Measuring effective diffusivities in porous pellets: experimental procedures and FFT calculations

摘要

The effective diffusivity, D[subscript] e, of He in N[subscript]2 in ZnS pellets was determined using a dynamic Wicke-Kallenbach diffusion cell at 1.5 atm and 23°C. Pellets were pressed from reagent powders in rings with internal diameters of 25.4 mm. Pellet lengths ranged from 3.43 to 23.34 mm;Diffusivities were calculated by minimizing the sum-of-squares error, in the frequency domain, between the experimental data and a model for diffusion through the pellet. The experimental data were transformed to the frequency domain via the fast Fourier transform (FFT) and the Laplace transform solution of the diffusion equations was used for comparison to the experimental data;Four different configurations of the experimental apparatus were used and the effects of apparatus dead-volumes, chamber volumes, bottom chamber gas feed position, and flow rate through the thermal conductivity detector (TCD) on D[subscript] e, are shown. The height of the tracer dispersion cone above the top face of the pellet affected the value of D[subscript] e calculated. A model was proposed which separated the upper chamber volume into a region of complete mixing and into another region where mass transfer occurred by gaseous diffusion. The model did not adequately describe the effect of cone height above the pellet on the values of D[subscript] e obtained. Sample loop volumes ranged from 0.15 x 10[superscript]-6 to 5.21 x 10[superscript]-6 m[superscript]3 and He/N[subscript]2 tracer gas concentrations of 9.9, 52.5, and 100% He were used, but the differences among the diffusivities attributed to using the different gas concentrations and sample loops were less than 10%;Diffusivities were found to increase with pellet length unless the response-time of the TCD and the dispersion of the gas flow through the apparatus were included in the diffusion model. D[subscript] e values calculated in this manner agreed within 5% of the values calculated using the difference in moments procedure;Tortuosity factors, [tau], were calculated by summing the diffusional contributions over all pores sizes using the results obtained from mercury porosimetry experiments. The best low and high estimates for the tortuosity factor gave reasonable values of 1.56 and 2.84, respectively.