Application of Automatic Fragment Modeling in Plastic Pyrolysis

Detailed kinetic models are well developed for predicting light hydrocarbon systems. However, for more complicated modeling processes such as waste plastic pyrolysis, new approaches are required. To overcome difficulties including exponentially growth of number of possible isomers and reactions or the uncertain ability of extrapolation for lumping methods, we develop a fragment-based framework called Auto-Fragment Modeling which targets to build detailed kinetic models for heavy hydrocarbons. This proposed fragment-based method treats large molecules as independent fragments. Each fragment represents different part of the original molecules which share common structures. It is assumed that similar structures will conduct similar reactions.

This method combines with open-source software Reaction Mechanism Generator (RMG) software package and estimates rate coefficient and thermochemistry parameters derived from high-accuracy quantum chemistry. Previously we demonstrated several examples to show promising results by capturing important chemistry as well as product distributions. The new application system elucidates the reaction mechanisms leading to several of the functional groups formed in this important process. Some technical challenges that still need to be overcome to allow accurate predictions for systems containing macromolecules are identified.