Sandia LabNews

Tragic gas explosion propels potential pipeline safety technologies onto national priority list

Sen. Bingaman asks Congress for R&D funding to enhance pipeline safety

In a flash the scene out of a Norman Rockwell painting — a family of 12 fishing on the banks of the Pecos River on a summer afternoon — went from idyllic to tragic.

Nobody knows what happened, exactly. Perhaps natural gas hissing from a leak in the pipe somehow got sparked, igniting a blast that killed nine members of the family instantly — a leak that might have been detected with new technology.

While firefighters attempted to reach the victims, it took workers almost an hour to manually shut off a valve that fed thousands of cubic feet of methane into the blaze — a valve that might have sensed the leak and shut off automatically.

After the flames were out, investigators recovered pieces of 50-year-old pipe that showed signs of corrosion — corrosion that might have been detected.

A job for technology

Six days after the Aug. 19 accident, Dan Horschel (6233) toured the scene with Sen. Jeff Bingaman, D-N.M. Dan saw the 20-foot-deep crater where the pipe used to be. He noted that paint on signs and a bridge railing more than 300 feet away from the blaze was scorched.

"Evidence of the intense heat was everywhere," he says.

Pipeline robot moving through a pipe
PIPE PROWLER — Mobile robot technology is one area of expertise that could be applied to safety problems in the natural gas transmission industry, according to a Sandia report written in the wake of the tragic Aug. 19 gas pipeline explosion in southern New Mexico. Here Jim Buttz (15252) inspects a small mobile robot developed for other applications. (Photo by Randy Montoya)

Predictably, in the wake of such a tragedy, people were asking how it might have been prevented. Bingaman, during a planned visit to Sandia’s Carlsbad facility days after the accident, asked how technology could make the nation’s 2 million miles of gas transmission lines, and many more millions of miles of urban distribution lines, safer.

Predictably, Sandia had some ideas.

By Sept. 1, 13 days after the accident, and with the help of scores of Sandians, Dan had compiled a seven-page white paper titled "Capabilities and Areas of Research and Development to Enhance Pipeline Safety in the US" and briefed Bingaman’s staff.

The paper noted that with some focused R&D work, and in cooperation with safety experts in the gas-distribution industry, technology and expertise at Sandia and other national labs could help.

"We don’t have all the answers," says Dan. "But there is some overlap between the safety issues and areas where technology might help. We want the best that technology can offer to be available to the gas industry as it looks for ways to improve safety."

The next week, Bingaman-sponsored language sailed through Congress attached to a bill proposing stepped up gas-line safety. Bingaman’s rider proposed new pipeline safety R&D. The bill passed the Senate on Sept. 7. The House is still debating it.

Inhibiting corrosion

As he gathered ideas about how technology could help, Dan found that Sandia already had examined a few pipeline reliability issues as part of its Critical Infrastructure Protection initiative managed by Energy and Critical Infrastructure Center 6200.

Researchers in Catalysis & Chemical Technologies Dept. 6245 have worked with DOE, industry, and Russian institutes to develop fuel cells that provide power to "active cathodic protection" systems — essentially applying an electric current to a pipe’s metal to inhibit corrosion — along remote trans-Siberian pipelines. That work continues, with help from DOE’s National Energy Technology Lab (NETL), says Al Sylwester (6245).

In addition, Center 6200 has an ongoing program to improve the security and reliability of SCADAs (Supervisory Command and Data Acquisition Systems), such as those used to monitor and control power distribution on the electric grid. The natural gas "grid" employs SCADAs as well.

Future gas transmission systems, if well instrumented, could use secure SCADAs to monitor sudden pressure drops in gas lines and ensure that data and control signals being transmitted are timely, accurate, and genuine. That would minimize false alarms and enable rapid or automatic shutdown of leaking gas lines, says Jason Stamp (6237).

Detecting leaks

Since 1992 Tom Kulp of Diagnostics and Remote Sensing Dept. 8356 and a Sandia team has worked with the natural gas industry, regulators, and oil companies to adapt commercially available laser-imaging tools to detect low-pressure leaks in urban gas distribution lines and at oil refineries. These imagers paint a scene with laser light of a wavelength tailored to the gas being sought. If a plume of the gas is present, it absorbs the light and creates a shadow in the video image.

In 1997 members of the team drove the streets of Atlanta using a van-mounted laser-imaging system tailored to methane, detecting one potentially dangerous gas leak in a manhole. More recent tests at oil refineries have successfully detected leaks as well.

Tom says the team currently is close to completing a miniaturized, portable leak-detection system that is carried in a backpack.

With some development work, he says, the technology could be applied to detecting leaks in gas transmission lines, which have much higher pressures (making plumes more visible) but typically are buried deeper (gas that percolates to the surface might be diluted).

Buried transmission lines would likely need to be inspected by aircraft, so detecting methane from greater stand-off distances would be a technical challenge as well, he says.

A pig in a pipe

Dan’s paper also suggests that Sandia use its robotics and sensors expertise to miniaturize and improve "pigs" used in the gas industry to inspect pipes from the inside, and to develop robots that could deliver sensors to suspect areas of a pipe’s interior.

The pigs, essentially sensor-laden logs that get pushed by flowing gas through a pipe to detect cracks, corrosion, or damage, are too large to get through smaller pipes, tight turns, and some valves. They also provide limited information about a pipe’s condition. As a result many gas lines, including the section of pipe that exploded Aug. 19, go uninspected, according to gas-industry sources after the accident.

Smarter, smaller, more agile pigs — "Smart Piglets" — could be developed at Sandia, says Mark Vaughn of Mobile Robotics Dept. 15252, that could store image data on board; after the pig is extracted, a map of the pipe’s interior could be reconstructed.

For detailed inspection of pipe walls, Mark suggests intelligent or remotely controlled robots that deliver sensor packages directly to suspect areas of the pipe. Sandia already is working with the US Navy on a shipboard inspection robot that has magnetic wheels, which might be useful for navigating a pipe’s interior.

To create a pipe-inspection robot, researchers might need to address issues such as communication of sensor and robot-control data through pipe walls and navigation over valves and other in-pipe obstacles, he says.

Sensors for safety

Better sensors that detect gas vapors associated with leaks, corrosion or its byproducts, and other telltale signs of poor pipe integrity could be adapted from Sandia technology as well, says Steve Martin of Microsensor R&D Dept. 1744. Inexpensive surface acoustic wave (SAW) sensors now under development at Sandia, for instance, could detect leaks by trapping particles of the leaking gas in a thin film that acts like a chemical fly paper, then using acoustic signals to detect the presence of the gas species on the film.

Chemically selective coatings on the pinpoint ends of fiber optic lines also can be used to detect gases or corrosion byproducts, he says. A beam of light transmitted down the fiber would be reflected differently when the coating has absorbed a gas or vapor, making it possible to identify the chemical species. Sandians already are developing such sensors to detect leaks in underground storage tanks.

To detect dangerous hydrogen leaks in rocket boosters, Labs researchers also have worked with NASA engineers to develop field-effect transistor (FET) sensors that recognize changes in the flow of electricity across a gate of specialized material. Such devices have been integrated onto chips with signal processing and control capabilities.

And the latest microsensors with tiny pumps and valves — being developed to detect trace amounts of toxic chemicals in the air — could be applied to the challenge of detecting corrosion or gas leakage as well.

Arrays of such sensors could be placed along new pipe lengths or at valves and fittings and could be networked using fiber optics or RF signals. They could be delivered to a corroded area of an existing pipe aboard a robot or pig. Or they could be used by an inspector to manually check a pipe from the outside, adds Steve.

Better inspection

To detect hidden flaws in metals, fringe-field sensors detect irregularities in an electromagnetic field coursing between two electrodes. Such sensors, originally developed at Sandia to identify metal burrs on workpieces being machined by robot, might help pipeline inspectors find cracks, wall thinning, and other flaws in pipelines, he says.

Remote-field eddy current devices, similar to those used at Sandia’s Airworthiness Assurance Center to find flaws in aging aircraft structures, might be used to manually inspect gas pipes for signs of pitting that indicates severe corrosion, says David Moore of Airworthiness Assurance Dept. 6252. Eddy current devices essentially measure the changes in electromagnetic fields in metals to locate hidden flaws.

Ranking safety priorities

But to make a real impact on pipeline safety, Sandia would need to come up with solutions that not only improve safety but also save the gas industry money. That’s where statistical tools and software employing probabilistic risk assessment might come into the equation.

Sandia employs the tools of probabilistic risk assessment — using statistics describing past failures to rank the probabilities of future failures — in its quest to keep US nuclear weapons among the safest and most reliable complex systems in the world.

The same assessment techniques embodied in a set of industry-specific software tools could help gas-line operators develop inspection and maintenance strategies, says Bob Cranwell of Systems Reliability Dept. 6411. These tools would rank failure scenarios by their likelihoods and consequence severities, which could help gas industry officials choose optimal inspection and replacement cycles and send maintenance workers to the riskiest sections of pipe.

Such tools could help the industry maintain an acceptable level of safety and reliability in its two million-plus miles of pipe and avoid the costly "shotgun approach."

Jeff Braithwaite of Corrosion Science & Technology Dept. 1832 suggests incorporating materials degradation models into risk assessments of pipelines, in much the same way materials performance models are applied to Sandia’s weapons surety work.

"Essentially it’s adding real-time sensor data, our fundamental understanding of materials behavior, and statistical uncertainty to identify what we call ‘vulnerabilities’ in weapons systems," he says. "You could combine models describing corrosion, thermally induced stresses on materials, and the ages, types, and locations of sections of pipe to create a code that predicts and prioritizes trouble spots."

Already Sandia has employed simple "predictive risk assessment" methodologies to manufacturing tools, gun boxes on helicopters, and other DoD aircraft.

Bettering our national infrastructures

"This is the kind of thing that really floats our boats — using the wealth of expertise, unique facilities, and good people at the Labs to solve national problems," says Dan. "It’s important that we get chances to tackle these kinds of high-profile problems."

"The surety of the oil and gas distribution system in the US has been a major concern of our Critical Infrastructure Protection initiative since we started the program four years ago," says Sam Varnado, Center 6200 Director. "We applaud Sen. Bingaman’s initiative in trying to get increased federal funding for R&D in this area. We are working to build partnerships with NETL and the DOT so that we can be more effective in bringing our technology to bear."

"Our Critical Infrastructure Protection team had already looked at issues of gas pipeline performance, so we were able to respond [to Bingaman’s request] knowledgeably with less than 24 hours’ notice," adds Bob Eagan, VP of Energy, Information, and Infrastructure Surety Div. 6000. "As the Carlsbad explosion makes all too clear, catastrophic failures of our systems can have immediate and tragic consequences. We look forward to helping regulators and the pipeline industry by developing and applying technologies that enhance the safety of this critical infrastructure."

Last modified: Oct. 9, 2000