*3.3. Mechanical Properties*

A significant change in the microstructure after ECAP alters the mechanical strength of the investigated material. In this report, we investigated microhardness as a multiaxial loading as well as compression/tensile tests as a uniaxial loading.

The resulting microhardness measurements shown in Table 2 reveal that the mechanical strength increased with the increasing number of ECAP passes, and in the final condition (8P), the microhardness was 1.5 times higher than in the as-extruded condition. Additionally, the yield compression strength (YCS) determined from flow curves (shown in Figure 8a) increased more rapidly (see Table 2). The YCS of the 8P sample was roughly 2.4 times higher than in the case of the as-extruded condition. The differences in the microhardness evolution and the evolution in ultimate tensile strength in uniaxial loading are usually caused by the changes in texture. The flow curves of all samples deformed in compression exhibited an S-shape character, which indicates the activation of twinning [29]. Moreover, the sharp yield point was observed in the 8P sample, which is quite unusual in magnesium-based materials. In order to reveal the possible effect of texture, compression tests in two mutually perpendicular directions (ED, transverse direction (TD)) were performed, and a tension test was performed along the

processing direction (ED). The resulting flow curves are shown in Figure 8b. The tensile curve did not exhibit the S-shape character, suggesting that twinning does not occur in tension in the processing direction (ED). However, the strain anisotropy was very low despite the different characters of the flow curves.

Note that unlike ED flow curves, the flow curves along TD and normal direction (ND) did not exhibit the sharp yield point. In order to reveal the microstructural changes that led to the appearance of the sharp yield point, EBSD was performed on a sample deformed to 4% of plastic deformation along its ED. The resulting EBSD parent matrix/twin map is shown in Figure 9. The microstructure is heavily twinned, and the twinned grains are aligned in stripes declined by ~45◦ from the loading ED.


**Table 2.** Results of mechanical tests.

**Figure 8.** True stress–true strain compression curves of (**a**) all investigated conditions measured along the processing direction (extrusion direction (ED)) and (**b**) of the 8P sample measured in ED, normal direction (ND), and transverse direction (TD), see Figure 1.

**Figure 9.** Parent/twin map calculated from EBSD, measured on a longitudinal section of the 8P sample, deformed in compression up to 4% of plastic deformation. Loading direction is horizontal. (Blue color corresponds to parent grains, red color corresponds to twins.)

#### **4. Discussion**

The as-extruded condition exhibited a fully recrystallized and homogeneous microstructure with relatively small grains and weak texture. Comparable texture was previously observed in the same alloy [30] and also in the binary Mg–Nd alloy [31], and is directly connected with the presence of the rare earth elements in the material. TEM investigation revealed that the secondary phase particles are mainly formed along the grain boundaries and belong to the stable γ-phase. Their inhomogeneous distribution probably resulted from the slow cooling rate after the extrusion.

Processing by ECAP resulted in additional grain refinement. Homogeneous microstructures with an average grain size of 1.5 μm were achieved after eight passes of the ECAP, which is common for similarly processed magnesium alloys [27,28,32–35]. However, the studied ZN11 alloy showed enhanced strength despite the comparable grain refinement. As mentioned above, the measured increase in microhardness was by a factor of 1.5, while YCS increased by a factor of 2.5. A larger relative effect on yield stress than on microhardness is rather common; however, note that the initial material has a different grain structure in different studies [27,28,32,36]. Therefore, the effect of the processing on the enhanced microhardness can be analyzed with the use of the Hall–Petch (HP) plot shown in Figure 10.

**Figure 10.** Hall–Petch plot showing the dependence of the microhardness on (grain size)−1/2. (Grain size was calculated from the average density of grain boundaries in the case of bimodal grain size distribution.)

The HP plot has a linear character, which indicates the grain boundary validity of the Hall–Petch relation for the studied processed samples. However, the HP slope coefficient calculated from Figure 10 was ~50 HV.μm1/2, which is a higher value than in theMg–Al–Zn- andMg–Al–RE-based alloys processed similarly [28,37]. Due to lower alloying, the studied alloy has lower microhardness in its coarse-grained condition, and the effect of grain refinement is enhanced. Additionally, other strengthening mechanisms may act concurrently with the grain refinement. For instance, dislocation density observed in the 8P sample was significant, contrary to the previously studied alloys in [28,37]. Furthermore, TEM analysis showed that ECAP processing of the studied alloy caused the formation of GP zones. Consequently, the higher HP coefficient can be explained by the synergic effect of these three phenomena acting together. However, the contribution of each strengthening factor cannot be differentiated from the investigated samples.

On the other hand, the increase in YCS was significantly higher than the increase in microhardness. This discrepancy can be explained by the texture evolution after ECAP. Note that the formation of the "rare earth texture" after the extrusion causes texture softening during uniaxial deformation along the extrusion direction [30]. A high volume fraction of grains had their c-axis rotated ~45◦ from the processing direction. Therefore, the activation of a basal slip system during uniaxial deformation along the extrusion direction is significantly facilitated. Figure 11a,b shows calculated distributions of the Schmid factor (SF) for all major deformation mechanisms for both extrusion and 8P conditions, when loaded in compression along the processing direction. The calculation was performed in MTEX software from raw EBSD data [23]. The SF for each point and each deformation mode was calculated and the resulting histograms were plotted.

A relatively sharp distribution of the SF for basal slip, with a maximum at SF = 0.5, turned to broad distribution, with maximum at SF = 0 after ECAP. It should be stressed that this texture development in the investigated alloy is quite uncommon compared to other Mg alloys. Usually, the initial as-extruded condition has a sharp - 1010 fiber texture, causing low values of SF for basal slip but high values of SF for twinning. ECAP processing results in the formation of a sharp and strong basal slip texture component, representing grains having their c-axis rotated by ~45◦ from all major axes. Consequently, significant texture softening occurs, which usually overwhelms grain boundary strengthening [27,28]. Therefore, the reason for the significantly higher increase of YCS as compared to *HV* values in the investigated alloy is the different texture evolution during ECAP. In magnesium alloys, the texture is typically formed during ECAP by the pronounced activation of a basal slip, or by combined activation of basal and non-basal slip systems [28,32]. The texture development in the investigated alloy can be explained according to the recently reported model on a ZN12 alloy. It was shown that a prismatic slip system has a relatively high activity during the hot rolling of the ZN12 alloy, and its effect on the texture is maintained by retarded recrystallization kinetics [38]. High activity of prismatic slip systems during ECAP causes a rotation of grains with their c-axis perpendicular to the processing direction (for details, see [28], for example). Consequently, significant grain boundary strengthening is even more promoted by the texture strengthening resulting from the processing.

However, a high volume fraction of grains having their c-axis perpendicular to the processing direction is favorable for the activation of tensile twinning during compression along the processing direction. This is also clearly visible in Figure 11b, where the distribution of the SF for tensile twinning is shown as well. High activity of twinning during the compression test also explains the formation of the sharp yield point in the ED-curve. It was shown previously that the formation of twin bands in the microstructure (Figure 9) can cause a drop in stress [39]. In order to compare changes in the deformation direction with the calculated SF for twinning, see Figures 11 and 12, which prove unambiguously that the probability of activation twinning in tension is significantly inferior to that in compression.

**Figure 11.** Schmid factor (SF) distribution calculated for major deformation mechanisms during the compression loading along the ED of (**a**) as-extruded sample (EX) and (**b**) 8P sample.

[

**Figure 12.** Schmid factor distribution calculated for major deformation mechanisms during the tensile loading along the ED of the 8P sample.
