Monday, May 31, 2010

Stopping the BP Oil Spill

Being from Texas, I have been distraught watching the environmental disaster that the Deepwater Horizon Oil Spill has become.  Being also an engineer, I can't help but start brainstorming ideas for how to fix the problem.  Below is a copy of a proposal I have submitted to the Mineral Management Website for technical proposals - it will be interesting to see if I get any response and if so what it is.  I highly encourage others that have ideas regarding stopping the spill or ways to manage the clean up to submit proposals.
While British Petroleum has tried several methods to mitigate or stop the flow of oil from the submerged vestiges of the sunken Deepwater Horizon offshore drilling platform, to date none have been successful.  It is believed highly likely that the relief wells being drilled now will permanently terminate the flow, but it is expected that this operation will not be complete for at least 4 more weeks during which time the size of the spill will double or more.  Below I roughly sketch out a stop-gap measure for significantly reducing the flow of oil form the well head by super-cooling the riser pipe using readily available and inexpensive liquid nitrogen.  British Petroleum most likely already has much of the infrastructure required to do such an operation and assuming the deep-water operations are successful the probability of a significant reduction in flow is high.
Brief Background

Crude oil and two-phase crude oil/gas mixtures have a very strong viscosity dependency on temperature.  At lower temperatures (especially below 0˚F) the viscosity of crude oil and its mixtures increase very rapidly.  As can be seen in the figure I generated below and to the left, the mass flowrate of a fluid through a pipe at constant pressure drop is a very strong function of the viscosity of the fluid flowing in the pipe.  Next to it is a plot of viscosity vs. temperature for a two-phase mixture of light crude and natural gas (as the flow coming out of the Deepwater Horizon wellhead is) at a bunch of different pressures.
Figure on left shows mass flowrate dependency on viscosity of fluid.  Figure on right shows the strong dependence of viscosity in oil/gas mixture on temperature (Right figure credit: Lyons, William C., and Joseph Zaba. Standard Handbook of Petroleum & Natural Gas Engineering. Houston, Tex.: Gulf Pub., 1996. Print.)

If one can control the temperature, and thus the viscosity, of the flow of crude from a well the flowrate can be reduced dramatically.
The Proposal

Using common and relatively low cost industrial materials, components and techniques, a cooling jacket can be applied around the broken riser in order to significantly lower the temperature of the oil flowing from the well.  This will increase its viscosity and lower the flowrate in an accelerating feedback loop:
Several companies make clamp-on liquid nitrogen collars for freezing fluids in pipes.  They are off the shelf cryogenic components:
Cryo-lator pipe freezing collar (courtesy

An alternate method would involve pumping liquid nitrogen directly into the well head below the break in the hope of forming hydrate and extremely thick crude “sludge” that would plug up the riser, slowing or haulting flow:
This method would likely also be worth considering although it would require somehow breaching the riser pipe and installing a leak-tight connection to a very high pressure liquid nitrogen line which could be problematic at these depths.
Required Equipment
  1. Large, clamp-on liquid nitrogen circulator for the riser pipe.  This would probably be a custom component manufactured for the size pipe being used at the sea floor.
  2. Long stainless steel or cryogenic-quality aluminum pipes to transport liquid nitrogen from the surface to the sea floor.  Stainless steel or aluminum are required for cryogenic liquids as they don’t become brittle at low temperatures.  It is likely that much of the plumbing BP already uses is stainless steel so they probably have this available.
  3. Large cryogenic holding vessels for liquid nitrogen.  These are readily available from gas suppliers such as AirGas and could be plumbed together in a large array on a barge.  Alternatively, a cryogenic transport ship for liquid natural gas (LNG) could be filled with liquid nitrogen instead, storing huge amount.
  4. Large cryogenic pumps.  These are commonly used in industry for moving large quantities of liquid nitrogen and liquid oxygen.  They are also very commonly used in the petroleum industry so chances are that BP has ships with these already installed.
  5. Capable deep water ROV’s to install the collar and associated plumbing on the riser.  BP has already been using such devices to try their other fixes.  They should be capable of this operation.
Potential Problems

  1. The single largest reason why this proposal may not work is the two-phase nature of the flow coming out of the well.  Gas mixed with liquid behaves very different than pure liquids and some analysis would need to be done to show that cooling this mixture would reduce flowrate significantly.  It is likely that it would - the plot shown above for temperature vs. viscosity is for a methane/crude mixture with specific gravity .7 and it can be clearly seen that the viscosity is still greatly affected by temperature.
  2. Getting liquid nitrogen to the ocean floor a mile beneath ships requires a well insulated pipe.  Fortunately, this will happen naturally as ice forms on the pipe once liquid nitrogen begins flowing.  Initially only gas will reach the ocean floor as liquid nitrogen boils in the pipes, cooling them down.  This is actually beneficial as well since it will dry out the entire nitrogen circuit making internal ice build-up unlikely.  As the pipes cool and ice forms insulating their outer walls, the liquid nitrogen will stay in a liquid state to deeper and deeper depths until it reaches the sea floor.  Getting to this point will take time and significant amount of liquid nitrogen, but it is an extremely cheap substance as it is made of air.
  3. The ice buildup on the pipes could provide structural problems for them.  Ice is less dense than water so it would provide a buoyant force trying to pull the pipes upward.  Analysis would have to be done to show that the pipes are strong enough to withstand this.
    While there are some potential pitfalls with this proposal, it seems to have enough merit to at least warrant further analysis by those in charge at BP and the federal government.  It is my hope that by writing this white-paper, someone will take notice and at least consider this as an option.