The crossover from single file to Fickian diffusion

Jimaan Sané, Johan T. Padding, Ard A. Louis

    Research output: Contribution to journalConference articleAcademicpeer-review

    24 Citations (Scopus)

    Abstract

    The crossover from single-file diffusion, where the mean-square displacement scales as 〈x2〉∼ t½, to normal Fickian diffusion, where 〈x2 ∼ t, is studied as a function of channel width for colloidal particles. By comparing Brownian dynamics to a hybrid molecular dynamics and mesoscopic simulation technique, we can study the effect of hydrodynamic interactions on the single file mobility and on the crossover to Fickian diffusion for wider channel widths. For disc-like particles with a steep interparticle repulsion, the single file mobilities for different particle densities are well described by the exactly solvable hard-rod model. This holds both for simulations that include hydrodynamics, as well as for those that do not. When the single file constraint is lifted, then for particles of diameter σ and pipe of width L such that (L − 2σ)/σ = δc ≪ 1, the particles can be described as hopping past one-another in an average time thop. For shorter times t ≪ thop the particles still exhibit sub-diffusive behaviour, but at longer times t ≫ thop, normal Fickian diffusion sets in with an effective diffusion constant Dhop-1/√thop. For the Brownian particles, thop ∼ δc−2 when δc ≪ 1, but when hydrodynamic interactions are included, we find a stronger dependence than δc−2. We attribute this difference to short-range lubrication forces that make it more difficult for particles to hop past each other in very narrow channels.
    Original languageEnglish
    Pages (from-to)285-299
    Number of pages15
    JournalFaraday discussions
    Volume144
    DOIs
    Publication statusPublished - 2010
    EventFaraday Discussion on Multiscale Modelling of Soft Matter 2009 - University of Groningen, Groningen, Netherlands
    Duration: 20 Jul 200922 Jul 2009

    Fingerprint Dive into the research topics of 'The crossover from single file to Fickian diffusion'. Together they form a unique fingerprint.

    Cite this