Modeling of Carbonyl/Ammonium Sulfate Aqueous Brown Carbon Chemistry via UV/Vis Spectral Deconvolution

Proper characterization of atmospheric brown carbon, its formation, and its light absorbance effects are critical to understanding the aggregate effect that these materials have on overall aerosol climate forcing. In aqueous systems, the expansive number of reaction products that can arise from dicarbonyl/ammonium chemistry can limit the practicality of identifying every specific chromophoric species that comprises an aerosol's overall absorbance spectrum as a reaction system ages. Here, we propose an alternative method of representing UV-visible absorbance spectra as a composite of multiple, dynamically-shifting spectral line shapes. Multiple numbers of curves, and dynamic elements with varying complexity, are compared. This modeling approach is applied to several single-carbonyl/ammonium reaction systems (methylglyoxal, glyoxal, glycolaldehyde) and compared to complex mixtures containing multiple organic compounds in a single reaction system. Kinetic information is inferred from these reaction systems based on fitted time-dependent model coefficients; while estimating kinetics using peak magnitudes in the 282 nm region yields rate constants consistent with previous literature, an overlap in modeled curves in the 208 and 220 nm region, containing two distinct rate constants, is observed. This overlap may result in mischaracterization of reaction rate constants if using 208 nm absorbance measurements as a basis for tracking kinetics in dicarbonyl/ammonium systems, due to the competing effects of two reactions contributing to a single reading.