Correlation of Poly(methylene)-Rich Amorphous Aliphatic Domains in Humic Substances with Sorption of a Nonpolar Organic Contaminant, Phenanthrene J.-D. MAO, † L. S. HUNDAL,* ,‡ M. L. THOMPSON, ‡ AND K. SCHMIDT-ROHR* ,† Department of Chemistry and Department of Agronomy, Iowa State University, Ames, Iowa 50011 The structural makeup of natural organic matter plays a major role in regulating its capacity to retain nonionic hydrophobic organic compounds (HOCs). We used a model HOCsphenanthrenesto investigate the correlations between sorption capacity, specifically the modified Freundlich coefficient (K′ f ), and compositional data of humic acids, humins, and a peat obtained from quantitative 13 C solid-state NMR spectroscopy. A positive correlation between K′ f and the weight fraction of amorphous poly- (methylene) in the sorbents was observed. In contrast, the correlation between phenanthrene sorption capacity and aromaticity or polarity indices of the sorbents was insignificant.The nonpolar aliphatic carbon fraction,excluding poly(methylene), was only partially correlated with K′ f . Detailed NM R analyses of the sorbents using 1 H inversion- recovery experiments showed that 10-nm diameter domains of branched nonpolar aliphatic groups, which account for 20-50% of all nonpolar aliphatic segments and may be associated with the poly(methylene), were responsible for the partial correlation. The correlation between K′ f and the amorphous nonpolar aliphatic domains including amorphous poly(methylene) was strong. The rubbery, relatively low-density, and amorphous nonpolar aliphatic domains can be expected to offer an excellent environment for the sorption of phenanthrene by partitioning. These observations suggest that the domains of amorphous poly(methylene) and branched nonpolar aliphatics, which make up 2-9 wt % of the organic fraction in our samples, may serve as good descriptors for the potential of natural organic matter to retain HOCs in the natural environments. Introduction Retention of nonionic hydrophobic organic compounds (HOCs) by soil materials governs their mobility and bio- availability. It has been demonstrated by many researchers that the HOC retention capacity of soils, sediments, and aquifermaterialsisdominantlycontrolled bythe totalorganic matter content, especially if it is higher than trace levels (1-3). It has also been well documented that the capacity and affinity of the natural organic matter (NOM) to retain aqueous-phase HOCs vary significantly among samples of different origins (4-8). It is of significant interest to gain an understanding of how the chemical structure of NOM controls the fate and behavior of HOCs in the natural environment. It has been reported that sorption capacity is negatively correlated with certain polarityindices ofNOM derived from elementalratios (9, 10). Recently, Kile et al. (8)used an index of the polar functional groups (sum of carbohydrate and carboxyl -amide -ester carbons) in NOM to explain differ- ences in the carbon tetrachloride sorption capacity and affinityof19soilsand nine freshwatersediments.Theyfound that partition coefficients were poorly correlated with the polarityindexand concluded that aromaticityor aliphaticity of NOM was a poor predictor of HOC sorption capacity. In a previous study, Chiou et al. (5) had suggested enhanced HOC sorption capacity for sorbents with NOM containing a higher fraction of aromatic carbon. A positive correlation between HOC sorption capacity and aromaticity of humic and fulvic acids has also been reported by Chin et al. (11) and Perminova et al. (7). In contrast, Chefetz et al. (6) concluded that the aliphatic character of NOM could contribute significantly to sorption ofpyrene. These studies clearlyemphasize the importance ofidentifyingthe structural components of NOM that determine the HOC sorption capacity and affinity of soils and sediments. But there is a little consensus about which domains of NOM are most important for retention ofHOCs in the naturalenvironment. We recently discovered semicrystalline poly(methylene) domains in NOM samples from various sources (12). These domains exhibited several characteristics that are similar to synthetic polyethylenes. Relatively rigid crystalline regions of 3-nm lamellar thickness, with melting points around 75 °C, were found adjacent to amorphous regions with rubber- like segmental mobility (12). The overall domains of sem- icrystalline poly(methylene) in humin were more than 10 nm in diameter. The hydrophobicity and relatively low den- sity of the amorphous poly(methylene) regions are similar to those of alkane solvents. Therefore, we hypothesize that thepoly(methylene)domainsmayofferan idealenvironment for retention of HOCs in NOM. To test this hypothesis, we investigated correlations between the modified Freundlich coefficient (K′f)ofamodel HOCsphenanthrene sand amorphous poly(methylene) con- tents of various humins, humic acids, and a peat sample, determined by quantitative magic-angle spinning (MAS), solid-state 13 C NMRspectroscopy. For comparison, we have also tested the correlation of K′f with the aromaticity,nonpolar aliphaticityexcludingpoly(methylene),and polarityobtained from quantitative 13 C NMR data. By 1 H inversion -recovery with 13 C NMR detection, we identified domains of poly- (methylene) and of branched nonpolar aliphatic segments and correlated their weight fractions with phenanthrene sorption capacity. The correlations between K′f and various components of NOM quantified by NMR can be used to identify the component that dominantly regulates the HOC sorption capacity of NOM and thus provide a better understanding of the relationship between sorption and structural makeup of NOM. Experimental Section Sorbents. We selected seven sorbents of widely differing compositions: one whole peat,two humins,two peat humic *Corresponding authors fax: (515)294-3163; e-mail: lhundal@ iastate.edu (L.S.Hundal);phone: (515)294-6105;fax: (515)294-0105; e-mail: srohr@iastate.edu (Klaus Schmidt-Rohr). † Department of Chemistry. ‡ Department of Agronomy. Environ. Sci. Technol. 2002, 36, 929-936 10.1021/es011054r CCC: $22.00  2002 American Chemical Society VOL. 36, NO. 5, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 929 Published on Web 01/26/2002