Carbon Dioxide Corrosion In Oil And Gas Production 3w5b

<5 1a ⇔ RDS(A)⇒ ⇒ ⇒ 60/1.5=40 –1 1

pH> 5 1b RDS⇒ ⇒ ⇒ ⇒ 60/0.5= 120 0 1

RDS = rate-determining step.

being reduced directly. The possible RDS in cathodic reactions, therefore, are summarized as follows: Schwenk

(HCO–3 ): H + + e– ⇒ H,

2H ⇒ H2

(2)

de Waard and Milliams H2CO3 + e– ⇒ H + HCO–3

RDS

HCO–3 + H + ⇔ H2CO3 and 2H ⇒ H2

(3) (4)

Ogundele and White

HCO–3 + e– ⇒ H + CO23–

RDS

HCO–3 + H + e– ⇒ H2 + CO23–

(5) (6)

While the Ogundele mechanism only deals with alkaline pH conditions and the Schwenk and deWaard mechanisms are only a possible hypothesis, recent reactions put forward by Crolet, et al.,19 are the most likely mechanism encountered. Based on the reaction proposed by Crolet, the anodic dissolution of iron in CO2-containing media is summarized in Table 1 with respective RDS at different pH conditions.19-20 This table goes beyond summarizing the anodic dissolution reactions and includes the respective Tafel slopes derived from each pH condition. Whichever is strictly mechanistically correct, it is evident that the concentrations of dissolved CO2 species in solution and the mass transport of dissolved CO2 to the steel surface have a critical influence on the reaction and subsequent corrosion rate13 and that every dissolved species present in the media can contribute to the cathodic reaction (Ref. 9 and Pots21 and Nes˘ic´, et al.22). Hence, there is a clear need to be able to characterize solution chemistry with respect to CO2 dissolution where the resulting acidification will depend also on water composition and its buffering capacity.14

CORROSION—Vol. 59, No. 8

Types of CO2 Corrosion Damage CO2 corrosion occurs primarily in the form of general corrosion and three variants of localized corrosion (pitting, mesa attack, and flow-induced localized corrosion).9 In studying CO2 corrosion, a clear distinction should be made between pure CO2 corrosion and a combined interaction of erosion/CO2 corrosion, which may be exacerbated by CO2 corrosion. The latter characterizes itself in the form of ripple marks, horseshoes, comet tails, and dinosaur foot prints,8,18 whereas the former is described in this paper. Pitting — Pitting occurs at low velocities and around the dew-point temperatures in gas-producing wells. In the field, only occasional pits have been observed, which were either accidental adjacent to nonmetallic inclusions or incipient mesa attack.7-8,18 The pitting susceptibility increases with temperature and CO2 partial pressure. In several related works on pitting, Schmitt and coworkers15-17 discussed the effects of temperature, Cl– concentration, nature of anions and cations, as well as corrosion inhibitors on the pit initiation during the first stages of CO2 corrosion of pure iron and low-alloy steels at 5 bar CO2. They also reported that literally all alloys of technical interest might undergo pitting corrosion in CO2 environments at the right conditions. Finally, they showed that Pb additions inhibited localized corrosion (including pitting) through deposition at local anodes. On the other hand, Videm and coworkers23-24 concluded that the pitting of carbon steel in CO2-containing environments was almost independent of the chloride content. Therefore, it was different from many other metal environment combinations. However, the “harmful ion,” if any, responsible for the pitting of carbon steels in CO2 solutions was not identified. The discussion on pitting of carbon steels in sweet environments has not reached a convincing conclusion. Various authors attribute the pitting initiation and its propagation to different factors and there is no generally applicable rule for its prediction. Mesa-Type Attack — Mesa attack is a type of localized corrosion and occurs in low to medium flow

661

CRITICAL REVIEW OF CORROSION SCIENCE AND ENGINEERING

FIGURE 1. A typical example of severe mesa attack experienced on internal surfaces of pipelines transporting CO 2 -containing hydrocarbon fluids (courtesy of Institute for Energy Technology, IFE).

conditions where the protective iron carbonate film forms but it is unstable to withstand the operating regime. It manifests itself in large flat-bottom steps with sharp edges. Corrosion damage in these locations is well in excess of the surrounding areas.9 Figure 1 is an example of mesa attack experienced on an internal surface of pipelines exposed to CO2containing environments. Crolet, et al.,25 proposed that the microstructurally formed galvanic coupling between steel (its ferrite phase) and the cementite (Fe3C) layer was one possible cause to promote mesa attack in sweet environments. According to the same author,18 mesa attack was observed in mature oil wells or, conversely, in young gas wells under high pressure of acid gases. Even in the presence of high fluid flow rates, its characteristics were totally different from the typical features of erosion corrosion. Mesa attack appeared to be a little sensitive to the velocity of water in the pipe, but extremely dependent on fluid composition.18 Ikeda, et al.,26 attributed the initiation of mesa attack to the competitive film formation reactions between ferrous carbonate (FeCO3) and magnetite (Fe3O4). Although, in actual field conditions, Fe3O4 has not been detected.18,27 The codeposition of both compounds could initiate the mesa corrosion by disturbing the protective film formation. They concluded that the initiation mechanism of mesa corrosion was closely related with the formation of a poorly protective FeCO3 film or the localized destruction of a protective film. Videm and coworkers28-30 showed that flowinduced mesa attack could occur in water saturated with FeCO3 under turbulent flow conditions where film formation is prevented locally. In similar work, Dugstad and coworkers30-31 demonstrated that the

662

initiation of mesa attack was a result of the marginal film stability of FeCO3. However, as indicated earlier, chemical instability of the film has a more pronounced influence on the formation of mesa attack than any mechanically driven effect by fluid dynamics.27 Dugstad then discussed the relation between Fe2+ content of the environment and the initiation of mesa attack for sweet environments.31 According to him, when mesa attack has initiated, a galvanic cell would probably establish where the film-covered surface was cathodic and the mesa-attacked areas were anodic.31 As described later, trace concentrations of Cr alloying element when added to carbon steel reduced mesa attack to a great extent.32-33 Again, the exact conditions under which mesa attack forms are poorly understood and further systematic studies are necessary to prevent its occurrence in the field. Flow-Induced Localized Corrosion — This form of corrosion starts from pits and/or sites of mesa attack above critical flow intensities. It then propagates by local turbulence created by the pits or steps at the mesa attack or by protruding geometry. The local turbulence combined with the stresses produced during scale growth may destroy existing scales.15-17 Once the scale is damaged or destroyed, the flow conditions then may prevent reformation of protective scale on the exposed metal.9,16-17 Flow-induced localized corrosion attack is principally observed in laboratory experiments in the absence of complete control of fluid chemistry.

KEY FACTORS INFLUENCING CO2 CORROSION CO2 corrosion is influenced by a number of parameters, including environmental, physical, and metallurgical variables as illustrated in Figure 2. The majority of these have been covered extensively by a number of authors and captured elsewhere.6,8-9 Notable parameters affecting CO2 corrosion include: —fluid makeup as affected by water chemistry, pH, water wetting, hydrocarbon characteristics, and phase ratios9,34-36 —CO2 and H2S content9,37-40 —temperature9,34-35,41 —steel surface, including corrosion film morphology, presence of wax, and ashphaltene9 —fluid dynamics9,29,39-41 —steel chemistry32-33,38,42-45 All parameters are interdependent and can interact in many ways to influence CO2 corrosion. In the present article, emphasis has been placed on two key issues relating to metallurgical variables and surface film as key parameters not covered systematically elsewhere. Other influential parameters are covered extensively by others8-9,34 and have been given only a brief mention here.

CORROSION—AUGUST 2003


FIGURE 2. CO2 corrosion, the influential parameters.

Environmental Parameters Environmental factors that affect the inherent corrosivity of the aqueous phase therefore will affect CO2 corrosion. These include solution chemistry, CO2 partial pressure, temperature, the in-situ pH, H2S, and the effect of organic acids. Solution Chemistry and Supersaturation — While there are still limited debates about the mechanism of CO2 corrosion in of which dissolved species are involved in the corrosion reaction, it is evident that the resulting cathodic corrosion rate is dependent on the partial pressure of CO2 gas and temperature. Partial pressure of CO2 gas thus will determine solution pH and the concentration of dissolved species for a given temperature. Supersaturation plays a vital role in the formation and stability of a protective corrosion layer. Supersaturation is defined as “log [A+][B–]/Ksp” for an insoluble salt (AB) having an equilibrium reaction of AB = A+ + B–, where A+ and B– are ionic species and Ksp is the solubility product. In any sweet environment, a particularly insoluble salt can play an important role in reducing the corrosion rate. High supersaturation of A+ and B– leads to precipitation of a corrosion layer/film that therefore would reduce the corrosion rate through several kinds of effects, including:18 —provision of a “diffusion barrier” (extended diffusion length between the metal substrate and the corrosive medium);


—formation of a low-porosity protective layer (lowering the exposed surfaces compared with the steel surface and hence less areas to be corroded); —creation of concentration gradients of the principal chemical species (Fe++ and HCO3–). This is potentially the most influential and particularly difficult to model. Ingress of solution to soak the porosity leads to steep concentration gradients, which may induce a significant shift of the local pH and water chemistry from the bulk conditions, and therefore, a genuine effect of “liquid surface state.” Altogether, the precipitation rate and protective characteristics of any scale depend heavily on the supersaturation of the bulk solution. Hence, any variations in this level of supersaturation could affect the severity of the corrosion. For iron carbonate systems, this can be portrayed as a reaction similar to “Fe(HCO3)2 ⇔ FeCO3 + H2CO3.”46 The interdependency of supersaturation with the respective rates of iron dissolution and reprecipitation is shown in Figure 3 as a function of temperature. Figure 3 illustrates that while iron carbonate solubility to achieve saturation remains a little dependent on temperature, the limit of supersaturation is achieved at lower Fe2+ concentrations with increasing temperature and hence facilitates the formation of FeCO3.31 CO2 Partial Pressure — CO2 partial pressure has been used in pH calculations and corrosion rate

663


FIGURE 3. Fe++ solubility in pure water as a function of temperature at 0.1 MPa CO2 partial pressure. Open squares represent results obtained with corroding steel present, where the water volume to steel surface area ratio is 4 cm 3/cm 2. Filled triangles are Fe++ concentrations calculated from pH. Filled squares are results obtained with solid iron carbonate. The dashed line is iron carbonate solubility calculated from IUPAC data.31

measurements by many authors.6,9,11,36,37,47-49 In most cases, a relationship has been developed between the CO2 partial pressure and corrosion rate. However, with some exception,36,48 the majority of pH calculations in these relationships do not take on board that produced water cannot stay supersaturated in CaCO3. It also should be noted that for gas wells with increasing pressure, the non-ideality of the natural gas will play an increasing role. Instead of the CO2 partial pressure, the CO2 fugacity (fco2) should be used.50 The difficulty is that for the purpose of homogeneity between nonideal phases, when fugacity is used for the gas phase, an activity coefficient (γco2) should also be introduced for the water phase. Unfortunately, fco2 is often available from “PVT” data, but rarely γco2,36,48 and this may be the reason for the simplistic use of fco2 in the prediction of corrosion rates. Operating Temperature — The operating temperature strongly affects the nature, characteristics, and morphology of surface film, which, in turn, influences the CO2 corrosion process. At temperatures in excess of ca. 80°C, the solubility of FeCO3 in the solution is decreased and high supersaturation leads to FeCO3 precipitation.41 At low temperature ranges (< ca. 70°C), corrosion rate progressively increases with temperatures up to an intermediate temperature range (between 70°C to 90°C), after which the corrosion rate then diminishes. However, at sites where breakdown in the formation of FeCO3 occurs,

664

corrosion process proceeds unhindered, which may lead to severe localized attack. This is an important design consideration.9 It is thought that the increase of corrosion rate in the low-temperature range is due to an increase of mass-transfer rate as a result of flow effect and slow FeCO3 formation rate.27 Consequently, after the formation of the protective scale, a diffusion process may become the RDS in the corrosion process.19 However, as described in the Solution Chemistry and Supersaturation Section, this does not rule out the potential for the occurrence of localized corrosion or the formation of nonprotective corrosion scales (i.e., when the aspects of solid surface states [diffusion barrier and porous layers] are superseded by those of liquid surface states [membrane effect and local water chemistry]27), as described in the section on surface film. In Situ pH — Solution pH plays an important role in the corrosion of carbon steels by influencing both the electrochemical reactions that lead to iron dissolution and the precipitation of protective scales that governs the various transport phenomena associated with the former. Under certain conditions, solution constituents of the aqueous phase will buffer the pH, which can lead to precipitation of a corrosion scale and the possible lowering of corrosion rates. It should be noted that, as described in the section on film formation, in certain circumstances the corrosion layer can even be corrosive and increase the severity of attack.8-9,19,27 As an example, by increasing pH from 4 to 5, the solubility of Fe++ is reduced five times; for an increase from pH 5 to 6, this reduction is around 100 times.8 A low solubility can correspond to higher supersaturation, which therefore accelerates the precipitation process of FeCO3 film. For pH values >5, the probability of film formation is thus increased and that can contribute to the lower corrosion rates observed. It must be noted that the solubility of the FeCO3 should not be confused with that of the iron ion.8-9,18 Effect of H2S — Ignoring the cracking aspects of corrosion problems associated with sour service, low levels of H2S can affect CO2 corrosion in different ways. H2S can either increase CO2 corrosion by acting as a promoter of anodic dissolution through sulfide adsorption and affecting the pH or it can decrease sweet corrosion by forming a protective sulfide scale. The exact interaction of H2S on the anodic dissolution reactions (Table 1) is not clear. For similar conditions, oil and gas installations could experience lower corrosion rates in sour conditions compared to completely sweet systems. This is attributable to the fact that the acid created by the dissolution of H2S is about three times weaker than that of carbonic acids, but H2S gas is about three times more soluble than CO2 gas. As a result, the effect of both CO2 and H2S gases on lowering the solution pH and potentially increasing corrosion rate



are fundamentally the same. In addition, H2S might play a significant role on the type and properties of the corrosion films, improving or undermining them.8-9 Videm, et al.,38 and Mishra, et al.,37 have reported two opposing results concerning H2S. While the former has reported that very small amounts of H2S in CO2-containing water augmented the corrosion rate, the latter has argued that small amounts of H2S had some inhibitive effect on CO2 corrosion of steels. They attributed this to the formation of an iron sulfide film that apparently was more protective than FeCO3. Many papers have been published on the interaction of H2S with low-carbon steels.6,8-9,39-40,51-52 However, literature data on the interaction of H2S and CO2 is still limited since the nature of the interaction with carbon steel is complex. The majority of open literature does indicate that CO2 corrosion rate is reduced in the presence of H2S at ambient temperatures. Nevertheless, it must be emphasized that H2S also might form a nonprotective layer and that it might catalyze the anodic dissolution of bare steel. Steels may experience some form of localized corrosion in the presence of H2S, although very little information is available. Published laboratory work has not been conclusive, indicating that there is a need to carry out further studies to clarify the mechanism. A recent failure showed how the corrosion rate in the presence of a high concentration of H2S might be higher than predicted using CO2 corrosion prediction models. However, in spite of the work on H2S corrosion of steels, no equations or models are available to predict corrosion, as is the case for CO2 corrosion of steels.9 Effect of Acetic Acid — Organic acids present in production fluids has long been considered to significantly influence and complement CO2 corrosion. The influence has been shown to occur systematically in all field conditions where CO2 corrosion was observed.18,27,53 Addition of acetic acid (HAc) to the test environment reduces the protectiveness of the films and increases the sensitivity to mesa attack. This attributes to a lower Fe2+ supersaturation in the corrosion film and at the steel surface. Significant reduction in film stability was observed when the concentration of undissociated HAc in the solution was increased from 0.05 mmol to 0.2 mmol, but the results are too few to give more accurate threshold values.19,53-56 Crolet and Bonis14,55 make the point that CO2induced acidification also can cause partial re-association of anions, such as acetates and propionates, to form organic acids. Such weak acids then will increase the oxidizing power of H+ by raising the limiting diffusion current for cathodic reduction (cf. Reaction [2]). The presence of such acids also will tend to solubilize the dissolving iron ions and sup-


press FeCO3, or oxide film formation, which can otherwise ivate the steel surface.25,35,55 It is also often observed, at least in laboratory tests, that water or brine acidified with CO2 to a given pH produces a more corrosive solution than acidifying to the same pH with mineral acid. This is generally attributed to the fact that because carbonic acid (H2CO3) is not fully dissociated in solution, it provides a reservoir of H+ ions over and above that determined by the solution pH (–log[H+]). In essence, this is the same effect as that once cited by Crolet and Bonis14 in the presence of organic acids (weak acids) (i.e., increasing the oxidizing power of H+). For low concentrations of HAc (very few mM), however, the effect cannot but remain negligible with respect to the tens or hundreds of mM of dissolved CO2.19 In addition, the presence of HAc may change the mechanism of the anodic dissolution of iron through competitive adsorption of acetate ions, CH3COO– (or Ac–) and HCO3–, although this was shown to have only a slight inhibiting effect.19,55 Generally, the presence of HAc caused a significant increase in the corrosion rates in CO2 environments.53,56 HAc (along with other organic acids) could jeopardize the protective corrosion product scales formed in top-of-the-line corrosion.57 At low CO2 partial pressure, CO2 corrosion disappears, but in certain fields, it can be replaced by a genuine “HAc corrosion.” It has been shown that this was not caused by any influence of the HAc, either on the cathodic reaction of H+ or on the anodic dissolution of iron, but rather by its effect on the protectiveness of the corrosion layer. In the presence of traces of free HAc, the majority of corrosion layers on bare metal was no longer FeCO3, but iron acetate, which had a much greater solubility.19 In a similar work,25 it has been reported that at a given pH, any replacement of a concentration or a flux of bicarbonate by an equivalent quantity of acetate would considerably increase the local solubility of iron. This decreases the protectiveness of the corrosion layer in proportion, by increasing iron concentration gradients, and therefore allowing and subsequently raising the fluxes of corrosion products, which potentially can be removed through the layer. An overview of the concentration gradient of acetate ions close to the metal surface is shown in Figure 4.23 The presence of HAc is a key issue in CO2 corrosion, which requires further extensive studies.

Physical Parameters Along with environmental and metallurgical parameters, physical parameters play an important role in CO2 corrosion of carbon and low-alloy steels by influencing hydrodynamics of the system and the interface between the environment and the steel substrate. These include water wetting, wax effect,

665


FIGURE 4. Illustration of possible HAc enrichment in a corrosion layer as a result of internal acidification and galvanic coupling between steel and cementite.25

surface films, crude oil, and fluid dynamics and are mostly covered by others.8-9 Their interactive and complementary influences affect the onset of film formation and removal. These effects have been highlighted briefly here. Water Wetting — CO2 corrosion occurs when water is present in the system and it wets the steel surface. The intensity of CO2 corrosion attack increases with the time during which the water phase is in with the steel surface. Therefore, the water content (water cut) and the notion of water wetting are important variables. There are at least three different notions of water wetting as follows:58 —Hydrodynamic concept focuses on modeling a continuous water phase at the fluid/wall interface, which is primarily over the corrosion layer. However, it is evident that corrosion does not occur over the corrosion layer, but beneath it. This concept therefore cannot be directly relevant to corrosion modeling as influenced by water wetting. —Electrochemistry and surface physics concept relates to liquid in direct with the metallic phase. This can be, in part, highly influential in the modeling of water wetting. —CO2 corrosion-related concept in which liquidsoaked porous film continues to hold water, even if the bulk phase in with the wall is temporarily either pure oil (in oil lines) or just a thin, wet film (in a pure gas line without any ongoing water condensation). This provides a favorable boosting for the sealing of cementite or hydrated mill scales during the corrosion process and, as a consequence, facilitates the onset of protectiveness in an originally nonprotective film.27 Such circumstances can occur during shutdown periods or in slug flow conditions.

666

Therefore, the influence of water cut on corrosion rate should be considered in association with the flow velocity and the flow regime effects in the context of the above notions of water wetting, particularly using a combination of the second and third notions.27 It is known that emulsions can form in oil/water systems. If a water-in-oil emulsion is formed and the water is held in the emulsion, then the water wetting of steel is prevented or greatly reduced, causing the corrosion rate to decrease. If, on the contrary, an oil-in-water emulsion is formed, then water wetting of steel will happen. The transition from a water-inoil emulsion to oil-in-water occurs at around 30% to 40% water in many oil lines and, in a straight pipe with emulsified liquids, an obvious increase in the corrosion rate can be observed.59 So, as a rule of thumb, for water cuts <30 wt%, the corrosion rate is often significantly reduced,9,59 although there are many exceptions to this.60 The water cut threshold depends primarily on the maturity of hydrocarbon (immature oils still contain natural surfactants) and the nature of its kerogen (I, II, III)27 as explained in a later section. Corrosion Film Characteristics — Characteristics of corrosion product significantly affect the CO2 corrosion process. Formation of surface film can provide subsequent protection, enhanced corrosion, or uncontrolled reaction, all subject to the nature, morphology, and growth habit of the corrosion product. Due to the importance of this subject, it is described in a separate section on film formation. Effect of Wax — The presence of wax in oil pipelines can influence CO2 corrosion damage in two ways, either exacerbating the damage or retarding the process. These are dependent on the nature of wax layer and subject to flow dynamics, temperature, and other physical parameters. Based on data gathered in sweet oil lines in the U.S., a layer of wax (paraffin) deposited on a carbon steel substrate caused heavy pitting in anaerobic sweet environments. The proposed corrosion mechanism is the diffusion of CO2 through the wax layer, which can provide a large cathodic area that promotes anodic dissolution of steel at discontinuities of the wax layer.9 However, generally, wax can provide a degree of protection, albeit its protection is not reliable. Effect of Crude Oil — Corrosion tests conducted on steel in brine environments in the absence of crude oil do not give an accurate representation of the behavior of the steel corrosion in a crude oil/ brine production environment. This can lead to gross errors when using the test results to estimate potential corrosion problems.61-63 Although there has not been any specific investigation on the effect of crude oil type on protectiveness of FeCO3 scale, it has been determined that crude oils modify the morphology, composition, and



compaction of corrosion products for the different oil/water ratios evaluated.64 Hydrocarbon appears to destabilize the formation of a ive FeCO3 film, accelerating the “localized” corrosion.27,61-62 Competitive wetting of steel surfaces by water and then by oil has been reported to play an important role.65-66 A combination of water cut and hydrocarbon type play a key role in affecting CO2 corrosion. In this, the maturity of hydrocarbon (immature oils still contain natural surfactants) and the nature of its kerogen (I, II, III)27 are main issues. While this information is extremely important, it is rarely recorded or reported. It should be noted that some crude oils also release natural corrosion inhibitors, which aid in lowering the corrosion rate.27 Crude oil plays an important and differing role on CO2 corrosion and has not received due attention. Flow and Erosion — The effect of flow on CO2 corrosion still remains a contentious area in the predictive processes. This effect should be discarded in predictive modeling unless it is fully defined and controlled by the production process throughout the production life. However, it is not only fluid content that needs to be determined—the flow regime is of equal importance, which, for multi- or mixed phase fluids, will determine whether water wetting of the steel surface will occur. Clearly, if continuous hydrocarbon wetting occurs, then the corrosion risk will be extremely low.59 Key factors here are oil/water ratio and emulsion tendency/stability. For many crude oils, the presence of a water cut > ≈30% will lead to water becoming the continuous phase for a fully mixed oil/water system, such that corrosion then becomes a continuous potential risk.59,62 Similarly, if the gas/oil ratio (GOR) is >5,000, then continuous water wetting by the condensed water can be expected.8-9 The flow parameter currently favored for determining the effect of velocity on corrosion rate and scale and inhibitor film formation/stability is liquid shear stress at the pipe wall. Although there is limited reported data on upper limits regarding shear stress, a figure of 100 Pa for C-steel above which disruption to surface films becomes a concern is considered by some as appropriate.67 However, it must be recognized that for specific situations it may be necessary to conduct laboratory tests under simulated flow conditions. Laboratory testing becomes particularly critical where erosion, as a result of the presence of particulates, is a concern. There are no industry guidelines that adequately cover this situation. The commonly cited API RP-14E68 recommended practice strictly refers to pure gas-in-liquid-induced erosion (i.e., no particulates present) and applies the basic formula: Ve = C /

(

ρm


)

(7)

where Ve, mixed velocity (ft/s or m/s), ρm, mixed fluid density (lb/ft3 or kg/m3), and C are constants. The relationship is essentially empirical as is the value of the constant (C) used for a given material. This is more often than not based on the individual operator experience.

Metallurgical Parameters Chemical composition, heat treatment, and microstructural features play important roles on corrosion of carbon steels in CO2 environments. While most authors have reported the beneficial effects of chromium additions,26,29,42-45,68-79 there is not yet a consensus on the optimum amount of Cr in the steel structure. Apart from Cr, molybdenum has been found to improve the corrosion resistance of carbon steels.42 In a related work, Videm and Dugstad23 showed that small amounts of Cu, Ni, Cr (and possibly Mo) increases the corrosion potential of carbon steels, making it more noble. However, Cu additions may have a side effect on inhibitor efficiency as reported by Gulbrandsen and Nyborg.43 A laboratory study has shown that the sulfur content of carbon steels appears to influence the CO2 corrosion rate as well. Certain high-S carbon steels were more corrosion resistant than low-S carbon steels in low-shear-stirred CO2 corrosion tests,74 although the steel samples used in this work were not representatives of oil industry grades and the practical implication of the work is uncertain. Work by Kermani and coworkers44,80 has paved the way to developing an optimum metallurgy of carbon and low-alloy steels for both downhole and transportation facilities through addition of microalloying elements like V, Ti, Mo, Cu, and Cr.

ALLOYING ELEMENTS It is now well established that small quantities of chromium (0.5 wt% to 3 wt%) can offer improved corrosion resistance of low-alloy steels in CO2-containing media by promoting the formation of a stable, protective chromium oxide film.8-9,33,72,77 It also has been appreciated recently that for carbon and lowalloy steels, there may be a correlation between protectiveness of the corrosion layer in the active state and a possible “ivation” by a “super protective” layer.8,72 The development of novel carbon and lowalloy steels with superior resistance to CO2 corrosion using metallurgical conditioning recently has been made in the laboratory and subsequently by industrial casts covering a wide range of parameters including microalloying constituents, heat treatment processing, and steel production scenarios. Corrosion performance and properties of optimum steels developed in this project have been verified through evaluation of the industrial casts,80 offering superior CO2 corrosion while tolerating H2S corrosion for ap-

667


FIGURE 5. Schematic presentation of relative effect of additional microalloying elements on corrosion rate.44,80

FIGURE 6. Schematic overview illustration of different categories of steel ranked according to their Cr level.44,80

plication in Region 2 of ISO-15156.81 The philosophy underlying the work of Kermani and coworkers44,80 was based on a combination of two principles: —lowering C and adding carbide-forming alloying elements to maximize the effect of a given addition of chromium and molybdenum, by ensuring that they remain in solid solution; —achieving the desired properties by microalloying additions and mechanical and heat treatments. Steel compositions were designed with lowcarbon contents and contain microalloying additions of stronger carbide-forming elements (V, Ti, and Nb). The intention was that these microalloying elements should preferentially combine with the carbon in a given steel, leaving Cr and Mo uncombined in the ferrite to provide enhanced corrosion resistance. In addition, the presence of Si can lead to bainite formation under normalized conditions. Thus, this element, together with Ni, was used to bring back the strength caused by the loss of carbon. The transfor-

668

mation characteristics of the steel and the heat treatment following initial cooling was studied to allow measures to prevent heat-affected zone (HAZ) cracking during welding. The strength and toughness requirements were met through grain size refinement, the promotion of bainitic microstructure, and the precipitation strengthening effect of the alloy carbides. Extensive metallurgical and corrosion characterization of laboratory heats44 and industrial casts,80 produced with new but economically realistic ranges of alloy contents, has led to the development of novel categories of low C, Cr-containing steels with superior resistance to CO2 corrosion. The work utilized the existing metallurgical knowledge of modern steel properties and extensive knowledge of steel corrosion behavior to identify the role of alloying elements to define optimum steel compositions likely to meet the strength, properties, weldability, and corrosion resistance targets required by the oil industry. Based on the outcome of this extensive study, certain compositional trends were confirmed. The results are summarized in of individual alloying elements in Figure 5, as follows: Cr — An optimum Cr content had a significant beneficial role on the CO2 corrosion performance of the steels. They categorized the effect of Cr as follows: —5% Cr category: the lowest corrosion rate —3% Cr category —1.5% Cr category —1% Cr category —0.02% Cr category: the highest corrosion rate An overview of these categories are presented schematically in Figure 6, illustrating a progressive reduction in corrosion rate with increasing Cr content, the extent of which is subject to other alloying constituents and heat treatment. While 3% Cr proved to offer a 10-times reduction in corrosion rate, 1.5% Cr was not sufficient to ensure this level of resistance. The optimum level of Cr addition was not determined, albeit a level between 2% to 3% Cr was considered essential to achieve the expected improvement in corrosion performance subject to additional microalloying constituents. V — V had a major beneficial effect on reducing corrosion rate.45,80 Ti — Ti had some beneficial effects on the corrosion rate, although inconsistent, and some unsatisfactory effects on mechanical properties. Control of properties in the Ti steels proved difficult, although Ti additions could help to reduce HAZ hardness. Mo — Mo had no effect on the corrosion rate in the active state at low pH, but helped to get a genuine ivation in case of upward pH shifts beneath an already protective corrosion scale. Si and Cu — These microalloying elements showed beneficial effects on CO2 corrosion complementary to the effect of Cr, albeit subject to the



microstructure, heat treatment, their interaction, and corrosion conditions. It is concluded that V-microalloyed steel containing Cr, Si, Mo, V, and Cu is the most promising composition in of corrosion resistance and mechanical properties.44-45,80 This steel offered good levels of strength and toughness and good hot ductility, making it suitable for continuous casting. Weldability is an issue for this composition, making it only readily suitable at this time for downhole applications (e.g., threaded connections).80 Further studies are underway to improve the weldability by reducing C, V, and Mo contents, without compromising corrosion resistance. A very small addition of Ti (below stoichiometry with nitrogen) could be used to reduce HAZ hardness.78 While corrosion rates of reference steels (X70, X65, or L80) increased with time, the new composition steels exhibited progressively reduced corrosion rates with time, stabilizing after the initial exposure—a clear indication of the progressively protective nature of the corrosion film that formed on the experimental steels and the necessity to carry out long-term corrosion experiments (in excess of 7 days) to allow steady-state conditions. The conclusion drawn is that, as expected, Cr is effective above a certain level, below which it is detrimental to the anodic reaction on bare steel. Inconsistency in the corrosion performance of low-Cr-containing steels has been experienced by a number of operators.8-9,73,76 The data demonstrated that the calculated value of “free Cr and V,” generally, proved to be good indicators of CO2 corrosion performance—nevertheless, microalloying constituents and resultant microstructure have influential and complementary roles.44,80

SURFACE FILMS; CORROSION LAYERS CO2 corrosion of carbon and low-alloy steels is strongly dependent on the surface films formed during the corrosion processes. The protectiveness, rate of formation/precipitation, and the stability of the film control the corrosion rate and its nature (general corrosion or localized corrosion, especially mesa attack). Precipitation kinetics of FeCO3 film is affected by the iron and carbonate concentrations, and its subsequent formation and growth are extremely temperature sensitive.46 It is not the thickness of the film but the structure and its morphology that leads to low corrosion and protectiveness.9,19 It is interesting to note that a corrosion layer containing the same solid components can be either extremely protective82 or not very protective, or can even be corrosive.19,25 It has not been very clear why under some conditions these scales form and mitigate further corrosion and sometimes, in spite of favorable thermodynamic conditions for their formation, they do not precipitate at all and the corrosion continues unhindered.


In general, the protective characteristics of a corrosion film/layer depend on both the carbon steel characteristics (microstructure, heat treatment history, alloying elements) and environmental variables (solution pH, temperature, solution composition, flow rate, etc.). The former has been covered earlier through influencing and modification of steel chemistry and treatment processes.8-9,41-44,47,70-79 This section focuses on the latter and discusses how film/scale properties are influenced by the environmental factors, bearing in mind: —their properties and effects on corrosion rate —effects of various variables on the properties of the film/layer —modification of the surface films and its growth habits Further studies that are necessary for the improvement of the corrosion of carbon and low-alloy steels in CO2-containing environments through enhancing the properties of the surface film are proposed.

Film Formation Based on extensive observations made by many workers, corrosion films in the 5°C to 150°C temperature range in water containing CO2 can generally be divided into four main classes: —transparent films —iron carbide (Fe3C) films —iron carbonate (FeCO3) films —iron carbonate plus iron carbide (Fe3C + FeCO3) films These are reviewed in this section and their overall characteristics are summarized in Table 2. Transparent Films — Transparent films are rarely cited in the literature. They are <1 µ m thick and only observed at around room temperature. The film appears to form faster by reducing the temperature below ambient. This class of film is not thermodynamically the most stable solid corrosion product and can form in CO2-containing water with a very low ferrous ion concentration. Increasing the ferrous ion concentration makes this film more protective, slowing down the corrosion rate by about an order of magnitude and possibly more after long exposures.25,35,55,77,83 Carbon steels protected by this transparent film can be susceptible to crevice and chloride pitting corrosion in a similar manner to ivated stainless steels.44,56,72,83 Auger electron spectroscopy showed that this film did not contain carbonate, but rather iron and oxygen ions roughly in the proportion of 1:2. Etching indicated a constant ratio between iron and oxygen in the full thickness of the film.83 It is important to examine this film in view of the controversy on the existence of metastable solid FeCO356 and the well-known dehydration of FeOOH in the steps leading to the classical ivation of iron.84 As these observations were not repeated, it is now ques-

669


TABLE 2 Characteristics of Corrosion Films Corrosion Film Class

Temperature Range of Formation (°C)

Characteristics/ Nature

Growth Habit and Composition

Transparent

Forms at room temperature and below

<1 µm thick, transparent— once formed, it is very protective

Forming fast as temperature reduces to < room temperature mainly consisting of Fe and oxygen

Iron carbide

No range

<100 µm thick, metallic, conductive, and nonadherent

Spongy and brittle, consisting of Fe and C

Iron carbonate

Min. required in laboratory conditions 50°C to 70°C

Adherent, protective, and nonconductive

Cubic morphology, consisting of Fe, C, and O

Iron carbonate + iron carbide

Maximum 150°C (higher temperatures not studied)

All depends on how FeCO3 is blended with Fe3C

Consisting of ferrous carbide and ferrous carbonate

FIGURE 7. A pure iron carbide layer formed at 60°C and 1 to 3 times supersaturation.72

FIGURE 8. An overview of corrosion layers seen on a tubular showing a tough, empty cementite layer adjacent to gouges caused by wire line runs (courtesy of J.L. Crolet).

tionable whether this O/Fe ratio of 2:1 actually corresponded to FeII or FeIII.58 Transparent films have been ignored by many researchers and a systematic study is needed to con-

670

firm or otherwise invalidate its formation and its effect on FeCO3 formation. Iron Carbide—Cementite (Fe3C) Films — Anodic dissolution of carbon steel leads to the formation of dissolved ferrous ions. This process leaves behind some uncorroded Fe3C film (cementite) that accumulates at the surface. Unlike FeCO3 scale, the Fe3C film can be fragile and porous and therefore susceptible to flow conditions,6 or it can be a tough cementite network similar to “graphitization” of pig iron in acidic waters.58 A tough cementite layer adjacent to gouges caused by wire line runs is shown in Figure 7.27 At fast flow rates in unbuffered CO2containing water, the corrosion film consists mainly of Fe3C plus constituents of some alloying elements from the substrate. A reduction of the flow rate may increase the amount of Fe3C, but it also leads to the presence of FeCO3 in the film.28-29,31,35 Figure 8 shows an overview of a pure Fe3C film referred to as “empty cementite.”72 An empty cementite network forms a conductive porous sponge layer on which the cathodic reaction can take place. It is very adherent, with a metallic to black appearance. Its thickness is up to 100 µ m when formed under laboratory conditions19 and millimeters in the field.27 The morphology shown in Figure 9 was thus observed at the region of an unlifted water slug at the bottom of a gas well in Nigeria in very agitated fluid conditions.27 This shows nonprotective tubercules of profuse FeCO3 deposits above a thick base of empty cementite. Fe3C film significantly affects the corrosion process and increases the corrosion rate by a factor of 3 to 10 by playing a number of roles, including: —Galvanic coupling: Fe3C has a much lower overpotential for the cathodic reactions than iron and thus galvanic between the two can accelerate the dissolution of iron by accelerating the cathodic reaction in the presence of <<1 ppm Fe2+ in water.25,72,77



—Local acidification: Cathodic reactions can take place preferentially at Fe3C sites, thus physically separating the anodic and cathodic corrosion reactions. This leads to changes in the water composition with the aqueous phase at cathodic regions becoming more alkaline and that at the anodic sites more acidic.18-19,25,27,72,77 This can cause internal localized acidification and promote corrosion on the metal surface. —Fe2+ Enrichment: Having left behind the carbide layer, ferrous ions dissolving at the metal surface diffuses over a larger distance, so that the continuous concentration gradient necessary for this diffusion ends in a larger Fe++ enrichment at the metal surface. This increases local supersaturation of ferrous ions and facilitates the formation of FeCO3. —Film Anchoring: In certain conditions, corrosion film consists of a combination of Fe3C and FeCO3. In these films, Fe3C acts as a framework anchoring the precipitated FeCO3 on the surface-enhancing film properties with improved tolerance to mechanical shear at high flow rates. In these situations, localized corrosion is greatly reduced.38,79,83 Despite a high concentration of ferrous ions, local acidification at the surface may lead to unfavorable conditions for the precipitation of FeCO3.19,25,56,77,82 This may lead to FeCO3 precipitating within or on the top of the cementite layer. This type of film forms either a corrosion layer with poor and bonding to the metal surface or with unfilled regions between metal surface and the corrosion film. This corrosion film provides little protection, so corrosion rates can be high. A local high corrosion rate tends to increase the local pH difference between adjacent anodic and cathodic regions, which subsequently favors further development of the nonprotective film.19,25,72,79,82 In general, the accumulation of Fe3C has a contrasting role on corrosion behavior, depending on its manner of formation and its dominance within the film structure and process. On one hand, by preventing the diffusion of ferrous ions from the surface, it promotes the formation of FeCO3 film to offer a degree of protection. By blending uniformly into FeCO3 film, it will enhance its properties and protectiveness, as described in the following section. On the other hand, Fe3C could provide local acidification and facilitate galvanic corrosion and hence increased rate of attack. Invariably, steel microstructure governs carbide distribution, which thus affects film stability or instability. This has been postulated to govern corrosion behavior as described elsewhere in the paper (Alloying Elements Section44,80). Iron Carbonate—Siderite (FeCO3) Films — In of corrosion mitigation, FeCO3 or siderite is the most important film that can grow on carbon steels in sweet environments. Film formation is strongly


FIGURE 9. Iron carbide layer forming on the metal surface followed by partial siderite sealing leading to a nonprotective film.56

dependent on the thermodynamics and kinetics of FeCO3 precipitation. Supersaturation plays the most important role in FeCO3 film growth and its morphology. A high supersaturation of FeCO3 is necessary to form a protective film, particularly at low temperatures.41,72 In principle, the precipitation process comprises two steps, nucleation and particle growth. The morphology of the film therefore depends on the dominating step.72,77 Once the film is formed, however, it will remain protective at a much lower supersaturation.41 Protective film formation is accelerated by measures that restrict the transport of reaction products from the surface.72 The adherence and thickness of the FeCO3 scale depend on the metal microstructure.8-9 The scale grown on normalized steels with a pearlitic/ferritic microstructure was more adherent, having larger crystals more densely packed and thicker than those formed on quenched and tempered steels.85 This is explained further in the following section. FeCO3 reduces the corrosion rate by reducing and virtually sealing film porosity. With altering neither the local phase compositions nor the concentration gradient, this restricts the diffusion fluxes of the species involved in the electrochemical reactions. Moreover, even prior to sealing cementite, its precipitation can lead to coverage and, therefore, can limit its electrochemical activity. This is explained further in the section on operating temperatures. All authors agree that increasing the temperature would improve the protectiveness of the FeCO3 scale as well as its adhesion and hardness35 and that the higher the temperature, the more improved the protectiveness. However, there is little agreement on a practical “threshold” temperature. Some have reported that the maximum corrosion rate observed for carbon steel in sweet environments was from 60°C to 70°C and then it started to decline due to growth of protective FeCO3 films.86-87 In another work,35 it has been suggested that the lowest temperature necessary to obtain FeCO3 films that would reduce the corrosion

671


rate significantly was 50°C and the protectiveness was increased also by increasing the pH. It has been argued that the protective films formed at higher temperatures and pressures provided better protection than those formed at low temperatures and pressures. The level of protection increased with exposure time,88 which is the time dependency of the process. Some key environmental parameters that influence the iron carbonate film formation/precipitation are described here. —Temperature Effect: As mentioned earlier, growth of FeCO3 scale is a slow temperature-dependent process.41,46 It has been reported that FeCO3 scale was more protective at higher temperatures (60°C to 100°C) with little protection at lower temperatures6,33,89 (<60°C) with Fe3C predominating the film. Above 100°C, a tight and adherent film was formed with suggestions6 that magnetite was the stable scale above 121°C and the dominant scale at 250°C,90 although magnetite has not been detected in field conditions. An increased precipitation rate at high temperatures (>60°C) could explain why the corrosion rate went through a maximum in the temperature range from 60°C to 90°C. —pH Effect: pH is the most important factor in FeCO3 scale formation. It influences the solubility of FeCO3, and, under suitable conditions, increasing pH decreases the FeCO3 solubility and promotes its precipitation in the environment, resulting in lower corrosion rates. In CO2-saturated environments, formation of FeCO3 would decrease the corrosion rate by diminishing the diffusion of reactive species through the precipitated FeCO3 film.33,41,91 The pH increase also improves the protective properties of the FeCO3. Videm and Dugstad35 reported that at pH 6.00 good protection could be obtained by FeCO3 films even at room temperature. In a related work, they demonstrated that an increase in pH also made film formation more likely as a result of a reduction in the solubility of Fe2+.83 Similarly, de Moraes, et al.,92 reported that protective layers could be observed only for pH values >5—very protective layers were present only at high temperatures (93°C) and high pH values (pH ≥ 5.5), although these results are now disputed.18 —Fe2+ Concentration Effect: Siderite formation occurs in conditions where the concentration of Fe2+ ions in the aqueous phase exceeds the solubility range of FeCO3, bearing in mind that other species also influence this process. It is also noteworthy that there are indeed two corrosion products, namely Fe++, resulting from the anodic reaction and HCO3–, from the cathodic reaction.27 In uniform corrosion, they are released consistently so that different concentrations of released ferrous ions also correspond to different concentrations of released bicarbonate, which will lead to local changes in pH.27,29 A ferrous ion

672

concentration well below the solubility limit of FeCO3 not only prevented the formation of FeCO3-containing films, but also dissolved the existing films.35,81 Videm and Dugstad28 concluded that a change of 30 ppm Fe2+ can sometimes affect the corrosion rate to the same degree as a change in CO2 concentration of 1,000 ppm (2 bar) at 90°C. —H2S Effect: In the presence of both H2S and CO2, simplified calculations indicate that iron sulfide may constitute the corrosion product layer when the H2S/CO2 ratio exceeds about 1/5,000—sour systems considerations then would be expected to apply.93 Occasionally, it has been reported94 that the corrosion product film formed under these conditions (containing FeS or Fe2S depending upon the H2S partial pressure) was apparently more protective than FeCO3. The opposite also has been reported, especially at very low H2S concentrations, wherein a mixture of layers forms comprising inner carbonate and outer sulfide.95-96 Iron Carbonate (FeCO3) Plus Iron Carbide (Fe3C) Films — This type of film is the most common film found on carbon and low-alloy steel surfaces in CO2containing environments. During CO2 corrosion of carbon steels, the Fe3C phase is cathodic (corrosion resistant) and may embed within the FeCO3 film. The structure of the film then will depend on where and when the FeCO3 precipitation takes place. On one hand, if it occurs directly and integrates within the carbide phase, then a protective and stable film will form that often can stand high fluid flow conditions. On the other hand, initial formation of a cementite layer on the surface followed by partial FeCO3 sealing close to or beyond the external limit of cementite can lead to a nonprotective film. An example of this is shown in Figure 10.25,27,72 In contrast, if cementite phase effectively sealed the siderite layer formed in with the metal surface, an incomplete sealing or a partial redissolution of FeCO3 anywhere else is not detrimental and the corrosion film remains protective, as shown in Figure 11.25,27,72 Crolet, et al.,25 have categorized the morphologies of corrosion film formation as affecting protectiveness. This is shown in a simplified diagram in Figure 12. The diagram is based on an analysis of the diffusion/precipitation issues and the resulting pH shifts, and is backed by observations of the actual morphologies of protective and nonprotective corrosion layers. As mentioned earlier, the structure of the mixed film plays an important role on the formation and breakdown of protective carbonate films. This is influenced by the carbon content and the size and distribution of carbides, which therefore is dependent of the microstructure of steel.85 The ferritic-pearlitic steels have a carbide structure, which gives a good framework for the buildup of protective carbonate films.79 Quenched and tempered steels and ferritic



FIGURE 11. A cementite layer sealed by siderite forming a protective film.25,72

FIGURE 12. Different morphologies observed for protective and nonprotective corrosion layers (after Crolet, et al.25).

FIGURE 10. An empty, thick, and porous cementite network formed on tubing recovered from the field, which was covered by a porous but hard and mechanically resisting siderite tubercule (courtesy of J.L. Crolet27).

steels with low carbon content have a finely distributed carbide structure that does not give an integrated framework to anchor and enhance the formation of protective carbonate films.32,38,77,79 Ex-


periments with the plain carbon steel after different heat treatments showed that both corrosion rate and the ability to form protective films decreases with increasing tempering temperature, indicating that the carbide structure of the steel plays an important role for the formation of protective corrosion films.77 As mentioned earlier, it is likely that Fe3C lamellae at the metal/corrosion film interface can anchor the film and contribute to corrosion protection.27,35,79 This anchoring effect of the carbide phase is believed to be the main factor influencing the adhesion of the FeCO3 film to the surface and has been reported by many workers.9,33,55,77,79

Film Breakdown and Localized Corrosion A sensitive balance between the formation and destruction of the protective film governs the subsequent corrosion progress and determines whether uniform or localized attack will take place.30 Mesa attack can be initiated by local breakdown of the protective film or by the growth of pits beneath

673


the porous corrosion film by a local cell mechanism. The remaining corrosion film on top is removed stepwise by the turbulent flow before the pit can continue to grow into a larger mesa attack. When the film has been torn away in one area, the corrosion attack continues further beneath the film close to the edge until another piece of the film is removed. It is assumed that a galvanic cell is set up between the anodic, film-free metal in the bottom of the mesa attack and the cathodic, film-covered metal outside the mesa attack, leading to a high corrosion rate in the base of the mesa-attacked area.33,43,97 Enhancement in the formation of protective films affects the corrosion rate in the base of the mesa by enabling the formation of carbonate films, which can then stop further growth of the mesa attack. Shallow mesa attack covered with corrosion films have been found in experiments on precorroded samples, and linear polarization resistance (LPR) measurements have shown that the growth of this mesa attack slows down gradually during the experiment and eventually stops. This shows that reformation of protective films in mesa-attacked areas is possible when the growth rate of the mesa attack is not too high.31,41,72 With a better quality of the initial protective films, the effect of galvanic coupling between the mesa-attacked area and the surrounding area with intact corrosion films can be smaller, resulting in a slower and less catastrophic mesa attack. Rate of growth at mesa-attacked areas has been shown to be excessively higher than the adjacent unattacked areas, reaching some 20 mm/y to 30 mm/y in water with around 2 bar CO2, pH 5.5 to 5.8, and 3% NaCl at 80°C.31,41,72,97 In laboratory tests, mesa attack occurred under these conditions at a flow rate of 7 m/s but not below 2.5 m/s. Low-chromium-containing steels (0.5% to 1% Cr) did not suffer deep mesa attack under these conditions.97 Fluid Flow Effects — The effect of flow rate on the corrosion rate originates primarily from localized flow-induced changes introduced by the corrosion film. Fe3C particles remaining on the surface, the FeCO3 buildup, and the accumulation of alloying elements in the corrosion film are flow-dependent phenomena.30 As mentioned earlier, a clear distinction should be made between pure CO2 corrosion and a combined interaction of erosion-CO2 corrosion. Flow rate of water containing CO2 has two principal effects on film formation and corrosion rate and form. First, it prevents the formation and slows down its growth by reducing the local supersaturation. Second, flow can damage film locally to cause localized corrosion, especially mesa attack. When the surface corrosion film is poorly adherent to the surface, the effect of flow rate on localized corrosion is most serious.47,69 In low-temperature conditions from 30°C to 60°C, films might not form or might not become pro-

674

tective for a long period in continuous flow loop experiments. This therefore could lead to fast uniform corrosion, although protective films can form at these temperatures when water stagnation on steel is prevalent.70,83 In the presence of solids (sand), erosion-corrosion can produce higher wall penetration rates than erosion or corrosion alone. Three typical behaviors have been found. At low velocities, a protective FeCO3 scale formed over all surfaces of an elbow, and corrosion rates were very low. At high velocities, impingement on elbow surfaces by sand particles entrained in the flow prevented protective layers forming in the elbow. Corrosion rates were high and uniform over the entire surface. At intermediate velocities, protective layers formed over the entire elbow surface, except at very localized points where impinging sand particles prevented scale formation. Deep pits formed at these locations and wall penetration rates were extremely high. These conditions are damaging but can be avoided by reducing or increasing flow velocities. A computational model for the prediction of sand erosion in piping systems was used to simulate experiments to explain the three observed behaviors and predict conditions defining boundaries between them.98 Siderite (FeCO3) formed only in low-velocity, single-phase flows where pH increased above 5.0. Corrosion rates after scale formation decreased by almost 1 order of magnitude. No FeCO3 scale formed in tests involving high-velocity, low-pH, single-phase flows or two-phase flows.85 In fact, at medium or high pH conditions, protective films always form unless the following parameters are met contly in laboratory testing:58 —initial bare surface —room temperature —uninterrupted high flow rate from the start of the test In the absence of each, films do form and become protective and their protectiveness remains unaffected by flow. This basic irreversibility of corrosion layers82 explains the reported temperature effect, which essentially appears as a laboratory artifact and most unexplained good performance of the film experienced in real field conditions, which can be the result of inevitable shutdown periods.58 The influence of solid particles (sand) and their interactions with flow-induced corrosion is not fully understood, and predictive models do not include corrosion-enhanced erosion by solids.

Iron Carbonate/Siderite (FeCO3) Scale: Recent Progress The significance of steel characteristics and environmental variables on the protective nature of FeCO3 (siderite) corrosion films was emphasized earlier. A recent study has demonstrated the intricacy of



FIGURE 13. Iron carbonate scale precipitated on carbon steel samples at 75°C exposed to a solution at pH 6.30 at 1 bar CO2 showing a somewhat cubic appearance (magnification: X6,000).99

FIGURE 14. Iron carbonate scale formed after 260 h of immersion in a solution (at pH 6.30, 75°C, and 1 bar CO2) showing a compact appearance.99

how corrosion protection offered by a precipitated FeCO3 scale on carbon steel substrate can be improved through varying or adjusting some of the environmental parameters such as pH, temperature, solution composition, and periodic exposure to air.99 Among these, pH was shown to be the most influential variable in affecting the precipitation of FeCO3. The key factor preventing the growth of a protective siderite scale was that the nucleation and growth of crystals of this compound were extremely temperature sensitive and slow with a rather high activation energy, confirming an earlier indication by others.46 In this respect, for siderite as well as calcite and other forms of calcium carbonates, the limits of the domain of precipitation are not driven by saturation (thermodynamic stability) but by supersaturation (kinetics). Also, it is worth noting that scaling inhibition is crystal specific and there was an indication that some scale inhibitors could prevent the formation of calcium carbonate but not siderite.99 Key points of observation emerging from this work on corrosion films/layers formed under siderite-forming conditions (pH range from 5.60 to 6.30, 75°C, and 1 bar CO2 pressure) are highlighted:90 —The film consisted primarily of FeCO3 and resembled a perfect cubic morphology (bearing in mind that siderite is a rhombohedric crys-

tal) as shown in Figure 13 as also previously seen by others.41 —The film was a few micrometers thick and formed just after 24 h immersion. Scale thickness, as measured by a scanning electron microscope (SEM), increased progressively with time, becoming more compact (Figure 14). Siderite “cube” dimensions play an important role here, as suggested by others,8,18,27 in affording further protection by hindering the ingress of corrosive species toward the metal substrate. It should be noted that the key issue of the liquid surface state has not been observed by SEM. —Periodic exposure to the atmosphere, or more precisely to oxygen (through removal and reexposure of samples), improved significantly the scale adhesion to the substrate. —The work reiterated that FeCO3 precipitation coincided with a sudden drop in weight change and its associated corrosion rate. —Similar immersion tests conducted with pure iron (99.5%) samples produced a patchy, cracked, and loosely adherent FeCO3 scale. This was attributed to the absence of a carbide network in pure iron. —Solution chemistry played an important role in altering the protective characteristics of the


675


TABLE 3 Effect of Various Parameters on Scale Protectiveness 99 Parameter

Surface Coverage

Stability

Adhesion

Density

— — * * —

* * — — *

* * * * *

— — — * —

Oxygen Magnesium Temperature pH Carbide phase

* Improvement in certain characteristics of the scale. — No improvement was observed. Note: Stability should not be confused with adhesion. Stability means the endurance of the scale toward environmental variables like solution pH and flow velocity.

precipitating FeCO3 scale. In these series of tests, scales developed in distilled water were less adherent to the substrate than those developed in a solution containing trace Mg2+ ions simulating produced fluid chemistry, and Mg2+ ions improved siderite scale protection and adhesion. However, magnesium-related trends have not been observed in field evaluations.14,36 —The nucleation and growth of FeCO3 crystals was a slow, temperature-dependent process; consequently, any environmental changes that retarded the precipitation process therefore led to progressively increased corrosion. The work highlighted that the “protectiveness” or “corrosion-resistant properties” of a scale were loosely defined definitions. Therefore, the protective characteristics of FeCO3 scale were translated through the following specific properties allowing better characterization: —scale density (or, in other words, pore distribution in the scale) —adhesion —stability —surface coverage Based on this, the effect of various environmental parameters affecting scale characterization and protectiveness have been tabulated. These are summarized in Table 3, demonstrating that solution pH is the most influential parameter in the precipitation of FeCO3 scale and seemingly the only parameter that affects scale density.99 The work of Nes˘ic´, et al., has led to modeling of the possibility of predicting the influence of FeCO3 film.20

CORROSIVITY As a general statement, the corrosivity of the aqueous phase will be dependent on CO2 partial pressure, temperature, solution chemistry, and insitu pH. On one hand, at a given partial pressure of CO2, condensed waters (associated with gas produc-

676

tion) are in general more corrosive than formation waters since they do not contain any buffering or scale-forming species and inherently have a lower insitu pH. On the other hand, at a given pH, the simple chemistry of condensed water at low CO2 partial pressures may be less corrosive than some complex chemistries of reservoir water under higher CO2 partial pressures.27 Software and spreadsheet-based programs now exist for characterizing water chemistry such that concentrations of dissolved species and solution pH can be readily calculated.36,100-104 Nevertheless, the oil and gas industry has grown up assessing the presence of dissolved CO2 in water in of CO2 partial pressure. This simply derives from the application of Henry’s Law for a gas in equilibrium with a liquid and the approximation of ideality, where the thermodynamic activity of CO2 in water is assessed through its partial pressure in the gas. This remains valid up to high CO2 partial pressures, particularly on a logarithmic pressure scale, which is that related to pH variations. Viewing it starting from reservoir conditioning, Crolet and Bonis14,36,55,105 consider variations in CO2 concentration resulting simply from differences in the solubilities and activity coefficients in the phases present. Thus, for example, in three-phase production, which is generally highly turbulent, even though the CO2 equilibrium is never perfect, any deviations remain relatively insignificant. Consequently, it is more reasonable, and certainly much easier, to express the amount of CO2, not in of its concentration in each phase, but as its “chemical potential” (or thermodynamic activity or fugacity). Provided conditions are not close to the gas liquefaction point, this is simply equivalent to the partial pressure of CO2 in the gas phase. By extrapolation, reference then can be made to CO2 partial pressure even in systems where there is no free gas phase present. Experimentally, this “virtual partial pressure” corresponds to the real partial pressure of CO2, which would appear in the nascent gas phase if the system were to be depressurized. At the bottom of an oil well, this “virtual partial pressure” will be the very first bubble at the “bubble pressure” of the reservoir fluid; in a gas free line downstream of the gas separator, this will be the partial pressure at the last gas in the separator.18 Variations of dissolved CO2 along the flow line will affect the solubility of potential scaling compounds, particularly CaCO3, which can interfere with the corrosion process and possibly its inhibition, if precipitated. The in-situ pH level can also affect the solubility of ferrous oxides and hydroxides, increasing their solubility with decreasing pH. However, of particular importance are conditions that favor deposition of FeCO3 (siderite). Formation of FeCO3 serves to complicate the picture further, as it can, under



certain conditions, ivate the corrosion reaction making it even more difficult to accurately predict corrosion rates.58

CO2 CORROSION PREDICTION Over the years much effort has been expended in studying factors controlling the performance of materials in production environments in an attempt to define the safe operating limits of candidate materials in of the environmental and mechanical conditions. This knowledge is gained, in part, from some comprehensive, empirical-based field statistics, such as those by API,7,68,106 laboratory-based information translated by quantitative regression to predictive a corrosion rate,9,91,100-104 and comprehensive well statistics based upon the actual field values of relevant scientific quantities.36,105 The laboratory-based equations have been conventionally based using the de Waard and Milliams nomogram.9,96-97,100 Nonetheless, in the past, it was imperative that candidate materials were assessed in the laboratory prior to their application in the field so that costly and potentially hazardous failures are avoided. However, the relevance of some of these short-term tests to long-term performance was exceedingly difficult to establish other than by longterm experience or evaluation. The inherent corrosivity of acidic gases has long been recognized and has prompted extensive studies of the subject.6-9,100 These have led to the development of predictive models or criteria that attempt to correlate corrosion to acidic gas partial pressure. However, in merely using these predictive models, there is a danger of working with a too simplistic or clear-cut picture of the situation. Any predictions that are made are not only influenced by the validity of the model, but also the quality and reliability of the supplied production and operational data used to calculate the corrosion risk. The engineer ideally wants a predictive tool that can be readily applied and is suitable for application at all stages of project development and subsequent operation. This may seem a tall order but it may nevertheless be argued that the fundamentals of the CO2 corrosion process will be common to all situations; it is the overlying effects of such factors as flow regime, film formation/deposition, hydrocarbon phase, realistic and representative testing, steel chemistry, and corrosion inhibitor that cloud or complicate the picture. A true industry standard approach to predicting CO2 corrosion does not exist, although there are aspects of commonality between the various approaches/models in use by the industry. There is no professional body or agency such as NACE International providing standard basic guidelines to work from—unlike the situation for H2S.81,107-108 While it


may be argued that such guidelines, because of the complexity of the problem, would either be too generalized and/or too conservative for certain situations, their presence nevertheless would have a unifying influence to a degree and at least provide a common reference point from which to work. Having determined the water chemistry, this then needs to be translated into a corrosion risk, preferably expressed quantitatively as a rate and type of attack. The easiest parameter to relate to (and in general, measure) experimentally and in the field is the partial pressure of CO2. Thus, over the years, a number of empirical relationships have evolved with varying levels of complexity and theoretical substance. In the simplest form, API7 in the late 1950s provided a “Rule of Thumb” criteria for carbon and low-alloy steels. After a first period of high popularity, this rule endured several exceptions and fell into oblivion with just periodic quotes, such as that by Mudge and Levesque,109 who cite the following “Rule of Thumb” criteria for carbon and low-alloy steels, or the complementary note from Crolet:18 —PCO2 < 0.5 bar (7 psi) Corrosion Unlikely: implying that corrosion is uniform and the rate is <0.1 mm/y —0.5 bar (7 psi) < PCO2 < 2 bar (30 psi) Corrosion Possible: implying that corrosion rate might be between 0.1 mm/y to 1 mm/y and design consideration is based on life expectancy —PCO2 > 2 bar (30 psi) Corrosion likely: implying that corrosion rate may exceed 1 mm/y, which is unacceptable where partial pressure of CO2 is denoted as PCO2 These are based on field experience, principally in the U.S. No subsequent standard guidelines exist as to the courses of action necessary for each condition, although experience will trigger the corrosion engineer to look initially toward certain possible options (e.g., use of a corrosion inhibitor or more corrosion-resistant alloys). However, deg on such a basis is not a very satisfactory way to operate. Rules of thumb may be viewed as: —an aid to first- materials selection —offering qualitative and generalized assessment More specific questions on service life, risk analysis, corrosion allowance, and inhibited corrosion rates require the ability to conduct a quantitative assessment (i.e., the ability to predict corrosion rates for a given set of conditions). In addition, consequence of corrosion is a key issue when corrosivity assessment is carried out. Nevertheless, the relationship between corrosion rate and CO2 partial pressure has formed the basis of a number of predictive models/equations for carbon steels, even though claims have been made that PCO2 bears little relation to the decision by the field operator.110 Experimental work has lead to a number of equations or models being developed relating corro-

677


sion rate of carbon and low-alloy steels to CO2 partial pressure,11,36,91,94,100-104,111 from which individual operators have each developed their own approach or philosophy. Most equations or models have been derived from a range of experimental conditions. For example, the de Waard and Milliams equation or nomogram11 was originally derived for a 0.1% NaCl solution test medium at a flow velocity of ca. 1 m/s, temperatures up to 80°C, and PCO2 up to 1 bar for X52 C-steel. Greco and Wright94 used 0.04% NaCl solution at 86°F and PCO2 up to 4.5 bar; Videm and Dugstad28 used distilled water, up to 120°C, 6.5 bar PCO2, and 20 m/s flow velocity. The models developed by Fang, et al.,102 and Liu and Erbar104 are based on theoretical analysis together with the use of data taken from the literature—both of these models were specifically developed for gas well corrosion. While the use of simple NaCl brine might eventually mimic water in gas systems, when considering brines co-produced with crude oil, their water chemistries are far more complex and their influence (buffering) thus needs to be corrected when using these equations. Many parameters, such as temperature, pH, partial pressures of CO2 and H2S, gas/oil ratio (GOR) and water production (water/oil ratio—WOR), water composition, together with flow rate and regime, have a greater or lesser effect on the corrosion rate. Furthermore, not all of these parameters are independent of each other, pH being a good example.8-9 Finally, direct field validation of the predicted corrosion rates often is difficult to unequivocally substantiate. This is particularly true when a model needs detailed data, which are seldom gathered in the field. Furthermore, these data are not totally defined throughout the service life of equipment.18 Several prediction models for CO2 corrosion of oil and gas pipelines are available. These models have very different approaches in ing for oil wetting and the effect of protective corrosion films, and this s for much of the differences in behavior between the models. All the models are capable of predicting the high corrosion rates found in systems with low pH and moderate temperature, while the models can predict quite different results for situations at high temperature and high pH, where protective corrosion films may form. An overview of capabilities and shortfalls of the available models has been provided by Nyborg100 and is also summarized in EFC Publication 23.9 An intermediate approach that places direct emphasis on knowledge of the water chemistry is the prediction tool developed by Crolet and Bonis.36,105 This tool is an “advanced” field database that calculates the factual input data, especially pH, free HAc, and potential corrosivity, in of corrosion risk for the occurrence of failure within the life of the well. It defines the severity of CO2 corrosion in wells

678

in of “low” and “high” risk as a function of workover statistics, plus a “medium” category for borderline or undocumented conditions. Crolet and Bonis compared these with all the basic information available at the time, combining both field experiences and laboratory data. Key parameters included in this multiple correlation were the “potential corrosivity” (what is observed in the laboratory on bare surfaces but just taken as an index of electrochemical reactivity and in no way a corrosion rate prediction) and the field features (in-situ pH, Ca2+/HCO3– ratio, concentration of dissolved organic acids [mainly HAc], and, to a lesser extent, flow velocity). This tool expresses two distinctive tables indicating corrosion risks, one for oil-producing wells and one for the gas-producing wells. It complements the API rules7 and explains their exceptions; notably that CO2 corrosion is indeed ubiquitous at high PCO2 and it progressively disappears at low PCO2, unless H2CO3 is supplemented by free organic acids and the “CO2 corrosion” by “HAc corrosion.”19,27,81 This, to a great deal, concludes the original debates launched in 19447-8,19,106 on the concept of damaging mechanisms caused by either CO2 or organic acids when the first decision was made in 1958 in favor of a CO2-only mechanism7 followed by the origin of the great many exceptions to the API rules.7,106

Industry Practice While an industry standard approach to predict CO2 corrosion damage does not exist per se, the work of Shell8-9,11,91,101 in this area has provided a strong reference statement from or against which to work. The de Waard and Milliams equation or nomogram has been developed as an engineering tool. It presents in a simple form the relationship between “potential corrosivity” (worst case) of aqueous media for a given level of dissolved CO2, defined by its partial pressure at a given temperature. The current form of the basic Shell equation is as follows:8-9,91,101,112 1 / Vcor = 1 / Vr + 1 / Vm log( Vr ) = 5.07 – 1,119 / ( t + 273) + 0.58 log(PCO2 ) – 0.34( pHactual – pHCO2 )

Vm = 2.7( Vliq 08 / diam 0.2 ) PCO2

(8)

(9)

(10)

where Vcor is the corrosion rate (mm/y); Vr and Vm represent the maximum kinetic reaction and masstransfer rates, respectively; t is temperature (°C), PCO2 is partial pressure of CO2; pHactual is the actual pH; pHCO2 is pH resulting from CO2 only; Vliq is liquid velocity (m/s); and diam is the hydraulic diameter (m). The important influence of temperature on CO2 corrosion is depicted in Figure 15.8,112



There are appropriate correction factors for gas fugacity, corrosion scales/surface films (i.e., Fe3O4 and/or FeCO3), the influence of flow (mass-transfer effects), and steel composition. Correlation with field experience suggests that the de Waard approach generally predicts the worst-case corrosion rate. The correction factors for pH and scale (scale factor applied at temperatures > 60°C) are simplified versions, but they nevertheless appear suitable for many applications, especially for formation waters. The relative simplicity of the de Waard and Milliams approach and its ease of use have undoubtedly been positive factors in its broad acceptance, although its complexity has grown over the years as better appreciation of the CO2 corrosion process and influence of water chemistry have evolved together with correlation with field experience.36,100,103,111 There would appear to be a trade-off between a model’s relative ease of use vs availability, detail, and reliability/accuracy of necessary input data/conditions vs degree of accuracy/absoluteness required of the assessment of the corrosion risk. The latter will also be influenced by the ease and sensitivity of subsequent corrosion monitoring and inspection. On the other hand, the well database of Crolet and Bonis36,105 also requires some tedious calculations for the figures used as input in the correlation, and this is also made through a spreadsheet called “CORMED.” This may have given the false impression that CORMED is just a model like others, whereas its goal is to translate the analytical sheets of the raw field data into the requested factual inputs (pH, free HAc, etc.) and automatically compare them to their respective critical values in the published tables. CORMED places direct emphasis on knowledge of the water chemistry and defines severity in of low, medium, or high risk. If early workovers in the first 10-year period have been reported in similar conditions, it is concluded that the corrosion risk is respectively high or otherwise low. Considering that a tubing is ~10 mm thick, this means that the average corrosion rate will be respectively above or below 0.1 mm/y, which is precisely the information given by conventional corrosion tables. Both the de Waard and Milliams nomogram and CORMED approach have been developed from a basic consideration of the CO2 corrosion reactions— the former is more empirical in origin and the latter is more theoretical. Both have attempted to for the overlying effects either by applying correction factors (de Waard and Milliams) or through field correlation (CORMED). Finally, it is worth restating that all available models/equations only apply to carbon and low-alloy steels and are not strictly valid in the presence of H2S due to formation of protective sulfide films, which may nevertheless be susceptible to localized breakdown under long-term exposure.


FIGURE 15. Effect of temperature on CO2 corrosion.8,112

EFC Publication 239 sets out a process for determining CO2 corrosion risk/rate. In achieving this, a key element is the interaction between the corrosion engineer and the petroleum engineer to ensure that the service conditions are correctly defined and relevant.

CO2 CORROSION CONTROL The wide-ranging environmental conditions prevailing in the oil and gas industry necessitates the appropriate and cost-effective materials choice and corrosion control measures. Corrosion can impose a significant cost penalty on the choice of material at the design stage, and its possible occurrence also has serious safety and environmental implications. Furthermore, as production conditions tend to become more corrosive, they require a more stringent corrosion management strategy. Corrosion control may be affected in a number of ways, either singularly or in combination. In the case of oil and gas production, the corrosion control methods include: —change/modify operational parameters (e.g., flow, temperature, remove water) or system design (e.g., remove sharp bends, deadlegs, crevices) —change/modify chemistry of the environment (e.g., remove O2 from any liquid reinjected in the production flow, lower CO2 partial pressure, scavenge H2S, add corrosion inhibitor, increase pH) —change/modify interfacial conditions of metal (surface modifications or possible implementation of cathodic protection)

679


—apply organic coating (e.g., fusion-bonded epoxy, phenolic-modified epoxy) or use a liner (e.g., polypropylene) to isolate metal from corrosive environment —use more corrosion-resistant materials (e.g., 13% Cr steel, duplex stainless steels) either in the solid form or as cladding on carbon steel —use nonmetallics (e.g., fiber-reinforced plastics) The relative merits of each approach must be viewed in the context of the application, the required service life, and the severity of the conditions. No hard and fast rules exist in a general sense, and many decisions are made based on past experience and individual preference. Clearly, each action is associated with a cost that highlights the need for good initial design together with a sound understanding of the possible corrosion processes if cost-effective corrosion management is to be achieved without compromising safety. Finally, the value and importance of corrosion monitoring and inspection should not be forgotten and goes hand-in-hand with the above for the most part and certainly where carbon and low-alloy steels are used. In addition, the role of top-of-the-line corrosion needs specific attention. Corrosion inhibitors are used extensively throughout the oil and gas industry offering, where appropriate, a cost-effective means of corrosion control, particularly for flowlines, pipelines, and trunklines. For the most part they are used to inhibit the reactivity of carbon and low-alloy steels ing the aqueous phase co-produced with oil and gas.

TECHNOLOGY GAPS This review has highlighted key areas of progress both mechanistically and industrially based on which a number of key messages recommending areas for additional research has been drawn. These can lead to further the understanding of CO2 corrosion mechanisms to enable improved predictive capabilities for the effective use and deployment of carbon and low-alloy steel in oil and gas production. Notable areas of technology gap requiring systematic studies include: —CO2 corrosion testing: a complex procedure and a topic of ongoing investigation. There is a great need to develop a field reproducible and consistent procedure to evaluate CO2 corrosion. This should truly reproduce the known field observation of corrosion damage from the worst cases to those providing adequate protection and incorporate all key parametric sensitivities. —Interaction of CO2 corrosion with H2S: the significance of H2S promoting or retarding corrosion has received little due attention and is a major source of concern in deg new development.

680

—Presence of hydrocarbon phase: hydrocarbon plays a vital role in CO2 corrosion in of its phase ratio, its partitioning capability for CO2, water wetting, and its inhibiting effect on CO2 corrosion. Further focused research to characterize the nature of hydrocarbon phase can lead to better predicative capabilities. —Steel chemistry: while major studies have led to the development of carbon and low-alloy steels with superior resistance to CO2 corrosion, these require fine-tuning to produce steels with inherent resistance to CO2 corrosion. —Flow modeling: further emphasis on flow modeling to better quantify water wetting/shedding can lead to improved predictive modeling, bearing in mind flow regime and their changes throughout the life of the field. —Welds: the performance of welds and preferential weld corrosion remain a major operational obstacle, particularly in the absence of corrosion inhibition. This subject has not been reviewed in the present article and requires focused attention. —HAc: while some recent work has discussed the influential role of HAc, its role requires further study to determine its complementary effect on CO2 corrosion. —Siderite formation: siderite, when formed, offers superior protection. Studies on early nucleation and growth of siderite can be of significant benefit in corrosion mitigation. —Erosion-corrosion: the presence of solids to induce erosion and its interaction with corrosion is poorly understood and the subject is in need of clarification to develop a cont erosioncorrosion model taking on board solid content.

CONCLUSIONS This overview article has led to a number of key conclusions emphasizing the importance of the subject in providing integrity management for oil and gas production facilities. Corrosion failures for 25% of all safety incidents, 2.8% turnover, 2.2% tangible asset, 8.5% increase on capital expenditure, 5% of lost/deferred production, 11.5% increase to the lifting costs or more quantitatively ca. 0.30 $/boe. The majority of these are associated with CO2 corrosion of carbon and low-alloy steels with implications on capital and operational expenditures (CAPEX and OPEX) and health, safety, and the environment (HSE) and operator dependency. The industry continues to lean heavily on the extended use of carbon and low-alloy steels, which are readily available and are able to meet many of the mechanical, structural, fabrication, and cost requirements. Their technology is well developed and they



represent for many applications an economic materials choice. However, a key obstacle for their effective use is their limited corrosion performance. CO2 corrosion of carbon and low-alloy steel remains a complex phenomenon and, despite several extensive studies over the past 4 decades, its mechanism, realistic predication, and control are in need of being addressed effectively. Its understanding, prediction, and control remain key challenges to sound facilities design, operation, and subsequent integrity assurance. CO2 corrosion is influenced by a large number of parameters, including environmental, physical, and metallurgical variables. All parameters are interdependent and can interact in many ways to influence CO2 corrosion. Ignoring the corrosion problems associated with sour service, low levels of H2S can affect CO2 corrosion in different ways, either complementing CO2 corrosion by acting as a weak acid or have a positive effect on sweet corrosion through the formation of a protective sulfide scale. Presence of HAc or, more generally, organic acids reduces the protectiveness of the films and increases the sensitivity to mesa attack. This is attributed to a lower Fe2+ supersaturation in the corrosion film and at the steel surface in the presence of organic acids. Corrosion scales, when formed under certain conditions, can afford superior protection. While their formation and growth have been the subject of many studies, favorable conditions for the development of a truly protective film to provide subsequent effective protection need further scrutiny. Precipitation kinetics of FeCO3 is affected by the iron and carbonate concentrations, and its subsequent formation and growth are extremely temperature sensitive. These can be categorized as follows: —At low temperature conditions (<75°C), growth is slow with consequent need for extremely long testing to address it effectively —At intermediate temperatures (75°C to 110°C), the precipitation reaction may influence corrosion reactions since there is no requirement for supersaturation —At high temperatures (>110°C), FeCO3 precipitation is likely to be rapid and growth is diffusioncontrolled. Iron released by the corrosion process therefore may re-precipitate immediately at the surfaces, providing a complete adherent protective layer. Steel chemistry plays a significant role in providing protection against CO2 corrosion and can lead to substantial economical gains.

ACKNOWLEDGMENTS The authors acknowledge J.L. Crolet (formerly with TotalFinaElf) for his significant input, help with


correcting errors, and addition of pertinent information. Helpful contributions of D. Harrop (BP), Y. Zhu (Windsor Scientific Ltd), and D. Williams (University College London) are also acknowledged. REFERENCES 1.

2.

3. 4.

5.

6. 7. 8.

9.

10. 11. 12. 13.

14. 15. 16.

17.

18.

19.

20.

21.

22.

23. 24. 25.

26.

27.

M.B. Kermani, D. Harrop, “The Impact of Corrosion on the Oil and Gas Industry,” J. SPE Production Facilities 8 (1996), p. 186-190. M.B. Kermani, “Materials Optimization for Oil and Gas Sour Production,” CORROSION/2000, paper no. 156 (Houston, TX: NACE International, 2000). P. McIntyre, Corrosion Management 46, 3/4 (2002): p. 19. M. Bonis, P. Thiam, “Measurement of Corrosion Costs within Elf’s Exploration and Production,” Eurocorr 2000, Conf. European Federation of Corrosion (London, U.K.: Institute of Materials, 2000). M.B. Kermani, D. Harrop, “Materials Selection for Oil and Gas Production and Transportation Facilities,” 3rd Int. Corros. Cong., held May 10-12 (Tehran, Iran: University of Tehran, 1993). R.H. Hausler, H.P. Gaddart, eds., Advances in CO2 Corrosion, vol. 1 (1985) and 2 (1986) (Houston, TX: NACE). Corrosion of Oil and Gas Well Equipment (Dallas, TX: American Petroleum Institute, 1958). M.B. Kermani, L.M. Smith, eds., Predicting CO2 Corrosion in the Oil and Gas Industry, European Federation of Corrosion Publication no. 13 (London, U.K.: Institue of Materials, 1994). M.B. Kermani, L.M. Smith, eds., CO2 Corrosion Control in Oil and Gas Production—Design Considerations, European Federation of Corrosion Publication no. 23 (London, U.K.: Institue of Materials, 1997). W. Schwenk, Werst. Korros. 25 (1974): p. 643. C. de Waard, D. Milliams, Corrosion 31 (1975): p. 131. G.I. Ogundele, W.E. White, Corrosion 43 (1987): p. 665. D. Harrop, J.W. Martin, C.W. White, ASTM/NPL Symp. on “Use of Synthetic Environments for Corrosion Testing” (Teddington, U.K.: National Physical Laboratory, 1986). J.L. Crolet, M.R. Bonis, Corrosion 39 (1983): p. 39. G. Schmitt, “Fundamental Aspects of CO2 Corrosion,” CORROSION/83, paper no. 43 (Houston, TX: NACE, 1983). G. Schmitt, S. Feinen, “Effect of Anions and Cations on the Pit Initiation in CO2 Corrosion of Iron and Steel,” CORROSION/2000, paper no. 1 (Houston, TX: NACE, 2000). G. Schmitt, D. Engels, “SEM/EDX Analysis of Corrosion Products for Investigations on Metallurgy and Solution Effects in CO2 Corrosion,” CORROSION/98, paper no. 149 (Houston, TX: NACE, 1988). J.L. Crolet, “Which CO2 Corrosion, Hence Which Prediction?,” in Predicting CO2 Corrosion in the Oil and Gas Industry, European Federation of Corrosion Publication no. 13 (London, U.K.: Institute of Materials, 1994), p 1. J.L. Crolet, N. Thevenot, A. Dugstad, “Role of Free Acetic Acid on the CO2 Corrosion of Steels,” CORROSION/99, paper no. 24 (Houston, TX: NACE, 1999). S. Nes˘ic´, J.L. Crolet, D.M. Drazic, “Electrochemical Properties of Iron Dissolution in the Presence of CO2—Basic Revisited,” CORROSION/96, paper no. 3 (Houston, TX: NACE, 1996). B.F.M. Pots, “Mechanistic Models for the Prediction of CO2 Corrosion Rates under Multi-Phase Flow Conditions,” CORROSION/95, paper no. 137 (Houston, TX: NACE, 1995). S. Nes˘ic´, R. Nyborg, A. Stangeland, “Mechanistic Modeling for CO2 Corrosion with Protective Iron Carbonate Films,” CORROSION/2001, paper no. 40 (Houston, TX: NACE, 2001). K. Videm, A. Dugstad, “Galvanic Influence of CO2 Corrosion,” CORROSION/89, paper no. 468 (Houston, TX: NACE, 1989). K. Videm, A.M. Koren, Corrosion 49, 9 (1993): p. 746-754. J.L. Crolet, N. Thevenot, S. Nes˘ic´, “Role of Conductive Corrosion Products on the Protectiveness of Corrosion Layers,” CORROSION/96, paper no. 4 (Houston, TX: NACE, 1996). A. Ikeda, S. Mukai, M. Ueda, “Prevention of CO2 Corrosion of Line Pipe and Oil Country Tubular Goods,” CORROSION/84, paper no. 289 (Houston, TX: NACE, 1984). J.L. Crolet, “Corrosion in Oil and Gas Production,” in Corrosion and Anticorrosion, eds. G. Beranger, H. Mazille (Paris, : Hermes Science, 2002).

681


28. K. Videm, A. Dugstad, “Effect of Flow Rate, pH, Fe2+ Concentration, and Steel Quality on the CO2 Corrosion of Carbon Steels,” CORROSION/87, paper no. 42 (Houston, TX: NACE, 1987). 29. K. Videm, A. Dugstad, Mater. Perform. 3 (1989): p. 63-67. 30. K. Videm, A. Dugstad, L. Lunde, “Influence of Alloying Elements upon the CO2 Corrosion Rate of Low-Alloyed Carbon Steels,” CORROSION/91, paper no. 473 (Houston, TX: NACE, 1991). 31. A. Dugstad, “The Importance of FeCO3 Supersaturation on the CO2 Corrosion of Carbon Steels,” CORROSION/92, paper no. 14 (Houston, TX: NACE, 1992). 32. M. Ueda, A. Ikeda, “Effect of Microstructure and Cr Content in Steel on CO2 Corrosion,” CORROSION/96, paper no. 13 (Houston, TX: NACE, 1996). 33. R. Nyborg, A. Dugstad, P. Dronen, “Effect of Chromium on Mesa Corrosion Attack of Carbon Steel,” Eurocorr 1997, Conf. of European Federation of Corrosion, held Sept. 22-25 (London, U.K.: Institute of Materials, 1997). 34. B. Mishra, S. Al-Hassan, D.L. Olson, M.M. Salama, Corrosion 53, 11 (1997): p. 852-859. 35. K. Videm, A. Dugstad, Mater. Perform. 4 (1989): p. 46-50. 36. J.L. Crolet, M.R. Bonis, SPE Prod. Eng. 11 (1991): p. 449-453. 37. B. Mishra, D.L. Olson, S. Al-Hassan, M.M. Salama, “Physical Characteristics of Iron Carbonate Scale Formation in Line Pipe Steels,” CORROSION/92, paper no. 13 (Houston, TX: NACE, 1992). 38. K. Videm, J. Kvarekvaal, T. Perez, G. Fitzsimons, “Surface Effects on the Electrochemistry of Iron and Carbon Steel Electrodes in Aqueous CO2 Solutions,” CORROSION/98, paper no. 1 (Houston, TX: NACE, 1998). 39. B. Kaasa, T. Ostvold, “Prediction of pH and Mineral Scaling in Waters with Varying Ionic Strength Containing CO2 and H2S for 0 <200 and 1

<500,” CORROSION/98, paper no. 62 (Houston, TX: NACE, 1998). 40. J.L. Crolet, M. Pourbaix, A. Pourbaix, “The Role of Trace Amount of Oxygen on the Corrosivity of H2S Media,” CORROSION/91, paper no. 22 (Houston, TX: NACE, 1991). 41. A. Dugstad, “Formation of Protective Corrosion Films during CO2 Corrosion of Carbon Steel,” Eurocorr 1997, Conf. of European Federation of Corrosion, held Sept. 22-25 (London, U.K.: Institute of Materials, 1997). 42. S. Al Hassan, B. Mishra, D.L. Olson, M.M. Salama, Corrosion 54, 6 (1998). 43. E. Gulbrandsen, R. Nyborg, “Effect of Steel Microstructure and Composition on Inhibition of CO2 Corrosion,” CORROSION/ 2000, paper no. 23 (Houston, TX: NACE, 2000). 44. M.B. Kermani, J.C. Gonzales, C. Linne, M. Dougan, R. Cochrane, “Development of Low-Carbon Cr-Mo Steels with Exceptional Corrosion Resistance for Oilfield Applications,” CORROSION/2001, paper no. 01065 (Houston, TX: NACE, 2001). 45. R.M. Grau, “Mechanisms for Improved Corrosion Resistance of Steels in Carbonic Acid Environments” (PhD thesis, University of Leeds, 2000). 46. M.L. Johnson, M.B. Tomson, “Ferrous Carbonate Precipitation Kinetics and its Impact on CO2 Corrosion,” CORROSION/91, paper no. 268 (Houston, TX: NACE, 1991). 47. K. Videm, A. Dugstad, L. Lunde, “Parametric Study of CO2 Corrosion of Carbon Steel,” CORROSION/94, paper no. 14 (Houston, TX: NACE, 1994). 48. J.L. Crolet, M.R. Bonis, MP 24, 5 (1984): p. 35-42. 49. S. Nes˘ic´, J. Postlethwaite, “Modeling of CO2 Corrosion Mechanisms,” in Advanced Research Workshop on Modeling Aqueous Corrosion: From Individual Pits to System Management, Proc. of the NATO Advanced Research Workshop, eds. K.R. Threthewey, P.R. Roberge, held Sept. 6-8 (Kluwer Academic Publishers, 1994). 50. J.R. Shadley, S.A. Shirazi, E. Dayalan, E.F. Rybicki, F.D. de Moraes, “CO2 Corrosion Prediction in Pipe Flow under FeCO3 Scale Forming Conditions,” CORROSION/98, paper no. 51 (Houston, TX: NACE, 1998). 51. K. Videm, J. Kvarekvaal, Corrosion 51, 4 (1995): p. 260-269. 52. A. Ikeda, M. Ueda, S. Mukai, “Influence of Environmental Factors on Corrosion, Advances in CO2 Corrosion,” vol. 2, Proc. CORROSION/84 and CORROSION/85 Symp. on CO2 Corrosion in the Oil and Gas Industry (Houston, TX: NACE, 1985). 53. Y. Garsany, D. Pletcher, B. Hedges,”The Role of Acetate in CO2 Corrosion of Carbon Steel: Has the Chemistry Been Forgotten?,” CORROSION/2002, paper no. 02273 (Houston, TX: NACE, 2002).

682

54. J.D. Garber, K.A. Sangita, “Factors Affecting Iron Carbonate Scale in Gas Condensate Wells Containing CO2,” CORROSION/98, paper no. 19 (Houston, TX: NACE, 1998). 55. J.L. Crolet, M.R. Bonis, “The Role of Acetate Ions in CO2 Corrosion,” CORROSION/83, paper no. 160 (Houston, TX: NACE, 1983). 56. J.L. Crolet, S. Olsen, W. Wilhelmsen, “Influence of a Layer of Indissolved Cementite on the Rate of the CO2 Corrosion of Carbon Steel,” CORROSION/94, paper no. 4 (Houston, TX: NACE, 1994). 57. B.F.M. Pots, E.L.J.A. Hendriksen, “CO2 Corrosion under Scaling Conditions—The Special Case of Top-of-the-Line Corrosion in Wet Gas Pipelines,” CORROSION/2000, paper no. 31 (Houston, TX: NACE, 2000). 58. J.L. Crolet, consultant, correspondence to M.B. Kermani (August through November 2002). 59. U. Lotz, L. van Bodengom, C. Ouwehand, “Effect of Oil or Gas Condensate on Carbonic Acid Corrosion,” CORROSION/90, paper no. 41 (Houston, TX: NACE, 1990). 60. A.S. Green, B.V. Johnson, H.J. Choi, “Flow-Related Corrosion in Large Diameter Multiphase Flowlines,” 65th SPE Annual Technical Conf. and Exhibition of the Society of Petroleum Engineers, paper no. SPE 20685 (Richardson, TX: Society of Petroleum Engineers [SPE], 1990). 61. K.D. Efird, R.J. Jasinski, Corrosion 45, 2 (1989): p. 165-171. 62. U. Lotz, L. Van Bodegom, C. Ouwehand, Corrosion 47, 8 (1991): p. 635-645. 63. H.J. Choi, T. Tonsuwannarat, “Unique Roles of Hydrocarbons in Flow-Induced Sweet Corrosion of X-52 Carbon Steel in Wet Gas Condensate Producing Wells,” CORROSION/2002, paper no. 02559 (Houston, TX: NACE, 2002). 64. M. Castillo, H. Rincon, S. Duplat, J. Vera, E. Baron, “Protective Properties of Crude Oils in CO2 and H2S Corrosion,” CORROSION/2000, paper no. 5 (Houston, TX: NACE, 2000). 65. C. de Waard, L. Smith, B. Craig, “Influence of Crude Oil on Well Tubing Corrosion Rates,” Eurocorr 2001, Conf. of European Federation of Corrosion (London, U.K.: Institute of Materials, 2001). 66. J. Smart, “Wetability—A Major Factor in Oil and Gas System Corrosion,” CORROSION/93, paper no. 70 (Houston, TX: NACE, 1993). 67. B. Hedges, D. Paisley, R. Woollam, “The Corrosion Inhibitor Availability Model,” CORROSION/2000, paper no. 34 (Houston, TX: NACE, 2000). 68. American Petroleum Institute, Recommended Practice 14E (API RP 14E), “Recommended Practice for Design and Installation of Offshore Production Platform Piping Systems” (Washington DC: API, 1981). 69. A. Dugstad, L. Lunde, K. Videm, “Influence of Alloying Elements upon the CO2 Corrosion Rate of Low-Alloyed Carbon Steels,” CORROSION/91, paper no. 473 (Houston, TX: NACE, 1991). 70. E.W.J van Hunnik, B.F.M Pots, E.L.J.A. Hendriksen, “The Formation of Protective FeCO3 Corrosion Product Layers in CO2 Corrosion,” CORROSION/96, paper no. 6 (Houston, TX: NACE, 1996). 71. T. Rogne, T.G. Eggen, U. Steinsmo, “Corrosion of C-Mn-Steel and 0.5% Cr Steel in Flowing CO2-Saturated Brines,” CORROSION/96, paper no. 33 (Houston, TX: NACE, 1996). 72. A. Dugstad, “Mechanism of Protective Film Formation during CO2 Corrosion of Carbon Steel,” CORROSION/98, paper no. 31 (Houston, TX: NACE, 1998). 73. N. Blackburn, “Downhole Materials Selection for Clyde Production Wells: Theory and Practice,” Society of Petroleum Engineers, European Production Operations Conf. and Exhibition, paper no. SPE 27604, held March 15-17 (Richardson, TX: SPE, 1993). 74. A.J. McMahon, Mater. Perform. 36, 12 (1997): p. 50-53. 75. P.I. Nice, H. Takabe, M. Ueda, “The Development and Implementation of a New Alloyed Steel for Oil and Gas Production Wells,” CORROSION/2000, paper no. 154 (Houston, TX: NACE, 2000). 76. P.I. Nice, M. Ueda, “The Effect of Microstructure and Chromium Alloying Content to the Corrosion Resistance of LowAlloy Steel Well Tubing in Seawater Injection Service,” CORROSION/98, paper no. 3 (Houston, TX: NACE, 1998). 77. A. Dugstad, H. Hemmer, M. Seiersten, “Effect of Steel Microstructure upon Corrosion Rate and Protective Iron Carbonate Film Formation,” CORROSION/2000, paper no. 23 (Houston, TX: NACE, 2000).



78. R. Cochrane, Leeds University, work in progress. 79. D.E. Cross, “Mesa-Type CO2 Corrosion and its Control,” CORROSION/93, paper no. 118 (Houston, TX: NACE, 1993). 80. M.B. Kermani, J.C. Gonzales, G.L. Turconi, D. Edmonds, G. Dicken, L. Scoppio, “Development of Superior Corrosion Resistance 3%Cr Steels for Downhole Applications,” CORROSION/2003, paper no. 03116 (Houston, TX: NACE, 2003). 81. Petroleum and Natural Gas Industries—Materials for Use in H2S-Containing Environments in Oil and Gas Production— Part 1, ISO 15156 (Geneva, Switzerland, International Organization for Standardization [ISO], 2002). 82. J.L. Crolet, S. Olsen, W. Wilhelmsen, “Observation of Multiple Steady States in the CO2 Corrosion of Carbon Steel,” CORROSION/95, paper no. 188 (Houston, TX: NACE, 1995). 83. K. Videm, A. Dugstad, “Film-Covered Corrosion, Film Breakdown, and Pitting Attack of Carbon Steels in Aqueous CO2 Environments,” CORROSION/90, paper no. 186 (Houston, TX: NACE, 1990). 84. S.H. Drissi, J.P. Mike, J.M. Genin, Corros. Sci. (1996): p. 623. 85. C.A. Palacios, J.R. Shadley, Corrosion 47, 2 (1991). 86. J.K. Heuer, J.F. Stubbins, Corrosion 54, 7 (1998): p. 566-577. 87. S. Rajappa, R. Zhang, M. Gopal, “Modeling the Diffusion Effects through the Iron Carbonate Layer in the Carbon Dioxide Corrosion of Carbon Steel,” CORROSION/98, paper no. 16 (Houston, TX: NACE, 1998). 88. Y.J.T. Kinsella, S. Bailey, Corrosion 54, 10 (1998): p. 835-842. 89. J.R. Shadley, S.A. Shirazi, E. Dayalan, E.F. Rybicki, G. Vani, “Modeling CO2 Corrosion of Carbon Steels in Pipe Flow,” CORROSION/95, paper no. 118 (Houston, TX: NACE, 1995). 90. D.W. Shannon, “Role of Chemical Components in Geothermal Brine on Corrosion,” CORROSION/78, paper no. 57 (Houston, TX: NACE, 1978). 91. C. de Waard, U. Lotz, A. Dugstad, “Influence of Liquid Flow Velocity on CO2 Corrosion: A Semi-Empirical Model,” CORROSION/1995, paper no. 128 (Houston, TX: NACE, 1995). 92. F.D. de Moraes, J.R. Shadley, J. Chen, E.F. Rybicki, “Characterization of CO2 Corrosion Product Scales Related to Environmental Conditions,” CORROSION/2000, paper no. 30 (Houston, TX: NACE, 2000). 93. Metals Handbook, “Corrosion,” 9th ed., vol. 13 (Materials Park, OH: ASM International, 1987), p. 1,233. 94. E.C. Greco, W.B. Wright, Corrosion 18 (1962): p. 119t. 95. P.R. Rhodes, Electrochemical Society Extended Abstracts, vol. 76, no. 2 (Pennington, NJ: Electrochemical Society, 1976), p. 300. 96. J.L. Crolet, “Protectiveness of Corrosion Layers,” in Modeling Aqueous Corrosion: From Individual Pits to System Manage-

CORROSION Author's Guide* Vol. 56, No. 1

Contents General ................................... 1 Submission of Manuscripts ...... 1 Acceptance Criteria ................. 1 Acceptable Forms of Articles .... 2

THE JOURNAL OF SCIENCE AND ENGINEERING

General Style Requirements for All NACE Publications ........ 3

Influence of Overaging Treatment on Localized Corrosion of Al 6056 V. Guillaumin and G. Mankowski

Anticorrosion, Antiscale Coatings Obtained on the Surface of Titanium Alloys by Microarc Oxidation Method and Used in Seawater S.V. Gnedenkov, P.S. Gordienko, S.L. Sinebrukhov, O.A. Khrisanphova, and T.M. Skorobogatova

Abbreviations, Acronyms, and Symbols ........................... 3

Prediction of Stress Corrosion Cracking Susceptibility of Stainless Steels Based on Reivation Kinetics H.S. Kwon, E.A. Cho, and K.A. Yeom

In-Situ Imaging of Chloride Ions at the Metal/Solution Interface by Scanning Combination Microelectrodes C.-J. Lin, R.-G. Du, and T. Nguyen

Galvanostatic Pulse Measurements of ive and Active Reinforcing Steel in Concrete

Equations and Footnotes...... 4-5

D.W. Law, S.G. Millard, and J.H. Bungey

Variation of Slow Strain Rate Test Fracture Mode of Type 304L Stainless Steel in 288°C Water N. Saito, Y. Tsuchiya, F. Kano, and N. Tanaka

Corrosion Behavior of High-Purity Fe-Cr-Ni Alloys in the Transive Condition M. Mayuzumi, J. Ohta, and K. Kako

Graphics ................................. 5

CORROSION ENGINEERING

80 90

100.

101. 102.

103.

104.

105.

106. 107. 108.

109. 110.

111.

112.

A

n updated, free guide for the prospective CORROSION author, with tips on manuscript preparation, format, style, color artwork, and editorial policies.

Publication Procedures ............ 2

Founded 1945

Path Dependence of the Potential-Current Density State for Cathodically Polarized Steel in Seawater W.H. Hartt and S. Chen

70

99.

Manuscript Format ................. 2

CORROSION SCIENCE

12 24 32 41 48 57

98.

January 2000

CORROSION 3

97.

ment, eds. K.R. Trethewey, P.R. Roberge, NATO ASI Series, Series E: Applied Sciences, vol. 266 (Dordrecht NL, Kluwer Academic Publishers, 1994), p. 1-28. R. Nyborg, A. Dugstad, “Mesa Corrosion Attack in Carbon Steel and 0.5% Chromium Steel,” CORROSION/1998, paper no. 29 (Houston, TX: NACE, 1998). J.R. Shadley, S.A. Shirazi, E. Dalayan, M. Ismail, E.F. Rybicki, Corrosion 52, 9 (1996). A. Morshed, “Surface Modification for Corrosion Protection of Steel Pipes” (PhD thesis, University College London, 2002). R. Nyborg, “Overview of CO2 Corrosion Models for Wells and Pipelines,” CORROSION/2002, paper no. 02233 (Houston, TX: NACE, 2002). C. de Waard, U. Lotz, D.E. Milliams, Corrosion 47, 12 (1991): p. 976-985. C.S. Fang, J.D. Garber, R. Perkins, J.R. Reinhardt, “Computer Model of a Gas Condensate Well Containing Carbon Dioxide,” CORROSION/89, paper no. 465 (Houston, TX: NACE, 1989). Y.M. Gunalton, “Combining Research and Field Data for Corrosion Rate Prediction,” CORROSION/94, paper no. 14 (Houston, TX: NACE, 1994). G. Liu, R.C. Erbar, “Detailed Simulation of Gas Well Downhole Corrosion in Carbon Steel Tubulars,” CORROSION/90, paper no. 30 (Houston, TX: NACE, 1990). M.R. Bonis, J.L. Crolet, “Basics of the Prediction of the Risks of CO2 Corrosion in Oil and Gas,” CORROSION/89, paper no. 466 (Houston, TX: NACE, 1989). Condensate Well Corrosion, NGAA, Tulsa, OK, 1953. NACE MR0175, “Sulfide Stress Cracking Resistant Metallic Materials for Oilfield Equipment” (Houston, TX: NACE, 2001). Guidelines on Materials Requirements for Carbon and LowAlloy Steels for H2S-Containing Environments in Oil and Gas Production, Publication no. 16 (London, U.K.: The Institute of Materials, European Federation of Corrosion, 1995). K. Mudge, C.J. Levesque, Oil Gas J. 27 (1982): p. 245. J.B. Bradburn, “Water Production: An Index to Corrosion,” NACE South Central Region Conf. (Houston, TX: NACE, 1977). J. Kolt, E. Buck, D.D. Erickson, M. Achour, “Corrosion Prediction and Design Considerations for Internal Corrosion in Continuously Inhibited Wet Gas Pipelines,” U.K. Corrosion Conf. 1990, vol. 3 (London, U.K.: Institute of Materials, 1990), p. 217. C. de Waard, U. Lotz, “Prediction of CO2 Corrosion of Carbon Steel,” CORROSION/93, paper no. 69 (Houston, TX: NACE, 1993).

Effect of Biomineralized Manganese on the Corrosion Behavior of C1008 Mild Steel B.H. Olesen, P.H. Nielsen, and Z. Lewandowski

An Electrochemical Approach to Predicting Long-Term Localized Corrosion of Corrosion-Resistant High-Level Waste Container Materials

Metric Measurements .............. 6

THE JOURNAL OF SCIENCE AND ENGINEERING

CORROSION

D.S. Dunn, G.A. Cragnolino, and N. Sridhar

Numbers and Punctuation ... 6-7 References ............................... 7 *New Guidelines Effective Jan. 1, 2000. Item # 32143

Trade Names and Author Affiliations ............ 9 Index ..................................... 12 Author Checklist .................... 14


To order, : NACE hip Services Department, 1440 South Creek Drive, Houston, TX 77084-4906; Phone: 281/228-6223 or Fax: 281/228-6329; ask for Item no. 32143. For immediate delivery by fax, call 1-800-327-3134. This information is also available on the NACE Web page at www.nace.org.

683

Carbon Dioxide Corrosion In Oil And Gas Production 3w5b

Overview 26281t

More details 6y5l6z

Related Documents 3h463d

Carbon Dioxide Corrosion In Oil And Gas Production 3w5b

Corrosion In Oil And Gas Industry Ppt 155u6a

Corrosion And Its Mitigation In The Oil And Gas Industry.pdf x4m4o

Oil And Gas Production Handbook Ed3x0_web_part11 123f31

Carbon Dioxide Scrubber 1o5u6u

More Documents from "Josemar Costa" 2oj1j

Carbon Dioxide Corrosion In Oil And Gas Production 3w5b

4p345f

Nor Ma Iso 5807 2k4y70

4p345f

Ge-9e Maintenance Manual 241y4g

Catalogo Bosch Bomba Injetora Pecas Diesel 4e24g