Abstract (eng)
New particle formation by gas-to-particle conversion is observed all over the globe. It is a
major contributor to cloud condensation nuclei concentrations and therefore impacts global
climate, as the lifetime and albedo of clouds are affected by the availability of cloud droplet
seeds. For a newly formed cluster containing only a few molecules, it is crucial that the
initial growth by further vapor addition is fast enough to prevent losses to larger background
aerosol. Recent studies have shown that not only sulfuric acid, but also organic molecules of
biogenic and anthropogenic origin can contribute to initial particle growth. However, it is still
under debate, how the different condensing vapors interact with each other and how external
conditions like temperature influence the role of organics. Moreover, complete mass balance
between observed gas-phase compounds and particle phase growth has yet to be demonstrated
for any organic system.
In this thesis, a new instrument, the DMA-train, for particle size distribution measurements in
the cluster-particle transition regime below 10nm is presented. It uses existing particle sizing
technology in a new parallel approach to overcome the problem of low measurable concentrations
of ultrafine particles and it therefore can investigate the dynamics of initial growth.
Moreover, this allows for the development of a new analysis method, providing size- and
time-resolved insights into the aerosol dynamics. The analysis method is applied to a chamber
experiment, where new particle formation from the ozonolysis products of two biogenic
volatile organic compounds is compared, revealing that particle growth rates below 10nm
show strong variations. A clear curvature effect is identified in two additional studies, using
the CLOUD experiment at CERN. It is shown that some highly oxygenated molecules from
the ozonolysis of the monoterpene alpha-pinene are already able to condense onto the smallest
molecular clusters. However, only with increasing particle size and a reduced Kelvin-barrier,
more and more less oxygenated products can contribute, increasing the particle growth rate.
This is further enhanced in the presence of isoprene, where the oxidation chemistry of alpha-
pinene is altered leading to less products with extremely low volatility, suppressing initial
but not later growth. On the other hand, it is found that organics do play a decisive role
over a wide range of temperatures. The volatility of all oxidation products is strongly reduced
at cold temperatures counterbalancing the slower chemical reaction rates. Moreover,
mass balance between sensitive gas-phase measurements and precision particle growth rates
is demonstrated in an aerosol growth model.
In summary, this thesis shows that the DMA-train can help resolving some long debated
topics: It indicates why particle growth rates in the atmosphere accelerate with increasing
size, why new particle formation is rarely observed in isoprene-rich environments like the
Amazon and that organics are still important for new particle formation at high altitudes,
e.g. in the outflow regions of deep convective clouds.