Biomass is a sustainable and renewable energy source with a huge potential to provide different types of gaseous and liquid biofuels as well as other specialty chemicals. Among several existing techniques to convert biomass into biofuels or chemicals, biomass thermochemical conversion is particularly promising for non-food sources of biomass, such as wood, agricultural residue or energy crops. This conversion process results from a strong coupling among several chemical and physical processes over a wide range of spatial and temporal scales, making it difficult to comprehend. The design and scale-up of the reactors for biomass thermochemical conversion, typically fluidized bed reactors, are mostly based on empirical correlations, relying heavily on expensive and lengthy pilot-scale reactor studies. The recent advances made in high performance computing (HPC) and computational fluid dynamics (CFD), show a great potential in using CFD tools for the design and optimization of these conversion reactors. However, at present, the usage of these tools is severely limited. This work improves the current state-of-the-art modeling and simulation tools for biomass thermochemical conversion by focusing on some of their most limiting aspects. To this end, we focus on the modeling of chemical kinetics and particle-scale processes of biomass thermochemical conversion, and explore the effect of multiphase flow (gas-solid flow) on the particle-scale processes in the conversion reactor. In the first part of the work, a compact chemical kinetic model is developed for the reactions of the biomass conversion and is shown to be computationally affordable to use with CFD tools. The second contribution includes the modeling of particle-scale processes of biomass conversion under large uncertainties in the model parameters, allowing for a rigorous validation study of the model against detailed experimental measurements. Apart from developing models for different processes of biomass conversion, the influence of multiphase dynamics on the particle-scale processes is also evaluated by performing several three-dimensional detailed CFD simulations of a laboratory-scale conversion reactor. As these detailed CFD simulations would be prohibitively expensive to perform for large-scale reactors, a reduced-order model, in the form of a differential equation, is developed to predict the performance of the reactor. The modeling and simulation tools developed in this work allow probing biomass thermochemical conversion process in great detail.