Explaining relationships between chemical structure and tar-rich coal pyrolysis products yield based on Pearson correlation coefficient

Abstract

Elucidating the relationship between the structure of tar-rich coal and its pyrolysis products, as well as developing an accurate predictive model for these products, can effectively establish a correspondence between tar-rich coal and high-value utilization pathways. Therefore, this study systematically investigates the structural characteristics of tar-rich coals through proximate analysis, ultimate analysis, and ¹³C NMR, and establishes the correlation between coal structure and pyrolysis products using Pearson correlation analysis. Subsequently, ridge regression is applied to construct a predictive model for pyrolysis products using the structural characteristics of tar-rich coals. The results indicate that there is a significant correlation between the pyrolysis yield, product distribution, and the functional groups of tar-rich coal structure. Specifically, tar yield is highly correlated with V_daf, terminal methyl groups, methyl groups on rings, and oxygen-attached aromatic carbon. The formation of tar components, including light oil, phenolic oil, and naphthalene oil, is influenced by aliphatic carbon, methylene, aromatic methyl, and methoxy. Additionally, the correlations between M_ad, O/C ratio, methyl group on rings (Ar-CH₃), and fixed carbon (FC_d) are closely related to the yields of pyrolysis gas and semi-coke. Based on these correlations, a pyrolysis yield prediction model was constructed, with the R² value exceeding 0.87 and the prediction accuracy greater than 88.06%. This study provides valuable theoretical insights into the directional transformation of oils and gases during pyrolysis.