Multiscale Modeling of Plastic Pyrolysis with a Neural Network-Inspired Pyrolysis Kinetic Model and Coarse-Grained DEM-CFD

Jiangfeng Shen,Jianjian Dai,Hairui Lin,Sijie Li,Shanshan Gao,Youwei Cheng,Michael Patrascu,Xi Gao
DOI: https://doi.org/10.1021/acs.iecr.4c01882
2024-01-01
Industrial & Engineering Chemistry Research
Abstract:To further understand pyrolysis kinetics and sand-plastic binary fluidization behavior in fluidized bed reactors, this study proposed a comprehensive mathematical model for investigating the complex multiphase reaction systems. A neural network-inspired pyrolysis kinetic model for high-density polyethylene (HDPE) was developed using experimental data obtained from a thermogravimetric analyzer (TGA) and Pyrolysis-GC-MS (Py-GC-MS) experiments. A coarse-grained discrete element method-computational fluid dynamics (DEM-CFD) fluidization model for sand-HDPE binary mixtures was developed and validated with fluidization experiments. Additionally, a multiscale model was developed for plastic pyrolysis in a fluidized bed reactor by integrating neural network-inspired kinetics, particle-scale model, and coarse-grained DEM-CFD simulations. The validated model was applied for the analysis of particle mixing and segregation, axial distribution and residence time, Lacey mixing index, and pyrolysis products. The findings of this study contribute to the experimental and theoretical foundations required for the design of fluidized bed reactors and the advancement of HDPE thermochemical conversion.
What problem does this paper attempt to address?