Single and Multicomponent Gas Phase Diffusion in a Porous Media: Modeling and Laboratory Measurements

Subsurface vapor migration of volatile chemicals may impact ambient and indoor air quality, increasing the importance to investigate the fate and transport of these chemicals. This project involved both modeling and experimental work to study the vapor phase transport behavior of single, binary, and tertiary component systems present in the gas phase. The experimental phase resulted in the development of a diffusion cell for measuring vapor phase transport. Three organic compounds (toluene, cumene, and isooctane) common to petroleum-based products were selected. The objective of this research project was to evaluate how the rate of a component diffusing alone in a stagnant gas mixture compares to the rate of the same component when diffusing in the presence of multiple diffusing species. The equipment was first validated by measuring the unobstructed gas phase diffusion fluxes for each organic compound. The diffusion coefficients were then calculated from the experimentally measured diffusive fluxes using Fick's Law at 20 and 25°C and compared to the respective literature values. The experimental/literature (E/L) ratio was calculated for toluene, cumene, and isooctane. The range of the average E/L ratio for the single component data sets is 0.93 to 1.05. The validation data provided the baseline for extending the research to multicomponent data. The multi-component systems research was characterized as either binary systems or a three-component system. The binary systems were either isooctane/tolu-ene or isooctane/cumene. The three-component system consisted of a mixture of all three compounds. For both temperatures and all compounds the flux rate decreased for any single component due to the dilution effect by incorporation into a mixture. Applying Fick's Law to calculate the effective diffusion coefficient for each compound that corresponded to the resulting concentration gradient by the mixture, an enhancement in the diffusive flux of each individual species was observed. This enhancement can be explained by a compositional coupling of each component to all others which results in a total vapor phase mass flux comprised of both diffusive and pseudo-advective mass transport. This pseudo-advective component is attributed to simultaneous diffusion of other species in the presence of the one of interest. Since this research project incorporated a mixture of toluene and cumene present in a background carrier solvent of isooctane, by calculating the ratios Dexp(3-component)/ Dexp(2-component) and Dexp(2-compo-nent)/Dexp(single component), an estimate is obtained of the enhancement effect due to the advective component of simultaneously diffusing chemicals. The diffusivity ratios for the three-component system compared to the dual component system ranged from 0.8 to 3.7. The diffusivity ratios for individual compounds were for 1.5-3.7 cumene, 0.8-1.2 for toluene, and 1.0-1.2 for isooctane. The diffusivity ratios for the dual component system to the single component systems ranged from 0.8 to 4.0. The range of diffusivity ratios for individual compounds were for 2.0-4.0 for cumene, 0.8-1.6 for toluene, and 1.1-1.4 for isooctane. A ratio greater than 1.0 indicated an enhancement effect on the molecular diffusion rate due to the presence of one or more additional diffusing chemical species present. The majority of fate and transport models are based on single component behavior modeled by Fick's Law using the pure gas phase diffusion coefficient. The enhancement of the individual diffusive flux in a multicomponent mixture observed in this study and accounted for by pseudo-advec-tive mass transport results in an under-prediction of the actual multicomponent diffusive fluxes. It is recommended that a more rigorous diffusion equation such as the Stefan-Maxwell equation be considered for incorporation into vapor phase transport models when modeling multicomponent/ contaminant systems.