Underground » Ventilation, Gas Drainage and Monitoring
Gas content measurement is a principal tool for the assessment and management of outburst risk by the Australian underground coal industry. Most gassy underground mines use a gas content threshold as a major determinant as to whether mining is permitted to take place. Gas content measurement is also fundamental to the evaluation of mine gas emission and has secondary roles for the assessment of methane resource evaluation and fugitive emission sources of methane for greenhouse gas impacts from mining.
While the fast desorption technique is widely used for gas content measurement, concerns have been raised regarding the reproducibility of results between different laboratories. This resulted in an Australian Coal Association Research Program (ACARP) supported study coordinated by the CSIRO Division of Energy Technology, completed in 1998. The test results indicated individual laboratories were performing measurements such that their own results were within the recommended bounds of error for the procedure (< 10%). However, their results were not being consistently reproduced at other laboratories, with variability of up to 17%.
The underground coal mining industry and the organisations measuring gas content to service the industry are both concerned with the situation. As a result of this study a further project was supported by ACARP. The project aims were defined as:
- Determine the source(s) of the variability between laboratories;
- Propose modifications to the fast desorption test method that will produce a robust test and improve cross-laboratory reliability.
A project team was assembled based on the team that participated in the C6023 study, plus other stakeholders that had been involved in the development of an Australian Standards Guideline for the technique. The study team comprised CSIRO, GeoGAS Pty Ltd, BHP Billiton and the NSW Department of Minerals Resources. The specific laboratories considered in the study were CSIRO Division of Energy Technology, GeoGAS and BHP Billiton Illawarra Coal.
After considering the minor variations in techniques and equipment used at each of the three laboratories, and the nature of the results from the previous study, possible mechanisms associated with individual laboratory practices were, in approximate order of significance:
- Variable partial pressure conditions at the end of the measurement,
- Temperature rise during crushing, creating an overestimate of gas content,
- Dissolution of carbon dioxide in water, leading to underestimation of gas content,
- Oxidation of the coal during the measurement, particularly in the crushing phase, most likely leading to an underestimate of gas content.
It was also recognised that sampling techniques for the coal can also lead to variability in results.
Despite there being the potential for a significant difference in gas content values due to partial pressure, of the order of 1.0 m3/t in some samples, when the particular conditions in the mill for each laboratory are considered along with the experimental results there is little evidence to support the conclusion that it is the principal contributor to the variability between laboratories.
Temperature rise during crushing can lead to over-estimation of gas content by as much as 0.7m3/t. However, this effect is subject to an existing correction, which appears to be adequate to remove the potential for variability.
Cross-laboratory comparison results of gas composition from this project and the previous ACARP project do not support the proposition that there is a significant carbon dioxide dissolution effect in the gas content measurement of the three laboratories concerned.
The conditions at one laboratory provide the greatest potential for coal oxidation. The experimental data suggests that the oxidation at this laboratory leads to underestimations of typically 0.2 to 0.4 m3/t. In many cases, particularly those with higher gas contents, the relative error from this source will be below the 10% target of maximum variability from the Australian Standard.
Overall, the variations in laboratory practice between CSIRO, GeoGAS and BHP Billiton do not appear to provide a single key source of the variability. That is, variability resulting from any of the factors is less than 0.5m3/t in absolute terms. Given the potential for significant gas content gradients, of the order of 0.5m3/t per metre, at locations where samples for gas content testing are taken underground, there is not a strong driver for putting in place processes to correct the minor variability identified in this project.
Two approaches are possible that will remove the coal oxidation or reduce the partial pressure effects from the laboratories. The first is for all laboratories to replace the air in their mills with an inert gas, prior to commencing the final crushing phase of the measurement. The second approach is for all laboratories to replace the air in their mills with the dominant gas of the coal sample, that is CH4 or CO2. As the benefits are expected to be modest, the decision to implement either of these approaches is left to the concerned laboratories.