Journal of Materials Research and Technology Journal of Materials Research and Technology
Original Article
Corrosion behaviour of Ti–Ni–Al alloys in a simulated human body solution
Mara Jetzabet Garcia-Ramireza, Roy Lopez-Sesenesb, Isai Rosales-Cadenaa, Jose Gonzalo Gonzalez-Rodrigueza,,
a Uiversidad Autonoma del Esatdo de Morelos, CIICAp, Av. Universidad 1001, Col. Chamilpa, 62209-Cuernavaca., Mor., Mexico
b Uiversidad Autonoma del Esatdo de Morelos, Facultad de Ciencias Quimicas e Ingenieria, Av. Universidad 1001, Col. Chamilpa, 62209-Cuernavaca., Mor., Mexico
Received 03 February 2017, Accepted 06 July 2017
Abstract

The effect of Al additions (5, 10, 15 and 20at.%) on the corrosion behaviour of a 50at.% Ti–50Ni alloy in Hank solution has been investigated by using potentiodynamic polarization curves, electrochemical impedance spectroscopy (EIS), electrochemical noise (EN) and linear polarization resistance (LPR) measurements. Microstructure evolution was observed from columnar grains to equiaxed grains in samples with low and high Al additions, respectively. All the different techniques indicated that with the addition of either 5 or 10at.% Al not only decreases the uniform corrosion rate, but also the susceptibility to pitting corrosion. EIS technique has shown that the corrosion resistance of these alloys is due to a TiO2 layer with a duplex structure. Results have been explained in terms of the difference in microstructures for the different alloys.

Keywords
TiNiAl alloy, Hank solution, Chloride corrosion
This article is only available in PDF
1Introduction

Ti and Ti-alloys are widely used for orthopaedic and dental applications due to their low density, high corrosion resistance, good mechanical properties and biocompatibility [1–4]. Because NiTi alloys are used in the fabrication of components for orthodontic and orthopaedic applications, mechanical considerations have to be taken into account in order to establish the best alloys performance such as hardness, toughness, compression, flexion and wear resistance (among others) in agreement to the alloy design for a specific application. It is important to consider that in the human body, each part is exposed to different stress types, in such way that the component need to be evaluated in agreement to the particular condition not only in mechanical point of view but in the fluid type that will affect the corrosion resistance and for consequence the element performance. Depending upon the environment used, these alloys spontaneously develop oxides, mainly TiO2 which have an important role in determining their biocompatibility and corrosion resistance of these alloys [5–10]. Namely, the formation of a stable passive oxide film on metallic implants considerably reduces their corrosion [4]. The composition and thickness of the formed oxide determine the corrosion resistance of biometallic materials and release of metal ions from the implant that can affect health [11–13]. Various oxides are reported to form on Ti alloys surface [4,14]. The passive film on Ti implants is mainly composed of TiO2, existing in a rutile-type tetragonal structure [4].

The presence of this thermodynamically stable and adherent oxide contributes to high corrosion resistance [14]. However, the structural changes in passive films of various Ti-based implants may alter their corrosion resistance. Thus, the V2O5 in the passive film formed on Ti–6Al–4V alloy dissolves and results in the lower film corrosion resistance [1]. On the other hand, the concentration of metal release is remarkably reduced by the addition of niobium (Nb), zirconium (Zr) and tantalum (Ta) [1,3]. Since their oxides are less soluble than those of aluminium (Al) and vanadium (V), the passive oxide film on the implant surface is more inert consisting of a dense rutile structure, providing greater protection to the underlying implant material [15,16]. On the other hand, in alloys such as Ti–6Al–7Nb, the incorporation of formed oxides such as Al2O3, Nb2O5 or NbO2 into the TiO2 matrix increased this oxide corrosion resistance [17]. Consequently, new Ti alloys, containing elements such as Nb, Zr and Ta are receiving a great deal of attention [3,15].

TiNi alloys is an alloy which is widely used for orthopaedic and dental applications also due to their mechanical and corrosion resistance [18,19]. The corrosion resistance is due to the formation of a TiO2 layer which has a duplex structure, with an inner compact layer which acts as a barrier against the ion dissolution, so that the corrosion resistance increase with its thickness. There is also an outer porous layer, which is less corrosion resistance, but the incorporation of components from the electrolyte into its pores favours osteo-integration [20,21]. However, despite of its high corrosion resistance, TiNi alloys suffer from corrosion in human body fluids, with the release of Ni2+ ions which can be allergenic and toxic, which can be avoided by the addition of some alloying elements [22,23]. Light materials are currently being needed for structural applications mainly due to the low density and ductility properties [24], is for such reason that aluminium additions is the most popular element to enhance the applications for several type of materials. For Ti–Ni alloys, aluminium additions in low proportions, allows to obtain a low density material without change the biomaterial condition, in other words, to keep the biocompatibility property [25]. Aluminium additions in NiTi alloys can be considered as a high potential option to produce a material with low density characteristics, specifically in leg bones implants and hand prosthesis applications. Additionally, aluminium can be considered non-toxic since very little gets into the blood stream and it is poorly absorbed within the gastrointestinal tract [26]. Thus, in this study, the effect of Al addition on the corrosion behaviour of a TiNi alloy will be evaluated in the Hank solution by using electrochemical techniques.

2Experimental procedure.2.1Testing material

Material tested in this study was a 50at.% Ti–50Ni base alloy with the addition of 5, 10, 15 and 20at.% Al. During the alloying, Ti contents remained constant, with a value of 50at.%, with Al replacing Ni. Alloys with nominal high purity Ti–Ni–Al elements (99.98%) were fabricated by induction melting technique. In order to obtain lower amount of impurities a quartz crucible was used for melting the alloys inside of a chamber under vacuum atmosphere (10−3Torr). After reaching the liquid state, alloys were cooled down in the same crucible. After this, specimens were annealed at 700°C during 60min in order to relieve any obtained stress and to get a more homogenous microstructure.

2.2Corrosion tests

As corrosive agent, the Hank solution with a chemical composition given in Table 1 was used at 36.5°C. This solution simulates the physiological media of the human body. To perform the corrosion tests, specimens of 5×5×3mm were machined, encapsulated in epoxy resin and then ground with 600 grade emery paper. Electrochemical experiments were performed using an ACM-Instruments potentiostat controlled by a personal computer. Potentiodynamic polarization curves were obtained by varying the applied potential from −1500mV with respect to the free corrosion potential, Ecorr, up to +1000mV at a scan rate of 1mV/s. Before the experiments, the Ecorr value was measured for approximately 30min, until it was stable. All the potentials were measured using a saturated calomel electrode (SCE) as reference electrode. The counter electrode was a graphite rode. Corrosion rates were calculated in terms of the linear polarization resistance curves, LPR, which were obtained by polarizing the specimen from +10 to −10mV, respect to Ecorr, at a scan rate of 1mV/s, a standard scanning rate for this kind of experiments, to get the polarization resistance, Rp. Electrochemical impedance spectroscopy (EIS) measurements were carried out at the Ecorr value by applying a signal with an amplitude of 10mV in a frequency interval of 100mHz–100kHz. A model PC4 300 Gamry potentiostat was used for this. The susceptibility to pitting corrosion, was evaluated by using electrochemical noise (ECN) readings in both current and potential. For the ECN measurements, a zero resistance ammeter (ZRA) from ACM Instruments was used, controlled using a desk top computer. Time records consisted of blocks of 1024 readings, taken at 1s intervals, using two identical working electrodes and a SCE arrangement. All the tests were carried out at 37°C. Finally, the noise resistance value, Rn, was calculated as the ratio of potential noise standard deviation, σv, over current noise standard deviation, σi. Corroded specimens were observed in the scanning electronic microscope (SEM).

Table 1.

Chemical composition of the Hank solution (g/L).

NaCl  CaCl2  KCl  Glucose  NaHCO3  MgCl2·6H2Na2HPO4·2H2KH2PO4  MgSO47H2
0.14  0.4  0.35  0.06  0.06  0.06 
3Results and discussion3.1Microstructures

Microstructures of the different TiNiAl alloys are shown in Fig. 1a–e. For alloys containing 0, 5 and 10at.% Al, Fig. 1a–c, samples present grains with columnar type structure, and a grain size of 5, 10 and 15μm for 0, 5 and 10at.% Al, respectively. Some second phases in low proportion can be observed in the surface sample which was a consequence of the aluminium oversaturation. For samples with 15 and 20at.% Al, Fig. 1d–e, grains with an equiaxed structure and an average size of 10μm could be observed, as well as the presence of a much higher amount of second phases than that found for lower Al concentrations. For samples with 0, 5, 10, 15 and 20at.% Al addition, an average area fraction of 3, 5, 8, 13 and 18 percent was obtained respectively of the Al2NiTi secondary phase, being NiTi the predominant primary phase (B2 structure with Al in solid solution).

Fig. 1.
(0.64MB).

SEM micrographs showing the microstructures of the as-received (a) 50Ti–50Ni, (b) 50Ti–45Ni+5Al, (c) 50Ti–40Ni+10Al, (d) 50Ti–35Ni+15Al, and (e) 50Ti–30Ni+20Al alloys.

3.2Polarization curves

The effect of Al concentration in the polarization curves for the 50Ti–50Ni alloy in the Hank solution is shown in Fig. 2. It can be seen that unalloyed, base 50Ti–50Ni alloy displays an active–passive behaviour, with an Ecorr value of −850mV and a corrosion current density value close to 1.9E–5A/cm2. Similar results have reported by other authors [18,19]. A wide passive zone can be observed from −850mV up to −160, where a pitting potential, Epit is observed, reaching a passive current density mean value of 1.0E−4A/cm2. This passive zone has been attributed to the formation of a TiO2 layer, maybe with some other Ti-oxides such as TiO and Ti2O3[18,19] which has a duplex structure, with an inner compact layer, which acts as a barrier against the ion dissolution, and an outer porous layer, which, according to literature, is less corrosion resistance [27–29]. As soon as Al was added to the 50Ti–50Ni base alloy, the Ecorr value shifted towards nobler values, whereas the Icorr value decreased, maybe due to the incorporation of the Al2O3 oxide into the TiO2 film reinforcing it [17,18]. For instance, Okazaki et al. [30] found that, when Al is added to Ti and tested in physiological environments, the passive layer of the Ti alloy is reinforced by Al2O3 either in the interstitial or substitutional sites in the TiO2 matrix. The passive formed layer acts as a barrier layer against the ion dissolution. The Ecorr value reached its noblest value with the addition of 10% Al, as shown in Table 2, whereas the lowest corrosion current density value was obtained with this Al contents also.

Fig. 2.
(0.16MB).

Polarization curve for Ti–Ni–Al alloys immersed in the Hank solution at 37°C.

Table 2.

Electrochemical parameters obtained from the polarization curves for Ti–Ni–Al alloys immersed in Hank solution.

Alloy  Icorr (A/cm2Ecorr (mV)  Ipas (A/cm2Epit (mV)  Corrosion rate (mpy) 
50Ni–50Ti  1.9E−5  −840  3.4E−4  –  10 
45Ti–50Ni+5Al  2.0E−6  −850  1.0E−04  160 
50Ti–40Ni+10Al  1.6E−7  −667  9.1E−05  – 
50Ti–35Ni+15Al  5.68E−6  −342  5.0E−07  100 
50Ti–30Ni+20Al  8.78E−6  −652  6.2E−06  −42 

With a further increase in the Al contents up to 15 or 20at.%, the Ecorr shifted in the opposite direction, and the corrosion current density increased. The Icorr value for the alloy containing 20at.% Al was very similar to that for the base, unalloyed 50Ti–50Ni alloy. Corrosion current density values, Icorr, can be used to calculate the penetration rate by using Faraday's law [31]:

where d is the thickness of the removed metal, t the time, M the molecular mass, z the number of electrons involved in the corrosion reaction and ρ the metal density. The corrosion current density to get a penetration rate of 1mmy−1 for Ti, Ni and Al are 86.25×10−6, 92.07×10−6 and 91.81×10−6A/cm2 respectively. Calculated penetration rates are listed in Table 2.

The passive corrosion current density value also decreased with the addition of Al whereas the Epit value increased only with the addition of either 5 or 10at.% Al. With the addition of either 15 or 20at.% Al, the Ipas value increased, whereas the Epit value decreased, perhaps due to galvanic effect due to the presence of second phases into the alloy. Thus, the beneficial effect of the Al addition to the 50Ti–50Ni alloy is only with the addition of either 5 or 10at.% Al. A further increase in the Al contents brings a detrimental effect on the corrosion resistance of the 50Ti–50Ni alloy in the Hank solution.

3.3Linear polarization resistance measurements

In order to have a better insight of the Al effect on the corrosion behaviour on the 50Ti–50Ni alloy in the Hank solution, some 24h LPR tests were carried out, and the results are shown in Fig. 3. It can be seen that the 50Ti–50Ni base alloy has an Rp value of 4000Ωcm2 during most of the testing time, indicating the stability of the passive TiO2 film. When either 5 or 10at.% Al were added to the 50Ti–50Ni alloy, the Rp value increased, reaching values of 9E4 and 1E6Ωcm2 respectively. These values did not remain constant as time elapsed, instead they showed a decrease in a certain period of testing time indicating that the film became less protective as time elapsed. However, they remained higher than the Rp values for unalloyed 50Ti–50Ni alloy. On the other hand, the Rp value for the alloy containing 15at.% Al was lower than that for the alloy containing 10at.% Al, and it reached its lowest value with a further increase in the Al contents up to 20%, increasing, thus, the corrosion rate of the 50Ti–50Ni alloy. Maybe this fact is due to the microstructure of the different alloys. Alloys with contents of Al lower than 15at.%, microstructures contain very few second phases, but for Al contents than 15at.%, the number of second phases increases tremendously as shown in Fig. 1, inducing galvanic pairs. The number of second phases is greatest for the alloy containing 20at.%, increasing the number of galvanic cells, and thus, the corrosion rate. Thus, the beneficial effect of the Al contents on the corrosion resistance only up to 10at.% and the detrimental effect of adding either 15 or 20at.% Al was confirmed with these experiments.

Fig. 3.
(0.21MB).

Change in the Rp values with time for Ti–Ni–Al alloys immersed in the Hank solution.

3.4EN readings

The susceptibility towards localized type of corrosion such as pitting was evaluated in terms of noise in current and potential measurements. The main use of this technique is that it can tell us mainly if a metal or alloy is susceptible to a localized type of corrosion such as pitting, crevice, etc. but it can also tell us if the metal is susceptible to uniform or mixed type of corrosion, and, very important, by calculating the ratio of the potential noise standard deviation, σv, over the current noise standard deviation, σi, a parameter equivalent to the polarization resistance, in this case it is called noise resistance, Rn, it can be used to calculate the metal corrosion rate. Fig. 4 shows the time series for the noise in both current and in potential for the 50Ti–50Ni base alloy in the Hank solution, where the presence of anodic transients in the current time series simultaneously with some cathodic transients in the potential time series can be observed. The current transients intensity was around 0.1μA/cm2, whereas the potential transients intensity was lower than 1mV. This type of transients is due to the rupture of the passive layer and its repasivation, and, thus, to the nucleation of localized type of corrosion such as pits [32]. The noise in current time series for the Ti–Ni alloys with different Al contents is shown in Fig. 5. The alloy containing 5at.% Al, Fig. 5a, shows anodic current transients with lower intensity to that for unalloyed 50Ti–50Ni alloy but with a higher frequency, indicating that the protective formed film in this alloy has a lower susceptibility to be disrupted, and thus, to pitting corrosion. When the Al contents increases to 10at.%, Fig. 5b, the current transients intensity decreased markedly for up to two orders of magnitude as compared to that for unalloyed 50Ti–50Ni alloy, indicating that the susceptibility to pitting type of corrosion is significantly reduced. For alloys containing either 15 or 20at.% Al, Fig. 5c and d, the intensity of the noise transients increases once again, reaching, however, values lower or similar to those found for unalloyed 50Ti–50Ni base alloy, indicating a decreased susceptibility towards pitting type of corrosion with these alloys. Thus, in general terms, by adding Al, the susceptibility of the Ti–Ni alloy to a localized type of corrosion is greatly decreased except with the addition of 20at.% Al which does not affect it.

Fig. 4.
(0.16MB).

Noise in current and in potential for 50Ti–50Ni alloy after one day of exposure to the Hank solution.

Fig. 5.
(0.67MB).

Noise in current for (a) 50Ti–45Ni+5Al, (b) 50Ti–40Ni+10Al, (c) 50Ti–35Ni+15Al, and (d) 50Ti–30Ni+20Al alloys during the first day of exposure to the Hank solution.

The combination of standard deviation in current, σi, and standard deviation in potential, σv, allow us to calculate the noise resistance, Rn, by dividing σv over σi and thus, calculate the overall corrosion rate since Rn is inversely proportional to Icorr. The change in the Rn values with time for the different inhibitor concentrations are shown in Fig. 6, where it can be seen that addition of either 5 or 10at.% Al to the base 50Ti–50Ni alloy increases its Rn value, increasing its corrosion resistance, whereas the additions of either 15 or 20at.% Al decreases the Rn value, increasing its corrosion resistance. A very similar behaviour was observed with the Rp value, Fig. 3, which is very encouraging, since different techniques give similar results. However, since Rn is related to localized events, and thus, to localized type of corrosion. Therefore, the addition of contents lower than 10at.% Al not only increase the resistance towards uniform type of corrosion, but also towards localized type of corrosion, whereas additions of Al contents higher than 10at.% increases both the uniform corrosion rate and the susceptibility to pitting corrosion.

Fig. 6.
(0.21MB).

Change in the Rn values with time for Ti–Ni–Al alloys immersed in the Hank solution.

3.5EIS measurements

EIS measurements in both Nyquist and Bode formats for the different Ti–Ni–Al alloys are shown in Fig. 7. Nyquist plots, Fig. 7a, indicate that for unalloyed, base 50Ti–50Ni and alloy containing 20at.% Al data display a capacitive loop at high frequency values followed by a straight line at about 45°, indicating a diffusion controlled process. Alloys containing 5, 10 and 15at.% Al display, at all frequency values, a single, capacitive-like, depressed loop, which means a corrosion process controlled by charge transfer. The semicircle diameter increases with increasing the Al contents, reaching its highest value for the alloy containing 10at.% Al, decreasing with a further increase in the Al contents. Bode plots, Fig. 7b, indicates a near capacitive behaviour with a phase angle close to −80° for alloy containing 15at.% Al over a wide frequency range, indicating a passive film acting as an insulator, suggesting that a highly stable film is formed on tested alloy in the electrolyte used. In this case, phase angles during the first day of exposure is around −75°. Alloy containing 5at.% Al showed a very similar behaviour to this alloy, but the phase angle was −70° and did not remain as constant with the frequency as that for the alloy containing 10at.% Al did, indicating a passive film less stable. Alloy containing 10at.% Al exhibited two time constants illustrated by two phase angles at −81 and −79°, whereas for alloy containing 20at.% Al the two time constants were at −52 and −38°. According to the literature [8–12]. The high corrosion resistance of these alloys is due to a dense porous layer of TiO2 which has a duplex structure, with an external porous outer layer and an inner compact layer which acts as a barrier against the ion dissolution; thus, the two time constants correspond to these two layers. The corrosion resistance is attributed to the inner compact layer. Normally, a highly capacitive behaviour, typical of passive materials, is indicated from medium to low frequencies by phase angles approaching −90°, close to the phase angles shown by the alloys containing 10 or 15at.% Al. Unalloyed, base 50Ti–50Ni alloy also displayed two time constants at −50 and −38°.

Fig. 7.
(0.32MB).

EIS data in the (a) Nyquist and (b) Bode formats for the different Ti–Ni–Al alloys after one day of exposure to the Hank solution.

EIS data were simulated by the equivalent electric circuits shown in Fig. 8. For alloys containing 5, 10 or 15at.% Al, which exhibited a charge transfer controlled corrosion process, electric circuit contains two time constants, Fig. 8a, whereas for 50Ti–50Ni base alloy and for alloy containing 20at.% Al, electric circuits contain the diffusion Warburg impedance, W, Fig. 8b. In these circuits, Rs is the solution resistance, Ro and Co are the outer porous layer resistance and capacitance respectively, and Rin and Cin the inner compact layer resistance and capacitance respectively. Capacitance values can be calculated with following equation [9,24]:

where Y is the admittance. Parameters used to simulate EIS data for the different alloys are given in Tables 3 and 4. In Table 3, Rd is the resistance of the diffusion layer. From Tables 3 and 4, it can be seen that the outer, porous layer resistance values, Ro, are lower than the inner, compact layer resistance, Rin, supporting the idea that the corrosion resistance of the TiO2 layer is given by the internal, compact layer [8–12]. The highest Rin value, thus, the lowest corrosion rate, was for the alloy containing 10at.% Al, whereas the lowest values were for the 50Ti–50Ni base alloy as well as for the alloy containing 20at.% alloy. Very similar trends in these values were given by the Rp and Rn values (Figs. 3 and 6). The thickness of the scales found on corroded specimens after 30 days of exposure are given in Table 5, where it can be seen that thickest scale, and thus the specimen with the highest corrosion rate, is for base 50Ti–50Ni alloy, whereas the thinnest scale was for alloy containing 10at.% Al, as predicted by polarization data in Table 2.

Fig. 8.
(0.12MB).

Electric circuits to simulate EIS data for (a) 50Ti–50Ni and 50Ti–30Ni+20 Al alloys and (b) 50Ti–45Ni+5Al, 50Ti–40Ni+10Al and 50Ti–35 Ni+15Al alloys.

Table 3.

Electrochemical parameters used to simulate EIS data for 50Ni–50Ti and 50Ti–30Ni+20Al alloys.

Alloy  Rscm2Yo−1snno  Rocm2Rdcm2nd  Yin−1snnin  Rincm2
50Ti–50Ni  19  2.3E−4  0.66  2.0E+2  4.8E+3  0.63  3.40E−4  0.65  4.97E+3 
50Ti–30Ni+20Al  26  1.74E−5  0.81  0.6E+2  1.48E+2  0.25  9.10E−4  0.91  2.8E+3 
Table 4.

Electrochemical parameters used to simulate EIS data for 50Ti–45Ni+5Al, 50Ti–40Ni+10Al and 50Ti–30Ni+20Al alloys.

Alloy  Rscm2Yo−1snno  Rocm2Yin−1snnin  Rincm2
50Ti–45Ni+5Al  21  4.33E−5  0.80  8.6E+4  1.23E−5  0.82  8.6E+5 
50Ti–40Ni+10Al  33  8.4E−6  0.90  6.7E+4  1.4E−6  0.91  5.5E+7 
50Ti–35Ni+15Al  41  1.40E−5  0.87  295.3  7.5E−6  0.87  4.33E+6 
Table 5.

Thickness values of scales formed on corroded Ti–Ni–Al alloys immersed in Hank solution during 30 days.

Alloy  Scale thickness (μm) 
50Ni–50Ti  21 
45Ti–50Ni+5Al  10 
50Ti–40Ni+10Al 
50Ti–35Ni+15Al  11 
50Ti–30Ni+20Al  16 

Nominally, the thickness of an oxide layer, δ, can be calculated by using following expression:

where ?? and ??o are the dielectric constants for the oxide and vacuum (8.85E−12F/m) and A the effective surface area [9]. By using Eq. (3) and the admittance values given in Tables 3 and 4, it can be seen that the inner layers were thicker and more compact than the outer porous layer, which may be due to the formation of aluminium oxides and their incorporation in to the TiO2 passive layer [17,18,30], which gives more protection against the diffusion of the electrolyte. The base 50Ti–50Ni alloy and the alloy containing 20at.% Al had the highest Yin values, and, thus, the thinnest inner layer, therefore the lowest corrosion resistance. Due to a thicker and porous outer layer in these alloys, slower diffusion processes appear, which is the rate controlling step. The nin values were in all cases either similar or lower than those for no, indicating a less capacitive behaviour of the inner layer, suggesting that the mass diffusion through the inner layer is the rate controlling step [22,25]. Only the alloy containing 10at.% Al had an n value close to 1.0 (0.91) for the inner layer, indicating a nearly ideal capacitive behaviour, resulting in a highly corrosion resistant alloy. For this alloy, the porous layer was thicker than the inner layer, probably due to the incorporation of the electrolyte components into the pores of the outer porous layer, resulting in an increased corrosion resistant of this oxide layer.

4Conclusions

The effect of adding Al (5–20at.%) to a 50at.% Ti–50Ni base alloy in its corrosion resistance in a Hank solution has been evaluated by electrochemical techniques. All different techniques have shown that the best corrosion resistance properties were achieved with the addition of either 10 or 15at.% Al. Also, the pitting resistance increased for Al contents lower than 15at.%. Finally, EIS data indicated that corrosion mechanism is under diffusion control for 0 and 20at.% Al, and under charge transfer control for the rest of the alloys. Modelling the EIS data are consistent with a duplex structure of the passive TiO2 layer, consisting of an inner, and an outer, porous layer.

Conflicts of interest

The authors declare no conflicts of interest.

References
[1]
S. Miyazaki,H.Y. Kim,H. Hosoda
Development and characterization of Ni-free Ti-base shape memory and superelastic alloys
Mater Sci Eng A, 438 (1998), pp. 18-24
[2]
J.A. Hunt,M. Stoiche
Biomaterials: drug delivery systems
Curr Opin Solid State Mater Sci, 5 (2001), pp. 161-162
[3]
C. Fonseca,M.A. Barbosa
Corrosion behaviour of titanium in biofluids containing H2O2 studied by electrochemical impedance spectroscopy
Corros Sci, 43 (2001), pp. 547-559
[4]
M. Long,H.J. Rac
Titanium alloys in total joint replacement—a materials science perspective
Biomaterials, 19 (1998), pp. 1621-1639
[5]
L. Bukovinová,B. Hadzima
Electrochemical characteristics of magnesium alloy AZ31 in Hank's solution
Corros Eng Sci Technol, 47 (2012), pp. 357-362
[6]
L. Mohan,C. Anandan,N. Rajendran
Electrochemical behaviour and bioactivity of self-organized TiO2 nanotube arrays on Ti–6Al–4V in Hanks’ solution for biomedical applications
Electrochim Acta, 155 (2015), pp. 411-420
[7]
C. Hong,J. Tu,D. Liu,R. Li,C. Gu
The electrochemical and mechanical properties of Ti incorporated amorphous carbon films in Hanks’ solution
Appl Surf Sci, 256 (2010), pp. 4859-4866
[8]
N.A. Al-Mobarak,A.A. Al-Swayih
Development of titanium surgery implants for improving osseointegration through formation of a titanium nanotube layer
Int J Electrochem Sci, 9 (2014), pp. 32-45
[9]
M. Geetha,A.K. Singh,R. Asokamani,A.K. Gogia
Ti based biomaterials, the ultimate choice for orthopaedic implants – a review
Prog Mater Sci, 54 (2009), pp. 397-425
[10]
T.C. Niemeyer,C.R. Grandini,L.M.C. Pinto,A.C.D. Angelo,S.G. Schneider
Corrosion behaviour of Ti–13Nb–13Zr alloy used as a biomaterial
J Alloy Compd, 476 (2009), pp. 172-175
[11]
S.G. Steinemann
Corrosion of surgical implants – in vivo and in vitro. Tests
Evaluation of biomaterials – advances in biomaterials,
[12]
P.G. Laing,A.B. Ferguson Jr.,E.S. Hodge
Tissue reaction in rabbit muscle exposed to metallic implants
J Biomed Mater Res, 1 (1967), pp. 135-149 http://dx.doi.org/10.1002/jbm.820010113
[13]
D.R.C. McLachlan,B. Farnell,H. Galin
Aluminum in human brain disease
Biological aspects of metals and metal-related diseases,
[14]
R. Singh,N.B. Dahotre-Bhola
Corrosion degradation and prevention by surface modification of biometallic materials
J Mater Sci: Mater M, 18 (2007), pp. 725-751
[15]
M.A. Khan,R.L. Williams,D.F. Williams
The corrosion behaviour of Ti–6Al–4V, Ti–6Al–7Nb and Ti–13Nb–13Zr in protein solutions
Biomater, 20 (1999), pp. 631-637
[16]
J.A. Davidson,A.K. Mishra,P. Kovacks,R.A. Poggie
New surface-hardened, low modulus, corrosion-resistant Ti–13Nb–13Zr alloy for total hip arthroplasty
Biomed Mater Eng, 4 (1994), pp. 231-243
[17]
I. Milosev,T. Kosec,H.H. Strehblow
XPS and EIS study of the passive film formed on orthopaedic Ti–6Al–7Nb alloy in Hank's physiological solution
Electrochim Acta, 53 (2008), pp. 3547-3558
[18]
S.A. Shabalovskaya
On the nature of the biocompatibility and medical applications of NiTi shape memory and superelastic alloys
Biomed Mater Eng, 6 (1996), pp. 267-289
[19]
S.A. Shabalovskaya
Surface, corrosion, and biocompatibility aspects of nitinol as an implant material
Biomed Mater Eng, 12 (2002), pp. 69-109
[20]
E. Vasilescu,P. Drob,D. Raducanu,I. Cinca,D. Mareci,J.M. Calderon Moreno
Effect of thermo-mechanical processing on the corrosion resistance of Ti6Al4V alloys in biofluids
Corros Sci, 51 (2009), pp. 2885-2896
[21]
M.V. Popa,D. Raducanu,E. Vasilescu,P. Drob,D. Cojocaru,C. Vasilescu
Mechanical and corrosion behaviour of a Ti–Al–Nb alloy after deformation at elevated temperatures
Mater Corros, 59 (2008), pp. 919-928
[22]
L. Mohan,D. Durgalakshmi,M. Geetha,T.S.N. Sankara Narayanan,R. Asokamani
Electrophoretic deposition of nanocomposite (Hap/TiO2) on titanium alloy for biomedical applications
Ceram Int, 34 (2012), pp. 3438-3443
[23]
V.A. Alves,R.Q. Reis,I.C.B. Santos,D.G. Souza,T.F. Gonçalves,M.A. Pereira-da-Silva
In situ impedance spectroscopy study of the electrochemical corrosion of Ti and Ti–6Al–4V in simulated body fluid at 25 and 37°C
Corros Sci, 51 (2009), pp. 2473-2482
[24]
M. Jaradeh,T. Carlberg
Effect of titanium additions on the microstructure of DC-cast aluminium alloys
Mater Sci Eng, 413A (2005), pp. 277-282
[25]
T. Kurita,H. Matsumoto,K. Sakamoto,K. Tanji,H. Abe
Effect of aluminum addition on the transformation of NiTi alloy
J Alloy Compd, 396 (2005), pp. 193-196
[26]
S. Piazza,G. Lo Biundo,M.C. Romano,C. Sunseri,F. Di Quatro
In situ characterization of passive films on Al–Ti alloy by photocurrent and impedance spectroscopy
Corros Sci, 40 (1998), pp. 1087-1108
[27]
I.C. Lavos-Valereto,S. Wolynec,I. Ramires,A.C. Guastaldi,I. Costa
Electrochemical impedance spectroscopy characterization of passive film formed on implant Ti–6Al–7Nb alloy in Hank's solution
J Mater Sci: Mater M, 15 (2004), pp. 55-59
[28]
J. Pan,H. Liao,C. Leygraf,D. Thierry,J. Li
Variation of oxide films on titanium induced by osteoblast-like cell culture and the influence of an H2O2 pre-treatment
J Biomed Mater Res, 40 (1998), pp. 244-256
[29]
X. Liu,P.K. Chu,C. Ding
Critical overview of nitinol surfaces and their modifications for medical applications
Mater Sci Eng, 47R (2004), pp. 49-121
[30]
Y. Okazaki,T. Tateishi,Y. Ito
Corrosion resistance of implant alloys in pseudo physiological solution and role of alloying elements in passive films
Mater Trans JIM, 38 (1997), pp. 78-86
[31]
L.L. Shreir
Corrosion
Butterworth and Co. Publishers, (1977)
[32]
K. Hladky,J.L. Dawson
The measurement of localized corrosion using electrochemical noise
Corros Sci, 22 (1982), pp. 231-237
Corresponding author. (Jose Gonzalo Gonzalez-Rodriguez ggonzalez@uaem.mx)