Research

The Distribution of Trace Elements Co and Pb at the Oxic-Anoxic Transition of a Stratified Lake:
Analytical Speciation and Modeling

Research	Resume	Publications	Teaching	Opportunities	Group	Geochemistry	Home

Abstract

This study presents an investigation of the biogeochemical processes regulating the distribution of trace elements at the oxic-anoxic transition of a stratified lake. The chemical speciation of Co and Pb was determined in the water column of Paul lake (MI). The research focused on the interactions of metals with hydrous oxides of manganese and iron using the natural environment as the laboratory.

Biological processes are responsible of major chemical changes in the water column. O_2(aq), SCO₂, and pH profiles reflect the combined effect of photosynthesis and mineralization of organic matter. Sulfate disappears below the oxycline and is accompanied by the formation of sulfides, but the precipitation of metal sulfides was not observed. Manganese oxides (MnO_x) and hydrous iron oxides (Fe_p) form two distinct peaks in the mixolimnion and in the suboxic waters, respectively. Mn and Fe particles were characterized by TEM. MnO_x form manganese crusts around bacteria, while Fep forms complex entities with fibrils of encapsulated polysaccharides (EPS).

The analytical speciation of Co at the microparticle level and in bulk water samples shows that the cycling of Co is regulated by MnO_x. There is a concurrent remobilization of Mn²⁺ and Co²⁺. The Co:Mn ratio obtained on individual microparticle is rather constant (~ 2%), and is greater than that of the dissolved species demonstrating that Co is preconcentrated in the solid phase.

The analytical speciation of dissolved Pb shows that it is completely complexed in the hypolimnion, probably by an organo-sulfide ligand. TEM analyses and batch reactor experiments show that Pb is scavenged by an Fep-EPS entity. The latter experiment demonstrates that Pb is removed by co-entrainment during the oxidation of iron rather than by adsorption subsequent to particulate iron formation.

A dynamic model was built to predict the distribution of Fe and Pb in the water column. It was based on a mass balance approach depicting transport and reaction. Simulations indicate that the processes regulating the cycling of Pb depend on the good understanding of the mechanisms affecting the distribution of iron. This exercise confirms the importance of good observations and field data to fully characterize the particulate and dissolved phase and to predict the chemical distribution of trace elements in aquatic systems.

Characteristics of the Sampling Site

Paul lake (MI) is located in UNDERC (46^o13' N; 89^o32' E). Take a look at few pictures of the site and facilities.

Principal Findings

Paul Lake is extremely well stratified: its hypolimnion may be permanently anoxic

Chemical signatures (e.g., O₂, pH, Alkalinity, SCO₂, DOC, POC, SH_4,SiO₄, Na⁺, K⁺, Mg²⁺, Ca²⁺, Cl^-, NO₃^-, SO₄^2-, PO₄^3-, NH₄⁺, Mn(II), Mn_p, Fe(II), Fe_p, SH₂S) of principal biogeochemical processes are pronounced and stable

Manganese oxides form crusts around bacteria above the chemocline

Hydrous iron oxides form complex aggregate of amorphous iron oxides with encapsulated polysaccharides below the oxic-anoxic transition

The distributions of Co and Pb follow two different cycles:

- Co is linked to the manganese cycle (Lienemann et al., 1997)
- Pb is regulated by the iron cycle and is complexed by a dissolved organo-sulfide ligand (Taillefert et al., 2000)

The analytical speciation of Co and Pb using different techniques (e.g., TEM-EDS, voltammetry, ion exchange chromatography (Taillefert and Gaillard, 1999), ICP-MS, GFAAS) provides with a clear picture of the cycling of these species.

Thermodynamic calculations, using surface complexation models, could not predict the distribution of trace elements among the dissolved phase and hydrous iron oxides in the water column. This indicates that it is extremely difficult to predict the fate of trace elements in natural systems using parameters derived from laboratory experiments performed with synthetic phases, because the models proposed by these experiments consider that the reacting surfaces are already present in solution.

A transport and reaction model was built (Taillefert and Gaillard, 2000) to predict the distribution of Pb among the hydrous iron oxides and the dissolved phases using co-entrainement kinetics determined in batch reactor experiments. These calculations clearly show that the processes regulating the cycling of Pb in the water column are more complex than generally admitted.

Acknowledgments

This research was performed under the direction of Dr. J.-F. Gaillard at Northwestern University, and in collaboration with Dr. C.-P. Lienemann and Dr. D. Perret from the University of Lausanne (Switzerland). This project was sponsored by a NSF grant (BES 93-09349). I would like to thank all the people who participated in the field studies: E. Rose, B. Greidanus, C. Féliers, and Dr. L. Méjanelle.