On-line version ISSN 0104-6632
Braz. J. Chem. Eng. vol. 15 n. 4 São Paulo Dec. 1998
PREPARATION AND CHARACTERIZATION OF Ti-Al-Nb ALLOYS FOR ORTHOPEDIC IMPLANTS
V. OLIVEIRA, R.R. CHAVES, R. BERTAZZOLI and R. CARAM
Departamento de Engenharia de Materiais, Faculdade de Engenharia Mecânica, UNICAMP
C.P. 6122, Campinas, SP, 13083-970, Brasil - Phone: +55-19-788-3309,
Fax: +55-19-289-3722 Email: email@example.com
(Received: November 28, 1998; Accepted: August 25, 1998)
Abstract - Pure titanium shows very interesting characteristics such as high strength-to-weight ratio, very good corrosion resistance and excellent biocompatibility, which make this material appropriate for use in orthopedic and dental implants. Due to the mechanical properties of pure titanium, its use in implants is restricted to applications which involve moderate mechanical stress, such as dental implants. In applications where high mechanical strength is necessary, like orthopedic implants, it is appropriate to employ titanium-based alloys, which have better properties than pure titanium. The present work is related to the microstructure and corrosion resistance characterization of the Ti-6Al-7Nb alloy, designed to be used in orthopedic prostheses.
Keywords: Corrosion resistance, titanium, microstructure, orthopedic implants.
The use of titanium has increased in importance since it was first applied by the aerospace industry in the 1950s (Betner, 1985). Today, a large number of alloys with variable compositions and microstructures is available, and they can be employed in different fields like in the manufacturing of aerospace components and orthopedic implant devices.
One of the most important applications of titanium is the one related to the artificial hip substitution. In the human body, the hip joint, which links the femur and the pelvis, plays an important role in providing the necessary rotary movement to the body. As a result of a disease or a fracture that can cause a detachment between the femur and its head, the hip joint needs to be replaced in order to allow the person to regain natural movement (Callister Jr., 1994).
Pure titanium and titanium alloys show very interesting characteristics, such as high strength-to-weight ratio, very good corrosion resistance and excellent biocompatibility, which make such materials appropriate for use in orthopedic and dental implants (Callister Jr., 1994). Furthermore, when the titanium and its alloys are placed in contact with bone, it does not result in the generation of a fibrous tissue barrier. Due to the mechanical properties of pure titanium, its use in implants is restricted to the applications that involve moderate mechanical stress, such as dental implants. In applications where high mechanical strength is required, like orthopedic implants, it is appropriate to use titanium alloys which have better mechanical properties than pure titanium.
Most of the titanium alloys are the result of an allotropic phenomenon. Titanium undergoes an allotropic transformation at 883oC (Polmear, 1995; Porter, 1995). Above this temperature, titanium has the BCC crystal structure (b phase), while at lower temperatures it has the HCP crystal structure (a phase) form (Flower, 1990). This transformation temperature can be severely changed by the level of the interstitial content as well as by alloying elements. The addition of a series of elements may change the a /b structure transformation temperature (Okazaki, 1993). While elements such as oxygen and aluminum stabilize the HCP structure, elements such as vanadium and niobium stabilize the BCC structure. Consequently, different types of titanium alloys are commercially available. According to their microstructure, these alloys are divided into a , a+b or b alloys (Porter, 1995). By adding a proper amount of aluminum and vanadium, one is able to obtain a titanium alloy with a microstructure presenting a combination of the a and b phases. Since such an alloy can be plastically deformed, it is possible to attain high tensile strength values (Stubbington, 1974).
Among the commercial titanium alloys, the Ti-6Al-4V alloy is the most extensively employed titanium alloy as it associates high strength along with workability and good fatigue resistance, good corrosion resistance, and high biocompatibility (Callister Jr., 1994). The Ti-6Al-4V alloy is a a+b alloy type, where the a phase is stabilized by aluminum, and the b phase is stabilized by vanadium. The a+b alloy type contains enough stabilizing elements to modify the a+b phase field in a way that both phases are present at room temperature. Such an alloy, which was first utilized in the aerospace industry, is one of the most useful materials in obtaining orthopedic prostheses. However, recent research led to the conclusion that vanadium is toxic to the human body. According to Ito et al. (1995) and Okazaki et al. (1996), vanadium can accumulate in human body parts, such as bone, kidneys and liver. Also, when compared with other metals like nickel and chromium, vanadium may be more toxic. An interesting alternative to solve this problem is the replacement of vanadium with niobium in a titanium a+b alloy type. By using a tissue reaction analysis, niobium is classified into a vital class of material, while vanadium belongs to a toxic class. The present work deals with the preparation in a vacuum arc furnace and characterization of a titanium alloy containing aluminum and niobium which is used in orthopedic implants. In such an alloy, the niobium is utilized to stabilize the b phase, substituting the vanadium. Considering that it is essential to find new alternatives for the classic Ti-4Al-6V alloy, the effect of niobium on the microstructure and on the corrosion resistance was studied.
In order to analyze the possibility of using niobium as a substitute for vanadium, titanium alloys with aluminum and niobium were prepared in an arc furnace in an inert atmosphere. A titanium alloy with 6% (weight) aluminum and 7% niobium was utilized as a reference alloy. This amount of niobium was chosen as a result of previous studies available in the literature (Semlitsch, 1987; Semlitsch, 1992). Such a content is believed to stabilize nearly 10% in volume of the b phase at room temperature. Also, titanium alloys with 6% aluminum and variable content of niobium were prepared.
The aim of the experiments was to analyze the microstructure, the solute distribution and the corrosion resistance of the alloys with niobium.
The ingots were prepared by weighing a proper amount of 99.9% pure titanium, 99.9% pure Al and 99.9% pure Nb and melting them together by arc-melting in an argon atmosphere. The samples used in the corrosion resistance study were melted three times, while those used in the microstructure investigation were melted five times. In both cases, the ingots were turned upside down between two consecutive meltings in order to increase the homogeneity. It was learned that the loss of aluminum (due to its high volatility) during the entire alloy preparation procedure was negligible.
After melting and solidification, the ingots weighed 25 g, were 40 mm long and were 15 mm in diameter. They were removed from the water cooled copper crucible and hot worked by isothermal forging at 950 oC. This temperature is just below the (a+b)/b transus temperature. For titanium alloys with a content of 10% of aluminum and b-stabilizers, the (a+b)/b transus temperature is close to 1.000oC. This forging procedure was applied in order to break the microstructure, and hence to obtain finer grains. An area reduction close to 10% area was utilized. The samples were cut along the longitudinal direction and perpendicularly to it. The resulting cross and longitudinal sections from several locations along the ingot were fixed in bakelite by hot compression, mechanically polished and chemically etched with a solution of 3 ml HF, 6 ml HNO3 and 100 ml H2O for 10 s.
The range of potentials within which the materials remain in the passive state, as well as the passivation current density, were determined by comparing the electrochemical behavior of Ti-6Al-7Nb and the electrochemical behavior of other materials. In the corrosion test, the samples were made of pure titanium, hot rolled Ti-6Al-4V alloy and hot rolled Ti-6Al-7Nb alloy. The sample geometry was a cylinder of 4.0 mm in diameter and 40.0 mm in length and embedded in Teflon in such a way that a 0.126 cm2 cross section was exposed to the electrolyte. These measurements were carried out by using a three-electrode cell with a platinum wire counter electrode and a saturated calomel as reference. Samples were polished with emery paper (# 1200), degreased in acetone and rinsed three times in distilled water. A 5g/l NaCl, pH 4 solution was used as an electrolytic medium with no previous deaeration. In the anodic polarization measurements, the potential was scanned at 0.1 mV.s-1 rate using a EG&G 273A potentiostat.
The aluminum and niobium distribution along the samples was determined by using electron probe microanalysis. Such analysis showed that a very uniform solute distribution resulted from arc furnace preparation. In the case where both aluminum and niobium are used to stabilize the a and b phases, respectively, a proper amount of these elements allows one to obtain a mixture of such phases over a wide range of temperatures. However, any further heat treatment alters the final microstructure (Ivasishin, 1996; Niinomi, 1996; Okazaki, 1996). Therefore, after preparing the alloy by melting the constituents and solidifying the alloy, the way it is cooled down will affect the microstructure and hence, the alloy properties.
Figures 1.a, b and c show the micrographs of the as cast Ti-6Al-5Nb, Ti-6Al-7Nb and Ti-6Al-9Nb alloys, respectively. In such microstructures, the white grains are a-phase, while the dark grains are b-phase. The acicular shape of the a-phase is present in all the figures in an arrangement known as basket-weave which characterizes the Widmanstätten structure. This type of structure is obtained by the rapid cooling from the b field, promoted by the water-cooled copper crucible used to melt the alloy in the arc melting furnace. When a a+b titanium alloy is cooled from a high temperature (above 1000oC), the formation of the a phase from the b phase occurs. Such a process is similar to the formation of ferrite from the g phase during the cooling of a carbon steel. In such a microstructure, the b phase is found between the a plates. The same amplification used in all three figures allows one to observe that an increase in the niobium quantity produces a higher volume fraction of the b-phase, as well as the fact that a more refined structure of the a-phase is obtained. These micrographs show a set of preferential directions to the a-phase growth from the b-phase.
Figure 1: As cast microstructure of the Ti-Al-Nb alloy in a basket-weave arrangement, where the a phase is the white grains, while the b phase is the dark grains: (a) Ti-6Al-5Nb; (b) Ti-6Al-7Nb; (c) Ti-6Al-9Nb.
The as cast macrostructure analysis of the ingots showed very coarse grains on top, and more refined grains in the interface with the crucible. If the cooling rate is decreased, the microstructure becomes coarser.
In order to be transformed into final products, generally, titanium alloys are hot worked in the a+b field. This procedure allows one to modify the microstructure by obtaining a fine array of the a and b phases. In the present study, a broken grain structure was obtained by isothermal forging the considered TiAlNb alloys. This deformation process resulted in a microstructure where the a phase changes its morphology, leading to a broken structure of a with refined distribution of b, as shown in Figure 2. Although it is expected that a higher content of niobium results in a larger quantity of stabilized b phase, the microstructures of the Ti-6Al-5Nb and Ti-6Al-7Nb alloys (figures 2.a and 2.b, respectively) do not show a significant difference in the volume fraction of the b phase. In order to explain this phenomenon, the temperature of the isothermal forging should be considered. Both figures show a deformed structure formed by refined a grains in a fine matrix of the b phase. This situation is changed in the Ti-6Al-9Nb microstructure (figure 2.c), where a very deformed array of constituents is shown. In this case, the microstructure is formed by coarse a grains in a more b-rich matrix. The amount of a and b phases, after cooling from b field to room temperature, depends on the amount of a and b stabilizing elements (aluminum and niobium).
According to Semlitsch et al. (1985) in a Ti-Al-Nb alloy with the niobium quantity in the range of 6.5 to 7.5% of the total weight, the volume fraction of the b-phase is nearest to that presented by the traditional Ti-6Al-4V alloy (9-12,5%). By comparing the microstructures shown in Figure 2, one is able to affirm that the Ti-6Al-7Nb composition produces an interesting balance of a and b phases, with a proper volume of b grains and refined a grains.
Figure 2: Isothermal forged titanium alloys at 950oC and air cooled: (a) Ti-6Al-5Nb alloy; (b) Ti-6Al-7Nb and (c) Ti-6Al-9Nb.
Figures 3 to 5 present the anodic polarization curves for pure titanium, for the Ti-6Al-4V alloy and for the Ti-6Al-7Nb alloy, respectively. The evolution of the current intensity as a function of the potential (versus saturated calomel electrode) can be used in the evaluation of the corrosion resistance of a given material in a corrosive medium. As a common procedure, all polarization tests started at a negative potential of -1.0 V vs. SCE, increasing up to more positive values. The initial portion of the anodic polarization curves is related to the start of the corrosion process, where the metal directly interacts with the medium, leading to active corrosion. In the following stage, a passivation layer is formed, and then the current density becomes constant. This layer consists of metallic oxides, which protect the metal from further corrosion, and the kinetics of their nucleation and growth have been studied (Rosa et al., 1996). As a common feature Figures 3 to 5 show that in a third stage, shifting the potential to a more positive value leads to an increasing current density. This behavior may be attributed to the rupture of the passivation layer, and hence to the cessation of the metal protection giving birth to pitting corrosion. However, this hypothesis is not confirmed during the reverse scanning of potential. Polarization curves obtained during forward and backward scans of potential are superimposed and no pitting potential is observed. In the literature, pitting potential is shown to occur at potentials higher than 6.0 V vs. SCE in titanium alloys (Aragon et al., 1972). Furthermore, it is possible that the portion of the curves where an increasing current density is observed indicates oxidation of the aqueous media with a consequent oxygen evolution. Considering curves from Figures 3 to 5 this event is observed within a potential value of 1.35 ± 0.05 V vs. SCE, and it is known that formal equilibrium potential for this reaction is 1.48 V vs. SCE (Bard et al., 1988).
By using the anodic portion of the polarization curve, it is possible to estimate the corrosion behavior of a given metal in an electrolytic medium, which can be achieved by measuring the extension of the range of potentials in which the passivation layer exists, as well as by determining the current intensity of passivation.
A comparison of the anodic polarization curves shown in Figures 3 to 5 leads to the observation that the passivation layer of pure titanium is formed at -0.25 V and current density in the passive state was greater than 10-6 A/cm2. On the other hand, the polarization curve for the Ti-6Al-4V alloy shows formation of the passivation layer at -0.25 V and current density in the region of potentials over which the alloy remains in the passive state was smaller than 10-7 A/cm2. Finally, the observation of the Ti-6Al-7Nb voltammogram showed that the passivation layer is obtained at -0.4 V, with a current density of passivation also smaller than 10-7 A/cm2. As a common feature, the range of potential in which the materials investigated remain in the passive state extends over more than 1.0 V, while only 0.5 V above -0.25 V vs. SCE is enough for a metallic material to show an adequate performance as an orthopedic implant (Semlitsch et al., 1992).
Figure 3: The polarization curve for the pure titanium. Scan rate of 0.1 mV.s-1.
Figure 4: The polarization curve for the Ti-6Al-4V alloy. Scan rate of 0.1 mV.s-1.
Figure 5: The polarization curve for the Ti-6Al-7Nb alloy. Scan rate of 0.1 mV.s-1.
Furthermore, in the case of the Ti-6Al-7Nb alloy, an interesting phenomenon was observed. The passivation layer apparently undergoes a continuous process of partial rupture and regeneration. This event is probably related to the presence of niobium in the alloy, which is characterized by its behavior as a valve metal (Gatti, 1991) probably catalyzed by the presence of chloride ions in the support electrolyte.
As observed, these materials show outstanding resistance to corrosion. However, by comparing the three materials (pure titanium and Ti-6Al-4V and Ti-6Al-7Nb alloys), one is able to state that the alloys show better behavior than the pure metal, as the current density of passivation for the pure titanium is larger than the current intensities for the titanium alloys containing niobium or vanadium. On the other hand, the niobium alloy presented features similar to the Ti-6Al-4V once they presented equivalent corrosion resistance in the chloride medium. When the potential extension in which the metal remains in the passive state is considered, the results lead to the assumption that the three materials have the same behavior.
Titanium alloys containing aluminum and niobium were arc-melted, hot worked at 950oC, and a characterization was carried out. The microstructure analysis revealed that an a+b structure was obtained at room temperature. Samples in the as-cast condition, which were rapidly cooled from the b phase field, presented a basket-weave lamellar structure of the a-phase. After hot working, the samples presented a finer structure of broken a grains. The use of different amounts of niobium, ranging from 5 to 9wt% led to different levels of volume fraction of the b phase at room temperature. An increase in the amount of niobium caused an increase in the amount of the b phase. Evaluation of the electrochemical performance resulted in a conclusion that the Ti-6Al-7Nb alloy has a corrosion behavior as good as the Ti-6Al-4V alloy behavior. Based upon the results obtained, it can be affirmed that the Ti-6Al-7Nb alloy is a potential alloy candidate for replacing the traditional Ti-6Al-4V alloy in orthopedic implants.
This research was supported by CNPq, FAEP-UNICAMP and FAPESP.
Aragon, P.J.; Hulbert, S.F., Corrosion of Ti-6Al-4V in Simulated Body Fluids and Bovine Plasma, Journal of Biomedical Materials Research, Vol. 6, p. 155 (1972). [ Links ]
Bard, A.J.; Parsons, R.; Jordan, J., Standard Potentials in Aqueous Solutions. Marcel Decker ed., New York (1988). [ Links ]
Betner, D.R.; Bradwell, R.G.; Malone, R.F.; Newcomer, R.E.; Ruckle, D.L.; Redden, T.K.; Shaw, J.J.; Tobin, A.; Wilson, D.; Wood, R.A.; Zanoni, R.; Zoeller, H.W. and Benjamin, D., Metals Handbook, 9th ed., ASM, Metals Park OH, Vol. 3 (1985). [ Links ]
Callister Jr., W.D., Materials Science and Engineering, An Introduction, 3rd ed., John Wiley & Sons, Inc, New York (1994).
Flower, H.M., Microstructural Development in Relation to Hot Working of Titanium Alloys, Materials Science and Technology, Vol. 6, p. 1082 (1990). [ Links ]
Gatti, J.A.B., Estudo do Comportamento de Anodos Insolúveis no Processo de Eletrodeposição de estanho sobre chapas de aço carbono. Masters thesis, Faculdade de Engenharia Mecânica. Universidade Estadual de Campinas (1991). [ Links ]
Ito, A.; Okazaki, Y.; Tateishi, T. and Ito, Y., In Vitro Biocompatibility, Mechanical Properties, and Corrosion Resistance of Ti-Zr-Nb-Ta-Pd and Ti-Sn-Nb-Ta-Pd Alloys, Journal of Biomedical Materials Research, Vol. 29, p. 893 (1995). [ Links ]
Ivasishin, O.M. and Markovsky, P.E., Enhancing the Mechanical Properties of Titanium Alloys with Rapid Heat Treatment, Journal of Metals, Vol. 48, p. 48 (1996). [ Links ]
Niinomi, M. and Kobayashi, T., Materials Science and Engineering, Vol. A213, p. 16 (1996). [ Links ]
Okazaki, Y.; Ito, Y.; Ito, A. and Tateishi, T., Effect of Alloying Elements on Mechanical Properties of Titanium Alloys for Medical Implants, Materials Transactions, JIM, Vol. 34, p. 1217 (1993). [ Links ]
Okazaki, Y.; Ito, Y. and Tateishi, T., Effects of Heat Treatment on Mechanical Properties and Corrosion Fatigue Strength in Physiological Saline Solution of New Titanium Alloys for Medical Implants, Materials Transactions, JIM, Vol. 37, p. 843 (1996). [ Links ]
Okazaki, Y.; Ito, Y.; Kyo, K. and Tateishi, T., Corrosion Resistance and Corrosion Fatigue Strength of New Titanium Alloys for Medical Implants without V and Al, Materials Science and Engineering, Vol. A213, p. 138 (1996). [ Links ]
Polmear, I.J., Light Alloys, Metallurgy of the Light Metals, 3rd ed., Arnold Publishers, London (1995). [ Links ]
Porter, D.A. and Easterling, K.E., Phase Transformations in Metals and Alloys, 2nd ed., Chapman & Hall, London (1995).
Rosa, S.; Vivanco, M.J.F. and Bertazzoli, R., Proceedings of the 12th Brazilian Congress of Engineering and Materials Science (CBECIMAT), Vol. 2, p. 668, (1996). [ Links ]
Semlitsch, M.; Staub, F. and Weber, H., Titanium-Aluminium-Niobium Alloy Development for Biocompatible, High-strength Surgical Implants. Biomed. Technik, Vol. 30, p.334 (1985). [ Links ]
Semlitsch, M.F., Titanium Alloys for Hip Joint Replacements, Clinical Materials, Vol. 2, p. 1 (1987). [ Links ]
Semlitsch, M.F.; Weber, H.; Streicher, R.M. and Schön, R., Joint Replacement Components Made of Hot-forged and Surface-treated Ti-6Al-7Nb, Biomaterials, Vol. 13, p. 781 (1992). [ Links ]
Stubbington, C.A. and Bowen, A.W., Improvements in the Fatigue Strength of Ti-6Al-4V Through Microstructure Control, Journal of Materials Science, Vol. 9, p. 941 (1974). [ Links ]