CHEMICAL AND STOCHASTIC MODELLING OF COMPLEX REACTIONS: A LIGNIN DEPOLYMERIZATION EXAMPLE (MONTE CARLO)

A new approach to the analysis of a broad class of complex reaction systems is described. Both the inherent complexity and physical phenomena associated with the reactions of macromolecular reactants to multicomponent product spectra obscure the underlying pathways, kinetics and mechanisms. Analysis of cogent model systems permits resolution of fundamental information but does so at the expense of practical relevance. The transfer of model system information in the analysis of real systems is thus a key issue. The chemical and stochastic modelling described herein allows development of the rules for this transfer as well as its execution. Lignin depolymerization provides a specific example.;The direct application of model compound information to a macromolecule can lead to erroneous results. A methodology for the incorporation of model compound information into a mathematical model of the reaction of a macromolecule that explicitly accounts for the differences between model compounds and macromolecules was developed. The principles of this model, which utilizes a stochastic interpretation of chemical kinetics, are demonstrated for simple reactions of small molecules and polymers.;This model was used to predict the results of lignin pyrolysis and reaction over a Ni/SiO(,2)-Al(,2)O(,3) heterogeneous catalyst. The model is truly a priori since only model compound information is incorporated. The predictions of this mathematical model can be used to evaluate novel process strategies.;While lignin was chosen to demonstrate this model, the principles and philosophy are intended to be applicable to any macromolecular substrate, not solely to lignin.;The chemical modelling comprises five steps. Scrutiny of lignin structure allowed selection of model compounds (veratrylglycol-b-ether and guaiacylglycol-b-ether) that mimic the important b-ether linkages within lignin. Reaction engineering experiments with these model compounds aimed at resolution of pathways and kinetics allowed the development of strategies to alter the operative pathways of lignin depolymerization. These strategies were then tested experimentally with these model compounds to infer the intrinsic modified pathways of the macromolecule. The model compound information was then assembled into a probabilistic mathematical model to predict actual lignin results.