Establishing an evaluation model for coalification degree based on quantitative characterization of nanoporous morphology on coal surfaces of different ranks: an AFM and multiple regression analysis study

Accurate characterization of coal rank is crucial for the efficient utilization of coal resources, yet traditional methods struggle to simultaneously correlate with its nanoscale surface structure. This study aims to explore a novel approach for the quantitative characterization of coal rank by integrating Atomic Force Microscopy (AFM) with multivariate statistical models. We collected nine samples ranging from medium-volatile bituminous coal to anthracite and extracted multidimensional parameters via AFM experiments, including pore quantity, average pore size, areal porosity, shape factor, as well as surface roughness and nano-ring roughness (Rr). The results indicate that as coal rank increases, the pore quantity and areal porosity in coal significantly rise, while the average pore size, surface roughness, and Rr all exhibit a linear decline, revealing a trend toward densification and ordering of the organic matter structure during coalification. Based on these findings, we developed a multivariate linear regression model for predicting coal rank, particularly the “4 R model” that incorporates Ro, surface roughness, and Rr. This model demonstrates significantly improved prediction accuracy compared to traditional models that do not consider nano-ring factors. This study provides new technological tools and theoretical foundations for the scientific classification of coal resources and the investigation of their microscopic mechanisms.