Abstract— In this study, the phenomenon of coal spontaneous combustion and its influencing factors were revisited with a focus on imported steam coals. A total of 34 steam coal samples were investigated for their spontaneous combustion liabilities through comprehensive experimental analyses. These included proximate analysis, ultimate analysis (C, H, S), grindability analysis (HGI), ash fusibility testing (AFT), ash chemical composition, free swelling index, and petrographic examination (for selected samples). Statistical correlation analyses revealed that moisture and volatile matter contents were the most influential parameters affecting spontaneous combustion, with their product showing a strong relationship with spontaneous combustion indicators. Specifically, predictive models for CPT (Crossing Point Temperature), ATI (Adiabatic Temperature Index), and FCC (Final Combustion Characteristic) were developed using moisture and volatile matter, yielding high correlation coefficients (R² = 0.89 for CPT, 0.83 for ATI, and 0.89 for FCC). CPT values ranged from 147°C to 175°C, while ATI and FCC values ranged between 0.63–1.18 and 3.62–8.03, respectively. Although particle size distribution (PSD) analysis was also conducted, no significant correlation was found between PSD parameters and combustion liability, indicating that low-temperature oxidation may depend more on pore structure and oxygen adsorption capacity than on particle size. This study proposes a practical, data-driven approach for assessing the combustion liability of imported coals, addressing a notable gap in the literature and offering predictive equations applicable in industrial contexts.

Abstract— In this study, the phenomenon of coal spontaneous combustion and its influencing factors were revisited with a focus on imported steam coals. A total of 34 steam coal samples were investigated for their spontaneous combustion liabilities through comprehensive experimental analyses. These included proximate analysis, ultimate analysis (C, H, S), grindability analysis (HGI), ash fusibility testing (AFT), ash chemical composition, free swelling index, and petrographic examination (for selected samples). Statistical correlation analyses revealed that moisture and volatile matter contents were the most influential parameters affecting spontaneous combustion, with their product showing a strong relationship with spontaneous combustion indicators. Specifically, predictive models for CPT (Crossing Point Temperature), ATI (Adiabatic Temperature Index), and FCC (Final Combustion Characteristic) were developed using moisture and volatile matter, yielding high correlation coefficients (R² = 0.89 for CPT, 0.83 for ATI, and 0.89 for FCC). CPT values ranged from 147°C to 175°C, while ATI and FCC values ranged between 0.63–1.18 and 3.62–8.03, respectively. Although particle size distribution (PSD) analysis was also conducted, no significant correlation was found between PSD parameters and combustion liability, indicating that low-temperature oxidation may depend more on pore structure and oxygen adsorption capacity than on particle size. This study proposes a practical, data-driven approach for assessing the combustion liability of imported coals, addressing a notable gap in the literature and offering predictive equations applicable in industrial contexts.