To improve the in-service behaviour of Cr Mo (V) steel grades used for the pressure vessels operating in hydrogen environment at high temperature for the oil industry, the manufacture of heavy forgings needs a high quality. Improvement of the standard and enhanced strength (ASME Case 1960) 2 l/4 Cr 1 Mo steel grades may be achieved by reducing drastically the impurities (S, P, X, etc...) to extra low level and avoiding segregates at the inner surface of the shells. For high temperature operation, new V modified steel grades are proposed (ASME case 1961, Code Case 1973 and Code Case 2098). Their conventional mechanical properties are similar to those of enhanced strength 2 1/4 Cr 1 Mo but they offer higher creep properties and improved resistance to hydrogen damage.

The objective of the research, through both an experimental and a modelling approach, was to determine the parameters controlling the nitrogen level in a weld. A second objective was to study the relationship between the weld microstructure and the corrosion properties. More particularly, the potential interest of microelectrode techniques has been investigated. TIG welding has been investigated through both an experimental and a modelling approach. TIG and A TIG tests have confirmed tnat it is necessary to add nitrogen in the shielding gas in order to prevent nitrogen loss during welding. For duplex stainless steel, 2.5 % nitrogen in the shielding gas is sufficient, whereas for high nitrogen content austenitic stainless steels higher levels are necessary. It has also been shown that, for a given grade, the nitrogen content increases when the penetration increases. Penetration depends on the material composition, with a beneficial effect of surface active elements (0, S, etc.). The model developed was based on the nitrogen exchange between the plasma, the weld pool and the shielding gas. It was first developed to describe nitrogen evolution during a stationary arc situation. The results were in good agreement with experiments. The model was then adapted to the case of welding with an active flux. An attempt was made to describe the traveling arc situation. However, some improvements are still necessary. Pitting corrosion tests have confirmed the influence of nitrogen content on the corrosion sensitivity of TIG welds. Microelectrode techniques have been used to characterise the local corrosion behaviour of welds. It has been shown that the scanning vibrating electrode technique was of limited utility to study corrosion resistance of highly alloyed stainless steels. More promising results have been obtained with microcapillary technique which make local electrochemical measurements possible. Finally, MIG tests have been performed in order to study the influence of the shielding gas composition on nitrogen content in the weld and also on the formation of porosities. For superduplex stainless steel, it has been demonstrated that nitrogen must be added in the gas to prevent nitrogen loss. It has also been shown that the number of porosities in the weld depends on the C02 content in the gas and not on the nitrogen content.

Duplex stainless steels become more and more attractive, both from technical and economical points of view. However, use of duplex stainless steels is limited to 300 °C. Metallurgical transformation and phase precipitations can affect the mechanical and corrosion properties over long-term exposure at elevated temperatures. In some cases, however, it can be interesting to submit a duplex stainless steel to a thermal cycle. In this work the influence of short thermal cycles on the properties of duplex stainless steel has been studied. Four duplex stainless steel grades have been selected: two superduplex grades, one standard grade and one lean duplex. The influence of thermal cycles has been studied on base materials and welds. The influence of plastic deformation has also been investigated. The influences of thermal cycles have been investigated on toughness and corrosion properties. The microstructural evolutions have also been studied. Results show that thermal cycles can affect toughness properties and corrosion properties in different ways. For example, at the lower temperature, toughness can be strongly reduced while localised corrosion resistance is not affected. At the higher temperature, both the corrosion resistance and the toughness can be significantly reduced. This is because of the microstructural evolution in the steels. The possible influence of thermal cycles on stress relaxation of welded structures has been studied experimentally and with the help of thermal simulation. It has been demonstrated that rate of stress relaxation up to 70 % can be achieved. The results of this work can be used to determine whether thermal cycles are suitable for a particular application. The thermal cycles acceptable depend on the base material, on the weld technique used and on the material specification.