Analysis of Statistically Predicted Rate Constants for Pyrolysis of High-Density Plastic Using R Software

Abstract

The surge in plastic waste production has forced researchers to work on practically feasible recovery processes. Pyrolysis is a promising and intriguing option for the recycling of plastic waste. Developing a model that simulates the pyrolysis of high-density polyethylene (HDPE) as the most common polymer is important in determining the impact of operational parameters on system behavior. The type and amount of primary products of pyrolysis, such as oil, gas, and waxes, can be predicted statistically using a multiple linear regression model (MLRM) in R software. To the best of our knowledge, the statistical estimation of kinetic rate constants for pyrolysis of high-density plastic through MLRM analysis using R software has never been reported in the literature. In this study, the temperature-dependent rate constants were fixed experimentally at 420 °C. The rate constants with differences of 0.02, 0.03, and 0.04 from empirically set values were analyzed for pyrolysis of HDPE using MLRM in R software. The added variable plots, scatter plots, and 3D plots demonstrated a good correlation between the dependent and predictor variables. The possible changes in the final products were also analyzed by applying a second-order differential equation solver (SODES) in MATLAB version R2020a. The outcomes of experimentally fixed-rate constants revealed an oil yield of 73% to 74%. The oil yield increased to 78% with a difference of 0.03 from the experimentally fixed rate constants, but light wax, heavy wax, and carbon black decreased. The increased oil and gas yield with reduced byproducts verifies the high significance of the conducted statistical analysis. The statistically predicted kinetic rate constants can be used to enhance the oil yield at an industrial scale.