On the relationship between and conjugate twins and double extension twins in rolled pure Mg

Abstract

The paper presents a new type of twin-like objects observed in rolled pure magnesium. They have {1126} and {1122} habit planes and their misorientations to the matrix are close to 56˚ and 63˚ about 10 10  axis, respectively. The ad hoc performed theoretical analysis and atomic simulations allow to interpret the objects as {1012} {1012}  double twins formed by the simultaneous action of two twinning shears with completely re-twinned volume of primary twin. The observed inclinations from the ideal misorientations for such double twins can be explained by the compliance of the strain invariant condition in the twin boundary. It seems plausible that, once the double twin is formed, its twin boundaries are hard to move by glide of twinning disconnections. If so, these twins represent obstacles for the motion of crystal dislocations increasing the hardness of the metal.

A one-step mechanism for new twinning modes in magnesium and titanium alloys modelled by the obliquity correction of a (58°,a+ 2b) prototype stretch twin

“…The supercell chosen for the calculations is twice that used in Cayron (2017a,b), which introduces more complexity and has prevented us for the moment from finding the analytical equations of the atomic trajectories. These results are the same as those recently obtained by Ostapovets et al (2017), despite the difference in the initial assumption of the mechanism. Despite these limitations, interesting information could be extracted.…”

“…As pointed out in x1, a similar observation was made by Ostapovets et al (2017) by using a double-shear model. For a hard-sphere packing ratio ¼ ffiffiffiffiffiffiffi 8=3 p , this angle takes the value arccos 5=11 ð Þ'62:96 .…”

“…There is a third point. Like Ostapovets et al (2017), our theoretical analysis led us to establish a link between these twinning modes, but our work differs from that of Ostapovets et al in the mechanism. 3 of Nave & Barnett (2004), or 56 as in Fig.…”

“…Why in titanium and zirconium alloys are the f1122g twins more frequent than f1102g extension twins if the shear value of the former (s ' 0.219) is larger than that of the latter (s ' 0.174)? Ostapovets et al (2017) very recently published an interesting experimental and theoretical paper that establishes such a link. These questions could not be raised between 1970 and 1990, because at that time EBSD statistics were not available.…”

“…For magnesium alloys, some histograms of misorientations extracted from EBSD maps reported in literature exhibit a peak corresponding to rotation of angle close to 56 and axis close to a + 2b (Nave & Barnett, 2004;Lentz et al, 2014), and Ostapovets et al (2017) The two possible obliquity corrections required to transform the (58 , a + 2b) stretch twin into a simple shear twin, by maintaining untilted (a,b) the plane (2112), and (c,d) the plane (2116). For a hard-sphere packing ratio ¼ ffiffiffiffiffiffiffi 8=3 p , this angle takes the value arccos 19=35 ð Þ'57:12 .…”