for better prediction of separated flows. For applications where the universal wall law is not valid, low Reynolds number turbulence models may be used at the expense of more computations. The reactor engineer may be interested in microscale flow characteristics near the feed nozzle in order to understand various selectivity issues in mixing sensitive reactions. An accurate simulation of energy dissipation rates is then of primary importance (these issues will be discussed in detail in the chapter on modeling reactive flow processes). When the objective is to quickly screen alternative reactor configurations to minimize overall mixing time (which is mostly dominated by the large-scale convective flows), it is sufficient to obtain the correct prediction of the mean velocity field. The choice of turbulence model is then not as critical as in some of the cases mentioned above.

For multiphase flow processes, turbulent effects will be much larger. Even oper-ability will be controlled by the generated turbulence in some cases. For dispersed fluid-fluid flows (as in gas-liquid or liquid-liquid reactors), the local sizes of dispersed phase particles and local transport rates will be controlled by the turbulence energy dissipation rates and turbulence kinetic energy. The modeling of turbulent multiphase flows is discussed in the next chapter.

It must be realized that most of the available turbulence models obscure the actual physical processes such as eddies, high vorticity regions, large structures which stretch and engulf and so on. However, the cautious application and interpretation of turbulence models has proved to be a valuable tool in engineering research and design, despite their physical deficiencies. The other important issue one must remember concerns the appropriate selection of a turbulence model to achieve the required objectives. Launder (1989) argued the case of different turbulence models by drawing an analogy between these and a variety of transport machines: from bicycles, automobiles to airplanes. Just as each transport machine has its own role, different turbulence models can play mutually complementary roles in developing quantitative models for simulating turbulent flow processes. The discussion in this chapter, hopefully, provides a road map with which a reactor engineer can plan an exploration of the world of turbulence modeling. Specific case studies of the application of turbulence models to reactor engineering are discussed in Part IV of this book.

0 0

Post a comment