274
Ecotoxicological Issues
1
INTRODUCTION
Polycyclic aromatic hydrocarbons (PAR) comprise a highly diverse class of
widespread environmental pollutants ([1], [2], [3]) They are formed as trace byproducts of combustion and pyrolysis of organic materials including fossil fuels
(gas, oil, coal), wood, plastics, food, municipal waste, etc. Figure 1 depicts
molecular structures for a small number of the PAR with particular attention
devoted to the structural diversity of this class.
A number of these compounds (e.g. benzo[a]pyrene, 18) are strong carcinogens,
while other closely-related species (e.g. benzo[e]pyrene, 19) are non-carcinogens.
Source apportionment for receptor modelling is a traditional discipline of
environmental research. Following analysis of the complex mixture of PAR
(» 100 compounds) in air particulate collected at a given location, its task is to
apportion the contributions arising from diesel-powered vehicles, automobiles,
space heating (gas, oil, wood), smoking, etc.
Obviously, the most straightforward scenario would be one in which each source
emitted a unique "marker" PAR at a constant (ng/BTU) rate or a consistent PAR
"signature" and that these marker PAR would have similar physical properties (e.g.
nonvolatile and thus totally associated with particulate matter) and would be
chemically unreactive under ambient environmental conditions.
Unfortunately, these conditions are never met: PAR mixtures are formed from
different sources with substantial overlaps in composition and many of the
individual PAR are highly reactive under ambient conditions. Indeed, as Grimmer
has noted [4] :
"Irrespective of the type of material to be burned, surprisingly similar profiles are
formed at a defined temperature. For example, thermal decomposition of pit coal,
cellulose, tobacco, and also of polyethylene, polyvinyl chloride and other plastic
materials which is carried out at 1000 °C (in air or nitrogen), yields the same
PAR in very similar ratios ... In a model installation, under comparable conditions,
similar PAR profiles are formed from different organic materials. Consequently,
PAR profiles seem to depend predominantly on temperature."
Nevertheless, many researchers have attempted to apply statistical techniques to
source apportionment modelling. An ambitious paper, published almost a decade
ago, attempted a critical evaluation of the utility of particulate-phase organic
compounds for receptor source apportionment [5]. Although the paper is
appropriately cautious, it concludes optimistically concerning the prospective use
of PAR for receptor source apportionment:
"PAR profiles of sources which have been repeatedly sampled by the
same researcher (italics added) appear to be fairly reproducible."
•
"The PAR compounds, in particular, appear most promising. Their
relative proportions in emissions from a given source type frequently
vary over several orders of magnitude which enhances their potential
usefulness for source discrimination."
"There also appear to be several unique or almost unique organic tracers
which may be useful such as levoglucosan, C-14 and cyclopenta (c,d)
pyrene."
Ecotoxicological Issues
1
INTRODUCTION
Polycyclic aromatic hydrocarbons (PAR) comprise a highly diverse class of
widespread environmental pollutants ([1], [2], [3]) They are formed as trace byproducts of combustion and pyrolysis of organic materials including fossil fuels
(gas, oil, coal), wood, plastics, food, municipal waste, etc. Figure 1 depicts
molecular structures for a small number of the PAR with particular attention
devoted to the structural diversity of this class.
A number of these compounds (e.g. benzo[a]pyrene, 18) are strong carcinogens,
while other closely-related species (e.g. benzo[e]pyrene, 19) are non-carcinogens.
Source apportionment for receptor modelling is a traditional discipline of
environmental research. Following analysis of the complex mixture of PAR
(» 100 compounds) in air particulate collected at a given location, its task is to
apportion the contributions arising from diesel-powered vehicles, automobiles,
space heating (gas, oil, wood), smoking, etc.
Obviously, the most straightforward scenario would be one in which each source
emitted a unique "marker" PAR at a constant (ng/BTU) rate or a consistent PAR
"signature" and that these marker PAR would have similar physical properties (e.g.
nonvolatile and thus totally associated with particulate matter) and would be
chemically unreactive under ambient environmental conditions.
Unfortunately, these conditions are never met: PAR mixtures are formed from
different sources with substantial overlaps in composition and many of the
individual PAR are highly reactive under ambient conditions. Indeed, as Grimmer
has noted [4] :
"Irrespective of the type of material to be burned, surprisingly similar profiles are
formed at a defined temperature. For example, thermal decomposition of pit coal,
cellulose, tobacco, and also of polyethylene, polyvinyl chloride and other plastic
materials which is carried out at 1000 °C (in air or nitrogen), yields the same
PAR in very similar ratios ... In a model installation, under comparable conditions,
similar PAR profiles are formed from different organic materials. Consequently,
PAR profiles seem to depend predominantly on temperature."
Nevertheless, many researchers have attempted to apply statistical techniques to
source apportionment modelling. An ambitious paper, published almost a decade
ago, attempted a critical evaluation of the utility of particulate-phase organic
compounds for receptor source apportionment [5]. Although the paper is
appropriately cautious, it concludes optimistically concerning the prospective use
of PAR for receptor source apportionment:
"PAR profiles of sources which have been repeatedly sampled by the
same researcher (italics added) appear to be fairly reproducible."
•
"The PAR compounds, in particular, appear most promising. Their
relative proportions in emissions from a given source type frequently
vary over several orders of magnitude which enhances their potential
usefulness for source discrimination."
"There also appear to be several unique or almost unique organic tracers
which may be useful such as levoglucosan, C-14 and cyclopenta (c,d)
pyrene."