Multi-Component Mass Transfer and Chemical Oxidation

Abstract

The remediation of soil and groundwater contaminated with multi-component non-aqueous phase liquids (NAPLs) such as coal tars from former manufactured gas plants (MGPs) is associated with a number of challenges. Due to thermodynamic considerations, the presence of more than one compound within multi-component NAPLs (especially when they are structurally dissimilar) can restrict intra-NAPL diffusion. Since diffusion is the dominant process in the dissolution of organic compounds, any diffusion limitations can restrict mass transfer between the NAPL and the aqueous phase. Consequently, the efficiency of conventional water-based remediation methods can be restricted. In situ chemical oxidation (ISCO) has been a possible remediation technology touted for the treatment of multi-component NAPLs. However, chemical oxidation occurs only in the aqueous phase and consequently the mass transfer between NAPLs and the aqueous phase indirectly controls the overall treatment efficiency. The primary objective of this research effort was to theoretically and experimentally investigate mass transfer processes from complex multi-component NAPLs subjected to water and chemical oxidants. For the purpose of this evaluation, the feasibility of chemical oxidants to degrade MGP residuals needs to be quantified. A series of physical model trials supported by a host of aqueous and slurry batch experiments were conducted to assess the performance of two chemical oxidants (persulfate and permanganate) using impacted sediments collected from a former MGP site. The results indicated that dissolved components were readily degraded with persulfate or permanganate (except for benzene) in the aqueous batch systems. In addition, in the well-mixed slurry systems when contact with the oxidant was achieved, permanganate, unactivated persulfate, and alkaline activated persulfate were able to degrade >95%, 45% and 30% of the initial mass quantified, respectively. However, insignificant quantifiable mass was lost in all physical models under dynamic conditions which are more representative of in situ conditions. A simple single-cell numerical model was constrained by the experimental results and used to investigate treatment expectations and the potential long-term behaviour of dissolved phase concentrations as a result of treatment using 6 pore volumes of oxidant. A specified inlet oxidant concentration and NAPL composition (22 compounds (34 %), and bulk mass (66 %) composed of unidentified material) were prescribed, and the effluent concentrations of the known soluble constituents were estimated from mass balance considerations. A variety of long-term simulation scenarios were performed. In general, for a NAPL saturation of 6 %, the results indicated that the effluent profiles over a 10-year period were reduced temporality as a result of the oxidant injection and then rebounded to a profile that was coincident with a no-treatment scenario. Based on a sensitivity analyses, neither water velocity or oxidant concentration affected the long-term behavior of dissolved phase concentrations; however, increasing the mass transfer rate coefficient had a dramatic impact, and chemical oxidant injections were only effective for low NAPL saturations (<1 %). Intra-NAPL diffusion is one of the most critical processes which can influence NAPL-water mass transfer processes. A comprehensive experimental and computational study was performed to investigate the role of intra-NAPL diffusion on the mass transfer between multi-component NAPLs and water, and to identify some of the controlling situations where this process should be considered. A diffusion-based numerical model was developed, and two different physical systems were simulated; a spherical single NAPL blob with total surface area available for mass transfer, and an isolated rectangular NAPL with only one side available for mass transfer. A series of batch and physical model experiments were conducted using coal tars collected from a former MGP site to capture multi-component diffusion-limited mass transfer behavior under static and dynamic conditions, respectively. This series of experiments was intended to focus on the direct interaction of multi-component NAPLs with water and a persulfate solution without the presence of sediment. The results from the static experiments indicated that under the diffusion-controlled mass transfer conditions, the estimated mass transfer rate coefficients were lower than typical mass transfer rate coefficients determined under continuous mixed conditions. Although, no overall trend was observed between the mass transfer rate coefficients for the various organic compounds identified, an inverse dependency between the mass transfer rate coefficient and molecular weight was clear but different for BTEX and some PAHs compounds suggesting that the intra-NAPL diffusion behavior of these two organic compound classes are different. The results indicated that molecular weight and concentration of each component are the most important parameters affecting intra-NAPL diffusion coefficients. A combination of NAPL composition, NAPL geometry, and interphase mass transfer rate may result in the depletion of more soluble compounds at the interface which can restrict NAPL-water mass transfer. When the main intra-NAPL diffusion coefficients are in the range of the self-diffusion coefficients, dissolution is not limited by internal diffusion except for high interphase mass transfer rates or long diffusional distances. In the case of complex and highly viscous NAPLs, smaller intra-NAPL diffusion coefficients are expected and even the low range of mass transfer rates can result in the depletion of more soluble compounds at the NAPL-water interface and diffusion-limited dissolution. Depending on the NAPL properties (i.e., constituent components, viscosity, temperature), interfacial depletion of the more soluble compounds can vary and influence mass transfer and dissolved phase concentrations. The comparison of experimental and simulated results indicated that rate-limited intra-NAPL diffusion within complex multi-component NAPLs as well as persulfate-NAPL interactions can restrict mass loss and chemical oxidation efficiency compared to the no-treatment scenario. It was determined that during 64 days of persulfate injection the multi-component mass transfer rate coefficients were ~70 % smaller than those estimated during an equivalent water injection period. The experimental and computational effort described in this study is the first effort to provide comprehensive information about the role of intra-NAPL diffusion on dissolution of multi-component NAPLs and the direct interaction of persulfate with MGP residuals. The diffusion-based model developed in this study provides a realistic platform to capture the temporal and spatial mass fluxes and compositional changes within complex NAPLs. While chemical oxidants (persulfate or permanganate) are able to degrade MGP residuals in well-mixed conditions, rate-limited NAPL-water mass transfer restricts treatment in systems more representative of in situ conditions. Therefore, methods to overcome the mass transfer limitations and intra-NAPL resistances are required for the remediation of complex multi-component NAPLs.