PSI - Issue 13

Johannes Tlatlik et al. / Procedia Structural Integrity 13 (2018) 243–248 Johannes Tlatlik, Dieter Siegele / Structural Integrity Procedia 00 (2018) 000 – 000

244

2

1. Introduction

The probabilistic assessment of cleavage fracture of ferritic-bainitic steels can be conducted by macroscopic or local numerical concepts. These various concepts were developed for quasi-static testing conditions, so generally a good agreement with the experimental results under these conditions is obtained. Recent investigations regarding dynamic loading conditions with a crack tip loading rate of about 10 3 to 10 5 MPa√m/s , however, have shown that the macroscopic Master Curve concept reveals severe shortcomings for the assessment of cleavage fracture at elevated loading rates (i.e. Schindler (2012), Reichert and Tlatlik (2017a)). To be more exact, the Master Curve appears to have a steeper slope under these conditions. This phenomenon is often linked to adiabatic heating processes in the crack tip region (Schindler (2012)) which locally increase cleavage fracture resistance. To account for this mechanism, a general adjustment of the exponent of the “dynamic” Master Curve from p = 0.019 to p = 0.030 /°C produces a steeper, and therefore more suitable Master Curve shape. However, is has been shown by Reichert et al. (2017b) that the distribution function associated with the experimental fracture toughness values under dynamic loading is also changed, so the mentioned adjustment is not sufficient to describe the fracture behavior completely. The physical reason for this situation is proven to be connected to local crack arrest incidences found on the fracture surface of the dynamically loaded specimens, and elaborately discussed by Tlatlik (2017a). The present paper deals with the brief investigation of the conditions, as well as the implications on fracture behavior regarding the occurrence of these local crack arrest incidences under dynamic loading conditions. Furthermore, an existing local cleavage fracture model is used to numerically assess fracture probability for various crack tip loading rates. Finally, a micromechanically motivated model modification is proposed by the implementation of a local crack arrest condition in order to significantly improve local cleavage fracture assessment concepts under dynamic loading conditions.

Nomenclature a 0

initial crack length h stress triaxiality coefficient ̇ crack tip loading rate K Jcd dynamic fracture toughness P ini

probability of cleavage fracture initiation

P f T ε e σ I

probability of failure testing temperature

accumulated plastic equivalent strain

pl

maximum principal stress

γ crack γ mat

crack energy

material resistance CMOD crack mouth opening displacement

2. Experimental Database and Material

An experimental database of dynamic fracture mechanics experiments for the ferritic-bainitic reactor pressure vessel steel 22NiMoCr3-7 was used from the research projects Böhme et al. (2012) and Reichert et al. (2017b). The relevant experiments involved SE(B)40/20 specimens with a crack depth ratio of a 0 /W = 0.3, 10% side grooves, and a fatigue pre-crack according to ASTM E1921. Dynamic testing was performed with varying crack tip loading rates of about 10 3 , 10 4 , and 10 5 MPa√m/s, and different testing temperatures of -20, 0, and +20 °C. The mentioned research projects provided dynamic fracture toughness values K Jcd,1T in accordance with ASTM E1921, as well a complete Master Curve evaluation.