BP's engineers have tried several methods to block off the flow from the well that is causing the oil spill in the Gulf of Mexico.
But oil continues to flow into the ocean, with potentially disastrous consequences for the environment.
Now a former US military scientist has proposed a seemingly radical solution: detonating a massive bomb underwater to seal the leaking well.
It sounds like an extreme measure, but could such a plan actually succeed?
One independent engineer contacted by BBC News agreed that explosives could seal off oil wells. But he also said the plan could make the spill worse.
But Franz Gayl says that the explosion could effectively cauterise the well shutting off the flow of oil into the Gulf. However, he points out that, without detailed computer modelling, he is using educated guesswork.
Mr Gayl, a former civilian science adviser to the US Marines, proposed the idea in an interview with an internet blog.
He says that the GBU-43 Massive Ordnance Air Blast (MOAB) bomb, delivered to the site and then detonated underwater would generate a big enough shock wave to force the leaking pipes shut.
The idea relies on the "implosive" effect of exploding the 8,500kg (18,500lb) munition underwater, collapsing the metal plumbing of the well to seal it up. A Vietnam-era Daisy Cutter bomb could also be used, he said.
Mr Gayl says that, as the MOAB consumes all its own fuel, the idea is a greener option than the nuclear weapons allegedly used by the Soviet Union to shut down leaking wells.
"The exploding MOAB or Daisy Cutter would have an incredible implosive-sealing effect on oil plumbing within the immediate vicinity of the detonation," he told the War is Boring blog.
"That devastating shock wave will treat any metal cavity like soft Play-Doh, sealing every perceived cavity with a crushing force thousands of times greater than even the ambient water pressure.
"The oil plumbing is filled with rapidly flowing oil that has at any moment a lower density than the surrounding and effectively incompressible water through which the shock wave moves. Not only is crude oil less dense, but it also is compressible, unlike the water surrounding it."
In an interview with BBC News, Franz Gayl said that researchers would have to use computer software to model the outcome. Only then could its workability be properly assessed.
Many different parameters would have to be taken into account in order to evaluate the plan's chances, he said. These include the explosive force and dimensions of the bomb, the exact construction of the well head and the propagation of the shock wave.
Since the weekend, Mr Gayl has been responding to a flurry of technical queries about the method posed by internet users and experts. He stressed that the idea was his own and that he was not speaking on behalf of his employers.
He is well known in the military technology community: Mr Gayl became a whistleblower over what he described as failures in the supply of equipment to troops in Iraq and Afghanistan.
BBC News asked Michael E Webber, assistant professor in the department of mechanical engineering at the University of Texas, Austin, whether the method had merit. Dr Webber said that "explosives can definitely be used to seal off wells".
But he added that the risks were numerous: "There's a risk that detonation would exacerbate the leak by opening up the well in ways that enable higher flow rates.
"The unintended consequences to the marine environment are also difficult to predict, and the risk of dispersing explosive materials into an area that is already suffering from the threat of long term environmental damage is intimidating."
Another researcher said he considered detonating explosives too risky.
Professor Julius Langlinais, from Louisiana State University in Baton Rouge, told BBC News: "In the case of a blowing out [an] oil well, we do not want to blow things apart, but rather perhaps squeeze the casings together to seal the well."
But he added: "Bottom line, I would not advise or suggest such an action. The odds of success are small and the odds of making things worse are high."
However, Professor Webber said that despite the challenges, "the solution remains possible".
He added: "The longer the spill continues unabated, the trade-offs start to stack up in favor of drastic action.
"Simply put, can the devastation of a weapon really stop the devastation of a massive oil leak?
"At what point do the actual environmental and political damages from the unplugged well outweigh the possible environmental and political damages from the use of a weapon?"