Journal of Materials Processing Technology 245 (2017) 46–69

Contents lists available at ScienceDirect

Journal of Materials Processing Technology

journal homepage: www.elsevier.com/locate/jmatprotec

Prospects of laser welding technology in the automotive industry: A

review
Kyung-Min Hong, Yung C. Shin ∗
Center for Laser-based Manufacturing, School of Mechanical Engineering, Purdue University, West Lafayette, IN, USA

a r t i c l e i n f o a b s t r a c t

Article history: The prospects for the use of lightweight materials (aluminum alloys, magnesium alloys, and titanium
Received 8 December 2016 alloys) in high volume vehicle manufacturing are discussed. Laser welding of galvanized steel is com-
Received in revised form 14 February 2017 pared to resistance spot welding of galvanized steel, along with ongoing efforts to improve the quality of
Accepted 15 February 2017
laser welding of galvanized steel by altering the weld configuration, changing the element composition,
Available online 20 February 2017
utilizing a pulse laser, and removing the zinc coating. The feasibility of implementing these techniques in
the industrial setup is discussed. Microstructure changes and defects encountered during laser welding
Keywords:
of these materials are described, and mechanical properties of welds such as hardness, shear and tensile
Laser welding
Keyhole welding
strength are analyzed.
Galvanized steel © 2017 Elsevier B.V. All rights reserved.
Aluminum alloys
Magnesium alloys
Titanium alloys
Porosity
Solidification cracking
Intermetallic
Dissimilar materials

1. Introduction materials and designs are implemented into chassis development,

the need for more advanced joining methods is growing.
Aslanlar (2006) reported that in the automotive industry, one Holliday et al. (1996) indicated that resistance spot welding of
of the oldest and most widely used processes in body-in-white galvanized steel sheets is known to incur high maintenance costs
(BIW) manufacturing is resistance spot welding. Chao (2003a, b) for electrodes, due to the higher current, the longer time required
stated that the number of spot welds in a single vehicle can range for welding, and the tendency of zinc coating to stick to the elec-
from 2000 to 5000, which signifies the importance of resistance trodes. An experimental study by Zhang et al. (2008) showed that
spot welding in automotive assembly. Zuniga and Sheppard (1997) electrode-face diameter enlargement, both axially and radially, was
pointed out that vehicle crashworthiness depends on the integrity much faster for welding zinc coated DP600 than for uncoated steels.
of the individual spot welds. Donders et al. (2005, 2006) described The electrode diameter was enlarged by 1.9 mm after 1200 welds
that the individual spot welds can transmit dynamic push and pull on a 0.8 mm thick DP600 steel, while the electrode diameter was
forces, as well as shear moments, which heavily influence the noise, enlarged by only 0.52 mm after 1500 welds on a non-coated steel.
vibration, and hardness (NVH) of automobiles. Despite the problems with resistance spot welding of galvanized
Jou (2003) mentioned that the heavy reliance on resistance spot steel, the change was unavoidable, as zinc coating provides supe-
welding in the automotive industry is due to the simplicity of the rior resistance to corrosion compared to non-coated steel sheets.
welding mechanism and the ease of implementation in robotic Fujita and Mizuno (2007) reported that manufacturing of parts with
systems. Pouranvari and Marashi (2013) pointed out that the high- galvanized steel has allowed major European automakers to pro-
output low operation cost in mass production of resistance welding vide voluntary warranties for corrosion protection lasting as long
makes this a suitable choice for the automotive industry. As newer as 12 years.
In addition to galvanized steel, interest in alternative
lightweight materials has been growing in the automotive indus-
try. Reducing the overall weight of a vehicle is the most effective
∗ Corresponding author. way to increase fuel economy without losses in performance. One
E-mail address: [email protected] (Y.C. Shin).
of the most promising lightweight materials is aluminum, due to

http://dx.doi.org/10.1016/j.jmatprotec.2017.02.008
0924-0136/© 2017 Elsevier B.V. All rights reserved.
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 47

its high strength-to-weight ratio. Miller et al. (2000) proclaimed chassis. It is important to note that research studies conducted by
that Audi was able to reduce body weight by 40% in manufactur- Tzeng (2006) and Graham et al. (1994) show that the defects are not
ing the A8, by employing an aluminum-intensive space frame. Even limited to a specific type of laser used for welding, but arise with
high volume vehicles such as Ford’s F-150 began featuring an all- every kind of laser, including CO2 , Nd:YAG and fiber lasers. There-
aluminum body and cargo box, as of 2015. Joining aluminum using fore, in order for the automotive industry to replace conventional
resistant spot welding presents a challenge as electrode tip wear is resistance spot welding with laser welding, it is crucial to find a
even worse than the welding of galvanized steel. Matsumoto and solution to this problem in a cost effective manner.
Mochizuki (1994) pointed out that compared to welding of steel, It is also important to take into account the strong interest in
aluminum requires an electric current three times higher, which developing lightweight vehicles in the automotive industry. Cur-
results in a tenfold reduction in the electrode’s lifetime. rently, most manufacturers depend on developing efficient chassis
Consequently, there has been a large amount of research in find- design through the utilization of different grades of advanced high-
ing better performing alternative joining methods. Of the many strength steel, rather than using alternative materials, due to cost
joining processes available, the high energy density and small heat restrictions. However, once a limitation to efficient design is rec-
affected zone of keyhole laser welding makes this method most ognized, and prices of alternative materials reduce, lightweight
attractive. Moreover, through technological advancements, it has materials, such as aluminum alloys, magnesium alloys, and tita-
become easier to acquire laser equipment with sufficient power nium alloys will increasingly replace conventional materials.
density for use in laser keyhole welding. Quintino et al. (2007) The aim of this paper is to review the recent progress in laser
reported that the new fiber lasers are more efficient than lamp or welding of zinc coated steel, magnesium alloys, aluminum alloys,
diode pumped rod lasers and have a compact design and good beam titanium alloys, and dissimilar materials. Less emphasis will be put
quality. Multi-kilowatt power levels can be achieved by combining on discussing optimized laser welding parameters (laser power,
the output of several single mode fiber lasers. The study conducted welding speed, focal plane position, shielding gas condition, surface
by Assunção et al. (2010) showed that fiber lasers give the highest preparation) due to their great variability. Instead, this paper will
welding speeds and lowest cost per hour compared to Nd:YAG and focus on the metallurgical defects encountered in laser welding and
CO2 lasers. critically review recent works that suggest solutions.
In laser welding, two regimes are possible depending on the
laser power density: keyhole mode welding produced by high
2. Materials for consideration
power densities (typically above 106 W/cm2 ), and characterized
by strong evaporation and cavity; and conduction mode welding
Before discussing various joining mechanisms, it is important to
with low power densities. Tobar et al. (2010) have pointed out that
identify the materials that have potential for use in vehicle manu-
compared to keyhole welding conduction welding is more robust
facturing, as different materials require different processes. Davies
against defects as its melt pool is stable; however, the maximum
(2012) listed the main criteria in material selection for automobile
penetration depth of keyhole welding is far superior to that of con-
assembly as ease of manufacturing, environmental friendliness,
duction welding. A high penetration depth is possible for keyhole
cost, design parameters, and manufacturing consistency. Consider-
welding because laser energy is transferred deep into the keyhole
ing these criteria, the material of choice in the automotive industry
bottom rather than onto the surface. Ki et al. (2002) were able to
has been steel. This predominant usage of steel can be attributed
model the laser keyhole welding process, which showed the laser
to the low cost and consistency of supply. Therefore, developments
beam reaching the keyhole bottom through multiple reflections on
in material joining mechanisms have been focused on high-tensile
the keyhole wall. This results in an increase in the overall energy
steels. Yang and Lee (1999) reported that as the automobile indus-
transfer from the laser to the workpiece. The high penetration depth
try is aiming for lightweight construction to comply with emission
of keyhole welding is especially advantageous in the overlap weld-
legislation and environmental requirements, steel is steadily being
ing configuration, where high penetration depth is required. In
replaced by alternative materials. As the industry replaces steel
addition, keyhole laser welding has a narrow heat affected zone,
with lightweight materials, research into joining mechanisms for
which minimizes the area that incurs phase changes.
both steel and alternative materials needs to coexist for a smooth
According to Optech Consulting (2013), the automotive sec-
transition. In another words, research studies should be conducted
tor’s consumption of the global laser system accounts only for
on both improving the current welding process for steel and devel-
disproportionately low 15% as of 2012. The reason for the low
oping new methods for joining alternative materials. In this section
implementation rate of laser welding is due to several challenges
materials such as magnesium, aluminum and titanium are inves-
that have not yet been overcome. Of the various challenges, the
tigated to determine the feasibility of replacing steel in vehicle
main difficulty is attributed to the galvanized steel used in the chas-
manufacturing.
sis assembly. While the zinc coating provides excellent protection
during the use of automobiles, it creates difficulties in the weld-
ing of overlapping configurations because of the low vaporization 2.1. Magnesium
temperature of zinc (Tvapor,Zn = 1180 K) compared to the melting
temperature of steel (Tmelt,Fe = 1800 K). More specifically, during Magnesium-based alloy is a promising material for reducing
keyhole welding, zinc, which is sandwiched between the metal vehicle weight, as it has a lower mass density (1.74 g/cm3 ) than
plates, begins to evaporate after the beam irradiates the surface. aluminum. However, as pointed out by Blawert et al. (2004) and
Without a gap between the plates, the least resistive way for the Mordike and Ebert (2001), significant galvanic corrosion and poor
vapors to escape is through the melt pool. Schmidt et al. (2008) creep resistance at high operating temperatures have deterred
were able to observe the degassing process through a high-speed companies from utilizing magnesium in automotive applications.
camera recording. Typically, zinc vapor is degassed through a chan- Particularly for components with operating temperatures exceed-
nel generated at a range of approximately 1 mm to 4 mm behind ing 100 ◦ C, such as engine castings, magnesium is not applicable
the keyhole wall. without special treatment. Koike et al. (2003) reported that even at
The high dynamic pressure of zinc causes instability in the melt low temperatures, creep is known to occur in magnesium within
pool, resulting in blowouts and porosity. The existence of poros- the grains and sub-grains through basal slip. McQueen et al. (2000)
ity results in low tensile strength and premature fatigue failure, pointed out that forming magnesium sheets requires an elevated
which are detrimental to the structural integrity of an automobile temperature to activate the additional slip systems (101̄1 pyramid
48 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Fig. 1. Development strategy of magnesium components.

Reproduced with permission from Friedrich and Schumann (2001), © Elsevier 2001.

planes) of the hexagonal structure of magnesium. Froes et al. (1998)

indicated that the cost of magnesium is approximately 1.9 times the
cost of aluminum, significantly increasing the cost of production.
As a result, research has been conducted in improving mag-
nesium processing, alloy development, and corrosion protection.
Schumann (2005) reported that the first development in creep
resistant magnesium for automotive applications was the Mg-Al-Si
alloy (AS21) produced by Volkswagen AG in the 1970s. In recent
years, BMW (2007) successfully manufactured die-cast engine
blocks based on Mg-Al-Sr alloys that can withstand high opera-
tion temperatures. How the addition of rare earth elements affects
the mechanical properties of magnesium alloys has been studied by
Jain and Koo (2007). Their research showed that the addition of Rhe-
nium to Mg-8Al alloy suppressed the precipitates of the thermally
less stable ␤ phase at high temperatures. Ishizaki et al. (2009) were
able to reduce corrosion on the AZ31 magnesium alloy by applying
a single-phase system in the form of newberyite coating.
Most applications of magnesium alloys have been limited to the
interior of vehicles, where the components are less susceptible to
corrosion. Friedrich and Schumann (2001) reported on the use of
magnesium alloys for high-load safety-related components, such Fig. 2. Aluminum alloys used for the Honda NSX.
as steering wheels and seat structures, as well as low load compo- Reproduced with permission from Muraoka and Miyaoka (1993), © Elsevier 1993.

nents like instrument panels. On the contrary there has been very
limited success in using magnesium alloys on outer body panel
components. The Volkswagen AG’s strategy of magnesium tech- such as engine blocks, transmission casings, and wheels (Becker,
nology development shows that it is more challenging to produce 1999). Kelkar et al. (2001) claimed that the utilization of aluminum
magnesium sheets for use in exterior components (Fig. 1). Kainer alloys as a replacement for steel in body in white (BIW) assembly
et al. (2015) reported the successful employment of a magnesium promises the most weight reduction, as BIW can account for 27% of
alloy roof, but the application is limited to ultra-niche products such vehicle mass. For example, Muraoka and Miyaoka (1993) reported
as the Porsche 911. To achieve mass production of magnesium alloy that the BIW of the Honda NSX, which was manufactured with alu-
components, relevant processes such efficient joining technology minum, was 140 kg lighter the than BIW manufactured with the
will need to be developed. steel of an equivalent specific strength and specific rigidity (Fig. 2).
In the Honda NSX BIW assembly, the frame components were made
by variable thickness extrusion and the sheet components were
2.2. Aluminum made by stretch forming.
In addition, the research conducted by Ito and Kobayashi (2006)
Among lightweight materials, aluminum has gained traction as shows that porous aluminum yields favorable results for structural
a candidate to replace steel sheet as it satisfies the torsional and elements, such as front side members, b pillars, and crash boxes,
stiffness requirements for chassis construction. Utilization of alu- as it shows small load fluctuations in crash tests. As presented by
minum has been focused mostly on casting individual components Dieffenbach (1999), the two most popular approaches in build-
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 49

Fig. 3. Laser welding used in hydroforming of aluminum sheets.

Reproduced with permission from Novotny and Geiger (2003), © Elsevier 2003.

ing aluminum vehicle chassis are space frame construction, with the automotive industry. In resistance spot welding, sheet metal is
extruded aluminum, and unibody construction with stamped body positioned in an overlapping configuration and electrodes come
parts. Between the two methods, it is more cost effective to utilize in contact with the outside of the sheets. Heat is generated by
stamping for large-scale production. passing a large electrical current between the electrodes. Williams
For producing hollow aluminum automotive parts, Novotny and and Parker (2004) reported that the Joule’s first law as shown in
Geiger (2003) proposed the process of using hydroforming and equation (1) can express the heat generated by the electrodes:
laser welding. The new process, as shown in Fig. 3, simplifies the
conventional process of several deep drawing, trimming, position- H = R · I2 · t (1)
ing, and joining steps.
In current production, as an intermediate and cost-effective
where H is heat output in joules, I is current in amperes, R is resis-
method, steel and aluminum hybrid structures have been adopted
tance in ohms and t is time in seconds. To ensure the appropriate
for BIW construction. For example, some vehicles use aluminum
flow of current, pressure is applied to the copper alloy electrodes.
alloys in nonstructural outer panels, such as vehicle hoods, fend-
The electrode force and current density parameters are controlled
ers, and truck lids. The advantage of using aluminum alloys in
with the goal of producing molten metal without expulsion.
these individual components is the utilization of conventional
joining methods used in current manufacturing lines. These com-
ponents are designed to attach to the body skeleton using bolts. As
3.1. Resistance spot welding of lightweight alloys
aluminum prices decrease and efficient processes are developed,
an increased adoption of aluminum in BIW construction can be
High maintenance is required for spot welding of aluminum
expected. Consequently, further research on welding is necessary
alloys. Deposits of aluminum form easily on the electrode tips, and
to improve the manufacturability of aluminum BIW construction.
as a result, the electrodes require frequent dressing or replacement.
Muraoka and Miyaoka (1993) stated that the accelerated wear is
2.3. Titanium due to the higher instantaneous current requirement of aluminum:
20,000–50,000 A, compared to 7000–12,000 A for steels. In addi-
Titanium is another material that is considered for use in the tion, higher pressing forces are required from the electrodes for
automobile assembly. The high cost of titanium has however welding of aluminum, as the body panels are generally thicker than
restricted its application in automobiles to a small number of com- steel. Mathers (2002) reported that in extreme cases where abso-
ponents, despite titanium’s high strength-to-weight ratio and good lute welding quality is required, such as in the aerospace industry,
corrosion resistance (Faller and Froes, 2001). Froes et al. (2004) it is a common practice to clean electrodes after every 20 spot welds
reported that the cost of titanium is high due to the large amount on aluminum alloys.
of energy required to separate titanium and oxygen atoms. Tita- Luo et al. (2011) carried out resistance spot welding on magne-
nium alloys are used only in components where the benefits can sium alloys AZ31B and AZ91D. Their study found that liquidation
be maximized. Faller (2002) stated that titanium is used in rotat- cracking and surface expulsion were prevalent in welding of AZ31B
ing and reciprocating masses such as connecting rods, valves, and and a large number of voids and cracks were created in the nugget
camshafts, as the characteristic high stiffness at elevated tempera- for AZ91D. One of the physical properties of magnesium that causes
tures allows the engine to operate at high speeds. Schauerte (2003) a problem in resistance spot welding is the large coefficient of ther-
reported on the examples of titanium alloys being used such as mal expansion, which results in large deformation, distortion, and
exhaust systems, suspension components, and brake systems. thermal stress after welding. The most significant issue arose from
the expulsion of the material in all radial directions at the faying
3. Resistance spot welding interface, which shortens the life of the electrodes on the weld-
ing machine. In conclusion, resistance spot welding of magnesium
Resistance spot welding, which emerged in the 1950s, has been alloys is not a recommended method because of both the weld
the principal method of choice for joining vehicle components in quality and the accelerated wear of the electrodes.
50 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Table 1 esthetics provided by high quality laser seam welding secondary

Welding time for resistance spot welding (Papkala, 1992).
processes such as grinding and finishing can be eliminated.
Metal sheet thickness (mm) Welding speed (welding spots/min)

0.7–0.8 40–60
0.9–1.0 30–50 4. Laser welding of galvanized steel
1.0–1.2 30–40
1.25–1.6 20–30 As mentioned previously, laser welding has many advantages in
1.6–2.0 15–25
automobile assembly such as vehicle weight reduction, esthetically
pleasing surface finishes, and the non-contact nature of welding.
However, similar to the problems of spot welding of zinc coated
steel sheets, the low vaporization temperature of the zinc coat-
3.2. Resistance spot welding vs. laser welding ing of galvanized steel creates problems in laser lap welding. The
vaporized zinc coating at the faying surface can cause disruption
One of the distinctive features of laser welding is the non- of the keyhole stability, and spatter and void formation will also
contact nature of welding. Buehrle et al. (2013) pointed out that be inevitable. The keyhole behavior during laser welding of zinc
laser remote welding has the advantages of low processing time coated steels has been investigated by Kim et al. (2015). The top sur-
(via reduction of index time) and a reduced number of clamping face images of the keyhole motion were analyzed, which revealed
fixtures. Laser welding can produce custom shape welds to maxi- that the zinc coated steel exhibited more instability, in terms of
mize structural integrity, while resistance spot welding is limited keyhole diameter change, than non-coated steel (blue markers in
to only spot welds. In addition, multi-kilowatt laser keyhole weld- Fig. 4). Fabbro et al. (2006) reported that the saturation pressure of
ing is known to produce deep penetrating welds. The difference in zinc varies from about 50 bars to 100 bars at temperatures rang-
processing time between laser and resistance spot welding is more ing from 1800 K to 2000 K. Kim et al. (2016) revealed that the laser
apparent when joining thicker metals as welding time increases beam position relative to the keyhole position was a critical factor
drastically for resistance spot welding (Table 1). in obtaining sound welds. When the laser beam is positioned on the
Laser welding also has the advantage of reducing the overall front keyhole wall, the zinc coating can be directly heated causing
weight of the vehicle by reducing flange widths. Resistance spot the zinc vapor to reach the critical point temperature of 2939 K. At
welding usually requires flanges about 16 mm in width for spot the critical point temperature, Martynyuk (1983) reported that the
gun access, which adds weight and material cost to non-structural corresponding critical point pressure of zinc is 2460 bar. The large
features. In contrast, the flange width for laser welding can be sig- vapor pressure will definitely disturb the intricate keyhole pressure
nificantly reduced because of the non-contact process. In some balance among evaporation recoil pressure, hydrostatic pressure,
areas, flanges can be eliminated by using laser welding in a butt and capillary pressure.
welding configuration. Most importantly, the single access feature Although a lot of research has been devoted to finding solutions
of laser welding can increase the torsional rigidity by eliminating to the instability of zinc coating, only a few studies have been con-
the access openings that are necessary for resistance spot welding. ducted on analyzing the effects of laser type. Hiraga et al. (2002)
A study conducted by Klinger (2012) showed that simply convert- found that the wavelengths of the laser source affect the laser beam
ing from a spot welded structure to a laser welded structure allowed absorption on the workpiece, as well as on the plasma plume. Tan
a mass reduction of 12.2 kg in BIW. In addition, Barnes and Pashby and Shin (2014) reported that metal particles formed in the plume
(2000) pointed out that highly focused and high energy density by condensation will absorb and scatter the incident laser beam,
laser beam induces very little thermal distortion. Lastly, due to the which would otherwise be absorbed by the substrate. The amount

Fig. 4. Keyhole diameter measurement vs. frame number (a) DP 590 steel, (b) galvanized DP 590 steel.
Reproduced with permission from Kim et al. (2015), © Elsevier 2015.
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 51

of absorption and scattering by the metal particles is lower for lasers sheets. The key parameter in this welding operation is the offset of
with a lower wavelength, such as fiber lasers. the laser beam, relative to the origin in the horizontal direction.
Mei et al. (2013) found that, in comparison to the CO2 laser, Pieters et al. (2006) proposed an edge lap configuration with-
the fiber laser weld resulted in higher tensile strength and hard- out a gap between the plates. The continuous wave (CW) Nd:YAG
ness. This result can be attributed to the longer wavelength of the laser was configured to target the laser beam onto the edge of the
CO2 laser, which increases the absorption of laser energy into the plates with an incident angle of 45◦ . This configuration allowed the
plasma rather than into the workpiece. Compared to CO2 lasers, zinc vapor to escape through an alternative route, and the amount
fiber lasers are more suitable for the automotive industry as they of spatter was reduced. However, the edge lap configuration is
are more energy efficient, and, as mentioned by Jeong et al. (2004), severely limited in that it can only be joined at the ends of the
high optical power can be transmitted through fiber-optic cables. plates.
Therefore, the main frame of the fiber laser system can be located Chen et al. (2009) suggested pre-drilling vent holes in the bot-
remotely, which increases flexibility in production line set-up. tom sheet using a laser beam. Holes with a nominal diameter of
There has been a lot of research conducted on the subject of 70 ␮m were drilled into the bottom plate using an Nd:YAG laser
finding the solutions to the problems of laser welding related to with an average power of 200 W and a pulse width of 1 ms. The mul-
zinc coating. In the next section, the welding techniques will be tiple vent holes were aligned along the weld seam and reduced the
discussed in these categories: weld configuration based method, pressure buildup of zinc vapor. The critical parameter that affected
change in composition, pulsed lasers, and removal of zinc coating. the welding quality was the distance between the vent holes. Com-
pared to no-gap overlap configurations, welding with the vent holes
4.1. Weld configuration based method resulted in less porosity and higher tensile strengths.
Iqbal et al. (2010) proposed the use of dual laser beams to
Creating an appropriate gap in the overlap joint between the sequentially cut an exhaust slot in the plate and perform the fusion
faying surfaces of the galvanized steel sheets can provide an alter- of top and bottom metal sheets. The exhaust slots were produced
native channel through which the evaporated zinc can readily in alignment with the weld seam. The novelty of this solution is the
escape. This method shows good welding quality because the vapor reduced overall process time by simultaneous cutting and welding.
can de-gas laterally and not disturb the weld pool. Akhter et al. The authors proposed a minimum slot width requirement, based on
(1989) proposed an equation, based on the volume of zinc vapor Bernoulli’s theorem, that provides a sufficient gap for zinc vapor
generated between the sheets that calculates the optimal gap venting as described by:
between the faying surfaces. 
w tp
−1/2 g = w + 2b − 0.5145 (3)
g = K · V · tzn · tp (2) v
where w is the keyhole width, b is the width of the zinc boiling
In Eq. (2), K is a laser-dependent constant (18.25 s m−1/2 for CO2
isotherm, v is the welding speed, and tp is the thickness of the plate.
laser), V is the welding speed, tzn is the thickness of the zinc coating,
and tp is the plate thickness. The gap size is critical in producing
sound welds. Chen et al. (2013) conducted an experimental study 4.2. Change in chemical composition
using a CO2 laser with a power range of 1.6–2 kW to find the key
factors influencing laser overlap welding. The study showed that Instead of modifying the welding configuration to lessen the
sound welds could be produced only when the width of the gap was pressure buildup of the zinc vapor, researchers attempted to change
kept within a precise range. When the gap was below 0.15 mm for the chemical composition of the zinc by introducing foreign ele-
galvanized steel sheets with single-sheet thickness of 1.2 mm, the ments. For example, Dasgupta and Mazumder (2006) proposed
weld surface showed pores, while the gaps above 0.5 mm resulted adding copper between the two sandwiched plates to alter the com-
in the sheets not joining together. position. The added copper, which has a low melting temperature
In order to create the interfacial gap, Graham et al. (1994) (Tmelt,Cu = 1358 K) compared to steel, would alloy with the zinc dur-
proposed placing shims between the top and bottom sheets or ing the welding process. The alloy of zinc and copper would prevent
stamping protrusions on the top sheet. Gu (2010) proposed cre- the violent behaviors of zinc in the weld pool, ultimately reducing
ating laser-pulse generated humps on the bottom plate to create a the porosity from 10% to 2%. A reduction in the spectra intensity
gap. The basic idea behind this method is to use a short laser pulse of zinc from spectroscopic measurements during welding confirms
(20 ms) while the beam is traversing in order to push the molten the reduction in zinc vapor when alloyed with copper. The down-
pool to the rear of the keyhole. When the molten pool reaches the side of this method is that adding copper will alter the mechanical
rear keyhole wall, the laser pulse will shut off, resulting in solidifi- properties of the weld. More research is needed to test mechanical
cation of a small bump. Under optimum processing conditions the failure modes such as tensile strength, corrosion fatigue, fracture,
height of the dimples ranged between 0.15 mm and 0.18 mm. Lee etc.
et al. (2013) were able to produce three sheet lap welds of accept- Instead of using copper, Li et al. (2007) suggested using alu-
able quality with upper and lower gaps of 0.1 mm and 0.1 mm. minum foil at the interface of the zinc coated sheets. The addition
Graham et al. (1996) showed that maintaining the gap toler- of aluminum in the fusion zone adds stability to the welding pro-
ance level is however very difficult, even in a research setup, not cess because of its low melting temperature (Tmelt,Al = 933 K) and
to mention in mass automobile production lines. Mei et al. (2015) high vaporization temperature (Tvapor,Al = 2792 K). A 25 ␮m alu-
emphasized the importance of achieving a consistent gap size by minum sheet was inserted between two steel plates. Initial shear
improving the structure of the clamping devices and increasing the testing showed an improvement in shear strength (about 83.3%)
part precision in automobile production lines. To overcome the dif- due to many factors, including decreased porosity in the fusion
ficulty of maintaining a tight tolerance, Bley et al. (2007) proposed zone, greater weld penetration, and improved thermal conductiv-
a special sheet geometry that allows the zinc vapor to be chan- ity. However, having loose contact between the aluminum foil and
neled sideways between the sheets. The special sheet has a large zinc coating resulted in inconsistent stability of the weld pool. Suffi-
radius, which can be produced during the stamping process. In this cient force was required from the clamps in contact with the metal
configuration, the laser beam is offset from the contact point of the sheet to produce sound welds. Alternatively, Li et al. (2007) pro-
two sheets in the horizontal direction to provide a gap between the posed using a cold spray technology to apply aluminum onto the
52 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

surface to avoid the dependency on the clamping force for process 5. Laser welding of magnesium alloys
stability.
Laser welding has attracted the most interest for joining mag-
nesium alloys because of the superior penetration depth, narrow
fusion zone, and high efficiency compared to gas tungsten arc weld-
4.3. Pulsed lasers
ing, electron beam welding, and friction stir welding. But despite
these advantages of laser welding, the laser welding of magnesium
The working principle of pulsed lasers is analogous to resistance
encounters many issues, such as oxide inclusions, pore formation,
spot welding. In contrast to CW lasers, which produce seam weld-
loss of alloying elements, and liquation and solidification cracking.
ing, pulsed lasers create partially overlapping spot welds. Tzeng and
In this paper welding of the most common magnesium alloys AZ31
Chen (2001) and Tzeng (2006) investigated the viability of using
and AZ91 will be discussed.
a pulsed laser to produce a defect free weld in gap-free overlap-
ping configurations. This technique is based on the idea of carefully
5.1. Microstructure
controlling the heating, melting, and vaporizing processes by alter-
ing the pulse rate, duty cycle, and peak power of a CO2 laser, to
The base metal of AZ31 contains hexagonal, closely packed ␣-
minimize the production of zinc vapor, and significantly increase
Mg. Chowdhury et al. (2012) reported that the fusion zone (FZ)
keyhole stability. Through the optimization of control parameters,
of magnesium alloy AZ31B-H24 weld produced by fiber laser con-
welds with defect-free exteriors are achievable. However, even the
tains numerous equiaxed dendritic structure and divorced eutectic
welds with sound surface appearance, when cross-sectioned, dis-
␤-Mg17 Al12 precipitates existing in the interdendritic and inter-
played internal pores due to trapped zinc vapors. In conclusion, it
granular regions. Chowdhury et al. (2011) pointed out similar
is very difficult to achieve defect-free welding using a pulsed laser
microstructures can be observed with the use of other lasers such
source.
as diode lasers. The non-equilibrium rapid cooling associated with
laser welding leads to the formation of equiaxed dendrites with
finer grains than that of the base metal. The grain size transi-
4.4. Removal of zinc coating tions from equiaxed grains in the weld center to columnar grains
near the FZ/heat affected zone (HAZ) boundary. Padmanaban and
Out of all considered techniques, the most intuitive solution is Balasubramanian (2011) observed that the ␤-Mg17 Al12 intermetal-
eliminating the root cause of the problem by pre-removing the lic eutectic compounds are homogeneously distributed in the
zinc coating. Removal of the zinc coating will significantly reduce magnesium matrix.
blow outs, spatter, and porosity during the welding of overlapping The average dendritic grain size depends on heat input condi-
steel sheets. The AWS WZC/D19.0-72 (1972) manual suggests the tions. Wang et al. (2011) conducted butt welding experiments on an
straightforward solution of mechanically removing zinc for weld- extruded AZ31B magnesium alloy with thickness of 5 and 10 mm
ing of lap-jointed galvanized steel. As a more advanced approach, using a fiber laser. The paper reported that higher heat input led
laser ablation of zinc has been suggested by Ma et al. (2013). This to dendrites in the FZ with larger average grain sizes due to the
technique involves a double pass of a 4 kW fiber laser for both pre- longer time available for grain coarsening. For example, the aver-
heating and welding. For the pre-heating process, the laser beam is age grain size of 45, 37, and 30 ␮m corresponded to the heat input
defocused to allow for a wider beam diameter and decreased inten- of 85.7, 62.5, 44 J/mm, respectively. A decrease in laser power or a
sity. After the optimization of the pre-heating speed and defocused faster welding speed leads to a faster cooling time and subsequently
laser beam offset distance, sound welds were obtainable. Energy finer grain sizes. The finer grain sizes induce an increase of material
Dispersive Spectroscopy (EDS) measurements of the cross section hardness. In contrast, Padmanaban and Balasubramanian (2010)
of the steel sheets confirmed that zinc at the faying surface was pointed out that shorter solidification time leads to a reduction in
melted or partially vaporized. concentration of brittle ␤-Mg17 Al12 precipitates. The implication
Similarly, a hybrid process using gas tungsten arc welding and of these competing factors will be discussed in the next section.
fiber laser welding was proposed by Yang and Kovacevic (2009). Bailey et al. (2015) analyzed the average grain size and
In this method, an auxiliary gas tungsten arc heat source was uti- secondary dendrite arm spacing of wrought magnesium alloy
lized to preheat the metal plates to transform the zinc coating to AZ31B-H24 welds produced by a fiber laser. The analysis showed
zinc oxide, and a fiber laser was used to weld the plates. Kim et al. that the average grain size and average secondary dendrite arm
(2008) pointed out that zinc oxide produced in the interface has spacing in the FZ decreased with increase in welding speed. Weld-
a much higher melting temperature (Tmelt,ZnO = 2248 K) than zinc, ing speed was more influential than laser power on the change
which facilitates a reduction in vapor pressure. Similar in nature to in average grain size and average secondary dendrite arm spac-
the hybridization of gas tungsten arc welding and fiber laser weld- ing. Li and Liu (2013) found that the grain refining in the FZ was
ing, Milberg and Trautmann (2009) proposed a bifocal dual laser improved by using a filler wire in the laser-arc hybrid welding. On
setup, utilizing a high power diode laser for the zinc removal and a the contrary, Coelho et al. (2008) reported that there is no signifi-
Nd:YAG laser for the fusion. cant difference in the grain size of ␣-Mg in the HAZ.
Even though the removal of zinc coating produces greater weld-
ability in steel plates, it increases the susceptibility to corrosion. 5.2. Hardness and tensile properties
Pennington (1987) proposed adding a nickel coating after remov-
ing the zinc coating at the faying surface. During the preheating Chowdhury et al. (2012) reported that the microhardness of the
process, fine nickel powder was injected into the laser beam to 2 mm thick fiber laser welded AZ31B-H24 magnesium alloy joints
coat the layer when solidified. Unlike zinc, nickel, which has a exhibited lower hardness than that of the parent material. The
higher melting temperature (Tmelt,Ni = 1728 K), behaves integrally hardness value decreased gradually from 73 HV of the base metal to
with steel. approximately 54 HV at the center of the weld as shown in Fig. 5(a).
The methods discussed above are schematically summarized The reduction in the average hardness from the base metal to the
in Table 2 for the weld configuration method, and Table 3 for the weld zone is due to the transformation of the microstructure from
change in chemical composition method, pulsed laser, and removal deformed and elongated grains in the partially annealed H24 state
of zinc coating method. to the dendritic solidification structure. As shown in Fig. 5(a) higher
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 53

Table 2
Summary of the methods in joining zinc coated steel (weld configuration).

Method Specifics Configuration Reference

Weld configuration Gap using spacers Graham et al. (1994) and Chen et al. (2013)
method

Gap generated by laser Gu (2010)

Large radius sheets Bley et al. (2007)

Edge welding Pieters et al. (2006)

Predrilled holes using laser beam Chen et al. (2009)

Exhaust slot created by leading laser beam Iqbal et al. (2010)

54 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Table 3
Summary of the methods in joining zinc coated steel (change in chemical composition, pulsed laser, removal of zinc coating).

Method Specifics Configuration Reference

Change in chemical Copper inner layer Dasgupta and Mazumder (2006)

composition

Aluminum inner layer Li et al. (2007)

Pulsed lasers Re-melting using pulsed lasers Tzeng and Chen (2001) and Tzeng (2006)

Removal of zinc coating Laser ablation of zinc Ma et al. (2013)

welding speed resulted in higher hardness due to the smaller grain The base metal hardness of the as-extruded AZ31B (Fig. 5(b)) at
sizes. The presence of more grain boundaries and finer dendrite arm approximately 54 HV is much lower than the 73 HB of the par-
spacing leads to stronger resistance to plastic deformation. Tensile tially annealed AZ31B-H24. Therefore, Wang et al. (2011) reported
testing results showed a reduction in both the strength and elon- that average hardness of the FZ of the fiber laser welded as-
gation after butt welding. The ultimate tensile strength reduced by extruded AZ31B was slightly higher than 54 HV of the base metal.
89% and 91% for welding speed of 50 mm/s and 100 mm/s, respec- As shown in Fig. 5(b) the average hardness increased from con-
tively, compared to the parent material. dition #1 (P = 2 kW, V = 1.4 m/min, Q = 85.7 J/mm) to #6 (P = 6 kW,
In contrast to the partially annealed H24 magnesium alloy, the V = 7.8 m/min, Q = 46.1 J/mm). The lower heat input Q of condition
base metal of the as-extruded AZ31B is mostly composed of ␣-Mg. #6 results in reduced solidification time for the grains to grow;
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 55

Fig. 5. Microhardness profile (a) Fiber laser welded AZ31B-H24 butt joints P = 2 kW V = 3, 6 m/min. Reproduced with permission from Chowdhury et al. (2012), © Springer
2012. (b) Fiber laser welded as-extruded AZ31 butt joints #1: P = 2.0 kW V = 1.4 m/min, #2 P = 2.5 kW V = 2.4 m/min, #6 P = 6.0 kW V = 7.8 m/min. Reproduced with permission
from Wang et al. (2011), © Elsevier 2011.

4.5–5.5 m/min investigated in the study, exhibited a yield strength

of 170 MPa, ultimate strength of 212 MPa, and maximum joint
efficiency of 98.6%. This result is in accordance with the smallest
average equiaxed dendritic structure achieved at higher welding
speed and lower laser power. Similarly, Wang et al. (2007) found
that when lower heat input is used the ultimate and tensile strength
can be improved. When the heat input was lowered to 48 J/mm, the
joint failed at the base metal rather than the FZ.
Scintilla et al. (2010) reported that the elongation to fracture for
Nd:YAG laser welded AZ31 was around 50–55% of the base metal.
The local strain analysis revealed that the fracture occurred due to
localization of deformation in the weld. The defects on the weld
surface such as marginal incisions, undercuts, and depression were
predicted to be the reason for the strain localization.

5.3. Defects

Fig. 6. Microstructure of AZ31 produced by a fiber laser (P = 6.0 kW V = 7.8 m/min). 5.3.1. Oxide inclusions
Reproduced with permission from Wang et al. (2011), © Elsevier 2011. It is well known that magnesium has a strong tendency to oxi-
dation. The oxidation rate is significantly increased during laser
therefore, leading to higher average hardness. As shown in Fig. 6 the welding due to the surge in workpiece temperature. Cao et al.
FZ is composed of equiaxed dendrites with finer grains than that of (2006) pointed out that the oxidation rate of magnesium intensifies
the base metal. Similarly, Coelho et al. (2008) used a Nd:YAG laser at temperatures above 450 ◦ C. During laser welding oxide films are
to join 2 mm thick rolled magnesium allow AZ31B and observed readily formed at the interface between the molten pool and the
that the HAZ and fusion zone show a slightly higher microhard- vapor region. The increase in oxygen content in the weld has been
ness value of 57 HV compared to 53 HV of the base metal. Wahba quantified by Zhu et al. (2005). The EDS of a CO2 laser welded AZ31
et al. (2012b) measured the microhardness of CW disk laser welded sheet showed a 14.1% increase in oxygen content in the fusion zone,
die-cast AZ91D magnesium alloy with a thickness of 1.3 mm. The compared to the base metal. The brittle oxides in the magnesium
average hardness from the base material to the FZ increased from alloy joints are known to cause a reduction in the weld strength.
63.5 ± 1.2 HV 200 gf to 77.6 ± 1.6 HV 200 gf. Microstructure analy- Cao et al. (2006) reported that the discontinuities in the oxide film
sis showed refinement of the grains and a higher volume fraction can function as nucleation sites that promote the formation of new
of the hard intermetallic compounds. cracks. Srinivasan et al. (2009) conducted a slow strain rate tensile
Scintilla et al. (2010) reported that the precipitation of ␤- test on a 2.5 mm thick AZ31HP magnesium alloy that was joined by
Mg17 Al12 compounds can occur at an elevated temperature adding an Nd:YAG laser and concluded that premature failure was likely
to the material hardening. Therefore, the increase of heat input Q caused by the initial cracking of the oxide layer, which propagated
can work both ways – coarsening of the grain can result in material through the workpiece.
softening, but precipitation of intermetallic compounds can lead to To minimize oxidation during laser welding, it is important
slight increase in hardness. to protect the workpiece from the surrounding air by using a
Padmanaban and Balasubramanian (2011) investigated the shielding gas. To identify the most suitable shielding gas, Weisheit
effect of laser welding parameters on the mechanical properties et al. (1998) conducted bead-on-plate laser beam welding on AZ91
of AZ31B magnesium alloy. According to the investigation, faster using helium, argon, and nitrogen. Helium gas proved to be the
welding speed and lower laser power resulted in higher joint effi- most effective shielding gas, based on surface quality, weld depth,
ciency. The joint created with the laser power of 2.5 kW and welding and aspect ratio. Helium gas was also effective in suppressing
speed of 5.5 m/min, which is the lowest power within the range the formation of plasma. Helium gas has a high ionization poten-
of 2.5–3.5 kW and the highest welding speed within the range of tial (24.5 eV), which limits plasma formation (Dhahri et al., 2001).
56 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Plasma plume absorption through the inverse bremsstrahlung (IB)

mechanism would lead to a loss in effective laser power. It is espe-
cially important to use helium as the shielding gas when using
long wave length lasers such as CO2 lasers (10.6 ␮m). Hiraga et al.
(2002) used 2 kW CO2 and Nd:YAG lasers to compare the effects of
shielding gases on wrought AZ31B-H24. Using argon as the center
shielding gas in CO2 laser welding led to a deterioration in bead
shape, while sound welds were produced when argon was used
in Nd:YAG laser welding, as the argon is less susceptible to the IB
effect in short wavelength (1064 nm).
Bannour et al. (2012) reported that the shielding gas affects the
peak temperature of the workpiece during the welding operation
of AZ91. Computational fluid dynamics analysis also showed that
cooling during the solidification process was improved when the
gas flow increased. Within the power range of 0.7–2 kW and weld-
ing speed of 1–10 m/min, it was recommended to use a gas flow
rate of more than or equal to 20 l/min to avoid oxidation.

5.3.2. Porosity
In laser welding of magnesium alloy, porosity can originate from
the oxide layer, pre-existing gaseous pores, or the periodic collapse
of unstable keyholes. You et al. (2001) reported that the amor-
phous structure of the magnesium oxide film on the surface of
Fig. 7. Cross-section of the weld produced by AZ91D die-cast magnesium alloy.
magnesium alloys provides a protective layer that prevents fur-
Reproduced with permission from Wahba et al. (2012b), © Elsevier 2012.
ther oxidation. Fournier et al. (2002) described that the magnesium
oxide layer is unstable and absorbs moisture from the atmosphere
in high humidity environments. The magnesium oxide reacts with
water molecules to produce the more stable magnesium hydroxide. (2012b) were able to capture the coalescing and expansion of pores
As stated by Halikia et al. (1998), this magnesium hydroxide then in real time using a high-speed X-ray transmission real-time imag-
decomposes at an approximate temperature of 300 ◦ C to release ing system. The weld cross-section of the weld bead containing
water through the following reaction: pores by the expanded gaseous elements is shown in Fig. 7. The
laser source employed was a CW disk laser with a power range
Mg(OH)2  MgO + H2 O (4)
of 1–4 kW, and the workpiece was a 3 mm thick die-cast AZ91D
Harooni et al. (2012) discovered that the decomposition of mag- magnesium alloy. Zhao and Debroy (2001) observed that the area-
nesium hydroxide into water vapor causes pore formation during percent porosity in the weld metal could amount to 11–17 times
laser welding, mainly near the interface of lap joints. During the that of the base metal, depending on the welding speed. The net
welding process, the water vapor dissolves into hydrogen gas in increase in total pore volume is due to the decrease in surface
the molten pool of magnesium, which then creates pores inside tension as the coalesced pores increase in radius.
the weld bead due to the decrease in the solubility of hydrogen To reduce the expansion of pre-existing pores, Marya and
during solidification. Edwards (2000) suggested minimizing the interaction time
To prevent pore formation, it is important to remove the oxide between the laser beam and the workpiece. Using a high welding
layer from the workpiece prior to welding. The oxide layer can speed and a small beam diameter will enhance the cooling rate,
be mechanically removed using stainless-steel brushes, aluminum which shortens the time for the pores to expand and coalesce.
wool, and sanding. Alternatively, Harooni et al. (2014) proposed a In addition, reduction in interaction time, using a high-density
dual laser beam, first to preheat the sample with a leading laser, beam with a high scanning speed, is recommended as it suppresses
and then to produce a weld using a secondary laser. The signifi- the nucleation and growth of pores. Pastor et al. (2000), using an
cance of this process is that the leading laser beam is also effective Nd:YAG laser at a power level of 3.0 kW, found that changing the
in decomposing the magnesium hydroxide prior to welding, which welding speed from 96 mm/s to 120 mm/s resulted in a drop from
eliminates the need for a separate cleaning process. Harooni et al. 40 to 6 area-percent porosity in the weld. In the butt laser weld-
(2012) also found that preheating the sample by using a plasma arc ing of AZ91D, Wahba et al. (2012b) demonstrated that inserting
torch was effective in preventing porosity. Sahul et al. (2016) fur- an extruded, pore-free alloy sheet between the die-cast parts can
ther studied the effects of the oxide layer on laser welding, involving reduce the weld porosity by up to 50%.
AZ31 magnesium alloy with a thickness of 1.7 mm, and a 2.0 kW Another possible cause of porosity in the laser welding of mag-
disk laser. Tensile testing results showed that the maximum shear nesium alloys is the instability of the keyhole. Maintaining a stable
force was higher for the welded workpiece that had the oxide layer keyhole in the laser welding of magnesium alloys is relatively
mechanically removed prior to welding. easy compared to in aluminum alloys. Zhao and Debroy (2001)
Weisheit et al. (1998) found that high pressure die-cast alloys stated that magnesium alloys have a strong vapor pressure and
such as AZ91 and AM60 also resulted in highly porous welds low surface tension, which allow the keyhole to remain open.
because of pre-existing gaseous elements inside the lattice. The However, deep keyholes are still undesirable as higher surface ten-
gaseous elements in the lattice originate from the turbulent flow sion and greater hydrostatic pressure promote keyhole instability.
and rapid cooling experienced during the die-casting process. Matsunawa (2001) was able to observe pore formation, induced by
Wahba et al. (2012b) analyzed the gas entrapped inside pores using keyhole instability, using X-ray transmission imaging. The X-ray
a quadrupole mass analyzer and found that it mainly consisted of images showed that large bubbles were formed near the bottom of
nitrogen gas (air). During melting, Pastor et al. (2000) reported that the keyhole as the keyhole fluctuated violently. The bubbles devel-
the pre-existing small pores in the base material coalesced into oped into pores as they became trapped in the solidifying wall
large pores and expanded to release the pressure. Wahba et al. during floatation.
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 57

Fig. 8. Causes of defects and preventive measures in laser welding of magnesium alloys.

Therefore, it is very important to find the optimized processing elements will alter the chemical composition of the fusion zone.
parameters to stabilize the keyhole. Bailey et al. (2015) conducted Research on Nd:YAG laser welding of die-cast AZ91 conducted
a parametric study on laser welding of magnesium alloy AZ31 with by Marya and Edwards (2000) showed a high concentration of
a fiber laser to find the necessary operating parameters for creating aluminum on the surface, caused by the faster vaporization of
viable, defect-free welds. A logarithmic trend line, defined in Eq. (5), magnesium. Leong et al. (1998) reported that the loss of alloying
has been established to find the welding cutoff speed that assures elements increases with high heat input, i.e., with slower weld-
defect-free welds: ing speeds and higher laser power. The loss of alloying elements
also causes a weld defect known as weld bead undercut. To reduce
Vc = 0.361 · exp(0.0010421 · I) (5) changes in both chemical composition and weld appearance during
laser keyhole welding, it is recommended to optimize the process
where Vc is the cutoff speed and I is the laser intensity. Modeling
conditions by minimizing the heat input. The various defects and
results showed that welding speeds higher than the cutoff would
remedies in laser welding of aluminum alloys are summarized in
lead to the keyhole frequently fluctuating between full penetration
Fig. 8.
and partial penetration.
Gao et al. (2009) proposed using hybrid fiber laser-MIG welding,
which can further reduce porosity. Increasing the width and depth 6. Laser welding of aluminum alloys
of the arc zone prolongs the solidification, which allows longer time
for the bubbles to escape. However, Gao et al. (2012) pointed out The inherent deficiencies of Metal Inert Gas (MIG) and Tungsten
that due to the low power density of the arc and the nature of ther- Inert Gas (TIG) welding of aluminum, such as the slow processing
mal conduction, the arc has a limited effect on thick materials. It speeds, porosity, high shrinkages, large heat affected zones, and
was not possible to achieve pore-free welds for 8 mm thick samples high residual stress, have renewed the interest in laser welding.
using hybrid welding. Despite the numerous advantages of laser welding, such as the
narrow heat affected zone and high processing speed, aluminum
5.3.3. Loss of alloying elements is however remarkably more difficult to laser-weld than steel due
Magnesium and zinc have lower boiling temperatures around to the lower absorptance of laser beam at the surface of this metal.
1000 K and higher vapor pressure than aluminum. Therefore, dur- Without optimized laser welding process parameters, solidification
ing keyhole welding, the difference in the evaporative loss of these cracking, a loss of alloying elements, and porosity can occur.
58 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Fig. 9. Microhardness profiles of the AA7075-T561 (P = 3.5 kW, V = 0.6 m/min).

Reproduced with permission from Ola and Doern (2015), © Elsevier 2015.

6.1. Microstructure magnesium for Alloy 5182 and Alloy 5754 were 0.74% and 0.22%,
respectively. As a result, a yield strength reduction of 14% and 6%
The laser welding process will affect the microstructure of the was predicted for AA5182 and AA5754 respectively, based on the
aluminum alloy weld joints. At high welding speeds, the FZ will pre- correlation between yield strength and magnesium concentration
dominately consist of an equiaxed dendritic structure. Oladimeji reported by Hatch (1984). In addition, Cieslak and Fuerschbach
and Taban (2016) attributed the equiaxed dendritic structure in (1988) reported that the loss of magnesium during laser welding
the FZ to the heterogenous nucleation mechanism initiated by the can lead to a reduction in weld hardness. The average hardness
presence of nucleating particles in the center of the FZ. In contrast, of the AA5456 aluminum alloy, which had a large loss in magne-
at lower welding speeds the microstructure will contain columnar sium, reduced from 88.1 HKnoop of the base metal to 65.0 HKnoop .
grains due to the lower cooling rate. In contrast, the AA5086 aluminum alloy, which had a small loss in
Ancona et al. (2007) reported that higher welding speeds will magnesium, only reduced from 80.5 HKnoop of the base metal to
lead to faster cooling rates and consequently finer grain structure. 76.0 HKnoop .
Subbaiah et al. (2013) observed fine grains in the FZ and columnar Ola and Doern (2015) reported that dissolution of precipitate
grains in the HAZ due to the different cooling rate distribution along phases can reduce the hardness of precipitation hardened alu-
the cross-section of the weld. minum alloys, including 2000, 6000, and 7000 series alloys. The
optical image and the corresponding microhardness profile of the
6.2. Hardness and tensile properties laser welded AA7075-T651 are shown in Fig. 9. To reduce the
change in chemical composition during laser keyhole welding, it
The mechanical properties of aluminum alloy weld joints can is recommended to optimize the process conditions by adjusting
be impaired by the loss of strain hardening, dissolution of the pre- the incident laser beam power and welding speed. Cieslak and
cipitate phases, and change in microstructure. In laser welding of Fuerschbach (1988) suggested using a higher welding speed to
aluminum alloys, columnar grains can grow in the HAZ approxi- minimize the magnesium depletion.
mately normal to the welding direction. During tensile testing these When the loss of magnesium and precipitation phase dissolu-
columnar grains resist rotation and elongation due to the adjoin- tion is not severe, the grain size is the major factor in the hardness
ing grains. Subbaiah et al. (2013) discovered that the change to of the weld. Sánchez-Amaya et al. (2013) reported that the hard-
columnar grains in the heat affected zone resulted in a decrease ness of AA5083 and AA6082 aluminum alloys in the FZ was on
of ductility up to 54% of that of the base metal (AA5083-H321) average about 10 HV higher than the hardness of the base metal.
value. Ramasamy and Albrigh (2000) reported that the columnar The increase in hardness was attributed to the grain refinement
grain orientation in the HAZ was affected by the welding speed. in the weld due to the high cooling rate. Similarly, Ancona et al.
When the welding speed was reduced, the columnar grains curved (2007) reported that the CO2 laser welded AA5083 aluminum alloy
and became parallel to the welding direction, leading to improved showed an increase in overall hardness ranging from 10% to 16%
ductility in the longitudinal direction. with respect to the base metal.
Pastor et al. (1999) predicted that the tensile strength would Ancona et al. (2007) reported that the tensile strength is also
reduce by the selective loss of alloying elements. The experiment influenced by porosity. Experimental results showed that the ten-
showed that after keyhole welding the weight loss percentages of sile strength of the AA5083 aluminum alloy dropped below 80% of
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 59

the base metal when the porosity was higher than about 5% due to Hu and Richardson (2006) conducted an experimental and
the reduction in the cross-sectional area. numerical study to investigate the mechanism and possible solu-
tion for the transverse solidification cracking in laser welding of
AA7075. They found that the transverse cracking was related to
6.3. Surface reflectivity and efficiency
the longitudinal length of temperature contour. A larger transverse
plastic strain was formed in the mushy zone with the elongated
Sánchez-Amaya et al. (2013) pointed out that the reflectivity of
temperature distribution. They suggested using a low welding
aluminum alloys can be higher than 80%. One of the difficulties in
speed and low laser power to minimize the risk of transverse
welding of aluminum alloys is the high possibility of damaging the
cracking. Similarly, Wang et al. (2014) used numerical modeling
laser heads, as the laser beam reflects back into the collimator lens.
to reveal that higher inter-beam spacing of side-by-side dual laser
To prevent damage, Paleocrassas and Tu (2007) suggested using a
beams increased the length of the hot cracking susceptible mushy
tilted laser head setup or fitting an optical isolator onto the laser
zone. Therefore, to minimize the mushy zone and thus the risk
head. In addition, Kutsuna et al. (2006) pointed out that it is prefer-
of centerline solidification cracking, they recommended using a
able to use a short wavelength laser such as an ytterbium fiber-optic
small inter-beam spacing. Moraitis and Labeas (2008) developed
laser and Nd:YAG laser, as it is absorbed into the sample better than
a numerical model to predict residual stresses in laser welding of
a CO2 laser in laser welding of aluminum alloys. Cao et al. (2003b)
the AA6061-T6 aluminum alloy. Using the model, they were able
reported that the plasma refraction and absorption are less severe
to design welding parameters that resulted in minimum residual
for short wavelength lasers because the absorptive coefficient in
stresses in the weld.
the plasma scales with the square of the wavelength.
Tirand et al. (2013) proposed a hot cracking criterion suitable
The absorption of the laser beam can be improved significantly
for laser welding. According to the hot cracking criterion, the prob-
with formation of a keyhole. Paleocrassas and Tu (2010), using a
ability of solidification cracking was increased with a reduction in
photo diode, could observe a significantly reduced reflected beam
interdendritic spacing, which is affected by solidification speed.
intensity once a keyhole was formed. The welding efficiency can
Therefore, higher welding speed, which leads to higher solidifi-
also be improved by using a CW laser. Jin et al. (2012) developed
cation speed, increases the probability of solidification cracking.
an experimental setup with an aluminum alloy sample clamped
Steenbergen and Thornton (1970) suggested that the basic param-
between two pieces of glass to detect the plasma inside the keyhole.
eters that contribute to HAZ cracking in aluminum are influenced by
The electron temperature distribution inside the keyhole showed
the cooling rate and the peak temperature. Based on this premise,
that the role of multi-reflections and Fresnel absorption on the key-
Sheikhi et al. (2015) reported that it is possible to avoid solidi-
hole wall in CW laser welding was stronger than that of the pulse
fication cracking using pulse shaping in PW laser welding of the
wave (PW) laser due to the continuous energy input.
AA2024 aluminum alloy. The method was to carefully control the
Gao et al. (2015) suggested that the plasma shielding effect
pulse ramp down shape, which resulted in a decrease in the thermal
must be taken into consideration in the laser welding of aluminum
load and a reduced solidification rate. The modification of the cool-
alloys. When the power of the fiber laser increased to 5 kW, a
ing rate was also suggested by Ola and Doern (2015), who improved
large amount of incident power was absorbed by the plasma plume
the resistance to intergranular cracking of the AA7075-T651 alloy
through inverse bremsstrahlung, causing the effective power den-
by utilizing a laser arc hybrid welding system, which resulted in
sity to decrease to 76% of the incident laser power density. In
slower cooling.
contrast, when the laser power was lower than 5 kW, the laser
Solidification cracking can also be suppressed by using a dual
power absorbed by the plasma was no more than 5%.
beam laser system. The premise of this method is to relieve the
The surface condition of the aluminum alloy also has a great
thermal strains created by a primary laser beam by use of a trailing
influence on laser absorption. To improve laser energy coupling
beam. Witzendorff et al. (2015) were able to reduce solidification
during welding, El-Batahgy and Kutsuna (2009) suggested using a
cracking in the AA6082 aluminum alloy welds by superimposing
stainless-steel brush and nitric acid solution to increase the surface
and synchronizing Nd:YAG and diode laser pulses with a specific
roughness of the workpiece.
delay. The dual beam system was able to produce crack free butt
welding with full penetration of 0.5 mm thick AA6082 aluminum
6.4. Defects sheets.
The weld profile also plays an important role in crack sensitivity.
6.4.1. Solidification and liquidation cracking El-Batahgy and Kutsuna (2009) used a 5 kW CO2 to produce bead-
Dausinger et al. (1996) indicated that the following defects can on-plate welds of AA5052, AA5083, and AA6061 aluminum alloys.
occur in the laser welding of aluminum: sagging seams, undercuts, For all the investigated aluminum alloys, cracks were observed
hot cracks, hydrogen pores, cavities, irregular roots, and blowholes. when the welding speed was higher than 5 m/min. The solidifi-
In addition to the various defects aluminum alloys are highly vul- cation cracking occurred due to the excessive penetration leading
nerable to solidification cracking after laser welding due to the to a concave weld surface. This severe undercut of the weld caused
high thermal expansion (approximately twice that of steel) and the high stress concentration at the center. As a remedy, El-Batahgy
large change in volume upon solidification. Near the solidus tem- and Kutsuna (2009) found that crack-free welds were possible by
perature, the alloys pass through a temperature range where the adding a backing strip from the same base metal, which improved
ductility is low. In this temperature range, solidification cracking the weld shape by converting to a convex bead. The weld profile
will occur when the material is subjected to thermal tensile strains can be also manipulated using an electromagnetic weld pool sys-
exceeding its endurance limit. tem. Avilov et al. (2012) were able to utilize a contactless inductive
Cao et al. (2003a) reported that during solidification of alu- electromagnetic weld pool system to suppress the gravity dropout
minum alloys low melting point alloy and impurity segregates are of the AW 5754 aluminum alloy melt and eliminate the sagging of
rejected into grain boundaries by the solidifying dendrites through the weld pool root side surface.
a micro-segregation mechanism. The rejected liquid film in the Huang et al. (2011) devised an experiment that added alloy
mushy zone repairs the cracks by filling in the void. Zhao et al. powders during laser welding of pure aluminum. According to the
(1999) pointed out that slow movements of the residual liquid, experimental results, adding the Si element could reduce the crack
which is controlled by its fluidity, can lead to insufficient filling tendency of aluminum welding by improving the fluidity of the
in the void. weld metal leading to fill-up of the fine cracks. Adding the Mg ele-
60 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

ment also facilitated the forming of low melting point eutectic ␣ sity decreased and the AIO spectra intensity increased. Therefore,
(Al) + Mg2 Si, which suppresses solidification cracking. careful control of the beam power density is critical to minimizing
selective alloy loss.
6.4.2. Porosity
Haboudou et al. (2003) found two kinds of porosity in Nd:YAG
7. Laser welding of titanium alloys
laser welding of AA5083 and A356 alloys. The micro porosities in
the range of 50–200 ␮m were ascribed to the hydrogen solubility
The major conventional welding methods for joining titanium
in aluminum. The hydrogen solubility in aluminum was signifi-
alloys are gas metal arc welding (GMAW), electron beam welding
cantly different between the liquid (0.65 ml/100 g) and solid alloy
(EBW), and gas tungsten arc welding (GTAW). Lathabai et al. (2001)
(0.034 ml/100 g). Therefore, hydrogen was rejected and formed
reported that the conventional methods however have inherent
porosity during the solidification due to the reduced solubility.
drawbacks, such as poor welding quality and high heat input result-
Kutsuna and Yan (1999) pointed out that the hydrogen induced
ing in greater distortion. As demonstrated by Saresh et al. (2007),
porosity was small in size since the high cooling rate associ-
electron beam welding (EBW) requires the welding to be conducted
ated with laser welding prevents growth. Matsunawa et al. (2000)
inside a high vacuum enclosure. Therefore, as pointed out by Lacki
reported that the hydrogen induced porosity in laser welding of
and Adamus (2013), the major advantage of laser welding is that
aluminum alloys can be suppressed by eliminating the source, and
the welding can be conducted in a non-vacuum environment. Gao
indicated that the hydrogen content in the base metal of the A5052
et al. (2013) compared laser welding to TIG welding and found
aluminum alloy was reduced from 10.4% to 3.6% by wire brushing
that laser welding resulted in lower residual distortion. Yunlian
the metal surface. It was also beneficial to use shielding gas com-
et al. (2000) reported that the grain size of welds produced by TIG
pletely void of hydrogen as the hydrogen content in the weld metal
welding was larger than the grain size of those produced by laser
decreased from 10.7% to 5.2% by changing the shielding gas from
welding. Bergmann (2013) found that the average grain size in the
Ar + 1%H2 to pure Ar.
middle of the weld produced by an Nd:YAG laser is 10 times smaller
Haboudou et al. (2003) reported that larger pores in the range
than those in GTAW weldments.
of 300–600 ␮m can be attributed to the keyhole closure, shrink-
The grain size of the weld is influenced by the cooling rate asso-
ing, and process instabilities. Matsunawa et al. (2000) observed the
ciated with the welding mechanism. The lower overall heat input,
porosity formation process in CO2 laser welding of the AA5083 alloy
faster processing time and small beam focus characteristic associ-
by an X-ray transmission imaging system, where the large bubbles
ated with laser beam welding lead to a faster cooling rate, which
that formed at the bottom of the keyhole became trapped in the
consequently induces a finer microstructure in the weld. The small
solidifying wall during floating. Haboudou et al. (2003) stated that
diameter of laser welding also minimizes the width of the HAZ.
using dual laser beams with the distance between the spots set
Balasubramanian et al. (2013) reported that the widths of the HAZ
between 0–0.9 mm enhanced keyhole stability. In contrast to single
were observed to be 4.0, 1.5 and 0.2 mm for GTAW, EBW, and
beam laser welding, coaxial camera images revealed that keyhole
laser beam welding joints, respectively. Among the various tita-
extinctions (collapse) did not appear for dual laser beam welding.
nium alloys this paper will focus on the two-phase ␣ + ␤ Ti6Al4V as
The surface porosity of A356 reduced from 8% to 0.5% by using dual
it is the most widely used titanium alloy in the automotive industry.
laser beams.
Pastor et al. (1999, 2001) revealed that setting the welding speed
too close to the keyhole-to-conduction transition regime led to key- 7.1. Microstructure
hole instability and macro porosity when using a Nd:YAG laser.
They suggested choosing a welding speed that is strictly in either In the laser welding of Ti6Al4V, the main problems arise from
the conduction or keyhole regime. the formation of hard and brittle martensite structures in the FZ and
HAZ. The narrow, focused beam characteristic and high energy of
6.4.3. Loss of alloying elements laser welding cause a large thermal gradient from the base metal
In laser welding of aluminum alloys, the vaporization of magne- to the melt pool. Squillace et al. (2012) reported that the spatial
sium is more pronounced due to the low boiling point temperature and temporal distribution of temperature produces a non-uniform
(1373 K). The decrease of magnesium should be avoided as it may microstructure. In addition, the high self-quenching rate of the laser
lead to the loss of precipitation strengthening, causing reduction in beam welding promotes the microstructure to transform from the
tensile strength and hardness. ␤ to the martensite ␣ phase.
Sibillano et al. (2006) demonstrated that insufficient shielding The base Ti6Al4V consists of an original, equiaxed ␣ phase and
results in a heavy loss of magnesium in the CO2 laser welding of a granular ␤ phase. The original ␣ phase is stabilized by aluminum,
AA5083. More specifically, the magnesium loss varied from 13% while the ␤ phase is stabilized by vanadium. As the heating cycle
to 10%, from the top surface to the middle of the keyhole, with in the laser welding process progresses, the vanadium diffuses out
insufficient shielding gas. As magnesium has a lower boiling point of the body-centered cubic ␤ phase and saturates the hexagonal
than other alloys, it is more likely to vaporize during laser welding. close-packed ␣ phase, and then ␤ dendritic grains grow. Wang
On the other hand, with sufficient shielding gas, the magnesium et al. (2003) observed that the body-centered cubic ␤ phase grains
loss was in the range of 2.5% to 3%, with respect to the base metal grow in the direction of the heat flow when the temperatures are
concentration throughout the keyhole. The presence of shielding above the ␤ transus temperature of 995 ◦ C. The ␤ phase increases
gas (helium) suppresses the vaporization rate with the pressure during the heating process until complete melting occurs at the
and lower temperature, and, therefore, prevents the recombination liquidus temperature of 1654 ◦ C. The FZ will undergo a complete
of magnesium alloys and oxygen. diffusionless ␤ phase transformation upon solidification. The HAZ
Kim et al. (2004) conducted spectroscopic analysis to identify will undergo a homogenization of the ␤ phase through vanadium
the plasma plume produced during the PW laser welding of A5083. diffusion.
The spectrum intensity analysis revealed that the vaporization of Al During the cooling cycle, the FZ and HAZ experience a transfor-
and Mg may change depending on the temperature of the molten mation to the martensitic ␣ phase and the transformed ␣ phase. A
pool. When the power density was set to 5.2 × 109 W/m2 , the major similar transformation can be achieved by quenching from above
species were MgO molecular spectra. In contrast, when the power the ␤ transus temperature. The degree of martensite formation
density was increased to 7.1 × 109 W/m2 , the MgO spectrum inten- will be governed by the cooling rate during the solidification pro-
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 61

blocky ␣, martensitic ␣ , and granular ␤ phase. Liu et al. (2012)

affirmed the existence of the original ␤ phase in the HAZ using
a micro focus X-ray diffraction system (Micro-XRD). Kabir et al.
(2010) reported that the HAZ can be distinguished by the region
near the base metal (cooled above the ␤ transus temperature of
995 ◦ C) and near the FZ (cooled below the liquidus temperature of
1605 ◦ C). The region near the FZ consists mainly of ␣ with a small
amount of primary ␣ while the region near the base metal consists
of primary ␣, intergranular ␤ phase, and ␣ .
The laser welding process conditions are known to have an effect
on the microstructure of the weld. Liu et al. (2012) observed that the
␣ martensite phase is only observed at fast welding speeds. Faster
welding speeds reduce the interaction time between the laser beam
and workpiece, which then increase the cooling rate. In addition,
the average width of the ␣ martensite phase became finer, from
1.39 ␮m to 1.09 ␮m as the welding speed increased from 2 m/min
to 8 m/min. Similarly, Akman et al. (2009) observed that the grain
sizes in both the FZ and HAZ became larger with an increase in the
Fig. 10. Continuous cooling diagram for Ti6AL4V from the reference.
Reproduced with permission from Ahmed and Rack (1998), © Elsevier 1998. average power of the PW Nd:YAG laser. This is due to the higher
overall heat input, resulting in a lower cooling rate. Akbari et al.
(2014) pointed out that the penetration depth and HAZ width were
cess. Ahmed and Rack (1998) reported that full transformation to decreased by increasing the welding speed.
martensite will be obtained when the cooling rate is higher than
410 ◦ C/s. As shown in the continuous cooling diagram (Fig. 10), 7.2. Hardness and tensile properties
the microstructure will go through partial transformation to the
␣ phase when the cooling rate is between 410 ◦ C/s and 20 ◦ C/s. The hardness distribution in the weld is determined by the
The boundaries between the base metals, HAZ and FZ can be microstructure, and the hardening in the FZ and HAZ is attributed
identified by the difference in microstructure as shown in Fig. 11(c). to the formation of ␣ martensite phase. The presence of ␣ marten-
As shown in the SEM image in Fig. 11(d), the base metal consists of site phase clearly increases hardness at the expense of toughness
the equiaxed original ␣ phase and granular ␤ phase. The fusion zone and ductility. Hong and Shin (2016) reported that the hardness in
as shown in Fig. 11(a) consists of the ␣ martensite phase with an the HAZ rapidly increased toward the center of the weld due to
acicular morphology as well as the gray block transformed ␣ phase. the increase in the ␣ martensite phase. A very high spatial gradi-
Hong and Shin (2016) showed that the HAZ shown in Fig. 11(b) and ent of microhardness is created due to the small width of the HAZ.
(d) exhibits a more complex structure compared to the FZ. The HAZ Casalino et al. (2015) reported that the Ti6Al4V weld joint, created
microstructure consists of a heterogeneous mixture of primary ␣, by a CW fiber laser with a power of 1.2 kW and a welding speed of

Fig. 11. Microstructure of the regions in the Ti6Al4V weld (a) FZ, (b) HAZ, (c) weld cross-section, (d) base metal.
Reproduced with permission from Hong and Shin (2016), © Elsevier 2016.
62 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Fig. 12. Average micro-hardness values of Ti6Al4V weld (P = 1000 W,

V = 2.66 m/min) with error bars showing a 95% confidence interval.
Reproduced with permission from Hong and Shin (2016), © Elsevier 2016.

2.5 m/min, resulted in an increase in peak hardness from 340 HV to

410 HV. Similarly, Gao et al. (2014) using a PW Nd:YAG laser found
an average increase of 45 HV in Vickers hardness in the FZ. Costa
Fig. 13. Sketch of the melt pool flow of the “pre-humping” regime resulting in
et al. (2007) reported that the hardness values measured near the
undercut.
weld root were higher than those measured near the surface.
Reproduced with permission from Fabbro (2010), © IOP Publishing 2010.
Kabir et al. (2010) observed an increase in the average hard-
ness of the weld with increasing welding speed. This trend was
also reported by Squillace et al. (2012) and is related to the larger
performed on Ti6Al4V sheets with a thickness of 2 mm using a PW
volume fraction ␣ martensite phase produced by the faster cool-
Nd:YAG laser. For the V-shaped welded joint, the plastic strain was
ing rate. Hong and Shin (2016) predicted the weld geometry and
non-uniformly distributed along the HAZ and base metal boundary.
microhardness of the weld using the thermal history from the
The non-uniform distribution of a strain of the V-shaped welded
multi-physics model. According to the experimental and simula-
joint promoted high local stress leading to lower tensile strength
tion results maximum hardness was consistently measured in the
of the V-shaped welded joints than in the H-shaped welded joints.
FZ and HAZ boundary. The micro-hardness distribution of the weld
cross-section produced by a fiber laser with a power level of 1000 W
and a welding speed of 2.66 m/min is shown in Fig. 12. 7.3. Defects
The tensile properties are heavily influenced by the weld
microstructure, porosity, and, weld profile. Balasubramanian et al. 7.3.1. Undercuts
(2013) conducted transverse tensile testing of butt welded sam- In laser welding, porosity and undercut are highly undesirable
ples produced by a CO2 laser at a power level of 3500 W and a as they can act as stress concentrations and subsequently initiate
welding speed of 50 mm/s. The yield strength and tensile strength fatigue cracks. According to the AWS D17.1/D17.1M:2010-AMD1
of the laser beam welded joint were 959 and 982 MPa, respectively, (2012)AWS D17.1, 2012AWS D17.1/D17.1M:2010-AMD1 (2012)
which are slightly lower than 969 and 1002 MPa of the parent standard, which was developed for general purpose fusion weld-
material. Balasubramanian et al. (2013) reported that the slight ing, the undercut should be less or equal to 0.07T (where T is the
decrease in yield strength and tensile strength could be attributed thickness of the workpiece) for a class A weld.
to the acicular morphology of the joint. Liu et al. (2014) pointed out Ahn et al. (2016) reported that the depth of undercut increased
that the heterogeneity of the microstructure in the laser welded with increasing laser power up to a maximum of around 0.15 mm.
Ti6Al4V alloy promoted crack growth and fatigue damage. Casalino The main cause of the undercut was high laser power, which causes
et al. (2015) reported tensile testing results of 2 mm thick Ti6Al4V strong evaporation and expulsion of the molten material. In fact,
plates that were fiber laser welded in butt welding configura- an undercut produced by laser power above 1.5 kW was all greater
tion. The fractured surface of the tensile tested welds exhibited than the limit set by the AWS D17.1 standard. Squillace et al. (2012)
a non-homogenous surface with sharp edges, and the random ori- reported that the undercut depth with respect to welding speed
entations of ␣ martensite deflected the crack path. Similar crack first increased and reached a maximum value, and then decreased
propagation behavior was also reported by Bergmann (2013). In with the welding speed. Similarly, Kabir et al. (2010) reported that
contrast, the tensile testing of the parent metal resulted in a uni- the undercut depth was the largest at a welding speed of 1.5 m/min
form ductile fracture. and then decreased with a further increase in welding speed.
Cao and Jahazi (2009) found the presence of micropores and alu- Squillace et al. (2012) explained this phenomenon by categoriz-
minum oxides at the fracture surface, which led to loss of ductility. ing the welding speed into two regimes. At the low welding speed
Casalino et al. (2015) reported that the existence of porosity was regime (before reaching a maximum undercut depth) the central
probably the reason for the reduction in ultimate tensile strength emerging flow collides with the side flows emitted from the front
and final ductility of the welded specimens by about 20% and 60 to keyhole wall. The collision of the two flows prevents the emerging
70%, respectively. melt from reaching the sample surface. As shown in Fig. 13, Fabbro
Gao et al. (2014) compared the tensile strength between butt (2010) defined this phenomenon as occurring in the pre-humping
welds with V- and H-shaped cross-section profiles. The welds were regime. At the higher welding speed regime (after reaching a max-
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 63

imum undercut depth), the side flow is overpowered by the central Table 4
Mechanical properties of Fe-Al intermetallic compounds (Rathod and Katsuna,
emerging flow.
2004).
Kawahito et al. (2006) developed an adaptive control system
that controls the peak power and pulse duration of Nd:YAG laser Type of intermetallic compound FeAl3 Fe2 Al5 FeAl Fe3 Al
Vickers hardness 1 kfg load (HV) 892 1013 470 330
based on the heat radiation monitoring signal. The adaptive con-
trol system was was able to reduce the undercut depth by one half,
compared to those made with a conventional rectangular pulse
shape. in high residual stresses. In addition, Katayama (2004) reported that
when welding aluminum to steel, brittle aluminum-rich phases
such as FeAl3 , FeAl2 , and Fe2 Al5 dominate the intermetallic phases
7.3.2. Porosity in the rapid heating and cooling processes. Massalski et al. (1990)
Keyhole stability during laser welding is important in avoiding reported that Fe3 Al, FeAl, FeAl2 , Fe2 Al5 , and FeAl3 intermetallic
porosity induced by keyhole collapses. Wang et al. (2007) pointed phases easily form in the molten Al-Fe mixture as per the binary
out that shielding gas helps stabilize the welding process and Al-Fe phase diagram. Zhou et al. (2016) reported that the Fe-Al com-
reduce the plasma plume size. According to their numerical model- pound easily forms in the order of FeAl, FeAl3 , Fe2 Al5 , and Fe3 Al,
ing, the effective shielding zone (zone with assist gas mass fraction based on the first-principle plane-wave pseudopotential method.
of 0.83 or higher) of helium gas was larger than that of argon gas The typical hardness of the intermetallic phases was reported by
when the equivalent gas flow was used. In addition, Casalino et al. Rathod and Katsuna (2004) as shown in Table 4.
(2005), using Taguchi analysis, found that helium gives a lower As a result of the formation of brittle intermetallic phases, par-
level of imperfection in the weld than argon for the parameters ticularly the aluminum-rich intermetallic phases, Ding et al. (2006)
considered in their investigation. Zhang et al. (2011) developed reported that the weld zone became more prone to cracks. Schubert
a numerical model for investigating the effect of assistant gas on et al. (2001) pointed out that controlling the diffusion process is
the stability of keyhole and molten pool during laser welding of important in laser welding of dissimilar materials because it gov-
Ti6Al4V. According to the simulation results, the impingement of erns the intermetallic formation.
assisting gas flow on the laser/material interaction region helps
the front of the molten pool surface to become concave. The con-
8.1. Single laser beam welding
cave shape around the keyhole reduces the molten pool around
the keyhole aperture and thus expands and stabilizes the keyhole
Dissimilar materials can be laser welded in a variety of joint
aperture.
configurations. The most common joint configurations are butt and
Porosity can also be produced by gas bubbles being trapped in
lap welding. Sun and Ion (1995) suggested that in the butt weld-
the weld pool during solidification. Gao et al. (2014) reported that
ing configuration, the intermetallic formation can be minimized by
increasing the overlapping factor in PW laser welding can assist in
precisely aligning the laser beam to only melt the aluminum, thus
the bubbles escaping the melt pool. Khaled (1994) indicated that
producing a brazed joint. It is important to emphasize that, for over-
hydrogen pores can be formed in the FZ due to the decreased hydro-
lap welding, the diffusion process is clearly affected by whichever
gen solubility of Ti6Al4V during solidification. Cao and Jahazi (2009)
material is on top. For example, Sierra et al. (2007) could achieve a
stated that titanium alloys are generally not susceptible to FZ solidi-
defect-free overlap weld with a penetration up to 500 ␮m for DC 04
fication cracking due to the absence of secondary phase dispersoids
steel-on-6016 aluminum. However, an aluminum-on-steel config-
and impurities at grain boundaries.
uration always resulted in cracks in the interface due to the thicker
intermetallic compound. Tensile shear testing of the joint exhibited
8. Laser welding of dissimilar materials a fracture at the intermetallic region, which is associated with high
brittleness and hardness.
Welding of dissimilar metals, such as combinations of alu- Borrisutthekul et al. (2007) proposed placing a copper heat sink
minum alloys, magnesium alloys, titanium alloys, and steel, has on the back side of the aluminum in the lap welding configura-
always been a difficult and sometimes impossible task because of tion to reduce the thickness of the intermetallic layer. The FEM
the formation of intermetallic compounds. Laser welding of press thermal analysis showed that adding a backing block reduced the
hardened steels also encounters difficulties similar to joining of molten time of aluminum alloys, thus suppressing the growth of
dissimilar metals as the steel sheets are usually coated with Al-Si the intermetallic layer. The joint strength improved to 70 MPa from
alloys. However, many recent studies using fiber lasers, Nd:YAG about 40 MPa by adding a copper backing block to the aluminum.
lasers, and CO2 lasers have been conducted with some success. In addition, Liedl et al. (2011) found that the position of the shield-
This section describes the research carried out on laser welding ing gas supply influenced the cooling conditions and growth of the
of dissimilar materials. intermetallic layer.
In recent years, parts made of steel, aluminum, and magnesium Peyre et al. (2007) discovered that when joining zinc coated
hybrids have increasingly been used in automobiles for weight sav- DC04 steel to 6016 aluminum alloy, the presence of the zinc layer
ing purposes. Yao et al. (2013) reported that press hardened steels favored liquid aluminum spreading and suppression of Fe-Al inter-
are used in structural elements such as A-pillars, B-pillars, and metallic formation. This is due to the high reactivity between
roof rails. Lightweight materials are used in front and rear fend- aluminum and zinc. Joint strength of a 1 mm thick sheet improved
ers, hood, and trunk. As mentioned in the materials section, the from 110 MPa to 230 MPa with zinc coating. Sierra et al. (2008)
majority of these lightweight parts are attached to the body skele- analyzed the influence of using a cleaning flux on joining low-
ton by mechanical means, such as screwing or riveting. However, carbon steel DC 04 with zinc coating and 6016-T4 aluminum in the
greater weight reductions can be achieved by directly welding steel lap welding configuration. The flux provided an additional ther-
and aluminum or magnesium together. mal barrier between the galvanized steel and aluminum alloy, thus
Yan et al. (2010) reported that welding of aluminum or mag- impeding the intermetallic growth. In addition, porosities were
nesium to steel is difficult as the metals’ thermal properties are prevented in the FZ by suppression of zinc vaporization. When a
different and a brittle intermetallic compound layer can form at the flux was used, the strength improved from 150 N/mm to 250 N/mm,
interface. Watanabe et al. (2006) pointed out that the large differ- which is close to the 260 N/mm ultimate tensile strength of the
ence in thermal expansion between the different materials results 6016-T4 base material.
64 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Table 5
Summary of the laser welding of dissimilar materials investigations.

Material type Thickness Laser type Welding conditions Intermetallic thickness Reference

DC04 steel-AA6016, 1.2–1 mm, 1.2–1.3 mm Nd:YAG 2.25–3.5 kW 5–20 ␮m Sierra et al. (2007)
DC04 steel-AA6056 4–6 m/min
Dual-ten 590 1.2–1.6 mm – 3 kW 2.5 ␮m Borrisutthekul et al.
steel-AA6022-O 0.03–1.4 m/min (2007)
DC04 steel-AA6016-T4 1.2–1 mm Nd:YAG 2.5–3 kW 2–15 ␮m Peyre et al. (2007)
0.6–1.4 m/min
DC04 steel-AA6016-T4 1.2–1 mm Nd:YAG 3 kW 2–40 ␮m Sierra et al. (2008)
0.6–1.4 m/min
JCS270CC 0.8–1.2 mm (1) CW Nd:YAG (1) 390 W 10 ␮m Yan et al. (2010)
steel-AA6111-T4 (2) PW Nd:YAG 0.06 m/min
(2) Peak
2.61 kW
0.06 m/min
DP590 0.75–1 mm Double pass Fiber laser (1) 4 kW 5 ␮m Ma et al. (2014)
steel-AA6061-T6 40–60 mm/s
(2) 3–3.5 kW
100 mm/s
DX54 steel-AA5754 1–1 mm Double pass Fiber laser (1) 600 W – Chen et al. (2011)
75–100 mm/s
(2) 150–250 W
75 mm/s
DP600 0.77–1 mm Nd:YAG 20–180 J/mm 3–23 ␮m Dharmendra et al.
steel-AA6016-T4 0.2–1.2 m/min (2011)
Filler: Zn–15%Al
DX56D+Z 140 MB 0.9–1 mm, 0.9–1 mm Nd:YAG 1–1.6 kW Up to 5 ␮m Laukant et al. (2005)
steel-AA6016, DC04+ 0.5–1.5 m/min
ZE 75/75 steel-AA6016
Filler: Zn–2%Al
H220YD steel-AA 6016 1.2–1.15 mm Fiber laser 2.3–2.6 kW 1.5–13 ␮m Zhang et al. (2013)
Filler: Al–5%Si 1 m/min
Q235 steel-AA6061 2.5–2.5 mm Fiber laser 5–5.2 kW – Sun et al. (2015)
Filler: Al–5%Si 1 m/min
Low-carbon – Nd:YAG 2.2–3.0 kW 5–20 ␮m Mathieu et al. (2006)
steel-AA6016-T4 –
Filler: Al–12%Si
Galvanized steel-5A02 1.2–1 mm (1) Nd:YAG (1) 1 kW – Qin et al. (2014)
Al (2) GMAW 1–2.5 m/min
Filler: ER4043
SP781 steel-AZ31B 1.2–3 mm Disk laser 2 kW Up to 450 nm Wahba and Katayama
2–4 m/min (2012a)

Intermetallic formation is also an issue for laser welding of Al- erated. Under optimum processing parameters, the intermetallic
Si coated press hardened steel in overlap configuration. Kim et al. layer thickness could be controlled at around 5 ␮m.
(2011) showed that the Al-Si coating diffused into the melt pool, Chen et al. (2011) proposed a double pass welding scheme to
leading to Fe-Al intermetallic phase formation along the fusion line. improve the weld bead appearance. The investigation involved
The diffusion of Al-Si coating led to the reduction of tensile shear fiber laser welding of 1.0 mm DX54 Zn-coated steel and 1.0 mm
strength from 17.3 kN of the non-coated steel to 13.4 kN. Saha et al. thick EN-AW-5754 aluminum in a lap welding configuration. The
(2016) reported that the mixing of the Al-Si coating during laser second pass welding (with lower power than the first pass) clearly
welding could promote ␦-ferrite formation in the fusion zone based improved the weld appearance and its corrosion resistance. How-
on the Fe-Al binary phase diagram. ever, upon closer inspection, the weld still contained porosity with
the diameters ranging from 26.9 to 112.3 ␮m located near the
8.2. Dual laser beam and double pass laser welding interface. Micro-cracks were observed near the interfacial reaction
area of the fusion zone. It is important to point out that using N2
Yan et al. (2010) suggested using a dual beam Nd:YAG laser rather than argon as the shielding gas improved the shear strength,
setup with CW and PW, which enhances the bonding strength reduced the intermetallic formation, and lowered the hardness
between A6111-T4 and low carbon steel by creating a PW-induced variation. Due to the high reactivity, it is possible that N2 gas reacted
root-shape structure at the interface. The dual laser setup was with aluminum to form the AlN phase.
effective in preventing blowholes, but the shearing strength of the
dissimilar joint was greatly reduced from 244 MPa of A6111 to 8.3. Laser welding-brazing
128 MPa. The microhardness of the Fe-Al intermetallic compound
layer was 437 HV, which was much higher than 76 HV and 150 HV Dharmendra et al. (2011) investigated the laser brazing of zinc
of A6111 and steel, respectively. coated steel to aluminum alloy with a zinc based filler. No flux was
Ma et al. (2014) utilized a two-pass welding process to join gal- necessary due to good metallurgical compatibility between the zinc
vanized DP590 steel to 6061-T6 aluminum alloy in the lap welding based filler and aluminum. The sheet materials used in the investi-
configuration. The preheating process was critical to obtaining a gation were DP600 steel with zinc coating and 6061 T4 aluminum,
high strength joint. The appropriate amount of zinc needs to be and the filler wire was Zn–15%Al alloy. The thickness of the inter-
removed during the preheating process to prevent zinc vaporiza- metallic layer varied from 3 to 23 ␮m depending on the process
tion causing spatter and blowholes. However, if too much of the zinc parameters. The dominant intermetallic phase was Fe-rich Fe3 Al
layer is removed, an Fe-Al intermetallic phase layer could be gen- due to most aluminum forming the intermetallic phase with zinc.
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 65

Table 6
Summary of the configurations used in the laser welding of dissimilar materials investigations.

Method Configuration Characteristics Reference

Single beam welding Lap welding with laser at a 90◦ Produced defect free steel-on Sierra et al. (2007)
irradiation angle (steel on top aluminum lap weld up to
of aluminum) 500 ␮m of penetration depth
Lap welding with laser at a 90◦ Added a backing block (heat Borrisutthekul et al. (2007)
irradiation angle (steel on top sink) to suppress the formation
of aluminum) of intermetallic layer
Lap edge welding with laser at Zinc coating of the steel Peyre et al. (2007)
a 30◦ irradiation angle suppressed the formation of
(aluminum on top of steel) intermetallic layer
Lap edge welding with laser at Added flux to provide an Sierra et al. (2008)
a 30◦ irradiation angle additional thermal barrier
(aluminum on top of steel) between the galvanized steel
and aluminum alloy

Dual laser beam and Lap welding with laser at a 90◦ Utilized a PW laser to induce a Yan et al. (2010)
double pass laser irradiation angle (steel on top root-shape structure at
welding of aluminum) interface
Lap welding with laser at a 90◦ Preheated the workpiece (first Ma et al. (2014)
irradiation angle (steel on top pass) to control the zinc
of aluminum) coating thickness
Lap welding with laser at a 90◦ Second pass of laser beam to Chen et al. (2011)
irradiation angle (steel on top Improve weld-bead
of aluminum) appearance

Laser welding-brazing Lap edge welding with laser at Laser brazing using Zn-15%Al Dharmendra et al. (2011)
a 35◦ irradiation angle alloy filler wire
(aluminum on top of steel)
Lap edge welding (steel on top Laser brazing using Zn-2%Al Laukant et al. (2005)
of aluminum) and filled flange alloy filler wire
welding
Butt welding Laser brazing using Al-5Si alloy Zhang et al. (2013)
filler wire
Butt welding Laser brazing using Al-5Si alloy Sun et al. (2015)
filler wire
Butt welding Preheated filler wire to Mathieu et al. (2006)
improve adhesion of the filler
material to the work piece
Flange welding Utilized laser-gas metal hybrid Qin et al. (2014)
welding

Conduction mode laser Lap welding with laser at a 90◦ Conduction welding to Wahba and Katayama (2012a)
welding irradiation angle (steel on top minimize intermetallic layer
of magnesium)

The intermetallic layer thickness of between 8 and 12 ␮m exhib- sion of the filler material to the workpiece. Mathieu et al. (2007)
ited the maximum joint strength. When the intermetallic layer was pointed out that the global geometry of the joints such as concavity
too thin, the joint did not have a sufficient bonding area, while a and wetting angle is a significant factor in weld strength.
thick intermetallic layer led to crack propagating through the brit- Qin et al. (2014) proposed a laser-gas metal arc brazing-fusion
tle interface. Similarly, Laukant et al. (2005) proposed using the welding technique using a large spot laser to stabilize arc and pre-
filler wire Zn–2%Al alloy to join hot dip galvanized DX56D+Z 140MB heat steel, while the trailing electric arc was the main heat source.
steel and EN AW-AA6016 aluminum alloy. The maximum thickness The materials used in the experiment were hot-dip galvanized steel
of the intermetallic layer was maintained below 5 ␮m. The tensile plate and 5A02 aluminum alloy with a ER4043 (Al-5Si) filler wire.
strength and elongation of the joint was 80% and 40% with respect One of the failure modes in tensile testing appeared at the HAZ
to the base material AA6016 in T6 state, respectively. aluminum alloy. This was due to the localized annealing by grain
Shah and Ishak (2014) reported that Si rich fillers can suppress growth during the welding process.
the growth of brittle Al-Fe phases by Al-Fe-Si ternary phases due
to the lower formation enthalpy. Zhang et al. (2013) investigated
laser butt welding of H220YD automotive galvanized steel and 8.4. Conduction mode laser welding
6016 aluminum alloy with eutectic 4043 aluminum alloy (Al–5%Si)
filler wire. Dendritic eutectic structure was formed at the center Instead of keyhole laser welding, Wahba and Katayama (2012a)
of the weld due to the rapid cooling. The intermetallic layer was proposed conduction mode welding of AZ31B magnesium alloy to
composed of Al8 Fe2 Si, Al13 Fe4 , and Al2 Fe, and the thickness was SP781 zinc coated steel in lap welding configuration. Compared to
maintained below 13 ␮m. Sun et al. (2015) butt joined AA6061 keyhole welding, conduction welding was much more stable and
aluminum allow and Q235 low-carbon mild steel with ER4043 (Al- did not have hot cracking. Staying in the conduction mode regime
5%Si) filler wire. All the specimens failed at the Fe-Al intermetallic prevented the perturbation caused by the evolution Fe-plume and
compound layer during tensile testing and XRD analysis revealed the keyhole instability. Furthermore, as with the welding of alu-
Fe2 Al5 at the fracture surface. Therefore, it is likely that the brittle minum alloy to steel, Fe3 Al intermetallic compound was formed at
Fe2 Al5 is the source of crack initiation and propagation. the interface of the magnesium alloy and steel, but the thickness
Mathieu et al. (2006) suggested use of preheated filler wire to of the compound layer was reduced down to 450 nm. As expected,
prevent imperfections that could have been caused by poor adhe- EDS analysis showed no diffusion between Fe and Mg. This implies
that Al was rejected from the magnesium alloy and interacted with
66 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

the Fe surface. It is important to point out that the Zn-coating layer Becker, M.N., 1999. Aluminum: new challenges in downstream activities. JOM 51
improved the joint strength by pre-dissolving the oxide film on the (11), 26–38.
Bergmann, J.P., 2013. Mechanical behaviour of overlap joints of titanium. Sci.
AZ31B and pushing the byproducts outside the weld zone. Table 5 Technol. Weld. Join., 50–60.
summarizes the material type, welding conditions, etc. and Table 6 Blawert, C., Hort, N., Kainer, K.U., 2004. Automotive applications of magnesium and
summarizes the configurations used in the dissimilar material laser its alloys. Trans Indian Inst. Met. 57 (4), 397–408.
Bley, H., Weyand, L., Luft, A., 2007. An alternative approach for the cost-efficient
welding investigations. laser welding of zinc-coated sheet metal. CIRP Ann. Manuf. Technol. 56 (1),
17–20.
Borrisutthekul, R., Yachi, T., Miyashita, Y., Mutoh, Y., 2007. Suppression of
9. Conclusions intermetallic reaction layer formation by controlling heat flow in dissimilar
joining of steel and aluminum alloy. Mater. Sci. Eng. A 467 (1), 108–113.
Buehrle, J., Bea, M., Brockmann, R., 2013. Laser remote process technology on
• Vaporization of zinc during laser welding of galvanized steel automotive manufacture. In: Proceedings of the FISITA 2012 World
can cause disruption of the keyhole stability, spatter, and void Automotive Congress, Berlin, Heidelberg, pp. 89–97.
Cao, X., Jahazi, M., 2009. Effect of welding speed on butt joint quality of Ti–6Al–4V
formation. The adverse effect of zinc vapor can be reduced by
alloy welded using a high-power Nd:YAG laser. Opt. Lasers Eng. 47 (11),
pre-removing the zinc coating, changing the weld configuration, 1231–1241.
altering chemical composition, and using pulse laser beams. Cao, X., Jahazi, M., Immarigeon, J.P., Wallace, W., 2006. A review of laser welding
• Magnesium alloys are susceptible to oxide inclusion and coa- techniques for magnesium alloys. J. Mater. Process. Technol. 171 (2), 188–204.
Cao, X., Wallace, W., Immarigeon, J.P., Poon, C., 2003a. Research and progress in
lescing of pre-existing pores in die-cast alloys. The interaction laser welding of wrought aluminum alloys II. Metallurgical microstructures,
time between the laser and magnesium alloy workpiece should defects, and mechanical properties. Mater. Manuf. Process. 18 (1), 23–49.
be minimized to suppress the growth of pores. Cao, X., Wallace, W., Poon, C., Immarigeon, J.P., 2003b. Research and progress in
laser welding of wrought aluminum alloys I. Laser welding processes. Mater.
• Aluminum alloy weldments are susceptible to solidification and
Manuf. Process. 18 (1), 1–22.
liquidation cracking and hydrogen induced porosity. Slowing the Casalino, G., Curcio, F., Minutolo, F.M.C., 2005. Investigation on Ti6Al4V laser
solidification rate is effective in preventing solidification cracking welding using statistical and Taguchi approaches. J. Mater. Process. Technol.
167 (2), 422–428.
of aluminum alloy weldments. Casalino, G., Mortello, M., Campanelli, S.L., 2015. Ytterbium fiber laser welding of
• The high cooling rate associated with laser welding can cause Ti6Al4V alloy. J. Manuf. Process. 20, 250–256.
phase transformation of Ti6Al4V welds leading to reduction in Chao, Y.J., 2003a. Failure mode of spot welds: interfacial versus pullout. Sci.
Technol. Weld. Join. 8 (2), 133–137.
ductility.
Chao, Y.J., 2003b. Ultimate strength and failure mechanism of resistance spot weld
• Laser welding of dissimilar metals, such as steel to aluminum subjected to tensile, shear, or combined tensile/shear loads. J. Eng. Mater.
alloy or steel to magnesium alloy, is difficult due to formation of Technol. 125 (2), 125–132.
Chen, G., Mei, L., Zhang, M., Zhang, Y., Wang, Z., 2013. Research on key influence
brittle intermetallic phases.
factors of laser overlap welding of automobile body galvanized steel. Opt. Laser
Technol. 45, 726–733.
Chen, H.C., Pinkerton, A.J., Li, L., Liu, Z., Mistry, A.T., 2011. Gap-free fibre laser
References welding of Zn-coated steel on Al alloy for light-weight automotive
applications. Mater. Design 32 (2), 495–504.
Ahmed, T., Rack, H.J., 1998. Phase transformations during cooling in ␣+ ␤ titanium Chen, W., Ackerson, P., Molian, P., 2009. CO2 laser welding of galvanized steel
alloys. Mater. Sci. Eng. A 243 (1), 206–211. sheets using vent holes. Mater. Design 30 (2), 245–251.
Ahn, J., Chen, L., Davies, C.M., Dear, J.P., 2016. Parametric optimisation and Chowdhury, S.H., Chen, D.L., Bhole, S.D., Powidajko, E., Weckman, D.C., Zhou, Y.,
microstructural analysis on high power Yb-fibre laser welding of Ti–6Al–4V. 2012. Fiber laser welded AZ31 magnesium alloy: the effect of welding speed
Opt. Lasers Eng. 86, 156–171. on microstructure and mechanical properties. Metall. Mater. Trans. A 43 (6),
Akbari, M., Saedodin, S., Toghraie, D., Shoja-Razavi, R., Kowsari, F., 2014. 2133–2147.
Experimental and numerical investigation of temperature distribution and Chowdhury, S.M., Chen, D.L., Bhole, S.D., Powidajko, E., Weckman, D.C., Zhou, Y.,
melt pool geometry during pulsed laser welding of Ti6Al4V alloy. Opt. Laser 2011. Microstructure and mechanical properties of fiber-laser-welded and
Technol. 59, 52–59. diode-laser-welded AZ31 magnesium alloy. Metall. Mater. Trans. A 42 (7),
Akhter, R., Steen, W.M., Cruciani, D., 1989. Laser welding of zinc coated steel. 1974–1989.
Proceedings of 6th International Conference Lasers in Manufacturing, 105–120. Cieslak, M.J., Fuerschbach, P.W., 1988. On the weldability, composition, and
Akman, E., Demir, A., Canel, T., Sınmazçelik, T., 2009. Laser welding of Ti6Al4V hardness of pulsed and continuous Nd:YAG laser welds in aluminum alloys
titanium alloys. J. Mater. Process. Technol. 209 (8), 3705–3713. 6061, 5456, and 5086. Metall. Trans. B 19 (2), 319–329.
Ancona, A., Lugara, P.M., Sorgente, D., Tricarico, L., 2007. Mechanical Coelho, R.S., Kostka, A., Pinto, H., Riekehr, S., Kocak, M., Pyzalla, A.R., 2008.
characterization of CO2 laser beam butt welds of AA5083. J. Mater. Process. Microstructure and mechanical properties of magnesium alloy AZ31B laser
Technol. 191 (1), 381–384. beam welds. Mater. Sci. Eng. A 485 (1), 20–30.
Aslanlar, S., 2006. The effect of nucleus size on mechanical properties in electrical Costa, A., Miranda, R., Quintino, L., Yapp, D., 2007. Analysis of beam material
resistance spot welding of sheets used in automotive industry. Mater. Design interaction in welding of titanium with fiber lasers. Mater. Manuf. Process. 22
27 (2), 125–131. (7–8), 798–803.
Assunção, E., Quintino, L., Miranda, R., 2010. Comparative study of laser welding in Dasgupta, A., Mazumder, J., 2006. A novel method for lap welding of automotive
tailor blanks for the automotive industry. Int. J. Adv. Manufacturing Technol. sheet steel using high power CW CO2 laser. Proceedings of the 4th
49 (1–4), 123–131. International Congress on Laser Advanced Materials Processing, 1–5.
Avilov, V.V., Gumenyuk, A., Lammers, M., Rethmeier, M., 2012. PA position full Dausinger, F., Rapp, J., Beck, M., Faisst, F., Hack, R., Hügel, H., 1996. Welding of
penetration high power laser beam welding of up to 30 mm thick AlMg3 plates aluminum: a challenging opportunity for laser technology. J. Laser Appl. 8 (6),
using electromagnetic weld pool support. Sci. Technol. Weld. Join. 17 (2), 285–290.
128–133. Davies, G., 2012. Materials for Automobile Bodies. Elsevier, Oxford, pp. 94–143.
AWS D17.1/D17.1M: 2010-AMD1, 2012. Specification for fusion welding for Dhahri, M., Masse, J.E., Mathieu, J.F., Barreau, G., Autric, M., 2001. Laser welding of
aerospace applications. American Welding Society. AZ91 and WE43 magnesium alloys for automotive and aerospace industries.
AWS WZC/D19.0-72, 1972. Welding zinc-coated steel. American Welding Society. Adv. Eng. Mater. 3 (7), 504–507.
Bailey, N.S., Tan, W., Shin, Y.C., 2015. A parametric study on laser welding of Dharmendra, C., Rao, K.P., Wilden, J., Reich, S., 2011. Study on laser
magnesium alloy AZ31 by a fiber laser. J. Manuf. Sci. Eng. 137 (4), 041003. welding–brazing of zinc coated steel to aluminum alloy with a zinc based filler.
Balasubramanian, T.S., Balakrishnan, M., Balasubramanian, V., Manickam, M.A.M., Mater. Sci. Eng. A 528 (3), 1497–1503.
2013. Effect of welding processes on joint characteristics of Ti–6Al–4V alloy. Dieffenbach, J.R., 1999. Not the Delorean Revisited: An Assessment of the
Sci. Technol. Weld. Join., 702–708. Competitive Position of a Stainless Steel Body-in-White. SAE Technical Paper,
Bannour, S., Abderrazak, K., Mhiri, H., Le Palec, G., 2012. Effects of 1999-01-3239.
temperature-dependent material properties and shielding gas on molten pool Ding, J.J., Huang, H.J., Peyre, P., Fabbro, R., 2006. Temperature criterion of laser
formation during continuous laser welding of AZ91 magnesium alloy. Opt. welding for joining aluminum alloy with low-carbon steel. Mater. Manuf.
Laser Technol. 44 (8), 2459–2468. Process. 21 (1), 59–61.
Barnes, T.A., Pashby, I.R., 2000. Joining techniques for aluminium spaceframes used Donders, S., Brughmans, M., Hermans, L., Liefooghe, C., Van der Auweraer, H.,
in automobiles: Part I—solid and liquid phase welding. J. Mater. Process. Desmet, W., 2006. The robustness of dynamic vehicle performance to spot
Technol. 99 (1), 62–71. weld failures. Finite Elem. Anal. Design 42 (8), 670–682.
Bavarian Motor Works (BMW), 2007. Magnesium fosters rebirth of an automotive Donders, S., Brughmans, M., Hermans, L., Tzannetakis, N., 2005. The effect of spot
engine. International Magnesium Association: The Global Voice for weld failure on dynamic vehicle performance. Sound Vibr. 39 (4), 16–25.
Magnesium, pp. 1–3.
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 67

El-Batahgy, A., Kutsuna, M., 2009. Laser beam welding of AA5052, AA5083, and Kabir, A.S.H., Cao, X., Medraj, M., Wanjara, P., Cuddy, J., Birur, A., 2010. Effect of
AA6061 aluminum alloys. Adv. Mater. Sci. Eng., 1–9. welding speed and defocusing distance on the quality of laser welded
Fabbro, R., 2010. Melt pool and keyhole behaviour analysis for deep penetration Ti–6Al–4V. In: Proceedings of the Materials Science and Technology 2010
laser welding. J. Phys. D: Appl. Phys. 43 (44), 445501. Conference, Houston, TX, pp. 2787–2797.
Fabbro, R., Coste, F., Goebels, D., Kielwasser, M., 2006. Study of CW Nd-Yag laser Kainer, K.U., Hoppe, R., Bohlen, J., Kurz, G., Yi, S.B., Letzig, D., 2015. Challenges and
welding of Zn-coated steel sheets. J. Phys. D: Appl. Phys. 39 (2), 401–409. solutions in the development of magnesium sheet for sustainable vehicle
Faller, K., 2002. Management justification to select titanium automotive concepts. Mater. Sci. Forum 828, 15–22.
components. SAE Technical Paper, 2002-01-0363. Katayama, S., 2004. Laser welding of aluminium alloys and dissimilar metals.
Faller, K., Froes, F.S., 2001. The use of titanium in family automobiles: current Weld. Int. 18 (8), 618–625.
trends. JOM 53 (4), 27–28. Kawahito, Y., Kito, M., Katayama, S., 2006. In-process monitoring and adaptive
Fournier, V., Marcus, P., Olefjord, I., 2002. Oxidation of magnesium. Surf. Interface control for laser spot and seam welding of pure titanium. J. Laser
Anal. 34 (1), 494–497. Micro/Nanoeng. 1 (3), 269–274.
Friedrich, H., Schumann, S., 2001. Research for a “new age of magnesium” in the Kelkar, A., Roth, R., Clark, J., 2001. Automobile bodies: can aluminum be an
automotive industry. J. Mater. Process. Technol. 117 (3), 276–281. economical alternative to steel? JOM 53 (8), 28–32.
Froes, F.H., Eliezer, D., Aghion, E., 1998. The science, technology, and applications of Khaled, T., 1994. An investigation of pore cracking in titanium welds. J. Mater. Eng.
magnesium. JOM 50 (9), 30–34. Perform. 3 (1), 21–36.
Froes, F.H., Friedrich, H., Kiese, J., Bergoint, D., 2004. Titanium in the family Ki, H., Mohanty, P.S., Mazumder, J., 2002. Multiple reflection and its influence on
automobile: the cost challenge. JOM 56 (2), 40–44. keyhole evolution. J. Laser Appl. 14 (1), 39–45.
Fujita, S., Mizuno, D., 2007. Corrosion and corrosion test methods of zinc coated Kim, C., Choi, W., Kim, J., Rhee, S., 2008. Relationship between the weldability and
steel sheets on automobiles. Corros. Sci. 49 (1), 211–219. the process parameters for laser-TIG hybrid welding of galvanized steel sheets.
Gao, M., Chen, C., Hu, M., Guo, L., Wang, Z., Zeng, X., 2015. Characteristics of plasma Mater. Trans. 49 (1), 179–186.
plume in fiber laser welding of aluminum alloy. Appl. Surf. Sci. 326, 181–186. Kim, C., Kang, M.J., Park, Y.D., 2011. Laser welding of Al-Si coated hot stamping
Gao, M., Tang, H.G., Chen, X.F., Zeng, X.Y., 2012. High power fiber laser arc hybrid steel. Procedia Eng. 10, 2226–2231.
welding of AZ31B magnesium alloy. Mater. Design 42, 46–54. Kim, J., Oh, S., Ki, H., 2015. A study of keyhole geometry in laser welding of
Gao, M., Zeng, X.Y., Tan, B., Feng, J.C., 2009. Study of laser MIG hybrid welded AZ31 zinc-coated and uncoated steels using a coaxial observation method. J. Mater.
magnesium alloy. Sci. Technol. Weld. Join. 14 (4), 274–281. Process. Technol. 225, 451–462.
Gao, X.L., Zhang, L.J., Liu, J., Zhang, J.X., 2013. A comparative study of pulsed Kim, J., Oh, S., Ki, H., 2016. Effect of keyhole geometry and dynamics in zero-gap
Nd:YAG laser welding and TIG welding of thin Ti6Al4V titanium alloy plate. laser welding of zinc-coated steel sheets. J. Mater. Process. Technol. 232,
Mater. Sci. Eng. A 559, 14–21. 131–141.
Gao, X.L., Zhang, L.J., Liu, J., Zhang, J.X., 2014. Effects of weld cross-section profiles Kim, J.D., Oh, J.S., Lee, M.H., Kim, Y.S., 2004. Spectroscopic analysis of plasma
and microstructure on properties of pulsed Nd:YAG laser welding of Ti6Al4V induced in laser welding of aluminum alloys. Mater. Sci. Forum 449, 429–432.
sheet. Int. J. Adv. Manuf. Technol. 72 (5–8), 895–903. Klinger, J., 2012, Automotive body structure assembly: Mass & cost saving
Graham, M.P., Hirak, D.M., Kerr, H.W., Weckman, D.C., 1994. Nd:YAG laser welding potential of laser welding compared to spot welding (Doctoral dissertation),
of coated sheet steel. J. Laser Appl. 6 (4), 212–222. www.diva-portal.org/smash/get/diva2:608422/FULLTEXT01.pdf.
Graham, M.P., Kerr, H.W., Weckman, D.C., 1996. Laser welding of Zn-coated sheet Koike, J., Ohyama, R., Kobayashi, T., Suzuki, M., Maruyama, K., 2003.
steels. Proc. SPIE 2703, 170–183. Grain-boundary sliding in AZ31 magnesium alloys at room temperature to
Gu, H., 2010. Laser lap welding of zinc coated steel sheet with laser-dimple 523 K. Mater. Trans. 44 (4), 445–451.
technology. J. Laser Appl. 22 (3), 87–91. Kutsuna, M., Kitamura, S., Shibata, K., Sakamoto, H., Tsushima, K., 2006.
Haboudou, A., Peyre, P., Vannes, A.B., Peix, G., 2003. Reduction of porosity content Improvement of the joint performance in laser welding of aluminium alloys.
generated during Nd:YAG laser welding of A356 and AA5083 aluminium Weld. World 50 (1–2), 22–27.
alloys. Mater. Sci. Eng. A 363 (1), 40–52. Kutsuna, M., Yan, Q., 1999. Study on porosity formation in laser welds of
Halikia, I., Neou-Syngouna, P., Kolitsa, D., 1998. Isothermal kinetic analysis of the aluminium alloys (Report 2) Mechanism of porosity formation by hydrogen
thermal decomposition of magnesium hydroxide using thermogravimetric and magnetism. Weld. Int. 13 (8), 597–611.
data. Thermochim. Acta 320 (1), 75–88. Lacki, P., Adamus, K., 2013. Numerical simulation of welding thin titanium sheets.
Harooni, M., Carlson, B., Kovacevic, R., 2014. Dual-beam laser welding of AZ31B Key Eng. Mater. 549, 407–414.
magnesium alloy in zero-gap lap joint configuration. Opt. Laser Technol. 56, Lathabai, S., Jarvis, B.L., Barton, K.J., 2001. Comparison of keyhole and conventional
247–255. gas tungsten arc welds in commercially pure titanium. Mater. Sci. Eng. A 299
Harooni, M., Kong, F., Carlson, B., Kovacevic, R., 2012. Mitigation of pore generation (1), 81–93.
in laser welding of magnesium alloy AZ31B in lap joint configuration. ASME Laukant, H., Wallmann, C., Müller, M., Korte, M., Stirn, B., Haldenwanger, H.G.,
2012 International Mechanical Engineering Congress and Exposition, 919–927. Glatzel, U., 2005. Fluxless laser beam joining of aluminium with zinc coated
Hatch, J.E., 1984. Aluminum: properties and physical metallurgy. ASM steel. Sci. Technol. Weld. Join. 10 (2), 219–226.
International, Materials Park, Ohio, pp. 209. Lee, S.J., Katayama, S., Kawahito, Y., Kinoshita, K., Kim, J.D., 2013. Weldability and
Hiraga, H., Inoue, T., Kamado, S., Kojima, Y., 2002. Effects of the shielding gas and keyhole behavior of Zn-coated steel in remote welding using disk laser with
laser wavelength in laser welding magnesium alloy sheets. Weld. Int. 16 (6), scanner head. J. Laser Appl. 25 (3), 032008.
442–450. Leong, K.H., Kornecki, G., Sanders, P.G., Keske, J.S., 1998. Laser beam welding of
Holliday, R., Parker, J.D., Williams, N.T., 1996. Relative contribution of electrode tip AZ31B-H24 magnesium alloy. Proceedings of ICALEO’98, 28.
growth mechanisms in spot welding zinc coated steels. Weld. World 4 (37), Li, C., Liu, L., 2013. Investigation on weldability of magnesium alloy thin sheet
186–193. T-joints: arc welding, laser welding, and laser-arc hybrid welding. Int. J. Adv.
Hong, K.M., Shin, Y.C., 2016. Analysis of microstructure and mechanical properties Manuf. Technol. 65 (1–4), 27–34.
change in laser welding of Ti6Al4V with a multiphysics prediction model. J. Li, X., Lawson, S., Zhou, Y., Goodwin, F., 2007. Novel technique for laser lap welding
Mater. Process. Technol. 237, 420–429. of zinc coated sheet steels. J. Laser Appl. 19 (4), 259–264.
Hu, B., Richardson, I.M., 2006. Mechanism and possible solution for transverse Liedl, G., Bielak, R., Ivanova, J., Enzinger, N., Figner, G., Bruckner, J., Pasic, H., Pudar,
solidification cracking in laser welding of high strength aluminium alloys. M., Hampel, S., 2011. Joining of aluminum and steel in car body manufacturing.
Mater. Sci. Eng. A 429 (1), 287–294. Phys. Procedia 12, 150–156.
Huang, A.G., Zhang, H., Liu, J.F., Yu, W., Li, Z.Y., Li, H., 2011. Study on solidification Liu, H., Nakata, K., Yamamoto, N., Liao, J., 2012. Microstructural characteristics and
crack criterion during laser welding pure aluminum and ZL114A aluminum mechanical properties in laser beam welds of Ti6Al4V alloy. J. Mater. Sci. 47
alloy. Adv. Mater. Res. 308, 852–858. (3), 1460–1470.
Iqbal, S., Gualini, M.M., Rehman, A., 2010. Dual beam method for laser welding of Liu, J., Gao, X.L., Zhang, L.J., Zhang, J.X., 2014. A study of fatigue damage evolution
galvanized steel: experimentation and prospects. Opt. Laser Technol. 42 (1), on pulsed Nd:YAG Ti6Al4V laser welded joints. Eng. Frac. Mech. 117, 84–93.
93–98. Luo, H., Hao, C., Zhang, J., Gan, Z., Chen, H., Zhang, H., 2011. Characteristics of
Ishizaki, T., Shigematsu, I., Saito, N., 2009. Anticorrosive magnesium phosphate resistance welding magnesium alloys AZ31 and AZ91. Weld. J. 90, 249–257.
coating on AZ31 magnesium alloy. Surf. Coat. Technol. 203 (16), 2288–2291. Ma, J., Harooni, M., Carlson, B., Kovacevic, R., 2014. Dissimilar joining of galvanized
Ito, K., Kobayashi, H., 2006. Production and fabrication technology development of high-strength steel to aluminum alloy in a zero-gap lap joint configuration by
aluminum useful for automobile lightweighting. Adv. Eng. Mater. 8 (9), two-pass laser welding. Mater. Design 58, 390–401.
828–835. Ma, J., Kong, F., Carlson, B., Kovacevic, R., 2013. Two-pass laser welding of
Jain, C.C., Koo, C.H., 2007. Creep and corrosion properties of the extruded galvanized high-strength dual-phase steel for a zero-gap lap joint
magnesium alloy containing rare earth. Mater. Trans. 48 (2), 265–272. configuration. J. Mater. Process. Technol. 213 (3), 495–507.
Jeong, Y.E., Sahu, J.K., Payne, D.N., Nilsson, J., 2004. Ytterbium-doped large-core Martynyuk, M.M., 1983. Critical-point parameters of metals. Zh Fiz Khim, 810–821.
fiber laser with 1.36 kW continuous-wave output power. Opt. Exp. 12 (25), Marya, M., Edwards, G.R., 2000. The laser welding of magnesium alloy AZ91. Weld.
6088–6092. World 44 (2), 31–37.
Jin, X., Zeng, L., Cheng, Y., 2012. Direct observation of keyhole plasma Massalski, T.B., Okamoto, H., Subramanian, P.R., Kacprzak, L., 1990. Binary Alloy
characteristics in deep penetration laser welding of aluminum alloy 6016. J. Phase Diagrams, 3. ASM International, pp. p1485.
Phys. D: Appl. Phys. 45 (24), 245205. Mathers, G., 2002. The Welding of Aluminium and Its Alloys. Woodhead
Jou, M., 2003. Real time monitoring weld quality of resistance spot welding for the publishing, Cambridge.
fabrication of sheet metal assemblies. J. Mater. Process. Technol. 132 (1),
102–113.
68 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69

Mathieu, A., Pontevicci, S., Viala, J.C., Cicala, E., Matteï, S., Grevey, D., 2006. Laser Saha, D.C., Biro, E., Gerlich, A.P., Zhou, Y.N., 2016. Fiber laser welding of
brazing of a steel/aluminium assembly with hot filler wire (88% Al, 12% Si). Al-Si-coated press-hardened steel. Weld. J. 95, 147–156.
Mater. Sci. Eng. A 435, 19–28. Sahul, M., Sahul, M., Lokaj, J., 2016. Effect of surface layer on the properties of AZ31
Mathieu, A., Shabadi, R., Deschamps, A., Suery, M., Matteï, S., Grevey, D., Cicala, E., magnesium alloy welded joints. Mater. Today: Proc. 3 (4), 1150–1155.
2007. Dissimilar material joining using laser (aluminum to steel using Sánchez-Amaya, J.M., Delgado, T., De Damborenea, J.J., Lopez, V., Botana, F.J., 2013.
zinc-based filler wire). Opt. Laser Technol. 39 (3), 652–661. Laser welding of AA 5083 samples by high power diode laser. Sci. Technol.
Matsumoto, J., Mochizuki, H., 1994. Spot welding of aluminium alloy-electrode life Weld. Join., 78–86.
for various electrodes. Weld. Int. 8 (6), 438–444. Saresh, N., Pillai, M.G., Mathew, J., 2007. Investigations into the effects of electron
Matsunawa, A., 2001. Problems and solutions in deep penetration laser welding. beam welding on thick Ti–6Al–4V titanium alloy. J. Mater. Process. Technol.
Sci. Technol. Weld. Join. 6 (6), 351–354. 192, 83–88.
Matsunawa, A., Seto, N., Kim, J.D., Mizutani, M., Katayama, S., 2000. Dynamics of Schauerte, O., 2003. Titanium in automotive production. Adv. Eng. Mater. 5 (6),
keyhole and molten pool in high-power CO2 laser welding. Adv. High-Power 411–418.
Lasers Appl., 34–45. Schmidt, M., Otto, A., Kägeler, C., 2008. Analysis of YAG laser lap-welding of zinc
McQueen, H.J., Myshlaev, M., Sauerborn, M., Mwembela, A., 2000. Flow stress coated steel sheets. CIRP Ann.-Manuf. Technol. 57 (1), 213–216.
microstructures and modeling in hot extrusion of magnesium Alloys. In: Schubert, E., Klassen, M., Zerner, I., Walz, C., Sepold, G., 2001. Light-weight
Kaplan, H.I., Hryn, J.N., Clow, B.B. (Eds.), Magnesium Technology 2000. John structures produced by laser beam joining for future applications in
Wiley & Sons, Hoboken, NJ, pp. 355–362. automobile and aerospace industry. J. Mater. Process. Technol. 115 (1), 2–8.
Mei, L., Chen, G., Yan, D., Xie, D., Ge, X., Zhang, M., 2015. Impact of inter-sheet gaps Schumann, S., 2005. The paths and strategies for increased magnesium
on laser overlap welding performance for galvanised steel. J. Mater. Process. applications in vehicles. Mater. Sci. Forum 488, 1–8.
Technol. 226, 157–168. Scintilla, L.D., Tricarico, L., Brandizzi, M., Satriano, A.A., 2010. Nd:YAG laser
Mei, L., Yan, D., Yi, J., Chen, G., Ge, X., 2013. Comparative analysis on overlap weldability and mechanical properties of AZ31 magnesium alloy butt joints. J.
welding properties of fiber laser and CO2 laser for body-in-white sheets. Mater. Mater. Process. Technol. 210 (15), 2206–2214.
Design 49, 905–912. Shah, L.H., Ishak, M., 2014. Review of research progress on aluminum–steel
Milberg, J., Trautmann, A., 2009. Defect-free joining of zinc-coated steels by bifocal dissimilar welding. Mater. Manuf. Process. 29 (8), 928–933.
hybrid laser welding. Prod. Eng. 3 (1), 9–15. Sheikhi, M., Ghaini, F.M., Assadi, H., 2015. Prediction of solidification cracking in
Miller, W.S., Zhuang, L., Bottema, J., Wittebrood, A., De Smet, P., Haszler, A., pulsed laser welding of 2024 aluminum alloy. Acta Mater. 82, 491–502.
Vieregge, A., 2000. Recent development in aluminium alloys for the Sibillano, T., Ancona, A., Berardi, V., Schingaro, E., Basile, G., Lugarà, P.M., 2006. A
automotive industry. Mater. Sci. Eng. A 280 (1), 37–49. study of the shielding gas influence on the laser beam welding of AA5083
Moraitis, G.A., Labeas, G.N., 2008. Residual stress and distortion calculation of laser aluminium alloys by in-process spectroscopic investigation. Opt. Lasers Eng.
beam welding for aluminum lap joints. J. Mater. Process. Technol. 198 (1), 44 (10), 1039–1051.
260–269. Sierra, G., Peyre, P., Beaume, F.D., Stuart, D., Fras, G., 2008. Galvanised steel to
Mordike, B.L., Ebert, T., 2001. Magnesium: properties—applications—potential. aluminium joining by laser and GTAW processes. Mater. Characterization 59
Mater. Sci. Eng. A 302 (1), 37–45. (12), 1705–1715.
Muraoka, Y., Miyaoka, H., 1993. Development of an all-aluminum automotive Sierra, G., Peyre, P., Deschaux-Beaume, F., Stuart, D., Fras, G., 2007. Steel to
body. J. Mater. Process. Technol. 38 (4), 655–674. aluminium key-hole laser welding. Mater. Sci. Eng. A 447 (1), 197–208.
Novotny, S., Geiger, M., 2003. Process design for hydroforming of lightweight metal Squillace, A., Prisco, U., Ciliberto, S., Astarita, A., 2012. Effect of welding parameters
sheets at elevated temperatures. J. Mater. Process. Technol. 138 (1), 594–599. on morphology and mechanical properties of Ti–6Al–4V laser beam welded
Ola, O.T., Doern, F.E., 2015. Fusion weldability studies in aerospace AA7075-T651 butt joints. J. Mater. Process. Technol. 212 (2), 427–436.
using high-power continuous wave laser beam techniques. Mater. Design 77, Srinivasan, P.B., Riekehr, S., Blawert, C., Dietzel, W., Kocak, M., 2009. Slow strain
50–58. rate stress corrosion cracking behaviour of as-welded and plasma electrolytic
Oladimeji, O.O., Taban, E., 2016. Trend and innovations in laser beam welding of oxidation treated AZ31HP magnesium alloy autogenous laser beam weldment.
wrought aluminum alloys. Weld. World 60 (3), 415–457. Mater. Sci. Eng. A 517 (1), 197–203.
Optech Consulting, 2013. www.optics.org/news/1/1/16. Steenbergen, J.E., Thornton, H.R., 1970. Quantitative determination of the
Padmanaban, G., Balasubramanian, V., 2010. Optimization of laser beam welding conditions for hot cracking during welding for aluminum alloys. Weld. J. 49 (2),
process parameters to attain maximum tensile strength in AZ31B magnesium 61–68.
alloy. Opt. Laser Technol. 42 (8), 1253–1260. Subbaiah, K., Manivasagam, G., Shanmugarajan, B., Rao, S.R., 2013. Effect of
Padmanaban, G., Balasubramanian, V., 2011. Effects of laser beam welding welding speed on CO2 laser beam welded aluminum-magnesium alloy 5083 in
parameters on mechanical properties and microstructure of AZ31B H321 condition. Adv. Mater. Res. 685, 259–263.
magnesium alloy. Trans. Nonferrous Met. Soc. China 21 (9), 1917–1924. Sun, J., Huang, J., Yan, Q., Li, Z., 2015. Fiber laser butt joining of aluminum to steel
Paleocrassas, A.G., Tu, J.F., 2007. Low-speed laser welding of aluminum alloy using welding-brazing method. Int. J. Adv. Manuf. Technol., 1–12.
7075-T6 using a 300-W single-mode ytterbium fiber laser. Weld. J. 86 (6), Sun, Z., Ion, J.C., 1995. Laser welding of dissimilar metal combinations. J. Mater. Sci.
179–186. 30 (17), 4205–4214.
Paleocrassas, A.G., Tu, J.F., 2010. Inherent instability investigation for low speed Tan, W., Shin, Y.C., 2014. Analysis of multi-phase interaction and its effects on
laser welding of aluminum using a single-mode fiber laser. J. Mater. Process. keyhole dynamics with a multi-physics numerical model. J. Phys. D: Appl.
Technol. 210 (10), 1411–1418. Phys. 47 (34), 345501.
Papkala, H., 1992. Technological problems in spot welding of galvanized car body Tirand, G., Arvieu, C., Lacoste, E., Quenisset, J.M., 2013. Control of aluminium laser
sheet. Weld. Int. 6 (5), 341–346. welding conditions with the help of numerical modelling. J. Mater. Process.
Pastor, M., Zhao, H., DebRoy, T., 2000. Continuous Technol. 213 (3), 337–348.
wave-Nd:yttrium–aluminum–garnet laser welding of AM60B magnesium Tobar, M.J., Lamas, M.I., Yáñez, A., Sánchez-Amaya, J.M., Boukha, Z., Botana, F.J.,
alloy. J. Laser Appl. 12 (3), 91–100. 2010. Experimental and simulation studies on laser conduction welding of
Pastor, M., Zhao, H., DebRoy, T., 2001. Pore formation during continuous wave AA5083 aluminium alloys. Phys. Procedia 5, 299–308.
Nd:YAG laser welding of aluminium for automotive applications. Weld. Int. 15 Tzeng, Y.F., 2006. Gap-free lap welding of zinc-coated steel using pulsed CO2 laser.
(4), 275–281. Int. J. Adv. Manuf. Technol. 29 (3–4), 287–295.
Pastor, M., Zhao, H., Martukanitz, R.P., DebRoy, T., 1999. Porosity, underfill and Tzeng, Y.F., Chen, F.C., 2001. Effects of operating parameters on the static
magnesium lose during continuous wave Nd:YAG laser welding of thin plates properties of pulsed laser welded zinc-coated Steel. Int. J. Adv. Manuf. Technol.
of aluminum alloys 5182 and 5754. Weld. J. 78, 207–216. 18 (9), 641–647.
Pennington, E.J., 1987. U.S. Patent No. 4,642,446. Wahba, M., Katayama, S., 2012a. Laser welding of AZ31B magnesium alloy to
Peyre, P., Sierra, G., Deschaux-Beaume, F., Stuart, D., Fras, G., 2007. Generation of Zn-coated steel. Mater. Design 35, 701–706.
aluminium–steel joints with laser-induced reactive wetting. Mater. Sci. Eng. A Wahba, M., Mizutani, M., Kawahito, Y., Katayama, S., 2012b. Laser welding of
444 (1), 327–338. die-cast AZ91D magnesium alloy. Mater. Design 33, 569–576.
Pieters, R.R.G.M., Bakels, J.G., Hermans, M.J.M., Den Ouden, G., 2006. Laser welding Wang, H., Shi, Y., Gong, S., Duan, A., 2007. Effect of assist gas flow on the gas
of zinc coated steels in an edge lap configuration. J. Laser Appl. 18 (3), 199–204. shielding during laser deep penetration welding. J. Mater. Process. Technol.
Pouranvari, M., Marashi, S.P.H., 2013. Critical review of automotive steels spot 184 (1), 379–385.
welding: process, structure and properties. Sci. Technol. Weld. Join. 18 (5), Wang, S.H., Wei, M.D., Tsay, L.W., 2003. Tensile properties of LBW welds in
361–403. Ti–6Al–4V alloy at evaluated temperatures below 450C. Mater. Lett. 57 (12),
Qin, G., Lei, Z., Su, Y., Fu, B., Meng, X., Lin, S., 2014. Large spot laser assisted GMA 1815–1823.
brazing–fusion welding of aluminum alloy to galvanized steel. J. Mater. Wang, X., Wang, H.P., Lu, F., Carlson, B.E., Wu, Y., 2014. Analysis of solidification
Process. Technol. 214 (11), 2684–2692. cracking susceptibility in side-by-side dual-beam laser welding of aluminum
Quintino, L., Costa, A., Miranda, R., Yapp, D., Kumar, V., Kong, C.J., 2007. Welding alloys. Int. J. Adv. Manuf. Technol. 73 (1–4), 73–85.
with high power fiber lasers–a preliminary study. Mater. Design 28 (4), Wang, Z., Gao, M., Tang, H., Zeng, X., 2011. Characterization of AZ31B wrought
1231–1237. magnesium alloy joints welded by high power fiber laser. Mater.
Ramasamy, S., Albright, C.E., 2000. CO2 and Nd:YAG laser beam welding of 6111-T4 Characterization 62 (10), 943–951.
aluminum alloy for automotive applications. J. Laser Appl. 12 (3), 101–115. Watanabe, W., Onda, S., Tamaki, T., Itoh, K., Nishii, J., 2006. Space-selective laser
Rathod, M.J., Kutsuna, M., 2004. Joining of aluminum alloy 5052 and low-carbon joining of dissimilar transparent materials using femtosecond laser pulses.
steel by laser roll welding. Weld. J. 83 (1), 16–26. Appl. Phys. Lett. 89 (2), 021106.
 K.-M. Hong, Y.C. Shin / Journal of Materials Processing Technology 245 (2017) 46–69 69

Weisheit, A., Galun, R., Mordike, B.L., 1998. CO2 laser beam welding of Zhang, L., Zhang, J., Zhang, G., Bo, W., Gong, S., 2011. An investigation on the effects
magnesium-based alloys. Weld. Res. Suppl. 77 (4), 149–154. of side assisting gas flow and metallic vapour jet on the stability of keyhole and
Williams, N.T., Parker, J.D., 2004. Review of resistance spot welding of steel sheets molten pool during laser full-penetration welding. J. Phys. D: Appl. Phys. 44
Part 1 Modelling and control of weld nugget formation. Int. Mater. Rev. 49 (2), (13), 135201.
45–75. Zhang, M.J., Chen, G.Y., Zhang, Y., Wu, K.R., 2013. Research on microstructure and
Witzendorff, P., Hermsdorf, J., Kaierle, S., Suttmann, O., Overmeyer, L., 2015. mechanical properties of laser keyhole welding–brazing of automotive
Double pulse laser welding of 6082 aluminium alloys. Sci. Technol. Weld. Join. galvanized steel to aluminum alloy. Mater. Design 45, 24–30.
20 (1), 42–47. Zhang, X.Q., Chen, G.L., Zhang, Y.S., 2008. Characteristics of electrode wear in
Yan, S., Hong, Z., Watanabe, T., Jingguo, T., 2010. CW/PW dual-beam YAG laser resistance spot welding dual-phase steels. Mater. Design 29 (1), 279–283.
welding of steel/aluminum alloy sheets. Opt. Lasers Eng. 48 (7), 732–736. Zhao, H., DebRoy, T., 2001. Pore formation during laser beam welding of die-cast
Yang, S., Kovacevic, R., 2009. Laser welding of galvanized DP980 steel assisted by magnesium alloy AM60B-mechanism and remedy. Weld. J. 80 (8), 204–210.
the GTAW preheating in a gap-free lap joint configuration. J. Laser Appl. 21 (3), Zhao, H., White, D.R., DebRoy, T., 1999. Current issues and problems in laser
139–148. welding of automotive aluminium alloys. Int. Mater. Rev. 44 (6), 238–266.
Yang, Y.S., Lee, S.H., 1999. A study on the joining strength of laser spot welding for Zhou, D., Xu, S., Zhang, L., Peng, Y., Liu, J., 2016. Microstructure, mechanical
automotive applications. J. Mater. Process. Technol. 94 (2), 151–156. properties, and electronic simulations of steel/aluminum alloy joint during
Yao, Y., Meng, J.P., Ma, L.Y., Zhao, G.Q., Wang, L.R., 2013. Study on Hot Stamping deep penetration laser welding. Int. J. Adv. Manuf. Technol., 1–11.
and Usibor 1500P. Appl. Mech. Mater. 320, 419–425. Zhu, J., Li, L., Liu, Z., 2005. CO2 and diode laser welding of AZ31 magnesium alloy.
You, B.S., Kim, M.H., Park, W.W., Chung, I.S., 2001. Change in oxide film Appl. Surf. Sci. 247 (1), 300–306.
characteristics of magnesium alloy by the addition of calcium and minor Zuniga, S., Sheppard, S.D., 1997. Resistance spot weld failure loads and modes in
element. In: Proceedings of the International Conference on Light Materials for overload conditions. Am. Soc. Test. Mater. 1296, 469–489.
Transportation System (LIMAT-2001), Pusan, Korea, pp. 943–948.
Yunlian, Q., Ju, D., Quan, H., Liying, Z., 2000. Electron beam welding, laser beam
welding and gas tungsten arc welding of titanium sheet. Mater. Sci. Eng. A 280
(1), 177–181.