Much of the oil that gushed into the Gulf of Mexico during the 2010 Deepwater Horizon oil spill may have shattered into droplets too small to rise to the surface.
Droplets with an average diameter of about 80 micrometers would have formed as the oil violently rushed out of the well and into the crushing pressure of 1,500-meter-deep water, researchers report in the May 4 Chemical Engineering Science. Such tiny spheres lack the buoyancy to resist ocean currents and rise through the water. The finding could help researchers account for oil still missing from the spill. It could also help in future spills to decide whether to apply chemical dispersants, which responders use to create similarly small droplets, the authors say. They suggest that use of dispersants should be considered on a case-by-case basis and that they may not be necessary for such deep, violent spills.
Science News headlines, in your inbox
Headlines and summaries of the latest Science News articles, delivered to your email inbox every Thursday.
Thank you for signing up!
There was a problem signing you up.
In 2010, spilled oil hit the Gulf’s water in a dramatic crash, says chemical engineer Zachary Aman of the University of Western Australia in Crawley. “It’s like hitting a wall at 70 miles an hour,” he says.
At the time of the spill, scientists knew little about what to expect from the oil and had few options for studying it at depth. To mimic the oil bursting out of an opening into high-pressure water, Aman and colleagues set up an apparatus called a sapphire autoclave, which pressurizes liquids while mixing them, like a powerful blender.
Using a cocktail of chemicals that matched the ingredients of the 2010 spill, researchers set the oil spinning under a pressure of 11 megapascals, or about 110 times that of the atmosphere. Then, through a pane of sapphire glass, the researchers used a high-speed camera to film the tiny droplets and measure their size.
The autoclave couldn’t actually replicate the pressure and mixing power of the spill in the Gulf, which would have been up to 25megapascals; the researchers couldn’t afford such a powerful apparatus. So they measured droplet sizes as the mixer ran at one of eight speeds. Then they fed the trend data into a supercomputer simulation, which extrapolated to match conditions in the Gulf.
Subscribe to Science News
Get great science journalism, from the most trusted source, delivered to your doorstep.
GUSHER Oil gushing out of the well during the 2010 BP oil spill transformed into a swarm of tiny particles, many too small to escape to the surface. U.S. Department of Energy/Youtube
The researchers found that the oil would have broken into droplets with an average size of about 80 micrometers, which would have been trapped in the deep layers of the oceans and pushed around by ocean currents. Aman hopes that the data will help researchers refine their estimates of what happened to the approximately 5 million barrels of oil that spilled according to estimates by the federal government, around 75 percent of which is missing (SN: 4/18/15, p. 22).
During the spill, responders were hoping that the oil would form small droplets that would be unlikely to surface or coat coastlines and animals. Responders sprayed chemical dispersants, detergent-like chemicals that separate oil into small droplets, at the surface of the water and at depth, at the wellhead.
When Aman and colleagues added chemical dispersants into their simulation of the wellhead, they found that average droplet size dropped from about 80 micrometers to 45 micrometers. This decrease would have reduced the amount of surfacing oil by at most 3 percent, Aman and his colleagues calculated.
Dispersants might not be useful in such deep, violent spills, the researchers concluded. But biogeochemist David Valentine of the University of California, Santa Barbara says it’s too early to say for sure. Valentine points out that the experimental and computer simulations might not accurately mimic how the oil gushed into the Gulf. “This result is useful, but other experimental approaches must be applied,” he says.
Aman agrees that more work needs to be done to verify the results. He and colleagues are currently refining their simulation.