Journal Article
Atmospheric Chemistry and Physics, vol. 18, iss. 16, pp. 12595-12612, 2018
Authors
Amy L. Bondy, Daniel Bonanno, Ryan C. Moffet, Bingbing Wang, Alexander Laskin, Andrew P. Ault
Abstract
Abstract. Aerosols in the atmosphere are chemically complex with thousands of chemical
species distributed in different proportions across individual particles in
an aerosol population. An internal mixing assumption, with species present in
the same proportions across all aerosols, is used in many models and
calculations of secondary organic aerosol (SOA) formation, cloud activation,
and aerosol optical properties. However, many of these effects depend on the
distribution of species within individual particles, and important
information can be lost when internal mixtures are assumed. Herein, we show
that – as found during the Southern Oxidant and Aerosol Study (SOAS) in
Centreville, Alabama, at a rural, forested location – aerosols frequently
are not purely internally mixed, even in the accumulation mode
(0.2–1.0 µm). A range of aerosol sources and the mixing state were
determined using computer-controlled scanning electron microscopy with
energy-dispersive X-ray spectroscopy (CCSEM-EDX) and scanning transmission
X-ray microscopy–near-edge X-ray absorption fine structure spectroscopy
(STXM-NEXAFS). Particles that were dominated by SOA and inorganic salts
(e.g., ammonium sulfate) were the majority of particles by number fraction
from 0.2 to 5 µm with an average of 78 % SOA in the
accumulation mode. However, during certain periods contributions by sea spray
aerosol (SSA) and mineral dust were significant to accumulation (22 % SSA
and 26 % dust) and coarse-mode number concentrations (38 % SSA and
63 % dust). The fraction of particles containing key elements (Na, Mg, K,
Ca, and Fe) were determined as a function of size for specific classes of
particles. Within internally mixed SOA/sulfate particles < 5 %
contained Na, Mg, K, Ca, or Fe, though these nonvolatile cations were present
in particles from the other sources (e.g., SSA and dust). Mass estimates of
the aerosol elemental components were used to determine the extent of
internal versus external mixing by calculating the mixing state index (χ). The aerosol population was more externally mixed than internally mixed
during all time periods analyzed. Accumulation mode aerosol ranged from more
internally mixed during SOA periods to mostly externally mixed during dust
periods. Supermicron aerosols were most externally mixed during SOA time
periods, when more SOA particles added a distinct supermicron class, and more
internally mixed when dominated by a single particle type (e.g., SSA or
dust). These results emphasize that neither external nor internal mixtures
fully represent the mixing state of atmospheric aerosols, even in a rural,
forested environment, which has important implications for air quality and
climate modeling.
