Analyses based on the Hall-Petch relation and tensile loading-unloading tests were made to clarify the deformation mechanisms in pure Mg and AZ31 alloy polycrystals. The relation between 0.2% yield stress and the inverse square root of average grain size at the range of 77 to 523K for AZ31 alloys is shown in Fig. 1, where each experimental point indicates the mean value of three measurements. The obtained graph enables us to confirm the Hall-Petch relation. Similarly, this relationship is also applied to the case of 77 to 673K for pure Mg. The Hall-Petch slope k_y of both specimens is almost independent of strain, however the increase of the frictional stress ₀ is remarkable with strain because of the work hardening. Furthermore, it is found that the values of k_y and ₀ decrease with increasing temperature, especially at higher than 423K. Owing to solution strengthening and twinning, the slope k_y of AZ31 alloys increases in comparison with that for pure Mg. Fig. 2 shows the temperature dependence of the k_y values for pure Mg and AZ31 alloys derived from the experimental data. The slope k_y of both specimens tends to decrease with an increase of temperature, becoming below one-third of that for 293K at the temperatures exceeding 423K. It is worth noting that the critical resolved shear stress (CRSS) of non-basal slip systems for pure Mg single crystal decreases rapidly with increasing temperature, in spite of the temperature dependence of that for the basal slip system is hardly to be seen. According to the original definition by Armstrong et al., the prominent decrease of k_y at elevated temperatures is considered to be associated with the contribution of non-basal slip to the basal slip mainly operated at room temperature, reducing the value of orientation factor which is related to the number of slip system. On the other hand, the authors performed the calculation employing the Schmid factor of 0.5 or Taylor factor of 6.5, together with the CRSS for pure Mg single crystal. Fig. 3 shows the fraction of the basal slip and non-basal slip systems activated in polycrystalline Mg at the range of 293 to 523K. The ratio of contribution of the basal slip and non-basal slip systems to the deformation is found to be 6:4 at 293K for pure Mg by using Schmid factor of 0.5, and 9:1 by using Taylor factor of 6.5. Percentage of the non-basal slips to the total slip increases with temperature. This fact indicates that the non-basal slips considerably contribute the plastic deformation of pure Mg polycrystals at room temperature and their involvement intensifies with increasing temperature.

To clarify the deformation mechanisms, the tensile loading-unloading tests were carried out as shown in the schematic stress-strain curves of Fig. 4, where _p is plastic strain, _er denotes elastic recovery strain, and _t is total strain. Fig. 5 show the relation between the elastic recovery strain ratio _er/_t and the plastic strain _p at the range of 77 to 523K for pure Mg composed of the grain sizes of 43 and 172m, respectively. It is noticed that the ratio _er/_t of both specimens decreases with an increase of _p, becoming almost constant at the range of _p higher than about 1.5%. Furthermore, the ratio _er/_t decreases sharply with increasing temperature, and the value at 523K is below half of that for 293K. This means that the non-basal slips activated in the plastic deformation at 293K, resulting in the mobile non-basal dislocations intersecting the forest basal ones. An increase of _p as well as deformation temperature causes to interfere with the back motion of piled-up dislocations on basal slip planes during unloading, due to the increasing activity of non-basal slips. Consequently, it is considered that the marked decrease of _er/_t is observed under the above experimental conditions, although the twinning is also a significant deformation mode. Fig. 6 shows the relation between the ratio _er/_t and the strain _p at 293K for AZ31 alloys with the grain size of 35m, together with pure Mg having those of 43, 60 and 172m. The temperature dependence of _er/_t at 1% plastic strain for each specimen is shown in Fig. 7. From both graphs, it is confirmed that the ratio _er/_t at 293K increases with alloying and grain coarsening whose effect decreases at elevated temperatures, however their tendency similar to the temperature dependence of the CRSS of non-basal slip systems is recognized. The obtained results are in agreement with the recent investigations, in particular with the outcome of computer simulation by Agnew et al., who used viscoplastic self-consistent model to analyze both mechanical behavior and texture development of pure Mg and its alloys.

[Published in Materials Science Forum, Vols. 488-489, (2005), pp. 555-558]