Environmental Engineering Reference
In-Depth Information
Since the group contribution scheme of Hine and Mookerjee is of limited
applicability for compounds with multiple polar groups, Meyland and Howard (1992)
proposed an alternative, but more reliable group contribution scheme whereby
log K aw =
a i q i +
b j Q j .
(4.7)
i
j
Appendix 6 lists q i and Q j values for several bonds.
4.1.2 E XPERIMENTAL D ETERMINATION OF H ENRY'S L AW C ONSTANTS
In Chapter 3, we noted that
V w
RT
P i
x i
,
K aw =
ยท
(4.8)
which means that K aw can be obtained from a ratio of the saturated vapor pressure
and aqueous solubility of a sparingly soluble compound.
Based on the above premise, there are several methods for K aw measurements
described in the literature. The different methods fall generally into three categories:
(1) from a ratio of vapor pressure and aqueous solubility that is independently mea-
sured; (2) static methods; and (3) mechanical recirculation methods. In the first
category of methods, the errors in independent measurements of vapor pressure and
aqueous solubility lead to additive errors in K aw . In the second category of meth-
ods, because of difficulties associated with the simultaneous precise measurements
of concentrations in both air and water, they are restricted to compounds with large
vapor pressures and aqueous solubilities. Such methods are ideally suited for solu-
ble solutes, such as CO 2 ,SO 2 , and several volatile organics. The EPICS method of
Gossett (1987) and the direct measurement technique of Leighton and Calo (1981)
fall under this category. In reality, both categories 1 and 2 methods should be superior
to the third since they require the measurement in both phases which would allow
mass balance closure if the initial mass of organic solute introduced into the system
is accurately known. The errors in these methods arise directly from the concentra-
tion measurements. The third class of methods (e.g., batch air stripping, wetted wall
column, and fog chamber) suffers from a serious deficiency, and that is the necessity
to ascertain the approach to equilibrium between phases.
Several investigators have compiled Henry's constants for a variety of compounds
of environmental interest (e.g., Mackay and Shiu, 1981; Ashworth et al., 1988). The
values reported show considerable scatter for a single compound. Table 4.1 is an
example of the degree of agreement reported by different workers. It is reasonable
to expect a great deal of scatter, especially for those compounds that have low vapor
pressures and aqueous solubilities. The standard deviation in H c is only 2% of the
mean for benzene, whereas it is 12% of the mean for chloroform and 17% of the
mean for carbon tetrachloride. The standard deviation is 67% of the mean for lindane,
which has a low H c value. For most environmental purposes, it has been suggested
that a reasonable standard error in H c is about 5-10% (Mackay and Shiu, 1981).
 
Search WWH ::




Custom Search