Experiments and Fracture Modeling of High-Strength Pipelines for High and Low Stress Triaxialities
Kirki Kofiani
1
, Aida Nonn
2
, Tomasz Wierzbicki
1
, Christoph Kalwa
3
and Carey Walters
4
1
Massachusetts Institute of Technology, Department of Mechanical Engineering, Cambridge, MA, USA
2
Salzgitter Mannesmann Forschung GmbH, Duisburg, Germany
3
EUROPIPE, Mülheim an der Ruhr, Germany
4
Centre for Mechanical and Maritime Structures TNO Built Environment and Geosciences, Delft, Netherlands
ABSTRACT
This paper provides results from a comprehensive study on mechanical
characterization of high-strength pipeline steel, grade X100 using
experimental and numerical methods. The material was characterized
for anisotropic plasticity, fracture initiation for various states of stress,
(pre-cracked) fracture toughness and uncracked ductility. The
experimental program included tests on flat butterfly-shaped, central
hole, notched and circular disk specimens for low stress triaxiality
levels; as well as tests on round notched bar specimens and SENT
fracture mechanics tests, for high values of stress triaxiality. This
program covered a wide range of stress conditions and demonstrated its
effect on the material resistance. Parallel to the experimental study,
detailed numerical investigations were carried out to simulate all
different experimental tests. Using an inverse method, a 3-parameter
calibration was performed on the Modified Mohr-Coulomb (MMC)
fracture model. Subsequently, the predictive capabilities of the MMC
were evaluated by the comparison to the fracture toughness tests
results, used extensively in the pipeline industry. The capabilities of the
MIT fracture model have been demonstrated on an example of high
strength offshore steel, X100. The outcome of this study was not only
to provide, the overall characterization of the fracture behavior of this
material as an example, but also to present the methodology on how to
use the MMC model as a practical tool in pipeline design.
KEY WORDS: ductile fracture; SENT tests, high-strength pipeline
material; X100; Modified Mohr-Coulomb model; triaxiality and Lode
angle dependence.
INTRODUCTION
The growing need for energy resources located in remote regions with
complex ground and extreme climate conditions has led to increasing
demands on pipeline material to ensure safe and cost-efficient design of
gas transmission pipelines. One of the main challenges of research in
the pipeline industry is to achieve both high strength and high
toughness material properties which will allow pipelines to sustain
large deformations, even under multi-axial loading. However, there are
still open issues regarding the deformation and fracture performance of
pipelines when subjected to different loading scenarios.
In recent decades, numerous fracture models have been developed with
the aim of characterizing ductile fracture of steel materials. The major
characteristic of these models is their capability to describe failure
mechanisms by taking into account the influence of local stress and
strain variables. One of the models which is widely used for the
simulation of the damage evolution in the pipelines is Gurson-
Tvergaard-Needleman (GTN) model (Gurson, 1977; Tvergaard, 1982;
Tvergaard and Needleman, 1984). The simulation of ductile fracture for
X100 material model has been performed recently using GTN model in
high stress triaxiality range (Ishikawa, Sueyoshi and Igi, 2010; Nonn
and Kalwa, 2010; Tanguy, Luu, Perrin, Pineau and Besson, 2008).
Although this micromechanics-based model describes the ductile
failure mechanism of void nucleation, growth and coalescence
adequately, it requires the calibration of numerous parameters. The
recent modifications of GTN, e.g. to account for shear-stress dominated
fracture (Jackiewicz, 2011; Nahshon and Hutchinson, 2008; Nielsen
and Tvergaard, 2011) , lead to an increase of the additional parameters
and hence the challenge for finding a representative parameter set.
Furthermore, the work of Dunand and Mohr (2011) shows that this
large amount of parameters is not necessary for achieving a good
correlation.
As an alternative to the GTN model, the modified Mohr-Coulomb
(MMC) model has been introduced by (Bai and Wierzbicki, 2010) and
successfully applied for the prediction of ductile fracture initiation and
propagation in advanced high-strength sheet and low-strength steel
materials. This extended model describes the equivalent strain to
fracture depending on stress triaxiality and Lode angle, parameter
related to the normalized third invariant of stress deviator. To derive the
three-dimensional fracture locus, it is necessary to calibrate only three
model parameters by at least the same number of the experiments. The
suitability of the MMC model to characterize the ductile fracture has
been demonstrated in many publications, primary for the ranges of low
stress triaxiality (η <0.7) see (e.g. Bai, 2010, Dunand and Mohr, 2011).
On the other hand, there is no sufficient knowledge about the accuracy
of the model prediction for the high triaxiality levels η >1.0, which is
very important for analysis of cracked structures.
This paper seeks to close this gap by providing the results from
experimental and numerical characterization of the ductile fracture
511
Proceedings of the Twenty-second (2012) International Offshore and Polar Engineering Conference
Rhodes, Greece, June 17–22, 2012
Copyright © 2012 by the International Society of Offshore and Polar Engineers (ISOPE)
ISBN 978-1-880653-94–4 (Set); ISSN 1098-6189 (Set)
www.isope.org