Research on naturally-derived bedrock contaminants (e.g. arsenic, uranium, alpha radiation) requires integration of numerous analytical approaches in order to understand the concentrations of the trace elements in different rock types as well as their speciation (i.e. what mineral is the source of the arsenic?) and behavior in aquifers (e.g. what controls the release and migration of arsenic from its source mineral into groundwater?).  In Vermont, approximately 40 % of residents obtain drinking water from bedrock water wells and emerging evidence suggests that some areas of the state are prone to elevated arsenic, uranium, alpha radiation and other bedrock-derived trace elements; in particular, previous research by Middlebury College students and faculty in collaboration with the Vermont Geological Survey has revealed useful information on the spatial distribution and general geology and geochemistry of these trace elements.  For example, elevated arsenic occurs in slates of the Taconic region of SW Vermont but appears to be absent in higher-grade phyllites (Thompson, 2011, Middlebury College thesis), and uranium is associated with phosphate-rich dolomites and phyllites in NW Vermont, but not in most limestones (McDonald, 2012, Middlebury College thesis). What has been lacking is precise data on the speciation of the trace elements, and this has hindered interpretation and prediction of their abundance in groundwater.

The current project has focused on integrating X-ray diffraction (XRD), scanning electron microscopy-energy dispersive spectrometry- wavelength dispersive spectrometry (SEM-EDS-WDS), Fourier transform infrared analysis (FTIR) and inductively coupled plasma-atomic emission spectrometry-mass spectrometry (ICP-AES-MS)  to better understand mineralogical/geochemical/hydrochemical relationships associated with bedrock-derived trace elements.  Figure 1 provides an example of how some of these methods have been used to study arsenic in the Taconic bedrock aquifer of SW Vermont.

Combined with ICP-AES-MS, results shown in Figure 1 indicate that (1) arsenic is disseminated in pyrite, but does not occur in arsenic sulfides (speciation which is distinct from zinc, copper and other trace metals, which are observed to occur in ZnS, CuS, e.g.), and (2) that arsenic concentration in pyrite decreases significantly (p < 0.05) with increasing metamorphism across a low-grade shale-slate-phyllite region (Studwell, 2013, Middlebury College thesis). This tells us that elevated arsenic in bedrock wells should occur primarily in shales and slates (but not in phyllites), data which is consistent with well water analysis, and that the primary controls on release of arsenic to groundwater will be dissolution of pyrite.  This latter observation agrees with groundwater chemistry (Ryan et al., in review, Applied Geochemistry).  

We have also used a combination of these approaches to study speciation of arsenic (Figure 2) in the bedrock aquifer system of north-central Vermont, where our findings indicate that the dominant source of arsenic in groundwater is serpentinite, specifically the minerals magnesite and antigorite (Ryan et al., 2011, Applied Geochemistry 26, 444-457).  

In a study of elevated uranium and alpha radiation in NW Vermont (McDonald, 2012, Middlebury College thesis), the combination of XRD, SEM and ICP revealed that the source of uranium and alpha in dolostones of the Clarendon Springs Formation is fluoroapatite and the resulting geochemical-geological model is helping to understand and predict distribution of U and alpha across the region.

Additional ongoing research objectives of the project include application of these instrumental methods to research on tropical soil mineralogy and geochemistry (Hobbs, 2012, Middlebury College thesis; Ryan and Huertas, in rev...

Please report errors in award information by writing to: awardsearch@nsf.gov.