Applied Surface Science 393 (2017) 340–347

Contents lists available at ScienceDirect

Applied Surface Science

j o u r n a l ho m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / a p s u s c

Full Length Article

Corrosion of Ti6Al4V pins produced by direct metal laser sintering

J.J. de Damborenea a,∗ , M.A. Arenas a , Maria Aparecida Larosa b,c, André Luiz Jardini b,c ,
Cecília Amélia de Carvalho Zavaglia b,d , A. Conde a
a
Centro Nacional de Investigaciones Metalúrgicas (CENIM-CSIC), Avenida Gregorio del Amo, 8, E-28040 Madrid, Spain
b National Institute of Biofabrication (INCT-BIOFABRIS), State of University of Campinas (UNICAMP), Campinas, Brazil
c School of Chemical Engineering, State of University of Campinas (UNICAMP), Campinas, Brazil
d Faculty of Mechanical Engineering, State of University of Campinas (UNICAMP), Campinas, Brazil

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

Article history: Direct Metal LaserSintering (DMLS) technique allows the manufacturing awide variety of medical devices
Received 5 July 2016 for any type of prosthetic surgery (HIP, dental, cranial, maxillofacial) as well as for internal fixation
Received in revised form devices (K-Wires or Steinmann Pins). There are a large number of research studies on DMLS, including
14 September 2016
microstructural characterization, mechanical properties andthose based on production quality assurance
Accepted 5 October 2016
but the influence of porosity in the corrosion behavior of thesematerials not been sufficiently considered.
Available online 8 October 2016
In the present paper, surgical pins of Ti6Al4V have been produced by DMLS. After testing in aphosphate
buffered saline solution, thesurface ofthe titanium alloy appeared locallycovered by a voluminous white
Keywords:
Internal fixation device oxide. This unexpected behavior was presumably due to the existence of internal defects in the pins as
Ti6Al4V result of the manufacturing process. The importanceof these defects—that mightact as crevice nucleation
Cyclic voltammetry sites- has been revealed by electrochemical techniques and confirmed by computed tomography.
SEM © 2016 Elsevier B.V. All rights reserved.
Tomography

1. Introduction behaviour in Ringer’s solution of Ti-6Al-4V alloy manufactured by

DMLS. According to their findings, the DMLS samples presented
In the last decade, the emergence of new production techniques lower corrosion resistance than wrought samples due to the forma-
based in the use of lasers has drastically increased. Especially inter- tion of microscopic galvanic cells at the ␣/␤ boundaries. However,
esting is the direct fabrication of three-dimensional final products after annealing, the growth and agglomeration of ␤ phase would
through different additive processes, which attracted the atten- reduce the overall area of ␣/␤ boundaries, improving the corro-
tion of the manufacturing industry worldwide [1]. Direct Metal sion behaviour. Similarly, Dai et al. [11] have recently stressed
Laser Sintering (DMLS) is one of the most promising because of the importance of microstructure in the corrosion behaviour of
its technical versatility which allows generating products with Ti-6Al-4V samples produced by selective laser melting (SLM) in
diverse geometries that can solve a wide variety of problems, from comparison to the commercial Grade 5 alloy. As in the previous
biomedicine to aerospace industry with fully functional properties case, lower corrosion resistance was found in the laser samples
[2,3]. Among these applications, the fabrication of specific medical resulting of large acicular ␣’ and less ␤–Ti phase in the microstruc-
implants for orthopedic surgery is one of the major areas of interest ture compared to the Grade 5 sample. Nevertheless,it isworthwhile
and development [4–6]. DMLS is also able to manufacture devices to mention that these results are slightly different to those found
used in orthopedic surgery for fracture fixation, widely known as by the Authors [12], where the shape of the polarisation curves
the Steinmann pins or the Kirschner wires. Most of the papers in corresponding to the medical grade alloy and the DMLS specimens
literature on DMLS focus on design and manufacturing of differ- become almost identical.
ent prostheses and orthopedic products as well as on the study On the other hand, and due to the intrinsic characteristics
of structural and mechanical properties of these devices [7–9]. of the laser processing, defects (i.e. cracks, porosity, unmelted
However, research on corrosion behaviour is much less. Xu and powders.. .) can be easily formed in additively manufactured Ti-
co-workers [10] studied the effect of annealing on the corrosion 6Al-4V alloy. It is important to emphasize the role of these defects
because can alter its functionality, leading to failures or malfunc-
tioningduring service, particularlyon thefatigue behaviour [13,14].
Despite the achievements made in characterising the proper-
∗ Corresponding author.
ties of titanium alloys produced by DMLS, there were found no
E-mail address: [email protected] (J.J. de Damborenea).

http://dx.doi.org/10.1016/j.apsusc.2016.10.031
0169-4332/© 2016 Elsevier B.V. All rights reserved.
 J.J. de Damborenea et al. / Applied Surface Science 393 (2017) 340–347 341

Table 1
DMLS parameters.

Laser power 200 W

Laser spot diameter 50 ␮m

Scan speed 1250 mm/s (≈5 mm3 /s)
Layer thickness 0.050 mm
Protective atmosphere Ar Grade 4.8 (O2 <5 ppm)
Powder morphology Spherical; particle size ± 50 ␮m
Alloy composition (wt%) Ti (balance); Al (5.5–6.75%); V (3.5–4.5%); O
(< 2000 ppm); N (< 500 ppm);C (< 800 ppm); H
(< 150 ppm); Fe (< 3000 ppm)

references concerning the influence of the defects on the corrosion

behaviour of these alloys.
The present paper compares the corrosion behavior of Ti6Al4V
pins fabricated by DMLS and samples used for internal fixation
surgery, known as Kirschner wires. Tests were done in a phosphate
buffered saline solution. The results showed the unexpected and
random appearance of corrosion oxides on the surface on the DMLS
samples. This behavior has not been described or even mentioned
in literature despite it might be crucial for further development and
applications of these materials, especiallyin biomedicaluses. More-
over, according to Wycisk et al. [15] the literature has neglected to
investigate in detail the influence of these defects on mechanical
properties and much less on corrosion properties.

2. Materials and methods

Pins were produced by the Direct Metal Laser Sintering (DMLS)

technique using aEOSINT M 270 machine from EOS GmbH Electro
Optical Systems [12]. Specimens were fabricated from a commer-
cial pre-alloyed powder of the Ti6Al4V alloy (EOS Titanium Ti64).
Table 1 summarizes the main fabrication parameters. The pins
were designed as solid rods of 20 mm length and 1.5 mm diame-
ter, Fig. 1a. After processing, the pins were subjected to a stress
relief treatment at a temperature of 650 ◦C for 3 h (partial anneal-
ing) in a high vacuum furnace (10 −5 mbar) with subsequent furnace
cooling.
Once the pins were processed, they were chemically polished
in a mixture of HF (40 wt.%):HNO3 (70wt.%):H 2 O with volume ratio
1:4:5 at room temperature under continuous agitation at 400 rpm
for 5 min, rinsed in distilled water and dried in cold air.
Hardness and reduced modulus values from the samples were Fig. 1. a) Ti6Al4VPins processed by DMLS, 3 before chemically polished and c) after
recorded through nanoindentation measurements (Nanotest 600 chemically polished.
from Micromaterials) equipped with a Berkovich type indenter,
using 500 mN load.For that, the samples were cut lengthwise down
study (0.9cm2 area) were cut from asurgical wire of 400 mm length
the midline. Ten measurementswere taken alongthe polished cross
and 3 mm diameter. A slight pickling was made in order to remove
section of the pin to obtain average values of the Young’s modulus
the anodic layer.
and hardness.
SEM analysis was made by field emission gun scanning elec-
The electrochemical measurements were done in triplicate in
tron microscopy (FEG-SEM) utilizing a Hitachi S 4800 J instrument
a conventional three-electrode cell. The working electrode was
equipped with energy dispersive X-ray -EDX-detector. Topography
the pin of the Ti6Al4V alloy, a Ag/AgCl electrode (saturated KCl)
and roughness measurements were performed using a Sensofar pl␮
was used as reference electrode and a platinum wire as counter
2300 confocal profilometer.
electrode. Corrosion behavior was evaluated by potentiodynamic
Finally,the pore structure of the specimens wasexamined using
polarisation using a Gamry Reference 600 potentiostat. The elec-
an X-raycomputed tomography system (Nikon XT H160). The spec-
trolyte was a phosphate buffered saline solution (PBS), containing
imens weremounted on a rotary stage andscanned intheir entirety,
8.77 g/L of NaCl, 1.3 6 g/L of KH 2 PO4 , and 1.28 g/L of Na2 HPO 4 ; with
being rotated 360◦ in 0.36 ◦ equiangular steps. The images consisted
a pH range between 7.2 and 7.4. All the electrochemical tests were
of 1000 slices with a voxel size of 4.28 ␮m in all three axes.
carried out in naturally aerated conditions and at 37 ± 2 ◦C.
The potentiodynamic curves were conducted from −0.3 V vs.
OCP to +3 V vs. Ag/AgCl electrode (saturated KCl) at a scan rate of 3. Results and discussion
0.167 mV/s. Before starting the scan, the sample remained in the
solution until the open circuit potential (OCP) stabilization (up to The samples showed a rough surface characterised by the pres-
15min). ence of spherical particles on the surface of the pin (Fig. 1b). These
Corrosion test were also carried out in a commercial Kirschner particles are formed from the melt splashes of the powder that
wire of Ti6Al4 V widely used in orthopedics. The samples under adhere to the substrate. The presence of these particles in prosthe-
342 J.J. de Damborenea et al. / Applied Surface Science 393 (2017) 340–347

Fig. 2. a) Confocal image of pin surface (measured area 636.61 ×477.25 ␮m2); b) Waviness profile; c) roughness profile.
 J.J. de Damborenea et al. / Applied Surface Science 393 (2017) 340–347 343

in accordance with ASTM for Ti6Al4 V titaniumalloys Grade 5, since

martensitic transformation on Ti does not produce hardness effect
as in ferrous alloys does. The average of maximum indentation
depth was of 2.52 ±0.04 ␮m. The modulus of elasticity is slightly
lower than the annealed conventional alloy (around 110 GPa). This
structure could noticeably decrease the fatigue resistance of the
alloy even though martensitic plates are strong obstacles for micro-
crack propagation.
Fig. 4a compares the polarisation curves of the DMLS pin
and commercial Ti6Al4 V Kirschner wire. As it can be seen
both curves are similar, characterised by a long anodic passive
branch (from −0.5 to 2.5 V) with a passive current density of
8·10 −7 –1·10−6 A/cm2 . Both specimens presented a pitting poten-
tial close to 2.5 V vs Ag/AgCl. The value of the corrosion potential
for the commercial wire was of −0.425mV while for the DMLS pin
was of −0.360 mV. It is also noticeable that the polarisation curve
of the pins processed by laser showed some spikes probably due
to the surface roughness of the sample as it was demonstrated by
Burstein and Pistorious [19]. These authors, working on a passive
304 stainless steel, concluded that the greater the roughness the
greater the number of transients, due to the increase of the number
of sites where the growth of metastable corrosion pits may occur.
Unexpected behaviour was observed in DMLS samples during
the polarisation curves at potentials over 2 V. As it is pictured in
Fig. 4b, some white-orange oxide is formed on the surface during
the anodic sweep. It appears highly localised in a single point of
the surface. Further polarisation induces the growth of the oxide
very quickly until it reaches a huge volume that fall down to the
bottom of the beaker. After the removal of the sample from the
solution, a void with orange traces is seen on the pin surface, Fig. 4c.
Unfortunately, the XRD diffractograms showed that the powders
were amorphous and the stoichiometry of the oxide grown could
not be properly identified.
SEM analysis of the surface after the polarisation curves, Fig. 5,
showed the presence of a porous oxide layer primarily composed of
oxygen(41.5 wt.%) and titanium (51.96 wt.%) with a minor contri-
Fig. 3. Microstructure of the pin a) Fully acicular martensitic microstructure com-
bution of Al and V as well as traces of chlorides and sodium. These
prised of fine needles of ␣/␣’phase and b) Fine ␤-phase dispersed throughout the
matrix. figures may lead to the assumption that it is mainly a hydrated
titanium oxide which formed on the surface.
In order to establish more precisely the value of the potential at
sis devices may lead to adverse effects on human health because which these oxides are formed potentiostatic pulse tests were con-
they could become dislodged from the implant and have negative ducted, Fig. 6. Potentiostatic pulse tests were performed applying
effects elsewhere in the body. To minimise this risk, the pins were potential steps of 1.4, 1.8, 2.0, 2.3, 2.5, 2.7 and 2.9 V for 300s. The
chemically polished in a HF/HNO3 /H 2O acid solution. Then, the sur- results revealed that at 1.8 V, a small amount of white oxide grows
face appeared free of melted powders, although still revealed an locally on the surface. Contrary to that observed in the potentio-
uneven surface(Fig. 1c). As it is known, the profile is a combination dynamic curve, the oxide did not increase in size and volume with
of the form, the waviness and the roughness of the surface. Fig. 2a further potential steps and it did not change colour either. More-
pictured a representative confocal microscopy picture of the sur- over the current corresponding tothe subsequent pulses increases
face of the pin. While for flat specimens the waviness is practically just slightly. This result suggests the sweep rate and time influence
negligible, the pins presented significant values. The average wavi- on the stability of the oxide grown and its protective properties.
ness was of 5.09 ␮m (Fig. 2b), while the average roughness was of In order to verify it, cyclic voltammograms at 25 mV/s were con-
1.41 ␮m (Fig. 2c). For implants, it is thought that these values could ducted anodically from −1 to 3 V vs. Ag/AgCl electrode (saturated
promote the strength of prosthesis–body interlock [16]. KCl) and return, repeating the cycle four times. These voltammo-
This roughness can be further minimisedwith amore aggressive grams revealed a current increase at the potential range between
chemical polishing, by combining blasting and chemical etching 1.5 and 2.4 V. At 2.4 V, the current reached a peak of 8 mA/cm2 ,
[17] or even by jet blasting process [18]. Fig. 7a, developing white oxide located on the surface of the sam-
The microstructure of the pin has been analysed both, on the ple. Further increases of the potential over 2.4 V, plot a decrease in
surface and in the cross section. As it can be seen in Fig.3a, it is char- current that continues during the reverse sweep. Moreover, subse-
acterised by a fully acicular martensitic microstructure comprised quent cycles do not show this peak anymore, suggesting that the
of fine needles of ␣/␣’phase. At higher magnifications, Fig. 3b, a oxide grown during the first cycle is stable and protective, Fig. 7b.
very fine␤-phase dispersed throughout the matrix can beseen. This The amount of oxide formed and its colour remain steady until the
structure is the result of the rapid quenching achieved by the laser end of the four cycles. With reference to the cathodic scan, Fig. 7b,
beam leading to a martensitic transformation of ␤-phase, which it can be seen that after the first cycle emerges a peak between
finally causes a very fine needle-like microstructure. −0.750 and −1.25 V (vs. Ag/AgCl). This peak remains constant at the
The as-processed pins showed an average Young’s modulus of same potential and charge and it could be attributed to an oxygen
99.61 ± 1.18GPa and a hardness of 4.11 ± 0.16GPa. Both values are reduction mechanism.
344 J.J. de Damborenea et al. / Applied Surface Science 393 (2017) 340–347

Fig.4. a) Polarisation curves of the DMLS pin and commercial Ti6Al4V Kirschner wire, b) Oxide film growth andc) Attacked area after pickling.

Fig. 5. a) SEM analysis of the surface after the polarisation curves and b)qualitative composition of the oxide film.
 J.J. de Damborenea et al. / Applied Surface Science 393 (2017) 340–347 345

Fig. 6. Potentiostatic pulse test at steps of 1.4, 1.8, 2.0, 2.3, 2.5, 2.7 and 2.9 V for 300 s.

the laser additive process. In a very recent paper, Khairallah et co-

workers [20] explained—using a new mesoscopic approach- the
formation of denudation zones, spattering, and open and trapped
pore defects. These defects inherent tothe laser bed-fusion process
and the DMLS technique must be minimised to reach the quality
standards required by the industry at large, and particularly in the
biomaterials sector.
To confirm the presence of such defects, the samples were
observed by X-ray computed tomography. Fig. 8 presents the CT
slices of the transversal section (Fig. 8a) and longitudinal sections
(Fig. 8b and c) as well as the CT reconstruction of the pin (Fig.8d). As
it can be appreciated, the X-ray Computed Tomography (X-ray CT)
scans evidenced the existence of surface porosity, which is located
in the surface as well as sub-surface internal voids although with-
out connection among them. The transversal slice showed open
cylindrical pores at around 10–20 ␮m width (or even higher) and
can reach up to 100 ␮m depth. In addition, the samples presented
a rough surface.
The presence of such defects in the surface appears to condi-
tion the corrosion behaviour of the pins and be the responsible for
the anomalous behaviour, in form of voluminous and low adherent
oxides, observed on the electrochemical tests.
Such defects can behave as crevices, so the dissolved oxygen
or other oxidizing species in the bulk solution, are depleted in
the restricted volume of solution in sufficient oxidizing conditions.
Titanium chlorides formed within the crevice are unstable and
tend to hydrolyse, forming hydrochloric acid. Because of the small,
restricted volume of solution within the defects low pH levels can
develop, inducing a fast localised active corrosion within crevices
[21].
Accordingto the potential-pH equilibrium diagram[22] inacidic
conditions, the most stable species in the region of 1.2 V (1.0 V vs.
SHE) is the tetravalent ion TiO2+ . However when a specific con-
centration of these ions is reached in the solution, the following
reaction can take place:
Fig. 7. Cyclic voltammograms at 25 mV/s conducted anodically from −1 to 3 V vs.
Ag/AgCl electrode (saturated KCl), a) Developing white oxide located on the surface TiO2+ (aq) → TiO 2·H 2 O(s)
of the sample and b) 2–4 cycles.
Therefore, the passive stage observed in the polarisation curves
at the potential range of about 1.4 to 2.4 V (1.2 to 2.2 V vs. SHE)
The literature review carried out shows that neither direct and
can be attributed to local growth of the TiO2 ·H 2O, −white oxide-.
explicit references were found to explain the influence of defects
Despite this oxide is protective its solubility is still higher than TiO2.
in the corrosion behaviour of biomaterials developed by DMLS nor,
At potential values over 2.4 V, this oxide may experience further
any other evidence related to this kind of corrosion products was
oxidation to Ti (VI) according to the reaction:
mentioned. This behaviour was not observed in the samples fab-
ricated by conventional methods. The fact that this anomalous TiO2 ·H2 O(s) → TiO 2 ++(aq)
behaviour is only seen in laser processed surfaces suggests that
might be associated with the way the samples have been processed. Then an abrupt increase in the current density is depicted in
It is well known that pores and other defects are formed during the polarisation curves as result of the dissolution of the oxide.
346 J.J. de Damborenea et al. / Applied Surface Science 393 (2017) 340–347

Fig. 8. X-ray computed tomography a) CT slices of the transversal section, b,c) longitudinal sections and d) CT reconstruction of the pin.

These ions, TiO2 ++ , can deposit as a orange precipitate of peroxide, 4. Conclusions

TiO 3 ·H2 O, with increasing pH.
In early 70’s, Hintermann et al. [23,24] studied the dissolution By mean of DMLS technique is possible to fabricate pins com-
products formed due to the anodic dissolution of titanium carbide parable to those currently used in surgery similar to the Kirschner
in 2N H 2 S04. According to their findings, the corrosion of the TiC wires. Even though the mechanical properties and corrosion kinet-
produced TiO2+ . At a potential above 0.91 V vs SHE the formation ics are similar to the parent materials, an unexpected and highly
of solid oxide (TiO2 ·H2 O) started, and at higher potentials (>1.75 V) localised formation of voluminous oxides has been observed on the
the oxidation of Ti4+ to produce Ti6+ ions it is expected, resulting a defects on the pins generated during the laser melting of the pow-
severe corrosion process. In present case, these considerations are der. The mechanism of the formation of the oxides appears to be
perfectly valid. The polarisation curve (made at slow scan rate) of related to de oxidation of the TiO2 ·H2 O formed on the surface to Ti
Fig. 3 would represent the fast dissolution of the hydrated oxide (VI) at potential values over 2.4 V which is evidenced by the abrupt
film. Conversely, at the high sweep rates used in the voltammo- increase in the current density depicted in the polarisation curves.
grams (25 mV/s), just the white oxide is formed during the anodic This aspect should be duly taken into account in the formulation of
peak plotted in the first cycle, suggesting that there is afast repas- new medical devices produced by DMLS.
sivation of the defect during the anodic scan.
This would explain why a pitting potential is exhibited in the Acknowledgements
polarisation curves, while there is not current increases in the suc-
cessive cycles of the voltammograms. Both the first reverse scan Authors wish to thank to the National Council for Scientific and
of the voltammogram and the successive cycles proceed through Technological Development of the Ministry of Science, Technol-
lower current values since the complete removal of hydrated oxide ogy and Innovation (CNPq/MCTI) program Ciência Sem Fronteira
does not occur and thereforethe surface remains passive. Although (400613/2013-1) and the Project 201560E004 funded by CSIC
the thermodynamic conditions for the formation of orange oxide, for supporting this research. X-ray tomography was done by Ms
TiO 3 · H2 O, are achieved in voltammograms, it appears that kinetics Cristina Paradela at the MNCN-CSIC.
restrictions prevent the transformation of TiO 2 ·H2 O at high sweep
speeds and at potentiostatic conditions. This result would also be
References
fully in accordance with those of Peláez-Abellán et al. [25] that sug-
gested the formation of unstable titanium peroxide, TiO3 , during [1] S. Ford, Additive manufacturing technology: potential implications for U.S.
anodisation. manufacturing competitiveness, J. Int.Commerce Econ. (2014) (Published
These results suggest the role that defects could have on the electronically September) http://www.usitc.gov/journals.
[2] S.M. Thompson, L. Bian, N. Shamsaei, A. Yadollahi, An overview of Direct Laser
corrosion behaviour of biomaterials fabricated by laser additive
Deposition for additive manufacturing; Part I: transport phenomena,
techniques. The presence of surface/internal defects could be con- modelling and diagnostics, Addit. Manuf. 8 (2015) 36–62.
sidered as aserious drawback for the further development of DMLS. [3] S. Amin Yavari, R. Wauthle, A.J. Böttger, J. Schrooten, H. Weinans, A.A.
Zadpoor, Crystal structure and nanotopographical features on the surface of
However, it should be noted that the present behaviour has been
heat-treated and anodized porous titanium biomaterials produced using
revealed atvery high potentials that arenot reachedat human body selective laser melting, Appl. Surf. Sci. 290 (2014) 287–294.
conditions, although this should not minimise the risk of corrosive [4] A.L. Jardini, M.A. Larosa, C.A.C. Zavaglia, L.F. Bernardes, C. Salles-Lambert, P.
Kharmandayan, D. Calderoni, R. Maciel-Filho, Customised titanium implant
attack.
fabricated in additive manufacturing for craniomaxillofacial surgery, Virtual
 J.J. de Damborenea et al. / Applied Surface Science 393 (2017) 340–347 347

Phys. Prototyping 9 (2) (2014) 115–125, http://dx.doi.org/10.1080/17452759. [15] E. Wycisk, A. Solbach, S. Siddique, D. Herzog, F. Walther, C. Emmelmann,
2014.900857. Effects of defects in laser additivemanufactured Ti-6Al-4 V on fatigue
[5] A. Bedogni, G. Bettini,G. Ferronato, S. Fusetti, G. Saia, Replacement of fractured properties, Phys. Procedia 56 (2014) 371–378.
reconstruction plate with customized mandible implant: a novel technique, [16] P. Kwasniak, J. Pura, M. Zwolinska, P. Wiecinski, H. Skarzynski, L. Olszewski, J.
Laryngoscope 124 (2014) 401–404, http://dx.doi.org/10.1002/lary.24230. Marczak, H. Garbacz, K.J. Kurzydlowski, Laser and chemical surface
[6] F. Mangano, L. Chambrone, C. van Noort, P. Hatton, C. Mangano, Direct metal modifications of titanium grade 2 for medical application, Appl. Surf. Sci. 336
laser sintering titanium dental implants: a review of the current literature, (2015) 267–273.
Int. J. Biomater. 2014 (2014) 11, http://dx.doi.org/10.1155/2014/461534 [17] G.A. Longhitano, M.A. Larosa, A.L. Jardini, C.A.C. Zavaglia,M.C.F. Ierardi,
(Article ID 461534). Surface finishes for Ti-6Al-4 V alloy produced by direct metal laser sintering,
[7] L.S. Bertol, W. Kindlein Jr., F.P. da Silva, C. Aumund-Kopp, Medical design: Mater. Res. 18 (2015) 838–842, http://dx.doi.org/10.1590/1516-1439.014415.
direct metal laser sintering of Ti–6Al–4V, Mater. Des. 31 (2010) 3982–3988. [18] T.B. Kim, S. Yue, Z. Zhang, E. Jones, J.R. Jones, P.D. Lee, Additive manufactured
[8] T.M. Mower, M.J. Long, Mechanical behavior of additive manufactured: porous titanium structures: through-process quantification of pore and strut
powder-bed laser-fused materials, Mater. Sci. Eng. A 651 (2016) 198–213. networks, J. Mater. Process. Technol. 214 (2014) 2706–2715.
[9] I. Abd-Aziz, B. Gabbitas, M. Stanford, Microstructure and tensile strength of [19] G.T. Burstein, P.C. Pistorius, Surface roughness and the metastable pitting of
rapid manufacturing parts, Adv. Mater. Res. 903 (2014) 114–117. stainless steel in chloride solutions, Corrosion 51 (1995) 380–385.
[10] Y. Xu, Y. Lu, K.L. Sundberg, R.D. Sisson, Jr, The effect of annealing on the [20] S.A. Khairallah, A.T. Anderson, A. Rubenchik, W.E. King, Laser powder-bed
microstructure, mechanicalproperties and corrosion behavior of Ti-6Al-4 V fusion additive manufacturing: physics of complex melt flow and formation
via Direct Metal Laser Sintering, Mater. Sci. Technology Conf. Exhibition 2015 mechanisms of pores spatter, and denudation zones, Acta Mater. 108 (2016)
1 (2015) 85–92. 36–45.
[11] N. Dai, L.C. Zhang, J. Zhang, Q. Chen, M. Wu, Corrosion behavior of selective [21] R. Bhola, S.M. Bhola, B. Mishra, D.L. Olson, Corrosion in titanium dental
laser melted Ti-6Al-4 V alloy in NaCl solution, Corros. Sci. 102 (2016) 484–489. Implants/Prostheses – a review, Trends Biomater. Artif. Organs 25 (1) (2011)
[12] J. de Damborenea, M.A. Larosa, M.A. Arenas,J.M. Hernández-López, A.L. 34–46.
Jardini, M.C.F. Ierardi, C.A.C. Zavaglia, R. Maciel-Filho, A. Conde, [22] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions,
Functionalization of Ti6Al4 V scaffolds produced by direct metal laser for Pergamon press, CEBELCOR Brussels, 1996, pp. 213–221.
biomedical applications, Mater. Des. 83 (2015) 6–13. [23] R.D. Cowling, H.E. Hintermann, The corrosion of titanium carbide, J.
[13] N. Shamsaei, A. Yadollahi, L. Bian, S.M. Thompson, An overview of Direct Laser Electrochem. Soc. 117 (1970) 1447–1449.
Deposition for additive manufacturing. Part II: mechanical behavior, process [24] H.E. Hintermann, A.C. Riddiford, R.D. Cowling, J. Malyszko, The anodic
parameter optimization and control, Addit. Manuf. 8 (2015) 12–35. behaviour of titanium carbide in sulphuric acid solutions, Electrodeposition
[14] S. Leuders, M. Thöne, A. Riemer, T. Niendorf, T. Tröster, H.A. Richard, H.J. Surf. Treat. 1 (1972) 59–69.
Maier, On the mechanical behaviour of titanium alloy TiAl6V4 manufactured [25] E. Peláez-Abellán, L. Rocha-Sousa, W.-D. Müller, A.C. Guastaldi,
by selective laser melting: fatigue resistance and crack growth performance, Electrochemical stability of anodic titanium oxide films grown at potentials
Int. J. Fatigue 48 (2013) 300–307. higher than 3 V in a simulated physiological solution, Corros. Sci. 49 (2007)
1645–1655.