A number of large-scale fire experiments with detailed quantitative analysis
of polycyclic aromatic hydrocarbon (PAH) including PAH congener distribution
have been conducted by SP. This data is reviewed here and is further assessed with
regard to toxicity applying a Toxic Equivalency Factor (TEF) model for estimation
of cancer potential. The PAH yield data from the large-scale fire experiments is also
compared to emission factors from other combustion sources. The study shows that
full-scale fire experiments with different products exhibit a large variation in total
PAH yields. Fires with products containing flame retardants were shown to produce
the highest yields and generally a more toxic mixture of PAHs than fires with nonflame
retarded products. The distribution of individual PAH congeners is generally
quantitatively dominated by low molecular weight PAHs, whereas a small number of
medium to high molecular weight PAHs are most important in determining the toxicity
of the PAH mixture. The large-scale fire experiments indicate that fires normally
produce orders of magnitudes higher yields compared to, e.g. modern residential
combustion devices. The relative distribution of individual PAHs, which determines
the toxicity of the PAH mix, is similar for the fires and open burning data studied, in
that benzo(a)pyrene and dibenz(a,h)anthracene dominate the toxicity of the mix as a
whole.