Influence of inclusion size on fatigue behavior of high strength steels in the gigacycle fatigue regime

https://doi.org/10.1016/j.ijfatigue.2006.06.004Get rights and content

Abstract

The fatigue properties of four high strength steels with same strength class but containing different inclusion sizes were investigated using an ultrasonic fatigue testing machine in the gigacycle fatigue regime. The fracture surfaces were observed using field emission scanning electron microscopy (FESEM) and element distributions at the crack origins were measured by an electron probe microanalyzer (EPMA). The fatigue behavior can be divided into three categories:

  • (a)

    The S–N curve displays a continuous decline and the internal cracks initiated from the large oxide inclusions for commercial 50CrV4 steel in which the average inclusion size is about 29 μm.

  • (b)

    Step-wise S–N curves were observed for clean 54SiCrV6 and clean 50CrV4 steels in which the average inclusion sizes are about 3.0 and 2.4 μm respectively. Most fatigue failures originated from the VC inclusion clusters at the lower stress amplitudes.

  • (c)

    For clean 54SiCr6 steel in which the inclusion size is smaller than 1 μm, the fatigue cracks did not initiated from inclusions or inclusion clusters but from the region enriched with carbon. S–N curve shows that the fatigue failure hardly occurs from 106 to 109 cycles, in other words, the fatigue reliability can be substantially improved in the super long fatigue life regime.

Introduction

For low and medium strength steels, specimens do not fail after testing to 107cycles, so the fatigue S–N curve is assumed to be a horizontal asymptote according to the conventional fatigue testing for steels. However, recent investigations have indicated that fatigue failure does occur in the regime of N > 107 cycles when the applied stress amplitude is lower than conventional fatigue strength in high strength steels [1], [2], [3], [4], [5], [6], [7], [8]. Stickler et al. [8] and Weiss et al. [9] have made the numerous investigation of the fatigue behavior of PM steel and founded that crack initiation occurred mainly at singularities such as single larger, isolated pores or also at pore clusters. For low alloy high strength steels, the fatigue crack generally initiates from a larger non-metallic inclusions and the fracture surface displays a fish-eye pattern [1], [2], [3], [4], [5], [6], [7]. The stress localization at the interface between the inclusion and matrix is the origin of fatigue cracking. This arises from the differential thermal contraction coefficients of inclusions and matrix during cooling and the concentration of applied stresses due to the different elastic constants between the matrix and the inclusions. As a result of stress concentration due to the factors mentioned above, cracks can initiate easily at the interface between inclusions and matrix [10]. In clean steels the size and number of non-metallic inclusions decrease greatly and mechanical properties can be improved substantially. Kiedssling [11] supposed that crack would not initiate from inclusions when the size of non-metallic inclusion is less than 5 μm in clean steels from the point of view of fracture toughness. Similarly under fatigue conditions, there must be a critical inclusion size, below which the fatigue fracture origins will not initiate from the inclusions. In this paper, a commercial and three clean steels of same strength class containing different inclusion sizes were prepared and the fatigue properties were investigated using an ultrasonic fatigue testing machine in the gigacycle fatigue regime. The effects of inclusion sizes over or below the critical value on fatigue behaviors were discussed.

Section snippets

Materials and experiments

Chemical compositions of four steels are shown in Table 1. The contents of harmful elements of S and P in clean steels (clean 50CrV4, clean 54SiCrV6 and clean 54SiCr6) are much less than that in commercial 50CrV4 steel. The heat-treatment procedure was austenized at 1123 K for 30 min followed by oil quenching and subsequently tempered at 713 K for 1 h for four steels. The microstructure of the four steels is the tempered martensite.

Standard tensile specimens were prepared after the heat-treatment

S–N curves

The S–N curves of four steels in the gigacycle regime are shown in Fig. 2. The open circle represents that the crack initiation site is at the surface matrix of the specimen and the solid circle with an arrow and a number indicates how many specimens pass the fatigue test at 109 cycles without failure. The meaning of other symbols can be seen in the figures. It is shown that at higher stress amplitudes, fatigue fracture origins of three clean steels all initiated from the surface matrix of

Discussions

Generally speaking, for high strength steels, inclusion competes with surface matrix to be the fatigue crack initiation site, the smaller the inclusion size is, the lower the probability of the crack initiation from the inclusion. The fatigue crack will initiate from surface matrix, for example, from the persistent slip bands or grain boundaries at the root of machining trace. Many results indicated that fatigue crack will not originate from the inclusions when inclusion size below the critical

Conclusions

From the fatigue testing results of high strength steels above, some conclusions may be drawn:

  • 1.

    For clean 50CrV4 steel, the fatigue strength is substantially improved compared with its commercial counterpart due to the average inclusion size decreasing.

  • 2.

    For clean 54SiCrV6 and clean 50CrV4 steels, in which the inclusion size is smaller than the critical size, the fatigue failure originates from the inclusion clusters.

  • 3.

    For clean 54SiCr6 steel, in which the inclusion size is smaller than 1 μm, the

Acknowledgement

This work is financially supported by the Project ’Fundamental research of new generation of steels in China. (No. 2004CB619100).

References (21)

There are more references available in the full text version of this article.

Cited by (159)

View all citing articles on Scopus
View full text