Water Condensation and Freezing in the Atmosphere: Exploring Deliquescence and Ice Nucleation

Abstract

Atmospheric aerosols play key roles in numerous atmospheric processes. They affect human health, are substrates and components for atmospheric chemistry, and via their contribution to clouds affect the water cycle and the energy balance of the planet. The role of aerosols in these processes involves large uncertainty and this partly comes from an incomplete understanding of the aerosol phase state in the atmosphere. In fact, as aerosols travel through the atmosphere they may be exposed to temperature and humidity conditions that alter their properties. Although some of the mechanisms involved such as the deliquescence of soluble particles and the nucleation of cloud droplets are reasonably well understood. Others, like hygroscopic growth below deliquescence and heterogeneous ice nucleation remain poorly understood. This thesis aims to fill some of these knowledge gaps. For clarity, the report is thematically separated between Pre-Deliquescence that has contributed to three published papers (Papers I, II and III) and Ice Nucleation. Ice Nucleation involves significant instrumentation development and has contributed to one measurement report (Paper IV) and one technical paper presenting the developed instrument (Paper V). I. Pre-Deliquescence Soluble particles are characterized by their ability to dissolve into liquid water. They take part in unique processes such as their complete solvation into brine droplets below water saturation (deliquescence); and their surface solvation into thin liquid films below deliquescence (pre-deliquescence). The transition from solid soluble particles to liquid droplets is typically captured by Köhler theory which describes a modified equilibrium vapor pressure due to (i) mixing entropy (Raoult’s law) and (ii) droplet geometry (Kelvin effect). However, this description omits the existence of a pre-deliquesced state. Therefore we develop a more complete model that accounts for interfacial forces giving rise to predeliquescence, in a manner akin to surface melting. The validity of the model is tested against previous hygroscopicity measurements of sodium chloride particles and results show that the model is able to reproduce observations using a set of physically realistic parameters. In Papers II and III the surface chemical composition of two common atmospheric salts, sodium acetate and ammonium sulfate, is observed at pre-deliquescence conditions using an ambient pressure X-ray photoelectron spectroscopy method. On the sodium acetate sample, reversible water uptake and salt dissociation is observed as humidity increases. Furthermore, a sodium depletion is observed at the sample surface after completing a deliquescence/efflorescence cycle. This is attributed to the formation of acetic acid and its enrichment at the liquid/gas interface. In the case of ammonium sulfate, species other than the salt ions are detected showing that chemistry takes place within the pre-deliquesced film. S0, HS-, HONO, and NH3(aq) appear simultaneously to salt solvation and a chemical mechanism is suggested to explain the presence of these species. II. Ice Nucleation and PINCii The ice phase represents ~ 65% of the total condensed water of clouds over the planet and is thus a key component in understanding the role of clouds on the radiative budget of the planet and on the hydrological cycle. In most tropospheric conditions (0°C ≥ T ⪆ -37°C), ice crystals result from heterogeneous nucleation processes requiring the presence of specific particles. On average, only 1 in 10^6 atmospheric particles can act as a suitable surface for ice nucleation and identifying these particles is key to understanding atmospheric ice nucleation and the development of cloud models. For this purpose, a large part of my PhD was invested in the development of a new portable ice nucleation chamber - PINCii. The details and evaluation of the PINCii instrument are presented in Paper V showing that PINCii is able to reproduce well defined activation processes such as homogeneous freezing, deliquescence and droplet formation with great accuracy. Results also show that PINCii can perform both ice nucleation and water droplet formation experiments with lower instrumental uncertainties than previous instruments of the same type. The HyICE-2018 field campaign based at the Smear II station in Hyytiälä (Finland) is presented in Paper IV. This campaign gathered various types of ice nucleation instruments in order to quantify the concentrations and identify the sources of ice nucleating particles in the boreal environment. Specific days were selected for ice nucleation instruments to run in parallel and at similar experimental conditions for inter-comparison purposes. Paper IV summarizes the aerosol properties and meteorological conditions during the campaign. It also presents the results of the inter-comparisons showing the amplitude of the deviations between the instruments present on site.