Articles | Volume 7, issue 19
04 Oct 2007
04 Oct 2007

Efficiency of immersion mode ice nucleation on surrogates of mineral dust

C. Marcolli, S. Gedamke, T. Peter, and B. Zobrist

Abstract. A differential scanning calorimeter (DSC) was used to explore heterogeneous ice nucleation of emulsified aqueous suspensions of two Arizona test dust (ATD) samples with particle diameters of nominally 0–3 and 0–7 μm, respectively. Aqueous suspensions with ATD concentrations of 0.01–20 wt% have been investigated. The DSC thermograms exhibit a homogeneous and a heterogeneous freezing peak whose intensity ratios vary with the ATD concentration in the aqueous suspensions. Homogeneous freezing temperatures are in good agreement with recent measurements by other techniques. Depending on ATD concentration, heterogeneous ice nucleation occurred at temperatures as high as 256 K or down to the onset of homogeneous ice nucleation (237 K). For ATD-induced ice formation Classical Nucleation Theory (CNT) offers a suitable framework to parameterize nucleation rates as a function of temperature, experimentally determined ATD size, and emulsion droplet volume distributions. The latter two quantities serve to estimate the total heterogeneous surface area present in a droplet, whereas the suitability of an individual heterogeneous site to trigger nucleation is described by the compatibility function (or contact angle) in CNT. The intensity ratio of homogeneous to heterogeneous freezing peaks is in good agreement with the assumption that the ATD particles are randomly distributed amongst the emulsion droplets. The observed dependence of the heterogeneous freezing temperatures on ATD concentrations cannot be described by assuming a constant contact angle for all ATD particles, but requires the ice nucleation efficiency of ATD particles to be (log)normally distributed amongst the particles. Best quantitative agreement is reached when explicitly assuming that high-compatibility sites are rare and that therefore larger particles have on average more and better active sites than smaller ones. This analysis suggests that a particle has to have a diameter of at least 0.1 μm to exhibit on average one active site.

Final-revised paper