Laser welding of AA5083 samples by High Power Diode Laser

Abstract

Laser welding is a very attractive technique to join different alloys at the industrial level, due to Its low heat input, high flexibility, high weld quality and high production rate. In this work, the weldability of the aluminium alloy AA 5083 with a high power diode laser has been tested. Concisely, samples were subjected to lineal treatments of laser radiation, with the objective of studying the properties of the bead on plate welds generated. The main objective of the present work has been to study the influence of both the processing rate and the superficial treatment of the AA 5083 samples, on the morphological, microstructural and corrosion properties of the laser weld beads. The sizes of the welds were higher as the processing rate was decreased. The weld beads were seen to have better behaviour against corrosion than the base metal due to the microstructural refinement. It was also verified that a blasting process before processing gave beads with lower size but better corrosion resistance than the application of a black layer, due to the minimisation of the magnesium evaporation In this former superficial treatment

Full text

Laser welding of AA 5083 samples by high
power diode laser
J. M. Sa
´nchez-Amaya*
1
, T. Delgado
1
, J. J. De Damborenea
2
, V. Lopez
2
and
F. J. Botana
1
Laser welding is a very attractive technique to join different alloys at the industrial level, due to its
low heat input, high flexibility, high weld quality and high production rate. In this work, the
weldability of the aluminium alloy AA 5083 with a high power diode laser has been tested.
Concisely, samples were subjected to lineal treatments of laser radiation, with the objective of
studying the properties of the bead on plate welds generated. The main objective of the present
work has been to study the influence of both the processing rate and the superficial treatment of
the AA 5083 samples, on the morphological, microstructural and corrosion properties of the laser
weld beads. The sizes of the welds were higher as the processing rate was decreased. The weld
beads were seen to have better behaviour against corrosion than the base metal due to the
microstructural refinement. It was also verified that a blasting process before processing gave
beads with lower size but better corrosion resistance than the application of a black layer, due to
the minimisation of the magnesium evaporation in this former superficial treatment.
Keywords: Aluminium alloy, Laser welding, Bead on plate, Microstructure, Corrosion
Introduction
Laser welding (also known as laser beam welding, LBW)
is at the frontier of the welding technology, being
attractive to a lot of industrial sectors because of the
possibility of processing materials of different thick-
nesses.
1
Nowadays, the commercially available and
stable high power laser systems allow LBW to be a very
attractive and viable welding technique for joining and
repairing materials in industrial production. The advan-
tages of the LBW are its low heat input, high localisation
ability (thus leading to processed samples with small
heat affected zones), high welding speed, high flexibility,
high weld quality and high production rate.
1–4
It is widely known that there are two laser welding
mechanisms, keyhole (density power higher than
10
6
Wcm
22
) and conduction (density power lower than
10
6
Wcm
22
).
1,5
Between these two modes, keyhole
welding is most used
3,4,6,7
because it produces welds
with high aspect ratios and narrow heat affected zones.
However, keyhole welding mechanism can lead to some
problems, such as instability, keyhole oscillation and
intermittent closure of the keyhole that often produces
porosity due to gas entrapment. Additionally, in some
alloys, the high weld speed (and hence high rates of
cooling) can lead to embrittlement in the weld or heat
affected zone.
1
Advancements in laser keyhole welding
and investigations into its difficulties are well documen-
ted in literature.
5,8
Analytical, numerical and empirical
studies are being undertaken in order to achieve better
understanding of the process.
1
On the other hand, conduction welding is a more
stable process, since metal vaporisation occurs at a lower
level than at keyhole mode.
1
In addition, it offers an
alternative way of welding traditionally difficult materi-
als such as aluminium alloys.
1
Different studies have
been carried out concerning the laser welding of
aluminium alloys by means of conduction mechanism.
1
Thus, the applicability of the diode laser to the
conduction welding of AA 2024 and titanium alloys is
analysed and interpreted by thermal modelling.
9
Zhao
and DebRoy
10
developed a model to study the
vaporisation rate and weld metal composition change
of automotive aluminium alloys, based on principles of
transport phenomena, kinetics and thermodynamics.
Although several studies have reported many advan-
tages in using laser welding, there still exists the
fundamental problem that the welding reliability of
aluminium alloys is lower than that of other industrial
metals, such as steels, due to its higher reflectivity, higher
thermal conductivity and lower viscosity.
4
In fact, the
thermal conductivity of the aluminium alloys is about
one order of magnitude higher than that of steels. Thus,
the thermal conductivity of low carbon steel is
y14 W m
21
K
21
, being 143 W m
21
K
21
for the alu-
minium alloy AA 5083.
11
Another important point to
1
Departamento de Ciencia de Materiales e Ingenierı´a Metalu´rgica y
Quı´mica I norga´ nica, Grupo de Ensayos, Corrosio´n y Proteccio´n,
Universidad de Ca´ diz, Lab. 712, CASEM, Avda. Repu´ blica Saharaui s/n,
11510 Puerto Real, Ca´ diz, Spain
2
Departamento de Corrosio´n y Proteccio´n, Centro Nacional de
Investigaciones Metalu´ rgicas (CENIM-CSIC), Av. Gregorio del Amo 8,
28040 Madrid, Spain
*Corresponding author, email josemaria.sanchez@uca.es
ß2008 Institute of Materials, Minerals and Mining
Published by Maney on behalf of the Institute
Received 31 March 2008; accepted 1 August 2008
DOI 10.1179/136217108X347629 Science and Technology of Welding and Joining 2009 VOL 14 NO 178

take into account is the high reflectivity of the
aluminium alloys, which can be higher than 80%, being
more reflective as the aluminium alloy is purer. Thus, the
high reflectivity makes the aluminium alloys absorb little
fraction of the incident radiation, the high thermal
conductivity provokes a rapid heat transfer, avoiding
the concentration of energy in the weld pool and finally,
the lower viscosity of the welding pool limits the
expansion of the pool before the solidification.
Some investigations can be found regarding the study
of mechanical properties and formability of the base
metal or the laser welding characteristics of the bead on
plate or butt welding processes on aluminium alloys.
12–17
However, few studies have been published relating to the
formation of welds on these alloys.
4,18
Recently, the
influence of the laser pulse modes on the properties of
butt welds of AA 5754 and AA 6022 is analysed using a
2?5 kW high power Nd:YAG laser.
4
Concisely, two
welding modes are studied:
4
continuous welding (CW)
and pulse welding (PW). In the first one, the laser power
is kept constant at 1500 W, while in the PW, two
different powers (1000 and 2000 W; 1300 and 1700 W)
are alternating every 5 ms. This study shows that CW
mode leads to a higher penetration than the PW mode.
The alternating periods in the PW mode is thought to
induce an unstable heat input, which would result in a
dispersion of the energy in the specimen. By contrast,
the energy input remains uniform and stable when
welding is performed in the CW mode, generating
deeper penetration beads. These results are attributed to
the very high thermal conductivity of aluminium alloys,
since the heat transfer via conduction mechanism is very
efficient in these alloys.
4
Therefore, it is difficult to
concentrate the laser energy on the irradiated area, as
the heat is quickly transferred to the whole processed
piece of aluminium alloy. Taking into account their
properties (high reflectivity, high thermal conductivity
and low viscosity), it can be deduced that a relatively
high power density, in continuous laser welding mode, is
required to get an effective coupling of the laser beam
and aluminium, thus leading to an appropriate bead on
plate or butt welding of these alloys.
During the welding process of aluminium alloys, some
defects can appear in the weld bead and in the heat
affected zone, such as shallow penetration, porosities,
blowholes, inappropriate microstructural changes and
solidification cracks.
4,19,20
These defects usually decrease
the mechanical strength of the weld. Therefore, when the
laser welding process is used to join aluminium alloys, it
is essential to control the experimental laser welding
conditions in order to minimise the appearance of these
defects.
4
The optimisation of the LBW of aluminium
alloys usually involves the generation of high penetra-
tion beads and the prevention of welding defects.
Additionally, a study of the microstructure evolution
and of the mechanical properties of the welded joints
helps the evaluation of the obtained joint.
6
Aluminium alloys of the 5XXX
21,22
and 6XXX
23,24
series have been employed in the fabrication of car body
panels because of their excellent formability character-
istics and their good corrosion behaviour in comparison
with other commercial aluminium alloys.
4,20,25
These
alloys can be joined by non-conventional welding
technologies such as LBW or friction stir welding. The
great interest of these techniques is that they allow one
to obtain high quality welds of similar or even dissimilar
materials.
20
A relatively high research has been per-
formed on welding aluminium alloys with high power
CO
2
or Nd:YAG lasers, but few reports on diode laser
welding of these alloys have been found.
25
Among them,
a high power diode laser is successfully employed to
weld aluminium alloys of the series 5XXX and 6XXX.
25
This work centred its attention on the butt welding of
1 mm thickness samples, analysing the influence of the
processing rate on the depths and widths of the welds
obtained.
In the present work, AA 5083 samples have been
processed with a high power diode laser of 2?8kWin
order to generate bead on plate welds. The main
objective of the present work has been to study the
influence of both the superficial treatment and the
processing rate on the properties of the weld beads
generated by the laser radiation. In order to characterise
the beads, both the widths and the depths were
measured. In these conditions, defects are not observed
in the welds. Additionally, the microstructure and the
microhardness of the different zones of the beads (fusion
zone and base material) have been studied. Although
mechanical properties of aluminium alloys welds have
been previously reported in literature, little information
is available about other properties such as the behaviour
against corrosion. For this reason, in the present paper,
the corrosion behaviour of these zones in NaCl solutions
is also studied by means of electrochemical techniques.
Experimental
In this work, a high power diode laser of Rofin-Sinar,
with a maximum power of 2800 W, has been employed
to generate bead on plate welds on aluminium alloy
AA 5083 samples, whose composition has been included
in Table 1. The size of the samples was approximately
7062065 mm. According to the conclusions previously
commented in the Introduction section, the laser treat-
ments were performed in continuous mode and at a
relatively high value of power, 1500 W. All laser treat-
ments were performed at the beam focus spot (69?3mm
from the focusing lens). The laser focus spot shape is
included in Fig. 1, being 1?772 and 2?272 mm in
diameter in the Xand Yaxes respectively. The influence
of the processing rate on the characteristics of the
generated weld beads has been evaluated. Two different
laser processing rates were employed, v
1
51?5 m min
21
and v
2
53 m min
21
.
Different superficial treatments were applied in order
to determine the one that most increase the radiation
absorption of the alloy AA 5083. These aluminium alloy
samples were not affected by the laser radiation when
they were simply ground or polished, due to the high
reflectivity. To solve this problem, different superficial
treatments can be used in order to improve the laser
radiation absorption. In this work, three different
superficial treatments were applied to AA 5083 samples
Table 1 Chemical composition of aluminium alloy
AA 5083, wt-%
Si Fe Cu Mn Mg Zn Cr Ti Al
0?155 0?295 0?068 0?471 4?85 0?074 0?135 0?022 93?883
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 179

before being irradiated with laser, which are described as
follows:
(i) Treatment I. The samples were firstly ground up
to 80 grits, obtaining a superficial roughness of
2?2mm. Later, the samples were covered by a
black marker layer
(ii) Treatment II. The samples were initially ground
up to 80 grits and later subjected to the following
chemical cleaning treatment: the samples were
immersed in an alkaline solution of the commer-
cial product Turco 4215-NCLT, raised with
distilled water, later immersed in an acid solution
of Ardrox and finally raised again with distilled
water. The superficial roughness was 2?2mm.
Finally, as in treatment I, the samples were
covered by a black marker layer
(iii) Treatment III. The AA 5083 samples were
blasted with corindon particles, increasing their
roughness up to 50 mm.
To sum up, two laser processing rates and three
superficial treatments were employed to generate bead
on plate welds in samples of AA 5083. The codes
employed in this paper to identify each condition have
been included in Table 2. Figure 2 shows a front view of
the weld generated on one of the samples encoded as
TIIIv
2
.
Microhardness measurements
The microhardness measurements were performed on
the transversal sections of the samples, starting from the
outer zone of the weld bead (Fig. 3). The profiles of 11
microhardness values were obtained in the six conditions
abovementioned: three superficial treatments and two
laser processing rates. The Standard UNE-EN
ISO 6507-1 has been followed to perform the Vickers
microhardness profile shown in Fig. 3. The charge
employed was 0?245 N. The procedure employed
allowed the measurement of some microhardness values
in the bead zone (between three and six values, in
function of the depth of the weld bead), the rest of the
values were obtained in the base metal.
Determination of corrosion behaviour
In order to determine the corrosion behaviour of the
weld beads generated in different conditions, polarisa-
tion tests were carried out. The polarisation curves were
registered after leaving the AA 5083 samples in
3?5%NaCl at open circuit potential for 1 h, with the
objective of letting the systems corrosion potential
stabilise. After this open circuit, the samples were
polarised from 20?050 to z0?400 V with respect to
the open circuit potential, at a polarisation rate of
0?1667 mV s
21
. These measurements were performed
with a potentiostat 1287 of Solartron, controlled with
the program CorrWare 2?0 of Scribner. From these
polarisation curves, the polarisation resistance R
p
was
estimated. This parameter is inversely proportional to
the corrosion rate strictly in the case of systems which
corrode uniformly under activation control.
26
However,
R
p
is always inversely proportional to the corrosion
activity, the systems with lower electrochemical activity
having higher R
p
values.
In most AA 5083 samples processed, the width of the
weld beads obtained ranged from 2 to 3 mm. This
constitutes a limitation when the corrosion behaviour
was intended to be estimated, since the area exposed to
the NaCl solution must be small. The disadvantage of
employing small areas in electrochemical tests is that the
results show higher scattering than when higher
exposure areas are used. The area of the weld bead
samples exposed to the NaCl solution and later
polarised was 0?12 cm
2
. For comparative purposes,
samples with the same three treatments TI, TII and
1 Laser focus spot shape
Table 2 Codes employed in this paper to identify AA 5083 samples subjected to laser radiation
Sample code Superficial treatment Laser processing rate, m min
21
TIv
1
TI: ground to 80 gritszblack marker layer v
1
51?5
TIv
2
TI: ground to 80 gritszblack marker layer v
2
53
TIIv
1
TII: ground to 80 gritszchemical cleaning treatmentzblack marker layer v
1
51?5
TIIv
2
TII: Ground to 80 gritszchemical cleaning treatmentzblack marker layer v
2
53
TIIIv
1
TIII: blasted v
1
51?5
TIIIv
2
TIII: blasted v
2
53
2 Front view of weld generated on one TIIIv
2
sample
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 180

TIII without being treated with laser, were also
subjected to the polarisation tests. In these latter cases,
a higher area (1 cm
2
) was exposed in order to decrease
the scattering in the R
p
values.
Results and discussion
In continuation, the results concerning different proper-
ties of the weld beads are described: depth and width,
metallographic observations, microhardness measure-
ments and the comparison between the corrosion
behaviours of the beads and the base metal. All
treatments and measurements have been made in
triplicate to assure the reproducibility.
Depth and width of weld bead
A metallographic image of the weld bead cross-section
of the TIv
1
sample has been included in Fig. 4. In this
image, the width and depth of the weld bead have been
marked with arrows. Both parameters have been
measured in AA 5083 samples subjected to the six
treatments described before (TIv
1
,T1v
2
, TIIv
1
, TIIv
2
,
TIIIv
1
and TIIIv
2
). From the analysis of the images
obtained from all the samples, it can be concluded that
the employed conditions allow one to obtain weld beads
without signs of any defects.
The obtained values of the weld bead depth and width
have been included in Figs. 5 and 6 respectively. It can
be seen in these figures that for the three superficial
treatments, both the width and depth values are higher
when the processing rate is lower. It can be appreciated
in Figs. 5 and 6 that in samples with TI treatment, the
depth decreases from 800 mmatv
1
to 678 mmatv
2
; and
the width is 2637 mmatv
1
and 2443 mmatv
2
. Similarly,
in TII treatment, the depth and width change from 829
and 2777 mmatv
1
to 688 and 2378 mmatv
2
respectively.
Finally, in samples with TIII, the depth and width
decrease from 651 and 2251 mmatv
1
to 493 and
2015 mmatv
2
respectively. Therefore, when the rate is
lower (v
1
), the interaction time between the laser
radiation and the sample is higher and consequently,
the welding melt pool is of bigger size, provoking bead
on plate welds with higher values of depth and width.
Concisely, when the processing rate is decreased from v
2
to v
1
, the penetration increases by 18% for TI, 20% for
TII and 32% for TIII.
On the other hand, it can be also seen in Figs. 5 and 6
that at the two laser processing rates, the TIII treatment
leads to lower values of both depth and width than TI
and TII. This can be due to the black marker layer
applied in treatments I and II, which encourages the
4 Metallographic image of weld bead cross-section of
one sample of AA 5083-TIv
1
3 Schematic of experimental procedure followed to mea-
sure microhardness values in different points of weld
beads
5 Mean and standard deviation values of weld beads
depths obtained in different superficial treatments and
laser processing rates
6 Mean and standard deviation values of weld beads
widths obtained in different superficial treatments and
laser processing rates
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 181

radiation absorption. It seems that the black layer
produces higher radiation absorption that the blasting
process used in treatment III. Although the blasting
process leads to less wide and deep beads, it should be
underline that it clearly improves the radiation absorp-
tion in the absence of any black layer, since when the
samples were only ground, no weld beads were
generated. Additionally, the blasting process is consid-
ered as a clean treatment compared with other
treatments based on the application of films. Indeed,
in these latter treatments, the composition of the weld
bead could change under some conditions, because of
the possible incorporation to the bead of chemical
components of the applied coatings.
Metallographic analysis of weld beads
In Fig. 7, it has been included a micrograph (650) of
the cross-section of the weld bead of the sample
AA 5083-TIv
1
. It has been appreciated that the fusion
zone is homogeneous under the six studied conditions
(TIv
1
,T1v
2
, TIIv
1
, TIIv
2
, TIIIv
1
and TIIIv
2
). It has been
also verified that the microstructures of the beads are
similar in the six working conditions, the values of width
and depth of the fusion zone being the only differences
found when the superficial treatment and the laser
processing rate are varied. In all cases, the microstruc-
ture of the bead consists of a fine precipitation of the
second phases in a solid solution matrix of Mg in Al, as
shown in the micrograph (6500) of the fusion zone of
the sample TIv
2
(Fig. 8).
Figure 9 shows an image (6200) of the interface base
metal/fusion zone of the sample AA 5083-TIv
1
. It can be
observed in this figure that the zones of the bead close to
the base metal (central zone of the image) are
characterised by showing a dendritic growth, which
correspond to the zones with higher solidification rates.
In these zones, it has been observed that intermetallic
compounds are not completely dissolved. Although
these types of precipitates are observed in the base
metal and in the zones with dendritic structure close to
the interface base metal/fusion zone, they do not appear
in the central zones of the beads (fusion zone, Fig. 8).
According to the results obtained with EDS analysis,
these particles are rich in Al, Mn, Fe, Si and Cr, as can
be seen in the spectrum of Fig. 10a.
27–29
For compara-
tive purposes, the EDS of the aluminium matrix has
been included in Fig. 10b. The compositions of the
aluminium matrix and of these intermetallic particles
have been included in Table 3. A SEM image of the
interface base metal/fusion zone can be seen in Fig. 11,
where the Al–Mn–Fe–Si–Cr compound appears in white
colour. In the fusion zone, these intermetallic com-
pounds are dissolved during the laser processing, so that
a fine microstructure is generated.
Microhardness measurements of weld beads
In Tables 4 and 5, the Vickers microhardness values of
the samples processed at v
1
and v
2
respectively, have
been included. Note that each condition has been
repeated in triplicate in order to assure an appropriated
reproducibility. The indentations made in the weld
beads have been italicised, while the indentations in
the base metal have not been. The metallographic
observation and the microhardness values of the
processed samples have not shown evidence of heat
affected zone between the base metal and the weld
beads. In general terms, it can be observed that for all
the systems, the HV values taken in the weld beads are
higher than those measured in the base metal. In Fig. 12,
all microhardness values have been represented for a
better view of the obtained data. In this figure, it can be
seen that the microhardness values of the weld beads are
slightly higher than those measured in the base metal. In
Fig. 12, it can be seen that the difference between these
zones are y10 HV, most bead data belonging to the
7 Metallographic image (650) of cross-section of welded
bead of sample TIv
1
8 Metallographic image (6500) of cross-section of fusion
zone of sample TIv
2
9 Metallographic image (6200) of cross-section of inter-
face base metal/fusion zone of sample TIv
1
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 182

interval between 85 and 100 HV, while most metal base
data range between 75 and 90 HV. The increase in the
hardness in the weld bead is due to the microstructural
changes, reported and discussed in the previous section,
provoked by the laser heat treatment. Thus, the higher
values of hardness measured in the fusion zone could be
a consequence of the microstructure refinement.
Corrosion behaviour of weld beads
The corrosion behaviour of the weld beads formed on
AA 5083 samples with the three superficial treatments
described before (TI, TII and TIII) has been studied.
Since the samples processed at the rates v
1
and v
2
have
bead on plate welds with similar microstructure and
(a)
(b)
aintermetallic compound; baluminium matrix
10 EDS spectra of intermetallic compounds found in base metal and in interface base metal/fusion zone and of alumi-
nium matrix
Table 3 EDS semiquantitative analysis of Al–Fe–Mn–Si–
Cr intermetallic compound and aluminium matrix,
wt%
Mg Al Si Cr Mn Fe
Intermetallic compound 1?39 66?57 3?80 1?49 8?84 17?90
Aluminium matrix 4?66 94?68 0?04 0?13 0?41 0?09 11 SEM image of interface base metal/fusion zone
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 183

microhardness values, it was decided to study only the
samples at one rate (3 m min
21
). Initially, polarisation
tests were performed over single scans of laser radiation.
After these preliminary tests, the samples surfaces show
similar characteristics to the sample included in Fig. 13.
It could be seen that samples show two zones with
different activities (Fig. 14). In both figures, it can be
observed that the central zone, which belongs to weld
bead, does not show corrosion products. Meanwhile, the
outer area, which belongs to base metal, shows black
corrosion products. Therefore, it can be deduced that
the weld bead generally behaves as cathode, while the
base metal surrounding the bead normally behaves as
anode. These results agree with those recently pub-
lished,
30
where the weld fusion zone was seen to behave
as cathode and the base metal as anode within a bead on
plate laser weldment of AA 6061 base metal and
AA 4043 filler wire. The differences in the activities
between the base metal and the weld beads can be due to
the microstructure refinement produced during the laser
processing. As a consequence, this weld bead area can
behave as a cathode with respect to the base metal.
As Fig. 14 summarised, the preliminary tests per-
formed on single lineal laser treatments evaluate the
corrosion behaviour of the pair weld bead/base metal.
As commented, if the samples with only one laser scan
are subjected to the polarisation tests, the electrochemi-
cal response is not related to the weld beads, but to the
galvanic coupling between the inner and the outer zones.
In order to study the corrosion behaviour of the weld
beads, several parallel laser scans were made with the
objective of having more area of beads to perform the
polarisation tests. The separation between the consecu-
tive beads was 2 mm for the three superficial treatments,
obtaining a 25% of overlapping zones between beads.
Subsequently, the samples with various parallel beads
were subjected to the linear polarisation tests. After
these tests, the samples did not show any signs of
galvanic coupling.
The R
p
values of AA 5083 samples with different
superficial treatments and later subjected to parallel
laser scans have been plotted in Fig. 15. For each
treatment, the corrosion behaviours of the base metal
and of the parallel weld beads have been studied. When
the three non-processed samples are compared, it can be
observed that the chemical cleaning (treatment II) leads
to very low R
p
values, revealing the high activity of the
system. Therefore, before being processed with laser, the
treatment II provokes a higher susceptibility to corro-
sion than the others. On the contrary, the higher R
p
values of the base metal are reached when the samples
are blasted (treatment III). After the samples are
processed with laser, the weld beads are generated,
changing the corrosion properties. It can be appreciated
in Fig. 15 that the corrosion behaviour is clearly
Table 4 Vickers microhardness measurements at
0?025 kgf of samples processed at v
1
:
indentations made inside weld beads have been
italicised
Samples TIv
1
Samples TIIv
1
Samples TIIIv
1
96?999?1 104?792?686?8 102?291?493?985?7
91?8 101?9 104?893?793?0 101?990?293?990?2
94?697?897?795?189?299?195?491?189?1
92?093?997?790?291?495?794?492?693?9
85?389?193?989?187?995?588?189?190?2
86?889?685?785?196?184?384?287?884?3
87?983?585?782?581?585?084?684?095?1
86?886?984?578?587?985?384?284?480?9
88?179?583?580?587?485?384?286?282?7
84?186?879?582?581?579?883?990?391?2
81?983?582?885
?784?686?280?589?184?6
Table 5 Vickers microhardness measurements at
0?025 kgf of samples processed at v
2
:
indentations made inside weld beads have been
italicised
Samples TIv
2
Samples TIIv
2
Samples TIIIv
2
107?895?196?493?992?891?495?192?691?4
100?8 101?397?794?198?191?995?196?494?7
97?695?193?993?298?090?293?994?887?2
97?297?291?191?995?792?695?585?385?4
97?489?486?883?991?681?488?488?187?7
85?087?980?579?680?984?685?889?086?4
90?783?187?885?784?980?587?083?285?8
86?687?677?082?487?391?890?289?288?2
86?482?582?883?684?384?385?086?488?1
87?083?582?587?188?885?487?485?885?3
88?277?690?289?481?985?782?482?184?1
12 Vickers microhardness values taken at 0?025 kgf of
weld beads and base metal of AA 5083 samples
13 Image of zone subjected to preliminary linear polarisa-
tion test, corresponding to sample of AA 5083 with
superficial treatment II
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 184

improved in all conditions. Thus, in the three treat-
ments, the R
p
values of the weld beads are higher than
those of the metal base. This corrosion resistance
improvement is thought to be due to the microstructural
refinement that takes place during the laser treatment.
In the recent literature, it has been observed that the
weld beads of 5XXX alloys can have lower magnesium
content than the metal base. Obviously, the magnesium
composition of the weld depends on the extent of the
magnesium evaporation, which is closely related to the
processing conditions. The magnesium depletion of weld
beads generated under different laser keyhole welding
conditions has been measured in literature for different
Al–Mg alloys. Thus, weld beads of AA 5754 samples
ranged between 2?32 and 2?55 wt-%, depending on the
welding conditions, the base metal being 2?9 wt-%.
4
This
means that the magnesium loss during the laser welding
under keyhole regime ranged between 12 and 20%.
4
In
AA 6061 samples, a magnesium loss of y20% has been
also observed in the weld fusion zone, with respect to the
base metal,
31
while in AA 5083 samples, a loss of Mg
between 13 and 22% in the middle of the keyhole is
reported.
32
The contents of magnesium contained in the base
metal and in the weld beads of the AA 5083 samples
studied here have been measured by means of EDS
semiquantitative analysis. These results, given as per-
centage of magnesium and percentage of magnesium
loss, have been included in Table 6. As can be observed
in the table, the loss of magnesium in the samples
processed under conduction regime ranged between 1
and 4%, values much lower than those reported in
literature for keyhole regime. These data are expected
since the keyhole mode involves a higher input energy
density than the conduction mode. Indeed, AA 5083
samples welded under keyhole regime leaded to beads
with lower Mg content (because of the higher evapora-
tion during the laser processing) than those samples
welded under conduction mode.
1
In Table 6, the percentage of magnesium loss in each
laser treatment can be compared. It can be appreciated
that the blasting treatment (TIII) leads to lower
magnesium evaporation than the other two treatments
(TI and TII). It has been recently reported that the
magnesium evaporation during the laser processing can
cause composition changes in the weld, modifying
consequently both the mechanical properties and the
corrosion behaviour of the joint.
33
The loss of alloying
elements is normally not desirable, as the properties of
the melted zone can be different from the base metal.
34
Although very few corrosion data of weld beads are
available in literature, some results concerning mechan-
ical properties can be found. Thus, weld beads of
AA 5083 and AA 5754 with lower magnesium evapora-
tion have showed better tensile strength than those with
high evaporation.
1,4,35
Additionally, it is reported that
the main factors affecting the tensile properties and
formability of aluminium joints are the microstructure,
the magnesium evaporation and the porosity.
4
Thus, a
microstructure refinement, a lower Mg evaporation and
a porosity reduction usually improve the quality of the
weld bead.
4
The analysis of the weld beads obtained in
the present work for the different processing conditions
showed no porosities and similar microstructure refine-
ment. However, as can be seen in Table 6, lower
magnesium evaporation was measured for the blasting
treatment (TIII). Therefore, the higher magnesium
contained in these weld beads seems to be the reason
why the TIII samples show better behaviour against
corrosion than TI and TII (Fig. 15).
To sum up, the laser treatments on AA 5083 samples
lead to surfaces with lower corrosion activity than the
base metal for the three superficial treatments studied.
The microstructural refinement provokes improvements
on the corrosion behaviour of the weld beads with
respect to the base metal. Additionally, when comparing
the corrosion behaviour of the weld beads generated
under different conditions, it can be seen that the
treatment III, the blasting process, is the treatment that
leads to samples with a higher corrosion resistance in
NaCl solutions, due to the low magnesium evaporation
during the laser treatment.
14 Model of galvanic coupling behaviour between weld
bead and base metal observed during preliminary lin-
ear polarisation tests 15 R
p
values of both base metal and weld bead, obtained
with different superficial treatments
Table 6 Percentages of magnesium (in base metal and in
weld beads) and percentage of magnesium loss
in welds with respect to base metal
Sample Mg, % Mg loss, %
Base metal 4?85
Weld bead TI 4?73 2?47
Weld bead TII 4?66 3?92
Weld bead TIII 4?80 1?03
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 185

Conclusions
In the present paper, a high power diode laser working
at a continuous power of 1500 W has been employed to
generate bead on plate welds on aluminium alloy
AA 5083 samples. The influence of both the processing
rate and the superficial treatments on the morphological,
microstructural and corrosion properties of the weld
beads has been analysed.
The penetration of the weld beads were between 18
and 32% higher at 1?5 m min
21
than at 3 m min
21
. The
superficial treatments which incorporated a black
marker layer leaded to larger values of both depth and
width due to the higher radiation absorption.
The microstructural phases of the beads were similar
for the six laser processing conditions studied, not
showing porosity. The microstructure of the inner part
of the bead consisted of a fine equiaxed grains
microstructure, while the zones of the bead close to
the base metal showed a dendritic growth. Because of
the microstructural refinement of the fusion zone with
respect to the metal base, the microhardness values of
the weld beads formed in all conditions were slightly
higher than those of the metal base.
After the samples were processed with laser, the
corrosion behaviour was clearly improved for all condi-
tions, due to the microstructural refinement. The blasting
process was the superficial treatment that leaded to bead
on plate welds with better corrosion behaviour, as with
this treatment, the magnesium evaporation is minimised.
The obtained results allow one to confirm that
AA 5083 samples can be welded by means of a high
power diode laser. However, further research is neces-
sary to optimise the laser welding experimental condi-
tions, in order to obtain high quality bead on plate and
butt welds of similar and dissimilar alloys.
Acknowledgement
The present work has been financially supported by
the Ministerio de Educacio
´n y Ciencia (project
no. DPI2005-09244-C04-02) and by the Junta de
Andalucı
´a.
References
1. P. Okon, G. Dearden, K. Watkins, M. Sharp and P. French: Proc.
21st Int. Cong. on Applications of Lasers and Electro-Optics
(ICALEO 2002), Scottsdale, AZ, USA, October 2002, Laser
Institute of America, 233–241.
2. R. G. Ding, O. A. Ojo and M. C. Chaturvedi: Intermetallics, 2007,
15, 1504–1510.
3. R. Akhter, L. Ivanchev and H. P. Burger: Mater. Sci. Eng. A, 2007,
A447, 192–196.
4. T. Y. Kuo and H. C. Lin: Mater. Sci. Eng. A, 2006, A416, 281–289.
5. W. W. Duley: ‘Laser welding’, Chapter 4; 1998, New York, John
Wiley & Sons.
6. Y. Shi, F. Zhonga, X. Li, S. Gong and L. Chen: Mater. Sci. Eng. A,
2007, A465, 153–159.
7. P. Bassani, E. Capello, D. Colombo, B. Previtali and M. Vedani:
Composites Part A, 2007, 38A, 1089–1098.
8. J. F. Ready and D. F. Farson (eds.): ‘LIA handbook of laser
materials processing’; 2001, Orlando, FL, Magnolia Publishing.
9. S. W. Williams, G. Scott and N. J. Calder: Laser Opto, 2001, 33,
50–54.
10. H. Zhao and T. DebRoy: Metall. Trans. B, 2001, 32B, 163–172.
11. H. Ahmed, M. A. Wells, D. M. Maijer, B. J. Howes and M. R. van
der Winden: Mater. Sci. Eng. A, 2005, A390, (1–2), 278–290.
12. M. Pastor, H. Zhao, R. P. Martukanity and T. DebRoy: Weld. J.,
1999, 78, 207s–216s.
13. X. J. Wang, S. Katagama and A. Matsunawa: Weld. J., 1997, 76,
70s–73s
14. S. Venkat, C. E. Albright, S. Ramasamy and J. P. Hurley: Weld. J.,
1997, 76, 275s–283s.
15. S. Ramasamy and C. E. Albright: Sci. Technol. Weld. Join., 2001,
6, (3), 182–190.
16. K. M. Zhao and J. K. Lee: J. Eng. Mater. Technol., 2001, 123, (3),
287–292.
17. A. Punkari, D. C. Weckman and H. W. Kerr: Sci. Technol. Weld.
Join., 2003, 8, (4), 269–281.
18. B. Irving: Weld. J., 1991, 71, 39–45.
19. S. Ramasamy and C. E. Albright: J. Laser Appl., 2000, 12, (3), 101–
115.
20. P. Ne`gre, D. Steglich and W. Brocks: Eng. Fract. Mech., 2004, 71,
2365–2383.
21. P. J. Bolt, N. A. P. M. Lamboo and R. P. J. C. Mozier: J. Mater.
Process. Technol., 2001, 115, (1), 118–121.
22. M. Jain, J. Allin and M. J. Bull: Mater. Sci. Eng. A, 1998, A256, (1–
2), 69–82.
23. M. Z. Wang and M. E. Kassner: J. Mater. Eng. Perform., 2002, 11,
(2), 166–168.
24. R. G. Kamat, J. F. Butler, Jr, S. J. Murtha and F. S. Bovard:
Mater. Sci. Forum, 2002, 396–402, 1591–1596.
25. N. Abe, M. Tsukamoto, K. Maeda, K. Namba and J. Morimoto: J.
Laser Appl., 2006, 18, (4), 289–293.
26. J. M. Sa´nchez-Amaya, R. A. Cottis and F. J. Botana: Corros. Sci.,
2005, 47, 3280–3299.
27. A. Aballe, M. Bethencourt, F. J. Botana, M. J. Cano and M.
Marcos: Corros. Sci., 2001, 43, 1657–1674.
28. A. Aballe, M. Bethencourt, F. J. Botana, M. Marcos and J. M.
Sa´nchez-Amaya: Corros. Sci., 2004, 46, 1909–1920.
29. M. Bethencourt, F. J. Botana, M. J. Cano, M. Marcos, J. M.
Sa´nchez-Amaya and L. Gonza´ lez-Rovira: Corros. Sci., 2008, 50,
(5), 1376–1384.
30. A. B. M. Mujibur Rahman, S. Kumar and A. R. Gerson: Corros.
Sci., 2007, 49, 4339–4351.
31. A. B. M. Mujibur Rahman, S. Kumar and A. R. Gerson: Corros.
Sci., 2008, 50, 1267–1273.
32. T. Sibillano, A. Ancona, V. Berardi, E. Schingaro, P. Parente and
P. M. Lugara`: Opt. Lasers Eng., 2006, 44, 1324–1335.
33. A. Ancona, P. M. Lugara`, D. Sorgente and L. Tricarico: J. Mater.
Process. Technol., 2007, 191, 381–384.
34. T. Sibillano, A. Ancona, V. Berardi and P. M. Lugara`: Opt.
Commun., 2005, 251, 139–148.
35. L. Tricarico, R. Spina, D. Sorgente, A. Ancona, T. Sibillano and
G. Basile: Key Eng. Mater., 2007, 344, 745–750.
Sa
´nchez-Amaya et al. Laser welding of AA 5083 samples by high power diode laser
Science and Technology of Welding and Joining 2009 VOL 14 NO 186