Abstract

The link between environmental exposure to aluminium, mammalian toxicity and agricultural productivity has focused attention on the relationship between the environmental geochemistry of Al and its potential toxicity. Recent work has questioned the accuracy of previously reported baseline levels of aluminium in tropical well waters. In this work we investigate the validity of these observations and their influence on criteria for the determination of water quality.

Data are presented for two contrasting study areas in Uganda in which c. 50 water samples were collected from a variety of water sources and subsequently analysed to identify concentrations of dissolved Al and of ligands that potentially control its environmental mobility and toxicty. Determinands included: pH, temperature, redox potential, conductivity, alkalinity, major and trace cations and anions and TOC.

Determination of Al by ICP-AES before and after sonication confirmed that sonication of samples filtered to 0.40 μm during sampling resulted in increased Al concentrations in solution. This effect is considered to be consistent with the aggregation of a non-labile colloidal Al on the addition of acid and confirms observations made by other researchers in Malawi. However, on-site analysis of samples using pyrocatechol violet indicated negligible free or labile Al present in the sampled waters and suggests that the majority of Al undetected by routine ICP-AES analysis is present in a non-labile form of relatively low potential toxicity.