search

PMEAL / OpenPNM

A Python package for performing pore network modeling of porous media
http://openpnm.org
MIT License
453 stars 174 forks source link

Normalized effective diffusivity > 1, is this right? #274

Closed TomTranter closed 9 years ago

TomTranter commented 9 years ago

I've generated a Delaunay network and run a single-phase Fickian Diffusion algorithm on it. The effective diffusivity normalised by phase['pore.diffusivity'] is greater than one in the in-plane directions and less than one in the through-plane direction. To me this suggests that the path taken by the fluid (in-plane) is less than that in open space meaning some errors in my calculation of conduit length. I've been checking the conduit lengths and running the tortuosity algorithm and I get sensible values i.e tortuosity always greater than 1 and correct domain areas and lengths. Can anyone think how else my effective diffusivity could be higher than bulk if that is what phase['pore.diffusivity'] represents? I've uploaded the network and a test script to dropbox if you would like to run the code. https://www.dropbox.com/sh/m59rkb3bdzkuitg/AAA4uYGuHL5pd-o_0ztptTMxa?dl=0

TomTranter commented 9 years ago

I don't think there is anything wrong with the values I'm getting for domain length and area or for throat length and area. I think the problem may be that hydraulic and diffusive conductance are too high as the formula assumes circular throats. If you introduce a shape factor (alpha) such as this one http://journals.aps.org/pre/abstract/10.1103/PhysRevE.71.057301 it roughly halves the conductances and I get effective diffusivity < 1.

jgostick commented 9 years ago

Ah, yah...duh. I should've thought about that. Sorry for wasting your time.

On Wed, Dec 10, 2014 at 6:31 PM, Tom Tranter notifications@github.com wrote:

I don't think there is anything wrong with the values I'm getting for domain length and area or for throat length and area. I think the problem may be that hydraulic and diffusive conductance are too high as the formula assumes circular throats. If you introduce a shape factor (alpha) such as this one http://journals.aps.org/pre/abstract/10.1103/PhysRevE.71.057301 it roughly halves the conductances and I get effective diffusivity < 1.

— Reply to this email directly or view it on GitHub https://github.com/PMEAL/OpenPNM/issues/274#issuecomment-66544635.