Steam generators in nuclear power plants experienced varying degrees of under-deposit pitting corrosion. A probabilistic model to accurately predict pitting corrosion is necessary for effective life-cycle management of steam generators. This thesis presents an advanced probabilistic model of pitting corrosion characterizing the inherent randomness of the pitting process and measurement uncertainties of the in-service inspection (ISI) data obtained from eddy current (EC) inspections. A Bayesian method is developed for estimating the model parameters. The proposed model is able to estimate the number of actual pits, the actual pit depth as well as the maximum pit depth, which is the main interest of the pitting corrosion model. A MATLAB program of the Markov chain Monte Carlo technique is developed to perform the Bayesian estimations. Simulation experiments are performed to check the behavior of the Bayesian method. Results show that the MCMC algorithm is an effective way to estimate the model parameters. Also, the effectiveness and efficiency of Bayesian modeling are validated. A comprehensive case study is also presented on the in-service inspection data of pitting corrosion in a steam generator unit. The Weibull distribution is found to be an appropriate probability distribution for modeling the actual pit depth in steam generators.

Steam generators in nuclear power plants experienced varying degrees of under-deposit pitting corrosion. A probabilistic model to accurately predict pitting corrosion is necessary for effective life-cycle management of steam generators. This thesis presents an advanced probabilistic model of pitting corrosion characterizing the inherent randomness of the pitting process and measurement uncertainties of the in-service inspection (ISI) data obtained from eddy current (EC) inspections. A Bayesian method is developed for estimating the model parameters. The proposed model is able to estimate the number of actual pits, the actual pit depth as well as the maximum pit depth, which is the main interest of the pitting corrosion model. A MATLAB program of the Markov chain Monte Carlo technique is developed to perform the Bayesian estimations. Simulation experiments are performed to check the behavior of the Bayesian method. Results show that the MCMC algorithm is an effective way to estimate the model parameters. Also, the effectiveness and efficiency of Bayesian modeling are validated. A comprehensive case study is also presented on the in-service inspection data of pitting corrosion in a steam generator unit. The Weibull distribution is found to be an appropriate probability distribution for modeling the actual pit depth in steam generators.

Highly oxidized species like perchlorate and nitrate that are released into the environment by anthropogenic activities are a source of concern as they have been known to contaminate groundwater. These species are extremely soluble in water and can migrate through aquifer systems, travelling substantial distances from the original site of contamination. Due to their high solubility, these oxy-anions cannot be treated using conventional treatment processes like filtration and sedimentation. Several treatment technologies are currently available to abate the human health risk due to exposure to perchlorate and nitrate. However, most of the existing treatment processes are expensive or have limitations, like generation of brines with high concentrations of perchlorate or nitrate. Aluminum can effectively reduce perchlorate and nitrate, if the protective oxide film that separates the thermodynamically reactive Al0 from most environments is removed. Aluminum was activated by pH change and electrochemically induced, pitting corrosion to remove the passivating oxide layer and expose the underlying, thermodynamically reactive, zero-valent aluminum. A partially oxidized species of aluminum, like monovalent aluminum, is believed to bring about the reduction of perchlorate and nitrate. This research studied the reduction of perchlorate and nitrate by aluminum that was activated by these two mechanisms. Results indicated that aluminum activated by pH change resulted in an instantaneous decrease in perchlorate concentration without any increase in chlorate or chloride concentrations, which suggests that the perchlorate might be adsorbed on the aluminum oxide surface. However, aluminum activated by electrochemically induced pitting corrosion can effectively reduce perchlorate to chlorate. Nitrate, on the other hand, was reduced completely to ammonia by both treatment mechanisms. The studies conducted in this dissertation suggest that aluminum can be effectively used as a reducing agent to develop a treatment process to reduce perchlorate and nitrate.

In this study, a electrical pitting tester and SEM are employed to investigate the effects of supply voltage, supply current, and oil film thickness on the electrical behavior, the action forces, and the formation mechanism of electric pitting for the lubricated surface of steel pair at sliding speed of 1£gm/sec using an additive of MoS2 in paraffin base oil under DC electric field. According to the experimental results and the observations of the surface pitting, two electrical pitting regimes are found under the influences of shaft voltage, oil film thickness, and particle concentration of additive, namely, pitting and no- pitting regimes in static condition. The area of pitting regime increases with increasing additive concentration and supply current. Furthermore, The ratio of pitting area to the interface power increases rapidly with increasing additive concentration and oil film thickness. This results from the molten plateau that directly connects two specimens, and the interface power is mainly consumed at the heating of the plateau and the interfacial materials. However, the weld strength of the plateau isn¡¦t influenced with additive concentration. It is known from the observations of the surface pitting in dynamic pitting occurs that the pitting width increases with increasing oil film thickness and additive concentration. Finally, the formation processes of electric pitting on the lubricated surface for both static and sliding conditions are deduced from the results of the normal force, the friction force, the interface impedance and the observations of the surface pitting.

Abstract The electrical pitting often occurs at the bearing of the ro-tating machinery due to the actions of the shaft voltage and the shaft current resulting in the arcing effect on the lubricated surface and causing the bearing failure. Since the mechanism of the electrical pitting cannot be microscopically observed in process, it is difficult to prevent the bearing damage. Hence, this study uses a static electrical pitting tester with sub -micrometer accuracy to experimentally investigate the effects of supply voltage, supply current, oil film thickness, and ad-ditive on the threshold condition of electrical pitting under the conventional bearing material pairs. Moreover, according to the SEM micrograph and EDS analysis, the mechanism of the pitted surfaces is investigated. According to the experimental results and the surface ob-servations of steel/steel pair using a paraffin base oil, three electrical pitting regimes are found under the influences of shaft voltage and oil film thickness, namely, pitting, transition, and no-pitting regimes. In the electrical pitting regime, the interface voltage, interface impedance, and interface power increases slightly with increasing oil film thickness at a certain supply current. However, the interface voltage and interface power increases with increasing supply current, and the inter-face impedance decreases with increasing supply current at a certain film thickness. Furthermore, the pitting area versus the interface power relationship is a cubic function. According to the experimental results and the surface ob-servations of babbitt alloy/steel pair using a paraffin base oil, two electrical pitting regimes are found under the influences of shaft voltage, oil film thickness, and melting point of material, namely, pitting and no-pitting regimes. The mechanism of electrical pitting on the babbitt alloy surface is significantly influenced by the interface power and the oil film thickness. At the smaller oil film thickness, the eroded surface of babbitt alloy exhibits a concave crater with a few micro-porosity in the vicinity of center region with a plateau on its surrounding, especially at high supply current. The polished track can be observed at the plateau. A large amount of tin element trans-fers to the steel ball surface because the molten tin contacts the ball. At the higher oil film thickness, only a little amount of metal element transfers to each other. The major pitting area of the babbitt alloy is caused at the initial stage of the arc dis-charge. With increasing arc discharge time, the pitting area increases slightly, and finally reaches a saturated value. According to the experimental results and the surface ob-servations of babbitt alloy/steel pair using an additive of MoS2 in a paraffin base oil, two electrical pitting regimes are found under the influences of shaft voltage, oil film thickness, and particle concentration of additive, namely, pitting and no-pitting regimes. The area of pitting regime increases with increasing additive concentration and supply current. Fur-thermore, the ratio of pitting area to the interface power in-creases with increasing additive concentration and supply current at the oil film thickness smaller than 6 mm. However, this ratio increases rapidly to about 10 times with increasing additive concentration and supply current as the oil film thickness increases from 6 mm to 10 mm. This results from the molten plateau that directly connects two specimens, and the interface power is mainly consumed at the heating of the pla-teau and the interfacial materials. According to the above re-sults, the growth model of the plateau on the pitting surface is proposed at the lubricated condition using an additive of MoS2 in paraffin base oil.

Nickel based alloys have been developed as a material offering superior general and localized corrosion resistance compared to the more traditionally used in chemical and oil plant in the most aggressive environment such as hydrochloric acid and ferric chloride. Hence the addition of Cr and Mo to Ni creates alloys with exceptional corrosion resistance in a diverse range of environments. This study examines the roles of Cr and Mo in the corrosion behavior of Ni alloys. The performance of three nickel-base alloys IN600, IN601 and C22 was examined in increasing saline solution severity of sodium chloride, concentrated hydrochloric acid and ferric chloride solution at pH = 0.0. The passive corrosion and breakdown behavior of these alloys suggests that Cr is the primary element influencing general corrosion resistance, while the repassivation potential is strongly dependent on the Mo content. This indicates that Cr plays a strong role in maintaining the passivity of the alloy, while Mo acts to stabilize the passive film after a localized breakdown event. Corrosion fatigue test results indicate that fatigue life of IN600, IN601 and C22 specimens tested in 3.5 % sodium chloride solution are essentially the same as for specimens tested in air. Test results also showed that for IN600, IN601 and C22 alloys, the number of cycles to failure was highest in air and sodium chloride solution, followed by specimens fatigued in hydrochloric acid, and was least in specimens fatigued in ferric chloride solution. No evidence of surface pitting was found on C22 specimens in all three solutions whereas IN600 and IN601 were both pitted. However, pits were generally larger in IN600 likely due to lower Cr content than in IN601.

Nickel based alloys have been developed as a material offering superior general and localized corrosion resistance compared to the more traditionally used in chemical and oil plant in the most aggressive environment such as hydrochloric acid and ferric chloride. Hence the addition of Cr and Mo to Ni creates alloys with exceptional corrosion resistance in a diverse range of environments. This study examines the roles of Cr and Mo in the corrosion behavior of Ni alloys. The performance of three nickel-base alloys IN600, IN601 and C22 was examined in increasing saline solution severity of sodium chloride, concentrated hydrochloric acid and ferric chloride solution at pH = 0.0. The passive corrosion and breakdown behavior of these alloys suggests that Cr is the primary element influencing general corrosion resistance, while the repassivation potential is strongly dependent on the Mo content. This indicates that Cr plays a strong role in maintaining the passivity of the alloy, while Mo acts to stabilize the passive film after a localized breakdown event. Corrosion fatigue test results indicate that fatigue life of IN600, IN601 and C22 specimens tested in 3.5 % sodium chloride solution are essentially the same as for specimens tested in air. Test results also showed that for IN600, IN601 and C22 alloys, the number of cycles to failure was highest in air and sodium chloride solution, followed by specimens fatigued in hydrochloric acid, and was least in specimens fatigued in ferric chloride solution. No evidence of surface pitting was found on C22 specimens in all three solutions whereas IN600 and IN601 were both pitted. However, pits were generally larger in IN600 likely due to lower Cr content than in IN601.

A resistência à corrosão das ligas de alumínio está relacionada ao meio de exposição, composição química, presença de intermetálicos e microestrutura do metal, entre outras. Neste trabalho investigou-se a corrosão nas ligas de alumínio 3105 H16 e 5052 H34, comumente utilizadas na indústria de carrocerias de ônibus, em meio aquoso continham íons agressivos, tais como cloretos, sulfatos e hidroxilas. Avaliou-se comparativamente o comportamento destas ligas com os revestimentos de proteção de Nanocerâmico, Cromo VI, Cromo III e Sistema de Pintura em exposição atmosférica acelerada e imersão em diferentes meios agressivos. Estudos mais aprofundados restringiram-se às ligas sem tratamento, para as quais foram realizados ensaios de exposição acelerada em névoa salina neutra, névoa salina acética, câmara de umidade e exposição atmosférica natural. Microscopia eletrônica de varredura por emissão de campo foi empregada para caracterizar a superfície das ligas antes e após os ensaios de névoa salina neutra e acética. O comportamento eletroquímico das ligas sem tratamento foi determinado através do monitoramento do potencial de circuito aberto, curvas de polarização potenciodinâmica anódicas e espectroscopia de impedância eletroquímica. Os resultados mostraram que o tratamento de Cromo VI foi o que obteve melhor desempenho na proteção de ligas de alumínio, particularmente para a liga 5052. De um modo geral, a liga 5052 apresentou maior resistência à corrosão nos meios testados em presença ou não de tratamentos superficiais. Na exposição à névoa salina, a corrosão manifestou-se principalmente na forma de pites. Ficou comprovado que o efeito do íon Cl- é importante, porém o fator preponderante do comportamento à corrosão das ligas de alumínio está relacionado ao pH do meio de exposição, sendo este recomendado para avaliar revestimentos protetores. Além das condições do meio ao qual o metal será exposto, na seleção de tratamentos anticorrosivos é importante considerar as particularidades de cada liga, uma vez que suas características composicionais e microestruturais exercem relevante influência no desempenho à corrosão. / The corrosion resistance of aluminum alloys is related to the exposure medium, chemical composition, presence of intermetallic particles and metallic microstructure, among others. This work investigated corrosion of 3105 H16 and 5052 H34 aluminum alloys commonly used in the bus body industry in aqueous media containing aggressive ions, such as chlorides, sulfates and hydroxyls. The behavior of these alloys with the protective coatings of Nanoceramic, Chromium VI, Chromium III and Paint System in accelerated atmospheric exposure and immersion in different aggressive media was evaluated comparatively. Further studies were restricted to untreated alloys for which accelerated exposure tests were performed on neutral salt spray, acetic salt spray, moisture chamber and natural atmospheric exposure. Field scanning electron microscopy was used to characterize the alloys surface before and after neutral and acetic salt spray tests. The electrochemical statement of the untreated alloys was determined by monitoring the open circuit potential, anodic potentiodynamic polarization curves and electrochemical impedance spectroscopy. Results have shown that the treatment with Chromium VI was the one that obtained better performance in the protection of the aluminum alloys, particularly for 5052 alloy. In general, 5052 aluminum alloy presented greater resistance to corrosion in all tested media, with or without surface treatments. Under salt spray exposure, the corrosion attack appeared mainly in the form of pitting. It has been proven that the effect of Cl- ion is important, but the predominant factor on the corrosion behavior of aluminum alloys is related to the pH of the exposure medium, which is recommended to evaluate protective coatings. Besides the conditions of the medium to which the metal will be exposed, in the selection of anticorrosive treatments it is important to consider the particularities of each alloy, since its compositional and microstructural characteristics exert a relevant influence on the corrosion performance.

Resumo: A resistência e o mecanismo de corrosão das ligas de alumínio 2024, 7010, 7050 e 7475 foram estudados em solução de NaCl. Os efeitos do tratamento térmico nas ligas, concentração de oxigênio, pH, adição de oxi-ânions e temperatura do meio constituem algumas das variáveis estudadas. Primeiramente, procedeu-se à caracterização físicoquímica dos materiais através de análise química e metalográfica, mediante microscopia eletrônica de varredura e espectroscopia de energia dispersiva. As ligas 2024, 7010, 7050 e 7475 como recebidas, recozidas e envelhecidas, revelaram a existência de partículas ternárias e quaternárias, constituídas por Al:Cu:Fe e Mg ou Zn. Também foi observado um maior número de partículas pequenas de composição variável situando-se, preferencialmente, nos contornos dos grãos. Os resultados dos ensaios de corrosão em meios aerados e desaerados indicam que o cromato é efetivo como inibidor da corrosão localizada em ambas ligas e que o molibdato somente na liga 7050. O efeito inibidor do tungstato se revela em meio desaerado e é comparativamente menor daquele observado com os outros oxi-ânions. As análises quantitativas de superfície das ligas após os ensaios de imersão indicam que ainda na presença de inibidor, se pites foram nucleados, eles crescem. / Abstract: Corrosion resistance and mechanism of 2024, 7010, 7050 and 7475 aluminium alloys have been studied in NaCl solution. Heat treatment effect on the alloys, oxygen concentration, pH, oxy-anions addition and environment temperature constitutes some of the studied variables. First, it has been proceeded the materials physical chemistry characterization through chemical and metalography analysis, by means of scanning electron microscopy and energy dispersive X-ray spectroscopy. The 2024, 7010, 7050 and 7475 alloys as-received, annealed and aged, have disclosed ternary and quaternary particle existence, consisting of Al:Cu:Fe and Mg or Zn. A major small particle number of variable composition preferentially localized in grain boundaries has also been observed. Corrosion tests results in aerated and deaerated media indicate that chromate are effective as localized corrosion inhibitor in both alloys whereas molibdate is only in 7050 alloy. Tungstate inhibiting effect is revealed in deaerated medium and is comparatively lesser than that observed with the other oxy-anions. Quantitative alloys surface analyses after immersion tests indicate that despite inhibitor presence, when pits have been nucleated, they grow.

Resumo: Foram caracterizados os comportamentos eletroquímicos e avaliadas as resistências à corrosão das ligas aeronáuticas 2024-T351 e 7050-T7451 em soluções aquosas de cloreto contendo cromato, molibdato e tungstato. Foram realizados ensaios de corrosão não-eletroquímicos de imersão prolongada acompanhados de análise metalográfica de superfície por microscopia óptica e identificação dos produtos de corrosão por difratometria de raios-X. A análise quantitativa de superfícies das ligas após a imersão, indica que os pites formados têm áreas médias similares. Os pites são mais largos do que profundos e de geometria, predominantemente, cônica ou quase-cônica e irregular. Em todos os produtos de corrosão de cada liga foi encontrado hidróxido de alumínio, em suas diferentes formas cristalinas. Medidas de perda de dureza, como uma conseqüência da deterioração superficial, também foram determinadas. Além disso, ensaios eletroquímicos como medidas de potencial em circuito aberto, curvas de polarização e voltametria cíclica complementaram este estudo. Em meio aerado os resultados obtidos mediante medidas eletroquímicas são consistentes com aqueles obtidos nos ensaios de imersão, em particular o efeito do CrO42- e do MoO42-. O WO42- mostrou-se agressivo em períodos prolongados de imersão. Apesar dos ensaios revelarem uma redução parcial de MoO42- em ambas as ligas, o efeito desse oxi-ânion parece ser diferente sobre cada liga. Em meio desaerado as ligas apresentam passivação em todos os eletrólitos. A adição dos oxi-ânions não modificou significativamente o potencial de pite para a liga 7050, enquanto que para a liga 2024 ele foi deslocado levemente para valores mais positivos. / Abstract: It has been characterized the electrochemical behavior and evaluated the 2024-T351 and 7050-T7451 aircraft alloys corrosion resistance in chloride aqueous solutions containing chromate, molybdate and tungstate. It has been carried out non-electrochemical long immersion corrosion testings accompanied by surface metalography analysis achieved by light microscopy and corrosion products identification by X-ray difratometry. Surfaces quantitative analysis upon the alloys after immersion, indicates that formed pits have similar average area. Pits are widther than deeper and own predominantly a conical or quasi-conical and irregular geometry. In all corrosion products of each alloy it has been found aluminum hydroxide in its different crystalline ways. Hardness loss measurements have also been determined. In addition, electrochemical testings such as open circuit potential measures, polarization curves and cyclical voltammetry have completed this study. In aerated means the obtained results before electrochemical mesurements are similar to those obtained in the immersion tests, in particular CrO42- and MoO42- effects. WO42- has been found to be aggressive in very long immersion period. Though tests display a MoO42- partial reduction in both alloys, this oxi-anion effect seems to be different upon each alloy. In de-aerated means alloys present passivation in all eletrolytes. Oxi-anion addition has not changed significantly pit potential for 7050 alloy, while for 2024 alloy it has been dislocated, slightly, for more positive values. / Orientador: Eduardo Norberto Codaro / Coorientador: Roberto Zenhei Nakazato / Banca: Luis Rogerio de Oliveira Hein / Banca: Gilberto Luis Jardim Pinto da Silva / Mestre