Roland Loge | Searcharxiv

arXiv2017

Evidence of new twinning modes in magnesium questioning the shear paradigm

Twinning is an important deformation mode of hexagonal close-packed metals. The crystallographic theory is based on the 150-years old concept of simple shear. The habit plane of the twin is the shear plane, it is invariant. Here we present Electron BackScatter Diffraction observations and crystallographic analysis of a millimeter size twin in a magnesium single crystal whose straight habit plane, unambiguously determined both the parent crystal and in its twin, is not an invariant plane. This experimental evidence demonstrates that macroscopic deformation twinning can be obtained by a mechanism that is not a simple shear. Beside, this unconventional twin is often co-formed with a new conventional twin that exhibits the lowest shear magnitude ever reported in metals. The existence of unconventional twinning introduces a shift of paradigm and calls for the development of a new theory for the displacive transformations

cond-mat.mtrl-sci↗

arXiv2017

A crystallographic model of the {557} habit planes in low-carbon martensitic steels

Low-alloy steels are constituted of twenty-four variants of lath martensite that exhibit gradients of orientations from Kurdjumov-Sachs (KS) to Nishiyama-Wassermann (NW). They are structured into four packets on each of the common close-packed plane {111}fcc// {110}bcc; and each packet is composed of three blocks constituted by pairs of low-misoriented variants. The habit planes reported in literature for this type of martensite are {557}fcc, but it is not clear whether they correspond to the laths or to the blocks. In this paper, we present crystallographic calculations proving that the average of the two KS distortions associated with the variants in a block is exactly a NW distortion. A new method of averaging distortion matrices was introduced for this purpose. It is also shown that the {575}fcc planes are let untilted by this NW distortion, and are thus good theoretical candidates for the observed habit planes. The predicted {575}fcc planes, however, do not contain any of the common close-packed directions of the two variants in the block, which is in apparent contradiction with the current view. In order to clarify this point, some Electron BackScatter Diffraction (EBSD) maps were acquired on different low-carbon steels; the prior austenitic grains were automatically reconstructed and the traces of the habit planes predicted by the different models were analyzed and compared to the morphologies. This experimental work shows that {575}fcc planes are the habit plane of the blocks, and that the habit plane of one block often dominates the others, which impedes to discriminate the different models. The advantages of our model are its simplicity, the absence of fitting parameters, and the symmetric role played by the variants the blocks.

cond-mat.mtrl-sci↗