Chapter 3
Selection of Corrosion-Resistant Alloys
Testing for Resistance to
Testing for Resistance to
Coatings
Cathodic Protection
Types of Cathodic Protection Systems
Application of Cathodic Protection to Oil Field Equipment
Inhibitors
Selection of Inhibitors
Application of Inhibitors
Nonmetallic Materials
Joining Methods
Advantages and Disadvantages
Typical Applications
Environmental Control
Mechanical Methods
When larger volumes of water are to be treated, it may be more economical to produce SO2 gas by burning sulfur. This gas is then dissolved in a sidestream of the water to be treated, pumped through the packed column, and then back into the main line.
Oxygen Exclusion
Corrosion Control Methods
Selection of Corrosion-Resistant Alloys
Traditionally, carbon and low-alloy steels were virtually the only metals used in the production of oil and gas. This was because large quantities of metal are required in petroleum production, and until a few years ago, crude oil and gas were relatively low-value products. In addition, insurmountable corrosion problems were not encountered.
This situation changed when gas and oil prices increased drastically and deeper wells were drilled that encountered corrosive environments of greatly increased severity. The final factor that made the current widespread use of corrosion-resistant alloy possible was the development of high-strength forms of these alloys. This allowed thinner pipe and vessel walls and greatly reduced the amount of material required.
The result of this situation is that essentially all high tonnage uses of corrosion-resistant alloy in oil and gas production involve alloys in high-strength form. Yield strengths typically span the range of 550 to 1250 MPa (80 to 180 ksi), but can reach nearly 1750 MPa (250 ksi) in wire lines. Table 2 lists the compositions of some corrosion-resistant alloys used in oil and gas production.
Alloy selection, from a corrosion standpoint, can be considered to be a three-step process. First, resistance to general corrosion must be ensured. This is primarily a function of the chromium content of the alloy. Second, resistance to localized attack also must be ensured. This is primarily a function of molybdenum content. Finally, resistance to environmental stress cracking is sought at the highest feasible strength level. Nickel content plays a principal role in this instance, particularly in providing resistance to anodic cracking.
The close correlation between pitting resistance and resistance to anodic cracking should be noted. This apparently results from the ease of crack initiation under the low-pH high-chloride conditions found in pits. Therefore, higher molybdenum can also increase resistance to anodic cracking.
With the procedures given below, regions of alloy applicability can be shown schematically as a qualitative function of environmental severity.
This has been attempted in Fig. 12, in which an aqueous, C02-containing environment (hence low pH) has been assumed and the effects of temperature, chloride, and 1125 concentration are illustrated. The effect of yield strength is not shown, but if environmental cracking is the limiting factor, reducing the yield strength should extend applicability to more severe environments.
The reader should be cautioned that a diagram such as Fig. 12 is really more of a guide to alloy qualification than to direct selection for a particular application. Therefore, it may aid in developing a more efficient approach £o alloy testing.
Testing for Resistance to
Environmental Stress Cracking
The most directly applicable results are obtained by exposing samples of commercially produced alloys to an environment simulating as closely as possible that expected in actual production operations. Fortunately, an understanding of the principles involved allows considerable simplification to be nude without significantly altering the value of the results.
Testing for Resistance to
Environmental Stress Cracking
The most directly applicable results are obtained by exposing samples of commercially produced alloys to an environment simulating as closely as possible that expected in actual production operations. Fortunately, an understanding of the principles involved allows considerable simplification to be nude without significantly altering the value of the results.
Two simplifications can be readily made in the environmental parameters, as follows. First, only the CO2 and H2S partial pressures are usually reproduced. The overburden of methane pressure to create the actual total pressure of the environment is dispensed with. This can substantially lower the pressure ratings of test vessels. Second, only the chloride content of the brine is typically reproduced, rather than trying to simulate the total ionic spectrum of the produced fluids.
Often, no reliable analysis is available. Assumption of a saturated sodium chloride solution is then a relatively conservative approach.
A wide variety of test specimens are used, but for high-strength corrosion-resistant alloy tubulars, C-rings are particularly convenient. Double cantilever beam specimens are very attractive because they can provide a quantitative measure of fracture toughness, which can then be used in mechanical design. Other specimens that can be statically loaded include U-bend specimens, tensile bars, and bent beams. The most widely used dynamic test is the slow strain-rate test. The advantages and disadvantages of different test methods are reviewed in Ref 29.
U-bends are made from sheet and bent into a “U” shape to stress the material plastically. The specimen is usually held in this position by a bolt while taking care to ensure electrical isolation between the two if they are not of the same material. The technique is discussed in ASTM G 30 (Ref 30). Specimens are exposed to the environment for the desired time and then examined with metallographic techniques. This test provides information only on whether or not the material cracked during the exposure. The advantage of this test is that a large number of specimens can be run at the same time, which lends itself to the screening of various alloys.
Tensile bars, C-rings, and bent beams are all stressed to known values, which are usually reported as a percentage of yield strength. Use of any of the three of these specimen types results in a threshold stress below which cracking is not expected. This is a useful way of ranking materials; a material with a high threshold is assumed to perform better in held environments.
Of the three specimen types listed above, only the tensile bar is associated with a standard test method for sour environments. However, a NACE committee (T- 1 F-9) is currently developing suitable standards for all three specimen types.
A great deal of testing has been conducted using tensile bars in accordance with the NACE TM-0l-77 test method (Ref 31). This method uses a test solution (called NACE solution) consisting of 5% NaCl, 0.5% acetic acid, and 1 atm H2S being bubbled through the solution at room temperature. The test is run for 720 h with the time-to-failure being plotted at various stress levels. From this plot, a threshold stress can be determined. The test specimens can be cut from almost any product form.
C-rings are made from tubular products and are bolt loaded to the desired stress level. ASTM o 38 (Ref 32) gives a procedure for making and stressing these samples. These specimens are usually placed in environments that more closely simulate field conditions than does the NACE solution. This requires the use of high-temperature and high-pressure autoclaves. The specimens are examined metallographically at the conclusion of the test for evidence of cracking. Again, a threshold stress can be determined for the exposure period in order to rank alloys.
Bent beams can be stressed in three-point or four-point bending. The stress level is again reported as a percentage of yield strength, and samples are tested in autoclaves to determine the susceptibility to cracking of the material over some particular time period. Beams can be machined from almost any product form.
Double-cantilever beam specimens are fracture mechanics specimens that allow development of crack-propagation data after the samples are exposed to the test environment (Ref 33). Other types of fracture mechanics specimens could be used, but the double-cantilever beam is the most common and probably the most convenient. These specimens are precracked by fatigue loading, then wedge-loaded to develop an initial stress intensity at the crack tip. If the crack propagates during the test, the stress intensity decreases, which reduces the driving force for further crack propagation.
After a certain period of time, the crack is arrested. By measuring the length of the crack at arrest, a threshold value for the stress intensity can be calculated. This value is rarely used in minimum flaw tolerance design of drilling equipment. Instead, it provides a numerical value to be used for ranking alloys.
The slow strain-rate test, or constant extension rate test (CERT), is a technique that is used to determine the susceptibility of a material to cracking under dynamic loading conditions. The specimen is a tensile bar with a gauge diameter of about 2.5 mm (0.1 in.). A specimen is tested to failure in the environment of interest, and another is tested in an inert environment, such as air. Strain rates between 10-4 and 10-8 s-1 can be used, with rates between l0-6 and 4 ´l0-6 s-1 be-lag common.
The slow strain-rate test is often considered to be a severe test because it will usually indicate embrittlement when other tests do not. This may be related to the fact that any passive films on the specimen surface are continually broken during the test, which exposes fresh surface to the environment. Detailed information on environmental cracking test specimens, procedures, and results can be found in the article “Evaluation of Stress-Corrosion Cracking” in Volume 13 of the ASM Handbook.
The National Association of Corrosion Engineers has published a procedure for testing steels to determine their hydrogen-reduced cracking (HIC) resistance. Standard TM-02-84 (Ref 34) calls for exposing unstressed coupons to a synthetic seawater solution saturated with hydrogen sulfide at ambient temperature at a pH between 4.8 and 5.4. The test is run for 96 h and is designed to accelerate the formation of cracks.
At the conclusion of the test, the samples are polished, and any cracks are measured at 100x. Crack thickness, length, and sensitivity ratios are calculated and are used to rank the various materials.
Coatings
Internal protective coatings have been used to protect tubing, downhole equipment, wellhead components, Christmas trees (manifolds used to control the rate of production, receive the produced fluids under pressure, and direct the produced (kids to the gathering point I, and various downstream how lines and pressure vessels for more than 30 years. Because internal coatings are subject to damage, successful use is usually accompanied by chemical inhibition or cathodic protection as part of the entire protection program. Most of the coating use has been below 175 oC (345 oF).
Tubing. The benefits derived from coating tubing depend on the coating remaining intact. Because no coating can be applied and installed 100% holiday-free, inhibition programs are commonly employed to accommodate holidays and minor damage. The suitability of the service is dependent on specific testing and an effective quality control program.
Inhibitor and volume will not change even though an operator decides to use coated tubing rather than bare tubing. The use of coated tubing improves the protection in shielded areas that are inaccessible to inhibitors. The two greatest dangers to coated tubing are wireline damage and improper joint selection. Wireline damage can be minimized by adjusting running procedures to include wireline guides and to slow wireline speed (<0.5 m/s, or <100 ft/min). Inhibiting immediately after wireline work is good practice.
Proper joint selection involves choosing a joint that allows coating to be applied around the pin nose into the first few pin threads and from the first few coupling threads into the coupling body. The proper joint allows the coating to remain undamaged. Often, a corrosion barrier compression ring is used to accomplish this end. Metal-to-metal sealing joint designs are not joints that can be coated.
The coatings used for tubing protection are polyurethane, phenolformaldehyde, epoxized cresol novolac, and epoxy resins. Suitability for service is and should be based on laboratory testing using the specific environment proposed for the service.
The quality control parameters of concern are tubing surface finish/preparation, application techniques, coating thickness, holiday detection, joint condition, and inspection. Inspection is required to ensure the suitability of the other parameters. Quality control and surveillance are as much a part of a successful protective coating program as choosing the appropriate coating. The production coating must be applied in the same way the coating was applied to the test specimens.
Coated pipe and couplings must be carefully handled after coating, during shipping, in storage, and at the well site. The threads must be protected from impact with other pipe and objects.
Wireline work is necessary. These operations can be accomplished with a minimal amount of damage to the tubing if the wireline speeds are kept to 0.5 m/s (100 ft/mm) or less, if all sharp edges are removed from the tools, if all tools are plastic coated or covered with a plastic sleeve, and if wheeled centralizers are used on all center hole jobs. Using the above precautions, many wireline trips can be made with little or no damage. The fact that coating should not be used because the wireline will cut the coating and cause accelerated corrosion in the wireline track is not true. When the wireline cuts uncoated tubing, it causes cold work. The wireline track then becomes anodic to the surrounding bare steel, and corrosion is accelerated. In coated tubing, the coating electrically insulates the cathodic areas so that the corrosion rate in the track is essentially the same as that for uncoated steel without wireline damage.
Wellheads, Christmas Trees, and Down-hole Equipment. Exposed surfaces of wellhead equipment, Christmas frees, and downhole equipment must be coated or manufactured of corrosion-resistant materials. This starts with the tubing hanger, which is threaded onto the production tubing, and continues through the tubing bonnet (tubing adapter), the master valve(s), the tee or cross, and the wing and crown valve(s) and into the choke. Ring gasket grooves, valve seat pockets and other compression fitted parts must not be coated. These areas can be overlaid with corrosion-resistant alloys, and the coating can be applied over a transition area up to the corrosion-resistant alloy overlay. Valve internal cavities need not be coated.
Generally, tubing hangers are difficult to coat and are therefore made of corrosion-resistant alloys for corrosive service. Hangers without back pressure valve threads can be coated, but the cost of coating the carbon steel hanger may be equivalent to the cost of a corrosion-resistant alloy hanger.
Downhole equipment (nipples, polished bore receptacles, seal subs, tie backs, millout subs, packers, and so on) use the same standards as tubing. Most downhole equipment is considered uncoatable because it was not designed for coating. Such equipment is usually coated as an assembled unit, rather than as individual pieces. If installation is below the packers, both internal and external surfaces should be coated.
Vessels. The vessel must be designed as a coated vessel. Coating a vessel that was not designed for coating is rarely successful. Vessel design for coating involves welded hangers for anodes in the fluid zone, flanged access, removable internals, smooth internal surfaces, and good access to all internal surfaces to be coated. As a general rule, coating over bolted assemblies is not successful.
Cathodic Protection
Corrosion occurs when an anode and a cathode are electrically connected in the presence of an electrolyte, and electrical currents leave a metal and go into an electrolyte. If another power source can be used to oppose these corrosion currents sufficiently, the metal will be protected from corrosion. This technique is known as cathodic protection. The entire metal surface is converted into a cathode, while the corrosion currents are transferred to an auxiliary anode in which corrosion can proceed. Cathodic protection has been used in the oil field to protect pipelines, well casings, tanks and production vessels, and offshore platforms. More information on principles and applications of cathodic protection is available in the section “Anodic and Cathodic Protection” of the article “Corrosion Protection Methods for the Petrochemical industry” in this Volume.
Types of Cathodic Protection Systems
There are two methods of applying eathodic protection: the sacrificial anode method and the impressed-current method. Because some metals are less noble (more electronegative) than the most common oil field material-steel-in the galvanic series, they will become the anode (site of corrosion attack) when coupled to steel in the presence of an electrolyte. The most common of these materials are magnesium, zinc, and aluminum; these are called sacrificial anodes. These types of anodes are used when:
• Current requirements are relatively low
• Electric power is not readily available
• Short system life dictates a low capital investment
The anode is usually electrically connected by a wire or steel strap to the structure to be protected. Magnesium and zinc are usually used in soils, while zinc can also be used in brine environments.
In the impressed-current method, an external energy source produces an electric current that is sent to the impressed-current anodes. The most common types of these materials include graphite, high-silicon cast iron, lead-silver alloy, platinum, and even scrap steel rails.
These types of anodes are used when:
• Current requirements are high
• Electrolyte resistivity is high
• Fluctuation in current requirements will occur. These types of systems can be adjusted to compensate for varying current requirements
• Electrical power is readily available, although this is not now as severe a limitation as it was in past years
In a typical impressed-current system, alternating current from a power line flows into a rectifier where it is converted into direct current. The dc current then flows to the anode groundbed. Other means of supplying this electrical current include solar energy and thermoelectric generators. These methods are applicable at locations where conventional electric power is not economically available.
Solar energy has powered cathodic protection for well casings in Kansas (Ref 35) and Saudi Arabia (Ref 36); segments of a 480-km (300-mile) long, 0.8-rn (32-in.) pipeline in Libya (Ref 37); and segments of a natural gas distribution system in Washington (Ref 3 8). In these systems, silicon semiconductor devices convert sunlight directly into dc electricity, which is then used for the anode groundbed and to charge batteries. These batteries provide current to the anode groundbeds during periods of little or no sunlight. The batteries must be checked periodically for proper electrolyte levels. Solar panels can be easily replaced or increased to attain a higher current output because they are fabricated in modules.
One oil company uses solar energy to protect about 800 well casings in western Kansas
A rheostat controls the rate of current flow from the batteries, and the voltage regulator controls the battery charge rate. The batteries can provide electrical current to the anodes for 10 days even if the sun is completely blocked.
Another unconventional electrical current source is the thermoelectric generator. One type of thermoelectric generator is a system that uses a burner to heat an organic liquid in a vapor generator. This vapor then expands through a turbine wheel, thus producing power to a shaft to drive an alternator, where ac power is produced. The vapor then passes through a condenser where it is cooled and condensed back into a liquid to start the cycle again. Once the an power is produced by the alternator, it is then sent to the rectifier to be converted to dc. Figure 13 shows an illustration of this type of thermoelectric generator. These systems can produce a maximum of 80 Aat2l V(Ref 39).
Application of Cathodic Protection to Oil Field Equipment
Pipelines. Cathodic protection of pipelines is very common in oil field operations. As discussed earlier in this article, sacrificial anodes or impressed current can be used for cathodic protection (Fig. 14). If the pipeline is well coated and not very long, the current requirements will probably be achieved with sacrificial anodes. If bare, a steel pipeline could require 1.1 mA/m2 (0.1 mA/fl2) in soil, while a very well coated pipeline could require only 0.003 mA/m2 (0.0003 mA/fl2) or less for cathodic protection (Ref 40).
The resistivity of soils (and therefore their corrosivity) will also vary with location. Differences in aeration, soil composition (sand or clay), and the presence of chemical spills are just a few of the factors that will affect the corrosivity of the soil. Sometimes, the resistivity of the surface soils is so high that conventional groundbeds (1.8 in, or 6 ft, deep) cannot be used. Conventional groundbeds are normally used when the soil resistivities near the surface of the ground are less than 5000 W . cm (Ref4 1).
In high-resistivity soils, deep groundbeds can be used where the anodes are installed vertically in holes at depths of 15 in (50 fl) or more. Deep groundbeds (Fig. 15) are normally used to provide a better distribution of current than conventional groundbeds. These types of groundbeds also minimize right-of-way problems and are essentially unaffected by seasonal moisture variations. They are more expensive to install than conventional groundbeds, and it is usu~l1y impossible to repair any damage to the cable insulation. These systems use the same type of anodes as the conventional groundbeds. The major difference is that a perforated vent pipe can be installed to prevent chlorine gas from accumulating around the anodes. If these gases collect around the anodes, they form an insulating barrier that increases the resistance of the groundbed and eventually causes the groundbed to become ineffective.
Well Casings. The first step in externally protecting well casing is to cement through any corrosive zones. Cement acts as a coating and will significantly reduce, but not completely stop, corrosion of the casing. Therefore, cathodic protection is needed to supplement the cement (Fig. 16). The experience of one company with the cathodic protection of well casings showed an 88% success rate in preventing predicted casing failures (Ref 44). Although a single anode bed for a buried pipeline may protect as much as 80 km (50 miles), the maximum amount of casing that needs to be protected usually does not exceed 2.4 to 3.2 km (1.5 to 2 miles). One company coated nine casing strings in 3500-m (1l -500-ft) wells with fusion-bonded epoxy in Florida
Another method of reducing the current requirement is to place an insulating joint in the flow line at the wellhead. This joint prevents current from the flow line from flowing into the wellhead and down the casing. This current would leave the casing at low-resistivity zones (Ref 43).
Two methods are generally used to determine current requirements for well casing: casing potential profile and E-log I. The casing potential profile is measured by using a tool that consists of two spring-loaded probes approximately 7.6 m (25 ft) apart. This tool is pulled through the casing, and voltage readings are taken between the probes as they contact the casing every 15 or 30 m (50 or 100 ft). A plot of potential versus depth is made (Fig. 17). A slope upward and to the left indicates an anodic area, while a slope upward and to the right indicates a cathodic area. Current from a temporary groundbed is then applied to the casing for protection, and another potential profile is taken. Current is increased until the profile slopes are upward and to the right. Providing a profile slope to the right does not necessarily mean that all of the casing is protected, but it does mean that all gross corrosion areas have been eliminated.
Another technique for determining the required current is the E-log I curve. It is less expensive and does not require the disturbance of subsurface equipment. The flow line to the well, however, must be isolated from the well casing. The pipe-to-soil potential relative to the Cu-CuSO4 reference electrode is measured, and a small amount of current from a temporary groundbed is applied. The current is then interrupted, and the pipe-to-soil potential is measured as quickly as possible.
The current is then increased a small amount, and the process is repeated to obtain a curve similar to that shown in Fig. 18. Generally, the current required corresponds to the break in the curve.
Tanks and Production Vessels. Internal corrosion of water-handling tanks and vessels can be controlled by the use of cathodic protection. Even if a coating has been applied to the interior of a water storage tank, there will always be imperfections where corrosion can occur; therefore, cathodic protection is needed. Sacrificial anodes can be suspended from the top of the tank as shown in Fig. 19 to offer protection to the portion of the tank that is covered with water. Cathodic protection will not help in the vapor area of the tank. Coatings must be used to protect this area. Sacrificial anodes have also been placed on concrete blocks in tanks. These blocks insulate the anode from the tank and allow decomposition products to fall away from the anode. In most cases, a lead wire is brought outside of the tank and welded to the tank. These anodes should also not be allowed to touch the sides of-the tank and should be uniformly distributed within the tank to give uniform current distribution.
In vessels that have several sections separated by steel plates, the anodes might be shielded from protecting all of the vessel. In these cases, the only safe procedure is to install an anode in each compartment. Fire tubes in similar tanks without cathodic protection in one southern Texas
Offshore Platforms. The subsea zone of an offshore platform includes the area from the splash zone to and including the pilings below the mudline. Cathodic protection is the principal means of preventing corrosion in this zone, but some companies also use coatings in conjunction with cathodic protection (Ref 48). The amount of electric current required to protect the bare steel varies with location. Typical current density values range from 54 to 65 mA/m2 (5 to 6 mA/ft2) in the Gulf of Mexico, 86 tol60mA/m2 (8 to 15 mA/ft2) in the North Sea, and as high as 375 to 430 mA/m2 (35 to 40mA/ft2) in the Cook Inlet (Ref 40). Current densities in the Cook Inlet are high because of the 8-knot tidal currents experienced there. In the mud zone, current densities of 10.8 to 32 mA/m2 (1 to 3 mA/ft2) are needed for protection, and an allowance of 3 A per well is customary for well casings (Ref 49). Environmental factors, such as oxygen content, water salinity, temperature, velocity, erosive effects, marine growth, and calcareous deposits, are largely responsible for the differences in current densities. It is very important that the current demand be conservatively estimated.
Partial protection of the steel in seawatcr usually means that the area of corrosion is reduced, while the unprotected areas continue to corrode at a high rate. Some companies report pits as deep as 13 to 16 mm (0.5 to 0.625 in.) and, in many cases, holes in platform members after less than 5 years on location without adequate cathodic protection. This result corresponds to a corrosion r4e of 2.5 to 3.2 mm/yr (100 to 125 mils~1yt) (Ref 50).
As in all applications of corrosion control on offshore platforms, the first step for cathodic protection in the subsea zone is design. Tubular members should be used whenever possible. Recessed corners in channels and I-beams are difficult to protect. Even crevices formed by placing channels back-to-back and noncontinuous welded joints cannot be protected. Bolted and riveted fittings should be avoided. Piping such as grout lines, discharge lines, water supply casings, and pipeline risers, if clustered around a platform leg, can cause shielding and interfere with the flow of cathodic protection current. If economically feasible, piping that is not necessary for platform operations should be removed. A minimum clear spacing of 1 1/2 diameters of the smaller pipe should be provided, and coatings on the pipe can also be used to minimize shielding. Corrosion will be negligible on the internal surfaces of structural members that are sealed and have no contact with either the atmosphere or the seawater. During launch, some structural members are flooded for the life of the platform. To prevent any internal corrosion, the flooding valves should be closed to isolate the flooded chambers from contact with fresh seawater or oxygen in the atmosphere.
Sacrificial Anode Systems. The early offshore platforms installed in the late 1940s and early 1950s used 45- and 68-kg (100- and 150-lb) magnesium anodes supported from horizontal braces. A low-carbon steel wire rope connected the anode to the brace. These anodes had a 2-year design life, which was normally shortened to I year or less because of hurricane or rough weather losses. These swinging anodes tangled with subsea braces, shorted, and rubbed the conductor wires to failure. A variable resistor was also connected in series with the anode and the connection at the brace. This resistor was used to regulate the current output of the anode to achieve maximum efficiency. Unfortunately, this system could never be maintained. Magnesium anodes in seawater have a high current output and corrode rapidly; therefore, they must be replaced frequently. This type of system has been discontinued for offshore use.
Zinc anodes have been used since the early nineteenth century. However, impurities, such as iron, were responsible for erratic performance. Virtually all zinc anodes are now fabricated from high-purity zinc meeting the military specification MIL-A-1 8001-H. More efficient aluminum anodes have also been developed. A mercury-zinc-aluminum alloy anode provides 21/2 times more current output than a zinc anode on a pound-for-pound basis. An anode weight of 330 kg (725 lb) is common for the initial system, while 150kg (325 lb) is common for the replacement system. Control of impurities in the anode is essential for best performance.
The number of anodes needed depends on the size of the anode and its useful life. A design life of 20 years is common. The distribution of anodes is also important, because poor distribution and the use of too few anodes will result in under-protection, particularly at welded joints. Individual anodes should be mounted at least 0.3 m (12 in.) from the structure, or a dielectric shield should be used beneath them to improve the current distribution. Anodes should not be located in either the splash zone or on bottom bracings. The anode will not function properly if it is intermittently in and out of the seawater, and mercury-containing aluminum anodes will passivate and not function if covered by mud. Some of the earliest anodes used were prone to being knocked off of the platform during installation because the standoff posts were either too small or did not provide adequate area for contact welding of the anode to the member. Gussets and doubler plates-can help obtain better anode attachment. A large fraction of premature cathodic protection failures have been traced to an inadequate number of anodes installed or excessive losses during pile driving. Some of these losses during pile driving are due to poor weld quality in attaching the anodes to the members.
Impressed-Current Systems. In an impressed current system, the three essential components are the rectifier, the anodes, and the cable joining them together. The anode materials include graphite, high-silicon cast iron, lead-silver, and platinum wound on a niobium rod. Permanently mounted anodes, retrievable anodes, and remote anode sleds have all been successfully used. Because anodes in impressed-current systems generally produce considerably more current than sacrificial anodes, there may be only 6 or 8 impressed anodes on a structure that might do the same job as 50 to 70 sacrificial anodes. The location of these impressed-current anodes is very important in order to ensure that adequate current distribution is obtained to cover the entire surface. The connecting wiring is the critical part of an impressed-current system, especially in the splash zone, where the cable can be subjected to severe wave pounding if it is not housed in a protective conduit. Even these conduits can be torn away from the platform during a hurricane if they have been under designed. It is often necessary to protect the structural member near the anode with fiberglass coating or a wrap called a dielectric shield. This shield prevents excessive current consumption at this area.
Whether sacrificial or impressed anodes are used, cathodic protection currents will promote the formation of hydroxyl (OW) ions at cathodic areas (the entire platform, it is hoped) and cause a pH shift in the seawater near the platform. Also, the concentration of calcium and magnesium ions tends to increase in the film of seawater over the cathode. As a result of these changes, the solubility of calcium carbonate and magnesium hydroxide is exceeded, and a calcareous coating is deposited. These mineral deposits provide the primary corrosion control, and the cathodic protection current demand drops to a level sufficient to repair this coating when it is damaged. For example, if a current density of 40mA/m2 (50 mA/ft2) is applied to a platform for the first 5 days on location, protection can be maintained with a current density of 32 mA/m2 (3 mA/ft2).
There appears to be less tendency for these mineral deposits to form in the deep ocean. In an experiment conducted by the U.S. Navy, sacrificial anodes were effective at providing cathodic protection to bare steel in seawater at depths of 1700 m (5600 ft) (Ref 51). However, the anodes were consumed more rapidly than if they were located near the surface. Because the pH is lower at great depths and the calcium carbonate concentration is below saturation, higher currents are required to achieve protection.
Inhibitors
Corrosion inhibitors are materials that, when present in a system in relatively small quantities, produce a reduction in metal loss due to corrosion attack. These inhibitors can interfere with the anodic or cathodic reaction, can form a protective barrier on the metal surface against corrosive agents, or can work by a combination of these actions. For oil field corrosion inhibitors, organic compounds containing nitrogen (amines) dominate because of their effectiveness and availability. These inhibitors usually contain three elements:
• One or more active inhibitor components
• A solvent base
• Certain additives, such as surfactants, dispersants, demulsifiers, and defoamers
Solvents are used to dilute inhibitors to control physical characteristics (such as viscosity and pour point), to aid in obtaining proper inhibitor concentration and placement during treating, to assist inhibition, and to maintain a reasonable cost per unit volume (Ref 52).
Physical Characteristics
Physical characteristics of inhibitors must be considered when evaluating a potential application. These include:
• Physical form
• Solubility
• Emulsion forming tendencies
• Thermal stability
• Compatibility with other chemicals
Physical Form. Inhibitors may take either a solid or liquid form. Solid inhibitors have been made in the shape of a stick that will sink to the bottom of a well and then slowly dissolve and be produced back. These sticks are rarely used. Most corrosion inhibitors are liquid form and have densities that range from 840 to 1440 g/L (7 to 12 lb/gal) (Ref 53). These liquids must not freeze when exposed to the coldest of field conditions and must be stable with a minimum loss to the vapor state when exposed to the hottest of field conditions.
Solubility. The formation of an inhibitor film and its life are primarily governed by the solubility of that product in the system. There are three categories of solubility: soluble, insoluble, and dispersible.
A product is soluble in a fluid when it forms a clear mixture that does not separate. A product is insoluble in a fluid when it will separate after mixing to form an identifiable layer. Materials are dispersible in a fluid if they form a mixture that is not clear and separates slowly, if at all.
Different solubilities are required for different applications of corrosion inhibitors. An inhibitor to be added continuously to a water-flood should be water soluble or highly dispersible.
Similarly, an inhibitor to be used for a squeeze treatment should be completely soluble in the carrying fluid to facilitate placement of the inhibitor without plugging the formation. On the other hand, where the only method of application is a periodic treatment, continuing protection requires some degree of insolubility of the inhibitor in the fluids to which it is exposed. In practical terms, this means that an inhibitor used in tubing displacement cannot be completely soluble in the well fluids. Also, a dispersion must be stable enough to remain intact until the inhibitor reaches the metal surface to be protected.
Emulsion-Forming Tendencies. Because of the chemical nature of most corrosion inhibitors, there is a positive tendency in water-oil systems to form emulsions. Some of these emulsions will break down quite readily, while others are extremely stable and practically impossible to break. When squeezed, an incorrectly selected inhibitor can form an emulsion in the formation that blocks or severely restricts further production.
The inclusion of a demulsifier in a corrosion inhibitor is no guarantee against the formation of stable emulsions. Produced fluids from each field must be tested to provide reasonable assurance that no stable emulsion will be formed upon application of a specific corrosion inhibitor.
Thermal Stability. Corrosion inhibitors generally have temperature limits above which they will lose their effectiveness and change their chemical composition. This temperature may be variable for any one inhibitor, depending on such conditions as pressure and presence of water. A typical example is that of an acid-amine salt. Under atmospheric conditions, this salt will yield water and form an amide at 70to 90 oC (160 to 190 oF) However, this chemical can be used in oil wells in the presence of water at these temperatures with no apparent degradation. Of course, exposure to high temperatures at low pressures will result in the vaporization of the solvent systems in these inhibitors.
Compatibility with Other Chemicals. The compatibility of corrosion inhibitors with other chemicals is ordinarily not troublesome when the inhibitor and the other chemicals are present in parts per million concentrations. However, chemical users frequently want to mix various chemicals so that a single chemical pump can be used for injection. Many products are not compatible with corrosion inhibitors, because of variations in solvent systems, type of chemicals (cationic versus anionic), and so on.
Most oil field corrosion inhibitors are cat-ionic to some extent: that is, they carry a positive electrical charge. Mixing a cationic inhibitor with an anionic chemical, such as a scale inhibitor or certain surfactants, will likely produce a reaction product that can have characteristics that are entirely different from those of either of its parent products. At best, the new material may function poorly: at worst, it may not function at all or may even form deposits in the system. When the two chemicals must be used, this problem can be prevented by using separate injection points that are not closely spaced. It should probably be standard practice never to mix any two different products. This would avoid any potential problems.
A final example of operating problems can be found where a conventional inhibitor is used in a gas stream upstream of a compressor. The nonvolatile components in the inhibitor could be left behind to foul the valves of the compressor.
Selection of Inhibitors
Many factors are involved in the selection of inhibitors, including the following:
• Identification of the problem to be solved
• Corrosives present
• Type of system (which influences the treatment method)
• Pressure and temperature
• Velocity
• Production composition
Although problems such as rod breaks and flow line leaks may initially be seen as purely corrosion failures, the actual cause of the problem could be oxygen, scale, or bacteria. Rod coupling failures could be caused by poor assembly or corrosion fatigue. Overstressing of rods greatly accelerates these failures. In such cases, mechanical measures could reduce or eliminate the need for chemical treatment. If attack is due to oxygen entry, installation of gas blankets or closing of the casing valve could greatly reduce corrosion.
The presence of corrosives such as H2S and CO2 greatly influences the choice of an inhibitor. Some inhibitors perform best in sweet fluids, while other inhibitors work best in sour fluids. Even the concentration of NaCl has a bearing on the choice of an inhibitor. With increasing NaCl content, some inhibitors will become insoluble and deposit.
The type of system also has an effect on the selection of an inhibitor. The correct inhibitor to use is determined by whether the system is a pumping oil well, a gas-lift well, a gas well, a waterflood system, or a flow line. For example, a weighted inhibitor is seldom recommended in dry gas wells, because water is required to release the inhibitor before it becomes effective (Ref 54). Also, when a gas-lift well is treated, the inhibitor is injected into the gas-lift lines. Therefore, the inhibitor must not have any tendency to form gunky deposits.
Both temperature and pressure have an influence on inhibitor selection. Bottom hole temperatures and pressures may get so high that inhibitors polymerize and form a sludge. Pressure influences the corrosivity of CO2 and H2S.
Velocity is yet another factor to consider. With pipelines, low velocity might be insufficient to displace water from low areas in the line.
In the case of dry gas pipelines with low velocity, a water-soluble inhibitor should be selected and should be injected continuously. If the velocity is high enough to prevent any accumulations of water in low areas of a dry gas line, then an oil-soluble inhibitor should be batch treated.
The composition of the produced water also determines the choice of inhibitor. Criteria such as water/oil ratio, salinity of water, and acidity of the water and oil are vital to the correct selection of the inhibitor.
Tests are conducted in the laboratory using the standard wheel test. Although not infallible, this test attempts to duplicate field conditions as closely as possible. Parameters such as temperature, water-cut, batch or continuous treatment, and whether the system is sweet or sour are reproduced as closely as possible to field conditions. Usually, these tests involve weight loss coupons.
The rate at which an inhibitor forms a film is completely dependent on the product and its environment. However, it can generally be said that film formation is a function of time and is not instantaneous. The concentration of inhibitor required to develop an adequate film is also directly related to the characteristics of the product and system. Many factors affect the dosage and frequency of treatment, including the following:
• Severity of corrosion
• Total amount of fluid produced
• Percentage of water
• Nature of corrodent
• Chemical selected
• Fluid level in the casing annulus
Because no laboratory test can take into account all of the conditions imposed by the oil well, the dosage and frequency of treatment must be constantly reviewed.
There are two general rules to follow for dosages. First, for continuous injection, a dosage of 10 to 20 ppm based on total produced fluid is used as a starting point. Second, for batch treatment, weekly batch frequency is used with a starting dosage of 3.8 L (I gal) per week for each 100 barrels of daily fluid production.
Also, if the corrosivity of the system is known, the following general criteria can be used to define more accurately a treatment rate for continuous injection (Ref 55):
Mild corrosion 10-15 ppm
Moderate corrosion 15-25 ppm
Severe corrosion >25 ppm
Moderate corrosion 15-25 ppm
Severe corrosion >25 ppm
Several major problems can occur with inhibitors, including foaming, emulsions, scale removal and plugging, and safety and handling. Corrosion of other metals can also be a problem.
The most appropriate action to take in avoiding difficulty from foaming is to determine where foam-forming conditions exist in the system. These will consist of places where the inhibitor-containing fluid is agitated with a gas, such as in a gas separator, a countercurrent stripper, or an aerator. The next step is to obtain a sample of the fluid and gas from the process step, add the inhibitor in question, adjust the temperature to that corresponding to the process step, and shake vigorously. If this test produces a stable foam, a potential problem exists.
There are three alternative remedies. First, an antifoaming agent can be added (this must be tested also); second, tests can be conducted to select an inhibitor that does not cause foaming; and third, the system can be shut down periodically and treated with a slug of persistent inhibitor. The last two remedies are the least palatable because the need for an inhibitor is at hand and there are few processes that can be shut down with sufficient frequency to maintain effective inhibition by slug treatment.
Emulsions are another problem that can occur when the wrong inhibitor is used. The use of other chemicals, heat, or both can usually break these emulsions. A great variety of chemicals are used for this purpose, but no one material has proved effective for all emulsions.
A system can be plugged as the result of an inhibitor loosening scale and suspending it in the fluid. This problem is best avoided by planing. The best preventive measure is to clean the system thoroughly, if possible, before inhibitor is applied. An alternative or supplementary method in systems that are very sensitive to suspended solids is to protect the sensitive parts with temporary filters.
As with most industrial chemicals handled in large volume on a regular basis, oil field corrosion inhibitors should be treated with respect from a safety standpoint. Although these products are not generally highly toxic (many acid corrosion inhibitor formulations are toxic), they can produce reactions because of the amines and aromatic solvents present. Reactions usually consist of skin burns from contact and dizziness from inhalation of the vapors. Repeated contact with amines will cause the development of sensitization to these products in some individuals. To avoid these problems, any contacted body areas should be washed, and contaminated clothing should be changed as soon as possible. Any accumulation of vapors should be eliminated from confined spaces.
Another possible adverse effect of inhibition is an increased rate of corrosion of a metal in the system other than the one for which the inhibitor was selected to protect. For example, some amines protect steel admirably, but will severely attack copper and brass. Nitrites may attack lead and lead alloys, such as solder. In some cases, the inhibitor may react in the system to produce a harmful product. An illustration of this is the reduction of nitrate inhibitors to form ammonia, which causes SCC of copper and brass. The only way to avoid these problems is to know the metallic components of a system and to be thoroughly familiar with the properties of the inhibitor to be used.
Application of Inhibitors
Choosing the proper inhibitor for treating a corrosion problem in the oil field is important; however, it is equally important to select the correct treating method. The best inhibitor available will not successfully control corrosion if it does not reach the trouble area. To be effective and economical, a corrosion inhibitor:
· Must be present at an initial concentration sufficient to promote complete coverage of all steel surfaces.
· Must be replenished as necessary to repair washed-away portions of the protective inhibitor film.
Batch treatments (Fig. 20) are commonly used in producing wells and, in some cases, in gas lines and crude flow lines. Inhibitor can be batched down the tubing-casing annulus, through the tubing, or between pigs (in the case of a pipeline). The various types of batch treatment are:
• Standard batch
• Extended batch
• Annular slug
• Tubing displacement
• Between pigs batch
Standard Batch. This method is used for producing wells that are not equipped with packers. The inhibitor is put into the annulus, and the well is placed on circulation to distribute the inhibitor throughout the system (Fig. 20a). Normally, the longer the well is circulated, the better the inhibitor film. The application of this treatment in low fluid level wells depends on the fluid level maintained in the annulus. The method would not be recommended in wells that pump off. It would be estimated that a fluid level of at least 46 m (150 ft) should be maintained. In placing the treatment in operation in these wells, it would be recommended that the initial treatment be immediately displaced into the tubing and that a second batch of inhibitor be placed in the annulus.
Extended Batch. This method is a variation of the standard batch treatment, but in this case, the inhibitor is left in the annulus (Fig. 20 b). As the annular fluid level fluctuates, small amounts of inhibitor are carried in the oil into the tubing, thus giving the well periodic treatments weeks or months after the actual treatment. This type of treatment has lasted up to 6 months in some wells of Oklahoma
Annular Slug. There is one technique for batch treating pumping wells that allows the well to continue full production while being treated. A water-dispersible or water-soluble inhibitor is mixed with water and placed in the annulus (Fig. 20c). This mixture will fall through the oil phase in the annulus. This technique will work if there is little or no water level in the annulus, but probably will not work if there is a substantial water level in the annulus. Frequency of treatment ranges from twice weekly to monthly.
Tubing Displacement. Wells that are set on packers or gas-lift wells are frequently treated by tubing displacement. The inhibitor is either dispersed or put in solution in water or hydrocarbon. The water may be fresh or produced. The hydrocarbon may be produced, or it may be a refined product, such as kerosene or diesel. The inhibitor is usually used at about 10% concentration in the water or hydrocarbon. The desired amount of this mixture is then introduced into the tubing (Fig. 20d). If the well is a dry gas well, the mixture will fall to the bottom if sufficient shut-in time is given (from several hours to overnight, depending on the depth of the well). If the tubing contains liquids, the mixture must be displaced to the bottom of the well by pumping liquid (usually produced fluids) in behind the mixture. The amount of displacing liquid is calculated by determining the volume of the tubing and subtracting the volume of inhibitor mixture. After the inhibitor has been displaced to the bottom, the well is usually shut in for 2 to 24 h. The well is then put into normal operation in the usual manner.
The tubing displacement technique is also known as a kiss squeeze. This type of treatment will last from a week to several months, depending on the system and the inhibitor, and is normally used on flowing oil wells.
Between Pigs Batch. This method is used to control corrosion in gas pipelines and is only used by itself in moderately corrosive systems.
The volume of inhibitor mixture needed to give a 3-mu-thick coating can be calculated from an equation that takes into account the pipe diameter and length (Ref 57, 58).
Continuous treatment is used on producing wells, injection wells, pipelines, and flow lines. Continuous treatment simply involves introducing inhibitor on a continuous basis so that its concentration in the corrosive fluids is maintained at a level sufficient to prevent or reduce corrosion. This concentration may vary from a few parts per million to 50 ppm or more, depending on the severity of attack. There are many ways to continuously treat producing wells:
• Introduce inhibitor into the line that bypasses part of production into the annulus
• Inject inhibitor into the power oil of a subsurface hydraulic pump
• Inject inhibitor into a small string of tubing that is run downhole (small-bore treating string)
Squeeze Treatment. This is a combination batch-continuous method in which the inhibitor solution is placed into the formation. The inhibitor and diluent are displaced down the tubing and into formation by 25 to 75 drums of displacing fluid, which is usually clean crude, diesel fuel, or nitrogen. When the well is returned to production after a squeeze, the initial concentration of chemicals in the returned fluid is high and decreases very rapidly. The inhibitor continuously returns from the formation to repair any breaks in the inhibitor film. The second squeeze and successive treatments all give a longer treatment life than the first squeeze. Possibly, a portion of the chemical used in the first squeeze is trapped in the formation and cannot return to the well bore. This action is shown in Fig. 21.
The advantages of squeeze treatment include:
• Can be used in tubingless or multiple completion wells
• Treating frequency is reduced and ranges from 6 to 18 months, depending on the inhibitor, the formation, placement technique, and the fluids being produced
The disadvantages of squeeze treatment are as follows:
• High cost
• Possible clay swelling
• Emulsion blocks that restrict production
• Injection pressure must be kept below the pressure necessary to fracture the formation
This method is used on gas-lift wells having a high-pressure, a high gas-oil ratio, and a high rate of water production. More information on the use of inhibitors in the oil patch is available in the section “Use of Inhibitors” of the article “Corrosion Protection Methods for the Petrochemical Industry” in this Volume.
Nonmetallic Materials
In recent years, there has been an increased use of nonmetallic materials in oil field operations. These materials are being used because they do not corrode in the environments in which steel readily corrodes. They are also lightweight, suitable for rapid installation, and, in most cases, less expensive than steel. In a 1982 American Gas Association survey of 56 gas utility companies, it was found that nonmetallic pipe systems failed at only 13.2% the rate of metallic pipe systems when excavation damage is excluded (Ref 59). Nonmetallic pipe can be classified into three major categories: thermoplastic materials, fiber-reinforced materials, and cement-asbestos.
Thermoplastic materials can be repeatedly heated, softened, and reshaped without destruction. The most commonly used thermoplastic pipe materials are (Ref 47):
• Polyvinyl chloride (PVC)
• Chlorinated polyvinyl chloride (CPVC)
• Polyethylene (PE) Polyacetal (PA)
• Acrylonitrile-butadiene-styrene (ABS)
• Cellulose acetate butyrate (CAB)
Glass fiber-reinforced thermoset materials are chemically set and cannot be softened or reshaped by the application of heat. There are two major classes of these materials in oil field use:
• Fiberglass-reinforced epoxy (FRE)
• Fiberglass-reinforced polyester (FRP)
Cement-asbestos is the oldest nonmetallic material in use in the oil field. It is a combination of Portland cement, asbestos fibers, and silica. It can be obtained with an epoxy lining, but most of this pipe currently in use is unlined (Ref 60).
Joining Methods
The methods used to join various types of nonmetallic pipe are shown in Table3. The heat welding method uses a heating element to soften the ends of the joints, which are then pushed together and held until the joint cools. About 25% of all thermoplastic pipe joints are made by this method (Ref 47). Solvent welding can be used on some of the thermoplastic pipe materials and on both of the thermosetting materials. This method uses both a solvent and a glue to hold the joints together. Finally, threads can be used on all nonmetallic pipe materials.
Advantages and Disadvantages
The advantages of nonmetallic materials include the following (Ref47):
• They are generally immune to corrosion in aqueous systems
• They are lightweight and are therefore easier to handle
• Nonmetallic pipe is quickly joined and installed
• No external protection, such as coatings or cathodic protection, is required
• The smooth internal surface of nonmetallic pipe results in lower fluid friction loss
Among the disadvantages of nonmetallic materials are (Ref 47):
• Nonmetallic pipe has a more limited working temperature and pressure. These limits are also more difficult to predict with assurance than the limits of steel pipe
• Careful handling is required in loading, unloading, and installation
• Nonmetallic pipe should be buried to protect it from sunlight, mechanical damage, freezing, and fire.
• Nonmetallic pipe has very low resistance to vibration and pressure surges
Typical Applications
Thermoplastics have seen use in flow lines, gathering lines, saltwater disposal lines, liners for steel pipe in high-pressure operations, and fuel lines for gas engines. Polyvinyl chloride has a maximum temperature limit of 65 0C (150 0F) and a maximum operating hoop stress of 27.5 MPa (4000 psi). Polyethylene has a maximum operating temperature of4O 0C (100 0F) and a maximum operating hoop stress of 4.3 MPa (625 psi) (Ref 43).
Glass-fiber-reinforced thermoset materials have also seen use in flow lines, gathering lines, saltwater disposal lines, liners for steel pipe in high-pressure operations, and fuel lines for gas engines. They have also been used for tubing in disposal and injection wells. Neither FRE nor FRP should be used for a well production flow line or gas gathering system at pressures above 2.1 MPa (300psi) and temperatures of 65 “C (150 “F). These materials should not be used in vacuum systems or where repetitive vacuum surges are likely to occur and in lines handling sand-laden fluid.
In addition, FRP has been used for stock tanks and barrels ranging in size from small chemical tanks of 1890 L (500 gallons) or less to 500 barrels or larger. Even sucker rods have had their bodies made of FRP.
Cement-asbestos materials have been used in low-pressure saltwater disposal lines. They have a maximum temperature rating of95 “C (200 oF).
Environmental Control
Oxygen dissolved in oil field water is one of the primary causes of corrosion. Dissolved oxygen is needed at 25 oC (75 oF) for an appreciable corrosion rate in neutral waters, while even in high-salinity brines at 150 “C (300”'F), the corrosion rate is low once the oxygen is removed (Ref 45). This type of corrosion is usually a localized form of attack, such as pitting, rather than a uniform attack. Oxygen also causes the growth of aerobic bacteria, algae, and slime, which can create plugging and enhance pitting.
Also, mixing an oxygen-containing water with oil field waters containing dissolved iron or hydrogen sulfide can cause precipitation of iron oxides, iron hydroxides, or free sulfur, thus causing serious plugging problems. In one case, some injection wells of a waterflood in west Texas
Both mechanical and chemical means have been used to remove dissolved oxygen from oil field waters. The mechanical means are counter current gas stripping and vacuum deaeration, while the chemical means include sodium sulfite, ammonium bisulfite, and sulfur dioxide. The choice of oxygen removal method depends on economics. Usually, the mechanical means are used when large quantities of oxygen are to be removed. Chemical removal is usually employed to remove small quantities of oxygen and even sometimes for the removal of residual oxygen after the mechanical means have been used.
Mechanical Methods
Gas stripping is performed in either a packed column or a perforated tray column. Perforated tray columns are preferred because they are not as easily fouled with suspended solids or bacterial slime as packed columns. Figure 22 illustrates a tray-type gas stripping column. Oxygenated water flows into the top of the column, while the stripping gas flows through the bottom inlet. As the gas bubbles up through the water, oxygen comes out of solution. The trays or packing in the column increases the contact area. These systems are designed to use not more than 0.06 m3 (2 ft3) of gas per barrel of water being stripped (Ref 30). The gas source should be free of both oxygen and H2S. Either natural gas or exhaust gas from engines is commonly used.
The principle of removal is to reduce the concentration of oxygen in the gas coming in with the water by dilution with the stripping gas.
In vacuum deaeration, a vacuum is created in a packed tower, and as the oxygenated water is passed over the packing, the low pressure causes the oxygen to bubble out of solution. The vacuum pump pulls the oxygen, water vapor, and other gases from the top of the tower. The tower usually consists of several different pressure stages, as shown in Fig. 23. In a packed column, each stage consists of a height of packing, which is sealed from the stage below by a layer of water in the bottom of the packing (Ref 46). A single-stage tower will economically remove oxygen only to a lower limit of 0.1 ppm because of the excessive vacuum pump horsepower required to achieve lower concentrations. Therefore, multistage columns are needed. Dissolved oxygen concentrations as low as 0.01 ppm have been achieved in three-stage towers (Ref 46).
Combination Vacuum Deaeration and Gas Stripping. Vacuum deaeration with the use of 0.003 m3 (0.1 ft3) of natural gas has been used to reduce the oxygen content of water from 5 to 0.05 ppm or less in a 40000 barrel per day water-flood in west Texas. Single-stage vacuum deaeration reduced the oxygen content of the water to 0.17 ppm, while the gas stripping further reduced the oxygen content to 0.05 ppm. Corrosion rates in the water were reduced from 0.36 to 0.04 mm/yr (14 to 1.6 mils/yr) (Ref 63).
Chemical Methods
Sodium sulfite is used to scavenge oxygen from water and is available as a liquid or as a powder. It reacts with oxygen according to Eq 4:
Approximately 8 ppm of Na2SO3 is required to react with 1 ppm 02. A 10% excess is usually required for complete reaction, and a catalyst such as cobalt chloride (0.1 ppm) is needed to scavenge to acceptable levels within a few minutes. Because Na2SO3 solutions will react with atmospheric oxygen, an inert gas blanket is required on the storage tank.
Aininonium bisulfite is a liquid scavenger and reacts with oxygen according to Eq 5:
An 80% solution ofNILtHSO3 requires a 10:1 ratio by weight for the reaction. A 10% excess is needed to complete the reaction. Ammonium bisulflte does not react with air and can be stored in open containers. A catalyst is not usually needed for oil field brines. Because the chemical is supplied as a solution with a pH of 4 to 4.5, it must be stored in a corrosion-resistant vessel. Type 304 stainless steel is commonly used (Ref 46).
Sulfur dioxide is a chemical scavenger that can be either supplied as a liquefied gas under pressure in a cylinder or generated by burning sulfur. The reaction between sulfur dioxide and oxygen proceeds according to Eq 6:
A quantity of 4 ppm by weight of SO2 is required to remove 1 ppm of oxygen. A 10% excess and a catalyst such as cobalt chloride are needed to complete the reaction. Sulfur dioxide from cylinders is applied by using a bypass line that handles approximately 10% of the total fluids, as shown in Fig. 24. The scavenger is added to the bypass fluids. The materials used in this bypass line should be resistant to acid attack because of the low pH formed from the reaction. Use of SO2 cylinders is most advantageous in small systems (less than 10,000 barrels per day) or where small concentrations of dissolved oxygen are encountered in larger systems (Ref 65).
Precautions
Some precautions involving oxygen scavengers should be noted (Ref 46):
• Oxygen scavengers will react with chlorine and hypochlorite (ClO-), which are added to injection water for bacterial control. Therefore, these chemicals should be added downstream of the point of scavenger injection to allow completion of the scavenger-oxygen reaction
• Any organic chemicals, such as biocides, scale inhibitors, and corrosion inhibitors, can possibly interfere with the scavenger-oxygen reaction and should be selected with care.
• Oxygen scavengers cannot normally be used in sour systems. If H2S is present, it may react with the cobalt chloride catalyst to form insoluble sulfides (Ref46)
Oxygen Exclusion
It is usually more economical to exclude oxygen from oil field equipment than to remove it after it has entered the system. The most common means of excluding oxygen is through the use of gas blankets on water supply wells and water storage tanks. Maintenance of valve stems and pump packing is also important.
All tanks handling air-free water should be blanketed with an oxygen-free gas such as natural gas or nitrogen. Most tanks require only a few ounces of pressure (Ref 62). The regulator should be sized to supply gas at a rate adequate to maintain pressure when the fluid level drops.
Oil blankets should not be used in place of gas blankets. Oxygen may be 5 to 25 times as soluble in hydrocarbons as in oil field waters (Ref 62). Oil blankets will coat precipitates in the water, which can lead to well plugging problems. Some bacteria will even thrive at the oil/water interface (Ref 46).
Even supply wells and producing wells may need to be gas blanketed to prevent oxygen entry.
If these wells are operated cyclically without gas blankets, oxygenated air will be drawn into the annulus every time the well is turned on and the fluid level drops.
Oxygen can also enter a pump on its suction side if a net positive suction head is not maintained. If the seals start to leak, air can then be sucked into the pump (Ref 62).
Monitoring
Once the well, line, or vessel is treated, it is necessary to evaluate the effectiveness of the treatment program and to determine when to retreat or change dosage levels. The methods used include iron counts, weight loss coupons, test nipples and spools, electrical resistance and linear polarization methods, and waiting for the tubing to fail. This last method is not popular with operators. However, the collection of failure data, such as tubing failures, is a valuable source of monitoring information.
Iron counts consist of taking a representative sample of produced water and testing for iron content. The sample must be representative. The iron counts can be plotted for easier understanding. Some computer programs will present iron counts on a computer plot in color. Care must be taken so that the iron from the formation is not assumed to be metal loss from the tubing. Some of the deep, hot Tuscaloosa Trend wells produce water with over 100 ppm of iron. A base count should be conducted on a downhole sample, if possible.
Corrosion coupons may be flat or cylindrical and may be installed in any accessible location. It must be remembered that coupons measure corrosion only where they are placed. Coupons show corrosion that has already taken place, and a single coupon will not show whether the corrosion was uniform or occurred all at once.
Different types of coupon holders or chucks are used, depending on the system, the pressure, the location, or other factors. Most coupons are run in a 25- or 50-mm (1or2 inch.) threaded plug. Flat coupon holders hold two coupons, while cylindrical coupon chucks may contain eight or more. The multichuck coupons allow a coupon to be pulled at intervals to see if the corrosion rate is uniform or not.
High-pressure systems require a special coupon check and insertion device. The insertion tool fits into a special attachment on the pipe or vessel that has a high-pressure chamber with a valve on each end. The inner valve is closed, the retrieval tool inserted, and the inner valve opened. The tool is then run in and left. The procedure is reversed to remove the coupon.
The industry guide for preparing, installing, and interpreting coupons is NACE standard RP07-75 (Ref 66). The primary consideration is that all coupons be treated exactly alike. A method of preparation that does not alter the metallurgy of the coupon is required. Grinding and sanding of coupons should be controlled to avoid metallurgical changes and to provide a consistent and reproducible surface finish.
Coupons should be handled carefully and stored in non-corrosive envelopes until they are installed. Rust spots caused by improper handling, fingerprints, and so on, may initiate a pit that is not representative of the system being evaluated. Prior to installation, the weight, serial number, date installed, name of system, location of coupon, and orientation of the coupon and holder should be recorded. The coupons are left in the system for a predetermined number of days and then removed.
When the coupons are removed, the serial number, date removed, observations of any erosion or mechanical damage, and appearance should be recorded. A photograph of the coupon may be valuable in some cases. The coupons should then be placed in a moisture-proof envelope impregnated with a vapor phase inhibitor and taken immediately to the laboratory for cleaning and weighing. The coupons can be blotted (not wiped) dry prior to placing in the envelope.
The laboratory receives the coupon and inspects, cleans, and weighs it. A report is issued showing the thickness loss, any pitting observations, and any other observations of interest.
Electrical resistance probes function by reading the resistance to current flow of a thin loop of metal installed in the system. The loop of metal is part of an electrical bridge circuit. As the loop corrodes and loses cross-sectional area, electrical resistance increases, and the current flow decreases. This imbalances the bridge and reads out directly on a meter. It may also be recorded on a strip recorder. As the reading changes, the points are plotted on a graph. The slope of the line is translated into a corrosion rate. The slope will change when a well is treated or when any event occurs that changes corrosion rate.
The metal loop is fragile, and it can be broken if foreign objects, chunks of scale, and similar obstructions are present in the how. The loop may become coated with paraffin and will not give a true reading. Pitting rates cannot be determined with a resistance probe.
Linear polarization instruments apply a voltage to a pair of electrodes, compare it to a reference voltage, and then read and record the current flow. The voltage is impressed in a forward direction until breakdown occurs, that is, until a small increase in voltage causes a large increase in current. The voltage is increased some distance past breakdown, and the polarity is reversed. The positive electrode then becomes the negative electrode.
The voltage is then increased in the opposite direction past the original voltage starting point. These voltages are recorded and plotted continuously on graph paper. The position and shape of the curves will show the corrosion rate, maximum corrosion rate, and the influence of oxygen.
The principal advantage of this instrument is that a corrosion rate can be determined immediately, and a pitting rate can be measured without waiting a month or so to pull a coupon or calculate a slope with a resistance probe. Probe locations are very important, because the probe must be immersed in water to give an accurate reading. Probes can become coated with paraffin and will show an erroneous corrosion rate. They should be located in the bottom of a line or in a bypass loop so that they are continuously immersed in water. The probe may also short out if a piece of metal is across the probe or if corrosion by-products and other deposits collect on the probe.
A test loop can be installed in a system for better control. A test mop is simply a bypass with valves for controlling flow, and it may contain weight loss coupons as well as probes. The system can be monitored, and different inhibitors can be evaluated at the same time.
A caliper survey can be conducted to determine if pitting and general metal loss have been halted. The caliper log can be easily compared with a log run before the treatment is begun.
Other monitoring methods include hydrogen probes, galvanic probes, and electromagnetic logging devices. Chemical analysis of produced water for alloying metals, such as manganese and chromium, can be conducted.
Collection of Field Data. It is important to collect and chart failure rates. Some failures may occur over a period of time and may erroneously indicate that the treatment is not effective. However, through proper charting and comparing with previous failure rates, the effectiveness of proper treating will be shown.
Post a Comment