Analysis of measurement uncertainty for verification/calibration of particulate matter measuring instruments using the gravimetric method

Abstract

As demand for particulate matter (PM) monitoring increases for scientific and regulatory purposes, the need for an accurate calibration process for these instruments will also increase. This article focuses on studying the factors that cause measurement uncertainty during the calibration process with the calibration facility (ManDust) for reference gravimetric instruments. Measurement uncertainty in this article included type A and type B uncertainties. Type A uncertainty mostly included repeatability and reproducibility while volume uncertainty predominately contributed to type B uncertainty. To clearly examine how these factors contributed to the combined uncertainty, the calibration process was carried out at a high concentration range from 0.4 to 1.6 mg m−3. In general, the measurement uncertainty ranged from 1.95% to 6.65%. Further analysis showed that type A uncertainty contributed more to the total uncertainty than type B. In particular, uncertainty from volume calculation and measurement contributed over 80% of type B uncertainty. During the PM filter weighing process, the uncertainty during the conditioning process contributed the most to type B uncertainty. This article also conducted a sensitivity analysis and found that besides volume measurement, the filter weighing procedure was also sensitive to changes even though it only contributed less than 5% of type B uncertainty. Using high-quality devices and careful handling during these two procedures should be emphasized to reduce the uncertainty of the calibration process.