Hi all,

I am concerned with the numerical simulation of incompressible reactive flows. The parallel-competitive reaction scheme (i) A+B->P, (ii) A+C->S where (i) occurs quasi-instantaneous and (ii) is still very fast (chemical time scale much smaller than hydrodynamical time scales) but much slower than (i) means that the ratio of product amounts P/S is a function of the (mixing) underlying flow field, and in particular, of the degree of mixedness on the molecular length scale where chemical reactions occur (so-called "micromixing"). Since the numerical simulations cannot resolve all length scales down to the Batchelor scale (too expensive), micromixing models have to be employed. Now, since the need for such a model stems from the fact that the numerical grid is too coarse, could it be possible to derive a (possibly new?) subgrid-scale (SGS) model for the crucial effect of micromixing, as done in large-eddy simulations (LES) where e.g. subgrid-scale stresses have to be modeled? And could such a SGS model work for LAMINAR flows? My idea is that the principle of LES where one takes the effect of unresolved spatial length scales on the resolved scales into account by means of a SGS model is valid for laminar flows as well - am I right or wrong?

Any comments or suggestions welcome

and many thanks in advance for your help,

Ingo Meisel

Hi Ingo,

The recent book from R.O. Fox 'Computational models for turbulent reacting flows' (Cambridge University Press) might be interesting for you. Most of the developments on SGS models for reacting flows are at the moment happening in the field of turbulent combustion. SGS models using the flamelet approach or the Conditional Moment Closure (CMC) are perhaps interesting, see for instance the article of Steiner & Bushe in Physics of Fluids (2001), Vol. 13, pp. 754-769 and the references therein. More information can be found in the book 'Theoretical and Numerical Combustion' from Poinsot & Veynante (Edwards).

I hope this helps,

Tom

Thank you very much, I'll check it out!

Ingo

By the way, it is not only the micromixing that needs to be modelled in LES. The chemical source term for LES of reacting flows is also a problem, but you will find that out if you read these books and articles.

Tom

Check this out - " Subgrid-scale turbulent micromixingynamic approach" J. Reveillon and L. Vervisch

AIAA Journal, Vol. 36, No. 3, March 1998.