Abstract
The nitrate radical (NO(3)) oxidation of isoprene is an important contributor to secondary organic aerosol (SOA). Isoprene has two double bonds which allow for multigeneration oxidation to occur. The effects of multigeneration chemistry on the gas- and particle-phase product distributions of the isoprene + NO(3) system are not fully understood. In this study, we conduct chamber experiments by varying the ratio of N(2)O(5) (precursor of NO(3)) to isoprene concentration from 1:1 to 14:1 to investigate the formation of products in both phases under different oxidation levels. Multigeneration chemistry leads to the formation of gas-phase products which then partition into particle phase depending on the product volatility; first-generation products (15-36% of total SOA) such as C(5)H(9)NO(5) and C(10)H(16)N(2)O(9) have volatility (log(10)C* = 1.0-3.0 using the partitioning method and log(10)C* = 2.6-4.5 using the formula method) 1-5 orders of magnitude higher than second-generation products (37-57% of total SOA, log(10)C* = -0.8-2.1 using the partitioning method and log(10)C* = -3.7-1.8 using the formula method) such as C(5)H(8,10)N(2)O(8), C(5)H(9)N(3)O(10), and C(10)H(17)N(3)O(13). The fast reaction rate constants of first-generation products (estimated to be on the order of 10(-13) cm(3) molecules(-1) s(-1) at 295 K) and the lower volatility of second-generation products result in increased SOA yields when NO(3) availability increases and multigeneration chemistry is enhanced. Specifically, an increase of up to 300% in SOA yield is observed when the N(2)O(5)/isoprene ratio increases from 1:1 to 3:1; from 5.7% (organic aerosol mass concentration, ΔM (o) = 4.2 μg/m(3)) to 16.3% (ΔM (o) = 11.9 μg/m(3)) when the reacted isoprene concentration is 25 ppb and from 3.1% (ΔM (o) = 1.2 μg/m(3)) to 12.4% (ΔM (o) = 5.4 μg/m(3)) when the reacted isoprene concentration is 15 ppb. The maximum SOA yield occurs when the N(2)O(5)/isoprene ratio is greater than or equal to 3:1 as a combined result of multigeneration chemistry and peroxy radicals (RO(2)) fate. We encourage future studies to consider both factors, which can vary under different laboratory and ambient conditions, when comparing SOA yields to better understand any differences observed. Our results highlight that multigeneration chemistry and the updated parameters including reaction rate constants and volatility distribution of products should be considered to enable a more comprehensive representation and prediction of SOA formation from NO(3) oxidation of isoprene in atmospheric models.