Properties study on Ti/Al butt joining by GMAW/GTAW hybrid welding-brazing

For adjusting the distribution of welding heat to obtain sound Ti/Al butt joints, Gas Metal Arc Welding (GMAW)/Gas Tungsten Arc Welding (GTAW) hybrid welding-brazing was introduced to join 5A06 to TC4. Then two kinds of tensile strength was tested. And the weld microstructure was observed by Scanning Electron Microscope (SEM), X-ray diffraction (XRD) and Energy Dispersive Spectrometer (EDS). Effects of welding parameters on weld reinforcement and interface characteristics were investigated. Results revealed TiAl3 IMCs layer was formed in the weld area of the titanium alloy side. The other area of weld was composed of α-Al matrix and Mg2Si+Al+Si eutectic. The back welding reinforcement was improved with the welding heat of the joint back increasing. But this may lead to a large number of molten metal flowing to the back of weld, the face reinforcement of the weld became poor. However, effects of welding parameters on the interface bonding strength was different. It was depended on the thickness and morphology of IMCs. When serrated IMCs of 2–6μm thickness were formed, the highest tensile strength without reinforcement was 240.3 MPa. Through comprehensive analyzing, when the welding parameters were welding speed of 15 mm s−1, GTAW current of 80–90 A and GMAW/GTAW position of −1 mm, the weld reinforcement and interface bonding strength were also high. So the excellent butt joints of TC4/5A06 were achieved. The maximum average value of TC4/5A06 joints tensile strength with reinforcement was 265 MPa with fracture inside the weld far from IMCs layer.


Introduction
By virtue of the low specific gravity, high strength and low cost, there are a lot of applications of Ti/Al hybrid parts or structures in the aviation and automobile industries [1][2][3][4].Thus the high quality dissimilar junction between Ti alloys and Al alloys plays an important role in these hybrid structures or parts.However, Ti-Al intermetallic compounds (IMCs) with high brittleness will be produced at the elevated temperature [5].On the other hand, the significant differences of physical properties exist between Ti alloys and Al alloys [6].It is hard to achieve the joint of Ti/Al by the conventional fusion welding [7][8][9].
Welding-brazing for joining dissimilar metals has been developed in recent years [10,11].During the welding process, the brazing and fusion welding are combined and used for joining dissimilar materials: on the one hand, a joint of the fusion welding is formed on the side of low melting point base metal; on the other hand, a brazing joint is achieved on the side of high melting point base metal [12,13].The main advantage of the process is that the IMCs thickness can be controlled effectively because of its low welding heat input [14].Some studies have applied the welding-brazing process to acquire the overlap and butt joints of dissimilar metals, and the shear and tensile strength of the joints were studied.Researches suggest that the mechanical properties of the dissimilar metal joints are improved significantly [10,11,15].Therefore, it is a possible method to achieve a high-performance joint of Ti/Al dissimilar metals.Some attempts have been carried out to finish the joints between Ti alloys and Al alloys, such as laser welding-brazing and arc welding-brazing [10,[16][17][18][19][20][21].The results shows that the IMCs formation is controlled effectively.From the studies above, we can know that most welding-brazing methods above are used for achieving the dissimilar metal lap joints.While it is still difficult to realize the the dissimilar metal butt joints.The reason is that weld reinforcement and interface characteristics are the key factors to get the sound Ti/Al dissimilar metal joint in the butt welding-brazing.However, in singlesided Ti/Al welding-brazing, since the weld back is not heated directly by heat source, the temperature of the weld back is lower than that of the weld face.So it is hard to make the molten metals wet and then spread on the base metal back with high melting point [22,23].Even that the increase of welding heat input would improve the welding root reinforcement, but the time staying in high temperature of the weld is too long so that the IMCs grows very rapidly.So the integrity of the joint is damaged, and then the mechanical properties of the joints are deteriorated.At the same time, in single-sided welding-brazing of Ti/Al, the welding temperature gradient is large along the weld thickness direction, so the inhomogeneous thickness of IMCs is caused.At the top of the joint, IMCs of Ti-Al is too thick because it is near the welding heat source, but at the bottom, an ineffective metallurgical bonding may be formed because it is far from the heat source.This is detrimental to the joint strength [23,24].
In GMAW/GTAW hybrid welding-brazing, two separate power sources are used rather than just one [22].One feature of this method is the protection to the both sides of the joint, as there are GMAW and GTAW placed on the front and back sides of the joint separately to protect the joint by the shielding gas [25,26].The second feature is that the advantages of GMAW and GTAW are combined: the assembly requirements of the base material and the impact of misalignment on welding quality are reduced by the front GMAW; the weld formation is improved further by the cathode cleaning of the back GTAW.The third feature of this method is that the joint are heated at both sides.At the same welding heat input, the welding temperature distribution is different from the common arc welding.There are sufficient welding heat on the joint back to help the forming of the weld by adjusting welding parameters.Also, smaller welding heat input is required to form the joint when the material and thickness of the base mental remain unchanged.The IMCs is restricted more effectively, and the welding residual stress is also decreased or even eliminated.Therefore GMAW/GTAW hybrid welding-brazing becomes a promising way to achieved Ti/Al joints with high quality.
In this study, GMAW/GTAW hybrid welding-brazing was adopted to realize the joints between 5A06 and TC4.Interface bonding strength and loading capacity of joints were taken as the evaluation criteria for interface characteristics and the quality of joints, respectively.The loading capacity and interface bonding strength of joints were characterized by testing two kinds of joints tensile strength (one with reinforcement and the other without reinforcement).A new parameter, form efficiency, was defined to characterize weld reinforcement for evaluating its contribution to the joint quality.In the test, the specimens were obtained by cutting the weldment in the vertical direction to the weld.The metallographic specimens were finished by grinding and sectioning.The weld microstructure was observed by Scanning Electron Microscope (SEM), X-ray diffraction (XRD) and Energy Dispersive Spectrometer (EDS).Effects of main parameters on weld reinforcement and interface characteristics of joints were analyzed.Finally, the welding parameters were optimized by synthetical analyzing of weld reinforcement and interface characteristics.

Experimental materials and methods
In this study, the materials of base metal are alloy sheets of TC4 and 5A06.The thickness of the base mental is 3 mm.The length and width of the base mental are 200 mm and 60 mm.The tensile strengths of TC4 alloy sheet and 5A06 alloy sheet are 920 MPa and 400 MPa, respectively.The Aluminum welding wire is ER4043.Its diameter is 0.8 mm.The electrode of GTAW is the Cerium tungsten electrode.The pure argon is used as the shielding gas.The compositions of main experimental materials in this paper are achieved by chemical analysis methods, as shown in table 1.The welding power sources of GMAW and GTAW are Miller Invision 352 MPa and Miller Syncrowave 350LX, respectively.
Before welding, the base metals were cleaned by acetone and clamped on the workbench suitably.As shown in figure 1(a)), the GMAW/GTAW hybrid welding-brazing was as follows: TC4 and 5A06 alloy sheets were placed on the workbench horizontally.There was no groove on both of them, and a 0.5 mm gap was reserved between the base metals.GMAW and GTAW were placed on the front and back of the alloy sheets separately.The value ranges of all welding parameters were selected after multiple tests firstly.Then main welding parameters and their values were obtained by orthogonal tests.Based on the weld with good face and root reinforcement, each welding parameter was taken as a group.During welding process, only one parameter was changed to study its effects on weld reinforcement and microstructure of TC4/5A06 joints shown in table 2. The other welding parameters were as follows: GTAW position was 0 mm, GMAW wire feeding speed was 9.5 m min −1 , GMAW welding voltage was 13 V, GMAW position was −0.2 mm.In the study, cleaning and fixation of base metal and adjustment of welding parameters were finished manually.While the welding process was completed automatically.Some welding parameters above were explained in figure 1.Along the vertical direction of welding, the distance from the center of the weld to the center of the tungsten was GTAW position.The distance from the center of the weld to the center of the welding wire was GMAW position.GMAW/GTAW position was the distance from the center of the welding wire to the center of the tungsten parallel to the welding direction.There were two directions for the three welding parameters.In order to study their effects, the positive and negative values were defined shown in figures 1(b)) and (c)).
After GMAW/GTAW hybrid welding-brazing, in order to study the parameters influence on TC4/5A06 joints, specimens cut transversely along the weld were prepared, tested and observed.Emery papers from 100 to  2500 grit size were applied to polish metallographic specimens, then they were further polished by diamond paste.The microstructure of joints was observed by SEM.The XRD and EDS were used to analyse the IMCs phase.For the purpose to realize the characterization of the interface bonding strength and loading capacity of the joints, the specimens of tensile tests with reinforcement and the tensile tests without reinforcement were prepared as shown in figure 2. The tensile tests of every joint were carried out 5 times, then the maximum, the minimum and the average values were taken.

Characterization methods
In the present study, the evaluation criterion of joint quality is the joint loading capacity.It is characterized by the joint tensile strength with reinforcement σ 1 shown in figure 2(a)), which takes the effects of weld reinforcement and interface bonding strength on the quality of joints into account.In the tensile test, most butt weldingbrazing joints without reinforcement of TC4/5A06 are fractured at the interface, so the interface bonding strength is characterized by tensile strength of the joint without reinforcement σ 2 shown in figure2(b)).Here, the face and reinforcements are removed by machining to make both sides of the joint flush with the base materials.Therefore, the weld reinforcement is characterized by the equation as follow: In equation (1), f represents form efficiency characterizing weld reinforcement in Ti/Al GMAW/GTAW hybrid welding-brazing, the bigger f is, the greater the contribution of the weld reinforcement to the joint quality is.
1 F 1 is the tension at the fracture of the joint with reinforcement, S 1 is the area of tensile specimen base metal perpendicular to the tensile direction (the cross section area of base mental).
F 2 is the tension at the fracture of the joint without reinforcement, S 2 is the fracture area perpendicular to the tensile direction (the cross section area of base mental), so = S S .

Results and discussion
4.1.The influence of main welding parameters on welding reinforcement and form efficiency For the butt joints between Ti alloys and Al alloys made by the welding-brazing, adequate heat on the back of titanium is a critical factor to form sound joints.Therefore, GTAW current is one of main welding parameters in this study.The heat on the back of TC4 base mental can be changed by adjusting GTAW current.Figure 3 shows the weld reinforcement and form efficiency of the weld obtained under various GTAW currents.As shown, with increasing GTAW current, the wetting of molten Al alloy on the surface of Ti alloy was improved and its spreading on the back of Ti alloy was better.The reason was that when GTAW current was 70A, the heat on back of TiAlloy was so low that the activity of the molten alloy was reduced.And GTAW arc pressure was also low.The liquid metal flows downwards under the action of gravity.So the wetting and spreading of the molten alloy on the back of Ti alloy was reduced.When the GTAW current was increased to 80A, the heat on the back of Ti alloy and GTAW arc pressure were sufficient to improve the wetting and spreading of the molten alloy, thus a sound weld with excellent face and root reinforcement was achieved.When the GTAW current was improved to 90 A, a large number of welding wire and Al alloy are melted.Although GTAW arc pressure was high and the surface tension of molten metal was low, the influence of the gravity was greater because of too much molten metal.This led to more molten metal flowing towards the joint back, while the joint face could not be be supplemented in time.Hence the face reinforcement of the weld became poor.In other words, with GTAW current from 70A to 90A, the form efficiency first increased and then decreased.Figure 4 presents the weld reinforcement and the form efficiency at various GMAW/GTAW positions.It affected the weld reinforcement and the form efficiency of the joints by changing the heat distribution when the welding heat input was constant.As shown, when GMAW/GTAW position was −2 mm, the distance between  GMAW and GTAW was to large, the effects of GTAW on the molten metal were not obvious.So the weld reinforcement, especially the root reinforcement, was poor.With increase of GMAW/GTAW position, the interaction between GMAW and GTAW was increased to improve the activity of the molten alloy.At the same time, GTAW arc pressure on the molten alloy was also improved.Then the root reinforcement was improved.While the GMAW/GTAW position was 0 mm, the effect of the gravity is greater than that of the surface tension and arc force.A large number of molten alloy flowed to the joint back, a shallow undercut at the face surface occurred.That is to say, when GMAW/GTAW position was increased, the form efficiency increased firstly and then decreased.
Effects of welding speed on the weld reinforcement and form efficiency were also investigated shown in figure 5.The welding speed affected the wettability and spreadability of the molten alloy through the changes of the welding heat input on both sides.As shown, the weld reinforcement got worse with the welding speed increasing.It was the same as the changes of form efficiency.The reason was that when the welding speed was 13 mm s −1 , welding heat on both sides of the joint was enough to improve the wettability and spreadability of the molten alloy.But when the welding speed became higher, the welding heat input of both GMAW and GTAW was reduced.So surface tension was high, the arc pressures of GMAW and GTAW were low.These led to the deterioration of the wettability and spreadability of the molten alloy, thus the weld reinforcement got worse.

The influence of main welding parameters on the interface characteristics
From previous studies of Ti/Al welding-brazing, one of the main factors affecting the joint quality is the Ti-Al brazing interface made up of Ti-Al IMCs with different kinds and morphologies [10,11,[13][14][15][16][17], so it is an effective way to acquire sound joints by controlling Ti-Al IMCs forming.Therefore, the Ti-Al IMCs layers under different welding parameters were observed in this paper.Those welding parameters were shown in table 2. And effects of parameters on the brazing interface microstructures were analyzed.Then the relationship between them was established to achieve sound joints of TC4/5A06.As shown in figure 6, under the same process parameters, since the reaction temperature in the middle was lower for the heating of GMAW and GTAW placed at double sides of the weld.The brazing interface IMCs of the middle was thinner than that of the bottom and the top.In the whole, the thickness and the morphology of brazing interface IMCs were determined by welding temperature and the time stating at high temperature.When the welding speed decreased from 17 mm s −1 to 13 mm s −1 , the GTAW current increased from 70A to 90A or GMAW/GTAW position was changed from −2 mm to 0 mm, the heat concentration, the reaction temperature and the reaction time were improved.The IMCs thickness of the brazing interface was increased.At the same time, the morphology of the IMCs also varied from lamellar to serration.Then the EDS analysis of the IMCs was carried out.The results are shown in table 3. From that, we can know the IMCs are mainly composed of aluminum and titanium, and their atomic ratio is close to 3:1.XRD pattern of the joints made at No. 2 in table 2 is shown in figure 7. Combining EDS with XRD, it is deduced that the brazing interface of the weld consisted almost of TiAl 3.According to the thermodynamic, kinetic conditions and the brazing interface products above, the interface reaction process of Ti/Al brazing in this paper is shown in figure 8.It can be divided into four parts: (1) Ti is dissolved in liquid metal continuously.the welding temperature and the time stating at high temperature are improved, the content of Ti along the solid/liquid interface increases.So the condition for the formation of compounds is created.(2) When the solubility of Ti reaches saturation, the reaction of Ti and  Al is finished to form TiAl 3 layer at the front of the solid/liquid interface.
(3) After that, it is the the cooling process.At the interface of the solid/liquid a lot of Ti still exists, so TiAl 3 continues to nucleate and grow along the temperature gradient.(4) Some serrated TiAl 3 grows too fast and break in the high temperature liquid metal.With the further reduction of the welding temperature, the liquid metal solidifies to form a welded joint.
To make clear the effect of the brazing interface IMCs layer on the mechanical properties of TC4/5A06 butt welding joints, the joint tensile strength without reinforcement at the parameters in table 2 was investigated  shown in figure 9. Combining with figure 6, we can see that the joint tensile strength without reinforcement were in accord with the morphology and thickness of IMCs.When a thin and lamellar Ti-Al IMCs was obtained at GTAW current of 70 A, GMAW/GTAW position of −2 mm or welding speed of 17 mm s −1 , the joint tensile strength without reinforcement was low.The reason was that the bonding area between the weld and the brazing interface IMCs was reduced.And when welding speed was 13 mm s −1 , the Ti-Al IMCs so thick that the uniformity of TC4/5A06 joints were destroyed because of the high IMCs brittleness, thus the joint tensile strength without reinforcement was low too.When the welding parameters were appropriate, the IMCs were controlled effectively.In the study, they were welding speed of 15 mm s −1 , GTAW current of 80-90 A and GMAW/GTAW position of −1-0 mm.Serrated IMCs with the 2-6μm thickness were obtained, and the average value of the joint tensile strength without reinforcement was 240.3 MPa with the fracture inside the weld.

Comprehensive analysis of weld reinforcement and the interface characteristics
Based on the analysis results above, we can know that effects of various welding parameters on weld reinforcement are different from that on the interface characteristics.Hence it is necessary to combine weld reinforcement with the interface characteristics to obtain the joint with the excellent quality.
The comprehensive analysis of weld reinforcement and the interface characteristics is shown in figure 10.When welding speed was 17 mm s −1 , GTAW current was 70 A, or GMAW/GTAW position was −2 mm, the TC4/5A06 joint tensile strength was low due to low interfacial bonding strength and poor weld reinforcement.When welding speed was 13 mm/s or GMAW/GTAW position was −0 mm, the TC4/5A06 joint tensile strength was low as well because low interfacial bonding strength or poor weld reinforcement respectively.By comprehensive analysis, the optimization parameters were welding speed of 15 mm s −1 , GTAW current of 80-90 A and GMAW/GTAW position of −1 mm.The average value of the joint tensile strength with reinforcement was 265 MPa. Figure 11 shows the surface and position of the fracture under the optimized welding parameters.As shown in figure 11(a)), lots of tearing ribs were yielded on the fracture surface.It showed that plastic deformation of the joints appeared and then the fracture occurred.In order to further determine the location of the fracture, EDS analysis of the fracture surface marked A was finished.The results showed that zone A was composed of 79.3 at% Al, 14.4 at% Si and 6.2 at% Mg.It was almost consistent with the weld (Mg 2 Si+α-Al+Si) shwon in table 4. The microhardness of the joint was also achieved shown in figure 12(a)).It was measured every 0.2 mm from the fracture position to both sides of the base metals, as shown in figure 11(b)).Then the metallographic structure and microstructure of the weld were also analysed to explain the changes of the joint microhardness shown in figures 12(b)) and (c)).From the the fracture position to TC4, the main microhardness was the weld, so it was little changed.As the IMCs layer was thin, the last microhardness was located on TC4 base metal.From the the fracture position to 5A06, the metallographic structure was divided into four regions: I was the weld equiaxial crystal zone, II was the weld columnar crystal zone, III was HAZ of 5A06 base metal, and IV was the initial structure of 5A06 base metal.As shown figure 12(b)), II and III were small.So the fluctuation of the microhardness was little when the test points were locate on the weld.Then it decreased suddenly from the weld to 5A06 base metal.At last, it fluctuated within a very small range when the test points were locate on 5A06 base metal.From figure 12(c)), we could that B was distributed on matrix C. EDS analysis of B and C was finished shown in table 4. The weld may be composed of α-Al matrix and Mg 2 Si+Al+Si eutectic.And the weld microhardness was improved for the distribution of Mg 2 Si and Si on α-Al.Therefore, the crack of joints under suitable parameters may occur and then extend inside the weld.

Conclusions
GMAW/GTAW hybrid welding-brazing was introduced to join 5A06 to TC4.Effects of welding parameters on weld reinforcement and interface characteristics were investigated.And then the welding parameters were optimized to obtain the sound joints.The results can be summarized as following: (1) Form efficiency, defined to characterize weld, could commendably reflect the contribution of weld reinforcement to the quality of joints.It was depended welding heat inputting and distribution.Excellent weld reinforcement with form efficiency of the weld 8.1% was achieved.
(2) Morphology of IMCs was determined by welding temperature distribution and staying time at high temperature.When the heat concentration or the welding heat input was improved, the IMCs morphology was changed from layered to serrated and its thickness increased.The highest tensile strength of TC4/5A06 joints without reinforcement was 240.3 MPa with serrated IMCs of 2-6μm thickness.
(3) The weld seam consisted of two areas.TiAl 3 IMCs layer was formed in the weld area of the titanium alloy side.The other area of weld was composed of α-Al matrix and Mg 2 Si+Al+Si eutectic.The weld microhardness was improved for the distribution of Mg 2 Si and Si on α-Al.
(4) The quality of joints was determined by weld reinforcement and interface characteristics.However, effects of welding parameters on them were different.By comprehensive analysis, the excellent TC4/5A06 butt joints were obtained.And the average value of the joints tensile strength with reinforcement was 265 MPa with the fracture occurring inside the weld far from IMCs layer.

1 2 σ
b represents the tensile strength of 5A06.

Figure 3 .
Figure 3. Effects of GTAW current on the weld reinforcement and the form efficiency

Figure 4 .
Figure 4. Effects of GMAW/GTAW position on the weld reinforcement and the form efficiency.

Figure 5 .
Figure 5. Effects of welding speed on the weld reinforcement and the form efficiency.

Figure 6 .
Figure 6.Morphology of brazing interface of various welding parameters.

Figure 7 .
Figure 7. XRD pattern of the typical joints made at No.2 in table 2 welding heat input.

Figure 8 .
Figure 8. Formation of interface reaction layer: (a) Dissolution of Ti; (b) Formation of TiAl 3 layer; (c) Growth of TiAl 3 layer; (d) Formation of a welded joint.

Figure 9 .
Figure 9. Effects of welding parameters on the joints tensile strength without reinforcement.

Figure 11 .
Figure 11.Fracture position and surface of the TC4/5A06 joint under the appropriate welding parameter: a) the fracture surface; b) the fracture position.

Figure 12 .
Figure 12.Microhardness of the TC4/5A06 joint under the appropriate welding parameter: a) the joint microhardness; b) the metallographic structure of the weld and 5A06; c) the microstructure of the weld.

Table 3 .
Chemical composition of the typical brazing interface(at %).

Table 4 .
Chemical composition of B and C(at %).