06/01/2010 03:22 pm ET Updated May 25, 2011

BP's Top Kill Attempt Fails: Here Is a Method That Could Really Work

I'm sure that everyone was disappointed with the failure of Top Kill to stem the flow of oil gushing into the Gulf of Mexico. After three days of pumping 30,000 barrels of drilling mud enhanced with non-toxic agents to increase its weight, supplemented with a junk shot, in which golf balls and rubber scraps were added in hopes of clogging up the pipeline, the bleeding never stopped. In retrospect, it seems as if the attempt were akin to trying to stop a runaway fire hose by stuffing dirt into the nozzle. According to a BP technician interviewed who spoke to the NY Times on condition of remaining anonymous, "Simply too much of what we pumped in was escaping."

The company seems to be all out of new ideas. Perhaps it is time to open this up to the thousands of innovative minds we have in this country. I propose that BP offer a $10 million reward for the originator of any idea that proves to be successful. That ought to be enough to get the creative juices flowing, while it would be a tiny drop in the bucket compared to the additional costs the company will be faced with if they have to wait until a relief well is completed in August, or longer, to stop the leak.

At this point the company plans to go back to the idea of using a containment dome. This cap will be smaller than the previous one, and they will attempt to connect it directly to the failed riser after they cut off the broken pipe. The maneuver will be difficult and complex and the company has already said it will not be a perfect seal, so at best they will capture "most" of the oil coming out. White House Energy Czar, Carol Browner has raised concerns that this new approach could actually end up increasing the size of the leak if the pipe is cut and the cap is unsuccessful.

The previous containment dome failed when ice-like crystals began to form as they began pumping out the dome. The combination of ice and congealed oil clogged up the outlet.
It was this failure that gave 35-year veteran chemical engineer Alan Olson his idea for stopping the leak. Olson is a Professional Engineer registered in the state of Ohio. I think this idea could work.

Here is a description of the idea in Olson's own words [my comments in brackets]:

"Form the plug by inserting two lances, (small diameter pipes, maybe 2 inches in diameter) into the riser pipe. Start flowing water (could be seawater) through one pipe. Flow liquid nitrogen through the other pipe. The liquid nitrogen is cold, and will [further] cool as it expands, freezing the water and congealing the oil. Refrigeration could also be clamped around the outside of the pipe [if needed]. The plug has to stay frozen long enough to cap the well [with concrete]. Straightforward chemical engineering calculations can be used to engineer the pipes, flows and cooling. If liquid nitrogen does not seem the right coolant, it would be very quick to check other substances such as carbon dioxide. There is also expertise at NASA for understanding how liquefied gases behave as they expand to gases and cool."

Let's go back to our runaway fire hose analogy. Imagine placing a small tube through the nozzle and some distance down the length of the hose. Then, we begin to flow a very cold gas out of the tube. Initially the gas will be swept away by the rushing water. But after time, since the gas is coming out of one fixed point, a region of cold will be established around the tube. If sufficient gas is injected, the water will begin to slow and then freeze, which will further slow the flow and eventually a solid plug of ice will form.

The fact that oil congeals when cold will help the process as it will move more slowly through the pipe. As the flow slows down, the freezing will speed up.

Olson feels, that, implemented correctly, this approach has a 95% chance of success. As a 28-year veteran mechanical engineer myself, I believe he could be right. To date, he has submitted the idea to EPA and BP but has heard nothing in response.

RP Siegel is a professional engineer with 45 US patents to his name. He is also an author and the President of Rain Mountain LLC.