Micromechanical characterization of casting-inducedinhomogeneity in an Al0.8CoCrCuFeNi high-entropy alloy

Zhiyuan Liu,a Sheng Guo,a Xiongjun Liu,a Jianchao Ye,a Yong Yang,a Xun-Li Wang,b

Ling Yang,b Ke Anb and C.T. Liua,

aDepartment of Mechanical Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, PR ChinabNeutron Scattering Science Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA

Received 14 December 2010; revised 12 January 2011; accepted 12 January 2011

Available online 15 January 2011

The microstructural features and micromechanical behavior of individual phases in a cast Al0.8CoCrCuFeNi high-entropy alloy(HEA) were characterized by high-resolution scanning electron microscopy and micro-compression tests. Use of neutron diractionenabled the detection of a new phase which was otherwise unobservable by conventional X-ray diraction. The identied phase con-stitution agreed well with the compositional analysis and the micro-compression results. The delicate microscale characterization ofindividual phase provides new insights for the design of novel HEAs with desirable mechanical properties. 2011 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Keywords: High-entropy alloy; Solidication microstructure; Neutron diraction; Compression test

High-entropy alloys (HEAs) provide a novel al-loy concept that signicantly expands the scope of tradi-tional alloy design [13]. HEAs typically consist of atleast ve principal metallic elements in near-equimolarratios, and they often form a single solid solution struc-ture, instead of many intermetallic compounds as ex-pected from general physical metallurgy principles.HEAs show great potential for engineering applicationsdue to their high hardness, wear resistance, high-temper-ature softening resistance and oxidation resistance [2,4].The commonly used alloying elements include face-cen-tered cubic (fcc)-type Cu, Al, Ni, body-centered cubic(bcc)-type Fe, Cr, Mo, V and hexagonal close packed(hcp)-type Ti, Co [510]. Over the years considerable ef-forts have been devoted to the development of new HEAsystems with improved mechanical and functional prop-erties. However, few studies have been carried out toinvestigate the mechanical inhomogeneity of theseHEAs even though inhomogeneity is a common issuein the cast structure for crystalline materials, and theexisting experimental observations have generally re-vealed the dendritic structure and compositional segre-gation in various HEA systems [6,11].

The AlxCoCrCuFeNi (in atomic proportion) system isa widely studied HEA system, and the phase constitutionin the as-cast material can be accurately adjusted by con-trolling the Al addition [6]. When x 6 0.5, only one singlefcc solidsolution phase is observed [6]; bcc solidsolutionphase starts to appear at x = 0.8, and at x > 2.8 a singlebcc phase is obtained although some minor phases (notdetectable via conventional X-ray diraction (XRD)techniques) do exist [6]. The mechanism behind the addi-tion of fcc-typeAl promoting the transformation betweenfcc and bcc phases is still unclear, although presumablythis can be rationalized as the alloying of the larger Alatoms lowers the atomic packing eciency [12]. Previousreports suggest that the lattice parameters for both fccand bcc phases increase with increasing x, but signicantincrease of hardness is seen only at x > 1.0 [6,13]. The for-mation of bcc phase will enhance the strength of the orig-inally purely fcc solid solution, but it also causes theembrittlement issue at ambient temperature [14]. Natu-rally, it is of interest to study the mechanical behaviorof Al0.8CoCrCuFeNi (the composition at which the bccphase just starts to appear) from the structurepropertycorrelation perspective. In addition, a secondary fccphase has been detected in as-cast Al1.0CoCrCuFeNi[11,15] by conventional XRD, but not yet in Al0.8CoC-rCuFeNi. In this work we used neutron diraction to suc-cessfully detect the secondary fcc phase. The existence of

1359-6462/$ - see front matter 2011 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.scriptamat.2011.01.020

Corresponding author. Tel.: +852 2766 6644; fax: +852 2365

4703; e-mail: mmct8tc@inet.polyu.edu.hk

Available online at www.sciencedirect.com

Scripta Materialia 64 (2011) 868871

www.elsevier.com/locate/scriptamat

three phases (two fcc phases and one bcc phase) can bewell correlated with the three distinct regions observedin the microstructure. We also performed careful micro-compression tests to characterize these three regions toobtain their individual mechanical properties. To the bestof our knowledge, this is the rst time a micro-compres-sion test has been used to identify the mechanical proper-ties of the constitutive phase components in HEAs.

The target alloy used in this work has the nominal com-position of Al0.8CoCrCuFeNi (in atomic proportion). Thealloys were prepared by arc-melting a mixture of the con-stituent elements with purity better than 99.9% in a Ti-get-tered high-purity argon atmosphere. Repeated melting wascarried out at least ve times to improve the chemicalhomogeneity of the alloy. The molten alloy was drop-castinto a 10 mm diameter copper mold. The phase constitu-tion of the alloywas examined by bothXRDusing Co radi-ation (Bruker AXS D8 Discover) and neutron diractionusing the VULCAN diractometer (Spallation NeutronSource, Oak Ridge National Laboratory, TN, USA [16]).Because neutrons are highly penetrating, the neutron dif-fraction measurements are representative of the bulk,rather than from the surface. As a state-of-the-art engineer-ing diractometer, VULCAN can be exibly congured ineither high-intensity or high-resolution mode [17,18]. Thepresent measurements were made in the high-resolutionmode, in which a Dd/d 0.2% resolution is maintainedover a wide range of d-spacing. This allowed easy identi-cation of themultiple phases in the sample. Themicrostruc-ture of the alloy was characterized by scanning electronmicroscopy (SEM) and energy dispersive spectrometry(EDS) using a Leo 1530 FEG microscope operated at5 kV. For the microstructure observation, the sample sur-face was sequentially polished down to 0.1 lm grit aluminasuspension nish and then etched with aqua regia solution.To investigate the micromechanical behavior, micropillarswith a top diameter of 0.8 lm and aspect ratios rangingfrom 2:1 to 5:1 were fabricated using a Quanta 200 3Ddual-beam scanning electron microscope/focused ion beam(FIB) system (FEI Company, Hillsboro, OR, USA).

Following the well-established sequential-milling ap-proach [19,20], the micropillars were prepared with greatcare particularly in the interdendritic regions, to ensurethat the as-cut micropillars came exclusively from thetarget regions (see below). The micro-compression testsof the micropillars were carried out using a low-loadTriboindentere Nanoindentation system (HysitronInc., Minneapolis, MN, USA) equipped with a 10 lmat-end diamond punch at a load-controlled mode. Toensure the reproducibility of the micro-compressiondata, at least three micropillars were fabricated andcompressed from each characteristic region. Vickershardness was also measured by applying a load of 1 kgfor 10 s using a Future-Tech microhardness tester.

Figure 1a shows the XRD pattern of the as-castAl0.8CoCrCuFeNi. It is seen that the alloy has a simplesolid solution structure in which one fcc and one bccphase were identied. Judging from the relative inten-sity, the main phase of the alloy is fcc. This is in agree-ment with the previous results reported by Tong et al.[6]. It should be mentioned here that due to the dendriticstructure (see below), the XRD intensities present tex-ture feature and hence some characteristic peaks are

abnormally weak, although the intensity decrease couldalso be caused by the highly distorted atomic planes inthe solid solutions [21]. To ascertain that the bcc phasealso exists in the bulk, the alloy was examined using neu-tron diraction and the pattern is shown in Figure 1b.Interestingly, apart from one fcc and one bcc phaseidentied using conventional XRD, a secondary fccphase was also detected. To dierentiate these two fccphases, they are termed fcc1 and fcc2 hereinafter. Thefcc2 phase is the minor phase judging from its volumefraction. Based on the neutron diraction results, thelattice parameters for the fcc1, fcc2 and bcc phases areestimated to be 3.603, 3.631 and 2.874 A, respectively.It is worth pointing out that the fcc1 and fcc2 phaseswere also identied in the as-cast Al0.5CoCrCuFeNi al-loy using neutron diraction (results not given here),and the lattice parameters for the fcc1 and fcc2 (3.596and 3.628 A, respectively) are almost the same as thosein Al0.8CoCrCuFeNi. In addition, the lattice parametersfor the fcc1, fcc2 and bcc phases in Al0.8CoCrCuFeNimeasured in this work are very close to those of the cor-responding phases in Al1.0CoCrCuFeNi (3.59, 3.62 and2.87 A, respectively) reported by Singh et al. using XRD[11]. It is noted here that our measurement of the latticeparameters together with those from Singh et al. are dif-ferent to those reported by Tong et al. [6] for the samealloy compositions, in which our measured latticeparameters are less sensitive to the Al composition. This

30 40 50 60 70 80 90 100

Inte

nsity

(cps

)

2 (degree)

fccbcc

(a)

0.5 1.0 1.5 2.0 2.5

0

5

10

15

20

25

30

Nor

mal

ized

Inte

nsity

D-spacing (Angstrom)

Experimental Refinement Difference FCC1 FCC2 BCC

(b)

Figure 1. (a) XRD pattern of the as-cast Al0.8CoCrCuFeNi alloy; (b)

neutron diraction pattern of the as-cast Al0.8CoCrCuFeNi alloy.

Figure 2. SEM images of the as-cast Al0.8CoCrCuFeNi alloy: (a) low -

magnication image; (b) magnied image of the indicated zone in (a)

showing the three distinct regions A, B and C. (c and d) Magnied

images for regions B and C, respectively.

Z. Liu et al. / Scripta Materialia 64 (2011) 868871 869

is most likely due to the fact that Tong et al. only iden-tied one fcc phase in their XRD analyses, and the hid-den fcc peaks would cause the inaccurate peak ttingfrom which the lattice parameters were calculated. Thelattice parameters reect the lattice distortion informa-tion and our results can better account for the hardnessvariation as a function of Al addition: from x = 0.5 to0.8, the alloying of Al causes the partial transformationfrom fcc to bcc phase while the fcc lattice remains nearlyinvariant; at x = 0.8, the main phase is still fcc and theamount of bcc phase is insucient to cause signicantstrengthening (the Vickers hardness obtained using1 kg load for Al0.5CoCrCuFeNi and Al0.8CoCrCuFeNiare 258 and 280, respectively); however, when x reaches1.0, the main phase becomes bcc [11], and the largeamount of harder bcc phases causes the hardness to in-crease to 531.

Figure 2 shows themicrostructures of the as-cast alloy.Dendritic and interdendritic structures typical of castHEAs [6] are clearly observed. According to Tong et al.[6], there are only two distinctive regions in the micro-structure and the dendritic regions have a fcc structure,while the interdendritic regions have a mixed fcc + bccstructure. However, our high-resolution SEM observa-tion, Figure 2, apparently shows the existence of threedistinctive regions: a dendritic region (A) and two inter-dendritic regions (B and C). The dendritic region A hasa relatively homogeneous contrast, which is consistentwith that reported by Tong et al. [6], and corresponds tothe single fcc structure. In contrast, regions B and C ap-pear to containmore than one phase. Based on themicro-structural features observed in Figure 2, the formingprocess of these three distinctive regions can be envisagedas follows: when the liquid cools down, the primary phaseforms and this is region A. The remaining liquid then sep-arates into two compositionally dierent liquids, Cu-richliquid andCu-depleted liquid, and these two liquid phasesthen solidify separately. The Cu-depleted liquid decom-poses into the mixed bcc and fcc phases (region B) prob-ably following a eutectic reaction as suggested by Tonget al. [6], while the Cu-rich liquid solidies and then fur-ther decomposes into two fcc phases via the spinodaldecomposition as indicated by the modulated structureseen in Figure 2d. The Cu-depleted liquid solidies at

higher temperature compared to the Cu-rich liquid, andthis can account for the coarser microstructure in regionB than in region C. Although apparently more detailedwork needs to be carried out to conrm the hypothesisof the above solidication process, it is a reasonable onewhich can explain the as-obtained microstructure well.Since it is generally perceived that the bcc phase is a hard-er phase compared with the fcc phase, region B wouldhave a higher hardness (or strength) than that of regionA. This is in fact supported by the micro-compressiontests discussed below. From the neutron diraction andcomposition analyses we already know that the fcc1 isthe main phase, and we can further infer that region Ahas the fcc1 structure, region B has the fcc1 + bcc struc-ture and region C has the fcc1 + fcc2 structure.

The average chemical compositions for the threeregions from multiple-point EDS analysis are listed inTable 1. It can be seen that these three regions haveclearly dierent chemical compositions: region A is en-riched in Co, Cr and Fe but decient in Al and Ni andCu; region B is enriched in Al and Ni but decient inthe other elements; region C is signicantly enriched inCu but decient in Co, Cr and Fe. Ni tends to accom-pany Al in the interdendritic regions due to the large neg-ative mixing enthalpy (DHmix) between these elements[22]. This possibly results in the formation of NiAl-type(B2) phase in region B [6,11]. Cu has a tendency to segre-gate from other alloying elements because it has a posi-tive DHmix with other elements (apart from Al) [22].

In addition to identifying the phase composition forthe three distinctive regions, we have gone further andcharacterized their individual mechanical propertiesusing the well-established micro-compression technique[23,24]. Micropillars were machined from the three re-gions using the FIB technique and then compressedusing a at-end diamond punch under the load-con-trolled condition. Typical micropillars before and afterthe micro-compression tests are shown in Figure 3, andrepresentative nominal stressstrain curves are plottedin Figure 4. Here the stress is taken as the applied loaddivided by the initial top area of the pillar, and the strainis taken as the indenter displacement divided by theinitial height of the pillar. As shown in Figure 4, all spec-imens exhibit a linear response upon loading before the

Table 1. Chemical composition and yield strength for the three distinctive regions in the as-cast Al0.8CoCrCuFeNi alloy.

Element (at.%) Structure Al Co Cr Cu Fe Ni Yield strength (MPa)

Nominal 13.79 17.24 17.24 17.24 17.24 17.24

Region A fcc1 10.39 21.26 22.21 8.56 22.12 15.47 764 47

Region B fcc1 + bcc 22.82 13.97 13.54 15.76 12.52 21.39 958 80

Region C fcc1 + fcc2 14.20 6.62 5.16 53.83 6.11 14.08 825 60

Figure 3. SEM images of the micropillars before (a, c and e) and after

(b, d and f) the micro-compression tests. (a and b), (c and d) and (e and

f) correspond to regions A, B and C, respectively.

0

200

400

600

800

1000

1200

1400 Region A Region B Region C

Stre

ss (M

Pa)

Strain

0.05

Figure 4. Stressstrain curves for the micropillars fabricated in the

three dierent regions as shown in Figure 2.

870 Z. Liu et al. / Scripta Materialia 64 (2011) 868871

yielding point, i.e. the rst observable strain burst ap-pears [25]. The extracted yield strengths for the three re-gions are listed in Table 1. The micropillars cut from theinterdendritic region B have the highest yield strength of958 MPa, which is about 20% higher than that of thedendritic region A (764 MPa). The much higherstrength of region B most probably derives from the pre-cipitation of the bcc phase in this region. Region C has aslightly higher yield strength (825 MPa) than that of re-gion A, in agreement with our previous assumption thatthis region has a mixed two-fcc-phase structure resultingfrom the spinodal decomposition. It is noted here thatthe single-phase region A has a high yield strength of764 MPa, which is much higher than the strength (i.e.ultimate tensile strength) of any constituent metal ele-ment (Al, 47; Co, 255; Cr, 483; Cu, 220; Fe, 289; Ni,407 MPa [26]). In addition, the yield strength presentsan apparent size eect in that the yield strengths mea-sured from the micropillars are much higher than thatfrom the macrocompression test (r0.2 470 MPa), andthe larger the size of the micropillars, the closer the yieldstrength measured from the micropillars and the macros-amples. The mechanism behind the size eect in HEAsshould be dierent to the dislocation starvation responsi-ble for the size eect in single crystals [19,27]. More de-tailed results and discussions on the size eect in HEAswill be reported elsewhere.

After yielding, the pillars deform plastically via theemission of sporadic strain burst under the load-con-trolled condition, separated by elastic-like deformationsegments. Similar phenomena have also been observedfrom the micro-compression of a variety of single-crystalmetals [28,29]. From the nominal stressstrain curves, itis noticed that smaller strain bursts tend to occur for re-gions B and C compared to those in region A. This couldbe attributed to the blocking and storage of dislocationsby the phase boundaries in the two-phase regions, B andC, and the interaction between the high density of dislo-cations and the phase boundaries can account for themuch less jumpy deformation behavior [30].

In summary, microstructural features and microme-chanical behavior for each individual phase in a castAl0.8CoCrCuFeNi alloy have been determined usinghigh-resolution SEM, together with delicate FIB-assistedmicro-compression tests. Both compositional and micro-mechanical analyses revealed the existence of three dis-tinct regions, in agreement with the phases identied byneutron diraction. The bcc-phase-containing regionclearly has a higher yield strength than the other two re-gions. The understanding of the mechanical behavior ofindividual phase provides important input for the designof new HEAs with desirable mechanical properties.

This research was supported by the internalfunding from HKPU. The Spallation Neutron Sourceis operated with support from the Division of ScienticUser Facilities, Oce of Basic Energy Sciences, USDepartment of Energy, under contract DE-AC05-00OR22725 with UT-Battelle, LLC.

[1] J.W. Yeh, Ann. Chim.-Sci. Mat. 31 (2006) 633.[2] W.H. Wu, C.C. Yang, J.W. Yeh, Ann. Chim.-Sci. Mat. 31

(2006) 737.

[3] J.W. Yeh, S.K. Chen, S.J. Lin, J.Y. Gan, T.S. Chin, T.T.Shun, C.H. Tsau, S.Y. Chang, Adv. Eng. Mater. 6 (2004)299.

[4] X.F. Wang, Y. Zhang, Y. Qiao, G.L. Chen, Intermetallics15 (2007) 357.

[5] H.Y. Chen, C.W. Tsai, C.C. Tung, J.W. Yeh, T.T.Shun, C.C. Yang, S.K. Chen, Ann. Chim.-Sci. Mat. 31(2006) 685.

[6] C.J. Tong, Y.L. Chen, S.K. Chen, J.W. Yeh, T.T. Shun,C.H. Tsau, S.J. Lin, S.Y. Chang, Metall. Mater. Trans.36A (2005) 881.

[7] J.W. Yeh, S.K. Chen, J.Y. Gan, S.J. Lin, T.S. Chin, T.T.Shun, C.H. Tsau, S.Y. Chang, Metall. Mater. Trans. 35A(2004) 2533.

[8] C.Y. Hsu, W.R. Wang, W.Y. Tang, S.K. Chen, J.W.Yeh, Adv. Eng. Mater. 12 (2010) 44.

[9] G.Y. Ke, S.K. Chen, T. Hsu, J.W. Yeh, Ann. Chim.-Sci.Mat. 31 (2006) 669.

[10] Y. Zhang, G.L. Chen, L. Gan, J. ASTM Int. 7 (2010)JAI102527.

[11] S. Singh, N. Wanderka, B.S. Murty, U. Glatzel, J.Banhart, Acta Mater. 59 (2011) 182.

[12] F.J. Wang, Y. Zhang, G.L. Chen, J. Alloy Compd. 478(2009) 321.

[13] C.J. Tong, M.R. Chen, S.K. Chen, J.W. Yeh, T.T. Shun,S.J. Lin, S.Y. Chang, Metall. Mater. Trans. 36A (2005)1263.

[14] C.W. Tsai, M.H. Tsai, J.W. Yeh, C.C. Yang, J. AlloyCompd. 490 (2010) 160.

[15] U.S. Hsu, U.D. Hung, J.W. Yeh, S.K. Chen, Y.S. Huang,C.C. Yang, Mater. Sci. Eng. A 460 (2007) 403.

[16] T.E. Mason, D. Abernathy, I. Anderson, J. Ankner, T.Egami, G. Ehlers, A. Ekkebus, G. Granroth, M. Hagen,K. Herwig, J. Hodges, C. Homann, C. Horak, L.Horton, F. Klose, J. Larese, A. Mesecar, D. Myles, J.Neuefeind, M. Ohl, C. Tulk, X.L. Wang, J. Zhao, PhysicaB 385386 (2006) 955.

[17] X.L. Wang, T.M. Holden, G.Q. Rennich, A.D. Stoica,P.K. Liaw, H. Choo, C.R. Hubbard, Physica B 385(2006) 673.

[18] X.L. Wang, T.M. Holden, A.D. Stoica, K. An, H.D.Skorpenske, A.B. Jones, G.Q. Rennich, E.B. Iverson,Mater. Sci. Forum 652 (2010) 105.

[19] J.R. Greer, W.C. Oliver, W.D. Nix, Acta Mater. 53(2005) 1821.

[20] Y. Yang, J.C. Ye, J. Lu, F.X. Liu, P.K. Liaw, ActaMater. 57 (2009) 1613.

[21] J.W. Yeh, S.Y. Chang, Y.D. Hong, S.K. Chen, S.J. Lin,Mater. Chem. Phys. 103 (2007) 41.

[22] A. Takeuchi, A. Inoue, Mater. Trans. 46 (2005)2817.

[23] M.D. Uchic, D.A. Dimiduk, Mater. Sci. Eng. A 400(2005) 268.

[24] J.C. Ye, J. Lu, Y. Yang, P.K. Liaw, Intermetallics 18(2010) 385.

[25] H. Bei, S. Shim, E.P. George, M.K. Miller, E.G. Herbert,G.M. Pharr, Scripta Mater. 57 (2007) 397.

[26] D. Henkel, A.W. Pense, Structure and Properties ofEngineering Materials, fth ed., McGraw-Hill, New York,2001.

[27] W.D. Nix, J.R. Greer, G. Feng, E.T. Lilleodden, ThinSolid Films 515 (2007) 3152.

[28] M.D. Uchic, P.A. Shade, D.M. Dimiduk, Annu. Rev.Mater. Res. 39 (2009) 361.

[29] E.M. Nadgorny, D.M. Dimiduk, M.D. Uchic, J. Mater.Res. 23 (2008) 2829.

[30] S.I. Rao, D.M. Dimiduk, T.A. Parthasarathy, M.D.Uchic, M. Tang, C. Woodward, Acta Mater. 56 (2008)3245.

Z. Liu et al. / Scripta Materialia 64 (2011) 868871 871

Micromechanical characterization of casting-induced inhomogeneity in an Al0.8CoCrCuFeNi high-entropy alloyack2References