(490a) Thermal Stability of Hydrcarbon Fuels

The increasing use of simulations as a tool in combustion technology has focused interest in the use of fundamental or transferable inputs into these programs. The is a reflection of the great advances in Computational Fluid Dynamics that can combine fluid dynamics and realistic chemistry. The thrust of this work is to aid in the development of the fundamental chemical kinetics information base that are necessary ingrediants into such programs. The unique aspect is the emphasis on real fuels. It is obvious that a kinetic database based on real fuels will be able to simulate their behavior in real combustion devices. This means the creation of a new tool for innovation and optimization. It will bring combustion technology to the stage of other cutting age areas where the design process is intimately linked with simulations.

Real fuels are complex mixtures of many organic compounds. They belong to a very limited set of classes of compounds. There is agreement among those who work in the fuel area that a limited palette of compounds can effectively represent the combustion behavior of real fuel mixtures. This has become an increasing focus of research in combustion technology. In terms of chemical kinetics database development the emphasis is on fundamental and transferable reaction rate constants and expressions. This work is focused on increasing coverage to include the newer and larger structures when real fuels are involved. A particular challenge is the size of the molecules. The need is for fuels containing as many as 16 carbon atoms (hexadecane in diesel). It is necessary to determine correlations for the various types of reactions that are possible, At the experimental level the work is similar to that carried out to determine Structural Activity Relations. For these relatively simple systems, transition state theory serves as a guide and the added features arising from chemical activation, fall-off effects and tunneling can be used to further refine the results. An important feature of fundamental or transferable kinetic data is their use in mixtures. It is simple to consider new formulations and compounds by the use of different mixture fractions or through addition of new species into the database. The present work has focused on pyrolytic decomposition. The rationale for this emphasis is the lack of attention on this aspect of combustion in current databases. They have focused on oxidative degradation since the emphasis is on matching ignition behavior. The processes involved in pyrolysis are of key importance for the treatment of Soot/PAH formation, since the products of such reactions are their precursors. A pyrolysis database complements and extends the range of current fuel mechanisms. From an experimental point of view it is difficult to study oxidation at high temperatures without a thorough understanding of the competing pyrolytic processes.

The quantitative details of the kinetics of radical forming processes from fuel molecules are well established. The present work on mechanism and rate constants of fuel radical degradation is a natural follow up to the earlier studies. In practically all cases fuel radicals are generated from an appropriate organic iodide. Advantage is taken of the weak C-I bond and the appropriate radicals are released into the high temperature shock tube environment. The analysis of this data and associated information is the focus of the present work. The results are unimolecular rate expressions for beta bond scission and isomerization. The latter is a particular challenge due to the necessity of considering tunneling. This has not been considered earlier and is a major scientific accomplishment of this work and has now been confirmed by ab initio calculations.

The work under this project involves the analysis of the data in terms of the rate constants for the various decomposition and isomerization channels. The results include not only the high pressure rate expressions but also the dependence of the rate constants on pressure obtained through the solution of the master equation. It is clears that the amount of work required to analyze each set of data increases dramatically as the radical becomes larger and, through substitution, loses symmetry.

In terms of the compounds in a particular fuel mixture, the radicals are derived from linear, branched (as from Fischer-Tropsch fuels), cyclic and olefinyl fuels. The only important class of compounds not treated are the branched aromatic fuels. The data on the linear fuels and properties of the reaction of aromatic compounds in the literature is sufficient to make predictions regarding the radicals derived from such compounds. General features of the results provide a guide to the soot forming proclivities of the various compounents of a real fuel mixture.