SPECIATION AND DISTRIBUTION OF ARSENIC
IN THE MAHOMET AQUIFER, ILLINOIS
Thomas R. Holm, Walton R. Kelly, Steven D. Wilson, George S. Roadcap
Illinois State Water Survey
Jonathon Talbott, John Scott
Illinois Waste Management and Research Center
Abstract
The Mahomet Aquifer in central Illinois is one of the principal aquifers in the state. It is the source of drinking water for many communities, including Champaign-Urbana, and thousands of private homes. This study was designed to characterize the distribution and speciation of arsenic in two areas of the aquifer. A reliable, robust method for arsenic speciation in groundwater was developed based on separation of As(III), As(V), monomethylarsonic acid (MAA), and dimethylarsinic acid (DMAA) by HPLC or of As(III) and As(V) by IC with ICP-MS detection. The mobile phase was chosen not only to separate the As species but also to avoid precipitation of CaCO3 and Fe(OH)3. Groundwater samples were collected from approximately 50 wells in Tazewell County, an area where arsenic has been found in many domestic wells, and northwest Champaign County, an area for which very little arsenic data was available. Roughly equal numbers of wells were finished near the bottom, middle, and upper part of the aquifer. A few of the wells were finished in the Glasford Aquifer, which lies above the Mahomet. The samples were analyzed for total arsenic, arsenic species, metals, anions, alkalinity, ammonia, and organic carbon. In both counties, wells with arsenic concentrations above the U. S. E. P. A. maximum contaminant level (MCL) of 10 micrograms per liter (ppb) (133 nM) were spread throughout the study area.
In Champaign County, 40% of the wells had nondetectable (<1 ppb) arsenic, 90% had less than the MCL, and one well had over 50 ppb. In Tazewell County, 25% of the wells had nondetectable arsenic, 55% had less than the MCL, and 10% had over 50 ppb. In both counties the spatial distribution of arsenic was complex. Wells with high concentrations were often located less than one mile from wells with nondetectable arsenic. In Tazewell County, the percentages of wells with arsenic concentrations above the MCL were roughly equal in the shallow, intermediate, and deep parts of the Mahomet Aquifer. A higher percentage of samples from the Glasford Aquifer exceeded the MCL, but there were relatively few samples. In most samples As(III) made up over 90% of the total dissolved arsenic with the balance being As(V). There was no detectable MAA or DMAA. Particulate arsenic (retained by a 0.45 micro-m filter) made up less than 10% of the total in unfiltered samples. The arsenic speciation was consistent with thermodynamic calculations based on the measured pH and oxidation-reduction (redox) potential values. Conversely, redox potentials calculated from arsenic speciation agreed very well with measured potentials. Although there were no good correlations between concentrations of arsenic and of other analytes, high arsenic concentrations were associated with low sulfate concentrations and with high concentrations of bicarbonate, ammonium, fluoride, and organic carbon.