On Simulating the Pyroprobe-Induced Pyrolysis of Hydrocarbons with a Model Having a Detailed Finite-Rate Chemical Kinetics Mechanism: Results for 1,5,9-Decatriene

Author(s):

Michael McQuaid, Chiung-Chu Chen, Rose Pesce-Rodriguez, Christopher Stone, and Rodger Cornell

Abstract:

As part of a program to develop and apply machine learning (ML) methods with the potential to significantly reduce the time and cost needed to create detailed finite-rate chemical kinetics mechanisms, we sought to demonstrate a capacity to model the pyroprobe-induced pyrolysis of hydrocarbons with eight or more carbon atoms. A homogeneous reactor model was formulated to represent the process. It was coupled with a chemical kinetics mechanism developed to represent the pyrolytic decomposition of 1,5,9-decatriene and employed to simulate a set of desorption/pyrolysis–gas-chromatography/mass spectroscopy (D/P–GC/MS) experiments whose results were previously reported. Although there were some discrepancies between measurement-derived and model-predicted concentrations of pyrolysis products that remain to be resolved, the overall agreement between the results indicated that the mechanism–model combination was sufficiently representative of the process to warrant use for ML model training. In addition, the model’s results for the temporal dynamics of the experiments were analyzed, and insights were obtained that could be exploited to design D/P–GC/MS heating protocols that better probe the decomposition chemistry of interest.