## Surfactantenhanced Desorption

To quantify the rate of solute desorption from soil, the following mathematical expression is often used:

where 5 is the concentration of the solute sorbed to soil (M/M), a is the mass-transfer rate coefficient (1/T), Kq is the sorption distribution coefficient (L3 /M), and C is the aqueous phase concentration of the solute (M/L3) (Deitsch and Smith 1995). For nonionic organic contaminants at equilibrium,

Therefore, equations (1) and (2) indicate that the rate of contaminant desorption from soil is a function of the mass-transfer coefficient, a, and the concentration gradient between the soil and aqueous phases. Surfactant-enhanced desorption focuses on these two factors. First, the interaction of surfactants with the soil organic matter is hypothesized to increase the magnitude of a (Deitsch and Smith 1995; Yeom et al. 1996; Sahoo and Smith 1997; Sahoo et al. 1998). Second, surfactant solutions may affect the apparent magnitude of K», thus altering the concentration gradient driving desorption from the solid phase to the aqueous phase (Deitsch and Smith 1995; Yeom et al. 1995; Yeom et al. 1996; Sahoo and Smith 1997; Sahoo et al. 1998). Equation (1) shows that the rate of desorption is directly proportional to the magnitude of the concentration gradient. These two mechanisms are now discussed in detail.

### 3.1 Surfactant Effects on a

Equation (1) quantifies the rate of desorption using a mass-transfer rate coefficient. An increase in the magnitude of a indicates a corresponding increase in the rate of solute diffusion within the sorbent matrix. The use of surfactants to increase desorption rates focuses on increasing the rate of solute diffusion through the soil organic matter. For a surfactant solution to increase the rate of solute diffusion through the condensed phase of the soil organic matter, the surfactants must cause conformational and structural changes within the soil organic matter.

Recent findings support the hypothesis that the glassy domain of soil organic matter is capable of undergoing conformational changes. First, Brusseau et al. (1991b) determined that the rate of organic pollutant desorption was increased by the ability of organic cosolvents to swell soil organic matter and reduce diffusive resistances. In another study (LeBoeuf and Weber 1997), conformational changes within the glassy domain of soil organic matter were assumed because the glass transition temperature, Tg, of the sorbent was decreased by solvent/sorbent interactions. As reported by LeBoeuf and Weber (1997) "Tg marks a second-order phase transition in which there is continuity of the free energy function and its first partial derivatives with respect to state variables such as temperature and pressure, but there is a discontinuity in the second partial derivatives of free energy." (The reader is referred to the preceding reference for a more detailed discussion of ) The decrease in represents a decrease in the degree of polymer cross-linking and as a result, there is a corresponding increase in the proportion of "soft" to "hard" carbon.

Such a mechanism could explain the effects of methanol on the rate of solute mass-transfer. LeBoeuf and Weber (1997) observed that a humic acid (solubilityparameter, ap= 11.5 (cal/cm3)05) swelled to greater degree in the presence of water (ctp=23.4 (cal/cm3)05) than did a poly(isobutyl methacrylate) polymer (ctp=8.63 (cal/cm3)0 5). The magnitude of the solubility parameter is a measure of the phase's intermolecular bonding energy. Phases with similar solubility parameters will interact more favorably than phases with dissimilar solubility parameters. Thus, the increased swelling was attributed to the water being able to interact more favorably with the humic acid than with the poly(isobutyl methacrylate) polymer, based on the magnitude of the solubility parameters. Although for a natural soil will not be as well defined as for a manufactured humic acid or polymer, it is reasonable to assume that ap for a natural soil could be within this range. Within this context, methanol with a (Jpof 14.5 (cal/cm3)05 would be expected to interact more favorably than water with the glassy domain, therefore causing a reduction in the average of the soil organic matter. Thus, the methanol facilitates the transformation of the glassy region into an amorphous region where solute diffusion may be orders of magnitude higher.

Several recent studies have demonstrated that surfactant solutions can increase the rate of contaminant diffusion within the sorbent matrix during desorption compared to non-surfactant solutions (Deitsch and Smith 1995; Yeom et al. 1996; Sahoo and Smith 1997; Sahoo et al. 1998). Deitsch and Smith (1995) demonstrated that Triton X-100 solutions at different concentrations increased the magnitude of for TCE desorption from a laboratory-contaminated peat soil by a factor of 2 to 3 compared to water. In another study (Yeom et al. 1995), researchers at the University of Tennessee showed that several different nonionic surfactants increased the magnitudes of polycyclic aromatic hydrocarbon diffusion coefficients for a weathered, coal tar-contaminated soil. Diffusion coefficients were up to two orders of magnitude higher in surfactant solutions compared to non-surfactant solutions. Sahoo and Smith (1997) quantified the rate of TCE desorption from a long-term field-contaminated soil. Experiments with Triton X-100 solutions showed that the mean value of a was increased 11.2% in batch experiments and 16.5% in column experiments compared to experiments sans Triton X-100. In a small-scale field test, Sahoo et al (1998) determined that flushing a section of a TCE-contaminated sandy aquifer with 300 mg/L of Triton X-100 increased the magnitude of a for TCE desorption by approximately 30%. Finally, Noordman et al (1998) reported that a rhamnolipid biosurfactant increased the rate constants associated with phenanthrene desorption in column experiments. In all of these studies, the increased rates of mass-transfer were attributed to the surfactants swelling the soil organic matter and increasing the rate of solute diffusion within the soil organic matter.

The hypothesized mechanism of enhanced desorption described in the previous paragraph can be understood within the framework of surfactant interactions with the soil organic matter. Depending upon the ap of a surfactant/water solution, surfactant solutions may transform portions of the "glassy" region of the soil organic matter into amorphous regions. Using an approach similar to LeBoeuf and Weber (1997), it may be possible to quantify the effect of different surfactant solutions on the TRof commercial humic acid or other polymer matrixes. Such experimental data would provide compelling evidence for the hypothesized mechanism of surfactant-

enhanced diffusion rates. Given the heterogeneous nature of natural sorbents, quantification of distinct values of Tg for natural soils is not possible.

0 0