Environmental Fatigue Testing of Type 304L Stainless Steel U-Bends in Simulated PWR Primary Water

Abstract

Environmental fatigue testing of small-scale austenitic stainless steel components under simulated pressurized water reactor (PWR) operating conditions was sponsored by the EPRI Materials Reliability Program (MRP) Fatigue Issue Task Group and the U.S. Department of Energy to fill in some important gaps in the knowledge base. An analysis and assessment of existing fatigue data for stainless steel exposed to the PWR primary environment identified a lack of data with respect to flow rate effects. The majority of existing data has been gained under static or quasi-static flow conditions, where the tendency to environmental enhancement of cyclic crack growth is generally expected to increase. However plant experience — where high-flow conditions prevail — shows significantly lower susceptibility to corrosion fatigue cracking. The main objective of the present test program was the identification of flow-rate effects on the initiation and growth of low-cycle corrosion fatigue (LCF) cracks in cold-drawn, 304L stainless steel tube U-bend specimens undergoing cyclic loading and simultaneous exposure to simulated PWR primary water on the inside of the tube. Use was made of a pre-existing facility at F-ANP in Germany where the experimental concept had been previously tested on carbon-steel U-bends. The test equipment was set up so as to allow direct comparison of results obtained under quasi-stagnant conditions (∼0.005 m/s flow rate at the internal surface) with relatively high-flow conditions (∼ 2.2 m/s flow rate). In accordance with literature data, PWR primary water was shown to have a significant effect on the high-temperature fatigue behavior of the bends, as demonstrated by the observed change in failure mode and location of cracking between baseline specimens (tested in nitrogen) and those bends exposed to simulated reactor coolant. Metallography and fractography suggest that the environment is acting by affecting both crack initiation and crack growth. In contrast to the situation for carbon steel, no beneficial effect of higher flow rate on the number of cycles to failure (leakage) was observed. The paper discusses further details of the environmental fatigue data obtained, including the effects of strain amplitude, strain rate and surface condition. It also provides a comparison of test results with the current ASME Section III fatigue curves.