Redox transformation, solid phase speciation and solution dynamics of copper during soil reduction and reoxidation as affected by sulfate availability

Abstract

In periodically flooded soils, interactions of Cu with biogenic sulfide formed during soil reduction lead to the precipitation of sparingly soluble Cu-sulfides. In contaminated soils, however, the amounts of Cu can exceed the amount of sulfate available for microbial reduction to sulfide. In laboratory batch experiments, we incubated a paddy soil spiked to ∼4.4 mmol kg^-1 (280 mg kg⁻¹) Cu(II) to monitor temporal changes in the concentrations of dissolved Cu and the speciation of solid-phase Cu during 40 days of soil reduction and 28 days of reoxidation as a function of initially available reducible sulfate (0.06, 2.09 or 5.92 mmol kg⁻¹). Using Cu K-edge EXAFS spectroscopy, we found that a large fraction of Cu(II) became rapidly reduced to Cu(I) (23–39%) and Cu(0) (7–17%) before the onset of sulfate reduction. Combination with results from sequential Cu extraction and chromium reducible sulfur (CRS) data suggested that complexation of Cu(I) by reduced organic S groups (S_org) may be an important process during this early stage. In sulfate-depleted soil, Cu(0) and Cu(I)–S_org remained the dominant species over the entire reduction period, whereas in soils with sufficient sulfate, initially formed Cu(0) and (remaining) Cu(II) became transformed into Cu-sulfide during continuing sulfate reduction. The formation of Cu(0), Cu(I)–S_org, and Cu-sulfide led to an effective decrease in dissolved Cu concentrations. Differences in Cu speciation at the end of soil reduction however affected the dynamics of Cu during reoxidation. Whereas Cu(0) was rapidly oxidized to Cu(II), more than half of the S-coordinated Cu fraction persisted over 14 days of aeration. Our results show that precipitation of Cu(0) and complexation of Cu(I) by reduced organic S groups are important processes in periodically flooded soils if sulfide formation is limited by the amount of available sulfate or the duration of soil flooding. The speciation changes of Cu described in this study may also affect the speciation and solubility of other chalcophile metals in redox-dynamic wetland soils.

Introduction

Copper (Cu) is an essential trace element for most living organisms. It is a constituent of a variety of proteins and enzymes that carry out fundamental biological functions (Rubino and Franz, 2012). However, elevated concentrations of Cu are toxic for microorganisms (Baath, 1989) and plants (Fernandes and Henriques, 1991). Ecosystem functioning can therefore be adversely affected as a result of soil contamination with Cu from anthropogenic activities such as metal mining or untreated waste water discharge.

While the behavior of Cu in well-drained oxic soils has been extensively studied, its biogeochemistry in periodically flooded soils, such as riparian wetlands and rice paddy soils, is less well-understood. In oxic soils, Cu dominantly occurs as bivalent (cupric) Cu(II). Cu(II) forms stable complexes with natural organic matter (Karlsson et al., 2006, Manceau and Matynia, 2010) and adsorbs on surfaces of Mn(III, IV) and Fe(III) (oxyhydr)oxides and clay minerals (Parkman et al., 1999, Strawn et al., 2004). However, Cu is a redox-active element and it may be reduced from cupric Cu(II) to cupreous Cu(I) or even Cu(0) by abiotic or biotic processes. Redox changes resulting from periodic overbank flooding and variations in groundwater level (Kirk, 2004) or flood irrigation (Koegel-Knabner et al., 2010) are therefore expected to strongly influence the chemical speciation and solubility of Cu in soils (Du Laing et al., 2009, Borch et al., 2010). Under sulfate-reducing conditions, the reduction of Cu(II) by dissolved sulfide (Luther et al., 2002, Ciglenecki et al., 2005) and subsequent precipitation as sparingly soluble metal sulfides (Pattrick et al., 1997, Morse and Luther, 1999) may substantially decrease Cu solubility. Recent studies on trace metal dynamics in a contaminated river floodplain soil showed that even at low levels of available sulfide and in the presence of other chalcophile metals (e.g., Hg, Cd, Pb and Fe), Cu can be effectively sequestered as Cu sulfide, due to the high thermodynamic stability of Cu sulfide minerals (Weber et al., 2009a, Hofacker et al., 2013).

In general, total S contents in freshwater floodplain soils are considerably lower than in marine or brackish wetlands and comprise only a small fraction (<5%) of inorganic S (Howarth et al., 1992). Riparian floodplain soils are aerated during most of the year, which prevents the accumulation of reduced inorganic S. Sulfate, on the other hand, is only weakly retained due to weak adsorption at pH > 6 (Tabatabai, 2005), and the mineralization of organic S is too slow to maintain high sulfate contents (Tabatabai and Alkhafaji, 1980, Zhou et al., 2005). As a result, the concentrations of chalcophile trace metals in contaminated freshwater floodplain soils may often exceed the amounts of reducible sulfate, resulting in the competition of different metal cations for reaction with limited amounts of sulfide formed by microbial sulfate reduction during periodic soil flooding (Weber et al., 2009a). Copper is expected to play a key role under sulfate-limited conditions, because it forms much more stable metal sulfides than Fe or trace metals like Pb, Cd, or Zn and typically occurs at much higher concentrations than Hg or Ag that would form even less soluble sulfides than Cu. However, in addition to the thermodynamic stability of metal sulfides, the solubility of other metal bearing phases and the kinetics of metal desorption, diffusion, and precipitation are also expected to strongly influence the formation of metal sulfide minerals (Weber et al., 2009a).

At low sulfide levels, Cu(II) may also be reduced by microorganisms (Wakatsuki, 1995), by reaction with dissolved Fe(II) (Matocha et al., 2005), or by redox-active functional groups of natural organic matter as recently shown (Pham et al., 2012). Biotic or abiotic reduction of Cu(II) to Cu(I) in sulfate-limited environments may have two major effects: the soft metal cation Cu⁺ may form strong complexes with thiol-groups of natural organic matter (R-SH), as reported for other soft metal cations like Ag⁺, Cd²⁺ and Hg²⁺ (Smith et al., 2002, Karlsson et al., 2005, Skyllberg et al., 2006). Association of Cu(I) with thiol-containing ligands has long been known to play a role in marine environments (Boulegue et al., 1982, Leal and van den Berg, 1998), but has hardly been considered for terrestrial soils. Another consequence of Cu reduction in sulfate-limited environments may be the formation of metallic Cu(0), since Cu(I) can rapidly disproportionate to Cu(II) and Cu(0) if the concentrations of Cu(I)-stabilizing ligands (e.g., sulfide, chloride and thiol) are too low. Metallic Cu has been observed in and near roots of wetland plants (Manceau et al., 2008) and in organic-rich bog soils (Lovering, 1927, Lett and Fletcher, 1980). Recently, Cu(0) formation has also been observed in microcosm flooding experiments with a contaminated floodplain soil (Weber et al., 2009b, Hofacker et al., 2013). In these studies Cu(0) was formed as intermediate phase, which was completely sulfidized during subsequent sulfate reduction. Under sulfate-limited conditions, however, Cu(0) may accumulate in reduced soils.

Variations in Cu speciation changes during soil reduction may also affect Cu dynamics during subsequent soil reoxidation, because different reduced Cu species may vary with respect to their reactivity under oxic conditions. While Cu sulfide clusters and Cu(I)-thiol complexes were shown to be stable over days to weeks in oxic waters (Leal and van den Berg, 1998, Rozan et al., 2000, Laglera and van den Berg, 2003, Sukola et al., 2005), Cu(0) oxidation occurs within tens of minutes (Kanninen et al., 2008).

Knowledge on the behavior of Cu during soil reduction and reoxidation at low sulfate availability is very limited (Weber et al., 2009a), and studies on Cu mobilization during transition from anoxic to oxic conditions were mainly conducted with sulfide-rich sediments and soils, focusing on the oxidative dissolution of Cu sulfides (e.g., Simpson et al., 1998, Carroll et al., 2002, Caetano et al., 2003). Thus, the aim of the present study was to elucidate the effect of sulfate availability on Cu dynamics in solution and solid phases during soil reduction and subsequent reoxidation. We conducted a time-resolved laboratory incubation experiment, in which we adjusted an uncontaminated paddy soil from Bangladesh to various sulfate-to-Cu ratios prior to an anoxic incubation period of 40 days, followed by reoxidation for 28 days. The dynamics of Cu in the solution phase and changes in Cu solid phase speciation were investigated by Cu K-edge X-ray absorption spectroscopy (XAS), sequential metal extractions, and chromium-reducible sulfur (CRS) distillation.