Measurement report: The Fifth International Workshop on Ice Nucleation Phase 1 (FIN-01): Intercomparison of Single Particle Mass Spectrometers

Abstract. Knowledge of chemical composition and mixing state of aerosols at a single particle level is critical for gaining insights into atmospheric processes. One common tool to make these measurements is single particle mass spectrometry. There remains a need to compare the performance of different single particle mass spectrometers (SPMSs). An intercomparison of SPMSs was conducted at the Aerosol Interaction and Dynamics in the Atmosphere (AIDA) chamber at the Karlsruhe Institute of Technology (KIT) in November 2014, as the first phase of the Fifth International Workshop on Ice Nucleation (FIN-01). In this paper we compare size distributions and mass spectra of atmospherically relevant particle types measured by five SPMSs. These include different minerals, desert and soil dusts, soot, bioaerosol (Snomax; protein granule), secondary organic aerosol (SOA) and SOA coated mineral particles. All SPMSs reported similar vacuum aerodynamic diameter (dva) within typical instrumental ranges from ~100‒200 nm (lower limit) to ~2‒3 μm (upper limit). In general, all SPMSs exhibited a wide dynamic range (up to ~103) and high signal to noise ratio (up to ~104) in mass spectra. Common spectral features with small diversities in mass spectra were found with high average Pearson’s correlation coefficients, i.e., for average positive spectra ravg-pos = 0.74 ± 0.12 and average negative spectra ravg-neg = 0.67 ± 0.22. The highest correlation of average mass spectra across all instruments was observed for Snomax (ravg-pos = 0.92 ± 0.04 and ravg-neg = 0.90 ± 0.05), attributed to the prevalence of common markers and similar spectral patterns. The poorest correlation was found for propane soot (ravg-pos = 0.51 ± 0.23 and ravg-neg = 0.35 ± 0.26), primarily because of low detection efficiency (DE) due to small particle size. We found that instrument-specific DE was more dependent on particle size than particle type. We also found that particle identification favored the use of bipolar, rather than monopolar, instruments. Particle classification from “blind experiments” showed that all instruments differentiated SOA, soot, and soil dust, and detected subtle changes in the particle internal mixing, but had difficulties differentiating among specific mineral types and dusts. This study helps to further understand the capabilities and limitations of the single particle mass spectrometry technique in general, as well as the specific instrument performance in characterizing atmospheric aerosol particles. We propose that intercomparison workshops should continue as new SPMSs are developed and, ideally, should include both laboratory and field activities.