The adsorption of cyanobacterial hepatotoxins from water onto soil during batch experiments.

Public health concerns associated with cyanobacteria, both chronic and acute, arise from their ability to produce toxins. Rural communities within Australia and those in developing countries require an inexpensive and low-cost method for removing toxins from drinking water. A candidate technology is bank filtration. Adsorption of cyanobacterial hepatotoxins was measured in batch studies to determine the applicability of bank filtration as an efficient removal strategy. Five soils with different physicochemical properties were collected from regions around South Australia. The soils were mixed with either nodularin or microcystin-LR in distilled water and buffered solutions (pH 4.8, 6.9 and 9.1). Additionally, nodularin was mixed in unbuffered solutions (pH 4.8 and 8.9). The three soils with the high clay and/or organic carbon contents (Paringa A, McLaren Flat and Lakes Plains) had the higher nodularin adsorption coefficients, ranging from 0.2 to 16.59 L kg-1. Soil suspensions in acetate buffer (pH 4.8) generally produced significantly higher nodularin adsorption coefficients, when compared to the other buffer systems. The background interference from the ionic strength of the buffers, however, made interpretation of the effect of pH on toxin sorption difficult. Increases in solution ionic strength, from freshwater to seawater, resulted in corresponding increases in the nodularin adsorption coefficients for all sites, except the sandy Hallett Cove site. The implications for bank filtration are that higher water pH values and lower salinities will enhance the in situ mobility of the toxins, resulting in an increased distance of filtration through the river bank before toxin free water could be abstracted for human consumption.