During the Deepwater Horizon oil spill, the unprecedented injection of millions of liters of chemical dispersant at the wellhead generated large quantities of submillimeter oil droplets that became entrained in a deep sea plume. The unexpected generation of these droplets has resulted in many studies in the last decade aiming to understand their transport and fate during and after the spill. Complicating matters, the plume coincided with a microbial bloom, and in addition to ocean dynamics these droplets were subjected to biological processes such as biodegradation and microbial aggregation. A lack of field observations and laboratory experiments using relevant conditions has left our understanding of these biotic processes and the role they played in the fate of the oil droplets poorly constrained. Furthermore, while biodegradation has been incorporated into drop transport models using available data, the effects of microbial aggregation involving extracellular polymeric substances (EPS) on their transport has seldom been incorporated into modeling efforts particularly due to our lack knowledge of these processes. We use a microfluidic platform to observe bacterial suspensions interacting with a single ~200 μm oil drop in conditions relevant to the drop rising through the microbial bloom. We observe the development of individual, invisible bacterial EPS threads extending from the drop surface which can capture additional passing bacteria and form bacteria-EPS aggregates. Using high speed imaging, we make high resolution flow measurements both with and without EPS threads present and analyze the momentum balance to elucidate the hydrodynamic impact of these filaments. Surprisingly, these thin individual EPS filaments alter significantly the pressure field around the drop and increase the drag, which would drastically reduce the drop's rising velocity in the water column. We demonstrate that this mechanism which plausibly occurred in the deep sea plume would have major impacts on both the drop and bacteria transport during and after the Deepwater Horizon oil spill.

The association between phytoplankton blooms and oil spills is still controversial despite numerous studies. Surprisingly, to date, there have been no studies on the effect of bacterial communities (BCs) exposed to crude oil on phytoplankton growth, even though crude oil changes BCs, which can then affect phytoplankton growth and species composition. Co-culture with crude oil-exposed BCs significantly stimulated the growth of Prorocentrum texanum in the laboratory. To gain more direct evidence, oil-degrading bacteria from oil-contaminated sediment collected after the Texas City “Y” oil spill were isolated, and changes in dinoflagellate growth when co-cultured with single bacterial isolates was investigated. The oil-degrading bacterial isolates significantly stimulated the growth of dinoflagellates (axenic and xenic cultures) through releasing growth-promoting substances. This study provides new evidence for the potential role of oil-degrading bacteria in the formation of phytoplankton blooms after an oil spill.

Bacteria are important examples of active or self-propelled colloids. Because of their directed motion, they accumulate near interfaces. There, they can become trapped and swim adjacent to the interface via hydrodynamic interactions, or they can adsorb directly and swim in an adhered state with complex trajectories that differ from those in bulk in both form and spatiotemporal implications. We have adopted the monotrichous bacterium Pseudomonas aeruginosa PA01 as a model species and have studied its motion at oil–aqueous interfaces. We have identified conditions in which bacteria swim persistently without restructuring the interface, allowing detailed and prolonged study of their motion. In addition to characterizing the ensemble behavior of the bacteria, we have observed a gallery of distinct trajectories of individual swimmers on and near fluid interfaces. We attribute these diverse swimming behaviors to differing trapped states for the bacteria in the fluid interface. These trajectory types include Brownian diffusive paths for passive adsorbed bacteria, curvilinear trajectories including curly paths with radii of curvature larger than the cell body length, and rapid pirouette motions with radii of curvature comparable to the cell body length. Finally, we see interfacial visitors that come and go from the interfacial plane. We characterize these individual swimmer motions. This work may impact nutrient cycles for bacteria on or near interfaces in nature. This work will also have implications in microrobotics, as active colloids in general and bacteria in particular are used to carry cargo in this burgeoning field. Finally, these results have implications in engineering of active surfaces that exploit interfacially trapped self-propelled colloids.

Inhalation of PM2.5, particles with an aerodynamic diameter <2.5 mm, from sea spray after crude oil spills could present serious health concerns. The addition of dispersants to effectively spread the crude oil throughout the water column has been practiced in recent years. Here, we investigated the possibility of an increase in the toxic content of fine PM after adding dispersant. A laboratory setup consisted of a vertical tank filled with seawater, 31.5 L airspace for aerosol sampling, and a bubble generating nozzle that aerosolized the oily droplets. Four different cases were studied: no slick, 0.5-mm-thick slick of pure crude oil (MC252 surrogate), dispersant (Corexit 9500A) mixed with crude oil at dispersant to oil ratio (DOR) 1:25, and DOR 1:100. The resulting airborne droplets were sampled for gravimetric and chemical analyses through development of a gas chromatography and mass spectrometry technique. Also, PM2.5 particles were size-fractioned into 13 size bins covering <60 nm to 12.1 mm using a low-pressure cascade impactor.
The highest PM2.5 concentration (20.83 ± 5.21 mg/m3) was released from a slick of DOR 1:25, 8.83x greater than the case with pure crude oil. The average ratio of crude oil content from the slick of DOR 1:25 to the case with pure crude oil was 2.37 (1.83 vs 0.77 mg/m3) that decreased to 1.17 (0.90 vs 0.77 mg/m3) at DOR 1:100. For particles <220 nm, the resultant crude oil concentrations were 0.64 and 0.29 mg/m3 at DOR 1:25 and 1:100, both higher than 0.11 mg/m3 from the slick of pure crude oil.

In this work, a large-eddy simulation of bubble plumes in linearly stratified environments is presented. The gas bubbles are treated as Lagrangian particles. The intrusion and peeling are clearly manifested in the computed flow fields. The results of about 50 simulations with different parameters reveal the importance of bubble source area for plumes on the laboratory scale. A new type of bubble plume with rapid and distinct peelings is observed which is favored by large source areas. With a proper normalization, the present data points collapse onto a single straight line after applying a virtual-source correction which reflects the source-area effect. These results provide a plausible explanation for the scatter of the previous experimental and computational data in literature. A simple relation between the trap height and the peel height is observed and its mechanism is discussed