The effect of thickness in the through-diffusion experiment

This report contains an experimental study of diffusion in the water filled pores of rock samples. The samples studied are rapakivi granite from Loviisa, southern Finland. The drill-core sample was sectioned perpendicularly with diamond saw and three cylinder formed samples were obtained. The nominal thicknesses (heights of the cylinders) are 2, 4 and 6 cm. For the diffusion measurement the sample holders were pressed between two chambers. One of the chambers was filled with 0.0044 molar sodium chloride solution spiked with tracers. Another chamber was filled with inactive solution. Tritium (HTO) considered to be water equivalent tracer and anionic 36Cl were used as tracers. The data has been modelled by assuming that the porosity can be divided in two categories: through-transport and dead-end porosity. The rock-capacity factor a is also divided in the same ratio in those two categories. The numeric computation software MATLAB was used to perform the fittings. Simplex-algorithm was used to minimize the sum of squares of the difference between the measured and calculated values. The effective diffusion coefficient De is the strongest fitting parameter and a decreasing tendency is seen, when the thickness of the sample increases. The values for 36Cl are smaller than for tritium both in the case of De and a. The results suggest that a+ is decreasing as a function of thickness. As to the length of the dead-ended pores lde, no clear tendency is seen. Neither the fraction nor the length of the dead-ended pores show clear differences between 3H and 36Cl