ELEMENTS by Cellulosic Internal Capsule Sampler: METHOD 7306, Issue 1, dated 10 September 2015 - Page 4 of 11
Zefon International, Ocala, FL). Solu-sert™ capsules were dosed with 33 elements at three different spiking levels (spike levels certified by High-Purity Standards, North Charleston, SC), listed in Table 3. Sets of the spiked samples were conveyed to participating volunteer laboratories and analyzed by ICP-AES after sample dissolution. A variety of sample preparation procedures were used by the labs including NIOSH 7300, 7301, 7302, 7303, hot plate digestion utilizing nitric acid, sulfuric acid and hydrogen peroxide (modified NIOSH 7300), and microwave assisted digestion using nitric acid and hydrogen peroxide (modified NIOSH 7302). Results were received from 9 laboratories; however, not every spiked element was reported by each laboratory. Individual sample results may be found in the backup data report [17]. Statistical calculations were performed using SAS Software (version 9.2, SAS Institute Inc., Cary, NC). For each data subset, Grubbs’ test at 1% confidence level was used to identify outliers, which if identified, were removed prior to further statistical calculations. No statistically significant differences were found between the laboratory results for the spiked samples; therefore, all reported data (regardless of the sample preparation method) are included in Tables 3 & 4. Recoveries were quantitative (as defined in Kennedy, et al. [18]) and interlaboratory variability was <0.20 for most elements and loading levels (Table 3). Mean overall recoveries below 90% were found only for Cr, K, and W at low loadings and for Ag at medium and high spike levels and for In at the high spike level. RSD values > 0.20 were found only for Sn at the low spike level, Ag at medium and high spike levels, and In and K at all loadings. Results for precision, bias and accuracy are summarized in Table 4. With the domed top, the internal capsule can be difficult to fit into standard sample digestion vessels. One way to achieve this is to place the samplers into the vessels by bending them slightly inward (using coated forceps) and to push them into the bottom of the vessel to ensure they are covered with the digestion acids. Care must be taken to ensure that sample is not lost in this process. Additional guidance is available from the internal capsule manufacturer [19]. While no statistically significant differences were found based upon the sample preparation, it is important to note that interlaboratory variation is included in those calculations. Some differences in the sample preparation methods may have been statistically significant without that variability. Of particular importance are the less than quantitative recoveries for Sb, Sn, and Ti using NIOSH 7300. This sample preparation (as written) may not be amenable to the analysis of those elements. While this method lists several options for sample preparation, it is imperative that the suitability of the particular sample preparation method for the analytes of interest be considered. Appreciable (>0.5 µg) media background levels were reported by Certified Reference Material (CRM) provider and/or the participating laboratories for several elements, notably Al, Ca, Cr, Fe, In, K, Mg, P, Sb, Se, and Tl. Trace media background levels of a few other elements, i.e., Ba, Cu and Zn, were also obtained. This background was effectively corrected for, as evidenced by the quantitative recoveries obtained for the vast majority of the elements and loading levels. Where the background levels may pose a greater influence is in the calculation of the method LOD. Method LODs (calculated using the standard deviation of blank responses) for internal capsule samples were greater than those calculated for MCE filters alone for many of the elements [17]. Care should be taken in choosing the appropriate media in concert with the expected sample concentrations. REFERENCES: [1] ACGIH [2014]. TLVs and BEIs based on the documentation of the Threshold Limit Values for chemical substances and physical agents and biological exposure indices. Cincinnati, Ohio: American Conference of Governmental Industrial Hygienists [www.acgih.org]. Date accessed: May 2015. [2] Institut fur Arbeitsschutz der Deutschen Gesetzlichen Unfallversicherung [2013]. GESTIS database on hazardous substances (German Social Accident Insurance). Sankt Augustin, FRG: [1]. Date accessed: May 2015.
NIOSH Manual of Analytical Methods (NMAM), Fifth Edition
