Environmental Engineering Reference
In-Depth Information
et al. , 2002a, 2002c), or emissions of terpenes in the boreal forests (Kavouras and
Stephanou, 2002) and may also include reaction products of ozone and anthropo-
genic terpenes released from fresheners, cleaners and fragrances in the indoor
environment (Weschler and Shields, 1999; Rohr et al. , 2003 ).
Very recently, Kulmala et al. (2007) showed that at a remote site in Finland,
neutral nucleation dominates over ion-induced mechanisms. Also recently, studies
of nucleation in the urban atmosphere of Atlanta, USA, came to the conclusion
that newly-formed particles (
10 nm) consist of ammonium and sulfate alone
(McMurry and Eisele, 2005; Smith et al. , 2005 ; Sakurai et al. , 2005 ; Jung et al. , 2006 ).
All these recent fi ndings demonstrated the frequency of nucleation mechanisms
involving sulfuric acid in various atmospheres.
The formation of new particles needs the existence of thermodynamically stable
clusters (Kulmala et al. , 2000) that are mostly neutral clusters of about 2 nm in
diameter (Kulmala et al. , 2007). The work of Kulmala et al. (2007) showed that these
clusters are continuously scavenged by coagulation and their ubiquitous presence
in the atmosphere implies continuous nucleation. Their detection with traditional
aerosol sizing instruments requires their growth to above 3 nm.
Large pre-existing particle surface area decreases nucleation rates because it
favours the coagulation scavenging of small nuclei and it depletes the non-volatile
vapour necessary for the growth of small nuclei. As a consequence, new particle
formation is rather infrequent in urban areas (Alam et al. , 2003 ). However, because
the concentrations of non-volatile vapours are proportional to the concentration
ratio between their precursor gases and pre-existing aerosol particles, nucleation
can take place in both clean and polluted environments (Kulmala et al. , 2004 ). This
could be illustrated by observations at two rural British sites with different condi-
tions of both gas phase precursor concentrations and particle surface area. New
particle formation occurred in similar westerly maritime air mass situations at
Weybourne, a rural site located on the east coast of England (Harrison et al. , 2000a ),
and at Harwell, a rural site located in southern England (Charron et al. , 2007 ).
Westerly maritime air masses cross polluted areas of England before reaching
Weybourne, while they cross uninhabited areas of Cornwall and Wales before
reaching Harwell. As a consequence, at Weybourne nucleation occurred in air
masses with large particle surface area and large concentrations of precursors such
as sulfur dioxide, while at Harwell nucleation occurred in clean air masses.
Kulmala et al. (2004) reviewed more than 100 investigations and concluded that
formation rates are often in the range 0.01- 10 cm − 3 s − 1 in the boundary layer and
that they could exceed 100 cm
<
3 s − 1 in urban areas, in industrial plumes or in coastal
areas where high concentrations of condensable species are available. They also
concluded that growth rates are generally in the range 1- 20 nm h − 1 except in coastal
areas and polar areas where growth rates could be signifi cantly higher for the
former or lower for the latter.
Particle growth occurs through condensation of supersaturated vapours on the
surface of small nuclei. This requires a lower degree of supersaturation than nucle-
ation, and condensation of semi-volatile compounds may reduce the rate of particle
formation if both processes involve the same species. For this reason, Holmes
Search WWH ::




Custom Search