metals
Article On Additive Manufactured AlSi10Mg to Wrought AA6060-T6: Characterisation of Optimal- and High-Energy Magnetic Pulse Welding Conditions
Moshe Nahmany 1,2, Victor Shribman 3, Shlomo Levi 1, Dana Ashkenazi 4,* and Adin Stern 2,5 1 Department of Materials, Nuclear Research Center-Negev, Beer Sheva 8410900, Israel; [email protected] (M.N.); [email protected] (S.L.) 2 Department of Materials Engineering, Ben-Gurion University of the Negev, Beer Sheva 8410501, Israel; [email protected] 3 Bmax Srl., Hod Hasharon 4501306, Israel; [email protected] 4 School of Mechanical Engineering, Tel Aviv University, Ramat Aviv 6997801, Israel 5 Department of Mechanical Engineering, Afeka Academic College of Engineering, Tel Aviv 6910717, Israel * Correspondence: [email protected]; Tel.: +972-3-6405579
Received: 23 August 2020; Accepted: 11 September 2020; Published: 14 September 2020
Abstract: This novel research aims to examine the macro and microstructural bonding region development during magnetic pulse welding (MPW) of dissimilar additive manufactured (AM) laser powder-bed fusion (L-PBF) AlSi10Mg rod and AA6060-T6 wrought tube, using both optimal- and high-energy welding conditions. For that purpose, various joint characterisation methods were applied. It is demonstrated that high-quality hermetic welds are achievable with adjusted MPW process parameters. The macroscale analysis has shown that the joint interfaces are deformed to a waveform shape; the interface is starting relatively planar, with waves forming and growing in the welding direction. The observed thickening of the flyer’s wall after welding is the result of its diametral inward deformation, taking place during the process. A slight increase in microhardness was adjacent to the faying interfaces; a higher increase was measured on the AlSi10Mg material side, while a smaller one was observed on the AA6060 side. Along the wavy interfaces, resolidified “pockets” of material or occasionally discontinuous short layers exhibiting different morphologies, were detected. The jet residues are typically located towards the end of the weld, confirming a temperature rise that exceeds the melting temperature of both alloys. Far from the weld zone, extremely thin-film deposits were clearly observed on the inner flyer surfaces. The formation of isolated Si particles and thin-film deposits may point out that the local increase in temperatures leads to melting or even evaporation vaporisation of superficial layers from the colliding parts. It is worth noting that this type of jet residue was discovered for the first time in the present research. The current research work is expected to provide an understanding of weld formation mechanisms of additively manufactured parts to conventional wrought parts conforming to existing wrought/wrought weld knowledge.
Keywords: additive manufacturing laser powder-bed fusion; AlSi10Mg alloy; magnetic pulse welding; optimal-energy welding conditions; high-energy welding conditions; wrought AA6060-T6
1. Introduction Joining by welding plays an important role in the fabrication of mechanical assemblies created from parts with different properties. Explosive welding, gas gun welding, laser impact welding, vaporising foil actuator welding, and magnetic pulse welding (MPW) are the five main joining techniques based on impact phenomena proven to date. For example, Bataev and his colleagues [1] studied explosion-welded steel plates; for this purpose, they used the hydrodynamics of particles for
Metals 2020, 10, 1235; doi:10.3390/met10091235 www.mdpi.com/journal/metals Metals 2020, 10, x FOR PEER REVIEW 2 of 21
Metals 2020, 10, 1235 2 of 20 techniques based on impact phenomena proven to date. For example, Bataev and his colleagues [1] studied explosion-welded steel plates; for this purpose, they used the hydrodynamics of particles for simulating the plastic deformation and generated heat heat.. Lee Lee et et al. al. [2] [2] reported reported the the effect effect of internal stress waves in vaporising foil actuator welding, using Cu-110 as the stationary part and constant thickness CP-TiCP-Ti flyers.flyers. ExperimentsExperiments and and simulations simulations show show that that increasing increasing target target thickness thickness leads leads to the to theincreasing increasing interfacial interfacial wavelength. wavelength. Flyer Flyer velocity velocity and wavy and interfacewavy interface morphology morphology were also were studied. also studied.Nassiri et Nassiri al. [3] studiedet al. [3] solid-statestudied solid-state welding processeswelding processes by combined by combined numerical numerical simulations, simulations, diffusion diffusioncalculations, calculations, and interfacial and characterisationinterfacial characterisa techniques.tion techniques. Determination Determination of the relations of the between relations the betweendevelopment the development of defects within of defects the joining within area the andjoining melting area and phenomena melting wasphenomena concluded. was concluded. In the last decade, special attention has been focused towards the technologies of dissimilar welding and joining processes [[4,5].4,5]. Stern Stern et et al. [6] [6] studied the microstructure development of the joining zone in MPWMPW ofof similarsimilar andand dissimilardissimilar workpieces.workpieces. A A continuous continuous transition transition region region or or a disrupted “pocket” type zone along the weld interf interfaceace is explained in terms of limited-area melting followed by rapid rapid solidifi solidification.cation. A A noteworthy noteworthy hardness hardness increase increase was was detected detected at at the the interface interface layer. layer. Stern and his colleagues colleagues [7] [7] investigated investigated the the molten molten welding welding pool pool during during the the MPW MPW of of Al/Mg Al/Mg couples. couples. They concluded that the temperature of the welding zone was locally greater than the melting points of Al and Mg, thus forming moltenmolten regions.regions. They found that the fusion zonezone contains equivalent amounts of both Al and Mg. The MPW method is ba basedsed on flyer flyer acceleration and collision attained through magnetic pressure, pressure, formed formed by by the the rapid rapid di dischargescharge of of a a capacitor bank bank through through a a tool-coil, tool-coil, located adjacent to the flyerflyer (Figure1 1).).
Figure 1. TheThe experimental experimental magnetic magnetic pulse pulse welding welding (MPW) (MPW) system: system: (a ()a schematic) schematic illustration illustration of ofa tubulara tubular configuration configuration of ofthe the MPW MPW setup; setup; (b) (theb) thecoil coilsetup setup (centre (centre of image); of image); (c) the (c )cylindrical the cylindrical non- conductivenon-conductive green green part partcentres centres the two the parts two parts (side (side view). view).
In the initial state, the to-be-joined componen componentsts (a (a movable movable flyer flyer and a stationary part) are positioned at a standoff, standoff, which definesdefines the acceleration distance. Lorentz Lorentz forces forces created by the eddy currents induced into the flyer, flyer, drive the flyer flyer away from from the the coil coil towards towards the the stationary stationary component. component. The imposed plasticplastic deformationdeformation and and interface interface pressures pressures of of approximately approximately several several GPa, GPa, taking taking place place for forseveral several microseconds, microseconds, create create the MP the joint. MP The joint. hydrodynamic The hydrodynamic phenomena phenomena occurring atoccurring the propagating at the propagatingcollision zone collision are expected zone to are eject expected a jet, containing to eject surfacea jet, containing materials, gases,surface and materials, contaminants, gases, while and
Metals 2020, 10, 1235 3 of 20 bringing the faying virgin surfaces into intimate contact. Kakizaki et al. [8] investigated the jet emission and the weld interface of similar and dissimilar metals. Numerical simulation of oblique collision between plates was made for various thicknesses, velocities, and angles. The composition of the jet was ruled by the degree of density difference between the two metals. For example, in cases of large density differences, the jet mainly contained the metal element with lower density. Stern et al. [9] studied the composition of the jet emitted in MP welding utilized on similar and dissimilar metal lap joints: Al/Al and Al/Mg welds. The composition of the jet remnants was governed by the density difference of the two components, as mentioned above [8]. The estimated thickness of the films peeled in the MPW process was also calculated; an average thickness of 17 µm was evaluated for the Al-Al pair and for Al-Mg pairs the values were of about 10 µm. Wang and his colleagues [10] investigated explosive/impact welding, to expand the understanding of the aspects that govern the quality of explosive-welded joints. The phenomenon of jetting and the interfacial waves detected in explosive welding were quite well replicated in the simulations. The jets are typically described as a mixture of materials from the flyer and stationary parts, their oxides, gases present in the closing gap, and other surface contaminants. The occurrence of jet phenomenon has been proven by trapping accumulated ejected materials [9], by numerical analysis [10], and by simulation of the MPW process. Ejected material may be in the form of solid debris; however, the ejection phenomena can also occur as fluid jetting when the temperature exceeds the melting temperature of the materials at the interface. The fragments from the interface are usually observed near the end of the weld zone and at the inner faying surfaces. The impact of thermal effects occurring at the interface in the course of MPW, such as local melting, has been brought into focus by numerous scholars, pointing to the identification of the major joining phenomena/mechanisms. Bellmann et al. [11] estimated the typical high velocity impact flash during MPW, using phototransistors, with the purpose of impact’s time measurement. The outcomes are in good agreement with the well-known PDV (photon Doppler velocimetry) demonstrating good repeatability; for example, Lueg-Althoff and his colleagues [12] showed “that the minimum radial impact velocity required for welding with the same geometrical setup can be reduced significantly at low discharge frequencies compared to high one”. Pourabbas et al. [13] studied the MPW of AA4014 to AA7075 and found that variety of welding parameters were carefully chosen to acquire acceptable welds. In addition, three modes of welding interfaces with wavy, molten wavy, and porous morphologies were detected. Of the three, welding with the wavy morphology showed the maximum mechanical strength. A short flash was visible for all successful MPW runs; it looks like the shock compression of the gas present in between the components is the source of the instant light emission. The hydrodynamic flow of material ejects superficial layers of both parts from the closing gap, as demonstrated by Shribman et al. [14], and that the jet that was ejected was formed from the welded faying faces of both workpieces, confirmed by Stern et al. [9]. The jet formation cleans and activates the metal surfaces prior to welding and leaves them chemically pure, preferring the establishment of metallic bonds under the prevailing interface pressure. Many of the latest MPW studies have shown evidence of melted and rapidly cooled regions along the joint interface, e.g., amorphous structures that can be attributed to rapid solidification with cooling rates in the magnitude of 107 K s 1. Recrystallised nanometric grain · − size, as well as localised melting in the range of micron-sized “pockets” were observed by several studies [3,7,15–17]. Bellmann et al. [18] examined the cloud of CoP particles, which is expelled as an outcome of the high-speed impact between the two metals. MPW tests were performed in vacuum, with diverse collision conditions to suppress the interaction with air, for a better-quality process monitoring. Böhme et al. [19] reported on the microstructure of aluminium–steel specimens produced by MPW. As one layer could be identified as Al solid solution, a very thin layer close to the steel side of the compound was found. Deeper study of this layer showed a mixture of nanocrystalline and amorphous parts. Geng et al. [20] studied an amorphous structure and the transition zone that were found in the MP-welded Al-Fe. Simulations and theoretical analyses were performed to understand the formation mechanism of such a morphology and succeeded in reproducing the wave morphology Metals 2020, 10, 1235 4 of 20 in the Al-Fe interface, and local melt was detected in the weld interface. Sapanathan et al. [21] reported the 3D simulations of MP welds: one turn coil combined with a field shaper. The forecast temperature distributions show the phenomena of Joule heating and plastic heat dissipation. It can be concluded that a limited “liquid-state welding” is a probable joining process for the MPW method, and the expulsion of debris, together with liquefied metal, is likely to happen during jetting. Sharafiev et al. [22] studied the interface microstructure of MPW of Al/Al joints; the interface displayed a wavy weld geometry. The created transitional layer consists of ultrafine grains and, in some regions, exhibits a columnar grain structure, and it is suggested that the microstructure was generated by rapid melting and solidification. The latest descriptions of MPW joints have several common aspects, e.g., regions in the form of thin layers and/or “pockets” of melted and rapidly resolidified material were frequently observed. This means, according to some researchers, that local “liquid-state welding” can additionally be considered an active joining mechanism for the MPW process, which is basically regarded as a solid-state technique [3,7]. If the local temperatures are too low before solidification has been completed, the joint quality may be lower than expected. In addition, localised melting and rapid solidification under high pressure enable non-equilibrium solidification and formation of interfacial metastable phases affecting the joint properties. MPW joints are strongly influenced by a variety of interacting physical phenomena, such as extensive plastic deformation accompanied by material mixing, local melting, and even vaporisation, facts leading to the possible formation of interfacial discontinuities in the form of voids and cracks. Additive manufacturing (AM) of metal alloys includes several technologies using different heat sources and differ in how the raw material is supplied. For example, Aboulkhair et al. [23] reviewed some recent developments in the AM field and highlights some key issues needing attention for further advance. Debroy et al. [15] describe the evolving study on AM of metallic materials and offered a wide-ranging overview of the physical procedures and properties of the printed parts. Zhang et al. [24] review the present range of alloys available for metal AM, including aluminium, titanium, and other common alloys, and compositionally complex alloys. The emphasis is the association between processing, compositions, microstructures, and properties of each system. Zuback et al. [25] discuss the role of cooling rate, alloy composition, microstructure, and post-process heat-treatment on the hardness values of AM aluminium, and other alloys. Hardness data for aluminium and steel alloys produced by AM and welding are associated to comprehend the relative roles of engineering processes. Mertens et al. [26] present the current progresses and target to identify challenges and prospects for forthcoming work. Most of the structural materials, e.g., Ti and Ni alloys, have been typically processed by both electron and laser beam; the latter is the dominant heat source for powder-bed fusion processing of Al alloys. Rosenthal et al. [27] reported the additive manufactured (AM) laser powder-bed fusion (L-PBF) AlSi10Mg response to a widespread range of strain rates, spanning from 2.77 10 6 to 2.77 10 1 S 1 and this alloy presented strain-rate sensitivity, × − × − − including weighty changes of flow stress and strain hardening. That conclusion is opposed to that reported for “conventional” Al alloys. The ongoing trend, in transport, automotive, and aerospace domains, to develop components with improved strength-to-weight ratio, has led to wide use of Al alloys and particularly of the AlSi10Mg alloy in these industries. The L-PBF technique has been broadly investigated and is currently the most used technique in AM of Al-alloys, in terms of high-quality near full-density parts. This technology enables the fabrication of complex aluminium components exhibiting unique microstructures that are hard to create by conventional methods. Very high solidification rates (of 0.1 103–5 103 mm s 1) and high cooling rates (typically 5 105 K s 1 and higher) accompanying × × · − × · − the L-PBF processing of AlSi10Mg components result in the formation of extremely fine microstructures and lead to improved quasi-static mechanical properties that can even outperform those of parts manufactured by conventional fabrication processes. Awd and his colleagues [28] examined fatigue cracking of AM L-PBF-AlSi10Mg by X-ray computed tomography and studied the influence of microstructural homogeneity on fatigue strength. The discontinuities originating during the AM-L-PBF process, such as pores, incomplete powder melting, and residual oxide layers significantly affect the Metals 2020, 10, 1235 5 of 20 mechanical properties of the alloys, such as fatigue [26,27]. The additively manufactured parts are normally subjected to further post-processing, which enable tailoring AlSi10Mg alloy’s mechanical properties. Rosenthal et al. [29] studied the correlation between the mechanical properties of AlSi10Mg (Z-oriented, or vertical) specimens exposed to diverse post-processing conditions, both thermal and thermo-mechanical. One can see the changes in the fracture mechanism and in the properties in relation to these treatments. Tradowsky et al. [30] report the influence of thermal post-processing using Hot Isostatic Pressure (HIP) with or without aging treatment and the effect of build orientation on the microstructure and mechanical properties in AM L-PBF-AlSi10Mg. The specimens show fine columnar grains, with a fine Si-enriched cellular dendritic morphology, leading to an increase in tensile strength, compared to castings. The fabrication of large hybrid modules, built from L-PBF-processed components and wrought parts, require the application of emerging joining methods to support the joining requirements. Biffi et al. [31] successfully used fusion welding of AM-AlSi10Mg parts by the laser beam technique. Nahmany et al. [32] reported a detailed investigation of the mechanical response of EBW-welded AM-AlSi10Mg, where the welded samples presented similar properties when compared to the AM-built samples. Zhang et al. [33] discussed and compared laser and Tungsten Inert Gas (TIG) welding of L-PBF and cast AlSi10Mg. It was established that L-PBF-AlSi10Mg has very high pore susceptibility, when compared to the cast alloy. Porosity creation is the main problem in welding AM L-PBF-AlSi10Mg. According to Zhang and his colleagues [33]: “The large pores distribute at the boundary of the weld for TIG welding, while the large pores distribute at the upper part of the weld for laser welding”. Nahmany et al. [34] presented, for the first time, effective electron beam welding of additively manufactured AlSi10Mg parts. That work initiated further studies in the field of welding AM parts. Nahmany et al. [35] studied and proved the feasibility of producing small thin walled AM-built pressure vessels welded by electron beam. Fusion welding of AM-Ti6Al4V parts was also demonstrated. Tavlovich et al. [36] reported successful fusion welding of AM-Ti6Al4V parts to themselves and to wrought Ti6Al4V parts and presented simulation of their process. Wits et al. [37] discussed the weldability of AM-titanium. They utilised pulsed laser beam-keyhole welding on conventionally manufactured Ti parts with comparison to AM-Ti parts. The work displayed that more specific energy (per unit weld length) is essential to achieve a comparable keyhole geometry for AM-Ti specimens. The availability of reliable low-cost welding technology, such as MPW, may offer superior design flexibility for modules that are, at present, barely possible to produce. The results of our preliminary work [14] show that L-PBF AlSi10Mg components can be successfully MP-welded to parts fabricated from wrought material, to overcome the significant limitations of the printed components size. The current study is part of continuing research on joining of AM to wrought components; the work deals with jetting phenomena, microstructural characterisation, and bonding mechanisms. For the sake of compatibility with previous experiments, the same alloys were investigated, e.g., AM L-PBF AM-AlSi10Mg rods to wrought AA6060-T6 tubes. The novelty of the present work focused mainly on the characterisation of both optimal- and high-energy welding conditions to check weldability under different MP welding conditions.
2. Experimental Part
2.1. Materials and Magnetic Pulse Welding The combination of printed and wrought material was selected for this work based on its relevance to the present and future applications of hybrid components in the manufacturing sector. The flyer component (i.e., the outer tube) consists of a wrought aluminium alloy AA6060-T6, while the stationary part (i.e., the inner rod) is made of L-PBF additively manufactured AlSi10Mg alloy. The chemical compositions of the flyer and stationary part alloys are presented in Table1. The microstructure of the as-supplied AA6060-T6 alloy consists of different shape and size inclusions; Adamczyk-Cie´slak and his colleagues [38] reported and distinguished between large inclusions with irregular shapes Metals 2020, 10, 1235 6 of 20
(Al5FeSi) and clusters of smaller and more regular form α-Al(Mn,Fe)Si particles. The stationary part was produced from AlSi10Mg alloy, due to the promising results in terms of processability Metalsby L-PBF; 2020, 10 the, x FOR amount PEER REVIEW of R & D undertaken on AlSi10Mg greatly surpasses that for most6 other of 21 alloys. This is motivated by the success in processing the AlSi10Mg alloy, and its use in different differentindustrial industrial applications. applications. The solidification The solidification mechanism inmechanism the L-PBF AlSi10Mgin the L-PBF alloy isAlSi10Mg cellular-dendritic, alloy is cellular-dendritic,resulting in a particularly resulting fine in microstructure. a particularly fine The microstructure. cellular structure The is stimulatedcellular structure when a is high-velocity stimulated whensolidification a high-velocity front is coupledsolidification with constitutionalfront is couple undercooling;d with constitutional both occurring undercooling; in L-PBF. both occurring in L-PBF. Table1. The chemical composition of the stationary part (additive manufactured (AM) laser powder-bed Tablefusion 1. (L-PBF)-AlSi10Mg) The chemical composition and the flyerof the (wrought stationary AA6060-T6) part (additive component manufactured (wt%). (AM) laser powder- bed fusion (L-PBF)-AlSi10Mg) and the flyer (wrought AA6060-T6) component (wt%). Composition (wt%) Alloy Composition (wt%) Alloy Al Si Mg Fe Cu Mn Zn V Ti Ni Al Si Mg Fe Cu Mn Zn V Ti Ni AM L-PBF AlSi10Mg Bal. 9.63 0.32 0.14 0.01 0.01 – – 0.01 0.01 AM L-PBF AlSi10Mg Bal. 9.63 0.32 0.14 0.01 ≤ ≤0.01 – – 0.01 ≤0.01≤ Wrought AA6060-T6 Bal. 0.51 0.40 0.21 0.03 0.05 0.02 0.02 0.02 0.01 Wrought AA6060-T6 Bal. 0.51 0.40 0.21 0.03 0.05 0.02 0.02 0.02 0.01
The stationary specimens, specimens, 30 30 mm mm in in diameter diameter by by 18 18 mm mm in inlength length (Figure (Figure 2a,b),2a,b), were were built built onon an L-PBFan L-PBF machine machine (EOSINT (EOSINT M280 M280 machine, machine, EOS, EOS, Krailling Krailling Germany) Germany) equipped equipped with with a 250 a 250× 250 250× 300 × × mm3003 mm build-platform3 build-platform with withan up an to up400 to W 400 continuous W continuous Yb fibre Yb laser. fibre The laser. parts The (Figure parts (Figure2) were2 printed) were printedfrom a frompre-alloyed a pre-alloyed AlSi10Mg AlSi10Mg powder powder with witha particle a particle size sizein the in therange range of of20–63 20–63 µmµm and and a a layer thickness ofof 3030 µµm,m,with with similar similar particle particle diameters diameters used used by by Inberg Inberg et al.et al. [39 [39].]. The The rods rods were were built built in the in verticalthe vertical (Z) direction(Z) direction (Figure (Figure2c) in2c) an in argon an argon atmosphere atmosphere with with maximum maximum oxygen oxygen content content of 0.12%. of 0.12%.
Figure 2. TheThe flyer flyer and stationary components and th thee integrated AM L-PBF-AlSi10Mg and wrought AA6060-T6 parts: ( (aa)) Computer-aided Computer-aided design design (CAD) (CAD) model model of of the the two two parts (flyer (flyer and stationary components) prior to the MPW welding process; ( b) CAD CAD model model of of the the welded welded specimen specimen simulation; simulation; c (c) the the L-PBF-AlSi10Mg L-PBF-AlSi10Mg stationary stationary component component after after surface surface machining, machining, build-direction build-direction included; included; (d) (d) the AlSi10Mg and wrought AA6060-T6 assembly after MPW. the AlSi10Mg and wrought AA6060-T6 assembly after MPW. The build-plate temperature was about 35 C (recommended parameter for the EOSINT M280 The build-plate temperature was about 35 ◦°C (recommended parameter for the EOSINT M280 printer), and the specimens underwent a post-processing T5 heat treatment at 300 C for 2 h, with similar printer), and the specimens underwent a post-processing T5 heat treatment at◦ 300 °C for 2 h, with conditions as described by others [29,40]. The laser beam spot diameter was about 80 µm with Gaussian similar conditions as described by others [29,40]. The laser beam spot diameter was about 80 µm with intensity distribution in the process plane and with an applied scanning velocity of about 1 m/s. Gaussian intensity distribution in the process plane and with an applied scanning velocity of about The build-strategy was the strip (8 mm wide) scanning strategy, based on a laser scan pattern rotating 1 m/s. The build-strategy was the strip (8 mm wide) scanning strategy, based on a laser scan pattern by an angle of 67 between successive layers along the Z-direction. The hatch distance of about 200 µm rotating by an angle◦ of 67° between successive layers along the Z-direction. The hatch distance of was applied with a 33% overlap. about 200 µm was applied with a 33% overlap. The components were cleaned in ethanol to remove residue from their surfaces, prior to MPW joining experiments. Preliminary tests were conducted to identify the minimum energy required for a leak-free continuous joint along the tube perimeter. The experimental runs (Figure 1a–c), with diverse charging energies, were performed on a Bmax Model 50/25 Pulse Generator (Bmax, Toulouse, France and Hod-Hasharon, Israel) with the characteristic values listed in Table 2. The current was
Metals 2020, 10, x FOR PEER REVIEW 7 of 21
measured for each trial with a Rogowski current probe, and the maximum current amplitude was calculated. Illustrations of samples, before and after the MPW welding process, are shown in Figures 2 and 3. Specimens 1, 2, 3, 4 were produced under high-energy conditions, and specimens 5, 6, 7 were Metalsproduced2020, 10, 1235 under optimal-energy conditions, as shown in Table 2. The optimal parameters were7 of 20 achieved by carrying out tests to establish collision angle, gap, positioning in coil, as well as frequency and kV. The components were cleaned in ethanol to remove residue from their surfaces, prior to MPW joining experiments.Table 2. MPW Preliminary system parameters tests werefor the conducted high and optimal to identify welding the energy minimum samples, energy AM L-PBF required for a leak-free continuousAlSi10Mg stationary joint along and AA6060-T6 the tube flyer perimeter. pairs: gap The 2.5 mm experimental (specimen No. runs 1—gap (Figure 1.5 mm).1a–c), with diverse charging energies,Specimen were Type performed and No. on a BmaxEnergy Model (kJ) 50/ Voltage25 Pulse (kV) Generator Current-I (Bmax,max Toulouse, (kA) France and Hod-Hasharon,High energy: Israel) 1, 2, with 3, 4 the characteristic 9.7 values listed 11 in Table2. The current 478 was measured for each trialOptimal with a energy: Rogowski 5, 6, current7 probe, 8.0 and the maximum 10 current amplitude 421 was calculated. Illustrations of samples, before and after the MPW welding process, are shown in Figures2 and3. Specimens2.2. MPW 1, 2, Joint 3, 4 Macro-Characterisation were produced under high-energy conditions, and specimens 5, 6, 7 were produced under(a) optimal-energy The MPW specimens conditions, were asvisually shown tested in Table (VT)2 .to The determine optimal the parameters existence of were macroscopic achieved by carrying outdiscontinuities tests to establish including collision jet residue angle, on the gap, AA6060-T6 positioning tube in inner coil, surfaces. as well as frequency and kV. (b) The as-welded samples were cemented to a commercial vacuum connector shown in Figure 3a. TableThe 2. HeMPW leak system tests were parameters performed for by the the high “spray-probe” and optimal technique welding to energy estimate samples, the weld AM quality. L-PBF AlSi10MgAn EDWARDS stationary andELD500 AA6060-T6 Helium flyer leak pairs:detector gap was 2.5 mmcoupled (specimen directly No. to 1—gap the vacuum 1.5 mm). system, to evacuate the system including the sample. The external surface of the sample was finely sprayed with Helium,Specimen particularly Type and No.around Energy the joint (kJ) zone. Voltage Any tiny (kV) leak Current-Icaused bymax defective(kA) welds, damagedHigh gaskets, energy: etc., 1, 2,will 3, 4allow helium 9.7 to penetrate and 11 be detected by 478the leak detector, as described in DIN EN 1779:1999-10, 2011 [41] standard, and as described and performed Optimal energy: 5, 6, 7 8.0 10 421 previously by Shribman, Nahmany et al. [14].
FigureFigure 3. General 3. General view view of MP-weldedof MP-welded specimen specimen No. 6 6 used used for for leak leak test: test: (a) (L-PBFa) L-PBF AlSi10Mg AlSi10Mg rod is rod is seen onseen the on upperthe upper side side and and AA6060-T6 AA6060-T6 tube tube inin thethe mi middle;ddle; the the leak leak test test adaptor adaptor appears appears on the on lower the lower side;side; (b) the (b) AA6060-T6the AA6060-T6 tube tube is is “wrapping” “wrapping” the the L-PBFL-PBF prin printedted rod. rod. The The fully fully coll collapsedapsed region region length length for all the samples is about 10.5 mm. for all the samples is about 10.5 mm.
2.2. MPW(c) Peel Joint tests Macro-Characterisation were performed to determine the quality and failure type of joints produced by diverse welding methods, such as resistance spot welding, etc.; ISO/AWI 23,598 standard [42] specifies (a) Thethe MPW geometry specimens of test werespecimens visually and the tested testing (VT) procedures to determine for such the a test. existence A manual of macroscopicpeel test discontinuitieswas used to assess including the macro jet residue weld quality on the of AA6060-T6 the welded tubespecimens inner at surfaces. four positions around the (b) The as-welded samples were cemented to a commercial vacuum connector shown in Figure3a.
The He leak tests were performed by the “spray-probe” technique to estimate the weld quality. An EDWARDS ELD500 Helium leak detector was coupled directly to the vacuum system, to evacuate the system including the sample. The external surface of the sample was finely sprayed with Helium, particularly around the joint zone. Any tiny leak caused by defective welds, damaged gaskets, etc., will allow helium to penetrate and be detected by the leak detector, as described in DIN EN 1779:1999-10, 2011 [41] standard, and as described and performed previously by Shribman, Nahmany et al. [14]. Metals 2020, 10, 1235 8 of 20
(c) Peel tests were performed to determine the quality and failure type of joints produced by diverse welding methods, such as resistance spot welding, etc.; ISO/AWI 23,598 standard [42] specifies the geometry of test specimens and the testing procedures for such a test. A manual peel test Metalswas used2020, 10 to, x FOR assess PEER the REVIEW macro weld quality of the welded specimens at four positions around8 of 21 the circumference:circumference: 0◦ ,0°, 90 90°,◦, 180 180°,◦, and and 270270°◦ (Figure 4).4). The The isometric isometric view view of the of the specimen specimen after after a peel a peel testtest is shown is shown in Figurein Figure4a,b. 4a,b. The The position position atat the co coil’sil’s slot slot (0°) (0 ◦is) of is special of special interest interest during during process process parameterparameter development, development, due due to to locally locally reduced magnetic magnetic field field intensity intensity in this in thisspecific specific region. region. (d) (d)The The macroscopic macroscopic deformation deformation ofof thethe flyer flyer tube tube wall, wall, after after MPW, MPW, was was measured measured at different at diff erent positionspositions along along the the joint joint and and far far enough enough to reach reach the the initial initial wall’s wall’s tube tube thickness. thickness.
Figure 4. The AlSi10Mg and AA6060-T6 MPW-welded specimen (sample No. 1): (a) and (b) isometric Figure 4. The AlSi10Mg and AA6060-T6 MPW-welded specimen (sample No. 1): (a) and (b) isometric viewview of the of specimenthe specimen after after peel peel test; test; (c ()c) top top viewview of the the welded welded specimen. specimen. 2.3. Material Characterisation 2.3. Material Characterisation (a) (a)Metallographic Metallographic samples samples of of the the welded welded parts were were prepared prepared according according to the to ASTM-E3 the ASTM-E3 standard. standard. TheThe surface surface preparation preparation included included sectioning,sectioning, grinding, grinding, and and fine fine polishing polishing steps steps (up to (up 0.05 to µm). 0.05 µm). TheThe microstructure microstructure was was etched etched with with aa modifiedmodified Flick’s Flick’s reagent reagent (10% (10% HF-90% HF-90% H2O). H 2O). (b) (b)Light Light microscopy microscopy (LM) (LM) examination examination was carriedwas carried out with out a Zeiss,with Aalen,a Zeiss, Germany, Aalen, metallographicGermany, microscope,metallographic in order microscope, to characterise in order the to microstructure characterise the of microstructure the base materials of the andbase the materials joint, as and well as to observethe joint, and as well estimate as to observe flyer tube and deformation. estimate flyer tube deformation. (c) Vickers microhardness (HV) indentation tests were performed with a Buehler MMT-7 tester (c) Vickers(Bühler, microhardness Uzwil, Switzerland), (HV) indentation using a 100 g tests load were and 10 performed s. duration with load, a in Buehler order to MMT-7 estimate tester (Bühler,joint Uzwil,mechanical Switzerland), properties. using a 100 g load and 10 s. duration load, in order to estimate joint (d)mechanical Scanning properties. electron microscopy (SEM) and energy dispersive spectroscopy (EDS) were conducted (d) Scanningwith micro electron FA Quanta microscopy 200FEI (SEM) SEM instrument and energy and dispersive an environmental spectroscopy SEM (Quanta (EDS) were 200FEG conducted E- withSEM, micro Thermo FA Quanta Fisher 200FEI Scientific, SEM instrumentWaltham, Ma andssachusetts, an environmental United SEMStates) (Quanta in high 200FEG vacuum E-SEM, Thermocondition Fisher and Scientific, a secondary Waltham, electron Massachusetts, detector. The UnitedSEM-EDS States) analysis in highwas vacuumperformed condition to andcharacterise a secondary the electron microstructure detector. and Thecomposition SEM-EDS of the analysis base materials was performed and the joint to characterise and detect the discontinuities, such as molten “pockets” and cracks. microstructure and composition of the base materials and the joint and detect discontinuities, 3.such Results as molten and Discussion “pockets” and cracks.
3. Results3.1. Leak and and Discussion Peel Tests
3.1. Leak3.1.1. and Leak Peel Tests Tests
3.1.1. LeakSelected Tests MP-welded samples were subjected to fine leak testing to determine the hermeticity of the joints. Typical results of He leak tests (see Section 2.2 (b) above) were as follows: 5·× 10−9 std-cc·s−1 HeSelected (specimen MP-welded No. 6) and samples 1·× 10 were−9 std-cc·s subjected−1 He to(specimen fine leak No. testing 4). Characteristically, to determine the aerospace hermeticity of the joints.applications Typical are results required of Heto have leak a tests leak (see rate Sectionbetter than 2.2 (b)5·× above)10−6 std-cc·s were−1 He, as follows: in order 5to protect10 9 std-cc the s 1 × − · − He (specimenequipment No.from 6)contaminants. and 1 10 Both9 std-cc weldeds 1samplesHe (specimen have a leak No. rate 4). better Characteristically, than 5·× 10−9 std-cc·s aerospace−1 × − · − applicationsHe, more are than required meeting toaerospace have a leakhermeticity rate better specifications. than 5 10The6 leakstd-cc testss demonstrate1 He, in order that, to with protect a the × − · −
Metals 2020, 10, 1235 9 of 20 equipment from contaminants. Both welded samples have a leak rate better than 5 10 9 std-cc s 1 × − · − He, more than meeting aerospace hermeticity specifications. The leak tests demonstrate that, with a correct set of process parameters, it is feasible to produce hermetically sealed assemblies based on additively manufactured components welded to wrought parts by MPW.
3.1.2. Peel Tests The mechanical macro-quality of the joints was assessed by manual peel test, as previously performed by others [12–14,43]. The outcome at the coil’s slot (0◦) is of special interest due to the locally reduced magnetic field intensity in the area, leading to possible local discontinuities and low mechanical properties (Figure4). Strips of the flyer material with a width of approximately 12 mm were axially cut and radially bent. The welded specimen was clamped, and a tension force was applied to the strips normal to the weld seam. In the case of a defective weld, the strips would separate from the stationary part. When adequate joints were obtained, the weld seam is either able to withstand the load, or the strip is expected to fail in the base material. Four peel tests were carried out on sample 1 (Figure4), and all strips failed in the base material, with resulting strip lengths of 0 ◦—9 mm, 90◦—9 mm, 180◦—10.5 mm, and 270◦—9.5 mm. In MPW, the effect of the coil slot is such that magnetic pressure varies in different positions around the coil. In general, magnetic pressure, and thus, flyer velocity is highest at 180 degrees from the slot, and so the weld length is expected to be highest at the 180◦ position and lowest at the slot (0◦). This is borne out by the above-presented results. The fact that the results of 0◦ and the 90◦ are identical, maybe due to a slight mechanical asymmetry of the parts or stray magnetism.
3.2. Flyer Tube Deformation Typical interface morphology of welded AlSi10Mg AM L-PBF rods to AA6060-T6 wrought tubes is presented in Figure5a–c (LM observation). The interface is relatively flat at the beginning of the joint and then becomes gradually wavy in the middle, before becoming flat near the end of the joint. This gradual change in the interface morphology was typical for all successful AM/wrought joints (Figure5). On selected samples, the total joint length was measured and compared to the wavy zone length (Table3). It is shown that the higher the applied energy, the greater is the joint length. Higher energy produces a higher flyer velocity. Therefore, the weld length along the interface, where the minimum flyer velocity to provide adequate welding conditions is at least maintained (as it is falling off), is increased, thus producing a longer weld.
Table 3. Total MPW average joint lengths, including the planar and wavy regions.
Specimen No. Measured Distance 4 (High Energy) 5 (Optimal Energy) Joint length (µm) 8029 5789 Wavy length (µm) 6049 4945 Collapsed flyer thicknesses of welds (µm) 1775 1827
In case of MPW of specimen No. 8, where the AM-AlSi10Mg material served as flyer and AA6060-T6 alloy as stationary part, no weld was obtained. Although acceptable deformation occurred in both components and the typical flat and wavy faying interfaces were identified on both the flyer and the stationary parts, the weld failed (Figure6, LM observation). Weak partial welding may have occurred, but the parts were probably separated by reverberating shock waves and the joint failed. Metals 2020, 10, x FOR PEER REVIEW 9 of 21
correct set of process parameters, it is feasible to produce hermetically sealed assemblies based on additively manufactured components welded to wrought parts by MPW.
3.1.2. Peel Tests The mechanical macro-quality of the joints was assessed by manual peel test, as previously performed by others [12–14,43]. The outcome at the coil’s slot (0°) is of special interest due to the locally reduced magnetic field intensity in the area, leading to possible local discontinuities and low mechanical properties (Figure 4). Strips of the flyer material with a width of approximately 12 mm were axially cut and radially bent. The welded specimen was clamped, and a tension force was applied to the strips normal to the weld seam. In the case of a defective weld, the strips would separate from the stationary part. When adequate joints were obtained, the weld seam is either able to withstand the load, or the strip is expected to fail in the base material. Four peel tests were carried out on sample 1 (Figure 4), and all strips failed in the base material, with resulting strip lengths of 0°—9 mm, 90°—9 mm, 180°—10.5 mm, and 270°—9.5 mm. In MPW, the effect of the coil slot is such that magnetic pressure varies in different positions around the coil. In general, magnetic pressure, and thus, flyer velocity is highest at 180 degrees from the slot, and so the weld length is expected to be highest at the 180° position and lowest at the slot (0°). This is borne out by the above-presented results. The fact that the results of 0° and the 90° are identical, maybe due to a slight mechanical asymmetry of the parts or stray magnetism.
3.2. Flyer Tube Deformation Typical interface morphology of welded AlSi10Mg AM L-PBF rods to AA6060-T6 wrought tubes is presented in Figure 5a–c (LM observation). The interface is relatively flat at the beginning of the joint and then becomes gradually wavy in the middle, before becoming flat near the end of the joint. This gradual change in the interface morphology was typical for all successful AM/wrought joints (Figure 5). On selected samples, the total joint length was measured and compared to the wavy zone length (Table 3). It is shown that the higher the applied energy, the greater is the joint length. Higher energy produces a higher flyer velocity. Therefore, the weld length along the interface, where the Metals 2020minimum, 10, 1235 flyer velocity to provide adequate welding conditions is at least maintained (as it is falling 10 of 20 off), is increased, thus producing a longer weld.
Metals 2020, 10, x FOR PEER REVIEW 10 of 21
practically similar results for specimens 4, 5, 6; (c) general view of specimen No. 6 interface morphology; the weld includes a wavy region and two flat areas, before and after the wavy zone.
Table 3. Total MPW average joint lengths, including the planar and wavy regions.
Specimen No. Measured Distance 4 (High Energy) 5 (Optimal Energy) Joint length (µm) 8029 5789 Wavy length (µm) 6049 4945 Collapsed flyer thicknesses of welds (µm) 1775 1827
In case of MPW of specimen No. 8, where the AM-AlSi10Mg material served as flyer and AA6060-T6 alloy as stationary part, no weld was obtained. Although acceptable deformation Figure 5. Joint cross-section: (a) radial deformation behaviour of the flyer: thickening effect of the flyer Figureoccurred 5. Joint in both cross-section: components ( aand) radial the typical deformation flat and behaviourwavy faying of interfaces the flyer: were thickening identified e ffonect both of the tube; indentation depth of the flyer into the stationary part is ~0.3 mm; (b) local wall thickness; flyerthe tube; flyer indentationand the stationary depth parts, of the the flyer weld into failed thestationary (Figure 6, LM part observation). is ~0.3 mm; Weak (b) local partial wall welding thickness; practicallymay have similar occurred, results but for the specimens parts were 4, 5,probably 6; (c) general separated view by of specimenreverberating No. shock 6 interface waves morphology; and the thejoint weld failed. includes a wavy region and two flat areas, before and after the wavy zone.
Figure 6.FigureWavy 6. Wavy surfaces surfaces appearance appearance of of No. No. 88 MPWMPW spec specimen,imen, which which underwent underwent normal normal deformation, deformation, but no weld was obtained: (a) L-PBF flyer tube; (b) wrought material stationary part. The waves are but no weld was obtained: (a) L-PBF flyer tube; (b) wrought material stationary part. The waves are smaller at the beginning, increasing in weld direction. smaller at the beginning, increasing in weld direction. The overall radial deformation of the specimens and the tube-wall thickening at the end of MPW Theare overall exhibited radial in Figures deformation 3 and 5. of Specimens the specimens 4 and 5 and display the tube-wallsimilar performance thickening with at no the visible end of MPW are exhibitedexternal in cracks Figures or discontinuities3 and5. Specimens in the collapsed 4 and 5 displayregions of similar both flyers. performance The change with in the no flyer’s visible wall external cracks orthickness, discontinuities measured in in the the collapsedwelding direction regions (specimen of both flyers.No. 5), Thewas assessed change infrom the the flyer’s initial wallimpact thickness, measuredzone, in along the welding the weld direction zone and (specimenthe inclined No.tube’s 5), region, wasassessed up to the fromundefo thermed initial original impact tube. zone, An along the weldexample zone andof the the thickness inclined evolution tube’s region, of the flyer up to wall the is undeformed presented in originalFigure 5a,b. tube. The An initial example wall of the thickness of 1.5 mm slightly decreases adjacent to the free flyer edge and then gradually increases, thickness evolution of the flyer wall is presented in Figure5a,b. The initial wall thickness of 1.5 mm reaching ~1.8 mm in most of the weld zone (Figure 5b). Under the reasonable assumption that almost slightlyno decreases material was adjacent lost (during to the jetting), free flyer the edgeexperi andmental then values gradually agree well increases, with the reachingcalculated ~1.8and mm in most ofexpected the weld thickening zone (Figure of5 b).the Under tube’s the wall, reasonable when assumptiontaking into thataccount almost that no materialthe major was lost (duringcompression/deformation jetting), the experimental has valuesoccurred agree in the well fully with collapsed the calculated region of ~10.5 and expectedmm. The thickening of the tube’s wall,effect when of the takingflyer’s intowall accountis clearly thatdisplayed the major in Figure compression 5, from an/ deformationinitial wall thickness has occurred of 1.5 mm, in the fully through 1.65 mm and 1.75 mm, reaching a maximum value of 1.8 mm in most of the weld zone. A collapsed region of ~10.5 mm. The thickening effect of the flyer’s wall is clearly displayed in Figure5, summary of the walls’ average thicknesses, measured at two or four locations, is shown in Table 3. from anThe initial average wall indentation thickness ofdepth 1.5 mm,of the through flyer into 1.65 the mm stationary and 1.75 part mm, was reaching ~0.3 mm a(Figure maximum 5a,b). value of 1.8 mmComparing in most of the the indentation weld zone. data A and summary bearing in of mind the that walls’ different average applied thicknesses, energy values measured were used at two or four locations,in the welding is shown experiments, in Table3 .no The major average differences indentation were identified. depth of theThis flyer is a intopositive the stationaryfinding, part was ~0.3demonstrating mm (Figure 5thea,b). low Comparing sensitivity of the the indentation joint quality datato the and welding bearing parameters. in mind Furthermore, that di fferent the applied energygeometrical values were design used chosen in the for welding the parts experiments,was found stable no and major can be di well-consideredfferences were for identified. scaling-up This is and even industrialisation of the MPW process for joining the current AM alloys to wrought alloys. a positive finding, demonstrating the low sensitivity of the joint quality to the welding parameters. The indentation depth of the flyer into the stationary part is very much determined by the mechanical
Metals 2020, 10, 1235 11 of 20
Furthermore, the geometrical design chosen for the parts was found stable and can be well-considered for scaling-up and even industrialisation of the MPW process for joining the current AM alloys to wrought alloys. The indentation depth of the flyer into the stationary part is very much determined by the mechanical properties of the stationary component material. The minor indentation of the flyer intoMetals the 2020 AM, 10, x L-PBF FOR PEER material, REVIEW having relatively high mechanical properties when11 compared of 21 to the wrought material, is an advantage in favour of the hybrid assemblies. The thickening effect of the properties of the stationary component material. The minor indentation of the flyer into the AM L- flyer’s wall, accompanied by a minor indentation in the stationary part, lowers the overall dimensional PBF material, having relatively high mechanical properties when compared to the wrought material, changesis of an the advantage as-welded in favour assembly. of the hybrid assemblies. The thickening effect of the flyer’s wall, accompanied by a minor indentation in the stationary part, lowers the overall dimensional changes 3.3. Microhardnessof the as-welded Across assembly. the Interface In order to evaluate the local changes adjacent to the interface, LM and SEM analyses were 3.3. Microhardness Across the Interface supported by Vickers microhardness tests (Figure7a, SEM observation). The microhardness In order to evaluate the local changes adjacent to the interface, LM and SEM analyses were measurements were conducted in several locations along the joint and were performed perpendicularly supported by Vickers microhardness tests (Figure 7a, SEM observation). The microhardness to the weldmeasurements line, crossing were theconducted interface in fromseveral AlSi10Mg-BM locations along to the AA6060-T6 joint and BM.were Typical performed results are presentedperpendicularly in Figure7, to where the weld location line, crossing “0” representsthe interface from the jointAlSi10Mg-BM interface to AA6060-T6 and negative BM. Typical locations are situatedresults in the are AM-AlSi10Mg presented in material.Figure 7, where Only location a slight change“0” represents in microhardness the joint interface was detectedand negative adjacent to the jointlocations interface, are with situated a smaller in the increaseAM-AlSi10Mg observed material. in the Only AA6060-T6 a slight change material in microhardness (Figure7b,c). was The small increasedetected in microhardness adjacent to the can joint be interface, attributed with to a the smaller deformation increase observed and local in the heating AA6060-T6 of the material alloys during (Figure 7b,c). The small increase in microhardness can be attributed to the deformation and local the impact welding. heating of the alloys during the impact welding.
Figure 7.FigureVickers 7. Vickers microhardness microhardness across across thethe weld interface: interface: (a) SEM (a) SEM image image of the indentation of the indentation mark at mark at the interface;the interface; (b) ( highb) high energy energy (specimen (specimen No. 4); 4); (c ()c )optimal optimal energy energy (specimen (specimen No. 5); No. “0” 5);location “0” location marks themarks joint the interface. joint interface.
Metals 2020, 10, 1235 12 of 20
3.4. Microanalysis of the Joint Metals 2020, 10, x FOR PEER REVIEW 12 of 21
3.4.1.3.4. Interfacial Microanalysis Layers of the or Joint “Pockets” All faying interfaces show a metal continuity that can either be flat or wavy; the interface is 3.4.1. Interfacial Layers or “Pockets” initially relatively flat, and waves form and grow in the welding direction. According to Zhang [43], the wavyAll morphology faying interfaces of the show weld a rises meta inl continuity the intimate that contact can either zone be andflat or helps wavy; interlocking the interface between is the twoinitially surfaces, relatively creating flat, and much waves more form robust and grow joints. in the Ben-Artzy welding direction. et al. [16 According] described to Zhang that the [43], wavy interfacethe wavy of the morphology weld is typically of the weld formed rises in in a the periodic intimate manner contact with zone defined and helps wavelength interlocking and between amplitude. the two surfaces, creating much more robust joints. Ben-Artzy et al. [16] described that the wavy Sridharan et al. [17] studied the joint in Al and Fe welds using a vaporising foil actuator welding interface of the weld is typically formed in a periodic manner with defined wavelength and (VFAW) system. They presented a pronounced hierarchical nature at the interface. A wavy interface amplitude. Sridharan et al. [17] studied the joint in Al and Fe welds using a vaporising foil actuator was found,welding as (VFAW) the result system. of high They plastic presented deformation. a pronounced A hierarchical SEM study nature indicated at the the interface. development A wavy of a liquidinterface layer at was the interface,found, as alsothe result as mentioned of high inplastic previous deformation. studies [A7 ,14SEM,22 ]study andin indicated ISO/AWI the 23,598 (2019)development [42]. Most researchers of a liquid layer believe at the that interface, the interfacial also as mentioned waviness isin producedprevious studies by the [7,14,22] Kelvin–Helmholtz and in instabilityISO/AWI mechanism, 23,598 (2019) i.e., [42]. the Most reflected researchers shock believe waves th interactat the interfacial with the waviness collision is point produced at the by interface, the whereKelvin–Helmholtz interferences are instability the source mechanism, for wave initiation. i.e., the reflected Collision shock energy waves and interact joint geometrywith the collision are believed to havepoint the at most the interface, significant where influence interferences on the are interfacial the source wave for morphology;wave initiation. high Collision energy energy use resultsand in long wavelengthjoint geometry and are high believed wave amplitude.to have the most significant influence on the interfacial wave morphology; high energy use results in long wavelength and high wave amplitude. Figure8a,b illustrate the evolution of the wave area and amplitude as a function of weld length Figure 8a,b illustrate the evolution of the wave area and amplitude as a function of weld length for thefor two the two applied applied energies. energies. High-energy High-energy welds welds present larger larger waves waves than than the the optimal-energy optimal-energy welds welds (Figures(Figures8 and 89 and). In 9). Figure In Figure9, the 9, wavelengththe wavelength is plottedis plotted against against thethe weld length; length; the the figure figure shows shows that that the wavelengththe wavelength increases increases with with increasing increasing weld weld energy. energy. The The optimal-energy optimal-energy welds welds present present more more waves, i.e., awaves, shorter i.e., wavelength. a shorter wavelength. The results The are results in accord are within accord the previous with the experimental previous experimental observations, as reportedobservations, by Ben-Artzy as reported et al.by [Ben-Artzy16]. et al. [16].
FigureFigure 8. Evolution 8. Evolution of waveof wave area area and and amplitude: amplitude: ( a)) Wave Wave area area distribution distribution of high-energy of high-energy (specimen (specimen No. 4)No. and 4) optimal-energyand optimal-energy (specimen (specimen No. No. 5) welds5) welds along along the the joint. joint. Wave Wave area area waswas measuredmeasured as the area underarea the under wave; the (b )wave; wave (b amplitude) wave amplitude distribution distribution for high-energy for high-energy (specimen (specimen No. No. 4) and4) and optimal-energy optimal- (specimen No. 5) welds along the joint. Wave amplitude was measured peak-to-peak (error 5 µm). ± “0” indicates the beginning of the wavy region. MetalsMetals 20202020,, 1010,, xx FORFOR PEERPEER REVIEWREVIEW 1313 ofof 2121
energyenergy (specimen(specimen No.No. 5)5) weldswelds alongalong thethe joint.joint. WavWavee amplitudeamplitude waswas measuredmeasured peak-to-peakpeak-to-peak (error(error ±± Metals 2020, 10, 1235 13 of 20 55 µm).µm). “0”“0” indicatesindicates thethe begibeginningnning ofof thethe wavywavy region.region.
Figure 9. Development of wavelength in the weld direction for optimal- and high-energy joints FigureFigure 9.9. DevelopmentDevelopment ofof wavelengthwavelength inin thethe weldweld directiondirection forfor optimal-optimal- andand high-energyhigh-energy jointsjoints (error(error (error 5 µm). ±± 55 µm).µm).± Along the wavy joints, “pockets” or occasional discontinuous short layers, exhibiting different AlongAlong thethe wavywavy joints,joints, “pockets”“pockets” oror occasionaloccasional discontinuousdiscontinuous shortshort layers,layers, exhibitingexhibiting differentdifferent morphologies, were detected and analysed. Typically, the optimal-energy weld presents fewer morphologies,morphologies, werewere detecteddetected andand analysed.analysed. TypicaTypically,lly, thethe optimal-energyoptimal-energy weldweld presentspresents fewerfewer “pockets” than high-energy welds, as demonstrated in Figure 10a,b. A comparison was made between “pockets”“pockets” thanthan high-energyhigh-energy welds,welds, asas demonstrateddemonstrated inin FigureFigure 10a,b.10a,b. AA comparisoncomparison waswas mademade high-energy welds, (specimen No. 4) and “optimal-energy” welds (specimen No. 6) and the results betweenbetween high-energyhigh-energy welds,welds, (specimen(specimen No.No. 4)4) andand “optimal-energy”“optimal-energy” weldswelds (specimen(specimen No.No. 6)6) andand thethe are reflected in Figures 11 and 12. No detectable heat-affected zone was observed adjacent to the resultsresults areare reflectedreflected inin FiguresFigures 1111 andand 12.12. NoNo detectabledetectable heat-affectedheat-affected zonezone waswas observedobserved adjacentadjacent toto faying interfaces of both high-energy welds and optimal-energy welds. Cracks and/or pores were thethe fayingfaying interfacesinterfaces ofof bothboth high-energyhigh-energy weldswelds anandd optimal-energyoptimal-energy welds.welds. CracksCracks and/orand/or porespores werewere occasionally found (Figures 10b and 12b, LM and SEM observation, respectively) in the “pockets” of occasionallyoccasionally foundfound (Figures(Figures 10b10b andand 12b,12b, LMLM andand SEMSEM observation,observation, respectively)respectively) inin thethe “pockets”“pockets” ofof high-energy welds, while the optimal-energy welds (specimen No. 6) presented no crack comparable high-energyhigh-energy welds,welds, whilewhile thethe optimal-energyoptimal-energy weldswelds (specimen(specimen No.No. 6)6) presentedpresented nono crackcrack comparablecomparable “pockets”. This phenomenon was also reported previously, probably when employing suboptimal “pockets”.“pockets”. ThisThis phenomenonphenomenon waswas alsoalso reportedreported previously,previously, probablyprobably whenwhen employingemploying suboptimalsuboptimal welding parameters, by the same method described in Shribman et al. [14]. weldingwelding parameters,parameters, byby thethe samesame methodmethod describeddescribed inin ShribmanShribman etet al.al. [14].[14].
FigureFigure 10.10. 10. WavyWavyWavy interfaceinterface interface SEMSEM SEM micrographs:micrographs: micrographs: ((aa)) AnAn ( aoptimal-energyoptimal-energy) An optimal-energy weldweld (specimen(specimen weld (specimen No.No. 5);5); ((bb) No.) aa high-high- 5); energy(energyb) a high-energy weldweld (specimen(specimen weld (specimenNo.No. 4).4). L-PBFL-PBF No. build-orientationbuild-orientation 4). L-PBF build-orientation isis oppositeopposite istoto opposite thethe weldweld to direction.direction. the weld TheThe direction. whitewhite arrowsThearrows white showshow arrows thethe meltedmelted show the pockets.pockets. melted pockets.
Element maps were acquired from the polished cross-sections adjacent to the interface (Figure 11a, SEM) showing the spatial distribution of Al, Si, and Mg (Figure 11b–d, EDS analysis). The element maps were found to be beneficial for displaying element distributions and compositional zonation in the regions where “pockets” or discontinuous short layers were detected. Si “depletion” is evident (marked with arrows in Figure 11), when compared to the AM-AlSi10Mg BM composition, proving that local melting has occurred during the welding process.
Metals 2020, 10, 1235 14 of 20 Metals 2020, 10, x FOR PEER REVIEW 14 of 21
Figure 11. SEM–energySEM–energy dispersive dispersive spectroscopy spectroscopy (EDS) (EDS) element element mapping mapping at the at theMPW MPW interface: interface: (a) (generala) general SEM SEM view, view, (b) ( bAl,) Al, (c) ( cSi,) Si, and and (d ()d Mg.) Mg. molten molten “pockets” “pockets” are are visible visible in in Figure Figure 11c11c and and marked marked by arrows.
Figure 12. SEMSEM micrographs micrographs of of selected selected “pockets” “pockets” found found adjacent adjacent to to the the wavy wavy interface interface (high-energy (high-energy weld specimen No. No. 4): 4): (a ()a area) area 5, 5,and and (b) ( bpoints) points 6, 7 6, designate 7 designate the thelocations locations of EDS of EDSmicroanalysis. microanalysis. The whiteThe white arrows arrows in Figure in Figure 12b indicate12b indicate local local interface interface cracking. cracking. 3.4.2. EDS of Interfacial Layers or “Pockets” Element maps were acquired from the polished cross-sections adjacent to the interface (Figure 11a, EDSSEM) microanalysis showing the acrossspatial the distribution peak and valleyof Al, ofSi, the and waves, Mg (Figure within “pockets”11b–d, EDS (specimen analysis). No. The 6) elementand/or intermediate maps were layersfound (specimen to be beneficial No. 4) wasfor di performedsplaying toelement determine distributions (Figure 12 and) the compositional local chemical zonation in the regions where “pockets” or discontinuous short layers were detected. Si “depletion”
Metals 2020, 10, 1235 15 of 20 composition of selected regions; the results are shown in Table4. In both specimens, the Si content (point 5 in Figure 12a and points 6 and 7 in Figure 12b) is roughly half the Si content in the AM-AlSi10Mg BM. At these locations, layers from both base metals melt and mix, and the resolidified zones present an average Si concentration of 5.2 wt%, roughly averaging the Si content (~0.5 wt% and ~10 wt%) of the base metals. These findings may lead to the conclusion that the “pockets” consist of a resolidified molten mix of base metals in roughly equal quantities of both alloys, as suggested previously [14]. Some cracking has been occasionally noticed along the wavy joint, with an example illustrated in Figure 12b. Local cracking was reported in the literature, but on a much larger scale, probably related to the inherent properties of the welded alloys and the applied welding parameters (ISO/AWI 23598, 2019).
Table 4. SEM-EDS microanalysis (wt%) of typical “pockets” found adjacent to the wavy interface of MP welds (see points 5, 6, 7 in Figure 12).
Measurement Locations, (wt%) Element 5 6 7 AA6060-BM AM-BM Mg 0.1 0.1 0.1 0.2 0.2 Al 94.2 94.2 95.2 99.8 89.3 Si 5.7 5.7 4.7 – 10.6
3.5. Jet Characterisation In previous works done by Stern et al. [9] and Shribman et al. [14], jet residues located towards the end of the weld zone were observed and examined. The agglomeration of the ejected remains from the interface near the end zone of the weld confirm a temperature rise that exceeds the melting temperature of the alloys. In the current study, a few local accumulations of Si particles (93.7 wt%) were also detected between the jet residues. The SEM-EDS microanalyses of the debris indicate that the elemental composition values are intermediate between the compositions of the two base metals; the jet contains approximately 7.5 wt% Si and some traces of Mg. Further investigations of jet residue located near the end of the weld zone and on the inner flyer surfaces were conducted. Residue in the form of faded black areas and lines was clearly observed, visually, on the interior flyer surface (Figures4b and 13) . Those faded findings were ejected during the welding process and deposited on the tube interior surface as the black films (Figures4b and 13). The most noticeable black area is located adjacent to the coil slot (coil origin), as a result of higher material ejection from this specific location, possibly due to decreased energy facing the slot and, thus, reduced impact and collision point velocity, enabling more time for jet production (see arrows in Figure 13a and enlarged in Figure 13b). The EDS-SEM analysis could not identify any differences in the chemical composition, probably because of the jet film being extremely thin. In the present study, some jet residue in the form of a very thin metal film (less than 0.1µm) was discovered in between the welded parts (Figure 14a–d, SEM observation); the element distribution in selected regions is displayed in Table5. This type of thin metal film jet, discovered for the first time in this work, is probably connected to the fluid metal jetting phenomena. During the MPW process, high impact energies are concentrated in a small material volume and generate a considerable local increase in temperature, leading to melting or even vaporisation of thin superficial layers from both colliding parts. The solidification of liquefied/vaporised jet material probably formed this type of thin metal foil. The thin film phenomenon might be attributed to the weld process with the AM-AlSi10Mg alloy, which has a lower melting point compared to the wrought AA6060 alloy (~577 ◦C, ~650 ◦C respectively). Nevertheless, the phenomenon was only detected when high-energy welding conditions were utilised and was not reported earlier, when welding “conventional” alloys. Fluid metal jetting, leaving behind elevated melting temperature particles, is also the reason for the formation of tiny Si agglomerates within the bonding zone, as shown in Figure 14d and Table5. It is worth noting that jet residue, in the form of thin films, was only observed in the samples welded using high-energy parameters. Figure 15 shows an AM local breakage that occurred after the weld, away from the end of the weld. Metals 2020, 10, 1235 16 of 20
It is noteworthy that these phenomena are different from the observation in the previous work done by Shribman et al. [14], probably due to the optimal-energy weld specimen No. 6, in contrast to high-energy weld specimen No. 4 at this point. The compositions of jet residue were investigated by SEM-EDS spot analyses, and the results are listed in Table5. The jet fragments contain alloying elements found in both components (Si, Mg), including small amounts of S, Fe, and Cu. Copper’s source found in some locations is uncertain. The copper contamination may come from Cu-containing parts of the MPW apparatus, such as the coil. Areas 36–38 (white dash rectangular) were examined by
EDS (TableMetals 20205)., 10, x FOR PEER REVIEW 16 of 21
FigureFigure 13. Jet 13. residueJet residue found found on on the the inner inner surface surface of the flyer flyer tube: tube: (a () aGeneral) General view; view; (b) zooming (b) zooming at jet at jet residueresidue origin. origin. The The arrows arrows refer refer to to the the parallel parallel locationlocation of of the the coil coil origin, origin, showing showing a local a local discontinuity discontinuity and a significant jetting source. andMetals a 2020 significant, 10, x FOR jetting PEER REVIEW source. 17 of 21
In the present study, some jet residue in the form of a very thin metal film (less than 0.1µm) was discovered in between the welded parts (Figure 14a–d, SEM observation); the element distribution in selected regions is displayed in Table 5. This type of thin metal film jet, discovered for the first time in this work, is probably connected to the fluid metal jetting phenomena. During the MPW process, high impact energies are concentrated in a small material volume and generate a considerable local increase in temperature, leading to melting or even vaporisation of thin superficial layers from both colliding parts. The solidification of liquefied/vaporised jet material probably formed this type of thin metal foil. The thin film phenomenon might be attributed to the weld process with the AM-AlSi10Mg alloy, which has a lower melting point compared to the wrought AA6060 alloy (~577 °C, ~650 °C respectively). Nevertheless, the phenomenon was only detected when high-energy welding conditions were utilised and was not reported earlier, when welding “conventional” alloys. Fluid metal jetting, leaving behind elevated melting temperature particles, is also the reason for the formation of tiny Si agglomerates within the bonding zone, as shown in Figure 14d and Table 5. It is worth noting that jet residue, in the form of thin films, was only observed in the samples welded using high-energy parameters. Figure 15 shows an AM local breakage that occurred after the weld, away from the end of the weld. It is noteworthy that these phenomena are different from the observation in the previous work done by Shribman et al. [14], probably due to the optimal-energy weld specimen No. 6, in contrast to high-energy weld specimen No. 4 at this point. The compositions of jet residue were investigated by SEM-EDS spot analyses, and the results are listed in Table 5. The jet fragments contain alloying elements found in both components (Si, Mg), including small amounts of S, Fe, and Cu. Copper’s source found in some locations is uncertain. The copper contamination may come from Cu-containing parts of the MPW apparatus, such as the coil. Areas 36–38 (white dash rectangular) were examined by EDS (Table 5).
FigureFigure 14. SEM14. SEM micrographs micrographs of of typicaltypical jet jet thin thin films films foun foundd close close to the to end the of end the weld of the (specimen weld (specimen No. No. 4),4), showingshowing the locations examined examined by by EDS EDS (Table (Table 5):5 ):(a) ( pointsa) points 27–31; 27–31; (b) points (b) points 41–43; 41–43; (c) point (c) point33; 33; (d) points(d) points 34, 35.34, 35.
Table 5. Chemical composition by SEM-EDS (wt%) in several locations of specimen No. 4; composition of both base metals is shown in Table 1.
Measurement Locations, (wt%) Element 27 28 29 30 31 33 34 35 36 37 38 42 43 44 Mg 0 0.2 0.1 0 0.1 0.2 0.3 0.3 0.3 0.1 0.1 0.2 0.4 0.13 Al 74.9 92.4 94.3 86.5 84.2 95.2 85.3 96.0 99.4 99.0 89.3 89.4 96.4 94.7 Si 4.5 2.7 5.5 8.1 3.8 4.6 6.2 3.7 0.3 0.9 10.6 8.8 3.3 4.1
S 7.1 – – – – – 3.0 – – – – – – – Cu 13.6 4.8 – 4.9 – – 5.2 – – – – 1.6 – 1.1 Fe – – – 1.5 11.9 – – – – – – – – –
Figure 15. Local breakage adjacent to unwelded interface, aside from the joint region.
Metals 2020, 10, x FOR PEER REVIEW 17 of 21
Figure 14. SEM micrographs of typical jet thin films found close to the end of the weld (specimen No. Metals4),2020 showing, 10, 1235 the locations examined by EDS (Table 5): (a) points 27–31; (b) points 41–43; (c) point 33;17 of 20 (d) points 34, 35.
Table 5.5.Chemical Chemical composition composition by SEM-EDSby SEM-EDS (wt%) (wt%) in several in locationsseveral locations of specimen of No.specimen 4; composition No. 4; compositionof both base metalsof both isbase shown metals in Table is shown1. in Table 1.
MeasurementMeasurement Locations, Locations, (wt%) (wt%) ElementElement 27 2728 2829 29 30 30 31 31 33 33 34 34 35 35 36 36 37 37 38 38 42 42 43 43 44 44 MgMg 0 0 0.2 0.2 0.1 0.1 0 0 0.1 0.1 0.2 0.2 0.3 0.3 0. 0.33 0.3 0.3 0.1 0.1 0.1 0.1 0.2 0.2 0.4 0.4 0.13 0.13 AlAl 74.9 74.9 92.4 92.4 94.3 94.3 86.5 86.5 84.2 84.2 95.2 95.2 85. 85.33 96.0 96.0 99.4 99.4 99.0 99.0 89.3 89.3 89.4 89.4 96.4 96.4 94.7 94.7 SiSi 4.5 4.5 2.7 2.7 5.5 5.5 8.1 8.1 3.8 3.8 4.6 4.6 6. 6.22 3.7 3.7 0.3 0.3 0.9 0.9 10.6 10.6 8.8 8.8 3.3 3.3 4.1 4.1 SS 7.1 7.1 – –– – – – – –– –3.0 3.0 –– – –– –– –– –– –– – CuCu 13.6 13.6 4.8 4.8 – – 4.9 4.9 – – – –5.2 5.2 –– – – – – – –1.6 1.6 – –1.1 1.1 FeFe – –– –– –1.5 1.5 11.9 11.9 – –– ––– – –– –– –– –– –– –
FigureFigure 15. 15. LocalLocal breakage breakage adjacent adjacent to to unwelded unwelded interface, interface, aside aside from the joint region.
4. Conclusions L-PBF of metals is an AM expanding technology that, up to this point, is hindered by the significant limitations of the printed component size. This novel contribution addresses PBF machine size limitations by developing a high-quality joining method between L-PBF and conventionally fabricated materials. The current study is part of continuing research dealing with joining of AM to wrought material and/or AM to AM. This novel contribution is mainly focused on the analyses of both optimal- and high-energy welding conditions. The current study demonstrates that it is possible to hermetically join L-PBF additive manufactured components to wrought parts by the MPW technique when using an adjusted set of process parameters. The use of L-PBF AlSi10Mg eutectic material—having a lower melting temperature than the wrought material, while exhibiting relatively high mechanical properties—enables the formation of high-quality welds at lower MPW energies. Thus, the loading on the MPW tool coils provides substantially reduced lifetime effects. The macro and microstructural evolution of the joint shows an increase in amplitude, area, and wavelength of the wavy interfaces, for both optimal- and high-energy joints, as the weld propagates. The joint formation is accompanied by flyer wall thickening, the result of the significant diametral inward deformation of the tube occurring during the welding process. The wall thickening and the wavy interface development are believed to be physical indicators of process parameters optimisation, as they contribute to the mechanical properties of the joint. Microhardness indentation tests performed perpendicularly to the weld line, crossing the interface between AlSi10Mg rod and AA6060-T6 wrought tube, revealed that the microhardness has increased slightly, adjacent to the interface. This phenomenon is probably due to high-velocity impact occurring in MPW, causing major deformation of the flyer tube, and in the current case, minor deformation of the stationary part. Metals 2020, 10, 1235 18 of 20
Due to the excessive kinetic energy derived from the impact, localised interfacial melting occurs; resolidified regions in the form of singular “pockets” have been observed. In the high-energy MP welds, discontinuous resolidified layers have been also detected. The EDS-SEM microanalysis results suggest that superficial layers from both base metal’s melt and mix during welding, forming the resolidified regions. Jet residue showing irregular morphologies was detected, occasionally in form of small conglomerates containing Si particles (93.7 wt%) and as very thin films (less than 100 nm thick) located in between the components. High-energy MPW, leading to a significant temperature increase, may cause the emission of the vaporised jets creating discrete Si particles and thin films. It is worth noting that this type of jet residue was discovered for the first time in the current research. The study shows that magnetic pulse welding provides metallurgical joints with a very small environmental footprint, offering high-quality bonding of dissimilar AlSi10Mg AM L-PBF rod and AA6060-T6 wrought tube, while exhibiting a negligible heat-affected zone. The study provides additional information concerning weld formation mechanisms of additively manufactured parts to conventional wrought parts by MPW, which conform to existing weld knowledge. Although the AM-to-wrought MP-weld mechanism looks similar to the known MPW mechanism, it is a note-worthy outcome that AM alloys can be readily MP-welded. The “thin film” phenomenon needs to be explored further.
Author Contributions: Conceptualization, M.N., V.S. and A.S.; methodology, M.N., V.S. and A.S.; validation, M.N., V.S., D.A. and A.S.; formal analysis, M.N., V.S., D.A. and A.S.; investigation, M.N. and S.L.; resources, M.N., V.S., D.A. and A.S.; data curation, M.N., V.S.; writing—original draft preparation, M.N., V.S., D.A. and A.S.; writing—review and editing, M.N., V.S., D.A. and A.S.; visualization, M.N. and D.A.; supervision, A.S.; project administration, D.A.; All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Acknowledgments: The authors are grateful to thank Y. Sharon of “Sharon Tuvia (1982)” Ltd., Ness Tziona, Israel, for supplying the AM L-PBF samples. The authors would also like to thank I. Benishti of NRCN, E. Millionshckik of the Ben-Gurion University of the Negev, O. Atiya formerly of Bmax Srl. and Z. Barkay of the Wolfson Applied Materials Research Centre, Tel Aviv University, for their valuable technical contributions. Conflicts of Interest: The authors declare no conflict of interest.
References
1. Bataev, I.A.; Ivanov, I.V.; Malyutina, Y.N.; Emurlaev, K.I.; Emurlaeva, Y.Y. High plastic deformations and high cooling rates at the interface of explosion-welded materials. Met. Sci. Heat Treat. 2019, 60, 674–679. [CrossRef] 2. Lee, T.; Zhang, S.; Vivek, A.; Daehn, G.; Kinsey, B. Wave formation in impact welding: Study of the Cu–Ti system. CIRP Ann. 2019, 68, 261–264. [CrossRef] 3. Nassiri, A.; Abke, T.; Daehn, G. Investigation of melting phenomena in solid-state welding processes. Scr. Mater. 2019, 168, 61–66. [CrossRef] 4. Carlone, P.; Astarita, A. Dissimilar metal welding. Metals 2019, 9, 1206. [CrossRef] 5. Raoelison, R.N.; Sapanathan, T.; Buiron, N.M.; Rachik, M. Magnetic pulse welding of Al/Al and Al/Cu metal pairs: Consequences of the dissimilar combination on the interfacial behaviour during the welding process. J. Manuf. Process. 2015, 20, 112–127. [CrossRef] 6. Stern, A.; Aizenshtein, M. Bonding zone formation in magnetic pulse welds. Sci. Tech. Weld. Join. 2002, 7, 339–342. [CrossRef] 7. Stern, A.; Shribman, V.; Ben-Artzy, A.; Aizenshtein, M. Interface phenomena and bonding mechanism in magnetic pulse welding. J. Mater. Eng. Perform. 2014, 23, 3449–3458. [CrossRef] 8. Kakizaki, S.; Watanabe, M.; Kumai, S. Simulation and experimental analysis of metal jet emission and weld interface morphology in impact welding. Mater. Trans. 2011, 152, 1104181357. [CrossRef] 9. Stern, A.; Becher, O.; Nahmany, M.; Ashkenazi, D.; Shribman, V. Jet composition in magnetic pulse welding: Al–Al and Al-Mg couples. Weld. J. 2015, 94, 258s–264s. Metals 2020, 10, 1235 19 of 20
10. Wang, X.; Zheng, Y.; Liu, H.; Shen, Z.; Hu, Y.; Li, W.; Gao, Y.; Guo, C. Numerical study of the mechanism of explosive/impact welding using smoothed particle hydrodynamics method. Mater. Des. 2012, 35, 210–219. [CrossRef] 11. Bellmann, J.; Beyer, E.; Lueg-Althoff, J.; Gies, S.; Tekkaya, A.E.; Schulze, S. Measurement of collision conditions in magnetic pulse welding processes. J. Phys. Sci. Appl. 2017, 7, 1–10. [CrossRef] 12. Lueg-Althoff, J.; Bellmann, J.; Gies, S.; Schulze, S.; Tekkaya, A.E.; Beyer, E. Influence of the flyer kinetics on magnetic pulse welding of tubes. J. Mater. Process. Tech. 2018, 262, 189–203. [CrossRef] 13. Pourabbas, M.; Abdollah-Zadeh, A.; Sarvari, M.; Pouranvari, M.; Miresmaeili, R. Investigation of structural and mechanical properties of magnetic pulse welded dissimilar aluminum alloys. J. Manuf. Processes 2019, 37, 292–304. [CrossRef] 14. Shribman, V.; Nahmany, M.; Levi, S.; Atiya, O.; Ashkenazi, D.; Stern, A. MP welding of dissimilar materials: AM laser powder-bed fusion AlSi10Mg to wrought AA6060-T6. Prog. Addit. Manuf. 2019, 5, 71–181. [CrossRef] 15. Debroy, T.; Wei, H.L.; Zuback, J.S.; Mukherjee, T.; Elmer, J.W.; Milewski, J.O.; Beese, A.M.; Wilson-Heid, A.; De, A.; Zhang, W. Additive manufacturing of metallic components—Process, structure, and properties. Prog. Mater. Sci. 2018, 92, 112–224. [CrossRef] 16. Ben-Artzy, A.; Stern, A.; Frage, N.; Shribman, V.; Sadot, O. Wave formation mechanism in magnetic pulse welding. Int. J. Impact Eng. 2010, 37, 397–404. [CrossRef] 17. Sridharan, N.; Poplawsky, J.; Vivek, A.; Bhattacharya, A.; Guo, W.; Meyer, H.; Mao, Y.; Lee, T.; Daehn, G. Cascading microstructures in aluminum-steel interfaces created by impact welding. Mater. Character. 2019, 151, 119–128. [CrossRef] 18. Bellmann, J.; Lueg-Althoff, J.; Schulze, S.; Hahn, M.; Gies, S.; Beyer, E.; Tekkaya, A.E. Thermal effects in dissimilar magnetic pulse welding. Metals 2019, 9, 348. [CrossRef] 19. Böhme, M.; Sharafiev, S.; Schumacher, E.; Böhm, S.; Wagner, M.F.X. On the microstructure and the origin of intermetallic phase seams in magnetic pulse welding of aluminum and steel. Mater. Werkst. 2019, 50, 958–964. [CrossRef] 20. Geng, H.; Mao, J.; Zhang, X.; Li, G.; Cui, J. Formation mechanism of transition zone and amorphous structure in magnetic pulse welded Al-Fe joint. Mater. Lett. 2019, 245, 151–154. [CrossRef] 21. Sapanathan, T.; Yang, K.; Chernikov, D.; Raoelison, R.N.; Gluschenkov, V.; Buiron, N.; Rachik, M. Thermal effect during electromagnetic pulse welding process. In Materials Science Forum; Trans Tech Publications: Stafa-Zurich, Switzerland, 2017; Volume 879, pp. 1662–1667. 22. Sharafiev, S.; Pabst, C.; Wagner, M.F.; Groche, P. Microstructural characterization of interfaces in magnetic pulse welded aluminum/aluminum joints. In Proceedings of the IOP Conference Series: Materials Science and Engineering, Rourkela, Bangalore, India, 14–16 July 2016; IOP Publishing: Bristol, UK, 2016; Volume 118, p. 012016. 23. Aboulkhair, N.T.; Simonelli, M.; Parry, L.; Ashcroft, I.; Tuck, C.; Hague, R. 3D printing of aluminium alloys: Additive manufacturing of aluminium alloys using selective laser melting. Prog. Mater. Sci. 2019, 106, 100578. [CrossRef] 24. Zhang, D.; Sun, S.; Qiu, D.; Gibson, M.A.; Dargusch, M.S.; Brandt, M.; Qian, M.; Easton, M. Metal alloys for fusion based additive manufacturing. Adv. Eng. Mater. 2018, 20, 1700952. [CrossRef] 25. Zuback, J.S.; DebRoy, T. The hardness of additively manufactured alloys. Materials 2018, 11, 2070. [CrossRef] [PubMed] 26. Mertens, A.I.; Delahaye, J.; Lecomte-Beckers, J. Fusion-based additive manufacturing for processing aluminum alloys: State-of the-art and challenges. Adv. Eng. Mater. 2017, 19, 1700003. [CrossRef] 27. Rosenthal, I.; Stern, A.; Frage, N. Strain rate sensitivity and fracture mechanism of AlSi10Mg parts produced by selective laser melting. Mater. Sci. Eng. A 2017, 682, 509–517. [CrossRef] 28. Awd, M.; Siddique, S.; Johannsen, J.; Emmelmann, C.; Walther, F. Very high-cycle fatigue properties and microstructural damage mechanisms of selective laser melted AlSi10Mg alloy. Int. J. Fatigue 2019, 124, 55–69. [CrossRef] 29. Rosenthal, I.; Shneck, R.; Stern, A. Heat treatment effect on the mechanical properties and fracture mechanism in AlSi10Mg fabricated by additive manufacturing selective laser melting process. Mater. Sci. Eng. A 2018, 729, 310–322. [CrossRef] Metals 2020, 10, 1235 20 of 20
30. Tradowsky, U.; White, J.; Ward, R.M.; Read, N.; Reimers, W.; Attallah, M.M. Selective laser melting of AlSi10Mg: Influence of post-processing on the microstructural and tensile properties development. Mater. Des. 2016, 105, 212–222. [CrossRef] 31. Biffi, C.A.; Fiocchi, J.; Tuissi, A. Laser weldability of AlSi10Mg alloy produced by selective laser melting: Microstructure and mechanical behavior. J. Mater. Eng. Perform. 2019, 28, 1–6. [CrossRef] 32. Nahmany, M.; Hadad, Y.; Aghion, E.; Stern, A.; Frage, N. Microstructural assessment and mechanical properties of electron beam welding of AlSi10Mg specimens fabricated by selective laser melting. J. Mater. Process. Tech. 2019, 270, 228–240. [CrossRef] 33. Zhang, C.; Bao, Y.; Zhu, H.; Nie, X.; Zhang, W.; Zhang, S.; Zeng, X. A comparison between laser and TIG welding of selective laser melted AlSi10Mg. Optic. Laser Tech. 2019, 120, 105696. [CrossRef] 34. Nahmany, M.; Stern, A.; Benishti, I.; Rosenthal, I.; Frage, N. Electron beam welding of AlSil0Mg workpieces produced by selected laser melting additive manufacturing technology. Addit. Manuf. 2015, 8, 63–70. 35. Nahmany, M.; Stern, A.; Aghion, E.; Frage, N. Structural properties of EB-welded AlSi10Mg thin-walled pressure vessels produced by AM-SLM technology. J. Mater. Eng. Perform. 2017, 26, 4813–4821. [CrossRef] 36. Tavlovich, B.; Shirizly, A.; Katz, R. EBW and LBW of additive manufactured Ti6Al4V products. Weld. J. 2018, 97, 179S–190S. 37. Wits, W.W.; Becker, J.J. Laser beam welding of titanium additive manufactured parts. Procedia CIRP 2015, 28, 70–75. [CrossRef] 38. Adamczyk-Cie´slak,B.; Mizera, J.; Kurzydłowski, K.J. Microstructures in the 6060 aluminium alloy after various severe plastic deformation treatments. Mater. Character. 2011, 62, 327–332. [CrossRef] 39. Inberg, A.; Ashkenazi, D.; Kimmel, G.; Shacham-Diamand, Y.; Stern, A. Gold–silver electroless plating on laser powder-bed fusion additively printed AlSi10Mg parts. Metals 2020, 10, 557. [CrossRef] 40. Inberg, A.; Ashkenazi, D.; Kimmel, G.; Shacham-Diamand, Y.; Stern, A. Gold plating of AlSi10Mg parts produced by a laser powder-bed fusion additive manufacturing technique. Prog. Addit. Manuf. 2020, 1–10. [CrossRef] 41. DIN EN. 1779:1999-1. Non-destructive testing Leak testing-Criteria for the method and technique selection. In Din 1779, New European Standard for the Selection of A Suitable Method for Leak Detection and Leak Tightness Testing; Beuth: Berlin, Germany, 2011. 42. ISO/AWI 23598 (under development). Mechanical Joining—Destructive Testing of Joints—Specimen Dimensions and Test Procedure for Mechanized Peel Testing of Single Joints; ISO: Geneva, Switzerland, 2019. 43. Zhang, Y. Investigation of Magnetic Pulse Welding on Lap Joint of Similar and Dissimilar Materials. Ph.D. Thesis, The Ohio State University, Columbus, OH, USA, 2010.
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).