LIFE ESTIMATION OF HIGH LEVEL WASTE TANK STEEL FOR F-TANK FARM CLOSURE PERFORMANCE ASSESSMENT

High level radioactive waste (HLW) is stored in underground storage tanks at the Savannah River Site. The SRS is proceeding with closure of the 22 tanks located in F-Area. Closure consists of removing the bulk of the waste, chemical cleaning, heel removal, stabilizing remaining residuals with tailored grout formulations and severing/sealing external penetrations. A performance assessment is being performed in support of closure of the F-Tank Farm. Initially, the carbon steel construction materials of the high level waste tanks will provide a barrier to the leaching of radionuclides into the soil. However, the carbon steel liners will degrade over time, most likely due to corrosion, and no longer provide a barrier. The tank life estimation in support of the performance assessment has been completed. The estimation considered general and localized corrosion mechanisms of the tank steel exposed to the contamination zone, grouted, and soil conditions. The estimation was completed for Type I, Type III, and Type IV tanks in the F-Tank Farm. The tank life estimation in support of the F-Tank Farm closure performance assessment has been completed. The estimation considered general and localized corrosion mechanisms of the tank steel exposed to the contamination zone, grouted, and soil conditions. The estimation was completed for Type I, Type III, and Type IV tanks in the F-Tank Farm. Consumption of the tank steel encased in grouted conditions was determined to occur either due to carbonation of the concrete leading to low pH conditions, or the chloride-induced de-passivation of the steel leading to accelerated corrosion. A deterministic approach was initially followed to estimate the life of the tank liner in grouted conditions or in soil conditions. The results of this life estimation are shown in Table 1 and Table 2 for grouted and soil conditions respectively. The tank life has been estimated under conservative assumptions of diffusion rates. However, the same process of calculation can be followed, once a better understanding of the concrete degradation and consequent diffusion rates is developed. A stochastic approach was also followed to estimate the distributions of failures based upon the same mechanisms of corrosion, but accounting for variances in each of the independent variables. The recommended distributions for time to failure of the tank liner for use in stochastic modeling are shown in Table 3. The distributions are based upon based upon known parameters of SRS waste tank construction and groundwater analyses. The results recommended for modeling are based upon discrete diffusion coefficients of 1 x 10{sup -6} cm{sup 2}/sec for oxygen diffusion as well as CO{sub 2} diffusion.