Atmospheric aerosol phase transitions: measurements and implications

Many properties of atmospheric aerosols, such as optical and chemical behavior, are strictly affected by their ability to adsorb moisture from the atmosphere, i.e. hygroscopicity, and to change their physical phase. Indeed, aerosols deliquesce when a solid-to-liquid phase transition occurs due to an increase in relative humidity (RH) conditions; the opposite process, i.e. crystallization, implies a decrease in RH. Whereas aerosols hygroscopicity is widely studied, the lack of a well-standardized method which allows to characterized the overall behavior of PMx fraction, leaded most of the authors to investigate the phase transitions of simplified aerosol systems, such as pure or simple mixture of compounds, and not real atmospheric aerosols which are extremely complex in chemical composition. Thus, the aim of this work is to describe a new conductance method which allows the identification of the phase transition of atmospheric aerosols sampled as PM2.5 fraction in a urban site (Milan, Po Valley, Italy). In addition, seasonal variability of deliquescence and crystallization relative humidity (DRH and CRH, respectively) is discussed in connection with the variability of chemical composition. The results highlights that atmospheric aerosols show a hysteresis when RH conditions are increased up to DRH and then decreased. In particular, deliquescence occurs at higher RH values in respect of CRH, since crystallization is kinetically-dependent and it occurs at supersaturated conditions. The averaged hysteresis amplitude is found to account about 14% RH and nitrates and sulfates content is found to mainly drive the deliquescence and crystallization conditions. The obtained data are then used to estimate the atmospheric aerosols physical state estimated on the basis of the RH conditions and history in Milan for the 2006-2014 period. Thus, the results are discussed to different applications. Firstly, the implications for energy saving of data centers are shown. Since this centers represent about 2% of the global energy demand, the importance in reducing this amount was stated. This can be allowed by improve the energy efficiency of data centers, mainly by means of decreasing the energy consumption of the cooling systems. The Direct-Free Cooling systems, which is one of the most used answer, forced filtered ambient airflow to remove heat produced by circuits activity. On the other hand, atmospheric contaminants are deposited on circuits surface and corrosion can occurred when they promote an electrolyte solutions formation due to hygroscopicity and deliquescence. Thus, a modelling and experimental approaches are compared to estimate the energy saving of ENI-Green Data Center (Sannazzaro de’ Burgondi, Po Valley, Italy). The effects of atmospheric contaminants are then discussed on their corrosive effects on electronic circuits. The results show that deliquesced aerosols promote corrosion and electrochemical migration far from condensing conditions and this can induce circuits failures. This states the importance in considering aerosols contamination and their phases for circuits reliability. Finally, the aerosols hygroscopic growth factors are evaluated within the hysteresis loop with the aim to provide an estimation of the error which affects the widely-used algorithms for PM2.5 concentrations from remote sensing data retrieval.