3.1 Introduction
The rate of sorption/desorption of volatile organic compounds (VOCs) plays an important role in pollutant fate modeling and contaminated soil remediation.
Irreversible VOC sorption to or slow VOC desorption from soils in laboratory studies or field scales has been reported (Aochi and Farmer, 1995; Pingatello and Xing, 1996).
It is found that a portion of VOCs desorbs very slowly from soil particles and a certain fraction of it is retained strongly by the particles (Steinberg et al., 1987; Pingatello, 1990). The retention of VOCs by soil components affects the fate of pollutants in the environment and the effectiveness of contaminated soil and aquifer remediation.
Delineation of the sorption rate and of the reversibility of the sorption process with essential soil components is necessary to better predict the rates of pollutant migration and attenuation in soil.
Soil, a chemically heterogeneous matrix, contains various inorganic and organic components that each exhibits unique sorption behavior for pollutants (Chiou, 1998).
The chemical heterogeneity complicates the prediction of the sorption or desorption rates of pollutants in the soil. Recent studies reveal the existence of a slow sorption/desorption of some organic compounds with soils (Luthy et al., 1997;
Pignatello and Xing, 1996). This phenomenon has been attributed in part to a slow diffusion of the compounds through the micropores of soil particles (Farrell and Reinhard, 1994; Lin et al., 1994) and in part to their slow migration through soil organic matter (SOM) (Brusseau et al., 1991; Fu et al., 1994).
Sorption to humic substances is contended to contribute to the irreversible retention of xenobiotic compounds by soil (Cheshire, 1979). However, the sorption of toluene to humic acid, an integral member of soil humic substance, is found to be reversible and diffusion-controlled (Chang et al., 1997). Since humin represents a highly stable, recalcitrant, and high-molecular-weight fraction of SOM (Almendros et al., 1996; Hatcher et al., 1985), it may behave differently than the humic acid in the kinetics of sorption. Due to its cross-linked structure, humin may be capable of retaining VOCs for a significantly longer time and thus contribute to the slow or apparent irreversible sorption.
Although, the sorption of nonpolar solutes by humin shows a slightly nonlinear
effect, due to the existence of a small amount of high-surface-area carbonaceous material (Chiou et al., 2000), the soil organic matter, including humin, behaves by and large like a partition medium for VOCs (Boyd et al., 1988; Chiou, 1988; Chiou et al., 1990; Chiou and Kile, 1994; Chiou, 1998). The amorphous humic structure provides a
“solvent-like” medium that organic molecules can enter into or escape from it according to the thermodynamic gradient.
Humin is defined as the portion of humic materials that is insoluble in an aqueous solution at any pH. To be separated from the inorganic minerals in soils, humin is obtained by extensive digestion of soils with a mixture of concentrated HF and HCl (Stevenson, 1982). The HF/HCl extraction method has been widely applied for humin preparation (Chefetz et al., 2000; Grasset and Ambles, 1998; Guthrie et al., 1999;
Lichtfouse et al., 1998a; Lichtfouse et al., 1998b), while the MIBK (methyl isobutyl ketone) method, as suggested by Rice and MacCarthy (1990), has also been applied for extraction of humin. In the MIBK method, humic substances are allowed to partition between water and MIBK as a function of the pH in water phase. Humin is then isolated from fulvic acid and humic acid. However, this method is so selective that some humin component could not be retrieved by MIBK. To meet the purpose of our experiment on the sorption/desorption rate of toluene with near natural humin, the solvent extraction was not adopted in order to prevent a significant alteration of the humic composition.
A spectroscopic approach has been adopted to address the interaction between VOCs and humic substances by Aochi and Farmer (1997). The authors investigated the sorption/desorption behavior of 1,2-dicholoethane on humic acid and fulvic acid under dry conditions. They found that an absorbance band increases continuously after days of desorption and the sorbed chemical was strongly retained. However, the sorption kinetics of VOCs on humin in a system resembling the natural level of humidity on a short time scale (say, minutes) has not been investigated. In this study, toluene, a model nonpolar, mononuclear hydrocarbon, was used as the sorbate to delineate the sorption behavior of VOCs with humin. Humin disks were prepared and used to study the rate of transport of toluene in humin matrix with artificially exaggerated mass transfer distance. Thin humin films were also used to mimic natural humin in a near-natural soil environment.
3.2 Materials and Methods 3.2.1 Soil Sample
Approximately ten kilograms of Yamingshan soil, classified as medial, thermic, Pachic Melanudands according to the definition by USDA (USDA-NRCS, 1993), were
air-dried, freed of large plant debris, and screened through a 20-mesh (0.84 mm) sieve.
Small plant debris was further removed by flotation using ethanol. Then, soil sample was mixed well.
3.2.2 Humin
Humin was extracted according to the procedure developed by Rigol et al. (1998) and Russell et al. (1983) except that an ethanol/hexane mixture (1:1 v/v) was used to remove fats and waxes to avoid the interference from toluene, which is the target VOC to be studied. In short, soil was refluxed to remove fats and waxes and extracted by sodium hydroxide to remove humic acids and fulvic acids. The solid residue was separated by centrifugation, neutralized with 6 M HCl, washed with 0.1 M HCl and deionized water, and finally freeze-dried. The humin fraction was isolated from the solid residue by sequentially removing the mineral matter with a three-step digestion procedure: first, suspension in a 1:1 mixture of 0.2 M HF and 0.2 M HCl (20 ml/g) for 64 hours; subsequently, digestion in a 1:1 HF (5.5 M) and HCl (1.1 M) mixture for 1 hour three times; and, finally, digestion in 5.5 M HF four times for 16 hours each time.
After centrifuging and washing with 0.1 M HCl and water three times, the final residue designated as humin was freeze-dried and ready for use.
A previous investigation by IR spectroscopy (Rigol et al., 1998) suggested that the treatment of soils with HF decreases the structural mineral matter content with relatively little influence on the nature of humin. Despite the challenges presented by modification, humin obtained by HF-extraction procedure were used for some sorption experiments (Chefetz et al., 2000; Gurthrie et al., 1999; Rigol et al., 1998). The
13C-NMR spectrum (not presented) of the solid residue after removal of waxes, humic and fulvic acids, and treatment by weak acids before severe HF/HCl extraction, is the same as that of the humin product after the severe HF/HCl treatment. The severe treatment procedure removed most of soil inorganic components while the humin components were preserved.
3.2.3 Sample Characterization
Element Analysis. Elements such as C, H, and N of humin were quantified in
triplicate samples using an Element Analyzer (EA) (Perkin-Elmer CHN-2400).Inorganic carbon was removed according to Ball et al. (1990). To determine the contents of major elements such as Fe, Al, Si, and Ca, samples were pretreated by fusion with LiBO2 at 1000oC for 30 min. The product was dissolved in 0.9 M HNO3
and diluted to 0.3 M HNO3. The major elements in the solution were quantified in
triplicates by inductively coupled plasma optical emission spectroscopy (ICP-OES) (Ingamells, 1970; Rigol et al., 1998).
Solid-State
13C-NMR Spectrometry. The cross polarization/magic angle spinning
(CP/MAS) 13C spectra of samples were measured on a Bruker DSX400WB NMR Spectrometer with a 7-mm diameter probe. The spinning rate was 7000 Hz. The acquisition parameters included contact time of 1 ms, pulse delay of 1 s, and pulse width of 4.2 µs.The 13C-NMR spectra were analyzed according to the chemical-shift assignments made by Perminova et al. (1999) and Chefetz et al. (2000): 5-50 ppm, aliphatic H and C-substituted C atoms; 50-108 ppm, aliphatic O-substituted C atoms; 108-145 ppm, aromatic H and C-substituted C atoms; 145-163 ppm, aromatic O-substituted atoms;
163-190 ppm, C atoms of carboxylic, esteric, and amide groups. The distribution of C in each structural group was calculated as the percentage to the total carbon. The region between 5 to 108 ppm was calculated as aliphatic C and 108 to 163 ppm as aromatic C.
The total aromaticity was calculated by expressing the aromatic C as a percentage of the sum of aliphatic and aromatic C; the total aliphaticity was calculated as the percentage of aliphatic C to the sum of aliphatic and aromatic C (Hatcher et al., 1981; Hatcher et al., 1983). The regions 50-108 ppm and 145-190 ppm were calculated as O- or N- substituted C atoms. The polarity was assigned based on the percentage of the sum of O- and N- substituted C atoms.
3.2.4 Preparation of Humin Disks
Humin disks were prepared by pressing the humin powder under a pressure of 12.7 N/m2 for 1.5 min (Chang et al., 1997). Four disks, all being 12.45 mm in diameter, were 0.34 mm, 0.44 mm, 0.61mm, and 0.64 mm in thickness, and weighed 61.5 mg, 72.8 mg, 100.2 mg, and 102.5 mg, respectively. Their bulk densities were 1.49 g/cm3, 1.36 g/cm3, 1.35 g/cm3, and 1.32 g/cm3, respectively. The disks were oven-dried (105 oC) overnight and stored in a desiccator before use.
The scanning electron microscopy (SEM) photographs were taken with a Hitachi S-800 SEM. Figure 3-1 shows the SEM photographs of one of the disks prepared by the above-mentioned procedure. The surface morphology (Fig. 3-1a) and the exposed inner surface of a broken edge with only a few pores and cracks display the homogeneity of the disk.
3.2.5 Sorption/Desorption Experiment
Gravimetric Method. The apparatus was shown in Fig. 3-2 and the procedure used
for sorption have been described elsewhere (Chang et al., 1997). Briefly, the experimental apparatus was maintained in a thermostatic room at 15±0.1oC, 25±0.1oC, or 35±0.1oC. The set temperature was closely monitored for at least one day to ensure its stability before initiation of the experiment. The disk was hung on the sample side of a Cahn 200 electric microbalance enclosed in a glass chamber. The toluene mass flux to the disk was significantly larger than the maximum toluene removal rate, which keeps the vapor concentrations inside the chamber at virtually fixed levels. The experiment was terminated when the change of weight could not be distinguished from the base noise of the microbalance, which was about 2 µg/5hr. The concentration of the toluene was determined with GC/FID (Hewlett-Packard 5890II).
Sorption/Desorption Experiments Traced by FTIR. A drop of the humin
suspension in water was placed on the inner surface of a ZnSe window of a gas cell and dried in a desiccator. The absorbance spectra of IR beam passing through the gas cell windows were recorded on an IR spectrometer (BIO-Rad FTS 40) by averaging 200 scans at 2 cm-1 resolutions. The sample cell was purged with nitrogen gas carrying a constant toluene vapor concentration and relative humidity (RH below 1% for dry conditions and above 95% for humid conditions) during sorption experiments and purged with nitrogen gas without toluene during desorption experiments (Fig. 3-3).3.2.6 Estimating Diffusivity
The diffusion model and its incorporation into the gravimetric method has been described in detail by Chang et al. (1997). The model development is briefly summarized here. The one-dimensional mass conservation equation is
2
where q (mg/cm3) is the sorbate concentration in the disk at a distance x from the center plane of the disk and at time t; D is the apparent diffusivity of the sorbate inside the disk;