Coal Preparation » Fine Coal
This project has explored the use of an automated sample changer on a micro-CT unit to enable faster density analysis. The intention was to develop the methodology for faster partition curve analysis for fine coal density separation units.
Early testing of the 16-sample autochanger unit purchased to fit an existing micro-CT unit suggested that the automation of the scans could be readily accomplished. However, it was also found that simply scanning samples of -2+0.125mm material did not yield particularly good results. This uncertainty resulted from the presence of larger particles (or lack of them) having an over-riding influence on density distribution.
The final methodology used four size fractions (-2+1mm, -1+0.5mm, -0.5+0.25mm and 0.25+0.125mm) scanned using a previous “pre-separation” technique on a adhesive Styrofoam film.
A comparison of the micro-CT generated data with commercial float/sink results showed similar broad trends across a Spiral sample suite of Feed, Product, Middlings and Reject.
The density distributions were then used to perform a partition curve analysis to evaluate the impacts carried through with the micro-CT data.
The project also showed that it could overcome beam hardening issues with high density minerals by altering the scanning conditions.
Overall, this study has shown that an automated micro-CT can provide faster analysis for partition curve evaluation than commercial float-sink testing. The use of sized fractions was found to improve particle resolution but also generated a significantly improved data set for partition curve analysis. The generated data on every individual particle provides for a wealth of information beyond “floatsink” with measured density metrics being within the reproducibility values of current standards.