It’s been more than a decade since Michael Gollner and his colleagues first watched a viral YouTube video of a fire tornado fueled by Jim Beam bourbon.
A warehouse in Kentucky had just been struck by lightning, funneling almost a million gallons of the flammable spirit into a nearby retention pond. As the flames whipped across the surface of the water, however, something in the atmospheric stars aligned: The flames coalesced into a towering fire whirl, more commonly known as a fire tornado.
“We saw that and went, ‘Wow, that would be a neat application’” for cleaning up oil spills, said Gollner, a mechanical engineering professor at the University of California, Berkeley Fire Research Lab. “I wonder if we could do that on purpose.”
They could, in fact. As Gollner and his collaborators recently reported in Fuel, fire whirls offer the potential to clean up oil spills more quickly and cleanly than existing methods.
Oil spill responses depend on fast, immediate action. After just 24 hours, crude oil naturally absorbs water and begins to sink beneath the waves, wreaking havoc on marine life.
Alongside other major techniques, such as containment and recovery and chemical dispersal, in situ burning via “fire pools” has been adopted as an imperfect but unavoidable tool for addressing oil spills. Fire pools stop the spread of an oil spill but send clouds of smoke into the atmosphere and leave behind a layer of tar that sinks to the seafloor.
Fire Away
If it’s far from shore, there are few methods other than basically corralling it up and burning it.”
Environmental agencies like the Bureau of Safety and Environmental Enforcement (BSEE) “were very excited about the concept of putting a change to what had been the standard for cleanup since the Exxon Valdez,” Gollner said. “There’s good knowledge, there’s an oil spill conference every year.…But if it’s far from shore, there are few methods other than basically corralling it up and burning it.”
In May 2023, Gollner, Texas A&M aerospace engineering professor Elaine Oran, and two dozen others congregated at the Texas A&M Engineering Extension Service’s (TEEX) Brayton Fire Training Field in collaboration with BSEE. The team erected a trio of 5-meter walls that would channel air flow above a central pool of water, about 3 meters square and 1.2 meters deep, topped by either a 15- or 40-millimeter layer of oil. The scale of the setup was a far cry from traditional fire whirl experiments, which take place mostly in laboratories.
“Everything’s bigger in Texas,” Gollner said.
The three walls, constructed with gaps in just the right places, caused air drawn in by the flames to spiral into a swirling, combusting tower. The intense whirlwinds effectively acted as a vortex furnace, increasing burning rates by 40% compared to traditional fire pools while also vaporizing many of the particles that would have polluted the air: Emissions of PM2.5, or particles smaller than 2.5 micrometers across that can be harmful to human health, were 40% lower in the fire whirl experiments than in pool fires.
Why these soot reductions occur is still largely a mystery; probing this question would require building a novel laboratory apparatus to take measurements from within the flame itself, Gollner explained. In the field experiments, meanwhile, one of the fire whirls managed to consume 95% of the available fuel, though the remaining tests extinguished prematurely, lowering the overall rates. Ambient wind conditions on the days of the experiments may also have had some effect.
Summoning a fire whirl in even semi-ideal conditions on the outskirts of College Station, Texas, remains a far simpler task than manifesting one in the thick of a disaster: Towing a three-walled tornado generator onto open water becomes as much a question of marine and naval engineering as of fire science. In the experiment at TEEX, the captive firenado rose to the full length of the 5-meter walls; lower walls could make a floating rig easier to transport, but the resulting mix of oxygen and fuel could actually make subsequent air pollution worse, not better.
Ali Rangwala, a professor of fire protection engineering at Worcester Polytechnic Institute (WPI) who was not involved with the project, also encourages scientific due diligence. A fire whirl “works very well if the boundary conditions are fixed and well-engineered,” he said in an email to Eos, adding that these whirls have yet to be tested on open water with waves and that the required infrastructure may be costly. (Rangwala helped conduct fire whirl experiments with Gollner at WPI but has not maintained a relationship with the project.)
“The honest fact is that this is a disaster-driven field,” Gollner said. One of the largest oil spills in history, the Deepwater Horizon spill, unleashed more than 750 million liters (200 million gallons) into the Gulf of Mexico. That was in 2010. “We haven’t seen a big oil spill for a long time, and interest in it has wavered.…We require a more interdisciplinary team and more testing. Does anyone have the appetite? Unfortunately, I think it will come with time, when we have another accident.”
Blazing a Trail
Gollner stressed the critical value of fundamental research—of lines of inquiry driven by fascination, not just application. What started as a pure appreciation of a natural wonder has the potential to change fields in ways that researchers have yet to imagine.
“Swirling or not, flames are beautiful,” Gollner said. “It is a natural flow tracer. I can see the fluid mechanics and the combustion interacting.…All the physics, all in one: It’s just beautiful.”
—Jonathan Feakins, Science Writer