PhreeqcUsers Discussion Forum

Reactive Transport => Reactive Transport Modelling => Topic started by: Hosseif on October 24, 2017, 08:59:01 AM

Title: Correct timestep for RUN_CELLS in reactive-Transport Modelling
Post by: Hosseif on October 24, 2017, 08:59:01 AM
Hi,

I'm using IPhreeqc module with a our transport solver. As stated in the phreeqc manual, every cell has its own number '1' to 'n' and the chemical reactants for these cells are saved as identification numbers 1 through n in an IPhreeqc module. For equilibrium reactions it's working well. Now I want to define kinetic reactions for some cells. My question is regarding the time for kinetic reactions in every cell. The time for kinetic reaction in every cell should be equal to the timestep for transport, right? and if yes, how can I implement it in RUN_CELLS block? is it like following (Assuming that I have no -steps in the kinetic blocks):

RUN_CELLS
-cells 1-n
-time Transport_Time_Step

Thanks in advance for your time,
Hossein



Title: Re: Correct timestep for RUN_CELLS in reactive-Transport Modelling
Post by: dlparkhurst on October 25, 2017, 03:26:56 PM
Yes, you can set the time steps with -time_step in RUN_CELLS. It will override any definitions of time step in KINETICS. Only cells with KINETICS definitions will have kinetics calculations.

Note that PhreeqcRM is a version of PHREEQC designed to couple with transport codes. It should require less work to set up initial conditions and run the set of cells at each transport step. In addition, it includes parallelization by either OpenMP or MPI.
Title: Re: Correct timestep for RUN_CELLS in reactive-Transport Modelling
Post by: Hosseif on October 27, 2017, 11:16:32 AM
Thank you David.

Trying to do a transport simulation in phreeqc. I'm seeing a behavior which seems a bit strange to me. in the attached phreeqc file, in TRANSPORT data block when I set time step as 0.006 seconds, the amount of calcite after nearly 3.5 seconds will be zero. While if I change time step to 0.06 seconds, then at time 3.5 seconds I still have nearly 29 moles of Calcite left!

How this can happen?

2nd question: in addition to example 15, do you have another reactive-transport benchmark (using kinetic reactions)?

3rd question: in my first post above, I used -time in RUN_CELLS and it works; but I did not find such a keyword in manual for RUN_CELLS. Could you please comment on it?

Best,
Hossein
Title: Re: Correct timestep for RUN_CELLS in reactive-Transport Modelling
Post by: dlparkhurst on October 27, 2017, 03:45:08 PM
The flow velocity is directly related to the time step. If you keep the cell lengths the same and increase the time step by a factor of 10, then the flow velocity is 10 times faster.

The manual for version 3 describes RUN_CELLS. It is distributed in the doc directory of the batch version, and is available from the help button in PhreeqcI.

KINETICS is used in examples 6, 9, and 15. I am sure other examples can be found at Tony Appelo's web site hydrochemistry.eu.
Title: Re: Correct timestep for RUN_CELLS in reactive-Transport Modelling
Post by: Hosseif on October 27, 2017, 04:19:38 PM
Many thanks for helpful comments!

Best,
Hossein