Calibration Uncertainty of 23nm Engine Exhaust Condensation Particle Counters with Soot Generators: a European Automotive Laboratory Comparison

Abstract

Calibration of condensation particle counters (CPC) to measure non-volatile particle number (PN) from vehicle emissions is a significant source of uncertainty of the regulated particle number measurements. In this work, the calibration uncertainty of automotive and calibration laboratories was determined in a first-of-its-kind comparison. For this purpose, the counting efficiency of a reference CPC for automotive exhaust emission measurements was determined at seven participants across Europe with ten soot aerosol generators. Calibration uncertainty was found to be very different in the CPC’s cut-off regime (around the D50 of 23nm) with a coefficient of variation (CoV) of 11% and the plateau regime (from the D90 of 41nm upwards) with a CoV of 4.5%. The uncertainty was higher for a group of soot generators with poorly optimized operating points with a CoV of 31% at 23nm and 5.8% at ≥41nm. Specific influence factors on the calibration uncertainty (measured as the inter-lab variability) could be identified. The calibration of the laboratories’ reference counters accounted for most of the variability in the plateau regime, while 20% of the variability was attributed to the sample flow measurement. Differences between soot generators were the main cause of variability in the cut-off regime due to the increased material sensitivity of the CPC at this particle size but had only secondary relevance in the plateau regime. The calibration uncertainty found in this inter-laboratory exercise should be a guideline for users and legislators, as it provides a typical value for the expected measurement uncertainty of a CPC for automotive exhaust PN.