Simulation failes

160 views
Skip to first unread message

reisi...@gmail.com

unread,
Feb 4, 2025, 5:10:47 PMFeb 4
to pflotran-users

Hello,

My simulation runs smoothly up to 1 year, but then it fails. However, when I remove the Update_porosity keyword, it no longer fails. Could you help me figure out what the issue might be? I think the problem is volume fraction which goes above the 1.0?

Thank you

Fatemeh

Hammond, Glenn E

unread,
Feb 6, 2025, 10:30:50 AMFeb 6
to pflotra...@googlegroups.com

Fatemeh,

 

It is very possible that the porosity is approaching zero. I would need an input file to run the case myself or screen/file output, ideally the first.

 

Glenn

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of reisi...@gmail.com <reisi...@gmail.com>
Date: Tuesday, February 4, 2025 at 2:10
PM
To: pflotran-users <pflotra...@googlegroups.com>
Subject: [pflotran-users: 8240] Simulation failes

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/9b16e010-63aa-4821-91a5-f99851801f3fn%40googlegroups.com.

Fatemeh

unread,
Feb 6, 2025, 6:23:52 PMFeb 6
to pflotra...@googlegroups.com, pflotra...@googlegroups.com

Dear Glenn,

Thank you for your email, the data I used in the model is confidential and I am not allowedto share. Can I share the results? Porosity by time? Or any other results that you need?

Bests,
Fatemeh



On Feb 6, 2025, at 12:00, 'Hammond, Glenn E' via pflotran-users <pflotra...@googlegroups.com> wrote:



Hammond, Glenn E

unread,
Feb 6, 2025, 6:26:57 PMFeb 6
to pflotra...@googlegroups.com, pflotra...@googlegroups.com

Please send the porosity over time and the screen output (around the time that  the code is failing).

 

Fatemeh

unread,
Feb 6, 2025, 6:38:00 PMFeb 6
to pflotra...@googlegroups.com

Thank you for your prompt reply.

Attached are the images, with the heat map displaying the domain at the final time step where the models successfully converged (1.6 years).

Best regards,
Fatemeh


Screenshot from 2025-02-06 20-05-12.png
Porosity_Z_X.png
Porosity_vs_Time.png



Fatemeh

unread,
Feb 6, 2025, 6:48:07 PMFeb 6
to pflotra...@googlegroups.com

My model represents a 1D transport simulation (no flow) of a cement core plug with 15% porosity, exposed to CO₂-saturated brine at its boundaries.

Best regards,
Fatemeh

Hammond, Glenn E

unread,
Feb 7, 2025, 11:17:36 AMFeb 7
to pflotra...@googlegroups.com

The porosity has not approached zero. Is this simulation, solely reactive transport (no flow) with only diffusion? Can you provide more screen output, going back to several previous time steps. I need to begin from where the simulation was converging to where it fails. It would be best if you sent the output in an ASCII text file.

 

Glenn

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Thursday, February 6, 2025 at 3:48
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8249] Simulation failes

My model represents a 1D transport simulation (no flow) of a cement core plug with 15% porosity, exposed to CO₂-saturated brine at its boundaries.

Best regards,
Fatemeh

 

On Thu, Feb 6, 2025 at 8:07PM Fatemeh <reisi...@gmail.com> wrote:

Thank you for your prompt reply.

Attached are the images, with the heat map displaying the domain at the final time step where the models successfully converged (1.6 years).

Best regards,
Fatemeh




 

Fatemeh

unread,
Feb 8, 2025, 4:03:12 PMFeb 8
to pflotra...@googlegroups.com
Dear Glenn 

Please see the attached output file.

Best regards,
Fatemeh



output.out

Hammond, Glenn E

unread,
Feb 10, 2025, 12:13:56 PMFeb 10
to pflotra...@googlegroups.com

Fatemeh,

 

Based on the information below, there is a major change to the update in reactive transport time step 1673 where the infinity norm of the update is 5 (the code actually truncates to 5 in natural log space). At that point, I believe that the simulation is unrecoverable. It is interesting that the 2 norm of the residual (2r) does not diverge during that step, instead of the following. Unfortunately, there is too little information in the output to help me figure this out. I would need to step through the code in the debugger.

 

Glenn

 

== REACTIVE TRANSPORT ==========================================================

  0 2r: 2.09E-07 2x: 0.00E+00 2u: 0.00E+00 ir: 1.36E-07 iu: 0.00E+00 rsn:   0

  1 2r: 1.71E-08 2x: 8.24E+02 2u: 1.04E+00 ir: 1.18E-08 iu: 6.55E-01 rsn: itol_res

 

Step   1672 Time=  1.64400E+00 Dt=  1.00000E-03 [y] conv_reason: 10

  newton =   1 [    1750] linear =     1 [      1750] cuts =  0 [   9]

  --> SNES Linear/Non-Linear Iterations = 1 / 1

  --> SNES Residual: 0.171419E-07 0.171419E-07 0.117816E-07

  --> max change:  dcmx=   1.6489E-04  dc/dt=   5.2288E-09 [mol/s]

                  dvfmx=   2.9138E-03 dvf/dt=   9.2397E-08 [1/s]

 

== REACTIVE TRANSPORT ==========================================================

  0 2r: 6.65E-08 2x: 0.00E+00 2u: 0.00E+00 ir: 4.30E-08 iu: 0.00E+00 rsn:   0

  1 2r: 3.27E-08 2x: 8.23E+02 2u: 1.01E+01 ir: 2.80E-08 iu: 5.00E+00 rsn: itol_res

 

Step   1673 Time=  1.64500E+00 Dt=  1.00000E-03 [y] conv_reason: 10

  newton =   1 [    1751] linear =     1 [      1751] cuts =  0 [   9]

  --> SNES Linear/Non-Linear Iterations = 1 / 1

  --> SNES Residual: 0.327157E-07 0.327157E-07 0.279507E-07

  --> max change:  dcmx=   7.0903E-02  dc/dt=   2.2483E-06 [mol/s]

                  dvfmx=   2.8546E-03 dvf/dt=   9.0520E-08 [1/s]

 

== REACTIVE TRANSPORT ==========================================================

  0 2r: 5.94E-07 2x: 0.00E+00 2u: 0.00E+00 ir: 5.30E-07 iu: 0.00E+00 rsn:   0

  1 2r: 4.46E-04 2x: 8.22E+02 2u: 1.60E+01 ir: 4.33E-04 iu: 5.00E+00 rsn:   0

  2 2r: 3.13E-05 2x: 8.22E+02 2u: 1.99E+01 ir: 1.98E-05 iu: 5.00E+00 rsn:   0

  3 2r: 2.95E-05 2x: 8.20E+02 2u: 1.94E+01 ir: 1.97E-05 iu: 5.00E+00 rsn:   0

  4 2r: 6.23E-04 2x: 8.21E+02 2u: 1.73E+01 ir: 5.65E-04 iu: 5.00E+00 rsn:   0

  5 2r: 1.92E-03 2x: 8.20E+02 2u: 2.19E+01 ir: 1.85E-03 iu: 5.00E+00 rsn:   0

  6 2r: 1.74E-03 2x: 8.20E+02 2u: 1.66E+01 ir: 1.65E-03 iu: 5.00E+00 rsn:   0

  7 2r: 3.28E+14 2x: 8.21E+02 2u: 2.20E+01 -diverged

-> Cut time step: snes= -9 icut= 1 [10] t= 1.64500E+00 dt= 5.00000E-04 [y]

 

Message has been deleted

Fatemeh

unread,
Feb 10, 2025, 6:22:02 PMFeb 10
to pflotra...@googlegroups.com

Hi Glenn,

I also noticed that changing the number of grids significantly affects the results, but altering the rate constant has no impact.

Best regards,
Fatemeh


On Mon, Feb 10, 2025 at 6:21 PM Fatemeh <reisi...@gmail.com> wrote:

Dear Glenn,

Thank you for your email. I made some modifications to the data and can send the input file. I’m still encountering the same issue with this file.

Please find the attached input file.

Best regards,
Fatemeh


Hammond, Glenn E

unread,
Feb 11, 2025, 2:06:14 PMFeb 11
to pflotra...@googlegroups.com

Fatemeh,

 

We recently refactored the update of mineral surface area. Please pull the latest source code and update your input deck to use the new functionality.

https://documentation.pflotran.org/user_guide/cards/subsurface/chemistry/mineral_kinetics_card.html#mineral-kinetics-card

I believe that you can set the SURFACE_AREA_FUNCTION to VOLUME_FRACTION_RATIO and set the SURFACE_AREA_VOL_FRAC_POWER to 1.0 for each mineral that you want to update. The following is an example

https://bitbucket.org/pflotran/pflotran/src/1a8b9fd282cfba4340ca097748d35d8253c102d4/regression_tests/default/column/mineral_prefactor.in#lines-145

 

You may need to specify the SPECIFIC_SURFACE_AREA for each mineral within the MINERAL_KINETICS block, though you do not use MINERAL MASS as the surface area function…so maybe not necessary.

 

Once you have completed the above, see if you still experience the same issues (most likely, you will still have issues). Then, please send that input deck.

 

Thanks,

 

Glenn

 

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Monday, February 10, 2025 at 1:51
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8258] Simulation failes

Dear Glenn,

Thank you for your email. I made some modifications to the data and can send the input file. I’m still encountering the same issue with this file.

Please find the attached input file.

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.

Fatemeh

unread,
Feb 11, 2025, 5:51:52 PMFeb 11
to pflotra...@googlegroups.com
Dear Glenn,

After adding the functions I received this error:
ERROR: Keyword "SURFACE_AREA_FUNCTION" not recognized in CHEMISTRY,MINERAL_KINETICS

This is the version I am using, should I update? (HEAD -> maint/v6.0, tag: v6.0.0, tag: v6.0, origin/maint/v6.0)


Bests,
Fatemeh 

Hammond, Glenn E

unread,
Feb 11, 2025, 6:24:19 PMFeb 11
to pflotra...@googlegroups.com

cd PFLOTRAN_DIR/src/pflotran

git checkout master

git pull

make clean; make -j4 pflotran

 

 

Glenn

 

Fatemeh

unread,
Feb 11, 2025, 7:11:44 PMFeb 11
to pflotra...@googlegroups.com
Hi Glenn,

I faced the following error:

Screenshot from 2025-02-11 20-40-42.png
Bests,
Fatemeh 

Fatemeh

unread,
Feb 11, 2025, 7:50:49 PMFeb 11
to pflotra...@googlegroups.com
Dear Glenn,

I think I have the latest version now, but still the same problem:

Screenshot from 2025-02-11 21-16-26.png

Hammond, Glenn E

unread,
Feb 12, 2025, 10:59:43 AMFeb 12
to pflotra...@googlegroups.com

The PFLOTRAN header still lists v6.0. You need to recompile:

 

cd $PFLOTRAN_DIR/src/pflotran

git checkout master

git pull

make clean

make pflotran

 

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Tuesday, February 11, 2025 at 4:50
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8266] Simulation failes

Dear Glenn,

 

I think I have the latest version now, but still the same problem:

 

On Tue, Feb 11, 2025 at 8:41PM Fatemeh <reisi...@gmail.com> wrote:

Hi Glenn,

 

I faced the following error:

Bests,

Fatemeh 

Message has been deleted

Hammond, Glenn E

unread,
Feb 14, 2025, 1:50:41 AMFeb 14
to pflotra...@googlegroups.com

Fatemeh,

 

Why not assign the FLOW_CONDITION used to initialize the Sc-CO2 zone INITIAL_CONDITION to a BOUNDARY_CONDITION at the top. That should provide the infinite source that you desire. Is that not possible?

 

Michael can correct me if I am wrong, but for supercritical CO2, you will not use a REFERENCE_PRESSURE or REFERENCE_TEMPERATURE.

 

Glenn

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Wednesday, February 12, 2025 at 6:02

PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8275] Simulation failes

Hi Glenn,

Thank you very much! After implementing the surface area function, my code is now running.

I have a question regarding my setup. I have a supercritical CO₂ cap, similar to the image, and I want to model an infinite CO₂ source. Essentially, as CO₂ is consumed by reactions, it should continuously dissolve into the brine. Could you guide me on how to implement this?

Additionally, could you explain how to set the system pressure and temperature using REFERENCE_PRESSURE and REFERENCE_TEMPERATURE?

Bests,

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.

Fatemeh

unread,
Feb 15, 2025, 3:59:56 PMFeb 15
to pflotra...@googlegroups.com
Dear Glenn

Thank you very much for your help, I will try it. Could  you please guide me on pressure and temperature for ScCO2?

Bests,
Fatemeh

Hammond, Glenn E

unread,
Feb 18, 2025, 11:10:54 AMFeb 18
to pflotra...@googlegroups.com

Michael,

 

Can you provide advice regarding the below? We should be able to assign a SCO2 gas pressure and saturation at the boundary to create an infinite source, correct?

 

Glenn

 

Nole, Michael

unread,
Feb 18, 2025, 12:42:35 PMFeb 18
to pflotra...@googlegroups.com

Applying a two-phase boundary condition on the top boundary with SCO2 mode should work. You’d give it a GAS_PRESSURE, LIQUID_PRESSURE, and SALT_MASS_FRACTION. Just make sure the LIQUID_PRESSURE is a little bit below the GAS_PRESSURE to make it two-phase.

 

Michael

 

reisi...@gmail.com

unread,
Feb 19, 2025, 4:41:49 PMFeb 19
to pflotran-users
Thank you very much Michael and Glenn!

reisi...@gmail.com

unread,
Mar 11, 2025, 7:58:28 PMMar 11
to pflotran-users
Dear Glenn and Michael,

Is there anyway to define the pressure and temperature for the system when I am only using transport mode?

Bests,

Fatemeh

Hammond, Glenn E

unread,
Mar 12, 2025, 4:39:56 PMMar 12
to pflotra...@googlegroups.com

Fatemeh

 

Add:

 

REFERENCE_PRESSURE 2.d5

REFERENCE_TEMPERATURE 35.

 

to the input file. You cannot output liquid pressure and temperature when running solely transport, but I confirmed that they are being used. Happy to enable the output of these variables (to a snapshot or observation file) if you desire the confirmation.

 

Glenn

 

Image removed by sender.

Bests,

Fatemeh

 

 

On Wed, Feb 12, 2025 at 12:29PM 'Hammond, Glenn E' via pflotran-users <pflotra...@googlegroups.com> wrote:

The PFLOTRAN header still lists v6.0. You need to recompile:

 

cd $PFLOTRAN_DIR/src/pflotran

git checkout master

git pull

make clean

make pflotran

 

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Tuesday, February 11, 2025 at 4:50
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8266] Simulation failes

Dear Glenn,

 

I think I have the latest version now, but still the same problem:

Image removed by sender.

 

On Tue, Feb 11, 2025 at 8:41PM Fatemeh <reisi...@gmail.com> wrote:

Hi Glenn,

 

I faced the following error:

Image removed by sender.

Bests,

Fatemeh 

 


Image removed by sender.
Image removed by sender.
Image removed by sender.

 

Fatemeh

unread,
Mar 12, 2025, 5:52:56 PMMar 12
to pflotra...@googlegroups.com

Fatemeh

unread,
Apr 29, 2025, 9:55:39 PMApr 29
to pflotra...@googlegroups.com

Hi Glenn,

I hope you're doing well.

I defined the pressure and temperature in my model using the following keywords:

REFERENCE_PRESSURE 20.68d7  
REFERENCE_TEMPERATURE 93.d0

However, in the .out file, the reported pressure and temperature appear to default to atmospheric conditions (25°C and 1 atm), and the Log(K) values also seem to be evaluated at 25°C.

Could you please advise on how to address this issue?

Screenshot from 2025-04-29 23-24-47.png



Bests,

Fatemeh





Fatemeh

unread,
May 3, 2025, 4:48:17 PMMay 3
to pflotra...@googlegroups.com
Dear Glenn,

I am sending this as gentle reminder.

Best regards,

Fatemeh 

Hammond, Glenn E

unread,
May 6, 2025, 12:20:23 PMMay 6
to pflotra...@googlegroups.com

Fatemeh,

 

Please send me an example input deck that replicates the issue. Depending on the flow mode, the code will use the initial pressure and temperature at the grid cell before it uses the reference values.

 

Glenn

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Saturday, May 3, 2025 at 1:48
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8375] Simulation failes

Dear Glenn,

I am sending this as gentle reminder.

 

Best regards,

 

Fatemeh 

 

On Tue, Apr 29, 2025 at 11:25PM Fatemeh <reisi...@gmail.com> wrote:

Hi Glenn,

I hope you're doing well.

I defined the pressure and temperature in my model using the following keywords:

REFERENCE_PRESSURE 20.68d7  
REFERENCE_TEMPERATURE 93.d0
 

However, in the .out file, the reported pressure and temperature appear to default to atmospheric conditions (25°C and 1 atm), and the Log(K) values also seem to be evaluated at 25°C.

Could you please advise on how to address this issue?

 

Bests,

Fatemeh

 

 

 

 

Error! Filename not specified.

Bests,

Fatemeh

 

 

On Wed, Feb 12, 2025 at 12:29PM 'Hammond, Glenn E' via pflotran-users <pflotra...@googlegroups.com> wrote:

The PFLOTRAN header still lists v6.0. You need to recompile:

 

cd $PFLOTRAN_DIR/src/pflotran

git checkout master

git pull

make clean

make pflotran

 

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Tuesday, February 11, 2025 at 4:50
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8266] Simulation failes

Dear Glenn,

 

I think I have the latest version now, but still the same problem:

Error! Filename not specified.

 

On Tue, Feb 11, 2025 at 8:41PM Fatemeh <reisi...@gmail.com> wrote:

Hi Glenn,

 

I faced the following error:

Error! Filename not specified.

Bests,

Fatemeh 

 


Error! Filename not specified.
Error! Filename not specified.
Error! Filename not specified.

 

Message has been deleted

Hammond, Glenn E

unread,
May 20, 2025, 5:08:40 PMMay 20
to pflotra...@googlegroups.com

Fatemeh,

 

The REFERENCE_PRESSURE and REFERENCE_TEMPERATURE cards must be placed within the SUBSURFACE block. Otherwise, they are ignored. I note that you will run into other error messages when you make this fix, due to logKs not being present for several database reactions.

 

Glenn

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Sunday, May 18, 2025 at 3:15
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8396] Simulation failes

Dear Glenn,

 

I am modeling a batch simulation and I have no flow in my system, please see the attached files.

 

Bests,

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.

reisi...@gmail.com

unread,
Jun 14, 2025, 7:17:35 PMJun 14
to pflotran-users
Dear Glenn,

As I previously mentioned, I was attempting to define an infinite CO₂ source at the boundary in a batch reactor simulation. Since I am using transport mode only (without flow), I’m working within the GIRT Mode.

To represent infinite source of supercritical CO₂ at the boundary, I defined the following constraint:

CONSTRAINT west_constraint
  CONCENTRATIONS   
    O2(aq)    1.0d-20   T        
    H+        1.0d-20   T   
    Ca++      1.0d-20   T        
    Fe++      1.0d-20   T         
    Mg++      1.0d-20   T         
    Mn++      1.0d-20   T        
    K+        1.0d-20   T         
    Na+       1.0d-20   T         
    SiO2(aq)  1.0d-20   T      
    Cl-       1.0d-20   T       
    CO2(aq)   20.0d0     G CO2(g)      
    SO4--     1.0d-20   T      
    Al+++     1.0d-20   T       
  /


I then applied the following transport condition:

TRANSPORT_CONDITION west_transport
  TYPE DIRICHLET_ZERO_GRADIENT
  TIME_UNITS y
  CONSTRAINT_LIST
   0.d0 brine_constraint
    1.d0 west_constraint
  /


However,  I suspect this setup is not accurately representing the system, Could you please guide me on this?

Regards,
Fatemeh

Hammond, Glenn E

unread,
Jun 17, 2025, 7:36:50 PMJun 17
to pflotra...@googlegroups.com

Fatemeh,

 

We equilibrate the SCO2 with the aqueous phase using equations of state

https://bitbucket.org/pflotran/pflotran/src/ea5eefd414c85b903b252f7770c9bf0dbdf48392/src/pflotran/reaction_co2.F90#lines-220

You can run a coupled problem, determine the aqueous co2 concentration in equilibrium with SCO2, and assign that concentration at the boundary for the transport only case. Note that unless you are using deionized water, the concentrations of other species will likely be larger than 1.d-20.

 

Glenn

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of reisi...@gmail.com <reisi...@gmail.com>
Date: Saturday, June 14, 2025 at 4:17

PM

Image removed by sender.

Bests,

Fatemeh

 

 

On Wed, Feb 12, 2025 at 12:29PM 'Hammond, Glenn E' via pflotran-users <pflotra...@googlegroups.com> wrote:

The PFLOTRAN header still lists v6.0. You need to recompile:

 

cd $PFLOTRAN_DIR/src/pflotran

git checkout master

git pull

make clean

make pflotran

 

 

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Fatemeh <reisi...@gmail.com>
Date: Tuesday, February 11, 2025 at 4:50
PM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Subject: Re: [pflotran-users: 8266] Simulation failes

Dear Glenn,

 

I think I have the latest version now, but still the same problem:

Image removed by sender.

 

On Tue, Feb 11, 2025 at 8:41PM Fatemeh <reisi...@gmail.com> wrote:

Hi Glenn,

 

I faced the following error:

Image removed by sender.

Bests,

Fatemeh 

 

Image removed by sender.
Image removed by sender.
Image removed by sender.

 

 

On Thu, Feb 6, 2025 at 7:56PM 'Hammond, Glenn E' via pflotran-users <pflotra...@googlegroups.com> wrote:

Fatemeh

unread,
Jun 23, 2025, 3:57:04 PMJun 23
to pflotra...@googlegroups.com
Hi Glenn,

Thank you for your email. Just to confirm, the steps I should follow are as follows right?

  • Set up a coupled simulation (flow + reactive transport).

  • Use a small domain with SCCO₂ in contact with brine.

  •  After equilibrium the end, output the aqueous CO₂ concentration near the SCCO₂ boundary.

  • Use that CO₂(aq) concentration as a fixed concentration boundary in transport-only simulation. 

Also, I have a question regarding the brine composition: if I want to apply a constant CO₂ concentration at the boundary (to define an infinite CO2 Source), but I don’t want the ion concentrations (e.g., Ca++) to remain fixed throughout the simulation, how should I define this? My concern is that using a Dirichlet condition and specifying a value for Ca++ would keep it constant at the boundary for the entire simulation, which isn’t realistic in this case.

Best regards,
Fatemeh


reisi...@gmail.com

unread,
Jun 27, 2025, 6:50:52 PMJun 27
to pflotran-users
Dear Glenn,

I am sending this as a gentle reminder.

Thank you for your time and consideration.

Regards,
Fatemeh 

Hammond, Glenn E

unread,
Jun 28, 2025, 10:31:59 AMJun 28
to pflotra...@googlegroups.com
Fatemeh,

  1. Set up a batch SCO2 + reactive transport simulation with a single grid cell and equilibrate.
  2. Assign the solute concentrations at the boundary of the transport only simulation

To not constrain the other species concentrations (e.g,. Ca++), the code would need to be modified.

Glenn

reisi...@gmail.com

unread,
Jun 30, 2025, 2:48:58 PMJun 30
to pflotran-users

Dear Glenn,

I followed the two steps you suggested, and it worked very well, thank you so much.

However, could you please provide further clarification on how to not constrain the other species concentrations?

Best regards,
Fatemeh

Hammond, Glenn E

unread,
Jul 2, 2025, 11:26:13 AMJul 2
to pflotra...@googlegroups.com
Fatemeh,

PFLOTRAN would need to be modified to constraint solely one species at the boundary while calculating the other species concentrations based on concentrations in the boundary cell. This is not currently supported.

Glenn

Fatemeh

unread,
Jul 2, 2025, 6:16:38 PMJul 2
to pflotra...@googlegroups.com
Dear Glenn,

Thank you for your email. If I set the initial brine composition throughout my entire system and specify a very small amount of species (~0) at one boundary (which I want to define the infinite source of CO2), like how I set up my model. Could you help me understand why this approach isn’t working?

Regards, 
Fatemeh

Fatemeh

unread,
Jul 7, 2025, 1:19:12 PMJul 7
to pflotra...@googlegroups.com

Dear Glenn,

I think I understand, using this approach to set up the model creates an extremely steep concentration gradient. Since my experiments are batch experiments with no flow, would using a FLOW_CONDITION to define the scCO₂ gas be an appropriate way to simulate my experiments (using a flow condition to define my infinite SCSO2 source)?

Best regards,
Fatemeh

Hammond, Glenn E

unread,
Jul 8, 2025, 11:17:58 AMJul 8
to pflotra...@googlegroups.com
Fatemeh,

I have very little information for consideration, but I suspect that the small concentrations at the boundary are causing solute to diffuse out of the domain. Without modifying the code, you cannot hold solely CO2 concentration constant at the boundary.

Glenn

Hammond, Glenn E

unread,
Jul 8, 2025, 11:20:44 AMJul 8
to pflotra...@googlegroups.com
Fatemeh,

I would need for Michael Nole to weigh in on this regarding whether boundary conditions can truly be two-phase with sCO2. If so, that should resolve the issue.

Glenn

Michael Nole

unread,
Jul 8, 2025, 11:57:45 AMJul 8
to pflotra...@googlegroups.com
Yes, you should be able to set a two phase boundary condition with SCO2 mode by specifying a GAS_PRESSURE and CO2_PRESSURE. As long as GAS_PRESSURE is greater than CO2_PRESSURE (the CO2 partial pressure in the CO2-rich phase), the boundary will be two-phase (aqueous/CO2-rich). For an example, you could check out the shortcourse directory in exercises/co2/sco2/buoyant_flow.in. For more details on some of your options with SCO2 mode, I’d refer you to the preprint of our manuscript on SCO2 mode here:


Hope this helps,

Michael

Michael Nole

unread,
Jul 8, 2025, 12:09:57 PMJul 8
to pflotra...@googlegroups.com
Sorry, that wasn’t quite right. You need to specify GAS_PRESSURE and LIQUID_PRESSURE, where GAS_PRESSURE is greater than LIQUID_PRESSURE. GAS_PRESSURE greater than CO2_PRESSURE would give you a gas boundary with some water vapor in it.

Michael

On Jul 8, 2025, at 9:57 AM, Michael Nole <michae...@gmail.com> wrote:

Yes, you should be able to set a two phase boundary condition with SCO2 mode by specifying a GAS_PRESSURE and CO2_PRESSURE. As long as GAS_PRESSURE is greater than CO2_PRESSURE (the CO2 partial pressure in the CO2-rich phase), the boundary will be two-phase (aqueous/CO2-rich). For an example, you could check out the shortcourse directory in exercises/co2/sco2/buoyant_flow.in. For more details on some of your options with SCO2 mode, I’d refer you to the preprint of our manuscript on SCO2 mode here:

Fatemeh

unread,
Jul 8, 2025, 8:14:36 PMJul 8
to pflotra...@googlegroups.com
Thank you very much Glenn, and Michael for your help. I will try it.

You received this message because you are subscribed to a topic in the Google Groups "pflotran-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pflotran-users/kpDclcurJZY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to pflotran-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/D9A21AF3-E888-4AD5-9DCC-E44A52ADCD39%40gmail.com.
Message has been deleted

Michael Nole

unread,
Aug 2, 2025, 2:22:19 PMAug 2
to pflotra...@googlegroups.com
What if you start out with a tiny bit of gas everywhere? i.e., for your initial condition set a GAS_PRESSURE 20.65001d7 instead of CO2_MASS_FRACTION 0.0?

Michael

On Aug 2, 2025, at 10:50 AM, Fatemeh <reisi...@gmail.com> wrote:

Dear Glenn and Michael,

I applied the flow condition to define the infinite CO2 source, but my initial CO2 is being consumed after a few days. It seems the CO₂(aq) is being consumed without any replenishment. Could you please advise on how to fix this?

Best regards,
Fatemeh

<Total_CO2(aq) [M]_vs_Time.png>


On Tue, Jul 8, 2025 at 1:40 PM Michael Nole <michae...@gmail.com> wrote:
You received this message because you are subscribed to a topic in the Google Groups "pflotran-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pflotran-users/kpDclcurJZY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to pflotran-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/D9A21AF3-E888-4AD5-9DCC-E44A52ADCD39%40gmail.com.

-- 
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.

Fatemeh

unread,
Aug 3, 2025, 11:55:04 AMAug 3
to pflotra...@googlegroups.com
Hi Michael,

It solved the issue, thank you very much!

Best regards,

Fatemeh

Reply all
Reply to author
Forward
0 new messages