Elsevier

Acta Materialia

Volume 163, 15 January 2019, Pages 68-77
Acta Materialia

Full length article
Quantification of precipitate hardening of twin nucleation and growth in Mg and Mg-5Zn using micro-pillar compression

https://doi.org/10.1016/j.actamat.2018.10.009Get rights and content

Abstract

In polycrystalline materials, the stress corresponding to twin nucleation is difficult to separate from twin growth because these events occur concurrently during deformation. In this work, we separate the nucleation stress and growth stress of {101¯2} twinning by compression of micro-pillars containing pre-existing twins through their centre. Micro-pillar compression results showed a strong size effect on both twin nucleation and twin growth stresses for pure Mg and Mg-5Zn alloys. Taking this into account, the critical stress for growth of twins in pure magnesium is found to be ∼7 MPa which is consistent with previously published measurements on macroscopic single crystals. These experiments have been used to deduce the precipitate hardening of twin growth, and for the present precipitate dispersion this has been measured to be ∼30 MPa. Back-stress calculations based on elastic bending of the precipitates showed close agreement to the measured precipitate hardening, and this model therefore accounts well for the observed strengthening. Site-specific atom probe tomography of the twin boundaries showed that room temperature ageing is sufficient to produce segregation of zinc to the twin boundary. This was found to immobilise the twin, and is believed to be the first report of solute locking of twins from room temperature exposure.

Introduction

In magnesium alloys, the hexagonal close-packed structure deforms by both dislocation slip and deformation twinning [1]. Deformation twinning often controls yield asymmetry, texture development, ductility and work hardening rates of magnesium alloys [2]. These behaviours are therefore of critical importance for understanding the plastic flow of these materials in engineering applications. Despite its technological significance, deformation twinning is not so well described. One of the difficulties of studying twinning is the discrimination between the twin nucleation and the subsequent twin growth. Twin nucleation requires a rather high and localized stress level, while twin growth is known to require a much lower stress level [3,4]. In a polycrystalline material, it is not possible to elucidate the difference between the twin nucleation and twin growth stresses from a flow curve because both twin nucleation and twin growth occur concurrently throughout plastic deformation, and cannot be decoupled.

With the development of micro-pillar compression, mechanical behaviour of single crystals along specific orientations can be studied [[5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]]. Deformation twinning and dislocation slip in magnesium alloys can be activated separately by compressing micro-pillars in various orientations [[17], [18], [19], [20], [21], [22]]. The orientation and size effects in magnesium alloys have been widely studied using micro-pillar compression [[22], [23], [24], [25], [26], [27]]. However, the quantification of twin growth stress for magnesium alloys is still challenging, because the nucleation of twins in magnesium micro-pillars leads to strain bursts in the flow curves [18,20,22,24]. Since the data points cannot be tracked closely during strain bursts, the true stress value for twin growth cannot be accurately quantified.

Therefore, in the present paper, we have designed a set of experiments to specifically measure the twin growth stress for pure Mg, solution-treated Mg-5Zn, and peak-aged Mg-5Zn using the micro-pillar compression test technique. The micro-pillars already contain a pre-twinned region, and upon compression the pre-twin expands and provides a measure of the twin growth stress. Compression of a second set of samples that do not contain any pre-twin boundaries allows a separate measurement of the twin nucleation stress to help benchmark the twin growth values. Thus, we are able to extract the intrinsic twin growth stress, and can compare values obtained in pure Mg with those obtained in Mg-5Zn alloys containing Zn in solid solution, and also the twin growth stress in the presence of precipitates. In the latter case, this particular precipitate dispersion has been shown to have a strong hardening effect in polycrystals [28], and the experiments described here further allow us to separate the precipitate strengthening of the individual deformation modes.

Section snippets

Experimental methods

Two alloys were used in this study, pure Mg (99.96% purity) and Mg-5Zn (5 wt% Zn). The Mg-5Zn alloy was examined in two conditions, i.e. solution-treated and peak-aged. Both conditions began with a solution treatment at 400 °C for 6 h in a flowing argon atmosphere, followed by a cold water quench. Those peak-aged samples requiring precipitate hardening were then aged at 150 °C for 8 days. Cylindrical specimens of 5.4 mm in diameter and 8 mm in height were machined from the starting materials

Twinning in pure Mg

As the compression axis is approximately parallel to the [101¯0] orientation of Mg, {101¯2} twinning is expected to be the preferred deformation mode. For the micro-pillars containing pre-twins, no twin nucleation was required, and growth of the pre-twin was observed (Fig. 1e and f). Under compression, the pre-twin propagated in both upward and downward directions, and eventually reached the top surface and substrate. The t-EBSD map (Fig. 2 a and c) and dark-field TEM image (Fig. 2d) confirmed

Micro-pillar size effect on twin growth

Care must be taken when using micro-pillars to predict macroscopic behaviour because there is a well-documented size effect whereby smaller samples show large strength values [5,7,8]. It has been suggested in a previous paper that the twin growth stress in pure Mg is relatively size insensitive [27]. However, the present results do not agree with that conclusion, Fig. 3. It is important to consider the reasons for the different conclusions. If we compare the flow curves from the present study

Conclusions

The nucleation and growth stresses of {101¯2} deformation twin have been investigated separately by compression of twin-free and pre-twinned micro-pillars for pure Mg, solution-treated Mg-5Zn, and peak-aged Mg-5Zn. The main conclusions can be summarised as follows.

  • 1.

    The twin growth stress was measured by compression of micro-pillars containing pre-twins through their centre. TEM and t-EBSD characterization confirmed that the pre-twins grew in response to the compressive load. The flow curves of

Acknowledgements

The authors would like to acknowledge the financial support of the Australian Research Council [Grant number: DP160101540] and Deakin University. The present work was carried out with the support of the Deakin Advanced Characterisation Facility. The authors are also grateful for assistance with the experimental work from Dr. Ross Marceau, Mr. Rob Pow, Mr. Dave Gray, Dr. Adam Taylor, Dr. Mark Nave, Mrs Rosey Squire, and Dr. Andrew Sullivan. The continued support of Prof. Emily Hilder and Prof

References (38)

Cited by (44)

View all citing articles on Scopus
View full text