Search

Functionalisation of colloidal particles with DNA provides a powerful and flexible path towards self-assembly of ordered materials. Given the nearly limitless possibilities for constructing DNA-functionalised particles, and the wide range of conditions under which they can be assembled, it is crucial to gain an understanding of the principles governing self-assembly of these particles and how their properties affect the structures produced. A number of computational models for DNA-functionalised systems have successfully described their properties, and molecular simulation techniques have provided a unique insight into the factors underlying their assembly. Here, we discuss a variety of efforts using simulations to solve an important design problem in DNA-mediated assembly: how the properties of individual DNA-functionalised particles affect their interactions with each other, and ultimately how these interactions determine what structures can be produced. [ABSTRACT FROM AUTHOR]

We perform molecular dynamics simulations to study the transport of geometrically modified water models through channels of carbon nanotube (CNT) membranes. We use two modifications to an existing water model (extended simple point charge SPC/E) as representative surrogates of molecular fluids: (1) bent model (model B) in which the HOH angle is varied while keeping the dipole moment constant by adjusting the OH bond length and (2) modified bent model (model MB) in which the HOH angle changes without any change in OH bond length thereby changing the dipole moment. Interestingly, we find that the fluid transport is a nonmonotonic function of the bond angle for both fluid models. This observed trend is not anticipated based on the fluid density as a function of the bond angle inside and outside of the nanotube channel. However, the average residence time of transmitted molecules through the channel provides an approximately inverse linear correlation with the observed flux, independent of the fluid model. Based on these correlations, we have developed an empirical design parameter connecting fluid transport through CNTs as a function of average occupancy (number of fluid molecules inside the nanotube) and the average residence time. Our results suggest that transport through carbon nanotubes can be sensitive to small changes in the structure of fluid molecules that can potentially be utilised for mixture separation. [ABSTRACT FROM PUBLISHER]