Using TOUGH ECO2N for Core flooding
Hi all,
I am curious to know if TOUGH2 is able to model a core flooding. Suppose that there is a core sample of length 30 CM.
I have tried, however my simulation didn't converge. I really appreciate it if anybody can share his/her experience in this regard.
Thank you,
Best regards
8 replies
-
Do you mean injecting CO2 into a water saturated core or injecting water into a CI2 saturated core? ECO2N should work well for both cases, if temperature and pressure are the reasonable range. Many reasons can cause the convergence problem. It is difficult to tell without looking at the model data.
-
Hi Kenny,
Thanks for your reply. I have attached a copy of input and Mesh file. The injection rate is 12 ml/min, the core length is 30 cm. injection temperature and pressure are 85 oC and 40 bar.
Best regards
-
I forgot to mention that the salinity is 12%, and I have removed it from the input file.
Best regards
-
It seems the temperature dropped to a lower limit ( several degree C) of the ECO2N near the injection gridblocks causes the convergence problem. You need input the corresponding enthalpy (GENER.1 EX) of injected CO2 to prevent temperature drop.
-
Thanks for your useful input.
I added 532000 J/kg for the corresponding injection condition. The time step is quite small. Is it ok?
-
I think the problem is high amount of salt precipitation, it reaches to almost 1 and SG to 0. How can I solve this issue? Should I inject small amount of water with CO2?
-
As the model uses small gridblocks, it is normal to have small time step size. CO2 injection causes dry out near the wellbore and leads to salt precipitation. This could be real situation. You may play around with lower salinity, or lower injection rate. Co-injection with water could be an effective approach for preventing the dry out, but I am not sure.
-
I agree with all Kenny's suggestions, and in addition you could try the following:
1. Different relative permeability function - liquid relative permeability krl controls how much water can flow toward the well and deposit its salt there when it evaporates. I think less mobile water (smaller van Genuchten m value, for example) will produce less salt precipitation. Weaker capillary pressure might accomplish the same thing (less pulling of liquid water toward the well).
2. Let intrinsic permeability be reduced as salt precipitation occurs (IE(11)>0). This will not solve your precipitation problem, but with permeability reduction, pressure will increase for a given injection rate. If pressure gets too big, you know that the injection rate you are using is not realistic.