Atomistic simulation of cellulose pyrolysis in fire conditions using ReaxFF molecular dynamics: Pathway towards a dynamic wood pyrolysis model development

A physics-based framework for modeling Lignocellulosic materials pyrolysis is proposed by analyzing reactive molecular dynamics (MD), ReaxFF, simulations of cellulose under fire-relevant conditions and translating atomistic insights into tabulated descriptors for continuum fire models. As a proof-of-concept demonstration isothermal simulations spanning 700–2000 K are used to illustrate the methodology, showing a distinct decomposition onset at 1300-1400 K, with bond scission and volatile formation occurring above this threshold. Time-resolved analysis of C-C bond breaking yields Arrhenius kinetics with Ea = 89.0 kJ mol −1 and A = 1.28 × 10 13 s −1 (R 2 = 0.92), demonstrating that MD-derived rates can provide quantitatively meaningful parameters for reduced-order models pending experimental validation. The methodology produces temperature-dependent pyrolysis rates, volatile yields, and char fractions derived from atomistic mechanisms, offering a potential physics-based alternative to empirical correlations typically used in computational fluid dynamics (CFD) pyrolysis sub-models. A compact lookup table format maps temperature to normalized pyrolysis rates, species yields, and material properties, providing a direct interface for CFD integration. This approach reframes pyrolysis modeling as an emergent, computable process rather than an imposed kinetic scheme, establishing a transparent pathway from molecular-scale mechanisms to engineering-scale fire simulations for predictive combustion modeling of lignocellulosic materials (e.g., wood).