Sorption and diffusion behavior of selenium in tuff

Sorption and diffusion behavior of Se in tuff was studied by batch sorption and through-diffusion experiments. Tuff samples used in the experiments were obtained from the Toki Lignite-Bearing Formation, Tono area, Gifu, central Japan. The experiments were carried out in a nitrogen glove-box with an oxygen level less than 1 ppm at room temperature. Batch sorption experiments of Se on crushed tuff samples and its constituent minerals were carried out in the pH range between 2 and 13. The fraction of Se sorbed on tuff was above 90% in the pH range below 8 and decreased to below 30% with increasing pH. Comparison of pH dependencies of sorption between tuff and its constituent minerals suggested that Fe minerals such as Fe-oxyhydroxide and pyrite contributed to Se sorption on tuff. Through-diffusion experiments of Se through intact tuff samples were carried out at the pH values of 8 and 11. Effective diffusion coefficients obtained at both pH values were almost the same, on the other hand, the distribution coefficient obtained at pH 8 was higher than that at pH 11 by one order of magnitude. This result was consistent qualitatively with the pH dependency of Se sorption by the batch experiments. However, the distribution coefficients obtained by the diffusion experiments were lower than those by the batch experiments and the differences were one order of magnitude. Correcting the difference of specific surface areas for crushed and intact tuff samples could not explain the difference between distribution coefficients obtained by batch sorption and diffusion experiments. Consideration of the pore size distribution by the mercury porosimetry suggested that the microscopic pores with size below several tens of nanometers might contribute slightly to sorption. The assumption that the pores with size below 20 nm are not concerned with sorption leads to the consistency of distribution coefficients between batch sorption and diffusion experiments.