Reaction heat
 

Good morning,
Once again, I will disturb you with my CFD model for a packed bed reactor.

I am implementing a CFD model for a packed bed reactor (PBR) for methanol production and comparing it with the performances predicted by a simpler 2D model.
When I switch the chemical reactions off, the two systems perform basically in an identical way. In this case, the chemical reactor behaves as a sort of heat exchanger, where the inlet mixture is progressively heated up by the medium that surrounds the reactor tube.

Then, when I try to add the chemical reactions to the CFD model, the behaviour of the system is completely different from what is normally expected from a chemical reactor (and from my 2D model). I start by saying that the chemical reactions implemented in the CFD model are correct and identical to the ones I have used in my simplified 2D model (I have checked them many times with the "Evaluate" tab), so the error should not lie there.
More in detail, even if I switch the chemical reactions on but I do not introduce any Energy "Source" in the model, the system undergoes strong increases in the temperature profiles, as if heat were produced by the chemical reactions. Moreover, such heat production is not compatible with the involved reactions, as it is definitely too high (3 / 4 times bigger than expected). Two questions arise at this point:
1) Is CFX capable of predicting the reaction heat once chemical reactions are provided and hence there is no need to specify it? In other words, when I select the material "Sources" option in my subdomain (i.e., species formation / consumption due to the reaction), there is no further need to tell the program the corresponding variation of heat?

2) Even if the answer to the first question is positive, why is CFX predicting such a high value for the reaction heat?

I have attached the file of interest (zip. folder).
Let me thank you in advance for your help.

Have a nice day,
Riccardo

Interesting, you decide to go manual with the reaction model; therefore, you now have to do a lot more work, i.e. verify all your sources are introduced correctly in all the proper equations.

Since you have not posted the reaction details, and not trivial to rebuild them from the expressions, I am surprised CH4 will pass through the system without reacting at all with any of the other species (Missing source in the sub-domain)

Also, you also be careful on how the properties are defined for each material you are including in your mixture. Since you have no energy source, you are modeling the heat of reaction "internally" by using the appropriate reference enthalpy for each material.

Even more concerning is that you added material sources, but no linearization at all (that is a recipe for problems)

Hope I provided enough information for you to diagnose your setup.

Reply
 

Dear Opaque,
thanks a lot for your answer.

As for the reaction details, I have just attached a Word document (2 pages) with all the details of the chemical kinetics utilized. I have checked everything again, but I cannot find any mistake.
Methane (CH4), in my case, is an inert species, since I am modelling a system for the production of methanol (that derives from CO and CO2). Hence, I think that it should be correct not to put any material source related to CH4 in the model.

I decided to go "manual", because I thought I had no alternative (or, better, I was not able to find many). I have tried to utilize the "Reaction" tab, but it seems that pressure is not accepted as a variable there. If I try to put my reaction rate (that, as every reaction rate in heterogeneous catalysis, depends on pressure), I get a message stating that the reaction rate cannot depend on pressure. Are there any alternatives to this solution?

Probably I am making a mistake in the definition of the materials, as you suggested. I have chosen all the materials from the Peng Robinson database of CFX (and I have checked that the reference state is the same for every species: 25 °C and 1 atm). The only exception is methanol (CH3OH), that is not present in the Peng Robinson database, and hence I have taken it from the combustion models (Real gas with Peng Robinson equation of state). But, even for that case, the reference state is set at 25°C and 1 atm. The only difference is that the Nasa Format is utilized instead of the 4th order polynomial. Could this be the issue?

As for linearization (I hope I got your point), as I am really a beginner, I did not know what to put as a source coefficient (I have also tried to read the reference guide on the topic, but I was unable to find a proper number).

I have worked on that reactor for a few weeks, but there are still some issues (i.e. thermal profile) that I cannot fix (and have no idea on how to do it!).


Thanks,
Riccardo

If you are a beginner using Ansys CFX, you are doing really well, but you are trying to go too fast.

Do NOT use real gases data until you get standard ideal gas working. You are introducing too many unknowns in your model w/o the proper expertise to understand how to tame those issues

Use the KISS approach (https://en.wikipedia.org/wiki/KISS_principle ).. What you learn by doing incremental steps will save a ton of time in the long run.