Mineral precipitation and hydrochemical evolution through evaporitic processes in soda brines (East African Rift Valley)

Abstract

Soda lakes of the East African Rift Valley are hyperalkaline, hypersaline lakes extremely enriched in Na+, K+, Cl−, CO32−, HCO3−, and SiO2. In this paper, we investigate the chemical evolution in these lakes and the production of chemical sediments by salt precipitation via evaporation. Water samples from tributary springs and three lakes (Magadi, Nasikie Engida and Natron) have been experimentally studied by in-situ X-ray diffraction during evaporation experiments to characterize the sequence of mineral precipitation. These data are complemented by ex-situ diffraction studies, chemical analyses and thermodynamic hydrochemical calculations producing detailed information on the activity of all solution species and the saturation state of all minerals potentially generated by the given composition. Major minerals precipitating from these samples are sodium carbonates/bicarbonates as well as halite. The CO3/HCO3 ratio, controlled by pH, is the main factor defining the Na‑carbonates precipitation sequence: in lake brines where CO3/HCO3 > 1, trona precipitates first whereas in hot springs, where CO3/HCO3 ≪ 1, nahcolite precipitates instead of trona, which forms later via partial dissolution of nahcolite. Precipitation of nahcolite is possible only at lower pH values (pCO2 higher than −2.7) explaining the distribution of trona and nahcolite in current lakes and the stratigraphic sequences. Later, during evaporation, thermonatrite precipitates, normally at the same time as halite, at a very high pH (>11.2) after significant depletion of HCO3− due to trona precipitation. The precipitation of these soluble minerals increases the pH of the brine and is the main factor contributing to the hyperalkaline and hypersaline character of the lakes. Villiaumite, sylvite, alkaline earth carbonates, fluorapatite and silica are also predicted to precipitate, but most of them have not been observed in evaporation experiments, either because of the small amount of precipitates produced, kinetic effects delaying the nucleation of some phases, or by biologically induced effects in the lake chemistry that are not considered in our calculations. Even in these cases, the chemical composition in the corresponding ions allows for discussion on their accumulation and the eventual precipitation of these phases. The coupling of in-situ and ex-situ experiments and geochemical modelling is key to understanding the hydrogeochemical and hydroclimatic conditions of soda lakes, evaporite settings, and potentially soda oceans of early Earth and other extraterrestrial bodies.