Richard M. Baldwin
Southwest Research Institute
San Antonio
Black powder can collect in gas pipelines, leading to flow loss and premature failure of compressors, instruments, and other equipment. (Photograph courtesy of Southwest Research Institute, San Antonio)
Recent research into black powder problems experienced by natural-gas transmission pipelines indicates steps that can be taken to control the material in gas pipelines, gathering systems, and storage facilities.
(The accompanying box on p. 52 reports the most recent, focused efforts to address the problem.)
Black powder (the common term for various forms of iron sulfide mixed with contaminants) is the least understood but most prominent contaminant in natural-gas pipelines and compression equipment in pipeline, refinery, gathering, and storage applications.
For many years, pipeline companies have observed its presence and effects but have viewed it only as a nuisance. They have devoted almost no attention to assessing the costs of the problem or of dealing with it.
In fact, black powder collects in gas pipelines, creates wear and reduced efficiency in compressors, clogs instrumentation and valves, and leads to flow losses.
Its sources have been understood by corrosion engineers, but the problem is not broadly understood in other parts of gas companies where decisions are made that affect the formation of black powder.
The material may be wet and have a tar-like appearance, or it can occur as a dry, very fine powder that is almost like smoke.
Black powder can be any of several forms of iron sulfide or iron oxide. It may be mechanically mixed or chemically combined with such other contaminants as water, liquid hydrocarbons, salts, chlorides, sand, and dirt.
In addition, while some pipelines have black-powder problems, others, usually smaller systems, do not.
Black powder has occurred in both dry and wet lines. Even in parallel lines, one line may show evidence of the problem while the other does not. No known pipeline has been able to eliminate the black powder problem once it has begun.
Iron sulfide has both a chemical and a microbial source in pipelines. Microbiologically influenced corrosion (MIC) is initiated by microbes that find a habitable environment in natural-gas piping. Iron sulfide is not easily filtered out of the flow stream, and it is pyrophoric (can catch fire) under some conditions.
No clear or universal solutions to eliminate the iron sulfide problem have been determined, but several approaches should aid in controlling it.
Sources
Black powder is created inside natural-gas wells and pipelines. The components and conditions necessary to create it can be found at some point in many pipelines.One of two mechanisms creates iron sulfides:
- Chemical reaction of constituents present in the pipeline, usually hydrogen sulfide
- Microbial assimilation of chemical constituents in the pipe and the production of both iron sulfides and pipe-wall pitting.
Iron sulfide, and apparently many of its variations, can quickly and efficiently occur in a gas pipeline from the chemicals naturally available in many systems. Hydrogen sulfide (H2S) easily reacts with the iron in piping to form iron sulfide:
H2S + Fe ? FeS + H2 (1)
The presence of a small percentage of water (approximately 3-9%) aids this reaction. Under more oxidizing conditions, pyrite (FeS2) may form:
2H2S + Fe ? FeS2 + 2H2 (2)
In addition to hydrogen sulfide, sulfur can also react to form iron sulfides. Sulfur compounds are abundant in gas and oil wells.
Iron sulfide, either attached to the wall of a pipe or collected in the bottom of a pipe, increases roughness, decreases flow area, and increases pressure drop. Over time, it thins the pipe wall and reduces the margin of safety.
Microbiologically influenced corrosion is a serious form of corrosion that can occur in pipelines to produce black powder or iron sulfide. MIC is pitting of the pipe wall because of the activities of microbial communities in areas that provide their required habitat.
The microbes that cause pipe corrosion are of two families: sulfate-reducing bacteria (SRB) and acid-producing bacteria (APB).
SRB microbes consume sulfates and produce hydrogen sulfide. These anaerobic bacteria use the reduction of sulfate as a source of energy and oxygen, according to such reactions as:
2H+ + SO4-2 + CH4
(H2S + CO2 + 2H2O) (3)
In transmission and distribution lines in which the gas has been treated to remove H2S, presence of H2S in amounts greater than the maximum allowable likely indicates microbes, although it could also mean the failure of a treatment plant.
Low molecular-weight mercaptans may also form iron sulfides, and some natural-gas streams (for example, from West Texas or New Mexico fields) have high natural mercaptans content. These usually are not removed because they serve as odorant for distribution.
Although SRB directly produce H2S instead of FeS, the conversion of the first to the second is direct and prompt at the site of the microbial activity where iron is present. In a dry or otherwise inhospitable environment, SRB can become largely dormant.
When injection waters used for secondary recovery (or other sources of water) enter the pipe or well formation, massive activation of these and other microbes can occur, plugging the pores of the deposit, the piping mechanisms, and the instrumentation with organisms and metal-sulfide precipitates.
APB consume organic nutrients and produce short-chain volatile fatty acids (VFA). Carbon dioxide and hydrogen are produced by organic fermentation.
SRB consume VFA as a carbon source and produce acetic acid and carbon dioxide in addition to hydrogen sulfide; therefore, the two microbial communities support each other and frequently co-exist at favorable sites. Although APB do not produce black-powder components, they are a reliable indicator of SRB that produce iron sulfide. APB may form oxides, depending upon the pH of the liquids present.
VFA are key natural components of reservoir fluids and the nutrient source for microbial species. Introducing nitrates into a reservoir can change the environment to favor indigenous and beneficial denitrifying bacteria (DNB). DNB are more voracious competitors for the VFA food supply and thereby inhibit detrimental SRB growth.
DNB produce nitrogen and benefit the reservoir by increasing the well pressure, preventing SRB reservoir plugging, and preventing hydrogen sulfide production. This process, referred to as "biocompetitive exclusion,"1 2 may aid sour reservoirs and storage facilities that tend to sour.
Filtering
The most common way of dealing with black powder is to filter it immediately before it enters a compressor station or processing plant. Filters are usually specified and supplied as a part of the compressor design package when the unit or piping is installed.Regardless of what is requested or promised, it is usually impossible to obtain field measurements that accurately determine the performance of filters and separators at any one condition.
The only identifiable measure of filter performance is an increase in pressure drop measured across a particle-removal device, indicating the filter is collecting material and beginning to clog.
It is difficult to identify the quantity and description of what gets through the filter. Ideally, filters are cleaned or back-flushed at optimum points indicated by pressure drop to minimize flow resistance, penetration, or overfill.
Several designs exist for removing solid or liquid material from a gas pipeline. Many trap particular materials; not all are good for every type of material or service. Because iron sulfide shears so easily, it frequently breaks down into submicron-sized particles, which require even-finer mesh filters.
Along with the filtration of iron sulfide, filters should be able to handle liquids that are present in many systems either as water or hydrocarbons.
Filtering (more about which presently) is a necessary but short-term, superficial solution and does not address root cause or offer a permanent solution.
Gas quality
Natural gas, as a mixture of hydrocarbons, sometimes contains non-combustible material components, in a gaseous state. The amount of each component in the gas stream can significantly affect measurement, pipeline operation and efficiency, and customer use.The objective of gas-quality standards has been to maximize heating value, but it is also necessary to minimize detrimental effects, such as corrosion and damage to pipeline equipment. Trace components that have negligible influence on heating value can cause formation of iron sulfide (black powder) and resulting equipment problems.
Of the generally desirable hydrocarbon components, iso-pentane and heavier components can drop out as free liquids at some point in normal operating conditions. Liquids of any source are not desirable because they cause equipment problems.
Other naturally occurring or frequently found components are hydrogen sulfide, sulfur, carbon dioxide, water, and nitrogen.
All occupy space and absorb energy in being moved through the system, thus reducing pipeline efficiency. They either do not support combustion or absorb energy when the natural gas is burned.
All except nitrogen can form acidic compounds that contribute to corrosion.
Hydrogen sulfide and sulfur, as major constituents in the formation of iron sulfide, are of primary interest in efforts to control black-powder formation. Some pipelines reportedly allow hydrogen sulfide as high as 1 grain/100 cu ft of gas, but 0.25-0.3 grains is more common.
Carbon dioxide is a strong acid former and can cause significant pipeline corrosion where water dropout occurs. Allowable levels for this component may be as high as 3 vol %.
Free oxygen is not naturally occurring in natural gas but can intrude through leaks in low-pressure systems. It can cause significant corrosion in small concentrations and combustion in larger portions.
The interaction of oxygen, carbon dioxide, hydrogen sulfide, and water in combinations in natural-gas pipelines is discussed in detail in PRCI-funded research. That important work defines some limits for these undesirable natural-gas components.
Oxygen concentrations of 10 ppm (vol) have little effect on steel corrosion in the presence of stagnant water inside pipelines transporting natural gas, while 100 ppm (vol) produces fairly high corrosion rates.
Transmission-pipeline operators should consider limiting maximum oxygen concentrations to 10 ppm (vol).
Significant corrosion of steels can occur at carbon dioxide partial pressures as low as 7 psi, or about 1.4 vol %. No minimum at less than which pitting will not occur has been established. Normal carbon dioxide gas quality levels are about 2-3%.
Gas-composition testing involves collecting a sample of the gas from a specific location in the pipe and testing for its constituent gases. Normally, liquids and particularly solids are not collected and in fact are even avoided.
Many gas-composition tests look only for hydrocarbons and a few other common constituents to verify compliance with heating-value specifications. These components (methane, iso-butane, propane, ethane, and others) normally make up 100% of the sample at levels down to hundredths of a percent.
Sulfur, hydrogen sulfide, oxygen, and water vapor, however, some in parts per million, can be important in the formation of iron sulfide and other corrosion products.
It is important to test gas-composition samples for these lesser components to monitor the likelihood of iron sulfide formation.
This effort to identify and remove the small quantities of these iron sulfide-producing chemicals may be the most effective approach to preventing formation of iron sulfide and its damage.
Even within the limits of these reasonable restrictions on trace components, consider the following mass balance: Hydrogen sulfide at 0.25 grains/ 100 std. cu ft (a common specification limit) in a continuous gas stream of 10 MMcfd will produce more than 3,500 lb of iron sulfide in a year (if all is converted to iron sulfide).
Thus, even gas that meets many H2S specifications can produce large amounts of FeSx and higher quantities can make the situation much worse.
Testing
Water is necessary for microbes to thrive and multiply. It is also a factor in direct oxidation of the walls of carbon steel pipe. MIC cannot occur in the absence of water.Changing atmospheric temperature or other pipeline environmental temperatures can cause water vapor to condense quickly and collect at the low points.
If water is not drained off frequently, it can cause the microbial spores that may be dormant in the pipe to bloom grossly and rapidly and cause MIC.
SRB and APB prefer to exist at the interface between a water puddle and a drier area. They create what is referred to as a "biofilm."
Thus, the most likely locations to find MIC evidence are at low places in piping where water is likely to collect. Avoidance or frequent elimination of these sites can help avoid pipe damage and iron sulfide creation.
Ideally, the water should be sampled from a pipeline without exposing it to air and tested before the microbe count has a chance to change significantly.
Often, it is difficult or impossible to sample and test in the field with a high degree of accuracy. In this case, as much care as possible should be taken to minimize oxidation or contamination of the sample and to reduce the time before it is tested.
The water sample should be tested for microbial content and the type determined by culture.
If there were no water in a pipe, it would be unnecessary to test water for evidence of the problems it causes. If there is a low place in the pipe that collects water, it should be removed frequently to reduce chances for corrosion, microbial blooming, and biofilm growth.
When black powder is found in piping, instrumentation, or compression equipment, it is sometimes helpful to verify its composition by chemical testing to identify or verify its likely origin. The origin could be chemical combination or microbial corrosion.
The most positive indication of MIC is the presence of pits in the pipe wall.
Normally, a larger pit will have smaller individual pits inside it. In more-developed cases, the pits may overlap each other. Striations and tunneling parallel to the pipe axis are also pointers to MIC, but it can also occur without these signs.
Sometimes a nodule of material, composed of the microbes and the material they consume and deposit, covers the pits. As pits deepen and become more numerous, individual pitting becomes less obvious.
Purely chemical formation of the iron sulfide is indicated by presence of iron sulfide and hydrogen sulfide without the identification of short-chain fatty acids.
An easy test to confirm black powder as a form of iron sulfide is to acidify it by placing a portion of the powder in a tube of hydrochloric acid. The hydrogen sulfide smell (rotten eggs) is proof of the presence of sulfide.
Removal, handling, disposal
Once black powder is generated in a pipeline, removal, handling, and disposal must continue until the material is prevented from forming. Removal. As stated previously, the most common method for removing black powder from a pipeline is filtration at a compressor station.
Most iron sulfide tends to adhere to the pipe wall where it forms. In some cases, it is held in place by an encrustation of salt. The powder adheres to wet surfaces and collects in pools of water or hydrocarbon.
Factors that affect the movement of iron sulfide are dryness, changes in flow rate induced by pressure changes, and abrasive actions that shear the powder from pipe walls.
This cleaning action is accomplished by abrasives moving through the pipe, water or other liquids, particularly in slugs, or pigging.
Liquids are commonly used to wash down piping and remove iron sulfide.
Although inexpensive and common, water can support MIC growth and oxidation, if left in the pipe. Diesel fuel or alcohol have been used as cleaning agents because they will not promote microbial growth and are relatively inexpensive and disposable.
Commercial suppliers offer other cleaning agents formulated specifically for cleaning piping of black powder.
Pigging involves moving an object with scraping action through the pipe by gas pressure. The cleaning action of the pig is to push loose material ahead of it and out of the pipe.
At the 1996 Southern Gas Association black-powder summit, one of the more-experienced companies indicated that using a heavy pig was more effective in breaking loose and moving black powder. Other companies indicated scraper pigs were effective, while others had less faith in their capabilities.
Participants related that, after several pigs were sent through a line, an intelligent pig passing through would then pick up much black powder. Many iron sulfides are magnetic, and after being broken loose from the pipe wall into small pieces, the fine iron sulfide particles are apparently attracted to the magnetic field around an intelligent pig.
Pigs are also helpful in removing water and other liquids collected in piping. The pig pushes the liquid down the pipe, sometimes causing it to combine with powder to form a sludge.
Sections of piping that cannot be pigged or otherwise opened for cleaning and inspection are candidates for MIC growth to the extent of pipe penetration. Collection of water in such a section or "deadleg" is an even greater inducement to MIC and oxidation corrosion.
One difficulty with inserting anything, solid or liquid, into the pipe for cleaning is that it must have means of getting in and being removed.
Such means and location strongly limit the ability to clean piping where the design does not support that capability. This applies to pigging as well as liquid washing.
Construction of deadlegs, side branches, and unpiggable piping should be avoided from the standpoint of black-powder formation. Even instrumentation lines should include a means of isolating and cleaning out the line to prevent clogging and loss of signal.
Handling; disposal. Once black powder is removed from piping, whether by filter cleanout, pigging, or other access to piping, compressor, or instrumentation internals, some attention must be given to handling the powder safely. The main hazard is the material's flammability.
The greatest hazard exists when the material is dry and fine. Collecting the material in aggregates or chunks causes less risk of combustion because the surface-to-mass ratio is much smaller. Wet black powder, whether from water or hydrocarbons, is less likely to combust because the liquid serves as a heat sink.
General practice has been to water down the powder as it is removed from piping or filters to retard combustion. But when the material dries out, the risk of combustion returns. This may occur in the first depository after removal from the pipe, during transportation when it may be exposed to forced draft that induces drying and oxygen exposure, or upon deposition in a landfill or other "permanent" site.
There are several reports of black powder smoldering or catching fire in the back of a truck transporting it to a landfill. Some pipelines report they store removed filters in a steel box until they have smoldered and stopped before taking them to a landfill.
Potassium permanganate (KMnO4) is unique among oxidants that could be used to treat iron sulfide. It is safe to use, easy to apply, and usually disposable.
When applied as a solution (<4%), it is relatively harmless if it contacts the skin, requires no special materials of construction, and does not form harmful or potentially explosive byproducts (OGJ, Feb. 24, 1997, p. 65).
References
- Giangiacomo, L.A., and Dennis, M.D., "Field Testing of the Biocompetitive Exclusion Process for Control of Iron and Hydrogen Sulfides," SPE paper 38351, May 18-21, 1997, Casper, Wyo.
- Hitzman, D.O., Sperl, G.T., and Sandbeck, K.A., "Method for Reducing the Amount of and Preventing the Formation of Hydrogen Sulfide in an Aqueous System," U.S. Patent No. 5,405,531, Apr. 11, 1995.
- Lyle, Jr., F.F., Carbon Dioxide/ Hydrogen Sulfide Corrosion Under Wet Stagnant Gas Pipeline Conditions in the Presence of Biocarbonate, Chloride, and Oxygen, Annual Report PRC PR-15-9313 for the Pipeline Research Committee International of the American Gas Association, June 1996.
The Author
Richard M. Baldwin is a senior research engineer in the mechanical and fluids engineering division at Southwest Research Institute, San Antonio. For the past 3 years, he has led Gas Machinery Research Council research into ingestion of foreign materials by pipeline compressors, with particular emphasis on iron sulfide formation and microbial-influenced corrosion. He has a BS (1967) in engineering science from Trinity University, San Antonio, and an MSME (1970) from Georgia Institute of Technology, Atlanta. He is a registered professional engineer in Texas and a Fellow of the ASME.
Joint black-powder research
Recent research into the problem of black powder in gas pipelines has been a cooperative effort by gas-pipeline companies, sponsored by the Southern Gas Association and its subsidiary the Gas Machinery Research Council (GMRC) and carried out by the mechanical and fluids engineering division of Southwest Research Institute (SwRI), San Antonio.The effort began when black powder was on the agenda at the 1996 Southern Gas Association's Offshore Gas Operations Conference. Pipeline company representatives found they had a common problem with the material. Some companies said they had had the problem for years but it was getting worse; others reported recent first-time encounters.
At the conclusion of the conference, Tennessee Gas Pipeline asked SGA to organize a forum later that year to explore the issue further.
Approximately 50 representatives met in December 1996 at Lafayette, La. Anecdotal evidence presented at the meeting persuaded Pete Harrell, senior research engineer at SwRI, to examine the problem to see if its extent and possible solution could be determined.
Investigations were made into the characteristics and sources of the material; pipeline filtering; sensors for particles in the gas flow; gas quality controls; gas, water, and powder testing; and powder removal, handling, and disposal.
It was found that considerable work had been done on the cause of the problem and treatments for it, but that the information mainly resided with corrosion engineers, research labs, and commercial vendors of black-powder control products such as filters, biocides, pigging, and chemical laboratories.
More knowledge was needed in the operations, gas control, maintenance, management, and tariff portions of the companies. The problem is too broad to be controlled by a few corrosion people in a company when many others have influence on the factors that control black-powder formation.
SGA sponsored a Black Powder Round Table in Houston in May 1997 where SwRI presented findings to date. The data were based in large part on research already under way at SwRI on foreign-material ingestion by compressors, sponsored by SGA subsidiary Pipeline and Compressor Research Council (now the GMRC).
"Having pipeline engineers from around the country discussing black powder helped a lot in defining the problem," said Marsha Short, director of operating and member services for SGA/GMRC. "A suggestion that the problem got worse the closer the system was to Louisiana fields was rejected when a pipeline company in Michigan reported significant problems with black powder."
Although attendees to the conference were all from U.S. companies, evidence was found that the black-powder problem exists in plants and pipelines in many countries.
The research into black powder was combined with the foreign ingestion study. Facilities operated by several SGA member companies served as test sites during this research.
The complete report is available at no charge from the Gas Machinery Research Council's home page, www.gmrc.org, and at SwRI's web site, www.swri.edu.
Copyright 1998 Oil & Gas Journal. All Rights Reserved.