Brian Robinson

Flume tank studies to elucidate the fate and behavior of diluted bitumen spilled at sea.

An economical alternative to conventional crudes, Canadian bitumen, harvested as a semi-liquid, is diluted with condensate to make it viable to transport by pipeline to coastal areas where it would be shipped by tankers to global markets. Not much is known about the fate of diluted bitumen (dilbit) when spilled at sea. For this purpose, we conducted dilbit (Access Western Blend; AWB and Cold Lake Blend; CLB) weathering studies for 13 days in a flume tank containing seawater. After six days of weathering, droplets detached from the AWB slick and were dense enough to sink in seawater. The density of CLB also increased, but at a slower rate compared to AWB, which was attributed to the high concentration of alkylated polycyclic aromatic hydrocarbons in it, which are more resistant to weathering. An empirical, Monod-type model was introduced and was found to closely simulate the increase in oil density with time. Such a model could be used within oil spill models.

Droplet and bubble formation of combined oil and gas releases in subsea blowouts

Underwater blowouts from gas and oil operations often involve the simultaneous release of oil and gas. Presence of gas bubbles in jets/plumes could greatly influence oil droplet formation. With the aim of understanding and quantifying the droplet formation from Deepwater Horizon blowout (DWH) we developed a new formulation for gas-oil interaction with jets/plumes. We used the jet-droplet formation model VDROP-J with the new module and the updated model was validated against laboratory and field experimental data. Application to DWH revealed that, in the absence of dispersant, gas input resulted in a reduction of d50 by up to 1.5mm, and maximum impact occurred at intermediate gas fractions (30-50%). In the presence of dispersant, reduction in d50 due to bubbles was small because of the promoted small sizes of both bubbles and droplets by surfactants. The new development could largely enhance the prediction and response to oil and gas blowouts.

Field Trials of in-situ Oil Spill Countermeasures in Ice-Infested Waters

An oil spill response technique in ice-infested waters based on the application of fine minerals in a slurry with mixing by propeller-wash to promote the formation of oilmineral aggregates (OMA) has been proposed. This process promotes the physical dispersion of mineral-fine stabilized oil droplets into the water column that support higher rates of oil degradation by natural bacteria. To validate the operational effectiveness of this technique a controlled oil spill experiment was conducted from a Canadian Coast Guard ice-breaker in the St. Lawrence Estuary (offshore of Matane, Quebec, Canada). Following the release of the test crude oil and the application of experimental treatments, time-series changes in oil concentrations were monitored to quantify dispersion effectiveness. Field samples were also recovered for laboratory microcosm studies on the biodegradation of petroleum hydrocarbons by monitoring CO2 production and the depletion of specific hydrocarbon components. Detailed chemical analysis (GC/MS with hopane normalization) from these studies showed that more than 60% of the total petroleum hydrocarbon, 75-88% of total alkanes, and 55-65% total PAHs, were degraded after 56 days of incubation at 0.5 o C. The alkylated PAH was degraded to a greater extent following the addition of mineral fines. This technique offers several operational advantages as a spill countermeasure for use under Arctic conditions such as reduced numbers of personnel required for its application, no need for waste disposal sites, and cost effectiveness.

Toxicity Effects of Chemically-Dispersed Crude Oil on Fish

ABSTRACT Studies were conducted to: (1) evaluate the toxicity of polycyclic aromatic hydrocarbons (PAHs), the water accommodated fraction (WAF) and chemically enhanced water accommodated fraction (CEWAF) from Alaska North Slope (ANS), Medium South American (MESA) and Arabian Light (AL) crude oils on life stages of commercially valuable herring and cod; (2) use biomarkers as indicators of exposure and effects; (3) assess environmental, physical, chemical, biological and temporal factors affecting toxicity; and (4) compare data using identical protocols for different oils and dispersants (Corexit 9500 and SPC-1000). Chronic LC50s for Pacific herring embryos were 3 (ANS) and 1.5 (MESA) mg-petroleum-hydrocarbons/L. ANS was 2.5–3 times more toxic than AL to Atlantic herring embryos. The apparent greater toxicity of a medium compared to a light crude might have been due to higher concentration and increased degree of alkyl PAHs with three or more rings, but the difference was not statistically significant. Incr...

Impact of mixing time and energy on the dispersion effectiveness and droplets size of oil.

The effects of mixing time and energy on Alaska Northern Slope (ANS) and diluted bitumen Cold Lake Blend (CLB) were investigated using EPA baffled flask test. Dispersion effectiveness and droplet size distribution were measured after 5-120xa0min. A modeling method to predict the mean droplet size was introduced for the first time to tentatively elucidate the droplet size breakup mechanism. The ANS dispersion effectiveness greatly increased with dispersant and mixing energy. However, little CLB dispersion was noted at small energy input (εxa0=xa00.02 Watt/kg). With dispersant, the ANS droplet size distribution reached quasi-equilibrium within 10xa0min, but that of CLB seems to reach quasi-equilibrium after 120xa0min. Dispersants are assumed ineffective on high viscosity oils because dispersants do not penetrate them. We provide an alternative explanation based on the elongation time of the droplets and its residence in high intensity zones. When mixing energy is small, CLB did not disperse after 120xa0min, long enough to allow the surfactant penetration. Our findings suggest that dispersants may disperse high viscosity oils at a rougher sea state and a longer time. The latter could determine how far offshore one can intervene for effective responses to a high viscosity oil spill offshore.

Application of Microbiological Methods to Assess the Potential Impact of Produced Water Discharges

Microbial production and activity in produced water directly recovered from the discharge stream of offshore oil and gas production facilities off the east coast of Canada were examined before and after aeration in a series of concentrations to determine the effect of dilution at sea. Aeration and dilution resulted in reduced toxicity due to volatilization and oxidation of the lighter hydrocarbons including polycyclic aromatic hydrocarbons (PAHs), alkylated PAHs, benzene, toluene, ethylbenzene, xylene, and short-chain alkanes (C10–C14). A fraction of the detrimental effects on microbial productivity and activity could also be attributed to the elevated salinity associated with produced water. These results suggest that caution should be used in the manipulation of produced water samples used for toxicity/risk assessment studies.

Settling of dilbit-derived oil-mineral aggregates (OMAs) & transport parameters for oil spill modelling

The size and settling velocity of oil-mineral aggregates (OMAs) derived from diluted bitumen are primary constituents in predictive models for evaluating the potential fate of oil spilled in the aquatic environment. A series of low sediment concentration (15mg·L-1), colder water (<10°C) wave tank experiments designed to measure variability in these parameters in naturally-formed OMAs in response the presence or absence of chemical dispersant are discussed. Corresponding lab experiments revealed settling velocities of artificially formed OMAs on the order of 0.1-0.4mm·s-1. High-resolution imagery of settling particles were analyzed for particle size, density and settling velocity. In situ formation of OMAs in the wave tank was unsuccessful. Possible effects of chemical dispersant on natural sediment flocculation, the size of suspended oil droplets and clearance rates of suspended particles are discussed.

Reynolds number scaling to predict droplet size distribution in dispersed and undispersed subsurface oil releases.

This study was aimed at testing the applicability of modified Weber number scaling with Alaska North Slope (ANS) crude oil, and developing a Reynolds number scaling approach for oil droplet size prediction for high viscosity oils. Dispersant to oil ratio and empirical coefficients were also quantified. Finally, a two-step Rosin-Rammler scheme was introduced for the determination of droplet size distribution. This new approach appeared more advantageous in avoiding the inconsistency in interfacial tension measurements, and consequently delivered concise droplet size prediction. Calculated and observed data correlated well based on Reynolds number scaling. The relation indicated that chemical dispersant played an important role in reducing the droplet size of ANS under different seasonal conditions. The proposed Reynolds number scaling and two-step Rosin-Rammler approaches provide a concise, reliable way to predict droplet size distribution, supporting decision making in chemical dispersant application during an offshore oil spill.

Interfacial film formation: influence on oil spreading rates in lab basin tests and dispersant effectiveness testing in a wave tank.

Test facilities such as lab basins and wave tanks are essential when evaluating the use of chemical dispersants to treat oil spills at sea. However, these test facilities have boundaries (walls) that provide an ideal environment for surface (interfacial) film formation on seawater. Surface films may form from surfactants naturally present in crude oil as well as dispersant drift/overspray when applied to an oil spill. The objective of this study was to examine the impact of surface film formation on oil spreading rates in a small scale lab basin and on dispersant effectiveness conducted in a large scale wave tank. The process of crude oil spreading on the surface of the basin seawater was influenced in the presence of a surface film as shown using a 1st order kinetic model. In addition, interfacial film formation can greatly influence chemically dispersed crude oil in a large scale dynamic wave tank.

A New Mechanism of Sediment Attachment to Oil in Turbulent Flows: Projectile Particles

The interaction of oil and sediment in the environment determines, to a large extent, the trajectory and fate of oil. Using confocal microscope imaging techniques to obtain detailed 3D structures of oil-particle aggregates (OPAs) formed in turbulent flows, we elucidated a new mechanism of particle attachment, whereby the particles behave as projectiles penetrating the oil droplets to depths varying from ∼2 to 10 μm due to the hydrodynamic forces in the water. This mechanism results in a higher attachment of particles on oil in comparison with adsorption, as commonly assumed. The projectile hypothesis also explains the fragmentation of oil droplets with time, which occurred after long hours of mixing, leading to the formation of massive OPA clusters. Various lines of inquiry strongly suggested that protruding particles get torn from oil droplets and carry oil with them, causing the torn particles to be amphiphillic so that they contribute to the formation of massive OPAs of smaller oil droplets (<∼5-10 μm). Low particle concentration resulted in large, irregularly shaped oil blobs over time, the deformation of which without fragmentation could be due to partial coverage of the oil droplet surface by particles. The findings herein revealed a new pathway for the fate of oil in environments containing non-negligible sediment concentrations.