Organic ice nuclei

In two recent experimental studies, the influence of organic ice nuclei (IN) on heterogeneous ice nucleation has been investigated.

In a first study, heterogeneous ice freezing points of aqueous solutions containing various immersed solid dicarboxylic acids (oxalic, adipic, succinic, phthalic and fumaric) have been measured with a differential scanning calorimeter. The results show that only the dihydrate of oxalic acid (OAD) acts as a heterogeneous ice nucleus, with an increase in freezing temperature between 2 and 5 K depending on solution composition. In several field campaigns, oxalic acid enriched particles have been detected in the upper troposphere with single particle aerosol mass spectrometry. Simulations with a microphysical box model indicate that the presence of OAD may reduce the ice particle number density in cirrus clouds by up to ~50% when compared to exclusively homogeneous cirrus formation without OAD. Using the ECHAM4 climate model we estimate the global net radiative effect caused by this heterogeneous freezing to result in a cooling as high as−0.3Wm−2 (pdf of the article).

In a second study, the heterogeneous ice nucleation rate coefficient (jhet) of water droplets coated with a monolayer of 1-nonadecanol (a long chain alcohol, C19H39OH) was determined from multiple freezing/melting cycles. Freezing was monitored optically with a microscope for droplet radii between 31 and 48 μm and with a differential scanning calorimeter for radii between 320 and 1100 μm. The combination of these two techniques allows the surface area of the 1-nonadecanol nucleating agent to be varied by more than a factor of 1000, showing that jhet increases only by ~5 orders of magnitude over a temperature range of 18 K. This is roughly 5 times less than the change in the ice nucleation rate coefficient for homogeneous ice freezing at around 238 K or for heterogeneous ice freezing in the presence of a solid ice nucleus, such as Al2O3. This temperature dependence of jhet can be reconciled with the framework of classical nucleation theory, when assuming a reduced compatibility of the alcohol monolayer with the ice embryo as the temperature decreases. We attribute this finding to an enhanced ability of the alcohol monolayer to adapt to the ice structure close to the ice melting point due to larger thermal density fluctuations in the monolayer, which in turn makes the monolayer serve as a better ice nucleus (article online).