ABSTRACT The measurement and prediction of atmospheric corrosivity is an important element for determining the risk of corrosion damage to aircraft. Together with smart structure-type corrosion sensors, the prediction of atmospheric corrosivity is integral to an idealized aircraft corrosion surveillance strategy. A comprehensive methodology for predicting corrosivity based on atmospheric parameters, developed by ISO, has been utilized for the Canadian maritime Greenwood and inland Trenton bases. Evidence is presented that the important parameters affecting atmospheric corrosivity fluctuate significantly with time. In order to optimize corrosion control efforts, this time variability should be to be taken into account for procedures such as aircraft washing, the application of corrosion inhibitors, aircraft storage and dehumidification. The measurement of critical parameters for predicting the corrosion risk to aircraft requires simple practical procedures.
INTRODUCTION In a previous paper [l], the concept of corrosion surveillance for aging military aircraft was described. The detection, prediction and ultimately the prevention of corrosion damage are central themes in the corrosion surveillance philosophy. The three main application areas of aircraft corrosion surveillance are the reduction of unnecessary inspections, the optimization of maintenance and inspection schedules and materials performance evaluation under actual operating conditions. An idealized corrosion surveillance program for military aircraft can be formulated, as shown in Figure 1. Essentially the scheme presented in Figure 1 revolves around predicting where and when the risk of corrosion damage is greatest and to tailor corrosion control efforts accordingly. The principle and importance of linking selected maintenance and inspection schedules to the prevailing atmospheric corrosivity has been described in detail previously [1,2]. An underlying consideration in these recommendations is that military aircraft spend the vast majority of their lifetime on the ground and most corrosion damage occurs at ground level. In this paper, the focus is on the utilization of atmospheric parameters for predicting atmospheric corrosivity at ground level and the rationalization of this data for optimizing aircraft washing schedules and other corrosion control efforts (Figure 1). The work reported in this paper pertains mainly to the Canadian Forces (CF) Greenwood air base, where corrosion concerns are heightened due to close proximity to the sea (Figure 2).
THE DETERMINATION OF ATMOSPHERIC CORROSIVITY
Two fundamental approaches to atmospheric corrosivity classification, based on direct measurement and the use of atmospheric parameters respectively, have been proposed . These two approaches of environmental classification can be used in a complimentary manner, to derive relationships between atmospheric corrosion rates and the dominant atmospheric variables. Ultimately, the value of atmospheric corrosivity classifications is enhanced if they are linked to estimates of actual corrosion rates of different metals/alloys.
The major benefits of taking the atmospheric parameter route on an air base are that corrosivity predictions can, in principle, be made in short time frames and the required experimental
ABSTRACT It is shown that through careful experimental design and proper monitoring, assessment of the corrosion rate beneath thin-film electrolytes in industrial environments is possible. This paper discusses the various problems associated with performing corrosion rate measurements in these often times transient, thin film environments through the review of several industrial case studies where Linear Polarization Resistance (LPR) techniques were utilized to assist in monitoring corrosion.
INTRODUCTION Many industrial environments exist where the corrosive is only present as a thin aqueous film on the surface of the metal. Often the existence of these films are extremely dependent upon the process variables (time, temperature, products, etc.) and therefore, the prediction of corrosion is often difficult. In many cases the films are only present during short time periods of system upset or during certain portions of a batch process. Corrosion rates obtained from conventional weight-loss measurements in systems where the films are not continuously present often underestimate the rate unless the period for which the films are present is known and the rate calculations are based upon that exposure period.
Electrochemical techniques that make use of polarization resistance, galvanic coupling, AC impedance and electrochemical noise have been used successfully to correlate corrosion initiation and propagation to changes in environment that may be due to plant operations. Successful implementation of any of these techniques require that the monitoring probes be properly designed. Several probe design issues and measurement errors associated with monitoring in thin-film electrolytes are discussed.
Proper probe design is critical to accurately determining corrosion rates in thin film electrolytes. Geometric effects of thin films produce ohmic potential drop errors even in highly conductive (low resistivity) environments. These errors must be accounted for if accurate corrosion rates are to be determined. In addition, the ohmic resistance between the working, reference and counter electrode elements often results in non-uniform polarization, where only adjacent edges of electrodes are actually polarized. These issues force the probe design to incorporate relatively small, closely spaced elements to minimize potential drop errors while maximizing current throwing power and current distribution. Two different probe designs were utilized in the case studies discussed below. Probe design A, shown in Figure 1, consists of a circular center electrode disk surrounded by circular ring sleeve electrode. The electrodes have equal exposed surface areas and are separated by an appropriate dielectric for the environment in which the probes will be exposed. The probe is ground and polished to assure that the surfaces of the electrodes and dielectric are co-planar so no disruption of the thin film electrolyte occurs. This probe is installed flush with respect to the wall within the equipment being monitored.
Probe design B is shown in Figure 2. This probe consists of alternating rings of electrodes separated by an appropriate dielectric. The electrodes are of equal exposed surface areas and are separated equally by the dielectric, Several different materials can be include
ABSTRACT This paper discusses a test program to in-situ evaluate the comparative performance of six centrifugally cast heater tube alloys in an ethylene pyrolysis heater. The unusually high creep growth of conventional HP-40 radiant tubes leading to development of the test program is described, as well as the performance of the test tubes, as measured by post exposure creep growth, carburization behavior, microstructural analysis and time to failure. The performance of an aluminized tube in a separate pyrolysis heater is also described.
Background In the mid-1970?stwo ethylene units were brought on stream by a major petrochemical company. The heaters were designed using what was then, conventional, slow-residence-time (SRT) technology. Each heater contained six coil passes arranged in a vertical, serpentine pattern (Figure 1).The initial radiant tube metallurgy was centrifugally cast ACI HK-40, at that time the industry standard alloy for ethylene radiant tubes. All heaters cracked liquid feedstock.
In the 1980?sthe heaters were modified to allow cracking of gas feedstocks. To adapt to gas cracking, larger diameter outlet pass radiant tubes were installed. The larger diameter tubes permitted longer residence times and lower hydrocarbon partial pressures, altering the cracking selectivity and increasing the value of the cracked products. The outlet passes were retubed with alloy HP-40 modified with 10/0niobium. HP-40 was selected based on it being the primary alloy successor upgrade when tube-metal temperatures were expected to exceed those recommended for HK-40 alloy (-982°C/ 1800F).In addition to resistance to high temperature oxidation, stress rupture strength was historically the primary criteria for selecting materials for radiant section ethylene cracking service. The HP-40 lNb alloy was chosen specifically due to improved stress-rupture properties vs. HK-40 and equal or better mechanical properties compared with other modified HP class alloys.
Approximately nine months after installation, several HP-40 tubes in gas cracking service had already elongated from axial creep to the point that they had to be shortened and the U-bends reinstalled to prevent the tubes from interfering with the heater floor. This accelerated creep induced growth was considered highly unusual and initiated an extended effort 10learn its cause. The result of that study suggested that carburization and extended decoke times were significant contributors to the creep growth.(?) The study culminated in an effort to develop information on relative radiant tube performance in aggressive gas cracking service by installing different tube metallurgies in a designated test heater.
ABSTRACT To date, hydrogen flux monitoring devices have been only qualitatively used in the field to gauge environmental severity with respect to identifying changes from normal operating conditions, This paper discusses a methodology for extending the use of hydrogen flux monitoring equipment to quantitatively assess the severity of hydrogen charging and relate the response directly to the materials resistance or susceptibility to wet H2Scracking, This methodology incorporates the value of measured hydrogen flux, operating equipment variables, material of construction and its inherent susceptibility to cracking to determine the risk of cracking based on the severity of the operating environment. This methodology will allow more applicability of the hydrogen flux monitoring equipment to determine safety margins with respect to intermittent upset conditions or more prolonged changes in hydrogen charging severity resulting from changing production or process conditions,
INTRODUCTION The severity of hydrogen charging and the related problem of wet H2S cracking of steels is a problem of current interest. Atomic hydrogen, produced by the sulfide corrosion reaction, diffuses into the steel and can either (1) diffuse through the fill thickness and exit the O,D, surface or (2) recombine to form H2 at inclusions or other microstructural discontinuities leading to the formation of hydrogen blisters, hydrogen induced cracking (HIC) or stress oriented hydrogen induced cracking (SOHIC). Diffusion of atomic hydrogen can also lead to sulfide stress cracking (SSC) provided the combination of hardness and microstructure is susceptible, However, SSC can occur at substantially reduced levels of hydrogen charging compared to that required to produce blistering, HIC and/or SOHIC ?1>21.Hence hydrogen flux monitoring is rarely used to predict its occurrence. Generally, equipment operating in wet H2Sservice is constructed to minimize the occurrence of SSC.
Several methods have been employed to measure hydrogen flux in the field, These methods and applications were described by Kane,They are either intrusive or non intrusive, Intrusive probes require penetration of the equipment wall and the sensing element is placed in the environment, For the purposes of intrusive hydrogen flux monitoring, only the finger probe is currently available, Hydrogen flux is measured based on the pressure build up within the probe, surface area exposed to the environment and internal volume of the probe, There are several nonintrusive probes available for use including (1) patch or barnacle cells, (2) external foil patches equipped with pressure gages, (3) weld-o-lets equipped with pressure gages and (4) solid state electrochemical cells,
Hydrogen flux monitoring devices have been only qualitatively used in the field to gauge environmental severity with respect to identifying changes from normal operating conditions. There has been little quantitative work conducted that has related their measurements to cracking severity. The ability to more quantitatively use hydrogen charging data collected under field or plant conditions could allow for the direct assessment of the severity of service environments and the influence of operating conditions, system upsets and chemical treatments on cracking seventy.
This paper describes the methodology utilized to calculate a subsurface hydrogen concentration, CO, from in-service hydrogen flux data and estimate the hydrogen concentration gradient through-wall, This hydrogen concentration gradient is further compared to the critical hydrogen concentration, CC,, to determine safety margins, particularly with respect to upset conditions or more prolonged changes in hydrogen charg
ABSTRACT In a previous work, a relationship was established between a vacuum loss method and a conventional electrochemical technique for measuring hydrogen permeation through metals. In this paper, mathematical expressions are derived and used in order to quantitatively compare electrochemical transients typically obtained in the laboratory, with the vacuum loss data that could be obtained in the field. The effects of capillary tubing length, temperature and metal wall thickness on both the time and magnitude response are assessed. These expressions will allow to use the vacuum loss hydrogen patch probe as an on-line sulfide stress cracking or hydrogen induced cracking susceptibility monitor, based on correlation found in the laboratory using electrochemical techniques.
INTRODUCTION One of the methods commonly used in the field to monitor internal corrosion rates of pipelines or reactors, non intrusively, is known as the Hydrogen Permeation Method. The principles behind this method, advantages and disadvantages have been extensively discussed in the literature?. The use of this technique for quantifying internal corrosion rates can be considered a controversial issue, because the relationship between the corrosion rate and the hydrogen flux is not straight forward. This relationship depends not only on the corrosion kinetics, but also on the parameters controlling hydrogen transport through the materials and interfaces involved. Besides, them are many other techniques for measuring corrosion rates directly, which compete with the hydrogen permeation method.
However, the hydrogen permeation method has one competitive advantage over other techniques, when it is important to assess the susceptibility of materials to fail from hydrogen related phenomena, such as sulfide stress cracking (SSCC), hydrogen embrittlement (HE) or hydrogen induced cracking (HIC, SOHIC).
Many testing methods have been developed to assess in the laboratory the SSCC susceptibility of candidate materials for sour service. Most of these methods are included in standard procedures, such as NACE TMO177-90?. However, the need of such practices to be accelerated leads to the use of aggressive environments that do not represent in an adequate way the conditions commonly found in the oil field. Gn the other hand, the constant extension rate, CERT?, and the fracture mechanics approach using double cantilever beam specimens, DCB4 can assess the SSCC susceptibility in different testing solution and conditions. Therefore, they have the potential of been used in establishing environmental envelopes inside which a given steel will exhibit crackingS. However, none of these methods can be used on-line to determine in the field the susceptibility of steels to suffer SSCC.
If a correlation between SSCC susceptibility of a given material with hydrogen permeation is known, it may be possible to extrapolate laboratory results to other set of field conditions that induce similar hydrogen flux or concentration inside the metal. For example, a correlation has been reported between normalized strain to failure and the normalized hydrogen, showing a trend that is dependent on the environmental parameters involved in the test6,?. This relationship has been extended to other OCTG materials at different temperatures, obtaining a c
ABSTRACT To reduce the air pollution caused by fossil energy boilers, flue gas desulfurization units are more and more used in North America, European countries and Asean. The most common technology consists of scrubbing the polluted gas with a slurry of lime or limestone in water. In certain zones of scrubber (absorber) where the reaction between polluted gas and the solution is not complete, acidic condensation can occur and, combined with high temperatures, chlorides and/or fluorides, lead to very aggressive conditions.
Generally, metallic materials present the best solution in terms of reliability ad cost. Since the corrosion resistance of standard stainless steels, such as 316L, is very limited in such environments, highly alloyed stainless steels or nickel based alloys are generally used for the most corrosive conditions.
Building scrubber units require welded materials. Welded joints are made-up of different zones : thermal cycles induce structural modifications, filler materials induce chemical composition variations, and weld beads induce geometric variations. Welds are very often the weak point for the corrosion resistance.
To increase the corrosion resistance of the welds, new stainless steel materials with improved weldability have been developed proposing higher corrosion resistance properties (> 6 MO grades, NO8926 or S32050) or high corrosion resistance and high mechanical properties (duplex S3 1803/S32205 and superduplex S3255O/S32520). More recently, a high nitrogen overalloyed austenitic grade (S31266) providing very high corrosion resistance and high mechanical properties has been developed. This new grade with high nitrogen content (0.45% by weight) exhibits exceptional corrosion resistance properties in both unwelded and welded conditions.
Nickel based alloys have been also investigated both in solid and clad materials. The aim of this paper is to evaluate and compare the behaviour of these materials in simulated FGD environments, particularly in welded conditions. Several tests representative of industrial conditions have been selected. Test conditions simulating the very corrosive environments of gas cleaning systems : low pH, high temperature and high chloride levels have been investigated. The critical conditions have been determined for each material in unwelded and welded conditions. The results are discussed in terms of technical efficiency and potential applications.
INTRODUCTION Improving the quality of life and the environment by combining industrial progress and pollution control is obviously the most exciting challenge of our decade. We need progress, but without causing damage to our beautiful earth. Air, water and land have to be considered as the world?s capital received from our parents. Our responsibility is to prevent it from injury so that our children may also enjoy life as well as we have.
Pollution control has thus become a major concern of industrial countries. This has led to the development of pollution control equipment, which is becoming a substantial part of the recent investments in industrial countries. This results, for some industries, higher operating costs which in turn makes productivity a major concern.
Here, downtime means costly production losses. Material se
ABSTRACT Rubber linings have been applied as a corrosion protection measure for steel surfaces, particularly in the absorbers, in the flue gas desulphurization plants of a large number of power stations in Europe and have decidedly proven their effectiveness. The rubber linings applied consist of either precured and/or cold-curing rubber sheets. In the course of the past five to seven years, the eastern European states have also begun retro- fitting their existing power stations with flue gas desulphurization plants. As the first of its kind, a scrubber in the flue gas desulphurization plant of the Konin Power Station in Poland, which operates on the basis of the limestone-gypsum process, was constructed of concrete. In this case also, the corrosion protection measures implemented consisted in the application of a precured rubber lining on the basis of butyl rubber. A surface area measuring 1500m2 of the concrete absorber was protected by means of this corrosion protection system.
INTRODUCTION Linings are frequently applied as an industrial corrosion protection measure in a varied range of industrial plants. Such linings are defined as surface coatings consisting of molded materials, such as sheets, plating and pipes for example. The rubber lining system is one of the oldest forms of corrosion protection lining that has been utilized in apparatus engineering. It has found a broad spectrum of application in the most varied of industrial branches. Rubber linings are utilized as a protective coating against corrosion wherever acids, alkaline or saline solutions are being worked with.
In Europe, rubber linings have also been applied as a corrosion protection measure in flue gas desulphurization plants (FGD plants) with great success since the late 1970?s*. Up to now, the energy producing power plants in Europe have been fitted almost exclusively with flue gas desulphurization plants operating on the basis of the wet absorption process, and in particular the limestone-gypsum-process3. In these instances, rubber linings are applied particularly in all sections that are subjected to wet loads such as absorbers, clean gas ducts, vessels and pipelines. The structural material most widely utilized is carbon steel. For the first time however, a flue gas scrubber made of concrete as opposed to steel has now been erected. Here again, a rubber lining was selected as the corrosion protection system for the absorber.
FLUE GAS DESULPHURIZATION PLANTS IN POLAND
At the beginning of this decade, a number of former Eastern-bloc countries also began retro-fitting existing energy producing power plants with systems designed for the desulphurization of flue gases. In addition to the Czech Republic, Poland has also been at the forefront of this development. At the following power plant sites in
· Belchatow · Jaworzno III · Opole · Konin · eleven flue gas scrubbers are already constructed and operational. Additional plants are currently either under construction (in Polaniec for example) or in advanced stages of planning. In addition to coal, the use of lignite as fuel in the electric power stations of Poland is also quite widespread. Approximately 66 % of the 32.000 MW capacity yield of the power plants is won by coal and roughly 28 % by lignite.
ABSTRACT To facilitate the treatment of spent decontamination and chemical cleaning solutions, this study explored the feasibility of replacing stable complexing agents (such as EDTA) by complexing agents which can be readily decomposed using chemical or thermal treatment. Acetohydroxamic acid was found to be a promising candidate for such a role. Preliminary experiments carried out on actual corrosion products from steam generators indicated that 15% acetohydroxamic acid solutions had similar effectiveness to that of the EPRI/SGOG iron solvent (which contains 15% EDT A) with respect to certain magnetite-rich deposits. The addition of malonic acid and hydrazine enhanced the dissolution of magnetite in acetohydroxamic acid solution. EDT A and acetohydroxamic acid were similarly effective in the decontamination of stainless steel surfaces superficially contaminated with 60CO. It was found that acetohydroxamic acid in spent solvents can be completely decomposed through acid-catalyzed hydrolysis using nitric or hydrochloric acid. Complete decomposition can also be achieved through oxidation of acetohydroxamic acid with active manganese dioxide or potassium permanganate. Regeneration of spent, contaminated acetohydroxamic acid solutions was demonstrated using a cation exchange resin to remove ferric or ferrous ions ITom the solution. Cobalt ions were removed from such solutions using charcoal. Based on the experimental findings, acetohydroxamic acid solutions appear to be promising candidates for use in decontamination or chemical cleaning, followed by decomposition or regeneration of the spent solutions.
INTRODUCTION Many studies have been carried out on developing and improving processes for chemical cleaning of accumulated sludge and deposits in heat exchangers, boilers, and steam generators, as well as for decontamination of surfaces in the primary loops of nuclear plants and in medical, industrial and research facilities.1,2 Many different cleaning solvents have been formulated for use under widely different conditions. For instance, the EPRI/SGOG process has been developed for chemical cleaning in the secondary loop of nuclear power plants. This process is based on the use of high concentrations (5 - 20%) ofethylenediaminetetraacetic acid (EDTA).3 Another example is the Can¬Decon decontamination process, which is characterized by the use of chelating reagents such as EDT A and citric acid ligands) at concentrations of the order of 1 g/L.2 In general, most cleaning solutions are based on the use of a complexing or chelating agent, frequently EDT A, with some other additives such as pH controlling compounds, redox agents, surfactants, and abrasives.
The disposal of secondary wastes generated in decontamination or chemical cleaning operations is a serious problem/ because such wastes very often contain radioactive contaminants (e.g. 60CO) and hazardous species (e.g. Cr) in the presence of high concentrations of organic complexants such as EDT A, NT A, or citric acid. The solidification of the wastes in cement in the presence of organic complexants is difficult, and it was found that the presence of EDT A in burial sites can cause re-solubilization and re-mobilization of radioactive species.4 Various methods for the destruction of organic complexants have been investigated, but such destruction requires extreme conditions because the complexants commonly used have high chemical stability. On the other hand, the regeneration of complexants using ion exchange media is limited to dilute solutions of such complexants. The replacement of currently used, stable chelating agents, such as EDT A, by degradable chelants could greatly facilitate solidification of radioactive or hazardous contaminants
ABSTRACT The initiation and development of mesa corrosion attack during CO2 corrosion of carbon steel has been studied in flow loop experiments performed at 80 C and pH 5.8. Video recordings of growing mesa attacks have been performed in a test section with a glass window in the corrosion loop. These observations have shown that the mesa attack can grow both laterally and in depth below a lid of original corrosion film before the film is tom away stepwise by the flow. Possible mechanisms for initiation of mesa corrosion attack are discussed based on the observations from the video recordings. Mesa attacks can result from several small local attacks growing together into one large mesa attack.
INTRODUCTION The application of carbon steel in oil and gas pipelines and production tubulars depends to a large degree on either formation of protective corrosion product films or use of corrosion inhibitors. When the corrosion films do not offer complete protection a local attack known as mesa attack can develop. This type of attack is characterized by a deep and often flat-bottomed local attack without protective corrosion film and covering a large surface area. In the mesa attacked area the local corrosion rate can be several mm/year. The edge of the mesa attack can be very sharp as it is surrounded by areas with intact corrosion films and low corrosion rate.
The mechanisms for initiation and growth of mesa corrosion attack have not been fully explained as it has been difficult to observe the development of this type of attack in-situ. Reproducing mesa attack in laboratory experiments requires the use of flow loops operating under pressure at high temperature and with strict control of the water chemistry. The first in-situ visual observations of localized CO2 corrosion in the author?s laboratory were done in a cooperation between Conoco, the University of Oslo and Institute for Energy Technology and have been presented previously by Joosten et al.?. The subsequent experiments reported here were carried out in the multiclient research project Kjeller Sweet Corrosion IV, where the experimental technique was improved for better visualization of the localized corrosion attack with emphasis on the details during initiation and early growth of a mesa attack. Detailed results from the Kjeller Sweet Corrosion III and IV projects are presented elsewhere 3. In the present paper it is concentrated on the visual observation and video recording of growing mesa attack.
Flow loop experiments were performed at 80 C, 1.8 bar CO2 partial pressure and pH value around 5.8. Different carbon steels were tested as flat coupons exposed to flow rates from 0.1 to 7 m/s. Corrosion rates were measured for all specimens by weight loss and by measurement of depth of mesa attack or pits after the experiment. The corrosion rates of some of the specimens were followed during the experiment by linear polarization resistance measurements (LPR).
The desired pH value was obtained by addition of sodium bicarbonate. The experiments were done with 3 % NaCl in the water. The content of ferrous ions was kept above the iron carbonate solubility, thus enabling precipitation of iron carbonate films on the specimen. The Fe2+ conte
ABSTRACT Grain boundary Cr concentration influences resistance to intergranular stress corrosion cracking of stainless steel reactor core components. Therefore it is important to understand (1) how initial thermal processing affects thermal nonequilibrium segregation (TNES), (2) how service exposure affects radiation-induced segregation (RIS), and (3) how temperature affects post-irradiation annealing (PIA). In this study, each process is shown to be controlled by Cr-vacancy interactions. Mechanistic predictions are in accord with measured kinetics of changes in grain boundary Cr concentration for RIS and PIA but not for TNES. Inconsistencies between TNES model predictions and TNES measured segregation suggest that multiple solute-defect interactions must be considered and not just the simple interaction between Cr and vacancies. Furthermore, analysis of RIS of enriched TNES profiles suggests that Cr is chemically bound to the grain boundary such that the boundary concentration is influenced both by RIS in the matrix outside of the boundary plane and by chemistry within the TNES enriched boundary. Mechanistic understanding of changes in Cr concentration at grain boundaries may suggest mitigating measures for reduced in-core cracking by control of heat treatments, alloy chemistry and PIA.
INTRODUCTION Grain boundary Cr concentration critically affects cracking of stainless steels in oxidizing environments. A well established link between Cr concentration and intergranular stress corrosion cracking (IGSCC) has been established for unirradiated alloys that have grain boundary carbides associated with the Cr depletion.1 A much weaker relationship between Cr concentration and irradiation-assisted (IA) SCC has been found for irradiated stainless steels where Cr depletion is produced without IG carbides.2 Thermal heat treatments before and after irradiation can also affect grain boundary Cr composition and SCCs-5 As a result of these observations, it is essential to understand how grain boundary composition changes during material processing, fabrication and service can affect SCC and how methods for mitigating IASCC can be established.
In the absence of grain boundary carbides, grain boundary Cr concentration can (1) be enriched by rapid cooling from solution-anneal temperatures6 (2) be depleted by radiation- induced segregation @IS),7 and (3) be restored to the bulk concentration by post-irradiation annealing (PlA).s These three processes are controlled by Cr-vacancy interactions that result in solute redistribution. Thermal nonequilibrium segregation (TNES) has been described by Doig and Flewitts as well as Karlsson4 for grain boundary enrichment of solute during cooling from a solution-anneal temperature. Binding of Cr-vacancy complexes at solution-anneal temperatures and intermediate temperatures allows migration of complexes to grain boundaries during cooling resulting in enriched grain boundary Cr concentrations. On the other hand, RIS has been described by the inverse-Kirkendal effect during which the flow of irradiation-induced vacancies to grain boundaries causes depletion of the fast diffusing solute, Cr. Finally, isothermal PIA and conventional substitutional diffusion (Cr-vacancy exchange) eliminates concentration gradients caused by nonequilibrium processes of cooling and/or irradiation.
Recently, high-resolution analytical transmission electron microscopy (ATEM) observationsW have shown that Cr (and MO) can be significantly enriched at grain boundaries in solution- annealed austenitic stainless steels. Presegregation of this type appears to delay subsequent radiation-induced Cr depletion and impact IASCC susceptibility.s-1s This delay in depletion is not predicted f