Page 131 - Partition & Adsorption of Organic Contaminants in Environmental Systems

122 CONTAMINANT SORPTION TO SOILS AND NATURAL SOLIDS

solutes in SOM here is promoted by the low solubilities (i.e., poor compati-
bilities) of the solutes in water, which in effect accounts for the large differ-
ence between the sorption coefficients of DDT and benzene, as discussed
earlier. The weak adsorption of relatively nonpolar compounds on minerals
from water has been documented, for instance, for lindane and dieldrin on
sand (Boucher and Lee, 1972); lindane on Ca-bentonite (Chiou et al., 1984);
DDT on montmorillonite (Pierce et al., 1974); and phenanthrene on alumina,
kaolinite, silica, and silica gel (Huang et al., 1996). These results reflect
the strong adsorptive competition of water for minerals, in keeping with the
water-versus-benzene vapor adsorption data presented in Chapter 6. The
weak adsorption of phenanthrene from water on minerals (Huang et al.,
1996) results in an essentially linear isotherm, in which the heat of adsorption
is found to be less exothermic than the heat of condensation from water
(-DH w) (Huang and Weber, 1997). This suggests that a nonpolar solute in
water could only concentrate to some extent near the mineral surface without
being condensed to form a separate (condensed) phase.
Even for the sorption of more polar solutes, such as parathion on clay
(Saltzman et al., 1972), phenol on goethite (Yost and Anderson, 1984), and
2,4-D on montmorillonite (Haque and Sexton, 1968), the sorption also tends
to be rather weak. In most previous solute-sorption studies with clays, small
amounts of organic matter in the samples were usually ignored. Since a trace
amount of organic matter in unpurified clay minerals would have a significant
impact on the solute uptake from water, neglecting this effect could seriously
distort the data interpretation. Thus the relatively high uptake of 2,4-D by
unpurified illite clay (Haque and Sexton, 1968) might partly be an artifact of
the organic impurities in the clay.
Despite that the sorption data of low-polarity compounds in water are illus-
trative of the dominance of solute partition in soil/sediment organic matter,
the situation may be more complicated for relatively polar organic contami-
nants under some conditions. For example, a close inspection of the isotherms
of 2-chlorophenol and 2,4-dichlorophenol on soils, as reported by Boyd et al.
(1989),reveals that the isotherms display a discernible concave-downward cur-
vature at low relative concentrations (C e/S w) but a good linearity at high C e/S w.
In other words, the sorption coefficient is enhanced at low C e/S w but remains
nearly constant at high C e/S w. A similar effect was documented for several
polar substituted ureas (herbicides) on soils by Spurlock and Biggar (1994),
where the lowering of solute C e /S w by about three orders of magnitude to
-4
-5
levels of 10 to 10 resulted in an increase of the sorption coefficients by about
a factor of 3.5. Boyd (1982) found that the sorption coefficients of some
phenolic compounds at low C e /S w in their single-solute systems were reduced
by some 10 to 30% in binary- and ternary-solute systems.
To account for the enhanced sorption coefficients of polar ureas at low C e /S w
on soils, Spurlock and Biggar (1994) proposed a general nonlinear solute–
SOM partition model that couples a linear partition to SOM matrix with a
nonlinear specific interaction with active SOM groups. This specific interac-