QUANTITATIVE FRACTOGRAPHY IN BIAXIAL FATIGUE OF MATERIALs
Final Report (AKTION 41p3)
Prof. Dr. Jaroslav Pokluda
Institute of Physical Engineering, Faculty of Mechanical Engineering
Brno University of Technology, Czech Republic
&
Assoc. Prof. Dr. Reinhard Pippan
Erich Schmid Institute for Materials Science, Austrian Academy of Sciences
Leoben, Austria
Research activities
The quantitative fractography (QF) is widely used for investigation of fatigue fracture
surfaces in order to solve problems concerning both the failure analysis and the structural
design. Unlike in the uniaxial fatigue, the application of QF to biaxial fatigue problems is still
rather in a pioneering stage. Information obtained by means of the standard two-dimensional
QF is highly insufficient owing to an extreme tortuosity of the crack front and diversity of
local crack propagation directions under biaxial loading. Methods based on
stereophotogrammetry and QF were used to obtain an extended set of morphological
characteristics as profile and surface roughness or fractal dimensions, developing under
biaxial (bending-torsion) fatigue loading. These characteristics were determined as functions
of loading parameters and the distance from the crack initiation site. The first part of the
results was presented at 10th International Conference on Physics of Materials (ISPMA 10) in
Prague, August 28 - September 2, 2005 [1], and it was submitted to the scientific journal
“Materials Science & Engineering. The second part was already accepted for Proceedings of
the 2nd international conference “Crack Paths 2006” in Parma, September 14-16, 2006 [2].
An intensive international discussion concerning the relevancy of a pure mode III
loading (one of the three basic loading modes) for the fatigue crack growth in metallic
materials has broken up as a consequence of published results obtained by the applicants in
the frame of the AKTION project during the years 2001-02 [3, 4]. We believe that cracks
propagating under the macroscopically pure mode III loading propagate, in fact, under either
microscopically pure mode I or mixed mode I+III. Therefore, a special emphasize was
devoted to the 3D micromorphology of fracture surfaces created under a macroscopically pure
mode III loading (pure torsion). An extended experiment focused on a direct comparison of
crack growth rates in modes II and III was prepared in Leoben. Special cylindrical specimens
of austenitic steel containing two artificial circumferential cracks will be cyclically loaded by
means of a fatigue stand under conditions preventing a superposition of mode I loading.
During the year 2005, static loading was applied to identify the critical crack tip opening
displacement (CTOD) for modes II and III as well as an eventual superposition of the mode I.
Fatigue experiments will be performed during the year 2006.
Theoretical modeling of crack closure micromechanisms in fatigue loading under plain
strain was performed based on synthesis of dislocation and statistical models previously
developed by R. Pippan et al. [5] and J. Pokluda et al. [6]. A common paper concerning the
roughness induced closure component was prepared for the “10th International Conference
on Physics of Materials” in Prague, 2005, and submitted to Materials Science & Engineering
A [7]. A more general paper concerning both roughness- and plasticity induced crack closure
was accepted for Proceedings of the 9th international congress “Fatigue 2006” in Atlanta,
May 14-19, 2006 [8].
Results
The most important results of the scientific research in the frame of the project
AKTION 41p3 can be summarized into the following points.
1) A great influence of torsion loading component on the surface topography was
quantified by identification of a critical value of the loading ratio r 0.5. For loadings of r 0.5
(a lower portion of the torsion component) surface roughness seems to be almost the same,
despite the visual differences between fracture surfaces. Once the critical value is exceeded,
fatigue crack propagates in much more complicated manner resulting in a distinctly higher
surface roughness.
2) Some of studied roughness parameters, namely the hybrid linear roughness RL and
the linear fractal dimension DL, have shown the sensitivity to the profile orientation. This is
believed to be caused by their dependence on both the asperity amplitude and spacing. The
parameter DL quantifies profile roughness in a wide range of its resolutions.
3) The micromorphology of mode II and mode III parts of static fracture surfaces on
cylindrical specimens was qualitatively the same. An asymmetry of mode II fracture parts
appeared at opposite sides of specimens. The reason was a superimposed mode I loading due
to a slight bending moment. This severe influence has to be corrected in subsequent fatigue
experiments.
4) Values of CTOD were determined for mode II and mode III loading of austenitic
specimens. Although these values exhibit a rather big scatter, they can be used for
approximate setting up of loading parameters in subsequent fatigue experiments.
5) The total crack closure ratio in the whole fatigue crack-growth region can be
expressed by a simple formula. It contains two fitting constants C and SRc in the order of 101, possessing a clear physical meaning. The theory shows that, for all engineering materials,
the total level of roughness-induced crack closur e rapidly decreases with increasing applied
stress intensity range K. For nanomaterials, the ratio can even be neglected in the whole crack
growth region. The plasticity induced component lies in the close range of 0.2 – 0.3 for all
metallic materials.
The financial statement
The list of costs in CZK, spent in 2005 and reprinted from the economical department of
BUT, is shown in the Appendix. The costs planned for the Austrian partners in the Czech
Republic were not fully consumed due to their extreme working stress to the end of the year.
On the other hand, the costs for Czech participants in CZK (as well as in EUR) were fully
utilized.
References:
[1] Slámečka K., Ponížil P. and Pokluda J. : Quantitative Fractography in BendingTorsion Fatigue (submitted to Materials Science & Engineering A).
[2] Major Š., Ponížil P., Sabirov I., Slámečka K. and Pokluda J.: 3D-Fractography in
Bending-Torsion Fatigue (accepted for “Crack Paths 2006”).
[3] Pokluda J., Pippan R., Slámečka K. and Kolednik O.: Fatigue Crack Growth in
Metals under Pure Mode III: Reality or Fiction? In: "Fatigue Crack Paths (FCP 2003)", Ed.
Carpinteri A., Pook, L.P., University of Parma, Parma 2003, p. 92.
[4] Pokluda J. and Pippan R.: Can a Pure Mode III Fatigue Loading Contribute to Crack
Propagation in Metallic Materials? Fatigue & Fracture of Engineering Materials &
Structures 28 (2005), 179.
[5] R. Pippan, G. Strobl, H. Kreuzer and C. Motz: Asymmetric crack wake plasticity – a
reason for roughness induced crack closure. Acta Materialia 52 (2004), 4493.
[6] Pokluda J., Šandera P. and Horníková J.: Statistical Approach to RoughnessInduced Shielding Effects. Fatigue & Fracture Engineering Materials & Structures 27
(2004), 141.
[7] Pokluda J. and Pippan R.: Analysis of Roughness-Induced Crack Closure Based on
Asymmetric Crack-Wake Plasticity and Size Ratio Effect (submitted to Materials Science &
Engineering A).
[8] Pokluda J., Šandera P. and Pippan R.: Analysis of Crack Closure Level in Terms of
Crack-Wake Plasticity (accepted for “Fatigue 2006”).
Brno, 26. 2. 2006
Prof. Dr. Jaroslav Pokluda