Interlaboratory Comparison of the Centrifugal Particle Mass Analyzer-Electrometer Reference Mass System (CERMS)
RYM MEHRI, Robert T. Nishida, Timothy Sipkens, Jason S. Olfert, Joel Corbin, Gregory Smallwood, National Research Council Canada
Abstract Number: 472
Working Group: Carbonaceous Aerosols
Abstract
CERMS is a combination of instruments capable of producing an aerosol of a known mass concentration, independent of particle properties. It consists of (1) a unipolar diffusion aerosol charger (UDAC) that uses a corona discharge to charge the particles; (2) a centrifugal particle mass analyzer (CPMA) to classify the particles based on their mass-to-charge ratio; and (3) an aerosol electrometer to measure the charge concentration. This system has the proven ability to calibrate nonvolatile particulate matter (nvPM) mass concentration instruments, including those measuring equivalent black carbon, refractory black carbon, or total carbon via thermal-optical analysis (TOA). CERMS shows repeatability and intermediate precision of approximately 1% and 2%, respectively, when tested in laboratory settings, however reproducibility has not been investigated. TOA has been associated with large uncertainty (k=2) for elemental carbon mass concentration (17-20%).
This work seeks to robustly define the uncertainty associated with CERMS measurements and aims to quantify the reproducibility of this system in different settings using different sources. For this purpose, an interlaboratory comparison of CERMS systems from three independent laboratories was conducted in two parts. First, a laboratory comparison was conducted using a miniature inverted soot generator (MISG) for the particle source to provide initial demonstration of the study feasibility. Second, the same procedure was applied to assess CERMS reproducibility during a campaign using a Start Cart and J85 gas-turbine engine as particle sources.
For both components of the interlaboratory comparison, results show high repeatability (k=2) for the electrometer (0.9-1.3%) and CPMA (1.6-2.3%) compared to CERMS repeatability (k=2) (6.3-11.2%). The reproducibility (k=2) of CERMS ranged from 6.3 to 15.4%.