Bubble-facilitated Mobilization of Trapped Dense Non-aqueous Phase Liquid (DNAPL) at Residual Saturation

Gas bubbles can significantly affect contaminant transport in groundwater systems. These bubbles can be produced by natural processes and remediation activities. Mobilized gas bubbles can enhance vertical transport of volatile organic compounds (VOCs), increasing vapor intrusion and air emissions. In some circumstances, gas bubble mobilization can also result in the transport of DNAPL as coatings on the gas bubble surface. This could move a greater mass of DNAPL to the groundwater table, potentially increasing the NAPL-impacted area. However, it could also serve as a removal mechanism, shortening source lifetimes as part of a remediation strategy. Despite this potential enhanced transport, there has been little controlled laboratory study to investigate bubble-facilitated DNAPL transport and to estimate its associated mass flux.

A series of experiments was conducted in a two-dimensional glass flow cell (10 × 20 ×1 cm³). It was packed with medium sand (d₅₀ = 0.7 or 0.5 mm) and contained a 4.2 cm-thick creosote source zone below a clean sand layer (11.5, 7.0 or 4.5 cm thick). A free water layer was placed above the clean sand. Gas (air or nitrogen) was injected through the creosote source zone at 1 mL/min through a stainless needle using either a syringe pump or a mass flow controller. Short-term experiments were conducted for 8 hours to estimate DNAPL flux, and a long-term experiment was conducted for several weeks to investigate DNAPL removal. The cell was backlit with a LED panel to facilitate visualization of gas and NAPL. Heptane was added above the water layer as a solvent trap to collect creosote mass transported through the sand by gas bubbles, and was analyzed by gas chromatography (GC) for petroleum hydrocarbons (PHC) and priority polycyclic aromatic hydrocarbons (PAHs).

Results showed that gas bubbles leaving the sourze zone were often coated in creosote, and that creosote travelled with the gas bubbles as either thin films or tails below the bubbles. PHC concentrations measured above the clean sand indicated that creosote mass was constantly released out of the sand pack over 8 hours (1.25-10 mg/hr), with more mass being released through thinner and finer clean sand layers. In the long-term experiment, clearly visible evidence has also suggested significant depletion of the source zone.