Effect of grid size on diffusion in heterogeneous media

[Won Woo Yoon]

Hello all,

I am currently seeking advice regarding a phenomenon I observed while testing various grid sizes for a diffusion problem in a heterogeneous domain.

The system is divided into a source zone, where uraninite dissolves and releases species, and a bentonite zone, where diffusion takes place.

I observed that using a finer grid size results in a progressively larger amount of uranium being released from the source into the bentonite.

Comparing the results of the attached input files (test_hr_1D_diss.in vs. test_lr_1D_diss.in), the Total UO2++ concentration remains identical within the source zone, while the concentration in bentonite zone is higher in the finer grid case.

This phenomenon occurs when the release from the source is driven by mineral dissolution. In the cases where I assigned a high initial uranium concentration to the source without mineral dissolution (test_hr_1D_const.in and test_lr_1D_const.in), the Total UO2++ profiles were identical across all domains regardless of grid size.

I have attached the input files and the database used.

I would appreciate any advice or suggestions on how to resolve this grid dependency issue.

Best regards,

Won Woo Yoon

[Hammond, Glenn E]

Won Woo,

I apologize or the late response. I left on vacation on Dec 13 and returned yesterday.

There is much information in the below. Can draw a simple diagram of the conceptual model including boundary and initial conditions. See a few comments/questions below in red.

Glenn

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Won Woo Yoon <yys...@yonsei.ac.kr>
Date: Monday, December 15, 2025 at 1:10 AM
To: pflotran-users <pflotra...@googlegroups.com>
Subject: [pflotran-users: 8637] Effect of grid size on diffusion in heterogeneous media

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

Hello all,

I am currently seeking advice regarding a phenomenon I observed while testing various grid sizes for a diffusion problem in a heterogeneous domain.

The system is divided into a source zone, where uraninite dissolves and releases species, and a bentonite zone, where diffusion takes place.

I observed that using a finer grid size results in a progressively larger amount of uranium being released from the source into the bentonite.

If by what you mean by “larger amount of U being release” is that the flux is higher for the finer grid, this is not surprising given a finer grid can handle steeper/larger gradients resulting in larger fluxes, but this should only be the case during early times when the gradient is step. At steady state, the instantaneous flux should be identical (or close to identical).

Comparing the results of the attached input files (test_hr_1D_diss.in vs. test_lr_1D_diss.in), the Total UO2++ concentration remains identical within the source zone, while the concentration in bentonite zone is higher in the finer grid case.

I would need to see the diagram of the conceptual model and associated boundary conditions to provide comment.

This phenomenon occurs when the release from the source is driven by mineral dissolution. In the cases where I assigned a high initial uranium concentration to the source without mineral dissolution (test_hr_1D_const.in and test_lr_1D_const.in), the Total UO2++ profiles were identical across all domains regardless of grid size.

Again, need to see the conceptual model diagram.

I have attached the input files and the database used.

I don’t have time to run the simulations and plot the results.

I would appreciate any advice or suggestions on how to resolve this grid dependency issue.

Best regards,

Won Woo Yoon

--
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/8529f62c-dd0b-4924-8233-10c26ec58212n%40googlegroups.com.