I have often touted the importance of sample preparation in mining. By preparing your sample, you can help ensure you are receiving the most accurate results when you are onsite. Portable x-ray fluorescence (XRF) analyzers work best with finely disseminated, homogeneous samples. The samples can be analyzed using plastic or paper bags — or pellets if you have the appropriate equipment. (View this Sample Collection and Preparation Tools video to see how a direct rock sampler as well as a hammer mill kit are used to prepare mining samples for analysis – including crushing, milling, and making pellets.)
Depending on the thickness of bag and its composition, the number of x-rays reaching the detector will be lower than with direct assay. In such cases, calibration of the instrument using “cal-factors” is necessary. However, the overall time savings gained by eliminating various sample preparation steps can be significant in some projects. How can you ensure the time spent preparing the sample is worth it, and the analysis through paper or plastic is accurate?
Testing the Rock Samples
We tested the method using pulverized samples from a sedimentary-hosted lead-zinc deposit. The samples were analyzed in both plastic (0.06 mm thick or 2-2.5 mils, 38 samples) and paper (0.11 mm thick or 4-5 mils; 18 samples) bags and the results were compared to their lab assays. Samples were crushed and pulverized in mill steel to 95% passing 105 micron mesh. The laboratory assay methods used were a combination of instrumental neutron activation (INAA) with a 30g aliquot and inductively coupled plasma optical emission spectrometry (ICP-OES), with a four acid, near total sample digestion of 0.25 g.
The portable analyzer used was equipped with a PIN detector using a filter time of 30 seconds per filter with analysis through the sample bags returned from the lab. The analyzer was operated in different modes, including Mining and Soil. Mining Mode is based on a Fundamental Parameters Calibration and relies on the detector’s response to pure element spectra, whereas Soil Mode uses Compton Normalization where scatter (Compton Peak) in spectrum is related to sample matrix. Mining Mode provides suitable assay data on samples tested in a plastic bag. Preliminary investigation of the samples and their assay data indicate that Soil Mode is not suitable due to high concentration of elements such as Fe, Pb and Zn.
Test Results
Results show there are good correlations between lab assay data and portable XRF results. Mining Mode is a preferred method for elements in the ranges of concentrations shown in the below table. If an application does not need very low levels of detection, the analyzer can be calibrated to the sample through various media. The below table shows recommended bag type based on concentration and composition of samples used in this study. The concentration column shows the maximum concentration in the samples used in this study. Limits of Detection (LOD) are shown for each bag type using Mining Mode.
This case study shows that for many applications, portable XRF analysis of samples through a plastic bag or paper bag yields valuable and reliable data that can be used to make timely decisions in the field. Furthermore, the study documents that Mining Mode (for most sample matrices) is preferred if the user wishes to analyze samples in a bag. See the Effects of Paper and Plastic Bags on the Performance of Portable XRF Analyzers Application Note for more details, including methods, results, and spectra.
nitonxrf says
Hi Gus,
Thanks for your informative comment. As you said, in each case study, it is important to find out the correlation between data from samples assayed in a bag with data from drill core assays. For gold (with very heterogeneous distribution), this correlation would be weak. I am wondering if the results in your case were acceptable? And which elements you were looking for in the samples in bag?
Regards,
Ali
Gus Van Weert says
Hi Ali: Oops, sorry. I was not looking for the gold content, but the silver. This work goes back to the early days of Veladero operation, before the ten million dollar challenge was fomulated. Results were acceptable for silver content; what to do with them was another kettle of fish……
Rgds, GUS
nitonxrf says
As you know, it is always hard to track precious metals mineralization. Strip logs (in drill cores) or contour maps (in surface sampling) may help to see the trend (if there is any). Being familiar with geology and hydrothermal alteration of specific mineralization can be very crucial.
-Ali