Large-eddy simulations (LES) of turbulent mixing with complex chemical reactions and their experimental verifi cation

Aim of project
For quick homogenous reactions decisive are both the contact type of reagents and the mixing velocity, as well as the kinetics of chemical reactions. Therefore, selection of mixing conditions can also influence selectivity of occurrence of many chemical reactions and their efficiency. In order to design a reactor correctly or to improve the operation of an existing system, normal simulations of system hydrodynamics are conducted, using computational fluid dynamics (CFD). In recent years, these methods have proved to be of wide practical use. On the other hand, development of non-invasive laser measurement methods has enabled a verification of the used models (LDA, PIV and PLIF). It is believed that in case of continua, turbulent flows can be fully described with Navier-Stokes formulas. Attempts of direct numerical solution of these equations with the method of Direct Numerical Simulations (DNS) are limited to relatively small Reynolds numbers and systems of simple geometry. What is left is turbulence modeling and this requires the use of the so-called closing hypotheses. We solve this problem using the moment method or the method of probability distribution function. An example of the first method are k-ε models and models of Reynolds strain; an example of the second method is the use of a predefined function describing the distribution of concentration fluctuation. A compromise between closing hypotheses and DNS is the use of the method of large-eddy simulation (LES). In the project, a closing hypothesis was used for simple and complex chemical reactions in industrial reactors, connected with the LES model, along with experimental verification of computational results.

Expected results
The result of the project will be the elaboration of a mathematical model describing mixing of a fluid, along with chemical reactions occurring in chemical reactors of complex geometry. The calculations will be made using the method of large-eddy simulation. The computational results will be verified experimentally. The obtained results will provide new knowledge, necessary to design processes on an industrial scale.