Diffusion dynamics of water controlled by topology of potential energy surface inside carbon nanotubes

The effect of topology of single-walled carbon nanotubes (CNTs) on the diffusion dynamics of water confined in the armchair and zigzag CNTs was investigated. It was found that the activation energy of molecular diffusion in zigzag CNTs is greater than that in armchair ones at similar diameters, which leads to water molecules in zigzag CNT diffusing much slower than in armchair CNT. Further calculations of potential energy surfaces (PESs) of water in these two types of CNTs explain the mechanism of diffusion dynamics of confined water. This research shows that the topology of PESs inside CNT could control the diffusion behaviors of water inside it. Since topology of PESs inside CNT could be modified by various chemical and/or physical methods, this work indicates the possibility of controlling the diffusion behaviors inside the CNTs by molecular engineering approach, without changing its pore size.

Figure 1

Diffusivity components of water confined in (a) armchair and (b) zigzag CNTs as a function of tube diameter for the outer and inner shells at $298\mathrm{K}$ and $1.0\mathrm{g}{\mathrm{cm}}^{-3}$.

Figure 2

Diffusivity components of water confined in armchair (14, 14) and zigzag (24, 0) CNTs as a function of temperatures for the outer and inner shells at $1.0\mathrm{g}{\mathrm{cm}}^{-3}$. (a) In the axial direction and (b) in the radial direction.

Figure 3

Fraction of water molecules confined in armchair and zigzag CNTs as a function of tube diameter for the outer and inner shells at $298\mathrm{K}$ and $1.0\mathrm{g}{\mathrm{cm}}^{-3}$.

Figure 4

(Color online) (a) Experimental and extrapolated diffusion coefficients from Ref. 23 as a function of temperature and comparison with the bulk water. (b) Logarithm of axial diffusivity as a function of the reciprocal of temperature in armchair (14, 14) and zigzag (24, 0) CNTs. (●) simulated in armchair, (—) fitted in armchair, (◼) simulated in zigzag, and (⋯⋯) fitted in zigzag.

Figure 5

(Color online) (a) Frame of the cylindrical coordinates used to describe the molecular probe in the interior surface of both armchair (14, 14) and zigzag (24, 0) CNTs. (b) Snapshot of molecular probe in interior surface of armchair (14, 14) CNT (b1) and zigzag (24, 0) CNT (b2).

Figure 6

(Color online) 3D and 2D plots of potential energy of armchair (14, 14) and zigzag (24, 0) CNTs by MM. The lowest potential energy in the scanning scale was taken to be zero. Blue, green, yellow, and red colors indicate the energy from low to high. (a1) 3D plot of $E(6.29,\theta ,z)$ in armchair (14, 14) CNT. (a2) 2D plot of $E(6.29,\theta ,z)$ in armchair (14, 14) CNT. (b1) 3D plot of $E(6.55,\theta ,z)$ in zigzag (24, 0) CNT. (b2) 2D plot of $E(6.55,\theta ,z)$ in zigzag (24, 0) CNT.

Figure 7

(Color online) Contour plots of potential energy $E(\theta ,z)$ of (a) armchair (14, 14) CNT and (b) zigzag (24, 0) CNT by MM. Arrows indicates the possible diffusion pathway.

Figure 8

(Color online) 3D plots of energy $E(\theta ,z)$ of (a) armchair (14, 14) CNT and (b) zigzag (24, 0) CNT by quantum mechanics in the extended Huckel level of theory. Arrows indicates the favorable diffusion pathway.