Dear Lorenzo, You might be interested to look at the implementation of adsorption (following very much the approach described by Timo) for the oil-water system with discrete fractures:
https://git.iws.uni-stuttgart.de/andrian/dumux/tree/master/dfm-ow3c Best regards, Nikolai -----Original Message----- From: Dumux <dumux-boun...@listserv.uni-stuttgart.de> On Behalf Of Timo Koch Sent: Monday, February 17, 2020 08:57 To: DuMuX User Forum <dumux@listserv.uni-stuttgart.de> Subject: Re: [DuMuX] adsorption/desorption term Viele Grüße, Timo > Am 17.02.2020 um 07:55 schrieb lc <campoli.lore...@gmail.com>: > > Dear Dumux community, > > speaking about adsorption in Dumux, I'd like to ask some clarification. > > With respect to the following answer and considering a 1p2c model, in > equilibrium. > > I need to implement an adsorption term such as d(c+A(c))/dt ... where c is > the generic concentration (mass or molar fraction component) and A is the > adsorption function term, which could Langmuir-like A(c) = k1*c/(1+k2*c). > > So, clearly, this term should enter in time derivative and cannot be a > source/sink term. Does the computeStorage function fit for this? As I > understand it may compute the partial time derivative (rate of change) of all > conservation quantities in the sub-volume, so it should be appropriate. But > maybe it depends on the model? > > Which custom LocalResidual should I consider for my 1p2c equilibrium model? > Dear Lorenzo, you implement a new LocalResidual class which inherits from the CompositionalLocalResidual and just overloads the computeStorage function. In the implementation of computeStorage you can call CompositionalLocalResidual::computeStorage first and then add your A(c) term to the storage. Then set the type of the property LocalResidual to your new class. Good luck Timo > > Thank you, > > Lorenzo > >> On 18.11.2019 18:58, Timo Koch wrote: >> 3) Depends on if you want to model an equilibrium process or a >> non-equilibrium process. For non-equilibrium, you can use the mineralization >> models in Dumux (e.g. 2pncmin). They add another equation for balancing the >> mass of surfactant adsorbed on the solid matrix. The exchange between solid >> and fluid is implemented as a regular source term in the "Problem" class. If >> you want an equilibrium process, you probably just want to add something to >> your storage term. To this end, you can implement a custom "LocalResidual" >> and add the term in the function "computeStorage". > _______________________________________________ > Dumux mailing list > Dumux@listserv.uni-stuttgart.de > https://listserv.uni-stuttgart.de/mailman/listinfo/dumux _______________________________________________ Dumux mailing list Dumux@listserv.uni-stuttgart.de https://listserv.uni-stuttgart.de/mailman/listinfo/dumux _______________________________________________ Dumux mailing list Dumux@listserv.uni-stuttgart.de https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
