A Smarter Way to Detect Fugitive Emissions

Chemical Engineering Progress, Dec 2007 by Parkinson, Gerald

Early next year, the U.S. Environmental Protection Agency (EPA; Washington, DC; www.epa.gov) plans to issue an amended regulation to make it simpler and cheaper for process plants to detect fugitive emissions. Under existing rules, emissions monitoring is a time-consuming task in which plant personnel use sensitive gas-sampling instruments to check for leaks from individual pumps, valves and other components. In contrast, the new regulation will allow the option of optical imaging, using a hand-held infrared camera or a similar device to obtain an image of a gas cloud and of plant equipment, so that the source of the leak can be more readily identified.

The amended regulation is the result of about 10 years of investigation and testing that was initiated by the American Petroleum Institute (API; Washington, DC; www.api.org) and subsequently expanded to include other groups, such as equipment vendors and state agencies, in cooperation with EPA and the U.S. Dept. of Energy (DOE, Washington, DC; www.doe.gov). However, the key technological factor has been the commercialization of hand-held infrared cameras that can produce clear images of otherwise invisible gas clouds.

At present, the procedure for detecting leaks of volatile organic compounds (VOCs) in the chemical process industries is defined in EPA's Reference Method 21 (M21). This typically involves quarterly monitoring of individual plant components using a gas analyzer (commonly called a "sniffer"), followed by the repair of leaking components, a sequence known as LDAR (for Leak Detection And Repair). The trigger point or threshold for a leak for many components is 500 ppm - i.e., if the reading is over the threshold, the component is considered to be leaking and is labeled a "leaker."

Under the amended regulation, one threshold for optical imaging will be a mass flowrate of 60 g/h, with a maximum interval between tests of two months. Higher thresholds will be allowed if monitoring is done more frequently than every two months. The threshold is approximately equivalent to that for M21 LDAR, says David Markwordt, an environmental engineer with EPA's Office of Air Quality Planning and Standards (Research Triangle Park, NC), who has headed EPA's years-long effort to develop the amendment.

Optical imaging will not replace M21 LDAR, but will be an alternative work practice, says Markwordt He also points out that while infrared (BR) imaging is the only commercially available optical method at present, other systems may be permitted as long as they meet the regulatory requirements.

Smart LDAR

The advantage of optical imaging, such as IR emissions monitoring, is that it allows viewing of a larger plant area to identify high-rate leaks more quickly, rather than checking each component one by one, as is done under M21. The benefits, say its proponents, are that it is less expensive than M21, and it reveals more large leaks, not just those from the components normally monitored, so it enhances plant safety.

The cost of using M21 LDAR at a large U.S. process plant often exceeds $1 million/yr, says Jeffrey Siegell, coordinator of the aforementioned API project and a senior engineering associate with ExxonMobil Research and Engineering Co. (Fairfax, VA; www.exxonmobil.com). He adds that not only is M21 LDAR labor-intensive, but most of the effort is wasted, since typically less than 1-2% of all components are found to be leaking.

The API project was established to identify a more-effective way to control fugitive emissions in refineries. The API working group analyzed five years of leak-monitoring data from seven California refineries and found that more than 92% of controllable emissions came from about 0.13% of the piping components. From diis was developed the concept of Smart LDAR - the more-rapid detection and repair of high-rate leaks.

Since a Smart LDAR program using an infrared camera focuses on locating the small population of high-emission-rate leakers, me cost is only about half that of M21 LDAR, says David Furry, president of Leak Surveys, Inc. (Early, TX; www.leaksurveysinc.com), which uses ER cameras to detect leaks for clients. "A person using a sniffer can check about 500 components per day," he says, "whereas one person with an IR camera can survey several thousand components per day."

A field test that compared IR imaging side-by-side with M21 concluded that the two methods are fairly comparable in performance, in terms of the estimated annual emissions from the leaks they detected. The six-month test program, which ended in October, was conducted by ExxonMobil Chemical Co. at its Baton Rouge, LA, plant, in cooperation with the Louisiana Dept. of Environmental Quality (Baton Rouge).

The testing was done on an isopropyl alcohol unit that has 3,542 fugitive emission components (as specified by EPA), says Wayne Sadik, a senior staff environmental engineer with ExxonMobil Chemical in Baytown, TX. Each testing method was used three times, with M21 being done at the prescribed three-month intervals and optical imaging every two months.