EXPERIMENTAL INVESTIGATION OF CEPHAPIRIN ADSORPTION TO SANDS: IMPLICATIONS FOR TRANSPORT OF ANTIBIOTICS IN GROUNDWATER

The search for trace amounts of pharmaceuticals in surface and groundwater is a hot field in hydrogeology, environmental science and analytical chemistry. Antibiotics in natural waters is a rapidly-growing concern among scientists, engineers and policy makers, because of the potential for inducing antibiotic-resistance in strains of pathogenic bacteria. While antibiotic contamination is being detected in watersheds across the nation, there is an emerging need for experimental data on the fate and transport of these compounds in various aquifer materials.

This presentation reports the results of batch experiments on the adsorption of cephapirin (a widely-used veterinary antibiotic) on various size sands. Two types of sand were investigated. The first type was nearly pure quartz filter sands separated into 14 sieved fractions ranging from 3.35 mm to 0.30 mm diameter, with soil organic matter (SOM) ranging from 0.078-0.395 wt.% The second type was a Lake Michigan dune sand separated into 4 sieved fractions ranging from less than 0.30 mm to 0.06 mm diameter, with SOM contents from 0.356-0.745 wt.% This sand consisted of at least 8 common rock-forming minerals, but was dominated by sub-equal proportions of quartz, carbonate and iron oxides. Aqueous solutions were buffered by the sand to a near neutral pH and cephapirin concentrations were determined by LC/MS.

Linear adsorption distribution coefficients (Kd) for the filter sands were determined to be from 0.21 to 0.76 L/kg for sands 0.60 mm diameter and larger, with several Kd's statistically indistinguishable. Statistically different Kd's range from 1.95 to 1.31 L/kg for sands 0.50mm to 0.30 mm diameter. Kd's for dune sands (diameters less than 0.30 to 0.06 mm) range from 1.02 to 3.83 L/kg, with small standard deviations. These values are 2-3 orders of magnitude smaller than Kd's determined in previously published studies of tetracycline adsorption to clays. Our preliminary data suggest that an adsorption threshold may exist in sands about 0.6mm diameter. At grain sizes less than the threshold, adsorption is related to grain size; at grain diameters larger than the threshold, adsorption is relatively unaffected by particle size. Further experiments are underway to test this hypothesis and the possibility that non-linear equations may describe the data more accurately.