Effects of Yttrium Addition on Plastic Deformation of Rolled Magnesium

Abstract

Tensile tests were applied to rolled Mg–Y alloy sheets with three at% of yttrium - Mg–0.54 at% Y, Mg–0.9 at% Y, and Mg–1.21 at% Y - to investigate effects of yttrium addition on activities of basal slip and non-basal slip systems so as to clarify the relationship between tensile properties and activities of slip systems. 0.2% proof stress of Mg–Y alloys increased with increasing yttrium content ranging from 0.5 to 1.2 at%. Ductility increased with increasing yttrium content until 0.9 at% but decreased when 1.2 at% was added. Frequencies of basal and non-basal slips increased by yttrium addition. The frequency of $\{ 10\bar{1}1\} \langle 11\bar{2}3\rangle $ first order pyramidal slip (FPCS) increased with increasing yttrium content until 0.9 at% and decreased when 1.2 at% was added. With increasing yttrium content, the frequency of $\{ 11\bar{2}2\} \langle 11\bar{2}3\rangle $ second order pyramidal slip (SPCS) decreased, while that of $\{ 10\bar{1}0\} \langle 11\bar{2}0\rangle $ prismatic slip (PS) increased. The highest total frequency of non-basal slips was observed in Mg–0.9Y, showing the highest ductility. Enhancement of ductility on magnesium was caused by activation of both basal and non-basal slips through yttrium addition.

 

This Paper was Originally Published in Japanese in J. Japan Inst. Met. Mater. 84 (2020) 44–49.

Fig. 5 Optical micrographs of slip lines in 0.5Y-1 (ε = (a) 1.7% and (b) 5.5%) and 1.2Y-2 (ε = (c) 2.0% and (d) 5.2%). Slip lines were identified to be BS, PS, FPCS and SPCS using the trace analysis. I and II indicate connecting points of BS and FPCS, and BS and SPCS, respectively.

1. Introduction

Light weight and high specific strength magnesium alloys are attractive materials for use in transport industries. However, magnesium shows low ductility at room temperature. Since the number of {0001}$\langle 11\bar{2}0\rangle $ basal slip (hereafter, BS), which is the main slip system of magnesium,¹⁾ is insufficient for von Mises criterion,²⁾ non-basal slips in the ⟨c + a⟩ direction: $\{ 10\bar{1}1\} \langle 11\bar{2}3\rangle $ first order pyramidal ⟨c + a⟩ slip (FPCS) and/or $\{ 11\bar{2}2\} \langle 11\bar{2}3\rangle $ second order pyramidal ⟨c + a⟩ slip (SPCS) must be activated to deform a crystal to any arbitrary shape. Therefore, activation of pyramidal slip systems in the ⟨c + a⟩ direction is key to improving ductility of magnesium.

Stanford et al.³⁾ evaluated effects of yttrium addition on the critical resolved shear stress (CRSS) of each slip system and twin system in Mg–Y alloys using in-situ neutron diffraction measurement, suggesting that ⟨c + a⟩ slip was activated. However, whether ⟨c + a⟩ slips were caused by FPCS or SPCS is unclear. Sandlöbes et al.⁴⁾ reported that Mg–3.0 mass%Y alloy polycrystals at room temperature had five times higher ductility than pure magnesium and tensile strength identical to pure magnesium. Also, cause of high ductility was concluded to be that ⟨c + a⟩ dislocations were activated by yttrium addition, based on TEM observations after tensile tests. However, which slip system types caused ⟨c + a⟩ dislocations and effects of yttrium addition on slip systems were unclear.

Trace analysis is suitable to distinguish slip system types.^5,6) Here, slip systems can be determined by observing angles of slip lines in each grain by optical microscopy and comparing those with crystal orientations of grains analyzed by electron backscatter diffraction (EBSD). As trace analysis can distinguish which FPCS and SPCS slip systems caused ⟨c + a⟩ slip, the effects of active slip systems on mechanical properties can be investigated. Using trace analysis, Rikihisa et al.⁷⁾ performed tensile tests of Mg–Y alloy single crystals and polycrystals and reported the effect of yttrium addition on pyramidal slips of magnesium. Mg–Y alloy single crystals yielded due to FPCS when yttrium addition was in the range of 0.6–1.1 at% while yielding due to $\{ 10\bar{1}0\} \langle 11\bar{2}0\rangle $ prismatic slip (PS) when yttrium addition exceeded 1.1 at%. Also, ductility was reported to improve due to increased activities of FPCS based on results of tensile tests of Mg–0.9 at% Y alloy polycrystals. However, such investigation was carried out solely in Mg–0.9 at%Y, not in Mg–Y alloys with other yttrium contents. Thus, neither effects of yttrium addition on activities of non-basal slip systems nor relationships between tensile properties and activation of slip systems remain unclarified. In this study, tensile tests were applied to rolled Mg–Y alloy sheets with various yttrium contents to investigate the effects of yttrium addition at% on activities of basal slip and non-basal slip systems so as to clarify relationship between tensile properties and activities of slip systems.

2. Experimental Procedures

Mg–0.54 at% Y and Mg–1.21 at% Y alloys were cast using magnesium and yttrium ingots with a high-frequency induction heating vacuum furnace; their compositions are shown in Tables 1 and 2. The composition of each alloy was analyzed by ICP emission spectroscopy. Cast alloys were hot-rolled at 523 K with rolling reduction approximately 10% per pass, totaling a reduction of 70∼75%. Figure 1 shows schematic illustrations of tensile specimens; tensile direction was parallel to rolling direction (RD). Tensile specimens were annealed in an argon atmosphere so as to obtain grain sizes ranging from 50 to 60 µm. Sample name, specimen number, annealing conditions, and mean grain sizes are summarized in Table 3. Four, one, and three specimens were prepared for Mg–0.54 at% Y (0.5Y), Mg–0.9 at% Y (0.9Y), and Mg–1.21 at% Y (1.2Y), respectively. Data of pure magnesium (Mg) and Mg–0.9 at% Y alloy (0.9Y-2)⁷⁾ were used for comparison. 0.9Y-2 was also produced by casting, rolled and annealing.⁷⁾ Specimens were mechanically polished using emery paper (# 400-4000), an MD-Nap (Struers), and diamond spray 1 mm (Struers). Specimen surfaces after mechanical polishing were chemically polished to mirror using a polishing cloth soaked in a solution (HNO₃:H₂O₂:C₂H₅OH = 5:7:20). For trace analysis, initial crystal orientations of the normal direction (ND) plane of specimens were analyzed using a FE-SEM (JEOL Ltd. JSM-7100F) and an EBSD camera (TexSEM Laboratories corp.) before tensile tests. Tensile tests were carried out at 293 K with the initial strain rate, $\dot{\varepsilon }$ = 8.3 × 10⁻⁴/s. Slip lines were observed using a Nomarsky type optical microscope (Nikon corp. ECLIPSE LV 150N) after tensile tests.

Table 1 Chemical composition of magnesium ingots. (mass%)

Table 2 Chemical composition of yttrium ingots. (mass%)

Fig. 1

Schematic illustration of tensile test specimen used in this study.

Table 3 Sample names, specimen numbers, annealing conditions and grain sizes of pure Mg and Mg–Y alloys.

3. Results

Figure 2 shows inverse pole figure (IPF) maps and (0002) pole figures of 0.5Y and 1.2Y before tensile tests. Basal texture of Mg–Y alloy has been reported to be weaker than that of Mg.^3,4,7) Rikihisa et al.⁷⁾ reported that Mg has basal texture with the maximum intensity of approximately 23. On the other hand, as shown in Fig. 2, texture intensity of Mg–Y alloys weakened when yttrium additions were 0.5 at% and 1.2 at%. However, texture intensity was independent of the amount of yttrium addition.

Fig. 2

IPF maps and (0002) pole figures of 0.5Y-1 ((a) and (b)) and 1.2Y-2 ((c) and (d)) before tensile tests.

Figure 3 shows typical stress-strain curves of 0.5Y and 1.2Y; arrows indicate 0.2 proof stress σ_0.2 and fracture strain ε_f. Curves of Mg⁷⁾ and 0.9Y⁷⁾ are also shown for comparison. Mg yielded at approximately 80 MPa and then showed rapid work hardening, while 0.9Y yielded at approximately 100 MPa and showed gradual work hardening behavior.⁷⁾ 0.5Y and 1.2Y showed gradual work hardening similar to 0.9Y. σ_0.2 and ε_f were 84 MPa and 20.5% in 0.5Y and were 123 MPa and 15.5% in 1.2Y. Figure 4 shows σ_0.2 and ε_f as a function of yttrium content. Note that the fracture strain of 0.9Y-1 was not plotted as the tensile test was discontinued before fracture. σ_0.2 increased with increasing yttrium addition. On the other hand, ε_f increased with increasing yttrium content until 0.9 at% but decreased when 1.2 at% was added.

Fig. 3

Stress-strain curves of Mg,⁷⁾ 0.5Y-4, 0.9Y-2⁷⁾ and 1.2Y-3.

Fig. 4

0.2% proof stress and fracture strain as a function of yttrium content.

Figure 5 shows optical micrographs of 0.5Y (ε = 1.7% and ε = 5.5%) and 1.2Y (ε = 2.0% and ε = 5.2%). In both specimens, slip lines were observed, and they were identified to be BS, FPCS, SPCS and PS by trace analysis. Also, non-basal slips seem to occur from grain boundaries. As shown in Fig. 5(b), $\{ 10\bar{1}2\} $ twins were observed, but the area was quite limited compared to Mg.⁷⁾ ⟨c + a⟩ slips were caused by FPCS or SPCS is clear. In addition, (a) I and (b) II in Fig. 5 indicate bent points of slip lines and BS slip lines and non-basal slip lines seem to connect. In particular, connections of BS and FPCS or SPCS were frequently observed.

Fig. 5

Optical micrographs of slip lines in 0.5Y-1 (ε = (a) 1.7% and (b) 5.5%) and 1.2Y-2 (ε = (c) 2.0% and (d) 5.2%). Slip lines were identified to be BS, PS, FPCS and SPCS using the trace analysis. I and II indicate connecting points of BS and FPCS, and BS and SPCS, respectively.

Figure 6 shows the relationship between strain and frequencies of dislocation slip in Mg–Y alloys and Mg.⁷⁾ Here, slip frequency was defined as the ratio of the number of grains with slip lines to the number of observed grains. The number of observed grains was 110 in 0.5Y-1, 152 in 0.9Y-1 and 122 in 1.2Y-2. The frequency of BS in Mg was 49% (ε = 4.0%).⁷⁾ On the other hand, in Mg–Y alloys, the frequency of BS was over 90% at a strain of approximately 2.0%. When yttrium was added, even in the early stage of deformation, it was found that the frequency of BS was high and that the frequency of non-basal slip increased. The frequency of FPCS increased with increasing yttrium content no more than 0.9 at% and decreased when 1.2 at% was added. With increasing yttrium content, the frequency of SPCS decreased, while that of PS increased. The total frequency of non-basal slip in 0.5Y (ε = 12.2%), 0.9Y (ε = 12.5%), and 1.2Y (ε = 13.0%), were 45%, 48%, and 42%, respectively. Sandlöbes et al.⁴⁾ reported that not only ⟨a⟩ dislocations but many ⟨c + a⟩ dislocations were observed by TEM after tensile tests in Mg–Y alloys. The TEM observation results agree with the present study that the slip frequency in Mg–Y alloys was higher than that in Mg.

Fig. 6

Relationship between strain and frequencies of BS, FPCS, SPCS and PS in Mg,⁷⁾ 0.5Y-1, 0.9Y-1 and 1.2Y-2.

Schmid factors (S.F.) of observed grains were calculated in each slip system, including the equivalent. S.F. for the actually activated slip system is S.F._act and the maximum is S.F._max. Figure 7 shows the relationship between S.F._act and S.F._max in each slip system in 0.5Y and 1.2Y. Plots are on the dashed lines when S.F._max and S.F._act are the same, that is, slip systems with maximum S.F. were activated. As shown in Fig. 7, most of plots of both FPCS and SPCS were not on the dashed lines and the slip system with S.F._max was found to be not necessarily activated. Conversely, most of plots of PS were on the dashed line, which was found that the slip system with S.F._max was activated. In addition, resolved shear stress (RSS) for each non-basal slip at a nominal strain of approximately 2.0% was calculated using the calculated S.F._act and the loading stress in tensile tests. They are shown in Fig. 8. Lines in Fig. 8 indicate CRSSs^7,8) for FPCS, SPCS, and PS of Mg and Mg–Y alloy single crystals calculated in tensile tests. In Mg, at a fracture strain of 4.0%, some RSSs for FPCS and SPCS were lower than CRSSs. On the other hand, in Mg–Y alloys, most FPCS and PS were activated even when their RSSs were lower than their CRSSs at a strain of approximately 2.0% at low loading stress. Here, CRSS for SPCS of Mg–Y alloy has been not reported. However, CRSS for SPCS is expected to be higher than that for FPCS since only FPCS was activated in tensile tests⁷⁾ of Mg–0.5 at% Y single crystals. Therefore, SPCS would be also activated when RSS for SPCS is lower than CRSS.

Fig. 7

Relationship between Schmid factors for actually activated slip systems, S.F._act, and maximum Schmid factors for non-basal slips, S.F._max, in 0.5Y-1 and 1.2Y-2: FPCS ((a) and (d)); SPCS ((b) and (e)); PS ((c) and (f)). Schmid factors of observed grains were calculated in each slip system, including the equivalent. S.F._max means maximum Schmid factors. When slip lines were actually observed, the S.F. of the slip system was defined as S.F._act. Plots are on the dashed lines when slip systems with maximum Schmid factor were activated.

Fig. 8

Resolved shear stresses at a nominal strain of approximately 2.0% in pure magnesium and Mg–Y alloys. Lines indicate CRSSs for FPCS, SPCS and PS.

4. Discussion

Yttrium addition has been reported to increase CRSS for BS in Mg–1.13 at% Y alloys⁹⁾ and CRSS for both FPCS and PS in Mg–(0.5∼1.3) at%Y alloy single crystals.⁷⁾ These results indicate that increasing σ_0.2 results from increasing CRSSs for slip systems due to yttrium addition. Also, BS and $\{ 10\bar{1}2\} $ twins were mainly activated but non-basal slips were hardly activated in plastic deformation of Mg.⁷⁾ However, the frequency of $\{ 10\bar{1}2\} $ twin decreased and those of both BS and non-basal slips increased when yttrium was added. Therefore, increasing ε_f would result from activities of both BS and non-basal slips in Mg–Y alloys. Here, the total frequency of non-basal slips was the highest when 0.9% yttrium was added, resulting in the highest ε_f of 0.9Y. Furthermore, changes in ε_f and frequency of FPCS as a function of yttrium content were in agreement, and the activity of FPCS in the ⟨c + a⟩ direction likely contributes to ductility.

Frequency of BS increased by yttrium addition as shown in Fig. 6, resulting from basal texture weakened by yttrium addition compared to Mg as shown in Fig. 2(c) and (d) and greater ease of BS activation. Also, FPCS frequency increased with increasing yttrium content until 0.9 at% and decreased when 1.2 at% was added, while PS frequency increased until 1.2 at% addition of yttrium, as shown in Fig. 6. It has been reported that Mg–Y alloy single crystals yielded due to FPCS when yttrium is added ranging from 0.6 to 1.1 at% of yttrium and yielded due to PS when yttrium addition exceeds 1.1 at%.⁷⁾ Therefore, easily activated slip systems likely change at a critical yttrium content of 0.9 at% in also Mg–Y alloy polycrystals.

Changes in activation of slip systems would be caused due to changes in CRSS for non-basal slip systems by yttrium addition, as described above. However, these slips were found to be activated even below CRSSs, as shown in Fig. 8 for reasons discussed below. Most of the slip lines causing by FPCS or SPCS occurred from grain boundaries; BS slip lines were observed near the grain boundaries in the neighboring grains, as shown in Fig. 5. Also, FPCS and SPCS with low S.F. were activated, as shown in Fig. 7. Therefore, non-basal slips would be activated though RSSs were lower than CRSSs due to stress concentration at grain boundaries by activation of BS and relaxations of mismatch strains at grain boundaries. Conversely, S.F. of activated PS were high, resulting in that connecting slip lines were barely observed compared to FPCS and SPCS. Therefore, to elucidate why non-basal slip systems were activated, m′ were calculated to evaluate orientation relationships between grains with BS lines and neighboring grains with non-basal slip lines.   

\begin{equation} m'=\cos\theta \cdot \cos\lambda \end{equation} (1)

Here, m′ is a value used to obtain a geometric orientation relationship between neighboring grains.^5,6) θ is the angle between two normal directions of slip planes and λ is the angle between two slip directions in neighboring slip systems. m′ of the actually activated slip system is expressed as m′_act, while maximum m′ in the equivalent of the actually activated non-basal slip is expressed as m′_max. Figure 9 shows the relationship between m′_max and m′_act in 0.5Y and 1.2Y. Plots are on the dashed lines when m′_act and m′_max were approximately the same - that is, slip systems with m′_max were activated. However, most plots in this study are not on the dashed lines. Therefore, though non-basal slips would be caused by the activation of BS in the neighboring grain, the cause could not be explained by m′. However, m′ were calculated using crystal orientation relationships between BS in a grain and non-basal slips in the neighboring grain. BS was actually also activated at the same time even in the grains where non-basal slips were activated. Therefore, activation of both BS and non-basal slips related to mismatch strains at grain boundaries will be evaluated and discussed near future.

Fig. 9

Relationship between maximum m′ of non-basal slips, m′_max, and m′ of activated slip system in (a) 0.5Y-1 and (b) 1.2Y-2. Plots are on the dashed lines when slip systems with m′_max were activated.

5. Conclusions

Tensile tests were applied to rolled Mg–Y alloy sheets with various contents of yttrium to investigate the correlation between yttrium contents and slip systems to clarify the effects of activities of slip systems on strength and ductility. Main results are as follows:

1. (1)    Yield stress increased with increasing yttrium content until 1.2 at% addition. Ductility increased with increasing yttrium content until 0.9 at% but decreased when 1.2 at% was added.
2. (2)    FPCS, SPCS and PS were activated when yttrium was added in the range between 0.5 to 1.2 at%.
3. (3)    Frequency of BS increased due to yttrium addition. Frequency of FPCS increased with increasing yttrium content until 0.9 at% but decreased when 1.2 at% was added. Also, frequency of SPCS increased until 0.5 at% of yttrium content but decreased with increasing yttrium content ranging from 0.9 to 1.2 at%, while frequency of PS increased with yttrium content until 1.2 at%.
4. (4)    The total frequency of non-basal slips in Mg–0.9 at%Y alloys was the highest, resulting in the highest fracture strain.
5. (5)    Resolved shear stresses (RSSs) for non-basal slips were lower than critical resolved shear stresses (CRSSs). This would be caused by stress concentrations due to activations of BS, but the cause of non-basal slip activation due to BS in the neighboring grain was not explained using m′.

Acknowledgements

A portion of the present study was financially supported by “The Light Metal Educational Foundation, Inc.”. The authors are very grateful for the support.

REFERENCES

• 1)   H.  Yoshinaga and  R.  Horiuchi: Trans. JIM 3 (1962) 220–226.
• 2)   R.  von Mises: Z. Angew. Math. Mech. 8 (1928) 161–185.
• 3)   N.  Stanford,  R.  Cottam,  B.  Davis and  J.  Robson: Acta Mater. 78 (2014) 1–13.
• 4)   S.  Sandlöbes,  S.  Zaefferer,  I.  Schestakow,  S.  Yi and  R.  Gonzalez-Martinez: Acta Mater. 59 (2011) 429–439.
• 5)   Y.  Guo,  T.B.  Britton and  A.J.  Wilkinson: Acta Mater. 76 (2014) 1–12.
• 6)   H.  Yu,  C.  Li,  Y.  Xin,  A.  Chapuis,  X.  Huang and  Q.  Liu: Acta Mater. 128 (2017) 313–326.
• 7)   H.  Rikihisa,  T.  Mori,  M.  Tsushida,  H.  Kitahara and  S.  Ando: Mater. Trans. 58 (2017) 1656–1663.
• 8)   H.  Tonda and  S.  Ando: Metall. Mater. Trans. A 33 (2002) 831–836.
• 9)   A.  Kula,  X.  Jia,  R.K.  Mishra and  M.  Niewczas: Int. J. Plast. 92 (2017) 96–121.