**

aKey Laboratory of Interface Science and Engineering in Advanced Materials,Ministry of Education,Taiyuan University of Technology,Taiyuan 030024,China

bSchool of Materials Science and Engineering,Taiyuan University of Technology,Taiyuan 030024,China

cKorea Institute of Materials Science,Changwon 642-831,Republic of Korea

Strengthening mechanisms of indirect-extruded Mg-Sn based alloys at room temperature

Wei Li Chenga,b,*,Quan Wei Tianb,Hui Yuc,Hua Zhanga,b,*,Bong Sun Youc

aKey Laboratory of Interface Science and Engineering in Advanced Materials,Ministry of Education,Taiyuan University of Technology,Taiyuan 030024,China

bSchool of Materials Science and Engineering,Taiyuan University of Technology,Taiyuan 030024,China

cKorea Institute of Materials Science,Changwon 642-831,Republic of Korea

The strength of a material is dependent on how dislocations in its crystal lattice can be easily propagated.These dislocations create stress f i elds within the material depending on their intrinsic character.Generally,the following strengthening mechanisms are relevant in wrought magnesium materials tested at room temperature:f i ne-grain strengthening,precipitate strengthening and solid solution strengthening as well as texture strengthening.The indirect-extruded Mg-8Sn(T8)and Mg-8Sn-1Al-1Zn(TAZ811)alloys present superior tensile properties compared to the commercial AZ31 alloy extruded in the same condition.The contributions to the strengthen of Mg-Sn based alloys made by four strengthening mechanisms were calculated quantitatively based on the microstructure characteristics,physical characteristics,thermomechanical analysis and interactions of alloying elements using AZ31 alloy as benchmark.

Mg alloys;Microstructure;Mechanical properties;Strengthening mechanism

1.Introduction

Mg-Sn system is known as a precipitation system,which has a relatively high solubility limit(3.35 at%)at 834 K and low solubility at ambient temperature.This is an age hardenable system with the potential to form 10 vol.%of Mg2Sn.The Mg2Sn precipitate(FCC,a=0.676 nm,point group m 3m)has a high melting temperature(770°C)and alloys based on this system are thought to show some promise for use in applications requiring elevated temperature creep resistance[1-3].In recent years,there is an increasing interest in the development of high strength wrought Mg-Sn based alloys.The results indicated that the Mg-Sn based alloys show higher tensile strength and similar elongations compared to a commercial AZ31 alloy[4-6].

The strength of a material is dependent on how dislocations in its crystal lattice can be easily propagated.These dislocations create stress f i elds within the material depending on their intrinsic character.Generally,the following strengthening mechanisms such as grain boundary strengthening(GBS), precipitates strengthening(PS),texture strengthening(TS)and solid solution strengthening(SS)are relevant in extruded and rolled magnesium materials tested at room temperature [7-10].The contributions to the strengthen of the alloy made by four strengthening mechanisms were calculated quantitatively based on the microstructure characteristics,physical characteristics,thermomechanical analysis and interactions of alloying elements using AZ31 alloy as benchmark and compared with those of the experimental results.

2.Experimental

Analyzed compositions of the T8,TAZ811 and AZ31 were Mg-8.0 wt.%Sn,Mg-7.95 wt.%Sn-0.95Al-0.95Zn and Mg-2.90 wt.%Al-0.69 wt.%Zn-0.32 wt.%Mn,respectively. Alloys were prepared from high purity(99.99%)Mg,Sn,Zn and Al by induction melting in a cemented graphite crucible at approximately 720°C under a CO2+SF6atmosphere and casting into a steel mold pre-heated to 200°C.After casting, T8 and TAZ811 alloys were homogenized at 500°C for 3 h and then water-quenched to induce a supersaturated solid solution.In the case of AZ31,the billet was homogenized at 400°C for 24 h and cooled by air.Afterward,the billets were extruded at 250°C with a ram speed of 1.3 mm/s(corresponding to extrusion speeds of 2 m/min)and extrusion ratio of 25.

Microstructural and textural examinations were conducted in the longitudinal section parallel to the extrusion direction (ED).For microstructure observations,specimens were etched after polishing in a solution of picric and acetic acid for 10 s. The average grain size was analyzed from several micrographs using a liner intercept measurement.Scanning electron microscopy(SEM)and transmission electron microscopy(TEM) were used to study the morphology,characterization and volume faction of second-phases particles.Texture measurements were taken via X-ray diffraction in the back ref l ection mode with monochromatic Cu Kα radiation.

Tensile test was conducted using an Instron 4206 universal testing machine equipped with 10 mm gauge extensometer. The extruded rods were machined into tensile samples with a gage length of 25 mm and a gage diameter of 5 mm.All tensile tests were carried out with an initial strain rate of 1 × 10-3s-1at ambient temperature,where the tensile direction was the same as the extrusion direction.

3.Results and discussion

3.1.Tensile properties of the indirect-extruded alloys

Typical stress-strain curves of the three alloys are shown in Fig.1.Related tensile properties are summarized in Table 1. There is a signif i cant difference in the mechanical behavior of the three alloys.TAZ811 showed the highest tensile strength and medium ductility compared to the other two alloys,having a failure strain of 17.5%.T8 showed a similar ultimate tensile strength(UTS)to AZ31(261 MPa vs 263 MPa);however,T8 showed a higher yield point compared to AZ31,having yield stress of 180 MPa.AZ31 showed a much lower yield point compared to the other two alloys,being about 73%of the value in the cases of TAZ811 alloy,while AZ31 shows a signif i cantly increased ductility of 23.5%,being almost 1.3 times as ductile as TAZ811.Generally speaking,grain ref i nement of materials can lead to the improvement in strength and ductility simultaneously.However,the Mg-Sn based alloys(d=3.2 and 4.7 μm)accompanying with the deteriorated ductility exhibited higher strength than AZ31 alloy(d=9.6 μm).This is due to the following aspects:

Fig.1.Stress-strain curves of the three alloys at room temperature tensile deformation.

(i)The easily occurred double twinning related to stronger if ber texture(see Section 3.2)accelerates cracking, which is induced by dislocation pile-ups at the twinmatrix interface[5].

(ii)The value of friction stress and the stress concentration factor in Hall-Petch(H-P)relationship are texture dependence.Stronger fi ber texture(see Section 3.2) results in grains in “hard orientation”(dif fi cult activation for basal slip,namely poor ductility),which leads to better strengthening,i.e.,higher friction stress and H-P slope[6].

3.2.Microstructure characteristics of the indirectextruded alloys

The main differences among the three alloys lied in the grain size,texture and second-phase particles,as shown in Figs.2 and 3.As shown in Fig.2,AZ31 shows the average grain size of 9.6 μm,while the grain size is similar in T8 and TAZ811,but appear to be less in TAZ811 with average grain size being 3.2 μm.In addition,second-phase particles below 1 μm were observed along the grain boundaries and within the Mg matrix in Mg-Sn based alloys.These second-phase particles are visible in Fig.2(b)as white regions and dark regions in Fig.2(d)and(f).It is also noticeable that the volume fraction of the second-phase particles in the three alloys is quite different.In the case of AZ31 alloy,few particles are visible,indicating that dynamic precipitate ability is weak for AZ31 during extrusion processing.While,Mg-Sn based alloys show a signif i cant amount of Mg2Sn particles inside thegrain and at grain boundary,but this is more extensive as in TAZ811.It has been reported that intermetallic phase particles can have a retardation effect on moving grain boundaries during grain growth[2].So it is believed the differences of grain size between TAZ811 and AZ31 are attributed to the different amount of second-phase particles.

Table 1Tensile properties and grain sizes of the three extruded alloys.

Fig.2.Optical,SEM and TEM micrographs of the three alloys,where ED is parallel to the scale bar.(a,b)AZ31,(c,d)T8(inset:[-112]Mg2Sn diffraction pattern) and(e,f)TAZ811.

The inverse pole f i gures(IPFs)referring to the ED are provided in Fig.3.They all reveal a type of f i ber texture in which basal poles are preferentially perpendicular to the ED and the maximum intensity is centered at[10-10],which is typical of extruded Mg alloys[2,6,10].As indicated,TAZ811 alloy shows the strongest f i ber texture with the maximal texture intensity of 4.4,T8 shows the moderate texture with the maximal texture intensity of 3.9 and AZ31 shows the weakest texture with the maximal texture intensity of 2.8.The relatively strong texture of Mg-8Sn based alloys was mainly due to the presence of deformed grains elongated in the ED based on the previous results[2,6].In general,the texture of HCP materials is determined by theirc/aratio as well as by the active slip systems or twinning during deformation.It is wellknown that a development of strong basal texture during plane strain compression is mainly caused by the activation of basal slip and tension twinning,and,in particular,that the occurrence of extension twins induces a large reorientation of gains [12].Subsequent recrystallization would lead to the formation or strengthening of a basal texture[3].

3.3.Strengthening mechanisms in the indirect-extruded alloys

3.3.1.Grain boundary strengthening

Grain boundary strengthening presented by the well-known Hall-Petch relation[6],is an established method of increasing the yield stress and is the main contributor to the improved mechanical properties.Calculation by the Hall-Petch relation was applied to extruded Mg alloys,

where σ0andkis experimental constants anddis the grain size in μm.The value ofkis determined by Ref.[11]:

Fig.3.Inverse pole f i gures of the three alloys in the ED(a)TAZ811,(b)T8 and(c)AZ31 alloys.

wheremis the Taylor factor, τcis the empirical stress for slip to break through grain boundaries, υ=0.29,andbis the magnitude of the Burgers vector(3.21 × 10-10m)for Mg [12].

When τcis the constant at the condition ofd＞ 1 μm,kis the constantfor Mg.The calculated results indicate that the increment in yield strength due to grain ref i nement from 9.6 to 4.7 and 3.2 μm is about 38.8-44.3 MPa and 66-75 MPa,respectively.It suggests that strengthening of the Mg-Sn based alloys compared to AZ31 isnotjustcoming from the grain boundary strengthening.

3.3.2.Precipitates strengthening

Precipitates of metastable transition or equilibrium phases are often key strengthening constituents in many magnesium alloys[13,14].The precipitation strengthening involves the following four processes with the corresponding contributions [15]:

1)The dislocation-particle interaction associated with the Orowan process,Δσ0;

2)The load transfers from the matrix to particles, ΔσT;

3)The generation of dislocations due to the difference between the thermal expansions of the matrix and particles, Δσg;

4)The generation of dislocations due to the geometric requirements during deformation, Δσf.

The strength of the extruded solid material can be represented in the form:

The version of the Orowan equation currently applied for the stress required to bow slip dislocations across spherical particles is[16,17]:

whereGis the shear modulus(=1.66 × 104MPa for magnesium),bis the magnitude of the Burgers vector (3.21 × 10-10m),ν is the Poisson ratio(0.29 for magnesium),λ is the interparticle spacing,dpis the effective particle diameter ontheslipplaneandr0isthedislocationcoreradius(commonly, the approximationr0≈bis used).The contributions from Orowan equation for TAZ811,T8 and AZ31 are 3.2 MPa, 1.70 MPa and 0.7 MPa,respectively.So the relative values for TAZ811 and T8 are 2.5 MPa and 1.0 MPa,respectively.

The contribution of the load transferring from the matrix to particles can be def i ned by the relationship:

where σs(100-150 MPa for Mg alloys[15])is the yield strength of the matrix(determined by the supersaturated magnesium solid solution)and the volume fraction(fv)of dispersed particles in TAZ811,T8 and AZ31 are 14.7%,8.7% and 3.2%,respectively(see Fig.2).The contributions of this process are approximately equal to 7.4-11 MPa,4.4-6.5 MPa and 1.2-2.3 MPa for TAZ811,T8 and AZ31,respectively.So the relative values for TAZ811 and T8 are 6.2-8.7 MPa and 3.2-4.2 MPa,respectively.

For the globular precipitate in present alloys,dpcan be calculated by Eq.(3-6)[16]:

The strengthening due to the difference between the thermal expansions of the matrix and particles is given by Refs. [15,18,19]:

where α is a constant, ΔTis the temperature increment, ΔC（CMg-CMg2Sn）is the difference between the thermal expansion coeff i cients of the matrix and particles under investigation,anddpis the mean planar diameter of the obstacles. When α = 1.25 andfv= 12.6%,CMg=2.61 × 10-5K,andCMg2Snis about 4.5 × 10-6K,dp= 0.32 μm for Mg2Sn precipitate in TAZ811 anddp=0.68 μm for Mg2Sn precipitate in T8.In the present study,all the tests were carried at room temperature and then this part of the contributions can be neglected.

The contribution associated with the generation of dislocations due to the geometric requirements during deformation can be written as[15,19]:

where γ is the shear strain calculated using the Taylor factor. Then the contribution is estimated to be 17,10.8 MPa and 6.3 MPa for TAZ811,T8 and AZ31,respectively.

In summary,the relative contributions from precipitate strengthening are about 19.4-22.3 MPa for TAZ811 and 8.7-9.7 MPa for T8,respectively.

3.3.3.Texture strengthening

The texture inf l uences the strength of hexagonal close packed (HCP)polycrystalsthrough itseffecton the Hall-Petch constants[20].The nature and intensity of texture decides the magnitude of the orientation factormin Eq.(3-9).

In Eq.(3-9),the value of the Taylor orientation factormis related to the basal texture in the material.When the texture is unfavorable for the occurrence of basal slip,the value ofmincreases and the material is strengthened.In randomly oriented HCP polycrystal(without texture),mhas a value of～6.5.The extruded texture in magnesium materials(basal planes parallel to the extrusion direction)is unfavorable for the occurrence of basal slip and hence themvalues are larger than 6.5.If texture strengthening is the only factor contributing to σ0,the value ofmcan be calculated by knowing the texture intensity.For AZ31 alloy,mis 18.2,6.5 times the texture intensity.Similarly,m=25.35 for T8,m=28.6 for TAZ811 alloy.On the basis of Eq.(3-9)the σ0values for Mg-Sn based alloys may be calculated knowing the relevant single-crystal resolved stress(τ0)and using the abovemvalues.Based on the previous reports[12],CRSSs of basal slip,prismatic slip,pyramidal slip and tension twinning of Mg at elevated temperatures are ranged from 0.6 to 4 MPa. Due to the lack of data at room temperature,a mean value of 2.3 was adopted in this study.Therefore,the relative contributions from texture strengthening are about 2.77 and 3.61 MPa for T8 and TAZ811,respectively.

3.3.4.Solid solution strengthening

It has been reported[21-23]that solutes such as Al,Zn and Sn increase the critical resolved shear stress for basal slip by an amount that is proportional toc2/3,wherecis the atomic concentration.If we assume here that Sn,Al and Zn atoms are present together without interacting with each other,the strengthening effect due to multiple alloying additions in Mg-Sn-Al-Zn ternary alloys might be determined using the method of Gypen and Deruyttere[24-26]:

wherenis a constant,Ciis the concentration of solutei,andkiis the strengthening constant for solutei.Theoretical treatments indicatencould equal 2/3,1 or 1/2[23,24].Thenvalue was determined as 2/3 in this study.For solid solution strengthening calculation,the parameters for Zn,Sn and Al atom in Mg are determined and shown in Table 2.

The amount of Sn atoms dissolved into matrix can be determined by Eq.(3-11).

where ρMg-Sn=1.92 g/cm3(densities of Mg-Sn based alloys), ρMg2Sn=3.59 g/cm3(densities of Mg2Sn phase).

Therefore,the solid solution strength ening caused by Sn in T8 and TAZ811 are bout 34 MPaand 19 MParespectively.It should be noted that the higher strengths of TAZ811 compared with those of T8 is mainly attributed to the grain boundary strengthening and solid solubility strengthening as well as texture strengthening mechanisms resulting from the smaller grain size and higher solid solubility of Al and Zn in Mg matrix as well as stronger f i ber intensities leading to grains with “hard orientation”in Mg matrix.

3.3.5.Strengthening effect due to multiple alloying additions of Al and Zn

Based on the previous literature and phase diagrams [1,5,6,27],it is assumed that all the Al and Zn atoms were dissolved into matrix,so theCAlandCZnin AZ31 is 2.6 at% and 0.37 at%,CAlandCZnin TAZ811 is 0.96 at%and 0.39 at%,respectively.So the compensation strength caused by Al and Zn between AZ31 and TAZ811 alloys is about7.4 MPa.While the compensation strength caused by Al and Zn between AZ31 and T8 alloys is about 41.2 MPa.

Table 2Solid solution strengthening parameters for Zn,Sn and Al atoms in Mg.

Table 3The respective contributions of the four strengthening mechanisms to Mg-Sn based alloys using AZ31 alloy as benchmark.

The respective contributions of the four strengthening mechanisms are summarized in Table 3.As indicated,there is some deviation between calculated and experimental values due to the deviations of the variates and parameters used in the equations and the possible defects which come from the preparation process of the alloy.Further work is required to determine the accurate values of the parameters in the above mentioned equations.Anyway,the variation tendency in the contribution of the four strengthen mechanisms can be referred as a basis in design and development of high strength wrought Mg alloys.

4.Summaries

1)Grain size, recrystallization fraction, precipitate morphology and texture of the indirect-extruded alloys were greatly affected by the compositions,resulting in the Mg-Sn based alloys showing higher tensile strengths than AZ31 alloy.

2)Mg-Sn based alloys were mainly strengthened by grain boundary strengthening,solid solution strengthening, second-phase precipitates strengthening as well as texture strengthening mechanisms.

3)When the alloyswere strengthened fully by four strengthening mechanism using AZ31 as benchmark,the calculated total contribution should reach to 43.27-112.5 MPa which is considerably larger than the experimental values.The difference between the calculated values and experimental values comes from the error of the parameters chosen in calculation and the possible defects which come from the preparation process of the alloy.

Acknowledgments

This work was supported by National Natural Science Foundation of China(Grant nos.51404166 and 51201112), Shanxi Province Science Foundation for Youths(2013021013-4),Research Project Supported by Shanxi Scholarship Council of China(2014-023),Technological Innovation Programs of Higher Education Institutions in Shanxi(Grant nos.2014120), and the Advanced Programs of Department of Human Resources and Social Security of Shanxi Province for Returned Scholars(2013068).

[1]S.S.Park,W.N.Tang,B.S.You,Mater.Lett.64(2010)31-34.

[2]W.L.Cheng,S.S.Park,B.S.You,B.H.Koo,Mater.Sci.Eng.A 527 (2010)4650-4653.

[3]Mingbo Yang,Hongliang Li,Chengyu Duan,Jia Zhang,J.Alloys Compd.579(2013)92-99.

[4]T.T.Sasaki,J.D.Ju,K.Hono,K.S.Shin,Scr.Mater.61(2009)80-83.

[5]W.L.Cheng,H.S.Kim,B.S.You,B.H.Koo,S.S.Park,Mater.Lett.65 (2011)1525-1527.

[6]W.Yuan,S.K.Panigrahi,J.-Q.Su,R.S.Mishra,Scr.Mater.65(2011) 994-997.

[7]T.Laser,C.Hartig,M.R.Nurnberg,D.Letzig,R.Bormann,Acta Mater. 56(2008)2791-2798.

[8]Y.N.Wang,J.C.Huang,Acta Mater.55(2007)897-905.

[9]C.L.Mendis,K.Ohishi,Y.Kawamura,T.Honma,S.Kamado,K.Hono, Acta Mater.57(2009)749-760.

[10]L.B.Tong,X.Li,D.P.Zhang,L.R.Cheng,J.Meng,H.J.Zhang,Mater. Charact.92(2014)77-83.

[11]Z.Yang,J.P.Li,Y.C.Guo,T.Liu,F.Xia,Z.W.Zeng,M.X.Liang,Mater. Sci.Eng.A 454-455(2007)274-280.

[12]H.J.Frost,M.F.Ashbv,Deformation-mechanism Maps,Pergamon Press, Oxford,1982,p.44.

[13]J.F.Nie,B.C.Muddle,Acta Mater.48(2000)1691-1703.

[14]A.Singh,A.P.Tsai,M.Nakamura,Philos.Mag.Lett.83(2003) 543-551.

[15]M.Mabuchi,K.Higashi,Acta Mater.44(1996)4611-4618.

[16]J.F.Nie,Scr.Mater.48(2003)1009-1015.

[17]R.M.Aikin Jr.,L.Christodoulou,Scr.Metall.Mater.25(1991)9-14.

[18]W.S.Miller,F.J.Humphreys,Scr.Metall.Mater.25(1991)33-38.

[19]J.W.Luster,M.Thumann,R.Baumann,Mater.Sci.Technol.9(1993) 853-862.

[20]G.S.Rao,Y.V.R.K.Prasad,Metall.Trans.A13(1982)2219-2226.

[21]C.H.Caceres,A.Blake,Phys.Status Solidi A 194(2002)147-158.

[22]L.Gao,R.S.Chen,E.H.Han,J.Alloys Compd.481(2009)379-384.

[23]S.M.He,X.Q.Zeng,L.M.Peng,X.Gao,J.F.Nie,W.J.Ding,J.Alloys Compd.427(2007)316-323.

[24]L.A.Gypen,A.Deruyttere,J.Mater.Sci.12(1977)1028-1033.

[25]H.A.Roth,Metall.Mater.Trans.A 28(1997)1329-1335.

[26]B.Q.Shi,R.S.Chen,W.Ke,J.Alloys Compd.509(2011)3357-3362.

[27]W.L.Cheng,S.S.Park,W.N.Tang,B.S.You,B.H.Koo,Trans. Nonferrous Met.Soc.China 20(2010)2246-2252.

Received 24 September 2014;revised 14 November 2014;accepted 18 November 2014 Available online 9 December 2014

*Corresponding authors.Key Laboratory of Interface Science and Engineering in Advanced Materials,Ministry of Education,Taiyuan University of Technology,Taiyuan 030024,China.

E-mail addresses:chengweili7@126.com(W.L.Cheng),zhanghua2009@ 126.com(H.Zhang).

Peer review under responsibility of National Engineering Research Center for Magnesium Alloys of China,Chongqing University.

http://dx.doi.org/10.1016/j.jma.2014.11.003.

2213-9567/Copyright 2014,National Engineering Research Center for Magnesium Alloys of China,Chongqing University.Production and hosting by Elsevier B.V.All rights reserved.

Copyright 2014,National Engineering Research Center for Magnesium Alloys of China,Chongqing University.Production and hosting by Elsevier B.V.All rights reserved.