Gas desorption law of granular coal in negative-pressure environment and calculation of gas loss during negative-pressure sampling

Abstract

• The gas desorption law at different AEPs and NP conditions was revealed. • The variation law of the diffusion coefficient was obtained by inversion. • The method of gas loss in the process of negative pressure sampling is analyzed. To explore the gas desorption law of granular coal during negative-pressure (NP) slag discharge sampling, atmospheric-pressure (AP) desorption, BP desorption and negative-to-atmospheric pressure (NP-AP) desorption experiments were performed. Based on the experimental results, the variation laws of the gas diffusion coefficient in the NP and NP-AP environments were analyzed through numerical simulation, and the variation laws of gas desorption in the processes of NP desorption and AP desorption were comparatively analyzed. The results show that the gas diffusion coefficient under the NP condition conforms to a power function. At the same adsorption equilibrium pressure (AEP), the initial diffusion coefficient monotonically increases with the rise of NP, that is, the NP environment facilitates gas desorption. At the same desorption environment pressure, the diffusion coefficient is smaller and varies more stably under a higher AEP. Within the same NP sampling time, a higher AEP corresponds to a larger volume of cumulative gas desorption, and a higher NP corresponds to a greater gas loss in the sampling process. This proves that shortening the sampling time can reduce the gas loss at the NP stage and weaken the influence on the desorption at AP. During NP sampling, the logarithmic method is more accurate than the t method in terms of gas loss calculation, its relative error ranging from −56.25% to 16.45%. The relative error is pretty small in the high NP environment and decreases with the passage of the sampling time. Based on the linear relationship between the environmental NP value and the calculation error, the gas loss calculation formulas corrected by error analysis were obtained. The research results can serve as a guideline for further mastering the NP sampling technology.