Stirred flow-through reactor: A powerful tool for studying adsorption and desorption kinetics of solutes on soils

GEOC 37

Guillaume Limousin, guillaume.limousin@cea.fr1, Stéphanie Szenknect, stephanie.szenknect@cea.fr1, Véronique Barthès, veronique.barthes@cea.fr1, Jean-Paul Gaudet, gaudet@hmg.inpg.fr2, Mohamed Krimissa, mohamed.krimissa@edf.fr3, and Laurent Charlet, laurent.charlet@ujf-grenoble.fr4. (1) Tracers Technology Laboratory, Atomic Energy Commission, 38054 Grenoble Cedex, France, (2) Laboratoire d’étude des Transferts en Hydrologie et Environnement (CNRS-INPG-IRD-UJF), BP 53, 38041 Grenoble Cedex, France, (3) Division Recherche et Développement, Laboratoire National d’Hydraulique et d’Environnement - P78, Electricité de France, 6 quai Watier, 78401 Chatou, France, (4) Laboratoire de Géophysique Interne et Techtonophysique (CNRS-IRD-LCPC-UJF-Université de Savoie), BP 53, 38041 Grenoble Cedex, France
Adsorption and desorption kinetics is not always easy to study because the concentration cannot be monitored continuously in batch, while column investigations are often too long. Cobalt retention on a soil was chosen to demonstrate the suitability of the stirred flow-through reactor to characterize adsorption and desorption kinetics. Adsorption in batch showed that equilibrium seemed to be reached after 1 h for a 10-4 mol.L-1 initial concentration, while a slow reaction happened for a 3.5x10-6 mol.L-1 initial concentration. These results were confirmed by stirred flow experiments with high residence time (low flow rate): for a 10-4 mol.L-1 injected concentration, adsorption and desorption were instantaneous; while a 3.5x10-6 mol.L-1 injected concentration provided a slow adsorption reaction and a much slower desorption. This hysteretic retention was satisfactory reproduced by a two-reaction model with a fast, instantaneously reversible adsorption and a slow, first-order, irreversible adsorption. These slow reactions were strongly lowered as residence time decreased from 20 h to 1 h, but were enhanced by shifting the pH of the injected solution from 4.5 (soil pH) to 5.8. These results are consistent with the existence of two types of sites: (i) a high density of surface charges with low affinity for cobalt and low pH-dependence (e.g. outer-sphere ion exchange), and (ii) a low density of surface charges with high affinity for cobalt and higher pH-dependence (e.g. inner-sphere surface complexation). In conclusion, the stirred flow-through reactor provided consistent results compared with batch. Furthermore, continuously-monitored desorption allows to verify the existence of kinetic hysteresis. Contrary to batch techniques, the composition of the solution can be kept constant, allowing to test the influence of chemical factors (such as pH or ionic background) in open-flow conditions where the competing species are flushed out.
 

General Poster Session
7:00 PM-9:00 PM, Sunday, 10 September 2006 Moscone Center -- Hall D, Poster

Division of Geochemistry

The 232nd ACS National Meeting, San Francisco, CA, September 10-14, 2006