| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a School of Environmental Engineering, Faculty of Environmental Sciences, Griffith University, Nathan, QLD, 4111, Australia
b CRC for Coastal Zone, Estuary and Waterway Management, Indooroopilly Sciences Centre, 80 Meiers Road, Indooroopilly, QLD, 4068, Australia
c School of Australian Environmental Studies, Faculty of Environmental Sciences, Griffith University, Nathan, QLD, 111, Australia
* Corresponding author (e.burton{at}griffith.edu.au)
Received for publication May 11, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
It is becoming increasingly accepted that total trace metal concentrations in sediment may give insufficient information on mobility and availability, and therefore not allow a full assessment of the environmental impacts of metal-enriched sites (Australian and New Zealand Environment and Conservation Council/Agriculture and Resource Management Council of Australian and New Zealand, 2000). Knowledge of contaminant partitioning between sediment and pore water, and exchange at the sediment–water interface allows a more complete description of environmental risk (Ryssen et al., 1999). Such partitioning is strongly influenced by association with various geochemical fractions (e.g., carbonate minerals, organic matter, sulfides, or Fe oxides) and environmental processes acting on the sediment (e.g., bioturbation, resuspension, burial, and pH and Eh changes).
Sequential extractions involve the selective extraction of trace metals from operationally defined sediment solid fractions (Tessier et al., 1979; Gleyzes et al., 2002). A sediment sample is subjected to a series of increasingly aggressive, phase-specific reagents under certain conditions. Sequential extraction techniques are not without limitations and have been criticized by Nirel and Morel (1990). Of particular importance in this regard is the potential for reagents to be non-phase specific and for metal ions solubilized by a reagent to readsorb onto a different sediment phase. Nevertheless, sequential extractions may still provide useful information on trace metal mobility and reactivity in a particular environmental context.
Sequential extraction techniques have been applied to study the geochemical partitioning of trace metals in contaminated soils (Hickey and Kittrick, 1984; Basta and Gradwohl, 2000), riverine sediments (Pardo et al., 1990; Jain, 2004), and estuarine sediments (Jones and Turki, 1997; Morillo et al., 2004). Geochemical partitioning results have also been used as an aid in predicting potential contaminant mobility and bioavailability (Phillips and Chapple, 1995; Kabala and Singh, 2001; Pueyo et al., 2003). It is clear that such information aids our understanding of trace metal behavior in the environment, and is therefore of widespread interest.
This study quantifies the solid–solution partitioning and geochemical fractionation of Cu, Pb, and Zn in three benthic sediment profiles from southeastern Queensland, Australia. The objective was to gain insight into the processes controlling trace metal partitioning in benthic sediments.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The study area can be regarded as a typical bar-built estuary, formed when off-shore barrier sand islands build above sea level and form a chain between headlands broken by one or more inlets. As a result, the majority of the study area comprises deep, siliceous sand accumulations derived from long-shore ocean currents. Land adjacent to the western shores of the Broadwater is also predominantly composed of siliceous beach sands.
Previous work has identified locations in this area where total sediment Cu, Pb, and Zn concentrations exceed natural background levels as well as Australian sediment quality guideline trigger values (Burton et al., 2004). This study characterizes the geochemical partitioning of Cu, Pb, and Zn in benthic sediment profiles at three previously identified sites of trace metal enrichment (Sites 1, 2, and 3). Sites 1 and 2 are located in residential canals and receive trace metals mainly from urban stormwater runoff. Site 3 is located within a commercial marina and receives metal inputs from boat maintenance activities, the use of antifouling paints, and urban stormwater runoff.
Sampling and Sample Preparation
Sediment cores were collected with a push-tube coring device (100-mm i.d.), similar to that described by Doyle et al. (1995). This type of sampling device allows the collection of relatively undisturbed sediment profiles from both muddy and sandy substrates. Triplicate cores were collected from each of the three study sites. The cores were sealed at the bottom with a rubber bung, filled with water from the study site, and sealed at the top with a layer of plastic. The cores were transported upright, in the dark, and on ice to the laboratory within 5 h of collection. They were stored at 4°C until sectioned, within 24 h, into 2-cm depth intervals to a total depth of 20 cm. The triplicate depth segments were then homogenized under a flow of N2, and stored frozen under N2 until analysis.
Sediment Characterization
The pH and Eh of the saturated sediment samples were recorded with a Beckman (Fullerton, CA) 
50 m and appropriate electrode. The syringe technique described by Percival and Lindsay (1997) was used for the simultaneous determination of water content, bulk density, and porosity. Subsamples of sediment were wet-sieved through a 63-µm stainless steel sieve, to determine "percent mud" on a mass basis (Loring and Rantala, 1992). A further separate subsample of the saturated sediments was used for determination of total organic C, according to the rapid dichromate oxidation technique described by Nelson and Sommers (1996). Amorphous and crystalline Fe oxyhydroxide mineral content was determined by analysis of Fe extracted in Steps 3 and 4, respectively, of the geochemical fractionation technique described below. Oven-dry samples (105°C, 24 h) were digested with HNO3–H2O2–HCl according to USEPA Method 3050B (USEPA, 1986) and analyzed for total Cu, Pb, and Zn concentrations.
Sediment pore water was extracted by centrifugation under N2 at 4000 x g for 30 min. Following centrifugation, pore water was removed with a syringe, filtered (0.45 µm), and acidified to pH < 2 with HNO3. It should be recognized that all pore water collection and processing methods have been shown to alter pore-water chemistry (Schults et al., 1992; Bufflap and Allen, 1995; Sarda and Burton, 1995). However, in a comparison of methods for collecting pore water, Schults et al. (1992) found that "the centrifuge method is the most accurate and precise for most chemicals."
The benthic profiles sampled in the present study were relatively sandy, only moderately reducing (see Results and Discussion, below), and exhibited no qualitative evidence for significant sediment sulfide concentrations [e.g., H2S odor, black Fe(II) regions]. Preliminary analysis of acid-volatile sulfide (AVS), performed as described by Simpson (2001), for the Site 1 sediment profile revealed <2 µmol/g AVS. Consequently, no further measurements of reduced S species were made during this work.
Geochemical Fractionation
The technique developed by Tessier et al. (1979) is one of the most frequently employed sequential extraction schemes, and forms the basis for the metal fractionation scheme employed in this study (Table 1). To obtain a greater understanding of potential trace metal mobility, the original single extraction targeting Fe and Mn oxides as described by Tessier et al. (1979) was replaced with two separate extractions (Amacher, 1996). Amorphous Fe and Mn oxides were extracted with an acidic (pH 2) hydroxylamine hydrochloride solution. This was complemented by dissolution of crystalline Fe and Mn oxides with a more aggressive ammonium oxalate extraction.
|
Analysis
Analysis of Cu, Fe, Pb, and Zn was performed using a SpectraAA 20-Plus flame atomic absorption spectrometer (AAS) with an air/acetylene flame (Varian, Palo Alto, CA). Calibration for AAS analysis was achieved with prepared external standards via the standard curve approach. For metal analysis in each step of the sequential extraction, analytical standards were prepared in the corresponding extractant to minimize matrix effects. Standards for pore-water analysis were prepared in major-ion artificial seawater. Full calibration was performed after every set of 10 samples, with resloping of the calibration curve performed after every set of five samples. The method detection limit (MDL) for metal analysis was defined as three times the standard deviation of 10 replicate blank measurements. Typical MDLs were approximately 0.02 mg/L for Cu, 0.04 mg/L for Pb, and 0.01 mg/L for Zn in artificial seawater.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The organic C content at Sites 1 and 2 was relatively high, ranging from 1.6 to 5.8% and 2.4 to 5.8%, respectively. Organic C content was lower at Site 3, with 3.3% in the 0- to 2-cm depth interval, decreasing to 0.8% lower in the profile. A notable feature of profiles from Sites 1 and 2 from the residential canals was the presence of coarse, recalcitrant organic material, derived from stormwater inputs of native vegetation.
The pore-water pH was relatively constant for all profiles (pH 7.0–8.1). This is consistent with Berner (1981), who recorded hundreds of pH measurements in estuarine and marine sediments and never found pH values outside the range of pH 6 to 8. This pH range is also consistent with previous work in the study area (Burton et al., 2004).
Redox conditions can influence trace metal behavior in benthic sediments either directly or indirectly through a change in the oxidation state of a ligand capable of complexing the metal. Changes in redox conditions can also cause the decomposition of mineral species (e.g., amorphous Fe-oxyhydroxides or Fe-sulfides) that may sorb trace metals. The measured Eh values were similar for all three profiles, ranging from +260 to +120 mV at Site 1, +250 to +130 mV at Site 2, and +240 to +150 mV at Site 1. According to Sposito (1989), Eh values in the range +120 to +414 mV indicate suboxic, moderately reducing conditions.
Under the observed Eh conditions, the sediment redox status is primarily controlled by Mn and Fe redox reactions (Sposito, 1989). These elements exist as stable oxyhydroxide minerals in oxic sediment. However, at Eh less than approximately +250 and +100 mV reductive dissolution of Mn and Fe oxyhydroxide minerals, respectively, begins to occur (Patrick and Jugsujinda, 1992). At Eh values less than approximately –120 mV the sediment is considered anoxic, and reduction of SO42– to H2S occurs (Bartlett, 1999). Given the measured Eh range (+120 to +260 mV), sulfide accumulation is unlikely in the sediments examined in this study. This is supported by a lack of qualitative evidence for significant sediment sulfide concentrations, such as H2S odor and black Fe(II) regions.
The sediment texture, organic C content, water content, and porosity were significantly interrelated (P < 0.05) as assessed by correlation between properties. Correlation between several sediment properties probably reflects the importance of particulate deposition and sediment mixing rather than diagenetic processes (which are largely controlled by a redox gradient).
Solid–Water Partitioning of Copper, Lead, and Zinc
The ranges observed in the present study for total metal concentrations are comparable with those observed previously at Sites 1, 2, and 3, and other nearby sites (Burton et al., 2004). Total Cu concentrations ranged from 8.5 to 31.9 mg/kg at Site 1, 8.3 to 35.8 mg/kg at Site 2, and 82.5 to 194 mg/kg at Site 3 (Fig. 1) . Background Cu concentrations in Queensland estuarine sediments (based on <63-µm size fraction) are 5 to 23 mg/kg (Moss and Costanzo, 1998). The trigger level for sediment contamination as presented in the interim Australian sediment quality guidelines is 65 mg/kg Cu (Australian and New Zealand Environment and Conservation Council/Agriculture and Resource Management Council of Australian and New Zealand, 2000). The range of total concentrations indicates that Cu levels at Sites 1 and 2 are only slightly above the background levels but are below the interim value presented by the Australian and New Zealand Environment and Conservation Council/Agriculture and Resource Management Council of Australian and New Zealand (2000). Total Cu at the marina Site 3 greatly exceeds both the background values and the sediment quality guidelines. Highly elevated Cu concentrations at Site 3 can be attributed to inputs of Cu derived from activities performed within the commercial marina surrounding this site. Copper is employed as an active ingredient in antifouling paints for recreational and commercial vessels moored and maintained at Site 3.
|
Total Zn concentrations ranged from 36.9 to 127 mg/kg for Site 1, 30.1 to 89.3 mg/kg for Site 2, and 103 to 220 mg/kg for Site 3 (Fig. 1). The values for Sites 1 and 2 are fairly consistent with background values for Queensland estuaries (37–110 mg/kg) and meet the sediment quality guideline of 200 mg/kg. The background range and sediment quality guideline are exceeded in some depth intervals at the marina Site 3. This may be attributed to past inputs of Zn from vessel maintenance activities, as well as runoff from local metal-based manufacturing industries (such as galvanizing works, vehicle repair yards, and paint and ink manufacturing).
It is important to note that the background Cu, Pb, and Zn concentrations quoted above reflect Cu, Pb, and Zn concentrations in the <63-µm size fraction of Queensland estuarine sediments (Moss and Costanzo, 1998). In contrast, the total values in this study are based on the whole sediment with only very coarse (>2 mm) detritus removed. It is generally accepted that contaminants are likely to be almost entirely associated with particle sizes of <63 µm (i.e., silt- and clay-sized particles, collectively defined as "mud" by Loring and Rantala, 1992) and that sand-sized particles act to dilute metal contamination in sediments. In the three profiles examined in the present study, the <63-µm size fraction generally comprises <30% of each profile. If the total Cu, Pb, and Zn concentrations were corrected for the abundance of <63-µm particles, then a relatively large proportion of depth intervals do indeed exceed the background values. This supports the assertion that sediment texture should be considered when assessing metal enrichment in sediments (Loring and Rantala, 1992).
In the present study, total Cu, Pb, and Zn concentrations at each site were strongly related to sediment texture (r2 = 0.83–0.95), as characterized by percent mud (Fig. 2) . This suggests that the vertical profiles for total Cu, Pb, and Zn shown in Fig. 1 are largely a consequence of vertical trends in sediment texture.
|
Pore-water Cu concentrations ranged from 70 to 80 µg/L at canal Sites 1 and 2 and up to 280 µg/L at the marina Site 3 (Fig. 1). Pore-water Pb ranged from 400 to 600 µg/L at Sites 1 and 2, and from undetectable to 400 µg/L for Site 3 (Fig. 1). Pore-water Zn was generally <200 µg/L, but with peak concentrations up to 800 µg/L in the 12- to 16-cm depth intervals at all three sites (Fig. 1). These values are comparable with pore-water Cu (100–200 µg/L) and Zn (approximately 100 µg/L) in estuarine sediments as reported by Sullivan and Taylor (2003). The observed pore-water Cu, Pb, and Zn concentrations are relatively high in comparison with the Australian marine water quality guideline trigger values for protection of 80% of species (Cu = 8 µg/L, Pb = 12 µg/L, and Zn = 43 µg/L; Australian and New Zealand Environment and Conservation Council/Agriculture and Resource Management Council of Australian and New Zealand, 2000).
A common approach to derivation of sediment quality guidelines is to assume that the primary exposure route for benthic organisms is via pore water. In this approach sediment quality is assessed by comparing water quality guidelines, developed from toxicity testing of free-swimming organisms, against measured pore-water concentrations. It is clear that the measured pore-water Cu, Pb, and Zn concentrations exceed the water quality guidelines for several depth intervals, thus suggesting that the sediment may represent a risk to benthic biota.
In strongly anoxic sediments (Eh less than –120 mV), the production of sulfide via bacterial sulfate reduction would be expected to buffer pore-water metal concentrations to very low levels via precipitation–dissolution of monosulfide minerals. For example, Simpson et al. (2002) report low pore-water concentrations of Cu, Pb, and Zn (<2.5 µg/L) for fine-textured, sulfidic sediment. Several researchers have also shown that the ratio of acid-volatile sulfide (AVS) to simultaneously extracted metals (SEM) largely dictates metal bioavailability and thus toxicity in anoxic sediment (Di Toro et al., 1990, 1992; Berry et al., 1996). The profiles examined in the present study exhibit Eh values well above the value where sulfate reduction has been observed. As such, pore-water sulfide or AVS is unlikely to be an important binding phase for trace metals in the sediments discussed in this study.
In sediments that do not contain significant pore-water sulfide or AVS, hydroxide or carbonate minerals may govern the maximum pore-water Cu, Pb, and Zn concentrations. By assuming equilibrium conditions, it is possible to calculate the aqueous concentration of Cu, Pb, and Zn in equilibrium with possible minerals derived from these metals and species such as hydroxide and carbonate that are present in oxic seawater. Possible minerals for Cu, Pb, and Zn in a seawater matrix at pH 7 include the carbonates Cu2(OH)2CO3 (malachite), Pb(OH)2:PbCO3 (hydrocerussite), and ZnCO3:H2O, and the hydroxides Cu2(OH)2, Pb(OH)2, and Zn(OH)2. The solubility in seawater of Cu, Pb, and Zn in equilibrium with these carbonate and hydroxide minerals was modeled with the PHREEQC code (Parkhurst, 1995) employing the MINTEQA2 database (Allison et al., 1991). The calculated aqueous equilibrium concentrations were 0.15 and 0.52 mg/L Cu, 1.23 and 0.09 mg/L Pb, and 13.8 and 33.2 mg/L Zn for the above carbonate and hydroxide minerals, respectively. In general, the measured pore-water Cu and Pb concentrations shown in Fig. 1 are relatively consistent with mineral solubilities calculated by PHREEQC. This comparison provides tentative evidence that precipitation–dissolution reactions may be limiting the maximum pore-water Cu and Pb concentrations. It should be noted that further work is required to verify the notion that hydroxy-carbonate minerals are controlling Cu and Pb solubility in the profiles examined in the present study. The measured pore-water Zn concentrations were well below the calculated solubility reported above. This indicates that adsorption reactions are likely to be important for Zn behavior in the observed sediment profiles.
The partitioning of trace metals between solid and aqueous phases is often quantified by a distribution coefficient (Kd). An estimate of the Kd value can be made by:
 |
The Kd,obs values for some depth intervals for this study are, however, relatively low for pH 7 to 8 conditions. The presence of fine inorganic colloids (e.g., Fe and Mn oxyhydroxides) that pass through a 0.45-µm filter and dissolved organic material have been shown to substantially enhance dissolved metal concentrations via complexation reactions (Elderfield, 1981). The relatively low Kd,obs values for some depth intervals may reflect reduced sorption due to complexed pore-water metals. However, future aqueous speciation work is required to verify the relative importance of pore-water metal complexation.
Geochemical Fractionation
Metal Recovery for the Sequential Extraction Technique
The metal recovery of the sequential extraction analysis was assessed by comparison of the sum of fractions with total metal digestion concentrations for each sediment sample (Fig. 3)
. This comparison reveals an excellent agreement between the total (as determined with the use of USEPA Method 3050B; USEPA, 1986) and the sum of fractions for Cu, Pb, and Zn sequential extraction analysis throughout the benthic profiles. This outcome suggests that the sequential extraction analysis allows a quantitative recovery of total Cu, Pb, and Zn. At worst, then, the sequential extraction data simply reflect ease of extractability (which may be related to potential mobility and bioavailability), and at best provide a valuable insight into the geochemical mode of trace metal retention.
|
Fujiyoshi et al. (1996) also found that Zn sorbed on marine sediment was not extractable with 1 M CH3COONH4, implying that sorption did not occur via a nonspecific, outer-sphere, electrostatic mechanism. This contrasts with results for metal fractionation in freshwater sediments, where previous work has found that exchangeable metals are often not negligible (Jain, 2004). This reflects the importance of the ionic strength of the aqueous matrix, which in some freshwater systems allows aqueous trace metals to occupy cation exchange sites. Emmerson et al. (2001) have shown that metals sorbed to soil are released from exchange sites (via displacement by major seawater cations) when soil is mixed with seawater. This is consistent with work by Calmano et al. (1992), who showed that increased salinity enhanced desorption of trace metals from sediments. The present results are therefore consistent with previous work in sediment–seawater systems indicating that nonspecific sorption to negatively charged colloid surfaces is of little importance to trace metal retention in saline environments.
Carbonate Fraction
Trace metals recovered with the use of 1 M NaOAc adjusted to pH 5 are associated with the operationally defined carbonate fraction (i.e., associated with carbonate minerals). However, Gleyzes et al. (2002) caution that metals extracted from soils or sediments with 1 M NaOAc adjusted to pH 5 may have also been specifically sorbed to low energy sites on the surfaces of clay minerals, organic matter, and oxide minerals, as well as coprecipitated with carbonate minerals. Therefore, it is acknowledged that metals associated with the operationally defined carbonate fraction may also be weakly sorbed to other noncarbonate phases.
It is clear that metals recovered within the carbonate fraction, whether truly associated with carbonates or not, are not strongly bound to the sediment solids. Metals associated with this fraction are likely to be released to the sediment pore water if acidic conditions (pH < 5) were to develop in the field.
Turner and Olsen (2000) determined extractability of trace metals in contaminated estuarine sediments by chemical and enzymatic extractions. Of the chemical reagents that they examined, acetic acid best represented the fraction that was likely to be bioavailable to sediment ingesting biota. These researchers defined bioavailability based on pepsin extractability, which provides an indication of metal extraction by digestive fluids within the gut of sediment-ingesting biota. Trace metals extractable with 1 M NaOAc adjusted to pH 5 (with acetic acid) are therefore likely to be bioavailable to sediment-ingesting, benthic organisms (Tessier and Campbell, 1987).
In general, the carbonate fraction was of negligible importance to Cu retention (<4%; Fig. 4)
, and of low to moderate importance to Pb (0–62%; Fig. 5)
and Zn (1–34%; Fig. 6)
retention in the three profiles examined in this study. While there was considerable variation between sites and depth intervals in terms of trace metal association with carbonate minerals, the results reflect the overall trend of association with the carbonate phase as:
 |
|
|
|
Amorphous Oxide and Crystalline Oxide Fractions
Oxyhydroxide (oxide) minerals, along with organic matter, have long been recognized as the predominant metal sorbents in aquatic systems. In comparison with carbonate minerals, amorphous oxide minerals have relatively large surface areas (oxides up to 300 m2/g, organic matter up to 1900 m2/g, and carbonates usually <1 m2/g) and surface site density that is three to four orders of magnitude greater (Forstner and Wittmann, 1979; Benjamin and Leckie, 1981; Bilinski et al., 1991).
The geochemical fractionation results from the present study are consistent with the high affinity of trace metals for amorphous oxide minerals. Association with amorphous oxide minerals (defined by extraction with 0.25 M NH2OH·HCl, 0.25 M HCl [pH 2] for 30 min at 50°C) was highly important to Cu (Fig. 4), Pb (Fig. 5), and Zn (Fig. 6) retention in all three profiles. Retention by amorphous oxide minerals followed the same general trend between individual trace metals as for carbonate minerals, with 8 to 58% of Cu, 22 to 80% of Pb, and 32 to 89% of Zn associated with amorphous oxides.
In contrast to amorphous minerals, only 3 to 11% of Cu, 0 to 31% of Pb, and 3 to 15% of Zn were associated with the operationally defined crystalline oxide fraction. This probably reflects the much greater surface area of amorphous minerals in comparison with more crystalline material (Kampf et al., 2000).
The observed trends in the association of Cu, Pb, and Zn with amorphous and crystalline oxide minerals were moderately well explained by the abundance of amorphous and crystalline Fe oxides, respectively. The linear regression (through the origin) parameters describing these relationships are presented in Table 2, and provide information that aids in the explanation of the observed fractionation trends. In particular, the gradient of the regression lines (expressed as mg Cu, Pb, or Zn per g Fe oxide as Fe) represents the stoichiometry of Cu, Pb, and Zn interaction with the amorphous and crystalline oxide fractions. This relationship provides a quantitative indication of the affinity of a given metal for the amorphous oxide or crystalline oxide fractions. For retention by amorphous oxide minerals, the observed stoichiometry ranged from 5.2 to 23.7 mg/g for Cu, 12.8 to 21.5 mg/g for Pb, and 23.1 to 85.7 mg/g for Zn (Table 2). The corresponding values for retention by crystalline Fe oxides were in general an order of magnitude less than for amorphous minerals, ranging from 0.51 to 0.84 mg/g for Cu, 0.49 to 1.06 mg/g for Pb, and 1.61 to 1.87 mg/g for Zn. These data indicate that the lesser importance of the crystalline oxide fraction in comparison with the amorphous oxide fraction for Cu, Pb, and Zn retention is not due to a lesser abundance of crystalline Fe oxides. Rather, the results indicate that the amorphous oxide minerals have a much greater stoichiometric affinity for Cu, Pb, and Zn than do crystalline oxide minerals. In terms of the affinity of different metals for both amorphous and crystalline oxides (based on the stoichiometric affinity), the results indicate:
 |
|
Metals associated with oxide minerals are likely to be released if these minerals were to dissolve. Reductive dissolution of the oxide mineral occurs at Eh less than approximately +250 mV for Mn oxides and +100 mV Fe oxides (Patrick and Jugsujinda, 1992). Given that measured Eh values ranged from +120 to +260 mV in the sediments examined in this study, it is clear that relatively small changes in Eh toward reducing conditions will cause reduction of Fe and Mn species. This will cause dissolution of Fe and Mn oxide minerals, thereby allowing release of associated Cu, Pb, and Zn.
Organic Fraction
The H2O2–extractable fraction is assumed to reflect trace metals strongly bound to sediment organic material. The results indicate that the organic fraction was very important for Cu retention, with 13 to 69% associated with this fraction (Fig. 4). In contrast, relatively low proportions of Pb (0.4–24%; Fig. 5) and Zn (2–13%; Fig. 6) were associated with organic matter in the profiles examined in this study. As with the oxide mineral fractions, the trends in association with the organic fraction were explained moderately well by the abundance of organic matter (Table 2). The stoichiometric relationships describing trace metal retention by organic matter ranged from 17.6 to 54.0 mg/g for Cu, 6.1 to 9.6 mg/g for Pb, and 6.4 to 16.4 mg/g for Zn. The affinity for retention by organic matter, based on these stoichiometric relationships (Table 2) as well as the proportions of total Cu associated with the organic-bound fraction (Fig. 4, 5, and 6), followed the order:
 |
The tendency of Cu to be associated with the oxidizable fraction (which implies association with organic matter in the sediments examined in this study) has been reported in several previous studies (Pardo et al., 1990; Lopez-Sanchez et al., 1996; Galan et al., 2003). This is attributed to the greater stability of organo–Cu complexes when compared with Pb and Zn (Stumm and Morgan, 1996).
Residual Fraction
The residual fraction represents metals occluded within the crystal structure of recalcitrant minerals. This fraction is not available to biological or diagenetic processes except over very long time scales (Tessier et al., 1979). It has been suggested that this fraction is most important for noncontaminated sediments. In the present study, moderate proportions of Cu (10–35%; Fig. 4), low to moderate proportions of Pb (0–20%; Fig. 5), and low proportions of Zn (2–8%; Fig. 6) were associated with the residual fraction. In uncontaminated settings, the residual fraction is usually the most important geochemical phase for trace metal retention. The association between trace metals and the residual fraction of uncontaminated soils is so strong that metal association with nonresidual fractions has been used as an indicator of anthropogenic enrichment (Arakel and Hongjun, 1992; Sutherland et al., 2000). The minor importance of the residual fraction in the present study supports work by Burton et al. (unpublished data, 2004), who found that sediment from the sites examined in the present study was enriched with Cu, Pb, and Zn in comparison with background levels and geochemical normalization against Al.
General Geochemical Fractionation Trends
Overall, the order of importance of the individual geochemical fractions was:
Cu: amorphous oxides 
organic matter > residual > crystalline oxides >> carbonates > exchangeable
Pb: amorphous oxides > organic matter residual > crystalline oxides > carbonates >> exchangeable
Zn: amorphous oxides >> carbonate > crystalline oxides residual > organic matter > exchangeable
These findings suggest that the order of potential metal mobility in the examined sediment profiles was:
|
|
This sequence is the reverse of the trend in hydrolysis constants for these metals, which are 10–7.8, 10–8.0, and 10–9.0 for Cu, Pb, and Zn, respectively (Basta and Tabatabai, 1992). McBride (1989) has shown that chemisorption of trace metals to mineral and organic surfaces is related to the hydrolysis of metal cations in solution. Sorption studies also typically show that metal sorption affinity is strongly related to hydrolysis (Stumm and Morgan, 1996). The estimated solubilities of Cu and Pb hydroxide minerals, as presented above, also reflect the importance of hydrolysis as an influence on metal behavior in aquatic systems. Overall, the trend in potential mobility of Cu, Pb, and Zn based on geochemical fractionation is in accordance with that expected on the basis of hydrolysis trends reported previously (McBride, 1989; Stumm and Morgan, 1996).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The association of Cu, Pb, and Zn with the operationally defined amorphous oxide, crystalline oxide, and organic fractions was linearly dependent on the abundance of each respective geochemical phase. The linear nature of this relationship suggests that the binding capacity of these phases for Cu, Pb, and Zn was not exceeded. By quantifying this relationship it was possible to assess the stoichiometric affinity of each metal for a given fraction. This approach is useful because (i) it helps to explain the observed trends in the geochemical fractionation, and (ii) it allows the differences between metal affinities for a given fraction to be quantified.
Future work will focus on quantifying metal partitioning in oxic, estuarine sediments under controlled, laboratory conditions. It is anticipated that such work will assist in our understanding of the partitioning behavior of trace metals in field situations.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in JEQ:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
