Molecular simulations are a powerful tool and they have provided predictions in a wide variety of physical and chemical systems that have later been verified by experiments. They can even be used to study how experimental probes disturb physical systems and in some cases they can be used to design molecules with desired physical properties.
The correctness of simulation protocol is crucial for obtaining physically correct and predictive results. If not chosen carefully, one can obtain quite spectacular but physically incorrect results. In the reference below [1], we demonstrate that there are critical several issues that need special attention especially in carbon nanotube systems. In particular, the treatment of charges is of extreme importance. In addition, thermostatting in Molecular Dynamics simulations may create additional problems. In addition to carbon nanotubes, our results also apply to biological systems such as aquaporin and other passive channels.
Water molecules inside the carbon nanotube are coloured to better illustrate their motions. The movie covers 2ns.
Reference:
1. Static charges cannot drive a continuous flow of water molecules through a carbon nanotube, J. Wong-ekkabut, M. Miettinen, C. Dias, M. Karttunen, Nature Nanotechnology 5, 555-557 (2010).
http://dx.doi.org/10.1038/nnano.2010.152
Link to this comment:
All Comments (0)