A Laboratory Investigation of the Ice Nucleation Ability of Mineral and Soil Dust

Ice nucleation is an important pathway for cloud formation and initiation of precipitation in the atmosphere, thus affecting the Earth’s hydrological cycle and energy and radiative balance (Mülmenstädt et al., 2015). Mineral dust has long been known to efficiently act as an ice nucleating particle (INP), primarily so due to its large size and morphology, and it is commonly associated with the formation of cirrus clouds. Its importance as an INP is increased by its ability to be subject to long-range transport after emission. While the ice nucleating ability of mineral dust increases with its size, there is also a strong dependence on its mineralogical composition, indicating a large and uncertain variability of the ice nucleation activity (INA) of mineral dust from various global sources.

Three samples of dust are being investigated with respect to their ice nucleation efficiency. The three dust samples include Iceland glaciogenic silt collected from the glacial river sediments in southern Iceland, China dust collected at a remote location 100 km north-west of Hohhot in China, and Himalaya dust collected in eastern Himalayas at the top of a glacier at an approximate altitude of 5 km. All three dust samples were collected at the surface and, therefore, can be subject to long-range transport.

Submicron size selected (monodisperse) dust samples of 100, 200 and 400 nm in diameter are analysed in the Portable Ice Nucleation Chamber (PINC), an instrument for ice nucleation measurements based on the principle of the Continuous Flow Diffusion Chamber (CFDC) (Rogers, 1988). The measurements of INA are conducted in water sub- and super-saturated regimes to represent deposition nucleation and condensation freezing modes, respectively. The temperature range investigated is 233–248K. In order to investigate the effect of various dust constituents on its INA, three dust treatments were carried out in addition to the untreated samples. These include heating to 573K for 2 hours and treating the samples with H₂O₂ aiming to denature biological proteinaceous matter and digest organic matter, respectively, as well as dissolving the samples in deionised water, which is expected to remove the soluble material.

The results of the experiments with untreated dust samples indicate no significant INA for 100 nm particles at all temperatures and at 248K for all particle sizes. It is found that the INA of dust samples increases significantly between particle diameters of 100 and 200 nm; however, no significant further increase in INA is observed between particles of 200 and 400 nm in diameter. Of the three dust samples examined, China dust exhibits the highest INA for all dust sizes and temperatures, followed by Himalayan and the Icelandic dusts. Reasons for these differences are currently under investigation using the results of scanning electron microscopy (SEM), elemental and mineralogical composition analyses.

Preliminary results of the measurements of the heated samples of China dust indicate that heating significantly reduces the INA of 200 and 400 nm particles at 238K with the difference in INA increasing at higher RH_w(Fig. 1, left). At a lower temperature of 233K such difference disappears, i.e. heating has no effect on the INA of 200 and 400 nm China dust particles at this temperature (Fig. 1, right).

Lastly, ice nucleation active surface (INAS) densities will be presented to compare efficiencies of the dust samples to previous INP parameterisations from laboratory and field studies. The results of this work are expected to shed more light on the effect of mineralogy, organic content and soluble material of mineral and soil dust on its ability to heterogeneously nucleate ice.

References

Mülmenstädt, J., Sourdeval, O., Delanoë, J, and Quaas, J.: Frequency of occurrence of rain from liquid-, mixed-, and ice-phase clouds derived from A-Train satellite retrievals, Geophys. Res. Lett., 42, 6502−6509, 2015.

Rogers, D.: Development of a continuous flow thermal gradient diffusion chamber for ice nucleation studies, Atmos. Res., 22, 149−181, 1988.

Figure 1. Activated fractions of untreated and heated China dust samples of two sizes, shown for two temperatures. Data points are averages of three scans binned into 1% RH_w intervals and error bars are standard deviations. Vertical line is the water saturation line.