Metal dusting is a kind of catastrophic corrosion phenomenon that can be observed in several of petrochemical processes. It occurs on iron-, nickel- and cobalt-base metals in carbonaceous atmospheres at high temperature when gaseous carbon activity is higher than one. The process is particularly rapid for ferritic alloys The aim of this project was to compare the dusting kinetics of pure iron and a 2.25Cr-1Mo alloy steel under CO-H2-H2O atmosphere at 650??C. Polished (3??m) samples of iron and the steel were exposed to flowing CO-H2-H2O gas atmospheres at 650??C, when the gases were supersaturated with respect to graphite. The partial pressure of CO was varied between 0.25 and 0.9 atm, and the carbon activity was varied from 2.35 to 16, in order to obtain a series of experimental conditions. In most experiments, pO2 was less than 7.37E-24 atm, and no iron oxide could form. However, Cr2O3 would always have been stable. When exposed to these gases, both iron and steel developed a surface scale of Fe3C which was buried beneath a deposit of carbon, containing iron-rich nanoparticles (the dust). Examination by Scanning Electron Microscopy allowed the observation of fine and coarse carbon nanotubes, and also spiral filaments. However, the morphology of the graphitic carbon was not sensitive to pCO and aC. Moreover, the carbon deposit was gas permeable, allowing continuing gas access to the underlying metal. At a fixed=4.5, the carburizing rate clearly increased with CO content from 0.25 to 0.68 atm. However, increasing the CO content to higher value led to decreased rates, indicating that carburizing rate reaches a maximum value at pCO=0.68 atm. When pCO was fixed at 0.25 atm and 0.68 atm, and carbon activity was varied. The induction period was extended by the formation of protective oxide layers at low values of carbon activity (aC= 2.35 and 2.55) where pO2 exceed the iron oxide formation value. For other reaction conditions, the carbon uptake rate for iron and steel did not increase with aC. The present work showed that the carbon deposition rates were not proportional to pCO or pCOpH2. Instead, the rate was affected by the partial pressure of all three reaction gases, and the carbon uptake rate for both materials could be expressed at r=k1pCOpH2+k2pCO2+k3pH22 and the rate constant k3 has a negative value, corresponding to coke gasification. From XRD analyses, it was found that cementite was the only iron-containing phase in the dusting product. The cementite particles acted as catalysts for carbon deposition from the gas. The same deposition process at the surface of the cementite layer led to its disintegration, thereby producing the particles. This disintegration process was faster on the steel than on pure iron. Consequently, the rates of both metal wastage and coke accumulation were faster for the steel. It is concluded that chromium and molybdenum do not stabilize the carbide but accelerate its disintegration process. It is suggested that Cr2O3 fine particles in the cementite layers provide more nucleation sites in the cementite layer on steel, explaining its more rapid dusting kinetics. However, appropriate methods of proving this assumption, such as TEM and FIB, are required.

Metal dusting is a kind of catastrophic corrosion phenomenon that can be observed in several of petrochemical processes. It occurs on iron-, nickel- and cobalt-base metals in carbonaceous atmospheres at high temperature when gaseous carbon activity is higher than one. The process is particularly rapid for ferritic alloys The aim of this project was to compare the dusting kinetics of pure iron and a 2.25Cr-1Mo alloy steel under CO-H2-H2O atmosphere at 650??C. Polished (3??m) samples of iron and the steel were exposed to flowing CO-H2-H2O gas atmospheres at 650??C, when the gases were supersaturated with respect to graphite. The partial pressure of CO was varied between 0.25 and 0.9 atm, and the carbon activity was varied from 2.35 to 16, in order to obtain a series of experimental conditions. In most experiments, pO2 was less than 7.37E-24 atm, and no iron oxide could form. However, Cr2O3 would always have been stable. When exposed to these gases, both iron and steel developed a surface scale of Fe3C which was buried beneath a deposit of carbon, containing iron-rich nanoparticles (the dust). Examination by Scanning Electron Microscopy allowed the observation of fine and coarse carbon nanotubes, and also spiral filaments. However, the morphology of the graphitic carbon was not sensitive to pCO and aC. Moreover, the carbon deposit was gas permeable, allowing continuing gas access to the underlying metal. At a fixed=4.5, the carburizing rate clearly increased with CO content from 0.25 to 0.68 atm. However, increasing the CO content to higher value led to decreased rates, indicating that carburizing rate reaches a maximum value at pCO=0.68 atm. When pCO was fixed at 0.25 atm and 0.68 atm, and carbon activity was varied. The induction period was extended by the formation of protective oxide layers at low values of carbon activity (aC= 2.35 and 2.55) where pO2 exceed the iron oxide formation value. For other reaction conditions, the carbon uptake rate for iron and steel did not increase with aC. The present work showed that the carbon deposition rates were not proportional to pCO or pCOpH2. Instead, the rate was affected by the partial pressure of all three reaction gases, and the carbon uptake rate for both materials could be expressed at r=k1pCOpH2+k2pCO2+k3pH22 and the rate constant k3 has a negative value, corresponding to coke gasification. From XRD analyses, it was found that cementite was the only iron-containing phase in the dusting product. The cementite particles acted as catalysts for carbon deposition from the gas. The same deposition process at the surface of the cementite layer led to its disintegration, thereby producing the particles. This disintegration process was faster on the steel than on pure iron. Consequently, the rates of both metal wastage and coke accumulation were faster for the steel. It is concluded that chromium and molybdenum do not stabilize the carbide but accelerate its disintegration process. It is suggested that Cr2O3 fine particles in the cementite layers provide more nucleation sites in the cementite layer on steel, explaining its more rapid dusting kinetics. However, appropriate methods of proving this assumption, such as TEM and FIB, are required.

A comparative study of the fatigue crack growth rate (FCGR) behaviour of five alloys in air and in aqueous environments has been performed. The alloys tested include: mild steel as a reference material, a corrosion resistant dual phase steel, 3CR12, a proprietary martensitic stainless steel, AISI 431, a newly developed 8% Cr martensitic steel, Alloy 825, and a newly developed corrosion-abrasion resistant metastable austenitic alloy, 1210. Tests were conducted in laboratory air, distilled water at rest potential, 500 ppm chloride solution at rest potential, 1000 ppm chloride solution at rest potential, and 1000 ppm chloride solution at -1200 m V see; solution temperatures were maintained at 25⁰ C. Crack growth rate tests were performed using sinusoidal loading at a load ratio R = 0.1, a frequency of 3Hz in the laboratory air, and a frequency of 1 Hz in the aqueous environments. At the completion of testing, fracture surfaces were studied using a scanning electron microscope. In air, the mild steel and 3CR12 display comparable rates of cracking and exhibit a greater resistance to fatigue crack propagation than the martensitic AISI 431 and Alloy 825; Alloy 825 shows the least resistance to fatigue crack propagation. The deformation induced transformation in 1210 gives this alloy the greatest resistance to fatigue crack propagation in air. Fatigue crack growth rates were all enhanced in the aqueous environments. The greatest overall rate of environmentally assisted cracking was shown by alloy 825 while the lowest was shown by the mild steel. Although the rate of cracking of 1210 in the aqueous environments was less than that of Alloy 825, 1210 was influenced the most by the aqueous environments. An environmentally assisted cracking index shows that the rate of fatigue crack propagation in 1210 is increased by 32 times in the 500 ppm chloride solution at low stress intensities. The fatigue crack growth rates of mild steel and AISI 431 were significantly influenced by the cathodically polarised conditions in the 1000 ppm chloride solution, compared to the rest potential conditions. In these cases hydrogen was seen to be evolved from the specimen surfaces. Changes in the fatigue crack growth rate behaviour were accompanied by changes in the fracture surface morphologies. The observed changes varied for each alloy and for each environment, and were manifest by the degree of intergranular cracking, cleavage, quasi cleavage, and increased coarseness of the transgranular cracking. The fracture surface morphologies are reported and discussed in detail. In general, the fracture surface morphologies could be directly related to the relative degrees of environmental influence on the rate of cracking; results are explained in terms of existing hypotheses. It is suggested that the environmentally assisted cracking of mild steel and AISI 431 at cathodic potentials in the 1000 ppm chloride solution could only be attributed to hydrogen assisted cracking. Similarly, it is suggested that the large crack growth rate acceleration of 1210 in the aqueous environments could also be attributed to hydrogen. The similar fracture surface morphologies observed on the other specimens after tests in the aqueous environments suggests-that hydrogen could be responsible for the environmentally assisted cracking of all the steels in aqueous environments.

Bibliography: pages 96-103. / The mechanisation of the working stapes in South African gold mines has required the introduction of a fundamentally new technology, hydro-power, in which machines are powered hydraulically using mine water fed from above ground. Mine water is aggressive and has a variable acidity and pH, and contains high concentrations of sulphate, chloride and nitrate ions. In order to minimise the pitting corrosion of piping and stoping machinery a compromise between selecting a suitable corrosion resistant material and treating the mine water to an acceptable level of corrosiveness is being sought.

An electrochemical investigation of the corrosion of iron and Fe18Cr based stainless steel alloys was undertaken with particular emphasis on the nucleation and growth of surface films. Chronoamperometry was shown to be a sensitive technique to investigate the initial stages of film formation and growth. In a variety of acidic (pH < 7), alkaline and alkaline cyanide electrolytes, providing dissolution of the substrate metal could occur rising current transients, similar to those reported in electrocrystallisation studies, were observed when the electrode was stepped to the appropriate potential. This indicated that at these potentials the surface film formed via the nucleation and growth of discrete nuclei. A significant aspect of this study was visual evidence of this nucleation and subsequent growth of the film provided by scanning electron microscopy which supported the electrochemical data and interpretation thereof. Existing electrocrystallisation models were used to evaluate the experimental rising current transients. While these models gave an indication as to the prevailing nucleation and growth mechanism, they were found to be inadequate in describing anodic oxide formation on an oxidising substrate. A qualitative model was proposed. In acidic electrolytes, rising chronoamperometric transients were observed for Fe, Cr and Fe18Cr at passive potentials and for FexCr (x = 16,18, 20,23% Cr) and alloys 444, 4732, 4733, 304L and 316L at transpassive potentials. The transients were shown to be sensitive to variations of potential, temperature, electrolyte and alloy composition. A systematic investigation of the influence of temperature (20 0 C - 1200 C) on the chronoamperometric, cyclic voltammetric and rotating ring - disc electrode behaviour of Fe in O.5M and 1.0M NaOH was also undertaken. In alkaline electrolytes, the formation of a duplex surface film was proposed, with x-ray photoelectron spectroscopy indicating that the protective base layer consisted of FeO while Fez03 and FeOOH constituted the upper layer. Base layer formation was favoured with increasing temperature and increasing hydroxide ion concentration of the electrolyte. Addition of OAM NaCN to O.5M and 1.0M NaOH had a marked effect on the electrochemistry of the system, with CN- inhibiting surface film formation, particularly of the upper layer. A mechanism for the oxidation of Fe in alkaline and alkaline cyanide electrolytes was proposed. / Thesis (Ph.D.)-University of Natal, 1994.

[Truncated abstract] Corrosion of steel structures caused by sulphide is a common engineering problem encountered by many industries, such as the petroleum, chemical processing, mining and mineral processing industries. The control of sulphide corrosion is still a controversial topic among corrosion engineers. There is an absence of guideline for a reliable acceptable limit of sulphide level in service and each processing industry has its own empirical values. Selection of inhibitors in the sulphide environment depends on laboratory testing before its actual application in pipelines and reaction vessels. Many investigators have postulated the corrosion mechanisms due to sulphide based on operating envelopes such as pH, chloride, manganese, hydrogen sulphide, sulphate reducing bacteria levels and inhibitor concentration. It is recommended in the literature that the batch dosing of inhibitor and biocide needs to be evaluated in regards to sulphide reducing bacteria (SRB) level, which may produce sulphide concentrations up to 2000 ppm. Although sulphide scale formation may protect the base metal by providing a physical barrier, the detrimental effects of sulphide are often inevitable, such as stress corrosion cracking, hydrogen embrittlement, etc. Currently, there are many chemicals that are used as inhibitors to prevent corrosion by scavenging the sulphide from the environment. Cerium, a rare-earth element, is not used as inhibitor in the sulphide environment. Also, there are no previous research findings on the effects of compounds of rare-earth metals, such as cerium chloride (CeCl3), in sulphide environment. This research examines the corrosion behaviour of 0.4Mo-0.8Cr steel, a High Strength Low Alloy (HSLA) steel, in sulphide-polluted artificial seawater with the addition of CeCl3 and glutaraldehyde. ... It is postulated that the moderate inhibiting effect of CeCl3 is due to the scavenging effect thereby forming Ce2S3 complex. Further reaction of sulphide with steel resulted in ferrous sulphide, leading to an increased corrosion rate. It is also concluded that the CeCl3 interferes with both anodic and cathodic reactions in deaerated conditions. Addition of glutaraldehyde in the sulphide-polluted seawater was found to decrease the corrosion rate. According to the electrochemical measurements conducted, the concurrent addition of glutaraldehyde and CeCl3 appeared to have an added effect on reducing the corrosion of the steel, as evidenced by the increase of the open circuit potential during the short-term testing. From the weight loss measurements after 60 days, sulphide pollution in deaerated seawater was found to increase corrosion rate. This is attributed to the increase of sulphide activity whereby continual dissolution of steel was encountered. From the weight loss tests, it was found that the addition of CeCl3 and glutaraldehyde reduced the corrosion rate of the steel in the solutions containing 0-10 ppm sulphide. There is no noticeable corrosion rate decrease for the solution containing 100 ppm sulphide. The added effect of CeCl3 and glutaraldehyde to the SRB medium has resulted in lower corrosion rates. Further detailed experimentation is required to elucidate the corrosion reduction mechanism in glutaraldehyde-containing environments.