Friction-stir Dissimilar Welding of Aluminium Alloy to High Strength Steels--Mechanical Properties and Their Relation to Microstructure

Materials Science & Engineering A 556 (2012) 175183

Contents lists available at SciVerse ScienceDirect

Materials Science & Engineering A

0921-50

http://d

n Corr

Energie

E-m

journal homepage: www.elsevier.com/locate/msea

Friction-stir dissimilar welding of aluminium alloy to high strength steels:Mechanical properties and their relation to microstructure

R.S. Coelho a,b,n, A. Kostka c, J.F. dos Santos d, A. Kaysser-Pyzalla a,b

a Helmholtz-Zentrum Berlin fur Materialien und Energie GmbH, 14109 Berlin, Germanyb Former at Max-Planck-Institut fur Eisenforschung GmbH, 40237 Dusseldorf, Germanyc Max-Planck-Institut fur Eisenforschung GmbH, 40237 Dusseldorf, Germanyd Helmholtz-Zentrum Geesthacht GmbH, Zentrum fur Materialforschung und Kustenforschung, Institute of Materials Research, Materials Mechanics, Solid-State Joining Processes,

21502 Geesthacht, Germany

a r t i c l e i n f o

Article history:

Received 2 March 2012

Received in revised form

19 June 2012

Accepted 20 June 2012Available online 6 July 2012

Keywords:

Friction stir welding

Dissimilar joint

Aluminium alloy

High strength steel

Microstructure

EBSD

Mechanical properties

93/$ - see front matter & 2012 Elsevier B.V. A

x.doi.org/10.1016/j.msea.2012.06.076

esponding author at: Helmholtz-Zentrum

GmbH, 14109 Berlin, Germany. Fax: 49 30ail address: rodrigo.coelho@helmholtz-berlin.

a b s t r a c t

The use of light-weight materials for industrial applications is a driving force for the development of

joining techniques. Friction stir welding (FSW) inspired joints of dissimilar materials because it does

not involve bulk melting of the basic components. Here, two different grades of high strength steel

(HSS), with different microstructures and strengths, were joined to AA6181-T4 Al alloy by FSW. The

purpose of this study is to clarify the influence of the distinct HSS base material on the joint efficiency.

The joints were produced using the same welding parameter/setup and characterised regarding

microstructure and mechanical properties. Both joints could be produced without any defects.

Microstructure investigations reveal similar microstructure developments in both joints, although

there are differences e.g. in the size and amount of detached steel particles in the aluminium alloy (heat

and thermomechanical affected zone). The weld strengths are similar, showing that the joint efficiency

depends foremost on the mechanical properties of the heat and the thermomechanical affected zone of

the aluminium alloy.

& 2012 Elsevier B.V. All rights reserved.

1. Introduction

Energy and environmental issues in transportation systemshave a strong impact on material selection and on thedevelopment of joining techniques [17]. The incorporation oflight-weight materials in many structures (e.g. automotive, ship-building and aerospace) allow a reduction of weight and conse-quently fuel consumption. In this regard, dissimilar jointsbetween light-weight materials such as aluminium alloys (Alalloy) and steels are receiving increased interest both in scienceand industrial application [820]. However, the substitution ofone material for another is not straightforward and highlyefficient products require appropriate joining processes.

Dissimilar fusion welding between Al alloy and steels is achallenge in welding control because of the large differences inmelting temperature and physical and mechanical properties ofthe alloys involved. The process often results in complex weldpool shapes, inhomogeneous solidification microstructures andsegregations in addition, the extremely low solubility of Fe in Al

ll rights reserved.

Berlin fur Materialien und

8062 15752.

de (R.S. Coelho).

leads to the formation of brittle and excessive Al-rich FexAlyintermetallic phases [21,22] which are detrimental for themechanical properties of the joint [812,20,22].

Friction stir welding (FSW) is based on extreme plasticdeformation in the solid-state where no associated bulk meltingis involved [2326]. At early stages of the process development,FSW appears especially attractive for joining Al alloys and otherlight-weight materials like Mg alloys [2431]. This is connectedwith two main reasons:

(1)

(2)

For an application of the FSW process on steels and titanium,an optimisation of tool material and geometry is needed; con-tinuing attempts can be found in the literature [3236]. However,FSW application for these materials is still limited by the cost ofeffective available tools [37].

In the case of dissimilar joints, FSW appears as a suitable andpromising process to minimise problems related to materials

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183176

incompatible with respect to melting temperatures and theformation of brittle intermetallic phases [27,38]. Recent studieshave shown that welds between dissimilar materials such as Mgalloys to Al alloys [3944], Mg alloys to steels [4547], Al alloys toTitanium [48] and Al alloys to steels [1320] can be produced byFSW. The high quality of dissimilar welds produced between Al orMg alloys to steels or titanium in a butt-joint configuration isassociated to the smart idea of positioning the tool pin centreshifted towards the Al or Mg alloys (fixed on the retreating side)[1420,48] or fixing Al or Mg alloys on the top side in the case oflap joints [13,4547]. Thus, the tool barely makes contact withthe steel and minimum tool wear occurs, improving the costefficiency of the process.

In the present study, two grades of high strength steel (HSS)with significantly different microstructure, and strengths wereselected to be joined to AA6181-T4 Al alloy by FSW. In order toaccess the influence of the distinct HSS base material on jointefficiency, the joints were produced by applying the same weld-ing parameters and by shifting the tool pin centre towards the Alalloy. Early investigations were conducted in one of these jointsand presented elsewhere [14]. Those studies [14] were focussedon the analysis of the complex material flow based on micro-structural observations applying SEM-EBSD technique. Here, wediscuss the influence of distinct HSS base material on the jointefficiency and microstructure formation. We show that indepen-dent of the HSS chosen the joint efficiency is determined by theheat-affected zone of the Al alloy, which controls the mechanicalproperties of the joints.

2. Experiments

2.1. Materials

Commercially available materials that are suitable for auto-motive structures and reinforcement parts were selected for thisstudy. DP600 and HC260LA HSS plates were chosen to be joinedto AA6181-T4 Al alloy by FSW. The chemical composition and themechanical properties of the steels and the Al alloy are presentedin Tables 1 and 2, respectively. It can be seen that the Al alloytensile strength is substantially less than those of the HSSs.

The microstructure of the base materials (BMs) is presented inFig. 1. The individual microstructural characteristics can besummarised as following:

Table 2Obtained yield, strength and strain values for each material selected for this study.

TablChem

Ma*A

Ma**H***D

n

n

n

The values are an average of at least three tensile tests conducted for each

material.

Materials Yield strength (MPa) Tensile strength (MPa) Elongation (%)

the HC260LA HSS shows a-Fe (ferrite) grains with a size ofabout 40 mm and presence of pearlite on the grainboundaries;

AA6181-T4 12775 25975 2671

the dual phase steel DP600 HSS shows a-Fe grains with a sizeof about 15 mm (appear dark in Fig. 1c) and the presence ofmartensite as a second phase (appears bright in Fig. 1c).

e 1ical composition of the materials selected for this study (wt.%).

terials Si Fe Cu Mn MA6181-T4 0.85 0.25 0.06 0.09 0

terials Cmax Mnmax Simax Pmax SC260LA 0.10 0.60 0.50 0.025 0

P600 0.10 1.40 0.15 0.07 0

[49]n norm DIN EN 10268, Edition 10.06.nn norm SEW 097 Part 1.

2.2. Welding procedure

The joints were produced in a butt-joint configuration at theHelmholtz-Zentrum Geesthacht, Germany, using a gantry systemequipped with a mechanical clamping table [50]. The weldingsetup is schematically illustrated in Fig. 2(a) and the processparameters follow in Table 3.

A FSW tool, the TungstenRhenium WRe25, consisting of a13 mm diameter concave shoulder and a 5 mm cylindrical non-threaded pin, was selected for this welding configuration. The pinedge was offset about 1 mm from the weld centre line towardsthe AA6181-T4 Al alloy (Fig. 2b). Minimum wear is expected inthis configuration once the pin is plunged into the softer AA6181-T4 Al alloy and does not come in contact with the HSS.

2.3. Microstructure assessment

The microstructure was investigated using optical microscopy(OM), scanning electron microscopy (SEM) electron backscatterdiffraction (EBSD) technique and transmission electron micro-scopy (TEM). SEM-EBSD characterisations were conducted using aZeiss Neon 40 field emission gun SEM equipped with the HikariEDAX/TSL EBSD system and a Jeol JSM-6490 Tungsten filamentSEM equipped with the Pegasus EDAX/TSL EBSD system. Theanalyses were carried out on the sample cross section and on thetop weld surface. The samples were extracted by spark erosion afew centimetres behind the start point of the weld, avoiding anyproblems related with the process that might not be completelystable at its start. The specimen preparation for OM and SEMconsisted of standard metallographic procedure followed by ashort etching in a 5% Nital solution. Consequently, the grainboundaries in the HSS microstructure were highlighted. Theassessment of the Al alloy microstructure was conducted mainlyby EBSD. The specimens for EBSD were prepared in the standardmetallographic way with careful final polishing and withoutchemical etching. Such sample preparation allows the assessmentof both the HSS and the Al alloy microstructure [7,14].

TEM work was performed using a Jeol JEM-2200FS operated at200 kV. The specimens were prepared from regions of interestalong the Al alloyHSS interface using a Jeol JEM-9320 focussedion beam system (FIB) operated at 30 kV.

g Cr Ni Zn Ti Al.74 0.01 0.002 0.012 0.023 Bal.

max Almin Timax Nmax.025 0.015 0.15

.008 0.02 0.009

Fig. 1. Micrographs of the alloy microstructures for each base materials involved in the presented work: (a) AA6181-T4 Al alloy, (b) HC260LA HSS and (c) DP600 HSS.

Fig. 2. Schematic illustration of joint setup studied here (a) Al alloy and steeljoined by FSW in a butt-joint configuration and (b) the relation of real micro-

structure and tool offset position applied to produce the welds analysed in this

study [14].

Table 3Welding parameters.

Main welding parameters

Travel speed (mm/s) 8.0

Rotation speed (rpm) 1600

Down-force (kN) 5.0

Main tool propertiesPin diameter (mm) 5.0

Shoulder diameter (mm) 13

Pin length (mm) 1.35

Pin offset Dy (mm) 1.0Tool material WRe25

Pin feature No-threaded

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183 177

2.4. Mechanical tests

Universal microhardness tests and tensile tests were con-ducted to assess the mechanical properties of the joints. Regard-ing the tensile tests, standard flat specimens with gauge sectionsof 1.5 mm10 mm45 mm were extracted by spark erosioncutting from the base materials and the weld in the transversaldirection of the welding direction. At least three samples fromeach base material and from the joints were tested. All weldedspecimens were machined on their surface in order to remove themarks and the crown features left by the tool shoulder and toavoid any influences of them on the results. Afterwards, failuremechanisms of the flat tensile test specimens were studied in theSEM by a fracture surface analysis.

Microhardness was measured on the specimen cross sectionaccording to DIN 50359-1:1997, Universal hardness (HU) stan-dard, applying a 0.02 N load. The measurements were performedin three different cross section depths in order to check forthrough-thickness variance. The measurements crossed allregions of interest, from the Al alloy side through to the HSSside. One specimen for each joint setup was tested.

3. Results

3.1. Microstructure of the joints

On the macroscopic scale, the as-welded joints revealed a goodweld surface quality containing neither macro voids/cracks norimperfections regarding the weld alignment (Fig. 2b).

Through the weld cross section, both analysed joints revealedthe same microstructure features showing no evidence of mixingbetween the Al alloy and the HSS. In both cases, a small amount of

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183178

detached particles of HSS transported into the Al alloy was observed(Fig. 3). However, comparing both joints, it is evident that the softerHC260LA HSS shows a more strongly deformed interface (dash linein Fig. 3b and d) and slightly larger HSS detached particles than theharder DP600 HSS (arrows in Fig. 3a and c).

A systematic investigation of the welding setup and materialflow of the joint analysed in this paper was previously conducted onDP600 HSS [14]. Some insight into the material flow and micro-structure formation can be visualised and summarised by the energydispersive X-ray (EDX) analysis shown in Fig. 4. These results wereobtained for the joint using the HC260LA HSS and confirm that thereis no evidence of mixing between both materials and also highlightthe presence of the vortex-like structures on the advancing side ofthe tool (Zn mapEDX analysis, Fig. 4d). The vortex featuresbecomes smooth approaching the bottom part of the cross sectionand implies the movement of the HSS detached particles.

Fig. 3. Cross section overview of both investigated joints: (a) Al alloy to HC260LA HSS a(d) the thermo-mechanically deformed interface.

Fig. 4. EDX analysis of element distributions: (a) analysed region, (b) Al distribution instructure on the advancing side in red. (For interpretation of the references to colour i

A close view of both joint interfaces reveals also very similarfeatures (Fig. 5). The high shear strain and friction heating duringthe process supports the intermetallic phase formation and theinterlocking between both materials. In both cases, crack-freebonding between both materials occurs. The microstructure atthe interface is complex and non-smooth, characterised by fineequiaxed a-Fe (ferrite) grains and a very thin layer of interme-tallic FexAly compounds formed by chemical reactions and diffu-sion between the Al and Fe.

Complementary TEM investigations (Fig. 6) revealed in detailsthe intermetallic compound present in both analysed joints. Theintermetallic compounds form very thin stripes (around 50 nmthick) embedded approximately 500 nm deep into the steelmatrix. Fig. 6(c), which reveals the Fe2Al5-phase, summarisesseveral attempts of unambiguous phase identification via electrondiffraction from selected areas of the interfaces of both joints. The

nd (b) the thermo-mechanically deformed interface; (c) Al alloy to DP600 HSS and

blue, (c) Fe distribution in pink and (d) Zn distribution highlighting the vortex-like

n this figure legend, the reader is referred to the web version of this article.)

Fig. 5. Secondary electron SEM micrographs of joint interface highlighting similar features: intermetallic formation and non-smooth interface resulting in mechanicalinterlocking between both materials; (a) Al alloy to HC260LA HSS and (b) Al alloy to DP600 HSS.

Fig. 6. TEM investigation of the Al alloyHSS interface: (a) Al alloy to DP600 HSS, (b) Al alloy to HC260LA HSS and (c) selected area electron diffraction pattern from theFe2Al5 reaction product on the interface between Al alloy and DP600 HSS.

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183 179

low volume fraction of the intermetallic phase, the presence ofchemical gradients (up to 20 at% on 300 nm) and the crystal-lographic lattice strain made the phase identification verychallenging.

3.2. Mechanical properties

3.2.1. Microhardness assessments

Microhardness profiles were measured on the weld cross sectionat three different depths: top, middle and bottom (Fig. 7). For bothinvestigated joints, the results do not show any variance crossingthe Al alloy welding regions: base material (BM), heat-affectedzone (HAZ), thermo-mechanically-affected zone (TMAZ) and stirzone (SZ). On the other hand, the HSS side shows a hardness

change crossing the thermomechanically affected regions. The jointproduced with the softer HSS (Fig. 7a) shows a thermomechanicalarea with length of about 4 mm, while in the other joint it is lessthan 1 mm (Fig. 7b). The highest values in both cases were observedat the weld interface.

3.2.2. Tensile properties (stressstrain curves)

Fig. 8 shows the typical tensile stressstrain curves of allinvestigated specimens. The curves can be considered as anaverage property since at least three specimens were preparedfor each BM and joint, and all of them revealed the same trend.The black and blue curves plotted in Fig. 8 are for Al alloy and HSS,respectively. Comparing both HSS curves, differences in yield

Fig. 7. Universal microhardness measurements horizontally along the Al alloy and the HSS side: (a) Al alloy to HC260LA HSS and (b) Al alloy to DP600 HSS.

Fig. 8. Typical stressstrain curves obtained by tensile test for each BMs (blackand blue curves) and joints (red curves) analysed in this study: (a) Al alloy to

HC260LA HSS and (b) Al alloy to DP600 HSS. (For interpretation of the references

to colour in this figure legend, the reader is referred to the web version of this

article.)

Table 4Obtained yield, strength and strain values for each joint investigated in this study.

Joint Yield strength(MPa)

Tensile strength(MPa)

Elongation(%)

*Al AlloyHC260LA 11272 20078 871*Al alloyDP600 11972 21172 771

n One specimen slide off during the test, therefore only 2 were considered.

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183180

strength and ultimate tensile strength are evident, even thoughboth of them show high elongation to fracture (see Table 2).

Typical tensile stressstrain curves of both Al alloyHSS jointsare ploted in red on the diagrams. The results show that the jointbehaviour follows the tendency of the Al alloy stressstrain curve.The main findings regarding the tensile strengths of the joints aresummarised in Table 4. It can be concluded that both joints revealalmost the same yield strengths as well as tensile strengths.Additionally, regarding the elongation to fracture, both joints alsoreveal similarities showing almost 30% of the Al alloy BMproperty (see Table 2). Please note the different scales for stress

in Fig. 8(a) and (b) which have been chosen for improved clarityin the presentation of results.

3.2.3. Fractography analyses

After the tensile test the fractured samples were submitted tofractography examinations in order to assess the origin of thespecimen failure and also to determine which fracture mechan-isms occurred. Both analysed samples show the same fracturefeatures, hence, only the results obtained for the harder DP600HSS joint are presented.

Fig. 9(a) shows a top view micrograph of the fractured samplehighlighting the expected schematic welding tool position (tooloffset position). The analysis reveals that the failure occurs at theretreating side of the Al alloy crossing the BM-HAZ-TMAZ zonesfar away from the joint interface and from the weld centre point.Analysis of the fractured surface through the cross section revealsa ductile fracture mode characterised by the presence of equiaxeddimples (Fig. 9b and c).

Additional investigations of the top view reveal that few crackswere found close to the joint interface (Fig. 9d). The analysissuggests that those cracks which were found propagated follow-ing the HSS particles [13]. Chemical composition mapping con-ducted through EDX-analysis confirms that the cracks areconnected to the HSS particles (Fig. 9eg).

Additional information regarding the joint failure region (Al alloyBM-HAZ-TMAZ) were obtained by EBSD analysis. Those mostrelevant regions were analysed in the as-welded samples crosssection and the results reveal that a large difference in grainmorphology and distribution can be seen crossing these regions(arrow in Fig. 10a). While the Al alloy BM shows elongated grainswith a size of 70 mm (Fig. 1a), the Al alloy SZ shows characteristics offine equiaxed grains 5 mm diameter in size. Further, top surfaceanalysis conducted on the interface between welded zone and the Alalloy BM reveals the strong grain deformation (arrows in Fig. 10b)caused by the movement of the FSW tool shoulder [14]. Theanalysis, conducted on the top weld surface after standard metallo-graphic sample preparation (approx. 200 mm layer removal),reveals that the surface shear layer caused by the shoulder

Fig. 9. SEM micrographs showing the fractured tensile test specimen: (a) the schematic illustration of the relationship sample and tool position; (bc) the microscopicanalysis of the fractured surface and (dg) the EDX element distribution analysis highlighting the cracking presence close to the joint interface (dg).

Fig. 10. Inverse pole figures maps from the critical interface Al alloy BM-HAZ-TMAZ-SZ: (a) cross section and (b) top analysis [14]. The arrows highlight the

gradient in grains characteristics.

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183 181

movement extends several hundred micrometres (300 mm) fromthe top weld surface.

4. Discussion

In this study, dissimilar joints between Al alloy and two grades ofHSS were investigated in terms of microstructure and mechanical

properties. The purpose of the study is to clarify the influence ofdistinct HSS BMs on the joint efficiency. The results have shown thatboth investigated samples show the same microstructure andproperties since a very similar microstructure evolution occurredduring the welding process. The welding parameter chosen (tooloffset position) meant that severe plastic deformation occurredmainly in the Al alloy and the pin barely moved into the HSS, whichis consistent with the microstructure characteristics presented here.The microstructure formation in FSW is directly connected to thematerial flow, and in the presented study, both joints revealed thesame material flow feature (Fig. 4) [14].

In terms of microstructure, it is evident that joints producedwith the softer steel (HC260LA) reveal a slightly larger deformedinterface region in the Al-alloy and a larger amount/size of theHSS detached particles (Fig. 3). Such differences were alsoobserved in the hardness profiles presented in Fig. 7(a) wherethe HSS side shows a strong hardness gradient which is connectedwith different levels of thermo-deformation. The highest hardnessvalues in both cases were observed in the weld interface wherethe smallest HSS grains were formed together with the thin stripsof intermetallic compounds.

Regardless of the HSS used, the microstructure features of thejoints were characterised by fine equiaxed a-Fe (ferrite) grainswith a small amount of thin intermetallic Fe2Al5 compounds andrough interface Al alloyHSS. The very small quantity, complexmorphology (thin stripes of intermetallic reaction products wereincorporated with steel BM) and strong chemical gradients madeunambiguous phase identification of the intermetallic compoundsvery difficult. Only in the joint with the DP600 HSS it was possibleto report presence of the Fe2Al5-phase. In terms of the binary AlFe equilibrium diagram, a couple of intermetallic phase com-pounds, FeAl3 (y-phase), Fe2Al5 (Z-phase) and FeAl2 (x-phase),might be expected to be formed [21]. Springer et al. [20]investigated the formation of intermetallic reaction layers whichform at the interface of FSW joints between low C (0.12 wt% C)steel and both pure Al (99.5%) and Al-5 wt.% Si. They claimed thatin the as-welded state no intermetallic reaction product could beobserved at the interface, and only after annealing the interme-tallic compounds make detailed investigations possible.

Analysis of the stressstrain tensile curves revealed no influ-ence of the joint interface on the joint efficiency. The results

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183182

revealed that the joint curves (red in Fig. 8) follow the tendency ofthe Al alloy curve showing almost the same yield and strength. Byevaluating the results presented in Tables 2 and 4, it becomesclear that the measured ultimate tensile strength for the jointsdoes not reach the yield strength of the HSS. This fact suggeststhat all deformation during the tensile test took place solely in theAl alloy. Additionally, no evidence of plastic deformation wasobserved in micrographs of the tensile specimens, which suggestsonce again that the Al alloy is responsible for the overall efficiencyof the joint.

Fractography analysis showed that failure always occurs faraway from the joint interface on the retreating side crossing theinterface of the Al alloy BM-HAZ-TMAZ-SZ (Fig. 9). Those regions(arrow in Fig. 10a) reveal the presence of highly deformed grainsand evidence of sub-grain development. No significant differencesin hardness were observed crossing those regions (Fig. 7). Thestrong differences in a grain size distribution and grain shapeshown in Fig. 10 originate from each region of the joint (BM-HAZ-TMAZ-SZ) being exposed to different thermo-mechanical cycles.The gradual accumulation of strain accompanied by the frictionalheating leads to the nucleation and growth of new grains during theongoing deformation process (non-homogeneous dynamic recrys-tallisation). While accumulated plastic deformation increases themechanical strength of the TMAZ and SZ, the grains located in theHAZ reveal relatively small defect densities (concentration ofdislocations and sub-grains) and control the mechanical behaviourof the joint as the weakest elements of the structure.

Additionally, the fracture analysis suggests that the crackpropagation follows the HSS inclusions. These fine-grained HSSinclusions are surrounded by the intermetallic compound and actas stress concentrators upon load. The control of their amount iscrucial for joint efficiency. Hence, the welding setup and para-meter chosen in this study were crucial for the good acquiredmechanical properties by controlling the tool pin position andconsequently the amount of HSS detached particles and theinterface formation (Figs. 3 and 5) [13].

5. Conclusions

In the present study, friction-stir dissimilar joints between twogrades of HSS (DP600 and HC260LA) and AA6181-T4 Al alloy wereproduced applying the same welding parameters and setup (tooloffset position). The studies focussed on the influence of distinctHSS BMs on the joint efficiency. The conducted analysis can besummarised as follows:

(a)

(b)

(c)

(d)

(e)

Acknowledgements

The authors gratefully acknowledge the Helmholtz Associationof German Research Centres for financial support via the virtualinstitute Improving Performance and Productivity of IntegralStructures through Fundamental Understanding of MetallurgicalReactions in Metallic Joints (VI-IPSUS). The authors also wouldlike to thank Mr. Martin Preilowski for help in sample preparationduring his undergraduate studies at MPIE.

References

[1] T.A. Barnes, I.R. Pashby, J. Mater. Process. Technol. 99 (2000) 6271.[2] T.A. Barnes, I.R. Pashby, J. Mater. Process. Technol. 99 (2000) 7279.[3] A. Kochan, Assembly Automation 22 (1) (2002) 2935.[4] G. Michalos, S. Makris, N. Papakostas, D. Mourtzis, G. Chryssolouris, CIRP J.

Manuf. Sci. Technol. 2 (2010) 8191.[5] M. Marya, Mater. Sci. Forum 580-582 (2008) 155158.[6] N.S. Ermolaeva, M.B.G. Castro, P.V. Kandachar, Mater. Des. 25 (2004)

689698.[7] R.S. Coelho, Joining of Light-Weight Materials by Friction Stir Welding and

Laser Beam Welding, Ph.D. Thesis, Ruhr-Universitat Bochum, Germany, 2008.[8] R. Borrisutthekul, T. Yachi, Y. Miyashita, Y. Mutoh, Mater. Sci. Eng. A 467

(2007) 108113.[9] A. Mathieu, R. Shabadi, A. Deschamps, M. Suery, S. Mattei, D. Grevey, E. Cicala,

Opt. Laser Technol. 39 (2007) 652661.[10] P. Peyre, G. Sierra, F. Deschaux-Beaume, D. Stuart, G. Fras, Mater. Sci. Eng. A

444 (2007) 327338.[11] L. Agudo, D. Eyidi, C.H. Schmaranzer, E. Arenholz, N. Jank, J. Bruckner,

A.R. Pyzalla, J. Mater. Sci. 42 (2007) 42054214.[12] L. Agudo Jacome, S. Weber, A. Leitner, E. Arenholz, J. Bruckner, H. Hackl,

A.R. Pyzalla, Adv. Eng. Mater. 11 (5) (2009) 350358.[13] R.S. Coelho, A. Kostka, S. Sheikhi, J.F. dos Santos, A.R. Pyzalla, Adv. Eng. Mater

10 (10) (2008) 961972.[14] R.S. Coelho, A. Kostka, J.F. dos Santos, A.R. Pyzalla, Adv. Eng. Mater 10 (12)

(2008) 11271133.[15] T. Tanaka, T. Morishige, T. Hirata, Scr. Mater. 61 (2009) 756759.[16] T. Watanabe, H. Takayama, A. Yanagisawa, J. Mater. Process. Technol. 178

(2006) 342349.[17] C.M. Chen, R. Kovacevic, Int. J. Mach. Tools Manuf. 44 (2004) 12051214.[18] H. Uzun, C.D. Donne, A. Argagnotto, T. Ghidini, C. Gambaro, Mater. Des. 26

(2005) 4146.[19] W.-B. Lee, M. Schmuecker, U.A. Mercardo, G. Biallas, S.-B. Jung, Scr. Mater. 55

(2006) 355358.[20] H. Springer, A. Kostka, J.F. dos Santos, D. Raabe, Mater. Sci. Eng. A 528 (2011)

46304642.[21] O. Kubaschewski, Iron-Binary Phase Diagrams, Springer-Verlag, Berlin, 1982.[22] H. Springer, A. Kostka, E.J. Payton, D. Raabe, A. Kaysser-Pyzalla, G. Eggeler,

Acta Mater. 59 (2011) 15861600.[23] W.M. Thomas, E.D. Nicholas, J.C. Needham, M.G. Murch, P. Temple-Smith, C.J.

Dawes, Friction Stir Butt Welding, International Patent Application no. PCT/GB92/02203, December 1991.

[24] R.S. Mishra, Z.Y. Ma, Mater. Sci. Eng. R 50 (2005) 178.[25] R. Nandan, T. DebRoy, H.K.D.H. Bhadeshia, Prog. Mater. Sci. 53 (2008)

9801023.[26] D. Lohwasser, Z. Chen (Eds.), Friction Stir Welding: From Basics to Applica-

tions, Published by Woodhead Publishing Limited and CRC Press LLC, 2010.[27] L.E. Murr, J. Mater. Eng. Perform. 19 (8) (2010) 10711089.[28] C. Leit~ao, B. Emlio, B.M. Chaparro, D.M. Rodrigues, Mater. Des. 30 (2009)

32353242.[29] N.T. Kumbhar, S.K. Sahoo, I. Samajdar, G.K. Dey, K. Bhanumurthy, Mater. Des.

32 (2011) 16571666.[30] U.F.H.R. Suhuddin, S. Mironov, Y.S. Sato, H. Kokawa, C.-W. Lee, Acta Mater. 57

(2009) 54065418.[31] W. Xunhong, W. Kuaishe, Mater. Sci. Eng. A 431 (2006) 114117.[32] M.P. Miles, T.W. Nelson, R. Steel, E. Olsen, M. Gallagher, Sci. Technol. Weld.

Join. 14 (3) (2009) 228232.[33] T. Saeid, A. Abdollah-zadeh, T. Shibayanagi, K. Ikeuchi, H. Assadi, Mater. Sci.

Eng. A 527 (2010) 64846488.[34] M. Ghosh, K. Kumar, R.S. Mishra, Mater. Sci. Eng. A 528 (2011) 81118119.[35] H.K.D.H. Bhadeshia, T. DebRoy, Sci. Technol. Weld. Join. 14 (3) (2009)

193196.[36] S. Mironov, Y.S. Sato, H. Kokawa, Acta Mater. 57 (2009) 45194528.[37] R. Rai, A. De, H.K.D.H. Bhadeshia, T. DebRoy, Sci. Technol. Weld. Join. 16 (4)

(2011) 325342.

R.S. Coelho et al. / Materials Science & Engineering A 556 (2012) 175183 183

[38] T. DebRoy, H.K.D.H. Bhadeshia, Sci. Technol. Weld. Join. 15 (4) (2010)266270.

[39] R.S. Coelho, A. Kostka, H. Pinto, A. Rothkirch, J. Dos Santos, A.R. Pyzalla.Proceedings of the 2008 Denver X-ray Conference, Colorado, USA, 48 August2008 (manuscript for electronic publication).

[40] A. Kostka, R.S. Coelho, J. dos Santos, A.R. Pyzalla, Scr. Mater. 60 (2009) 953956.[41] Y. Yong, Z. Da-tong, Q. Cheng, Z. Wen, Trans. Nonferrous Met. Soc. China 20

(2010) s619s623.[42] P. Venkateswaran, Z.-H. Xu, X. Li, A.P. Reynolds, J. Mater. Sci. 44 (2009)

41404147.[43] M.A. Mofid, A. Abdollah-zadeh, F. Malek Ghaini, Mater. Des. 36 (2012)

161167.

[44] A.C. Somasekharan, L.E. Murr, Mater. Charact. 52 (2004) 4964.[45] Y. Weia, J. Lib, J. Xiong, F. Huang, F. Zhang, Mater. Des. 33 (2012) 111114.[46] Y.C. Chen, K. Nakata, Mater. Des. 30 (2009) 39133919.[47] S. Jana, Y. Hovanski, G.J. Grant, Metall. Mater. Trans. A 41A (2010)

31733182.[48] U. Dressler, G. Biallas, U.A. Mercado, Mater. Sci. Eng. A 526 (2009) 113117.[49] J.F. dos Santos, in: G. Amancio , J.F. dos Santos (Eds.), Proceedings of an

International Seminar on Friction Stir Welding of Steels, 2223 November2007, Geesthacht, Germany.

[50] J. Bacalhau, S. Sheikhi, J.F. dos Santos, in: G. Amancio, J.F. dos Santos (Eds.),Proceedings of an International Seminar on Friction Stir Welding of Steels,2223 November 2007, Geesthacht, Germany.