Page 106 - Partition & Adsorption of Organic Contaminants in Environmental Systems

IMPROPER SURFACE-AREA MEASUREMENT 97

known surface areas (Chiou et al., 1993); the latter is then used for convert-
ing Q m(N 2) to Q m(EGME) eq by Eq. (6.4), which then gives Q m(EGME) eq =
1.30Q m(N 2). Whereas a comparison of Q m(EGME) ap with Q m(EGME) eq pro-
vides an interesting test on the extension of the BET method with a polar
vapor for systems in which the solvent exhibits no significant penetration into
the bulk solid, it must be understood that the Q m monolayer values so deter-
mined with different vapors (including N 2) do not necessarily occur at the
same P/P°.
Inspection of the data in Table 6.2 shows that the Q m(EGME) ap values for
such solids as sand, hematite, alumina, kaolinite, and synthetic hydrous iron
oxide (SHIO) determined by use of EGME data with the BET equation are
virtually the same as the Q m(EGME) eq values based on the respective Q m(N 2)
values by the standard BET-N 2 method. In other words, if one were to deter-
mine the surface areas of these solids using Q m(EGME) ap together with
a m(EGME), the results would be practically the same as obtained by the stan-
dard BET-N 2 method. This finding manifests the fact that uptake of EGME
by these solids is confined essentially to external solid surfaces without any
significant solvent penetration into the solid (and without any significant
molecular sieving by the solid). For these solids, the results validate and extend
the utility of the BET model for surface-area determination with EGME (or
with similar polar vapors) as long as there is no significant bulk–solid pene-
tration, a basic assumption and requirement of the BET model. It is noted,
however, that while the observed (P/P°) m values for N 2 at Q = Q m(N 2) fall into
the common range (0.05 to 0.30) for inert gases, the observed (P/P°) m values
for EGME at Q = Q m(EGME) ap are generally lower. This effect suggests that
polar EGME adsorbs more efficiently than does N 2 (or other nonpolar vapors)
onto the surfaces of these solids through additional polar and/or possibly H-
bonding forces.
For the remaining solids, one sees that the ratios of Q m(EGME) ap to
Q m(EGME) eq are larger than 1, ranging from 1.5 for illite, 3.2 for Woodburn
soil, and 7.6 for Ca-SAz-1. In this case, if one were to determine the surface
areas of these solids by use of the respective Q m(EGME) ap values, the calcu-
lated surface areas would be higher than those by the standard BET-N 2
method to varying extents. Such discrepancies are caused by EGME penetra-
tion into solids, which is not a surface phenomenon and thus violates the basic
assumption of the BET model for surface-area determination.
For a strongly expanding clay (e.g., Ca-montmorillonite), the excessive
uptake of a polar solvent (e.g., water, EG, and EGME) actually occurs by
cation solvation (McNeal, 1964; Dowdy and Mortland, 1967; Tiller and Smith,
1990) rather than by forming internal monolayers that are intercalated
between silicate layers of the clay as assumed previously (Dyal and Hendricks,
1950; Quirk, 1955; Carter et al., 1965). Here the ion-dipole complex formed
2+
initially as a result of the Ca solvation with EGME may be considered as
“forming a new compound,” according to the view of Brunauer (1945), since
2+
the Ca solvation force is very potent. In a more recent study by Chiou and