How to attack complex gas phase combustion systems

Abstract

Starting from detailed investigations of elementary chemical reactions a strategy for experimental and theoretical analysis of complex combustion systems is described. Experimental results on the effect of selective translational excitation of reactants in the reaction H + O 2 → OH + O are compared with theoretical studies on ab initio potential energy surfaces. For the reaction CN + H 2O → HCN + OH direct measurements of thermal rates in shock tubes are compared with data obtained from a laser-photolysis-laser-induced fluorescence technique. The attack of practical combustion problems with homogeneous models including detailed chemistry is described for the NO x reduction in power plants and the autoignition of hydrocarbonair mixtures in relation to engine knock. Next the interaction of elementary chemical processes with steady and unsteady laminar flows is considered. Laminar counterflow diffusion flames constitute an important basis set for the simulation of more complicated turbulent combustion processes. Having a library of precalculated “flamelets” of different strain rate and fueloxidizer mixture compositions the numerical treatment of turbulent flame structures can be performed on existing computers. Detailed calculations on pure and partially premixed CH 4air counterflow diffusion flames are compared with results of nonintrusive CARS measurements for various strain rates. For the unsteady case as a simple test system the ignition of O 2O 3 mixtures by irradiation with a CO 2 laser along the axis of a cylindrical vessel is considered. Mathematical simulation of the ignition process is done by solving the corresponding system of conservation equations. Spatial retization using definite differences leads to a system of ordinary differential and algebraic equations that can be solved numerically. Experimental data are presented for velocity components of the flame front from IR-UV double resonance experiments. The last part describes two approaches for linking complex chemistry to models of turbulent flow field predictions applied to an enclosed turbulent jet diffusion flame. One is a conserved-scalar approach employing a stretched-laminar-flamelet model, and the other involves a direct closure of the chemical production terms by a pdf method. Numerical results for turbulent COair diffusion flames and experimental results on imaging of laminar and turbulent flame fronts by OH radical and acetaldehyde fluorescence in an internal combustion engine are presented.