Particle Solver, CHEMKIN

The motive here is to solve for soot (using one of method of moments, sectional methods) for ethylene diffusion flames. I am using a large mechanism consisting of around 2000 species. I had no issues in solving for gas phase conditiom. However, things get really tough when the soot mechanism is coupled. The soot mechanism includes reactions for nucleation, surface growth (HACA), surface condensation (with PAHs) and oxidation. I have tried different things to make it work, but nothing worked for me. Some of them are as follows:

These are some best practices suggested in literature. They include, using smaller scaling factors for moments and surface species, using a value close to 1 for relaxation factor in segregated scheme, higher number of iterations for segregated scheme, allowing higher number of Pseudo time steps, small negative numbers for minimum bonds for species fractions and particle moments. I have also tried using continuations by scaling the surface reaction rate multiplier (SRRM). I mean, the first iteration uses a very small value of SRRM (~1.0E-4) which further is increased to 1.0 in small steps by using several continuations. However, the CHEMKIN's particle solver fails to converge.

Interestingly, I noticed that solution converges when I have only the nucleation reaction. So, I tried to initialize the solution from previously converged solution with the help of "external source for initializing." But things don't work when I include reactions for surface condensation and oxidation. 

I am reaching out to the community for help in resolving this issue. Thank you for your time and consideration.

Tagged:
Sign In or Register to comment.