Analysis of Flow of Materials in Dissimilar FSW …ijeem.org/Papers/Jun2015/Analysis of Flow of Materials in...Analysis of Flow of Materials in Dissimilar FSW between Al Alloy and

International Journal of Engineering, Economics and ManagementISSN: 2319-7927, Volume 3, Issue 4

14

Analysis of Flow of Materials in Dissimilar FSW betweenAl Alloy and Mg Alloy at a Low Rotational Speed

C. B. Jagadeesha

Abstract: The aim of this study is to analyze the flowbehavior of materials in FSW of Mg alloy to Al alloymaterials at a low rotational speed of 300 rpm and at a lowwelding speed of 50 mm/min. It was observed that duringthe dissimilar welding at 300 rpm, some of the material ofthe advancing side rotated almost 3600 and transferred tothe position just behind the previous position i.e. initialadvancing side position and some of the material of leadingedge region transferred to the trailing edge region and sucha change continued. Wrinkles formation begins at the frontside of the forward moving tool and as the materialtransferred to backside they will become onion ringspattern, and this is inevitable because the formation ofonion rings at once in the backside region of the tool isclearly impossible. Below 295 rpm the welds werecompletely defective; this was owing to insufficient plasticdeformation of the materials at lesser (< 295 rpm) rpm. 295rpm is the minimum rpm at which the flow stress of thematerial in the weld volume is just enough to form goodweld.

Keywords: AA 2024-T3 Al alloy, AZ31B-O Mg alloy,Dissimilar friction stir welding, Material flow

I. INTRODUCTION

FSW is a solid state joining process. The materialsespecially light metals such as Al [1], Mg and their alloys[2]-[7], which cannot be welded properly by fusionwelding can be welded effectively by FSW [8]. In a FSWtool, advancing side (AS) lies where linear velocityvector of rotating tool and welding direction are one andthe same, whereas the retreating side (RS) lies wherelinear velocity vector of rotating tool and weldingdirection are opposite to each other. The AS of the toolmoves counter to material flow during FSW. In FSW, thefront portion of the moving tool is the leading edge (LE)and the backside of the moving tool is the trailing edge(TE).

As the rotating FSW tool pierces down into the abuttedplates the following things occur consecutively. Thebeginning heat generation occurs owing to frictionbetween tool pin and the workpiece; as the heatgeneration increases and material around the tool softens,simultaneously tool gets heated. Within a short timeseizure occurs between tool surface and plasticizedmaterial around the tool. Plasticization of materialadjacent to tool surface is highest and decreases, radiallyoutwards up to interface of base metal and weldvolume’s outer surface; between these two surfaces or

*E-mail,address:[email protected].: +91 9880529105; fax: +91 80 2360 0648

regions in weld volume, sticking and shearing betweenlayers of materials occurs leading to heat generationamong the layers and thus there is increase oftemperature in weld volume. At this juncture there willbe no friction between tool and material and also no heatgeneration owing to nullified friction.

The analysis of material flow is an important tool tounderstand the role of FSW tool on the weld formation[9]. There have been many experimental studies [10]-[14] and computational procedures [15]-[17] performedto analyze the material flow, and it has been reported thatthe material flow in FSW is quite complex and notclearly understood. The computed streamlines in thehorizontal planes around the tool pin showed thepresence of nearly circular closed streamlines, indicatingthe presence of a plug of material [15]. The tangentmovement of the material takes the main contributions tothe flow of the material in FSW [16].

Fratini et al. [18] made an important contribution tomaterial flow visualization by relating material flow withmicrostructural evolution. Formation of onion ringpattern in FSW has been studied by Fratini et al. [18] andKrishnan [19]. Krishnan [19] described this using a claymodel, and pointed out that this was a geometric effect.He reported that semi cylindrical sheets of material wereextruded during each rotation of the tool and crosssectional slices through a set of semi cylinders wouldform onion ring pattern. Studies show visible patterns incolder welds, with no observable ring pattern at hotterwelds [20]. There is no correlation between onion ringpattern and the resulting quality of the weld nugget [13],[21]. There is no experimental evidence that confirms themechanism of onion ring formation proposed byresearchers [9].

Reynolds [22] mentioned that understanding of the flowof the material is critical to determine the accuratethermo mechanical processing conditions during FSW.Tool geometry is the prime parameter in type of flowduring FSW. Analysis of flow of material is an importanttool to understand the mechanism of weld formation andphysical insight of the process, which in turn is necessaryto manipulate the welding parameters to produce defectfree welds. Li et al.[13] &[ 23] studied the flow ofmaterial in joint of AA2024 and A6061 using differentialetching technique. They suggested that the dynamicrecrystallization must takes place to provide superplasticeffect to assist the instantaneous solid state flow. They

mailto:address:[email protected]


15

studied microstructure and revealed the complexintercalated, lamellar like flow patterns in the weldvolume forms vortex, whorl, and swirl features which arecharacteristic of chaotic dynamic mixing.

Schmidt et al. [11] took copper as a marker material invertical, longitudinal and combined configurationscombined with X ray computer tomography. Theyclassified the material flow zones into three zones,namely, deflection zone, transition zone and rotationzone. Material in rotation zone rotates with the tool pin,deflection zone slightly away from tool is affected byslight deformation, and transition zone is between thesetwo zones. As the tool advances, the base metal in the LEenters the deflection zone, then transition zone andfinally the rotation zone.

M. Guerra et al.[12] inserted 0.1 mm thick high puritycopper in between Al alloy plates and traced back bymetallographic means after instantaneous stopping oftool. They reported two types of material flow, first, ASmaterial entered into the rotating zone and advancessimultaneously with the pin, which is sloughed offbehind the pin in arc shaped feature. Second flow isextrusion of the material around the tool, on the RS. Theformer is controlled by rpm and the latter is controlled bywelding speed. Seidel et al. [24] and Reynolds [14]studied the material flow in butt welded AA2195-T8using markers made of AA5454-H32. They found thatthe material in the AS moves down, while the material inthe RS moves up. Reynolds [22] states that the frictioncoefficient will strongly affect the material flow andmicrostructure. He pointed that material flow occurs inthree horizontal parallel zones.

Sinha et al. [25] suggested that material flow occurs intwo modes, first, layer by layer flow caused by theshearing action of the tool shoulder and second, extrusionof the plasticized metal around the tool pin. Schneiderand Nunes Jr. [26] (Nunes kinematic model) classifiedthe material flow as straight through current whichbypasses the tool pin and maelstrom current which goesseveral rotation around and with the pin. Abregast metalworking model [27] regards the FSW as a materialworking process that involves five zones: preheat, initialdeformation, extrusion, forging, and post weld cooldown.

Pooya pourahmad et al. [28] performed weld between Al6013/Mg, and observed micrographs of cross section ofwelds revealing three regions. The first upper regionexperienced extensive extrusion and stirring action wherezigzag interface, interlocking and islanding wereobserved. The second middle region experienced lessextrusion of Al into Mg than the upper portion where theweld interface was smoother. The third bottom region isaround the weld root and the extrusion of Al into Mg wasthe least.

In this report the role of the threaded tool on the FSWformation of dissimilar materials (AZ31B-O Mg alloy to2024-T3 Al alloy) and the resulted material flow havebeen analyzed at a low rotational speed (300 rev min-1).

II. EXPERIMENTAL METHODS

Indigenously developed computer controlled FSWmachine (BiSS Bangalore) was used for all FSWexperiments. The base materials used were 2024- T3 Alalloy and AZ31B-O Mg alloy. Composition of 2024-T3Al alloy: 4.3 – 4.5%, copper; 0.5 – 0.6 %, manganese;1.3 – 1.5 %, magnesium and less than a 0.5 % of silicon,zinc, nickel, chromium, lead and bismuth. This hastensile strength = 400 to 427 MPa, yield tensile strength= 269 to 275 MPa, elongation = 10 to 15 %, young’smodulus = 73 GPa. Composition of AZ31B-O Mg alloy:2.5 – 3.5 %, Aluminum; 0.7 – 1.3 %, zinc and 0.20 – 1.0%, manganese. This has tensile strength = 240 MPa,yield tensile strength = 140 MPa, elongation = 10 %,young’s modulus = 45 GPa.

The size of Al alloy and Mg alloy plates: 250 mm lengthx 80 mm breadth x 5 mm thickness. During thedisplacement controlled FSW, 2024-T3 Al alloy andAZ31B-O Mg alloy plates were kept in AS and RS of theFSW tool respectively. FSW Tool material was HDS andtool had threaded (two and a half rounds of left handthreads to a little depth on the pin surface) pin withrounded tip, top pin diameter 6 mm, bottom pin diameter4 mm, pin length 4.67 mm and smooth tool shoulder ofdiameter 15 mm. Initially FSW performed on dissimilarmaterials by setting welding speed to 100 mm/min, zeroIO, tool plunge depth to 4.88 mm, rpm varied from 300to 400, but a through hole at the bottom of the WNobserved. This through hole did not disappear even for80 mm/min and 60 mm/min welding speeds.


16

Fig.1. Weld coupon obtained at parameters, 300 rev min-1 to 400 rev min-1, 50 mm min-1 optimized welding speed; scale incms.

Fig.2. Optical micrograph of cross section sample, at 305 rev min-1 and 50 mm min-1 welding speed. Arrow coming out ofpage is the welding direction.

Fig.3. Surface morphology of the FSW of Al alloy to Mg alloy weld coupon, obtained by interface offset varying weldingwith optimized parameters: rotational speed = 300 rev min-1, welding speed = 50 mm min-1.

Fig. 4 Cross sectional macrostructure of joint sample at zero interface offset, cut from interface varied weld coupon shownin Fig 3.

AS

2 mm


16





AS

2 mm


16





AS

2 mm


17

Finally for 50 mm/min welding speed defect free soundweld was obtained; this was owing to increased forgingaction on weld volume by reduction of welding speedfrom 100 mm/min to 50 mm/min. So, for sound weld thefollowing FSW parameters were set: zero interface offset(IO), welding speed = 50 mm/min, tool plunge depth =4.88 mm, rotational speed varied from 300 rev min-1 to400 rev min-1 (Fig. 1). After welding, metallographicsamples were obtained from the weld coupon at every 12mm interval (Fig. 1). Defect free weld was obtained for asample at 305 rev min-1 (Fig. 2).

Varying interface offset FSW (Fig. 3) was conducted on2024-T3 Al alloy and AZ31B-O Mg alloy dissimilarmaterial combination for the following optimizedparameters: rotational speed = 300 rev min-1, startinginterface offset was 2 mm in RS and ending interfaceoffset was 2 mm in AS, welding speed = 50 mm s-1,plunge depth = 4.88 mm, weld length = 200 mm and tooltilt angle = 2o. After welding, metallographic samplesand tensile samples were extracted alternatively fromweld coupon (Fig. 3). In metallographic samples, Alalloy side was etched by using Kellers Reagent (1 ml HF+ 1.5 ml HCl + 2.5 ml HNO3 + 95 ml distilled water) for30 to 50 seconds and Mg alloy side was etched bysolution (14 ml out of ‘2g picric acid + 20 ml ethanol’solution + 2 ml acetic acid + 2 ml water) for 5 seconds[29].

III. RESULTS AND DISCUSSION

A. Flow Analysis by Steel Shot TracerTechnique [10]

Small steel balls (0.38 mm diameter or 0.015 in) wereused as a tracer material embedded at different positionswithin butt joint welds of 6.4 mm thick 6061-T6 and7075 – T6 plate. A weld was run along the length of theseeded butt joint and stopped at a point along the tracerpattern. By stopping the forward motion of the weldingtool while it is still in the seeded material, the steel shotdistribution around the welding tool is preserved in theend of the weld, revealing the path that the tracermaterial took in travelling around the welding tool. Postweld placing of the steel balls in each weld, asinvestigated by X ray radiography indicated an orderlyflow of the material around the tool pin. Based on theopening into the weld zone, only some of the flow ofmaterial seemed to be forced downward by the threadedpin, while the rest seemed to be simply rotated from thefront to the back of tool pin [10]; in Ref. [10], weldingparameters have not been stated.

If one observes Fig 3 (please see this Fig., in Ref.[10])and Fig 4 (please see this Fig., in Ref.[10]), the offsets(with respect to tool centerline) of steel shot tracers,1,5,9; 2,6,10,13; 3,7,11,14; and 4,8,12 are same, i.e. shots1,5,9 have same offset with tool centerline, similarly,

shots 2,6,10,13 have same offset, and so on. Let usdenote offset position of shots 1, 5, 9 as ‘A’; of shots 2,6, 10, 13 as ‘B’; of shots 3, 7, 11, 14 as ‘C’ and of shots4,8,12 as ‘D’ (please see these in Fig. 5, here, in thispaper).

By observing Fig 6 (please see this Fig., in Ref. [10]) andFig 8 (please see this Fig., in Ref. [10]), the tracers’positions or points before welding can be located at A, B,C, D (please see these in Fig. 5, here). After welding thetracers are carried or deposited approximately topositions A|, B|, C|, D| respectively, i.e. A to A|, B to B|, Cto C|, D to D|; note that positions A|, B|, C|, D| (please seethese in Fig. 5, here) are averaged positions of steel shotsafter welding as shown in Fig.6 (please see this Fig., inRef. [10]) and Fig.8 (please see this Fig., in Ref. [10]).

B. Flow Analysis in Dissimilar Material FSW

During FSW, FSW tool has two distinct motions, one isthe forward linear motion of the tool, and another is therotational motion of the tool. The forward motion of thetool, being highest at LE, which (the tool) compresses orsqueezes the material ahead of LA. The combination ofthese two motions transfers the material ahead of tool tobackside region of the tool. Extrusion of the materialoccurs owing to rotating tool at and from AS, through LEand through RS, i.e. extrusion occurs in the curvedregion ‘AS-LE-RS’. Extrusion reaches maximum at RS,i.e. why maximum temperature develops at RS [30], andsince there are more grouping of flow lines in RS region,weld volume becomes unsymmetrical. The extrudedmaterial will be laid in ‘RS- TE- AS curved region atbackside of the forward moving tool.

Defect free FSW was obtained at 300 rev min-1and 50mm min-1 welding speed between 2024-T3 Al alloy andAZ31B-O Mg alloy materials; arrow coming out of pageis the welding direction (in Fig. 2 and Fig. 4 here).Conceive that you are sitting inside the tool pin while thewelding is going on, and you are looking in the weldingdirection. At the front left side of the material, the Alalloy and at the front right side of the material from youis the Mg alloy (in Fig. 2 and Fig. 4 here). If you watchthe developments taking place in the rear part of you overthe material, from the place where you are sitting,motionless or with fixed body, you can see that the sameAl alloy from the front left part of the material istransferred to the rear left part of the material and thesame Mg alloy from the front right part of the material istransferred to the rear right part of the material. For thistype of transfer mechanism to take place material flowpattern and material transfer has to be as follows.

Material at the point A begins to seize up the tool (Fig. 6,here and Fig. 7, here) and again material begins to seizeup the tool at the point B over the layer of the material


18

stuck previously to the tool at the point A and again atthe point C the material seizes up the tool over the layerof the material previously stuck to the tool at the point B.The same process occurs continuously till the point H.The tool carries the material already stuck to it at A up toA| and leaves it being stuck over the material at that pointA|; and it carries the layer of material already stuck to itat B up to B| and leaves it being stuck over the material atthe point B|; and further it carries the layer of materialstuck to the point C and leaves it being stuck over thematerial at C|; and further it carries the layer of thematerial stuck to it at D up to D| and leaves it being stuckover the material at D|. This process continues up to pointH; but the layer of the material stuck to the tool at Hremains stuck to the material at the same position, beingcarried nowhere. H.W. Zhang et al [16] reports that, theflow of the material in front of the pin on the RS is fasterbut the flow behind the pin on the RS is slower.

This aforementioned type of flow of material has beenreported by authors (Reynolds [14] and London [31]) intheir studies with markers. It is logical to come to theconclusion that material at AS rotates almost 3600 andgets deposited at the rear AS position. Only this logicalargument can explain the type of material transferoccurring in the weld volume. Because if one thinks thatAS (position A in Fig. 6) material comes to TE ( positionD|) position and LE (position D) material comes to AS(A|), then this thinking is false because this can neverhappen really, since these motions of materials crisscrosseach other in the region behind the forward moving tool.At higher rpms a plugof material rotates around the tool,so at lower rpm material makes one revolution i.e.material at AS (in front of the tool) will be laid back toAS (behind the tool) i.e. material rotates almost by 3600.

So the flow of materials in weld volume are same forFSW of Al alloy with embedded steel shots (Fig. 5,here), as well as for dissimilar material welding (Fig. 7,here); in Fig. 5, here, see the various positions ofmaterials points before welding (A, B, C, D), andtransferred positions after welding (A|, B|, C|, D|) and inFig. 7, here, see the various positions of materials pointsbefore welding (A, B,… G), and transferred positionsafter welding (A|, B|,… G|). So one can conclude that theexplanation of material flow behavior, given previouslyfor dissimilar welding is correct and this explanation orhypothesis predicts the materials flow in weld volume,which is very same as the material flow really happeningin the FSW of Al alloy with embedded steel shots ordissimilar FSW process at low rotation speed (here 300

rev min-1). All this aforementioned analysis of flow ofmaterial in the weld volume equally applies to FSW ofsimilar materials also.

Threaded tool performs two functions, one is pushing ofthe material downwards other is effectively breaking theinterface of the two butt plates and plastically deformingthe interface, which may not be properly achieved byunthreaded tool pin. At higher rpm both functions will beperformed to a higher degree by tool pin. But at lowerrpm (such as 300 rpm here) former function of pin is oflesser extent than later function. Also materials at thefront side of the moving tool, as they are, are transferredto the backside of the tool. This is possible if and only ifthere is no mixing within the materials and if there is noconsiderable top down and bottom up movement of thematerial in the weld volume especially at regions slightlyaway from tool pin. So one can conclude that, at lowerrpm transfer of material occurs in almost horizontallayers, parallel to top and bottom surfaces of the plates.This is justified in Fig. 4 where the material transfer isuniform, consecutive and systematic in horizontal planes,throughout top to bottom of WN. So at low rpm almostextrusion type of flow of materials in horizontal layers,occur in the WN. Similar observations had been made byColligan [10]; see section 3.1 first paragraph. AfterColligan [10] subsequent studies have seen at insertedcopper foil, plated surfaces, and composite markers tofurther find these observations [12], [24], [31]. Allstudies pointed that the flow of material was orderly,with the material seeming to flow along defined path.

Okayasu et al. [32] stated that the difference in storedenergy in the material due to difference in deformationleads to differential etching contrast in the region ofonion rings. Since the extruded material is in crescentshaped cross section with wider top and narrow bottom[19] [33] (Fig.8 here), during rotation of tool from AS toRS through leading edge, there occurs the formation ofhorizontal layers (but perpendicular to pin axis) withfolds or wrinkles (Fig. 8) in the ‘extruded material increscent shape’. Those wrinkles, rucked up from top tobottom of the crescent in that said material are formeddue to the slanted orientation of the crescent shapedmaterial on which the shoulder pressure is being applied.These wrinkles, with more degrees of deformation,become onion rings after they reach the trailing edgeregion of the tool. And this process will be elaborated inmore detail further.


19

Fig.5. Plan view of tool pin; angular arrow is therotational direction and arrow is the welding direction.

Fig.6. Top view of tool pin from the shoulder side;angular arrow is the direction of rotation.

Fig.7. Exaggerated top view of tool pin (around middlelength of pin); angular arrow is direction of rotation;arrow is the welding direction.

Fig.8. Extruded material in crescent shape

While the layers of the material were being extruded atthe various points like A to A| etc. (Fig. 6 and Fig. 7), themaximum pressure occurs on the material in the trailingedge central position i.e. at point D| (Fig. 7). Theoccurrence of maximum pressure is caused by thebackward tilted tool. And also the void slot occurringbeneath the shoulder in the trailing edge is caused by theforward motion of the tool pin. On either side of slot,there is, not plasticized base metal at lower region andover this there are the HAZ plus TMAZ plus plasticizedmaterial. Owing to lesser resistance offered at the slottedportion compared to the resistance offered at the region

of BM plus HAZ plus TMAZ plus plasticized material,the layers of material which reach the trailing edgeregion and filling the slot sag more at the slot or WNregion than, at the region of base metal plus HAZ plusTMAZ plus plasticized material. That is why the onionrings being shaped in that process get sagged at thecentral and the bottom portion of the weld nugget. A fewstrata of onion rings, owing to eddy flow of material asseen in fig.2, at the bottom portion of weld nuggetsupport the present argument.

Welding direction

wrinkles

Shoulder pressure


19







Welding direction

wrinkles

Shoulder pressure


19







Welding direction

wrinkles

Shoulder pressure


20

Another possibility of wrinkles formation is, in the weldvolume, on the surfaces of contact i.e. on pin surface andon shoulder surface where highest velocity occur, owingto seizure between material and tool surface. Material atthe contact has high velocity and the velocity of materialin weld volume goes on decreases radially and vertically(from top to bottom) until the TMAZ zone of the weldvolume is reached, where the velocity of the materialsreach zero. Owing to the different velocity at differentplaces in the weld volume, shearing between materialslayers occur or shear bands occur in horizontal surfacesfrom top to bottom of the weld volume. This shearbanding also occurs radially, owing to rotating pin, butthe intensity of this is lower than that of caused byshoulder having high velocity along its surface. Theseshear bands will become onion rings when the material istransferred to backside region of the rotating tool.

During the forward motion of the tool, pin drivenmaterial is in front of the shoulder driven material; firstpin driven material is deposited backside i.e. in the cavitycreated by forward moving tool. Over this material theshoulder driven material is deposited. During this type ofdeposition the rings formed owing to wrinkles willbecome curved and bent upward and fall above thesagged rings forming onion rings in the WN, e.g asobserved by Kumar [33]. Wrinkles formation begins atthe front side of the forward moving tool and as thematerial transferred to backside they will become onionrings pattern, and this is inevitable because the formationof onion rings at once in the backside region of the tool isclearly impossible.

The compelling evidence of flash can be seen in weldcoupon (Fig. 2), which accentuates flash of materialoccurred on top portion (i.e. at RS) of weld coupon alongthe length of the weld. This flash formation can beanalyzed as follows. Flash occurred at retreating side,first due to distending of material at leading edge portioncaused by the backward tilted tool with built-in orprotruded tool pin, being ploughed into the material inwelding direction and while the material was distending,some of it was pushed aside to the retreating side by theshoulder and second, when the tool shoulder was beingangularly pushed over the material between trailing edgeand advancing side, some portion of the material at andaround leading edge got bulged and pushed aside to theretreating side by the shoulder. H.W. Zhang et al. [16]reports that the material in front of the pin moves upwardand the material behind the pin moves downward inFSW. If the plunge depth is more, then flash occurs onboth sides i.e. on AS, as well as on RS, of the tool.

Sound weld did not form for rpm less than 295 rpmirrespective of welding speed. The author had done theFSW experiment by varying rpm from 280 to 320 rpm,but below 295 rpm the welds were completely defective;

this was owing to insufficient plastic deformation of thematerials at lesser (< 295 rpm) rpm. 295 rpm is theminimum rpm at which the flow stress of the material inthe weld volume is just enough to form good weld.Moreover reducing the welding speed did not alter thetype of flow considerably.

IV. CONCLUSIONS

It is observed that during the welding, some of thematerial of the AS rotated 3600 and transferred to theprevious position i.e. AS and some of the material ofleading edge transferred to the trailing edge and such achange continues. This flow mechanism applies forsimilar as well as dissimilar material combination FSW.

In the low rev min-1 weld the material transfer is uniform,consecutive and systematic in horizontal planes, parallelto top and bottom surfaces of the workpiece throughouttop to bottom of WN. So at low rpm almost extrusiontype of flow of materials in horizontal layers, occur in theweld nugget.

Wrinkles formation begins at the front side of theforward moving tool and as the material transferred tobackside they will become onion rings pattern, and this isinevitable because the formation of onion rings at once inthe backside region of the tool is clearly impossible.

Below 295 rpm the welds were completely defective; thiswas owing to insufficient plastic deformation of thematerials at lesser (< 295 rpm) rpm. 295 rpm is theminimum rpm at which the flow stress of the material inthe weld volume is just enough to form good weld.Moreover reducing the welding speed did not alter thetype of flow considerably.

It has to be taken note of that thorough mixing ofmaterials in FSW, cannot be achieved as there thematerial in the weld volume is in the solid state (likepaste) but not in the liquid state.

ACKNOWLEDGMENT

The author is grateful to IISc,Bangalore and DRDO,India (Grant No. DRDO/MME/SVK/0618) forsponsoring this project. The author would like to thankProf. Satish V. Kailas, Mechanical Department,IISc, forhis valuable suggestions.

REFERENCES

[1] C.J. Dawes, W.M. Thomas, Weld J. 75 (1996) 41-45.[2] A. C. Somasekharan, L. E. Murr, Mater. Characterization. 52 (2004)

49-64.[3] Y. S. Sato, C. H. Seung, M. Michiuchi, H. Kokawa, Scr. Mater. 50

(2004) 1233–1236.[4] J. Yan, Z. Xu, Z. Li, L. Li, S. Yang, Scr. Mater. 53 (2005) 585-589.[5] S. Hirano, K. Okamoto, M. Doi, H. Okamura, M. Inagaki, Y. Aono,Quart. J. Jpn. Weld. Society. 21 (2003) 539- 545.[6] R. Zettler, A.A.M. da Silva, S. Rodrigues, A. Bianco, J.F. dos

Santos, Advanced Eng. Mater. 8 (2006) 415–421.[7] Saad Ahmed Khodir, Toshiya Shibayanagi, Mater. Trans. Jpn. Inst.

Met. 48 (2007) 2501 to 2505.


21

[8] W.M. Thomas, E.D. Ncholas, Mater. Des. 18 (1997) 269 – 73.[9] K. Kumar, Satish V. Kailas, Mater. Des. 29(2008) 791–797.[10] K. Colligan, Weld. J. 78 (1999) 229s-237s.[11] H.N.B. Schimidit, T.L. Dickerson, J.H. Hattel, Acta Mater. 54

(2006) 1199-1209[12] M. Guerra, C. Schmidta, J.C.McClure, L.E. Murr, A.C. Nunesb,

Mater. Charact. 49 (2003) 95-101[13] Y. Li, L.E. Murr, J.C. McClure, Scr. Mater. 40 (2000) 1041–1046.[14] A.P. Reynolds, Sci. Technol. Weld. Join. 5 (2000) 120–124.[15] R. Nandan, G.G. Roy, T.J. Lienert, T. Debroy, Acta Mater. 55

(2007), 883–895.[16] H.W. Zhang, Z. Zhang, J.T. Chen, J. Mater. Process. Technol. 183

(2007) 62–70.[17] P.A. Colegrove, H.R. Shercliff, J. Mater. Process. Technol. 169

(2005) 320–327.[18] L. Fratini, G. Buffa,D. Palmeri, J. Hua, R. Shivpuri, J. Eng. Mater.

Technol.128 (2006) 428–435.[19] K.N. Krishnan, Mater. Sci. Eng. A 327 (2002) 246–251.[20] G. Biallas, R. Braun, C. Dalle Donne, G. Staniek, W. A. Kaysser,

First Int. Conf. on FSW, June 1999 (Thousand Oaks, CA).[21] W. Tang, X. Guo, J.C. McClure, L.E. Murr, A.Nunes, J. Mater.

Process. Manuf. Sci.,7 (1998) 163 – 172.[22] A. P. Reynolds, Scripta Mater., 58 (2008) 338 – 342.[23] Y.Li, L. E. Murr, J. C. McClure, Mater. Sci. Eng. A 271 (1999)

213 – 223.[24] T. U. Seidel, A. P.Reynolds, Metallurgical Mater. Trans. A 32 No.

11 2879 – 2884.

[25] P. Sinha, S. Muthukumaran, S. K. Mukherjee, J. Mater. Process.Technol., 197 (2008) 17 – 21.

[26] J. A. Schneider, A. C. Nunes, Jr., Metallurgical and Mater. Trans.B, 35 (2004) 777 – 783.

[27] W. J. Arbegast, Z. Jin, Ed., TMS, 2003.[28] Pooya pourahmad, Mehrdad Abbasi, Trans. Nonferrous Met. Soc .

China 23(2013) 1253-1261.[29] P. Venkateswaran, A.P. Reynolds, Mater. Sci. Eng. A 545 (2012)

26-37.[30] J.E. Gould, Z. Feng, J. Mater. Process. Manuf. Sci.7 (1998) 185-

194.[31] B. London, M. Mahoney, W. Bingel, M. Calabrese, R. H. Bossi,

D. Waldron, Proc. Symp. On FSW and Processing II.[32] M. Okayasu, Z. Wang, D. L. Chen, Mater. Sci. Technol. 25 (2005)

530-538.[33] K. Kumar, Satish V. Kailas, Mater. Sci. Eng. A 485 (2008) 367–

374.

AUTHOR’S PROFILE

C. B. Jagadeesha

PhD student.Department of Mechanical Engineering, Indian Institute ofScience, Bangalore, Bengaluru- 560012, India

Analysis of Flow of Materials in Dissimilar FSW …ijeem.org/Papers/Jun2015/Analysis of Flow of Materials in...Analysis of Flow of Materials in Dissimilar FSW between Al Alloy and

Documents