Natural gas is racing neck and neck with coal to be the top fuel for generating electricity in the U.S. thanks to increased use of techniques, such as hydraulic fracturing, that extract the gas from unconventional sources. Since burning natural gas emits half as much carbon dioxide as does burning coal, the natural gas boom has helped lower per capita carbon emissions in the U.S. over the past decade.
However, a major downside of natural gas is that it leaks. Millions of tons every year—roughly 9 million tons in the U.S. alone—spill into the atmosphere during extraction, storage, and transport. Because natural gas comprises 95% methane, a greenhouse gas that traps 86 times as much heat as does carbon dioxide over a 20-year period, leaking just 2–3% of the gas that’s produced worldwide can wipe out its environmental benefits over coal. “The short-term climate punch of waste methane is equivalent to over 200 coal-fired power plants,” says Ben N. Ratner, a director at the nonprofit Environmental Defense Fund (EDF).
Methane is also highly flammable. Leaks have caused fatal explosions in San Bruno, Calif.; New York City; and elsewhere. Plus, leaking natural gas is like scattering money into the wind. According to the policy analysis firm Rhodium Group, methane leaks cost companies about $30 billion per year.
These leaks continue in part because there are no regulations requiring their containment. The U.S. Environmental Protection Agency has only a voluntary program to reduce leaks. And detecting leaks isn’t easy: Current devices cost tens of thousands of dollars and are cumbersome to employ across long pipelines crossing remote landscapes.
Cheap, accurate, and compact sensors that continuously look for methane leaks could enable widespread monitoring, even along pipelines. These could nip climate-harming leaks in the bud and help companies’ bottom lines. Innovators are making strides on improving sensors thanks to challenges issued by EDF and the U.S. Department of Energy’s Advanced Research Projects Agency-Energy (ARPA-E). Some natural gas producers and utility companies are starting to test advanced leak detectors. These new technologies could make it easier and cheaper for them to greatly reduce the carbon footprint of natural gas.
In the U.S., natural gas producers extract the gas from roughly half a million well pads: sites where engineers bore pipes deep into the ground. Monitoring these pads and miles of pipelines for invisible, odorless methane is daunting. The distinctive rotten-egg smell that alerts you to gas leaks at home is due to a mercaptan, which is added later by utility companies for safety. “If methane emissions were purple, this would’ve been solved years ago,” Ratner says.
Today, oil and gas companies and utilities hire inspectors who walk around facilities with infrared cameras that visualize methane swirls. Methane absorbs infrared light, so when light from the sun travels through millions of methane molecules, it looks different than light just passing through air.
These cameras can spot about 80% of emissions from a leaking well under ideal conditions—low wind, warm weather, clear skies—and at an imaging distance of 10 meters, according to researchers from Stanford University (Environ. Sci. Technol. 2016, DOI: 10.1021/acs.est.6b03906). But conditions are not ideal most of the time. Besides, these cameras can detect methane only at levels greater than 10,000 ppm, so they’re incapable of catching small leaks, and most companies perform inspections about once every quarter, so leaks can go undetected for months.
Newer, laser-based sensors can detect 5 ppm concentrations from 50 meters. These spectrometers shoot pulses from tunable infrared lasers at a target—a storage tank or pipe, for example. Some of the light bounces back to the detector, which senses changes due to absorption by methane. These laser spectrometers can scan large areas from farther away than can infrared cameras, but they have a hefty price tag of more than $75,000.
The goal for next-generation detectors is to detect methane at parts-per-million levels from a distance; recognize that the methane is not coming from, say, the rotting garbage in nearby landfills; pinpoint where the leak is; and estimate how much gas is escaping. And they need to be affordable. Researchers are moving toward these goals by using less expensive, chip-based laser and detector technologies and automated monitoring. “The vision is to detect potent natural gas leaks with a speed we would all expect in the digital age,” Ratner says. “Automated, continuous methane monitoring that alerts the operator could allow personnel to be dispatched at the right place and time to fix leaks in minutes.”
To help this vision become reality, EDF launched its Methane Detectors Challenge program in 2014, offering innovators $10,000 and a chance to test their methane-sensing technologies with big oil and gas companies. To qualify, the sensors needed to detect methane at a level of 2 ppm for a market cost of $1,000. Two promising technologies have emerged from the program and are now undergoing pilot testing.
One is a spectroscopy system made by Longmont, Colo.-based start-up Quanta3. The company took low-cost, low-power, chip-based near-infrared tunable laser diodes used in fiber-optic communications and repurposed them to detect methane. The system is sensitive to parts-per-billion levels and should cost $3,000 per site per year, says company founder Dirk Richter. Norwegian energy company Statoil began testing the system at its production facility in Eagle Ford, Texas, in January.
The second device is made by San Francisco start-up Acutect. Californian utility Pacific Gas & Electric (PG&E) is testing the device at its natural gas storage facility in Los Medanos, Calif. The innovation of this device is to include a reflector that bounces back to the detector an infrared laser beam shined from up to 30 meters away. Methane passing between the reflector and detector decreases the reflected signal. Both Quanta3’s and Acutect’s systems are solar powered and send data to the cloud, where the information can be analyzed and used to immediately alert operators to leaks.
Companies such as PG&E and Statoil are taking methane leak detection seriously. PG&E owns more than 10,000 km of large transmission pipelines and 72,000 km of smaller neighborhood distribution lines, says François Rongere, PG&E gas operations R&D and innovation manager. PG&E tested Acutect’s sensor and is also testing a lightweight laser spectrometer originally made by the National Aeronautics & Space Administration’s Jet Propulsion Laboratory for the Mars Curiosity rover to look for signs of life on the red planet. Methane plumes detected there could come from microbes. Now the lab is adapting the technology for detecting methane on Earth.
The sensor uses an efficient, ultracompact laser called an interband cascade laser that emits near-infrared light. NASA engineers shrank it down further and mounted it on a drone (C&EN, April 11, 2016, page 9). It is sensitive to parts-per-billion methane levels and runs on little power, Rongere says. Monitoring remote pipelines with a drone would be far cheaper than having a person drive around looking for leaks.
The EDF program isn’t the only one nurturing novel methane-sensing technologies. ARPA-E’s $60 million MONITOR (Methane Observation Networks with Innovative Technology to Obtain Reductions) program is funding a dozen technologies for monitoring well pads. The goal is to accurately detect small leaks for an annual cost of $3,000 and estimate the leaks’ locations and flow rates.
In one project, IBM scientists and engineers made cheap, compact, silicon-chip-based tunable diode lasers and photodetectors. Each 5- by 5-mm sensor should cost about $300, says team leader Hendrik F. Hamann. IBM can fabricate the sensors on silicon wafers using the same technology for putting transistors on a computer chip, which should drastically reduce manufacturing costs and be easy to scale up. Wireless sensors, placed around a well pad, will send data to cloud-based computers. On the basis of the methane reading combined with weather data, the software will pinpoint the location of the leak and quantify it, Hamann says. Houston-based oil and gas company Southwestern Energy will soon run pilot tests of the device.
Although they have become the industry standard for gas sensing, optical techniques aren’t the only way to go for methane emissions monitoring. Jeffrey T. Glass of Duke University has developed a specialized mass spectrometer that can detect methane and other volatile organics found in natural gas. This ability should help it distinguish between gas leaking from wells and gas leaking from nearby farms since the two sources would have different chemical signatures, he says.
Mass spectrometry is “neglected for methane monitoring because it’s hard to get a field-portable instrument,” Glass says. Mass specs are typically refrigerator-sized beasts, but Glass has managed to shrink the instruments down to shoe-box size.
In a spectrometer, ionized molecules pass through a slit into a magnetic field. A smaller spectrometer means a smaller slit that allows fewer ions through, decreasing sensitivity. Glass and his team have designed a small instrument with multiple slits that allow up to 12 times as many ions to pass into the machine. But that change on its own would lead to “a jumble of information at the end, and you wouldn’t know what you’re detecting,” Glass says.
To make sense of the ions, Glass’s instrument has a bar-code-like array of slits of different widths. The pattern acts like a code, revealing precisely which spot the ions came from, giving order to the barrage of atoms hitting the detector. The instrument is about half as precise as full-size mass spectrometers, which typically have low parts-per-million sensitivity for recognizing methane. But this should be sufficient for detecting most methane leaks, Glass says.
Meanwhile, researchers at the Palo Alto Research Center are taking a completely different approach with a nanotube-based sensor. To make the devices, they print arrays of carbon nanotubes in 1-mm2patches. The nanotubes in each patch are decorated with different chemical groups and nanoparticles that react with various trace chemicals in natural gas, including ethane, propane, hydrogen sulfide, and ammonia. Molecules collecting on the surface change the conductivity of the sensor.
Like the IBM team, the Palo Alto group also plans to deploy multiple sensors around a well pad in a network to pinpoint leak location within 1 meter. The printed sensors are cheap, about $10 each, says David E. Schwartz, who is leading the development. “ARPA-E wants total monitoring cost per well per year to be $3,000,” he says. “Our target is a few hundred dollars.”
These and other ARPA-E-funded technologies are now undergoing a slew of tests at a mock natural gas well pad in Fort Collins, Colo. By simulating big and small gas leaks from different equipment and locations under real-world conditions, researchers are testing whether the technologies are up to par.
One piece of good news for companies hoping to waste less methane via leaks is that the largest 5% of leaks are responsible for more than half of methane emissions. Equipment leaks, open tank hatches, corroded pipelines, or other “superemitters” gush 100 to 1,000 tons of methane per year (C&EN Online Latest News, April 25, 2016). Monitoring technologies that help companies detect and fix these giant leaks could make a huge dent in total emissions.
While many companies are pursuing leak detection for the benefit of their bottom lines and to reduce greenhouse gas emissions, others may not have a choice but to look for leaks in the future. In 2016, EPA and the U.S. Bureau of Land Management finalized rules to limit the oil and gas industry’s methane emissions from gas venting and leaks. Congress under the Trump Administration unsuccessfully tried to roll those rules back. In July, a federal appeals court blocked EPA’s attempt to delay implementation of the agency’s new rules (C&EN, July 10, page 18) and then reaffirmed the order later that month (C&EN, Aug. 7, page 15). Meanwhile, California and Colorado have set their own stringent methane emissions standards. Ratner explains, “Well-crafted regulation can be a strong driver for innovation.”
Prachi Patel is a freelance writer. A version of this story appeared in ACS Central Science: cenm.ag/methanedetectors.