SEASONAL VARIATIONS IN THE CHEMICAL SPECIATION OF COPPER ALONG THE HURON RIVER

 

Imar Mansilla-Rivera and Jerome O. Nriagu

Department of Environmental Health Sciences, School of Public Health

The University of Michigan, 109 S. Observatory St.

Ann Arbor, MI 48109-2029, USA

Email:  imar@umich.edu

 

Abstract

The industrial development in the Huron River watershed has increased the levels of copper in the area and this is of major concern for 445,000 inhabitants of southeastern Michigan that are affected by the water quality of the river.  A combination of anodic stripping voltammetry and atomic absorption spectrometry methods was used to study the chemical speciation of copper along the Huron River in two different seasons of the year.  The concentrations are found to be above known toxic thresholds for sensitive organisms in lotic environments and surface runoff seems to be a major factor determining the copper distribution along the river.  Two types of voltammograms were obtained, which can be attributed to differences in organic material excreted by phytoplankton after a season of high biological productivity.  These results suggest that the strong complexation in the Huron River ecosystem can reduce the risk of copper toxicity to aquatic organisms. 

 

Introduction

            While copper speciation in marine and estuarine environments has received considerable attention, studies in freshwater systems are limited.  This can be explained by the fact that a major portion of ligands which complex with copper (such as humic acids) are not well characterized.  Equally important is the fact that many cycling processes affecting the complexation of copper in freshwater environments are quite different from those in marine systems.  Therefore, there is a profound need for the determination of copper speciation in freshwater ecosystems. 

The Huron River was selected as a freshwater ecosystem model for study.  It is of major importance in the area because it provides water for drinking, agriculture, and industry (Hay-Chmielewski et al. 1995).  It is also used for recreational purposes and for hydroelectricity generation for towns and cities in the area.  In addition, the Huron River is a critical habitat for a variety of organisms, especially mammalian and bird species (Hay-Chmielewski et al. 1995).  Therefore, the water quality of this river affects, in major portion, the lives of many people, as well as those of other living organisms in the area.  Nevertheless, there is a lack of information on chemical speciation as well as on transport or fate of trace metals in this aquatic system. 

Given the importance of the Huron River in southeastern Michigan and the impact that industries have on this river, there was a critical need for conducting such an investigation in this area.  This study of copper in the Huron River is of great interest because it can provide a basis for comparison with other reported data in similar environments and helps to clarify some aspects of the bioavailability of copper in this system. 

Methods

            Sampling, storage, and handling process of all samples were conducted using ultra clean techniques to minimize contamination.  The sampling sites were chosen so that differences presented by the characteristic longitudinal zonation of lotic environments could be measured.  Wetland, lake, and stream samples were collected during March and April of 1999 (the spring) and in early November of the same year (the fall).  There were nine stations chosen:   Flatrock, Lower Huron Metropark, Belleville Lake, Ford Lake Park, Gallup Park, wetland, Delhi Metropark, Evergreen Rd, and Kent Lake;  and these were located at approximately 8, 25, 40, 45, 60, 65, 80, 110, and 130 Km from the mouth of the river, respectively.

            Once the samples were collected, they were transported to a clean laboratory and filtered through polycarbonate membranes (Nucleopore or Millipore) with 0.4 mm pore size.  Then they were split into acidified and untreated fractions.  Both fractions were kept in a refrigerator at 4 oC until analysis was conducted (within six weeks after collection).  Dissolved copper was determined from the filtered and acidified (0.2 % HNO3, final concentration) fraction using a Perkin-Elmer 4100ZL graphite furnace atomic absorption spectrometer (GFAAS).  The particulate concentration of copper was obtained by dissolving the metal residue from the membranes used to filter each sample and they were analyzed with the atomic absorption spectrometer described before.

Differential pulse anodic stripping voltammetry (DPASV) was used to determine the concentration of labile copper present in the samples.  It was performed by using a hanging mercury drop electrode, a silver/silver chloride reference electrode, and a platinum wire counter electrode with a Princeton Applied Research 394 analyzer.  A final concentration of 0.02 M KNO3 was used as supporting electrolyte.  To minimize the time of analysis without jeopardizing its sensitivity, a 540-s deposition step at –0.7 V was used.  Labile copper was measured on samples that had been spiked with a copper standard solution and were left overnight to reach equilibrium.  Since labile copper may be related to bioavailable copper, a mathematical formula previously derived (Wu 1998) was used to estimate the fraction of copper that may be bioavailable to organisms.

 

Results and Discussion

            All results are presented in Table 1.  The average concentrations of total dissolved copper in the Huron River for the spring and fall samplings were 20 ± 3 nM (1.3 ± 0.2 ppb) and 47 ± 24 nM (3.0 ± 1.5 ppb), respectively.  The highest levels were found in both samplings at stations close to the mouth of the river.  This may indicate that copper is being transported downstream in the dissolved load of the river.  However, no specific trend of copper levels was found along the river.  This oscillation may be the result of surface runoff from the urban areas located in the watershed. 

The most significant difference between the spring and fall samplings is the considerable increase in dissolved copper during the fall for all nine stations.  High suspended solids in the spring adsorptively removes the dissolved copper and may be responsible for the observed difference in total dissolved copper concentrations between seasons.  While the concentration of copper along the river was fairly stable during the spring, in the fall they were very variable.  Since baseflow conditions are reached in this season, the impact that tributaries and point source discharges have on the river are more significant.  Mill Creek (which drains 144 mi2 in agricultural use), the Ann Arbor Waste Water Treatment Plant, and the industries and airports located upstream from the Lower Huron Metropark, may be responsible for the high values of dissolved copper observed in Delhi Metropark, Ford Lake, and Lower Huron Metropark, respectively.

The overall average concentration of particulate copper in the river decreased from 5 ± 3 nM during the spring to 3 ± 1 nM in the fall.  This reduction in the particulate levels of copper during the fall was observed in most of the stations (Belleville Lake and Lower Huron Metropark were the exceptions).  During spring runoff, the river is very turbid and most of the copper should be in particulate form because of higher suspended solid concentration and its consequent increased adsorption of dissolved copper.  Similarly, the ratio of particulate to total copper is significantly higher during the spring in most of the samples (19%) (Belleville Lake is the only exception) than it is in the fall (6 %).  Comparable ratios of particulate to total copper are observed in other rivers in Michigan (Hurley et al. 1996). 

Two types of voltammograms were obtained when conducting the titrations of the Huron River samples.  For the purposes of discussion, we will refer to those as “type A” and “type B”.  The type A voltammogram presents only one reoxidation peak and it was observed in all samples taken in the spring and three of the samples (Delhi Metropark, Ford Lake Park, and Belleville Lake) collected in the fall.  On the other hand, the type B voltammogram presents two reoxidation peaks and was primarily found in the majority of the samples collected in the fall. 

In the type A voltammogram the peak for the reoxidation of copper occurs at a variable potential (from -0.130 V to -0.180 V), depending on the conditions of each sample.  However, it is important to mention that this variation in potential is only observed between samples and not within samples, implying that the voltammogram is a characteristic feature of the samples analyzed rather than analytical artifact.  Variation in potentials within samples was always less than 5 %.  In contrast, in the type B voltammogram there is one peak that starts increasing at a less negative potential (from -0.078 V to -0.100 V).  As copper continues to be added, this peak stops increasing and a separate peak at a more negative potential appears (from -0.146 V to -0.164 V).  This more negative peak is the one that shows a linear relation with the concentration of copper added.  As in the type A voltammogram, variations in potentials within samples were always less than 5 %.

We attribute the first peak appearing in the type B voltammogram (the less negative one) to the reoxidation of a copper(I) complex adsorbed on the surface of the electrode.  The stabilizing ligand responsible for the formation of this complex can be of biogenic origin since the double peak response was observed in samples collected during the fall, after a high biological productivity season.  The second peak can be a result of two different processes:  the reoxidation of Cu(0) to Cu(II), or that of Cu(0) to Cu(I).  Although the order of the peaks in this study is similar to the case observed in seawater (Boussemart et al. 1993), we believe it is unlikely for copper to be stabilized as Cu(I)Cl2- (as it is stabilized in seawater) in the Huron River.  However, the high concentrations of humic and fulvic acids (which contain sulfhydryl groups) in the samples may prevent the further reoxidation of Cu(I) to Cu(II).  These ligands are produced from many phytoplankton species and have been detected in natural waters at relatively high levels (Croot et al. 1999).  Other possible Cu(I) stabilizing agents are compounds with N donor ligands, which form stronger complexes with Cu(I) than with Cu(II) if the complex provides a more stable stereochemistry (Croot et al. 1999).  Therefore, it is also conceivable that the organic compounds involved are of industrial origin since the river is highly polluted.  However, the fact that the potential at which this more negative peak appears is very similar to the potential at which the peak in the type A voltammogram occurs may suggest that the reoxidation is complete from Cu(0) to Cu(II), although the presence of organic ligands in the sample cause a shift on the potential at which the oxidation peak appears.

The errors of the complexation capacity, CL, and log of the stability constant, K’, were estimated to be approximately 12 %.  The calculated values of CL or log K’ did not present any definite spatial trend in any of the seasons.  The complexation capacity increased significantly in the fall at the Evergreen Rd., Ford Lake Park, and Lower Huron Metropark stations.  This increase might be a result of a higher biological productivity in these stations during the fall, which suggest that the copper complexation capacity at these particular sites in the Huron River is mainly produced by biota in the area.  Otherwise, the values were similar in both seasons.  The stability constants remained fairly constant in most of the samples throughout the time of the study.  The fraction of bound copper initially present in all samples is very high (> 97 %) in both seasons.  This strong complexation may reduce the risk of toxicity to organisms living in the Huron River ecosystem. 

 

Table 1.  Concentrations and complexation parameters of copper in the Huron River

 


 

 


References

Boussemart, M, Menargues, L, Benaïm, JY (1993), Electroanalysis 5: 125-133.

Croot, PL, Moffett, JW, Luther, GW (III) (1999), Mar. Chem. 67: 219-232.

Hay-Chmielewski, EM, Seelbach, PW, Whelan, GE, Jester, DB Jr. (1995), Huron River Assessment, Fisheries Division Special Report No. 16.  State of Michigan, Department of Natural Resources, Lansing.

Hurley, JP, Shafer, MM, Cowell, SE, Overdier, JT, Hughes, PE, Armstrong, DE (1996), Environ. Sci. Technol. 30: 2093-2098.

Wu, Y (1998), Voltammetric studies on copper, lead, and cadmium speciation in the Great Lakes.  Ph.D. Thesis.  The University of Michigan, Ann Arbor.