The sensitivity of ice-nucleating minerals to heat and implications for the detection of biogenic ice-nucleating particles

Ice nucleating particles (INP) are a subset of atmospheric aerosols which strongly influence the radiative properties and precipitation onset in mixed phase clouds. Mineral dust and biogenic particles such as bacteria, pollen and fungal spores can act as INP and are present in the atmosphere as internal or external mixtures. However, the sources, abundance and distribution of INP are poorly understood. The current widely accepted method of determining the relative contributions of mineral and biogenic INP is to treat INP samples with heat. This is based on the hypothesis that proteinaceous biogenic INP will be deactivated and mineral INP will be unaffected. However the hypothesis that mineral INP are never deactivated by heating not been tested to date. Mineral surfaces may undergo a range of geochemical reactions when heated in air or water and the potential effects of this on their ice nucleating activity is not known. We therefore subjected a range of atmospherically relevant minerals and atmospheric dust analogues to heat treatments equivalent to those used in the past studies. The samples were heated both as aqueous suspensions (100°C for 30 minutes) and in air as dry powders (250°C for 4 hours) and their ice nucleating activity was tested before and after treatment with a microlitre droplet freezing assay. We found that silica based samples showed a significant response to aqueous heating (reduction of median freezing temperature of a 1% suspension of up to 5.4°C) but little response to dry heating. Similar responses were seen in Arizona Test Dust and calcite. In contrast, K-feldspar samples were largely unaffected by aqueous heating but some showed mild deactivations when dry heated. Notably, K-feldspar was sensitive to longer heat treatments. Overall this survey shows that the assumption that mineral INP are completely inert to heat should be reconsidered in the context of using heat to ‘detect’ biogenic INP. We conclude that while INP heat tests are an effective method for positive detection of biogenic INP in ambient air samples they have the potential to produce false positives. For example, in a silica-rich mineral dust a deactivation may be related to the mineral component. Nevertheless, in samples where the mineral ice nucleating activity is determined by K-feldspar the aqueous heat test provides a valid qualitative test for proteinaceous biological INP.