High-humidity tandem differential mobility analyzer for accurate determination of aerosol hygroscopic growth, microstructure, and activity coefficients over a wide range of relative humidity

Abstract

Abstract. Interactions with water are crucial for the properties, transformation, and climate effects of atmospheric aerosols. Here we present the high-humidity tandem differential hygroscopicity analyzer (HHTDMA) and a new method to measure the hygroscopic growth of aerosol particles with in situ restructuring to minimize the influence of particle shape. With this approach, growth factors can be measured with an uncertainty of 0.3 %–0.9 % over a relative humidity (RH) range of 2 %–99.6 % and with an RH measurement accuracy better than 0.4 %. The HHTDMA instrument can be used in hydration, dehydration, and restructuring modes of operation. The restructuring mode allows us to investigate the effects of drying conditions on the initial microstructure of aerosol particles and specifies the optimal parameters that provide their rearrangements into compact structures with a nearly spherical shape. These optimal parameters were used in hygroscopic growth experiments by combining the restructuring mode with a conventional hydration or dehydration mode. The tandem of two modes allowed us to measure the particle growth factors with high precision as well as to determine the thickness of the water adsorption layer on the surface of compact crystalline particles. To verify the HHTDMA instrument we compared the measured ammonium sulfate growth factors with those obtained from the E-AIM-based Köhler model (E-AIM: Extended Aerosol Inorganics Model). Averaged over the range of 38 %–96 % RH, the mean relative deviations between measurements and model results is less than 0.5 %. We demonstrate this precision by presenting data for glucose, for which bulk thermodynamic coefficients are available. The HHTDMA-derived activity coefficients of water and glucose were obtained for both dilute and supersaturated solutions and are in good agreement with those reported in the literature. The average deviation between the measured activity coefficients and those obtained by the bulk method is less than 4 %. For dilute solution in water with an activity range of 0.98–0.99, the hygroscopicity parameter of glucose and the molal osmotic coefficient were obtained with an uncertainty of 0.4 % and 2.5 %, respectively.