Deformation and damaging processes analyzed on the micro scale of VHCF-loaded nodular cast iron by 3D digital volume correlation of tomographic images
Jens Nellesen, RIF e.V. - Institut für Forschung und Transfer, Dortmund, GermanyWolfgang Tillmann, Lehrstuhl für Werkstofftechnologie, Fakultät Maschinenbau, Technische Universität Dortmund, Dortmund, Germany
In order to comprehend micro deformation and micro damaging processes running before the macroscopic failure it is inevitable to analyze these processes on the micro scale. By 3D imaging techniques, local deformation and the initiation and propagation of micro cracks can be detected in the bulk. As micro cracking is usually preceded by the strain localization, the origin of the final failure can be predicted in an early stage of mechanical load.
In the present work, it is illustrated by the example of a VHCF-loaded multi-phased material how deformation fields and damage can be quantified by 3D digital volume correlation (DVC) of tomographic images. The material used in the fatigue tests is nodular cast iron, which consists of a ferritic matrix with spherical graphite inclusions sized approx. 30 µm. The ex situ VHCF tests were carried out on tiny axisymmetric dog-bone-shaped specimens whose circular cross-section has a diameter in the range of 3 to 5 mm. To fix the specimens on the horn of the piezoelectric ultrasonic (US) processor, used for the VHCF loading, one of the specimen's heads was provided with a threaded stud. Thus, the specimens oscillated freely in the tension–compression mode (load ratio R = −1, frequency f ≈ 20 kHz) and their inertial mass was utilized for loading.
For the acquisition of the tomographic images, both monochromatic synchrotron radiation and polychromatic X-ray tube radiation were exploited for the μCT experiments. Using an in-house developed digital volume correlation software, the 3D strain fields and damage were analyzed in different stages of VHCF load in relation to the unloaded state. In doing so, the gradients of reconstruction values in the tomograms are evaluated which are caused by the attenuation contrast between the phases constituting the microstructure.
The results evidence that strain values < 1% can be measured by the DVC analysis method, which allows to characterize the inhomogeneity of the irreversible deformation structure. Moreover, it is shown that the inhomogeneous deformation structure can be detected already in an early deformation stage and is developing step-by-step in the course of the VHCF loading. As might be expected, the highest strain maxima were found in the region of the smallest cross-section of the specimen. The possibilities and limitations of the tomographic imaging techniques with the subsequent digital volume correlation are discussed.