# Fileset

[JIM,66,2025,p905.pdf](https://mdr.nims.go.jp/filesets/cbc6fd2e-3982-447a-8759-6d079a687ca1/download)

## Creator

Elango Chandiran, [Yukiko Ogawa](https://orcid.org/0000-0002-7830-1597), [Rintaro Ueji](https://orcid.org/0000-0001-6969-3165), [Hidetoshi Somekawa](https://orcid.org/0000-0001-5007-5834)

## Rights

©2025 The Japan Institute of Metals and Materials[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Effect of Loading Direction on Compression Behaviour of Pure Magnesium at Different Grain-Size and Strain-Rates](https://mdr.nims.go.jp/datasets/240059f0-b0aa-483b-a397-3709e04c8328)

## Fulltext

Effect of Loading Direction on Compression Behaviour of Pure Magnesium at Different Grain-Size and Strain-RatesEffect of Loading Direction on Compression Behaviour of Pure Magnesium atDifferent Grain-Size and Strain-RatesElango Chandiran+, Yukiko Ogawa, Rintaro Ueji and Hidetoshi SomekawaResearch Center for Structural Materials, National Institute for Materials Science, Tsukuba 305-0047, JapanThis study investigates the effect of loading direction on the compression behaviour of extruded pure magnesium with different grain-sizesand at different strain-rate. At the same grain-size level, samples compressed at 45 degrees to the extrusion direction have lower yield stress thansamples compressed parallel to the extrusion direction. However, the loading direction has a negligible effect on the dominant deformationmodes in the studied conditions. [doi:10.2320/matertrans.MT-M2025008](Received January 27, 2025; Accepted March 24, 2025; Published April 18, 2025)Keywords: pure magnesium, grain-size, loading direction, compression, flow stress1. IntroductionWrought processing of magnesium (Mg) and its alloysresults in the development of a strong crystallographictexture, which significantly affects its room-temperatureformability and limits their use in commercial applications[1, 2]. It has been shown that modifying the crystallographictexture can improve the formability and specific strength-ductility balance of Mg and its alloys [2, 3]. The changes ininitial texture caused by variations in the angle between theloading direction and the c-axis of the crystal, can activatedifferent deformation mechanism [4]. The activation of non-basal slip mechanisms promotes more homogenous deforma-tion, improving ductility [5]. Peng et al. [6], in an extrudedAZ31 Mg alloy, showed that the dominant deformationmodes changed gradually during compression, with morepronounced increase in the angle between the loadingdirection and the c-axis during the tension testing.In addition to texture and loading direction, other factors,such as grain-size and strain-rates, significantly influencethe deformation behaviour of Mg alloys [4–8]. For instance,grain boundary sliding (GBS) is observed at room-temper-ature in a fine grain-size structures, particularly at lowerstrain-rates [7, 8]. However, the contribution of GBS is tooverall ductility remain limited, as dislocation slip remainsthe dominant deformation mechanism in the extruded Mg atroom-temperature [8]. Moreover, the influence of loadingdirection on the deformation behaviour further complicatesthe materials response in extruded samples. Despite advancesin understanding the effects of loading direction, grain-size,and strain-rate on deformation behaviour, their combinedeffect is not well understood. This study investigates howloading direction, grain-size, and strain-rate influence thecompression behaviour of pure Mg at room-temperature,focusing on yield stress and deformation modes to identifyfactors that could improve the mechanical properties of Mgalloys.2. Materials and methodsCommercial grade pure Mg (99.96% purity) was used inthis study. The cast pure Mg was extruded into bar of 8mm indiameter at 623K, 423K, and 378K with an extrusion ratioof 25:1 at an extrusion speed of 0.2mm/s. From the extrudedbars, cylindrical samples with compression axis (CA) parallelto the extrusion direction, (ED) (referred to as CA//ED) and45° to the extrusion direction (referred to as CA//45°ED)were prepared. The sample size was either 4mm in diameterand 8mm in height or 3mm in diameter and 6mm in height.These samples were compressed at room-temperature atstrain-rates of 1 © 10¹3 s¹1, 1 © 10¹4 s¹1 and 1 © 10¹5 s¹1.For selected deformation conditions, the samples werecompressed to intermediate strains ranging from 0.08 to0.12. The microstructures of the samples were examined ina plane parallel to the compression axis using a scanningelectron microscope equipped with a backscatter electrondetector. For microscopic observation, the samples weremechanically polished and etched in a 6% aqueous solutionof hydrochloric acid. The detailed sample preparationmethods for microstructure observation have been reportedelsewhere [8].3. Results and DiscussionFigure 1 shows representative microstructure and inversepole figure triangles along the compression axis (CA) ofsamples with CA//ED and CA//45°ED. The microstructuralfeatures, obtained from statistically sufficient EBSD measure-ments, show equiaxed grains, with grain-size decreasing asthe extrusion temperature decreases. In the coarse grain-size(d = 88 µm) sample with CA//ED, the ©10�10ª direction ofgrains are predominantly parallel to the CA, followed by©2�1�10ª (Fig. 1(b)). This indicates that the basal c-axes ofmost grains are nearly perpendicular to the CAwhich suggesta basal texture. On the other hand, in the sample with CA//45°ED (Fig. 1(c)), the grains were random along the loadingdirection. Similar behaviour is observed with decreasing thegrain-size (Fig. 1(d)–(i)). Therefore, observable differencesin the compression behaviour of these samples can be solelyattributed primarily to the grain-size.+Corresponding author, E-mail: CHANDIRAN.elango@nims.go.jpMaterials Transactions, Vol. 66, No. 7 (2025) pp. 907 to 911©2025 The Japan Institute of Metals and Materials RAPID PUBLICATIONhttps://doi.org/10.2320/matertrans.MT-M2025008Figure 2, summarizes the compression behaviour of thesamples. Note that the stress-strain curves for the coarse(d = 88 µm) and fine (d = 1.4 µm) grain-size samples withCA//ED are adapted from our previous work [8]. For thesame grain-size level, both the yield stress and the maximumstrength are lower in samples with CA//45°ED than sampleswith CA//ED (Fig. A1(a)). At a strain-rate of 1 © 10¹3 s¹1,strain hardening is observed after yielding in both coarsegrain-size (d = 88 µm) and meso grain-size (d = 6.8 µm)samples with CA//45°ED (Fig. 2(a)). On the other hand, thestrain hardening is less significant after yielding in the finegrain-size sample (d = 1.4 µm) (Fig. 2(a)). The decrementin strain hardening with grain refinement observed in theCA//45°ED samples is similar to that observed in the CA//ED samples. At a strain-rate of 1 © 10¹5 s¹1, the stress-strainbehaviour of the CA//45°ED sample is again similar to thatof the CA//ED samples (Fig. 2(b)).Figure 2(c) shows the variations in flow stress at 0.02plastic strain for different strain-rates. At strain-rates above1 © 10¹4 s¹1, the flow stress increases with grain refinementat the same strain-rate. Additionally, for the same grain-sizelevel, the flow stress is lower in samples with CA//45°EDcompared to those with CA//ED. For grain-sizes smallerthan 6.8 µm, the flow stress decreases significantly as strain-rates are reduced. At strain-rates below 1 © 10¹4 s¹1, thedecrease in the flow stress is considerably increased in thefine grain samples (d = 1.4 µm). The variation of flow stressat different strain-rate is similar to the yield stress(Fig. A1(a)). Moreover, the yield stress of the fine grainsample (d = 1.4 µm) at low strain-rate (1 © 10¹5 s¹1) arecomparable to that of the coarse grain-size (d = 88 µm) andmeso grain-size (d = 6.9 µm) samples.These results demonstrated that both yield and flow stressare strongly influenced by loading direction, grain-size, andstrain-rates. However, in terms of the slope of strain-rate vsflow stress, no significant difference is observed betweensamples with CA//45°ED and those with CA//ED. Sincethe slope of strain-rates vs flow stress is related to strain-ratesensitivity factor (m) [9], the similar slope observed in bothCA//45°ED and CA//ED sample sets, suggest that the thatdominant deformation modes are less likely to be influencedby the loading direction during compression.Figure 3 summarizes the deformation behaviour duringcompression of pure Mg with CA//45°ED. At a strain-rateof 1 © 10¹3 s¹1, the formation of deformation twins isfrequently observed in the coarse grain-size sample (d =88 µm) (Fig. 3(a)). On the other hand, in the fine grain-sizesample, such deformation twins are not observed (Fig. 3(b)).This suggest that grain refinement retards the formation ofdeformation twins since grain refinement is known toincrease the stress required for deformation twin formation[10].50 m10 m5 mIPF- Map CA//ED CA//45°EDd = 88 m (ET =350 °C)d = 6.9 m (ET =150 °C)d = 1.4 m (ET =105 °C)(a) (b) (c)(d) (e) (f)(g) (h) (i)Max – 3.6Max – 2.8Max – 2.4Max – 1.7Max – 1.7Max – 1.5Compression AxisFig. 1 (a) Inverse pole figure (IPF) map, (b) IPF triangles along compression axis parallel to extrusion direction (CA//ED) and (c) IPFtriangles along compression axis 45° to extrusion direction (CA//45°ED) for coarse grain-size sample (d = 88µm); Similarly (d), (e) and(f ) for meso grain-size sample (d = 6.9µm); and (g), (h) and (i) for fine grain-size sample (d = 1.4 µm); ET: extrusion temperature; ED:extrusion direction; CA: compression axis. (online color)E. Chandiran, Y. Ogawa, R. Ueji and H. Somekawa908At a strain-rate of 1 © 10¹5 s¹1, the formation of thedeformation twin is confirmed in the coarse grain-size sample(d = 88 µm) with CA//45°ED (Fig. A1(b)). At a strain of0.10, the twin fraction ( ftwin) was determined as 0.17, whichis similar to that of a coarse grain-size sample (d = 88 µm)with CA//ED ( ftwin = 0.16). In the fine grain-size sample(d = 1.4 µm) with CA//45°ED, deformation twins are notobserved (Fig. A1(c)). Furthermore, the surface of the finegrain-size sample (d = 1.4 µm) with CA//45°ED, exhibitswavy-like features (Fig. 3(c)), which are attributable to theGBS effect [7]. For Mg, m-values over 0.2 is reported toindicate a contribution from GBS to overall deformation atroom-temperature [11]. In the fine grain-size sample (d =1.4 µm) with CA//ED an m-value ² 0.3, was observed atstrain-rates below (1 © 10¹4 s¹1), demonstrating furthercontribution from GBS [8]. In the fine grain-size sample(d = 1.4 µm) with CA//45°ED, similar slopes were observedas in the case of the CA//ED sample (Fig. A1(a)). These20 m2 m50 m(a) d = 88 m, CA//45°ED(IQ)= 1x10-3 s-1ecomp = 0.08 = 1x10-3 s-1ecomp = 0.12 = 1x10-5 s-1ecomp = 0.5(b) d = 1.4 m, CA//45°ED(IQ)twins(c) d = 1.4 m, CA//45°ED(SEM)Compression AxisFig. 3 Image quality map (IQ) of sample with CA//45°ED at 1 © 10¹3 s¹1 grain (a) coarse grain-size sample (d = 88µm) after strain of0.12, (b) fine grain-size sample (d = 1.4 µm) after strain of 0.08, (c) SEM image of deformed surface of fine grain-size sample withCA//45°ED at 1 © 10¹5 s¹1.0501001502002503003504000 10 20 30 40 50 0 10 20 30 40 50050100150200250300350400Nominal stress (MPa)Nominal plastic strain (%)Nominal stress (MPa)Nominal plastic strain (%)d = 88 md = 6.9 md = 1.4 mSolid lines  – CA//45°EDDotted lines – CA//EDd = 88 md = 6.9 md = 1.4 mSolid lines  – CA//45°EDDotted lines – CA//ED(a) (b)10−6 10−4 10−2050100150200250Flow stress (eplastic= 0.02) (MPa)Strain rated = 88 md = 6.9 md = 1.4 mSolid lines  – CA//45°EDDotted lines – CA//ED(c)[8][8][8][8]Fig. 2 Nominal stress-plastic strain curves at strain-rates (a) 1 © 10¹3 s¹1 and (b) 1 © 10¹5 s¹1; (c) variation in flow stress at differentstrain-rates. (online color)Effect of Loading Direction on Compression Behaviour of Pure Magnesium at Different Grain-Size and Strain-Rates 909results suggests that in fine grain-size sample (d = 1.4 µm)CA//45°ED dislocation slip dominates deformation behav-iour, with GBS also contributing at low strain-rate (1 ©10¹5 s¹1). At strain-rates over 1 © 10¹4 s¹1, the m-valuedecreases, leading to a lesser contribution from GBS, and theoverall deformation is dominated by dislocation slip [12].In contrast, in the coarse grain-size samples (d = 88 µm),deformation is dominated by dislocation slip and twinning,irrespective of loading direction. Similar results werereported by Peng et al. [6], where for compression at 0° to61.9° from the extrusion direction of AZ31 with grain-size13 µm, deformation mechanism is dominated by basal slipand twinning in large volume fraction of grains. It is evidentthat the loading direction has a less significant effect onthe dominant deformation modes during room-temperaturecompression behaviour of extruded pure Mg, whereas grain-size has a significant influence.The deformation modes of pure Mg at ambient-temper-ature include dislocation slip on the basal, prismatic, andpyramidal planes, as well as twinning. Among these modes,basal slip is the primary deformation mode in polycrystallinemagnesium due to its typically lower critical resolved shearstress (CRSS) required for activation [13, 14]. Consequently,the yield stress is strongly influenced by the activation ofbasal slip, which is significantly dependent on the angularrelationship between the mechanical loading direction andthe crystallographic texture in wrought Mg [6, 14].According to the Schmid law for a single crystal, the yieldstress is inversely proportional to the critical resolved shearstress for a given direction [15].To understand the effect of loading direction on the yieldstress, the number fraction distribution of the Schmid factor(for basal slip) was determined for samples with CA//45°ED and CA//ED for both coarse grain-size (d = 88 µm)(Fig. 4(a)) and fine grain-size (d = 1.4 µm) (Fig. 4(b)). As,the change in loading directions (with CA//45°ED andCA//ED) was prepared by cutting the extruded samples at0° and 45° relative to the extrusion direction. The sampleswere fully recrystallized structures (Fig. 1). This suggeststhat changes in the yield stress is primary influenced by theloading direction and grain-size.In the coarse grain-size sample (d = 88 µm) with CA//45°ED, a large fraction of grains exhibits a higher Schmid factorfor basal slip compared to the samples with CA//ED(Fig. 4(a)). The average Schmid factor for basal slip in thesample with CA//45° ED is 0.33 (as shown in the insertedtable), which is higher than that in the sample with CA//ED(0.16). In the fine grain-size sample (d = 1.4 µm) with CA//45°ED, a larger fraction of grains also exhibits higher Schmidfactors for basal slip than the sample with CA//ED. Thisexplains the observed lower yield stress in the samplewith CA//45°ED compared to the sample with CA//ED(Figs. 1(a), 1(b) and Fig. A1(a)). Most grains in the CA//45°ED sample are oriented such that basal slip is readilyactivated upon compression. These results suggest that, forthe same grain-size, the loading direction significantly affectsthe yield stress in extruded bars of pure Mg (i.e., for sampleswith the same grain-size, CA//45°ED results in lower yieldstress than CA//ED). Furthermore, it is noteworthy thatdislocation-slip is primarily responsible for the initiation ofyielding, irrespective of grain-size, strain-rate, or loadingdirection.4. ConclusionThe effect of loading directions on the room-temperaturecompression behavior of extruded pure Mg with differentgrain-sizes was investigated. The loading direction affectsyield stress, with samples compressed at 45° to the extrusiondirection having lower yield stress. This lower yield stress isdue to a higher proportion of grains with a high Schmidfactor for basal slip. In the coarse grain sample (d = 88 µm),the deformation mode was dominated by slip anddeformation twinning. On the other hand, in a fine grainsample (d = 1.4 µm), the deformation is dominated by slip,and at low strain-rate (1 © 10¹5 s¹1), in addition to slip,grain boundary sliding also contributes to the deformationbehavior. Furthermore, in the studied conditions, thedominant deformation modes were found to be less affectedby the loading direction during room-temperature compres-sion of pure Mg.00.020.040.060.080.10.120.140.1 0.2 0.3 0.4 0.50.1 0.2 0.3 0.4 0.500.020.040.060.080.10.120.14Number fraction (%)(b) d = 1.4 mNumber fraction (%)Schmid factor(a) d = 88 mSchmid factorCA//45°EDCA//EDCA//EDCA//45°EDFig. 4 Number fraction distribution of Schmid factor for basal slip in (a) coarse grain-size sample (d = 88µm) and (b) fine grain-sizesample (d = 1.4µm).E. Chandiran, Y. Ogawa, R. Ueji and H. Somekawa910REFERENCES[1] T.T.T. Trang, J.H. Zhang, J.H. Kim, A. Zargaran, J.H. Hwang, B.-C.Suh and N.J. Kim: Designing a magnesium alloy with high strengthand high formability, Nat. Commun. 9 (2018) 2522.[2] J.F. Nie, K.S. Shin and Z.R. Zeng: Microstructure, deformation, andproperty of wrought magnesium alloys, Metall. Mater. Trans. A 51(2020) 6045–6109.[3] T. Mukai, M. Yamanoi, H. Watanabe and K. Higashi: Ductilityenhancement in AZ31 magnesium alloy by controlling its grainstructure, Scr. Mater. 45 (2001) 89–94.[4] Y. Wang and H. Choo: Influence of texture on Hall–Petch relationshipsin an Mg alloy, Acta Mater. 81 (2014) 83–97.[5] R. Gehrmann, M.M. Frommert and G. Gottstein: Texture effects onplastic deformation of magnesium, Mater. Sci. Eng. A 395 (2005)338–349.[6] J. Peng, Z. Zhang, H. Chen, C. Long, Y. Wu, W. Zhou and Y. Wu:Investigation on the anisotropy of mechanical properties alongdifferent orientations of an AZ31 hot-extrusion bar, J. Alloy. Compd.854 (2021) 157108.[7] H. Somekawa, A. Singh, R. Sahara and T. Inoue: Excellent roomtemperature deformability in high strain rate regimes of magnesiumalloy, Sci. Rep. 8 (2018) 656.[8] E. Chandiran, Y. Ogawa, R. Ueji and H. Somekawa: An inverse Hall-Petch relationship during room-temperature compression of commer-cially pure magnesium, J. Alloy. Compd. 930 (2023) 167443.[9] M. Barnett, Z. Keshavarz, A. Beer and D. Atwell: Influence of grainsize on the compressive deformation of wrought Mg–3Al–1Zn, ActaMater. 52 (2004) 5093–5103.[10] F.A. Nichols: Plastic instabilities and uniaxial tensile ductilities, ActaMetall. 28 (1980) 663–673.[11] Z. Zeng, J.F. Nie, S.W. Xu, C.H.J. Davies and N. Birbilis: Super-formable pure magnesium at room temperature, Nat. Commun. 8(2017) 972–977.[12] H. Somekawa, A. Singh and T. Inoue: Development of Isotropic andAccordion-like Deformable Magnesium Alloys, Mater. Trans. 58(2017) 1089–1092.[13] H. Yoshinaga and R. Horiuchi: On the Flow Stress of α Solid SolutionMg-Li Alloy Single Crystals, Trans. JIM 4 (1963) 134–141.[14] S. Kleiner and P.J. Uggowitzer: Mechanical anisotropy of extrudedMg–6% Al–1% Zn alloy, Mater. Sci. Eng. 379 (2004) 258–263.[15] X.-L. Nan, H.-Y. Wang, L. Zhang, J.-B. Li and Q.-C. Jiang:Calculation of Schmid Factors in Magnesium: Analysis ofDeformation Behaviors, Scr. Mater. 67 (2012) 443–446.Appendix2 m50 m(c) d = 1.4 m, CA//ED(IQ)(b) d = 88 m, CA// 45°ED(IQ)= 1x10-5 s-1ecomp = 0.5 = 1x10-5 s-1ecomp = 0.12 twinsCompression Axis10−6 10−4 10−2050100150200250Yield stress (MPa)Strain rateSolid lines  – CA//45°EDDotted lines – CA//EDd = 88 md = 6.9 md = 1.4 m(a)Fig. A1 (a) Variation in yield stress at different strain-rates. IQ map of sample with CA//45°ED at 1 © 10¹5 s¹1 (b) coarse grain-sizesample (d = 88µm) after strain of 0.12, (c) fine grain-size sample (d = 1.4 µm) compressed to strain of 0.5. (online color)Effect of Loading Direction on Compression Behaviour of Pure Magnesium at Different Grain-Size and Strain-Rates 911https://doi.org/10.1038/s41467-018-04981-4https://doi.org/10.1007/s11661-020-05974-zhttps://doi.org/10.1007/s11661-020-05974-zhttps://doi.org/10.1016/S1359-6462(01)00996-4https://doi.org/10.1016/j.actamat.2014.08.023https://doi.org/10.1016/j.msea.2005.01.002https://doi.org/10.1016/j.msea.2005.01.002https://doi.org/10.1016/j.jallcom.2020.157108https://doi.org/10.1016/j.jallcom.2020.157108https://doi.org/10.1038/s41598-017-19124-whttps://doi.org/10.1016/j.jallcom.2022.167443https://doi.org/10.1016/j.actamat.2004.07.015https://doi.org/10.1016/j.actamat.2004.07.015https://doi.org/10.1016/0001-6160(80)90144-3https://doi.org/10.1016/0001-6160(80)90144-3https://doi.org/10.1038/s41467-017-01330-9https://doi.org/10.1038/s41467-017-01330-9https://doi.org/10.2320/matertrans.M2017108https://doi.org/10.2320/matertrans.M2017108https://doi.org/10.2320/matertrans1960.4.134https://doi.org/10.1016/j.msea.2004.02.020https://doi.org/10.1016/j.scriptamat.2012.05.042