Reevaluation of general partitioning model for sorption of hydrophobic organic contaminants by soil and sediment organic matter

Abstract

Abstract-This evaluation of the polymer-based Flory-Huggins sorption theory has shown that the model predicts relatively concentration-independent values for rubbery matrices, concentration-dependent values for more glassy matrices, and highly concentration-dependent, largely negative interaction parameters for sorption of phenanthrene onto Green River kerogen, Ohio Shale II kerogen, Illinois No. 6 and Wyoming coals, and poly(phenyl methacrylate). These greatly negative enthalpic contributions to the interaction parameter are only to be expected where specific sorbate-sorbent interactions exist, such as in polar sorbatepolar sorbent systems. The failure of the nonlinear partitioning model is explained in terms of its inability to predict these increased interactions of sorbing solutes within microvoids present within glassy matrices. Calculations of sorbent solubility parameters through use of this model do, however, provide good agreement with literature solubility parameter values for specific sorbents; it may thus prove useful for partial characterization of sorbents for which solubility parameters are unknown.

Soil organic matter beyond molecular structure Part I: Macromolecular and supramolecular characteristics

“…Comparably negative χ values were, however, also observed by LeBoeuf and Weber (1999) for the sorption of phenanthrene to kerogens, coal, and poly(phenylmethacrylate), where strong hydrogen bonds can be excluded. To explain this contradiction, LeBoeuf and Weber (1999) suggest that the sorbate is enclosed in rigid micropores, which then leads to an increase of sorbent load due to fixation in the pores instead of specific interactions and thus to an only apparently negative interaction parameter.…”

“…The degree of nonlinearity of the sorption is expressed by m and describes the dependence of sorbate partial molar free energy in the sorbed phase (~ln c 1 ) on sorbed-phase loading (Spurlock and Biggar, 1994a). The Flory-Huggins nonlinear partitioning model was applied to NOM for the sorption of substituted phenylureas (Spurlock and Biggar, 1994a) in soil samples and for phenanthrene sorption in biopolymers and humic substances (LeBoeuf, 1998;LeBoeuf and Weber Jr., 1999). For sorption of phenylureas to soils, the Freundlich exponents vary between 0.9 and 0.6, and the standard-state interaction parameters χ* range from 0.95 to 3.28, while at low loadings, χ became strongly negative (up to -2) for fenuron, monuron, and diuron (Spurlock and Biggar, 1994b).…”

Validation of a Modified Flory–huggins Concept for Description of Hydrophobic Organic Compound Sorption on Dissolved Humic Substances

“…The Flory interaction parameter is composed of an enthalpic and an entropic component: ϭ h ϩ s . For humic organic matter, s was assumed to be in the range of 0.25 [13] to 0.34 [16,20], independent of the interacting solute. By varying the assumed value of s in the range of 0.1 to 0.5, Chin and Weber [20] concluded that their model was relatively insensitive to deviations from the ideal entropy of mixing and agreed with Chiou et al [13], who stated that h comprised a significant part of the total Flory parameter.…”

“…In case of sorption on particulate OM (soils or sediments), several more sophisticated approaches for prediction of K OC have been proposed, based, for example, on the Flory-Huggins theory [13][14][15][16], molecular connectivity index [17], linear solvation energy relationships (LSERs) [18], or quantum-mechanical universal solvation models [19]. For describing sorption on DOM beyond empirical K DOC to K OW correlations, so far only models based on Flory-Huggins theory [15,20,21] as well as an LSER approach [22] restricted to phenols have been applied.…”

“…LeBoeuf and Weber [16] stated that the Flory-Huggins model is not able to describe nonlinear isotherms for sorption on SOM, which may be caused by mechanisms other than partitioning (e.g., filling of microvoids within glassy polymers). Although natural DOM may be similar to SOM from the chemical point of view, it is assumed to be much more flexible, and it does not have such a relatively rigid pore network.…”