Intercomparison of number concentration measurements by various aerosol particle counters

Abstract

Total aerosol particle number concentrations, as measured by means of 16 different measurement systems, have been quantitatively compared during an international workshop at the Institute for Experimental Physics of the University of Vienna, Austria, which was coordinated within the Committee on Nucleation and Atmospheric Aerosols (ICCP-IUGG). The range of measuring instruments includes Pollak counters (PCO) in use already for several decades, presently available commercial particle counters, as well as laboratory prototypes. The operation of the instruments considered was based on different measurement principles: (1) adiabatic expansion condensation particle counter, (2) flow diffusion condensation particle counter, (3) turbulent mixing condensation particle counter, (4) laser optical particle counter, and (5) electrostatic particle measurement system. Well-defined test aerosols with various chemical compositions were considered: DEHS, sodium chloride, silver, hydrocarbons, and tungsten oxide. The test aerosols were nearly monodispersed with mean particle diameters between 4 and 520 nm, the particle number concentrations were varied over a range from about 4×10 1 to 7×10 6 cm −3. A few measurements were performed with two-component aerosol mixtures. For simultaneous concentration measurements, the various instruments considered were operated under steady state conditions in a linear flow system. A series of at least 10 single concentration measurements was performed by each individual instrument at each set of test aerosol parameters. The average of the concentration data measured by the various instruments was defined as a common reference. The number concentrations obtained from the various instruments typically agreed within a factor of about two over the entire concentration range considered. The agreement of the measured concentrations is notable considering the various different measurement principles applied in this study, and particularly in view of the broad range of measurement instruments used. Significant deviations and nonlinear response were observed only in a few cases and are possibly related to calibration errors. For certain conditions, a dependence of aerosol counter response on particle composition has been found. The scatter of the number concentrations obtained from each individual instrument during measurements with constant test aerosol typically did not exceed 20% to 25%. At concentrations below 10 3 cm −3, however, several of the instruments, including electrostatic particle measurement systems, tend to show increased experimental scatter.