Discussion

Most particles in the atmosphere take up water when exposed to increasing relative humidity (RH), and this behavior is described and quantized by a hygroscopicity parameter, kappa. Hygroscopic Tandem Differential Mobility Analysis (HTDMA) can be used to measure the physicochemical properties of aerosol particles and study phenomena that lead to size changes in submicron aerosol particles. Specifically, HTDMA measures how aerosol particles of different initial dry sizes grow or shrink when exposed to changing RH. HTDMA uses two differential mobility analyzers (DMAs) and a humidification system to make these measurements. One DMA selects a narrow size range of dry aerosol particles which are exposed to varying RH conditions in the humidification system. The second humidified DMA scans the particle size distribution after the particles have been conditioned by the humidification system. Scanning a wide range of particle sizes enables the second DMA to measure changes in size or growth factor (growth factor = humidified size/dry size) due to water uptake by the particles. A condensation particle counter (CPC) downstream of the second DMA counts particles as a function of selected sizes to obtain the number size distribution of particles exposed to different RH conditions. A new HTDMA system was designed for a set of RH conditions with user-friendly data-analyzing software. Two DMAs were connected, with one humidification system in the second DMA sheath flow, and one humidification system in the particle flow line before the aerosol entered the first DMA. The outflow from the second DMA was running into a CPC, and the data collected from CPC was input into software for particle size and distribution analysis. Different humidifiers and dry/wet air mixed flows were experimented with to reach specified RHs. New HTDMA software was developed to analyze the data on hygroscopicity. The software uses supplied parameters to determine the theoretical distribution of dry particles while performing a least squares fit using a curve that is designed to fit multiple Gaussian distributions on a log-normal scale. Our approach accurately captures multiple peaks even in the presence of "shoulders", i.e. when two peaks are close in diameter. A user interface was designed using the Qt library in Python, allowing the user to explore the fitted data. The designed HTDMA system was then calibrated by pure ammonium sulfate, a well-characterized salt that can be used to calibrate these systems. The hygroscopic behavior of aerosol mixtures was determined.