A comparison of portable XRF and ICP-OES analysis for lead on air filter samples from a lead ore concentrator mill and a lead-acid battery recycler
Martin Harper,Bruce Pacolay,Patrick Hintz,Michael E Andrew
DOI: https://doi.org/10.1039/b518075a
Abstract:Personal and area samples for airborne lead were taken at a lead mine concentrator mill, and at a lead-acid battery recycler. Lead is mined as its sulfidic ore, galena, which is often associated with zinc and silver. The ore typically is concentrated, and partially separated, on site by crushing and differential froth flotation of the ore minerals before being sent to a primary smelter. Besides lead, zinc and iron are also present in the airborne dusts, together with insignificant levels of copper and silver, and, in one area, manganese. The disposal of used lead-acid batteries presents environmental issues, and is also a waste of recoverable materials. Recycling operations allow for the recovery of lead, which can then be sold back to battery manufacturers to form a closed loop. At the recycling facility lead is the chief airborne metal, together with minor antimony and tin, but several other metals are generally present in much smaller quantities, including copper, chromium, manganese and cadmium. Samplers used in these studies included the closed-face 37 mm filter cassette (the current US standard method for lead sampling), the 37 mm GSP or "cone" sampler, the 25 mm Institute of Occupational Medicine (IOM) inhalable sampler, the 25 mm Button sampler, and the open-face 25 mm cassette. Mixed cellulose-ester filters were used in all samplers. The filters were analyzed after sampling for their content of the various metals, particularly lead, that could be analyzed by the specific portable X-ray fluorescence (XRF) analyzer under study, and then were extracted with acid and analyzed by inductively coupled plasma optical emission spectroscopy (ICP-OES). The 25 mm filters were analyzed using a single XRF reading, while three readings on different parts of the filter were taken from the 37 mm filters. For lead at the mine concentrate mill, all five samplers gave good correlations (r2 > 0.96) between the two analytical methods over the entire range of found lead mass, which encompassed the permissible exposure limit of 150 mg m(-3) enforced in the USA by the Mine Safety and Health Administration (MSHA). Linear regression on the results from most samplers gave almost 1 ratio 1 correlations without additional correction, indicating an absence of matrix effects from the presence of iron and zinc in the samples. An approximately 10% negative bias was found for the slope of the Button sampler regression, in line with other studies, but it did not significantly affect the accuracy as all XRF results from this sampler were within 20% of the corresponding ICP values. As in previous studies, the best results were obtained with the GSP sampler using the average of three readings, with all XRF results within 20% of the corresponding ICP values and a slope close to 1 (0.99). Greater than 95% of XRF results were within 20% of the corresponding ICP values for the closed-face 37 mm cassette using the OSHA algorithm, and the IOM sampler using a sample area of 3.46 cm2. As in previous studies, considerable material was found on the interior walls of all samplers that possess an internal surface for deposition, at approximately the same proportion for all samplers. At the lead-acid battery recycler all five samplers in their optimal configurations gave good correlations (r2 > 0.92) between the two analytical methods over the entire range of found lead mass, which included the permissible exposure limit enforced in the USA by the Occupational Safety and Health Administration (OSHA). Linear regression on the results from most samplers gave almost 1 ratio 1 correlations (except for the Button sampler), indicating an absence of matrix effects from the presence of the smaller quantities of the other metals in the samples. A negative bias was found for the slope of the button sampler regression, in line with other studies. Even though very high concentrations of lead were encountered (up to almost 6 mg m(-3)) no saturation of the detector was observed. Most samplers performed well, with >90% of XRF results within +/- 25% of the corresponding ICP results for the optimum configurations. The OSHA algorithm for the CFC worked best without including the back-up pad with the filter.