A kinetic reaction model for biomass pyrolysis processes in Aspen Plus. (15th February 2017)

• Record Type: Journal Article
• Authors: Peters, Jens F.; Banks, Scott W.; Bridgwater, Anthony V.; Dufour, Javier
• Is Part Of: Applied energy. Volume 188(2017)
• Page Start: 595
• Page End: 603
• Publication Date: 2017-02-15
• Subjects: Aspen Plus -- Bio-oil -- Lignocellulosic biomass -- Process simulation -- Pyrolysis -- Reaction kinetics; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2016.12.030
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: Predictive kinetic reaction model applicable to any lignocellulosic feedstock. Calculates pyrolysis yields and product composition as function of reactor conditions. Detailed modelling of product composition (33 model compounds for the bio-oil). Good agreement with literature regarding yield curves and product composition. Successful validation with pyrolysis experiments in bench scale fast pyrolysis rig. Abstract: This paper presents a novel kinetic reaction model for biomass pyrolysis processes. The model is based on the three main building blocks of lignocellulosic biomass, cellulose, hemicellulose and lignin and can be readily implemented in Aspen Plus and easily adapted to other process simulation software packages. It uses a set of 149 individual reactions that represent the volatilization, decomposition and recomposition processes of biomass pyrolysis. A linear regression algorithm accounts for the secondary pyrolysis reactions, thus allowing the calculation of slow and intermediate pyrolysis reactions. The bio-oil is modelled with a high level of detail, using up to 33 model compounds, which allows for a comprehensive estimation of the properties of the bio-oil and the prediction of further upgrading reactions. After showing good agreement with existing literature data, our own pyrolysis experiments are reported for validating the reaction model. A beech wood feedstock is subjected to pyrolysis under well-defined conditions at different temperatures and the product yields and compositions are determined. Reproducing the experimental pyrolysis runs with the simulation model, a high coincidence is found for the obtained fraction yields (bio-oil, char and gas), for the water content and for the elemental composition of the pyrolysis products. The kinetic reaction model is found to be suited for predicting pyrolysis yields and product composition for any lignocellulosic biomass feedstock under typical pyrolysis conditions without the need for experimental data.