pH error in OSRT

[Won Woo Yoon]

Hi all,

Currently I'm developing an AI surrogate model for the reaction part of RTM using PFLOTRAN. In the process, I identified an issue with the OSRT mode.

The system is a simple 1D transport problem involving uranium and a single exchange site. I observed that the system's pH could not exceed 7.0 during the simulation, even when I set the initial and boundary condition pH to over 10. Under the exact same condition using GIRT, the model correctly simulated the pH above 7.0.

I suspect this is due to the total concentration masking logic within RStep() in reaction.F90. The code appears to clamp total concentration lower than 1e-40 to 1e-40. Therefore, when the pH is high enough to result in a negative Total H+, it is reset to 1e-40, causing a loss of information regarding Total H+. I was able to obtain the correct pH range after I disabled the line 'rt_auxvar%total(i,1) = 1.d-40'.

I suspect a similar issue might occur for O2(aq), though I have not verified this. 

Input and database I used are attached.

I would greatly appreciate any further discussion to solve this problem.

Best regards,

Won Woo Yoon

[Hammond, Glenn E]

Won Woo,

Thank you for reporting this issue. I will work on resolving it over the coming weeks. It may take some time.

Glenn

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Won Woo Yoon <yys...@yonsei.ac.kr>
Date: Tuesday, November 25, 2025 at 5:26 PM
To: pflotran-users <pflotra...@googlegroups.com>
Subject: [pflotran-users: 8622] pH error in OSRT

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

--
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/48c76990-bb54-43b1-aec3-2b7d155977ecn%40googlegroups.com.

[Won Woo Yoon]

Dear Glenn,

I was wondering if there has been any progress or updates on this issue since.

Best regards,

Won Woo Yoon

2025년 12월 5일 (금) AM 1:40, 'Hammond, Glenn E' via pflotran-users <pflotra...@googlegroups.com>님이 작성:

To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/SA1PR09MB7838B4AD591C5B178BF6067F9AA6A%40SA1PR09MB7838.namprd09.prod.outlook.com.

[Hammond, Glenn E]

Won Woo,

I apologize for responding earlier. You are correct that the truncation is incorrect for total component concentrations that can fall below zero. This was a significant error on my part. I have since modified the code to allow for negative totals for primary species with negative stoichiometric contributions to secondary aqueous complexes. 

https://bitbucket.org/pflotran/pflotran/pull-requests/1354/diff

Please pull the latest source code and see if that resolve the issue.

Regards,

Glenn

To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/CAKarx-xKZL8_EBO_-5sXvewNqiGUkGgNmTEpGNcJP60i%3DAsJ5Q%40mail.gmail.com.