Discussion Forum

Hello Alberto Amato

Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.

If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.

just a follow-up to my post: I sent an e-mail to the COMSOL support: they confirmed that the CHEMKIN import utility has some bugs that prevent it to correctly interpret pressure dependent reaction rates. A fix to this bug should be available in future updates this fall.

Hi,

I am wondering if you've find a way to avoid this error when importing CHEMKIN format kinetic and thermodynamic data?

now I have the same problem with 4.3b version. I've imported CHEMKIN format of reactions and species thermodynamic input data. When it comes to calculate rate constants for reactions that involve third part, th efunction changes to something strange. For instance if the reaction is H+HCO(+M)<=>CH2O(+M), naming the species as H= s7, CHO= s2, CH2O= s6, and M is the third part element (AR), also we have the value of the Arrhenius expressions : Forward frequency factor = A_f, forward temperature exponent: nf , forward activation energy: Ef
then it gives a forward rate constant as following!!

1.35E12*exp(-Ef_36/(R_const*T))*exp(-5962.2/(R_const*T))*(0.7*c_s1+c_s2+2*c_s3+c_s4+c_s5+c_s6+c_s7+c_s8+c_s9+c_s10+c_s11+c_s12+2*c_s13+c_s14+c_s15+c_s16+c_s17+2*c_s18+3*c_s19+6*c_s20+c_s21+c_s22+c_s23+1.5*c_s24)*10^(log10(exp(-1.5220700152207003E-4*T)+0.7824*exp(-3.629764065335753E-4*T)+0.21760000000000002*exp(-T/271))/(1+(-0.4+log10(1.35E12*T^(-2.57-nf_36)*exp(-5962.2/(R_const*T))*(0.7*c_s1+c_s2+2*c_s3+c_s4+c_s5+c_s6+c_s7+c_s8+c_s9+c_s10+c_s11+c_s12+2*c_s13+c_s14+c_s15+c_s16+c_s17+2*c_s18+3*c_s19+6*c_s20+c_s21+c_s22+c_s23+1.5*c_s24)*exp(Ef_36/(R_const*T))/Af_36)-0.67*log10(exp(-1.5220700152207003E-4*T)+0.7824*exp(-3.629764065335753E-4*T)+0.21760000000000002*exp(-T/271)))^2/(0.75-0.14*(-0.4+log10(1.35E12*T^(-2.57-nf_36)*exp(-5962.2/(R_const*T))*(0.7*c_s1+c_s2+2*c_s3+c_s4+c_s5+c_s6+c_s7+c_s8+c_s9+c_s10+c_s11+c_s12+2*c_s13+c_s14+c_s15+c_s16+c_s17+2*c_s18+3*c_s19+6*c_s20+c_s21+c_s22+c_s23+1.5*c_s24)*exp(Ef_36/(R_const*T))/Af_36)-0.67*log10(exp(-1.5220700152207003E-4*T)+0.7824*exp(-3.629764065335753E-4*T)+0.21760000000000002*exp(-T/271)))-1.27*log10(exp(-1.5220700152207003E-4*T)+0.7824*exp(-3.629764065335753E-4*T)+0.21760000000000002*exp(-T/271)))^2))*exp(Ef_36/(R_const*T))/((1+1.35E12*T^(-2.57-nf_36)*exp(-5962.2/(R_const*T))*(0.7*c_s1+c_s2+2*c_s3+c_s4+c_s5+c_s6+c_s7+c_s8+c_s9+c_s10+c_s11+c_s12+2*c_s13+c_s14+c_s15+c_s16+c_s17+2*c_s18+3*c_s19+6*c_s20+c_s21+c_s22+c_s23+1.5*c_s24)*exp(Ef_36/(R_const*T))/Af_36)*T^2.57)


I appreciate if someone could guide of a way to solve this problem.

Regardsm,
Zari

The post that I wrote a few months ago was for COMSOL 4.2a. I just recently tried to import mechanisms in 4.3b and some errors in the "Import Chemkin files" utility have been corrected compared to the previous version but not all of them as you pointed out. In the end to solve my problem I just corrected manually the wrong reaction rate expressions and verified that a plug-flow calculation with Chemkin gave the same answer as Comsol. My reaction mechanism wasn't too big (only 10 species) so this strategy was manageable.
Hope this was helpful
Good luck!

Hi Alberto,

Thank you so much for your reply.
I have 24 species and 104 reactions. I think it is risky to change the rate expressions manually. The reaction engineering part be solve alone, but when it comes to combine it with CFD module interfaces (Turbulent flow+ heat transfer+ diluted species transport) then I get an error about log 10 of a negative number that comes from the Troe coefficients at rate expressions.

I hope I could fix the errors soon.
Cheers
Zari

I rember having a similar problem, it had something to do with how I set up intial conditions, but I don't remeber exactly, let me know if you don't manage to solve the problem, I can take a better look to what I did. Have you already contacted the Comsol support about this problem? If you haven't, can you do me a favor and contact them? If they see that many people have the same problem hopefully they will eventually correct the software errors (these errors have been around for a few years and it is annoying) Thanks and good luck.

Hi Alberto,

I appreciate if you could again look at your model and tell me how you managed the problem. My Chemkin kinetic s includes 104 reactions (most of them reversible) and 24 species (most of them are radicals and dissociation species ). I thought that if I remove the Troe coefficients from the Chemkin file then I will have the rates according to the Lindemann which does not have a Log 10 and will be a simpler way.
The kinetic was again being solved with this but I still have errors solving CFD, at mass transport level. which points to reaction rate.

I have contacted Comsol support from 2-3 months ago and it\s still an active case :(
May I have your email and send you my model maybe you could have a look at that, if.if.if you have time?

Best,
Zari

Hi everyone,

is there someone with experience of importing chemkin files in comsol 5.0 and with chemical reactions in comsol in general? I'm having a hard time with this and wanted to ask, if someone has an answer. What I wonder is, how three body reactions are calculated in comsol.

What I did was this: I imported the grimech30.dat reaction mechanism and seperated the first three body reaction: 2O + M <=> O2 + M. Now comsol recognizes the species, O, O2 and M. The first two species are clear, but why is "M" considered a species? "M" is just a collision partner and should not be specified. But comsol wants thermodynamic data for it.

Thanks in advance

Göksel Özuylasi