Experimental and Numerical Study on Microstructure and Mechanical Properties of Ti-6Al-4V/Al-1060 Explosive Welding

Article Experimental and Numerical Study on Microstructure and Mechanical Properties of Ti-6Al-4V/Al-1060 Explosive Welding

Yasir Mahmood, Kaida Dai * , Pengwan Chen *, Qiang Zhou, Ashfaq Ahmad Bhatti and Ali Arab

State Key Laboratory of Explosion Science and Technology, Beijing Institute of Technology, Beijing 100081, China; [email protected] (Y.M.); [email protected] (Q.Z.); [email protected] (A.A.B.); [email protected] (A.A.) * Correspondence: [email protected] (K.D.); [email protected] (P.C.); Tel.: +86-010-6891-8740 (K.D. & P.C.)

Received: 15 October 2019; Accepted: 2 November 2019; Published: 5 November 2019

Abstract: The aim of this paper is to study the microstructure and mechanical properties of the Ti6Al4V/Al-1060 plate by explosive welding before and after heat treatment. The welded interface is smooth and straight without any jet trapping. The disturbances near the interface, circular and random pores of Al-1060, and beta phase grains of Ti6Al4V have been observed by Scanning electron microscopy (SEM). Heat treatment reduces pores significantly and generates a titanium-island-like morphology. Energy dispersive spectroscopy (EDS) analysis results show that the maximum portion of the interfacial zone existed in the aluminium side, which is composed of three intermetallic phases: TiAl, TiAl2 and TiAl3. Heat treatment resulted in the enlargement of the interfacial zone and conversion of intermentallic phases. Tensile test, shear test, bending test and hardness test were performed to examine the mechanical properties including welding joint qualities. The results of mechanical tests show that the tensile strength and welding joint strength of the interfacial region are larger than one of its constituent material (Al-1060), the microhardness near the interface is maximum. Besides, tensile strength, shear strength and microhardness of heat treated samples are smaller than unheat treated. Smooth particle hydrodynamic (SPH) method is used to simulate the transient behaviour of both materials at the interface. Transient pressure, plastic deformation and temperature on the flyer and base side during the welding process were obtained and analyzed. Furthermore, the numerical simulation identified that almost straight bonding structure is formed on the interface, which is in agreement with experimental observation.

Keywords: explosive welding; Ti6Al4V/Al-1060; microstructure; mechanical properties; smooth particle hydrodynamic (SPH)

1. Introduction Welding is a useful technique for joining two similar or different materials. Nowadays, various welding techniques i.e., gas welding, arc welding, laser welding, friction welding, explosive welding etc. have been widely used in many industries. The explosive welding is a solid state welding process in which a flyer plate is accelerated by an explosive and welded with other material in a short interval of time. During the welding process, a high velocity jet removes the impurities on the material surfaces [1]. Explosive welding was used in many mechanical related industries, especially in power plants and aerospace industry. Moreover, it was successfully applied to produce biocompatible materials in the medical related fields [2,3]. This method is advantageous to weld different kinds of metals and alloys that cannot easily be welded by some other means of welding. Materials having excellent mechanical properties (i.e., corrosive resistance, high strength to density ratio, good conductance and

Metals 2019, 9, 1189; doi:10.3390/met9111189 www.mdpi.com/journal/metals Metals 2019, 9, 1189 2 of 17 heat resistance, etc.) can be welded with materials having low mechanical properties. This type of welding makes the materials more reliable and cost effective. Both aluminium and titanium are widely used in the transportation and aerospace industries. Al-1060 belongs to commercially pure aluminium. It is highly malleable and corrosion resistant. However, it is very reactive with air at high temperature and has low mechanical strength. Titanium and its alloys have the highest strength to mass ratio in all naturally existing materials. So their combination can be used as an excellent industrial application. Explosive welding is one of the best choice for bonding these two metal alloys because the two materials have a big difference in their melting points, and mechanical strengths [4]. Furthermore, this combination has been received special attention due to the formation of titanium aluminides at the interface. Titanium aluminides have low density, good oxidation ignition resistance and excellent mechanical properties at high temperatures. Due to these features titanium aluminides get special attention and researchers practised different techniques to develop titanium aluminides. Previously, Perusko et al. [5] investigated the intermetallic formation during Al/Ti welding by using Ar+ ion implementation. Adeli et al. [6] used induction preheating process to synthesis the Ti/Al powder and prepared TiAl phase. Arakawa et al. [7] formed titanium aluminide foam with the help of self-propagating high-temperature synthesis (SHS). They created a foam which contains 60% to 70% porosity due to TiAl3. Titanium aluminides percentage is increased by using diffusion and combustion [8]. DC magnetron sputtering method was opted by Ramos et al. [9] to prepare γ-TiAl. Furthermore, they controlled activation energy by silver foil. Romankov et al. [10] prepared TiAl3 intermetallic by using thermal deposition technique. Kahraman et al. [11] studied the complex microstructure of explosively welded Ti6Al4V and aluminium alloy using different charge to mass ratios. They found that hardness and corrosion were increased at the interface with the increase of the charge to mass ratios. E et al. [12] evaluated the tensile properties of the explosively welded Ti/Al sample with a load applied along parallel and perpendicular directions to the interface. Xia et al. [13] observed the micro grains and recrystallisation of titanium alloy in the interface. Fronczek et al. [14] and Bazarnik et al. [15] studied mechanical properties and microstructure of the explosively welded titanium alloy with aluminium alloy. Additionally, the Ti/Al interface study is essential to check the quality of welded joints. According to Ege et al. [16], at the Ti/Al interface, the aluminium percentage content reinforces the Ti/Al joints, providing stability and increasing the strength up to 825 MPa. Inal et al. [17] found that the straight interface is more suitable than the wavy interface because it accompanied by excessive heat generation and could produce weak and brittle intermetallics. Ege et al. [18] presented that the heat treatment of the Ti/Al welded sample did not affect the stability, therefore could be used for high temperature environment. Multilayered combination for explosive welding of titanium and aluminium is one of the best options to fabricate titanium aluminides. Lazurenko et al. [19] welded forty layers and Mali et al. [20] joined 23 of Ti-Al by using explosive welding technique. They produced TiAl3 stable intermetallics by using sintering at 640 ◦C temperature and 3MPa pressure. Furthermore, Bataev et al. [21] studied the thickness of TiAl3 from top to bottom plates and compared the results with post heat treated samples. Likewise, Foaden et al. [22] analysed the TiAl3 formation process after annealing of the explosively welded composite plate. Fan at al. [23] prepared multilayered TiAl foils by using vacuum hot pressing and studied the foil thickness effect. Similarly, Fan et al. [24] welded 5 Ti-Al plates by using an explosive welding technique to analyse the microstructure and mechanical properties of composite plates. Simulation is an excellent tool to predict the explosive welding conditions, i.e., impact velocity, pressure, plastic strain, etc. Many authors applied various approaches to simulate explosive welding phenomenon. Recently, Mousavi et al. [25] used ANSYS AUTODYN to simulate the straight, wavy, jetting morphology and humps formation during the welding process. Nassiri et al. [26] replicated the jetting phenomena with the help of SPH method, and exercised Arbitrary Lagrangian-Eulerian (ALE) method to obtain other welding parameters, i.e., interface shape, temperature, the velocity of material, shear stress. Wang et al. [27] imitated the shear stress and effective plastic strain of materials by using Metals 2019, 9, x FOR PEER REVIEW 3 of 17 Metals 2019, 9, 1189 3 of 17 by using the SPH approach. Wang et al. [28] reproduced the whole explosive welding phenomena withthe the SPH help approach. of material Wang point et al. method [28] reproduced (MPM) in the C++ whole program. explosive welding phenomena with the help of materialIn this paper, point microstructure method (MPM) and in Cmechanical++ program. properties of the welded composite plate (Ti6Al4V and AlIn-1060) this paper, were microstructure investigated. and The mechanical SPH approach properties with of the welded help of composite LS-DYNA plate was (Ti6Al4V used toand understandAl-1060) were the investigated.welding conditions The SPH i.e., approach interface with morphology, the help of transient LS-DYNA pressure, was used temperature to understand and the plasticwelding deformation. conditions i.e., interface morphology, transient pressure, temperature and plastic deformation.

2. 2.Experimental Experimental Procedure Procedure

2.1.2.1. Materials Materials and and Explosive Explosive Welding Welding Setup Setup Ti6Al4VTi6Al4V was was used used as as a flyer a flyer plate plate and and Al Al-1060-1060 as as a base a base plate plate because because titanium titanium had had lower lower thermal thermal diffusivity than aluminium [29]. Both plates with the size of 200 mm 150 mm 3 mm were welded diffusivity than aluminium [29]. Both plates with the size of 200 mm ×× 150mm × 3× mm were welded inin air. air. The The schematic schematic experimental experimental setup setup is isshown shown in in Fig Figureure 1.1 .Both Both plates plates were were arranged arranged in in parallel parallel configurationconfiguration with with a standoff a stando ffdistancedistance of of 5 5mm. mm. Ammonium Ammonium nitrate nitrate and and fuel fuel oil oil (ANFO) (ANFO) with with 30 30 mm mm 3 thicknessthickness (packing (packing density density is is 670 670 Kg/m Kg/m3 andand detonation detonation velocity isis 26002600 mm/s)/s) was was used used to to accelerate accelerate the theflyer flyer plate. plate. The The sand sand was was employed employed as as an an anvil. anvil. The The welded welded samples samples were were heat heat treated treated at at 525 525◦C °C for for4 h4 andh and then then cooled cooled in in open open air. air. Inal Inal et al.et [al.17 ][17] reported reported that that the bondthe bond strength strength remains remains constant constant at this atheating this heating temperature temperature and timeand time duration. duration.

Figure 1. Schematic diagram of explosive welding. Figure 1. Schematic diagram of explosive welding. 2.2. Microstructure Test 2.2. Microstructure Test For microstructure analysis, samples were cut from the middle of the welded plate approximately 100For mm microstructureaway from the detonating analysis, point. samples The sampleswere cut were from mounted the middle in an epoxy of such the thatwelded the welding plate approximatelyinterface was parallel100 mm to away the detonation from the direction.detonating The point. samples The weresamples cleaned were with mounted sandpapers in an (Gritepoxy 320, such1000, that 1500, the 2000),welding and interfac then polishede was parallel with ceramic to the detonation powder. Finally, direction. the The samples samples were were soaked cleaned in the withKroll sandpapers reagent (distilled (Grit 320, water 1000, 92 1500, mL, nitric 2000), acid and 6 mL,then hydrofluoricpolished with acid ceramic 2 mL) forpowder. 20 s to Finally, examine the the samplesapparent were microstructure. soaked in the Kroll reagent (distilled water 92 mL, nitric acid 6 mL, hydrofluoric acid 2 mL) forThe 20 microstructure s to examine the and apparent elements microstructure. distribution at the interface was observed by using scanning electronThe microstructure microscope (SEM, and elements Hitachi S-4800, distribution Tokyo, at Japan)the interface at an accelerationwas observed voltage by using of 15scanning kV. SEM electronelemental microscope scan images (SEM were, Hitachi further S-4800, edited Tokyo, to combineJapan) at bothan acceleration elements (Ti-Al)voltage using of 15 CoralDrawkV. SEM elemental(Coral Corporation, scan images Ottawa, were further ON, Canada).edited to combine both elements (Ti-Al) using CoralDraw (Coral Corporation, Ottawa, ON, Canada) 2.3. Mechanical Test 2.3. MechanicalThe Vickers Test hardness tests were conducted along the perpendicular direction of the interface. TheThe hardness Vickers tests hardness were carried tests were out onconducted a microhardness along the machine perpendicular using a 0.98direction N load of forthe 15 interface. s. The hardnForess mechanical tests were tests, carried stratified out on samples a microhardness were obtained machine from theusing welded a 0.98 plate N load before for and15 s. after heat treatment.For mechanical Tensile test tests, and stratified shear test samples samples were (as seen obtained in Figure from2a,b) the were welded performed plate onbefore material and testingafter heatmachine treatment. (MTS) Tensile with atest loading and shear rate test of 5 samples mm/min (as and seen 1.8 in mm Figure/min 2a,b) respectively were performed to measure on thematerial tensile testingproperties machine of the (MTS) welded with portion. a loading For rate three of 5 point mm/min bending and 1.8 test, mm/min the sample respectively with a thickness to measure of 5 the mm tensileand a properties length of 100of the mm welded was conducted portion. For along three the point perpendicular bending test, direction the sample of the with flyer-base a thickness interface. of 5 mm and a length of 100 mm was conducted along the perpendicular direction of the flyer-base interface.

Metals 20192019,, 9,, x 1189 FOR PEER REVIEW 44 of 17 17

Figure 2. SchematicSchematic d diagramiagram for for ( (aa)) tensile tensile test test ( (bb)) shear shear tensile tensile test test.. 3. Simulation of Explosive Welding 3. Simulation of Explosive Welding ANFO has non-ideal detonation behavior [30]. It is difficult to simulate it with Jones-Wilkins-Lee ANFO has non-ideal detonation behavior [30]. It is difficult to simulate it with Jones-Wilkins- (JWL) equation because it has a long reaction zone, so energy and momentum at Von-Neuman Spike Lee (JWL) equation because it has a long reaction zone, so energy and momentum at Von-Neuman cannot be neglected [31]. Therefore empirical formula was used to calculate the plate velocity and Spike cannot be neglected [31]. Therefore empirical formula was used to calculate the plate velocity impact angle. and impact angle. 3.1. Calculation of Plate Velocity 3.1. Calculation of Plate Velocity Deribas et al. [32] and Manikandan et al. [33] developed a distance dependent empirical relation to calculateDeribas the et al. impact [32] and angle Manikandan between flyer et al. and [33] base developed plate. The a distance impact angledependent is defined empirical as, relation to calculate the impact angle between flyer and base plate. The impact angle is defined as,  r   k + 1 π R β(rad) =  1 (1)  푘k+ 11 − 휋2 R + A푅 + Bte 훽(푟푎푑) = (√ − − 1) S (1) 푘 − 1 2 퐵푡푒 (푅 + 퐴 + ⁄푆) where A = 2.71, B = 0.184, te is the thickness of explosive, S is the standoff distance between the flyer and base plate. Polytropic exponent k can be measured with the by Gurney velocity [32]. where A = 2.71, B = 0.184, 푡푒 is the thickness of explosive, 푆 is the standoff distance between the flyer and base plate. Polytropic exponent k can bes measured with the by Gurney velocity [32]. D2 k = + 1 (2) 퐷22 Vg 푘 = √ 2 + 1 (2) 푉푔 where Vg represents the Gurney velocity, which can be calculated by using the following formula [34–36]. Where Vg represents the Gurney velocity, which can be calculated by using the following formula [34–36]. D √2E(m/s) = 600 + 0.52 p (3) γ + 1 퐷 where γ is a ratio of specific heat constant.√2퐸(푚/ For푠) = ANFO,600 + the0.52 value of γ is 2.881 [37]. Plate velocity can(3) be √훾 + 1 calculated by Crossland relation [38]. Where γ is a ratio of specific heat constant. For ANFO,! the value of 훾 is 2.881 [37]. Plate β velocity can be calculated by Crossland relationV = [38].2Dsin (4) p 2

Using the above equations for the current experimental훽 conditions, plate velocity Vp with 5 mm 푉푝 = 2퐷 푠𝑖푛 ( ) (4) standoﬀ distance is 707 m/s and the impact angle β is 15.042 ◦.

3.2. EquationUsing the of above State and equations Constitutive for the Model current experimental conditions, plate velocity Vp with 5 mm standoffMei distance Gruneisen is 707 equation m/s and of the state impact is mostly angle used β is for 15.04°. shock wave propagation [27]. This equation gives us a relation between pressure and volume under shock conditions at a given temperature. 3.2. Equation of State and Constitutive Model Johnson-Cook material model was used for the simulation of explosive welding, because it can Mei Gruneisen equation of state is mostly used for shock wave propagation [27]. This equation gives us a relation between pressure and volume under shock conditions at a given temperature. Johnson-Cook material model was used for the simulation of explosive welding, because it can

Metals 2019, 9, x FOR PEER REVIEW 5 of 17

Metalssuccessfully2019, 9, 1189 predict the high deformation and Von-Misses yield stress of the material. Johnson-Cook5 of 17 equation can be written as

푛 ∗푚 successfully predict the high deformation휎 = (퐴 + and퐵휀 ) Von-Misses(1 + 퐶 푙푛휀푝̇ )( yield1 − 푇 stress) of the material. Johnson-Cook(5) equationwhere 휀 canis plastic be written strain, as C is strain rate constant, 휀̇ is plastic strain rate, n is hardening exponent, n 푝 . m σ = (A + Bε ) 1 + C lnεp (1 T∗ ) (5) A is yield strength of the material, m is softening exponent, B −is strain hardening coefficient and T* is . where ε is plastic strain, C is strain rate constant,∗ (푇 − 푇ε푟표표푚is) plastic strain rate, n is hardening exponent, homologous temperature and equal to 푇 = p ⁄(푇 − 푇 ). The parameters of material A is yield strength of the material, m is softening exponent,푚푒푙푡B is strain푟표표푚 hardening coeﬃcient and T* models for Ti6Al4V and Al-1060 are listed in Table 1. (T Troom) is homologous temperature and equal to T = − . The parameters of material models for ∗ (T Troom) melt − Ti6Al4V and Al-1060 areTable listed 1. Parameters in Table1. of material model and equation of state.

Table 1. Parameters of material model andDensity equation of state. Material A (MPa) B (MPa) N C M Co (m/s) G S (Kg/m3) Ti-6Al-4V [39] 1098 A 1092B 0.93 0.014 1.1 Density4430 Co 5130 1.23 1.028 Material N CM 3 GS Al-1060 [40] 66.56(MPa) 108.8(MPa) 0.23 0.029 1 (Kg2707/m ) (m/s)5386 1.97 1.339 Ti-6Al-4V [39] 1098 1092 0.93 0.014 1.1 4430 5130 1.23 1.028 3.3. SmoothAl-1060 Particle [40] Hydrodynamics 66.56 108.8 (SPH) 0.23 0.029 1 2707 5386 1.97 1.339 Numerical gridless lagrangian hydrodynamics simulation was carried out with the SPH method 3.3. Smooth Particle Hydrodynamics (SPH) in ANSYS/LS-DYNA (Ansys 14.5, LSTC, Livermore, CA, USA). In SPH method, Kernel approximation of fieldNumerical variable gridlessat given lagrangian points is applied hydrodynamics to simplify simulation the conservation was carried equations. out with Discrete the SPH methodparticle inhelps ANSYS to find/LS-DYNA out field (Ansys information 14.5, LSTC, while Livermore, neighboring CA, USA). particles In SPH are usedmethod, to solveKernel the approximation integrals. If ofneighboring field variable par atticles given arepoints expressed is applied by to simplify subscript the j conservation, then field equations. variable for Discrete non particle zero Kernel helps toapproximation find out field can information be described while as neighboring follow: particles are used to solve the integrals. If neighboring particles are expressed by subscript j, then field variable for non zero Kernel approximation can be 푚푝 described as follow: 푓(푟) ≅ ∑ 푓푗푊(|푟 − 푟푗| , ℎ|푋|, ℎ) (6) X휌m푝p f (r) fjW( r rj , h X , h) (6) ρp | − | | | Where mp,ρp are particle mass and density respectively, f(r) is a field variable, r represents the where mp, ρp are particle mass and density respectively, f (r) is a field variable, r represents the location of particle, h shows the length at which particle is effected by the neighboring particles. For location of particle, h shows the length at which particle is effected by the neighboring particles. For the current simulation of the explosive welding, it is simplified with the collision of the flyer and base the current simulation of the explosive welding, it is simplified with the collision of the flyer and plate at a certain velocity and angle. These two parameters are calculated by Equation (1) and base plate at a certain velocity and angle. These two parameters are calculated by Equation (1) and Equation (4), respectively. The schematic model for SPH simulation is shown in Figure 3, with a Equation (4), respectively. The schematic model for SPH simulation is shown in Figure3, with a particle size of 0.5 μm. particle size of 0.5 µm.

Figure 3. Model for SPH simulation. Figure 3. Model for SPH simulation. 4. Results and Discussion 4. Results and Discussion 4.1. Microstructure of Welding Interface before Heat Treatment 4.1. MicrostructureThe SEM results of Welding of the welded Interface sample before Heat before Treatment heat treatment are displayed in Figure4. It shows that weldingThe SEM interface results isof smooththe welded and ﬂatsample without before delamination. heat treatment This are type displayed of smooth in Ti Figure/Al interface 4. It show joints patternthat welding was previously interface is observed smooth byandBazarnik flat without et al. [delamination.15] for Ti6Al4V This/Al2519 type welding of smooth and Ti/AlEge etinterface al. [16] for Ti6Al4V/Al6061 multilayer welding.

Metals 2019, 9, x FOR PEER REVIEW 6 of 17

Metalsjoint2019 ,pattern9, 1189 was previously observed by Bazarnik et al. [15] for Ti6Al4V/Al2519 welding and Ege et6 of 17 al. [16] for Ti6Al4V/Al6061 multilayer welding.

FigureFigure 4. SEM 4. SEM image image of of explosively explosively weldedTi6Al4V weldedTi6Al4V/Al1060/Al1060 interface interface before before heat heat treatment treatment at at magnificationsmagnifications of ( aof) 500(a) 500×,(b, )(b 1000) 1000×,(, c(c)) 20002000×,,( (dd) )5000×. 5000 . × × × × SeeingSeeing from from Figure Fig4a–d,ure 4a both–d, metals both metals have somehave somedisturbances disturbances in microstructures in microstructures near nearthe interface, the interface, especially aluminium side shows the maximum disturbances. High pressure and especially aluminium side shows the maximum disturbances. High pressure and deformation during deformation during explosive welding process may generate recrystallization of microstructure. explosive welding process may generate recrystallization of microstructure. These phenomena near the These phenomena near the interface formed under high pressure were previously observed by Zhang interfaceet al. formed [41] and under Gloc et high al. [42]. pressure Furthermore, were previously Figure 4d observedshows that by AlZhang-1060 contains et al. [41 some] and circularGloc et and al. [42]. Furthermore,random pores. Figure These4d shows pores thatare formed Al-1060 during contains severe some deformatio circularn and and random high temperature pores. These gradient, pores are formedcausing during rupture severe of deformationlocal surfaces. andRaoelison high temperatureet al. [43] analyzed gradient, the formation causing rupture of pores of during local surfaces.the Raoelisonwelding et al.process [43] analyzedand stated the that formation these pores of might pores be during the origin the weldingof the crack process propagation. and stated The same that these porespores might were be observed the origin by of Su the et crackal. [44] propagation.during the welding The sameprocess pores of Fe were-Al. As observed shown in by Figure Su et 4d, al. [44] duringTi6Al4V the welding microstructure process of near Fe-Al. the Asinterface shown is in also Figure altered,4d, Ti6Al4V especially microstructure beta phase grains near thethat interface are is alsoelongated altered, toward especially the betadetonation phase direction. grains that During are elongated the explosive toward welding the process, detonation due direction.to high strain During and pressure, the temperature of the interfacial zone is abruptly raised. This temperature increment the explosive welding process, due to high strain and pressure, the temperature of the interfacial zone is high enough to change the phase of the plate [25]. Therefore, Ti6Al4V is most likely to be converted is abruptlyto the beta raised. phase This during temperature the collisi incrementon. However, is highthe cooling enough rate to is change very high the in phase this ofprocess, the plate the [25]. Therefore,titanium Ti6Al4V phase change is most process likely toonly be takes converted a short to interval the beta and phase quickly during turns theback collision. to the alpha However, phase the cooling[38]. rate According is very high to Tomashchuk in this process, et al. the [45] titanium the titanium phase beta change phase process is more only likely takes to a react short wi intervalth and quicklyaluminium turns than back alpha to the phase alpha to phase form titanium [38]. According aluminide. to Tomashchuk The intense plastic et al. [ deformation45] the titanium and beta phaseelongation is more likely were toalso react observed with aluminiumby Murr et al. than [46] alpha and Kacar phase et toal. form[47]. titanium aluminide. The intense plastic deformationThe interface and between elongation titanium were and also aluminium observedhas by Murra very et complicatedal. [46] and Kacar structure, et al. so [ 47 EDS]. Thetechnique interface is used between to understand titanium and the microstructure. aluminiumhas Figure a very 5acomplicated shows an elemental structure, scan so EDS near technique the is usedinterface to understand region, indicating the microstructure. that the Ti6Al4V/Al Figure5-1060a shows interface an elemental is smooth scan and nearflat. Figure the interface 5b reveals region, indicating that the Ti6Al4V/Al-1060 interface is smooth and flat. Figure5b reveals that aluminium element counts in the base plate start decreasing about 0.86 µm away from the interface. This decrease in aluminium counts continues in flyer plate and gets normal counts after 0.24 µm thickness from the interface. So the total width of the interfacial zone is 1.1 µm with maximum portion existing in the Metals 2019, 9, x FOR PEER REVIEW 7 of 17 thatMetals aluminium2019, 9, 1189 element counts in the base plate start decreasing about 0.86 μm away from7 the of 17 interface. This decrease in aluminium counts continues in flyer plate and gets normal counts after 0.24 μm thickness from the interface. So the total width of the interfacial zone is 1.1 μm with maximumbase side (aluminium). portion existing Slope in di thefference base side shows (aluminium). that aluminium Slope counts difference near interface shows that decline aluminium faster in countsthe base near plate interface as compared decline to faster the flyerin the plate base side. plate as compared to the flyer plate side.

(a) (b)

FigureFigure 5. 5. (a)( aElemental) Elemental scan scan of aluminium of aluminium and and titanium titanium elements elements along along with with welding welding interface, interface, (b) graphical(b) graphical representation representation of linear of linear element element analysis analysis at the at the interface, interface, (c) ( c EDS) EDS point point spectrum spectrum at at the the interface,interface, (d (d) )Ti Ti-Al-Al atomic atomic distribution distribution near near interface interface at at selected selected points points..

ParasithiParasithi et et al. al. [48] [48 ]reported reported that that materials materials with with highe higherr thermal thermal conductivity conductivity keep keep a a maximum maximum percentagepercentage in in interfacial interfacial area. area. Since Since aluminium aluminium has has almost almost more more than than 25 25 times times higher higher thermal thermal conductivityconductivity as as compared compared to to titanium, titanium, that that is is why why maximum maximum microstructural microstructural changes changes are are observed observed inin the the base base plate. plate. Manikandan Manikandan et et al. al. [33] [33 ] gave gave another another demonstration demonstration about about the the interfacial interfacial zone. zone. AccordingAccording to to their their study, study, at at the the interface, interface, the the atomic atomic concentration concentration of of base base material material is is relatively relatively higher higher thanthan the the flyer flyer material. material. DueDue to to the the extreme extreme conditions conditions in in explosive explosive welding welding proces process,s, chemical chemical equilibrium equilibrium was was not not achievedachieved.. Consequently, Consequently, this this may may exhibit exhibit various various kinds kinds of of intermetallics intermetallics at at different different metastable metastable equilibriumequilibrium states states [49]. [49]. Ti Ti-Al-Al has has three three main main titanium titanium aluminides aluminides TiAl, TiAl, TiAl TiAl33,, and and Ti Ti3Al.Al. TiAl TiAl2 2isis anotheranother equilibrium equilibrium phase phase of of titanium titanium aluminide. aluminide. However, However, this this phase phase is is formed formed under under the the overlap overlap conditioncondition with with the the TiAl TiAl phase phase [14]. [14]. Figure Figure 5c5c shows shows that that the the atomic atomic percentage percentage of of aluminium aluminium in in the the interfaceinterface is is 60% 60% and and titanium titanium is is about about 40%. 40%. It It indicates indicates that that this this region region is is formed formed by by o overlappingverlapping of of twotwo different different phases phases of of Ti Ti-Al,-Al, i.e., TiAlTiAl+ + TiAlTiAl22.. Fronczek etet al.al. [[14]14] observedobserved this this phase phase with with the the help help of ofX-ray X-ray synchrotron synchrotron reflection reflection method. method. Similarly, Similarly, in Figurein Figure5d, EDS5d, EDS point point scan scan exhibits exhibits that pointthat point 2 has 2 a haschance a chance of overlapped of overlapped phase TiAlphase+ TTiAliAl2 +and TiAl point2 and 5 may point contains 5 may containsTiAl3 phase. TiAl While3 phase. the While remaining the remainingpoints show points no external show no interference external interference to other elements. to other elements.

4.2. Microstructure of Welding Interface after Heat Treatment

Metals 2019, 9, 1189 8 of 17

4.2. Microstructure of Welding Interface after Heat Treatment Metals 2019, 9, x FOR PEER REVIEW 8 of 17 Figure6a–d with di fferent selected area magnifications show that the interfacial zone becomes widerFig afterure 6a 4 h–d heat with treatment.different selected Pores area are magnifications significantly decreasedshow that th ande interfacial grains arezone rearranged. becomes High deformationwider after 4 canh heat produce treatment. pores Pores and are cracks significantly during explosive decreased welding.and grains Consequently, are rearranged. when High the stress ofdeformation the composite can produce plate ispores relieved and cracks by heatduring treatment explosive process,welding. Consequently, then these cracks when the can stress generate the islandof the/ peninsula-like composite plate morphology. is relieved by Furthermore,heat treatment theprocess, formation then these of this cracks island can/ peninsula generate the pattern is influencedisland/peninsula by the-like detonation morphology. force Furthermore, and metal vortex the formation flow. Previously, of this island/peninsula many researchers pattern observed is this typeinfluenced of morphology by the detonation [16,48,49 force]. Figure and metal6d shows vortex this flow. titanium-island-like Previously, many researchers shape (marked observed by this red line) in type of morphology [16,48,49]. Figure 6d shows this titanium-island-like shape (marked by red line) the aluminium rich area. These types of islands may affect the mechanical performance of the welded in the aluminium rich area. These types of islands may affect the mechanical performance of the plate [50,51]. welded plate [50,51].

FigureFigure 6. 6. SEMSEM images images ofof explosively explosively welded welded Ti6Al4V Ti6Al4V/Al1060/Al1060 interfaces after after heat heat treatment at at different different magnifications of selected regions (a) 500×, (b) 1000×, (c) 2000×, (d) 5000×. magnifications of selected regions (a) 500 ,(b) 1000 ,(c) 2000 ,(d) 5000 . × × × × The elemental scan of Ti-Al interface (Figure 7a) shows that at the interface, an island (as The elemental scan of Ti-Al interface (Figure7a) shows that at the interface, an island (as indicated indicated by arrow) is formed with titanium as the dominant element while the surrounding area by arrow) is formed with titanium as the dominant element while the surrounding area consists of consists of titanium aluminides. Their detailed atomic percentage can be determined by using the titaniumEDS line aluminides.and point scans Their (Figure detailed 7b–d). atomic As shown percentage in Figure can 7b, be two determined positions are by selected using the for EDSline line and pointscan EDS scans. One (Figure line 7passesb–d). through As shown islands in Figure and the7b, neighboring two positions area areof interface selected (Figure for line 7c), scan while EDS. One linethe passessecond line through passes islands through and interface the neighboring and surroundings area (Figure of interface 7d). (Figure7c), while the second line passesFigure through 7b describes interface the and point surroundings elemental scan (Figure near7 inted). rface and island morphology. The point scansFigure 3 and 75b show describes that the the area point near the elemental interfacescan contains near titanium interface aluminides. and island Based morphology. on EDS atomic The point scanspercentage 3 and analysis, 5 show there that is the a possibility area near of the the interfaceexistence of contains most stable titanium titanium aluminides. aluminide (TiAl Based3). on EDS atomicIt can furt percentageher be verified analysis, from thereline scan is ain possibility Figure 7c, which of the shows existence that region of most 2 has stable higher titanium titanium aluminide counts as compared to the base plate side. It indicates that the post heat treatment process influences (TiAl ). It can further be verified from line scan in Figure7c, which shows that region 2 has higher the aluminides3 equilibrium state. Points 1, 2 and 4 scans show that there is no reasonable external titanium counts as compared to the base plate side. It indicates that the post heat treatment process involvement of any element in each side. Particularly point 4 is situated in the mid of island influencesmorphology. the This aluminides point shows equilibrium that the island state. is formed Points from 1, 2 and the flyer 4 scans plate. show that there is no reasonable

Metals 2019, 9, 1189 9 of 17 external involvement of any element in each side. Particularly point 4 is situated in the mid of island morphology.Metals 2019, This9, x FOR point PEER shows REVIEW that the island is formed from the flyer plate. 9 of 17 EDS line scan (Figure7c) shows that the interfacial zone is expended up to approximately 11.55 µm, which is almostEDS line 10 scan times (Figure larger 7c) than shows unheated that the inte samplerfacial (Figure zone is5 b).expended In this up interfacial to approximately area, the 11.55 titanium islandμm, surrounded which is almost by Ti-Al 10 intermetallicstimes larger than is observed,unheated sample which has(Figure a width 5b). In of this 5 µm. interfacial Titanium area, aluminides the titanium island surrounded by Ti-Al intermetallics is observed, which has a width of 5 μm. Titanium of different equilibrium phases have been observed in the area between island and interface, which aluminides of different equilibrium phases have been observed in the area between island and µ is expandedinterface, up which to 4.6is expandedm. Heat uptreatment to 4.6 μm. Heat reduces treatment the number reduces ofthe element number of counts element in counts the unit in area, increasesthe unit the area, titanium increases penetration the titanium in aluminium penetration and in aluminium stretched theand interfacial stretched the zone. interfacial Furthermore, zone. the equilibriumFurthermore, phases the of titaniumequilibrium aluminide phases of are titanium formed. aluminide Especially, are TiAl formed.3 is dominated Especially, as comparedTiAl3 is to all otherdominated phases. as This compared kind ofto variationsall other phases. were This also kind observed of variations by Gloc were et al.also [42 observed] and Fronczek by Gloc et al. al. [14]. Figure[42] and7 dFronczek shows et that al. [14]. interfacial zone becomes more extensive than the sample before heat treatment.Figu Althoughre 7d shows titanium that is interfacial penetrated zone into becomes aluminium more upextensive to 3 µ m, than the the high sample level before of Ti-Al heat mixing is intreatment. the range Although of 1.7 µm titanium with a is maximum penetrated areainto aluminium existed in up the to aluminium3 μm, the high side. level Comparisonof Ti-Al mixing of the Figureis 7inc–d the indicates range of that1.7 μ them with interfacial a maximum area area has aexisted di fference in the ofaluminium about 9.8 side.µm. Comparison It shows that of the post Figure 7c–d indicates that the interfacial area has a difference of about 9.8 μm. It shows that the post heat treatment process makes the interface irregular shape. Additionally, it is noticed that titanium heat treatment process makes the interface irregular shape. Additionally, it is noticed that titanium aluminidesaluminides have have existed existed in in the the area area between the the titanium titanium island island and andthe flyer the plate. flyer This plate. area This did areanot did not directlydirectly exposeexpose during the the collision collision of of the the plates. plates. It Itimplies implies that that titanium titanium aluminides aluminides was wascreated created duringduring the heatthe heat treatment treatment process. process.

(a) (b)

FigureFigure 7. Pictures 7. Pictures and and EDS EDS after after heat heat treatmenttreatment ( aa)) Ti Ti-Al-Al mapping mapping near near interface, interface, (b) Ti (b-)Al Ti-Al atomic atomic distributiondistribution near near interface interface at selectedat selected points, points, ( c()c) linear linear distributiondistribution of of Ti Ti-Al-Al near near interface interface (at (atposition position 1 in Figure1 in Figure7b), ( d 7b),) Linear (d) Linear distribution distribution of Ti-Alof Ti-Al near near interface interface (at (at positionposition 2 2 in in Figure Figure 7b)7b)..

4.3. Mechanical4.3. Mechanical Tests Tests TensileTensile tests tests were were conducted conducted to study to study the mechanical the mechanical response response of the of welded the welded material. material. Two Two samples weresamples prepared were from prepared normal from and normal heat treated and heat welded treated plates,welded as plates, shown as shown in Figure in Figure8. 8.

Metals 2019, 9, x FOR PEER REVIEW 10 of 17

Metals 2019, 9, 1189 10 of 17 Metals 2019, 9, x FOR PEER REVIEW 10 of 17

Figure 8. Tensile test samples (a) before the test; (b) after the test. Figure 8. Tensile test samples (a) before the test; (b) after the test. Figure 8. Tensile test samples (a) before the test; (b) after the test. For successful bonding, the tensile strength of the welded sample should be higher than the For successful bonding, the tensile strength of the welded sample should be higher than the weakerFor material successful used bonding, for welding the tensile [50]. strength Ege et al.of the[16] welded proposed sample that should the yield be higher strength than ofthe welded weaker material used for welding [50]. Ege et al. [16] proposed that the yield strength of welded materialweaker depend materials onused aluminium for welding volumetric [50]. Ege et percentage al. [16] proposed and interfacial that the yield density. strength Fig ofure welded 9 shows the material depends on aluminium volumetric percentage and interfacial density. Figure9 shows the stressmaterial-strain depend curvess onof aluminiumthe tested volumetricsamples before percentage and afterand interfacial heat treatment. density. AfterFigure heat 9 shows treatment, the the stress-strainstress-strain curves curves of of the the testedtested samples before before and and after after heat heat treatment. treatment. After After heat treatment, heat treatment, the the tensile strength of the welded sample was decreased, while the elongation and plasticity were tensiletensile strength strength of the of the welded welded sample sample was was decreased, decreased, while while the theelongation elongation and and plasticity plasticity were were increased. increased. Furthermore, Figure 9 shows that a small jerk occurred at the end of elastic limit of the Furthermore,increased. Furthermore, Figure9 shows Figure that 9 shows a small that jerk a small occurred jerk occurred at the end at the of end elastic of elastic limit limit of the of unheatedthe unheated sample. This jerk state indicates that failure begins from the weak material (Al-1060) before sample.unheated This sample. jerk state This indicates jerk state thatindicates failure that begins failure from begins the from weak the materialweak material (Al-1060) (Al-1060) before before the tensile the tensile strength is reached. The experimental results of the tensile tests were compared with the strengththe tensile is reached. strengthThe is reached. experimental The experimental results of results the tensile of the tests tensile were tests compared were compared with thewith mechanical the mechanicalmechanical properties properties of of thethe Al 10601060 [51] [51] and and Ti6Al4V Ti6Al4V [52] [52] based based on their on their percentage percentage thickness thickness in the in the properties of the Al 1060 [51] and Ti6Al4V [52] based on their percentage thickness in the welded plate weldedwelded plate plate (Table (Table 2). 2). ThisThis approximation helps helps us usto estimateto estimate the tensilethe tensile properties properties of the ofwelded the welded (Table2). This approximation helps us to estimate the tensile properties of the welded sample [ 16,53]. samplesample [16,53]. [16,53]. Detailed Detailed resultsresults areare shown shown in in Table Table 2. 2. Detailed results are shown in Table2. It isIt isindicated indicated that that bothboth samples have have better better results results than than the thecalculated calculated values. values. Furthermore, Furthermore, It is indicated that both samples have better results than the calculated values. Furthermore, experimentalexperimental results results exhibit exhibit that the the tensile tensile strength strength and and yield yield stress stress of heat of heattreated treated samples samples are are experimentalreduced as resultscompared exhibit to the that unheated the tensile sample, strength but elongation and yield is stressimproved of heat from treated 20% to samples 23%. are reduced reduced as compared to the unheated sample, but elongation is improved from 20% to 23%. as compared to the unheated sample, but elongation is improved from 20% to 23%.

Figure 9. Engineering stress-strain profile before and after heat treatment. Figure 9. Engineering stress-strain proﬁle before and after heat treatment. TableFigure 2. Tensile 9. Engineering test results ofstress the welded-strain profilesamples before together and with after specific heat materials.treatment . Table 2. Tensile test results of the welded samples together with speciﬁc materials. Samples UTS (MPa) Yield Stress (MPa) Elongation (%) Table 2. Tensile test results of the welded samples together with specific materials. TiSamples-6Al-4V [52] UTS (MPa)947 Yield Stress (MPa)872 Elongation13 (%) Al-1060 [51] 97 110 28 Ti-6Al-4VCalculatedSamples [52 value ] 947UTS509 (MPa) 872Yield446 Stress (MPa) 13 20.7Elongation (%) BeforeAl-1060Ti-6Al heat-4V [treatment51 [52]] 97560 ±947 4 110486 ± 8872 28 20 ± 1 13 CalculatedAfterAl -heat1060 treatment value[51] 509525 ± 971 446462.5 ± 8110 20.7 23 ± 2 28 Before heat treatment 560 4 486 8 20 1 Calculated value ± 509 ± 446 ± 20.7 After heat treatment 525 1 462.5 8 23 2 Before heat treatment ± 560 ± 4 ± 486 ± 8 ± 20 ± 1 After heat treatment 525 ± 1 462.5 ± 8 23 ± 2 Three points bending test results show that no fractures or cracks were observed in any samples for both before and after heat treatment, as seen in Figure 10a,b. MetalsMetals 2019 2019, ,9 9, ,x x FOR FOR PEER PEER REVIEW REVIEW 1111 of of 17 17

ThreeThree points points bending bending test test results results show show that that no no fractures fractures or or cracks cracks were were observed observed in in any any sample sampless Metals 2019, 9, 1189 11 of 17 forfor both both before before and and after after heat heat treatment, treatment, as as seen seen in in Figure Figure 10a,b. 10a,b.

Figure 10. Three points bending test (a) side view of 90◦and 180◦ bent sample; (b) bending interface. FigureFigure 10 10. .Three Three point pointss bending bending test test ( a(a) )s sideide view view of of 90°and 90°and 180° 180° bent bent sample; sample; ( b(b) )bending bending interface interface. . In addition to the bending test and tensile test, the ﬂat shear test was conducted to verify the InIn addition addition to to the the bending bending test test and and tensile tensile test, test, the the flat flat shear shear test test was was conducted conducted to to verify verify the the welding joint strength for both samples before and after heat treatment. Figure 11a indicates that the weldingwelding joint joint strength strength for for both both samples samples before before and and after after heat heat treatment. treatment. Figure Figure 11a 11a ind indicatesicates that that the the tensile strength of aluminium decreases after the heat treatment. Figure 11b shows that the deformation tensiletensile strength strength of of aluminium aluminium decreases decreases after after the the heat heat treatment. treatment. Figure Figure 11b 11b shows shows that that the the starts in the aluminium and the joint displays no disturbance. This result reveals that the joint is deformationdeformation start startss in in the the aluminium aluminium and and the the joint joint displays displays no no disturbance. disturbance. This This result result reveals reveals that that stronger than the Al-1060 tensile strength ~40MPa. thethe joint joint is is stronger stronger than than the the Al Al-1060-1060 tensile tensile s strengthtrength ~40MPa. ~40MPa.

(a(a) ) (b(b) ) Figure 11. (a) Shear force versus displacement (b) fracture morphology after shear test. FigureFigure 11. 11. ( a(a) )Shear Shear force force versus versus displacement displacement ( b(b) )fracture fracture morphology morphology after after shear shear test test. . Figure 12 shows that near the interface, the microhardness is maximum. Its value is about ~412, FigureFigure 12 12 shows shows that that near near the the interface, interface, the the microhardness microhardness is is maximum. maximum. Its Its value value is is about about ~412, ~412, higher than the flyer and base standard values (flyer plate ~350, base plate ~30), which is due to high higherhigher than than the the flyer flyer and and base base standard standard values values (flyer (flyer plate plate ~350, ~350, base base plate plate ~30), ~30), which which is is due due to to high high value of heat produced during explosive welding caused annealing at the contact point [47]. Heat valuevalue of of heat heat produced produced during during explosive explosive welding welding caused caused annealing annealing at at the the contact contact point point [47]. [47]. Heat Heat treatment relaxed the interface and widened the interfacial zone, which affected the hardness value and treatmenttreatment relaxed relaxed the the interface interface and and widened widened the the interfacial interfacial zone, zone, which which affected affected the the hardness hardness v valuealue causedand caused it to decrease it to decrease to ~390 to at ~390 the interface. at the interface. As moving As moving away from away the from interface, the interface, hardness hardness becomes andMetals caused 2019 ,it 9 , tox FOR decrease PEER REVIEW to ~390 at the interface. As moving away from the interface, hardness12 of 17 approachingbecomesbecomes approaching approaching to the typical to to the the values typical typical of values flyervalues and of of flyer baseflyer and plateand base base plate plate

Figure 12. Microhardness (Hv) proﬁles before and after heat treatment. Figure 12. Microhardness (Hv) profiles before and after heat treatment.

4.4. Simulation Results Figure 13a shows that smooth interface obtained by numerical simulation, which is in good agreement with the experimental results (see in Figure 4). Since Ti6Al4V is a very hard material and Al-1060 is soft, the flyer can easily penetrate into the base material. Furthermore, Figure 13a explains that some partial wavy shapes patterns appear, but due to high impact velocity and density difference, these waves are suppressed and became almost flat. Figure 13b exhibits that jet is formed during this process and the maximum portion of the jet consisted of base plate (Al-1060).

Figure 13. Simulation results for (a) Ti6Al4V/Al-1060 interface and (b) jet formation.

Figure 14a,b show that near the contact point, the pressure distribution is behind the jet. It means that the collision velocity is subsonic, which is an essential requirement for the bonding of materials. According to Blazynski et al. [4], the bonding contact pressure should be higher than material yield strength. Pressure contour plots (Figure 14a,b) illustrate that the transient pressure profile at all contact points during the collision is more than 10 GPa. Figure 14c supports that, at time 0.256 μs, the peak flyer transient pressure rises to 13.1 GPa, which is considerably higher than the yield strength of the flyer and base plate. Additionally, it was observed (Figure 14c) that in the base plate, the pressure and impulse were slightly lower than that of the flyer. According to Holtzman et al. [54] investigation, if the base plate had a greater contribution to the jetting, there would be more transient pressure in the flyer side. Similarly, Mousavi et al. [37] simulated the same pressure difference between the flyer and base plates. Furthermore, Figure 14c shows that there is a slight hump in the pressure-time curves before reaching the peak value. This hump indicates that the deformity had begun before the arrival of the

Metals 2019, 9, x FOR PEER REVIEW 12 of 17

Metals 2019, 9, 1189 12 of 17 Figure 12. Microhardness (Hv) profiles before and after heat treatment.

4.4.4.4. Simulation Results FigureFigure 13 13aa shows shows that that smooth smooth interface interface obtained obtained by by numerical numerical simulation, simulation, which is in good good agreementagreement withwith thethe experimentalexperimental resultsresults (see(see inin FigureFigure4 4).). SinceSince Ti6Al4VTi6Al4V isis aa veryvery hardhard materialmaterial andand Al-1060Al-1060 isis soft,soft, thethe flyerflyer cancan easilyeasily penetratepenetrate intointo thethe basebase material.material. Furthermore, Figure 1313aa explains thatthat some some partial partial wavy wavy shapes shapes patterns patterns appear, appear, but due but to due high to impact high velocity impact and velocity density and di ff densityerence, thesedifference, waves these are suppressed waves are suppressed and became and almost became flat. almost Figure 13flat.b exhibitsFigure 13b that exhibits jet is formed that jet during is formed this processduring this and pr theocess maximum and the portion maximum of the portion jet consisted of the jet of consisted base plate of (Al-1060). base plate (Al-1060).

Figure 13. Simulation results for (a) Ti6Al4V/Al-1060 interface and (b) jet formation. Figure 13. Simulation results for (a) Ti6Al4V/Al-1060 interface and (b) jet formation. Figure 14a,b show that near the contact point, the pressure distribution is behind the jet. It means that theFigure collision 14a,b velocityshow that is subsonic,near the contact which point, is an essentialthe pressure requirement distribution for theis behind bonding the ofjet. materials. It means Accordingthat the collision to Blazynski velocity et is al. subsonic, [4], the bonding which is contactan essential pressure requirement should be for higher the bonding than material of materials. yield strength.According Pressure to Blazynski contour et plotsal. [4], (Figure the bonding 14a,b) illustratecontact pressur that thee transientshould be pressure higher profilethan material at all contact yield pointsstrength. during Pressure the collision contour is plots more (Figure than 10 14a,b) GPa. Figure illustrate 14 c that supports the transient that, at time pressure 0.256 profileµs, the at peak all flyercontact transient points pressureduring the rises collision to 13.1 isGPa, more which than 10 is GPa. considerably Figure 14c higher supports than that, the yield at time strength 0.256 μs, of thethe flyerpeak andflyer base transient plate. pressure rises to 13.1 GPa, which is considerably higher than the yield strength of theAdditionally, flyer and base it wasplate. observed (Figure 14c) that in the base plate, the pressure and impulse were slightlyAdditionally, lower than it that was of observed the flyer. (Figure According 14c) tothat Holtzman in the base et al.plate, [54] the investigation, pressure and if theimpulse base platewere hadslightly a greater lower contributionthan that of the to theflyer. jetting, According there to would Holtzman be more et al. transient [54] investigation, pressure in if the base flyer plate side. Similarly,had a greater Mousavi contribution et al. [37 ]to simulated the jetting, the there same would pressure be di morefference transient between pressure the flyer in and the baseflyer plates. side. Similarly, Mousavi et al. [37] simulated the same pressure difference between the flyer and base Metals Furthermore,2019, 9, x FOR PEER Figure REVIEW 14c shows that there is a slight hump in the pressure-time curves13 before of 17 reachingplates. the peak value. This hump indicates that the deformity had begun before the arrival of Furthermore, Figure 14c shows that there is a slight hump in the pressure-time curves before shockthe shock wave. wave. Cowen Cowen et al. et [55] al. [ 55 reported] reported that that deformation deformation and and jet jet formation formation were were necessary necessary for for reaching the peak value. This hump indicates that the deformity had begun before the arrival of the materialmaterial bonding. bonding.

Figure 14. Numerica Simulation results for (a,b) pressure contour and (c) pressure-time graph at Figure 14. Numerica Simulation results for (a,b) pressure contour and (c) pressure-time graph at the the interface. interface.

The high impact creates shearing at the interface that causes to rise in heating and produce a bond between different metals. Kinetic energy during impact is converted into plastic work that causes the rise of temperature. Plastic deformation is not possible to witness experimentally, so simulation is an excellent way to explain the process. Simulation results show that along with interface (Figure 15a–d), the maximum plastic strain is increased up to 7, while averagely its value is more than 5 throughout the interface. This strain value is enough to verify that at this impact velocity, the welding should be possible (the minimum plastic deformation required for welding should be more than 0.25 (Ti-MS) [37]). Figure 15a,b show that high plastic deformation expands up to 2 μm near the interface, which is similar to the experimentally measured interfacial area. It indicates that maximum deformation of the region generates the interfacial zone. Figure 15c explains that this deformation is purely localized. As we move away from the interface, this deformation abruptly decreases to the minimum level. Furthermore, after 10 μm the flyer shows no plastic deformation, while the base plate has accommodated this deformation up to 30 μm. Figure 15d illustrates that at the same impact point, the flyer and base plates have different values of plastic strain. The base plate has a plastic strain value almost double than that of the flyer plate. It is just because of the differences in their density and mechanical properties.

Metals 2019, 9, x FOR PEER REVIEW 13 of 17

shock wave. Cowen et al. [55] reported that deformation and jet formation were necessary for material bonding.

Metals 2019, 9, 1189 13 of 17 Figure 14. Numerica Simulation results for (a,b) pressure contour and (c) pressure-time graph at the interface. The high impact creates shearing at the interface that causes to rise in heating and produce a bond The high impact creates shearing at the interface that causes to rise in heating and produce a between different metals. Kinetic energy during impact is converted into plastic work that causes the bond between different metals. Kinetic energy during impact is converted into plastic work that risecauses of temperature. the rise of temperature. Plastic deformation Plastic deformation is not possible is not to possible witness to experimentally, witness experimentally, so simulation so is an excellentsimulation way is an to excellent explain way the process.to explain the process. SimulationSimulation results results show show that that along along with with interface interface (Figure (Figure 15a–d), 15 thea–d), maximum the maximum plastic strain plastic is strain isincreased increased up up to 7, to while 7, while averagely averagely its value its is value more isthan more 5 throughout than 5 throughout the interface. the This interface. strain value This strain valueis enough is enough to verify to that verify at this that impact at this velocity, impact the velocity, welding the should welding be possible should (the be minimum possible (theplastic minimum plasticdeformation deformation required requiredfor welding for should welding be more should than be 0.25 more (Ti-MS) than [37]). 0.25 Figure (Ti-MS) 15a,b [37 show]). Figurethat 15a,b showhigh plastic that high deformation plastic deformationexpands up to 2 expands μm near upthe interface, to 2 µm which near theis similar interface, to the whichexperimenta is similarlly to the experimentallymeasured interfacial measured area. It interfacial indicates area.that maximum It indicates deformation that maximum of the region deformation generates of the the region interfacial zone. generates the interfacial zone. Figure 15c explains that this deformation is purely localized. As we move away from the interface,Figure this 15 cdeformation explains that abruptly this deformation decreases to is the purely minimum localized. level. AsFurthermore, we move awayafter 10 from μm thethe interface, thisflyer deformation shows no plastic abruptly deformation, decreases while to the the minimumbase plate has level. accommodated Furthermore, this after deformation 10 µm the up flyer to shows no30 plastic μm. deformation, while the base plate has accommodated this deformation up to 30 µm. FigureFigure 1515dd illustratesillustrates that at the same impact point, point, the the flyer flyer and and base base plates plates have have different different values ofvalues plastic of strain.plastic Thestrain. base The plate base hasplate a has plastic a plastic strain strain value value almost almost double double than than that that of of the the flyer flyer plate. It is justplate. because It is just of because the diff oferences the differences in their densityin their density and mechanical and mechanical properties. properties.

Metals 2019, 9, x FOR PEER REVIEW 14 of 17

Figure 15. (a,b) Distribution of plastic deformation contour along with the interface, (c) graphical Figurerepresentation 15. (a,b) Distribution of plastic strain of plastic along deformation the vertical contour distance along of interface, with the interface, and (d) plastic(c) graphical strain plot with representationrespect to time of atplastic the interface. strain along the vertical distance of interface, and (d) plastic strain plot with respect to time at the interface. During explosive welding, high pressure and plastic strain raise the temperature locally and During explosive welding, high pressure and plastic strain raise the temperature locally and abruptly because this phenomenon occurs in a very short interval of time and the cooling rate during abruptly because this phenomenon occurs in a very short interval of time and the cooling rate during this process is very high about 105–107 K/s [38]. this process is very high about 105–107 K/s [38]. In contour contour plots plots of oftemperature temperature (Figure (Figure 16a– 16c), a–c),it is indicated it is indicated that the that temperature the temperature increment increment is is enoughenough to to melt melt for for both both metals. metals. However, However, both boththe temperature the temperature increments increments of flyer and of flyer base andplates base plates areare not the the same. same. At At the the flyer flyer plate plate side, side, the average the average maximum maximum temperature temperature is 3000 K, is while 3000 K,thewhile base the base hashas an an average average of of1200 1200 K. It K. is also It is observed also observed that along that with along the interface, with the the interface, peak temperature the peaks have temperatures haveno fixed no value. fixed value.Figure 16d Figure shows 16 dthat shows the flyer that has the attained flyer has its attainedmelting point its melting temperature point in temperature a very in a short time during the impact and then cools down immediately. While at the base plate side, the melting temperature sustains longer as compared to flyer. Obtained the melting point implies that the interfaces of both materials are more likely to form intermetallics. Furthermore, the interfacial zone behaves like the fluid flow and gets elongated toward detonation direction. Additionally, this process causes the refinement of grains near the interface.

Figure 16. Plot of the temperature distribution at (a) the interface, (b) flyer side, (c) base side and (d) temperature profile of flyer and base plates at the interface.

5. Conclusions Ti6Al4V was successfully welded with Al-1060 by explosive welding. The welded interface between Ti and Al was smooth and straight without any jet trapping. The maximum portion of the interfacial zone existed in the base side (Al-1060) where different phases of titanium aluminide were observed. Mechanical results, i.e., tensile test, bending test, shear test and Vickers hardness test, showed that welding quality was not highly affected by these titanium aluminides.

Metals 2019, 9, x FOR PEER REVIEW 14 of 17

Figure 15. (a,b) Distribution of plastic deformation contour along with the interface, (c) graphical representation of plastic strain along the vertical distance of interface, and (d) plastic strain plot with respect to time at the interface.

During explosive welding, high pressure and plastic strain raise the temperature locally and abruptly because this phenomenon occurs in a very short interval of time and the cooling rate during this process is very high about 105–107 K/s [38]. In contour plots of temperature (Figure 16a–c), it is indicated that the temperature increment is enough to melt for both metals. However, both the temperature increments of flyer and base plates are not the same. At the flyer plate side, the average maximum temperature is 3000 K, while the base hasMetals an2019 average, 9, 1189 of 1200 K. It is also observed that along with the interface, the peak temperatures14 have of 17 no fixed value. Figure 16d shows that the flyer has attained its melting point temperature in a very short time during the impact and then cools down immediately. While at the base plate side, the very short time during the impact and then cools down immediately. While at the base plate side, melting temperature sustains longer as compared to flyer. Obtained the melting point implies that the melting temperature sustains longer as compared to flyer. Obtained the melting point implies that the interfaces of both materials are more likely to form intermetallics. Furthermore, the interfacial the interfaces of both materials are more likely to form intermetallics. Furthermore, the interfacial zone zone behaves like the fluid flow and gets elongated toward detonation direction. Additionally, this behaves like the fluid flow and gets elongated toward detonation direction. Additionally, this process process causes the refinement of grains near the interface. causes the refinement of grains near the interface.

Figure 16. Plot of the temperature distribution at (a) the interface, (b) flyer side, (c) base side and (d) Figure 16. Plot of the temperature distribution at (a) the interface, (b) flyer side, (c) base side and (d) temperature profile of flyer and base plates at the interface. temperature profile of flyer and base plates at the interface. 5. Conclusions 5. Conclusions Ti6Al4V was successfully welded with Al-1060 by explosive welding. The welded interface betweenTi6Al4V Ti and was Al successfully was smooth welded and straight with Al without-1060 by any explosive jet trapping. welding. The The maximum welded portion interface of betweenthe interfacial Ti and zone Al was existed smooth in the and base straight side (Al-1060)without any where jet trapping. different phasesThe maximum of titanium portion aluminide of the interfacialwere observed. zone existed Mechanical in the results, base side i.e., (Al tensile-1060) test, where bending different test, shearphases test of andtitanium Vickers aluminide hardness were test, observed.showed that Mechanical welding quality results, was i.e., not tensile highly test, aff ected bending by these test, sheartitanium test aluminides. and Vickers hardness test, showedHeat that treatment welding processquality stretchedwas not highly the interfacial affected by zone these with titanium some titanium aluminides. island /peninsula like shape. Due to this, strength of welded material was decreased as compared to the normal welded sample, but ductility was improved. Numerical simulation depicted that impact pressure at all contact points had larger values than the yield strength of both welded materials, which is one of the basic requirements to meet the welding conditions. Furthermore, simulation results showed that in the interfacial zone, plastic deformation had values more than 5 and both materials obtained their melting points during impact. Melting of both materials provide a reason to form titanium aluminides. Since pressure, plastic deformation and temperature distribution for both materials (flyer and base) had different values, therefore, both materials had different interfacial thickness.

Author Contributions: Conceptualization, Y.M. and P.C.; explosive welding, Y.M. and Q.Z.; microstructure examination and mechanical tests, Y.M., A.A.B. and A.A.; numerical simulation, Y.M., K.D. and Q.Z.; writing–original draft manuscript, Y.M.; writing–review and editing, K.D. and P.C. Funding: This research was funded by the National Natural Science Foundation of China (Grant No. 11521062) and State Key Laboratory of Explosion Science and Technology, Beijing Institute of Technology (Grant No. ZDKT18-01). Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Findik, F. Recent developments in explosive welding. Mater. Des. 2011, 32, 1081–1093. [CrossRef] 2. Wang, B.F.; Luo, X.Z.; Wang, B.; Zhao, S.T.; Xie, F.Y. Microstructure and its formation mechanism in the interface of Ti/NiCr explosive cladding bar. J. Mater. Eng. Perform. 2014, 24, 1050–1058. [CrossRef]

3. Wang, B.F.; Xie, F.Y.; Wang, B.; Luo, X.Z. Microstructure and properties of the Ti/Al2O3/NiCr composites fabricated by explosive compaction/cladding. Mater. Sci. Eng. C 2015, 50, 324–331. [CrossRef][PubMed] 4. Blazynski, T.Z. Explosive Welding, Forming and Compact; Springer: Dordrecht, The Netherlands, 1983. Metals 2019, 9, 1189 15 of 17

5. Perusko, D.; Petrovic, S.; Stojianovic, M.; Mitric, M.; Cizmovic, M.; Panjan, M.; Milosavljevic, M. Formation of intermetallics by ion implantation of multilatered Al/Ti nano-structures. Nucl. Instrum. Methods Phys. Res. 2012, 282, 4–7. [CrossRef] 6. Adeli, M.; Seyedein, S.H.; Aboutalebi, M.R.; Kobashi, M.; Kanetake, N. A study on the combustion synthesis of titanium aluminide in the self-propagating mode. J. Alloys Compd. 2010, 497, 100–104. [CrossRef] 7. Arakawa, Y.Y.; Kobashi, M.; Kanetake, N. Foaming behavior of long-scale Al-Ti intermetallic foam by SHS mode combustion reaction. Intermetallics 2013, 41, 22–27. [CrossRef] 8. Goda, D.J.; Richards, N.L.; Caley, W.F.; Chaturvedi, M.C. The effect of processing variables on the structure and chemistry of Ti-aluminide based LMCS. Mater. Sci. Eng. A 2002, 334, 280–290. [CrossRef] 9. Ramos, A.S.; Calinas, R.; Vieira, M.T. The formation of γ-TiAl from Ti/Al multilayers with different periods. Surf. Coatings Technol. 2006, 200, 6196–6200. [CrossRef] 10. Romankov, S.E.; Mukashev, B.N.; Ermakov, E.L.; Muhamedshina, D.N. Structural formation of aluminide phases on titanium substrate. Surf. Coat. Technol. 2004, 180, 280–285. [CrossRef] 11. Kahraman, N.; Gulenc, B.; Findik, F. Corrosion and mechanical-microstructural aspects of dissimilar joints of Ti-6Al-4V and Al plates. Int. J. Impact Eng. 2007, 34, 1423–1432. [CrossRef] 12. E, J.C.; Huang, J.Y.; Bie, B.X.; Sun, T.; Fezzaa, K.; Xiao, X.H.; Sun, W.; Lou, S.N. Deformation and fracture of explosion-welded Ti/Al plates: A synchrotron-based study. Mater. Sci. Eng. A 2016, 674, 308–317. [CrossRef] 13. Xia, H.B.; Wang, S.G.; Ben, H.F. Microstructure and mechanical properties of Ti/Al explosive cladding. Mater. Des. 2014, 56, 1014–1019. [CrossRef] 14. Fronczek, D.M.; Wojewoda-Budka, J.; Chulist, R.; Sypien, A.; Korneva, A.; Szulc, Z.; Schell, N.; Zieba, P. Structural properties of Ti/Al clads manufactured by explosive welding and annealing. Mater. Des. 2016, 91, 80–89. [CrossRef] 15. Bazarnik, P.; Adamczyk-Cieslak, B.; Galka, A.; Plonka, B.; Sniezek, L.; Cantoni, M.; Lewandowska, M. Mechanical and microstructural characteristics of Ti6Al4V/AA2519 and Ti6Al4V/AA1050/AA2519 laminates manufactured by explosive welding. Mater. Des. 2016, 111, 146–157. [CrossRef] 16. Ege, E.S.; Inal, O.T.; Zimmerly, C.A. Response surface study on production of explosively-welded aluminum-titanium laminates. J. Mater. Sci. 1998, 33, 5327–5338. [CrossRef] 17. Inal, O.T.; Szecket, A.; Vigueras, D.J.; Pak, H. Explosive welding of Ti–6Al–4V to mild-steel substrates. J. Vac. Sci. Technol. A Vac. Surf. Film 1985, 3, 2605–2609. [CrossRef] 18. Ege, E.S.; Inal, O.T. Stability of interfaces in explosively-welded aluminum-titanium. J. Mater. Sci. Lett. 2000, 19, 1533–1535. [CrossRef] 19. Lazurenko, D.V.; Bataev, I.A.; Mali, V.I.; Bataev, A.A.; Maliutina, I.N.; Lozhkin, V.S.; Esikov, M.A.; Jorge, A.M.J. Explosively welded multilayer Ti-Al composites: Structure and transformation during heat treatment. Mater. Des. 2016, 102, 122–130. [CrossRef] 20. Pavliukova, D.V.; Bataev, I.A.; Bataev, A.A.; Smirnov, A.I. Formation of the intermetallic layers in Ti-Al multilayer composites. Adv. Mater. Res. 2011, 311, 236–239. 21. Bataev, I.A.; Bataev, A.A.; Pavliukova, D.V.; Yartsev, P.S.; Golovin, E.D. Nucleation and growth of titanium aluminide in an explosion welded laminate composite. Phys. Met. Metallogr. 2012, 113, 998–1007. [CrossRef] 22. Fan, M.Y.; Domblesky, J.; Jin, K.; Qin, L.; Cui, S.Q.; Guo, X.Z.; Kim, N.; Tao, J. Effect of original layer thicknesses on the interface bonding and mechanical properties of Ti-Al laminate composites. Mater. Des. 2016, 99, 535–542. [CrossRef] 23. Fan, M.Y.; Yu, W.W.; Wang, W.T.; Guo, X.Z.; Jin, K.; Miao, R.J.; Hou, W.Q.; Kim, N.; Tao, J. Microstructure and mechanical properties of thin-multilayer Ti/Al laminates prepared by one-step explosive bonding. J. Mater. Eng. Perform. 2017, 26, 277–284. [CrossRef] 24. Foadian, F.; Soltanieh, M.; Adeli, M.; Etinanbakhsh, M. A study on the formation of intermetallics during the heat treatment of explosively welded Al-Timultilayers. Metall. Mater. Trans. A 2014, 45, 1823–1832. [CrossRef] 25. Mousavi, A.A.A.; Al-Hassani, S.T.S. Numerical and experimental studies of the mechanism of the wavy interface formations in explosive/impact welding. J. Mech. Phys. Solids 2005, 53, 2501–2528. 26. Nassiri, A.; Kinsey, B. Numerical studies on high-velocity impact welding: smoothed particle hydrodynamics (SPH) and arbitrary Lagrangian–Eulerian (ALE). J. Manuf. Process. 2016, 24, 376–381. [CrossRef] Metals 2019, 9, 1189 16 of 17

27. Wang, X.; Zheng, Y.Y.; Liu, H.X.; Shen, Z.B.; Hu, Y.; Li, W.; Gao, Y.Y.; Guo, C. Numerical study of the mechanism of explosive/impact welding using Smoothed Particle Hydrodynamics method. Mater. Des. 2012, 35, 210–219. [CrossRef] 28. Wang, Y.; Beom, H.G.; Sun, M.; Lin, S. Numerical simulation of explosive welding using the material point method. Int. J. Impact Eng. 2011, 38, 51–60. [CrossRef] 29. Saravanan, S.; Raghukandan, K.; Hokamoto, K. Improved microstructure and mechanical properties of dissimilar explosive cladding by means of interlayer technique. Arch. Civ. Mech. Eng. 2016, 16, 563–568. [CrossRef] 30. Tanaka, K. Numerical studies on the explosive welding by smoothed particle hydrodynamics (SPH). Shock Compress. Condens. Matter. 2007, 955, 1301–1304. 31. Brows, W.B.; Braithwaite, M. Analytical reprsentation of the adiabatic equation for detonation products based on statistical mechanics and intermolecular forces. Phil. Trans. Roy. Soc. A 1992, 339, 345–353. 32. Deribas, A.A. Acceleration of metal plates by a tangential detonation wave. J. Appl. Mech. Tech. Phys. 2000, 41, 824–830. [CrossRef] 33. Manikandan, P.; Hokamoto, K.; Fujita, M.; Raghukandan, K.; Tomoshige, R. Control of energetic conditions by employing interlayer of different thickness for explosive welding of titanium/304 stainless steel. J. Mater. Process. Technol. 2008, 195, 232–240. [CrossRef] 34. Cooper, P. Explosives Engineering; Wiley: New York, NY, USA, 1996. 35. Zukas, J.A.; Walters, W.P. Explosive Effects and Applications; Springer: New York, NY, USA, 1998. 36. Kamlet, M.J.; Jacobs, S.J. Chemistry of detonations. I. A simple method for calculating detonation properties of C-H-N-O explosives. J. Chem. Phys. 1968, 48, 23–35. [CrossRef] 37. Mousavi, S.A.A.A.; Al-Hassani, S.T.S. Finite element simulation of explosively-driven plate impact with application to explosive welding. Mater. Des. 2008, 29, 1–19. [CrossRef] 38. Crossland, B. Explosive Welding of Metals and Its Application; Oxford University Press: New York, NY, USA, 1982. 39. Lesuer, D. Experiment Investigations of Material Models for Ti–6Al–4V Titanium and 2024-T3 Aluminum; Final Report, DOT/FAA/AR-00/25; US Department of Transportation, Federal Aviation Administration: Washington, DC, USA, 2000. 40. Ye, L.; Zhu, X. Analysis of the effect of impact of near-wall acoustic bubble collapse micro-jet on Al 1060. Ultrason. Sonochem. 2017, 36, 507–516. [CrossRef] 41. Zhang, L.J.; Pei, Q.; Zhang, J.X.; Bi, Z.Y.; Li, P.C. Study on the microstructure and mechanical properties of explosive welded 2205/X65 bimetallic sheet. Mater. Des. 2014, 64, 462–476. [CrossRef] 42. Gloc, M.; Wachowski, M.; Plocinski, T.; Kurzydlowski, K.J. Microstructural and microanalysis investigations of bond titanium grade1/low alloy steel st52-3N obtained by explosive welding. J. Alloys Compd. 2016, 671, 446–451. [CrossRef] 43. Raoelison, R.N.; Racine, D.; Zhang, Z.; Buiron, N.; Marceau, D.; Rachik, M. Magnetic pulse welding: Interface of Al/Cu joint and investigation of intermetallic formation effect on the weld features. J. Manuf. Process. 2015, 16, 427–434. [CrossRef] 44. Su, Y.C.; Hua, X.M.; Wu, Y.X. Quantitative characterization of porosity in Fe-Al dissimilar materials lap joint made by gas metal arc welding with different current modes. J. Mater. Process. Technol. 2014, 214, 81–86. [CrossRef] 45. Tomashchuk, I.; Sallamand, P.; Cicala, E.; Peyre, P.; Grevey, D. Direct keyhole laser welding of aluminum alloy AA5754 to titanium alloy Ti6Al4V. J. Mater. Process. Technol. 2015, 217, 96–104. [CrossRef] 46. Murr, L.E.; Ferreyra, E.; Pappu, T.S.; Garcia, E.P.; Sanchez, J.C.; Huang, W.; Rivas, J.M.; Kennedy, C.; Ayala, A.; Niou, C.S. Novel deformation processes and microstructures involving ballistic penetrator formation and hypervelocity impact and penetration phenomena. Mater. Charact. 1996, 37, 245–276. [CrossRef] 47. Kacar, R.; Acarer, M. An investigation on the explosive cladding of 316L stainless steel-din-P355GH steel. J. Mater. Process. Technol. 2004, 152, 91–96. [CrossRef] 48. Prasanthi, T.N.; Kirana, R.; Saroja, S. Explosive cladding and post-weld heat treatment of mild steel and titanium. Mater. Des. 2016, 93, 180–193. [CrossRef] 49. Chulist, R.; Fronczek, D.M.; Szulc, Z.; Wojewoda-Budka, J. Texture transformations near the bonding zones of the three-layer Al/Ti/Al explosively welded clads. Mater. Charact. 2017, 129, 242–246. [CrossRef] Metals 2019, 9, 1189 17 of 17

50. Szecket, A.; Inal, O.T.; Vigueras, D.J.; Rocco, J. A wavy versus straight interface in the explosive welding of aluminum to steel. J. Vac. Sci. Technol. A 1985, 3, 2588–2593. [CrossRef] 51. Wang, Z.J.; Wang, Z.; Li, M.X. Failure analysis of Al 1060 sheets under double-sided pressure deformation conditions. Key Eng. Mater. 2007, 353, 603–606. [CrossRef] 52. Kimura, M.; Iijima, T.; Kusaka, M.; Kaizu, K.; Fuji, A. Joining phenomena and tensile strength of friction welded joint between Ti–6Al–4V titanium alloy and low carbon steel. J. Manuf. Process. 2016, 24, 203–211. [CrossRef] 53. Boroński,D.; Kotyk, M.; Maćkowiak, P.; Snie˙zek,´ L. Mechanical properties of explosively welded AA2519-AA1050-Ti6Al4V layered material at ambient and cryogenic conditions. Mater. Des. 2017, 133, 390–403. [CrossRef] 54. Holtzman, A.H.; Cowan, G.R. Bonding of metals with explosives. Weld. Res. Counc. 1965, 104, 1–40. 55. Cowan, G.R.; Holtzman, A.H. Flow configurations in colliding plates: explosive bonding. J. Appl. Phys. 1962, 34, 928–939. [CrossRef]

© 2019 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/).