Transport diffusivities of fluids in nanopores by non-equilibrium molecular dynamics simulation

Hendrik Frentrup, Carlos Avendaño, Martin Horsch, Alaaeldin Salih, Erich A. Müller

    Research output: Contribution to journalArticlepeer-review

    211 Downloads (Pure)

    Abstract

    We present a method to study fluid transport through nanoporous materials using highly efficient non-equilibrium molecular dynamics simulations. A steady flow is induced by applying an external field to the fluid particles within a small slab of the simulation cell. The external field generates a density gradient between both sides of the porous material, which in turn triggers a convective flux through the porous medium. The heat dissipated by the fluid flow is released by a Gaussian thermostat applied to the wall particles. This method is effective for studying diffusivities in a slit pore as well as more natural, complex wall geometries. The dependence of the diffusive flux on the external field sheds light on the transport diffusivities and allows a direct calculation of effective diffusivities. Both pore and fluid particle interactions are represented by coarse-grained molecular models in order to present a proof-of-concept and to retain computational efficiency in the simulations. The application of the method is demonstrated in two different scenarios, namely the effective mass transport through a slit pore and the calculation of the effective self-diffusion through this system. The method allows for a distinction between diffusive and convective contributions of the mass transport. © 2012 Taylor & Francis.
    Original languageEnglish
    Pages (from-to)540-553
    Number of pages13
    JournalMOLECULAR SIMULATION
    Volume38
    Issue number7
    DOIs
    Publication statusPublished - 1 Jun 2012

    Keywords

    • diffusion
    • Lennard-Jones potential
    • non-equilibrium molecular dynamics
    • slit pore

    Fingerprint

    Dive into the research topics of 'Transport diffusivities of fluids in nanopores by non-equilibrium molecular dynamics simulation'. Together they form a unique fingerprint.

    Cite this