to fluctuate rapidly between two minima in the potential-energy curve, corresponding to rapid making and breaking of hydrogen bonds. This is more like the kind of dynamics seen in liquid water than in ice. As the temperature rises, this motion becomes even more pronounced as the potential-energy well flattens out. There is no sudden jump in the mean-square displacement, as there is at the melting transition for bulk ice/water.
So whereas confinement of other fluids in nanoscale pores typically leads to an enhanced preference for the dense condensed phases — the phenomena of capillary condensation and capillary freezing — it seems that this situation is quite different for water in nanotubes, owing to the complex, hydrogen-bonded nature of water. It wouldn't be surprising if, in view of the lightness of hydrogen, this picture were to be modified significantly by quantum effects, which the present studies cannot take into account. A quantum treatment of water channels in bacteriorhodopsin, for example, shows that hydrogen atoms can become symmetrically shared between two water molecules6 – something that is predicted to happen in ice only under extremely high pressures.
Bizarre things happen at the nano-scale level: Water that doesn't freeze, non-metals that act like metals and vice-versa, electro/chemical/physical properties gone haywire. I don't pretend to understand it, but it looks like nanotech is going to give us a huge new class of materials with very, very bizarre, otherworldly properties, whose final uses we can only guess at at the moment. Bring it on, I say (I've really got my hopes up for thin, soft, form-fitting, bulletproof, Buck Rogers type space suits)!
Uh-oh. Ice 9 isn't far away...
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.