NIST Time|NIST Home|About NIST|Contact NIST

HomeAll Years:AuthorKeywordTitle2005-2010:AuthorKeywordTitle

Multiple Parameter Mixture Fraction With Two-Step Combustion Chemistry for Large Eddy Simulation.

pdf icon Multiple Parameter Mixture Fraction With Two-Step Combustion Chemistry for Large Eddy Simulation. (202 K)
Floyd, J. E.; McGrattan, K. B.

Volume 2;

Interflam 2007. (Interflam '07). International Interflam Conference, 11th Proceedings. Volume 2. September 3-5, 2007, London, England, 907-918 pp, 2007.


simulation; mixture fraction; combustion chemistry; formulation; combustion models; laminar flames; diffusion flames; experiments; carbon monoxide; fire models; equations; validation; burners; extinction


A common approach for treating combustion in practical fire models is to use the mixture fraction, a conserved scalar to which all gas species can be related. Typically, infinitely fast chemistry is assumed, in which case the technique works well for fires scenarios in which there is an adequate supply of oxygen. A somewhat more complex approach is to create flamelet libraries that map temperature and mixture fraction to species mass fractions. This has been shown to work well in small scale simulations and is widely used in the combustion community. However, for simulations of fires in large structures, the inability to resolve flame temperatures and scalar dissipation rates, regardless of the turbulence model used, make detailed flamelet models impractical. Therefore, we seek a methodology that allows us to describe incomplete combustion and flame extinction at large scale while staying within the basic framework of the mixture fraction. In the proposed new framework, the mixture fraction retains its classic definition as the mass fraction of gas that originates as fuel. However, with a single value of the mixture fraction it is not possible to account for products of incomplete combustion, or even the mixing of unburned fuel and oxygen. Instead, we need to decompose the mixture fraction into constitutive parts that represent the products of the different reactions. The number of components depends on the complexity of the phenomena. For example, to account for local flame extinction and also the production/destruction of CO, we need to decompose the mixture fraction into three components. This paper will document the new mixture fraction approach and test it against three sets of experimental data of varying scale: a slot burner, a hood experiment, and a compartment fire experiment. All three sets of experiments involve relatively clean burning fuels because the emphasis is on CO, not soot, production.