The chemical dispersion of heavy or viscous oils and the upper limit of viscosity for the successful dispersion of viscous oils have been studied by a number of researchers including past work at Ohmsett (SL Ross 2005, 2008). Most of this work has focused on determining what the limiting viscosity is for chemical dispersion not what the primary reason is for the inability to chemically disperse the oil. Two dispersion processes may be affected by the oil’s viscosity. One is the ability to get the applied dispersant to penetrate into the viscous oil in sufficient quantity to be effective before it is washed away in the dominant water phase (we will call this “mixing one”). The other is internal visco-elasticity of the oil that may prevent the oil from being broken into small droplets under the prevailing energy conditions even if a significant quantity of dispersant has penetrated into the oil (we will call this “mixing two”).

This proposed research project will investigate which of these processes is the limiting factor for a number of viscous oils. Preliminary laboratory scale work using the SL Ross wave tank will be conducted using pre-mixed viscous oil and dispersant samples to determine when the internal oil viscosity is the controlling factor. A range of dispersant doses will be pre-mixed with a number of viscous oils and oil-dispersant mix viscosities will be monitored to separate the surfactant and viscosity influences on the dispersion process. The viscous oils that disperse when pre-mixed with dispersant will be tested using a spray application of dispersant at similar (and possibly higher) doses to determine if “mixing one” is the limiting factor in the chemical dispersion of the oil.

Results from a recent study on low-dose repeat application of dispersants
(SL Ross 2009) indicate that multiple low-dose applications of dispersant to viscous oil slicks may in fact result in a better dispersion outcome than one single dose. Multiple low-dose applications on the viscous oils where dispersion is not limited by the “mixing two” process will also be proposed in this project to study the merits of this strategy in improving the “mixing one” process. Initial small scale testing in the SL Ross wave tank will be proposed to be followed by full scale testing at Ohmsett where more realistic dispersant application and mixing regimes can be studied.

Crude oils from twelve wells located offshore Southern California and in the Gulf of Mexico were sourced and delivered to the Ohmsett facility for use in this project. Samples of each crude oil were sent to Environment Canada in Ottawa, On for physical and chemical analyses and for laboratory dispersant effectiveness (DE) testing.  Samples of the same crude oils were sent to the US EPA in Cincinnati for DE testing using the Baffled Flask Test (TAR Project 666) and, and SL Ross Environmental Research Ltd. in Ottawa, ON for chemical analyses and small scale DE testing in their wave tank.  Small scale DE testing was completed by SL Ross and the results used to develop the test plan for full scale dispersant testing at Ohmsett April 19-23, 2010.

The first study goal was to investigate pre-mixed oil-dispersant viscosities and final dispersant dosages to determine limiting viscosities for successful “mixing-two” at different dispersant dosages.  The success achieved at the high dose for the most viscous oil indicates that the viscosity limit for dispersion is a function of both the amount of dispersant mixed into the oil (the final DOR) and the oil-dispersant mix viscosity.  Based on the small- and large-scale test results of this study oil-dispersant mixes with final viscosities of about 10,000 cP appear to need a DOR of 1:10 or better to achieve significant dispersion. Both the reduction in viscosity due to the addition of the more fluid dispersant and the presence of more surfactant improve the final DE. 

The second study goal was to investigate if there is an initial oil viscosity that prevents successful “mixing one” or the penetration of the dispersant into the oil during spray applications prior to it being washed away by water.  In large-scale Ohmsett tests, dispersant applied by spraying produced high levels of effectiveness in all three oils and tests with pre-mixed oil and dispersant (at similar DORs) resulted in lower dispersant effectiveness (DE) than those where the dispersant was applied by spray. The results of large-scale test at Ohmsett did not confirm that effectiveness is controlled by the simple effect of oil viscosity preventing the penetration of the dispersant into the oil where it could do its work (mixing-one). 

Twenty-five years of experience in dispersant effectiveness testing and the small-scale test results from this study would have predicted that the opposite result should have prevailed. Evidence that the data collected was valid includes 1) both pre-mixed and spray–applied tests were successfully replicated in the test program, 2) visual observations matched the measured test DE and 3) the in-water LISST and C3 fluorometry results were consistent with the DE measurements.  Because the spray applications were at least somewhat effective in all cases and were more effective than pre-mixed tests with similar dispersant dosages the “mixing-one” limitation on dispersant effectiveness was not evident for the oils and conditions used in the large-scale Ohmsett tests.   

Dispersant experiments were conducted at Ohmsett November 1-5, 2010 to confirm the results obtained in the April 2010 test program.  The results from the Ohmsett dispersant experiments are currently being analyzed.  The final report from this project will be updated to include these experimental results.   

Report:  Please add the following report.  It will be report AA