13th International Conference on Fracture June 16–21, 2013, Beijing, China Furthermore, additional experiments allow the authors to show experimental evidences of coupling between corrosion and cyclic loading even under ultrasonic loading frequency. 2. Material and experimental conditions The material studied hereafter is a non-standard hot rolled low alloy steel grade (named R5 according to the International Classification Societies of Off-shore Systems) with a typical fine grain microstructure composed by tempered martensite and bainite as illustrated in [12]. Its chemical composition is shown in Table 1. This steel is used after a double quenching in water at 920°C then 880°C and tempering at 650°C with water cooling. After this heat treatment its mechanical properties are: hardness 317 HB, yield strength 970 MPa, UTS=1018 MPa, Young modulus E=211 GPa. It has to be noticed that this steel is vacuum degassed with low hydrogen content (1 ppm maximum in the liquid steel after the vacuum treatment). Furthermore, very low non-metallic inclusions are present in this steel (see details in [12]). Table 1: composition of R5 steel (% weight, Fe balance) C Mn Si P S Cr Ni Mo V O 0.22 1.22 0.3 0.009 0.003 1.07 1.07 0.5 0.09 12 2.1. Fatigue test conditions and specimen geometry 2.1.1. Testing machine and specimen geometry All the fatigue tests (crack initiation and crack growth) presented hereafter were carried out with an ultrasonic fatigue testing machine [1] operating continuously at 20 kHz (no pulse loading) under fully reversed tension (R= -1) (see [1, 12] for details). Figure 1 shows the dimensions of the two types of specimens used for (i) VHCF crack initiation tests and (ii) fatigue crack growth (FCG) tests. The geometry of the FCG specimens was designed according to the works of Wu [13] and Sun [14]. The crack growth was measured with an optical binocular microscope with a maximum magnification ×200. The roughness of the tested area of the VHCF specimens was Ra=0.6 µm and a few additional specimens with a better roughness (Ra=0.1 µm) were tested too for investigating the effect of roughness. The VHCF specimens were tested under three different conditions: (i) without any corrosion (virgin state), (ii) after pre-corrosion and (iii) under real time artificial sea water flow. Figure 1: Specimen geometry for (a) VHCF tests, Kt=1.02 in tension, (b) crack growth tests (dimensions in mm). All the VHCF tests were calibrated by using a wide band (0 to 100 kHz) strain gauge conditioner and a longitudinal strain gauge glued on the specimen surface. Such calibration allows the authors to be certain of the local strain (elastic stress) amplitude and mean value in the narrowest cross section of the specimen. These tests were carried out until a decrease of the resonance frequency of 0.5 kHz that indicates the presence of a macrocrack. 2.1.2. Corrosion of the specimens The pre-corrosion of the specimens was done according to the ASTM G85 standard: 600 hours in a salt fog corrosion chamber under temperature and humidity control (35°C with 95% of humidity). The salt solution contained 5% of NaCl, its pH was 6.6 and it was applied in the chamber with a
RkJQdWJsaXNoZXIy MjM0NDE=