WM’01 Conference, February 25-March 1, 2001, Tucson, AZ EXPERIMENT-BASED PROGNOSTICATION OF WASTE BOROSILICATE GLASS BEHAVIOR IN A LOAMY SOIL OF THE WET REPOSITORY SITE M.I. Ojovan, N.V. Ojovan, I.V. Startceva, G.N.Chuikova, A.S. Barinov Scientific and Industrial Association ‘Radon’, the 7-th Rostovsky Lane 2/14, Moscow, RUSSIA Fax: (095) 248 1941, E-mail: Oj@tsinet.ru ABSTRACT Series of borosilicate glass blocks containing intermediate-level NPP-operational waste have been manufactured in the mid-80s, when a pilot vitrification plant had been brought into operation, and placed for testing in experimental repositories and on an open site. Presented are some results of processing the data of 12yr field testing the assemblage of six glass blocks weighing about 180 kg in sum. A model developed for description of waste glass corrosion has been applied to assess a cumulative radionuclide release from the glass blocks for time periods of several hundred years. In the model, the radionuclide release is assumed to be controlled by the processes of diffusion limited ion exchange and glass network dissolution. Some parameters of the model were calculated on base of experimental results. Radionuclide release curves were plotted for field data and calculation results. According to the calculations performed for a proposed institutional control period of 300 years, f max = 2.3 10 -3 % of the radioactivity initially present in the waste glass will be released into the environment under saturated conditions of the near-surface repository site in the absence of additional engineered barriers. For the same period, f max = 2.110 -2 % of the initial activity will be released into the environment from the same glass at the open site due to leaching and weathering. A simplified version of the model that does not consider the glass network dissolution gives f max values of 1.810 -3 % and 1.110 -2 %, respectively. Unsealing of the experimental repository in summer 1999 has been performed with the aim to examine the waste glass condition and to collect the samples of the glass, backfill and soil for analysis. Some data on the repository near-field contamination are also presented. INTRODUCTION Waste forms disposed of in a near-surface wet repository eventually come into contact with groundwater. Engineered structures used or designed to prevent or postpone such contact and the subsequent radionuclide release, are complex and often too expensive. Development of vitrification technologies by the beginning of the 1970s provided waste forms with excellent resistivity to corrosion and gave the basic possibility of maximal simplification of engineered barrier systems. The most simple disposal option is to emplace the waste form packages directly into earthen trenches provided the host rock has the necessary sorbing and confinement properties. Such an approach has been implemented on an experimental scale as part of the research program of SIA ‘Radon’ on the encapsulation of radioactive waste in a glass matrix. The program was initiated in the mid-70s. Since then pilot and industrial vitrification plants have been constructed based on the use of a ceramic Joule-heated melter, plasma melter, and induction cold-crucible melting process. Vitrification techniques were applied to liquid and solid LILW received by the site from various sources including the Moscow wastewater purification plant, nuclear power plants (PPS) and minor producers.