Page 64 - Partition & Adsorption of Organic Contaminants in Environmental Systems

OCTANOL–WATER SYSTEMS 55

K ow and S w values for a number of substituted aromatic compounds at room
temperature is given in Table 5.1.
Consider first the relation between logK ow and logS w for different organic
solutes; the solutes have relatively small S w values (i.e., large g w values) which
span over several orders of magnitude, as shown in Table 5.1. By contrast, the
solutes are usually very soluble in (i.e., highly compatible with) most organic
solvents. If the solutes form ideal solutions in water-saturated octanol and if
the solute solubility is the same in water and in octanol–saturated water, the
last two terms in Eq. (5.1) drop out, and what remains is a linear plot of
logK ow versus logS w, with a slope of -1 and an intercept of -logV * o. The inter-
cept in this case is essentially constant for all solutes in dilute solution. If one
sets K° ow as the partition coefficient from the ideal line, defined as

logK° ow =-logS w - logV * o (5.2)

then the difference between logK° ow and logK ow for a solute with a given log
S w expresses the effects of logg* o and log(g w/g* w) on logK ow. As noted in Eq.
(3.10), the term g w/g* w expresses the extent of solute solubility enhancement in
water by the dissolved organic solvent (in this case, octanol).
According to Eqs. (5.1) and (5.2), logK° ow - logK ow = logg* o + log(g w/g* w)
must be satisfied if the measured values are accurate and if Raoult’s law is
valid. For p,p¢-DDT and hexachlorobenzene (HCB), two highly insoluble
solutes, the supercooled logS w (mol/L) are -6.74 and -5.57 at 25°C, respec-
tively, and their (logK° ow - logK ow) values are 1.30 and 0.99. The respective
experimental g* o values, based on measured solute solubilities in water-
saturated octanol, are 7.8 and 5.4, or logg* o = 0.89 and 0.73 at 24 to 25°C; the
respective experimental g w/g* w values, based on measured solubilities in
octanol-saturated water and pure water, are 2.8 and 1.9, or log(g w/g* w) = 0.45
and 0.27 at 24 to 25°C. Thus, the data substantiate the expectation well. The
results show that the relative effects of the terms on the right of Eq. (5.1) on
K ow are, in decreasing order, water solubility (S w), compatibility with water-
saturated octanol (g* o), and the influence of dissolved octanol on water solu-
bility (g w/g* w). The major effect of S w is evidenced by the small solubility (or
the large g w) of relatively nonpolar organic solutes in water. The effect of g* o,
which increases with decreasing S w, is less than 10 for practically all solutes.
The effect of octanol saturation in water on solute water solubility (g w /g* w ),
which also increases with decreasing S w , is significant only for extremely water-
insoluble solutes (as liquids or supercooled liquids). We shall see later that the
solubility-enhancement effect for low-S w solutes by a dissolved organic sub-
stance is influenced not only by the concentration of the dissolved organic
substance but also profoundly by its molecular weight, polar-group content,
and molecular conformation.
In light of the fact that S w is the dominant factor in determining the
magnitude of K ow , a linear correlation should exist between these two para-
meters. A plot of logK ow versus logS w for compounds in Table 5.1 is shown in