Environmental Engineering Reference
In-Depth Information
than 95% by unburned lubricating oil; Kittelson et al. (2006a) showed that nuclei
mode particles mainly consisted of heavy hydrocarbons with volatility close to that
of C 24 - C 32 n - alkanes (as the fi ngerprint of lubricating oil). Very recently, Rö nkk ö
et al. (2007) demonstrated that the nucleation mode particles had a non-volatile
core formed before the dilution process and semi-volatile hydrocarbons condensed
on it.
The relative contribution of sulfates and hydrocarbons to nucleation of nanopar-
ticle emissions and favourable conditions for particle formation are areas that are
still poorly understood. Maricq (2007) notes that the nucleation of sulfates prefer-
entially occurs in the exhaust of light duty diesel vehicles, while the nucleation of
lubricant oil hydrocarbons preferentially occurs in the exhaust of heavy duty diesel
vehicles. He interprets this difference from the different oil consumption rates and
the presence or absence of an oxidation catalyst that controls the exhaust sulfate
to hydrocarbon ratio.
Another possible mechanism of formation of nucleation mode particles is related
to the addition of a fuel additive, such as cerium or iron. Since the exhaust tem-
perature of a diesel engine is low, the diesel particle fi lter requires catalytic assis-
tance for its regeneration and this could be done either by the incorporation of a
catalyst into the particle fi lter substrate or into the fuel. Recent research (Miller
et al. , 2007) has shown that the addition of iron (ferrocene) in diesel is responsible
for the formation of 5-10 nm NPs via homogeneous nucleation of iron and subse-
quent agglomeration (Miller et al. , 2007). The number and size of these NPs increase
with the level of doping. The addition of cerium oxide NPs to the diesel fuel reduces
signifi cantly the number concentration of particles in the accumulation mode
(mostly above 80 nm) and then the particle mass. However, it leads to a dramatic
increase of nucleation mode particles that are mostly comprised of cerium (Skillas
et al. , 2000 ; Jung et al. , 2005 ).
Uncertainties still remain in number emission and chemical composition of NPs
from diesel vehicles and in understanding the nucleation process under different
ambient conditions. Also, most of the laboratory chassis dynamometer studies have
been based upon small numbers of vehicles and studies have shown that volatile
nanoparticle emissions were highly dependent on the vehicle (Mathis et al. , 2005 ).
A modelling approach of particle exhaust emissions (Vouitsis et al. , 2005 ) demon-
strated that the nucleation mode is signifi cantly suppressed by the presence of the
soot mode due to the high surface of the latter. As vehicle exhaust dilutes and cools
in the atmosphere, semi-volatile species may either nucleate to form new particles
or condense onto existing particles. This suggests that the reduction of the particle
mass using devices such as diesel particle fi lters may lead to an increase of particle
number concentrations if volatile species are not appreciably removed.
Far fewer studies are available on petrol engines. Petrol engines have been
reported to emit a lower total number concentration of particles than diesel engines
(Kittelson, 1998) but higher concentrations of very small particles than diesel
engines (Maricq et al. , 1999a, 1999b ). Ultrafi ne particles emitted by petrol engines
comprise carbonaceous agglomerates ranging from 10 to 80nm, consisting of a
carbon core with various condensed compounds (Morawska and Zhang, 2002). The
peak in the number size distributions were reported to be in the 30- 60 nm size
range (Ristovski et al. , 1998 ; Maricq et al. , 1999b) with particles also containing a
Search WWH ::




Custom Search