The Evaporation of Caribbean Seawater: Composition of Brines, Precipitates and Fluid Inclusions

Heide Zimmermann IGDL, Universität Göttingen, Goldschmidtstr.3, D-37077 Göttingen, Germany

Christine Wehebrink IGDL, Universität Göttingen, Goldschmidtstr.3, D-37077 Göttingen, Germany


The evolution of major components during the evaporation of modern seawater was thoroughly studied by Herrmann et al. (1973), Brantley et al. (1984), and McCaffrey et al. (1987). Except Br there is little known on the distribution of trace elements between evaporating brines and precipitating minerals. During the last decade
an increasing effort was made to understand paleo seawater composition by studying fluid inclusions in halite (e.g. Horita et al., 1991; Ayora et al., 1994). But there exists only little information on the composition of fluid inclusions in modern halite. That is why we went to the solar salt production facility on Inagua, Bahamas, to investigate distribution patterns of trace components (B, Ba, Cs, I, Li, Mo, Rb, and Sr) between evaporating seawater, precipitating mineral phases and fluid inclusions trapped during this process.

Analytical Procedures

Halite samples were milled (< 125 µm) and cleaned with ethanol to remove all trapped fluid. Brines and dissolved halite samples were analyzed for Cl-, SO42-, Br-, F-, Na+, K+, Mg2+, Ca2+, and Li+ by ion chromatography (IC) using a DIONEX 500. B, Ba, Cs, I, Li, Mo, Rb, and Sr were measured with a FISONS INSTRUMENTS Plasmaquad mass spectrometer PQ2+ (ICP-MS). Fluid inclusions > 200 µm were extracted using a microdrill and a capillary with a uniform inner diameter of 25 µm. 100 µl of distilled water were added to the extracted brine and then analyzed by IC (e.g. Lazar & Holland, 1988).


The collected brines represent degrees of evaporation (D.E.) between 1 and 30. The precipitation of aragonite starts at D.E.=1.5, gypsum crystallizes at D.E.=4, and finally halite at D.E.=10. This agrees well with the studies of Brantley et al. (1984) and McCaffrey et al. (1987). Herrmann et al. (1973) observe aragonite precipitation at D.E.=4 and gypsum precipitation at D.E.=6. But the re-interpretation of their data seems to show that both minerals start to precipitate more or less simultaneously at D.E.=4. Up to D.E.=30 potassium and magnesium are the only main components that are concentrated quantitatively in the evaporating brines.

Sr, Ba, and I are removed from the brine at a D.E.of about 6. The decrease of Sr and especially Ba contents do not seem to be related to the precipitation of gypsum. It is more likely that there occurs an independant Sr phase, such as coelestite. The observed decrease of iodine contents during evporation is not yet resolved. Li, Rb, B, and Br are almost quantitatively accumulated in the evaporating brines, whereas the continuously increasing concentration of Mo reaches only 80 % of the expected value at D.E.=10.

During evaporation the molar ratios of Rb/Br (0.0025) and B/Br (0.5) remain fairly constant, so that Carribean seawater and evaporating brines plot on a distinct area in a Rb/Br - B/Br diagram. River water and seawater show very different Rb/Br and B/Br ratios. That is why NaCl saturated brines, derived from dissolution of halite either by meteoric water or by seawater, seem to keep their original imprint of the different sources of water. Thus the introduced Rb/Br - B/Br diagram could prove to be a powerful tool for distinguishing between different kinds of reservoirs and genetic origin of natural brines.

Cs, Li, Rb, Ba, I, and Mo contents of halite were below detection limit of ICP-MS. B and Br show a regular distribution pattern between brine and precipitating halite. The DBr value of 0.038 agrees well with the data of Braitsch and Herrmann (1963), Herrmann et al (1973), and McCaffrey et al. (1987).

First analyses of fluid inclusions show significantly higher K, SO4 ,and Br contents compared to the evaporating parent brine (Zimmermann & Wehebrink, 1995). Since
there is no reasonable explanation for a selective chemical fractionation during the trapping process, this might indicate that, e.g. due to rainfall, brine and trapped fluid represent different degrees of evaporation.


Ayora, C., Garçia-Veigas, J., & Pueyo, J.-J., Geochim. Cosmochim. Acta, 58, 43-55 (1994).

Braitsch, O., & Herrmann, A. G., Geochim. Cosmochim. Acta, 27, 361-391 (1963).

Brantley, S. L., Crerar, D. A., Møller, N. E., & Weare, J. H., J. Sed. Petrology, 54, 447-462 (1984).

Herrmann, A. G., Knake, D., Schneider, J., & Peters, H., Contr. Mineral. Petrol., 40, 1-24 (1973).

Horita, J, Friedman, T. J., Lazar, B., & Holland, H. D., Geochim. Cosmochim. Acta, 55, 417-432 (1991).

Lazar, B., & Holland, H. D., Geochim. Cosmochim. Acta, 52, 485-490 (1988).

McCaffrey, M. A., Lazar, B., & Holland, H. D., J. Sed. Petrology, 57, 928-937 (1987).

Zimmermann, H., & Wehebrink, C., Eur. J. Mineral., submitted (1995).