Multiscale kinetic modeling of biomass fractionation in an experiment: Understanding individual reaction mechanisms and cellulose degradation

To avoid over-consumption and wastage of virgin fibers, the quality of pulp products is as important as their productivity. To this end, both the quality and quantity of pulp should be considered together during the pulping processes. Multiple experimental studies have highlighted that maintaining a high degree of polymerization (DP) for cellulose microfibers in the wood chip ensures a good quality pulp product. However, in the pulping process, the applied reagents and severe conditions can cause a certain degree of cellulose degradation, accompanying unwanted lowering of fiber grades. In order to mitigate cellulose degradation during this step, it is crucial to control the process conditions such as reagent concentration and temperature. Also, to establish the optimum operating strategies, it is necessary to understand how the operating conditions impact the DP of cellulose microfibers. Therefore, we have proposed a novel multiscale model which predicts mesoscopic properties (e.g., the lignin content and fiber morphology) alongside microscopic properties (e.g., the DP of the cellulose microfibers). The proposed model incorporates a multi-layered kinetic Monte Carlo (kMC) framework that allows us to capture the temporal evolution of lignin content, fiber morphology, and cellulose DP, occurring at disparate timescales, as a function of reaction conditions in a computationally tractable fashion. Furthermore, the model predictions are validated with the experimental results so that it gives us a detailed picture of the pulp production processes. Overall with the proposed model, we aim to maximize productivity and maintain a high quality of cellulose fibers from the wood chips during the pulping process. •A multiscale model for cellulose degradation in woodchips is proposed.•Hydrolysis of glycosidic linkages has been considered in the multiscale model.•Properties related to fiber morphology are predicted.•Model predictions are validated with experimental results.•Model’s sensitivity to reaction conditions sheds light into degradation control.