The microstructure and tensile properties of mitrogen containing vacuum atomized alloy 690

While nickel-based alloys have been widely used for power plants due to corrosion resistance and good mechanical properties, during the last couple of decades, failures of nuclear components increased gradually. One of main degradation mechanisms was primary water stress corrosion cracking at dissimilar metal welds of piping and reactor head penetrations. In this context, precise estimation of welding effects became an important issue for ensuring reliability of them. The present study deals with a series of finite element analyses and crack growth rate evaluation of Alloys 690/152. Firstly, variation of residual stresses and equivalent plastic strains was simulated taking into account welding of a cylindrical block. Subsequently, extraction and pre-cracking of compact tension (CT) specimens were considered from different locations of the block. Finally, crack growth curves of the alloys and heat affected zone were developed based on analyses results combined with experimental data in references. Characteristics of crack growth behaviors were also discussed in relation to mechanical and fracture parameters.

The effects of filler metal (FM) composition on inclusions and inclusion defects for ER NiCrFe-7 weldments have been investigated and analyzed. Results show that as Al, Ti content in FM increases from 0.14 wt% Al, 0.30 wt% Ti to 0.42 wt% Al, 0.92 wt% Ti, the Al, Ti reduction will increase during welding. Inclusion defects (point-like defects named by welding workers) are prone to form in the high Al, Ti content weldments. Inclusion defects with Mg, Ca, Al, and Ti as major metallic elements have been found on the surface and interior of the weldments, as Al, Ti content in FM is over 0.29 wt% Al, 0.62 wt% Ti. Less Ti content in FM cannot prevent ductility-dip-cracking (DDC) through producing enough intragranular precipitates and lessening intergranular M₂₃C₆ precipitates. Nb can be used to replace Ti to reduce the sensitivity of the DDC in the NiCrFe-7 alloy weldments.

The segregation and precipitation behavior of Alloy 690 containing 0.001–0.11 wt% nitrogen during isothermal solidification at 1370 and 1355 °C have been investigated using optical microscopy (OM), electron probe microanalysis (EPMA) and transmission electron microscopy (TEM). The results indicate that the volume fraction of TiN-type nitride formed during isothermal solidification increases with the nitrogen content of Alloy 690. Segregation of Ti and Cr exists in samples solidified at 1370 and 1355 °C. The Ti content in the residual liquid markedly decreases and the concentration of Cr increases when the nitrogen content of Alloy 690 increases. Furthermore, N and S also show segregation to some extent in the residual liquids at 1355 °C. Accompanying by the segregation of Cr, Ti, C, N and S, sulfides and chromium nitrides form. In a low nitrogen content Alloy 690, sulfur segregates and precipitates in the form of Ti₄C₂S₂ and (Cr, Ti)S, but in the form of (Cr, Ti)S or CrS in a high nitrogen content Alloy 690. (Cr, Ti)N-type nitrides with an fcc crystal structure have been identified in a sample with 0.11 wt% nitrogen.

In this paper we compare and contrast the crack growth rate of a nickel-base superalloy (Alloy 690) in the Pressurized Water Reactor (PWR) environment. Over the last few years, a preponderance of test data has been gathered on both Alloy 690 thick plate and Alloy 690 tubing. The original model, essentially based on a small data set for thick plate, compensated for temperature, load ratio and stress-intensity range but did not compensate for the fatigue threshold of the material. As additional test data on both plate and tube product became available the model was gradually revised to account for threshold properties. Both the original and revised models generated acceptable results for data that were above 1 × 10⁻¹¹ m/s. However, the test data at the lower growth rates were over-predicted by the non-threshold model. Since the original model did not take the fatigue threshold into account, this model predicted no operating stress below which the material would effectively undergo fatigue crack growth. Because of an over-prediction of the growth rate below 1 × 10⁻¹¹ m/s, due to a combination of low stress, small crack size and long rise-time, the model in general leads to an under-prediction of the total available life of the components.

In this paper, the corrosion-fatigue data on Alloy 690 plate documented in the open literature is compared with test data obtained using steam generator tubing. The results on plate material documented in the open literature is not representative of actual behavior of steam generator tubing due to the mutually interactive influences of microstructural features, section thickness and processing variables. This paper also examines if response of steam generator tubing in a pressurized water reactor environment can be modeled using test data obtained for the material on plate stock. The fatigue crack growth rate data for alloy IN 690 used in actual steam generator tubing in low dissolved oxygen PWR environment was generated using a circumferentially through-wall cracked tube specimen. The laboratory test data is compared with the data published in the open literature using a modified corrosion-fatigue model. Test results reveal the tube material to have near similar fatigue crack growth rate behavior when compared to the plate material. This provides conclusion that the standard specimen geometry, such as compact tension specimen, can be used to characterize the response of the tube in adverse environments.

This paper describes the use of a crack growth rate superposition model. In this model, the total crack growth rate is taken to be the sum of the crack growth rate in air plus the local corrosion rate. For purpose of statistical data analysis, the fatigue threshold was incorporated into the model. The fatigue threshold for the chosen material (INCONEL 690) is defined and taken to be the crack length per unit cycle of 1.0 × 10⁻¹⁰ m/cycle. To account for threshold values in fatigue crack growth, Klesnil and Lukas developed a modified equation to the Paris Law. By incorporating the modified equation into the model developed and put forth by Argonne National Laboratory, the test data in the threshold region became statistically significant. A comparison of the model developed both with and without the threshold term in the environments of ambient temperature laboratory air and pressurized water reactor (PWR), or low dissolved oxygen (DO) (<10 ppb) water, is made. Results reveal that by incorporating threshold terms into the model, a better correlation between model projection and the actual experimental test data obtained on the material in the ambient temperature laboratory air environment are obtained. The results obtained from using the superposition model are presented and discussed.