(368a) CFD Modeling of Turbulent Reactive Flows with Bourne Chemistry Comparisons | AIChE

(368a) CFD Modeling of Turbulent Reactive Flows with Bourne Chemistry Comparisons

Authors 

Yuan, Q. - Presenter, The Dow Chemical Company
Gillis, P. A., The Dow Chemical Company
Kuchibhatla, S. C., The Dow Chemical Company
Pressler, J., The Dow Chemical Company
Reactive Computational Fluid Dynamics (CFD) simulation can be a powerful tool in designing industrial mixers. Many industrially-relevant reactive flows involve multiple parallel, series and competing reactions of different time scales. Slow reactions (Damköhler number<<1) are limited by the kinetics rate. Faster reactions (Damköhler number >>1) are limited by the mixing rate of the reactants. An accurate representation of the key competitive reactions as well as the turbulence effect, in a CFD model or during experiments using a model chemistry, is of key importance for mixer/reactor design.

This presentation aims to both highlight how these issues have been addressed from an industrial standpoint as well as expose the technological gaps that exist in the current experimental and modeling tools. The study case consists of a converging jet in cross-flow and includes a kinetic scheme with series-parallel reactions. A laboratory representation of a portion of the industrial mixer/reactor was constructed and reactive (the fourth Bourne chemistry: 2, 2 DMP and caustic neutralization with acid) flow experiments were conducted. The concentration of the NaOH, which reacts with the limiting reagent (acid), was varied. Several turbulent mixing-limited reactions models (Eddy Breakup Model, Eddy Contact Model and Multiple-Time-Scale Model) were compared with the reactive flow data (2,2 DMP conversion to acetone and methanol).

This work has revealed a number of gaps in the current modeling and experimental approaches. For example, the mixing-limited reagent is often diluted with a solvent and reacted with an excess co-reagent. These engineering controls are not accurately depicted by some of the mixing-limited reaction models. Other key issues include the resolution the turbulent flow field and the need for models to simulate the mixing and reaction of intermediates. This presentation will explain these gaps and invite the academic community to provide solutions.