Dissolution Of A Spatially Variable Non-Aqueous Phase Liquid Source: Experimental Study And Model Devolpment
2009 (English)In: PROCEEDINGS, TOUGH Symposium 2009: Lawrence Berkeley National Laboratory, Berkeley, California, September 14-16, 2009, 2009Conference paper (Other academic)
Mass transfer from subsurface occurrences of non-aqueous phase liquids (NAPLs) to the surrounding groundwater is a key process, both for persistent contamination problems originating from organic liquid sources, and for dissolution trapping in geological CO2 sequestration. While the dissolution typically is governed by processes that occur on very small scales (i.e., the pore scale), there is a need to model the mass transfer coupled to transport of dissolved chemical components over field scales that are several orders of magnitude larger. Upscaled models, linking the small-scale characteristics to the total mass transfer from sources of dissolved chemicals, are therefore needed. Well-controlled laboratory experiments conducted in test cells allow for the generation of accurate data to validate such upscaling methods before applying them to much more complex field systems.
Following this approach, a set of experiments were conducted in a two-dimensional sand tank, wherethe dissolution of a spatially variable, 5 cm by 5 cm DNAPL tetrachloroethene (PCE) source was carefully monitored in space and time. With a resolution of 0.2´0.2 cm, NAPL saturations were measured using x-ray attenuation techniques at approximately 1,000 individual pixels in the source zone. By continuously measuring the NAPL saturations, the temporal evolution of DNAPL mass loss by dissolution to groundwater could be measured at each pixel. The rate of dissolution varied spatially and temporally within the source, and was found to be correlated to NAPL morphology, groundwater flow velocity, and position within the source. The dissolution process was modeled using iTOUGH2/T2VOC under assumption of local equilibrium (LE) between the DNAPL and dissolved PCE. A preliminary model of rate-limited (RL) dissolution, based on a Gilland-Sherwood type relation implemented in MODFLOW/RT3D, was also tested. It was found that the LE model could not capture the observed dissolution patterns, although it predicted the total rate of mass transfer well for the given source conditions. The RL model showed potential to better capture the dissolution pattern after further model development. Ongoing work is aimed at addressing these issues by correlating the detailed measurements of NAPL entrapment morphology and local dissolution rates, with the final goal of developing an upscaled model of the total mass transfer from the source zone.
Place, publisher, year, edition, pages
Earth and Related Environmental Sciences
IdentifiersURN: urn:nbn:se:uu:diva-144077OAI: oai:DiVA.org:uu-144077DiVA: diva2:392369
TOUGH Symposium 2009