Prediction accuracy in modelling beech wood pyrolysis at different temperatures using a comprehensive, CFD-based single particle pyrolysis model

CFD modelling is a novel approach to overcome problems in predicting the pyrolysis outcome in a reliable and repeatable way. It allows through real-time, model-based investigation the assessment of parameters that are impossible to be analysed experimentally. The aim of this study was to establish a comprehensive 2D single particle model of beech wood pyrolysis, which would be a reliable tool for process optimization with respect to the properties of the resulting liquid and solid pyrolysis products (biochar). The model comprised the primary biomass degradation according to the RAC kinetic scheme (48 compounds). Modelled wood cylinders were in a dry state and had a size of O8 mm x 10mm, with 660 kg/m3 bulk density and 1430 kg/m3 true density. The model was validated with experimental data of pyrolysis of dry beech wood cylinders, conducted in a single-particle reactor at 5 different temperatures (300, 400, 500, 700, 900 °C). The validation dataset consisted of the evolution of particle’s center and surface temperatures, mass loss, and composition of 14 evolved volatiles (CO2, CO, H2O, CH4, C2H4, formaldehyde, acetic acid and furfural, among others). Prediction of the particle’s temperature and mass loss evolution were deemed accurate. For all products, up to 500 °C, the predictions differed from the experimental data by a few %. For the temperature range between 700 °C and 900 °C, the model however, strongly over-predicted the yield in bio-oil at the expense of pyrolysis gases. The implemented primary kinetic scheme showed satisfactory results in the investigated temperature range. However, the model did not reflect very well the pyrolysis products evolution at higher temperatures (thermal tar cracking and gasification range), so implementation of accurate secondary kinetic schemes is deemed necessary.