Environ. Sci. Technol. 2002, 36, 4052-4057
Sorption Behavior of Nonylphenol in Terrestrial Soils ROLF-ALEXANDER DU ¨ RING,* S E B A S T I A N K R A H E , A N D S T E F A N G A¨ T H Department of Agricultural Ecology and Natural Resources Management, Division of Waste Management and Environmental Research, Justus-Liebig-University, Heinrich-Buff-Ring 26-32, D-35392 Giessen, Germany
Nonylphenol (NP) as an intermediate from anaerobic degradation of widely used nonionic surfactants occurs widespread in the environment. Partition behavior of this toxic and endocrine-disrupting chemical between soil and water was not examined until yet. The objective of this investigation was to quantify sorption and desorption behavior of 4-nonyl[14C]phenol in a set of 51 soils using the batch equilibrium approach. Kinetic studies indicated apparent equilibrium within 20 h. Sorption was influenced by sorbate structure as could be shown with branched 4-nonyl[14C]phenol and the linear 4-n-NP, respectively. Linear 4-n-NP behaves differently from the branched isomers of 4-NP. Sorption of 4-nonyl[14C]phenol tested with five different initial concentrations resulted in linearly fitted isotherms that provided calculation of sorption partition coefficients (KP). Desorption partition coefficients (KP-des) revealed hysteresis independent of soil properties but decreasing with decreasing initial NP concentrations. KP values were correlated with organic carbon content of the soils yielding a log KOC of 3.97.
Sorptive and desorptive behavior of NP should be quantified since its enrichment in sediments and sewage sludge provides pools to release NP into the environment (17). Furthermore, uptake of NP by plants from sludge-amended soil depends on its bioavailability, which is determined by sorptive behavior. NP is known to be lipophilic with reported octanol-water partition coefficients (POW) in a range of log POW ) 3.01-4.48 (18, 19). Regarding different log POW - log KOC (organic carbon-normalized partition coefficient) correlations existing for a set of organic pollutants, moderate to high adsorption potential of this compound to the soil matrix can be expected (20). NP behaves as a weak acid with a pKa of 10.7 (21), and ionization under neutral to alkaline conditions probably influences its solubility and sorption to solid matrixes as already determined for chlorinated phenols (22). NP consists of a number of isomers with differently branched nonyl chains. Nonetheless, there are still no published data on distinctive estrogenic activity of each isomer, which can be postulated to be effectuated to a large extent by different branching. Significant estrogenic activity was found for the branched octylphenol and NP isomers whereas no significant induction was seen with the two linear isomers (23). In a work on the environmental fate of NP, a group of NP isomers was assumed to degrade more slowly than other isomers (24). Different log KOC for the linear and branched isomers, respectively, were recently expected (17). Estimated log POW values differing from 4.78 to 5.72 for singular proposed structures of some distinct isomers of 4-NP indicate that bioconcentration of isomers will likely differ (25). The present paper reports the sorption behavior of 4-nonyl[14C]phenol considering adsorption and desorption kinetics at different concentration levels. Some isomerspecific phenomena are also illustrated. Partition coefficients for the ad- and desorption of 4-nonyl[14C]phenol were obtained for a large set of soils.
Introduction
Experimental Section
The environmental occurrence of alkylphenols such as nonylphenol (NP) is proven since the late 1970s (1), and its aquatic toxicity and potential to disrupt the endocrine system was determined several times in the last two decades (2-4). Different environmental media like waters, sediments, sludge, biota, and air were analyzed (5-12). Within wastewater treatment, the degradation of nonylphenolethoxylates (NPEO) results in an emission of NP. NP accumulates mainly in anaerobically digested sewage sludge (5, 13, 14). It is widespread distributed via the application of sludge and, to a smaller extent, pesticide formulations (2) onto agricultural land. Furthermore, NP can be introduced by atmospheric deposition onto soils (7, 15). Partition of NP in the environment is assessed to be more than 60% in sediment, >10% in soil, and approximately 25% in the water phase (16). Recent results on the distribution and behavior of NP in different waters and sediments revealed that approximately 20% of NP was found in the particulate phase (11). However, besides the results from the monitoring of NP in different environmental media, no fundamental laboratory data exist on its sorption behavior in soils or sediments. To assess the environmental fate of any pollutant, the investigation of its sorption/partition behavior is essential.
Sorbents. Soil material was taken from field sites located in different parts of Germany. This pool of soils included mainly agricultural soils exhibiting pH values from 5.2 to 7.8. In addition, some “extreme” soils with lower pH derived from forests were considered (Figure 1). Organic matter content was determined as follows: Carbonate-free ground soil samples were weighed (approximately 20 mg) into tin capsules. The samples were burned and detected with a thermal conductivity detector by a carbon/nitrogen analyzer (Carlo-Erba, NA 1500, Milan, Italy). Soil pH was measured in a 1:2.5 (w/w) mixture of soil with CaCl2 solution (0.01 mol L-1). Sorbates and Chemicals. Because of the relatively low water solubility and high surface affinity of alkylphenols (26, 27) and thus limited suitability of classical analytical techniques, experiments were carried out with radiolabeled, branched 4-NP. The substance was synthesized according to a method of ref 28: A mixture of isomers of nonene was added to phenol, which was spiked with uniformly 14C-labeled phenol (12.2 mCi mmol-1, International Isotopes Munich, Germany) in a ratio of 25:1 under acidic conditions to produce 4-nonyl[14C]phenol. Nonlabeled linear 4-n-NP (Riedel de Hae¨n, Seelze, Germany) was added within this synthesis. The product was purified by preparative normal-phase HPLC (mobile phase: dichloromethane/methanol 95%/5%; column: LiChroSphere, Si60, 10 µm particle size, 20 mm i.d., 250 mm length; MZ-Analysentechnik, Mainz, Germany) to
* Corresponding author telephone: +49 641 9937393; fax: +49 641 9937389; e-mail:
[email protected]. 4052
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 19, 2002
10.1021/es0103389 CCC: $22.00
2002 American Chemical Society Published on Web 08/15/2002
FIGURE 1. Properties of examined soil samples (n ) 51). Corg ) organic carbon.
TABLE 1. Concentrations of 4-Nonylphenol (mg L-1) in 0.01 M CaCl2 Solution Applied within the Sorption Experiments 4-nonyl[14C]phenol
4-nonyl[12C]phenol
4-n-nonyl[12C]phenol
0.15 0.30 0.45 0.60 0.75
3.75 7.5 11.25 15.0 18.75
0.29 0.59 0.88 1.17 1.46
separate remaining phenol and other reaction products. That mixture should provide to assess sorption behavior of the different isomers by means of GC-MS analysis. Specific radioactivity of the product 4-nonyl[14C]phenol was 594 µCi mmol-1. NP solutions were prepared as follows: 1-5 µL of the labeled NP-mixture was dissolved in 5 mL of acetone (Riedel de Hae¨n, Pestanal). This NP solution was filled up to 1000 mL with 0.01 M CaCl2 (Riedel de Hae¨n, p.a.) aqueous solution, resulting in five concentration levels given in Table 1. Octanol-Water Partition Coefficient. To further characterize the test substance, log POW was determined by the shake flask method according to OECD Guideline 107 (29). After the experiments, activity in both of the separated phases was quantified by liquid scintillation counting (see below). Sorption Experiments. Sorption experiments were carried out in conformity to OECD Guideline 106 (30). A total of 3 g of air-dried and sieved (