Effect of replacement of vanadium by iron on the

REVISTA DE METALURGIA, 45 (1)ENERO-FEBRERO, 32-41, 2009

ISSN: 0034-8570eISSN: 1988-4222

doi: 10.3989/revmetalm.0750

32

EEffffeecctt ooff rreeppllaacceemmeenntt ooff vvaannaaddiiuumm bbyy iirroonn oonn tthhee eelleeccttrroocchheemmiiccaallbbeehhaavviioouurr ooff ttiittaanniiuumm aallllooyyss iinn ssiimmuullaatteedd pphhyyssiioollooggiiccaall mmeeddiiaa((••))

D. Mareci*, V. Lucero** and J. Mirza**

AAbbssttrraacctt The electrochemical behaviour of Ti6Al4V, Ti6Al3.5Fe and Ti5Al2.5Fe alloys has been evaluated in Ringer’s solutionat 25 °C. The effect of the substitution of vanadium in Ti6Al4V alloy has been specifically addressed. The evaluationof the corrosion resistance was carried out through the analysis of the open circuit potential variation with time,potentiodynamic polarization curves, and electrochemical impedance spectroscopy (EIS) tests. Very low currentdensities were obtained (order of nA/cm2) from the polarization curves and EIS, indicating a typical passive behaviourfor all investigated alloys. The EIS results exhibited relative capacitive behaviour (large corrosion resistance) withphase angle close to –80° and relative high impedance values (order of 105 Ω·cm2) at low and medium frequencies,which are indicative of the formation of a highly stable film on these alloys in Ringer’s solution. In conclusion, theelectrochemical behaviour of Ti6Al4V is not affected by the substitution of vanadium with iron.

KKeeyywwoorrddss Titanium; Alloys; Biocompatibility; Corrosion; Polarization; Impedance.

EEffeeccttoo ddee llaa ssuusstt ii ttuucciióónn ddee vvaannaaddiioo ppoorr hhiieerrrroo eenn eell ccoommppoorr ttaammiieennttooeelleeccttrrooqquuíímmiiccoo ddee aalleecciioonneess ddee ttiittaanniioo eenn uunn mmeeddiioo ffiissiioollóóggiiccoo ssiimmuullaaddoo

RReessuummeenn El comportamiento electroquímico de las aleaciones Ti6Al4V, Ti6Al3.5Fe y Ti5Al2.5Fe fue evaluado en una disolu-ción Ringer a 25 °C. Se ha estudiado especialmente el efecto de la sustitución del vanadio en la aleación Ti6Al4V. Laevaluación de la resistencia a la corrosión se ha llevado a cabo a través del análisis de la variación del potencial de un cir-cuito abierto con el tiempo, las curvas de polarización potenciodinámicas y los ensayos de espectroscopía de impedan-cia electroquímica (EIS). Se han obtenido densidades de corriente muy bajas (del orden de nA/cm2) en las curvas de po-larización y EIS, indicando un comportamiento pasivo típico para todas las aleaciones investigadas. Los resultados de laEIS mostraron un comportamiento capacitivo relativo (gran resistencia a la corrosión) con ángulos de fase próximos a–80° y valores de impedancia relativamente altos (del orden de 105Ω·cm2) a frecuencias bajas e intermedias, lo cual esindicativo de la formación de una película altamente estable sobre estas aleaciones en solución Ringer. En resumen, elcomportamiento electroquímico de Ti6Al4V no se ve afectado si se sustituye el vanadio por el hierro.

PPaallaabbrraass ccllaavvee Titanio; Aleaciones; Biocompatibilidad; Corrosión; Polarización; Impedancia.

11.. IINNTTRROODDUUCCTTIIOONN

Since the 1980s, the titanium alloys are intensivelyapplied in the manufacturing of biomedical deviceswhere the most important factors are, firstly,biocompatibility, corrosion resistance, mechanicalbehaviour and osseointegration (facility to boneingrowths or fixes in the metal implant with thepatient’s bone structure) [1-6].

Corrosion resistance is a critical property for thiskind of application because physiological fluids are

chloride containing solutions, with concentrationof approximately 1% wt. NaCl. Furthermore,corrosion products are largely responsible for limitedbiocom patibility and can produce undesirablereactions in implant-adjacent tissues. Corrosionresistance plays a decisive role in determining thesuccessful use of metal alloys as biomaterials.

Titanium is difficult to cast because it has a highmelting point value (1,682 oC) and presentstendency to contamination due to an increasedreactivity.

(•) Trabajo recibido el día 25 de septiembre de 2007 y aceptado en su forma final el día 28 de julio de 2008.* Technical University of Iasi, Faculty of Chemical Engineering, D. Mangeron nº71, 700050, Iasi, Romania.** Las Palmas de Gran Canaria University, Dept. Mechanical Engineering, 35017 Las Palmas de Gran Canaria, Spain.

EFFECT OF REPLACEMENT OF VANADIUM BY IRON ON THE ELECTROCHEMICAL BEHAVIOUR OF TITANIUM ALLOYS IN SIMULATED PHYSIOLOGICAL MEDIA

EFECTO DE LA SUSTITUCIÓN DE VANADIO POR HIERRO EN EL COMPORTAMIENTO ELECTROQUÍMICO DE ALECIONES DE TITANIO EN UN MEDIO FISIOLÓGICO SIMULADO

REV. METAL. MADRID, 45 (1), ENERO-FEBRERO, 32-41, 2009, ISSN: 0034-8570, eISSN: 1988-4222, doi: 10.3989/revmetalm.0750 33

In order to decrease the quantity of melting impu -rities and to increase the wear resistance, titaniumis alloyed with various elements. Titanium alloyshaving biphasic structure (α+β) are used on alarge scale. Among the alloying elements, Al isα-stabilizer and it dissolves in the β-phase increasingthe stability domain of this phase. Mo, V, Nb, Ta areβ-stabilizers elements which decrease thetemperature of allotropic transformation, formingcontinuous series of solid solutions with β-Ti buttheir solubility in α-Ti is limited. Fe is a β-stabilizerelement, which forms with the titanium eutectic andeutectoids phases.

Ever since the pioneer metal alloys have been useas biomaterials, lack of biocompatibility has beenextensively reported and research on improvedmaterials with appropriate mechanical behaviourand adequate biocompatibility was developed. TheTi6Al4V alloy was the first titanium alloy registeredas implant material[7]. Further studies have indicatedthat vanadium, used to stabilize the β-phase,produces harmful oxides for the human body[8 and 9].The toxicity of vanadium pushed forward the searchfor materials to replace Ti6Al4V. It has beensuggested by Khan et al.[10] that Ti6Al7Nb can be abetter alternative to Ti6Al4V because of its corrosionresis tance and resistance to loss the mechanicalproperties with changes in pH in simulated bodyfluid environ ment.

Electrochemical behaviour of titanium and itsalloys has been reported based on biocompatibilityin various physiological mediums. Pan et al.[11]

investigated the electrochemical behaviour ofcommercially pure titanium in phosphate-bufferedsaline with and without H2O2 additions. Gonzalezand Mirza-Rosca[12] investigated the electroche micalbehaviour of titanium and several of its alloys inRinger’s solution using the electrochemical impedan -ce spectroscopy. Silva et al.[13] carried out electroche -mical characterization of oxide films formed onTi6Al4V alloy implanted with Ir in phosphate-buffered saline solution.

In the present study, the electrochemical andcorrosion behaviour of Ti6Al4V has been comparedwith that of titanium alloys in which vanadium wasreplaced with Fe. The Ti5Al2.5Fe alloy was evaluatedbecause it has been suggested as a candidate humanbody implant material[14].

Two electrochemical techniques were used forthis purpose: potentiodynamic polarization andelectrochemical impedance spectroscopy (EIS).Standard corrosion parameters such as corrosioncurrent density (icorr), corrosion potential (Ecorr) andpolarization resistance (Rp), were evaluated fromthese experiments.

22.. MMAATTEERRIIAALLSS AANNDD MMEETTHHOODDSS

22..11.. MMaatteerriiaallss aanndd ssaammppllee pprreeppaarraattiioonn

Titanium alloys were purchased from Rare and Non-Ferrous Metals Institute, Bucharest, as cast. Thetitanium-based materials used in this work were:Ti6Al3.5Fe, Ti5Al2.5Fe and Ti6Al4V alloys.

The testing medium was an aerated Ringer’ssolution whose composition is: NaCl – 8.6 g/l, KCl– 0.3 g/l, CaCl2 – 0.48 g/l.

The pH was measured with a multiparameteranalyzer CONSORT 831C. The pH of this referencesolution was 6.5.

From each material were cut three samples of 1cm2 size and brass nut was attached to each sampleusing conductive paint to ensure electricalconductivity. The assembly was then embedded intoan epoxy resin disk. Then the samples were groundwith SiC abrasive paper up to 1000 grit, finalpolishing was done with 1 µm alumina suspension.The samples were degreased with ethyl alcoholfollowed by ultrasonic cleaning with deionised waterand dried under a hot air stream.

22..22.. MMiiccrroossttrruuccttuurraall cchhaarraacctteerriissaattiioonn

The structural studies were performed with an opticalmicroscope, Olympus PME 3-ADL. The microstruc -tures were revealed by etching in 10 % HF + 5 %HNO3 solution for 3-5 s at 25

0C and were analysedusing Scion Image software.

22..33.. PPootteennttiiooddyynnaammiicc ppoollaarriizzaattiioonn ssttuuddiieess

For all the electrochemical tests, experiments wereperformed using an aerated Ringer’s solutionmaintained at 25 ± 1 oC as an electrolyte. Theassembled specimen was placed in a glass corrosioncell, which was filled with freshly prepared electrolyte(within 24 h). A saturated calomel electrode (SCE)was used as the reference electrode and a platinumcoil as the counter electrode.

The measurements were performed with aPrinceton Applied Research potentiostat (Model 263A) controlled by a personal computer and a specificsoftware (PowerCorr, Princeton Applied Research).

For each specimen, 24 h of open circuit potential,EOC, measurement was performed initially followedby the linear polarization measurement. Linearpolariza tion was conducted from EOC –150 mV toEOC + 150 mV at a scanning rate of 0.2 mV/s, to

D. MARECI, V. LUCERO AND J. MIRZA

34 REV. METAL. MADRID, 45 (1), ENERO-FEBRERO, 32-41, 2009, ISSN: 0034-8570, eISSN: 1988-4222, doi: 10.3989/revmetalm.0750

identify the ba (Tafel slopes for the partial anodicprocesses) and bc (Tafel slopes for the partial cathodicprocesses), the icorr (corrosion current density) andthe Ecorr (corrosion potential). It was then followedby the general polarization tests from -600 mV to1200 mV at a scanning rate of 0.5 mV/s for evaluatingthe stability of passivation. From the obtained anodicpolarization curves, ipass (passivation current density)was determi ned.

22..44.. EElleeccttrroocchheemmiiccaall iimmppeeddaannccee ssppeecc --ttrrooss ccooppyy ssttuuddiieess ((EEIISS))

The electrochemical impedance spectroscopy (EIS)was performed 24 h after immersion in aeratedsolutions at the open circuit potential using aPrinceton Applied Research potentiostate (Model263 A) connected with a Princeton Applied Research5210 lock-in amplifier.

The spectra were recorded in the 10–2 Hz to 105

Hz frequency range. The applied alternating potentialsignal had an amplitude of 10 mV.

Data acquisition and analysis were performed witha personal computer. The spectra were interpretedusing the ZSimpWin program. The programme uses avariety of electrical circuits to numerically fit themeasured impedance data. The program is capableof conducting analysis of heavily convoluted frequencydispersion data by deconvoluting the complex responseinto those of simple subcom ponents. This approachcombined with the general nonlinear least squares fitprocedure allowed us to construct equivalent circuit(EC), whose simulated responses fit well the actuallymeasured data. Impedance data were represented inimpedance amplitude and phase angle plots (Bodeplots). The Bode representation shows the logarithmof the impedance modulus (Zmod) and phase angle asa function of the logarithm of the frequency. Thefrequency dependence of the phase angle indicateswhether one or more time constants occur and can beused to determine the values of the parameters inthe equivalent circuit (EC).

22..55.. SSEEMM aanndd XXRRDD ssttuuddiieess

Surface morphology after the electrochemicaltreatments of the samples was studied with anresearch electronic microscope of Tesla BS 300 type.

XRD patterns were measured using a PhilipsAnalytical PW3710 XPERT system with a Ni filterand Cu Kα (λ1.5418 A) radiation (40 kV, 30 mA)

at 0.05° steps at the rate of 10s per step over the range10°<2θ<70°.

33.. RREESSUULLTTSS

33..11.. MMiiccrroossttrruuccttuurraall cchhaarraacctteerriizzaattiioonn

Surfaces of the three alloys were characterized fromthe structural point of view. As cast, titanium-basedmaterials have a structure that is determined by thecooling from α-phase domain at a moderate coolingrate. The investigated titanium alloys have aWidmanstätten-type structure that consists of α-phaselamellae. The optical images from figure 1 a), b) andc) show a basket weave morphology of α-phaseregardless of chemical compositions of titanium alloys.

In all the microstructures the ‚ phase appears darkand the α phase light. All the studied alloys werealpha-beta alloys. Alpha was the dominant phase inall the alloys: 80.3% in Ti6Al4V, 78% in Ti6Al3.5Feand 84% in Ti5Al2.5Fe.

In the case of the alloys with vanadium the widthof the lamellas is bigger than in the case of the twoalloys with iron; it can be supposed that this alloy ismore susceptible to corrosion because the compositionaldifference across the grain boundaries is bigger andleads to the galvanic cell formation.

33..22.. PPootteennttiiooddyynnaammiicc ppoollaarriizzaattiioonn ssttuuddiieess

The metals immersed in an electrolytic environ mentgenerate an electric potential that varied with time.After a period of immersion it stabilises around astationary value. This potential may vary with timebecause changes in the nature of the surface of theelectrode occur (oxidation, formation of the passivelayer or immunity). The open circuit potential is usedas a criterion for the corrosion behaviour. The opencircuit potential is a parameter whish indicates thethermodynamically tendency of a material toelectrochemical oxidation in a corrosive medium.

Figure 2 shows the EOC curves for all the threesamples immersed in Ringer’s solution at 25 oC.

Open circuit potential variation is similar for allthe samples. Initially, the potential of the twotitanium-iron based alloys presents approximatelythe same value: -547 mV for Ti6Al3.5Fe and -540mV for Ti5Al2.5Fe while Ti6Al4V alloy has the mostnegative potential: –667 mV.

During the first moments of immersion, an abruptEOC displacement towards positive potentials wasnoticed in figure 2 during a period of 1-2 h. This initial




increase seems to be related to the formation andthickening of the oxide film on the metallic surface,improving its corrosion protection ability[15-18].Afterwards, the EOC increases slowly suggesting thegrowth of the film onto the metallic surface.

The studied samples did not exhibit potentialdrops associated with surface activation during 24h exposure in the Ringer’s solution. This kind ofbehaviour strongly suggests that the air-formed nativeoxide is thermodynamically resistant to chemicaldissolution in Ringer’s solution. Open circuit potentialvalues are presented in table I at the beginning of thetest and after 24 h of immersion in Ringer’s solution.

Standard techniques were used to extract Ecorr andicorr values from the potentiodynamic polarizationplots. The Tafel slopes (ba and bc) were determinedby fitting the theoretical polarisation curve to theexperimental polarisation curve plotted in a range of± 150 mV vs. EOC. The two Tafel slopes intercept atthe point of the coordinates (Ecorr, icorr). In most ofthe aerated neutral solutions the corrosion rate iscontrolled by the oxygen diffusion at the surface ofthe electrode when bc >> ba. This supposition isconfirmed by the values presented in the table II withthe exception of the Ti5Al2.5Fe alloy. The corrosioncurrent density (icorr) is representative for the

FFiigguurree 11.. Metalographic microscopy images for:a) Ti5Al2.5Fe, b) Ti6Al3.5Fe, c) Ti6Al4V. Thesamples were etched using 10 % HF + 5 %HNO3 solution at 25

oC.

Figura 1. Imagenes de microscopia metalografi-ca para: a) Ti5Al2.5Fe, b) Ti6Al3.5Fe, c) Ti6Al4V.Las muestras se han atacado con disolucion 10% HF + 5%HNO3 a 25

oC.

0 4 8 12 16 20 24

Time (h)

-700

-600

-500

-400

-300

-200

-100

Ti6Al4V alloyTi6Al3.5Fe alloy

Ti5Al2.5Fe alloy

EO

C (

mV

vs.

SC

E)

FFiigguurree 22. Variation of open circuit potential (EOC)with time for the three titanium alloys in Ringer’ssolution.

Figura 2. Variación del potencial en circuito abier-to (EOC) con el tiempo para tres aleaciones detitanio en disolución Ringer.



degradation degree of the alloy. The average valuesba, bc, Ecorr and icorr from three different polarizationcurves determined by the PowerCorr program arepresented in Table II. The very low icorr valuesobtained for the three tested titanium alloys aretypical of passive materials.

The corrosion current densities for all the alloyswere of the same order of magnitude (nA/cm2): 210nA/cm2 for Ti6Al4V alloy, 185 nA/cm2 for Ti6Al3.5Fealloy and 165 nA/cm2 for Ti5Al2.5Fe alloy.

Figure 3 compares typical potentiodynamic curvesin a semi-logarithmic version between -600 mV and+1,200 mV of the three titanium alloys tested inRinger’s solution aerated at 25°C after being immersedfor 24 h.

The nature of the potentiodynamic polarizationcurves indicated that all the samples have beenpassivated immediately after the immersion in theRinger’s solution; all three materials translated directlyfrom the “Tafel region” into a stable passive state,without exhibiting a common active-passive transition.

For the potentials above approximately 100 mVup to approximately 300 mV all the three anodic

curves indicates behaviour typical of activationpolarization showing a well defined linear range,suggesting conformity to the Tafel’s Law. From 300mV the curves show a passive behaviour. Passivecurrent density (ipass) was also determined from thepotentiodynamic anodic diagram of each specimenin Ringer’s solution. Passive current densities (ipass)are obtained around the middle of the passive rangeand are listed in Table II. The lowest ipass is observedon Ti6Al4V alloy (at about 4.5 µA/cm2) and thelargest on the Ti5Al2.5Fe (at about 8.5 µA/cm2). The

TTaabbllee IIII.. Electrochemical parameters of corrosion process after 24 h of immersion in Ringer’ssolution

Tabla II. Parámetros electroquímicos del proceso de corrosión después de 24 h de inmersión endisolución Ringer

AAllllooyy EEccoorrrr ((mmVV)) bbaa((mmVV//ddeecc..)) bbcc((mmVV//ddeecc..)) iiccoorrrr ((nnAA//ccmm22)) iippaassss((µµAA//ccmm22))

Ti6Al4V 78 ± 10 85 ± 7 ∞ 210 ± 7 4.5 ± 0.3Ti6Al3.5Fe 96 ± 18 82 ± 6 ∞ 195 ± 10 6.3 ± 0.3Ti5Al2.5Fe 75 ± 15 90 ± 10 250 ± 10 185 ± 10 8.5 ± 0.5

-0.8 -0.4 0 0.4 0.8 1.2

Potential, E (V vs SCE)

1E-007

1E-006

1E-005

0.0001

Ti6Al4V alloyTi6Al3.5Fe alloy

Ti5Al2.5Fe alloy C

urre

nt d

ensi

ty, i

(A

/cm

2 )FFiigguurree 33. The potentiodynamic polarisationcurves for titanium alloys after 24 h of immersionin Ringer’s solution at 25 oC. Scan rate 0.5 mV/s.

Figura 3. Curvas de polarización potenciodiná-mica para aleaciones de titanio después de 24horas de inmersión en disolución Ringer a 25oC. Velocidad de barrido 0,5 mV/s.

TTaabbllee II.. Open circuit potential values of thetested samples: initially and after 24 h of

immersion in Ringer’s solution

Tabla I. Valores de potencial de circuitoabierto: inicial y después de 24 h de la

inmersión de las muestras de aleación endisolución Ringer

OOppeenn cciirrccuuiitt EEOOCCAAllllooyy ppootteennttiiaall,, ((mmVV vvss.. SSCCEE))

IInniittiiaall AAfftteerr 2244 hh

Ti6Al4V –667 ± 15 –192 ± 6Ti5Al2.5Fe –540 ± 11 –165 ± 5Ti6Al3.5Fe –547 ± 12 –161 ± 5




passive current densities (ipass) for all three alloysamples are higher than corrosion current densities(icorr) and suggest that the protective Ti oxide filmcan be more defective[19 and 20]. From the polarizationcurves it was found that all the alloys are maintainedin their passive state at 1.2 V potential too. Nobreakdown was observed for any of the alloys.

No significant differences were found in Ecorr, icorror ipass for all the samples. The results indicate thatthe polarization curves behaviour of the Ti6Al3.5Feand Ti5Al2.5Fe alloys resembles that of the Ti6Al4Valloy. These alloys had a good corrosion resistancewithin the potential range used in this study.

Figure 4 shows the SEM images of Ti6Al4V andTi5Al2.5Fe alloys after the electrochemicaltreatments. After electrochemical treatments saltdepositions were formed on the surface of the alloy.The Ringer’s solution contains calcium ion, whichgenerally precipitates on titanium alloys surface. Thissupposition is confirmed by XRD spectra, presentedin figure 5 from Ti5Al2.5Fe alloy after 24 h immersionin Ringer’s solution and electrochemical treatments.

33..33.. EElleeccttrroocchheemmiiccaall iimmppeeddaannccee ssppeecc --ttrrooss ccooppyy ((EEIISS)) ssttuuddiieess

The corrosion resistance can also be estimated bymeans of the impedance method known asElectrochemical Impedance Spectroscopy (EIS). Thistechnique requires minimal invasive procedures,neither the oxidation nor the reduction was forced

to take place in the open circuit mode. From the EISdata we can obtain information about the passivefilm dissolution (the rate and type of the control)and the passive film characteristics (film thickness,passive film dielectric constant, diffusion coefficientof the diffusing species etc.).

Figures 6 and 7 shows Bode diagrams (Phase anglevs. log Frequency and Z modulus vs. log Frequency)of the titanium alloys at the open circuit potentialafter 1 min and 24 h respectively, of immersion inRinger’s solution at 25 °C.

FFiigguurree 44. The salts depositions on the surfaces of the: a) Ti5Al2.5Fe alloy, b) Ti6Al4V alloy, afterpolarization tests from –600 mV to 1,200 mV at a scanning rate of 0.5 mV/s.

Figura 4. Las deposiciones de sales sobre la superficie de: a) aleacion Ti5A12,5Fe, b) aleacion Ti6Al4V,después de los ensayos de polarización desde –600 mV hasta 1.200 mV a una velocidad de barridode 0,5 mV/s.

FFiigguurree 55. XRD pattern of the Ti5Al2.5Fe alloysurface after 24 h immersion in Ringer’s solutionand electrochemical tests.

Figura 5. Difractograma de Rayos X en la super-ficie de la aleación Ti5Al2.5Fe después de 24 hde inmersión en disolución Ringer y ensayoselectroquímicos.



From the Bode spectra it is possible to observe thepresence of a compact passive film if: (a) the phaseangle is close to –90° over a wide frequency range,and (b) if the spectrum shows linear portions atintermediate frequency.

From figure 6, the phase angle observed at lowfrequency for immediate immersion (1 min) wasnearly –60°. However, at intermediate frequency, thephase angle shifted to 80° and remained constantover a wide range of frequency, indicating a nearcapacitive response for all the alloys. The highimpedance values and phase angle of –80° areindicative of a single thin passive oxide film presenton the surface at 1 min of immersion. The phaseangles approaching –80° in a large frequency rangeare characteristic of a compact passive film[12].

After 24 h of immersion, at low frequency, thephase angle values were shifted to –70°, but therewasn’t a significant difference in the response atmedium frequencies.

For the interpretation of the electrochemicalbehaviour of a system from EIS spectra, an appropriatephysical model of the electrochemical reactionsoccurring on the electrodes is necessary. Theelectrochemical system may be represented by anequivalent circuit (EC). An EC consists of variousarrangements of resistances, capacitors and othercircuit elements, and provides the most relevantcorrosion parameters applicable to the subs tra -te/electrolyte system. The usual guidelines for theselection of the best-fit EC were followed:

— a minimum number of circuit elements wereto be employed;

— the ¯2 error was suitably low, and the errorassociated with each element were up to 5 %.

The impedance spectra were fitted using theZsimpWin software. The results of the analysis areshown in Table III. Generally, for TiO2 films a

TTaabbllee IIIIII.. Electrical parameters of theproposed equivalent circuit obtained for the allstudied samples at 1 minute of immersion in

Ringer’s solution

Tabla III. Parámetros eléctricos del circuitoequivalente propuesto para las muestras

estudiadas después de 1 minuto de inmersiónen disolución Ringer

AAllllooyy RR11 nn11 110055 CCPPEE11(( ccmm22)) ((SS ssnn ccmm--22))

Ti6Al4V 1.5 105 0.85 6.6 ± 0.2Ti5Al2.5Fe 1.5 105 0.88 8.1 ± 0.2Ti6Al3.5Fe 1.6 105 0.86 7.2 ± 0.2

0.01 0.1 1 10 100 1000 10000100000

Frequency (Hz)

1

10

100

1000

10000

100000

1000000

Zm

od (

ohm

cm

2 )

0

20

40

60

80

-Pha

se a

ngle

(de

gree

)Ti6Al4V alloy (experimental)

Ti6Al3.5Fe alloy (experimental)Ti5Al2.5Fe alloy (experimental)

Simulation

FFiigguurree 66. Bode plots of tested materials recordedat open circuit potential after immersion for 1 minin Ringer’s solution at 25 °C.

Figura 6. Graficas Bode de los materiales ensa-yados a potencial de circuito abierto despuésde la inmersión durante 1 min en disoluciónRinger a 25 °C.

0.01 0.1 1 10 100 1000 10000

Frequency (Hz)

100000

1

10

100

1000

10000

100000

1000000

Zm

od (

ohm

cm

2 )

0

20

40

60

80

-Pha

se a

ngle

(de

gree

)

Ti6Al4V alloy (experimental)Ti6Al3.5Fe alloy (experimental)Ti5Al2.5Fe alloy (experimental)Simulation

FFiigguurree 77. Bode plots of tested materials recordedat open circuit potential after immersion for 24 hin Ringer’s solution at 25 °C.

Figura 7. Graficas Bode de los materiales ensa-yados registrados en potencial de circuito abier-to después de la inmersión durante 24 h en diso-lución Ringer a 25 °C.




distributed relaxation feature is commonly observed.Due to this fact, in this study a constant phaseelement (CPE) was used for data analysis instead ofan ideal capacitor. Impedance of CPE is given by ZCPE= 1/[Q(jw)n] where Q is the combination of propertiesrelated to both the surfaces and electroactive speciesindependent of frequency; n is related to a slope ofthe lg Zmod vs lg Frequency in Bode-plots, ω is theangular frequency and j is imaginary number (). Q isan adjustable parameter used in the fitting routine:when the value of n is equal to 1, the CPE acts as apure capacitor[21].

Gonzalez and Mirza Rosca[12] proposed Rs(CPE1R1)as EC model to fit the EIS data in case of a singlepassive film present on the metal surface. Similarly,in the EC seen in figure 8 a) for titanium alloys at 1min of immersion in Ringer’s solution, consists of thefollowing elements: Rsol – ohmic resistance of theelectrolyte, CPE1 – constant phase element of thedouble layer and R1 – resistance of the barrier layer.From Table III, it can be seen that the values of nwere close to one (n > 0.85) for all titanium alloys at1 min of immersion in Ringer’s solution. Thisindicated a near ideal capacitance of the passive filmsformed on the titanium alloys.

The EC shown in figure 8 a) did not shows a goodfit of the experimental results for the titanium alloys at24 h of immersion in Ringer’s solution. As the immersiontime increased (24 h), the EC used for titanium alloys(Figure 8b) was Rsol(R1(CPE1(R2CPE2)), where CPE1represents the constant phase element of the outer layerand CPE2 the constant phase element of the inner layer,R1 and R2 are the resistance of the outer layer(precipitate layer) and inner layer. The results of thefittings are presented in table IV.

Dissolution of the titanium oxide film in waterfrom a thermodynamic point of view can berepresented by reaction:

TiO2solid + 2H2O = Ti(OH)4(aq) (1)

The majority of the dissolved species has beenfound to be uncharged[22]. The presence of a very smallamount of charged corrosion products is expected toinduce minimal unwanted biochemical reactions suchas strong protein binding and unfolding:

TiO2solid + 2H2O = [Ti(OH)3](aq) + OH– (2)

TiO2solid + 2H2O = [Ti(OH)–](aq) + H3O– (3)

The calcium ions present in the Ringer solutiontends to attach to titanium oxide surface uponimmersion:

a)

b)

CPE1

R1

Rsol

Rsol

CPE 1

CPE 2

R1

R2

FFiigguurree 88. Equivalent circuit (EC) used in the ge-neration of simulated data for titanium alloys inRinger’s solution at different immersion time: (a)1 min and (b) 24 h.

Figura 8. Circuito equivalente (EC) utilizado en lageneración de datos simulados para las alea-ciones de titanio en disolucion Ringer a diferen-tes tiempos de inmersion: (a) 1 min y (b) 24 h.

TTaabbllee IIVV.. Electrical parameters of the proposed equivalent circuit obtained for the all studiedsamples

Tabla IV. Parámetros eléctricos del circuito equivalente propuesto por todas aleaciones de titanioestudiados

AAllllooyy RR11 nn11 110055 CCPPEE11 RR22 nn22 110055 CCPPEE22 iiccoorrrr((Ω ccmm22)) ((SS ssnn ccmm--22)) ((WW ccmm22)) ((SS ssnn ccmm--22)) ((nnAA//ccmm22))

Ti6Al4V 2.1 103 0.9 2.6 ± 0.1 2.1 105 0.7 4.6 ± 0.2 174 ± 15Ti5Al2.5Fe 1.1 103 0.92 6.1 ± 0.3 2.5 105 0.75 6.7 ± 0.3 115 ± 11Ti6Al3.5Fe 1.7 103 0.9 7.2 ± 0.3 2.3 105 0.72 2.7 ± 0.2 155 ± 12



TiO2solid + Ca2+(aq) = [TiO–… Ca2+(aq)]solid + H+(aq) (4)

The same value for Rsol, equal to 41 ± Ω and 35± 2 W, was observed for all of the specimens after 1min and 24 h of immersion and was not inserted intables III and IV.

After 24 h of immersion large values of R2 at opencircuit potential (order of 105Ω cm2) were obtainedsupporting the passivation of the titanium alloys inRinger’s solution. R2 was greater than R1 by a factorof nearly 100, showing that the resistance of thepassive film on the titanium alloys is due to the innerlayer.

Polarization resistance (Rp) is represented by thesum of the resistance of the inner layer and outerlayer, R1 + R2. A high Rp value corresponds to a highresistance to passive dissolution. From the Stern-Geary equation[23]:

(5)

where: ba and bc are the Tafel slopes for the partialanodic and cathodic processes, respectively and B isa constant:

(6)

The polarization resistance of the alloys after 24h of immersion is large indicating a sufficient stabilityof titanium alloys in Ringer’s solution. The corrosioncurrent densities for the titanium alloys were of thesame order of magnitude (nA/cm2) and are inagreement with the polarization data.

The exponents of the CPE1, n1, take valuesbetween 0.9 and 0.92 while the exponents of theCPE2, n2, are between 0.7 and 0.75. The n valuesof nearly one suggest that the behaviour of outer layerapproached that of an ideal capacitor.

The fitted circuit for the obtained impedance datahas an error distribution less than 1% and the circuitshowed a perfect fitting for titanium alloys data inRinger’s solution.

DDIISSCCUUSSSSIIOONNSS

When an alloy is placed in a biological environment,an electrochemical interaction (corrosion) betweenthe two takes place. The effect of this interaction maybe manifested as: (1) release of soluble metallic ionsinto the environment; (2) formation of corrosion

products on the alloy surface; or (3) a combination ofboth. These effects, depending on the altered alloysurface and/or the nature of released metallic ions,may trigger adverse biological reactions such asallergy[24]. For a specific environment, corrosiondepends on the structure and the composition ofthe alloy. The titanium alloys of the present studyhave different compositions. All the studied alloyswere alpha-beta alloys and have a Widmanstätten-typestructure. Alpha was the dominant phase in all thealloys. The analysis of obtained results in all performedelectrochemical tests indicates that all three Ti alloysbehaved in a very similar way. The differences arerather small, and this can be ascribed to the fact thatthe passive film formed on these alloys is essentiallyof the same nature. The open circuit potential curvesshown in figure 2 indicate that the potential of the Tialloys in the Ringer’s solution at 250C increases rapidlywith time and this initial increase seems to be relatedto the thickening of the oxide film improving itscorrosion protection ability[25 and 26].

The very low corrosion current densities (order10–7A/cm2) obtained for the three tested titaniumalloys are typical of passive materials. The increasein current density with potential could occur if theincrease in potential is accompanied by acorresponding thickening film, or could be due to theoxidation of TiO or Ti2O3 to TiO2

[19 and 20].Nevertheless, it appears that the film formed above300 mV is different from that formed below thispotential. Its higher ipass value indicates it is a moredefective oxide.

High impedance values (order of 105 Ω·cm2) wereobtained from medium to low frequencies for all Tialloys suggesting a high corrosion resistance in theelectrolyte. This result supports the passive behaviourof the alloys, indicated by the polarization curves andopen circuit potential variation with time curves.

The EIS results indicated the presence of twolayers after 24 h immersion for all three Ti alloys.According to the proposed model, the passive filmconsists of two layers, the inner barrier layer (titaniumoxide) and the outer layer (precipitate layer). Theresistance values, R2, are significantly larger than thevalues associated to the precipitate layer as tableIV shows. These results indicate that the protectionprovided by the passive layer is predominantly dueto the inner layer.

CCOONNCCLLUUSSIIOONNSS

The electrochemical techniques used in thisinvestigation led to the following conclusions:

( )a c

corrp a c p

b b Bi

2.3R b b R= =

+

( )a c

a c

b bB

2.3 b b=

+

( )a c

corrp a c p

b b Bi

2.3R b b R= =

+

( )a c

a c

b bB

2.3 b b=

+




— All three titanium alloys has provided apassive film formation during corrosion testsin Ringer´s solution, which has been observedby the very low experimental current densities.

— The EIS spectra results indicated that the filmformed on the titanium materials is composedof a bi-layered consisting of an inner barrierlayer associated to high impedance andresponsible for corrosion protection and anouter precipitate layer.

— These alloys exhibited good corrosionresistance within the potential range used inthe present study.

— The electrochemical behaviour of the Ti5Al2.5Feand Ti6Al3.5Fe alloys examined in the presentstudy resembles that of Ti6Al4V alloy.

The EIS technique is a powerful method to cha-racterize the influence of the alloying elements oncorrosion behaviour.

AAcckknnoowwlleeddggeemmeennttss

The authors would like to acknowledge the financialsupport of this work by MEC (Ministerio de Educa -cion y Ciencia, Spain), Ref. NAN2006-27753-E.

RREEFFEERREENNCCEESS

[1] J.R. Newman, D. Eylon and J.K. Thorne, MetalsHandbook, ASM International, Metals Park,Ohio, USA, 1988, pp. 824.

[2] C. Eriksson, K. Ohlson, K. Richter, N.Billerdahl, M. Johansson and H. Nygren, J.Biomed. Mater. Res. 83A (2007) 1062-1069.

[3] E.B. Taddei, V.A.R. Henriques, C.R.M. Silvaand C.A.A. Cairo, Mater. Sci. Eng. 24C (2004)683-687.

[4] H.S. Kim, S.H. Lim, I.D. Yeo and W.Y. Kim,Mater. Sci. Eng. A 449-451 (2007) 322-325.

[5] Y. Mantani and M. Tajima, Mater. Sci. Eng.442A (2006) 409-413.

[6] C.R.M. Afonso, G.T. Aleixo, A.J. Ramírez and R.Caram, Mater. Sci. Eng. C 27 (2007) 908-913.

[7] T. Akahori and M. Niinomi, Mater. Sci. Eng.A243 (1998) 237-243.

[8] M.A. Khan, R.L. Williams and D.F. Williams,Biomaterials 20 (1999) 631-637.

[9] T.I. Kim, T.H. Han, I.S. Lee, K.H. Lee, M.C.Shin and B.B. Choi, Bio-Med. Mater. Eng. 7(1997) 253-263.

[10] M.A. Khan, R.L. Williams and d. F. Williams,Biomaterials 17 (1996) 212-217.

[11] J. Pan, D. Thierry and C. Leygraf, J. Biomed.Mater. Res. 28 (1994) 113-122.

[12] J.E.G. González, J.C. Mirza Rosca, J. Electroanal.Chem. 471 (1999) 109-115.

[13] T.M. Silva, J.E. Rito, A.M.P. Simoes, M.G.S.Ferreira, M. da Lunha Belo and K.G. Watkins,Electrochim. Acta 43 (1998) 203-211.

[14] J. Mathew Jr., Titanium: a Technical Guide, 2nded., ASM International, 2000, pp.143-282.

[15] S. Canary, N. Hersek, A. Culha and S. Bilgic,J. Oral Rehabil. 25 (1998) 759-764.

[16] T. Eliades, Int. J. Oral Maxillofac. Implants 12(1997) 621-627.

[17] J. Gluszek, J. Masalski, P. Furman and K. Nitsch,Biomaterials 18 (1997) 789-794.

[18] S. P. Kedici, A. A. Aksut, M.A. Kilicarslan, G.Bayramoglu and K. Gokdemir, J. Oral Rehabil.25 (1998) 800-808.

[19] S. Luiz de Assis, S. Wolynec and I. Costa,Electrochim. Acta 51 (2006) 1815-1819.

[20] A. Cremasco, W.R. Osorio, C.M.A. Freire, A.Garcia and R. Caram, Electrochim. Acta (2008)in press.

[21] I.C. Lavos-Valereto, S. Wolynec, I. Ramires,A.C. Guastaldi and I. Costa, J. Mater. Sci: Mater.Med. 15 (2004) 55-59.

[22] S.G. Steinemann, Periodontology, 17 (1998) 7-14.[23] M. Stern and A. Geary, J. Electrochem. Soc.

104 (1957) 56-63.[24] K. Merrit and S.A. Brown, Hypersensitivity to

metallic biomaterials. In: D.F. Williams, editor.Systemic aspects of biocompatibility, vol. 2, BocaRaton, FL: CRC Press, 1981, pp. 33-48.

[25] I.C. Lavos-Valereto, J. Costa and S. Wolynec,J. Biomed. Mater. Res. (2002) 664-672.

[26] S.Y. Yu and J.R. Scully, Corrosion 53 (1997)965-976.

Documents

Effect of replacement of vanadium by iron on the