Assessing the contribution of water to the mass closure of PM10

The data obtained during a number of field studies aimed at determining the chemical composition of atmospheric particulate matter (PM) have shown that the measurement of the main PM components (main elements, ions, elemental carbon, organic carbon) was generally sufficient to obtain a reasonable mass closure. Notwithstanding, a wide gap between PM mass concentration and reconstructed mass was observed in two peculiar environmental conditions: desert dust intrusion and severe atmospheric stability episodes characterized by very high ammonium nitrate concentration. In these two cases, the mass closure improved significantly by adding the concentration of PM-bound water. Water was determined by using a coulometric Karl-Fisher system equipped with a controlled heating device; the method was able to separate different water contributions released in different temperature ranges from 50 to 250 °C. In our field studies the amount of water associated to ammonium salts in winter stability conditions was mostly dependent on ammonium nitrate concentration and constituted up to 22% of the total PM10 mass; the specific water contribution linked to ammonium salts (released in the temperature range 180-250 °C) constituted up to 30% of the ammonium nitrate mass. It was confirmed that in these extreme conditions quartz and Teflon filters behave differently: when measured on quartz filters, PM concentration was lower than on Teflon, the mass closure was satisfactory and the concentration of water was presumably very low. In the case of desert dust episodes, water was up to 10% of total PM10 mass; the specific water contribution linked to desert dust (released in the temperature range 100-180 °C) constituted about 5% of the mass of soil components. In other environmental situations, such as urban environments, marine atmosphere and rural areas, the concentration of PM-bound water was below 2-3 μg/m3.