USING PASSIVE ULTRASENSITIVE HYDROCARBON DETECTION TO ELUCIDATE NASCENT PIPELINE LEAKS: A COLUMBIA PIPELINE CASE STUDY

Abstract

Underground and above ground hydrocarbon transport pipelines often contain carbon dioxide (CO2), hydrogen sulfide (H2S), water and chlorine which cause corrosion. Corrosion often begins as pinpoint leaks that expand over time. These leaks are often difficult to detect using conventional methods until a major event occurs. Pressure testing can determine a leak to be present, but does not pinpoint the location of the leak. Pipeline pigs normally only detect leaks after they become significant and costly. The use of methane detectors has also been utilized with the recent popularity of drones. However, the use of airborne methane detectors has been less than successful due to the limited linear range of the methane detectors and poor sensitivity. Passive ultrasensitive sorbent modules have been used to detect nascent leaks at parts per billion ( ppb) levels, which is 1,000 times more sensitive than traditional methods. Passive ultrasensitive sorbent modules contain a specially engineered oleophilic (i.e. oil loving) adsorbent encased in a microporous membrane. These membrane pores are small enough to prevent the entrance of soil particles or water, but are large enough to allow hydrocarbon vapor molecules to pass through and concentrate on the adsorbent material within. The result is a 1,000-fold increase in concentration allowing for ppb level detection. The Columbia natural gas condensate pipeline case study took place in 2007 just southwest of Pittsburgh and involved a pipeline buried at a depth of approximately 6 ft. Ultrasensitive passive modules were installed at the surface above the pipeline. A battery operated hand drill was used to drill a 1 inch hole in the ground to an approximate depth of 3 ft. The module was inserted into the hole, covered with dirt, and left for 4 days. After retrieval the modules were analyzed by thermal desorption/mass spectrometry. The objectives of the survey were to: examine potential fingerprints for evidence of gas condensate leakage, determine if nascent leaks could be distinguished from baseline readings, compare results with pipeline maintenance records for ground-truthing purposes. The results of the project showed: several locations along the pipeline exhibited strong potential as leakage points, the results were validated with a known leak along the pipeline, the data helped to monitor the efficiency of prior pipeline repair work, baseline levels of hydrocarbons were statistically derived from the data, potential nascent leak points were identified along the pipeline.