(7.8)--Origin of the reversed yield asy机械工程材料机械工程材料.pdf

《(7.8)--Origin of the reversed yield asy机械工程材料机械工程材料.pdf》由会员分享，可在线阅读，更多相关《(7.8)--Origin of the reversed yield asy机械工程材料机械工程材料.pdf（13页珍藏版）》请在得力文库 - 分享文档赚钱的网站上搜索。

1、Origin of the reversed yield asymmetry in Mg-rare earth alloys at hightemperatureP.Hidalgo-Manrique,a,V.Herrera-Solaz,bJ.Segurado,a,bJ.Llorca,a,bF.Ga lvez,bO.A.Ruano,cS.B.Yidand M.T.Pe rez-PradoaaIMDEA Materials Institute,C/Eric Kandel 2,28906 Getafe,Madrid,SpainbDepartment of Materials Science,Poly

2、technic University of Madrid,E.T.S.de Ingenieros de Caminos,28040 Madrid,SpaincDepartment of Physical Metallurgy,CENIM-CSIC,Av.Gregorio del Amo 8,28040 Madrid,SpaindMagnesium Innovation Centre MagIC,Helmholtz-Zentrum Geesthacht,Max-Planck-Strasse 1,D-21502 Geesthacht,GermanyReceived 10 October 2014;

3、revised 30 March 2015;accepted 30 March 2015Available online 22 April 2015AbstractThe mechanical behaviour in tension and compression of an extruded Mg1 wt.%Mn1 wt.%Nd(MN11)alloy was studied along theextrusion direction in the temperature range?175?C to 300?C at both quasi-static and dynamic strain

4、rates.Microstructural analysis revealed thatthe as-extruded bar presents a recrystallized microstructure and a weak texture that remain stable in the whole temperature range.A remarkablereversed yield stress asymmetry was observed above 150?C,with the compressive yield stress being significantly hig

5、her than the tensile yield stress.The origin of this anomalous reversed yield stress asymmetry,which to date remains unknown,was investigated through the analysis of the macroand microtexture development during deformation,as well as by means of crystal plasticity finite element simulations of a rep

6、resentative volumeelement of the polycrystal.The critical resolved shear stresses of slip and twining for simulated single crystals were obtained as a function of thetemperature by means of an inverse optimisation strategy.Experimental and simulation results suggest that the reversed yield asymmetry

7、 may beprimarily attributed to the non-Schmid behaviour of pyramidal hc+ai slip,which is the dominant deformation mechanism at high temperatures.It is proposed,furthermore,that the asymmetry is enhanced at quasi-static strain rates by the stronger interaction of hc+ai dislocations withthe diffusing

8、solute atoms and particles in compression than in tension.?2015 Acta Materialia Inc.Published by Elsevier Ltd.All rights reserved.Keywords:Magnesium alloys;Reversed yield stress asymmetry;Non-Schmid;PortevinLe Chatelier;Finite element simulation1.IntroductionMg can deform by crystallographic slip an

9、d mechanicaltwinning 1,2.Slip may take place both along the h11?20ior hai direction mainly on basal and prismatic 10?10planes as well as along the h11?23i or hc+ai direction onpyramidal 11?22 planes.Twinning occurs predominantlyon the pyramidal 10?12 planes 3,4 and it plays a key roleduring deformat

10、ion.Indeed,due to the low symmetry ofthe hcp lattice,the number of independent slip systemscapable of accommodating deformation along the c-axisis limited and twinning has to occur in order to avoid inter-granular incompatibilities.As the c/a ratio of Mg is lowerthan the ideal one,pyramidal twinning

11、 is denominated ten-sion or extension twinning,since it can only be activatedwhen the resolved applied stress results in an extension ofthe c-axis 5.Different combinations of deformationsystems may be activated under different deformationmodes for a given texture due to the polar nature of themechan

12、ical twinning.Wrought processes,such as extrusion or rolling,gener-ally give rise to a crystallographic texture where a signifi-cant fraction of grains have their 0001 basal planespreferentially oriented parallel to the extrusion(ED)orthe rolling direction(RD)6.Basal slip,which is usuallythe most ea

13、sily activated deformation mechanism at roomtemperature(RT)2,is severely hindered when loadingalong the ED or the RD due to the low Schmid factor ofthe basal planes in this condition 7.In contrast,tensiontwinning is easily activated under compression,while pris-matic slip is mainly activated under t

14、ension 7.Since thecritical resolved shear stress(CRSS)of non-basal systemsat RT is much higher than that of basal slip and twinning8,9,Mg wrought products are characterised by a distinctRT tensioncompression asymmetry 4,5.The addition of certain rare earth(RE)elements,even indilute concentrations,ha

15、s been recently claimed to be aneffective method for reducing the RT yield asymmetry ofMg wrought products 1015.This is related to the relativeweak texture of RE-containing Mg alloys as compared withpure Mg or conventional Mg alloys.The mechanismresponsible for the texture weakening is not still cle

16、arhttp:/dx.doi.org/10.1016/j.actamat.2015.03.0531359-6462/?2015 Acta Materialia Inc.Published by Elsevier Ltd.All rights reserved.Corresponding author.Available online at ScienceDirectActa Materialia 92(2015) the intense efforts devoted in the past few years toinvestigate MgRE alloys.It has been pro

17、posed that REelements could alter the activity of the different deforma-tion modes leading to weaker textures during hot rollingand/or extrusion.In particular,the activation of non-basalhc+ai slip and/or a greater propensity for contraction anddouble twinning would favour a broader distribution of t

18、hebasal planes 16.This has been sometimes related to theformation of shear bands,which are inclined towardsthe direction of the metal flow and where the basal planesare closely aligned with the band plane 1719.However,the effect of REs on the deformation behaviour is notenough to explain the texture

19、 change in MgRE alloys.Moreover,some investigations suggest that RE additionsalter the nature of static or dynamic recrystallization pro-cesses 2024,resulting in the observed texture randomiza-tion during hot processing.This has been mainly associatedwith preferred nucleation of grains at particles

20、10 andshear bands 21 and with retardation of grain growth byparticles or solutes 20.The influence of RE additions on the mechanical beha-viour of Mg alloys during uniaxial loading at RT is alsonot well understood.Some studies 12,14 suggest that theweak texture alone cannot explain the elimination of

21、 theRT yield asymmetry in RE-containing Mg alloys,whichwould also require a change of the CRSS for the deforma-tion modes with respect to non-RE Mg alloys.It has beenrecently shown that the RE additions lead to a decrease inthe ratio of the CRSS for non-basal slip to the CRSS fortwinning,leading to

22、an enhancement of the activity ofnon-basal slip systems at the expense of twinning 15,which clearly contributes to the reduced mechanical asym-metry of the RE-containing Mg alloys.This effect has beenascribed to changes in the c/a ratio 16,in the Peierlspotentials 19 or in the stacking fault energie

23、s(SFEs)19,25 and even to the anisotropic strengthening effect ofprecipitates 12 and solutes,as well as of the grain refine-ment relative to non-containing RE alloys 26.It is clearthat,whether an effect of texture weakening or an effectof non-basal slip promotion,the reduction of the yieldstress asym

24、metry caused by the RE additions is based ona lower incidence of twinning or,in other words,a higherreliance on slip to accommodate the deformation.Significantly fewer efforts have been devoted to investi-gate the influence of temperature on the mechanical beha-viour of RE-containing alloys 2734.Und

25、erstanding thedeformation mechanisms of these materials at moderatetemperatures is critical,as most deformation processingoperations take place in this temperature range.It has beenobserved that the RE-containing alloys present serrationsin the stressstrain curve 27,29,31,32,indicating the occur-ren

26、ce of the PortevinLe Chatelier effect during plasticdeformation.This has been usually ascribed to dynamicstrain ageing(DSA),originated by the attractive interac-tion between mobile dislocations and diffusing solute atoms35,36.It has been proposed that RE elements enhanceDSA phenomena as a consequenc

27、e of their higher tendencyto segregate to dislocations 29,37.Moreover,RE elementsseem to displace the DSA active range to higher tempera-tures due to their slow diffusivity in Mg 27.However,the fundamental mechanisms of DSA in MgRE alloyshave still not been established and several aspects stillrequi

28、re further investigation,as the influence of the testingmode or of the hydrostatic stresses on serrations 31.A sec-ond aspect that requires clarification is the observation ofan unexpected yield stress asymmetry at high temperatures,where the compressive yield stress(CYS)is higher than thetensile yi

29、eld stress(TYS)28,30.Despite its relevance,thisphenomenon has not been extensively explored and its ori-gin is still not understood.Thus,additional efforts to fullyunderstand the high temperature mechanical behaviour ofRE-containing Mg alloys are very timely.The aim of this work is to investigate th

30、e influence oftemperature on the microstructure,the texture and themechanical behaviour of an extruded MN11(Mg1 wt.%Mn1 wt.%Nd)alloy.With that purpose,uniaxial tensileand compressive tests were carried out along ED atquasi-static and dynamic strain rates in the temperaturerange from?173?C to 300?C.T

31、he microstructure andthe texture of the MN11 alloy before and after deformationwereexaminedbytransmissionelectronmicroscopy(TEM),X-ray diffraction(XRD)and electron backscat-tered diffraction(EBSD).In addition,the CRSS for thevarious deformation mechanisms at different temperaturesand quasi-static st

32、rain rates were estimated by means ofnumerical simulations based on computational homogeni-sation.The activity of the different deformation modes asa function of temperature and loading mode(tension orcompression)is inferred from the microstructure,textureand CRSS data.Finally,the origin of the vari

33、ations ofthe tensioncompression yield asymmetry with temperatureas well as the unexpected asymmetric nature of DSA arediscussed on the light of the above findings and related tothe dominant deformation mechanisms.2.Experimental procedureThe material under study is a Mg MN11 alloy withchemical compos

34、ition(wt.%)1 Mn,1 Nd,Mg(balance).Billets for extrusion produced by gravity casting weremachined up to a diameter of 93 mm.The billets werehomogenised at 350?C during 15 h before extrusion.Indirect extrusion was carried out at 275?C at 8.8 mm/sto produce round bars of 17 mm in diameter,which corre-sp

35、onds to an extrusion ratio of 1:30.Tensile and compressive specimens were machined fromthe as-extruded bar with their loading axis parallel to theED.The tensile specimens had a cylindrical geometry witha gauge section of 3 mm in diameter and 10 mm in length.The compressive specimens were also cylind

36、rical with 3 mmin diameter and 4.5 mm in height.Quasi-static uniaxial ten-sile and compressive tests at relative low strain rates wereperformed until failure in a Servosis universal testingmachine at?173?C,50?C,150?C,250?C and 300?Cand at initial strain rates of 5?10?2s?1and 10?3s?1.Prior to testing

37、,the specimens were kept for?20 min atthetesttemperature,whichwasmeasuredusingathermocouple clamped close to the specimens.The com-pression tests were performed using lubrication in orderto minimise friction between the sample and the anvils.The yield stress(YS)corresponding to each test was calcu-l

38、ated as the true stress at 0.2%engineering strain.For thetests at?175?C,the samples were tested in a liquid nitro-gen bath.Otherwise,an elliptical furnace furnished withfour quartz lamps in air was used and,upon completionof the tests,the specimens were immediately water-cooledto preserve the micros

39、tructure.Additionally,several speci-mens were deformed in compression at 50?C and 250?Cand at an initial strain rate of 10?3s?1up to intermediate266P.Hidalgo-Manrique et al./Acta Materialia 92(2015)265277strains to study the evolution of microstructure and texturewith deformation.Moreover,one tensil

40、e test and one com-pressive test were conducted at 250?C in an Instron univer-sal testing machine using the strain rate change(SRC)method in order to calculate the strain rate sensitivity(m lnr=ln _ e)at selected temperatures.During theSRC tests,the strain rate was consecutively reduced in sev-eral

41、steps from 10?2s?1to 10?5s?1and,then,consecu-tively increased from 10?5s?1to 10?2s?1.From the SRCcurves,thetruestress(r)-truestrainrate(_ e)pairscorresponding to the decreasing strain rate steps wereextracted for the determination of the m values.Anotherset of high strain rate(?103s?1)tensile and co

42、mpressivetests was carried out at about 50?C,150?C,250?C and300?C using a Hopkinson bar equipped with a radiant fur-nace.Self-heating during the dynamic tests was estimatedto reach a maximum value of about 50?C.Microstructural analysis was performed by opticalmicroscopy(OM)in an Olympus BX-51 micros

43、cope andby TEM at an accelerating voltage of 200 kV in a JEOLJEM 2100 microscope.Elemental analysis of the differentphases present was conducted using an energy dispersivespectrometer(EDS)coupled to the TEM.The macrotex-ture was analysed by XRD.The(0001),(10?10),(10?11),(10?12),(10?13)and(11?20)pole

44、 figures were measuredusing Cu Karadiation in a XPert PRO ALPHA1PANalytical diffractometer furnished with a PW3050/60goniometer.From these experimental data,the orientationdistribution function(ODF)and the calculated pole figureswere obtained using the MATLAB toolbox MTEX 38.The inverse pole figures

45、(IPFs)were then derived fromthe calculated direct pole figures.EBSD was also carriedout using a FEI Helios NanoLab 600i field emission gunscanning electron microscope(FEG-SEM)operated at15 kV,with a sample tilt of 70?and a working distanceof 8 mm.For orientation mapping,a step size of 0.13 lmor 0.15

46、 lm was used.The EBSD analyses were completedusing the commercially available HKL Channel 5 software.In preparation for OM,EBSD and XRD,samples wereground and polished using standard metallographic tech-niques.Afterwards,the specimens for OM were chemicallyetched in an acetic picral solution to reve

47、al the grainboundaries,while the specimens for EBSD inspections werechemically polished in a solution based on hydrochloricand nitric acids.For TEM investigations,discs of 3 mmdiameter were thinned to perforation using a Gatan preci-sion ion polishing system.3.Computational homogenisationThe effecti

48、ve behaviour of the MN11 Mg alloy at differ-ent temperatures was determined by means of the crystalplasticity finite element(CPFE)simulation of a representa-tivevolumeelement(RVE)ofthepolycrystallinemicrostructure.The crystal plasticity model of each grainis presented in detail in 39 and is briefly

49、summarised herefor the sake of completion.It accounts for the dominantdeformation modes of Mg alloys:basal hai,prismatic haiand pyramidal hc+ai slip as well as extension twinning.The plastic slip rate for a given slip system follows apower-law,according to_ ci _ c0jsisnjsi?1msignsisn1where _ c0is a

50、reference shear strain rate(equal to the aver-age experimental strain rate of 10?3s?1in this paper),sitheCRSS of the slip system i,m the rate-sensitivity exponentand sisnthe resolved shear stress on the slip system i.Similarly,the twinning rate,_fa,also follows a viscouslaw:_fa_f0hsasnisi?1mwithhsas