Environ. Sci. Technol. 2003, 37, 2855-2860
Comparison of Sorption Domains in Molecular Weight Fractions of a Soil Humic Acid Using Solid-State 19 F NMR M. KHALAF,† SCOTT D. KOHL,‡ E . K L U M P P , † J A M E S A . R I C E , * ,‡ A N D E . T O M B AÄ C Z § Institute of Chemistry and Dynamics of the Geosphere IV, Agrosphere, Research Center Ju ¨ lich, D-52428 Ju ¨ lich, Germany, Department of Chemistry and Biochemistry, South Dakota State University, Box 2202, Brookings, South Dakota 57007-0896, and Department of Colloid Chemistry, University of Szeged, H-6720 Szeged, Hungary
Humic acid was fractionated into eight different molecular size components using ultrafiltration. Solid-state CPMAS 13C NMR demonstrated that fractions larger than 100 000 Daltons were primarily aliphatic in character, while fractions smaller than 30 000 Daltons were predominantly aromatic in character. Solid-state 19F NMR examination of the sorptive uptake of hexafluorobenzene (HFB) by HA and each of the fractions gave spectroscopic evidence for the existence of at least three sorption sites in the smaller molecular size fractions, while two predominant sorption sites could be established in the larger molecular size fractions. Sorbed HFB displayed higher mobility in the smaller, more aromatic fractions while HFB in the larger, more aliphatic fractions displayed lower mobility. The relative mobilities of HFB in each sorption domain suggest that the rigid domain may be composed of aliphatic carbon rather than aromatic carbon moieties. In larger size fractions, this domain may be the result of rigid, glassy regions composed of aliphatic molecules or side chains.
Introduction A wide variety of diagenetic processes acting on a diverse assemblage of starting materials are responsible for the formation of humic materials (1-3). Humic acid (HA) is the alkaline soluble, acid insoluble fraction of humic materials. Generally, the properties of HA are the result of the interactions of a polydisperse mixture of aliphatic, aromatic, carbohydrate, and amino containing components. Conceptual models and simulations suggest that these component molecules form inter- and intramolecular hydrophobic domains (4, 5). These domains are an important consideration in understanding the environmental fate and transport of hydrophobic organic compounds (HOC) (6). The broad range of characteristics, size, and the lack of a definitive structure (7, 8) make it difficult to elucidate the interaction mechanisms between HOCs and HA. An alternate method to gain insight into these mechanisms is to fractionate the HA mixture to decrease the chemical heterogeneity prior * Corresponding author phone: (605)688-4252; fax: (605)688-6364; e-mail:
[email protected]. † Research Center Ju ¨ lich. ‡ South Dakota State University. § University of Szeged. 10.1021/es0206386 CCC: $25.00 Published on Web 05/28/2003
2003 American Chemical Society
to contaminant interaction studies. Among the methods available, ultrafiltration (UF) is a reasonably simple method to fractionate polydisperse mixtures of molecules (9, 10). Several studies are available on the chemical heterogeneity of HAs in relation to molecular size distribution (11-18). While somewhat contradictory, these studies indicate that the low molecular weight humic fractions are generally thought to be more hydrophilic and mobile than the high molecular weight, hydrophobic fractions (19). This work also suggests that chemical composition varies with molecular size and that differences in the chemical composition between size fractions may have significant consequences for the environmental chemistry and geochemistry of humic substances. Current contaminant-organic matter models can be differentiated by their mechanism. Two mechanisms, “partition” and “sorption”, have been frequently employed to describe HA-HOC association (e.g. refs 18-21). Partitioning views HA as a gel-like polymer that represents a hydrophobic environment where HOC will preferentially be “extracted” from an aqueous environment (22-24). In the “sorption” model, HA may be viewed as a mixture of macromolecules possessing specific hydrophobic domains where HOCs, thermodynamically excluded from the aqueous phase due to their hydrophobicity, may preferentially “bind” (6, 25). Schlautman and Morgan (25) noted that the binding of HOCs by dissolved humic materials depends on the hydrophobicity as well as the size of the solute molecules which are both important properties in determining their ability to fit into proposed hydrophobic cavities in humic materials. This reasoning alludes to a host/guest phenomenon as being a likely mechanism for the association of polycyclic aromatic compounds with humic substances. Conceptually, interactions occurring by the “partition” mechanism can be compared to the sorption of HOCs to rubbery polymers (26), and interactions occurring by the “sorption” mechanism can be compared to sorption of HOCs to glassy polymers (27). Despite the large number of studies supporting a partitioning mode of interaction based on linear HOC sorption isotherms on HA (e.g. refs 28-30), many recent studies give evidence for sorption character by reporting nonlinear sorption isotherms (31-37), solute-solute competition (27, 32, 38), and desorption hysteresis (39-42). These observations cannot be adequately explained by a purely partitioning model and have been explained by attributing “dual-mode” sorption properties to soil or sediment organic matter SOM (43-47). Polymers have been shown to possess both rubbery and glassy states and have been described as dual-mode sorbents (48, 49). The glassy state is characterized by a more condensed three-dimensional structure where the polymer segments have measurably higher cohesive forces than in the rubbery state (50-53). Sorption of gases and organic molecules to the rubbery state occurs by dissolution, while sorption to the glassy state occurs by concurrent dissolution and hole-filling mechanisms. The holes are postulated to be local regions of physical voids having molecular-sized dimensions where a limited number of sorbate molecules may undergo an adsorption-like interaction with an internal surface. A number of investigators have drawn analogies between HA and synthetic organic polymers. And while HA is certainly not the only organic matter fraction in a soil or sediment involved in HOC sorption, its ability to interact strongly with a variety of HOCs makes it an interesting component with which to initiate a study of the effect of organic matter chemical characteristics on HOC sorption. Humic substances have been described as having expanded and condensed VOL. 37, NO. 13, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2855
regions (44, 46-47), which may be analogous to rubbery and glassy states found in polymers. This theoretical treatment of polymers states that sorption into rubbery (“soft”) domains is fast and displays linear isotherms while that into glassy (“hard”) domains is slow and displays nonlinear isotherms (36, 43, 45, 54). There has been significant discussion regarding, among other experimental data, linear and nonlinear sorption isotherms of contaminants to soil systems. These discussions have produced a number of studies that have resulted in the proposition of several different conceptual models for soil humic matter that can broadly be encompassed under the term “dual-mode” sorption. “Dual-mode” sorption models make specific claims about the presence of microscopic HA domains (27, 31-33, 35-36, 39, 42, 44-47, 55, 56). The existence of these microscopic domains cannot directly be observed through macroscopic sorption experiments. Among these claims there is question over the actual existence of different organic matter phases (57) or distinctly different local chemical moieties that are cited as being responsible for nonlinearity in sorption isotherms, competitive sorption, desorption hysteresis, and other results that seems to contradict a “partition” mechanism. A recent review of the chemical interactions of hydrophobic organic contaminants with soil organic matter recognized the need for “direct observational data” of the location of sorption interaction (56). Such data are difficult to obtain due to the heterogeneity of SOM. In light of this heterogeneity, solid-state NMR may be the best technique for these observations because of its ability to describe the local chemical environment of NMR-active nuclei. Incorporating such a nuclei in a suitable model HOC molecular probe allows the direct observation and description of different chemical domains that may be present in a sorbent such as HA. Using this technique, dual-mode sorption domains in soil or sediment (SOM) have been observed by examination of the sorption of HFB to SOM (58). In this study, we reduce the heterogeneity of a soil HA by UF. The fractions obtained from the UF fractionation are characterized by 13C CPMAS to compare relative carbon type differences between the fractions. The sorption of the HFB probe molecule is observed via solid-state NMR to obtain data about the local chemical environments of sorption domains of the HFB after interaction with HA. The sorption data is correlated with the 13C spectrum and molecular-size range for each fraction to assess the impact of size fractionation on the interaction between HFB and HA. This correlation gives insight into the different substructures that should be further investigated in study of the interaction of HA with hydrophobic organic contaminants.
Materials and Methods Hexafluorobenzene (99%) was obtained from Aldrich. It has a water solubility of 342 mg/L and a log Kow of 2.55 (59). All other solvents and chemicals were reagent grade or better and used as received. Extraction and Purification of Humic Acid. Soil was collected from the Ap horizon of the Orthic Luvisol, Merzenhausen, Germany (60). The field-moist soil was passed through a 5-mm sieve and stored at -17°C. Humic acid was isolated using standard extraction and purification procedures as recommended by the International Humic Substances Society (61). Briefly, 500 g of moist soil material was suspended in 5 L of argon-purged 0.1 M NaOH, shaken for 24 h at 25 °C, and centrifuged for 15 min at 13 000 rpm. The supernatant containing fulvic and humic acids was filtered through glass wool and acidified to pH 1.0 with HCl to precipitate the HA. To achieve complete separation of fulvic acid, the precipitated HA was redissolved in 0.1 M NaOH and precipitated with HCl twice as described above. The HA fractions were collected together and treated three times 2856
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 13, 2003
TABLE 1. Ultrafiltration Humic Acid Size Fractions and Mass Percent Basis of Whole Humic Acid Found in Each Fraction size fraction
wt %
F0 (>0.2 µm) F1 (300 kDa-0.2 µm) F2 (100-300 kDA) F3 (50-100 kDa) F4 (30-50 kDa) F5 (10-30 kDa) F6 (3-10 kDa) F7 (0.5-3 kDa)
16.5 ( 4.9 7.8 ( 0.5 8.5 ( 2.6 12.3 ( 1.2 6.1 ( 0.4 6.0 ( 0.6 10.7 ( 0.9 19.9 ( 1.5
with a 0.1 M HCl-0.3 M HF solution to remove mineral impurities. The precipitated HA was dissolved again in 0.1 M NaOH. The ash-free HA was dialyzed against deionized water using Spectra/Por 6 tubing (MWCO 1000) for 1 month. The extracted HA was freeze-dried and stored in the dark at 3 °C. Size Fractionation of Humic Acid. The isolated and purified HA was separated into eight molecular size fractions using ultrafiltration (62, 63). An alkaline solution of HA (1 g/50 mL) was filtered through a 0.2 µm Nylon filter using an Amicon ultrafiltration stirred cell (model 8050) at a pressure of 2.5 bar (argon). The retentate (F0) was washed with small portions of Millipore water until a colorless liquid passed through the microfilter. The filtrate obtained was then fractionated by using an Amicon ultrafiltration stirred cell (model 8400) and a series of Amicon membranes of successively smaller pore size to obtain HA fractions with different nominal molecular size which were classified as F0, larger than 0.2 µm; F1, 0.2 µm-300 000 Da; F2, 300 000100 000 Da; F3, 100 000-50 000 Da; F4, 50 000-30 000 Da; F5, 30 000-10 000 Da; F6, 10 000-3000; and F7, 3000-1000 Da. All separations were performed under identical conditions. The complete separation process was repeated at least three times to assess the reproducibility of the fractionation results. The HA fractions were freeze-dried and stored in the dark at 3 °C. The mass balance of the eight fractions showed an average recovery of ∼87.8%. Table 1 lists the weight percent of the HA fractions obtained from the fractionation process. 13C NMR Parameters. Comparative solid-state 13C NMR spectra of the dried humic materials were obtained on a Bruker DSX 200 operating at a frequency of 50.3 MHz using 7 mm OD zirconia rotors with KEL-F caps. A ramped cross polarization magic angle spinning technique (6.8 kHz spinning rate) 1H-pulse was used to circumvent spin modulation of Hartmann-Hahn conditions (64). A contact time of 1 ms and a 90° 1H-pulse width of 5.3 µs were used for all spectra. These conditions do not guarantee quantitative spectra, but the data do indicate relative changes. The 13C-chemical shifts were referenced to tetramethylsilane () 0 ppm), using glycine as an external standard (COOH: 176.04 ppm). Between 2 × 104 and 1 × 105 scans were accumulated using a pulse delay of 400 ms (65). Prior to Fourier transformation, a line broadening of 0-75 Hz was applied, depending on the sensitivity of the sample. The relative intensities of the peak areas were obtained by integration of chemical shift ranges typically employed for HA (e.g., ref 66) using an integration routine supplied with the instrument software. Sample Preparation for 19F NMR Experiments. Water was added to the different HAs to achieve a moisture level of 40% of their paste saturation capacities. The sample was then stirred by hand with a microspatula to homogenize it. The sample was placed in a 4 mm OD zirconia NMR rotor equipped with a vespel endcap. Liquid HFB was placed directly onto the HA samples at room temperature at levels corresponding to 7600, 7300, 7300, 7100, 6900, 7100, 5500, 4200, and 4300 mg/kg for the whole HA, F0, F1, F2, F3, F4, F5, F6, and F7, respectively. The HA fractions and the HFB
FIGURE 1. fractions.
13C NMR spectra for whole humic acid and ultrafiltration
were in contact for 24 h prior to characterization by 19F solidstate NMR. 19F NMR Parameters. Initial 19F NMR experiments were performed in the single-pulse mode to optimize the acquisition rate and 90 degree pulse angle while avoiding saturation of the nuclei due to inappropriately short recycle or delay times. Subsequent static and magic angle spinning (MAS) 19F NMR experiments were performed in the single-pulse mode (4.3 ms pulse, 49 kHz window, 0.334 s acquisition, 1.00 s delay between pulses) on a Bruker Avance 300 spectrometer equipped with a 4 mm solid state probe at a frequency of 282.414 MHz and externally referenced to CFCl3 (neat HFB taken as -163.0 ppm) (68). Rotor spin rates were held constant for each spectra acquired but were varied between 4 and 15 kHz in separate experiments for identification of spinning sidebands. A line broadening of 50 Hz was applied to all spectra except the fine detail inserts, which have no line broadening applied.
Results and Discussion Whole Humic Acid. Integration of the unfractionated HA 13C NMR spectrum (Figure 1, HA and Table 2) shows a carbontype distribution that is typical of HA.
The 19F spectrum obtained for the interaction of HFB with whole HA shows two distinct resonances (Figure 2, HA) that have been attributed to different sorption domains (58). The most intense peak is centered at -159.7 ppm. This resonance also has a broad, low intensity shoulder upfield from the peak maximum (-170 to -175 ppm). The sideband intensities from this peak are unevenly distributed. The downfield sidebands are more intense than those upfield; this is particularly evident when comparing the two sidebands nearest to the central peak. This resonance contains the majority of the 19F signal, and thus the majority of the sorbed HFB in the system. The broadness of this resonance indicates a significant 19F chemical shift anisotropy. Due to this anisotropy, the -159.7 ppm resonance is attributed to immobile or motionally restricted HFB. The -168.1 ppm peak has no discernible sidebands and an 18 Hz line width at half-height (LWHH), significantly narrower than that of the -159.7 ppm resonance (1500 Hz LWHH). It is very sharp in both the 8 kHz MAS experiments and the static experiments. The fact that this peak does not appreciably broaden in static experiments (spectra not shown) indicates that the resonance is caused by highly mobile HFB molecules. This resonance is not due to neat HFB because its 19F resonance is -163.0 ppm, and the resonance is not observed after ∼6 h contact time. This resonance is also not attributable to HFB dissolved in water because its resonance was measured at -159.3 ppm. The presence of HFB dissolved in water in these experiments would result in a very narrow peak with the intensity of that signal concentrated accordingly. To put it another way, the 19F resonance of dissolved HFB in these experiments would be a solution-state spectrum. As such, it is likely that even a small amount of HFB would give a very intense signal because HFB in water would lack the chemical shift anisotropy that is clearly apparent in the -159.7 ppm resonance. In addition, the presence of such a peak is not observed in any of the spectra observed in Figure 2, even in those spectra where the -159.7 ppm resonance is absent or of comparatively intensity. The most probable source of this resonance is highly mobile HFB that is physisorbed or somehow weakly attracted to surface groups in such a manner that allows the HFB molecules to rapidly “hop” from site to site. The rapid movement between surface sites would result in a very sharp and well-defined resonance. Fractionated HA. The general progression of changes in the chemical character of the fractionated HA with decreasing molecular size is a decrease in aliphatic character and an increase in aromatic and carboxylic character. Of these chemical characteristics, the carboxylic character has the highest variability but clearly increases relative to the decreasing apparent molecular weight (Table 2). Humic acid fractions can generally be divided into three different groups based on their molecular weight, 13C, and 19F NMR spectra. The first group contains HA high molecular weight fractions F0, F1, and F2. These fractions have, on average, approximately 16% more aliphatic, 39% less aromatic, and 10% less carboxylic carbon relative to the whole HA. These fractions account for 33% of the mass of the whole HA (Table 1). These fractions all display a strong, broad signal in the -160 ppm HFB sorption domain similar to the whole HA. This is the dominant sorption domain in this HA size range and is termed “rigid” due to its very broad nature (48, 58, 68). The broad nature of the resonance indicates the likely presence of a wide variety of different interactions between the HA size fraction and HFB. This seems to be a reasonable speculation since the largest molecules would likely have the most heterogeneous composition, structural variability, and potential for varied intra- and intermolecular interactions. The rigid sorption domains all have asymmetrical peaks with higher intensities being observed on the downfield VOL. 37, NO. 13, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2857
TABLE 2. Carbon Type Distribution for Ha and It Size Fractions Based on Integration of the 13C Spectraa sample
0-45 alkyl C
45-110 O-alkyl C
110-160 aromatic C
140-160 phenolic C
160-185 carboxyl C
185-220 carbonyl C
Carom/ Caliph
HA Fr0 Fr1 Fr2 Fr3 Fr4 Fr5 Fr6 Fr7
32.8 39.7 38.8 36.3 20.3 23.4 19.8 19.7 18.8
37.1 39.6 40.6 40.9 48.9 44.4 44.9 44.5 36.9
19.6 10.7 11.3 14.0 19.3 19.9 21.6 22.6 27.9
5.4 3.02 2.97 3.62 5.28 4.97 6.16 6.84 8.28
9.2 8.9 8.1 7.5 9.0 9.6 10.6 9.8 12.8
1.3 1.1 1.2 1.3 2.5 2.7 3.1 3.4 3.6
0.28 0.13 0.14 0.18 0.28 0.29 0.33 0.35 0.50
a The ratio C arom/Caliph was calculated as the ratio: aromatic C (110-160 ppm)/(alkyl C (0-45 ppm) + O-alkyl (45-110 ppm)). A smaller value of the ratio indicates a material with a higher aliphatic carbon content.
side. This suggests that this sorption domain is skewed to moieties that have lower electron density around the fluorine nuclei. This would likely result from sites with high electron-withdrawing abilities. Structures such as carbonyl or other oxygen bearing moieties would be likely candidates to produce such results. This domain narrows slightly going from the heavier to lighter fractions suggesting a loss of heterogeneity in available sorption domains with decreasing molecular size. The -168 ppm resonance is very narrow in all of the spectra acquired. Due to the narrowness of this peak it is attributed to loosely bound, mobile HFB (48, 58). It is worth noting that the sorption spectrum of the largest fraction (i.e. F0), though accounting for only 16.5% of the mass of HA, is almost identical to the sorption spectrum of the whole HA. This suggests that the largest molecules can be used as a reasonable surrogate for studying the sorption properties for whole HA for hydrophobic compounds. The second group contains the intermediate molecular weight HA fractions F3 and F4. These fractions have, on average, carbon contents that are approximately 34% less aliphatic, and with similar aromatic and carboxylic contents relative to the whole HA. These fractions account for 18.4% of the mass of the whole HA. These fractions display spectra that can be perceived as transitioning from the sorption domains found in the large fractions to those observed in the smaller fractions. The -160 ppm sorption domain changes significantly from a very broad, somewhat featureless domain, to a smaller, narrower domain with multiple peaks and slightly different chemical shifts. Fraction 4 (Figure 2, F4) displays this transition the most clearly. All peaks observed in all the other fractions are present to some degree in this fraction. The highly mobile peak attributed to HFB at -168 ppm is clearly evident. The -164 ppm peak observable in the lighter fractions is evident as two weak resonances. The -160 ppm resonance is actually composed of two peaks, one broad resonance and a second, sharper resonance (Figure 2, F4). These two resonances represent the averages of two different chemical environments, the broader domain centered at -159.9 ppm corresponds to the broad domain in the heavier fractions and the narrower resonance centered at -160.6 ppm corresponds to the broad domain in the lighter fractions. This observation suggests that there are different sorption domains in HA which may be dependent upon the organization of HA components with different molecular weights and/or structural characteristics. It is also an indication that the different size components of HA have different chemical and/or structural differences that have significant impact on the type of sorption environment available to nonionic, nonpolar compounds. The third group contains the low molecular weight HA fractions F5, F6, and F7 that together represent 37% of the total HA sample. These fractions have, on average, carbon contents that are approximately 41% less aliphatic, 21% more 2858
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 13, 2003
aromatic, and 20% more carboxylic relative to the whole HA. These fractions have lower aliphatic, higher aromatic, and higher carboxylic contents than the whole HA. These fractions display three unique sorption domains. The -164 ppm sorption domain is very sharp; however, it still results in an observable spinning sideband (-136 ppm). This spinning sideband is evidence that this sorption domain is not due to a liquidlike environment but a more rigid environment with a broad chemical shift range. This conclusion is also supported by the static spectra (not shown) that show no evidence of a highly mobile resonance at -164 ppm. In comparison, the -168 ppm domain is much more narrow with a clear signal present in the static spectrum. The -164 ppm resonance is composed of two peaks, a sharper peak with the same average chemical shift as the broader peak. These peaks suggest two different sorption domains in which HFB finds a similar local chemical environment that differs only in its relative mobility. The rigid sorption domain predominates in samples that have higher aliphatic character and larger average molecular weights (i.e. whole HA and fractions F0, F1, and F2). Clauss et al. (50) have shown that polymer systems containing long aliphatic side-chains (saturated C16 chains) and different backbone configurations can form aliphatic crystalline phases. The crystalline domains were found to be the result of main chain (backbone) packing in planar arrays allowing ordered interactions between aliphatic side chains. Young and Weber (35) have proposed that adsorption onto microcrystalline structures in SOM that are formed during diagenesis are important sites for adsorption. We propose that the rigid sorption domain observed in these studies is due to rigid glassy moieties composed primarily of aliphatic molecules. The lighter fractions also have aliphatic character, but it is possible that the glassy regions are likely produced predominately in the higher molecular weight fractions because their larger relative sizes allow for the formation of this domain. This larger size and higher relative aliphatic content results in the necessary combination of aliphatic character and molecular organization that allows rigid domains or microvoids to form. Fraction F0 and whole HA have almost identical sorption spectra. It is possible that the rigid sorption domain of glassy moieties produces regions of such strong affinity for HOCs that HFB preferentially interact almost exclusively with those domains. However, in the absence of these rigid domains, the sorption profile looks much more like partitioning, as observed in the lighter fractions. The mobile sorption domain predominates in samples that have lower molecular weights and higher aromatic character. These smaller molecules either do not have the necessary molecular size and/or aliphatic content to form glassy regions (it is likely a combination of deficiencies in both properties). As a result the molecules interact together
It is unclear from the data obtained whether the different sorption domains found in the smaller molecular weight fractions are also present in the larger fractions. If they are present, the large signal from the rigid domain observed in the larger fractions overwhelms these domains making it impossible to determine their presence using these experiments. This study provides evidence for a role for aliphatic components of natural organic matter in creating what has been variously referred to as the rigid, immobile, or glassy domain in soil and sediment organic matter. It also suggests that such a domain only exists in higher molecular size components of the organic matter in soils and sediments. These observations provide initial insights into HOC organic matter interactions based on direct observation of a HOC with an organic substrate that are relevant to understanding the fate and transport of HOC in the environment. They form a basis for the continued study of the molecular interactions between HOC and organic matter.
Conclusions Humic acid molecules have many different chemical environments into which nonionic molecules such as HFB can sorb. The chemical and physical characteristics of these domains are highly dependent upon the type of carbon present in the HA. Small HA molecules have at least three sorption sites. Those sorption sites are more clearly defined and homogeneous than the sorption domains found in larger HA molecules. The larger molecular weight fractions are dominated by a rigid sorption domain. It is proposed that this sorption domain may result from regions composed of aliphatic moieties. The presence of these different sorption domains gives evidence that the interactions of HA and HOC cannot be strictly viewed as a simple partitioning process.
Acknowledgments This project has been funded by the Office of Naval Research’s Harbor Processes Program under grant number N00014-991-0587. It has not been subjected to the agency’s peer and administrative review and therefore may not necessarily reflect the views of the agency. No official endorsement should be inferred. We gratefully thank Dr. H. Knicker for performing the 13C NMR measurements.
Literature Cited
FIGURE 2. 19F NMR spectra for sorption of hexafluorobenzene to humic acid and ultrafiltration fractions. A “1” denotes sidebands from -160 ppm resonance and a “2” denotes sidebands from -164 ppm resonance. “T” denotes sidebands from PTFE (isotropic peak at -120 ppm) which is a contaminant introduced by abrasion of Teflon-coated stir-bars used in the HA extraction. to produce a sorbent whose interactions with HOC are best described as partitioning. Hexafluorobenzene sorbed to these fractions showed an intense mobile sorption domain that would be expected from partitioning-type interactions.
(1) Stevenson, F. J. In Cycles of Soil: John Wiley & Sons: 1986; p 346. (2) Flaig, W. In Report of FAO/IAEA Technical Meeting; Pergamon: Elmsford, NY, 1966; p 103. (3) Kononova, M. M. Soil Organic Matter; Pergamon: Elmsford, NY, 1966. (4) Engebretson, R.; Amos, T.; von Wandruszka, R. Environ. Sci. Technol. 1996, 30, 390. (5) Schnitzer, M.; Schulten, H. R. Adv. Agron. 1995, 55, 167. (6) Schulten, H. R.; Schnitzer, M. Naturwissenschaften 1993, 80, 29. (7) Swift, R. S. In Humic Substances in Soil, Sediment and Water; Aiken, G. R., McKnight, D. M., Wershaw, R. L., McCarthy, P., Eds.; Wiley: New York, 1985; p 387. (8) Stevenson, F. J. In Wiley: New York, 1994; p 285. (9) Burba, P.; Aster, B.; Nifant’eva, T.; Shkinev, V.; Spivakov, B. Ya. Talanta 1998, 45, 977. (10) Nifant’eva, T. I.; Shkinev, V. M.; Spivakov, B. Y.; Burba, P. Talanta 1999, 48, 257. (11) Tomba´cz, E.; Szekeres, M.; Klumpp, E.; Narres, H.-D. In Understanding and managing organic matter in soils, sediments, and waters; Swift, R. S., KMS, Ed.; IHSS: St. Paul, MN, 2001; pp 61-68. (12) Tomba´cz, E. Soil Sci. 1999, 164, 814. (13) Silva, S.; Zanetti, B.; Burzoni, E.; Dell’Agnola, G.; Nardi, S. Agrochim. 1981, 25, 131. (14) Summers, R. S.; Cornel, P. K.; Roberts, P. V. Sci. Total Environ. 1987, 62, 27. VOL. 37, NO. 13, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2859
(15) Swift, R. S.; Leonard, R. L.; Newman, R. H.; Theng, B. K. G. Sci. Total Environ. 1992, 117/118, 53. (16) Lakshman, S.; Mills, R.; Patterson, H. Anal. Chim. Acta 1993, 282, 101. (17) Lobartini, J. C.; Orioli, G. A.; Tan, K. H. Commun. Soil Sci. Plant Anal. 1997, 28, 787. (18) Cristl, I.; Knicker, H.; Kogel-Knabner, I.; Kretzschmar, R. Eur. J. Soil Sci. 2000, 51, 617. (19) Davis, J. A.; Gloor, R. Environ. Sci. Technol. 1981, 15, 1223. (20) Calderbank, A. Rev. Environ. Contamin. Toxicol. 1989, 108, 71. (21) Keoleian, G. A.; Curl, R. L. In ACS Symposium Series 219; Suffel, I. H., MacCarthy, P., Eds.; 1989; p 231. (22) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. In Environmental Organic Chemistry; John Wiley: New York, 1993; p 255. (23) Chiou, C. T. Theoretical considerations of the partition uptake of nonionic organic compounds by soil organic matter. In Reactions and Movement of Organic Chemicals in Soils; Sawhney, B. L., Brown, K., Eds.; Soil Science Society of America, American Society of Agronomy: Madison, WI, 1989; p 1. (24) Karickhoff, S. W. J. Hydraul. Eng. 1984, 110, 707. (25) Schlautman, M. A.; Morgan, J. J. Environ. Sci. Technol. 1993, 27, 961. (26) Flory, P. J. J. Chem. Phys. 1942, 19, 51. (27) Xing, B.; Pignatello, J. J.; Gigliotti, B. Environ. Sci. Technol. 1996, 30, 2432. (28) Karickhoff, S. W.; Brown, D. S.; Scott, T. A. Water Res. 1979, 13, 241. (29) Rutherford, D. W.; Chiou, C. T.; Kile, D. E. Environ. Sci. Technol. 1992, 26, 336. (30) Chiou, C. T.; Peters, L. J.; Freed, V. H. Science 1979, 206, 831. (31) McGinley, P. M.; Katz, L. E.; Weber, W. J., Jr. Environ. Sci. Technol. 1993, 27, 1524. (32) Xing, B.; Pignatello, J. J. Environ. Toxicol. Chem. 1996, 15, 1282. (33) Xing, B.; Pignatello, J. J. Environ. Sci. Technol. 1997, 31, 792. (34) Kohl, S. D.; Rice, J. A. Org. Geochem. 1999, 30, 929. (35) Young, T. M.; Weber, W. J., Jr. Environ. Sci. Technol. 1995, 29, 92. (36) Leboeuf, E. J.; Weber, W. J., Jr. Environ. Sci. Technol. 1997, 31, 1697. (37) Huang, W.; Young, T. M.; Schlautman, M. A.; Yu, H.; Weber, W. J., Jr. Environ. Sci. Technol. 1997, 31, 1703. (38) Pignatello, J. J. In Organic Substances and Sediments in Water: Lewis Publishers: Chelsea, MI, 1991; p 291. (39) Kan, A. T.; Fu, G.; Tomson, M. B. Environ. Sci. Technol. 1994, 28, 859. (40) Aochi, Y. O.; Farmer, W. J. Environ. Sci. Technol. 1997, 31, 2520. (41) Cornelissen, G.; Van Noort, P. C. M.; Govers, H. A. J. Environ. Sci. Technol. 1998, 32, 3124. (42) Weber, W. J. J.; Huang, W.; Yu, H. J. Contam. Hydrol. 1998, 31, 149. (43) Brusseau, M. L.; E., J. R.; Rao, P.; Suresh, C. Environ. Sci. Technol. 1991, 25, 134. (44) Huang, W.; Schlautman, M. A.; Weber, W. J., Jr. Environ. Sci. Technol. 1996, 30, 2993. (45) Pignatello, J. J.; Xing, B. Environ. Sci. Technol. 1996, 30, 1. (46) Weber, W. J., Jr.; McGinley, P. M.; Katz, L. E. Environ. Sci. Technol. 1992, 26, 1955.
2860
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 13, 2003
(47) Weber, W. J., Jr.; Huang, W. Environ. Sci. Technol. 1996, 30, 881. (48) Toscano, P. J.; Waechter, J.; Schermerhorn, E. J.; Zhou, P.; Frisch, H. L. J. Polym. Sci.: Part A 1993, 31, 859. (49) Bandis, A.; Cauley, B. J.; Inglefield, C. E.; Wen, W.-Y.; Ingelfield, P. T.; Jones, A. A., et al. J. Polym. Sci. Part B 1993, 31, 447. (50) Clauss, J.; Schmidt-Rohr, K.; Adam, A.; Beoffel, C.; Spiess, H. W. Macromolecules 1992, 25, 5208. (51) Immergut, E. H.; Mark, H. F. In Plasticization and Plasticizer Processes; Gould, R. F., Ed.; Washington, DC, 1964; p 1. (52) Darby, J. R.; Sears, J. K. In Encyclopedia of Polymer Science and Technology; Mark, H. F., Gaylord, N. G., Bikales, N. M., Eds.; Interscience: New York, 1969; p 228. (53) Hill, R. O. J. In Encyclopedia of Polymer Science and Technology; Mark, J. F., Gaylord, N. G., Bikales, N. M., Eds.; Interscience: New York, 1965; p 354. (54) Pignatello, J. J. In Reactions and Movements of Organic Chemicals in Soils; Sawhney, B. L., Brown, K, Eds.; Soil Science Society of America: Madison, WI, 1989; Chapter 3. (55) Weber, W. J. J.; McGinley, P. M.; Katz, L. E. Environ. Sci. Technol. 1992, 26, 1955. (56) Graber, E. R.; Borisover, M. D. Environ. Sci. Technol. 1998, 32, 3286. (57) Luthy, R. G.; Aiken, G. R.; Brusseau, M. L.; Cunningham, S. D.; Gschwend, P. M.; Pignatello, J. J., et al. Environ. Sci. Technol. 1997, 31, 1, 3341. (58) Kohl, S. D.; Toscano, P. J.; Hou, W.; Rice, J. A. Environ. Sci. Technol. 2000, 34, 204. (59) In Handbook of Physical Properties of Organic Chemicals; Howard, P. H., Meylan, W, M., Eds.; Lewis Publishers: New York, 1997; p 322. (60) Mittelstaedt, W.; Fuhr, F. Field lysimeter studies with 14-C pyridate; Report No.: Report IRA 11/91 E, Study-Nr. Al 2/89; Research Center Julich: 1991. (61) Swift, R. S. In Methods of Soil Analysis: Part 3, Chemical Methods; Sparks, D. L., Bartels, J. M., Bigham, J. M., Eds.; Soil Science Society of America: Madison, WI, 1996; p 1018. (62) Droppo, I.; Krishnappan, B. G.; Rao, S. S.; Ongley, E. D. Environ. Sci. Technol. 1995, 29, 546. (63) Shin, H.-S.; Monsallier, J. M.; Choppin, G. R. Talanta 1999, 50, 641. (64) Peersen, O. B.; Wu, X.; Kustanovich, I.; Smith, S. O. J. Magn. Reson. 1993, 104. (65) Fru ¨ nd, R.; Lu ¨ demann, H.-D.; Gonza´lez-Vila, F. J.; Almendros, G.; del Rı´o, J. C.; Martı´n, F. Sci. Tot. Environ. 1989, 81/82, 187. (66) Wilson, M. A. NMR Techniques and Applications in Geochemistry and Soil Chemistry; Pergamon Press: Oxford, 1987; pp 182216. (67) Emsley, J. W.; Phillips, L. Prog. Nucl. Magn. Reson. Spectrosc. 1971, 7, 1. (68) Toscano, P. J.; Frisch, H. L. J. Polym. Sci. 1991, 29, 1219.
Received for review March 11, 2002. Revised manuscript received March 25, 2003. Accepted April 7, 2003. ES0206386
