Materials Science Forum Vols. 595-598

Abstract: In the frame of its research work on nuclear fuel safety, the French “Institut de Radioprotection et de Sûreté Nucléaire” (IRSN) has highlighted the importance of cladding tube oxidation on its thermomechanical behavior. The occurrence of radial cracking and spallation has been observed as the main mechanisms for the zirconia layer degradation during transient experiments. A study of these two mechanisms has been jointly launched by IRSN and Areva-NP. Thus laboratory air oxidations of fully recrystallized or stress-relieved low-tin Zircaloy-4 cladding tubes have been performed. Representative oxide layer thicknesses varying from 10 to 100 0m have been obtained. SEM micrographs of the obtained oxidised samples show that short circumferential cracks are periodically distributed in the oxide thickness. For specimens with oxide film thickness greater than 30 0m, radial cracks are initiated from the outer surface of the oxide layer and propagated radially. Veins characterised by the lack of circumferentially orientated crack are evidenced. All these phenomena are mainly linked to high compressive stress levels in the zirconia layer. A model describing the stress evolution in the oxide and in the cladding has been developed. This model takes into account the influence of elasticity, cladding creep, oxide growth and thermal expansion. Deflection tests data [15] are used to calibrate the oxide growth modelling. The model enables the evaluation of strain or stress profile in the oxide layer and in the base metal. Numerical results are in good agreement with a large set of axial and circumferential strains measurements. Further a better understanding of cracking mechanisms is achieved considering the good agreement between experimental and numerical analysis.

Abstract: Above a given temperature called TA, the chromium rich oxide which has been developed on the surface of Haynes 230® and model NiCrWC alloys at a lower temperature becomes unstable in impure helium: carbon monoxide is released. Actually, oxide is reduced by carbon from the alloy. A thermodynamic model is developed to rationalize the variation of TA as a function of the partial pressure of CO in the gas phase. It was found that, at the early stages of the scale reduction, the relevant reaction occurs at the oxide/metal interface between chromia and carbon from the alloy. The interfacial activity of carbon in the alloy can be calculated based on measurements of the interfacial weight percentage of chromium and using ThermoCalc® software. Excellent agreement is observed between experimental values of TA and theoretical predictions.

Abstract: Nickel base alloys Haynes 230 and Inconel 617 are of interest for gas cooled reactors. At high temperature in impure helium, they generally form surface chromium-rich oxides. However above a critical temperature called TA, the scales are not stable anymore and the chromia destruction comes with a production of carbon monoxide. Reactivity tests on model alloys, with and without carbon, prove that chromia is reduced by the carbon from the alloy. TA vs P(CO) curves were also plotted for the two commercial alloys based on the experimental determination of TA in various atmospheres with increasing partial pressures of carbon monoxide. Unexpectedly, both materials exhibit an almost identical behavior although a basic equilibrium approach suggests that the chromia scale would be reduced in different conditions due to the thermodynamic particularity of the interfacial alloy/scale system.

Abstract: Two Ni-Fe-Cr ternary alloys have been oxidized in simulated pressurized water reactor primary water at 360°C for 1000 h. The chemical composition of those alloys were chosen in order to be representative of the one of chromium depleted areas under the oxide scale of industrial alloys (e.g. alloy 600) exposed in the same conditions. The resulting oxidized structures (corrosion scale and underlying metal) were characterized using complementary analytical methods (FEG-SEM, TEM, SIMS, optical microscopy). On the one hand, the characterized external oxide layer is very close to the one observed on industrial nickel-base alloys, hence validating the use of such model alloys. On the other hand, both free oxygen and oxides have been detected at grain boundaries several micrometers under the metal/oxide interface. Implications of such a finding on the involved transport mechanisms for oxygen and the intergranular stress corrosion cracking resistance of nickel-base alloys are then discussed.

Abstract: To get a better understanding of oxidation behavior of Ni-base alloys in PWR primary water, a numerical model for oxide scale growth has been developed. The final aim of the model is to estimate the effects of possible changes of experimental conditions. Hence, our model has not been restricted by the classical hypothesis of quasi-steady state and can consider transient stages. The model calculates the chemical species concentration profiles, but also the vacancy concentration profiles evolution in the oxide and in the metal as a function of time. It treats the elimination of the possible supersaturated vacancies formed at the metal/oxide interface by introducing a dislocation density at the interface and in the metal bulk. This latter density can be related to the cold-working state. Its effect on the vacancy profile evolution is studied in the case of a pure metal. Eventually an extension of the present model to the oxidation of Ni-base alloys is discussed regarding a recent vacancy diffusion model adjusted on Ni-base alloys.

Abstract: The knowledge of the quinary Pb–Bi–O–Fe–Hg is necessary for understanding the degradation mechanisms of the T91 steel used as structural material in future ADS nuclear reactors. In this device, the steel will be in direct contact with the liquid spallation target (which is constituted by lead or lead-bismuth eutectic) surrounded by a reduced oxygen pressure atmosphere. In the present work, the characterization of the pseudo-binary PbO–Fe2O3 cut has been performed. In order to complete the available data in the literature, some experimental investigations by DTA, isothermal annealing, SEM and EPMA have been done. These results have allowed proposing a thermodynamic assessment using the Calphad method by the ThermoCalc software.

Abstract: Resistance to corrosion of the structural materials is a key factor for nuclear applications that use molten fluorides. Low chromium, nickel-base alloys are regarded as the most suitable metallic materials. In a first approach, corrosion of some pure metallic constituents Ni, Mo, W and Fe, was studied by electrochemical techniques. Linear voltammetry was applied in LiF-NaF and LiF-AlF3, in the temperature range 900-1100°C. The relative stability of the metals in LiF-NaF is established. To determine the corrosion current density, three methods are presented, two based on the Tafel extrapolation method and the third one being the polarization resistance method. Results regarding corrosion rates are compared. Two corrosion behaviors are observed: on the one side, Ni, Mo and W and on the other side Fe. The difference might come either from different corrosion mechanisms or from a different number of exchanged electrons. The corrosion rate increases with temperature following the Arrhenius law. However, further experiments are needed in order to identify the key parameters that influence the corrosion in the different melts.

Abstract: Due to the specific in service VHTR conditions, the corrosion behaviour of Inconel 617, candidate alloy for the IHX design, has been investigated at elevated temperatures in representative helium containing impurities (CO, H2O, H2 and CH4) in the range of 'bar. The role of Al in the corrosion behaviour of IN617 is investigated using Ni-22Cr-9Mo base model alloys containing different Al levels (from 0 to 2wt.%).

Abstract: Nickel base alloys 617 and 230 are promising candidates for the Intermediate Heat eXchanger (IHX) of GenIV Very High Temperature gas cooled Reactors. The capability to maintain an oxide layer as an efficient barrier against corrosion under mechanical loading is investigated through SEM in situ tensile test. The mechanical properties of external oxide layers are so compared between the two alloys. Cracks and spallation are observed. Few differences could be observed between these two alloys when pre oxidized in impure helium.

Abstract: HAYNES® 230® and 617 alloys are competing for use on Generation IV, high temperature gas cooled reactor components because of good high temperature creep strength in the temperature range between 760°C and 982°C and resistance to oxidation in the gas cooled reactor environment. A review of the metallurgy affecting the properties in each alloy will be discussed. Grain size and carbide precipitation developed during fabrication effect short term and long term ductility, fatigue, and creep. For example, 230 alloy has a finer grained structure which promotes fatigue strength with a slight sacrifice in creep strength. The 617 alloy has a coarser grain structure which provides slightly higher creep resistance while sacrificing some fatigue strength. Thermal aging also introduces gamma prime precipitation to the 617 alloy as well as grain boundary carbides, and this, in addition to grain boundary oxidation, reduces the low cycle fatigue strength of 617 alloy compared to 230 alloy. Independent studies have shown that 230 alloy possesses higher resistance to thermal fatigue than 617 alloy. However, welds of both base metals with similar weld composition have about the same thermal fatigue life. Cooling rates from solution annealing temperatures during processing effect the ductility and creep strength of these alloys with the highest cooling rates preferred for retention of ductility and creep strength. The reason; slow cooling rates promote carbide precipitation in the grain boundaries which reduces ductility and creep strength.