Underground » Geology
This project was aimed at developing and validating new technologies for detailed non-destructive chemical analysis of cored and in-situ coal seams for the Australian coal industry, using both a laboratory-based core-scanning system (Itrax core scanner) and a hand-held field-portable X-ray fluorescence (fp-XRF) analyser. The results have shown that either technique can be used to map variations in ash percentage, ash composition and total sulphur in coal seams at a much higher spatial resolution than that available from conventional sampling programs, and that the techniques, especially the fp-XRF instrument, provide an improved basis for rapid but relatively comprehensive coal quality assessment in cores and other exposed faces for mine geology and resource evaluation programs.
Both techniques are based on energy-dispersive X-ray fluorescence (ED-XRF) analysis, which measures the abundance of individual inorganic elements based on the X-ray fluorescence spectrum emitted from the sample. Light elements up to and including sodium cannot be measured by the systems used in the study, and magnesium is also below detection limits at the concentrations usually present in coals, unless long counting times are used as was done when using the fp-XRF. However, the concentrations of the other major elements that make up the ash residues of coal, including silicon, aluminium, potassium, calcium, titanium and iron, as well as the total sulphur, can be measured non-destructively within the coal by these ED-XRF analysis techniques, either in cores or other exposed faces.
A suite of Australian and international coals with a wide range of compositional parameters was collected and independently analysed as reference materials for the study. Each of these was prepared as a pressed-powder disk to approximate the packing density in solid coal, and the suite used to develop calibration curves and equations for both types of core-scanning instrument. Tests were also carried out to examine whether suites of synthetic reference standards could be developed, based on blending low-ash coal with different proportions of selected mineral components.
As well as individual profiles of element concentrations, the Itrax laboratory scanner has the capacity to produce optical and X-radiograph images of the core, co-ordinated in position with the element concentration data. This provides a permanent record of the core scanned, and also a more in-depth basis for interpreting the scan results. A limiting factor in the use of the Itrax system, however, is that the core surface exposed to the instrument should be smooth and unbroken; gaps in the core need to be infilled with masking tape or “Blu-tac”, and as a result data are not available from those intervals. Cores that are significantly broken cannot be scanned using the Itrax system.
The Itrax scanner also provides data on Compton backscatter, which is inversely related to the total inorganic oxide components. This parameter may assist in linking Itrax data to profiles from down-hole density logs.
Scans were carried out on a test core with the Itrax instrument using different sampling increments and measuring times, to identify appropriate balances between spatial resolution, measurement precision, and the time and cost of core-scanner analysis. These have shown that an increment spacing of 2 mm provided an adequate resolution of the quality variation in the core studied. Setup, X-radiography and automated measurement of a core 1.75 m long (the maximum for the instrument) at this resolution can be completed in a single overnight (approximately 6 hour) run.
Scanning was also carried out on this and other cores using the fp-XRF analyser, to establish optimum scanning times and point spacings. Measurement times of 30 seconds at each point were found to give adequate results for most elements, and at the same time allowed the individual measurements to be carried out using hand-held operation. Measurement times of 255 seconds were also tested, using a jig to hold the instrument steady; these allowed the reading of magnesium and appeared to provide slightly better results for phosphorus, but produced no significant difference for the other elements studied. Point spacings of 100 mm were used to provide an overview of quality trends in thick coal seams, and spacings of 10 mm (1 cm) were used for more detailed profile studies. Testing of the fp-XRF unit on large coal blocks also showed that the instrument could identify variations in the inorganic chemistry of individual coal lithotypes (e.g. vitrain bands) within the seam, as well as potentially significant variations in the composition of individual non-coal layers.
Check analyses were conducted on samples from several scanned cores using conventional methods, with the results being compared to the composition indicated by density-weighted averaging of measurements over the same intervals obtained by the two scanning techniques. Although there were some variations due to external factors, these showed significant agreement for a range of coal and non-coal materials. While the precision was not as high as that from conventional methods, the scan data allowed the materials to be categorised at a level that is of value for many exploration and mine geology applications.
Several issues were identified that may affect the reliability of the data obtained from the instruments. These include variations in moisture content, especially of claystone bands, and in some cases development of mineral coatings or salt deposits on the measured surfaces. Such factors need to be taken into account in the practical use of these core scanning techniques.
Despite being incomplete due to the absence of data on sodium and magnesium (the latter with the exception of one fp-XRF calibration), the “total oxides” calculated from the scan data (sum of the SiO2, Al2O3, K2O, CaO, Fe2O3, TiO2 percentages as a fraction of the coal) appear to provide an acceptable estimator of the ash yield, as long as other factors that may affect the results are taken into account. With the same proviso, the percentage of sulphur measured by the scanners appears to represent a reasonable estimate of the total sulphur as determined by conventional methods.
Difficulties were encountered in measuring phosphorus in the coals studied with the Itrax, due to the low concentrations typically present, and as such its exclusion is recommended. Although a good relationship was shown in the calibration studies, difficulty was also encountered in measuring potassium below levels of around 1% K2O using the fp-XRF instrument.
Profile data from both instruments can be imported into spreadsheet programs such as Excel, allowing further manipulation, cross-checking, and integration with other information for individual projects. The profiles produced by either system can provide a guide for conventional seam sampling, and may possibly be extended to facilitate tailored mining or preparation strategies for individual coal seams. They also provide a permanent record of the seam profile from a quality point of view, accessible even after the coal itself has been broken up for analysis.