The applicability and limitations of thermodynamic geochemical models to simulate trace element behaviour in natural waters. Lessons learned from natural analogue studies

摘要

Natural analogue investigations aim to understand key phenomena and processes in natural systems related to those expected to occur in radioactive waste repositories. One of the key applications of natural analogue studies has been the possibility to test the geochemical models to be used to describe the migration of radionuclides in a future radioactive waste repository system. To this end, several geochemical modelling testing exercises (commonly denoted as blind predictive modelling, BPM) have formed an integral part of these studies over the last decade. We have reviewed, discussed and compared the results obtained from geochemical modelling BPM exercises carried out within six natural analogue studies: Pocos de Caldas, Cigar Lake, Maqarin, El Berrocal, Oklo and Palmottu. To make this comparison meaningful, we present the main geochemical characteristics of each site in order to highlight the most relevant mineralogical and hydrochemical differences. The elements selected for discussion are: Sr, Ba, Sn, Pb, Se, Ni, Zn, REEs, Th and U. We have based our discussion on the results obtained from the calculated aqueous speciation as well as by comparing solubility calculations with the actually observed concentrations. Results can be differentiated into two categories of elemental behaviour: 1. those elements Re Th and U under reducing conditions that can be fairly well described by assuming solubility control exerted by pure solid phases as their oxyhydroxides; 2. elements such as Sr, Zn, REEs and U under oxidising conditions for which the association to major geochemical components of the system must be considered in order to explain their concentrations in groundwaters. Additionally, we discuss the main improvements made to the thermodynamic databases and the geochemical calculation methodologies due to the BPM exercises. Furthermore, the most important characterisation geochemical data needed to complete predictive solubility and speciation calculations are identified. (C) 2002 Elsevier Science B.V. All rights reserved. [References: 70]