JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 9 January 2007
Published in J Environ Qual 36:61-69 (2007)
DOI: 10.2134/jeq2006.0148
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schuwirth, N.
Right arrow Articles by Hofmann, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schuwirth, N.
Right arrow Articles by Hofmann, T.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Schuwirth, N.
Right arrow Articles by Hofmann, T.
Related Collections
Right arrow Mining Wastes
Right arrow Heavy Metals
Right arrow Soil Pollution

TECHNICAL REPORTS

Heavy Metals in the Environment

Vertical Distribution and Speciation of Trace Metals in Weathering Flotation Residues of a Zinc/Lead Sulfide Mine

Nele Schuwirtha, Andreas Voegelinb, Ruben Kretzschmarb and Thilo Hofmannc,*

a Institute of Geosciences, Mainz Univ., Becherweg 21, D-55099, Mainz, Germany. Present address: Swiss Federal Institute of Aquatic Science and Technology (Eawag), CH-8600 Duebendorf, Switzerland
b Institute of Biogeochemistry and Pollutant Dynamics, ETH Zurich, Universitätstrasse 16 CHN, CH-8092 Zurich, Switzerland
c Vienna Univ., Center of Earth Sciences, Dep. of Environmental Geoscience, Althanstraße 14, A-1090 Vienna, Austria

* Corresponding author (thilo.hofmann{at}univie.ac.at)

Received for publication April 13, 2006.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Sulfide-bearing mine tailings are a serious environmental problem around the world. In this study, the vertical distribution and speciation of Zn and Pb in the fine-grained flotation residues of a former sulfide ore mine in Germany were investigated to assess the inorganic weathering processes that effect the environmental risk arising from this site. Total metal contents were determined by X-ray fluorescence spectroscopy (XRF). Mobilizable fractions of Zn, Pb, Fe, and Mn were quantified by sequential chemical extractions (SCE). Furthermore, the speciation of Zn was analyzed by Zn K-edge extended X-ray absorption fine structure spectroscopy (EXAFS) to identify the residual Zn species. The variations in pH and inorganic C content show an acidification of the topsoil to pH 5.5. EXAFS results confirm that Zn is mainly bound in sphalerite in the subsoil and weathering reactions lead to a redistribution of Zn in the topsoil. A loss of 35% Zn and S from the topsoil compared with the parent material with 10 g kg–1 Zn and neutral pH has been observed. If acidification proceeds it will lead to a significant release of Zn, S, and Pb to the ground water. In contrast to Zn, Pb is enriched in the mobile fraction of the topsoil by more than a factor of two compared with the subsoil which contains a total of 2 g kg–1 Pb. Thus, the high bioavailability of Pb and the potential for Pb uptake by plants and animals currently represent the most severe threat for environmental health.

Abbreviations: SCE, sequential chemical extractions • EXAFS, extended X-ray adsorption fine structure spectroscopy • LCF, linear combination fit • TOC, total organic carbon • TIC, total inorganic carbon


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
SULFIDE-BEARING mine tailings are a serious environmental problem in many regions worldwide (Dudka and Adriano, 1997; van Geen et al., 1999; Shu et al., 2001). Pyrite (FeS2)-containing mine tailing piles undergoing oxidation are a major source of acid mine drainage (AMD) containing metal contaminants (Impellitteri, 2005). Metal ore processing usually leads to multi-elemental contamination of the environment (Dudka and Adriano, 1997). Typical contaminants occurring in association with Zn and Pb ore mining activities include Al, As, Cd, Cu, and Hg (Bridge, 2004).

The weathering of sulfidic mine tailings, caused by an interaction of oxidation and acidification, often results in the release of sulfate, metals, and As into solution (Mascaro et al., 2001). The released trace elements may subsequently react with organic or mineral surfaces or leach into surface and ground water. In oxidizing environments, Fe oxides are likely to precipitate and it is possible that certain metals are coprecipitated in, or adsorbed onto, these oxides, whereas others are leached out of the soil (Sohlenius and Öborn, 2004). The oxidation of sulfide minerals may cause a decrease in pH, which can also affect the mobility of heavy metals and As (Simpson et al., 1998; Cappuyns and Swennen, 2005).

The critical pH value below which the desorption of Zn becomes relevant is around 5, whereas the desorption of Pb requires pH values 1 to 2 magnitudes lower (Alloway, 1999).

While oxidative surface reactions on Fe sulfide minerals have been extensively studied (Rimstidt and Vaughan, 2003), far fewer studies have focused on the dissolution of other sulfide minerals such as sphalerite (ZnS) (Salomons, 1995; Weisener et al., 2003). The direct oxidation of sphalerite (ZnS + 2O2 -> Zn2+ + SO42–) does not lead to acidification. At low pH (< 3), on the other hand, the oxidative dissolution of sphalerite may be catalyzed by abiotic Fe3+ reduction and subsequent oxidation of Fe2+ by Thiobacillus ferrooxidans (Garcia et al., 1995). If the Fe2+ is not completely reoxidized or if it is leached with the soil water, Fe3+–mediated sphalerite oxidation results in a net acid release.

Soil pH is a result of both acid-producing and acid-neutralizing processes. The dissolution of carbonate minerals such as calcite (CaCO3), dolomite CaMg(CO3)2, or carbonates of Sr or Mn can neutralize AMD. However, only a fraction of the total amount of carbonates may be available for buffering acid input because the precipitation of Fe hydroxide and Ca sulfate can form coatings around the particles and prevent further neutralization (Salomons, 1995). Field and experimental evidence indicates that carbonates dissolve in the order calcite, dolomite, ankerite (Ca(Fe, Mg, Mn)(CO3)2), siderite (FeCO3) (Al et al., 2000). While initial siderite dissolution may exert a neutralizing effect, continued dissolution may produce additional acidity through the oxidation of the released Fe2+ and subsequent Fe oxide precipitation (Paktunc, 1999). The neutralization of acidity by carbonates is a relatively fast process; further acidity may be neutralized by slower reactions of protons with silicate minerals (Salomons, 1995).

In the Devonian sedimentary rocks of the low mountain ranges of Germany, Zn and Pb ore deposits can be found. Mining activities lead to a large number of unprotected small-scale mine tailings impoundments, which result in considerable heavy metal contamination of soil and ground water (Wieber and Knoblich, 1995).

In this work, we studied the fine-grained Zn- and Pb-containing flotation residues from one of these mines. The objective was to gain a more detailed understanding of the inorganic weathering processes that determine the environmental fate of Zn, Pb, and other trace elements in weathering fine-grained flotation residues under field conditions. For this purpose, we analyzed sample cores from the flotation residues for the vertical distribution of major and trace elements, the variations in pH, and inorganic and organic C content. The chemical behavior of Zn, Pb, Fe, and Mn were further characterized by sequential chemical extractions (SCE) and the speciation of Zn by Zn K-edge EXAFS spectroscopy. The results from this study allow for a more detailed assessment of present and future environmental risks arising from the weathering of this type of flotation residue tailings.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Field Site Description
In this study, we investigated a former tailings pond of a Zn/Pb/Ag ore mine in northern Rhineland-Palatinate, Germany, which was operated until 1961. The original ore deposit mainly consisted of sphalerite, silver-bearing galena (PbS), and chalcopyrite (CuFeS2). Pyrite occurred as accessory mineral. Further accessory minerals were siderite, ankerite, dolomite, calcite, and limonite (FeO(OH) nH2O). Gangue was quartz (Hünermann, 1955). To extract the ore, the material was crushed, ground, and subsequently processed in a flotation plant using organic agents such as Flotol, Sapinol, and potassium ethyl xanthate to separate the ore from the spoil based on different surface properties (Slotta, 1983). For deposition of the flotation residues, a 70 m long and 25 m high dam was built (Fig. 1). The tailings were pumped through a pipeline over the dam crest to be deposited by several discharges. With increasing distance to the dam the flow velocity decreased, leading to a distal deposition of the fine grained material. The small creek flowing in the narrow valley was redirected in a brick pipe. The pipe ran underneath the tailings pond and ended below the dam, directing the creek water back in to its former bed (Fig. 1). Outlet discharge structures drained the clarified sewage of the tailings pond directly into the creek canal (Herbst, 1962). The flotation dump was used from 1946 until the mining activities were stopped in 1961. The largest part of the former tailings pond was later on covered with coarse detritus. Only the rear area was still directly accessible, which was covered with different grasses and rushes. In this part, the thickness of the fine grained flotation residues ranged from 3 to 7 m. The residues showed high water contents and water logging could be observed at the surface.


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 1. Schematic sketch of the deposition of the flotation tailings (not in scale) with the position of the soil core P3.

 
Soil Sample Collection and Characterization
A soil core (P3) with an 80-mm diameter and 5-m depth was collected in the rear part of the flotation dump (Fig. 1). The core was sectioned into samples with 30-cm length in the field. The core samples were wrapped firmly in plastic and transferred into a glove box under oxygen-free conditions (N2/H2 95:5 v/v%) within 4 h to preserve redox conditions. Additional samples were taken at different spatial resolution.

The total metal contents of the soil samples were analyzed by energy-dispersive X-ray fluorescence spectroscopy (Spectro, X-Lab 2000, Kleve, Germany, 32-mm pressed wax-pellets) and the total C concentration was measured using a CHNS elemental analyzer (LECO, CHNS-932, St. Joseph, MI). The carbonate contents were determined by extracting 0.5 to 1.0 g of soil in 50 mL of boiling 1 M H2SO4 and CO2 trapping using NaOH beads. The entrapped CO2 was quantified gravimetrically using an analytical balance (AT 261 Delta Range, Mettler-Toledo, Greifensee, Switzerland). Organic C was calculated as the difference between total and inorganic C. X-ray diffraction (XRD) spectra were recorded from powdered soil samples (Seifert, XRD 3000 TT, Ahrensburg, Germany, Cu-K{alpha}-radiation, 3 to 100°2{theta}, 0.02° steps). The soil pH was determined at a solid/solution ratio of 2 g/5 mL in 0.01 M CaCl2 solution using a glass electrode.

For the selection of samples to be investigated for metal speciation and fractionation, the Ti and Zr contents were used as reference, because Ti and Zr are elements which are resistant to weathering reactions. Titanium concentration of the fine-grained gray flotation material averaged 0.1 mol kg–1 (Fig. 2). Zirconium averaged 1.5 mmol kg–1 and showed the same distribution pattern as Ti (not shown). Only soil samples from the first centimeters and rust colored, sandy soil samples (e.g., P3–0.34), which appeared within the first 50 cm as intermediate layers, showed lower Ti and Zr concentrations. This may have been due to higher concentrations of quartz sand since Si concentrations were also elevated in these particular samples (Fig. 2). For further investigations, only samples from the fine-grained gray material with similar Ti and Zr concentrations were chosen for SCE to improve the comparability and to avoid misinterpretations due to different original compositions.


Figure 2
View larger version (30K):
[in this window]
[in a new window]

  		Fig. 2. Soil pH (0.01 M CaCl2), total inorganic carbon (TIC), Ca, Mg, Zn, Mn, Cd, S, Pb, total organic carbon (TOC), Fe, and Ti contents in mmol kg–1 dry mass of the soil core P3. Samples additionally investigated by SCE and Zn K-edge extended X-ray adsorption fine structure (EXAFS) spectroscopy are shown in gray symbols.

 
Sequential Chemical Extractions
To characterize the extractability of Zn, Pb, Fe, and Mn, a five-step SCE was performed on three samples from the soil profile. The homogenized sample from 13- to 16-cm depth (P3–0.16) was chosen to represent the topsoil. The sample from 60- to 70-cm depth (P3–0.7) was selected as a sample from the transition zone where the soil pH had already decreased, and a sample from 230- to 300-cm depth (P3–3.0) was chosen to represent the anoxic subsoil. The same samples were also analyzed by XRD for the determination of the bulk mineralogical composition. The speciation of Zn in these three samples and two additional samples from the topsoil was further investigated by Zn K-edge EXAFS spectroscopy. The SCE were conducted in a glove box under nitrogen atmosphere using oxygen-free extractant solutions to preserve redox conditions. Agitation and centrifugation were conducted outside the glove box in air-tight centrifugation vessels. The SCE was started with six replicates. After each extraction step, one subsample was withdrawn for EXAFS analysis, resulting in triplicates for the last two steps. All chemicals used were of analytical grade. Deionized water was produced with the water purification system Arium 611 DI (Sartorius). All materials in contact with the samples were precleaned with 0.14 M HNO3 solution for at least 24 h and subsequently rinsed two times with demineralized and once with deionized water. Blank samples were treated in the same way as the samples for the whole SCE process.

Field moist soil corresponding to 8 g dry mass was initially weighed into 400 mL air-tight polyethylene (PE) centrifuge tubes and subsequently subjected to five extraction steps. Since we wanted to avoid the usage of oxalate due to a potential complexation of Pb ions (Calmano et al., 2001) the following procedure was chosen (extraction parameters and references given in parentheses):

Step (1) Extraction with 200 mL 1 M NH4NO3 (24 h, 20°C, [Zeien and Brümmer, 1989]).
Step (2) Extraction with 200 mL 1 M NH4OAc (24 h, 20°C, pH 6, [Zeien and Brümmer, 1989]).
Step (3) Extraction with 200 mL 0.04 M NH2OH*HCl in 25% HOAc (6 h, 96 ± 3°C, pH 3, [Tessier et al., 1979]).
Step (4) Extraction with 0.025 M NH4–EDTA (1.5 h, 20°C, pH 4.6 [Zeien and Brümmer, 1989]).
Step (5) In this final step, the residual fraction was determined by total digestion. For this purpose, the extracted samples were dried and 200 mg of each sample were added to 4 mL of 48% HF, 2 mL of 65% HNO3, and 4 mL of approximately 72% HClO4 in a Teflon beaker and heated in 10 h to 150°C. This temperature was kept constant for 2 h. After cooling, the mixture was fumed at 180°C in a closed system for 12 h. This residue was redissolved by heating with 2 mL HCl, 0.6 mL HNO3, and 10 mL deionized water to 150°C for 2 h. After cooling to room temperature, the solution was diluted to 100 mL and transferred into polyethylene flasks. No visible residues were left after this procedure. Five standard samples were processed in an analogous manner for quality control. All acids used were subboiled. Trace and major ions were analyzed by ICP–OES (Fisons Instruments, Maxim I, Manchester, UK). Analyses of blank extractions were below the analytical detection limit for all extraction steps.

After each extraction (steps 1 through 4), the solutions were centrifuged (15 min, 770 x g, 20°C) and the supernatant withdrawn, filtered through 0.45-µm membrane filters, and acidified to pH < 2. One hundred mL of the extractant from the previous step were used to wash the soil sample for 10 min at 20°C to reduce artifacts due to redistribution of metals remaining in the entrapped extractant solutions. After the separation of the washing solution (centrifugation, filtration, acidification), the extracts were combined and analyzed for Ca, K, and Na using flame photometry (Eppendorf, ELEX 6361, Hamburg, Germany) and for Zn, Pb, Fe, Mn, and Mg using flame atomic absorption spectrometry (Solar, Unicam 989, Cambridge, UK).

Zinc K-Edge Extended X-Ray Adsorption Fine Structure (EXAFS) Spectroscopy
Zinc K-edge (9659 eV) EXAFS spectra of selected untreated or sequentially extracted soil samples were measured at the X-ray absorption (XAS) beamline at the Angstroemquelle Karlsruhe (ANKA), Germany. The ring was operated at 2.5 GeV with a beam current of 130 mA. The Si(111) monochromator was detuned by 40% using a software-controlled monochromator stabilization. Soil and reference samples were prepared as pellets and measured in transmission and/or fluorescence mode using a five-element solid state detector. All measurements were performed at room temperature. The EXAFS spectrum of the sequentially extracted soil sample from 60- to 70-cm depth was measured by Dr. Maarten Nachtegaal at the Dutch Belgian Beamline (DUBBLE) at the European Synchrotron Radiation Facility (ESRF) in Grenoble, France. The raw data were transformed into k-space using the WinXAS software (Ressler, 1998). The pre- and post-edge background correction was performed by fitting a first- and second-order polynomial function, respectively. The E0 was determined as the inflection point of the Zn K-edge (second derivative equal to zero). The µ0 was fit using a cubic spline function and minimizing the amplitude of the Fourier transform at radial distances (r) < 0.9 Å. Linear combination fit (LCF) was based on an extensive collection of reference spectra including Zn complexed by organic substances, Zn adsorbed to metal oxides and clay minerals, and Zn in primary and secondary minerals. Based on preliminary data analyses and soil chemical constraints, the following four spectra were considered in the final LCF fits: ZnS (sphalerite), Zn adsorbed to montmorillonite (SWy-2, pH 5.7, 48 000 mg kg–1 Zn), Zn in trioctahedral smectite (25% Zn and 75% Mg in octahedral sheet, synthesized according to Decarreau [1981]), Zn-containing goethite (2% Zn and 98% Fe, synthesized according to Schwertmann and Cornell [1991]), and Zn sorbed to purified humic acid (pH 6.0, 350 mg kg–1 Zn). Linear combination fit analysis was started with the reference giving the best one-component fit (defined as the one with the lowest normalized sum of squared residuals [NSSR = {Sigma}i(datai – fiti)2/{Sigma}idatai2]). Additional references were considered to significantly improve the fit if the NSSR decreased by more than 10%. The software package from the beamline 10.3.2 of the Advanced Light Source (ALS), Berkeley, USA (Marcus et al., 2004) was used to perform the LCF analysis.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Vertical Distribution of Major and Trace Elements and Variation of pH
Figure 2 shows the vertical profiles of soil pH, total inorganic carbon (TIC), total organic carbon (TOC), and the total contents of Ca, Mg, Zn, Mn, Cd, S, Pb, Fe, and Ti. Below 2 m, the flotation residues had a neutral pH. A pronounced decrease in pH from 7.0 to 6.4 was followed by a further slow decrease to pH 6.0 at 50-cm depth. The topsoil sample was again clearly more acidic with a pH of 5.5. Similar to the pH, the TIC decreased toward the surface from values around 170 mmol kg–1 in the subsoil down to 22 mmol kg–1 in the topsoil. Calcium concentrations increased almost linearly with depth. For Mg no clear trend could be observed along the soil profile. Zinc, Mn, Cd, and S showed a relatively homogeneous distribution in the subsoil below about 60-cm depth. These elements decreased toward the surface and showed lowest contents in the top 20 cm. In contrast to the other metals, Pb showed no depletion but an enrichment in the top cm of the topsoil, where the TOC concentrations were highest.

With XRD analysis, the minerals quartz, illite, muscovite, and zeolite could be identified in all three soil samples. In addition, sphalerite could be identified in the middle and deep sample (P3–0.7 and P3–3.0) but not in the topsoil sample (P3–0.16).

Characterization of Metal Speciation by Sequential Chemical Extractions
Data from SCE are depicted in Fig. 3. There is a general agreement that results from SCE are operationally defined (Filgueiras et al., 2002) and do not necessarily reflect true chemical species. Metals extracted in the first step (NH4NO3) were usually assumed to represent water-soluble and exchangeable cations. The second step (NH4OAc) mainly extracted weakly complexed metals and metals bound by carbonates (easily mobilizable). The third step (NH2OH*HCl) was designed to extract metals bound by Fe/Mn (hydr)oxides. However, Peltier et al. (2005) showed that in step 3 freshly formed amorphous sulfides can also be dissolved whereas primary minerals such as sphalerite and galena seemed to be unaffected (Svete et al., 2001; Peltier et al., 2005). The fourth step (NH4–EDTA) was assumed to extract metals bound by organic matter. In the last step, the residual fraction was extracted by total digestion.


Figure 3
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 3. Total extracted amount of Zn, Pb, Fe, and Mn of all five steps in mmol kg–1, percentage of Zn, Pb, Fe, and Mn extracted by sequential chemical extractions in samples P3–0.16, P3–0.7, and P3–3.0; F1 (NH4NO3), F2 (NH4OAc), F3 (NH2OH*HCl), F4 (NH4–EDTA), and F5 (HF/HNO3/HClO4).

 
The different soil acidity of samples from different depths was reflected in the pH of the first leachate. In the topsoil sample, the pH in the first step was acidic (pH 4.1) whereas the deeper samples show nearly neutral pH (6.6). This agreed well with soil pH measurements (Fig. 2) and had a major impact on desorption of the different metals which strongly depends on pH.

In all three samples, Fe was mainly released in step 3 and step 5, the steps for the removal of Fe/Mn (hydr)oxides and residual mineral phases. In the topsoil sample, no Fe was released in the first two steps, according to low solubility of Fe oxides at oxic conditions (Cornell and Schwertmann, 1996; Scheinost et al., 2002). Differences between the investigated samples from different depths were small (Fig. 3). It was likely that released Fe subsequently reprecipitated as (hydr)oxides in contact with oxygen.

In the two subsoil samples, more than 90% of Mn was released during the first three steps with 50 to 60% in step 3, the step for the removal of Fe/Mn (hydr)oxides. This indicates that Mn was mobilizable to a large extent in the parent material. In the topsoil sample, all Mn was found in the residual step. Manganese contents in this sample were considerably lower than in the parent material (Fig. 2, Table 1), which indicated that large amounts had been mobilized and leached out due to oxidation and weathering of the tailing. The amount of Mn found in the residual phase of the two deeper samples corresponded well with the total content of the topsoil sample (Fig. 3). This suggested that all Mn extracted in the first three steps was already released from topsoil which was now depleted except for the residual fraction.


View this table:
[in this window]
[in a new window]

  		Table 1. Soil characteristics of samples studied by sequential chemical extractions (SCE), X-ray diffraction (XRD), and extended X-ray adsorption fine structure (EXAFS) spectroscopy.{dagger}

 
Zinc
In the two subsoil samples, Zn was released in all extraction steps except for step 4, the step designed for the release of metals bound to organic matter. The topsoil sample from 13- to 16-cm depth (P3–0.16) contained far less Zn (4 mmol kg–1) than the deeper samples (150 mmol kg–1). In the topsoil sample, 23% of Zn was extracted in the third step and 62% in the last step of the SCE, suggesting that Zn is predominantly bound in residual minerals and Fe (hydr)oxides (since only 0.4 mmol kg–1 of Mn were found in this sample, extracted in the residual step). Thirteen percent of Zn was extracted in step 1 and can be regarded as mobile or easily mobilizable. No Zn was extracted in the second step, designed for the extraction of weakly complexed metals and metals bound by carbonates. This was in agreement with carbon measurements (Fig. 2) which showed that the topsoil was already depleted in TIC. More than 60% of Zn was extracted in the last step in all three samples, indicating that most Zn was bound by residual minerals such as sphalerite.

Lead
More than 65% of Pb in the subsoil was extracted in step 3, the step designed for the dissolution of Fe/Mn (hydr)oxides. In the topsoil sample, Pb was predominantly extracted in the first step (73%), indicating that the major part of Pb was present as mobile or easily mobilizable species. In the first step of the SCE, 80% of all cations released were Pb, which means that in the topsoil the largest fraction of the cation exchange capacity was occupied by Pb. Thirteen percent of Pb in the topsoil was released in the third step and 7% in both the second and the residual step.

Originally, Pb has been present as galena in the ore deposit (Hünermann, 1955). However, it is not unlikely that galena has already been oxidized during flotation (Harvey and Yen, 1998; Madhuchhanda et al., 2000) to form anglesite (PbSO4) (da Silva et al., 2003). This would explain why the residual fraction of Pb was small in all three samples.

Zinc K-Edge Extended X-Ray Adsorption Fine Structure (EXAFS) Spectroscopy
Zinc was the most relevant contaminant of the investigated site in terms of quantity. To get complimentary information about the residual Zn species in the different depths on a molecular scale, the three samples from SCE and two additional samples from the topsoil were investigated by Zn K-edge EXAFS spectroscopy.

The Zn K-edge EXAFS spectra of untreated and sequentially extracted soil samples and the corresponding Fourier transforms are shown in Fig. 4A and 4B, respectively. The figure also shows the spectra of the reference compounds which were used for LCF. The results of the LCF analysis are listed in Table 2 and the fitted EXAFS spectra are shown in Fig. 4A (offset thin dashed lines).


Figure 4
View larger version (53K):
[in this window]
[in a new window]

  		Fig. 4. (A) Zinc K-edge extended X-ray adsorption fine structure (EXAFS) spectra of untreated and sequentially extracted soil samples from increasing profile depths and reference spectra (solid lines). Linear combination fit (LCF) spectra of the soil samples are offset for clarity (thin dashed lines). Details indicated by arrows are mentioned in the text. (B) Fourier transforms of the EXAFS spectra of soil samples and references (solid lines = magnitude; thin dashed lines = imaginary part). Vertical lines mark the approximate positions of the peaks for first shell O and S coordinated with Zn in the reference spectra.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Linear combination fit (LCF) results for untreated and sequentially extracted soil samples from different sampling depths. Reference spectra and the LCF spectra are contained in the Fig. 4A.

 
The untreated samples from 230- to 300-cm and 60- to 70-cm depth clearly exhibited the characteristic features of the sphalerite EXAFS spectrum (Fig. 4A). Also, the first shell peak in the Fourier transform approximately matched the position of the first shell peak of sphalerite, which corresponded to the nearest S atoms coordinated to Zn. However, closer inspection of the Fourier transforms showed that for the untreated samples from 60- to 70-cm and 230- to 300-cm, the first shell peak was slightly shifted to the left of the Zn-S peak from sphalerite, indicating a minor contribution from Zn species with Zn coordinated to O atoms. In the spectra of the untreated soil samples collected at 13- to 16-cm and 26- to 34-cm depth, the spectral features of sphalerite had mostly disappeared and the first-shell peak in the Fourier transform approximately coincided with the position of the Zn-O peaks of the references for adsorbed and crystalline Zn species (Fig. 4). However, the peak in the EXAFS spectra at 7.5 Å–1 indicated that some sphalerite was still present in these samples (Fig. 4A, arrows). In the topmost sample from 4- to 6-cm depth, finally, no spectral contribution from sphalerite was visible anymore, and Zn appeared to be exclusively present in adsorbed or precipitated species containing Zn coordinated to O.

While the identification of sphalerite was unequivocal from the EXAFS spectra collected at 60- to 70-cm and 230- to 300-cm depth, the selection of further references for LCF analysis was complicated by the limited number of sample spectra and by the strong contribution from sphalerite in the majority of the spectra. The selection of references was therefore based on preliminary testing of an extensive set of reference spectra of adsorbed and crystalline Zn species and on chemical considerations discussed below. However, the choice of the additional references was found to have little influence on the LCF results for sphalerite.

The LCF of the sample from 230- to 300-cm depth indicated that Zn in the unweathered flotation residue was predominantly contained in sphalerite. The fitted sphalerite fraction (73%) closely corresponded to the residual fraction from the SCE (71%, Fig. 3). Furthermore, the best LCF of the extracted sample from 230- to 300-cm depth was obtained by using the sphalerite reference only. In combination, the LCF and SCE results for that sample suggested that during the SCE, sphalerite was mostly retained in the residual fraction. In the spectra from 60- to 70-cm and 13- to 16-cm depth, the sphalerite fraction decreased to 65 and 15%, respectively. In contrast to the extracted sample from 230- to 300-cm depth, the extracted samples from 60- to 70-cm and 13- to 16-cm depth revealed an increasing contribution from other residual Zn species in which Zn is coordinated to O. The LCF of the spectra of the untreated and treated samples from 13- to 16-cm depth and the sample from 26- to 34-cm depth consistently improved if the reference spectrum of Zn-substituted goethite was added. In contrast, reference spectra for Zn adsorbed to ferrihydrite or coprecipitated with ferrihydrite did not improve the LCF. Incorporation of Zn into newly forming goethite is possible, as the sample from 26- to 34-cm depth exhibited a distinct rusty-red color indicating Fe oxide formation and as Fe oxide-rich zones were also found around root channels in the 13- to 16-cm depth. From the interpretation of the SCE protocol, on the other hand, goethite would have been expected to dissolve during the third step of the SCE, while the LCF indicated a contribution of 40% Zn-goethite in the extracted sample from 13- to 16-cm depth. The improvements in the LCF obtained by the Zn-goethite reference might therefore also be due to the presence of another recalcitrant species with similar spectral features. The EXAFS spectrum of the sample from 4- to 6-cm depth was clearly distinct from the spectra of deeper soil samples (Fig. 4A) and the LCF did not indicate any contribution from sphalerite in that spectrum (Table 2). While the EXAFS spectrum did not exhibit the distinct spectral features of the ZnMg-clay reference containing 25% Zn and 75% Mg in its octahedral sheets, its envelope and frequency match that of the ZnMg-clay reference spectrum. This suggests that the spectral contributions of Zn adsorbed to soil clay minerals or soil organic matter reduced any sharp features of Zn contained in ordered structures. Such an interpretation was supported by the LCF, which yielded 53% contribution from the Zn sorbed montmorillonite reference and 38% contribution from the ZnMg-clay reference. The ZnMg-clay reference spectrum used in the LCF analysis was chosen from a wide range of references, which exhibit similar spectral features depending on the ratio of light and heavy atoms in their octahedral sheets (ZnxMg1-x-clays with x ranging from 0.03 to 1; Zn in layered double hydroxides [LDH], Zn sorbed in the hydroxy-aluminum interlayers of clay minerals or in the gibbsitic sheets of lithiophorite). All of these phases have been identified before in pristine or contaminated soils (Manceau et al., 2000; Scheinost et al., 2002; Juillot et al., 2003; Manceau et al., 2004; Panfili et al., 2005; Voegelin et al., 2005). The formation of Zn-bearing LDH was unlikely due to the relatively low pH of the topsoil samples, and the formation of Zn-bearing lithiophorite can be excluded as major species based on the small Mn content (Table 1). We therefore assume that the spectral contribution fit by the ZnMg-clay reference resulted from the presence of Zn in the octahedral sheets of clay minerals and/or Zn sorbed into the Al-hydroxy interlayers of clay minerals. These species likely also resisted the first three steps of the SCE and their presence in the spectrum from the extracted sample from 13- to 16-cm depth would be chemically realistic. Considering the organic carbon content of 24.2 g kg–1 in the sample from 4- to 6-cm depth, some association of Zn with organic matter could be expected. However, considering the similarities in the reference spectra for Zn adsorbed to montmorillonite and humic acid, this fraction was likely included in the fraction fit with the montmorillonite reference.

Distribution and Fate of Zinc and Lead in the Fine-Grained Flotation Residues
Zinc
The analysis of the EXAFS spectra showed that in the unaltered flotation residues Zn was predominantly present as sphalerite, and that the fraction of sphalerite decreased with decreasing profile depth. The stability of sphalerite mainly depends on the redox conditions. Under anoxic conditions, sphalerite has a low solubility (log Ksp = –20.6 for ZnSs -> Zn2+aq + S2–aq at 25°C in water [Weisener et al., 2003]). Electron supply by O2 may lead to a transformation of sulfide into soluble sulfate (Garcia et al., 1995) and a significant depletion of the soil in S and Zn due to leaching. Significant amounts of dissolved sulfate in soil and ground water may be the consequence.

The EXAFS spectra further indicated that sphalerite was retained in the residual fraction of the SCE. However, with decreasing profile depth, this residual fraction also contained increasing amounts of other Zn species. The exact identification of these species was complicated by the limited number of spectra and the strong contribution of the sphalerite signal in most spectra. However, spectra collected from samples of the oxidized topsoil (<34 cm) suggested the presence of Zn-substituted goethite (or a species with similar spectral features) and of Zn bound in the octahedral sheets of Zn-bearing phyllosilicate and/or in the hydroxy-aluminum interlayers of clay minerals.

The SCE results indicated that Zn was partly redistributed into mobilizable species (Fig. 3). The acidic pH of the topsoil lead to a transfer to the aqueous phase by desorption and/or dissolution and leaching to the subsoil, where the pH is still neutral and metals were retarded due to adsorption, surface complexation, and/or (co)precipitation.

Originally, pyrite was present in the ore deposit (Hünermann, 1955). The oxidation of pyrite to a great extent would lead to considerable acidification. If other sulfides with a lower rest potential are coexistent, pyrite oxidation will be diminished due to galvanic interactions (Domènech et al., 2002; Jeong and Lee, 2003; Kwong et al., 2003). It is possible that the coexistence of sphalerite retards the oxidation of pyrite and has prevented much lower pH conditions so far.

The processes we observed to occur in fine-grained sulfidic flotation residues were similar to the processes occurring in acid sulfate soils (ASS). In ASS of coastal regions of Finland and Sweden, the oxidation of sulfide minerals (particularly pyrite) leads to an acidification of the topsoil (Åström, 1998; Sohlenius and Öborn, 2004). The element distribution showed a similar pattern. Zinc and also Cd were probably mobilized from ASS due to the low pH or partly due to the oxidation of discrete sulfide minerals (Sohlenius and Öborn, 2004). It is suggested that the immobilization in the transition zone was caused either by the increase in pH, the decrease in O2 supply, or by the combined action of both of these changes. In the reduced zone, where O2 was unavailable and the pH was high, the concentration of Zn and Cd was higher than in the acidified topsoil, whereas distinct concentration maxima were located in the narrow transition zone of most but not all investigated sites (Åström, 1998).

Lead
Lead was enriched in the topsoil in the mobile fraction compared with the unweathered subsoil (Fig. 2). Results from total content analysis suggested that this was due to an interaction with soil organic matter. Diffusion processes and/or plant uptake may have played a major role controlling Pb enrichment at the surface. The transfer of Pb to the soil solution strongly depends on pH. pH-dependent sorption behavior of the different metals strongly correlated with hydrolysis constants (Herms and Brümmer, 1984; Voegelin et al., 2003). Lead has one of the largest first hydrolysis constants of any divalent cation (pK 7.6 for Pb2+ + H2O -> PbOH+ + H+ [Alloway, 1999]). In the presence of abundant surface sites, as expected for this soil, Pb should be quickly removed from solution by adsorption (Wang and Benoit, 1996). However, a further decrease in pH below 4 could lead to significant desorption of Pb and substantial pollution of the ground water.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Extended X-ray adsorption fine structure spectroscopy results confirmed that Zn was mainly bound in sphalerite in the unweathered subsoil. Weathering reactions lead to a redistribution of Zn in the topsoil and a loss of 35% of the Zn and S from the weathered flotation residues in the topsoil. Lead, on the other hand, was enriched by a factor of two in the topsoil relative to the unweathered flotation residues, most likely due to its strong association with the soil organic matter. Due to the low pH of the topsoil, most Pb was present in readily mobile or mobilizable form. Overall, Pb accounted for about 80% of the effective cation exchange capacity of the topsoil. Thus, the risk of Pb uptake by plants and its transfer via the food chain is of major concern.

By combining the SCE and EXAFS, we were able to more accurately characterize the speciation and reactivity of Zn in the weathering flotation residues than would have been possible with only one of these methods. While the SCE allowed quantification of the mobilizable Zn species, the EXAFS analyses allowed us to characterize the residual Zn phases which could not be differentiated by SCE.

Oxidation of sulfide minerals lead to an acidification of the flotation residues to a pH of 5.5, where buffering minerals were already depleted. Continued acidification would lead to a significant release of Zn, S, and Pb and cause a substantial pollution of the ground water.


   ACKNOWLEDGMENTS
 
We acknowledge three anonymous reviewers for their helpful comments. The Angströmquelle Karlsruhe GmbH (ANKA, Karlsruhe, Germany) is acknowledged for providing beamtime at the XAS beamline. We thank Stefan Mangold for his assistance with the EXAFS measurements at the ANKA. Dr. Maarten Nachtegaal is acknowledged for measuring the EXAFS spectrum of one of the soil samples at the DUBBLE beamline at the ESRF in Grenoble. Olivier Jacquat (ETH Zurich) is acknowledged for the synthesis of the Zn(Mg)-clay and the Zn-goethite references. Christian Bitterli (ETH Zurich) helped with the XRF and carbon analyses. We thank the working group of Prof. König (especially Uta Limper and Martina Schlander) from the Institute of Microbiology and Wine Research at Mainz University for the allocation of the glovebox and technical support. We thank the working group of Prof. D. Schenk at Mainz University for their contribution. The last step of the sequential chemical extractions was conducted by the laboratory of the Department of Sedimentology/Environmental-Geology at the Centre of Earth Science, University of Goettingen.

Part of this study was funded by the State of Rhineland-Palatinate, Germany, represented by the Office for Environment, Water Economy, and Trade Supervisory (Landesamt für Umwelt, Wasserwirtschaft, und Gewerbeaufsicht Rheinland-Pfalz).


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schuwirth, N.
Right arrow Articles by Hofmann, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schuwirth, N.
Right arrow Articles by Hofmann, T.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Schuwirth, N.
Right arrow Articles by Hofmann, T.
Related Collections
Right arrow Mining Wastes
Right arrow Heavy Metals
Right arrow Soil Pollution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome