Orientador: Roseana da Exaltação Trevisan / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica / Made available in DSpace on 2018-08-05T03:36:29Z (GMT). No. of bitstreams: 1 Fraga_FranciscoEdsonNogueira_M.pdf: 7668507 bytes, checksum: 806c3c771652dfe2aa90373e2fca42ba (MD5) Previous issue date: 2005 / Resumo: Trinca induzida por hidrogênio é um dos defeitos mais graves que ocorrem em juntas soldadas de diversos aços. Esta falha mecânica ocorre com bastante freqüência na zona afetada pelo calor (ZAC) destes materiais. Com o desenvolvimento de novos aços, o risco da ocorrência de trincamento na ZAC tem se reduzido cada vez mais e passado a ocorrer com mais freqüência no metal de solda (MS). O teste G-BOP (Gapped Bead-on-Plate) destaca-se dentre os vários testes autodestrutivos pela grande aplicação em avaliar susceptibilidade de ocorrência de trincas de hidrogênio somente no cordão de solda. As principais vantagens deste teste são: baixo custo quando comparado a outros testes, simplicidade na execução e facilidade em quantificar trincas a frio no MS. Apesar de todas as vantagens, este teste tem sérias limitações e uma delas é o fato de não ser normalizado. Visando contribuir com informações que possam agregar maior confiabilidade e contribuir para a normalização do teste G-BOP, este trabalho teve como objetivo principal estudar de maneira sistemática e científica a influência das principais variáveis do teste G-BOP (dimensão do rebaixo, energia nominal de soldagem e temperatura de preaquecimento) sobre a variável de resposta, que é o percentual de trinca induzida por hidrogênio na seção transversal do metal de solda, As três variáveis foram estudadas segundo uma análise estatística de variância, identificando a influência individual de cada uma e a interação entre elas sobre os resultados do teste. Para o desenvolvimento experimental foi utilizado como material de base um aço ASTM A-285 grau C, como metal de adição, um arame tubular de classificação AWS E71T-1 e o 'CO IND. 2¿ como gás de proteção auxiliar. Para os níveis de cada uma das variáveis analisadas aqui, identificou-se que a energia nominal de soldagem e a temperatura de preaquecimento são variáveis que tem influência significativa sobre os resultados do teste G-BOP e que a variável dimensão do rebaixo não tem influência significativa. Identificou-se ainda que a interação entre estas variáveis também não apresenta influência significativa sobre os resultados do teste / Abstract: Hydrogen induced cracking (HIC) is a serious defect that occurs in welded joints of several steel types. This mechanical failure occurs frequently on the heat affected zone (HAZ). With the development of new steels the probability of HIC occurring on the HAZ has reduced, however it has started to occur on the weld metal (WM). The Gapped bead-on-plate test (G-BOP) stands out from several other self-restraint tests for its great application to evaluate HIC only on the weld metal. The main advantages of this test are: low cost, simple execution and crack quantification on WM. Despite its advantages, this test has a serious limitation that it is not normalized. To contribute to getting information that can add greater trustworthiness to G-BOP test and help to normalize it, the objective of this study is to evaluate the influence of the main variable of the G-BOP test (gap, welding heat and preheat temperature) in the output variable (HIC %). A variance analysis was used to identify the influence of these variables in the test results. For the experimental development the ASTM A-285 grade C steel was used as base metal as well as AWS E71T-1 flux core and 'CO IND. 2¿ shielding. It was concluded that the heat input and the preheat temperature have significant influence in the test result. The gap and the interaction between these variables don't have any influence in the test result / Mestrado / Materiais e Processos de Fabricação / Mestre em Engenharia de Fabricação

A tecnologia dos aços microligados para transporte de gás natural e petróleo tem sido pressionada pelo descobrimento das novas jazidas e o aumento da demanda no mundo. As solicitações ambientais e de resistência mecânica são os parâmetros para o desenvolvimento de aços de alta resistência baixa liga para o transporte de gás e petróleo a menor custo e de forma segura. Neste contexto esta pesquisa investiga, em um aço microligado para tubos API 5L X80, o efeito das transformações de fase obtidas por resfriamentos controlados na fratura induzida por hidrogênio Hydrogen Induced Cracking-(HIC) e nas propriedades mecânicas. Os testes de HIC foram realizados no material como recebido, na espessura da chapa submetida a resfriamentos contínuos e em amostras do material tratadas de forma a simular as regiões de grão grosso da zona afetada pelo calor (GGZAC). Segundo o ciclo de resfriamento, os aços microligados têm microestruturas complexas, como é caso do aço microligado em estudo, onde sua microestrutura, estudada em trabalhos anteriores, é formada principalmente por ferrita, bainita, perlita e microconstituinte austenita/martensita (AM). A morfologia, tamanho, quantidade e distribuição dos produtos de transformação na chapa mudam as propriedades do aço. Esses fenômenos são de grande interesse tecnológico em aços microligados para a fabricação de tubos soldados para o transporte de gás e petróleo, tanto quando a solda é realizada em campo como também durante o encurvamento por indução; aqui as propriedades mecânicas do tubo decorrentes do processo de fabricação termomecânico podem ser degradadas pela ação do aquecimento e dos resfriamentos experimentados na zona afetada pelo calor (ZAC), principalmente na região de GGZAC. A simulação dos ciclos térmicos para o estudo da HIC na espessura da chapa foram realizados em CP austenitizados a 900ºC e submetidos a resfriamentos contínuos no dilatômetro de têmpera. Para simular os ciclos térmicos com resfriamentos controlados focados na GGZAC e a seguir obter CP de tamanho adequado para testes de tração e Charpy, foi necessário fazer os tratamentos térmicos a 1300ºC e resfriamentos contínuos em um simulador termomecânico e dilatômetro Gleeble. O maior tamanho da amostra tratada termicamente neste último equipamento permitiu extrair amostras para avaliar as propriedades mecânicas e a HIC do material, pois as diferentes regiões da ZAC em uma solda real são restritas e não permitem este tipo de ensaios em uma região específica da ZAC. Os resultados permitiram identificar a suscetibilidade de cada microestrutura produto da transformação da austenita na espessura da chapa, sendo a região central da chapa a mais sensível ao hidrogênio no aço como recebido e quando tratado a baixas taxas de resfriamento de 0,5°C/s após austenitizado a 900°C. As bandas grosseiras formadas por estruturas de maior dureza que a matriz na região central diminuíram a resistência à HIC. Da mesma forma nos corpos de prova que simulam a região GGZAC, a fratura induzida pelo hidrogênio foi localizada na região central da espessura embora apresente bainita e ferrita acicular. A falha possivelmente se deve aos elementos remanescentes segregados nesta região central e partição de carbono para os sub contornos de grão da bainita e ferrita que cresceram a partir a austenita primária. As inclusões e precipitados, segundo seu tipo, forma e localização na microestrutura, participam ou não da nucleação e propagação da trinca, sendo a posição mais crítica quando localizadas dentro das estruturas bandeadas. Não foi observada a nucleação de trincas na presença de hidrogênio em precipitados de Nb e Ti. / The technology of microalloyed steels for the transportation of natural gas and oil has been pressed by the discovery of new deposits and the increased demand in the world. Environmental requests for safety and ever increasing mechanical strength are the parameters for the development of high strength low alloy steels for transporting gas and oil at lower cost and safely. In this context, this research investigates, in a microalloyed steel pipe API 5L X80, the effect of phase transformations obtained by controlled cooling on the behavior when loaded with hydrogen - Hydrogen Induced Cracking - (HIC) and in the mechanical properties. HIC tests were performed on as-received material, on samples extracted from the thickness of the plate and subjected to continuous cooling and on samples of the material treated to simulate the coarse-grained regions of heat affected zone (CGHAZ). According to the cooling cycle, the microalloyed steels have complex microstructures: in the steel under evaluation its microstructure, studied in a previous work, consists mainly of ferrite, bainite, pearlite and austenite/martensite constituent (AM). The morphology, size, quantity and distribution of the products of transformation change the properties of plate steel. These phenomena are of great technological interest in microalloyed steels for the fabrication of welded tubes for the transport of gas and oil, when the welding is performed in the field as well as during hot bending; here the mechanical properties of the tube from the process of thermomechanical fabrication can be degraded by the action of heating and cooling experienced in the heat affected zone (HAZ), mainly in the region of CGHAZ. Simulations of thermal cycles for the study of HIC on sheet thickness were performed in coupons subjected to austenitization at 900ºC followed by continuous cooling in the dilatometer. To simulate the thermal cycles with controlled cooling, focused in the CGHAZ, and getting suitable sample sizes for tensile testing and Charpy, it was necessary to austenitize at 1300ºC followed by continuous cooling using the thermal and thermomechanical simulator in a Gleeble dilatometer. Samples heat treated in this equipment were suitable to evaluate the mechanical properties and the HIC of the material for different regions of HAZ, while a real weld would not have enough material to allow this type of testing on a specific region of HAZ. The results showed the susceptibility of each microstructure product of austenite transformation and of the position on the plate thickness. The central region of the plate was more sensitive to hydrogen in the steel as-received and when treated at low cooling rates of 0.5°C/s after austenitization at 900°C. The bands formed by coarse structures of greater hardness than the matrix in the central region decreased the resistance to HIC. Likewise in coupons that simulate the CGHAZ region, the fracture induced by hydrogen was located in the central thickness line, even when the microstructure were bainite and acicular ferrite. Failure there was possibly due to remnants of segregated elements in this central region and carbon partition to the subboundaries of the bainite and ferrite grain that grew from the primary austenite. Inclusions and precipitates, according to their type, shape and location in the microstructure, participating or not in the nucleation and propagation of the crack, were more critical when located within the banded structures; crack nucleation in the presence of hydrogen was not observed at Nb and Ti precipitates.

Com a descoberta de novas fontes de petróleo e gás, em regiões remotas e de difícil acesso, tem-se a necessidade do desenvolvimento de novas tecnologias para garantir a eficácia da exploração destes recursos. Essa exploração e extração muitas vezes se dão em ambientes altamente corrosivos e os equipamentos devem apresentar propriedades que garantam um fator de segurança em serviço. Os aços de alta resistência e baixa liga (ARBL) são utilizados em tubulações para o transporte de gás natural e petróleo. Estes estão constantemente expostos a ambientes ácidos os quais são compostos de umidade e sulfeto de hidrogênio (H2S), podendo causar falha induzida pela presença de hidrogênio (Hydrogen Induced Cracking HIC). Este tipo de falha é normalmente abordado na literatura através de ensaios em solução contendo ácido acético e/ou sais (cloreto de sódio, entre outros), sempre com a injeção de H2S. Há vários mecanismos propostos, no entanto, o assunto não está totalmente resolvido. As alterações de composição química dos aços, processos de refino do aço e processos de conformação mecânica são responsáveis pela microestrutura final e determinantes da resistência à fragilização por hidrogênio. O objetivo deste trabalho é analisar e comparar o comportamento quanto à resistência à corrosão e resistência à HIC de quatro materiais: tubo X65 sour, sua região de solda, tubo X65 não-sour e uma chapa destinada a confecção de tubo X65. Os eletrólitos empregados foram: solução A (ácido acético contendo cloreto de sódio) e a solução B (água do mar sintética), os quais correspondem às soluções recomendadas pela norma NACE TM0284-2003. Os materiais foram submetidos a: ensaios de polarização (Polarização Linear para determinação da Resistência de Polarização - Rp) e ensaios de resistência a HIC segundo a norma NACE TM0284-2003; exames em microscópio óptico e eletrônico de varredura para caracterização da morfologia da corrosão e do trincamento. Os ensaios de Rp revelaram que a solução A é mais agressiva do que a solução B, sendo isso explicado pela diferença de pH entre estas duas soluções. Os resultados mostraram ainda que a máxima resistência à corrosão sempre é obtida para o tubo sour, enquanto a mínima ocorreu para o tubo não-sour. Após o ensaio de resistência a HIC os exames em microscópio óptico revelaram que, em ambas as soluções, o tubo de X65 sour, e a sua solda não apresentaram trincas, bem como a chapa destinada a tubo X65; já o tubo de X65 não-sour apresentou trincamento principalmente na região central. Os exames das trincas revelaram que a presença de cementita intergranular e a estrutura bandeada foram as causas do trincamento. No caso do tubo sour, o bom desempenho foi discutido em termos da microestrutura de ferrita poligonal, acicular e microconstituinte M/A. Já o comportamento distinto encontrado para a chapa (para tubo X65), foi discutido levando-se em conta que esta chapa apresentou menor quantidade de cementita intergranular, uma vez que, sua microestrutura é bandeada e não foi encontrado trincamento. Os resultados também revelaram que a solução B, como no caso da resistência à corrosão, é uma solução menos agressiva, pois o trincamento obtido foi muito menor. / The discovery of new oil and gas reserves, at remote and hard to reach locations, makes imperative the development of new technologies to ensure effective exploitation of these resources. This exploitation is often performed at highly corrosive environments and equipment such as pipelines should have special mechanical and corrosion properties to guarantee safety levels in service. High-Strength Low Alloy (HSLA) steels are used in pipelines for transporting gas and oil. These steels are in constant exposure to acid environments containing hydrogen sulfide (H2S) and water, that can cause pipeline failures due to Hydrogen-Induced Cracking - HIC. The literature reports that Hydrogen-Induced Cracking in steels is normally tested in solutions containing acetic acid and/or, salts (sodium chloride and others) with addition of H2S. Chemical composition, steel refining processes and metal forming processes are responsible for the final microstructure of the steel and have effect on the hydrogen embrittlement resistance. The purpose of this work is to analyze and compare the corrosion resistance and HIC resistance, and compare of four materials: pipeline steel API 5L X65 for sour service, its welded junctions, pipeline steel API 5L X65 for non-sour service and pipeline steel plate API 5L X65. The materials were submitted to linear polarization test (Rp) and HIC resistance test according to NACE TM0284-2003 standard. Both tests were carried out with two different electrolytes: the solution A (acetic acid and sodium chloride) and solution B (synthetic seawater). Subsequently; the surface of the steels were evaluated by optical microscope and scanning electron microscopy in order to characterize the cracking modes and corrosion morphology. The Rp tests showed that the solution A is more aggressive than solution B, behavior attributed to the pH difference between solutions. Steel API 5L X65 for sour service had the highest corrosion resistance and pipeline steel API 5L X65 for non-sour service had the lowest. The HIC test and the surface examination revealed that in both solutions, pipeline steel API 5L X65 for sour service, the welded junctions and the pipeline steel plate API 5L X65 showed no cracks. On the other hand, pipeline steel API 5L X65 for non-sour service presented cracking mainly in the central region. The tests revealed that the cracks nucleated at the intergranular cementite in the banded structure. The good performance of the pipeline steel API 5L X65 for sour service was discussed in terms of the microstructure, formed by polygonal ferrite, acicular ferrite and M/A microconstituent. The performance of steel plate (for pipeline API 5L X65) was different. This material did not exhibit cracks in the matrix in spite of its banded microstructure. This result was discussed taking into account that the plate studied had a small amount of intergranular cementite. The results also showed that the solution B, as in the case of corrosion resistance tests, was less aggressive than solution A, because the cracks produced were smaller.

Abstract From the largest discontinuities to the smallest of the elements, various factors can threaten structural integrity. Susceptibility to these factors elevates with higher yield strengths. As-quenched low-carbon steels with a martensitic or martensitic-bainitic microstructure are modern ultra-high-strength structural steels. They can possess sufficient toughness, formability, and weldability, and are typically used in weight-critical and high-performance structures. Common problems with as-quenched steels with a yield strength of 900 MPa or more are that they do not obey the conventional correlation between the fracture toughness reference temperature T₀ and the impact toughness transition temperature T₂₈ᴊ used in many standards and structural integrity assessment procedures, and a lack of design rules in general. This thesis studies the relationship between the T₀ and T₂₈ᴊ to provide additional knowledge for future standardisation, the microstructural features governing the toughness at these temperatures on both global and local scale, and whether hydrogen embrittlement is present at subzero temperatures. It uses steels produced via laboratory rolling and quenching as well as from pilot-scale and full-scale industrial production, studying them with standardised toughness tests, microstructural characterisation, fractography, and cohesive zone modelling. As-quenched steels have a distinct correlation between T₀ and T₂₈ᴊ. An improved general T₀ – T₂₈ᴊ correlation applies to a wide range of steels. T₂₈ᴊ correlates closely with a dynamic reference toughness, which can be used together with the fraction of detrimental {100} cleavage planes near the main fracture plain to effectively estimate the transition temperatures. On a local scale, centreline segregation decreases the effective coarse grain size, which more than compensates for the harmful effects associated with the higher hardness and inclusion content of the centreline, resulting in increased fracture toughness. Hydrogen embrittlement causes a decrease in fracture toughness and local deformability, thereby increasing T₀ while leaving T₂₈ᴊ unaffected. Overall, the results show that high toughness demands good control of effective coarse grain size and hydrogen content. / Tiivistelmä Tekijät suurimmista epäjatkuvuuskohdista aina pienimpään alkuaineeseen voivat uhata rakenteellista eheyttä, minkä lisäksi alttius näille kasvaa materiaalin myötölujuuden kasvaessa. Modernit karkaistun tilan ultralujat matalahiiliset rakenneteräkset voivat silti omata riittävän sitkeyden, muovattavuuden ja hitsattavuuden. Tyypillisiä käyttökohteita näille ovat painon suhteen kriittiset ja korkean suorituskyvyn rakenteet. Yleinen ongelma myötölujuudeltaan noin ja yli 900 MPa karkaistun tilan teräksillä on se, että ne eivät noudata perinteistä murtumissitkeyden referenssilämpötilan T₀ ja iskusitkeyden transitiolämpötilan T₂₈ᴊ välistä korrelaatiota, jota käytetään useissa standardeissa ja suunnitteluohjeissa, jotka eivät myöskään vielä salli näin lujien terästen käyttöä. Tämä väitöstyö tutkii transitiolämpötilojen T₀ ja T₂₈ᴊ välistä suhdetta edistääkseen näiden terästen sisällyttämistä standardeihin, haurasmurtuma-sitkeyteen vaikuttavia mikrorakenteellisia tekijöitä sekä yleisellä että paikallisella tasolla, ja vetyhaurautta matalissa lämpötiloissa. Koeteräkset ovat laboratoriovalmisteisia, tuotantokokeita ja tuotantolaatuja. Niitä tutkitaan standardisoiduilla sitkeyskokeilla, mikrorakenteen karakterisoinnilla, fraktografialla ja koheesiovyöhykettä hyödyntävällä mallinnuksella. Tulokset osoittavat karkaistun tilan terästen omaavan erityisen korrelaation T₀ ja T₂₈ᴊ välillä. Muokattu, ultralujat teräkset huomioiva yleinen T₀ – T₂₈ᴊ -korrelaatio soveltuu laajalti eri terästyypeille. T₂₈ᴊ korreloi läheisesti dynaamisen referenssisitkeyden kanssa, jonka avulla yhdessä haitallisten {100} lohkomurtumatasojen osuuden kanssa voidaan estimoida joukko transitiolämpötiloja. Paikallisella tasolla keskilinjasuotauma pienentää efektiivistä karkeiden rakeiden kokoa, mikä suotauman suurista sulkeumista ja kovuudesta huolimatta parantaa murtumissitkeyttä. Vetyhauraus taas huonontaa sitkeyttä ja paikallista muodonmuutoskykyä myös matalissa lämpötiloissa nostaen T₀ lämpötiloja. Kokonaisuutena erinomainen transitiolämpötilasitkeys vaatii efektiivisen karkearaekoon ja vetypitoisuuden minimointia.

Dans de nombreuses applications technologiques des alliages métalliques structurés, la fragilisation par l'hydrogène (HE) est une préoccupation majeure car elle peut pénétrer dans la plupart des métaux, dégrader leurs propriétés et conduire à des défaillances prématurées. Malgré de nombreux efforts durant ces dernières décennies, au cours desquelles de nombreux mécanismes microscopiques ont été proposées, une compréhension claire des mécanismes de fragilisation H n'a pas encore été atteint. Depuis qu'il processus se produisent sur une échelle atomique, les mécanismes exacts conduisant à HE ne sont pas facilement identifiés expérimentalement. Une amélioration possible serait d'utiliser des simulations à l'échelle atomique pour essayer de capturer les détails des processus de déformation et de rupture au niveau atomique, permettant l'enquête du mécanisme microscopique pertinente. Dans un tel contexte, l'objectif de ce travail de thèse est de comprendre et de quantifier les interactions H avec des défauts comme postes vacants, les luxations et les fissures dans les métaux cfc par la modélisation multi-échelle. L'étude est organisé en quatre parties principales. Dans la première partie, nous avons utilisé des premiers calculs principe (basée sur la théorie fonctionnelle de la densité) pour décrire H interaction avec une vacance de Nickel. Plus précisément, les énergies de ségrégation de plusieurs atomes d'hydrogène en un seul et di-postes vacants ont été calculés. Deux énergies caractéristiques qui ont été trouvés à clarifier les pics expérimentaux observés chez Spectra désorption thermique dans la littérature. Les concentrations de groupes H-inoccupation équilibre a ensuite été évalué, dans des conditions pertinentes à SE et de corrosion sous contrainte (CSC) des alliages à base de Ni (industrie nucléaire), par des simulations de Monte Carlo et un modèle thermodynamique développé à partir de nos données DFT. Dans la deuxième partie, nous avons quantifié l'effet de piégeage de postes vacants sur H diffusion dans Nickel. Avec DFT barrières de saut calculé, liés à H piégeage et dépiégeage dans les postes vacants, nous avons utilisé accéléré Kinetic Monte Carlo (KMC) simulations pour évaluer le coefficient H de diffusion en fonction de la concentration de vacance et de la température. Dans la troisième partie, nous avons étudié la diffusion des grappes H-vacance de Ni, basée sur la combinaison de DFT et une méthode statistique. Calculs DFT de barrières de saut d'inoccupation ont été effectuées pour les clusters contenant de un à six H à l'intérieur du poste vacant. Avec ces barrières calculés et les concentrations calculées précédentes de grappes H-inoccupation, un modèle simple stochastique similaire à la procédure KMC a été développé pour estimer le coefficient de diffusion de grappes H-inoccupation en fonction de la concentration et de la température M. Dans la dernière partie, nous avons étudié l'interaction de l'hydrogène avec une pointe de fissure émoussée en aluminium par EAM combinée (potentiel interatomique semi-empirique) et calculs DFT. Méthode atome embarqué (EAM) simulations potentiels ont été réalisées pour évaluer l'effet de H sur la dislocation émission d'une pointe de fissure émoussée en mode mixte chargement. Ce phénomène peut être comprise par le changement induit H de l'empilement instable faute de l'énergie (γus) dans le modèle de Rice. Par conséquent, DFT et EAM calculs de γus ont été effectuées, y compris les effets de H et des charges en mode mixte. Il est montré que l'effet de la charge perpendiculaire au plan de glissement est très forte, contrairement à l'effet de la surface sous-H, ce qui est négligeable. / In many technological applications of structured metallic alloys, hydrogen embrittlement (HE) is a major concern as it can penetrate in most metals, degrade their properties and lead to premature failures. Despite numerous efforts in the past decades during which many microscopic mechanisms were proposed, a clear understanding of H embrittlement mechanisms has not been achieved yet. Since HE processes occur on an atomic-scale, the exact mechanisms leading to HE are not easily identified experimentally. One possible improvement would be to use atomic-scale simulations to try to capture details of deformation and fracture processes at the atomic level, enabling the investigation of relevant microscopic mechanism. In such context, the goal of this PhD work is to understand and quantify H interactions with defects like vacancies, dislocations and cracks in fcc metals through multi-scale modeling. The study is organized in four main parts. In the first part, we employed first principle calculations (based on density functional theory) to describe H interaction with a vacancy in Nickel. More specifically, the segregation energies of multiple H atoms in a single and di-vacancies were computed. Two characteristic energies were found which clarify the experimental peaks observed in Thermal Desorption Spectra in the literature. The equilibrium concentrations of H-vacancy clusters was then evaluated, under conditions relevant to HE and stress corrosion cracking (SCC) of Ni based alloys (nuclear industry),by Monte Carlo simulations and a thermodynamic model developed from our DFT data. In the second part, we quantified the trapping effect of vacancies on H diffusion in Nickel. With DFT computed jump barriers, related to H trapping and detrapping in vacancies, we employed accelerated Kinetic Monte Carlo (KMC) simulations to evaluate the H diffusion coefficient as a function of vacancy concentration and temperature. In the third part, we studied the diffusion of H-vacancy clusters in Ni, based on the combination of DFT and a statistical method. DFT calculations of vacancy jump barriers were performed for clusters containing from one to six H inside the vacancy. With these computed barriers and previous calculated concentrations of H-vacancy clusters, a simple stochastic model similar to the KMC procedure was developed to estimate the diffusion coefficient of H-vacancy clusters as a function of H concentration and temperature. In the last part, we studied the interaction of hydrogen with a blunted crack tip in Aluminum by combined EAM (semi-empirical interatomic potential) and DFT calculations. Embedded atom method (EAM) potential simulations were performed to evaluate the H effect on dislocation emission from a blunted crack tip under mixed mode loading. This phenomenon can be understood by the H induced change of the unstable stacking fault energy (γus ) in Rice’s model. Therefore, DFT and EAM calculations of γus were performed including the effects of H and of the mixed mode loads. It is shown that the effect of the load perpendicular to the glide plane is very strong, contrary to the effect of sub-surface H, which is negligible

L'objectif de ce travail est de prédire les risques de fissuration par H2S d'aciers au carbone à très haute limite d'élasticité utilisés en environnement pétrolier. Nous nous intéressons aux conditions de sévérité modérées, caractérisées par des teneurs faibles mais non nulles en H2S dans le milieu corrosif. Dans ces environnements, le chargement en hydrogène dans les aciers est lent, et les temps d'incubation avant apparition de fissures internes de type HIC peuvent donc être longs. En complément des essais standards de tenue à la fissuration des aciers, il est donc important de pouvoir disposer de méthodes rapides et fiables d'évaluation des risques de fissuration. La principale technique expérimentale mise en oeuvre est la perméation électrochimique d'hydrogène. Elle est mise à profit afin d'étudier le chargement, la diffusion et le piégeage de l'hydrogène dans différents aciers à haute limite d'élasticité en environnement faiblement concentré en H2S. En complément des essais de perméation, nous effectuons une étude de la fissuration assistée par l'environnement dans différentes conditions de pH et de PH2S. Ces essais permettent d'abord de vérifier l'évolution de la cinétique de fissuration interne en fonction des conditions environnementales. Dans le même temps, nous réalisons des mesures de la quantité d'hydrogène absorbée par le métal, en distinguant les parts d'hydrogène libre, piégé réversiblement et irréversiblement. La confrontation entre les données de diffusion et de piégeage issues des essais de perméation, et des données directes de fissuration et de quantités d'hydrogène absorbé permet l'établissement de seuils de concentrations critiques en hydrogène. Ces seuils sont ensuite utilisés afin de mettre en place une approche empirique prédictive de la fissuration HIC dans les aciers / The work of this thesis applies to flexible pipelines, which are used as risers or flowlines in offshore production. More specifically, it is dedicated to study the risks of hydrogen embrittlement of carbon steel wires used as armors if flexible pipes. The study is focused on low H2S containing medium inducing slow hydrogen entry in steels, thus, potentially long incubation times before embrittlement becomes effective. The first goal of this work is to study the entry, diffusion and trapping of hydrogen into different grades of carbon steel in low H2S concentration environment. The permeation technique will be used, as well as immersion tests of variable duration with characterization of the crack process. Analysis of permeation transients gives information on hydrogen diffusion and hydrogen trapping into steels in function of experimental conditions (pH, PH2S). Immersion tests will be performed complementary to permeation tests in various conditions of pH and PH2S. We verify the cracking evolution with environmental conditions. We also quantify the lattice hydrogen, the reversibly and irreversibly trapped hydrogen absorbed by the materials. Then, using both diffusion properties from permeation tests and cracking data from immersion experiments, we set up a predictive approach to link the risks of H2S cracking for each steel grade in function of the severity of the environment. Applying this method should also allow to establish faster and more reliable comparisons of the hydrogen induced cracking resistance of different steel grades. It should be used as complimentary tool for qualification purposes

Slat-band chains are used as conveyors by the food industry, breweries and bottling plants. The operating conditions require abrasion resistance and strength which are at the limit of the capabilities of the current material of choice, cold worked type 430. In an unconventional way of strengthening this material, Mintek developed a process in which the cold worked material is aged between 450°C and 500°C. The present work aims to elucidate the strengthening mechanism, using type 430 stainless steel containing 16.42% Cr and 0.036% C, in the cold-rolled condition (38% reduction in area), with and without prior solution heat treatment. The Cr-rich precipitate α" may form in the 450°C to 500°C range (due to the miscibility gap in the Fe-Cr system), resulting in the increased hardness and lowered ductility. Mossbauer studies confirmed that the α", at this composition and temperature, forms through the process of nucleation and growth. Hardening due to α" precipitation was only observed after aging for 64 hours or more, however. After increasing the dissolved interstitial content by solution heat treatments (in the vicinity of 900°C), increases in Vickers hardness of 30-50 kg/mm2 could be obtained after only 8 minutes at 475°C. This hardness increase corresponds to an increase in tensile strength of more than 100 MPa. The increased hardness does not appear to be caused by strain aging, and presumably results from fine carbide or nitride precipitation. Solution treatment at 930°C also introduced some martensite (α') into the microstructure, which raised the hardness of the unaged cold worked material. Overaging of the carbide and nitride precipitates was observed at 475°C, but not at 450°C, probably due to the lower diffusion rates at the lower temperature. No averaging of the α" precipitates occurred, for aging times up to 2072 hours. Samples aged for selected periods of time at 475°C had low impact strengths - even well before the formation of α" - and revealed predominantly cleavage fracture with some ductile fracture areas, mostly at grain boundaries. Both impact strength and lateral expansion indicated that embrittlement accompanies the increased hardness obtained by aging. Calculation of critical crack lengths from the impact data, however, revealed that a maximum flaw length of 0.8 mm, for specimens solution treated at 880°C, could be tolerated before catastrophic failure. Since it is not expected that flaws of that size would exist in the as manufactured links, fatigue will probably determine the lifetime of the chains, although the lower K1c values indicate that less crack propagation will be tolerated before brittle fracture. During the aging treatment, the strength may be lowered by recrystallisation of the coldworked material. Transmission electron microscopy (TEM) revealed the start of recovery, but no recrystallisation. Some large precipitates (around lμm in diameter) were present. These were identified, through their diffraction patterns, as M23C6; these carbides were present in both aged and unaged material and hence represent precipitates which had not dissolved during the initial solution treatments. The α" precipitates- and the presumed newly formed nitride and carbide precipitates - were too fine for detection by TEM. Potentiodynamic testing of the treated material in a 0.5M H2SO4 solution indicated that, although the probable hardening mechanisms imply localised Cr depletion of the matrix, the general corrosion resistance and passivation behaviour were not affected. It is concluded that the strength of the chain may be increased markedly by short-term heat treatments at 475°C, with lowered toughness, but with no decrease in corrosion resistance. Martensite, work hardening, and precipitation of carbides and nitrides all contribute to the final strength, with α" formation only becoming significant after longer aging times. / Dissertation (MEng)--University of Pretoria, 2013. / Materials Science and Metallurgical Engineering / unrestricted

During high pressure pulsation tests repeated cracks of CP3. housings occur. High pressure pump CP3 is part of injection system Common Rail. Cracks can be caused by segregation of phosphorous from a washing medium on grain boundaries during subsequent heat treatment process. The present master´s thesis focuses on analysis of phosphorous compounds on specimen surface made from steel 20MnCr5, which were soaked in solutions with different concentration of phosphorous. Master’s thesis is also focused on potencial influence of phosphorous on structure and properties of this material. The first part of the thesis provides a summary of current knowledge in the studied field, experimental results can be found in the second part. The surface analysis XPS, the light microscopy, the scanning electron microscopy equipped by energy dispersive chemical analysis, the glow discharge optical emission spectroscopy, the methods of the phase analysis by means of the X-ray diffraction and impact test were used.

Les matrices en carbure de Tungstène et Cobalt (WC‐Co) sont utilisées dans les procédés de frappe à froid de l’acier pour leur exceptionnelle capacité à résister aux phénomènes d’usure. Ces travaux ont pour objectif de mieux comprendre les mécanismes complexes qui entrainent finalement la dérive des matrices vers l’état usé. Cette complexité vient des liens étroits entre la microstructure et les propriétés mécaniques macroscopiques de ces matériaux. Pour la compréhension des mécanismes de dérive vers l’usure, une stratégie de travail en quatre étapes est établie. La première étape est le prélèvement de matrices de frappe, avec différentes durées de vie, directement sur la chaîne de production. La deuxième étape est l’identification de la rhéologie. Elle s’accompagne de la modélisation numérique du procédé de frappe pour calculer le champ des contraintes et des déformations plastiques. La troisième étape est la caractérisation localisée de l’évolution de la surface selon trois axes : les propriétés tribologiques, morphologiques, et mécaniques. On quantifie ainsi la dégradation progressive des conditions de contact corrélée avec une fragilisation des surfaces et la décohésion de grains de carbures WC. Pour comprendre les mécanismes qui conduisent à la décohésion de grains, une stratégie de modélisation numérique à l'échelle mésomécaniques 2D est mise en place. L’énergie de rupture entre un grain et le reste du matériau est modélisée par des éléments cohésifs. Ces modèles montrent que la sensibilité de chaque grain à l’arrachement dépend non seulement des conditions de contact et de la ténacité du matériau, mais également de la taille et de la configuration du grain au voisinage de la surface. / Tungsten carbide and Cobalt (WC‐Co) dies are used for cold forming processes of steel because of their exceptional performances in resisting wear phenomena. This work aims to a better understanding of the complex damage mechanisms that eventually cause wear. This complexity comes from the existing relationships between their microstructure and their macroscopic mechanical properties. For a better understanding of the damage mechanisms leading towards wear, a four‐step strategy is presented. The first step is the cold heading dies sampling directly on the production line. They are collected at different lifetimes. The second step is the identification of the die rheology. It is followed by numerical modeling of the forging process to calculate the stress field and plastic strain magnitude. The third step is to characterize the local evolution of the surface properties along three axes: the tribological, the morphological and mechanical aspects. These analyses quantify the progressive decrease of contact conditions correlated with surface embrittlement and WC carbide grains debonding. To understand the mechanisms that lead to the grains debonding, a set of 2D mesoscale contact models are performed. The fracture energy between a WC grain and the rest of the material is computed using cohesive elements. These models show that the sensitivity to debonding depends not only on the contact conditions and the material fracture toughness, but also on the grain size and grain configuration in the vicinity of the surface.

In this work, Slow Strain Rate Test (SSRT) testing, Light Optical Microscopy (LOM) and Scanning Electron Microscopy (SEM) were used to study the effect of micro-structure, corrosive environments and cathodic polarisation on stress corrosion cracking (SCC) of two grades of high strength steels, Type A and Type B. Type A is manufactured by quench and tempered (Q&T) method. Type B, a normalize steel was used as reference. This study also supports electrochemical polarisation resistance method as an effective testing technique for measuring the uniform corrosion rate. SSRT samples were chosen from base metal, weld metal and Heat Affected Zone (HAZ). SSRT tests were performed at room temperature under free corrosion potential and cathodic polarisation using 4 mA/cm2 in 1 wt% and 3.5 wt% NaCl solutions. From the obtained corrosion rate measurements performed in 1 wt% and 3.5 wt% NaCl solutions it was observed that increased chloride concentration and dissolved oxygen content enhanced the uniform corrosion for all tested materials. Moreover, the obtained results from SSRT tests demonstrate that both Q&T and normalized steels were not susceptible to SCC in certain strain rate(1×10-6s-1) in 1 wt% and 3.5 wt% NaCl solutions under free corrosion potential. It was con-firmed by a ductile fracture mode and high reduction in area. The weld metal of Type A with acicular ferrite (AF), pro-eutectoid (PF) and bainite microstructure showed higher susceptibility to hydrogen assisted stress corrosion cracking compared to base metal and HAZ. In addition, typical brittle intergranular cracking with small reduction in area was observed on the fracture surface of the Type A due to hydrogen charging.