AAAR 35th Annual Conference October 17 - October 21, 2016 Oregon Convention Center Portland, Oregon, USA
Abstract View
Atmospheric Modeling of Cyclic Volatile Methyl Siloxanes
NATHAN JANECHEK, Charles Stanier, University of Iowa
Abstract Number: 354 Working Group: Urban Aerosols
Abstract Cyclic siloxanes are common chemicals in personal care products, especially antiperspirants. These are volatile chemicals that are released into the atmosphere by personal care product use. Since people are the emission source, highest concentrations are found in urban and indoor locations. In the atmosphere, cyclic siloxanes are oxidized by hydroxyl radicals (OH) forming oxidation products that can form aerosol species. The oxidation products have largely been unexplored, and production of nanoparticles could have important environmental and health consequences. In this work, we use the Community Multiscale Air Quality (CMAQ) atmospheric chemistry transport model to simulate gas phase cyclic siloxane behavior over North America. The model has been modified to include the most common cyclic siloxanes (D4, D5, and D6) and their OH oxidation products, emissions and seasonally varying boundary conditions calculated using measurement ratios accounting for differential oxidative aging, and wet and dry deposition for the cyclic siloxane species. This work represents the highest resolution modeling to date, and for the first time expected concentrations and the spatial distribution of the oxidation products, which can likely form particles, are reported. Using the model, seasonal concentration trends are explored by simulating January, April, July, and October months. Modeling results show urban parent cyclic siloxane seasonal concentrations are not sensitive to seasonally varying OH concentrations, and instead are associated with chemical transport. Conversely, rural parent cyclic siloxane concentrations, and both rural and urban oxidation products, show sensitivity to seasonally varying OH concentrations. Model results were evaluated to published spring cyclic siloxane measurements with calculated fractional bias values of -0.41, -0.033, -0.90, and fractional error values of 0.95, 0.66, 0.98 for D4, D5, and D6 respectively. While these error metrics remain high, they improve over coarser resolution models previously published at hemispheric scale.