PSI - Issue 13

M. Seleznev et al. / Procedia Structural Integrity 13 (2018) 2071–2076 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

2072

2

1. Introduction

Due to the increasing demands on fatigue life of constructions, improvement of steel purity is one of the important topics, since non-metallic inclusions in steel (NMIs) act as stress raisers, initiating cracks and promoting fracture [1]. Progress in refining techniques allows to increase steel purity, reducing the population of NMIs. However, in some cases maximum size of NMI can grow along with the total NMI content reduction [2]. This is harmful for fatigue life of steel (and other alloys as well), because crack-inducing stress concentrations correlate with NMI size [3]. Thus, not only the total impurity content control, but also inclusion shape and size control is crucial for high quality steel manufacturing [4]. Among traditional refining methods, metal melt filtration by novel ceramic foam filters shows promising results in steel cleaning [5]. In order to study the reaction of filter surface with the steel melt, one can introduce a filter inside the melt for some limited time – this is so called “finger testing” (FT) [6]. Recently, Storti et al. reported on efficient reduction of alumina (Al 2 O 3 ) particles population in 42CrMo4 steel after FT time of only 10 s [7]. Steel after described treatment is of high interest to material science due to presence of NMIs with special plate morphology, whose influence on fatigue properties was not reported in the literature so far. Stress concentration on common ellipsoid-type inclusions is proportional to the cross-section area regardless of chemical compound [8][9], whereas shapes of greater complexity are reported scarcely. Few works report on fatigue crack initiation from clusters of NMIs [10,11]. It is shown [11] that equivalent area of NMI cluster can be estimated using the fine granular area (FGA) size, which appears around the crack initiating flaw in the very high cycle fatigue (VHCF) regime (>10 7 cycles) [12]. Influence of planar discontinuities on fatigue was investigated experimentally by introducing a notch to the specimen [13], also theoretical calculation of stress state of two notches is helpful [14]. However, both of these works [13,14] focus on semi-elliptical planar surface defects aligned perpendicular to the load direction, which differs sufficiently from plate-like inclusions distributed and aligned randomly in the volume. The aim of the present work is to experimentally investigate the influence of plate-like NMIs on mechanical behaviour of 42CrMo4 steel in VHCF regime.

2. Experimental procedures 2.1. Material and specimen preparation

Several batches of sulfur-lean (FT2, 3, 4, 7, 8) and sulfur-rich (FT5, 6) 42CrMo4 steel were remelted, oxidized and deoxidized in a specially designed casting simulator [6]. Subsequently, carbon-bonded alumina (AC5) foam filters with different coatings were introduced into the melt for 10 s with further solidification of steel [7]. Coating abbreviations in Table 1 denote carbon nano-tubes (CNT), alumina nano-sheets (ANS) and calcium hexaluminate (CA6). Industrial batch FT2 was only oxidized and deoxidized without filter and taken as a reference state.

Table 1. Properties of investigated 42CrMo4 fatigue specimens after “finger test” experiments and heat treatment

Chemical compound, mass %

Core hardness, HV10

NHD, mm

Code name

AC5 filter coating

C

Cr

Mn

Si

S

P

Al

O

FT2 FT3 FT4 FT5 FT6 FT7 FT8

No filter

0.34 0.33 0.32 0.33 0.22 0.34 0.33

0.95 0.97 0.96 1.03 1.04 0.97 0.97

0.71 0.20 0.001 0.021 0.70 0.21 0.002 0.017

0.007 0.003 0.003 0.006

313 310 319 314 290 323 320

0.22 0.34 0.26 0.37 0.32 0.27 0.27

Al 2 O 3

AC5

0.68 0.20 0.002 0.016 <0.001 0.004

No coating

0.66 0.21 0.037 0.014

0.009 0.003

CNT

0.65 0.20 0.028 0.015 <0.001 0.005

CNT+ANS

0.75 0.23 0.002 0.023 0.75 0.22 0.001 0.023

0.001 0.004 0.002 0.004

CA6

After solidification, each batch was hot isostatically pressed (HIP) to close porosity. Specimens for fatigue testing and metallographic inspection (sections) were machined from each batch parallel to each other after HIP. Austenitization + quenching, followed by two-step plasma-nitriding (1 h at 420 °C and 2 h at 570 °C) were applied to