H42E-02
Co-boiling of NAPLs and water during thermal remediation: experimental and modeling study

Thursday, 17 December 2015: 10:35
3016 (Moscone West)
Magdalena Krol1, Cindy Zhao2, Kevin G Mumford3, Brent E Sleep4 and Bernard H Kueper3, (1)York University, Toronto, ON, Canada, (2)Morrison Hershfield, Ottawa, ON, Canada, (3)Queen's University, Kingston, ON, Canada, (4)University of Toronto, Civil Engineering, Toronto, ON, Canada
Abstract:
The persistence of non-aqueous-phase liquids (NAPLs) in the subsurface has led to the development of several remediation technologies to address this environmental problem. One such group of technologies (in situ thermal treatment) uses heat to volatilize contaminants. Subsurface temperature measurements are often used to monitor progress and optimize contaminant removal. However, when NAPL and water are heated together, gas is created at a temperature lower than the boiling point of either liquid (co-boiling), which can affect temperature observations.

To examine the effect of co-boiling on observed temperatures and NAPL mass removal, a series of heated laboratory experiments were performed using single and multi-component NAPLs. The experiments consisted of glass jars filled with a mixture of sand, water, and NAPL mixed to obtain an approximately uniform NAPL distribution within the jar. The experiments were heated from the outside and interior temperatures were measured using a thermocouple. The tests showed that local-scale temperature measurements are unreliable in indicating the end of co-boiling and may not indicate complete mass removal. This is because a well-defined co-boiling plateau does not exist when heating a multi-component NAPL and the temperature is dependent on the proximity of NAPL to the monitoring point.

To further investigate temperature distributions and the potential to use gas production as a complementary indicator of NAPL removal, a 2D finite-difference mass transport model was used that incorporated heat transport, latent heat, phase change, and a multicomponent gas phase and used a macroscopic invasion percolation (MIP) model to simulate gas movement. Latent heat was calculated by multiplying specific latent heat, which is an intrinsic property of a substance, by the amount of liquid mass being vaporized and its incorporation into the model allowed for the simulation of co-boiling plateaus (during single component NAPL boiling). The developed model was used to examine the spatial distributions of gas production and temperature during heating of water and NAPL, for a variety of NAPL source configurations, to better understand the processes occurring during thermal remediation activities and how to improve the interpretation of measured temperatures.