On the mechanism controlling fluoride concentration in groundwaters of the south of the Province of Buenos Aires, Argentina: adsorption or solubility?

High levels of fluoride in drinking water affect millions of people due to health problems associated with water consumption. Fluoride concentrations in groundwaters from the unsaturated and saturated zones of a watershed located in the Chaco-Pampean Plain of Argentina are reported and investigated to evaluate the mechanism that controls these concentrations. Two hypotheses were formulated: (a) the adsorption hypothesis, which assumes that fluoride is controlled by adsorption/desorption processes on mineral oxides, and (b) the dissolution hypothesis, which assumes that the solubility of fluorite is the controlling process. Calculations with the CD-MUSIC surface complexation model to test the adsorption hypothesis rendered very poor results. The affinity of fluoride for surface sites is so low that it is easily displaced by carbonate, and thus the calculated fluoride–carbonate trends are very different to the experimentally found with the studied water samples. On the contrary, a comparison of reaction quotients, Q, with equilibrium constants, K, for fluorite dissolution, calcite dissolution and a combination of both processes indicates that fluorite and calcite are in equilibrium or close to equilibrium with their corresponding ions in solution, giving strength to the dissolution hypothesis. This hypothesis can also explain, at least qualitatively, why fluoride correlates positively with arsenate and carbonate. The conclusions could be extrapolated to many other systems worldwide, which present these positive correlations.