Diffusion Coefficients and Boundary Conditions

[reisi...@gmail.com]

Dear Glenn,

 I would greatly appreciate your advice on two issues:

1. Diffusion Coefficient

   • In PFLOTRAN, is it possible to assign species-specific molecular diffusion coefficients for solutes in brine, or is diffusion treated as a single bulk coefficient for the fluid?

   • If species-specific coefficients are not supported, what would be the best practice for handling ions with different diffusivities?

   • Additionally, the diffusion coefficient defined under liquid properties — should be Dw or  Deff?
     

2. Boundary Conditions for Core Plug in Batch Reactor

   • I am attempting to model a cement core plug immersed in brine within a batch reactor, without advective flow.

   • For this setup, what transport boundary condition would be most appropriate to define at the brine–core interface and within the core domain?

   • Should I apply no-flux boundaries, or would a Dirichlet condition be more suitable?

Best regards,
Fatemeh

[Hammond, Glenn E]

See my answers in red.

Glenn

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of reisi...@gmail.com <reisi...@gmail.com>
Date: Tuesday, August 19, 2025 at 3:38 PM
To: pflotran-users <pflotra...@googlegroups.com>
Subject: [pflotran-users: 8532] Diffusion Coefficients and Boundary Conditions

Check twice before you click! This email originated from outside PNNL.

Dear Glenn,

 I would greatly appreciate your advice on two issues:

1. Diffusion Coefficient

• In PFLOTRAN, is it possible to assign species-specific molecular diffusion coefficients for solutes in brine, or is diffusion treated as a single bulk coefficient for the fluid?

• Yes, it is possible to assign individual diffusion coefficients. By default, all species have the same diffusion coefficient (to ensure charge balance). When individual anions and cations are assigned differing diffusion coefficients, charge balance is not conserved (unless you use the Nernst-Planck capability: https://www.sciencedirect.com/science/article/pii/S0098300422001200 Search on AQUEOUS_DIFFUSION_COEFFICIENTS at https://documentation.pflotran.org/user_guide/cards/subsurface/chemistry_card.html. Also, see the following regression test for an example: https://bitbucket.org/pflotran/pflotran/src/2d5169f01591ed7c34917c211a24d34d6d2a7f6e/regression_tests/default/column/multiphase_transport_pulse.in#lines-61

• If species-specific coefficients are not supported, what would be the best practice for handling ions with different diffusivities?

• Additionally, the diffusion coefficient defined under liquid properties — should be Dw or  Deff?

• Dw. The effective diffusion coefficient is calculated internally as a function of porosity, liquid saturation, tortuosity and the Dw. See Eqs. 11 and 13 at https://documentation.pflotran.org/theory_guide/mode_reactive_transport.html#governing-equations.

1. Boundary Conditions for Core Plug in Batch Reactor

• I am attempting to model a cement core plug immersed in brine within a batch reactor, without advective flow.

• For this setup, what transport boundary condition would be most appropriate to define at the brine–core interface and within the core domain?

• Should I apply no-flux boundaries, or would a Dirichlet condition be more suitable?

• I would apply a DIRICHLET (note: not DIRICHLET_ZERO_GRADIENT) at the boundary.

Best regards,
Fatemeh

--
You received this message because you are subscribed to the Google Groups "pflotran-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pflotran-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/39788e44-20df-4bb3-9fb6-38401cc9c521n%40googlegroups.com.