To investigate how dynamic fluctuations in oxygen concentration—induced by air leakage flow in the goaf—affect the oxidation and spontaneous combustion behavior of residual coal along the airflow path, particularly considering the catalytic and inhibitory roles of CO and CO2 generated during coal oxidation, a series-connected dual coal sample tank experimental system was developed. Experiments were conducted under controlled thermal conditions: isothermal operation in the upstream coal sample tank and programmed temperature ramping in the downstream tank. Coal oxidation indicators—including O2 consumption rate, CO/CO2 generation profiles, heat release rate, and apparent activation energy—were systematically quantified under dynamically varying atmospheric conditions and benchmarked against those obtained under fresh air and fixed-O2 reference conditions. The results reveal that under dynamic atmospheres—characterized by declining O2 concentration coupled with accumulating CO and CO2—coal oxidation deviates markedly from behavior observed under stable, high-O2 conditions. Crucially, CO and CO2 are not merely passive oxidation products; they actively modulate reaction kinetics. Specifically, they suppress the dominant chain-propagation reactions of low-temperature oxidation, thereby reducing both oxygen consumption and heat release rates relative to fixed-O2 controls at equivalent initial O2 levels. Concurrently, they accelerate the CO-producing pathway, resulting in disproportionately elevated CO yields, even under thermally mild conditions. This decoupling between thermal activity and gaseous hazard implies a heightened risk of CO poisoning and combustible gas accumulation, potentially preceding detectable temperature rise. Accordingly, conventional single-parameter risk assessment frameworks—especially those relying solely on temperature or O2 depletion—are insufficient for early hazard identification in such complex, transient airflow environments. We recommend integrating real-time CO concentration monitoring as a critical, proactive parameter in spontaneous combustion early-warning systems.
xing et al. (Sat,) studied this question.