The IAEA proficiency test on evaluation of methods for 90Sr measurement in a mineral matrix. preliminary evaluation of sources of Bias and measurement uncertainties

In order to assess the effect of various analytical operations and measurement routines on the quality of the 90Sr data, the IAEA Analytical Control Services have organised a proficiency test using a mineral sample spiked at three different levels with known amounts of 90Sr. Each participant received a total of nine samples, three containing approximately 0.01 Bq 90Sr, three containing approximately 0.1 Bq, one containing approximately 5 Bq, together with a blank sample and a 90Sr reference solution. The standard sample size was 50 g, however smaller sample sizes of 10 g or 5 g were made available to analysts whose methods could not handle large samples. Analysts were instructed not to take sub-samples but to analyze the entire sample. Analysts were also provided with a set of handling instructions, an operation checklist, a data sheet for reporting the results and a questionnaire for documenting their radioanalytical procedure. The samples were distributed between December 1998 and end of March 1999. Although the original deadline for the submission of the results was June 30, 1999, it was extended to the end of September 1999 to accommodate the requests of many participants. This proficiency test generated considerable interest from the radioanalytical community as a total of 192 sets of samples were distributed to 158 radio-analytical laboratories world-wide. The major sources of bias leading to overestimated values were: ineffective purification procedures; high background values; and a lack of statistical control over background values. The major sources of bias leading to an underestimation were: neglect to correct for stable Sr in the sample in calculating recovery factors and neglect to correct for quenching in liquid scintillation counting. The participants to the proficiency test currently receive individual feedback on their performance within one month and are provided with suggestions where to modify and improve their radio-analytical procedures in the appropriate cases. (authors)