A unique approach to determine the ice nucleating potential of soot-containing aerosol from biomass combustion

Ice nucleating particles (INP) play a crucial role in cloud and precipitation development in mixed phase clouds by catalyzing ice formation at temperatures warmer than -36 C. Despite their importance, however, there is still considerable uncertainty as to the sources and chemical nature of INP. Water insoluble particles such as mineral dust and certain biological aerosols have been shown to be efficient ice nuclei, and soot particles have also been suggested as potential INP. Biomass burning, such as wildfires and prescribed burning, is a large contributor to atmospheric soot concentrations, and could therefore be a potentially important source of INP. Both laboratory and field studies have detected enhanced INP concentrations in smoke plumes; however, the chemical composition of these INP is still uncertain as fires emit and loft a complex mixture of aerosol particles. In this work we employ a novel approach to selectively remove soot aerosol from the sample stream to determine the specific contribution of soot to INP concentrations. A number of commonly consumed biomass fuels were burned in the U.S. Forest Service combustion laboratory during the FLAME-4 (Fire Laboratory At Missoula Experiment - 4) study. Number concentrations of INP acting in the condensation and immersion freezing modes and total aerosol greater than 500 nm in diameter (N500) were measured using the Colorado State University Continuous Flow Diffusion Chamber (CFDC). To determine the contribution of soot to INP concentrations, the sample stream was passed through a Single Particle Soot Photometer (SP2; Droplet Measurement Technologies) which employs laser induced incandescence (LII) to detect soot containing particles and total soot mass. During LII, soot containing particles are vaporized and removed from the sample while non-soot containing particles pass through the instrument unaffected. By sampling the exhaust of the SP2 with the CFDC and alternately cycling laser power on and off we were able to estimate the contribution of soot to total INP. Reductions in both N500 and INP were observed when the laser power was on, indicating both the presence of soot in the total aerosol and the INP fraction of these particles. However, considerable variability was observed in the fraction of INP composed of soot-containing particles with a range from ~0 - 70% for the biomass types and combustion conditions examined.