2.4.1 Materials Susceptible to Mechanical Fatigue
Fatigue design curves (magnitude of cyclic stress against logarithmic scale of cycles to failure, S-N curves) have been established for different materials or material classes used in nuclear power plants (NPPs). The basis for the S-N curve in air for carbon steels, low alloy steels and austenitic stainless steels in the US codes and standards [NUREG-CR-6909] are experimental data from various establishments and research institutions (including Argonne National Laboratory, General Electric and Japan Nuclear Energy Safety Organisation) performed with different types of steels. The following material classes and their explicit design curves can be found in codes and standards:
- Carbon Steel (CS);
- Low Alloy Steel (LAS, such as A508Cl3 and A533-B);
- Austenitic Stainless Steel (ASS, such as AISI 304, AISI 304L, AISI 316, AISI321 and AISI347), may also be used for Ni-Cr-Fe Alloys (such as Alloy600).
In general, an approach as used in [NUREG-CR-6909] can be applied to build design S-N curves, in which a mean through the experimental fatigue test data is fitted and then the curve is shifted by the most conservative of a factor of 2 on stress or 20 on cycles at each alternating stress. In the case of the ASME mean curves, the best-fit curve through the experimental data represents the number of cycles needed to develop a 3-mm crack corresponding to the size of specimens and to the criterion of a 25% decrease of the tensile stress from its peak or steady state value, which was used to define fatigue life for the experiments. The adjusting factors (2 and 20) take into account effects and uncertainties like data scatter, size effects, surface finish and atmosphere [NUREG-CR-6909]. As a result of a review of literature data and Monte Carlo simulations, new fatigue design curves (Argonne National Laboratory model curves) have been developed for carbon steels, low alloy steels and austenitic steels with new adjusting factors (2 and 12) to reduce conservatism [NUREG-CR-6909]. These adjusting factors are accepted by US regulators [NRC-RG-1.207] but even higher factors can be used for carbon and low alloy steels. Different mean curves and adjusting factors have been applied for example in German codes and standards for the austenitic stainless steels AISI 321 and AISI 347, i.e. a factor of 1.88 on stress for temperatures <= 80°C and 1.79 for temperatures > 80 °C and a factor of 12 on cycles for all temperatures [KTA3201.2] [KTA3211.2].
An overview of fatigue S-N data for carbon steels, low alloy steels and austenitic stainless steels can be obtained from different reports, where although the environmental influence on fatigue is discussed, include data for mechanical fatigue in air and describe some dependencies [NUREG-CR-6583] [NUREG-CR-6787] [NUREG-CR-6909] [NUREG-CR-6909-1].
In general the fatigue life of carbon steels and low allow steels is comparable at less than 5x105 cycles, whereas the latter performs better at 106 cycles. The material composition affects the influence of strain rate on the fatigue life. The lives of welded materials of both types are slightly lower than for non-welded materials [NUREG-CR-6583] [NUREG-CR-6909-1].
The fatigue life of austenitic stainless steels AISI 304, AISI 304L and AISI 316 is comparable. Different behaviour was observed between weld material and the cast or wrought material. Austenitic stainless steels of the type AISI 321 or AISI 347 were not included in the data for the fatigue design curve [NUREG-CR-6787] [NUREG-CR-6909-1] but can be found for example in separate KTA design curves [KTA3201.2] [KTA3211.2] because of the common use of these materials in German NPPs.
The European Coal and Steel Community (now RFCS) and EC-funded research projects within the timeframe of FP4 to FP7 (including 7210-MA/131, 7210-MA/823, 7210-MA/951, 7210-PR/303, GRETE and NET) were focused on austenitic stainless steels (ASS).
2.4.2 Austenitic Stainless Steels
The experimental data used for the fatigue design curve indicate that fatigue lives of AISI 304, AISI 304L and AISI 316 are comparable and no significant different behaviour could be observed between cast or wrought material. However, weld material of austenitic stainless steels exhibited lower fatigue lives in the low cycle fatigue regime (<104 cycles) and generally longer in the high cycle regime [NUREG-CR-6909].
Experiments with AISI 316 have been performed in [JRC8609] at different temperatures (373, 473, 573 K) with unirradiated material and under simultaneous proton irradiation (6x10-7 dpa s-1, fluence < 1 dpa). The applied load was kept at approximately 190 MPa with a load ratio of 0.1 and a sinusoidal waveform with a frequency of 1 Hz.
Experiments with 20 % cold-worked and annealed AISI 316L material at 400 °C and under simultaneous deuteron irradiation conditions (1 x 10-6 dpa s-1 and 6 x 10-6 dpa s-1) have been performed in [JRC12412] [JRC16627]. Under irradiation conditions, different hold times (20 s, 50 s, 1000 s) at the minimum strain value have been applied to promote irradiation creep. Control tests under similar conditions but without irradiation were performed to compare the in-beam and out-of-beam results.
AISI 304L, AISI 321 and AISI 347 have been investigated in the GRETE project [EUR22282] to evaluate NDE (non-destructive examination) techniques. The AISI 304L was tested in the “as-received” and in the “cold-worked” condition (33 % reduction of thickness), whereas for the AISI 321 and AISI 347 material only the “as-received” condition was investigated. The AISI 321 steel was the only one which revealed a significant increase of martensitic phase under cyclic load.
Welding and weld toe improvement techniques for AISI 304L as well as other stainless steels, i.e. duplex steel S31803 (1.4462) and austenitic Cr-Mn steel (1.4376), were investigated in the project 7210-PR/303 funded by the RFCS [EUR22809]. Recommendations could be derived for welding and weld toe improvement techniques for AISI 304L steel as well as the other materials. Within preceding RFCS projects 7210-MA/131, 7210-MA/823 and 7210-MA/951, AISI 304L steel as well as the welding material AISI 308L have been investigated [EUR19972]. Fatigue tests have been performed for the base metal as well as the heat affected zone (HAZ) and the welds themselves.