Numeral simulation of thermal conversion of biomass in a rotary kiln

Abstract

This thesis presents the development of a one-dimensional (1D) steady-state numerical model to simulate the pyrolysis of pinewood biomass in an indirectly heated rotary kiln. The model supports the objectives of the HybrEEn project at the German Aerospace Center (DLR), which aims to integrate renewable energy into biomass thermochemical conversion processes. Thermogravimetric analysis (TGA) data for pinewood, measured at DLR under controlled inert conditions, were used to extract pyrolysis kinetic parameters. A combination of isoconversional methods and master plot techniques was applied to determine a reliable kinetic triplet, namely activation energy, pre-exponential factor, and reaction model, based on experimental decomposition behavior. A temperature-dependent volatile gas yield function was developed to capture the staged release of volatiles during moisture evaporation, active pyrolysis, and passive pyrolysis. The 1D rotary kiln model incorporates conductive, convective, and radiative heat transfer, along with mass loss due to pyrolysis reactions. It was validated under both non-reactive and reactive conditions using published literature data, showing good agreement in thermal and mass conversion profiles. The model simulates the important reactor-scale dynamics, including axial temperature gradients, volatile gas generation, and changes in solid bed height and residence time. A comprehensive sensitivity analysis was conducted to identify the most influential geometric and operational parameters affecting volatile gas yield and thermal performance. Kiln wall temperature, inner kiln diameter, and kiln length were found to have the greatest impact, while inlet solid feed rate, inlet fill fraction, and kiln rotational speed showed moderate effects. Overall, this work provides a systematic methodology for linking lab-scale thermal decomposition data with reactor-scale process simulations. The insights from the sensitivity analysis offer direct guidance for optimizing reactor design and operational strategies to achieve desired volatile gas yields and thermal performance. This computationally efficient and physically representative 1D modeling approach can thus support both academic studies and industrial applications in biomass pyrolysis.