High power fiber laser arc hybrid welding of AZ31B magnesium alloy

High power fiber laser arc hybrid welding of AZ31B magnesium alloy

Materials and Design 42 (2012) 46–54 Contents lists available at SciVerse ScienceDirect Materials and Design journal homepage: www.elsevier.com/loca...

3MB Sizes 3 Downloads 18 Views

Materials and Design 42 (2012) 46–54

Contents lists available at SciVerse ScienceDirect

Materials and Design journal homepage: www.elsevier.com/locate/matdes

High power fiber laser arc hybrid welding of AZ31B magnesium alloy Ming Gao a,⇑, Hai-Guo Tang a, Xiao-Feng Chen b, Xiao-Yan Zeng a a b

Wuhan National Laboratory for Optoelectronics, School of Optoelectronics Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China Research and Development Office, Huazhong University of Science and Technology, Wuhan 430074, PR China

a r t i c l e

i n f o

Article history: Received 11 April 2012 Accepted 18 May 2012 Available online 27 May 2012 Keywords: A. Non-ferrous metals and alloys C. Lasers D. Welding

a b s t r a c t High power fiber laser–metal inert gas arc hybrid welding of AZ31B magnesium alloy was studied. The fusion zone consisted of hexagonal dendrites, where the secondary particle of Al8Mn5 was found at the center of dendrite as a nucleus. Within hybrid weld, the arc zone had coarser grain size and wider partial melted zone compared with the laser zone. The tensile results showed the maximum strength efficiency of 5 mm thick welds was up to 109%, while that of 8 mm thick welds was only 88%. The fracture surface represented a ductile–brittle mixed pattern characterized by dimples and quasi-cleavages. On the fracture surface some metallurgical defects of porosity and MgO inclusions around with secondary cracks were observed. Meanwhile, a strong link between the joint strength and weld porosity were demonstrated by experimental results, whose relevant mechanism was discussed by the laser–arc interaction during hybrid welding. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction Considering the fuel economy in automotive and aerospace industries, magnesium (Mg) and its alloys would be extensively used in the future due to their lightweight and excellent specific strength [1,2]. The welding of Mg alloys is then becoming more and more important. Among the modern welding techniques of Mg alloys, laser welding has been considered more potential because of high welding speed, precise control of power output, narrow joints with reduced heat affected zone (HAZ), low distortion and excellent environment adaptability [3]. For example, Wahaba studied keyhole stability and porosity mechanism in laser welding of AZ series Mg alloys [4,5]. Padmanaban studied the fatigue behavior of CO2 laser welded AZ31 Mg alloy [6]. Scintilla investigated the mechanical properties of Nd:YAG laser welded AZ31 Mg alloy [7]. As a new kind of laser with increased power density, good beam quality, and higher efficiency compared to equivalent power Nd:YAG lasers [8], fiber laser was recommended to join Mg alloys, which showed the tensile strength of welded joints could be stronger than base metal [9,10]. However, the metallurgical defects such as weld porosity and undercut also occurred easily in fiber laser welds of AZ31 [10], AMCa403 [11], AE42 and AS41 [12] Mg alloys. In a way, the occurrence of porosity limited the application of fiber laser welding of Mg alloys. Laser–arc hybrid welding has received significant attention in joining Mg alloys because of higher welding efficiency and stronger ⇑ Corresponding author. Tel.: +86 27 87792404; fax: +86 27 87541423. E-mail address: [email protected] (M. Gao). 0261-3069/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.matdes.2012.05.034

ability to avoid weld metallurgical defects compared with pure laser welding [13]. Liu had first developed low power Nd:YAG laser–tungsten inert gas (TIG) hybrid welding of AZ series Mg alloys [14,15], showing it was effective to suppress weld defects and improve the joint quality. Our group had studied high power CO2 laser–metal inert gas (MIG) hybrid welding of AZ31 Mg alloy and obtained accepted joints with the maximum tensile strength efficiency of 98.7% [16,17]. The results also demonstrated high power laser–MIG hybrid welding would be more potential for Mg industry because of deeper penetration depth, fuller weld appearance and lower weld defects. During CO2 laser–arc hybrid welding, it was found that expensive inert gas of helium must be used to prevent the plasma suppressed effect [18]. However, the helium would increase the instability of arc because the high ionization potential of helium makes the arc ionized more difficultly [19]. Wohlfahrt claimed conventional MIG arc of Mg alloys was essentially instable due to their high vaporization pressure and low vaporization temperature [20]. Therefore, CO2 laser–MIG hybrid welding of Mg alloys is sensitive to instability, although a narrow parameter window was obtained in previous studies [21]. Because of short wavelength of 1.07 lm, fiber laser beam is insensitive to plasma suppressed effect and then pure argon could be used as a shielding gas during fiber laser welding. Accordingly, during fiber laser–MIG hybrid welding, the pure argon shielding gas could be used, which is favorable to ionize and stabilize the arc. It indicates fiber laser hybrid welding is potential to get a more stable process compared with CO2 laser hybrid welding. However, the research on fiber laser–MIG hybrid welding of Mg alloys has not yet been reported. This article


M. Gao et al. / Materials and Design 42 (2012) 46–54 Table 1 Chemical composition (wt.%) of base metal and filler wire. Alloying element









Base metal Filler wire

3.5–4.5 3.0–4.0

0.8–1.4 0.2–0.8

0.3–0.6 0.15–0.5

0.1 0.1

0.05 0.05

0.05 0.05

0.005 0.005

Balance Balance

Fig. 2. Wire feeding rate and arc voltage as a function of arc current.

Fig. 1. Arrangement of experimental set-up.

aims to provide an in-depth understanding of high power fiber laser–arc hybrid welding of AZ31B Mg alloy by analyzing the characterization of welded joints.

2. Experimental equipments and methods An IPG YLR-6000 fiber laser with the maximum power of 6 kW, wavelength of 1070 nm and beam parameter product of 6.9 mm mrad was employed. The laser beam was transmitted by a fiber with 200 lm core diameter and focused by a 250 mm mirror. The welder used was a Panasonic pulse MIG welder with the maximum arc current of 350 A. The base material (BM) was commercial AZ31B wrought Mg alloy. The filler wire was also the AZ31B Mg alloy with the diameter of 1.2 mm. Table 1 shows the chemical compositions of BM and filler wire. Two thicknesses plates (5 mm and 8 mm) were used in current study. Before welding, the plates were milled as 200  75 mm2 and cleaned by acetone. Fig. 1 presents the arrangement of laser beam and welder torch. The incident laser beam was inclined by 10°, and the angle of welder torch to workpiece was 40°. The shielding gas of welder torch was pure argon of 20 l min1. In addition, a gas nozzle using pure argon of 8 l min1 was placed over the weld torch to suppress the climbing of laser induced plume during welding, and a root gas nozzle using pure argon of 10 l min1 was under weld seam to prevent the weld root from oxidizing. In the investigation to reveal the relationships between welding parameters and bead shape, only 8 mm thick plates were welded in bead-on-plate configuration. Where, laser power (P) was in the range of 0.8–3.5 kW, arc current (I) was in the range of 90–200 A, welding speed (v) was in the range of 0.4–3.5 m min1 and the distance between laser beam and wire tip (DLA) is in the range of 1– 6 mm. The wire feeding speed and arc voltage corresponding to each arc current was illustrated in Fig. 2. In the investigation of weld microstructure and mechanical properties, both 5 mm and

8 mm thick samples were welded in butt configuration. Table 2 shows the welding parameters for butt joints. Throughout the study, laser defocused distance kept 0 mm and the DLA kept 3 mm except it was discussed as a variable. After welding, the metallurgical sample was etched by a solution of 4.2 g picric acid, 10 ml acetic acid, 10 ml water and 100 ml ethanol with an etching time of 20 s. The microstructure and fracture surface were observed by scanning electron microscopy (SEM). The chemical compositions were examined by energy dispersive spectrometry (EDS) technique. According to the standard of ISO 6507-1:2005 [22], the Vickers microhardness was measured by a digitalized microhardness tester using a loading force of 100 g (HV0.2) and a load time of 10 s. According to the standard of ISO 6892-1:2009 [23], all the tensile samples including the thickness of 5 mm and 8 mm were first prepared as the drawing in Fig. 3a and tested at the room temperature at a constant travel speed of 1 mm min1. However, some 5 mm-thick joints fractured in the BM were tested again by the arc-notched samples shown in Fig. 3b to reveal the real strength of weld metal. The tensile results were the average of two samples. 3. Results and discussion 3.1. Weld appearance Fig. 4 shows the joints with regular and beautiful appearances are obtained by high power fiber laser–MIG welding, although some pores appear within the weld. The weld shows a ‘wine-cup’ shape. Usually, as shown in Fig. 4c, the wide upper zone is defined as ArcZ because its shape is like that of arc weld, while the narrow lower zone is defined as LaserZ because its shape is like that of laser weld [24]. In Fig. 5a, the penetration depth of hybrid weld is proportional to laser power. Under equivalent laser power, the penetration depth of hybrid weld is higher than that pure laser weld by 10%, indicating the favorable laser–arc synergic effects is achieved during hybrid welding. According to the weld shape, it is interest to observe the welding mode changes from heat conduction mode to laser keyhole mode as the laser power increases to 1.0 kW. In previous CO2 laser–MIG hybrid welding of Mg alloy, the threshold


M. Gao et al. / Materials and Design 42 (2012) 46–54 Table 2 Welding parameters selected for butt joints, where U denotes arc voltage, Q denotes heat input and d denotes plate thickness. Sample

d (mm)

P (kW)

U/I (A/V)

v (m min1)

#1 #2 #3 #4 #5 #6 #7 #8 #9

8 8 8 8 5 5 5 5 5

2.0 2.5 3.0 3.5 1.0 1.5 2.0 2.5 3.0

110/19.4 130/20.4 180/22.6 200/24.4 110/19.4 110/19.4 130/20.4 150/21.4 150/21.4

0.5 0.85 1.5 2.1 0.4 0.85 1.45 1.95 2.4

Q (J mm1) Laser



240 176 120 100 150 105 83 77 75

256 187 163 139 320 150 110 99 80

496 363 283 239 470 255 193 176 155

pen the ArcZ. As shown in Fig. 5c, both the penetration depth and width of hybrid weld decrease with the welding speed because the heat input of both the laser and arc reduces with welding speed. In addition, it was observed that the process stability of hybrid welding would be deteriorated when the welding speed is too fast because the wandering arc root cannot be suppressed and stabilized by laser–arc interaction at this stage. The experimental results demonstrate the welding speed should be limited to lower than 2.4 m min1 to guarantee a stable process. Fig. 5d shows the optimal DLA is in the range of 2–3 mm, by which the weld penetration depth reaches the maximum.

3.2. Microstructure

Fig. 3. Illustrations of tensile samples, (a) standard samples with thickness of 5 mm and 8 mm, (b) arc-notched samples with thickness of 5 mm.

of laser power inducing this welding mode transfer is 2.0 kW [21]. Obviously, fiber laser hybrid welding has more advantages in improving welding efficiency and saving cost because of the higher energy utilization rate. Fig. 5b shows both the width and penetration depth of ArcZ increase with arc current. It is due to the increasing heat input and arc pressure with increasing arc current. The bigger heat input would melt more BM to widen the ArcZ, and meanwhile the stronger arc pressure would push down the melted metal and then dee-

Fig. 6 shows the BM is composed of equiaxed a-Mg grains with the average size of 25 lm. Between the BM and fusion zone (FZ), a partial melted zone (PMZ) characterized by equiaxed a-Mg grains but with some precipitates is found. As shown in Fig. 6a and b, the PMZ of ArcZ within the joint #4 is about 60 lm wide, while that of LaserZ disappears. During hybrid welding, most of the arc heat only acts on the upper molten pool (ArcZ) because of its heat conduction characterization, while the laser beam act on the total molten pool because of the laser keyhole penetrating the weld [25]. That is, the heat of hybrid welding prefers to accumulate in the ArcZ. It results in above phenomenon that within one hybrid joint, the PMZ of the ArcZ is wider than that of LaserZ. However, with increasing laser heat input the PMZ of LaserZ appears and widens gradually as shown in Fig. 6c and d. As shown in Fig. 7, the FZ is composed of equiaxed dendrites exhibiting a typical hexagonal structure where the angle between primary arms is 60°. Bottger [26,27] claimed the formation of this

Fig. 4. Weld appearances of typical joints, (a) #1, (b) #2, (c) #4.

M. Gao et al. / Materials and Design 42 (2012) 46–54


Fig. 5. Effects of welding parameters on weld shape, (a) laser power, (b) arc current, (c) welding speed, (d) distance between laser and arc, DLA.

hexagonal dendrite is due to the hexagonal-close-packed lattice of Mg element. As shown in Fig. 7a and b, the average of semi-axis length of ArcZ and LaserZ within the joint #5 is about 60 and 45 lm respectively, which well agrees with the heat input 320 and 150 J mm1. For different joints with laser heat input in the sequence of 150, 105 and 75 J mm1, as shown in Fig. 7b–d, the average of semi-axis length of LaserZ decreases from 45 to 35, 30 lm. Obviously, the zones with higher heat input prefer to obtain coarse grains. It is because that the corresponding to higher heat input, the molten pool would solidify more slowly, and then the grains has more time to grow up. As arrowed in Fig. 8a and b, some secondary particles are found at the center of hexagonal dendrites as heterogeneous nuclei. The EDS result of point P1 in Fig. 8c shows they are Al–Mn particles. According to Al–Mn binary phase diagram and the molar ratio (nearly 8:5) of Al to Mn in this nucleus particle, they are identified as Al8Mn5 and considered as the un-melted BM Al–Mn inclusions that were still present in the weld pool. This Al8Mn5 particle was also found in resistance-spot-welded and pure laser welded AZ31 Mg joints [9,28]. In conjunction with these Al8Mn5 particles, a large number smaller precipitates arrowed by P2 and P3 in Fig. 8c appear between the dendrite arms. Fig. 8d and e shows they

are rich in Al and Zn element. According to the EDS analysis and Xray diffraction test results [9,10,29,30], these smaller precipitates are usually identified as divorced eutectic Mg17(Al, Zn)12 formed by fast nonequilibrium cooling of welding pool. Usually, the burning-loss of Zn and Mg is usually serious during welding of AZ series Mg alloys because of their low boiling point (Zn-907 °C and Mg-1107 °C) [3]. To reveal this behavior of different zones within hybrid weld, the ArcZ and LaserZ of the joint shown in Fig. 4b were analyzed by EDS. The area A of ArcZ contains 93.65 Mg–2.02Al–1.54Zn–2.79Mn (wt.%), while the area B of LaserZ contains 94.45 Mg–2.26Al–1.73Zn–1.57Mn (wt.%). It indicates the ArcZ has a more serious burning loss of Mg and Zn because of the heat accumulation in this zone. 3.3. Microhardness Fig. 9 presents the microhardness distribution of joint #1 and #4, where the test points of ArcZ are under the weld upper surface of 2.0 mm and those of LaserZ are over the weld bottom of 2.0 mm. With increasing heat input, overall, the FZ microhardness decreases gradually and the HAZ microhardness becomes lower than that of BM, indicating a softening of HAZ as shown in Fig. 9a.


M. Gao et al. / Materials and Design 42 (2012) 46–54

Fig. 6. Microstructure of PMZ, (a) ArcZ of joint #4 with total heat input 239 J mm1, (b) LaserZ of joint #4 with laser heat input 100 J mm1, (c) LaserZ of joint #2 with laser heat input 176 J mm1, (d) LaserZ of joint #1 with laser heat input 240 J mm1.

Fig. 7. Microstructure of FZ, (a) ArcZ of joint #5 with arc heat input 320 J mm1, (b) LaserZ of joint #5 with laser heat input 150 J mm1, (c) LaserZ of joint #6 with laser heat input 105 J mm1, (d) LaserZ of joint #9 with laser heat input 75 J mm1.

Within the joint #4 where laser heat input is lower than arc heat input, the FZ microhardness of LaserZ is slightly higher than that of ArcZ. Within the joint #1 where the heat inputs of laser and arc are almost equal, the FZ microhardness of LaserZ and ArcZ are also almost the same. As discussed in above section, the higher heat input is corresponding to coarser grain size, and then causes a lower microhardness based on the well-known Hall–Petch relationship. Similarly, the HAZ with higher heat input prefers to ob-

tain a lower microhardness due to overheated and recrystallization, resulting in the softening of HAZ. Although scattering exists, the average grain size of FZ and microhardness were measured and drawn in Fig. 10 to reveal their Hall–Petch relationship represented by following formula: 0:5

Hv ¼ 36:7 þ 133:5dg

where Hv is microhardness and dg is grain size.


M. Gao et al. / Materials and Design 42 (2012) 46–54


Fig. 8. EDS analysis of precipitates in the FZ, (a) SEM photo with low magnification, (b) details of the precipitates, (c) P1, (d) P2, (e) P3.

Fig. 9. Microhardness of joint (a) #1 and (b) #4.

3.4. Tensile properties The tensile properties are listed in Table 3. The ultimate tensile strength (UTS) efficiency was calculated by following formulas:

UTS efficiency ¼

Joint UTS  100% BM UTS


It can be seen that some of 5 mm thick welds are stronger than the BM. The biggest UTS efficiency of 5 mm thick welds is up to

109% and the lowest also reaches 88%. Since the maximum UTS efficiency was only 96% in previous CO2 laser hybrid welds of 5 mm thick AZ31 Mg alloy [21], it assumes that fiber laser–MIG hybrid welding is superior to improve the strength of Mg alloys. However, all of the 8-mm thick welds are weaker than BM. Most of the UTS efficiency of 8 mm thick welds is 88% but the lowest is only 76%. Why the UTS efficiencies of 5 mm and 8 mm thick welds have obvious difference will be discussed in following section by fractograph.


M. Gao et al. / Materials and Design 42 (2012) 46–54

Fig. 10. Microhardness as a function of grain size.

Table 3 Tensile properties (#1–#4: 8 mm thick sample, #5–#9: 5 mm thick sample). Sample

#1 #2 #3 #4 #5 #6 #7 #8 #9


Tensile efficiency (%)




188 219 219 224 252 208 218 232 263

192 223 218 217 253 211 221 226 258

190 221 219 221 253 210 220 229 261

76 88 88 88 105 88 92 95 109

Note: The UTS of 5 mm and 8 mm thick BM is 240 MPa and 250 MPa, respectively.

3.5. Fractograph and metallurgical defects As shown in Fig. 11, the fracture surface shows a mixed ductile– brittle feature characterized by quasi-cleavage steps and dimples, and the fraction of dimples to total fracture surface increases with the joint strength. A few secondary particles are found in the dimples as shown in Fig. 12a. The EDS result of area C on this secondary practice shows it contains 42.57 Mg–36.47Al–20.96Mn (at.%), suggesting this particle is Al8Mn5. Fig. 12b shows some pores appear on the fracture surface, while Fig. 12c and d shows many grained inclusions form near the cavity. The EDS result of area D indicates the grained inclusion contains 18.54O–78.88 Mg–2.59Al (at.%). Then, it is identified as MgO. In addition, obvious secondary cracks are found in the Al8Mn5 particles, around the pores and cavities, indicating the joints originally crack at these defects. It is of interest to observe that many pores appear on the fracture surface of the joints weaker than BM, while almost no pores appear on the fracture surface of the joint stronger than BM. As shown in Fig. 13a, for the joint #1 with the lowest UTS efficiency of 76%, the pore diameter is within the range of 1.2–1.56 mm and the porosity (the fraction of pore area to whole fracture surface) is 22.9%. As shown in Fig. 13b, c, e, for the joints with the UTS efficiency of 90% or so, the diameter of most of the pores are lower than 0.8 mm or even smaller and their porosity is within the range of 6.2–8.5%. However, for the joint with the UTS efficiency more than 100%, as shown in Fig. 13d and f, the fracture surface is pore free except for few micro-pores with the diameter lower than 0.15 mm. Obviously, there is a strong link between the porosity and joint strength: the higher the porosity, the lower the UTS efficiency. Because 8 mm thick welds could not obtain

Fig. 11. Micro SEM photos of fracture surface of typical joints with different strength, (a) #6, (b) #9.

pore free weld although all possible welding parameters were tried, their maximum UTS efficiency is only 88%. During the hybrid welding, the arc plays a big role in the pore reduction by increasing the heat input and speeding the melt flow in the molten pool [17], but its impact depth of the arc is limited due to its low power density and heat conduction characteristic. By this viewpoint, why 5 mm thick hybrid welds can obtain the pore free welds but 8 mm thick welds cannot is explained as following. When the plate is 5 mm thick, the large portion of the weld is composed of ArcZ as shown in Fig. 14a where both the depths of LaserZ and ArcZ are 3 mm. It indicates the effects of arc heat and arc pressure cover most of the molten pool. In this stage, the arc plasma could entrap into laser keyhole to stabilize the keyhole, which reduces the sensitivity of bubble formation in the molten pool [16,17]. Meanwhile, as shown in Fig. 14b, the bubbles formed in LaserZ could float into the wider ArcZ easily by the melt flow and subsequently escape into the air. Consequently, the 5 mm thick samples could obtain pore free welds by laser–arc synergic effects during hybrid welding. When the plate is 8 mm thick, however, the ArcZ becomes the small portion of the welds as shown in Fig. 14c where the depth of ArcZ and LaserZ is 3.8 mm and 5.5 mm, respectively. It indicates the arc impact depth almost keeps constant although the depth of laser keyhole increases by 1.6 times. Then, the deeper part of LaserZ is hard to be effected by the arc, and

M. Gao et al. / Materials and Design 42 (2012) 46–54


Fig. 12. Typical defects found on fracture surface, (a) particle in the dimple, (b) pores, (c) secondary cracks, (d) oxidized inclusions.

Fig. 13. Marco SEM photos of pores arrowed on fracture surface, (a) #1, (b) #2, (c) #4, (d) #5, (e) #7, (f) #9.

prefers to keep the characteristics of pure laser welding which is sensitive to the bubble formation in the molten pool. Though some of the bubbles could escape into the air by the melt flow during welding, many bubbles are trapped by the solidifying front to form weld pores due to the long escaping distance caused by the narrow and deep LaserZ as shown in Fig. 14d. 4. Conclusions High power fiber laser–arc hybrid welding of AZ31B Mg alloy was carried out in current study. The characterization of welded joints was analyzed. Major conclusions of this study could be summarized as follows.

(1) Accepted AZ31B Mg alloy joints were obtained by high power fiber laser–MIG hybrid welding. It was found that the threshold of laser power changing the welding from heat conduction mode to keyhole mode was 1.0 kW, and the welding speed should be lower than 2.4 m min1 to guarantee the process stability. (2) In the FZ, the secondary particle of Al8Mn5 was found at the center of hexagonal dendrites as a nucleus, while the smaller precipitates of Mg17(Al, Zn)12 were found between the dendrite arms. The grain size of FZ and the width of PMZ both increase with heat input, and within one hybrid weld ArcZ has coarser grain size and wider PMZ width compared with LaserZ.


M. Gao et al. / Materials and Design 42 (2012) 46–54

Fig. 14. Characteristic comparison of 5 mm and 8 mm thick joints, (a) weld cross-section of 5 mm-thick weld, (b) schematic molten pool of 5 mm-thick weld, (c) weld crosssection of 8 mm-thick weld, (d) schematic molten pool of 8 mm-thick weld.

(3) Increasing heat input increased the softening of HAZ and meanwhile decreased the FZ microhardness. The Hall–Petch relationship between the microhardness and the grain size was formulated as Hv = 36.7 + 133.5dg0.5. (4) The pore free weld could be obtained as the sample is 5 mm thick, but it could not be obtained as the sample is 8 mm thick. It was found that the joint UTS efficiency depends on weld porosity: the higher the porosity, the bigger the UTS efficiency. Accordingly, the maximum UTS efficiency of 5 mm-thick welds was up to 109% because of the pore-free, while that of 8 mm-thick welds was only 88% due to the high porosity.

References [1] Friedrich HE, Mordike BL. Magnesium Technol. 1st ed. Berlin: Springer; 2006. [2] Luo AA. Magnesium: Current and potential automotive applications. JOM 2002;54:42–8. [3] Cao X, Jahazi M, Immarigeon JP, Wallace W. A review of laser welding techniques for magnesium alloy. J Mater Process Technol 2006;171:188–203. [4] Wahaba M, Mizutani M, Kawahito Y, Katayama S. Keyhole stability in disc laser welding of AZ31B and AZ61A magnesium alloys and weld metal properties. Sci Technol Weld Join 2010;15:559–66. [5] Wahaba M, Mizutani M, Kawahito Y, Katayama S. Laser welding of die-cast AZ91D magnesium alloy. Mater Des 2012;33:569–76. [6] Padmanaban G, Balasubramanian V, Madhusudhan Reddy G. Fatigue crack growth behaviour of pulsed current gas tungsten arc, friction stir and laser beam welded AZ31B magnesium alloy joints. J Mater Process Technol 2011;211:1224–33. [7] Scintilla LD, Tricarico L, Brandizzi M, Satriano AA. Nd:YAG laser weldability and mechanical properties of AZ31 magnesium alloy butt joints. J Mater Process Technol 2010;210:2206–14. [8] Quintino L, Costa A, Miranda R, Yapp D, Kumar V, Kong CJ. Welding with high power fiber lasers – a preliminary study. Mater Des 2007;28:1231–7. [9] Chowdhury SM, Chen DL, Bhole SD, Powidajko E. Microstructure and mechanical properties of fiber-laser-welded and diode-laser-welded AZ31 magnesium alloy. Metall Mater Trans A 2011;42A:1974–89. [10] Wang ZM, Gao M, Tang HG, Zeng XY. Characterization of AZ31B wrought magnesium alloy joints welded by high power fiber laser. Mater Charact 2011;62:943–51. [11] Yu L, Nakata K, Liao J. Weld porosity in fibre laser weld of thixomolded heat resistant Mg alloys. Sci Technol Weld Join 2009;14:554–8.

[12] Liu J, Dong JH, Wu YJ. Analysis of welding procedure and welded joint’s microstructure of fiber laser welding Mg alloy. J Mater Eng 2009(1):176–9 [in Chinese]. [13] Bagger C, Olsen FO. Review of laser hybrid welding. J Laser Appl 2005;17:2–14. [14] Liu LM, Wang JF, Song G. Hybrid laser-TIG welding, laser beam welding and gas tungsten arc welding of AZ31B magnesium alloy. Mater Sci Eng A 2004;381:129–33. [15] Liu LM, Song G, Zhu ML. Low-power laser-arc hybrid welding behavior in AZbased Mg alloys. Metall Mater Trans A 2008;39A:1702–11. [16] Gao M, Cao Y, Zeng XY, Lin TX. Mechanical properties and microstructures of hybrid laser MIG welded dissimilar Mg–Al–Zn alloys. Sci Technol Weld Join 2010;15:638–45. [17] Gao M, Zeng XY, Tan B. Study of laser MIG hybrid welded AZ31 magnesium alloy. Sci Technol Weld Join 2009;14(4):274–81. [18] Gao M, Zeng XY, Hu QW. Effects of gas shielding parameters on weld penetration of CO2 laser–TIG hybrid welding. J Mater Process Technol 2007;184:177–82. [19] Campana G, Ascari A, Fortunato A. Hybrid laser–MIG welding of aluminum alloys: the influence of shielding gas. Appl Surf Sci 2009;255:5588–90. [20] Wohlfahrt H, Rethmeier M, Bouaifi HB. Metal-inert gas welding of magnesium alloy. Weld Cutt 2003;55:80–4. [21] Gao M, Mei SW, Wang ZM, Li XY. Process and joint characterizations of laser– MIG hybrid welding of AZ31 magnesium alloy. J Mater Process Technol 2012;212:1338–46. [22] ISO 6507–1:2005. Metallic materials–Vickers hardness test – Part 1: test method. ISO; 2005. [23] ISO 6892–1:2009. Metallic materials–Tensile testing – Part 1: method of test at room temperature. ISO; 2009. [24] EL Rayes M, Walz C, Sepold G. The influence of various hybrid welding parameters on bead geometry. Weld J 2004;83:147s–53s. [25] Mahrle A, Beyer E. Hybrid laser beam welding – classification, characteristics, and applications. J Laser Appl 2006;18:169–80. [26] Bottger B, Eiken J, Munekazu O. Controlling microstructure in magnesium alloys: a combined thermodynamic experimental and simulation approach. Adv Eng Mater 2006;8(4):241–7. [27] Bottger B, Eiken J, Steinbach I. Phase field simulation of equiaxed solidification in technical alloys. Acta Mater 2006;54(4):2697–704. [28] Xiao L, Liu L, Zhou Y, Esmaeili S. Resistance-spot-welded AZ31 magnesium alloys: Part I. Dependence of fusion zone microstructures on second-phase particles. Metall Mater Trans A 2010;41A:1511–22. [29] Rose Razal A, Manisekar K, Balasubramanian V, Rajakumar S. Prediction and optimization of pulsed current tungsten inert gas welding parameters to attain maximum tensile strength in AZ61A magnesium alloy. Mater Des 2012;37:334–48. [30] Quan YJ, Chen ZH, Yu ZH. Characteristics of laser welded wrought Mg–Al–Zn alloy. Mater Charact 2008;59:1799–804.