Friction assisted solid state lap seam welding and additive manufacturing method

Prasad Rao KALVALA*，Javed AKRAM，Mano MISRA

Dept.of Metallurgical Engineering，University of Utah，Salt Lake City，UT 84112，USA

Friction assisted solid state lap seam welding and additive manufacturing method

Prasad Rao KALVALA*，Javed AKRAM，Mano MISRA

Dept.of Metallurgical Engineering，University of Utah，Salt Lake City，UT 84112，USA

Thispaperdescribesresultsofseamweldingofrelativelyhightemperaturemeltingmaterials，AISI304，C-Mnsteels，Ni-basedalloys，CPCu，CP Ni，Ti6Al4Vandrelativelylowtemperaturemeltingmaterial，AA6061.Itdescribestheseamweldingofmulti-layeredsimilaranddissimilarmetallic sheets.Themethoddescribedandinvolvedadvancingarotatingnon-consumablerod （CPMoorAISI304）towardtheuppersheetofametallicstack clamped under pressure.As soon as the distal end of the rod touched the top portion of the upper metallic sheet，an axial force was applied.After an initial dwell time，the metallic stack moved horizontally relative to the stationery non-consumable rod by a desired length，thereby forming a metallurgical bond between the metallic sheets.Multi-track and multi-metal seam welds of high temperature metallic sheets，AISI 304，C-Mn steel，Nickel-based alloys，Cp Cu，Ti6Al4V and low temperature metallic sheets，AA6061 were obtained.Optical and scanning electron microscopy examinationand180degreeU-bendtestindicatedthatdefectfreeseamweldscouldbeobtainedwiththismethod.Tensile-sheartestingshowedthat the seam welds ofAISI 304，C-Mn steel，Nickel-based alloy were stronger than the starting base metal counterparts whileAA6061 was weaker due tosoftening.Themetallurgicalbondingattheinterfacebetweenthemetallicsheetswasattributedtolocalizedstickandslipattheinterface，dynamic recrystallization and diffusion.The method developed can be used as a means of welding，cladding and additive manufacturing.

Lap seam welding；Friction assisted welding；Solid-state welding；Cladding；Additive manufacturing

1.Introduction

Conventional lap seam welding processes can be classif i ed as fusion-based which include resistance seam，laser beam，electron beam，plasma arc welding，soldering，and brazing and solid-state-based which include ultrasonic welding and roll bonding.The lap seam welds produced by fusion based methods are often associated with a variety of problems，including cracking，high porosity，deleterious metallurgical changes，and high residual stresses.Resistance seam welding is diff i cult to apply to aluminum alloys because of their high conductivity，low strength at temperature，and tendency to degrade the electrodes ［1］and aluminum and copper require more energy because of their low electrical resistance.Steels with high carbon equivalents need additional post-weld annealing treatments and some combinations of dissimilar metal resistance welds can form intermetallics resulting in poor mechanicalproperties or liquid metal induced embrittlement［2］.Cracking，expulsion of molten metal，and unclean work-piece surfaces can all cause defective resistance seam welds.Laser welds are sensitive to heat input.High laser pulse energy resulted in poor mechanical properties and increased discontinuities in weld joints ［3］.When the heat input was too high，craters and pores appeared in the fusion zone of AZ 31 alloy ［4］.Limitations of ultrasonic welding include an inability to weld large and thick base metals and a tendency of base metals to bond to the anvil or sonotrode ［5］.Further ultrasonic welding method has not yet been fully optimized and a number of issues remain to be addressed ［6］.Roll bonding of alloys such as Titanium to other alloys such as steels will result in the formation of titanium oxide and brittle intermetallic compounds.These metallurgical changes reduce their interface bond strength ［7］.

Cladding refers to the deposition of a f i ller metal on a substrate metal to impart corrosion，wear resistance or some desired property that is not possessed by the substrate metal. Examples of cladding include hard facing for the purpose of reducing wear，abrasion，impact，erosion，galling，or cavitation，weld cladding for the purpose of providing a corrosion-resistantsurface and buttering for the purpose of adding one or more layers of weld metal to the face of the joint or surface to be welded.Conventional cladding processes can be classif i ed as fusion-based including electric arc welding processes，brazing，electron beam welding，and laser beam welding and solid-state welding processes including explosive cladding，friction surfacing and roll bonding.The clad metals produced by conventional fusion-based methods are often associated with a variety of problems，including cracking，high porosity，deleterious metallurgical changes，and high residual stresses.High percentages of dilution，viz.，the amount of base metal in the clad metal can occur in such fusion-based clad metals.Dilution percentages are typically very high，as high as 20%-50%，in most commonly used arc cladding processes such as submerged arc. Explosion cladding is typically restricted to metals with minimum elongation of 10%-15%and a notch toughness value above 30 J at bonding temperature ［8］.Friction surfacing is limited by its inability to produce larger clad areas in less time.

Additive manufacturing methods and additive and selective subtractive manufacturing methods are conventionally used to fabricate layered，multi-material structural components.Layerby-layer fabrication of three-dimensional components can be directly fabricated from a computer-aided design，CAD model of an object.Laser Engineered Net Shaping，Direct Metal Deposition，Selective Laser Melting，and Electron Beam Melting are among a number of processes being considered as additive manufacturing methods by which material addition is achieved through melting and solidif i cation.Due to melting and solidif i cation involved in these processes，the parts made by these techniques suffer from the following limitations ［9-14］:（a）unmelted zones resulting in lack of bonding between powder particles；（b）porosity；（c）solidif i cation cracking susceptibility；（d）a cast microstructure and micro-segregation leading to compositional in-homogeneities；（e）signif i cant tensile residual stress build-up；（f）long production times for large components，often less than 1 gram/minute build rates；and （g）stiffness problems.Many dissimilar metal combinations cannot be deposited by these processes as the resultant deposited layers crack.

Solid-state additive manufacturing methods，where liquid to solid transformation is absent，have been used to address many of the shortcomings of liquid-to-solid-based additive manufacturing methods.Ultrasonic consolidation was the f i rst solid state additive and selective subtractive manufacturing method to be applied，and was shown to overcome some of the limitations of fusion-based methods.In contrast to the fusion based additive manufacturing methods，ultrasonic consolidation is a typical additive and selective subtractive manufacturing method to build up a near-net shape part which is then machined to its f i nal dimensions using an integrated，3-axis CNC milling machine.Ultrasonic consolidation suffers from limitation such as the formation of inter-foil defects ［15］.Further，in ultrasonic consolidation process，if the substrate is not stiff enough，no friction can occur between the foil being deposited and the substrate ［16］.Additionally，a conventional ultrasonic consolidation process is conducted at 175 °C by employing a heated base plate.A major limitation of this process is that it can be applied to metallic layers which are few microns thick. Thus，conventional fusion-based and solid-state additive manufacturing methods and additive and selective subtractive manufacturing methods suffer from physical，metallurgical，and mechanical limitations.

Fig.1.Friction spot weld-AISI 304-3.5 mm.

Thus，there is a need to develop alternative methods for forming strong lap seam welds between a variety of hard and soft materials while avoiding the physical and metallurgical def i ciencies of conventional welding methods.There is a further need for applying these lap seam welds for cladding and additive manufacturing applications.Our motivation for the work on lap seam welding emanated from our experimental work on friction spot welding method:Systems and Methods for Friction Spot Welding and Friction Seam Welding，S/N PCT/US2014/012355，F(xiàn)iled January 21，2014.Our results of this work have shown that it is possible to develop metallurgically bonded spot welds of relatively thicker materials up to 3.25 mm，both low melting temperature alloys such as aluminum，magnesium，copper and high melting temperature alloys such as steel，stainless steel，Ni-based alloys.Typical spot welds of AISI 304 sheets are shown in Fig.1.The success of this process motivated us to extend the same method with control of an additional parameter，viz.，and traverse of the rod along the x-axis （Fig.2（a））.The idea was found to be successful in obtaining a longitudinal seam weld between metallic sheets. The results of the longitudinal friction lap seam welding of different metallic sheets are reported in this paper.

2.Experimental

2.1.Materials

The following metallic sheets were used in this work:AISI 304；AISI 1012；Ni-based alloys:IN 600，IN 625，HX；AA 6061；commercially pure CP Cu，CP Ni，and Ti6Al4V. The thickness ranges of metallic sheets were 0.5-5.0 mm. The dimensionsofthe sheetsused were asfollows: 190 mm × 200 mm × thickness.Non-consumable rods，CP Mo with 25.4 mm dia.and AISI 304 with 25.4 mm dia.and Ra1-2 μm were used to obtain the seam welds.CP Mo rod was used for seam welding of high temperature melting metallic sheets，AISI 304，AISI 1012，In 600，IN 625 and HX sheets，CP Cu，CP Ni，and Ti6Al4V.AISI 304 rod was used for low temperature melting metallic sheets，AA 6061.It can be noted that the non-consumable rods used were pin less in contrast to thetool used with pin for the conventional friction stir welding.The chemical composition of the sheets and rods used are shown in Table 1.

Fig.2.Schematic.

2.2.Single track seam welding （Fig.2（a））

In this method，an upper metallic sheet was overlaid on to a lower metallic sheet which was clamped under pressure to an anvil.In some cases，an intermediate metallic sheet was also positionedinbetweenupperandlowermetallicsheets.Theseam method involved advancing the rotating non-consumable rod，CP Mo or AISI 304，toward the upper sheet of a metallic stack clamped under pressure.As soon as the distal end of the rod touchedthetopportionoftheuppermetallicsheet，anaxialforce was applied generating frictional heat and plasticizing a portion of the upper metallic sheet.After an initial dwell time，the metallic stack moved relative to the stationery non-consumable rod by a desired length，thereby forming a metallurgical bond between the metallic sheets.Following completion of the weld，the rod is retracted.The length of each seam weld was about 125 mm long and 25 mm wide.

2.3.Multi-track seam welding （Fig.2（b））

Multi-track seam welding was formed by repeating the above steps after the clamped metallic sheet stack was axiallyshifted perpendicular relative to the welding direction by a desired length.The second seam weld overlapped the f i rst seam weld relative to the transverse axis.The amount of overlap between the f i rst seam weld and the second seam weld was between 10%and 20%of the width of the f i rst seam weld.The lap seam welding method described above was repeated to obtain multi-layer stacks.The sequence described for two consecutive metal sheets was continued by adding a third sheet to the pre-existing two sheet welded stack.These metallic sheets comprised similar as well as dissimilar metal combinations.In another series of experiments，2 mm thick Inconel 625 sheets with dimensions 150 mm × 150 mm，were used to fabricate a metallic micro reactor.Micro-channels with dimensions:depth of 1 mm with a length of 25 mm and width of 2 mm were milled over one sheet using a CNC milling machine.Two holes with a diameter of 3 mm were drilled holes at the edges of the linear micro-channel milled.By overlapping another plain sheet of Inconel 625 on the milled sheet，seam welding was performed as described above.To view the concealed microchannel a radiograph of a friction seam weld of Inconel 600 was taken.To assess the continuity of the micro-channel，colored water was injected from one hole to exit through another hole via the micro-channel of seam weld.

Table 1Chemical compositions of metallic sheets used.

The longitudinal length from the collet end to the distal end of the rod ranged from 50 mm to 100 mm.The process parameter range used was as follows:Non-consumable rod rotational speed 800-1600 RPM；axial force 5-12 KN and weld traverse speed 15-20 mm/min.A commercially available friction stir welding machine was used to achieve the seam welds.Tables 2 and 3 show the weld process parameters for similar and dissimilar metal welds.

2.4.Temperature measurements

The temperature generated due to the frictional heat was measured between two overlapped metallic sheets，AISI 304，1.5 mm thick each using K-type thermocouples （Fig.3）.It was not possible to place thermocouples exactly under the rod as the thermocouple will be destroyed due to frictional heat androtation.It was also not possible to place thermocouple at the weld interface again for the same reason.Therefore，thermocouples were placed at some def i ned distances from the center of the rod both on top side of AISI 304 and the interface between two sheets.

Table 2Seam weld process parameters （similar metals-single and multi-track）.

2.5.Evaluation of seam welds

Bond integrity was evaluated using optical and scanning electron microscopes and 180 degree U-bend test.Microstructure of weld samples was evaluated using optical，scanning electron microscopes （SEM）and SEM-Electron backscatter diffraction （EBSD）.EBSD studies were conducted on seam welds using an FEI Quanta-600 scanning electron microscope equippedwith TSL-OIM software.Thescanareawas 150 μm × 150 μm.Tensile-shear tests was conducted to assess the tensile strength of the seam welds （Fig.4）with tensile test axis perpendicular to the direction of seam welding.The tensile-shear test was conducted on single seam welds of similar materials having the same thickness.It was not done for multi layered welds.Whether it is single seam or multi layered seam weld，with the existing shear test method，the bond strength of only one interface can be tested.For comparative purposes，the base materials，without any weld，were also subjected to tensile testing using the reduced cross section sheet tensile test.Vickers hardness test was carried on limited welds across the cross section.AISI 304 portion of AISI 304/C-Mn steel dissimilar weld was subjected to intergranular corrosion testing as per ASTM-A262 Practice A:10%Oxalic acid electrolytic etch.

3.Results and discussion

3.1.Similar metal single track seam welds

Table 2 shows the optimized parameters for single seam welds between two layers of AISI 304；AISI 1012；HX alloy； Inconel 625 and Inconel 600 alloy and AA6061.Typical single seam weld，AA6061 is shown in Fig.5（a）.

Fig.3.Thermocouple arrangement-schematic.

3.2.Similar metal multi-track seam welds

Applying the single track weld parameters，multi-track welds were made as per procedure described earlier.After fabricating multi-track weld sheets between two overlapped sheets，a third sheet was overlapped to make three layered seam weld stack.Multi-track seam welds were made on two and three sheets ofAISI 304；AA1012；HX alloy andAA6061 sheets.The length and width of each seam weld was about 125 mm and 25 mm respectively.Fig.5（b）shows a typical multi-track and multi-layered seam welded AISI 304.

3.3.Dissimilar metal single and multi-track seam welds

The process parameters for dissimilar metal single track welds are shown in Table 3.Dissimilar metal welds were made between two sheets and also three sheets.Fig.6（a）shows typical single-track seam weld between CP Cu （0.5 mm）sandwiched in between two AISI 304，1.5 mm sheets.Fig.6（b）shows a single-track seam weld of an Inconel HX alloy，2.0 mm and an AISI 304 stainless steel，1.5 mm sheet.Fig.7 shows typical multi-track seam weld between Ni-200 sheet，1.5 mm）to AISI 304，1.5 mm sheet.

Table 3Seam weld process parameters （dissimilar metals）.

Fig.4.Tensile-shear test specimen dimensions.

3.4.Micro-reactor

Fig.8-1and2depictthemacrographsandFig.8-3radiograph ofamicro-reactorseamweldofInconel600sheetsdepictingthe drilled holes and the micro-channel as dark patches.Fig.8-4 depicts the color of water exiting through the right hole，via the micro-channelofseamweld，wheninjectedthroughthelefthole，indicating the presence of a concealed micro-channel.The micro-channel weld interface between Inconel 600 sheets was found to be free of any physical defects.The grain size at the interface was found to be 5-10 μm compared to the original grain size of its base metal counterpart，60 μm.This could be attributed to the dynamic recrystallization mechanism.

3.5.Bond integrity evaluation

Fig.5.Single seam weld AA6061 seam welding and multi-track and multilayered seam welding-AISI 304.

Fig.6.Single-track seam welding-CP Cu and single-track seam HX alloy.

Optical and scanning electron metallography and 180 degree U-bend test showed that the seam welds were free of any no-bond zones or any types of physical defects，such as porosity and cracking.Typical U-bent specimens of seam welds are shown in Fig.9（a）1，2 and 3.

3.6.Tensile-shear test

Fig.7.Multi-track seam Ni-200 with AISI 304.

Fig.8.1 and 2-depict the macrographs；3-radiograph of a micro-reactor；4 -color of water exiting through the right hole.

In the tensile-shear tests，all the welds，except AA6061，failed in the base metal and not in the weld，indicating that the seam welds were stronger than the base metal counterparts. The weld integrity of these seam welds can be attributed to the metallurgical bonding between the upper and lower sheets. However，AA6061 seam weld failed in the heat affected zone（HAZ）.AA6061 was welded in aT-6 aged condition and due to the thermal cycles experienced，the HAZ showed a softening effect.Fig.9（b）shows typical tensile-shear tested specimen of AISI 304.It failed in the base metal indicating that the seam weld was stronger than the base metal.Same trend was shown by all the welds except AA6061 which showed inferior tensile strength attributed to softening.AA6061 was in T-6 artif i cially aged condition and due to thermal cycles，the precipitates would have gone in to solution softening the alloy.The tensile values obtained by the welds is shown in Table 4.

Fig.9.Bend tested seam welds.

Table 4Tensile test results.

3.7.Intergranular corrosion testing

The microstructure of the ASI 304 sheet cladded to AISI 1012 steel substrate showed a “step”structure，similar to the one showed in Fig.6，indicating that the frictional heat generated by the seam welding did not lead to any “ditch” structure and demonstrating the absence of intergranular corrosion in the cladded 304 sheet.Microstructural:EBSD results showed that there is a signif i cant reduction in the grain size at the seam weld interface compared with their starting base metal sheets（Fig.10）.For example，the weld interface grain size was found to be 2-4 μm compared to the original grain size of Inconel 625 which was 25-30 μm.This is an indication that there was an increase in temperature which can be attributed to the frictional heat as well as adiabatic heating due to the plastic deformation at the weld interface.The f i nest microstructure of the weld interface area indicates that this location was subjected to the most severe plastic deformation.Further，it was noticed that the grain size started to increase as one moves away from the weld interface toward the outer surface of the sheet.As one moves toward the outer surface of the sheet，the original grain size of the sheet was more or less retained.The f i ner grains at the weld interface ref l ected in the hardness values also，f i ner grains exhibiting relatively higher hardness than the coarser （Fig.11）. The hardness at the weld interface was found to be 360 VHN compared to 236 VHN of base metal.

The fact that the grain size changed from the interface of the seam weld to the outer surface of the top and bottom sheets indicates that the weld interface was subjected to a different degree of plastic f l ow，induced by the rotation and axial forceexerted by the non-consumable rod，compared to the rest of the sheet thickness.It also shows that there must be a gradation in plastic deformation from the weld interface toward the outer surface of the sheets.

Fig.10.EBSD micrograph of Inconel 625 seam weld.

Fig.11.Hardness prof i les for seam welded IN 625 （HV0.5）.

Recrystallization isa restoration process within the deformed microstructure with the formation of new dislocation free grains.Recrystallization taking place during deformation at elevated temperatures is termed as dynamic recrystallization（DRX）and is known to be exhibited in relatively low-tomedium stacking fault energy metals such as steels，nickel based alloys，copper etc.Recrystallization removes most of the dislocations，but the material still contains grain/sub-grain boundaries，which are thermodynamically unstable ［17］.The driving force for recrystallization is known to be supplied by the stored energy toward the nucleation of new grains both within the interior of the grain and at the grain boundaries ［18］. Metals with high stacking fault energy such as Al alloys，are known to exhibit continuous dynamic recrystallization ［19］especially under hot working/thermo mechanical processing where as metals with medium stacking fault energy such as Copper are known to exhibit discontinuous dynamic recrystallization under hot working conditions ［20］.Similar DRX phenomenon was found for other welds studied.Typical optical micrographs are shown in Fig.12 （Inconel 625）.

3.8.Temperature measurements

Fig.12.DRX phenomenon Inconel.

Fig.13.Temperatures on the top sheet.

Fig.13（a）shows the temperature prof i les recorded on the top sheetofaseamweldofAISI304sheets.Thetemperaturesforthe 3 points were plotted against known distances and extrapolated to the periphery of the rod，10 mm （Fig.3）.Based on the extrapolated value of temperature，the temperature attained between rod and upper sheet is presumed to be ～1050 °C（Fig.13（b））.Similarresultswereobtainedfortemperaturesatthe interface portion of the two sheets （Fig.3）.Based on the extrapolation，the temperature between two overlapping sheets near the weld interface is presumed to be ～900 °C.Peak temperatureofabout1200 °Cisreportedforfrictionsurfacingof austenitic stainless steels ［20］and 1050 °C for tool steel ［21］. These values can be applied to 3 layered seam weld （Fig.13（c）），AISI 304/CP Cu/AA6061.The fact that CP Cu showed wrought metal structure at the weld interface of AISI 304 and CP Cu indicates that the temperature attained was below 1040 °C than theextrapolatedvalue900 °C.ThefactthatAA6061meltedand showed a cast dendritic structure （Fig.13（c））indicates that the temperatures at the interface between CP Cu andAA6061 were above ～600 °C.These results suggest that it is possible to seam weld metals with extremely different melting temperatures such as Cp Cu and Al alloy-AA 6061.However，this weld cannot be called a solid state weld，becauseAA 6061 became liquid due to the high heat generated in the process.Some more work is underway，to prepare dissimilar metal welds to f i nd out which metal combinations can be and cannot be friction seam welded. 3.9.Mechanism of bonding

Fig.14.AISI 304 seam weld-Grains extending from one metallic sheet to the other metallic sheet indicating diffusion.

It is likely that the metallurgical bonding at the seam weld interface is formed by the formation of f i ner grains between two abutting metallic sheets because of dynamic recrystallization. Inter diffusion between two abutting surfaces could also have played a role in the formation of metallurgical bonding.The grains extending from one metallic sheet to the other metallic sheet are indicative of the diffusion between the two metallic sheets （Fig.14）.Fig.15 shows SEM micrograph and EDS prof i les of CP Cu-CP Ni seam weld at the weld interface.It was found that both Cu and Ni diffused in to each other at the interface.Because，Cu and Ni have complete solid solubility in each other，they form solid solution and therefore，no deterioration in mechanical properties such as bend ductility is expected. However，if two dissimilar metals are likely to form intermetallic phases during diffusion，then one should control process parameters very closely to see to it that diffusion does not take place. Detailed studies are under way to study the diffusion aspects of various dissimilar metal welds to understand the likelihood of formation of intermetallic phases near the weld interface.

Fig.15.SEM-EDS prof i les of Cu and Ni near the weld interface of Cu-Ni seam weld.

Fig.16.AISI 1012 steel weld interface-stopped just before weldingformation of spiral f l ow lines due to material.

The rotational rate and an axial load applied to the rod can impart rotational force to the plasticized metal on the top surface of the upper sheet.It is contemplated that the imparted rotational force can cause elasto-plastic deformation due to the shear forces at the interface between the bottom surface of upper metallic sheet and top surface of lower metallic sheet. More particularly，tangential，centrifugal or rotational forces，can lead to the formation of spiral f l ow lines that emanate from a center portion of the metallic sheets to the peripheral portions of the metallic sheets （Fig.16）.The plastic deformation could be due to the relative movement between the two abutting metallic sheets at the weld interface.It is contemplated that the sudden rise in temperature at the distal end of the rod and the upper metallic sheet can result in extensive plastic deformation. This sudden rise in temperature could also be due to the adiabatic heating resulting from viscous heat dissipation during plastic deformation.It is contemplated that highly localized stick and slip at the interface can break up surface oxides and surface f i lms.This，in turn，can facilitate metal-to-metal contact between contacting surfaces at the interface.The metal-tometal contact area can grow due to the continued stick and slip mechanism，resulting in a continuous metallurgically bonded interface.Further，there can be a rise in localized temperature at the interface between the upper and lower metallic sheets due to the combined effects of conduction，imparted from the rotating plasticized metal，interfacial stick-slip mechanism，and plastic deformation.The rise in temperature at the interface can help solid state diffusion to take place at the interface between the upper and lower metallic sheets.Thus，the metallurgical bonding at the interface between the upper and lower metallic sheets can result from one or more operating mechanisms，including localized stick and slip at the interface，which tends to break up surface oxides and surface f i lms，by the formation of f i ner grains between two abutting metallic sheets because of dynamic recrystallization and diffusion between the upper and lower metallic sheets while they are held together under high static axial force applied through the rod.

4.Summary

This paper describes results of seam welding of AISI 304，C-Mn steels，Ni-based alloys，CP Cu，CP Ni，Ti6Al4V，AA6061. It describes the solid state seam welding of multi-layered similar and dissimilar metallic sheets.Multi-track and multimetal seam welds were obtained which were defect free and metallurgically bonded.Tensile-shear testing showed that the seam welds were，in general，stronger than the starting base metal counterparts.The metallurgical bonding at the interfacebetween the metallic sheets was attributed to localized stick and slip at the interface，dynamic recrystallization and diffusion. The method developed can be used as a means of welding，cladding and additive manufacturing.

5.Conclusions

1）Friction assisted solid state lap seam welding method is reported for metal joining，welding，cladding and multilayered structures as part of additive manufacturing.

2）Multi-track，multi metal and multi-layered friction seam welds were successfully fabricated using AISI 304；AISI 1012；Ni-based alloys:IN 600，IN 625，HX；CP Cu，CP Ni，Ti6Al4V and AA6061.Micro-channeled microreactor was successfully fabricated using the friction seam welding method.

3）The metallic sheets were found to be metallurgical bonded by this method which exhibited higher tensile strength values than their base metal counterparts.

4）The seam weld interface exhibited dynamic recrystallization phenomenon with f i ner grains compared to the starting metal sheets which showed relatively higher hardness values.

5）The metallurgical bonding at the interface between the upper and lower metallic sheets was proposed to result from one or more operating mechanisms，including （a）localized stick and slip at the interface，which tends to break up surface oxides and surface f i lms，（b）by the formation of f i ner grains between two abutting metallic sheets because of dynamic recrystallization and （c）diffusion between the upper and lower metallic sheets.

［1］Prangnell PB，Bakavos D.Novel approaches to friction spot welding thin aluminium automotive sheet.Mater Sci Forum 2010；638-42:1237-42.

［2］Liu L，Xiao L，Chen DL，F(xiàn)eng JC，Kim S，Zhou Y.Microstructure and fatigue properties of Mg-to-steel dissimilar metal spot welds.Mater Des 2013；45:336-42.

［3］Ventrella VA，Berretta JR，de Rossi W.Pulsed Nd:YAG laser seam welding of AISI 316L stainless steel thin foils.J Mater Process Technol 2010；210:1838-43.

［4］Quan YJ，Chen ZH，Gong XS，Yu ZH.Effects of heat input on microstructure and tensile properties of laser welded magnesium alloy AZ31.Mater Charact 2008；59:1491-7.

［5］Popoola OO，Skogsmo JB，Wexler A，Reatherford LA，Vries E，Wilkosz DE.Ultrasonic welding apparatus.US Patent US 6523732 B1；2003.

［6］Uematsu Y，Tokaji K，Tozaki Y，Kurita T，Murata S.Effect of re-f i lling probe hole on tensile failure and fatigue behavior of friction stir spot welded joints in Al-Mg-Si alloy.Int J Fatigue 2008；30:1956-66.

［7］Bae DH，Jung SJ，Cho YR，Jung WS，Jung HS，Kang CY，et al.Effect of pre-heat treatment on bonding properties in Ti/Al/STS clad materials. J Korean Inst Met Mater 2009；47:573-9.

［8］Banker JG，Reineke EG.Explosion welding.ASM International Hand Book，vol.6.USA:Explosive Fabricators，Inc.；1993.

［9］Gaytan SM，Murr LE，Medina F，Martinez E，Lopez MI，Wicker RB. Advanced metal powder based manufacturing of complex components by electron beam melting.Mater Technol Adv Perform Mater 2009；24: 180-90.

［10］Gong H，Raf iK，Starr T，Stucker B.Effect of defects on fatigue tests of as-builtTi-6Al-4V parts fabricated by selective laser melting，Presented at International Solid Free form Fabrication Symposium，The University of Texas at Austin，August 6-8，2012.

［11］Attar E.Simulation of selective electron beam melting processes ［Ph.D. thesis］.Universit at Erlangen-Nurnberg； 2011 ＜http://d-nb.info/ 1013101472/34＞.

［12］Loeber L，Biamino S，Ackelid U，Sabbadini S，Epicoco P，F(xiàn)ino P，et al.，Comparison of selective laser and electron beam melted titanium aluminides，Presented at International Solid Free form Fabrication Symposium，The University of Texas at Austin，The University of Texas in Austin on August 6-8. ＜http://sffsymposium.engr.utexas.edu/ Manuscripts/2011/2011-43-Loeber.pdf＞；2011 ［accessed 7.12.15］.

［13］Martukanitz R，Whitney E，Kelly S，Donnellan T.Laser additive manufacturing， Penn State University， ＜http://www.nmc.ctc.com/ useruploads/f i le/events/PA_Thomas_Donnellan.pdf＞； 2010 ［accessed 7.12.15］.

［14］Kong CY，Soar RC，Dickens PM.Characterisation of aluminum alloy 6061 for the ultrasonic consolidation process.Mater Sci Eng A 2003；363:99-106.

［15］Janaki Ram GD，Yang Y，Stucker BE.Effect of process parameters on bond formation during ultrasonic consolidation of aluminum alloy 3003. J Manuf Syst 2006；25:221-38.

［16］Robinson J，Janaki Ram GD，Stucker BE.Role of substrate stiffness in ultrasonic consolidation.Int J Rapid Manuf 2011；2:162-77.

［17］Ganapathysubramanian S，Zabaras N.Deformation process design for control of microstructure in the presence of dynamic recrystallization and grain growth mechanisms.Int J Solids Struct 2004；41:2011-37.

［18］Mariani E，Ghassemieh E.“Microstructure evolution of 6061 O Al alloy during ultrasonic consolidation:an insight from electron backscatter diffraction”.Acta Mater 2010；58:2492-503.

［19］El Wahabi M，Gavard L，Montheillet F，Cabrera JM，Prado JM.Effect of initial grain size on dynamic recrystallization in high purity austenitic stainless steels.Acta Mater 2005；53:4605-12.

［20］Puli R，Janaki Ram GD.Dynamic recrystallization in friction surfaced austenitic stainless steel coatings.Materials Characterization 2012；74: 49-54.

［21］Prasad Rao K， Veera sreenu A，Khalid Raf i H， Libin MN，Balasubramaniam K.Tool steel and copper coatings by friction surfacing-A thermography study.J Mater Process Technol 2012；212:402-7.

Received 3 October 2015；revised 2 November 2015；accepted 3 November 2015 Available online 24 November 2015

Peer review under responsibility of China Ordnance Society.

*Corresponding author.Tel.:+1 415 722 8105.

E-mail address:Prasad.Kalvala@utah.edu （P.R.KALVALA）.

http://dx.doi.org/10.1016/j.dt.2015.11.001

2214-9147/? 2015 China Ordnance Society.Production and hosting by Elsevier B.V.All rights reserved.

? 2015 China Ordnance Society.Production and hosting by Elsevier B.V.All rights reserved.