Programs in Physics & Physical Chemistry
|[Licence| Download | New Version Template] aekb_v1_0.tar.gz(62 Kbytes)|
|Manuscript Title: An Open-Source Library for the Numerical Modeling of Mass-Transfer in Solid-Oxide Fuel Cells|
|Authors: Valerio Novaresio, María García-Camprubí, Salvador Izquierdo, Pietro Asinaria, Norberto Fueyo|
|Program title: multiSpeciesTransportModels|
|Catalogue identifier: AEKB_v1_0|
Distribution format: tar.gz
|Journal reference: Comput. Phys. Commun. 183(2012)125|
|Programming language: C++.|
|Computer: Any x86 (the instructions reported in the paper consider only the 64 bit case for the sake of simplicity).|
|Operating system: Generic Linux (the instructions reported in the paper consider only the opensource Ubuntu distribution for the sake of simplicity).|
|Keywords: Solid Oxide Fuel Cell, Multicomponent, Mass Transfer, Porous Media, OpenFoam®.|
External routines: OpenFOAM® (version 1.6-ext) (http://www.extend-project.de)
Nature of problem:
This software provides a library of models for the simulation of the steady state mass and momentum transport in a multispecies gas mixture, possibly in a porous medium. The software is particularly designed to be used as the mass-transport library for the modeling of solid oxide fuel cells (SOFC). When supplemented with other submodels, such as thermal and charge-transport ones, it allows the prediction of the cell polarization curve and hence the cell performance.
Standard finite volume method (FVM) is used for solving all the conservation equations. The pressure-velocity coupling is solved using the SIMPLE algorithm (possibly adding a porous drag term if required). The mass transport can be calculated using different alternative models, namely Fick, Maxwell-Stefan or dusty gas model. The code adopts a segregated method to solve the resulting linear system of equations. The different regions of the SOFC, namely gas channels, electrodes and electrolyte, are solved independently, and coupled through boundary conditions.
When extremely large species fluxes are considered, current implementation of the Neumann and Robin boundary conditions do not avoid negative values of molar and/or mass fractions, which finally end up with numerical instability. However this never happened in the documented runs. Eventually these boundary conditions could be reformulated to become more robust.
From seconds to hours depending on the mesh size and number of species. For example, on a 64 bit machine with Intel Core Duo T8300 and 3 GBytes of RAM, the provided test run requires less than 1 second.
|Disclaimer | ScienceDirect | CPC Journal | CPC | QUB|