Ice nucleation ability of aged aerosol particles under cirrus conditions

Cold, high-altitude cirrus clouds, which consist entirely of ice crystals, play an important role
in the Earth’s radiation budget. They form either through homogeneous freezing of aqueous
solution droplets, or through heterogeneous ice nucleation on a small subset of atmospheric
aerosol particles known as ice-nucleating particles (INPs). INPs lower the energy barrier
for ice formation, enabling ice nucleation at lower supersaturations and warmer sub-zero
temperatures. Freshly emitted or formed INPs undergo substantial physical, morphological,
and chemical aging through a variety of processes, including internal mixing through vapor
condensation, surface and structural changes during ice cloud processing, and chemical
modification through reactive uptake of precursor gases. These aging pathways have the
potential to profoundly alter the ice nucleation ability of INPs, and there is still ample scope
for in-depth experimental studies of these processes and their representation in atmospheric
models for cirrus cloud formation.
This thesis took an experimental approach to investigating the influence of three key aging
pathways on the ice nucleation ability of atmospherically relevant INPs under cirrus conditions.
... mehr
The experiments were carried out at two large-scale cloud chamber facilities: the Aerosol
Interaction and Dynamics in the Atmosphere (AIDA) facility at the Karlsruhe Institute of
Technology (KIT), and the Cosmics Leaving Outdoor Droplets (CLOUD) facility at the
European Organization for Nuclear Research (CERN). The two cloud chambers were used
to simulate atmospheric aging at low temperatures over long timescales, offering extensive
instrumentation to resolve the chemical and structural changes of the INPs during aging.
Meanwhile, the ice nucleation ability of the particles was continuously measured using
Continuous Flow Diffusion Chambers (CFDCs).
The first aging mechanism investigated was internal mixing by condensation of organic and
inorganic precursor vapors onto primary aerosol particles, particularly K-feldspar particles as
a proxy for mineral dust. It was found that the intrinsic heterogeneous ice nucleation ability of
the K-feldspar particles was completely masked by a large number of coating types, including
aqueous sulfuric acid, aqueous nitric acid, and viscous secondary organic material (SOM)
from the oxidation of isoprene. A novel aspect of this thesis was the neutralization of acidic
nitrate and sulfate coating layers by ammonia. In the case of ammonium nitrate, this led to
the reappearance of a heterogeneous ice nucleation mode because the K-feldspar particles
triggered the heterogeneous crystallization of the coating layer, which by itself then acted as the ice nucleating entity. However, no heterogeneous ice nucleation was observed in the case
of ammonium sulfate, presumably due to the absence of heterogeneous crystallization of the
coating layer.
The second mechanism studied was the morphological aging of secondary organic aerosol
(SOA) particles from isoprene oxidation during ice cloud processing. Through a complex
sequence of processes, including liquefaction of the organic material, homogeneous freezing,
re-vitrification of the organic material within the ice lattice, and sublimation of ice, the initially
compact SOA particles transformed to irregularly shaped particles with internal cavities
and pores. Although having a much larger surface area and potential sites to promote ice
formation via the pore condensation and freezing (PCF) mechanism, these ice cloud processed
SOA particles were found to be only marginally better INPs than their pristine, unprocessed
counterparts. This finding corroborates previous experiments conducted with SOA particles
from 𝛼-pinene oxidation and suggests that, irrespective of their history, pure SOA particles are
generally poor INPs under cirrus conditions. However, ice cloud processing may be relevant
for SOM-coated INPs, as it can disrupt an initially uniform coating layer, thereby re-exposing
active sites that were previously shielded.
The third set of experiments investigated chemical aging mechanisms by exposing two
types of seed particles, i.e., sulfuric acid solution droplets and ammonium sulfate crystals,
to either glyoxal or isoprene, both under dark and UV-illuminated conditions. In the case
of the chemically processed sulfuric acid solution droplets, mass spectrometry revealed the
presence of glyoxal-related particle-phase fragments or isoprene-derived oxidized organic
material, but these did not alter the homogeneous freezing mode of the seed particles. In
the case of crystalline ammonium sulfate seed particles with an intrinsic heterogeneous ice
nucleation ability, chemical aging produced different results depending on the type of organic
precursor. A coating of isoprene-derived SOM deactivated the seed particles. Reactive
glyoxal uptake led to the formation of organonitrogen and imidazole-related compounds.
These altered the surface or near-surface properties of the ammonium sulfate crystals during
early-stage carbonyl-ammonium chemistry, thereby transiently enhancing the seed particles’
ice nucleation ability.
Overall, the results demonstrate that the ice nucleation ability of pristine INPs does not
change in a consistent manner as a consequence of aging. The ice nucleation activity of
aged aerosol particles is influenced by coating composition, phase state, cloud-processing
history, and chemical reactivity, which can cause it to increase or decrease compared to the
unprocessed, pristine particles. Incorporating these complex effects of aerosol aging into
atmospheric models is challenging but ultimately necessary to improve predictions of cirrus
cloud formation.