PSI - Issue 13

Kohei Kishida et al. / Procedia Structural Integrity 13 (2018) 1032–1036 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

1033

2

1. Introduction

Non-propagation of a small fatigue crack is a key phenomenon controlling the fatigue limit of steels. The factors affecting the crack non-propagation limit in steels are crack closure and local hardening at the crack tip. In particular, the effect of local hardening at the crack tip in steels can be enhanced significantly by utilizing dynamic strain aging (DSA) (Koyama et al., 2017b; Oates and Wilson, 1964; Wilson and Tromans, 1970). In our previous studies (Li et al., 2016, 2017), the threshold stress intensity factor range for small crack growth in Fe – C fully ferritic alloys at room temperature was confirmed to increase with the increase in the content of supersaturated solute carbon via water quenching treatment, which is associated with enhanced DSA. However, the DSA-driven fatigue limit in the Fe – C alloy was easily deteriorated by increasing the temperature to 433 K, owing to the occurrence of dynamic precipitation during the fatigue test (Li et al., 2018). Therefore, precipitation must be suppressed to utilize the DSA effect enhanced by the supersaturated carbon. In this regard, we focused on improving the robustness of the high fatigue limit associated with DSA in steels with supersaturated carbon. It was expected that the DSA effect at a high temperature could be improved via the suppression of dynamic precipitation. In this context, the key solute element is Si, because it has been recognized to prevent cementite formation in terms of energetic and kinetic problems. More specifically, the carbide-formation tendencies of Si, Al, Co, and Ni are small, and hence, they can increase the free energy of formation of cementite (Bain, 1940; Kaneko et al., 1963). From the viewpoint of kinetics, as the distribution coefficient of Si in cementite/ferrite is low (Sato and Nishizawa, 1955), cementite formation and growth require Si diffusion in ferrite, which delays the occurrence of precipitation. These expected effects maintain the solute state of carbon, which improves the robustness of the fatigue properties of Fe – C alloy against the environmental temperature.

2. Experimental procedure 2.1. Material

The material selected in this study is a Fe-0.016C-1.0Si alloy. The chemical compositions including that of the Fe- 0.017C alloy used in the previous study (Li et al., 2016, 2017) are listed in Table 1. An ingot of the Fe-0.016C- 1.0Si alloy was prepared via vacuum induction melting. This alloy was first hot-forged and rolled at 1523 K and subsequently air-cooled. Subsequently, the rolled bar was solution-treated at 973 K for 3600 s and thereafter water- quenched. Subsequently, the Fe-0.016C-1.0Si alloy was stored in a refrigerator at 193 K to prevent carbon segregation to a large extent, except during the specimen preparation time and fatigue testing.

Table 1. Chemical composition of alloy. [mass%] (*mass ppm for B, N, and O)

Material

C

Si

Mn

P

S

Ni

Ti

Al

B*

N*

O*

Fe-0.016C-1.0Si

0.016 0.017

0.979 0.003

0.001 0.003

0.0007

0.0016 0.0003

0.048 0.052

11

13 15

Fe-0.017C

0.002

0.003

0.002

3

9

2.2. Tension, fatigue and hardness tests Tension tests were conducted at the initial strain rates of 10 -3 and 10 -5 s -1 and at 293 and 433 K. Figure 1(a) illustrates the specimen geometry for the tensile tests. The Vickers hardness at 293 K was measured at an indentation load of 9.8 N and holding time of 30 s. Fatigue tests were performed by the tension – compression fatigue machine with a stress ratio of −1 and frequency of 30 Hz. The test temperature was con trolled using a thermostatic chamber equipped with the fatigue machine. Figure 1(b) shows the specimen geometry for the fatigue testing. At the center of the fatigue specimen, a micro-notch shown in Fig. 1(c) was introduced via micro- drilling and focused ion beam machining. The crack propagation behavior was recorded using the replica technique, and the replicas were observed using optical microscopy.

3. Results and discussion 3.1. Temperature dependence of stress – strain response

Figure 2 shows nominal stress – strain curves at 293 and 433 K. No significant serrated flow was observed at 10 -3 s -1 at 293 K, but it was observed upon decreasing the strain rate to 10 -5 s -1 . The appearance of serrations at 293 K indicates the occurrence of DSA.