In Situ Raman Spectroscopic Observations of Gas-Saturated Rising Oil droplets: Simulation with Decane as an Oil-Equivalent Substitute

Edward T Peltzer, Peter M Walz and Peter G Brewer, MBARI, Moss Landing, CA, United States
Abstract:
Oil droplets rising from the sea floor, whether from seeps or well leakage, contain very large quantities of dissolved gas that profoundly affects their density and critical oil-water interfacial characteristics. The primary dissolved gas is methane which may be up to 30% of the molar volume. This can create a hydrate skin as the methane gas is shed from the oil as it rises through the water column, thus decreasing in pressure and increasing in temperature, and steadily changing the rising droplet buoyancy. We have explored this phenomenon by executing controlled ROV based experiments with a “bubble cup” technique in which a small volume of gas saturated decane (saturated with pure methane, a mix of methane and nitrogen , or a mix of methane and CO2) is interrogated by laser Raman spectroscopy. The use of decane as an oil “substitute” is required since natural oil samples are highly fluorescent due to the presence of polycyclic aromatic hydrocarbons. We have devised Matlab techniques for extracting the spectroscopic dissolved methane signal from the thicket of decane peaks that surround it. We have directly observed the rate at which gases are lost from the “oil” per unit area at depths in the water column that are both within and outside the hydrate forming phase boundary. We have compared the behavior of both a non-hydrate forming dissolved gas (nitrogen) with CO2 where the hydrate phase boundary is at significantly shallower depth. The results indicate complex interfacial behavior and physical chemistry. We did not observe direct gas bubble formation on the decane outer surface but did observe gas bubble formation within the oil droplets as they rose through the water column. Because there are significant energy barriers for homogeneous bubble formation within the decane phase, we took this as evidence of significant gas super-saturation within the oil droplet. The gas loss rates increased significantly in all cases when the hydrate phase boundary was crossed.