Size effects on the ductile/brittle fracture properties of the pressure vessel steel 20 g

https://doi.org/10.1016/j.tafmec.2008.07.005Get rights and content

Abstract

Detailed un-standard experiments of fracture toughness in which SENB specimens of five different thicknesses were included were carried out to investigate the size effect on the ductile and brittle fracture. It is found that the ductile fracture toughness parameter-the critical strain energy release rate increases with the size of the specimens which have the similar geometry shape and then decreases gradually to the plane strain fracture toughness. On the upper shelf the critical strain energy release rate increases with the size in the range of 4–16 mm for the increment of the plastic deformation zone size and plastic fracture strain under general yielding conditions, and then drops down from 16 mm to 22 mm for the plastic deformation zone size not changing much which is less than the residual ligament width and the increase of the proportional of the high stress triaxiality zone to the whole specimen. While on the lower shelf the critical strain energy release rate increases with the thickness in the range of 4–8 mm for the plastic deformation zone size increasing under small scale yielding conditions, and then drops down from 8 mm to 22 mm for the increase of the high out-of-plane constraint. Theoretical analysis with the primary definition of the strain energy release rate, the constraint level and the plastic deformation volume was performed to investigate the different size effects for different temperatures. FEA simulations with continuum damage model of GTN models to get the results of the stress triaxiality as an important factor of fracture toughness increasing with the reduction of the thickness.

Introduction

It is an important topic of research for material science when the fracture properties of materials processed under the form of plates must be characterized, and for structural mechanics when the integrity of structures made of different thickness plates or pipes must be assessed [1]. And furthermore, it has been found that fracture properties as compared to elastic and plastic material properties, depends on the specimen size, crack depth, geometry, and loading conditions [2], [3], [4], [5], [6], [7], [8]. Detailed fracture toughness data base of deferent un-standard geometries especially of smaller specimens for the oldest reactors which have limited numbers of larger specimens, together with the method to apply this information in integrity assessment is urgently required in design and operation of the defect tolerant fusion structures.

A number of fracture toughness parameters based on energetic considerations have been proposed to characterize material resistance to crack propagation, such as the J–R curve, strain energy release rate G and the crack tip opening angle/displacement (CTOA/CTOD) obtained from the relevant load–geometry parameters. Besides the plane strain fracture toughness, the normal fracture toughness of different sizes and temperatures has direct importance to evaluate the safety of the structures with the same size. Un-standard normal fracture toughness of true structures is also suggested in ASTM especially considering that the uncertain of the transferability of the plane strain toughness (material constant/the real toughness) to thicker/thinner plates and pipes cannot be ensured.

Specimen size effects on the normal fracture toughness were usually interpreted in terms of the variation of the plain strain state region on the lower shelf for the constraint increment and plastic zone size at the crack tip in the specimens on the upper shelf [9], [10]. Experimental results show that, the fracture toughness increases at lower specimen size when the stress state ahead of the crack tip is nearly plane stress state [11], reaches a maximum and then decreases gradually to a constant value with increasing size [12], [13] for the stress state ahead of the crack tip is changed from plane stress state prevailing to nearly plane strain state as shown in Fig. 1 schematically. While the critical transition size B0 of the ductile is much larger than that of brittle fracture from the experimental results [14], [15].

Some theoretical models and simulations were proposed to investigate the size effect on the experimental fracture toughness based on the stress triaxiality [3]. The level of stress triaxiality factor which represents the typical triaxial stress state determines the constraint level, which was used to explain the size effect on the fracture toughness in many papers [10]. The fraction of high constraint area in the middle of the specimen cross-section under stable crack growth increases in proportion to the shear area with increasing triaxial stress state and increasing test temperature [16], while the fracture strain decreases with increasing stress state triaxiality and with increasing temperature following the Rice and Tracey model for void growth [17].

In this paper, the typical fracture toughness parameter, strain energy release rate GC was obtained experimentally by SENB specimens of five different thicknesses together with the basic material properties under wide temperature range to characterize material resistance to crack propagation. Theoretical analysis with the primary definition of the strain energy release rate, the constraint level and the plastic deformation volume was performed to investigate the different size effects for different temperatures. FEA simulations with continuum damage models were carried out to study the distribution of the stress triaxiality at the critical loading level, which is considered as an important factor determining the critical ductile fracture strain and ductile fracture toughness.

Section snippets

Materials

The material used for the experimental investigations is a low carbon low alloy ferric steel which has been used in pressure vessels and piping structures for several decades, with Chinese designation 20 g. Its chemical composition is following as: C 0.2125, Si 0.3380, Mn 0.5859, S 0.0287, Cr 0.0427, Al 0.4389, Mo 0.049. The microstructure of the steels is shown in Fig. 2 with the average grain size of 20–30 μm by SEM. Levels about mechanical properties of the material at 20 °C are as follows:

Size effect on the upper shelf

The critical strain energy release rate for the SENB specimens at the critical loading point can be calculated with Eq. (3):GC=2UCB(W-a)where UC is the work from the loading system, and equal to the fracture energy absorbed by the material in the fracture affected volumeUC=0cPdΔ=VaEfdVwhere Ef is the fracture energy density integrated on the loading history which can be divided into elastic, plastic energy density and surface energy density:Ef=We+Wp+Wswhere the elastic energy density can be

Concluding remarks

Previously, many studies were performed on the pipeline steels with various microstructures, which were tested as full-thickness or even thicker plates (range: 25.4–400 mm) with the aim of investigating the size effect on the brittle fracture toughness. In this study, fracture toughness tests of smaller size from 4 mm to 22 mm were executed under wider range of temperatures.

As the configuration is far from pure plane strain conditions, the parameters characterizing crack initiation and propagation

Acknowledgement

The financial support from the Framatome-ANP under the Joint Research Project agreement N 10-M and the Major State Basic Research Projects (2006CB605003) is greatly acknowledged.

References (33)

  • J.R. Rice et al.

    Journal of the Mechanics and Physics of Solids

    (1969)
  • G. Bertolino et al.

    Influence of the crack-tip hydride concentration on the fracture toughness of Zircaloy-4

    Journal of Nuclear Materials

    (2006)
  • G.R. Odette et al.

    Cleavage fracture irradiation embrittlement of fusion reactor alloys: mechanisms, multiscale models, toughness measurements and implications to structural integrity assessment

    Journal of Nuclear Materials

    (2003)
  • Shiro Jitsukawa et al.

    Effect of size and configuration of 3-point bend bar specimens

    Journal of Nuclear Materials

    (1999)
  • S. Chandrakanth et al.

    An isotropic damage model for ductile material

    Engineering Fracture Mechanics

    (1995)
  • D. Chae et al.

    Damage accumulation and failure of HSLA-100 steel

    Materials Science and Engineering A

    (2004)
  • Cited by (17)

    • Investigation of the effects of material parameters on the relationship between crack tip constraint and CTOD fracture toughness

      2020, Theoretical and Applied Fracture Mechanics
      Citation Excerpt :

      To ensure the safe operation of pipelines, it is important to measure accurate fracture toughness. However, it is well known that the fracture toughness strongly depends on the constraint level ahead of the crack tip [1–3]. Moreover, there are many factors that can affect the relationship between the constraint and the fracture toughness, including structure size and material parameters.

    • The effect of crack-tip constraint in some problems of fracture mechanics

      2020, Engineering Failure Analysis
      Citation Excerpt :

      In contrast to in-plane constraint, out-of-plane constraint is due to thickness. A change of crack-tip constraint parameters leads to considerable variation of the fracture toughness (e.g., [1–5]). For example, a standard specimen, such as compact tension specimen, is usually of high crack-tip constraint, while most of non-standard specimens or real cracked engineering components have low crack-tip constraint.

    • Numerical simulation on fracture resistance and factors affecting toughness for welded joint of low-alloy steel

      2019, Advances in Engineering Software
      Citation Excerpt :

      Shlyannikov et al [3] quantitated out-of-plane constraint by applying constraint parameter (T-stress) to characterize stress filed near the crack front based on finite element analyses. Wang et al [4] found that the critical strain energy release rate (ductile fracture toughness parameter) increases with the size of the SENB specimens and then decreases gradually to the plane strain fracture toughness for low carbon low alloy ferric steels. Shahani et al [5] and Newton [6] have also proposed similar effect of thickness on fracture toughness for alloy steel COST08Ch22N6T and copper foils respectively, while Meshii and Tanaka [7] reported that fracture toughness at the maximum force Jcmax was proportional to B−0.5 for 0.55% carbon steel (JIS S55C) after conducting fracture toughness tests on CT specimens with different thickness.

    • Influence of energy dissipation at the interphase boundaries on impact fracture behaviour of a plain carbon steel

      2018, Theoretical and Applied Fracture Mechanics
      Citation Excerpt :

      The non-linearity of the latter may be taken into account by considering the independent 2D-subsystem (surface and interface boundaries) and 3D-crystal subsystem (inside the grain body) [1]. The ductility and fracture toughness of pipe steels is heavily dependent on the microstructure where the distribution of phases and grain sizes plays a key role [18–22]. In the case of brittle and/or mixed fracture, micro-cracks initiate primarily at the grain boundaries.

    • Two-parameter J-A concept in connection with crack-tip constraint

      2017, Theoretical and Applied Fracture Mechanics
      Citation Excerpt :

      In contrast to in-plane constraint, out-of-plane constraint is due to thickness. A change of crack-tip constraint parameters leads to considerable variation of the fracture toughness (e.g., [1–7]). For example, a standard specimen, such as compact tension specimen, is usually of high crack-tip constraint, while most of non-standard specimens or real cracked engineering components have low crack-tip constraint.

    View all citing articles on Scopus
    View full text