2014 GSA Annual Meeting in Vancouver, British Columbia (19–22 October 2014)

Paper No. 157-3
Presentation Time: 1:30 PM


VINSON, David S.1, LUNDY, James R.2, ALVAREZ, Tom2 and VENGOSH, Avner3, (1)Department of Geography & Earth Sciences, University of North Carolina at Charlotte, 9201 University City Blvd., McEniry 324, Charlotte, NC 28223, (2)Minnesota Department of Health, 625 North Robert St, P.O. Box 64975, St. Paul, MN 55164, (3)Division of Earth and Ocean Sciences, Nicholas School of the Environment, Duke University, Durham, NC 27708, dsvinson@uncc.edu

Radium (Ra) differs from other naturally-occurring trace elements of health significance (e.g. As) by its behavior - a balance of radioactive and chemical processes. Here we report results of an investigation conducted during a pilot aquifer storage and recovery study in the Cambrian Mt. Simon sandstone of Minnesota (USA), an aquifer known for elevated levels of naturally-occurring Ra in groundwater. 63 × 106 gal (2.4 × 105 m3) of water were injected during ~4.5 months and stored for ~3 months after injection ended. Recovery of 52 × 106 gal (2.0 × 105 m3) over ~2 months resulted in storage time of 3-9 months.

The recovered water resembled injected water in terms of nonreactive ions such as Cl-, implying only minor mixing of native groundwater and injected water. However, Ra in recovered water exceeded what would result from mixing of injected water (226Ra+228Ra < 2 pCi/L) with native groundwater (226Ra+228Ra > 6 pCi/L). During recovery, short-lived 224Ra increased to a stable value within hours. Long-lived 226Ra and 228Ra increased gradually during recovery (0.7-2.2 and 1.0-2.3 pCi/L, respectively).

Here, Ra is influenced mainly by alpha recoil (the release of Ra from α-decay of solid-phase Th) and/or chemical mechanisms of radium release and removal. Alpha recoil: The rapid increase of 224Ra, and 224Ra/228Ra ratios >1 after the first 2 hours of recovery (1.5-2.6), are consistent with alpha recoil. 224Ra/228Ra >1 occurs because 224Ra is replaced by α-recoil more rapidly than the longer-lived 228Ra. 228Ra/226Ra decreased during the recovery period (1.8-1.0), approaching the ratio of native groundwater and apparently approaching steady-state activities with increasing storage time. Chemical mechanisms: Concentrations of barium (Ba) were higher in injected water (0.19-0.24 mg/L) than in native groundwater (<0.1 mg/L). As Ra increased during recovery, Ba concentrations decreased (0.21-0.14 mg/L). Barium in injected water (present at 108-times higher concentration than Ra) may have competed with Ra for adsorption sites. Such competition, if occurring, could affect the efficiency of Ra adsorption after its release into water by α-recoil. These results imply that the chemistry of injected water and native groundwater - and the duration of storage, even if only a few months - should be evaluated during aquifer storage.