Assessing the fate of radioactive nickel in cultivated soil cores Se ´ bastien Denys a, b, * , Guillaume Echevarria a , Louis Florentin a , Elisabeth Leclerc b , Jean-Louis Morel a a Laboratoire Sols et Environnement, UMR 1120 ENSAIA-INPL/INRA, 2, av. de la Foreˆt de Haye, BP.172, 54 505 Vandoeuvre-le`s-Nancy, France b Andra, Agence Nationale pour la gestion des de´chets radioactifs,1-7 rue Jean Monnet, 92 290 Chatenay-Malabry, France article info Article history: Received 21 October 2008 Received in revised form 19 May 2009 Accepted 22 June 2009 Available online 25 July 2009 Keywords: Risk assessment Nickel-63 Distribution coefficient Concentration ratio Undisturbed soil cores Maize Water drainage abstract Parameters regarding fate of 63 Ni in the soil–plant system (soil: solution distribution coefficient, K d and soil plant concentration ratio, CR) are mostly determined in controlled pot experiments or from simple models involving a limited set of soil parameters. However, as migration of pollutants in soil is strongly linked to the water migration, variation of soil structure in the field and seasonal variation of evapo- transpiration will affect these two parameters. The aim of this work was to explore to what extent the downward transfer of 63 Ni and its uptake by plants from surface-contaminated undisturbed soil cores under cultivation can be explained by isotopic dilution of this radionuclide in the pool of stable Ni of soils. Undisturbed soil cores (50 cm  50 cm) were sampled from a brown rendzina (Rendzic Leptosol), a colluvial brown soil (Fluvic Cambisol) and an acidic brown soil (Dystric Cambisol) using PVC lysimeter tubes (three lysimeters sampled per soil type). Each core was equipped with a leachate collector. Cores were placed in a greenhouse and maize (DEA, Pioneer Ò ) was sown. After 44 days, an irrigation was simulated at the core surfaces to supply 10 000 Bq 63 NiCl 2 . Maize was harvested 135 days after 63 Ni input and radioactivity determined in both vegetal and water samples. Effective uptake of 63 Ni by maize was calculated for leaves and kernels. Water drainage and leaching of 63 Ni were monitored over the course of the experiment. Values of K d in surface soil samples were calculated from measured parameters of isotopic exchange kinetics. Results confirmed that 63 Ni was strongly retained at the soil surface. Prediction of the 63 Ni downward transfer could not be reliably assessed using the K d values, since the soil structure, which controls local water fluxes, also affected both water and Ni transport. In terms of 63 Ni plant uptake, the effective uptake in undisturbed soil cores is controlled by isotope dilution as previously shown at the pot experiment scale. Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved. 1. Introduction Chemical and radioecological human health risk assessments have numerous common assumptions (Shaw, 2005). Among them, the key role played by the soil–plant compartment in overall human exposure is of great importance. In the framework of radi- oecological risk assessment, a current conceptual scheme in the case of deep underground repositories of radioactive waste considers the input of radionuclides to soils either through a discharge of radionuclides in rivers and subsequent flooding of soils or through soil irrigation using radionuclide-contaminated groundwater. Fluxes of radionuclides in the soil–plant system are closely related to their phytoavailability, i.e. the fraction of soil pollutants taken up by the plants. Soil: liquid partitioning coeffi- cients (K d ) estimate the pool of the pollutant which is theoretically available for plant uptake (Sheppard et al., 1992; Vandenhove et al., 2007). Transfer factor (TF), Concentration Ratio (CR) or Bio- concentration Factor (BCF), i.e. the ratio of the concentration of pollutant in plant tissues over its concentration into the soil, and Effective Uptake (EU), i.e. the ratio of the quantity of radionuclides taken up by plants over the total amount introduced into the soil, are parameters used to quantify plant uptake (Ng, 1982; Sheppard and Evenden, 1990; Gerzabek et al., 1994; Echevarria et al., 1998; Denys et al., 2002; Vandenhove and Van Hees, 2007). Definition and calculation of K d and CR values assume equilibrium of the radionuclide between the compartments (Sheppard and Evenden, 1991). Values for these parameters, as well as their variability related with soil or plant factors, are often determined in labora- tory-controlled conditions using model materials such as sieved soil samples and uniform soils contamination (IUR, 1989; Albrecht et al., 2002; Vandenhove et al., 2007). In this case, the conditions simulated in these experiments can be far from the contamination event (irrigation at the soil surface or soil flooding) considered in the actual radioecological risk assessment. In particular, heteroge- neity due to anisotropic conditions within a soil profile is not * Corresponding author. Present address: INERIS, Parc Technologique Alata, BP 2, 60 550 Verneuil en Halatte, France. Tel.: þ33 3 55 44 61 89; fax: þ33 3 55 44 65 56. E-mail address: Sebastien.Denys@ineris.fr (S. Denys). Contents lists available at ScienceDirect Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad 0265-931X/$ – see front matter Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvrad.2009.06.019 Journal of Environmental Radioactivity 100 (2009) 884–889