Abstract
XRF bulk sorting is put forward as a credible intender for pre-concentration methods in mining. This paper presents a preliminary study employing a bench-scale approach to calibrate XRF surface sensing to a target concentration of interest related to its volume so as to assess the feasibility of a case barren copper ore “slots” to bulk sorting. Firstly, various top sizes, along with different operators and specimens’ weight/volumes, were examined to quantify the minimum readings needed (MRN) for XRF spot scanning at this scale. Multiple linear regression with a correlation coefficient of 0.921 was validated to be a reliable algorithm for calibrating XRF surface response to Cu volumetric concentration, owing to its consideration of elemental interacts. With the aid of graphical representations of calibrated data on recovery/grade curves, cut-off grade of 0.23% was chosen to enable 30% mass below the threshold to be discarded as waste and the remainder to go for further processing.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
