Content uploaded by Javier Botana
Author content
All content in this area was uploaded by Javier Botana on Jul 08, 2015
Content may be subject to copyright.
Influence of Surface Pre-treatments on Laser Welding
of Ti6Al4V Alloy
J.M. Sa
´nchez-Amaya, M.R. Amaya-Va
´zquez, L. Gonza
´lez-Rovira, M. Botana-Galvin, and F.J. Botana
(Submitted October 15, 2013; in revised form February 7, 2014; published online April 8, 2014)
In the present study, Ti6Al4V samples have been welded under conduction regime by means of a high
power diode laser. The main objective of the work has been to determine the actual influence of the surface
pre-treatments on the laser welding process. Thus, six different pre-treatments were applied to Ti6Al4V
samples before performing bead-on-plate and butt welding treatments. The depth, width, microstructure,
and microhardness of the different weld zones were deeply analyzed. Grinding, sandblasting, and chemical
cleaning pre-treatments lead to welds with the highest depth values, presenting high joint strengths.
Treatments based on the application of dark coatings generate welds with lower penetration and worse
mechanical properties, specially the graphite-based coating.
Keywords laser beam welding, microstructure, surface treat-
ments, Ti6Al4V
1. Motivation/State of the Art
Some interesting properties for welding of titanium alloys are their
low thermal conductivity (lower than 22 W/cm/K), which prevents
heat dissipation, and their low coefficient of thermal expansion (around
8.6 10
6
/m/m/K), which reduces thermal stresses during welding (Ref
1). Therefore, titanium alloys generally have low heat input, which
minimizes welding stresses and reduces distortion (Ref 2). Titanium
alloys also present a good rate of laser beam absorption (0.4%) and a
high melting point (around 1670 C) (Ref 1). These properties imply
that high energy must be used to weld these alloys. Therefore, taking
into account these considerations, laser welding technology is an
adequate technology to join these alloys, as it allows a high locali zation
and low size of the melting pool, reducing considerably the required
energy in comparison with other welding technologies.
In other materials, such as aluminium alloys, various surface
treatments such as sandblasting or the application of dark coatings
are reported to improve the laser absorption, leading to significant
increase of weld penetration (Ref 3-7). Different surface pre-
treatments have been also applied to titanium alloys before
welding in order to avoid surface oxidation and to improve laser
absorption. Thus, in (Ref 1) a study focussed on improving laser
beam absorption in TiG1 (commercially pure atitanium alloy)
samples was carried out, concluding that sandblasted surfaces lead
to the highest laser beamabsorption, because they reduce the beam
reflectivity. In other work (Ref 8), the influence of several surface
preparations on the penetration of laser welds of commercially
pure titanium has been studied. Black marker and air-abrasion pre-
treatments are reported to lead to significantly deeper welds than
mirror polished samples (Ref 8). Unfortunately, such kinds of
studies are not available for other titanium alloys. In this context,
there is nota clear consensus aboutthe real influence of surface pre-
treatments on laser welding process of Ti6Al4V titanium alloy, in
terms of penetration, microstructure, and mechanical properties of
the welds. The aim of the present paper has been to cover this gap
of knowledge. Thus, the influence of different surface pre-
treatments on the laser welding process of Ti6Al4V samples has
been analyzed. In order to avoid the possible appearance of defects
(such as porosities), the laser welding treatments have been
performed with a high power diode laser (HPDL) under conduc-
tion regime. Six pre-treatments were applied to Ti6Al4V samples
before performing bead-on-plate and butt welding treatments:
ground at 80 grits (T1), sandblasted (T2), chemical cleaned (T3),
painted with black marker (T4), painted with black spray paint
(T5), and painted with graphite-based coating (T6). The beads size
and shape (depth and width), the microstructure, and microhard-
ness of the different zones were subsequently analyzed. Finally,
tensile tests were performed on some welds to analyze the weld
strength.
2. Experimental
Bead-on-plate and butt welding treatments were performed
on Ti6Al4V samples using a HPDL under conduction regime
This article is an invited submission to JMEP selected from presentations
at the Symposia ‘‘Wetting,’’ ‘‘Interface Design,’’ and ‘‘Joining
Technologies’’ belonging to the Topic ‘‘Joining and Interface Design’’
at the European Congress and Exhibition on Advanced Materials and
Processes (EUROMAT 2013), held September 8-13, 2013, in in Sevilla,
Spain, and has been expanded from the original presentation.
J.M. Sa´nchez-Amayaand L. Gonza´lez-Rovira, LABCYP,Departamento
de Ciencia de los Materiales e Ingenierı´a Metalu´rgica y Quı´mica
Inorga´nica, CASEM, Universidad de Ca´diz, Campus Rı´o San Pedro,
11510 Puerto Real (Ca´diz), Spain; and Titania, Ensayos y Proyectos
Industriales S.L. Parque Tecnolo´gico TecnoBahı´a Edif. RETSE, Nave 4.
Ctra. Sanlu´car Km 7, 11510 El Puerto de Santa Marı´a (Ca´diz), Spain;
M.R. Amaya-Va´zquez and F.J. Botana, LABCYP, Departamento de
Ciencia de los Materiales e Ingenierı´a Metalu´ rgica y Quı´mica Inorga´nica,
CASEM, Universidad de Ca´diz, Campus Rı´o San Pedro, 11510 Puerto
Real (Ca´diz), Spain; and M. Botana-Galvin, Titania, Ensayos y
Proyectos Industriales S.L. Parque Tecnolo´gico TecnoBahı´a Edif.
RETSE, Nave 4. Ctra. Sanlu´ car Km 7, 11510 El Puerto de Santa Marı´a
(Ca´diz), Spain. Contact e-mail: josemaria.sanchez@uca.es.
JMEPEG (2014) 23:1568–1575 ASM International
DOI: 10.1007/s11665-014-0967-y 1059-9495/$19.00
1568—Volume 23(5) May 2014 Journal of Materials Engineering and Performance
(ROFIN-SINAR DL028S), working at the focal distance
(69.3 mm from the focusing lens). The sample surface was
always placed at the focus position, i.e., in all experiments the
distance between the focussing lens and the samples surface
was 69.3 mm. In this condition, the HPDL provides a spot size
on surface of 2.2 91.7 mm
2
. Figure 1shows an image of this
laser equipment, the mobile XY table, and the shielding gas
system. The composition of the Ti6Al4V samples is shown in
Table 1. Three types of welds were generated: bead-on-plate of
small samples (70 914 93 mm); butt welds of small samples
(70 914 93 mm); and butt welds of bigger samples
(50 950 93 mm, Fig. 2a), to obtain standard tensile speci-
mens (Fig. 2b). Table 2shows the six surface pre-treatments
realized on Ti6Al4V samples before laser welding. Laser
treatments consisted of a linear laser scan at a constant travel
speed of 1 m/min and a power of 2500 W. Therefore, the power
density (Irradiance) was 0.067 MW/cm
2
(laser power/area of
the laser beam) and the fluence 6.82 KJ/cm
2
(laser power/
welding speed 9spot width) (Ref 6,9). Big welded samples
(for tensile tests) were performed with higher fluence to assure
full penetration (33 KJ/cm
2
). The shielding gas was argon with
a flow rate of 15 NL/min (NL are normal liters, i.e., liters
measured in normal conditions, at 1 atm and 0 C), applied
with a conical coaxial nozzle (as shown in Fig. 1).
The beads size and shape (depth and width), the micro-
structure, and the microhardness of the different zones were
analyzed. For this study, mounted cross-sections of samples
were evaluated, after polishing and etching for 10 s with Krolls
reagent (6 mL HNO
3
, 2 mL HF, 92 mL H
2
O). The morphology
and size of the weld beads were analyzed with a Leica
microscope (Model. MST53) controlled by LAS V4.2 software.
As this software can correlate measured pixels and real
distances of images, it allows the measurements of depth and
width of the welds (calibrated equipments allowed one to verify
this correlation). In order to assure the data accuracy, all
welding experiments were performed in triplicate.
Scanning electron microscopy (SEM), x-ray energy-disper-
sive spectroscopy (EDS), and x-ray diffraction (XRD) were
employed to analyze in detail the composition of some welds.
EDS chemical analyses were performed using a Phoenix-
EDAX equipment coupled at a SEM Quanta 200-FEI (FEI
Company, Hillsboro, OR). XRD was performed by a Bruker
instrument, model D8 Advance (radius 250 mm).
Microhardness measurements were accomplished with a
Duramin microhardness tester of Struers, employing a charge
of 2.945 N (0.3 HV). Finally, selected welds were subjected to
tensile tests in a Shimadzu universal testing machine (100 kN)
to evaluate the weld strength, fixing a deformation speed of
0.005 mm/min at the elastic deformation regime, and 1.6 mm/
min at the plastic deformation regime. These tensile tests were
Fig. 1 High Power Diode Laser (HPDL) head, mobile XY table
and shielding gas system
Table 1 Chemical composition of Ti6Al4V samples (wt.%)
Element Al V Fe C Element remains, maximum % Element remains, total % Ti
Ti6Al4V 5.67 4.50 0.18 0.01 £0.10 £0.40 Balance
Fig. 2 Standard tensile specimens (b) obtained from butt welds (a)
Table 2 Surface pre-treatments applied before laser
welding
Surface pre-treatments Identification
Sandblasted T1
Ground at 80 grit T2
Chemical cleaned T3
Painted with black marker T4
Painted with black spray paint T5
Painted with graphite-based coating T6
Journal of Materials Engineering and Performance Volume 23(5) May 2014—1569
performed on standard samples extracted from the welds
(Fig. 2b), to fulfill the requirements of the ‘‘small size’’
specifications of ASTM E8/E8M-11.
3. Results and Discussion
3.1 Size and Shape of Bead-On-Plate and Butt Welds
Metallographic images at low magnification (309) of bead-
on-plate beads cross-sections are included in Fig. 3. Regardless
of the applied pre-treatment, three different zones are always
identified: the fusion zone (FZ), the heat-affected zone (HAZ),
and the base metal (BM). In both bead-on-plate and butt
welding experiments, the weld shape is always close to a
semicircle, allowing one to confirm that the weld beads are
always generated under conduction regime. As can be seen in
the metallographic images, this regime gives rise to beads with
Fig. 3 Metallographic images at 309of cross-sections of bead-on-plate weld beads
Table 3 Width and depth of bead-on-plate and butt welds
Pre-treatments
Bead-on-plate Butt welding
FZ + HAZ FZ FZ + HAZ FZ
Width, mm Depth, mm Width, mm Depth, mm Width, mm Depth, mm Width, mm Depth, mm
T1 4.50 ±0.08 1.73 ±0.06 4.09 ±0.08 1.21 ±0.04 4.56 ±0.16 1.69 ±0.02 4.06 ±0.22 1.23 ±0.03
T2 4.400.03 1.72 ±0.04 4.04 ±0.05 1.25 ±0.05 4.43 ±0.20 1.71 ±0.03 4.04 ±0.19 1.27 ±0.07
T3 4.70 ±0.12 1.73 ±0.01 4.28 ±0.07 1.25 ±0.06 4.79 ±0.03 1.67 ±0.05 4.36 ±0.02 1.27 ±0.04
T4 4.76 ±0.03 1.74 ±0.02 4.35 ±0.09 0.93 ±0.04 4.75 ±0.12 1.61 ±0.05 4.22 ±0.03 0.93 ±0.07
T5 4.74 ±0.08 1.65 ±0.10 4.29 ±0.13 0.84 ±0.01 4.72 ±0.09 1.59 ±0.12 4.26 ±0.09 0.85 ±0.10
T6 4.76 ±0.09 1.67 ±0.03 4.24 ±0.03 0.83 ±0.01 ÆÆÆ ÆÆÆ ÆÆÆ ÆÆÆ
Fig. 4 Depth of (FZ + HAZ) and (FZ) of bead on plate and butt
welding
1570—Volume 23(5) May 2014 Journal of Materials Engineering and Performance
much lower porosity than the weld beads that are usually
obtained under keyhole regime (Ref 1,6,9-11).
Table 3reports the depth and width values of the obtained
welds (bead-on-plate and butt welds). In order to analyze the
welds in detail, data (depth and width) from fusion zone only
(FZ) and data from the whole heat-affected zone (including
FZ + HAZ) are reported in Table 3. The average and the
standard deviation values included in Table 3have been
Fig. 5 Examples of metallographic images (at 5009) of BM and HAZ of the welded samples
Fig. 6 Metallographic images (at 5009) of FZ of bead-on-plate beads obtained with different surface pre-treatments
Journal of Materials Engineering and Performance Volume 23(5) May 2014—1571
estimated from the analysis of three welds obtained in each
condition.
Figure 4plots the averaged penetration of the FZ and
FZ + HAZ of the different welds. As expected, the depths of
the bead-on-plate samples present the same tendency than the
butt welds, showing also similar penetration values. The depths
of the whole heat-affected zone (FZ + HAZ) are observed to be
similar in all welds, regardless of the surface treatment.
However, the penetration values of the FZ are clearly different
for each treatment. According with the microstructure and
depth of the FZ (Fig. 3,4), the welds can be classified into three
types: Type 1 corresponding to pre-treatments T1, T2, and T3;
Fig. 7 Metallographic images (at 2009and 5009) of FZ of broken T6 butt weld
Fig. 8 SEM and EDS of FZ of broken T6 butt weld
1572—Volume 23(5) May 2014 Journal of Materials Engineering and Performance
Type 2 corresponding pre-treatments T4 and T5; and finally
Type 3, corresponding to pre-treatment T6. The deepest FZ is
obtained in Type 1 welds (sandblasted, ground and chemical
cleaned samples), reaching penetration values between 1.2 and
1.3 mm. Type 2 welds (with treatments based on application of
black coatings) presented FZ with lower penetrations (0.8-
1.0 mm) than Type 1 welds. Finally, Type 3 welds
(with graphite-based coating) showed the lowest values
(0.8-0.9 mm). Note that the depths of the butt welds of this
latter treatment are not reported because they broke during the
metallographic preparation.
In summary, samples with treatments T1, T2, and T3 have
the deepest FZ, meaning that the effective heat input within the
laser welding is higher than that involved in samples with
treatments T4, T5, and T6. The higher heat input provokes
higher temperatures in the weld pool, and consequently, greater
FZs. The relatively high heat transfer in T1 and T2 is thought to
be due to the high roughness of the surfaces (encouraging the
energy absorption), while in the case of T3, the reason seems to
be related to a cleaning effect (the chemical cleaning dissolve
oxides and others contaminants from surface, increasing the
direct laser absorption). Samples with treatments T4, T5, and
T6 seem to receive lower effective heat input, probably because
of the ablation of the coatings applied to the surface.
Accordingly, the evaporated coatings material can react with
the incoming laser source and generate plasma, which can
absorb part of the beam energy. This process reduces the
effective energy received by the sample to generate the weld
pool. As a consequence, the temperatures of the liquid welds in
T4, T5, and T6 samples would be lower than in T1, T2, and T3
samples, leading to smaller melting pools and, therefore, to
smaller FZ.
3.2 Microstructure of Bead-On-Plate and Butt Welds
Figure 5includes the typical metallographic images (at
5009) of BM and HAZ of the welded samples. Similar
microstructure is observed in all bead-on-plate and butt welds
obtained with the six pre-treatments. In agreement with (Ref 9,
12), the microstructure of Ti6Al4V BM is observed to consist
of inter-granular beta phase (b, BCC structure) in an equiaxed
alpha phase (a, HCP structure). Meanwhile, the HAZ micro-
structure consists of a mixture of martensitic a¢and primary a.
Figure 6reports metallographic images at high magnifica-
tion (500X) of the FZ of bead-on-plate welds obtained with the
different pre-treatments. The FZ microstructures of all samples
(including bead-on-plate and butt welds) are identified as a¢
martensite (microhardness values presented later support the
identification of martensite). Martensitic phase could be
obtained because the welding pool temperature reached the b
transus (980 C for Ti6Al4V) and the cooling rate was higher
than 410 C/s (Ref 13,14). A careful observation of the images
allows to state that some differences can be found between the
martensitic microstructure of the six pre-treatments. Thus, FZ
of T1, T2, and T3 (Type 1) consists of acicular a¢martensite,
microstructure similar to that obtained by authors in previous
work (Ref 9). Both T4 and T5 (Type 2) presented a plate-like
martensitic microstructure with a higher density of dark
acicular particles than T1-T3. These darker particles are also
present at grain boundaries located inside the FZ of T4 and T5
samples. FZ of T6 samples (Type 3) is clearly different from the
others; small dark elongated zones are extensively incorporated
to the martensite microstructure. Deeper analyses were per-
formed to this sample in order to evaluate the possible
formation of TiC or graphite inclusions (Ref 15), as these
particles can provoke important welding embrittlement. In fact,
butt welds of T6 could not be obtained because they broke
easily. Micrographs of butt welds of T6 broken samples can be
observed in Fig. 7. Figure 8and 9report the results of the
analyses performed by SEM-EDS and XRD. The EDS spectra
(Fig. 8) did not detect the presence of carbon, while x-ray
Fig. 9 XRD of FZ of broken T6 butt weld
Fig. 10 Metallographic images (at 2009) of broken T6 butt weld
Journal of Materials Engineering and Performance Volume 23(5) May 2014—1573
diffractograms (Fig. 9) only revealed the presence of TiO
2
.
Therefore, neither TiC nor graphite was detected in T6 butt
weld sample. The reason of the premature failure of the weld is
thought to be due to the formation of alpha case. Besides the
TiO
2
layer, an alpha case layer can be clearly observed in
metallographic images included in Fig. 10. Alpha case is an
oxygen-enriched layer, hard, brittle, and considered detrimental
(Ref 16). This undesirable superficial microstructure is nor-
mally formed in titanium alloys when exposed to air at
temperatures higher than 540 C (813 K).
3.3 Microhardness Profile of Bead-On-Plate and Butt Welds
Figure 11 reports the microhardness profile of both bead-on-
plate and butt welds. As expected, in all samples, FZ has higher
microhardness values than HAZ and BM, as a consequence of
the martensitic formation. T6 treatment is observed to provide
the hardest FZ microstructure, probably due to the dark
elongated zones incorporated to the martensite microstructure
(Fig. 6).
3.4 Strength of Selected Butt Welds
Standard specimens were extracted from Type 1 butt welds,
as shown in Fig. 2, to fulfill the requirements of ASTM E8/
E8M-11. Tensile tests were performed to these samples and
also to BM probes, with the aim of verifying that the strength of
the welded sample is similar to the reference untreated sample
(BM). Yield strength (YS) and ultimate tensile strength (UTS)
values were estimated from the tests. Similar values for both
parameters were obtained (Untreated Sample: YS = 990 ±8
MPa, UTS = 1058 ±9 MPa; Weld: YS = 914 ±17 MPa,
UTS = 1050 ±20 MPa). Images of broken samples are in-
cluded in Fig. 12. Welded samples break at the BM, far from
the joint. According to these results, it can be stated that welded
samples are as strong as the reference unwelded sample, as they
show similar strength values (in fact, the welded sample broke
at the BM, confirming that the HAZ and FZ are stronger than
the BM in the welded samples).
4. Conclusions
In the present study, the influence of the surface pre-treatments
on the laser welding process of Ti6Al4V samples has been
analyzed. Six different pre-treatments were studied as follows:
ground at 80 grit (T1), sandblasted (T2), chemical cleaned (T3),
painted with black marker (T4), painted with black spray paint
(T5), and painted with graphite-based coating (T6). The beads
size and shape (depth and width), the microstructure, and the
microhardness of the different zones were analyzed. Although
martensite was detected in the FZ of all welds, the detailed
microstructure and microhardness of welds are sensitive to the
surface pre-treatment. T1 to T3 leaded to the welds with the
deepest FZ, presenting also similar strength in tensile tests than
reference unwelded samples. Treatments involving the applica-
tion of dark coatings lead to welds with lower penetration and
worse mechanical properties, especially T6, which provoke weld
hardening and embrittlement.
References
1. C. Bertrand, O. Laplanche, J.P. Rocca, Y. Le Petitcorps, and S.
Nammour, Effect of the Combination of Different Welding Parameters
on Melting Characteristics of Grade 1 Titanium with a Pulsed Nd-Yag
Laser, Lasers Med. Sci., 2007, 22, p 237–244
2. A. Costa, R. Miranda, L. Quintino, and D. Yapp, Analysis of Beam
Material Interaction in Welding of Titanium with Fiber Lasers, Mater.
Manuf. Process., 2007, 22, p 798–803
3. P. Okon, G. Dearden, K.Watkins, M. Sharp, and P. French, Laser
Welding of Aluminium Alloy 5083, 21st International Congress on
Applications of Lasers and Electro-Optics, Scottsdale, Oct 14-17,
(ICALEO 2002) ISBN 0-912035-72-2
Fig. 11 Microhardness profiles of bead-on-plate and butt welds in
function of the surface pre-treatment
Fig. 12 Tensile specimens after tests
1574—Volume 23(5) May 2014 Journal of Materials Engineering and Performance
4. W.W. Duley, Laser Welding, Chapter 4, Wiley, New York, 1999, p 69
5. K. Howard, S. Lawson, and Y. Zhou, Welding Aluminium Sheet Using
a High-Power Diode Laser, Weld. J., 2006, 85(5), p 101–110
6. J.M. Sa´nchez-Amaya, T. Delgado, J.J. De Damborenea, V. Lo´ pez, and
F.J. Botana, Laser Welding of AA 5083 Samples by High Power Diode
Laser, Sci. Technol. Weld. Join., 2009, 14(1), p 78–86
7. J. On˜oro, R. Bermejo, and Y.L. Sa´nchez-Ibarzabal, Ana´lisis de la
influencia del estado superficial en la energı´a absorbida en la
interaccio´n la´ser CO
2
-aluminio, Rev. Met. Madrid, 2005, 41, p 40–45
8. I. Watanabe, N. Baba, J. Chang, and Y. Chiu, Nd:YAG Laser
Penetration into Cast Titanium and Gold Alloy with Different Surface
Preparations, J. Oral Rehabil., 2006, 33, p 443–446
9. M.R. Amaya-Va´ zquez, J.M. Sa´ nchez-Amaya, Z. Boukha, and F.J.
Botana, Microstructure, Microhardness and Corrosion Resistance of
Remelted TiG2 and Ti6Al4V by a High Power Diode Laser, Corros.
Sci., 2012, 56, p 36–48
10. J.M. Sa´nchez-Amaya, T. Delgado, L. Gonza´lez-Rovira, and F.J.
Botana, Laser Welding of Aluminium Alloys 5083 and 6082 Under
Conduction Regime, Appl. Surf. Sci., 2009, 255(23), p 9512–9521
11. T.Y. Kuo and H.C. Lin, Effects of Pulse Level of Nd-YAG Laser on
Tensile Properties and Formability of Laser Weldments in Automotive
Aluminium Alloys, Mater. Sci. Eng. A Struct., 2006, 416, p 281–289
12. E. Akman, A. Demir, T. Canel, and T. Sınmazcelik, Laser Welding of
Ti6Al4V Titanium Alloys, J. Mater. Process. Technol., 2009, 209,
p 3705–3713
13. X. Cao and M. Jahazi, Effect of Welding Speed on Butt Joint Quality of
Ti-6Al-4V Alloy Welded Using a High-Power Nd:YAG Laser, Opt.
Laser Eng., 2009, 47, p 1231–1241
14. T. Ahmed and H.J. Rack, Phase Transformations During Cooling in
a+bTitanium Alloys, Mater. Sci. Eng. A, 1998, 243, p 206–211
15. A.F. Saleh, J.H. Abboud, and K.Y. Benyounis, Surface Carburizing of
Ti-6Al-4V Alloy by Laser Melting, Opt. Laser Eng., 2010, 48, p 257–
267
16. J.M. Sanchez-Amaya, M.R. Amaya Va´zquez, and F.J Botana, Hand-
book of Laser Welding Technologies. Chapter 8 Laser Welding of Light
Metal Alloys: Aluminium and Titanium Alloys. S. Katayama, Eds.,
Osaka University, Japan. Woodhead Publishing Series in Electronic
and Optical Materials No. 41, p 215–254
Journal of Materials Engineering and Performance Volume 23(5) May 2014—1575
