Crystalline materials: Twin behaviour and size.

D anny de Vito would be happy to hear that for twin formation the paradigm 'smaller is stronger' still prevails. Twinning is a deformation process in crystals defined as the collective shearing of one portion of the crystal with respect to the rest. However, compared with plasticity based on dislocation glide, twinning has been scarcely studied in the deformation of metals. This is because twinning occurs (when slip by dislocation glide is suppressed) mostly for a few hexagonal-close-packed metals such as Ti, or at very low temperatures. Thus, twinning in metals has been considered of rather small technical importance. Furthermore, twinning is difficult to study as it occurs as a collective mechanism, and its origin and evolution in space and time remain puzzling. Writing in Nature, Yu et al. 1 now describe size effects on the deformation of small, micrometre-and submicrometre-sized Ti alloy crystals and, in doing so, provide new insights and questions relating to the deformation twinning process. It has long been known that size effects have a prominent role in the deformation and strength of metals. In the 1950s, it was observed that grain size has a strong impact on the strength of pure metals and alloys. Also, it was demonstrated impressively that specimen size is important and that small whiskers, presumably dislocation-free, can have very high strength. Over the decades, such size effects have been observed for many samples and testing conditions, including metal thin films and, most recently, for small single crystals in microcompression tests 2–4. Although not all details are understood, it is now recognized that these size effects relate to the confinement of the motion of dislocations and the availability of dislocations and dislocation sources 5. Moreover, it has been shown that nanoscale growth twins can lead to high-strength materials because the twins confine dislocation motion 6. This observation, however, is not to be confused with twinning as a deformation mechanism, which is discussed here. Intuitively, one might imagine that there is no size effect for twinning of a small single crystal, which can freely deform. It could be argued that a smaller crystal twins more easily because a smaller volume needs to be sheared compared with a large crystal. However, Yu et al. show that the opposite is true: the smaller the crystal, the greater the stress required for deformation twinning (Fig. 1). For submicrometre-sized crystals, the flow stress is found to be …