Page 89 - Partition & Adsorption of Organic Contaminants in Environmental Systems

80 CONTAMINANT PARTITION AND BIOCONCENTRATION

solubility in water-saturated butanol decreases (or the g* o increases) more
rapidly than that in water-saturated octanol because of the higher butanol
polarity and high water content in butanol phase.
It is recognized that using logK ow as the reference partition constant tends
to reduce the level of data scattering in the correlation analysis for a group of
solutes. To understand this effect, one recalls that the observed correlation of
logK ow with logS w [Eq. (5.3)] applies satisfactorily to a wide variety of solutes
(except for highly polar ones), which is attributed to the unique polar versus
nonpolar balance of the octanol molecule. As a consequence, the use of
octanol–water as the reference system for a mixed class of compounds tends
to enhance the goodness of correlation over the use of other solvent–water
systems as the reference. In essence, K ow is much like an inverse of S w ,
which accounts to a large extent for a solute’s partition magnitude with a
particular organic phase. Thus, the correlation analysis using logK ow as the
reference minimizes differences in the behavior of a diverse group of
solutes with the reference system that could otherwise contribute to the over-
all data scattering.

5.8 BIOCONCENTRATION OF ORGANIC CONTAMINANTS

A major concern for environmental contamination is the extent to which pol-
lutants concentrate from water into aquatic organisms such as fish. The extent
of such concentration, termed the bioconcentration factor (BCF), is given by
the ratio of the pollutant concentration in fish to that in water. For nonionic
organic contaminants, there is good reason to believe that their bioconcen-
tration would occur by partition into certain biological components. It has
been observed that the concentrations of several refractory chlorinated con-
taminants (e.g., DDT and PCBs) in different fish species or in different tissues
of a fish correlate well with the lipid contents in whole fish or in its tissues
(Reinert, 1970; Roberts et al., 1977; Sugiura et al., 1979). This finding suggests
that the polar biological components, such as protein and carbohydrate, have
a relatively poor affinity for nonionic (especially, nonpolar) contaminants. The
BCFs of some chlorinated benzenes with guppies, rainbow trout, and bluegill
in laboratory systems were measured by Könemann and van Leeuwen (1980),
Oliver and Nimii (1983), and Banerjee et al. (1984). Chiou (1985) measured
the K tw (triolein–water) of a large set of organic solutes, including many chlo-
rinated benzenes, and found a high correlation between K tw and reported
BCFs when BCFs are normalized to the fish lipid content.
In most laboratory BCF studies, the test fish and contaminant are brought
to equilibrium either in a static-water system with a fixed initial load of con-
taminant or in a flow-through system in which the contaminant level in water
is kept constant during equilibration. Contaminant concentrations in fish and
water (the latter being frequently fixed) are monitored until equilibrium is
reached. A one-dimensional pharmacokinetic model has also been conceived