Bubble Migration in Pore Networks of Uniform Geometry

The behavior of bubbles migrating in porous media is a critical factor in several soil remediation operations such as in situ air sparging, supersaturated water injection, bioslurping, trench aeration and up-flow operation of moving bed sand filters as well as in the oil and gas industry. Groundwate...

Full description

Bibliographic Details
Main Author: Ghasemian, Saloumeh
Language:en
Published: 2012
Subjects:
Online Access:http://hdl.handle.net/10012/6502
Description
Summary:The behavior of bubbles migrating in porous media is a critical factor in several soil remediation operations such as in situ air sparging, supersaturated water injection, bioslurping, trench aeration and up-flow operation of moving bed sand filters as well as in the oil and gas industry. Groundwater aquifers are constantly polluted by human activity and a common threat to fresh water is the contamination by non-aqueous phase liquids (NAPL). In many NAPL removal technologies, gas bubbles carrying NAPL residuals move upwards through the water-saturated porous media and thus play an essential role in contaminant recovery. The mobilization of the residual oil blobs in oil reservoirs is another important application for rising bubbles in porous media. After an oil field is waterflooded, a significant fraction of oil, referred to as waterflood residual oil, remains trapped. A potential mechanism to recover this residual oil is the mobilization of oil by gas bubbles moving upwards in water-wet systems. The main focus of this work was to measure the velocity of bubbles of various lengths during their migration through a water-wet porous medium. Experiments were conducted in a saturated glass micromodel with different test liquids, air bubbles of varying lengths and different micromodel elevation angles. More than a hundred experimental runs were performed to measure the migration velocity of bubbles as a function of wetting fluid properties, bubble length, and micromodel inclination angle. The results showed a linear dependency of the average bubble velocity as a function of bubble length and the sine of inclination angle of the model. Comparisons were made using experimental data for air bubbles rising in kerosene, Soltrol 170 and dyed White Oil. The calculated permeability of the micromodel was obtained for different systems assuming the effective length for viscous dissipation is equal to the initial bubble length. It was found that the calculated permeability had an increasing trend with increasing bubble length. Laboratory visualization experiments were conducted for air bubbles in White Oil (viscosity of 12 cP) to visualize the periodic nature of the flow of rising bubbles in a pore network. The motion of the air bubbles in saturated micromodel was video-recorded by a digital camera, reviewed and analyzed using PowerDVD ™11 software. An image of a bubble migrating in the porous medium was obtained by capturing a still frame at a specific time and was analyzed to determine the bubble shape, the exact positions of the bubble front and bubble tail during motion and, thus, the dynamic length of the bubble. A deformation in the shape of the bubble tail end was observed for long bubbles. The dynamic bubble lengths were larger than the static bubble lengths and showed an increasing trend when increasing the angle of inclination. The dynamic bubble lengths were used to recalculate the bubble velocity and permeability. A linear correlation was found for the average bubble velocity as a function of dynamic bubble length. Numerical simulation was performed by modifying an existing MATLAB® simulation for the rise velocity of a gas bubble and the induced pressure field while it migrates though porous media. The results showed that the rise velocity of a gas bubble is affected by the grid size of the pore network in the direction perpendicular to the bubble migration. In reality, this effect is demonstrated by the presence of other bubbles near the rising bubble in porous media. The simulation results showed good agreement with experimental data for long bubbles with high velocities. More work is required to improve the accuracy of simulation results for relatively large bubbles.