2011 Ground Water Summit and 2011 Ground Water Protection Council Spring Meeting

Fate and Transport of Carbamazepine and Sulfamethoxazol in Natural Aquifer Sediment and an Artificial Palygorskite-Clay Filter

Monday, May 2, 2011: 1:50 p.m.
Constellation D (Hyatt Regency Baltimore on the Inner Harbor)
Tedros Berhane, Miami University;
Jonathan Levy, Miami University;
Mark Krekeler, Miami University;
John Hawes, Miami University;

Abstract

 Carbamazepine, an anticonvulsant and analgesic, and the antibiotic, sulfamethoxazol, are emerging contaminants with potentially harmful ecosystem and human health impacts. Of all the many US rivers sampled in a recent study, the Great Miami River (GMR) in southwest Ohio had the largest number and the greatest concentrations of pharmaceuticals and endocrine-disrupting compounds. The GMR aquifer is a sole-source aquifer utilizing induced infiltration of GMR water for communities between Dayton and Cincinnati.  Two important questions arise: to what extent does riverbank filtration attenuate these contaminants are there inexpensive technologies that might help prevent such river contamination in the first place thereby protecting ecosystem as well as human health?  In this study, sorption and laboratory-transport experiments were performed on two media: glacial-outwash from the GMR aquifer and Florida palygorskite clay (PF1) known for its high sorption capacity for a wide variety of compounds.  PF1 was investigated as a potential filter material that could be used at sewage treatment plants. The outwash was investigated to assess the efficacy of riverbank filtration on the attenuation of carbamazepine and sulfamethoxazol before groundwater production.

      Water was analyzed with high performance liquid chromatography coupled with double mass spectrometry.  Sorption experiments developed sorption isotherms under a variety of pH and temperature conditions.  Intact sediment cores and palygorskite granules were incorporated into columns for pulse-source transport experiments.  Hydraulic conductivity of all columns was determined and bromide was used as a conservative tracer to determine sediment dispersivity and effective porosity.  Carbamazepine and sulfamethoxazol were then injected and column effluent was collected over time.  Observed breakthrough curves were simulated with a 1-D advection-dispersion-reaction model and, using the sorption results, appropriate transport parameter values were determined through calibration. Laboratory-derived values were used to make predictions made at the field scale. Predictions will be compared to actual field data in the future.