Reduction of Polychlorinated Ethanes and Carbon Tetrachloride by

May 4, 2009 - (b) δ13C-values of HCA (black circles) and PCE (red squares) in reduced smectite suspension at pH 8.5. Solid lines were calculated usin...
0 downloads 0 Views 1MB Size
Environ. Sci. Technol. 2009, 43, 4082–4089

Reduction of Polychlorinated Ethanes and Carbon Tetrachloride by Structural Fe(II) in Smectites ANKE NEUMANN, THOMAS B. HOFSTETTER,* MARITA SKARPELI-LIATI, AND RENÉ P. SCHWARZENBACH Institute of Biogeochemistry and Pollutant Dynamics (IBP), ETH Zurich, Switzerland

Received January 20, 2009. Revised manuscript received April 5, 2009. Accepted April 7, 2009.

Ferrous iron associated with clay minerals can be important for the reductive transformation of organic contaminants in anoxic soils and groundwaters. We investigated the reactivity of structural Fe(II) in ferruginous smectite for the reduction of a series of polychlorinated alkanes (hexa-, penta-, 1,1,1,2- and 1,1,2,2tetrachloroethane, and carbon tetrachloride (CCl4)) in laboratory batch reactors. Evaluation of reaction kinetics, product distribution, and C-isotope fractionation suggest that polychlorinated ethanes containing three R-Cl atoms reacted via reductive β-elimination to the corresponding ethenes while CCl4reduction leads predominantly to the formation of chloroform. Reduction kinetics followed a typical biphasic behavior characteristic of the presence of two types of Fe(II) species exhibiting different reactivity in the octahedral sheet of smectites, and reaction rate constants were pH-independent. Dehydrochlorination reactions of chloroethanes containing at least one β-H atom were found to compete with or even dominate over the reduction reaction with increasing suspension pH. Reference experiments in homogeneous solution and with nonreduced smectite performed in the pH range of 5.5-8.5 suggest that the HCl-elimination is not catalyzed at mineral surfaces. From the observed slow transformation of chloroalkanes, we hypothesize that structural Fe(II) in smectites will be important mainly as a reductant in the subsurface once iron(hydr)oxides have been reductively dissolved.

Introduction Ferrous iron species associated with clay minerals can be of great importance for the reductive transformation of pollutants in the anoxic subsurface. Clay minerals are ubiquitously present in soils and sediments and most clay minerals contain some iron which can be available as reactive Fe(II) upon microbial reduction (1). Moreover, due to the lower susceptibility of structural Fe(III) in clay minerals for reductive dissolution as compared to Fe(III)(oxy)hydroxides (2, 3), iron in clay minerals provides a renewable source of reduction equivalents for contaminant transformation (4). The few studies in which the reduction of organic and inorganic contaminants by Fe(II) associated with clay minerals was investigated found that structural Fe(II) species in the octahedral sheet of smectites were the predominant reductants (5-9). Fe(II) surface hydroxyl complexes, in contrast, did not contribute significantly to the overall reactivity. * Corresponding author e-mail: [email protected]. 4082

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 11, 2009

We recently proposed a conceptual model that includes the presence of two different types of structural Fe(II) sites for the reduction kinetics of nitroaromatic compounds (NACs) to the corresponding aromatic amines in suspensions of smectites with structural Fe content varying from 3 to 13 wt% (8). Smectites containing high amounts of structural Fe (>12 wt%) enabled a fast initial reduction of the NACs followed by a much slower transformation, which is characteristic for the presence of two distinct types of reactive Fe(II) sites. NAC reduction rate constants for the two sites differed by 3 orders of magnitude regardless of the contaminant’s intrinsic reactivity and time scale of the experiment. In case of clay minerals with low structural Fe content, the formation of such reactive octahedral Fe(II) arrangements is impaired and slower NAC reduction following pseudo-first order behavior was observed. Whether the same types of structural Fe(II) species in smectites are accessible to other classes of important reducible organic groundwater contaminants such as polychlorinated alkanes and whether similar reactivity trends for structural Fe(II) species exist is currently unclear. For example, the (eco-)toxic solvents 1,1,2,2-tetrachloroethane, 1,1,1-trichloroethane, and other polychlorinated ethanes can be reduced by Fe(II) species bound to Fe(III)(oxy)hydroxides (10, 11). However, in suspensions of reduced, Fe-bearing smectite, the experimental findings were not conclusive. Reduction was not observed for penta- and hexachlorinated ethanes. Instead, transformation of contaminants containing at least one H atom was proposed to occur predominantly as a surface-catalyzed, non-reductive dehydrochlorination (β-elimination, (12)). Structural Fe(II) contents of the smectite exceeded 10 wt%, which would have favored a reductive transformation. Hence, it remains unclear whether the same types of Fe(II) species found responsible for NAC reduction (8) facilitate the transformation of other organic compounds and how the reactivity of structural Fe(II) in smectites compares for different contaminant classes. The objective of this study was to identify the factors governing the reductive transformation of polychlorinated aliphatic hydrocarbons by structural Fe(II) in smectites. To this end, we investigated whether a series of chlorinated ethanes (hexa- and pentachloroethane as well as the 1,1,1,2and 1,1,2,2-tetrachloroethane isomers) and carbon tetrachloride (CCl4) react with the same types of structural Fe(II) species of ferruginous smectite (SWa-1) that were identified previously (8). In suspensions of chemically reduced ferruginous smectite, we examined whether the reduction kinetics of chlorinated ethanes can be assessed with a recently proposed model for Fe(II) site reactivity in iron-rich smectites (8). To account for the competing non-reductive, basecatalyzed dehydrochlorination of chlorinated ethanes containing at least one H atom, reference experiments were conducted at different pH values in the presence and absence of non-reduced smectite. The identity of the reductive (reductive R- or β-elimination, hydrogenolysis) and nonreductive transformation pathways was examined via analysis of the reaction rates and product evolution and via evaluation of changes in C isotope signatures in reactant and products.

Materials and Methods Kinetic Experiments. For a complete list of chemicals used in this study see Supporting Information (SI). Suspensions containing reduced ferruginous smectite (SWa-1) were prepared under anoxic conditions as described previously (6). Briefly, suspensions of clay mineral were reduced according to the citrate-bicarbonate-dithionite method (13) 10.1021/es9001967 CCC: $40.75

 2009 American Chemical Society

Published on Web 05/04/2009

followed by cation exchange with 1 M NaCl. Homoionized smectite stock suspensions were diluted to yield 7.8 g L-1 of reduced mineral using anoxic solutions of 10 mM MES (2morpholino-ethane-sulfonic acid), PIPES (piperazine-1,4bis(2-ethanesulfonic acid)), MOPS (morpholino-propanesulfonic acid), or TAPS (N-[tris(hydroxymethyl)methyl]-3aminopropane sulfonic acid) buffered at pH 5.5 ( 0.1, 6.5 ( 0.1, 7.5 ( 0.1, and 8.5 ( 0.1, respectively. Reactors contained 20 mL of smectite suspension and were kept in the dark on a roller apparatus at 25 ( 1 °C during the experiments. Reactors containing non-reduced ferruginous smectite for the evaluation of chloroalkane adsorption to smectite surfaces and of smectite surfaces for dehydrochlorination were prepared following the same procedure but without the addition of dithionite. Homogeneous reference reactors contained only buffer and 0.05 M NaCl solution. All experiments were carried out in duplicate. Kinetic experiments were initiated by the addition of methanolic stock solution of the chlorinated ethane or carbon tetrachloride (CCl4) to the reactors yielding initial aqueous concentrations of 20 µM (final methanol content