ABSTRACT Permeability of methane, carbon dioxide and water in plasticized polyvinylidene fluoride (PVDF) and plasticized polyarnid 11 (PA11 ) has been measured for a number of temperatures and pressures in a small scale test apparatus and permeability coefficients have been calculated. The results have been used to predict if the annulus of flexible pipes will be water wet. For verification of the small scale test, a large scale test has also been carried out in a 50 mm flexible pipe. Both test methods show that the annulus of flexible pipes will be water wet when carrying gas and water. This implies that the conditions in the annulus will be corrosive when pipes are carrying gas which contains carbon dioxide or hydrogen sulphide. The corrosive conditions and corrosion fatigue must be taken into account when the fatigue life of flexible risers is calculated.
INTRODUCTION Flexible pipes are widely used as flowlines and risers for offshore oil and gas production. Its structure consists of concentric extruded polymer and reinforcing helical metallic layers as shown in Figure 1. The steel carcass prevents collapse in case of pressure drop in the flexible pipe and it resists the external hydrostatic pressure. The inner thermoplastic sheath is the pressure barrier.
Outside the pressure sheath there are several layers with armor wires that shall sustain the internal pressure and axial loads. The external thermoplastic sheath shall prevent water ingress in the structure.
Gases such as methane, carbon dioxide, hydrogen sulfide and water vapor permeate the inner thermoplastic sheath and flow into the annulus with armor wires. A corrosive annulus environment means that the armor wires can be subject to general corrosion, sulfide stress cracking and corrosion fatigue. Flexible risers are exposed to heavy dynamic loads, and fatigue calculations constitute an essential part of life time calculations for flexible risers.
A test program has therefore been carried out to investigate the corrosion fatigue performance of armor steel in de-oxygenated water containing carbon dioxide. The test was designed to simulate the case when the annulus is water wet. The results showed a significant reduction in cycles to failure (life time) in the carbon dioxide containing environments compared to air. Furthermore, the lifetime was significantly shorter in the environment with 0.1 bar carbon dioxide compared to 0.01 bar carbon dioxide, indicating that the tendency to corrosion fatigue increases as the environmental corrosivity increases.
Another test program was therefore started to determine the permeation coefficients for these gases with the purpose to investigate if corrosive conditions will be established in the annulus. The test program has included both small scale testing and testing on a 50 mm flexible pipe for verification of the small scale tests.
ABSTRACT WC based coatings with high alloy binders were investigated with respect to structure, corrosion and wear. The coatings were made by HVOF spraying of different powders. All powders studied were made by agglomeration/sintering, i.e. agglomeration of metal particles with WC particles with subsequent sintering. Some powders were made using pre-alloyed metal particles. A blend of ceramic- metallic powder and pure metallic powder was also studied. Different methods were used for characterisation of the powders and coatings.
INTRODUCTION WC based coatings are well known for their wear resistance and are often applied on surfaces to improve the wear properties. Numerous publications concerning pure wear properties like erosion and abrasion of WC coatings are available. Less attention has been paid to corrosion and synergy between corrosion and wear. The corrosion properties of WC coatings are of major importance if these coatings are used in corrosive environments. Corrosion of the binder will undermine the WC particles, resulting in fall-outs of particles or are they are pulled out by the erosive media. Typical examples of WC coatings with poor corrosion resistance are those with binders of pure cobalt or pure nickel.
The corrosion properties can be improved by addition of corrosion restricting elements like chromium and molybdenum. When selecting a coating material the corrosion properties of the coating should be comparable to that of surrounding materials. Different corrosion properties will give galvanic actions between the coating and the surrounding materials. The binder will act either a cathode or an anode compared to the surrounding material depending on the nobleness and corrosion properties. Typical examples are WC-Co applied in carbon steel and stainless steel systems. WC-Co applied in a carbon steel system will act as a cathode because the carbon steel is less noble than the Co binder. However, applied in a stainless steel system the steel will act as a cathode and the WC-Co as an anode, which increases the corrosion rate of the WC-Co binder. Hence, the corrosion properties of the coating are crucial for the expected lifetime.
ABSTRACT An investigation of accelerated atmospheric corrosion of copper by using a four-point- probe method and the Scanning Electron Microscope (SEM) is described An aggressive atmosphere containing 100ppm of H2Sand a relative humidity (RH) above MYko was used to accelerate the corrosion process under a controlled environment. The test consisted of two container testing, in triplicate, 25~m thick copper wire. The inhibited samples showed less increases in the measured voltage than the specimens not inhibited. The decrease in the voltage could be correlated to the corrosion rate and the efficacy of the VCI. The SEM images revealed different surface changes in the oxide formation on the protected and non- protected copper surface. We expect that this test presents a possible alternative to existing test methods, i.e. Japanese and German based on the commonly used Federal Test Method 4031, 101C a.k.a. Vapor Inhibiting Ability of VCI Vapors (VIA). Both the VIA test and this test are quite simple to conduct, ?however, quantitatively which is not possible with the VIA.
Volatile Corrosion Inhibitor (VCIS) are increasingly gaining more and more acceptance in industrial and consumer applications. Uses are widespread from water treatment to oil and gas line production to packaging, electronics, artifact procurement and a host of other uses. In recent years there have been several studies aimed at revising and updating the methods to test for the efficacy of VCIS and their detection of them on a surface.1%Success has been quite good for the detection of VCIS and also measuring the efficacy of them in harsh environments. l? 214There is though a desire for a simpler way to accomplish the same goal, Industrial users do not always have the sophisticated equipment and the facilities like a research laboratory to determine if there is a VCI on the surface or even to determine if a host of a VCI is efficient as a VCI carrier. For most companies the only method to test a VCI, i.e. crystals, liquids, or paper/polyethylene or other carriers of VCIS was to use a test method developed over 15 years ago called the Federal Test Method 4031, 101C a.k.a. Vapor Inhibiting Ability of VCI Vapors (via)? or similar methods which follow the same procedure.
The test method has come under some criticism due to the procedure of testing and due to the evaluation guidelines. They have been criticized because of it?s poor reproducibility and inconsistency of results. Here area few of the criticisms of the test method: There are inherent problems associated with this method such as the condensation portion of the test. In this portion of the test, Section 126.96.36.199, states, ...Cold water at a temperature 40 F below ambient shall be added to the aluminum tubes until full. After 3 hours, the water shall be removed from the steel tubes and the steel specimen evaluated for evidence of rust... The condensation part of this testis in part erroneous due to the very fast thermal dynamics changes experienced in the metallic substrate. The shock value to the specimen is difficult to reproduce and can sometimes change depending on the metal specimen preparation. The test method also does not fully outline polishing method which properly cleans the surface of the samples.
Alodan and Smyrl10have described a very thorough method for cleaning the surface of aluminum, which can also be applied towards other metals, after polishing using an ultrasonic cleaner. This method addresses cleaning out the embedded particles within the peaks and valleys experienced during polishing. The VIAmethod only recommends wiping with surgical cloth and dipping the part into hot mineral spirits followed with a boiling methanol dip. This method can be dangerous as
ABSTRACT In the arid or semi-arid western part of the United States, there are special parameters that may require investigation to determine whether the right-of-way soils are corrosive to cementitious-coated pipe. Cyclical wetting and drying, especially in arid environments, can cause an increase in chloride concentration in the capillary system of mortar by a factor of at least five. Under the right circumstances, with cyclical wetting and drying, a soil chloride content of 140 mgkg will exceed the corrosion threshold of 700 mg/kg and initiate corrosion on prestress wire. Several actual field examples encountered in an investigation on a PCCP failure are described. Also included is a simple field test to quickly and economically determine which combinations of gravel, sand, and salty clay that are helpful in predicting both the likelihood and severity of cyclical wetting and drying stimulated corrosion on prestress concrete cylinder pipe. Other parameters examined were mortar properties of thickness, compressive strength, absorption, density, pH and total alkalinity vs. chloride penetration into the mortar, and degree of corrosion on prestress concrete cylinder pipe.
SOIL RESISTIVITY The most common method employed to determine if a right-of-way (ROW) is potentially corrosive to prestressed concrete cylinder pipe (PCCP) is a soil resistivity survey. This is always a good first step. As shown in Table 1, a soil resistivity survey is often classified into resistivity categories of severely corrosive, corrosive, moderately corrosive, and mildly corrosive soils. Sometimes overlooked is that these terms were borrowed from soil classifications pertaining to dielectrically-coated or bare ferrous pipe material such as steel or ductile iron. Without additional information, these resistivity ranges are not necessarily indicative of a propensity for corrosion on cementitious-coated pipe including PCCP. However, the soil resistivity~ range is a useful first step in determining what additional tests may be required to achieve the desired life expectancy of PCCP in the right-of-way under investigation.
ABSTRACT In certain regimes of atmospheric corrosion, the corrosion rate is limited not by electrochemical reactions but by the rate of mass transfer of pollutants. In these cases, amass transfer model that accounts for the transport of pollutants, such as a marine salt aerosol, provides a theoretical and predictive framework for assessing corrosivity severity. Such a model of the transport of a marine aerosol fairly near the ground and well within the planetary boundary layer was developed. The predicted aerosol concentration as a function of distance for 1500 m from a steady source was consistent with published data on steel corrosion and salinity rates near an ocean. Implications from the model regarding objects that are exposed to aerosol-containing wind include: (i) increasing wind speed increases the aerosol deposition rate and therefore the corrosion rate, (ii) objects that are in the lee of prevailing winds from an aerosol source will corrode faster than objects on the windward side of an aerosol source, and (iii) smaller objects can be expected to corrode faster because of a greater capture efficiency of salt aerosols.
INTRODUCTION Design factors such as proximity to a source of pollution, degree of wind exposure and an object?s size affect the rate of pollution deposition to an object because, in certain regimes of atmospheric corrosion, the corrosion rate is limited by the rate of mass transfer of pollutants and not by the rate of electrochemical reactions. In these cases, a mass transfer model that accounts for the transport of pollutants, such as a marine salt aerosol, provides a theoretical and predictive framework for assessing atmospheric corrosivity. A desire to understand the effects of distance from an aerosol source, object size, and wind speed and direction on the corrosivity of microenvironment prompted the development of a first-generation model of the transport and deposition of marine-type aerosols. The corrosive effect of salt aerosols that are carried by the wind from salt-water bodies is well recognized. In the Pacer Lime 1algorithm, locations that are within 4.5 km from a sea are given the highest of four corrosivity ratings without any other considerations. In a statistical study, atmospheric damage functions for four metals were developed that each included a term for the chloride deposition rate2. The chloride deposition rate was either measured by a salt candle or calculated as wet deposition from rainfall rates and average chloride concentration in precipitation. Also, the 1S0 atmospheric classification algorithm requires a chloride deposition rate from salt candle measurements as well as time-of-wetness and sulfur dioxide measurements3.
However, there are limitations to the above approaches for characterizing local corrosivities. The Pacer Lime approach does not take into-account graduations in the effects of salt deposition within 4.5 km from a coastline or in the vicinity beyond 4.5 km. Published wet candle measurements are scarce relative to time-of-wetness data which can be fairly readily obtained from meteorological data. The use of wet chloride deposition rates from precipitation data, which is fairly accessible, is limited because it is only a measure of aerosol concentration and not deposition rate which depends on other factors such as wind speed.
Previous work has revealed the relationship between the mass transfer characteristics of pollutants to corrosion severity to some degree. An atmospheric corrosion study found that wire samples corroded at nearly twice the rate as sheet samples in a manner consistent with the turbulent mass transfer of gaseous pollutants. In another atmospheric corrosion study, the effects of dry dep
ABSTRACT A novel downhole corrosion monitoring system was used to monitor corrosion rates, and verify corrosion inhibitor effectiveness in the production tubing of a CO2 flood in the Oklahoma panhandle. The monitoring system was placed in the front tubing joint immediately above the electrical submersible pump. This location was deemed most corrosive, and therefore requiring the highest inhibitor concentration, due to high CO2 partial pressure, the elevated temperature, and the extremely turbulent flow. Laboratory evaluations had indicated the approximate effective inhibitor concentration required to attain the desired target corrosion rate under similar environmental and turbulence conditions. The complex problem of translating laboratory flow (high speed autoclave test) to field conditions was attempted empirically using established correlation for the rotating cylinder and tubular flow.
INTRODUCTION The Pestle Field, situated in the Oklahoma Panhandle, north of the city of Guymon, has been in operation since 1958. The field produces form the Morrow Sand at a depth of 6,100 R and was placed on CO2 flood in 1995. The field differs from Mobil?s other CO flood operations in hat it is a sandstone formation, while most of Mobil?s West Texas C@ flood operations to date have been in limestones.
Initial corrosion control treatment for the Pestle followed the program which had been developed for the West Texas limestone floods with a weekly batch treatment into the annulus at a rate of 20 to 25 ppm for inhibitor D]). The treatment procedure was changed horn batch treatment (inhibitor with over flush) to continuous when C02 breakthrough occurred, but without changing the treatment rate.
Shortly after CO2 breakthrough occurred the field began to require more frequent well workovers. Severe corrosion on both internal and external surfaces of the production tubing was found with corrosion rates, in some instances being in excess of 300 mpy. Field personnel observed that high corrosion rates appeared to be associated only with wells where large quantities of CO2 would be produced though the corrosion pattern was complex and initially difficult to predict.
The Pestle CO2 fl@ being a sandstone formation, is inherently more corrosive than limestone floods mainly because of the low bicarbonate concentration (200ppm) in the produced water from sandstone formations (vs. 2000 ppm bicarbonate from limestone formations), and the attendant lower pH that results in the presence of CO2
ABSTRACT Constant strain rate tests were conducted on welded austenitic AISI 304 (UNS 30400) in boiling MgC12 solutions at concentrations stainless steel of 32°/0 and 37°/0 weight respectively. Potentiodynamic anodic polarization tests were carried out on unstressed welded coupons. Open circuit potential measurements on stressed specimens showed that weldments exhibited free corrosion behavior in the active-passive region. Residual ferrite phase in weldments was detected using experimental and analytical methods. The presence of residual delta ferrite in weldments is hypothesized to have been responsible for the formation of microgalvanic cells that lead to the selective dissolution of the ferrite phase in favor of the austenitic matrix. This situation provided suitable nucleation sites for cracking, and its simultaneous occurrence promoted crack propagation. Optical and electron microscopy investigations support this hypothesis and show that weldment interdendritic spaces suffered selective dissolution and that crack propagation followed a preferential path along delta ferrite rich interdendritic arms. It was also revealed by the optical and electron microscopy investigations that crack growth took place by the union of longitudinally aligned interdendritic arms. The results are in agreement with the Parkins model of crack nucleation by simultaneous action of passive film formation, film rupture.
INTRODUCTION Austenitic stainless steels are commonly joined by welding rods that contain from 5% to 10°/0residual &ferrite, which is of known benefit in halting hot fissuring. ?1) During initial stages of the joining process, and under high rates of heat loss, molten steel starts to solidify into dendritic &ferrite grains. With progressive heat loss, several changes take place. Dendritic grains grow and close up on each other creating a network of interdendritic spaces, where the remaining non-solidified metal is trapped. Further descent down the temperature scale leads the solidified grains into a solid state phase transformation form &ferrite to y-austenite. In this stage, interdendritic spaces host ferrite stabilizing elements rejected from the austenitic matrix, and become rich in retained &ferrite phase.
When the entire weldment solidifies, retained &ferrite that formed within interdendritic spaces shall not transform into austenite and shall remain trapped in these spaces. ?2) Being more active than austenite ferrite shall assume the anodic electrode in ?3) It is hypothesized that the presence of such galvanic case a galvanic couple forms. action between constituent phases of the weldment created a suitable situation for incubation of cracks.
Electrochemical investigation of the corrosion behavior of austenitic stainless steel weldments in boiling MgC12 solutions comprised of polarization and free corrosion, open circuit potential tests. Results of electrochemical tests showed that weldments under given environmental conditions exhibited a preferred free corrosion potential in the zone of passivity-activity.
Cracked specimens were subjected to optical and electron microscopy investigation, which showed that the pattern of crack initiation and propagation came in agreement with the proposed Parkins model for crack initiation by continuous action of passive film formation-dissolution. ?4)
ABSTRACT Although chloride is the main source for corrosion of reinforcing steel in coastal buildings, concrete carbonation leads to a uniform corrosion of the steel that would accelerate the crack formation and decrease the structure service life. Therefore, carbonation induced corrosion could be minimized by improving the concrete quality and cover thickness. The objective of this investigation was to study the effect of carbonation on public buildings located up to 800 m from the seashore. Preliminary results based on the analysis of carbonation, resistivity and porosity data suggested the need to increase the concrete cover thickness and concrete quality proportional to the distance from the seashore as well as to their elevation. In general, the higher carbonation coefficients corresponded to the top sections of the evaluated buildings where the measured relative humidity was lower. However, the concrete cracks due to corrosion were found in the lower sections where humidity was higher. Data from lab specimens exposed to the same environment buildings. ,
INTRODUCTION It is well known that in marine environments, chloride is the element with the strongest influence on reinforced concrete corrosion. However, acidification of concrete due to COZ can occur in places where the right climate conditions are available so that a uniform corrosion can develop. This uniform carbonation-induced corrosion accelerates the concrete crack formation and decreases the residual service life of the concrete structures. Both aggressive agents can interact in marine environments and lead to a much faster deterioration than if either one acts alone. In the North of the Yucatan Peninsula as in other places with tropical marine climate, the aggression of the atmosphere allows a combined action of chloride and carbon dioxide. Humidity and temperature conditions of this region2 promote the advance of the carbonation front3. The carbonation rate will depend on several factors4 such as the type and amount of cement, porosity of the concrete, type and quantity of pozzolanic additions,5 etc. Moreover, modifications in concrete properties as compressive strength, superficial hardness and resistance to aggressive agents (e.g. sulfates) are produced.
Chlorides or carbon dioxide can reach the reinforcement causing depassivation and shortening the propagation period? in which any rehabilitation action will be very expensive. From the engineering point of view, an alternative is to design and construct structures according to their geographic orientation, marine breeze direction, sources of humidity and insulation* and elevation and distance regarding the sea (micro-climate) among others. However, few works have been found with this approach?-l? and neither one correlate data from real structures with those from laboratory specimens exposed at the same microclimate.
In our case, a research to propose modifications to the design of concrete structures situated up to 800 m from the sea based on their susceptibility for corrosion initiation as a function of concrete quality and micro-climate conditions has been initiated. This paper discusses preliminary carbonation data obtained in columns from 10 buildings situated at different distances from the seashore. The results are compared with those from cylindrical specimens that were exposed to the same conditions for five years. 12-14
ABSTRACT This paper summarizes 1995 through 1998 laboratory, outdoor exposure facility, and field data on the subject concrete rehab system. The system shows promise as a means of providing cathodic protection to the reinforcing, as a chloride removal process, as a re-alkalization process, and/or as a lithium injection procedure to minimize alkali-aggregate reactions in the concrete. Unique characteristics of the system include: 1. Surrounding each galvanic anode with a highly corrosive liquid which maintains it (the anode) at peak output voltage throughout its life; and 2. Placing an ionic transfer layer between the anode and the concrete surface that is high volume, low resistivity and deliquescent (i.e. pulls water vapor out of the air at relative humidities of 35??ZOor higher). The ionic transfer layer typically consists of sponge, felt or sand loaded with calcium chloride (and/or other chemicals such as sodium hydroxide, potassium acetate, and lithium-salts). In some cases it also contains a wetting agent and is encapsulated (fully or partially) in vapor permeable, but water impermeable materials. The ionic transfer layer will not freeze at temperatures as low as -20 C (?5 ?F), and provides sufficient space for all anode corrosion products, thus preventing undesirable stresses on the concrete, the anode assembly and any cosmetic covering.
INTRODUCTION Cathodic protection is technically proven means of mitigating the corrosion of steel in waters, soils and moist, chloride contaminated, reinforced and prestressed concrete structures both above and below water and soil.l?2
Work on adapting galvanic (sacrificial) anodes to reinforced and prestressed concrete structures began in the mid- 1970s.3 Although such anodes could be made to work, problems with low current ~~put and accommodation of the anode corrosion products prevented widespread use of the technology .
The alternative to a galvanic anode system is an impressed current cathodic protection system in which power from an outside source is used in concert with low corrosion rate anodes. 1-8 This alternative overcomes the difficulties of low power output and accommodation of anode corrosion products. Many impressed current systems have been installed. However, they are somewhat complicated and generally require continuing monitoring.7?8
Present galvanic anode systems for atmospherically exposed concrete, such ass rayed zinc, will work in hot, moist environments, such as the Florida Keys and similar coastal areas.9 -* However, many are too low power for complete corrosion control on high corrosion rate, above-water and soil concrete structures in the central and northern portions of the United States, and on structures with multiple mats of reinforcing steel.
The causes of low power output area low voltage or potential difference between the sacrificial anode and the corroding steel in salty concrete, oftentimes only 0.5 volts or less. In addition, concrete, even when wet, has a higher resistivity (resistance per unit area) than most wet soils and natural waters, up to 100,000 ohm-cm or more; and significant contact resistance exists between many sacrificial anode systems and concrete. These factors create a large circuit (anode to cathode) resistance that results in low current output.
Temperature greatly affects concrete resistivity. The lower the temperature, the higher the resistivity and thus the lower the current output. The resistance is increased if the concrete layer immediately beneath the sacrificial anode dries out. Additionally, some sacrificial systems have the shortcoming of a finite life that often is less than the life of the structure involved. If the sa
ABSTRACT A series of erosion-corrosion (E-C) tests was carried out on the NiA1-A1203 interrnetallic-ceramic coatings deposited with high-velocity oxygen-fiel method (HVOF). The tests attempted to simulate the erosion conditions at the heat exchanger tubes in coal-fired boilers. The E-C behavior of these coatings was investigated and compared with other thermal sprayed coatings. It was found that comparing to other coatings, eroded by the bed ash at 300°C, the HVOF NiA1-40A1203 coating exhibited the lowest thickness loss at a 90° impact angle, and was the second best at a 30° impact angle. Eroded by the fly ash under test temperatures 450-600°C, the HVOF NiA1-40A1203 coating demonstrated the highest erosion-corrosion resistance at all impact angles of testing. At temperatures below 200°C, the E-C wastage of the HVOF NiA1-40A1203 coating had essentially no dependence on temperature. From 200C to 600°C the coating thickness loss increased and from 600 to 800°C the thickness loss decreased with temperature. The HVOF NiA1-40A1203 coating eroded by cracking and chipping brittle mechanism.
INTRODUCTION Some thermal sprayed coatings are used on structural steels in energy conversion and utilization systems to prevent surface degradation by elevated temperature erosion or combined erosion-corrosion. Many coatings are now undergoing testing in aggressive environments of boilers. An arc- sprayed amorphous Fe-Cr-B-Si coating (Armacor M) has been accepted by some plants as the preferred erosion protection in the refractory interface of fluidized bed combustion (FBC) boilersl?2. Recent achievements of the high velocity oxygen fbel spraying process (HVOF) have increased the demand for HVOF sprayed coatings. The HVOF process involves low temperatures and high velocities, thus producing high density and low oxidized coatings. The HVOF chromium carbide-based cermet coatings have shown superiority to arc-sprayed Fe-Cr-B-Si coatings in preventing the erosion3. However, while achieving the stable and rather high mechanical properties, the carbide cermet coatings are still sensitive to possible temperature fluctuations over 450C during service. The nickel aluminide intermetallic /aluminum oxide materials are very attractive because of their excellent high temperature corrosion resistance.
The purpose of this work was to understand the high temperature erosion-corrosion behavior of the NiA1-40A1203 intermetallic-ceramic coating on a mild steel. The effect of temperature on the coating erosion was investigated. The high temperature erosion behavior of this coating was compared with that of 1018, T-22 steels and other thermal sprayed coatings.